Elafibranor Inhibits Chronic Kidney Disease Progression in NASH Mice

Identification of new pharmacological approaches to inhibit the excessive fat intake-induced steatohepatitis and chronic kidney disease (CKD) is important. High-fat diet (HFD)-induced steatohepatitis and CKD share common pathogenesis involving peroxisome proliferator-activated receptor (PPAR)-α and -δ. Elafibranor, a dual PPARα/δ agonist, can ameliorate the HFD-induced steatohepatitis. Nonetheless, the effects of HFD-induced CKD had not yet explored. This study investigated the effects of elafibranor (elaf) on the progression of HFD-induced CKD in mice. In vivo and in vitro renal effects were evaluated in HFD-elaf mice receiving 12 weeks of elafibranor (from 13th to 24th week of HFD feeding) treatment. In elafibranor-treated HFD mice, increased insulin sensitivity, reduced obesity and body fat mass, decreased severity of steatohepatitis, increased renal expression of PPARα, PPARδ, SIRT1, and autophagy (Beclin-1 and LC3-II) as well as glomerular/renal tubular barrier markers [synaptopodin (podocyte marker), zona occludin-1, and cubulin], reduced renal oxidative stress and caspase-3, and less urinary 8-isoprostanes excretion were observed. Aforementioned benefits of elafibranor were associated with low renal tubular injury and tubulointerstitial fibrosis scores, less albuminuria, low urinary albumin-to-creatinine ratio, and preserved glomerular filtration rate. Acute incubation of podocytes and HK-2 cells with elafibranor or recombinant SIRT1 reversed the HFD-sera-induced oxidative stress, autophagy dysfunction, cell apoptosis, barrier marker loss, albumin endocytosis, and reuptake reduction. Besides hepatoprotective and metabolic beneficial effects, current study showed that elafibranor inhibited the progression of HFD-induced CKD through activation of renal PPARα, PPARδ, SIRT1, autophagy, reduction of oxidative stress, and apoptosis in mice with steatohepatitis.


Introduction
High-fat diet (HFD) intake and obesity have been associated with onset and progression of steatohepatitis and chronic kidney disease (CKD) [1][2][3][4]. In obesity, hemodynamic and morphological changes, together with other factors such as systemic inflammation, oxidative stress, and metabolic dyshomeostasis, may result in steatohepatitis and CKD and 2 BioMed Research International ultimately lead to cirrhosis and ESRD. Obesity was the second most highly predictive factor to predict end-stage renal disease (ESRD), even independent of diabetes and hypertension [4,5].
The perioxisome proliferator-activated receptors (PPAR)-훼 and PPAR훿 are crucial for the regulation of inflammation, oxidative stress, and metabolic dyshomeostasis in obese individuals with steatohepatitis [8,9]. Elafibranor is a novel dual PPAR훼/훿 agonist to reduce impaired metabolism, inflammation, and fibrosis in obese patients and animals with steatohepatitis [10,11].
Accordingly, this study evaluated the PPAR훼/훿-activated SIRT1-mediated molecular mechanism and effects of chronic elafibranor treatment on the progression of CKD in HFD-fed obese mice with steatohepatitis.

Materials and Methods
Additional information was included in the supplementary materials and methods (available here) . . Animals. Male C57BL/6 mice (Jackson Laboratories, Bar Harbor, ME), 8 weeks old, were housed in temperature-and humidity-controlled rooms, kept on 12 h light/dark cycle, and provided unrestricted amounts of food and water. This study was approved by the Animal Experiments Committee of Yang-Ming University. Mice were provided with normal chow (NC, Laboratory Autoclavable Rodent Diet 5010) as NC group or a high-fat-diet (HFD, 60% kcal in fat) as HFD groups. In concordance with previous reports [20,21], CKD including albuminuria (renal damage) and decreased creatinine clearance (poor renal function) were observed in HFD-24w mice [mice feeding with 24 weeks of HFD] in preliminary experiments (n=3).
The groups (Figure 1(a)) of C57BL/6 mice included the following: NC-w group (n=5)/HFDw group (n=8) continuously fed NC/HFD for 24 weeks were administered vehicle for 12 weeks from 13 th to 24 th week of NC/HFD feeding; HFD-elaf group (n=8) continuously fed HFD for 24 weeks were administered by oral gavage elafibranor (3mg/kg/day) for 12 weeks from 13 th to 24 th week of HFD feeding, at which time steatohepatitis, albuminuria, and a decrease in GFR developed. This dose of elafibranor (3mg/kg/day) has been demonstrated previously to decrease the progression of steatohepatitis [10,12].
. . Blood Pressure, Metabolic Demands, and Renal Function. The mouse was placed in a metabolic cage and had free access to diet and water. Then, 24-hour urine sample was collected at 3 consecutive days, and the average of 3-day urine was calculated. The supernatant of centrifuged 3day urine samples was used for various analyses, including albumin, creatinine, and 8-isoprostane [marker of oxidative stress]. Urinary albumin-to-creatinine ratio (ACR) was calculated as ACR = urine albumin/urine creatinine (휇g/mg). One day later, mouse GFR was measured after a single injection of FITC-inulin. The GFR was calculated using a two-compartment model of two-phase exponential decay. All above measurements were undergone in Taiwan Mouse Clinic (National Phenotyping and Drug Testing Center) on week 24 of the feeding regimen for HFD-24w/HFD-elaf/NC-24w groups.
Mouse GFR was measured by single injection of FITCinulin clearance as described previously, modified to minimize plasma volume. The GFR was calculated using a twocompartment model of two-phase exponential decay. Briefly, dialyzed FITC-inulin (3.74 휇l/g body wt, Sigma-Aldrich, Inc., St. Louis, MO) was injected retro-orbitally under light anesthesia induced using isoflurane (Baxter Pharmaceutical Products, Deerfield, IL). The anesthesia lasted ∼20s. Approximately 20휇l of blood was collected via the saphenous vein at 10, 35, 55, and 75 min after injection of FITC-inulin for the determination of FITC concentration.
. . Basal Measurements. Glucose tolerance test (GTT) was performed after overnight (16 h) fasting by intraperitoneal injection of D-glucose (2 mg/g body weight, Sigma-Aldrich, Inc., St. Louis, MO). Blood glucose was measured at 0, 30, 90, and 120 minutes using blood obtained by tail nicking using a One Touch glucometer (One Touch Ultra2, Life Scan, Johson&Johson, USA). All animals continued their initial feeding regimen until scarification. Under anesthesia, 2 days after stabilization and overnight fasting, heparinizedblood (from the inferior-vena-cava, abdominal-aorta, and heart-chamber) and the liver/kidney were collected and weighted.
. . Serum and Tissue Metabolic and Inflammatory Profiles. Serum biochemistry data, triglyceride, insulin, TNF훼, and IL-6, caspase-3/7 activity as well as renal IL-6/TNF-훼 levels, myeloperoxidase (MPO) and SIRT1/caspase 3 activities, and hepatic SIRT1 activity were measured.  . . Histologic Analysis. Nonalcoholic fatty liver disease activity score (NAS) was measured by H-E-stained liver section. The H-E and periodic acid-Schiff (PAS)-stained renal section was evaluated to score the renal tubular damage and tubulointerstitial fibrosis. With an ApopTag Peroxidase In Situ Apoptosis Detection Kit (Chemicon, CA, USA), glomerular and tubules cells undergoing apoptosis were calculated.

. . Renal Electron Microscopic and Immunofluorescence
Analysis. Then, the isolated membranes and autophagosomes on the ultrathin section in the proximal renal tubule of kidney were calculated by electron microscopy at 1,200x magnification. Meanwhile, each slide was evaluated for the numbers of cubulin/synaptopodin (+) cells per 1mm 2 in the FITC images.  Table 1) were measured in podocytes/HK-2 cell lysates, glomerular and tubular fractions of renal homogenates.

. . Roles of SIRT -Autophagy on Elafibranor-Related Effects on HFD-Sera-Pretreated
Podocytes and HK-Cells. HFD/ NC-sera were obtained from NC-24w and HFD-24w mice. To mimic the impacts of circulating factors of HFD mice on abnormalities of renal microenvironment, various measurements were undertaken in 10% HFD-sera-pretreated podocytes/HK-2 cells. Significantly, 10% HFD-sera incubation suppressed the SIRT1 activity in cell lysates of podocytes/HK-2 cells. A preliminary dose-finding experiment revealed that, among different concentrations (5, 10, 15, and 30휇M) of elafibranor, maximal stimulation of SIRT1 activity on HFD-sera-pretreated cells was noted at 15휇M of elafibranor. Meanwhile, siSIRT1 was transfected into cells and maximal blockade of elafibranor-activated SIRT1 activity was noted at 100휇M.
Meanwhile, we found that rSIRT1 (300휇M) had similar effects as elafibranor (15휇M) to reverse HFD-sera-suppressed SIRT1 activity in cells. To examine the SIRT1-mediated effects of elafibranor on autophagy, HFD-sera-pretreated cells were

. . Albumin Endocytosis or Albumin Reuptake of HFD-Sera-Pretreated Podocytes and HK-Cells.
For albumin endocytosis and albumin reuptake experiments, pretreated podocytes/HK-2 (1x10 5 cells) cells were incubated with 1.5 mg/ml human FITC-albumin (MP Biomedicals, Santa Ana, CA) in Ringer solution at either 4 ∘ C or 37 ∘ C for 1 hour. Afterward, for spectrofluorometric measurements, podocytes/HK-2 were lysed in 20 mM MOPS with 0.1% Triton X-100. FITC-fluorescence was measured using an excitation wavelength of 490 nm and an emission wavelength of 540 nm by a fluorescence plate reader (Synergy HT; Biotek Instruments, Winooski, VT). The amount of protein in the lysates was measured using the bicinchoninic acid (BCA) assay (Pierce, Rockford, IL) and the amount of cell associated FITC-albumin was expressed as FITC-albumin (mg/mL) to compare degree of albumin endocytosis/reuptake between groups.

. . SIRT -Autophagy Protein and mRNA Levels in Cultured
Podocyte and HK-Cells. Notably,same protocol in albumin endocytosis (podocyte) and reuptake (HK-2) experiments was used to prepare cells for this part. After supernatants were collected for 8-isoprostane, caspase 3/7 activity was measured by the ELISA and luminescent substrate assay (Caspase-Glo assay; promega). Proteins and mRNAs were extracted, and cell lysates were used for SIRT1 activities measurement by Biomol SIRT1 fluorescence assay kit (AK-555; Biomol, Farmingdale, NY, USA).
Specifically, for calculation of autophagy flux index, cells were treated with 125nM of Bafilomycin A1 [inhibitors for maturation step of autophagosome including lysosomal enzyme activity and fusion of autophagosomes with lysosomes], in DMSO 2 hours before harvest to obtain cell lysates for measurement of LC3-II protein expression. Then, autophagy flux index was calculated by the formula [autophagy flux index=LC3-II (indicator of autophagosomes formation) expression levels with Bafilomycin A1 (100ng/mL)/LC3-II expression levels without Bafilomycin-A1]. LC3-II expression was normalized by its GADPH expression level.

. . Apoptotic and Barrier Markers (+) Podocytes and HK-
Cells. Cells were stained with Annexin V and 7-aminoactinomycin D (7-AAD; BioLegend, San Diego, CA) and analyzed by flow cytometry with FCSCanto II (BD Biosciences, Mississauga, ON). For immunofluorescence (IF) staining, cells were fixed in paraformaldehyde followed by permeabilization with 0.025% digitonin in PBS. After washing, the cells were subsequently incubated at RT with synaptopodin/cubulin/LC3-II antibodies, FITC-conjugated secondary antibody. After washing with PBS, optical section data for % of synaptopodin/cubulin/LC3-II (+) area on each slide were evaluated.
. . Statistical Analysis. Results are presented as means±SD. Data were analyzed by ANOVA and Student-Newman-Keuls tests for multiple comparisons or by Student's 푡-test for unpaired data between two groups. Statistical significance was accepted at the P < 0.05 level.

Results
. . Chronic Elafibranor Treatments Improve Metabolic Profiles in HFD Mice. In comparison with NC-24w group, 24 weeks of HFD feeding induced hyperglycemia, hyperinsulinemia, abnormal GTT, higher homeostasis model assessment-insulin-resistance (HOMA-IR) index, more food consumption, greater incremental trend of body weight, higher serum/hepatic triglyceride level, and higher whole body fat mass were observed in HFD-24w group (Figures 1, 2(b), and 2(e), Table 2). Nonetheless, water consumption   and metabolic demands [respiratory quotient (average whole body CO 2 production/O 2 consumption) and energy expenditure] were not different between NC group and HFD-group (Figures 2(a), 2(c), and 2(d)).
In HFD group, the beneficial effects of elafibranor with respect to hepatic steatosis, whole body fat mass, and GTT were counteracted by EX527 (a specific SIRT1 inhibitor), but not in body weight changes (Figures 1(b)-1(d) and 2(e), Table 2).,
Although no difference in water consumption and urine output was observed, significantly, increased kidney weight, renal MPO activity, and urinary 8-isoprostanes excretion were observed in HFD-group with CKD compared with that in the NC group (Table 2, Figures 2(a), 5(d), and 5(e)).

. . Elafibranor Normalizes Hepatic and Renal SIRT Expression in HFD Mice with Steatohepatitis and CKD.
In the HFD group, a general reduction in liver, adipose tissue, and renal PPAR훼 and PPAR훿 expression than in the NC group was observed (Figure 3(c)). In our study, the expression of other energy and nutrient sensors (AMPK훼1/2 and SIRT3) in the liver, small intestine, adipose tissue, and kidney was not different between the HFD-group and the NC-group (Figures 3(d)-3(f)).
A similar decreasing trend of PPAR훼, PPAR훿, and SIRT1 expression in the liver and kidney was observed in HFD mice with steatohepatitis and CKD (Figures 3(c), 3(e), 3(f), and 4(a)). Nonetheless, the expression of SIRT1 in the small intestine and adipose tissue was not different between the NC group and the HFD group. Remarkably, simultaneous activation of hepatic and renal PPAR훼 and PPAR훿 by preventive or therapeutic elafibranor treatment restored hepatic and renal SIRT1 expression in the HFD group. Elafibranor-related decrease in severity of steatohepatitis was accompanied by an improvement of CKD (Figures 1(c), 3(b), 5(a), and 5(b)). In the electron microscopic images of renal section, a decrease in double membrane structure and autophagosome was observed in the proximal renal tubule of kidney of HFD mice compared to the NC group, which was increased after elafibranor treatment (Figures 4(b) and 4(c)).
. . Elafibranor Improves Inflammatory and Apoptotic Profiles in HFD Mice with Steatohepatitis and CKD. In HFD mice, the high serum/renal TNF훼 levels, serum caspase 3/7, renal Tunnel stain-assessed apoptotic activity, serum IL-6, serum AST, and ALT were significantly suppressed by elafibranor treatment, and the effect was reversed by EX-527 (Table 2 and Figures 5(f) and 5(g)). Nonetheless, renal IL-6 levels were not different between HFD mice with and without elafibranor treatment.

. . Renal SIRT Activation by Elafibranor Is Accompanied by Normalization of Renal Barrier Markers in HFD Mice.
Decreased renal p-SIRT1 expressions were accompanied by the reduction of the expression of glomerular [synaptopodin (marker of podocyte)/ZO-1] and tubular [cubulin] barrier markers in the HFD-group (Figures 4(a), and 6(a)-6(d)).
In cultured HK-2 cells monolayer, the effects of abovementioned treatment were similar to those in cultured podocytes. In particular, the changes in the renal tubular barrier marker, cubulin, were similar as changes in glomerular barrier markers (synaptopodin/ZO-1) in podocytes (Table 3 and Figure 8).

Discussion
Chronic elafibranor treatment inhibits the progression of HFD-induced CKD in mice in this study. Reduction in GFR and increasing albuminuria are initial markers for the detection of the progression to ESRD [22]. In this study, the renoprotective effects of chronic elafibranor treatment, including preserved GFR and decreased albuminuria, were observed in HFD-induced obese mice with steatohepatitis.
In PPAR훼 knockout and diabetic mice, heavy albuminuria is associated with significant renal inflammation, apoptosis, and fibrosis [13,23]. In renal tubular cells, PPAR훼 activation protects cells from gentamicin-induced oxidative stress and apoptosis [24]. Both PPAR훼 and PPAR훿 are highly expressed in kidney [11,13,24]. Activation of PPAR훿 ameliorates tubulointerstitial inflammation in mice with proteinuric kidney disease [25]. PPAR훿 activation protects cardiomyoblasts from oxidative stress-induced apoptosis [26]. Accordingly, it is reasonable that elafibranor, through renal PPAR훼 and PPAR훿 activation, improves CKD through inhibition of renal oxidative stress, inflammation, fibrosis, and apoptosis in HFD-induced obese mice with steatohepatitis in our study.
In fact, the renoprotective effects of 12 weeks of PPAR훼 agonist treatment had been reported in HFD-fed obese mice with CKD [27]. In our study, the effects and mechanisms of renoprotective effects of 12 weeks of elafibranor treatment were explored in HFD-fed NASH mice with CKD. Nonetheless, it is mandatory to explore the shortest period of renoprotective effects of elafibranor which need to be evaluated in future studies. Antiapoptosis and antioxidative stress effects of SIRT1 are accompanied by its anti-inflammatory effects in animals with CKD and steatohepatitis [3, 5, 7, 14-19, 21, 28]. Hepatic and renal SIRT1 were reduced in animals with steatohepatitis and CKD [14,15]. In our study, concomitant SIRT1 inhibitor (EX527) treatment reversed elafibranor-related benefits, indicating that these effects are SIRT1-dependent. Primarily, antioxidative stress effect is responsible for the activation of renal SIRT1 in HFD-fed mice treated with elafibranor.
By increasing peroxisome function, SIRT1 activation, which is reciprocally stimulated by upregulated PPAR훼 and PPAR훿, can prevent the drug-induced renal cell apoptosis and acute kidney injury in mice [29][30][31][32][33]. Decreased SIRT1 expression on podocyte increases cell apoptosis and albuminuria in mice [16,32]. In our study, siSIRT1 coincubation eliminates PPAR훼/훿 agonist elafibranor-related suppression of HFD-sera-induced apoptosis. Meanwhile, rSIRT1 coincubation has similar effects as elafibranor on the reversal of HFD-sera-induced apoptosis. So, in current study, the renoprotective effect of the PPAR훼/훿 agonist elafibranor, at least partly, is attributable to SIRT1-mediated inhibition of HFD-induced circulating factors on renal cells and the kidney.
Autophagy promotes cell survival by elimination of damage organelles, which is initiated by increased Beclin-1/LC3-II levels, resulting in increased autophagic flux. Suppression of autophagy (reduced Beclin-1/LC3-II) and autophagy flux induce cell apoptosis. PPAR훼 activation protects the liver from acute inflammation and failure by activating autophagy [33]. PPAR훿 activation protects human cardiac cells from ER stress-induced injury by stimulating autophagy [34]. Inhibition SIRT1 exacerbates oxidative stress-suppressed autophagy  in stem cells [35]. Downregulation of SIRT1 signals is involved in the HFD-induced renal dysfunction in mice [36]. SIRT1 activator suppresses hyperglycemia-induced apoptosis of podocytes via autophagy activation in diabetic mice with nephropathy [37]. In in vitro experiments, we revealed that the coincubation with autophagy flux blocker (BAF) abolished elafibranor-related SIRT1-mediated inhibition of NASH-sera-induced pathogenic signals in podocytes and HK-2 cells. Impaired glomerular protein endocytosis and reduced tubular reuptake of leakage protein can lead to albuminuria. Podocytes are epithelial cells of the outer membrane of renal glomeruli that maintain its integrity. Podocytes damage not only impairs glomerular barrier but also collapses its architecture and leads to advanced renal injury and albuminuria. Restoration of glomerular barrier protein expressions including ZO-1 and synaptopodin (marker of podocyte) avoids albuminuria in diabetic mice [38]. Proximal tubule cells have a capacity to uptake glomeruli-leaked albumin and prevent final leakage. In proximal renal tubules, cubulin mediates the reuptake of leakage albumin from glomeruli to avoid tubulointerstitial inflammation/fibrosis [39]. Oxidative stress-related downregulation of barrier markers worsens albuminuria and tubulointerstitial inflammation/fibrosis [40,41]. Our study revealed that CKD-related oxidative stress and albuminuria were associated with the downregulation of renal barrier markers in HFD-induced obese mice with steatohepatitis.
In summary, through systemic in vivo and in vitro approaches, our study revealed that kidney-specific protective effects of elafibranor are attributable to the preservation of glomerular/tubular barrier protein, maintenance of structure, antioxidative stress, and antiapoptosis effects via activation of SIRT-autophagy-mediated protective signals ( Figure 9).
This study suggested that elafibranor and strategies aimed at activating SIRT-autophagy are promise for treating high fat consumption which induces steatohepatitis and CKD.

Data Availability
All data supporting the results reported in the article can be found in Division of General Medicine, Department of Medicine, Taipei Veterans General Hospital, Taipei, Taiwan, and can be seen after asking the corresponding author. various barrier and autophagic markers. Additionally, the roles of SIRT1-autophagy on elafibranor-related effects were evaluated in HFD-sera-pretreated podocytes and HK-2 cells system. The additional detail materials and methods are included in supplementary materials and methods. (Supplementary Materials)