Abstract Saponins from bitter melon (BMS) are well-known to have various biological activities, especially in the field of fat-lowering. However, many gaps remain in our knowledge of BMS-induced fat reduction and health benefits. Here, we aimed to investigate the precise mechanism of BMS in alleviating fat accumulation in C. elegans and HepG2 cell line. Results indicated that BMS showed strong fat-lowering and lifespan-extension properties. Lipidomic analysis illustrated that BMS could alter the lipid profile, especially represented by phosphatidylethanolamine (PE) increase, which plays an essential role in autophagy. Furthermore, we applied gene-deficient mutants and RNAi technology to confirm that BMS largely depended on daf-16/FoxO1 and hlh-30/TFEB mediated lipophagy to reduce fat deposition. In addition, BMS could ameliorate oil acid (OA)-induced fat accumulation in HepG2 cells by induction of autophagy-related proteins, such as the phosphorylated AMPK and LC3B. In conclusion, our results elucidated the underlying mechanism of bitter melon saponins interfering with lipid metabolism from the autophagy point of view, which provide new insights into a nutraceutical to mitigate obesity. Keywords: Bitter melon saponin, Fat accumulation, Lipidomics, Autophagy, Daf-16/ hlh-30 Graphical abstract We aimed to investigate the precise mechanism of saponins from bitter melon (BMS) in alleviating fat accumulation in C. elegans. Results indicated that BMS showed strong fat-lowering and lifespan-extension properties. Lipidomic analysis illustrated that BMS could alter the lipid profile, especially represented by PE increase, which plays an essential role in autophagy. Furthermore, we applied gene-deficient mutants and RNAi technology to confirm that BMS could inhibit fat deposition through daf-16/FoxO1 and hlh-30/TFEB mediated lipophagy. In addition, BMS could ameliorate oil acid (OA)-induced fat accumulation in HepG2 cells by induction of AMPK phosphorylation and LC3B expression, which were involved in autophagy. Thus, we concluded that BMS could ameliorate fat accumulation by inducing autophagy. [45]Image 1 [46]Open in a new tab Highlights * • Bitter melon saponin (BMS) could inhibit fat accumulation and extended the lifespan of C. elegans. * • Lipidomics analysis predicted autophagy may be a key pathway involved in the fat-lowering effects of BMS. * • BMS induced daf-16/hlh-30 mediated lipophagy to confer fat-lowering benefit. * • BMS regulated autophagy via activating AMPK phosphorylation and LC3B expressions in HepG2 cells. 1. Introduction Bitter melon (Momordica charantia L.) is recognized as a medicinal and edible plant for obesity and diabetes prevention due to its numerous active ingredients, like polysaccharides, saponins, peptides or flavonoids ([47]Bai et al., 2018). Saponins, the main source of bitter taste in bitter melon, are reported to hold similar hypoglycemic property with insulin, and the underlying mechanisms include the activation of AMPK and insulin receptor-1 (IRS-1), promotion of downstream glucose transporter (GLUT4), or elevation of intracellular glycogen synthase kinase-3 β (GSK-3β) ([48]Han et al., 2018; [49]Tan et al., 2016). In addition, BMS could inhibit the differentiation of preadipocytes by down-regulating PPARγ in 3T3-L1 preadipocytes, and effectively reduce plasma lipid levels in rat ([50]Senanayake et al., 2012). However, the detailed mechanism of BMS in fat-lowering effect remains elusive. Caenorhabditis elegans (C. elegans), a whole-organism, has recently gained considerable interest in studying the molecular mechanism of lipid metabolism and aging due to their short lifespan, broad availability of mutants and ease of genetic manipulation ([51]Bai et al., 2021). Commonly, researches in the C. elegans lipid field have been ongoing to characterize lipid biosynthetic genes, regulatory genes affecting lipid synthesis, storage, and breakdown with some specific lipids alteration, such as triacylglycerol (TAG), phosphatidylethanolamine (PE), or phosphatidylcholine (PC), which are considered crucial for membrane integrity and signaling pathways ([52]Kimura et al., 2016). In order to identify and quantify the lipid species relevant for various biological processes, lipidomic technologies have been largely used to characterize the lipid profile of C. elegans, through the analysis of structure, function, or interaction of cellular lipids, which also play an essential role in nutritional research ([53]Kim et al., 2019). In the process of lipid hydrolysis, the lipid autophagy (i.e. lipophagy) connects autophagy and lipid metabolism, equally important as lipolysis. Lipid droplets (LDs) are cytoplasmic organelles containing mostly TAG, surrounded by a monolayer mainly of PC and PE. AMPK (homolog of aak-2 in C. elegans) is an energy sensor modulating fat catabolism, including fatty acid oxidation, lipolysis and autophagy. Adipose Triglyceride Lipase (ATGL) is a well-known cytosolic lipase to break down TAG into FFAs and glycerol, which means lipolysis ([54]Obrowsky et al., 2013; [55]Li et al., 2022). Except for lipolysis, LDs can also be accessed by lipophagy, a specific subset of selective autophagy that catabolizes the components of LDs. Lipophagy-associated genes are primarily controlled by the master regulator of lysosomal biogenesis transcription factor EB (TFEB)/helix-loop-helix (hlh)-30, the forkhead homeobox transcription factor FoxO1/daf-16 as well, which are required for LD clearance in multiple systems ([56]Gary et al., 2015). In order to investigate the precise mechanism of BMS inhibiting fat accumulation, we applied the lipidomics to uncover another pathway additive to that of fat synthesis. As expected, the lipidomic analysis predicted that autophagy was involved in the effects of BMS, characterized by the regulation of daf-16/hlh-30 mediated lipophagy. In addition, BMS increased the level of autophagy in HepG2 cell lines with OA treatment. This study can enrich the mechanism of the biological activity of bitter melon saponins, and provide potential applications in the field of food nutrition. 2. Materials and methods 2.1. Chemicals and reagents Fresh bitter melons were collected from Lvjian Agricultural Station (Yangzhong City, China) and authenticated by Jiangsu Academy of Agricultural Science. HepG2 cell was purchased from Shanghai institute of biochemistry and cell biology. Oil Red O staining, triglyceride (TG) reagents and protein assay kit were purchased from Nanjing Jiancheng Bioengineering Institute (Nanjing, China). Levamisole hydrochloride and chemicals for C. elegans cultivation were purchased from Shanghai Alighting Biochemical Technology Co., LTD (Shanghai, China). Reagents for qRT-PCR were from Takara Bio Inc. (DaLian, China). 2.2. BMS preparation and chemical constituents analysis Fresh and unripe bitter melons were washed thoroughly and the seeds were removed. After sliced thinly, bitter melon slices were frozen-dried and milled to the powder (diameter < 105 μm). BMS was prepared as previously described ([57]Bai et al., 2020b), as the following: the milled powder was added with 75% ethanol for reflux extraction twice at 80 °C, and the filtrate was collected. Then n-butanol saturated with water was applied to extract from the filtrate for 3 times until the concentrated n-butanol phase was brown and sticky. Then methanol and acetone were added to produce precipitation, which was lyophilized as the BMS. In our study, the saponin content in the stocking solution dissolved by ethanol was 54.60 μg/mL ginsenoside Rg1 equivalent detected by vanillin–acetic acid method ([58]Lin et al., 2020). Stock solution of BMS was prepared in ethanol at 10 mg/mL and stored at −20 °C. Next, we applied ultra-high performance liquid chromatography tandem with high resolution mass spectrometry (UPLC-HRMS) (Thermo-Fisher Scientific, USA) to analyze the components, following the previous literature ([59]Zhao et al., 2021). Using ginsenoside Rg1 as internal standard, 7 main components of saponins from bitter melon were identified ([60]Figure S1 A). Most of the saponins were cucurbitane triterpenoids, including momordicosides P (RT=7.312) and L (RT=1.081), Kuguacin N (RT=9.065) and momordicine I (RT=8.397). The basic structure of cucurbitane aglycone was preliminarily drawn ([61]Figure S1 B). 2.3. C. elegans strains and cultivation Escherichia coli OP50 and C. elegans strains including N2, Bristol (wild type), CF1038 (daf-16 (mu86)), JIN1375 (hlh-30 (tm1987) IV), TJ356 (DAF-16:GFP), DA2130 (LGG-1:GFP) were obtained from Caenorhabditis Genetics Center (CGC) (Minneapolis, MN, USA). Worms were propagated on nematode growth medium (NGM) spread with E. coli OP50 bacterial. Experiments were performed at 20 °C. L1 worms were synchronized by using bleach and NaOH according to previously describe ([62]Bai et al., 2020a), which were applied in all assays. Synchronized L1 worms were supplemented with ddH[2]O as control or BMS (100 μg/mL or 200 μg/mL, final concentrations) mixed with E. coli OP50 on NGM plate. 2.4. Oil red O (ORO) staining and TG assay Synchronized L1 worms were treated with BMS from L1 to late L4 stage. After 48 h, 300-500 adult worms in each group were washed with M9 buffer for three times, followed by fixation and dehydration with 60% isopropanol for 15 min. ORO stocking solution (5 mg/mL) was diluted with ddH[2]O to 60% working solution and filtered to dye the fixed worms for 2 h at room temperature. Next, worms were washed with M9 three times to remove the extra dye, then mounted and imaged with a microscope outfitted with DIC optics (Leica, Wetzlar, Germany). Image J was used to quantify the ORO intensities ([63]Xiao et al., 2020a). For TG quantification, worms (∼1000 worms each group) were washed from NGM for three times with M9 buffer to collect worm pellets, and homogenized using ultrasonic disruption to release the triglyceride, followed by centrifugation at 8000 rpm for 5 min to obtain the supernatant for TG assay. TG content was normalized by the protein content, which was determined according to the instructions of Coomassie brilliant blue kit. 2.5. Body size, locomotive activity, growth rate and lifespan assay Worm sizes were represented by the body length and width. C. elegans were treated with BMS from L1 to L4 larvae. Photographs were taken of the worms after anesthetization by 100 μM levamisole hydrochloride. The body length and body width of each worm (more than 30 worms) was quantified by using Image J software. The test was performed at least 3 times. For locomotive behavior, head thrashes and body bends were assessed. C. elegans were treated with BMS from L1 to L4 larvae. For thrashing frequency of the head, the head of worms from one side to the other and then back within 1 min on clean NGM was recorded under a stereo microscope. For the frequency of body bends, worms placed on a NGM without E. coli forwarded body walks in a shape of wavelength, and the number of wavelength shapes within 1 min worms were recorded. Twenty nematodes in each group were examined per experiment. For the growth rate, the number of worms at each developmental stage after 48 h was counted (n > 100 worms per group) as previously described ([64]Shen et al., 2017). Lifespan assay was performed at 20 °C as previously described ([65]Bai et al., 2020a). L1 worms were cultivated to L4 stage on NGM and then transferred to the Transwell®-24 well permeable plates (BD Biosciences Inc., NY, USA) supplemented with ddH[2]O or BMS (100 and 200 μg/mL) and 2% glucose. Fluorodeoxyuridine (FUDR, 120 mM) was added to prevent eggs from hatching after worms reached L4 stage/young adult stage during the treatment period. The medium was changed every two days. Survivals were recorded every other day until all the worms died. The day when we treated the L4 worms was defined as day 0 of adult age. 2.6. Detection of proteins with GFP labeled expressions Synchronized transgenic worms and DAF-16:GFP (TJ356) and DA2130 (LGG-1:GFP) were treated with BMS for 48h (∼500 worms per each group), then anesthetized with 100 μM levamisole hydrochloride at room temperature and put onto a glass slide. Images were captured by an Olympus IXplore (Olympus Sales Service Co., Ltd. Beijing, China) fluorescence microscope equipped with a GFP (emission 500–515 nm) filter (10 × magnification). Three independent experiments were performed. Quantification of fluorescence intensity was evaluated with the ImageJ (NIH) software to measure the percentage of GFP-positive nuclei in transgenic worms. 2.7. Lipidomic profiling by LC-MS/MS and data analysis For lipidomics, worms (∼10000 worms each duplicate, 6 replicates per group) were transferred from NGM and washed five times with M9 buffer to collect worm pellets. 480 μL extract solution (MTBE: MeOH= 5: 1) was added sequentially, then homogenized at 35 Hz for 4 min (Ultra-Turraz homogenizer IKA LaborTechinik) and sonicated for 5 min (IKA LaborTechnik) in ice-water bath, which were repeated for 3 times. Then the samples were incubated at −40 °C for 1 h and centrifuged at 3000 rpm for 15 min at 4 °C. 300 μL of supernatant was transferred to a fresh tube and dried in a vacuum concentrator at 37 °C. Then, the dried samples were reconstituted in 200 μL of 50% methanol in dichloromethane by sonication for 10 min in ice-water bath. The constitution was then centrifuged at 13000 rpm for 15 min at 4 °C, and 75 μL of supernatant was transferred to a fresh glass vial for LC/MS analysis. The quality control (QC) sample was prepared by mixing an equal aliquot of the supernatants from all of the samples. The UHPLC separation was carried out using an ExionLC Infinity series UHPLC System (AB Sciex), equipped with a Kinetex C18 column (2.1 * 100 mm, 1.7 μm, Phenomen). The mobile phase A consisted of 40% water, 60% acetonitrile, and 10 mmol/L ammonium formate. The mobile phase B consisted of 10% acetonitrile and 90% isopropanol, which was added with 50 mL 10 mmol/L ammonium formate for every 1000 mL mixed solvent. The analysis was carried with elution gradient as follows: 0–12.0 min, 40%–100% B; 12.0–13.5 min, 100% B; 13.5–13.7 min, 100%–40% B; 13.7–18.0 min, 40% B. The column temperature was 55 °C. The auto-sampler temperature was 6 °C, and the injection volume was 2 μL (pos) or 4 μL (neg), respectively. The TripleTOF 5600 mass spectrometer was used for its ability to acquire MS/MS spectra on an information-dependent basis (IDA) during an LC/MS experiment. In this mode, the acquisition software (Analyst TF 1.7, AB Sciex) continuously evaluates the full scan survey MS data as it collects and triggers the acquisition of MS/MS spectra depending on preselected criteria. In each cycle, the most intensive 12 precursor ions with intensity above 100 were chosen for MS/MS at collision energy (CE) of 45 eV (12 MS/MS events with accumulation time of 50 msec each). ESI source conditions were set as following: Gas 1 as 60 psi, Gas 2 as 60 psi, Curtain Gas as 30 psi, Source Temperature as 600 °C, Declustering potential as 100 V, Ion Spray Voltage Floating (ISVF) as 5000 V or −3800 V in positive or negative modes, respectively. With the MS/MS spectrum, lipid identification was achieved through a spectral match using LipidBlast library. Dataset containing the information of peak number, sample name and normalized peak area was imported to SIMCA16.0.2 software package (Sartorius Stedim Data Analytics AB, Umea, Sweden) for multivariate analysis, including principle component analysis (PCA), orthogonal projections to latent structures-discriminate analysis (OPLS-DA). The value of variable importance in the projection (VIP) of the first principal component in OPLS-DA analysis was obtained. It summarizes the contribution of each variable to the model. The metabolites with VIP>1 and p<0.05 (student t-test) were considered as significantly changed metabolites. In addition, commercial databases including KEGG ([66]http://www.genome.jp/kegg/) and MetaboAnalyst ([67]http://www.metaboanalyst.ca/) were used for pathway enrichment analysis. 2.8. Quantitative real time reverse-transcription PCR (qRT-PCR) Total RNA of worms (∼3000 worms per sample) was extracted by a TaKaRa MiniBEST Universal RNA extraction kit (Takara Bio, DaLian, China), which was reverse-transcribed to cDNA using a PrimeScript RT Master Mix kit (Takara Bio, DaLian, China). Quantification of gene expression by real-time PCR conducted on a CFX96 real-time PCR detection system (BioRad, California, America) based on the SYBR Premix Ex Taq kit (Takara Bio, DaLian, China). Primer sequences used in this study were listed in [68]Table S1, and act-1 was used as an internal control. Quantitative PCR amplification for all of the genes included pre-incubation at 95 °C for 10 s, followed by 40 cycles at 95 °C for 5 s, and 95 °C for 30 s, and then calculation of the fold change by the 2^-ΔΔCt method ([69]Xiao et al., 2020b). 2.9. RNA interference (RNAi) E. coli HT115 (DE3) harboring empty vector L4440 or expressing double stranded RNA corresponding to hlh-30 was transferred into LA broth containing isopropyl 1-thio-β-d-galactopyranoside (IPTG, 5 mM). L1-larvae were fed with E. coli HT115 carrying double stranded RNA for hlh-30 gene or harboring empty vector L4440 as the control ([70]Yang et al., 2020). Once they developed into gravid, worms were transferred to a fresh RNAi plate to lay eggs and obtain the second generation for treatment of BMS. The primer of hlh-30 for RNAi was forward: CCGCTCGAGTCTTCCCATCTATTCACGGC; Reverse: CCCAAGCTTCTCTGATGTGTTCTTCGGCA. qRT-PCR was employed to confirm the RNAi efficiency ([71]Fig. S3). 2.10. HepG2 cell culture and western blotting analysis The cells were cultured in monolayers up to 80% confluence in DMEM supplemented with 10% heat-inactivated fetal calf serum and 1% penicillin/streptomycin at 37 °C in a humidified incubator supplied with 5% CO[2.] Cells in the logarithmic growth phase were used for all the studies described below. 0.5 mM oleic acid (OA)was used for 24 h to induce fat accumulation of HepG2 cells as the model group ([72]Song et al., 2022), and BMS was treated for another 24 h. Proteins of HepG2 cells were extracted by RIPA lysate with protease and phosphatase inhibitors (Beijing Solarbio Science & Technology Co., Ltd., Beijing, China) and centrifuged at 12000 rpm for 20 min to obtain the supernatant. Protein concentrations were determined by BCA assay kit (Nanjing Jiancheng Bioengineering Institute, Nanjing, China). Briefly, the lysate in loading buffer (30 μg protein/lane) was separated by sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) and transferred to a polyvinylidene fluoride (PVDF) membrane. The membranes were then blocked and probed with antibodies against AMPK, p-AMPK, LC-3B and β-actin (Cell Signaling Technology, Inc., USA), and incubated with the membranes at 4 °C overnight. Bands were detected using electrochemiluminescence (ECL) reagents (Millipore) according to the manufacturer's instructions. All primary and secondary antibodies were obtained from Abcam (Abcam, Cambridge, MA). Quantity One software (BioRad) was used to quantify band densities. 2.11. Statistical analysis All data were expressed as means ± standard error (SE). Statistical analysis was performed using one-way analysis of variance, followed by Tukey's multiple range test to compare between groups. The significance of differences was defined at the p < 0.05 level. 3. Results and discussion 3.1. BMS decreased fat accumulation and extended lifespan in C. elegans To understand the function of BMS in lipid metabolism, we firstly estimated the effect of BMS on fat deposition in C. elegans judged by ORO. As shown in [73]Fig. 1A and B, the dosages of 100 and 200 μg/mL BMS we have reported ([74]Bai et al., 2020b) could profoundly decrease the overall fat compared to the control group, accompanied by the reduction of TG content. Consistent with the study of [75]Lin et al. (2020), the inhibitory effects of momordica saponins extracts on fat accumulation in C. elegans were also observed. Fig. 1. [76]Fig. 1 [77]Open in a new tab Effects of BMS on overall fat accumulation and physiological indicators in wild-type (N2) C. elegans. Groups were divided into Control, BMS at 50 μg/mL, 100 μg/mL and 200 μg/mL, respectively. Worms were treated from L1 to L4 stage for 48 h followed by ORO staining (A) and TG assay (B). Fat intensities were analyzed by Image J from 15 to 20 worms (n = 3 sets/group). For the impacts of BMS on worm size, locomotive activities, growth rate and lifespan, L1 worms were treated with BMS (100 μg/mL and 200 μg/mL) for 48 h. Worm length(C) and worm width (D) were analyzed by Image J software (n > 30 worms per group). Head thrashes(E) and body bends (F) were counted under a microscope for 30 s (n > 30 worms per group). (G) The total number of worms at each developmental stage was counted (n > 100 per group). YA: young adult. (H) Lifespan was conducted from late L4 and the number was counted until all dead (Control, n = 126; BMS 100, n = 108; BMS 200, n= 138). Data were expressed as mean ± S.E. Different letters (a–b) denote meant values that were statistically different at P < 0.05. Since lipid metabolism is linked to changes in physiology, the growth and the survival of C. elegans ([78]Sun et al., 2016), we next determined whether BMS treatment we used affected worm size, locomotive behavior, growth rate and lifespan. Results indicated that BMS at both 100 and 200 μg/mL showed no effects on worm length and width, the locomotive activities as an indicator of energy expenditure including head thrashed and body bends as well ([79]Fig. 1C–F). In addition, no significant effects were observed in the growth rate of worms treated with BMS ([80]Fig. 1G). Surprisingly, worms treated with 100 and 200 μg/mL BMS exhibited anti-aging characterized by increased mean lifespan by 16.64% and 13.83% ([81]Fig. 1H and [82]Table 1, p < 0.001), respectively. In accordance with Lin's findings ([83]Lin et al., 2020), the ability of momordica saponins extracts improving lifespan was attributed to the stress resistance effect. Beyond that, we speculated that the altered lipid metabolism by BMS could make an essential impact on the extension of longevity, in which the mechanism needed further confirmation. Table 1. Effects of BMS on survival rate in C. elegans. Strains BMS (μg/mL) No. Animals Median Lifespan (days) Mean lifespan (days±SE)^a MaximumLifespan (days) P-value^b N2 0 126 18 17.07 ± 0.35 24 ＜0.001 100 108 20 19.91 ± 0.36 26 200 138 20 19.43 ± 0.35 26 ＜0.001 daf-16 0 160 8 8.81± 0.3 18 100 144 6 7.13± 0.2 18 <0.0001 hlh-30 0 160 12 11.71±0.46 26 100 129 12 12.14±0.47 26 0.8026 hlh-30 (RNAi) 0 110 20 18.02±0.5 26 100 109 14 14.99±0.57 26 0.0009 [84]Open in a new tab Worms were treated with BMS from the late L4 stage. FUDR (120 μM) was added to sterilize the worms during the treatment period. Survivals were recorded every other day until all of the worms died. ^a The mean lifespan was referred to the exact time when the survival rate reached to 50%, which was generated by the OASIS application. ^b P-values were analyzed by log-rank (Mantel-Cox method) tests. Significant differences were defined at p< 0.05. 3.2. Differential metabolites of lipidomics predicted autophagy may be a key pathway involved in the fat-lowering effects of BMS In C. elegans, animal development, metabolism, and lifespan are complexly connected with lipid molecules, which are regarded essential for signaling pathways and membrane integrity ([85]Kimura et al., 2016). Herein, after confirmation of the fat-lowering impact of BMS, we further conducted the lipidomics to overview the effects of 100 μg/mL BMS on the profile of lipids in C. elegans. In all, 1190 and 1245 metabolites with identified names were detected in the positive and negative ion modes, respectively ([86]Fig. S2). OPLS-DA was applied to examine changes in lipid metabolites patterns between the control and BMS groups as well as identify key metabolites that contributed to metabolic pattern changes. OPLS-DA indicated that the two groups formed clear clusters based on the lipids in both positive and negative modes ([87]Fig. 2A and B), and the cross-validation showed that the original model had good robustness and there is no overfitting phenomenon (R^2 = 0.995, Q^2 = 0.455 in positive; R^2 = 0.990, Q^2 = 0.365). Fig. 2. [88]Fig. 2 [89]Open in a new tab Lipidomics analysis of C. elegans treated with BMS. L1 worms (∼10000 worms each duplicate, 6 replicates per group) were treated with BMS (100 μg/mL) for 48 h. (A) OPLS-DA plots in positive ion mode, (B) OPLS-DA plots in negative ion mode, (C) Heatmaps in positive ion mode, (D) Heatmaps plots in negative ion mode, (E) Bubble plots in positive ion mode, (F) Bubble plots in negative ion mode (dots in grey indicated metabolites without significance), and (G) PE involved autophagy pathway in KEGG analysis (PE was indicated as purple dots). (For interpretation of the references to colour in this figure legend, the