NLRP3 Inhibitor Tranilast Attenuates Gestational Diabetes Mellitus in a Genetic Mouse Model

C57BL/KsJ+/+ (wild type, WT) and C57BL/KsJdb/+ (db/+) mice were purchased from Shanghai Model Organisms (Shanghai, China). Female db/+ mice aged 8–12 weeks were mated with age-matched male WT mice, and mating was confirmed by the presence of a copulatory plug on the next day, which was designated as gestational day (GD) 0. Pregnant db/+ mice (GDM mice) were randomly divided into three groups, and orally treated with phosphate-buffered saline (control), 20 mg/kg of metformin (metformin), or 20 mg/kg of tranilast (tranilast), respectively, from GD 0 daily for 2 weeks. The doses of metformin or tranilast were based on previous publications [16, 17]. Pregnant WT female mice, which were mated with WT male mice, were used as controls (WT). On GD 1, GD 8, and GD 15, pregnant mice from all groups were weighed. Non-fasting blood glucose levels were measured from the tail vein by a glucometer (Roche Diagnostics, Risch-Rotkreuz, Switzerland). Insulin levels in the serum were analyzed using the Ultra Sensitive Mouse Insulin ELISA kit according to the manufacturer’s instructions (Alpco Diagnostics, Salem, NH, USA) [18]. After delivery, the litter size of offspring born from each pregnant mouse was recorded, and the body weight at birth of their offspring was weighed [19, 20]. The study was approved by the Ethics Committee of Tianjin First Central Hospital.

A glucose tolerance test and an insulin tolerance test were performed as previously described [21]. On GD 16, mice were intraperitoneally injected with 2 g/kg of glucose after 6 hours of fasting, then blood glucose levels were measured at indicated timepoints (0, 30, 60, 90, and 120 minutes after the glucose injection) using a glucometer. For insulin tolerance tests, mice were intraperitoneally injected with 1 mU/kg of insulin after a 1-hour fast, then blood glucose levels were measured and recorded at indicated timepoints [22].

Visceral adipose tissues and placentas were collected from pregnant mice and digested in RIPA buffer supplemented with a protease inhibitor cocktail (Sigma-Aldrich, St Louis, MO, USA) to extract protein. Western blot was performed as previously described [23]. Briefly, the same amounts of protein were loaded into SDS-PAGE gel, and transferred onto a PVDF membrane. After blocking in 5% milk, the membrane was incubated with a primary antibody followed by the incubation of a secondary antibody. The antibodies used were anti-NLRP3 (Santa Cruz Biotechnology, Inc., Santa Cruz, CA, USA, catalog: sc-134306), TNF-α (Abcam, Cambridge, MA, USA, catalog: ab6671), IL-6 (Abcam, catalog: ab208113), and β-actin (Sigma-Aldrich, clone number: AC-15). The protein bands were quantified using ImageJ software (National Institutes of Health, Bethesda, MD, USA) [24].

Data were analyzed by one-way analyses of variance with a Tukey post-hoc test used for multiple comparisons, and a Student’s t-test was used for two-group comparisons. Data are represented as means ± standard deviations. P < 0.05 was considered as a significant difference.

We first explored the activation of NLRP3 inflammasomes in the VAT and placenta of GDM mice, where there were accumulated macrophages during pregnancy. In the VAT and placenta of WT pregnant mice, the NLRP3 protein level was slightly increased on GD 15 with no difference in the early stage of pregnancy (GD 1 and GD 8). Gestational diabetes mellitus mice had greatly elevated expressions of NLRP3 in the VAT (Fig. 1A–C) and placenta (Fig. 1D–F) on GD 8, and its expression was further elevated on GD 15. These data indicated that NLRP3 inflammasomes were gradually activated in the VAT and placenta of GDM mice.

Next, GDM mice were orally administered 20 mg/kg of metformin, 20 mg/kg of NLRP3 inhibitor tranilast, or phosphate-buffered saline daily from GD 1 for 2 weeks as shown in Fig. 2A. Compared with WT mice, GDM mice in the control group gradually displayed GDM symptoms from GD 1 to GD 15, including elevated maternal body weight and blood glucose as well as decreased insulin levels (Fig. 2B–D). Metformin has proven to be a safe and effective drug to treat GDM [25]; therefore, we compared the efficacies of metformin and tranilast on GDM. Compared with GDM mice, metformin remarkably decreased the maternal body weight and blood glucose, and increased insulin levels, but all of these GDM symptoms could be further attenuated by treatment with tranilast. Furthermore, metformin ameliorated glucose and insulin intolerance in GDM mice, and tranilast further increased glucose and insulin sensitivity (Fig. 3A, B). Altogether, these data indicated that tranilast significantly attenuated the symptoms of GDM, and its therapeutic effect on GDM symptoms was even better than that of metformin.

To explore the role of the NLRP3 inhibitor on GDM, we first analyzed NLRP3 expressions in the VAT and placenta of mice with GDM. As expected, GDM mice displayed a higher expression of NLRP3 than WT mice, which could be significantly restored by tranilast both in the VAT (Fig. 4A, C) and placenta (Fig. 4C, D). However, metformin had no effect on the increases in NLRP3 expression.

Next, we evaluated the protective role of tranilast on inflammation in GDM mice. Inflammation activated in GDM mice, as reflected by the increased proteins and genes of proinflammatory cytokines, including TNF-α and IL-6, was reduced by metformin in the VAT (Fig. 5A–B and E) and placenta (Fig. 5C–D and F). Moreover, secreted TNF-α and IL-6 in the serum, which were increased in GDM mice, were significantly decreased by tranilast or metformin (Fig. 5G–H). Notably, compared with metformin, tranilast further decreased the expressions of TNF-α and IL-6. Thus, tranilast significantly suppressed the inflammatory responses in GDM mice.

Next, we measured the effect of tranilast on the reproductive outcomes of pregnant mice. The offspring of GDM mice had a smaller litter size and heavier body weight than the offspring of WT mice. Tranilast significantly increased the litter size and decreased their body weight; however, metformin had no effect on the reproductive outcomes of GDM mice (Fig. 6A–B).