Skip to content

Advertisement

  • Oral presentation
  • Open Access

Regulation of collagenase-3 (MMP-13) gene and protein expression in human cartilage: when simple things get complicated

  • 1,
  • 1 and
  • 1
Arthritis Res Ther20035 (Suppl 3) :47

https://doi.org/10.1186/ar848

  • Published:

Keywords

  • Collagenase
  • Hepatocyte Growth Factor
  • Transform Growth Factor Beta
  • Alternative Transcription Start Site
  • Collagenase Gene

In cartilage, collagen type II is of particular importance as its breakdown results in the irreversible loss of structural integrity of the tissue. Within the protease, the collagenases have a major involvement in this collagen network degradation. Evidence demonstrates that collagenase-3 is the major enzyme accounting for collagen degradation in osteoarthritic (OA) cartilage.

Collagenase-3 has a greater effect (five to 10 times) on type II collagen than collagenase-1, and, in OA, is localized predominantly in the lower intermediate and deep layers of the cartilage, where type II collagen fibers are of the largest size and chondrocytes possess the most efficient capacity to reconstitute the extracellular matrix. Collagenase is upregulated in OA cartilage and is suggested to be implicated in cartilage remodeling in pathological conditions. Various factors induce its transcription, including proinflammatory cytokines and growth factors such as IL-1β, IL-17, tumor necrosis factor alpha, transforming growth factor beta and hepatocyte growth factor. Interestingly, we recently reported that transforming growth factor beta, but not IL-1β, treatment of normal cartilage mimicked the in situ collagenase-3 distribution in OA cartilage.

The proximal promoter sequence contains a TATA box, as well as AP-1, Ets/PEA-3, and OSE-2 binding sites. The AP-1 site is essential for both basal and proinflammatory cytokine inducible transcription, and the PEA-3 site exerts a cooperative effect. We also demonstrated that some AP-1 proteins play a different role in terms of collagenase-3 production according to the stimulator. Indeed, IL-17-induced collagenase-3 resulted in FosB activation, whereas IL-1β stimulated c-Fos, which may explain the different capacities of these cytokines in producing this enzyme. Our data also suggest that JunB protein plays a rate-limiting step in cytokine-induced collagenase-3 production in OA chondrocytes. Moreover, we recently identified a novel protein-binding site on the collagenase-3 promoter that appears to be directly implicated in the repression of its basal transcription. This site was designated AGRE for AG-rich element. This site was not found in other human metalloprotease genes or in the mouse collagenase.

Contrary to the other human collagenase genes that are transcribed into one mature mRNA, human cells expressed collagenase-3 transcripts of 3.0, 2.5 and 2.2/2.0 kb as demonstrated by northern blot. We recently identified five different collagenase-3 RNA species in humans; each could be translated in a cellular environment, indicating that they could be synthesized in response to specific cellular events. For two of the RNA species, the enzyme synthesized would differ from the original collagenase-3 and will have potentially different function/activity. Moreover, one of the transcripts appears to be an alternative transcription start site. Start sites are known to regulate gene expression by affecting the level of transcription initiation, the translation efficiency of the mRNA produced, and the generation of protein isoforms differing at their amino termini, or may respond differently to the cellular environment.

At present, a therapeutic intervention based on the inhibition of metalloproteases is under intensive investigation, and collagenase-3 appears to be an attractive target for the development of disease-modifying OA drugs. However, the human collagenase-3 is subjected to different levels of regulation and constitutes a more complex system than originally thought. It would be interesting to verify whether compounds that block only one pathway, such as the synthesis/activity of the original collagenase-3, are sufficient to block all the effects of this enzyme, or whether all the enzyme transcripts should rather be targeted in order to achieve maximum therapeutic efficacy.

Authors’ Affiliations

(1)
Osteoarthritis Research Centre, University of Montreal Hospital Centre, Notre-Dame Hospital, Montreal, Quebec, Canada

Copyright

© The Author(s) 2003

Advertisement