Sulforaphane induced NRF2 activation in obese pregnancy attenuates developmental redox imbalance and improves early-life cardiovascular function in offspring

Abstract

In adverse pregnancy a perturbed redox environment is associated with abnormal early-life cardiovascular development and function. Previous studies have noted alterations in the expression and/or activity of Nuclear Factor E2 Related Factor 2 (NRF2) and its antioxidant targets during human gestational diabetic (GDM) pregnancy, however to our knowledge the functional role of NRF2 in fetal 'priming' of cardiovascular dysfunction in obese and GDM pregnancy has not been investigated. Using a murine model of obesity-induced glucose dysregulated pregnancy, we demonstrate that NRF2 activation by maternal sulforaphane (SFN) supplementation normalizes NRF2-linked NQO1, GCL and CuZnSOD expression in maternal and fetal liver placental and fetal heart tissue by gestational day 17.5. Activation of NRF2 in utero in wild type but not NRF2 deficient mice improved markers of placental efficiency and partially restored fetal growth. SFN supplementation was associated with reduced markers of fetal cardiac oxidative stress, including Nox2 and 3-nitrotyrosine, as well as attenuation of cardiac mass and cardiomyocyte area in male offspring by postnatal day 52 and improved vascular function in male and female offspring by postnatal day 98. Our findings are the first to highlight the functional consequences of NRF2 modulation in utero on early-life cardiovascular function in offspring, demonstrating that activation of NRF2 affords cardiovascular protection in offspring of pregnancies affected by redox dysregulation.

Keywords: Cardiovascular; Diabetes; NRF2; Obesity; Pregnancy; Sulforaphane.

Graphical abstract

Fig. 1

Study design Proven breeder C57BL/6 wild type (WT) and NRF2 knockout (NRFKO) dams were fed a normal chow (RM1) diet (Lean) or were switched to an obesogenic (Ob) diet prior to pregnancy. Maternal metabolic dysregulation was confirmed when dams had gained ∼30% of their original body weight and developed glucose intolerance assessed ipGTT. Dams were randomly assigned to receive either dietary vehicle (corn oil, WT-LeanVeh/WT-ObVeh/NRF2KO-ObVeh) or SFN (2.5 mg/kg, WT-ObSFN/NRFKO-ObSFN) for 6 days/week throughout pregnancy and lactation. With the presence of a morning copulation plug signifying gestational day 0.5 (GD0.5), WT-LeanVeh dams fed RM3 and a subset of Ob dams were sacrificed at GD17.5 to assess fetal outcomes and placental development. Male and female offspring from the remaining Ob dams were weaned onto control RM1 diet without further veh/SFN intervention and sacrificed on postnatal day 52 (PD52, 7.5 weeks – males only) or PD98 (14 weeks) to assess cardiac morphology (PD52) and/or vascular reactivity (PD98).

Fig. 2

Obesogenic diet induces pre-gestational obesity and glucose intolerance in WT and to a lesser extent in NRF2 KO dams WT (A-C) and NRF2 KO (D-F) dams were maintained on normal chow RM1 (Lean) or fed an obesogenic diet (Ob) prior to mating. For obese groups of each genotype, littermate dams were randomly assigned to one of two groups; Group 1 (G1) subsequently went on to receive vehicle, while Group 2 (G2) subsequently received dietary SFN (2.5 mg/kg) during pregnancy. Prior to pregnancy and veh/SFN intervention, dam body weight was recorded weekly during Ob or Lean RM3 diet exposure (A, D). During the final week of pre-pregnancy diet exposure, dams were fasted for 6 h to assess fasting blood glucose (FBG), glucose tolerance (ipGTT 2 g/kg, 0–120 min) with data displayed as time course and corresponding area under curve (AUC) (B, E) and fasting insulin (C, F). Data are summarised as mean ± S.E.M., n = 4–20, with data points representing individual dams. *P < 0.05, **P < 0.01 and ***P < 0.0001 in WT-Ob vs WT-Lean, ####P < 0.0001 Nrf2KO-Ob vs WT-Ob †P < 0.05 Nrf2KO-Ob vs WT-Lean.

Fig. 3

SFN restores expression of inducible and constitutive NRF2-linked antioxidant genes in placenta from WT but not NRF2 deficient dams At mating, WT-lean dams were supplemented with veh and WT and NRFKO obese dams were supplemented with veh or SFN (2.5 mg/kg) for 6 days/week throughout pregnancy before termination at gestational day 17.5 (GD17.5). A-D, Representative immunoblots (A) and summary data showing normalised protein expression of NQO1 (B) and GCLM (C) and constitutive CuZnSOD (D) in placenta. Data denote mean ± S.E.M of n = 4 individual dam litters. *P < 0.05 and **P < 0.01.

Fig. 4

Maternal dietary SFN during obese pregnancy reduces cardiomyocyte area in hearts of male offspring Hearts from the 7-week old male offspring of WT-Ob and NRF2KO-Ob dams supplemented with veh or SFN (2.5 mg/kg) throughout pregnancy and weaning were fixed and frozen sections (12 μm) stained with wheat germ agglutinin (WGA) for measurements of cardiomyocyte cross-sectional area. Representative WGA immunofluorescence grayscale images (A) and quantitative analysis of cardiomyocyte area (B) determined transverse sections. Data denote mean ± S.E.M from measurements of at least 500 cells and 3 separate FOV from n = 3–6 individual dam litters per group. *P < 0.05.

Fig. 5

SFN suppresses oxidative stress in fetal hearts of WT obese litters resulting in normalization of NRF2-linked antioxidant enzyme expression At mating, WT-Lean dams were supplemented with veh and WT and NRF2KO obese dams were supplemented with veh or SFN (2.5 mg/kg) for 6 days/week throughout pregnancy before termination at gestational day 17.5 (GD17.5). A-B, Representative immunoblots and densitometric analysis of 3-NT and Nox2 (markers of oxidative stress, panel A), constitutive CuZnSOD and inducible NRF2 targets NQO1 and GCLC (B) expression relative to GAPDH. Data denote mean ± S.E.M of n = 4 individual dam litters per group. *P < 0.05, **P < 0.01 and ***P < 0.001.

Fig. 6

Dietary SFN attenuates resistance artery contractility in WT male and female 3-month-old offspring and improves nitric oxide independent vasorelaxation only in males A-G, Wire myography used to assess vascular reactivity of small mesenteric arteries isolated at PD98 from male (A-D) and female (E-G) offspring of WT and NRF2KO obese dams given Veh or SFN (2.5 mg/kg) throughout pregnancy and weaning. Vasoconstriction in response to increasing concentrations of noradrenaline (NA, 10⁻⁹-10⁻⁴ M) as an isometric tension (A, E). Vasorelaxation of mesenteric arteries pre-constricted with NA in response to increasing concentrations of acetylcholine (Ach, 10⁻⁹-10⁻⁴ M, panels B, F), NO donor sodium nitroprusside (SNP, 10⁻⁹-10⁻⁴ M, panels C, G) or Ach in the presence of eNOS inhibitor l-NAME (10⁻⁵ M, 30 min, panels D, H). Results are expressed as % of NA pre-constriction tone. Data denote mean ± S.E.M. of measurements in offspring from n = 5–9 individual litters per group. *p < 0.05, ***P < 0.001 and ****P < 0.0001 vs maximal responses Veh vs SFN, ^#P < 0.05 ^####P < 0.0001 WT vs NRF2KO corresponding condition.