Skip to main content
PMC home page
Nature Portfolio logoLink to Nature Portfolio
. 2013 Apr 23;37(7):907–913. doi: 10.1038/ijo.2013.47

Associations between antibiotic exposure during pregnancy, birth weight and aberrant methylation at imprinted genes among offspring

A C Vidal 1,2, S K Murphy 3, A P Murtha 4, J M Schildkraut 2,5, A Soubry 2, Z Huang 3, S E B Neelon 5,6, B Fuemmeler 5, E Iversen 7, F Wang 2,5, J Kurtzberg 8, R L Jirtle 8, C Hoyo 1,*
PMCID: PMC3705584  PMID: 23609933

Abstract

Objectives:

Low birth weight (LBW) has been associated with common adult-onset chronic diseases, including obesity, cardiovascular disease, type II diabetes and some cancers. The etiology of LBW is multi-factorial. However, recent evidence suggests exposure to antibiotics may also increase the risk of LBW. The mechanisms underlying this association are unknown, although epigenetic mechanisms are hypothesized. In this study, we evaluated the association between maternal antibiotic use and LBW and examined the potential role of altered DNA methylation that controls growth regulatory imprinted genes in these associations.

Methods:

Between 2009–2011, 397 pregnant women were enrolled and followed until delivery. Prenatal antibiotic use was ascertained through maternal self-report. Imprinted genes methylation levels were measured at differentially methylated regions (DMRs) using bisulfite pyrosequencing. Generalized linear models were used to examine associations among antibiotic use, birth weight and DMR methylation fractions.

Results:

After adjusting for infant gender, race/ethnicity, maternal body mass index, delivery route, gestational weight gain, gestational age at delivery, folic acid intake, physical activity, maternal smoking and parity, antibiotic use during pregnancy was associated with 138 g lower birth weight compared with non-antibiotic use (β-coefficient=−132.99, s.e.=50.70, P=0.008). These associations were strongest in newborns of women who reported antibiotic use other than penicillins (β-coefficient=−135.57, s.e.=57.38, P=0.02). Methylation at five DMRs, IGF2 (P=0.05), H19 (P=0.15), PLAGL1 (P=0.01), MEG3 (P=0.006) and PEG3 (P=0.08), was associated with maternal antibiotic use; among these, only methylation at the PLAGL1 DMR was also associated with birth weight.

Conclusion:

We report an inverse association between in utero exposure to antibiotics and lower infant birth weight and provide the first empirical evidence supporting imprinted gene plasticity in these associations.

Keywords: newborns, antibiotics, pregnancy, birth weight, epigenetics, race

Introduction

Early in utero exposures can lead to altered organogenesis and fetal development. Antibiotics, including penicillins, nitrofurans, and cephalosporins, are commonly prescribed for a wide range of ailments during the periconceptional and prenatal periods.1, 2, 3 Over the past 30 years, first-trimester use of prescription drugs increased >60%. In 2010, 94% of pregnant women took at least one medication during pregnancy, and 82% did so in the first trimester; among these medications, antibiotics were within the top 20 most frequently used, amoxicillin being top of the list, followed by azithromycin (No. 10), nitrofurantoin (No. 12), sulfamethoxazole-thrimethroprim (No. 14), penicillin (No. 16) and cephalexin (No. 19).4 Maternal antibiotic use has been associated with changes in infant birth weight,5, 6, 7, 8, 9 but in the few studies that reported associations, the direction of these changes was inconsistent.5, 6 Czeizel et al.5 reported lower birth weights among infants born to women who took antibiotics, whereas Jepsen et al.6 reported higher birth weights among infants born to antibiotic users in pregnancy. Both high birth weight (HBW), including large for gestational age, and low birth weight (LBW), including small for gestational age (SGA), are consistent risk factors for childhood and adult obesity,10 type 2 diabetes,11, 12 cardiovascular disease13, 14 and some cancers, such as breast15 and prostate16 cancers and esophageal adenocarcinomas.17 For reasons that are as yet unclear, since 1999 the US incidence of large for gestational age births has decreased, whereas that of SGA births has increased.18

Because antibiotics are the most commonly prescribed compound during pregnancy4 and the incidence of SGA is on the rise,18 understanding the relationship between birth weight and antibiotic use during pregnancy and mechanisms underlying these associations remains important. Improved understanding of this relationship may lead to focused interventions in pregnant women who may have other risk factors or exposures for LBW outcomes. Epigenetic mechanisms have been hypothesized to underlie these associations, but potential targets for the study are unknown. Many imprinted genes are critically involved in regulating early growth and development and are regulated by allele-specific methylation at differentially methylated regions (DMRs). The methylation patterns at imprinted DMRs are established in the gamete and early embryo and maintained through germ layer specification and tissue differentiation, resulting in highly similar levels of methylation that is present in all the tissues.19 Any shift in the process of reprogramming or lack of fidelity in maintaining these marks through the reprograming process can lead to shifts that are evident in somatic tissues. In contrast to most biallelically expressed genes where there are two active alleles, imprinted genes, with only one active allele, are at increased vulnerability to deregulated expression, and this can be mediated through epigenetic changes at the regulatory DMRs.20, 21, 22 As imprinted genes occur in clusters, their regulation may be under network-level control. This class of genes is also over-selected for growth effectors. For example, the pleomorphic adenoma gene-like 1 (PLAGL1) is a paternally expressed gene located at chromosome 6q24-q25, in the same chromosomal region as the imprinted transient neonatal diabetes mellitus locus, that encodes a zinc finger tumor-suppressor protein.23 In mice studies, Plagl1 was associated with intrauterine growth restriction, altered bone formation and neonatal lethality. Plagl1 was also proposed to serve as a major regulatory hub that coordinates the expression of many other genes, including a large number that are imprinted.24 We chose the nine imprinted marks in this study because these DMRs regulate genes already known to be important to appropriate growth and development,25, 26 neurological function27 and to social and maternal nurturing behaviors (from studies in mice)25 (for review, see Murphy et al.28). In addition, our previous studies have shown that these regions exhibit differential malleability in methylation profiles in response to antidepressant use during pregnancy,29 maternal depression,30 smoking,31 folic acid intake26 and paternal obesity.32 To our knowledge, no studies have provided mechanistic insights into the association between antibiotic intake during pregnancy and birth weight, although epigenetic mechanisms have been hypothesized. Here, we report associations between antibiotic intake during pregnancy, birth weight, and newborn methylation status at nine imprinted gene DMRs, including PLAGL1, Insulin-like Growth Factor 2 (IGF2) and noncoding H19, Neuronatin (NNAT), Delta-like homolog 1 and noncoding Maternally Expressed Gene 3 (DLK1/MEG3), Paternally Expressed Gene 3 (PEG3), the regulatory region regulating both Epsilon Sarcoglycan and Paternally Expressed Gene 10 (SGCE/PEG10) and Paternally Expressed Gene 1/Mesoderm-Specific Transcript (MEST).19

Subjects and methods

Subjects

Pregnant women were enrolled as part of NEST (Newborn Epigenetic STudy), a prospective study of women and their offspring in the Southeastern United States. Methods for enrollment of study participants have been previously described.30 Briefly, between 2009 and 2011, pregnant women were recruited from five prenatal clinics and obstetric facilities serving Duke University and Durham Regional Hospitals. To be eligible, women had to be ⩾18 years of age, pregnant and intending to use one of the participating obstetric facilities for delivery. Exclusion criteria were plans to relinquish custody of the child and plans to move away from the area in the subsequent 3 years. Women with established HIV-I infection were also excluded due to uncertainty about how this might alter DNA methylation profiles.

We consented and enrolled 1700 (66.7% response rate) of the 2548 women who were approached. Women enrolled in the study were similar to the 848 women who declined with respect to age (P=0.66) but more likely to be Asian and Native American women than African American or White (P<0.001). Among the 1700 women who consented, we excluded infant deaths before, during or soon after birth (n=115), 281 who were either illiterate, underage, refused further participation or could not be found, so that 1304 (76.7%) remained in the study. An additional 50 had missing information on antibiotic use, such that an effective sample size of 1254 remained for these analyses. These analyses are limited to the 397 women and their infants for whom data on infant DNA methylation were available. The study protocol was approved by the Duke University Institutional Review Board.

Data collection

Participants completed a self-administered questionnaire at the time of enrollment that included sociodemographic characteristics, reproductive factors, lifestyle factors such as cigarette smoking, anthropometric measurements and antibiotic use in the 6 months before enrollment; a time period that includes periconception. Medical records were abstracted at delivery to obtain information on birth outcomes, including birth weight and length. Infant cord blood specimens were collected at birth.

Assessment of birth outcome

Birth weight in grams and gestational age at delivery were abstracted from medical records following delivery.

Assessment of antibiotic exposure

To ascertain antibiotic use, participants were first asked a ‘Yes/No' question: ‘In the past six months did you take antibiotics?' if the answer was ‘Yes', pregnant women were asked to provide the name of the antibiotic used. To ensure inclusion of the periconceptional period, a vulnerable period for the fetus, the questionnaire was administered at enrollment so that antibiotic exposure included the 3 months period before conception. The mean gestational age at enrollment was 12 weeks (range 4–32 weeks). Study participants were unaware of the study hypothesis.

Measurement of co-variables

Current maternal body mass index (BMI) may reduce the effect of antibiotic use or may alter birth weight.14 Maternal BMI was calculated from self-reported weight and height at the time of enrollment and converted to kilograms and meters and expressed in kilograms per square-meter (kg m−2). Because birth weight also varies by race/ethnicity, women were asked to self-report their race/ethnicity. Maternal level of education attained, also a risk factor for birth weight, was self-reported.

DNA methylation analysis

Infant cord blood specimens were collected in EDTA-containing tubes and centrifuged using standard protocols to allow for collection of plasma and buffy coat for DNA extraction (Qiagen; Valencia, CA, USA); samples were stored at −80°C until required. 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, USA). Infant genomic DNA (800 ng) was modified by treatment with sodium bisulfite using the Zymo EZ DNA Methylation kit (Zymo Research; Irvine, CA, USA). Bisulfite treatment of denatured DNA converts all unmethylated cytosines to uracils but leaves methylated cytosines unchanged, allowing quantitative definition of cytosine methylation status. Pyrosequencing was performed using one of two Pyromark Q96 MD pyrosequencers (Qiagen). Nine imprinted DMRs for infants were analyzed: the IGF2 DMR, the H19 DMR, the NNAT DMR, the DLK1/MEG3 DMR, the DLK1/MEG3-IG DMR, the PLAGL1 DMR, the SGCE/PEG10 DMR, the PEG3 DMR, and the MEST DMR. Pyrosequencing assay design, genomic coordinates, assay conditions and assay validation have been previously described in detail.19, 33 Briefly, assays were designed to query established imprinted gene DMRs using the Pyromark Assay Design Software (Qiagen). PCR conditions were optimized to produce a single, robust amplification product by adjusting annealing temperature and magnesium chloride concentrations. 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's r>0.99 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 following bisulfite modification), keeping the thermocycler and pyrosequencer constant. Percentage of methylation for each CpG cytosine was determined using Pyro Q-CpG software (Qiagen). We interrogated between 3 and 10 CpG sites per DMR: 3 for IGF2, 4 for H19, 4 for MEST, 10 for PEG3, 6 for PLAGL1, 6 for SGCE/PEG10, 4 for MEG3-IG, 3 for NNAT, and 8 for MEG3. There was a high correlation between the values of CpGs within a DMR site (Cronbach's alphas for these regions were 0.95–0.99).

Statistical analyses

Infant birth weight showed no evidence of departure from normality and was used as both a continuous and a categorical variable. Categories were LBW (<2500 g), normal birth weight (2500–⩽4000 g), and HBW (>4000 g). We also calculated birth weight for gestational age z-scores as previously described.34 We examined the distribution of factors that have been previously associated with birth weight and DNA methylation at birth, including maternal age (categorized as <25, 25–29, 30–34 and ⩾35 years); BMI dichotomized at <30, 30–35 and >35 kg m−2; race/ethnicity (White, African American, Hispanic, others); maternal education (less than high school, high school graduate and some college-/college graduate-level education); maternal smoking (yes, no and quit during pregnancy); early prenatal folic acid intake (yes or no); gestational weight gain obtained from medical records; and gestational age at delivery, in relation to birth weight. Antibiotic use was dichotomized into non-users and users and further categorized by penicillins use and any other antibiotic use, including metronidazoles, nitrofurans, macrolides, tetracyclines, cephalosporines and quinolones. Factors previously associated with birth weight were compared with respect to antibiotic use during pregnancy using chi-square tests. Normality for each DMR was examined using the Kolmogorov–Smirnov test. With the exception of PLAGL1 (P<0.01), PEG3 (P<0.01), DLK1/MEG3-IG (P<0.01) and MEST (P=0.03), all other DMRs showed no evidence of departure from normality (P>0.05).20 We used F-tests for parametric and Kruskal–Wallis and Wilcoxon–Mann–Whitney tests to examine DNA methylation levels by antibiotic use. Linear regression models were used to examine the associations between antibiotic use and birth weight outcomes. DMR methylation values were included in mixed models to allow unrestrained entry of individual CpGs into refined models, one at a time, to examine the extent to which antibiotic use influences methylation levels. All statistical analyses were conducted in SAS v9.3 (SAS Institute, Cary, NC, USA).

Results

Table 1 summarizes the demographic characteristics of study participants. Twenty one percent (n=82) of the 397 women reported antibiotic use during pregnancy. Of these 82, 26% reported use of penicillins and 74% either did not report antibiotic names (48.6%) or reported use of nitrofurans (9%), metronidazoles (7%), quinolones (4%), tetracylclines (3%), macrolides (2%) and cephalosporines (0.4%). Most women (90%) were <35 years old, and maternal age did not vary by antibiotic use during pregnancy (P=0.50). About a third of reported antibiotic users (n=27) were obese (BMI >30 kg m−2), similar to 32% who did not report antibiotic use (P=0.92). More African-American women reported no use of antibiotics (41%), compared with White (19%) and Hispanic (31%) women (P=0.28). Overall, women's education level (<high school, high school graduate and some college/college graduate) was not associated with antibiotic use (P=0.97). More antibiotic users (17.07%) reported folic acid intake compared with non-users (7%) (P=0.007), while the proportions of antibiotic users who reported maternal smoking (P=0.83), being exposed to environmental smoke (P=0.73), and engaging in physical activity >30 min a day (P=0.93) were similar compared with non-users. Twenty eight percent of primiparous women reported use of antibiotics compared with 29% who did not report antibiotic use. Mean gestational age at enrollment was 12 weeks (s.d.=4.70), mean gestational weight gain was 12.96 kg (s.d.=8.35), and infants were all singletons. Infants not exposed to antibiotics and exposed infants were different with respect to delivery route (P=0.02), but their gender distribution was similar (P=0.22). A slightly higher proportion of infants whose mothers reported antibiotic use during pregnancy were born LBW (5%) versus infants not exposed to antibiotics (3%) (P=0.47). Conversely, slightly more unexposed than exposed infants (8% versus 5%) were born with HBW (P=0.15; Table 1).

Table 1. Characteristics of the 397 NEST study participants by antibiotic use.

Characteristic None (n=315) N (%) All antibiotics (n=82) N (%) Penicillins (n=21) N (%) Any other antibioticsa (n=61) N (%)
Maternal age (years)   P=0.50 P=0.58 P=0.74
 <25 (n=126) 105 (33.33) 21 (25.61) 4 (19.05) 17 (27.87)
 25–29 (n=122) 95 (30.16) 27 (32.93) 7 (33.33) 20 (32.79)
 30–34 (n=108) 85 (26.98) 23 (28.05) 7 (33.33) 16 (26.23)
 35+ (n=41) 30 (9.52) 11 (13.41) 3 (14.29) 8 (13.11)
Maternal BMI   P=0.92 P=0.89 P=0.77
 <30 (n=269) 214 (67.94) 55 (67.07) 14 (66.67) 41 (67.21)
 30–35 (n=67) 52 (16.15) 15 (18.29) 3 (14.29) 12 (19.67)
 ⩾30 (n=61) 49 (15.56) 12 (14.63) 4 (19.05) 8 (13.11)
Race   P=0.28 P=0.74 P=0.39
 African American (n=163) 130 (41.27) 93 (36.76) 8 (38.10) 25 (40.98)
 Caucasian (n=83) 60 (19.05) 88 (34.78) 6 (28.57) 17 (27.87)
 Hispanic (n=116) 97 (30.79) 44 (17.39) 5 (23.81) 14 (22.95)
 Other (n=35) 28 (8.89) 28 (11.07) 2 (9.52) 5 (8.20)
Gestational age at enrollment   P=0.70 P=0.60 P=0.82
 First trimester (0–13 weeks, n=222) 173 (54.92) 49 (59.76) 14 (66.67) 35 (57.38)
 Second trimester (>13 weeks, n=171) 139 (44.13) 32 (39.02) 7 (33.33) 25 (40.98)
 Third trimester (>26 weeks, n=4) 3 (0.95) 1 (1.22) 0 (0.00) 1 (1.64)
Education   P=0.97 P=0.85 P=0.99
 <High-school graduate (n=72) 57 (18.10) 15 (18.29) 4 (19.05) 11 (18.03
 High school (n=174) 139 (44.13) 35 (42.68) 8 (38.10) 27 (44.26)
 Some college/college graduate (n=151) 119 (37.38) 32 (39.02) 9 (42.86) 23 (37.70)
Gender of infants   P=0.22 P=0.97 P=0.15
 Male (n=194) 149 (47.30) 45 (54.88) 10 (47.62) 35 (49.73)
 Female (n=203) 166 (52.70) 37 (45.12) 11 (52.38) 26 (50.27)
Delivery route   P=0.02 P=0.08 P=0.07
 Vaginal (n=270) 223 (70.79) 47 (57.32) 11 (52.38) 36 (59.02)
 Caesarean section (n=127) 92 (29.21) 35 (42.68) 10 (47.62) 25 (40.98)
Folate intake   P=0.007 P=0.25 P=0.009
 Yes (n=37) 23 (7.30) 14 (17.07) 3 (14.29) 11 (18.03)
 No (n=360) 292 (92.70) 68 (82.93) 18 (85.71) 50 (81.97)
Maternal smoking   P=0.83 P=0.44 P=0.42
 Yes (n=34) 27 (8.57) 7 (8.54) 3 (14.29) 4 (6.56)
 Quit during pregnancy (n=78) 60 (19.05) 18 (21.95) 2 (9.52) 16 (26.23)
 No (n=285) 228 (72.38) 57 (69.51) 16 (76.19) 41 (67.21)
Environmental smoke   P=0.73 P=0.99 P=0.68
 Yes (n=77) 60 (20.73) 17 (20.73) 4 (19.05) 13 (21.31)
 No (n=320) 255 (80.95) 65 (79.27) 17 (80.95) 48 (78.69)
Physical activity   P=0.93 P=0.70 P=0.76
 Yes (n=110) 87 (27.62) 23 (28.05) 5 (23.81) 18 (29.51)
 No (n=287) 228 (72.38) 59 (71.95) 16 (76.19) 43 (70.49)
Parity   P=0.83 P=0.21 P=0.68
 Primiparous (n=115) 92 (29.21) 23 (28.05) 8 (38.10) 15 (24.59)
 Non-primiparous (n=282) 223 (70.79) 59 (71.95) 13 (61.90) 46 (75.41)
Gestational weight gain (kg)   P=0.42 P=0.61 P=0.20
(mean, s.d.) 13.39 (8.14) 12.53 (8.57) 14.34 (5.12) 11.92 (9.43)
Birth weight (g)   P=0.02 P=0.50 P=0.02
(mean, s.d.) 3374 (479.63) 3236 (461.1) 3299 (595.42) 3214 (408.60)
Birth weight   P=0.47 P=0.15 P=0.29
 Normal BW 2500–4000 g (n=356) 282 (89.52) 74 (90.24) 16 (76.19) 58 (95.08)
 LBW <2500 g (n=13) 9 (2.86) 4 (4.88) 2 (9.52) 2 (3.28)
 HBW >4000 g (n=28) 24 (7.62) 4 (4.88) 3 (14.29) 1 (1.64)
Open in a new tab

Abbreviations: BMI, body mass index; BW, birth weight; HBW, high birth weight; LBW, low birth weight; NEST, Newborn Epigenetic STudy.

Numbers not necessarily add up due to missing values.

a

Any antibiotics include: nitrofurans, metronidazoles, tetracyclines, macrolides, cephalosporins, quinolones and any other antibiotic.

Associations between antibiotic use and birth weight

Mean birth weight in this cohort was 3346 g (s.d.=479, range 1500–5500 g). Table 2 shows coefficient estimates (β's) and their s.e.s. for the multivariate linear regression of birth weight on maternal antibiotic use, in which unexposed infants served as the baseline category. These models were adjusted for infant gender, race/ethnicity, maternal BMI, delivery route, gestational weight gain, gestational age at delivery, folic acid intake, physical activity, parity and maternal smoking. We observed a significant inverse association between antibiotic use during pregnancy and birth weight (β=−132.99, s.e.=50.70, P=0.008), corresponding to 138 g lower birth weight among infants born to antibiotic users (Table 2). This difference was more pronounced in newborns exposed to antibiotics other than penicillins (which included nitrofurans, metronidazoles, tetracyclines, macrolides, cephalosporins, quinolones; β=−135.53, s.e.=57.38, P=0.02). The most commonly reported antibiotic, penicillin, was not significantly associated with birth weight (β=−125.91, s.e.=90.45, P=0.16).

Table 2. Adjusted regression coefficients and s.es. for the associations between antibiotic use during pregnancy and birth weighta.

  Any antibiotics (n=82) Penicillins (n=21) Any other antibiotics (n=61)
  β-coefficient, s.e., P-value β-coefficient, s.e., P-value β-coefficient, s.e., P-value
All (n=396) −132.99, 50.70, 0.008 −125.91, 90.45, 0.16 −135.53, 57.38, 0.02
Blacks (n=162) −137.04, 76.43, 0.07 −292.11, 139.37, 0.04 −86.09, 85.16, 0.31
Whites (n=83) −343.12, 101.65, 0.0007 −253.16, 179.32, 0.16 −379.28, 117.55, 0.001
Hispanics (n=116) +20.14, 97.13, 0.84 +11.33, 183.02, 0.95 +23.02, 109.53, 0.83
Maternal BMI <30 (n=268) −204.75, 62.51, 0.001 −138.24, 122.14, 0.22 −227.76, 70.27, 0.001
Maternal BMI 30–35 (n=67) −20.82, 134.17, 0.84 −390.25, 265.55, 0.14 +50.68, 140.63, 0.72
Maternal BMI >35 (n=353) −14.64, 91.28, 0.87 −156.68, 160.99, 0.33 35.03, 102.07, 0.73
Male infants (n=193) −197.15, 68.87, 0.004 −141.33, 130.54, 0.28 −214.23, 76.74, 0.005
Female infants (n=203) −70.27, 74.66, 0.35 −138.92, 125.85, 0.27 −41.71, 85.68, 0.62
Open in a new tab

Abbreviation: BMI, body mass index.

a

Adjusted for infant gender, race, maternal BMI, delivery route, gestational weight gain, gestational age at delivery, folic acid intake, physical activity, parity and maternal smoking.

Because the potential effects of any drug will depend on maternal obesity, we repeated these analyses among women with pre-pregnancy BMI 30–35, BMI ⩾35 and in those with BMI <30 kg m−2. We found that these associations were limited to infants born to women with BMI<30 kg m−2 (β-coefficient=−204.75, s.e.=62.51, P=0.001). However, the prevalence of BMI ⩾30 in African-American women was higher (45%) compared with Whites (25%) and Hispanics (24%). Restricting these analyses by race revealed a more pronounced difference in Whites (β=−343.12, s.e.=101.65, P=0.0007), compared with African Americans (β=−137.04, s.e.=76.43, P=0.07). The direction of the association between antibiotics and birth weight was in the opposite direction for Hispanic women (β=+20.14, s.e.=97.13 P=0.84). We stratified analyses by infant gender and found that these associations were stronger in male infants (β=−197.15, s.e.=68.87, P=0.004). These analyses were restricted to gestational age >37 weeks. These analyses were repeated in a larger sample of 1254 subjects, where 286 women reported antibiotic use, and the association between antibiotic use and birth weight was similar (all antibiotics: β=−90.22, s.e.=35.88, P=0.01; penicillins (n=78): β=−25.00, s.e.=64.00, P=0.69; and any other antibiotic (n=208): β=−58.21, s.e.=20.43, P=0.004). Furthermore, using birth weight for gestational age z-scores as the outcome to account for the non-linear growth in utero, we also found a strong inverse association between antibiotic use and birth weight (β=−0.28, s.e.=0.11, P=0.01) that was also stronger in infants exposed to any antibiotics (β=−0.28, s.e.=0.12, P=0.02) than to penicillins (β=−0.27, s.e.=0.19, P=0.16).

DNA methylation at DMRs regulating imprinted genes, prenatal antibiotic use and birth weight

We also examined whether DMR methylation of imprinted genes involved in growth and development mediated the association between antibiotic use during pregnancy and birth weight. To allow for unconstrained model entry of individual CpGs at each DMR, we used mixed models to first evaluate associations between DNA methylation levels at each of the nine DMRs and maternal antibiotic use. Using a liberal cutoff P<0.2, we found associations between DNA methylation and maternal antibiotic use for MEG3 (P=0.006), IGF2 (P=0.05), PLAGL1 (P=0.01), PEG3 (P=0.08) and H19 (P=0.15), after adjusting for race/ethnicity, maternal education, gestational age at delivery, folic acid intake, maternal smoking and infant gender (Table 3). After accounting for multiple comparisons using Bonferroni's correction, methylation at only two DMRs, PLAGL1 and MEG3, remained statistically significant. We then examined whether DNA methylation at these DMRs was also associated with birth weight and whether inclusion of PLAGL1 or MEG3 DMRs in the base model altered the association between antibiotic use and birth weight. We found that a 1% increase in DNA methylation at the PLAGL1 (P=0.04) but not at the MEG3 (P=0.42) DMR was significantly associated with a 10-g increase in birth weight (Table 4). After adjusting for infant gender, race/ethnicity, maternal BMI, delivery route, gestational weight gain, gestational age at delivery, folic acid intake, physical activity, parity and maternal smoking (Table 5), including PLAGL1 DMR methylation into the base model depicting the association between antibiotic use and birth weight revealed a somewhat stronger inverse association between antibiotic use and birth weight (β=−151.38, s.e.=71.59, P=0.03 in all antibiotic users, β=−216.42, s.e.=76.36, P=0.005, in users of antibiotics other than penicillins). Because DMR methylation may be less malleable in late pregnancy, we also repeated these analyses among 222 women who were enrolled in the first trimester, and these findings remained unaltered (data not shown).

Table 3. Mean DNA methylation percentages, regression coefficients (β, s.e.) and P-values for the association between antibiotic exposure and DMR methylation among 396 infantsa.

DMR Mean methylation % (s.d.) Any antibiotics (n=82) β-coefficient, s.e., P-value Penicillins (n=21) β-coefficient, s.e., P-value Any other antibiotics (n=61) β-coefficient, s.e., P-value
IGF2 51.30 (4.37) +0.11, 0.32, 0.73 +1.24, 0.56, 0.05 −0.32, 0.36, 0.40
H19 47.85 (3.86) +0.35, 0.23, 0.15 +0.18, 0.40, 0.65 +0.42, 0.27, 0.15
PLAGL1 57.34 (6.53) +0.85, 0.32, 0.01 +1.84, 0.56, 0.003 +0.47, 0.36, 0.21
MEG3 72.37 (5.51) +0.17, 0.43, 0.69 +2.43, 0.79, 0.006 −0.57, 0.48, 0.25
MEG3_IG 49.45 (3.33) +0.17, 0.46, 0.72 +0.06, 0.85, 0.94 +0.20, 0.52, 0.70
NNAT 55.08 (5.84) +0.14, 0.52, 0.79 +0.83, 0.85, 0.35 −0.17, 0.60, 0.77
PEG3 36.09 (3.34) +0.31, 0.17, 0.08 +0.41, 0.32, 0.21 +0.28, 0.19, 0.15
SGCE/PEG10 45.07 (5.44) +0.17, 0.27, 0.55 +0.01, 0.49, 0.98 +0.22, 0.31, 0.48
MEST 43.41 (4.55) −0.21, 0.30, 0.49 +0.22, 0.51, 0.67 −0.38, 0.34, 0.28
Open in a new tab

Abbreviations: DMR, differentially methylated region; IGF2, insulin-like growth factor 2; MEG3, maternally expressed gene 3; MEST, mesoderm-specific transcript; NNAT, neuronatin; PEG3, paternally expressed gene 3; PLAGL1, pleomorphic adenoma gene-like 1; SGCE, epsilon sarcoglycan.

a

Adjusted for infant gender, race/ethnicity, maternal education, maternal smoking, folic acid intake and gestational age at delivery.

Table 4. Adjusted regression coefficients and s.es. for the associations between DMR methylation and birth weight among 396 infantsa.

DMR Any antibiotics (n=82) β-coefficient, s.e., P-value
MEG3 −4.7, 5.85, 0.42
PLAGL1 +10.47, 5.22, 0.04
Open in a new tab

Abbreviations: DMR, differentially methylated region; MEG3, maternally expressed gene 3; PLAGL1, pleomorphic adenoma gene-like 1.

a

Adjusted for infant gender, race/ethnicity, maternal education, maternal smoking, folic acid intake and gestational age at delivery.

Table 5. aAdjusted regression coefficients and s.e.s. for the associations between antibiotic use during pregnancy and birth weight when DMR methylation is introduced into the model.

DMR Any antibiotics (n=82) Penicillins (n=21) Any other antibiotics (n=61)
All (n=396) β-coefficient, s.e., P-value β-coefficient, s.e., P-value β-coefficient, s.e., P-value
Base model −132.99, 50.70, 0.008 −125.91, 90.45, 0.16 −135.53, 57.38, 0.02
PLAGL1 −151.38, 71.59, 0.03 +128.99, 142.99, 0.36 −216.42, 76.36, 0.005
Open in a new tab

Abbreviations: DMR, differential methylated region; PLAGL1, pleomorphic adenoma gene-like 1.

a

Adjusted for infant gender, race/ethnicity, maternal body mass index, delivery route, gestational weight gain, gestational age at delivery, folic acid intake, physical activity, maternal smoking and parity.

Discussion

Antibiotics are the most commonly prescribed medication during pregnancy,4 and in livestock, antibiotic use has been associated with rapid growth.35 We examined whether antibiotic use in pregnancy up to 30 weeks gestation was associated with birth weight in a cohort of 397 newborns. Our key finding was that newborns exposed to antibiotics during pregnancy had a lower birth weight when compared with infants not prenatally exposed to antibiotics. These differences were most apparent in males and white infants and in newborns born to women with BMIs <30 kg m−2. We also found that PLAGL1 DMR methylation was associated with both antibiotic exposure and birth weight. Previous studies have found that infants with lower birth weights (even within normal birth weights) have more rapid postnatal catch-up growth and overweight in early infancy, a condition likely to persist into adulthood.36 These findings support imprinted gene plasticity in response to antibiotic exposure that may affect intrauterine growth.37

Our findings are consistent with other studies showing that antibiotic exposure during pregnancy is associated with reduced birth weight.5 Czeizel's group5 reported an average decrease in birth weight of 40 g for infants exposed to penicillins, compared with unexposed infants—similar to the 75-g difference reported here. However, these findings contrast with a report where a 57-g increase in birth weight was reported among infants exposed to antibiotics during pregnancy, compared with infants not exposed to antibiotics,6 although antibiotic use for most women in this study was after the third trimester of pregnancy.6 Our finding is consistent with the US trend in reduced birth weight observed between 1990 and 2005 (−52 g in the overall population and −79 g in a homogeneous low-risk subgroup),18 as antibiotic use in pregnant women increased.4 In our study, even strong predictors of HBW, such as maternal overweight and obesity, did not influence the inverse association between antibiotic exposure during pregnancy and birth weight.

Our finding that antibiotic exposure in non-obese mothers was inversely associated with birth weight is intriguing, as antibiotics are growth stimulants in livestock38 and have been suggested as contributors to the obesity epidemic.38 However, in the womb, the infant gut is relatively sterile39 and thus antibiotics could not be exerting the same effect as in fully colonized guts. In addition, our findings of reduced birth weight also have implications for postnatal development, particularly catch-up growth in early life. Children who are SGA and exposed to antibiotics early in life have a higher risk of overweight later in life, by age 7 years.40, 41 Antibiotics may alter the diversity and composition of the maternal gut microbiota and possibly reducing the population of obesogenic microbes. The latter include the Bacteroides, Clostridium, and Staphylococcus genuses, which are found in higher concentrations in overweight pregnant women and high gestational-weight-gain women, as well as in overweight children.42 According to this hypothesis, newborns of mothers exposed to antibiotics would inherit a less obesogenic maternal microbiota, which is consistent with evidence of a reduced risk of overweight for infants born to obese mothers who were administered antibiotics in the early postnatal period.43 Taken together, our findings and other studies support the hypothesis that early exposure to antibiotics may prematurely alter the diversity of microbiota in the fetus. It has been suggested that the cumulative effect of pre- and postnatal antibiotic exposure may increase the risk of chronic diseases in adulthood, as gut microbiota composition is intimately linked to a healthy immune system,44 via epigenetic adjustment evident at some loci.

Reasons for the present increase in SGA births in the United States are not known, and although racial and ethnic disparities and variation by geography and familial aggregation have been suggested as potential contributors,45 the effect of prescription drug intake during pregnancy on birth weight has not been thoroughly investigated. Although associations between antibiotic use during pregnancy and birth weight outcomes have been described recently in a handful of studies,5, 6 and epigenetics has been hypothesized to mediate these relationships, to our knowledge, this is the first study to examine a potential epigenetic mediation to birth weight outcomes when infants are exposed to antibiotics in utero. Clarifying the relationship between antibiotic use during pregnancy and birth weight outcomes has the potential to further our understanding of whether frequently prescribed medications taken during fetal development permanently alter the epigenome and thus increase the risk of chronic diseases in childhood and adulthood. Methylation differences between antibiotic exposed and unexposed infants were found at MEG3, PEG3, PLAGL1, IGF2 and H19 DMRs; however, only methylation at PLAGL1 DMR was statistically significantly associated with birth weight, suggesting that DNA methylation alterations at this DMR may be potentially mediating the association between antibiotic use and birth weight. The effect of PLAGL1 DMR has implications for follow-up studies in larger populations to determine whether these epigenetic signatures become permanent and influence these imprinted genes' expression and thus infant development in early postnatal life.

Our findings should be interpreted in the context of their limitations. Even though the NEST cohort includes a large number of pregnant women, antibiotic use was self-reported with no follow-up metabolite analysis. Less than half of participants who used antibiotics recalled the name of the antibiotic prescribed. However, it is unlikely that women would report antibiotic use when none were used. In addition, although we did not measure antibiotic dosage nor duration, or metabolites, most women reported antibiotic use in the 24 weeks previous to enrollment, a time frame that included periconception and the first 12 weeks of pregnancy, when fetal vulnerability is highest during organogenesis. Another limitation is the possibility that there could have been confounding by indication. However, sub-analysis of reported morbidity did not suggest that morbidity was associated with birth weight. The small numbers of infants with DNA methylation data did not allow us to present data further stratified by sex and race/ethnicity, although these findings did not diverge from what is reported here. Moreover, because this is an ongoing prospective study, we plan to confirm these initial results in larger samples. Another limitation is that we were not able to measure gut microbiota composition in antibiotic exposed versus unexposed infants, nor did we analyze transcript levels of the imprinted genes, although such studies are planned.

Although small numbers limit our ability to make inferences, our results suggest that prenatal exposure to antibiotics is associated with lower birth weight, an association that may be mediated, at least in part, by alterations in the differential methylation at regulatory regions of imprinted genes. Larger studies using agnostic approaches to epigenetic targets are needed to confirm these findings.

Acknowledgments

We are grateful to the participants in the NEST study project, without whom this work would not have been possible. We thank Carole Grenier, Erin Erginer, Allison Barrat and Cara Davis for excellent technical assistance, and Francine Overcash and Stacy Murray for their recruiting efforts. This work was funded by NIH R01-ES016772. The Newborn Epigenetics Study (NEST) was funded through the US National Institutes of Health (agreement no. R01ES016772) and R01CA142983-02S1.

The authors declare no conflict of interest.

References

  1. Andrade SE, Gurwitz JH, Davis RL, Chan KA, Finkelstein JA, Fortman K, et al. Prescription drug use in pregnancy. Am J Obstet Gynecol. 2004;19:398–407. doi: 10.1016/j.ajog.2004.04.025. [DOI] [PubMed] [Google Scholar]
  2. Kaatsch P, Scheidemann-Wesp U. Maternal use of antibiotics and cancer in the offspring: results of a case-control study in Germany. Cancer Causes Control. 2010;21:1335–1345. doi: 10.1007/s10552-010-9561-2. [DOI] [PubMed] [Google Scholar]
  3. Crider KS, Cleves MA, Reefhuis J, Berry RJ, Hobbs CA, Hu DJ. Antibacterial medication use during pregnancy and risk of birth defects. Arch Pediatr Adolesc Med. 2009;163:978–985. doi: 10.1001/archpediatrics.2009.188. [DOI] [PubMed] [Google Scholar]
  4. Mitchell AA, Gilboa SM, Werler MM, Kelley KE, Louik C, Hernandez-Diaz S. National Birth Defects Prevention Study. Medication use during pregnancy, with particular focus on prescription drugs: 1976-2008. Am J Obstet Gynecol. 2011;205:e1–e8. doi: 10.1016/j.ajog.2011.02.029. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Czeizel AE, Rockenbauer M, Olsen J. Use of antibiotics during pregnancy. Eur J Obstet Gynecol Reprod Biol. 1998;81:1–8. doi: 10.1016/s0301-2115(98)00138-9. [DOI] [PubMed] [Google Scholar]
  6. Jepsen P, Skriver MV, Floyd A, Lipworth L, Schonheyder HC, Sorensen HT. A population-based study of maternal use of amoxicillin and pregnancy outcome in Denmark. Br J Clin Pharmacol. 2002;55:216–221. doi: 10.1046/j.1365-2125.2003.01750.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Benn CS, Thorsen P, Jensen JS, Kjaer BB, Bisgaard H, Andersen M, et al. Maternal vaginal microflora during pregnancy and the risk of asthma hospitalization and use of antiasthma medication in early childhood. J Allergy Clin Immunol. 2002;110:72–77. doi: 10.1067/mai.2002.125833. [DOI] [PubMed] [Google Scholar]
  8. McKeever TM, Lewis SA, Smith C, Hubbard R. The importance of prenatal exposures on the development of allergic disease: a birth cohort study using the West Midlands General Practice Database. Am J Respir Crit Care Med. 2002;166:827–832. doi: 10.1164/rccm.200202-158OC. [DOI] [PubMed] [Google Scholar]
  9. Martel MJ, Rey E, Malo JL, Perreault S, Beauchesne MF, Forget A, et al. Determinants of the incidence of childhood asthma: a two-stage case-control study. Am J Epidemiol. 2009;169:195–205. doi: 10.1093/aje/kwn309. [DOI] [PubMed] [Google Scholar]
  10. Yu ZB, Han SP, Zhu GZ, Zhu C, Wang XJ, Cao XG, et al. Birth weight and subsequent risk of obesity: a systematic review and meta-analysis. Obes Rev. 2011;12:525–542. doi: 10.1111/j.1467-789X.2011.00867.x. [DOI] [PubMed] [Google Scholar]
  11. Hales CN, Barker DJP. Type 2 (non-insulin-dependent) diabetes mellitus: the thrifty phenotype hypothesis. Diabetologia. 1992;35:595–601. doi: 10.1007/BF00400248. [DOI] [PubMed] [Google Scholar]
  12. Whincup PH, Kaye SJ, Owen CG, Huxley R, Cook DG, Anazawa S, et al. Birth weight and risk of type 2 diabetes: a systematic review. JAMA. 2008;300:2886–2897. doi: 10.1001/jama.2008.886. [DOI] [PubMed] [Google Scholar]
  13. Drake AJ, Walker BR. The intergenerational effect of fetal programming: non-genomic mechanisms for the inheritance of low birth weight and cardiovascular risk. J Endocrinol. 2004;180:1–16. doi: 10.1677/joe.0.1800001. [DOI] [PubMed] [Google Scholar]
  14. Barker DJP, Osmond C, Golding J, Kuh D, Wadsworth. MEJ. Growth in utero, blood pressure in childhood and adult life, and mortality from cardiovascular disease. Br Med J. 1989;298:564–567. doi: 10.1136/bmj.298.6673.564. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Wu AH, McKean-Cowdin R, Tseng CC. Birth weight and other factors and risk of breast cancer in Asian-Americans. Breast Cancer Res Treat. 2011;130:917–925. doi: 10.1007/s10549-011-1640-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Nilsen TIL, Romundstad PR, Troisi R, Vatten LJ. Birth size and subsequent risk for prostate cancer: A prospective population-based study in Norway. Int J Cancer. 2005;113:1002–1004. doi: 10.1002/ijc.20674. [DOI] [PubMed] [Google Scholar]
  17. Kaijser M, Akre O, Cnattingius S, Ekbom A. Preterm birth, low birth weight, and risk for esophageal adenocarcinoma. Gastroenterology. 2005;128:607–609. doi: 10.1053/j.gastro.2004.11.049. [DOI] [PubMed] [Google Scholar]
  18. Donahue SMA, Kleinman KP, Gillman MW, Oken E. Trends in birth weight and gestational length among singleton term births in the United States. Obstet Gynecol. 2010;115:357–364. doi: 10.1097/AOG.0b013e3181cbd5f5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Murphy SK, Huang Z, Hoyo C. Differentially methylated regions of imprinted genes in prenatal, perinatal and postnatal tissues. PLoS One. 2012;7:e40924. doi: 10.1371/journal.pone.0040924. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Cui H, Onyango P, Brandenburg S, Wu Y, Hsieh CL, Feinberg AP. Loss of imprinting in colorectal cancer linked to hypomethylation of H19 and IGF2. Cancer Res. 2002;62:6442–6446. [PubMed] [Google Scholar]
  21. Cui H, Cruz-Correa M, Giardiello FM, Hutcheon DF, Kafonek DR, Brandenburg S, et al. Loss of IGF2 imprinting: a potential marker of colorectal cancer risk. Science. 2003;299:1753–1735. doi: 10.1126/science.1080902. [DOI] [PubMed] [Google Scholar]
  22. Murphy SK, Huang Z, Wen Y, Spillman MA, Whitaker RS, Simel LR, et al. Frequent IGF2/H19 domain epigenetic alterations and eleveated IGF2 expression in epithelial ovarian cancer. Mol Cancer Res. 2006;4:283–292. doi: 10.1158/1541-7786.MCR-05-0138. [DOI] [PubMed] [Google Scholar]
  23. Kamiya M, Judson H, Okazaki Y, Kusakabe M, Muramatsu M, Takada S, et al. The cell cycle control gene ZAC/PLAGL1 is imprinted- a strong candidate gene for transient neonatal diabetes. Hum Mol Genet. 2000;9:453–460. doi: 10.1093/hmg/9.3.453. [DOI] [PubMed] [Google Scholar]
  24. Varrault A, Gueydan C, Delalbre A, Bellmann A, Houssami S, Aknin C, et al. Zac1 regulates an imprinted gene network critically involved in the control of embryonic growth. Dev Cell. 2006;11:711–722. doi: 10.1016/j.devcel.2006.09.003. [DOI] [PubMed] [Google Scholar]
  25. Lefebvre L, Viville S, Barton SC, Ishino F, Keverne EB, Surani MA. Abnormal maternal behaviour and growth retardation associated with loss of the imprinted gene Mest. Nat Genet. 1998;20:163–169. doi: 10.1038/2464. [DOI] [PubMed] [Google Scholar]
  26. Hoyo C, Murtha AP, Schildkraut JM, Jirtle RL, Demark-Wahnefried W, Forman MR, et al. Methylation variation at IGF2 differentially methylated regions and maternal folic acid use before and during pregnancy. Epigenetics. 2011;6:928–936. doi: 10.4161/epi.6.7.16263. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Rajiv J, Dou D, Tsang W. Molecular cloning of a novel mRNA (Neuronatin) that is highly expressed in neonatal mammalian brain. Biochem Biophys Comm. 1994;201:1227–1234. doi: 10.1006/bbrc.1994.1836. [DOI] [PubMed] [Google Scholar]
  28. Murphy SK, Jirtle RL. Imprinting evolution and the price of silence. Bioessays. 2003;25:577–588. doi: 10.1002/bies.10277. [DOI] [PubMed] [Google Scholar]
  29. Soubry A, Murphy SK, Huang Z, Murtha AP, Schildkraut JM, Jirtle RL, et al. The effects of depression and use of antidepressive medicines during pregnancy on the methylation status of the IGF2 imprinted control regions in the offspring. Clin Epigenetics. 2011;3:2. doi: 10.1186/1868-7083-3-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Liu Y, Murphy SK, Murtha AP, Fuemmeler BF, Schildkraut JM, Huang Z, et al. Depressed mood in pregnancy, birthweight, and DNA methylation of imprint regulatory elements. Epigenetics. 2012;7:746. doi: 10.4161/epi.20734. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Murphy SK, Adigun A, Huang Z, Overcash F, Wang F, Jirtle RL, et al. Gender-specific methylation differences in relation to prenatal exposure to cigarette smoke. Gene. 2012;494:36–43. doi: 10.1016/j.gene.2011.11.062. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Soubry A, Schildkraut JM, Murtha A, Wang F, Huang Z, Bernal A, et al. Paternal obesity is associated with IGF2 hypomethylation in newborns: results from a Newborn Epigenetics Study (NEST) cohort. BMC Med. 2013;11:29. doi: 10.1186/1741-7015-11-29. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Nye M, Hoyo C, Gammon M, Huang Z, Vidal AC, Wang F, et al. Associations between methylation of Paternally Expressed Gene 3 (PEG3), cervical intraepithelial neoplasia and invasive cervical cancer. PLoS One. 2013;8:e56325. doi: 10.1371/journal.pone.0056325. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Oken E, Kleinman KP, Rich-Edwards J, Gillman MW. A nearly continuous measure of birth weight for gestational age using a United States national reference. BMC Pediatrics. 2003;3:6. doi: 10.1186/1471-2431-3-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Gaskins HR, Collier CT, Anderson DB. Antibiotics as growth promotants: mode of action. Anim Biotechnol. 2002;13:29–42. doi: 10.1081/ABIO-120005768. [DOI] [PubMed] [Google Scholar]
  36. Karaolis-Danckert N, Buyken AE, Bolzenius K, Perim de Faria C, Lentze MJ, Kroke A. Rapid growth among term children whose birth weight was appropriated for gestational age has a longer lasting effect on body fat percentage than on body mass index. Am J Clin Nutr. 2006;84:1449–1455. doi: 10.1093/ajcn/84.6.1449. [DOI] [PubMed] [Google Scholar]
  37. Liu Y, Hoyo C, Murphy S, Huang Z, Overcash F, Thompson J, et al. DNA methylation at imprint regulatory regions in preterm birth and infection Am J Obst Gynecol 2013. piiS0002-937800142–00147. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Ternak G. Antibiotics may act as growth/obesity promoters in humans as an inadvertent result of antibiotic pollution. Med Hypothesis. 2005;64:14–16. doi: 10.1016/j.mehy.2004.08.003. [DOI] [PubMed] [Google Scholar]
  39. Gosalbes MJ, Llop S, Valles Y, Moya A, Ballester F, Francino MP. Meconium microbiota types dominated by lactic acid or enteric bacteria are differentially associated with maternal eczema and respiratory problems in infants. Clin Exp Allergy. 2012;43:198–211. doi: 10.1111/cea.12063. [DOI] [PubMed] [Google Scholar]
  40. Ajslev TA, Andersen CS, Gambor M, Sorensen TIA, Jess T. Childhood overweight after establishment of the gut microbiota: the role of delivery mode, pre-pregnancy weight and early administration of antibiotics. Int J Obes. 2011;35:522–529. doi: 10.1038/ijo.2011.27. [DOI] [PubMed] [Google Scholar]
  41. Kalliomaki M, Collado MC, Salminen S, Isolauri E. Early differences in fecal microbiota composition in children may predict overweight. Am J Clin Nutr. 2008;87:534–538. doi: 10.1093/ajcn/87.3.534. [DOI] [PubMed] [Google Scholar]
  42. Collado MC, Isolauri E, Laitinen K, Salminen S. Distinct composition of gut microbiota during pregnancy in overweight and normal-weight women. Am J Clin Nutr. 2008;88:894–899. doi: 10.1093/ajcn/88.4.894. [DOI] [PubMed] [Google Scholar]
  43. Dominguez-Bello MG, Costello EK, Contreras M, Magris M, Hidalgo G, Fierer N, et al. Deliver mode shapes the acquisition and structure of the initial microbiota across multiple body habitats in newborns. Proc Natl Acad Sci USA. 2010;107:11971–11975. doi: 10.1073/pnas.1002601107. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Round JL, Mazmanian SK. The gut microbiota shapes intestinal immune responses during health and disease. Nat Rev Immunol. 2009;9:313–323. doi: 10.1038/nri2515. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Luyckx VA, Brenner BM. The clinical importance of nephron mass. J Am Soc Nephrol. 2010;21:898–910. doi: 10.1681/ASN.2009121248. [DOI] [PubMed] [Google Scholar]

Articles from International Journal of Obesity (2005) are provided here courtesy of Nature Publishing Group

RESOURCES