Adenosine-rich extract of Ganoderma lucidum: A safe and effective lipid-lowering substance

Summary

Ganoderma lucidum is a traditional Chinese medicine with a variety of active compounds and possesses adequate lipid-lowering and anti-atherosclerotic effects. However, its main active components and potential mechanisms still remain unclear. Here, we evaluated the anti-hyperlipidemic effect of the adenosine extract from Ganoderma lucidum (AEGL) in high-fat-diet (HFD)-induced hyperlipidemic ApoE^−/− mice and explored the underlying biological mechanism by multi-omics analysis. Treatment with AEGL for 8 weeks significantly decreased the serum levels of total cholesterol (TC), triglyceride (TG), and low-density lipoprotein cholesterol (LDL-c) by 45.59%, 41.22%, and 39.02%, respectively, as well as reduced liver TC and TG by 44.15% and 76.23%, compared with the HFD-only group. We also observed significant amelioration of hepatic steatosis without liver and kidney damage after AEGL treatment. Regulating the expression and acetylation/crotonylation of proteins involved in the PPAR signaling pathway may be one of the potential mechanisms involved in the observed lipid-lowering effects of AEGL.

Graphical abstract

Highlights

• •AEGL significantly reduced LDL, TC, and TG
• •AEGL improved hepatic steatosis without impairing liver and kidney function
• •AEGL regulated the expression and modification of proteins involved in the peroxisome proliferator-activated receptors (PPAR)

Alternative medicine; Lipid; Physiology; Omics;

Introduction

Hyperlipidemia is characterized by abnormal high levels of low-density lipoprotein cholesterol (LDL-c) and/or triglycerides (TG) in blood, and as a key risk factor for global cardiovascular and metabolic diseases, it has been responsible for the increased incidence of acute cardiovascular and cerebrovascular events in recent years (Nicholls et al., 2018; Klimchak et al., 2020; Bai et al., 2021). Deaths from cardiovascular diseases have steadily increased from 12.1 million (95% UI:11.4 to 12.6 million) in 1990, reaching 18.6 million (95% UI: 17.1 to 19.7 million) in 2019 (Du et al., 2016), and it is also estimated that there are currently more than 400 million patients with hyperlipidemia in China ([2016 Chinese guideline for the management of dyslipidemia in adults], 2016). A clinical study of 170,000 individuals has reported that the average annual incidence of major cardiovascular disease events was decreased by about 20% when a decrease of 1.0 mmol/L in LDL-c level existed in a patient individual (Baigent et al., 2010). Hence, relieving hyperlipidemia-related symptoms is a crucial strategy to hamper the progress of cardiovascular and cerebrovascular complications. However, the existing lipid-lowering drugs are plagued by various factors, including adverse reactions, drug resistance, individual differences in treatment responses, or poor efficacy (Hirota et al., 2020). Therefore, it is urgently needed to discover novel hypolipidemic agents with more efficiency and less adverse effects.

Ganoderma lucidum is a Basidiomycete fungus with high medicinal value in traditional Chinese medicine (TCM). It contains a variety of essential bioactive compounds and thus has unique advantages in treating or preventing many diseases. Previous studies have found that polysaccharide and ganoderic acid extract from Ganoderma lucidum can help to reverse obesity and effectively alleviate insulin resistance (Chang et al., 2015; Yang et al., 2018). In addition, the ethanol extract of Ganoderma lucidum spore has been found to possess lipid-lowering and anti-atherosclerotic effects that occur owing to the extract’s increasing expression of LXRα and downstream genes related to cholesterol transport and metabolism (Lai et al., 2020). Adenosines are an important bioactive component in Ganoderma lucidum, which have several derivatives (Khan et al., 2015). Previous researches have shown that adenosine can ameliorate metabolic disorders by activating brown adipose tissues (Cypess et al., 2015; Gnad et al.,2014, 2020). However, the side effects of pure adenosine limit its clinical application. Therefore, as an adenosine-rich extract from a medicinal plant used in TCM, whether adenosine extract of Ganoderma lucidum (AEGL) possesses a safe anti-hyperlipidemia effect and the potential mechanisms are worth exploring.

Protein post-translational modifications (PTMs) are crucial biological processes that add polypeptides or chemical groups to protein amino acid residues in order to control various cellular processes by altering protein stability, localization, activity, or interactions (Keenan et al., 2021). Several kinds of physiological activities and diseases are not only associated with the abundance of certain proteins but are essentially regulated by PTMs of various other proteins as well (Yang et al., 2017). The metabolism of lipids involves an accurate and complex regulation network, such as gene expression, transcription and translation, and protein PTMs (Harney et al., 2021), and as one of the most common PTMs, protein acetylation modification, in particular, plays an important role in lipid metabolism. Previous studies have shown that the acetylation of several potentially important proteins, such as Hadha, Acat1, and Ehhadh, may be crucial regulatory factors in the development of fatty liver disease (Zhang et al., 2020). Lysine crotonylation (Kcr) is another type of PTM that has been newly identified on histones and has been shown to be related to transcription activation and cell signaling pathways (Yang et al., 2021). However, the biological functions related to lipid metabolism of histone crotonylation have not yet been reported.

In this work, we first evaluated the lipid-lowering effects of adenosine-rich AEGL in HFD-induced ApoE^−/− mice and then explored its underlying mechanisms based on proteomics, protein PTMs, and lipidomic analysis.

Results

Identification of the nucleosides profile of Ganoderma lucidum by UPLC/Q-TOF-MS

The AEGL was firstly extracted through 30% aqueous MeOH (v/v) and isolated by various column chromatographic methods as well as HPLC, then the content of adenosine was preliminarily calculated. After calculation, the peak area percentage of adenosine components accounted for 72.68% at the wavelength of 259nm in the HPLC spectrum (Figure 1A). We then identified its composition by UPLC-Q-TOFMS. The typical base peak intensity (BPI) chromatograms of AEGL in both positive and negative ion modes are presented in Figure 1B. Based on the retention time, MS₂ fragmentation patterns of molecular ions in positive mode and negative mode, and comparison with reference standards (Gao et al. (2007) Ge et al., 2019), 12 nucleoside-type compounds were detected and identified as (1) uracil1; (2) N6-(2-hydroxyethyl)adenine; (3) cytidine; (4) guanine; (5) hypoxanthine; (6) uridine; (7) isouridine; (8) adenine; (9) inosine; (10) guanosine; (11) thymidine; and (12) adenosine (Figure 1C and Table 1).

Figure 1.

The adenosine profile of Ganoderma lucidum detected by UPLC/Q-TOF-MS

(A) The HPLC chromatogram of AEGL at 259nm.

(B) The typical base peak intensity (BPI) chromatograms of adenosine extract of Ganoderma lucidum (AEGL) scanned in both positive and negative ion mode.

(C) The typical MS spectrums of compounds 1 through 12 from AEGL.

Table 1.

The adenosine-type compounds identified from the Ganoderma lucidum extraction by UPLC-Q-TOFMS

No.	tR (min)	m/z	Adduct	Structure identification
1	1.45	112.9989	[M + H]+	Uracil20
2	2.72	180.0750	[M−H]-	N6-(2-hydroxyethyl)adenine21
3	2.87	244.0939	[M + H]+	Cytidine20
4	3.58	152.0553	[M + H]+	Guanine20
5	3.64	137.0451	[M + H]+	Hypoxanthine20
6	4.96	243.0713	[M−H]-	Uridine21
7	5.21	267.0580	[M + Na]+	Isouridine20
8	6.84	136.0643	[M + H]+	Adenine20
9	9.22	267.0811	[M−H]-	Inosine20
10	9.78	284.0978	[M + H]+	Guanosine20
11	11.35	265.0813	[M + Na]+	Thymidine20
12	12.95	268.1094	[M + H]+	Adenosine20

Note: 20 and 21represent corresponding references.

Adenosine extract of Ganoderma lucidum alleviates hyperlipidemia in ApoE^−/− mice

Currently, several mouse models including high-fat-diet (HFD)-fed wildtype mice, ob/ob obese mice, LDLR^−/− mice and ApoE^−/− mice, are popularly used to investigate the anti-hyperlipidemic effects of various drugs. As the response of wildtype mice to HFD-induced hypertriglyceridemia is unstable (Gao et al., 2010) and previous studies have shown that Ganoderma lucidum and its related components can improve hyperlipidemia through modulating LDLR and leptin (Meneses et al., 2016), so we chose ApoE^−/− mice to investigate the anti-hyperlipidemic effects and potential mechanisms of AEGL on HFD-induced hyperlipidemia. We fed ApoE^−/− mice with HFD for 4 weeks then exposed them to low and high doses of AEGL (7.63 mg/kg/d and 15.26 mg/kg/d) and distilled water for 8 weeks. Serum levels of total cholesterol (TC), triglycerides (TG), and low-density lipoprotein cholesterol (LDL-c) in both the low- and high-dose AEGL groups were significantly lower than those in the HFD group, while the serum levels of high-density lipoprotein cholesterol (HDL-c) of three groups had no significant differences (Figure 2A). Statistically, TC, TG, and LDL-c were decreased by 30.24%, 43.42%, 35.96% and 28.67%, 29.14%, 38.35% in the low- and high-dose AEGL groups, respectively, as compared to the HFD group (Figure 2B). Liver function, as assessed by serum levels of alanine aminotransferase (ALT) and aspartate aminotransferase (AST) did not appear to be affected by AEGL (Figure 2C). In addition, neither dosage of AEGL caused renal function injury as reflected by the measurements of serum creatinine (Cr), blood urea nitrogen (BUN), and uric acid (UA) (Figures 2D-2F).

Figure 2.

AEGL treatment ameliorates the symptoms of hyperlipidemia in ApoE^−/− mice

(A) The serum levels of TC, TG, LDL-c, and HDL-c in the chow, HFD, low dose, and high dose of AEGL groups (n = 10).

(B) The percentage change in serum TC, TG, LDL-c, and HDL-c from low and high dose of AEGL as compared to the HFD group.

(C) The serum levels of AST and ALT in the chow, HFD, low dose, and high dose of AEGL groups (n = 10).

(D-F) The serum levels of CR, BUN, and UA in the chow, HFD, low dose, and high dose of AEGL groups (n = 10). All data are presented as mean ± SEM, ∗p < 0.05.

To explore the lipid-lowering effects of AEGL, we designed a study where 6-week-old male ApoE^−/− mice were fed with a HFD for 4 weeks and subsequently treated with a 7.63 mg/kg/d dosage of AEGL for 8 weeks. As expected from the above results, both bodyweight and bodyweight gain (Figures 3A and 3B) were significantly lower after AEGL treatment, and the serum TC, TG, and LDL-c were decreased by 45.59%, 41.22%, and 39.02%, respectively, in the AEGL group as compared to the HFD group (Figures 3C and 3D).

Figure 3.

AEGL treatment alleviates liver lipid accumulation in ApoE^−/− mice

(A) Bodyweight and (B) bodyweight gain of ApoE^−/− mice from 1 day to 95 days in the HFD group and AEGL group.

(C) The serum levels of TC, TG, LDL-c, and HDL-c in the HFD group and AEGL group (n = 10).

(D) The percentage change in serum TC, TG, LDL-c, and HDL-c in the AEGL group as compared to the HFD group.

(E) The liver levels of TC and TG in the HFD group and AEGL group.

(F) The percentage change in liver TC and TG in the AEGL group as compared to the HFD group.

(G) Oil red O staining of the liver in the HFD group and AEGL group (×200). All data are presented as mean ± SEM, ∗p < 0.05, ∗∗∗p < 0.001.

Subsequently, we focused on the adipose tissues, one of the major sites of lipid metabolism, and on the activation of uncoupling protein 1 (UCP1), a protein involved in the anti-obesity effects of adenosine (Cypess et al., 2015; Gnad et al.,2014, 2020).^. Hematoxylin/eosin (H&E) staining of scapular and inguinal adipose tissue and Western blot analysis of UCP1 both showed that AEGL had no significant effect on adipose histopathology but reduced UCP1 expression in scapular and inguinal adipose tissue compared to the HFD group (Figures S1A-S1C), suggesting that AEGL did not stimulate adipose browning. Contrastively, the liver TC and TG in the AEGL group were significantly lower, by 44.15% and 76.23%, respectively, compared to the HFD group (Figures 3E and 3F). Oil red O staining also showed that AEGL dramatically ameliorate hepatic steatosis (Figure 3G), indicating that AEGL may alleviate hyperlipidemia via the modulation of liver lipid metabolism.

Adenosine extract of Ganoderma lucidum alters hepatic protein expression and post-translational modifications involved in fatty acid metabolism

Regulations on protein expression and PTMs (e.g. phosphorylation, acetylation, and lysine crotonylation) have been widely indicated to participate in various physiological and pathological processes (Radamaker et al., 2021). We, therefore, performed proteomics on liver protein, aiming to uncover the underlying mechanism of AEGL’s lipid-lowering effects. Principal component analysis (PCA) showed that the protein profiles in both the HFD and AEGL groups had marked separation. Similarly, levels of the acetylation and lysine crotonylation modification were also evidently altered by AEGL treatment (Figure 4A). Moreover, we performed KEGG gene sets and GO biological process analysis to explore certain signaling pathways associated with AEGL’s lipid-lowering effects. The KEGG gene sets showed that the differential proteins were mostly enriched in the lysosome, peroxisome proliferator-activated receptors (PPAR) signaling pathways, as well as peroxisome and steroid biosynthesis (Figure 4B). GO biological process analysis showed that both the small-molecule metabolic process and lipid metabolic process were affected by AEGL (Figure 4C), suggesting that AEGL can regulate lipid metabolism, especially fatty acid metabolism.

Figure 4.

AEGL treatment regulates the protein expression and PTMs involved in fatty acid metabolism

(A) Principal component analysis (PCA) of proteomic, acetylation, and crotonylation modification.

(B and C) Bubble map of the most enriched pathway based on KEGG gene sets and GO biological process analysis in proteomics.

(D and E) Bubble map of the most enriched pathway based on KEGG gene sets and GO biological process analysis in acetylation modification.

(F and G) Bubble map of the most enriched pathway based on KEGG gene sets and GO biological process analysis in crotonylation modification.

To improve the accuracy of the analysis, we further narrowed down candidate proteins with a score greater than 60 in the modification-specific proteomics study. KEGG gene sets analysis showed that the acetylated modified proteins were mostly enriched in the PPAR signaling pathway and fatty acid metabolism (Figure 4D), and the small-molecule metabolic process, organic acid metabolic process, lipid metabolic process, fatty acid metabolic process, lipid catabolic process, lipid oxidation, and fatty acid beta-oxidation were all affected by AEGL to some degree (Figure 4E). Additionally, the differential crotonylated modification proteins were enriched in fatty acid metabolism and the PPAR signaling pathway as well (Figure 4F). Finally, our GO biological process once again showed that the lipid metabolic process was affected by AEGL (Figure 4G). Together, our proteomics and PTMs data suggest that the activation of the PPAR signaling pathway plays a key role in AEGL’s lipid-lowering effects.

To explore further how AEGL affects protein expression or PTMs related to fatty acid metabolism, we combined proteomic and PTMs in HFD-induced ApoE^−/− mice with or without AEGL treatment in order to understand the anti-hyperlipidemia effects of AEGL better. Based on a fold change (AEGL/HFD) greater than 1.5 and a p value less than 0.05, we obtained 189 up-regulated proteins and 119 down-regulated proteins; 203 up-regulated acetylated modified positions in 176 proteins and 38 down-regulated acetylated modified positions in 35 proteins; as well as 59 up-regulated lysine crotonylated modified positions in 49 proteins and 70 down-regulated lysine crotonylated modified positions in 66 proteins (Figures 5A–5C). We primarily detected the protein expression and modification levels in fatty acid metabolism, and the results showed that the proteins in the PPAR signaling pathway, such as Fabp1, Acsl5, Abcd3, Ech1, Acox1, Cpt2, and Hadh were mostly down-regulated. However, the acetylation modifications of those proteins were mostly up-regulated, while crotonylation modification levels were regulated to varying degrees (Figure 5D and Table 2). We then assessed the expression levels of several proteins in the table by Western blot analysis and Parallel reaction monitoring (PRM) (Figure S2 and Table S1), and evaluated modifications of these proteins by Co-immunoprecipitation and Western blot (Figure S3). Our results showed that the profiles of protein expression and modification were basically consistent with proteomics.

Figure 5.

Multi-omics analysis reveals that AEGL promotes fatty acid metabolism in HFD-fed ApoE^−/− mice

(A-C) Volcano plot of differentially expressed proteins based on a ratio of AEGL/HFD greater than 1.5 and a p value less than 0.05 in proteomic, acetylation, and crotonylation modification.

(D) Ciucus plot showing the difference in protein expression among proteomic, acetylation, and crotonylation modification. The outermost circles represent different groups. The second circle is the subcellular localization annotation of the protein presented in the axis, showing only the top 5 and the other set as “Others.” The number is the cumulative total number of amino acids of the protein. The third circle is the number of modifications occurring on each protein, presented as a heatmap. The fourth circle is the abundance of protein (mean for three replicates of the same group and converted based on log₂) presented in a bar graph. The fifth through the innermost circles are sequentially different types of modifications presented as a heatmap, with each vertical line representing 1 protein. The depth of color indicates the number of corresponding modification sites.

Table 2.

The levels of proteins and PTMs in the PPAR signaling pathway

Protein accession	Gene name	Expression (p value)	Acetylation modification Position (p value)	Crotonylation modification Position (p value)
Q05816	Fabp5	up(∗∗)	–	–
Q9CZW4	Acsl3	up(∗)	–	–
P12710	Fabp1	down(∗∗∗)	up: 57(∗)/46(∗∗∗)	up: 46(∗)
P55050	Fabp2	down(∗∗∗)	–	down: 28(∗)
P41216	Acsl1	down(∗∗)	–	down: 387(∗)
Q8JZR0	Acsl5	down	up: 361(∗∗)/616(∗)/125(∗)	up: 412(∗)
Q3UNX5	Acsm3	down(∗∗)	–	–
Q8BGA8	Acsm5	down(∗)	–	–
P55096	Abcd3	down(∗)	up: 260(∗∗)	–
Q9DBM2	Ehhadh	down(∗∗)	–	down: 240(∗)
O35459	Ech1	down(∗∗)	up: 316(∗∗)	–
Q921G7	Etfdh	down(∗∗)	up: 222(∗)	–
Q99LC5	Etfa	down(∗∗)	up: 206(∗)	–
Q9R0H0	Acox1	down(∗∗)	up: 260(∗∗)	up: 260(∗)
Q921H8	Acaa1a	down(∗)	up: 198(∗); down: 292(∗)	–
Q8VCH0	Acaa1b	down(∗)	–	–
Q8BWN8	Acot4	down(∗∗)	–	–
Q8BWT1	Acaa2	down(∗∗)	up: 214(∗)	–
Q8QZT1	Acat1	down(∗∗)	up: 265(∗)	–
P50544	Acadvl	down(∗)	down: 557(∗)	up: 277(∗); down: 128(∗)
P52825	Cpt2	down(∗∗)	up: 62(∗)/305(∗∗)/239(∗)	up: 544(∗∗)/69(∗)
Q8BMS1	Hadha	down(∗∗)	up: 284(∗)/634(∗)	–
Q99JY0	Hadhb	down(∗∗∗)	up: 409(∗∗)	up: 292(∗∗)
O08756	Hsd17b10	down(∗)	–	–
P54869	Hmgcs2	down(∗∗)	–	up: 427(∗)
Q8R0Y8	Slc25a42	down(∗)	–	–
Q99PG0	Aadac	down(∗∗)	–	–
Q8VI47	Abcc2	down	up: 764(∗∗)	–
Q8K3K7	Agpat2	up(∗∗∗)	–	down: 216(∗)

Lipidomics analysis of fatty acid metabolism

We further performed lipidomics analysis to examine the readouts of lipid metabolism regulated by AEGL. As shown in Figure 6A, glycerides, free fatty acids, and carnitines in the AEGL group were markedly separated from the HFD group. A volcano plot of glyceride profile revealed that among 230 glycerides, 62 were decreased in AEGL-treated mice (AEGL/HFD ratios > 2-fold; p value < 0.05) (Figure 6B). Among these, TAG56:1 (22:0), TAG54:1 (18:1), TAG55:2 (17:0), TAG56:1 (18:1), TAG51:2 (17:0), TAG52:1 (18:1), TAG53:3(17:0), TAG53:4 (18:3), TAG53:2 (19:0), TAG50:3 (16:2), TAG53:2 (19:1), TAG54:1 (18:0), TAG56:2 (22:1), and TAG55:2 (19:0) were markedly decreased by AEGL (Figure 6C). Compared with the HFD group, the saturated fatty acids FFA 14:0 and FFA 17:0, as well as monounsaturated fatty acids FFA 16:1, FFA 17:1, and FFA18:1 were reduced after AEGL treatment, while the polyunsaturated fatty acids such as FFA 20:3, FFA 20:4, FFA 20:5, FFA 22:4, FFA 22:5, FFA 18:2, and FFA 22:6 were enriched in the AEGL group (Figure 6D). Carnitine is a key substance in mediating fatty acid oxidation, and we found that 16:1-carnitine was significantly increased in the AEGL group, while 20:0-carnitine, 11:0-carnitine, and 10:0-carnitine were markedly decreased after AEGL treatment (Figures 6E and 6F). Given that acetyl CoA generated by fatty acid oxidation can enter the tricarboxylic acid (TCA) cycle, we then measured the levels of organic acids in order to investigate the effects of AEGL on the TCA cycle. Malate and aconitate were found to be increased in the AEGL group (Figure 6G), supporting that AEGL, indeed, affected the TCA cycle. These results are in line with GO biological process analysis regarding acetylation and crotonylation modification.

Figure 6.

The levels of glyceride, fatty acid, carnitine, and organic acid in liver based on lipidomics analysis

(A) PCA analysis of glyceride, fatty acid, and carnitine in the HFD and AEGL group.

(B) Volcano plot of differentially glyceride based on a ratio of AEGL/HFD greater than 2 and a p value less than 0.05.

(C) The significantly decreased glyceride in the AEGL group.

(D) The clearly decreased or increased free fatty acid in the AEGL group.

(E) Volcano plot of differentially expressed carnitine based on a ratio of AEGL/HFD greater than 2 and a p value less than 0.05.

(F) The clearly decreased or increased carnitine and CoA in the AEGL group.

(G) The clearly decreased or increased organic acids in the AEGL group. ∗p < 0.05.

Discussion

As a major cause of cardiovascular and cerebrovascular complications, hyperlipidemia is threatening global public health with increasing incidence (Klimchak et al., 2020). Among the current lipid-lowering drugs, statins can effectively reduce cholesterol levels, but their effect on triglyceride levels is not significant (Nicholls et al., 2018). Bates can promote the degradation of triglycerides, yet exhibiting no obvious effect on reducing cholesterol levels (Nicholls et al., 2018). Consequently, there is still room to exploit lipid-lowering drugs with multi-targets and more effectiveness. Ganoderma lucidum, a widely used TCM to treat metabolic diseases in China, possesses various active compounds, such as polysaccharides, organic acids, and nucleosides. Compared with polysaccharides and organic acids, little is known about the lipid-lowering effects of the nucleoside compounds in Ganoderma lucidum.

Our previous work reported that AEGL was favored to reduce cholesterol and triglycerides. ApoE^−/− mice fed with high-fat diet usually accompany increased serum TG, TC, LDL-c, and decreased HDL-c, and are commonly used hyperlipidemia model (Wu et al., 2014). We hypothesized that the efficacy of AEGL may be related to UCP1 which partially depends on the signaling pathway of leptin. Therefore, ApoE^−/− mice are superior to ob/ob (leptin^−/−) and Ldlr^−/− mice in investigating the lipid-lowering effect of AEGL and exploring its potential mechanism. Hence, in the present study, we established hyperlipidemia in ApoE^−/− mice with HFD-feeding for 4 weeks and evaluated AEGL’s lipid-lowering activity by pharmacodynamics, and then preliminarily explored its mechanism based on multi-omics. The results showed that AEGL could significantly reduce lipid levels in blood and liver, as well as significantly ameliorate hepatic steatosis without causing significant damage to liver and kidney. By integrating analysis of proteomics, modification-omics, and lipidomics, we found that AEGL mainly exerted its lipid-lowering effect by promoting fatty acid metabolism.

The liver and adipose tissue are the most critical organs to regulate lipid metabolism, and studies have shown that synthetic adenosine plays a lipid-lowering role primarily by activating adipose tissue (Gnad et al.,2014, 2020). However, the pathological examination of mouse adipose tissue and the expression of UCP1 in our study indicated that AEGL did not exert lipid-lowering effects by acting on adipose tissue. In particular, liver oil red O staining showed that AEGL was able to marked reduce the accumulation of lipid droplets induced by HFD feeding. Consistently, liver TC and TG were also reduced upon AEGL treatment. Therefore, we conclude that AEGL has a lipid-lowering effect by affecting liver fat metabolism.

Fatty acids (FAs) are important raw materials for the synthesis of triglycerides and cholesterol (Donnelly et al., 2005; Zhang et al., 2013). In hepatocytes, fatty acids are esterified to glycerol 3-phosphate (G3p) and cholesterol to produce TG and cholesterol esters, respectively. The lipidomics analysis in our study showed that the level of hepatic triacylglycerol (TAG) was significantly decreased by AEGL, and the oil red O staining showed that hepatic steatosis was alleviated by AEGL, indicating that AEGL can either reduce the synthesis or accelerate the decomposition of hepatic TAG. However, in the results from our multi-omics analysis, the key enzymes of liver TG degradation had no obvious change, but many proteins related to FA metabolism changed significantly.

In our present results, the crotonylation of ACSL1 at k387 was significantly down-regulated, and the acetylation of FASN at k287 was significantly up-regulated. Previous studies have shown that post-translational modification can affect the function of these proteins. For instance, histone acetylation can affect the function of FASN by impairing the binding of srebp-1csre and chrebpchore (Du et al., 2017), and the modification levels of myristoylation, acetylation, and phosphorylation of ACSL1 protein have been found to change under enhanced fatty acid oxidation (Zhang et al., 2021). However, the question of whether the elevation of acetylation modification level at the k287 site of FASN and the down-regulation of crotonylation modification level at the k387 site of ACSL1 affect their functions remains unanswered. Therefore, whether the lipid-lowering effect of AEGL is related to these modifications will be a question to be future studied.

The results of lipidomics showed that 62 kinds of TAGs in mouse liver decreased significantly after treatment with AEGL. Among the 62 TAGs significantly down-regulated by AEGL, TAG54:2 (18:0), TAG51:0 (17:0), TAG52:0 (18:0), TAG56:2 (22:1), TAG53:1 (19:0), and TAG52:4 (16:2) were the most significantly down-regulated. The FA chains of these tags primarily included FFA18:0, FFA17:0, FFA16:2, and FFA19:0. Decreased synthesis or increased the decomposition of these TAGs can lead to the accumulation of corresponding free FAs (FFAs) in the liver, and FFAs can then be used to synthesize TAG, thus increasing the TG storage of the liver and the over-production of VLDL, resulting in hepatic steatosis and the increase of serum TG, TC, and LDL (Donnelly et al., 2005). Interestingly, the amount of these FAs in the present study did not change significantly in the liver, and some FAs even showed a downward trend, suggesting that the rate of FA decomposition in the liver was greatly accelerated.

As presented in our results, many proteins related to FA metabolism changed significantly, and most of them were enriched in the peroxidase proliferators activated receptors (PPAR) signaling pathway (Figure 7). In particular, FA binding proteins (FABPs) were expressed in various tissues including adipocytes and liver. These proteins mediate lipid transport and inflammation by preventing excessive accumulation of long-chain FAs (Du et al., 2017). Researchers have reported that increasing the expression of FABP in the liver can enhance the uptake of medium or long-chain FAs and promote the process of intracellular targeting to the nucleus (Huang et al., 2002; Baran et al., 2019), and our study showed that FABP5 protein was highly expressed in the AEGL group, suggesting that AEGL may promote the uptake of unesterified long-chain FAs and then interact with PPAR to start transcription in the nucleus.

Figure 7.

The mechanistic figure of AEGL

Our Functional enrichment analysis showed that the acetylation and crotonylation of proteins related to FA metabolism were significantly changed, especially the related proteins in the PPAR signal pathway. Acetylation and lysine crotonylation are one of the most common epigenetic modifications and have been shown to be the primary factors that regulate the oxidative metabolism of FAs. The acetylation modification levels of these proteins were significantly up-regulated, while the crotonylation modification levels of these proteins were significantly up-regulated or down- regulated. Among the proteins with significant modification changes, liver carnitine palmitoyltransferase 1a (CPT1A) and peroxisome ABC transporter ABCD3 are essential in β-oxidation and peroxisome oxidation, respectively (Cacciola et al., 2021; Violante et al., 2019). Existing studies have shown that ABCD3, acyl CoA oxidase 2 (ACOX2), Acaa1a, and Ehhadh proteins are the key regulatory enzymes of peroxisome oxidation (Kersten and Stienstra, 2017; Nenicu et al., 2007), and the expression levels of CPT2, ACADVL, ACAA2, HADHA, ACADSB, and ACOX1 contribute to FAs β-Oxidation. We found that the acetylation modification levels of the above proteins, as well as the crotonylation modification levels of ACOX1 and ACSL5 were up-regulated after AEGL, while the crotonylation modification levels of FABP2, ACSL1, ACSM3, and ehhadh were down-regulated, indicating that the acetylation and crotonylation modification of lipid oxidation related proteins partly play a role in AEGL’S lipid-lowering activity.

In addition, from our lipidomics results, we found that the key substances 20:0-carnitine, 11:0-carnitine, and 10:0-carnitine required for FA oxidation were significantly reduced in the AEGL treatment group, which also indicated that FA oxidation was promoted. Combined with the results of proteomics and modification omics, we can see that the changes in acetylation and crotonylation modification in the FA oxidation pathway can, eventually, accelerate FA β-oxidation.

Furthermore, based on lipidomics, we found that AEGL increased the content of polyunsaturated FAs (PUFAs), but decreased the content of saturated and monounsaturated FAs. Studies have shown that increased PUFAs can inhibit the occurrence of hyperlipidemia. However, the sharp decrease of malonyl CoA and polyunsaturated FA (PUFAs) caused by ACC inhibitor, a drug for the treatment of nonalcoholic fatty liver, is one of the primary causes of side effects of hyperlipidemia treatment. Reduced PUFAs lead to an increase in the expression of SREBP and its downstream adipogenesis and VLDL secretion genes, and lower PUFAs also significantly inhibit FA oxidation, thus preventing lipid degradation in plasma (Zhang et al., 2022). Other studies have shown that 2 ω-3PUFA can reduce plasma triglycerides by 20% to 30% (Zhang et al., 2022). In animal models, dietary intake rich in linoleic acid (LA) can increase the expression of LDL receptors in the liver and effectively reduce LDL (Dekker et al., 2000). These results suggest that the increase in liver PUFAs by AEGL may be AEGL’s potential mechanism in reducing cholesterol and improving hyperlipidemia.

Whether acetyl coenzyme A, the prestigious metabolite of FA β-oxidation, can promote the TCA cycle merits its own discussion. To begin, aconitate is an intermediate from citric acid to isocitrate, and malic acid is also an important metabolite in the TCA cycle. Our lipidomics results showed that these substances were significantly increased, demonstrating that AEGL promoted the process of the TCA cycle. The enhancement of TCA also indicates an increase in fatty acid oxidation. Excess lipids, eventually, can be decomposed into CO₂ and H₂O to produce ATP (Choi et al., 2021), which improves hyperlipidemia and hepatic steatosis.

Our results indicate that AEGL is a lipid-lowering drug worthy of in-depth research and exploration. Because the action mechanism of AEGL is different from current mainstream lipid-lowering agents, AEGL provides possibilities for therapeutic combinations with other medications to manage blood lipids more effectively and safely. If AEGL can be deeply studied and developed, it may provide new therapeutic drugs for hypertriglyceridemia, hypercholesterolemia, and even nonalcoholic fatty liver.

Conclusions

This study showed that AEGL effectively reduced the levels of lipid in both serum and liver in HFD-fed ApoE^−/− mice. The possible biological mechanism of AEGL’s lipid-lowering was found to occur through regulating the expression of FA-metabolism-related proteins, as well as the up-regulating of acetylation and regulation of crotonylation modification levels of these proteins. Our study not only presents a new candidate for the treatment of hyperlipidemia but also provides ideas for mechanistic studies of multi-target drugs based on a multi-omics research model.

Limitations of the study

In this study, our limitations are that we did not explore the intrinsic logic of significant changes in protein quantity and modification level in the PPAR signaling pathway, and we did not further evaluate the toxicity of AEGL through cell experiments. In addition, there are some unknown adenosine components in AEGL, which need to be further separated by HPLC and identified by NMR and other spectroscopy techniques to identify the most active monomer compounds, which is also the future research direction.

Availability of data and material

All data involved in this article can be presented on request.

STAR★Methods

Key resources table

REAGENT or RESOURCE	SOURCE	IDENTIFIER
Antibodies
UCP1 Rabbit Polyclonal Antibody	proteintech	23673-1-AP
CPT2 Rabbit Polyclonal Antibody	proteintech	26555-1-AP
FABP5 Rabbit Polyclonal Antibody	proteintech	12348-1-AP
FABP1 Rabbit Polyclonal Antibody	proteintech	13626-1-AP
ACSL1 Rabbit Polyclonal Antibody	proteintech	13989-1-AP
ACSL5 Rabbit Polyclonal Antibody	proteintech	15708-1-AP
FABP2 Rabbit Polyclonal Antibody	proteintech	21252-1-AP
Recombination anti-ACOX1 Antibody	abcam	ab184032
Acetylated-Lysine Antibody	CST	9441s
Abti-Crotonyllysine MOuse mAb	PTM-BIO	PTM-502
Experimental models:Organisms/strains
Male ApoE−/− mice, age 6 weeks	Beijing Huafukang Biotechnology Co., Ltd. (Beijing, China)	N/A
Software and algorithms
Sangerbox 3.0	http://vip.sangerbox.com/home.html	N/A
MetaboAnalyst	https://www.metaboanalyst.ca/	N/A

Resource availability

Lead contact

Information and requests for resources should be directed to and will be fulfilled by the corresponding author, S. Wu (mark.wushx@sina.com).

Materials availability

No plasmids and mouse lines were generated as a part of this study. AEGL in this study is a new Ganoderma lucidum extract. Please refer to the STAR methods section for its extraction method. All sourced commercially or from academic core facilities on a fee-for-service basis. All other reagents were purchased commercially from the vendors described in the key resources table. All experiments met relevant regulatory standards.

Experimental model and subject details

The serum, adipose tissue and liver tissue involved in this study were obtained from the ApoE^−/− mice. These biological samples were used to generate the data listed in Figures 2, 3, 4, 5, and 6 and Table 2. This research study met all appropriate regulatory standards.

Mouse studies were conducted using 6 week-old male ApoE^−/− mice. All mice were purchased from a qualified supplier, Beijing Huafukang Biotechnology Co., Ltd. (Beijing, China), and were maintained in clean conditions while on-study. All mouse studies were conducted using protocols approved the IACUC for the study site (CHORI), and met all institutional and national standards for studies on live vertebrates. Animal experiments were approved by the Animal Care & Welfare Committee of Dongzhimen Hospital of Beijing University of Chinese Medicine (No. of the tab of Animal Experimental Ethical Inspection:21–37).

Method details

Preparation of Ganoderma lucidum extract

The crude drugs Ganoderma lucidum were purchased from Dongzhimen Hospital Beijing University of Chinese Medicine, and the preparation of Ganoderma lucidum adenosine extract was 30% aqueous MeOH (v/v). The main constituents of Ganoderma lucidum adenosine extraction were analyzed and identified by using ultra-performance liquid chromatography coupled with quadruple-time-of-flight mass spectrometry (UPLC-Q-TOF/MS).

First, the Ganoderma lucidum was crushed, and then sieved through 40–80 meshes to obtain the Ganoderma lucidum power. The powder was soaked in 50–80% methanol aqueous solution (0.05–0.2 g/mL), and then extracted 3–6 times with ultrasonic at the power of 400–600W.

After the evaporation of the solvent under reduced pressure, the residue was fractionated by Xad-16n macroporous ( MeOH-H2O, 40% and 60%). The subfraction was further separated by preparative HPLC (MeOH-H2O, from 0% to 30%; Elution time 150–200 min). The eluents of all chromatographic peaks at the detection wavelength of 259 nm were collected, combined and concentrated under reduced pressure.

UPLC-MS analysis

Ultra-performance liquid chromatography (Waters ACQUITY UPLC, USA) coupled with quadruple-time-of-flight mass spectrometry (Waters Xevo™ G2 Q/TOF, USA) was applied for Ganoderma lucidum extraction analysis. We purchased HPLC-grade methanol (MeOH) from Fisher Scientific (Fairlawn, NJ, USA), and our deionized water was purified using a Milli-Q device (Millipore, Milford, MA, USA). Chromatographic separation was performed using an ACQUITY UPLC HSS T3 analytic column (2.1 × 100 mm, 1.8 μm, Waters, USA) maintained at 25°C. The mobile phase consisted of solutions A (MeOH) and B (water) with a gradient elution of 0% A (0–3 min), 0–20% A (3–18 min), and 20–60% A (18–25min). The flow rate was 0.3 mL/min, and the temperature of the autosampler was set at 4°C. Injection of 1 μL aliquot of sample supernatant was carried out after equilibration.

Mass spectrometry was executed on a Xevo™ G2 Q/TOF (Waters Corp., Milford, MA, USA) equipped with electrospray ionization. The MS parameters were as follows. The MS scan m/z range was set to 100 to 1000 Da in either positive or negative ionization mode; the capillary voltage in positive and negative mode was set to 3.0 kV, 2.5 kV, respectively; the sampling cone voltage was set to 40; and the flow rate of desolvation gas was set to 800 L/h with a temperature of 400°C. All the data were collected in continuum mode and real-time corrected using a Lock Spray reference to ensure the accuracy. All the acquisitions were precisely controlled by Waters Masslynx v4.1 software (Waters Corporation, Milford, MA, USA).

Animal experiment design

A total of 40 male ApoE^−/− mice that were 6 weeks old were purchased from Beijing Huafukang Biotechnology Co., Ltd. (Beijing, China). The mice were housed in a temperature-controlled (22°C) chamber with a 12-h light/dark cycle. All operations were approved by the Key Laboratory of Dongzhimen Hospital of Beijing University of Traditional Chinese Medicine. The animals were randomly divided into 4 groups. After adaptive feeding for one week, 10 mice were fed with chow (chow group, n = 10), and 30 mice were treated with a high-fat diet(HFD) (21% lipid and 0.15% cholesterol) for 4 weeks to induce hyperlipidemia (HFD, HFD + AEGL (7.63 mg/kg/d) and HFD + AEGL (15.26 mg/kg/d) group, n = 10). The chow group and HFD group were intraperitoneally injected with distilled water, and the HFD + AEGL (7.63 mg/kg/d) and HFD + AEGL (15.26 mg/kg/d) groups were treated with AEGL at the dosages of 7.63 mg/kg/d and 15.26 mg/kg/d, respectively After 8 weeks, all mice were fasted for 12 h and then euthanized. The blood was collected and then centrifuged at 8000 rpm for 15 min at 4°C, and the serum was then stored at −80°C for further biochemical analysis.

To find the optimal dose of the AEGL, we randomly divided the 20 male ApoE^−/− mice into 2 groups (HFD and HFD + AEGL group) and fed them a HFD for 4 weeks to induce hyperlipidemia. Then the HFD group was intraperitoneally injected with distilled water, while the HFD + AEGL group was treated with AEGL at a dosage of 7.63 mg/kg/d. After 8 weeks, all mice were fasted for 12 h and then euthanized, and blood and tissue samples were collected for further examination.

Serum and liver lipid examination

The levels of serum total cholesterol (TC), low-density lipoprotein-cholesterol (LDL-c), high-density lipoprotein-cholesterol (HDL-c), triglycerides (TG), alanine aminotransferase (ALT), aspartate aminotransferase (AST), creatinine (Cr), blood urea nitrogen (BUN), and uric acid (UA), as well as liver TC and TG, were measured using a fully automated biochemical analyzer (Beckman, AU-480).

Histopathological examination of liver and adipose tissue

The liver and adipose tissue were fixed in 10% formalin solution for 72 h, rinsed with distilled water, and washed once with 60% isopropyl alcohol. Paraffin-embedded liver and adipose tissue were then cut into 5 μm thick sections for hematoxylin and eosin (HE) staining, and then the sections were observed with a pathological section panoramic scanner (Leica Aperio AT2). Liver embedded in optimal cut-temperature compound (Tissue-Tek, Laborimpex) was used for oil-red O staining in order to assess hepatic steatosis.

Proteomics analysis

Extraction of protein

Frozen liver tissue was pulverized in liquid nitrogen and ultrasonic cleavage with 4 times lysis buffer (8 M urea, 1% protease inhibitor, 3 μM TSA, 50 mM NAM). The resulting homogenate was then centrifuged at 12,000g for 10 min at 4°C, and the supernatant was collected for quantification using a BCA Protein Quantification Kit (Vazyme, China).

Trypsin enzymolysis

An equal amount of proteins were adjusted to the same volume with lysate, and slowly added to 20% TCA and mixed. The sample mixture was precipitated at 4°C for 2h and then centrifuged at 4,500 g for 5 min. The supernatant was discarded, and the precipitate was washed 2 to 3 times with pre-cooled acetone. After drying the precipitate, TEAB with a final concentration of 200 mM was added and then sonicated. The protein suspension was enzymolyzed with trypsin at a 1:50 ratio (protease: protein, m / m) overnight. Disulfur threitol (DTT) was added to gain a final concentration of 5 mM and reduced by 56°C for 30 min. Finally, iodiacetamide (IAA) was added to gain a final concentration of 11 mM and the suspension was incubated for 15 min at room temperature under darkness.

HPLC-MS/MS analysis

Peptides were resolved by mobile phase A, separated using an HPLC system (Thermo Fisher Scientific EASY-nLC 1000; Waltham, MA, USA) with a flow rate of 450 nL/min, and identified using a timsTOF Pro mass spectrometer (Thermo Fisher Scientific). The separation system comprised mobile phase A (0.1% formic acid and 2% acetonitrile) and mobile phase B (0.1% formic acid and 100% acetonitrile). Mobile phase B was used to separate the peptides over a 1.5 h separation time. From 0 min to 70 min, the linear gradient of mobile phase B was 6% to 24%; from 70 min to 84 min, the gradient of mobile phase B was 24% to 32%; and from 84 min to 87 min, the gradient of mobile phase B was 32% to 80%. Thereafter, mobile phase B remained at 80% from 87 min to 90 min. Our MS analysis was set to positive ion detection mode. The ion source voltage was set to 1.6 kV, and the scanning range for the secondary mass spectrometry was 100–1700 m/z. Additionally, the data acquisition used a parallel accumulation serial fragmentation (PASEF) mode. A secondary spectrogram with the number of parent ion charges in the range 0 to 5 after 10 PASEF mode acquisition after primary mass spectrometry as well as the dynamic exclusion time of tandem MS scans were set to 30 s s to avoid repeat scans of the parent ion.

Acetylation and lysine crotonylation analysis

Modification enrichment: Peptides were dissolved in IP buffer solution (100 mM NaCl, 1 mM EDTA, 50 mM Tris-HCl, 0.5% NP-40, pH 8.0), and the supernatant was transferred to prewashed anti-acetyllysine antibody resin (PTM-104, Hangzhou Jingjie Biotechnology Co., Ltd., PTM Bio) and anti-crotonyllysine antibody resin (PTM-503, Hangzhou Jingjie Biotechnology Co., Ltd., PTM Bio) and placed in a rotary shaker at 4°C, which was gently shaking and incubated overnight. After incubation, the resin was washed sequentially with IP buffer solution 4 times and with deionized water twice. Finally, the resin-bound peptide was eluted 3 times using 0.1% trifluoroacetic acid and the eluate was collected and vacuum freeze-drained. The salt was removed according to C18 ZipTips instructions and vacuum freeze-drained for HPLC-MS/MS analysis.

HPLC-MS/MS analysis: Peptides were resolved by mobile phase A and separated using the NanoElute UPLC system (Thermo Fisher Scientific EASY-nLC 1000; Waltham, MA, USA) with a flow rate of 450 nL/min and identified using a timsTOF Pro mass spectrometer (Thermo Fisher Scientific). The separation system comprised mobile phase A (0.1% formic acid and 2% acetonitrile) and mobile phase B (0.1% formic acid and 100% acetonitrile). Mobile phase B was used to separate the peptides over a 1.5 h separation time. From 0 min to 43 min, the linear gradient of mobile phase B was 6% to22%; from 43 min to 56 min, the gradient of mobile phase B was 22% to30%; and from 56 min to 58 min, the gradient of mobile phase B was 30% to80%. Thereafter, mobile phase B remained at 80% from 56 min to 80 min. Our MS analysis was set to positive ion detection mode. The ion source voltage was set to 1.8 kV, and the scanning range of the secondary mass spectrometry was 100–1700 m/z. The data acquisition mode used a parallel accumulation serial fragmentation (PASEF) mode. A secondary spectrogram with the number of parent ion charges in the range 0–5 after 10 PASEF mode acquisition after primary mass spectrometry and the dynamic exclusion time of tandem MS scans were set to 30 s s to avoid repeat scans of the parent ion.

Lipidomics analysis

Metabolites extraction

Lipids were extracted from approximately 30mg mouse liver samples using a modified version of Bligh and Dyer’s method as described previously. Briefly, mouse livers were homogenized in 900 μL of chloroform: methanol: MilliQ H₂O (3:6:1) (v/v/v) on a bead ruptor (OMNI, USA). The homogenate was then incubated at 1500 rpm for 1 h at 4°C. At the end of the incubation, 350 μL of deionized water and 300 μL of chloroform were added to induce phase separation. The samples were then centrifuged, and the lower organic phase-containing lipids were extracted into a clean tube. Lipid extraction was repeated once by adding 500 μL of chloroform to the remaining tissues in aqueous phase, and the lipid extracts were pooled into a single tube and dried in the SpeedVac under OH mode. Samples were then stored at −80°C until further analysis.

Itaconic acid and TCA cycle metabolites were extracted from mouse liver tissue using acetonitrile: water (1:1) and derivatized using 3-nitrophenylhdyrazones. Extraction of acyl-CoAs from 100mg mouse liver samples was carried out as previously described for lipid extraction but with some modifications. Briefly, 300 μL of extraction buffer containing isopropanol, 50 mM KH2PO4, and 50 mg/mL BSA (25:25:1 v/v/v) acidified with glacial acetic acid was added to the cells. Next, 19:0-CoA was added as an internal standard, and lipids were extracted by incubation at 4°C for 1 h at 1,500 rpm. Following this, 300 μL of petroleum ether was added, and the sample was centrifuged at 12,000 rpm for 2 min at 4°C. The upper phase was then removed, and the samples were then extracted two more times with petroleum ether as described above. To the lower phase finally remaining, 5 μL of saturated ammonium sulfate was added followed by 600 μL of chloroform: methanol (1:2 v/v). The sample was then incubated on a thermomixer at 450 rpm for 20 min at 25°C, followed by centrifugation at 12,000 rpm for 5 min at 4°C. Clean supernatant was transferred to a fresh tube and subsequently dried in the SpeedVac under OH mode (Genevac). Dry extracts were resuspended in appropriate volume of methanol: water (9:1 v/v) prior to liquid chromatography–mass spectrometry (LC-MS) analysis on a Thermo fisher U3000 DGLC coupled to a Sciex QTRAP 6500 Plus.

Lipidomics analysis

All lipidomic analysis was conducted at LipidALL Technologies using an Agilent 1290 II UPLC coupled with a Sciex QTRAP 6500 PLUS as mentioned above. For normal phase analysis of polar lipids, individual species were separated using a Phenomenex Luna 3μm-silica column (internal diameter 150 × 2.0 mm) under the following conditions: mobile phase A, (chloroform: methanol: ammonium hydroxide, 89.5: 10: 0.5) and mobile phase B, (chloroform: methanol: ammonium hydroxide: water, 55: 39: 0.5: 5.5). Free FAs were quantified using d31-16:0 (Sigma-Aldrich) and d8-20:4 (Cayman Chemicals) as internal standards.

Glycerol lipids including diacylglycerols (DAG) and triacylglycerols (TAG) were quantified using a modified version of reverse phase HPLC/MRM. Separation of neutral lipids was achieved on a Phenomenex Kinetex-C18 2.6 μm column (i.d. 4.6 × 100 mm) using an isocratic mobile phase containing chloroform:methanol:0.1 M ammonium acetate 100:100:4 (v/v/v) at a flow rate of 170 μL for 17 min. Levels of short-, medium-, and long-chain TAGs were calculated by referencing to spiked internal standards of TAG(14:0)3-d5, TAG(16:0)3-d5, and TAG(18:0)3-d5 obtained from CDN isotopes, and DAGs were quantified using d5-DAG17:0/17:0 and d5-DAG18:1/18:1 as internal standards (Avanti Polar Lipids).

Itaconic acid and TCA cycle metabolites were analyzed on a Jasper HPLC coupled to a Sciex 4500 MD system. In brief, individual itaconic acid and TCA cycle metabolites were separated on a Phenomenex Kinetex C18 column (100 × 2.1 mm, 2.6 μm) using 0.1% formic acid in water as mobile phase A and 0.1% formic acid in acetonitrile as mobile phase B. Finally, d4-succinic acid, d4-citric acid, d3-malic acid, 13C-3-lactate acid, d3-pyruvate acid, d4-fumarate acid purchased from Cambridge Isotope Laboratories were used as internal standards for quantitation.

Parallel reaction monitoring (PRM)

Protein Extraction: The sample was grinded by liquid nitrogen into cell powder and then transferred to a 5-mL centrifuge tube. After that, four volumes of lysis buffer (8 M urea, 1% Triton-100, 10 mM dithiothreitol, and 1% Protease Inhibitor Cocktail) was added to the cell powder, followed by sonication three times on ice using a high intensity ultrasonic processor (Scientz). The remaining debris was removed by centrifugation at 20,000 g at 4°C for 10 min. Finally, the protein was precipitated with cold 20% TCA for 2 h at −20°C. After centrifugation at 12,000 g 4°C for 10 min, the supernatant was discarded. The remaining precipitate was washed with cold acetone for three times. The protein was redissolved in 8 M urea and the protein concentration was determined with BCA kit according to the manufacturer’s instructions.

Trypsin Digestion: For digestion, the protein solution was reduced with 5 mM dithiothreitol for 30 min at 56°C and alkylated with 11 mM iodoacetamide for 15 min at room temperature in darkness. The protein sample was then diluted to urea concentration less than 2M. Finally, trypsin was added at 1:50 trypsin-to- protein mass ratio for the first digestion overnight and 1:100 trypsin-to-protein mass ratio for a second 4 h-digestion.

LC-MS/MS Analysis: The tryptic peptides were dissolved in 0.1% formic acid (solvent A), directly loaded onto a home-made reversed-phase analytical column. The gradient was comprised of an increase from 6% to 23% solvent B (0.1% formic acid in 98% acetonitrile) over 38 min, 23% to 35% in 14 min and climbing to 80% in 4 min then holding at 80% for the last 4 min, all at a constant flow rate of 700 nL/min on an EASY-nLC 1000 UPLC system. The peptides were subjected to NSI source followed by tandem mass spectrometry (MS/MS) in Q ExactiveTM Plus (Thermo) coupled online to the UPLC. The electrospray voltage applied was 2.0 kV. The m/z scan range was 350 to 1000 for full scan, and intact peptides were detected in the Orbitrap at a resolution of 35,000. Peptides were then selected for MS/MS using NCE setting as 27 and the fragments were detected in the Orbitrap at a resolution of 17,500. A data-independent procedure that alternated between one MS scan followed by 20 MS/MS scans. Automatic gain control (AGC) was set at 3E6 for full MS and 1E5 for MS/MS. The maxumum IT was set at 20 ms for full MS and auto for MS/MS. The isolation window for MS/MS was set at 2.0 m/z.

Data Analysis: The resulting MS data were processed using Skyline (v.3.6). Peptide settings: enzyme was set as Trypsin [KR/P], Max missed cleavage set as 2. The peptide length was set as 8–25, Variable modification was set as Carbamidomethyl on Cys and oxidation on Met, and max variable modifications was set as 3. Transition settings: precursor charges were set as 2, 3, ion charges were set as 1, 2, ion types were set as b, y, p. The product ions were set as from ion 3 to last ion, the ion match tolerance was set as 0.02 Da.

Co-immunoprecipitation (IP) of endogenous proteins

For co-immunoprecipitation, cells were collected 24 h after transfection and lysed in lysis buffer supplemented with a protease inhibitor cocktail. After centrifugation for 15 min at 12000 rpm and 4°C, the supernatant was collected and incubated with Protein A/G Sepharose beads (SC-2003, Santa Cruz) coupled to specific antibodies overnight with slow rotation at 4°C. The next day, beads were washed three times with highsalt wash buffer and three times with low-salt wash buffer, and finally eluted by boiling for 10 min with 5× sample buffer (as indicated). Precipitates were fractionated using SDS-PAGE at appropriate concentrations (as indicated).

Western blot

Protein concentration was adjusted with water and SDS-PAGE Loading Buffer (5×) (GenePool/GPP1820) and denatured by boiling at 100°C for 10 min. An equal amount of protein (50 μg) was separated by 12% SDS-PAGE, after it was transferred into PVDF membranes. The membrane was then sealed with Milk Blocking Buffer (GenePool/GPP1819)/BSA Blocking Buffer (GenePool/GPP1818) for 1 h, followed by the addition of specific primary antibody, with β-actin as the internal reference. The membrane was incubated overnight at 4°C and washed 3 times with TBST (GenePool/GPP1822). Anti-mouse HRP-labeled secondary antibody was added and incubated for 50 min and washed 4 times with TBST (GenePool/GPP1822). Target protein bands were visualized using the ImageQuantLAS4000 chemiluminescence imaging system, and densitometry was performed using the Image J 1.80.

Quantification and statistical analysis

All data were transformed to mean ± standard error of the mean (S.E.M.), and Sangerbox 3.0 (http://vip.sangerbox.com/home.html) was used to preform functional annotation and pathway enriched analysis based on Gene Ontology (GO) and the Kyoto Encyclopedia of Genes and Genomes (KEGG). MetaboAnalyst (https://www.metaboanalyst.ca/) was used for lipidomics analysis. To test the statistical significance of differences between groups, we used t-tests, and we used one-way ANOVA to analyze the differences between multiple groups. GraphPad Prism 8.0.1 was used for all analysis, and we consider a test result to be statistically significant if its two-sided p-value is less than 0.05.

Published: November 18, 2022

Data and code availability

• •All data reported in this paper will be shared by the lead contact upon request.
• •All data obtained by mass spectrometry in this study have been submitted to the PRIDE archive with accession number PXD036478.
• •This paper does not report original code.
• •Any additional information required to reanalyze the data reported in this paper is available from the lead contact on request.