Variants of human BMP7 protein

Abstract

The present invention relates to novel variants of human BMP7 protein. The invention embodies vectors and host cells for the propagation of nucleic acid sequences encoding said proteins and the production thereof. Also disclosed are methods for the treatment of cancer, cartilage damage and degeneration, pain associated with osteoarthritis, or bone healing.

Claims

1. A protein comprising a polypeptide comprising the amino acid sequence of: TABLE-US-00020 (SEQ ID NO: 3) STGSKQRSQNRSKTPKNQEALRMANVAENSSSXaa.sub.33QRQXaa.sub.37CK KHELYVSFRDLGWQDWIIAPXaa.sub.60GYAAXaa.sub.65YCEGECAFPLNSY MNATNHAXaa.sub.86Xaa.sub.87QXaa.sub.89LXaa.sub.91HXaa.sub.93Xaa.sub.94NPETVP KPCCAPTQLXaa.sub.110AISXaa.sub.114LYFDDXaa.sub.120SNVILKKXaa.sub.128 RNMXaa.sub.132VXaa.sub.134ACGCH, wherein: Xaa.sub.33 is D or M; Xaa.sub.37 is A or P; Xaa.sub.60 is E or Q; Xaa.sub.65 is Y, S, or G; Xaa.sub.86 is I, V, or L; Xaa.sub.87 is V or L; Xaa.sub.89 is T, S, or A; Xaa.sub.91 is V or M; Xaa.sub.93 is F or V; Xaa.sub.94 is I, F, or M; Xaa.sub.110 is G; Xaa.sub.114 is V or M; Xaa.sub.120 is S or Q; Xaa.sub.128 is Y, F, or W; Xaa.sub.132 is V, Q, or S; and Xaa.sub.134 is R or K.

2. The protein of claim 1, wherein Xaa.sub.33 is D; Xaa.sub.37 is A; Xaa.sub.60 is E; Xaa.sub.87 is V; Xaa.sub.89 is T or A; Xaa.sub.91 is V; Xaa.sub.94 is I; Xaa.sub.120 is S; Xaa.sub.132 is V; and Xaa.sub.134 is R.

3. The protein of claim 2, wherein Xaa.sub.65 is Y or G; and Xaa.sub.86 is I or L.

4. The protein of claim 3, wherein Xaa.sub.65 is G; Xaa.sub.86 is L; Xaa.sub.89 is T; Xaa.sub.93 is V; and Xaa.sub.128 is F or W.

5. The protein of claim 4, wherein Xaa.sub.114 is V.

6. The protein of claim 5, wherein the amino acid sequence of the polypeptide is SEQ ID NO: 8.

7. The protein of claim 1, wherein the amino acid sequence of the polypeptide is selected from the group consisting of SEQ ID NO: 4, SEQ ID NO: 5, SEQ ID NO: 6, SEQ ID NO: 7, SEQ ID NO: 9, and SEQ ID NO: 10.

8. The protein of claim 1, wherein the N-terminus of the polypeptide is covalently fused to the C-terminus of a human BMP7 pro-domain polypeptide comprising an amino acid sequence of SEQ ID NO: 18.

9. The protein of claim 1, wherein the amino acid sequence of the polypeptide is selected from the group consisting of SEQ ID NO: 12, SEQ ID NO: 13, SEQ ID NO: 14, SEQ ID NO: 15, and SEQ ID NO: 16.

10. A pharmaceutical composition comprising a protein of claim 1, and one or more pharmaceutically acceptable excipients, diluents, or carriers.

11. The pharmaceutical composition of claim 10, wherein the protein is a disulfide linked homodimer.

12. The pharmaceutical composition of claim 11, wherein the composition further comprises a polypeptide having the amino acid sequence of SEQ ID NO: 18.

13. A method of treatment for cartilage damage and degeneration, pain associated with osteoarthritis, or bone healing comprising administering to a human patient in need thereof an effective amount of the protein of claim 1.

14. A method of treatment for cartilage damage and degeneration, pain associated with osteoarthritis, or bone healing comprising administering to a human patient in need thereof an effective amount of the pharmaceutical composition of claim 10.

15. A method of treatment for glioblastoma comprising administering to a human patient in need thereof an effective amount of the protein of claim 1 and an effective amount of all-trans retinoic acid (ATRA).

16. A method of treatment for glioblastoma comprising administering to a human patient in need thereof an effective amount of the pharmaceutical composition of claim 10 and an effective amount of all-trans retinoic acid (ATRA).

Description

EXAMPLE 1

Expression Methods

(1) The human BMP7 proteins and variants of human BMP7 proteins of the present invention may be produced by culturing a host cell transformed with an expression vector containing a nucleic acid encoding a human BMP7 protein or variant thereof, under the appropriate conditions to induce or cause expression of the human BMP7 protein or variant thereof. Either transient or stable transfection methods may be used. The conditions appropriate for expression of BMP7 proteins or variants thereof may vary with the choice of the expression vector and the host cell, and may be ascertained by one skilled in the art through routine experimentation.

EXAMPLE 2

Stimulation of RAR Alpha and RAR Gamma by BMP7

(2) 3T3-L1 fibroblasts and MFc7 mouse renal myofibroblasts may be maintained under routine cell culture conditions in DMEM/10% calf serum medium or OPTI-MEM/10% FBS medium, respectively. The cells may be plated in 10 cm tissue culture plates at a seeding density of 500,000 cells/plate. After 48 hours, the cells may be washed with PBS prior to the addition of either DMEM/2% dialyzed FBS or OPTI-MEM/2% dialyzed FBS medium containing 1 μg/mL human BMP7 protein or 1 μg/mL of a human BMP7 protein variant. The cells may be incubated for 48 hours and then rinsed with ice-cold PBS, and then lysed with Protein Extraction Reagent. Then lysates may be centrifuged at 14,000×g for 15 minutes. A Bradford protein assay may be performed to determine the protein concentration of each lysate. Equivalent amounts of protein (for example, 25 μg) may be resolved by SDS-PAGE and then transferred to nitrocellulose membranes for Western Blot analysis. The membranes may be probed with an anti-RAR alpha antibody and/or an anti-RAR gamma antibody.

(3) In experiments conducted essentially as described above in this Example 2, the intensity of the Western Blots demonstrated that with both MFc7 and 3T3-L1 cell lines, human BMP7 protein and human BMP7 protein variants, including human BMP7 protein variant F9, stimulated RAR alpha and RAR gamma protein expression (data not shown). Accordingly, RAR-alpha and RAR-gamma appear to be involved in a major signaling pathway for BMP7.

EXAMPLE 3

Combination Treatment with Human BMP7 Proteins and ATRA in 3T3-L1 and MFc7 Cells

(4) 3T3-L1 fibroblasts and MFc7 mouse renal myofibroblasts may be grown essentially as described in Example 2 above. After 48 hours the cells may be washed with PBS prior to the addition of either DMEM/2% dialyzed FBS or OPTI-MEM/2% dialyzed FBS medium containing control vehicle, 1 μg/ml human proBMP7 protein, 1 μM ATRA or a combination of both human BMP7 protein (1 μg/ml) and ATRA (1 μM) may be added for 48 hours. After treatment the cells may be rinsed with PBS and stored at −80° C. for approximately 20 minutes. The cells may be thawed at 37° C. for 30 minutes and 100 μl of PNPP may be added to every well. The plate may be incubated at 37° C. for 1 hour. The absorbance at 405 nm may be read on a plate reader.

(5) In experiments conducted essentially as described above in this Example 3, the combination treatment of wild type human pro-BMP7 protein and ATRA resulted in a synergistic stimulation of alkaline phosphatase. The data shown in Table 1 represent the fold change in alkaline phosphatase stimulation relative to control vehicle treated cells.

(6) TABLE-US-00004 TABLE 1 Treatment 3T3-L1 MFc7 Control 1.0 1 Wild type human pro-BMP7 1.2 21 ATRA 2.9 8 COMBO 12.4 103

EXAMPLE 4

SOX2 and Ki67 Expression in GBM Cells Treated with a Human BMP7 Protein Variant in Combination with ATRA

(7) SOX2 is a transcription factor expressed by neural stem cells. Its expression is lost when a cell differentiates. Therefore, SOX2 is considered to be a marker for terminal differentiation of multipotent brain tumor cells.

(8) Ki67 is a nuclear protein expressed in proliferating cells. It is preferentially expressed in late G1, S, G2 and M-phase of the cell cycle, and G0 or quiescent cells are negative for this protein. Fast growing cell lines have a high percentage of Ki67 positive cells. Ki67 expression is reduced or lost as cells differentiate to indicate that growth is slowing down as the cell population becomes terminally differentiated.

(9) Preparation of Glioblastoma Stem Cell Culture:

(10) Glioblastoma multiforme stem cells (GBMs) may be obtained from patients with primitive brain tumors undergoing complete or partial surgical resection. These cells may be maintained as neurospheres in defined media with 3.34 g/L of DMEM and 2.66 g/L F12 reconstituted in sterile distilled water and containing 1% glucose, 0.12% sodium bicarbonate, 5 mM hepes, 2 mM L-glutamine, 4 mg/L heparin, 10 ng/mL bFGF, 20 ng/mL EGF, 0.4% BSA, 100 μg/mL apotransferrin, 25 μg/mL insulin, 60 uM putrescine, 30 nM sodium selenite, and 20 nM progesterone. Cells may be cultured at 37° C. in 5% CO.sub.2. Cells may be plated by enzymatically dispersing spheres into single cells with a 2-5 minute incubation at 37° C. with TrypLE™ Express cell dissociation enzyme. The enzyme may be quenched with Dulbecco's Phosphate Buffered Saline (DPBS) containing Ca+ and Mg+. Then the cells may centrifuged to remove TrypLE and DPBS and resuspended as single cells in defined media and counted by a Coulter Z2 Cell and Particle counter.

(11) Experimental Procedure:

(12) Single GBM cells may be plated in defined media at 2×10.sup.6 cells/15 mL/T75 flask for high content imaging or 5×10.sup.5 cells/2 mL/well in 6-well plates for light microscopy. Cells may be treated with 0.01% DMSO, 1 μg/mL ATRA, 100 ng/mL of a human BMP7 protein variant of the present invention (e.g., human BMP7 protein variant F9 (SEQ ID NO: 8)), or a combination of 1 μg/mL ATRA and 100 ng/mL human BMP7 protein variant of the present invention. For light microscopy imaging of the effect of control BMPs (BMP2 and BMP4), cells may be plated as above in 6-well dishes and treated with 0.01% DMSO, 1 μg/mL ATRA, 100 ng/mL BMP2, 50 ng/mL BMP4, or a combination of BMPs at the same concentrations plus 1 μg/mL ATRA. Example images may be captured at 3, 7, and 30 days post-treatment on a Leica DMIRM inverted microscope using a 20× objective, for example. For growth longer than 7 days, media and treatment may be changed approximately every 10 days beginning at day 10.

(13) After collecting medium containing floating neurospheres for each condition, the spheres may be pelleted and incubated with TrypLE to disperse to single cells as described in Preparation 2 while attached cells for each condition may be detached using TrypLE and are then added back to the dispersed neurospheres. Cells may be plated into poly-D-lysine coated 96-well plates at a density of 5,000-10,000 cells/well in 100 μL defined media. Cells may be treated again with 0.01% DMSO, ATRA, human BMP7 protein, or a combination of the two as above, and incubated for an additional 48 hours at 37° C. in 5% CO.sub.2. Cells may be fixed with 3.7% formaldehyde for approximately 20 minutes. All dilutions and washes may be performed in PBS. Cells may be permeabilized with 0.1% Triton X-100 (polyethylene glycol octylphenyl ether) for approximately 10 minutes at 25° C. and then washed. Cells may be blocked for 1 hour in 1% bovine serum albumin (BSA) then incubated overnight with 2 μg/mL of a mouse monoclonal anti-SOX-2 antibody in 1% BSA or rabbit monoclonal anti-Ki67 diluted 1:500 in 1% BSA. Cells may be washed further (e.g., two times) then incubated for approximately one hour with goat α-mouse-Alexa-488 IgG or goat α-rabbit-Alexa-488 IgG and 200 ng/mL Hoechst 33342 diluted in 1% BSA solution. Cells may be washed again (e.g., two times) and cell images may be captured using an ArrayScan Vti (Cellomics, Pittsburgh, Pa.) using a 10× objective. Two-channel analysis may be performed with the Target Activation Bioapplication.

(14) In experiments conducted essentially as described above in this Example 4, GBM cells (1000-2000 cells measured per condition) were treated with either control vehicle, human pro-BMP7 protein variant alone, ATRA alone or a combination of human pro-BMP7 protein variant and ATRA. Values were normalized to vehicle control and reflect percent responders of a population. The data summarized in Table 2 represent the treatment of two different clones of GBM cells, CL-61 and CL-1. Briefly summarized, at Day 3, a clear synergistic effect was observed in CL-61 with the combination treatment of human pro-BMP7 protein variant F9 and ATRA with both the SOX2 and the Ki67 markers, indicating that the GBM stem cells are in a terminally differentiated, benign state. A similar synergistic effect was observed with the SOX2 marker on CL-1, but at the 25 day readout instead of Day 3 as seen with C61. These data illustrate the growth variability and differential response to differentiation agents of GBM stem cells, but more importantly, show surprising synergistic effects of the combination treatment of human BMP7 protein variant F9 and ATRA on biomarkers of terminal differentiation of multipotent brain tumor cells.

(15) TABLE-US-00005 TABLE 2 CL-61 CL-1 Cl-1 Day 0 Day 3 Day 25 SOX2 Control vehicle 100 100 100 ATRA (1 μg/ml) 89 97 97 Pro-BMP7 variant F9 84 100 72 (100 ng/mL) Combination 19 94 38 Ki67 Control vehicle 100 100 100 ATRA (1 μg/ml) 83 99 101 Pro-BMP7 variant F9 69 88 78 (100 ng/mL) Combination 39 66 76
Surprisingly, the combination of a variant of human pro-BMP7 protein (i.e., F9) with ATRA produced significant synergy with regard to loss of SOX2 as indicated in the treatment of glioblastoma stem cells.

EXAMPLE 5

Characterization of BMP7 Protein Variants Using the MFc7 Bioassay

(16) MFc7 mouse renal myofibroblasts (mouse kidney fibroblast cell line) may be maintained under routine cell culture conditions in OPTI-MEM/10% FBS medium. The cells may be plated in 10 cm tissue culture plates at a seeding density of 500,000 cells/plate. After approximately 48 hours, the cells may be washed with phosphate buffered saline (PBS) prior to the addition of OPTI-MEM/2% dialyzed fetal bovine serum (FBS) medium containing 1 μg/ml BMP7 protein or 1 μg/ml of a BMP7 protein variant and incubated for 48 hours. After treatment the cells may be rinsed with PBS and stored at −80° C. for 20 minutes. After thawing the cells at 37° C. for 30 minutes 100 μl of para-nitrophenyl phosphate (PNPP) may be added. The plate may be incubated at 37° C. for 1 hour. The absorbance at 405 nm may be read utilizing a plate reader. The average relative potency (EC.sub.50 of wild type BMP7 protein/EC.sub.50 of the BMP7 protein variant) of certain human BMP7 protein variants of the present invention may be calculated based on such readings.

(17) In experiments conducted essentially as described above in this Example 5, various human BMP7 protein variants of the present invention have an increased average relative potency or specific activity relative to the corresponding wild type human BMP7 protein. More specifically, the increases in average relative potency or specific activity relative to wild type human BMP7 for various human BMP7 protein variants with multiple amino acid position changes are provided in Table 3 while Table 4 provides the same for various human BMP7 protein variants with a single amino acid position change. These data demonstrate that certain human BMP7 protein variants of the present invention have an increased average relative potency or specific activity relative to the corresponding wild type human BMP7 protein.

(18) TABLE-US-00006 TABLE 3 Name of Human Average Human BMP7 BMP7 Relative protein variant SEQ ID NO: protein Potency F93V/N110G SEQ ID NO: 4 9 Y65G/I86L/T89A/ SEQ ID NO: 5 10 N110G Y65G/I86L/N110G/ SEQ ID NO: 6 7 Y128F Y65G/I86L/N110G/ SEQ ID NO: 7 16 Y128W Y65G/I86L/F93V/ SEQ ID NO: 8 F9 69 N110G/Y128W Y65G/T89A/N110G/ SEQ ID NO: 9 6.2 Y128F Y65G/I86L/N110G SEQ ID NO: 10 8.7 Y65G/V114M SEQ ID NO: 11 5.0

(19) TABLE-US-00007 TABLE 4 Human BMP7 protein variant Average (SEQ ID NO: 2 having the single Relative amino acid mutation shown here) Potency D33M 3.3 A37P 7.5 E60Q 5.5 Y65G 1.3 I86L 5.9 I86V 17.4 T89A 3 V91M 3.3 F93V 16 I94F 7.5 N110G 1.7 V114M 9.4

EXAMPLE 6

Solubility Assay for Human BMP7 Protein and Human BMP7 Protein Variants

(20) Solubility/physical stability of human BMP7 protein and human BMP7 protein variants may be measured in a stir-induced aggregation assay. Proteins are diluted to 40 μg/mL with assay buffer (50 mM sodium phosphate, 150 mM NaCl, pH 7.4) to a total volume of 2 mL. This solution may be put in a 7 mL glass vial containing one ‘flea’ stir bar and stirred at 400-rpm at room temperature. At periodic intervals (typically 0, 30, 60, 90, 120, and 150 minutes) a 150 μL aliquot may be withdrawn and centrifuged for 2 minutes at 16,000×g in a 1.5 mL tube. The supernatant (120 μL) may be transferred to an HPLC vial and the amount of remaining protein may be determined by reversed-phase HPLC under the following conditions: Zorbax C8 SB-300® column (3.5 micron, 4.6×50 mm), mobile phase: A buffer=0.1% TFA (v/v) in water, B buffer=0.085% TFA (v/v) in acetonitrile; flow rate at 1 mL/minute; column heated to 60° C.; autosampler cooled to 10° C., 214 nm UV detection, 80 μL injection, and 20 minute run time using the linear gradient indicated below. From the HPLC chromatograms (not shown) the protein peak or peaks (e.g., mature and pro domains) are integrated and percent change from initial peak area may be calculated.

(21) As shown in Table 5 below, wild type human mature BMP7 is insoluble from the initiation of the assay. In contrast, the human mature BMP7 protein variant F9 (i.e., SEQ ID NO: 8) is significantly more soluble, even at the 150 minute time point. Thus, the BMP7 protein variant F9 (i.e., SEQ ID NO: 8) provides significantly improved solubility relative to wild type mature BMP7.

(22) TABLE-US-00008 TABLE 5 Aggregation Assay: Percent of Material Precipitated Human mature Wild type Human Human mature BMP7 protein Time human pro-BMP7 BMP7 WT variant (minutes) pro-BMP7 Variant F9 (SEQ ID NO: 2) (SEQ ID NO: 8) 0 0 0 100 0 30 8 9 100 7 60 42 23 100 23 90 51 36 100 24 120 65 43 100 37 150 72 55 100 39

EXAMPLE 7

Thermal Unfolding Assay

(23) The effect of temperature on the conformational stability of human BMP7 protein variants may be followed by circular dichroism (CD) on a Jasco J-810 instrument equipped with a thermoelectric sample compartment. Briefly stated, 0.2-1.0 mg/mL pro-BMP7 may be formulated in storage buffer (10 mM citrate, 300 mM NaCl, pH 7.4) and loaded into a 0.02 cm path-length CD cuvette. The sample may be heated from 20 to 80° C. at a linear rate of 1° C./minute and the resulting CD signal at 208 nm may be recorded every 0.2° C. using a 1 second signal response time. A non-linear fit to equation 1 may be performed by the program JMP (SAS Institute Inc, Cary, N.C.) to obtain the thermal unfolding mid-point (Tm).

(24) $\begin{array}{cc}\frac{\begin{array}{c}\left(\mathrm{Yu}+\mathrm{Mu}*T\right)*\mathrm{Exp}\left[-\left[\left(\mathrm{Hm}-T*S\right)/\left(1.987*T\right)\right]\right]+\\ \mathrm{Yf}+\mathrm{Mf}*T\end{array}}{1+\mathrm{Exp}\left[-\left[\left(\mathrm{Hm}-T*S\right)/\left(1.987*T\right)\right]\right]}& \mathrm{Equation}1\end{array}$
where, Yu and Yf are fit and represent the Y-intercept of the pre- and post-transition baselines, respectively. Mu and Mf are fit and represent the slope of the pre- and post-transition baselines, respectively. Hm and S are fit and represent enthalpy and entropy, respectively. T is the measured temperature in Kelvin units.

(25) In experiments conducted essentially as described above in this Example 7, the Tm values for human pro-BMP7 wild type and human pro-BMP7 protein variant V114M are 61° C. and 64.5° C., respectively, indicating that a variant of human pro-BMP7 of the present invention has an increased thermal unfolding stability, i.e., is more stable than wild type human pro-BMP7.

EXAMPLE 8

Ectopic Bone Formation Model

(26) Ectopic bone formation (EBF) may be measured in female CB17SC-M SCID mice upon subcutaneous administration of certain human BMP7 protein variants of the present invention compared to wild type human mature BMP7.

(27) Briefly described, the mice may be anesthetized with 3% isoflurane and are injected with 3 μg/100 μl of the human BMP7 protein variant, wild type human mature BMP7 protein or a vehicle control on the back left flank. The back right flank of each mouse may serve as a control (no injection). Vehicle (pH 4.5): 0.5% Sucrose, 2.5% Glycine, 5 mM L-Glutamic Acid, 5 mM NaCl, 0.01% polysorbate 80, and, 0.1% BSA.

(28) EBF may be measured by a CT scan on Day 13 after injection of the human BMP7 protein variant or wild type human mature BMP7 protein.

(29) The data obtained from experiments conducted essentially as described above in this Example 8 suggests that despite increased potency of certain human BMP7 protein variants of the present invention, they demonstrate less EBF capability (see Table 6).

(30) TABLE-US-00009 TABLE 6 Ectopic Bone Formation Volume Group (mm.sup.3) Wild type human mature BMP7 (SEQ ID NO: 2) 78 Human mature BMP7 variant (SEQ ID NO: 7) 11 Human mature BMP7 variant F9 (SEQ ID NO: 8) 7

EXAMPLE 9

MIA Rat Model of Osteoarthritic Pain

(31) The effect of human mature BMP7 protein variant F9 (SEQ ID NO: 8) may be evaluated in the monosodium iodoacetate (MIA)-induced rat model of OA pain (Bove, et. al., Osteoarthritis and Cartilage, 2003, 11: 821-830).

(32) Briefly stated, male Lewis rats aged 7-8 weeks at the time of MIA injection may be used for this study. For induction of osteoarthritis, the rats may be anesthesized with isoflurane and given a single intra-articular injection of 0.3 mg/50 μl MIA through the infrapatellar ligament of the right knee. The left contralateral control knee may be injected with 50 μl sterile saline. The animals are treated with wild type human mature BMP7 protein at a dose of 1.0 μg/knee and human mature BMP7 protein variant F9 (SEQ ID NO: 8) at 0.175 μg and 0.7 μg/knee via intra-articular injections on day 9 and 14 post-MIA.

(33) Pain measurements may be made by incapacitance testing on days 17, 24, 31, 38, 45, 52, 58, and 66 post-MIA injection using an incapacitance tester (Columbus Instruments International, Columbus, Ohio). Changes in hind paw weight distribution between the osteoarthritic (right) and contralateral control (left) limb may be utilized as an index of joint discomfort (measure of pain) in the osteoarthritic knee. Results may be presented as the difference in weight bearing between the contralateral control (left) and the osteoarthritic (right).

(34) In experiments conducted essentially as described above in this Example 9, treatment with human mature BMP7 protein variant F9 (SEQ ID NO: 8) at a dose of 0.175 μg/knee shows a significant decrease in pain beginning at day 38 and persisting, with the exception of day 45, until day 66 post-MIA compared to vehicle-treated controls. On the other hand, wild type human mature BMP7 (SEQ ID NO: 2) at a dose of 1 μg only shows a significant decrease in pain beginning at day 58. The data indicates that human mature BMP7 protein variant F9 (SEQ ID NO: 8) is more potent in decreasing pain compared to wild type human mature BMP7 (SEQ ID NO: 2)(see Table 7).

(35) TABLE-US-00010 TABLE 7 Treatment Day 17 Day 24 Day 31 Day 38 Day 45 Day 52 Day 58 Day 66 Vehicle 23.28 ± 3.60.sup.a 23.30 ± 2.01 23.15 ± 2.24 22.48 ± 0.81 22.19 ± 1.07 22.75 ± 0.49 21.72 ± 1.69 22.14 ± 1.83 Wild type 27.52 ± 2.81 26.05 ± 2.81 23.46 ± 1.23 22.68 ± 0.61 20.95 ± 0.87 20.37 ± 0.71 17.66 ± 1.79.sup.b 17.59 ± 2.68.sup.b BMP7 1.0 μg SEQ ID NO: 8 23.29 ± 1.31 24.32 ± 2.97 21.87 ± 2.14 20.82 ± 1.03.sup.b 19.44 ± 0.80 19.35 ± 0.77 .sup.b 17.42 ± 3.16.sup.b 15.00 ± 1.84.sup.b 0.175 μg SEQ lD NO: 8 36.75 ± 6.45.sup.b 29.78 ± 8.39 26.62 ± 6.07 23.63 ± 0.90 21.75 ± 4.34 20.51 ± 3.03 16.72 ± 0.79.sup.b 13.06 ± 1.78.sup.b 0.7 μg Values are mean ± SD n = 6 per group; .sup.aDifference in hind paw weight bearing (g) .sup.bP <0.008 vs vehicle using a repeated measures analysis and treatment comparison by time point

EXAMPLE 10

Meniscal Tear Rat Model of Cartilage Degeneration

(36) The effect of human mature BMP7 protein variant (SEQ ID NO: 8) may be evaluated using an osteoarthritic (OA) pain in the meniscal tear (MT) induced rat model of cartilage degeneration. Male Lewis rats (approximately 25 weeks of age) may be used for such a study. Briefly described, rats may be anesthetized with 3% isoflurane prior to surgery. The right knee may be flexed and a transverse medial incision is made along the proximal antero-medial aspect of the tibia exposing the medial contralateral ligament. The contralateral ligament and the joint capsule may be incised simultaneously freeing the tibial attachments of the medial meniscus. Approximately 3 mm of the meniscus may be freed from its attachment to the margin of the tibial plateau and the meniscus may be transected to simulate a complete tear. The incision may be closed with surgical glue.

(37) The animals may be treated with wild type human mature BMP7 at a dose of 350 ng/knee or human mature BMP7 protein variant F9 (SEQ ID NO: 8) at two doses of 49 ng or 245 ng/knee in 50 μl phosphate buffered saline (PBS), pH 7.4. The treatment may be initiated 3 weeks after MT surgery and continued for approximately 8 weeks. Animals may be administered weekly intra-articular injections of wild type human mature BMP7 or human mature BMP7 protein variant F9 (SEQ ID NO: 8) or vehicle at various doses for 5 weeks into the operated knee. Pain measurements may be made by incapacitance testing on days 18 (baseline), 25, 31, 42 and 53 post MT surgery using an incapacitance tester (Columbus Instruments International, Columbus, Ohio). Changes in hind paw weight distribution between the osteoarthritic (right) and contralateral control (left) limb may be utilized as an index of joint discomfort (measure of pain) in the osteoarthritic knee. Results may be presented as the difference in weight bearing between the contralateral control (left) and the osteoarthritic (right).

(38) In experiments conducted essentially as described above in this Example 10, MT surgery in the right knee results in an increase in joint discomfort as defined by change in the hind paw weight distribution (measure of pain)(see Table 8). Human mature BMP7 protein variant F9 (SEQ ID NO: 8) at a dose of 245 ng/knee or a dose of 49 ng/knee showed a significant decrease in pain compared to vehicle-treated controls beginning at day 31 and persisting until day 53. Treatment with wild type human mature BMP7 at a dose of 350 ng/knee showed a decrease in pain at only 42 days post-MT surgery compared to vehicle treated controls. The data indicates that treatment with human mature BMP7 protein variant F9 (SEQ ID NO: 8) is more effective in decreasing pain compared to wild type human mature BMP7 in the rat MT-induced model of OA pain.

(39) TABLE-US-00011 TABLE 8 Baseline Day 25 Day 31 Day 42 Day 53 A .sup. 51.62 ± 3.08.sup.a 51.35 ± 1.34 53.42 ± 2.10 .sup. 52.5 ± 1.34 49.57 ± 2.35 B 52.33 ± 2.45 50.76 ± 1.09 49.97 ± 1.10 46.81 ± 1.34.sup.b 47.77 ± 1.78 C 52.48 ± 2.11 49.96 ± 0.66 .sup. 49.52 ± 1.10.sup.b  44.8 ± 3.04.sup.b 45.70 ± 2.77 D 52.02 ± 2.19 49.77 ± 1.00 .sup. 49.54 ± 2.77.sup.b 42.37 ± 1.28.sup.b .sup. 38.68 ± 1.68.sup.b Treatment A: Vehicle Treatment B: Wild type human mature BMP7 (SEQ ID NO: 2); 350 ng Treatment C: Human mature BMP7 protein variant F9 (SEQ ID NO: 8); 49 ng Treatment D: Human mature BMP7 protein variant F9 (SEQ ID NO: 8); 245 ng Values are mean ± SD, n = 6 per group .sup.aDifference in hind paw weight bearing (g) .sup.bP < 0.05 vs vehicle using a repeated measures analysis and treatment comparison by time point

EXAMPLE 11

Proteoglycan Synthesis in Human Articular Osteoarthritic Chondrocytes

(40) Proteoglycan synthesis in human articular osteoarthritic (OA) chondrocytes is an n vitro model for chondrocyte activity. The effect of human mature BMP7 protein variants may also be evaluated on proteoglycan synthesis m human articular OA chondrocytes in vitro and compared to wild type human mature pro-BMP7. The proteoglycan synthesis may be measured using .sup.35S incorporation. Arthritic human knee cartilage is obtained from donors at surgery. The cartilage pieces may be finely chopped and chondrocytes may be isolated from the associated matrix by enzymatic digestions. The cartilage may be first digested with 1 mg/ml pronase in DMEM/PRF (phenol red free) media with 5% Fe supplemented calf serum (FCS) and 2% penicillin-streptomycin-antimycotic for 60 minutes followed by overnight digestion with 1 mg/ml collagenase II in DMEM/PRF media with 5% FCS) and 2% penicillin-streptomycin-antimycotic at 37° C. The cells may be washed with DMEM/F-12 media and then resuspended in DMEM/F-12 with 5% FCS and counted with a Coulter counter. The cells may be plated at a density of 30,000 cells/well in 96-well collagen coated CytoStar T plate in growth media containing DMEM, 5% FCS. ITS (insulin, transferin, selenium) and 1% penicillin-streptomycin-antimycotic. After 24 hours, the media may be replaced with 100 μl of growth media containing 10 μCi/ml (1 μCi/well) of .sup.35S and treated with wild type human pro-BMP7 or a human pro-BMP7 protein variants at various doses. Then the cells may be incubated for approximately 7 days at 37° C. with 5% CO.sub.2. At the end of the treatment period the media may be removed, replaced with phosphate buffered saline and .sup.35S incorporation is counted using a Wallac 1450 MicroBeta TriLux Liquid Scintillation Counter & Luminometer.

(41) In experiments conducted essentially as described above in this Example 11, cells treated with human pro-BMP7 protein variant (SEQ ID NO: 16) demonstrate a significant increase in proteoglycan synthesis at doses ranging between 1.2 to 300 ng/ml (see Table 9). In contrast, wild type human pro-BMP7 shows a significant increase only at the 300 ng/ml dose. Thus, human pro-BMP7 protein variant (SEQ ID NO: 16) is more potent in stimulating proteoglycan synthesis compared to wild type human pro-BMP7 in rat primary human OA articular chondrocytes.

(42) TABLE-US-00012 TABLE 9 Proteoglycan Synthesis* (CPMs) Human wild SEQ ID Treatment (ng) Control type pro-BMP7 NO: 16 0 580 ± 84 NT NT 0.1 NT 679 ± 120 654 ± 93  0.4 NT 724 ± 120 .sup. 729 ± 133 1.2 NT 679 ± 120 1197 ± 98.sup.a  3.7 NT 666 ± 130 1991 ± 224.sup.a 11 NT 723 ± 72  3822 ± 263.sup.a 33 NT 799 ± 65  4191 ± 355.sup.a 100 NT 1500 ± 309  4235 ± 202.sup.a 300 NT 3325 ± 237.sup.a  4219 ± 75.sup.a  Values are mean ± SD; n = 3/group *Proteoglycan synthesis measured by .sup.35S incorporation .sup.aP < 0.05 vs untreated control CPM—counts per minute NT—Not tested

EXAMPLE 12

Chondrogenesis Assay

(43) The effect of wild type human mature BMP7 and variants thereof may be evaluated for differentiation of chondrocytes in an in vitro model utilizing rat primary articular chondrocytes (RPACs). To obtain RPACs, articular cartilage may be isolated from 2-3 day old rats. The cartilage may be digested with 0.4% collagenase for 2 hours and then the resultant cells may be washed with phosphate buffered saline and subsequently cultured in media containing DMEM, 10% FBS and 1% penicillin/streptomycin in humidified air with 5% CO.sub.2 at 37° C.

(44) To assess the chondrogenic differentiation, a pellet culture system may be used. Approximately 2×10.sup.5 cells from passage 2-3 may be placed in 1.5 ml tubes and centrifuged at 500 g for 10 minutes. The pellets may be cultured at 37° C. with 5% CO.sub.2 in 500 μl of medium. The effect of wild type human mature BMP7 or human mature BMP7 protein variants may be tested at various concentrations, such as 0.02, 0.2 and/or 2 μM. The media may be replaced twice per week for 2 weeks. After 2 weeks in culture, the pellets may be harvested and macroscopic analysis was performed by stereomicroscopic procedures. The images may be analyzed and the pellet sizes may be calculated in a 2-dimensional image.

(45) In experiments conducted essentially as described above in this Example 12, pellets treated with human mature BMP7 protein variant F9 (SEQ ID NO: 8) are significantly larger compared to those that were treated with wild type human mature BMP7 or untreated controls (see Table 10). The increases in pellet sizes are observed in a dose-dependent manner, but are most prominent at 0.14 and 1.4 nM concentrations. The data demonstrates that human mature BMP7 protein variant F9 (SEQ ID NO: 8) is more potent in increasing pellet sizes compared to wild type human mature BMP7 in rat primary articular chondrocytes (see Table 10). Because cell pellet sizes are indicative of increased cell proliferation, human mature BMP7 protein variant F9 (SEQ ID NO: 8) appears to be much more potent in increasing the proliferation of rat primary articular chondrocytes compared to wild type human mature BMP7.

(46) TABLE-US-00013 TABLE 10 Pellet Size Human mature (mm.sup.2) BMP7 protein Treatment WT human variant F9 (nM) Control mature BMP7 (SEQ ID NO: 8) 0 0.51 ± 0.05 NT NT 0.014 NT NT 0.67 ± 0.16 0.14 NT NT .sup. 1.56 ± 0.46.sup.a 1.4 NT NT .sup. 2.61 ± 0.18.sup.a 0.02 NT 0.64 ± 0.15 NT 0.2 NT 0.70 ± 0.03 NT 2 NT 0.83 ± 0.13 NT Values are mean ± SD; n = 3/group .sup.aP < 0.05 vs untreated control NT—Not tested

EXAMPLE 13

Bone Healing Model

(47) The effect of wild type human mature BMP7 and variants thereof may be evaluated on bone regeneration and repair or bone healing in a rat surgical fracture model. Animals may be ovariectomized at 6 months of age and allowed to lose bone for two months before fracture surgery. Cortical defect surgery may be performed essentially as previously described (Komatsu, et al., Endocrinology, 150:1570-1579, 2009). Briefly, the procedure involves incising the skin over the lateral femoral aspect and blunt dissection of the quadriceps to expose the distal femoral diaphysis. Cortical defects may be then generated through the anterior and posterior cortexes using a Dremel tool (Robert Bosch Tool Corp, Gerlingen, Germany) equipped with a 2 mm orthopedic drill bit (Zimmer Inc, Warsaw, Ind.). The muscles may be subsequently repositioned, and the skin may be closed with tissue adhesive. Groups of animals may be treated with various amounts of wild type human mature BMP7 or variants thereof prepared in sodium citrate buffer pH 3.0 and adsorbed to Helistat type-1 collagen sponges in a volume of 50 μl. The treatments may be administered locally at the site of defect during surgery and treated for 35 days. A control group may receive the vehicle containing the same constituents in collagen sponge without the therapeutic proteins. In vivo fracture repair and bone mineral content (BMC) may be evaluated by quantitative computed tomography (QCT) using a GE Locus Ultra CT scanner (GE Healthcare, London, Ontario, Canada) as described previously (Komatsu et al., 2008).

(48) In experiments conducted essentially as described above in this Example 13, treatment with 16.5 μg human mature BMP7 protein variant F9 (SEQ ID NO: 8) showed a significant increase in BMC and cortical area by forming new cortical shell on day 35 after treatment (see Table 11). On the other hand, treatment with 16.5 μg wild type human mature BMP7 did not show a significant change in BMC on day 35. The data demonstrates that human mature BMP7 protein variant F9 (SEQ ID NO: 8) is potent in increasing BMC and cortical area compared to wild type human mature BMP7 in a rat cortical defect bone healing model.

(49) TABLE-US-00014 TABLE 11 % Increase From Treatment BMC (mg) Vehicle Control A 14.11 ± 1.50 100 B 16.25 ± 1.60 115 C 16.19 ± 1.09 114 D .sup. 18.13 ± 2.29.sup.a 128 Treatment A: Vehicle Treatment B: Wild type human mature BMP7 (SEQ ID NO: 2); 16.5 μg Treatment C: Human mature BMP7 protein variant F9 (SEQ ID NO: 8); 2.0 μg Treatment D: Human mature BMP7 protein variant F9 (SEQ ID NO: 8); 16.5 μg Values are mean ± SEM; n = 6-8/group .sup.aP < 0.05 vs untreated vehicle

EXAMPLE 14

Established Cord Assay

(50) An in vitro endothelial cord formation assay, a surrogate of angiogenesis, may be used to investigate the role of human BMP7 proteins on various growth factor established cords including, but not limited to, VEGF, basic FGF, and EGF established cords. Endothelial cord forming cells (ECFCs; passage 4-10 suitable for cord formation) may be cultured in EGM-2 MV (Lona) media containing a final concentration of 10% FBS and passaged onto type 1 collagen (fibrillar) coated flasks prior to seeding into the cord formation assay in vitro. Adipocyte derived stem cells (Zen-Bio, cells frozen at passage 4; cells at passage 5 or greater not assayed) may be cultured in EGM-2 MV (Lonza) media prior to plating at 50,000 cells per well (into 96-well black poly-D-lysine coated plates) in co-culture media [for example, MCDB-131 media (Invitrogen) supplemented with 30 μg/ml L-ascorbic acid 2-phosphate, 1 μM dexamethasone, 50 μg/ml tobramycin, 10 μg/ml insulin (all from Sigma-Aldrich), and 10 μg/ml cell prime r-transferrin AF (Millipore, Billerica, Mass.)] for 24 hours. Adipocyte derived stem cell (ADSC) media may be removed and 5,000 ECFCs (Lonza) per well may be over seeded. Approximately 4 hours following ECFC plating, the cords may be treated with 10 ng/ml VEGF, 10 ng/ml bFGF, or 10 ng/ml EGF (all from Biosource International) and exposure to growth factors may be continued for approximately 120 hours. Then, PBS controls, human BMP7 proteins including human BMP7 protein variants of the present invention (100 ng/ml), or sunitinib may be added and incubated for 96 hours. All cell culture incubations may be conducted at 37° C., 5% CO.sub.2. Then the cells may be directly fixed for 10 minutes with 3.7% formaldehyde (Sigma Aldrich) followed by ice-cold 70% ethanol for 20 minutes at 25° C. Cells may be rinsed once with PBS, blocked for 30 minutes with 1% BSA and immuno-stained for 1 hour with antiserum directed against CD31 (R&D Systems. Minneapolis, Minn.) diluted to 1 μg/ml in 1% BSA. Then the cells may be washed 3 times with PBS and incubated for 1 hour with 5 μg/ml donkey α-sheep-Alexa-488 (Invitrogen), α-Smooth Muscle Actin Cy3 conjugate (1:200, Sigma-Aldrich), and 200 ng/ml Hoechst 33342 (Invitrogen) in 1% BSA. Afterwards, the cells may be washed with PBS, then imaged using the cord formation algorithms on the Cellomics ArrayScan VTI at an image magnification of 5× (Thermo Fisher Scientific, Pittsburgh, Pa.).

(51) In experiments conducted essentially as described above in this Example 14, sunitinib (100 nM), an agent with an anti-angiogenic mechanism of action, was used as a positive control. Human pro-BMP7-F9 (100, 50, 25, and 12.5 ng/ml) was determined to reduce established cord connected tube area (% PBS control) of endothelial cords that were allowed to form for 120 hours in the presence of the indicated growth factors prior to compound treatment (growth factors remained for duration of experiment).

(52) Human Pro-BMP7 Variant F9 Reduced VEGF-Driven Cord Formation

(53) An ADSC/ECFC co-culture was unstimulated (basal) or stimulated with 10 ng/ml VEGF and treated simultaneously with PBS or 100 ng/ml human pro-BMP7 variant F9 or 100 nM sunitinib for 96 hours prior to immunohistochemistry for CD31 (green), α-smooth muscle actin (red), and Hoechst 33342 to stain all nuclei (blue). Resulting CD31-positive endothelial cords were visualized and quantitated using high content imaging. Results are shown in Table 12. Values represent % PBS controls, mean±SD; n=8/treatment group, n=16 in PBS control groups.

(54) TABLE-US-00015 TABLE 12 Connected Tube Area - VEGF (% PBS) PBS  100.0 ± 14.90% BMP7F9 100 ng/ml 18.66 ± 3.72% 100 nM sunitinib 26.14 ± 9.88%
Human Pro-BMP7 Variant F9 Reduced FGF-Driven Cord Formation

(55) An ADSC/ECFC co-culture was unstimulated (basal) or stimulated with 10 ng/ml bFGF and treated simultaneously with PBS or 100 ng/ml human pro-BMP7 variant F9 or 100 nM sunitinib for 96 hours prior to immunohistochemistry for CD31 (green), α-smooth muscle actin (red), and Hoechst 33342 to stain all nuclei (blue). Resulting CD31-positive endothelial cords were visualized and quantitated using high content imaging. Results are shown in Table 13. Values represent % PBS controls, mean±SD; n=8/treatment group, n=16 in PBS control groups.

(56) TABLE-US-00016 TABLE 13 Connected Tube Area - FGF (% PBS) PBS 100.0 ± 18.67% BMP7F9 100 ng/ml 4.39 ± 2.80% 100 nM sunitinib 43.01 ± 18.28%
Human Pro-BMP7 Variant F9 Reduced EGF-Driven Cord Formation

(57) An ADSC/ECFC co-culture was unstimulated (basal) or stimulated with 10 ng/ml EGF and treated simultaneously with PBS or 100 ng/ml human pro-BMP7 variant F9 or 100 nM sunitinib for 96 hours prior to immunohistochemistry for CD31 (green), α-smooth muscle actin (red), and Hoechst 33342 to stain all nuclei (blue). Resulting CD31-positive endothelial cords were visualized and quantitated using high content imaging. Results are shown in Table 14. Values represent % PBS controls, mean±SD; n=8/treatment group, n=16 in PBS control groups.

(58) TABLE-US-00017 TABLE 14 Connected Tube Area - EGF (% PBS) PBS 100.0 ± 22.16% BMP7F9 100 ng/ml 13.21 ± 2.59%  100 nM sunitinib  9.89 ± 8.059%

(59) These results indicate that human pro-BMP7 protein variant F9 reduces existing, established-growth factor induced endothelial cords in a surrogate in vitro angiogenesis assay.

EXAMPLE 15

BMP7 Protein Variant Inhibition of Tumor Growth in Mouse Xenograft Models for Colon Cancer

(60) Xenograft animal models may be used to assess the effectiveness of the human BMP7 protein variants of the present invention (as compared to the corresponding wild type human BMP7 protein) against specific types of cancer. More specifically, compounds may be tested on staged tumor growths that have been engrafted via subcutaneous or orthotopic inoculation in an immunocompromised mouse or rat model. Xenograft studies can be highly complex, starting with the selection of the appropriate animal model, choice of tumorigenic cell line, administration method, administration regimen, dosing, analysis of tumor growth rates and tumor analysis (histology, mRNA and protein expression levels).

(61) The effect of various BMP7 protein variants on xenograft models of specific cancer were tested as described below.

(62) Female athymic nude mice age 6- to 7-weeks old are available commercially, including from Harlan Laboratories (Indianapolis, Ind.). The mice are allowed to acclimate for one week and fed ad libitum on a normal low fat (4.5%) diet, which may be continued for the duration of the study. Tumor cells are available for purchase from ATCC and may be cultured in cell culture media such as RPMI 1640 (Life Technologies) with L-glutamine, 25 mM HEPES supplemented with 10% FBS and 1 mM Na Pyruvate. Cells may be detached, washed with serum free medium and then resuspended at a final concentration of 50×106 cells/mL in serum free RPMI 1640. Tumor cells, approximately 5×106 may be injected subcutaneously in the rear flank of subject mice in a 1:1 mixture of serum free growth medium and Matrigel (Becton Dickinson, Bedford, Mass.). Tumor and body weight measurements may be performed twice weekly. Prior to treatment, mice can be randomized based on tumor size using a randomization algorithm. Treatments may be started when the average tumor volume reaches 100 mm3. The randomized mice were separated into different groups and dosed with compounds through tail vein injection once a week.

(63) All test or control proteins are prepared in phosphate Buffered Saline (PBS) prior to dose. Tumor size may be determined by caliper measurements. Tumor volume (mm3) may be estimated from the formula A2×B×0.536, where A is the smaller and B is the larger of perpendicular diameters. Tumor volume data can be transformed to a log scale to equalize variance across time and treatment groups. Log volume data can be analyzed with two-way repeated measures ANOVA by time and treatment using SAS PROC MIXED software (SAS Institutes Inc, Cary, N.C.). Treatment Groups are Compared with the Specified Control Group at Each Time Point.

(64) Immunodeficient mice bearing DLD1 C5 tumor xenografts (Wnt/TCF-1 driven colon cancer model) may be generated as described above in this Example 15 and treated with either vehicle control, the human mature BMP7 variant F9, irinotecan, or the combination of the human mature BMP7 variant F9 and irinotecan, three times/week (BMP7 variant F9) or twice/week (irinotecan) for approximately 3-6 consecutive weeks.

(65) In a DLD1 C5 mouse xenograft tumor model conducted essentially as described above in this Example 15, tumor regression was observed in a DLD1 C5 mouse xenograft model when they were pre-treated with human mature BMP7 variant F9 for three weeks followed by 3 weeks of therapy with irinotecan. Specifically, the administration of the human mature BMP7 variant F9 for 3 weeks followed by 3 weeks of therapy with irinotecan (dosed IP at 0.04 mg/kg MWF, and 20 mg/kg, M and Th, respectively) induced DLD1 C5 tumor sensitivity to the chemotherapeutic agent irinotecan (immediate tumor regression) relative to saline-treated, then irinotecan treated animals (which resulted in tumor growth).

EXAMPLE 16

Endogenous BMP7 Inhibitor Activity on BMP7 Protein Variants

(66) Endogenous BMP antagonists such as noggin, Sost, follistatin, twisted gastrulation (TSG), chordin, and members of the gremlin, Cerberus and Dan families of proteins for example act on BMP family of proteins in vivo to modulate the BMP activities including those relating to growth, differentiation and activity of a range of cells. As with BMPs themselves, expression of the various antagonists is under tight spatiotemporal control. (see, Bone Morphogenic Proteins and Their Antagonists, Vitamins and Hormones (2015), volume 99, pages 63-90; Editor Gerald Litwack). The effect of various BMP antagonists on human BMP7 proteins and variants thereof may be tested in an in vitro assay as described below in this Example 16.

(67) Briefly described, Hep3B2 cells stably transfected with a hepcidin promoter luciferase construct (hereinafter, referred to as Hep3B2_HepPro_luc cells) were used for BMP inhibitor experiments. Hep3B2_HepPro_luc cells may be plated at 30,000 cells per well in a tissue culture treated 96 well plate in DMEM (Hyclone) 5% FBS (Gibco) supplemented with non-essential amino acids (Hyclone) and 200 μg/ml gentecin (Hyclone) for 24 hours. Cells were then starved in OMEM+0.2% BSA for 5 hours. Cells were treated with a mixture of human BMP7 proteins and variants thereof in OMEM (Gibco)+0.2% BSA (Gibco) for 18 hours then developed for luciferase activity utilizing Luciferase reporter Gene Assay Kit (Roche). BMP7 proteins were added at concentrations in the linear range of the assay, at 1 nM and 100 pM. Inhibitors were added at various concentrations in molar excess of the particular BMP7 protein being tested. All BMP inhibitors except Follistatin were purchased from R&D Systems. Follistatin and BMP7 proteins and variants thereof were generated at Eli Lilly and Company.

(68) In experiments conducted essentially as described above in this Example 16, wild type BMP7 proteins were significantly more susceptible to inhibition by certain BMP antagonists as compared to the corresponding (i.e., pro- or mature) form of the BMP7 F9 variant of the present invention (see Table 15).

(69) TABLE-US-00018 TABLE 15 % inhibition Molar excess of inhibitor 100X 10X 1X 0.1X 0.01X BMP Inhibitor BMP7-F9 Pro 22.5* 11.7 −4.3 −2.6 Follistatin BMP7 WT Pro 79.8 29.5 5.9 −10.4 BMP7-F9 mature 23.8 10.0 10.6 3.9 BMP7 WT mature 54.7 21.1 −11.9 −16.6 BMP7-F9 Pro 24.9* 8.8* 2.4* 3.5* Chordin-like 2 BMP7 WT Pro 88.2 87.5 55.9 8.9 BMP7-F9 mature 9.0 −1.0 −3.2 −1.5 BMP7 WT mature 71.9 69.5 49.0 −13.7 BMP7-F9 Pro −28.1* −23.6* −18.3 −15.2 Noggin BMP7 WT Pro 87.3 84.8 24.8 −5.3 BMP7-F9 mature −1.2 −6.9 −7.2 −2.6 BMP7 WT mature 73.7 72.8 36.7 −12.2 BMP7-F9 Pro −17.7* −1.3 −5.3 −5.9 Chordin BMP7 WT Pro 76.5 18.9 −2.2 −15.1 BMP7-F9 mature 5.6 5.2 6.8 3.7 BMP7 WT mature 70.9 21.3 −9.2 −13.1 BMP7-F9 Pro 754.sup.# 7.0 −27.5 −35.4 Coco BMP7 WT Pro 81.6 17.4 4.0 −10.2 BMP7-F9 mature 28.5 7.3 7.0 2.3 BMP7 WT mature 77.4 37.6 −25.9 −23.6 *indicates inhibition of human BMP7 F9 variant form was significantly lower than inhibition the corresponding wild type form of human BMP7 protein .sup.#indicates human BMP7 F9 variant form and human WT form were inhibited about equally.

(70) TABLE-US-00019 LISTING OF VARIOUS SEQUENCES SEQ ID NO: 1 MHVRSLRAAAPHSFVALWAPLFLLRSALADFSLDNEVHSSFIHRRLRSQERREMQ REILSILGLPHRPRPHLQGKHNSAPMFMLDLYNAMAVEEGGGPGGQGFSYPYKA VFSTQGPPLASLQDSHFLTDADMVMSFVNLVEHDKEFFHPRYFIHREFRFDLSKIP EGEAVTAAEFRIYKDYIRERFDNETFRISVYQVLQEHLGRESDLFLLDSRTLWASE EGWLVFDITATSNHWVVNPRHNLGLQLSVETLDGQSINPKLAGLIGRHGPQNKQ PFMVAFFKATEVHFRSIRSTGSKQRSQNRSKTPKNQEALRMANVAENSSSDQRQ ACKKHELYVSFRDLGWQDWIIAPEGYAAYYCEGECAFPLNSYMNATNHAIVQTL VHFINPETVPKPCCAPTQLNAISVLYFDDSSNVILKKYRNMVVRACGCH SEQ ID NO: 2 STGSKQRSQNRSKTPKNQEALRMANVAENSSSDQRQACKKHELYVSFRDLGWQ DWIIAPEGYAAYYCEGECAFPLNSYMNATNHAIVQTLVHFINPETVPKPCCAPTQ LNAISVLYFDDSSNVILKKYRNMVVRACGCH SEQ ID NO: 3 STGSKQRSQNRSKTPKNQEALRMANVAENSSSXaa.sub.33QRQXaa.sub.37CKKHELYVSFRD LGWQDWIIAPXaa.sub.60GYAAXaa.sub.65YCEGECAFPLNSYMNATNHAXaa.sub.86Xaa.sub.87QXaa.sub.89 LXaa.sub.91HXaa.sub.93Xaa.sub.94NPETVPKPCCAPTQLXaa.sub.110AISXaa.sub.114LYFDDXaa.sub.120SNVILKK Xaa.sub.128RNMXaa.sub.132VXaa.sub.134ACGCH Wild type pro-domain + variant mature BMP7 (402 aa)  F93V/N110G SEQ ID NO: 12 DFSLDNEVHSSFIHRRLRSQERREMQREILSILGLPHRPRPHLQGKHNSAPMFMLD LYNAMAVEEGGGPGGQGFSYPYKAVFSTQGPPLASLQDSHFLTDADMVMSFVN LVEHDKEFFHPRYHHREFRFDLSKIPEGEAVTAAEFRIYKDYIRERFDNETFRIS VYQVLQEHLGRESDLFLLDSRTLWASEEGWLVFDITATSNHWVVNPRHNLGLQLSV ETLDGQSINPKLAGLIGRHGPQNKQPFMVAFFKATEVHFRSIRSTGSKQRSQNRSK TPKNQEALRMANVAENSSSDQRQACKKHELYVSFRDLGWQDWIIAPEGYAAYY CEGECAFPLNSYMNATNHAIVQTLVHVINPETVPKPCCAPTQLGAISVLYFDDSS NVILKKYRNMVVRACGCH Wild type pro-domain + variant mature BMP7 (402 aa)  Y65G/I86L/T89A/N110G SEQ ID NO: 13 DFSLDNEVHSSFIHRRLRSQERREMQREILSILGLPHRPRPHLQGKHNSAPMFMLD LYNAMAVEEGGGPGGQGFSYPYKAVFSTQGPPLASLQDSHFLTDADMVMSFVN LVEHDKEFFEIPRYHHREFRFDLSKIPEGEAVTAAEFRIYKDYIRERFDNETFRIS VYQVLQEHLGRESDLFLLDSRTLWASEEGWLVFDITATSNHWVVNPRHNLGLQLSV ETLDGQSINPKLAGLIGRHGPQNKQPFMVAFFKATEVHFRSIRSTGSKQRSQNRSK TPKNQEALRMANVAENSSSDQRQACKKHELYVSFRDLGWQDWIIAPEGYAAGY CEGECAFPLNSYMNATNHALVQALVHFINPETVPKPCCAPTQLGAISVLYFDDSS NVILKKYRNMVVRACGCH Wild type pro-domain + variant mature BMP7 (402 aa)  Y65G/I86L/Y128F/N110G SEQ ID NO: 14 DFSLDNEVHSSFIHRRLRSQERREMQREILSILGLPHRPRPHLQGKHNSAPMFMLD LYNAMAVEEGGGPGGQGFSYPYKAVFSTQGPPLASLQDSHFLTDADMVMSFVN LVEHDKEFFEIPRYHHREFRFDLSKIPEGEAVTAAEFRIYKDYIRERFDNETFRIS VYQVLQEHLGRESDLFLLDSRTLWASEEGWLVFDITATSNHWVVNPRHNLGLQLSV ETLDGQSINPKLAGLIGRHGPQNKQPFMVAFFKATEVHFRSIRSTGSKQRSQNRSK TPKNQEALRMANVAENSSSDQRQACKKHELYVSFRDLGWQDWIIAPEGYAAGY CEGECAFPLNSYMNATNHALVQTLVHFINPETVPKPCCAPTQLGAISVLYFDDSS NVILKKFRNMVVRACGCH Wild type pro-domain + variant mature BMP7 (402 aa)  Y65G/I86L/Y128W/N110G SEQ ID NO: 15 DFSLDNEVHSSFIHRRLRSQERREMQREILSILGLPHRPRPHLQGKHNSAPMFMLD LYNAMAVEEGGGPGGQGFSYPYKAVFSTQGPPLASLQDSHFLTDADMVMSFVN LVEHDKEFFEIPRYHHREFRFDLSKIPEGEAVTAAEFRIYKDYIRERFDNETFRIS VYQVLQEHLGRESDLFLLDSRTLWASEEGWLVFDITATSNHWVVNPRHNLGLQLSV ETLDGQSINPKLAGLIGRHGPQNKQPFMVAFFKATEVHFRSIRSTGSKQRSQNRSK TPKNQEALRMANVAENSSSDQRQACKKHELYVSFRDLGWQDWIIAPEGYAAGY CEGECAFPLNSYMNATNHALVQTLVHFINPETVPKPCCAPTQLGAISVLYFDDSS NVILKKWRNMVVRACGCH Wild type pro-domain + variant mature BMP7 (402 aa)  Y65G/I86L/F93W/N110G/Y128W SEQ ID NO: 16 DFSLDNEVHSSFIHRRLRSQERREMQREILSILGLPHRPRPHLQGKHNSAPMFMLD LYNAMAVEEGGGPGGQGFSYPYKAVFSTQGPPLASLQDSHFLTDADMVMSFVN LVEHDKEFFEIPRYHHREFRFDLSKIPEGEAVTAAEFRIYKDYIRERFDNETFRIS VYQVLQEHLGRESDLFLLDSRTLWASEEGWLVFDITATSNHWVVNPRHNLGLQLSV ETLDGQSINPKLAGLIGRHGPQNKQPFMVAFFKATEVHFRSIRSTGSKQRSQNRSK TPKNQEALRMANVAENSSSDQRQACKKHELYVSFRDLGWQDWIIAPEGYAAGY CEGECAFPLNSYMNATNHALVQTLVHVINPETVPKPCCAPTQLGAISVLYFDDSS NVILKKWRNMVVRACGCH SEQ ID NO: 17 STGSKQRSQNRSKTPKNQEALRMANVAENSSSXaa.sub.33QRQXaa.sub.37CKKHELYVSFRD LGWQDWIIAPXaa.sub.60GYAAXaa.sub.65YCEGECAFPLNSYMNATNHAXaa.sub.86VQXaa.sub.89L Xaa.sub.91HXaa.sub.93Xaa.sub.94NPETVPKPCCAPTQLXaa.sub.110AISXaa.sub.114LYFDDSSNVILKK Xaa.sub.128RNMVVRACGCH wherein at least one amino acid substitution that is: Xaa.sub.33 is D or M; Xaa.sub.37 is A or P; Xaa.sub.60 is E or Q; Xaa.sub.65 is  Y, S, or G; Xaa.sub.86 is I, V, or L; Xaa.sub.89 is T, S, or A;    Xaa.sub.91 is V or M; Xaa.sub.93 is F or V; Xaa.sub.94 is I, F or M; Xaa.sub.110 is N, A, S, or G; Xaa.sub.114 is V or M; and, Xaa.sub.128 is Y, F or W. SEQ ID NO: 18 DFSLDNEVHSSFIHRRLRSQERREMQREILSILGLPHRPRPFILQGKHNSAPMFMLD LYNAMAVEEGGGPGGQGFSYPYKAVFSTQGPPLASLQDSHFLTDADMVMSFVN LVEHDKEFFHPRYFIHREFRFDLSKIPEGEAVTAAEFRIYKDYIRERFDNETFRIS VYQVLQEHLGRESDLFLLDSRTLWASEEGWLVFDITATSNHWVVNPRHNLGLQLSV ETLDGQSINPKLAGLIGRHGPQNKQPFMVAFFKATEVHFRSIR (DNA SEQUENCE CORRESPONDING TO SEQ ID NO: 7) SEQ ID NO: 19 ATGCACGTGCGCAGCCTGCGCGCCGCCGCCCCCCACAGCTT CGTGGCCCTGTGGGCCCCCCTGTTCCTGCTGCGCAGCGCCCTGGCCGACT TCAGCCTGGACAACGAGGTGCACAGCAGCTTCATCCACCGCCGCCTGCGC AGCCAGGAGCGCCGCGAGATGCAGCGCGAGATCCTGAGCATCCTGGGCCT GCCCCACCGCCCCCGCCCCCACCTGCAGGGCAAGCACAACAGCGCCCCCA TGTTCATGCTGGACCTGTACAACGCCATGGCCGTGGAGGAGGGCGGCGGC CCCGGCGGCCAGGGCTTCAGCTACCCCTACAAGGCCGTGTTCAGCACCCA GGGCCCCCCCCTGGCCAGCCTGCAGGACAGCCACTTCCTGACCGACGCCG ACATGGTGATGAGCTTCGTGAACCTGGTGGAGCACGACAAGGAGTTCTTC CACCCCCGCTACCACCACCGCGAGTTCCGCTTCGACCTGAGCAAGATCCC CGAGGGCGAGGCCGTGACCGCCGCCGAGTTCCGCATCTACAAGGACTACA TCCGCGAGCGCTTCGACAACGAGACCTTCCGCATCAGCGTGTACCAGGTG CTGCAGGAGCACCTGGGCCGCGAGAGCGACCTGTTCCTGCTGGACAGCCG CACCCTGTGGGCCAGCGAGGAGGGCTGGCTGGTGTTCGACATCACCGCCA CCAGCAACCACTGGGTGGTGAACCCCCGCCACAACCTGGGCCTGCAGCTG AGCGTGGAGACCCTGGACGGCCAGAGCATCAACCCCAAGCTGGCCGGCCT GATCGGCCGCCACGGCCCCCAGAACAAGCAGCCCTTCATGGTGGCCTTCT TCAAGGCCACCGAGGTGCACTTCCGCAGCATCCGCAGCACCGGCAGCAAG CAGCGCAGCCAGAACCGCAGCAAGACCCCCAAGAACCAGGAGGCCCTGCG CATGGCCAACGTGGCCGAGAACAGCAGCAGCGACCAGCGCCAGGCCTGCA AGAAGCACGAGCTGTACGTGAGCTTCCGCGACCTGGGCTGGCAGGACTGG ATCATCGCCCCTGAGGGCTACGCCGCCGGCTACTGCGAGGGCGAGTGCGC CTTCCCCCTGAACAGCTACATGAACGCCACCAACCACGCCCTGGTGCAGA CCCTGGTGCACTTCATCAACCCCGAGACCGTGCCCAAGCCCTGCTGCGCC CCCACCCAGCTGGGCGCCATCAGCGTGCTGTACTTCGACGACAGCAGCAA CGTGATCCTGAAGAAGTGGCGCAACATGGTGGTGCGCGCCTGCGGCTGCC AC (DNA SEQUENCE CORRESPONDING TO SEQ ID NO: 8) SEQ ID NO: 20 ATGCACGTGCGCAGCCTGCGCGCCGCCGCCCCCCACAGCTTCGTGGCCCT GTGGGCCCCCCTGTTCCTGCTGCGCAGCGCCCTGGCCGACTTCAGCCTGG ACAACGAGGTGCACAGCAGCTTCATCCACCGCCGCCTGCGCAGCCAGGAG CGCCGCGAGATGCAGCGCGAGATCCTGAGCATCCTGGGCCTGCCCCACCG CCCCCGCCCCCACCTGCAGGGCAAGCACAACAGCGCCCCCATGTTCATGC TGGACCTGTACAACGCCATGGCCGTGGAGGAGGGCGGCGGCCCCGGCGGC CAGGGCTTCAGCTACCCCTACAAGGCCGTGTTCAGCACCCAGGGCCCCCC CCTGGCCAGCCTGCAGGACAGCCACTTCCTGACCGACGCCGACATGGTGA TGAGCTTCGTGAACCTGGTGGAGCACGACAAGGAGTTCTTCCACCCCCGC TACCACCACCGCGAGTTCCGCTTCGACCTGAGCAAGATCCCCGAGGGCGA GGCCGTGACCGCCGCCGAGTTCCGCATCTACAAGGACTACATCCGCGAGC GCTTCGACAACGAGACCTTCCGCATCAGCGTGTACCAGGTGCTGCAGGAG CACCTGGGCCGCGAGAGCGACCTGTTCCTGCTGGACAGCCGCACCCTGTG GGCCAGCGAGGAGGGCTGGCTGGTGTTCGACATCACCGCCACCAGCAACC ACTGGGTGGTGAACCCCCGCCACAACCTGGGCCTGCAGCTGAGCGTGGAG ACCCTGGACGGCCAGAGCATCAACCCCAAGCTGGCCGGCCTGATCGGCCG CCACGGCCCCCAGAACAAGCAGCCCTTCATGGTGGCCTTCTTCAAGGCCA CCGAGGTGCACTTCCGCAGCATCCGCAGCACCGGCAGCAAGCAGCGCAGC CAGAACCGCAGCAAGACCCCCAAGAACCAGGAGGCCCTGCGCATGGCCAA CGTGGCCGAGAACAGCAGCAGCGACCAGCGCCAGGCCTGCAAGAAGCACG AGCTGTACGTGAGCTTCCGCGACCTGGGCTGGCAGGACTGGATCATCGCC CCTGAGGGCTACGCCGCCGGCTACTGCGAGGGCGAGTGCGCCTTCCCCCT GAACAGCTACATGAACGCCACCAACCACGCCCTGGTGCAGACCCTGGTGC ACGTGATCAACCCCGAGACCGTGCCCAAGCCCTGCTGCGCCCCCACCCAG CTGGGCGCCATCAGCGTGCTGTACTTCGACGACAGCAGCAACGTGATCCT GAAGAAGTGGCGCAACATGGTGGTGCGCGCCTGCGGCTGCCAC