Basal Autophagy Is Necessary for A Pharmacologic PPARα Transactivation

Autophagy is a conserved cellular process of catabolism leading to nutrient recycling upon starvation and maintaining tissue and energy homeostasis. Tissue-specific loss of core-autophagy-related genes often triggers diverse diseases, including cancer, neurodegeneration, inflammatory disease, metabolic disorder, and muscle disease. The nutrient-sensing nuclear receptors peroxisome proliferator-activated receptor α (PPARα) plays a key role in fasting-associated metabolisms such as autophagy, fatty acid oxidation, and ketogenesis. Here we show that autophagy defects impede the transactivation of PPARα. Liver-specific ablation of the Atg7 gene in mice showed reduced expression levels of PPARα target genes in response to its synthetic agonist ligands. Since NRF2, an antioxidant transcription factor, is activated in autophagy-deficient mice due to p62/SQSTM1 accumulation and its subsequent interaction with KEAP1, an E3 ubiquitin ligase. We hypothesize that the nuclear accumulation of NRF2 by autophagy defects blunts the transactivation of PPARα. Consistent with this idea, we find that NRF2 activation is sufficient to inhibit the pharmacologic transactivation of PPARα, which is dependent on the Nrf2 gene. These results reveal an unrecognized requirement of basal autophagy for the transactivation of PPARα by preventing NRF2 from a nuclear translocation and suggest a clinical significance of basal autophagy to expect a pharmacologic efficacy of synthetic PPARα ligands.


Introduction
Autophagy is an evolutionarily conserved catabolic and adaptive process induced by diverse conditions, including the deprivation of nutrient or growth factors, pathogen infection, hypoxia, and even exercise [1][2][3][4][5][6][7]. This leads to nutrient recycling, prevents cellular damage, and maintains tissues homeostasis in response to energy shortage and constant external and internal harmful insults. Therefore, autophagy has inevitable cytoprotective roles and needs to be precisely and tightly controlled to properly respond to various stimuli. It has been shown that autophagy dysfunctions are intimately associated with a wide range of human pathogenesis, including neurodegeneration, inflammatory and metabolic diseases, and cancers [3,8,9]. Recent studies reveal that autophagy used to be considered a nonselective process is able to selectively eliminate unwanted, potentially cytotoxic molecules, such as damaged organelles or aggregated proteins, serving as a primary cytoprotective function [10,11]. This leads to intensive research to modulate autophagy activity due to its therapeutic potential in many human diseases [12].

RNA Purification, cDNA Synthesis, and qPCR Analysis
Total RNA was purified from the snap-frozen mouse liver or AML12 cells using Trizol Reagent according to the manufacturer's instructions (Invitrogen: Waltham, MA, USA). RNA concentration and quality were determined by Nanodrop. A total of 1 µg of total RNA was converted to cDNA using PrimeScript TM 1st strand cDNA Synthesis kit (Takara Bio Inc.: Kusatsu, Shiga, Japan). qPCR was performed with RbTaq TM qPCR 2X PreMIX (SYBR Green with high ROX) (Enzynomics: Daejeon, Korea) on StepOnePlus Real-Time PCR systems (Applied Biosystems: Waltham, MA, USA). All reactions were carried out in either duplicate or triplicate, and C t values were obtained. mRNA levels were normalized to the expression levels of the housekeeping gene 36b4 (also known as Rplp0) using the standard curve method. Information of mouse primer sequences for qPCR analysis is listed in Table S1.

Immunoblot Analysis
100 mg of frozen liver tissue was homogenized in 1 mL of ice-cold RIPA buffer (25 mM Tris-HCl, pH 7.6, 150 mM NaCl, 1% NP-40, 1% sodium deoxycholate, 1 mM EDTA) supplemented with cOmplete protease and PhosSTOP inhibitors (Roche: Basel, Switzerland), followed by brief sonication in ice. The lysates were centrifuged at 13,000 rpm for 15 min at 4 • C to collect supernatants, of which protein concentration was determined by the Pierce BCA protein assay kit (ThermoFisher Scientific: Waltham, MA, USA). A total of 25 µg of protein lysates were loaded on a 10% SDS-PAGE gel to be resolved, and the proteins were transferred onto an Immuno-Blot PVDF membrane (Bio-Rad Laboratories: Hercules, CA, USA). The membrane was blocked with 5% BSA/TBST (Tris-buffered saline with 0.1% Tween-20) solution for 1 h at room temperature. Primary antibodies were diluted in 5% BSA/TBST solution, which was applied onto the membranes at 4 • C for 16 h. The membrane was washed three times at room temperature (each wash for 15 min; the first and second washes with 5% BSA/TBST solution; the third wash with TBST solution). After washes, the membrane was incubated with an anti-rabbit IgG HRP-linked antibody (Cell Signaling Technology (Danvers, MA, USA)) at room temperature for 1 h, followed by three washes as shown above. The bands were visualized using the Pierce ECL Western Blotting Substrate (ThermoFisher Scientific: Waltham, MA, USA).

Statistical Analysis
All values are shown as mean ± s.e.m. and error bars were derived from biological replicates rather than technical replicates. Significant differences between the two groups were evaluated using a two-tailed, unpaired Student t-test, which was found to be appropriate as groups displayed a normal distribution, and comparable variance of p < 0.05 was considered statistically significant.

Autophagy Inhibition Impairs a Pharmacologic PPARα Transactivation
To examine any crosstalk between autophagy and nuclear receptor signaling, liverspecific Atg7 knockout (Atg7 LKO ) mice were generated by crossing male Alb-Cre mice with female Atg7 F/F mice. Consistent with previous reports, hepatomegaly phenotypes were observed in Atg7 LKO mice compared with Atg7 F/F control littermates ( Figure S1a). As expected, hepatic expression and protein levels of the Atg7 gene were also markedly reduced in Atg7 LKO mice ( Figure S1b,c).
A previous study has shown that increased serum levels of β-hydroxybutyrate, one species of ketone bodies in Atg7 F/F control littermates upon treatment of a synthetic PPARα agonist GW7647 [38], were blunted in Atg7 LKO mice. These results allowed us to set up a hypothesis that core autophagy-related genes are essential for the normal function of PPARα. To test this, hepatic expression levels of PPARα target genes (Acox1, Acot3, Fgf21, Pdk4, Acot1, and Cidec) were determined in fed or fasted Atg7 F/F control littermates and Atg7 LKO mice in response to intraperitoneal treatment of GW7647 (Figure 1a). Consis-tent with a previous report, pharmacologic inductions of hepatic PPARα target genes were markedly increased in Atg7 F/F mice compared with those of vehicle-treated Atg7 F/F mice. However, these inductions were largely decreased in GW7647-treated Atg7 LKO mice (Figure 1b). To investigate whether these in vivo results recapitulate in cultured cells, we used mouse embryonic fibroblasts (MEFs) derived from wild-type, Atg5 −/− , or Atg7 −/− mice ( Figure S2a) [17,39]. As expected, expression levels of Atg5 or Atg7 were almost completely absent in either Atg5 −/− or Atg7 −/− MEFs. Treatment of Wy-14,643, a synthetic PPARα agonist in wild-type MEFs robustly increased expressions of PPARα target genes Acox1, Pdk4, Acot3, and Ucp2, but these responses were markedly blunted in either Atg5 −/− or Atg7 −/− MEFs ( Figure S2b,c). These results suggest that core-autophagy genes are required for the pharmacological PPARα activation.
To examine whether pharmacologic inhibition of autophagy also shows similar results, mouse hepatocyte cell line AML12 cells were dose-dependently treated with 3-methyladenine or bafilomycin A1, well-known autophagy inhibitors with or without synthetic PPARα agonist ligands GW7647 or Wy-14,643 ( Figures S2d and 1c) [40,41]. Consistent with in vivo mouse results shown in Figure 1b, inductions of most of PPARα target genes upon GW7647 or Wy-14,643 treatment were dramatically blunted by a dose-dependent treatment of 3-methyladenine or bafilomycin A1 ( Figures S2f and 1d,e). All these in vivo and cell culture results indicate that autophagy inhibition can impair a pharmacologic PPARα activity.
Previous reports have also demonstrated that genetic ablations of core autophagyrelated genes, including Atg7, lead to the accumulation of p62, an autophagy adaptor protein [42]. We could also confirm marked accumulations of p62 in the livers of Atg7 LKO mice ( Figure S1d). This causes interactions with KEAP1, an E3 ubiquitin ligase, which liberates NRF2 from the ubiquitin-mediated proteasomal degradation system. Free cytoplasmic NRF2 is then translocated into the nucleus, where it acts as an antioxidant transcription factor to remove reactive oxygen species [21]. As expected, we also observed dramatically increased expression levels of the Nqo1 gene, an NRF2 target gene in the livers of Atg7 LKO mice compared with those of Atg7 F/F mice ( Figure S1e). Atg5 −/− and Atg7 −/− MEFs also increased Nqo1 and Hmox1 expressions ( Figure S2a). Based on these data, we concluded that our Atg7 LKO mice also increased hepatic NRF2 activation. It is of interest to note that the pharmacologic PPARα activation by GW7647 treatment does not significantly alter hepatic Nqo1 expression compared with vehicle-treated counterparts ( Figure S1e). We also found that a dose-dependent treatment of bafilomycin A1 in AML12 cells markedly increased expression levels of the Hmox1 gene, another known NRF2 target gene. However, expression levels of the Nqo1 gene showed opposite results, which might reflect different expression kinetics of two known NRF2 target genes in response to autophagy inhibition ( Figure S1f,g). Treatment of 3-methyladenine in AML12 cells showed similar results ( Figure S2d,e).

SFN-Mediated NRF2 Activation in a Dose-Dependent Manner Impairs a Pharmacologic PPARα Transactivation
To examine whether NRF2 activation directly affects PPARα activity, AML12 cells were dose-dependently treated with sulforaphane [43] or dimethylfumarate [44], potent NRF2 activators in the presence or absence of synthetic PPARα agonist ligands GW7647 or Wy-14,643 (Figures 2a and S2a). As expected, sulforaphane or dimethylfumarate dramatically increases the expression of NRF2 target genes Nqo1 and Hmox1, which are not altered by the treatment of synthetic PPARα agonists (Figures 2b,c and S2b,c). These data indicate that a pharmacologic PPARα activation may not affect sulforaphane or dimethylfumaratemediated NRF2 activation. Intriguingly, however, completely opposite results were observed in terms of a pharmacologic PPARα activation in the presence of NRF2 activators. We found that dose-dependent treatments of sulforaphane or dimethylfumarate markedly downregulated expression of PPARα target genes Pdk4, Acot3, and Ucp2 in response to GW7647 or Wy-14,643 treatment (Figures 2d,e and S2d,e). All these data suggest that NRF2 activation might mediate the suppression of a pharmacologic PPARα activation.

SFN-Mediated NRF2 Activation in a Time-Dependent Manner Impairs a Pharmacologic PPARα Transactivation
To further thoroughly investigate whether NRF2 activation directly affects PPARα activity, AML12 cells were time-dependently treated with sulforaphane or dimethylfumarate in the presence or absence of GW7647 or Wy-14,643 (Figures 3a and S3a). As expected, these NRF2 activators dramatically increase expression of the Nqo1 gene in a time-dependent manner (Figures 3b and S3b). However, as mentioned above, we again observed different expression kinetics of two known NRF2 target genes, Nqo1 and Hmox1. A total of 6 h treatment of NRF2 activators showed the highest expression levels of Hmox1 gene whose mRNA levels were gradually decreased by further increasing treatment time of NRF2 activators (Figures 3b and S3b). Nevertheless, time-dependent treatments of sulforaphane or dimethylfumarate in AML12 cells dramatically suppress the expression of PPARα target genes Pdk4, Acot3, and Ucp2 in response to GW7647 or Wy-14,643 (Figures 3c,d and S3c,d). All these data support that acute NRF2 activation might mediate the suppression of the pharmacologic PPARα activation.

NRF2 Activation Is Necessary and Sufficient for Suppressing a Pharmacologic PPARα Transactivation
To further confirm these results in vivo, C57BL/6J wild-type mice were orally gavaged with butylated hydroxyanisole (BHA), an NRF2 activator, for 3 days [45]. On day 3, these mice were intraperitoneally treated with vehicle or GW7647, as shown in Figure 4a. After that, hepatic expression levels of PPARα target genes Pdk4, Ucp2, Acot3, Acox1, and Fgf21 were determined. As expected, GW7647 treatment markedly increased mRNA levels of these PPARα target genes, but these responses were significantly blunted in the livers of BHA-treated mice (Figure 4b). These data indicate that BHA-mediated NRF2 activation is able to suppress a pharmacologic PPARα activation.
To genetically validate these results, we generated liver-specific Keap1 knockout (Keap1 LKO ) mice by crossing male Alb-Cre mice with female Keap1 F/F mice. Since floxed Keap1 alleles in Keap1 F/F mice turned out to be hypomorphic, leading to the spontaneous activation of NRF2, we used Alb-Cre control littermates as control mice instead of Keap1 F/F mice (Figure 4c) [46]. As expected, Keap1 LKO mice showed almost an absence of hepatic Keap1 expressions but marked increased mRNA levels of Nqo1 and Gasta1, indicating that a potent NRF2 activation occurs in the livers of Keap1 LKO mice. Once again, hepatic expression levels of PPARα target genes Pdk4, Ucp2, Cidec, Acot1 to 3, and Acox1 were significantly downregulated in GW7647-treated Keap1 LKO mice (Figure 4d). These data suggest that genetic NRF2 activation leads to the inhibition of the pharmacological PPARα activation.
To genetically validate whether these blunted pharmacological PPARα activations mediated by NRF2 activators depend on NRF2 proteins, we first generated liver-specific Nrf2 knockout (Nrf2 LKO ) mice by crossing male Alb-Cre mice with female Nrf2 F/F mice. Nrf2 F/F control littermates and Nrf2 LKO mice were then orally gavaged with BHA for 3 days. On day 3, these mice were intraperitoneally treated with vehicle or GW7647, as shown in Figure 5a. As expected, hepatic Nrf2 mRNA levels were markedly reduced in Nrf2 LKO mice, confirming a genetic ablation of the Nrf2 gene in the liver (Figure 5b). Like the results obtained in C57BL/6J mice (Figure 4), BHA treatment also robustly increased hepatic expression levels of NRF2 target genes Nqo1 and Gasta1 in Nrf2 F/F control littermates, but these responses were significantly blunted in Nrf2 LKO mice (Figure 5b).  The 8-to 9-week-old male C57BL/6J wild-type mice were orally gavaged with vehicle or butylated hydroxyanisole (BHA, 200 mg/kg BW) once a day for 3 days. During the last 24 h, all mice were intraperitoneally injected with vehicle (0.1% DMSO in 4:1 ratio of PEG 400 and Tween 80) or GW7647, a synthetic PPARα agonist (10 mg/kg BW) twice a day (first injection: 00:00 am, second injection: 12:00 pm). A total of 5 h after last treatment, all mice were killed to collect livers of which total RNAs were prepared for qPCR analysis. (b) Hepatic expression levels of PPARα target genes Pdk4, Upc2, Acot3, Acox1, and Fgf21 were determined as shown in panel (a) by qPCR analysis. n = 8 per group, * p < 0.05, ** p < 0.01 vs. WT treated with vehicle. (c) A schematic diagram of an experimental procedure in mice. The 8-to 9-week-old male Alb-Cre/+ and Keap1 LKO mice were intraperitoneally injected with vehicle (0.1% DMSO in 4:1 ratio of PEG 400 and Tween 80) or GW7647 (10 mg/kg BW) twice a day. A total of 5 h after last treatment, all mice were killed to collect livers, of which total RNAs were isolated for qPCR analysis. (d) Hepatic expression levels of Keap1, NRF2 target genes (Nqo1 and Gsta1), and PPARα target genes Pdk4, Upc2, Cidec, Acot1 to 3, and Acox1 were determined shown in panel (c) by qPCR analysis. n = 4 per group, ** p < 0.01 vs. Alb-Cre/+ treated with vehicle. # p < 0.05, ## p < 0.01. Data represent mean ± s.e.m. and are plotted as fold change. Each dot indicates an individual mouse. Statistics by a two-tailed, unpaired Student t-test. WT, wild-type. Figure 5. NRF2 activation is necessary and sufficient for suppressing pharmacologic PPARα transactivation. (a) A schematic diagram of an experimental procedure in mice. The 8-to 9-week-old male Nrf2 F/F and Nrf2 LKO mice were orally gavaged with vehicle or butylated hydroxyanisole (BHA, 200 mg/kg BW) once a day for 3 days. During the last 24 h, all mice were intraperitoneally injected with vehicle (0.1% DMSO in 4:1 ratio of PEG 400 and Tween 80) or GW7647, a synthetic PPARα agonist (10 mg/kg BW) twice a day (first injection: 00:00 am, second injection: 12:00 pm). A total of 5 h after last treatment, all mice were killed to collect livers of which total RNAs were prepared for qPCR analysis. (b) Hepatic expression levels of Nrf2 and its target genes Nqo1, Gsta1, and Hmox1 were determined in AML12 cells shown in panel (a) by qPCR analysis, * p < 0.05. (c) Hepatic expression levels of PPARα target genes Pdk4, Upc2, Cidec, and Acot1 to 3 were determined as shown in panel (a) by qPCR analysis. n = 3-4 per group, ** p < 0.01 vs. Nrf2 F/F treated with vehicle. # p < 0.05, ## p < 0.01. Data represent mean ± s.e.m. and are plotted as fold change. Each dot indicates an individual mouse. Statistics by a two-tailed, unpaired Student t-test.
These results demonstrate that induction of Nqo1 and Gsta1 genes by BHA treatment largely depends on NRF2. Moreover, GW7647 treatment in Nrf2 F/F control littermates also robustly induced hepatic expressions of PPAR target genes Pdk4, Ucp2, Cidec, and Acot1 to 3, but these inductions were significantly blunted in BHA-treated Nrf2 F/F counterparts. However, these suppressed expression levels of PPAR target genes were almost completely lost in BHA-treated Nrf2 LKO mice, demonstrating that blunted pharmacological transactivation of PPARα mediated by BHA treatment indeed requires NRF2 proteins (Figure 5c).

Discussion
In this study, we showed that inhibition of basal autophagy impairs a pharmacological activation of the fasting-activated nuclear receptor PPARα. Moreover, our data also suggested that NRF2 known to be activated in autophagy-defective conditions might mediate this effect. In the clinical aspects, this phenomenon could be critical, in particular for hyperlipidemia patients taking fibrate derivatives. Simultaneous activation of both NRF2 and PPARα transcription factors might diminish the beneficial effects of fibrate medications.
Similar results have also been reported by the Komatsu and Panasyuk laboratories in regards to the impairment of PPARα transactivation by autophagy defects, although their molecular mechanisms are distinct from ours [47,48]. They have found that NCoR1 corepressor proteins are remarkedly accumulated in autophagy-defective hepatocytes obtained from Atg5, Atg7, or Vps15 conditional knockout mice. These increased levels of NCoR1 tend to be recruited to PPARα, which leads to suppression of its target gene expression. In terms of the detailed molecular mechanism by which autophagy defects impede PPARα signaling, we speculate that there are many more mechanisms yet to be discovered.
Similar to our findings, sulforaphane treatment in 3T3-L1 cells inhibits adipocyte differentiation by reducing the expression of PPARγ, a master regulator of adipogenesis, indicating that NRF2 activation might impair functionalities of other PPARs [30]. Conversely, NRF2 activation also increases the expression of PPARs. Treatment of either BHA or sulforaphane has been shown to induce hepatic PPARγ or PPARβ/δ expression in livers of wild-type mice [31,32]. Consistently, in our experiments in Figure 4a, we could confirm that a 3-day treatment of BHA increases expression of both Pparγ and Pparβ/δ genes (data not shown). Moreover, cooperative actions between PPARγ and NRF2 have also been reported. In oxidative stress conditions, both transcription factors reciprocally induce their expression as an upstream regulator, resulting in increased expressions of many cytoprotective genes encoding HO-1, CAT, SOD, and GST. These enzymes have been known to alleviate reactive oxygen species in cells and tissues [33,34].
It has been known that heme oxygenase-1 (HO-1) encoded by the Hmox1 gene, a known NRF2 target gene, produces biliverdin from heme, and biliverdin reductase then converts biliverdin to bilirubin [49]. Interestingly, the Hinds laboratory has shown that bilirubin might act as an endogenous agonist ligand for PPARα, although its physiological relevance to the fasting-activated PPARα signaling should be confirmed [50,51]. According to this scenario, PPARα target genes might be induced rather than reduced via an HO-1/bilirubin/PPARα axis. However, we found that treatment of sulforaphane, dimethylfumarate, or BHA, or liver-specific Keap1 deletion did not increase basal expression levels of most PPARα target genes at least we examined, indicating that NRF2 activation might not affect a basal activity of PPARα.
In the future, the generation of liver-specific Atg7 and Nrf2 double knockout (Atg7 LKO ; Nrf2 LKO ) mice may provide a valuable resource to address whether NRF2 activation in Atg7 LKO mice is genetically required for suppressing the pharmacological activation of the nuclear receptor PPARα. Moreover, the generation of liver-specific Nrf2 and Keap1 double knockout (Nrf2 LKO ; Keap1 LKO ) mice may also give us important insights to examine whether the blunted pharmacological activation of PPARα in Keap1 LKO mice is genetically dependent on NRF2.
Further investigations are also necessary to define detailed molecular mechanisms by which nuclear NRF2 proteins interfere with the functions of the nuclear receptor PPARα. This could be achieved in a variety of mechanisms, including a direct protein-protein interaction, squelching coactivators, interfering with the recruitment of coactivators, promoting recruitment of corepressors, inhibiting nuclear receptor binding to its response element of target genes, and so on.

Conclusions
In summary, we propose an essential role of basal autophagy activity for the pharmacological functions of the nuclear receptor PPARα. Autophagic defects lead to NRF2 translocation from the cytoplasm to the nucleus, where it acts as an antioxidant transcription factor by regulating many cytoprotective genes. We speculate that these nuclear NRF2 transcription factor proteins could interfere with the activities of many other transcription factors, including PPARα, via a variety of mechanisms ( Figure 6). This study establishes the unexpected role of basal autophagy for the therapeutic effects of pharmacologically targeting the nuclear receptor PPARα, which may be useful for the treatment of diverse metabolic diseases. Figure 6. Working model of basal autophagy defects or NRF2 activators that impair pharmacological transactivation of the nuclear receptor PPARα. Similar to NRF2 activators, autophagy inhibition by pharmacologic or genetic manipulations liberates NRF2 from KEAP1-mediated proteasomal degradation (1,2). This event leads to the nuclear translocation of NRF2 from the cytoplasm, which acts as a potent antioxidant transcription factor regulating the expression of many cytoprotective genes (3). On the other hand, the nuclear NRF2 interferes with a pharmacologic nuclear receptor signaling, including PPARα (4). ARE, antioxidant response element; PPRE, peroxisome proliferator response element; DR1, direct repeat 1. This schematic diagram was created in BioRender.com (accessed on 6 January 2022).

Data Availability Statement:
The data presented in this study are available on request from the corresponding author.