Main

Peroxisome proliferator-activated receptor alpha (PPARα) is a nuclear receptor transcription factor that regulates lipid metabolism. When bound to a ligand, PPARα forms a heterodimer with retinoid X receptor (RXR) that binds to the peroxisome proliferator response element in the promoter region of target genes and modulates transcription as depicted in Figure 1 . Activation of PPARα inhibits nuclear factor κB (NFκB) activation, and consequently repressing inflammation (1,2). Hepatic nuclear receptor RXR has also been shown to be important in hepatic responses to inflammation (3). Alternatively, NFκB activation increases expression of antiapoptotic genes, such as the caspase-8 inhibitor c-FLIP, which could lead to resistance to Fas ligand and/or tumor necrosis factor-dependent apoptosis (4). Previous studies have shown omega-3 long-chain polyunsaturated fatty acids (ω3PUFA) to be a PPARα ligand, thereby inhibiting NFκB activation (2). In studies where PPARα null mouse hepatocytes treated with nafenopin (a peroxisome proliferator), apoptosis was not suppressed, although treatment with tumor necrosis factor-α was able to suppress apoptosis in PPARα null hepatocytes (5). These studies demonstrated that PPARα mediates hepatocyte apoptosis, yet high concentrations of tumor necrosis factor-α can modulate and/or negate this response (5). These results suggest that PPARα inhibition of NFκB may play an important role in the overall cellular innate immune response and cell survival. The pathways affected by PPARα and NFκB are complex and in some cases, seemingly contradictory. These contradictory effects of NFκB on apoptosis have perplexed investigators studying interferon-mediated activation of NFκB, where this results in antiviral and proapoptotic activity. There is also evidence that micro-RNAs may have an important role in this complex signaling pathway (6). Finally, PPARα is known to inhibit apoptosis via NFκB-independent pathways, such as upregulation of the 14-3-3 proteins (7). The suppression of proinflammatory cytokine transcription by PPARα via NFκB would support our hypothesis, but the NFκB-mediated suppression of antiapoptotic pathways would suggest that ω3PUFA activation of PPARα could result in increased apoptosis.

Figure 1
figure 1

PPARα transcriptional activity. This figure is an illustration of PPARα transcription and its effects on lipid homeostasis and NFκB activation. PPARα, peroxisome proliferator-activated receptor alpha.

PowerPoint slide

Full size image

Omega-3 PUFA have been shown to be antiinflammatory and immunomodulatory in the attenuation of cellular injury (8,9,10,11,12,13). Recently, there has been much interest in the use of fish oil contacting ω3PUFA to treat parenteral nutrition-associated liver disease (PNALD) in children (14). Interestingly, ω3PUFA have been shown to paradoxically inhibit apoptosis and sensitize the cells to apoptosis (15,16,17,18). It is established that the composition and levels of free fatty acids can significantly alter cell survival pathways. Free fatty acids have been shown to inhibit serum-starved apoptosis in a murine cell model via the G-coupled protein receptor, GPR120 (19). In these experiments, both saturated and unsaturated free fatty acid-enhanced cell survival, but linolenic acid, palmitoleic acid, and docosahexaenoic acid (DHA) had the most potent effects on cell survival (20). In addition, GPR120 has recently been shown to be a receptor or sensor for ω3PUFA, mediating regulation of inflammation, and inhibiting both toll-like receptor 2/3/4 and tumor necrosis factor-α cascades (21). Both NFκB and c-Jun N-terminal kinases have been shown to play a role in the upregulation of death receptors in the presence of noxious stimuli (22,23).

In the present studies, we hypothesized that ω3PUFA would modulate PPARα activaton and/or mRNA levels resulting in the attenuation of hydrophobic bile acid-induced apoptosis via PPARα downstream effects. Expression of liver X receptor alpha (LXRα) and farnesoid X receptor (FXR) was also studied, as these nuclear receptors are recognized mediators of diverse metabolic pathways in liver, including inflammation, apoptosis and fibrosis (24,25).

Results

PPARα Activity

As shown in Figure 2 , treatment of Human hepatocellular carcinoma (HepG2) cells with ω3PUFA alone resulted in increased PPARα activity compared with controls and levels were statistically equal to that observed with positive control (recombinant PPARα). Treatment with ω3PUFA was increased 37% over untreated cells and DNA binding of PPARα was decreased by 45% in the presence of chenodeoxycholic acid (CDCA) (P < 0.001). There was a modest, trend to restoration (only decreased by 30% as compared with 45%) of PPARα activity to untreated control levels with the addition of ω3PUFA in the setting of CDCA treatment.

Figure 2
figure 2

PPARα activity. PPARα activity in HepG2 cells treated with CDCA with and without ω3PUFA. Values are based on a fold change relative to the vehicle control. Statistical significance was determined using an ANOVA with Tukey’s LSD. Each bar represents mean ± SEM for data from three culture wells. Statistical significance was determined using an ANOVA with Tukey’s LSD. Means of data from various treatment conditions with different letters above the bars are significantly different from each other at P < 0.05 by ANOVA. Those bars sharing the same symbol or no symbol are not significantly different from each other. PPARα, peroxisome proliferator-activated receptor alpha; CDCA, chenodeoxycholic acid; ω3PUFA, omega-3 long-chain polyunsaturated fatty acids. LSD, least significant difference.

PowerPoint slide

Full size image

PPARα, LXRα, FXR, and RXR mRNA Levels

Treatment with CDCA alone resulted in 67% reduction in PPARα mRNA, as compared with cells treated with a vehicle control ( Figure 3a ). Treatment with CDCA alone resulted in a twofold increase in LXRα mRNA that was attenuated by the addition of ω3PUFA ( Figure 3b ). No significant changes in RXR or FXR mRNA levels were observed with either treatment (data not shown).

Figure 3
figure 3

PPARα and LXRα mRNA levels. (a) PPARα, (b) LXRα mRNA levels were measured by quantitative RT-PCR. Values are based on a fold change relative to the vehicle control. Statistical significance was determined using an ANOVA with Tukey’s LSD. Each bar represents mean ± SEM for data from three culture wells. Statistical significance was determined using an ANOVA with Tukey’s LSD. Means of data from various treatment conditions with different letters above the bars are significantly different from each other at P < 0.05 by ANOVA. Those bars sharing the same symbol or no symbol are not significantly different from each other. PPARα, peroxisome proliferator-activated receptor alpha; LXRα, liver X receptor alpha; RT-PCR, reverse-transcriptase PCR. LSD, least significant difference.

PowerPoint slide

Full size image

Role of PPARα in the Attenuation of Bile Acid-Induced Apoptosis

As shown in Figure 4 , HepG2 cells treated with ω3PUFA, in addition to the CDCA challenge, exhibited a 47% attenuation of apoptosis, as reflected by caspase-3/7, when compared with cells treated with CDCA alone. This ω3PUFA protective effect was not observed with blockade of PPARα with specific antibody. With antibody dose escalation, there was no significant difference in caspase-3/7 with increasing doses of antibody. These results suggest that the attenuation of apoptosis occurs via PPARα-dependent pathways.

Figure 4
figure 4

Caspase-3/7 activity. Bile acid-induced apoptosis evaluated by caspase-3/7 with and without the addition of an antibody specific to PPARα with ω3PUFA treatment. Data are represented based on relative fluorescence units above control. Each bar represents mean ± SEM for data from three independent experiments. Statistical significance was determined using an ANOVA with Tukey’s LSD. Means of data from various treatment conditions with different letters above the bars are significantly different from each other at P < 0.05 by ANOVA. Those bars sharing the same symbol or no symbol are not significantly different from each other. PPARα, peroxisome proliferator-activated receptor alpha; ω3PUFA, omega-3 long-chain polyunsaturated fatty acids. LSD, least significant difference.

PowerPoint slide

Full size image

Discussion

The present studies support a role for PPARα in protection from bile acid-induced apoptosis in hepatocytes and may, at least partially, mediate the attenuation of apoptosis by ω3PUFA, which may represent one mechanism for the positive effect of ω3PUFA treatment in PNALD. Activation and mRNA levels of PPARα were reduced by treatment with the lipophilic bile acid, CDCA, at a dose associated with hepatocyte apoptosis. When hepatocytes were exposed to ω3PUFA alone, PPARα activity and mRNA levels were increased. When cells were treated with CDCA with the addition of ω3PUFA, PPARα mRNA levels were similar to those of control cells, and there was a trend to restoration of PPARα activation. Our findings are consistent with studies of CFTR-knockout mice and isolated macrophages from CFTR-knockout mice in which pretreatment with DHA resulted in increased PPARα mRNA levels. This increase was blocked by the addition of a PPARα antagonist (26,27). In contrast, studies in a mouse model of PNALD using PPARα-knockout and wild-type mice resulted in increased hepatic triglyceride concentrations in both WT and KO mice treated with PN with the addition of ω3PUFA and ω6PUFA as compared with chow fed mice (28). However, animals receiving ω3PUFA had lower hepatic triglyceride content compared with animals fed ω6PUFA in both WT and KO mice. Therefore, the investigators concluded that ω3PUFA attenuation of hepatic triglyceride accumulation in this model of PNALD is not via PPARα-dependent pathways (28). While the effects of ω3PUFA on hepatic triglyceride accumulation may not be PPARα dependent in this in vivo model of PNALD, we have shown in the present studies in an in vitro model that bile acid-induced apoptosis often associated with PNALD is attenuated by ω3PUFA via PPARα-dependent pathways. This is not surprising in light of the complex and multifactorial nature of PNALD pathogenesis, which involves hepatocellular injury, inflammation, and apoptosis in addition to steatosis. Our data support the hypothesis that ω3PUFA are functioning, at least partially, via PPARα-dependent mechanisms to attenuate direct bile acid-induced apoptosis. The mechanisms by which ω3PUFA reduce hepatic steatosis may indeed occur independently of PPARα.

For these experiments, we chose to use an equal molar ratio of eicosapentaenoic acid to DHA, because our group has previously shown eicosapentaenoic acid and DHA to be synergistic in the attenuation of bile acid induce apoptosis (29). We observed a 67% reduction in PPARα mRNA levels when cells were treated with CDCA as compared with cells treated with a vehicle control. The addition of ω3PUFA resulted in restoration to the level of control cells. We did not observe a significant difference in RXR mRNA levels, which was not unexpected, since abundance of RXR and its ligand, cis retinoic acid, typically precludes it from becoming rate limiting in its role as a heterodimer partner with transcription factors.

Treatment with CDCA alone resulted in a twofold increase in LXRα mRNA levels that was attenuated by the addition of ω3PUFA. It is noteworthy that LXRα has been shown to have specific roles in regulating the innate immune response (30). Also, studies in LXRα null macrophages have shown that LXRα has an important role in both innate and adaptive immunity, specifically in apoptosis (31). Cancer cells of diverse origin may be programmed to enter apoptosis by LXRα activation (32).Therefore, it is possible that LXRα may be an important mediator of CDCA-induced apoptosis, and its suppression by ω3PUFA may represent an additional pathway for the beneficial effect of ω3PUFA on apoptosis. Interestingly, FXR mRNA abundance was not affected by CDCA or ω3PUFA treatment.

In summary, these studies suggest roles for PPARα, as well as LXRα, in the regulation of bile acid-induced apoptosis in hepatocytes and its suppression by ω3PUFA. It is possible that PPARα activation and LXRα suppression may be promising mechanisms by which ω3PUFA attenuate the bile acid-induced hepatocellular injury that occurs in cholestatic liver disease, such as that seen in PNALD. However, these conclusions must be tempered by several caveats. First, these studies were performed in a cell culture system using hepatoma-derived hepatocytes. Second, we explored but one mechanism of cell injury, bile acid toxicity, and its induction of apoptosis, whereas multiple pathogenic mechanisms are operative in clinical PNALD. Finally, more work will be required to further define additional potential pathways involved in these observed ω3PUFA effects and to translate these findings to useful therapy for PNALD in human infants.

Methods

Cell Culture

Because this was an in vitro study and did not involve the use of human subjects or animals, the University of Tennessee Health Science Center did not require approval by the Institutional Review Board. Cell culture experiments were performed using HepG2 cells obtained from the American Type Culture Collection (Rockville, MD). Cell culture experiments used standard medium, incubation conditions, and passages according to the supplier’s instructions. HepG2 cells were treated with CDCA (200 μmol/l) (Sigma-Aldrich, St. Louis, MO) with and without the addition of ω3PUFA (5 μmol/l eicosapentaenoic acid combined with 5 μmol/l DHA) (Sigma-Aldrich). Concentrations of CDCA and ω3PUFA for these experiments were based on our previously published work (29).

PPARα Activation

Activation of PPARα was evaluated by isolation of nuclear protein after 4-h treatment and measured by assessing transcription factor activity in a nonradioactive, sensitive ELISA method for detecting PPARα DNA binding activity in nuclear extracts (PPARα Transcription Factor Assay Kit, Cayman Chemical, Ann Arbor, MI). This kit contained a positive control of clarified cell lysate and a competitor of double-strand DNA.

PPARα mRNA Analysis

HepG2 cells were treated with 200 μmol/l CDCA ± ω3PUFA for 0.5 h, and PPARα, LXRα, FXR, and RXR mRNA levels were evaluated by reverse-transcriptase PCR using an ABI 7500 Sequence Detector and SYBR Green PCR Master Mix according to the manufacturer’s instructions. Levels of mRNA were normalized to 18s ribosomal RNA quantified simultaneously in a multiplex reverse-transcriptase PCR reaction. Specific primers were designed for the genes of interest (PPARα, FXR, LXRα, and RXR). See Table 1 for primer sequences. All samples were analyzed in triplicates.

Table 1 Reverse-transcriptase PCR (RT-PCR) primer sequences
Full size table

Caspase 3/7 Activity

In order to determine if the attenuation of bile acid-induced apoptosis by ω3PUFA occurs via PPARα-dependent pathways, HepG2 cells were treated with a PPARα polyclonal antibody (PPARα H-98: sc-9000) purchased from Santa Cruz Biotechnology (Dallas, TX), and apoptosis was evaluated using caspase-3/7. Cultured HepG2 cells were treated with ω3PUFA (eicosapentaenoic acid 5 μmol/l + DHA 5 μmol/l) alone; 200 μmol/l CDCA alone; and then with the combinations ω3PUFA, and varying concentrations of PPARα antibody (3–10 µg/ml). Apoptosis was evaluated using the Apo-ONE Homogeneous Caspase-3/7 Assay purchased from Promega Corporation (Madison, WI) and performed according to the manufacturer’s instructions. HepG2 cells were treated for 12 h followed by the addition of caspase-3/7 reagent and incubation for 4 h in the dark on a rocking shaker at low speed. Results were read at fluorescein 485/535nm with a Victor 2, Perkin-Elmer Wallace 1420 multilabel counter (Shelton, CT) as previously described (29).

Statistical Analysis

All data represent at least three separate and independent experiments. Data are depicted as mean ± SEM. A one-way ANOVA was used to compare differences between groups and a post-hoc Tukey’s HSD (honestly significant difference) test was used to correct for multiple comparisons. A two-tailed P value of 0.05 was used to reject the null hypothesis.

Statement of Financial Support

This research was supported by a Le Bonheur Children’s Hospital Junior Faculty Grant, Memphis, TN.

Disclosure

The authors of this manuscript do not hold any financial ties to products in the study or potential/perceived conflicts of interest; there are no disclosures.