The MXL-3/SBP-1 Axis Is Responsible for Glucose-Dependent Fat Accumulation in C. elegans

Chronic exposure to elevated glucose levels leads to fatty acid accumulation, which promotes the development of metabolic diseases such as obesity and type 2 diabetes. MXL-3 is a conserved transcriptional factor that modulates the inhibition of lipolysis in Caenorhabditis elegans. However, the role of MXL-3 in lipid metabolism during nutrient excess remains unknown. We hypothesized that inhibition of MXL-3 prevents glucose-dependent fat accumulation. Nematodes from wild-type N2, MXL-3::GFP and sbp-1 or mxl-3 null strains were grown on standard, high glucose or high glucose plus metformin plates for 24 h. Using laser-scanning confocal microscopy, we monitored the glucose-induced activation of MXL-3 labeled with GFP (MXL-3::GFP). Lipid levels were determined by Oil Red O (ORO) staining and gas chromatography/mass spectrometry, and gene expression was assessed by qRT-PCR. We found that high glucose activated MXL-3 by increasing its rate of nuclear entry, which in turn increased lipid levels via sterol regulatory element-binding protein (SBP-1). This activated critical genes that synthesize long chain unsaturated fatty acids (MUFAs and PUFAs) and repress lipolytic genes. Interestingly, the anti-diabetic drug metformin inhibited MXL-3 activation and subsequently prevented glucose-dependent fat accumulation. These findings highlight the importance of the MXL-3/SBP-1 axis in the regulation of lipid metabolism during nutritional excess and provide new insight into the mechanism by which metformin prevents lipid accumulation. This study also suggests that inhibition of MXL-3 may serve as a potential target for the treatment of chronic metabolic diseases, including obesity, type 2 diabetes, and cardiovascular disease.


Introduction
Hyperglycemia is a key risk factor for the development of diabetes mellitus and metabolic syndrome. Nowadays, both diseases have a substantial impact on public health and safety because and strongly suggest that MXL-3/SBP-1 may be a novel target for the development of therapeutic procedures that ameliorate or even prevent metabolic diseases.

Strains
All strains were obtained from the Caenorhabditis Genetics Center (CGC, University of Minnesota, Minneapolis, MN, USA) strain bank. The wild-type strain was C. elegans variant Bristol, N2 strain. The following mutants were used: RB1588 mxl-3 (ok1947) X, VC1886 sbp-1 (ok2363) III and UL1439 MXL-3::GFP unc-119 (ed3) III. Worms were cultured and handled according to standard laboratory protocols, as described previously [22]. Briefly, worms were grown on NGM-lite plates and maintained at 20 • C with Escherichia coli OP50-1 as the food source. When required, synchronized nematodes were maintained until the L4 larval state and transferred to new experimental NGM plates without glucose (control), with 100 mM glucose (glucose) or with 100 mM glucose plus 50 mM metformin (glucose/metformin). Glucose plates were prepared at a final concentration of 100 mM by adding glucose to the solid media and then allowing to spread for at least one day; metformin (Sigma-Aldrich, St. Louis, MO, USA) was added to melted media [23]. All experimental media included 49 µM 5-fluorodeoxyuridine (FUDR, Sigma-Aldrich). After 24 h, nematodes were collected and excess bacteria were quickly removed with M9 buffer. Unless otherwise mentioned, samples were frozen in liquid nitrogen and stored at −70 • C until analysis.

Fat Storage Staining
Fat storage was assayed as described previously using the Oil red O (Sigma-Aldrich) staining method with some modifications [24,25]. Worms were collected with M9 buffer, washed twice with 1X PBS buffer, and suspended in 100 µL of 1X PBS, to which an equal volume of 2X buffer (160 mM KCl, 40 mM NaCl, 14 mM Na 2 EGTA at pH 7.4) containing 2% paraformaldehyde was added. Worms were then washed three times with 1X PBS. After allowing the worms to settle, the buffer was aspirated, and 500 µL of freshly prepared ORO solution was added (0.5 g of ORO powder in 100 mL of 60% isopropanol; filter: 0.2 µm [26]). Animals were incubated overnight with rocking. The dye was removed, and the samples were washed with 200 µL of 1X PBS/0.01% Triton X-100. Worms were mounted on prepared agar slides. Images were acquired under differential interference contrast (DIC) and a rhodamine filter using a confocal FluoView FV1000 microscope (Olympus, Tokyo, Japan).

RNA Extraction and Quantitative RT-PCR
Prior to total mRNA quantification, worms were lysed by adding 100 µL of lysis buffer (5 mM Tris pH 8.0, 0.5% Triton X-100, 0.5% Tween 20, 0.25 mM EDTA and 2 mg/mL proteinase K) and incubating at 65 • C for 10 min. To inactivate the proteinase K, samples were heated to 85 • C for 1 min and immediately cooled on ice [28]. Total RNA was extracted by adding 300 µL of Trizol to 100 µL of worm lysate following the manufacturer's instructions. cDNA was obtained using random hexamer oligonucleotides and M-MLV reverse transcriptase (Invitrogen, Carlsbad, CA, USA). Expression analyses were performed using Taqman (Applied Biosystems, Foster City, CA, USA) probes, which were detected using the StepOne (Thermo Fisher Scientific, Inc., Waltham, MA, USA) System and StepOne Software v2.2 (ABI instruments, Carlsbad, CA, USA). mRNA levels were normalized to eukaryotic 18S rRNA gene expression as an endogenous control.

Lipid Extraction and Gas Chromatography/Mass Spectrometry Assays
Fatty acid composition was determined by producing fatty-acid methyl esters (FAMEs). Total fatty acids were extracted via the chloroform-methanol (2:1, v/v) method and incubated at 95 • C for 1 h in 0.2 mL of toluene, 1.76 mL of methanol and 0.04 mL of H 2 SO 4 [29,30]. The concentration and composition of FAMEs were evaluated, as described previously [31]. Briefly, Gas Chromatography/Mass Spectrometry (GC/MS) was performed using an Agilent 6890N gas chromatograph (Agilent Technologies, Santa Clara, CA, USA) equipped with an HP-5MS column (30 m × 0.25 mm I.D. × 0.25 µm, Agilent) containing ultrapure helium as a carrier gas, and 1 µL of each sample was injected at a 10:1 split ratio. The initial oven temperature was set to 150 • C, increased to 310 • C at 8 • C per min, and held for 15 min at this temperature. Mass detection was performed as previously described for a JMS-GC System [31]. Fatty acid abundance was expressed as the percent of total fatty acids in the N2 control strain for each sample.

MXL-3 Is Involved in Lipid Accumulation Due to High Glucose Exposure in C. elegans
To investigate whether the MXL-3 transcription factor facilitates the excess lipid accumulation induced by high glucose exposure in C. elegans, semi-quantitative analysis of neutral lipids stained with ORO was performed in fixed wild-type (N2) and mxl-3 null mutant (mxl-3/ok1947) animals. Consistent with previous reports [7], we found that high glucose exposure for 24 h significantly induced lipid accumulation in N2 animals, as demonstrated by an increase in the intensity of ORO staining ( Figure 1A,B). Next, we determined whether lipid accumulation due to a high-glucose diet is influenced by MXL-3. Ablation of mxl-3 substantially reduced ORO staining compared with that observed for the N2 control strain ( Figure 1A,B), whereas the augmented lipid storage observed in N2 under high glucose was blunted in MXL-3-null animals, even when the mutant was grown under the same conditions, suggesting that fat accumulation is mediated by the MXL-3 transcription factor. StepOne Software v2.2 (ABI instruments, Carlsbad, CA, USA). mRNA levels were normalized to eukaryotic 18S rRNA gene expression as an endogenous control.

Lipid Extraction and Gas Chromatography/Mass Spectrometry Assays
Fatty acid composition was determined by producing fatty-acid methyl esters (FAMEs). Total fatty acids were extracted via the chloroform-methanol (2:1, v/v) method and incubated at 95 °C for 1 h in 0.2 mL of toluene, 1.76 mL of methanol and 0.04 mL of H2SO4 [29,30]. The concentration and composition of FAMEs were evaluated, as described previously [31]. Briefly, Gas Chromatography/Mass Spectrometry (GC/MS) was performed using an Agilent 6890N gas chromatograph (Agilent Technologies, Santa Clara, CA, USA) equipped with an HP-5MS column (30 m × 0.25 mm I.D. × 0.25 μm, Agilent) containing ultrapure helium as a carrier gas, and 1 μL of each sample was injected at a 10:1 split ratio. The initial oven temperature was set to 150 °C, increased to 310 °C at 8 °C per min, and held for 15 min at this temperature. Mass detection was performed as previously described for a JMS-GC System [31]. Fatty acid abundance was expressed as the percent of total fatty acids in the N2 control strain for each sample.

MXL-3 Is Involved in Lipid Accumulation Due to High Glucose Exposure in C. elegans
To investigate whether the MXL-3 transcription factor facilitates the excess lipid accumulation induced by high glucose exposure in C. elegans, semi-quantitative analysis of neutral lipids stained with ORO was performed in fixed wild-type (N2) and mxl-3 null mutant (mxl-3/ok1947) animals. Consistent with previous reports [7], we found that high glucose exposure for 24 h significantly induced lipid accumulation in N2 animals, as demonstrated by an increase in the intensity of ORO staining ( Figure 1A,B). Next, we determined whether lipid accumulation due to a high-glucose diet is influenced by MXL-3. Ablation of mxl-3 substantially reduced ORO staining compared with that observed for the N2 control strain ( Figure 1A,B), whereas the augmented lipid storage observed in N2 under high glucose was blunted in MXL-3-null animals, even when the mutant was grown under the same conditions, suggesting that fat accumulation is mediated by the MXL-3 transcription factor.

MXL-3 Actively Shuttles between the Cytoplasm and Nucleus in Response to High Glucose
MXL-3 actively shuttles between the cytoplasm and nucleus in response to nutrient availability [19]. Therefore, it seems plausible that glucose might affect the cellular localization of MXL-3. To determine whether high glucose exposure modifies the nucleocytoplasmic shuttling of MXL-3::GFP, transgenic worms were grown on media containing high glucose (100 mM) or no glucose (control) for 24 h. Glucose exposure resulted in the major accumulation of MXL-3::GFP in the nucleus compared to that in the controls, suggesting that glucose augments its regulatory activity (Figure 2A,B). To verify if nuclear localization correlates with transcriptional activity, we measured several known target genes of MXL-3 by performing quantitative RT-PCR (qRT-PCR). The expression levels of the specific lysosomal lipases lipl-1 and lipl-3 were markedly downregulated after N2 nematodes were fed high glucose for 24 h compared to the levels in worms grown under control conditions; this effect was abolished in mxl-3 null mutants ( Figure 2C). Thus, MXL-3 nuclear localization induced by glucose enhances the transcriptional repression of these genes.

MXL-3 Actively Shuttles between the Cytoplasm and Nucleus in Response to High Glucose
MXL-3 actively shuttles between the cytoplasm and nucleus in response to nutrient availability [19]. Therefore, it seems plausible that glucose might affect the cellular localization of MXL-3. To determine whether high glucose exposure modifies the nucleocytoplasmic shuttling of MXL-3::GFP, transgenic worms were grown on media containing high glucose (100 mM) or no glucose (control) for 24 h. Glucose exposure resulted in the major accumulation of MXL-3::GFP in the nucleus compared to that in the controls, suggesting that glucose augments its regulatory activity ( Figure  2A,B). To verify if nuclear localization correlates with transcriptional activity, we measured several known target genes of MXL-3 by performing quantitative RT-PCR (qRT-PCR). The expression levels of the specific lysosomal lipases lipl-1 and lipl-3 were markedly downregulated after N2 nematodes were fed high glucose for 24 h compared to the levels in worms grown under control conditions; this effect was abolished in mxl-3 null mutants ( Figure 2C). Thus, MXL-3 nuclear localization induced by glucose enhances the transcriptional repression of these genes. (B) Nuclear MXL-3::GFP intensity by glucose relative to nematodes fed with control diet. (C) mRNA quantification of lipl-1 and lipl-3 was performed using qRT-PCR in the N2 and mxl-3 mutant (ok1947 null) nematode strains. The relative expression of each gene was normalized to endogenous 18S rRNA gene expression. Data represent the mean ± SEM of three independent experiments; * p < 0.05, ** p < 0.01, *** p < 0.001, one-way ANOVA with Bonferroni's post hoc test using GraphPad Prism.

MXL-3 Regulates Lipid Synthesis Through SBP-1
As observed in mammals, the transcription factor SREBP ortholog in C. elegans, SBP-1, induces lipid synthesis by activating lipogenic genes such as acetyl-CoA carboxylase (pod-2), fatty acid synthase (fasn-1), elongase (elo-2) and ∆12 desaturase (fat-2) [32]. A previous study employing an assay of transcriptional activity proposed that MXL-3 positively regulates SBP-1 expression [21], suggesting a new physiological role for MXL-3 in the lipid biosynthesis signal transduction pathway. To determine if SBP-1 induces lipid accumulation in an MXL-3-dependent manner under high glucose conditions, we first measured the mRNA expression levels of sbp-1 during glucose excess. As expected, the expression of sbp-1 mRNA was significantly higher in nematodes treated with high glucose ( Figure 3A), suggesting augmented activity of this transcription factor. To establish MXL-3-and SBP-1-dependent gene regulation, we grew mxl-3 null mutants (ok1947 strain) on high glucose and measured the mRNA levels of both sbp-1 and some of their known target genes. MXL-3 was necessary for full sbp-1 expression, as the mxl-3 (ok1947) null mutant failed to properly activate sbp-1 gene expression ( Figure 3A). Similarly, genes that were highly expressed under glucose, specifically pod-2, fasn-1, and fat-2, were blunted by the absence of MXL-3 ( Figure 3B-D), indicating that this factor is an additional element required for the appropriate expression of SBP-1-dependent genes. elo-2 expression was decreased under high glucose, even in the N2 and mxl-3 null mutant strains. To determine to what extent MXL-3 contributes to glucose-regulated gene expression through SBP-1, similar studies were performed in sbp-1 mutants. As expected, the expression levels of pod-2, fasn-1, elo-2 and fat-2 were not induced in response to a high-glucose diet ( Figure 3). Taken together, our data suggest that MXL-3 might be important for promoting a selective lipogenic program that is dependent on a high-glucose diet through SBP-1. assay of transcriptional activity proposed that MXL-3 positively regulates SBP-1 expression [21], suggesting a new physiological role for MXL-3 in the lipid biosynthesis signal transduction pathway. To determine if SBP-1 induces lipid accumulation in an MXL-3-dependent manner under high glucose conditions, we first measured the mRNA expression levels of sbp-1 during glucose excess. As expected, the expression of sbp-1 mRNA was significantly higher in nematodes treated with high glucose ( Figure 3A), suggesting augmented activity of this transcription factor. To establish MXL-3-and SBP-1-dependent gene regulation, we grew mxl-3 null mutants (ok1947 strain) on high glucose and measured the mRNA levels of both sbp-1 and some of their known target genes. MXL-3 was necessary for full sbp-1 expression, as the mxl-3 (ok1947) null mutant failed to properly activate sbp-1 gene expression ( Figure 3A). Similarly, genes that were highly expressed under glucose, specifically pod-2, fasn-1, and fat-2, were blunted by the absence of MXL-3 ( Figure 3B-D), indicating that this factor is an additional element required for the appropriate expression of SBP-1-dependent genes. elo-2 expression was decreased under high glucose, even in the N2 and mxl-3 null mutant strains.
To determine to what extent MXL-3 contributes to glucose-regulated gene expression through SBP-1, similar studies were performed in sbp-1 mutants. As expected, the expression levels of pod-2, fasn-1, elo-2 and fat-2 were not induced in response to a high-glucose diet ( Figure 3). Taken together, our data suggest that MXL-3 might be important for promoting a selective lipogenic program that is dependent on a high-glucose diet through SBP-1.

MXL-3 Is Required to Maintain Fat Stores and De Novo Fat Synthesis
Since MXL-3 promotes transcriptional changes under a high-glucose diet through SBP-1, we determined whether the observed changes were also reflected in fat and lipid metabolism. Using GC/MS, we measured the composition of free fatty acids (FFA) in the mxl-3 mutant strain. Consistent and (E) fat-2 in N2 wild-type or mxl-3 and sbp-1 mutant strains. Mutation of mxl-3 significantly decreased sbp-1 mRNA levels under both control and high-glucose conditions. Similar results were observed for elo-2 and fasn-1, whereas no effects on pod-2 and fat-2 were observed. The relative expression of each gene was normalized to endogenous 18S rRNA gene expression. Data represent the mean ± SEM of three independent experiments; * p < 0.05, ** p < 0.01, *** p < 0.001, one-way ANOVA with Bonferroni's post hoc test using GraphPad Prism.

MXL-3 Is Required to Maintain Fat Stores and De Novo Fat Synthesis
Since MXL-3 promotes transcriptional changes under a high-glucose diet through SBP-1, we determined whether the observed changes were also reflected in fat and lipid metabolism. Using GC/MS, we measured the composition of free fatty acids (FFA) in the mxl-3 mutant strain. Consistent with ORO staining (Figure 1A), worms lacking MXL-3 showed decreased FFA content (Figures 4A and A1). We also analyzed the three FFA lipid classes, saturated fatty acids (SFAs), mono-unsaturated fatty acids (MUFAs), and poly-unsaturated fatty acids (PUFAs), and found that MXL-3 triggers a metabolic shift that decreases SFAs and PUFAs ( Figure 4B,D); MUFAs were not modified by the absence of MXL-3 ( Figure 4C). In mxl-3 mutants, we observed a decrease in stearic acid (18:0), a fatty acid obtained by elongation that is dependent on ELO-2, an SBP-1-dependent elongase ( Figure 4F). Additionally, in the absence of MXL-3, the 18:1n9/18:2n6 (oleic acid/linoleic acid) ratio was decreased ( Figures 4G and A2), while oleic acid was increased ( Figure 4H), suggesting FAT-2 (∆12) desaturase blockage controls the conversion of fatty acids to PUFAs ( Figure 4E). When the mxl-3 mutant was subjected to a high-glucose diet and its lipid profile was analyzed by GC/MS, we did not observe FFA accumulation or SFA/PUFA changes, which were normally observed in the wild-type strain. MUFA content was slightly increased by glucose ( Figure 4A-D and Figure A1). Together, these data suggest that MXL-3 is as necessary as SBP-1 for the adequate synthesis and elongation of fatty acids, and MXL-3 is involved in the same steps mediated by SBP-1. These findings reinforce the hypothesis that MXL-3 acts through SBP-1 to regulate lipid metabolism. with ORO staining (Figure 1A), worms lacking MXL-3 showed decreased FFA content ( Figure 4A and Figure A1). We also analyzed the three FFA lipid classes, saturated fatty acids (SFAs), mono-unsaturated fatty acids (MUFAs), and poly-unsaturated fatty acids (PUFAs), and found that MXL-3 triggers a metabolic shift that decreases SFAs and PUFAs ( Figure 4B,D); MUFAs were not modified by the absence of MXL-3 ( Figure 4C). In mxl-3 mutants, we observed a decrease in stearic acid (18:0), a fatty acid obtained by elongation that is dependent on ELO-2, an SBP-1-dependent elongase ( Figure 4F). Additionally, in the absence of MXL-3, the 18:1n9/18:2n6 (oleic acid/linoleic acid) ratio was decreased ( Figure 4G and Figure A2), while oleic acid was increased ( Figure 4H), suggesting FAT-2 (Δ12) desaturase blockage controls the conversion of fatty acids to PUFAs ( Figure  4E). When the mxl-3 mutant was subjected to a high-glucose diet and its lipid profile was analyzed by GC/MS, we did not observe FFA accumulation or SFA/PUFA changes, which were normally observed in the wild-type strain. MUFA content was slightly increased by glucose ( Figure 4A-D and Figure A1). Together, these data suggest that MXL-3 is as necessary as SBP-1 for the adequate synthesis and elongation of fatty acids, and MXL-3 is involved in the same steps mediated by SBP-1. These findings reinforce the hypothesis that MXL-3 acts through SBP-1 to regulate lipid metabolism.

Metformin Decreases Lipid Overload through MXL-3
Metformin, a drug that is broadly prescribed to treat type-2 diabetes and obesity, is beneficial because it modulates several transcriptional elements that affect lipid storage [33]. To evaluate the effects of metformin on the glucose-dependent accumulation of lipid droplets, the wild-type N2 strain was grown with high glucose and simultaneously treated with metformin (glucose-metformin). Glucose-dependent lipid droplet accumulation was diminished in response to metformin, as evidenced by the reduced fluorescence of ORO staining close to control levels ( Figure 5A). Since we showed that glucose availability modulated the cellular location of MXL-3, we next evaluated the subcellular distribution of MXL-3 upon treatment with metformin (24 h). Interestingly, metformin-treated worms grown in glucose-containing media showed a diminishment of nuclear content of MXL-3::GFP from their nuclei, resulting in cytoplasmic accumulation similar to untreated worms ( Figure 5B). Thus, the localization of MXL-3 in the cytoplasm upon metformin treatment in the presence of glucose may be one mechanism underlying reduced lipid storage. Consistent with the exit of MXL-3 from the nucleus, expression of the MXL-3 target gene lipl-1 increased to the same extent, while lipl-3 expression did not recover ( Figure 5C,D). Similarly, when MXL-3 was inhibited by metformin, the MXL-3-dependent expression of SBP-1 target genes, including fasn-1, pod-2, and fat-2, returned to control levels in N2 nematodes ( Figure 5E-G). Furthermore, based on MXL-3 cytoplasmic localization, when wild-type worms were grown in the presence of both glucose and metformin and then lipid-related gene expression and lipid stores were assessed, FFAs were decreased by metformin versus the glucose group and mimicked the lipid levels observed under control conditions, even in the presence of high glucose (Figures 5H and A1). This effect was more evident for MUFAs and PUFAs, which returned to control levels ( Figure 5I-K). Specifically, the fatty acid moieties that had a clear tendency to decrease to control levels under glucose-metformin conditions were the myristic, stearic, oleic, linoleic and eicosapentaenoic acids (14:0, 18:0, 18:1n9, 18:2n6, and 20:5n3, respectively, Figure 5L). This lipid profile was surprising because it was similar to the mxl-3 null mutant profile ( Figure A1). Based on these results, metformin decreased lipid storage and modified the FFA profile via an MXL-3-and SBP-1-dependent mechanism.

Discussion
In this study, we determined that MXL-3 is activated by a high-glucose diet and, together with SBP-1, promotes fat accumulation. This effect was mediated by the increased expression of genes that mediate fatty acid synthesis and desaturation. Furthermore, lipid accumulation under glucose-rich conditions was also dependent on MXL-3-dependent lipolysis. Moreover, the anti-diabetic drug metformin prevented MXL-3 activation, suggesting a new key role for MXL-3 in the mechanism underlying metformin action on lipid accumulation. Thus, MXL-3 is a transcription factor that acts at the nexus of metabolic pathways to stimulate fat accumulation.

MXL-3 Senses Nutrient Availability and Regulates the Obese Phenotype during High Glucose Exposure
The key finding of our study was the identification of MXL-3 as a primary activator of an obese phenotype that has previously been shown by others to be derived from lipid accumulation via high glucose ingestion [6][7][8][9]. Myc-associated factor X contains a basic helix-loop-helix motif (b-HLH) and a leucine zipper domain and functions at the core of a transcription factor network that controls diverse members of the Myc family. Indeed, the heterodimer MAX-Myc binds to E-box sequences

Discussion
In this study, we determined that MXL-3 is activated by a high-glucose diet and, together with SBP-1, promotes fat accumulation. This effect was mediated by the increased expression of genes that mediate fatty acid synthesis and desaturation. Furthermore, lipid accumulation under glucose-rich conditions was also dependent on MXL-3-dependent lipolysis. Moreover, the anti-diabetic drug metformin prevented MXL-3 activation, suggesting a new key role for MXL-3 in the mechanism underlying metformin action on lipid accumulation. Thus, MXL-3 is a transcription factor that acts at the nexus of metabolic pathways to stimulate fat accumulation.

MXL-3 Senses Nutrient Availability and Regulates the Obese Phenotype during High Glucose Exposure
The key finding of our study was the identification of MXL-3 as a primary activator of an obese phenotype that has previously been shown by others to be derived from lipid accumulation via high glucose ingestion [6][7][8][9]. Myc-associated factor X contains a basic helix-loop-helix motif (b-HLH) and a leucine zipper domain and functions at the core of a transcription factor network that controls diverse members of the Myc family. Indeed, the heterodimer MAX-Myc binds to E-box sequences (CACGTG) to activate or repress the transcription of several genes, including those for glycolysis [10,11]. Importantly, in several processes such as the cell cycle and apoptosis, there is a strict requirement for MAX to ensure adequate Myc function. However, in Drosophila spp., a Myc mutant lacking a Max-interaction domain retained partial activity [34]. C. elegans contains two MAX orthologs, MXL-1 and MXL-3, the latter of which regulates lipolysis in response to nutrient levels [19]. However, the role of MXL-3 in the development of metabolic diseases, such as obesity dependent on a high-glucose diet, remains poorly studied. Therefore, the importance of MXL-3 activity in the control of lipid accumulation under high glucose exposure was evaluated in this study.
According to O'Rourke and Ruvkun, MXL-3 moves to the cytoplasm during nutrient deprivation and de-represses lysosomal lipolysis (lipophagy). This is a critical mechanism by which most cells catabolize lipids as an energy source [19]. Interestingly, we observed that MXL-3::GFP shuttled from the cytoplasm to the nucleus under high-glucose conditions, suggesting MXL-3 is a sensor of intracellular glucose metabolism; indeed, during nutrient deprivation, MXL-3 was expelled from the nucleus, whereas under high-glucose conditions, MXL-3 occupies the promoters of its targets to modulate gene expression. Consistent with the glucose-dependent occupancy of MXL-3 in the nuclei, the expression levels of lipl-1 and lipl-3 were diminished by glucose excess, resulting in repression of the lipophagic process. These findings suggest that MXL-3 activation is responsible for lipid accumulation.

MXL-3 Converges in the Complex Lipid Biosynthesis Signaling Pathway
Energy homeostasis is maintained by an extremely complex network of signaling pathways that operate to mediate different aspects of the fatty acid balance to maintain energetic resources. SBP-1, known as SREBP1c in mammals, acts as a critical lipogenic transcriptional factor that promotes lipid synthesis by inducing genes such as pod-2, fasn-1 and (elo-2 [32]. However, less is known about its upstream regulation. Recently, employing a transcriptional activity assay, MacNeil et al. proposed that MXL-3 positively regulates sbp-1 expression [21]. Since MXL-3 increased lipid storage and given the key role of SBP-1 in lipid metabolism, we tested the possible relationship between both transcription factors. Interestingly, in animals that lacked mxl-3, we showed diminished sbp-1 gene expression. The effects of glucose observed in wild-type strains and the near doubling of sbp-1 expression in the N2 strain with glucose were avoided in mxl-3 mutants. Similarly, we observed a dependence on MXL-3 for the adequate expression of SBP-1-dependent lipogenic genes, including pod-2, fasn-1 and fat-2, when glucose was administered. Based on these findings, transcriptional lipogenic processes are not only regulated by SBP-1 but also by MXL-3. Thus, we propose a new critical physiological role for MXL-3 in the complex lipid biosynthesis signal transduction pathway. Fat regulation in living organisms, including C. elegans, involves a complex network of genes and signaling pathways. Here, we showed that MXL-3 is in part responsible for the observed transcriptional changes that result in lipid accumulation during a high-glucose diet. Our analysis of fatty acid composition using GC/MS showed that the absence of mxl-3 decreases total FFAs in a similar fashion as previously reported in an sbp-1 RNAi strain [9]. When we analyzed FFAs grouped in families according to their saturation levels, we showed that the absence of the MXL-3 transcription factor decreased SFAs and PUFAs but did not change MUFAs. These results might be explained by blockage of the FAT-2 (∆12) desaturase, an enzyme that catalyzes a critical SBP-1-dependent step, because there was a decrease in calculated activity leading to a decrease in the conversion of fatty acids to PUFAs. This led to decreased de novo synthesis from acetyl-coA because the genes that carry out this process were also reduced.
For de novo synthesis, two multifunctional enzymes, acetyl-CoA carboxylase (POD-2) and fatty acid synthase (FASN-1), both of which are SBP-1 targets, are necessary to synthesize palmitic acid (16:0) [32,35]. Palmitic acid is integrated into TAGs or is modified by fatty acid elongases and desaturases (elo and fat genes, respectively) to form a variety of long-chain UFAs [35]. In mxl-3 mutants, we observed a decrease in stearic acid (18:0), a fatty acid obtained by elongation dependent on ELO-2, a palmitic acid elongase that is itself dependent on SBP-1 [32]. In addition, the ratio of 18:1n9/18:2n6 (oleic acid/linoleic acid) was decreased while oleic acid increased, confirming the decreased synthesis of PUFAs in mxl-3 mutants; similar levels of PUFAs were observed in an sbp-1 RNAi strain [9]. These changes in fatty acid composition may affect energy storage functions as well as membrane properties associated with signal transduction mechanisms.
Thus, the key finding of our study is the identification of MXL-3, which acts in concert with SBP-1, as an essential component for fatty acid metabolism. Interestingly, the overlapping functions of these two transcription factors might be explained by the fact that both are helix-loop-helix proteins with canonical E-box binding activities [36], thus suggesting a possible interplay between these transcription factors to regulate lipid metabolism.

MXL-3 Is a New Player That Mediates the Protective Effect of Metformin against Lipid Accumulation
Metformin is one of the most commonly prescribed drugs for type II diabetes and obesity [33]. Previously, metformin was found to decrease lipid concentrations in C. elegans [23]. Therefore, we speculated that metformin might be responsible for regulating the biological function of MXL-3. We demonstrated that metformin enhance cytoplasmic localization of MXL-3, even in glucose-treated worms. These results are consistent with the fact that metformin triggers caloric restriction-like effects in C. elegans [37], promoting changes in MXL-3 cellular localization that have been observed during dietary restriction [19]. Similar effects on nucleocytoplasmic shuttling have also been described with other transcription factors, such as FOXO1 (Forkhead box protein O1) and CRTC2 (CREB Regulated Transcription Coactivator 2) [38][39][40]. Thus, our studies strongly suggest that regulation of MXL-3 pathway crosstalk with SBP-1 by metformin plays an important role in ameliorating lipid accumulation, a mechanism that was unknown until this work.
MXL-3 might modulate the cellular localization of DAF-16/FOX0, a pro-longevity and lipid-mobilizing transcription factor of the FOXO-family that is cytoplasmic under physiological conditions, because loss of MXL-3 function induces its nuclear localization, suggesting a relationship between both transcription factors [41]. Interestingly, a recent study suggested a positive feedback loop involving DAF-16 and AMPK, the latter of which is an evolutionarily conserved sensor of cellular energy status under low energy levels or metformin [42][43][44]. Consistent with this, we speculate that metformin, via AMPK induction, may inhibit MXL-3 to activate DAF-16 and trigger the activation of genes required for β-oxidation and desaturation, such as nhr-49, to diminish systemic lipid accumulation. These possibilities warrant further exploration.

Conclusions
In summary, our data showed that the increased expression of the transcription factor MXL-3 is associated with an increased activity of SBP-1, suggesting that this couple mediates the upstream regulation of lipid metabolism in nematodes exposed to a high-glucose diet. These observations imply that increased MXL-3 activity in the organism is an important factor in the development of pathogenesis of lipid accumulation. Therefore, we reasonably infer that MXL-3 antagonist (such as metformin) may serve a suitable mechanism for the pharmacology. Figure A1. The striking lipid profile shift after glucose and metformin exposure is controlled by MXL-3. A subset of 17 fatty acids is shown in the N2 and mxl-3 mutant strains treated with high glucose and high glucose plus metformin. (A) FFAs that increased due to glucose exposure showed a clear tendency to return to the levels of the control group in the presence of metformin. (B) FFAs are unable to change in the mxl-3 mutant due to glucose or metformin. Graphs depict the mean ± SEM of three independent experiments; * p < 0.05, *** p < 0.001 vs. fatty acids in the control group, one-way ANOVA with Bonferroni's post hoc test using GraphPad Prism. Figure A1. The striking lipid profile shift after glucose and metformin exposure is controlled by MXL-3. A subset of 17 fatty acids is shown in the N2 and mxl-3 mutant strains treated with high glucose and high glucose plus metformin. (A) FFAs that increased due to glucose exposure showed a clear tendency to return to the levels of the control group in the presence of metformin. (B) FFAs are unable to change in the mxl-3 mutant due to glucose or metformin. Graphs depict the mean ± SEM of three independent experiments; * p < 0.05, *** p < 0.001 vs. fatty acids in the control group, one-way ANOVA with Bonferroni's post hoc test using GraphPad Prism. Figure A2. Indirect measurement of the fatty acid biosynthetic pathway. Specific steps in the fatty acid biosynthetic pathway were indirectly measured from substrate/product ratios. If the relationship increased, a decrease in activity was inferred. Each graph shows the mean ± SEM of three independent experiments; * p < 0.05, ** p < 0.01 vs. fatty acids in the control group, one-way ANOVA with Bonferroni's post hoc test using GraphPad Prism. Figure A2. Indirect measurement of the fatty acid biosynthetic pathway. Specific steps in the fatty acid biosynthetic pathway were indirectly measured from substrate/product ratios. If the relationship increased, a decrease in activity was inferred. Each graph shows the mean ± SEM of three independent experiments; * p < 0.05, ** p < 0.01 vs. fatty acids in the control group, one-way ANOVA with Bonferroni's post hoc test using GraphPad Prism.