|Home | About | Journals | Submit | Contact Us | Français|
Relaxin (Rlx) has antifibrotic effects in a number of tissues. Many of these effects are similar to those induced by the activators of peroxisome proliferator-activated receptor γ (PPARγ), raising the possibility that a mechanism for Rlx’s antifibrotic effects may involve activation of the PPARγ pathway. This study investigates the effect of Rlx on PPARs and their mechanism of upregulation. It shows that Rlx stimulates ligand-independent PPAR activation in a dose-dependent manner. The combined effect of Rlx and PPARγ agonists was superadditive, suggesting that both agents might be used together for an increased antifibrotic effect. Rlx caused increased expression of the PPARγ target genes CD36 and LXRα. Rlx’s effect was mimicked by forskolin, and partially blocked by pertussis toxin, suggesting that Rlx is working through a cAMP/PKA pathway to activate PPARγ. A better understanding of this pathway might help in the amelioration of fibrotic diseases.
Rlx is known to be a reproductive hormone by virtue of its action in softening of the cervix and vagina1. Its action involves widespread remodeling of the extracellular matrix involving altered secretion and degradation of matrix components2. The role of Rlx has been extended to nonreproductive tissues. Rlx has demonstrated antifibrotic actions on a number of cells, tissues and organs including lung, liver, heart and kidney2, 3. Experimentally induced fibrosis in mouse lung and rat kidney was restored to a normal phenotype after Rlx treatment2. Rlx-null mice developed pulmonary, cardiac and renal fibrosis, which could be reversed with Rlx treatment, suggesting a protective function for Rlx against fibrosis4. In the liver and liver fibroblasts, Rlx increased matrix metalloproteinase activity, decreased TIMP expression and reduced collagen production, suggesting a useful role in the treatment of hepatic fibrosis5, 6. Many of these effects are similar to those seen with agonists of PPARγ.
PPARs are nuclear transcription factors that heterodimerize with the 9-cis retinoic acid receptor (RXR) to induce transcription of target genes through their binding to PPAR response elements (PPRE)7. Three types of PPARs (PPARα, β/δ and γ) have been identified, and all bind to similar PPRE sequences7. Recent studies have implicated PPARs in the control of fibrosis8–10. Treatment with PPAR ligands, or expression of PPARs, suppresses fibrosis in organs such as heart, liver, kidney, and lung8. PPARγ agonists have been shown to prevent experimentally-induced fibrosis in some of these organs9, 11.
Because Rlx and PPARγ share a number of antifibrotic effects, it is possible that they possess common signaling pathways. Activation of PPARγ may be one possible downstream mechanism for the antifibrotic effects of Rlx. Therefore, we examined the effect of Rlx on PPARγ activity.
HEK-293T cells and THP1 monocytes were used for this study. To measure PPRE activity, 293T cells were transiently transfected with a RXFP1 expression plasmid12 (kindly provided by Aaron Hsueh, Stanford University), along with a reporter plasmid containing three tandem copies of the acyl CoA oxidase PPRE upstream of the firefly luciferase gene13 (ACO-PPRE-luc) (kindly provided by Brian Seed, Harvard University), and a renilla luciferase construct as a transfection control. Fugene-6 (Roche) was used as transfection reagent. Cells were treated for 24 hours with porcine Rlx (kindly provided by O.D. Sherwood, University of Illinois at Urbana-Champaign) or human InsL3 or Rlx-3 (Phoenix Pharmaceuticals). In some experiments, agonists of PPARα (WY14643, Calbiochem), PPARβ/δ (GW0742), PPARγ rosiglitazone (rosi), 15-deoxyδ12,14-prostaglandinJ2 (15dPGJ2), and the PPARγ ligand blocker (GW9662) (Cayman) were used. To elucidate signaling pathways, cells were treated with Forskolin, isobutylmethyl xanthine (IBMX), L-NAME (Sigma), Pertussis toxin, PD98059, LY294002 (Calbiochem), or PD169316 (Cayman). Firefly and renilla luciferase activities were measured using Dual-Glo luciferase assay kit (Promega). The gene expression levels of CD36 and LXRα were determined using Taqman real-time RT-PCR (Applied Biosystems), using 18S ribosome of RNA as an endogenous control, as shown previously14. Western blots for CD36 were performed using a rabbit polyclonal antibody (Abcam), as shown previously5.
A PPRE reporter assay was performed to investigate the effects of Rlx, Rlx-3 and InsL3 on PPARs. Rlx increased PPRE activation in a concentration-dependent manner, maximally at 1nM, with an EC50 of 24 pM. In contrast, InsL3, which has low affinity for RXFP1 did not activate PPRE even at a concentration of 100Nm, and Rlx-3, which has lower affinity for RXFP1 as compared to Rlx15, activated PPRE only at the highest concentration tested (100 nM) (data not shown). HEK293T cells which were not transfected with RXFP1 did not show any change in PPRE activity in response to Rlx (data not shown). These results suggest that Rlx increases the activity of one or more PPARs, and that the effect of Rlx is consistent with activation of RXFP1.
To further study the involvement of individual PPARs, specific PPAR agonists were used in the PPRE reporter assay. Rlx alone, in the absence of exogenous PPAR agonists, increased the activity by more than 70% as compared to control. PPARα and PPARβ/δ agonists (WY14643 and GW0742 respectively) had little or no effect on activity of PPRE compared to control (data not shown). The combination of PPARα or PPARβ/δ agonists with Rlx produced an activity which was similar to that of Rlx alone (data not shown). In contrast, the PPARγ agonist rosi increased PPRE activity, and the combined effect of Rlx with rosi was more than additive (Rlx 169.2 ± 9.1%, rosi 158.7 ± 15.0%, Rlx + rosi 270 ± 25.9%). Similar effects were also seen with 15dPGJ2, another PPARγ agonist (15dPGJ2 135.5 ± 22.1%, Rlx + 15dPGJ2 248 ± 41.6%). Because PPARγ agonists in combination with Rlx had a superadditive increase in PPRE activity, we chose to focus on the effect of Rlx on PPARγ.
The previous data showed Rlx activation of PPARγ in the absence of exogenous ligand. PPARγ is activated by binding of its ligand, but can also be regulated in a ligand independent manner by coactivators, corepressors and post translational modifications such as phosphorylation16–18. To determine whether the increase in PPRE activation by Rlx is dependent on ligand binding, we studied Rlx-mediated activation of PPARγ using a PPARγ ligand binding blocker (GW9662). In the presence of GW9662, PPRE activation in response to rosi was reduced, but there was no effect on the Rlx-induced response. GW9662 partially blocked the response of Rlx and rosi combined, reducing the response to the level of Rlx alone. This suggests that Rlx stimulates ligand-independent PPARγ activation.
An increase in PPARγ activity by Rlx should result in an increase in the expression level of PPARγ target genes such as CD36 and LXRα. To determine the effects of Rlx on target genes, HEK293T cells transiently transfected to express RXFP1 were treated with Rlx, and the expression levels of CD36 and LXRα were determined by real-time RT-PCR. Rlx increased the gene expression of PPARγ target genes as compared to control. These results were confirmed in THP1 cells, which naturally express RXFP-1. Furthermore, Rlx combined with rosi caused a synergistic increase in the gene expression level of CD36. The increase in CD36 by Rlx was further validated by western blotting using THP-1 cells. These data suggest the activation of PPARγ in response to Rlx.
Rlx has been reported to activate a number of signaling pathways1. To investigate possible signaling pathways triggered by Rlx to stimulate PPARγ activity, we exposed 293T cells expressing RXFP1 to various signaling inhibitors. Inhibitors of MAPK (PD98059, 20uM), p38-MAPK (PD169316, 1uM), PI-3K (LY294002, 50uM), or nitric oxide (L-NAME, 3mM) had no effect on Rlx-stimulated PPARγ activity. In contrast, the adenylate cyclase stimulator forskolin (2uM) mimicked the Rlx effects. There was no effect of the PDE inhibitor IBMX (200uM), but the Gi pathway inhibitor pertussis toxin (100ng/Ml) partially blocked Rlx’s effects. Taken together, these data suggest that Rlx acts through a cAMP-mediated mechanism that may involve Gs and Gi pathways.
Rlx has diverse functions. It has been shown to activate many pathways making it a pleiotrophic hormone1. We have shown for the first time evidence of Rlx activation of the nuclear receptor PPARγ. All the three PPARs known so far can recognize similar PPRE region in genes and activate them. Rlx enhanced the PPRE activity in a dose- dependent manner suggesting that Rlx increased the activity of one or more PPARs. Rlx-3 only activated PPRE at a high concentration (100nM), approximately 1000 times less potent than Rlx. This was surprising, because Rlx-3 is only 30–50 fold less potent than Rlx at stimulating cAMP production through RXFP115. One possible explanation could be because of Rlx-3 triggering a signaling pathway different from that of Rlx. Rlx did not stimulate PPRE in untransfected cells, which do not express RXFP1. Taken together, this suggests that Rlx activation of PPRE activity is mediated through RXFP1.
PPRE activity was increased by PPARγ agonists, and when combined with Rlx caused a superadditive effect. In contrast, PPARα and PPARβ/δ agonists had little or no effect on PPRE activity with or without Rlx. This suggests that PPARγ is the likely target of Rlx, therefore, we focused on PPARγ. However, a possible role for the remaining two PPARs can not be ruled out at this stage of investigation. Rlx increased the gene expression of the PPARγ target genes CD36 and LXRα, in cells overexpressing RXFP1. Rlx and the PPARγ agonist rosi caused a synergistic increase in the expression of CD36 in THP1 cells. This suggests a relationship between Rlx and PPARγ.
PPARs are transcription factors consisting of a ligand binding domain and a DNA binding domain. One of the possible ways Rlx could activate PPARγ is through ligand binding. On ligand binding, PPARγ heterodimerizes with RXR to bind and stimulate PPRE activation7. The presence of coactivators, corepressors or phosphorylation can also affect PPARγ activity16, 17. The ligand blocker GW9662 modifies the ligand binding site of PPARγ and blocks the binding of ligand to PPARγ. GW9662 had no inhibitory effect on Rlx activation of PPRE, suggesting that Rlx works independent of a ligand binding mechanism. Ligand-independent activation of PPARγ by Rlx might involve other regulators of PPARγ activity, such as PGC1α, p300, CBP, RXR, its ligand retinoic acid and many others, yet to be studied19, 20.
Rlx has been shown to activate cAMP/PKA, nitric oxide, MAPK and other pathways making it very diverse in action1. Rlx activation of PPARγ seems to involve a cAMP-mediated mechanism that may involve the Gs and Gi pathways, because forskolin mimicked Rlx effects and pertussis toxin partially blocked Rlx’s effects. However, as stated earlier, Rlx-3-induced cAMP production through RXFP1 is only 30–50 fold less potent than relaxin15, yet in our study Rlx-3 only detectably stimulated PPRE activtion at a very high concentration (100nM), suggesting that cAMP production alone may not be responsible for relaxin’s stimulation of PPARγ. However, in our study, Rlx does not seem to act on PPARγ activity through the MAPK, P38-MAPK, PI3K, or Nitric oxide pathways, as inhibitors of these pathways did not reduce the PPRE activity.
In summary, we have shown that Rlx increases the activity of PPARγ through a ligand-independent mechanism. This activation of PPARγ seems to involve cAMP/PKA pathway. The combination of Rlx with PPARγ agonists produced a synergistic effect on expression of PPARγ target genes. PPARγ has diverse biological implications in diseases ranging from obesity, diabetes, immunological disorders, asthma, atherosclerosis, cancer, fibrosis and Alzheimer’s disease21–24. The synergistic effects of Rlx with PPARγ agonist and better understanding of the signaling pathway will help to develop new therapies in diseases involving PPARγ.
This work was funded in part by the Department of Veteran Affairs, National Institute on Alcohol Abuse and Alcoholism (NIAAA), Bly Memorial Research Fund.