A potential role for insulin treatment during pregnancy in reducing postpartum psychological distress in maternal obesity: an administrative population health study.

BACKGROUND: Studies have found an association between obesity and an increased risk for peripartum depression, which has also been linked to decreased placental lactogen levels. In addition, women with obesity treated for gestational diabetes with insulin were found to have increased levels of placental lactogen. Treatment options exist for perinatal and postpartum depression however they pose a risk to the developing offspring. Thus, prevention as well as markers for early identification of peripartum depression are needed. Therefore, our study objective is to identify the association between insulin treatment in pregnancy and the risk of postpartum psychological distress (abbreviated here as PPD) among cohorts of women with and without obesity. METHODS: Administrative health data (2002/03-2018/19) were used to identify a cohort of women (age 15+ years) who gave birth (N = 250,746) and had no pre-existing mood/anxiety disorders or diabetes (N = 222,863 excluded). Women were then divided into two groups: lean (N = 17,975) and with obesity (N = 9908), which was identified by a recorded maternal weight of > 38 to < 65.6 kg and ≥ 85 to < 186 kg (respectively). The risk of PPD within one year after delivery with and without insulin treatment was assessed by Poisson regression analysis. Models were adjusted for maternal age group (at pregnancy start date) and area-level income (at delivery). RESULTS: The unadjusted risk of PPD was higher in the obesity group (8.56%; 95% CI 8.00-9.15) than in the lean group (6.93%; 95% CI 6.56-7.33). When no insulin treatment was given during pregnancy, mothers with obesity had a significantly higher risk of PPD than the lean group (aRR 1.27; 95% CI 1.17-1.39; p < 0.0001). However, when women with obesity and insulin treatment were compared to the lean group with no insulin treatment, no significant difference in the risk of PPD was observed between the groups (aRR 1.30; 95% CI 0.83-2.02; p = 0.248). CONCLUSION: This is the first study to demonstrate a positive association between insulin treatment in pregnancy among women with obesity and reduced PPD rates, suggesting insulin as a possible preventative measure. However, the biological mechanism behind the observed positive effect of insulin on PPD rates remains to be investigated.

Obesity and regulation of human placental lactogen production in pregnancy.

The four genes coding for placental members of the human (h) growth hormone (GH) family include two that code independently for placental lactogen (PL), also known as chorionic somatomammotrophin hormone, one that codes for placental growth hormone (PGH) and a pseudogene for which RNA but no protein product is reported. These genes are expressed preferentially in the villus syncytiotrophoblast of the placenta in pregnancy. In higher primates, the placental members, including hPL and PGH, are the result of multiple duplication events of the GH gene. This contrasts with rodents and ruminants, where PLs result from duplication of the prolactin (PRL) gene. Thus, unlike their mouse counterparts, the hPL and PGH hormones bind both lactogenic and somatogenic receptors with varying affinity. Roles influenced by nutrient availability in both metabolic control in pregnancy and maternal behaviour are supported. However, the effect maternal obesity has on the activation of placental members of the hGH gene family, particularly the expression and function of those genes, is poorly understood. Evidence from partially humanised hGH/PL transgenic mice indicates that both the remote upstream hPL locus control region (LCR) and more gene-related regulatory regions are required for placental expression in vivo. Furthermore, a specific pattern of interactions between the LCR and hPL gene promoter regions is detected in term placenta chromatin from women with a normal body mass index (BMI) in the range 18.5-25 kg m-2 by chromosome conformation capture assay. This pattern is disrupted with maternal obesity (class II BMI > 35 kg m-2 ) and associated with a > 40% decrease in term hPL RNA levels, as well as serum hPL but not PRL levels, during pregnancy. The relative importance of the chromosomal architecture and predicted properties for transcription factor participation in terms of hPL production and response to obesity are considered, based on comparison with components required for efficient human pituitary GH gene expression.

A useful model to compare human and mouse growth hormone gene chromosomal structure, expression and regulation, and immune tolerance of human growth hormone analogues.

Human (h) pituitary growth hormone (GH) is both physiologically and clinically important. GH reaches its highest circulatory levels in puberty, where it contributes to energy homeostasis and somatogenic growth. GH also helps to maintain tissues and organs and, thus, health and homeostasis. A reduction in the rate of hGH production begins in middle age but if GH insufficiency occurs this may result in tissue degenerative and metabolic diseases. As a consequence, hGH is prescribed under conditions of GH deficiency and, because of its lipolytic activity, stimulation of hGH release has also been used to treat obesity. However, studies of normal GH production and particularly synthesis versus secretion are not feasible in humans as they require sampling normal pituitaries from living subjects. Furthermore, human (or primate) GH structure and, as such, regulation and potential function, is distinct from non-primate rodent GH. As a result, most information about hGH regulation comes from measurements of secreted levels of GH in humans. Thus, partially humanized hGH transgenic mice, generated containing fragments of human chromosome 17 that include the intact hGH gene locus and many thousands of flanking base pairs as well as the endogenous mouse (m) GH gene provide a potentially useful model. Here we review this mouse model in terms of its ability to allow comparison of hGH versus mGH gene expression, and specifically: (i) GH locus structure as well as regulated and rhythmic expression; (ii) their ability to model a clinical assessment of hGH production in response to overeating and hyperinsulinemia as well as a possible effect of exercise, and (iii) their hGH-related immune tolerance and thus potential for testing hGH-related analogue immunogenicity.

Chromosomal architecture and placental expression of the human growth hormone gene family are targeted by pre-pregnancy maternal obesity.

The human (h) placental lactogenic hormone chorionic somatomammotropin (CS) is highly produced during pregnancy and acts as a metabolic adaptor in response to maternal insulin resistance. Maternal obesity can exacerbate this "resistance", and a >75% decrease in CS RNA levels was observed in term placentas from obese vs. lean women. The genes coding for hCS ( hCS-A and hCS-B) and placental growth hormone ( hGH-V) as well as the hCS-L pseudogene and pituitary growth hormone (GH) gene ( hGH-N) are located at a single locus on chromosome 17. Three remote hypersensitive sites (HS III-V) located >28 kb upstream of hGH-N as well as local hCS gene promoter and enhancer regions are implicated in hCS gene expression. A placenta-specific chromosomal architecture, including interaction between HS III-V and hCS but not hGH gene promoters, was detected in placentas from lean women (BMI <25 kg/m2) by using the chromosome conformation capture assay. This architecture was disrupted by pre-pregnancy maternal obesity (BMI >35 kg/m2), resulting in a predominant interaction between HS III and the hGH-N promoter, which was also observed in nonplacental tissues. This was accompanied by a decrease in hCS levels, which was consistent with reduced RNA polymerase II and CCAAT/enhancer-binding protein-ß association with individual hCS promoter and enhancer sequences, respectively. Thus, pre-pregnancy maternal obesity disrupts the placental hGH/CS gene locus chromosomal architecture. However, based on data from obese women who develop GDM, insulin treatment partially recapitulates the chromosomal architecture seen in lean women and positively affects hCS production, presumably facilitating prolactin receptor-related signaling by hCS.

Expression of Placental Members of the Human Growth Hormone Gene Family Is Increased in Response to Sequential Inhibition of DNA Methylation and Histone Deacetylation.

The genes coding for human (h) chorionic somatomammotropin (CS), hCS-A and hCS-B, and placental growth hormone (GH-V), hGH-V, are located at a single locus on chromosome 17. Efficient expression of these placental genes has been linked to local regulatory (5' P and 3' enhancer) sequences and a remote locus control region (LCR), in part, through gene transfer in placental and nonplacental tumor cells. However, low levels of endogenous hCS/GH-V transcripts are reported in the same cells compared with term placenta, suggesting that chromatin structure, or regulatory region accessibility, versus transcription factor availability contributes to the relatively low levels. To assess individual hCS-A, CS-B, and GH-V gene expression in placental and nonplacental tumor cells and the effect of increasing chromatin accessibility by inhibiting DNA methylation and histone deacetylation using 5-aza-2'-deoxycytidine (azadC) and trichostatin A (TSA). Low levels of hCS-A, CS-B, and GH-V were detected in placental and nonplacental tumor cells compared with term placenta. A significant >5-fold increase in activity was seen in placental, but not nonplacental, cells transfected with hybrid hCS promoter luciferase genes containing 3' enhancer sequences. Pretreatment of placental JEG-3 cells with azadC resulted in a >10-fold increase in hCS-A, CS-B, and GH-V RNA levels with TSA treatment compared with TSA treatment alone. This effect was specific as reversing the treatment regimen did not have the same effect. An assessment of hyperacetylated H3/H4 in JEG-3 cells treated with azadC and TSA versus TSA alone revealed significant increases consistent with a more open chromatin structure, including the hCS 3' enhancer sequences and LCR. These observations suggest that accessibility of remote and local regulatory regions required for efficient placental hGH/CS expression can be restricted by DNA methylation and histone acetylation status. This includes restricting access of the hCS 3' enhancer sequences to available placental enhancer transcription factors.

Embryonic survival and severity of cardiac and craniofacial defects are affected by genetic background in fibroblast growth factor-16 null mice.

Disruption of the X-chromosome fibroblast growth factor 16 (Fgf-16) gene, a member of the FGF-9 subfamily with FGF-20, was linked with an effect on cardiac development in two independent studies. However, poor trabeculation with lethality by embryonic day (E) 11.5 was associated with only one, involving maintenance in Black Swiss (Bsw) versus C57BL/6 mice. The aim of this study was to examine the potential influence of genetic background through breeding the null mutation onto an alternate (C57BL/6) background. After three generations, 25% of Fgf-16(-/Y) mice survived to adulthood, which could be reversed by reducing the contribution of the C57BL/6 genetic background by back crossing to another strain. There was no significant difference between FGF-9 and FGF-20 RNA levels in Fgf-16 null versus wild-type mice regardless of strain. However, FGF-8 RNA levels were reduced significantly in Bsw but not C57BL/6 mice. FGF-8 is linked to anterior heart development and like the FGF-9 subfamily is reportedly expressed at E10.5. Like FGF-16, neuregulin as well as signaling via ErbB2 and ErbB4 receptors have been linked to trabeculae formation and cardiac development around E10.5. Basal neuregulin, ErbB2, and ErbB4 as well as FGF-8, FGF-9, and FGF-16 RNA levels varied in Bsw versus C57BL/6 mice. These data are consistent with the ability of genetic background to modify the phenotype and affect embryonic survival in Fgf-16 null mice.

Inhibition of TGFbeta signaling potentiates the FGF-2-induced stimulation of cardiomyocyte DNA synthesis.

OBJECTIVE: Added transforming growth factor beta (TGFbeta) inhibits the proliferation of immature cardiomyocytes. We have now examined the hypothesis that suppression of endogenous TGFbeta signaling will boost the proliferative response (DNA synthesis) of cardiac myocytes to serum and/or to the mitogenic factor fibroblast growth factor-2 (FGF-2). METHODS AND RESULTS: Overexpression of a kinase-deficient TGFbeta type II receptor (TGFbetaRIIDeltaKD) resulted in a 2.8-fold increase in cardiomyocyte DNA synthesis in serum-rich cultures, an effect requiring active FGFR-1 since it was not observed in the presence of excess kinase-deficient FGFR-1. This finding suggested that endogenous TGFbeta-TGFbetaRII suppressed endogenous FGFR-1-mediated signals that stimulate or are permissive for DNA synthesis. TGFbeta had no effect, however, on the FGF-2-induced acute stimulation of extracellular signal regulated kinase1/2. FGF-2, added in the absence or presence of TGFbeta inhibition, elicited a 3- or a 13-fold stimulation of DNA synthesis, respectively, pointing to a synergistic effect. CONCLUSION: Inhibition of TGFbetaRII-transduced signaling upregulates the proliferative response of cardiomyocytes to serum, and greatly potentiates the stimulatory effect of FGF-2. A combinatorial strategy including activation of FGF-2 and inhibition of TGFbeta-triggered signal transduction may be required for maximal stimulation of immature cardiomyocyte DNA synthesis.