Cell-based tissue engineering strategies are currently in clinical use and continue to be developed at a rapid pace for the repair of cartilage defects. Regardless of the repair methodology, newly created joint cartilage will still be susceptible to the abnormal inflammatory and mechanical environments that underlie osteoarthritic disease, potentially compromising its integration, function, and longevity. Chondroprotection measures must therefore be designed for the chondrocytes of regenerated cartilage, which are in close proximity to matrix proteases, bioactive extracellular matrix fragments, and inflammatory mediators responsible for degenerative OA. Previous reports have indicated that many of the biologic changes in articular chondrocytes during the progression of OA are similar to those in the cartilage growth plate during developmental endochondral ossification. Specifically, articular chondrocytes reinitiate a sequence of phenotypic changes that begin with hypertrophy and progress towards mineralization and apoptosis9, 10, 22, 23
. Increased production of catabolic proteases, including MMP-1, MMP-3, MMP-9, and MMP-13, has been observed both in the growth plate and OA cartilage, resulting in matrix degradation in both processes12, 24–28
Because the progression of the degenerative cascade is thought to involve the loss of inhibitory control over terminal differentiation genes, we hypothesized that an anti-hypertrophic therapy may protect healthy chondrocytes from arthritic disease. PTHrP is an important signaling factor in the regulation of endochondral ossification in the epiphyseal growth plate of long bones. Within cells, PTHrP is processed by members of the family of prohormone convertases to at least three fragments with different biologic functions29, 30
. Several direct effects of these fragments make it an attractive option for chondroprotection therapy. Suppression of hypertrophy and terminal differentiation is mediated through parathyroid hormone receptor 1 (PTHR1). Activation of PTHR1 leads to subsequent activation of the PKA signaling pathway, increasing chondrocyte proliferation by decreasing p57 and Runx2 levels and increasing Sox9 phosphorylation31
. Because the N-terminal regions of PTHrP and PTH are similar and share the same receptor (PTHR1), previous reports have shown the ability of these peptides to suppress hypertrophic changes and terminal differentiation of articular chondrocytes induced by T3 and 5-azacytidine32, 33
. Additionally, PTHrP has a well-documented anti-apoptotic effect through its translocation to the nucleus via a mid-region bipartite nuclear targeting sequence (NTS)17, 34–36
. Thus, PTHrP plays an important role in maintaining the survival and chondrogenesis of proliferating chondrocytes and suppressing terminal differentiation in the growth plate.
In the present study, the effects of PTHrP in suppressing early arthritic-like changes were ascertained in a mechanical CTS model. Because the half-life of the N-terminal fragment is only 6 to 8 min37
, sustained delivery of PTHrP will eventually be needed to inhibit the effects of a chronic disease like OA to achieve optimal therapeutic results. Such a mechanism of sustained PTHrP delivery was tested by non-viral gene transfer of hPTHrP cDNA into the articular chondrocytes themselves. This study contributes several novel approaches and results. First, high magnitude CTS model induced an arthritic phenotype with a similar pattern of changes as OA chondrocytes, including increased gene expression of the collagenases and aggrecanases, as well as increased production of NO and PGE2
. Additionally, high magnitude CTS was discovered to increase gene expression of Col10a1, a marker of chondrocyte hypertrophy, and this phenomenon may be mediated by Runx2 due to their concurrent increase. Endogenous bPTHrP gene expression increased with CTS, consistent with increased PTHrP secretion seen in human osteoarthritic cartilage38, 39
and thought to be a self-repair response to damaged cartilage40
. Interestingly, gene expression of PTH1R, a marker of pre-hypertrophy, decreased initially for the first 12 hours with CTS but steadily increased for the next 36 hours (data not shown). This bimodal regulation may arise from CTS overriding the autocrine feedback down-regulation of receptor expression in cells producing PTHrP. However, cellular changes that occur during recovery from trypsinization may also modulate this response. Gene expression of ALP, a recognized marker for terminal differentiation, did not change significantly with 48 hours of CTS. It may be possible that only early hypertrophic genes were upregulated within the first 48 hours of CTS, while late hypertrophic genes would have been upregulated with a longer duration of CTS. Second, sustained PTHrP delivery to chondrocytes was achieved by means of FuGENE6-mediated transfection of hPTHrP constructs. As expected, chondrocytes expressing exogenous hPTHrP showed higher secreted levels of total PTHrP (bovine and human) than controls. Chondrocytes expressing isoform 1-141 generally secreted more total PTHrP than those expressing isoform 1-173, which is consistent with previous data showing that serial truncation of the C-terminal end increases the amount of secreted PTHrP21
Third, overexpression of both hPTHrP isoforms inhibited CTS-induced and basal Col10a1 expression in articular chondrocytes. Although CTS significantly increased Runx2 mRNA levels, this change was so slight that whether inhibition of Col10a1 expression by hPTHrP was via Runx2 could not be ascertained. hPTHrP had no significant inhibitory effect on CTS-induced MMP-13 gene expression. MMP-13 is expressed by hypertrophic chondrocytes to degrade the cartilage matrix, to facilitate vascular invasion and matrix mineralization by osteoblasts. Because MMP-13 is expressed in only the most terminally differentiated hypertrophic chondrocytes41
, the immediate MMP-13 up-regulation observed in this CTS model is most likely attributed to mechanotransduction rather than phenotype modulation. hPTHrP overexpression was not able to immediately inhibit the catabolic changes along with the hypertrophic-like changes, suggesting that the two pathways are regulated separately.
Lastly, CTS increased NO and PGE2
production, which both act as strong catabolic signals in cartilage by altering chondrocyte function and enhancing chondrocyte apoptotic potential42
. hPTHrP isoform 1-141 potentiated the production of these mediators with CTS by upregulating transcriptional levels of iNOS and COX-2. Because these effects were not observed with isoform 1-171, the underlying mechanism may be specific to the amount of secreted PTHrP or to the 140–141 region that differs between the two isoforms. Although PTHrP is known to stimulate NO release from endothelial cells for local regulation of vascular tone43
, a possible mechanism for PTHrP-mediated NO release in chondrocytes has not been elucidated. Additionally, studies have found that PGE2
increases PTHrP production in chondrocytes44, 45
, but the reverse has not been reported. Increased production of pro-inflammatory intermediaries is expected to promote apoptosis through regulation of Bcl-2 and Bax gene expression in chondrocytes. Bax forms heterodimers with Bcl-2, and when overexpressed, counters the anti-apoptotic effect of Bcl-2, causing accelerated cell death. Therefore, it is the ratio of Bcl-2 to Bax that determines the apoptotic fate of a cell. Our results show that Bcl-2 and Bax gene up-regulation by CTS follows a similar temporal pattern as hypertrophic chondrocytes in the growth plate16, 46
. Anti-apoptotic Bcl-2 gene expression was immediately increased, possibly in a rescue attempt for survival, while pro-apoptotic Bax gene expression was not increased until 12 to 24 hours of CTS, when the ratio of the expression level of Bcl-2 and Bax gradually shifts in favor of Bax. Intracellular hPTHrP expression was expected to effect a pro-survival advantage to chondrocytes due to translocation of the NTS to the nucleus, as shown in previous work in the literature17, 34–36
. However, hPTHrP overexpression had no significant effect on Bcl-2 and Bax gene expression. This result may be attributed to the short testing period. The intact viability of the majority of cells after 48 hours of CTS indicates that the only the initial stages of the apoptotic cascade were activated. Longer follow-up times may be needed to appreciate any apoptosis and cell death cause by CTS and thus, any anti-apoptotic effects of hPTHrP overexpression. Interestingly, Bcl-xL gene expression was upregulated with CTS and augmented by expression of hPTHrP 1-141 at 48 hours (data not shown).
Several studies have reported the use of inflammation-responsive promoters to regulate OA gene therapy47–49
. Regulation of hPTHrP expression may be necessary to selectively activate targeted pathways because high concentrations of PTHrP can facilitate activation of other heterotrimeric G proteins, including the Gq
family. The effect of Gq
activation is opposite from that of Gs
activation in that it mildly accelerates chondrocyte differentiation50
. The purpose of these opposing actions may be to regulate the proliferation of chondrocytes based on their location, thus creating a functional gradient of PTHrP across the growth plate31
. PTHrP overexpression may expose chondrocytes to concentrations that are too high, simultaneously activating Gs
pathways. Therefore, we are currently testing the use of an inflammation-responsive, self-limiting promoter consisting of NF-κB repeat sequences to optimize PTHrP therapy.
This study demonstrates that PTHrP overexpression inhibits CTS-induced hypertrophic-like changes of articular chondrocytes. Several gene therapy approaches are currently being investigated for the repair of cartilage defects and the inhibition of OA progression. However, most of these approaches (i.e., MMP inhibitors) target only one of several players in the disease pathogenesis, and none have attempted to target the early hypertrophic-like changes seen in OA. Ultimately, a combination of these approaches may be needed for effective chondroprotection of regenerated cartilage. Because CTS only initiated early stage arthritic-like changes, the long-term protective effects of PTHrP may be even greater than indicated in this study. Future work will examine longer-term treatment regimens to better assess the effects of PTHrP overexpression on overall matrix metabolism and chondrocyte survival under mechanical loading conditions.