Placental mTOR complex 1 regulates fetal programming of obesity and insulin resistance in mice

Abstract

Fetal growth restriction, or low birth weight, is a strong determinant for eventual obesity and type 2 diabetes. Clinical studies suggest placental mechanistic target of rapamycin (mTOR) signaling regulates fetal birth weight and the metabolic health trajectory of the offspring. In the current study, we used a genetic model with loss of placental mTOR function (mTOR-KOPlacenta) to test the direct role of mTOR signaling on birth weight and metabolic health in the adult offspring. mTOR-KOPlacenta animals displayed reduced placental area and total weight, as well as fetal body weight at embryonic day (E) 17.5. Birth weight and serum insulin levels were reduced; however, β cell mass was normal in mTOR-KOPlacenta newborns. Adult mTOR-KOPlacenta offspring, under a metabolic high-fat challenge, displayed exacerbated obesity and metabolic dysfunction compared with littermate controls. Subsequently, we tested whether enhancing placental mTOR complex 1 (mTORC1) signaling, via genetic ablation of TSC2, in utero would improve glucose homeostasis in the offspring. Indeed, increased placental mTORC1 conferred protection from diet-induced obesity in the offspring. In conclusion, placental mTORC1 serves as a mechanistic link between placental function and programming of obesity and insulin resistance in the adult offspring.

Keywords: Diabetes; Endocrinology; Islet cells; Metabolism; Obesity.

Figure 1. Reduced placental weight and fetal body weight in mTOR-KO^Placenta embryos.

Schematic diagram and timeline of the mTOR-KO study (A). Endogenous GFP reporter (CAG-ZsGreen1) in control (CYP19Cre^–), positive control (CYP19Cre⁺) experimental placentas (B). Representative Western blot of E15.5 placenta lysates for mTOR and phosphorylated S6 (Ser 240) and respective quantifications normalized to vinculin (C). H&E staining of control (Ctrl) and mTOR-KO^Placenta (D, junctional zone [JZ] and labyrinth zone [LZ]). Placental weight of E17.5 males (n = 25 Ctrl, 16 mTOR-KO^Placenta) and females (n = 24 Ctrl, 18 mTOR-KO^Placenta, E). E17.5 fetal weight separated by male (n = 25 Ctrl, 16 mTOR-KO^Placenta) and female (n = 23 Ctrl, 18 mTOR-KO^Placenta, F). P0 (within 12 hours of birth) body weight separated by male (n = 13 Ctrl, 5 mTOR-HET^Placenta, 11 mTOR-KO^Placenta) and female (n = 10 Ctrl, 4 mTOR-HET^Placenta, 5 mTOR-KO^Placenta, G). Random blood glucose was measured from trunk blood of P0 males (n = 13 Ctrl, 5 mTOR-HET^Placenta, 10 mTOR-KO^Placenta) and females (n = 10 Ctrl, 4 mTOR-HET^Placenta, 5 mTOR-KO^Placenta, H). Random insulin serum was measured from trunk blood of P0 Ctrl, mTOR-HET^Placenta, and mTOR-KO^Placenta pups (n = 25, 13, 10, respectively) from 5 dams (I). Random insulin serum was measured from trunk blood of P0 Ctrl, mTOR-HET^Placenta, and mTOR-KO^Placenta males (n = 17, 6, 10, respectively) and females (n = 10, 7, 9, respectively) from 5 dams (J). Fetal ³H-Leucine uptake after 10 minutes of leucine placental perfusion in utero (n = 4–6) from 3 dams (K). Statistical analysis was performed using 2-tailed Mann-Whitney (C, I, and K) and 2-way ANOVA with Sidak’s multiple comparisons (E–G, H, and J). Error bars represent mean ± SEM. *P < 0.05 Ctrl vs. mTOR-KO^Placenta. Scale bars in images (B and D) are 200 μm.

Figure 2. mTOR-KO^Placenta females become obese, glucose intolerant, and insulin resistant after HFD treatment.

Weight of Ctrl versus mTOR-KO^Placenta over 90 days of HFD starting at 17 weeks of age (n = 6, 7, respectively; A). EchoMRI body composition measurements after 11 weeks of HFD (n = 6, 7, B) and body fat percentage (C). Food consumed per gram body weight over 48 hours (n = 4, 6, D). VO₂ of Ctrl versus mTOR-KO^Placenta at 11 weeks of HFD (n = 6, E). VCO₂, RER, and energy expenditure (n = 6, F–H). Random blood glucose of HFD (n = 6, 7, I). Intraperitoneal glucose tolerance test (IPGTT) after 16 hours of fasting on HFD at 6 weeks (n = 6, 7, J). Insulin tolerance test (ITT) after 6 hours of fasting at 8 weeks of HFD (n = 6, 7, K). ITT after 6 hours of fasting at 12 weeks of HFD (n = 6, 7, L). Random serum insulin at 10 weeks of HFD (n = 6, 7, M). β Cell mass at 13 weeks of HFD (n = 4 each group, N). In vivo GSIS and secretory ratio at 10 weeks of HFD (n = 4 Ctrl, 5 mTOR-KO^Placenta, O and P). In vivo GSIS on NCD, 12 weeks old (n = 8 each group, Q). In vitro GSIS on normal chow, 12 weeks old (n = 5 each group, R), and islet insulin content (n = 5 each group, S). β Cell mass at 12 weeks old (n = 3 Ctrl, 3 mTOR-KO^Placenta, T). Random insulin serum of normal chow at 70 weeks old (n = 4 each group, U). In vivo and in vitro GSIS of normal chow 70-week-old mice (n = 4 each group, V and W). Islet insulin content of normal chow 70-week-old mice (n = 4 each group, X). LG, low glucose; HG, high glucose. Statistical analysis was performed using 2-tailed Mann-Whitney (C, D, M, N, P, S–U, and X) and 2-way ANOVA Sidak’s multiple comparisons (A, B, E–L, O, Q, R, V, and W) with repeated measures when appropriate. Error bars represent mean ± SEM. *P < 0.05, **P < 0.01, ***P < 0.001, Ctrl vs. mTOR-KO^Placenta. ^#P < 0.05, ^##P < 0.01, ^###P < 0.001 vs. time point 0 or LG (within genotype).

Figure 3. TSC2-KO^Placenta newborn offspring have normal placental weight and β cell mass.

Schematic diagram and timeline of the TSC2-KO study (A). H&E staining of E17.5 Ctrl and TSC2-KO^Placenta (B). Placental weights of male Ctrl versus TSC2-KO^Placenta (n = 7, 6) and female Ctrl versus TSC2-KO^Placenta (n = 7, 8) mice at E17.5 (C). Fetal weights of male (n_E17.5 = 8, 5; n_P0 = 10, 6) and female (n_E17.5 = 7, 8; n_P0 = 6, 9) Ctrl vs. TSC2-KO^Placenta mice at E17.5 and P0 (D). Pancreas weights of male (n_E17.5 = 8, 5; n_P0 = 8, 6) and female (n_E17.5 = 7, 8; n_P0 = 6, 9) mice measured at E17.5 and P0 (E). Random blood glucose of males (n = 8, 6) and females (n = 6, 9) at P0 (F). Random serum insulin of males (n = 4, 6) and females (n = 4, 6) at P0 (G). β Cell mass of males (n = 4 Ctrl, 2 TSC2-HET^Placenta, 3 TSC2-KO^Placenta) and females at P0 (n = 7 Ctrl, 4 TSC2-HET^Placenta, 5 TSC2-KO^Placenta, H). Whole pancreatic insulin content of males (n = 4 Ctrl, 6 TSC2-KO^Placenta) and females (n = 5 Ctrl, 6 TSC2-KO^Placenta) at P0 (I). Statistical analysis was performed using 2-way ANOVA with Sidak’s multiple comparisons (C–I). Error bars represent mean ± SEM. Scale bars in images are 500 μm. *P < 0.05 Ctrl vs. TSC2-KO^Placenta.

Figure 4. Adult male TSC2-KO^Placenta mice are protected from obesity-induced metabolic dysfunction.

Weight progression of Ctrl versus TSC2-KO^Placenta male mice fed with HFD over 9 weeks starting at 17 weeks of age (n = 4, 5, respectively, A). Food intake measurements of HFD male mice (n = 4, 5, B). Random blood glucose of HFD male mice (n = 4, 5, C). IPGTT after 16 hours of fasting at 2 weeks HFD (n = 4, 5, D) and corresponding random and fasting blood glucose (n = 4, 5, E). IPGTT after 16 hours of fasting at 4 weeks of HFD (n = 4, 5, F) in addition to random and fasting blood glucose (n = 4, 5, G). ITT at 5 weeks of HFD after 6 hours of fasting (n = 4, 5, H). HOMA-IR of HFD males at 6 weeks (n = 4, 5, I). Random insulin serum levels at 1 and 4 weeks of HFD (n = 4, 4–5, J). In vivo GSIS on males at 6 weeks of HFD (n = 4, 5, K) and concurrent IPGTT (n = 4, 5, L). β Cell mass of males at 10 weeks of HFD (n = 4, 5, M). Statistical analysis was performed using 2-tailed Mann-Whitney (I and M) and 2-way ANOVA with Sidak’s multiple comparisons (A–H and J–L) with repeated measures when appropriate. Error bars represent mean ± SEM. *P < 0.05, **P < 0.01 Ctrl vs. TSC2-KO^Placenta. ^##P < 0.01 vs. time point 0 (within genotype).

Figure 5. Adult female TSC2-KO^Placenta offspring are protected from obesity-induced metabolic dysfunction.

Weight progression of female mice fed with HFD over the course of 12 weeks starting at 8 weeks old (n = 6 Ctrl, 4 TSC2-KO^Placenta, A). Random blood glucose of HFD females over the duration of treatment (n = 6, 4, B). Female IPGTT after 16-hour fast at 6 weeks of HFD (n = 6, 4, C). Female ITT after 6-hour fast at 8 weeks of HFD (n = 6, 4, D). Blood glucose and insulin serum measurements at random and fasting times of 8-week HFD females (n = 6, 4, E; n = 5, 3, F). HOMA-IR of HFD females at 8 weeks (n = 5, 3, G). Pancreas weight (n = 6, 4, H) and β cell mass (n = 6, 4, I) of females at 12 weeks of HFD. In vivo GSIS of aged Ctrl versus TSC2-KO^Placenta females under normal diet (n = 3, 5, J). In vitro GSIS of 12-week-old female islets under low glucose, high glucose, high glucose plus palmitate, and KCl stimulation (n = 7, 6, K). Insulin content normalized to DNA (n = 7, 6, L). β Cell mass of female Ctrl versus TSC2-KO^Placenta under normal chow (n = 3, 4, M). Statistical analysis was performed using 2-tailed Mann-Whitney (G–I, L, and M) and 2-way ANOVA with Sidak’s multiple comparisons in (A–F, J, and K) with repeated measures when appropriate. Error bars represent mean ± SEM. *P < 0.05, **P < 0.01, ***P < 0.001 Ctrl vs. TSC2-KO^Placenta. ^#P < 0.05 vs. time point 0 or LG (within genotype).