Numerous evidence indicates that inflammation in adipose tissue is the primary cause of systemic insulin resistance induced by obesity. Obesity-associated changes in circulating LPS level and hypoxia/HIF-1α activation have been proposed to be involved in boosting obesity-induced inflammation. However, there is poor understanding of what triggers obesity-induced inflammation. In this study, we pinpoint lactate as a key trigger to mediate obesity-induced inflammation and systemic insulin resistance. Specific deletion of Slc16a1 that encodes MCT1, the primary lactate transporter in adipose tissues, robustly elevates blood levels of proinflammatory cytokines and aggravates systemic insulin resistance without alteration of adiposity in mice fed high-fat diet. Slc16a1 deletion in adipocytes elevates intracellular lactate level while reducing circulating lactate concentration. Mechanistically, lactate retention due to Slc16a1 deletion initiates adipocyte apoptosis and cytokine release. The locally recruited macrophages amplify the inflammation by release of proinflammatory cytokines to the circulation, leading to insulin resistance in peripheral tissues. This study, therefore, indicates that lactate within adipocytes has a key biological function linking obesity to insulin resistance, and harnessing lactate in adipocytes can be a promising strategy to break this link.

Lactate, once regarded as anaerobic “metabolic waste” in the past, is now considered to be a key player in metabolic homeostasis under aerobic conditions. Lactate serves three major biological functions: as a major fuel to feed tricarboxylic acid cycle, as a precursor of gluconeogenesis, and as a signaling molecule (1,2). In an experiment using 13C-labeled nutrients to trace the fluxes of circulating metabolites, it was discovered that the circulatory turnover rate of lactate in mice is 1.1-fold and 2.5-fold more than that of glucose on a molar basis under feeding and fasting conditions, respectively (3). This elegant experiment clearly illustrated that lactate is a major metabolic fuel that feeds the tricarboxylic acid cycle under aerobic conditions. Lactate shuttling between different cells and different intracellular compartments is actively involved in the distribution of energy substrate (1,2). Lactate producers, such as skeletal muscle, skin, and adipose tissue, provide energy for working muscle, heart, and brain and serve as a gluconeogenic precursor for the liver (1,2). For example, lactate can exchange between white-glycolytic and red-oxidative fibers in skeletal muscle and heart (4,5) and between astrocytes and neurons (6,7). In addition, lactate may act as a signaling molecule to modulate reactive oxygen species production and increase transcription of genes involved in mitochondrial biogenesis. Lactate can bind a cell surface receptor GPR81 to mediate the antilipolytic action of insulin in adipocytes (8,9). Lactate can also control immune invasion through paracrine activation of GPR81 on antigen-presenting cells in the tumor microenvironment (10), partly explaining the Warburg effect in tumor cells.

The functional importance of lactate in adipose tissue has started to be recognized in recent years (11). Adipose tissue is an active producer of lactate, and 50–70% of ingested glucose is converted to lactate in adipose tissue (12). A series of early studies demonstrated that the magnitude of lactate production from adipose tissue is similar to that of skeletal muscle in humans (1315). It is noteworthy that the increased size of adipocytes in obesity is accompanied by an increase in lactate production in adipose tissue (12). Such an increase in lactate production is considered to be caused by hypoxia in obese adipocytes (16). In adipose tissue, lactate can induce browning of white adipose cells dependent on PPARγ signaling pathway (17). Lactate can also induce production in adipocytes of FGF21, an important regulator of metabolism in the whole body (18). Detailed analysis with white adipose tissue illustrated that monocarboxylate transporter 1 (MCT1), a major transporter of lactate, is expressed in a subpopulation of the white adipose tissue and can be considered as a marker of beiging adipocytes (19). In summary, the currently known functions of lactate in adipose tissue are as follows: 1) adipose tissue is an important source of lactate production, which is increased in obese conditions; 2) lactate can induce browning of beige adipocytes and serves as an oxidative substrate for thermogenesis in adipocytes; and 3) lactate functions as a signaling molecule to mediate insulin inhibition of lipolysis via activating GPR81 (8,9).

To further explore the biological function of lactate in adipose tissue, we generated a mouse model with adipose-specific deletion of Slc16a1 that encodes MCT1, an important lactate transporter among the MCT family, as MCT1–4 mediates lactate transport (20). Intriguingly, high-fat diet (HFD)-induced obesity is not affected by Slc16a1 deletion in adipose tissue. However, insulin resistance is markedly aggravated by Slc16a1 deletion. Through a series of in vivo and in vitro studies, we demonstrate that lactate plays a key role in mediating obesity-induced insulin resistance via modulating inflammation in adipose tissue.

Antibodies and Reagents

Antibodies used in this study were purchased as follows: anti–caspase 3 (cat. no. 9662s), anti–cleaved caspase 3 (9661s), anti-HSP90 (4874s), anti-AKT (4691s), anti–phosphorylated AKT (4060s), and anti-F4/80 (70076s) were from Cell Signaling Technology (Boston, MA), and anti-GAPDH (AC033) and anti-MCT1 (A3013) were from ABclonal (Boston, MA). The secondary antibodies used in the immunoblotting assays were goat anti-mouse IgG (Jackson, Philadelphia, PA) and goat anti-rabbit IgG (Jackson). Antibodies used for flow cytometry were anti-mouse CD45 FITC, anti-mouse F4/80 (BM8) PE-eFlour 610, anti-mouse CD11b (M1/70) FITC, and anti-mouse CD206 allophycocyanin from Thermo Fisher Scientific (Waltham, MA); anti-mouse CD11c BV421 and anti-mouse CD45 APC-Cy7 were from BD Pharmingen (Franklin Lake, NJ). Insulin was a gift from Xiaoying Li’s laboratory at Zhongshan Hospital, Shanghai, China. The selective MCT1 inhibitor AZD3965 (CSN13789) was purchased from CSNpharm (Chicago, IL). The caspase inhibitors Asperosaponin VI (HY-N0265) and Ac-DEVD-CHO (HY-P1001) were purchased from MedChemExpress (Monmouth Junction, NJ). l-lactate was purchased from Sangon (Shanghai, China). IBMX (13347) was purchased from Cayman Chemical (Ann Arbor, MI). Collagenase I (C0130), dexamethasone (D4902), and rosiglitazone (R2408) were purchased from Sigma-Aldrich (St. Louis, MO). The kit to detect the activity of caspase 3 was purchased from Beyotime Biotechnology (Shanghai, China). The kit to detect triglyceride, total cholesterol, and HDL was from ShenSuo UNF (Shanghai, China).

Plasmids and Cell Transfection

Slc16a1-knockout plasmid was generated with LentiCRISPRV1 containing the sequence of single guide RNA 5′-GGATGGATTTGGGAAATGCAT-3′ for murine Slc16a1 knockout. The plasmid Laconic/pcDNA3.1 for lactate detection was purchased from Addgene. Slc16a1 deletion in 3T3-L1 cells were carried out by lentivirus infection as previously reported (21). Lipofectamine 2000 was used in transient transfection for lactate detection in 3T3-L1 cells. Fluorescence was measured in 48 h after the transfection.

Animal Studies

The Slc16a1-flox mice were developed by Shanghai Model Organisms Center (Shanghai, China), and aP2-Cre mice were purchased from the same company. The Slc16a1-knockout mouse was generated by homologous recombination. Briefly, a plasmid containing 4.0 kb 5′ homologous arm, 3.3 kb flox sequence, PGK-Neo-polyA, 4.0 kb 3′ homologous arm, and MC1-TK-polyA selection marker (Supplementary Fig. 1A) was transfected into mouse JM8A3 embryonic stem (ES) cells. The homologous arms were designed from the genomic sequence in the 3rd and 5th introns of the Slc16a1 gene. The ES cells were screened with G418 and GANC and 144 positive clones were identified. Long-segment PCR was later used to select out six ES clones. These ES cells were injected into blastula to generate first-generation chimera mice, which were then used to generate Slc16a1-flox mice (Supplementary Fig. 1B). Successful homologous recombination was confirmed by PCR and sequencing using genomic DNA. All animals were maintained and used in accordance with the guidelines of the Institutional Animal Care and Use Committee of the Shanghai Institute of Nutrition and Health, Chinese Academy of Sciences, (approval no. SINH-2020-CY-1). Mice were maintained on a 12-h light/dark cycle at 25°C. Only male mice were used in the study, and all of the mice were on C57/B6J background. For HFD-induced obesity, mice were fed an HFD (with 60% of kilocalories from fat, cat. no. D12492; Research Diets) starting from 6 weeks of age. Weights of food and mice were observed every week, and metabolic efficiency was detected by the Comprehensive Lab Animal Monitoring System (CLAMS-16; Columbus Instruments) at 14th week after HFD. Glucose tolerance test (GTT) (with 2 g/kg glucose i.p.) was performed at 14th week after HFD and insulin tolerance test (ITT) (with 1 unit/kg insulin i.p.) was performed at 15th week after HFD. The blood glucose level was determined at 0, 15, 30, 60, 90, and 120 min after injection for both GTT and ITT. As for detection of insulin-stimulated AKT phosphorylation in tissues, mice fasted for 6 h were injected with a dose of insulin (3 units/kg body wt) and the tissues were collected 8 min later. Basal level of AKT phosphorylation was detected with tissues from mice injected with saline.

Cell Culture and 3T3-L1 Differentiation

Human embryonic kidney (HEK)293T cells were cultured in DMEM with 100 units/mL penicillin/streptomycin and 10% FBS at 37°C with 5% CO2. RAW264.7 cells (provided by Dr. Wei Lv, Shanghai Institute of Nutrition and Health, Chinese Academy of Science) were cultured in RPMI medium with 10% FBS at 37°C with 5% CO2. 3T3-L1 cells provided by Qiurong Ding, Shanghai Institute of Nutrition and Health, Chinese Academy of Science, were kept in DMEM with 10% newborn cattle’s serum at 37°C with 5% CO2. For 3T3-L1 differentiation, confluent 3T3-L1 cells were cultured with DMEM with FBS for 2 days and then incubated with 0.5 mmol/L IBMX, 1 μmol/L dexamethasone, 20 nmol/L rosiglitazone, and 10 μg/mL insulin for 2 days. The medium was then replaced every other day with DMEM containing 10 μg/mL insulin. For coculture experiment, 3T3-L1 cells were allowed to differentiate for 6 days and then the medium was replaced with DMEM containing 10% FBS without insulin. RAW264.7 cells were added in the upper chamber of transwell with 5 × 105 per chamber with RPMI medium containing no FBS. We performed coculture experiments with a chamber of 0.4 μm aperture and macrophage migration assay with a chamber of 8 μm aperture. Isolation of peritoneal macrophages (PMs) was performed as previously described (21).

Lactate Detection and ELISA

Lactate inside cells were isolated by multigelation. Cells were digested and counted and then suspended with 300 μL diluent water per 106 cells, followed by repeated freezing and thawing, three times. Secreted lactate was directly detected with culture medium or serum. Lactate was detected with a lactate measurement kit purchased from Nanjing Jiancheng Bioengineering Institute (Nanjing, China). The concentrations of IL-6, CCL2, TNFα, and IL-1β were detected with ELISA kits purchased from Shanghai Enzyme-linked Biotechnology (Shanghai, China).

RNA Isolation, RT-PCR, and RNA Sequencing

RNA isolation, RT-PCR, and real-time quantitative PCR were carried out as previously described (21). RNA sequencing (RNA-seq) and analysis were performed by Majorbio (Shanghai, China).

Protein Extraction and Western Blotting

Protein extraction from cells and immunoblotting have previously been described (21). For tissue extract, the mouse tissues were cut and grinded in radioimmunoprecipitation assay with 1% proteinase inhibitor cocktail. The tissue suspension was centrifuged at 12,000 rpm/min for 15 min, and the supernatant was used for immunoblotting.

Immunohistochemistry and Immunofluorescent Staining

Mouse adipose tissues were cut and fixed with 4% PFA for 24 h for paraffin-embedding. Hematoxylin-eosin staining and immunohistochemistry for F4/80 were performed by Servicebio (Wuhan, China). The mounted slides were imaged with an Olympus X51 microscope. For immunofluorescence analysis, tissue sections were deparaffinized in xylene and rehydrated via a graded ethanol series (100%, 90%, 80%, 70%, and 50%) and water. The antigen was retrieved by heat treatment with 0.1 mol/L citrate buffer (pH 6.0), and the sections were blocked with blocking buffer (PBS + 1% normal goat serum + 0.1% trixton-100). The sections were incubated with antibody against cleaved caspase 3 and then incubated with a secondary antibody. The images of the stained sections were captured with an LSM 510 confocal microscope (ZEISS, Jena, Germany).

Flow Cytometry

Stromal vascular fraction (SVFs) were isolated from epididymal white adipose tissue (eWAT) of the mice as previously described (22). The adipose tissue was minced and digested with 1 mg/mL collagenase I for 30 min at 37°C with shaking. The digests were filtered through 100 μm cell strainer and centrifuged at 600g for 5 min. Pellets were resuspended with red cells lysis buffer (BD Pharmingen) for 5 min and then centrifuged at 800g for 3 min. Pellets containing immune cells were collected for flow cytometry. Cells were incubated with antibodies for 30 min and then stained with Live/Dead Aqua (BD Pharmingen) for 10 min and washed twice with PBS before detection. Cells were suspended in PBS for flow cytometry. The cell surface markers used were CD45 for leukocytes, CD11b and F4/80 for macrophages, and CD11c and CD206 for M1 and M2-like subpopulations respectively. Data were analyzed with FlowJo 7.6.

RNA-seq of Mouse Samples

The eWATs from the mice were collected and flash frozen with liquid nitrogen, and RNA was extracted using Trizol reagent. RNA quality was determined by 2100 Bioanalyser (Agilent, Santa Clara, CA) and quantified with ND-2000 (Thermo Fisher Scientific). RNA-seq transcriptome library was prepared according to the instructions for the TruSeq RNA Sample Prep Kit from Illumina (San Diego, CA) and then sequenced on the Illumina HiSeq Xten NovaSeq 6000 sequencer (2 × 150 base pairs read length). The raw paired end reads were trimmed and quality controlled by SeqPrep (https://github.com/jstjohn/SeqPrep) and Sickle (https://github.com/najoshi/sickle) with default parameters. Then clean reads were separately aligned to reference genome with orientation mode using HISAT2 software. The mapped reads of each sample were assembled by StringTie in a reference-based approach. Differentially expressed genes were analyzed with DESeq2 with transcripts per kilobase of exon model per million mapped reads (TPM) as the expression level. The data were analyzed on the free online platform of Majorbio Cloud Platform (www.majorbio.com) (Shanghai Majorbio Bio-pharm Technology Co., Ltd).

Statistical Analysis

All data are shown as mean ± SEM. Quantitation results were analyzed with Student t test. Values of P < 0.05 were considered statistically significant.

Data and Resource Availability

The RNA-seq data were deposited in National Center for Biotechnology Information’s Gene Expression Omnibus (GEO) database (accession no. GSE184760). All other data sets generated during or analyzed during the current study are available from the corresponding author on reasonable request.

MCT1 Mediates Lactate Efflux in Adipocytes and Affects Adipocyte Differentiation

For clarity, we used Slc16a1 to annotate the gene and MCT1 to annotate the protein. We first detected the mRNA expression level of Slc16a1 in different tissues in C57BL/6 mice. Slc16a1 had the highest expression in brown adipose tissue (BAT) among the tissues analyzed (Fig. 1A). Slc16a1 had moderate expression level in inguinal white adipose tissue (iWAT), eWAT, and colon (Fig. 1A). As the other two members of the MCT family, MCT4 (encoded by Slc16a3) and MCT2 (encoded by Slc16a7) were also important transporters in mice, as previously reported (20,23). We compared the expression levels of Slc16a3 and Slc16a7 with those of Slc16a1 in three adipose tissues. Slc16a1 had the highest expression level in all of the adipose tissues among the three MCT family members, with Slc16a7 almost undetectable and Slc16a3 only moderately expressed in iWAT (Fig. 1B). We next analyzed the expression pattern of Slc16a1 during differentiation of 3T3-L1 cells. The expression of Slc16a1 but not Slc16a3 was significantly increased during adipocyte differentiation, similar to the change of an adipocyte differentiation marker, Adipoq (Fig. 1C). Such a result was also consistent with public data sets that revealed that Slc16a1 expression was elevated during adipocyte differentiation in both mouse and human cells (Fig. 1D). Taken together, these results indicate the potential functional significance of MCT1 in adipose tissues.

Figure 1

Slc16a1 is highly expressed in adipose tissues, is upregulated during adipocyte differentiation, and affects lactate efflux in adipocytes. A: mRNA level of Slc16a1 in various tissues of C57BL/6J mice (n = 4 for each group). B: mRNA levels of Slc16a1, Slc16a3, and Slc16a7 in adipose tissues (n = 4 for each group). C: mRNA levels of Slc16a1 and Slc16a3 at different days (D) during 3T3-L1 differentiation (n = 6 for each group). D: Heat map of gene expression of MCT family from GEO data sets GSE20696 and GSE20697. Data are shown in Log2TPM. E: Lactate concentration in culture medium during 3T3-L1 differentiation (n = 4 for each group). F: Western blotting to detect MCT1 knockout efficiency in differentiated 3T3-L1 cells. G: Lactate concentration in culture medium (left) and adipocytes (right) in control and Slc16a1-deleted 3T3-L1 cells after 8 days’ differentiation (n = 6 for each group). H: Lactate concentration in culture medium (left) and adipocytes (right). 3T3-L1 cells were differentiated for 6 days and then treated with 1 μmol/L AZD3965 (AZD) for 2 days (n = 6 for each group). I: Determination of intracellular lactate level (ratio of OD490 to OD562) by a fluorescent dye. 3T3-L1 cells were differentiated for 4 days and then transfected with plasmids, following by treatment with DMSO or 1 μmol/L AZD3965 for 48 h (n = 6 for each group). J: Representative images of Oil Red O staining of control or Slc16a1-deleted 3T3-L1 cells after differentiation for 8 days. Quantification of the staining is shown on the right (n = 4 for each group). Oil Red O was redissolved with isopropanol and detected at OD 510 nm. K: Triglyceride (TG) level of the cells as in J. L: Relative mRNA levels of Adipoq, Fabp4, and Apoe of the cells as in J. M: Representative images of Oil Red O staining of 3T3-L1 cells treated with DMSO or 1 μmol/L AZD3965 for 4 days. N: Triglyceride level of the cells as in M (n = 4 for each group). O: Relative mRNA levels of Adipoq, Fabp4, and Apoe in cells as in M. All the mRNA expression levels were relative to β-actin. All the quantitative data were analyzed with Student t test and are shown as mean ± SEM. *P < 0.05; **P < 0.01; ***P < 0.001. CTR, control.

Figure 1

Slc16a1 is highly expressed in adipose tissues, is upregulated during adipocyte differentiation, and affects lactate efflux in adipocytes. A: mRNA level of Slc16a1 in various tissues of C57BL/6J mice (n = 4 for each group). B: mRNA levels of Slc16a1, Slc16a3, and Slc16a7 in adipose tissues (n = 4 for each group). C: mRNA levels of Slc16a1 and Slc16a3 at different days (D) during 3T3-L1 differentiation (n = 6 for each group). D: Heat map of gene expression of MCT family from GEO data sets GSE20696 and GSE20697. Data are shown in Log2TPM. E: Lactate concentration in culture medium during 3T3-L1 differentiation (n = 4 for each group). F: Western blotting to detect MCT1 knockout efficiency in differentiated 3T3-L1 cells. G: Lactate concentration in culture medium (left) and adipocytes (right) in control and Slc16a1-deleted 3T3-L1 cells after 8 days’ differentiation (n = 6 for each group). H: Lactate concentration in culture medium (left) and adipocytes (right). 3T3-L1 cells were differentiated for 6 days and then treated with 1 μmol/L AZD3965 (AZD) for 2 days (n = 6 for each group). I: Determination of intracellular lactate level (ratio of OD490 to OD562) by a fluorescent dye. 3T3-L1 cells were differentiated for 4 days and then transfected with plasmids, following by treatment with DMSO or 1 μmol/L AZD3965 for 48 h (n = 6 for each group). J: Representative images of Oil Red O staining of control or Slc16a1-deleted 3T3-L1 cells after differentiation for 8 days. Quantification of the staining is shown on the right (n = 4 for each group). Oil Red O was redissolved with isopropanol and detected at OD 510 nm. K: Triglyceride (TG) level of the cells as in J. L: Relative mRNA levels of Adipoq, Fabp4, and Apoe of the cells as in J. M: Representative images of Oil Red O staining of 3T3-L1 cells treated with DMSO or 1 μmol/L AZD3965 for 4 days. N: Triglyceride level of the cells as in M (n = 4 for each group). O: Relative mRNA levels of Adipoq, Fabp4, and Apoe in cells as in M. All the mRNA expression levels were relative to β-actin. All the quantitative data were analyzed with Student t test and are shown as mean ± SEM. *P < 0.05; **P < 0.01; ***P < 0.001. CTR, control.

Close modal

Consistent with the observation that Slc16a1 was elevated during adipocyte differentiation, the lactate concentration in the culture medium was increased during differentiation of 3T3-L1 cells (Fig. 1E). To investigate whether MCT1 had a functional role during adipocyte differentiation, we knocked out the Slc16a1 gene in 3T3-L1 cells with the CRISPR/Cas9 system (Fig. 1F). Deletion of Slc16a1 resulted in a decrease of lactate concentration in the culture medium, together with an increase of lactate concentration inside adipocytes (Fig. 1H). A similar phenomenon was observed when MCT1 was inhibited by a specific inhibitor, AZD3965 (Fig. 1H). Detection of intracellular lactate using a fluorescence marker (24) also revealed that the lactate accumulation inside adipocytes was elevated by AZD3965 treatment (Fig. 1I). These results, therefore, demonstrated that deletion of Slc16a1 or inhibition of MCT1 reduced the net efflux of lactate in adipocytes. As lactate transport via MCT family members is mainly dependent on concentration gradient of lactate and proton, this result also indicated that the lactate concentration inside the adipocytes is higher than that in culture medium.

We next analyzed whether MCT1 affects adipocyte differentiation in 3T3-L1 cells. Deletion of Slc16a1 reduced lipid accumulation in 3T3-L1 cells (Fig. 1J and K). In addition, the expression of three markers of adipocyte maturation was decreased by Slc16a1 deletion (Fig. 1L). Similarly, inhibition of MCT1 by AZD3965 reduced lipid accumulation and the expression of adipocyte maturation markers (Fig. 1M–O). Together, these results indicated that MCT1 has a functional role in adipogenesis.

Deletion of Slc16a1 Reduces Circulating Lactate Level and Has No Effect on HFD-Induced Adiposity

To explore the in vivo function of MCT1 in adipose tissues, we generated a mouse model with specific deletion of Slc16a1 in adipose tissues (Slc16a1-AKO) using the Cre-loxP system (Fig. 2A and Supplementary Figure 1). Successful deletion of Slc16a1 in the mice was confirmed by regular RT-PCR and quantitative RT-PCR, which revealed that Slc16a1 was markedly reduced in all three adipose tissues but not in skeletal muscle (Fig. 2B and C). Consistent with our results revealing that MCT1 mediated lactate efflux in adipocytes (Fig. 1), deletion of Slc16a1 led to a reduction of blood lactate level in mice fed with either normal chow or HFD (Fig. 2D). On the other hand, Slc16a1 deletion elevated the lactate level in eWAT in the mice under both normal chow and HFD conditions (Fig. 2E). Thus, these observations were in agreement with those of previous reports that adipose tissue is an important source of circulating lactate (1315). As MCT1 can transport other short-chain fatty acids in addition to lactate, we analyzed the blood levels of a few representative metabolites including acetate, propionate, and butyrate. We found that Slc16a1 deletion in adipose tissues could not alter the blood levels of these metabolites (Supplementary Fig. 2A), indicating Slc16a1 deletion in adipose tissue mainly affects lactate concentration in the blood. In addition, the concentrations of acetate, propionate, and butyrate in eWAT were not altered by Slc16a1 deletion (Supplementary Fig. 2B). We next investigated the metabolic features of the mice. Slc16a1 deletion slightly reduced the body weight of mice at age 14 weeks old with normal chow feeding (Supplementary Fig. 3B), together with a decrease in food intake (Supplementary Fig. 3B). Thus, the slight reduction of body weight under normal chow is likely caused by the decrease in food intake. In addition, we found no differences between Slc16a1-AKO mice and wild-type mice in oxygen consumption, CO2 production, respiratory exchange ratio (RER), or activities (Supplementary Fig. 3C–E), indicating that the overall metabolic rate was not significantly altered by Slc16a1 deficiency.

Figure 2

Slc16a1 deletion in adipose tissue has no effect on HFD-induced adiposity. A: A diagram to depict the generation of adipose tissue–specific Slc16a1-deleted mice (Slc16a1-AKO) with P1 and P2 denoting two pairs of primers used for PCR identification. B: Confirmation of Slc16a1 deletion by RT-PCR with two sets primers as indicated in A. C: Quantitative RT-PCR to confirm Slc16a1 deletion. D: Lactate concentration in serum of mice fed with normal chow or HFD (n = 5 for normal chow and n = 8 for HFD). E: Lactate concentration in eWAT of the mice fed with normal chow or HFD (n = 4 for each group). F: Body weight of the mice fed with HFD for 15 weeks (n = 5 for WT and n = 12 for Slc16a1-AKO). G: Food intake of the mice. d, days. H: Body composition of the mice at the end of experiments. I: Representative images of hematoxylin-eosin staining of eWAT of the mice. J: Relative mRNA levels of genes involved in lipid metabolism in the eWAT of the mice. K: Oxygen consumption and CO2 production of the mice. hr, hour. L: RER of the mice. The rate was normalized with lean weight. M: Locomotor activity of the mice on x- and z-axes (X-TOT and Z-TOT). The data in KM were measured during a 24-h period including a light and dark cycle. N: The level of triglyceride (TG), total cholesterol (TC), and HDL cholesterol (HDL-C) in the serum of the mice. WT, Slc16a1fl/fl mice; AKO, Slc16a1fl/flaP2-Cre+/− mice. All the mRNA expression levels were relative to β-actin. One-way ANOVA was used for F and G, and Student t test was used for other data. All the quantitative data are shown as mean ± SEM. *P < 0.05; **P < 0.01; ***P < 0.001.

Figure 2

Slc16a1 deletion in adipose tissue has no effect on HFD-induced adiposity. A: A diagram to depict the generation of adipose tissue–specific Slc16a1-deleted mice (Slc16a1-AKO) with P1 and P2 denoting two pairs of primers used for PCR identification. B: Confirmation of Slc16a1 deletion by RT-PCR with two sets primers as indicated in A. C: Quantitative RT-PCR to confirm Slc16a1 deletion. D: Lactate concentration in serum of mice fed with normal chow or HFD (n = 5 for normal chow and n = 8 for HFD). E: Lactate concentration in eWAT of the mice fed with normal chow or HFD (n = 4 for each group). F: Body weight of the mice fed with HFD for 15 weeks (n = 5 for WT and n = 12 for Slc16a1-AKO). G: Food intake of the mice. d, days. H: Body composition of the mice at the end of experiments. I: Representative images of hematoxylin-eosin staining of eWAT of the mice. J: Relative mRNA levels of genes involved in lipid metabolism in the eWAT of the mice. K: Oxygen consumption and CO2 production of the mice. hr, hour. L: RER of the mice. The rate was normalized with lean weight. M: Locomotor activity of the mice on x- and z-axes (X-TOT and Z-TOT). The data in KM were measured during a 24-h period including a light and dark cycle. N: The level of triglyceride (TG), total cholesterol (TC), and HDL cholesterol (HDL-C) in the serum of the mice. WT, Slc16a1fl/fl mice; AKO, Slc16a1fl/flaP2-Cre+/− mice. All the mRNA expression levels were relative to β-actin. One-way ANOVA was used for F and G, and Student t test was used for other data. All the quantitative data are shown as mean ± SEM. *P < 0.05; **P < 0.01; ***P < 0.001.

Close modal

We next focused on the potential effect of Slc16a1 deletion on HFD-induced obesity and metabolic changes. Both wild-type and Slc16a1-AKO mice were fed with HFD for 15 weeks starting at 6 weeks old. During HFD feeding, the body weight gain, food intake, and body composition did not significantly differ between the two groups of mice (Fig. 2F–H). Meanwhile, the adipocyte size in eWAT was not altered by Slc16a1 deletion (Fig. 2I). Similarly, the mRNA levels of a series of genes responsible for lipogenesis and lipolysis in eWAT were not changed by Slc16a1 deletion (Fig. 2J). Analyses with metabolic cage revealed that Slc16a1 deletion did not result in changes in oxygen consumption, CO2 production, RER, or activities in the mice (Fig. 2K–M). Measurement of energy expenditure of the mice with the CalR method also showed no difference between the two groups of mice (25) (Supplementary Fig. 4). However, Slc16a1-deleted mice did show significant increases in serum levels of triglyceride, total cholesterol, or HDL (Fig. 2B). In addition, the blood level of free fatty acids was elevated by Slc16a1 deletion (Supplementary Fig. 2C). On the other hand, the overall morphology and expression of major genes related to lipid metabolism/inflammation in iWAT and BAT were largely not altered by Slc16a1 deletion (Supplementary Figs. 5 and 6). Collectively, these data indicate that Slc16a1 deletion has little effect on adiposity and metabolic rate under HFD condition, while it raised the blood levels of triglyceride, total cholesterol, HDL, and free fatty acid.

Slc16a1 Deletion Aggravates Glucose Tolerance in Peripheral Tissues and Inflammation in Adipose Tissue Under HFD

As obesity is a most important contributor to insulin resistance, we analyzed whether Slc16a1 deletion impacted glucose metabolism. Under the normal chow condition, glucose tolerance and insulin sensitivity were comparable between wild-type and Slc16a1-deleted mice (Supplementary Fig. 3F and G). Surprisingly, both glucose tolerance and insulin tolerance were significantly impaired by Slc16a1 deletion in HFD-fed mice (Fig. 3A and B). Consistently, insulin-stimulated AKT phosphorylation in the skeletal muscle, liver, and eWAT was decreased by Slc16a1 deletion in HFD-fed mice (Fig. 3C). In addition, the concentration of blood insulin level was elevated by Slc16a1 deletion in the HFD-fed mice (Fig. 3D). These data thus indicate that Slc16a1 deletion aggravated HFD-induced insulin resistance while not affecting HFD-induced obesity.

Figure 3

Deletion of Slc16a1in adipose tissue aggravates glucose tolerance and insulin resistance and increases local and systemic inflammation after HFD. A: GTT of the mice at 14th week after HFD (n = 5 for Slc16a1fl/fl [WT] and n = 9 for HFD). Quantification of area under curve (AUC) is shown on the right. B: ITT at 15th week after HFD. AUC is shown on the right. C: Western blotting to detect AKT phosphorylation (p-AKT) in the skeletal muscle, liver, and eWAT of the mice at the end of the experiment. Quantification of phosphorylated AKT relative to total AKT from two independent experiments is shown in the right panel. D: Serum levels of insulin in the mice fed with normal chow (NC) for 19 weeks (male, n = 3 for each group) or HFD for 12 weeks (male, HFD started at 6 weeks old, n = 5 for each group). E: Relative mRNA levels of genes of four adipokines in eWAT of the mice. F: Serum levels of TNFα, IL-1β, and IL-6 in the mice fed with HFD for 12 weeks (male, HFD started at 6 weeks old, n = 6 for each group). G: Relative mRNA levels of genes of macrophage markers and proinflammation cytokines in eWAT of the mice. H, I, and K: Flow cytometry analysis of eWAT SVFs from Slc16a1fl/fl and Slc16a1-AKO mice fed HFD for 12 weeks as in F (n = 9 for Slc16a1fl/fl and n = 8 for Slc16a1-AKO) as follows: the ratio of total immune cells (CD45+) to SVFs (H), the ratio of total macrophages (CD11b+F4/80+) to total immune cells (I), the ratio of CD11c+CD206M1-like and CD11cCD206+ M2-like macrophages to total macrophages (K [scatter plot image is on the left and the quantitation is on the right]). J: Representative images of F4/80 staining of eWAT sections of mice fed HFD for 15 weeks as in B. AKO, Slc16a1fl/flaP2-Cre+/− mice. All the mRNA expression levels were relative to β-actin. All the quantitative data were analyzed with Student t test and are shown as mean ± SEM. *P < 0.05; ***P < 0.001.

Figure 3

Deletion of Slc16a1in adipose tissue aggravates glucose tolerance and insulin resistance and increases local and systemic inflammation after HFD. A: GTT of the mice at 14th week after HFD (n = 5 for Slc16a1fl/fl [WT] and n = 9 for HFD). Quantification of area under curve (AUC) is shown on the right. B: ITT at 15th week after HFD. AUC is shown on the right. C: Western blotting to detect AKT phosphorylation (p-AKT) in the skeletal muscle, liver, and eWAT of the mice at the end of the experiment. Quantification of phosphorylated AKT relative to total AKT from two independent experiments is shown in the right panel. D: Serum levels of insulin in the mice fed with normal chow (NC) for 19 weeks (male, n = 3 for each group) or HFD for 12 weeks (male, HFD started at 6 weeks old, n = 5 for each group). E: Relative mRNA levels of genes of four adipokines in eWAT of the mice. F: Serum levels of TNFα, IL-1β, and IL-6 in the mice fed with HFD for 12 weeks (male, HFD started at 6 weeks old, n = 6 for each group). G: Relative mRNA levels of genes of macrophage markers and proinflammation cytokines in eWAT of the mice. H, I, and K: Flow cytometry analysis of eWAT SVFs from Slc16a1fl/fl and Slc16a1-AKO mice fed HFD for 12 weeks as in F (n = 9 for Slc16a1fl/fl and n = 8 for Slc16a1-AKO) as follows: the ratio of total immune cells (CD45+) to SVFs (H), the ratio of total macrophages (CD11b+F4/80+) to total immune cells (I), the ratio of CD11c+CD206M1-like and CD11cCD206+ M2-like macrophages to total macrophages (K [scatter plot image is on the left and the quantitation is on the right]). J: Representative images of F4/80 staining of eWAT sections of mice fed HFD for 15 weeks as in B. AKO, Slc16a1fl/flaP2-Cre+/− mice. All the mRNA expression levels were relative to β-actin. All the quantitative data were analyzed with Student t test and are shown as mean ± SEM. *P < 0.05; ***P < 0.001.

Close modal

It was reported that adipokines secreted by adipose tissues influence insulin sensitivity of peripheral tissues (26). Thus, we examined the expression levels of a few key adipokines involved in regulating glucose homeostasis. We found that the mRNA levels of adiponectin, resistin, leptin, and RBP4 in eWAT were not altered by Slc16a1 deletion (Fig. 3E).

Another key mechanism that links obesity to insulin resistance is proinflammatory cytokines produced in adipose tissue (2730). Most importantly, using ELISA assays, we found that the blood levels of a few key proinflammatory cytokines, including TNFα, IL-1β, and IL-6, were all significantly elevated by Slc16a1 deletion (Fig. 3F). Consistently, the mRNA levels of Adgre1 and Cd68, two important markers of M1-like macrophages in adipose tissues, were elevated by Slc16a1 deletion (Fig. 3G). Meanwhile, the mRNA level of Ccl2 gene, which encodes a cytokine involved in macrophage recruitment, was also significantly increased in Slc16a1-deleted adipose tissue (Fig. 3G). Flow cytometry analysis with stromal vascular cells isolated from eWAT revealed that the ratio of CD45+ immune cells was much higher in Slc16a1-deleted mice than the ratio in wild-type mice (Fig. 3H). Meanwhile, the proportion of CD11b+F4/80+ adipose tissue macrophages (ATMs) among the immune cells was significantly elevated by Slc16a1 deletion (Fig. 3I). Immunostaining of adipose tissue sections with F4/80 also demonstrated an increase in ATMs, which formed a crown-like structure around adipocytes in Slc16a1-deleted eWAT (Fig. 3J). Further analysis revealed that the M1-like proinflammatory macrophages (CD11c+CD206), but not the M2-like anti-inflammatory macrophages (CD11cCD206+), were increased by Slc16a1 deletion (Fig. 3K). Collectively, these results indicate that Slc16a1 deletion increases M1-like macrophages in adipose tissue and elevates production of proinflammatory cytokines from adipose tissue. As a result, the elevated blood levels of inflammatory cytokines aggravate insulin resistance in the Slc16a1-deleted mice.

Slc16a1 Deletion Alters Cytokine Production in Adipocytes and Cocultured Macrophages

We next explored how MCT1 alters cytokine production at the cellular level. In 3T3-L1 adipocytes, Slc16a1 knockout significantly elevated the mRNA levels of Ccl2 and Il6 (Fig. 4A), two important cytokines responsible for macrophage recruitment and inflammation initiation. Consistently, the protein levels of CCL2 and IL-6 were also increased in the culture medium of Slc16a1-deleted adipocytes (Fig. 4B). The mRNA levels of Ccl2 and Il6 in 3T3-L1 adipocytes were elevated by MCT1 blockage with AZD3965 (Fig. 4C). Likewise, the secretion of these two cytokines from 3T3-L1 adipocytes was increased by AZD3965 treatment (Fig. 4D). Thus, loss of MCT1 function can directly increase the production of CCL2 and IL-6 in adipocytes.

Figure 4

Loss of MCT1 increases cytokine production in adipocytes and promotes migration of macrophages. A: Relative mRNA levels of genes of proinflammatory cytokines of control and Slc16a1-deleted differentiated 3T3-L1 adipocytes (n = 4 for each group). B: Levels of CCL2 and IL-6 in culture medium of cells in A. C: Relative mRNA levels of genes of proinflammatory cytokines in differentiated 3T3-L1 cells with or without treatment of 1 μmol/L AZD3965 starting at the 4th day of differentiation (n = 4 for each group). D: Levels of CCL2 and IL-6 in culture medium of cells in C. E: A diagram to depict the coculture experiment using transwell. RAW264.7 cells can migrate to the other side at the bottom of the apical chamber upon stimulation. RAW264.7 and differentiated 3T3-L1 cells were cocultured for 24 h for FH. F: Representative images of migrated RAW264.7 cells stained with crystal violet. The cocultured 3T3-L1 adipocytes were either wild type or deleted of Slc16a1. The quantitation is shown on the right. G: Relative mRNA levels of genes of proinflammatory cytokines in cocultured RAW264.7 cells as in F (n = 4 for each group). H: Representative images of migrated RAW264.7 cells stained with crystal violet. The cocultured 3T3-L1 adipocytes were pretreated with or without 1 μmol/L AZD3965. The quantitation is shown on the right. I: Relative mRNA levels of genes of proinflammatory cytokines in cocultured RAW264.7 cells as in H (n = 4 for each group). All the mRNA expression levels were relative to β-actin. All the quantitative data were analyzed with Student t test and are shown as mean ± SEM. *P < 0.05; **P < 0.01; ***P < 0.001. CTR, control; KO, knockout.

Figure 4

Loss of MCT1 increases cytokine production in adipocytes and promotes migration of macrophages. A: Relative mRNA levels of genes of proinflammatory cytokines of control and Slc16a1-deleted differentiated 3T3-L1 adipocytes (n = 4 for each group). B: Levels of CCL2 and IL-6 in culture medium of cells in A. C: Relative mRNA levels of genes of proinflammatory cytokines in differentiated 3T3-L1 cells with or without treatment of 1 μmol/L AZD3965 starting at the 4th day of differentiation (n = 4 for each group). D: Levels of CCL2 and IL-6 in culture medium of cells in C. E: A diagram to depict the coculture experiment using transwell. RAW264.7 cells can migrate to the other side at the bottom of the apical chamber upon stimulation. RAW264.7 and differentiated 3T3-L1 cells were cocultured for 24 h for FH. F: Representative images of migrated RAW264.7 cells stained with crystal violet. The cocultured 3T3-L1 adipocytes were either wild type or deleted of Slc16a1. The quantitation is shown on the right. G: Relative mRNA levels of genes of proinflammatory cytokines in cocultured RAW264.7 cells as in F (n = 4 for each group). H: Representative images of migrated RAW264.7 cells stained with crystal violet. The cocultured 3T3-L1 adipocytes were pretreated with or without 1 μmol/L AZD3965. The quantitation is shown on the right. I: Relative mRNA levels of genes of proinflammatory cytokines in cocultured RAW264.7 cells as in H (n = 4 for each group). All the mRNA expression levels were relative to β-actin. All the quantitative data were analyzed with Student t test and are shown as mean ± SEM. *P < 0.05; **P < 0.01; ***P < 0.001. CTR, control; KO, knockout.

Close modal

We next used a coculture assay with transwell to analyze how change of MCT1 in adipocytes affects macrophages (Fig. 4E). Deletion of Slc16a1 in adipocytes significantly increased the migration of cocultured macrophages (Fig. 4F). Interestingly, the cocultured macrophages also had significantly increased Ccl2 and Il6 expression (Fig. 4G). Likewise, AZD3965 treatment elevated macrophage migration (Fig. 4H) and elevated the expression of Ccl2 and Il6 (Fig. 4I). Collectively, these results suggest that loss of MCT1 function in adipocytes can promote macrophage recruitment and boost CCL2 and IL-6 production in macrophages.

Transcriptome Analysis Reveals That Slc16a1 Deletion Is Closely Associated With Functional Changes in Inflammation and Apoptosis

To explore the global effects of Slc16a1 deletion on adipose tissues, we analyzed transcriptional profiling of eWAT in wild-type and Slc16a1-deleted mice fed with HFD. Principal components analysis showed that the transcriptomes of the two groups of mice were divergent (Fig. 5A). Analysis of genes by volcano plot revealed that ∼350 genes were significantly different between the two groups of the mice (Fig. 5B). We further performed analysis with all the transcripts. With the standards of average TPM >1, we identified 25,750 transcripts, among which 357 were significantly upregulated and 185 were significantly downregulated in Slc16a1-deleted eWAT (Fig. 5C).

Figure 5

RNA-seq analysis of tissues reveals features of inflammation and apoptosis in Slc16a1-deleted adipose tissue. A: PCA analysis showing RNA-seq results of eWAT from Slc16a1fl/fl (WT) and Slc16a1-AKO mice fed HFD for 15 weeks. B: Volcano plot showing differentially expressed genes. Red plots, genes significantly upregulated (P < 0.05 and fold change >2) in Slc16a1-AKO compared with Slc16a1fl/fl mice. Green plots, genes downregulated (P < 0.05 and fold change <0.5). C: Venn diagram showing changes of transcripts. Among 25,750 transcripts with average TPM >1, 357 transcripts were significantly upregulated and 185 were significantly downregulated in Slc16a1-deleted eWAT (with absolute log2fold change >1). D and E: GO analysis (D) and KEGG analysis (E) of the 542 transcripts as in C. F: A heat map showing z scores of the genes in a few selected pathways from GO analysis. G: Relative mRNA levels of a few selected genes from the list in F. The expression level was relative to β-actin, and data are shown as mean ± SEM. *P < 0.05 and **P < 0.01. H and J: Gene set enrichment analysis (GSEA) analysis of immune response pathways (H), macrophage activation pathways (I), and apoptotic process (J). AKO, Slc16a1fl/faP2-Cre+/− mice; ER, endoplasmic reticulum; NES, normalized enrichment score; PC1, principal component 1; PC2, principal component 2.

Figure 5

RNA-seq analysis of tissues reveals features of inflammation and apoptosis in Slc16a1-deleted adipose tissue. A: PCA analysis showing RNA-seq results of eWAT from Slc16a1fl/fl (WT) and Slc16a1-AKO mice fed HFD for 15 weeks. B: Volcano plot showing differentially expressed genes. Red plots, genes significantly upregulated (P < 0.05 and fold change >2) in Slc16a1-AKO compared with Slc16a1fl/fl mice. Green plots, genes downregulated (P < 0.05 and fold change <0.5). C: Venn diagram showing changes of transcripts. Among 25,750 transcripts with average TPM >1, 357 transcripts were significantly upregulated and 185 were significantly downregulated in Slc16a1-deleted eWAT (with absolute log2fold change >1). D and E: GO analysis (D) and KEGG analysis (E) of the 542 transcripts as in C. F: A heat map showing z scores of the genes in a few selected pathways from GO analysis. G: Relative mRNA levels of a few selected genes from the list in F. The expression level was relative to β-actin, and data are shown as mean ± SEM. *P < 0.05 and **P < 0.01. H and J: Gene set enrichment analysis (GSEA) analysis of immune response pathways (H), macrophage activation pathways (I), and apoptotic process (J). AKO, Slc16a1fl/faP2-Cre+/− mice; ER, endoplasmic reticulum; NES, normalized enrichment score; PC1, principal component 1; PC2, principal component 2.

Close modal

To identify pathways altered by Slc16a1 deletion under HFD, we performed gene ontology (GO) analysis (Fig. 5D) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways analysis (Fig. 5E) with the genes corresponding to the identified 542 significantly changed transcripts. We found that Slc16a1 deletion in eWAT resulted in alterations of various biological processes, especially those involved in immune response and cell growth (Fig. 5D and E). By visualizing the individual transcriptional changes in pathways of interest with heat map, we noted that numerous genes related to innate immune response and apoptosis, such as Nfkb1, Aida, Hck, Akt1, Diablo, and Dapk3, were upregulated in Slc16a1-AKO mice (Fig. 5F). Such changes were further verified by quantitative RT-PCR (Fig. 5G). Furthermore, gene set enrichment analysis of transcripts revealed that pathway of immune response was among the top enriched pathways in Slc16a1-deleted eWAT compared with the wild-type eWAT (Fig. 5H). We also found substantial upregulation of genes related to macrophages activation (based on GSE22935 data set) in Slc16a1-deleted eWAT (Fig. 5I). In addition, an enrichment of genes involved in apoptotic process was found in Slc16a1-deleted eWAT (Fig. 5J). These results collectively indicate that Slc16a1 deletion is associated with a global increase in expression of genes involved in inflammation and apoptosis in white adipose tissue.

Lactate Accumulation in Adipocytes Promotes Apoptosis That Is Required for Inflammatory Response

As transcriptional profiling indicated that Slc16a1 deletion is associated with an increase of apoptosis in eWAT, we next analyzed whether MCT1 could regulate apoptosis in adipocytes. The level of cleaved caspase 3, a marker of cell apoptosis, was clearly elevated by Slc16a1 deletion in 3T3-L1 adipocytes (Fig. 6A). Treatment of 3T3-L1 adipocytes with MCT1 inhibitor AZD3965 also dose-dependently elevated the level of cleaved caspase 3 (Fig. 6B). Furthermore, deletion of Slc16a1 or treatment with AZD3965 increased the caspase 3 activity in 3T3-L1 adipocytes (Fig. 6C). Consistently, cleaved caspase 3 was increased in Slc16a1-deleted eWAT as analyzed by immunoblotting and immunofluorescent staining (Fig. 6D and E). These results thus demonstrate that loss of MCT1 function in adipocytes enhances cell apoptosis. It is noteworthy that even though the expression of Slc16a1 was very high in BAT, deletion of Slc16a1 appeared to have no effect on apoptosis in BAT (Supplementary Fig. 6C).

Figure 6

Inhibition of MCT1 leads to lactate accumulation and promotes apoptosis in adipocytes. A: Western blotting to detect cleaved caspase 3 in control and Slc16a1-deleted 3T3-L1 adipocytes. Quantification of cleaved caspase 3 relative to HSP90 is shown on the right. B: Western blotting to detect cleaved caspase 3 in 3T3-L1 adipocytes treated with different concentrations of AZD3965 for 4 days. C: Relative activity of caspase 3 in control, Slc16a1-deleted, and AZD3965-treated (1 μmol/L) 3T3-L1 adipocytes (n = 3 for each group). D: Western blotting to detect cleaved caspase 3 in eWAT of Slc16a1fl/fl (WT) and Slc16a1fl/flaP2-Cre+/− (AKO) mice fed HFD for 15 weeks. Quantification of cleaved caspase 3 relative to HSP is shown on the right. E: Representative images of immunofluorescent staining of cleaved caspase 3 in eWAT sections from WT and AKO mice fed HFD for 15 weeks. F: Western blotting to detect cleaved caspase 3 in 3T3-L1 adipocytes treated with different concentrations of lactate for 4 days. G: Relative caspase 3 activity in differentiated 3T3-L1 cells administered with DMSO (CTR), 1 μmol/L AZD3965 (AZD), 1 μmol/L AZD3965 plus 50 μmol/L Asperosaponin VI (AZD+ASP), and 1 μmol/L AZD3965 plus 2 μmol/L Ac-DEVD-CHO (AZD+ADC) for 4 days (n = 3 for each group). H: Relative mRNA levels of Ccl2 and Il6 in adipocytes as in G (n = 6 for each group). I: A model to depict the role of lactate in linking obesity to inflammation and insulin resistance. All the mRNA expression levels were relative to β-actin. Intensity of cleaved caspase 3 was analyzed with Student t test, while caspase 3 activity and mRNA levels in C, G, and H were analyzed with one-way ANOVA. All the quantitative data are shown as mean ± SEM. *P < 0.05; **P < 0.01; ***P < 0.001. KO, knockout.

Figure 6

Inhibition of MCT1 leads to lactate accumulation and promotes apoptosis in adipocytes. A: Western blotting to detect cleaved caspase 3 in control and Slc16a1-deleted 3T3-L1 adipocytes. Quantification of cleaved caspase 3 relative to HSP90 is shown on the right. B: Western blotting to detect cleaved caspase 3 in 3T3-L1 adipocytes treated with different concentrations of AZD3965 for 4 days. C: Relative activity of caspase 3 in control, Slc16a1-deleted, and AZD3965-treated (1 μmol/L) 3T3-L1 adipocytes (n = 3 for each group). D: Western blotting to detect cleaved caspase 3 in eWAT of Slc16a1fl/fl (WT) and Slc16a1fl/flaP2-Cre+/− (AKO) mice fed HFD for 15 weeks. Quantification of cleaved caspase 3 relative to HSP is shown on the right. E: Representative images of immunofluorescent staining of cleaved caspase 3 in eWAT sections from WT and AKO mice fed HFD for 15 weeks. F: Western blotting to detect cleaved caspase 3 in 3T3-L1 adipocytes treated with different concentrations of lactate for 4 days. G: Relative caspase 3 activity in differentiated 3T3-L1 cells administered with DMSO (CTR), 1 μmol/L AZD3965 (AZD), 1 μmol/L AZD3965 plus 50 μmol/L Asperosaponin VI (AZD+ASP), and 1 μmol/L AZD3965 plus 2 μmol/L Ac-DEVD-CHO (AZD+ADC) for 4 days (n = 3 for each group). H: Relative mRNA levels of Ccl2 and Il6 in adipocytes as in G (n = 6 for each group). I: A model to depict the role of lactate in linking obesity to inflammation and insulin resistance. All the mRNA expression levels were relative to β-actin. Intensity of cleaved caspase 3 was analyzed with Student t test, while caspase 3 activity and mRNA levels in C, G, and H were analyzed with one-way ANOVA. All the quantitative data are shown as mean ± SEM. *P < 0.05; **P < 0.01; ***P < 0.001. KO, knockout.

Close modal

Due to the tight relationship between intracellular pH and apoptosis (31), we hypothesized that the observed increase of apoptosis by loss of MCT1 function was likely due to accumulation of intracellular lactate. We treated 3T3-L1 cells with different concentrations of lactate during adipocyte differentiation. We found that lactate could dose dependently elevate the level of cleaved caspase 3 (Fig. 6F), thus indicating that intracellular lactate accumulation caused by loss of MCT1 function promotes apoptosis in adipocytes.

As cell apoptosis is the primary cause of inflammatory response (32), we next addressed whether apoptosis is required for the production of proinflammatory cytokines in adipocytes. We analyzed the combined effect of MCT1 inhibitor and apoptosis inhibitor. We found that the enhanced adipocyte apoptosis by AZD3965 was completely blocked by apoptosis inhibitors Asperosaponin VI and Ac-DEVD-CHO (Fig. 6G). Importantly, AZD3965-induced expression of Ccl2 and Il6 in adipocytes was completely overridden by Asperosaponin VI or Ac-DEVD-CHO (Fig. 6H). These data indicate that the increased lactate accumulation by Slc16a1 deletion or MCT1 inhibition can promote apoptosis, and apoptosis is required for production of CCL2 and IL-6 from adipocytes.

Finally, it is noteworthy that aP2-Cre used in our study can recombine in a number of nonadipocyte cells (33). To rule out the potential effect of Slc16a1 deletion in macrophages in our mouse model, we isolated both SVFs in eWAT and PMs and analyzed the expression of Slc16a1 in these cells. As shown in Supplementary Fig. 7B and C, the mRNA level of Slc16a1 in SVFs and PMs was not altered in the Slc16a1-deleted mice made from aP2-Cre. We also did a preliminary experiment to analyze macrophage polarization using PMs from Slc16a1-deleted mice. We found that LPS-stimulated M1 polarization in PMs was not changed (Supplementary Fig. 7C). Together, these results indicated that the observed changes in the inflammation level in the adipose tissue of the knockout mice were not caused by deletion of Slc16a1 in macrophages per se but, rather, were a consequence of deletion of Slc16a1 in adipocytes.

In these studies, we identified that lactate is a key player that links obesity to insulin resistance. Among the MCT family members involved in lactate transport, Slc16a1 has the highest expression in adipose tissues. At the cellular level, deletion of Slc16a1 or inhibition of MCT1 attenuates adipocyte differentiation. Specific deletion of Slc16a1 in adipose tissues has no effect on HFD-induced adiposity; however, it aggravates systemic insulin resistance, accompanied by a marked increase in blood levels of proinflammatory cytokines, including TNFα, IL-1β, and IL-6. Mechanistically, MCT1 mediates lactate efflux in adipocytes, such that the loss of MCT1 function results in intracellular lactate accumulation, which is aggravated by obesity. The elevated intracellular lactate level can lead to an increase in expression of cytokines, such as CCL2 and IL-1β. In a coculture experiment, lactate accumulation in adipocytes increased macrophage recruitment and expression of cytokines from macrophages. Thus, intracellular lactate accumulation in adipocytes leads to a robust inflammatory response via a direct pathway in which lactate stimulates cytokine expression in adipocytes per se and an indirect pathway in which cytokines released from adipocytes recruit macrophages to boost the local inflammation. The macrophage involvement in this process is also partly mediated by apoptosis of adipocytes caused by lactate accumulation. Collectively, lactate accumulation in adipocytes can elicit a vicious cycle of inflammatory response via the functional interaction between adipocytes and macrophages and likely other immune cells. Consequently, this leads to a robust increase in circulating levels of proinflammatory cytokines, which in turn leads to systemic insulin resistance (Fig. 6J).

Accumulating evidence has clearly pinpointed that inflammation in obese adipose tissue is the primary cause of systemic insulin resistance. However, what triggers obesity-induced inflammation is not completely understood (27,29,30). One hypothesis is that obesity leads to elevated circulating LPS level due to increased intestinal penetrance (34). However, LPS produced by the gram-positive bacteria inside intestine lumen can boost the whole-body inflammatory response but does not serve as a direct trigger of inflammation in adipose tissue. Another mechanism that underlies obesity-induced inflammation is obesity-induced hypoxia in adipose tissue (16). Hypoxia develops as adipose tissue expansion, and adipocyte enlargement is accompanied by reduced local blood flow and increased oxygen consumption. Hypoxia can initiate inflammation and insulin resistance via transcription factor HIF-1α (22,35). It is noteworthy that hypoxia is the primary drive for anaerobic glycolysis and lactate production in physiology. We postulate that obesity-induced hypoxia favors lactate production in adipocytes, and the elevated lactate triggers inflammatory response in adipose tissue. One possible mechanism is that elevation of intracellular lactate (likely via lowering pH) can initiate apoptosis of adipocytes, which in turn kicks off the inflammatory response (32). This assumption is supported by our study showing that inhibition of apoptosis alleviates proinflammatory cytokine production upon MCT1 blockage. However, we cannot rule out the possibility that lactate retention in adipocytes can initiate inflammatory response independent of apoptosis. We observed that Slc16a1 deletion or MCT1 blockage can induce expression of Ccl2 expression in adipocytes. CCL2 is a primary factor for macrophage recruitment; thus, the elevated lactate can indirectly promote inflammatory response via recruitment of macrophages, which then release proinflammatory cytokines. At present we are not sure whether stimulation of Ccl2 production by lactate in adipocytes is dependent on apoptosis.

Although a large majority of studies have indicated that inflammation in adipose tissue is linked to insulin resistance, a few recent reports have challenged this point of view. It was found that upon HFD exposure, the proinflammatory signaling in adipocytes is required for adipose tissue remodeling and expansion (36). Suppression of proinflammatory potential in adipose tissues actually promoted insulin resistance in mice (36,37). Overexpression of Cidea in adipose tissue caused severe obesity together with an enhanced insulin sensitivity (38). The blood DPP4 activity is associated with obesity but not glucose homeostasis (39). On the other hand, it was found that lipid delivery to ATMs, the critical player in mediating obesity-induced inflammation in adipose tissue, is mediated not by autophagy but, rather, by exosome-sized, lipid-filled vesicles (AdExos) (40,41). In addition, it was found that insulin resistance in mTORC2-deleted mice caused inflammation in adipose tissue via production of MCP1 that activates local proinflammatory macrophages (42). Therefore, obesity is not necessarily associated with insulin resistance. In this study, we discovered that MCT1 deficiency aggravated obesity-induced accumulation of lactate in adipocytes, whereby causing activation of local macrophages, elevation of blood levels of proinflammatory cytokines, and systemic insulin resistance in peripheral tissues. Combining with other people’s findings, we propose that inflammation in adipose tissue likely plays two distinct roles in obesity and insulin resistance dependent on the pathophysiological stage. During early phases of obesity, an increase of proinflammatory signaling in adipose tissue is likely an adaptive response to prevent further development of insulin resistance via its effect on adipose tissue expansion and remodeling. At the same time, the local lactate concentration is not reaching a critical level to provoke apoptosis and overt inflammation due to an adaptive increase in the expression of MCT1. After obesity is worsened in late stages when hypoxia-induced production of lactate is elevated beyond a level MCT1 can handle, a robust local inflammation occurs, leading to release of proinflammatory cytokines and systemic insulin resistance. Under this scenario, keeping local lactate production in check would be an intriguing strategy to disassociate obesity from insulin resistance.

Our study corroborated previous investigations showing that adipose tissue is a significant lactate producer in the body. Approximately 50–70% of glucose is converted to lactate in adipose tissue (12). Numerous studies demonstrated that lactate production from adipose tissue is similar to that of skeletal muscle in humans (1315). It is noteworthy that lactate production is elevated in obese adipocytes due to the enlargement of adipocyte size (12). Adipose tissue is now considered as a prioritized organ for lactate production (43). In our study, the blood lactate concentration dropped from 2.13 to 1.65 mmol/L (a 23.4% reduction) with normal chow and dropped from 2.50 to 1.77 mmol/L with HFD conditions (a 29.3% reduction) (Fig. 2C). However, the efficiency of Slc16a1 knockout was 81.0% in iWAT and 68.9% in eWAT (Fig. 3B) (average 74.9% for both WATs; BAT is not included as its total mass is quite small). Combining the percentages of lactate change in blood and knockout efficiency, we estimate that ∼31.2% and 39.1% of circulating lactate is from adipose tissue (mediated by MCT1) with normal chow and HFD, respectively. In addition, we found that the expression of Slc16a1 was increased in white adipose tissues by HFD (Supplementary Fig. 7A), likely indicating an adaptive response of white adipose tissues to accommodate the obesity-induced increase of local lactate level. Therefore, our results, combined with other groups’ studies, support the conclusion that adipose tissue is a significant lactate provider in the body. It has been postulated that lactate can shuttle between different cells and between different intracellular compartments and such shuffling is actively involved in the distribution of energy substrate in the body (1,2). Lactate producers, such as muscle, skin, and adipose tissue, provide energy for working muscle, the heart, and the brain, as well as provide a gluconeogenic precursor for the liver (1,2). For example, lactate can exchange between white-glycolytic and red-oxidative fibers in skeletal muscle and heart (4,5) and between astrocytes and neurons (6,7). As adipose tissue is an active lactate producer, we propose that the lactate produced from the adipose tissue can also be involved in lactate shuffling. Globally, lactate from adipose tissue can serve as a substrate to be used by other tissues such as liver (as a gluconeogenic substrate) and skeletal muscle (as an oxidative fuel). Locally, lactate derived from adipocytes can provide an oxidative fuel to aid in browning of beige cells and likely brown adipocytes. Interestingly, MCT1 is localized in a subpopulation of beige adipose tissue and is required for cold-stimulated UCP1 expression and thermogenesis (19). Thus, the lactate produced from adipocytes around the MCT1-positive cold-inducible beige adipocytes can provide an oxidative substrate to boost browning of the beige cells.

In summary, the results of our study indicate that lactate within adipocytes functions as a trigger to initiate inflammatory response in adipose tissues, thus linking obesity to systemic insulin resistance. Therefore, exploiting this function of lactate in adipocytes can be a promising strategy to break the deadly link between obesity and insulin resistance, as well as serve as a potential treatment of obesity. Interestingly, it was found that fasting every other day can elevate browning of adipose tissues and reduce obesity in mice, and such effects are proposed to be mediated by lactate produced from gut microbiota (44). Bearing in mind that lactate can induce browning in beige adipocytes dependent on MCT1 (17,19), a strategy to boost lactate-induced browning has the potential to control obesity. In addition, as elevated lactate within adipocytes can trigger inflammatory responses in adipose tissue and insulin resistance in peripheral tissues, harnessing lactate production/accumulation in adipocytes can be another useful strategy to control obesity-induced insulin resistance.

This article contains supplementary material online at https://doi.org/10.2337/figshare.17837834.

Acknowledgments. The authors thank Qianqian Chu and Lin Qiu from Institutional Center for Shared Technologies and Facilities of SINH, CAS, for technical assistance.

Funding. This study was funded by National Natural Science Foundation of China (31630036 to Y.C.) and the Ministry of Science and Technology of China (2016YFA0500103 to Y.C.).

Duality of Interest. No potential conflicts of interest relevant to this article were reported.

Author Contributions. Y.L. and Y.C. conceptualized and designed the study. Y.L. performed the experiment. M.B, S.W., L.C., Z.L., C.L., and P.C. provided technical assistance. Y.L. and Y.C. wrote the manuscript and prepared the figures. All authors read and approved the manuscript. Y.C. is the guarantor of this work and, as such, had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

1.
Brooks
GA
.
The science and translation of lactate shuttle theory
.
Cell Metab
2018
;
27
:
757
785
2.
Brooks
GA
,
Arevalo
JA
,
Osmond
AD
,
Leija
RG
,
Curl
CC
,
Tovar
AP
.
Lactate in contemporary biology: a phoenix risen
.
J Physiol
.
10 February 2021 [Epub ahead of print]. DOI: 10.1113/JP280955
3.
Hui
S
,
Ghergurovich
JM
,
Morscher
RJ
, et al
.
Glucose feeds the TCA cycle via circulating lactate
.
Nature
2017
;
551
:
115
118
4.
Gertz
EW
,
Wisneski
JA
,
Stanley
WC
,
Neese
RA
.
Myocardial substrate utilization during exercise in humans. Dual carbon-labeled carbohydrate isotope experiments
.
J Clin Invest
1988
;
82
:
2017
2025
5.
Bergman
BC
,
Tsvetkova
T
,
Lowes
B
,
Wolfel
EE
.
Myocardial glucose and lactate metabolism during rest and atrial pacing in humans
.
J Physiol
2009
;
587
:
2087
2099
6.
Pellerin
L
,
Pellegri
G
,
Bittar
PG
, et al
.
Evidence supporting the existence of an activity-dependent astrocyte-neuron lactate shuttle
.
Dev Neurosci
1998
;
20
:
291
299
7.
Liu
L
,
MacKenzie
KR
,
Putluri
N
,
Maletić-Savatić
M
,
Bellen
HJ
.
The glia-neuron lactate shuttle and elevated ROS promote lipid synthesis in neurons and lipid droplet accumulation in glia via APOE/D
.
Cell Metab
2017
;
26
:
719
737.e6
8.
Liu
C
,
Wu
J
,
Zhu
J
, et al
.
Lactate inhibits lipolysis in fat cells through activation of an orphan G-protein-coupled receptor, GPR81
.
J Biol Chem
2009
;
284
:
2811
2822
9.
Ahmed
K
,
Tunaru
S
,
Tang
C
, et al
.
An autocrine lactate loop mediates insulin-dependent inhibition of lipolysis through GPR81
.
Cell Metab
2010
;
11
:
311
319
10.
Brown
TP
,
Bhattacharjee
P
,
Ramachandran
S
, et al
.
The lactate receptor GPR81 promotes breast cancer growth via a paracrine mechanism involving antigen-presenting cells in the tumor microenvironment
.
Oncogene
2020
;
39
:
3292
3304
11.
Carrière
A
,
Lagarde
D
,
Jeanson
Y
, et al
.
The emerging roles of lactate as a redox substrate and signaling molecule in adipose tissues
.
J Physiol Biochem
2020
;
76
:
241
250
12.
DiGirolamo
M
,
Newby
FD
,
Lovejoy
J
.
Lactate production in adipose tissue: a regulated function with extra-adipose implications
.
FASEB J
1992
;
6
:
2405
2412
13.
Jansson
PA
,
Smith
U
,
Lönnroth
P
.
Evidence for lactate production by human adipose tissue in vivo
.
Diabetologia
1990
;
33
:
253
256
14.
Hagström
E
,
Arner
P
,
Ungerstedt
U
,
Bolinder
J
.
Subcutaneous adipose tissue: a source of lactate production after glucose ingestion in humans
.
Am J Physiol
1990
;
258
:
E888
E893
15.
Hagström-Toft
E
,
Enoksson
S
,
Moberg
E
,
Bolinder
J
,
Arner
P
.
Absolute concentrations of glycerol and lactate in human skeletal muscle, adipose tissue, and blood
.
Am J Physiol
1997
;
273
:
E584
E592
16.
Trayhurn
P
.
Hypoxia and adipocyte physiology: implications for adipose tissue dysfunction in obesity
.
Annu Rev Nutr
2014
;
34
:
207
236
17.
Carrière
A
,
Jeanson
Y
,
Berger-Müller
S
, et al
.
Browning of white adipose cells by intermediate metabolites: an adaptive mechanism to alleviate redox pressure
.
Diabetes
2014
;
63
:
3253
3265
18.
Jeanson
Y
,
Ribas
F
,
Galinier
A
, et al
.
Lactate induces FGF21 expression in adipocytes through a p38-MAPK pathway
.
Biochem J
2016
;
473
:
685
692
19.
Lagarde
D
,
Jeanson
Y
,
Barreau
C
, et al
.
Lactate fluxes mediated by the monocarboxylate transporter-1 are key determinants of the metabolic activity of beige adipocytes
.
J Biol Chem
2021
;
296
:
100137
20.
Halestrap
AP
,
Meredith
D
.
The SLC16 gene family-from monocarboxylate transporters (MCTs) to aromatic amino acid transporters and beyond
.
Pflugers Arch
2004
;
447
:
619
628
21.
Lin
Y
,
Huang
M
,
Wang
S
,
You
X
,
Zhang
L
,
Chen
Y
.
PAQR11 modulates monocyte-to-macrophage differentiation and pathogenesis of rheumatoid arthritis
.
Immunology
2021
;
163
:
60
73
22.
Seo
JB
,
Riopel
M
,
Cabrales
P
, et al
.
Knockdown of Ant2 reduces adipocyte hypoxia and improves insulin resistance in obesity
.
Nat Metab
2019
;
1
:
86
97
23.
Halestrap
AP
,
Wilson
MC
.
The monocarboxylate transporter family--role and regulation
.
IUBMB Life
2012
;
64
:
109
119
24.
San Martín
A
,
Ceballo
S
,
Ruminot
I
,
Lerchundi
R
,
Frommer
WB
,
Barros
LF
.
A genetically encoded FRET lactate sensor and its use to detect the Warburg effect in single cancer cells
.
PLoS One
2013
;
8
:
e57712
25.
Mina
AI
,
LeClair
RA
,
LeClair
KB
,
Cohen
DE
,
Lantier
L
,
Banks
AS
.
CalR: a web-based analysis tool for indirect calorimetry experiments
.
Cell Metab
2018
;
28
:
656
666.e1
26.
Kusminski
CM
,
Bickel
PE
,
Scherer
PE
.
Targeting adipose tissue in the treatment of obesity-associated diabetes
.
Nat Rev Drug Discov
2016
;
15
:
639
660
27.
Hirosumi
J
,
Tuncman
G
,
Chang
L
, et al
.
A central role for JNK in obesity and insulin resistance
.
Nature
2002
;
420
:
333
336
28.
Olefsky
JM
,
Glass
CK
.
Macrophages, inflammation, and insulin resistance
.
Annu Rev Physiol
2010
;
72
:
219
246
29.
Saltiel
AR
,
Olefsky
JM
.
Inflammatory mechanisms linking obesity and metabolic disease
.
J Clin Invest
2017
;
127
:
1
4
30.
Reilly
SM
,
Saltiel
AR
.
Adapting to obesity with adipose tissue inflammation
.
Nat Rev Endocrinol
2017
;
13
:
633
643
31.
Lagadic-Gossmann
D
,
Huc
L
,
Lecureur
V
.
Alterations of intracellular pH homeostasis in apoptosis: origins and roles
.
Cell Death Differ
2004
;
11
:
953
961
32.
Van Opdenbosch
N
,
Lamkanfi
M
.
Caspases in cell death, inflammation, and disease
.
Immunity
2019
;
50
:
1352
1364
33.
Lee
KY
,
Russell
SJ
,
Ussar
S
, et al
.
Lessons on conditional gene targeting in mouse adipose tissue
.
Diabetes
2013
;
62
:
864
874
34.
Amar
J
,
Chabo
C
,
Waget
A
, et al
.
Intestinal mucosal adherence and translocation of commensal bacteria at the early onset of type 2 diabetes: molecular mechanisms and probiotic treatment
.
EMBO Mol Med
2011
;
3
:
559
572
35.
Lee
YS
,
Kim
JW
,
Osborne
O
, et al
.
Increased adipocyte O2 consumption triggers HIF-1α, causing inflammation and insulin resistance in obesity
.
Cell
2014
;
157
:
1339
1352
36.
Wernstedt Asterholm
I
,
Tao
C
,
Morley
TS
, et al
.
Adipocyte inflammation is essential for healthy adipose tissue expansion and remodeling
.
Cell Metab
2014
;
20
:
103
118
37.
Zhu
Q
,
An
YA
,
Kim
M
, et al
.
Suppressing adipocyte inflammation promotes insulin resistance in mice
.
Mol Metab
2020
;
39
:
101010
38.
Abreu-Vieira
G
,
Fischer
AW
,
Mattsson
C
, et al
.
Cidea improves the metabolic profile through expansion of adipose tissue
.
Nat Commun
2015
;
6
:
7433
39.
Varin
EM
,
Mulvihill
EE
,
Beaudry
JL
, et al
.
Circulating levels of soluble dipeptidyl peptidase-4 are dissociated from inflammation and induced by enzymatic DPP4 inhibition
.
Cell Metab
2019
;
29
:
320
334.e5
40.
Grijalva
A
,
Xu
X
,
Ferrante
AW
Jr
.
Autophagy is dispensable for macrophage-mediated lipid homeostasis in adipose tissue
.
Diabetes
2016
;
65
:
967
980
41.
Flaherty
SE
3rd
,
Grijalva
A
,
Xu
X
,
Ables
E
,
Nomani
A
,
Ferrante
AW
Jr
.
A lipase-independent pathway of lipid release and immune modulation by adipocytes
.
Science
2019
;
363
:
989
993
42.
Shimobayashi
M
,
Albert
V
,
Woelnerhanssen
B
, et al
.
Insulin resistance causes inflammation in adipose tissue
.
J Clin Invest
2018
;
128
:
1538
1550
43.
Krycer
JR
,
Quek
LE
,
Francis
D
, et al
.
Lactate production is a prioritized feature of adipocyte metabolism
.
J Biol Chem
2020
;
295
:
83
98
44.
Li
G
,
Xie
C
,
Lu
S
, et al
.
Intermittent fasting promotes white adipose browning and decreases obesity by shaping the gut microbiota
.
Cell Metab
2017
;
26
:
672
685.e4
Readers may use this article as long as the work is properly cited, the use is educational and not for profit, and the work is not altered. More information is available at https://www.diabetesjournals.org/journals/pages/license.