Engineered TGF-beta monomers and their use for TGF-beta signaling

Abstract

To provide a recombinant transforming growth factor (TGF)-β monomer modified to inhibit dimerization and block TGF-β signaling. [Solution] Recombinant TGF-β monomers lack the ability to bind and recruit TGF-β type I receptors (TβM) but retain the ability to bind to high-affinity TGF-β type II receptors (TβMI), and in some cases contain mutations that increase their affinity for TβMI. Nucleic acid molecules and vectors encoding the recombinant TGF-β monomers are also described. Isolated cells, such as T cells, can be reprogrammed with the TGF-β monomer-encoding nucleic acids or vectors to secrete the monomers. Uses of recombinant TGF-β monomers and / or cells producing recombinant TGF-β monomers to inhibit TGF-β signaling, for example, to treat disorders associated with aberrant TGF-β signaling, are also described.

Description

[Technical field]

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS This application claims the benefit of U.S. Provisional Application No. 62 / 423,920, filed November 18, 2016, which is incorporated herein by reference in its entirety.

[0002] Field The present disclosure relates to recombinant transforming growth factor (TGF)-β monomers modified to inhibit dimerization while retaining the ability to bind to the high affinity TGF-β type II receptor (TβRII). The present disclosure further relates to uses of the recombinant TGF-β monomers to inhibit TGF-β signaling.

[0003] Government support acknowledged This invention was made with Government support under Grant No. GM058670 awarded by the National Institutes of Health. The Government has certain rights in this invention. [Background technology]

[0004] background TGF-β is a multifunctional cytokine with diverse biological effects on cellular processes including cell proliferation, migration, differentiation, and apoptosis. The three mammalian TGF-β isoforms, TGF-β1, -β2, and -β3, exert their functions through cell surface receptor complexes composed of type I (TβRI) and type II (TβRII) serine / threonine kinase receptors. Receptor activation induces both SMAD proteins and other downstream targets including Ras, RhoA, TAK1, MEKK1, PI3K, and PP2A to result in a full TGF-β response (Roberts and Wakefield, Proc Natl Acad Sci USA 100:8621-8623, 2003; Derynck and Zhang, Nature 425:577-584, 2003; Massague, Cell 134:215-230, 2008).

[0005] TGF-β proteins are known to promote the progression of fibrotic disorders and certain types of cancer. In the context of fibrotic disorders, TGF-β potently stimulates the expression of extracellular matrix (ECM) proteins. Dysregulation of ECM remodeling can lead to pathological fibrosis. The role of TGF-β in cancer is multifaceted. TGF-β isoforms, TGF-β1, -β2 and -β3, are also known to suppress host immune surveillance, stimulate epithelial-mesenchymal transition, and enhance cancer progression and metastasis. [Prior art documents] [Non-patent literature]

[0006] [Non-Patent Document 1] Roberts and Wakefield, Proc Natl Acad Sci USA 100:8621-8623, 2003. [Non-Patent Document 2] Derynck and Zhang, Nature 425:577-584, 2003. [Non-Patent Document 3] Massague, Cell vol. 134: 215-230, 2008 Summary of the Invention [Means for solving the problem]

[0007] Abstract Described herein are engineered TGF-β monomers that can block TGF-β signaling. The engineered monomers inhibit TGF-β signaling by preventing TGF-β dimerization and recruitment of TβRI.

[0008] Provided herein is a recombinant TGF-β monomer comprising a cysteine ​​to serine substitution at amino acid residue 77; a deletion of amino acid residues 52-71; and at least one amino acid substitution that increases the net charge of the monomer. In some embodiments, the TGF-β monomer further comprises at least one amino acid substitution that increases the affinity of the TGF-β monomer for the TGF-β type II receptor (TβRII). The TGF-β monomer can be, for example, a TGF-β2, TGF-β1, or TGF-β3 monomer, such as a human, rat, mouse, or other mammalian TGF-β2, TGF-β1, or TGF-β3 monomer.

[0009] Fusion proteins comprising a TGF-β monomer and a heterologous protein are also provided. Further provided are compositions comprising a recombinant TGF-β monomer or fusion protein disclosed herein and a pharma- ceutically acceptable carrier, diluent, or excipient.

[0010] Also provided is a method of inhibiting TGF-β signaling in a cell by contacting the cell with a recombinant TGF-β monomer, fusion protein, or composition disclosed herein. In some embodiments, the method is an in vitro method. In other embodiments, the method is an in vivo method comprising administering a recombinant TGF-β monomer, fusion protein, or composition to a subject having a disease or disorder associated with aberrant TGF-β signaling.

[0011] Also provided are nucleic acid molecules and vectors encoding the recombinant TGF-β monomers disclosed herein. Further provided are isolated cells, such as isolated T lymphocytes, that contain a recombinant TGF-β monomer-encoding nucleic acid molecule or vector.

[0012] Further provided are methods of treating a disease or disorder associated with aberrant TGF-β signaling in a subject by administering to the subject an isolated cell (such as a T cell) containing the disclosed nucleic acid or vector.

[0013] The foregoing and other objects, features and advantages of the present invention will become more apparent from the following detailed description which proceeds with reference to the accompanying drawings. In certain embodiments, for example, the following are provided: (Item 1) A recombinant transforming growth factor (TGF)-β monomer, (i) a cysteine ​​to serine substitution at an amino acid residue corresponding to residue 77 of SEQ ID NO:2; (ii) a deletion of the α3 helix corresponding to amino acid residues 52-71 of SEQ ID NO:2; and (iii) at least one amino acid substitution, relative to wild-type TFG-β monomer, that increases the net charge of the recombinant TFG-β monomer; A recombinant TGF-β monomer comprising: (Item 2) 2. The recombinant TGF-β monomer of item 1, further comprising at least one amino acid substitution compared to wild-type TGF-β2 monomer that increases the affinity of the recombinant TGF-β monomer for the TGF-β type II receptor (TβRII). (Item 3) 3. The recombinant TGF-β monomer according to item 1 or 2, which is a human TGF-β2 monomer. (Item 4) The at least one amino acid substitution that increases the net charge of the monomer is A leucine to arginine substitution at an amino acid residue corresponding to residue 51 of SEQ ID NO:2; an alanine to lysine substitution at an amino acid residue corresponding to residue 73 of SEQ ID NO:2; or Both a leucine to arginine substitution at an amino acid residue corresponding to residue 51 and an alanine to lysine substitution at an amino acid residue corresponding to residue 73 of SEQ ID NO:2. 4. The recombinant human TGF-β2 monomer according to item 3, comprising: (Item 5) 5. The recombinant human TGF-β2 monomer of claim 3 or 4, wherein the at least one amino acid substitution that increases the affinity of the monomer for TβRII comprises a substitution at an amino acid residue corresponding to residue 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98 or 99 of SEQ ID NO:2, or any combination of two or more thereof. (Item 6) 6. The recombinant human TGF-β2 monomer of item 5, wherein the at least one amino acid substitution that increases the affinity of the monomer for TβRII comprises at least one substitution at residue 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36 or 37, and at least one substitution at residue 89, 90, 91, 92, 93, 94, 95, 96, 97, 98 or 99. (Item 7) 7. The recombinant human TGF-β2 monomer of item 6, wherein the at least one amino acid substitution that increases the affinity of the monomer for TβRII comprises lysine to arginine at residue 25, arginine to lysine at residue 26, leucine to valine at residue 89, isoleucine to valine at residue 92, asparagine to arginine at residue 94, threonine to lysine at residue 95, isoleucine to valine at residue 98, or any combination of two or more thereof. (Item 8) 8. The recombinant human TGF-β2 monomer of item 7, comprising lysine to arginine at residue 25, arginine to lysine at residue 26, leucine to valine at residue 89, isoleucine to valine at residue 92, asparagine to arginine at residue 94, threonine to lysine at residue 95, and isoleucine to valine at residue 98. (Item 9) 9. The recombinant human TGF-β2 monomer of any one of items 3 to 8, comprising the amino acid sequence of SEQ ID NO: 8 or SEQ ID NO: 10. (Item 10) 3. The recombinant TGF-β monomer according to item 1 or 2, which is a human TGF-β1 monomer. (Item 11) The at least one amino acid substitution that increases the net charge of the monomer is an isoleucine to arginine substitution at an amino acid residue corresponding to residue 51 of SEQ ID NO:1; an alanine to lysine substitution at an amino acid residue corresponding to residue 74 of SEQ ID NO:1; an alanine to serine substitution at an amino acid residue corresponding to residue 75 of SEQ ID NO:1; or an isoleucine to arginine substitution at the amino acid residue corresponding to residue 51, an alanine to lysine substitution at the amino acid residue corresponding to residue 74, and an alanine to serine substitution at the amino acid residue corresponding to residue 75 of SEQ ID NO:1 11. The recombinant human TGF-β1 monomer according to item 10, comprising: (Item 12) 12. The recombinant human TGF-β1 monomer according to item 10 or 11, comprising the amino acid sequence of SEQ ID NO:7. (Item 13) 3. The recombinant TGF-β monomer according to item 1 or 2, which is a human TGF-β3 monomer. (Item 14) The at least one amino acid substitution that increases the net charge of the monomer is A leucine to glutamic acid substitution at an amino acid residue corresponding to residue 51 of SEQ ID NO:3; an alanine to glutamic acid substitution at an amino acid residue corresponding to residue 72 of SEQ ID NO:3; an alanine to aspartic acid substitution at an amino acid residue corresponding to residue 74 of SEQ ID NO:3; or a leucine to glutamic acid substitution at the amino acid residue corresponding to residue 51, an alanine to glutamic acid substitution at the amino acid residue corresponding to residue 72, and an alanine to aspartic acid substitution at the amino acid residue corresponding to residue 74 of SEQ ID NO:3; 14. The recombinant human TGF-β3 monomer according to item 13, comprising: (Item 15) 15. The recombinant human TGF-β3 monomer of item 13 or 14, comprising the amino acid sequence of SEQ ID NO:9. (Item 16) 16. The recombinant TGF-β monomer of any one of items 1 to 15, which is PEGylated. (Item 17) 16. The recombinant TGF-β monomer of any one of items 1 to 15, which is glycosylated or hyperglycosylated. (Item 18) 18. A fusion protein comprising the recombinant TGF-β monomer of any one of items 1 to 17 and a heterologous protein. (Item 19) 19. The fusion protein of item 18, wherein the heterologous protein comprises a protein tag. (Item 20) 19. The fusion protein of item 18, wherein the heterologous protein comprises an Fc domain. (Item 21) 19. The fusion protein of item 18, wherein the heterologous protein comprises albumin or an albumin binding polypeptide. (Item 22) 19. The fusion protein of item 18, wherein the heterologous protein comprises an antibody, an antigen-binding fragment thereof, or a targeting moiety. (Item 23) 19. The fusion protein according to item 18, which is a single-chain polypeptide. (Item 24) 19. The fusion protein according to item 18, which forms a dimeric polypeptide. (Item 25) 19. The fusion protein of item 18, which is a heterodimer. (Item 26) 19. The fusion protein according to item 18, which is a multimer. (Item 27) 27. The recombinant TGF-β monomer of any one of items 1 to 17, or the fusion protein of any one of items 18 to 26, comprising a radiotherapeutic agent, a cytotoxic agent for chemotherapy, or a drug. (Item 28) 27. The recombinant TGF-β monomer of any one of items 1 to 17, or the fusion protein of any one of items 18 to 26, comprising an imaging agent, a fluorescent dye, or a fluorescent protein tag. (Item 29) 27. A composition comprising a recombinant TGF-β monomer according to any one of items 1 to 17, or a fusion protein according to any one of items 18 to 26, and a pharma- ceutically acceptable carrier, diluent, or excipient. (Item 30) 30. A method for inhibiting TGF-β signaling in a cell, comprising contacting the cell with a recombinant TGF-β monomer according to any one of items 1 to 17, a fusion protein according to any one of items 18 to 26, or a composition according to item 29. (Item 31) 31. The method according to item 30, which is an in vitro method. (Item 32) 31. The method of claim 30, which is an in vivo method comprising administering the recombinant TGF-β monomer, fusion protein or composition to a subject having a disease or disorder associated with aberrant TGF-β signaling. (Item 33) 30. A method for treating a disease or disorder associated with abnormal TGF-β signaling in a subject, comprising administering to the subject a recombinant TGF-β monomer according to any one of items 1 to 17, a fusion protein according to any one of items 18 to 26, or a composition according to item 29. (Item 34) The method of claim 32 or 33, wherein the disease or disorder associated with aberrant TGF-β signaling is a fibrotic disorder. (Item 35) The method of claim 32 or 33, wherein the disease or disorder associated with aberrant TGF-β signaling is breast cancer, brain cancer, pancreatic cancer, prostate cancer or skin cancer. (Item 36) The method of claim 32 or 33, wherein the disease or disorder associated with aberrant TGF-β signaling is an eye disease. (Item 37) The method of claim 32 or 33, wherein the disease or disorder associated with aberrant TGF-β signaling is an inherited disorder of connective tissue. (Item 38) 18. An isolated nucleic acid molecule encoding a recombinant TGF-β monomer according to any one of items 1 to 17. (Item 39) 39. The nucleic acid molecule of item 38, operably linked to a promoter. (Item 40) A vector comprising the nucleic acid molecule of item 38 or item 39. (Item 41) 41. An isolated cell comprising the nucleic acid molecule of item 38 or item 39, or the vector of item 40. (Item 42) 42. The isolated cell of item 41, which is a T lymphocyte. (Item 43) 43. A method for treating a disease or disorder associated with abnormal TGF-β signaling in a subject, comprising administering to the subject the isolated cell of item 41 or item 42. (Item 44) 44. The method of claim 43, wherein the disease or disorder associated with abnormal TGF-β signaling is a fibrotic disorder. (Item 45) 44. The method of claim 43, wherein the disease or disorder associated with abnormal TGF-β signaling is breast cancer, brain cancer, pancreatic cancer, prostate cancer or skin cancer. (Item 46) 44. The method of claim 43, wherein the disease or disorder associated with abnormal TGF-β signaling is an eye disease. (Item 47) 44. The method of item 43, wherein the disease or disorder associated with abnormal TGF-β signaling is a genetic disorder of connective tissue. [Brief description of the drawings]

[0014] [Figure 1ABC]Figures 1A-1F: Structure of the TGF-β signaling complex and sequence of an engineered TGF-β variant lacking the interfacial α-helix. (Figure 1A) Schematic showing the TGF-β signaling complex formed between human TGF-β3 homodimer and the extracellular ligand-binding domains of the human TGF-β type I and type II receptors (TβRI and TβRII (PDB 2PJY)) (Groppe et al., Mol Cell 29:157-168, 2008). Disulfide bonds, including a single interchain disulfide, that link the TGF-β monomers are shown. TGF-β monomers are shown as curling left hands, with a heel formed by a 3-1 / 2 turn α-helix (α3) and four fingers formed by β strands extending from a cystine knot that stabilizes each monomer. (Figure 1B) A close-up showing the packing interactions formed by hydrophobic residues protruding from the heel α-helix of one TGF-β3 monomer with hydrophobic residues from the palm region of the opposing TGF-β3 monomer. (Figure 1C) A close-up showing the ionic, hydrogen-bonding, and hydrophobic interactions that stabilize TβRI at the composite interface formed by both TGF-β3 and TβRII monomers. (Figure 1D) Sequence alignment of TGF-β1, -β2, and -β3 with monomeric variants in which Cys77, which normally forms the interchain disulfide bond, has been replaced by serine (mTGF-β2 and mTGF-β3) or mini-monomeric variants in which Cys77 has been replaced by serine, residues 52-71 have been deleted, and two or three additional residues have been replaced (mmTGF-β1, mmTGF-β2, and mmTGF-β3). The calculated net charge of the corresponding monomer at pH 7.0 is shown on the right. (FIG. 1E) Sequence alignment of the TβRII binding region of TGF-β1, -β3, -β2, mmTGF-β2 and mmTGF-β2-7M. Residues in the TβRII binding interface are indicated by shading.Residues substituted in mmTGF-β2-7M compared to mmTGF-β2 are indicated by boxes and include K25R, I92V, and N94R, which have previously been shown to be necessary and sufficient for TβRII binding with high affinity (Baardsnes et al., Biochemistry 48:2146-2155, 2009; De Crescenzo et al., J Mol Biol 355:47-62, 2006). (Figure 1F) Interface between TGF-β3 and TβRII. [Figure 1DEF]Figures 1A-1F: Structure of the TGF-β signaling complex and sequence of an engineered TGF-β variant lacking the interfacial α-helix. (Figure 1A) Schematic showing the TGF-β signaling complex formed between human TGF-β3 homodimer and the extracellular ligand-binding domains of the human TGF-β type I and type II receptors (TβRI and TβRII (PDB 2PJY)) (Groppe et al., Mol Cell 29:157-168, 2008). Disulfide bonds, including a single interchain disulfide, that link the TGF-β monomers are shown. TGF-β monomers are shown as curling left hands, with a heel formed by a 3-1 / 2 turn α-helix (α3) and four fingers formed by β strands extending from a cystine knot that stabilizes each monomer. (Figure 1B) A close-up showing the packing interactions formed by hydrophobic residues protruding from the heel α-helix of one TGF-β3 monomer with hydrophobic residues from the palm region of the opposing TGF-β3 monomer. (Figure 1C) A close-up showing the ionic, hydrogen-bonding, and hydrophobic interactions that stabilize TβRI at the composite interface formed by both TGF-β3 and TβRII monomers. (Figure 1D) Sequence alignment of TGF-β1, -β2, and -β3 with monomeric variants in which Cys77, which normally forms the interchain disulfide bond, has been replaced by serine (mTGF-β2 and mTGF-β3) or mini-monomeric variants in which Cys77 has been replaced by serine, residues 52-71 have been deleted, and two or three additional residues have been replaced (mmTGF-β1, mmTGF-β2, and mmTGF-β3). The calculated net charge of the corresponding monomer at pH 7.0 is shown on the right. (FIG. 1E) Sequence alignment of the TβRII binding region of TGF-β1, -β3, -β2, mmTGF-β2 and mmTGF-β2-7M. Residues in the TβRII binding interface are indicated by shading.Residues substituted in mmTGF-β2-7M compared to mmTGF-β2 are indicated by boxes and include K25R, I92V, and N94R, which have previously been shown to be necessary and sufficient for TβRII binding with high affinity (Baardsnes et al., Biochemistry 48:2146-2155, 2009; De Crescenzo et al., J Mol Biol 355:47-62, 2006). (Figure 1F) Interface between TGF-β3 and TβRII. [Figure 2A] Figures 2A-2D: Structure of mmTGF-β2. (Figure 2A) Assigned 1H-15N HSQC spectrum of mmTGF-β2 recorded at 800 MHz in 10 mM sodium phosphate, 10 mM CHAPS, 5% 2H2O, pH 4.70, 37 °C. Assigned backbone amide signals are shown by their residue numbers and one-letter amino acid codes. (Figure 2B) Crystal structure of mmTGF-β2 at 1.8 Å and superimposition of one monomer from the crystal structure of TGF-β2 (PDB 2TGI) at 1.8 Å. Major structural features are shown, along with the engineered loop of mmTGF-β2 substituting for the palm (α3) helix of TGF-β2. (Figure 2C) Superimposition of the two mmTGF-β2 chains (chains A and B) from the crystallographic asymmetric unit. Other details as in Figure 2B. (FIG. 2D) Overlap of mmTGF-β2 and TGF-β2 in FIG. 2B, except that the aligned portions are limited to residues 18-45 and 61-87 of fingers 1 / 2 and 3 / 4, respectively. [Figure 2BCD]Figures 2A-2D: Structure of mmTGF-β2. (Figure 2A) Assigned 1H-15N HSQC spectrum of mmTGF-β2 recorded at 800 MHz in 10 mM sodium phosphate, 10 mM CHAPS, 5% 2H2O, pH 4.70, 37 °C. Assigned backbone amide signals are shown by their residue numbers and one-letter amino acid codes. (Figure 2B) Crystal structure of mmTGF-β2 at 1.8 Å and superimposition of one monomer from the crystal structure of TGF-β2 (PDB 2TGI) at 1.8 Å. Major structural features are shown, along with the engineered loop of mmTGF-β2 substituting for the palm (α3) helix of TGF-β2. (Figure 2C) Superimposition of the two mmTGF-β2 chains (chains A and B) from the crystallographic asymmetric unit. Other details as in Figure 2B. (FIG. 2D) Overlap of mmTGF-β2 and TGF-β2 in FIG. 2B, except that the aligned portions are limited to residues 18-45 and 61-87 of fingers 1 / 2 and 3 / 4, respectively. [Fig. 3ABCD]3A-3H: Binding characteristics of mmTGF-β2 and mmTGF-β2-7M. (FIGS. 3A and 3B) Surface plasmon resonance (SPR) sensorgrams upon injection of a two-fold dilution series of TβRII from 0.047 to 12 μM to immobilized TGF-β2 (FIG. 3A) or mmTGF-β2 (FIG. 3B). (Figures 3C-3H) SPR sensorgrams upon injection of a 2-fold dilution series of 0.012-3 μM TβRII (Figures 3C and 3D), 0.008-1.024 μM TβRII (Figures 3E and 3F), or 0.008-1.024 μM TβRII (Figures 3G and 3H) in the presence of 2 μM TβRII in running buffer and injected samples, against immobilized avi-TGF-β3 (Figures 3C, 3E, and 3G) or avi-mmTGF-β2-7M (Figures 3D, 3F, and 3H). The sensorgrams shown in Figures 3C, 3D, and 3G were fitted to a 1:1 binding model, and raw data and fitted curves are shown. TGF-β2 and mmTGF-β2 were immobilized by direct carbodiimide-based amine coupling to the sensor surface, and avi-TGF-β3 or avi-mmTGF-β2-7M were immobilized by enzymatically biotinylated protein capture onto the surface of a streptavidin-coated sensor chip at high (approximately 8000 response units (RU)) density. [Fig. 3EFGH]3A-3H: Binding characteristics of mmTGF-β2 and mmTGF-β2-7M. (FIGS. 3A and 3B) Surface plasmon resonance (SPR) sensorgrams upon injection of a two-fold dilution series of TβRII from 0.047 to 12 μM to immobilized TGF-β2 (FIG. 3A) or mmTGF-β2 (FIG. 3B). (Figures 3C-3H) SPR sensorgrams upon injection of a 2-fold dilution series of 0.012-3 μM TβRII (Figures 3C and 3D), 0.008-1.024 μM TβRII (Figures 3E and 3F), or 0.008-1.024 μM TβRII (Figures 3G and 3H) in the presence of 2 μM TβRII in running buffer and injected samples, against immobilized avi-TGF-β3 (Figures 3C, 3E, and 3G) or avi-mmTGF-β2-7M (Figures 3D, 3F, and 3H). The sensorgrams shown in Figures 3C, 3D, and 3G were fitted to a 1:1 binding model, and raw data and fitted curves are shown. TGF-β2 and mmTGF-β2 were immobilized by direct carbodiimide-based amine coupling to the sensor surface, and avi-TGF-β3 or avi-mmTGF-β2-7M were immobilized by enzymatically biotinylated protein capture onto the surface of a streptavidin-coated sensor chip at high (approximately 8000 response units (RU)) density. [Fig. 4ABCD] Figures 4A-4D: Solubility of TGF-β2 and monomeric variants. (Figures 4A and 4C) TGF-β2 and mTGF-β2 (Figure 4A), as well as mmTGF-β2 and mmTGF-β2-7M (Figure 4C), were diluted from concentrated stocks in 100 mM acetic acid into PBS at 7.4 (neutral pH) or 100 mM acetic acid (acidic pH) and light scattering was measured at 340 nm. (Figures 4B and 4D) Samples of TGF-β2 and mTGF-β2 (Figure 4B), as well as mmTGF-β2 and mmTGF-β2-7M (Figure 4D), diluted in PBS or 100 mM acetic acid, were centrifuged at 20,000 x g for 5 min and protein absorbance at 280 nm was measured. [Figure 5A]Figures 5A-5E: Structures of mmTGF-β2-7M and the mmTGF-β2-7M:TβRII complex. (Figure 5A) Assigned 1H-15N HSQC spectrum of mmTGF-β2-7M recorded at 800 MHz in 10 mM sodium phosphate, 10 mM CHAPS, 5% 2H2O, pH 4.70, 37 °C. Assigned backbone amide signals are shown by their residue numbers and one-letter amino acid codes. (Figure 5B) Crystal structure of mmTGF-β2-7M at 1.8 Å and superposition of one monomer from the crystal structure of TGF-β2 (PDB 2TGI) at 1.8 Å. Major structural features are shown, along with the engineered loop of mmTGF-β2 substituting for the palm (α3) helix of TGF-β2. (FIG. 5C) Superposition of the three mmTGF-β2-7M chains (chains A, B, and C) from the crystallographic asymmetric unit. The dashed lines correspond to the deleted segment in the engineered loop region in chain C due to weak electron density. Other details are as in FIG. 5B. (FIG. 5D) Superposition of the crystal structure of the mmTGF-β2-7M:TβRII complex at 1.8 Å with one TGF-β3 monomer and its bound TβRII from the crystal structure of the TGF-β3:TβRII:TβRI complex at 3.0 Å (PDB 2PJY, TGF-β3 monomer and TβRII; TβRI not shown for clarity). Represents the engineered loop of mmTGF-β2 replacing the palm (α3) helix of TGF-β2. (Figure 5E) The overlay is as in Figure 5B but expanded to show near-identity to the critical hydrophobic and hydrogen-bonding / electrostatic interactions previously shown to be essential for high-affinity TGF-β3:TβRII binding. [Figure 5BC]Figures 5A-5E: Structures of mmTGF-β2-7M and the mmTGF-β2-7M:TβRII complex. (Figure 5A) Assigned 1H-15N HSQC spectrum of mmTGF-β2-7M recorded at 800 MHz in 10 mM sodium phosphate, 10 mM CHAPS, 5% 2H2O, pH 4.70, 37 °C. Assigned backbone amide signals are shown by their residue numbers and one-letter amino acid codes. (Figure 5B) Crystal structure of mmTGF-β2-7M at 1.8 Å and superposition of one monomer from the crystal structure of TGF-β2 (PDB 2TGI) at 1.8 Å. Major structural features are shown, along with the engineered loop of mmTGF-β2 substituting for the palm (α3) helix of TGF-β2. (FIG. 5C) Superposition of the three mmTGF-β2-7M chains (chains A, B, and C) from the crystallographic asymmetric unit. The dashed lines correspond to the deleted segment in the engineered loop region in chain C due to weak electron density. Other details are as in FIG. 5B. (FIG. 5D) Superposition of the crystal structure of the mmTGF-β2-7M:TβRII complex at 1.8 Å with one TGF-β3 monomer and its bound TβRII from the crystal structure of the TGF-β3:TβRII:TβRI complex at 3.0 Å (PDB 2PJY, TGF-β3 monomer and TβRII; TβRI not shown for clarity). Represents the engineered loop of mmTGF-β2 replacing the palm (α3) helix of TGF-β2. (Figure 5E) The overlay is as in Figure 5B but expanded to show near-identity to the critical hydrophobic and hydrogen-bonding / electrostatic interactions previously shown to be essential for high-affinity TGF-β3:TβRII binding. [Figure 5D]Figures 5A-5E: Structures of mmTGF-β2-7M and the mmTGF-β2-7M:TβRII complex. (Figure 5A) Assigned 1H-15N HSQC spectrum of mmTGF-β2-7M recorded at 800 MHz in 10 mM sodium phosphate, 10 mM CHAPS, 5% 2H2O, pH 4.70, 37 °C. Assigned backbone amide signals are shown by their residue numbers and one-letter amino acid codes. (Figure 5B) Crystal structure of mmTGF-β2-7M at 1.8 Å and superposition of one monomer from the crystal structure of TGF-β2 (PDB 2TGI) at 1.8 Å. Major structural features are shown, along with the engineered loop of mmTGF-β2 substituting for the palm (α3) helix of TGF-β2. (FIG. 5C) Superposition of the three mmTGF-β2-7M chains (chains A, B, and C) from the crystallographic asymmetric unit. The dashed lines correspond to the deleted segment in the engineered loop region in chain C due to weak electron density. Other details are as in FIG. 5B. (FIG. 5D) Superposition of the crystal structure of the mmTGF-β2-7M:TβRII complex at 1.8 Å with one TGF-β3 monomer and its bound TβRII from the crystal structure of the TGF-β3:TβRII:TβRI complex at 3.0 Å (PDB 2PJY, TGF-β3 monomer and TβRII; TβRI not shown for clarity). Represents the engineered loop of mmTGF-β2 replacing the palm (α3) helix of TGF-β2. (Figure 5E) The overlay is as in Figure 5B but expanded to show near-identity to the critical hydrophobic and hydrogen-bonding / electrostatic interactions previously shown to be essential for high-affinity TGF-β3:TβRII binding. [Figure 5E]Figures 5A-5E: Structures of mmTGF-β2-7M and the mmTGF-β2-7M:TβRII complex. (Figure 5A) Assigned 1H-15N HSQC spectrum of mmTGF-β2-7M recorded at 800 MHz in 10 mM sodium phosphate, 10 mM CHAPS, 5% 2H2O, pH 4.70, 37 °C. Assigned backbone amide signals are shown by their residue numbers and one-letter amino acid codes. (Figure 5B) Crystal structure of mmTGF-β2-7M at 1.8 Å and superposition of one monomer from the crystal structure of TGF-β2 (PDB 2TGI) at 1.8 Å. Major structural features are shown, along with the engineered loop of mmTGF-β2 substituting for the palm (α3) helix of TGF-β2. (FIG. 5C) Superposition of the three mmTGF-β2-7M chains (chains A, B, and C) from the crystallographic asymmetric unit. The dashed lines correspond to the deleted segment in the engineered loop region in chain C due to weak electron density. Other details are as in FIG. 5B. (FIG. 5D) Superposition of the crystal structure of the mmTGF-β2-7M:TβRII complex at 1.8 Å with one TGF-β3 monomer and its bound TβRII from the crystal structure of the TGF-β3:TβRII:TβRI complex at 3.0 Å (PDB 2PJY, TGF-β3 monomer and TβRII; TβRI not shown for clarity). Represents the engineered loop of mmTGF-β2 replacing the palm (α3) helix of TGF-β2. (Figure 5E) The overlay is as in Figure 5B but expanded to show near-identity to the critical hydrophobic and hydrogen-bonding / electrostatic interactions previously shown to be essential for high-affinity TGF-β3:TβRII binding. [Figure 6AB]Figures 6A-6B: Signaling activity of TGF-β1 and variants. (Figure 6A) TGF-β luciferase reporter activity for TGF-β1, mTGF-β3 and mmTGF-β2-7M shown as filled circles, squares and triangles, respectively. The solid lines correspond to the fitted curves and lead to the EC50 (mmTGF-β2-7M was not fitted due to the lack of signaling activity of this variant). (Figure 6B) TGF-β luciferase reporter activity upon addition of increasing concentrations of the indicated TGF-β monomeric variants to cells treated with a subsaturating concentration (8 pM) of TGF-β1 (mTGF-β3 and mmTGF-β2-7M shown as open squares and filled triangles, respectively). The solid lines correspond to the fitted curves for mTGF-β3 and lead to the EC50, and for mTGF-β2-7M and lead to the IC50. [Figure 7AB] Figures 7A-7B: Time-resolved fluorescence resonance energy transfer (TR-FRET) assay for ligand-mediated association of the TβRI:TβRII complex. (Figure 7A) Structure of the TGF-β3:TβRII:TβRI complex with a C-terminal tag and fluorescently labeled donor and acceptor proteins associated with the tag. TβRII has a hexahistidine tag (His6) at the C-terminus and binds to a Tb3+-cryptate-labeled anti-hexahistidine tag antibody (Cisbio, Bedford, MA). TβRI has a biotinylated avi tag at the C-terminus and binds to XL665-labeled streptavidin (Cisbio, Bedford, MA). The single lysine residue in the biotinylated TβRI C-terminal avi tag is designated "KB." (Figure 7B) TR-FRET ΔF values ​​for samples consisting of the indicated components added. In all cases, the buffer contained 25 mM Tris, 50 mM NaCl, pH 7.4, 2 nM Tb3+-cryptate-labeled anti-hexahistidine tag antibody and 30 nM XL665-labeled streptavidin. TR-FRETΔF corresponds to the ratio of acceptor (XL665) emission at 665 nm to donor (Tb3+-cryptate) emission at 490 nm. [Figure 8]Alignment of the amino acid sequences of the TGF-β used in this study. In the figure, the sequences are numbered so that the first residue after the N-terminal methionine is residue 1. [Figure 9AB] Figures 9A-9B: Secondary structure predictions and backbone 15N T2 relaxation times of mmTGF-β2. (Figure 9A) Secondary structure predictions were calculated using the program PECAN, based on backbone HN, NH, Cα and CO and side chain Cβ atoms. Predictions of β-strands and α-helices are plotted as positive and negative values, respectively. (Figure 9B) 15N T2 relaxation times were plotted as a function of residue number. Secondary structures shown above each graph correspond to those from the crystal structure of TGF-β2 (PDB 2TGI). [Figure 10AB] Figures 10A-10B: Secondary structure predictions and backbone 15N T2 relaxation times of mmTGF-β2-7M. (Figure 10A) Secondary structure predictions were calculated using the program PECAN, based on backbone HN, NH, Cα and CO and side chain Cβ atoms. Predictions of β-strands and α-helices are plotted as positive and negative values, respectively. (Figure 10B) 15N T2 relaxation times were plotted as a function of residue number. Secondary structures shown above each graph correspond to those from the crystal structure of TGF-β2 (PDB 2TGI). [Figure 11] Finite element fit of the reversible monomer-dimer model to sedimentation velocity experiments of mTGF-β3. The finite element fit overlaid experimental data is shown at the top and the residuals are shown at the bottom. [Figure 12] Finite element fit of the reversible monomer-dimer model to sedimentation velocity experiments of mmTGF-β2. The finite element fit overlaid experimental data is shown at the top and the residuals are shown at the bottom. [Figure 13] Finite element fit of the reversible monomer-dimer model to sedimentation velocity experiments of mmTGF-β2-7M. The finite element fit overlaid experimental data is shown at the top and the residuals are shown at the bottom. [Figure 14]TR-FRET assay to assess association of the TGF-β:TβRII:TβRI complex. The concentrations of the terbium-cryptate anti-hexahistidine tag antibody donor fluorophore and streptavidin-665 acceptor fluorophore were 2 nM and 30 nM, respectively. [Figure 15] Figures 15A-15B: Inhibition of TGF-β2 and TGF-β3 by mmTGF-β2. TGF-β luciferase activity in cells treated with a constant concentration of TGF-β2 (20 pM, Figure 15A) or TGF-β3 (10 pM, Figure 15B) and with increasing concentrations of mmTGF-β2-7M. The solid line corresponds to the fitted curve and derives the IC50. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0015] Sequence Listing The amino acid sequences listed in the attached sequence listing are shown using the standard three-letter code for amino acids, as defined in 37 C.FR1.822. The sequence listing was submitted as an 11.3KB ASCII text file, created on November 6, 2017, and is incorporated herein by reference. The contents of the attached sequence listing are as follows:

[0016] SEQ ID NO:1 is the amino acid sequence of wild-type human TGF-β1.

[0017] SEQ ID NO:2 is the amino acid sequence of wild-type human TGF-β2.

[0018] SEQ ID NO:3 is the amino acid sequence of wild-type human TGF-β3.

[0019] SEQ ID NO:4 is the amino acid sequence of human TGF-β3 with an N-terminal Avi tag.

[0020] SEQ ID NO:5 is the amino acid sequence of an engineered human TGF-β2 monomer designated mTGF-β2.

[0021] SEQ ID NO:6 is the amino acid sequence of an engineered human TGF-β3 monomer designated mTGF-β3.

[0022] SEQ ID NO:7 is the amino acid sequence of an engineered human TGF-β1 monomer designated mmTGF-β1.

[0023] SEQ ID NO:8 is the amino acid sequence of an engineered human TGF-β2 monomer designated mmTGF-β2.

[0024] SEQ ID NO:9 is the amino acid sequence of an engineered human TGF-β3 monomer designated mmTGF-β3.

[0025] SEQ ID NO:10 is the amino acid sequence of an engineered human TGF-β2 monomer designated mmTGF-β2-7M.

[0026] SEQ ID NO:11 is the amino acid sequence of mmTGF-β2-7M with an N-terminal Avi tag.

[0027] Detailed Description I. Abbreviations AUC Analytical Ultracentrifugation BSA Bovine Serum Albumin EC 50 50% effective concentration FBS Fetal Bovine Serum I C 50 50% inhibitory concentration NMR nuclear magnetic resonance RU Resonance unit SPR Surface Plasmon Resonance TβRI Transforming growth factor-β type I receptor TβRII Transforming growth factor-β type II receptor TGF-β Transforming growth factor-β TR-FRET Time-resolved fluorescence resonance energy transfer

[0028] II. Terms and Methods Unless otherwise noted, terminology is used according to conventional usage. Definitions of common terms in molecular biology can be found in Benjamin Lewin, Genes V, Oxford University Press, 1994 (ISBN 0-19-854287-9); Kendrew et al. (eds.), The Encyclopedia of Molecular Biology, Blackwell Science Ltd., 1994 (ISBN 0-632-02182-9); and Robert A. Meyers (ed.), Molecular Biology and Biotechnology: a Comprehensive Desk Reference, VCH Publishers, Inc., 1995 (ISBN 1-56081-569-8). It can be found.

[0029] In order to facilitate review of the various embodiments of the disclosure, the following explanations of specific terms are provided:

[0030] Aberrant (TGF-β signaling): Abnormal or dysregulated TGF-β signaling. In the context of this disclosure, "aberrant TGF-β signaling" refers to excessive (pathological) activation of the TGF-β signaling pathway.

[0031] Administration: Providing or giving an agent, such as a therapeutic agent (e.g., recombinant TGF-β), to a subject by any effective route. Exemplary routes of administration include, but are not limited to, injection or infusion (subcutaneous, intramuscular, intradermal, intraperitoneal, intrathecal, intravenous, intraventricular, intrastriatal, intracranial and spinal, etc.), oral, intraductal, sublingual, rectal, transdermal, intranasal, vaginal and inhalation routes.

[0032] Contacting: Placing in direct physical association; includes both solid and liquid forms. When used in the context of in vivo methods, "contacting" also includes administering.

[0033] Fibrosis: The formation of excess fibrous connective tissue in an organ or tissue during repair or reaction processes. Fibrosis can occur in many different body tissues (such as the heart, lungs and liver), typically as a result of inflammation or injury. Fibrotic disorders include, but are not limited to, pulmonary fibrosis, cystic fibrosis, idiopathic pulmonary fibrosis, interstitial lung disease, liver cirrhosis, renal fibrosis (such as from damage caused by diabetes), atrial fibrosis, endomyocardial fibrosis, atherosclerosis, restenosis and scleroderma. Fibrosis can also occur as a result of surgical complications, chemotherapy drugs, radiation, injury or burns.

[0034] Fusion protein: A protein that comprises at least a portion of two different (heterologous) proteins. In some embodiments herein, the fusion protein comprises a TGF-β monomer fused to a protein tag, an Fc domain (such as a human Fc domain), or albumin.

[0035] Glycosylation: The process of covalent attachment of a sugar moiety to an asparagine (N-glycosylation), or a serine or threonine residue (O-glycosylation). The level and type of glycosylation may vary depending on the host organism used for recombinant expression. New glycosylation sites may be engineered sequences by introducing glycosylation sequons into solvent-exposed regions of a protein. For example, the N-glycosylation sequon NX[S / T] may be introduced at one or more locations within the sequence of certain embodiments disclosed herein. Varying the type and extent of glycosylation has utility in modulating solubility, function and half-life, as well as allowing site-specific chemical conjugation.

[0036] Heterologous: Derived from separate genetic sources or species.

[0037] Isolated: An "isolated" biological component, such as a nucleic acid, a protein (including an antibody) or an organelle, is substantially separated or purified from other biological components, i.e., other chromosomal and extrachromosomal DNA and RNA, proteins, and organelles, in the environment in which the component naturally occurs (such as a cell). "Isolated" nucleic acids and proteins include nucleic acids and proteins purified by standard purification methods. The term also encompasses nucleic acids and proteins prepared by recombinant expression in a host cell, as well as chemically synthesized nucleic acids.

[0038] Monomer: A single molecular unit (such as a protein) that can combine with other molecular units to form dimers or polymers. In the context of this disclosure, a "TGF-β monomer" is a single TGF-β polypeptide chain whose wild-type version can combine with other TGF-β monomers to form dimers. In some embodiments herein, recombinant TGF-β monomers are engineered to prevent dimerization. In other embodiments herein, recombinant TGF-β monomers that are engineered to prevent direct dimerization can be fused to a heterologous protein (e.g., the Fc domain of an IgG) that can itself dimerize.

[0039] Neoplasm, malignancy, cancer or tumor: A neoplasm is an abnormal growth of tissue or cells resulting from excessive cell division. Neoplastic growth can result in a tumor. The amount of tumor in an individual is the "tumor burden," which can be measured as the number, volume, or weight of tumors. Tumors that do not metastasize are called "benign." Tumors that invade surrounding tissues and / or can metastasize are called "malignant."

[0040] PEGylation: The process of both covalent and non-covalent attachment or fusion of polyethylene glycol (PEG) polymer chains to molecules and macrostructures such as drugs, therapeutic proteins or vesicles, then referred to as PEGylation (or pegylation). PEGylation is routinely achieved by incubation of reactive derivatives of PEG with the target molecule. Covalent attachment of PEG to drugs or therapeutic proteins can mask the agent from the host's immune system (reducing immunogenicity and antigenicity), increase the hydrodynamic size (size in solution) of the agent, and extend circulation time by decreasing renal clearance. PEGylation can also confer water solubility to hydrophobic drugs and proteins.

[0041] Peptide or Polypeptide: A polymer in which the monomers are amino acid residues linked through amide bonds. When the amino acids are alpha amino acids, either the L-optical isomer or the D-optical isomer may be used, with the L-isomer being preferred. The terms "peptide", "polypeptide" or "protein" as used herein are intended to encompass any amino acid sequence and include modified sequences, including modified globulin proteins. The terms "peptide" and "polypeptide" are specifically intended to include naturally occurring proteins, as well as those produced recombinantly or synthetically.

[0042] Conservative amino acid substitutions are those substitutions that, when made, disturb the properties of the original protein the least, i.e., the structure and especially the function of the protein are preserved and not significantly altered by such substitutions. Examples of conservative substitutions are given below. [Table 5]

[0043] Conservative substitutions generally maintain (a) the structure of the polypeptide backbone in the area of ​​the substitution, for example as a sheet or helix, (b) the charge or hydrophobicity of the target site molecule, or (c) the bulk of the side chains.

[0044] In general, the substitutions expected to result in the greatest change in protein properties will be non-conservative, e.g., (a) a hydrophilic residue, e.g., serine or threonine, is substituted for (or by) a hydrophobic residue, e.g., leucine, isoleucine, phenylalanine, valine or alanine; (b) a cysteine ​​or proline is substituted for (or by) any other residue; (c) a residue with a positively charged side chain, e.g., lysine, arginine, or histidine, is substituted for (or by) a negatively charged residue, e.g., glutamine or aspartic acid; or (d) a residue with a bulky side chain, e.g., phenylalanine, is substituted for (or by) one without a side chain, e.g., glycine.

[0045] Pharmaceutically acceptable carriers: Useful pharma- ceutically acceptable carriers are conventional. Remington's Pharmaceutical Sciences, Mack Publishing Co., Easton, PA, 15th ed., 1975, by E. W. Martin, describes compositions and formulations suitable for pharmaceutical delivery of the compositions disclosed herein.

[0046] Generally, the nature of the carrier will depend on the particular mode of administration being employed. In addition to biologically neutral carriers, the pharmaceutical composition to be administered may contain minor amounts of nontoxic auxiliary substances such as wetting or emulsifying agents, preservatives, and pH buffering agents, and the like, for example, sodium acetate or sorbitan monolaurate.

[0047] Preventing, Treating, or Ameliorating a Disease: "Preventing" a disease refers to inhibiting the full development of the disease. "Treatment" refers to a therapeutic intervention that ameliorates a sign or symptom of a disease or condition after the disease or condition has begun to develop, such as reducing tumor burden or decreasing the number of metastases size. "Amelioration" refers to a reduction in the number or severity of signs or symptoms of a disease.

[0048] Recombinant: A recombinant nucleic acid or protein is one that has a sequence that is not naturally occurring or that is made by the artificial combination of two otherwise separate sequence segments. This artificial combination is often accomplished by chemical synthesis or the artificial manipulation of isolated segments of nucleic acid, e.g., by genetic engineering techniques. The term recombinant includes nucleic acids and proteins that have been altered by the addition, substitution, or deletion of portions of a naturally occurring nucleic acid molecule or protein.

[0049] Sequence identity / similarity: Identity between two or more nucleic acid sequences or between two or more amino acid sequences is expressed by the identity or similarity between the sequences. Sequence identity can be measured by identity percentage, the higher the percentage, the more identical the sequences. Sequence similarity can be measured by similarity percentage (taking into account conservative amino acid substitutions), the higher the percentage, the more similar the sequences. Homologs or orthologs of nucleic acid or amino acid sequences have a relatively high degree of sequence identity / similarity when aligned using standard methods. This homology is more pronounced when orthologous proteins or cDNAs are from more closely related species (such as human and mouse sequences) compared to more distantly related species (such as human and C. elegans sequences).

[0050] Methods for aligning sequences for comparison are well known in the art. Various programs and alignment algorithms are described in: Smith and Waterman, Adv. Appl. Math. 2:482, 1981; Needleman and Wunsch, J. Mol. Biol. 48:443, 1970; Pearson and Lipman, Proc. Natl. Acad. Sci. USA 85:2444, 1988; Higgins and Sharp, Gene 73:237-4 4, 1988; Higgins and Sharp, CABIOS 5:151-3, 1989; Corpet et al., Nuc. Acids Res. 16:10881-90, 1988; Huang et al. Computer Appls., Biosciences 8:155-65, 1992; and Pearson et al., Meth. Mol. Bio. 24:307-31, 1994. Altschul et al., J. Mol. Biol. 215:403-10, 1990, presents a detailed consideration of sequence alignment methods and homology calculations.

[0051] NCBI Basic Local Alignment Search Tool (BLAST) (Altschul et al., J. Mol. Biol. 215:403-10, 1990) ) is available from several sources, including the National Center for Biotechnology Information (NCBI), and on the Internet, for use in conjunction with the sequence analysis programs blastp, blastn, blastx, tblastn, and tblastx. Additional information can be found on the NCBI website.

[0052] Subject: Living multicellular organisms, including vertebrate organisms, a category that includes both human and non-human mammals.

[0053] Tag: A molecule that can be added to a protein or nucleic acid, such as for labeling, detection or purification purposes. In some embodiments, the tag is a protein tag. In some embodiments, the protein tag is an affinity tag (e.g., Avi tag, hexahistidine, chitin-binding protein, maltose-binding protein, or glutathione-S-transferase), an epitope tag (e.g., V5, c-myc, HA, or FLAG), or a fluorescent tag (e.g., GFP or another well-known fluorescent protein).

[0054] Therapeutically effective amount: A quantity of a compound or composition, e.g., recombinant TGF-β monomer, sufficient to produce a desired effect in a subject being treated. For example, this may be the amount necessary to inhibit or block TGF-β signaling in a cell.

[0055] Transforming growth factor-β (TGF-β): A secreted, multifunctional protein that controls proliferation, cell differentiation, and several other cellular functions. Many cells synthesize TGF-β, and almost all cells express receptors for TGF-β. The term "TGF-β" refers to three distinct protein isoforms, TGF-β1, TGF-β2, and TGF-β3, encoded by the genes TGFB1, TGFB2, and TGFB3, respectively.

[0056] TGF-β signaling pathway: A signaling pathway involved in many cellular processes, such as cell proliferation, differentiation and apoptosis. Members of the TGF-β pathway include, but are not limited to, TGF-β1, TGF-β2, TGF-β3, and TGF-β receptor type I and TGF-β receptor type II.

[0057] TGF-β receptor: The term "TGF-β receptor" includes TGF-β receptor type I (encoded by TGFBR1) and TGF-β receptor type II (encoded by TGFBR2). TGF-β receptors are serine / threonine protein kinases. Type I and type II TGF-β receptors form a heterodimeric complex upon binding to TGF-β, transducing the TGF-β signal from the cell surface to the cytoplasm.

[0058] Unless otherwise explained, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this disclosure belongs. The singular terms "a", "an", and "the" include plural referents unless the context clearly indicates otherwise. "Comprising A or B" means including A, or B, or A and B. It should be further understood that all base or amino acid sizes and all molecular weight or molecular mass values ​​given for nucleic acids or polypeptides are approximate and are provided for illustration purposes. Although methods and materials similar or equivalent to those described herein can be used in the practice or testing of this disclosure, suitable methods and materials are described below. All publications, patent applications, patents, and other references mentioned herein are incorporated by reference in their entirety. In case of conflict, the present specification, including explanations of terms, will control. Furthermore, the materials, methods, and examples are illustrative only and are not intended to be limiting.

[0059] III. Overview of Some Embodiments Disclosed herein are recombinant transforming growth factor (TGF)-β monomers modified to inhibit dimerization and type I receptor binding but retain the ability to bind to the high affinity TGF-β type II receptor (TβRII). The recombinant TGF-β monomers disclosed herein can be used to inhibit TGF-β signaling, such as for the treatment of diseases or disorders characterized by abnormal TGF-β signaling, such as fibrotic disorders, eye diseases, certain types of cancer, or genetic disorders of connective tissue. Furthermore, nucleic acid molecules encoding recombinant TGF-β monomers can be used to reprogram T cells to overproduce the recombinant protein. T cells engineered to overexpress recombinant TGF-β monomers can be used in gene therapy applications, such as for the treatment of diseases or disorders characterized by abnormal TGF-β signaling.

[0060] Provided herein are recombinant TGF-β monomers that include a cysteine ​​to serine substitution at amino acid residue 77 (with respect to SEQ ID NO:2); a deletion of amino acid residues 52-71 (with respect to SEQ ID NO:2); and at least one amino acid substitution (e.g., a substitution proximal to the deleted residue) relative to wild-type TGF-β monomer that increases the net charge of the monomer. The cysteine ​​to serine substitution prevents disulfide bond formation between TGF-β monomers. The deletion of amino acid residues 52-71 removes the α-helical 3 (α3) region (the major dimerization motif), as well as several flanking residues (FIG. 1D). When residues 52-71 are removed, the remaining residues form a loop containing polar and charged residues (FIGS. 1D and 2B).

[0061] In some embodiments, the TGF-β monomer is a human, mouse, rat or other mammalian TGF-β monomer.

[0062] In some embodiments, the TGF-β monomer further comprises at least one amino acid substitution compared to wild-type TGF-β2 monomer that increases the affinity of the TGF-β monomer for TβRII.

[0063] In some embodiments, the TGF-β monomer is a human TGF-β2 monomer. In some instances, the at least one amino acid substitution that increases the net charge of the human TGF-β2 monomer includes a leucine to arginine substitution at residue 51; an alanine to lysine substitution at residue 73; or both a leucine to arginine substitution at residue 51 and an alanine to lysine substitution at residue 73 (with respect to SEQ ID NO:2).

[0064] In some embodiments, the at least one amino acid substitution that increases the affinity of the human TGF-β2 monomer for TβRII includes a substitution at an amino acid residue corresponding to residue 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, or 99 of SEQ ID NO: 2, or any combination of two or more of these residues. In some examples, the at least one amino acid substitution that increases the affinity of the monomer for TβRII includes at least one substitution at residue 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, or 37, and at least one substitution at residue 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, or 99. In particular examples, the at least one amino acid substitution that increases the affinity of the human TGF-β2 monomer for TβRII includes lysine to arginine at residue 25, arginine to lysine at residue 26, leucine to valine at residue 89, isoleucine to valine at residue 92, asparagine to arginine at residue 94, threonine to lysine at residue 95, isoleucine to valine at residue 98, or any combination of two or more thereof, e.g., three or more, four or more, five or more, or six or more. In one non-limiting example, the recombinant human TGF-β2 monomer includes lysine to arginine at residue 25, arginine to lysine at residue 26, leucine to valine at residue 89, isoleucine to valine at residue 92, asparagine to arginine at residue 94, threonine to lysine at residue 95, and isoleucine to valine at residue 98.

[0065] In some examples, the amino acid sequence of the human TGF-β2 monomer is at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% identical to SEQ ID NO:8 or SEQ ID NO:10. In some cases, the human TGF-β2 monomer is at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% identical to SEQ ID NO:8 or SEQ ID NO:10 and contains only conservative amino acid substitutions. In certain non-limiting examples, the amino acid sequence of the human TGF-β2 monomer comprises or consists of SEQ ID NO:8 or SEQ ID NO:10.

[0066] In other embodiments, the recombinant TGF-β monomer is a human TGF-β1 monomer. In some examples, the at least one amino acid substitution that increases the net charge of the human TGF-β1 monomer includes an isoleucine to arginine substitution at residue 51; an alanine to lysine substitution at residue 74; an alanine to serine substitution at residue 75; or an isoleucine to arginine substitution at residue 51, an alanine to lysine substitution at residue 74, and an alanine to serine substitution at residue 75 (with respect to SEQ ID NO: 1).

[0067] In some examples, the amino acid sequence of the human TGF-β1 monomer is at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% identical to SEQ ID NO: 7. In some cases, the human TGF-β1 monomer is at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% identical to SEQ ID NO: 7 and contains only conservative amino acid substitutions. In certain non-limiting examples, the amino acid sequence of the human TGF-β1 monomer comprises or consists of SEQ ID NO: 7.

[0068] In other embodiments, the recombinant TGF-β monomer is a human TGF-β3 monomer. In some examples, the at least one amino acid substitution that increases the net charge of the human TGF-β3 monomer includes a leucine to glutamic acid substitution at residue 51; an alanine to glutamic acid substitution at residue 72; an alanine to aspartic acid substitution at residue 74; or a leucine to glutamic acid substitution at residue 51, an alanine to glutamic acid substitution at residue 72, and an alanine to aspartic acid substitution at residue 74 (with respect to SEQ ID NO:3).

[0069] In some examples, the amino acid sequence of the human TGF-β3 monomer is at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% identical to SEQ ID NO: 9. In some cases, the human TGF-β3 monomer is at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% identical to SEQ ID NO: 9 and contains only conservative amino acid substitutions. In certain non-limiting examples, the amino acid sequence of the human TGF-β3 monomer comprises or consists of SEQ ID NO: 9.

[0070] In some embodiments herein, the recombinant TGF-β monomer is PEGylated, glycosylated, hyperglycosylated, or contains another modification that enhances circulation time.

[0071] Also provided herein is a fusion protein comprising a TGF-β monomer and a heterologous protein. In some embodiments, the heterologous protein is a protein tag. In some examples, the protein tag is an affinity tag (e.g., Avi tag, hexahistidine, chitin-binding protein, maltose-binding protein, or glutathione-S-transferase), an epitope tag (e.g., V5, c-myc, HA, or FLAG), or a fluorescent tag (e.g., GFP or another well-known fluorescent protein). In other embodiments, the heterologous protein comprises an Fc domain, such as a mouse or human Fc domain. In certain embodiments, the heterologous protein promotes intermolecular association of the fusion protein into a homodimeric (e.g., Fc domain from human IgG1, IgG2, IgG3), heterodimeric (e.g., engineered Fc domain, E / K coiled coil), or multimeric (e.g., pentabody, nanoparticle) state. In other embodiments, the heterologous protein is albumin, an albumin binding protein or drug, or another protein that increases the circulation time of TGF-β monomers in vivo.

[0072] Also provided are recombinant TGF-β monomers or fusion proteins that include a radiotherapeutic agent, a cytotoxic agent for chemotherapy, or a drug. Also provided are recombinant TGF-β monomers or fusion proteins that include an imaging agent, a fluorescent dye, or a fluorescent protein tag.

[0073] Further provided herein are compositions, such as pharmaceutical compositions, comprising a recombinant TGF-β monomer or fusion protein disclosed herein and a pharma- ceutically acceptable carrier, diluent or excipient.

[0074] Also provided herein is a method of inhibiting TGF-β signaling in a cell. In some embodiments, the method comprises contacting the cell with a recombinant TGF-β monomer, fusion protein, or composition disclosed herein.

[0075] In some embodiments, the method is an in vitro method.

[0076] In other embodiments, the method is an in vivo method. In some examples, the in vivo method comprises administering a recombinant TGF-β monomer, fusion protein or composition to a subject having a disease or disorder associated with abnormal TGF-β signaling. In some examples, the recombinant TGF-β monomer, fusion protein or composition is administered by injection, such as subcutaneous, intramuscular, intradermal, intraperitoneal, intravenous or intratumoral injection.

[0077] Also provided are methods of treating a disease or disorder associated with aberrant TGF-β signaling. In some embodiments, the methods comprise administering to a subject a recombinant TGF-β monomer, fusion protein, or composition disclosed herein.

[0078] In some embodiments, the disease or disorder associated with abnormal TGF-β signaling is a fibrotic disorder such as, but not limited to, pulmonary fibrosis, cystic fibrosis, idiopathic pulmonary fibrosis, interstitial lung disease, liver cirrhosis, renal fibrosis (such as from injury due to diabetes), atrial fibrosis, endomyocardial fibrosis, atherosclerosis, restenosis, scleroderma, or fibrosis due to surgical complications, chemotherapy drugs, radiation, injury or burns.

[0079] In other embodiments, the disease or disorder associated with aberrant TGF-β signaling is breast cancer, brain cancer, pancreatic cancer, prostate cancer, skin cancer, bladder cancer, liver cancer, ovarian cancer, kidney cancer, endometrial cancer, colorectal cancer, gastric cancer, skin cancer (such as malignant melanoma), or thyroid cancer.

[0080] In other embodiments, the disease or disorder associated with aberrant TGF-β signaling is an ocular disease.

[0081] In other embodiments, the disease or disorder associated with aberrant TGF-β signaling is an inherited disorder of connective tissue.

[0082] Further provided is an isolated nucleic acid molecule encoding a recombinant TGF-β monomer disclosed herein. In some embodiments, the nucleic acid molecule is operably linked to a promoter, such as a T cell specific promoter.

[0083] Also provided is a vector comprising a TGF-β monomer-encoding nucleic acid molecule. In some embodiments, the vector is a viral vector, such as a lentiviral vector.

[0084] Further provided are isolated cells, such as, but not limited to, isolated T cells, comprising a nucleic acid molecule or vector encoding a recombinant TGF-β monomer as disclosed herein. The cells may be autologous to the subject, or the cells may be xenogeneic (allogeneic). Also provided are compositions comprising the isolated cells and a pharma- ceutically acceptable carrier.

[0085] Further provided is a method for treating a disease or disorder associated with abnormal TGF-β signaling in a subject. In some embodiments, the method comprises administering to a subject a nucleic acid molecule, vector or isolated cell disclosed herein. In some examples, the disease or disorder associated with abnormal TGF-β signaling is a fibrotic disorder. In other examples, the disease or disorder associated with abnormal TGF-β signaling is breast cancer, brain cancer, pancreatic cancer, prostate cancer or skin cancer. In other examples, the disease or disorder associated with abnormal TGF-β signaling is an eye disease. In yet other examples, the disease or disorder associated with abnormal TGF-β signaling is a hereditary disorder of connective tissue.

[0086] IV. Administration of engineered TGF-β monomers Provided herein are compositions, such as pharmaceutical compositions, that include recombinant human TGF-β monomer or fusion proteins. Also provided are compositions that include isolated cells, such as T cells, that include a vector encoding recombinant human TGF-β monomer. In some embodiments, the composition includes a pharma- ceutically acceptable carrier.

[0087] The pharma- ceutically acceptable carriers and excipients useful in this disclosure are conventional. See, for example, Remington: The Science and Practice of Pharmacy, The University of the Sciences in Philadelphia, editors Lippincott, Williams, and Wilkins, Philadelphia, PA, 21st ed. (2005). For example, parenteral formulations usually contain injectable fluids that are pharma- ceutical and physiologically acceptable fluid media, such as water, physiological saline, other balanced salt solutions, aqueous dextrose, glycerol, and the like. For solid compositions (e.g., powder, pill, tablet, or capsule forms), conventional non-toxic solid carriers can include, for example, pharmaceutical grades of mannitol, lactose, starch, or magnesium stearate. In addition to biologically neutral carriers, the pharmaceutical compositions to be administered can contain minor amounts of non-toxic auxiliary substances, such as wetting or emulsifying agents, preservatives, pH buffering agents, and the like (e.g., sodium acetate or sorbitan monolaurate). Excipients that may be included are, for example, human serum albumin or other proteins such as plasma preparations.

[0088] For administration of cells, various aqueous carriers, such as buffered saline, can be used to introduce cells. These solutions are sterile and generally free of undesirable substances. These compositions can be sterilized by conventional, well-known sterilization techniques. The compositions may contain pharma- ceutically acceptable auxiliary substances, such as pH adjusting and buffering agents, toxicity adjusting agents, etc. (e.g., sodium acetate, sodium chloride, potassium chloride, calcium chloride, sodium lactate, etc.), as needed to approximate physiological conditions. The concentrations in these formulations can vary widely and are selected primarily based on fluid volumes, viscosities, body weight, etc., according to the particular method of administration selected and the needs of the subject.

[0089] The dosage form of the composition is determined by the method of administration selected. For example, in addition to injectable fluids, topical, inhalation, oral and suppository formulations may be used. Topical preparations may include eye drops, ointments, sprays, patches, etc. Inhalation preparations may be liquid (e.g., solutions or suspensions) and may include mists, sprays, etc. Oral formulations may be liquid (e.g., syrups, solutions or suspensions) or solid (e.g., powders, pills, tablets, or capsules). Suppository preparations may also be in solid, gel, or suspension form. For solid compositions, conventional non-toxic solid carriers may include pharmaceutical grades of mannitol, lactose, starch, or magnesium stearate. Actual methods of preparing such dosage forms are known or will be apparent to those skilled in the art.

[0090] Compositions, such as pharmaceutical compositions, containing recombinant human TGF-β monomer can be formulated in unit dosage forms suitable for individual administration of precise dosage amounts.The amount of TGF-β monomer administered depends on the subject being treated, the severity of the disease, and the mode of administration, and is best left to the discretion of the prescribing physician.Within these ranges, the formulation administered contains an amount of active ingredient in an amount effective to achieve the desired effect in the subject being treated.

[0091] TGF-β monomers, or compositions thereof, can be administered in various ways to the tissues of humans or other animals where they are effective, such as topically, orally, intravenously, intramuscularly, intraperitoneally, intranasally, intradermally, intrathecally, subcutaneously, by inhalation or suppository. The particular method of administration and dosing regimen will be selected by the attending physician, taking into account the particulars of the case (e.g., the subject, the disease, the condition, and whether the treatment is preventive). Treatment may require a daily dose or multiple daily doses of the compound for several days to several months or even years.

[0092] The following examples are provided to illustrate particular features and / or embodiments, and should not be construed as limiting the disclosure to the particular features or embodiments described. EXAMPLES

[0093] Example 1 Engineered TGF-β monomers that block TGF-β signaling to function as dominant negatives This example describes an engineered TGF-β monomer that can block TGF-β signaling. The engineered TGF-β monomer, referred to herein as mmTGF-β2-7M, has three changes compared to the wild-type dimeric TGF-β2 monomer: (1) A cysteine ​​(Cys77) that normally forms an interchain disulfide was replaced by a serine ( Fig. 1D ); (2) the α3 helix was removed and replaced with a short loop carrying polar and charged residues ( Fig. 1D ); (3) Seven residues were substituted relative to TGF-β2, which enabled TβRII binding with high affinity (Figure 1E).

[0094] Characteristics of mmTGF-β2-7M and other engineered TGF-β variants disclosed herein are described below and listed in Table 1. The sequences of all engineered TGF-β variants are shown in FIG. 8 and are provided as SEQ ID NOs: 1-11.

[0095] background Previous studies have demonstrated that wild-type TGF-β1 and TGF-β3 monomers (TGF-β1 and TGF-β3 monomers in which the cysteine ​​residue Cys77 that normally forms the interchain disulfide is replaced by serine) have a midpoint stimulatory potency (EC ) of approximately 150 pM, despite being approximately 10-15 times less potent than the native disulfide-linked homodimers. 50 ) and have been shown to retain significant signaling activity (Amatayakul-Chantler et al., J Biol Chem 269:27687-27691, 1994; Zuniga et al., J Mol Biol 355:47-62, 2006).

[0096] Based on structural studies, it has been shown that the TGF-β type I receptor (TβRI), one of the two essential receptors that bind the growth factor, binds by spanning the TGF-β homodimer interface (Figure 1A) (Groppe et al., Mol Cell 29:157-1). 68, 2008), it was unclear why the TGF-β1 Cys77→Ser and TGF-β3 Cys77→Ser variants retained such pronounced signaling activity.

[0097] We hypothesized that TGF-β monomers noncovalently dimerize and signal by binding to receptors, thereby stabilizing the noncovalent dimer (due to the fact that at least one of the receptors, TβRI, binds across the dimer interface). To obtain TGF-β monomers that function as inhibitors rather than stimulators of TGF-beta signaling, we produced engineered monomers in which the major dimerization motif, the interfacial α-helix (α3), was replaced with a flexible loop (Figures 1A and 1D). This was thought to interfere with the ability of TGF-β to dimerize and recruit the TGF-β type I receptor (TβRI), by (a) limiting its ability to dimerize noncovalently due to hydrophobic contacts (Figure 1B) and (b) removing the majority of the contact surface for the TGF-β type I receptor (TβRI), which binds by spanning the TGF-beta dimer interface (Figure 1C).

[0098] method Protein expression and purification TGF-β1 was expressed as a secreted protein associated with the prodomain in stably transfected Chinese hamster ovary (CHO) cells. The cell line used to produce TGF-β1 and the subsequent method for isolating the mature disulfide-linked TGF-β1 homodimer from the conditioned medium have been previously described (Zou et al., 2003). (Yoshino and Sun, Protein Expr Purif 37, 265-272, 2004). Dimeric TGF-β2 (TGF-β2), human homodimeric TGF-β3 (TGF-β3), and avi-tagged (Cull and Schatz, Methods Enzymol. 326, pp. 430-440, 2000) homodimeric TGF-β3 (TGF-β3-avi), monomeric TGF-β2 (mTGF-β2), monomeric TGF-β2 (mTGF-β3), minimonomeric TGF-β1 (mmTGF-β1), minimonomeric TGF-β2 (mmTGF-β2), minimonomeric TGF-β3 (mmTGF-β3), minimonomeric TGF-β2 with seven substitutions that allow TβRII binding with high affinity (mmTGF-β2-7M), and an avi tag with seven substitutions that allow TβRII binding with high affinity (Cull and Schatz, Methods Enzymol. J. Am. Soc ... 326, 430-440, 2000) Variants including minimonomeric TGF-β2 (mmTGF-β2-7M) were expressed in E. coli and refolded from inclusion bodies to native folded disulfide-linked homodimers (TGF-β2, TGF-β3, TGF-β3-avi) or monomers (mTGF-β1, mTGF-β2, mTGF-β3, mmTGF-β1, mmTGF-β2, mmTGF-β3, mmTGF-β2-7M, mmTGF-β2-7M-avi) and purified to homogeneity using high-resolution cation exchange chromatography as described (Huang and Hinck, Methods Mol Biol 1344, 63-92, 2016) (Source Q, GE Healthcare, Piscataway, NJ). The nomenclature and key characteristics of the dimeric and monomeric TGF-βs used in this study are summarized in Table 1, and the full sequences are shown in FIG.

[0099] [Table 1-1] [Table 1-2]

[0100] The human TβRI ectodomain (TβRI), spanning residues 1-101 of the mature receptor, or a variant spanning residues 1-88 of the mature receptor with a 15 amino acid avi tag at the C-terminus (TβRI-ΔC-Avi) (Cull and Schatz, Methods Enzymol 326:430-440, 2000), was expressed in E. coli, refolded from inclusion bodies, and purified to homogeneity as previously described (Zuniga et al., J Mol Biol 354:1052-1068, 2005). The human TβRII ectodomain spanning residues 15-136 of the mature receptor (TβRII) or one bearing a hexahistidine tag at its C-terminus (TβRII-His) was expressed in E. coli, refolded from inclusion bodies, and purified to homogeneity as previously described (Hinck et al., J Biomol NMR 18:36). pp. 9-370, 2000).

[0101] Solubility assay TGF-β dimers and monomers were concentrated to 300 μM or higher in 100 mM acetic acid and diluted to the desired concentrations in 100 mM acetic acid or phosphate buffered saline (PBS, 10 mM Na2HPO4, 1.8 mM KH2PO4, 137 mM NaCl, 2.7 mM KCl, pH 7.4). The pH of the samples diluted in PBS was adjusted to a final pH of 7.4 with a small aliquot of NaOH. The light scattering of the samples at 340 nm was measured using an HP 8452 Diode Array Spectrophotometer (HP, Palo Alto, CA). The samples were transferred to microfuge tubes, centrifuged at 20,000×g for 5 min, and the absorbance of the supernatant at 280 nm was measured using a NANODROP™ Spectrophotometer (ThermoFisher, Waltham, MA). The results of the solubility assay are shown in Figures 4A-4D.

[0102] Nuclear magnetic resonance (NMR) spectroscopy 0.1% (w / v) 15 NH4Cl, or 0.1% (w / v) 15 NH4Cl and 0.03% (w / v) 13For NMR, bacterial cells were grown in M9 medium containing C-labeled glucose. 15 N, or 15 N and 13 C radiolabeled mmTGF-β2 and mmTGF-β2-7M samples were prepared. All NMR samples were prepared in 10 mM sodium phosphate, 10 mM 3-[(3-cholamidopropyl)dimethylammonio]-1-propanesulfonate (CHAPS), 5% 2 Protein concentrations of 0.2 mM to 0.4 mM were prepared in HO, 0.02% w / v sodium azide, pH 4.7. All NMR data were acquired at 700 or 800 MHz with a sample temperature of 37 °C. High sensitivity HNCACB (Wittekind and Mueller, J Magn Reson Ser B 101:201-205, 1993), CBCA(CO)NH (Grzesiek et al., J Magn Reson Ser B 101:114-119, 1993), C(CO)NH (Grzesiek and Bax, J Biomol NMR 3:185-204, 1993), H Retrieved by NCO (Kay et al., J Magn Reson 89:496-514, 1990) and analyze, 13 C. 15 Backbone resonance assignments for mmTGF-β2 and mmTGF-β2-7M were obtained as a data set with 25% non-uniform sampling (NUS) at N capture grid positions. 8:8972-8979, 1989) 1 Using the H detection pulse method, 70.95MHz 15 N frequency, 300°K, backbone amide in interleaving fashion 15 N T2 relaxation parameters were measured. Eight to ten delay times, varying from 16 to 192 ms, were used to collect each T2 data set. The relative peak intensity as a function of T2 delay time was fitted to a two parameter exponential decay to obtain the T2 relaxation time. NMRPipe was used to measure the NUS data set.13 C and 15 Data were processed by the SMILE algorithm (Ying et al., J. Biomol. NMR 2016), which is used to predict missing points in N dimensions (Delaglio et al., J Biomol NMR 6:277-293, 1995). Data analysis was performed using NMRFAM-SPARKY (Lee et al., Bioinformatics 31:1325-1327, 2015). .

[0103] SPR binding measurement SPR measurements with TGF-β2 and mmTGF-β2 shown in Figures 3A-3B were performed using a BIACORE™ 3000 SPR (GE Healthcare, Piscataway, NJ) instrument, where TGF-β2 or mmTGF-β2 was directly immobilized onto the surface of a CM5 sensor chip (GE Healthcare, Piscataway, NJ) using an amine (carbodiimide-based) coupling kit (GE Healthcare, Piscataway, NJ). SPR experiments with TGF-β3 and mmTGF-β2-7m shown in Figures 3C, 3E, and 3G, and Figures 3D, 3F, and 3H, respectively, were performed using a BIACORE™ X100 SPR instrument (GE Healthcare, Piscataway, NJ) with moderate density (50-200 RU) capture of biotinylated ligand onto a streptavidin-coated CM5 sensor chip (GE Healthcare, Piscataway, NJ). Biotinylated TGF-β3 or mmTGF-β2-7M was generated by expressing TGF-β3 or mmTGF-β2-7M with an N-terminal 15 amino acid avi tag (Cull and Schatz, Methods Enzymol 326:430-440, 2000). TGF-β3-avi or mmTGF-β2-7M-avi was biotinylated by binding to TβRII in 10 mM bicine at pH 8.0 and incubating with catalytic amounts of bacterially expressed BirA recombinase, biotin, and ATP for 2 hours at 37°C as described (Huang and Hinck, Methods Mol Biol 1344:63-92, 2016). Biotinylated avi-tagged TGF-β3 or avi-tagged TGF-β2-7m was bound to a C4 reversed-phase column equilibrated with 94.9% water / 5% acetonitrile / 0.1% trifluoroacetic acid and eluted with a linear gradient of acetonitrile.

[0104] SPR measurements shown in Figures 3A-3E were performed by injecting the indicated receptors in a two-fold dilution series over the indicated concentration range in HBS-EP buffer (10 mM HEPES, pH 7.4, 150 mM NaCl, 3 mM EDTA, 0.005% surfactant P20, GE Healthcare, Piscataway, NJ). Injections were performed in duplicate, including 10 buffer blank injections at the beginning of the experiment. Binding was performed at a flow rate of 100 mL / min, with association for 2-3 min, followed by dissociation for 1 min or longer. After each injection cycle, the cells were regenerated with 10 ml of 4 M guanidine·HCl, 2 M NaCl. Data were processed by subtracting the response from a blank flow cell and a buffer blank using the program Scrubber2 (Biologic software, Campbell, Australia). Kinetic fitting of the data was performed with Scrubber2 assuming a simple 1:1 binding model. The SPR measurements shown in Figures 3G and 3H were carried out similarly, except that 2 μM TβRII was included in the running buffer and in the injected sample. The results of the SPR measurements are shown in Table 2 and Figures 3A-3H.

[0105] [Table 2]

[0106] Crystallization, structure determination and refinement Crystals of mmTGF-β2 were formed in sitting drops at 25°C by mixing 0.2 μL of 7.9 mg / mL protein stock in 10 mM MES, pH 5.5 with 0.2 μL of precipitate from the well, 20% PEG 3350, and 0.2 M sodium thiocyanate. The harvested crystals were attached to a small nylon loop that was rid of excess mother liquor and flash cooled in liquid nitrogen before data collection. Advanced Photon Source: Data were obtained and accumulated at NE-CAT beamline 24-ID-C and published in XDS (Kabsch, Acta Crystallogr D Biol Crystallogr 66, 125-132, The analysis was performed using the PDB entry 2TGI (Daopin et al., Science 257:369-373, 1992) as a search model. , PHASER (McCoy et al., J Appl Crystallogr 40, 658-674, 200 The structure was determined by molecular replacement as implemented in the Sigma-Aldrich ...

[0107] Crystals of mmTGF-β2-7M:TβRII complex were formed in hanging drops by mixing 1.0 μL of a 7.4 mg / mL stock solution of the complex in 10 mM Tris, pH 7.4 with 1.0 μL of 0.1 M HEPES, pH 7.5, 60% v / v (+ / -)-2-methyl-2,4-pentanediol at 25°C. Recovered crystals were attached to nylon loops and flash cooled with liquid nitrogen before data collection. Data were acquired and collected on an Advanced Photon Source 24-ID-C and analyzed using an HKL2000 (Otwinowski and Minor, Method Enzymol 276:307-326, 1997). We used TβRII (PDB 1M9Z; Boesen et al., Structure 10, 913-919, 2002) and mmTGF-β2 as search models, and PHASER (McCoy et al. The structure was determined by molecular replacement as implemented in J Appl Crystallogr 40, 658-674, 2007). Coordinates were refined using PHENIX (Adams et al., Acta Crystallogr D Biol Crystallogr 66, 213-221, 2010) using COOT (Emsley et al., Acta Crystallogr D Biol Crystallogr 66, 4 The data collection and refinement statistics are shown in Table 3.

[0108] Crystals of mmTGF-β2-7M were formed in hanging drops by mixing 1.0 μL of a 10 mg / mL protein stock in 20 mM acetic acid with 0.8 μL of precipitant from the well, 100 mM sodium acetate dibasic trihydrate (pH 4.6), 25% 2-propanol, 400 mM anhydrous calcium chloride, and 0.2 μL of 5% n-octyl-β-D-glucoside at 25° C. Recovered crystals were attached to nylon loops, cryopreserved in well buffer containing 20% ​​glycerol, and flash cooled in a stream of nitrogen. Data were collected at 100K using a Rigaku FR-E Superbright generator with a Saturn 944 CCD detector and processed using MOSFLM (Battye et al., Acta Crystallogr D 67, 235-242, 2011) in CCP4 (Winn et al., Acta Crystallogr D 67, 271-281, 2011). The structure of mmTGF-β2-7M was solved by molecular replacement with TβRII using the structure of mmTGF-β2-7M derived from a co-crystal structure. Iterative model building and refinement were performed using COOT (Emsley et al., Acta Crystallogr D Biol Crystallogr D 67, 271-281, 2011), respectively. 66, pp. 486-501, 2010) and PHENIX. Data collection and refinement statistics are shown in Table 3.

[0109] The results of the structural analysis are shown in Figures 2A-2D, 5A-5E, 9A-9B, and 10A-10B.

[0110] [Table 3-1] [Table 3-2]

[0111] Luciferase assay CAGA 12 Human embryonic kidney 293 (HEK293) cells stably transfected with a TGF-β reporter were used for luciferase reporter assays (Thies et al., Growth Factors 18:251-259, 2001). Stably transfected CAGA cells were cultured in Dulbecco's modified Eagle's medium (DMEM) containing 10% fetal bovine serum (FBS) and 1% penicillin / streptomycin. 12 HEK293 cells containing a TGF-β reporter were maintained. Cells were treated for 16 h with a concentration series of TGF-β (TGF-β1, mTGF-β3 or mmTGF-β2-7M) or a concentration series of mmTGF-β2-7M in the presence of constant subsaturating concentrations of TGF-β (TGF-β1, 8 pM; TGF-β2, 20 pM; TGF-β3, 10 pM). Proteins were diluted in DMEM containing 0.1% (w / v) BSA. After 16 h, cells were lysed with Tropix lysis buffer (ThermoFisher, Waltham, MA) and luciferase activity was measured in a Promega GloMax luminometer (Promega, Madison, WI). Luciferase activity was normalized to total protein levels measured by bicinchoninic acid (BCA) protein assay. Data were analyzed using Graphpad Prism 6 to calculate ligand activity (EC 50 ) and ligand inhibitory activity (IC 50 ) was fitted to the standard model (Graphpad, La Jolla, Calif.) The results are shown in Figures 6A-6B.

[0112] Time-resolved FRET assay The ligand requirement for the formation of complexes containing TβRI and TβRII was examined using the following purified proteins: TGF-β3, mTGF-β3, mmTGF-β2-7M, biotinylated TβRI-ΔC-Avi and TβRII-His. Initially, binary complexes of TGF-β3:TbRII-His (1:2), mTGF-β3:TβRII-His (1:1) and mmTGF-β2-7M:TβRII-His (1:1) were formed at 20 μM in 50 mM Tris buffer (pH 7.5) and stored at 4°C. A time-resolved fluorescence resonance energy transfer (TR-FRET) assay based on the proximity-dependent transfer of fluorescence from a donor terbium cryptate labeled with an anti-His mAb (Tb-anti-His, CisBio, Bedford, MA) to an acceptor XL665-labeled streptavidin (SA-665, CisBio, Bedford, MA) was used to monitor the association of the ternary ligand TβRII-His:biotinylated TβRI-ΔC-Avi complex. 50 μL assays containing 100 nM or 250 nM TGF-β3:TβRII-His (1:2), mTGF-β3:TβRII-His (1:1), and mmTGF-β2-7M:TβRII-His (1:1) complexes were incubated with 50 nM biotinylated TβRI-ΔC-Avi. Each 50 μl ternary complex formation assay also contained 2 nM Tb-anti-His and 30 nM SA-665 and was incubated for 2 h at room temperature. Six replicates were tested for each condition. Buffer controls (n=6) contained only 2 nM Tb-anti-His and 30 nM SA-XL665. Buffer conditions for each assay were 50 mM Tris, 50 mM NaCl, pH 7.5. Assays were performed in Corning black 384-well low flange microplates (ThermoFisher, Waltham, MA). After 2 h of incubation, terbium / XL-665 TR-FRET was measured in the assay plates on a BMG Labtech Pherastar FS multimode plate reader (BMG Labtech Inc., Cary, NC).An optical module equipped with 337, 490 and 665 nm filters was used to monitor TR-FRET generating raw data of 337 / 490 (terbium emission) and 337 / 665 (XL-665) emission. The ratio of 665 emission / 490 emission was measured for each condition and then used to calculate ΔF, a measure reflecting sample vs. background signal. ΔF was calculated using the following equation: (Ratio. signal -Ratio negative / Ratio negative )×100. Ratio signal Ratio refers to assays including ternary complex or buffer controls. negative refers to two assay buffer controls (2 nM Tb-anti-His and 30 nM SA-665). For the buffer controls, two of the six replicates were assigned as negative controls for the calculation of ΔF. ΔF was calculated using the remaining four replicates of the buffer controls. The results are shown in Figures 7A, 7B, and 14.

[0113] Analytical Ultracentrifugation mTGF-β3, mmTGF-β2 and mmTGF-β2-7M were analyzed by sedimentation velocity to determine the equilibrium constant for the self-association of monomeric TGF-β to form homodimers. mTGF-β3, mmTGF-β2 and mmTGF-β2-7M were each measured at 280 nm on an epon 2-channel centerpiece fitted with quartz windows and centrifuged at 20°C and 42,000 rpm for 27 hours in 15 mM sodium phosphate buffer containing 100 mM NaCl and adjusted to pH 3.8. 300 scans were collected in strong mode using a Beckman Optima XL-I analytical ultracentrifuge at the CAUMA Center of UTHSCSA. Data analysis was performed with UltraScan release 2130 (Demeler and Gorbet, Analytical ultracentrifugation data analysis with Ultrascan-III, Analytical Ultracentrifugation: Instrumentation, Software, and Applications (Uchiyama, S., Stafford, W., and Laue, T., eds.), pp. 119-143, Springer, 2016; Demeler et al., A comprehensive data analysis package for analytical ultracentrifugation experiments, 2016) and T Calculations were performed on a Lonestar-5 at the exas Advanced Computing Center. (Demeler, Curr Protoc Protein Sci Chapter 7, Unit 7 1 Sedimentation velocity data were first fitted to two-dimensional spectral analysis to remove temporally and radially invariant noise from the raw data and to fit the meniscus position as described in (Cao and Demeler, Biophys J 95:54-65, 2008) and the data were then analyzed using an adaptive space-time finite element method (Cao and Demeler, Biophys J 95:54-65, 2008) and a genetic algorithm for parameter optimization (Demeler et al., Macromol Biosci 10:775-782, 2010). The monomer-dimer model was used to fit the individual monomer-dimer models, which account for mass action and reversible association behavior, and K D (hydrological parameters) as well as partial specific volumes were fitted while assuming predicted molar masses for either wild type or mutants. Monte Carlo analysis (Demeler and Brookes, Colloid Polym Sci 286:129-137, 2008) was performed for each data set with 100 iterations to obtain fitting statistics. Buffer density and viscosity were estimated by UltraScan based on the buffer composition, and all hydrodynamic values ​​were corrected for standard conditions (20°C and water). The fitting resulted in excellent fits with random residuals and very low RMSD values. All results are summarized in Table 4 and Figures 11-13.

[0114] [Table 4]

[0115] result Design of engineered minimonomeric TGF-β (mmTGF-β) Structure of the TGF-β receptor complex (Groppe et al., Mol Cell 29:157-168, 2008; Radaev et al., J Biol Chem 285:14806-14814, 201 0 years) and associated binding and cross-linking studies by TGF-β3 C77S (Zuniga et al., J Mol Biol 354:1052-1068, 2005; Groppe et al., Mol Cell 29:157-168, 2008; Huang et al., EMBO J 30:1263-1276, 2011) suggested that the signaling capacity of monomeric TGF-β (TGF-β1 C77S or mTGF-β1, and TGF-β3 C77S or mTGF-β3) derives from their ability to non-covalently dimerize and thereby bind their receptors (Figures 1A and 1C). This gave rise to the hypothesis that it should be possible to reduce or completely eliminate receptor complex association with monomeric TGF-β by removing or altering residues involved in dimerization and binding of TβRI. The structural motif that perhaps contributes most significantly to monomeric self-association is the "heel" α-helix (α-helix 3) (Figure 1A). This helix is ​​highly amphipathic, with numerous hydrophobic interactions with residues lining the "palm" of the opposing monomer (Figure 1B). This helix also forms the majority of the binding surface for TβRI (Figure 1C). It was therefore hypothesized that removal of α-helix 3 would disrupt monomer self-association and TβRI binding, but would not impair TβRII binding, as this occurs through the ligand fingertips far from α-helix 3 (Figure 1A).

[0116] To evaluate this hypothesis, we generated bacterial expression constructs for TGF-β1, TGF-β2, and TGF-β3 in which residues 52-71 were removed and Cys-77 was replaced with serine. This corresponds to the deletion of all α-helix 3, as well as five flanking residues on the N-terminus and three flanking residues on the C-terminus (Figure 1D). The length of the deletion was chosen to leave a sufficient number of residues between the last residue of β-strand 4 (Gly-48) and the first residues of β-strand 5 (Cys-77 / Ser-77) to form an unconstrained loop bridging strands 4 and 5. As a secondary consideration, we also replaced two (TGF-β2) or three (TGF-β1 and -β3) loop-forming residues, increasing the overall net charge of the full-length TGF-β1, -β2, and -β3 monomers at pH 7.0 from -0.9, +1.1, and +4.4, respectively, in the constructs with deleted α-helix 3 to -3.1, +3.9, and +6.1 (Figure 1D). The implication from this is that the solubility of the monomers, which is low at pH 4.5-9.5 like the homodimers (see Figures 4A and 4B), can be improved both by removing the hydrophobic α-helix 3 and by artificially increasing the net charge at pH 7.0.

[0117] Isolation and physical characterization of mmTGF-β2 The above-mentioned "minimonomers" of TGF-β1, -β2 and -β3, designated mmTGF-β1, mmTGF-β2 and mmTGF-β3, were expressed in Escherichia coli and accumulated in the form of insoluble inclusion bodies. After isolating the inclusion bodies, reconstituting them in denaturing agents and purifying them, the minimonomers were refolded by dilution into CHAPS-containing buffer at pH 9.0 as previously described (Huang et al., Methods Mol Biol 1344:63-92, 2013). 6). Folding of the minimonomers differed significantly; the majority of mmTGF-β2 remained soluble during folding and large amounts of monomeric protein were produced after purification by cation exchange chromatography, whereas only small amounts of mmTGF-β1 and mmTGF-β3 remained soluble during folding and no (TGF-β1) and very little (TGF-β3) monomeric protein was obtained after purification by cation exchange chromatography. This pattern mirrors that previously observed for folding of TGF-β homodimers from full-length wild-type monomers (Huang et al., Methods Mol Biol 1344: 63-92, 2016) and possibly reflecting differences in the inherent propensity of the monomers to properly form the four intramolecular disulfides specific for each monomer. mmTGF-β2 was the least desirable variant due to its predicted low binding affinity for TβRII. However, this was considered an addressable issue based on previous studies that demonstrated that substitution of Lys-25, Ile-92, and Lys-94 of TGF-β2 with the corresponding residues in TGF-β1 and TGF-β3 conferred on TGF-β2 the ability to bind TβRII with high affinity (Baardsnes et al., Biochemistry 48:2146-2155). pp. 2009; De Crescenzo et al., J Mol Biol 355:47-62, 2006 year).

[0118] To determine whether mmTGF-β2 was suitable for further development of the above method, it was characterized with respect to folding, solubility and receptor binding properties. 15 2. Preparing N-labeled Samples and 2D 1 H- 15The protein was tested by recording N-shift correlation spectra (Figure 2A). The results showed a highly dispersed spectrum, typical of proteins folded in the native state. The spectrum could be fully assigned, and analysis of the assigned chemical shifts, which identifies secondary structure trends, showed that the protein has the expected secondary structure, especially in the palm region formed by the cysteine ​​knot and finger regions where TβRII binds (Figure 9A). The analysis further showed that the newly formed loop between residues 47 and 56 has a close to zero probability of forming an α-helix or β-strand, suggesting that this loop is probably flexible, as expected for a loop of this length connecting two antiparallel β-strands. This is due to the fact that the backbone 15 This was directly confirmed by analysis of the T2 values ​​of N. These values, which provide information on behavior on fast (nanoseconds-picoseconds) and intermediate (microseconds-milliseconds) time scales, were significantly increased in regions corresponding to newly created loops relative to the rest of the protein (Figure 9B), i.e., except for the N-terminus and the short loop connecting α-helix 1 and β-strand 1, which appear to be structurally ordered.

[0119] To directly examine the three-dimensional structure, mmTGF-β2 was crystallized and its structure was determined to 1.8 Å resolution using molecular replacement (Table 3). The overall fold of mmTGF-β2 was shown to be very similar to that previously determined for TGF-β2, except for a newly created loop, which was shown to replace the predicted α-helix 3 (Figure 2B). Superimposition of mmTGF-β2 with the monomer from the structure of TGF-β2 indicates that there is a systematic displacement of up to about 1.5 Å in the finger region of mmTGF-β2 compared to TGF-β2. Such differences are likely the result of bending of the monomer near the center of the finger region and not a change in the structure of the finger region, since superimposition of the fingers alone shows that they correspond closely, with a backbone root-mean-square deviation of less than 0.2 Å and similar orientation of the side chains of several residues that pack and stabilize the fingers (Figure 2D). Such bending is also supported by the overlap of two molecules of mmTGF-β2 present in the crystallographic asymmetric unit, which also show a small but still noticeable displacement of the finger regions relative to each other (Fig. 2C).Consistent with the NMR analysis, not only was the electron density in the region corresponding to the newly created loop noticeably weaker, but the two molecules from the asymmetric unit were shown to be oriented in different directions (Fig. 2C).

[0120] The similar folding of mmTGF-β2 to TGF-β2 suggested that it also binds to TβRII in a similar manner, especially in the TβRII-binding finger region. To assess this, surface plasmon resonance (SPR) experiments were performed in which the same concentration series of TβRII was injected onto TGF-β2 and mmTGF-β2 immobilized on separate flow cells (Figures 3A and 3B). Although affinity could not be quantified due to weak binding, the sensorgrams showed similar shapes and concentration dependences. These sensorgrams were consistent with previous reports (Baardsnes et al., Biochemistry 48:2146-2155, 2011). Consistent with previous studies (2009), we demonstrated that the binding between mmTGF-β2 and TβRII was weak and qualitatively similar to that of TGF-β2.

[0121] The solubility of mmTGF-β2 appears to be significantly better than that of TGF-β2 and full-length TGF-β2 monomer (mTGF-β2), as samples of it can be easily prepared at concentrations of 2-3 mg / ml at pH 7.0 without noticeable precipitation, whereas samples of the latter two proteins were completely precipitated in these same conditions. To quantify solubility, TGF-β2, mTGF-β2 and mmTGF-β2 were prepared as concentrated stocks in 100 mM acetic acid (pH 2.9), which readily dissolved, and then diluted in PBS (pH 7.4). Light scattering was measured at 340 nM to assess precipitation, and then samples were centrifuged and absorbance was measured at 280 nM to assess protein concentration. This demonstrated that both TGF-β2 and mTGF-β2 were effectively insoluble at neutral pH over the entire concentration range evaluated (7-100 μM) (Figures 4A and 4B). This is consistent with the known poor solubility of TGF-β homodimers (Pellaud et al., J Biol Chem 274:7699-7704, 1999), which indicates that this property extends to full-length monomeric TGF-β. Minimonomeric TGF-β2 (mmTGF-β2), in contrast, exhibited modest light scattering and a corresponding modest reduction in the amount of soluble protein compared to that expected at protein concentrations of 40 μM or higher, i.e., indicating that mmTGF-β2 is, in fact, fairly, if not completely, soluble at neutral pH. This is reflected in the NMR spectrum, which shows that at concentrations of 100-200 μM, 15Despite the ease of preparing N mmTGF-β2 samples, the spectra were weak, showing that only detectable signals arose from residues in the flexible parts of the protein, namely the N-terminus, the exposed loop between α-helix 1 and β-strand 1, and the newly created loop between β-strands 4 and 5. The fact that signals were only detectable from the flexible parts of the protein suggested that mmTGF-β2 formed extensive soluble aggregates in these conditions. By trial and error, it was found that these soluble aggregates could be removed by the addition of the zwitterionic detergent CHAPS, with most NMR signals appearing at a CHAPS concentration of 5 mM, and all NMR signals appearing at 10 mM CHAPS. Thus, all NMR spectra, including the one shown in Figure 2A, were recorded in the presence of 10 mM CHAPS.

[0122] Isolation and physical characterization of mmTGF-β2-7M The results presented above indicate that mmTGF-β2, despite its native folding, has an intrinsic low affinity for binding to TβRII. To confer to mmTGF-β2 the ability to bind to TβRII with high affinity comparable to TGF-β1 and TGF-β3, three residues of mouse TGF-β2 previously shown to differ at the interface with TβRII, Lys-25, Ile-92, and Asn-94 (De Crescenzo et al., J Mol Biol 355:47-62, 2006; Hart et al., Nat Struct Biol 9:203-208, 2002), were replaced with the corresponding residues Arg-25, Val-92, and Arg-94 from TGF-β1 and -β3 (Figures 1E and 1F). Substitution of these three residues has been shown in previous studies to be sufficient to confer TβRII binding affinity to TGF-β2 equivalent to that of TGF-β1 and TGF-β3 (Baardsnes et al., Biochemistry 48:2146-2155, 2009; De Crescenzo et al., J Mol Biol 355:47-62, 2006). Four additional residues around the TβRII-binding site, which are different in TGF-β2 compared to TGF-β1, were also replaced with the corresponding residues in TGF-β1 (R26K, L89V, T95K and I98V) (Figures 1E and 1F). Although previous results suggested that this was not strictly necessary, we checked whether the actual orientation of the residues in the mmTGF-β2-binding site relative to TβRII matched as closely as possible with that of the high affinity TGF-β isoforms TGF-β1 and TGF-β3. The resulting construct with these seven amino acid substitutions, designated mmTGF-β2-7M (Figures 1E, 9 and Table 1), was expressed in E. coli in the form of insoluble inclusion bodies. As with mmTGF-β2, most of the protein remained in solution after reconstitution and dilution into the native folding buffer, and large amounts of homogenous monomer could be isolated (4-5 mg / L in E. coli medium).

[0123] The folding and homogeneity of the isolated mmTGF-β2-7M was assessed by NMR and demonstrated that, like mmTGF-β2, the protein is a 2D 1 H- 15 It was found to have the expected number of signals in the N shift correlation spectrum (Figure 5A), as well as secondary structure as determined by analysis of the second shift in NMR (Figure 10A). The solubility of mmTGF-β2-7M was evaluated as before, and as shown, its behavior was comparable or perhaps slightly better than that of mmTGF-β2 (Figures 4C and 4D). However, this slight improvement in macroscopic solubility did not change the solubility under microscopy as in NMR analysis, i.e., it was still necessary to include 10 mM CHAPS in the sample buffer to detect signals of all backbone amide resonances in the protein.

[0124] The three-dimensional structure of mmTGF-β2-7M was determined crystallographically at 2.75 Å resolution (Table 3), and as previously described, the overall fold was maintained compared to TGF-β2, with the only difference being a slight hinge bend in the monomer as described in mmTGF-β2 (Figures 5B and 5C). 15 The increase in N T2 relaxation time was comparable to that of mmTGF-β2 (Fig. 10B), suggesting that the loss of density observed in some chain A and mostly chain C in the three molecules of the asymmetric unit in the regions corresponding to newly formed loops in mmTGF-β2-7M was not due to increased dynamics but to other factors, most likely the lower resolution of the structure of mmTGF-β2-7M compared to the mmTGF-β2 structure (Table 3).

[0125] To determine whether mmTGF-β2-7M binds to TβRII with high affinity, mmTGF-β2-7M and TGF-β3 variants were produced with an N-terminal avi tag and, after biotinylation and immobilization on a streptavidin-coated SPR sensor, kinetic SPR experiments were performed to measure their binding affinity to TβRII (Figures 3C and 3D). The resulting sensorgrams were very different from those previously obtained for mmTGF-β2 and TGF-β2, showing a clear pattern of saturation. The sensorgrams further showed similar shapes and K D We show that the fitted parameters, including K values ​​(Table 2), are within experimental error of each other and, albeit on the high end, of previously reported K values ​​for TβRII binding to TGF-β1 and TGF-β3. D This was consistent with values ​​from previous studies (Huang et al., EMBO J 30:1263-1276, 2011; Baardsnes et al., Biochemistry 48:2146-2155, 2009; De Crescenzo et al., J Mol Biol 355:47-62, 2006).

[0126] To determine whether the interactions that allow TβRII binding with high affinity were maintained in mmTGF-β2-7M compared to TGF-β1 and TGF-β3, we crystallized the mmTGF-β2-7M·TβRII complex and determined its structure at 1.88 Å resolution (Table 3). The overall structure of the mmTGF-β2-7M·TβRII complex was shown to be highly similar to that of a single TβRII-bound monomer from the structure of the TGF-β3·TβRII·TβRI complex, in which TβRII bound to the fingertips of mmTGF-β2-7M in a manner essentially indistinguishable from that of TGF-β3 (Fig. 5D). The interactions known to contribute most significantly to high affinity binding were further shown to be fully maintained in the mmTGF-β2-7M·TβRII complex compared to the previously determined TGF-β1·TβRII and TGF-β3·TβRII complexes (TGF-β3·TβRII complex determined to 1.8 Å (Hart et al., Nat Struct Biol 9:203-208, 2002) is the highest resolution structure determined to date (Figure 5E), which includes the packing of Ile-53 from TβRII in the hydrophobic pocket between the TGF-β fingers and the hydrogen bond ion pair formed between TGF-βArg-25 and Arg-94, respectively, on the tip of the loop connecting fingers 1 / 2 and 3 / 4, and the carboxylate groups of Glu-119 and Asp-32 on TβRII (Figure 5E).

[0127] Inhibitory activity of mmTGF-β2-7M and its underlying mechanisms The above results indicate that mmTGF-β2-7M possesses one of the essential attributes required to function as a dominant-negative inhibitor of TGF-β signaling, namely the ability to bind to TβRII with high affinity comparable to that of TGF-β1 and TGF-β3. To directly determine whether mmTGF-β2-7M can signal and, if not, whether it can function as an inhibitor, we performed CAGA 12TGF-β signaling was assessed by treating HEK293 cells stably transfected with a TGF-β luciferase reporter under the control of a promoter with increasing concentrations of TGF-β (Thies et al., Growth Factors 18:251-259, 2001). The results showed that dimeric TGF-β1 (TGF-β1) and full-length monomeric TGF-β3 (mTGF-β3) increased the luciferase response in a sigmoidal manner, with no further increase in luciferase response measured by concentrations of approximately 25 pM TGF-β1 and 250 pM mTGF-β3. This is consistent with previous reports showing that (full-length) monomeric TGF-β1 and -β3 were 5-15 fold less potent than their dimeric counterparts (Zuniga et al., J Mol Biol 354:1052-1068, 2005; Amatayakul-Chantler et al., J Biol Chem 269:27687-27691, 1994). Normalized luciferase responses could be readily fitted to a standard model of ligand-dependent activation, revealing ECs of 12.4±1.5 pM for TGF-β1 and 182±16 pM for mTGF-β3. 50 The values ​​for TGF-β1 and mTGF-β3 were previously reported by Amatayakul-Chantler et al. (J Biol Chem 269:27687-27691, 1994) for TGF-β1 and mTGF-β3, respectively. The potent subnanomolar signaling activity observed with TGF-β1 and mTGF-β3 was in contrast to that of mmTGF-β2-7M, which had no detectable signaling activity at concentrations that produced a saturation response for mTGF-β3 (approximately 200 pM) or up to four orders of magnitude higher (Figure 6A). Thus, mmTGF-β2-7M either completely lost signaling activity or had signaling activity that was 10,000-fold less potent than mTGF-β3.

[0128] To further characterize mmTGF-β2-7M, the same HEK293 luciferase reporter cell line was cultured at constant EC 50 We performed competition experiments stimulating with submaximal concentrations of dimeric TGF-β1 (8.0 pM) and increasing concentrations of mTGF-β3 or mmTGF-β2-7M, showing that mTGF-β3 further stimulated signaling with a midpoint concentration similar to that of mTGF-β3 alone (Figure 6B). 50 The EC value is 182 ± 16 pM for the data shown in Figure 6A. 50 , and an EC of 194 ± 36 pM for the data shown in Figure 6B . 50 confirms this. mmTGF-β2-7M behaved very differently, with no detectable change in signaling activity when added up to a concentration of 10 nM, and a sharp decline to no detectable signaling activity when the concentration was increased to 100 nM (Figure 6B). This indicates that mmTGF-β2-7M is indeed devoid of signaling activity and can function to completely block and inhibit TGF-β signaling. The normalized luciferase response could be easily fitted to a standard model of ligand-dependent inhibition, yielding an IC of 68 ± 7 nM. 50 Similar experiments demonstrated that mmTGF-β2-7M also functions as a potent competitive inhibitor of other TGF-β isoforms, TGF-β2 and TGF-β3, with measured IC 50 Value (IC value of TGF-β2 50 was 19 ± 3 nM, and the IC 50 The IC values ​​(21 ± 8 nM) were 2-3 fold higher than those measured for TGF-β1 (Figures 15A and 15B). 50The values ​​are at the lower end of the range of affinities reported for the binding of high affinity TGF-β isoforms to TβRII, including mmTGF-β2-7M reported herein (Table 2). This suggests that mmTGF-β2-7M functions to inhibit TGF-β signaling by binding to and blocking endogenous TβRII. The fact that the measured potency is stronger than previously reported to be the strongest binding affinity of TGF-β1 and TGF-β3 to TβRII (140 nM) (9) suggests that other factors (e.g., nonspecific association of mmTGF-β2-7M with the plasma membrane) may help enhance its inhibitory activity.

[0129] The finding that mmTGF-β2-7M has no apparent signaling activity and functions as an inhibitor of TGF-β signaling at low nanomolar concentrations suggests that removal of α-helix 3 reduces the noncovalent association of monomers and greatly attenuates or abolishes TβRI binding. To directly assess this, SPR experiments were performed to determine whether mmTGF-β2-7M can recruit TβRI in the presence of TβRII. To do this, increasing concentrations of TβRI and a series of the same TβRI concentrations in the presence of near-saturating amounts of TβRII (2 μM) were injected over the same TGF-β3 and mmTGF-β2-7M-bearing SPR chip surfaces used for the TβRII binding measurements above. This showed that TβRI alone bound only negligibly to TGF-β3 and mmTGF-β2-7M (Figures 3E and 3F), but unlike TGF-β3, mmTGF-β2-7M bound to TβRII cannot recruit TβRI (Figures 3G and 3H). This is consistent with previous results reported by Huang et al. (EMBO J 30:1263-1276 (2011)) in which mTGF-β3 bound to TβRII was significantly or completely impaired in its ability to bind and recruit TβRI. This also provides further evidence that TGF-β monomers bound to TβRII cannot bind and recruit TβRI, but because mmTGF-β2-7M was immobilized on the surface of the sensor, this result alone does not provide any insight into whether mmTGF-β2-7M can non-covalently dimerize and bind and recruit TβRI.

[0130] To directly address these questions, two solution-based techniques were used: analytical ultracentrifugation (AUC) and time-resolved fluorescence resonance energy transfer (TR-FRET). AUC experiments were performed by measuring the total UV absorbance at 280 nm as a function of radial position and time when mTGF-β3, mmTGF-β2 and mmTGF-β2-7M were precipitated in acidic conditions (pH 3.8) where the monomers were fully solubilized. The AUC data showed van Holde-Weischet sedimentation coefficient distribution plots that formed parabolically for all three monomers, consistent with each forming dimers or other higher-order oligomers by reversible self-association. To more precisely determine which species may be present in solution, finite element analysis was used to fit the data to the simplest model possible: distinct monomer-dimer equilibria. The fitting procedure resulted in a near perfect fit of all three monomers to a simple monomer-dimer model, as indicated by (a) the close overlap between the fitted curves (red) with the raw data after removal of temporally and radially invariant noise (black), and (b) the absence of any systematic deviations in the residuals (Figures 11-13). The fitted parameters were further refined to determine the K D was an order of magnitude higher for mTGF-β3 compared with mmTGF-β2 and mmTGF-β2-7M. Thus, although removal of the heel helix (α3) reduces the self-association of dimer-forming monomers, it does not completely abolish dimer formation.

[0131] Using TR-FRET, we assessed the ability of dimeric and monomeric TGF-β to bind and bring together TβRI and TβRII. This was done by generating tagged forms of TβRII and TβRI, respectively, and conjugating these tags with proteins labeled with fluorescent donors and acceptors. TβRII was C-terminally His-tagged and bound to a terbium cryptate-labeled anti-His monoclonal antibody fluorescent donor, and TβRI was N-terminally avi-tagged and, after enzymatic biotinylation, bound to a dye-labeled (XL-665) streptavidin fluorescent acceptor (Figure 7A). Addition of TGF-β to the tagged receptors brings them together, resulting in a large increase in the ΔF value, defined as the ratio of acceptor and donor emitted fluorescence intensities. The TR-FRET assay is illustrated by the data shown in Figure 14 and was used here to compare the ability of TGF-β3 full-length monomer, mTGF-β3, and TGF-β2 minimonomer (mmTGF-β2-7M), which binds to TβRII with high affinity, to bind and bring together TβRI and TβRII. The TR-FRET signal for mTGF-β3 showed comparable results to TGF-β3 and was independent of whether the TGF-β concentration was 100 or 250 nM (Figure 7B). The TR-FRET signal for mmTGF-β2-7M, in contrast, was within the error range of the buffer control and was independent of the TGF-β concentration (Figure 7B). These results indicate that under these conditions, mTGF-β3 retains all of its ability to form non-covalent dimeric complexes with TβRI and TβRII, whereas under these same conditions, mmTGF-β2-7M does not. These results, together with the AUC results, indicate that removal of the heel helix had the expected effect, in that its removal reduced but did not eliminate dimer formation, and even when dimers were formed, they were unable to bind and recruit TβRI.

[0132] Consideration TGF-β plays a role in promoting the progression of many human diseases (Dietz et al., Nature 2013). 352, pp. 337-339, 1991; Biernacka et al., Growth Factors, vol. 29, pp. 196-202, 2011; Massague, Cell, vol. 134, pp. 215-230, 2008; Loeys et al., Pediatr Endocrinol Rev, vol. 10, pp. 417-423, 2013). In the last 20 years or so, preclinical and clinical trials have shown that no inhibitors have been approved for human use. The results presented herein demonstrate that an engineered TGF-β monomer (deleting Cys-77 and the heel α-helix (α3)) exhibits IC 50 The results show that the novel inhibitor functions to block and inhibit TGF-β1, -β2 and -β3 signaling at high concentrations (Figure 6B and Figure 15). This novel inhibitor has several attributes that overcome the limitations of other types of inhibitors, such as the naturally high specificity of TGF-β, and thus an inhibitor against TβRII can confer much higher specificity to TGF-βII compared to the much more promiscuous TGF-β kinase inhibitors, thus reducing undesirable side effects. Furthermore, the small size of the inhibitor (approximately 10 kDa) greatly enhances the ability of TGF-β to penetrate tumors and other dense tissues that enhance disease progression, a distinct advantage compared to IgG antibodies, which are much larger (approximately 150 kDa) and tend to occupy only the interstitial spaces of blood vessels and highly perfused organs (Meibohm (201 2), Therapeutic Proteins: Strategies to Modulate Their Plasma Half-Lives (Kontermann R. ed.), pp. 23-38, Wiley-Blackwell, Weinheim, Germany, Meibohm and Raeckman (2008), "Pharmacokinetics and Pharmacodynamics of Peptides and Protein Drugs", Pharmaceutical Biotechnology: Fundamentals and Applications (Crommelin DJA, Sindelar RD, and Meibohm B. (Eds.) , pp. 95-123, Informa Healthcare, New York). Other advantages of this novel inhibitor include its high inherent stability, due to the four intramolecular disulfide bonds that hold the four fingers together and the fact that, unlike the native TGF-β dimer or full-length TGF-β monomer, it is highly soluble in water at neutral pH.

[0133] The structure of the TGF-β receptor complex, taken together with previously published chemical cross-linking data, suggested that the potent signaling activity of TGF-β1 C77S and TGF-β3 C77S results from the ability of monomers to non-covalently dimerize and then associate into (TβRI·TβRII)2 heterotetramers. The results of this example, the AUC experiments used to assess non-covalent dimer formation, and the TR-FRET experiments used to assess complex association with TβRI and TβRII, provided further evidence of this. The AUC data showed that full-length monomeric TGF-β3, mTGF-β3, self-associated and formed dimers with a dimerization constant of 4.1 μM (Table 4). The TR-FRET data showed that mTGF-β3, at concentrations of 0.1 or 0.25 μM, associated TβRI·TβRII complexes to the same extent as dimeric TGF-β3 in the presence of appropriate concentrations of TβRI and TβRII ectodomains (Figure 7B). This suggests that the mTGF-β3 concentrations (0.1–0.25 μM, Figure 7B) are the K for self-association. D The fact that this occurs at conditions over an order of magnitude lower than the K value (4.1 μM, Table 4) indicates that receptor binding also significantly participates in the assembly of the TβRI·TβRII complex. Thus, the association of the TβRI·TβRII complex with mTGF-β3 and possibly also with mTGF-β1 appears to be a cooperative process (much like protein folding) in which multiple weak interactions, including monomer-monomer, non-covalent dimer-receptor and receptor-receptor interactions, cooperate to allow the formation of a thermodynamically stable TGF-β·TβRI·TβRII complex. This mode of cooperative association is likely due to the monomeric self-association K of mTGF-β1 and mTGF-β3. D More than four orders of magnitude lower than the value (EC 50 K value of approximately 0.1 nM vs. K for self-association D This is due to its ability to induce signal transduction at concentrations (value 4.1 μM).

[0134] Removal of the heel helix from TGF-β monomers was shown to be highly effective in terms of preventing cooperative association of the TβRI·TβRII complex, as shown by the TR-FRET data (Figure 7B) and cell-based signaling data (Figures 6A and 6B). The AUC data showed that removal of the heel helix resulted in a weakening of the monomer-monomer interaction by an order of magnitude (Table 4). The SPR data shown in Figures 3G and 3H further indicated that the TβRII-bound form of mmTGF-β2-7M was unable to bind and recruit TβRI, which would be expected based on reported structures of the TGF-β receptor complex, which show binding of TβRI to both chains of TGF-β as well as a composite interface formed by TβRII (Groppe et al., Mol Cell 29:157-168, 2008; Radaev et al., Mol Cell 29:157-168, 2008). (E. et al., J Biol Chem, vol. 285, pp. 14806-14814, 2010). The data indicate that the low tendency of the engineered monomers to self-associate, combined with the expected very weak binding of TβRI to any dimers formed, accounts for the inability of mmTGF-β2-7M to assemble TβRI·TβRII complexes, explaining the lack of signaling activity and, together with its retention of high affinity binding to TβRII, explaining its inhibitory activity.

[0135] Other family of type II receptors, activin type II receptor II, activin type II B receptor, BMP type II receptor and anti-Mullerian hormone type II receptor, have been shown or are predicted to bind GF knuckles and not GF fingertips, similar to TβRII (Hinck et al., Cold Spring Harb Perspect Biol 8:a022103, 2016). Nevertheless, they are not associated with all type I receptors. It shares the same property as TβRII of binding only by contacting residues from a single GF monomer, not both monomers, as has been shown or predicted in a family of receptors (Hinck et al., Cold Spring Harb Perspect Biol 8:a022103, (2016). This, taken together with the structures reported here where we show that α3 can be removed without affecting the overall structure of the monomer (Figures 2B-2D and 5B-5E), suggests that other GF monomers of the family may be made that lack the heel helix to function as inhibitors. Inhibitors of this type have a wide range of potential applications, from research tools to analyze the role of specific ligands in vivo, to clinically useful inhibitors to treat diseases that are driven by hyperactive signaling by other ligands of the family, such as activin-induced cancer cachexia (Coerver et al., Mol Endocrinol 10:534-543, 1996).

[0136] In view of the many possible embodiments to which the principles of the disclosed invention may be applied, it should be recognized that the illustrated examples are merely preferred examples of the invention and are not intended to limit the scope of the invention, which is rather defined by the following claims. It is therefore the sole object of the present invention to claim all that comes within the scope and spirit of these claims.

Claims

1. A recombinant transforming growth factor (TGF)-β monomer that blocks TGF-β signaling, comprising: (i) a cysteine ​​to serine substitution at an amino acid residue corresponding to residue 77 of SEQ ID NO:2; (ii) a deletion of the α3 helix corresponding to amino acid residues 52-71 of SEQ ID NO:2; and (iii) at least one amino acid substitution relative to a wild-type TGF-β monomer that increases the net charge of the recombinant TGF-β monomer, the amino acid substitution comprising alanine, valine, leucine, isoleucine, methionine, phenylalanine, tryptophan, serine, threonine, tyrosine, asparagine, glutamine, glycine, proline, aspartic acid, glutamic acid, or histidine to arginine or lysine; and Including, having at least 90% sequence identity with SEQ ID NO: 8 or 10; Recombinant TGF-β monomer.

2. 2. The recombinant TGF-β monomer of claim 1, further comprising at least one amino acid substitution compared to wild-type TGF-β2 monomer that increases affinity of the recombinant TGF-β monomer for the TGF-β type II receptor (TβRII), comprising a lysine to arginine substitution at the amino acid residue corresponding to residue 25 of SEQ ID NO:2, an arginine to lysine substitution at the amino acid residue corresponding to residue 26, a leucine to valine substitution at the amino acid residue corresponding to residue 89, an isoleucine to valine substitution at the amino acid residue corresponding to residue 92, a lysine to arginine substitution at the amino acid residue corresponding to residue 94, a threonine to lysine substitution at the amino acid residue corresponding to residue 95, or an isoleucine to valine substitution at the amino acid residue corresponding to residue 98 of SEQ ID NO:

2.

3. The recombinant TGF-β monomer of claim 1 or claim 2, which is a human TGF-β2 monomer.

4. The at least one amino acid substitution that increases the net charge of the monomer is A leucine to arginine substitution at an amino acid residue corresponding to residue 51 of SEQ ID NO:2; An alanine to lysine substitution at an amino acid residue corresponding to residue 73 of SEQ ID NO:2; or Both a leucine to arginine substitution at the amino acid residue corresponding to residue 51 and an alanine to lysine substitution at the amino acid residue corresponding to residue 73 of SEQ ID NO:

2.

4. The recombinant human TGF-β2 monomer of claim 3, comprising:

5. 5. The recombinant human TGF-β2 monomer of claim 3 or claim 4, wherein the at least one amino acid substitution that increases the affinity of the monomer for TβRII comprises: lysine to arginine at residue 25 of SEQ ID NO:2, arginine to lysine at residue 26, leucine to valine at residue 89, isoleucine to valine at residue 92, asparagine to arginine at residue 94, threonine to lysine at residue 95, isoleucine to valine at residue 98, or any combination of two or more thereof.

6. 6. The recombinant human TGF-β2 monomer of claim 5, comprising a lysine to arginine substitution at residue 25, an arginine to lysine substitution at residue 26, a leucine to valine substitution at residue 89, an isoleucine to valine substitution at residue 92, an asparagine to arginine substitution at residue 94, a threonine to lysine substitution at residue 95, and an isoleucine to valine substitution at residue 98.

7. A recombinant human TGF-β2 monomer according to any one of claims 3 to 6, comprising the amino acid sequence of SEQ ID NO:8 or SEQ ID NO:

10.

8. The recombinant TGF-β monomer of any one of claims 1 to 7, which is PEGylated.

9. A recombinant TGF-β monomer according to any one of claims 1 to 7, which is glycosylated or hyperglycosylated.

10. A fusion protein comprising a recombinant TGF-β monomer according to any one of claims 1 to 9 and a heterologous protein.

11. The fusion protein of claim 10 , wherein the heterologous protein comprises a protein tag.

12. The fusion protein of claim 10 , wherein the heterologous protein comprises an Fc domain.

13. The fusion protein of claim 10 , wherein the heterologous protein comprises albumin or an albumin-binding polypeptide.

14. The fusion protein of claim 10 , wherein the heterologous protein comprises an antibody, an antigen-binding fragment thereof, or a targeting moiety.

15. The fusion protein of claim 10 which is a single polypeptide chain.

16. The fusion protein of claim 10 which forms a dimeric polypeptide.

17. The fusion protein of claim 10 which is a heterodimer.

18. The fusion protein of claim 10 which is a multimer.

19. A recombinant TGF-β monomer according to any one of claims 1 to 9 or a fusion protein according to any one of claims 10 to 18, comprising a radiotherapeutic agent, a cytotoxic agent for chemotherapy, or a drug.

20. A recombinant TGF-β monomer according to any one of claims 1 to 9 or a fusion protein according to any one of claims 10 to 18, comprising an imaging agent, a fluorescent dye or a fluorescent protein tag.

21. A composition comprising a recombinant TGF-β monomer according to any one of claims 1 to 9, or a fusion protein according to any one of claims 10 to 18, and a pharma- ceutically acceptable carrier, diluent, or excipient.

22. A composition comprising a recombinant TGF-β monomer according to any one of claims 1 to 9 or a fusion protein according to any one of claims 10 to 18, or a composition according to claim 21, for use in a method of inhibiting TGF-β signaling in a cell, the method comprising contacting the cell with the recombinant TGF-β monomer, the fusion protein or the composition.

23. The composition of claim 22, wherein the method is an in vitro method.

24. 23. The composition of claim 22, wherein the method is an in vivo method comprising administering the recombinant TGF-β monomer, fusion protein or composition to a subject having a disease or disorder associated with aberrant TGF-β signaling.

25. A composition comprising a recombinant TGF-β monomer according to any one of claims 1 to 9 or a fusion protein according to any one of claims 10 to 18, or a composition according to claim 21, for treating a disease or disorder associated with abnormal TGF-β signaling in a subject.

26. The composition of claim 24 or claim 25, wherein the disease or disorder associated with aberrant TGF-β signaling is a fibrotic disorder.

27. 26. The composition of claim 24 or claim 25, wherein the disease or disorder associated with aberrant TGF-β signaling is breast cancer, brain cancer, pancreatic cancer, prostate cancer or skin cancer.

28. The composition of claim 24 or claim 25, wherein the disease or disorder associated with aberrant TGF-β signaling is an eye disease.

29. 26. The composition of claim 24 or claim 25, wherein the disease or disorder associated with aberrant TGF-β signaling is an inherited disorder of connective tissue.

30. An isolated nucleic acid molecule encoding a recombinant TGF-β monomer according to any one of claims 1 to 9.

31. 31. The nucleic acid molecule of claim 30 operably linked to a promoter.

32. A vector comprising the nucleic acid molecule of claim 30 or claim 31.

33. 33. An isolated cell comprising a nucleic acid molecule according to claim 30 or claim 31, or a vector according to claim 32.

34. 34. The isolated cell of claim 33, which is a T lymphocyte.

35. A composition for treating a disease or disorder associated with aberrant TGF-β signaling in a subject, the composition comprising an isolated cell according to claim 33 or claim 34.

36. The composition of claim 35, wherein the disease or disorder associated with aberrant TGF-β signaling is a fibrotic disorder.

37. 36. The composition of claim 35, wherein the disease or disorder associated with aberrant TGF-β signaling is breast cancer, brain cancer, pancreatic cancer, prostate cancer or skin cancer.

38. The composition of claim 35, wherein the disease or disorder associated with aberrant TGF-β signaling is an ocular disease.

39. 36. The composition of claim 35, wherein the disease or disorder associated with aberrant TGF-β signaling is an inherited disorder of connective tissue.

Images