Vitamin B6 deficiency disrupts serotonin signaling in pancreatic islets and induces gestational diabetes in mice

Abstract

In pancreatic islets, catabolism of tryptophan into serotonin and serotonin receptor 2B (HTR2B) activation is crucial for β-cell proliferation and maternal glucose regulation during pregnancy. Factors that reduce serotonin synthesis and perturb HTR2B signaling are associated with decreased β-cell number, impaired insulin secretion, and gestational glucose intolerance in mice. Albeit the tryptophan-serotonin pathway is dependent on vitamin B6 bioavailability, how vitamin B6 deficiency impacts β-cell proliferation during pregnancy has not been investigated. In this study, we created a vitamin B6 deficient mouse model and investigated how gestational deficiency influences maternal glucose tolerance. Our studies show that gestational vitamin B6 deficiency decreases serotonin levels in maternal pancreatic islets and reduces β-cell proliferation in an HTR2B-dependent manner. These changes were associated with glucose intolerance and insulin resistance, however insulin secretion remained intact. Our findings suggest that vitamin B6 deficiency-induced gestational glucose intolerance involves additional mechanisms that are complex and insulin independent.

Fig. 1. Vitamin B6 deficiency induces glucose intolerance during pregnancy.

Body weights (A) and fasting glucose levels (B) were measured in control and vitamin B6 deficient (VB6D) non-pregnant, pregnant, and 3 weeks postpartum mice. Glucose time graphs are shown for each timepoint GD 9.5 (C), GD 12.5 (D), and GD 16.5 (E) in which VB6D dams are glucose intolerant. Glucose area under the curve (AUC) was elevated in VB6D mice only during pregnancy (F). Data represent 3–8 mice per treatment group. Analysis of panels A, B, and F were done by an unpaired, two-sided, t-test. Panels C–E by two-way, repeated-measures ANOVA. In panels D and E, diet × time interaction effect was observed, and post hoc Tukey’s tests were performed. *p ≤ 0.05, **p ≤ 0.01.

Fig. 2. Vitamin B6-deficient dams have normal insulin secretion, but heightened insulin sensitivity during pregnancy.

At GD 12.5, VB6D dams have normal fasting insulin levels and insulin secretion response when glucose challenged compared to controls (A). At GD 12.5, VB6D has decreased insulin sensitivity compared to control (B). At GD 16.5, VB6D dams have normal fasting insulin levels and insulin secretion response when glucose challenged compared to control (C). At GD16.5, VB6D dams are more insulin sensitive compared to control indicated by a more significant reduction in glucose compared to controls (D). Data represent 5–8 mice per treatment group. Analysis of all panels was done by a two-way, repeated-measures ANOVA. A main effect of diet was observed in panel (B). In panel D, we identified a diet × time interaction effect, and a post hoc Holm–Sidak’s test was subsequently conducted which revealed that the percent fasting glucose at T = 15 and 30 min were significantly different between the groups. *p ≤ 0.05.

Fig. 3. Reduced islet and serum serotonin in vitamin B6-deficient dams at GD 12.5 and 16.5, respectively.

Representative ×20 immunofluorescent images of GD 12.5 pancreases from control and vitamin B6-deficient or VB6D dams (A). Corrected total cell fluorescence (CTCF) was used to calculate islet serotonin in the GD 12.5 pancreases (B). In (C), we used LC-HRMS to measure absolute islet serotonin levels. Relative serum serotonin levels in GD 12.5 (D) and 16.5 (E) control and VB6D dams were measured using LC-MS. We analyzed 40–50 islets per mouse (N = 3 mice) in panels (A) and (B), 6–10 mice in panel (C), and 5–7 mice in panels (D) and (E). Unpaired, one-sided, t-tests were performed in (B–E). *p ≤ 0.05; non-significant (ns) represents p values > 0.1.

Fig. 4. HTR2B receptor agonist treatment increases β-cell proliferation in vitamin B6-deficient dams and normalizes glucose tolerance at GD 16.5.

Shown in (A) is representative ×20 immunofluorescent images of GD 12.5 pancreases from PBS and HTR2B agonist injected control and vitamin B6-deficient (VB6D) mice. Proliferating β-cells are identified using MCM-2 and Insulin antibodies and calculated as percentage of total β-cells (B). Glucose AUC for PBS and HTR2B agonist-treated control and VB6D dams at GD 16.5 is shown in (C). Data for panels A and B are based on 40–50 islets per mouse (N = 3 mice), while panel C represents 5–7 mice. Two-way ANOVA tests performed in panels B and C revealed a diet × treatment interaction effect and post hoc Tukey’s multiple comparison tests were subsequently performed. *p ≤ 0.05, **p < 0.01, ****p < 0.0001.