BR112015023905B1 - USES OF AN ANTIBODY OR AN ANTIGEN-BINDING FRAGMENT THEREOF WHICH BINDS TRANSFORMING GROWTH FACTOR BETA IN THE TREATMENT OF OSTEOGENESIS IMPERFECTA - Google Patents

Abstract

USES OF AN ANTIBODY OR ANTIGEN-BINDING FRAGMENT THEREOF WHICH BINDS TRANSFORMING GROWTH FACTOR BETA IN THE TREATMENT OF OSTEOGENESIS IMPERFECTA. The present invention provides methods for treating and ameliorating symptoms of osteogenesis imperfecta (OI) in an individual by administering to the individual a therapeutically effective amount of a binding agent that binds to transforming growth factor beta (TGFbeta).

Description

Related Orders

[001] This application is related to Provisional Patent Application No. 61/803,647, filed March 20, 2013, U.S. Provisional Patent Application At the. 61/875,399, filed on September 9, 2013, and Provisional Patent Application No. 61/883,151, filed October 26, 2013, the contents of which are each incorporated herein by reference in their entirety.

Government Financial Support

[002] This invention was made with government support under P01 HD070394, P01 HD22657& R01 DE01771 granted by the ‘National Institutes of Health’. The United States government has certain rights in this invention.

Field of Invention

[003] The present invention relates to methods for treating osteogenesis imperfecta (osteogenesis imperfecta, OI). More specifically, the invention relates to methods for treating OI using a binding protein, e.g., an antibody or antigen-binding fragment thereof, that specifically binds to human transforming growth factor beta (TGFbeta) or isomers thereof.

Background

[004] Osteogenesis imperfecta (OI), also known as "brittle bone disease" or Lobstein syndrome, is a rare and debilitating congenital bone disease that affects one in 15,000 people. Although phenotypes vary between types of OI, common symptoms include incomplete ossification of bones and teeth, reduced bone mass, brittle bones, and pathological fractures. These common symptoms of OI are thought to be caused by genetic mutations that result in deficiencies in type I collagen or other proteins involved in bone matrix deposition or homeostasis. As a result of these symptoms and the propensity for fractures and fatal bone complications, the life expectancy of patients with OI is reduced when compared to the general population. Correspondingly, there is clearly an urgent need in the art to develop effective treatments for OI.

Brief Description of Figures

[005] Figures 1A (Western blot) and 1B-1F (graphs) demonstrate that TGFbeta signaling is elevated in bones derived from Crtap-/- mice when compared to wild-type controls.

[006] Figures 2A (luminescent image) and 2B-C (graphs) show that Crtap-/- mice crossed with TGFbeta reporter mice exhibit higher TGFbeta activity compared to wild-type controls.

[007] Figure 3 is a μCT image of vertebrae derived from Crtap-/- mice treated with mouse pan-specific anti-TGFbeta antibody 1D11.

[008] Figure 4 is a graph containing quantitative measurements of vertebrae derived from Crtap-/- mice treated with mouse pan-specific anti-TGFbeta antibody 1D11.

[009] Figure 5 is a histomorphometric analysis of vertebrae from Crtap-/- mice treated with mouse pan-specific anti-TGFbeta antibody 1D11.

[0010] Figures 6A and 6B are graphs demonstrating the biochemical properties (Figure 6A - maximum load, and Figure 6B - stiffness) of femurs derived from Crtap-/- mice treated with the mouse pan-specific anti-TGFbeta antibody 1D11, as determined by a three-point bending test.

[0011] Figures 7A and 7B (micrographic images) and 7C (graph) demonstrate the lung phenotype of Crtap-/- mice treated with the mouse pan-specific anti-TGFbeta antibody 1D11.

[0012] Figure 8 are micrographic images of wild-type and Crtap-/- lung stained with an anti-decorin antibody.

[0013] Figure 9 is a graph showing a decorin binding assay.

[0014] Figure 10 is a group of graphs, photos, and images showing increased TGFbeta signaling in Crtap-/- mice. Figure 10A is a series of three graphs showing the results of quantitative RT-PCR of TGFbeta target genes p21, PAI-1, and Col1a1. Graphs indicate increased TGFbeta signaling in the calvarial bone of P3 wild-type and Crtap-/- mice. Results are shown as 'fold changes' multiples of the wild-type group mean±SD; n=5 per group, *p<0.05. Figure 10B is a photograph of a Western blot analysis of calvarial P3 protein extracts, which shows increased amounts of activated Smad2 (pSmad2) relative to total Smad2 protein in Crtap-/- mice versus wild-type mice, which suggest increased TGFbeta signaling; n=3 per group. Figure 10C is a graph showing the quantification of the Western blot shown in Figure 10B. Results are shown as multiples of changes from wild-type±SD group mean, *p<0.05. Figure 10D is an image showing increased bioluminescence in regions that overlap skeletal structures in Crtap-/- mice compared to wild-type mice that were interbred with TGFbeta reporter mice (SBE-Luc mice). Representative images of 3 litters at P10 are shown. In 3 litters of Crtap -/- mice show an average bioluminescence signal of 2.86 times (SD±0.34) in the head/calvaria compared to wild-type mice (scale bar =1 cm). Figure 10E is a graph showing, using a TGFbeta reporter cell line, TGFbeta activity was evaluated in conditioned medium derived from wild-type and Crtap-/- bone marrow stromal cells cultured under osteogenic conditions for 3 days, demonstrating greater TGFbeta activity compared to medium derived from wild-type cells. Results are shown as multiples of changes from wild-type group mean±SD, n=5 per group, *p<0.05. Figure 10F is two images showing immunostaining of lungs (P10) for pSmad2, which show increased intracellular staining in wild-type and Crtap -/- mice (40X magnification). Representative images of n=3 mice per group are shown (scale bar=20 µm).

[0015] Figure 11 is a group of photos and graphs showing phenotypic correction of Crtap-/- mice after treatment with the TGFbeta-neutralizing antibody 1D11. Figure 11A is a series of three MicroCT images of L4 vertebral bodies from 6-week-old wild-type (wild-type) mice, control antibody-treated Crtap-/- mice, and 1D11-treated Crtap-/- mice after treatment for 8 weeks. Figure 11B is a group of three graphs showing MicroCT results of L4 vertebral bodies demonstrating increased bone volume/total volume (BV/TV), trabecular number (Tb.N), and thickness (Tb.Th) in wild-type mice, Crtap-/-e control mice, and 1D11-treated Crtap-/- mice. Results are shown as mean±SDs, n=8 per group, *p<0.05 for Crtap-/- vs. 1D11. Crtap-/- control, +p<0.05 for Crtap-/- vs. wild type. Figure 11C is a group of three graphs showing the results of histomorphometric analysis of L4 vertebrae, which show increased osteoclast (N.Oc/BS) and osteoblast (N.Ob/BS) numbers per bone surface in Crtap -/- mice compared to wild type. Reduced osteoblast and osteoclast number after 1D11 treatment indicates effective suppression of accelerated bone remodeling in Crtap-/- mice. Increased numbers of osteocytes per area of bone (N.Ot/B.Ar) in Crtap-/- mice are reduced to the wild-type level after treatment with 1D11. Results are shown as mean±SDs, n=6 per group, *p<0.05 for 1D11 of Crtap-/- vs. Crtap-/- control, +p<0.05 for Crtap-/- vs. wild type. Figure 11D is a series of three images showing hematoxylin/eosin staining of inflated lungs from 1 6-week-old wild-type (wild-type) mice, Crtap-/- control mice, and Crtap-/- mice treated as 1D11 after treatment for 8 weeks. control Crtap-/- mice show an increase in distal airway space compared to wild-type mice. After treatment with 1D11, there is a reduction in the diameter of the distal airways compared to control Crtap-/- mice. Representative images of n=8 mice per group are shown (scale bar=100 µm). Figure 11E is a graph showing quantification of distance between alveolar structures by the mean linear intersection (MLI) method, which demonstrates a significant reduction in distal airway space in Crtap-/- mice treated as 1D11 compared to Crtap-/- treated with control antibody and wild-type mice. Results are shown as mean±SDs, n=8 mice per group, 10 analyzed images per mouse, *p<0.05 for 1D11 Crtap-/- vs. Crtap-/- control, +p<0.05 for Crtap-/- vs. wild type.

[0016] Figure 12 is a series of graphs showing that decorin binding to type I collagen overlaps the P986 3Hyp site and is reduced in type I collagen from Crtap-/- mice. Figure 12A is a group of three graphs showing the results of quantitative RT-PCR of calvarial bone from P3 mice, which shows the difference in RNA expression of small leucine-rich proteoglycans decorin (Dcn), biglycan (Bgn), and asporin (Aspn) in the calvarial bone of Crtap-/- mice compared to wild-type. Results are given as multiples of changes from wild-type ± SD group mean, n = 5 per group. Figure 12B is a graph showing the results of surface plasmon resonance analysis, which indicates that binding of recombinant decorin core protein to type I collagen from Crtap -/- mice is about 45% less compared to type I collagen from wild type mice. Three independent experiments using 3, 5, and 12 µM decorin were performed. Response units (RU) of total amount of decorin bound normalized to type I collagen immobilized on the chip are shown. Mean reduction of decorin binding to Crtap-/- type I collagen is 44.6 ± 7.9%.

[0017] Figure 13 is a series of graphs and photos showing that inhibition of enhanced TGFbeta signaling improves bone phenotype in a mouse model of dominant OI that results from a G610C mutation in the Col1a2 gene (Col1α2tm1.1Mcbr). Figure 13A comprises two graphs showing the results of quantitative RT-PCR of p21 and PAI-1 TGFbeta target genes, which indicate increased TGFbeta signaling in the calvarial bone of wild-type P3 mice and Col1α2tm1.1Mcbr mice. Results are shown as multiples of changes from wild-type±SD group mean; n = 3 per group, *p<0.05. Figure 13B is a photo of Western blot analysis results, which show increased levels of activated Smad2 (pSmad2) relative to total Smad2 protein levels in P3 calvarium of wild-type mice and Col1α2tm1.1Mcbr mice compared to wild-type, which suggest increased TGFbeta signaling; n=3 per group. Figure 13C is a graph showing the quantification of the Western blot seen in Figure 13B. Results are shown as multiples of changes from wild-type±SD group mean; *p<0.05. Figure 13D is a photo series of MicroCT images of L4 vertebral bodies of 16-week-old wild-type (wild-type) mice, control antibody-treated Col1α2tm1.1Mcbr mice, and 1D11-treated Col1α2tm1.1Mcbr mice after treatment for 8 weeks. Figure 13E is a series of graphs of data from MicroCT of L4 vertebral bodies showing increased bone volume/total volume (BV/TV), trabecular number (Tb.N) and thickness (Tb.Th) in Col1α2tm1.1Mcbr mice after 1D11 treatment. Results are shown as mean±SDs, n=6 per group, *p<0.05 for Col1α2tm1.1Mcbr 1D11 vs. Control Col1α2tm1.1Mcbr, +p<0.05 for Col1α2tm1.1Mcbr vs. wild type.

[0018] Figure 14 is a graph of the weights curve, which shows a reduced weight of Crtap-/- mice compared to wild-types during the study period (p<0.05 for all time points, mean±SDs are shown). No statistically significant difference in weight was observed in Crtap-/- mice treated as 1D11 compared to control Crtap-/- mice.

[0019] Figure 15 is a series of graphs and tables showing no effect of TGFbeta inhibition on post-translational modification of abnormal type I collagen in Crtap-/- mice. Figure 15A is a series of three graphs showing tandem mass spectra of type I collagen extracted from the tibia of wild-type mice, control Crtap-/- mice, and 1D11-treated Crtap-/- mice (16 week old mice, after treatment for 8 weeks). The sequence at the top of the graph is SEQ ID NO: 19, the sequence in the middle of the graph is SEQ ID NO: 20, and the sequence at the bottom of the graph is SEQ ID NO: 21. Figure 15B is a table showing a summary of analyzes of tandem mass spectra. 1D11 treatment did not significantly affect the 3-hydroxylation status of Pro986 alpha 1(I) collagen residue in the bone samples. Shown is the mean percentage of 3-hydroxy residues (±SD), n=5 per group. Figure 15C is a group of three graphs showing that type I collagen from bone from Crtap-/- control mice and Crtap-/- mice treated with 1D11 exhibit changes in cross-linking levels of hydroxysilyl pyridinoline (HP) and lysyl pyridinoline (LP) and an increased HP/LP ratio compared to wild-type mice. 1D11 treatment of Crtap-/- mice did not significantly affect these parameters compared to control Crtap-/- mice. Results are given as mean±SDs, n = 4 mice per group, +p<0.05 for Crtap-/- vs. wild type.

[0020] Figure 16 is a series of graphs showing the serum bone turnover markers osteocalcin (OCN) and bone collagen C-terminal cross-linked telopeptide (CTX) at baseline (Figure 16A = 8 weeks old) and at the end of the treatment study (Figure 16B = 16 weeks old). Figure 16A is two graphs showing increased serum levels of OCN CTX in 8 week old Crtap -/- mice compared to wild-type mice at baseline indicate increased bone turnover in Crtap -/- mice. Results are given as mean±SDs, n=8 for wild type, n=14 for Crtap -/- mice, +p<0.05 for Crtap -/- vs. wild-type mice. Figure 16B is two graphs showing that 16 week old Crtap -/- mice treated as 1D11 show a trend towards reduced serum OCN levels and significantly reduced serum CTX levels compared to control Crtap -/- mice, indicating increased bone turnover suppression by TGFbeta inhibition. Results are given as mean±SDs, n=8 for wild type, n=7 for Crtap-/- group; *p<0.05 for Crtap-/- 1D11 vs. Crtap-/- control, +p<0.05 for Crtap-/- vs. wild type.

[0021] Figure 17 is a table showing the results of MicroCT analysis of L4 vertebral body of wild-type mice, control Crtap-/- mice, and 1D11-treated Crtap-/- mice (16-week-old mice, after treatment for 8 weeks). Mean±SDs are shown for bone volume/tissue volume (BV/TV), trabecular number (Tb.N), trabecular thickness (Tb.Th), trabecular separation (Tb.Sp), and bone volume bone mineral density (BMD BV); n=8 per group, + indicates one way Kruskal-Wallis ANOVA in classes where the test of equal variance failed. n.s.=not statistically significant.

[0022] Figure 18 is a table showing the results of MicroCT analysis of trabecular bone in proximal femurs for wild-type mice, control Crtap-/- mice, and 1D11-treated Crtap-/- mice (16-week-old mice, after treatment for 8 weeks). Mean±SDs are shown for bone volume/tissue volume (BV/TV), trabecular number (Tb.N), trabecular thickness (Tb.Th), trabecular separation (Tb.Sp), and bone volume bone mineral density (BMD BV); n=8 per group. + indicates a Kruskal-Wallis ANOVA way in the classes where the equal variance test failed. n.s.=not statistically significant.

[0023] Figure 19 is a table showing the results of MicroCT analysis of cortical bone in the central shaft of the femur for wild-type mice, control Crtap-/- mice, and 1D11-treated Crtap-/- mice (16-week-old mice, after treatment for 8 weeks). Mean±SDs are shown for cortical thickness, bone mineral density of bone volume (BMD BV), anteroposterior diameter, cross-sectional area (CSA), and cross-sectional moments of inertia (CSMI) for the mediolateral axis (m.l.) and anteroposterior axis (a.p.); n=8 per group. n.s.=not statistically significant.

[0024] Figure 20 is a table showing the results of mechanical testing of femurs by 3-point flexion (16-week-old mice, after treatment for 8 weeks). Compared with wild-type mice, control Crtap-/- mice exhibit significantly reduced biochemical parameters except elastic modulus and elastic displacement. Anti TGFbeta treatment with 1D11 resulted in significant improvements in maximum load and maximum force in Crtap -/- femurs, indicating increased tissue and whole bone resistance. However, no significant change in post-yield displacement was observed, indicating that 1D11 did not affect the increased bone fragility of OI. N=6 for wild type mice, n=4 for control Crtap -/- mice and n=3 for 1D11 treated Crtap -/- mice. n.s.=not statistically significant.

[0025] Figure 21 is a table showing the results of histomorphometric analysis of L4 vertebral bodies from wild-type mice, control Crtap-/- mice, and 1D11-treated Crtap-/- mice (16-week-old mice, after treatment for 8 weeks). Mean±SDs are shown for bone volume/tissue volume (BV/TV), trabecular number (Tb.N), trabecular thickness (Tb.Th), trabecular separation (Tb.Sp), number of osteoclasts/bone surface (N.Oc/BS), osteoclast surface/bone surface (Oc.S/BS), number of osteoblasts/bone surface (N.Ob/BS), osteoblast surface/bone surface bone (Oc.S/BS), and number of osteocytes/bone area (N.Ot/B.Ar); n=6 per group. + indicates a Kruskal-Wallis ANOVA way in the classes where the equal variance test failed. n.s.=not statistically significant.

[0026] Figure 22 is a table showing the results of MicroCT analysis of L4 vertebral body of wild-type mice, Col1α2tm1.1Mcbr control mice and Col1α2tm1.1Mcbr mice treated with 1D11 (16 week old mice, after treatment for 8 weeks). Mean±SDs are shown for bone volume/tissue volume (BV/TV), trabecular number (Tb.N), trabecular thickness (Tb.Th), trabecular separation (Tb.Sp), and bone volume bone mineral density (BMD BV); n=6 per group, n.s.=not statistically significant.

[0027] Figure 23 is a graph showing surface plasmon resonance analysis measuring recombinant decorin core protein binding to type I collagen of wild-type mice and Crtap-/- mice. Three technical copies at each of the indicated concentrations of decorin were made from two independent biological copies (♦ copy 1, ▲ copy 2). Results are shown as a percentage of the wild-type mean (bars indicate mean by group).

[0028] Figure 24 (micrographic images) demonstrates immunostaining for decorin in the distal femur metaphysis of wild-type mice and Crtap-/- mice at 20X, n=3 per genotype, scaled bar=100μm (Figures 24A-24C) and 40X magnification, n=3 per genotype, scaled bar=50μm (Figures 24) D-24F). Control femurs were incubated in the secondary antibody only.

[0029] Figure 25 (micrographic images) demonstrates immunostaining for TGFβ1 in the distal femur metaphysis of wild-type mice and Crtap-/- mice at 20X, n=3 per genotype, scaled bar=100 μm (Figures 25A-25C) and 40X magnification, n=3 per genotype, scaled bar = 50 μm (Figures 2) 5D-25F). Control femurs were incubated in the secondary antibody only.

[0030] Figure 26 (micrographic images) demonstrates immunostaining for TGFβ1 in the distal femur metaphysis of wild-type mice and Col1α2tm1.1Mcbr mice at 20X, n=3 per genotype, scaled bar=100 μm (Figures 26A-26C) and 40X magnification, n=3 per genotype, scaled bar=50 µm (figures 26D-26F). Control femurs were incubated in the secondary antibody only. Summary of the Invention

[0031] The present invention relates to a method for the effective treatment of osteogenesis imperfecta (OI). More specifically, the invention relates to methods for treating OI using a binding protein such as an antibody or an antigen-binding fragment thereof that specifically binds transforming growth factor beta (TGFbeta) or an isoform thereof. Preferably, the binding protein is pan-specific and binds to all three human isoforms of TGFbeta, i.e. TGFβ1, TGFβ2 and TGFβ3. More preferably, the binding protein specifically binds and neutralizes human TGFβ1, TGFβ2, and TGFβ3. In one aspect, the invention provides a method for treating OI in a subject in need thereof comprising administering to the subject a therapeutically effective amount of an antibody or antigen-binding fragment thereof that especially binds to TGFbeta.

[0032] In one embodiment, the antibody or antigen-binding fragment thereof comprises a heavy chain variable region comprising three complementarity determining regions (CDRs) having amino acid sequences selected from the group consisting of SEQ ID NOs: 4, 5, and 6; and a light chain variable region comprising three CDRs having amino acid sequences selected from the group consisting of SEQ ID NOs: 7, 8, and 9.

[0033] In another embodiment, the antibody or antigen-binding fragment thereof comprises a heavy chain variable region comprising the amino acid sequence of SEQ ID NO: 10, and a light chain variable region comprising the amino acid sequence of SEQ ID NO: 11.

[0034] In one embodiment, the antibody or antigen-binding fragment thereof further comprises a human IgG4 constant region. In one embodiment, the human IgG4 constant region comprises the amino acid sequence of SEQ ID NO: 12. In another embodiment, the antibody or antigen-binding fragment thereof further comprises the human k light chain constant region. In another embodiment, the human k light chain constant region comprises the amino acid sequence of SEQ ID NO: 13. In another embodiment, the antibody or antigen-binding fragment thereof further comprises a human IgG4 constant region, and a human k light chain constant region.

In another embodiment, the human IgG4 constant region comprises the amino acid sequence of SEQ ID NO: 12, and the human k light chain constant region comprises the amino acid sequence of SEQ ID NO: 13. In another embodiment, the antibody comprises a heavy chain comprising the amino acid sequence of SEQ ID NO: 14. In another embodiment, the antibody comprises a light chain comprising the amino acid sequence of SEQ ID NO: 15. the antibody comprises a heavy chain comprising the amino acid sequence of SEQ ID NO: 14, and a light chain comprising the amino acid sequence of SEQ ID NO: 15.

[0036] In another embodiment, the antibody or antigen-binding fragment thereof binds to human TGFβ1, TGFβ2, and TGFβ3. In another embodiment, the antibody or antigen-binding fragment thereof neutralizes human TGFβ1, TGFβ2, and TGFβ3.

[0037] In another embodiment, the antibody or antigen-binding fragment thereof improves a bone parameter selected from the group consisting of bone volume density (BV/TV), total bone surface (BS), bone surface density (BS/BV), trabecular number (Tb.N), trabecular thickness (Tb.Th), trabecular spacing (Tb.Sp), and total volume (Dens TV).

[0038] In another embodiment, the antibody or antigen-binding fragment thereof inhibits bone resorption.

[0039] In another embodiment, the antibody or antigen-binding fragment thereof reduces a serum biomarker of bone resorption selected from the group consisting of urinary hydroxyproline, urinary total pyridinoline (PYD), urinary free deoxypyridinololine (DPD), urinary type I collagen crosslinked N-telopeptide (NTX), serum or urinary collagen type I crosslinked N-telopeptide (CTX), sialo bone protein (BSP), osteopontin (OPN), and tartrate-resistant acid phosphatase 5b (TRAP).

[0040] In another embodiment, the antibody or antigen-binding fragment thereof increases a serum biomarker of bone deposition selected from the group consisting of such as total alkaline phosphatase, bone-specific alkaline phosphatase, osteocalcin, and type I procollagen (C-terminal/N-terminal).

[0041] In another embodiment, the antibody or antigen-binding fragment thereof inhibits bone resorption. In another embodiment, the antibody or antigen-binding fragment thereof promotes bone deposition. In another embodiment, the antibody or antigen-binding fragment thereof improves the function of a non-skeletal organ affected by OI selected from the group consisting of auditory function, lung function, and kidney function.

[0042] In another aspect, the invention provides a method for treating OI in a subject in need thereof comprising administering to the subject a therapeutically effective amount of an antibody or antigen-binding fragment thereof that binds to TGFbeta, wherein the antibody comprises a heavy chain comprising the amino acid sequence of SEQ ID NO: 14, and a light chain comprising the amino acid sequence of SEQ ID NO: 15.

[0043] In another aspect, the invention provides a method for treating OI in a subject in need thereof comprising administering to the subject a therapeutically effective amount of an antibody or antigen-binding fragment thereof that binds to TGFbeta in combination with at least one therapeutic agent. In another embodiment, the agent is a bisphosphonate. Detailed Description of the Invention A. Definitions

[0044] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art.

[0045] It is noted here that as used in this report and appended claims, the singular forms "an "an", "the" and "the" also include plural reference, unless the context clearly dictates otherwise.

[0046] The term "about" or "approximately" means within 10%, and more preferably within 5% (or 1% or less), of a given value or range.

[0047] The terms "administer" or "administration" refer to the act of injecting or otherwise physically delivering a substance when it exists outside the body (e.g., an antibody) into a patient, such as by mucosal delivery, intradermal, intravenous, subcutaneous, intramuscular, and/or any other method of physical delivery described herein or known in the art. When a disease, or a symptom thereof, is being treated, administration of the substance typically occurs after the onset of the disease or its symptoms. When a disease or its symptoms are being prevented, administration of the substance typically occurs before the onset of your disease or its symptoms.

[0048] A TGFbeta "antagonist" or "inhibitor" refers to a molecule that is capable of inhibiting or otherwise decreasing one or more of the biological activities of TGFbeta, such as in a cell that expresses TGFbeta or in a cell that expresses a TGFbeta ligand, or that expresses a TGFbeta receptor. In certain exemplary embodiments, antibodies of the invention are antagonist antibodies that inhibit or otherwise decrease TGFbeta activity in a cell having a TGFbeta receptor expressed on the cell surface (e.g., TGFbeta receptor 1, 2, or 3) when said antibody is brought into contact with said cell. In some embodiments, a TGFbeta antagonist (e.g., an antibody of the invention) can, for example, act by inhibiting or otherwise decreasing the activation and/or cell signaling pathways of the cell expressing a TGFbeta receptor, thereby inhibiting a TGFbeta-mediated biological activity of the cell relative to TGFbeta-mediated biological activity in the absence of antagonist. In certain embodiments of the invention, the anti-TGFbeta antibodies are fully human, monoclonal, antagonistic anti-TGFbeta antibodies, antagonistic anti-TGFbeta antibodies.

[0049] The terms "antibody", "immunoglobulin", or "Ig" may be used interchangeably herein. The term antibody includes, but is not limited to, synthetic antibodies, monoclonal antibodies, recombinantly produced antibodies, multispecific antibodies (including bispecific antibodies), human antibodies, humanized antibodies, chimeric antibodies, intrabodies, single chain Fvs (scFv) (e.g., including monospecific, bispecific, etc.), camelized antibodies, Fab fragments, F(ab') fragments, disulfide-linked Fvs (sdFv), anti- -idiotypic (anti-Id), and epitope-binding fragments of any of the above. In particular, antibodies include immunoglobulin molecules and immunologically active portions of immunoglobulin molecules, i.e., antigen-binding domains or molecules that contain an antigen-binding site that binds especially to a TGFbeta antigen (e.g., one or more complementarity determining regions (CDRs) of an anti-TGFbeta antibody). Anti-TGFbeta antibodies can be of any type (e.g., IgG, IgE, IgM, IgD, IgA, and IgY), any class (e.g., IgG1, IgG2, IgG3, IgG4, IgA1, and IgA2), or any subclass (e.g., IgG2a and IgG2b) of an immunoglobulin molecule. In certain embodiments, the anti-TGFbeta antibodies are humanized, such as humanized monoclonal anti-TGFbeta antibodies. In other embodiments, the anti-TGFbeta antibodies are fully human, such as fully human monoclonal anti-TGFbeta antibodies. In preferred embodiments, the anti-TGFbeta antibodies are IgG antibodies, such as IgG4 antibodies.

[0050] The terms "composition" and "formulation" are intended to include a product containing specific ingredients (e.g., an anti-TGFbeta antibody) in optionally specified amounts, as well as any product that results, directly or indirectly, from combining specific ingredients in optionally specified amounts.

[0051] The terms "constant region" or "constant domain" refer to a carboxy-terminal portion of the light and heavy chain that is not directly involved in antibody binding to antigen, but exhibits various effector functions, such as interaction with the Fc receptor. The terms refer to the portion of an immunoglobulin molecule having a more conserved amino acid sequence relative to the other portion of the immunoglobulin, the variable domain, which contains the antigen-binding site. The constant domain contains the CH1, CH2, and CH3 domains of the heavy chain, and the CHL domain of the light chain.

[0052] The term "epitope" refers to a region located on the surface of an antigen, such as a TGFbeta polypeptide or TGFbeta polypeptide fragment, which is capable of being bound to one or more antigen-binding regions of an antibody, and which has antigenic or immunogenic activity in an animal, preferably a mammal, and most preferably in a human, which is capable of eliciting an immune response. An epitope having immunogenic activity is a portion of a polypeptide that elicits an antibody response in an animal. An epitope having antigenic activity is a portion of a polypeptide to which an antibody specifically binds, as determined by any method well known in the art, for example, such as an immunoassay. Antigenic epitopes need not necessarily be immunogenic. Epitopes usually consist of chemically surface-active groupings of molecules, such as amino acids or sugar side chains, and have specific three-dimensional structural characteristics as well as specific charge characteristics. A region of a polypeptide that contributes to an epitope can be adjacent amino acids of the polypeptide or the epitope can be joined from two or more non-adjacent regions of the polypeptide. The epitope may or may not be a three-dimensional surface feature of the antigen. In certain embodiments, a TGFbeta epitope is a three-dimensional surface feature of a TGFbeta polypeptide (e.g., in a trimeric form of a TGFbeta polypeptide). In other embodiments, a TGFbeta epitope has the linear characteristic of a TGFbeta polypeptide (e.g., in a dimeric form or monomeric form of the TGFbeta polypeptide). Anti-TGFbeta antibodies can especially bind to an epitope of the monomeric form of TGFbeta, an epitope of the dimeric form of TGFbeta, or both the monomeric form and the dimeric form of TGFbeta.

[0053] The term "excipients" refers to inert substances that are commonly used as a diluent, vehicle, preservative, binder, stabilizing agent, etc. for drugs and includes, but is not limited to, proteins (e.g., serum albumin, etc.), amino acids (e.g., aspartic acid, glutamic acid, lysine, arginine, glycine, histidine, etc.), fatty acids and phospholipids (e.g., alkyl sulfonates, caprylate, etc.), surfactants (e.g., SDS, polysorbate, nonionic surfactant, etc.), saccharides (e.g., sucrose, maltose , trehalose, etc.) and polyols (eg mannitol, sorbitol, etc.). See also Remington's Pharmaceutical Sciences (1990) Mack Publishing Co., Easton, Pa., which is incorporated herein by reference in its entirety.

[0054] In the context of a peptide or a polypeptide, the term "fragment" refers to a peptide or a polypeptide that comprises less than the full-length amino acid sequence. Such a fragment may originate, for example, from an amino-terminal truncation, a carboxy-terminal truncation, and/or an internal deletion of a residue(s) of the amino acid sequence. Fragments may, for example, result from alternative RNA splicing or from in vivo protease activity. In certain embodiments, fragments of TGFbeta include polypeptides comprising an amino acid sequence of at least 50 to 100 amino acid residues, at least 125 adjacent amino acid residues, at least 150 adjacent amino acid residues, at least 175 adjacent amino acid residues, at least 200 adjacent amino acid residues, or at least 250 adjacent amino acid residues of the amino acid sequence of a TGFbeta polypeptide. In a specific embodiment, a fragment of a TGFbeta polypeptide or an antibody that specifically binds a TGFbeta antigen retains at least 1, at least 2, or at least 3 functions of the full-length antibody or polypeptide.

[0055] The terms "fully human antibody" or "human antibody" are used interchangeably herein and refer to an antibody comprising a human variable region, and more preferably a human constant region. In specific embodiments, the terms refer to an antibody comprising a variable region and constant region of human origin. "Fully human" anti-TGFbeta antibodies, in certain embodiments, can also include antibodies that bind TGFbeta polypeptides and are encoded by nucleic acid sequences that are naturally occurring somatic variants of a human germ line immunoglobulin nucleic acid sequence. In a specific embodiment, the anti-TGFbeta antibodies are fully human antibodies. The term "fully human antibody" includes antibodies having constant and variable regions that correspond to human germ line immunoglobulin sequences as described by Kabat et al., (See Kabat et al., (1991) Sequences of Proteins of Immunological Interest, 5th edition, U.S. Department of Health and Human Services, NIH Publication No. 91-3242). Methods of producing fully human antibodies are known in the art.

[0056] The phrase "recombinant human antibody" includes human antibodies that are prepared, expressed, created, or isolated by recombinant means, such as antibodies expressed using a recombinant expression vector transfected into a host cell, antibodies isolated from a human recombinant combinatorial antibody library, antibodies isolated from an animal (e.g., a mouse or cow) that is transgenic and/or transchromosomal for human immunoglobulin genes (see, for (e.g., Taylor, L.D. et al., (1992) Nucl. Acids Res. 20:6287-6295) or antibodies prepared, expressed, raised, or isolated by any other means that involves splicing human immunoglobulin gene sequences or other DNA sequences. Such recombinant human antibodies can have constant and variable regions derived from human germ line immunoglobulin sequences (See Kabat, E.A. et al., (1991) Sequences of Proteins of Immunological Interest, 5th edition, U.S. Department of Health and Human Services, NIH Publication No. 91-3242). In certain embodiments, however, such recombinant human antibodies undergo in vitro mutagenesis (or, when an animal transgenic for human Ig sequences is used, in vivo somatic mutagenesis) and thus the amino acid sequences of the VH and VL regions of the recombinant antibodies are sequences which, while derived from and related to human germ line VH and VL sequences, may not naturally exist within the human germ line repertoire antibody in vivo.

[0057] The term "heavy chain" when used in reference to an antibody refers to five distinct types, named alpha (α), delta (Δ), epsilon (ε), gamma (Y) and mi (μ), based on the amino acid sequence of the heavy chain constant domain. These distinct types of heavy chains are well known in the art and give rise to five classes of antibodies, IgA, IgD, IgE, IgG, and IgM, respectively, including four subclasses of IgG, namely IgG1, IgG1, IgG3, and IgG4. Preferably the heavy chain is a human heavy chain.

[0058] An "isolated" or "purified" antibody is substantially free of cellular material or other contaminating proteins from the cell or tissue source from which the antibody is derived, or substantially free of chemical precursors or other chemicals when chemically synthesized. The language "substantially free of cellular material" includes preparations of an antibody in which the antibody is separated from cellular components of the cells from which it is isolated or recombinantly produced. Thus, an antibody that is substantially free of cellular material includes antibody preparations having less than about 30%, 20%, 10%, or 5% (by dry weight) heterologous protein (also called a "contaminant protein" herein). When the antibody is recombinantly produced, it is also preferably substantially free of culture medium, i.e., culture medium represents less than about 20%, 10%, or 5% by volume of the protein preparation. When the antibody is produced by chemical synthesis, it is preferably substantially free of chemical precursors or other chemicals, that is, it is separated from chemical precursors or other chemicals that are involved in protein synthesis. Correspondingly, such antibody preparations have less than about 30%, 20%, 10%, 5% (by dry weight) precursor chemicals or compounds other than the antibody of interest. In a preferred embodiment, anti-TGFbeta antibodies are isolated and purified.

[0059] The terms "Kabat numbering," and similar terms are art-recognized and refer to a numbering system for amino acid residues that are more variable (i.e., hypervariable) than other amino acid residues in the light and heavy chain variable regions of an antibody, or an antigen-binding portion thereof (Kabat et al., (1971) Ann. NY Acad. Sci. 190:382-391 and, Kabat et al., (199) 1) Sequences of Proteins of Immunological Interest, 5th edition, U.S. Department of Health and Human Services, NIH Publication No. 91-3242). For the heavy chain variable region, the hypervariable region typically ranges from amino acid positions 31 to 35 for CDR1, amino acid positions 50 to 65 for CDR2, and amino acid positions 95 to 102 for CDR3. For the light chain variable region, the hypervariable region typically ranges from amino acid positions 24 to 34 for CDR1, amino acid positions 50 to 56 for CDR2, and amino acid positions 89 to 97 for CDR3.

[0060] The term "light chain" when used in reference to an antibody refers to two distinct subtypes, called kappa (K) from lambda (À), based on the amino acid sequence of the constant domains. Light chain amino acid sequences are well known in the art. In preferred embodiments, the light chain is a human light chain.

[0061] The terms "manage", "management", and "management" refer to the beneficial effects that an individual derives from a therapy (eg, a therapeutic or prophylactic agent) that does not result in a cure for the disease or disorder. In certain embodiments, a subject is administered one or more therapies (e.g., therapeutic or prophylactic agents) to "manage" a TGFbeta-mediated disease (e.g., OI), or one or more symptoms thereof, in order to prevent progression or worsening of the disease.

[0062] The term "monoclonal antibody" refers to an antibody obtained from homogeneous or substantially homogeneous antibodies, and each monoclonal antibody will typically recognize a single epitope on the antigen. In preferred embodiments, a "monoclonal antibody" is an antibody produced by a single hybridoma or other cell. The term "monoclonal" is not limited to any particular method for forming the antibody. For example, monoclonal antibodies can be made by the hybridoma method as described in Kohler et al.; Nature, 256:495 (1975) or can be isolated from phage libraries. Other methods for preparing clonal cell lines and monoclonal antibodies expressed in this way are well known in the art (see, for example, chapter 11 in: Short Protocols in Molecular Biology, (2002) 5th Ed.; Ausubel et al., eds., John Wiley and Sons, New York).

[0063] The term "pharmaceutically acceptable" means that it is approved by a federal or state government regulatory agency or listed on the U.S. Pharmacopeia, European Pharmacopeia, or Pharmacopeia generally recognized for use in animals, and more particularly in humans.

[0064] The term "pharmaceutically acceptable excipient" means any inert substance that is combined with an active molecule, such as a monoclonal antibody, for the preparation of a pleasant or convenient dosage form. The "pharmaceutically acceptable excipient" is an excipient that is non-toxic to recipients at the dosages and concentrations employed, and is compatible with other ingredients of the formulation comprising the monoclonal antibody.

[0065] The terms "prevent", "preventing", and "prevention" refer to the total or partial inhibition of the development, recurrence, onset, or spread of a TGFbeta-mediated disease and/or symptom related thereto, which results from the administration of a therapy or a combination of therapies provided herein (e.g., a combination of therapeutic or prophylactic agents).

[0066] The term "TGFbeta antigen" refers to that portion of a TGFbeta polypeptide to which an antibody specifically binds. A TGFbeta antigen also refers to an analogue or derivative of a TGFbeta polypeptide or fragment thereof to which an antibody specifically binds. In some embodiments, a TGFbeta antigen is a monomeric TGFbeta antigen or a dimeric TGFbeta antigen. A region of a TGFbeta polypeptide that contributes to an epitope can be adjacent amino acids of the polypeptide, or the epitope can come together from two or more non-adjacent regions of the polypeptide. The epitope may or may not be a three-dimensional surface feature of the antigen. A region located on the surface of a TGFbeta antigen that is capable of eliciting an immune response is a TGFbeta epitope. The epitope may or may not be a three-dimensional surface feature of the antigen. As used herein, a TGFbeta antigen "analog" refers to a polypeptide that has a similar or identical function to that of a TGFbeta polypeptide, a fragment of a TGFbeta polypeptide, or a TGFbeta epitope described herein. For example, the analogue can comprise a sequence that is at least 80%, at least 85%, at least 90%, at least 95%, or at least 99% identical to the amino acid sequence of a TGFbeta polypeptide (e.g., SEQ ID NO: 1, 2, or 3), a fragment of a TGFbeta polypeptide, a TGFbeta epitope, or an anti-TGFbeta antibody described herein. Additionally or alternatively, the polypeptide is encoded by a nucleotide sequence that hybridizes under stringent conditions to a nucleotide sequence encoding a TGFbeta polypeptide, a fragment of a TGFbeta polypeptide, or a TGFbeta epitope described herein.

[0067] The term "human TGFbeta," "hTGFbeta," or "hTGFbeta polypeptide" and similar terms refer to the polypeptides ("polypeptides", "peptides" and "proteins" are used interchangeably herein) that comprise the amino acid sequence of SEQ ID NO: 1, 2, or 3, and related polypeptides, including SNP variants thereof. Related polypeptides include allelic variants (eg, SNP variants); union variants; fragments; derivatives; substitution variants, deletion variants, and insertion variants; fusion polypeptides; and preferably interspecies homologues that retain TGFbeta activity and/or are sufficient to generate an anti-TGFbeta immune response. Also encompassed are soluble forms of TGFbeta that are sufficient to generate an anti-TGFbeta immune response. As those skilled in the art will appreciate, an anti-TGFbeta antibody can bind a TGFbeta polypeptide, polypeptide fragment, antigen, and/or epitope, since an epitope is part of the larger antigen, which is part of the larger polypeptide fragment, which, in turn, is part of the larger polypeptide. hTGFbeta can exist in a dimeric or monomeric form.

[0068] The terms "TGFbeta-mediated disease" and "TGFbeta-mediated disorder" are used interchangeably and refer to any disease or disorder that is completely or partially caused by or is the result of TGFbeta, e.g., hTGFbeta. In certain embodiments, TGFbeta is aberrantly expressed. In some embodiments, TGFbeta can be aberrantly up-regulated in a particular cell type. In other embodiments, normal, aberrant, or excessive cell signaling is caused by binding of TGFbeta to a TGFbeta receptor. In certain embodiments, the TGFbeta receptor (e.g., TGFbeta receptor 1, 2, or 3), is expressed on the surface of a cell, such as an osteoblast, osteoclast, or marrow stromal cell. In certain embodiments, the TGFbeta-mediated disease is a degenerative bone disease, such as osteogenesis imperfecta.

[0069] The terms "specially binds to" or "specially binds to" mean to specially bind to an antigen or a fragment thereof (eg, TGFbeta) and not specifically to bind to other antigens. An antibody that specifically binds to one antigen may bind to other peptides or polypeptides with lesser affinity, as determined by, for example, radioimmunoassays (RIA), enzyme-linked immunosorbent assays (ELISA), BIACORE, or other assays known in the art. In certain embodiments, an anti-TGFbeta antibody of the invention can specifically bind TGFbeta (e.g., hTGFbeta) with more than twice as much affinity as a different, non-TGFbeta antigen. Antibodies or variants or fragments thereof that bind especially to an antigen can be reactive with related antigens. For example, in certain embodiments an anti-TGFbeta antibody that cross-reacts with hTGFbeta and another TGFbeta antigen (e.g., a rodent or non-human primate TGFbeta antibody). Preferably, antibodies or variants or fragments thereof that especially bind to an antigen do not cross-react with other non-TGFbeta antigens. An antibody or variant or fragment thereof that specifically binds to a TGFbeta antigen can be identified, for example, by immunoassays, BIAcore, or other techniques known to those skilled in the art. Typically a specific or selective reaction will be at least twice background noise or signal, and more typically more than 10 times background. In some embodiments, the binding protein or antibody will bind its antigen, for example TGFbeta, with a dissociation constant of between 1x10-6 M and 1x10-7. In other embodiments, the dissociation constant is between 1x10-6 M and 1x10-8. See, for example, Paul, ed., 1989, Fundamental Immunology Second Edition, Raven Press, New York at pages 332-336 for a discussion regarding antibody specificity.

[0070] The terms "individual" and "patient" are used interchangeably. As used herein, a subject is preferably a mammal, such as a non-primate (e.g., cows, pigs, horses, cats, dogs, rats, etc.) or a primate (e.g., ape and human), more preferably a human. In one embodiment, the subject is a mammal, preferably a human, having a TGFbeta-mediated disease. In another embodiment, the subject is a mammal, preferably a human, at risk of developing a TGFbeta-mediated disease.

[0071] The term "therapeutic agent" refers to any agent that can be used in the treatment, management, or amelioration of a TGFbeta-mediated disease and/or a symptom related thereto. In certain embodiments, the term "therapeutic agent" refers to a TGFbeta antibody. In certain other embodiments, the term "therapeutic agent" refers to an agent other than a TGFbeta antibody. Preferably, a therapeutic agent is an agent that is known to be useful for, or has been, or is currently being used for the treatment, management, or amelioration of a TGFbeta-mediated disease, or one or more symptoms related thereto.

[0072] The term "therapy" refers to any protocol, method, and/or agent that can be used in the prevention, management, treatment, and/or amelioration of a TGFbeta-mediated disease (eg, OI). In certain embodiments, the terms "therapies" and "therapy" refer to biological therapy, supportive therapy, and/or other therapies useful in preventing, managing, treating, and/or ameliorating a TGFbeta-mediated disease known to those of skill in the art, such as medical personnel.

[0073] The terms "treat", "treatment", and "treating" refer to reducing or ameliorating the progression, severity, and/or duration of a TGFbeta-mediated disease (e.g., OI) that results from the administration of one or more therapies (including, but not limited to, the administration of one or more therapeutic or prophylactic agents). In specific embodiments, such terms refer to reducing or inhibiting TGFbeta binding to a TGFbeta receptor, reducing or inhibiting TGFbeta production or secretion from a cell that expresses a TGFbeta receptor from an individual, reducing or inhibiting TGFbeta production or secretion from a cell that does not express a TGFbeta receptor from an individual, and/or inhibiting or reducing one or more symptoms associated with a TGFbeta-mediated disease, such as OI.

[0074] The terms "variable region" or "variable domain" refer to a portion of the light and heavy chains, typically about the amino-terminal 120 to 130 amino acids in the heavy chain and about 100 to 110 amino acids in the light chain, that differ extensively in sequence between antibodies and are used in the binding and specificity of each particular antibody for its particular antigen. Sequence variability is concentrated in those regions called complementarity determining regions (CDRs), while the most highly conserved regions in the variable domain are called framework regions (FR). The light and heavy chain CDRs are primarily responsible for antibody-antigen interaction. Numbering of amino acid positions is according to the EU index, as in Kabat et al., (1991) Sequences of proteins of immunological interest. (U.S. Department of Health and Human Services, Washington D.C.) 5th ed. ("Kabat and others"). In preferred embodiments, the variable region is a human variable region.

B. Osteogenesis Imperfecta (OI)

[0075] OI includes a group of congenital bone disorders characterized by deficiencies in one or more proteins involved in bone matrix deposition or homeostasis. There are eight types of OI that are defined by their specific genetic mutation, and the resulting protein deficiency and phenotype of the affected individual. Although phenotypes vary between types of OI, common symptoms include incomplete ossification of bones and teeth, reduced bone mass, brittle bones, and pathological fractures.

[0076] Type I collagen is one of the most abundant connective tissue proteins in both calcified and non-calcified tissues. Exact synthesis, post-translational modification, and secretion of type I collagen are necessary for proper tissue development, maintenance, and repair. Most mutations identified in individuals with OI result in reduced synthesis of type I collagen, or incorrect synthesis and/or processing of type I collagen.

[0077] In addition to mutations in the type I collagen gene, other mutations in genes that participate in intracellular trafficking and processing of collagens have been identified in individuals affected with OI. These genes include molecular chaperones such as FK506 binding protein 10 (FKBP10) and heat shock protein 47 (HSP47) (Alanay et al., 2010; Christiansen et al., 2010; Kelley et al., 2011). Additional mutations have been identified in intermolecular collagen cross-linking genes, such as procollagen-lysine, 2-oxoglutarate 5-dioxygenase 2 (PLOD2), and in members of the collagen polyyl hydroxylase family of genes, including leucine-enriched proteoglycan (leprecan) (LEPRE1), peptidylprolyl isomerase B (cyclophilin B) (CYPB), and cartilage-associated protein (CRTAP) (Mor ello et al., 2006; Cabral et al., 2007; Baldridge et al., 2008; van Dijk et al., 2009; Choi et al., 2009; Barnes et al., 2010; Pyott et al., 2011). Mutations aside, proteins such as bone morphogenetic protein (BMP) and transforming growth factor β (TGFbeta) and their respective receptors are thought to participate in various OI phenotypes, although the extract mechanisms of their actions are unknown (Gebken et al., 2000).

[0078] In one embodiment, expression of TGFbeta is regulated by molecules that bind type I and type II collagen. In certain embodiments, a small leucine-rich proteoglycan (SLRP) regulates TGFbeta expression. In a specific embodiment, decorin regulates TGFbeta synthesis. In a certain embodiment, decorin does not bind type I or type II collagen in which the 3-hydroxyproline site is absent at position 986 of the type I and/or type II collagen molecules.

C. Bone Biology

[0079] The skeleton of vertebrates is made up of bone, which is a living, calcified tissue that provides structure, support, protection, and a source of minerals for the regulation of ion transport. Bone is a specialized connective tissue that is made up of both cellular and acellular components. The acellular extracellular matrix (ECM) contains both collagen and non-collagen proteins, both of which participate in the calcification process. A correctly secreted and aligned ECM is critical for proper bone formation. Pathology results when any of the ECM proteins are absent, malformed, or misaligned, as it is in osteogenesis imperfecta.

[0080] The term "cortical bone" or "compact bone" refers to the outer layer of bone, which is dense, rigid, and tough. The term "trabecular bone" or "spongy bone" is the porous inner layer of bone, which is lighter and less dense than cortical bone. The term "trabecular" refers to the microscopic structural unit of porous bone, which is of a bar-like shape and collagenous composition.

[0081] Bone is a dynamic tissue that undergoes constant remodeling. The term "osteoblast" refers to a terminally differentiated bone-forming cell that deposits an osteoid. The term "osteoid" refers to immature, demineralized bone that is composed primarily of type I collagen. The term "pre-osteoblast" refers to proliferating immature osteoblasts that are not fully differentiated. The term "osteoprogenitor" refers to a pluripotent cell that gives rise to various types of stromal cells, including osteoblasts. Osteoprogenitor cells, which are commonly called "mesenchymal stem cells" originate in the bone marrow and can be isolated in small numbers from the circulating blood. The term "osteoclast" refers to a terminally differentiated bone-resorting cell that is descended from a bone marrow monocyte. Osteoclasts can be identified by their expression of tartrate-resistant acid phosphatase (TRAP).

[0082] Under normal homeostatic conditions, osteoblasts and osteoclasts work in unison to maintain bone integrity. Pathology results when bone deposition and bone resorption become uncoupled. For example, osteopetrosis is a bone disease characterized by hard, overly dense bone that is a result of unresponsive osteoclasts, while osteoporosis is a bone disorder characterized by brittle, porous bones that can result from increased osteoclast activity. Evidence suggests that osteoclast activity may be increased in osteogenesis imperfecta, implicating these cell types as a potential target for therapeutic intervention. The present disclosure includes methods of inhibiting osteoclasts with a TGFbeta antibody.

[0083] Several methods can be used to measure and characterize bone structure, density, and quality, including histology and histomorphometry, atomic force microscopy, confocal Raman microscopy, nanoindentation, three-point curve test, X-ray imaging, and computed microtomography (μ-CT). In an exemplary embodiment, bones are measured and characterized by at least one of these methods.

[0084] The term "bone volume density" refers to the fraction of a given bone volume (total volume or TV) that is made up of calcified material (bone volume or BV). Therefore, bone volume density is calculated as BV/TV and is reported as a percentage. The term "specific bone surface" refers to the total bone surface (BS) per given volume of bone. Therefore, specific bone surface is calculated as BS/TV. Other common bone measurements include: bone area (B.Ar); trabecular number (Tb.N); trabecular spacing (Tb.Sp); N.Oc (osteoclast number); Oc.S (osteoclast surface area); Oc.S/BS; osteoblast number (N.Ob), osteoblast surface area (Ob.S), osteoblast perimeter (Ob.Pm), and derived from any of said measurements. A higher Oc.S/BS is an indicator of increased bone resorption by osteoclasts. D. Transforming growth factor beta (TGFbeta)

[0085] TGFβs are multifunctional cytokines that are involved in cell proliferation and differentiation, embryonic development, extracellular matrix formation, bone development, wound healing, hematopoiesis, and immune and inflammatory responses (Roberts et al., 1981; Border et al., 1995a). Secreted TGFbeta protein is cleaved into a latency-associated peptide (LAP) and a mature TGFbeta peptide, and is found in both latent and active forms. The mature TGFbeta peptide forms both homodimers and heterodimers with other members of the TGFbeta family. TGFbeta can be purified from any natural source, or it can be produced synthetically (eg, by using recombinant DNA technology). Preferably, the TGFbeta molecule is from a human, known as "hTGF".

[0086] There are three isoforms of human TGFbeta: TGFβ1, TGFβ2, and TGFβ3 (Swiss Prot accession numbers P01137, P08112, and P10600, respectively) which, in their biologically active state, are 25 kDa homodimers comprising two 112 amino acid monomers joined by an interchain disulfide bridge. TGFβ1 differs from TGFβ2 by 27, and from TGFβ3 by 22, mainly conservative amino acid changes. These differences were mapped onto the 3D structure of TGFbeta were determined by X-ray crystallography (Schlunegger et al., 1992; Peer et al., 1996) and receptor binding regions were defined (Griffith et al., 1996; Qian et al., 1996). hTGFβl (SEQ ID NO: 1) MPPSGLRLLL LLLPLLWLLV LTPGRPAAGL STCKTIDMEL VKRKRIEAIR GQILSKLRLA 60 ADYYAKEVTR VLMVETHNEI 120 KLKVEQHVEL YQKYSNNSWR 180 AHCSCCDSRDN TLQVDINGFT 240 TGRRGDLATI HGMNRPFLLL MATPLERAQ H LQSSRHRRAL DTNYCFSSTE KNCCVRQLYI 300 DFRKDLGWKW IHEPKGYHAN FCLGPCPYIW SLDTQYSKVL ALYNQHNPGA SAAPCCVPQA 360 LEPLPIVYYV GRKPKVEQLS NMIVRSCKCS 390 (SEQ ID NO: 1) hTGFβ2 (SEQ ID NO: 2) MHY CVLSAFL ILHLVTVALS RGQILSKLKL TSPPEDYPEP 60 LSTCSTLDMD QFMRKRIEAI EEVPPEVISI YNSTRDLLQE KEVYKIDMPP FFPSENAIPP 120 KASRRAAACE RERSDEEYYA TFYRPYFRIV RFDVSAMEKN VPEQRIELYQ ILKSKDLTSP 180 ASNLVKAEFR VF RLQNPKAR TQRYIDSKVV KTRAEGEWLS KISLHCPCCT FVPSNNYIIP 240 FDVTDAVHEW LHHKDRNLGF NKSEELEARF AGIDGTSTYT LLLMLLPSYR LESQQTNRRK 300 SGDQKTIKST RKKNSGKTPH KRALDAAYCF RNVQDNCCLR YNANFCAGAC PYLWSSDTQH 360 P LYIDFKRDL GWKWIHEPKG SRVLSLYNTI NPEASASPCC VSQDLEPLTI EQLSNMIVKS CKCS 414 (SEQ ID NO: 2) hTGFβ3 (SEQ ID NO: 3) LYYIGKTPKI MKMHLQRALV VLALLNFATV EAIRGQILSK LRLTSPPEPT 60 SLSLSTCTTL GHIKKKRV VMTH VPYQVL ALYNSTRELL YAKEIHKFDM IQGLAEHNEL 120 EEMHGEREEG CTQENTESEY AVCPKGITSK VFRFNVSSVE SKRNEQRIEL FQILRPDEHI 180 KNRTNLFRAE FRVLRVPNPS AKQRYIGGKN LPTRGTAEWL LEISIHCPCH TFQPNGDILE 240 SFDVTDTVRE WLLRRESNLG NIHEVMEIKF KGVDNEDDHG RGDLGRLKKQ KDHHNPHLIL MMIPPHRLDN PGQGGQRKKR 300 ALDTNYCFRN LEENCCVRPL YIDFRQDLGW KWVHEPKGYY ANFCSGPCPY LRSADTTHST 360 VLGLYNTLNP EASASPCCVP QDLEPLTILY YVGRTPKVEQ LSNMVVKSCK CS 41 2 (SEQ ID NO: 3)

[0087] There are three TGFbeta receptors in humans, TGFbeta receptor 1, 2, and 3, which can be distinguished by their structural and functional properties, including affinity for TGFbeta protein family members. Binding of a TGFbeta protein to a homodimeric or heterodimeric TGFbeta transmembrane complex activates the canonical TGFbeta signaling pathway mediated by intracellular SMAD proteins.

[0088] Dysregulation of TGFβs leads to pathological processes that, in humans, have been implicated in numerous conditions, such as birth defects, cancer, inflammatory diseases, autoimmune diseases, and fibrotic diseases (Border et al., 1994; Border et al., 1995b).

[0089] Human TGFβs are very similar to mouse TGFβs: human TGFβ1 was only one amino acid different from mouse TGFβ1; Human TGFβ2 has only three amino acid differences from mouse TGFβ2; and human TGFβ3 is identical to mouse TGFβ3. E. Molecules that inhibit transforming growth factor beta (TGFbeta)

[0090] The present invention includes methods comprising administering to a subject a molecule that binds to TGFbeta. The TGFbeta binder can be any binding molecule, such as an antibody, a fusion protein (e.g., an immunoadhesin), an siRNA, a nucleic acid, an aptamer, a protein, or a small molecule organic compound.

[0091] In certain embodiments, the invention includes an antibody that binds to TGFbeta (an anti-TGFbeta antibody), or a variant thereof, or an antigen-binding fragment thereof. Anti-TGFbeta antibodies specifically bind to a TGFbeta protein, polypeptide fragment, or epitope. The molecule that binds to TGFbeta can be of any species

[0092] In certain exemplary embodiments, the antibody that binds to TGFbeta is a humanized antibody, a fully human antibody, or variant thereof, or an antigen-binding fragment thereof. Preferred TGFbeta antibodies prevent binding of TGFbeta to its receptors and inhibit TGFbeta biological activity (eg, TGFbeta receptor-mediated intracellular SMAD signaling and resulting cellular activity).

[0093] In certain embodiments, the antibody, the antigen-binding fragment thereof, is Lerdelimumab (CAT-152), Metelimumab (CAT-192), Fresolimumab (GC-1008), LY2382770, STX-100, or IMC-TR1.

[0094] In certain specific embodiments, the TGFbeta-binding antibody comprises a heavy chain variable region (VH) comprising the amino acid sequence of any one or more of the following complementarity determining regions (CDRs): HCDR1 - SNVIS (SEQ ID NO: 4); HCDR2 - GVIPIVDIANYAQRFKG (SEQ ID NO: 5); or HCDR3 - TLGLVLDAMDY (SEQ ID NO: 6).

[0095] In other specific embodiments, the TGFbeta-binding antibody comprises a light chain variable region (VL) comprising the amino acid sequence of any one or more of the following complementarity determining regions (CDRs): LCDR1 - RASQSLGSSYLA (SEQ ID NO: 7); LCDR2 - GASSRAP (SEQ ID NO: 8); or LCDR3 - QQYADSPIT (SEQ ID NO: 9).

[0096] In a specific embodiment, the antibody that binds to TGFbeta comprises a heavy chain variable region (VH) comprising the amino acid sequences of SEQ ID NOs: 4, 5, and 6.

[0097] In another specific embodiment, the antibody that binds to TGFbeta comprises a light chain variable region (VL) comprising the amino acid sequences of SEQ ID NOs: 7, 8, and 9.

[0098] In more specific embodiments, the antibody that binds to TGFbeta comprises a heavy chain variable region comprising the amino acid sequences of SEQ ID NOs: 4, 5, and 6; and a light chain variable region comprising the amino acid sequences of SEQ ID NOs: 7, 8, and 9.

[0099] In a specific embodiment, the TGFbeta binding antibody comprises a heavy chain variable region comprising the amino acid sequence of SEQ ID NO: 10: QVQLVQSGAE VKKPGSSVKV SCKASGYTFS SNVISWVRQA PGQGLEWMGG VIPIVDIANY AQRFKGRVTI TADESTSTTY MELSSLRSED TAVYYCASTL GLVLDAMDYW GQGTLVTVSS (SEQ ID NO: 10).

[00100] In another specific embodiment, the antibody that binds to TGFbeta comprises a light chain variable region comprising the amino acid sequence of SEQ ID NO: 11: ETVLTQSPGT LSLSPGERAT LSCRASQSLG SSYLAWYQQK PGQAPRLLIY GASSRAPGIP DRFSGSGSGT DFTLTISRLE PEDFAVYYCQ QYADSPITFG QGTRLEIK (SEQ ID NO : 11).

[00101] In more specific embodiments, the antibody that binds to TGFbeta comprises a heavy chain variable region comprising the amino acid sequence of SEQ ID NO: 10; and a light chain variable region comprising the amino acid sequence of SEQ ID NO: 11.

[00102] In some embodiments, the antibody that binds to TGFbeta further comprises a constant region, for example, a human IgG constant region. In some embodiments, the constant region is a human IgG4 constant region. In further embodiments, the constant region is a modified human IgG4 constant region. Preferably, the IgG4 constant region comprises the amino acid sequence of SEQ ID NO: 12: ASTKGPSVFP LAPCSRSTSE STAALGCLVK DYFPEPVTVS WNSGALTSGV HTFPAVLQSS GLYSLSSVVT VPSSSLGTKT YTCNVDHKPS NTKVDKRVES KYGPPCPSCP APEFLGGPSV FLFPPKPKDT LMISRTPEVT CVVVDVS QED PEVQFNWYVD GVEVHNAKTK PREEQFNSTY RVVSVLTVLH QDWLNGKEYK CKVSNKGLPS SIEKTISKAK GQPREPQVYT LPPSQEEMTK NQVSLTCLVK GFYPSDIAVE WESNGQPENN YKTTPPVLDS DGSFFLYSRL TVDKSRWQEG NVFSCSVMHE ALHNHYTQKS LSL SLGK (SEQ ID NO: 12).

[00103] In other embodiments, the constant region is a human CK constant region. Preferably, the CK constant region comprises the amino acid sequence of SEQ ID NO: 13: RTVAAPSVFI FPPSDEQLKS GTASVVCLLN NFYPRAKVQ WKVDNALQSG NSQESVTEQD SKDSTYSLSS TLTLSKADYE KHKVYACEVT HQGLSSPVTK SFNRGEC (SEQ ID NO: 13).

[00104] In specific embodiments, the antibody which binds to TGFbeta comprises a heavy chain comprising the amino acid sequence of SEQ ID NO: 14: QVQLVQSGAE VKKPGSSVKV SCKASGYTFS SNVISWVRQA PGQGLEWMGG VIPIVDIANY AQRFKGRVTI TADESTSTTY (complementarity determining regions, according to a Kabat definition) are underlined. Positions 121-447: human IgG4 constant region (SwissProt IGHG4_HUMAN). In other specific embodiments, the TGFbeta-binding antibody comprises a light chain comprising the amino acid sequence of SEQ ID NO: 15: ETVLTQSPGT LSLSPGERAT LSCRASQSLG SSYLAWYQQK PGQAPRLLIY GASSRAPGIP DRFSGSGSGT DFTLTISRLE PEDFAVYYCQ QYADSPITFG QGTRLEIKRT VAAPSVFIFP PSDEQLKSGT ASV VCLLNNF YPREAKVQWK VDNALQSGNS QESVTEQDSK DSTYSLSSTL TLSKADYEKH KVYACEVTHQ GLSSPVTKSF NRGEC (SEQ ID NO: 15). Positions 1-108: light chain variable region (VL). The CDRs (complementarity determination regions, according to Kabat's definition) are underlined. Positions 109-215: human CK constant region.

[00105] In other embodiments, the antibody that binds to TGFbeta comprises a heavy chain comprising the amino acid sequence of SEQ ID NO: 14, and a light chain comprising the amino acid sequence of SEQ ID NO: 15.

[00106] In some embodiments, the antibody that binds to TGFbeta is expressed by host cells when comprising leader sequences. The leader sequence preferably comprises an amino acid sequence of 1-30 amino acids in length, more preferably 25-25 amino acids, and most preferably 19 amino acids. The heavy chain, light chain, or both the heavy chain and light chain may comprise a leader sequence.

[00107] For example, the light or heavy chain leader sequence can comprise the amino acid sequence of SEQ ID NO: 16: MGWSCIILFL VATATGVHS (SEQ ID NO: 16). Correspondingly, a host cell may express an unprocessed heavy chain may comprise the amino acid sequence of SEQ ID NO: 17: MGWSCIILFL VATATGVHSQ CKASGYTFSS 50 NVISWVRQAP GQGLEWMGGV VQLVQSGAEV IPIVDIANYA KKPGSSVKVS QRFKGRVTIT ADESTSTTYM 100 ELSSLRSEDT AVYYCASTLG LVLDAMDYW G QGTLVTVSSA STKGPSVFPL 150 APCSRSTSES TAALGCLVKD YFPEPVTVSW NSGALTSGVH TFPAVLQSSG 200 LYSLSSVVTV PSSSLGTKTY TCNVDHKPSN TKVDKRVESK YGPPCPSCPA 250 PEFLGGPSVF LFPPKPKDTL MISRTPEVTC VVVDVSQEDP EV QFNWYVDG 300 VEVHNAKTKP REEQFNSTYR VVSVLTVLHQ DWLNGKEYKC KVSNKGLPSS 350 IEKTISKAKG QPREPQVYTL PPSQEEMTKN QVSLTCLVKG FYPSDIAVEW 400 ESNGQPENNY KTTPPVLDSD GSFFLYSRLT VDKSRWQEGN VFSCSVMHE A 450 LHNHYTQKSL SLSLGK 466 (SEQ ID NO: 17). where positions 1-19: leader sequence Positions 20-139: heavy chain variable region (VH). The CDRs (complementarity determination regions, according to Kabat's definition) are underlined. Positions 140-466: human IgG4 constant region (SwissProt IGHG4_HUMAN).

[00108] In other exemplary embodiments, a host cell expressing an unprocessed light chain can comprise the amino acid of SEQ ID NO: 18: MGWSCIILFL VATATGVHSE TVLTQSPGTL SLSPGERATL SCRASQSLGS50 SYLAWYQQKP GQAPRLLIYG ASSRAPGIPD RFSGSGSGTD FTLTISRLEP 100 EDFAVYYCQQ YADSPITFGQ GTRLEIKRTV AAPSVFIFPP SDEQLKSGTA 150 SVVCLLNNFY PREAKVQWKV DNALQSGNSQ ESVTEQDSKD STYSLSSTLT 200 LSKADYEKHK VYACEVTHQG LSSPVTKSFN RGEC 234 (SEQ ID NO: 18). where Positions 1-19: leader sequence Positions 20-127: light chain variable region (VL). The CDRs (complementarity determination regions, according to Kabat's definition) are underlined. Positions 128-234: human CK constant region.

[00109] In an exemplary embodiment of the invention, the antibody that binds to TGFbeta is a humanized or fully human antibody. Examples of humanized or fully human antibody isotypes include IgA, IgD, IgE, IgG, and IgM. Preferably, the anti-TGFbeta antibody is an IgG antibody. There are four forms of IgG. Preferably, the anti-TGFbeta antibody is an IgG4 antibody. In one embodiment of the invention, the anti-TGFbeta antibody is a humanized IgG4 antibody. In another embodiment of the invention, the anti-TGFbeta antibody is a fully humanized IgG4 antibody.

[00110] In a more preferred embodiment of the invention, the anti-TGFbeta antibody is an IgG4 anti-TGFbeta antibody comprising a heavy chain comprising the amino acid sequence of SEQ ID NO: 14 and a light chain comprising the amino acid sequence of SEQ ID NO: 15. In an alternative more preferred embodiment of the invention, the anti-TGFbeta antibody is an anti-IgG4 TGFbeta antibody comprising a heavy chain variable region and a light chain variable region, the heavy chain variable region comprising 3 complementarity determining regions (CDRs) comprising the amino acid sequences of SEQ ID NOs: 4, 5, and 6, and the light chain variable region comprising 3 CDRs comprising the amino acid sequences of SEQ ID NOs: 7, 8, and 9. Identification, isolation, preparation, and characterization of TGFbeta antibodies, including anti-TGFbeta antibody comprising an amino acid sequence of heavy chain comprising SEQ ID NO: 14 and a light chain amino acid sequence comprising SEQ ID NO: 15, and the CDR sequences corresponding to SEQ ID NOs: 4-9 have been described in detail in U.S. Pat. At the. 7,723,486, and in U.S. patent At the. 8,383,780, each of which is incorporated herein by reference in its entirety.

[00111] Preferably, the antibody or antigen-binding fragment thereof is pan-specific and binds human TGFβ1, TGFβ2, and TGFβ3. More preferably, the antibody or antigen-binding fragment thereof binds human TGFβ1, TGFβ2, and TGFβ3, and acts as an antagonist. More preferably, the antibody or antigen-binding fragment thereof binds human TGFβ1, TGFβ2, and human TGFβ3, and neutralizes human TGFβ1, TGFβ2, and TGFβ3. Exemplary pan-specific monoclonal TGFbeta antibodies (mAbs) suitable for use in the methods of the invention are described in US Pat. 7,723,476 and 8,383,780, each of which is incorporated by reference herein in its entirety.

[00112] 1D11.16 is an exemplary mouse pan-specific anti-TGFbeta antibody that neutralizes human and mouse TGFβ1, TGFβ2, and TGFβ3 in a wide range of in vitro assays (Dasch et al., 1989; Dasch et al., 1996; R&D system product sheet for MAB1835, each of which is incorporated herein by reference in its entirety) and is effective in evidence-principles studies in animal models of fibrosis (Ling et al., 2003; Miyajima et al., 2000; Schneider et al., 1999; Khanna et al., 1999; Shenkar et al., 1994). However, since 1D11.16 is a mouse monoclonal antibody (Dasch et al., 1989; Dasch et al., 1996), it is not preferred for therapeutic use in humans. Correspondingly, in certain embodiments, variants or derivatives of the 1D11.16 antibody are employed in the methods of the invention.

[00113] As noted above, certain embodiments of the invention also include variants or derivatives of TGFbeta antibodies. Especially, the invention may include anti-TGFbeta antibody variants which is an IgG4 anti-TGFbeta antibody comprising a heavy chain comprising the amino acid sequence of SEQ ID NO: 14, and a light chain comprising the amino acid sequence of SEQ ID NO: 15. In other embodiments, the invention includes variants or derivatives of the 1D11.16 antibody. Variants of TGFbeta antibodies may have similar physicochemical properties based on their high similarity, and therefore, are also included within the scope of the invention. Variants are defined as antibodies with an amino acid sequence that is at least 80%, at least 90%, at least 95%, or at least 97%, e.g., at least 98% or 99% homologous to an anti-TGFbeta antibody described herein, and capable of competing for binding to a TGFbeta polypeptide, a TGFbeta polypeptide fragment, or a TGFbeta epitope. Preferably, the variants will enhance, neutralize, or otherwise inhibit TGFbeta binding to its receptors and TGFbeta biological activity (e.g., TGFbeta receptor-mediated intracellular SMAD signaling and resultant cellular activity). Determination of competition for target binding can be done by routine methods known to the person skilled in the art. Preferably the variants are human antibodies, and preferably are IgG4 molecules. In preferred embodiments, the variant is at least 90%, 95%, 96%, 97%, 98%, or 99% identical in amino acid sequence to the IgG4 anti-TGFbeta antibody comprising a heavy chain comprising the amino acid sequence of SEQ ID NO: 14, and a light chain comprising the amino acid sequence of SEQ ID NO: 15. The term "variant" refers to an antibody comprising an amino acid sequence that is altered by one or more amino acids compared to the anti-TGFbeta antibody amino acid sequences. The variant can have conservative sequence modifications, including amino acid substitutions, modifications, additions, and deletions.

[00114] Examples of modifications include, but are not limited to, glycosylation, acetylation, pegylation, phosphorylation, amidation, derivatization by known protecting/blocking groups, proteolytic cleavage, and binding to a cellular ligand or other protein. Amino acid modifications can be introduced by standard techniques known in the art, such as site-directed mutagenesis, molecular cloning, oligonucleotide-directed mutagenesis, and random PCR-mediated mutagenesis in the nucleic acid encoding the antibodies. Amino acid substitutions include those in which the amino acid residue is replaced with an amino acid residue having similar chemical or structural properties. Families of amino acid residues having similar side chains have been defined in the art. These families include amino acids with basic side chains (e.g. lysine, arginine, histidine), acidic side chains (e.g. aspartic acid, glutamic acid), uncharged polar side chains (e.g. asparagine, glutamine, serine, threonine, tyrosine, cysteine, tryptophan), non-polar side chains (e.g. glycine, alanine, valine, leucine, isoleucine, proline, phenylalanine, methionine ), unbranched side chains (e.g., threonine, valine, isoleucine), and aromatic side chains (e.g., tyrosine, phenylalanine, tryptophan). It will be clear to the skilled person that classifications of amino acid residue families other than that used above may also be employed. Furthermore, the variant may have non-conservative amino acid substitutions, for example, exchanging an amino acid for an amino acid residue having different chemical or structural properties. Similar minor variations may also include amino acid deletions or insertions, or both. Guidance in determining which amino acid residues can be substituted, modified, inserted or deleted without abolishing immunological activity can be found using computer programs well known in the art. Computer algorithms, such as, inter alia, Gap or Bestfit, which are known to a person skilled in the art, can be used to optimally align amino acid sequences as being compared and as defining similar or identical amino acid residues. Variants may have the same or different, or higher or lower, binding affinities compared to an anti-TGFbeta antibody, but are still capable of specifically binding TGFbeta, and may have similar, higher or lower biological activity as the anti-TGFbeta antibody.

[00115] Embodiments of the invention also include antigen-binding fragments of anti-TGFbeta antibodies. The term "antigen-binding domain," "antigen-binding region," "antigen-binding fragment," and similar terms refer to that portion of an antibody that comprises the amino acid residues that interact with an antigen and confer on the binding agent its specificity and affinity for the antigen (e.g., the complementarity determining regions (CDR)). The antigen binding region can be derived from any species of animal, such as rodents (e.g., rabbit, mouse or hamster) and humans. Preferably, the antigen binding region will be of human origin. Non-limiting examples of antigen-binding fragments include: Fab fragments, F(ab')2 fragments, Fd fragments, Fv fragments, light chain (scFv) Fv molecules, dAb fragments, and minimal recognition units consisting of the amino acid residues that mimic the hypervariable region of the antibody. F. Therapeutic Administration

[00116] The methods described herein comprise administering a therapeutically effective amount of an antibody that binds to TGFbeta to a subject. As used herein, the phrase "therapeutically effective amount" means a dose of antibody that binds to TGFbeta that results in a detectable improvement in one or more symptoms associated with OI or that causes a biological effect (e.g., a decrease in the level of a particular biomarker) that is correlated with the underlying pathological mechanism(s) giving rise to the condition or symptom(s) of osteogenesis imperfecta. For example, the dose of antibody that binds to TGFbeta that increases bone mineral density, increases bone mass and/or bone strength, reduces bone and/or tooth fractures, and/or improves any diagnostic measurement of OI is considered the therapeutically effective amount.

[00117] In one embodiment, bone mineral density, bone mass, and/or bone strength are increased by about 5% to about 200% after treatment with an antibody that binds to TGFbeta. In certain embodiments, bone mineral density, bone mass, and/or bone strength are increased by about 5% to about 10%, from 10% to about 15%, from 15% to about 20%, from 20% to about 25%, from 25% to about 30%, from 30% to about 35%, from 35% to about 40%, from 40% to about 45%, from 45% to about 50%, from 50% to about 55%, from 55% to about 60%, from 60% to about 65%, from 65% to about 70%, from 70% to about 75%, from 75% to about 80%, from 80% to about 85%, from 85% to about 90%, from 90% to about 95%, from 95% to about 100%, from 100% to about 105%, from 105% to about 110%, from 110% to about 115%, from 115% to about 120%, from 120% to about 125%, from 125% to about 130%, from 130% to about from 135%, from 135% to about 140%, from 140% to about 145%, from 145% to about 150%, from 150% to about 155%, from 155% to about 160%, from 160% to about 165%, from 165% to about 170%, from 170% to about 1 75%, 175% to about 180%, 180% to about 185%, 185% to about 190%, 190% to about 195%, or 195% to about 200% after treatment with an antibody that binds to TGFbeta.

[00118] In certain embodiments, the dose of an antibody that reduces serum biomarkers of bone resorption, such as urinary hydroxyproline, urinary total pyridinoline (PYD), urinary free deoxypyridinololine (DPD), urinary type I collagen crosslinked N-telopeptide (NTX), serum or urinary collagen type I crosslinked N-telopeptide (CTX), bone sialoprotein (BSP), osteopontin (OPN), and Tartrate Resistant Acid Phosphatase 5b (TRAP), is considered the therapeutically effective amount. In one embodiment, serum biomarkers of bone resorption are reduced by about 5% to about 200% after treatment with an antibody that binds to TGFbeta.

[00119] In one embodiment, serum biomarkers of bone resorption, such as urinary hydroxyproline, urinary total pyridinoline (PYD), urinary free deoxypyridinololine (DPD), urinary type I collagen crosslinked N-telopeptide (NTX), serum or urinary collagen type I crosslinked N-telopeptide (CTX), bone sialoprotein (BSP), osteopontin (OPN), and resistant acid phosphatase to tartrate 5b (TRAP), are decreased by about 5% to about 10%, from 10% to about 15%, from 15% to about 20%, from 20% to about 25%, from 25% to about 30%, from 30% to about 35%, from 35% to about 40%, from 40% to about 45%, from 4 5% to about 50%, from 50% to about 55%, from 55% to about 60%, from 60% to about 65%, from 65% to about 70%, from 70% to about 75%, from 75% to about 80%, from 80% to about 85%, from 85% to about 90%, from 90% to about 95%, from 95% to about 100%, from 100% to about 105%, from 105% to about 110%, from 110% to about 115%, from 115% to about 120%, from 120% to about 125%, from 125% to about 130%, from 130% to about 135%, from 135% to about 140%, from 140% to about 145%, from 145% to about 150%, from 150% to about 155%, from 155% to about 160%, from 160% to about 165%, from 165% to about 170%, from 170% to about 175%, from 1 75% to about 180%, 180% to about 185%, 185% to about 190%, 190% to about 195%, or 195% to about 200%, after treatment with an antibody that binds to TGFbeta.

[00120] In certain embodiments, the dose of an antibody that increases serum biomarkers of bone deposition, such as total alkaline phosphatase, bone-specific alkaline phosphatase, osteocalcin, and type I procollagen (C-terminal/N-terminal), is considered the therapeutically effective amount. In one embodiment, serum biomarkers of bone deposition are increased by about 5% to about 200% after treatment with an antibody that binds to TGFbeta.

[00121] In one embodiment, serum biomarkers of bone deposition, such as total alkaline phosphatase, bone-specific alkaline phosphatase, osteocalcin, and procollagen type I (C-terminal/N-terminal), are increased by about 5% to about 10%, from 10% to about 15%, from 15% to about 20%, from 20% to about 25%, from 25% to about 30%, from 30% to about 35%, from 35% to about 40%, from 40% to about 45%, from 45% to about 50%, from 50% to about 55%, from 55% to about 60%, from 60% to about 65%, from 65% to about 70%, from 70% to about 60% 75%, from 75% to about 80%, from 80% to about 85%, from 85% to about 90%, from 90% to about 95%, from 95% to about 100%, from 100% to about 105%, from 105% to about 110%, from 110% to about 115%, from 115% to about 120%, from 120% to about 125%, from 125% to about 130%, from 130% to about 135%, from 135% to about 140%, from 140% to about 145%, from 145% to about 150%, from 150% to about 155%, from 155% to about 160%, 160% to about 165%, 165% to about 170%, 170% to about 175%, 175% to about 180%, 180% to about 185%, 185% to about 190%, 190% to about 195%, or 195% to about 20 0% after treatment with an antibody that binds to TGFbeta.

[00122] Other embodiments include administering a therapeutically effective dose of an antibody that improves the function of non-skeletal organs affected by OI. For example, the dose of antibody that binds to TGFbeta that improves auditory function, lung function, and/or kidney function is considered the therapeutically effective amount.

[00123] In accordance with the methods of the present invention, the therapeutically effective amount of an antibody that binds to TGFbeta that is administered to an individual will vary depending on the age and size (e.g., body weight or body surface area) of the individual, as well as the route of administration, and other factors well known to those of ordinary skill in the art.

[00124] In certain exemplary embodiments, the anti-TGFbeta antibody is administered to the subject as a subcutaneous dose. Other exemplary modes of administration include, but are not limited to, intradermal, intramuscular, intraperitoneal, intravenous, intranasal, epidural, and oral routes. The composition can be administered by any convenient route, for example, by bolus injection or infusion, by absorption through epithelial or mucocutaneous linings (e.g., oral mucosa, rectal and intestinal mucosa, etc.), and can be administered together with other biologically active agents. Administration can be systemic or local. The TGFbeta antibody can be administered parenterally or subcutaneously.

[00125] Various delivery systems are known and can be used to administer the pharmaceutical composition, for example, encapsulation in liposomes, microparticles, microcapsules, receptor-mediated endocytosis (see, for example, Wu et al., (1987) J. Biol. Chem. 262:4429-4432 ). Therapeutic compositions will be administered with suitable vehicles, excipients, and other agents that are incorporated into formulations to provide improved transfer, delivery, tolerance, and the like. A multitude of suitable formulations can be found in the form known to all pharmaceutical chemists: Remington's Pharmaceutical Sciences, Mack Publishing Company, Easton, PA. These formulations include, for example, powders, pastes, ointments, jellies, waxes, oils, lipids, lipid-containing vesicles (cationic or anionic) (such as LIPOFECTIN™), DNA conjugates, anhydrous absorption pastes, oil-in-water and water-in-oil emulsions, carbowax emulsions (polyethylene glycols of various molecular weights), semisolid gels, and semisolid mixtures containing carbowax. See also Powell et al., "Compendium of excipients for parenteral formulations" PDA (1998) J Pharm Sci Technol 52:238-311.

[00126] Pharmaceutical compositions can be prepared in dosage forms in a unit dose suitable for adjusting a dose of the active ingredients. Such dosage forms in a unit dose include, for example, tablets, pills, capsules, injections (ampoules), suppositories, etc.

[00127] Pharmaceutical compositions may also be administered to the subject using any acceptable device or mechanism. For example, administration can be performed using a syringe and needle with an auto-injector and/or reusable pen delivery device. Methods of the present invention include the use of numerous auto-injector and/or reusable pen delivery devices to administer a TGFbeta binder (or pharmaceutical formulation comprising the binder). Examples of such devices include, but are not limited to, AUTOPEN™ (Owen Mumford, Inc., Woodstock, UK), DISETRONIC™ pen (Disetronic Medical Systems, Bergdorf, Switzerland), HUMALOG MIX 75/25™ pen, HUMALOG™, HUMALIN 70/30™ pen (Eli Lilly and Co., Indianapolis, IN), NOVOPEN™ I, II and III (Novo Nordisk, Copenhagen, Denmark), NOVOPEN JUNI OR™ (Novo Nordisk, Copenhagen, Denmark), BD™ pen (Becton Dickinson, Franklin Lakes, NJ), OPTIPEN™, OPTIPEN PRO™, OPTIPEN STARLET™, and OPTICLIK™ (sanofi-aventis, Frankfurt, Germany), to name just a few. Examples of autoinjector and/or disposable pen delivery devices having applications in the subcutaneous delivery of a pharmaceutical composition include, but are not limited to, the SOLOSTAR™ pen (sanofi-aventis), the FLEXPEN™ (Novo Nordisk), and the KWIKPEN™ (Eli Lilly), the SURECLICKTM autoinjector (Amgen, Thousand Oaks, CA), the PENLETTM (Haselmeier, Stuttgart, Germany), the EP IPEN (Dey, L.P.), and the HUMIRA™ Pen (Abbott Labs, Abbott Park, IL), to name but a few.

[00128] The use of a microinfuser to deliver a TGFbeta binder (or pharmaceutical formulation comprising the binder) to an individual is also contemplated herein. As used herein, the term "microinfuser" means a subcutaneous delivery device designed to slowly deliver large volumes (e.g., up to about 2.5 mL or more) of a therapeutic formulation over an extended period of time (e.g., about 10, 15, 20, 25, 30 or more minutes). See, for example, the U.S. at the. 6,629,949; US patent no. 6,659,982; and Meehan et al., J. Controlled Release 46:107-116 (1996). Microinfusers are particularly useful for delivering large doses of therapeutic proteins contained within high concentration (eg, about 100, 125, 150, 175, 200 or more mg/mL) and/or viscous solutions. G. Combination Therapies

[00129] In certain aspects, the invention includes methods for treating OI comprising administering to a subject in need of such treatment an antibody that binds to TGFbeta in combination with at least one additional therapeutic agent. Examples of additional therapeutic agents that can be administered in combination with an anti-TGFbeta antibody in practicing the methods of the present invention include, but are not limited to, bisphosphonates, calcitonin, teriparatide, and any other compound known to treat, prevent, or ameliorate osteogenesis imperfecta in an individual. In the present methods, the additional therapeutic agent(s) can be administered concurrently or sequentially with the antibody that binds to TGFbeta. For example, for concurrent administration, a pharmaceutical formulation can be made that contains both an antibody that binds to TGFbeta and at least one additional therapeutic agent. In one embodiment, the antibody that binds to TGFbeta is administered in combination with pharmaceutical bisphosphonates (e.g., Etidronate, Clodronate, Tiludronate, Pamidronate, Neridronate, Olpadronate, Alendronate, Ibandronate, Zoledronate, and Risedronate). In another embodiment, the antibody that binds to TGFbeta is administered in combination with a drug that stimulates bone formation, such as parathyroid hormone analogues and calcitonin. In yet another embodiment, the antibody that binds to TGFbeta is administered in combination with a selective estrogen receptor modulator (SERM). The amount of the additional therapeutic agent that is administered in combination with the antibody that binds to TGFbeta in practicing the methods of the present invention can be readily determined using routine methods known and readily available in the art.

EXAMPLES

[00130] OI is a generalized connective tissue disease in which affected individuals exhibit an abnormality in the formation of type I collagen fibrils due to mutations in the primary sequence of the alpha 1 or alpha 2 chain of type I collagen, also abnormalities in the translational modification of type I collagen and proteins that bind type I collagen fibrils. adequate repair, post-translational modification, and type I collagen secretion. Mutations in CRTAP are responsible for type VII osteogenesis imperfecta. Mice that lack the Crtap gene (Crtap-/-) exhibit a phenotype that mimics osteogenesis imperfecta, and are used as a model of this disease (Morello et al., 2006). Crtap -/- mice and age-matched wild-type littermate controls were used in the following examples.

Material and Methods Animals, anti-TGFbeta treatment and tissue collection

[00131] Crtap-/- mice were generated and maintained on a mixed C57Black/6J/129Sv genetic background. Mice harboring a G610C mutation in the Col1a2 gene (Col1α2tm1.1Mcbr) were obtained and bred to wild type C57Bl/6J mice. Mice that were heterozygous for the Col1α2tm1.1Mcbr allele were used for experiments. TGFbeta reporter mice expressing luciferase in response to the Smad2/3-dependent TGFbeta signaling pathway (SBE-Luc mice) were obtained and bred to Crtap+/- mice for 2 generations to generate Crtap-/- mice and wild-type littermates expressing the reporter transgene. All mice were housed in a vivarium and animal experiments were performed following Animal Care and Use Committee (IACUC) approved protocol.

[00132] For protein and RNA analyses, calvaria from P3 mice were isolated, cleared of extraskeletal tissue, and rapidly frozen in liquid nitrogen. For immunostaining of Crtap-/- P10 mouse lungs, the lungs of each mouse were similarly inflated immediately after gravity sacrifice with 4% paraformaldehyde at a constant pressure of 25 cm H2O and then sutured closed in the trachea. Lungs were then lightly dissected from the thorax and fixed in 4% paraformaldehyde overnight.

[00133] Eight-week-old female Crtap-/- and Col1α2tm1.1Mcbr mice were treated with the pan-TGFbeta neutralizing antibody 1D11 for 8 weeks (10 mg/kg body weight, I.P. injections 3 times each week). Control Crtap-/-, Col1α2tm1.1Mcbr, and wild-type mice received a control antibody (13C4) of the same IgG1 isotype. After treatment, mice were sacrificed, and lumbar spine and femurs were collected and fixed in 10% formalin for microCT and bone histomorphometry. Contralateral femurs from Crtap-/- mice were stored at -20°C and wrapped in saline-soaked gauze until mechanical testing was performed. Lungs from these Crtap -/- mice were similarly inflated, collected and fixed as described for P10 mice. Blinding was not possible during treatment, since 1D11 antibody or control antibody were injected according to the group allocation. In all subsequent analyzes investigators were blinded to genotype and treatment group.

Immunostaining

[00134] Protein was extracted from rapidly frozen P3 Calvary Samples, was transferred to 300 μL of lysis buffer (Tris-HCL0.0625 m to pH 7.5, 2% SDS, NAF to 5 mm, Na3VO4 to 2 mm and completed rock proteinase inhibitor) and were homogenized by 1 minute, followed by incubation at 9 5 ° C for 40 minutes. The supernatant was transferred to Centrifugal Filter Units/Amicon Ultra 3K (Millipore) and centrifuged to concentrate the protein. The total protein concentration of the lysate was measured using the Micro BCA reagent (Pierce) following the manufacturers instructions. 40 µg of calvaria protein extracts were suspended in a Laemmeli buffer containing 5% β-mercaptoethanol and separated on Mini Protean TGX SDS-PAGE gels (4-20% gradient; Bio-Rad) and transferred onto PVDF membranes for western blot analysis. PVDF membranes were incubated with pSmad2 monoclonal antibody (cell flag #3108, 1:750 in TBST containing 5% BSA overnight), followed by secondary HRP-linked anti-rabbit antibody (GE, 1:5000 in TBST containing 5% BSA for 2 hours), treated with ECL Plus Western Blotting (GE) detection system, and exposed to X-ray film. Subsequently, antibodies were separated from membranes using ReBlot Plus reagent (Millipore), and incubated with Smad2 monoclonal antibody (#5339 cell signaling, 1:2000 in TBST containing 5% BSA overnight), followed by similar secondary antibody incubation and ECL-mediated visualization. X-ray films were scanned and the density of each band was quantified using ImageJ software (National Institutes of Health). Quantitative Real Time PCR

[00135] Total RNA was extracted from rapidly frozen P3 mouse calvaria using Trizol reagent (Invitrogen). The Superscript III RT system (Invitrogen) was used to synthesize cDNA from total RNA according to the manufacturer's protocol. Quantitative RT-PCR was performed on a LightCycler.v 1.5 (Roche) using gene-specific primers and SYBR Green I reagent (Roche). β2-Microglobulin was used as the reference gene for normalizing cDNA concentrations.

In vivo Bioluminescence Imaging

[00136] P10 Ctrap-/- mice and wild-type littermates expressing the TGFbeta reporter transgene (SBE-Luc mice) were injected with D-luciferin (Goldbio, 150 mg/kg, IP), anesthetized with isoflurane, and imaged 10 minutes after injection using a bioluminescence (Xenogen) imaging system. Primary Osteoblast Culture, TGFbeta Reporter Cells

[00137] Bone marrow cells were isolated from tibiae and femurs of approximately 2-month-old Crtap-/- mice and wild-type mice and were cultured in α-MEM supplied with 10% FBS, 100 U/ml penicillin and 100 ug/ml streptomycin. Medium was changed every second day and unattached cells were discarded. After 7 days, the attached cells, defined as bone marrow stromal cells (BMSCs), were reseeded in 24-well plates 2.5 x104 cells per cm2 and were cultured in osteogenic medium (α-MEM, 10% FBS, 500 µM ascorbic acid, and 10 mM β-glycerophosphate) for 3 days. Conditioned medium was collected and incubated with PAI-luciferase reporter mink lung epithelial cells. After 24 hours, cell lysates were collected for luciferase activity assays, which were measured using the Dual-Luciferase Reporter System from Promega. Results were normalized to the amount of total protein quantified using the Micro BCA reagent (Pierce).

MicroCT, Bone Histomorphometry

[00138] Lumbar vertebrae and femurs were scanned ('scanned') using a Scanco μCT-40 microCT for quantification of trabecular and cortical bone parameters. Parameters of vertebral and trabecular femoral bones were analyzed using Scanco analysis software by manual contouring of the trabecular bone of the L4 vertebral body as well as the distal metaphyseal section of the femur. Cortical bone parameters at the center of the femoral central axis were quantified using the automated threshold algorithm included in the software.

[00139] Swept and undecalcified Crtap-/- mouse column samples were then embedded in plastic for sectioning. Toluidine Blue staining and TRAP staining were performed using standard for visualization and quantification of Ob's and Oc's, respectively, using the Bioquant Osteo image analysis system.

Immunostaining and histology

[00140] For immunohistochemistry, hindlimbs of P5 mice were collected, fixed overnight in 4% paraformaldehyde, and embedded in paraffin. After deparaffinization and rehydration, heat-induced antigen retrieval was performed (Dako, S1700) followed by hyaluronidase treatment for 30 min (2 mg/ml; Sigma). Endogenous peroxidase was blocked using 3% hydrogen peroxide for 10 min. After incubation with blocking solution (3% normal goat serum, 0.1% BSA, 0.1% Triton X-100 in PBS), sections were incubated in antibodies to TGFβ1 (G1221, Promega) and decorin (LF-113, kindly provided from Larry Fisher, National Institute of Dental and Craniofacial Research, Bethesda, MD, USA) for 60 min (1:25 dilution each in PBS, only control samples were incubated in PBS) at 37°C, and subsequently incubated with secondary antibody (SuperPicTure Polymer Detection Kit, Invitrogen). DAB substrate was added according to the manufacturer's recommendations and samples were dehydrated and mounted using Cytoseal XYL xylene-based mounting medium (Thermo Scientific). Sections from wild-type and mutant littermates were processed at the same time. Images of trabecular bone were taken with an electron microscope (Axioplan 2, Zeiss) using identical exposure times for wild-type and mutant littermates.

[00141] Lungs from 16-week-old P10 and Crtap-/- mice were similarly inflated during tissue collection, were fixed in 4% paraformaldehyde, and were embedded in paraffin. Lungs from P10 Crtap -/- and wild-type mice were used for immunostaining for pSmad2. Briefly, paraffin sections were treated with xylene, rehydrated and heated for 20 minutes for antigen retrieval (pH 6; Dako). Sections were then incubated in blocking solution (3% normal donkey serum, 0.1% BSA, 0.1% Triton X-100 in PBS), and subsequently incubated with rabbit anti-pSmad2 antibody (1:500) (Cell Signaling, #3108), donkey anti-rabbit secondary antibody was conjugated to Alexa fluor 594 (1:600) (Invitrogen), and was mounted with Prolong Gold anti-fade reagent with DAPI (Invitrogen). Fluorescent images from these sections were taken using a Zeiss microscope (Axiovision Software) using identical exposure times.

[00142] For histology and morphometry of 16-week-old mouse lungs, parasagittal sections were stained using a standard protocol for staining Hematoxylin and Eosin. The mean linear intersection (MLI) method was used to quantify the distance between alveolar structures. Briefly, 10 histological fields were captured per mouse at 20X magnification from all lobes of both lungs using an electron microscope (Axioplan 2, Zeiss). The MLI was measured using modified Image J software (National Institutes of Health, modified by Paul Thompson). After manual removal of blood vessels, large airways and other non-alveolar structures, the software automatically outlines the alveolar tissue in each image and superimposes a grid of lines consisting of 1,353 lines with each line measuring 21 pixels over the image. The number of lines intersecting alveolar structures was used to calculate the MLI.

Mechanical testing by 3-point bending

[00143] Crtap-/- and wild-type femurs were tested by three-point curve using a 6 mm span with an Instron 5848 device (Instron Inc., Norwood MA). All femurs were tested wet at room temperature. They were preloaded to 1 N at a rate of 0.05 N/s for 5 seconds. After preloading, the femurs were loaded to failure at a rate of 0.1 mm/sec. Loading and displacement data were captured at a rate of 40 Hz using BLUEHILL Software (Instron 5848).

[00144] To determine the yield point, a region was identified after preloading and before the maximum load on the load-displacement curve. This region was separated into 5 segments from which the fitted line of the segment with the greatest slope was drawn. Next, an offset of 0.012 mm was implemented over the line. The intersection point between the displacement line and the load-displacement curve was the yield point of displacement 0.012. This yield point more closely corresponded to a displacement stress of 0.2%, which is commonly chosen in the literature. The elastic region was identified with the region from preload completion to Yield Point. The post-yield region was identified as the region from the yield point to the point where the change in load exceeded -1N, indicating failure. Elastic displacement was the displacement during which a species remained in the elastic region. The post-yield displacement was the displacement during which the species remained in the post-yield region. The total displacement was calculated as the sum of the elastic displacement and the post-yield displacement. Using the trapezoidal numerical integration method, energy for a fault was calculated as the area under the load-displacement curve. Maximum loading was determined by finding the highest loading value recorded by BLUEHILL, before the species failed. To calculate Stiffness, the least squares fitting method was applied to the steepest segment of the elastic region of the load-displacement curve. Stiffness was the slope of the least squared fit line. Geometric data (diameter and moment of inertia) obtained from microCT analysis of the femoral midshaft were used to calculate intrinsic material properties: ultimate strength, strength to failure, and elastic modulus. Serum bone turnover markers

[00145] Serum Osteocalcin (OCN) was quantified using the EIA Mouse Osteocalcin Kit from Biomedical Technologies Inc. C-terminal bone collagen crosslinked telopeptide (CTX) was quantified using the RatLapsTM EIA kit from Immunodiagnostic Systems Ltd. Both analyzes were performed according to the manufacturer's protocols. Collagen SDS-PAGE, mass spectrometry and crosslink analysis

[00146] For mass spectrometry, type I collagen was prepared from Crtap-/- and wild-type tibiae. Bone was defatted with chloroform/methanol (3:1 v/v) and was demineralized in 0.5M EDTA, 0.05M Tris-HCl, pH 7.5, in all steps at 4°C. Bones were finely cut and collagen was solubilized by hot denaturation (90oC) in SDS-PAGE sample buffer. α-Collagen chains were cut from SDS-PAGE gels and subjected to in-gel trypsin digestion. Electrospray MS was performed on the tryptic peptides using a Deca XP LCQ ion trapping mass spectrometer equipped with an in-line liquid chromatography (LC) column (ThermoFinnigan) using a C8 capillary column (300 µm x 150 mm; Grace Vydac 208 MS5.315) eluted at 4.5 µL min. Sequest search software (ThermoFinnigan) was used for peptide identification using the NCBI protein databases.

[00147] Pyridinoline crosslinks (HP and LP) were quantified by HPLC after hydrolyzing demineralized bone in 6N HCl. Surface plasmon resonance analysis

[00148] Surface plasmon resonance experiments were performed using a BIACore X instrument (GE Healthcare Bio-Science Corp.). Purified native mouse tendon type I collagen from wild-type mice and Crtap -/- mice were immobilized on a CM5 sensor chip by amide coupling at a concentration of about 0.05 ng/mm2 (500 RU) and 0.08 ng/mm2 (800 RU), respectively. Experiments were conducted at a flow rate of 10 µl/min and 20°C in HBS-P buffer (10 mM Hepes buffer, pH 7.4, containing 150 mM NaCl and 0.005% Surfactant P20). Recombinant human decorin core protein (R&D systems) was injected over both CM5 type I chips. The concentration of the human decorin stock solution was determined by amino acid analysis. The decorin binding response to wild-type mouse type I collagen and Crtap -/- was normalized by the amount of type I collagen immobilized on the CM5 sensor chips. Three concentrations of decorin were used (3, 5 and 12 μM), for each concentration the analysis was repeated three times. This experiment was performed twice with collagen isolated from different mice each time Statistical methods

[00149] Two-group comparisons were performed using two unpaired, tailored Student's t-tests. For 3-group comparisons, a one-way analysis of variance (ANOVA) was performed if equal variance was confirmed, followed by multiple pairwise comparison totals using the Holm-Sidak method. If the equal variance test failed, a Kruskal-Wallis ANOVA on the ratings was performed, followed by multiple pairwise comparison totality using Tukey's test. A P-value less than 0.05 was considered statistically significant for Student's t-test, ANOVA, and one way Kruskal-Wallis One ANOVA on ratings. For multiple pairwise posthoc concentrations, each P-value was compared to a critical level depending on the P-value rank and the total number of comparisons made to determine whether differences between groups were significant. Sigma Plot V11.0 (Systat Software Inc.) was used for statistical analysis.

[00150] The effects of 1D11 on bone and lungs of mice with OI were unknown at baseline. To determine the initial sample size per group of mice we calculated that in order to detect a minimum difference of 20% in bone mass (BV/TV) by MicroCT between control and 1D11-treated OI mice at 90% power, a group size of 8 mice is required. Example 1: Altered TGFbeta signaling in Crtap-/- calvaria

[00151] Crtap-/- mice and age-matched wild-type (WT) littermate controls were analyzed for expression of activated pSmad 2, a member of the TGFbeta signaling pathway, as well as other downstream targets of TGFbeta. Calvarial bones were excised and RNA and protein were extracted and analyzed by Real-Time-PCR (Realtime-PCR, abbreviated RT-PCR) and Western blot, respectively. As seen in figure 1A and 1B, Crtap-/- mice had a higher ratio of 100% activated pSmad2 to total Smad2 compared to wild-type mice, as measured by Western blot and quantitated by densitometry, indicating that TGFbeta signaling is elevated in Crtap-/- mice. TGFbeta transcriptional targets, such as Col1a1 and p21, were elevated compared to wild-type controls, as measured by RT-PCR and demonstrated in Figure 1C and Figure 1D, respectively. Profibrotic ECM protein connective tissue growth factor (CTGF) was measured and found to be about 50% higher in Crtap -/- mice compared to wild-type controls, as determined by RT-PCR and demonstrated in Figure 1E. As shown in Figure 1F, RT-PCR analysis revealed that cyclin-dependent kinase inhibitor p27 expression was unchanged in either Crtap -/- mice or wild-type mice. Example 2: Increased TGFb activity in Crtap-/- mice in vivo and in crtap-/- osteoblastic cells

[00152] Crtap-/- mice were crossed with TGFbeta reporter mice that express luciferase in response to activation of TGFbeta signaling (Jackson Laboratory; B6.Cg-Tg(SBE/TK-luc)7Twc/J). P9 mice were injected with D-Luciferin substrate (150 mg/kg) 10 minutes before imaging (Xenogen; IVIS chamber system). As demonstrated in Figure 2A, Crtap-/- mice had considerably higher luminescence without their tails, long bones, and calvaria compared to wild-type controls, indicating increased TGFb activity in crtap-/- mice. Figure 2B, quantification of luciferase activity in calvaria.

[00153] Bone marrow stromal cells (BMSCs) were isolated from Crtap-/- mice and wild-type mice, were cultured under ex vivo osteogenic conditions, and the conditioned culture medium was analyzed for TGFbeta activity using a cell line that expresses luciferase in response to activation of TGFbeta signaling. As shown in Figure 2C, conditioned medium derived from Crtap -/- BMSCs resulted in a two-fold greater luciferase activity of the reporter cell line compared to BMSCs derived from wild-type mice. Together, these data indicate that TGFbeta secretion and activity are elevated in bone from osteoblastic cells of Crtap-/- mice. Example 3: μCT analysis of Crtap-/- vertebrates

[00154] Adult 8 week old Crtap -/- mice (N = 6 per group) were administered 1D11 (10 mg/kg, I.P., 3 times/week, 8 weeks total), a pan-specific antibody mouse surrogate that binds to TGFbeta comprising a heavy chain comprising the amino acid sequence of SEQ ID NO: 14 and a light chain comprising the amino acid sequence of SEQ ID NO: 1 5. An unrelated 13C4 antibody was administered to a separate group of Crtap -/- mice and wild-type mice as a control (N = 6). The L4 vertebral bodies of 16-week-old mice (treated for 8-16 weeks) were imaged by μ-CT. MicroCT data of L4 vertebral body derived from 8 week old Crtap -/- mice (n = 6 per group) that were treated with the TGFbeta neutralizing antibody 1D11 (Genzyme; 10 mg/kg, I.P., 3 times/week) for 8 weeks and wild-type (wild-type) and control Crtap -/- mice that were treated with a control antibody (13C4-placebo ) is shown in Figure 3. As shown in Figure 3, Crtap -/- vertebrae were porous compared to wild-type control vertebrae. However, treatment with 1D11 resulted in a skeletal phenotype that was comparable to the wild-type condition.

[00155] The data from figure 3 were quantified in figure 4, treatment with pan-specific anti-TGFbeta antibody rescued the skeletal phenotype of Crtap-/- mice. Vertebrae derived from treated Crtap-/- mice were statistically similar to wild-type control mice in measured parameters, including bone volume density (BV/TV), total bone surface (BS), bone surface density (BS/BV), trabecular number (Tb.N), trabecular thickness (Tb.Th), trabecular spacing (Tb.Sp), and total volume (Dens TV). Example 4: Histomorphometry of TGFbeta-treated anti-Crtap-/- vertebrates

[00156] In addition to μ-CT, vertebral bones from antibody-treated, placebo-treated, and wild-type mice were analyzed by histomorphometry. As shown in figures 5A and 5B, μ-CT results were confirmed by histomorphometric analysis. In addition, tissue sections were stained for expression of the osteoclast marker TRAP. Analysis revealed that there are more osteoclasts covering more of the bone surface in Crtap -/- mice compared to wild-type controls (N.Oc/BS and Oc.S/BS), indicating increased osteoclastic activity. Treatment with the 1D11 anti-TGFbeta reduced all osteoclast-specific parameters to lower wild-type numbers. Thus, osteoclasts were identified as a potential target for a TGFbeta antibody, and more specifically, pan-specific TGFbeta antibodies. Example 5: Three-point curve test of Crtap-/- femurs treated with anti-TGFbeta

[00157] Biochemical testing was performed on femurs excised from 16-week-old mice (after treatment for 816 weeks) using a standard three-point curve test with an Instron 5848 device (Instron Inc., Norwood MA) with a span of 6 mm, preloaded to 1 N at a rate of 1N/s for 5 seconds. After preloading, femurs were compressed to failure at a rate of 0.1 mm/sec. Load and displacement data were captured at a rate of 40 Hz using BLUEHILL Software (Instron 5848).

[00158] As demonstrated in Figure 6, femurs from Crtap-/- mice were less rigid and were able to withstand a significantly lower maximum load compared to wild-type control mice. Femurs from Crtap-/- mice treated with 1D11 showed a significant improvement in maximal load, and a tendency towards increased stiffness compared to control Crtap-/- mice.

[00159] Thus, treatment with the 1D11 pan-specific anti-TGFbeta antibody quantitatively, qualitatively, and biomechanically restored the skeletal phenotype of Crtap-/- mice. Example 6: Inhibition of TGFbeta signaling with 1D11 improves lung phenotype

[00160] Crtap-/- mice have a generalized connective tissue disease manifested by low bone mass, glomerulosclerosis, and pulmonary dysplasia (Baldridge et al.; PLoSone, 5(5):e10560 (2010)). Increased TGFbeta expression was seen in the lungs of Crtap -/- mice, as evidenced by positive immunostaining for pSmad2 and demonstrated in Figure 7A. Histologically, Crtap-/- mice exhibited increased distal airway space compared to wild-type mice, as shown in Figure 7B. Treatment with 1D11 (10 mg/kg, IP, 3x/week for 8 weeks) reduced pSmad2 expression in Crtap-/- mice and reduced distal airway space and improved lung phenotype, as demonstrated in Figure 7A and Figure 7B and quantified in Figure 7C (*P<0.05 vs. control Crtap-/-; 10 analyzed images per mouse, n = 8 mice per group ). Example 7: Decorin expression in Crtap-/- lungs

[00161] Transcriptional regulators of TGFbeta expression were investigated in order to understand the basis of dysregulated TGFbeta signaling in the bones and lungs of Crtap-/- mice. A major class of extracellular proteins that can regulate TGFbeta in ECM include small leucine-rich proteoglycan (SLRP), such as decorin. Immunostaining revealed increased decorin expression in Crtap -/- lungs compared to wild-type control lungs, as shown in figure 8.

[00162] Since decorin is a mature TGFbeta regulator, this finding suggests that altered post-translational modification of collagen, as occurs in OI, alters the interactions of ECM proteins, including SLRPs. Decorin binds to hydroxyproline sites, such as that located at amino acid residue 396 of type I and type II collagens, which are absent in Crtap-/- mice. A decorin binding assay was performed to determine whether decorin binding could be altered in OI, and whether this could be at least partially responsible for the observed phenotypes in bones and Crtap-/- mice. As shown in figure 9, binding of decorin to 3-hydroxylated collagen peptides (as in type I and II collagen) was greater than collagen peptides without 3-hydroxylation (as in type III collagen). Example 8: Increased TGFbeta signaling is a mechanism in Osteogenesis Imperfecta

[00163] OI is characterized by brittle bones, low bone mass, fractures and bone deformities. In addition, extraskeletal manifestations including lung abnormalities contribute substantially to morbidity and mortality. Most cases of OI are caused by autosomal dominant mutations in the genes encoding type I collagen (COL1A1 and COL1A2). In recent years, mutations in additional genes encoding proteins involved in the post-translational modification of collagen have been identified as causing recessive forms of OI. The first described was in cartilage-associated protein (CRTAP), a member of the polyyl-3-hydroxylase complex that is responsible for 3-hydroxylation of proline residue 986 α1(I) in collagen type I. Hypomorphic CRTAP mutations led to partial loss of 3-hydroxyproline (3Hyp) in fibrillar collagen as well as ultra-modification of other residues and result in recessive OI type VII, which clinically overlaps with dominant forms s of severe OI. The physiological function of 3Hyp is not fully understood, but biochemical studies suggest that it may be involved in collagen-protein interactions, rather than negatively affecting collagen stability.

[00164] The ECM is an important reservoir for signaling molecules and their regulators. In bone, TGFbeta acts as a central coordinator of bone remodeling by coupling the localized activity of bone-resorbing osteoclasts and bone-forming osteoblasts. TGFbeta is abundantly produced by osteoblasts, is predominantly secreted in inactive latent forms, and is deposited into the bone matrix. Here, it can be released and activated during bone resorption by osteoclasts. As an additional level of regulation, active TGFbeta can be bound by proteoglycans, which modulate its bioactivity in association with collagen fibrils. Since type I collagen is the most abundant ECM component in bone, this raises the intriguing hypothesis that the alteration of collagen structure observed in OI not only increases bone fragility, but also affects the signaling reservoir function of the bone matrix. Interestingly, Crtap-/- mice show phenotypic overlap with animal models of increased TGFbeta signaling. For example, overexpression of TGFbeta results in low bone mass. Furthermore, Crtap-/- mice exhibit an enlargement in the alveolar airway space in the lungs, which is similar to that seen in a mouse model of Marfan syndrome, where increased TGFbeta signaling has been shown to be a major contributor to lung pathology. Therefore, the status of TGFbeta signaling was studied in the Crtap-/- mouse model and recessive OI.

[00165] To assess the status of TGFbeta signaling in bone, the expression levels of TGFbeta target genes in the calvarial bone of Crtap-/- mice were evaluated. Compared to wild-type (wild-type) samples, Crtap -/- bone showed increased expression of the downstream TGFbeta targets p21 (cyclin-dependent kinase inhibitor 1), PAI-1 (plasminogen activator inhibitor 1), and Col1a1, consistent with elevated TGFbeta activity (Figure 10A). To confirm the increased activation in the intracellular TGFbeta signaling pathway, the status of Smad2, an intracellular second messenger protein, which becomes phosphorylated upon activation of TGFbeta receptors, was evaluated. Consistent with target gene expression, immunoblot analysis demonstrated a higher ratio of phosphorylated Smad2 (pSmad2) to total Smad2 in bone samples from Crtap -/- mice, indicating increased TGFbeta signaling (Figures 10B and 10C).

[00166] To determine whether these static measurements reflect increased in vivo TGFbeta activity, Crtap -/- mice were intercrossed with TGFbeta reporter mice expressing luciferase under the control of Smad binding elements responsible for TGFbeta (SBE-Luc mice). Compared to wild-type/SBE-Luc littermates, Crtap-/-/SBE-Luc mice showed an increase in bioluminescence of areas over skeletal structures, indicating increased in vivo TGFbeta activity (Figure 10D). In the 3 litters, Crtap -/- mice show an average 2.86-fold (SD±0.34) bioluminescence signal in the head/calvaria compared to wild-type mice. Furthermore, in order to test whether the increased TGFbeta/Smad signaling associated with Crtap loss is intrinsic to bone, i.e., autonomous tissue, bone marrow stromal cells (BMSCs) were differentiated to osteoblastic cells in vitro. By using a TGFbeta reporter cell line, conditioned medium from Crtap -/- BMSCs was found to exhibit higher TGFbeta activity compared to medium derived from wild-type BMSCs ( Figure 10E ). Together, these findings indicate that loss of Crtap increases TGFbeta signaling in bone in a tissue-autonomous manner.

[00167] Patients with severe OI may also exhibit intrinsic lung abnormalities, and respiratory failure is one of the leading causes of death for these individual individuals. Interestingly, Crtap-/- mice show a diffuse increase in alveolar airway space, a feature associated with increased TGFbeta signaling in other developing models. Correspondingly, lungs from Crtap -/- mice showed increased intracellular staining for pSmad2 in alveolar cells, indicating that increased TGFbeta activity is also present in extraskeletal tissues (Figure 10F).

[00168] To understand an increased TGFbeta signaling represents a causal mechanism contributing to lung and bone phenotypes in Crtap-/- mice, a rescue experiment was performed with a pan-TGFbeta neutralizing antibody (1D11). Eight week old Crtap -/- mice were treated with 1D11 for 8 weeks; control Crtap-/- mice and wild-type mice received a non-specific control antibody (13C4). 1D11 did not significantly change the body weight of treated Crtap-/- mice, indicating that TGFbeta inhibition did not affect overall nutritional status (Figure 14). Furthermore, crosslinking and mass spectrometric analyzes showed that 1D11 did not significantly change the 3-hydroxylation state of type I collagen P986 or collagen crosslinks in Crtap-/- mice, which suggest that dysregulated TGFbeta signaling is a consequence of altered molecular collagen structure, and not directly involved in the processing of extracellular collagen or extracellular fibril assembly (Figure 15). Crtap-/- mice exhibit reduced bone mass and abnormal trabecular bone parameters (Figures 11A and 11B). MicroCT image analysis of vertebrae demonstrated that compared to control Crtap-/- mice, inhibition of TGFbeta significantly increased trabecular bone parameters, including bone volume/total volume, trabecular number, and trabecular thickness to near wild-type levels (Figures 11A and 11B, and Figure 17). Similar beneficial effects were seen in trabecular femoral bone in Crtap-/- mice, where inhibition of TGFbeta significantly improved trabecular bone parameters (Figure 18). Skeletal TGFbeta inhibition effects with 1D11 have been previously reported in wild-type mice and in Esl-1-/- 11 mice, a model with increased TGFbeta activity due to a defect in the maturation of normal TGFbeta. Whereas 1D11 modernly increased trabecular BV/TV by 3% in the spine in wild-type mice, Esl-1-/- mice exhibited a 106% increase in BV/TV. This suggested that targeting TGFbeta in a pathophysiological situation where it is increased in the skeleton, could lead to a relatively more pronounced positive effect. In the present study, 1D11 increased trabecular BV/TV in the spine by 235% in Crtap-/- mice, which supports that dysregulated TGFbeta signaling is an important contributor to low bone mass in Crtap-/- mice. In the midshaft of the femur, cortical architectural parameters including cortical thickness, diameter, cross-sectional area, and cross-sectional moments of inertia in Ctrap-/- mice were significantly reduced compared to wild-type mice. After treatment with 1D11, these parameters were no longer significantly different from wild-type mice ( Figure 19 ). To test whether these changes in trabecular and cortical bone translated into improved bone strength, mechanical testing of the femurs by 3-point flexion was performed. It was found that inhibition of TGFbeta was able to increase maximal load and maximal force in Crtap-/- treated mice, indicating improved whole-bone and tissue strength and improved resistance to fracture. However, treatment with 1D11 had no effects on the increased fragility of the OI bone, as indicated by reduced post-yield displacement in both control mice and Crtap-/- mice treated with 1D11 (Figure 20). This likely reflects abnormal inert mineralization associated with altered collagen structure. Taken together, these findings indicate that increased TGFbeta signaling is a major contributor to the bone phenotype in recessive OI that results from Crtap deficiency and that inhibition of dysregulated TGFbeta signaling restores bone mass, microstructural parameters and improves overall bone strength.

[00169] To understand the inhibition effects of TGFbeta in Crtap-/- mice at the cellular level, histomorphometric analyzes or treated mice were performed. On sections of vertebral bodies in this study found increased osteoclast numbers (Oc) and osteoblast numbers (Ob) per bone surface in control Crtap-/- compared to wild-type mice, indicating increased bone remodeling in the spine (Figure 11C and Figure 21). Consistently, blood bone turnover markers osteocalcin (OCN) and C-terminal bone collagen cross-linked telopeptide (CTX) were elevated in eight-week-old Crtap-/- mice (both OCN and CTX) and 16-week-old control Crtap-/- (CTX only) mice (Figure 19). Similar changes in bone cellular composition have been described in patients with recessive and dominant OI, who show increased Oc and Ob numbers consistent with increased bone turnover. Interestingly, mouse models of increased TGFbeta signaling also show low bone mass with increased osteoclastic bone resorption and abnormal bone remodeling. Further reports of the effects of TGFbeta on bone cells are consistent with a model where TGFbeta can stimulate the initial recruitment and differentiation of Oc and Ob precursors at bone repair sites, followed by insulin-like growth factor-1 (IGF-1) mediated Ob differentiation. However, at persistently high doses, TGFbeta can inhibit Ob differentiation by repressing differentiation factor RUNX2. Given the crucial effects on the Oc/Ob interaction, fine tuning of TGFbeta availability is a key factor for the local coupling of bone resorption with formation during bone remodeling and its imbalance can lead to significant bone pathology.

[00170] In contrast to the findings in control Crtap-/- mice, bone sections from Crtap-/- mice treated with 1D11 revealed reduced Oc and Ob numbers, which were even lower than the values measured in wild-type mice, indicating a supraphysiological suppression of dysregulated bone remodeling as a result of TGFbeta inhibition at the dose of 1D11 used in this experiment (figure 11C). Consistent with a previous report, the observation of a reduction in Oc's and Ob's below wild-type levels also underscores the physiological requirement for local amounts of TGFbeta to normally coordinate Oc's and Ob's during the bone remodeling process. Our findings are different from previous studies in wild-type mice, where 1D11 treatment reduced Oc numbers but increased Ob numbers. This may reflect distinct cellular effects of TGFbeta inhibition in a pathophysiological situation with increased TGFbeta signaling and increased bone remodeling compared to normal bone in wild-type mice. TGFbeta has been shown to inhibit the differentiation of osteoblast precursor cells, and increased TGFbeta signaling could therefore lead to a higher proportion of immature osteoblast lineage cells. On the other hand, an increased number or a higher proportion of immature Ob's on the bone surface could result in an increased amount of TGFbeta secreted by these cells. The finding that inhibition of TGFbeta with 1D11 significantly reduces increased Ob numbers in Crtap -/- mice suggests that increased TGFbeta signaling causally contributes to the increase in osteoblast lineage cells.

[00171] In addition to the findings with reference to Oc and Ob numbers, increased osteocyte numbers (Ot) per bone area in control Crtap-/- mice were observed, which were reduced to levels comparable to those of wild-type mice in Crtap-/- mice treated with 1D11 (Figure 11C and Figure 21). In patients with OI, an increased Ot density was observed in individuals with more severe forms of the disease, probably reflecting the presence of immature primary bone due to a defect in physiological maturation in the OI bone. Consistent with our hypotheses that increased TGFbeta signaling contributes to bone pathology in OI, overexpression of TGFbeta in wild-type mice similarly results in an increased Ot density. As a possible explanation, TGFbeta may inhibit Ob opoptosis during the transition from Ob's to Ot's, and thereby leading to an increased Ot density. Collectively, these findings indicate that increased TGFbeta signaling contributes to high bone turnover status and decreased bone mutation in Crtap-/- mice and that inhibition of dysregulated TGFbeta signaling reverses these cellular changes.

[00172] Given these crucial effects on the Oc/Ob interaction, fine tuning of TGFbeta availability is a key factor for the local coupling of bone resorption with bone formation during bone remodeling and its imbalance can lead to significant bone pathology. Our findings indicate that inhibition of dysregulated TGFbeta signaling in Crtap-/- mice restores bone mass as well as microstructural parameters, fully improves bone strength, and reverses cellular changes observed in Crtap-/- mice. Therefore, dysregulation of TGFbeta signaling is an important contributor to the bone phenotype in this mouse model of recessive OI.

[00173] It is also of interest to know whether TGFbeta inhibition affected the lung phenotype of Crtap-/- mice. Lungs from control Crtap-/- mice show an increase in distal airway space compared to wild-type mice (Figure 11D). Interestingly, lungs from Crtap -/- mice treated with the TGFbeta neutralizing antibody showed a 60% improvement in the distance between alveolar structures (Figures 11D and 11E). This finding indicates that excessive TGFbeta signaling is also an important pathogenic contributor to the lung abnormalities present in Crtap-/- mice. Increased TGFbeta signaling was linked to developing lung abnormalities as well as disease in mature lungs. For example, overexpression of TGFbeta in the lungs results in decreased lung development with enlarged airway space areas, and increased TGFbeta signaling is a pathomechanistic contribution in lung abnormality in Marfan syndrome as well as in the development of emphysema and bronchial asthma. Our results indicate that excessive TGFbeta signaling is an important pathogenic contributor to the lung abnormalities present in Crtap-/- mice. Given the partial rescue of the 1D11 lung phenotype in Crtap-/- mice, it is possible that dysregulated TGFbeta signaling affects lung tissue development when anatomical structures are established, in addition to lung tissue maintenance at later stages when TGFbeta inhibition is able to improve the phenotype.

[00174] The next question asked was how changes in collagen due to Crtap loss (leading to loss of 3Hyp to P986 and post-translational over collagen modification) result in dysregulated TGFbeta signaling. Biochemical analyzes indicate that collagen prolyl-3-hydroxylation does not fundamentally affect the stability of collagen molecules, but may instead affect collagen-protein interactions. An alternative hypothesis is that 3Hyp loss could affect collagen interaction with small leucine-rich proteoglycans (SLRPs). SLRPs are known to be both type I collagen and TGFbeta, and thereby modulate TGFbeta activity. For example, SLRP decorin is able to inhibit distinct effects of TGFbeta in osteosarcoma cells while increasing TGFbeta activity in proteoblastic cells. The decorin-binding region of type I collagen is suggested to center on residues 961/962 of the triple helical domain, which is located in close proximity to residue P986, which is unhydroxylated in OI due to Crtap deficiency. Therefore, it is possible that the position of P986 3Hyp marks an interaction site for decorin binding to type I collagen, thereby mediating the sequestration of mature TGFbeta into collagen.

[00175] Therefore, it has been hypothesized that binding of decorin to collagen is critical for regulation of TGFbeta and that this binding is disrupted with altered collagen structure, for example by loss of post-translational 3-prolyl-hydroxylation modification of P986 in the α1 chain of type I collagen in the case of recessive OI. It was identified that although Crtap loss did not alter RNA expression of decorin and other SLRPs in calvarial bone (Figure 12A), nor the quantitative abundance of decorin in trabecular bone (Figure 24), it reduced binding of recombinant decorin core protein to type I collagen isolated from Crtap-/- mice versus wild-type mice (Figure 12B). Surface plasmon resonance analysis measurements of recombinant decorin core protein binding to type I collagen from wild-type mice and Crtap -/- mice demonstrated reduced binding in Crtap -/- mice at the three concentrations tested (Figure 23). Three technical copies at each of the indicated concentrations of decorin were made from two independent biological copies (♦ copy 1, ▲ copy 2). Results are shown as a percentage of the wild-type mean (bars indicate mean by group). The mean reductions in decorin binding to Crtap-/- collagen type I at 3, 5, and 12 µM decorin were 28.5%, 33.5%, and 38.1%, respectively.

[00176] This finding suggests that alterations of collagen-proteoglycan interactions may contribute to dysregulated TGFbeta signaling in bone and other collagen-rich tissues in OI. Based on the reported requirement of decorin-collagen binding to decorin to effectively reduce TGFbeta bioactivity, it is possible that defects in OI collagen lead to altered decorin binding, and therefore, its ability to sequester TGFbeta in the matrix and modulate TGFbeta functions. Therefore, altered proteoglycan-collagen interactions may contribute to dysregulated TGFbeta signaling in bone and other collagen-rich tissues in OI, yet in major changes in absolute TGFbeta levels are present (Figure 24 and Figure 25). This notation is supported by the finding that COL1A1 and COL1A2 mutations in more severe forms of OI cluster dominant in specific regions that are known to bind proteoglycans, further supporting a physiological relevance of proteoglycan-collagen interactions to normal bone homeostasis. This would also imply that other proteoglycans that are competing with decorin for the collagen binding site might also contribute to dysregulated TGFbeta activity, and that additional signaling pathways could be altered.

[00177] Because of the clinical overlap of some recessive and dominant forms of OI where defective structure of collagen fibers leads to brittle bones and extraskeletal manifestations, it is possible that dysregulation of TGFbeta signaling is a common pathophysiological disease mechanism. To address this hypothesis, the status of TGFbeta signaling in a mouse model of dominant OI was investigated. Knock-in mice that carry a G610C mutation in the Col1a2 gene phenocopy (Col1α2tm1.1Mcbr) of a mild, dominantly inherited form of OI that was originally identified in an Amish population. In bone samples from Col1α2tm1.1Mcbr mice, increased expression of p21 and PAI-1 TGFbeta target genes was found, indicating upregulation of TGFbeta signaling (Figure 13A). Consistently, immunoblot analysis of bone extract from Col1α2tm1.1Mcbr mice also showed an increased ratio of activated pSmad2/total Smad2, similar to our observation in Crtap-/- mice (Figures 13B and 13C).

[00178] To test whether increased TGFbeta signaling in this dominant OI model also represents a causal mechanism, 8-week-old Col1α2tm1.1Mcbr mice were treated with the TGFbeta-neutralizing antibody 1D11 for 8 weeks; control Col1α2tm1.1Mcbr mice and wild-type mice were treated with the control 13C4 antibody. Similar to findings in Crtap-/- mice, treatment with 1D11- restored trabecular bone parameters in the spine to wild-type levels (Figures 13D, 13E, and 22). Taken together, these findings indicate that dysregulated TGFbeta signaling is also an important contributor to the pathogenesis of dominant forms of OI, and that anti-TGFbeta therapy corrects the bone phenotype in dominant OI.

[00179] From the clinical translational perspective, potential negative effects of systemic TGFbeta inhibition in patients with OI have to be considered. While TGF-β1-/- mice develop severe multifocal inflammatory disease and dysregulation within the first few weeks of life, in both Crtap-/- and Col1α2tm1.1Mcbr mice treated with 1D11 we observed no obvious negative effects on overall health, behavior, or growth, which suggest that the effects of a partial pharmacological inhibition of TGFbeta ligands in adult mice are different from a complete loss of TGFβ1 during development. In humans, Fresolimumab (GC1008, Genzyme), which is similar to 1D11 in its affinity and specificity for the 3 isoforms of TGFbeta, was used in phase I studies in patients with treatment-resistant primary focal segmental glomerulosclerosis, idiopathic pulmonary fibrosis, and malignant melanoma or renal cell carcinoma. In these studies, Fresolimumab was generally well tolerated, with possible dose-related adverse events including rash or skin lesions, epistaxis, gingival bleeding and fatigue.

[00180] The molecular mechanisms of OI are incompletely understood. As a result, current treatment options for patients with OI are mostly limited to anti-resorptive therapies as used for the treatment of osteoporosis. Interestingly, a recently randomized, placebo-controlled trial of the anabolic agent teriparatide in adults with OI showed that severe OI types III/IV responded differently than those with moderate OI type I (Orwoll et al., 2014). This suggests genotypic differences in response to targeted therapies in modifying cell signaling and that TGFbeta-targeted treatment may be a promising option for further study in severe OI due to collagen and post-translational collagen-modifying gene mutations. Overall, our data support the concept of matrix-disregulated cell signaling as a mechanism in the pathogenesis of different genetically inherited forms of brittle bone disease and are indicative of a disease-specific mechanism-based strategy for the treatment of OI by counteracting ultra-active TGFbeta activity in skeletal and extraskeletal tissues.

Claims (21)

1. Use of an antibody or antigen-binding fragment thereof that especially binds to and neutralizes transforming growth factor beta (TGFβ) TGFβ1, TGFβ2 and TGFβ3, characterized in that it is for the manufacture of a medicament for the treatment of osteogenesis imperfecta (OI) in a subject in need thereof, wherein the antibody or antigen-binding fragment comprises a heavy chain variable region comprising SEQ ID NO: 10, and a light chain variable region comprising SEQ ID NO: 11. 2. Use according to claim 1, characterized in that the antibody or antigen-binding fragment further comprises a human IgG4 constant region. 3. Use according to claim 2, characterized in that the human IgG4 constant region comprises SEQ ID NO: 12. 4. Use according to any one of claims 1 to 3, characterized in that the antibody or antigen-binding fragment further comprises a human k light chain constant region. 5. Use according to claim 4, characterized in that the human k light chain constant region comprises SEQ ID NO: 13. 6. Use according to claim 1, characterized in that the antibody further comprises a human IgG4 constant region and a human k light chain constant region. 7. Use according to claim 6, characterized in that the human IgG4 constant region comprises SEQ ID NO: 12, and the human k light chain constant region comprises SEQ ID NO: 13. 8. Use according to claim 1, characterized in that the antibody comprises a heavy chain comprising SEQ ID NO: 14. 9. Use according to claim 1, characterized in that the antibody comprises a light chain comprising SEQ ID NO: 15. 10. Use according to claim 1, characterized in that the antibody comprises a heavy chain comprising SEQ ID NO: 14 and a light chain comprising SEQ ID NO: 15. 11. Use according to any one of claims 1 to 10, characterized in that the antibody improves a bone parameter selected from the group consisting of bone volume density (BV/TV), total bone surface (BS), bone surface density (BS/BV), trabecular number (Tb.N), trabecular thickness (Tb.Th), trabecular spacing (Tb.Sp), and total volume (Dens TV). 12. Use according to any one of claims 1 to 11, characterized in that the drug reduces a serum biomarker of bone resorption selected from the group consisting of urinary hydroxyproline, urinary total pyridinoline (PYD), urinary free deoxypyridinololine (DPD), urinary type I collagen cross-linked N-telopeptide (NTX), serum or urinary type I collagen cross-linked N-telopeptide (CTX), sia bone loprotein (BSP), osteopontin (OPN), and tartrate-resistant acid phosphatase 5b (TRAP). 13. Use according to any one of claims 1 to 12, characterized in that the drug increases a serum biomarker of bone resorption selected from the group consisting of total alkaline phosphatase, bone-specific alkaline phosphatase, osteocalcin and type I procollagen. 14. Use according to claim 13, characterized by the fact that the type I pro-collagen is a C-terminal/N-terminal type I pro-collagen. 15. Use according to any one of claims 1 to 14, characterized in that the drug inhibits bone resorption. 16. Use according to any one of claims 1 to 15, characterized by the fact that the drug promotes bone deposition. 17. Use according to any one of claims 1 to 16, characterized in that the medicine improves the function of the lungs affected by OI. 18. Use according to any one of claims 1 to 17, characterized in that the antibody or antigen-binding fragment thereof is combined with at least one bisphosphonate. 19. Use according to claim 18, characterized in that the bisphosphonate is selected from aledronate, pamidronate, ibandronate, etidronate, clodronate, zoledronate, tiludronate, neridronate, olpadronate and risedronate. 20. Use, according to any one of claims 1 to 19, characterized by the fact that the individual has type I, III, IV or VII OI. 21. Use of an antibody that binds to and neutralizes the transforming growth factor beta (TGFβ) TGFβ1, TGFβ2 and TGFβ3, characterized in that it is for the manufacture of a drug for the treatment of osteogenesis imperfecta (OI) in an individual in need thereof, wherein the antibody comprises a heavy chain comprising SEQ ID NO: 14, and the light chain comprising SEQ ID NO: 15.