Ethyl Gallate Dual-Targeting PTPN6 and PPARγ Shows Anti-Diabetic and Anti-Obese Effects

Abstract

The emergence of the high correlation between type 2 diabetes and obesity with complicated conditions has led to the coinage of the term "diabesity". AMP-activated protein kinase (AMPK) activators and peroxisome proliferator-activated receptor (PPARγ) antagonists have shown therapeutic activity for diabesity, respectively. Hence, the discovery of compounds that activate AMPK as well as antagonize PPARγ may lead to the discovery of novel therapeutic agents for diabesity. In this study, the knockdown of PTPN6 activated AMPK and suppressed adipogenesis in 3T3-L1 cells. By screening a library of 1033 natural products against PTPN6, we found ethyl gallate to be the most selective inhibitor of PTPN6 (K_i = 3.4 μM). Subsequent assay identified ethyl gallate as the best PPARγ antagonist (IC₅₀ = 5.4 μM) among the hit compounds inhibiting PTPN6. Ethyl gallate upregulated glucose uptake and downregulated adipogenesis in 3T3-L1 cells as anticipated. These results strongly suggest that ethyl gallate, which targets both PTPN6 and PPARγ, is a potent therapeutic candidate to combat diabesity.

Keywords: 3T3-L1 adipocyte; AMPK; PPARγ antagonist; PTPN6; adipogenesis; ethyl gallate; natural product.

Figure 1

The depletion of PTPN6 by siRNA knockdown increases AMPK phosphorylation. 3T3-L1 preadipocytes were transfected with PTPN6 siRNA for 48 h. After cell lysis, the expression of PTPN6 was analyzed by quantitative real-time-polymerase chain reaction (qRT-PCR) (A) and Western blotting (B,C). Changes in the phosphorylation of AMPK following the knockdown of PTPN6 (also named SHP1) were analyzed by Western blotting (B). Quantification was performed and normalized to control using CSAnalyzer 4 (C,D). Data are shown as mean ± standard deviation (n = 3). ** p < 0.01, *** p < 0.001. Statistical significance was analyzed by a two-tailed unpaired t-test.

Figure 2

siRNA knockdown of PTPN6 suppresses adipogenesis of 3T3-L1 preadipocytes. After the knockdown of PTPN6 using siRNA for 48 h, 3T3-L1 preadipocytes were differentiated by DMI-induction for 6 days (DMI: dexamethasone, methylisobutylxanthine, and insulin). Lipid droplets of 3T3-L1 adipocytes were stained using the Oil red O working solution and visualized. Scale bar, 100 μm. (A). Quantitative data on the lipid content were analyzed with extracted Oil red O using 100% isopropanol (B). For Western blot analysis, the cells were lysed after 48 h of differentiation. Western blot was performed using anti-PPARγ, C/EBPα, and β-actin antibodies (C). Quantification was performed for PPARγ (D), C/EBPα P42 (E), and C/EBPα P30 (F), and each was normalized to control using CSAnalyzer 4. Data are shown as mean ± standard deviation (n = 3). ** p < 0.01. Statistical significance was analyzed by a two-tailed unpaired t-test.

Figure 3

Selection and validation of ethyl gallate (EG) as a PTPN6 inhibitor. For the selection of PTPN6 inhibitors, a 1033 natural products library (20 μM for each compound) was screened against 20 PTPs. A substrate, 6,8-difluoro-4-methylumbelliferyl phosphate (DiFMUP), which is hydrolyzed to DiFMU after reactions with protein tyrosine phosphatases (PTPs), was used at the concentration of 2 × K_m for measuring enzymatic activities. The increase in fluorescence intensity at Ex/Em = 355/460 nm was monitored by a microplate reader using a 96-well black plate. Results of screening with over 60% inhibition for PTPN6 are reported as heatmap analysis using GraphPad Prism. The compound names corresponding to the compound numbers are listed in Table S1. (A). To determine the inhibition constant (K_i) and type of inhibition of EG (271), various concentrations of EG (2.5, 5, 7.5, and 10 μM) were added to the pH 7.0 reaction buffer containing different concentrations (0.5 × K_m, 1 × K_m, 2 × K_m, and 4 × K_m; K_m = 86.26 μM) of DiFMUP followed by treatment of 2.0 nM PTPN6. The K_i and type of inhibition were derived from a Dixon plot (B) and a Lineweaver-Burk plot (C) for EG inhibition.

Figure 4

Identification of PPARγ antagonistic effect of ethyl gallate (EG). After a 6 h co-transfection with pCMV6-hPPARG-GFP and pPPRE-TK-luc, CHO cells were treated with compounds in the presence of 2 μM rosiglitazone and incubated for 24 h. (A) PPARγ antagonistic activities against hit compounds, exhibiting PTPN6 inhibition. Cells were treated with each compound at 20 μM concentration. The compound names corresponding to the compound numbers are listed in Table S1. Data are shown as mean ± standard deviation (n = 2). (B) Concentration-dependent PPARγ antagonistic activities of EG. Cells were treated with EG or GWW9662 at various concentrations (0.0001, 0.001, 0.01, 0.1, 1, 5, 10, 25, 50, and 100 μM). The IC₅₀ values represent the concentrations of compounds that inhibited 50% of the response induced by 2 μM rosiglitazone. Data are shown as mean ± standard deviation (n = 3).

Figure 5

Docking model of ethyl gallate (EG) on PTPN6 and PPARγ. (A) Structure of gallate derivatives and screening results of gallate derivatives on PTPN6. (B) Docking model of EG (orange) and propyl gallate (pale green) on PTPN6 (white). Ligands are presented as ball-and-stick models. Important motifs and residues of PTPN6 are highlighted in color. Hydrogen bonds are represented as yellow dashed lines. (C) Surface representation of PTPN6 in gray. (D) Docking model of EG (orange) and the X-ray conformation of rosiglitazone (yellow) on PPARγ (sky blue). The orthosteric pocket is represented as a gray surface. (E) Multiple H-bonds between EG and residues in branch I. π–π stacking interaction is represented as blue dashed lines. For visual inspection, the representation of helix3 is simplified as a green CA trace line.

Figure 6

Ethyl gallate (EG) treatment upregulates glucose uptake and activates AMPK in 3T3-L1 adipocytes. Preadipocytes of 3T3-L1 cells were differentiated by DMI-induction (DMI: dexamethasone, methylisobutylxanthine, and insulin) for 6 days. Fully differentiated 3T3-L1 cells were incubated overnight with low-glucose Dulbecco’s modified Eagle’s medium (DMEM) for starvation, followed by treatment of EG (6 h) or insulin (30 min) in glucose-depleted DMEM with 0.1% DMSO. Subsequently, they were treated using 100 μM 2-NBDG for 1 h. Fluorescence intensity was detected at Ex/Em = 485/535 nm by confocal microscopy (A) or microplate reader (B) after washing with cold DPBS. Scale bar, 100 μm. For Western blot analysis, cells were lysed and analyzed after treatment of EG under the same condition as the 2-NBDG uptake assay. Western blotting was performed using anti-P-AMPK, T-AMPK, and β-actin antibodies (C). Quantification was performed for P-AMPK, T-AMPK, and normalized to control using CSAnalyzer 4 (D). Data are shown as mean ± standard deviation (n = 3 or 4). ** p < 0.01. *** p < 0.001. Statistical significance was analyzed by one-way ANOVA for multiple comparisons followed by Tukey–Kramer statistical test.

Figure 7

Anti-adipogenic effect of ethyl gallate (EG) at the early stage of differentiation. Preadipocytes of 3T3-L1 cells were differentiated by DMI-induction (DMI: dexamethasone, methylisobutylxanthine, and insulin) and treated with various concentrations of EG (12.5 μM, 25 μM, and 50 μM) simultaneously. To analyze which stages are affected by the treatment with EG, 3T3-L1 cells were treated with 50 μM EG for 0–6 days (1), 0–2 days (2), 2–4 days (3), or 4–6 days (4) during DMI-induced differentiation (C). On day 6 of differentiation, lipid droplets of 3T3-L1 adipocytes were stained using the Oil red O working solution and visualized. Scale bar, 100 μm (A,D). Quantitative data on the lipid content were analyzed using extracted Oil red O with 100% isopropanol (B,E). For analysis of early adipogenic factors, cells were lysed and analyzed by Western blot using anti-PPARγ, C/EBPα, and β-actin antibodies after treating with EG for 48 h during DMI-induced differentiation (F). Quantification was performed, and each was normalized to control using CSAnalyzer 4 (G). Data are shown as mean ± standard deviation (n = 3). * p < 0.05, ** p < 0.01, *** p < 0.001. Statistical significance was analyzed by one-way ANOVA for multiple comparisons followed by Tukey–Kramer statistical test.