Peripheral cathepsin L inhibition induces fat loss in C. elegans and mice through promoting central serotonin synthesis
Cathepsin L and some other cathepsins have been implicated in the development of obesity in humans and mice. The functional inactivation of the proteases reduces fat accumulation during mammalian adipocyte differentiation. However, beyond degrading extracellular matrix protein fibronectin, the molecular mechanisms by which cathepsins control fat accumulation remain unclear. We now provide evidence from Caenorhabditis elegans and mouse models to suggest a conserved regulatory circuit in which peripheral cathepsin L inhibition lowers fat accumulation through promoting central serotonin synthesis.
We established a C. elegans model of fat accumulation using dietary supplementation with glucose and palmitic acid. We found that nutrient supplementation elevated fat storage in C. elegans, and along with worm fat accumulation, an increase in the expression of cpl-1 was detected using real-time PCR and western blot. The functional inactivation of cpl-1 reduced fat storage in C. elegans through activating serotonin signaling. Further, knockdown of cpl-1 in the intestine and hypodermis promoted serotonin synthesis in worm ADF neurons and induced body fat loss in C. elegans via central serotonin signaling. We found a similar regulatory circuit in high-fat diet-fed mice. Cathepsin L knockout promoted fat loss and central serotonin synthesis. Intraperitoneal injection of the cathepsin L inhibitor CLIK195 similarly reduced body weight gain and white adipose tissue (WAT) adipogenesis, while elevating brain serotonin level and WAT lipolysis and fatty acid β-oxidation. These effects of inhibiting cathepsin L were abolished by intracranial injection of p-chlorophenylalanine, inhibitor of a rate-limiting enzyme for serotonin synthesis.
This study reveals a previously undescribed molecular mechanism by which peripheral CPL-1/cathepsin L inhibition induces fat loss in C. elegans and mice through promoting central serotonin signaling.
KeywordsPeripheral cathepsin L Central serotonin Fat accumulation C. elegans Mice
White adipose tissues
Nematode Growth Medium
- L4 stage
Fourth larval stage
- Cstl−/− mice
Cathepsin L-deficient mice
- Cstl+/+ mice
Littermate control mice
- L1 stage
First larval stage
Lysosomal cathepsins play an important role in various physiological processes and diseases, such as adaptive immunity, rheumatoid arthritis, cardiovascular diseases, and cancer . Over the past 15 years, some of the proteinases have been implicated in the development of obesity through controlling adipogenesis. The circulating levels of cathepsin L and S, which correlate significantly with body weight indexes, significantly increase in clinical obese individuals [2, 3, 4]. Obese humans and mice have higher expressions of cathepsin D, K, L, and S in white adipose tissues (WAT) than lean controls [2, 3, 5, 6, 7, 8]. In obese patients and/or mice undergoing weight loss, the reduction in body weight is accompanied by the decline in WAT and circulating levels of cathepsin K, L, and S [3, 4, 6, 8]. In vitro studies show that the expressions of cathepsin D, K, L, and S are elevated along with fat accumulation during human and murine adipocyte differentiation, and cathepsin inhibition can reduce adipogenesis of the cells [2, 5, 7, 9, 10]. In contrast, cathepsin L and K overexpression in 3T3-L1 pre-adipocytes and the treatment of pre-adipocytes with human recombinant cathepsin S enhance adipogenesis [2, 9, 10]. Utilizing genetic deficiency and pharmacological inhibition of cathepsin K and cathepsin L, our previous studies demonstrate their essential roles in mouse adipogenesis and body weight gain and attribute the actions to the degradation of extracellular matrix protein fibronectin [2, 9]. However, beyond degrading fibronectin, the detailed molecular mechanisms that cathepsins control fat accumulation remain unclear.
Over the past decade, Caenorhabditis elegans has emerged as a powerful animal model for exploring the molecular mechanism of fat metabolism and obesity [11, 12, 13]. As the first wholly sequenced multicellular organism, C. elegans shares at least 83% of its protein sequences with humans  and over 70% of its lipid genes have human orthologs . Most importantly, its core lipid metabolic pathways, such as insulin, TOR, serotonin, dopamine, and glutamate pathways, are highly conserved with mammals . Moreover, genetic tractability and transparent bodies of C. elegans make it highly suitable for investigating the complicated molecular pathways and tissue interaction with respect to fat metabolism. The possible involvement of worm cathepsin-like protein in fat storage in C. elegans remains untested.
In this study, we firstly established a C. elegans model of fat accumulation and utilized it to uncover that the expression of cathepsin L-like protease CPL-1 increased along with nutrient-induced fat accumulation in C. elegans. Moreover, we found that mutation and knockdown of cpl-1 reduced fat storage in C. elegans. Further, a feedback circuit, which CPL-1 inhibition in the intestine and hypodermis promoted central serotonin synthesis to induce fat loss, was demonstrated in C. elegans. Finally, this conserved circuit, linking the peripheral fat storage tissue and the central nervous system, was also confirmed in mice fed with a high-fat diet (HFD).
The supplementation of glucose or palmitic acid elevated fat storage and the expression of cathepsin L-like protease CPL-1 in C. elegans
During mammalian preadipocyte differentiation, the expressions of multiple cathepsins are also increased along with fat accumulation . To understand whether their homologous proteases were involved in nutrient-induced fat accumulation in C. elegans, we used real-time PCR to detect the expressions of candidate cathepsin-like genes in N2 worms grown on standard NGM plates and NGM plates with 5 mM glucose. As shown in Additional file 4: Table S2, 5-mM glucose supplementation significantly affected the expression of six genes, including cpr-1(C52E4.1), cpr-4(F44C4.4), cpr-5(W07B8.5), cpr-6(C25B8.3), asp-12(F21F8.4), and cpl-1(T03E6.7). And cpl-1(T03E6.7) demonstrated the highest induction in response to 5-mM glucose supplementation (Additional file 5: Figure S3A and Additional file 4: Table S2). Further, we examined the expressions of these six genes after glucose (at 1 mM and 5 mM) or palmitic acid (at 0.02 mM and 0.2 mM) treatments. Among the genes, only the expression of cpl-1(T03E6.7) was significantly elevated by nutrient supplementation in a dose-dependent manner (Additional file 5: Figure S3B). To examine the changes of mature (at 25 kD) and pro-enzyme (at 39 kD) protein of CPL-1, we further performed western blot analysis using anti-CPL-1 polyclonal antibodies. Supporting the results in real-time PCR analysis (Additional file 5: Figure S3 and Additional file 4: Table S2), the supplementation of glucose and palmitic acid also significantly elevated the levels of mature CPL-1 in a dose-dependent manner (Fig. 1e). The results suggest a substantial role of CPL-1 in fat accumulation in C. elegans.
Functional inactivation of CPL-1 suppressed fat accumulation in C. elegans
Functional inactivation of CPL-1 reduced fat storage through activating serotonin signaling in C. elegans
Knockdown of cpl-1 in the intestine and hypodermis promoted serotonin synthesis in ADF neurons in C. elegans
In C. elegans, fat depots are primarily stored in the intestinal and skin-like epidermal/hypodermal cells . As a widely distributed protein in worms, CPL-1 is also expressed in the intestine and hypodermis . Here, we introduced an extrachromosomal array (yqEx688, Pcpl-1cpl-1::mChOint) expressing CPL-1 fusion mChOint fluorescence protein into the RT258 (unc-119(ed3)III;pwls50) strain. The punctate of CPL-1::mChOint overlapped very well with the GFP-tagged lysosomal membrane protein LMP-1 (LMP-1::GFP) in the intestinal and hypodermal cells (Additional file 11: Figure S7), confirming the location of CPL-1 in lysosomes of the cells.
As a classical neurotransmitter, serotonin is only produced from several neurons in C. elegans. In the neurons, including head sensory neuron ADF, pharyngeal neuron NSM, and hermaphrodite-specific neuron HSN, tryptophan hydroxylase TPH-1 as a critical rate-limiting enzyme is required for serotonin biosynthesis . Thus, to assess the effects of CPL-1 inhibition in the intestine and hypodermis on central serotonin synthesis, we crossed the Ptph-1::gfp line with N2, the VP303, or the NR222 strains and performed knockdown of cpl-1 in the whole body, intestine, and hypodermis, respectively. Notably, an increased expression of TPH-1 was observed in ADF neurons in the cpl-1 knockdown worms (Fig. 4e). In C. elegans, serotonin from ADF neurons and serotonin from NSM neurons specifically regulate distinct physiological functions. Unlike the specific role of serotonin from NSM neurons in foraging regulation [22, 23], serotonin from ADF neurons specifically regulates body fat loss, feeding, and olfactory learning ability for avoiding pathogenic bacteria [20, 24, 25]. Consistent with this notion, we found that mutation of cpl-1 and knockdown of cpl-1 in the whole body, intestine, and hypodermis not only induced body fat loss (Figs. 2b, d and 4a) but also elevated worm pumping rates (Figs. 3d, h and 4f) and olfactory learning abilities for avoiding pathogenic bacteria (Figs. 3e, i and 4g). The results suggest that knockdown of cpl-1 in the intestine and hypodermis resembles systemic CPL-1 inactivation and induces body fat loss in C. elegans by promoting serotonin synthesis in ADF neurons.
Knockdown of cpl-1 induced body fat loss via a central serotonin signaling
Cathepsin L knockout promoted fat loss and central serotonin synthesis in HFD-fed mice
Like LFD-fed Ctsl−/− mice, HFD-fed Ctsl−/− mice also had a higher level of food intake than HFD-fed Ctsl+/+ mice (Additional file 14: Figure S10C). However, the HFD-fed Ctsl−/− mice had a lower body (Fig. 6b) and WAT weight gain and fat accumulation (Fig. 6c, d) than HFD-fed Ctsl+/+ mice. Further, we detected the energy expenditure in mice. In the HFD-fed mice, despite there were no differences in O2 consumption and CO2 production between Ctsl−/− and Ctsl+/+ mice when the metabolic data were normalized per animal (Additional file 15: Figure S11 A and S11B), Ctsl−/− mice had a significantly higher metabolic rates than Ctsl+/+ mice when the data were normalized to body weight (Fig. 6e, f). Correspondingly, the expressions of genes responsible for lipolysis and fatty acid β-oxidation (Fig. 6g) were enhanced in WAT in HFD-fed Ctsl−/− mice. Notably, compared with HFD-fed Ctsl+/+ mice, HFD-fed Ctsl−/− mice had higher brain serotonin levels (Fig. 6h) and central tryptophan hydroxylase Tph2 expression (Fig. 6i). Together, the results indicate that a genetic deficiency of cathepsin L enhances brain serotonin production and promotes fat loss in HFD-fed mice, suggesting a circuit including peripheral cathepsin L inhibition and central serotonin synthesis may also induce fat loss in HFD-fed mice.
Peripheral inhibition of cathepsin L induced fat loss in HFD-fed mice through promoting central serotonin synthesis
In mammals, adipose tissue is a primary fat storage organ. WAT lipometabolic imbalance, especially excess fat accumulation, is closely related to the development of obesity and associated insulin resistance, diabetes, and cardiovascular disease [27, 28, 29]. Thus, it is crucial to understand the regulation of fat storage or loss. In the present study, we select a classic model animal C. elegans to establish a model of fat accumulation. Further, utilizing to the model, we demonstrate an increased expression of lysosomal cathepsin L-like protease CPL-1 along with nutrient-induced fat accumulation and hence reveal a previously undescribed molecular circuit that CPL-1 inhibition in fat storage tissues promotes serotonin production in neurons and, in turn, induces body fat loss in C. elegans (Fig. 7f, left). Finally, a similar circuit is also recapitulated in HFD-fed mice (Fig. 7f, right).
Long-term excess energy intake, for example, an increase in dietary saturated fats and sugars, fundamentally results in the development of obesity and comorbidity metabolic diseases in mammals. Glucose is an essential nutrient and the main component of metabolism and energy production. In C. elegans, like its action in mammals, dietary glucose supplementation increases fat accumulation [30, 31] and hence leads to some pathophysiological changes, such as accelerated aging and life shorting [32, 33]. Palmitic acid is one of the most common saturated fatty acids in the human body  and the primary saturated fatty acid in E. coli OP50 membrane . Therefore, in this study, to establish a C. elegans model of nutrient-induced fat accumulation, glucose and palmitic acid are supplemented into the NGM plates.
Across all metazoans, fat accumulation is conserved controlled by a series of critical enzymes that catalyze multistep conversions from acetyl CoA to TAG . In C. elegans, some key rate-limiting enzymes, such as acetyl CoA carboxylase and fatty acid synthase, are expressed at all developmental stages, and their deficiency can result in early larval arrest, suggesting fat accumulation may be an essential requirement for larval and adult normal growth . On the other hand, the formation of dauer, which is a phenotype of post-embryonic developmental arrest in C. elegans, is characterized by increased fat accumulation and altered metabolism . Thus, in C. elegans, fat accumulation and post-embryonic development rate may influence each other. To avoid the influence, we must develop a C. elegans model with increased fat deposition and without altered developmental rate. In the present study, as shown in Fig. 1a, a short-term nutrient supplement from L4 to adult is performed. Utilizing such a short-term supplement, both glucose and palmitic acid induce fat accumulation and increase cpl-1 expression in N2 worms, yet they do not affect worm developmental rate (Fig. 1, Additional file 5: Figure S3, and Additional file 2: Table S1).
As a proteolytic enzyme localized in lysosomes, CPL-1 has been implicated in embryogenesis, development, and phagosomal degradation of apoptotic germ cells in C. elegans [16, 17]. In the previous studies, to enucleate the relative role of CPL-1 in worm development and embryogenesis, several cpl-1 mutants have been utilized. The embryos of cpl-1 mutant ok360 are lethal, suggesting that CPL-1 is essential for the degradation of yolk proteins during worm embryogenesis . However, high embryonic lethality does not mean abnormal post-embryonic development. Due to maternal providence of CPL-1, the heterozygous cpl-1(ok360) F1 progeny usually develops to adults without visible defects in growth rate, molting, and brood size [16, 17]. Moreover, two weak loss-of-function mutants of cpl-1, qx304, and yq89 also develop normally when cultured at 20 °C . Consistent with the previous observations, all tested worms in this study, including qx304 and yq89 mutants, homozygous cpl-1(ok360) F1 progenies generated by heterozygous mothers and cpl-1 knockdown worms, develop normally when cultured at 20 °C (Additional file 6: Table S3 and Additional file 7: Table S4). Utilizing the normally developed cpl-1-deficient and knockdown worms, we demonstrate that CPL-1 inhibition reduces basal and nutrient-induced fat accumulation (Fig. 2 and Additional file 8: Figure S4), although its role in regulating nutrient-induced fat accumulation is yet to be thoroughly studied in C. elegans.
Some conserved signaling pathways across metazoans, such as insulin, TOR, and serotonin pathways, have been reported to regulate body fat in C. elegans [11, 18]. As the most well-studied metabolic regulating pathway, insulin signaling connects nutrient levels to many cellular processes, such as fat metabolism, stress response, and longevity . In daf-2(e1370) mutants, the deficiency of insulin receptor-like gene increases body fat content [37, 38]. As a phosphatidylinositol kinase-related family member, TOR regulates cell growth and proliferation in response to nutrient levels . In C. elegans, rict-1 encodes a component of TOR complex, and the deficiency of rict-1(ft7) increases body fat . As a neuromodulator, serotonin controls various food-related behaviors and physiological processes in C. elegans [11, 18, 23]. In tph-1(mg280) mutants, the failure to serotonin synthesis elevates fat storage [24, 38, 40]. In contrast, exogenous serotonin addition into the NGM plates reduces TAG levels in N2 worms [19, 24, 26]. In this study, similar to those in N2 worms, knockdown of cpl-1 reduces body fat content in daf-2 and rict-1 mutants. In contrast, deficiency of tph-1 blocks the action of cpl-1 knockdown. Further, utilizing cpl-1-deficient and knockdown worms, we find that CPL-1 inhibition elevates worm serotonin signaling, suggesting serotonin signaling is required for CPL-1 inhibition-mediated fat loss.
In C. elegans, serotonin is only synthesized in a few central neurons, such as ADF and NSM neurons [41, 42]. Serotonin from ADF neurons specifically mediates body fat loss . Recently, concerning the serotonin-mediated fat loss, a central serotonin signaling, which integrates octopamine signaling with serotonin production, has been deciphered in C. elegans [24, 26]. In the central serotonin circuit, exogenous octopamine treatment and octopamine released from RIC neurons regulate serotonin synthesis in ADF neurons via octopaminergic receptor SER-6 in AWB neurons. In turn, serotonin from ADF neurons promotes the secretion of Phe-Met-Arg-Phe-NH2 (FMRFamide)-like neuroendocrine peptide FLP-7 via serotoninergic receptor MOD-1 in URX neurons. Further, FLP-7 stimulates lipolysis and fatty acid β-oxidation and hence leads to fat loss in the intestine via the NPR-22/NK2R receptor [24, 26]. The central serotonin circuit has well underlain serotonin-mediated fat loss in C. elegans, but it prompts us to test whether the peripheral signaling could be transmitted to initiate the central serotonin signaling. In this study, we demonstrate that CPL-1 inhibition in peripheral fat storage tissues is such an initiator. In the context of cpl-1 knockdown, we recapitulate the central serotonin signaling and demonstrate that peripheral CPL-1 inhibition indeed promotes central serotonin signaling to induce fat loss (Figs. 4 and 5), although a gap between peripheral CPL-1 inhibition and central serotonin synthesis is yet to be thoroughly studied in C. elegans.
In mammals, owing to the separation of the blood-brain barrier, there are two self-governed compartments of serotonin production and action. There are two rate-limiting enzymes that tryptophan hydroxylase Tph1 and Tph2 control peripheral and central serotonin synthesis, respectively . Although 95% serotonin in the body is produced from peripheral tissues, central serotonin substantially regulates mammalian energy balance . Central serotonin abatement, such as pharmacological inhibition of central serotonin synthesis in rats and induced necrosis of Pet-1+ serotonin neurons in mice, weakens thermogenesis and elevates fat accumulation in adipose tissue [46, 47]. Thus, like serotonin from ADF neurons in C. elegans, murine central serotonin also negatively regulates body fat accumulation. Supporting the notion, in this study, the lean HFD-fed Ctsl−/− mice has higher levels of brain serotonin and WAT lipolysis and fatty acid β-oxidation than the obese HFD-fed Ctsl+/+ mice (Fig. 6). In HFD-fed wild-type mice, peripheral administration of cathepsin L inhibitor CLIK195 reduces body weight gain and WAT adipogenesis and elevates brain serotonin levels and WAT lipolysis and fatty acid β-oxidation, while the intracranial injection of tryptophan hydroxylase inhibitor PCPA abrogates the effects of peripheral CLIK195 infusion (Fig. 7). Therefore, similar to that in C. elegans, a circuit, which peripheral cathepsin L inhibition promotes central serotonin synthesis, also induces fat loss in HFD-fed mice.
In conclusion, this study reveals a previously undescribed molecular mechanism that peripheral CPL-1/cathepsin L inhibition induces fat loss in C. elegans and mice through promoting central serotonin signaling (Fig. 7f). As potential therapeutic targets of obesity and associated metabolic disorders, the main modulators of the novel circuit merit combined evaluating in the future experimental animals and humans.
Worms were maintained according to the standard protocols  unless specifically noted. N2 Bristol was used as a wild-type reference strain, and the following mutant strains were obtained from Caenorhabditis Genetics Center (CGC): VC322 (+/nT1 IV;cpl-1(ok360)/nT1 V), daf-2(e1370) III, rict-1(ft7) II, tph-1(mg280) II, VP303 (rde-1(ne219) V;kzls7), NR222 (rde-1(ne219) V;kzls9), MAH23 (rrf-1(pk1417) I), WM118 (rde-1(ne300) V;nels9 X), VH624 (rhIs13 V; nre-1(hd20) lin-15B(hd126) X), RT258 (unc-119(ed3)III;pwls50), GR1366 mgls42[tph-1::GFP+pRF4(rol-6(su1006))], mod-1(ok103) V, and ser-6(tm2146)IV. The cpl-1 mutants qx304 and yq89 and CPL-1::mChOint transgenic line yqEx688 (Pcpl-1cpl-1::mChOint) were as gifts from Professor Yang, Chonglin (the Institute of Genetics and Developmental Biology, Chinese Academy of Sciences). The worms expressing lipid droplet marker DHS-3::GFP (LIU1 (ldrIs1 [dhs-3p::dhs-3::GFP + unc-76(+)])) were gifts from Professor Liu, Pingsheng (State Key Laboratory of Biomacromolecules, Institute of Biophysics, Chinese Academy of Sciences). All worms were synchronized for experiments by hypochlorite, hatched in M9 buffer (3 g KH2PO4, 6 g Na2HPO4, 5 g NaCl, 1 mL 1 M MgSO4, added H2O to 1 l) to the first larval (L1) stage, then seeded on plates.
To develop the nutrient-induced fat accumulation, we added glucose (at 1 mM and 5 mM) or palmitic acid (at 0.02 mM and 0.2 mM) into the NGM. The glucose was added into the sterile NGM after filtration sterilization. The palmitic acid was directly added into the NGM and sterilized by dry heat sterilization. For nutrient supplementation treatment, worms were added to the nutrient-supplemented plates since the L4 stage. After that, the worms were cultured for 16 h and harvested at adult.
C57BL/6 WT mice were purchased from Vital River Laboratory Animal Technology Co. Ltd. (Beijing, China). Cathepsin L-deficient Cstl−/− mice and their littermate control Cstl+/+ mice (males, C57BL/6/S129 background) were produced by Ctsl+/− heterozygous breeding pairs as the previous study . All research protocols were conducted with the approval of the Hefei University of Technology Standing Committee on Animals. Mice housed in a pathogen-free facility at 22 ± 2 °C with a 12/12-h light/dark cycle were given irradiated food and autoclaved water.
To induce obesity, male 6-week-old Cstl−/− and Cstl+/+ littermates were fed with HFD (D12451 diet containing 4.73 kcal/g and 45% kcal% fat; n = 10) and LFD (D12450B diet containing 3.85 kcal/g and 10% kcal% fat; n = 10) for 12 weeks. Their body weights were weekly measured. After 12 weeks, their energy expenditures were measured by a combined indirect calorimetry system (TSE Systems GmbH, Bad Homburg, Germany) for 2 days. Either oxygen consumption (VO2) or carbon dioxide production (VCO2) was continuously monitored during the next 24 h. At the end of the experiments, the mice were euthanized by CO2. All the tissues were rapidly isolated and weighed on ice, snap-frozen in liquid nitrogen, and stored at − 80 °C. Hematoxylin-eosin staining was performed in the epididymal fat and subcutaneous fat to show adipocyte size.
In order to examine the effects of peripheral cathepsin L inhibition on cerebral serotonin synthesis, 6-week-old male WT mice were fed with HFD for 5 weeks. At 9 weeks old, to inhibit cerebral serotonin synthesis, some of them were i.a. injected 30 mg kg−1 tryptophan hydroxylase inhibitor PCPA. To suppress the activities of peripheral cathepsin L, some of them were i.p. injected 100 mg kg−1 cathepsin L inhibitor CLIK195 at 9 and 10 weeks old. Since 9 weeks old, body weights of the mice were measured every 3 days. At 11 weeks old, the mice were euthanized by CO2, and their brains and adipose tissues were harvested.
Glucose and palmitic acid content in E. coli and C. elegans
The glucose quantification was performed as described  with minor modifications. Briefly, bacteria were seeded on control NGM plates and glucose-supplementing NGM plates for 1 day. And then, bacteria were pelleted and lysed in water by ultrasonic decomposition and centrifuged to remove the sediment. As to the analysis in C. elegans, approximately 20,000 young adult nematodes grown on control NGM plates and glucose-supplementing NGM were harvested, washed, and aliquots were removed for protein determination. The remaining nematodes were lysed in water and centrifuged to remove the sediment. The aqueous extracts of E. coli or C. elegans were measured by a blood glucose meter (Omnitest Plus, B.BRAUN, Germany). For the palmitic acid assay, bacteria seeded on control NGM plates and NGM plates with palmitic acid for 1 day were pelleted, and C15:0 standard was added to the pellet. The mixture was subjected to simultaneous extraction and transmethylation by incubating for 1 h at 70 °C in 1 mL of 2.5% H2SO4 in methanol. Then, palmitic acid was analyzed by gas chromatography as previously described . As to the analysis of palmitic acid in C. elegans, approximately 20,000 young adult nematodes grown on control NGM plates and NGM plates with 0.02 mM or 0.2 mM palmitic acid were harvested and washed. Palmitic acid in C. elegans was extracted with chloroform to methanol (1:1) and analyzed by gas chromatography as previously described . For the assay of glucose and palmitic acid, four independent growth experiments were performed, and each assay was repeated in triplicate for each growth.
Phenotypic analysis of the developmental rate
The developmental rate was observed as described  with minor modifications. Worms were grown on NGM plates with the OP50 diet at 20 °C. The eggs were collected by bleaching and washed three times with M9 buffer and allowed to hatch in M9 buffer for 18 h. After synchronization, the L1 larva was transferred to NGM plates and incubated at 20 °C. At 50 h (worms fed with OP50) or 60 h (worms fed with HT115) after synchronization, worms were washed off and mounted on agarose pads and examined on a compound microscope. The numbers of L4, adult, and gravid adult worms were visually counted based on the development of the vulva. For each condition, 3 independent experiments were performed, and at least 30 worms were scored in each test.
Oil Red O staining and quantitation
The Oil Red O staining was performed as described  with minor modifications; 0.5% Oil Red O solution was prepared in 1,2-propanediol and allowed to rest on a bench for a couple of days. On the day of staining, the Oil Red O solution was filtered through a 0.22-μm syringe filter. To conduct Oil Red O staining, day 1 adult worms were washed from NGM plates with M9 buffer for three times. After gravitational settling in tubes, 1000 μL M9 buffer and 50 μL 10% paraformaldehyde were added and immediately frozen in − 80 °C. And then, these tubes were thawed and refrozen for three cycles. At the third time, tubes were allowed to thaw on ice for 30–40 min. After complete thawing, worms were washed three times in cold M9 buffer. Then, worms were dehydrated in 1,2-propanediol for 5 min and stained with 1 mL Oil Red O solution for overnight at room temperature. After staining, worms were washed in turn with 85% 1,2-propanediol and PBS and mounted on a 2% agarose slide for imaging.
For Oil Red O quantification, the stained worms were fully suspended in 1 mL M9 buffer with 0.1% Tween-20. And then, we pipetted 10-μL suspension on slides and counted the worm numbers in the suspensions. To obtain accurate worm numbers in each 10-μL suspension, we repeated independently 3 times and took the mean to eradicate any discrepancies. According to the mean, 3000 worms were fetched in an appropriate volume. After centrifugation, stained lipid from the worms was extracted with 200 μL ethanol and quantitated at OD 510 nm.
Label-free quantitative analysis of lipid droplets using DHS-3::GFP worms
The label-free quantification of lipid droplets in DHS-3::GFP worms was performed as previously described . All worms were anesthetized in a droplet of 100 mM sodium azide and mounted on fresh 2% agarose slides before imaging. To evaluate the size of lipid droplet, images of DHS-3::GFP worms were used to measure the diameter of the lipid droplets in the posterior of the intestine with the same area by Image-Pro Plus Version 6.0 (Media Cybernetics, USA). Thirty animals were measured for each condition, and three independent experiments were repeated.
Thin-layer chromatography and measurement of triglycerides
Adult worms were washed from NGM plates with M9 buffer. After gravitational settling in tubes, the worms were incubated in 0.9% NaCl at room temperature for 20 min. And then, the worms were pelleted again by centrifugation and resuspended in 500 μL H2O. Using chloroform to acetone (1:1, v/v), lipids were extracted from 400-μL nematode suspension as described previously . The solvent was removed by pure nitrogen gas, and the lipids were dissolved in chloroform and separated on TLC plates in hexane to diethyl ether to acetic acid (80:20:1, v/v) for 40 min. TAG was visualized by dipping the plates into a developing reagent (0.63 g MnCl2, 60 mL water, 60 mL methanol, and 4 mL concentrated sulfuric acid) for 6 s. And then, the TLC plates were briefly dried and heated at 100 °C for 30 min. Utilizing the 100-μL rest nematode suspension, we assessed total protein contraction with the BCA assay kit (Sangong). TAG was quantitated by densitometric scanning at 400 nm with triolein (Sigma) as a standard and presented as TAG mass pre-microgram protein.
RNAi-mediated inactivation of cpl-1 was performed as described . The 1157-bp fragment of cpl-1(T03E6.7) was PCR amplified from worm cDNA with the forward primer (XhoΙ), 5′-GGCCTCGAGCCATTCAGCCAATACCGCAA, and the reverse primer (XbaΙ), 5′-CTCTCTAGAGCAAATAAAACTGACCCGTC. The PCR product was cloned into the T7 vector L4440 and transfected into HT115 bacteria. HT115 bacteria containing RNAi vector were cultured and added into the NGM plates. L1-stage wild-type or mutant worms were placed on the plates with RNAi bacteria and incubated at 20 °C to adulthood.
Quantitative real-time PCR
Total RNA was isolated from worms or mouse tissues using TRIzol reagent (TaKaRa). cDNA was synthesized using PrimeScript™ RT Master Mix (TaKaRa) according to the manufacturer’s protocol. The cDNA was used as the template for real-time PCR (Bio-Rad MyiQ2 Real-time PCR System) in the presence of SYBR® Premix Ex Taq™ II (TaKaRa). All primer sequences for C. elegans and mouse experiments were listed in Additional file 16: Table S5 and Additional file 17: Table S6, respectively. Data were processed using the ΔΔCT method. act-1 was used as the reference gene in C. elegans, and β-Actin was used as the reference gene in mice.
Western blot analysis
Total proteins were extracted from worms using RIPA lysis buffer with total protease inhibitor (1 μM PMSF) and phosphatase inhibitor (Sangon). The protein concentration was assessed with the BCA assay kit (Sangong). Samples in equal amounts were separated on SDS-PAGE and transferred to PVDF membranes (Millipore). Immunoblot analysis was performed with the corresponding primary and secondary antibodies. Then, reactive bands were developed using the ECL kit (Thermo) in ImageQuant LAS4000 mini (GE Healthcare) and quantified using ImageQuant TL 7.0 software (GE Healthcare). The antibodies included the following: anti-β-actin antibody; rabbit monoclonal antibody (Sigma-Aldrich, SAB5500001, 1:1000); GFP (D5.1) XP® rabbit monoclonal antibody (Cell Signaling Technology, Inc., #2956S, 1:1000); goat anti-rabbit IgG (H+L secondary antibody (Boster Biological Technology Co. Ltd., BA1054, 1:5000); and rabbit anti-CPL-1 antibody made by ourselves (1:5000). As to the production of rabbit anti-CPL-1 antibody, cpl-1 (NM_001269789) gene was amplified by PCR and constructed into the prokaryotic expression plasmid (pET28a). Then, the recombinant protein was expressed in the BL21 E. coli expression system and purified. After that, the recombinant protein was injected into the rabbit, and the serum was collected. Finally, we detected the titer at 1:256,000 using enzyme-linked immunosorbent assay and identified good specificity using western blot. The related data were listed in Additional file 18: Figure S12.
Determination of serotonin levels
The levels of serotonin in C. elegans and the whole mouse brain were measured by UPLC-MS/MS as previously described  with minor modifications. Briefly, the tissues were precisely weighed and fully lysed with a 20-fold (w/v) volume of 2% formic acid in methanol (v/v). After centrifugation at 13,000g for 25 min at 4 °C, 500 μL of supernatants was spiked with 10 μL of isoproterenol hydrochloride. And then, the mixtures were extracted using 490 μL methanol by vortexing for 30 s. After centrifugation at 13,000g for 5 min at 4 °C, the supernatants were directly detected. Their concentrations were calculated by Xcalibur software based on the standard samples.
Two-choice olfactory preference assays
The two-choice olfactory preference assays were performed as previously described  with minor modifications. Synchronized L1 worms were grown at room temperature. “Naive” animals were grown on standard NGM plates with OP50 or HT115. For training olfactory preference abilities of worms, 200 μL suspending pathogenic Pseudomonas aeruginosa PA14 was spread on one side of an NGM plate and 50 μL OP50 or HT115 suspension was used to make a small lawn on the other side of the plate. The training plates were incubated at 20 °C for 24 h before use. For the two-choice olfactory preference assays, OP50, HT115, and PA14 were resuspended at the same absorbance of 1.0 at 600 nm. And then, 25 μL PA14 were spotted onto one side of a plate and 25 μL OP50 or HT115 were spotted on the other side of the plate. The assay plates were air-dried for 5 h at room temperature. To perform two-choice olfactory preference assays, adult worms were firstly transformed to the training plates for 1 h or 2 h. And then, “trained” or “naive” worms were washed twice in S-basal buffer. One hundred to 300 worms were placed at the center of the assay plates and allow them to be equidistant from the 2 bacteria spots. After moving freely for 1 h, the numbers of worms on bacteria spots were counted.
Data were presented as mean ± SEM. As indicated in the figure legends, Student’s t test, one-way ANOVA, or two-way ANOVA was used to evaluate the difference among various C. elegans assay groups. As to the mouse experiments, the comparisons between the two groups were assessed with the nonparametric Mann-Whitney test due to our small sample sizes and data abnormal distribution. GraphPad Prism 7.01 (GraphPad Software, Inc.) was used as the statistical software, and p < 0.05 was considered as statistically significant.
We thank Professor Chonglin Yang (the Institute of Genetics and Developmental Biology, Chinese Academy of Sciences) for providing the cpl-1-mutant C. elegans, Professor Pingsheng Liu (State Key Laboratory of Biomacromolecules, Institute of Biophysics, Chinese Academy of Sciences) for providing the LIU1 (ldrIs1 [dhs-3p::dhs-3::GFP+unc-76(+)]) strain, and Professor Cheng-Gang Zou (State Key Laboratory for Conservation and Utilization of Bio-Resources in Yunnan, Yunnan University) for providing the PA14 bacteria. We also thank Professor Shouhong Guang (University of Science and Technology of China) for his advice regarding the experiments. Some C. elegans strains were provided by the CGC, which is funded by the NIH Office of Research Infrastructure Programs (P40 OD010440).
YL, BB, G-PS, and JL conceived and designed the experiments. YL, BB, HY, XW, AF, LZ, XN, and NY performed the experiments. YL and BB analyzed the data. YL, BB, and JL wrote the manuscript. All authors read and approved the final manuscript.
Our study was supported by the National Natural Science Foundation of China (31401204, U1532269 to BB; 31471320, 31671485 to LJ) and the Natural Science Foundation of Anhui Province (1408085QC48 to BB).
Ethics approval and consent to participate
The research of mouse experiments in the manuscript has been conducted under the guidance of international ethical standards. All research protocols were conducted with the approval of the Hefei University of Technology Standing Committee on Animals.
All authors declare that they have no competing interests.
- 5.Masson O, Prebois C, Derocq D, Meulle A, Dray C, Daviaud D, Quilliot D, Valet P, Muller C, Liaudet-Coopman E. Cathepsin-D, a key protease in breast cancer, is up-regulated in obese mouse and human adipose tissue, and controls adipogenesis. PLoS One. 2011;6(2):e16452.PubMedPubMedCentralCrossRefGoogle Scholar
- 8.Naour N, Rouault C, Fellahi S, Lavoie M-E, Poitou C, Keophiphath M, Eberlé D, Shoelson S, Rizkalla S, Bastard J-P, et al. Cathepsins in human obesity: changes in energy balance predominantly affect cathepsin S in adipose tissue and in circulation. J Clin Endocrinol Metab. 2010;95(4):1861–8.PubMedCrossRefGoogle Scholar
- 9.Yang M, Sun J, Zhang T, Liu J, Zhang J, Shi MA, Darakhshan F, Guerre-Millo M, Clement K, Gelb BD, et al. Deficiency and inhibition of cathepsin K reduce body weight gain and increase glucose metabolism in mice. Arterioscler Thromb Vasc Biol. 2008;28(12):2202–8.PubMedPubMedCentralCrossRefGoogle Scholar
- 11.Ashrafi K. Obesity and the regulation of fat metabolism. 2007. In: WormBook: The Online Review of C. elegans Biology [Internet]. Pasadena: WormBook; 2005-2018. Available from: https://www.ncbi.nlm.nih.gov/books/NBK19757/.
- 31.Mejia-Martinez F, Franco-Juarez B, Moreno-Arriola E, Hernandez-Vazquez A, Martinez-Avila M, Gomez-Manzo S, Marcial-Quino J, Carvajal K, Velazquez-Arellano A, Ortega-Cuellar D. The MXL-3/SBP-1 axis is responsible for glucose-dependent fat accumulation in C. elegans. Genes (Basel). 2017;8(11):307–21.PubMedCentralCrossRefPubMedGoogle Scholar
- 47.McGlashon Jacob M, Gorecki Michelle C, Kozlowski Amanda E, Thirnbeck Caitlin K, Markan Kathleen R, Leslie Kirstie L, Kotas Maya E, Potthoff Matthew J, Richerson George B, Gillum MP. Central serotonergic neurons activate and recruit thermogenic brown and beige fat and regulate glucose and lipid homeostasis. Cell Metab. 2015;21(5):692–705.PubMedPubMedCentralCrossRefGoogle Scholar
- 54.Suominen T, Uutela P, Ketola RA, Bergquist J, Hillered L, Finel M, Zhang H, Laakso A, Kostiainen R. Determination of serotonin and dopamine metabolites in human brain microdialysis and cerebrospinal fluid samples by UPLC-MS/MS: discovery of intact glucuronide and sulfate conjugates. PLoS One. 2013;8(6):e68007.PubMedPubMedCentralCrossRefGoogle Scholar
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.