Opposing Roles of Sphingosine 1-Phosphate Receptors 1 and 2 in Fat Deposition and Glucose Tolerance in Obese Male Mice

Abstract

Sphingosine 1-phosphate (S1P) is a bioactive sphingolipid that regulates fundamental cellular processes such as proliferation, migration, apoptosis, and differentiation through 5 cognate G protein–coupled receptors (S1P₁-S1P₅). We previously demonstrated that blockade of S1P₂ signaling in S1P₂-deficient mice attenuates high-fat diet-induced adipocyte hypertrophy and glucose intolerance and an S1P₂-specific antagonist JTE-013 inhibits, whereas an S1P₁/S1P₃ dual antagonist (VPC23019) activates, adipogenic differentiation of preadipocytes. Based on those observations, this study examined whether an S1P₁-specific agonist, SEW-2871, VPC23019, or their combination acts on obesity and glucose intolerance in leptin-deficient ob/ob mice. The oral administration of SEW-2871 or JTE-013 induced significant reductions in body/epididymal fat weight gains and epididymal/inguinal fat adipocyte sizes and improved glucose intolerance and adipocyte inflammation in ob/ob mice but not in their control C57BL/6J mice. Both SEW-2871 and JTE-013 decreased messenger RNA levels of tumor necrosis factor-α and CD11c, whereas they increased those of CD206 and adiponectin in the epididymal fats isolated from ob/ob mice with no changes in the levels of peroxisome proliferator activated receptor γ and its regulated genes. By contrast, VPC23019 did not cause any such alterations but counteracted with all those SEW-2871 actions in these mice. In conclusion, the S1P₁ agonist SEW-2871 acted like the S1P₂ antagonist JTE-013 to reduce body/epididymal fats and improve glucose tolerance in obese mice. Therefore, this study raises the possibility that endogenous S1P could promote obesity/type 2 diabetes through the S1P₂, whereas exogenous S1P could act against them through the S1P₁.

Obesity and its associated lifestyle diseases including type 2 diabetes, hypertension, dyslipidemia, and cardiovascular diseases are among the most pressing global health issues. In obesity, adipocyte hypertrophy leads to enhanced secretion of chemokine to recruit macrophages (1), especially proinflammatory M1 macrophages rather than anti-inflammatory M2 macrophages (2). Obesity-induced, low-grade white adipose tissue inflammation and resultant insulin resistance are considered the main pathology underlying these diseases; however, detailed mechanisms have not been fully elucidated (3).

Sphingosine 1-phosphate (S1P) is a bioactive sphingolipid produced by sphingosine kinase 1 and 2 and mediates numerous cellular signals including proliferation, differentiation, migration, and apoptosis through 5 types of cognate G protein–coupled S1P receptors, S1P₁ to S1P₅ (4‐6). The S1P receptor axis participates in a wide range of physiological process, including carcinogenesis/metastasis, and the regulation of circulatory and immune systems (3, 6‐9). Among the 5 receptors, the features and roles of S1P₁ and S1P₂ have been largely clarified in immune systems. S1P₁ is ubiquitously expressed among various tissues and coupled to Gα_i/o. S1P₁ regulates trafficking/migration of various cells including lymphocytes, neutrophils, dendritic cells, osteoclasts, and macrophages (10), and plays pivotal roles in maintaining the integrity of the vascular system (11). FTY720 (fingolimod), an immunosuppressant approved for the treatment of multiple sclerosis, activates S1P₁ and thereby downregulates it from the cell surface, which irreversibly prevents S1P₁ signaling and promotes lymphocyte homing (12). S1P₂ is also widely expressed and coupled with Gα_12/13 and others to activate a small G protein Rho. S1P₂ signaling could act against S1P₁ signaling (13), or in concert with it (14).

Unfortunately, limited studies have examined on the roles of the S1P receptor axis in obesity and diabetes. We have previously demonstrated the beneficial effect of S1P₂ blockade on adipocyte hypertrophy and insulin resistance induced by high-fat diet (HFD) (15). In that study, HFD-induced glucose intolerance, insulin resistance, adipocyte hypertrophy, and adipose tissue inflammation were ameliorated in S1P₂-deficient mice (16) compared with wild-type mice. Combined with results from in vitro pharmacological study using preadipocyte cell lines, we speculated that their proliferation is enhanced by S1P₁/S1P₃ signaling but attenuated by S1P₂ signaling, whereas its differentiation to adipocyte is accelerated by S1P₂ signaling but suppressed by S1P₁ signaling (17). These results raise the notion that S1P₁ and S1P₂ exert opposite effects on adipocyte functions. Because the S1P₂-specific antagonist JTE-013 (17) reduced body weight and prevented adipocyte hypertrophy in genetically obese/diabetic leptin-deficient ob/ob mice (15), this study examined how the S1P₁-specific agonist SEW-2871 (18) and the S1P₁/S1P₃ dual antagonist VPC23019 (19, 20) act on adipose hypertrophy/inflammation and glucose intolerance in ob/ob mice and control wild-type mice.

Materials and Methods

Animals

Male leptin-deficient ob/ob mice (C57BL/6JHamSlc-ob/ob) and control wild-type mice (C57BL/6JHamSlc), which were first established in Hamamatsu University in 2000, were purchased from Japan SLC. They were fed ad libitum a CE-2 standard powder dry rodent diet (Japan Clea) consisting of 51.4% (g/g) carbohydrates, 24.9% crude protein, 4.6% crude fat, and 3.7% crude fiber up to age 10 weeks, and then ad libitum CE-2 with/without 40 mg/kg JTE-013 (Cayman Chemical), SEW-2871 (Cayman Chemical), or VPC23019 (Santa Cruz) for 12 weeks. Mice were housed in an air-conditioned room (24-26 °C; 40%-60% humidity) kept on a 12-hour dark (8 Pm-8 Am)/light cycle. Body weight was measured at 10 Am every week, and daily food intake was measured 3 times (at the same time Monday, Wednesday, and Friday) a week. Only in some experiments were daily food intakes calculated from 4-cycle measurements of the sum values from 3 consecutive days between weeks 2 and 4 after drug administration. All procedures for animal experiments were conducted in accordance with protocols approved by the Gifu Institutional Animal Care Committee (Permission no. 23-4), and were performed according to the ethical guidelines adopted by Act on Welfare and Management of Animals.

Serum Biochemistry

Blood was collected by cardiocentesis from isoflurane anesthetized mice at 11 Am after 15 hours’ fasting and used for serum preparation. Serum levels of glucose, immunoreactive insulin, triglycerides (TGs), total cholesterol (T-Cho), aspartate transaminase (AST), and alanine transaminase (ALT) were measured using the Quick-auto-neo-GLU-HK (Shino-Test), Insulin Rat enzyme-linked immunosorbent assay (ELISA) kit (catalog No. ERINS, RRID: AB_2848197, Thermo Fisher Scientific), L-type Wako TG M, L-type Wako CHO M, L-type Wako AST J2, and L-type Wako ALT J2 (Fujifilm-Wako), respectively.

Glucose Tolerance Test

Glucose tolerance activity was evaluated by a conventional intraperitoneal (2 g/kg) glucose tolerance testing (ip-GTT) performed at 11 Am after 15 hours’ fasting. Blood was collected from the tail vein every 30 minutes, and blood glucose levels were measured using the Medisafe Mini GR102 (Terumo).

Histology

Epididymal/inguinal fat tissues were isolated from isoflurane anesthetized mice, fixed in 10% neutral-buffered formalin, and embedded in paraffin. Then, 3-µm-thick sections were cut, deparaffinized, and stained using rabbit anti-uncoupling protein-1 (UCP1) antibody (1/200 dilution; catalog No. ab10983, RRID: AB_2241462, Abcam), mouse anti-CD11c antibody (1/100; catalog No. 60258-1-Ig, RRID: AB_2881379, Proteintech), or goat anti-CD206 antibody (1/100; catalog No. sc-34577, RRID: 2144904, Santa Cruz) as primary antibodies, and Alexa Fluor 594-labeled donkey anti-rabbit immunoglobulin G (IgG) (1/200; catalog No. A21207, RRID: AB_141637, Invitrogen), Alexa Fluor 594-labeled donkey anti-mouse IgG (1/200; catalog No. A21203, RRID: AB_141633, Invitrogen), or Alexa Fluor 488-labeled donkey anti-goat IgG (1/200; catalog No. A11055, RRID: AB_2534102, Invitrogen) as secondary antibodies. The section slides were examined using an LSM710 confocal laser-scanning microscope (Zeiss). To evaluate adipocyte sizes, formalin-fixed adipose tissue sections were stained with hematoxylin/eosin, and the long axis of 300 independent mature adipocytes (per mouse) was measured by manual tracing using a BZ-II image analysis application (Keyence). Liver sections were examined by hematoxylin/eosin (HE) staining.

Quantitative Polymerase Chain Reaction

Quantitative polymerase chain reaction (PCR) was performed to determine messenger RNA (mRNA) levels of various adipocyte-specific genes. Total RNA was isolated using Trizol (Invitrogen) and purified using RNase-free DNase (Promega). Reverse transcription was performed using a PrimeScript Reverse Transcriptase (Takara), and the produced complementary DNA was amplified using a SYBR Premix Ex Taq Kit (Takara). PCR amplification was performed as follows: 1 cycle of 95 °C for 30 seconds and 40 cycles of 95 °C for 5 seconds and 60-62 °C for 30 seconds using a Thermal Cycler Dice (Takara). Oligonucleotide primers were designed based on sequences from the GenBank database: tumor necrosis factor-α (Tnfa), 5′-ACGGCATGGATCTCAAAGAC-3′ (f) and 5′-AGATAGCAAATCGGCTGACG-3′(r); CD11c (Cd11c), 5′-CTGGATAGCCTTTCTTCTGCTG-3′ (f) and 5′-GCACACTGTGTCCGAACTC-3′ (r); CD206 (Cd206); 5′-CTGTGTAGTTGTGATTGGTGGC-3′(f) and 5′-CTTTTGGAGTAGTGGTTGGAGAA-3′(r); adiponectin (Adipoq), 5′-AAGGACAAGGCCGTTCTCT-3′ (f) and 5′-TATGGGTAGTTGCAGTCAGTTGG-3′ (r); peroxisome proliferator activated receptor γ 2 (Pparγ2), 5′-GCATCAGGCTTCCACTATGGA-3′ (forward: f) and 5′-AAGGCACTTCTGAAACCGACA-3′ (reverse: r); fatty acid binding protein 4 (Fabp4), 5′-AAAGACAGCTCCTCCTCGAAGGTT-3′ (f) and 5′-TGACCAAATCCCCATTTACGC-3′ (r); lipoprotein lipase (Lpl), 5′-AGGATGCAACATTGGAGAAG-3′ (f) and 5′-TCTCTTGGCTCTGACCTTGT-3′ (r); uncoupling protein-1 (Ucp1), 5′-GCCTTCAGATCCAAGGTGAA-3′ (f) and 5′-TAAGCCGGCTGAGATCTTGT-3′; glyceraldehyde-3-phosphate dehydrogenase (Gapdh), 5′-GGCATTGTGGAAGGGCTCAT-3′ (f) and 5′-GACACATTGGGGGTAGGAACA-3′ (r). Expression levels were quantified using the comparative CT method with housekeeping Gapdh levels for normalization.

Statistics

All experimental data were presented as mean ± SD with sample numbers (n). Statistical comparisons were performed using an unpaired 2-tailed t test; significance was defined as P less than .05.

Results

SEW2871 Reduces Body/Fat Weight Gain in ob/ob Mice

Before drug administration at age 10 weeks, ob/ob mice displayed much higher body weights than their control wild-type (C57BL/6J) mice; 49.3 ± 3.1 g (mean ± SD [n = 12]) vs 25.9 ± 2.3 g (n = 12) (Fig. 1A). Both JTE-013 and SEW-2871 (with the same dosage of 40 mg/kg in their diets) did not affect body weight gains in wild-type mice until week 11 after their administration but significantly reduced them in ob/ob mice from week 4 (see Fig. 1A). Daily food intake levels, which originally differed between wild-type and ob/ob mice, were not altered by 4, 8, or 12 weeks of JTE-013 or SEW-2871 administration in either group of mice (Fig. 1B); however, both drugs significantly reduced epididymal fat deposition in ob/ob mice, but not in wild-type mice (Fig. 1C).

Figure 1.

Effects of JTE-013 (S1P₂ antagonist) and SEW-2871 (S1P₁ agonist) on body weight, food intake, and epididymal fat weight in wild-type (C57BL/6J) and ob/ob mice. A, Body weight change in wild-type (open symbols) and ob/ob (closed symbols) male mice fed the control diet (Con), JTE-013 (40 mg/kg)-formulated diet (JTE), or SEW-2871 (40 mg/kg)-formulated diet (SEW) after age 10 weeks. B, Daily food intake (g) at week 4, 8, and 12. Daily food intake was measured 3 times a week (Monday, Wednesday, and Friday) and then averaged. C, Epididymal fat weights at week 12. Data are mean ± SD (sample numbers [n] in parentheses) and the suppressive effects by JTE-013 or SEW-2871 are statistically significant at *P less than .05, **P less than .01, and ***P less than .001 in t test.

SEW2871 Improves Glucose Intolerance and Reduces Fasting Blood Glucose Levels in ob/ob Mice

The ip-GTT was conducted on wild-type and ob/ob mice that were treated with/without JTE-013 and SEW-2871 for 12 weeks. No significant alteration was found in the glucose level transition in wild-type mice among the 3 groups (Fig. 2A), although the area under the curve (AUC) values in JTE-013-treated mice were slightly (11.5%) but significantly (P < .05) smaller than nontreated mice (see Fig. 2A, inset). By contrast, both JTE-013 and SEW-2871 significantly improved glucose intolerance in ob/ob mice in some glucose level peaks (Fig. 2B) and the AUC in which 17.1% and 18.8% reductions were observed, respectively (see Fig. 2B, inset). Routine serum biochemistry revealed that ob/ob mice displayed much higher levels of fasting blood glucose, fasting blood (immunoreactive) insulin, TAG, T-Cho, AST, and ALT (Fig. 2C-2H, respectively). Both JTE-013 and SEW-2871 markedly reduced the levels of fasting blood glucose and TGs only in ob/ob mice without affecting insulin, total cholesterol, AST, and ALT levels, although SEW-2871 significantly reduced only the fasting blood insulin levels in wild-type mice (Fig. 2D).

Figure 2.

Effects of JTE-013 and SEW-2871 on glucose tolerance and serum biochemical parameters in wild-type and ob/ob mice. A and B, Intraperitoneal glucose tolerance test (ip-GTT) in A, wild-type and B, ob/ob mice fed the control diet (Con), JTE-013 diet (JTE), or SEW-2871 diet (SEW) for 12 weeks. Areas under the curve (AUC) are presented in inset graphs. C to H, Serum levels of C, fasting blood glucose; D, fasting blood (immunoreactive) insulin; E, triacylglycerol (TAG); F, total cholesterol (T-Cho); G, aspartate transaminase (AST); and H, alanine transaminase (ALT) in wild-type and ob/ob mice fed the aforementioned diets for 12 weeks. Data are mean ± SD (n in parentheses), and the suppressive effects by JTE-013 or SEW-2871 are statistically significant at *P less than .05 and **P less than .01 in t test.

SEW2871 Suppresses Adipocyte Hypertrophy and Inflammation

Adipose tissue expansion occurs mainly through 2 processes, namely, expansion of existing adipocytes (hypertrophy) and/or recruitment of new adipocytes (hyperplasia). Hypertrophic (rather than hyperplasic) obesity is known to correlate with insulin resistance and impaired adipocyte differentiation, inflammation, remodeling, and fibrosis (21). The administration of JTE-013 or SEW-2871 reduced adipocyte sizes in epididymal fats isolated from ob/ob mice but not those from wild-type mice (Fig. 3A and 3B). The administration of JTE-013 and SEW-2871 also reduced adipocyte sizes in epididymal fats isolated from ob/ob mice as well as those from wild-type mice (Fig. 3C and 3D). Immunohistochemistry revealed the predominant expression of an M1 macrophage marker CD11c (red) in epididymal fats of nontreated ob/ob mice, whereas the expression of an M2 macrophage marker CD206 (blue) was rather predominant in epididymal fats of JTE-013– or SEW-2871–treated ob/ob mice (Fig. 3E). In addition, liver histology (by HE staining) did not reveal any significant alteration on JTE-013 or SEW-2871 in ob/ob mice (Fig. 3F), suggesting that adipocyte infiltration into the liver was not affected by either drug.

Figure 3.

Effects of JTE-013 and SEW-2871 on adipocyte hypertrophy and inflammation in epididymal and inguinal fats of wild-type and ob/ob mice. A to D, Representative hematoxylin/eosin (HE)-stained image of cross-sectional A, epididymal and C, inguinal adipocytes and their cell area sizes (B and D, respectively) from wild-type and ob/ob mice fed the control diet (Con), JTE-013 diet (JTE), or SEW-2871 diet (SEW) for 12 weeks. The suppressive effects by JTE-013 or SEW-2871 are statistically significant at *P less than .05, **P less than .01, ***P less than .001 in t test. Bars: 100 µm. E, Immunochemical CD11c and CD206 staining of epididymal adipocytes isolated from ob/ob mice. The nuclei were stained with DAPI (4′,6-diamidino-2-phenylindole). Bars: 50 µm. F, HE-stained hepatic sections isolated from ob/ob mice. Bars: 100 µm.

VPC23019 Alone Did Not Affect Fat Deposition and Glucose Intolerance but Counteracted With SEW-2871

In our previous in vitro experiments, VPC23019 (10 µM) inhibited preadipocyte proliferation but promoted its differentiation to adipocytes whereas JTE-013 (10 µM) promoted its proliferation but inhibited its differentiation, and CYM-50358 (S1P₄-specific antagonist; 1 µM) had no effect (15). We examined the effect of the same dose (40 mg/kg) of VPC23019 on wild-type and ob/ob mice in another set of experiments. Although the body weights of both wild-type and ob/ob mice were lower than the previous respective mouse lot (in Fig. 1A), ob/ob mice again displayed much higher weights than control mice: 42.0 ± 2.5 (g) vs 22.1 ± 0.7 (g) (Fig. 4A). VPC23019 treatment did not alter body weight gain (see Fig. 4A); food intake (Fig. 4B); epididymal fat weight (Fig. 4C), glucose tolerance (Fig. 4D), fasting blood glucose (Fig. 4E); fasting blood insulin (Fig. 4F); serum levels of TGs (Fig. 4G), T-Cho (Fig. 4H), AST (Fig. 4I), and ALT (Fig. 4J); epididymal fat adipocyte hypertrophy (Fig. 4K and 4L); inguinal fat adipocyte hypertrophy (Fig. 4M and 4N); and CD11c/CD206 staining of epididymal fat adipocyte (Fig. 4O) and liver histology (Fig. 4P) in both wild-type and ob/ob mice.

Figure 4.

Systemic effects of VPC23019 (S1P₁/S1P₃ antagonist) on obesity, glucose intolerance, and adipocyte hypertrophy/inflammation. A, Body weight change in wild-type (open symbols) and ob/ob (closed symbols) male mice fed the control diet (Con or C) or VPC23019 (40 mg/kg)-formulated diet (VPC or V) after age 10 weeks. B, Daily food intake calculated from 4 cycles of the sum values from 3 consecutive days between weeks 2 and 4 after food change. C, Epididymal fat weights at week 12. D, Intraperitoneal glucose tolerance test (ip-GTT) test in wild-type (left) and ob/ob (right) mice fed Con or VPC diet for 12 weeks. Areas under the curve (AUC) are presented in inset graphs. E to J, Serum levels of E, fasting blood glucose; F, fasting blood (immunoreactive) insulin; G, triacylglycerol; H, total cholesterol; I, aspartate transaminase (AST); and J, alanine transaminase (ALT) in wild-type and ob/ob mice fed the aforementioned diets for 12 weeks. K to N, Representative hematoxylin/eosin (HE)-stained image of K, epididymal and M, inguinal adipocytes and their cell area sizes (L and N, respectively) from wild-type and ob/ob mice. Bars: 100 µm. O, Immunochemical CD11c and CD206 staining of epididymal adipocyte sections prepared from ob/ob mice. The nuclei were stained with DAPI (4′,6-diamidino-2-phenylindole). Bars: 50 µm. P, HE-stained hepatic sections prepared from ob/ob mice. Bars: 100 µm. Data are mean ± SD (n in parentheses) and the effects of VPC23019 were not statistically significant in any parameters.

Next, we examined the effect of VPC23019 on SEW-2871–induced such alterations. The VPC23019 coaddition to the diet prevented SEW-2871–induced body weight/epididymal fat weight loss without altering daily food intakes (Fig. 5A-5C). VPC23019 coaddition also prevented SEW-2871–induced improvement of glucose intolerance (Fig. 5D) and decreases in blood glucose, insulin, and TG levels (Fig. 5E-5G) without affecting T-Cho, AST, and ALT levels (Fig. 5H-5J) and epididymal/inguinal adipocyte atrophy (Fig. 5K and 5L).

Figure 5.

Preventive effects of VPC23019 (S1P₁/S1P₃ antagonist) on SEW-2871 (S1P1 agonist)-induced alterations in obesity, glucose intolerance, and adipocyte hypertrophy/inflammation. A, Body weight change in wild-type (open symbols) and ob/ob (closed symbols) male mice fed with the control diet (Con or C), SEW-2871 (40 mg/kg)-formulated diet (SEW or S), SEW-2871 and VPC23019 (40 mg/kg each)-formulated diet (SEW + VPC or SV) after age 10 weeks. B, Daily food intake at week 12. C, Epididymal fat weights at week 12. D, Intraperitoneal glucose tolerance test (ip-GTT) test in wild-type (left) and ob/ob (right) mice fed Con, SEW, or SEW + VPC diet for 12 weeks. Areas under the curve (AUC) are presented in inset graphs. E to J, Serum levels of E, fasting blood glucose; F, fasting blood (immunoreactive) insulin; G, triacylglycerol; H, total cholesterol; I, aspartate transaminase (AST); and J, alanine transaminase (ALT) in wild-type and ob/ob mice fed the aforementioned diets for 12 weeks. K and L, Representative cross-sectional hematoxylin/eosin (HE)-stained images and cell area sizes of K, epididymal and L, inguinal adipocytes from wild-type and ob/ob mice. Bars: 100 µm. Data are mean ± SD (n in parentheses) and the VPC cotreatments mostly prevented the actions by SEW.

Effects on the Expression of Genes Involved in Inflammation and Adipogenesis in Adipose Tissues

Next, we examined the effects of JTE-013, SEW-2871, and VPC23019 on the expressions of genes involved in inflammation (Tnfa, Cd11c, and Cd206) and adipogenesis (as Adipoq induction) both in epididymal and inguinal fats (Fig. 6A-6D, respectively). Gene expression of TNFα (a classic proinflammatory cytokine and the first “adipokine” that is produced from adipose tissue and regulates obesity and insulin sensitivity [22]) and CD11c (M1 polarization marker [23]) was highly upregulated in the epididymal but not in the inguinal fat of ob/ob mice, which was dramatically suppressed by JTE-013 or SEW-2871 but not VPC23019 (Fig. 6A-6D). By contrast, the gene expression of CD206 (M2 polarization marker [23]) was suppressed in the epididymal (not inguinal) fat of ob/ob mice, which was reversed by JTE-013 or SEW-2871 but not VPC23019 (Fig. 6E and 6F). Furthermore, the gene expression of adiponectin (the “adipokine” that is produced from adipose tissue and enhances insulin sensitivity [24]) was upregulated both in epididymal and inguinal fats of ob/ob mice on JTE-013 and SEW-2871 administration, although the control levels were not significantly altered between wild-type and ob/ob mice (Fig. 6G and 6H). Although Adipoq is one of the PPARγ-responsive genes, both JTE-013 and SEW-2871 did not influence the expression of Pparg (Fig. 7A and 7B) and other PPARγ-responsive genes such as Fabp4 (Fig. 7C and 7D) and Lpl (Fig. 7E and 7F) both in epididymal (Fig. 7A, 7C, and 7E) and inguinal (Fig. 7B, 7D, and 7F) fats of wild-type and ob/ob mice.

Figure 6.

Expression of adipocyte inflammation marker genes in epididymal and inguinal fats from wild-type and ob/ob mice fed the control, JTE-013, SEW-2871, or VPC23019 diet for 12 weeks. Messenger RNA levels of A and B, tumor necrosis factor α (Tnfa); C and D, CD11c (Cd11c); E and F, CD206 (Cd206); and G and H, adiponectin (Adipoq) in the A, C, E, and G, epididymal and B, D, F, and H, inguinal fats of wild-type and ob/ob mice were normalized by that of GAPDH (glyceraldehyde-3-phosphate dehydrogenase) and expressed with that of wild-type mice fed the same control diet as 1. Data are mean ± SD (n in parentheses); *P less than .05 and **P less than .01 vs wild-type samples (black asterisks) or control samples (red and blue asterisks for JTE-013 and SEW-2871 samples, respectively).

Figure 7.

Expression of Pparg and its regulated genes in epididymal and inguinal fats from wild-type and ob/ob mice fed the control, JTE-013, SEW-2871, or VPC23019 diet for 12 weeks. Messenger RNA levels of A and B, PPARγ (Pparg); C and D, fatty acid binding protein 4 (Fabp4); and E and F, lipoprotein lipase (Lpl) in the A, C, and E, epididymal and B, D, and F, inguinal (fats of wild-type and ob/ob mice were normalized by that of GAPDH (glyceraldehyde-3-phosphate dehydrogenase) and expressed with that of wild-type mice fed the same control diet as 1. Data are mean ± SD (n in parentheses) and no statistically significant differences were observed.

JTE-013/SEW-2871–induced suppression of glucose intolerance (see Fig. 2B) and inguinal adipocyte hypertrophy (not significant for JTE-013; see Fig. 3C and 3D) in ob/ob mice could be attributed to adipocyte browning, the transdifferentiation of UCP1-negative white adipocytes to UCP1-positive thermogenic beige cells in subcutaneous fats such as inguinal fats (25, 26); however, there was no specific UCP1 protein induction (Fig. 8A) or Ucp1 mRNA induction (Fig. 8B) on JTE-013, SEW-2871, or VPC23019 administration in inguinal fats of both wild-type and ob/ob mice. Those drugs also did not induce Ucp1 expression in epididymal fats (Fig. 8C).

Figure 8.

Expression of uncoupling protein-1 (UCP1) in inguinal and epididymal fats. A, Immunochemical UPC1 (Ucp1) staining of inguinal adipocyte sections isolated from wild-type and ob/ob mice fed the control, JTE-013, SEW-2871, or VPC23019 diet for 12 weeks. The nuclei were stained with DAPI (4′,6-diamidino-2-phenylindole). B and C, Messenger RNA levels of UCP1 in the B, inguinal and C, epididymal fats of wild-type and ob/ob mice were normalized by that of GAPDH (glyceraldehyde-3-phosphate dehydrogenase) and expressed with that of wild-type mice fed the same control diet as 1. Data are mean ± SD (n in parentheses) and no statistically significant differences were observed.

Discussion

Our previous in vitro study using adipocyte precursor 3T3F442A and 3T3-1 cells demonstrated that the blockade of S1P₂ signaling by JTE-013 upregulates cell proliferation while suppressing adipogenic differentiation, whereas the inhibition of S1P_1/3 signaling by VPC23019 induces completely opposite effects (15). These results indicate that S1P produced and secreted from those cells acts on the S1P₂ to suppress proliferation and promote differentiation and vice versa on the S1P₁/S1P₃. Furthermore, our previous in vivo (mouse) study revealed that HFD-induced glucose intolerance, epididymal adipocyte hypertrophy, and mRNA-level changes in Adipoq (a decrease) and Leptin (an increase) were all dampened in S1P₂-deficient mice and that JTE-013 treatment for 4 weeks resulted in weight reduction and improved glucose tolerance in ob/ob mice (15). These results raise the idea that specific S1P₁ activation with specific agonists might induce similar or possibly better therapeutic effects on obese mice because it would not be largely affected by endogenous S1P levels in situ compared with antagonists such as JTE-013. We have chosen SEW-2871 as the S1P₁-selective agonist that has a 50% effective concentration of 10 to 20 nM on human and mouse receptors (18). SEW-2871 can activate all known signals downstream of the S1P₁ and induce receptor internalization/recycling (like endogenous S1P) but not receptor degradation (like phosphorylated FTY720) (18). We have also selected VPC23019 as the S1P₁/S1P₃-selective antagonist because of its potent effects on both cell lines (15) and mice (27).

We found that SEW-2871, just like JTE-013, reduced body/epididymal fat weight gains without affecting food intake (see Fig. 1A-1C), improved glucose intolerance (see Fig. 2B), reduced fasting blood glucose (see Fig. 2C) and TG levels (see Fig. 2E), and inhibited adipocyte hypertrophy (see Fig. 3A-3D) and M1 polarization (see Figs. 3E and 6C) only in ob/ob mice without affecting food intake (see Fig. 1B), serum levels of insulin, T-Cho, AST, and ALT (see Fig. 2D, 2G, and 2H), and liver adipogenesis and inguinal adipocyte browning (see Fig. 8A and 8B). By contrast, VPC23019 failed to demonstrate all such changes (see Figs. 4A-4P and 8A-4C) but completely abolished the effects of SEW-2871 (see Fig. 5A, 5C-5G, 5K, and 5L). These results suggest that (relatively low levels of) endogenous S1P acts mainly through S1P₂ (rather than S1P₁/S1P₃) to promote adipogenesis and obesity associated with glucose intolerance and adipocyte inflammation, whereas (high levels of) exogenous S1P₁ agonist acts only or mainly through S1P₁ to prevent or counteract all those processes.

The effects of JTE-013 and SEW-2871 were most apparent in reducing body weight (see Fig. 1A), impairing epididymal fat deposition (see Fig. 1C) by visceral adipocyte hypertrophy (see Fig. 3A and 3B), and inhibiting glucose tolerance (see Fig. 2B) and adipocyte inflammation (see Figs. 3E and 5A, 5C, 5E, and 5G). Weight reduction is not always associated with improved glucose tolerance and hepatic steatosis. Mice overexpressing preadipocyte factor-1 were resistant to HFD-induced obesity but exhibited severe insulin resistance and diacylglycerol accumulation in skeletal muscle (28). Mice deficient in mitochondrial transcription in adipocytes had reduced body weight but developed severe insulin resistance, high blood TG levels, and hepatic steatosis (29). These mice were considered animal models of lipoatrophic diabetes that show lipotoxicity and insulin resistance due to the deposition of excessive lipids both in liver and skeletal muscle caused by adipocyte dysfunction. Both JTE-013 and SEW-2871 ameliorated adiposity accompanied by improved glucose intolerance, decreased fasting plasma glucose/TG levels without altering fasting insulin levels in ob/ob mice, which could be attributed to ameliorated epididymal adipocyte inflammation (see Figs. 3E and 6A, 6C, and 6E), and increased adiponectin expression (see Fig. 6G) in PPARγ-independent ways (30) (see Fig. 7A, 7C, and 7E). The latter result contrasted with our previous in vitro study in which JTE-013 suppressed expression of “PPARγ-regulated” genes such as Adipoq, Fabp4, and Lpl in 3T3-F442A preadipocytes (15).

Unexpectedly, both JTE-013 and SEW-2871 did not improve/prevent hepatic steatosis (see Fig. 3F) because JTE-013 has been shown to suppress adipogenesis in epididymal fats of ob/ob mice (15) and prevent palmitate-induced insulin resistance in rat hepatocytes (31). However, JTE-013 and SEW-2871 may improve insulin resistance in the liver independent of steatosis. Both drugs appear to act mainly on hypertrophied visceral adipocytes, although they acted on cultured adipocytes to prevent their adipogenesis (progression to hypertrophy) (15). Our experiment might deny the possibility that their weight-reducing effects were caused by the browning of subcutaneous adipocytes (see Fig. 8A-8C).

In our previous study, S1P₂-deficient mice displayed body and epididymal fat weight gains on HFD feeding just like wild-type mice (15). Therefore, even long-term administration of JTE-013 and SEW-2871 may not contribute to weight loss. However, adipocyte hypertrophy and insulin resistance were significantly ameliorated in S1P₂-deficient mice (15); thus, both drugs could still have such beneficial effects. Pharmacological blockade of S1P₂ could lead to the activation of other S1P receptor subtypes, and their effects largely depend on variable endogenous S1P levels (Fig. 9); notably, a statistically significant positive correlation was found between body mass index and plasma S1P levels (32), and plasma S1P levels are upregulated in streptozotocin-induced diabetic mice (33). Therefore, exogenous S1P₁ agonists might be a better therapeutic option against obesity and type 2 diabetes (see Fig. 9). Furthermore, a recent report shows that S1P₃-deficient mice also exhibited lipodystrophy, exacerbated insulin resistance, and glucose intolerance on an HFD (34); thus, the S1P₃-specific agonists could be another option.

Figure 9.

Proposed illustration of the actions of SEW-2871, JTE-013, and VPC23019 on obesity and glucose intolerance. Endogenously produced S1P acts mainly through S1P₂ to promote adipocyte hypertrophy and glucose intolerance in ob/ob mice, and the S1P₂ antagonist JTE-013 dampens its actions. By contrast, exogenously applied the S1P₁ agonist SEW-2871 activates only S1P₁ to improve adipocyte hypertrophy and glucose intolerance in ob/ob mice, which is prevented in the presence of the S1P1/S1P3 dual antagonist VPC23019. SEW-2871 could be a better option since it does not affect other S1P receptor subtypes and is not affected by endogenous S1P levels.

In conclusion, the S1P₁ agonist SEW-2871 acted like the S1P₂ antagonist JTE-013 in preventing obesity and associated glucose intolerance due to visceral adipocyte hypertrophy and inflammation in ob/ob mice (see Fig. 9). However, in this study, we did not evaluate the effect of JTE-013 or SEW-2871 administration on basal metabolism and oxygen consumption in mice. Therefore, it was not possible to explain the mechanism of weight loss caused by these drugs. We also showed that JTE-013 and SEW-2871 improve diabetes associated with obesity, but whether each can actually be used as an antidiabetic drug requires longer-term observation. Although JTE-013 and SEW-2871 had no effect on food intake, we do not know whether these drugs act through the central nervous system/hypothalamus or the peripheral system such as adipose tissues. Although S1P has been shown to regulate insulin secretion for pancreatic β cells and insulin signaling in the liver and skeletal muscles (35, 36), the clarification of S1P receptor subtypes involved in those processes and the effects of S1P receptor agonists/antagonists remain to be investigated by in vivo experiments like this study. When these questions are resolved, the role of S1P receptors in adipocytes will be better understood and may lead to their clinical application.

Abbreviations

Adipoq

adiponectin

ALT

alanine transaminase

AST

aspartate transaminase

AUC

area under the curve

DAPI

4′,6-diamidino-2-phenylindole

Fabp4

fatty acid binding protein 4

Gapdh

glyceraldehyde-3-phosphate dehydrogenase

HE

hematoxylin/eosin

HFD

high-fat diet

IgG

immunoglobulin G

ip-GTT

intraperitoneal glucose tolerance testing

Lpl

lipoprotein lipase

mRNA

messenger RNA

PCR

polymerase chain reaction

S1P

sphingosine 1-phosphate

T-Cho

total cholesterol

TGs

triglycerides

Tnfa

tumor necrosis factor-α

UCP1

anti-uncoupling protein-1

Financial Support

This work was supported in part by the joint research program of the Institute for Molecular and Cellular Regulation, Gunma University, and a Grant-in-Aid for Transformative Research Areas (A) (No. 22H05577 to I.I.) from the Japan Society for the Promotion of Science. This research also received funding from Nihon Eli Lilly, Sanofi KK, and Takeda Pharmaceutical.

Author Contributions

K.K. designed the research; M.A., M.F., T.K., and I.M. performed the experiments; and M.A., K.K., N.A., I.I., and H.M. wrote and reviewed the manuscript.

Disclosures

The authors have nothing to disclose.

Data Availability

Some or all data sets generated during and/or analyzed during the present study are not publicly available but are available from the corresponding author on request.