The transcription factor PPARγ has been shown to modulate a number of inflammatory and catabolic responses in articular joint tissues and was suggested to be protective in OA and other arthritic diseases [14
]. Although many stimuli have been reported to regulate the expression of PPARγ in several cell types (including chondrocytes) [8
], little is known about the details of the exact mechanisms that govern its expression.
In the present study, we investigated the roles of the transcription factors Egr-1 and Sp1 in the downregulation of PPARγ expression by IL-1. We demonstrated that IL-1-mediated downregulation of PPARγ coincided with the induction of Egr-1 expression. In addition, downregulation of PPARγ expression was preceded by Egr-1 recruitment to, and concomitant reduced Sp1 occupancy at, the PPARγ promoter. Overexpression of Egr-1 suppressed, whereas that of Sp1 enhanced, PPARγ promoter activity. Furthermore, Egr-1 silencing prevented the downregulation of PPARγ expression by IL-1. Together, these data indicate that Egr-1 mediates the suppressive effect of IL-1 on PPARγ expression, likely through displacement of Sp1.
The PPARγ promoter contains an overlapping Sp1/Egr-1-binding site. The transcription factor Sp1 is ubiquitously expressed in cell lines and tissues and generally functions as an activator of transcription [29
]. The transcription factor Egr-1 is not expressed in normal tissues but is rapidly induced by inflammatory cytokines and growth factors [30
]. In promoters containing overlapping Sp1/Egr-1-binding sites, Egr-1 can function as a transcriptional activator or repressor. For example, Egr-1 has been shown to compete with Sp1 for an overlapping region in the promoter of platelet-derived growth factor-A (PDGF-A) and activates transcription in vascular endothelial cells [34
]. Egr-1-mediated transcriptional activation through displacement of Sp1 was also observed for N-myc downregulated gene (NDRG1
] and tissue factor [36
]. In contrast, other studies reported that Egr-1 competes with Sp1 and represses the transcription of a number of genes, including the β-adrenergic receptor [37
], protein tyrosine phosphatase 1B [38
], sterol regulatory element-binding protein 1 (SREBP-1
], the adenosine 5'-triphosphate-binding cassette transporter 2 (ABCA2
], and type II collagen [31
Here, we found that treatment of chondrocytes with IL-1 led to a time-dependent increase in Egr-1 expression, whereas the expression of Sp1 was not altered. This is consistent with previous studies showing that IL-1 is a potent inducer of Egr-1 expression in the chondrocyte cell line C-28/I2 [31
]. We then examined the effect of IL-1 on the recruitment of Egr-1 and Sp1 to the PPARγ promoter. ChIP results demonstrated that IL-1 induced Egr-1 recruitment to the PPARγ promoter with a parallel reduction in Sp1 occupancy, indicating that Egr-1 displaced the binding of Sp1. It is noteworthy that these changes at the PPARγ promoter were concomitant with the decrease in PPARγ expression, suggesting that Egr-1 recruitment to the PPARγ promoter could mediate the suppressive effect of IL-1 on PPARγ expression.
Using reporter gene assays, we found that IL-1 down-regulated PPARγ promoter activity and this effect was further potentiated by co-transfection with an expression vector for Egr-1. In contrast, Sp1 overexpression mitigated the suppressive effect of IL-1. This confirms the respective negative and positive regulation of the PPARγ promoter by Egr-1 and Sp1.
It should be noted that, in the absence of IL-1, transfection with Egr-1 had no effect on PPARγ promoter activity, indicating that Egr-1 needs to be activated to achieve inhibition of PPARγ promoter activity. In this context, it has been reported that the effects of Egr-1 on transcription are modulated through its phosphorylation by casein kinase II [41
] and extracellular signal-regulated kinase (Erk) [42
]. Egr-1 activity can also be regulated through acetylation, methylation, and ubiquitination, which are known for their impact on the activity of a number of proteins, including transcription factors. Indeed, Egr-1 harbors several consensus sites for acetylation and methylation. Further studies are needed to determine whether the repressive effect of Egr-1 on PPARγ expression involves such post-translational modifications.
Collectively, these results strongly suggest that the induction of Egr-1 expression and its recruitment to the PPARγ promoter mediate the suppressive effect of IL-1 on PPARγ expression. This is further supported by the fact that siRNA-mediated silencing of Egr-1 blocked IL-1-induced downregulation of PPARγ protein expression.
There are a number of potential mechanisms through which Egr-1 could mediate the downregulation of PPARγ expression by IL-1. The first possibility is that Egr-1 can repress transcription by displacing prebound Sp1. This is corroborated by our finding that the recruitment of Egr-1 to the PPARγ promoter paralleled reduced Sp1 occupancy. Moreover, several studies have shown that, through competition with promoter-associated Sp1, Egr-1 represses transcription of genes that harbor overlapping binding sites for Egr-1/Sp1 [34
]. Secondly, Egr-1 may inhibit PPARγ expression through direct binding to Sp1 and inhibition of its transcriptional activity. In this context, Egr-1 has been shown to inhibit Sp1 transcriptional activity, independently of DNA binding, through mechanisms that involve protein-protein interactions [41
]. Thirdly, Egr-1 can also repress transcription by interfering with the interaction between Sp1 and TATA-binding proteins (TBPs). Indeed, Sp1 has been shown to interact with TBPs [43
], and Egr-1 was reported to inhibit the binding of TBPs to target promoters [44
]. Finally, Egr-1 can attenuate Sp1 activities by competing for limited amounts of general transcriptional co-activators. Of note, Egr-1 has been reported to repress transcription by disrupting the interaction between Sp1 and CREB-binding protein (CBP/p300) [31
]. It is noteworthy that the overlapping binding site for Sp1 and Egr-1 in the PPARγ promoter can also bind the transcription factor Sp3. Indeed, Sp3 and Sp1 recognize and bind to the same DNA element with similar affinity and their DNA-binding domains share over 90% DNA sequence homology. Therefore, it is possible that Sp3 contributes to the regulatory effect of IL-1 on PPARγ expression. Indeed, IL-1 induces Sp3 expression and Sp3 down-regulates the transcriptional activity of Sp1 in chondrocytes. Such a mechanism was documented in IL-1-induced downregulation of type II transforming growth factor-beta (TGF-β) receptor [45
]. In addition to containing the overlapping Sp1/Egr-1-binding sites, the PPARγ promoter contains binding sites for other transcription factors known to be activated by IL-1, including activation protein-1 (AP-1), nuclear factor-kappa-B (NF-κB), nuclear factor of activated T cells (NF-AT), and myogenic differentiation 1 (MyoD). Although the role of these elements in IL-1-mediated downregulation of PPARγ expression is still unknown, we cannot exclude the possibility that activation of these transcription factors by IL-1 also participates in the downregulation of PPARγ expression. This is supported by the observation that siRNA-mediated silencing of Egr-1 did not completely reverse the suppressive effect of IL-1 on PPARγ expression.
The involvement of Egr-1 in IL-1-mediated downregulation of PPARγ expression may be of relevance for other stimuli known to modulate PPARγ expression. For instance, TNF-α and oxidative stress are known to down-regulate PPARγ expression [46
]. Interestingly, TNF-α and oxidative stress are potent inducers of Egr-1 expression [30
]. Therefore, it is possible that the induction of Egr-1 expression is part of the mechanisms by which TNF-α and oxidative stress down-regulate PPARγ expression.
Several studies have suggested roles for Egr-1 in the regulation of several genes involved in the pathogenesis of arthritis. For example, Egr-1 was shown to mediate TNF-α-induced MMP-9 [32
], IL-1-mediated suppression of type II collagen [31
], and TNF-α-mediated suppression of aggrecan [33
]. Egr-1 was also shown to positively regulate several inflammatory responses. Indeed, Egr-1 mediates IL-1-induced mPGES-1 expression and PGE2
production in several cell types, including chondrocytes and synovial fibroblasts [22
]. Furthermore, Egr-1 contributes to lipopolysaccharide-induced transcription of suppressor of cytokine signaling-1 (SOCS-1), a key regulator of lipopolysaccharide-induced cytokine production [49
]. Egr-1 was also demonstrated to play a critical role in the induction of a number of chemokines [50
] and cytokines, including IL-2, TNF-α [51
], IL-6, granulocyte colony-stimulating factor, and intracellular adhesion molecule [52
]. In addition to inflammatory and catabolic responses, chondrocyte apoptosis plays a significant role in the pathogenesis of OA. Of importance, Egr-1 was shown to positively regulate the expression of several pro-apoptotic factors, including TNF-α-related apoptosis-inducing ligand (TRAIL) [53
] and phosphatase and tensin homolog (PTEN) [54
]. These data, together with our findings that Egr-1 mediates the suppressive effect of IL-1 on PPARγ expression, suggest that therapeutic interventions that control Egr-1 expression may have protective effects in OA. Further in vivo
studies will be required to elucidate the exact role of Egr-1 in cartilage integrity and the pathogenesis of OA.
Finally, we showed that OA cartilage expresses high levels of Egr-1 compared with normal tissue. Positive immunoreactive staining for Egr-1 was located primarily in chondrocytes of the superficial layers. Interestingly, the levels of IL-1, a key player in the pathogenesis of OA, were reported to be elevated in these regions [55
], suggesting that IL-1 may be responsible for the observed increase in Egr-1 in OA cartilage. This is consistent with our findings that IL-1 is a potent inducer of Egr-1 expression in cultured chondrocytes. Our results are consistent with the findings of Trabandt and colleagues [56
], who showed elevated Egr-1 expression in rheumatoid synovium, which is characterized by increased production of inflammatory cytokines. In contrast, Wang and colleagues [57
] reported reduced expression of Egr-1 in OA cartilage. These apparent discrepancies in the expression of Egr-1 may be due to differences in study design. Indeed, Wang and colleagues [57
] performed their immunohistochemical study by using cartilage from two donors: one OA and one normal. The discrepancies may also lie in differences in tissue processing, antibody concentrations, or staining detection methodology.