Lysosome-targeting degradation fusion design

Abstract

Provided herein is disclosure of a recombinant bifunctional protein or polypeptide capable of binding to a cell surface receptor for lysosome targeting that is made up of an N-glycosylated peptide comprising at least one N-glycan group and a protein of interest, or antibody or antibody fragment capable of binding to a protein of interest. Also provided herein are methods for producing said recombinant bifunctional protein. Also provided herein are methods for lysosomal degradation of a protein of interest comprising introducing to a cell the peptide sequence of the recombinant bifunctional protein.

Description

LYSOSOME-TARGETING DEGRADATION FUSION DESIGNCROSS REFERENCE TO RELATED APPLICATIONS

[0001] This application claims the benefit of priority to U.S. Patent Application No. 63 / 647,645, filed May 15, 2024, and claims the benefit of priority to U.S. Patent Application No. 63 / 685,944, filed August 22, 2024, the contents of which are incorporated herein by reference.BACKGROUND

[0002] Extracellular and membrane-associated proteins — the products of 40% of all proteinencoding genes — are key proteins involved in cell communication, regulation, growth, and function. The mutation, accumulation or overexpression of these proteins are associated with many diseases such as cancer, ageing-related diseases, and autoimmune disorders. A general strategy to selectively decrease the level of these proteins has the potential to improve human health.

[0003] Lysosomes are membrane-bound organelles found in animal cells; they are vesicles that contain hydrolytic enzymes that digest many kinds of biomolecules. Lysosomes are degradative organelles that act as the waste disposal system of the cell by digesting used or wrong materials generated from both inside and outside the cell. Material from outside the cell is taken up through endocytosis, while material from the inside of the cell is digested through autophagy. Cell-surface lysosome-targeting receptors (LTRs) have been reported that facilitate the transport of proteins to lysosomes, including the mannose 6-phosphate (M6P) receptor. Coutinho et al., Mol.Genet. Metab., 2012, Vol.l07(3):257-266.

[0004] Enzymes of the lysosomes are synthesized in the rough endoplasmic reticulum and exported to the Golgi apparatus; the enzymes are transported from the Golgi apparatus to lysosomes in small vesicles, which fuse with larger acidic vesicles. Enzymes destined for a lysosome are tagged with the molecule mannose 6-phosphate (M6P). The presence of the M6P tags allows for binding to mannose 6-phosphate receptors in the Golgi apparatus, a phenomenon that is crucial for proper packaging into vesicles destined for the lysosomal system. Martinez-Pomares, J. Leukocyte Biol., 2012, vol. 92: 1177-1186.

[0005] The cation-independent mannose 6-phosphate receptor / insulin-like growth factor 2 receptor (CI-MPR / IGF2R) is the most important receptor for the delivery of newly synthesized lysosomal enzymes and recycling of extracellular lysosomal enzymes to thelysosome. The CI-MPR binds to the mannose 6-phosphate (M6P) moiety, a phosphorylated high mannose N-glycan presented on the surface of lysosomal enzymes, at high binding affinity (~2 nM) at neutral pH. When the receptor and lysosome enzyme complex is transported to the lysosome, the low pH (pH 4.5 - 5.5) causes a conformation change of the receptor reducing its binding affinity for M6P and results in the release of lysosome enzymes in the endosome and lysosome. Then the receptor is recycled back to the Golgi and plasma membrane. The discovery of the M6P pathway in 1970-80s greatly facilitated the development of recombinant lysosomal enzymes to treat lysosomal storage disorders (LSDs). Addition of the M6P moiety to lysosomal enzymes requires the activity of GlcNAc-1- phosphotransferase (PTase).

[0006] Lysosome-targeting molecules have been described, e.g., Banik et al., Nature, 2020, vol. 584(7820):291 and WO / 2020 / 132100 Al. However, these molecules rely on chemically-synthesized large sugar groups (e.g., mannose 6-phosphate group, mannose group) and a follow-up conjugation to purified antibodies. The process is complex, with the added disadvantage that additional chemical bonds are introduced by the conjugation steps.

[0007] Accordingly, there is a need for naturally produced lysosome-targeting molecules that do not require complex chemical synthesis and conjugation steps.SUMMARY

[0008] The present invention is directed to compositions and methods for targeting a protein of interest for lysosomal degradation. The compositions for lysosome targeting generally comprise a recombinant molecule comprising a N-glycosylated peptide (NGP) sequence that is fused with a peptide or protein targeted for lysosomal degradation. Alternatively, the N- glycosylated peptide may be fused with an antibody or antibody fragment capable of binding to a molecule targeted for lysosomal degradation.

[0009] The designed N-glycosylated peptide contains variant natural N-glycosylation sites with Asparagine-x-Serine or Threonine modifications which allow cells to post- translationally add N-glycans to the peptide when a nucleic acid construct encoding the peptide is expressed in the cell.

[0010] In one aspect, the N-glycosylated peptide comprises a peptide sequence comprising one or more modifications at N-glycosylation sites, wherein the one or more modifications comprise Asparagine-X-Serine or Threonine modifications, and wherein X comprises anyamino acid except proline. In some embodiments, X comprises one or more of Serine (S), Alanine (A), Histidine (H), Glycine (G), Asparagine (N), or Threonine (T). In some embodiments, the composition comprises a peptide sequence selected from NXTGGSNXTGGGNXTGGSNXTMMSGNS (SEQ ID NO: 1), NXTGNXTSNXTGNXTSNXTGNXTMMST (SEQ ID NO: 2), NXTNXTNXTNXTNXTNXTNXTNXTMMS (SEQ ID NO: 3), NXTSNXTTNXTTNXTSNXTTNNXSMMS (SEQ ID NO: 4), and NNXSNNXTNNXTNNXSNNXTNNXSMMS (SEQ ID NO: 5), wherein X comprises any amino acid except proline. In one embodiment, X comprises one or more of Serine (S), Alanine (A), Histidine (H), Glycine (G), Asparagine (N), or Threonine (T). In some embodiments, peptide sequence is selected from NSTGGSNATGGGNSTGGSNATMMSGNS (SEQ ID NO: 13), NATGNSTSNSTGNSTSNSTGNATMMST (SEQ ID NO: 10), NSTNATNHTNGTNNTNTTNGTNSTMMS (SEQ ID NO: 14), and NNTSNNTTNNTTNNTSNNTTNNTSMMS (SEQ ID NO: 15). In one embodiment, the peptide sequence is encoded by a nucleic acid sequence selected from: AACAGCACCGGCGGCAGCAACGCCACCGGCGGCGGCAACAGCACCGGCGGCAG CAACGCCACCATGATGAGCGGCAACAGC (SEQ ID NO: 16), AACGCCACCGGCAACAGCACCAGCAACAGCACCGGCAACAGCACCAGCAACAG CACCGGCAACGCCACCATGATGAGCACC (SEQ ID NO: 17), AACAGCACCAACGCCACCAACCACACCAACGGCACCAACAACACCAACACCACC AACGGCACCAACAGCACCATGATGAGC (SEQ ID NO: 18), and AACAACACCAGCAACAACACCACCAACAACACCACCAACAACACCAGCAACAA CACCACCAACAACACCAGCATGATGAGC (SEQ ID NO: 19).

[0011] The present invention is also directed to a composition comprising a peptide sequence comprising an N-glycosylated peptide having one or more modifications at the N- glycosylation sites fused to a protein of interest that is targeted for lysosomal degradation, or a chimeric antibody or antibody fragment that recognizes a protein of interest targeted for lysosomal degradation. The protein of interest may be a lysosomal or non-lysosomal protein or polypeptide that is a fragment of a non-lysosomal protein.

[0012] In one embodiment, the protein of interest is a lysosomal protein or a polypeptide fragment of a lysosomal protein selected from P-glucocerebrosidase (GBA), Cathepsin D (CathD), Niemann-Pick disease type C2 (NPC2), P-hexosaminidase (HEXB), a-Galactosidase (GLA), P-Mannosidase (MANBA), alpha-L-iduronidase, iduronate sulfatase, arylsulfatase B, acid a-glucosidase (GAA), N-acetyl-a-glucosaminidase (NAGLU), and lysosomal acid a-mannosidase (LAMAN).

[0013] In some embodiments, the protein of interest is a non-lysosomal protein or a polypeptide fragment of a non-lysosomal protein, including, without limitation, a cytokine, a membrane receptor, or immune checkpoint molecule.

[0014] Alternatively, the peptide sequence may comprise an N-glycosylated peptide fused with an antibody or antibody fragment that is specific for and binds to a protein of interest. In some embodiments, the protein of interest is selected from Tumor necrosis factor-a (TNFa), interferons, interleukins, IL- 12, growth IGF, EGF, EGFR, insulin, PD-L1, PD-1, and adrenaline.

[0015] The present invention is also directed to methods for producing the synthetic N- glycosylated peptides and fusion constructs in which the peptide is joined or fused with a protein of interest, or antibody or antibody binding fragment of interest.

[0016] In one aspect, the fusion construct is produced by in vitro expression of a nucleic acid sequence encoding the N-glycosylated peptide joined or fused to a nucleic acid sequence encoding a protein of interest or a polypeptide that is a fragment of a protein of interest, or a chimeric antibody or antibody fragment that recognizes a protein of interest.

[0017] In one aspect, the method comprises expressing the nucleic acid sequence encoding the fusion construct in the presence of a mannosidase I enzyme inhibitor, or a gene knockout cell line for generating terminal mannose N-glycans. In one embodiment, the mannosidase inhibitor comprises kifunensine. In one embodiment, the cell line is a gene knockout cell line for generating terminal mannose N-glycans, for example, the cell line with deficient of P-1, 2- N-acetylglucosaminyltransferase 1 (MGAT1). In one embodiment, a modified cell line for generation of other types of N-glycans. This expression system produces a recombinant fusion protein comprising a peptide containing terminal mannose N-glycans fused to the protein of interest or a fragment of protein of interest, or antibody or antibody fragment that binds to a protein of interest.

[0018] In another aspect, the method for in vitro expression of the nucleic acid sequence encoding an N-glycosylated peptide fused to a protein of interest or a fragment of a protein of interest, or a chimeric antibody or antibody fragment that recognizes a protein of interest,comprises expressing the nucleic acid sequence in the presence of a GlcNAc-1- phosphotransferase, or a truncated form of GlcNAc-1 -phosphotransferase that retains the activity of GlcNAc-1 -phosphotransferase. In one embodiment, the truncated form of GlcNAc-1 -phosphotransferase comprises a polypeptide comprising the core catalytic domain of GlcNAc-1 -phosphotransferase. In one embodiment, the truncated form of GlcNAc-1- phosphotransferase comprises SI S3 phosphotransferase. This expression system produces a fusion polypeptide or protein comprising a phosphorylated high-mannose type N-glycan peptide fused to the protein or antibody or antibody fragment of interest.

[0019] The present invention is also directed to a method for lysosomal degradation of a protein of interest comprising introducing to a cell. A fusion protein comprising the present N-glycan peptide fused to a protein of interest or a fragment of protein of interest, or to a chimeric antibody or antibody fragment that recognizes a protein of interest. The N-glycan may comprise a high-mannose type N-glycan or a phosphorylated high-mannose type N- glycan.

[0020] The present disclosure is also directed to a composition comprising an expression vector comprising a nucleic acid sequence encoding the N-glycosylated peptide and a protein of interest or a fragment of protein of interest, or a chimeric antibody or antibody fragment that recognizes a protein of interest. In one embodiment, the expression vector further encodes GlcNAc-1 -phosphotransferase or a truncated form thereof. In one embodiment, the expression vector comprises a viral vector. Alternatively, a second expression vector encoding GlcNAc-1 -phosphotransferase, or truncated form thereof, may be co-expressed with the expression vector encoding the protein fusion construct.

[0021] The present invention also is directed to methods of treating a disease or condition that is characterized by expression, or overexpression, of a protein or polypeptide of interest that is implicated in the development and / or progression of the disease or condition. The protein of interest may be a mutant form of a protein of interest. In some embodiments, the disease is a proliferative disease, a lysosomal storage disease, infection, an immune disease, or an autoimmune disease.BRIEF DESCRIPTION OF THE DRAWINGS

[0022] For the purpose of illustrating the disclosure, there are depicted in the drawings certain embodiments of the disclosure. However, the disclosure is not limited to the precise arrangements and instrumentalities of the embodiments depicted in the drawings.

[0023] Figure l is a schematic representation of a high-mannose N-glycan (top left), monophosphorylated high-mannose N-glycan (top right), bis-phosphorylated high-mannose N- glycan (bottom left) and complex-type N-glycan (bottom right), and their sensitivity for endoglycosidase H (“Endo H”) and peptide glycosidase F (“PNGase F”) digestion.

[0024] Figure 2 depicts N-Glycosylated Peptide (“NGP”) designs with variant peptide sequences allowing for 4, 6, 8, or 12 N-glycosylation sites (“4NGP,” “6NGP,” “8NGP” and “12NGP”).

[0025] Figure 3 depicts a western blot of conditioned medium from CHO cells expressing 4NGP, 6NGP, 8NGP, and 12NGP fused to GFP following digestion with Endo H or PNGase F.

[0026] Figure 4 A depicts a schematic showing production of NCP-GFP and PNCP-GFP in CHO cells.

[0027] Figure 4B depicts a western blot of conditioned medium containing NCP-GFP or PNCP-GFP that was treated with Endo-H or PNGase-F.

[0028] Figure 4C depicts CI-MPR binding of the NCP-GFP and PNCP-GFP protein from Figure 4B. Binding was measured by GFP fluorescence signal. Data are mean ± SD. N=3

[0029] Figure 4D depicts the GFP fluorescence in the produced conditioned medium used for CI-MPR binding assay depicted in Figure 4C.

[0030] Figure 5A depicts a western blot of conditioned medium from CHO cells expressing NCP-GFP, treated with kifunensine, following digestion with Endo-H or PNGase F.

[0031] Figure 5B depicts total GFP fluorescent signal of the NCP-GFP produced in the presence or absence of kifunensine.

[0032] Figure 6 A depicts a schematic showing the production of NCA-FLAG and PNCA- FLAG in CHO cells.

[0033] Figure 6B depicts a western blot of the NCA-FLAG and PNCA-FLAG heavy chain N-glycosylation after Endo-H and PNGase-F treatment.

[0034] Figure 6C depicts purified NCA-FLAG produced in the presence or absence of SI S3 phosphotransferase.

[0035] Figure 6D depicts CLMPR binding of the NCA-FLAG and PNCA-FLAG. To visualize the binding signal, Alexa Fluor 594 was conjugated to NCA-FLAG or PNCA- FLAG for the assay. Data are mean ± SD. N=3.

[0036] Figure 7A depicts fluorescence of several concentrations of PNCA-FLAG-AF594 and Ab-FLAG-AF594. Alexa fluor 594 (AF594) conjugated antibodies were analyzed for total fluorescence.

[0037] Figure 7B depicts CLMPR affinity chromatograph of NCA-FLAG-AF594.

[0038] Figure 7C depicts CLMPR affinity chromatograph of PNCA-FLAG-AF594.

[0039] Figure 8A depicts internalization of NCA-FLAG or PNCA-FLAG in human fibroblast cells treated with at 20 or 100 nM each antibody for 24 hours. Purified antibodies were loaded as loading control.

[0040] Figure 8B depicts human fibroblast cell uptake of Alexa Fluor 594 conjugated NCA- GFP or PNCA-GFP following incubation with and without 2mM free Mannose 6-phosphate in the media for 6 hours. The uptake was examined by Alexa Fluor 594 fluorescence under fluorescent microscopy. Nuclei were stained with Hoechst for 10 min (blue).

[0041] Figure 8C depicts internalized PNCA-FLAG (red) co-localized with lysotracker (green) in fibroblast cells.

[0042] Figure 9 depicts a western blot of conditioned medium from CHO cells expressing NCA-FLAG, treated with kifunensine, following digestion with Endo-H or PNGase-F.

[0043] Figure 10 depicts western blots with three dilutions of conditioned media containing soluble GFP-FLAG protein. Each blot was blotted with anti-FLAG antibodies (commercial Sigma F3165 mouse anti-FLAG, NCA-FLAG (M2) or PNCA-FLAG (M2) antibodies. Signal was detected with HRP-conjugated anti -mouse IgG.

[0044] Figure 11 A depicts a schematic showing the procedure for detection of FLAG- mCherry and PNCA-FLAG complexes binding toward CLMPR.

[0045] Figure 1 IB depicts bound mCherry fluorescence from Figure 11 A for Ab-FLAG or PNCA-FLAG. N=3 or 4.

[0046] Figure 11C depicts fluorescent microscopy of HepG2 cells incubated with FLAG- mCherry and Ab-FLAG or PNCA-FLAG overnight with or without 2 mM M6P. Lysosomes were stained with LysoTracker (green) and Hoechst for Nuclei (blue).

[0047] Figure 1 ID depicts HexM internalization via HexM activity assay following incubation of HEK293T cells with FLAG-HexM enzyme with or without Ab-FLAG, PNCA- FLAG or 2 mM M6P. Dot represents each independent experiment. N=3-6

[0048] Figure 1 IE depicts HexM internalization via HexM activity assay following incubation of patient fibroblast cells with FLAG-HexM enzyme with or without Ab-FLAG, PNCA-FLAG or 2 mM M6P. Dot represents each independent experiment. N=3.

[0049] Figure 1 IF depicts HexM internalization via HexM activity assay following incubation of Daoy (HTB-186) cells with FLAG-HexM enzyme with or without Ab-FLAG, PNCA-FLAG or 2 mM M6P. Dot represents each independent experiment. N=3.

[0050] Figure 11G depicts HexM internalization via HexM activity assay following incubation of HepG2 cells with FLAG-HexM enzyme with or without Ab-FLAG, PNCA- FLAG or 2 mM M6P. Dot represents each independent experiment. N=3.

[0051] Figure 11H depicts HexM internalization via HexM activity assay following incubation of MDA-MB-231 cells with FLAG-HexM enzyme with or without Ab-FLAG, PNCA-FLAG or 2 mM M6P. Dot represents each independent experiment. N=6.

[0052] Figure 1 II depicts western blot of CI-MPR deficiency in HEK293T CI-MPR- / - cells.

[0053] Figure 11 J depicts internalized HexM activity in CI-MPR- / - cells. Dot represents each independent experiment. N=3.

[0054] Figure 12A depicts a schematic demonstrating production of Ab-TNFa and PNCA- TNFa in CHO cells.

[0055] Figure 12B depicts SDS-PAGE and Coomassie staining of purified Ab-TNFa or PNCA-TNFa to visualize heavy and light chains.

[0056] Figure 12C depicts a western blot of the Ab-TNFa and PNCA-TNFa heavy chain N- glycosylation after Endo-H and PNGase-F treatment.

[0057] Figure 12D depicts CI-MPR affinity chromatograph to determine the binding of Ab- TNFa. To visualize the binding signal, Alexa Fluor 594 was conjugated to Ab-TNFa. Thedash line indicates the concentration of free M6P for antibody elution from the CI-MPR column.

[0058] Figure 12E depicts CI-MPR affinity chromatograph to determine the binding of PNCA-TNFa. To visualize the binding signal, Alexa Fluor 594 was conjugated to PNCA- TNFa. The dash line indicates the concentration of free M6P for antibody elution from the CI-MPR column.

[0059] Figure 12F depicts CI-MPR plate binding profile to examine the binding Kd(Kd=7.16 nM) for purified PNCA-TNFa-AF594 and Ab-TNFa-AF594. AF594 conjugated antibodies were selected for the analysis.

[0060] Figure 12G depicts uptake of AF594 conjugated antibodies incubated with human fibroblasts with or without 2 mM M6P (right panels). Uptake of antibody is indicated by red signal. Hoechst staining (blue) for nuclei.

[0061] Figure 12H depicts a schematic showing internalization and degradation of extracellular TNFa by PNCA-TNFa through CI-MPR to the lysosome.

[0062] Figure 121 depicts mCherry-TNFa fusion proteins; 1’ full amino acid sequence expressed as higher molecular weight band on Coomassie gel (right); 2’ truncated that runs as a lower molecular weight. Photo of mCherry protein expression in media and cell pellet (bottom left).

[0063] Figure 12J depicts fluorescent microscopy of HepG2 cells incubated with Cherry- TNFa and Ab-TNFa or PNCA-TNFa overnight with or without 2 mM M6P and mCherry- TNFa signal (red) detected. Lysosomes were stained with LysoTracker (green) and Hoechst for Nuclei (blue).

[0064] Figure 12K depicts a western blot of internalized mCherry-TNFa protein in cell lysates using antibodies to detect mCherry, TNFa and GAPDH, as a loading control.

[0065] Figure 12L depicts a western blot of internalized mCherry-TNFa with the addition of 1% or 0.5% of lysosomal protease inhibitors (PI). Antibodies used to detect mCherry, TNFa and GAPDH, as a loading control.

[0066] Figure 13 depicts a western blot of cell lysates of HepG2 cells incubated with recombinant TNFa (rhTNFa) and Ab-TNFa or PNCA-TNFa with or without protease inhibitors. Blots were probed for TNFa, human IgG or GAPDH (loading control).

[0067] Figure 14A depicts SDS-PAGE and Coomassie stain of Conditioned media and purified antibody produced by co-transfection of antibody TNFa antibody sequence with SI S3 PTase (Ab-TNFa-S). M: conditioned media; FT: flowthrough; P: purified protein.

[0068] Figure 14B depicts a western blot of the purified Ab-TNFa-S from Figure 14A treated with PNGase-F or Endo-H enzymes.

[0069] Figure 14C depicts the CI-MPR column binding profile for Ab-TNFa-S.

[0070] Figure 15A depicts a schematic for constructs expressed to produce Ab-PD-Ll by heavy and light chain constructs or PNCA-PD-L1 by co-expression of SI S3 PTase with heavy-NGP and light chain constructs.

[0071] Figure 15B depicts a western blot of purified Ab-PD-Ll or PNCA-PD-L1 treated with PNGase-F or Endo-H enzymes and heavy and light chain, probed for human IgG.

[0072] Figure 15C depicts CI-MPR affinity chromatograph for purified AF594 conjugated Ab-PD-Ll -AF594.

[0073] Figure 15D depicts CI-MPR affinity chromatograph for purified PNCA-PD-L1- AF594.

[0074] Figure 16A depicts a schematic of the internalization of cell surface PD-L1 or EGFR by PNCA antibodies and CI-MPR to mediate receptor internalization and lysosomal degradation.

[0075] Figure 16B depicts the internalization of Ab-PD-Ll and PNCA-PD-L1 in fibroblast cells. AF594 labeled antibodies were incubated with the cells and co-stained with LysoTracker (green) and Hoechst (blue).

[0076] Figure 16C depicts a western blot on cell lysates of Daoy (HTB-186) cells incubated alone or with 1, 5 or 20 nM Ab-PD-Ll or PNCA-PD-L1 for 24 h. Western blot was performed on cell lysates to detect total PD-L1 or GAPDH, loading control.

[0077] Figure 16D depicts a western blot on cell lysates of A431 cells incubated alone or with 1, 5 or 20 nM Ab-PD-Ll or PNCA-PD-L1 for 24 h. Western blot was performed on cell lysates to detect total PD-L1 or GAPDH, loading control.

[0078] Figure 17A depicts a schematic of constructs expressed to produce Ab-EGFR by heavy and light chain constructs or PNCA-EGFR by co-expression of SI S3 PTase with heavy-NGP and light chain constructs.

[0079] Figure 17B depicts SDS-PAGE and Coomassie staining of purified Ab-EGFR and PNCA-EGFR.

[0080] Figure 17C depicts a western blot of purified Ab-EGFR and PNCA-EGFR treated with PNGase-F and Endo-H antibodies. Samples were probed for human IgG heavy and light chain.

[0081] Figure 17D depicts CI-MPR binding affinity of purified Ab-EGFR and PNCA-EGFR conjugated with alexa fluor 594 (AF594). Binding was detected by fluorescence for PNCA- EGFR-AF594 or Ab-EGFR-AF594.

[0082] Figure 17E depicts a CI-MPR affinity chromatograph for Ab-EGFR- AF594.

[0083] Figure 17F depicts a CI-MPR affinity chromatograph for PNCA-EGFR-AF594.

[0084] Figure 18A depicts fluorescent staining of fibroblast cells incubated with AF594 conjugated purified Ab-EGFR-AF594 or PNCA-EGFR-AF594 and treated with or without 2 mM M6P. Antibody uptake is indicated by red signal. Lysosomes were stained by incubation with lysotracker (green) and nuclei were stained using Hoechst (blue).

[0085] Figure 18B depicts fluorescent staining of HepG2 cells incubated with AF594 conjugated purified Ab-EGFR-AF594 or PNCA-EGFR-AF594 and treated with or without 2 mM M6P. Antibody uptake is indicated by red signal. Lysosomes were stained by incubation with lysotracker (green) and nuclei were stained using Hoechst (blue).

[0086] Figure 18C depicts western blot of Daoy (HTB-186) cells treated using 20 nM of Ab- EGFR antibody or PNCA-EGFR for 24 h, examined for total EGFR protein or GAPDH. Relative quantification of EGFR:GAPDH ratio is shown below EGFR signal

[0087] Figure 18D depicts a quantification of the EGFR signal observed in Figure 18C. Dot represents each independent experiment. N=4. Significant testing was performed by paired two-tailed t-tests.

[0088] Figure 19A depicts a western blot of Daoy cells incubated with 5 nM or 20 nM Ab- EGFR or PNCA-EGFR for 8 h or 24 h. Total EGFR and GAPDH were assessed. % indicates the amount of EGFR left compared to control group (data normalized to GAPDH level)

[0089] Figure 19B depicts a schematic of cell co-culture setup with transfected HEK293T cells producing anti-EGFR antibodies or co-expression with SI S3 to produce PNCA forms and incubation with Daoy cells in the upper chamber.

[0090] Figure 19C depicts a western blot examining EGFR levels in Daoy cells following 24 hours co-culture. GAPDH was used as a loading control.

[0091] Figure 19D depicts a quantification of the EGFR signal observed in Figure 19C. Dot represents each independent experiment. N=3. Significant testing was performed by paired two-tailed t-tests.

[0092] Figure 20A depicts a western blot of co-cultured Daoy cells probed for total PD-L1 or GAPDH.

[0093] Figure 20B depicts a quantification of the PD-L1 signal observed in Figure 20A.

[0094] Figure 21 is a graphical depiction of the single chain antibodies that will be expressed together with SI S3 in a viral vector.

[0095] Figure 22 is a schematic depiction showing human GlcNAc-1 -phosphotransferase (designated hGNPTAB) and exemplary truncated forms thereof, including SI S3. The core catalytic domains are the sequences represented by numbers 1, 2, 3 and 4. The structures designated G0051 and G0067— G0069 depict truncated forms of GlcNAcPTase in which the dictyostelium sequence and / or transmembrane or cytosolic domains have been removed.

[0096] Figure 23 depicts the results of a blocking assay comparing the ability of PNCA-PD- L1 to block the biding of PD-1 to its receptor PD-L1 compared to Atezolizumab.

[0097] Figure 24 depicts the results of an ELISA assay showing that PNCA-PD-L1 can induce the release of the proinflammatory cytokines IL-2 (left) and IFN-y (right) from T-cells comparable to or better than Atezolizumab.

[0098] Figures 25A-25B depict the ability of PNCA-PD-L1 to activate an anti -turn or effect in T-cells compared to Atezolizumab. Figure 25A shows representative images indicating reduced crystal violet signals in PNCA-PD-L1 treated RKO cells co-cultured with T cells; no changes were observed in the control RKO cells alone. Figure 25B shows that PNCA-PD-L1 introduces enhanced tumor cell killing at higher concentrations compared to the commercial product Atezolizumab.

[0099] Figure 26 the design parameters of an in vivo pharmacokinetics study in WT C57BL / 6 mice to compare the effects of regular human TNFa, hTNFa conjugated with an NGP peptide containing 1 mannose 6-phosphate site (PNCA- hTNFa-lNGP), and hTNFa conjugated with an NGP peptide containing 6 mannose 6-phosphate sites (PNCA- hTNFa-6NGP).

[0100] Figure 27 depicts the results of the in vivo pharmacokinetics study of Figure 26. The graph represents the results of an ELISA assay showing that the PNCA- hTNFa- 1NGP peptide has a similar half-life to regular human TNFa in mice, and the PNCA- hTNFa- 6NGP peptide design has a shorter serum half-life, indicating faster internalization of PNC A- hTNFa-6NGP.

[0101] Figure 28 depicts the design parameters of an in vivo efficacy study to assess MC38 tumor inhibition in PD-1 humanized mice by a phosphorylated PNCA-PD-L1 chimeric antibody containing 6 mannose 6-phosphate sites (PNCA-PD-L1-6NGP).

[0102] Figures 29A-29B depict the results of the in vivo efficacy study of Figure 28. Figure 29A shows the tumor volume post-treatment. Figure 29B shows the body weight posttreatment. The results show that the PNCA-PD-L1-6NGP reduced tumor volume without significant change in body weight compared to vehicle alone.DETAILED DESCRIPTION

[0103] The present invention is directed to novel lysosome-targeting compositions and methods that allow for any protein, peptide, antibody, antibody fragment, or antigen to be targeted to the lysosome for degradation. The present invention enables the development of a novel platform technology for producing bi-functional proteins or antibodies, or antibody binding fragments, with high levels of mannose N-glycans or mannose 6-phosphate glycans. The bifunctional proteins are obtained directly from producing cells, and exhibit high affinity binding to cell surface cation-independent mannose-6-phosphate receptors followed by internalization and delivery to lysosomes. This leads to efficient degradation of both extracellular soluble and membrane proteins. This approach provides a strategy to produce bi-functional proteins or antibodies for selective protein degradation.

[0104] The novel lysosome-targeting compositions, in short and described in further detail in the paragraphs below, comprise (1) an N-Glycosylated Peptide (“NGP”) allowingfor the addition of N-glycans to a peptide when a nucleic acid sequence encoding the peptide is expressed in a cell; (2) an NGP Chimeric Protein (“NCP”) comprising a fusion polypeptide or protein comprising an NGP joined or fused to protein or peptide sequence; and (3) an NGP Chimeric Antibody (“NCA”) comprising a fusion polypeptide or protein comprising an NGP joined or fused to an antibody, antibody fragment, or heavy and / or light chain sequence that recognizes a desired protein or peptide targeted for lysosomal degradation.

[0105] Figure l is a schematic representation of an example of a high-mannose N- glycan (left), a mono-phosphorylated high-mannose N-glycan (middle left), a bisphosphorylated high-mannose N-glycan (middle right), and a complex N-glycan (right), and their sensitivity for Endo H and peptide glycosidase F (PNGase F) digestion. Endo H is a highly specific endoglycosidase which cleaves asparagine-linked mannose rich oligosaccharides, but not processed complex oligosaccharides from glycoproteins. PNGase F can remove any type of N-glycans from glycoproteins. Freeze et al, Curr Protoc Mol Biol. 2010, doi: 10.1002 / 0471142727 (AH Note: this reference describes a general protocol for the use of both enzymes).

[0106] The NGPs contain variant natural N-glycosylation sites, which allow cells to add N-glycans post-translationally to the peptide when a nucleic acid sequence encoding the peptide is expressed in the cell.

[0107] The NCPs and NCAs are recombinantly produced by joining or fusing a nucleic acid sequence encoding the NGP to a nucleic acid sequence encoding a polypeptide or protein of interest to form a nucleic acid construct, and expressing the nucleic acid construct in a host cell. The NCPs and NCAs produced in cells contain multiple complex N- glycans. Complex N-glycans are problematic because they are not recognized by cell receptors.

[0108] To remedy this problem, NCPs and NCAs may be produced in the presence of an inhibitor of ER / Golgi mannosidases, which cause the N-glycans on the NGP portion of the fusion protein to be maintained as high mannose sugars, which contain terminal mannose residues and can bind to cell surface mannose receptors for lysosome targeting. The NCP or NCA then can be internalized to the lysosome in macrophage-related cell types.Alternatively, the NCPs or NCAs may be expressed in a gene knockout cell line with deficient of P-1, 2-N-acetylglucosaminyltransf erase 1 (MGAT1), for example, the cell linedescribed in US Pat. Pub. No.US20240167010A9). In one embodiment, the mannosidase inhibitor comprises kifunensine. In one embodiment, the gene knockout cell line is a cell line with deficient of P-1, 2-N-acetylglucosaminyltransferase 1 (MGAT1).

[0109] In an alternative approach, the NCPs or NCAs are expressed together with a nucleic acid sequence encoding a GlcNAc-1 -phosphotransferase, or a truncated form of GlcNAc-1 -phosphotransferase (“GlcNAc PTase”) that retains the activity of full-length GlcNAc-1-PTase. In one embodiment, the truncated form of GlcNAc- 1 -PTase comprises a nucleic acid sequence encoding a polypeptide comprising the core catalytic domains of GlcNAc-1-PTase. The term “core catalytic domain” refers to the minimum sequences which must be present in a truncated form of GlcNAc PTase to retain its activity. Specifically, the core catalytic domain comprises amino acids 75-86, 322-438, 955-1007 and 1149-1185 of human GlcNAc- 1 -PTase (see Fig. 22). Co-expression of GlcNAc-PTase, or a truncated version thereof, with the nucleic acid sequence encoding the NCP or NCA converts the complex N-glycans on the NGP portion of the recombinant fusion protein to phosphorylated mannose 6-phosphate glycans, which can be recognized by cell surface cation-independent mannose-6-phosphate receptors (CI-MPRs), followed by internalization and delivery to lysosomes.

[0110] In one embodiment, the truncated form of GlcNAc- 1 -PTase is S1S3 phosphotransferase. SI S3 phosphotransferase is described in Lin et al., Mol. Therapy: Methods & Clinical Develop., 2017, Vol. 5:59. In one embodiment, the truncated form of GlcNAcPTase is a truncated form comprising the core catalytic domains of GlcNAC-PTase. Exemplary truncated forms of GlcNAc-PTase based on the core catalytic domains are shown in Figure 22 as G0051 and G0067- G0069, and described in US Serial No. 63 / 678,729, filed Aug. 2, 2024.

[0111] Conditioned medium or purified NCPs or NCAs expressed with GlcNAc PTase, SI S3 phosphotransferase, or other truncated form of GlcNAc PTase containing the core catalytic domain, have glycans that are sensitive to endoglycosidase H (“Endo H”), and can bind to CI-MPRs. This approach allows any protein to be converted to a form having a phosphorylated sugar group recognizable by a lysosome. Stated another way, the present invention is a production system using NCPs or NCAs co-expressed with GlcNAc PTase, SI S3 phosphotransferase, or another truncated form of GlcNAc PTase containing the corecatalytic domain as a platform to produce proteins that can be targeted for lysosomal degradation. NCAs can be produced by joining or fusing a nucleic acid sequence encoding the NGP to a nucleic acid sequence encoding an antibody, or antibody fragment, such as the Fab domain of an antibody, that recognizes a desired protein or peptide sequence, and expressing the nucleic acid construct in a host cell. NCAs show the same recognition of their antigens as the original unmodified antibody, and gain the ability to bind to mannose receptors when produced with kifunensine or other mannosidase inhibitor, or mannose 6- phosphate receptors when expressed together with a GlcNAcPTase, such as SI S3 or other truncated version containing the core catalytic domains of GlcNAcPTase, for lysosome targeting of the antigen. This approach allows any antibody; or binding fragment thereof, to be converted to a form having a phosphorylated sugar group recognizable by a lysosome. Stated another way, the invention comprises a production system using NCAs co-expressed with SI S3 phosphotransferase (“PCNAs”) as a platform to produce antibodies (or binding fragments thereof) that bind with their target antigens / proteins to form a complex that can be targeted for lysosomal degradation.

[0112] The lysosome targeting design (NCP or NCA) provides the possibility for any protein or antibody (and / or its antigen), to be targeted to the lysosome for functional or degradation purposes.

[0113] NGPs. The NGPs of the present invention comprise peptide sequences containing variant natural N-glycosylation sites with Asparagine-x- Serine or Threonine modifications (where x can be any amino acid but not proline), which allow cells to add N- glycans to the peptide when a nucleic acid construct encoding the peptide is expressed in a cell. In one embodiment, the NGP may comprise any one of the following peptide sequences: NXTGGSNXTGGGNXTGGSNXTMMSGNS (SEQ ID NO: 1), NXTGNXTSNXTGNXTSNXTGNXTMMST (SEQ ID NO: 2), NXTNXTNXTNXTNXTNXTNXTNXTMMS (SEQ ID NO: 3), NXTSNXTTNXTTNXTSNXTTNNXSMMS (SEQ ID NO: 4), or NNXSNNXTNNXTNNXSNNXTNNXSMMS (SEQ ID NO: 5), wherein “X” indicates any amino acid except proline. In one embodiment, the NGP comprises any one of the peptide sequences in the previous sentence, wherein “X” indicates Serine (S), Alanine (A), Histidine (H), Glycine (G), Asparagine (N), or Threonine (T).

[0114] In one embodiment, the NGP comprises any one of the peptide sequences depicted in Figure 2. In one embodiment, the DNA sequence encoding the NGP peptide sequences depicted in Figure 2 comprise:

[0115] 4NGP:AACAGCACCGGCGGCAGCAACGCCACCGGCGGCGGCAACAGCACCGGCGGCAG CAACGCCACCATGATGAGCGGCAACAGC (SEQ ID NO: 6)

[0116] 6NGP:AACGCCACCGGCAACAGCACCAGCAACAGCACCGGCAACAGCACCAGCAACAG CACCGGCAACGCCACCATGATGAGCACC (SEQ ID NO: 7)

[0117] 8NGP:AACAGCACCAACGCCACCAACCACACCAACGGCACCAACAACACCAACACCACC AACGGCACCAACAGCACCATGATGAGC (SEQ ID NO: 8)

[0118] 12NGP:AACAACACCAGCAACAACACCACCAACAACACCACCAACAACACCAGCAACAA CACCACCAACAACACCAGCATGATGAGC (SEQ ID NO: 9)

[0119] NCPs. The NCPs of the present invention comprise NGPs joined or fused to a polypeptide or protein of interest. The polypeptide or protein of interest may be any polypeptide or protein for which targeting to the lysosome is desired. The NCPs are produced by expressing in a host cell a nucleic acid construct comprising a nucleic acid sequence encoding the NGP joined or fused to a nucleic acid sequence encoding the polypeptide or protein of interest.

[0120] The examples disclosed herein illustrate a proof-of-concept NCP in which the NGP is fused to a GFP protein (see, e.g., Figure 3). This example is for illustrative purposes and is not intended to limit the class of proteins that may be fused to NGPs according to the present disclosure. In one embodiment, the protein fused to NGP is selected from the list of lysosomal proteins known to be involved in LSDs disclosed in Table 1 A-1C ofWO202 1003442, the contents of which are incorporated fully herein. In another embodiment, the protein fused to NGP comprises non-lysosomal proteins, including without limitation, cytokines, membrane receptors or immune checkpoint molecules. Non-limiting examples of non-lysosomal proteins of interest include Tumor necrosis factor-a (TNFa), interferons,interleukins, IL-12, growth factors (e.g., IGF, EGFR), insulin, adrenaline, as well as any non- lysosomal proteins associated with the onset or progression of cancer.

[0121] In some embodiments, the NCP may comprise an antigen or ligand that will bind to autoantibodies. For example, some autoimmune diseases, including Type 1 diabetes, Antiphospholipid Syndrome and Lupus, are caused by the overexpression of autoantibodies which attack healthy tissues. The present invention enables production of NCPs in which the protein of interest is an antigen sequence specific for the autoantibodies having increased levels of M6P. The phosphorylated NCPs form a complex with the autoantibodies that steer the autoantibodies to the lysosome for degradation.

[0122] NCAs. The NCAs of the present invention comprise NGPs fused to a Lactase chimeric antibody, or antibody fragment. The term antibody is intended to encompass antibody fragments capable of binding an antigen. The chimeric antibody may be any chimeric antibody of interest for which targeting the antibody itself or its antigen to the lysosome is desired. The NCAs are produced by expressing in a host cell a nucleic acid construct comprising a nucleic acid sequence encoding the NGP joined or fused to a nucleic acid sequence encoding the antibody or antibody fragment of interest.

[0123] The examples disclosed herein illustrate a proof-of-concept NCA in which the NGP is fused to portions of the heavy and light chain of the anti-Flag antibody. This example is for illustrative purposes and is not intended to limit the class of chimeric antibodies that may be fused to NGPs according to the present disclosure. In one embodiment, the chimeric antibody fused to NGP is a chimeric antibody that recognizes one of the lysosomal proteins known to be involved in LSDs disclosed in Table 1 A-1C of W02021003442, the contents of which are incorporated fully herein. In another embodiment, the chimeric antibody fused to NGP is a chimeric antibody that recognizes antigens associated with diseases including, but not limited to, cancer, infectious diseases, Alzheimer’s disease, Crohn’s disease, rheumatoid arthritis, psoriatic arthritis, hereditary high cholesterol, atherosclerosis, psoriasis, systemic lupus, or erythematosus. Antigens associated with these diseases include, without limitation, amyloid beta, amyloid beta protofibrils, TNF-a, EGFR, PD-L1, PCSK9, IL-36R, IL- 17A / 17F, IL-23pl9, IL-17R, IL-17a, IL-12 / 23, CDl la, IFNAR1, and BLyS.

[0124] Mannosidase Inhibitors. The present disclosure is also directed to a method for producing NCPs or NCAs in which the N-glycans on the NGP portion of the NCP or NCA are maintained as high mannose sugars containing terminal mannose residues that can bind tocell surface mannose receptors for lysosome targeting. The method comprises (a) providing a nucleic acid construct comprising a nucleic acid sequence encoding the NGP joined or fused with a nucleic acid sequence encoding the polypeptide or protein of interest, or the antibody or antibody fragment of interest; (b) introducing the nucleic acid construct into a host cell together with an inhibitor of ER / Golgi mannosidase. Expression of the NCP or NCA in the presence of an inhibitor of ER / Golgi mannosidase, results in the N-glycans on the NGP portion of the NCP or NCA fusion protein to be maintained as high mannose sugars containing terminal mannose residues that can bind to cell surface mannose receptors for lysosome targeting. Any methods and / or compositions for generating terminal mannose residues can be used. Mannosidase inhibitors include, without limitation, swainsonine, kifunensine, 1-deoxymannojirimycin hydrochloride, (lS,2S,3R)-l,2,3-trihydroxy-4- cyclopropene 2,3 -cyclohexyl ketal, or N-butyldeoxymannojirimycin HC1. In one embodiment, the mannosidase inhibitor is kifunensine. Additionally, expression of the NCP or NCA in the presence of an inhibitor of the mannosidase 1 enzyme results in high-mannose glycan groups on the N-glycan portion of the NCP or NCA that can target mannose receptors on macrophages and dendritic cells. More efficient antigen delivery to macrophages and dendritic cells through mannose receptors could enhance the innate adaptive immune responses by high levels antigen presentation. Other methods to modify the N-glycans could be combined with the NCPs or NCAs for targeting other cell surface sugar receptors, as asialoglycoprotein receptors in liver cells.

[0125] The host cell may be any mammalian cell. In one embodiment, the host cell is a CHO cell, an HEK293 cell or an sf9 cell.

[0126] Alternatively the method comprises (a) providing a nucleic acid construct comprising a nucleic acid sequence encoding the NGP joined or fused with a nucleic acid sequence encoding the polypeptide or protein of interest, or the antibody or antibody fragment of interest; (b) introducing the nucleic acid construct into a host cell in which the MGAT gene has been knocked out. Expression of the NCP or NCA in a host cell lacking the mannosidase gene results in the N-glycans on the NGP portion of the NCP or NCA fusion protein to be maintained as mannose sugars containing terminal mannose residues that can bind to cell surface mannose receptors for lysosome targeting.

[0127] GlcNAc-1 -phosphotransferase. The present disclosure is also directed to expression of NCPs or and NCAs in the presence of native or modified versions of GlcNAc-1 -phosphotransferase. In some embodiments, NCPs and NCAs of the present disclosure are expressed in the presence of full length GlcNAcPTase, SI S3 PTase, or another truncated version of GlcNAcPTase comprising the core catalytic domains of GlcNAcPTase. The method comprises (a) providing first a nucleic acid sequence comprising a nucleic acid sequence encoding the NGP joined or fused with a nucleic acid sequence encoding the polypeptide or protein of interest, or the antibody or antibody fragment of interest; (b) providing a second nucleic acid sequence encoding full length GlcNAcPTase, SI S3 PTase, or a truncated version of GlcNAcPTase comprising the core catalytic domains of GlcNAcPTase; and (c) introducing the first and second nucleic acid sequences into a host cell. Coexpression of the NCP or NCA with GlcNAcPTase or a truncated version of GlcNAc-1- PTase, results in conversion of complex N-glycans to mannose 6-phosphate glycans in cells to produce phosphorylated NCPs (“PNCPs”) or phosphorylated NCAs (“PNCAs”). SI S3 PTase is described in Lin et al., Mol. Therapy: Methods & Clinical Develop., 2017, Vol. 5:59. Other truncated forms of GlcNAcPTase comprising the core catalytic domains are shown in Figure 22 as G0051 and G0067-69 and described in US Serial No. 63 / 678,729, filed Aug. 2, 2024, the entirety of which is incorporated herein by reference.

[0128] Alternatively, the method comprises providing a nucleic acid sequence that encodes both the NCP or NCA and_GlcNAcPTase or a truncated version of GlcNAc-1 -PTase, and introducing the nucleic acid sequence into a host cell. The host cell may be any mammalian cell. In one embodiment the host cell is a CHO cell, an HEK293 cell or an sf9 cell.

[0129] Expression Vectors. The present disclosure is also directed to an expression vector comprising a nucleic acid sequence encoding an NCP or NCA or PNCP or PNCA. The term “expression vector” as used herein refers to a vector comprising a nucleic acid sequence encoding a recombinant polynucleotide comprising expression control sequences operatively linked to a nucleotide sequence to be expressed. An expression vector comprises sufficient cis-acting elements for expression; other elements for expression can be supplied by the host cell or in an in vitro expression system. Expression vectors include all those known in the art, such as cosmids, plasmids (e.g., naked or contained in liposomes) and viruses (e.g., lentiviruses, retroviruses, adenoviruses, and adeno-associated viruses) that incorporate the recombinant polynucleotide. In some embodiments, the expression vector comprises a viral vector. In one embodiment, the expression vector further comprises expression of native or modified versions of GlcNAc-1 -phosphotransferase, such as SI S3 phosphotransferase fordelivery into a host cell, such as the two-promoter vector disclosed in WO 2023 / 150051 Al. The present disclosure is also directed to providing the expression vector to a cell. Expression vector technology is well known in the art and is described, for example, in Sambrook et al. (2001, Molecular Cloning: A Laboratory Manual, Cold Spring Harbor Laboratory, New York), and in other virology and molecular biology manuals. A cell comprising the expression vector may be used for protein expression and, optionally, purification. Methods for expressing and, optionally, purifying an expressed protein from a cell are standard in the art.

[0130] In some embodiments, the cell with the expression vector may be used to produce a polypeptide encoded by the NCA or NGP or PNCP or PNCA constructs of the disclosure. Generally, production of a polypeptide of the disclosure involves transfecting cells with an expression vector comprising an enzyme construct and then culturing the cells so that they transcribe and translate the desired polypeptide. The isolated cells may then be lysed to extract the expressed polypeptide for subsequent purification.

[0131] In some embodiments, the viral vector comprises an AAV vector or a lentiviral vector. In some embodiments, the AAV vector comprises a sequence isolated or derived from an AAV of serotype AAV1, AAV2, AAV3, AAV4, AAV5, AAV6, AAV7, AAV8 or AAV9. In some embodiments, the delivery vector comprises a non-viral vector. In some embodiments the non-viral vector comprises a liposome, a lipid nanoparticle (LNP), a micelle, a polymersome, a nanoparticle, a polymer nanoparticle, or an exosome.

[0132] In some embodiments, the vector is an adenoviral vector or an adeno- associated viral (AAV) vector. In some embodiments, the AAV vector comprises a serotype selected from the group consisting of AAV1, AAV2, AAV3, AAV4, AAV5, AAV6, AAV7, AAV8, and AAV9. In some embodiments, the AAV vector comprises a sequence encoding a capsid isolated or derived from one or more of a serotype selected from the group consisting of AAV1, AAV2, AAV3, AAV4, AAV5, AAV6, AAV7, AAV8, and AAV9. In some embodiments, the AAV vector comprises a sequence encoding at least one inverted terminal repeat (ITR) isolated or derived from one or more of a serotype selected from the group consisting of AAV1, AAV2, AAV3, AAV4, AAV5, AAV6, AAV7, AAV8, and AAV9.

[0133] In some embodiments of the compositions of the disclosure, the vector is a bicistronic vector.

[0134] In some embodiments of the compositions of the disclosure, the vector is a multi ci str onic vector.

[0135] In some embodiments of the compositions of the disclosure, the promoter comprises a ubiquitous promoter. In some embodiments, the promoter is capable of driving expression in a mammalian cell. In some embodiments, the promoter is capable of driving expression in a human cell.

[0136] Therapeutic Methods of Use. The bispecific NCPs and NCAs and their phosphorylated counterparts, PNCPs and PNCAs, described herein may be used as therapeutic molecules for the treatment numerous diseases and conditions involving proteins. For example many diseases are characterized by the expression or overexpression of certain proteins contribute to the onset or progression of the disease. The PNCPs and PNCAs of the present invention can be configures to a therapeutic humanized antibody for lysosomal degradation of proteins of interest, including cytokines, membrane receptors, and immune checkpoint inhibitors. Soluble cytokines, such as TNFa, and cell surface transmembrane proteins, such as PD-L1 and EGFR, have been shown to be implicated in cancer.

[0137] Targeting of soluble cytokines using monoclonal antibodies as inhibitors has been utilized therapeutically for many years. For example, Adalimumab works by binding and blocking the tumor necrosis factor-alpha (TNFa) interaction with its receptors. The PNCA technology of the present invention can be applied to produce a therapeutic humanized antibody for lysosomal degradation of soluble cytokines. As shown in the Examples, an N-glycosylated chimeric anti-TNFa (NCA-TNFa) sequence containing the human IgGl Fc region fused with the N-glycosylated peptide in the C-terminal of heavy chain of the adalimumab sequence. The resulting PNCA-TNFa was generated by coexpression of the chimeric antibody sequence with the SI S3 PTase in ExpiCHO cells. The results showed that the PNCA-TNFa retained its interaction with its soluble TNFa antigen, and the PNCA-antibody-antigen complex can be internalized to the lysosome for degradation.

[0138] In another Example, the present invention was used to create a PNCA with an immune checkpoint inhibitor, PD-L1 to demonstrate that the platform can be used to createmolecules that target cell surface transmembrane receptors for lysosomal degradation. In this Example, a variable sequence of an anti-PD-Ll antibody, Atezolizumab fused to our N- glycosylated peptide design. The resulting phosphorylated N-glycosylated peptide chimeric PD-L1 antibody (PNCA-PD-L1) was produced with co-expression of SI S3 PTase and purified from transient transfected ExpiCHO cells. The results showed that the PNCA-PD- L1 retained its interaction with its PD-L1 receptor, and the PNCA-antibody-antigen complex can be internalized to the lysosome for degradation.

[0139] The technology of the present invention can also be applied to antigen / ligand instead of antibody. Many autoimmune diseases, such as Type 1 Diabetes, Antiphospholipid Syndrome and Lupus, are caused by the overexpression of antibodies, which attack healthy tissues. The present invention can be applied to create NCPs or PNCPs in which the NGP is fused with a ligand specific for the autoantibodies. When expressed together with a GlcNAcPTase or truncated form thereof, the resulting PNCAs exhibit increased levels M6P groups that remove the autoantibodies to the lysosome for degradation.

[0140] The technology of the invention also may be applied to gene and cellular therapy applications. For these applications, the sequences encoding the NCP or NCA may be incorporated into viral constructs (such as lentivirus) or non-viral particles for cell and gene therapy. As demonstrated in the Examples, full antibody sequences, ScAb, ScFv or nanobody sequences fused with the chimeric NGP co-expressed with the a GlcNAc-PTase, such as SI S3 PTase, could be packaged and transduced to stem cells, T cells or B cells for long-term antibody expression in vivo to treat immune, neurodegenerative, or other diseases. Compared to the regular antibody infusion treatment, the gene and cell therapy would last for years and improve the patient’s quality of life.Examples

[0141] The disclosure is described with reference to the following Examples. These Examples are provided for the purpose of illustration only and the disclosure should in no way be construed as being limited to these Examples, but rather should be construed to encompass any and all variations which become evident as a result of the disclosure provided herein.

[0142] Without further description, it is believed that one of ordinary skill in the art can, using the preceding description and the following illustrative examples, make and utilize the compounds of the present disclosure and practice the claimed methods. The following working examples are not to be construed as limiting in any way the present disclosure.

[0143] The following materials and methods were used for the examples disclosed herein.

[0144] Construct and vector design: Protein sequences for antibody variable regions were obtained from anti-FLAG M2 (PDB ID: 2G60), anti-TNF adalimumab variable (US Patent US6258562), anti-EGFR cetuximab variable (US Patent US6217866B1), anti-PD-Ll atezolizumab variable (US Patent US8217149). Expression plasmids were generated by DNA fragments from Twist Bioscience and placed in pcDNA3.1, pTwist vector or variable regions ordered in pTwist CMV-hlgGl or pTwist CMV-hlgk for expression of heavy or light chain respectively. PCR primers were from Azenta Biosciences. PCR amplifications using Q5 polymerase (NEB, M0491S) and restriction digest for ligation or HiFi cloning of fragments (NEB, M0492S). Large scale DNA purification was performed using a Zymo Maxi-prep kit (Zymo, D4203). The N-glycosylated peptide (NATGNSTSNSTGNSTSNSTGNATMMST, SEQ ID NO: 10) was designed based on NX(S / T) N-glycosylation sequence. All plasmid sequences were confirmed by Sanger or Full plasmid sequencing.

[0145] Antibody production, purification and conjugation of Alexa Fluor-594: ExpiCHO cells (ThermoFisher, A29127) were cultured and transfected with plasmids for heavy chain and light chain expression with or without co-expression of SI S3 PTase plasmid following the manufacture protocol (ThermoFisher, A29131). Cells were cultured for 7-8 days, and conditioned media was harvested and filtered with 0.22 um filter. Antibodies were purified from the conditioned media using Nab Protein G spin columns (ThermoFisher, 89957) and eluted in Pierce IgG Elution Buffer (ThermoFisher, 21004). Purified antibodies were dialyzed in PBS buffer with 1 :300 - 500 ratios (sample volume to PBS) overnight, using slide-a-lyzer cassettes with 20 K molecular weight cut-off (ThermoFisher, 66005). 100 ug of purified antibodies were conjugated with Alexa Fluor-594 using the Alexa Fluor 594 antibody labeling kit (ThermoFisher, A20185) following manufacturer’s protocol.

[0146] Endoglycosidases analysis: 5 uL of GFP conditioned medium or 1 ug of each purified antibodies was treated with PNGase-F (NEB, P0704L) or Endo-H (NEB, P0702L) following the manufacturing protocol in a total 20 uL reaction volume. 1 ul (50 ng) of sample was loaded for SDS-PAGE for western blot analysis.

[0147] Cell culture: human fibroblast cells GM16648 (Coriell Institute for Medical Research) and Tay-Sachs patient fibroblast cells GM02968 (Coriell Institute for MedicalResearch) were cultured in MEM (Corning, 10-009-CV) with 15% FBS (Neuromics, FBS001). HepG2 (ATCC, HB-8065), Daoy (ATCC, HTB-186) and A431 (Sigma, 85090402) cells were cultured in MEM medium (Corning, 10-009-CV) with 10 % FBS. MDA-MB-231 (Sigma, 92020424) and HEK293T (ATCC, CRL-3216) cells were cultured in DMEM (Corning, 10-013 -CV) with 10% FBS.

[0148] Generation of CI-MPR knockout cell line: HEK293T were transfected with pX330 containing sgRNA to target human IGF2R / CI-MPR using the oligos (caccgAAGTGCAACCAGATCTCTCC (SEQ ID NO: 11) and aaacGGAGAGATCTGGTTGCACTTC (SEQ ID NO: 12)) (Addgene plasmid ID: 42230) following the protocol disclosed in Ran, F.A., et al., Genome engineering using the CRISPR- Cas9 system. Nat Protoc, 2013. 8(11): p. 2281-2308. Cells were selected with puromycin for 48 hours, and single cells plated in 96 well plates by dilution cells to 0.5 cell per 100 uL. Loss of CI-MPR expression was confirmed by Western blot.

[0149] HexM uptake and activity: Cells were seeded in 12-well plate overnight. HexM-producing conditional media (20ul / well) alone or combined with antibody (20nM, final concentration) with or without M6P (2mM, final concentration) were added accordingly into each well for 4 hours. Cells were then harvested and washed with PBS twice. Cell pellets were lysed in M-PER buffer (ThermoFisher, 78501) with protease inhibitor(ThermoFisher, 78440). HexM activity was detected by 4-methylumbelliferyl (4-MU) substrate method to a-Hexosaminidase activity (HexA). Briefly, 5 ul of cell lysate was incubated with 1 mM of 4-methylumbelliferyl 6-sulfo-N-acetyl-b-D-glucosaminide (Sigma, 454428) in 10 mM citrate phosphate buffer with 0.5% Triton-X-100, pH 4.2 for 1 hour. 4- MU fluorescence was measured by SpectraMax ID3 (Molecular Devices, Sunnyvale, CA) with 360 nm excitation, 460 nm emission. See Tropak, M.B., et al., Pharmacological enhancement of beta-hexosaminidase activity in fibroblasts from adult Tay-Sachs and Sandhoff Patients. J Biol Chem, 2004. 279(14): p. 13478-87.

[0150] Antibody internalization experiment: Human fibroblast cells (GM16648) were seeded in a 6-well plate with 3x105 cells per well 24 hours before staining. 10-20 nM Alexa Fluor 594 labeled antibodies were mixed with the culture medium and applied to the cells for 6 hours. To co-stain with nuclei and lysosomes, medium containing 10 ug / ml of Hoechst 33342 stain (ThermoFisher, H3570) and 1 uM LysoTracker Green DND-26 (ThermoFisher, L7526) was loaded to cells for 10 minutes. Then cells were washed once with PBS buffer(Corning, 21-040-CV). Images were taken under the Invitrogen EVOS M5000 Imaging System with 20x objective. Data was further processed by Fiji Image J. See Schindelin, J., et al., Fiji: an open-source platform for biological-image analysis. Nat Methods, 2012. 9(7): p. 676-82.

[0151] mCherry-FLAG internalization experiment: 1 xl06 HepG2 cells were plated in 12-well plate for 24 hours and then incubated with FLAG-mCherry (1 ug) with or without Ab-FLAG or PNCA-FLAG at concentrations of 20 nM for 24 h. Cells were washed three times with PBS buffer (Corning, 21-040-CV) before imaging and harvesting for western blotting analysis. Data was processed and quantified by Fiji Image J. See Schindelin, J., et al., Fiji: an open-source platform for biological-image analysis. Nat Methods, 2012. 9(7): p. 676- 82.

[0152] TNFa internalization experiment: 1 xl06 HepG2 cells were plated in 12-well plate 24 hours ahead and incubated with TNFa alone or TNFa plus its antibody Ab-TNFa or PNCA-TNFa at the indicated concentrations for 24 h. Cells were washed three times and harvested and lysed in M-PER (ThermoFisher, 78501) with Protease and Phosphatases inhibitor cocktail (ThermoFisher, 78440) for western blot analysis.

[0153] EGFR internalization experiment: 5 xl05 Daoy cells were plated in 12-well plate and incubated with Ab-EGFR or PNCA-EGFR at various concentrations indicated for different amount of culture time as indicated. Cells were washed three times with PBS buffer and harvested and lysed in RIPA buffer (ThermoFisher 89900) with Protease and Phosphatases inhibitor cocktail (ThermoFisher, 78440).

[0154] PD-L1 internalization experiment: 5 xl05 Daoy, A431 or MDA-MB-231 cells were plated in 12-well plate and incubated with Ab-PD-Ll or PNCA-PD-L1 at various concentrations indicated for different amount of culture time as indicated. Cells were washed three times with PBS buffer and harvested and lysed in RIPA buffer (ThermoFisher 89900) with Protease and Phosphatases inhibitor cocktail (ThermoFisher, 78440).

[0155] Co-culture experiment: One day before transfection, 1 xl06 HEK293T cells were seeded in a 6-well plate, and 5 xl05 Daoy cells were seeded in an insert, which was first cultured in a different 6-well plate. Regular anti-EGFR or anti-PD-Ll antibody plasmids orantibody plasmids containing a motif for the modified N-glycosylated peptide with the SI S3 PTase plasmid were co-transfected to HEK293T cells with Lipofectamine 3000 (ThermoFisher, L3000008) following the manufacturing protocol. 24-hour post-transfection, new medium was replaced in HEK293T cells and the insert containing Daoy cells was inserted to the well with the transfected HEK293T cells. Co-culture was performed for another 24 hours. The Daoy cells were washed three times with PBS buffer and harvested for further analysis.

[0156] SDS-PAGE and Western blot: Protein concentration was determined by BCA assay (ThermoFisher, 23227). 20 - 25 ug total protein were mixed with Pierce Lane Marker Reducing Sample Buffer (ThermoFisher, 39000) and boiled at 100 C for 5 minutes. Samples were loaded on Invitrogen NuPAGE 4-12% Bis-Tris Gels (ThermoFisher, NP0321BOX or NP0336BOX)) and separated by 150V. Gels for Coomassie staining was performed by InstantBlue Coomasssie Stain (Abeam, abl 19211). Gels for western were transferred to nitrocellulose membrane (Amersham, 10600011) overnight. After washing once with lx PBST buffer (VWR, MSPP-IBB-171), membrane was blocked with 5% non-fat milk in PBST or 5% BSA (Sigma-Aldrich, A9085) at RT for 1 hour. Primary antibodies Rabbit anti- mCherry (ThermoFisher, PA5-34974) with 1 : 1000 dilution, Rabbit anti-TNFa (Cell Signaling, 6945s) with 1 : 1000 dilution, Rabbit anti-EGFR (Cell Signaling, 4267s) with 1 :2000 dilution, Rabbit anti-PD-Ll (Abeam, ab228415) with 1 : 1000 dilution and Rabbit anti- GAPDH (ThermoFisher, PAI -987) with 1 :2000 dilution in blocking buffer were incubated for another 1 hour at room temperature. HRP -linked anti-Rabbit IgG (Cell Signaling, 7074) was diluted with 1 :5000 dilution for imaging under Azura. Data was quantified by Fiji Image J.

[0157] CI-MPR Affinity Chromatography Profiling for Antibodies: Pharmacia Fine Chemicals FPLC (Liquid Chromatography Controller LCC-500 and P-500 pumps) associated with 1 mL Bovine CLMPR agarose resin was used for the analysis. 3 ug of AF594 labeled antibody in 100 pL of buffer A (50mM Imidazole, 150mM NaCl, 2mM EDTA, 5mM betaglycerophosphate, 0.05% Tx-100, 0.02% NaN3, pH 6.8) was injected and 0.5 mL / min flow rate was applied with total 5 mL buffer A to flow through any unbound material. 5 mL of linear gradient of 0 mM to 2 mM of mannose 6-phosphate (Sigma, M3655) by mixing buffer A and buffer B (50mM NaOAc, 150mm NaCl, 2mM EDTA, 5mM beta-glycerophosphate, 0.05% Tx-100, 0.02% NaN3, 10 mM M6P, pH 4.8) was applied to elute any loosely boundantibody. Final elution was performed with 5 mL buffer B to fully elute any bound material in the CI-MPR column. The column was then re-equilibrated with 4 volumes of buffer A. Every 0.2 mL fraction was collected in a clear 96-well plate during the process. Fraction collection plate was measured for AF594 fluorescence using the SpectraMax ID3 (Molecular Devices, Sunnyvale, CA) with 580 nm excitation, 620 nm emission. Raw fluorescence units were graphed against collection volume to determine the % antibody that binds to CI-MPR.

[0158] CI-MPR plate binding assay: Purified bovine CI-MPR was immobilized on a clear flat bottom high binding 96-well plate (Costar, 3601) at a concentration of 1 ug / well. Following 2% bovine serum albumin (Sigma, A9085) blocking, samples were serially diluted into Hepes buffer (50 mM Hepes, 150 mM NaCl, 0.05% Tween-20, pH6.8) and incubated for one hour at room temperature on the CI-MPR coated plate. After incubation, the plate was washed three times with Hepes buffer and read for fluorescence using the SpectraMax ID3 (Molecular Devices, Sunnyvale, CA) with 580 nm excitation, 620 nm. Raw fluorescence units were graphed against the total concentration of antibody applied. The total fluorescence per dilution was also read to ensure antibodies at the same concentration yielded similar total fluorescence.

[0159] Example 1: Design ofNGP, NCP, and NCA for N-glycosylation

[0160] NGP sequences with 4, 6, 8, and 12 N-glycosylation sites (4NGP, 6NGP, 8NGP, and 12NGP, respectively) were designed as depicted in Figure 2 (SEQ ID NOs: 20- 23).

[0161] DNA constructs were designed to encode the NGPs of Figure 2 and GFP (see Figure 3, top, for schematic of constructs), the expression of which produces an NGP chimeric GFP protein, (“NCP-GFP”). CHO cells were transfected with the constructs and conditioned medium from the CHO cells then treated with Endo H or PNGase F. Western blotting with an anti -GFP antibody illustrates fusion of the NGPs to GFP (Figure 3). Endo H resistant / PNGase F sensitive glycans were identified for all constructs (Figure 3).

[0162] The peptide containing 6 potential N-glycan sites was tested by fusion to the C -terminus of enhanced green fluorescent protein (GFP), which does not contain any N- glycan sites along with the inclusion of the IgK signal sequence to make a secreted soluble GFP, described as an N-glycosylated peptide chimeric protein-GFP (“NCP-GFP”). Figure 4Ashows a schematic to produce NCP-GFP in ExpiCHO cells or phosphorylated N-glycosylated chimeric GFP protein (PNCP-GFP) by co-expression of the SI S3 PTase. To determine the types of N-glycans on the NCP-GFP and PNCP-GFP proteins, conditioned medium was harvested and treated with Peptide-N-Glycosidase F (PNGase-F) which cleaves off all N- glycans or Endoglycosidase H (Endo-H) which only cleaves high mannose glycans and analyzed by western blot using anti-GFP antibody. Untreated NCP-GFP and PNCP-GFP run as higher molecular weights (above 43 kDa) (Figure 4B). While PNGase-F treatment causes a shift to above 34 kDa by removal of all N-glycans, only PNCP-GFP is affected by Endo-H treatment, indicating only PNCP-GFP has high M6P / mannose content. To confirm that PNCP-GFP has M6P modifications, a CI-MPR plate binding assay was performed by incubation of the NCP-GFP or PNCP-GFP samples with a CI-MPR coated plate. Figure 4C left panel shows that only PNCP-GFP binds to CI-MPR as indicated by increased GFP fluorescence in the CI-MPR coated plate, compared to NCP-GFP. Both samples show similar total GFP fluorescence signals in conditioned medium applied for the binding assay (Figure 4D). These results demonstrate that the N-glycosylated peptide contains multiple N-glycans when produced in cells, and the N-glycans can be modified to M6P by co-expression of the SI S3 PTase.

[0163] Next, we demonstrated that kifunensine converts complex N-Glycans to high mannose type in the NGP-GFP construct. CHO cells were transfected with the NCP-GFP DNA construct with or without kifunensine present. Conditioned medium from the CHO cells was then treated with Endo H or PNGase F. Western blotting with an anti-GFP antibody illustrates that the kifunensine was able to maintain high mannose type N-glycan structure of NCP-GFP as evidenced by sensitivity to Endo H digestion (Figure 5A). Figure 5B illustrates the total GFP fluorescent signal of NCP-GFP produced without or with kifunensine.

[0164] Next, we determined if a chimeric antibody containing the N-glycosylated peptide could be produced like NCP-GFP. A schematic (Figure 6A) demonstrates a construct used for expression of mouse anti-FLAG M2 light chain under CMV promoter and the heavy chain fused with the N-glycosylated peptide under the EF-la promoter. These constructs were expressed with or without the SI S3 PTase to produce N-glycosylated peptide chimeric antibodies (NCAs). NCA-FLAG (produced without the SI S3 PTase) or PNCA-FLAG (produced with the SI S3 PTase) were purified by protein G beads and treated with Endo-H or PNGase-F and the antibody heavy chain molecular weight examined by western blotting (Figure 6B). NCA-FLAG is unaffected by Endo-H treatment, but there is a size shift afterPNGase-F treatment, indicating the presence of complex type glycans. By contrast, the apparent molecular weight of PNCA-FLAG is altered by both Endo-H and PNGase-F indicating high M6P / mannose type glycans. Figure 6C illustrates a size shift of the NCA- FLAG produced without or with SI S3 phosphotransferase. Thus, co-expression of NCA- FLAG with SI S3 phosphotransferase converts the N-glycans ofNCA-FLAG to phosphorylated high mannose type glycans. Next, the purified NCA-FLAG and PNCA- FLAG were conjugated with Alexa-Fluor 594 (AF594), and CLMPR plate binding assay and affinity chromatograph were performed. AF594 labeled PNCA-FLAG (PNCA-FLAG- AF594) shows a dose dependent increase of fluorescence in the CLMPR plate with high binding affinity (Kd=4.4 nM) (Figures 6D, 7A) compared to no binding for NCA-FLAG. CI- MPR affinity chromatograph also demonstrates more than 90% of PNCA-FLAG binds to CL MPR (Figures 7B, 7C), indicating enhanced M6P content on PNCA-FLAG.

[0165] Next, we demonstrated that kifunensine converts complex N-Glycans to high mannose type in the NCA-FLAG construct. CHO cells were transiently transfected with DNA construct encoding anti-FLAG antibody fused with 6NGP peptide with or without kifunensine. Conditioned medium from the CHO cells was then treated with Endo H or PNGse F. Western blotting with HRP conjugated sheep anti-mouse antibody illustrates that the heavy chain of NCA-FLAG is sensitive to Endo H and PNGase F digestion and that the IgG antibody contains high mannose type N-glycans when expressed with kifunensine as evidenced by Endo H sensitivity (Figure 9).

[0166] To determine if the enhanced M6P content on PNCA-FLAG can enhance its internalization into lysosomes, human fibroblasts were incubated with 20 and 100 nM NCA- FLAG or PNCA-FLAG for 6 hours. Uptake was examined by Western blotting with HRP conjugated anti-mouse secondary antibodies which detect signal for only PNCA-FLAG in the cell lysates (Figure 8A). The uptake was also examined by incubating AF594 conjugated NCA-FLAG or PNCA-FLAG for 6 hours in human fibroblasts and examining AF594 signal in cells by fluorescence microscopy. Images shown in Figure 8B demonstrate strong AF594 signal inside the cells treated with PNCA-FLAG-AF594, but not in NCA-FLAG-AF594. Further this uptake can be blocked by incubating with 2 mM M6P, indicating that the uptake of PNCA-FLAG-AF594 is through the CI-MPR pathway. In addition, lysosomal localization was determined by treating human fibroblast cells with PNCA-FLAG-AF594 (red) and costaining with lysotracker (green), a marker of the lysosome. The yellow signal (Figure 8C) inthe merger panel indicates colocalization of PNCA-FLAG and lysosomes, suggesting transport of the PNCA antibody to the lysosome. These results indicate that the N- glycosylated peptide fusion antibody produced with co-expression of the SI S3 PTase contains high levels of M6P, has high binding affinity to CI-MPR and can be efficiently internalized to the lysosome.

[0167] Example 2: PCNA-FLAG binds to antigen and is internalized to the lysosome through the CI-MPR receptor

[0168] The modification of high M6P on PNCA-FLAG does not alter its antigen recognition function tested by western blot (Figure 10). To examine if the PNCA-FLAG could mediate efficient antigen internalization, we expressed a soluble mCherry containing the FLAG tag in ExpiCHO cells and collected the conditioned media. The CI-MPR plate binding assay was performed by incubating the FLAG-mCherry and NCA-FLAG or PNCA- FLAG mixture in CI-MPR coated plates, schematic shown in Figure 11 A. Binding to CI- MPR was determined by measuring mCherry fluorescence on the plate after washing out unbound FLAG-mCherry. Figure 1 IB shows the mCherry fluorescence signal is dosedependent on the PNCA-FLAG, indicating that the PNCA-FLAG can bind to its antigen FLAG and CI-MPR. No mCherry signal was detected in the NCA-FLAG sample. Next HepG2 cells were incubated overnight with FLAG-mCherry (1 pg / mL) in the presence of 20 nM NCA-FLAG or PNCA-FLAG antibodies and uptake was examined by fluorescence microscopy. mCherry signal was detected only in cells treated with PNCA-FLAG but not with NCA-FLAG (Figure 11C). The uptake of mCherry signal in cells by PNCA-FLAG is inhibited by addition of 2 mM M6P suggesting that CI-MPR mediates the internalization. To better quantify the amount of internalization of extracellular protein, we created a construct expressing FLAG tagged HexM enzyme, a modified version of hexosaminidase with enhanced stability and ability to homodimerize. See Tropak, M.B., et al., Construction of a hybrid beta-hexosaminidase subunit capable of forming stable homodimers that hydrolyze GM2 ganglioside in vivo. Mol Ther Methods Clin Dev, 2016. 3: p. 15057. Our previous data has shown that using an overexpression system generates HexM with a minimum amount of M6P. See Benzie, G., et al., Increased phosphorylation of HexM improves lysosomal uptake and potential for managing GM2 gangliosidoses. BBA Adv, 2022. 2: p. 100032. FLAG- HexM was produced in ExpiCHO cells and incubated with 20 nM NCA-FLAG or PNCA- FLAG for 4 hours for uptake / internalization analysis. The uptake of FLAG-HexM intomultiple cell types was analyzed by measuring HexM enzyme activity in cell lysates. In the five cell lines (HEK293T, Patient Fibroblast, Daoy, HepG2, and MDA-MB-231) tested (Figures 1 ID-11H), the HexM alone could be internalized at a low background level. Coincubation with NCA-FLAG did not further increase the HexM activity in cells, but PNCA- FLAG boosted the FLAG-HexM internalization 2.5 - 9 fold higher than NCA-FLAG or HexM alone group as detected by the cellular HexM activity. The internalization was inhibited by 2 mM M6P in the culture medium. To further demonstrate the importance of the CI-MPR in this process, CI-MPR knockout cells (CLMPR- / -) cells were generated by CRISPR-Cas9 in HEK293T cells (Figure 1 II). Only background enzyme activity was detected in each group (Figure 11 J), demonstrating that the removal of CI-MPR abolished the uptake of HexM mediated by PNCA-FLAG. These data indicate that the phosphorylated N- glycan antibodies (PNCAs) bind to their antigen and internalize the antigen to cells and lysosomes through CI-MPR pathway.

[0169] Example 3: PNCA-TNFa enables extracellular soluble protein TNFa internalization and degradation in the lysosome

[0170] To demonstrate that the PNC A technology disclosed herein can be applied to a therapeutic humanized antibody for lysosomal degradation of soluble cytokines, we constructed a N-glycosylated chimeric anti-TNFa (NCA-TNFa) sequence containing the human IgGl Fc region with the N-glycosylated peptide in the C-terminal of heavy chain of adalimumab sequence. PNCA-TNFa was generated by co-expression of the chimeric antibody sequence with the SI S3 PTase in ExpiCHO cells. In addition, the human adalimumab antibody (Ab-TNFa) was generated in ExpiCHO as a control (Figures 12 A, 12B). N-glycosylation changes were determined by treatment of purified Ab-TNFa or PNCA-TNFa with Endo-H and PNGase-F and analysis by western blotting with HRP conjugated anti-human antibody. Normal human IgGl contains a single N-Glycan in the heavy chain (N297 reference), so a small shift was observed with PNGase-F treatment, however, PNCA-TNF shows large shifts following Endo-H and PNGase-F treatment indicating high M6P / mannose N-glycans in PNCA-TNFa (Figure 12C). Purified Ab-TNFa and PNCA-TNFa were conjugated with AF594 followed by CI-MPR affinity chromatograph. PNCA-TNFa antibody selectively bound to the CI-MPR (85%) (Figures 12D, 12E) with Kd=7 nM (Figure 12F). Internalization in human fibroblasts was efficient (Figure 12G),while no binding or internalization were observed for Ab-TNFa.

[0171] Next PNCA-TNFa mediated TNFa endocytosis and degradation was examined in cells as depicted in a schematic (Figure 12H). To monitor the TNFa internalization, a fusion protein of mCherry linked to the N-terminal of soluble human TNFa sequence (mCherry-TNFa) was expressed and purified from ExpiCHO cells (Figure 121). HepG2 cells were incubated with mCherry-TNFa and Ab-TNFa or PNCA-TNFa with or without 2mM M6P. Internalized mCherry fluorescence signal was detected only in the PNCA-TNFa treated group (Figure 12J). Uptake was also determined by Western blot analysis (Figure 12K) using anti-mCherry (top) or anti-TNFa (middle) antibodies. Anti- GAPDH was a loading control (bottom). Staining revealed detection of mCherry or TNFa only in the PNCA-TNFa treated cells, and this uptake is inhibited by the addition of 2mM M6P. Protease inhibitors inhibit protein degradation in the endosome / lysosome. See Liu, L. and B. Doray, Elevated mRNA expression and defective processing of cathepsin D in HeLa cells lacking the mannose 6-phosphate pathway. FEBS Open Bio, 2021. 11(6): p. 1695-1703. A protease inhibitor cocktail was included in the internalization experiments to study the degradation of TNFa and mCherry in cells. HepG2 cells were treated with mCherry-TNFa alone or with PNCA-TNFa with or without protease inhibitor cocktail. Samples were analyzed by western blot as in Figure 12L. mCherry-TNFa uptake is observed in the presence of PNCA-TNFa, as detected by anti-mCherry antibody, and protease inhibitors restored the detection of full length mCherry-TNFa (top). Similar to mCherry signal, a small amount of TNFa was detected in the PNCA-TNFa treated sample by anti-TNFa antibody (bottom). Further the protease inhibitors not only increase the TNFa level, but also increased the amount of mCherry-TNFa fusion protein level in cells (bottom). Instead of using mCherry- TNFa, recombined human TNFa was also applied to the internalization experiment, and western blot data in Figure 13 suggests it can be internalized with PNCA-TNFa and be degraded in the lysosome. These data show PNCA antibody retain their interaction with its soluble antigen (shown here for TNFa) and the antibody-antigen complex can be internalized to the lysosome for degradation.

[0172] We next sought to determine if the SI S3 PTase can increase M6P level in the antibody itself. The Ab-TNFa antibody sequences were expressed and produced with the SI S3 PTase in ExpiCHO cells, designated as Ab-TNFa-S and purified by Protein G resin (Figure 14 A). Endo-H treatment did shift the heavy chain of antibody produced with theSI S3 PTase to lower molecular weight (Figure 14B). The antibody was further conjugated with AF594 and subjected to CI-MPR affinity chromatography analysis. Only a small amount (19%) of the Ab-TNFa-S antibody binds to the CI-MPR (Figure 14C) which is much lower than PNCA-FLAG (85%, Figure 12E). These data suggest the original N-glycan in the human IgGl is not sufficient for the generation of M6P when combined with SI S3 PTase.

[0173] Example 4: PCNA antibodies enable degradation of membrane receptors

[0174] Programmed cell death 1 ligand 1 (PD-L1) is a biomarker expressed on the cell surface of tumor cells. See Wang, X., et al., PD-L1 expression in human cancers and its association with clinical outcomes. Onco Targets Ther, 2016. 9: p. 5023-39. To demonstrate PNCA mediated lysosome targeted degradation of cell surface transmembrane proteins, anti- PD-L1 antibody - Atezolizumab variable sequence was incorporated into the human IgGl fused to our N-glycosylated peptide design. The phosphorylated N-glycosylated peptide chimeric PD-L1 antibody (PNCA-PD-L1) was produced with co-expression of SI S3 PTase and purified from transient transfected ExpiCHO cells. A human IgGl with Atezolizumab variable domain sequences served as a control (Ab-PD-Ll) (Figure 15A). Glycan analysis by Endo-H and PNGase-F indicates the PNCA-PD-L1 contains high M6P / mannose sugars (Figure 15B). CI-MPR binding analysis using the AF594 conjugated antibody shows 80% of the PNCA-PD-L1 binds to CI-MPR (Figures 15C, 15D). Figure 16A depicts a schematic for PNCA-PD-L1 mediated lysosomal targeting / degradation of PD-L1 in the cell. To examine the PNCA-PD-L1 mediated PD-L1 internalization in cells, antibodies were conjugated with AF594 and applied to Fibroblast cells for 6 hours. AF954 signal was only detected in PNCA- PD-L1-AF594 treated fibroblast and it is colocalized with lysotracker (green) in the cell (Figure 16B). Daoy or A431 cells were treated with 1 to 20 nM Ab-PD-Ll or PNCA-PD-L1 for 24 hours followed by lysis and examined PD-L1 protein levels by western blot. The PNCA-PD-L1 antibody significantly reduces the total PD-L1 level in Daoy (Figure 16C) and A431 (Figure 16D) cells where or the Ab-PD-Ll antibody had no effect.

[0175] The epidermal growth factor receptor (EGFR) regulates epithelial tissue development and homeostasis and is expressed in a variety of human tumors. See Sigismund, S., D. Avanzato, and L. Lanzetti, Emerging functions of the EGFR in cancer. Mol Oncol, 2018. 12(1): p. 3-20. The commercial EGFR - Cetuximab antibody sequence was incorporated into our PNCA platform. Regular Ab-EGFR and PNCA-EGFR were purified(Figures 17A, 17B). Characterization data in Figures 17C-17F showed the PNCA-EGFR contains Endo-H sensitive N-glycan, binds to CI-MPR at high level (82%) with high binding affinity (Kd=10 nM), and was internalized into human fibroblast (Figure 18A) and HepG2 cells (Figure 18B). Ab-EGFR and PNCA-EGFR were applied to Daoy cells for examination of degradation of EGFR. The PNCA-EGFR antibody significantly reduces the total EGFR level in cells where the Ab-EGFR antibody had no effect (Figures 18C, 18D). PNCA-EGFR as low as 5 nM was able to reduce the EGFR level in cells within 24 hours (Figure 19A).

[0176] Example 5: NCAs and SI S3 PTase are capable of being packaged into a viral vector for gene and cell therapy

[0177] To explore the possibility of this technology for cell therapy purposes, we developed an in-vitro co-culture system (Figure 19B). HEK293T cells were seeded in the 6- well plate and transfected with plasmids encoding the regular EGFR antibody sequence or co-transfected with the N-glycosylated peptide chimeric EGFR antibody plasmid with the SI S3 PTase plasmid. Co-culture with HEK293T producing PNCA-EGFR or PNCA-PD-L1 (co-transfection with the SI S3 PTase) significantly reduced the total EGFR (Figures 19C, 19D) or PD-L1 level (Figures 20A, 20B) of Daoy cells in the insert.

[0178] To better express NCAs with SI S3 in a viral or non-viral vector (AAV, Lentiviral etc), small antibody versions are planned to be made as shown in Figure 21. The single chain antibodies will be expressed together with SI S3 PTase in mammalian cells (e.g., CHO cells) to test their expression. Those candidates showing good expression and activity will be incorporated into an expression construct such as the bicistronic constructs described in WO 2001 / 003442, or the dual-promoter constructs disclosed in WO 2023 / 150051) for viral or non-viral treatments.

[0179] Example 6: Phosphorylated PNC A Chimeric PD-L1 Antibody Has the Same Blockage for PD-1 and PD-L1 Binding as Atezolizumab

[0180] A PD-1 and PD-L1 blocking assay was performed by co-culturing CHO-K1 cells with stable PD-L1 overexpression (CHOK1-PD-L1) and Jurkat-PDl-NFAT cells in a 96-well plate. Briefly, CHOK1-PD-L1 cells were seeded and incubated overnight (16~20h) with PNCA-PD-L1 or Atezolizumab at concentrations of 300, 75, 18.75, 4.688, 1.172, 0.293, 0.0733, 0.0183, 0.0045, and 0 nM. Jurkat-PDl-NFAT cells were collected and added toCH0K1-PD-L1 cells and incubated at 37°C for 6 hrs. After incubation, 80 uL of ready-Glo luciferase assay buffer was added. After 5-30 minutes, read the plate by a microplate reader. As shown in Figure 23, PNCA-PD-L1 has the same binding characteristics as the approved anti-PD-11 monoclonal antibody Atezolizumab.

[0181] Example 7: Phosphorylated PNCA-PD-L1 Antibody Induces Release of Proinjlammatory Cytokines by T-Cells

[0182] A mixed lymphocyte reaction (MLR) assay was performed to assess the immune response of human T cells in vitro. Dendritic cells differentiated from CD14+ monocytes were cocultured with human T cells isolated from human peripheral blood mononuclear cells with various concentrations of the PNCA-PD-L1 or Atezolizumab for 72 hrs. Supernatant was collected for detection of IL-2 and IFN-y by ELISA analysis. As shown in Figure 24, the amount of IL-2 (left) and IFN-y (right) releases using the PNCA-PD-L1 chimeric antibody as comparable to or better than Atezolizumab.

[0183] Example 8: Phosphorylated Anti-PD-Ll Antibody Induces Activation of Immune Cells for Killing Tumor Cells

[0184] A T cell mediated tumor cell killing assay was used to investigate the effect of PNCA-PD-L1 to activate immune cells. RKO cancer cells were co-cultured with CD3+ T cells isolated from human peripheral blood mononuclear cells for 24 hrs with various concentrations of the PNCA-PD-L1 chimeric antibody or Atezolizumab. After three-times washing with PBS, cells were fixed by 4% paraformaldehyde and stained with crystal violet. Images were taken by microscopy. 3Three representative images were analyzed by Imaged software to calculate the density of crystal violet signals. The results are shown in Figure 25: the representative images in Figure 25 A show reduced crystal violet signals in PNCA-PD-L1 treated RKO cells co-cultured with T cells. No changes were observed in the control RKO cells alone. Figure 25B shows that PNCA-PD-L1 introduces enhanced tumor cell killing at higher concentrations compared to the commercial product Atezolizumab. The cell killing was calculated by crystal violet signal in Imaged. N=3.

[0185] Example 9: Pharmacokinetics Study Comparing the Rate of Internalization of Human TNFa compared to Phosphorylated PNC A- TNFa

[0186] An in vivo pharmacokinetics study was performed in wild-type mice to compare the half-life of a human TNFa antibody with a hTNFa-NGP peptide containing a 1 mannose 6-phosphate site (PNCA- TNFa-lNGP), and an hTNFa-NGP peptide containing 6 mannose 6-phosphate sites (PNCA- TNFa-6NGP). The study parameters are shown in Figure 26. The results, shown in Figure 27, depict ELISA data showing that the PNCA- TNFa- lNGP peptide design has a similar half-life to regular human TNFa in mouse, but the 6NGP PNCA- TNFa-6NGP peptide design has a shorter serum half-life, which is indicative of a higher rate of internalization.

[0187] Example 10: Initial Phosphorylated Anti-PD-Ll Antibody Enables In Vivo Tumor Inhibition With Minimal Effect on Body Weight

[0188] An in vivo efficacy study was conducted to assess MC38 tumor inhibition in mice by intraperitoneal injection of human PD-L1 antibody conjugated with an NGP peptide containing 6 mannose 6-phosphate sites (PNCA-PD-L1-6NGP) at dosage 5 mg / kg, twice a week treatment. The study design is shown in Figure 28.

[0189] Tumor size was measured every other day in the animals from the in vivo efficacy study. The results are shown in Figures 29A and 29B. Figure 29A shows that PNCA-PD-L1-6NGP slows down the tumor growth compared to vehicle group. N=8. Figure 29B shows that animal body weight of the two groups does not change significantly.

[0190] For ease of reference, the following Table 1 lists all nucleic acid and amino acid sequences disclosed herein.

Claims

CLAIMSWhat is claimed is:

1. A composition comprising a recombinant bifunctional protein or polypeptide comprising(a) an N-glycosylated peptide comprising at least one N-glycan group; and(b) a protein of interest, or an antibody or antibody fragment capable of binding to a protein of interest; wherein the composition is capable of binding to a cell surface receptor for lysosome targeting.

2. The composition of claim 1 wherein the N-glycan is high mannose type with a terminal mannose group.

3. The composition of claim 1 wherein the N-glycan is phosphorylated N-glycan with a terminal mannose-6-phosphate group.

4. The composition of claim 1, wherein the antibody or antibody fragment is a chimeric antibody or antibody fragment.

5. The composition of claim 1, wherein the N-glycosylated peptide comprises one or more N-glycosylation sites, wherein the one or more sites comprise Asparagine-X-Serine or Threonine modifications, wherein X comprises any amino acid except proline.

6. The composition of claim 5, wherein X comprises one or more of Serine (S), Alanine (A), Histidine (H), Glycine (G), Asparagine (N), or Threonine (T).

7. The composition of claim 5 comprising a peptide sequence selected from NXTGGSNXTGGGNXTGGSNXTMMSGNS (SEQ ID NO: 1), NXTGNXTSNXTGNXTSNXTGNXTMMST (SEQ ID NO: 2), NXTNXTNXTNXTNXTNXTNXTNXTMMS (SEQ ID NO: 3), NXTSNXTTNXTTNXTSNXTTNNXSMMS (SEQ ID NO: 4), and NNXSNNXTNNXTNNXSNNXTNNXSMMS (SEQ ID NO: 5), wherein X comprises any amino acid except proline.

8. The composition of claim 7 wherein X comprises one or more of Serine (S), Alanine (A), Histidine (H), Glycine (G), Asparagine (N), or Threonine (T).

9. The composition of any of claims 1-8, wherein the peptide sequence is selected from NSTGGSNATGGGNSTGGSNATMMSGNS (SEQ ID NO: 13), NATGNSTSNSTGNSTSNSTGNATMMST (SEQ ID NO: 10), NSTNATNHTNGTNNTNTTNGTNSTMMS (SEQ ID NO: 14), and NNTSNNTTNNTTNNTSNNTTNNTSMMS (SEQ ID NO: 15).

10. A nucleic acid sequence encoding a recombinant bifunctional protein comprising(a) an N-glycosylated peptide comprising at least one N-glycosylation site; and(b) a protein of interest, or an antibody or antibody fragment capable of binding to a protein of interest.

11. The composition of claim 10, wherein the nucleic acid sequence of the N- glycosylated peptide is selected from:AACAGCACCGGCGGCAGCAACGCCACCGGCGGCGGCAACAGCACCGGCGGCAG CAACGCCACCATGATGAGCGGCAACAGC (SEQ ID NO: 16), AACGCCACCGGCAACAGCACCAGCAACAGCACCGGCAACAGCACCAGCAACAG CACCGGCAACGCCACCATGATGAGCACC (SEQ ID NO: 17), AACAGCACCAACGCCACCAACCACACCAACGGCACCAACAACACCAACACCACCAACGGCACCAACAGCACCATGATGAGC (SEQ ID NO: 18), and AACAACACCAGCAACAACACCACCAACAACACCACCAACAACACCAGCAACAA CACCACCAACAACACCAGCATGATGAGC (SEQ ID NO: 19).

12. A composition comprising the nucleic acid sequence of any of claims 10-11 wherein the nucleic acid sequence is fused to a nucleic acid sequence encoding a protein of interest, or an antibody or antibody fragment capable of binding to a protein of interest.

13. The composition of claim 1, wherein the protein of interest is a lysosomal protein or a non-lysosomal protein.

14. The composition of claim 13, wherein the protein is a lysosomal protein selected from P-glucocerebrosidase (GBA), GalA, Cathepsin D (CathD), Niemann-Pick disease type C2 (NPC2), P-hexosaminidase (HEXB), a-Galactosidase (GLA), P-Mannosidase (MANBA), alpha-L-iduronidase, iduronate sulfatase, arylsulfatase B, acid a-glucosidase (GAA), and lysosomal acid a-mannosidase (LAMAN).

15. The composition of claim 13, wherein the protein of interest is a non-lysosomal protein.

16. The composition of claim 15, wherein the non-lysosomal protein is a cytokine, a membrane receptor, or immune checkpoint molecule.

17. The composition of claim 16, wherein the non-lysosomal protein is selected from TNFa, IL-12, growth IGF, EGF, insulin, and adrenaline.

18. The composition of claim 1 wherein the N-glycosylated peptide is fused to a chimeric antibody or antibody fragment that recognizes a protein of interest.

19. A method for producing a recombinant bifunctional protein of claim 1, comprising:(a) providing a nucleic acid sequence comprising a first portion encoding an N- glycosylated peptide and a second portion encoding a protein of interest, or a chimeric antibody or antibody fragment that recognizes a protein of interest;(b) contacting the nucleic acid sequence with a host cell under conditions sufficient to induce expression of the recombinant bifunctional protein.

20. The method of claim 19, further comprising expressing the recombinant bifunctional protein in the presence of a mannosidase inhibitor.

21. The method of claim 20, wherein the mannosidase inhibitor is selected from swainsonine, kifunensine, 1-deoxymannojirimycin hydrochloride, (1S,2S,3R)-1,2,3- trihydroxy-4-cyclopropene 2,3 -cyclohexyl ketal, or N-butyldeoxymannojirimycin HC1.

22. The method of claim 19, further comprising co- expressing the recombinant bifunctional protein with a GlcNAc-1 -phosphotransferase.

23. The method of claim 22, wherein the GlcNAc-1 -phosphotransferase comprises SI S3 phosphotransferase.

24. The method of claim 22, wherein the GlcNAc-1 -phosphotransferase comprises a truncated form comprising core catalytic domains of GlcNAc-1 -phosphotransferase25. The method of claim 22, wherein the nucleic acid sequence of step (a) comprises a third portion encoding a GlcNAc-1 -phosphotransferase, an SI S3 phosphotransferase or a truncated form comprising core catalytic domains of GlcNAc-1 -phosphotransferase.

26. The method of claim 22, comprising the additional step of providing a second nucleic acid sequence encoding a GlcNAc-1 -phosphotransferase, an SI S3 phosphotransferase or a truncated form comprising core catalytic domains of GlcNAc-1 -phosphotransferase.

27. The method of claim 19, wherein the host cell is a mammalian cell.

28. The method of claim 27, wherein the mammalian cell is a CHO, HEK293 or sf9 cell.

29. The method of claim 19 wherein the protein of interest is a lysosomal protein or polypeptide fragment thereof.

30. The method of claim 29, wherein the protein is a lysosomal protein selected from P- glucocerebrosidase (GBA), GalA, Cathepsin D (CathD), Niemann-Pick disease type C2 (NPC2), P-hexosaminidase (HEXB), a-Galactosidase (GLA), N-acetyl-a-glucosaminidase (NAGLU), P-Mannosidase (MANBA), alpha-L-iduronidase, iduronate sulfatase, arylsulfatase B, acid a-glucosidase (GAA), and lysosomal acid a-mannosidase (LAMAN).

31. The method of claim 19, wherein the protein of interest is a non-lysosomal protein or polypeptide fragment thereof.

32. The method of claim 31, wherein the non-lysosomal protein is a cytokine, a membrane receptor, or immune checkpoint molecule.

33. The method of claim 32, wherein the non-lysosomal protein is selected from TNFa, IL- 12, growth IGF, PD-L1, PD-1, EGF, EGFR, insulin, and adrenaline.

34. The method of claim 19 wherein the second portion comprises a nucleic acid sequence encoding a chimeric antibody or antibody fragment.

35. The method of claim 34, wherein the antibody or antibody fragment binds to a membrane receptor or soluble protein.

36. The method of claim 35, wherein the soluble protein comprises TNFa.

37. The method of claim 35, wherein the membrane receptor comprises EGFR or PD-L1.

38. A method for lysosomal degradation of a protein of interest comprising introducing to a cell the peptide sequence of any of claims 13-18.

39. A composition comprising an expression vector comprising a nucleic acid sequence encoding the peptide sequence of any of claims 13-18.

40. The composition of claim 39, wherein the expression vector further comprises a nucleic acid sequence encoding GlcNAc phosphotransferase, SI S3 phosphotransferase or a truncated form comprising core catalytic domains of GlcNAc- 1 -phosphotransferase.

41. The composition of claim 39 further comprising a second expression vector comprising a nucleic acid sequence encoding GlcNAc phosphotransferase, SI S3 phosphotransferase or a truncated form comprising core catalytic domains of GlcNAc-1- phosphotransferase.

42. The composition of any of claims 39-41, wherein the expression vector comprises a viral capsid.

43. The composition of claim 42, wherein the viral capsid comprises an AAV capsid.

44. The composition of any of claims 39-43, further comprising a promoter.

45. The composition of claim 44, wherein promoter is a CMV promoter, or an EF-la promoter.

46. A pharmaceutical composition comprising: the bifunctional recombinant protein or polypeptide of any one of claims 1 to 18; and a pharmaceutically acceptable carrier.

47. The pharmaceutical composition of claim 46, wherein the composition is formulated for parenteral administration.

48. A method comprising administering the pharmaceutical composition of claim 46 or 47 to a subject in need thereof.

49. A method of treating a subject suffering from a proliferative disease, a lysosomal storage disease, an infection, an immune disease, or an autoimmune disease comprising administering to the subject an effective amount of the pharmaceutical composition of claim 46 or 47.

50. A kit comprising the bifunctional recombinant protein or polypeptide of any one of claims 1 to 18.

51. The kit of claim 50, comprising a pharmaceutical composition comprising the bifunctional recombinant protein or polypeptide of any one of claims 1 to 18; and a pharmaceutically acceptable carrier.

52. The kit of claim 50 or 51, further comprising instructions for administering the bifunctional recombinant protein or polypeptide or pharmaceutical composition to a subject in need thereof.

Images