Uteroplacental insufficiency leads to intrauterine growth retardation (IUGR) and increases the risk of insulin resistance and hypertriglyceridemia in both humans and rats. Postnatal changes in hepatic gene expression characterize the postnatal IUGR rat, despite the transient nature of the initial in utero insult. Phenomena such as DNA methylation and histone acetylation can induce a relatively static reprogramming of gene transcription by altering chromatin infrastructure. We therefore hypothesized that uteroplacental insufficiency persistently affects DNA methylation and histone acetylation in the IUGR rat liver. IUGR rat pups were created by inducing uteroplacental insufficiency through bilateral uterine artery ligation of the pregnant dam on day 19 of gestation. The SssI methyltransferase assay and two-dimensional thin-layer chromatography demonstrated genome-wide DNA hypomethylation in postnatal IUGR liver. To investigate a possible mechanism for this hypomethylation, levels of hepatic metabolites and enzyme mRNAs involved in one-carbon metabolism were measured using HPLC with coulometric electrochemical detection and real-time RT-PCR, respectively. Uteroplacental insufficiency increased IUGR levels of S-adenosylhomocysteine, homocysteine, and methionine in association with decreased mRNA levels of methionine adenosyltransferase and cystathionine-β-synthase. Western blotting further demonstrated that increased quantities of acetylated histone H3 also characterized the IUGR liver. Increased hepatic levels of S-adenosylhomocysteine can promote DNA hypomethylation, which is often associated with histone hyperacetylation. We speculate that the altered intrauterine milieu associated with uteroplacental insufficiency affects hepatic one-carbon metabolism and subsequent DNA methylation, which thereby alters chromatin dynamics and leads to persistent changes in hepatic gene expression.
- intrauterine growth retardation
- one-carbon metabolism
- Barker’s fetal origins of adult disease hypothesis
barker’s “fetal origins of adult disease hypothesis” proposes that fetal adaptation to a deprived intrauterine milieu leads to permanent changes in cellular biology and systemic physiology. Intrauterine growth retardation (IUGR) predisposes affected newborns toward long-term morbidity from type 2 diabetes, as well as other components of the metabolic syndrome (2). Uteroplacental insufficiency, a morbidity associated with many common complications of pregnancy such as pre-eclampsia, induces low ponderal index or asymmetrical IUGR. In the rat, uteroplacental insufficiency induced through bilateral uterine artery ligation of the pregnant dam also results in asymmetrical IUGR and, similar to the human, causes fetal hypoinsulinemia, hypoglycemia, acidosis, and hypoxia (5, 14, 16, 39, 40). Juvenile IUGR rats demonstrate insulin resistance, and adult animals suffer overt diabetes with fasting hypertriglyceridemia, hyperglycemia, and hyperinsulinemia (30, 52, 58). Interestingly, although the initial insult is transient and the metabolic-endocrinological milieu of these animals is in flux throughout their life, these IUGR animals are characterized by postnatal changes in hepatic gene expression (29, 30). This suggests a process that induces a relative stable in transcriptional regulation, such as DNA methylation or histone acetylation, which induces a relatively stable change in gene expression without altering DNA sequence (63).
Methylation of the C-5 position of cytosine occurs in 60–90% of CpG dinucleotides within the vertebrate genome (63). CpG-rich regions of DNA reside in 60% of promoters utilized by human RNA polymerase II and are often found in association with housekeeping and tissue-specific genes. DNA methylation inversely correlates with histone acetylation. Histone acetylation alters the positioning of histone-DNA contacts and the affinity with which these histones bind to DNA (26). These covalent changes in DNA and histone structure affect the extent to which transcription machinery is able to access specific regions of DNA over extended periods of time (26).
Considering the ubiquitous nature of DNA methylation and histone acetylation in genome control pathways, these phenomena are possible mechanisms through which uteroplacental insufficiency could initiate a “metabolic imprint” by altering chromatin structure and contribute to the pathogenesis of the postnatal complications such as diabetes and dyslipidemia, which are associated with IUGR (59). We therefore hypothesized that uteroplacental insufficiency alters DNA methylation and histone acetylation in IUGR rats. To prove this hypothesis, control and IUGR rats were studied at day 0 and day 21 of life, prior to the onset of overt and severe diabetes. DNA methylation patterns were compared using methylation-sensitive arbitrarily primed PCR (MSAP-PCR), and genome-wide DNA methylation was determined using the SssI methyltransferase assay and two-dimensional thin-layer chromatography (2D-TLC).
mRNA levels of DNA methyltransferase 1 (DNMT1) were measured using real-time RT-PCR. DNMT1 is the most fully characterized DNA methyltransferase and performs both maintenance and de novo methylation (19). Acetylated histone H3 and acetylated histone H4 quantities were determined using Western blotting. (3).
To investigate a possible mechanism through which uteroplacental insufficiency alters DNA methylation, we measured levels of hepatic metabolites and enzyme RNAs that are involved in one-carbon metabolism (Fig. 1). The most relevant of these metabolites are S-adenosylhomocysteine (SAH) and S-adenosylmethionine (SAM), which play a role in the regulation of genome-wide DNA methylation (51). Enzyme selection was based upon their relevance to the physiological context of IUGR. Hypoxia significantly reduces methionine adenosyltransferase (MAT) expression, and both glucose and insulin affect cystathionine-β-synthase (CBS) and methylenetetrahydrofolate reductase (MTHFR) activity (Fig. 1) (7) (13).
All procedures were approved by the UCLA Chancellor’s Animal Research Committee. These surgical methods have been previous described (30, 52, 58). This is a well-characterized rat model of uteroplacental insufficiency and IUGR that is induced by ligating both uterine arteries of the pregnant rat. Fetal and neonatal rats in this model are significantly lighter than controls that undergo identical anesthesia and sham surgery, and litter size does not differ between control and IUGR groups (39). Like the human, the IUGR rat fetus is characterized by hypoxia, acidosis, hypoglycemia, and hypoinsulinemia (39). These values normalize in the perinatal period. At day 21 of life, both male and female IUGR rats weigh significantly less than sham-operated control counterparts (28).
In brief, on day 19 of gestation, the maternal rats were anesthetized with intraperitoneal xylazine (8 mg/kg) and ketamine (40 mg/kg), and both uterine arteries were ligated (IUGR) (n = 8 litters). Sham surgery was performed upon control animals that underwent identical anesthetic and surgical procedures except for the uterine artery ligation (control) (n = 8 litters). Day 0 pups were delivered by caesarian section (n = 4 litters control and IUGR, respectively). The remaining maternal rats were allowed to deliver spontaneously, and litters were randomly culled to 6. Day 21 animals were separated from their dams (to minimize individual hormonal variations associated with feeding), anesthetized, and killed (n = 4 litters control and IUGR, respectively). Liver was harvested and frozen in liquid nitrogen. Both male and female rats were included in the study in equal numbers.
DNA was isolated using standard procedures and was suspended in TE buffer after dialysis. The DNA concentration was determined by reading the absorbance at 260 nm.
Two micrograms control and IUGR liver genomic DNA was digested with either 20 U of RsaI or with 20 U of RsaI and HpaII, respectively, in 40-μl reaction volume (New England Biolabs, Beverly MA) (33). Two microliters of the reaction mixture was amplified with primers listed in Table 1. The PCR buffer includes 10 mM Tris·HCl (pH 8.3), 1.5 mM MgCl2, 50 mM KCl, 0.1% gelatin, 200 μmol dNTP, and 1 U of Taq polymerase. PCR cycling conditions depend upon the length of the primer. The reaction mixture was diluted with 5 μl of loading buffer, denatured at 90°C for 3 min, and cooled on ice. Then, 2.5 μl of each sample was loaded and separated on a 5% polyacrylamide sequencing gel. The gel was then dried, exposed to autoradiographic film, and analyzed (Amersham, Cleveland, OH).
SssI methyltransferase assay.
The SssI methyltransferase assay of Balaghi and Wagner has been previously reported (46). A 30-μl reaction volume was used to incubate 0.5 μg of genomic DNA with 3 μmol/l (74 kBq) [3H-methyl]methionine (NEN, Boston, MA), 3 μl of 10 × reaction buffer, and 3 U of SssI methyltransferase (New England Biolabs, Beverly, MA). The reaction was incubated at 30°C for 60 min, and then 15 μl of the reaction mixture was spotted onto DE81 paper circles (Whatman, Ann Arbor, MI). The paper circles were washed five times in 0.5 mol/l acidic phosphate buffer (pH 6.8) and dried in air. Radioactivity was determined by liquid scintillation counting. Blank values were determined from reactions without SssI methyltransferase enzyme.
Two-dimensional thin-layer chromatography.
A modified method of Bestor et al. (4) was used to quantify levels of cytosine (C) and methylcytosine (mC) in control and IUGR DNA. Genomic DNA was digested to completion with methylation-sensitive enzymes MspI and TaqI (New England Biolabs) and then labeled with [γ-32P]ATP (PerkinElmer) using T4-polynucleotide kinase (Promega) according to manufacturer instructions. DNA was precipitated and then completely digested with nuclease P1 (USB) at 50°C for 2 h. The samples were dried and resuspended in 10 μl first-dimension chromatography solvent. Samples of 1 μl were separated two-dimensionally on 20 × 20-cm cellulose chromatography plates (EM Science) with 66:33:1 of isobutyric acid:water:ammonium hydroxide in the first dimension and 70:15:15 of 2-propanol:water:hydrochloric acid in the second dimension. Plates were allowed to air dry after each separation and then exposed to autoradiography film. Levels of C and mC were quantified using densitometry, and the percent of mC per the total amount of cytosine was calculated.
One-carbon metabolite measurement.
Measurement of intracellular SAM, SAH, homocysteine, methionine, cysteine, and adenosine was accomplished utilizing HPLC with coulometric electrochemical detection as previously described in detail (35). Briefly, samples of frozen liver (10–15 mg wet wt) were homogenized with 200 μl of phosphate-buffered saline. To reduce sulfhydryl bonds, 50 μl freshly prepared 1.43 M sodium borohydride solution containing 1.5 μM EDTA, 66 mM NaOH, and 10 μl isoamyl alcohol was added. To precipitate proteins, 250 μl ice-cold 10% meta-phosphoric acid was added, mixed well, and the sample was incubated for 30 min on ice. After centrifugation at 18,000 g for 15 min at 4°C, the supernatant was filtered through a 0.2-μm nylon membrane filter (PGC Scientific, Frederick, MD). Metabolite elution was performed using HPLC with a Shimadzu solvent delivery system (ESA model 580) and a reverse-phase C18 column (5 μm; 4.6 × 150 mm; MCM, Tokyo, Japan) obtained from ESA (Chemlsford, MA). A 20-μl aliquot of plasma extract was directly injected onto the column using Beckman Autosampler (model 507E). All plasma metabolites were quantified using a model 5200A Coulochem II and CoulArray electrochemical detection systems (ESA) equipped with a dual analytical cell (model 5010), a four-channel analytical cell (model 6210), and a guard cell (model 5020). The concentrations of plasma metabolites were calculated from peak areas and standard calibration curves using HPLC software.
Total RNA was extracted from liver and quantified in triplicate using UV absorbance (10). Gel electrophoresis confirmed the integrity of the samples. RNA was treated to DNase (Ambion, Austin, TX).
Real-time RT-PCR with TaqMan probe.
Hepatic mRNA levels of DNMT1, MAT, MTHFR, and CBS were measured at day 0 and day 21. cDNA was synthesized using random hexamers and SuperScript II RT (Life Technologies, Gaithersburg, MD) from 0.05 μg of hepatic RNA. DNMT1, MAT, CBS, and MTHFR primers and probes were designed using Primer Express Software (Applied Biosystems, Foster City, CA) (Table 2); target probes were labeled with fluorescent reporter dye FAM. Prior to the performance of real-time PCR, all primer pairs are tested with serial Mg2+ and primer concentrations to determine the optimal reaction conditions and to demonstrate the specificity of each primer pair. Reporter dye emission is detected by an automated sequence detector combined with ABI Prism 7700 Sequence Detection System software (Applied Biosystems). An algorithm normalizes the reporter signal (Rn) to a passive reference and multiples the standard deviation of the background Rn in the first cycles by a default factor of 10 to determine threshold CT. CT has a linear relation with the logarithm of the initial template copy number (25). Real-time PCR quantification is then performed using the TaqMan glyceraldehyde-3-phosphate dehydrogenase (GAPDH) controls. Prior to the use of GAPDH as a control, serial dilutions of cDNA are quantified to prove the validity of using GAPDH as an internal control. Relative quantification of PCR products are then based upon value differences between the target and GAPDH control using the comparative CT method (36). Cycle parameters were 55°C for 5 min, 95°C for 10 min, and then 40 cycles of 95°C for 15 s to 60°C for 60 s. Each sample was analyzed in triplicate in assays performed on three occasions.
Liver was homogenized in lysis buffer containing 100 mM sodium butyrate, 0.2 M HCl, 10 mM HEPES (pH 7.9), 1.5 mM MgCl2, 10 mM KCl, 0.5 mM DTT, and 1.5 mM PMSF. After centrifugation, the supernatant was dialyzed in 0.1 M acetic acid twice for 2 h each time, then further dialyzed three times for 1 h, 3 h, and overnight, respectively. The protein samples were then separated on a bisacrylamide gel and transferred to nitrocellulose. After blocking for 30 min with PBS-milk (3%), the blot was incubated with 10 ng/ml of anti-acetyl H3 or H4 (Upstate Biotechnology, Lake Placid, NY) that had been diluted in PBS-milk (3%). After rinsing, the blot was further incubated with the secondary antibody (1:2,000) for 1.5 h at room temperature. The filters were washed, and detection was performed using ECL chemiluminescence (Amersham, Piscataway, NJ). The products were quantified by densitometry after standardization for loading. Each blot was replicated three times.
All data presented are expressed as mean percent of control ± SE. The SssI DNA methyltransferase statistical analyses were performed using the nonparametric Wilcoxon matched pair test. For real-time RT-PCR, Western blotting, HPLC, and 2D-TLC, statistical analyses were performed using ANOVA (Fisher’s protected least significance difference) and Student’s unpaired t-test.
Day 0 control and IUGR MSAP-PCR.
MSAP-PCR was performed to determine whether banding patterns differ between genomic DNA from control and IUGR day 0 liver. MSAP-PCR uses primers designed to amplify CpG islands and takes advantage of HpaII’s inability to cut the 5′…CCGG …3′ sequence when the inner cytosine is methylated. We found replicable differences between control and IUGR tissues in liver and skeletal muscle, respectively (n = 4 litters) (Fig. 2A).
Day 0 SssI DNA methyltransferase assay.
The SssI DNA methyltransferase assay uses the universal methyl donor SAM to transfer a tritium-labeled methyl group to unmethylated cytosines in CpG sites. As a result, higher counts demonstrate less genomic DNA methylation. Counts were significantly higher in genomic DNA from day 0 IUGR liver (17,034 ± 1,210, P < 0.05) vs. those from day 0 control liver (12,021 ± 1,380 cpm), which demonstrates a decrease in IUGR DNA methylation (n = 4 litters).
Day 0 and day 21 2D-TLC.
2D-TLC was performed on both day 0 and day 21 hepatic DNA from control and IUGR samples to measure genome-wide methylation in the target site 5′…CCGG…3′. For day 0 IUGR hepatic genomic DNA, the amount of cytosine methylation was significantly decreased to 58.3 ± 4% (P < 0.05) of the total labeled cytosine, vs. 70. 2 ± 5% for control DNA (n = 4 litters). Similarly, for day 21 IUGR hepatic genomic DNA, the amount of cytosine methylation was also significantly decreased to 61.6 ± 4% (P < 0.05) of the total labeled cytosine vs. 75.3 ± 7% for control DNA (n = 4 litters) (Fig. 2B).
DNMT1 mRNA levels.
DNMT1 mRNA levels were measured in day 0 and day 21 liver control and IUGR livers using real-time PCR (n = 4 litters). At day 0 of life, hepatic IUGR mRNA levels of DNMT1 were significantly decreased to 33 ± 9% of control values; in contrast, there was no significant difference between control and IUGR DNMT mRNA levels at day 21 of life (Fig. 3A).
Day 0 one-carbon metabolite and enzyme levels.
Measurement of intracellular SAM, SAH, homocysteine, methionine, cysteine, and adenosine was accomplished utilizing HPLC with coulometric electrochemical detection. Uteroplacental insufficiency significantly increased levels of SAH, adenosine, homocysteine, and methionine in day 0 IUGR liver (Table 3). In contrast, IUGR significantly decreased hepatic cysteine levels, without affecting methionine (Table 3). Real-time RT-PCR quantified mRNA levels of CBS, MTHFR, MAT1A, and MAT2A in day 0 control and IUGR liver. Uteroplacental insufficiency significantly decreased mRNA levels of CBS, MAT1A, and MAT2A to 61 ± 7% (P < 0.05), 38 ± 4% (P < 0.01), and 75 ± 6% (P < 0.05) of sham-operated control values, respectively (Fig. 3B). mRNA levels of MTHFR were not significantly altered by IUGR.
Day 0 and day 21 acetylated histone H3 and acetylated histone H4 levels.
Western blotting was performed on both day 0 and day 21 protein from control and IUGR samples. For acetylated histone H4 protein levels, there was no significant difference between control and IUGR samples at either day 0 or day 21 (n = 4 litters) (Fig. 4A). In contrast, IUGR day 0 acetylated histone H3 levels were significantly increased to 180 ± 22% of control values, and IUGR day 21 acetylated histone H3 levels were significantly increased to 233 ± 37% (P < 0.05) (n = 4 litters) (Fig. 4B).
This study demonstrates that the altered intrauterine milieu associated with uteroplacental placental insufficiency affects DNA methylation and histone H3 acetylation; moreover, this occurs within the context of elevated SAH and altered mRNA levels of key one-carbon metabolizing enzymes such as CBS and MAT. This suggests a series of in utero molecular mechanisms that may contribute to the pathogenesis of adult morbidities such as diabetes and dyslipidemia, which are associated with perturbations of the intrauterine milieu such as uteroplacental insufficiency.
Uteroplacental insufficiency is a significant cause of human morbidity and mortality that is associated with many common complications of pregnancy such as pre-eclampsia and the other hypertensive disorders of pregnancy. These disorders are a concern in both western and underdeveloped nations. Uteroplacental insufficiency causes hypoglycemia, hypoinsulinemia, acidosis, hypoxia, and decreased levels of the branched chain amino acids (15). This abnormal intrauterine milieu causes IUGR. For most IUGR infants, brain growth is spared relative to the rest of the body, and they are categorized as being “asymmetrically growth retarded.” As adults, asymmetrically growth retarded humans are at risk for multiple morbidities including the metabolic syndrome, which consists of dyslipidemia, insulin resistance, and cardiovascular disease (2). Similarly, bilateral uterine artery ligation of the pregnant rat results in asymmetrical IUGR rat pups (41).
Like the human, the IUGR rat fetus is characterized by hypoxia, acidosis, altered insulin-like growth factor (IGF) availability, decreased levels of branch chain amino acids, hypoglycemia, and hypoinsulinemia (39). These conditions alter the hepatic redox state of the IUGR fetus and lead to oxidative stress (41). Previous studies have demonstrated that oxidative stress, as measured by glutathione depletion, and free radical production alters DNA methylation in vitro and causes hepatic DNA hypomethylation in adult animals (31, 61). Similarly, after administration of 10% oxygen for 10 days, hepatic DNA methylation decreases in young adult rats (9). In contrast, Rees et al. (46) found that maternal protein deprivation is associated with hepatic DNA hypermethylation using the DNA SssI methyltransferase assay, and it was speculated that l-threonine deficiency induced changes in methionine metabolism, increased homocysteine production, and thereby led to changes in fetal DNA methylation.
In contrast, our findings of elevated SAH and DNA hypomethylation within the context of decreased CBS mRNA levels resemble those of Caudill et al. (8), who found that an elevation of SAH tissue levels consistently lead to decreased DNA methylation in CBS-deficient mice fed methyl-deficient diets, regardless of whether SAM tissue levels were unchanged or decreased.
Our findings are also consistent with previous reports of CBS activity in the fetal liver, as well as the direct relationship between CBS activity and glucose levels (13, 49). We speculate that alterations in CBS gene expression and activity contribute to the changes in one-carbon metabolism and thereby DNA methylation observed in the uteroplacental insufficiency model of IUGR.
Because of the heterogeneity of the human condition, both models of IUGR are relevant, although they differ in two key aspects. l-Threonine deficiency and normoxia characterize the protein malnutrition model; branched chain amino acid deficiency and hypoxia characterize the uteroplacental insufficiency model of IUGR (39, 47). The latter difference may be particularly important because in utero oxidative stress is linked to IUGR and insulin resistance in the human (44).
At least three different DNA methyltransferases contribute to DNA methylation patterns, and controversy exists over their regulation and specific functions (4). DNMT1 performs both maintenance and de novo methylation, and it is the most fully characterized (4, 19). De novo activity of DNMT1 is stimulated by binding of methylated DNA to the NH2-terminal of the enzyme and by the presence of DNMT3a (19, 20). DNMT3a and DNMT3b are necessary for de novo methylation in embryonic stem cells and early embryos (42). This study quantified DNMT1 mRNA levels because targeted mutations of the DNMT1 are recessive lethal, and DNMT1 represents the majority of methyltransferase activity in embryo lysates (32, 65). DNMT1 mRNA levels also correspond with methyltransferase activity level, and similarly, methyltransferase activity positively correlates with DNA methylation (54, 55). Decreased levels of hepatic IUGR DNMT1 mRNA thereby support our finding of relative hypomethylation in the IUGR fetal liver. We speculate that hepatic DNA hypomethylation in the postnatal day 21 rat persists because of the initial insult and does not reflect further changes in DNMT1 gene expression. Indeed, hepatic DNMT1 mRNA levels are comparable between control and IUGR animals at day 21 of life.
The increase in acetylated histone H3 in IUGR liver is novel and further supports the findings of IUGR fetal liver hypomethylation, because DNA hypomethylation is associated histone hyperacetylation (26). Differential acetylation between histone H3 and H4 has been previously found globally and in association with specific genes (11, 22). The relationship between DNA methylation and histone acetylation may occur through interactions between DNMT1 and the histone deacetylase (6). Another contributing factor to increased acetylated histone H3 may be hypoxia. Two other groups have found that hypoxia stimulates histone acetylation in tissues other than liver (43, 57).
Hypoxia and free radical production, characteristics of the IUGR fetal liver, also decrease MAT activity and gene expression (1, 12, 41). Two genes produce MAT: MAT1A and MAT2A (27). The adult hepatocyte expresses primarily MAT1A, which results in the production of the MATI/III enzyme; the fetal hepatocyte expresses MAT2A and subsequently produces the MATIII enzyme. Reduced expression of these enzymes likely contribute to the unremarkable change of in IUGR liver SAM levels, despite the increase in methionine levels.
Considering the known association between IUGR and adult cardiovascular morbidity, it is intriguing that uteroplacental insufficiency increases liver homocysteine levels. Human epidemiological studies suggest that increased levels of plasma homocysteine increase the risk for atherosclerotic vascular disease, and rats demonstrate a direct correlation between liver homocysteine and plasma homocysteine levels (23, 34, 37).
Together, the DNA hypomethylation and histone hyperacetylation represent significant alterations in chromatin infrastructure and provide an attractive mechanism for establishing and maintaining a stable state of transcriptional regulation. The overall implication of our findings is placed into context by Van de Vijver et al. (60), who made the following statement in an elegantly written review on epigenetics: “There is, however, a growing list of facts that no longer fit the linear, one-gene-only approach of the genome. This shows that it is no longer sufficient to restrict research to classical genetic analyses in term of genetic mutations, distinct phenotype-genotype distinctions, and metaphors of genetic programs. From this angle, multiple contexts, such as nuclear and intracellular contexts that directly impinge upon the genome…can no longer be neglected.”
A recent review by Ehrlich (17) reveals several instances in which DNA methylation regulates the expression of several genes during development, and this regulation ranges from across the continuum from major change to fine tuning of expression. Specific implications of our findings include the direct relationship between DNA hypomethylation and aging, particularly within the context of the “fetal origins of adult disease hypothesis” (50, 53, 62). Moreover, as summarized by Maier and Olek (33a), considerable indirect evidence exists that aberrant DNA methylation plays a role in the development of insulin resistance, not the least of which is finding of increased SAH levels in diabetes.
Our finding that DNA hypomethylation occurs in response to uteroplacental insufficiency demonstrates the vulnerability of the fetus to in utero perturbations and suggests folate as a key nutritional substrate in our scenario. Folate deficiency correlates with IUGR in human epidemiological studies, and folate supplementation favorably affects birth weight (21, 48, 56). Moreover, folate deficiency increases hepatic homocysteine and SAH levels, causes DNA hypomethylation, and affects the expression of genes such as MeCP2 (18, 24) (64).
We speculate that decreased levels of substrate in the intrauterine milieu (1: decreased folate leading to increased SAH levels; 2: hypoglycemia leading to decreased CBS and subsequent increased SAH levels, as well as oxidative stress; and 3: hypoxia leading to oxidative stress) induce DNA hypomethylation and histone hyperacetylation in this model of uteroplacental insufficiency and adult onset diabetes. Although this is a model of IUGR, the paradigm of an aberrant intrauterine environment affecting chromatin structure and thereby altering the functional genome’s response to latter environments is potentially applicable to situations such as gestational diabetes and maternal malnutrition, even though the specific effects may not be identical (Fig. 5). Caution is necessary of course when attempting to apply data from a rat model to human pathophysiology. The fetal and juvenile rat is physiologically immature relative to the human, and the insult imposed on the fetal rat in this model of uteroplacental insufficiency is severe and specific. In contrast, the timing and impact of uteroplacental insufficiency experienced by humans range across a continuum.
In summary, by identifying alterations in DNA methylation and acetylation, we introduce the concept that in utero perturbations may affect chromatin organization and thereby suggest a mechanism for establishing a relatively stable state of altered transcriptional regulation whose response to subsequent environmental stimuli lay the groundwork for future postnatal morbidities.
This research was supported by National Institute of Child Health and Human Development Grants K08-HD-01225, R01-HD-41075, the March of Dimes, and the American Diabetes Association through an ADA Research Grant.
Article published online before print. See web site for date of publication (http://physiolgenomics.physiology.org).
Address for reprint requests and other correspondence: R. H. Lane, Univ. of Utah, Division of Neonatology, 30 North 1900 East, Rm. 2A100, Salt Lake City, UT 84124-2202 (E-mail:).
- Copyright © 2004 the American Physiological Society