If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
1 Huan Wang and Yushang Zhao contributed equally to this work.
Huan Wang
Footnotes
1 Huan Wang and Yushang Zhao contributed equally to this work.
Affiliations
Immunology Research Center for Oral and Systemic Health, Beijing Friendship Hospital, Capital Medical University, Beijing, China, 100050Beijing Key Laboratory of Tolerance Induction and Organ Protection in Transplantation, Beijing, China, 100050Beijing Clinical Research Institute, Beijing, China, 100050
1 Huan Wang and Yushang Zhao contributed equally to this work.
Yushang Zhao
Footnotes
1 Huan Wang and Yushang Zhao contributed equally to this work.
Affiliations
Immunology Research Center for Oral and Systemic Health, Beijing Friendship Hospital, Capital Medical University, Beijing, China, 100050Beijing Key Laboratory of Tolerance Induction and Organ Protection in Transplantation, Beijing, China, 100050Beijing Clinical Research Institute, Beijing, China, 100050National Clinical Research Center for Digestive Diseases, Beijing, China, 100050Beijing Laboratory of Oral Health, Capital Medical University School of Basic Medicine, Beijing, China, 100069
Immunology Research Center for Oral and Systemic Health, Beijing Friendship Hospital, Capital Medical University, Beijing, China, 100050Beijing Key Laboratory of Tolerance Induction and Organ Protection in Transplantation, Beijing, China, 100050Beijing Clinical Research Institute, Beijing, China, 100050National Clinical Research Center for Digestive Diseases, Beijing, China, 100050Beijing Laboratory of Oral Health, Capital Medical University School of Basic Medicine, Beijing, China, 100069
Immunology Research Center for Oral and Systemic Health, Beijing Friendship Hospital, Capital Medical University, Beijing, China, 100050Beijing Key Laboratory of Tolerance Induction and Organ Protection in Transplantation, Beijing, China, 100050Beijing Clinical Research Institute, Beijing, China, 100050National Clinical Research Center for Digestive Diseases, Beijing, China, 100050Beijing Laboratory of Oral Health, Capital Medical University School of Basic Medicine, Beijing, China, 100069
2 Lead contact. Dong Zhang, M.D., Capital Medical University Affiliated Beijing Friendship Hospital, 95 Yongan Road, Xicheng District, Beijing, China, 100050. [email protected] or. [email protected]
Affiliations
Immunology Research Center for Oral and Systemic Health, Beijing Friendship Hospital, Capital Medical University, Beijing, China, 100050Beijing Key Laboratory of Tolerance Induction and Organ Protection in Transplantation, Beijing, China, 100050Beijing Clinical Research Institute, Beijing, China, 100050National Clinical Research Center for Digestive Diseases, Beijing, China, 100050Beijing Laboratory of Oral Health, Capital Medical University School of Basic Medicine, Beijing, China, 100069Beijing Laboratory of Oral Health, Capital Medical University School of Basic Medicine, Beijing, China, 100069General Surgery Department, Beijing Friendship Hospital, Capital Medical University, 100050
Immunology Research Center for Oral and Systemic Health, Beijing Friendship Hospital, Capital Medical University, Beijing, China, 100050Beijing Key Laboratory of Tolerance Induction and Organ Protection in Transplantation, Beijing, China, 100050Beijing Clinical Research Institute, Beijing, China, 100050National Clinical Research Center for Digestive Diseases, Beijing, China, 100050Beijing Laboratory of Oral Health, Capital Medical University School of Basic Medicine, Beijing, China, 100069Beijing Laboratory of Oral Health, Capital Medical University School of Basic Medicine, Beijing, China, 100069
1 Huan Wang and Yushang Zhao contributed equally to this work. 2 Lead contact. Dong Zhang, M.D., Capital Medical University Affiliated Beijing Friendship Hospital, 95 Yongan Road, Xicheng District, Beijing, China, 100050. [email protected] or. [email protected]
PLD1 is highly expressed in hepatocytes of NAFLD patients and HFD-fed mice.
●
Hepatocyte-specific deficiency of Pld1 ameliorates hepatic steatosis.
●
PLD1 promotes CD36 expression and alter lipid composition in hepatocytes.
●
PA, the downstream product of PLD1, upregulates CD36 expression via PPARγ.
Abstract
Background & aims
Phospholipase D1 (PLD1), a phosphatidylcholine-hydrolyzing enzyme, is involved in cellular lipid metabolism. However, its involvement in hepatocyte lipid metabolism and consequently hepatic steatosis has not been explicitly explored.
Methods
Hepatic steatosis was induced in hepatocyte-specific Pld1 knockout (Pld1(H)-KO) and control (Pld1-Flox) mice feeding a high-fat diet (HFD) for 20 weeks. Changes of the lipid composition in the liver were compared. Alpha mouse liver 12 (AML12) cells and mouse primary hepatocytes were incubated with oleic acid (OA) or sodium palmitate (SP) in vitro to explore the role of PLD1 in the development of hepatic steatosis. Hepatic PLD1 expression was evaluated in liver biopsy samples in patients with NAFLD.
Results
PLD1 expression levels were increased in the hepatocytes of patients with NAFLD and HFD-fed mice. Compared with Pld1-Flox mice, Pld1(H)-KO mice exhibited decreased plasma glucose and lipid levels as well as lipid accumulation in liver tissues after HFD feeding. Transcriptomic analysis showed that hepatocyte-specific deficiency of PLD1 decreased Cd36 expression in steatosis liver tissues, which was confirmed at the protein and gene levels. In vitro, specific inhibition of PLD1 with VU0155069 or VU0359595 decreased CD36 expression and lipid accumulation in OA- or SP- treated AML12 cells or primary hepatocytes. Inhibition of hepatocyte PLD1 significantly altered lipid composition, especially phosphatidic acid (PA) and lyso-PA (LPA) levels in liver tissues with hepatic steatosis. Furthermore, PA, the downstream product of PLD1, increased the expression levels of CD36 in AML12 cells, which was reversed by a PPARγ antagonist.
Conclusions
Hepatocyte-specific Pld1 deficiency ameliorates lipid accumulation and hepatic steatosis development by inhibiting the PPARγ/CD36 pathway. PLD1 may be a new target for the treatment of hepatic steatosis.
Impact and implications
The involvement of PLD1 in hepatocyte lipid metabolism and hepatic steatosis has not been explicitly explored. In this study, we found the inhibition of hepatocytes PLD1 exerted potent protective effects against HFD-induced hepatic steatosis, which were due to a reduction in PPARγ/CD36 pathway-mediated lipid accumulation in hepatocytes. Targeting hepatocytes PLD1 may be a new target for the treatment of hepatic steatosis.
The authors confirm that the data supporting the findings of this study are available within the article and/or its Supplementary Materials and methods. Any additional data are available from the corresponding authors upon reasonable request. The data reported in this work have been uploaded in the Gene Expression Omnibus (GEO) database under accession number GSE207281. The following secure token has been created to allow review of record GSE207281 while it remains in private status: qjofseiahjqztmh.
Competing interests
All authors declare no conflict of interest.
Financial support
Grants from the National Natural Science Foundation of China (No. 82270606, 81900784, 81970503 and 82070580), R&D Program of Beijing Municipal Education Commission (KZ202210025036), Natural Science Foundation of Beijing Municipality (7204248), Beijing Municipal Administration of Hospitals’ Ascent Plan (DFL20220103) and the Youth Beijing Scholar (No. 035) supported this work.
Authors' contributions
All listed authors participated meaningfully in the study and that they have seen and approved the submission of this manuscript. H.W. and Y.Z. participated in performing the research, analyzing the data, and initiating the original draft of the article. Y.P., A.Y., C.L., S.W., Z.D and M.L. participated in performing the research. D.Z. and G.S. established the hypotheses, supervised the studies, analyzed the data, and co-wrote the manuscript. Y.Z. participated in establishing the hypotheses and provided the conditional knockout mice. S.W. and Z.Z. participated in the review and editing of the manuscript.
1. Introduction
Non-alcoholic fatty liver disease (NAFLD) has become the leading chronic liver disease globally; however, no effective treatment has been approved [
]. Lipid droplets (LDs) accumulation in hepatocytes is a distinctive characteristic of NAFLD, a chronic, heterogeneous liver condition that can progress to liver fibrosis and hepatocellular carcinoma [
]. This lipid accumulation may occur due to increased de novo lipogenesis, increased fatty acid uptake, decreased redistribution of fatty acids to other tissues, or decreased utilization of lipids as energy substrates [
]. In mammalian cells, the PLD isoenzymes-PLD1 and PLD2-are major sources of phosphatidic acid (PA), a lipid second messenger that modulates diverse intracellular signaling [
]. PLD is involved in cell proliferation, inflammation, survival, redox signaling, mitochondrial function, and many pathophysiological actions; it has also been associated with neuronal ailments, cancer, thrombotic events, and infectious diseases [
A phospholipase D-dependent process forms lipid droplets containing caveolin, adipocyte differentiation-related protein, and vimentin in a cell-free system.
]. Further studies are needed to clarify the role of hepatic PLD1 in metabolic processes.
In this study, inducing hepatic steatosis in hepatocyte-specific Pld1-deficient mice using a high-fat diet (HFD), we provide evidence that PLD1 plays a key role in promoting hepatocyte lipid accumulation and steatosis development via the PPARγ/CD36 pathway. Furthermore, PA may be the main factor involved in the activation of the PPARγ/CD36 pathway during hepatocyte lipid accumulation.
2. Materials and Methods
Animals and experiment protocol
Pld1flox/flox mice were generated at the Shanghai Model Organisms Center, INC (China). To generate hepatocyte-specific Pld1 knockout (Pld1flox/flox; Alb-Cre+; Pld1(H)-KO) mice, Pld1flox/flox mice were crossed with Alb-Cre mice (Jackson Laboratory, 003574, USA). Littermate Pld1-Flox (Pld1flox/flox; Alb-Cre-) mice were used as controls. All animals were maintained in a pathogen-free, temperature-controlled environment under a 12-hour light/dark cycle at Beijing Friendship Hospital, and all animal protocols were approved by the Institutional Animal Care and Ethics Committee. Male Pld1-Flox and Pld1(H)-KO mice at 8-9 weeks were fed a normal control diet (NCD) or an HFD (60 kcal% fat; D12492, Research Diets, USA).
Clinical study
Twelve patients diagnosed with NAFLD by liver biopsy at Beijing Friendship Hospital, Capital Medical University. Patients with NAFLD who consumed < 10 g alcohol/day for women or 20 g alcohol/day for men and those with other causes of steatosis or chronic liver disease were excluded. The demographic and clinical characteristics of the NAFLD patients were showed in Table S2. For each biopsy, a NAS score summarizing the main histological lesions was defined based on the grade of steatosis, the grade of activity (hepatocytes ballooning and lobular inflammation) and the stage of fibrosis (Table S2) [
]. Healthy liver tissues were obtained from eight donors, whose livers were subsequently used for transplantation. Written informed consent was obtained from all patients before enrollment, and the study protocol was approved by the human institutional review board of the Beijing Friendship Hospital (No. 2017-P2-131-03).
Statistical analysis
Statistical analysis was performed using GraphPad Prism software (version 9.0), and values were expressed as mean ± SEM. Differences between two groups were compared by Student’s t test for normal distribution and Mann-Whitney test for abnormal distribution. One-way ANOVA with Tukey’s test for normal distribution and Kruskal-Wallis test for abnormal distribution were used for analyzing differences among three or more groups. Statistical significance was set at P < 0.05.
For further details regarding the materials and methods used, please refer to the CTAT table and supplementary materials and methods.
3. Results
3.1 Hepatocyte-specific deficiency of Pld1 ameliorated NAFLD in HFD-fed mice
After 20 weeks of HFD feeding, hepatocellular PLD activity was significantly upregulated compared to that in the NCD-fed control mice (Fig. 1A). The protein and mRNA expression levels of PLD1 were markedly increased in the hepatocytes obtained from HFD-fed mice than in those obtained from NCD-fed mice (Fig. 1B-D). However, the expression level of PLD2 did not change after HFD feeding (Fig. 1B, C, E). We further tested PLD1 expression in hepatocytes of patients with NAFLD and in those of healthy controls by immunofluorescence (Fig. 1F, G) and found the proportion of ALB+PLD1+ area to be higher in patients with NAFLD than in healthy controls. These findings suggest that hepatocellular PLD1 is involved in the development and progression of NAFLD in both mice and humans.
Fig. 1PLD1 expression was increased in the hepatocytes of HFD mice and patients with NAFLD. (A) PLD activity, (B) representative western blot images showing PLD1, PLD2, and GAPDH levels, (C) quantification of PLD1 and PLD2 by densitometry and gene expression levels of (D) Pld1 and (E) Pld2 in the hepatocytes of Pld1-Flox mice after 20 weeks of NCD or HFD feeding. (F) Representative immunofluorescence staining images in the liver of patients with NAFLD and healthy controls, ALB in red (546), PLD1 in green (488) and nuclei in blue (DAPI) (scale bar, 20 μm). (G) Quantification of ALB+PLD1+ area. N=8-12 subjects per group. Differences between two groups were compared by Student’s t test for normal distribution and Mann-Whitney test for abnormal distribution. **p < 0.01, ***p < 0.001. PLD, phospholipase D; NCD, normal control diet; HFD, high-fat diet; ALB, albumin.
To investigate the role of hepatocellular PLD1 in NAFLD development or progression, HFD was fed to Pld1(H)-KO and Pld1-Flox mice for 20 weeks. The protein and mRNA expression levels of PLD1 decreased in Pld1-deficient hepatocytes in both HFD- and NCD-fed mice (Figs. S1A–C). PLD1 deficiency did not alter the expression of PLD2 in hepatocytes after HFD feeding (Figs. S1D–F). As shown in Fig. 2A-C, Pld1(H)-KO mice gained remarkably less body weight, liver size, and liver weight than age-matched Pld1-Flox mice, although food intake among each group was not significantly different after HFD feeding (Fig. S1G). HFD-fed mice had insulin resistance with impaired OGTT compared with NCD-fed mice; these were improved by hepatocyte Pld1 knockout (Fig. 2D, E). Plasma TG and FFA levels did not differ between NCD- and HFD-fed mice (Fig. 2F). Plasma levels of TC, LDL-C, GLU, ALT and AST in mice with hepatocyte-specific deficiency of Pld1, were significantly reduced compared with those in Pld1-Flox mice after HFD-feeding (Fig. 2F-I). Hepatocyte-specific Pld1 deficiency significantly decreased liver steatosis score, lobular inflammation score, and the NAS in HFD-fed Pld1(H)-KO mice compared to HFD-fed Pld1-Flox mice (Fig. 2J, K, S1H-J). Oil Red O staining revealed decreased lipid accumulation in the livers of Pld1(H)-KO mice compared to those of Pld1-Flox mice after HFD feeding (Fig. 2J, L). BODIPY staining showed that LDs were increased in the livers of HFD-fed mice and decreased in hepatocyte Pld1 knockout mice (Fig. 2M, N). LDs deposition was also found in a portion of Desmin+ hepatic stellate cells (HSCs) in both NCD- and HFD-feeding mice (Fig. S1K). Furthermore, the deficiency of PLD1 could decrease liver-infiltrating CD45+, F4/80+ and CD3+ cells and the expression levels of inflammation related gene, such as Tnfa, Il1b, Il6 and Il17 after HFD feeding (Fig. 2J, O, P). Meanwhile, the levels of fibrosis-related genes, such as α-SMA, Col1a1, Col3a1 and Col6a1 and hydroxyproline were also downregulated in the liver tissues after PLD1 knockout (Fig. 2Q, R).
Fig. 2Hepatocyte-specific Pld1 deficiency alleviated NAFLD in mice. (A) Body weight was measured weekly in Pld1-Flox and Pld1(H)-KO mice after NCD or HFD feeding. (B) Representative photos of the liver, (C) liver weight, (D) oral glucose tolerance test (OGTT), (E) area under curve (AUC) of the OGTT and plasma (F) triglyceride (TG), free fatty acid (FFA), total cholesterol (TC), low density lipoprotein-cholesterol (LDL-C), (G) glucose (GLU), (H) alanine transaminase (ALT) and (I) aspartate transaminase (AST) levels in Pld1-Flox and Pld1(H)-KO mice after 20 weeks of NCD or HFD feeding. (J) Representative images of hepatic hematoxylin & eosin (H&E) (scale bar, 100 μm), Oil Red O (scale bar, 100 μm), CD45, F4/80 and CD3 staining (scale bar, 50 μm), (K) NAS, and (L) Oil Red O positive area per total area in the liver of Pld1-Flox and Pld1(H)-KO mice after 20 weeks of NCD or HFD feeding. (M) Representative immunofluorescence staining images of lipid droplet in green (BODIPY, 493/503) and nuclei in blue (DAPI) (scale bar, 100 μm), (N) BODIPY fluorescence intensity, (O) CD45+, F4/80+ and CD3+ cell numbers per field, (P) relative mRNA levels of proinflammatory cytokine and (Q) fibrosis-related gene, and (R) hydroxyproline levels in the liver of Pld1-Flox and Pld1(H)-KO mice after 20 weeks of NCD or HFD feeding. N=8-10 mice per group. One-way ANOVA with Tukey’s test for normal distribution and Kruskal-Wallis test for abnormal distribution were used for analyzing differences among three or more groups. *p < 0.05, **p < 0.01, ***p < 0.001. aaap < 0.001, Pld1-Flox-NCD mice vs. Pld1-Flox-HFD mice, bp <0.05, bbp < 0.01, Pld1(H)-KO-HFD mice vs. Pld1-Flox-HFD mice. PLD1, phospholipase D1; NCD, normal control diet; HFD, high-fat diet.
3.2 Hepatocyte-specific deficiency of Pld1 changed hepatic lipid metabolism in HFD-fed mice
Considering the substantial difference in liver Oil Red O and LD staining between HFD-fed Pld1(H)-KO and Pld1-Flox mice. We performed lipidomic analysis to examine lipid metabolism and lipidomic changes in the livers of Pld1(H)-KO and Pld1-Flox mice after HFD feeding. Our lipidomic data revealed 460 lipid species in the liver tissues, consisting of 181 triacylglycerols (TAG), 49 cardiolipins (CLs), 47 phosphatidylcholines (PC), 25 diacylglycerols (DAG), and other lipid classes (Fig. S2A). We visualized all significantly altered lipid species using a bubble map (Fig. S2B). Using a p-value of 0.05 as cut-off, a total of 207 species were significantly changed in the Pld1-deficient livers (Fig. S2B). The overall abundance of TAG and DAG, which are the most abundant lipid classes in the liver, was significantly decreased by Pld1 deficiency (Fig. S2C). As a key enzyme regulating phospholipid metabolism, hepatocyte Pld1 knockout had a tendency to increase the total levels of PC in the liver (Fig. S2D). The levels of PC34:0p, PC36:3p, PC32:2/0, PC34:2/1/0, PC36:4, and PC38:6 among 47 PC species were significantly increased by hepatocyte Pld1 knockout (Fig. S2E). In addition, there were significant changes in other glycerophospholipids, including a decrease in the abundance of lyso-phosphatidylethanolamines and bis (monoacylglycerol) phosphate and an increase in the abundance of phosphatidylethanolamines (Fig. S2F). Sphingolipids, including sphingomyelins and ceramides, were unaffected by Pld1 deficiency (Fig. S2G). The abundance of free fatty acids and cholesteryl esters was decreased by Pld1 deficiency (Figs. S2H and I). These results suggest that the inhibition of hepatocellular PLD1 affects the composition of lipids in liver tissues.
3.3 Transcriptome sequencing analysis showed that the expression of fatty acid translocase CD36 changed significantly in Pld1(H)-KO mice
To explore the impact of hepatocyte-specific Pld1 deficiency, mRNA transcriptome sequencing was performed in the liver tissues of Pld1(H)-KO and Pld1-Flox mice after NCD or HFD feeding (Figs. S3A and B). GO pathway analysis revealed that pathways related to lipid uptake, transport, and metabolism were up-regulated and lipase activity pathways were down-regulated in Pld1(H)-KO mice compared with Pld1-flox mice (Fig. S3C). Compared with Pld1-flox mice, pathways related to lipid activity were upregulated in Pld1(H)-KO mice and pathways related to lipid digestion, absorption, metabolism, and import cells were downregulated (Fig. 3A). Gene set expression analysis revealed that intestinal lipid absorption and LD were positively enriched in Pld1-Flox-HFD mice compared with Pld1-Flox-NCD mice (Fig. S3D), whereas cellular responses to fatty acids and LDs were negatively enriched in Pld1(H)-KO-HFD mice compared with Pld1-Flox-HFD mice (Fig. 3B). A total of 193 differential expressed genes were obtained from the intersection of the differential genes in Pld1-Flox-HFD vs. Pld1-Flox-NCD mice and in Pld1(H)-KO-HFD vs. Pld1-Flox-HFD mice. Among these genes, four lipid metabolism-related GO pathways were enriched: the arachidonic acid metabolic process, long-chain fatty acid metabolic process, fatty acid metabolic process, and regulation of plasma lipoprotein particle levels. Among the genes involved in these pathways, 10 were downregulated and five were upregulated in Pld1(H)-KO-HFD mice compared with Pld1-Flox-HFD mice. All genes were sequenced according to -log10 (p-value), and we found that among the genes involved in lipid metabolism, fatty acid translocase Cd36 was in a relatively high position (Fig. 3C). To verify these observations, CD36 expression levels were measured in vivo. The protein and mRNA expression levels of CD36 were increased in the liver tissues of HFD-fed mice and significantly reduced by hepatocyte-specific Pld1 deficiency (Fig. 3D-F). CD36 expression levels were also significantly higher in the liver tissues of patients with NAFLD than in those of healthy controls (Fig. 3G-H).
Fig. 3Hepatocyte-specific Pld1 deficiency induced changes in lipid and gene expression profiles. (A) Upregulated and downregulated GO pathway enriched in differentially expressed genes in Pld1(H)-KO-HFD mice compared with Pld1-Flox-HFD mice. (B) Negative enrichment plot for the gene set in Pld1(H)-KO-HFD mice compared with that of Pld1-Flox-HFD mice. (C) GO enrichment analysis was performed on the genes obtained from the intersection of Pld1-Flox-HFD vs. Pld1-Flox-NCD mice and Pld1(H)-KO-HFD vs. Pld1-Flox-HFD mice (upper left), and the genes involved in enriched pathways related to lipid metabolism were showed according to their expression levels and sorted according to -log10(p) values (left bottom and right). (D) Representative western blot images showing CD36 and GAPDH levels, (E) quantification of CD36 by densitometry and (F) gene expression levels of Cd36 in the livers of Pld1-Flox and Pld1(H)-KO mice after 20 weeks of NCD or HFD feeding. (G) Representative immunofluorescence staining images of CD36 in red (647) and nuclei in blue (DAPI) (scale bar, 20 μm), (H) quantification of CD36 by fluorescence intensity in the liver of patients with NAFLD and those of the livers of healthy controls. N=6-12 subjects per group. Differences between two groups were compared by Student’s t test for normal distribution and Mann-Whitney test for abnormal distribution. One-way ANOVA with Tukey’s test for normal distribution and Kruskal-Wallis test for abnormal distribution were used for analyzing differences among three or more groups. *p < 0.05, **p < 0.01, ***p < 0.001. PLD1, phospholipase D1; NCD, normal control diet; HFD, high-fat diet.
3.4 PLD1 inhibition and deficiency decreased the content of lipid and expression of CD36 in hepatocytes in vitro
To gain insight into PLD1-induced changes in lipid metabolism in hepatocytes, PLD1-specific inhibitors, VU01 and VU03, were used to interfere with the murine hepatocyte cell line AML12. The protein and mRNA expression levels of PLD1 and PLD activity were significantly increased after OA stimulation, and the most obvious increase was observed after 500 μM OA stimulation for 48 hours (Fig. 4A-D). Therefore, AML12 cells were incubated with 500 μM OA for 48 h for subsequent inhibition experiments. VU01 and VU03 reduced the proportion of CD36+ cells, MFI of CD36 and the mRNA levels of Cd36, which were increased by OA treatment (Fig. 4E-G). LDs and lipid accumulation were increased in the OA-treated AML12 cells and decreased by inhibition of PLD1 (Fig. 4H-K). VU01 and VU03 also decreased TG and TC levels in the OA treated AML12 cells (Fig. 4L, M). To determine the key role of CD36 in PLD1-mediated lipid accumulation, CD36 overexpression plasmids were transfected into AML12 cells. The mRNA levels of CD36 and the proportion of CD36+ cells elevated significantly after transfection, and could not be reduced by VU01 (Figs. S4A and B). Oil Red O and Nile Red staining revealed increased lipid accumulation after overexpression of CD36 in OA-treated AML12 cells, which could not be decreased by inhibition of PLD1 (Fig. 4N, O and S4C, D).
Fig. 4Inhibition of PLD1 reduced OA-induced increase in CD36 expression and lipid levels in hepatocytes. (A) Representative western blot images showing PLD1 and GAPDH levels, (B) quantification of PLD1 by densitometry, (C) gene expression levels of Pld1 and (D) PLD activity in OA-treated AML12 cells. (E) The ratio of CD36+ cells, (F) the MFI of CD36, (G) the mRNA expression levels of Cd36 and (H) the ratio of BODIPY+ cells in OA-stimulated AML12 cells with or without PLD1 inhibitor treatment. (I) Representative staining images of lipid droplet in green (BODIPY, 493/503), nuclei in blue (DAPI) and Oil Red O, (J) BODIPY fluorescence intensity and (K) Oil Red O positive area per total cell area, (L) triglyceride (TG) and (M) total cholesterol (TC) levels in the OA-stimulated AML12 cells with or without PLD1 inhibitor. (N) Representative staining images of Oil Red O and (O) Oil Red O positive area per total cell area in OA and VU01 treated AML12 cells after CD36 overexpression. (P) Gene expression levels of Pld1, (Q) representative western blot images showing CD36 and GAPDH levels, (R) quantification of CD36 by densitometry, (S) the mRNA expression levels of Cd36, (T) representative immunofluorescence staining images of lipid droplet in green (BODIPY, 493/503) and nuclei in blue (DAPI), (U) BODIPY fluorescence intensity and (V) the levels of TG and TC in OA treated primary hepatocytes from Pld1-Flox and Pld1(H)-KO mice. N=3-6 dishes of cells per group. Differences between two groups were compared by Student’s t test for normal distribution and Mann-Whitney test for abnormal distribution. One-way ANOVA with Tukey’s test for normal distribution and Kruskal-Wallis test for abnormal distribution were used for analyzing differences among three or more groups. *p < 0.05, **p < 0.01, ***p < 0.001. PLD1, phospholipase D1; VU0155069 (VU01) and VU0359595 (VU03), PLD1 specific inhibitors; NC, normal control; OA, oleic acid.
In addition, primary hepatocytes from Pld1(H)-KO and Pld1-Flox mice were treated with OA to verify the role of PLD1 in lipid accumulation of hepatocytes. The expression levels of Pld1 and CD36 increased after OA treatment in Pld1-Flox hepatocytes, while decreased in Pld1-deficient hepatocytes (Fig. 4P-S). Moreover, deficiency of PLD1 reduced the elevated levels of LD, TG and TC after OA treatment in mouse primary hepatocytes (Fig. 4T-V). These data suggested that PLD1 plays a major role in the regulation of CD36-mediated lipid accumulation in hepatocytes.
Similar results were also found in SP-treated hepatocytes. The protein and mRNA expression levels of PLD1 were significantly increased, and PLD1 activity peaked at 48 h after stimulation with 250 μM SP (Figs. S5A–D). Therefore, AML12 cells were incubated with 250 μM SP for 48 h for subsequent experiments. The proportion of CD36+ cells, MFI of CD36, the mRNA levels of Cd36, LD, TG, and TC, and lipid accumulation were significantly increased in AML12 cell treated with SP, and decreased after incubation with VU01 and VU03 (Fig. S5E-P). SP stimulation increased the proportion of 7AAD-Annexin V+ cells and 7AAD+ Annexin V+ cells were, which was decreased after VU03 treatment (Figs. S5Q–S). Consistently, deficiency of PLD1 significantly decreased the levels of Pld1 mRNA, CD36 protein and mRNA, TG, TC and LD in SP-treated Pld1(H)-KO hepatocytes compared to SP-treated Pld1-Flox hepatocytes (Fig. S5T-Y).
3.5 PA (a PLD1 product) regulates CD36 expression in hepatocytes
As a key enzyme regulating phospholipid metabolism, hepatocyte Pld1 knockout observably decreased the abundance of PA and LPA (Fig. 5A-B). The levels of PA32:2/1/0, PA34:2/1, and PA36:2/1 among 13 PA species and those of LPA16:0 and LPA18:2/1 among five LPA species were significantly decreased by hepatocyte Pld1 knockout (Fig. 5C). Meanwhile, the content of PA and LPA increased in OA-treated AML12 cells, and decreased by inhibition of PLD1 (Fig. 5D, E). The levels of PA32:1, PA34:2/1, PA36:2/1 and PA38:6/4/3 among 13 PA species and those of LPA16:0 and LPA18:1/0 among 3 LPA species were significantly decreased by VU01 and VU03 (Fig. 5F). SP treatment also elevated PA and LPA levels, while PLD1 inhibitor VU01 and VU03 downregulated PA and LPA levels respectively (Figs. S6A and B). The levels of PA34:1, PA36:2, PA38:6/4/3 and PA40:6/5 among 13 PA species and those of LPA18:1 among 3 LPA species were significantly decreased by VU01 and VU03 (Fig. S6C).
Fig. 5PLD1 inhibition or knockdown decreased the content of PA and LPA. The changes of (A) phosphatidic acid (PA) and (B) lysophosphatidic acid (LPA), and (C) heatmap of the different isoforms of PA and LPA determined by lipidomic analysis in the liver of Pld1-Flox and Pld1(H)-KO mice after 20 weeks of HFD feeding. The changes of (D) PA and (E) LPA, and (F) heatmap of the different isoforms of PA and LPA determined by lipidomic analysis in the AML12 cells after OA treatment. (G) Gene expression levels of Cd36 in PA- or LPA-treated AML12 cells. (H) The ratio of CD36+ cells to 7AAD- cells. N=8-10 mice per group; Differences between two groups were compared by Student’s t test for normal distribution and Mann-Whitney test for abnormal distribution. One-way ANOVA with Tukey’s test for normal distribution and Kruskal-Wallis test for abnormal distribution were used for analyzing differences among three or more groups. N=3-6 dishes of cells per group. *p < 0.05, **p < 0.01, ***p < 0.001. VU0155069 (VU01) and VU0359595 (VU03), PLD1 specific inhibitors; PA, phosphatidic acid; LPA, lysophosphatidic acid; NC, normal control; SP, sodium palmitate; OA, oleic acid.
Because PA and LPA are products of PLD1, we explored whether they regulate CD36 expression. We found that PA, but not LPA, increased the mRNA expression levels of Cd36 in AML12 cells (Fig. 5G). Meanwhile, CD36 expression, which was reduced by the inhibition of PLD1, was normalized by PA supplementation (Fig. 5H). These results suggest that PA, catalyzed by PLD1, plays a role in increasing CD36 expression in hepatocytes.
3.6 PA (a PLD1 product) impacts CD36 expression through PPARγ
To explore the specific regulatory mechanism by which PA regulates Cd36 expression, we performed KEGG pathway enrichment analysis on the differentially expressed genes in Pld1(H)-KO-HFD vs. Pld1-Flox-HFD mice and found that the PPAR signaling pathway occupied a prominent position (Fig. 6A). Among the genes that were enriched in the PPAR pathway and differentially expressed in both Pld1-Flox-HFD mice vs. Pld1-Flox-NCD mice and Pld1(H)-KO-HFD mice vs. Pld1-Flox-HFD mice, we found Pparg to have the same change trend as Cd36 (Fig. 6B). Similarly, both mRNA and protein levels of PPARγ were elevated after HFD-feeding and decreased by hepatocyte specific deficiency of Pld1 (Fig. 6C, D). In AML12 cells, inhibiting PLD1 using VU01 or VU03 decreased the mRNA expression levels of Pparg, which were increased by OA or SP treatment (Fig. 6E, S6D). The mRNA expression levels of Pparg were also increased in OA-treated Pld1-Flox hepatocytes, while decreased in Pld1(H)-KO hepatocytes (Fig. 6F). PA stimulation could increase Pparg gene expression in AML12 cells (Fig. 6G). Pparg reportedly participate in the transcriptional regulation of Cd36, and we also predicted that there were Pparg transcription factor-binding sites in the promoter region of Cd36 (Fig. 6H). CUT&Tag-PCR showed the presence of CD36 promoter region fragments in the DNA segment to which PPARγ protein binds in the AML12 cells. PA significantly increased the binding of PPARγ to the CD36 promoter region compared with the NC group (Fig. 6I). The upregulation of CD36 expression caused by PA was blocked by the PPARγ antagonist GW9662 (Fig. 6J-L).
Fig. 6PA promoted CD36 expression through PPARγ. (A) KEGG pathway enriched in differentially expressed genes in Pld1(H)-KO-HFD mice compared with Pld1-Flox-HFD mice. (B) Heatmap of genes related to PPAR signaling pathway according to their expression levels. (C) Representative western blot images showing PPARγ and GAPDH levels, and quantification of PPARγ by densitometry, (D) gene expression levels of Pparg in Pld1-Flox and Pld1(H)-KO mice after 20 weeks of NCD or HFD feeding. (E) Gene expression levels of Pparg in OA-stimulated AML12 cells with or without PLD1 inhibitor treatment. (F) Gene expression levels of Pparg in OA treated primary hepatocytes from Pld1-Flox and Pld1(H)-KO mice. (G) Gene expression levels of Pparg in PA treated AML12 cells. (H) Transcription factor binding site in Cd36 promoter region. (I) Binding of PPARγ on the Cd36 promoter region was detected by CUT&Tag-qPCR experiment in AML12 cells with or without PA stimulation. (J) Representative images of flow cytometry data showing the percentages of CD36+ cells to 7AAD- cells, (K) the ratio of CD36+ cells in 7AAD- cells and (L) MFI of CD36+ cells to 7AAD- cells in PA treated AML12 cells with or without GW9662. N=8-10 mice per group; N=3-6 dishes of cells per group. Differences between two groups were compared by Student’s t test for normal distribution and Mann-Whitney test for abnormal distribution. One-way ANOVA with Tukey’s test for normal distribution and Kruskal-Wallis test for abnormal distribution were used for analyzing differences among three or more groups. *p < 0.05, **p < 0.01, ***p < 0.001. PLD1, phospholipase D1; NCD, normal control diet; HFD, high-fat diet; VU0155069 (VU01) and VU0359595 (VU03), PLD1 specific inhibitors; PA, phosphatidic acid; OA, oleic acid; GW9662, PPARγ antagonist; FMO, fluorescence minus one.
PLD hydrolyzes phospholipids into fatty acids and lipophilic substances. Phospholipases can be divided into four categories according to their catalytic activity: phospholipases A, B, C, and D. There are two main subtypes of PLD in mammals: PLD1 and PLD2 [
A phospholipase D-dependent process forms lipid droplets containing caveolin, adipocyte differentiation-related protein, and vimentin in a cell-free system.
FOXM1-mediated activation of phospholipase D1 promotes lipid droplet accumulation and reduces ROS to support paclitaxel resistance in metastatic cancer cells.
]. Furthermore, PA (a PLD1 product) is converted to DAG, which is further metabolized to either TG or PC, followed by enhanced lipid accumulation in mouse embryonic fibroblasts [
]. In addition, the activation of PLD1 and formation of PA are critical for the assembly and output of very-low-density lipoproteins in the rat liver cell line McA-RH7777 [
]. These results suggest that PLD1 is an important regulator of lipid metabolism in various cells.
In our study, hepatocyte Pld1 specific knockout significantly decreased lipid accumulation in hepatocytes via downregulating CD36 expression. However, another study showed that Pld1 systemic knockout induced NAFLD by decreasing LD decomposition due to decreased autophagy [
]. Thus, systemic knockout of PLD1 might increase HSCs activation, promote NAFLD development and liver fibrosis, which may partially explain the discrepancy between Pld1 systemic and hepatocyte specific knockout mice. Further studies are needed to clarify the role of PLD1 in different liver cells during NAFLD development.
Overexpression of fatty acid uptake systems, such as CD36 scavenger receptors, is an important cause of intracellular lipid accumulation in non-adipose tissues [
]. CD36-mediated hepatic free fatty acid uptake and CCL2-induced inflammation jointly drive the progression of NAFLD to non-alcoholic steatohepatitis [
]. The rate of fatty acid uptake is controlled by CD36 on the cell surface, which is mainly regulated by subcellular vesicle circulation from the endosome to the plasma membrane [
]. In our study, CD36 was upregulated in Pld1-Flox-HFD mice and downregulated in Pld1(H)-KO-HFD mice, consistent with changes in the lipid accumulation phenotype and NAFLD symptoms. In AML12 cells, the expression levels of CD36 increased after SP treatment and then decreased by PLD1 inhibition. These results suggest that PLD1 may increase hepatocyte lipid accumulation by promoting the CD36 expression.
Lipotoxicity is defined as a condition in which accumulation of harmful lipids can lead to organelle dysfunction, cell injury and death, which are strongly associated with the progression from NAFLD to non-alcoholic steatohepatitis [
]. These toxic lipids may cause cellular damage through different mechanisms including the modification of intracellular organelle function, such as the endoplasmic reticulum and mitochondria, as well as the direct activation of death receptor signaling pathways [
]. In the present study, hepatocyte-specific Pld1 deficiency significantly decreased hepatic DAG, FFA and cholesteryl ester, which may reduce the hepatotoxicity induced liver inflammation and fibrosis in HFD-fed mice.
PLD1 specific deletion in hepatocytes could lower mice liver and body weight, but did not affect mice food intake. PLD1, as a member of the phospholipase family, plays important roles in regulating hepatocytes lipid metabolism. Hepatocytes lipid metabolism was closely related with plasma lipid level and body weight. For example, Scap/SREBP pathway is essential for the synthesis of fatty acids, triglycerides, and cholesterol in all organs. Hepatocyte-specific deletion of Scap in ob/ob mice could block hepatocytes fatty acids synthesis, prevent hepatic steatosis, decrease liver weight and body weight [
]. Carbohydrate Responsive Element-Binding Protein (ChREBP) also plays a key role in the control of lipogenesis through transcriptional regulation of lipogenic gens, which significantly increasing in ob/ob mice liver. Liver-specific inhibition of ChREBP improves hepatic steatosis by specifically decreasing lipogenic rates, and also downregulates liver and bodyweight [
]. Therefore, we speculate that hepatic specific Pld1 knockout could downregulate hepatocytes lipid metabolism firstly, and then alleviate hepatocytes lipid accumulation and hepatic steatosis, and ultimately lead to liver and body weight loss.
In this study we also found PLD1 knockout in HFD-fed mice could not only decrease lipid accumulation in hepatocytes, but also decrease plasma lipids and body weight. We speculate that PLD1 knockout in hepatocytes decreases CD36 expression, lowers lipid accumulation in hepatocytes, reduces lipotoxicity, and enhances hepatocytes viability and lipid metabolism functions, and finally reduces NAFLD and plasma lipid levels. Meanwhile, lipid mass spectrometry analysis also showed hepatocyte-specific Pld1 deficiency significantly decreased hepatic DAG, FFA and cholesteryl ester, which may also reduce the hepatotoxicity induced hepatocytes dysfunction and NAFLD in HFD-fed mice.
PPARγ plays a key role in adipogenesis and is important in a variety of cellular processes including cell cycle regulation, cell differentiation, and insulin sensitivity [
]. In addition, PPARγ can increase lipid uptake by affecting CD36. For instance, osteoprotegerin promotes CD36 expression by acting on the PPAR response element (PPRE) on the CD36 promoter to aggravate liver lipid accumulation in NAFLD [
]. On the one hand, the negative charged state of PA and the local accumulation of its negative head group promote the formation of curved membranes in the lipid bilayer, which is conducive for the formation of vesicles [
]. It is also worth mentioning that PA can be converted to DAG, which is further metabolized to TAG, followed by enhanced lipid accumulation in adipose tissue [
]. On the other hand, LPA (a downstream product of PA) stimulates the expression of the PPRE reporter factor and endogenous PPARγ controlling gene Cd36, and induces lipid accumulation in oxidized LDL monocytes [
Lysophosphatidic acid activates peroxisome proliferator activated receptor-gamma in CHO cells that over-express glycerol 3-phosphate acyltransferase-1.
]. Similarly, in the present study, we found that PA promoted the expression of PPARγ and CD36. Therefore, the mechanism by which PA promotes PPARγ and CD36 expression requires further investigation.
In summary, we found that the expression of PLD1 was significantly increased in the steatosis livers of mice and humans with NAFLD. Hepatocyte-specific depletion of Pld1 significantly reduced lipid accumulation in hepatocytes, both in vivo and in vitro. Pld1(H)-KO mice exhibited significantly lesser weight gain, lower fasting plasma lipid and glucose levels, and improved glucose tolerance, insulin sensitivity, and NAFLD. PA may be the key factor in activating the PPARγ/CD36 pathway during hepatocyte lipid accumulation.
In conclusion, the inhibition of hepatocytes PLD1 exerted potent protective effects against HFD-induced hepatic steatosis. The improved outcomes observed were due to a reduction in PPARγ/CD36 pathway-mediated lipid accumulation in hepatocytes. Therefore, PLD1 inhibition may be a new target for the treatment of hepatic steatosis.
Appendix A. Supplementary data
The following is the supplementary data to this article:
A phospholipase D-dependent process forms lipid droplets containing caveolin, adipocyte differentiation-related protein, and vimentin in a cell-free system.
FOXM1-mediated activation of phospholipase D1 promotes lipid droplet accumulation and reduces ROS to support paclitaxel resistance in metastatic cancer cells.
Lysophosphatidic acid activates peroxisome proliferator activated receptor-gamma in CHO cells that over-express glycerol 3-phosphate acyltransferase-1.
00057-5&id=ga1.jpg){kind=link}
00057-5&id=gr1.jpg){kind=link}
00057-5&id=gr2.jpg){kind=link}
00057-5&id=gr3.jpg){kind=link}
00057-5&id=gr4.jpg){kind=link}
00057-5&id=gr5.jpg){kind=link}
00057-5&id=gr6.jpg){kind=link}