Study participants

Study participants included all singleton infants born prematurely (<37 weeks) to the cohort of women enrolled in the Newborn Epigenetics STudy (NEST); a prospective study of mother/infant dyads aimed at investigating the effects of in utero exposures on epigenetic profiles and phenotypes in children. The target population and methods for participant identification and enrollment have been described in detail previously. Briefly, between 2007 and 2011, pregnant women were recruited during their first or second trimester visits to prenatal clinics serving Duke Hospital or Durham Regional Hospital, the only 2 obstetric facilities in Durham County. Eligibility criteria were age 18 years and older, intention to use Duke Hospital or Durham Regional Hospital for delivery, and English and Spanish speaking. Subjects were excluded if they planned to give up custody of their children or had HIV because of the limited research on the interaction of HIV infection with potential DNA methylation alterations in the offspring. Of the 181 singleton preterm births in this cohort, infection, complete parturition, and methylation data were available in a subset of 73 mother-infant pairs. These mother-infant pairs are similar to the larger group of 181 with respect to maternal age ( P = .3284), race ( P = .638), maternal body mass index (BMI) ( P = .8156), type of preterm birth ( P = .106), and proportion of births with chorioamnionitis ( P = .620) and funisitis ( P = .619). The study protocol was approved by the Duke University Institutional Review Board.

Data collection

Demographic information was collected from mothers at enrollment. Gestational age at enrollment ranged from 5-36 weeks (mean, 17.5 weeks; standard deviation [SD], 9.0). Gestational age at enrollment was slightly higher in the group of 73 mother-infant pairs (mean, 20.8 weeks; SD, 9.1) as compared with the larger group of 181 (mean, 17.5 weeks; SD, 9.0), P = .01. Pregnant women self-reported birthdate (maternal age was computed by subtracting delivery date from birth date), current health status, parity, marital status, and usual weight and height at last menstrual period (LMP) for BMI. Because of changes in some questions in more recent demographic surveys, this study only characterizes women as “black” or “white”, and the remaining women (ones who chose “Asian,” “Native-American,” “Hispanic,” or “Other”) are considered “other.” Women who chose “white” or “caucasian” were classified as “white.” Women were classified as “black” if they identified themselves as “black/African-American” or if they identified as “biracial/other” and their mother was “black/African-American.” Smoking status was harmonized between questionnaires to 3 categories, none, smoking during pregnancy, and quitting during pregnancy, as previously described. Education level was harmonized to less than high school, high school or GED, college, and graduate school.

After delivery of the infant, trained personnel abstracted parturition data from medical records including gestational age at birth, infant sex, delivery mode, and birthweight. PTB was defined as gestational age <37 weeks, and LBW was defined as birthweight <2500 g. Trained abstractors examined the medical records of all preterm infants to determine the type of PTB. Medically indicated PTB was defined as emergent delivery because of a medical complication, intrauterine growth restriction (IUGR), or preeclampsia (defined for this study as medical order for magnesium and 24-hour urine protein level >300 mg). PPROM was defined by a clinical diagnosis (listed on problem list) or presence of 2 of 3 of the following positive tests (pool, fern, or nitrazine). All other PTBs were defined as spontaneous. When the medical record was unclear, an obstetrician (A.P.M.) determined the phenotype based on clinical expertise. Standard pathologic reports from the placenta of preterm infants were used to assess infection status. Chorioamnionitis was defined as a pathologic diagnosis of Stage II or III chorioamnionitis. Funisitis was also a standardized pathologic diagnosis. All data were collected and deidentified in STATA 12.0 (StataCorp, College Station, TX).

DNA methylation analysis

Infant cord blood specimens were collected at birth. Samples were collected in EDTA-containing vacutainer tubes and centrifuged using standard protocols to allow for collection of plasma and buffy coat, with buffy coat used for DNA extraction (Qiagen, Valencia, CA). Specimens were stored at −80°C until time of analysis. DNA was extracted using Puregene reagents according to the manufacturer’s protocol (Qiagen) and quantity and quality assessed using a Nanodrop 1000 Spectrophotometer (Thermo Scientific, Wilmington, DE). Infant genomic DNAs (800 ng) were modified by treatment with sodium bisulfite using the Zymo EZ DNA Methylation kit (Zymo Research, Irvine, CA). Bisulfite treatment of denatured DNA converts all unmethylated cytosines to uracils, leaving methylated cytosines unchanged, allowing for quantitative measurement of cytosine methylation status. Pyrosequencing was performed using a Pyromark Q96 MD pyrosequencer (Qiagen). The bisulfite pyrosequencing assays were used to quantitatively measure the level of methylation at CpG sites contained within 9 imprinted DMRs. DMRs analyzed were the paternally methylated IGF2 DMR, H19 DMR, MEG3-IG DMR (located intergenic to DLK1 and MEG3) and MEG3 DMR (promoter), and the maternally methylated PEG3 DMR, MEST DMR, SGCE/PEG10 DMR, NNAT DMR, and PLAGL1 DMR. For all DMRs except PLAGL1, pyrosequencing assay design, genomic coordinates, assay conditions and assay validation are described in detail elsewhere. Assays were designed to query established imprinted gene DMRs using the Pyromark Assay Design Software (Qiagen). For amplification of PLAGL1 from bisulfite modified DNAs, polymerase chain reaction (PCR) primers were forward, 5′-GTA GGG TAG GTG TTT GGG TGT T-3′ and reverse, 5′-[biotin]CRA CAA AAA CAC ACC CTC CTC-3′. PCR was performed as previously described using the following conditions: 95°C for 15 minutes, 55 cycles of 94°C for 30 seconds, 68°C for 30 seconds. and 72°C for 30 seconds, and a final 10-minute extension at 72°C. Pyrosequencing was performed using primer 5′-TGA GGT GTT TGG GTG TT-3′ to analyze sequence at genomic coordinates 144,329,179 to 144,329,209 on chromosome 6 (UCSC Genome Browser, Feb. 2009 release, GRCh37/hg19). PCR conditions were optimized to produce a single, robust amplification product. Defined mixtures of fully methylated and unmethylated control DNAs were used to show a linear increase in detection of methylation values as the level of input DNA methylation increased (Pearson r > 0.98 for all DMRs). Once optimal conditions were defined, each DMR was analyzed using the same amount of input DNA from each specimen (40 ng, assuming complete recovery after bisulfite modification). Percent methylation for each CpG cytosine was determined using Pyro Q-CpG Software (Qiagen).

Statistical analysis

Fisher exact tests were used to determine associations between type of PTB and infection status (chorioamnionitis and funisitis). To examine the role of epigenetics, we then assessed each DMR for normality using the Kolmogorov-Smirnov test. We found that with the exception of SGCE/PEG10 ( P < .01), all other infant DMRs were normally distributed ( P > .05). Confirmatory factor analysis for individual maternal and infant CpGs revealed Cronbach’s alphas for all DMRs were > .74, suggesting mean methylation levels for each DMR could be used in models. One-way analyses of variance (ANOVA) were used to compare infant DNA methylation differences at the 9 DMRs by type of PTB (medically indicated, PPROM, or spontaneous PTL). t tests were used to compare infant DNA methylation differences at DMRs by infection status (chorioamnionitis or funisitis). Wilcoxon-rank sum tests were used to confirm differences found by t testing in all DMRs that were not normally distributed. For DMRs that differed by type of PTB or infection status, logistic regression models were fit to examine associations between PTB or infection status and DNA methylation at DMRS. Models were initially fit with all variables considered clinically relevant including maternal age at delivery, maternal race, health status, parity, marital status, maternal BMI, smoking, education level, delivery mode, infant sex, infant birthweight, and gestational age at delivery. A backward stepwise approach was used to refine the model, and log likelihood tests for individual covariates were used to create the final parsimonious model.

Demographic characteristics

Table 1 describes the characteristics of the 181 preterm mother-infant dyads. Maternal age ranged from 18–49, with mean age of 30 years and SD of 6.6 years. Thirty-eight percent of women were married, 31% were never married, and the rest were either living with a partner, divorced/separated or chose “other.” Fifty-four percent of the mothers were black, 28% were white, and 18% were grouped as other, which included women choosing a race/ethnicity of Hispanic, Asian, Native American, or other. The majority of the women were multiparous (68%). BMI ranged from 16–67 with a mean BMI of 28, SD of 7.7. The majority of the women endorsed good, very good, or excellent health with 17% endorsing fair or poor health. 20% of women smoked during pregnancy, whereas 17% reported quitting. The majority (>50%) of women had some college education or more. Fifty-one percent of the births were delivered vaginally, and 49% were delivered via cesarean section. Fifty-six percent of the infants delivered were male, and 44% were female. Of the PTBs, 37% were medically indicated, 24% were a result of PPROM, and 39% were spontaneous. Birthweight ranged from 580–3765 g with a mean birthweight of 2346 g (SD, 677 g). Gestational age at birth ranged from 24–36 weeks with a mean of 33.9 weeks (SD, 2.9 weeks). Sixteen percent of the PTB infants had chorioamnionitis, and 12% had funisitis. Of the 22 cases of chorioamnionitis and 16 cases of funisitis, 13 cases overlapped and had pathologic features of both.

Types of PTB and infection

When examining type of preterm birth by infection status, more cases of chorioamnionitis occurred in preterm births that were the result of PPROM (n = 14) or spontaneous PTL (n = 8) when compared with medically indicated PTB (n = 0), Fisher exact P value < .001, which was expected based on previous epidemiologic studies. A similar phenomenon was observed in those infants with funisitis with more cases in preterm births that occurred as a result of PPROM (n = 8) or spontaneous PTL (n = 7) when compared with medically indicated PTB (n = 1), Fisher exact P value = .003. In addition, chorioamnionitis and funisitis were highly correlated (correlation coefficient, 0.6449; P < .001).

Type of PTB, infection, and DNA methylation at imprint regulatory regions

When examining the role of epigenetics in our subset of 73 preterm infants with completed methylation data, infant DNA methylation levels did not differ by type of PTB (medically indicated, PPROM, and spontaneous PTL) at any of the 9 DMRs examined, P > .20 ( Figure 1 ). We found that infants born with pathologic evidence of chorioamnionitis had higher methylation levels at the DMRs regulating the imprinted genes PLAGL1 and PEG3 ( Table 2 , P < .05). We also found that infants born with pathologic evidence of funisitis had higher methylation levels at the DMR regulating the imprinted gene PLAGL1 ( Table 2 , P < .05). After Bonferroni correction for multiple testing, differences at PLAGL1 by chorioamnionitis persisted ( P = .0026). In logistic regression models ( Table 3 ), this association persisted, and increasing levels of methylation at PLAGL1 was associated with increased risk for chorioamnionitis, after adjusting for the significant covariates maternal BMI and gestational age at birth (odds ratio [OR], 1.22; 95% confidence interval [CI], 1.04–1.44; P = .015). Decreasing gestational age at birth and increasing maternal BMI were also associated with increased risk of chorioamnionitis. Infant birthweight and gestational age at birth were highly correlated (correlation coefficient, 0.7661; P < .0001). As a result, although both were associated with chorioamnionitis, only gestational age at birth was included in overall models. After adjusting for significant covariates maternal age and infant sex, increasing levels of methylation at PLAGL1 was associated with increased risk for funisitis (OR, 1.17; 95% CI, 1.03–1.32; P = .016). Increasing maternal age and male gender were also associated with increased risk for funisitis. Gestational age at birth did not differ for funisitis ( P = .380), and it was not included in overall models. Figure 2 shows the distribution of DNA methylation levels at PLAGL1 by infection status (chorioamnionitis or funisitis).

Distribution of DMRs by type of PTB

Median and interquartile range of infant DNA methylation levels at nine DMRs by type of PTB. There are no significant differences between mean methylation levels at any of the 9 DMRs by type of PTB, P > .20. Reference line represents 50% methylation.

Only gold members can continue reading. Log In or Register to continue

Share this:

• Click to share on Twitter (Opens in new window)
• Click to share on Facebook (Opens in new window)

Related

Related posts:

1. Prenatal treatment of congenital adrenal hyperplasia: risks outweigh benefits
2. Correction
3. Trends in the rate of invasive procedures after the addition of the intrauterine tamponade test to a protocol for management of severe postpartum hemorrhage
4. Modulation of lipopolysaccharide-induced chorioamnionitis by Ureaplasma parvumin sheep

Stay updated, free articles. Join our Telegram channel

Join