WISP1/CCN4 Function: Osteoarthritis and Cutaneous Wound healing Previous studies showed that Wisp1/CCN4 mRNA and Wisp1 protein are increased in both induced (mouse) and naturally occurring (human) OA (Blom et al Arthritis Rheum 60:501, 2009). Moreover, adenoviral overexpression of Wisp1 in nave mouse knee joints resulted in early OA-like cartilage lesions. Based on these findings we wanted to determine if Wisp1 is involved in the pathology that occurs in the osteoarthritic environment, either with aging or in experimentally induced osteoarthritis using Wisp1 KO and Wt mice. Wt and Wisp1 KO mice were either aged or experimental osteoarthritis was induced with 1) intraarticular collagenase injection, 2) destabilization of the medial meniscus or 3) anterior cruciate ligament transection. To our surprise, aging of Wisp1 KO mice did not result in decreased spontaneous cartilage degeneration compared with their Wt controls. However, we observed significantly decreased cartilage degeneration in Wisp1 KO mice after induction in all three induced experimental osteoarthritis models. Considering our previous work showing Wisp1 was needed for bone formation we next examined the subchondral bones of affected mice. While the degree of bone formed by osteophytes was comparable between Wt and Wisp1 KO mice, increased cortical thickness and reduced trabecular spacing were observed only in Wisp1 KO mice. In summary, these data point to the concept that Wisp1 plays an important role in the development of post-traumatic experimental osteoarthritis, and further, that a reduction in WISP1 could be could be considered as a novel therapy to ameliorate human OA. Investigations by several other labs suggested that members of the CCN family could play roles in cutaneous wound healing. suggested that members of the CCN family could play roles in cutaneous wound healing. When full-thickness skin defects were made in mice we found Wisp1 up-regulation during wound healing as early as 1 day after surgery, suggesting a role in inflammation and subsequent dermal migration and proliferation. To determine how WISP1 could regulate wound healing, we used Wisp1-KO mice and showed they had delayed wound closure accompanied by reduced expression of Col1a1 and Fn mRNA. Boyden chamber assays using Wisp1-deficient dermal fibroblasts showed they have reduced migration and proliferation compared to Wt counterparts. To confirm WISP1 has a role in proliferation and migration of dermal cells, siRNA knockdown and transduction of WISP1 adenoviral transduction were used and resulted in reduced or enhanced migration of human adult dermal fibroblast (hADF) cells, respectively. The induced migration of the dermal fibroblasts by WISP1 appears to work via a5b1 integrin receptors that further modulate downstream ERK/JNK signaling. The regulation of WISP1 by TNF- prompted us to look further at their potential relationship. Treatment of hADFs with WISP1 and TNF- alone or together showed WISP1 counteracted the effects of TNF-a on matrix mRNA expression and subsequent downstream NF-B/p-65 signaling. Taken together we showed WISP1 regulates dermal fibroblast cell migration, proliferation and modulation of TNF-a induction, all of which could regulate wound healing. SLRPs and osteoclastogenesis The small leucine-rich proteoglycan (SLRP) family is composed of 17 members sub-divided into classes (I-V) based on their amino acid sequence and genomic organization. All members of the SLRP family (excluding asporin) have extensive post-translational glycosylation on a relatively small protein core backbone composed of repeat structures rich in leucine. For years, evidence has been mounting about the importance of SLRPs in skeletal function. We have focused on the SRLP, biglycan (Bgn), because of its high level of expression in bones and teeth. Taken together, our work highlights the fact that Bgn is not needed for bone development but, rather, appears to play a role in skeletal aging. This has been demonstrated using mice unable to make bgn that are found to acquire early onset osteoporosis (osteopenia/low bone mass), osteoarthritis and ectopic bone in their tendons. Our first approach to this study was to carry out a thorough evaluation of the bones of Wt and DKO mice with aging. Previous work from our lab showed that they acquire osteoarthritis (OA) (Ameye et al, FASEB J 16:673, 2002) and ectopic bones in their tendons (Bi et al, Nature Medicine 13:1219, 2007); however, a comprehensive bone analysis had never been performed. Visual inspection of developing mice stained with alcian blue for cartilage and alizarin red for bone did not reveal developmental effects in the skeleton. However, our work so far using Dual Energy X-Ray Absorptiometry (DEXA) and micro-CT analysis confirmed our suspicions from earlier studies that the bones from the DKO have progressive and severe low bone mass in both cortical and trabecular bone, which worsens with aging in both male and female animals. Even though Bgn and Fmod are made by bone-forming osteoblasts, our dynamic labeling experiments indicate that the DKO mice have slightly higher bone formation rates compared to Wt controls. As predicted from our earlier studies on the TMJ and the periodontal ligament (PDL), we found the levels of TRAP-positive cells in the long bones of the DKO were significantly higher than in Wt long bones as early as 5 weeks of age. To understand the cellular basis for the diminished bone mass found in the DKO mice, we examined the differentiation capacity of BMSCs and calvarial-derived osteogenic cells because they both make copious amounts of Bgn and Fmod. We speculated that changes in the osteoprogenitors in the mutants and the microenviroment that they create could influence osteoclastogenesis. Our experiments showed that the number of CFU-Fs (colony forming unit-fibroblasts, an approximation of the number of skeletal stem/progenitor cells in bone marrow) is higher in the DKO mice compared to the WT, suggesting stem/progenitor number is not reduced. Furthermore, the differentiation of osteogenic precursors (both BMSCs and calvarial cells) shows a trend towards increased osteogenesis (based on RT-PCR of osteogenic genes), an observation that corresponds with the fact that bone formation rates were not lower in this mouse model. Considering the location of Bgn and Fmod in the matrices surrounding the osteoblasts, we proposed that the modulation of bone resorption by Bgn and Fmod may work by regulating the microenviroment and localization/stability of Osteoprotegerin (Opg), Rank decoy, or other key players in the osteoclastogenesis system. Using multiple approaches, we found that both Bgn and Fmod directly bind to TNF as well as Rankl in a dose-dependent manner and that despite expressing higher levels of both TNF and Rankl, Bgn/Fmod KO-derived osteoblasts cannot retain these cytokines in the vicinity of the cells, which leads to elevated TNF and Rankl signaling and enhanced osteoclastogenesis. Furthermore, adding either Bgn or Fmod to osteoclast precursor cultures significantly attenuated the cells ability to form TRAP positive, multinucleated giant cells. In summary, our data indicate that Bgn and Fmod expressed by osteogenic cells are novel coupling ECM components that control bone mass through sequestration of TNF and/or Rankl, thereby adjusting their bioavailability to regulate osteoclastogenesis. In summary, the CCN member WISP1 was shown to have numerous functions in the musculoskeletal system including an aggravating role cartilage loss in cases of induced osteoarthritis but a positive role in accelerating cutaneous wound healing. The SLRPs Bgn and Fmod on the other hand were shown to be positive regulators of bone mass by inhibiting the formation and activity of bone resorbing osteoclasts.
Showing the most recent 10 out of 48 publications
