Cardiovascular outcome in 12-month-old male and female offspring of metformin-treated obese mice

Abstract

Metformin is increasingly used to treat diabetes in pregnancy, but the effects on adult offspring health remain under-explored. The present study investigated the long-term cardiovascular effects in male and female offspring of maternal metformin treatment using a well-established mouse model of obese glucose intolerant pregnancy. Female mice were given chow, or an obesogenic diet with/without 300 mg kg^-1 day^-1 oral metformin during gestation. At 3, 6 and 12 months of age, male and female offspring were studied longitudinally with tail-cuff plethysmography and echocardiography. At 12 months, tissues were collected for wire myography, histology and molecular analyses. Female offspring of obese dams had elevated blood pressure throughout life, cardiac diastolic dysfunction at 3 months, and increased femoral vasoconstrictor reactivity and aortic wall remodelling at 12 months. Metformin treatment did not ameliorate these effects and led to obesity-induced hypertension at 12 months. Irrespective of metformin, male offspring of obese pregnancy had cardiac diastolic dysfunction from 6 months without changes in blood pressure. Male metformin-exposed offspring also showed cardiomegaly, increased cardiac collagen and vascular sympathetic hyperreactivity, suggesting metformin exposure worsened the cardiovascular phenotype. These findings show that maternal obesity caused sex-specific cardiovascular aberrations in aged offspring. Maternal metformin was not corrective and introduced further sex-dependent cardiovascular alterations. Further long-term offspring follow up of both sexes is needed for informed decisions about metformin during pregnancy. KEY POINTS: The oral medication metformin is increasingly used to treat diabetes in pregnancy. Metformin readily crosses the placenta, and long-term effects on offspring cardiovascular health remain unexplored in human and animal studies. In a mouse model of maternal diet-induced obesity with impaired glucose tolerance, female and male offspring developed hypertension and diastolic cardiac dysfunction, respectively, by 12 months of age (equivalent to middle age in humans). Maternal metformin treatment worsened the cardiovascular phenotype and introduced further sex-dependent cardiovascular alterations in both male (cardiac stiffening, vascular dysfunction) and female (obesity-induced hypertension) offspring. This work highlights that long-term cardiovascular follow up in offspring of both sexes from human pregnancies treated with metformin is crucial to make more informed decisions about metformin use in diabetic pregnancy.

Keywords: cardiac function; gestational diabetes mellitus; hypertension; metformin.

Figure 1. Animal model

Con = offspring from control‐fed dams; Ob, offspring from dams fed obesogenic diet; Ob‐Met, offspring from dams fed obesogenic diet treated with 300 mg kg⁻¹ day⁻¹ metformin according to previously published work (Schoonejans et al., 2021, 2022). HFHS, high‐fat high‐sugar diet. TD‐NMR, time‐domain nuclear magnetic resonance. *Offspring n = a/b refers to sibling pairs for which n = a siblings were used for in vivo physiological measurements and histology and n = b siblings for wire myography and molecular analyses. Created with BioRender.com.

Figure 2. Blood pressure and vascular function in 12‐month‐old offspring

A, systolic blood pressure (SBP) and heart rate (HR) measured in female offspring by tail cuff plethysmography. B–C, femoral arterial response curves to (B) increasing concentrations of K⁺ (n = 11 Con, n = 9 Ob, n = 11 Ob‐Met), or (C) increasing doses of phenylephrine (PE) normalised to 10⁻⁸  m PE and the vessel's maximal contraction to elevated K⁺ (n = 11 Con, n = 9 Ob, n = 11 Ob‐Met), measured in female offspring using wire myography. D, aortic tunica media thickness in 12‐month‐old female offspring as measured by histology and imaged using brightfield light microscopy (n = 10 Con, n = 10 Ob, n = 6 Ob‐Met) with representative images of aortas from Con, Ob and Ob‐Met females. Scale bar (lower left corner) = 50 µm. Arrows point to tunica media (TM) and tunica adventitia (TA). E, SBP and HR measured in male offspring by tail cuff plethysmography. F and G, femoral arterial response curves to (F) increasing concentrations of K⁺ (n = 9 Con, n = 9 Ob, n = 7 Ob‐Met), or (G) increasing doses of PE normalised to 10⁻⁸  m PE and the vessel's maximal contraction to elevated K⁺ (n = 8 Con, n = 5 Ob, n = 7 Ob‐Met), measured in male offspring using wire myography. P values reflect one‐way ANOVA with Tukey's multiple comparisons (A, D and E) or using repeated measures mixed model analysis with Tukey's multiple comparisons (B, C, F and G). Significant P values are shown in bold. Curves are the best non‐linear sigmoidal curve fit.

Figure 3. Left ventricular function in female offspring

A, isovolumetric relaxation time (IVRT, n = 7–12 per group). B, left ventricular myocardial performance index (MPI, n = 6–12 per group). C, deceleration time (DT, n = 7–12 per group). D, ratio between early and atrial contraction‐related ventricular filling velocity (E/A, n = 6–12 per group). E, stroke volume (SV, n = 8–12 per group). F, cardiac output (CO, n = 8–12 per group). G, end‐diastolic left ventricular volume (EDV, n = 9–12 per group) in the same 3‐, 6‐ and 12‐month‐old female offspring using echocardiography. H, cardiac weight as determined by dissection at 12 months of age (n = 11–12 per group). P values reflect one‐way ANOVA with Tukey's multiple comparisons, or non‐parametric alternative where appropriate. Significant P values are shown in bold.

Figure 4. Left ventricular function in male offspring

A, cardiac output (CO, n = 8–12 per group). B, stroke volume (SV, n = 8–12 per group). C, early ventricular filling velocity (E‐wave, n = 7–12 per group). D, atrial contraction‐related ventricular filling velocity (A‐wave, n = 7–12 per group). E, ratio between early and atrial contraction‐related ventricular filling velocity (E/A, n = 7–12 per group). F, deceleration time (DT, n = 7–12 per group) in 3‐, 6‐ and 12‐month‐old male offspring using echocardiography. G, expression of left ventricular collagen genes by real time quantitative PCR (n = 7–12 per group). H, correlation between mitral valve deceleration time (DT) and left ventricular expression of Col4a1, measured in sibling pairs (n = 26). P values reflect one‐way ANOVA with Tukey's multiple comparisons or non‐parametric alternative where appropriate (A–G and I); or two‐tailed Pearson correlation analysis (H). Significant P values are shown in bold.

Figure 5. Summary of cardiovascular outcomes in aged offspring

Female offspring (left) and male offspring (right). Ob (red), offspring of obesogenic diet‐fed dams; Ob‐Met (blue), offspring of dams fed an obesogenic diet supplemented with metformin in gestation. Created with BioRender.com.