Scholarly article on topic 'Intra-articular therapy with recombinant human GDF5 arrests disease progression and stimulates cartilage repair in the rat medial meniscus transection (MMT) model of osteoarthritis'

Intra-articular therapy with recombinant human GDF5 arrests disease progression and stimulates cartilage repair in the rat medial meniscus transection (MMT) model of osteoarthritis Academic research paper on "Biological sciences"

CC BY-NC-ND
0
0
Share paper
Academic journal
Osteoarthritis and Cartilage
OECD Field of science
Keywords
{Osteoarthritis / "Growth and differentiation factor" / "Intraarticular therapy" / GDF5 / "Medial meniscus transection" / "Cartilage repair"}

Abstract of research paper on Biological sciences, author of scientific article — W.R. Parrish, B.A. Byers, D. Su, J. Geesin, U. Herzberg, et al.

Summary Objective Investigation of osteoarthritis (OA) risk alleles suggests that reduced levels of growth and differentiation factor-5 (GDF5) may be a precipitating factor in OA. We hypothesized that intra-articular recombinant human GDF5 (rhGDF5) supplementation to the OA joint may alter disease progression. Methods A rat medial meniscus transection (MMT) joint instability OA model was used. Animals received either one intra-articular injection, or two or three bi-weekly intra-articular injections of either 30 μg or 100 μg of rhGDF5 beginning on day 21 post surgery after structural pathology had been established. Nine weeks after MMT surgery, joints were processed for histological analysis following staining with toluidine blue. Control groups received intra-articular vehicle injections, comprising a glycine-buffered trehalose solution. OA changes in the joint were evaluated using histopathological end points that were collected by a pathologist who was blinded to treatment. Results Intra-articular rhGDF5 supplementation reduced cartilage lesions on the medial tibial plateau in a dose-dependent manner when administered therapeutically to intercept OA disease progression. A single 100 μg rhGDF5 injection on day 21 slowed disease progression at day 63. A similar effect was achieved with two bi-weekly injections of 30 μg. Two bi-weekly injections of 100 μg or three bi-weekly injections of 30 μg stopped progression of cartilage lesions. Importantly, three biweekly injections of 100 μg rhGDF5 stimulated significant cartilage repair. Conclusions Intra-articular rhGDF5 supplementation can prevent and even reverse OA disease progression in the rat MMT OA model. Collectively, these results support rhGDF5 supplementation as an intra-articular disease modifying OA therapy.

Academic research paper on topic "Intra-articular therapy with recombinant human GDF5 arrests disease progression and stimulates cartilage repair in the rat medial meniscus transection (MMT) model of osteoarthritis"

Accepted Manuscript

Intra-articular therapy with recombinant human GDF5 arrests disease progression and stimulates cartilage repair in the rat medial meniscus transection (MMT) model of osteoarthritis

William R. Parrish, Benjamin A. Byers, Dongling Su, Jeffrey Geesin, Uri Herzberg, Scott Wadsworth, Alison Bendele, Brooks Story

PII: S1063-4584(16)30389-2

DOI: 10.1016/j.joca.2016.11.002

Reference: YJOCA 3886

To appear in: Osteoarthritis and Cartilage

Received Date: 2 June 2016 Revised Date: 23 September 2016 Accepted Date: 2 November 2016

Please cite this article as: Parrish WR, Byers BA, Su D, Geesin J, Herzberg U, Wadsworth S, Bendele A, Story B, Intra-articular therapy with recombinant human GDF5 arrests disease progression and stimulates cartilage repair in the rat medial meniscus transection (MMT) model of osteoarthritis, Osteoarthritis and Cartilage (2016), doi: 10.1016/j.joca.2016.11.002.

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

1 Intra-articular therapy with recombinant human GDF5 arrests disease

2 progression and stimulates cartilage repair in the rat medial meniscus

3 transection (MMT) model of osteoarthritis.

1 1 2 2 2 2

5 William R. Parrish *, Benjamin A. Byers , Dongling Su , Jeffrey Geesin , Uri Herzberg , Scott Wadsworth ,

6 Alison Bendele3 and Brooks Story1.

8 1) DePuy Synthes Mitek Sports Medicine, Raynham, MA.

9 2) Advanced Therapeutics and Regenerative Medicine, Johnson & Johnson, Somerville, NJ.

10 3) Bolder BioPATH, Boulder, CO.

11 Corresponding Author

13 William Parrish: wparrish@its.jnj.com

14 Benjamin Byers: bbyers@its.jnj.com

15 Dongling Su: sudongling@yahoo.com

16 Jeffrey Geesin: jeffrey@geesinconsulting.com

17 Uri Herzberg: uherzberg@celgene.com

18 Scott Wadsworth: scottawadsworth@comcast.net

19 Alison Bendele: Alison@bolderbiopath.com

20 Brooks Story: bstory1@its.jnj.com

22 Running Title: GDF5 stimulates cartilage regeneration.

23 Objective: Investigation of osteoarthritis (OA) risk alleles suggests that reduced levels of growth and

24 differentiation factor-5 (GDF5) may be a precipitating factor in OA. We hypothesized that intra-articular

25 recombinant human GDF5 (rhGDF5) supplementation to the OA joint may alter disease progression.

26 Methods: A rat medial meniscus transection (MMT) joint instability OA model was used. Animals

27 received either one intra-articular injection, or two or three bi-weekly intra-articular injections of either

28 30 ^g or 100 mg of rhGDF5 beginning on day 21 post surgery after structural pathology had been

29 established. Nine weeks after MMT surgery, joints were processed for histological analysis following

30 staining with toluidine blue. Control groups received intra-articular vehicle injections, comprising a

31 glycine-buffered trehalose solution. OA changes in the joint were evaluated using histopathological end

32 points that were collected by a pathologist who was blinded to treatment.

33 Results: Intra-articular rhGDF5 supplementation reduced cartilage lesions on the medial tibial plateau in

34 a dose-dependent manner when administered therapeutically to intercept OA disease progression. A

35 single 100 ^g rhGDF5 injection on day 21 slowed disease progression at day 63. A similar effect was

36 achieved with two bi-weekly injections of 30 ^g. Two bi-weekly injections of 100 ^g or three bi-weekly

37 injections of 30 ^g stopped progression of cartilage lesions. Importantly, three biweekly injections of

38 100 ^g rhGDF5 stimulated significant cartilage repair.

39 Conclusions: Intra-articular rhGDF5 supplementation can prevent and even reverse OA disease

40 progression in the rat MMT OA model. Collectively, these results support rhGDF5 supplementation as

41 an intra-articular disease modifying OA therapy.

43 Introduction:

44 Growth and differentiation factor 5 (GDF5) is a bone morphogenic protein (also known as BMP-14 and

45 CDMP-1) and a member of the transforming growth factor-R (TGF-R) superfamily1. GDF5 is a

46 developmental marker for early joint formation in the embryo, and is involved in the maintenance and

47 repair of bone and cartilage in the adult2;3. The importance of GDF5 in synovial joint development is well

48 established2, as is the importance of GDF5 for normal synovial joint function4. Mutations in GDF5 also

49 cause chondrodysplasias that prevent normal skeletal development and are associated with severe

50 articular abnormalities5. Interestingly, single nucleotide polymorphisms (SNPs) in the human GDF5 gene

51 have been identified by genome-wide association studies (GWAS) as significant genetic risk factors

52 associated with osteoarthritis by different investigators studying distinct populations of OA patients6-10.

53 For example, the rs143383 SNP that resides in the 5' untranslated region (UTR) of the GDF5 promoter

54 results in a C-T transition substitution that was originally identified in a GWAS cohort of Asian descent6.

55 The association of this rs143383 SNP with OA was subsequently replicated in a cohort of European

56 descent7. On a mechanistic level, the risk associated "T" allele has been shown to have significantly

57 reduced transcriptional activity compared to the "C" allele in chondrocytes, the resident cells that are

58 responsible for synthesizing and maintaining cartilage11. Therefore, the increased OA susceptibility

59 associated with the rs143383 SNP is believed to be mediated through reduced expression of GDF511-13.

61 The therapeutic potential of GDF5 in connective tissue (fibrous and cartilaginous) healing has been

62 demonstrated in several pre-clinical model systems. For example, stimulatory effects of exogenous

63 GDF5 on matrix synthesis in human articular chondrocytes have been demonstrated in vitro using cells

64 harvested from healthy individuals as well as OA patients14;15. In addition, animal studies have shown

65 that exogenous GDF5 treatment can stimulate rodent tendon healing16;17 and can positively impact the

66 degeneration of intervertebral discs in mouse, rabbit and bovine models18-20. Collectively, this

67 foundation of evidence supports our hypothesis that intra-articular supplementation of rhGDF5 in an

68 osteoarthritic joint may provide a therapeutic means to prevent disease progression and potentially

69 activate anabolic responses leading to cartilage repair. Therefore, we studied the effect of intra-

70 articular GDF5 supplementation in the rat medial meniscal transection (MMT) model of progressive

71 degenerative osteoarthritis. A distinct benefit of the MMT model is that it is a well-established model

72 for studying OA structural changes where the kinetics of disease progression are highly reproducible21;22.

73 We found that therapeutic intra-articular supplementation of rhGDF5 beginning on day 21 post MMT

74 surgery slowed disease progression with a single 100 ^g injection, whereas 2 bi-weekly 100 ^g injections

75 arrested disease progression and 3 bi-weekly 100 ^g injections stimulated cartilage repair. The effects

76 were dose-dependent, as 2 bi-weekly 30 ^g injections were necessary to achieve protective effects

77 similar to those observed for a single 100 ^g injection, and 3 bi-weekly 30 ^g injections were necessary

78 to provide levels of disease arrest similar to those observed with 2 bi-weekly 100 ^g injections. Taken

79 together, these data support the therapeutic potential of intra-articular supplementation with rhGDF5

80 as a disease modifying therapy for osteoarthritis.

82 Materials and Methods:

83 rhGDF5 preparation:

84 Bulk rhGDF5 was obtained from Biopharm GmbH at a concentration of 3.5 mg/ml in 10 mM HCl in a

85 frozen format stored at -80° C. The frozen bulk protein was thawed overnight at 2-8° C prior to

86 formulation. Bulk rhGDF5 was dialyzed against a 5 mM glycine-HCl buffer overnight through a 3000 MW

87 cutoff membrane at 2-8o C. Dialysis resulted in an increase in concentration to 3.8 mg/ml rhGDF5. This

88 rhGDF5 solution was then diluted to a final concentration of 0.5 mg/ml rhGDF5 in a 5 mM glycine-HCL

89 buffer containing 5% trehalose (W:V) at pH 3.0. This solution was then passed through a 0.22 micron

90 filter and further diluted with sterile 5 mM glyci ne-HCL buffer with 5% trehalose at pH 3.0 to obtain

91 working concentrations for injection. One ml of each formulation was dispensed directly into vials for

92 lyophilization to achieve the indicated treatment dosages when the lyophilized cake is resuspended in 1

93 ml of water for injection. Vials from sample lots were tested at random after reconstitution with water

94 for injection to confirm concentration, stability, and potency of rhGDF5 prior to initiation of animal

95 studies using OARSI (Osteoarthritis Research Society International) recommended methodologies

96 described in detail elsewhere23, with minor modifications to enhance the accuracy of the measurement

97 of collagen matrix loss (parameter #1 in the OARSI guidelines). Rather than employing the broad the 0,

98 50 and 100% loss scaling outlined by Gerwin, Bendele, Glasson, and Carlson21, a more sensitive range of

99 loss approach was used in order to provide a more specific and detailed look at the depths of

100 degeneration and/or repair. This modification includes tibial cartilage micrometer measurements for

101 minimal damage (superficial, affecting upper 10% only), mild damage (extending through 11-30% of the

102 cartilage thickness), moderate damage (extending thru 31-60% of the cartilage thickness), marked

103 damage (extending through 61-90% of the cartilage thickness), severe damage (total or near total loss of

104 collagen to tidemark, >90% thickness), and any damage (fibrillation ranging from superficial to full

105 thickness loss). "Any damage" corresponds to the 0% measure in the OARSI Guidelines, and "severe

106 damage" corresponds to the 100% measure. This expanded scaling provides the sensitivity to express

107 the micrometer measurements either individually or in sum combinations to examine potential effects

108 on cartilage sparing or repair in a therapeutic setting. None of the un-injected control animals at day 63

109 displayed lesions more shallow than the range of "moderate damage" (31-60% of the cartilage

110 thickness), therefore the analysis for this parameter was focused on the sum of cartilage degeneration

111 including moderate, marked, and severe damage so that this parameter indicates beneficial treatment

112 effects on matrix loss extending through 31% or greater of the cartilage thickness.

114 Animals:

115 This study was carried out in strict accordance with the standards set forth by the Animal Welfare Act

116 and the recommendations in 'Guide for the Care and Use of Laboratory Animals' (HHS Publication (NIH)

117 No. 85-23), The protocol was approved by the Institutional Animal Care and Use Committee at Bolder

118 BioPATH, Inc (Protocol# BBP-008). All surgery was performed under isoflurane anesthesia, and all efforts

119 were made to minimize suffering with pre and post-operative analgesia. Male Lewis rats weighing

120 approximately 300g (~12 weeks upon arrival) (Charles River, Wilmington, MA) were housed 3 per cage in

121 shoe-box polycarbonate cages with wire tops in sanitary ventilated animal rooms with temperatures

122 ranging between 67-76°F and relative humidity between 30-70%. Automatic timers provided 12 hours

123 of light and 12 hours of dark. Cages contained wood chip bedding and suspended food (Rodent chow,

124 Harlan Teklad, Indianapolis, IN) and access to water bottles ad libitum. Animals were acclimated for 7

125 days prior to being randomized into groups. An attending veterinarian was on site or on call during the

126 live phase of the study.

128 Induction of OA:

129 Animals were anesthetized with isofluorane and the right knee area was shaved and scrubbed in

130 preparation for surgery. A skin incision was made over the medial aspect of the knee and the medial

131 collateral ligament was exposed by blunt dissection, and then transected. The medial meniscus was

132 reflected medially to prevent damage to the articular cartilage, then cut through its full thickness with

133 small surgical scissors to simulate a complete meniscal tear. Skin and subcutis were closed with 4-0

134 VICRYL™ suture. All animals resumed weight bearing immediately post-surgery upon recovery from

135 anesthesia and there was no evidence of excessive post-operative swelling indicative of joint infection.

137 Treatments:

138 For all injected treatment groups (n=15), sterile solutions of rhGDF5 or vehicle were delivered directly to

139 the synovial fluid of the knee via intra-articular injection. Injections were carried out using 500^l insulin

140 syringes with 28 44" needles. In every case, dosing was conducted in a 50 ml volume according to the

141 assigned regimen as detailed in the study schematic shown in Table 1. The study design included low

142 (30 ^g) and high (100 ^g) dose treatments that have been shown to protect cartilage in the MMT model

143 in rats when administered each week for 5 weeks24. In the current study, doses were administered in an

144 exposure de-escalation format as either a single injection, or two or three bi-weekly injections in order

145 to study the durability of rhGDF5 effects (single injection at week 3) as well as the effects of repeat

146 dosing at week 5 and at weeks 5 and 7 in the rat knee joint following OA disease establishment. For

147 each dosing regimen, rhGDF5 vehicle control groups were included. An unun-injected (no treatment or

148 untreated), MMT surgery group terminated at 3 weeks post MMT surgery was also included to

149 determine the extent of OA pathological changes within the knee at the time when interventions were

150 initiated. All treatment groups were bench marked against an unun-injected (no treatment or

151 untreated) group that received MMT surgery and no further intervention. The veterinarian, surgeon,

152 histopathologist, and technical staff were blinded to the therapy assigned to each randomized group.

Histopathology:

Histopathological assessments were carried out in a blinded, unbiased fashion, essentially as described21. Briefly, the knee joints with patellae removed were fixed in 10% NBF for 3 days. Joints were then decalcified in Surgipath decalcifier I (Grayslake, IL) for approximately 1 week. When decalcification was complete, the knees were transected in the frontal plane at the level of the anterior cruciate ligament so the medial and lateral orientation was maintained and both halves were embedded in paraffin with the cut side on the face of the block. Three sections were cut from each knee at approximately 200 mm intervals. Sectioning in this manner provides simultaneous histological coverage of the anterior and posterior aspects of the central tibial plateau and the corresponding articulation of the femoral condyle across the weight bearing region where the most severe osteoarthritic lesions develop in this model21. The sections were stained with toluidine blue and photomicrographs were taken using a CoolSNAP™-Pro microscope camera. The three most severely affected sections from either the anterior or posterior side of the joint were evaluated for cartilage damage and osteophyte formation as described21. The reported values for each parameter represent the average across the 3 sections analyzed per knee joint. Parameters considered were severe + marked + moderate matrix degeneration width, significant tibial cartilage degeneration width, zone 2 lesion depth ratio, medial tibial osteophyte width, and medial collateral ligament thickness.

Statistics:

Statistical analysis of histopathology parameters was done by comparing group values using the Student's two-tailed t-test with significance set at p<0.05. This analysis was performed for each treatment paradigm compared to its corresponding vehicle control group, as well as, for each treatment or vehicle group compared to the un-injected controls (day 21 un-injected, day 63 un-injected).

178 Results:

179 Intra-articular supplementation with rhGDF5 prevented and reversed OA pathology in a dose and

180 administration regimen dependent manner. Photomicrographs taken from the median animal in each

181 group (n=15 per group) are shown in Figure 1 relative to vehicle and un-injected controls. Substantial

182 cartilage degeneration on the medial tibial plateau was apparent by day 21 following MMT surgery

183 (Figure 1 and Table 2), demonstrating the aggressive nature of osteoarthritis-like pathology in this

184 model. If left un-injected, OA-like lesions become progressively larger and deeper by day 63 post MMT,

185 and ultimately result in complete eburnation of the articular cartilage, sclerosis of the subchondral bone

186 with formation of bone marrow lesions, and the duplication and collapse of the calcified cartilage

187 tidemark21 (Figure 1b). The glycine-buffered trehalose vehicle did provide some limited protection to

188 the cartilage in this model compared with the day 63 un-injected group (Figure 1c, Table 2 and

189 Supplemental Tables 1-4). These beneficial effects were more evident for the multiple injection vehicle

190 groups where differences in histological parameters were statistically significant compared with day 63

191 un-injected controls (supplemental tables 1, 2, 3, and 4). However, substantial improvements in the

192 architecture of the cartilage were apparent only for the rhGDF5 treated groups, and were especially

193 evident upon repeat injection (Figures 1d-f and Table 2). Indeed, the three 100 ^g rhGDF5 injection

194 group clearly demonstrated reparative changes in the joint tissues resulting in thickening of the femoral

195 cartilage and significant repair of the medial tibial cartilage (Figure 1g, Table 2 and Supplemental Tables

196 1-4). Quantitative analysis of histopathological changes following intra-articular supplementation with

197 rhGDF5 on osteoarthritis progression in rats subjected to medial meniscus transection is displayed in

198 Table 2. These measures included the width of substantial tibial cartilage lesions involving greater than

199 50% of the normal cartilage thickness (Figure 2 and Supplemental Table 1), the width of tibial cartilage

200 matrix degeneration (Figure 3 and Supplemental Table 2), and the depth of cartilage lesions in zone 2,

201 the central zone, of the medial tibial plateau (Figure 4 and Supplemental Table 3). Substantial anabolic

202 effects on the thickness of the medial collateral ligament (MCL) and on medial osteophyte development

203 were also observed (Figures 5 and 6,Table 1 and Supplemental Table 4). However, different treatment

204 regimens differed greatly in their capacity to stimulate anabolic changes in the MCL and especially in

205 osteophyte growth, (Figure 6 and Table 2). For example, Figure 6 shows that significant osteophyte

206 enhancement was not evident in the 30 ^g rhGDF5 injection groups compared to Day 63 un-injected

207 controls, however each of these groups displayed an increase in the thickness of the repaired MCL

208 (Figure 6). In contrast, multiple injections of 100 ^g rhGDF5 did cause significant enhancement of

209 osteophyte growth compared to Day 63 un-injected controls (Figure 6). A single 100 ^g rhGDF5

210 injection did not significantly affect osteophyte growth, however significantly enhanced repair of the

211 MCL was evident in each 100 ^g rhGDF5 treatment group (Figure 6). These data illuminate a disparity

212 between anabolic effects of rhGDF5 on osteophytes and on the MCL.

214 Interestingly, although there were differences in the anabolic activity of rhGDF5 with respect to

215 osteophytes and the MCL, the effects of 2 injections of 100 ^g rhGDF5 were similar to the effects of 3

216 injections of 30 ^g rhGDF5 with respect to measures of cartilage protection. Similar trends were also

217 apparent in the comparison between 1 injection of 100 ^g rhGDF5 and 2 injections of 30 ^g rhGDF5

218 (Figures 4 and 6 and Table 2). Given the disparity between effects of rhGDF5 on osteophytes versus the

219 MCL, these data suggest that the observed protective and reparative effects on cartilage can be

220 attributed directly to activity of rhGDF5 on cartilage rather than as a secondary effect of increased joint

221 stability that might result from enlarged osteophytes or MCL hyperplasia. Regardless of the absolute

222 mechanism of action, these data demonstrate the potential of rhGDF5 to alter the natural history of

223 osteoarthritis in the rat MMT model.

225 Discussion:

226 Results of this study clearly indicate a dose- and injection regimen-dependent effect of rhGDF5 intra

227 articular supplementation on cartilage preservation and repair in the rat MMT model of osteoarthritis.

228 The ability of rhGDF5 to halt disease progression when therapeutically administered at 100 ^g 2x or 30

229 ^g 3x was profound. However, the ability of rhGDF5 to reverse the OA-like pathology and stimulate

230 cartilage repair by multiple histology measures indicates its potential as a potent disease modifying

231 agent when administered to MMT rats in 3 x 100 ^g doses. It remains possible that bi-weekly

232 administration is not the optimal therapeutic approach. Therefore, should clinical development of

233 rhGDF5 be pursued, dosing studies should investigate dose ranges and other injection schedules in

234 order to further optimize therapeutic efficacy.

236 The ability of the 3 injection regimen at 100 ^g dosing to stimulate cartilage repair should not

237 overshadow the observation that a single injection of 100 ^g rhGDF5 at day 21 provided a significant

238 durable cartilage protection measured at day 63 (Supplemental Tables 1-3). Durable cartilage

239 protection in the presence of the instability lesion (medial meniscal tear) that initiated the

240 osteoarthritis-like disease process clearly indicates that therapeutic administration of rhGDF5 stimulated

241 anabolic responses that altered the natural history of OA-like disease progression in the rat MMT model.

242 This is particularly important since this 100 ^g single injection therapy did not significantly increase

243 osteophyte size compared with the un-injected group (p=0.07). Of equal importance, none of the 30 ^g

244 rhGDF5 injection regimens significantly increased osteophyte size compared with the un-injected group

245 (p>0.18). These data indicate that osteophyte size and cartilage protection are not associated in these

246 therapy groups, and strongly suggests that direct stimulation of reparative responses in the cartilage are

247 responsible for the observed protective effects of rhGDF5 supplementation. Regardless of the absolute

248 mechanism of action, the results of this study clearly demonstrate the disease modifying potential of

249 intra-articular rhGDF5 supplementation and its ability to drive cartilage repair in the rat MMT model.

250 Acknowledgments:

251 Funding for this work was provided by DePuy Synthes Mitek Sports Medicine, Raynham, MA. Animal

252 studies were carried out by Bolder BioPATH, Inc. as a contract research organization.

253 Author contributions:

254 William R. Parrish, Benjamin A. Byers, Dongling Su, Jeffrey Geesin, Uri Herzberg, Scott Wadsworth, and

255 Brooks Story conceived and designed the experiments. Alison Bendele and William R. Parrish analyzed

256 the data. William R. Parrish wrote the first draft of the manuscript. William R. Parrish and Brooks Story

257 developed the structure and arguments for the manuscript. All authors agree with manuscript results

258 and conclusions, and all authors made critical revisions and approved the final version.

259 Competing interests:

260 William R. Parrish, Benjamin A. Byers, Dongling Su, Jeffrey Geesin, Uri Herzberg, Scott Wadsworth, and

261 Brooks Story are employees of Johnson and Johnson. Alison Bendele is an Employee of Bolder BioPATH.

262 References:

264 1. Wang RN, Green J, Wang Z, Deng Y, Qiao M, Peabody M, et al. Bone Morphogenetic Protein (BMP)

signaling in development and human diseases. Genes Dis. 2014;1:87-105.

266 2. Edwards CJ and Francis-West PH. Bone morphogenetic proteins in the development and healing of

synovial joints. Semin.Arthritis Rheum. 2001;31:33-42.

268 3. Luyten FP. Cartilage-derived morphogenetic protein-1. Int.J.Biochem.Cell Biol. 1997;29:1241-1244.

269 4. Daans M, Lories RJ, and Luyten FP. Dynamic activation of bone morphogenetic protein signaling in

collagen-induced arthritis supports their role in joint homeostasis and disease. Arthritis Res.Ther.

2008;10:R115.

272 5. Polinkovsky A, Robin NH, Thomas JT, Irons M, Lynn A, Goodman FR, et al. Mutations in CDMP1

273 cause autosomal dominant brachydactyly type C. Nat.Genet. 1997;17:18-19.

274 6. Miyamoto Y, Mabuchi A, Shi D, Kubo T, Takatori Y, Saito S, et al. A functional polymorphism in the

275 5' UTR of GDF5 is associated with susceptibility to osteoarthritis. Nat.Genet. 2007;39:529-533.

276 7. Chapman K, Takahashi A, Meulenbelt I, Watson C, Rodriguez-Lopez J, Eliq R, et al. A meta-analysis

277 of European and Asian cohorts reveals a global role of a functional SNP in the 5' UTR of GDF5 with

278 osteoarthritis susceptibility. Hum.Mol.Genet. 2008;17:1497-1504.

279 8. Dai J, Shi D, Zhu P, Qin J, Ni H, Xu Y, et al. Association of a single nucleotide polymorphism in

280 growth differentiate factor 5 with congenital dysplasia of the hip: a case-control study. Arthritis

281 Res.Ther. 2008;10:R126.

282 9. Xiao JL, Meng JH, Gan YH, Zhou CY, and Ma XC. Association of GDF5, SMAD3 and RUNX2

283 polymorphisms with temporomandibular joint osteoarthritis in female Han Chinese. J.Oral

284 Rehabil. 2015;42:529-536.

285 10. Williams FM, Popham M, Hart DJ, de Schepper E, Bierma-Zeinstra S, Hofman A, et al. GDF5 single-

286 nucleotide polymorphism rs143383 is associated with lumbar disc degeneration in Northern

287 European women. Arthritis Rheum. 2011;63:708-712.

288 11. Southam L, Rodriguez-Lopez J, Wilkins JM, Pombo-Suarez M, Snelling S, Gomez-Reino JJ, et al. An

289 SNP in the 5'-UTR of GDF5 is associated with osteoarthritis susceptibility in Europeans and with in

290 vivo differences in allelic expression in articular cartilage. Hum.Mol.Genet. 2007;16:2226-2232.

291 12. Reynard LN, Bui C, Canty-Laird EG, Young DA, and Loughlin J. Expression of the osteoarthritis-

292 associated gene GDF5 is modulated epigenetically by DNA methylation. Hum.Mol.Genet.

293 2011;20:3450-3460.

294 13. Egli RJ, Southam L, Wilkins JM, Lorenzen I, Pombo-Suarez M, Gonzales A, et al. Functional analysis

295 of the osteoarthritis susceptibility-associated GDF5 regulatory polymorphism. Arthritis Rheum.

296 2009;60:2055-2064.

297 14. Bobacz K, Gruber R, Soleiman A, Graninger WB, Luyten FP, and Erlacher L. Cartilage-derived

298 morphogenetic protein-1 and -2 are endogenously expressed in healthy and osteoarthritic human

299 articular chondrocytes and stimulate matrix synthesis. Osteoarthritis.Cartilage. 2002;10:394-401.

300 15. Chubinskaya S, Segalite D, Pikovsky D, Hakimiyan AA, and Rueger DC. Effects induced by BMPS in

301 cultures of human articular chondrocytes: comparative studies. Growth Factors 2008;26:275-283.

302 16. Forslund C, Rueger D, and Aspenberg P. A comparative dose-response study of cartilage-derived

303 morphogenetic protein (CDMP)-1, -2 and -3 for tendon healing in rats. J.Orthop.Res. 2003;21:617-

304 621.

305 17. Rickert M, Wang H, Wieloch P, Lorenz H, Steck E, Sabo D, et al. Adenovirus-mediated gene transfer

306 of growth and differentiation factor-5 into tenocytes and the healing rat Achilles tendon.

307 Connect.Tissue Res. 2005;46:175-183.

308 18. Liang H, Ma SY, Feng G, Shen FH, and Joshua LX. Therapeutic effects of adenovirus-mediated

309 growth and differentiation factor-5 in a mice disc degeneration model induced by annulus needle

310 puncture. Spine J. 2010;10:32-41.

311 19. Walsh AJ, Bradford DS, and Lotz JC. In vivo growth factor treatment of degenerated intervertebral

312 discs. Spine (Phila Pa 1976.) 2004;29:156-163.

313 20. Chujo T, An HS, Akeda K, Miyamoto K, Muehleman C, Attawia, et al. Effects of growth

314 differentiation factor-5 on the intervertebral disc--in vitro bovine study and in vivo rabbit disc

315 degeneration model study. Spine (Phila Pa 1976.) 2006;31:2909-2917.

316 21. Gerwin N, Bendele AM, Glasson S, and Carlson CS. The OARSI histopathology initiative -

317 recommendations for histological assessments of osteoarthritis in the rat. Osteoarthritis.Cartilage.

318 2010;18 Suppl 3:S24-S34.

319 22. Bendele AM. Animal models of osteoarthritis. J.Musculoskelet.Neuronal.Interact. 2001;1:363-376.

320 23. Garigapati V, Su D, and Lopez J. Protein formulations for use at elevated temperatures. (US

321 7,678,764). 3-16-2010. Ref Type: Patent

322 24. Byers BA, Parrish WR, Geesin J, Herzberg U, Wadsworth S, Bendele A, et al. Intra-articular

323 supplementation with recombinant human GDF5 arrests disease progression and stimulates

324 cartilage regeneration in the rat medial meniscus transection (MMT) model of osteoarthritis.

325 Osteoarthritis.Cartilage. 2014;22 Suppl S471-2

327 Figure legends.

328 Figure 1. Representative toluidine blue stained micrographs display histopathological changes present in

329 the medial tibio-femoral joint from each group relative to un-injected and vehicle controls. Note the

330 progressive improvement in cartilage architecture between the single injection and the 3 injection

331 regimens. a) no intervention at day 21 post MMT, b) no intervention at day 63 post MMT, c) 3 bi-weekly

332 vehicle injections, d) 3 biweekly 30 mg GDF5 injections, e) 2 bi-weekly 30 mg GDF5 injections, f) single 30

333 mg GDF5 injection, g) 3 bi-weekly 100 mg GDF5 injections, h) 2 bi-weekly 100 mg GDF5 injections, i) single

334 100 mg GDF5 injection. 50X original magnification.

336 Figure 2. Substantial cartilage degeneration width (OARSI histopathology guideline parameter #421).

337 The width of tibial cartilage in which 50% or greater of the original thickness (from surface to tidemark)

338 was seriously compromised was measured microscopically using a calibrated digital micrometer. Note

339 that the medial tibial cartilage in the rat is typically about 2000 microns wide across the load bearing

340 area that was analyzed. A cartilage area is considered to be substantially compromised when 50% or

341 more of chondrocytes are absent or necrotic. Graph data represent the group mean (n = 15) ± SEM.

342 Therapeutic benefit encompasses the range of improvement in histology parameters that falls between

343 the values for day 21 and day 63 un-injected groups. Disease arrest indicates histology parameters that

344 are substantially equivalent to the day 21 un-injected group where the net effect of therapy is no

345 progression of cartilage degeneration. Repair defines the degree of histological improvement that

346 shows decreased cartilage lesion size compared to the day 21 un-injected control group.

348 Figure 3. Cartilage matrix degeneration width (OARSI histopathology guideline parameter #121). The

349 width of tibial cartilage matrix loss were measured microscopically using a calibrated digital micrometer.

350 This measurement only takes cartilage matrix into account regardless of cellularity or viability of the

351 tissue. Graph data represent the group mean (n=15) + SEM. Therapeutic benefit encompasses the range

352 of improvement in histology parameters that falls between the values for day 21 and day 63 un-injected

353 groups. Disease arrest indicates histology parameters that are substantially equivalent to the day 21 un-

354 injected group where the net effect of therapy is no progression of cartilage degeneration. Repair

355 defines the degree of histological improvement that shows decreased cartilage lesion size compared to

356 the day 21 un-injected control group.

358 Figure 4. Zone 2 lesion depth ratio (OARSI histopathology guideline parameter #521). The lesion depth

359 ratio is calculated by dividing the depth of the lesion by the full thickness of the cartilage from projected

360 articular surface to tidemark. These measurements are taken at the midpoint of zone 2 using an ocular

361 micrometer. This parameter can be used to document cartilage thickening and increased matrix

362 deposition. Graph data represent the group mean (n = 15) ± SEM. Therapeutic benefit encompasses the

363 range of improvement in histology parameters that falls between the values for day 21 and day 63 un-

364 injected groups. Disease arrest indicates histology parameters that are substantially equivalent to the

365 day 21 un-injected group where the net effect of therapy is no progression of cartilage degeneration.

366 Repair defines the degree of histological improvement that shows decreased cartilage lesion size

367 compared to the day 21 un-injected control group.

370 Figure 5. Medial collateral ligament (MCL) thickness (OARSI histopathology guideline parameter #921).

371 Thickness of the MCL is a sensitive measure of soft tissue anabolic activity during the tissue repair

372 process. This measurement is made by digital micrometer through a non-tangential area of repair in the

373 tissue section. Graph data represent the group mean (n = 15) ± SEM. Normal growth encompasses the

374 increase in MCL thickness that falls between the day 21 and day 63 un-injected group values. Anabolic

375 stimulation defines the degree of increased MCL thickness that exceeds the day 63 un-injected control

376 group. Note the MCL thickness in a normal non-surgical condition was 272 +/- 8 microns (data not

377 shown), indicating the normal tissue repair response to surgical injury.

379 Figure 6. Differential anabolic effects of rhGDF5 on osteophytes and the MCL. Values for anabolic

380 effects of rhGDF5 on osteophytes (OARSI histopathology guideline parameter #621) and the MCL (OARSI

381 histopathology guideline parameter #921) were transformed to % of Day 63 un-injected control in order

382 to present values on the same axis. Data indicate that the magnitude of increased osteophyte size and

383 enhanced MCL repair are not directly correlated. N=15 animals per group; * indicates p< 0.05 by 2384 tailed Student's Ttest compared with Day 63 un-injected control group.

Table 1. Group assignments and treatment schedule.

group # treatment D=0 D=7 D=14 D=21 D=28 D=35 D=42 D=49 D=56 D=63

1 untreated Sx sac

2 untreated Sx ANX ANX ANX sac

3 V 3X Sx X X X sac

4 30 3X Sx X X X sac

5 100 3X Sx X X X sac

6 100 2X Sx X X ANX sac

7 30 2X Sx X X ANX sac

8 V2X Sx X X ANX sac

9 100 1X Sx X ANX ANX sac

10 V 1X Sx X ANX ANX sac

11 30 1X Sx X ANX ANX sac

V= vehicle treated; Sx= surgery; Anx= anesthethesia without intervention; X= injection as indicated per group assignment; sac= termination

Animals were randomly assigned to groups, and groups were then randomly assigned to treatment regimens. Intraarticular injections were administered under isofluorane anesthesia, and all groups were subjected to anesthesia at each intervention point regardless of their treatment group in order to maintain congruency in handling across groups. Veterinary and histopathology staff was blinded with respect to treatment.

DePuy Synthes Mitek Sports Medicine. Confidential.

Table 2: Summary of histopathological measurements and observations

treatment group Matrix degeneration width (severe + marked+ moderate) Substantial tibial cartilage degeneration width Zone 2 lesion depth ratio medial tibial osteophyte size MCL thickness

Day 63 No Treatment 513.3(59.1) 788.9(38.2) 0.42 (.02) 636.9(50.4) 611.6(35.9)

1 injection Vehicle 453.3(111.5) 773.3(73.1) 0.43 (0.05) 607.3(44.6) 604.4(25.7)

30 ug 364.4(104.1) 674.4(99.7) 0.34(0.07) 631.3(49.3) 732.4(39.1)

100 ug 173.3(89.5) 623.3(67.8) 0.29(0.06) 722.2(72.5) 922.7(71.0)

2 injection vehicle 360.3(89.4) 695.6(62.3) 0.36(0.06) 573.6(61.7) 616.0(17.1)

30 ug 223.3(96.2) 573.3(102.4) 0.31(0.04) 658.2(56.1) 799.1(28.6)

100 ug 149.4(88.1) 501.1(86.8) 0.25(0.03) 872.2(90.7) 989.3(75.4)

3 injection vehicle 437.7(118.6) 681.1(85.8) 0.37 (0.05) 579.8(70.6) 644.4(27.4)

30 ug 212.2(123.2) 498.9(133.3) 0.27(0.07) 696.7(69.8) 881.8(45.0)

100 ug 45.2(27.0) 389.3(89.4) 0.18(0.04) 881.0(111.6) 1213.3(93.2)

Day 21 No Treatment 120(61.7) 562.2(119.8) 0.33 (0.05) 450.2(27.1) 521.8(18.2)

Values represent the group mean (n= 15) and the 95% confidence interval in parentheses. Note that Matrix degeneration width corresponds with OARSI scoring parameter #1, Substantial tibial cartilage degeneration width corresponds with parameter #4, Zone 2 lesion depth ratio corresponds with parameter #5, medial tibial osteophyte size corresponds with parameter #6, and Medial Collateral Ligament (MCL) thickness corresponds with parameter #9.21

DePuy Synthes Mitek Sports Medicine. Confidential.

Supplemental Table 1: Student's 2-tailed Ttest probability matrix for substantial cartilage degeneration width

treatment group day 21 day 63 V 3x 30 3x 100 3x V 2x 30 2x 100 2x Vlx 30 lx

Day 63 No Treatment 0.001

3 injection Vehicle 30 ug 100 ug 0.125 0.494 0.033 0.033 0.000 0.000 0.032 0.000 0.198

2 injection vehicle 0.063 0.018 0.791 0.014 0.000

30 ug 0.951 0.000 0.082 0.510 0.024 0.032

100 ug 0.425 0.000 0.007 0.978 0.090 0.001 0.424

1 injection vehicle 0.006 0.714 0.120 0.001 0.000 0.124 0.002 0.000

30 ug 0.169 0.045 0.922 0.048 0.000 0.727 0.123 0.016 0.128

100 ug 0.392 0.000 0.309 0.114 0.000 0.135 0.301 0.038 0.006 0.413

Shaded boxes indicate pairwise comparisons with statistical differences.

DePuy Synthes Mitek Sports Medicine. Confidential.

Supplemental Table2: Student's 2-tailed Ttest probability matrix for cartilage matrix degeneration width

day 21 day 63 V 3x 30 3x 100 3x V 2x 30 2x 100 2x Vlx 30 lx

Day 63 No Treatment 0.000

3 injection Vehicle 30 ug 100 ug 0.000 0.200 0.044 0.273 0.000 0.000 0.015 0.000 0.018

2 injection vehicle 0.000 0.011 0.338 0.081 0.000

30 ug 0.087 0.000 0.010 0.890 0.002 0.060

100 ug 0.596 0.000 0.001 0.423 0.041 0.004 0.276

1 injection vehicle 0.000 0.359 0.853 0.008 0.000 0.233 0.005 0.000

30 ug 0.000 0.021 0.370 0.075 0.000 0.955 0.061 0.004 0.263

100 ug 0.344 0.000 0.002 0.620 0.015 0.010 0.462 0.712 0.001 0.011

Shaded boxes indicate pair-wise comparisons with significant differences.

DePuy Synthes Mitek Sports Medicine. Confidential.

Supplemental Table 3: Student's 2-talled Ttest probability matrix for zone 2 lesion depth ratio

treatment group day 21 day 63 V 3x 30 3x 100 3x V 2x 30 2x 100 2x Vlx

Day 63 No Treatment 0.004

3 injection Vehicle 0.291 0.052

30 ug 0.232 0.001 0.038

100 ug 0.000 0.000 0.000 0.048

2 injection vehicle 0.439 0.054 0.831 0.071 0.000

30 ug 0.661 0.000 0.088 0.326 0.000 0.192

100 ug 0.036 0.000 0.001 0.724 0.007 0.005 0.034

1 injection vehicle 0.011 0.727 0.081 0.001 0.000 0.073 0.001 0.000

30 ug 0.709 0.041 0.588 0.158 0.000 0.743 0.428 0.028 0.044

100 ug 0.358 0.000 0.054 0.708 0.007 0.107 0.518 0.353 0.001

Shaded boxes indicate pair-wise comparisons with significant differences.

DePuy Synthes Mitek Sports Medicine. Confidential.

Supplemental Table 4: Student's 2-talled Ttest probability matrix for MCL thickness

treatment group day 21 day 63 V 3x 30 3x 100 3x V 2x 30 2x 100 2x Vlx

Day 63 No Treatment 0.000

3 injection Vehicle 0.000 0.165

30 ug 0.000 0.000 0.000

100 ug 0.000 0.000 0.000 0.000

2 injection vehicle 0.000 0.828 0.096 0.000 0.000

30 ug 0.000 0.000 0.000 0.005 0.000 0.000

100 ug 0.000 0.000 0.000 0.023 0.001 0.000 0.000

1 injection vehicle 0.000 0.755 0.046 0.000 0.000 0.470 0.000 0.000

30 ug 0.000 0.000 0.001 0.000 0.000 0.000 0.012 0.000 0.000

100 ug 0.000 0.000 0.000 0.348 0.000 0.000 0.004 0.218 0.000

Shaded boxes indicate pair-wise comparisons with significant differences.

DePuy Synthes Mitek Sports Medicine. Confidential.

Figure 1: Histopathological changes in knee joints following MMT

Figure 2. Substantial cartilage degeneration width

vehicle ■ 30 ug GDF5 ■ 100 ug GDF5

Therapeutic benefit

Disease arrest

Repair

1 injection 2 injection 3 injection untreated untreated (21) (21 and 35) (21, 35 and day 21 day 63

DePuy Synthes Mitek Sports Medicine. Confidential. Figure 3. Cartilage matrix degeneration width

vehicle ■ 30 ug GDF5 ■ 100 ug GDF5

Therapeutic benefit

Disease arrest

Repair

1 injection

2 injection 3 injection day 21 day 63

untreated untreated

Figure 4. Zone 2 lesion depth ratio

o 'ä 0.

■a 0.

.2 0. "O

vehicle 30 ug ■ 100 ug

Therapeutic * benefit

Disease arrest

» Repair

1 injection 2 injection 3 injection day 21 day 63

untreated untreated

Figure 5. Anabolic effect of rhGDF5 on MCL thickness

I vehicle 30 ug ■ 100 ug

с 800

"E 600

Anabolic stimulation

Normal

1 injection 2 injection 3 injection day 21 day 63

untreated untreated

Figure 6. Differential anabolic effects of rhGDF5 on

osteophytes and MCL

O ■u

tu "g 160% E *J

§ 140%

£ 120%

1 injection 2 injection 3 injection

Osteophyte

vehicle

Osteophyte MCL 30 ug

- m 1 r

Osteophyte MCL 100 ug

Table 1. Group assignments and treatment schedule.

group # treatment D=0 D=7 D=14 D=21 D=28 D=35 D=42 D=49 D=56 D=63

1 untreated Sx sac

2 untreated Sx ANX ANX ANX sac

3 V 3X Sx X X X sac

4 30 3X Sx X X X sac

5 100 3X Sx X X X sac

6 100 2X Sx X X ANX sac

7 30 2X Sx X X ANX sac

8 V 2X Sx X X ANX sac

9 100 1X Sx X ANX ANX sac

10 V 1X Sx X ANX ANX sac

11 30 1X Sx X ANX ANX sac

V= vehicle treated; Sx= surgery; Anx= anesthethesia without intervention; X= injection as indicated per group assignment; sac= termination

Animals were randomly assigned to groups, and groups were then randomly assigned to treatment regimens. Intraarticular injections were administered under isofluorane anesthesia, and all groups were subjected to anesthesia at each intervention point regardless of their treatment group in order to maintain congruency in handling across groups. Veterinary and histopathology staff was blinded with respect to treatment.

Table 2: Summary of histopathological measurements and observations

treatment group Matrix degeneration width (severe + marked+ moderate) Substantial tibial cartilage degeneration width Zone 2 lesion depth ratio medial tibial osteophyte size MCL thickness

Day 63 No Treatment 513.3 (59.1) 788.9 (38.2) 0.42 (.02) 636.9 (50.4) 611.6 (35.9)

1 injection Vehicle 453.3 (111.5) 773.3 (73.1) 0.43 (0.05) Y 607.3 (44.6) 604.4 (25.7)

30 ug 364.4 (104.1) 674.4 (99.7) 0.34 (0.07) 631.3 (49.3) 732.4 (39.1)

100 ug 173.3 (89.5) 623.3 (67.8) 0.29 (0.06) 722.2 (72.5) 922.7 (71.0)

2 injection vehicle 360.3 (89.4) 695.6 (62.3) 0.36 (0.06) 573.6 (61.7) 616.0 (17.1)

30 ug 223.3 (96.2) 573.3 (102.4) 0.31 (0.04) 658.2 (56.1) 799.1 (28.6)

100 ug 149.4 (88.1) 501.1 (86.8) 0.25 (0.03) 872.2 (90.7) 989.3 (75.4)

3 injection vehicle 437.7 (118.6) 681.1 (85.8) Y 0.37 (0.05) 579.8 (70.6) 644.4 (27.4)

30 ug 212.2 (123.2) 498.9 (133.3) 0.27 (0.07) 696.7 (69.8) 881.8 (45.0)

100 ug 45.2 (27.0) 389.3 (89.4) 0.18 (0.04) 881.0 (111.6) 1213.3 (93.2)

Day 21 No Treatment 120 (61.7) 562.2 (119.8) 0.33 (0.05) 450.2 (27.1) 521.8 (18.2)

Values represent the group mean (n= 15) and the 95% confidence interval in parentheses. Note that Matrix degeneration width corresponds with OARSI scoring parameter #1, Substantial tibial cartilage degeneration width corresponds with parameter #4, Zone 2

lesion depth ratio corresponds with parameter #5, medial tibial osteophyte size corresponds with parameter #6, and Medial Collateral

Ligament (MCL) thickness corresponds with parameter #9.