The low molecular weight protein tyrosine phosphatase promotes adipogenesis and subcutaneous adipocyte hypertrophy

Abstract

Obesity is a major contributing factor to the pathogenesis of Type 2 diabetes. Multiple human genetics studies suggest that high activity of the low molecular weight protein tyrosine phosphatase (LMPTP) promotes metabolic syndrome in obesity. We reported that LMPTP is a critical promoter of insulin resistance in obesity by regulating liver insulin receptor signaling and that inhibition of LMPTP reverses obesity-associated diabetes in mice. Since LMPTP is expressed in adipose tissue but little is known about its function, here we examined the role of LMPTP in adipocyte biology. Using conditional knockout mice, we found that selective deletion of LMPTP in adipocytes impaired obesity-induced subcutaneous adipocyte hypertrophy. We assessed the role of LMPTP in adipogenesis in vitro, and found that LMPTP deletion or knockdown substantially impaired differentiation of primary preadipocytes and 3T3-L1 cells into adipocytes, respectively. Inhibition of LMPTP in 3T3-L1 preadipocytes also reduced adipogenesis and expression of proadipogenic transcription factors peroxisome proliferator activated receptor gamma (PPARγ) and CCAAT/enhancer-binding protein alpha. Inhibition of LMPTP increased basal phosphorylation of platelet-derived growth factor receptor alpha (PDGFRα) on activation motif residue Y849 in 3T3-L1, resulting in increased activation of the mitogen-associated protein kinases p38 and c-Jun N-terminal kinase and increased PPARγ phosphorylation on inhibitory residue S82. Analysis of the metabolome of differentiating 3T3-L1 cells suggested that LMPTP inhibition decreased cell glucose utilization while enhancing mitochondrial respiration and nucleotide synthesis. In summary, we report a novel role for LMPTP as a key driver of adipocyte differentiation via control of PDGFRα signaling.

Keywords: LMPTP; adipocyte; adipogenesis; phosphorylation; protein tyrosine phosphatase.

Fig. 1.. Adipocyte-specific LMPTP deletion reduces SubQ adipocyte size in obese mice.

To generate diet-induced obese (DIO) mice, male Acp1^fl/fl adiponectin-Cre+ (n=5) and Cre- (n=6) littermates were placed on high-fat diet for 12 months. (a) Mean±SEM mouse body weight. (b) Mean±SEM SubQ fat pad weight. (c-e) Subcutaneous (SubQ) adipose tissue (AT) was harvested and stained with H&E and adipocyte surface area calculated in ImageJ. (c) Representative H&E-stained images. (d) Mean±SEM adipocyte surface area. (e) Frequency distribution of SubQ adipocyte size. (a, b, d) *, p <0.05; NS, non-significant: unpaired t-test.

Fig. 2.. LMPTP promotes adipogenesis through its catalytic activity.

(a) Primary preadipocytes were isolated from the SubQ fat pad of wild type (WT; n=3) and LMPTP knockout (KO; n=4) mice and were subjected to an adipogenesis assay. Cells from 1 WT mouse were used as an undifferentiated (Undiff.) control. Mean±SEM is shown. (b) 3T3-L1 preadipocytes were subjected to an adipogenesis assay in the presence of 10 μM mouse LMPTP-targeting antisense oligonucleotide (ASO) or control non-targeting (Ctrl) ASO. Mean±SEM from 3 independent experiments is shown. (c) Primary human visceral preadipocytes were subjected to an adipogenesis assay in the presence of 10 μM human LMPTP-targeting ASO or Ctrl ASO. Mean±range from 2 independent experiments is shown. (d-e) 3T3-L1 were subjected to an adipogenesis assay in the presence of 10 μM LMPTP inhibitor Compd. 3 (d) or Compd. 23 (e) or dimethyl sulfoxide (DMSO). Mean±SEM from (d) 4 and (e) 14 independent experiments is shown. (a-e) Cells were stained with AdipoRed reagent and fluorescence was measured. Fluorescence relative to the control sample in each experiment is shown. Undiff, undifferentiated; *, p <0.05: (a-b) unpaired t-test or (c-e) unpaired t-test with Welch’s correction.

Fig. 3.. LMPTP promotes expression of pro-adipogenic transcription factors.

3T3-L1 were subjected to an adipogenesis assay in the presence of 10 μM Compd. 23 or DMSO. mRNA was measured at the indicated times by qPCR. Mean±SEM relative expression from 3 independent experiments following normalization to the housekeeping gene POLR2A is shown. *, p <0.05; NS, non-significant: two-way ANOVA treatment effect.

Fig. 4.. LMPTP inhibits activation of PDGFRα, p38 and JNK, and promotes activation of PPARγ in preadipocytes.

3T3-L1 were serum-starved overnight and incubated with DMSO or 10 μM Compd. 23 and phospho-analytes were assessed by Western blotting. Upper panels show representative blots. Lower panels show quantification of phospho-blots. (a) PDGFRα was immunoprecipitated (IP) and Y849 phosphorylation (pPDGFRα-Y849) was assessed in 5 independent experiments. (b) MAPK activation motif phosphorylation in cell lysates was assessed in 4 independent experiments. (c) PPARγ-S82 phosphorylation in cell lysates was assessed in 5 independent experiments. (a-c) Mean±SEM relative phospho-blot signal to (a) PDGFRα, (b) GAPDH, or (c) PPARγ is shown. *, p <0.05; NS, non-significant: unpaired t-test with Welch’s correction.

Fig. 5.. LMPTP inhibition alters the metabolome of adipocytes during differentiation.

(a) PCA (PC1 vs. PC2) obtained from the metabolomic and the fatty acid analyses of 3T3-L1 treated with Compd. 23 or DMSO at days 2, 6, and 10 of the differentiation protocol. (b) Volcano plot comparison of features from metabolomic and fatty acid analyses at day 10. Fold change cutoff, 1.5; q-value (FDR corrected p-value) cutoff, <0.05. (a-b) Data from 5 experimental replicates is shown.

Fig. 6.. LMPTP inhibition reduces glycolysis in adipocytes during differentiation.

(a-b) The cellular metabolome was assessed by HPLC-MS in differentiating 3T3-L1 treated with 10 μM Compd. 23 or DMSO. (a) Intensity of glucose, glycolytic intermediates, and lactate. (b) Intensity of 6-phosphogluconate, NADP+, and NADPH. (c) Hexokinase 2 (HK2) expression was assessed by qPCR in differentiating 3T3-L1 treated with 10 μM Compd. 23 or DMSO. (a, b) Data from 5 experimental replicates is shown. (c) Relative expression from 3 independent experiments following normalization to the housekeeping gene POLR2A is shown. (a-c) Mean±SEM is shown. *, p <0.05: two-way ANOVA interaction effect; ^#, p<0.05: two-way ANOVA treatment effect.

Fig. 7.. LMPTP inhibition enhances mitochondrial electron transport chain activity in adipocytes during differentiation.

The cellular metabolome was assessed by HPLC-MS in differentiating 3T3-L1 treated with 10 μM Compd. 23 or DMSO. (a) Intensity of mitochondrial electron transport chain (mETC) electron carriers. (b) Intensity of tricarboxylic acid cycle (TCA) substrates. (c) Intensity of nucleotides. (a-c) Mean±SEM is shown. Data from 5 experimental replicates is shown. *, p <0.05: two-way ANOVA interaction effect; ^#, p <0.05: two-way ANOVA treatment effect.

Fig. 8.. Proposed role for LMPTP in adipogenesis.

Our model suggests LMPTP dephosphorylates PDGFRα-Y849 in preadipocytes, reducing basal PDGFRα activity and leading to decreased p38/JNK activation. This enhances PPARγ activation by decreasing phosphorylation on inhibitory residue S82. In the dephosphorylated state, PPARγ is free to promote expression of genes associated with the differentiation of 3T3-L1 into mature adipocytes. Additional mechanisms downstream of PDGFRα and/or p38/JNK are also possible.