Volume 7 Supplement 1
Regulating apoptosis in fibroblasts
© BioMed Central Ltd 2005
Received: 11 January 2005
Published: 17 February 2005
Apoptosis constitutes a highly selective way of eliminating aged and injured cells and is a key mechanism for the balanced growth and regeneration of tissues. In addition to genotoxic stress and the withdrawal of growth factors, apoptosis can be induced by death receptors. Such receptors trigger apoptosis through an approximately 90 amino acid death domain. Prominent and best-characterized members of the death domain receptor family are Fas (CD-95/Apo-1) and the p55 TNF receptor (TNFRI). Alterations in receptor-mediated apoptosis may result in changes of tissue homeostasis and have been found in a variety of malignancies as well as in the inflamed and hyperplastic synovium in rheumatoid arthritis. Accumulating evidence suggests that the activation of rheumatoid arthritis synovial fibroblasts (RASF) is associated with alterations in apoptosis, especially at sites of invasion into cartilage and bone. Specifically, it has been demonstrated that RASF are less susceptible to Fas-induced cell death than osteoarthritis synovial fibroblasts but show higher expression levels of Fas.
Recent studies have shown that both cytokine-dependent and cytokine-independent mechanisms contribute to the resistance of RASF against Fas-induced apoptosis. Thus, tumor necrosis factor alpha (TNF-α) – a major inflammatory cytokine in the rheumatoid synovium – fails to induce apoptosis in RASF, but reduces the susceptibility of these cells to Fas-mediated cell death through the induction of the transcription factor NF-κB. Providing a link between altered apoptosis and cartilage destruction, we have shown that overexpression of TIMP-3 through gene transfer not only reduces the invasiveness of RASF but also modulates the apoptosis-inhibiting effects of TNF-α. RASF overexpressing TIMP-3 are sensitized strongly to Fas/CD95-mediated cell death by TNF-α, and gene transfer of TIMP-3 inhibits the TNF-α-induced activation of NF-κB. While these effects of TNF-α on apoptosis can be found in different fibroblasts, several anti-apoptotic molecules have been demonstrated to be upregulated specifically in RASF.
In this context, we have investigated the role of the small ubiquitin-like modifier SUMO-1 in the regulation of apoptosis. We found that increased levels of SUMO-1 in RASF but not in osteoarthritis synovial fibroblasts were associated with a reduced susceptibility of RASF to Fas-induced apoptosis. Using small interfering (si)RNA to knock-down the expression of SUMO-1 as well as retroviral gene transfer of SUMO-1, we could establish a functional relationship between the expression of SUMO-1 and the resistance of RASF against apoptosis. Moreover, it was demonstrated by gene transfer of the nuclear SUMO-protease SENP1 that, rather than by interacting directly with the Fas-associated death domain, SUMO-1 inhibits apoptosis through recruiting pro-apoptotic molecules such as DAXX into nuclear PML bodies, where they cannot exert their pro-apoptotic effects.
Taken together, there is growing evidence that in activated fibroblasts, such as in RASF, there is a close association between anti-apoptotic and destructive pathways. In addition to cytokine-dependent inflammatory mechanisms, the intrinsic upregulation of anti-apoptotic molecules contributes to the resistance of RASF against apoptosis. Post-translational modification of nuclear proteins through SUMO-1 appears to constitute an important mechanism of apoptosis regulation. Therefore, the inhibition of molecules that confer the resistance of RASF to apoptosis constitutes a most interesting therapeutic target.