HIV-targeted multispecific antigen-binding molecules and methods for their use

Description

FIELD

[0001] Provided are multi-specific antigen binding molecules and antigen binding fragments thereof for one or both of the treatment and the prevention of human immunodeficiency virus (HIV) infection.BACKGROUND

[0002] Human immunodeficiency virus (HIV) infection and related diseases are a major public health problem worldwide. Most currently approved therapies for HIV infection target the viral reverse transcriptase, protease enzymes, and integrase. Yet resistance of HIV to these existing drugs, long-term toxicity, and lack of patient adherence to daily dosing regimens have been associated with these therapies. Therefore, it is important to discover and develop new anti-HIV antibodies with advantageous properties suitable for therapeutic uses.

[0003] WO 2005 / 058963; WO 2010 / 107939; WO 2012 / 030904; WO 2012 / 158948; WO 2013 / 090644; WO 2013 / 016468; WO 2013 / 192589; WO 2014 / 063059 and WO 2018 / 125813 describe human anti-HIV antibodies derived from memory B cells of HIV-infected donors, which are capable of inhibiting infection by HIV-1 species from a plurality of clades. The therapeutic use of the antibodies is limited due to their intra-patient viral coverage, pharmacokinetics, induction of anti-drug antibodies, off-target binding (i.e., polyspecificity), and other properties that interfere with efficient manufacturing and storage, however.

[0004] Multi-specific antigen binding molecules are single molecules that can bind at least two different antigens. Bispecific antigen binding molecules are single molecules that can bind two different antigens. This property can be leveraged in a number of ways to improve the efficacy and / or selectivity of biotherapeutics, for example, by neutralizing the activity of two disease mediators instead of one, enhancing selective binding to disease over normal tissue, generating novel functions (e.g. Factor VIII mimicry of Emicizumab) or by directly recruiting immune cells for targeted killing (e.g. Blinatumomab recruitment of CD3+ T-cells to kill CD19+ B-cells). As such, the field of bispecific antibodies is rapidly growing, with potential applications in nearly every therapeutic area. There are currently three approved bispecific products (Blinatumomab, Ebmicizumab and Catumaxomab) and more than 50 clinical trials underway (antibodysociety.org).

[0005] A large number of different bispecific antibody formats have been described, with many of these being used to develop therapeutic molecules (e.g., reviewed in Spiess and Carter, 2015, Mol. Immunol, 67: 95-106). Production of bispecific antibodies is typically more complex than that of conventional antibodies. For instance, generation of Fab-arm exchanged bispecifics such as the Genmab Duobody platform (Labrijn et. al., 2013, PNAS, 110: 5145-5150) requires separate production of each half-antibody (as an IgG), then mixing these together under special conditions that enables Fab arm exchange of the two half antibodies to generate the desired bispecific molecule. The need for separate cell lines to produce each half antibody (or parental unreacted reduced antibody) IgG, purification of these intermediates, and optimization of the Fab arm exchange reaction and process to purify the target bispecific away from residual half antibody IgG adds significant time and complexity to research and development.

[0006] Other bispecific format strategies, such as those that pair an scFv fusion protein with a Fab-Fc fusion protein (e.g., WO 2016 / 086196; WO 2016 / 071004), ensure that there is only a single light chain as a strategy to avoid Fab arm exchange. Challenges with bispecific formats that include an scFv arise because the scFv moiety oftentimes does not bind the target antigen with desired affinity, can have undesirable off-target binding, and can contribute to a bispecific molecule that does not express well, is difficult to purify and has insufficient serum half-life to allow efficacy for an intended indication. Brozy, J. et al., Journal of Virology, 2018; 92(14): e00491-18 discuss antiviral activity of HIV gp120-targeting bispecific engager antibody constructs.SUMMARY

[0007] In one aspect, provided is a multi-specific antigen binding molecule that binds to human CD3 and HIV gp120, wherein the antigen binding molecule comprises: (a) a first antigen binding domain that comprises a first heavy chain variable domain (VH) and a first light chain variable domain (VL), wherein the first antigen binding domain binds to CD3, wherein the first antigen binding domain comprises a first VH-complementarity determining region (CDR) 1 (VH-CDR1), a first VH-CDR2, a first VH-CDR3, a first VL-CDR1, a first VL-CDR2 and a first VL-CDR3, comprising the following amino acid sequences, respectively (according to Kabat): SEQ ID NOs: 1, 12, 8, 4, 9 and 10, wherein the first VH and the first VL comprise amino acid sequences that are at least 99% identical to the amino acid sequences set forth, respectively: SEQ ID NOs: 51 and 56, and wherein the first antigen binding domain binds to CD3 with a KD of lower than 3.0 nM determined by surface plasmon resonance; and (b) a second antigen binding domain that binds to HIV gp120 comprising one or more extracellular (EC) domain of CD4, wherein the one or more EC domains of CD4 comprise a sequence that is at least 99% identical to:

[0008] The multi-specific (e.g., bispecific) antigen binding molecule that binds to human CD3 and a second antigen, comprises: (a) a first antigen binding domain that comprises a first heavy chain variable domain (VH) and a first light chain variable domain (VL), wherein the first antigen binding domain binds to CD3 and comprises: (i) a first VH-complementarity determining region (CDR) 1 comprising the amino acid sequence of TYAMN (SEQ ID NO:1); (ii) a first VH-CDR2 comprising the amino acid sequence of RIRSKYNNYATYYAX 1 SVKX 2 , wherein X 1 is A or D and X 2 is G or S (SEQ ID NO:2); (iii) a first VH-CDR3 comprising the amino acid sequence of HGNFGX 3 SYVSWFAY, wherein X 3 is H or N (SEQ ID NO:3); (iv) a first VL-CDR1 comprising the amino acid sequence of GSSTGAVTTGHYAN (SEQ ID NO: 4); (v) a first VL-CDR2 comprising the amino acid sequence of GTX 4 X 5 RAP, wherein X 4 X 5 is SN or NK (SEQ ID NO:5); and (vi) a first VL-CDR3 comprising the amino acid sequence of ALWYSNX 6 WV, wherein X 6 is L or R (SEQ ID NO:6), wherein the first VH-CDR1, the first VH-CDR2, the first VH-CDR3, the first VL-CDR1, the first VL-CDR2, and the first VH-CDR3 are according to Kabat; and (b) a second antigen binding domain that binds to a second antigen, wherein the first VH-CDR1, the first VH-CDR2, the first VH-CDR3, the first VL-CDR1, the first VL-CDR2 and the first VL-CDR3 comprise the following amino acid sequences, respectively (according to Kabat): SEQ ID NOs: 1, 12, 8, 4, 9 and 10.

[0009] The first VH-CDR1, the first VH-CDR2, the first VH-CDR3, the first VL-CDR1, the first VL-CDR2 and the first VL-CDR3 may comprise the following amino acid sequences, respectively (according to Chothia): SEQ ID NOs: 17, 18, 23, 20, 24 and 25.

[0010] The first VH-CDR1, the first VH-CDR2, the first VH-CDR3, the first VL-CDR1, the first VL-CDR2 and the first VL-CDR3 may comprise the following amino acid sequences, respectively (according to IMGT): SEQ ID NOs: 28, 29, 32, 31, 24 and 10.

[0011] The first VH-CDR1, the first VH-CDR2, the first VH-CDR3, the first VL-CDR1, the first VL-CDR2 and the first VL-CDR3 may comprise the following amino acid sequences, respectively (according to Honegger): SEQ ID NOs: 34, 43, 40, 37, 41 and 25.

[0012] The first VH comprises an amino acid sequence that is at least 99% identical to the amino acid sequence of SEQ ID NO: 51. In some embodiments, the first VH comprises the amino acid sequence of SEQ ID NO: 51. The first VL comprises an amino acid sequence that is at least 99% identical to the amino acid sequence of SEQ ID NO: 56. In some embodiments, the first VL comprises the amino acid sequence of SEQ ID NO: 56. The first VH and the first VL comprise the amino acid sequences set forth, respectively, or comprise amino acid sequences that are at least 99% identical to the amino acid sequences set forth, respectively: SEQ ID NOs: 51 and 56. In some embodiments, at least one of the first antigen binding domain and the second antigen binding domain independently comprise a Fab, an F(ab)2, Fv, a scFv, a sc(Fv)2, or a diabody. In some embodiments, the first antigen binding domain comprises a Fab and the second antigen binding domain comprises an extracellular domain of CD4. In some embodiments, the first antigen binding domain comprises a Fab and the second antigen binding domain comprises an extracellular domain of CD4, wherein the first VH and the first VL comprise the amino acid sequences set forth, respectively: SEQ ID NOs: 51 and 56. In some embodiments, the first antigen binding domain comprises a scFv and the second antigen binding domain comprises an extracellular domain of CD4. In some embodiments, the first antigen binding domain is a scFv comprising a cysteine (C) at position 44 in the scFv variable heavy domain; and a cysteine (C) at position 100 in the scFv variable light domain. In some embodiments, the first antigen binding domain is a scFv comprising a VH and a VL, the scFv comprising an amino acid sequence selected from SEQ ID NOs: 63.

[0013] With respect to further embodiments of the second antigen binding domain of the multi-specific (e.g., bispecific) antigen binding molecules targeting or binding to an HIV antigen, in some embodiments, the second antigen binding domain binds to an HIV envelope protein selected from the group consisting of gp120 and gp41. In some embodiments, the second antigen binding domain competes with or comprises VH and VL variable domains of a broadly neutralizing antibody (bNAb) that binds to an HIV antigen. In some embodiments, the second antigen binding domain binds to an epitope or region of gp120 selected from the group consisting of: third variable loop (V3) (e.g., high mannose patch) comprising a N332 oligomannose glycan; second variable loop (V2) (e.g., Env trimer apex); CD4 binding site (CD4bs); gp120 / gp41 interface; or silent face of gp120. In some embodiments, the second antigen binding domain binds to an epitope or region of gp120 in the CD4 binding site (CD4bs) and competes with or comprises VH and VL regions from an antibody selected from the group consisting of 3BNC117, GS-9723, GS-5423, 3BNC60, b12, F105, VRC01, VRC07, VRC07-523, VRC03, VRC06, VRC06b01 VRC08, VRC0801, NIH45-46, PGV04 (a.k.a., VRC-PG04); CH103, 44-VRC13.01, 1NC9, 12A12, N6, 1-18, N49-P7, NC-Cow1, IOMA, CH235 and CH235.12, N49P6, N49P7, N49P11, N49P9 and N60P25. In some embodiments, the second antigen binding domain binds to an epitope or region of gp120 in the CD4 binding site (CD4bs) and comprises one or more extracellular (EC) domains of CD4. In some embodiments, the one or more EC domains of CD4 comprise a sequence as set forth below, or a sequence that is at least 99% identical to a sequence selected from the group consisting of: (i) or (ii) In some embodiments, the EC domain of CD4 comprises a sequence that is at least 99% identical to the sequence of SEQ ID NO: 746. In some embodiments, the EC domain of CD4 comprises the sequence of SEQ ID NO: 746..

[0014] With respect to the first and second Fc regions or domains of the multi-specific antigen binding molecules some embodiments, the multi-specific (e.g., bispecific) antigen binding molecules comprise a heterodimeric human IgG1 or IgG4, comprising a first Fc region and a second Fc region. In some embodiments, the first Fc region and the second Fc region are derived from IgG1m17. In some embodiments, the multi-specific (e.g., bispecific) antigen binding molecules comprise a heterodimeric human IgG1 or IgG4, comprising a first Fc region and a second Fc region, wherein one or both of the first and second Fc regions comprise one or more of the following amino acids at the indicated positions (EU numbering): alanine at position 234, alanine at position 235; and serine at position 331. In some embodiments, the multi-specific (e.g., bispecific) antigen binding molecules comprise a heterodimeric human IgG1, comprising a first Fc region and a second Fc region, wherein both of the first and second Fc regions comprise one or more of the following amino acids at the indicated positions (EU numbering): alanine at position 234, alanine at position 235; and serine at position 331. In some embodiments, the multi-specific (e.g., bispecific) antigen binding molecules comprise a heterodimeric human IgG1, comprising a first Fc region and a second Fc region, wherein both of the first and second Fc regions comprise the following amino acids at the indicated positions (EU numbering): alanine at position 234, alanine at position 235; and serine at position 331. In some embodiments, the multi-specific (e.g., bispecific) antigen binding molecules comprise a heterodimeric human IgG1 or IgG4, comprising a first Fc region and a second Fc region, wherein one or both of the first and second Fc regions comprise the following amino acids at the indicated positions (EU numbering): tyrosine at position 252, threonine at position 254 and glutamic acid at position 256 (YTE); or leucine at position 428 and serine at position 434 (LS). In some embodiments, the multi-specific (e.g., bispecific) antigen binding molecules comprise a heterodimeric human IgG1, comprising a first Fc region and a second Fc region, wherein one of the first and second Fc regions comprise the following amino acids at the indicated positions (EU numbering): tyrosine at position 252, threonine at position 254 and glutamic acid at position 256 (YTE). In some embodiments, the multi-specific (e.g., bispecific) antigen binding molecules comprise a heterodimeric human IgG1, comprising a first Fc region and a second Fc region, wherein the second Fc region comprises the following amino acids at the indicated positions (EU numbering): tyrosine at position 252, threonine at position 254 and glutamic acid at position 256 (YTE). In some embodiments, the multi-specific (e.g., bispecific) antigen binding molecules comprise a heterodimeric human IgG1 or IgG4, comprising a first Fc region and a second Fc region, comprising the following amino acids at the indicated positions (EU numbering): the first Fc region comprises a tryptophan at position 366 (T366W); and the second Fc region comprises a serine at position 366 (T366S), an alanine at position 368 (L368A) and a valine at position 407 (Y407V); the first Fc region comprises a serine at position 366 (T366S), an alanine at position 368 (L368A) and a valine at position 407 (Y407V); and the second Fc region comprises a tryptophan at position 366 (T366W); the first Fc region comprises a cysteine at position 354 (S354C), a tryptophan at position 366 (T366W); and the second Fc region comprises a cysteine at position 349 (Y349C), a serine at position 366 (T366S), an alanine at position 368 (L368A) and a valine at position 407 (Y407V); the first Fc region comprises cysteine at position 349 (Y349C), a serine at position 366 (T366S), an alanine at position 368 (L368A) and a valine at position 407 (Y407V); and the second Fc region comprises a cysteine at position 354 (S354C), a tryptophan at position 366 (T366W). In some embodiments, the multi-specific (e.g., bispecific) antigen binding molecules comprise a heterodimeric human IgG1, comprising a first Fc region and a second Fc region, comprising the following amino acids at the indicated positions (EU numbering): the first Fc region comprises a tryptophan at position 366 (T366W); and the second Fc region comprises a serine at position 366 (T366S), an alanine at position 368 (L368A) and a valine at position 407 (Y407V); or the first Fc region comprises a serine at position 366 (T366S), an alanine at position 368 (L368A) and a valine at position 407 (Y407V); and the second Fc region comprises a tryptophan at position 366 (T366W). In some embodiments, the multi-specific (e.g., bispecific) antigen binding molecules comprise a heterodimeric human IgG1, comprising a first Fc region and a second Fc region, comprising the following amino acids at the indicated positions (EU numbering): the first Fc region comprises a serine at position 366 (T366S), an alanine at position 368 (L368A) and a valine at position 407 (Y407V); and the second Fc region comprises a tryptophan at position 366 (T366W). In some embodiments, the multi-specific (e.g., bispecific) antigen binding molecules comprise a heterodimeric human IgG1 or IgG4, comprising a first hinge region and a second hinge region, wherein one or both of the first and second hinge regions comprise a serine at position 220 (C220S) (EU numbering). In some embodiments, the multi-specific (e.g., bispecific) antigen binding molecules comprise a heterodimeric human IgG1 or IgG4, comprising a first Fc region and a second Fc region, wherein one of the first Fc region or the second Fc region comprise the following amino acids at the indicated positions (EU numbering): arginine at position 435 (H435R); or arginine at position 435 (H435R) and phenylalanine at position 436 (Y436F). In some embodiments, the multi-specific (e.g., bispecific) antigen binding molecules comprise a heterodimeric human IgG1, comprising a first Fc region and a second Fc region, wherein one of the first Fc region or the second Fc region comprise the following amino acids at the indicated positions (EU numbering): arginine at position 435 (H435R). In some embodiments, the multi-specific (e.g., bispecific) antigen binding molecules comprise a heterodimeric human IgG1, comprising a first Fc region and a second Fc region, wherein the first Fc region comprises arginine at position 435 (H435R). In some embodiments, the multi-specific (e.g., bispecific) antigen binding molecules comprise a heterodimeric human IgG1 or IgG4, comprising a first Fc region and a second Fc region, comprising the following amino acids at the indicated positions (EU numbering): the first Fc region comprises an alanine at position 234 (L234), an alanine at position 235 (L235A), a serine at position 331 (P331S) and a tryptophan at position 366 (T366W); and the second Fc region comprises an alanine a position 234 (L234), an alanine at position 235 (L235A), a serine at position 331 (P331S), a serine at position 366 (T366S), an alanine at position 368 (L368A), a valine at position 407 (Y407V), and an arginine at position 435 (H435R); the first Fc region comprises an alanine at position 234 (L234), an alanine at position 235 (L235A), a serine at position 331 (P331S) and a tryptophan at position 366 (T366W); and the second Fc region comprises an alanine a position 234 (L234), an alanine at position 235 (L235A), a serine at position 331 (P331S), a serine at position 366 (T366S), an alanine at position 368 (L368A), a valine at position 407 (Y407V), an arginine at position 435 (H435R) and a phenylalanine at position 436 (Y436F); the first Fc region comprises an alanine at position 234 (L234), an alanine at position 235 (L235A), a serine at position 331 (P331S), a serine at position 366 (T366S), an alanine at position 368 (L368A), and a valine at position 407 (Y407V); and the second Fc region comprises an alanine at position 234 (L234), an alanine at position 235 (L235A), a serine at position 331 (P331S) and a tryptophan at position 366 (T366W); the first Fc region comprises an alanine at position 234 (L234), an alanine at position 235 (L235A), a serine at position 331 (P331S), a tryptophan at position 366 (T366W), a leucine at position 428 (M428L) and a serine at position 434 (N434S); and the second Fc region comprises an alanine a position 234 (L234), an alanine at position 235 (L235A), a serine at position 331 (P331S), a serine at position 366 (T366S), an alanine at position 368 (L368A), a valine at position 407 (Y407V), and an arginine at position 435 (H435R); or the first Fc region comprises an alanine at position 234 (L234), an alanine at position 235 (L235A), a serine at position 331 (P331S), a tryptophan at position 366 (T366W), a tyrosine at position 252 (M252Y), a threonine at position 254 (S254T) and a glutamic acid at position 256 (T256E); and the second Fc region comprises an alanine a position 234 (L234), an alanine at position 235 (L235A), a serine at position 331 (P331S), a serine at position 366 (T366S), an alanine at position 368 (L368A), a valine at position 407 (Y407V), and an arginine at position 435 (H435R). In some embodiments, the multi-specific (e.g., bispecific) antigen binding molecules comprise a heterodimeric human IgG1, comprising a first Fc region and a second Fc region, comprising the following amino acids at the indicated positions (EU numbering): the first Fc region comprises an alanine a position 234 (L234), an alanine at position 235 (L235A), a serine at position 331 (P331S), a serine at position 366 (T366S), an alanine at position 368 (L368A), a valine at position 407 (Y407V), and an arginine at position 435 (H435R); and the second Fc region comprises an alanine at position 234 (L234), an alanine at position 235 (L235A), a serine at position 331 (P331S), a tryptophan at position 366 (T366W), a tyrosine at position 252 (M252Y), a threonine at position 254 (S254T) and a glutamic acid at position 256 (T256E). In some embodiments, the multi-specific (e.g., bispecific) antigen binding molecules comprise a heterodimeric human IgG1, comprising a first Fc region and a second Fc region, comprising amino acid sequences set forth, respectively, or comprising amino acid sequences that are at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% identical to the amino acid sequences set forth, respectively: SEQ ID NOs.: 696 and 697; SEQ ID NOs.: 697 and 696; SEQ ID NOs.: 696 and 698; SEQ ID NOs.: 698 and 696; SEQ ID NOs.: 699 and 700; SEQ ID NOs.: 700 and 699; SEQ ID NOs.: 701 and 698; SEQ ID NOs.: 698 and 701; SEQ ID NOs.: 702 and 703; SEQ ID NOs.: 703 and 702; SEQ ID NOs.: 704 and 698; SEQ ID NOs.: 698 and 704; SEQ ID NOs.: 705 and 703; SEQ ID NOs.: 703 and 705; SEQ ID NOs.: 706 and 704; SEQ ID NOs.: 704 and 706; SEQ ID NOs.: 707 and 703; SEQ ID NOs.: 703 and 707; SEQ ID NOs.: 708 and 704; SEQ ID NOs.: 704 and 708; SEQ ID NOs.: 709 and 710; or SEQ ID NOs.: 710 and 709. In some embodiments, the multi-specific (e.g., bispecific) antigen binding molecules comprise a heterodimeric human IgG1, comprising a first Fc region and a second Fc region, comprising amino acid sequences that are at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% identical to the amino acid sequences set forth, respectively: SEQ ID NOs.: 703 and 705 . In some embodiments, the multi-specific (e.g., bispecific) antigen binding molecules comprise a heterodimeric human IgG1, comprising a first Fc region and a second Fc region, comprising amino acid sequences that are at least 95% identical to the amino acid sequences set forth, respectively: SEQ ID NOs.: 703 and 705. In some embodiments, the multi-specific (e.g., bispecific) antigen binding molecules comprise a heterodimeric human IgG1, comprising a first Fc region and a second Fc region, comprising amino acid sequences that are at least 99% identical to the amino acid sequences set forth, respectively: SEQ ID NOs.: 703 and 705. In some embodiments, the multi-specific (e.g., bispecific) antigen binding molecules comprise a heterodimeric human IgG1, comprising a first Fc region and a second Fc region, comprising amino acid sequences set forth, respectively: SEQ ID NOs.: 703 and 705.

[0015] With respect to further embodiments of the multi-specific (e.g., bispecific) antigen binding molecules, in some embodiments, the multi-specific (e.g., bispecific) antigen binding molecule comprises (a) a first antigen binding domain that comprises a first heavy chain variable domain (VH) and a first light chain variable domain (VL), wherein the first antigen binding domain binds to CD3; and (b) a second antigen binding domain that binds to HIV gp120 comprising one or more extracellular (EC) domain of CD4, wherein the one or more EC domains of CD4 comprise a sequence as set forth below, or a sequence that is at least 99% identical to a sequence selected from the group consisting of: (i) or (ii) In some embodiments, the first antigen binding domain comprises a first VH-CDR1, a first VH-CDR2, a first VH-CDR3, a first VL-CDR1, a first VL-CDR2 and a first VL-CDR3 comprising the following amino acid sequences, respectively (according to Kabat): SEQ ID NOs: 1, 12, 8, 4, 9 and 10; and the second antigen binding domain comprises one EC domain of CD4 comprising an amino acid sequence that is at least 99% identical to a CD4 EC domain of SEQ ID NO:746. In some embodiments, the first antigen binding domain comprises a first VH-CDR1, a first VH-CDR2, a first VH-CDR3, a first VL-CDR1, a first VL-CDR2 and a first VL-CDR3 comprising the following amino acid sequences, respectively (according to Kabat): SEQ ID NOs: 1, 12, 8, 4, 9 and 10; and the second antigen binding domain comprises one EC domain of CD4 comprising the amino acid sequence of SEQ ID NO:746. In some embodiments, the first antigen binding domain comprises a first VH and a first VL comprising amino acid sequences that are at least 99% identical to the amino acid sequences set forth, respectively: SEQ ID NOs: 51 and 56; and the second antigen binding domain comprises one EC domain of CD4 comprising an amino acid sequence that is at least 99%, identical to a CD4 EC domain of SEQ ID NO:746. In some embodiments, the first antigen binding domain is a Fab that binds to CD3 and the second antigen binding domain is an EC domain of CD4 that binds to HIV gp120, wherein the first antigen binding domain comprises a first HC and a LC, and the second antigen binding domain comprises a second HC, the second HC, the first HC and the LC comprising the amino acid sequences set forth, respectively, or comprising amino acid sequences that are at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% identical to the amino acid sequences set forth, respectively: SEQ ID NOs: 751, 752 and 753; or SEQ ID NOs: 754, 752 and 753In some embodiments, the first antigen binding domain is a Fab that binds to CD3 and the second antigen binding domain is an EC domain of CD4 that binds to HIV gp120, wherein the first antigen binding domain comprises a first HC and a LC, and the second antigen binding domain comprises a second HC, the second HC, the first HC and the LC comprising the amino acid sequences set forth, respectively, or comprising amino acid sequences that are at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% identical to the amino acid sequences set forth, respectively: SEQ ID NOs: 751, 752 and 753.

[0016] With respect to further embodiments of the multi-specific (e.g., bispecific) antigen binding molecules, in some embodiments, the multi-specific antigen binding molecule is a bispecific antigen binding molecule. In some embodiments, the multi-specific antigen binding molecule binds to or targets human CD3 and HIV gp120. In some embodiments, the first VH and the first VL have at least 80%, 81%, 82%, 83%, 84%, 85%, or more, sequence similarity to a human germline VH and a human germline VL, respectively. In some embodiments, the first antigen binding domain has reduced or insignificant or substantially no binding to Protein A, or does not detectably bind to Protein A. In some embodiments, the first antigen binding domain binds to Protein A with a binding equilibrium dissociation constant (K D ) of greater than 10 -6< M. The first antigen binding domain binds to CD3 with a K D of lower than 3.0 nM (e.g., 2.5 nM). In some embodiments, the multi-specific (e.g., bispecific) antigen binding molecule has a serum half-life in a human or cynomolgus monkey of at least 3 days, e.g., at least 4 days, at least 5 days, at least 6 days, at least 7 days, at least 8 days, at least 9 days, at least 10 days, at least 11 days, at least 12 days, at least 13 days, at least 14 days, at least 15 days, at least 16 days, or longer. In some embodiments, the multi-specific (e.g., bispecific) antigen binding molecule has a serum half-life in a human or cynomolgus monkey of at least 7 days. In some embodiments, the multi-specific (e.g., bispecific) antigen binding molecule has a serum half-life in a human of at least 7 days. In some embodiments, the first antigen binding domain binds to CD3 with a K D of lower than 3.0 nM (e.g., 2.5 nM), and the multi-specific (e.g., bispecific) antigen binding molecule has a serum half-life in a human or cynomolgus monkey of at least 7 days. In some embodiments, the first antigen binding domain binds to CD3 with a K D of lower than 3.0 nM (e.g., 2.5 nM), and the multi-specific (e.g., bispecific) antigen binding molecule has a serum half-life in a human of at least 7 days. In some embodiments, at least 50%, at least 60%, at least 70%, least 80%, at least 85%, at least 90%, or more, N-linked glycosylation sites in at least one of the first VH, the first VL, the second VH and the second VL are sialylated. In some embodiments, the N-linked glycosylation sites in at least one of the first VH, the first VL, the second VH and the second VL have a sialic acid occupancy (e.g., a glycan comprising one or two terminal sialic acid residues) of at least 40%, at least 50%, at least 60%, at least 70%, least 80%, at least 85%, at least 90%, or more. In some embodiments, the sialylated N-linked glycosylation sites in at least one of the first VH, the first VL, the second VH and the second VL comprise from 1 to 5 sialic acid residues, e.g., from 1 to 4 sialic acid residues, e.g., from 1 to 3 sialic acid residues, e.g., from 1 to 2 sialic acid residues. In some embodiments, at least one of the first VH, the first VL, the second VH and the second VL are sialylated with N-acetylneuraminic acid (NANA). In some embodiments, the sialic acid residues are present in biantennary structures. In some embodiments, the sialic acid residues are present in complex N-linked glycan structures. In some embodiments, the sialic acid residues are present in hybrid N-linked glycan structures. In some embodiments, the glycans are terminally sialylated.

[0017] In another aspect, provided is a polynucleotide or multiple polynucleotides encoding the multispecific antigen binding molecule of the invention. The polynucleotide or multiple polynucleotides may encode at least the first VH and the first VL of the multi-specific (e.g., bispecific) antigen binding molecule, as described herein. In some embodiments, the polynucleotide or polynucleotides comprise a polynucleotide or multiple polynucleotides encoding the HC and LC of the first antigen binding domain that is a Fab, and the HC of the second antigen binding domain that is an EC domain of CD4, of the multi-specific (e.g., bispecific) antigen binding molecule. In some embodiments, the polynucleotide or polynucleotides encode a multi-specific antigen binding molecule, as described herein, comprising the following polynucleotide sequences, or polynucleotide sequences that are at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% identical to the polynucleotide sequences set forth, respectively: SEQ ID NOs: 995, 996 and 997; or SEQ ID NOs: 998, 999 and 1000. In some embodiments, the polynucleotide or polynucleotides encode a multi-specific antigen binding molecule, as described herein, comprising polynucleotide sequences that are at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% identical to the polynucleotide sequences set forth, respectively: SEQ ID NOs: 995, 996 and 997. In some embodiments, the polynucleotide or polynucleotides encode a multi-specific antigen binding molecule, as described herein, comprising polynucleotide sequences that are at least 95% (e.g., at least 96%, at least 97%, at least 98%, or at least 99%) identical to the polynucleotide sequences set forth: SEQ ID NOs: 995, 996 and 997. In some embodiments, the polynucleotide or polynucleotides encode a multi-specific antigen binding molecule, as described herein, comprising the following polynucleotide sequences: SEQ ID NOs: 995, 996 and 997. In some embodiments, the polynucleotide or polynucleotides encode a multi-specific antigen binding molecule, as described herein, comprising polynucleotide sequences that are at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% identical to the polynucleotide sequences set forth, respectively: SEQ ID NOs: 998, 999 and 1000. In some embodiments, the polynucleotide or polynucleotides encode a multi-specific antigen binding molecule, as described herein, comprising polynucleotide sequences that are at least 95% (e.g., at least 96%, at least 97%, at least 98%, or at least 99%) identical to the polynucleotide sequences set forth: SEQ ID NOs: 998, 999 and 1000. In some embodiments, the polynucleotide or polynucleotides encode a multi-specific antigen binding molecule, as described herein, comprising the following polynucleotide sequences: SEQ ID NOs: 998, 999 and 1000. In some embodiments, the polynucleotide or polynucleotides are comprised of DNA or RNA. In some embodiments, the polynucleotide or polynucleotides are comprised of mRNA. In some embodiments, the polynucleotide or polynucleotides comprise codon bias for efficient expression in a human cell. Also disclosed is a lipoplex, e.g., a lipid nanoparticle (LNP), comprising the polynucleotide or polynucleotides, as described herein. In a further aspect, provided is an expression cassette or multiple expression cassettes comprising one or more regulatory sequences operably linked to the polynucleotide or polynucleotides of the invention.

[0018] In a further aspect, provided is an expression vector or multiple expression vectors comprising one or more regulatory sequences operably linked to the polynucleotide or polynucleotides, or the expression cassette or expression cassettes, of the invention. In some embodiments, the expression vector or expression vectors comprise a plasmid vector or a viral vector. In some embodiments, the expression vector comprises three, four or five expression cassettes or cistrons. In some embodiments, the expression vector comprises, optionally in sequential order from 5' to 3': (i) a first expression cassette or cistron comprising a first polynucleotide encoding an anti-CD3 VL-CL fusion protein; (ii) a second expression cassette or cistron comprising a second polynucleotide encoding an anti-CD3 VH-Fc fusion protein; and (iii) a third expression cassette or cistron comprising a third polynucleotide encoding a CD4 extracellular (EC) domain-Fc fusion protein. In some embodiments, the anti-CD3 VL-CL fusion protein, the anti-CD3 VH-Fc fusion protein and the CD4 EC domain-Fc fusion protein comprise amino acid sequences set forth, respectively, below, or comprise amino acid sequences that are at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% identical to the amino acid sequences set forth, respectively, below: SEQ ID NOs: 753, 752 and 751; or SEQ ID NOs: 753, 752 and 754. In some embodiments, the anti-CD3 VL-CL fusion protein, the anti-CD3 VH-Fc fusion protein and the CD4 EC domain-Fc fusion protein comprise amino acid sequences that are at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% identical to the amino acid sequences set forth, respectively: SEQ ID NOs: 753, 752 and 751. In some embodiments, the anti-CD3 VL-CL fusion protein, the anti-CD3 VH-Fc fusion protein and the CD4 EC domain-Fc fusion protein comprise amino acid sequences set forth, respectively: SEQ ID NOs: 753, 752 and 751. In some embodiments, the first, second and third expression cassettes or cistrons each comprise a promoter of identical or equivalent transcription strength, e.g., a constitutive promoter, e.g., a promoter selected from cytomegalovirus (CMV), SV40, RSV, EF1a, UBC, PGK and CAGG. In some embodiments, the first, second and third expression cassettes or cistrons comprise one or more promoters of different transcription strength. In embodiments, the expression vector further comprises a fourth expression cassette or cistron positioned 5' to the first expression cassette or cistron comprising a polynucleotide encoding a eukaryotic selection marker protein, e.g., glutamine synthetase (GS). Generally, the fourth expression cassette or cistron, positioned 5' to the first expression cassette and comprising a polynucleotide encoding a eukaryotic selection marker protein, is translated from the same strand as the first, second and third expression cassettes or cistrons.

[0019] In a further aspect, provided is a cell or population of cells comprising the polynucleotide or polynucleotides of the invention, the expression cassette or multiple expression cassettes of the invention, or the expression vector or expression vectors of the invention. In some embodiments, the cell or population of cells comprises a eukaryotic cell. In some embodiments, the cell or population of cells comprises a mammalian cell, an insect cell, a plant cell or a yeast cell. In some embodiments, the mammalian cell is a Chinese Hamster Ovary (CHO) cell. In some embodiments, the mammalian cell is a human cell. In some embodiments, the cell is a human embryonic kidney cell. In some embodiments, the cell predominantly sialylates N-linked glycosylation sites in the variable domains (Fv) of expressed antigen binding molecules. In some embodiments, at least 50%, at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, or more, N-linked glycosylation sites in the variable domains (Fv) of expressed antigen binding molecules are sialylated.

[0020] In a further aspect, provided is a pharmaceutical composition comprising one or more multi-specific antigen binding molecules of the invention, and a pharmaceutically acceptable carrier. Also disclosed is a pharmaceutical composition comprising one or more polynucleotides, described herein, encoding one or more multi-specific antigen binding molecule, described herein, or the lipoplex (e.g., LNP) described herein, and a pharmaceutically acceptable carrier. In some embodiments, the pharmaceutical composition comprises an aqueous formulation. In some embodiments, the pharmaceutical composition comprises one or more multi-specific antigen binding molecules, described herein, at a concentration of from 0.1 mg / ml to 150 mg / ml, e.g., from 0.1 mg / ml to 100 mg / ml, e.g., from 1 mg / ml to 100 mg / ml, e.g., from 5 mg / ml to 60 mg / ml, e.g., from 20 mg / ml to 150 mg / ml, or from 10 mg / ml to 50 mg / ml. In some embodiments, the pharmaceutical composition is lyophilized. In some embodiments, the pharmaceutical composition is formulated for intravenous, intramuscular or subcutaneous administration. In some embodiments, the pharmaceutical composition further comprises a second agent for treating an HIV infection. In some embodiments, the pharmaceutical composition further comprises a toll-like receptor (TLR) agonist or an IL-15 receptor agonist. In some embodiments, the pharmaceutical composition comprises a multi-specific antigen binding molecule having a first antigen binding domain that binds to CD3, and a second antigen binding domain that binds to an epitope or region of gp120 in the CD4 binding site (CD4bs) and comprises one or more extracellular (EC) domains of CD4, and optionally, an IL-15 receptor agonist. In some embodiments, the EC domain of CD4 comprises a sequence that is at least 99% identical to the sequence of SEQ ID NO: 746. In some embodiments, the EC domain of CD4 comprises the sequence of SEQ ID NO: 746. In some embodiments, the first antigen binding domain comprises a first VH-CDR1, a first VH-CDR2, a first VH-CDR3, a first VL-CDR1, a first VL-CDR2 and a first VL-CDR3 comprising the following amino acid sequences, respectively (according to Kabat): SEQ ID NOs: 1, 12, 8, 4, 9 and 10; and the second antigen binding domain comprises one EC domain of CD4 comprising an amino acid sequence that is at least 99% identical to a CD4 EC domain of SEQ ID NO: 746. In some embodiments, the first antigen binding domain comprises a first VH-CDR1, a first VH-CDR2, a first VH-CDR3, a first VL-CDR1, a first VL-CDR2 and a first VL-CDR3 comprising the following amino acid sequences, respectively (according to Kabat): SEQ ID NOs: 1, 12, 8, 4, 9 and 10; and the second antigen binding domain comprises one EC domain of CD4 comprising the amino acid sequence of SEQ ID NO: 746. In some embodiments, the TLR agonist is a TLR2 agonist, a TLR3 agonist, a TLR7 agonist, a TLR8 agonist or a TLR9 agonist. In some embodiments, the pharmaceutical composition further comprises the TLR7 agonist is selected from the group consisting of vesatolimod, imiquimod, and resiquimod. In some embodiments, the pharmaceutical composition comprises a first multi-specific antigen binding molecule and second or additional antigen binding molecules, wherein the first multi-specific antigen binding molecule and the second or additional antigen binding molecules bind to different epitopes or regions of gp120 selected from the group consisting of: (i) the third variable loop (V3) (e.g., high mannose patch) comprising a N332 oligomannose glycan; (ii) the second variable loop (V2) (e.g., Env trimer apex); (iii) the CD4 binding site (CD4bs); (iv) the gp120 / gp41 interface; or (v) the silent face of gp120. In some embodiments, the first multi-specific antigen binding molecule binds to the third variable loop (V3) (e.g., high mannose patch) comprising a N332 oligomannose glycan and the second or additional antigen binding molecules bind to the CD4 binding site (CD4bs). In some embodiments, the first multi-specific antigen binding molecule competes with or comprises VH and VL regions from an antibody selected from the group consisting of GS-9722 (elipovimab), GS-2872, PGT-121, PGT-121.66, PGT-121.414, PGT-122, PGT-123, PGT-124, PGT-125, PGT-126, PGT-128, PGT-130, PGT-133, PGT-134, PGT-135, PGT-136, PGT-137, PGT-138, PGT-139, 10-1074, 10-1074-J, VRC24, 2G12, BG18, 354BG8, 354BG18, 354BG42, 354BG33, 354BG129, 354BG188, 354BG411, 354BG426, DH270.1, DH270.6, PGDM12, VRC41.01, PGDM21, PCDN-33A, BF520.1 and VRC29.03, and the second or additional antigen binding molecules compete with or comprise VH and VL regions from an antibody selected from the group consisting of GS-9723, GS-5423, 3BNC117, 3BNC60, b12, F105, VRC01, VRC07, VRC07-523, VRC03, VRC06, VRC06b01 VRC08, VRC0801, NIH45-46, PGV04 (a.k.a., VRC-PG04); CH103, 44-VRC13.01, 1NC9, 12A12, N6, 1-18, N49-P7, NC-Cow1, IOMA, CH235 and CH235.12, N49P6, N49P7, N49P11, N49P9 and N60P25. In some embodiments, the first multi-specific antigen binding molecule competes with or comprises VH and VL regions from an antibody selected from the group consisting of 10-1074, 10-1074-J, GS-9722 (elipovimab), GS-2872, PGT-121, PGT-121.66, PGT-121.414 and PGT-134, and the second or additional antigen binding molecules compete with or comprise VH and VL regions from an antibody selected from the group consisting of GS-9723, GS-5423, 3BNC117, VRC07 and VRC07-523. In some embodiments, the first multi-specific antigen binding molecule competes with or comprises VH and VL regions from an antibody selected from the group consisting of 10-1074, 10-1074-J, GS-9722 (elipovimab), GS-2872, PGT-121, PGT-121.66, PGT-121.414 and PGT-134, and the second or additional antigen binding molecules comprise an EC domain of CD4. In some embodiments, the first multi-specific antigen binding molecule binds to the CD4 binding site (CD4bs) and the second or additional antigen binding molecules bind to the third variable loop (V3) (e.g., high mannose patch) comprising a N332 oligomannose glycan. In some embodiments, the first multi-specific antigen binding molecule competes with or comprises VH and VL regions from an antibody selected from the group consisting of GS-9723, GS-5423, 3BNC117, 3BNC60, b12, F105, VRC01, VRC07, VRC07-523, VRC03, VRC06, VRC06b01 VRC08, VRC0801, NIH45-46, PGV04 (a.k.a., VRC-PG04); CH103, 44-VRC13.01, 1NC9, 12A12, N6, 1-18, N49-P7, NC-Cow1, IOMA, CH235 and CH235.12, N49P6, N49P7, N49P11, N49P9 and N60P25, and the second or additional antigen binding molecules compete with or comprise VH and VL regions from an antibody selected from the group consisting of GS-9722 (elipovimab), GS-2872, PGT-121, PGT-121.66, PGT-121.414, PGT-122, PGT-123, PGT-124, PGT-125, PGT-126, PGT-128, PGT-130, PGT-133, PGT-134, PGT-135, PGT-136, PGT-137, PGT-138, PGT-139, 10-1074, 10-1074-J, VRC24, 2G12, BG18, 354BG8, 354BG18, 354BG42, 354BG33, 354BG129, 354BG188, 354BG411, 354BG426, DH270.1, DH270.6, PGDM12, VRC41.01, PGDM21, PCDN-33A, BF520.1 and VRC29.03. In some embodiments, the first multi-specific antigen binding molecule competes with or comprises VH and VL regions from an antibody selected from the group consisting of GS-9723, GS-5423, 3BNC117, VRC07 and VRC07-523, and the second or additional antigen binding molecules compete with or comprise VH and VL regions from an antibody selected from the group consisting of 10-1074, 10-1074-J, GS-9722 (elipovimab), GS-2872, PGT-121, PGT-121.66, PGT-121.414 and PGT-134. In some embodiments, the first multi-specific antigen binding molecule competes with or comprises an EC domain of CD4, and the second or additional antigen binding molecules compete with or comprise VH and VL regions from an antibody selected from the group consisting of GS-9722 (elipovimab), GS-2872, PGT-121, PGT-121.66, PGT-121.414, PGT-122, PGT-123, PGT-124, PGT-125, PGT-126, PGT-128, PGT-130, PGT-133, PGT-134, PGT-135, PGT-136, PGT-137, PGT-138, PGT-139, 10-1074, 10-1074-J, VRC24, 2G12, BG18, 354BG8, 354BG18, 354BG42, 354BG33, 354BG129, 354BG188, 354BG411, 354BG426, DH270.1, DH270.6, PGDM12, VRC41.01, PGDM21, PCDN-33A, BF520.1 and VRC29.03. In some embodiments, the first multi-specific antigen binding molecule competes with or comprises an EC domain of CD4, and the second or additional antigen binding molecules compete with or comprise VH and VL regions from an antibody selected from the group consisting of 10-1074, 10-1074-J, GS-9722 (elipovimab), GS-2872, PGT-121, PGT-121.66, PGT-121.414 and PGT-134. In some embodiments, the pharmaceutical composition further comprises additional antigen binding molecules that compete with or comprise VH and VL regions from an antibody selected from the group consisting of GS-9723, GS-5423, 3BNC117, 3BNC60, b12, F105, VRC01, VRC07, VRC07-523, VRC03, VRC06, VRC06b01 VRC08, VRC0801, NIH45-46, PGV04 (a.k.a., VRC-PG04); CH103, 44-VRC13.01, 1NC9, 12A12, N6, 1-18, N49-P7, NC-Cow1, IOMA, CH235 and CH235.12, N49P6, N49P7, N49P11, N49P9 and N60P25. In some embodiments, the pharmaceutical composition further comprises additional antigen binding molecules that compete with or comprise VH and VL regions from an antibody selected from the group consisting of GS-9723, GS-5423, 3BNC117, VRC07 and VRC07-523. In some embodiments, the pharmaceutical composition comprises: (i) a multi-specific (e.g. bispecific) antigen binding molecule comprising a EC domain of CD4, as described herein; (ii) an antibody or multi-specific antigen binding molecule that competes with or comprises VH and VL regions from an antibody selected from the group consisting of 10-1074, 10-1074-J, GS-9722 (elipovimab), GS-2872, PGT-121, PGT-121.66, PGT-121.414 and PGT-134; and (iii) an antibody or multi-specific antigen binding molecule that competes with or comprises VH and VL regions from an antibody selected from the group consisting of GS-9723, GS-5423, 3BNC117, VRC07 and VRC07-523. In some embodiments, the pharmaceutical composition comprises an additional multi-specific antigen binding molecule that competes with or comprises VH and VL regions from an antibody that binds to the gp120 second variable loop (V2) (e.g., Env trimer apex). In some embodiments, the pharmaceutical composition comprises an additional multi-specific antigen binding molecule that competes with or comprises VH and VL regions from an antibody selected from the group consisting of PG9, PG16, PGC14, PGG14, PGT-142, PGT-143, PGT-144, PGT-145, CH01, CH59, PGDM1400, CAP256, CAP256-VRC26.08, CAP256-VRC26.09, CAP256-VRC26.25, PCT64-24E and VRC38.01. In some embodiments, the pharmaceutical composition comprises an additional multi-specific antigen binding molecule that competes with or comprises VH and VL regions from an antibody that binds to the gp120 / gp41 interface. In some embodiments, the pharmaceutical composition comprises an additional multi-specific antigen binding molecule that competes with or comprises VH and VL regions from an antibody selected from the group consisting of PGT-151, CAP248-2B, 35O22, 8ANC195, ACS202, VRC34 and VRC34.01. In some embodiments, the pharmaceutical composition comprises an additional multi-specific antigen binding molecule that competes with or comprises VH and VL regions from an antibody that binds to an epitope or region of gp41 in the membrane proximal region (MPER). In some embodiments, the pharmaceutical composition comprises an additional multi-specific antigen binding molecule that competes with or comprises VH and VL regions from an antibody selected from the group consisting of 10E8, 10E8v4, 10E8-5R-100cF, 4E10, DH511.11P, 2F5, 7b2, and LN01. In some embodiments, the pharmaceutical composition comprises an additional antigen binding molecule or antigen binding fragment thereof that at least one of binds, inhibits, and neutralizes HIV, or a polynucleotide encoding the additional antigen binding molecule or antigen binding fragment thereof, wherein the additional antigen binding molecule or antigen binding fragment does not compete for binding to gp120 with the one or more multi-specific antigen binding molecules.

[0021] Also provided is a kit comprising one or more containers comprising one or more of the multi-specific antigen binding molecules, the polynucleotide or polynucleotides, the expression vector or expression vectors, or the pharmaceutical composition of the invention, wherein the kit comprises: (a) one or more unitary doses of the one or more multi-specific antigen binding molecules, or the polynucleotide or polynucleotides, in one or more containers; (b) one or more unitary doses of the one or more multi-specific antigen binding molecules and a second agent for treating an HIV infection in separate containers; and / or (c) wherein the kit further comprises a multi-specific antigen binding molecule comprising an EC domain of CD4 of the invention. Also disclosed is a kit comprising one or more containers comprising one or more of the multi-specific (e.g., bispecific) antigen binding molecules, the polynucleotide or polynucleotides, the lipoplex (e.g., LNP), or the pharmaceutical composition, as described herein. In some embodiments, the kit comprises one or more unitary doses of the one or more multi-specific antigen binding molecules, or the polynucleotide or polynucleotides, in one or more containers (e.g., one or more vials, ampules, syringes). In some embodiments, the kit comprises one or more unitary doses of the one or more multi-specific antigen binding molecules and a second agent for treating an HIV infection in separate containers. In some embodiments, the kit further comprises at least one of a toll-like receptor (TLR) agonist and an IL-15 receptor agonist. In some embodiments, the kit comprises a multi-specific antigen binding molecule having a first antigen binding domain that binds to CD3, and a second antigen binding domain that binds to an epitope or region of gp120 in the CD4 binding site (CD4bs) and comprises one or more extracellular (EC) domains of CD4, and optionally, an IL-15 receptor agonist. In some embodiments, the EC domain of CD4 comprises a sequence that is at least 99% identical to the sequence of SEQ ID NO: 746. In some embodiments, the EC domain of CD4 comprises the sequence of SEQ ID NO: 746. In some embodiments, the TLR agonist is a TLR2 agonist, a TLR3 agonist, a TLR7 agonist, a TLR8 agonist or a TLR9 agonist. In some embodiments, the TLR7 agonist is selected from the group consisting of vesatolimod, imiquimod, and resiquimod. In some embodiments, the kit comprises a first multi-specific antigen binding molecule and second or additional antigen binding molecules, wherein the first multi-specific antigen binding molecule and the second or additional antigen binding molecules bind to different epitopes or regions of gp120 selected from the group consisting of: (i) the third variable loop (V3) (e.g., high mannose patch) comprising a N332 oligomannose glycan; (ii) the second variable loop (V2) (e.g., Env trimer apex); (iii) the CD4 binding site (CD4bs); (iv) the gp120 / gp41 interface; or (v) the silent face of gp120. In some embodiments, the first multi-specific antigen binding molecule binds to the third variable loop (V3) (e.g., high mannose patch) comprising a N332 oligomannose glycan and the second or additional antigen binding molecules bind to the CD4 binding site (CD4bs). In some embodiments, the first multi-specific antigen binding molecule competes with or comprises VH and VL regions from an antibody selected from the group consisting of GS-9722 (elipovimab), GS-2872, PGT-121, PGT-121.66, PGT-121.414, PGT-122, PGT-123, PGT-124, PGT-125, PGT-126, PGT-128, PGT-130, PGT-133, PGT-134, PGT-135, PGT-136, PGT-137, PGT-138, PGT-139, 10-1074, VRC24, 2G12, BG18, 354BG8, 354BG18, 354BG42, 354BG33, 354BG129, 354BG188, 354BG411, 354BG426, DH270.1, DH270.6, PGDM12, VRC41.01, PGDM21, PCDN-33A, BF520.1 and VRC29.03, and the second or additional antigen binding molecules compete with or comprise VH and VL regions from an antibody selected from the group consisting of GS-9723, GS-5423, 3BNC117, 3BNC60, b12, F105, VRC01, VRC07, VRC07-523, VRC03, VRC06, VRC06b01 VRC08, VRC0801, NIH45-46, PGV04 (a.k.a., VRC-PG04); CH103, 44-VRC13.01, 1NC9, 12A12, N6, 1-18, N49-P7, NC-Cow1, IOMA, CH235 and CH235.12, N49P6, N49P7, N49P11, N49P9 and N60P25. In some embodiments, the first multi-specific antigen binding molecule competes with or comprises VH and VL regions from an antibody selected from the group consisting of 10-1074, 10-1074-J, GS-9722 (elipovimab), GS-2872, PGT-121, PGT-121.66, PGT-121.414 and PGT-134, and the second or additional antigen binding molecules compete with or comprise VH and VL regions from an antibody selected from the group consisting of GS-9723, GS-5423, 3BNC117, VRC07 and VRC07-523. In some embodiments, the first multi-specific antigen binding molecule competes with or comprises VH and VL regions from an antibody selected from the group consisting of 10-1074, 10-1074-J, GS-9722 and PGT-121, and the second or additional antigen binding molecules comprise an EC domain of CD4. In some embodiments, the first multi-specific antigen binding molecule binds to the CD4 binding site (CD4bs) and the second or additional antigen binding molecules bind to the third variable loop (V3) (e.g., high mannose patch) comprising a N332 oligomannose glycan. In some embodiments, the first multi-specific antigen binding molecule competes with or comprises VH and VL regions from an antibody selected from the group consisting of GS-9723, GS-5423, 3BNC117, 3BNC60, b12, F105, VRC01, VRC07, VRC07-523, VRC03, VRC06, VRC06b01 VRC08, VRC0801, NIH45-46, PGV04 (a.k.a., VRC-PG04); CH103, 44-VRC13.01, 1NC9, 12A12, N6, 1-18, N49-P7, NC-Cow1, IOMA, CH235 and CH235.12, N49P6, N49P7, N49P11, N49P9 and N60P25, and the second or additional antigen binding molecules compete with or comprise VH and VL regions from an antibody selected from the group consisting of GS-9722 (elipovimab), GS-2872, PGT-121, PGT-121.66, PGT-121.414, PGT-122, PGT-123, PGT-124, PGT-125, PGT-126, PGT-128, PGT-130, PGT-133, PGT-134, PGT-135, PGT-136, PGT-137, PGT-138, PGT-139, 10-1074, VRC24, 2G12, BG18, 354BG8, 354BG18, 354BG42, 354BG33, 354BG129, 354BG188, 354BG411, 354BG426, DH270.1, DH270.6, PGDM12, VRC41.01, PGDM21, PCDN-33A, BF520.1 and VRC29.03. In some embodiments, the first multi-specific antigen binding molecule competes with or comprises VH and VL regions from an antibody selected from the group consisting of GS-9723, GS-5423, 3BNC117, VRC07 and VRC07-523, and the second or additional antigen binding molecules compete with or comprise VH and VL regions from an antibody selected from the group consisting of 10-1074, 10-1074-J, GS-9722 (elipovimab), GS-2872, PGT-121, PGT-121.66, PGT-121.414 and PGT-134. In some embodiments, the first multi-specific antigen binding molecule competes with or comprises an EC domain of CD4, and the second or additional antigen binding molecules compete with or comprise VH and VL regions from an antibody selected from the group consisting of GS-9722 (elipovimab), GS-2872, PGT-121, PGT-121.66, PGT-121.414, PGT-122, PGT-123, PGT-124, PGT-125, PGT-126, PGT-128, PGT-130, PGT-133, PGT-134, PGT-135, PGT-136, PGT-137, PGT-138, PGT-139, 10-1074, VRC24, 2G12, BG18, 354BG8, 354BG18, 354BG42, 354BG33, 354BG129, 354BG188, 354BG411, 354BG426, DH270.1, DH270.6, PGDM12, VRC41.01, PGDM21, PCDN-33A, BF520.1 and VRC29.03. In some embodiments, the first multi-specific antigen binding molecule competes with or comprises an EC domain of CD4, and the second or additional antigen binding molecules compete with or comprise VH and VL regions from an antibody selected from the group consisting of 10-1074, 10-1074-J, GS-9722 (elipovimab), GS-2872, PGT-121, PGT-121.66, PGT-121.414 and PGT-134. In some embodiments, the kits further comprise second or additional antigen binding molecules compete with or comprise VH and VL regions from an antibody selected from the group consisting of GS-9723, GS-5423, 3BNC117, 3BNC60, b12, F105, VRC01, VRC07, VRC07-523, VRC03, VRC06, VRC06b01 VRC08, VRC0801, NIH45-46, PGV04 (a.k.a., VRC-PG04); CH103, 44-VRC13.01, 1NC9, 12A12, N6, 1-18, N49-P7, NC-Cow1, IOMA, CH235 and CH235.12, N49P6, N49P7, N49P11, N49P9 and N60P25. In some embodiments, the kits further comprise second or additional antigen binding molecules compete with or comprise VH and VL regions from an antibody selected from the group consisting of GS-9723, GS-5423, 3BNC117, VRC07 and VRC07-523. In some embodiments, the kit comprises: (i) a multi-specific antigen binding molecule comprising a EC domain of CD4, described herein; (ii) an antibody or multi-specific antigen binding molecule that competes with or comprises VH and VL regions from an antibody selected from the group consisting of 10-1074, 10-1074-J, GS-9722 (elipovimab), GS-2872, PGT-121, PGT-121.66, PGT-121.414 and PGT-134; and (iii) an antibody or multi-specific antigen binding molecule that competes with or comprises VH and VL regions from an antibody selected from the group consisting of GS-9723, GS-5423, 3BNC117, VRC07 and VRC07-523. In some embodiments, the kit comprises two or more unitary doses, wherein the unitary doses are the same or different.

[0022] In a further aspect, provided are methods of producing a multi-specific (e.g., bispecific) antigen binding molecule, of the invention, wherein the methods comprises: (a) culturing a cell or population of cells, of the invention, transformed with the polynucleotide or polynucleotides, of the invention, or the expression cassette or multiple expression cassettes, of the invention, in a cell culture under conditions sufficient to express the multi-specific antigen binding molecules; and (b) isolating or purifying the antigen binding molecules from the cell culture. In some embodiments, the first antigen binding domain is a Fab and the second antigen binding domain is an EC domain of CD4. In some embodiments, the polypeptide comprising the first antigen binding domain and the polypeptide comprising the second antigen binding domain are expressed and assembled in the same cell. In some embodiments, the isolating or purifying step comprises Protein A affinity chromatography. In some embodiments, at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or more, of the multi-specific antigen binding molecules are isolated or purified. In some embodiments, the isolating or purifying step further comprises ion exchange chromatography. In some embodiments, at least 95%, 96%, 97%, 98%, 99%, or more, of the multi-specific antigen binding molecules are isolated or purified. In some embodiments, at least 95%, 96%, 97%, 98%, 99%, or more, of the multi-specific antigen binding molecules isolate or purify as non-aggregated soluble heterodimer as determined using size exclusion chromatography (SEC). In some embodiments, the isolated or purified multi-specific antigen binding molecules have increased homogeneity as assessed by analytical ion exchange chromatography, wherein the integrated area of a main peak representing an unmodified target species is at least 95%, 96%, 97%, 98%, or more, of the sum of all integrated protein peak areas. In some embodiments, the isolated or purified antigen binding molecules have fewer than 35%, 30%, 25%, 20%, 15%, 10%, 9%, 8%, 7%, or fewer, acidic contaminants. In some embodiments, the cell or population of cells are cultured in a culture volume of at least 2L, e.g., at least 5L, 10L, 50L, 100L, 150L, 200L, 250L, or more. In some embodiments, the methods further comprise formulating the antigen binding molecule into a sterile pharmaceutical composition suitable for administration to a human subject.

[0023] In another aspect, provided are one or more multi-specific antigen binding molecules of the invention, one or more polynucleotides of the invention, or the pharmaceutical composition of the invention for use in method of treating or preventing HIV in a human subject in need thereof. In some embodiments, the methods comprise administering to the subject an effective amount of one or more multi-specific (e.g., bispecific) antigen binding molecules, described herein, or the pharmaceutical composition, described herein. In a related aspect, provided are the one or more of the multi-specific antigen binding molecules, or a pharmaceutical composition thereof, for use in methods of preventing or treating an HIV infection or an HIV-related disease. In some embodiments, the methods comprises the steps of: identifying a patient in need of such prevention or treatment, and administering to said patient a first therapeutic agent comprising a therapeutically effective amount of at least one multi-specific (e.g., bispecific) antigen binding molecules, of the invention, or the pharmaceutical composition, of the invention. In some embodiments, the methods further comprise administering to the subject a second agent for treating an HIV infection. In some embodiments, the subject is not receiving antiretroviral therapy (ART) or ART is discontinued prior to administration of the one or more multi-specific antigen binding molecules. In some embodiments, ART is discontinued after one or more administrations of the one or more multi-specific antigen binding molecules. In some embodiments, the methods further comprise administering one or more antiretroviral therapy (ART) agents to the subject. In some embodiments, the methods further comprise administering to the subject at least one of a TLR agonist and an IL-15 receptor agonist. In some embodiments, the method comprises administering a multi-specific antigen binding molecule having a first antigen binding domain that binds to CD3, and a second antigen binding domain that binds to an epitope or region of gp120 in the CD4 binding site (CD4bs) and comprises one or more extracellular (EC) domains of CD4, and optionally, an IL-15 receptor agonist. In some embodiments, the EC domain of CD4 comprises a sequence that is at least 99% identical to the sequence of SEQ ID NO: 746. In some embodiments, the EC domain of CD4 comprises the sequence of SEQ ID NO: 746. In some embodiments, the TLR agonist is a TLR2 agonist, a TLR3 agonist, a TLR7 agonist, a TLR8 agonist or a TLR9 agonist. In some embodiments, the TLR7 agonist is selected from the group consisting of vesatolimod, imiquimod, and resiquimod. In some embodiments, the method comprises administering a first multi-specific antigen binding molecule and second or additional antigen binding molecules, wherein the first multi-specific antigen binding molecule and the second or additional antigen binding molecules bind to different epitopes or regions of gp120 selected from the group consisting of: (i) the third variable loop (V3) (e.g., high mannose patch) comprising a N332 oligomannose glycan; (ii) the second variable loop (V2) (e.g., Env trimer apex); (iii) the CD4 binding site (CD4bs); (iv) the gp120 / gp41 interface; or (v) the silent face of gp120. In some embodiments, the first multi-specific antigen binding molecule binds to the third variable loop (V3) (e.g., high mannose patch) comprising a N332 oligomannose glycan and the second or additional antigen binding molecules bind to the CD4 binding site (CD4bs). In some embodiments, the first multi-specific antigen binding molecule competes with or comprises VH and VL regions from an antibody selected from the group consisting of 10-1074, 10-1074-J, GS-9722 (elipovimab), GS-2872, PGT-121, PGT-121.66, PGT-121.414, PGT-122, PGT-123, PGT-124, PGT-125, PGT-126, PGT-128, PGT-130, PGT-133, PGT-134, PGT-135, PGT-136, PGT-137, PGT-138, PGT-139, VRC24, 2G12, BG18, 354BG8, 354BG18, 354BG42, 354BG33, 354BG129, 354BG188, 354BG411, 354BG426, DH270.1, DH270.6, PGDM12, VRC41.01, PGDM21, PCDN-33A, BF520.1 and VRC29.03, and the second or additional antigen binding molecules compete with or comprise VH and VL regions from an antibody selected from the group consisting of GS-9723, GS-5423, 3BNC117, 3BNC60, b12, F105, VRC01, VRC07, VRC07-523, VRC03, VRC06, VRC06b01 VRC08, VRC0801, NIH45-46, PGV04 (a.k.a., VRC-PG04); CH103, 44-VRC13.01, 1NC9, 12A12, N6, 1-18, N49-P7, NC-Cowl, IOMA, CH235 and CH235.12, N49P6, N49P7, N49P11, N49P9 and N60P25. In some embodiments, the first multi-specific antigen binding molecule competes with or comprises VH and VL regions from an antibody selected from the group consisting of 10-1074, 10-1074-J, GS-9722 (elipovimab), GS-2872, PGT-121, PGT-121.66, PGT-121.414 and PGT-134, and the second or additional antigen binding molecules compete with or comprise VH and VL regions from GS-9723, GS-5423, 3BNC117, VRC07 and VRC07-523. In some embodiments, the first multi-specific antigen binding molecule competes with or comprises VH and VL regions from an antibody selected from the group consisting of 10-1074, 10-1074-J, GS-9722 (elipovimab), GS-2872, PGT-121, PGT-121.66, PGT-121.414 and PGT-134, and the second or additional antigen binding molecules comprise an EC domain of CD4. In some embodiments, the first multi-specific antigen binding molecule binds to the CD4 binding site (CD4bs) and the second or additional antigen binding molecules bind to the third variable loop (V3) (e.g., high mannose patch) comprising a N332 oligomannose glycan. In some embodiments, the first multi-specific antigen binding molecule competes with or comprises VH and VL regions from an antibody selected from the group consisting of GS-9723, GS-5423, 3BNC117, 3BNC60, b12, F105, VRC01, VRC07, VRC07-523, VRC03, VRC06, VRC06b01 VRC08, VRC0801, NIH45-46, PGV04 (a.k.a., VRC-PG04); CH103, 44-VRC13.01, 1NC9, 12A12, N6, 1-18, N49-P7, NC-Cowl, IOMA, CH235 and CH235.12, N49P6, N49P7, N49P11, N49P9 and N60P25, and the second or additional antigen binding molecules compete with or comprise VH and VL regions from an antibody selected from the group consisting of GS-9722 (elipovimab), GS-2872, PGT-121, PGT-121.66, PGT-121.414, PGT-122, PGT-123, PGT-124, PGT-125, PGT-126, PGT-128, PGT-130, PGT-133, PGT-134, PGT-135, PGT-136, PGT-137, PGT-138, PGT-139, 10-1074, 10-1074-J, VRC24, 2G12, BG18, 354BG8, 354BG18, 354BG42, 354BG33, 354BG129, 354BG188, 354BG411, 354BG426, DH270.1, DH270.6, PGDM12, VRC41.01, PGDM21, PCDN-33A, BF520.1 and VRC29.03. In some embodiments, the first multi-specific antigen binding molecule competes with or comprises VH and VL regions from an antibody selected from the group consisting of GS-9723, GS-5423, 3BNC117, VRC07 and VRC07-523, and the second or additional antigen binding molecules compete with or comprise VH and VL regions from 10-1074, 10-1074-J, GS-9722 (elipovimab), GS-2872, PGT-121, PGT-121.66, PGT-121.414 and PGT-134. In some embodiments, the first multi-specific antigen binding molecule competes with or comprises an EC domain of CD4, and the second or additional antigen binding molecules compete with or comprise VH and VL regions from an antibody selected from the group consisting of GS-9722 (elipovimab), GS-2872, PGT-121, PGT-121.66, PGT-121.414, PGT-122, PGT-123, PGT-124, PGT-125, PGT-126, PGT-128, PGT-130, PGT-133, PGT-134, PGT-135, PGT-136, PGT-137, PGT-138, PGT-139, 10-1074, 10-1074-J, VRC24, 2G12, BG18, 354BG8, 354BG18, 354BG42, 354BG33, 354BG129, 354BG188, 354BG411, 354BG426, DH270.1, DH270.6, PGDM12, VRC41.01, PGDM21, PCDN-33A, BF520.1 and VRC29.03. In some embodiments, the first multi-specific antigen binding molecule competes with or comprises an EC domain of CD4, and the second or additional antigen binding molecules compete with or comprise VH and VL regions from an antibody selected from the group consisting of 10-1074, 10-1074-J, GS-9722 (elipovimab), GS-2872, PGT-121, PGT-121.66, PGT-121.414 and PGT-134. In some embodiments, the method further comprises administering second or additional antigen binding molecules compete with or comprise VH and VL regions from an antibody selected from the group consisting of GS-9723, GS-5423, 3BNC117, 3BNC60, b12, F105, VRC01, VRC07, VRC07-523, VRC03, VRC06, VRC06b01 VRC08, VRC0801, NIH45-46, PGV04 (a.k.a., VRC-PG04); CH103, 44-VRC13.01, 1NC9, 12A12, N6, 1-18, N49-P7, NC-Cow1, IOMA, CH235 and CH235.12, N49P6, N49P7, N49P11, N49P9 and N60P25. In some embodiments, the method further comprises administering second or additional antigen binding molecules compete with or comprise VH and VL regions from an antibody selected from the group consisting of GS-9723, GS-5423, 3BNC117, VRC07 and VRC07-523. In some embodiments, the methods comprise co-administering: (i) a multi-specific antigen binding molecule comprising a EC domain of CD4, described herein; (ii) an antibody or multi-specific antigen binding molecule that competes with or comprises VH and VL regions from an antibody selected from the group consisting of 10-1074, 10 1074-J, GS-9722 (elipovimab), GS-2872, PGT-121, PGT-121.66, PGT-121.414 and PGT-134; and (iii) an antibody or multi-specific antigen binding molecule that competes with or comprises VH and VL regions from an antibody selected from the group consisting of GS-9723, GS 5423, 3BNC117, VRC07 and VRC07-523. In some embodiments, the methods further comprise administering to the human subject an additional antigen binding molecule or antigen binding fragment thereof that at least one of binds, inhibits, and neutralizes HIV or a polynucleotide encoding the additional antigen binding molecule or antigen binding fragment thereof. In some embodiments, the method comprises administering an antibody or multi-specific antigen binding molecule that competes with or comprises VH and VL regions that bind to the gp120 third variable loop (V3) (e.g., high mannose patch) comprising a N332 oligomannose glycan, and the human subject is infected with an HIV expressing a gp120 comprising the following amino acid residues, wherein the positions and residues are with reference to SEQ ID NO:69: N332glycan, D325 and T63; N332glycan, D325 and L179; N332glycan, D325 and T320; N332glycan, D325 and H330; N332glycan, D325, T63 and L179; N332glycan, D325, T63 and T320; N332glycan, D325, T63 and H330; N332glycan, D325, L179 and T320; N332glycan, D325, L179 and H330; N332glycan, D325, T320 and H330; N332glycan, D325, T63, T320 and H330; N332glycan, D325, T63, L179 and T320; N332glycan, D325, T63, L179 and H330; N332glycan, D325, L179, T320 and H330; or N332glycan, D325, T63, L179, T320 and H330. In some embodiments, the method comprises administering an antibody or multi-specific antigen binding molecule that competes with or comprises VH and VL regions that bind to the gp120 CD4 binding site, and wherein the human subject is infected with an HIV expressing a gp120 comprising the following amino acid residues, wherein the positions and residues are with reference to SEQ ID NO:73: I201 and F353; I201, I108 and F353; I201, I108, A281 and F353; I201, E102, I108, A281 and F353; or I201, E102, I108, A281, Y318 and F353. In some embodiments, the method entails multiple administrations of the one or more multi-specific antigen binding molecules, optionally with a TLR agonist or an IL-15 receptor agonist, at predetermined intervals. In some embodiments, the method comprises administering a multi-specific antigen binding molecule having a first antigen binding domain that binds to CD3, and a second antigen binding domain that binds to an epitope or region of gp120 in the CD4 binding site (CD4bs) and comprises one or more extracellular (EC) domains of CD4, and optionally, an IL-15 receptor agonist, wherein the multi-specific antigen binding molecule and the IL-15 receptor agonist are administered independently at predetermined intervals. In some embodiments, the subject is chronically infected with HIV. In some embodiments, the one or more multi-specific antigen binding molecules, the polynucleotide, the vector, the LNP and / or the pharmaceutical composition are administered systemically or locally. In some embodiments, the one or more multi-specific antigen binding molecules, the polynucleotide, the vector, the LNP and / or the pharmaceutical composition is administered via a route selected from intravenous, subcutaneous, intramuscular, intradermal and mucosal (e.g. buccal, intranasal, intrarectal, intravaginal). In some embodiments, the one or more multi-specific antigen binding molecules, the polynucleotide, the vector, the LNP and / or the pharmaceutical composition and the one or more additional therapeutic agents are administered by the same routes of administration. In some embodiments, the one or more multi-specific antigen binding molecules, the polynucleotide, the vector, the LNP and / or the pharmaceutical composition and the one or more additional therapeutic agents are administered by different routes of administration. In some embodiments, the one or more multi-specific antigen binding molecules, the polynucleotide, the vector, the LNP and / or the pharmaceutical composition and the one or more additional therapeutic agents are co-administered according to the same schedule (e.g., co-administered at the same time intervals). In some embodiments, the one or more multi-specific antigen binding molecules, the polynucleotide, the vector, the LNP and / or the pharmaceutical composition and the one or more additional therapeutic agents are co-administered according to different schedules (e.g., co-administered at different time intervals). In some embodiments, the one or more multi-specific antigen binding molecules, the polynucleotide, the vector, the LNP and / or the pharmaceutical composition is administered at a dose in the range of from 1 µg / kg to 5 µg / kg, e.g., from 350 µg / kg to 550 µg / kg, e.g., from 0.3 mg / kg to 30 mg / kg, e.g., from 2 mg / kg to 10 mg / kg, e.g., from 1 µg / kg up to 2 µg / kg, 3 µg / kg, 4 µg / kg, 5 µg / kg, 10 µg / kg, 50 µg / kg, 100 µg / kg, 250 µg / kg, 300 µg / kg, 350 µg / kg, 400 µg / kg, 410 µg / kg, 420 µg / kg, 430 µg / kg, 440 µg / kg, 450 µg / kg, 460 µg / kg, 470 µg / kg, 480 µg / kg, 490 µg / kg, 500 µg / kg, 750 µg / kg, 1 mg / kg, 1.5 mg / kg, 2 mg / kg, 2.5 mg / kg, 3 mg / kg, 3.5 mg / kg, 4 mg / kg, 4.5 mg / kg, 5 mg / kg, 6 mg / kg, 7 mg / kg, 8 mg / kg, 9 mg / kg, 10 mg / kg, 15 mg / kg, 20 mg / kg, 25 mg / kg, 30 mg / kg, 35 mg / kg, 40 mg / kg, 45 mg / kg, or 50 mg / kg, body weight per administration. In some embodiments, the one or more multi-specific antigen binding molecules, the polynucleotide, the vector, the LNP and / or the pharmaceutical composition is administered at a dose in the range of 0.05 mg to 1000 mg per administration, e.g., from 0.05 mg to 150 mg per administration, e.g., from 0.05 mg to 0.35 mg per administration, e.g., from 25 mg to 50 mg per administration, e.g., from 30 mg to 35 mg per administration, e.g., from 10 mg to 1000 mg per administration, e.g., from 50 mg to 1000 mg per administration, e.g., from 100 mg to 700 mg per administration, e.g., at least 0.05 mg up to 0.1 mg, 0.2 mg, 0.3 mg, 0.35 mg, 0.4 mg, 0.5 mg, 0.6 mg, 0.7 mg, 0.8 mg, 1.0 mg, 5 mg, 10 mg, 25 mg, 30 mg, 35 mg, 40 mg, 45 mg, 50 mg, 60 mg, 70 mg, 80 mg, 90 mg, 100 mg, 200 mg, 300 mg, 400 mg, 500 mg, 600 mg, 700 mg, 800 mg, 900 mg, or 1000 mg per administration. In some embodiments, the method entails multiple administrations of the one or more multi-specific antigen binding molecules, the polynucleotide, the vector, the LNP and / or the pharmaceutical composition, optionally with one or more additional therapeutic agents, at predetermined intervals. In some embodiments, the method entails administering over a time period of at least about 2 weeks, 3 weeks, 1 month, 6, weeks, 2 months, 10 weeks, 3 months, 4 months, 5 months, 6 months, 7 months, 8 months, 9 months, 10 months, 11 months, 12 months, 13 months, 14 months, 15 months, 16 months, 17 months, 18 months, 19 months, 20 months, 21 months, 22 months, 23 months, 24 months, or longer. In some embodiments, the method entails administering one or more times at predetermined intervals spaced at least 1 week and up to at least 2 weeks, 3 weeks, 1 month, 2 months, 3 months, 4 months, 5 months or 6 months apart. In some embodiments, the one or more multi-specific antigen binding molecules, the polynucleotide, the vector, the LNP and / or the pharmaceutical composition is administered once weekly (i.e., QW), once bi-weekly (i.e. once every other week, or once every two weeks or Q2W), once thrice-weekly (i.e. once every three weeks or Q3W), once monthly (i.e., QM) or once bi-monthly dosing (i.e. once every other month, or once every two months or Q2M), once every three months (Q3M), once every four months (Q4M), once every five months (Q5M), once every six months (Q6M), or less often. In some embodiments, the one or more multi-specific antigen binding molecules, the polynucleotide, the vector, the LNP and / or the pharmaceutical composition is administered two, three, four, five, or more, times intravenously or subcutaneously at an interval or at intervals between once bi-weekly (i.e. once every other week, or once every two weeks or Q2W) to once thrice-weekly (i.e. once every three weeks or Q3W). In some embodiments, the method entails the one or more multi-specific antigen binding molecules have a serum half-life in a human of at least 3 days, e.g., at least 4 days, at least 5 days, at least 6 days, at least 7 days, at least 8 days, at least 9 days, at least 10 days, at least 11 days, at least 12 days, at least 13 days, at least 14 days, at least 15 days, at least 16 days, or longer. In some embodiments, the subject or the mammal is a human. In some embodiments, the subject does not exhibit symptoms of HIV or AIDS in the absence of anti-retroviral treatment (ART) for at least 6 months, at least 1 year, at least 2 years, at least 3 years, or more. In some embodiments, the subject has a viral load of copies / ml blood of less than 500, e.g., less than 400, less than 300, less than 200, less than 100, less than 50, in the absence of anti-retroviral treatment (ART) for at least 6 months, at least 1 year, at least 2 years, at least 3 years, or more.

[0024] In a related aspect, provided are one or more multi-specific antigen binding molecules of the invention, one or more polynucleotides of the invention, or the pharmaceutical composition of the invention for use in methods of treating or preventing HIV in a human subject in need thereof. In some embodiments, the methods comprise identifying a subject infected with an HIV or a population of HIV expressing a gp120 comprising the following amino acid residues: N332glycan, D325 and T63; N332glycan, D325 and L179; N332glycan, D325 and T320; or N332glycan, D325 and H330. In some embodiments, the methods comprise identifying a subject infected with an HIV or a population of HIV expressing a gp120 comprising the following amino acid residues: N332glycan, D325, T63 and L179; N332glycan, D325, T63 and T320; N332glycan, D325, T63 and H330; N332glycan, D325, L179 and T320; N332glycan, D325, L179 and H330; or N332glycan, D325, T320 and H330. In some embodiments, the methods comprise identifying a subject infected with an HIV or a population of HIV expressing a gp120 comprising the following amino acid residues: N332glycan, D325, L179, T320 and H330; N332glycan, D325, T63, T320 and H330; N332glycan, D325, T63, L179 and T320; or N332glycan, D325, T63, L179 and H330. In some embodiments, the methods comprise identifying a subject infected with an HIV or a population of HIV expressing a gp120 comprising the following amino acid residues: N332glycan, D325, T63 and H330; N332glycan, D325, T320 and H330; N332glycan, D325, L179, T320 and H330; or N332glycan, D325, T63, L179, T320 and H330. In some embodiments, the methods further comprise administering to the subject a second agent for treating an HIV infection. In some embodiments, the subject is not receiving antiretroviral therapy (ART) or ART is discontinued prior to administration of the one or more multi-specific antigen binding molecules. In some embodiments, ART is discontinued after one or more administrations of the one or more multi-specific antigen binding molecules. In some embodiments, the methods further comprise administering one or more antiretroviral therapy (ART) agents to the subject. In some embodiments, the methods further comprise administering to the subject at least one of a TLR agonist and an IL-15 receptor agonist. In some embodiments, the method comprises administering a multi-specific antigen binding molecule having a first antigen binding domain that binds to CD3, and a second antigen binding domain that binds to an epitope or region of gp120 in the CD4 binding site (CD4bs) and comprises one or more extracellular (EC) domains of CD4, and optionally, an IL-15 receptor agonist. In some embodiments, the TLR agonist is a TLR2 agonist, a TLR3 agonist, a TLR7 agonist, a TLR8 agonist or a TLR9 agonist. In some embodiments, the TLR7 agonist is selected from the group consisting of vesatolimod, imiquimod, and resiquimod. In some embodiments, the methods entail multiple administrations of the one or more multi-specific antigen binding molecules, optionally with the TLR agonist, at predetermined intervals. In some embodiments, the subject does not exhibit symptoms of HIV or AIDS in the absence of anti-retroviral treatment (ART) for at least 6 months, at least 1 year, at least 2 years, at least 3 years, or more. In some embodiments, the subject has a viral load copies / ml blood of less than 500, e.g., less than 400, less than 300, less than 200, less than 100, less than 50, in the absence of anti-retroviral treatment (ART) for at least 6 months, at least 1 year, at least 2 years, at least 3 years, or more.

[0025] Also disclosed is the use of the multi-specific (e.g., bispecific) antigen binding molecules or antigen binding fragment thereof, or the pharmaceutical compositions, as described herein, in a method of at least one of treating, preventing and inhibiting HIV in a human subject in need thereof. Also disclosed are the multi-specific (e.g., bispecific) antigen binding molecules or antigen binding fragment thereof, or the pharmaceutical compositions, as described herein, for use in a method of at least one of treating, preventing and inhibiting HIV in a human subject in need thereof. In some embodiments, the use further comprises administering to the subject a second agent for treating an HIV infection. In some embodiments, the use further comprises administering to the subject at least one of a TLR agonist and an IL-15 receptor agonist. In some embodiments, the use entails administering a multi-specific antigen binding molecule having a first antigen binding domain that binds to CD3, and a second antigen binding domain that binds to an epitope or region of gp120 in the CD4 binding site (CD4bs) and comprises one or more extracellular (EC) domains of CD4, and optionally, an IL-15 receptor agonist. ). In some embodiments, the TLR agonist is a TLR2 agonist, a TLR3 agonist, a TLR7 agonist, a TLR8 agonist or a TLR9 agonist. In some embodiments, the TLR7 agonist is selected from the group consisting of vesatolimod, imiquimod, and resiquimod. In some embodiments, the use comprises administering a first multi-specific antigen binding molecule and second or additional antigen binding molecules, wherein the first multi-specific antigen binding molecule and the second or additional antigen binding molecules bind to different first and second epitopes or regions of gp120 selected from the group consisting of: (i) third variable loop (V3) (e.g., high mannose patch) comprising a N332 oligomannose glycan; (ii) second variable loop (V2) (e.g., Env trimer apex); (iii) CD4 binding site (CD4bs); (iv) gp120 / gp41 interface; or (v) silent face of gp120. In some embodiments, the first multi-specific antigen binding molecule binds to the third variable loop (V3) (e.g., high mannose patch) comprising a N332 oligomannose glycan, and the second antigen binding molecule binds to the CD4 binding site (CD4bs). In some embodiments, the first multi-specific antigen binding molecule competes with or comprises VH and VL regions from an antibody selected from the group consisting of 10-1074, 10-1074-J, GS-9722 (elipovimab), GS-2872, PGT-121, PGT-121.66, PGT-121.414, PGT-122, PGT-123, PGT-124, PGT-125, PGT-126, PGT-128, PGT-130, PGT-133, PGT-134, PGT-135, PGT-136, PGT-137, PGT-138, PGT-139, VRC24, 2G12, BG18, 354BG8, 354BG18, 354BG42, 354BG33, 354BG129, 354BG188, 354BG411, 354BG426, DH270.1, DH270.6, PGDM12, VRC41.01, PGDM21, PCDN-33A, BF520.1 and VRC29.03, and the second antigen binding molecule competes with or comprises VH and VL regions from an antibody selected from the group consisting of b12, F105, VRC01, VRC07, VRC07-523, VRC03, VRC06, VRC06b01 VRC08, VRC0801, NIH45-46, 3BNC117, 3BNC60, PGV04 (a.k.a., VRC-PG04); CH103, 44-VRC13.01, 1NC9, 12A12, N6, 1-18, N49-P7, NC-Cow1, IOMA, CH235 and CH235.12, N49P6, N49P7, N49P11, N49P9 and N60P25. In some embodiments, the first multi-specific antigen binding molecule competes with or comprises VH and VL regions from an antibody selected from the group consisting of 10-1074, 10-1074-J, GS-9722 (elipovimab), GS-2872, PGT-121, PGT-121.66, PGT-121.414 and PGT-134, and the second antigen binding molecule competes with or comprises VH and VL regions from 3BNC117, GS-9723, VRC07 or VRC07-523. In some embodiments, the first multi-specific antigen binding molecule competes with or comprises VH and VL regions from an antibody selected from the group consisting of 10-1074, 10-1074-J, GS-9722 (elipovimab), GS-2872, PGT-121, PGT-121.66, PGT-121.414 and PGT-134, and the second antigen binding molecule comprises an EC domain of CD4. In some embodiments, the first multi-specific antigen binding molecule binds to the CD4 binding site (CD4bs) and the second or additional antigen binding molecules bind to the third variable loop (V3) (e.g., high mannose patch) comprising a N332 oligomannose glycan. In some embodiments, the first multi-specific antigen binding molecule competes with or comprises VH and VL regions from an antibody selected from the group consisting of GS-9723, GS-5423, 3BNC117, 3BNC60, b12, F105, VRC01, VRC07, VRC07-523, VRC03, VRC06, VRC06b01 VRC08, VRC0801, NIH45-46, PGV04 (a.k.a., VRC-PG04); CH103, 44-VRC13.01, 1NC9, 12A12, N6, 1-18, N49-P7, NC-Cow1, IOMA, CH235 and CH235.12, N49P6, N49P7, N49P11, N49P9 and N60P25, and the second or additional antigen binding molecules compete with or comprise VH and VL regions from an antibody selected from the group consisting of GS-9722 (elipovimab), GS-2872, PGT-121, PGT-121.66, PGT-121.414, PGT-122, PGT-123, PGT-124, PGT-125, PGT-126, PGT-128, PGT-130, PGT-133, PGT-134, PGT-135, PGT-136, PGT-137, PGT-138, PGT-139, 10-1074, 10-1074-J, VRC24, 2G12, BG18, 354BG8, 354BG18, 354BG42, 354BG33, 354BG129, 354BG188, 354BG411, 354BG426, DH270.1, DH270.6, PGDM12, VRC41.01, PGDM21, PCDN-33A, BF520.1 and VRC29.03. In some embodiments, the first multi-specific antigen binding molecule competes with or comprises VH and VL regions from an antibody selected from the group consisting of GS-9723, GS-5423, 3BNC117, VRC07 and VRC07-523, and the second or additional antigen binding molecules compete with or comprise VH and VL regions from 10-1074, 10-1074-J, GS-9722 (elipovimab), GS-2872, PGT-121, PGT-121.66, PGT-121.414 and PGT-134. In some embodiments, the first multi-specific antigen binding molecule competes with or comprises an EC domain of CD4, and the second or additional antigen binding molecules compete with or comprise VH and VL regions from an antibody selected from the group consisting of GS-9722 (elipovimab), GS-2872, PGT-121, PGT-121.66, PGT-121.414, PGT-122, PGT-123, PGT-124, PGT-125, PGT-126, PGT-128, PGT-130, PGT-133, PGT-134, PGT-135, PGT-136, PGT-137, PGT-138, PGT-139, 10-1074, 10-1074-J, VRC24, 2G12, BG18, 354BG8, 354BG18, 354BG42, 354BG33, 354BG129, 354BG188, 354BG411, 354BG426, DH270.1, DH270.6, PGDM12, VRC41.01, PGDM21, PCDN-33A, BF520.1 and VRC29.03. In some embodiments, the first multi-specific antigen binding molecule competes with or comprises an EC domain of CD4, and the second or additional antigen binding molecules compete with or comprise VH and VL regions from an antibody selected from the group consisting of 10-1074, 10-1074-J, GS-9722 (elipovimab), GS-2872, PGT-121, PGT-121.66, PGT-121.414 and PGT-134. In some embodiments, the use further comprises administering second or additional antigen binding molecules compete with or comprise VH and VL regions from an antibody selected from the group consisting of GS-9723, GS-5423, 3BNC117, 3BNC60, b12, F105, VRC01, VRC07, VRC07-523, VRC03, VRC06, VRC06b01 VRC08, VRC0801, NIH45-46, PGV04 (a.k.a., VRC-PG04); CH103, 44-VRC13.01, 1NC9, 12A12, N6, 1-18, N49-P7, NC-Cow1, IOMA, CH235 and CH235.12, N49P6, N49P7, N49P11, N49P9 and N60P25. In some embodiments, the use further comprises administering second or additional antigen binding molecules compete with or comprise VH and VL regions from an antibody selected from the group consisting of GS-9723, GS-5423, 3BNC117, VRC07 and VRC07-523. In some embodiments, the use comprises co-administering: (i) a multi-specific antigen binding molecule comprising an EC domain of CD4, as described herein; (ii) an antibody that competes with or comprises VH and VL regions from an antibody selected from the group consisting of 10-1074, 10 1074-J, GS-9722 (elipovimab), GS-2872, PGT-121, PGT-121.66, PGT-121.414 and PGT-134; and (iii) an antibody that competes with or comprises VH and VL regions from an antibody selected from the group consisting of GS-9723, GS 5423, 3BNC117, VRC07 and VRC07-523. In some embodiments, the use further comprises administering to the human subject an additional antigen binding molecule or antigen binding fragment thereof that at least one of binds, inhibits, and neutralizes HIV or a polynucleotide encoding the additional antigen binding molecule or antigen binding fragment thereof. In some embodiments, the human subject is infected with an HIV expressing a gp120 comprising the following amino acid residues: N332 / D325; N332 / D325 / H330; N332 / D325 / H330 / T320; N332 / D325 / H330 / T63; N332 / D325 / H330 / T63 / T320; or N332 / D325 / H330 / T63 / T320 / L179.

[0026] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. Although methods and materials similar or equivalent to those described herein can be used in the practice or testing of the present bispecific molecules described herein, the exemplary methods and materials are described below. In case of conflict, the present application, including definitions, will control. The materials, methods, and examples are illustrative only and not intended to be limiting.

[0027] Other features and advantages of the invention will be apparent from the following detailed description and from the claims.BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figures 1A and 1B illustrate a homology model of mSP34. The sidechains of potential solvent exposed sequence liabilities are shown as spheres and labelled using single letter amino acid nomenclature and numbered according to Kabat. Figures 2A-2B. Figure 2A illustrates a sequence alignment of the mSP34 heavy chain variable domain and five comparators found in clinical stage antibodies. Potential solvent exposed sequence liabilities (Figure 1) are underlined. Figure 2A depicts SEQ ID NOs: 1044-1049, respectively, in order of appearance. Figure 2B illustrates a sequence alignment of the mSP34 light chain variable domain and five comparators found in clinical stage antibodies. Figure 2B depicts SEQ ID NOs: 1050-1055, respectively, in order of appearance. Figure 3 illustrates a structural superposition of homology models of mSP34 (white) and the closest human germline (black) found using the IMGT domain gap align server. The models were created using Discovery Studio 2017r2 (Biovia) and superposed using PyMOL (Schrodinger, Inc.). The Cα backbone trace is shown. Figures 4A-4B. Figure 4A illustrates a heavy chain (HC) sequence alignment of mSP34, the closest human germline and the first round of huSP34 heavy chain variants (SEQ ID NOs: 1064-1075, respectively, in order of appearance). Figure 4B illustrates a light chain (LC) sequence alignment of mSP34, the closest human germline and the first round of huSP34 light chain variants (SEQ ID NOs: 1056-1063, respectively, in order of appearance). Figures 5A-5B. Figure 5A illustrates a heavy chain (HC) sequence alignment of mSP34, the closest human germline and the second round of huSP34 heavy chain variants ((SEQ ID NOs: 1082-1087, respectively, in order of appearance). Figure 5B illustrates a light chain (LC) sequence alignment of mSP34, the closest human germline and the second round of huSP34 light chain variants (SEQ ID NOs: 1076-1081, respectively, in order of appearance). Figures 6A-6D illustrate an analysis of anti-CD3 affinity (Octet BLI) (Fig. 6A), charge homogeneity (CX-1) (Fig. 6B), aggregation propensity (size exclusion column (SEC)) (Fig. 6C) and match to the human germline (Fig. 6D) for the first and second round huSP34 Fab variants, mSP34 human chimeric Fab and comparator huSP34 Fab molecules. Figures 7A-7D. Figure 7A illustrates examples of scFv-Fc / Fab-Fc bispecific antibodies designed as part of the present bispecific molecules. Figure 7B illustrates examples of CD4 ECD-Fc / Fab-Fc bispecific fusion proteins designed as part of the present bispecific molecules. Figure 7C illustrates examples of CD4 ECD-Fc / scFv-Fc bispecific fusion proteins designed as part of the present bispecific molecules. Figure 7D illustrates examples of CD4 ECD-Fc / scFv-Fc or CD4 ECD-Fc / Fab-Fc bispecific fusion proteins incorporating bivalent or tandem CD4 ECDs and designed as part of the present bispecific molecules. All technologies illustrated are optional and may be omitted or used in combinations beyond what is shown in these figures. Key: ΔEF = remove effector function (e.g., FcγR binding reduced or eliminated with at least one of L234A and L235A (LALA) mutations; C1q binding reduced or eliminated with P331S mutation; collectively "AAS"); knob and hole = engineered Fc heterodimer (e.g., "hole" (H) mutations include T366S, L368A and Y407V ("SAV"); "knob" (K) mutations include T366W ("W"); HLE = half-life extension (e.g., M252Y, S254T and T256E ("YTE") or M428L and N434S ("LS")); ΔProA = reduce or eliminate Protein A binding (e.g., H435R or H435R+Y436F ("RF")); linker = scFv linker (e.g., (GGGS) 4 (SEQ ID NO: 711) or (GGGGS) 4 (SEQ ID NO: 750)). Figures 8A and 8B. Figure 8A illustrates representative HiTrap ®< SP HP (Cytiva Life Sciences) cation exchange chromatography (employed to isolate the bispecific antibody or fusion protein Fc heterodimer from Fc homodimer or other contaminants with low isoelectric points (pI) using a gradient of 0-30% 1M NaCl in 20 mM sodium phosphate pH 7.0) to separate the desired bispecific antibody or fusion protein Fc heterodimer from Fc homodimer or other contaminants with low isoelectric points (pI)) and SDS-PAGE analysis for purification of the hPGT121.66 (see, e.g., WO 2018 / 237148) AAS+W / huSP34.13.10scFv AAS+SAV+RF Fab-scFv-Fc bispecific antibody (Ab 265; SEQ ID NOs: 848, 829 and 823) (bispecific molecules described herein are summarized in Table 53). Figure 8B illustrates representative HiTrap ®< SP HP (Cytiva Life Sciences) cation exchange chromatography results and SDS-PAGE analysis for purification of the hCD4 D1.22 Fc AAS+SAV+YTE / huSP34.3.13 AAS+W+YTE Fab-Fc-fusion bispecific molecule 185. Figure 9 illustrates the analytical CX-1 % main peak values for all round 1 huSP34 Fab variants. The variants containing asparagine (N) at heavy chain (HC) position 100 were compared with those containing histidine at heavy chain position 100 using an unpaired t test. The results suggest that H100 contributed significantly to improved charge homogeneity. Figure 10 illustrates representative binding of a bispecific to immobilized CD3 obtained from SPR experiments. Black lines denote injection of the bispecific 249 at various concentrations for a period of 120 seconds. Dissociation was then monitored for 80 seconds except for the highest concentration where dissociation was monitored for 10.5 minutes. Smoothed gray lines represent the fit to this data as obtained from a simple kinetic model. This model was used to derive k on , k off . and K D for these interactions. Figure 11 illustrates a representative binding concentration-response curve of PGT121.66 x huSP34 bispecific antibodies 265 (circle), 251 (square), 259 (upright triangle; dotted line). 258 (inverted triangle; solid line) and 260 (diamond) to CD4+ T cells (from Donor 4574). Bispecific molecules described herein are summarized in Table 53. Figure 12 illustrates a representative binding concentration-response curve of PGT121.66 x huSP34 bispecific antibodies 265 (circle), 251 (square), 259 (upright triangle; dotted line). 258 (inverted triangle; solid line) and 260 (diamond) to CD8+ T cells (from Donor 4574). Figures 13A-13B illustrate a representative binding concentration-response curve of (A) PGT121.66 x huSP34 bispecific antibodies 257 (solid circle), 274 (solid square), 273 (solid, upright triangle), 275 (solid inverted triangle), 256 (solid diamond), 243 (half-solid circle), 251 (half-solid square) and 261 (half-solid upright triangle) to human CD4+ T cells in PBMCs (from Donor hu 151) and (B) CD4 x huSP34 bispecific 180 to human CD4+ T cells in PBMCs (from Donor hu 921). Figures 14A-14B illustrate a representative binding concentration-response curve of (A) PGT121.66 x huSP34 bispecific antibodies 257 (solid circle), 274 (solid square), 273 (solid, upright triangle), 275 (solid inverted triangle), 256 (solid diamond), 243 (half-solid circle), 251 (half-solid square) and 261 (half-solid upright triangle) to monkey CD4+ T cells in PBMCs (from rhesus Donor rh 3563) and (B) CD4 x huSP34 bispecific 180 to monkey CD4+ T cells in PBMCs (from cynomolgus Donor cy 2177). Figures 15A-15B illustrate a representative binding concentration-response curve of (A) PGT121.66 x huSP34 bispecific antibodies 257 (solid circle), 274 (solid square), 273 (solid, upright triangle), 275 (solid inverted triangle), 256 (solid diamond), 243 (half-solid circle), 251 (half-solid square) and 261 (half-solid upright triangle) to human CD8+ T cells in PBMCs (from Donor hu 151) and (B) CD4 x huSP34 bispecific 180 to human CD8+ T cells in PBMCs (from Donor hu 921). Figures 16A-16B illustrates a representative binding concentration-response curve of (A) PGT121.66 x huSP34 bispecific antibodies 257 (solid circle), 274 (solid square), 273 (solid, upright triangle), 275 (solid inverted triangle), 256 (solid diamond), 243 (half-solid circle), 251 (half-solid square) and 261 (half-solid upright triangle) to rhesus CD8+ PBMCs (from Donor rh 3563) and (B) CD4 x huSP34 bispecific 180 to monkey CD8+ T cells in PBMCs (from cynomolgus Donor cy 2177). Figure 17 illustrates EC 50 values derived from concentration-response curves of PGT121.66 x huSP34 bispecific antibodies 257 (solid circle), 274 (solid square), 273 (solid, upright triangle), 275 (solid inverted triangle), 256 (solid diamond), 243 (open circle), 251 (open square) and 261 (open upright triangle) binding to human (hu) and rhesus (rh) CD4+ PBMCs. Figure 18 illustrates EC 50 values derived from concentration-response curves of PGT121.66 x huSP34 bispecific antibodies 257 (solid circle), 274 (solid square), 273 (solid, upright triangle), 275 (solid inverted triangle), 256 (solid diamond), 243 (open circle), 251 (open square) and 261 (open upright triangle) binding to human (hu) and rhesus (rh) CD8+ PBMCs. Figure 19 illustrates a representative killing concentration-response curve by PGT121.66 x huSP34 bispecific antibodies 265 (solid circle), 259 (upright triangle), 258 (inverted triangle), 251 (square), and 260 (half-solid circle) using PBMC effector cells. Figure 20 illustrates a Pearson correlation of PGT121.66 x huSP34 bispecific antibodies (from highest to lowest PBMC killing EC 50 value) 258, 260, 259, 251 and 265. huSP34 variant CD3 binding affinity vs. PBMC killing EC 50 values. The killing EC 50 value is the geometric mean EC 50 value of the 5 viruses tested with two donors. Figure 21 illustrates a Pearson correlation of PGT121.66 x huSP34 bispecific antibodies (from highest to lowest T cell killing EC 50 value) 260, 258, 259, 251 and 265. huSP34 variant CD3 binding affinity vs. isolated T cell killing EC 50 values. The killing EC 50 value is the geometric mean EC 50 value of the 5 viruses tested with two donors. Figure 22 illustrates a representative concentration-response curve of killing by 3BNC117.52.64 x huSP34 bispecific antibodies 237 (circle), 230 (square) and 232 (triangle) using PBMC effector cells. Figure 23 illustrates a Pearson correlation of 3BNC117.52.64 x CD3 bispecific antibodies (from highest to lowest PBMC cell killing EC 50 value) 238, 230 and 237. huSP34 variant CD3 binding affinity vs. PBMC killing EC 50 values. The killing EC 50 value is the geometric mean EC 50 value of the 5 viruses tested with two donors. Figure 24 illustrates a representative killing concentration-response curve of primary HIV-infected (virus 657) CD4+ cells by PGT121.66 x huSP34 bispecific molecules 265 (circle) 251 (square) and 259 (triangle) using PBMC effector cells (E:T = 3:1). Figure 25 illustrates a representative killing concentration-response curve of primary HIV-infected (virus 7552) CD4+ cells by PGT121.66 x huSP34 bispecific molecules 265 (circle) 251 (square) and 259 (triangle) using PBMC effector cells (E:T = 3:1). Figure 26 illustrates EC 50 values of killing curves plotted across CD4 ECD x α-CD3 bispecific molecules (from right-to-left: 212, 211, 213, 198, 187, 199 and 186). The geomean EC 50 value for each molecule ( - ) calculated from all viruses / donors are shown. Figure 27 illustrates antibody- and bispecific- mediated killing of primary CD4+ T cells infected with different HIV isolates. Primary CD4+ T cells were infected with a panel of 32 HIV isolates and incubated with the antibodies or antibody combinations indicated and autologous PBMCs as effector cells. Data on the graph represents the EC 50 values determined in the killing assay for each HIV isolate and antibody, bispecific or antibody / bispecific combination (from left to right: PGT121.42, h3BNC117.52.64, bispecific 180, PGT121.42 + h3BNC117.52.64 combination, PGT121.42 + h3BNC117.52.64 + bispecific 180 combination. Horizontal bars indicate the median and 95% confidence intervals. Figure 28 EC 50 values of binding curves for 24 virus isolate-infected cells plotted across test bispecific molecules 186, 255 and 230. Figure 29 illustrates pharmacokinetic (PK) profiles for PGT121.66 x huSP34 bispecific molecules 250 (circle), 252 (square), and 251 (triangle) dosed at 1 mg / kg IV to naive male cynomolgus monkeys (n=3). For comparison, dose normalized individual monkey PK profiles for Comparator #7 bispecific (dashed lines) dosed at 30 mg / kg IV on Day 0 and 7. Each symbol is the measured mean ± standard deviation (SD) serum concentration. Figure 30 illustrates PK profiles for PGT121.66 x huSP34 bispecific molecules 264 (solid circle), 265 (solid square), 261 (solid triangle), 262 (open circle), and 263 (open square) following 1 mg / kg IV dosing to naive male cynomolgus monkeys (n=3). Each symbol is the measured mean ± SD serum concentration. Figure 31 illustrates PK profiles for PGT121.66 x huSP34 bispecific molecules 257 (solid circle), 274 (open circle), 273 (solid square), 275 (open square), 256 (solid triangle), and 243 (open triangle) following 1 mg / kg IV dosing to naive male cynomolgus monkeys (n=2). Each symbol is the measured mean serum concentration. Figure 32 illustrates PK profiles for PGT121.66 x huSP34 bispecific molecules 249 (solid circle), 276 (solid square), 277 (solid triangle), and 218 (open circle) following 1 mg / kg IV dosing to naïve male cynomolgus monkeys (n=2 or 3). Each symbol is the measured mean ± SD serum concentration. Figure 33 illustrates PK profiles for CD4 ECD x huSP34 bispecific molecules 186 (solid circle), 187 (solid square), 180 (open circle), 182 (open square), and 185 (open triangle) following 1 mg / kg IV dosing to naive male cynomolgus monkeys (n=3). Bispecific molecules 186 and 187 were dosed on Day 0 and 14. Each symbol is the measured mean ± SD serum concentration. Figure 34 illustrates organization of an expression vector having polynucleotide sequences encoding the expression of the three polypeptide chains of an asymmetric bispecific molecule having an anti-HIV gp120 Fab and an anti-CD3 scFv. The vector was designed to drive the expression of three polypeptide chains: heavy chain and light chain of anti-gp120 Fab and scFv of anti-CD3 from three separate expression cassettes, each driven by its own promoter of equal transcription strength (e.g., a cytomegalovirus (CMV) promotor). An additional DNA cassette having a polynucleotide encoding a eukaryotic selection marker (e.g., glutamine synthetase (GS)) using a promoter of relatively weaker transcription strength (e.g., an SV40 promoter) was included in the expression vector. DNA encoding for a bacterial origin of replication (e.g., Ori) and an antibiotic selection marker (e.g., ampicillin (AmpR)) were utilized for expression vector production in E. coli. Figure 35 illustrates an evaluation of expression performance of cell lines expressing asymmetric bispecific molecules having an anti-HIV gp120 Fab and an anti-CD3 scFv. The expression performance was assessed by culturing the cells expressing the bispecific molecules in a fed-batch process and evaluating their performance at baseline (D0) and upon aging (D30). Expression of multiple species including the bispecific molecule of interest was measured using a ProA based Bio-layer interferometry and the relative levels of the heterotrimer (Het%, heterotrimer%) comprising the asymmetric bispecific molecule was monitored using size separation methods (non-reduced capillary electrophoresis). Figure 36 illustrates organization of an expression vector having polynucleotide sequences encoding the expression of the three polypeptide chains of an asymmetric bispecific molecule having an anti-CD3 Fab and a CD4 extracellular (EC) domain-Fc fusion protein. The vector was designed to drive the expression of three polypeptide chains: heavy chain and light chain of anti-CD3 Fab and CD4-Fc fusion protein from three separate expression cassettes, each driven by its own promoter of equal transcription strength (e.g., a cytomegalovirus (CMV) promotor). An additional DNA cassette having a polynucleotide encoding a eukaryotic selection marker (e.g., glutamine synthetase (GS)) using a promoter of relatively weaker transcription strength (e.g., an SV40 promoter) was included in the expression vector. DNA encoding for a bacterial origin of replication (e.g., Ori) and an antibiotic selection marker (e.g., ampicillin (AmpR)) were utilized for expression vector production in E. coli. Figure 37 illustrates an evaluation of expression performance of cell lines expressing asymmetric bispecific molecules having an anti-CD3 Fab and a CD4-Fc fusion protein. The expression performance was assessed by culturing the cells under the production mode (fed-batch process). Each dot represents the selected clones, where the ratio of the desired species (Het%, heterotrimer%) was monitored by size separation method (non-reduced capillary electrophoresis). The amounts of desired heterotrimer comprising the asymmetric bispecific molecule were calculated as the amount of total bispecific molecule (detected via ProA Biosensors) and then multiplied by the heterotrimer % (specific titer in g / L). Figure 38 illustrates PGT121 / anti-CD3 bispecific antibody pharmacokinetics in serum before ART discontinuation. Peak serum antibody levels are shown following each of the twelve infusions (2 infusions first with anti-CD3KO version followed by 10 of anti-CD3 version) of antibody and during the washout period. Dotted lines indicate limit of detection. Dotted line indicates limit of detection. Figure 39 illustrates SHIV viral loads following ART discontinuation. Plasma viral load for 168 days following ART discontinuation. Numbers and percentages of animals that did not show viral rebound are shown. Limit of detection is 1.7 log RNA copies per ml. Figure 40 illustrates plasma viral loads before and after CD8+ cell depletion in animals with no viral rebound (n=1) and animals with viral rebound (n=4) following ART discontinuation. Numbers of animals that remained aviremic and total number of animals are shown. Arrows indicate CD8+ cell depletion on day 245 (week 77) after ART discontinuation. Limit of detection is 1.7 log RNA copies per ml. DETAILED DESCRIPTION 1. Introduction

[0029] Provided are multi-specific or bispecific antigen binding molecules, targeting CD3 and an HIV antigen (e.g., gp120 or gp41), designed for improved manufacturing efficiency, reduced cost, improved drug-like properties (e.g., increased sequence identity to human germline and reduced off-target binding and inducement of anti-drug antibodies (ADAs)). The Fc-containing multi-specific or bispecific molecules described herein can be efficiently produced at high yield using a single cell line with a simplified purification process. We have developed multi-specific or bispecific molecules having an anti-CD3 antibody variable domain with (1) a high affinity for human and non-human primate (NHP) (e.g., to facilitate pre-clinical toxicity studies) CD3, (2) a high sequence similarity to the human germline (e.g., to reduce or eliminate risk of immunogenicity and anti-drug antibody (ADA) reactions in human patients), (3) IgG-like pharmacokinetic (PK) properties with no evidence of ADA in non-human primates (e.g., to facilitate pre-clinical efficacy and toxicity studies) (4) reduced product heterogeneity via removal of sequence liabilities (e.g., deamidation, aspartate isomerization) to improve manufacturing, (5) high thermodynamic stability (e.g., to ensure product stability), (6) low aggregate content (e.g., to reduce risk of immunogenicity), (7) low polyspecificity (to reduce the risk of immunogenicity and improve PK properties), and (8) a heavy chain variable region (VH) with low or no binding to Protein A affinity chromatography resin (e.g., to facilitate efficient purification of bispecific antibody heterodimers). Further the anti-CD3 antibody variable domain possesses all the foregoing desirable properties in both scFv and Fab formats, such that it can be incorporated into a variety of bispecific antibody formats containing three or fewer polypeptide chains, e.g., to limit light chain miss-pairing or other sources of bispecific antibody product heterogeneity.

[0030] Multi-specific or bispecific molecules can comprise a first antigen binding domain, targeting human CD3, that is a single chain variable fragment (scFv) fused to Fc, while the second antigen binding domain, targeting an HIV antigen (e.g., gp120 or gp41) is comprised of a Fab antigen binding fragment fused to Fc. The resulting molecules have three polypeptide chains (i.e., scFv heavy chain, Fab heavy chain, Fab light chain), which are co-expressed in a single cell line. Formation of the desired bispecific heterodimeric molecule is facilitated through use of mutations in the Fc region that limits unwanted homodimerization of either half molecule, while the use of a scFv fragment as one of the antigen binding arms eliminates the need to co-express two different light chains that can otherwise lead to heterogeneity resulting from incorrect light chain pairing. Purification of the multi-specific antigen binding molecules is simplified and improved by eliminating the ability of the variable regions and one the of Fc regions to bind Protein A. Employing this design strategy, one of the two possible homodimer impurities is not retained during Protein A-based capture of the bispecific from cell culture media.

[0031] In various embodiments of the present multi-specific antigen binding molecules the first and second Fc regions of native human IgG sequences can be modified to promote heavy chain heterodimerization, allow for simplified and efficient purification, and to reduce or remove binding to FcγR and C1q.

[0032] Heterodimerization of the first and second Fc regions can be facilitated by introduction of 'knobs-into-holes' mutations (Atwell et al. 1997. JMB 270:26-35). The 'hole' mutations (T366S, L368A and Y407V) are incorporated into one Fc-containing chain, the T366W 'knob' mutation is used in the other chain (Atwell et al., supra). In addition, the C220S mutation can be incorporated into the IgG1 hinge region of the scFv-containing arm to eliminate the free cysteine, which would otherwise form a disulfide bond with a corresponding cysteine in an immunoglobulin light chain. Co-transfection of such constructs into a single host cell promotes formation of a heterodimeric Fc, with low levels of homodimer contaminants. The S354C (made in the Fc containing the 'knob' mutations) and Y349C (made in the Fc containing the 'hole' mutations) mutations can optionally be incorporated to generate a covalent bond between the two halves of the heterodimeric Fc if additional thermodynamic stability is desired (Merchant et al. 1998. Nat. Biotechnol. 16: 677-81). As described herein, numbering of amino acid positions for substitutions in the first and second Fc regions is according to Eu index.

[0033] Purification of the heterodimeric molecule away from contaminating homodimeric products can be facilitated by introducing into the first or the second Fc region H435R or H435R+Y436F mutations to reduce or eliminate Protein A binding (Jendeberg, L. et al. 1997 J. Immunol. Methods 201:25-34), optionally combined with mutations in any variable heavy (VH) region of the same chain should such VH be derived from a human VH3 germline. For example, additional VH mutations can be made at one or more of positions R19, T57, G65, Q81 and N82a, as numbered according to the Kabat numbering system. Incorporation of such amino acid substitutions in the Fc region, and optionally in the VH region, can reduce or eliminate Protein A binding of the homodimer contaminants, and greatly simplifies purification of the desired heterodimer away from remaining homodimer contaminant via additional chromatography steps (e.g., ion exchange).

[0034] In certain embodiments, the first and second Fc domain or Fc region incorporate amino acid substitutions to reduce or eliminate one or more of effector function (antibody-dependent cellular cytotoxicity (ADCC), antibody-dependent cellular phagocytosis (ADCP), and complement-dependent cytotoxicity (CDC)). In these contexts, at least one of the first and second Fc region can incorporate one or both of the L234A and L235A mutations to reduce or eliminate FcγR binding (Chappel, M. S et al. 1991 PNAS 88:9036-9040), and / or the P331S mutation to reduce or eliminate C1q binding (Xu Y, et al. J Biol Chem. 1994. 269:3469-74).

[0035] scFv fragments represent a minimal antibody-derived antigen binding unit, and are generated by direct fusion of a variable heavy and variable light domain via a flexible polypeptide linker (Huston et al., 1988, PNAS 85:5879-5883). The sequence of this linker can contain 3 or 4 repeats of a GGGGS motif (SEQ ID NO: 712) (Desplancq et al. 1994, Protein Engineering 7:1027-1033). In various embodiments, the G44C mutation (variable heavy domain) and the G100C mutation (variable light domain) can be incorporated to generate a covalent disulfide bond between the VH and VL domains of the scFv if additional thermodynamic stability is desired (Brinkmann, U et al., 1993, PNAS 90: 7538-7542).

[0036] For a first antigen binding domain in the form of a scFv, and a second antigen binding domain in the form of a Fab, the three polypeptide chains can co-expressed in a single host cell and purified via Protein A chromatography. The desired Fab-scFv-Fc heterodimer is the dominant species observed in non-reducing SDS-PAGE analysis. Contaminating homodimeric species can be further mitigated using the H435R or H435R+Y436F mutations in the first or second Fc regions. Subsequent polishing using ion exchange chromatography and dialysis into a standard formulation buffer generates final material with high purity and homogeneity, e.g., that is at least 90%, e.g., at least 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or more, pure.2. Anti-CD3 Effector Arm

[0037] Cluster of Differentiation (CD3) is a multimeric protein complex that is composed of four distinct polypeptide chains: epsilon (ε) (CD3E; NCBI Gene ID: 916), gamma (γ) (CD3G; NCBI Gene ID: 917), delta (δ) (CD3D; NCBI Gene ID: 915) and zeta (ζ) (CD247; NCBI Gene ID: 919), that assemble and function as three pairs of dimers (εγ, εδ, ζζ). CD3 proteins have an N-terminal extracellular region, a transmembrane domain, and a cytoplasmic tail where the immunoreceptor tyrosine activation motifs (ITAMs) are located. The extracellular domains of CD3 ε, γ and δ contain an immunoglobulin-like domain and thus are considered part of the immunoglobulin superfamily. The CD3 / T-cell co-receptor helps to activate both CD8 +< T-cells and also CD4 +< T-cells.

[0038] The amino acid sequence of human CD3ε can be found at UNiProtKB-P07766 and is provided below (the signal sequence is underlined):

[0039] The amino acid sequence of human CD3δ can be found at UNiProtKB-P04234 and is provided below (the signal sequence is underlined):

[0040] Antibodies that bind human CD3 have been described, e.g., in Kuhn & Weiner, Immunotherapy, 8(8):889-906 (2016); WO 2015 / 001085; WO 2015 / 104346. OKT3 (Muromonab-CD3), an anti-CD3 antibody directed against CD3ε, has been clinically approved for use in humans for the induction of immunosuppression in solid organ transplantation for the prevention and treatment of rejection (Norman, Therapeutic Drug Monitoring, 17, 615-620

[0041] (1995)). Teplizumab, also known under the names hOKT3γ1 (Ala-Ala) and MGA031, is a humanized IgG1 antibody that was developed by grafting the complementarity determining region of OKT3 into a human IgG1 backbone. Introduction of two point mutations in its Fc portion decreases binding to FcR. Otelixizumab (ChAglyCD3, TRX4, GSK2136525) was derived from the rat antibody YTH12.5. This humanized IgG1 bears a single mutation in the γ1 Fc portion to avoid glycosylation and thus inhibit FcR binding. Visilizumab (Nuvion, HuM291) is a humanized IgG2 antibody that is rendered non-mitogenic by two point mutations in its Fc region. Foralumab (28F11-AE; NI-0401) is an entirely human anti-CD3 mAb; the Fc portion of this human IgG1 was mutated such that the mAb is non FcR binding in vitro and exhibits only minor cytokine release in vivo while maintaining modulation of the CD3 / TCR and T-cell depletion. Non-limiting examples of anti-CD3 antibodies are also disclosed in US 2016 / 0333095A1.

[0042] In certain embodiments, the anti-CD3 antigen binding domains described herein bind human CD3. In some instances, the anti-CD3 antigen binding domains described herein bind human CD3ε. In some embodiments, the anti-CD3 antigen binding domains described herein bind human CD3δ.Exemplary anti-CD3 Antigen Binding Domain Sequences

[0043] In various embodiments, the first antigen binding domain, targeting or binding to, or specifically binding to human CD3, comprises a first heavy chain variable domain (VH) and a first light chain variable domain (VL), comprising: (i) a first VH-complementarity determining region (CDR) 1 comprising the amino acid sequence of TYAMN (SEQ ID NO:1); (ii) a first VH-CDR2 comprising the amino acid sequence of RIRSKYNNYATYYAX 1 SVKX 2 , wherein X 1 is A or D and X 2 is G or S (SEQ ID NO:2); (iii) a first VH-CDR3 comprising the amino acid sequence of HGNFGX 3 SYVSWFAY, wherein X 3 is H or N (SEQ ID NO:3); (iv) a first VL-CDR1 comprising the amino acid sequence of GSSTGAVTTGHYAN (SEQ ID NO: 4); (v) a first VL-CDR2 comprising the amino acid sequence of GTX 4 X 5 RAP, wherein X 4 X 5 is SN or NK (SEQ ID NO:5); and (vi) a first VL-CDR3 comprising the amino acid sequence of ALWYSNX 6 WV, wherein X 6 is L or R (SEQ ID NO:6), wherein the first VH-CDR1, the first VH-CDR2, the first VH-CDR3, the first VL-CDR1, the first VL-CDR2, and the first VH-CDR3 are according to Kabat, wherein the first VH-CDR1, the first VH-CDR2, the first VH-CDR3, the first VL-CDR1, the first VL-CDR2 and the first VL-CDR3 comprise the following amino acid sequences, respectively: SEQ ID NOs: 1, 12, 8, 4, 9 and 10.

[0044] Illustrative sequences of complementarity determining regions (CDRs) of exemplary first antigen binding domains of the multi-specific antigen binding molecules, targeting human CD3, according to the Kabat definition, the Chothia definition, the IMGT definition and Honegger definition are provided in Tables A1, A2, A3 and A4, respectively. Multi-specific antigen binding molecules comprising the CDRs identified herein are encompassed by the present application. It is to be understood that this disclosure also encompasses multi-specific antigen binding molecules (e.g., anti-CD3 / anti-HIV antigen bispecific antibodies) comprising the CDRs according to any other CDR definition (e.g., Honegger definition, enhanced Chothia definition, Martin definition, Gelfand definition, AbM definition, contact definition, see, e.g., bioinf.org.uk / abs / #cdrdef and Dondelinger, et al., Front Immunol. (2018) 9:2278) of the anti-CD3 / anti-HIV multi-specific antigen binding molecules disclosed herein. TABLE A1 - CDRs for anti-CD3 binding arm (Kabat) Ab Name VH - CDR1 VH - CDR2 VH - CDR3 VL - CDR1 VL - CDR2 VL - CDR3 400TYAMN SEQ ID NO: 1RIRSKYNNYATYYAX 1 SVKX 2 HGNFGX 3 SYVSWFAYGSSTGAVTTGHYAN SEQ ID NO: 4GTX 4 X 5 RAPALWYSNX 6 WVX 1 is A or DX 3 is H or NX 4 X 5 is SN or NKX 6 is L or RX 2 is G or SSEQ ID NO:3SEQ ID NO:6SEQ ID NO:2SEQ ID NO:5401TYAMN SEQ ID NO: 1RIRSKYNNYATYYADSVKX 2 HGNFGHSYVSWFAY SEQ ID NO:8GSSTGAVTTGHYAN SEQ ID NO: 4GTSNRAP SEQ ID NO:9ALWYSNRWV SEQ ID NO:10X 2 is G or SSEQ ID NO:7402TYAMN SEQ ID NO:1RIRSKYNNYATYYADSVKG SEQ ID NO:11HGNFGHSYVSWFAY SEQ ID NO:8GSSTGAVTTGHYAN SEQ ID NO: 4GTX 4 X 5 RAPALWYSNRWV SEQ ID NO:10X 4 X 5 is SN or NKSEQ ID NO:5403TYAMN SEQ ID NO: 1RIRSKYNNYATYYADSVKG SEQ ID NO:11HGNFGHSYVSWFAY SEQ ID NO:8GSSTGAVTTGHYAN SEQ ID NO: 4GTSNRAP SEQ ID NO:9ALWYSNRWV SEQ ID NO:10404TYAMN SEQ ID NO:1RIRSKYNNYATYYADSVKS SEQ ID NO:12HGNFGHSYVSWFAY SEQ ID NO:8GSSTGAVTTGHYAN SEQ ID NO: 4GTSNRAP SEQ ID NO:9ALWYSNRWV SEQ ID NO:10405TYAMN SEQ ID NO: 1RIRSKYNNYATYYAASVKG SEQ ID NO:13HGNFGHSYVSWFAY SEQ ID NO:8GSSTGAVTTGHYAN SEQ ID NO: 4GTNKRAP SEQ ID NO:14ALWYSNLWV SEQ ID NO:15406TYAMN SEQ ID NO: 1RIRSKYNNYATYYAASVKG SEQ ID NO:13HGNFGNSYVSWFAY SEQ ID NO:16GSSTGAVTTGHYAN SEQ ID NO: 4GTNKRAP SEQ ID NO:14ALWYSNLWV SEQ ID NO:15407TYAMN SEQ ID NO:1RIRSKYNNYATYYADSVKG SEQ ID NO:11HGNFGHSYVSWFAY SEQ ID NO:8GSSTGAVTTGHYAN SEQ ID NO: 4GTNKRAP SEQ ID NO:14ALWYSNRWV SEQ ID NO:10 TABLE A2 - CDRs for anti-CD3 binding arm (Chothia) Ab Name VH - CDR1 VH - CDR2 VH - CDR3 VL - CDR1 VL - CDR2 VL - CDR3 408GFTFNTY SEQ ID NO:17SKYNNYGNFGX 3 SYVSWFASSTGAVTTGHY SEQ ID NO: 20GTX 4 WYSNX 6 WSEQ ID NO:18X 3 is H or NX 4 is N or SX 6 is L or RSEQ ID NO:19SEQ ID NO:21SEQ ID NO:22409GFTFNTY SEQ ID NO:17SKYNNY SEQ ID NO:18GNFGHSYVSWFA SEQ ID NO:23SSTGAVTTGHY SEQ ID NO:20GTX 4 WYSNRW SEQ ID NO:25X 4 is N or SSEQ ID NO:21410GFTFNTY SEQ ID NO:17SKYNNY SEQ ID NO:18GNFGHSYVSWFA SEQ ID NO:23SSTGAVTTGHY SEQ ID NO:20GTS SEQ ID NO:24WYSNRW SEQ ID NO:25411GFTFNTY SEQ ID NO:17SKYNNY SEQ ID NO:18GNFGHSYVSWFA SEQ ID NO:23SSTGAVTTGHY SEQ ID NO:20GTN SEQ ID NO:26WYSNLW SEQ ID NO:27412GFTFNTY SEQ ID NO:17SKYNNY SEQ ID NO:18GNFGNSYVSWFA SEQ ID NO:75SSTGAVTTGHY SEQ ID NO:20GTN SEQ ID NO:26WYSNLW SEQ ID NO:27413GFTFNTY SEQ ID NO:17SKYNNY SEQ ID NO:18GNFGHSYVSWFA SEQ ID NO:23SSTGAVTTGHY SEQ ID NO:20GTN SEQ ID NO:26WYSNRW SEQ ID NO:25 TABLE A3 - CDRs for anti-CD3 binding arm (IMGT) Ab Name VH - CDR1 VH - CDR2 VH - CDR3 VL - CDR1 VL - CDR2 VL - CDR3 414GFTFNTYA SEQ ID NO:28IRSKYNNYATVRHGNFGX 3 SYVSWFAYTGAVTTGHY SEQ ID NO:31GTX 4 ALWYSNX 6 WVSEQ ID NO:29X 3 is H or NX 4 is N or SX 6 is L or RSEQ ID NO:30SEQ ID NO:21SEQ ID NO:6415GFTFNTYA SEQ ID NO:28IRSKYNNYAT SEQ ID NO:29VRHGNFGHSYVSWFAY SEQ ID NO:32TGAVTTGHY SEQ ID NO: 31GTX 4 ALWYSNRWV SEQ ID NO:10X 4 is N or SSEQ ID NO:21416GFTFNTYA SEQ ID NO:28IRSKYNNYAT SEQ ID NO:29VRHGNFGHSYVSWFAY SEQ ID NO:32TGAVTTGHY SEQ ID NO: 31GTS SEQ ID NO:24ALWYSNRWV SEQ ID NO:10417GFTFNTYA SEQ ID NO:28IRSKYNNYAT SEQ ID NO:29VRHGNFGHSYVSWFAY SEQ ID NO:32TGAVTTGHY SEQ ID NO: 31GTN SEQ ID NO:26ALWYSNLWV SEQ ID NO:15418GFTFNTYA SEQ ID NO:28IRSKYNNYAT SEQ ID NO:29VRHGNFGNSYVSWFAY SEQ ID NO:33TGAVTTGHY SEQ ID NO: 31GTN SEQ ID NO:26ALWYSNLWV SEQ ID NO:15419GFTFNTYA SEQ ID NO:28IRSKYNNYAT SEQ ID NO:29VRHGNFGHSYVSWFAY SEQ ID NO:32TGAVTTGHY SEQ ID NO: 31GTN SEQ ID NO:26ALWYSNRWV SEQ ID NO:10 TABLE A4 - CDRs for anti-CD3 binding arm (Honegger) Ab Name VH - CDR1 VH - CDR2 VH - CDR3 VL - CDR1 VL - CDR2 VL - CDR3 420ASGFTFNTYA SEQ ID NO:34IRSKYNNYATYYAX 1 SVKX 2 RHGNFGX 3 SYVSWFASSTGAVTTGHY SEQ ID NO: 37GTX 4 NRAPX 7 VPARWYSNX 6 WX 1 is A or DX 3 is H or NX 4 is N or SX 6 is L or RX 2 is G or SSEQ ID NO:36X 7 is G or WSEQ ID NO:22SEQ ID NO:35SEQ ID NO:38421ASGFTFNTYA SEQ ID NO:34IRSKYNNYATYYADSVKX 2 RHGNFGHSYVSWFA SEQ ID NO:40SSTGAVTTGHY SEQ ID NO: 37GTSNRAPGVPAR SEQ ID NO:41WYSNRW SEQ ID NO:25X 2 is G or SSEQ ID NO:39422ASGFTFNTYA SEQ ID NO:34IRSKYNNYATYYADSVKX 2 RHGNFGHSYVSWFA SEQ ID NO:40SSTGAVTTGHY SEQ ID NO: 37GTSNRAPGVPAR SEQ ID NO:41WYSNRW SEQ ID NO:25X 2 is G or SSEQ ID NO:39423ASGFTFNTYA SEQ ID NO:34IRSKYNNYATYYADSVKGR SEQ ID NO:42HGNFGHSYVSWFA SEQ ID NO:40SSTGAVTTGHY SEQ ID NO: 37GTSNRAPGVPAR SEQ ID NO:41WYSNRW SEQ ID NO:25424ASGFTFNTYA SEQ ID NO:34IRSKYNNYATYYADSVKSR SEQ ID NO:43HGNFGHSYVSWFA SEQ ID NO:40SSTGAVTTGHY SEQ ID NO: 37GTSNRAPGVPAR SEQ ID NO:41WYSNRW SEQ ID NO:25425ASGFTFNTYA SEQ ID NO:34IRSKYNNYATYYAASVKGR SEQ ID NO:44HGNFGHSYVSWFA SEQ ID NO:40SSTGAVTTGHY SEQ ID NO: 37GTNKRAPWTPAR SEQ ID NO:45WYSNLW SEQ ID NO:27426ASGFTFNTYA SEQ ID NO:34IRSKYNNYATYYAASVKGR SEQ ID NO:44HGNFGNSYVSWFA SEQ ID NO:46SSTGAVTTGHY SEQ ID NO: 37GTNKRAPWTPAR SEQ ID NO:45WYSNLW SEQ ID NO:27427ASGFTFNTYA SEQ ID NO:34IRSKYNNYATYYADSVKGR SEQ ID NO:42HGNFGHSYVSWFA SEQ ID NO:40SSTGAVTTGHY SEQ ID NO: 37GTNKRAPGVPAR SEQ ID NO:47WYSNRW SEQ ID NO:25

[0045] Additionally, it has been reported that heavy chain variable domains derived from VH3 family germlines can exhibit direct binding to Protein A affinity chromatography resins (Bach, et al., J Chromatogr A. (2015) 1409:60-9). Accordingly, in certain embodiments, to reduce, or substantially or completely eliminate, binding to Protein A by one heavy chain-containing subunit of a multi-specific or bispecific antigen binding molecule, a H435R or H435R+Y436F mutation in the Fc region of the first or second heavy chain can be combined with one or more amino acid substitutions in the VH of the same heavy chain, should that VH region be derived from a human VH3 family germline. In some embodiments, the first VH (and / or a second VH derived from a human VH3 family germline) comprises one or more of the following amino acids at the indicated positions (as described herein, numbering of amino acid positions for substitutions in the first and second VH and VL regions is according to Kabat): the position corresponding to 19 is A, S, T or K; the position corresponding to 57 is A, E or T; the position corresponding to 65 is G, S or T; the position corresponding to 81 is E, K or T; and the position corresponding to 82a is S, T or R. In some embodiments, the first (and / or a second VH derived from a human VH3 family germline) VH comprises one or more of the following amino acids at the indicated positions (position numbering according to Kabat): the position corresponding to 19 is A or S; the position corresponding to 57 is A or E; the position corresponding to 65 is S; and the position corresponding to 81 is E. In some embodiments, the first VH (and / or a second VH derived from a human VH3 family germline) comprises one or more of the following amino acids at the indicated positions (position numbering according to Kabat): the position corresponding to 19 is S and the position corresponding to 57 is A; the position corresponding to 19 is A or S and the position corresponding to 57 is E; the position corresponding to 19 is A and the position corresponding to 57 is E; the position corresponding to 19 is S and the position corresponding to 57 is E; the position corresponding to 19 is S and the position corresponding to 65 is S; the position corresponding to 19 is S and the position corresponding to 81 is E; the position corresponding to 19 is K and the position corresponding to 81 is E; the position corresponding to 57 is A and the position corresponding to 81 is E; the position corresponding to 57 is A and the position corresponding to 65 is S; the position corresponding to 57 is E and the position corresponding to 65 is S; the position corresponding to 57 is E and the position corresponding to 81 is E; the position corresponding to 65 is S and the position corresponding to 81 is E; or the position corresponding to 81 is E and the position corresponding to 82a is S.

[0046] In some embodiments, the first antigen binding domain has reduced or insignificant or substantially no binding to Protein A, or does not detectably bind to Protein A. In some embodiments, the first antigen binding domain binds to Protein A with a K D of greater than 10 -6< M. Protein binding affinity can be determined by any method known in the art, e.g., by surface plasmon resonance (SPR), e.g., using Octet as described herein.

[0047] As used herein, numbering of a given amino acid polymer or nucleic acid polymer "corresponds to", is "corresponding to" or is "relative to" the numbering of a selected or reference amino acid polymer or nucleic acid polymer when the position of any given polymer component (e.g., amino acid, nucleotide, also referred to generically as a "residue") is designated by reference to the same or to an equivalent position (e.g., based on an optimal alignment or a consensus sequence) in the selected amino acid or nucleic acid polymer, rather than by the actual numerical position of the component in the given polymer.

[0048] In some embodiments, the first VH and the first VL comprise one or more of the following amino acid substitutions (numbering according to Kabat): position 81 of the first VH is Q or E; position 83 of the first VH is K or R; position 89 of the first VH is M or V; position 100 of the first VH is H; position 57 of the first VL is G or W; and / or position 75 of the first VL is I or L. In some embodiments, position 81 of the first VH is Q or E. In some embodiments, position 81 of the first VH is E. In some embodiments, position 81 of the first VH is Q. In some embodiments, one or more of position 81 of the first VH is Q or E; position 83 of the first VH is R; position 89 of the first VH is V; position 100 of the first VH is H; position 57 of the first VL is G; and / or position 75 of the first VL is I.

[0049] In some embodiments, the first VH comprises an amino acid sequence that is at least 99% identical to the amino acid sequence of SEQ ID NO: 51. In some embodiments, the first VH comprises the amino acid sequence of SEQ ID NO: 51. In some embodiments, the first VL comprises an amino acid sequence that is at least 99% identical to the amino acid sequence of SEQ ID NO: 56. In some embodiments, the first VL comprises the amino acid sequence of SEQ ID NO: 56..

[0050] In some embodiments, the first VH comprises the amino acid sequence of SEQ ID NO: 51 and the first VL comprises an amino acid sequence that is at least 99% identical to the amino acid sequence of SEQ ID NO: 56. In some embodiments, the first VH comprises an amino acid sequence that is at least 99% identical to the amino acid sequence of SEQ ID NO: 51 and the first VL comprises the amino acid sequence of SEQ ID NO: 56.

[0051] Position 100 of the first VH (numbering according to Kabat) is a histidine (H).

[0052] In some embodiments, the first VH comprises the amino acid sequence of SEQ ID NO: 51 and the first VL comprises an amino acid sequence that is at least 99% identical to the amino acid sequence of SEQ ID NO: 56. In some embodiments, the first VH comprises an amino acid sequence that is at least 99% identical to the amino acid sequence of SEQ ID NO: 51 and the first VL comprises the amino acid sequence of SEQ ID NO: 56. In some embodiments, the first VH and the first VL comprise amino acid sequences that are at least 99% identical to the amino acid sequences set forth, respectively: SEQ ID NOs: 51 and 56. In some embodiments, the first VH and the first VL comprise the amino acid sequences set forth, respectively: SEQ ID NOs: 51 and 56. In some embodiments, the first VH-CDR1, the first VH-CDR2, the first VH-CDR3, the first VL-CDR1, the first VL-CDR2 and the first VL-CDR3 comprise the following amino acid sequences, respectively (according to Kabat): SEQ ID NOs: 1, 12, 8, 4, 9 and 10, and the first VH and the first VL comprise amino acid sequences that are at least 99% identical to the amino acid sequences set forth, respectively: SEQ ID NOs: 51 and 56. In some embodiments, the first VH-CDR1, the first VH-CDR2, the first VH-CDR3, the first VL-CDR1, the first VL-CDR2 and the first VL-CDR3 comprise the following amino acid sequences, respectively (according to Chothia): SEQ ID NOs: 17, 18, 23, 20, 24 and 25, and the first VH and the first VL comprise amino acid sequences that are at least 99% identical to the amino acid sequences set forth, respectively: SEQ ID NOs: 51 and 56. In some embodiments, the first VH-CDR1, the first VH-CDR2, the first VH-CDR3, the first VL-CDR1, the first VL-CDR2 and the first VL-CDR3 comprise the following amino acid sequences, respectively (according to IMGT): SEQ ID NOs: 28, 29, 32, 31, 24 and 10, and the first VH and the first VL comprise amino acid sequences that are at least 99% identical to the amino acid sequences set forth, respectively: SEQ ID NOs: 51 and 56. In some embodiments, the first VH-CDR1, the first VH-CDR2, the first VH-CDR3, the first VL-CDR1, the first VL-CDR2 and the first VL-CDR3 comprise the following amino acid sequences, respectively (according to Honegger): SEQ ID NOs: 34, 43, 40, 37, 41 and 25, and the first VH and the first VL comprise amino acid sequences that are at least 99% identical to the amino acid sequences set forth, respectively: SEQ ID NOs: 51 and 56.

[0053] Illustrative sequences of the VH and VL of the first antigen binding domains of the multi-specific antigen binding molecules, targeting human CD3, are provided in Tables B1 and B2. SEQ ID NO: TABLE B1 -anti-CD3 binding HC variable regions (VH) 48X 1 is A or D;X 2 is G or S;X 3 is H or N;X 8 is E or Q;X 9 is K or R; andX 10 is M or V49X 2 is G or S; andX 8 is E or Q50VH351VH3952VH3453VH1 SEQ ID NO: TABLE B2 - anti-CD3 binding LC variable regions (VL) 54X 4 X 5 is SN or NK;X 6 is L or R;X 7 is G or W; andX 11 is I or L55X 4 X 5 is SN or NK56VL1357VL358VL8

[0054] "Homology" or "identity" or "similarity" as used herein in the context of nucleic acids and polypeptides refers to the relationship between two polypeptides or two nucleic acid molecules based on an alignment of the amino acid sequences or nucleic acid sequences, respectively. Homology and identity can each be determined by comparing a position in each sequence which may be aligned for purposes of comparison. When an equivalent position in the compared sequences is occupied by the same base or amino acid, then the molecules are identical at that position; when the equivalent site occupied by the same or a similar amino acid residue (e.g., similar in steric and / or electronic nature), then the molecules can be referred to as homologous (similar) at that position. Expression as a percentage of homology / similarity or identity refers to a function of the number of identical or similar amino acids at positions shared by the compared sequences. In comparing two sequences, the absence of residues (amino acids or nucleic acids) or presence of extra residues also decreases the identity and homology / similarity.

[0055] As used herein, "identity" means the percentage of identical nucleotide or amino acid residues at corresponding positions in two or more sequences when the sequences are aligned to maximize sequence matching, i.e., taking into account gaps and insertions. Sequences are generally aligned for maximum correspondence over a designated region, e.g., a region at least 20, 25, 30, 35, 40, 45, 50, 55, 60, 65 or more amino acids or nucleotides in length, and can be up to the full length of the reference polypeptide or polynucleotide sequence. For sequence comparison, typically one sequence acts as a reference sequence, to which test sequences are compared. When using a sequence comparison algorithm, test and reference sequences are input into a computer program, subsequence coordinates are designated, if necessary, and sequence algorithm program parameters are designated. Otherwise, standard parameters can be used. The sequence comparison algorithm then calculates the percent sequence identity for the test sequence(s) relative to the reference sequence, based on the designated program parameters.

[0056] When comparing polynucleotide and polypeptide sequences, two sequences are said to be "identical" if the sequence of nucleotides or amino acids in the two sequences is the same when aligned for maximum correspondence, as described below. Comparisons between two sequences are typically performed by comparing the sequences over a comparison window to identify and compare local regions of sequence similarity. A "comparison window" as used herein, refers to a segment of at least 20 contiguous positions, usually 30 to 75, 40 to 50, or the full length of a sequence, in which a sequence may be compared to a reference sequence of the same number of contiguous positions after the two sequences are optimally aligned.

[0057] Optimal alignment of sequences for comparison may be conducted by the local identity algorithm of Smith and Waterman (1981) Add. APL. Math 2:482, by the identity alignment algorithm of Needleman and Wunsch (1970) J. Mol. Biol. 48:443, by the search for similarity methods of Pearson and Lipman (1988) Proc. Natl. Acad. Sci. USA 85: 2444, by computerized implementations of these algorithms (e.g., GAP, BESTFIT, BLAST, FASTA, and TFASTA in the Wisconsin Genetics Software Package, Genetics Computer Group (GCG), 575 Science Dr., Madison, WI), or by inspection.

[0058] One example of algorithms that are suitable for determining percent sequence identity are the Basic Local Alignment Search Tool (BLAST), BLAST 2.0 and PSI-BLAST algorithms, which are described in Altschul, et al., J. Mol. Biol. (1990) 215: 403-410, Altschul, et al., Nucleic Acids Res. (1977) 25: 3389-3402, and Altschul, et al., Nucleic Acids Res. (1997) 25(17):3389-402, respectively. Software for performing BLAST analyses is publicly available through the National Center for Biotechnology Information (blast.ncbi.nlm.nih.govBlast.cgi).

[0059] In one illustrative example, cumulative scores can be calculated using, for nucleotide sequences, the parameters M (reward score for a pair of matching residues; always >0) and N (penalty score for mismatching residues; always <0). Extension of the word hits in each direction are halted when: the cumulative alignment score falls off by the quantity X from its maximum achieved value; the cumulative score goes to zero or below, due to the accumulation of one or more negative-scoring residue alignments; or the end of either sequence is reached. The BLAST algorithm parameters W, T and X determine the sensitivity and speed of the alignment. The BLASTN program (for nucleotide sequences) uses as defaults a word length (W) of 11, and expectation (E) of 10, and the BLOSUM62 scoring matrix (see Henikoff and Henikoff (1989) Proc. Natl. Acad. Sci. USA 89: 10915) alignments, (B) of 50, expectation (E) of 10, M=5, N=-4 and a comparison of both strands.

[0060] For amino acid sequences, a scoring matrix can be used to calculate the cumulative score. Extension of the word hits in each direction are halted when: the cumulative alignment score falls off by the quantity X from its maximum achieved value; the cumulative score goes to zero or below, due to the accumulation of one or more negative-scoring residue alignments; or the end of either sequence is reached. The BLASTP algorithm parameters W, T and X determine the sensitivity and speed of the alignment.

[0061] In one approach, the "percentage of sequence identity" is determined by comparing two optimally aligned sequences over a window of comparison of at least 20 positions, wherein the portion of the polynucleotide or polypeptide sequence in the comparison window may comprise additions or deletions (i.e., gaps) of 20 percent or less, usually 5 to 15 percent, or 10 to 12 percent, as compared to the reference sequences (which does not comprise additions or deletions) for optimal alignment of the two sequences. The percentage is calculated by determining the number of positions at which the identical nucleic acid bases or amino acid residues occur in both sequences to yield the number of matched positions, dividing the number of matched positions by the total number of positions in the reference sequence (i.e., the window size) and multiplying the results by 100 to yield the percentage of sequence identity.

[0062] Residue positions which are not identical can differ by conservative amino acid substitutions. Conservative amino acid substitutions refer to the interchangeability of residues having similar side chains. For example, a group of amino acids having aliphatic side chains is glycine (Gly, G), alanine (Ala, A), valine (Val, V), leucine (Leu, L), and isoleucine (Ile, I); a group of amino acids having aliphatic-hydroxyl side chains is serine (Ser, S) and threonine (Thr, T); a group of amino acids having amide-containing side chains is asparagine (Asn, N) and glutamine (Gln, Q); a group of amino acids having aromatic side chains is phenylalanine (Phe, F), tyrosine (Tyr, Y), and tryptophan (Trp, W); a group of amino acids having basic side chains is lysine (Lys, K), arginine (Arg, R), and histidine (His, H); and a group of amino acids having sulfur-containing side chains is cysteine (Cys, C) and methionine (Met, M). Further, glutamic acid (Glu, E) and aspartic acid (Asp, D) are conservative amino acid substitutions.

[0063] In some embodiments, the first antigen binding domain comprises a Fab and the second antigen binding domain comprises one or more extracellular domains of CD4, e.g., as set forth herein. In some embodiments, the first antigen binding domain comprises a scFv and the second antigen binding domain comprises one or more extracellular domains of CD4, e.g., as set forth herein. Embodiments comprising an scFv can have one or both of the following amino acid substitutions (numbering according to Kabat): (i) a cysteine (C) at position 44 in the scFv variable heavy domain; and (ii) a cysteine (C) at position 100 in the scFv variable light domain.

[0064] In some embodiments, the first antigen binding domain is a scFv comprising a VH and a VL, the scFv comprises an amino acid sequence selected from SEQ ID NO: 63. Illustrative sequences of scFv of the first antigen binding domains of the multi-specific antigen binding molecules, targeting human CD3, are provided in Table C. SEQ ID NO: Ab name TABLE C - scFv for anti-CD3 binding arm 59X 1 is A or D;X 2 is G or S;X 3 is H or N;X 8 is E or Q;X 9 is K or R;X 10 is M or V;X 4 X 5 is SN or NK;X 6 is L or R;X 7 is G or W; andX 11 is I or L60X 2 is G or S;X 4 X 5 is SN or NK; andX 8 is E or Q61X 2 is G or S; andX 8 is E or Q62 3.136339.136434.3651.3663.8

[0065] In some embodiments, the first antigen binding domain, e.g., in scFv or Fab format, binds to CD3 with a K D of lower than 3.0 nM, or lower. In some embodiments, the first antigen binding domain binds to CD3 with a K D of lower than 3.0 nM, or lower. Protein binding affinity can be determined by any method known in the art, e.g., by surface plasmon resonance (SPR), e.g., using Octet as described herein.3. Anti-HIV Antigen Arm

[0066] HIV-1 is the main family of HIV and accounts for 95% of all infections worldwide. HIV-2 is mainly seen in a few West African countries.

[0067] HIV viruses are divided into specific groups, M, N, O and P, of which M is the "major" group and responsible for majority of HIV / AIDS globally. Based on their genetic sequence, Group M is further subdivided into subtypes (also called clades) with prevalence in distinct geographical locations.

[0068] A Group M "subtype" or "clade" is a subtype of HIV-1 group M defined by genetic sequence data. Examples of Group M subtypes include Subtypes A-K. Some of the subtypes are known to be more virulent or are resistant to different medications. There are also "circulating recombinant forms" or CRFs derived from recombination between viruses of different subtypes, which are each given a number. CRF12_BF, for example, is a recombination between subtypes B and F. Subtype A is common in West Africa. Subtype B is the dominant form in Europe, the Americas, Japan, Thailand, and Australia. Subtype C is the dominant form in Southern Africa, Eastern Africa, India, Nepal, and parts of China. Subtype D is generally only seen in Eastern and central Africa. Subtype E has never been identified as a nonrecombinant, only recombined with subtype A as CRF01_AE. Subtype F has been found in central Africa, South America and Eastern Europe. Subtype G (and the CRF02_AG) have been found in Africa and central Europe. Subtype H is limited to central Africa. Subtype I was originally used to describe a strain that is now accounted for as CRF04_cpx, with the cpx for a "complex" recombination of several subtypes. Subtype J is primarily found in North, Central and West Africa, and the Caribbean Subtype K is limited to the Democratic Republic of Congo and Cameroon. These subtypes are sometimes further split into sub-subtypes such as A1 and A2 or F1 and F2. In 2015, the strain CRF19, a recombinant of subtype A, subtype D, and subtype G, with a subtype D protease was found to be strongly associated with rapid progression to AIDS in Cuba.

[0069] This disclosure provides, inter alia, multi-specific antigen binding molecules comprising second antigen binding domains derived from human anti-HIV neutralizing antibodies (e.g., broadly neutralizing Abs) that target the gp120 polypeptide on the surface of HIV-infected cells. Neutralizing antibodies against viral envelope proteins provide adaptive immune defense against HIV-1 exposure by blocking the infection of susceptible cells. Broad neutralization indicates that the antibodies can neutralize HIV-1 isolates from different clades. Thus, the multi-specific antigen binding molecules described herein have cross-clade binding activity.gp120

[0070] Envelope glycoprotein gp120 (or gp120) is a 120 kDa glycoprotein that is part of the outer layer of HIV. It presents itself as viral membrane spikes consisting of three molecules of gp120 linked together and anchored to the membrane by gp41 protein. Gp120 is essential for viral infection as it facilitates HIV entry into the host cell through its interaction with cell surface receptors. These receptors include DC-SIGN, Heparan Sulfate Proteoglycan, and the CD4 receptor. Binding to CD4 on helper T-cells induces the start of a cascade of conformational changes in gp120 and gp41 that lead to the fusion of the virus with the host cell membrane.

[0071] Gp120 is encoded by the HIV env gene. The env gene encodes a gene product of around 850 amino acids. The primary env product is the protein gp160, which gets cleaved to gp120 (about 480 amino acids) and gp41 (about 345 amino acids) in the endoplasmic reticulum by the cellular protease furin.

[0072] The amino acid sequence of an exemplary gp160 polypeptide of HIV clone WITO is provided below (the V3 hypervariable loop is boldened and the N332 potential N-linked glycosylation site is boldened and underlined):

[0073] The amino acid sequence of an exemplary gp160 polypeptide of HIV clone identified in NCBI Ref Seq No. NP_057856.1 is provided below (the V3 hypervariable loop is boldened and the N332 potential N-linked glycosylation site is boldened and underlined):

[0074] The amino acid sequence of an exemplary gp120 polypeptide of HXB2 subtype B HIV-1 isolate (GenBank Accession No. K0345; corresponding to residues 1-511 of NCBI Ref Seq No. NP_057856.1) is provided below (the V3 hypervariable loop is boldened and the N332 potential N-linked glycosylation site is boldened and underlined; signal peptide is underlined):

[0075] The amino acid sequence of an exemplary gp120 polypeptide is provided below:

[0076] The amino acid sequence of another exemplary gp120 polypeptide (see, bioafrica.net / proteomics / ENV-GP120prot.html) is provided below:

[0077] Genomic diversity among independent human immunodeficiency virus type 1 (HIV-1) isolates, to a lesser degree among sequential isolates from the same patients, and even within a single patient isolate is a well-known feature of HIV-1. Although this sequence heterogeneity is distributed throughout the genome, most of the heterogeneity is located in the env gene. Comparison of predicted amino acid sequences from several different isolates has shown that sequence heterogeneity is clustered in five variable regions (designated V1 through V5) of the surface glycoprotein, gp120. The V3 region, although only 35 amino acids long, exhibits considerable sequence variability. Interestingly, in spite of this variability, the V3 region includes determinants that mediate interactions with CD4 +< cells. The increase in gp120 variability results in higher levels of viral replication, suggesting an increase in viral fitness in individuals infected by diverse HIV-1 variants. Variability in potential N-linked glycosylation sites (PNGSs) also result in increased viral fitness. PNGSs allow for the binding of long-chain carbohydrates to the high variable regions of gp120. Thus, the number of PNGSs in env might affect the fitness of the virus by providing more or less sensitivity to neutralizing antibodies.

[0078] A consensus sequence of the V3 region of gp120 (Milich et al., J Virol., (1993) 67(9):5623-5634) is provided below: CTRPNNNTRKSIHIGPGRAFYTTGEIIGDIRQAHC (SEQ ID NO: 72).Anti-gp120 and Anti-gp41 Antigen Binding Domains

[0079] This disclosure features anti-gp120 or anti-gp41 multi-specific antigen binding molecules comprising a second antigen binding domain that targets and binds to gp120 or gp41, respectively. In certain embodiments, these multi-specific antigen binding molecules bind to HIV-1 antigens expressed on a cell surface and eliminate or kill the infected cell.

[0080] The second antigen binding domains in the herein described multi-specific antigen binding molecules can be derived from human neutralizing antibodies (e.g., monoclonal) that target HIV-1. A "neutralizing antibody" is one that can neutralize the ability of HIV to at least one of initiate and perpetuate an infection in at least one of a host in vivo and in target cells in vitro. The disclosure provides neutralizing monoclonal human antibodies, wherein the antibody recognizes an antigen from HIV, e.g., a gp120 polypeptide. A "neutralizing antibody" may inhibit the entry of HIV-1 virus, e.g., SF162 and / or JR-CSF, with a neutralization index >1.5 or >2.0 (Kostrikis LG et al., J. Virol, 70(1): 445-458 (1996)).

[0081] The second antigen binding domains in the herein described multi-specific antigen binding molecules can be derived from human broadly neutralizing antibodies (e.g., monoclonal) that target HIV-1. By "broadly neutralizing antibodies" are meant antibodies that neutralize more than one HIV-1 virus species (from diverse clades and different strains within a clade) in a neutralization assay. A broad neutralizing antibody may neutralize at least 2, 3, 4, 5, 6, 7, 8, 9 or more different strains of HIV-1, the strains belonging to the same or different clades. A broad neutralizing antibody may neutralize multiple HIV-1 species belonging to at least 2, 3, 4, 5, or 6 different clades. The inhibitory concentration of the multi-specific antigen binding molecule may be less than 0.0001 µg / ml, less than 0.001 µg / ml, less than 0.01 µg / ml, less than 0.1 µg / ml, less than 0.5 µg / ml, less than 1.0 µg / ml, less than 5 µg / ml, less than 10 µg / ml, less than 25 µg / ml, less than 50 µg / ml, or less than 100 µg / ml to neutralize 50% of the input virus in the neutralization assay.

[0082] The second antigen binding domain of the multi-specific antigen binding molecules may bind to an epitope or region of gp120 selected from the group consisting of: (i) third variable loop (V3) (e.g., high mannose patch) comprising a N332 oligomannose glycan; (ii) second variable loop (V2) (e.g., Env trimer apex); (iii) CD4 binding site (CD4bs); (iv) gp120 / gp41 interface; or (v) silent face of gp120. The foregoing epitopes or regions of gp120 bound by broadly neutralizing antibodies are described, e.g., in McCoy, Retrovirology (2018) 15:70; Sok and Burton, Nat Immunol. 2018 19(11):1179-1188; Possas, et al., Expert Opin Ther Pat. 2018 Jul;28(7):551-560; and Stephenson and Barouch, Curr HIV / AIDS Rep (2016) 13:31-37.

[0083] Additional broadly neutralizing antibodies that bind to gp120 in the third variable loop (V3) (e.g., high mannose patch) comprising a N332 oligomannose glycan are described, e.g., in WO 2012 / 030904; WO 2014 / 063059; WO 2016 / 149698; WO 2017 / 106346; WO 2018 / 075564, WO 2018 / 125813; WO 2018 / 237148, WO 2019 / 226829, WO 2020 / 023827, WO2020 / 056145 and Kerwin, et al., J Pharm Sci. 2020 Jan;109(1):233-246.

[0084] Additional broadly neutralizing antibodies that bind to gp120 in the second variable loop (V2) (e.g., Env trimer apex) are described, e.g., in WO 2010 / 107939; WO 2012 / 030904; WO 2018 / 075564 and WO 2018 / 125813.gp120 CD4 Binding Site

[0085] Antibody variants described herein bind to the CD4 binding site (CD4bs) of HIV gp120. The CD4 binding site (CD4bs) involves structurally conserved sites located within the β1-α1, loop D, β20-β21 (bridging sheet) and β24-α5 of gp120, which determine the CD4 binding and are involved in the epitopes of CD4bs-directed antibodies (Qiao, et al., Antiviral Res. 2016 Aug;132:252-61). The CD4bs of gp120 forms conformational epitopes recognized by anti-CD4bs antibodies involving one or more amino acid residues selected from Thr278, Asp279, Ala281, Thr283, Asp368, Trp427, Glu460, Ser461, Glu462, Leu452, Leu453 and Arg476. The amino acid residues and position numbering is with reference to HXB2 subtype B HIV-1 isolate, which corresponds to residues 1-511 of NCBI Ref Seq No. NP_057856.1, provided below. Residues Thr278, Asp279, Asn280, Ala281, Thr283, Asp368, Trp427, Leu452, Leu453, Gly459, Glu464, Ser465, Glu466, Ile467, Gly472, Gly473 and Arg476, which can contribute to the gp120 CD4bs, are boldened and underlined:

[0086] Tridimensional models depicting amino acid residues contributing to the gp120 CD4bs are provided, e.g., in Canducci, et al., Retrovirology. 2009 Jan 15;6:4; Falkowska, et al., J Virol. 2012 Apr;86(8):4394-403; and Li, et al., J. Virol. 2012 Oct;86(20):11231-41; Gristick, et al., Nat Struct Mol Biol. 2016 Oct;23(10):906-915; Kwon, et al., Nat Struct Mol Biol. 2015 Jul;22(7):522-31; Liu, et al., Nat Struct Mol Biol. 2017 Apr;24(4):370-378; Chen, et al., Science. 2009 Nov 20;326(5956):1123-7 and Lyumkis, et al., Science. 2013 Dec 20;342(6165):1484-90. In some embodiments, the second antigen binding domain binds to an epitope or region of gp120 in the CD4 binding site (CD4bs) and competes with or comprises VH and VL regions from an antibody selected from the group consisting of 3BNC117, GS-9723, GS-5423, 3BNC60, b12, CH103, 1NC9, 12A12, VRC01, VRC07, VRC07-523, N6, NIH45-46 and PGV04 (a.k.a., VRC-PG04) for binding to gp120 CD4bs. In some embodiments, the antibody variants described herein bind to an overlapping or identical epitope to the epitope bound by one or more of anti-CD4bs antibodies 3BNC117, GS-9723, GS-5423, 3BNC60, b12, CH103, 1NC9, 12A12, VRC01, VRC07, VRC07-523, N6, NIH45-46 and PGV04 (a.k.a., VRC-PG04).

[0087] Additional broadly neutralizing antibodies that bind to gp120 in the CD4 binding site (CD4bs) are described, e.g., in WO 2011 / 038290; WO 2012 / 158948; WO 2013 / 016468; WO 2013 / 192589; WO 2013 / 086533; WO 2015 / 128846; WO 2016 / 149698; WO 2016 / 149695; WO 2018 / 075564; WO 2018 / 125813; WO 2018 / 237357, WO 2020 / 010107, WO 2020 / 086446 and U.S. Patent Nos. 9,493,549 and 9,879,068.

[0088] In some embodiments, the second antigen binding domain binds to an epitope or region of gp120 in the CD4 binding site (CD4bs) and competes with or comprises VH and VL regions from 3BNC117 and comprises Phe-Asp-Phe-Asp (FDFD) (SEQ ID NO: 1040) at positions 74a, 74b, 74c, and 74d of the VH (position numbering according to Kabat). In some embodiments, the second antigen binding domain binds to an epitope or region of gp120 in the CD4 binding site (CD4bs) and competes with or comprises VH and VL regions from 3BNC117 and comprises a framework region 3 (FR3) of the VH comprising the following amino acid sequence RVSLTRHASFDFDTFSFYMDLKALRSDDTAVYFCAR (SEQ ID NO: 74). Crystallographic studies have shown that framework region 3 at VH Kabat position numbers 74a, 74b, 74c and 74d form part of the paratope of 3BNC117 variants, directly contacting the antigen target, gp120. See, e.g., Lee, et al., Immunity (2017) 46(4): 690-702 (Figure 1G, identifying residue W71d); Klein, et al., Cell. (2013) 153(1):126-38 (Figures 4 and 5); and Zhou, et al., (2013) Immunity (2013) 39 245-258 (Table 1); ribbon diagrams of crystallized structures of 5V8L, 5V8M, 4JPV and 4LSV can be viewed at rcsb.org. In some embodiments, at least 50%, at least 60%, at least 70%, least 80%, at least 85%, at least 90%, or more, N-linked glycosylation sites in at least one of the first VH, the first VL, the second VH and the second VL are sialylated. In some embodiments, the N-linked glycosylation sites in at least one of the first VH, the first VL, the second VH and the second VL have a sialic acid occupancy (e.g., a glycan comprising one or two terminal sialic acid residues) of at least 40%, at least 50%, at least 60%, at least 70%, least 80%, at least 85%, at least 90%, or more. In some embodiments, the N-linked glycosylation site at VL asparagine amino acid position 72 (N72) according to Kabat numbering is sialylated. In some embodiments, the sialylated N-linked glycosylation sites in at least one of the first VH, the first VL, the second VH and the second VL comprise from 1 to 5 sialic acid residues, e.g., from 1 to 4 sialic acid residues, e.g., from 1 to 3 sialic acid residues, e.g., from 1 to 2 sialic acid residues. In some embodiments, at least one of the first VH, the first VL, the second VH and the second VL are sialylated with N-acetylneuraminic acid (NANA). In some embodiments, the sialic acid residues are present in biantennary structures. In some embodiments, the sialic acid residues are present in complex N-linked glycan structures. In some embodiments, the sialic acid residues are present in hybrid N-linked glycan structures.

[0089] In some embodiments, the second antigen binding domain binds to an epitope or region of gp120 in the CD4 binding site (CD4bs) and comprises an EC domain of CD4 (e.g., domain 1 (D1), D1-D1 (tandem), D1-D2, D1-D2-D3-D3, or D1-D2-D3-D4). Illustrative CD4 extracellular domains that can be used in the herein described multi-specific antigen binding molecules are described, e.g., in WO2011146891, WO2014150748 and WO2016153572. The one or more EC domains of CD4 comprises an amino acid sequence that is at least 99% identical, or 100% identical, to a CD4 EC domain selected from: (i) see, e.g., Chen, et al., J Virol. 2014 Jan;88(2):1125-39); or (ii)

[0090] In some embodiments, the EC domain of CD4 comprises a sequence that is at least 99% identical to the sequence of SEQ ID NO: 746. In some embodiments, the EC domain of CD4 comprises the sequence of SEQ ID NO: 746.

[0091] In some embodiments, the first antigen binding domain comprises a first VH-CDR1, a first VH-CDR2, a first VH-CDR3, a first VL-CDR1, a first VL-CDR2 and a first VL-CDR3 comprising the following amino acid sequences, respectively (according to Kabat): SEQ ID NOs: 1, 12, 8, 4, 9 and 10; and the second antigen binding domain comprises one EC domain of CD4 comprising an amino acid sequence that is at least 99% identical to a CD4 EC domain of SEQ ID NO: 746. In some embodiments, the first antigen binding domain comprises a first VH-CDR1, a first VH-CDR2, a first VH-CDR3, a first VL-CDR1, a first VL-CDR2 and a first VL-CDR3 comprising the following amino acid sequences, respectively (according to Kabat): SEQ ID NOs: 1, 12, 8, 4, 9 and 10; and the second antigen binding domain comprises one EC domain of CD4 comprising the amino acid sequence of SEQ ID NO: 746.

[0092] In some embodiments, the first antigen binding domain comprises a first VH and a first VL comprising the amino acid sequences set forth, respectively: SEQ ID NOs: 51 and 56; and the second antigen binding domain comprises one EC domain of CD4 comprising an amino acid sequence that is at least 99% identical to a CD4 EC domain of SEQ ID NO: 746. In some embodiments, the first antigen binding domain comprises a first VH and a first VL comprising amino acid sequences that are at least 99% identical to the amino acid sequences set forth, respectively: SEQ ID NOs: 51 and 56; and the second antigen binding domain comprises one EC domain of CD4 comprising an amino acid sequence that is at least 99% identical to a CD4 EC domain of SEQ ID NO: 746. In some embodiments, the first antigen binding domain comprises a first VH comprising the amino acid sequence of SEQ ID NO: 51 and a first VL comprising an amino acid sequence that is at least 99% identical to the amino acid sequence of SEQ ID NO: 56; and the second antigen binding domain comprises one EC domain of CD4 comprising an amino acid sequence that is at least 99% identical to a CD4 EC domain of SEQ ID NO: 746. In some embodiments, the first antigen binding domain comprises a first VH comprising an amino acid sequence that is at least 99% identical to the amino acid sequence of SEQ ID NO: 51 and a first VL comprising the amino acid sequence of SEQ ID NO: 56; and the second antigen binding domain comprises one EC domain of CD4 comprising an amino acid sequence that is at least 99% identical to a CD4 EC domain of SEQ ID NO: 746. In some embodiments, the first antigen binding domain comprises a first VH and a first VL comprising the amino acid sequences set forth, respectively: SEQ ID NOs: 51 and 56; and the second antigen binding domain comprises one EC domain of CD4 comprising an amino acid sequence of SEQ ID NO: 746.

[0093] Additional broadly neutralizing antibodies that bind to gp120 in the gp120 / gp41 interface are described, e.g., in WO 2011 / 038290; WO 2012 / 030904 and WO2017 / 079479.

[0094] Additional broadly neutralizing antibodies that bind to gp41 in the MPER are described, e.g., in WO 2011 / 034582; WO 2011 / 038290; WO 2011 / 046623 and WO 2013 / 070776.

[0095] Additional broadly neutralizing antibodies are described, e.g., in U.S. Patent Nos. 8,673,307; 9,493,549; 9,783,594; 10,239,935; and patent publications numbers US2018371086, US2020223907, WO 2012 / 154312; WO2012 / 158948; WO 2013 / 086533; WO 2013 / 142324; WO2014 / 063059; WO 2014 / 089152, WO 2015 / 048462; WO 2015 / 103549; WO 2015 / 117008; WO2016 / 014484; WO 2016 / 154003; WO 2016 / 196975; WO 2016 / 149710; WO2017 / 096221; WO 2017 / 133639; and WO 2017 / 133640. Additional examples include those described in Sajadi, et al., Cell. (2018) 173(7):1783-1795; Sajadi, et al., J Infect Dis. (2016) 213(1):156-64; Klein et al., Nature, 492(7427): 118-22 (2012), Horwitz et al., Proc Natl Acad Sci U S A, 110(41): 16538-43 (2013), Scheid, et al., Science, 333 : 1633-1637 (2011), Scheid, et al., Nature, 458:636-640 (2009), Eroshkin et al, Nucleic Acids Res., 42 (Database issue):Dl 133-9 (2014), Mascola et al., Immunol Rev., 254(l):225-44 (2013), such as 2F5, 4E10, M66.6, CAP206-CH12, 10E8, 10E8v4, 10E8-5R-100cF, DH511.11P, 7b2, and LN01 (all of which bind the MPER of gp41); PG9, PG16, CAP256, CAP256-VRC26, CAP256-VRC26.25, CH01-04 (all of which bind V1V2-glycan), 2G12 (which binds to outer domain glycan); b12, F105, VRC01, VRC07, VRC07-523, VRC03, VRC06, VRC06b01 VRC08, VRC0801, NIH45-46, 3BNC117, 3BNC60, PGV04 (a.k.a., VRC-PG04); CH103, 44-VRC13.01, 1NC9, 12A12, N6, 1-18, N49-P7, NC-Cow1, IOMA, CH235 and CH235.12, N49P6, N49P7, N49P11, N49P9 and N60P25 (all of which bind to the CD4 binding site).

[0096] Illustrative sequences of complementarity determining regions (CDRs) of the first antigen binding domains of the multi-specific antigen binding molecules, targeting HIV gp120, are provided in Tables D1, D2, D3 and D4. Illustrative sequences of the VH and VL of the second antigen binding domains of the multi-specific antigen binding molecules, targeting HIV gp120, are provided in Table E. Table D1- CDRs (Kabat) for illustrative anti-HIV antigen binding arm Ab Name VH - CDR1 VH - CDR2 VH - CDR3 VL - CDR1 VL - CDR2 VL - CDR3 1DSYWS SEQ ID NO:76YVHKSGDTNYSPSLKS SEQ ID NO:77GEKSLGSRAVQ SEQ ID NO:79NNQDRPS SEQ ID NO:80HIWDSRVPTKWV SEQ ID NO:812DSYWS SEQ ID NO:76YVHKSGDTNYNPSLKS SEQ ID NO:82GEKSLGSRAVQ SEQ ID NO:79NNQDRPS SEQ ID NO:80HIWDSRVPTKWV SEQ ID NO:813NYYWT SEQ ID NO:83YISDRESATYNPSLNS SEQ ID NO:84GRQALGSRAVQ SEQ ID NO:86NNQDRPS SEQ ID NO:80HMWDSRSGFSWS SEQ ID NO:874NYYWT SEQ ID NO:83YISDRETTTYNPSLNS SEQ ID NO:88GRQALGSRAVQ SEQ ID NO:86NNQDRPS SEQ ID NO:80HMWDSRSGFSWS SEQ ID NO:875GRFWS SEQ ID NO:90YFSDTDRSEYNPSLRS SEQ ID NO:91GERSRGSRAVQ SEQ ID NO:93NNQDRPA SEQ ID NO:94HYWDSRSPISWI SEQ ID NO:956GRFWS SEQ ID NO:90YFSDTDRSEYNPSLRS SEQ ID NO:91GERSRGSRAVQ SEQ ID NO:93NNQDRPA SEQ ID NO:94HYWDSRSPISWI SEQ ID NO:957DNYWS SEQ ID NO:97YVHDSGDTNYNPSLKS SEQ ID NO:98GEESLGSRSVI SEQ ID NO:100NNNDRPS SEQ ID NO:101HIWDSRRPTNWV SEQ ID NO:1028DAYWS SEQ ID NO:103YVHHSGDTNYNPSLKR SEQ ID NO:104GKESIGSRAVQ SEQ ID NO:106NNQDRPA SEQ ID NO:94HIYDARGGTNWV SEQ ID NO:1079ACTYFWG SEQ ID NO:108NGTATNFVS SEQ ID NO:111GVDKRPP SEQ ID NO:112GSLVGNWDVI SEQ ID NO:11310ACDYFWG SEQ ID NO:114 TGTSNRFVS SEQ ID NO:117GVNKRPS SEQ ID NO: 118SSLVGNWDVI SEQ ID NO: 11911ACDYFWG SEQ ID NO:114TGNINNFVS SEQ ID NO:122GVNKRPS SEQ ID NO:118GSLAGNWDVV SEQ ID NO:12312ACNSFWG SEQ ID NO:124TGTSNNFVS SEQ ID NO:127DVNKRPS SEQ ID NO:128GSLVGNWDVI SEQ ID NO:11313GCDYFWG SEQ ID NO:129TGTSNNFVS SEQ ID NO:127GVNKRPS SEQ ID NO:118GSLVGNWDVI SEQ ID NO:11314TGHYYWG SEQ ID NO:132HIHYTTAVLHNPSLKS SEQ ID NO:133NGTSSDIGGWNFVS SEQ ID NO:135EVNKRPS SEQ ID NO:136SSLFGRWDVV SEQ ID NO:13715GTDWGENDFHYG SEQ ID NO:138SIHWRGRTTHYKTSFR S SEQ ID NO:139RASQNVKNNLA SEQ ID NO:141DASSRAG SEQ ID NO:142QQYEEWPRT SEQ ID NO:14316GGEWGDSDYHWG SEQ ID NO:144SIHWRGTTHYNAPFRG SEQ ID NO:145RASQSVKNNLA SEQ ID NO:147DTSSRAS SEQ ID NO:148QQYEEWPRT SEQ ID NO:14317GGEWGDKDYHWG SEQ ID NO:149SIHWRGTTHYKESLRR SEQ ID NO:150RASQNINKNLA SEQ ID NO:152ETYSKIA SEQ ID NO:153QQYEEWPRT SEQ ID NO:14318SDHSWT SEQ ID NO:154DIHYNGATTYNPSLRS SEQ ID NO:155SGAPLTSRFTY SEQ ID NO:157RSSQRSS SEQ ID NO:158QSSDTSDSYKM SEQ ID NO:15919DYFIH SEQ ID NO:160QANGYLN SEQ ID NO:163DGSKLER SEQ ID NO:164QVYEF SEQ ID NO: 16520DHFIH SEQ ID NO:166QRSDFWDFDV SEQ ID NO:167QANGYLN SEQ ID NO:163DGSKLER SEQ ID NO:164QVYEF SEQ ID NO: 16521NCPIN SEQ ID NO:168RTSQYGSLA SEQ ID NO:171SGSTRAA SEQ ID NO:172QQYEF SEQ ID NO: 17322NCPIN SEQ ID NO:168RTSQYGSLA SEQ ID NO:171SGSTRAA SEQ ID NO:172QQYEF SEQ ID NO: 17323DCTLN SEQ ID NO:175WLKPRGGAVNYARPLQ GSEQ ID NO:176GKNCDYNWDFEH SEQ ID NO:177RTSQYGSLA SEQ ID NO:171SGSTRAA SEQ ID NO:172QQYEF SEQ ID NO: 17324AHILF SEQ ID NO:178DRSYGDSSWALDA SEQ ID NO:180QTSQGVGSDLH SEQ ID NO:181HTSSVED SEQ ID NO:182QVLQF SEQ ID NO: 18325DDDTFTKYWTH SEQ ID NO:902RASQGLDSSHLA SEQ ID NO:905GTSNRAR SEQ ID NO:906QRYGGTPIT SEQ ID NO:90726RTELIH SEQ ID NO:908QKFYTGGQGWYFDL SEQ ID NO:910TAASYGHMT SEQ ID NO:911ATSKRAS SEQ ID NO:912QQLEF SEQ ID NO: 91327SGFDFSRQGMH SEQ ID NO:184FIKYDGSEKYHADSVW SEQ ID NO:185NGTSNDVGGYESVS SEQ ID NO:187KSLTSTRRRV SEQ ID NO:18928SGFTFHKYGMH SEQ ID NO:190LISDDGMRKYHSDSMW SEQ ID NO:191NGTSSDVGGFDSVS SEQ ID NO:193SSLTDRSHRI SEQ ID NO:19529DYYLH SEQ ID NO:196GAVGADSGSWFDP SEQ ID NO:198SGSKLGDKYVS SEQ ID NO:199ENDRRPS SEQ ID NO:200QAWETTTTTFV SEQ ID NO:20130SYAFS SEQ ID NO:202RASQTIHTYLN SEQ ID NO:205GASTLQS SEQ ID NO:206QQSYSTPRT SEQ ID NO:20731NHDVH SEQ ID NO:208LATHRAS SEQ ID NO: 212MQGLHSPWT SEQ ID NO:21332KYDVH SEQ ID NO:214LGSQRAS SEQ ID NO:218MQGLNRPWT SEQ ID NO:21933KYDVH SEQ ID NO:214LGSQRAS SEQ ID NO:218MQGLNRPWT SEQ ID NO:21934KYDVH SEQ ID NO:214LGSQRAS SEQ ID NO:218MQGLNRPWT SEQ ID NO:21935GYGMH SEQ ID NO:914SGNTSNIGNNFVS SEQ ID NO:917ETDKRPS SEQ ID NO:918ATWAASLSSARV SEQ ID NO:91936KYPMY SEQ ID NO:223EVSNRFS SEQ ID NO:227MQSKDFPLT SEQ ID NO:22837GLYAVN SEQ ID NO:229QIWRWKSSASHHFRG SEQ ID NO:230RASQSITGNWVA SEQ ID NO:232RGAALLG SEQ ID NO:233QQYDTYPGT SEQ ID NO:23438NDNYYWA SEQ ID NO:920TIYYSGTTYYNPSLRN SEQ ID NO:921RASQSVTKYLN SEQ ID NO:923GTYTLLS SEQ ID NO:924QQAHSTPWT SEQ ID NO:925 Table D2 - CDRs (Chothia) for illustrative anti-HIV antigen binding arm Ab Name VH - CDR1 VH - CDR2 VH - CDR3 VL - CDR1 VL - CDR2 VL - CDR3 39GASISD SEQ ID NO:235KSG SEQ ID NO:236EKSLGSRA SEQ ID NO:238NNQ SEQ ID NO: 239WDSRVPTKW SEQ ID NO:24040GDSMNNY SEQ ID NO:241DRE SEQ ID NO:242RQALGSRA SEQ ID NO:244NNQ SEQ ID NO: 239WDSRSGFSW SEQ ID NO:24541GGSISNY SEQ ID NO:246DRE SEQ ID NO:242RQALGSRA SEQ ID NO:244NNQ SEQ ID NO: 239WDSRSGFSW SEQ ID NO:24542NGSVSGR SEQ ID NO:248DTD SEQ ID NO:249ERSRGSRA SEQ ID NO:251NNQ SEQ ID NO: 239WDSRSPISW SEQ ID NO:25243NGSVSGR SEQ ID NO:248DTD SEQ ID NO:249ERSRGSRA SEQ ID NO:251NNQ SEQ ID NO: 239WDSRSPISW SEQ ID NO:25244GTLVRDN SEQ ID NO:254DSG SEQ ID NO:255EESLGSRS SEQ ID NO:257NNN SEQ ID NO: 258WDSRRPTNW SEQ ID NO:25945GASINDA SEQ ID NO:260HSG SEQ ID NO:261KESIGSRA SEQ ID NO:263NNQ SEQ ID NO: 239YDARGGTNW SEQ ID NO:26446GESTGACTY SEQ ID NO:265HCQSFWGSG SEQ ID NO:266DGEVLVYNHWPKPAWVD SEQ ID NO:267GTATNF SEQ ID NO:268GVD SEQ ID NO: 269LVGNWDV SEQ ID NO:27047GDSTAACDY SEQ ID NO: 271HCAGYYNTG SEQ ID NO:272DGEVLVYHDWPKPAWVD SEQ ID NO:273GTSNRF SEQ ID NO:274GVN SEQ ID NO:275LVGNWDV SEQ ID NO:27048GDSTAACDY SEQ ID NO:271HCAGYYNSG SEQ ID NO:276GGDVLVYHDWPKPAWVD SEQ ID NO:277GNINNF SEQ ID NO:278GVN SEQ ID NO:275LAGNWDV SEQ ID NO:27949GDSTAACNS SEQ ID NO:280HCASYWNRG SEQ ID NO:281GGEVLRYTDWPKPAWVD SEQ ID NO:282GTSNNF SEQ ID NO:283DVN SEQ ID NO:284LVGNWDV SEQ ID NO:27050GDSTAGCDY SEQ ID NO:285HCAGYYNTG SEQ ID NO:272DGEVLVYNDWPKPAWVD SEQ ID NO:286GTSNNF SEQ ID NO:283GVN SEQ ID NO:275LVGNWDV SEQ ID NO:27051GESINTGHY SEQ ID NO:287YTT SEQ ID NO:288GGDILYYYEWQKPHWFS SEQ ID NO:289GTSSDIGGWNF SEQ ID NO:290EVN SEQ ID NO: 291LFGRWDV SEQ ID NO:29252WRGR SEQ ID NO:294KYHDIFRVVPVAGWFD SEQ ID NO:295SQNVKNN SEQ ID NO:296DAS SEQ ID NO: 297YEEWPR SEQ ID NO:29853WRG SEQ ID NO:300KYHDIVMVVPIAGWFD SEQ ID NO:301SQSVKNN SEQ ID NO:302DTS SEQ ID NO:303YEEWPR SEQ ID NO:29854 WRG SEQ ID NO:300RHHDVFMLVPIAGWFD SEQ ID NO:305SQNINKN SEQ ID NO:306ETY SEQ ID NO:307YEEWPR SEQ ID NO:29855QDSRPSDH SEQ ID NO:308YNG SEQ ID NO:309GAPLTSRF SEQ ID NO:311RSS SEQ ID NO: 312SDTSDSYK SEQ ID NO:31356GYNIRDY SEQ ID NO:314PKTG SEQ ID NO:315RSDYWDFD SEQ ID NO:316NGY SEQ ID NO:317DGS SEQ ID NO:318YE SEQ ID NO:32757GYKISDH SEQ ID NO:320PKTG SEQ ID NO:315RSDFWDFD SEQ ID NO:321NGY SEQ ID NO:317DGS SEQ ID NO:318YE SEQ ID NO:32758GYEFINC SEQ ID NO:322PRGG SEQ ID NO:323KYCTARDYYNWDFE SEQ ID NO:324SQYGS SEQ ID NO:325SGS SEQ ID NO:326YE SEQ ID NO:32759GYEFINC SEQ ID NO:322PRHG SEQ ID NO:328KYCTARDYYNWDFE SEQ ID NO:324SQYGS SEQ ID NO:325SGS SEQ ID NO:326YE SEQ ID NO:32760GYEFIDC SEQ ID NO:329PRGG SEQ ID NO:323KNCDYNWDFE SEQ ID NO:330SQYGS SEQ ID NO:325SGS SEQ ID NO:326YE SEQ ID NO:32761GYTFTAH SEQ ID NO:331PQYG SEQ ID NO:332RSYGDSSWALD SEQ ID NO:333SQGVGSD SEQ ID NO:334HTS SEQ ID NO:335LQ SEQ ID NO:33662DDPYTDDDTFTKY SEQ ID NO:926PHFA SEQ ID NO:927PFGDRAPHYNYHMD SEQ ID NO:928SQGLDSSH SEQ ID NO:929GTS SEQ ID NO: 930YGGTPI SEQ ID NO:93163EDIFERTE SEQ ID NO:932TVTG SEQ ID NO:933KFYTGGQGWYFD SEQ ID NO:934ASYGH SEQ ID NO:935ATS SEQ ID NO: 936LE SEQ ID NO:93764GFDFSRQ SEQ ID NO:337YDGS SEQ ID NO:338GTSNDVGGYES SEQ ID NO:340DVS SEQ ID NO:341LTSTRRR SEQ ID NO:34265GFTFHKY SEQ ID NO:343DDGM SEQ ID NO:344GTSSDVGGFDS SEQ ID NO:346DVS SEQ ID NO:341LTDRSHR SEQ ID NO:34766GYSFIDY SEQ ID NO:348PENG SEQ ID NO:349AVGADSGSWFD SEQ ID NO:350GSKLGDKY SEQ ID NO:351END SEQ ID NO:352WETTTTTF SEQ ID NO:35367GGAFSSY SEQ ID NO:354PVFG SEQ ID NO:355RRVVPMATDNWLD SEQ ID NO:356SQTIHTY SEQ ID NO:357GAS SEQ ID NO:358SYSTPR SEQ ID NO:35968GNSFSNH SEQ ID NO:360HEGD SEQ ID NO:361SHSLQHSTGANY SEQ ID NO:363LAT SEQ ID NO:364GLHSPW SEQ ID NO:36569GNTFSKY SEQ ID NO:366HERD SEQ ID NO:367TQSLRHSNGANY SEQ ID NO:369LGS SEQ ID NO:370GLNRPW SEQ ID NO:37170GNTFSKY SEQ ID NO:366HEGD SEQ ID NO:361TQSLRHSNGANY SEQ ID NO:369LGS SEQ ID NO:370GLNRPW SEQ ID NO:37171GNTFRKY SEQ ID NO:372HEGD SEQ ID NO:361TQSLRHSNGANY SEQ ID NO:369LGS SEQ ID NO:370GLNRPW SEQ ID NO:37172QFRFDGY SEQ ID NO:938HDGI SEQ ID NO:939GNTSNIGNNF SEQ ID NO:941ETD SEQ ID NO: 942WAASLSSAR SEQ ID NO:94373DFPFSKY SEQ ID NO:374GDAW SEQ ID NO:375 SESLRQSNGKTS SEQ ID NO:377EVS SEQ ID NO:378SKDFPL SEQ ID NO:37974GVNTFGLY SEQ ID NO:380RW SEQ ID NO:381SQSITGNW SEQ ID NO:383RGA SEQ ID NO:384YDTYPG SEQ ID NO:38575GDSVSNDNY SEQ ID NO:944YSG SEQ ID NO:945PSHGFWSTSFSYWYFD SEQ ID NO:946SQSVTKY SEQ ID NO:947GTY SEQ ID NO: 948AHSTPW SEQ ID NO:949 Table D3 - CDRs (IMGT) for illustrative anti-HIV antigen binding arm Ab Name VH - CDR1 VH - CDR2 VH - CDR3 VL - CDR1 VL - CDR2 VL - CDR3 76GASISDSY SEQ ID NO:386VHKSGDT SEQ ID NO:387ARTLHGRRIYGIVAFNEWFTYFYMDV SEQ ID NO:388SLGSRA SEQ ID NO:389NNQ SEQ ID NO: 239HIWDSRVPTKWV SEQ ID NO:8177GDSMNNYY SEQ ID NO:390ISDRESA SEQ ID NO:391ATARRGQRIYGVVSFGEFFYYYSMDV SEQ ID NO:392ALGSRA SEQ ID NO:393NNQ SEQ ID NO:239HMWDSRSGFSWS SEQ ID NO:8778GDSMNNYY SEQ ID NO:390ISDRESA SEQ ID NO:391ARARRGQRIYGVVSFGEFFYYYSMDV SEQ ID NO:394ALGSRA SEQ ID NO:393NNQ SEQ ID NO: 239HMWDSRSGFSWS SEQ ID NO:8779GGSISNYY SEQ ID NO:395ISDRETT SEQ ID NO:396ATARRGQRIYGVVSFGEFFYYYYMDV SEQ ID NO:397ALGSRA SEQ ID NO:393NNQ SEQ ID NO: 239HMWDSRSGFSWS SEQ ID NO:8780NGSVSGRF SEQ ID NO:398FSDTDRS SEQ ID NO:399ARAQQGKRIYGIVSFGELFYYYYMDA SEQ ID NO:400SRGSRA SEQ ID NO: 401NNQ SEQ ID NO: 239HYWDSRSPISWI SEQ ID NO:9581NGSVSGRF SEQ ID NO:398FSDTDRS SEQ ID NO:399ARAQQGKRIYGIVSFGEFFYYYYMDA SEQ ID NO:402SRGSRA SEQ ID NO: 401NNQ SEQ ID NO: 239HYWDSRSPISWI SEQ ID NO:9582GASINDAY SEQ ID NO:403VHHSGDT SEQ ID NO:404ARALHGKRIYGIVALGELFTYFYMDV SEQ ID NO:405SLGSRS SEQ ID NO: 406NNN SEQ ID NO: 258HIWDSRRPTNWV SEQ ID NO:10283GTLVRDNY SEQ ID NO:407VHDSGDT SEQ ID NO:408ATTKHGRRIYGVVAFKEWFTYFYMDV SEQ ID NO:409SIGSRA SEQ ID NO: 410NNQ SEQ ID NO: 239HIYDARGGTNWV SEQ ID NO:10784GESTGACTYF SEQ ID NO:411LSHCQSFWGSGWT SEQ ID NO:412ARFDGEVLVYNHWPKPAWVDL SEQ ID NO:413ATNF SEQ ID NO: 414GVD SEQ ID NO: 269GSLVGNWDVI SEQ ID NO:11385GDSTAACDYF SEQ ID NO:415LSHCAGYYNTGWT SEQ ID NO:416ARFDGEVLVYHDWPKPAWVDL SEQ ID NO:417SNRF SEQ ID NO: 418GVN SEQ ID NO:275SSLVGNWDVI SEQ ID NO:11986GDSTAACDYF SEQ ID NO:415LSHCAGYYNSGWT SEQ ID NO:419ARFGGDVLVYHDWPKPAWVDL SEQ ID NO:420INNF SEQ ID NO: 421GVN SEQ ID NO:275GSLAGNWDVV SEQ ID NO:12387GDSTAACNSF SEQ ID NO:422LSHCASYWNRGWT SEQ ID NO:423ARFGGEVLRYTDWPKPAWVDL SEQ ID NO:424SNNF SEQ ID NO: 425DVN SEQ ID NO:284GSLVGNWDVI SEQ ID NO:11388GDSTAGCDYF SEQ ID NO: 426LSHCAGYYNTGWT SEQ ID NO:416ARFDGEVLVYNDWPKPAWVDL SEQ ID NO:427SNNF SEQ ID NO: 425GVN SEQ ID NO:275GSLVGNWDVI SEQ ID NO:11389GESINTGHYY SEQ ID NO: 428IHYTTAV SEQ ID NO:429VRSGGDILYYYEWQKPHWFSP SEQ ID NO:430SSDIGGWNF SEQ ID NO:431EVN SEQ ID NO: 291SSLFGRWDVV SEQ ID NO:13790GGSMRGTDWG ENDFH SEQ ID NO: 432IHWRGRTT SEQ ID NO:433ARHKYHDIFRVVPVAGWFDP SEQ ID NO:434QNVKNN SEQ ID NO: 435DAS SEQ ID NO: 297QQYEEWPRT SEQ ID NO:14391IHWRGTT SEQ ID NO:437VKHKYHDIVMVVPIAGWFDP SEQ ID NO:438QSVKNN SEQ ID NO: 439DTS SEQ ID NO:303QQYEEWPRT SEQ ID NO:14392IHWRGTT SEQ ID NO:437ARHRHHDVFMLVPIAGWFDV SEQ ID NO:441QNINKN SEQ ID NO: 442ETY SEQ ID NO:307QQYEEWPRT SEQ ID NO:14393IHYNGAT SEQ ID NO:444NAIRIYGVVALGEWFHYGMDV SEQ ID NO:445PLTSRF SEQ ID NO:446RSS SEQ ID NO:312QSSDTSDSYKM SEQ ID NO:15994INPKTGQP SEQ ID NO:448ARQRSDYWDFDV SEQ ID NO:449NGY SEQ ID NO: 450DGS SEQ ID NO:318QVYEF SEQ ID NO: 16595INPKTGQP SEQ ID NO:448ARQRSDFWDFDV SEQ ID NO:452NGY SEQ ID NO: 450DGS SEQ ID NO:318QVYEF SEQ ID NO:16596MKPRGGAV SEQ ID NO:454TRGKYCTARDYYNWDFEH SEQ ID NO:455QYGS SEQ ID NO: 456SGS SEQ ID NO:326QQYEF SEQ ID NO: 17397MKPRHGAV SEQ ID NO:457TRGKYCTARDYYNWDFEH SEQ ID NO:455QYGS SEQ ID NO: 456SGS SEQ ID NO:326QQYEF SEQ ID NO: 17398LKPRGGAV SEQ ID NO:459TRGKNCDYNWDFEH SEQ ID NO:460QYGS SEQ ID NO: 456SGS SEQ ID NO:326QQYEF SEQ ID NO: 17399GYTFTAHI SEQ ID NO:461IKPQYGAV SEQ ID NO:462ARDRSYGDSSWALDA SEQ ID NO:463QGVGSD SEQ ID NO: 464HTS SEQ ID NO:335QVLQF SEQ ID NO: 183100ISPHFARP SEQ ID NO:951ARDPFGDRAPHYNYHMDV SEQ ID NO:952QGLDSSH SEQ ID NO:953GTS SEQ ID NO: 930QRYGGTPIT SEQ ID NO:907101EDIFERTEL SEQ ID NO:954VKTVTGAV SEQ ID NO:955ARQKFYTGGQGWYFDL SEQ ID NO:956SYGH SEQ ID NO: 957ATS SEQ ID NO: 936QQLEF SEQ ID NO: 913102GFDFSRQG SEQ ID NO:465IKYDGSEK SEQ ID NO:466SNDVGGYES SEQ ID NO:468DVS SEQ ID NO:341KSLTSTRRRV SEQ ID NO:189103GFTFHKYG SEQ ID NO:469ISDDGMRK SEQ ID NO:470SSDVGGFDS SEQ ID NO:472DVS SEQ ID NO:341SSLTDRSHRI SEQ ID NO:195104GYSFIDYY SEQ ID NO:473IDPENGEA SEQ ID NO:474AAGAVGADSGSWFDP SEQ ID NO:475KLGDKY SEQ ID NO:476END SEQ ID NO:352QAWETTTTTFV SEQ ID NO:201105GGAFSSYA SEQ ID NO:477ITPVFGET SEQ ID NO:478TRDRRVVPMATDNWLDP SEQ ID NO:479QTIHTY SEQ ID NO: 480GAS SEQ ID NO:358QQSYSTPRT SEQ ID NO:207106GNSFSNHD SEQ ID NO:481MSHEGDKT SEQ ID NO:482LAT SEQ ID NO:364MQGLHSPWT SEQ ID NO:213107GNTFSKYD SEQ ID NO:485ISHERDKT SEQ ID NO:486LGS SEQ ID NO:370MQGLNRPWT SEQ ID NO:219108GNTFSKYD SEQ ID NO:485MSHEGDKT SEQ ID NO:482LGS SEQ ID NO:370MQGLNRPWT SEQ ID NO:219109GNTFRKYD SEQ ID NO:489MSHEGDKT SEQ ID NO:482LGS SEQ ID NO:370MQGLNRPWT SEQ ID NO:219110QFRFDGYG SEQ ID NO:958ISHDGIKK SEQ ID NO:959ETD SEQ ID NO: 942ATWAASLSSARV SEQ ID NO:919111DFPFSKYP SEQ ID NO:491ISGDAWHV SEQ ID NO:492EVS SEQ ID NO:378MQSKDFPLT SEQ ID NO:228112GVNTFGLYA SEQ ID NO:495IWRWKS SEQ ID NO:496TTTSTYDKWSGLHHDGVMAFSS SEQ ID NO:497QSITGNW SEQ ID NO:498RGA SEQ ID NO:384QQYDTYPGT SEQ ID NO:234113GDSVSNDNYY SEQ ID NO:962IYYSGTT SEQ ID NO:963VRMPSHGFWSTSFSYWYFDL SEQ ID NO:964QSVTKY SEQ ID NO: 965GTY SEQ ID NO: 948QQAHSTPWT SEQ ID NO:925 Table D4 - CDRs (Honegger) for illustrative anti-HIV antigen binding arm Ab Name VH - CDR1 VH - CDR2 VH - CDR3 VL - CDR1 VL - CDR2 VL - CDR3 114VSGASISDSY SEQ ID NO:499VHKSGDTNYSPSLKSR SEQ ID NO:500EKSLGSRA SEQ ID NO:238NNQDRPSGIPER SEQ ID NO:502WDSRVPTKW SEQ ID NO:240115VSGASISDSY SEQ ID NO:499VHKSGDTNYNPSLKSR SEQ ID NO:503EKSLGSRA SEQ ID NO:238NNQDRPSGIPER SEQ ID NO:502WDSRVPTKW SEQ ID NO:240116VSGDSMNNYY SEQ ID NO:505ISDRESATYNPSLNSR SEQ ID NO:506RQALGSRA SEQ ID NO:244NNQDRPSGIPER SEQ ID NO:502WDSRSGFSW SEQ ID NO:245117VSGGSISNYY SEQ ID NO:508ISDRETTTYNPSLNSR SEQ ID NO:509RQALGSRA SEQ ID NO:244NNQDRPSGIPER SEQ ID NO:502WDSRSGFSW SEQ ID NO:245118VSNGSVSGRF SEQ ID NO:511FSDTDRSEYNPSLRSR SEQ ID NO:512ERSRGSRA SEQ ID NO:251NNQDRPAGVSER SEQ ID NO:514WDSRSPISW SEQ ID NO:252119VSNGSVSGRF SEQ ID NO:511FSDTDRSEYNPSLRSR SEQ ID NO:512ERSRGSRA SEQ ID NO:251NNQDRPAGVSER SEQ ID NO:514WDSRSPISW SEQ ID NO:252120VSGTLVRDNY SEQ ID NO:516VHDSGDTNYNPSLKSR SEQ ID NO:517EESLGSRS SEQ ID NO:257NNNDRPSGIPDR SEQ ID NO:519WDSRRPTNW SEQ ID NO:259121VSGASINDAY SEQ ID NO:520VHHSGDTNYNPSLKRR SEQ ID NO:521KESIGSRA SEQ ID NO:263NNQDRPAGVPER SEQ ID NO:523YDARGGTNW SEQ ID NO:264122VSGESTGACTYF SEQ ID NO:524GTATNF SEQ ID NO: 268GVDKRPPGVPDR SEQ ID NO:527LVGNWDV SEQ ID NO:270123VSGDSTAACDYF SEQ ID NO:528 GTSNRF SEQ ID NO:274GVNKRPSGVPDR SEQ ID NO:531LVGNWDV SEQ ID NO:270124VSGDSTAACDYF SEQ ID NO:528GNINNF SEQ ID NO: 278GVNKRPSGVPDR SEQ ID NO:531LAGNWDV SEQ ID NO:279125VSGDSTAACNSF SEQ ID NO:534GTSNNF SEQ ID NO: 283DVNKRPSGVPDR SEQ ID NO:537LVGNWDV SEQ ID NO:270126VSGDSTAGCDYF SEQ ID NO:1090GTSNNF SEQ ID NO: 283GVNKRPSGVPDR SEQ ID NO:531LVGNWDV SEQ ID NO:270127VSGESINTGHYY SEQ ID NO:539IHYTTAVLHNPSLKSR SEQ ID NO:540GTSSDIGGWNF SEQ ID NO:290EVNKRPSGVPGR SEQ ID NO:542LFGRWDV SEQ ID NO:292128IHWRGRTTHYKTSFRSR SEQ ID NO:544ASQNVKNN SEQ ID NO:546DASSRAGGIPDR SEQ ID NO:547YEEWPR SEQ ID NO: 298129IHWRGTTHYNAPFRGR SEQ ID NO:549ASQSVKNN SEQ ID NO:1091DTSSRASGIPAR SEQ ID NO:551YEEWPR SEQ ID NO: 298130IHWRGTTHYKESLRRR SEQ ID NO:553ASQNINKN SEQ ID NO:555ETYSKIAAFPAR SEQ ID NO:556YEEWPR SEQ ID NO: 298131VSQDSRPSDHS SEQ ID NO:557IHYNGATTYNPSLRSR SEQ ID NO:558GAPLTSRF SEQ ID NO:311RSSQRSSGWSGR SEQ ID NO:560SDTSDSYK SEQ ID NO:313132ASGYNIRDYF SEQ ID NO:561INPKTGQPNNPRQFQGR SEQ ID NO:562QRSDYWDFD SEQ ID NO:563ANGY SEQ ID NO: 564DGSKLERGVPSRF SEQ ID NO:565YE SEQ ID NO:327133ASGYKISDHF SEQ ID NO:566INPKTGQPNNPRQFQGR SEQ ID NO:562QRSDFWDFD SEQ ID NO:1092ANGY SEQ ID NO: 564DGSKLERGVPAR SEQ ID NO:567YE SEQ ID NO:327134ASGYEFINCP SEQ ID NO:568MKPRGGAVSYARQLQGR SEQ ID NO:569TSQYGS SEQ ID NO: 571SGSTRAAGIPDR SEQ ID NO:572YE SEQ ID NO:327135ASGYEFINCP SEQ ID NO:568MKPRHGAVSYARQLQGR SEQ ID NO:573TSQYGS SEQ ID NO: 571SGSTRAAGIPDR SEQ ID NO:572YE SEQ ID NO:327136ASGYEFIDCT SEQ ID NO:574LKPRGGAVNYARPLQGR SEQ ID NO:575GKNCDYNWDFE SEQ ID NO:576TSQYGS SEQ ID NO: 571SGSTRAAGIPDR SEQ ID NO:572YE SEQ ID NO:327137TSGYTFTAHI SEQ ID NO:577IKPQYGAVNFGGGFRDR SEQ ID NO:578DRSYGDSSWALD SEQ ID NO:579TSQGVGSD SEQ ID NO: 580HTSSVEDGVPSR SEQ ID NO:581LQ SEQ ID NO:336138ISPHFARPIYSYKFRDR SEQ ID NO:967ASQGLDSSH SEQ ID NO:969GTSNRARGTPDR SEQ ID NO:970YGGTPI SEQ ID NO: 931139TSEDIFERTEL SEQ ID NO:971VKTVTGAVNFGSPDFRQ SEQ ID NO:972QKFYTGGQGWYFD SEQ ID NO:973AASYGH SEQ ID NO: 974ATSKRASGIPDR SEQ ID NO:975LE SEQ ID NO: 937140ASGFDFSRQG SEQ ID NO:582IKYDGSEKYHADSVWGR SEQ ID NO:583GTSNDVGGYES SEQ ID NO:340DVSKRPSGVSNR SEQ ID NO:585LTSTRRR SEQ ID NO:342141ASGFTFHKYG SEQ ID NO:586ISDDGMRKYHSDSMWGR SEQ ID NO:587GTSSDVGGFDS SEQ ID NO:346DVSHRPSGISNR SEQ ID NO:589LTDRSHR SEQ ID NO:347142VSGYSFIDYY SEQ ID NO:590IDPENGEARYAEKFQGR SEQ ID NO:591GAVGADSGSWFD SEQ ID NO:592GSKLGDKY SEQ ID NO:351ENDRRPSGIPER SEQ ID NO:593WETTTTTF SEQ ID NO:353143ASGGAFSSYA SEQ ID NO:594ITPVFGETKYAPRFQGR SEQ ID NO:595ASQTIHTY SEQ ID NO: 597GASTLQSGVPSR SEQ ID NO:598SYSTPR SEQ ID NO:359144ASGNSFSNHD SEQ ID NO:599MSHEGDKTGLAQKFQGR SEQ ID NO:600LATHRASGVPDR SEQ ID NO:603GLHSPW SEQ ID NO:365145ASGNTFSKYD SEQ ID NO:604ISHERDKTESAQRFKGR SEQ ID NO:605LGSQRASGVPDR SEQ ID NO:608GLNRPW SEQ ID NO:371146ASGNTFSKYD SEQ ID NO:604MSHEGDKTESAQRFKGR SEQ ID NO:609LGSQRASGVPDR SEQ ID NO:608GLNRPW SEQ ID NO:371147ASGNTFRKYD SEQ ID NO:610MSHEGDKTESAQRFKGR SEQ ID NO:609LGSQRASGVPDR SEQ ID NO:608GLNRPW SEQ ID NO:371148ASQFRFDGYG SEQ ID NO:976ISHDGIKKYHAEKVWGR SEQ ID NO:977GNTSNIGNNF SEQ ID NO:941ETDKRPSGIPDR SEQ ID NO:979WAASLSSAR SEQ ID NO:943149VSDFPFSKYP SEQ ID NO:612ISGDAWHVVYSNSVQGR SEQ ID NO:613 EVSNRFSGVSDR SEQ ID NO:616SKDFPL SEQ ID NO:379150AYGVNTFGLYA SEQ ID NO:617IWRWKSSASHHFRGR SEQ ID NO:618ASQSITGNW SEQ ID NO: 620RGAALLGGVPSR SEQ ID NO:621YDTYPG SEQ ID NO:385151VSGDSVSNDNYY SEQ ID NO:980IYYSGTTYYNPSLRNR SEQ ID NO:981ASQSVTKY SEQ ID NO:983GTYTLLSGVSPR SEQ ID NO:984AHSTPW SEQ ID NO: 949 Table E - VH / VL for illustrative anti-HIV binding arm Ab Name SEQ ID NO VH SEQ ID NO VL 320622623321624625322624626323627628324629630325631632326633634327635636328637638329639640330641642331643644332645646333647648334649650335651652336653654337655656338657658339659660340661662341663664342665666343667668344669670345671672346673670347674675348985986349987988350676677351678679352680681353682683354684685355686687356688689357690691358989990359692693360694695361991992 4. Fc Domains or Fc Regions

[0097] In certain embodiments, the Fc region or Fc domain of one or both of the CD3-targeting heavy chain and the HIV antigen-targeting heavy chain is from a human IgG. In one embodiment, the Fc region or Fc domain of the multi-specific antigen binding molecule is an IgG1, IgG2, IgG4, or chimeras thereof. In some embodiments, the antibody has a chimeric heavy chain constant region (e.g., having the CH1, hinge, and CH2 regions of IgG4 and CH3 region of IgG1). In some embodiments, the Fc region is from a human IgG1 or IgG4. In some embodiments, the Fc region is from a human IgG1. In some embodiments, the Fc region is from a human IgG4.

[0098] In certain embodiments the Fc region or Fc domain of one or both of the CD3-targeting heavy chain and the HIV antigen-targeting heavy chain, is directly linked to, or linked via an intervening amino acid sequence (e.g., a G-S linker), to a human IgG1 (e.g., a wild-type or mutant IgG1m3 sequence), IgG2, IgG3 or IgG4 with 1 to 10 (e.g., 1, 2, 3, 4, 5, 6, 7, 8, 9, 10) amino acid substitutions. In some embodiments, the Fc modifications can promote one or more of increased serum half-life or decreased antibody effector function of the molecule. In other embodiments, certain of these modifications, decrease antibody effector function and increase half-life of the antibody. In some embodiments, the Fc region or Fc domain of one or both of the CD3-targeting heavy chain and the HIV antigen-targeting heavy chain comprise two or more, three or more, four or more, five or more, six or more, six or fewer, five or fewer, four or fewer, three or fewer, two or fewer, or one modified Fc amino acid residue(s). As appropriate, the Fc region or Fc domain of one or both of the CD3-targeting heavy chain and the HIV antigen-targeting heavy chain can be the same or different. In some embodiments, the Fc region or Fc domain of one or both of the CD3-targeting heavy chain and the HIV antigen-targeting heavy chain independently can comprise two or more, three or more, four or more, five or more, six or more, six or fewer, five or fewer, four or fewer, three or fewer, two or fewer, or one modified Fc amino acid residue(s). Illustrative amino acid substitutions are described below.IgG1 Isotype Fc

[0099] In one embodiment, the Fc region comprises or is derived from a human IgG1. In some embodiments, the antibody has a chimeric heavy chain constant region (e.g., having the CH1, hinge, CH2 regions of IgG4 and CH3 region of IgG1).

[0100] IgG1 antibodies exist in various allotypes and isoallotypes. In various embodiments, one or both of the Fc region or Fc domain of the CD3-targeting heavy chain and the HIV antigen-targeting heavy chain include an IgG1 heavy chain having an allotype of G1m1; nG1m2; G1m3; G1m17,1; G1m17,1,2; G1m3,1; or G1m17. Each of these allotypes or isoallotypes is characterized by the following amino acid residues at the indicated positions within the IgG1 heavy chain constant region (Fc) (EU numbering): G1m1: D356, L358; nG1m1: E356, M358; G1m3: R214, E356, M358, A431; G1m17,1: K214, D356, L358, A431; G1m17,1,2: K214, D356, L358, G431; G1m3,1: R214, D356, L358, A431; and G1m17: K214, E356, M358, A431.

[0101] In certain embodiments, the Fc region or Fc domain of the multi-specific antigen binding molecule comprises a wild-type IgG1m3 sequence (e.g., SEQ ID NO: 1089), or has at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% identical to a wild-type IgG1m3 sequence (e.g., SEQ ID NO: 1089), as provided below.

[0102] In certain embodiments, the Fc region or Fc domain of one or both of the CD3-targeting heavy chain and the HIV antigen-targeting heavy chain is from an IgG1 isotype. In some embodiments, the Fc region or Fc domain of one or both of the CD3-targeting heavy chain and the HIV antigen-targeting heavy chain contains a human IgG1 constant region. In some embodiments, the human IgG1 Fc region contains one or more modifications. For example, in some embodiments, the Fc region contains one or more amino acid substitutions (e.g., relative to a wild-type Fc region of the same isotype). In some embodiments, the one or more amino acid substitutions are selected from N297A, N297Q (Bolt S et al. (1993) Eur J Immunol 23:403-411), D265A, L234A, L235A (McEarchem et al., (2007) Blood, 109:1185-1192), C226S, C229S (McEarchem et al., (2007) Blood. 109:1185-1192), P238S (Davis et al., (2007) J Rheumatol, 34:2204-2210), E233P, L234V (McEarchern et al., (2007) Blood, 109:1185-1192), P238A, A327Q, A327G, P329A (Shields R L. et al., (2001) J Biol Chem. 276(9):6591-604), K322A, L234F, L235E (Hezareh, et al., (2001) J Virol 75, 12161-12168; Oganesyan et al., (2008). Acta Crystallographica 64, 700-704), P331S (Oganesyan et al., (2008) Acta Crystallographica 64, 700-704), T394D (Wilkinson et al. (2013) MAbs 5(3): 406-417), A330L, M252Y, S254T, and T256E, where the amino acid position is according to the EU numbering convention. In certain embodiments, the Fc region further includes an amino acid deletion at a position corresponding to glycine 236 according to the EU numbering convention.

[0103] In some embodiments, the Fc region or Fc domain of one or both of the CD3-targeting heavy chain and the HIV antigen-targeting heavy chain has an IgG1 isotype with a heavy chain constant region that contains a C220S amino acid substitution according to the EU numbering convention.

[0104] In some embodiments, the Fc region comprises a human IgG1 isotype and comprises one or more amino acid substitutions in the Fc region at a residue position selected from the group consisting of: N297A, N297G, N297Q, N297G, D265A, L234A, L235A, C226S, C229S, P238S, E233P, L234V, P238A, A327Q, A327G, P329A, P329G, K322A, L234F, L235E, P331S, T394D, A330L, M252Y, S254T, T256E, M428L, N434S, T366W, T366S, L368A, Y407V and any combination thereof, wherein the numbering of the residues is according to EU numbering. In some embodiments, the Fc region comprises a human IgG1 isotype and comprises one or more amino acid substitutions in the Fc region at a residue position selected from the group consisting of: L234A, L234V, L234F, L235A, L235E, A330L, P331S, and any combination thereof, wherein the numbering of the residues is according to EU numbering. In some embodiments, the Fc region comprises a human IgG1 isotype and comprises the amino acid substitutions in the Fc region of L234A, L235A, and P331S, wherein the numbering of the residues is according to EU numbering. In some embodiments, the Fc region comprises a human IgG1 isotype and comprises the amino acid substitutions in the Fc region of M252Y, S254T, T256E, wherein the numbering of the residues is according to EU numbering. In some embodiments, the Fc region comprises a human IgG1 isotype and comprises the amino acid substitutions in the Fc region of T366S, L368A, Y407V, wherein the numbering of the residues is according to EU numbering. In some embodiments, the Fc region comprises a human IgG1 isotype and comprises the amino acid substitutions in the Fc region of L234A, L235A, P331S, T366S, L368A and Y407V, wherein the numbering of the residues is according to EU numbering. In some embodiments, the Fc region comprises a human IgG1 isotype and comprises the amino acid substitutions in the Fc region of L234A, L235A, P331S, M252Y, S254T and T256E, wherein the numbering of the residues is according to EU numbering.IgG4 Isotype Fc

[0105] For uses where effector function is to be avoided altogether, e.g. when antigen binding alone is sufficient to generate the desired therapeutic benefit, and effector function only leads to (or increases the risk of) undesired side effects, IgG4 antibodies may be used, or antibodies or fragments lacking the Fc region or a substantial portion thereof can be devised, or the Fc may be mutated to eliminate glycosylation altogether (e.g. N297A). Alternatively, a hybrid construct of human IgG2 (CH1 domain and hinge region) and human IgG4 (CH2 and CH3 domains) has been generated that is devoid of effector function, lacking the ability to bind the FcγRs (like IgG2) and unable to activate complement (like IgG4). (see, Rother et al. (2007) Nat. Biotechnol. 25:1256; Mueller et al. (1997) Mol. Immunol. 34:441; and Labrijn et al. (2008) Curr. Op. Immunol. 20:479, discussing Fc modifications to reduce effector function generally).

[0106] In certain embodiments, the Fc region or Fc domain of one or both of the CD3-targeting heavy chain and the HIV antigen-targeting heavy chain has an IgG4 isotype. In some embodiments, the Fc region or Fc domain of one or both of the CD3-targeting heavy chain and the HIV antigen-targeting heavy chain contains a human IgG4 constant region. In some embodiments, the human IgG4 constant region includes an Fc region. In some embodiments, the Fc region contains one or more modifications. For example, in some embodiments, the Fc region contains one or more amino acid substitutions (e.g., relative to a wild-type Fc region of the same isotype). In some embodiments, the one or more amino acid substitutions are selected from E233P, F234V, F234A, L235A, G237A, E318A, S228P, L235E, T394D, M252Y, S254T, T256E, N297A, N297G, N297Q, T366W, T366S, L368A, Y407V, M428L, N434S, and any combination thereof, where the amino acid position is according to the EU numbering convention. See, e.g., Hutchins et al. (1995) Proc Natl Acad Sci USA, 92:11980-11984; Reddy et al., (2000) J Immunol, 164:1925-1933; Angal et al., (1993) Mol Immunol. 30(1):105-8; U.S. Pat. No. 8,614,299 B2; Vafa O. et al., (2014) Methods 65:114-126; and Jacobsen et. al., J. Biol. Chem.(2017) 292(5):1865-1875. In some embodiments, the Fc region comprises a human IgG4 isotype and comprises one or more amino acid substitutions in the Fc region at a residue position selected from the group consisting of: F234V, F234A, L235A, L235E, S228P, and any combination thereof, wherein the numbering of the residues is according to EU numbering.

[0107] In some embodiments, an IgG4 variant of the present disclosure may be combined with at least one of an S228P mutation according to the EU numbering convention (Angal et al., (1993) Mol Immunol, 30:105-108) and with one or more mutations described in Peters et al., (2012) J Biol Chem. 13; 287(29):24525-33) to enhance antibody stabilization.IgG2 Isotype Fc

[0108] In certain embodiments, the Fc region or Fc domain of one or both of the CD3-targeting heavy chain and the HIV antigen-targeting heavy chain has an IgG2 isotype. In some embodiments, the Fc region or Fc domain of one or both of the CD3-targeting heavy chain and the HIV antigen-targeting heavy chain contains a human IgG2 constant region. In some embodiments, the human IgG2 constant region includes an Fc region. In some embodiments, the Fc region contains one or more modifications. For example, in some embodiments, the Fc region contains one or more amino acid substitutions (e.g., relative to a wild-type Fc region of the same isotype). In some embodiments, the one or more amino acid substitutions are selected from P238S, V234A, G237A, H268A, H268Q, H268E, V309L, N297A, N297G, N297Q, V309L, A330S, P331 S, C232S, C233S, M252Y, S254T, and T256E, where the amino acid position is according to the EU numbering convention (Vafa, et al., (2014) Methods 65:114-126).Fc Mutations that Increase Serum Half-Life

[0109] In some embodiments, the Fc region or Fc domain of one or both of the CD3-targeting heavy chain and the HIV antigen-targeting heavy chain comprise amino acid modifications that promote an increased serum half-life of the multi-specific antigen binding molecule. Mutations that increase the half-life of an antibody have been described. In one embodiment, the Fc region or Fc domain of one or both of the CD3-targeting heavy chain and the HIV antigen-targeting heavy chain comprise a methionine to tyrosine substitution at position 252 (EU numbering), a serine to threonine substitution at position 254 (EU numbering), and a threonine to glutamic acid substitution at position 256 (EU numbering). See, e.g., U.S. Patent No. 7,658,921. This type of mutant, designated as a "YTE mutant" exhibits a four-fold increased half-life relative to wild-type versions of the same antibody (Dall'Acqua, et al., J Biol Chem, 281: 23514-24 (2006); Robbie, et al., Antimicrob Agents Chemotherap., 57(12):6147-6153 (2013)). In certain embodiments, the Fc region or Fc domain of one or both of the CD3-targeting heavy chain and the HIV antigen-targeting heavy chain comprise an IgG constant domain comprising one, two, three or more amino acid substitutions of amino acid residues at positions 251-257, 285-290, 308-314, 385-389, and 428-436 (EU numbering). Alternatively, M428L and N434S ("LS") substitutions can increase the pharmacokinetic (PK) half-life of the multi-specific antigen binding molecule. In other embodiments, the Fc region or Fc domain of one or both of the CD3-targeting heavy chain and the HIV antigen-targeting heavy chain comprise a M428L and N434S substitution (EU numbering). In other embodiments, the Fc region or Fc domain of one or both of the CD3-targeting heavy chain and the HIV antigen-targeting heavy chain comprise T250Q and M428L (EU numbering) mutations. In other embodiments, the Fc region or Fc domain of one or both of the CD3-targeting heavy chain and the HIV antigen-targeting heavy chain comprise H433K and N434F (EU numbering) mutations.Fc Mutations that Reduce or Eliminate Effector Activity

[0110] In some embodiments, the Fc region or Fc domain of one or both of the CD3-targeting heavy chain and the HIV antigen-targeting heavy chain have amino acid substitutions that reduce or eliminate Fc effector function (including, e.g., antibody-dependent cellular cytotoxicity (ADCC), antibody-dependent cellular phagocytosis (ADCP), and complement-dependent cytotoxicity (CDC)).

[0111] In some embodiments, the Fc region is altered by replacing at least one amino acid residue with a different amino acid residue to reduce or eliminate effector function(s) of the antibody. For example, one or more amino acids selected from amino acid residues 234, 235, 236, 237, 297, 318, 320 and 322 (EU numbering) can be replaced with a different amino acid residue such that the multi-specific antigen binding molecule has decreased affinity for an effector ligand. The effector ligand to which affinity is altered can be, for example, an Fc receptor (e.g., at residue positions 234, 235, 236, 237, 297 (EU numbering)) or the C1 component of complement (e.g., at residue positions 297, 318, 320, 322 (EU numbering)). U.S. Pat. Nos. 5,624,821 and 5,648,260, both by Winter et al.

[0112] Fc modifications reducing or eliminating effector function include substitutions, insertions, and deletions, e.g., at one or more positions including 234, 235, 236, 237, 267, 269, 325, and 328, e.g., 234G, 235G, 236R, 237K, 267R, 269R, 325L, and 328R (EU numbering). Further, an Fc variant may comprise 236R / 328R. Other modifications for reducing FcγR and complement interactions include substitutions at positions 297A, 234A, 235A, 318A, 228P, 236E, 268Q, 309L, 330S, 331S, 220S, 226S, 229S, 238S, 233P, and 234V (EU numbering). These and other modifications are reviewed in Strohl (2009) Current Opinion in Biotechnology 20:685-691. Effector functions (both ADCC and complement activation) can be reduced, while maintaining neonatal FcR binding (maintaining half-life), by mutating IgG residues at one or more of positions 233-236 and 327-331, such as E233P, L234V, L235A, optionally G236A, A327G, A330S and P331S in IgG1; E233P, F234V, L235A, optionally G236A, in IgG4; and A330S and P331S in IgG2 (EU numbering). See Armour et al. (1999) Eur. J. Immunol. 29:2613; WO 99 / 58572. Other mutations that reduce effector function include L234A and L235A ("LALA") in IgG1 (Alegre et al. (1994) Transplantation 57:1537); V234A and G237A in IgG2 (Cole et al. (1997) J. Immunol. 159:3613; see also U.S. Pat. No. 5,834,597); and S228P and L235E for IgG4 (Reddy et al. (2000) J. Immunol. 164:1925). Another combination of mutations for reducing effector function in a human IgG1 include L234F, L235E and P331S. Oganesyan et al. (2008) Acta Crystallogr. D. Biol. Crystallogr. 64:700. See generally Labrijn et gal. (2008) Curr. Op. Immunol. 20:479. Additional mutations found to decrease effector function in the context of an Fc (IgG1) fusion protein (abatacept) include C226S, C229S and P238S (EU numbering). Davis et al. (2007) J. Immunol. 34:2204.

[0113] ADCC activity may be reduced by modifying the Fc region. In certain embodiments, sites that affect binding to Fc receptors may be removed, e.g., sites other than salvage receptor binding sites. In other embodiments, an Fc region may be modified to remove an ADCC site. Exemplary ADCC sites have been described with respect to ADCC sites in IgG1 (Sarmay, et al, (1992) Molec. Immunol. 29 (5): 633-9). In one embodiment, the G236R and L328R variant of human IgG1 effectively eliminates FcγR binding (Horton, et al. (2011) J. Immunol. 186:4223 and Chu, et al. (2008) Mol. Immunol. 45:3926). In other embodiments, the Fc having reduced binding to FcγRs comprises the amino acid substitutions L234A, L235E and G237A. Gross, et al. (2001) Immunity 15:289. Modifications in the IgG Fc region to decrease binding to FcγRI to decrease ADCC (e.g., 234A; 235E; 236A; G237A) identified in WO 88 / 007089 can be used in the present fusion proteins. See also Duncan & Winter (1988) Nature 332:563; Chappel et al. (1991) Proc. Nat'l Acad. Sci. (USA) 88:9036; and Sondermann et al. (2000) Nature 406:267 (discussing the effects of these mutations on FcγRIII binding). In some embodiments, the Fc having reduced binding to FcγRs comprises the amino acid substitutions L234A and L235A.

[0114] CDC activity may also be reduced by modifying the Fc region. Mutations at IgG1 positions D270, K322, P329 and P331, specifically alanine mutations D270A, K322A, P329A and P331A, significantly reduce the ability of the corresponding antibody to bind C1q and activate complement (Idusogie et al. (2000) J. Immunol. 164:4178; WO 99 / 51642. Modification of position 331 of IgG1 (e.g. P331S) has been shown to reduce complement binding (Tao et al. (1993) J. Exp. Med. 178:661; Xu Y, et al. J Biol Chem. 1994. 269:3469-74; and Canfield & Morrison (1991) J. Exp. Med. 173:1483). In another example, one or more amino acid residues within amino acid positions 231 to 239 are altered to thereby reduce the ability of the antibody to fix complement (WO 94 / 29351). Modifications in the IgG Fc region identified in WO 88 / 007089 that reduce or eliminate binding to complement component C1q, and therefore reduce or eliminate CDC (e.g., E318A or V / K320A and K322A / Q) can be used in the present fusion proteins.

[0115] In some embodiments, the Fc with reduced complement fixation has the amino acid substitutions A330S and P331S. Gross et al. (2001) Immunity 15:289. In some embodiments, the Fc with reduced complement fixation has the amino acid substitution P331S.

[0116] In certain embodiments, the Fc region or Fc domain of one or both of the CD3-targeting heavy chain and the HIV antigen-targeting heavy chain have essentially no effector function, e.g., one or both of the Fc domains have reduced or eliminated binding to FcγRs and reduced or eliminated complement fixation, e.g., is effectorless. An exemplary IgG1 Fc that is effectorless comprises the following five mutations: L234A, L235E, G237A, A330S and P331S (EU numbering) (Gross et al. (2001) Immunity 15:289). These five substitutions may be combined with N297A to eliminate glycosylation as well.

[0117] In certain embodiments, the multi-specific antigen binding molecules comprise the L234F, L235E, D264A mutations, which are collectively referred to as "FEA." The FEA mutations decrease or abrogate effector function. In certain embodiments, the multi-specific antigen binding molecules comprise the L234F, L235E, D264A, and F405L mutations, which are collectively referred to as "FEAL." In certain embodiments, the multi-specific antigen binding molecules comprise the L234F, L235E, D264A, and a mutation selected from the group consisting of F405L, F405A, F405D, F405E, F405H, F405I, F405K, F405M, F405N, F405Q, F405S, F405T, F405V, F405W, and F405Y. In certain embodiments, the multi-specific antigen binding molecules comprise the L234F, L235E, D264A, and K409R mutations, which are collectively referred to as "FEAR." In certain embodiments, FEAL and FEAR are comprised in a multi-specific antigen binding molecule described herein. In certain embodiments, the multi-specific antigen binding molecules additionally comprise the M428L and N434S mutations, which are collectively referred to as LS. In certain embodiments, the multi-specific antigen binding molecules comprise the L234F, L235E, D264A, F405L, M428L, and N434S mutations, which are collectively referred to as "FEALLS." In certain embodiments, the multi-specific antigen binding molecules comprise the L234F, L235E, D264A, M428L, and N434S mutations along with one further mutation selected from the group consisting of F405L, F405A, F405D, F405E, F405H, F405I, F405K, F405M, F405N, F405Q, F405S, F405T, F405V, F405W, and F405Y. In certain embodiments, the multi-specific antigen binding molecules comprise the L234F, L235E, D264A, K409R, M428L, and N434S mutations which are collectively referred to as "FEARLS." In certain embodiments, FEALLS and FEARLS are comprised in a multi-specific antigen binding molecule described herein.

[0118] Other Fc variants having one or both of reduced ADCC and reduced CDC are disclosed at Glaesner et al. (2010) Diabetes Metab. Res. Rev. 26:287 (F234A and L235A to decrease ADCC and ADCP in an IgG4); Hutchins et al. (1995) Proc. Nat'l Acad. Sci. (USA) 92:11980 (F234A, G237A and E318A in an IgG4); An et al. (2009) MAbs 1:572 and U.S. Pat. App. Pub. 2007 / 0148167 (H268Q, V309L, A330S and P331S in an IgG2); McEarchern et al. (2007) Blood 109:1185 (C226S, C229S, E233P, L234V, L235A in an IgG1); Vafa et al. (2014) Methods 65:114 (V234A, G237A, P238S, H268A, V309L, A330S, P331S in an IgG2) (EU numbering).

[0119] By reducing or abrogating effector function on the Fc domains of the CD3 X gp120 multi-specific / bispecific antigen binding molecule (i) T-cells bound by the molecule, including those not infected with HIV, are not killed by innate effector cells e.g., NK cells, macrophages; and (ii) T-cells are not activated in the absence of target cells due to reduced binding to FcγRs on innate effector cells. Activation of T-cells in the absence of target cells could lead to an intolerable cytokine response. Binding of the bispecific molecule to FcγRs on innate effector cells could lead to clustering of the CD3 molecules on the T-cells, resulting in antigen-independent T-cell activation.Fc Mutations that Facilitate Heterodimerization

[0120] In some embodiments, first and second Fc domains have mutations to facilitate heterodimerization. Mutations in Fc domain pairs that facilitate or promote heterodimerization are reviewed in Ha, et al., Front. Immunol. (2016) 7:394. In some embodiments, the first Fc domain and the second Fc domain comprise the following amino acid substitutions (EU numbering), respectively (or vice versa): T366W and T366S / L368A / Y407V; T366W / S354C and T366S / L368A / Y407V / Y349C; S364H / F405A and Y349T / T394; T350V / L351Y / F405A / Y407V and T350V / T366L / K392L / T394W; K360D / D399M / Y407A and E345R / Q347R / T366V / K409V; K409D / K392D and D399K / E356K; K360E / K409W and Q347R / D399V / F405T; K360E / K409W / Y349C and Q347R / D399V / F405T / S354C; or K370E / K409W and E357N / D399V / F405T. In some embodiments, the first Fc domain and the second Fc domain comprise the following amino acid substitutions (EU numbering), respectively (or vice versa): T366W and T366S / L368A / Y407V.

[0121] In some embodiments, Fc region heterodimerization of the two different heavy chain-containing species can be facilitated by so-called 'knobs-into-holes' mutations (Atwell et al. 1997. JMB 270:26-35). The 'hole' mutations (T366S, L368A and Y407V ("SAV")) are incorporated into one Fc-containing chain, the T366W 'knob' ("W") mutation is incorporated into the other chain. In addition, a C220S mutation can be incorporated into an IgG1 hinge region of a scFv-containing arm to eliminate a free cysteine that otherwise forms a disulfide bond with a corresponding cysteine in the light chain in a wild-type IgG1. Co-transfection of such constructs leads to preferential formation of a heterodimeric Fc, with low levels of homodimer contaminants. Additionally, incorporating a S354C mutation can be incorporated into the Fc containing the 'knob' mutations and a Y349C mutation into the Fc containing the 'hole' mutations can optionally be used to generate a covalent bond between the two halves of the heterodimeric Fc if additional thermodynamic stability is desired (Merchant et al. 1998. Nat. Biotechnol. 16: 677-81).

[0122] To facilitate purification of the heterodimeric molecule away from contaminating homodimeric products, the H435R ("R") or H435R+Y436F ("RF") mutations to reduce or eliminate Protein A binding can be introduced into one but not both of the Fc-containing chains (Jendeberg, L. et al. 1997 J. Immunol. Methods 201:25-34). This reduces or eliminates Protein A binding of the homodimer contaminant containing these mutations, and greatly simplifies purification of the desired heterodimer away from remaining homodimer contaminant via additional chromatography steps (e.g. ion exchange). In embodiments incorporating H435R (or H435R+Y436F) mutations in the first or second Fc region of a heavy chain, if the VH region in the same heavy chain is from a VH3 family variable region, this VH region can also include amino acid substitutions, as described herein, to reduce or eliminate Protein A binding of the entire heavy chain. In some embodiments, the H435R is introduced into one of the Fc-containing chains.

[0123] In certain embodiments, the one or more modifications are selected from the following Fc amino acid substitutions (EU numbering) or combinations thereof: L234F; L235E; G236A; S239D; F243L; D265E; D265A; S267E; H268F; R292P; N297Q; N297A; S298A; S324T; I332E; S239D; A330L; L234F; L235E; P331S; F243L; Y300L; V305I; P396L; S298A; E333A; K334A; E345R; L235V; F243L; R292P; Y300L; P396L; M428L; E430G; N434S; G236A, S267E, H268F, S324T, and I332E; G236A, S239D, and I332E; S239D, A330L, I332E; L234F, L235E, and P331S; F243L, R292P, Y300L, V305I, and P396L; G236A, H268F, S324T, and I332E; S239D, H268F, S324T, and I332E; S298A, E333A, and K334A; L235V, F243L, R292P, Y300L, and P396L; S239D, I332E; S239D, S298A, and I332E; G236A, S239D, I332E, M428L, and N434S; G236A, S239D, A330L, I332E, M428L, and N434S; S239D, I332E, G236A and A330L; M428L and N4343S; M428L, N434S; G236A, S239D, A330L, and I332E; and G236A and I332E. In certain embodiments, the one or more modifications is selected from the group consisting of: N297A, D265A, L234F, L235E, N297Q, and P331S. In certain embodiments, the one or more modifications is N297A or D265A. In certain embodiments the one or more modifications are L234F and L235E. In certain embodiments, the one or more modifications are L234F, L234E, and D265A. In certain embodiments, the one or more modifications are L234F, L234E, and N297Q. In certain embodiments, the one or more modifications are L234F, L235E, and P331S. In certain embodiments, the one or more modifications are D265A and N297Q. In certain embodiments, the one or more modifications are L234F, L235E, D265A, N297Q, and P331S. In some embodiments the modifications are L234A, L235A, P331S, T366S, L368A, Y407V and H435R. In some embodiments the modifications are L234A, L235A, P331S, T366W, M252Y, S254T and T256E.

[0124] Combined mutations that reduce Fc-receptor binding and find use in the present antigen binding molecules include, for example, N297A; N297Q; D265A; L234F / L235E; L234F / L235E / N297Q; L234F / L235E / P331S; D265A / N297Q; and L234F / L235E / D265A / N297Q / P331S (all EU numbering). In certain embodiments the multi-specific antigen binding molecules described herein comprise L234F and L235E mutations. In certain embodiments the multi-specific antigen binding molecules described herein comprise L234F, L235E, and D265A mutations. In certain embodiments the multi-specific antigen binding molecules described herein comprise L234F, L235E, and N297Q mutations. In certain embodiments the multi-specific antigen binding molecules described herein comprise an N297A or N297Q mutation. In certain embodiments the multi-specific antigen binding molecules described herein comprise an N297A or N297Q mutation as well as L234F, L235E, and D265A mutations. In certain embodiments, one, two, three, four, or more amino acid substitutions are introduced into a Fc region to alter the effector function of the antigen binding molecule. For example, these substitutions are located at positions selected from the group consisting of amino acid residues 234, 235, 236, 237, 265, 297, 318, 320, and 322, (according to EU numbering). These positions can be replaced with a different amino acid residue such that the antigen binding molecule has an altered (e.g., reduced) affinity for an effector ligand (e.g., an Fc receptor or the C1 component of complement), but retains the antigen binding ability of the parent antibody. In certain embodiments, the multi-specific antigen binding molecules described herein comprise one or more of E233P, L234V, L235A, and G236A mutations (EU numbering). In some embodiments, the multi-specific antigen binding molecules comprise one or more of A327G, A330S, and P331S mutations (EU numbering). In some embodiments, the multi-specific antigen binding molecules comprise K322A mutations (EU numbering). In some embodiments the multi-specific antigen binding molecules comprise E318A, K320A, and K322A (EU numbering) mutations. In certain embodiments, the multi-specific antigen binding molecules comprise a L235E (EU numbering) mutation.

[0125] In some embodiments, the Fc region or Fc domain of one or both of the CD3-targeting heavy chain and the HIV antigen-targeting heavy chain does not comprise a hinge region; or the hinge region is truncated or deleted, in whole or in part. The structural hinge region of human IgG1, IgG2 and IgG4 antibodies is a peptide linker of 19 to 23 amino acids containing two to four cysteine residues, is genetically encoded on the hinge exon together with the 5'-end of the CH2 exon, and allows for disulfide bridges between first and second Fc domains (Roux, et al., J. Immunol. (1998) 161:4083). The structural hinge region is comprised of amino acid residue positions 216-238 (EU numbering) or 226-251 (Kabat numbering) (identified on imgt.org). In some embodiments, the Fc region or Fc domain of one or both of the CD3-targeting heavy chain and the HIV antigen-targeting heavy chain comprises or is derived from a human IgG4 isotype and does not comprise the amino acid sequence ESKYGPPCPPCP (SEQ ID NO: 504). In some embodiments, the Fc region or Fc domain of one or both of the CD3-targeting heavy chain and the HIV antigen-targeting heavy chain comprises or is derived from a human IgG1 isotype and does not comprise the amino acid sequence EPKSCDKTHTCPPCP (SEQ ID NO: 993) or EPKSCDKTHTCPPCPAPELL (SEQ ID NO: 994). As appropriate, the hinge region of the Fc domain of one or both of the CD3-targeting heavy chain and the HIV antigen-targeting heavy chain can be the same or different.

[0126] In various embodiments, the Fc region or Fc domain of the CD3-targeting heavy chain and the HIV antigen-targeting heavy chain comprise a heterodimeric human IgG1, comprising amino acid sequences set forth, respectively, or comprising amino acid sequences that are at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% identical to the amino acid sequences set forth, respectively: SEQ ID NOs.: 696 and 697; SEQ ID NOs.: 697 and 696; SEQ ID NOs.: 696 and 698; SEQ ID NOs.: 698 and 696; SEQ ID NOs.: 699 and 700; SEQ ID NOs.: 700 and 699; SEQ ID NOs.: 701 and 698; SEQ ID NOs.: 698 and 701; SEQ ID NOs.: 702 and 703; SEQ ID NOs.: 703 and 702; SEQ ID NOs.: 704 and 698; SEQ ID NOs.: 698 and 704; SEQ ID NOs.: 705 and 703; SEQ ID NOs.: 703 and 705; SEQ ID NOs.: 706 and 704; SEQ ID NOs.: 704 and 706; SEQ ID NOs.: 707 and 703; SEQ ID NOs.: 703 and 707; SEQ ID NOs.: 708 and 704; SEQ ID NOs.: 704 and 708; SEQ ID NOs.: 709 and 710; or SEQ ID NOs.: 710 and 709.

[0127] In some embodiments, the Fc region or Fc domain of the CD3-targeting heavy chain and the HIV antigen-targeting heavy chain comprise a heterodimeric human IgG1, comprising amino acid sequences that are at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% identical to the amino acid sequences set forth, respectively: SEQ ID NOs.: 703 and 705. In some embodiments the Fc region or Fc domain of the CD3-targeting heavy chain and the HIV antigen-targeting heavy chain comprise a heterodimeric human IgG1, comprising amino acid sequences that are at least 95% identical to the amino acid sequences set forth, respectively: SEQ ID NOs.: 703 and 705. In some embodiments, the Fc region or Fc domain of the CD3-targeting heavy chain and the HIV antigen-targeting heavy chain comprise a heterodimeric human IgG1, comprising amino acid sequences that are at least 99% identical to the amino acid sequences set forth, respectively: SEQ ID NOs.: 703 and 705. In some embodiments, the Fc region or Fc domain of the CD3-targeting heavy chain and the HIV antigen-targeting heavy chain comprise a heterodimeric human IgG1, comprising amino acid sequences set forth, respectively: SEQ ID NOs.: 703 and 705.

[0128] Illustrative sequences of complementarity first and second Fc domain pairs of the multi-specific antigen binding molecules, targeting HIV gp120, are provided in Table F. In some embodiments, the first Fc domain in a Fc domain pair is fused to the first antigen binding domain, targeting CD3. In such embodiments, the second Fc domain in a Fc domain pair is fused to the second antigen binding domain, targeting an HIV antigen. In some embodiments, the second Fc domain in a Fc domain pair is fused to the first antigen binding domain, targeting CD3. In such embodiments, the first Fc domain in a Fc domain pair is fused to the second antigen binding domain, targeting an HIV antigen. TABLE F - Fc regions - Heterodimeric pairs Fc aa subst SEQ ID NO: Fc aa subst SEQ ID NO: L234A, L235A, P331S T366W696L234A, L235A, P331S T366S, L368A, Y407V H435R, Y436F697L234A, L235A, P331S T366W696L234A, L235A, P331S T366S, L368A, Y407V H435R698L234A, L235A, P331S T366S, L368A, Y407V699L234A, L235A, P331S T366W700L234A, L235A, P331S T366WM428L, N434S701 L234A, L235A, P331S T366S, L368A, Y407V H435R698 L234A, L235A, P331S T366W M428L, N434S702L234A, L235A, P331S T366S, L368A, Y407V H435R703L234A, L235A, P331S T366W M252Y, S254T, T256E704L234A, L235A, P331S T366S, L368A, Y407V H435R698L234A, L235A, P331S T366W M252Y, S254T, T256E705L234A, L235A, P331S T366S, L368A, Y407V H435R703L234A, L235A, P331S T366S, L368A, Y407V H435R706 L234A, L235A, P331S T366W M252Y, S254T, T256E704 L234A, L235A, P331S + T366W707L234A, L235A, P331S + T366S, L368A, Y407V + H435R703L234A, L235A, P331S T366S, L368A, Y407V M252Y, S254T, T256E708L234A, L235A, P331S T366W M252Y, S254T, T256E704L234A, L235A, P331S T366S, L368A, Y407V M252Y, S254T, T256E709 L234A, L235A, P331S T366W M252Y S254T, T256E710

[0129] In some embodiments, the multi-specific antigen binding molecule comprises a first antigen binding domain that is a Fab and a second antigen binding domain that is an EC domain of CD4 wherein the first antigen binding domain comprises a first HC and a LC, and the second antigen binding domain comprises a second HC, the second HC, the first HC and the LC comprising the amino acid sequences set forth, respectively, or comprising amino acid sequences that are at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% identical to the amino acid sequences set forth, respectively: SEQ ID NOs: 751, 752 and 753;or SEQ ID NOs: 754, 752 and 753. In some embodiments, the multi-specific antigen binding molecule comprises a first antigen binding domain that is a Fab and a second antigen binding domain that is an EC domain of CD4 wherein the first antigen binding domain comprises a first HC and a LC, and the second antigen binding domain comprises a second HC, the second HC, the first HC and the LC comprising the amino acid sequences set forth, respectively, or comprising amino acid sequences that are at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% identical to the amino acid sequences set forth, respectively: SEQ ID NOs: 751, 752 and 753. In various embodiments, the multi-specific antigen binding molecule has a potency, as measured by EC50, of less than 0.15 µg / mL, e.g., less than 0.14 µg / mL, 0.13 µg / mL, 0.12 µg / mL or 0.11 µg / mL, or less, against at least 30 different HIV isolates. In some embodiments, the multi-specific antigen binding molecule has a potency, as measured by EC50, of less than 0.11 µg / mL against at least 30 different HIV isolates.

[0130] In some embodiments, the multi-specific antigen binding molecule comprises a first antigen binding domain that is a Fab and a second antigen binding domain that is an EC domain of CD4 wherein the first antigen binding domain comprises a first HC and a LC, and the second antigen binding domain comprises a second HC, the second HC, the first HC and the LC comprising amino acid sequences that are at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% identical to the amino acid sequences set forth, respectively: SEQ ID NOs: 751, 752 and 753. In some embodiments, the multi-specific antigen binding molecule comprises a first antigen binding domain that is a Fab and a second antigen binding domain that is an EC domain of CD4 wherein the first antigen binding domain comprises a first HC and a LC, and the second antigen binding domain comprises a second HC, the second HC, the first HC and the LC comprising amino acid sequences that are at least 95% identical to the amino acid sequences set forth, respectively: SEQ ID NOs: 751, 752 and 753. In some embodiments, the multi-specific antigen binding molecule comprises a first antigen binding domain that is a Fab and a second antigen binding domain that is an EC domain of CD4 wherein the first antigen binding domain comprises a first HC and a LC, and the second antigen binding domain comprises a second HC, the second HC, the first HC and the LC comprising amino acid sequences that are at least 99% identical to the amino acid sequences set forth, respectively: SEQ ID NOs: 751, 752 and 753. In some embodiments, the multi-specific antigen binding molecule comprises a first antigen binding domain that is a Fab and a second antigen binding domain that is an EC domain of CD4 wherein the first antigen binding domain comprises a first HC and a LC, and the second antigen binding domain comprises a second HC, the second HC comprising an amino acid sequence that is at least 95% (e.g., at least 96%, at least 97%, at least 98%, or at least 99%) identical to the amino acid sequence of SEQ ID NO: 751, and the first HC and the LC comprising amino acid sequences that are at least 95% (e.g., at least 96%, at least 97%, at least 98%, or at least 99%) identical to the amino acid sequences set forth, respectively: SEQ ID NOs: 752 and 753.

[0131] In some embodiments, the multi-specific antigen binding molecule comprises a first antigen binding domain that is a Fab and a second antigen binding domain that is an EC domain of CD4 wherein the first antigen binding domain comprises a first HC and a LC, and the second antigen binding domain comprises a second HC, the second HC comprising an amino acid sequence that is at least 95% (e.g., at least 96%, at least 97%, at least 98%, or at least 99%) identical to the amino acid sequence of SEQ ID NO: 751, the first HC comprises an amino acid sequence that is at least 95% (e.g., at least 96%, at least 97%, at least 98%, or at least 99%) identical to the amino acid sequence of SEQ ID NO: 752, and the LC comprises an amino acid sequence that is at least 95% (e.g., at least 96%, at least 97%, at least 98%, or at least 99%) identical to the amino acid sequence of SEQ ID NO: 753. In some embodiments, the multi-specific antigen binding molecule comprises a first antigen binding domain that is a Fab and a second antigen binding domain that is an EC domain of CD4 wherein the first antigen binding domain comprises a first HC and a LC, and the second antigen binding domain comprises a second HC, the second HC comprising an amino acid sequence that is at least 95% (e.g., at least 96%, at least 97%, at least 98%, or at least 99%) identical to the amino acid sequence of SEQ ID NO: 751, the first HC comprises an amino acid sequence of SEQ ID NO: 752, and the LC comprises an amino acid sequence that is at least 95% identical to the amino acid sequence of SEQ ID NO: 753. In some embodiments, the multi-specific antigen binding molecule comprises a first antigen binding domain that is a Fab and a second antigen binding domain that is an EC domain of CD4 wherein the first antigen binding domain comprises a first HC and a LC, and the second antigen binding domain comprises a second HC, the second HC comprising an amino acid sequence that is at least 95% (e.g., at least 96%, at least 97%, at least 98%, or at least 99%) identical to the amino acid sequence of SEQ ID NO: 751, the first HC comprises an amino acid sequence of SEQ ID NO: 752, and the LC comprises an amino acid sequence that is at least 99% identical to the amino acid sequence of SEQ ID NO: 753. In some embodiments, the multi-specific antigen binding molecule comprises a first antigen binding domain that is a Fab and a second antigen binding domain that is an EC domain of CD4 wherein the first antigen binding domain comprises a first HC and a LC, and the second antigen binding domain comprises a second HC, the second HC comprising an amino acid sequence that is at least 95% (e.g., at least 96%, at least 97%, at least 98%, or at least 99%) identical to the amino acid sequence of SEQ ID NO: 751, the first HC comprises an amino acid sequence that is at least 95% identical to the amino acid sequence of SEQ ID NO: 752, and the LC comprises an amino acid sequence of SEQ ID NO: 753. In some embodiments, the multi-specific antigen binding molecule comprises a first antigen binding domain that is a Fab and a second antigen binding domain that is an EC domain of CD4 wherein the first antigen binding domain comprises a first HC and a LC, and the second antigen binding domain comprises a second HC, the second HC comprising an amino acid sequence that is at least 95% (e.g., at least 96%, at least 97%, at least 98%, or at least 99%) identical to the amino acid sequence of SEQ ID NO: 751, the first HC comprises an amino acid sequence that is at least 99% identical to the amino acid sequence of SEQ ID NO: 752, and the LC comprises an amino acid sequence of SEQ ID NO: 753. In some embodiments, the multi-specific antigen binding molecule comprises a first antigen binding domain that is a Fab and a second antigen binding domain that is an EC domain of CD4 wherein the first antigen binding domain comprises a first HC and a LC, and the second antigen binding domain comprises a second HC, the second HC, the first HC and the LC comprising amino acid sequences set forth, respectively: SEQ ID NOs: 751, 752 and 753.

[0132] Illustrative sequences of bispecific molecules targeting CD3 and HIV gp120, described herein, are provided in Table G. Bispecific molecules described herein are summarized in Table 53. TABLE G - amino acid sequences of illustrative CD3 / gp120-targeting bi-specific binding moleculesAb name / featuresunpaired HCFab Arm - HC (VH-Fc)Fab Arm - LC (VL-CL)180 hCD4 D1.22 Fc AAS+W+YT E / huSP34 .39.13 AAS+SAV+ R 181 hCD4 D1.22 Fc AAS+W / hu SP34.39. 13 AAS+SAV+ R 182 hCD4 D1.22 Fc AAS+SAV+ R / huSP34 .3.13 AAS+W+YT E 183 hCD4 D1.22 Fc AAS+SAV+ R / huSP34 .3.13 AAS+W 184 hCD4 D1.22 Fc AAS+SAV / huSP34.3 .13 AAS+W 185 hCD4 D1.22 Fc AAS+SAV+ YTE / huSP 34.3.13 AAS+W+YT E 186 CD4 D1.22 Fc AAS+W / hu SP34.1.3 AAS+SAV+ R 187 hCD4 D1.22 Fc AAS+W+LS / huSP34. 1.3 AAS+SAV+ R 190 hCD4 D1.22 (ta ndem) Fc AAS+W+YT E / huSP34 .39.13 AAS+SAV+ R 191 hCD4 D1.22 (ta ndem) Fc AAS+W / hu SP34.39. 13 AAS+SAV+ R 192 hCD4 D1.22 (ta ndem) Fc AAS+SAV+ R / huSP34 .3.13 AAS+W+YT E 193 hCD4 D1.22(ta ndem) Fc AAS+SAV+ R / huSP34 .3.13 AAS+W 194 hCD4 D1.22(ta ndem) Fc AAS+SAV / huSP34.3 .13 AAS+W 195 hCD4 D1.22(ta ndem) Fc AAS+SAV+ YTE / huSP 34.3.13 AAS+W+YT E 196 hCD4 D1.22 (ta ndem) Fc AAS+W / hu SP34.1.3 AAS+SAV+ R 197 hCD4 D1.22(ta ndem) Fc AAS+W+LS / huSP34. 1.3 AAS+SAV+ R 198 hCD4-D1.22(ta ndem) Fc AAS+W+LS / huSP34. 1.3seFv AAS+SAV+ R 199 hCD4-D1.22 (bi valent) AAS+W+LS / huSP34. 1 . 3s cFv AAS+SAV+ R 200 hCD4 D1.22 (bi valent) AAS+SAV+ R / huSP34 .3.13scF v AAS+W+YT E 201 hCD4 D1.22 (bi valent) AAS+SAV+ R / huSP34 .3.13scF v AAS+W 202 hCD4 D1.22(bi valent) AAS+SAV / huSP34.3 .13scFv AAS+W 203 hCD4 D1.22(bi valent) AAS+SAV+ YTE / huSP 34.3.13s cFv AAS+W+YT E 204 hCD4 D1D2 Fc AAS+W+YT E / huSP34 .39.13 AAS+SAV+ R 205 hCD4 D1D2 Fc AAS+W / hu SP34.39. 13 AAS+SAV+ R 206 hCD4 D1D2 Fc AAS+SAV+ R / huSP34 .3.13 AAS+W+YT E 207 hCD4 D1D2 Fc AAS+SAV+ R / huSP34 .3.13 AAS+W 208 hCD4 D1D2 Fc AAS+SAV / huSP34.3 .13 AAS+W 209 hCD4 D1D2 Fc AAS+SAV+ YTE / huSP 34.3.13 AAS+W+YT E 210 CD4 D1D2 Fc AAS W / huSP34 .1.3 AAS+SAV+ R 211 hCD4-D1D2 Fc AAS+W+LS / huSP34. 1.3 AAS+SAV+ R 212 hCD4-D1D2 Fc AAS+W+LS / huSP34. 1.3scFv AAS+SAV+ R 213 hCD4-D1D2 (bivalen t AAS+W+LS / huSP34. 1.3scFv AAS+SAV+ R 214 hCD4 D1D2 (biv alent) AAS+SAV+ R / huSP34 .3.13scF v AAS+W+YT E 215 hCD4 D1D2 (biv alent) AAS+SAV+ R / huSP34 .3.13scF v AAS+W 216 hCD4 D1D2 (biv alent) AAS+SAV / huSP34.3 .13scFv AAS+W 217 hCD4 D1D2 (biv alent) AAS+SAV+ YTE / huSP 34.3.13s cFv AAS+W+YT E 218 h3BNC117 .52.64 AAS+SAV+ YTE / huSP 34.3.13s cFv AAS+W+YT E 219 h3BNC117 .52.64 AAS+SAV+ R / huSP34 .3.13scF v AAS+W+YT E 220 h3BNC117 .52.64 AAS+SAV+ R / huSP34 .3.13scF v AAS+W 221 h3BNC117 .52.64 AAS+SAV / huSP34.3 .13scFv AAS+W 222 h3BNC117 .52.64 AAS+SAV+ YTE / huSP 34.39.13 scFv AAS+W+YT E 223 h3BNC117 .52.64 AAS+SAV+ R / huSP34 .39.13sc Fv AAS+W+YT E 224 h3BNC117 .52.64 AAS+SAV+ R / huSP34 .39.13sc Fv AAS+W 225 h3BNC117 .52.64 AAS+SAV / huSP34.3 9.13scFv AAS+W 226 h3BNC117 .52.64 AAS+W+YT E / huSP34 .39.13sc Fv AAS+SAV+ R 227 h3BNC117 .52.64 AAS+W / hu SP34.39. 13scFv AAS+SAV+ R 228 h3BNC117 .52.64 AAS+SAV+ YTE / huSP 34.1.3sc Fv AAS+W+YT E 229 h3BNC117 .52.64 AAS+W+YT E / huSP34 .1.3 scFv AAS+SAV+ R 230 h3BNC117 .52.64 AAS+W / hu SP34.1.3 scFv AAS+SAV+ R 231 h3BNC117 .52.64 AAS+SAV / huSP34.1 .3scFv AAS+W 232 h3BNC117 .52.64 AAS+W+LS / huSP34. 1.3 scFv AAS+SAV+ R 233 h3BNC117 .52.64 AAS+SAV+ YTE / huSP 34.3.8sc Fv AAS+W+YT E 234 h3BNC117 .52.64 AAS+SAV / huSP34.3 .8scFv AAS+W 235 h3BNC117 .52.64 AAS+SAV+ YTE / huSP 34.34.3s cFv AAS+W+YT E 236 h3BNC117 .52.64 AAS+SAV / huSP34.3 4.3scFv AAS+W 239 hPGT121. 66 AAS+SAV+ YTE / huSP 34.3.13s cFv AAS+W+YT E 240 hPGT121. 66 AAS+SAV+ R / huSP34 .3.13scF v AAS+W+YT E 241 hPGT121. 66 AAS+SAV+ R / huSP34 .3. 13scF v AAS+W 242 hPGT121. 66 AAS+SAV / huSP34.3 .13scFv AAS+W 243 hPGT121. 66 AAS+W / hu SP34.3.1 3 scFv AAS+SAV+ R 244 hPGT121. 66 AAS+SAV+ YTE / huSP 34.39.13 scFv AAS+W+YT E 245 hPGT121. 66 AAS+SAV+ R / huSP34 .39.13sc Fv AAS+W+YT E 246 hPGT121. 66 AAS+SAV+ R / huSP34 .39.13sc Fv AAS+W 247 hPGT121. 66 AAS+SAV / huSP34.3 9.13scFv AAS+W 248 hPGT121. 66 AAS+W+YT E / huSP34 .39.13sc Fv AAS+SAV+ R 249 hPGT121. 66 AAS+W / hu SP34.39. 13scFv AAS+SAV+ R 250 hPGT121. 66 AAS+SAV / huSP34.1 .3scFv AAS+W 251 hPGT121. 66 AAS+W / hu SP34.1.3 scFv AAS+SAV+ RF 252 hPGT121. 66 AAS+W / hu SP34.1.3 scFv AAS+SAV+ R 253 hPGT121. 66 AAS+W+YT E / huSP34 .1.3 scFv AAS+SAV+ R 254 hPGT121. 66 AAS+W+LS / huSP34. 1.3 scFv AAS+SAV+ R 255 hPGT121. 66 W / huSP34 .1.3scFv SAV+R 256 hPGT121. 66 AAS+W / hu SP34.3.8 scFv AAS+SAV+ R 257 hPGT121. 66 AAS+W / hu SP34.34. 3scFv AAS+SAV+ R 278 10-1074 AAS+SAV+ YTE / huSP 34.3.13s cFv AAS+W+YT E 279 10-1074 AAS+SAV+ R / huSP34 .3.13scF v AAS+W+YT E 280 10-1074 AAS+SAV+ R / huSP34 .3.13scF v AAS+W 281 10-1074 AAS+SAV / huSP34.3 .13scFv AAS+W 282 10-1074 AAS+W / hu SP34.3.1 3scFv AAS+SAV+ R 283 10-1074 AAS+SAV+ YTE / huSP 34.39.13 scFv AAS+W+YT E 284 10-1074 AAS+SAV+ R / huSP34 .39.13sc Fv AAS+W+YT E 285 10-1074 AAS+SAV+ R / huSP34 .39.13sc Fv AAS+W 286 10-1074 AAS+SAV / huSP34.3 9.13scFv AAS+W 287 10-1074 AAS+W+YT E / huSP34 .39.13sc Fv AAS+SAV+ R 288 10-1074 AAS+W / hu SP34.39. 13scFv AAS+SAV+ R 289 10-1074 AAS+SAV+ YTE / huSP 34.1.3sc Fv AAS+W+YT E 290 10-1074 AAS+SAV / huSP34.1 .3scFv AAS+W 291 10-1074 AAS+W+YT E / huSP34 .1.3 scFv AAS+SAV+ R 292 10-1074 AAS+W / hu SP34.1.3 scFv AAS+SAV+ R 293 10-1074 AAS+W+LS / huSP34. 1.3 scFv AAS+SAV+ R 294 10-1074 AAS+SAV+ YTE / huSP 34.3.8sc Fv AAS+W+YT E 295 10-1074 AAS+SAV / huSP34.3 .8scFv AAS+W 296 10-1074 AAS+SAV+ YTE / huSP 34.34.3s cFv AAS+W+YT E 297 10-1074 AAS+SAV / huSP34.3 4.3scFv AAS+W 298 PGT-134 AAS+SAV+ YTE / huSP 34.3.13s cFv AAS+W+YT E 299 PGT-134 AAS+SAV+ R / huSP34 .3.13scF v AAS+W+YT E 300 PGT-134 AAS+SAV+ R / huSP34 .3.13scF v AAS+W 301 PGT-134 AAS+SAV / huSP34.3 .13scFv AAS+W 302 PGT-134 AAS+W / hu SP34.3.1 3scFv AAS+SAV+ R 303 PGT-134 AAS+SAV+ YTE / huSP 34.39.13 scFv AAS+W+YT E 304 PGT-134 AAS+SAV+ R / huSP34 .39.13sc Fv AAS+W+YT E 305 PGT-134 AAS+SAV+ R / huSP34 .39.13sc Fv AAS+W 306 PGT-134 AAS+SAV / huSP34.3 9.13scFv AAS+W 307 PGT-134 AAS+W+YT E / huSP34 .39.13sc Fv AAS+SAV+ R 308 PGT-134 AAS+W / hu SP34.39. 13scFv AAS+SAV+ R 309 PGT-134 AAS+SAV+ YTE / huSP 34.1.3sc Fv AAS+W+YT E 310 PGT-134 AAS+SAV / huSP34.1 .3scFv AAS+W 311 PGT-134 AAS+W+YT E / huSP34 .1.3 scFv AAS+SAV+ R 312 PGT-134 AAS+W / hu SP34.1.3 scFv AAS+SAV+ R 313 PGT-134 AAS+W+LS / huSP34. 1.3 scFv AAS+SAV+ R 314 PGT-134 AAS+SAV+ YTE / huSP 34.3.8sc Fv AAS+W+YT E 315 PGT-134 AAS+SAV / huSP34.3 .8scFv AAS+W 316 PGT-134 AAS+SAV+ YTE / huSP 34.34.3s cFv AAS+W+YT E 317 PGT-134 AAS+SAV / huSP34.3 4.3scFv AAS+W

[0133] In some embodiments, one or more of the polypeptides comprising the multispecific antigen binding molecules described herin comprise an N-terminal signal peptide. The signal peptide can be an endogenous signal peptide (e.g., from a native or wild-type immunoglobulin or CD4 ECD protein), or from a heterologous polypeptide. In various embodiments, the signal peptide or leader sequence is from a source protein selected from a serum protein, an immunoglobulin, a cytokine, a chemokine, a chaperone protein, an invariant protein, and a protein that directs proteins to the lysosomal compartment. In various embodiments, the signal peptide or leader sequence is from a source protein selected from colony stimulating factor 2 (CSF2, GM-CSF), tissue type plasminogen activator (PLAT, t-PA), C-C motif chemokine ligand 7 (CCL7, MCP-3), C-X-C motif chemokine ligand 10 (CXCL10, IP-10), catenin beta 1 (CTNNB1), CD74 (p33; DHLAG; HLADG; immunoglobulin Kappa; Ia-GAMMA, invariant chain), serum albumin (ALB), SPARC (osteonectin), cwcv and kazal like domains proteoglycan 1 (SPOCK1); SPARC (osteonectin), cwcv and kazal like domains proteoglycan 2 (SPOCK2); polyubiquitin B / C (UBB / UBC), calreticulin (CALR), vesicular stomatitis virus G protein (VSV-G), lysosomal associated membrane protein 1 (LAMP-1) and lysosomal associated membrane protein 2 (LAMP-2). In some embodiments, the signal peptide is from a serum albumin signal peptide (e.g., comprising the amino acid sequence KWVTFISLLFLFSSAYS (SEQ ID NO: 1026)). In various embodiments, the signal peptide or leader sequence is selected from an amino acid sequence of any one of SEQ ID NOs: 1026-1039, or a sequence that is at least 95%, 96%, 97%, 98%, or 99% identical to any one of SEQ ID NOs: 1026-1039. Illustrative signal sequences that can be used in the present multi-specific binding proteins are provided in Table H. TABLE H - illustrative signal sequences SEQ ID NO: source protein name SEQUENCE 1026albuminKWVTFISLLFLFSSAYS1027IL-2MYRMQLLSCIALSLALVTNS1028SPOCK1MPAIAVLAAAAAAWCFLQVES1029SPOCK2MRAPGCGRLVLPLLLLAAAALA1030Ig KappaMETDTLLLWVLLLWVPG1031CSF2, GM-CSFMWLQSLLLLGTVACSISV1032PLAT, t-PAMDAMKRGLCCVLLLCGAVFVSAR1033CD74MHRRRSRSCREDQKPV1034β-cateninMRKAAVSHWQQQSYLDSGIHSGATTTAPSLS1035CCL7, MCP-31036ubiquitin1037calreticulinMLLSVPLLLGLLGLAVA1038VSV-GMKCLLYLAFLFIGVNC1039CXCL10, IP-10MNQTAILICCLIFLTLSGIQG

[0134] The signal peptide can be designed to be cleaved off, e.g., after secretion from the cell, to form a mature fusion protein. A modified human serum albumin signal peptide to secrete proteins in cells that can find use in expressing the present fusion proteins is described, e.g., in Attallah, et al., Protein Expr Purif. (2017) 132:27-33. Additional signal peptide sequences for use in expressing the herein described fusion proteins are described, e.g., in Kober, et al., Biotechnol Bioeng. (2013) 110(4):1164-73.

[0135] In various embodiments, multi-specific antigen binding molecules described herein, and / or the polynucleotides encoding such polypeptides, are provided in provided in isolated form. This means that such the polypeptide or polynucleotide is at least 50% w / w pure of interfering proteins, cellular and other contaminants arising from its production or purification but does not exclude the possibility that the agent is combined with an excess of pharmaceutical acceptable carrier(s) or other vehicle intended to facilitate its use. The term "isolated," when applied to a polypeptide or polynucleotide, as described herein, denotes that the polypeptide or polynucleotide is essentially free of cellular components with which it is associated in the natural state. It can be, for example, in a homogeneous state and may be in either a dry or aqueous solution. Purity and homogeneity can be determined using known methods, e.g., analytical chemistry techniques such as polyacrylamide gel electrophoresis, column chromatography, thin layer chromatography, or high-performance liquid chromatography (HPLC) analysis. A protein that is the predominant species present in a preparation is substantially purified. An "isolated" or "purified" polypeptide or polynucleotide is substantially free of other cellular material, or culture medium when produced by recombinant techniques, or chemical precursors or other chemicals when chemically synthesized. In various embodiments, purified polypeptides and / or polynucleotides are at least 60%, 70%, 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% (w / w), separated from, purified of, or free of interfering proteins and contaminants from production or purification. Often a multi-specific antigen binding molecule is the predominant macromolecular species remaining after its purification.5. Conjugated Multi-Specific Antigen Binding Molecules

[0136] Any of the multi-specific antigen binding molecules described herein may be conjugated antigen binding molecules which are bound to various molecules (e.g., labels) including without limitation macromolecular substances such as polymers (e.g., polyethylene glycol (PEG), polyethylenimine (PEI) modified with PEG (PEI-PEG), polyglutamic acid (PGA) (N-(2-Hydroxypropyl) methacrylamide (HPMA) copolymers), hyaluronic acid, radioactive materials (e.g. 90< Y, 131< I, 12s< I 35< S, 3< H, 121< In, 99< Tc ), fluorescent substances (e.g., fluorescein and rhodamine), luminescent substances (e.g., luminol), haptens, enzymes (e.g., glucose oxidase), metal chelates, biotin, avidin, and drugs.

[0137] The above-described conjugated multi-specific antigen binding molecules can be prepared according to known methods, e.g., performing chemical modifications on the antigen binding molecules or the lower molecular weight forms thereof described herein. Methods for modifying antibodies are well known in the art (e.g., US 5,057,313 and US 5,156,840).6. Polynucleotides Encoding Multi-Specific Antigen Binding Molecules

[0138] Provided are polynucleotides encoding the multi-specific antigen binding molecules, vectors comprising such polynucleotides, and host cells (e.g., mammalian cells, plant cells, yeast cells, bacteria cells including E. coli cells) comprising such polynucleotides or expression vectors. Provided herein are polynucleotides comprising nucleotide sequence(s) encoding any of the multi-specific antigen binding molecules provided herein, as well as expression cassettes and vector(s) comprising such polynucleotide sequences, e.g., expression vectors for their efficient expression in host cells, e.g., mammalian cells.

[0139] Further provided are polynucleotides or nucleic acid molecules encoding the multi-specific antigen binding molecule. In some embodiments, the polynucleotides encode an immunoglobulin heavy chain variable region (or a fragment thereof) and an immunoglobulin light chain variable region (or a fragment thereof), of one or both of the anti-CD3 binding domain and an anti-HIV antigen (e.g., anti-gp120, anti-gp41). In other embodiments, the polynucleotides or nucleic acid molecules are DNA, cDNA, or mRNA. In some other embodiments, the polynucleotides or nucleic acid molecules are codon-biased to enhance expression in a desired host cell.

[0140] Disclosed are polynucleotides encoding the VH, VL, or VH and VL of one or both of the first and second binding domains of the multi-specific antigen binding molecules or antigen binding fragments which bind to gp120. In certain instances, the polynucleotides encode one or more of CDRs, VH, VL, HC, LC and Fc regions comprising the amino acid sequences of the multi-specific antigen binding molecules set forth in one or more of Tables A1-A4, B1-B2, C, D1-D4, E, F and G, as described herein.

[0141] Disclosed herein are polynucleotides encoding the CDRs, VH, VL, light chain, or heavy chain of a multi-specific antigen binding molecule, described herein. The polynucleotides may comprise nucleotide sequences encoding a heavy chain or heavy chain variable domain targeting CD3 comprising the VH CDRs of antibodies described herein (see, e.g., Tables A1-A4, B1, C and G, herein). The polynucleotides may comprise nucleotide sequences encoding a light chain or light chain variable domain targeting CD3 comprising the VL CDRs of multi-specific antigen binding molecules described herein (see, e.g., Tables A1-A4, B2, C and G, herein). A polynucleotide described herein encodes a heavy chain variable region or a heavy chain comprising VH-CDRs targeting CD3 comprising the amino acid sequences set forth in Tables A1, A2, A3 or A4, respectively. A polynucleotide described herein encodes a light chain variable region or a light chain comprising VL-CDRs targeting CD3 comprising the amino acid sequences set forth in Tables A1, A2, A3 or A4, respectively. In some embodiments, the polynucleotide or polynucleotides encode a first heavy chain variable domain (VH) and a first light chain variable domain (VL) that target, bind to or specifically bind to CD3, comprising a first VH-CDR1, a first VH-CDR2, a first VH-CDR3, a first VL-CDR1, a first VL-CDR2 and a first VL-CDR3 comprising the following amino acid sequences (according to Kabat), respectively: SEQ ID NOs: 1, 12, 8, 4, 9 and 10.

[0142] The polynucleotide or polynucleotides may encode a first heavy chain variable domain (VH) and a first light chain variable domain (VL) that target, bind to or specifically bind to CD3, comprising a first VH-CDR1, a first VH-CDR2, a first VH-CDR3, a first VL-CDR1, a first VL-CDR2 and a first VL-CDR3 comprising the following amino acid sequences (according to Chothia), respectively: SEQ ID NOs: 17, 18, 23, 20, 24 and 25.

[0143] The polynucleotide or polynucleotides may encode a first heavy chain variable domain (VH) and a first light chain variable domain (VL) that target, bind to or specifically bind to CD3, comprising a first VH-CDR1, a first VH-CDR2, a first VH-CDR3, a first VL-CDR1, a first VL-CDR2 and a first VL-CDR3 comprising the following amino acid sequences (according to IMGT), respectively: SEQ ID NOs: 28, 29, 32, 31, 24 and 10.

[0144] The polynucleotide or polynucleotides may encode a first heavy chain variable domain (VH) and a first light chain variable domain (VL) that target, bind to or specifically bind to CD3, comprising a first VH-CDR1, a first VH-CDR2, a first VH-CDR3, a first VL-CDR1, a first VL-CDR2 and a first VL-CDR3 comprising the following amino acid sequences (according to Honegger), respectively: SEQ ID NOs: 34, 43, 40, 37, 41 and 25.

[0145] In some embodiments, the polynucleotide or polynucleotides encode a first VH and a first VL, the first VH and the first VL comprising amino acid sequences that are at least 99% identical to the amino acid sequence set forth, respectively: SEQ ID NO: 51 and 56. In some embodiments, the polynucleotide or polynucleotides encode a first VH and a first VL, the first VH and the first VL comprising amino acid sequences that are 100% identical to the amino acid sequence set forth, respectively: SEQ ID NO: 51 and 56. In some embodiments, the polynucleotide or polynucleotides encode a first VH and a first VL, the first VH comprising the amino acid sequence of SEQ ID NO: 51 and the first VL comprising an amino acid sequence that is at least 99% identical to the amino acid sequence of SEQ ID NO: 56. In some embodiments, the polynucleotide or polynucleotides encode a first VH and a first VL, the first VH comprising an amino acid sequence that is at least 99% identical to the amino acid sequence of SEQ ID NO: 51 and the first VL comprising the amino acid sequence of SEQ ID NO: 56.

[0146] The polynucleotide may encode a scFv comprising a VH and a VL, the scFv comprises an amino acid sequence that is at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or at least 100%, identical to an amino acid sequence selected from SEQ ID NOs: 59-66, e.g., SEQ ID NOs: 59-66, e.g., SEQ ID NOs: 59-63, e.g., SEQ ID NOs: 61, 62 or 63, e.g., SEQ ID NOs: 62 or 63.

[0147] Described herein are polynucleotides encoding the CDRs, VH, VL, light chain, or heavy chain of a multi-specific antigen binding molecule, described herein. The polynucleotides can comprise nucleotide sequences encoding a heavy chain or heavy chain variable domain targeting an HIV antigen (e.g., HIV gp120) comprising the VH CDRs of antibodies described herein (see, e.g., Tables D1-D4, E and G, herein). The polynucleotides can comprise nucleotide sequences encoding a light chain or light chain variable domain targeting an HIV antigen (e.g., HIV gp120) comprising the VL CDRs of multi-specific antigen binding molecules described herein (see, e.g., Tables D1-D4, E and G, herein). A polynucleotide described herein encodes a heavy chain variable region or a heavy chain comprising VH-CDRs targeting an HIV antigen (e.g., HIV gp120) comprising the amino acid sequences set forth in Tables D1, D2, D3 or D4, respectively. A polynucleotide described herein encodes a light chain variable region or a light chain comprising VL-CDRs targeting an HIV antigen (e.g., HIV gp120) comprising the amino acid sequences set forth in Tables D1, D2, D3 or D4, respectively.

[0148] A polynucleotide described herein encodes a heavy chain variable region and a heavy chain targeting an HIV antigen (e.g., HIV gp120) comprising the amino acid sequences set forth in Table E.

[0149] Polynucleotide or polynucleotides may encode a first Fc region and a second Fc region, as set forth in Table F.

[0150] In some embodiments, the polynucleotide or polynucleotides encode an unpaired heavy chain a Fab arm heavy chain and a Fab arm light chain as set forth in Table G. Illustrative polynucleotide sequences are provided in Table J. In some embodiments, the polynucleotide or polynucleotides encode an unpaired heavy chain a Fab arm heavy chain and a Fab arm light chain comprising amino acid sequences comprising amino acid sequences that are at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99% identical, or is 100% identical, to the amino acid sequences set forth, respectively: SEQ ID NOs: 995, 996 and 997; or SEQ ID NOs: 998, 999 and 1000.

[0151] In some embodiments, the polynucleotide or polynucleotides encode an unpaired heavy chain a Fab arm heavy chain and a Fab arm light chain comprising amino acid sequences that are at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99% identical, or is 100% identical, to the amino acid sequences set forth, respectively: SEQ ID NOs: 995, 996 and 997. In some embodiments, the polynucleotide or polynucleotides encode an unpaired heavy chain a Fab arm heavy chain and a Fab arm light chain comprising amino acid sequences that are at least 95% identical to the amino acid sequences set forth, respectively: SEQ ID NOs: 995, 996 and 997. In some embodiments, the polynucleotide or polynucleotides encode an unpaired heavy chain a Fab arm heavy chain and a Fab arm light chain comprising amino acid sequences that are at least 99% identical to the amino acid sequences set forth, respectively: SEQ ID NOs: 995, 996 and 997. In some embodiments, the polynucleotide or polynucleotides encode an unpaired heavy chain a Fab arm heavy chain and a Fab arm light chain comprising amino acid sequences that are 100% identical to the amino acid sequences set forth, respectively: SEQ ID NOs: 995, 996 and 997.

[0152] In some embodiments, the polynucleotide or polynucleotides encode an unpaired heavy chain a Fab arm heavy chain and a Fab arm light chain comprising amino acid sequences that are at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99% identical, or is 100% identical, to the amino acid sequences set forth, respectively: SEQ ID NOs: 998, 999 and 1000. In some embodiments, the polynucleotide or polynucleotides encode an unpaired heavy chain a Fab arm heavy chain and a Fab arm light chain comprising amino acid sequences that are at least 95% identical to the amino acid sequences set forth, respectively: SEQ ID NOs: 998, 999 and 1000. In some embodiments, the polynucleotide or polynucleotides encode an unpaired heavy chain a Fab arm heavy chain and a Fab arm light chain comprising amino acid sequences that are at least 99% identical to the amino acid sequences set forth, respectively: SEQ ID NOs: 998, 999 and 1000. In some embodiments, the polynucleotide or polynucleotides encode an unpaired heavy chain a Fab arm heavy chain and a Fab arm light chain comprising amino acid sequences that are 100% identical to the amino acid sequences set forth, respectively: SEQ ID NOs: 998, 999 and 1000.

[0153] Also encompassed by this disclosure are polynucleotides encoding an anti-gp120 antigen binding molecule or antigen binding fragment thereof, or an anti-CD3 antigen binding domain or antigen binding fragment thereof, that have at least one of codon-biased sequences for improved expression in a desired host cell, replacement heterologous signal sequences, and reduced or eliminated mRNA instability elements. Methods to generate codon-biased nucleic acids can be carried out by adapting the methods described in, e.g., U.S. Patent Nos. 5,965,726; 6,174,666; 6,291,664; 6,414,132; and 6,794,498. Preferred codon usage for expression of the multi-specific antigen binding molecules in desired host cells is provided, e.g., at kazusa.or.jp / codon / ; and genscript.com / tools / codon-frequency-table.

[0154] As appropriate, in certain embodiments, the 3'-end of the polynucleotides encoding the multi-specific antigen binding molecules comprises one or multiple tandem stop codons, e.g., two or more tandem TAG ("amber"), TAA ("ochre") or TGA ("opal" or "umber") stop codons. The multiple tandem stop codons can be the same or different. TABLE J - polynucleotide encoding three chains of illustrative CD3 / gp120-targeting bi-specific binding molecules Ab nameunpaired HCFab Arm - HCFab Arm - LC430SEQ ID NO: 995SEQ ID NO: 996SEQ ID NO: 997hCD4 D1.22 Fc AAS+W +YTE / huSP3 4.39. 13 AAS+S AV+R431SEQ ID NO: 998SEQ ID NO: 999SEQ ID NO: 1000hCD4 D1.22 Fc AAS+W +YTE / huSP34.39. 13 AAS+S AV+R432SEQ ID NO: 1001SEQ ID NO: 1002SEQ ID NO: 1003hCD4 D1.22 Fc AAS+S AV+R / huSP3 4.3.1 3 AAS+W +YTE433SEQ ID NO: 1004SEQ ID NO: 1005SEQ ID NO: 1000hCD4 D1.22 Fc AAS+S AV+R / huSP3 4.3.1 3 AAS+W +YTE434SEQ ID NO: 1006SEQ ID NO: 1002SEQ ID NO: 997hCD4 D1.22 FcAAS+S AV+YT E / huS P34.3 .13 AAS+W +YTE435SEQ ID NO: 1007SEQ ID NO: 1093SEQ ID NO: 1000hCD4 D1.22 Fc AAS+S AV+YT E / huS P34.3 .13 AAS+W +YTE436SEQ ID NO: 998SEQ ID NO: 1008SEQ ID NO: 1000hCD4 D1.22 Fc AAS+W +YTE / huSP3 4.40. 13 AAS+S AV+R437SEQ ID NO: 998SEQ ID NO: 1009SEQ ID NO: 1000hCD4 D1.22 Fc AAS+W +YTE / huSP3 4.41. 13 AAS+S AV+R438SEQ ID NO: 1010SEQ ID NO: 1011SEQ ID NO: 1012h3BNC 117.5 2.64 AAS+ SAV+Y TE / huSP34. 3.13 scFv AAS+W +YTE439SEQ ID NO: 1013SEQ ID NO: 1014SEQ ID NO: 1015h3BNC 117.5 2.64 AAS+S AV+YT E / huS P34.3 .13 scFv AAS+W +YTE440SEQ ID NO: 1016SEQ ID NO: 1017SEQ ID NO: 1012h3BNC 117.5 2.64 AAS+S AV / hu SP34. 3.13s cFv AAS+W441SEQ ID NO: 1018SEQ ID NO: 1019SEQ ID NO: 1012h3BNC 117.5 2.64 AAS+W +YTE / huSP3 4.39. 13scF v AAS+S AV+R442SEQ ID NO: 1018SEQ ID NO: 1020SEQ ID NO: 1012h3BNC 117.5 2.64 AAS+W / huSP 34.39 .13sc Fv AAS+S AV+R443SEQ ID NO: 1021SEQ ID NO: 1022SEQ ID NO: 1023hPGT1 21.66 AAS+W / huSP 34.3. 13scF v AAS+S AV+R444SEQ ID NO: 1024SEQ ID NO: 1025SEQ ID NO: 1023hPGT1 21.66 AAS+W / huSP 34.39 .13 scFv AAS+S AV+R

[0155] In some embodiments, the one or more polynucleotides encoding the antibodies or antigen-binding fragments, described herein, are formulated or encapsulated in a lipoplex, e.g., a lipid nanoparticle (LNP). As used herein, a "lipoplex" refers to cationic liposomes that are nonviral (synthetic) lipid carriers of DNA. In some embodiments the lipoplex is a lipid nanoparticle (LNP). As used herein, the term "lipid nanoparticle" refers to one or more spherical nanoparticles with an average diameter of between 10 to 1000 nanometers, and which comprise a solid lipid core matrix that can solubilize lipophilic molecules. In certain embodiments, the lipid core is stabilized by surfactants (e.g., emulsifiers), and can comprise one or more of triglycerides (e.g., tristearin), diglycerides (e.g., glycerol bahenate), monoglycerides (e.g., glycerol monostearate), fatty acids (e.g., stearic acid), steroids (e.g., cholesterol), and waxes (e.g., cetyl palmitate), including combinations thereof. Lipid nanoparticles are described, for example, in Petrilli et al., Curr Pharm Biotechnol. 15:847-55, 2014; and U.S. Patent Nos. 6,217,912; 6,881,421; 7,402,573; 7,404,969; 7,550,441; 7,727,969; 8,003,621; 8,691,750; 8,871,509; 9,017,726; 9,173,853; 9,220,779; 9,227,917; and 9,278,130. LNP-encapsulated mRNA molecules encoding a broadly neutralizing antibody are described, e.g., in Pardi, et al., Nat Commun. (2017) 8:14630. In certain embodiments, the one or more polynucleotides encoding the antibodies or antigen-binding fragments, described herein, are formulated or encapsulated in an LNP comprised of an ionizable cationic lipid / phosphatidylcholine / cholesterol / PEG-lipid, e.g., in molar ratios of about 50:10:38.5:1.5 mol mol -1< , respectively.7. Vectors and Host Cells

[0156] Further provided are vectors comprising one or more polynucleotides encoding one or more of the multi-specific antigen binding molecules. A vector can be of any type, for example, a recombinant vector such as an expression vector. Vectors include, but are not limited to, plasmids, cosmids, bacterial artificial chromosomes (BAC) and yeast artificial chromosomes (YAC) and vectors derived from bacteriophages or plant or animal (including human) viruses. Vectors can comprise an origin of replication recognized by the proposed host cell (e.g., including prokaryotic and eukaryotic host cells) and in the case of expression vectors, promoter and other regulatory regions recognized by the host cell. In additional embodiments, a vector comprises a polynucleotide encoding an antibody of the disclosure operably linked to a promoter and optionally additional regulatory elements. Certain vectors are capable of autonomous replication in a host into which they are introduced (e.g., vectors having a bacterial origin of replication can replicate in bacteria). Other vectors can be integrated into the genome of a host upon introduction into the host, and thereby are replicated along with the host genome. Vectors include, but are not limited to, those suitable for recombinant production of the antibodies disclosed herein.

[0157] The choice of the vector is dependent on the recombinant procedures followed and the host used. Introduction of vectors into host cells can be effected by inter alia calcium phosphate transfection, virus infection, DEAE-dextran-mediated transfection, lipofectamine transfection or electroporation. Vectors may be autonomously replicating or may replicate together with the chromosome into which they have been integrated. In certain embodiments, the vectors contain one or more selection markers. The choice of the markers may depend on the host cells of choice. These include, but are not limited to, kanamycin, neomycin, puromycin, hygromycin, zeocin, ampicillin (AmpR), thymidine kinase gene from Herpes simplex virus (HSV-TK), mammalian glutamine synthetase (GS) and dihydrofolate reductase gene from mouse (dhfr). Vectors comprising one or more nucleic acid molecules encoding the antibodies described herein, operably linked to one or more nucleic acid molecules encoding proteins or peptides that can be used to isolate the antibodies, are also covered by the disclosure. These proteins or peptides include, but are not limited to, glutathione-S-transferase, maltose binding protein, metal-binding polyhistidine, green fluorescent protein, luciferase and beta-galactosidase. In other embodiments, the vector that is used is or is based on pcDNA ™< 3.1+ (ThermoFisher, MA) or pCGS3. In various embodiments, the vector can have one, two, three, four or five open reading frames or expression cassettes. In various embodiments, the vector can have one, two, three, four or five cistrons. In some embodiments, the first, second and third expression cassettes (e.g., as set forth below) each comprise a promoter of identical or equivalent transcription strength. In varying embodiments, the promoter is a constitutive promoter. In some embodiments, the first, second and third expression cassettes (e.g., as set forth below) comprise one or more promoters of different transcription strength. Illustrative promoters of use include without limitation cytomegalovirus (CMV), SV40, RSV, EF1a, UBC, PGK and CAGG (see, e.g., Qin, et al., PLoS One. (2010) 5(5):e10611). In embodiments, the expression vector further comprises a fourth expression cassette positioned 5' to the first expression cassette comprising a polynucleotide encoding a eukaryotic selection marker protein, e.g., glutamine synthetase (GS). Generally, the promoter driving expression of the polynucleotide encoding a eukaryotic selection marker protein (fourth expression cassette) has a relatively weaker transcription strength in comparison to the promoters driving expression of the polynucleotide in the first, second and third expression cassettes.

[0158] In some embodiments, the expression vector or expression vectors comprise a plasmid vector or a viral vector. In some embodiments, the expression vector comprises three, four or five expression cassettes or cistrons.

[0159] In some embodiments, the expression vector comprises, optionally in sequential order from 5' to 3': (i) a first expression cassette comprising a first polynucleotide encoding an anti-CD3 VL-CL fusion protein; (ii) a second expression cassette comprising a second polynucleotide encoding an anti-CD3 VH-Fc fusion protein; and (iii) a third expression cassette comprising a third polynucleotide encoding a CD4 extracellular (EC) domain-Fc fusion protein. In some embodiments, the anti-CD3 VL-CL fusion protein, the anti-CD3 VH-Fc fusion protein and the CD4 EC domain-Fc fusion protein comprise amino acid sequences set forth, respectively, below, or comprise amino acid sequences that are at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% identical to the amino acid sequences set forth, respectively, below: SEQ ID NOs: 753, 752 and 751; or SEQ ID NOs: 753, 752 and 754. In some embodiments, the anti-CD3 VL-CL fusion protein, the anti-CD3 VH-Fc fusion protein and the CD4 EC domain-Fc fusion protein comprise amino acid sequences that are at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% identical to the amino acid sequences set forth, respectively: SEQ ID NOs: 753, 752 and 751. In some embodiments, the anti-CD3 VL-CL fusion protein, the anti-CD3 VH-Fc fusion protein and the CD4 EC domain-Fc fusion protein comprise amino acid sequences set forth, respectively: SEQ ID NOs: 753, 752 and 751.

[0160] In some embodiments, the anti-CD3 VL-CL fusion protein, the anti-CD3 VH-Fc fusion protein and the CD4 EC domain-Fc fusion protein comprise amino acid sequences that are at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% identical to the amino acid sequences set forth, respectively: SEQ ID NOs: 753, 752 and 751. In some embodiments, the anti-CD3 VL-CL fusion protein, the anti-CD3 VH-Fc fusion protein and the CD4 EC domain-Fc fusion protein comprise amino acid sequences that are at least 95% identical to the amino acid sequences set forth, respectively: SEQ ID NOs: 753, 752 and 751. In some embodiments, the anti-CD3 VL-CL fusion protein, the anti-CD3 VH-Fc fusion protein and the CD4 EC domain-Fc fusion protein comprise amino acid sequences that are at least 99% identical to the amino acid sequences set forth, respectively: SEQ ID NOs: 753, 752 and 751. In some embodiments, the anti-CD3 VL-CL fusion protein comprises an amino acid sequence that is at least 95% (e.g., 99%) identical to the amino acid sequence of SEQ ID NO: 753, the anti-CD3 VH-Fc fusion protein comprises an amino acid sequence that is at least 95% (e.g., 99%) identical to the amino acid sequence of SEQ ID NO: 752, and the CD4 EC domain-Fc fusion protein comprises an amino acid sequence that is at least 95% (e.g., 99%) identical to the amino acid sequence of SEQ ID NO: 751. In some embodiments, the anti-CD3 VL-CL fusion protein comprises an amino acid sequence that is at least 95% identical to the amino acid sequence of SEQ ID NO: 753, the anti-CD3 VH-Fc fusion protein comprises an amino acid sequence of SEQ ID NO: 752, and the CD4 EC domain-Fc fusion protein comprises an amino acid sequence that is at least 95% (e.g., 99%) identical to the amino acid sequence of SEQ ID NO: 751. In some embodiments, the anti-CD3 VL-CL fusion protein comprises an amino acid sequence that is at least 99% identical to the amino acid sequence of SEQ ID NO: 753, the anti-CD3 VH-Fc fusion protein comprises an amino acid sequence of SEQ ID NO: 752, and the CD4 EC domain-Fc fusion protein comprises an amino acid sequence that is at least 95% (e.g., 99%) identical to the amino acid sequence of SEQ ID NO: 751. In some embodiments, the anti-CD3 VL-CL fusion protein comprises an amino acid sequence of SEQ ID NO: 753, the anti-CD3 VH-Fc fusion protein comprises an amino acid sequence that is at least 95% identical to the amino acid sequence of SEQ ID NO: 752, and the CD4 EC domain-Fc fusion protein comprises an amino acid sequence that is at least 95% (e.g., 99%) identical to the amino acid sequence of SEQ ID NO: 751. In some embodiments, the anti-CD3 VL-CL fusion protein comprises an amino acid sequence of SEQ ID NO: 753, the anti-CD3 VH-Fc fusion protein comprises an amino acid sequence that is at least 99% identical to the amino acid sequence of SEQ ID NO: 752, and the CD4 EC domain-Fc fusion protein comprises an amino acid sequence that is at least 95% (e.g., 99%) identical to the amino acid sequence of SEQ ID NO: 751. In some embodiments, the anti-CD3 VL-CL fusion protein, the anti-CD3 VH-Fc fusion protein and the CD4 EC domain-Fc fusion protein comprise amino acid sequences set forth, respectively: SEQ ID NOs: 753, 752 and 751.

[0161] The disclosure also provides host cells comprising one or more recombinant polynucleotides or one or more vectors, as described herein. Any of a variety of host cells can be used. In one embodiment, a host cell is a prokaryotic cell, for example, E. coli. In another embodiment, a host cell is a eukaryotic cell, for example, a yeast cell, a plant cell, an insect cell, a mammalian cell, such as a Chinese Hamster Ovary (CHO)- based or CHO-origin cell (e.g., CHO-S, CHO DG44, ExpiCHO ™< , CHOZN ®< ZFN-modified GS- / - CHO cell line or CHO-K1, CHO-K1a cells), COS cells, BHK cells, NSO cells or Bowes melanoma cells. Examples of human host cells are, inter alia, HeLa, 911, AT1080, A549, 293, Expi293 ™< and HEK293T-cells.

[0162] The terms "polynucleotide" and "nucleic acid molecule" interchangeably refer to a polymeric form of nucleotides and includes both sense and anti-sense strands of RNA, cDNA, genomic DNA, and synthetic forms and mixed polymers of the above. As used herein, the term nucleic acid molecule may be interchangeable with the term polynucleotide. In some embodiments, a nucleotide refers to a ribonucleotide, deoxynucleotide or a modified form of either type of nucleotide, and combinations thereof. The terms also include, but are not limited to, single- and double-stranded forms of DNA. In addition, a polynucleotide, e.g., a cDNA or mRNA, may include either or both naturally occurring and modified nucleotides linked together by either or both of naturally occurring and non-naturally occurring nucleotide linkages. The nucleic acid molecules may be modified chemically or biochemically or may contain non-natural or derivatized nucleotide bases, as will be readily appreciated by those of skill in the art. Such modifications include, for example, labels, methylation, substitution of one or more of the naturally occurring nucleotides with an analogue, internucleotide modifications such as uncharged linkages (e.g., methyl phosphonates, phosphotriesters, phosphoramidates, carbamates, etc.), charged linkages (e.g., phosphorothioates, phosphorodithioates, etc.), pendent moieties (e.g., polypeptides), intercalators (e.g., acridine, psoralen, etc.), chelators, alkylators, and modified linkages (e.g., alpha anomeric nucleic acids, etc.). The above term is also intended to include any topological conformation, including single-stranded, double-stranded, partially duplexed, triplex, hairpinned, circular and padlocked conformations. A reference to a nucleic acid sequence encompasses its complement unless otherwise specified. Thus, a reference to a nucleic acid molecule having a particular sequence should be understood to encompass its complementary strand, with its complementary sequence. The term also includes codon-biased polynucleotides for improved expression in a desired host cell.

[0163] The term "operably linked" refers to two or more nucleic acid sequence elements that are usually physically linked and are in a functional relationship with each other. For instance, a promoter is operably linked to a coding sequence if the promoter is able to initiate or regulate the transcription or expression of a coding sequence, in which case, the coding sequence should be understood as being "under the control of" the promoter.

[0164] A "substitution," as used herein, denotes the replacement of one or more amino acids or nucleotides by different amino acids or nucleotides, respectively.

[0165] An "isolated" nucleic acid refers to a nucleic acid molecule that has been separated from a component of its natural environment. An isolated nucleic acid includes a nucleic acid molecule contained in cells that ordinarily contain the nucleic acid molecule, but the nucleic acid molecule is present extrachromosomally or at a chromosomal location that is different from its natural chromosomal location. "Isolated nucleic acid encoding an multi-specific antigen binding domain or fragment thereof" refers to one or more nucleic acid molecules encoding first antigen binding domain, and optionally second antigen binding domain, antibody heavy and light chains (or fragments thereof), including such nucleic acid molecule(s) in a single vector or separate vectors, and such nucleic acid molecule(s) present at one or more locations in a host cell.

[0166] The term "vector," as used herein, refers to a nucleic acid molecule capable of propagating another nucleic acid to which it is linked. The term includes the vector as a self-replicating nucleic acid structure as well as the vector incorporated into the genome of a host cell into which it has been introduced. Some vectors are suitable for delivering the nucleic acid molecule or polynucleotide of the present application. Certain vectors are capable of directing the expression of nucleic acids to which they are operatively linked. Such vectors are referred to herein as expression vectors.

[0167] The terms "host cell," "host cell line," and "host cell culture" are used interchangeably and refer to cells into which exogenous nucleic acid has been introduced, including the progeny of such cells. Host cells include "transformants" and "transformed cells," which include the primary transformed cell and progeny derived therefrom without regard to the number of passages. Progeny may not be completely identical in nucleic acid content to a parent cell, but may contain mutations. Mutant progeny that have the same function or biological activity as screened or selected for in the originally transformed cell are included herein.

[0168] A polynucleotide "variant," as the term is used herein, is a polynucleotide that typically differs from a polynucleotide specifically disclosed herein in one or more substitutions, deletions, additions and insertions. Such variants may be naturally occurring or may be synthetically generated, for example, by modifying one or more of the polynucleotide sequences described herein and evaluating one or more biological activities of the encoded polypeptide as described herein, e.g., using any of a number of techniques well known in the art.

[0169] A polypeptide "variant," as the term is used herein, is a polypeptide that typically differs from a polypeptide specifically disclosed herein in one or more substitutions, deletions, additions and insertions. Such variants may be naturally occurring or may be synthetically generated, for example, by modifying one or more of the above polypeptide sequences described herein and evaluating one or more biological activities of the polypeptide as described herein, e.g., using any of a number of techniques well known in the art.

[0170] The term "variant" may also refer to any naturally occurring or engineered molecule comprising one or more nucleotide or amino acid mutations. In one embodiment, the multi-specific antigen binding molecule is a bispecific antigen binding molecule. In one embodiment, the multi-specific antigen binding molecule is a bispecific antibody. For example, somatic variants may encompass all related naturally occurring antibodies that are part of or derived from the same B-cell lineage. Engineered variants may encompass all single mutations or combinatorial mutations made to an antibody.8. Methods of Producing Multi-Specific Antigen Binding Molecules

[0171] Multi-specific and bispecific antigen binding molecules that bind to an HIV antigen (e.g., gp120, gp41) and human CD3 (e.g., human CD3ε or human CD3δ) can be produced by any method known in the art for the synthesis of multi-specific antibodies, for example, by chemical synthesis or by recombinant expression techniques.

[0172] Methods of making monospecific antibodies are known. Methods of making bispecific antibodies are known and described, for example, in PCT Publ. Nos. WO2011 / 038290; WO2012 / 158818, WO2012 / 162067, WO2015 / 104346, WO2016 / 086189, WO2016 / 182751, WO2017 / 009442, WO2017 / 125897, WO2017 / 136659, WO2017 / 157305, WO2017 / 201493, WO2018 / 183139, WO2018 / 191438, WO2019 / 034580, WO2019 / 078697 and WO2019 / 143636; U.S. Pat. Nos. 5,731,168; 5,807,706; 5,821,333; and U.S. Appl. Publ. Nos. 2003 / 020734, 2002 / 0155537, 2014 / 242079, 2015 / 133640, 2016 / 297885 and 2017 / 037130. Bispecific tetravalent antibodies, and methods of making them are described, e.g., in WO 02 / 096948 and WO 00 / 44788. In addition, other publications relating to making bispecific antibodies include WO 91 / 00360; WO 92 / 08802; WO 92 / 05793, and WO 93 / 17715; Tutt et al., J. Immunol. 147:60-69 (1991); U.S. Pat. Nos. 4,474,893; 4,714,681; 4,925,648; 5,573,920; 5,601,819, 9,212,230 and 9,701,759; and Kostelny et al., J. Immunol. 148:1547-1553 (1992). Heterodimeric bispecific antibodies having an scFv first antigen binding domain and a Fab second antigen binding domain are described, e.g., in WO 2013 / 163427 and in U.S. Patent No. 9,701,759.

[0173] One method of making bispecific antibodies and the multi-specific antigen binding molecules described herein employs so-called "knobs-into-holes" technology (Ridgway et al., Protein Eng., 9:617-621 (1996); WO 2006 / 028936). The mispairing problem of Ig heavy chains that is a chief drawback for making bispecific antibodies is reduced in this technology by mutating selected amino acids forming the interface of the CH3 domains in IgG. At positions within the CH3 domain at which the two heavy chains interact directly, an amino acid with a small side chain (hole) is introduced into the sequence of one heavy chain and an amino acid with a large side chain (knob) into the counterpart interacting residue location on the other heavy chain. In some instances, multi-specific antigen binding molecules described herein have immunoglobulin chains in which the CH3 domains have been modified by mutating selected amino acids that interact at the interface between two polypeptides so as to preferentially form a bispecific antigen binding molecule. In some embodiments, the multi-specific antigen binding molecules can be composed of immunoglobulin chains of the same subclass or different subclasses. In one instance, a multi-specific antigen binding molecule that binds to gp120 and CD3 comprises a T366W (EU numbering) mutation in the "knobs chain" and T366S, L368A, Y407V (EU numbering) mutations in the "hole chain." In certain embodiments, an additional interchain disulfide bridge is introduced between the CH3 domains by, e.g., introducing a Y349C mutation into the "knobs chain" and a E356C mutation or a S354C mutation into the "hole chain." In certain embodiments, R409D, K370E mutations are introduced in the "knobs chain" and D399K, E357K mutations in the "hole chain." In other embodiments, Y349C, T366W mutations are introduced in one of the chains and E356C, T366S, L368A, Y407V mutations in the counterpart chain. In some embodiments. Y349C, T366W mutations are introduced in one chain and S354C, T366S, L368A, Y407V mutations in the counterpart chain. In some embodiments, Y349C, T366W mutations are introduced in one chain and S354C, T366S, L368A, Y407V mutations in the counterpart chain. In yet other embodiments, Y349C, T366W mutations are introduced in one chain and S354C, T366S, L368A, Y407V mutations in the counterpart chain (all EU numbering).

[0174] Another exemplary method of making bispecific antibodies is by using the Bispecific T-cell Engagers (BiTEs ®< ) platform. BiTEs are made by genetically fusing a first scFv (e.g., a scFv that binds gp120) to a second scFv (e.g., a scFv that binds human CD3) via a flexible peptide linker (e.g., GGGGS (SEQ ID NO: 1088)). See, e.g., Staerz et al., Nature, 314:628-631 (1985); Mack et al., PNAS, 92:7021-7025 (1995); Huehls et al., Immunol. Cell Biol., 93:290-296 (2015).

[0175] Another exemplary method of making bispecific antibodies is by using the Dual-Affinity Re-targeting (DART) platform. This technology is based on the diabody format of Holliger et al. (PNAS, 90:6444-6448 (1993)) and further improved for stability and optimal pairing of the VH and VL chains (Johnson et al., J Mol. Biol., 399: 436-449 (2010); Sung et al., J Clin Invest., 125(11): 4077-4090 (2015)).

[0176] Yet another exemplary method of making bispecific antibodies is by using the Trifunctional Hybrid Antibodies platform - Triomab ®< . This platform employs a chimeric construction made up of half of two full-length antibodies of different isotypes, mouse IgG2a and rat IgG2b. This technology relies on species-preferential heavy / light chain pairing associations. See, Lindhofer et al., J Immunol., 155:219-225 (1995).

[0177] A further exemplary method of making bispecific antibodies is by using the TandAb ®< platform. This technology is based on the diabody concept but are designed as a single polypeptide chain VH1-VL2-VH2-VL1 comprising short linkers to prevent intrachain pairing. Head-to-tail dimerization of this single chain results in the formation of a tetravalent homodimer (Kipriyanov et al., J Mol. Biol., 293: 41-56 (1999)).

[0178] Yet another method for making bispecific antibodies is the CrossMab technology. CrossMab are chimeric antibodies constituted by the halves of two full-length antibodies. For correct chain pairing, it combines two technologies: (i) the knob-into-hole which favors a correct pairing between the two heavy chains; and (ii) an exchange between the heavy and light chains of one of the two Fabs to introduce an asymmetry which avoids light-chain mispairing. See, Ridgway et al., Protein Eng., 9:617-621 (1996); Schaefer et al., PNAS, 108:11187-11192 (2011). CrossMabs can combine two or more antigen binding domains for targeting two or more targets or for introducing bivalency towards one target such as the 2:1 format.

[0179] In various embodiments, the multi-specific antigen binding molecules described herein may be produced in bacterial or eukaryotic cells. The multi-specific antigen binding molecules can also be produced in eukaryotic cells such as transformed cell lines (e.g., CHO, CHO-S, 293E, 293T, Expi293 ™< , COS, NIH3T3). In addition, the multi-specific antigen binding molecules described herein (e.g., Fabs, Fab-scFv, scFv's) can be expressed in a yeast cell such as Pichia (see, e.g., Powers et al., J Immunol Methods. 251:123-35 (2001)), Hanseula, or Saccharomyces. In one embodiment, the bispecific antibodies described herein are produced in a CHO-based or CHO-origin cell line (e.g., CHO-S, CHO DG44, ExpiCHO ™< , CHOZN ®< ZFN-modified GS- / - CHO cell line or CHO-K1, CHO-K1a) or a HEK293 (e.g., Expi293 ™< ) cell line. To produce the multi-specific antigen binding molecules of interest, one or more polynucleotides encoding the multi-specific antigen binding molecules is constructed, introduced into an expression vector, and then expressed in one or more suitable host cells. In some embodiments, three polynucleotides encoding an scFv heavy chain comprising the first antigen binding domain, a Fab heavy chain and a Fab light chain comprising the second antigen binding domain are co-expressed in a single host cell. In some embodiments, three polynucleotides encoding a Fab heavy chain and a Fab light chain comprising the first antigen binding domain, and an EC domain comprising the second antigen binding domain are co-expressed in a single host cell. Standard molecular biology techniques are used to prepare the recombinant expression vector, transfect the host cells, select for transformants, culture the host cells, and recover the multi-specific antigen binding molecules.

[0180] In some embodiments, the host cell predominantly sialylates N-linked glycosylation sites within the variable regions of an immunoglobulin antigen binding domain. In some embodiments, the polynucleotides encoding a multi-specific antigen binding molecule, as described herein, are expressed in a host cell that sialylates at least 50%, at least 55%, at least 60%, at least 65%, at least 70%, at least 75%, at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, or more, N-linked glycosylation sites in the variable domains (Fv) of expressed antigen binding molecules. In various embodiments, in multi-specific antigen binding molecules expressed from such host cells, at least 50%, at least 60%, at least 70%, least 80%, at least 85%, at least 90%, or more, N-linked glycosylation sites in at least one of the first VH, the first VL, the second VH and the second VL of the multi-specific antigen binding molecule are sialylated. In various embodiments, in multi-specific antigen binding molecules expressed from such host cells, the N-linked glycosylation sites in at least one of the first VL, the second VH and the second VL have a sialic acid occupancy (e.g., a glycan comprising one or two terminal sialic acid residues) of at least 40%, at least 50%, at least 60%, at least 70%, least 80%, at least 85%, at least 90%, or more. In some embodiments, the sialylated N-linked glycosylation sites in at least one of the first VL, the second VH and the second VL of the multi-specific antigen binding molecule comprise from 1 to 5 sialic acid residues, e.g., from 1 to 4 sialic acid residues, e.g., from 1 to 3 sialic acid residues, e.g., from 1 to 2 sialic acid residues. In some embodiments, at least one of the first VL, the second VH and the second VL are sialylated with N-acetylneuraminic acid (NANA). In some embodiments, the sialic acid residues are present in biantennary structures. In some embodiments, the sialic acid residues are present in complex N-linked glycan structures. In some embodiments, the sialic acid residues are present in hybrid N-linked glycan structures. In some embodiments, the glycans are terminally sialylated.

[0181] If the multi-specific antigen binding molecules are to be expressed in bacterial cells (e.g., E. coli), the expression vector should have characteristics that permit amplification of the vector in the bacterial cells. Additionally, when E. coli such as JM109, DH5α, HB101, or XL1-Blue is used as a host, the vector must have a promoter, for example, a lacZ promoter (Ward et al., 341:544-546 (1989), araB promoter (Better et al., Science, 240:1041-1043 (1988)), or T7 promoter that can allow efficient expression in E. coli. Examples of such vectors include, for example, M13-series vectors, pUC-series vectors, pBR322, pBluescript, pCR-Script, pGEX-5X-1 (Pharmacia), "QIAexpress system" (QIAGEN), pEGFP, and pET (when this expression vector is used, the host is preferably BL21 expressing T7 RNA polymerase). The expression vector may contain a signal sequence for secretion of the multi-specific antigen binding molecules. For production into the periplasm of E. coli, the pelB signal sequence (Lei et al., J. Bacteriol., 169: 4379 (1987)) may be used as the signal sequence for secretion of the multi-specific antigen binding molecules. For bacterial expression, calcium chloride methods or electroporation methods may be used to introduce the expression vector into the bacterial cell.

[0182] If the multi-specific antigen binding molecules are to be expressed in animal cells, e.g., such as CHO-based or CHO-origin cells, COS, and NIH3T3 cells, the expression vector includes a promoter useful for expression in these cells. In various embodiments, the promoter for expression of the multi-specific antigen binding molecules in mammalian cells is a constitutive promoter or an inducible promoter. Illustrative promoters for expression of the multi-specific antigen binding molecules in mammalian cells include without limitation an SV40 promoter (Mulligan et al., Nature, 277:108 (1979)), MMLV-LTR promoter, EF1α promoter (Mizushima et al., Nucleic Acids Res., 18:5322 (1990)), or CMV promoter. In addition to the nucleic acid sequence encoding the immunoglobulin or domain thereof, the recombinant expression vectors may carry additional sequences, such as sequences that regulate replication of the vector in host cells (e.g., origins of replication) and selectable marker genes. The selectable marker gene facilitates selection of host cells into which the vector has been introduced (see e.g., U.S. Pat. Nos. 4,399,216, 4,634,665 and 5,179,017). For example, typically the selectable marker gene confers resistance to drugs, such as G418, hygromycin, or methotrexate, on a host cell into which the vector has been introduced. Examples of vectors with selectable markers include pMAM, pDR2, pBK-RSV, pBK-CMV, pOPRSV, and pOP13.

[0183] In one embodiment, the multi-specific antigen binding molecules are produced in mammalian cells. Exemplary mammalian host cells for expressing multi-specific antigen binding molecules include Chinese Hamster Ovary (CHO cells) (including dhfr- CHO cells, e.g., described in Urlaub and Chasin (1980) Proc. Natl. Acad. Sci. USA 77: 4216-4220, used with a DHFR selectable marker, e.g., as described in Kaufman and Sharp (1982) Mol. Biol. 159:601 621 and glutamine synthetase (GS)- cells used with a GS selectable marker, e.g., described in Lin, et al., MAbs. (2019) 11(5):965-976; and Noh, et al., Sci Rep. (2018) 8(1):5361), human embryonic kidney 293 cells (e.g., 293, 293E, 293T, Expi293 ™< ), COS cells, NIH3T3 cells, lymphocytic cell lines, e.g., NS0 myeloma cells and SP2 cells, and a cell from a transgenic animal, e.g., a transgenic mammal. For example, the cell is a mammary epithelial cell. CHO and NS0 cell lines for recombinant antibody production are reviewed by Dhara, et al., BioDrugs. (2018) 32(6):571-584.

[0184] In an exemplary system for expression of the multi-specific antigen binding molecules, recombinant expression vectors encoding the first and second binding domains (e.g., VH and VL of an anti-CD3 targeting arm and VH and VL of an anti-gp120 targeting arm) are introduced into dhfr- CHO cells by calcium phosphate-mediated transfection. In a specific embodiment, the dhfr- CHO cells are cells of the DG44 cell line, such as DG44i (see, e.g., Derouaz et al., Biochem Biophys Res Commun., (2006) 340(4):1069-77). Within the recombinant expression vectors, the immunoglobulin heavy and light chain genes are each operatively linked to enhancer / promoter regulatory elements (e.g., derived from SV40, CMV, adenovirus and the like, such as a CMV enhancer / AdMLP promoter regulatory element or an SV40 enhancer / AdMLP promoter regulatory element) to drive high levels of transcription of the genes. The recombinant expression vectors also carry a DHFR gene, which allows for selection of CHO cells that have been transfected with the vector using methotrexate selection / amplification. The selected transformant host cells are cultured to allow for expression of the immunoglobulin heavy and light chains and the multi-specific antigen binding molecule is recovered from the culture medium. In one embodiment, all three polypeptides of a bi-specific antigen binding molecule, as described herein, are expressed in a single cell. In one embodiment, all three polypeptides of a bi-specific antigen binding molecule, as described herein, are expressed in a single cell from a single vector.

[0185] The multi-specific antigen binding molecules can also be produced by a transgenic animal. For example, U.S. Pat. No. 5,849,992 describes a method of expressing an antibody in the mammary gland of a transgenic mammal. A transgene is constructed that includes a milk-specific promoter and one or more polynucleotides encoding the multi-specific antigen binding molecule of interest and a signal sequence for secretion. The milk produced by females of such transgenic mammals includes, secreted-therein, the multi-specific antigen binding molecule of interest. The multi-specific antigen binding molecule can be purified from the milk, or for some applications, used directly. Animals are also provided comprising one or more of the nucleic acids described herein.

[0186] The multi-specific antigen binding molecules can be isolated from inside or outside (such as medium) of the host cell and purified as substantially pure and homogenous, non-aggregated multi-specific antigen binding molecules (e.g., heterodimeric bispecific antigen binding molecules). As appropriate or desired, the cell or population of cells are cultured in a culture volume of at least 2L, e.g., at least 5L, 10L, 50L, 100L, 150L, 200L, 250L, or more. Methods for isolation and purification commonly used for antibody purification may be used for the isolation and purification of herein described multi-specific antigen binding molecules, and are not limited to any particular method. The multi-specific antigen binding molecules may be isolated and purified by appropriately selecting and combining, for example, column chromatography, filtration, ultrafiltration, salting out, solvent precipitation, solvent extraction, distillation, immunoprecipitation, SDS-polyacrylamide gel electrophoresis, isoelectric focusing, dialysis, and recrystallization. Chromatography includes, for example, affinity chromatography, ion exchange chromatography, hydrophobic chromatography, gel filtration, reverse-phase chromatography, and adsorption chromatography (Strategies for Protein Purification and Characterization: A Laboratory Course Manual. Ed Daniel R. Marshak et al., Cold Spring Harbor Laboratory Press, 1996). Chromatography can be carried out using liquid phase chromatography such as HPLC and FPLC. Columns used for affinity chromatography include Protein A column and protein G column. Examples of columns using Protein A column include Hyper D, POROS, and Sepharose FF (GE Healthcare Biosciences). The present disclosure also includes multi-specific antigen binding molecules that are highly purified using these purification methods. In various embodiments, the isolating or purifying step comprises Protein A chromatography, and at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or more, of the multi-specific antigen binding molecules are isolated or purified. In various embodiments, the isolating or purifying step comprises Protein A chromatography, followed by ion exchange chromatography, and at least 95%, 96%, 97%, 98%, 99%, or more, of the multi-specific antigen binding molecules are isolated or purified. In various embodiments, at least 95%, 96%, 97%, 98%, 99%, or more, of the multi-specific antigen binding molecules isolate or purify as non-aggregated soluble heterodimer as determined using size exclusion chromatography (SEC). In some embodiments, the isolated or purified multi-specific antigen binding molecules have increased homogeneity as assessed by analytical ion exchange chromatography, wherein the integrated area of a main peak representing an unmodified target species is at least 95%, 96%, 97%, 98%, or more, of the sum of all integrated protein peak areas. In some embodiments, the isolated or purified antigen binding molecules have fewer than 35%, 30%, 25%, 20%, 15%, 10%, 9%, 8%, 7%, or fewer, acidic contaminants.9. Methods of Treating and Preventing HIV

[0187] Provided are one or more multi-specific antigen binding molecules of the invention, or one or more polynucleotides of the invention for use in methods for treating or preventing an HIV infection or a related disease or disorder in a subject in need thereof (e.g., a human subject), comprising providing to a subject in need thereof an effective amount of one or more of the multi-specific antigen binding molecules (e.g., bispecific antigen binding molecules, bispecific antibodies), or one or more polynucleotides encoding the multi-specific antigen binding molecules.

[0188] In some embodiments, the methods entail administering a multi-specific antigen binding molecule having a first antigen binding domain that binds to CD3, and a second antigen binding domain that binds to an epitope or region of gp120 in the CD4 binding site (CD4bs) and comprises one or more extracellular (EC) domains of CD4, and optionally, an IL-15 receptor agonist. In some embodiments, the EC domain of CD4 comprises a sequence that is at least 99% identical to the sequence of SEQ ID NO: 746. In some embodiments, the EC domain of CD4 comprises the sequence of SEQ ID NO: 746. In some embodiments, the first antigen binding domain comprises a first VH-CDR1, a first VH-CDR2, a first VH-CDR3, a first VL-CDR1, a first VL-CDR2 and a first VL-CDR3 comprising the following amino acid sequences, respectively (according to Kabat): SEQ ID NOs: 1, 12, 8, 4, 9 and 10; and the second antigen binding domain comprises one EC domain of CD4 comprising an amino acid sequence that is at least 99% identical to a CD4 EC domain of SEQ ID NO:746. In some embodiments, the first antigen binding domain comprises a first VH-CDR1, a first VH-CDR2, a first VH-CDR3, a first VL-CDR1, a first VL-CDR2 and a first VL-CDR3 comprising the following amino acid sequences, respectively (according to Kabat): SEQ ID NOs: 1, 12, 8, 4, 9 and 10; and the second antigen binding domain comprises one EC domain of CD4 comprising the amino acid sequence of SEQ ID NO:746. In some embodiments, the first antigen binding domain comprises a first VH-CDR1, a first VH-CDR2, a first VH-CDR3, a first VL-CDR1, a first VL-CDR2 and a first VL-CDR3 comprising the following amino acid sequences, respectively (according to Chothia): SEQ ID NOs: 17, 18, 23, 20, 24 and 25; and the second antigen binding domain comprises one EC domain of CD4 comprising the amino acid sequence of SEQ ID NO:746. In some embodiments, the first antigen binding domain comprises a first VH-CDR1, a first VH-CDR2, a first VH-CDR3, a first VL-CDR1, a first VL-CDR2 and a first VL-CDR3 comprising the following amino acid sequences, respectively (according to IMGT): SEQ ID NOs: 28, 29, 32, 31, 24 and 10; and the second antigen binding domain comprises one EC domain of CD4 comprising the amino acid sequence of SEQ ID NO:746. In some embodiments, the first antigen binding domain comprises a first VH-CDR1, a first VH-CDR2, a first VH-CDR3, a first VL-CDR1, a first VL-CDR2 and a first VL-CDR3 comprising the following amino acid sequences, respectively (according to Honegger): SEQ ID NOs: 34, 43, 40, 37, 41 and 25; and the second antigen binding domain comprises one EC domain of CD4 comprising the amino acid sequence of SEQ ID NO:746. In some embodiments, the first antigen binding domain comprises a first VH and a first VL comprising the amino acid sequences set forth, respectively: SEQ ID NOs: 51 and 56; and the second antigen binding domain comprises one EC domain of CD4 comprising an amino acid sequence that is at least 99% identical to a CD4 EC domain of SEQ ID NO:746. In some embodiments, the first antigen binding domain comprises a first VH and a first VL comprising amino acid sequences that are at least 99% identical to the amino acid sequences set forth, respectively: SEQ ID NOs: 51 and 56; and the second antigen binding domain comprises one EC domain of CD4 comprising an amino acid sequence that is at least 99% identical to a CD4 EC domain of SEQ ID NO:746. In some embodiments, the...

Claims

1. A multi-specific antigen binding molecule that binds to human CD3 and HIV gp120, wherein the antigen binding molecule comprises: (a) a first antigen binding domain that comprises a first heavy chain variable domain (VH) and a first light chain variable domain (VL), wherein the first antigen binding domain binds to CD3, wherein the first antigen binding domain comprises a first VH-complementarity determining region (CDR) 1 (VH-CDR1), a first VH-CDR2, a first VH-CDR3, a first VL-CDR1, a first VL-CDR2 and a first VL-CDR3, comprising the following amino acid sequences, respectively (according to Kabat): SEQ ID NOs: 1, 12, 8, 4, 9 and 10, wherein the first VH and the first VL comprise amino acid sequences that are at least 99% identical to the amino acid sequences set forth, respectively: SEQ ID NOs: 51 and 56, and wherein the first antigen binding domain binds to CD3 with a KD of lower than 3.0 nM determined by surface plasmon resonance; and (b) a second antigen binding domain that binds to HIV gp120 comprising one or more extracellular (EC) domain of CD4, wherein the one or more EC domains of CD4 comprise a sequence that is at least 99% identical to:

2. The multi-specific antigen binding molecule of claim 1, wherein the second antigen binding domain comprises one or more EC domains of CD4 comprising an amino acid sequence that comprises SEQ ID NO: 746.

3. The multi-specific antigen binding molecule of claim 1 or claim 2, wherein the first VH and the first VL comprising the amino acid sequences set forth, respectively: SEQ ID NOs: 51 and 56; and the second antigen binding domain comprises one EC domain of CD4 comprising an amino acid sequence of SEQ ID NO: 746.

4. The multi-specific antigen binding molecule of any one of claims 1 to 3 comprising: (i) a heterodimeric human IgG1 or IgG4, comprising a first Fc region and a second Fc region, wherein one or both of the first and second Fc regions comprise one or more of the following amino acids at the indicated positions according to EU numbering (A) Alanine at position 234 (B) Alanine at position 235; and (C) Serine at position 331; (ii) a heterodimeric human IgG1 or IgG4, comprising a first Fc region and a second Fc region, wherein one or both of the first and second Fc regions comprise the following amino acids at the indicated positions according to EU numbering: (A) Tyrosine at position 252, threonine at position 254 and glutamic acid at position 256 (YTE); or (B) Leucine at position 428 and serine at position 434 (LS); (iii) a heterodimeric human IgG1, comprising a first Fc region and a second Fc region, wherein both of the first and second Fc regions comprise the following amino acids at the indicated positions according to EU numbering : Tyrosine at position 252, threonine at position 254 and glutamic acid at position 256 (YTE); (iv) a heterodimeric human IgG1 or IgG4, comprising a first Fc region and a second Fc region, comprising the following amino acids at the indicated positions according to EU numbering: (A) the first Fc region comprises a tryptophan at position 366 (T366W); and the second Fc region comprises a serine at position 366 (T366S), an alanine at position 368 (L368A) and a valine at position 407 (Y407V); (B) the first Fc region comprises a serine at position 366 (T366S), an alanine at position 368 (L368A) and a valine at position 407 (Y407V); and the second Fc region comprises a tryptophan at position 366 (T366W); (C) the first Fc region comprises a cysteine at position 354 (S354C), a tryptophan at position 366 (T366W); and the second Fc region comprises a cysteine at position 349 (Y349C), a serine at position 366 (T366S), an alanine at position 368 (L368A) and a valine at position 407 (Y407V); (D) the first Fc region comprises cysteine at position 349 (Y349C), a serine at position 366 (T366S), an alanine at position 368 (L368A) and a valine at position 407 (Y407V); and the second Fc region comprises a cysteine at position 354 (S354C), a tryptophan at position 366 (T366W); (v) a heterodimeric human IgG1, comprising a first Fc region and a second Fc region, comprising the following amino acids at the indicated positions according to EU numbering: the first Fc region comprises a serine at position 366 (T366S), an alanine at position 368 (L368A) and a valine at position 407 (Y407V); and the second Fc region comprises a tryptophan at position 366 (T366W); (vi) a heterodimeric human IgG1 or IgG4, comprising a first Fc region and a second Fc region, wherein one of the first Fc region or the second Fc region comprise the following amino acids at the indicated positions according to EU numbering (A) arginine at position 435 (H435R); or (B) arginine at position 435 (H435R) and phenylalanine at position 436 (Y436F); and / or (vii) a heterodimeric human IgG1 or IgG4, comprising a first Fc region and a second Fc region, comprising the following amino acids at the indicated positions according to EU numbering: (A) the first Fc region comprises an alanine at position 234 (L234), an alanine at position 235 (L235A), a serine at position 331 (P331S) and a tryptophan at position 366 (T366W); and the second Fc region comprises an alanine a position 234 (L234), an alanine at position 235 (L235A), a serine at position 331 (P331S), a serine at position 366 (T366S), an alanine at position 368 (L368A), a valine at position 407 (Y407V), and an arginine at position 435 (H435R); (B) the first Fc region comprises an alanine at position 234 (L234), an alanine at position 235 (L235A), a serine at position 331 (P331S) and a tryptophan at position 366 (T366W); and the second Fc region comprises an alanine a position 234 (L234), an alanine at position 235 (L235A), a serine at position 331 (P331S), a serine at position 366 (T366S), an alanine at position 368 (L368A), a valine at position 407 (Y407V), an arginine at position 435 (H435R) and a phenylalanine at position 436 (Y436F); (C) the first Fc region comprises an alanine at position 234 (L234), an alanine at position 235 (L235A), a serine at position 331 (P331S), a serine at position 366 (T366S), an alanine at position 368 (L368A), and a valine at position 407 (Y407V); and the second Fc region comprises an alanine at position 234 (L234), an alanine at position 235 (L235A), a serine at position 331 (P331S) and a tryptophan at position 366 (T366W); (D) the first Fc region comprises an alanine at position 234 (L234), an alanine at position 235 (L235A), a serine at position 331 (P331S), a tryptophan at position 366 (T366W), a leucine at position 428 (M428L) and a serine at position 434 (N434S); and the second Fc region comprises an alanine a position 234 (L234), an alanine at position 235 (L235A), a serine at position 331 (P331S), a serine at position 366 (T366S), an alanine at position 368 (L368A), a valine at position 407 (Y407V), and an arginine at position 435 (H435R); or (E) the first Fc region comprises an alanine at position 234 (L234), an alanine at position 235 (L235A), a serine at position 331 (P331S), a tryptophan at position 366 (T366W), a tyrosine at position 252 (M252Y), a threonine at position 254 (S254T) and a glutamic acid at position 256 (T256E); and the second Fc region comprises an alanine a position 234 (L234), an alanine at position 235 (L235A), a serine at position 331 (P331S), a serine at position 366 (T366S), an alanine at position 368 (L368A), a valine at position 407 (Y407V), and an arginine at position 435 (H435R).

5. The multi-specific antigen binding molecule of any one of claims 1 to 4, wherein the multi-specific antigen binding molecule comprises the amino acid sequences set forth, respectively, below, or comprising amino acid sequences that are at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% identical to the amino acid sequences set forth, respectively, below: 1) SEQ ID NOs: 751, 752 and 753; or 2) SEQ ID NOs: 754, 752 and 753.

6. The multi-specific antigen binding molecule of any one of claims 1 to 5, wherein the multi-specific antigen binding molecule is a bispecific antigen binding molecule.

7. A polynucleotide or multiple polynucleotides encoding the multi-specific antigen binding molecule of claim 5.

8. The polynucleotide or polynucleotides of claim 7, comprising the following polynucleotide sequences, or polynucleotide sequences that are at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% identical to the polynucleotide sequences set forth, respectively, below: i. SEQ ID NOs: 995, 996 and 997; or ii. SEQ ID NOs: 998, 999 and 1000.

9. The polynucleotide or polynucleotides of any one of claims 7 to 8: (a) comprising: (i) the following polynucleotide sequences, or polynucleotide sequences that are at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% identical to the polynucleotide sequences set forth, respectively: SEQ ID NOs: 995, 996 and 997, optionally wherein the polynucleotide or polynucleotides comprises the following polynucleotide sequences, respectively: SEQ ID NOs: 995, 996 and 997; or (ii) the following polynucleotide sequences, or polynucleotide sequences that are at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% identical to the polynucleotide sequences set forth, respectively: SEQ ID NOs: 998, 999 and 1000, optionally wherein the polynucleotide or polynucleotides comprises the following polynucleotide sequences, respectively: SEQ ID NOs: 998, 999 and 1000; and / or (b) wherein: (i) the polynucleotide or polynucleotides are comprised of DNA or RNA; and / or (ii) the polynucleotide or polynucleotides are comprised of mRNA.

10. An expression cassette or multiple expression cassettes comprising one or more regulatory sequences operably linked to the polynucleotide or polynucleotides of any one of claims 7 to 9.

11. An expression vector or multiple expression vectors comprising one or more regulatory sequences operably linked to the polynucleotide or polynucleotides of any one of claims 7 to 9, or the expression cassette or expression cassettes of claim 10.

12. The expression vector or expression vectors of claim 11, wherein: (a) the expression vector or expression vectors comprise a plasmid vector or a viral vector; and / or (b) the first, second and third expression cassettes each comprise a promoter of equivalent transcription strength, e.g., a constitutive promoter, e.g., a promoter selected from cytomegalovirus (CMV), SV40, RSV, EF1a, UBC, PGK and CAGG.

13. A cell or population of cells, comprising the polynucleotide or polynucleotides of any one of claims 7 to 9, the expression cassette or multiple expression cassettes of claim 10, or the expression vector or expression vectors of any one of claims 11 to 12.

14. The cell or population of cells of claim 13, wherein the cell or population of cells comprises a eukaryotic cell.

15. A pharmaceutical composition comprising one or more multi-specific antigen binding molecules claimed in any one of claims 1 to 6 and a pharmaceutically acceptable carrier.

16. The pharmaceutical composition of claim 15, wherein: (a) the composition comprises an aqueous formulation; (b) the pharmaceutical composition is at a concentration of from 0.1 mg / ml to 150 mg / ml, e.g., 0.1 mg / ml to 100 mg / ml, e.g., 1 mg / ml to 100 mg / ml, e.g., from 5 mg / ml to 60 mg / ml, e.g., from 20 mg / ml to 150 mg / ml, or from 10 mg / ml to 50 mg / ml; and / or (c) the pharmaceutical composition is formulated for intravenous, intramuscular or subcutaneous administration.

17. A kit comprising one or more containers comprising one or more of the multi-specific antigen binding molecules claimed in any one of claims 1 to 6, the polynucleotide or polynucleotides of any one of claims 7 to 9, the expression vector or expression vectors of any one of claims 11 to 12, or the pharmaceutical composition of any one of claims 15 to 16, wherein the kit comprises: (a) one or more unitary doses of the one or more multi-specific antigen binding molecules, or the polynucleotide or polynucleotides, in one or more containers; (b) one or more unitary doses of the one or more multi-specific antigen binding molecules and a second agent for treating an HIV infection in separate containers, optionally wherein the kit further comprises at least one of a toll-like receptor (TLR) agonist and an IL-15 receptor agonist, for example wherein the kit comprises a TLR7 agonist selected from the group consisting of vesatolimod, imiquimod, and resiquimod; and / or (c) wherein the kit further comprises a multi-specific antigen binding molecule comprising an EC domain of CD4 of any one of claims 1 to 6.

18. A method of producing a multi-specific antigen binding molecule, the method comprising: a) culturing a cell or population of cells of any one of claims 13 to 14 transformed with the polynucleotide or polynucleotides of any one of claims 7 to 9, or the expression cassette or multiple expression cassettes of claim 10 in a cell culture under conditions sufficient to express the multi-specific antigen binding molecules; and b) isolating or purifying the antigen binding molecules from the cell culture.

19. The one or more multi-specific antigen binding molecules of any one of claims 1 to 6, one or more polynucleotides of any one of claims 7 to 9 or the pharmaceutical composition of any one of claims 15 to 16, for use in a method of treating or preventing HIV in a human subject in need thereof.

20. The one or more multi-specific antigen binding molecules, one or more polynucleotides, or the pharmaceutical composition, for use of claim 19, wherein: (a) the method further comprises administering to the subject a second agent for treating an HIV infection; (b) the method further comprises administering to the subject at least one of a TLR agonist and an IL-15 receptor agonist; and / or (c) the method comprises co-administering a multi-specific antigen binding molecule comprising an EC domain of CD4 of any one of claims 1 to 6.