Epigenetic Regulation of HSD11B2 Gene by Promoter Methylation in Glucocorticoid-Treated PatientsThe enzyme 11 beta-hydroxysteroid dehydrogenase type 2 11 beta HSD2 is selectively expressed in aldosterone target tissues, where it confers aldosterone selectivity for the mineralocorticoid receptor by inactivating 11 beta-hydroxyglucocorticoids. Variable activity of 11 beta HSD2 is relevant for blood pressure control and hypertension. The present investigation aimed to elucidate whether an epigenetic mechanism, DNA statistics of college athletes using steroids, accounts for the rigorous control of expression of the gene encoding epigenetic regulation of 11 beta-hydroxysteroid dehydrogenase type 2 expression beta HSD2, HSD11B2. CpG islands covering the promoter and exon 1 of HSD11B2 were found to be densely methylated in tissues and cell lines with low expression but not those with high expression of HSD11B2. Demethylation induced by 5-aza-2'-deoxycytidine and procainamide enhanced the transcription and activity of the 11 beta Eehydrogenase enzyme in human cells in vitro and in rats in vivo.
JCI - Citations to Epigenetic regulation of 11β-hydroxysteroid dehydrogenase type 2 expression
Maternal exposure to stress during pregnancy is associated with significant alterations in offspring neurodevelopment and elevated maternal glucocorticoids likely play a central role in mediating these effects. However, previous studies indicate that maternal adversity during the prenatal period can lead to a down-regulation of this enzyme.
In the current study, we examined the impact of prenatal stress chronic restraint stress during gestational days 14—20 in Long Evans rats on HSD11B2 mRNA in the placenta and fetal brain E20 and assessed the role of epigenetic mechanisms in these stress-induced effects.
Within individuals, we identified CpG sites within the HSD11B2 gene promoter and exon 1 at which DNA methylation levels were highly correlated between the placenta and fetal cortex. These findings highlight the tissue specificity of epigenetic effects, but also raise the intriguing possibility of using the epigenetic status of placenta to predict corresponding changes in the brain.
February 9, ; Accepted: May 31, ; Published: This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
The authors have declared that no competing interests exist. In humans, the experience of stress during pregnancy is associated with increased risk of preterm birth, reduced birth weight, and smaller head circumference  —  and has been implicated in the heightened risk of metabolic and psychiatric disorders  — .
The long-term consequences of maternal stress during gestation for offspring neurobiological and physiological functioning have likewise been demonstrated in laboratory studies using rodents  —  , suggesting a causal link between this form of prenatal adversity and various developmental outcomes.
Though there may be multiple pathways through which maternal stress during pregnancy shapes offspring development, stress-induced elevations in maternal glucocorticoids have emerged as a primary mechanism of prenatal stress effects  , supported by studies in rats in which the effects of prenatal stress are prevented amongst adrenalectomized females .
In mammals, the placenta serves as a critical interface between maternal and fetal physiology and forms a barrier to maternal glucocorticoids  , . In mice, mutation of the HSD11B2 gene leads to hypertension, excess mineralocorticoid activity, and increased anxiety-like behavior in adulthood  whereas HSD11B1 mutation leads to attenuated negative-feedback of the HPA response to stress and improved cognitive performance in aging .
In humans, heightened maternal anxiety assessed just prior to parturition was found to be negatively correlated with placental HSD11B2 mRNA levels . Reduced placental HSD11B2 mRNA levels have also been found associated with intrauterine growth retardation and pre-term birth  , suggesting that the transcriptional activity of this enzyme may be predicted by maternal adversity and predictive of high risk birth outcomes.
Regulation of gene expression through epigenetic mechanisms — factors that alter gene transcription without altering DNA sequence - is being increasingly explored within the context of environmentally-induced changes in neurobiology, metabolism, and disease risk  — .
Epigenetic changes involving DNA methylation at cytosine-guanine sequences — CpG sites , post-translational modification of histone proteins, and microRNAs can have dynamic effects on gene expression and may also account for the stability of experience-dependent effects on transcription  — .
Epigenetic dysregulation has been associated with prenatal intrauterine growth restriction  and disease in both humans and rodents  — . Moreover, there is increasing evidence that maternal adversity during gestation can induce epigenetic changes in placenta and fetal tissues that may account for the heightened HPA reactivity amongst offspring . In the current study, we examined the impact of gestational maternal stress in pregnant rats on the transcription and DNA methylation of the HSD11B2 gene to determine whether epigenetic factors may account for the down-regulation of placental HSD11B2 function in response to stress.
An important consideration in studies of environmental-induced epigenetic effects for which there has been limited empirical investigation, is tissue specificity, and here we compared the impact of gestational maternal stress on CpG methylation in placenta as well as fetal hypothalamus and cortex.
Moreover, we explored the potential role of the DNA methyltransferases DNMT1 and DNMT3a — enzymes that promote DNA methylation  ,  — in these tissues, to determine the possible mechanistic pathways through which stress-induced epigenetic variation is achieved. Finally, we also assessed the feasibility of using placental epigenetic variation in the HSD11B2 gene to predict DNA methylation levels of this gene in the fetal hypothalamus and cortex.
This study provides new insight into the molecular basis of the effects of maternal adversity and highlights issues that are critical for the study of epigenetic effects and the translation of epigenetic analysis to studies of human prenatal exposures. Food and water were available ad libitum and replenished daily by animal care staff at Animals were habituated to the facility for 2 weeks prior to mating.
Pair-housed virgin females were mated for one week. Timing of the pregnancy was confirmed by the presence of a vaginal plug designated as gestational day 0. Within the sample of 16 mated females, 12 females became pregnant following the 1-week mating period. Control females were left undisturbed throughout gestational days 14— The timing of stress exposure was randomized to prevent habituation.
At gestational day 20, pregnant dams were sacrificed 1 hour after restraint stress through rapid decapitation and trunk blood was collected for assay of corticosterone levels. Plasma corticosterone was assayed using an RIA kit MP Biomedicals and this assay confirmed elevated levels of corticosterone in stressed compared to control dams [control: Feti and placenta were extracted via caesarean section at gestational Day Pups were decapitated and whole brains extracted.
Placenta samples were dissected such that a pie slice including both the basal zone and inner labyrinth zone was used. Subsequent analysis of gene expression and CpG methylation was conducted using tissue from 1—2 offspring each of 4 control and 4 stress dams. Selection was based on average pup weight at the time of sacrifice.
Only litters in which average pup weight was greater than 2. Tissue hypothalamus, cortex, placenta from 2 pups per litter from 4 control and 4 stress dams were included in the gene expression analyses. Primer pairs are included in Table 1. Relative expression was normalized to control non-stress cortex samples.
Tissue hypothalamus, cortex, placenta samples from 1 pup per litter from control and stress dams were run in duplicate and included in the DNA methylation analyses. Samples were bisulfite converted and PCR amplified see Table 2 for primers. The region was assayed by two sets of pyrosequencing primers see Table 2. Paired samples t-test indicated no significant differences between duplicates and average level of methylation across duplicates was used for all subsequent analyses.
The rat HSD11B2 gene contains a promoter and five exons. The promoter lowercase letters and first part of exon 1 capital letters are indicated and the ATG site is bolded. Lanes expressing a bright band were identified as male.
Multiple SRY loci have been identified in the rat genome  and some variability in amplified nucleotide length was observed around the expected fragment length of bp. Analysis of average ddCT levels to determine differences in gene expression was conducted with a 2-way ANOVA with tissue type and stress treatment as factors. Correlation between DNA methylation levels at specific CpG sites was conducted with two-tailed Pearson correlation coefficients. Though sex-specific effects on gene expression and DNA methylation were not observed in this sample, sex was used as a covariate in the analyses.
Based on the tissue-specific effects of stress on HSD11B2 gene expression, we proceeded to determine whether there were tissue specific effects of stress on the expression of genes involved in epigenetic regulation of gene expression DNMT1 and DNMT3a. There were no significant effects of stress on DNMT1 expression in the placenta. Analysis of the de novo methyltransferase DNMT3a , indicated a placenta-specific effect of stress on the expression of this enzyme.
Relative gene expression levels were determined by the 2ddCT method using cyclophilin-A and beta-actin as internal standards. Moreover, previous studies in rats report environmentally-induced differences in methylation within this region associated with variation in gene transcription . Analysis of within-individual variation in CpG methylation of the HSD11B2 promoter and at sites within the first exon — see Figure 1 generally indicated a lack of correlation in the methylation status of sites between different tissues.
At sites 1 and 3 site 1: At sites 10 and 14 site At all other CpG sites, there were no differences between tissue types in the average level of DNA methylation. Within the fetal hypothalamus Figure 3a , prenatal maternal stress was associated with reduced CpG methylation at sites 2, 3, 4, 6, 7, 8, 15, and 33 and elevated CpG methylation at sites 21 and Within the fetal cortex Figure 3b , there was no effect of prenatal maternal stress on CpG methylation.
In the placenta Figure 3c , prenatal maternal stress was associated with increased CpG methylation at sites 4, 5, 7, 8, and In the current study we report a robust and tissue-specific effect of maternal stress during pregnancy in rats on mRNA levels of the enzyme HSD11B2. Stress, anxiety, and depression during pregnancy can have a long-lasting impact on the psychological health of children  , .
Chronic unpredictable prenatal stress in rodents has been shown to impact cognitive and stress-coping behaviors, particularly when stress is experienced in the third gestational week . Consistent with previous findings, chronic stress was associated with decreased transcription of HSD11B2 within the placenta in late gestation  , . In contrast, our analysis indicates decreased DNA methylation at several sites of the HSD11B2 gene promoter in fetal hypothalamic tissue associated with stress and no effect of this prenatal exposure on DNA methylation in the fetal cortex.
These findings highlight the tissue specificity of epigenetic effects. Overall, these findings provide novel evidence for the epigenetic regulation of HSD11B2 as a potential mechanism linking maternal stress during gestation, dysregulation of placental gene expression, and neurodevelopmental outcomes in offspring.
Heightened glucocorticoid exposure during fetal development may promote lung development and thus increase survival following pre-term birth — an obstetric outcome predicted by maternal stress  , . Thus, though heightened anxiety and HPA activity may be considered maladaptive, these outcomes may be the cost of promoting survival amongst the offspring of gestationally stressed females.
Though we do not find an increase in hypothalamic or cortical HSD11B2 mRNA levels, the stress-induced DNA hypomethylation of this gene that we observe in hypothalamic tissue may be an epigenetic precursor to these buffering effects.
Time course analysis of HSD11B2 gene expression and DNA methylation throughout the prenatal stress exposure and into the postnatal period may thus be a key strategy for determining the dynamics of glucocorticoid programming mechanisms.
Investigation of the epigenetic regulation of HSD11B2 has previously been explored outside the context of studies on maternal stress. In humans and in rats, the promoter and first exon of the HSD11B2 gene regions analyzed in the current study are rich in CpG sites and DNA methylation levels at CpG sites within this region are related to the expression of this gene .
One functional consequence of elevated DNA methylation is the inhibition of transcription factor binding to the promoter regions of target genes and in the case of HSD11B2 , CpG methylation at Sp1 and NF1 recognition sequences prevents binding of these transcription factors and diminishes the transcriptional activity of HSD11B2 . These epigenetic effects were found to be sex-specific and present both at birth and postnatal day 21, indicating long-term consequences of prenatal HSD11B2 dysregulation.
The sex-specificity of environmentally induced changes in DNA methylation is increasingly evident and indeed epigenetic modifications may be involved in the normal process of sexual differentiation  ,  , .
Though we do not find sex-specific effects of maternal stress on HSD11B2 expression or DNA methylation in the current study, it may be that large samples are needed to detect sex-differences in the epigenetic regulation of this gene to account for individual differences in hormonal exposure or that sex-differences are more likely to emerge in later development.
A candidate mechanism for the increased placental DNA methylation and increased HSD11B2 levels we have found in response to maternal stress during pregnancy may involve the up-regulation of DNA methyltransferase levels in the placenta of stressed offspring. This stress-induced effect has implications for genome-wide epigenetic changes, and may account for the diverse phenotypic outcomes associated with maternal adversity during pregnancy.
DNMT1 and DNMT3a are enzymes active throughout the lifespan and thus could potentially serve as a mechanism for long-term epigenetic regulation, though it is possible that DNMT3b only active early in development is similarly altered by prenatal stress to mediate HSD11B2 promoter methylation.
Our data are consistent with previous findings indicating an up-regulation of DNMTs in the placenta of mice exposed to 1 st trimester maternal stress . Targeted deletion of DNMT1 or DNMT3a in mice has been found to produce embryonic and postnatal lethality and widespread epigenetic changes — particularly amongst imprinted genes — in embryonic tissues  — . Amongst offspring that are heterozygous for a mutation in DNA methyltransferase 3-like DNMT3L protein, there are specific disruptions to placental development .
These results illustrate the challenge of in vivo studies of epigenetic effects and the complexity of the pathways through which transcriptional activation is achieved. Determining the temporal dynamics of stress-induced changes in DNMTs, DNA methylation, and gene expression in different tissue types may provide some insight into this complexity.
For example, elevations in DNMT3a and DNMT1 will likely precede elevated methylation of target genes, and the reduced fetal CpG methylation observed in hypothalamic tissue of stressed offspring will likely precede observable changes in mRNA in this brain region. The association of DNMT3a with de novo methylation and of DNMT1 with maintaining methylation marks in dividing cells adds a further layer of temporal complexity. Though in vivo imaging of these changes within individuals is not methodologically feasible, a cross-sectional design including offspring of varying embryonic and postnatal ages would help elucidate these dynamics.
In addition, it is important to consider the diversity of epigenetic mechanisms that can contribute to maternal stress effects on gene regulation, particularly histone modifications, and to further explore other physiological factors that have been demonstrated to alter 1HSD11B2 activity and expression, such as catecholamines  and proinflammatory cytokines .
The translation of in vitro and animal studies of environmentally-induced epigenetic variation in target genes may provide critical insights into the role of these mechanisms in long-term disease risk. In humans, maternal antenatal depression has been found to predict elevations in DNA methylation of the glucocorticoid receptor gene NR3C1 promoter in fetal cord blood samples and the degree of CpG methylation detected in these cells predicts salivary cortisol levels of infants at 3 three months of age .
In both cases, these studies have used peripheral tissues as a proxy for potential epigenetic changes that may be induced within the brain and account for elevations in the HPA response to stress associated with these adverse prenatal exposures.