Progenies of gestational diabetes mellitus exhibit sex disparity in metabolism after respective therapies of insulin, glibenclamide, and metformin in dams during pregnancy

Abstract

Background

The aim of this study was to compare the sex-dependent intergenerational effects of insulin, glibenclamide, and metformin on glucose and lipid metabolism in the offspring born to GDM mice.

Methods

The murine GDM was induced by high fat diet. The offspring were grouped based on the treatments in maternal mice. ITT and GTT were performed at 4th and 8th weeks of age, respectively. Serum levels of TC, TG, HDL-C, and LDL-C plus hepatic levels of TG and TC, were respectively determined by enzymatic kits. Western blotting was conducted to detect related proteins in the livers from offspring.

Results

The dyslipidaemia, hepatic lipid abnormality, and insulin insensitivity caused by GDM were persistently normalised in male adult offspring by the respective therapies of insulin, glibenclamide, and metformin during maternal pregnancy. Specifically, the decreases in plasma TC, TG, and LDL-C levels (29%, 37.8%, and 57.7%, respectively, p ˂ .05) and in hepatic lipid contents (TC 31.3% and TG 39.2%, p ˂ .05), the increases in hepatic phosphorylation levels of AKT, CPT1A, PPAR-α, and PPAR-γ (57.1%, 91.7%, 68%, and 173.3%, respectively, p ˂ .05) and the inhibition of G6Pase, PEPCK, and HMGCS1 (35.7%, 68.8%, and 77.3% respectively, p ˂ .05) were still observed in the male offspring born to treated GDM mice from 4th to 8th week of age. Unexpectedly, the aforementioned parameters in female progenies in different groups were not significantly changed compared with controls.

Conclusions

Respective treatments in GDM mice during pregnancy with insulin, glibenclamide, and metformin have the long-term persistent effects in male offspring, while female progenies born to untreated dams showed an autonomous inhibition of intergenerational relay of glucose and lipid dysregulation. Our current findings may imply a sex-dependent strategy of medical care for GDM mothers and their offspring.

Novelties

• Respective interventions of insulin, glibenclamide, and metformin on dams exerted the persisted effects on male progenies.

• Therapies of three drugs on dams had the similarly improved effects in offspring.

• Female offspring autonomously corrected their dysregulated glucose-lipid metabolism caused by gestational diabetes mellitus (GDM) in dams.

Keywords:

• GDM
• offspring
• glibenclamide
• metformin
• glucose and lipid metabolism

Background

GDM is defined as the onset of glucose intolerance during pregnancy (American Diabetes Association 2013), and uncontrolled hyperglycaemia during pregnancy affects foetal development and neonatal adaptation. Adequate treatment has a direct impact on maternal and perinatal outcomes (Crowther et al. 2005). The foetal exposure to maternal diabetes in utero increases the risk of obesity, glucose intolerance, and type 2 diabetes (T2DM) for offspring in later life (Krishnaveni et al. 2010, Portha et al. 2011). For 15–60% of women with GDM, routine care, such as dietary measures, physical activity, and glucose monitoring is not adequate to achieve glucose control (Langer 2002). Therefore, the medical therapies with hypoglycaemic agents during pregnancy may be very crucial for GDM women and their offspring to maintain health. Although insulin has been commonly used in the management of GDM (Ijas et al. 2015), it has many disadvantages including a complicated dosing schedule, an unfavourable route of administration, and reduced patient adherence. Oral hypoglycaemic agents are non-invasive and easy to administer (Ijas et al. 2015). Recently, International Federation of Gynaecology and Obstetrics (FIGO) and the UK National Institute for Health and Care Excellence (NICE), suggested insulin, glibenclamide, and metformin as appropriate first-line therapies for GDM (Hod et al. 2015, Webber et al. 2015). The rationale for the use of glibenclamide during pregnancy is based on the similarities of the pathophysiology in GDM and T2DM. Sulfonylurea (SU) drugs have been used to treat T2DM for many decades. Metformin was reintroduced more recently as treatment for GDM. Based on the available data regarding the short-term effects, glibenclamide and metformin appear to be safe and effective for the treatment of GDM (George et al. 2015, Nachum et al. 2017). But previous studies have all stressed the lack of intergenerational effect data on offspring. The unanswered questions are mainly related to offspring’s late development, growth, and glucose-lipid metabolism. Therefore, the intergenerational effects of glibenclamide and metformin on the metabolic regulation in GDM offspring are currently unknown and constitute an important barrier for their use. In this study, we used murine GDM model to explore the intergenerational impacts of respective interventions by insulin, glibenclamide, and metformin during pregnancy on the offspring born to GDM mice.

Material and methods

Mice handlings

C57BL/6J mice (8 weeks old; male mice: 20 ± 24 g; female mice: 17 ± 20 g) were obtained from Beijing Vital River. Mice were kept under observation for one week to acclimatise the new conditions and maintained under a 12 h light–dark cycle at constant temperature (23 ± 2)°C and relative humidity (50 ± 5)% with food (Beijing Huafukang Bioscience Co, LTD, Beijing 11027182, China) and water available ad libitum. The murine GDM model was established by being fed with high-fat diet (HFD; 60% kcal from fat) for 4 weeks before pregnancy (Nanobashvili et al. 2018, McIlvride et al. 2019). The pregnant mice were obtained by proesterous normal females being left for one night to copulate with the normal males (2:1). Onset of pregnancy was determined by the presence of a copulation plug, which was defined as gestational day zero (GD0). The total successful mating for GDM group was 62, while for normal controls were 11 (N0). The pregnant mice for drug treatments were fed continuously with HFD during pregnancy and diabetic state for GDM group was confirmed when the fasting plasma glucose concentration exceeded 8 mmol/l (150 mg/dl). The successful GDM models were randomly distributed into four groups (10 animals/group): GDM control without treatment (G0), GDM treated with insulin (GI0), GDM treated with glibenclamide (GG0), and GDM treated with metformin (GM0). The drug administration was conducted since the hyperglycaemia was confirmed during pregnancy.

The offspring’s groups were named based on their mothers’ groups, respectively (8–10 pups/group, randomly selected from the total numbers born to the dams in the same group), recorded as N, G, GI, GG, and GM. After delivery, the murine pups were obtained, and their weights, crown-rump lengths were measured. The offspring were fostered by their biological mothers until they were weaned after 3 weeks of age. All of offspring in experimental groups were weighed on days of 3, 7, 10, 14, 17, and 21, respectively, during lactation. Body weights and plasma glucose were measured once a week after weaning. ITT and GTT experiments were performed, respectively, at 4th and 8th weeks of age. Animal procedures were approved by the Animal Care and Use Committee, faculty of Science, Anhui Medical University, in accordance with the International Guidance principles for Biomedical Research Involving animals of CIOMS.

Insulin treatment

The therapeutic dose to the GI0 group was 7 to 10 U (human long-acting insulin, Novo Nordisk A/S, Denmark) for the initial subcutaneous injection (Shuster et al. 2014). Subsequent daily injections were adjusted according to the plasma glucose level with 1 U each time, until the FPG is less than 6 mmol/l.

Metformin and glibenclamide therapy

The mice were administered by gavage with 20 mg/kg/d of glibenclamide for GG0 group, or with 300 mg/kg/d of metformin for GM0 group. The volume of gavage administration was 1 ml/100 g body weight.

Glucose tolerance test (GTT) and insulin tolerance test (ITT)

GTTs were performed on murine offspring after 6 h fasting by injecting the glucose (2 mg/g body weight) into the intraperitoneal cavity and plasma glucose was measured via tail blood immediately before and after 0.5, 1, 1.5, 2, and 2.5 h injections using an Accu-Check glucometer (Roche, Mannheim, Germany). ITTs were performed on 4 h fasted conscious offspring, respectively, by injecting human insulin (4 weeks old, 0.45 mU/g; 8 weeks old, 0.65 mU/g body weight; Sigma, Saint Louis, MO) into the intraperitoneal cavity and plasma glucose was measured via tail blood immediately before and after 20, 40, 60, 80, 100, and 120 min injections using an Accu-Check glucometer (Roche, Mannheim, Germany).

Tissue collection and biochemical assay

After 8 weeks, blood samples were, respectively, collected from the orbital venous plexus in offspring when the mice were anaesthetised with ketamine (100 mg/kg), acepromazine (10 mg/kg), and xylazine (100 mg/kg). Whole blood obtained from offspring was centrifuged at 3000 g, 4 °C for 20 min to collect plasma. The livers were immediately removed and rinsed with cold phosphate-buffered saline (PBS) and stored at –80 °C after dip-freezing in liquid nitrogen. Insulin was analysed using an enzyme-linked ELISA kit (Elabscience, Wuhan, China). Plasma levels of TC, TG, HDL-C, LDL-C, and hepatic levels of TG and TC, were determined respectively using enzymatic kits (Nanjing Jiancheng Bioengineering Institute, Jiangsu, China).

Western blotting

Liver homogenates were prepared with RIPA lysis buffer containing a complete protease inhibitor. Lysates were centrifuged at 12,000 g, 4 °C for 15 min. The supernatant extracts were quantified for protein concentration by BCA. Of 30–50 μg of denatured proteins were loaded, resolved by 10–15% SDS-PAGE and transferred to PVDF membranes that were incubated in 5% non-fat milk at room temperature for 1 h, and then incubated with the appropriate primary antibodies respectively, overnight at 4 °C. Membranes were then washed and incubated with secondary antibodies at room temperature for 1–2 h. Target proteins were detected by enhanced chemiluminescence (ECL).

Statistical analysis

Data are expressed as mean ± SEM. The physiology and biochemistry parameters in dams were analysed using one-way ANOVA followed by SNK or two-way ANOVA followed by Tukeys multiple comparisons test. A value of p < .05 was considered to be statistically significant. ITT and GTT were additionally analysed using glucose area under the curve (AUC), calculated using trapezoidal integration.

Results

Respective interventions of insulin, glibenclamide, and metformin during pregnancy improved the hyperglycaemia and insulin resistance in GDM dams

The whole experimental scheme is shown in Figure 1(A). Based on current data, the body weights of dams in the untreated diabetic G0 group were significantly higher than those in the drug-treated groups (Figure 1(B)), indicating that the respective therapies of insulin, glibenclamide, and metformin just exerted substantial reduction effects on the maternal body weights during pregnancy (Figure 1(B)). In addition, the fasting plasma glucose levels in untreated G0 group were dramatically higher than those in N0 group, and the interventions of three drugs respectively and dramatically ameliorated this hyperglycaemia (Figure 1(C)).

Figure 1. Respective interventions of insulin, glibenclamide, and metformin during pregnancy corrected the plasma glucose and insulin resistance in GDM dams. (A) Shown is the whole experimental scheme starting from the establishment of maternal GDM model to the adulthood of offspring. (B) Demonstrated are the curves of body weights of GDM and control maternal mice during pregnancy. (C) Dramatic elevation of plasma glucose levels was induced by GDM for dams during pregnancy, which could be substantially improved by respective treatments of insulin, glibenclamide, and metformin. (D) GDM remarkably raised the plasma insulin levels in maternal mice, while the respective treatments normalised the levels. (E) GDM sharply enhanced the HOMA-IR levels in untreated maternal mice, the respective treatments, however, greatly improved the levels. The values represent the mean ± SEM, n = 10 dams/group. The columns with * are significantly different, **p < .01, ***p < .001 (two-way ANOVA by Tukeys for B and C; one-way ANOVA by SNK for D).

As expected, the GDM dams showed the significant hyperinsulinaemia, which could be normalised by the respective three-drug therapies during pregnancy (p < .05, Figure 1(D)). Similarly, untreated GDM dams showed significantly elevated HOMA-IR, which could be drastically alleviated by the respective drug therapies (p < .001, Figure 1(E)).

Respective interventions of three drugs during maternal pregnancy normalised pup’s body weights and improved offspring overweight during lactation

After G18, the dams delivered the pups which were collected and measured, respectively. The crown-rump length and body weights of the pups born to untreated diabetic G group were significantly higher than those in N group (Figure 2(A,B)), but the therapies of insulin, glibenclamide, and metform in dams, respectively, normalised these characteristics. Through the lactation (3–21 d), the growth rates of offspring born to G group were sharply faster than those in N group (p < .05, Figure 2(C)), suggesting that the untreated GDM in dams exerted the intergenerational impact on their offspring, while the respective interventions of three drugs significantly and similarly ameliorated the effects (p < .05, Figure 2(C)).

Figure 2. Insulin, glibenclamide, and metformin, respectively, normalised pup’s body weights and GDM-induced overgrowth in progenies during lactation. (A) Pup’s crown-rump length was significantly higher in untreated G group than that in control group, which could be normalised by three drug interventions. (B) GDM induced greatly higher pup’s body weights than those in healthy controls, which could be corrected by three-drug therapies. (C) The increases of body weights in offspring born to untreated dams were significantly higher than those in control and drug-treated groups during lactation. The values represent the mean ± SEM, n = 8–10 pups/group. The columns with * are significantly different, *p < .05, **p < .01, and ***p < .001 (one-way ANOVA by SNK for A and B; two-way ANOVA by Tukeys for C).

Respective interventions of three drugs during maternal pregnancy normalised the fasting plasma glucose and growth rate in male offspring at adulthood

After three weeks of lactation, male and female offspring were kept in separate cages. Body weights and plasma glucose were measured once a week. The body weights of female offspring in N, GI, GG, and GM groups were gradually reached the same levels including those in untreated G group (Figure 3(A)). Likewise, the fasting plasma glucose levels in female offspring showed the same convergent trend from 4 to 8 weeks of age. These results suggested a gradually autonomous correction of overgrowth and hyperglycaemia in female progenies (Figure 3(A,B)). However, the male offspring in untreated G group kept faster increase in body weights than those in N, GI, GG, and GM groups, and the respective interventions during maternal pregnancy showed significant intergenerational improvements on the body weights in male offspring at adulthood (Figure 3(C)). Similarly, the male progenies in untreated G group demonstrated the persisted rising hyperglycaemia compared to the N group, which could be normalised by the three respective therapies during maternal pregnancy (Figure 3(D)).

Figure 3. Respective therapies of three drugs during maternal pregnancy normalised the fasting plasma glucose levels in male adult offspring. (A) Body weight curves of female offspring in different groups from 4 to 8 weeks of age showed an autonomous correction of overweight caused by untreated GDM in dams. (B) Fasting plasma glucose changes in female offspring in different groups from 4 to 8 weeks of age indicated an autonomous improvement of hyperglycaemia induced by untreated GDM in dams. (C) Body weight curves of male offspring in different groups from 4 to 8 weeks of age exhibited significant inhibition of overweight by the respective three drug treatments. (D) Fasting plasma glucose changes in male offspring in different groups from 4 to 8 weeks of age suggested the dramatic suppression of hyperglycaemia by three-drug therapies during pregnancy. The values represent the mean ± SEM, n = 8–10/group. The bars with * are significantly different, *p < .05 (two-way ANOVA by Tukeys for C and D).

Respective therapies of three drugs in dams normalised the glucose tolerance and insulin sensitivity in all male offspring at adulthood

In order to examine the offspring’s insulin sensitivity after weaning and when they reach adulthood, we carried out an acute insulin challenge test in the female and male offspring, respectively. As shown in Figure 4, after intraperitoneal exogenous insulin injection in offspring with 4 weeks of age, the plasma glucose levels in the female offspring born to the untreated GDM dams did not show significant difference compared to those in N, GI, GG, and GM groups (Figure 4(A)), while the same tests in male progeny showed significant elevation in untreated group (Figure 4(B)). The results suggest that insulin resistance still exists in the male offspring with 4 weeks of age which could be totally corrected by the three respective drug therapies, and that female offspring may exhibit an autonomous alleviation trend. For the same test in the offspring with 8 weeks of age, after the insulin injection, the blood glucose levels in the female offspring born to the untreated G group were not different from those in intervention groups and normal control group (Figure 4(C)), indicating an automatic recovery of insulin sensitivity in female adult offspring. While the same test showed that the blood glucose levels in the male offspring born to the untreated dams in G group were still markedly higher than those in N, GI, GG, and GM groups, suggesting that the intergenerational intervention effects were still persisting for male adult offspring (Figure 4(D)). To explore the offspring’s ability to acutely handle the glucose loading, GTT experiments were performed in mice at 4th and 8th weeks of age, respectively. After exogenous glucose injection into female offspring with 4 weeks of age, the peak of glucose levels in the untreated group appeared at 30 min that was not significantly different from the 30-min-peaks in N group and other treated groups (Figure 4(E)). On the contrary, the GTT tests in male progenies at 4 weeks old demonstrated substantially higher blood glucose levels in untreated group than in N and other treated groups (Figure 4(F)). These results suggest that impaired glucose tolerance induced by untreated GDM in dams still exists in the male offspring with 4 weeks of age. Furthermore, the GTT tests conducted in female offspring with 8 weeks of age showed no any significant difference for postprandial glucose levels in all five groups (Figure 4(G)). Whereas, the same test performed in the male offspring with 8 weeks old revealed that the postprandial glucose levels in GI, GG, and GM groups were significantly lower than those in untreated G group, suggesting that three interventions in maternal mice still elicited the amelioration effects for male offspring even when they reach adulthood (Figure 4(H)).

Figure 4. Normalisation of glucose tolerance and insulin sensitivity by the respective interventions of three drugs in dams only persisted in male offspring. (A) Shown is the insulin tolerance test in female offspring in different groups at 4 weeks of age. The mice were fasted 6 h on 4th week and intraperitoneally injected with insulin at 0.45 mU/g body weight, and glucose levels were determined at 0, 20, 40, 60, 80, and 120 min after injection. Quantified results are expressed as the area under the curve. (B) Insulin tolerance tests were conducted in male offspring in different groups at 4 weeks of age. (C) Insulin tolerance tests were performed in female offspring in different groups on 8th week of age. (D) Insulin tolerance tests were done in male offspring in different groups on 8th week of age. (E) Intraperitoneal glucose tolerance tests were executed in female offspring in different groups on 4th week of age. The mice were fasted 6 h on 4th week and injected with glucose at 2 mg/g body weight, and plasma glucose levels were determined at 0, 30, 60, 90,120, and 150 min after injection. Quantified results are expressed as the area under the curve. (F) Shown are the intraperitoneal glucose tolerance tests in male offspring in different groups on 4th week of age. (G) Intraperitoneal glucose tolerance tests were conducted in female offspring in different groups on 8th week of age. (H) Exhibited are the intraperitoneal glucose tolerance tests in male offspring in different groups on 8th week of age. The same procedure as A was taken for B, C, D, respectively; and the same procedure as E was taken for F, G, H, respectively. The values represent the mean ± SEM, n = 8–10/group. The columns with * are significantly different, **p < .01, ***p < .001 (two-way ANOVA by Tukeys for B, D, F, and H).

The normalisation of hyperlipidaemia and hepatic lipid contents by respective drug therapies in dams during pregnancy still persisted in male offspring at adulthood

The lipid profiles in each group are shown in Figure 5. For the female offspring with 8 weeks of age, there were no significant differences in serum TG, TC, LDL-C, and HDL-C levels in all groups. These results indicate the possible autonomous correction of dyslipidaemia in adult female offspring (p > .05, Figure 5(A–D)). For the male progenies, compared with N group, serum TG, TC, and LDL-C levels were significantly elevated in those born to the untreated GDM dams (p < .05, Figure 5(E–G)), while there was no difference in HDL-C levels in each group (p > .05, Figure 5(H)). For the female offspring with 8 weeks of age, there were no significant differences in liver TG, TC levels in all groups, including those born to the untreated GDM maternal mice (p > .05, Figure 5(I,J)). For the male progenies, compared with N group, liver TG, and TC levels were significantly elevated in untreated G group (p < .05, Figure 5(K,L)).

Figure 5. Normalisation of blood and hepatic lipid levels by respective interventions of three drugs during maternal pregnancy only persisted in male offspring at adulthood. (A) Demonstrated are plasma TG levels in female offspring in different groups on 8th week. (B) Shown are plasma TC levels in female offspring in different groups on 8th week. (C) Plasma LDL-C levels were detected in female offspring in different groups on 8th week. (D) Plasma HDL-C levels were assayed in female offspring in different groups on 8th week. (E) Exhibited are plasma TG levels in male offspring in different groups on 8th week. (F) Plasma TC levels were analysed in male offspring in different groups on 8th week. (G) Shown are plasma LDL-C levels in male offspring in different groups on 8th week. (H) Detected are plasma HDL-C levels in male offspring in different groups on 8th week. (I) Hepatic TG levels were assayed in female offspring at 8 weeks of age. (J) Hepatic TC levels were analysed in female offspring at 8 weeks of age. (K) Hepatic TG levels were detected in male offspring at 8 weeks of age. (L) Shown are hepatic TC levels in male offspring at 8 weeks of age. The values represent the mean ± SEM, n = 8–10/group. The columns with * are significantly different, *p < .05, **p < .01, ***p < .001 (one-way ANOVA by SNK for E, F, K, and L).

The female adult offspring demonstrated the autonomous normalisation of insulin sensitivity, glucose, and lipid metabolism

To evaluate insulin resistance in offspring, we measured fasting plasma glucose and serum insulin in male and female progenies with 8 weeks of age. The results showed that there were no differences in fasting plasma glucose and serum insulin levels between all groups for all female offspring (p > .05, Figure S1(A,B)). And the HOMA-IR in female adult offspring from untreated G group was not different from those in other intervention groups (p > .05, Figure S1(C)). To explore the mechanism by which the interventions impact insulin sensitivity, we tested the hepatic levels of p-AKT (Thr473 and Ser308) and p-FOXO1. Our results showed that the levels of p-AKT and p-FOXO1 in female adult progenies with 8 weeks of age did not differ between all groups, including untreated G control group (Figure S1(D,E)). PEPCK and G6Pase catalyse committed steps of gluconeogenesis, thus playing important roles in glucose homeostasis. The protein expressions of PEPCK and G6Pase in female adult progenies did not differ between all groups either (Figure S1(F)). SREBP2 and HMGCS1 play an important role in the regulation of cholesterol synthesis. Our immunoblots showed that the protein expressions of hepatic SREBP2 and HMGCS1 in female adult progenies did not differ between the treated and untreated groups (Figure S1(G)). ACL plays a fundamental role in lipogenesis and steroidogenesis. Its activation provides the building blocks for fatty acid biosynthesis. The ratios of p-ACL to ACL did not differ between all groups (Figure S1(H)). CPT1A is a rate-limiting enzyme for fatty acid β-oxidation. It can transfer medium- and long-chain fatty acids into mitochondria for β-oxidation, which reduces the deposition of lipids in peripheral tissues. The hepatic protein expressions of CPT1A in female adult progenies did not differ between the untreated and treated groups (Figure S1(I)). PPAR-α and PPAR-γ are the nuclear transcription factors. They are not only key regulators of adipose tissue differentiation and lipid metabolism, but also necessary molecules to maintain insulin sensitivity. The hepatic protein expressions of PPAR-α and PPAR-γ in female adult progenies did not differ between the untreated and treated groups (Figure S1(J)).

The normalisation of insulin sensitivity, glucose and lipid metabolism by respective therapies of three drugs in dams persisted in male offspring at adulthood

The male adult offspring born to the untreated GDM dams showed significantly elevated HOMA-IR owing to the substantially enhanced FPG levels in the G group, which could be normalised by the respective therapies of insulin, glibenclamide, and metformin during maternal pregnancy (p < .05, Figure 6(A,C)), despite the similar plasma insulin levels between all groups for male adult offspring (p > .05, Figure 6(B)). Likewise, to explore the mechanism by which the interventions enhanced insulin sensitivity, we found that the hepatic p-AKT (Thr473 and Ser308) and p-FOXO1 ratios in male adult progenies born to the untreated GDM dams were still significantly reduced compared with the N group. However, the treatments restored p-AKT and p-FOXO1 levels in male adult offspring with 8 weeks of age, similar to those in N group (Figure 6(D,E)). The significant increases in PEPCK and G6Pase were observed in the livers of male adult offspring in untreated G group, compared with those in N group, which were normalised by the three-drug therapies during maternal pregnancy (Figure 6(F)), except that metformin has no inhibitory effect on the expression of G6Pase in the livers of male offspring (Figure 6(F)). SREBP2 in male adult progenies did not differ between all groups (Figure 6(G)). HMGCS1 was significantly increased in the livers of male adult offspring in untreated G group, compared with N group, and the elevations were sharply inhibited by three drugs, respectively (Figure 6(G)). The ratios of p-ACL to ACL were significantly decreased in untreated G group compared with N group (Figure 6(H)), suggesting the increased ACL activity in G group. However, the activity could be dramatically inhibited by three drug interventions compared with N group (Figure 6(H)). The protein expressions of PPAR-α and CPT1A were significantly decreased in the livers of male adult offspring in untreated G group, compared with those in N group (Figure 6(I,J)). Nevertheless, the PPAR-α and CPT1A levels could be normalised in male adult progenies born to the treated dams, indicating an intergenerational effect by the three-drug interventions during maternal pregnancy (Figure 6(I,J)). Likewise, the protein expression of PPAR-γ significantly decreased in untreated G group compared with N group (Figure 6(J)), which however could be significantly enhanced by the three-drug therapies in dams.

Figure 6. Normalisation of hepatic insulin signalling, gluconeogenesis and de novo lipogenesis by the drug interventions during maternal pregnancy still persisted in male offspring at adulthood. (A) Shown are the fasting plasma glucose in male offspring at 8 weeks of age in different groups. (B) Plasma insulin levels were detected in male offspring at 8 weeks of age in different groups. (C) HOMA-IR values were calculated in male offspring at 8 weeks of age in different groups. (D) Decreased levels of phosphorylated AKT (p-AKT S473 and p-AKT T308) caused by maternal GDM were drastically improved in the livers of drug-treated male offspring at 8 weeks of age. (E) Inhibited levels of phosphorylated FOXO1 caused by maternal GDM were dramatically enhanced in the livers of drug-treated male offspring at 8 weeks of age. (F) The elevated protein expressions of PEPCK and G6Pase in untreated G group were substantially suppressed in the livers of drug-treated male offspring at 8 weeks of age. (G) The boosted protein expression of HMGCS1 not SREBP2 caused by maternal GDM was sharply improved in the livers of drug-treated male offspring at 8 weeks of age. (H) The enhanced levels of phosphorylated ACL caused by maternal GDM were markedly improved in the livers of drug-treated male offspring at 8 weeks of age. (I) The suppressed protein expression of CPT1A caused by maternal GDM was substantially reinstated in the livers of drug-treated male offspring at 8 weeks of age. (J) The inhibited hepatic PPAR-α and PPAR-γ in the male progenies born to untreated GDM dams at 8 weeks of age were substantially restored by three drug treatments. All data in this figure are representatives of three independent experiments. The columns with * are significantly different, **p < .01, ***p < .001 (one-way ANOVA by SNK for A–G).

Discussion

The children born to the mothers with GDM are prone to develop chronic diseases, such as obesity, impaired glucose tolerance and T2DM in adulthood. In this study, our HFD-induced GDM mouse model just clearly demonstrated the dramatic elevation of blood glucose, the significant hyperinsulinaemia and insulin resistance that mimics the human GDM state (Figure 1(D,E)). In particular, the hyperglycaemia in our GDM mice could be significantly ameliorated by the glibenclamide, a representative of insulin secretion enhancer (Figure 1(C)), suggesting the presence of intact β-cells in our current GDM dams.

These results show that insulin, glibenclamide and metformin, respectively, alleviated pup’s overgrowth and improved offspring’s body weights during breastfeeding. These drugs possibly worked by improving intrauterine hyperglycaemia, inflammation and oxidation.

GTT is widely applied to assess glucose homeostasis. Glucose stimulates the pancreas to secrete insulin, which increases the utilisation of blood glucose. A reduction in glucose homeostasis in diabetic mice may result in marked hyperglycaemia by the GTT (Lebovitz 1999, Hanson et al. 2002). Correspondingly, ITT is widely applied to assess insulin sensitivity. In spite of the aforementioned advantages, euglucemic-hyperinsulinaemic clamp is usually regarded as better way to assess the insulin sensitivity. We did not use this method in this study due to the difficult practice in small C57BL/6j mice.

In this study, we unexpectedly found that the female offspring born to the untreated GDM dams automatically became normalised when reach their adulthood with 8 weeks of age, although the normalisation trend occurred since they were at 4 weeks old. These findings may be similar with the reports from Jun Ren et al. and Di Xiao et al, suggesting that female offspring might have better glucose tolerance and insulin sensitivity than male counterparts (Xiao et al. 2017, Ren et al. 2018). The rational explanation of the sex differences might be attributed to the sexual hormones.

The prevalence of T2DM has shown sex disparities. Premenopausal women exhibit enhanced insulin sensitivity and reduced incidence of T2DM compared with age-matched men, but this advantage disappears after menopause with disrupted glucose homeostasis, in part owing to a reduction in circulating 17β-estradiol (E2) (Danaei et al. 2011). Both clinical and animal studies show a strong correlation between oestrogen deficiency and metabolic dysfunction (Misso et al. 2003, Salpeter et al. 2006). The reduction of oestrogen in postmenopausal women accelerates the development of insulin resistance and T2DM (Louet et al. 2004). Clinical trials of oestrogen replacement therapy in postmenopausal women demonstrated an amelioration of insulin resistance and reductions of plasma glucose level and incidence of T2DM (Kim et al. 2013, Howard et al. 2013). Hui Yan et al. demonstrated that E2 improves insulin sensitivity and suppresses hepatic gluconeogenesis through inhibition of Foxo1 by activation of ERα-PI3K-Akt signalling in male and ovariectomised female control and liver-specific Foxo1 knockout mice (Yan et al. 2019). It was reported that oestrogen receptor a (ERa) regulated β-cell formation during pancreatic development (Yuchi et al. 2015). However, some studies also indicated that the serum testosterone level of male patients and male animals with obesity and related diseases was significantly lower than controls, which was closely related to the disorder of glucose and lipid metabolism (Boese AC et al. 2017, Moretti et al. 2017). Exogenous testosterone supplementation could alleviate the occurrence and development of obesity and related diseases and improve glucose and lipid metabolism, which confirmed the important role of androgen in the occurrence of these diseases and the regulation of glucose and lipid metabolism (Lee and Tillman 2016, Maseroli et al. 2016). However, unlike men, women with too high androgen levels, such as patients with polycystic ovary syndrome, may develop obesity, disorders such as glucose and lipid dysregulation, insulin resistance, and diabetes (Kelly et al. 2016, Boese et al. 2017, Moretti et al. 2017). Therefore, the better compensation of the glucose metabolic function in female offspring might be due to the relative high concentration of oestrogen in female pups. Di Xiao et al. also reported that FGR female offspring have higher mRNA expression of insulin receptor in livers than the male progenies (Xiao et al. 2017). Therefore, the improved hepatic insulin/IGF1 signalling pathways may be one of the reasons for the normalised insulin sensitivity. In addition, several studies indicated long-term exercise training improves insulin sensitivity in both liver and muscle (Murakami et al 1997). Female mice naturally exhibited a higher physical activity than male mice (Lightfoot 2008). We suspect that such a higher physical activity may explain this sex difference to a certain extent.

At hepatic level, activated PPAR- α enhances lipolysis, mitochondrial β-oxidation and triacylglyceride degradation (Minnich et al. 2001), while reducing lipogenesis through upregulation of PPAR- α with concomitant SREBP-1c downregulation (Hernández-Rodas et al. 2017, Ren et al. 2017, Lee et al. 2017). In human hepatocytes, upregulation of PPAR-γ repressed expression of SREBP-2 and HMGCR which are involved in cholesterol biosynthesis (Klopotek et al. 2006, Han et al. 2019). In addition, activation of PPAR-γ results in systemic insulin sensitisation through complex mechanisms involving multiple organs. In the adipose tissue, PPAR-γ promotes lipid uptake and storage. In the skeletal muscle, PPAR-γ ligands potentiate insulin action, and in the liver it acts to suppress gluconeogenesis, altogether leading to lower plasma glucose levels (Díaz-Delfín et al. 2007, Hevener et al. 2007, Odegaard et al. 2007). This data showed the good consistence with these findings.

Taken together, respective treatments in GDM dams during pregnancy with glibenclamide and metformin may be essential to increase insulin sensitivity, to lower blood lipid levels and to improve glucose metabolism in male offspring even when they reach adulthood. Glibenclamide and metformin interventions during maternal pregnancy may have the similar intergenerational effects as do insulin intervention for male offspring.

Conclusion

Respective treatments in GDM dams during pregnancy with insulin, glibenclamide and metformin have the long-term persisted effects in male offspring, while female progenies born to untreated dams showed an autonomous inhibition of intergenerational relay of glucose and lipid dysregulation. Our current findings may imply a sex-dependent strategy of medical care for GDM mother and offspring. Specifically, more medical care or more attention may be given to the pregnant woman with diabetes herself if she is having a female foetus, and the hyperglycaemia during pregnancy may be intervened by physical activity and life style change instead of drug or insulin treatment. After delivery, the female offspring may be treated as normal and healthy children. However, if a pregnant woman with diabetes is having a male foetus, the medical care may be applied to both mother and foetus, and hypoglycaemic drug or insulin intervention would be necessary for the mother and foetus during pregnancy.

Ethics approval and consent to participate

Animal procedures were approved by the Animal Care and Use Committee, faculty of Science, Anhui Medical University, in accordance with the International Guidance principles for Biomedical Research Involving animals of CIOMS.

Consent for publication

All authors listed have approved the manuscript for publication.

Author contributions

K.C and Y. L designed the research, provide financial support and revised/edited the manuscript; Y.L, Y.X provided partial research fund; Y. L^#, Y.J, and J.L performed all experiments; J.L analysed the data; Y. L^# and Y.J wrote the draft of manuscript.

Acknowledgements

The authors thank all the participants in this study.

Disclosure statement

The authors declared no conflicts of interest.

Availability of data and materials

All data in this study are available from the corresponding author on reasonable request.

Supplemental Material available at: DOI: https://doi.org/10.6084/m9.figshare.14617119 URL: https://figshare.com/s/c8fa04af4e7dc6f86676

Additional information

Funding

This work is supported by the National Natural Science Foundation of China [NSFC, 81570786 to K.C, 81770295 to Y.L], and The Key Research and Development Project of Anhui province of China [201904d07020003 to Y. X].