Cohort profile: Epigenetics in Pregnancy (EPIPREG) - population-based sample of European and South Asian pregnant women with epigenome-wide DNA methylation (850k) in peripheral blood leukocytes.

Pregnancy is a valuable model to study the association between DNA methylation and several cardiometabolic traits, due to its direct potential to influence mother's and child's health. Epigenetics in Pregnancy (EPIPREG) is a population-based sample with the aim to study associations between DNA-methylation in pregnancy and cardiometabolic traits in South Asian and European pregnant women and their offspring. This cohort profile paper aims to present our sample with genetic and epigenetic data and invite researchers with similar cohorts to collaborative projects, such as replication of ours or their results and meta-analysis. In EPIPREG we have quantified epigenome-wide DNA methylation in maternal peripheral blood leukocytes in gestational week 28±1 in Europeans (n = 312) and South Asians (n = 168) that participated in the population-based cohort STORK Groruddalen, in Norway. DNA methylation was measured with Infinium MethylationEPIC BeadChip (850k sites), with technical validation of four CpG sites using bisulphite pyrosequencing in a subset (n = 30). The sample is well characterized with few missing data on e.g. genotype, universal screening for gestational diabetes, objectively measured physical activity, bioelectrical impedance, anthropometrics, biochemical measurements, and a biobank with maternal serum and plasma, urine, placenta tissue. In the offspring, we have repeated ultrasounds during pregnancy, cord blood, and anthropometrics up to 4 years of age. We have quantified DNA methylation in peripheral blood leukocytes in nearly all eligible women from the STORK Groruddalen study, to minimize the risk of selection bias. Genetic principal components distinctly separated Europeans and South Asian women, which fully corresponded with the self-reported ethnicity. Technical validation of 4 CpG sites from the methylation bead chip showed good agreement with bisulfite pyrosequencing. We plan to study associations between DNA methylation and cardiometabolic traits and outcomes.

Introduction

Studies of epigenetic marks have in recent years gained increased interest in the context of human diseases [1]. Such studies may enhance our biological understanding of the aetiology of several diseases, increase our understanding of detrimental or protective mechanisms, or for prognosis and risk prediction [2]. One of the most studied epigenetic mechanisms is DNA methylation, which plays an important role in normal development, chromatin organization and gene expression [3]. Several studies have indicated that DNA methylation is associated with cardiovascular risk factors such as body mass index (BMI) [4-7], gestational diabetes (GDM) [8], type 2 diabetes (T2D) [9-13], lipid levels [14,15], hypertension [16,17], smoking [18-22] and alcohol intake [23-25], suggesting that cardiometabolic diseases have an epigenetic component.

Although scarcely studied, pregnant women provide a unique opportunity to study the association between blood DNA methylation and several phenotypes related to glucose homeostasis and cardiovascular traits. This is because pregnancy has been proposed as a stress test for metabolism in several organs [26], including the pancreatic beta-cells, since insulin resistance increase naturally in all pregnancies [27]. In the third trimester of pregnancy, this insulin resistance in many women reaches a level similar to that observed in type 2 diabetes, requiring the beta-cell to increase its insulin secretion considerably to compensate [28]. Similarly, pregnancy-induced hypertension is associated with increased risk of future cardiovascular disease [29].

In the Epigenetics in Pregnancy (EPIPREG) sample, we have quantified epigenome-wide DNA-methylation in peripheral blood leukocytes in women of European and South Asian origin attending the well-characterized, multi-ethnic and population-based STORK Groruddalen (STORK G) study [30]. The population based design and inclusion of women with Western and South Asian ethnicity allow us to study of a wide range of phenotypes to detect either ethnic-specific and/or common DNA methylation patterns in relation to phenotypes of interest. Furthermore, South Asians are of special interest due to their higher weight retention after pregnancy, increased prevalence of gestational diabetes, and increased risk for later type 2 diabetes compared to Europeans [31,32]. The aim of EPIPREG is to discover novel associations between DNA-methylation in pregnancy and cardiometabolic related traits in South Asian and European pregnant women and their offspring, which may have potential for prevention and treatment. This cohort profile paper aims to present our sample with genetic and epigenetic data and invite researchers with similar cohorts to collaborative projects, such as replication of ours or their results and meta-analysis.

Cohort description

Study population

EPIPREG (n = 480) is a sub-study of the larger STORK Groruddalen (STORK G) study, which is a population-based cohort of 823 healthy women with different ethnic origin (European, South Asian, African, Middle Eastern and South American) attending three public Child Health Clinics for antenatal care in the multi-ethnic area of Groruddalen, Oslo, Norway, 2008–2010 [30]. Briefly, women were eligible if they: 1) Lived in the study districts; 2) planned to give birth at one of two study hospitals; 3) were <20 weeks pregnant; 4) could communicate in Norwegian or any of the eight translated languages; and 5) were able to give an informed consent. Women with pre-gestational diabetes, or in need of intensive hospital follow-up during pregnancy, were excluded. The overall participation rate in STORK G was 74%, 81.5% for Europeans and 73% for South Asians [30].

Ethical approval

The STORK G study including genetic and epigenetic data is approved by the Norwegian Regional Committee for Medical Health Research Ethics South East (ref.number 2015/1035). We obtained written informed consent from all participants before any study-related procedure.

Data collection

Questionnaire data and anthropometrics

In STORK G, interviewer-administered questionnaires were completed at gestational week 15±3 (visit 1) and 28±2 during pregnancy (visit 2), and 12±2 weeks postpartum (visit 3) [30,33]. Questionnaire data included information on mother’s general health, physical activity and a dietary habits, in addition to some information about the father. Details about the questions used and data gathered have been described previously [30], and are available upon request to the corresponding.

Ethnic origin was defined by either the individual’s country of birth or their mother’s country birth, if the latter was born outside Europe [34]. We have detailed data on parity, pre-pregnant BMI, smoking status, alcohol intake, education, marital status and diet [30,33,35]. At all the three visits, we measured maternal height, body weight, fat mass with bioelectrical impedance (Tania-Weight BC-418 MA), skinfold thickness at three sites (Holtain T/W Skinfold Caliper, Holtain Ltd., Crymych) [31] and systolic and diastolic blood pressure (Omron HEM-7000-E M6 Comfort) [36].

Universal screening for gestational diabetes

All women underwent a 75 g oral glucose tolerance test at gestational week 28 ±2. Fasting and 2-hour glucose were analysed with a point-of-care instrument (HemoCue, Angelholm, Sweden). Women were diagnosed with gestational diabetes based on the WHO 1999 criteria (fasting glucose ≥ 7.0 mmol/l or 2-hour glucose ≥ 7.8 mmol/l) [37]. Furthermore, in retrospect and exclusively for research purposes, we also classified the samples using a slightly modified version of the WHO 2013 criteria (fasting glucose ≥ 5.1–6.9 mmol/l or 2-hour glucose ≥ 8.5–11 mmol/l, no data for 1-hour glucose) [38].

Laboratory data

Venous blood was drawn at the three visits into tubes with ethylenediaminetetraacetic acid (EDTA). Subsequently, the samples were aliquoted and biobanked or subject to routine laboratory analyses that were performed continuously during the study period. Fasting glucose, total Cholesterol, LDL-Cholesterol, HDL-Cholesterol and triglycerides levels were measured with a colorimetric method (Vitros 5.1 fs, Ortho clinical diagnostics) [35], HbA1c levels were assessed in full blood with HPLC (Tosoh G8) [34], fasting C-peptide and insulin were measured at the Hormone Laboratory, Oslo University Hospital, with non-competing immunoflurometric assays (DELFIA, PerkinElmer Life Sciences, Wallac Oy, Turku, Finland) [39]. Serum 25(OH)D was analysed by competitive RIA (DiaSorin) at visit 1 and visit 2 [40], and S-leptin was analyzed by HADCYMAG-61K based on Luminex® xMAP® technology [32], at the Hormone Laboratory, Oslo University Hospital. Serum vitamin B12 and folate were measured with Electrochemiluminescence (ECLIA) assays, Roche, at Medical Biochemistry, Oslo University Hospital. HOMA-IR and HOMA-B [39,41] were estimated by Oxford University HOMA Calculator 2.2 using fasting glucose and C-peptide.

Objectively measured physical activity

Physical activity (PA) was objectively measured from visit 1 to 3 using SenseWear™ Pro3 Armband (SWA) (BodyMedia Inc, Pittsbur, PA, USA) [42]. Data from women with at least one valid day (defined as ≥ 19.2h) were considered valid [43]. Physical activity was characterized as Sedentary behaviour (< 1.5 metabolic equivalents (METs)) light intensity (1.5 to <3 METS) or moderate or intense (≥ 3 METs) [42,44].

Offspring data

The STORK G study also collected abdominal circumference, head circumference, bi-parietal diameter, femur length and estimated fetal weight by ultrasound measured on three different time points during pregnancy, and gestational age at birth derived from the first day of the woman’s last menstrual period [45]. We have detailed anthropometric measurements at birth such as birthweight, head circumference, abdominal circumference, crown-heel length and neonatal skinfolds measured with a Holtain T/W Skinfold Caliper (Holtain Ltd., Crymych) [46]. Measurements of weight and length/height were collected during routine follow-up at the Mother—and Child Health Clinics when the children were 6 weeks old and thereafter at the 3, 6, 12, 15, 24 and 48 months visits. Further register-based follow-up is planned.

Venous serum cord blood samples were collected at birth and stored at -80°C. Several sections from the placenta and umbilical cord have been sampled, and stored as Formalin-Fixed Paraffin-Embedded (FFPE) blocks. Currently, an ongoing pilot study of 80 FFPE placentas (40 South Asian, 40 European) demonstrate that enough DNA can be extracted from the FFPE to successfully run pyrosequencing (Sletner, unpublished). Furthermore, we have frozen placenta biopsies of about 1/3 of the women’s offspring.

DNA extraction

In STORK G, at gestational week 28±2, DNA from peripheral blood leukocytes was extracted continuously throughout the data collection, at the Hormone Laboratory, Oslo University Hospital, using a salting out procedure [47], and stored at -80°C.

Genetic data

The samples were genotyped using the Illumina CoreExome chip, by the Department of Clinical Sciences, Clinical Research Centre, Lund University, Malmö, Sweden [48]. Of the 664 genotyped samples those with low call rate (i.e. < 95%, n = 0), extreme heterozygosity (> |mean± (3xSD)|, n = 1), mismatched gender (n = 24) or cryptic relatedness (i.e. one individual (chosen at random) from each related pair, defined as genome-wide Identity by descent (IBD) > 0.185 (n = 6) were excluded from analyses. Genetic ethnic origin was defined by ancestry informative principal component analysis based on the variance-standardized relationship matrix generated in PLINK 1.9 software package [49] (https://www.cog-genomics.org/plink/1.9/)).

Variants with call rate <95% (10081 SNPs), out of Hardy-Weinberg equilibrium (exact p<10−6, 1971 SNPs) or with low minor allele frequency (MAF) <1% (245221 SNPs) were removed before imputation. Quality control was performed using the PLINK 1.9 software package [49]. After quality control, 293914 variants were left for imputation.

Imputation in European and South Asian samples was performed as follows: The GWAS scaffold was mapped to NCBI build 37 of the human genome. Imputation to the 1000G reference panel (Phase3- http://www.well.ox.ac.uk/~wrayner/tools/) was performed separately in Europeans and South Asians using their respective panels from the 1000G. The populations that are included in the European panel are Utah residents with Northern and Western European ancestry, Iberian populations in Spain, Finish in Finland, British in England and Scotland, and Tuscany in Italy. In the South Asian panel the included populations are Bengali in Bangladesh, Gujarati Indian in Houston, Texas, Indian Telugu in the UK, Punjabi in Lahore, Pakistan and Sri Lankan Tamil in the UK. The software used for imputation was IMPUTE (version 2.3.2) [50].

Epigenome-wide DNA methylation

Europeans and South Asians were the largest ethnic groups in STORK G, and South Asians of special interest due to their higher weight retention after pregnancy, increased prevalence of gestational diabetes, and increased risk for later type 2 diabetes compared to Europeans [32]. In EPIPREG, we quantified DNA methylation in maternal peripheral blood leukocytes in gestational week 28±1.2 in all Europeans (n = 312) and South Asians (n = 168) participating in STORK G who were genotyped and had fasting glucose data recorded. DNA samples were bisulfite converted using EZ DNA MethylationTM Kit (Zymo Research, Tustin, CA, USA) before added onto Infinium MethylationEPIC BeadChip (Illumina, San Diego, CA, USA) at the Department of Clinical Sciences, Clinical Research Centre, Lund University, Malmö, Sweden. Raw signal intensities of each probe were extracted using Illumina’s GenomeStudio Software. The methylation level at each site was represented as a beta (β) value of the fluorescent intensity radio ranging from 0 (not methylated) to 1 (completely methylated). Meffil R package [51] (https://cran.r-project.org/) was used for quality control, normalization and reporting of beta values. During QC, we removed 8 samples: 1 due to sex mismatch (predicted sex outliers > 5SD), 1 outlier in control probes bisulfite 1 and bisulfite 2 (>5 SD), and 6 outliers from the methylated/unmethylated ratio comparison (>3 SD). Furthermore, 1299 probes with a detection p-value <0.01, and bead count <3 were removed. We used functional normalization, adjusting for effects of different batches, plates, columns and rows. A total of 307 European and 165 South Asian women and 864 560 probes of the array passed the QC.

Pyrosequencing

Random samples of 30 women were selected for technical validation of four CpGs sites by bisulfite pyrosequencing. The four CpG sites were chosen from preliminary top associations with fasting glucose (cg08098128, cg14120215), 2-hour glucose (cg19327414) and BMI (cg17148978). DNA samples were first bisulphite converted per QIAGEN Bisulfite conversion protocol [52] using 500 ng of DNA. Short DNA sequences that contained the CpG site of interest were amplified by PCR using PyroMark PCR kit from QIAGEN following the manufacturer instructions [53]. The PCR was performed on the 30 samples in duplicates with two positive controls, one was unmethylated converted DNA and the other was fully methylated converted DNA, both controls were commercially available from the EpiTect PCR Control DNA Set (100) by QIAGEN, and were used per the provider instructions. Also, a negative control only containing RNAse free water was used. Pyrosequecing was performed using the PyroMark Q48 Autoprep per the instructions provided in the user manual [54]. We used Bland-Altman plots to evaluate the agreement between pyrosequencing and the Infinium MethylationEPIC BeadChip, which have been used previously to assess agreement for this type of data [55,56]. To asses if there were proportional bias, we regressed the mean difference with the mean between methods. We also performed Pearson correlations between methods per CpG site and we followed a previous published approach which consist in pooling all the CpG sites for the correlation analysis [57].

Follow- up study of the women 10–12 years after delivery

A 10-year follow-up of the women who attended STORK G is currently ongoing and expected to finish in 2021. The main aims are to assess the incidence of prediabetes and T2D and explore changes in risk factors for T2D (and CVD) over the last 10 years. We measure weight, height, physical inactivity, and blood pressure, and collect data on self-reported smoking, and chronic diseases/conditions. Currently, dried blood spots are biobanked and about 60% meet for fasting blood samples, and DNA from buccal swabs is being collected as well. We estimate to reach a sample of 350 women– 50% of those eligible.

Results to date

In the EPIPREG sample, we excluded women without fasting plasma glucose available in week 28±2, and those without genotype data due to low DNA concentrations or problems with DNA extraction. For Europeans, we were able to quantify DNA methylation in 99% of the eligible samples (empty wells = 2 and full plates = 1), 87.2% of the total number of Europeans participating in STORK G at week 28±2. For South Asians, methylation status could be determined in 100% of the eligible subjects, representing 87.5% of total South Asians participating in STORK G at week 28±2 (Fig 1). Hence, EPIPREG resulted in DNA methylation data of 312 Europeans and 168 South Asians.

Fig 1

Flow chart of the EPIPREG sample.

E = European, SA = South Asian.

Some characteristics of the mothers and offspring of the EPIPREG sample are shown in Table 1. When comparing the clinical characteristics of the women and their offspring with and without DNA methylation data, the average of sedentary hours and percentage of truncal fat were significantly higher in the European subjects included in EPIPREG, whereas gestational week, hours of light and moderate-intense physical activity, and 25-hydroxyvitamin D levels were higher in the excluded samples (S1 Table). The differences were small and generally followed the same trend as the overall STORK G missing data analysis [34]. In South Asians we did not detect any significant differences between the women and their offspring included in EPIPREG vs the excluded individuals (S2 Table).

Table 1

Characteristics of the EPIPREG sample.

Variable	N	Europeans, n = 312	South Asians, n = 168
Age	480	30.1 (4.6)	28.2 (4.6)
Weeks’ gestation	480	28.1 (1.2)	28.2 (1.2)
Height (cm)	480	167.4 (5.8)	159.9 (5.7)
Smoking status	476
Current		20 (6.5)	1 (0.6)
3 months pre-pregnancy		80 (25.9)	2 (1.2)
Former*		87 (28.2)	9 (5.4)
Never		122 (39.5)	155 (92.8)
Pre-pregnancy alcohol intake, n (%)	472	234 (76.2)	5 (3.0)
Pre-pregnancy BMI (kg/m2)	474	24.6 (4.9)	23.8 (4.1)
Actual BMI (kg/m2)	478	27.8 (4.7)	26.8 (4.1)
Total fat (%)	465	29.6 (9.7)	26.2 (8.2)
Truncal fat (%)	465	15.9 (5.5)	14.0 (5.4)
Systolic blood pressure (mmHg)	480	107.0 (9.6)	101.1 (8.7)
Diastolic blood pressure (mmHg)	480	68.4 (7.1)	66.1 (6.9)
First degree relative with diabetes	474	43 (14.0)	79 (47.3)
GDM (WHO2013), n (%)	480	76 (24.4)	70 (41.7)
GDM (WHO1999), n (%)	478	37 (11.9)	25 (15.1)
Fasting glucose (mmol/L)	480	4.7 (0.6)	5.0 (0.6)
2 hour glucose (mmol/L)	477	6.0 (1.4)	6.4 (1.5)
HbA1c (%)	475	5.1 (0.3)	5.3 (0.3)
Fasting Insulin (pmol/L)	474	48.0 [33.0, 70.2]	71.0 [57.0, 100.5]
Fasting C-peptide (pmol/L)	474	712.0 [560.2, 901.8]	855.5 [688.0, 1067.5]
HOMA-B	474	173.2 [151.3, 199.7]	179.1 [154.9, 207.9]
HOMA-IR	474	1.5 [1.2, 1.9]	1.8 [1.4, 2.3]
Total cholesterol (mmol/L)	480	6.4 (1.1)	6.0 (1.0)
HDL-cholesterol (mmol/L)	480	1.9 (0.4)	1.9 (0.4)
LDL-cholesterol (mmol/L)	473	3.7 (1.0)	3.3 (0.9)
Triglycerides (mmol/L)	480	2.0 (0.7)	2.0 (0.
Folate (nmol/l)	471	13.0 [9.9, 18.0]	12.0 [9.4, 16.0]
Vitamin D (μmol/L)	472	69.9 (27.8)	45.6 (22.2)
Leptin (μmol/L)	476	1599.5 [966.9, 2497.7]	2276.9 [1532.8, 3295.8]
Folate (nmol/l)	471	13.0 [9.9, 18.0]	12.0 [9.4, 16.0]
Vitamin B12 (μmol/L)	472	211.0 [169.2, 253.0]	186.0 [150.0, 232.2]
Sedentary time (hours/day)**	382	18 (1.6)	17.8 (1.7)
Light physical activity (hours/day)**	382	4.3 (1.3)	4.6 (1.3)
Moderate-intense physical activity (hours/day)**	382	1.0 [0.7, 1.5]	0.8 [0.5, 1.2]
Offspring data
Gestational age (days)	475	280.9 (11.7)	277.1 (12.6)
Female sex, n (%)	462	147 (49.0)	79 (48.8)
Birth weight (g)	472	3582.2 (529.4)	3211.7 (512.2)
Birth length (cm)	432	50.1 (2.3)	49.3 (2.2)
Neonatal sum of skinfolds (mm)	359	18.8 (4.0)	17.0 (3.6)

Data from gestational visit 2 for cross-sectional associations with DNA methylation data, otherwise specified. Data are presented in mean (SD) for normally distributed variables and median [IQR] for non-normal variables. Categorical variables are presented by frequency (%). Despite 8 individuals did not pass the QC procedure, the data of the 480 individuals are presented for informative purposes. WHO = World health organization. GDM = gestational diabetes mellitus. BMI = body mass index. HDL = high-density lipoproteins. LDL = low-density lipoproteins.

* Ex-smokers and occasional smokers who did not smoke 3 months before pregnancy,

* *Women with at least one valid day of registered physical activity (Armband).

Genetic principal components distinctly separated Europeans and South Asian women (Fig 2), which fully corresponded with the self-reported ethnicity. South Asians were separated in two groups, the largest cluster mainly consisted of Pakistani women and the smaller group of Sri-Lankan women.

Fig 2

Scatter dot plot of genetic PC1 (GPC1) and PC2 (GPC2) (n = 438).

Blue dots are Europeans, red dots are South Asians, based on self-reported ethnicity. It can be noticed that South Asians were separated in two groups, being the upper largest group mainly composed of Pakistanis and the smaller lower group of Sir Lankans.

To assess the agreement between the Infinium MethylationEPIC BeadChip and the technical validation by pyrosequencing, we used Bland-Altman plots to illustrate agreement per CpG site (Fig 3). In Fig 3, we can see that most of the samples in cg17148978, cg14120215 and cg02098128 were within the limits of agreement (LAG). For cg17148978 the average mean difference was 0.74% (lower LAG:-0.56, upper LAG: 2.05%), for cg14120215 7.56% (lower LAG: -6.95, upper LAG: -22.09%) and for cg02098128–0.66% (lower LAG:-2.09, upper LAG 0.78%). We found no evidence of proportional bias in cg17148978, cg14120215 or cg02098128, meaning that the agreement in these CpG sites was good. In cg19327414, the mean difference was 7.79% (lower LAG: -12.32, upper LAG: 27.89%), and there was a significant proportional bias (Beta = -1.85, pval = <0.001). All the samples but one were within the lines of agreement. In the correlation analysis, the overall correlation when pooling all 4 CpG sites was high (r = 0.98, p<0.001) (Fig 4), while site specific correlations for cg17148978 (R = -0.23, p = 0.22), cg19327414 (R = 0.18, p = 0.40), cg14120215 (R = 0.18, p = 0.33) and cg02098128 (R = 0.30, p = 0.11) were weak.

Fig 3

Bland-Altman plots showing the mean difference between methods (y-axis) versus the mean between methods (x-axis) for each CpG site tested for technical validation.

Fig 4

Scatter plot showing the relationship between the DNAm values of the four selected CpG sites quantified with the Infinium MethylationEPIC BeadChip (x-axis) and with bisulphite pyrosequencing (BSP) (y-axis).

Dots colour code: Pink: cg02098128, green: cg17148978, blue: cg14120215, orange: cg19327414.

Strengths and limitations

EPIPREG is a population-based sample of European and South Asian pregnant women with epigenome-wide DNA methylation in maternal peripheral blood leukocytes. EPIPREG has detailed phenotype data from both the mother and the offspring, as well as genotype data. DNA methylation measured with the epigenome-wide chip showed high agreement with bisulphite pyrosequencing.

The inclusion of women with both European and South Asian ethnic background enables interesting studies into the role of DNA methylation in ethnic disparities in health. Furthermore, EPIPREG has both genome-wide genotype and DNA methylation data also allowing for methylation quantitative trait loci (mQTL) analysis.

Regarding the technical validation, correlations were low for each CpG site separately. However, correlations could be misleading for agreement analyses as they mainly measure the linearity of the variables irrespective of the data’s shape [58], and are sensitive to the range of values—the broader the range, the higher the correlation coefficient [59]. Bland-Altman plots are considered a better test of agreement between methods [59], especially when the range of values is low as for three of our four CpG sites. Therefore although our individual sites showed low correlation between the two methods, they had high agreement.

EPIPREG’s population-based design and comprehensive phenotyping allows for gaining representative data about the associations between DNA methylation and a wide range of phenotypic traits, exposures and outcomes. Hence, a major advantage of the cohort is the availability of several maternal phenotypes collected in gestational weeks 15 and 28, and 3 months after delivery. Furthermore, there is an ongoing follow-up in some women 10 to 12 years after pregnancy. In the offspring we have anthropometric data recorded in utero, at birth and during the first four years of life, as well as serum cord blood and placental tissue biobanked. Lastly, we have permission for linkage with Norwegian national registries using the personal identification number.

Since EPIPREG has a moderate sample size, our study has limited statistical power for EWAS, nevertheless our broad availability of phenotypes will allow us to perform several DNA methylation-phenotype association analyses. Also, our sample is well suited for meta-analysis efforts, or to serve as a replication cohort. Another limitation is that DNA methylation is only measured in gestational week 28±2.

Collaboration

EPIPREG may serve as a useful sample for generation of new hypotheses about associations between DNA-methylation and phenotypic traits relating to GDM, for replication of findings from other studies, and for meta-analysis efforts. We are currently welcoming collaborations with cohorts with similar data and researchers interested in collaboration are welcome to contact Christine Sommer, or visit our website: www.epipreg.no.

Mean comparison between the European samples included in the Infinium MethylationEPIC BeadChip versus their respective excluded samples.

(XLSX)

Click here for additional data file.

Mean comparison between the South Asian samples included in the Infinium MethylationEPIC BeadChip versus their respective excluded samples.

(XLSX)

Click here for additional data file.

24 May 2021

PONE-D-21-08932

Cohort Profile: Epigenetics in Pregnancy (EPIPREG) – population-based sample of European and South Asian pregnant women with epigenome-wide DNA methylation (850k) in peripheral blood leukocytes.

PLOS ONE

Dear Dr. Sommer,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

Please submit your revised manuscript by Jul 05 2021 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Lee-Ling Lim

Academic Editor

PLOS ONE

Additional Editor Comments:

Please address all reviewer comments.

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. Please include additional information regarding the survey or questionnaire used in the study and ensure that you have provided sufficient details that others could replicate the analyses. For instance, if you developed a questionnaire as part of this study and it is not under a copyright more restrictive than CC-BY, please include a copy, in both the original language and English, as Supporting Information.

3. We note that you are reporting an analysis of a microarray, next-generation sequencing, or deep sequencing data set. PLOS requires that authors comply with field-specific standards for preparation, recording, and deposition of data in repositories appropriate to their field. Please upload these data to a stable, public repository (such as ArrayExpress, Gene Expression Omnibus (GEO), DNA Data Bank of Japan (DDBJ), NCBI GenBank, NCBI Sequence Read Archive, or EMBL Nucleotide Sequence Database (ENA)). In your revised cover letter, please provide the relevant accession numbers that may be used to access these data. For a full list of recommended repositories, see http://journals.plos.org/plosone/s/data-availability#loc-omics or http://journals.plos.org/plosone/s/data-availability#loc-sequencing.

4. Thank you for stating the following in the Financial Disclosure section:

[EPIPREG is supported by the South Eastern Norway Regional Health Authority (grant number: 2019092), and the Norwegian Diabetes Association (grant number: N/A). G.H.M. is supported by the Norwegian Research Council (Post doctoral mobility research grant 287198), and have received funding support by Nils Normans minnegave (grant number: N/A)].

We note that you received funding from a commercial source: Nils Normans minnegave

Please provide an amended Competing Interests Statement that explicitly states this commercial funder, along with any other relevant declarations relating to employment, consultancy, patents, products in development, marketed products, etc.

Within this Competing Interests Statement, please confirm that this does not alter your adherence to all PLOS ONE policies on sharing data and materials by including the following statement: "This does not alter our adherence to PLOS ONE policies on sharing data and materials.” (as detailed online in our guide for authors http://journals.plos.org/plosone/s/competing-interests).  If there are restrictions on sharing of data and/or materials, please state these. Please note that we cannot proceed with consideration of your article until this information has been declared.

Please include your amended Competing Interests Statement within your cover letter. We will change the online submission form on your behalf.

Please know it is PLOS ONE policy for corresponding authors to declare, on behalf of all authors, all potential competing interests for the purposes of transparency. PLOS defines a competing interest as anything that interferes with, or could reasonably be perceived as interfering with, the full and objective presentation, peer review, editorial decision-making, or publication of research or non-research articles submitted to one of the journals. Competing interests can be financial or non-financial, professional, or personal. Competing interests can arise in relationship to an organization or another person. Please follow this link to our website for more details on competing interests: http://journals.plos.org/plosone/s/competing-interests

5. Please include captions for your Supporting Information files at the end of your manuscript, and update any in-text citations to match accordingly. Please see our Supporting Information guidelines for more information: http://journals.plos.org/plosone/s/supporting-information.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Partly

Reviewer #2: Yes

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: No

Reviewer #2: No

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: This paper describes a new cohort with pregnancy data on a set of women from two different ethnic groups.

The cohort is of interest however I am disappointed that the authors have not performed any basic analysis with any of the cardio-metabolic traits or other pregnancy related traits. Moreover, none of the methods (experimental nor statistical analysis) are novel therefore this is merely a report of a new collected cohort with some description of demographics.

Reviewer #2: This is an excellent presentation of a very interesting cohort. I have some minor comments which may help to improve the manuscript content.

The authors state several times that the cohort is of a large size although they do also say they have limited statistical power in their limitations section. I would argue the cohort is of modest size, however the quantity and range of data collected on each mother-child pair is sizable and the repeated measures are a major advantage to the cohort design.

Within the introduction the paper would benefit from a justification of why there is value in a cohort collection with 2 ethnic groups.

Additional minor comments:

Suggest referring to the EPIC array as Infinium MethylationEPIC BeadChip kit rather than just MethylationEPIC kit to make the resource more discoverable.

In the introduction there are other citations that the authors could consider including and which support the content presented. For example, there are other large studies showing the association between T2D and methylation eg Juvinao-Quintero DL et al (DOI: 10.1186/s13148-021-01027-3) and some studies which have a multi-ethnic study design relevant to EPIPREG eg Chambers JC et al (doi: 10.1016/S2213-8587(15)00127-8). For smoking citations (#16 or #17) it might be also useful to include eg Joehanes R et al (10.1161/CIRCGENETICS.116.001506), Wiklund P et al (DOI: 10.1186/s13148-019-0683-4) and/or Joubert et al (DOI: 10.1289/ehp.1205412). For alcohol intake there are other published papers that are relevant eg Dugue AP (blood)(DOI: 10.1111/adb.12855) and Xu K (although this is a study in saliva not blood)(DOI: 10.1111/acer.14168). Citation #8 is a review so should be acknowledged as such.

“The population based design and inclusion of a significant number of women with European and South Asian ethnicity allows us to study a wide range of phenotypes.” I would suggest removing the words “significant number” from this sentence as it is subjective.

The introduction is clearly written. It could however be improved by discussing the rationale for inclusion of women of European and South Asian ancestries.

In the “Study Population” section it would be more informative to report the specific participation rates in the two ethnic groups in EPIPREG rather than the range across all groups in STORK G.

Line 134: “…if the last was born outside Europe” suggest changing to “if the latter was born outside Europe”

Line 164: A citation to the Oxford University HOMA Calculator should be included.

Line 173: “intense?” typo ?

Line 186 (and 189): suggest amending “Formalin-fixated- paraffin embedded blocks” to “Formalin-Fixed Paraffin-Embedded (FFPE) blocks”

Line 215: for imputation of genetic data it isn’t clear if imputation was conducted in each ethnic group separately or not (or what reference panel was used).

Line 246: typo “QUIAGEN” -> “QIAGEN”

Line 247: suggest replacing the work “doublet” with “duplicate” since this is a more common way of describing replicates.

Line 248: It is unclear what the unmethylated and methylated controls are. Presumably these are commercially available samples(?) It needs to be made clear the negative control is a sample containing no DNA template (if this is the case).

Line 298: The R2 of 0.98 is misleading as it doesn’t actually tell us anything about correlation between methods on a per CpG basis. There should be 4x R2 measures reported, one for each CpG tested which would give an estimate of agreement between methods for each CpG.

Line 309-310: There are a number of other studies studying epigenetics perinatally (eg members of the PACE consortium https://www.niehs.nih.gov/research/atniehs/labs/epi/pi/genetics/pace/index.cfm) and cohorts who also have South Asian and European ancestry maternal and offspring samples eg Born in Bradford: https://borninbradford.nhs.uk/). I would argue that in terms of sample size EPIPREG is relatively small compared to some of these other cohorts.

Figure 2: There appears to be some population stratification which is particularly noticeable in the South Asian group (two groups are evident on both the 1st and 2nd PCs). It would be useful to comment if this can be explored further.

Figure 3: This figure isn’t very informative given that the CpG sites tested have very different distributions. A 4 panel figure showing the R2 for each of the 4 CpG sites would show the relationship between the two methods tested much more clearly. This plot could also be improved if the scales were labelled the same (ie both 0-1 or 0-100) and the labels were descriptive (eg “% methylation measured by bisulphite pyrosequencing” instead of “BSP”).

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #2: Yes

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

3 Jul 2021

Response to reviewers

Reviewer #1:

This paper describes a new cohort with pregnancy data on a set of women from two different ethnic groups.

The cohort is of interest however I am disappointed that the authors have not performed any basic analysis with any of the cardio-metabolic traits or other pregnancy related traits. Moreover, none of the methods (experimental nor statistical analysis) are novel therefore this is merely a report of a new collected cohort with some description of demographics.

Reply = Thank you for your comment! We did not include any association analyses of cardio-metabolic related traits in this manuscript, since the aim of this paper is to present the EPIPREG sample with epigenetics, genetics and a variety of interesting phenotypes. We noticed that this aim was only mentioned in the abstract, but have now clarified it and added the aim also to the introduction (lines 111-113 in Manuscript with track changes).

Reviewer #2:

This is an excellent presentation of a very interesting cohort. I have some minor comments which may help to improve the manuscript content.

The authors state several times that the cohort is of a large size although they do also say they have limited statistical power in their limitations section. I would argue the cohort is of modest size, however the quantity and range of data collected on each mother-child pair is sizable and the repeated measures are a major advantage to the cohort design.

Reply = Thanks for pointing this out, we have corrected the terminology and stated that the cohort has a moderate sample size throughout the manuscript.

Within the introduction the paper would benefit from a justification of why there is value in a cohort collection with 2 ethnic groups.

Reply = We have added a few lines to the introduction highlighting the importance of having two ethnic groups (lines 103-108).

Additional minor comments:

Suggest referring to the EPIC array as Infinium MethylationEPIC BeadChip kit rather than just MethylationEPIC kit to make the resource more discoverable.

Reply= The modification was done

In the introduction there are other citations that the authors could consider including and which support the content presented. For example, there are other large studies showing the association between T2D and methylation eg Juvinao-Quintero DL et al (DOI: 10.1186/s13148-021-01027-3) and some studies which have a multi-ethnic study design relevant to EPIPREG eg Chambers JC et al (doi: 10.1016/S2213-8587(15)00127-8). For smoking citations (#16 or #17) it might be also useful to include eg Joehanes R et al (10.1161/CIRCGENETICS.116.001506), Wiklund P et al (DOI: 10.1186/s13148-019-0683-4) and/or Joubert et al (DOI: 10.1289/ehp.1205412). For alcohol intake there are other published papers that are relevant eg Dugue AP (blood)(DOI: 10.1111/adb.12855) and Xu K (although this is a study in saliva not blood)(DOI: 10.1111/acer.14168). Citation #8 is a review so should be acknowledged as such.

Reply= Thanks for recommending these studies; they were added in the introduction, but Xu study was not included since we limited our scope to blood. Since we mainly used research papers for this part in the introduction, in citation 8 we replaced the review with a recently published EWAS of GDM (https://doi.org/10.2337/dc20-2960).

“The population based design and inclusion of a significant number of women with European and South Asian ethnicity allows us to study a wide range of phenotypes.” I would suggest removing the words “significant number” from this sentence as it is subjective.

Reply= As suggested, we removed “of a significant number” in all the manuscript

The introduction is clearly written. It could however be improved by discussing the rationale for inclusion of women of European and South Asian ancestries.

Reply= Thank you! We have added it to the introduction (lines 103-108).

In the “Study Population” section it would be more informative to report the specific participation rates in the two ethnic groups in EPIPREG rather than the range across all groups in STORK G.

Reply= The participation rates of Europeans and South Asians that participated in STORK G has been added (line 126-127).

Line 134: “…if the last was born outside Europe” suggest changing to “if the latter was born outside Europe”

Reply= Corrected

Line 164: A citation to the Oxford University HOMA Calculator should be included.

Reply= The citation was added

Line 173: “intense?” typo ?

Reply= Typo erased

Line 186 (and 189): suggest amending “Formalin-fixated- paraffin embedded blocks” to “Formalin-Fixed Paraffin-Embedded (FFPE) blocks”

Reply= Corrected

Line 215: for imputation of genetic data it isn’t clear if imputation was conducted in each ethnic group separately or not (or what reference panel was used).

Reply= We have now added that the imputation was done separately in EUR and SA, and that we used the 1000 genome panels specific for Europeans and South Asians populations (Lines 230-237).

Line 246: typo “QUIAGEN” -> “QIAGEN”

Reply= Typo corrected

Line 247: suggest replacing the work “doublet” with “duplicate” since this is a more common way of describing replicates.

Reply= Doublet was replaced with duplicate

Line 248: It is unclear what the unmethylated and methylated controls are. Presumably these are commercially available samples(?) It needs to be made clear the negative control is a sample containing no DNA template (if this is the case).

Reply= Specifications of the control used were added (Lines 271-274).

Line 298: The R2 of 0.98 is misleading as it doesn’t actually tell us anything about correlation between methods on a per CpG basis. There should be 4x R2 measures reported, one for each CpG tested which would give an estimate of agreement between methods for each CpG.

Reply= Thank you for your comment! We originally followed the approach from Ronn et al 2013 (https://doi.org/10.1371/journal.pgen.1003572), where they pooled their sites together for the correlation analysis. As suggested we have added correlation per CpG site. We have also added Bland-Altman plots as presented by Mamtani and colleagues (https://doi.org/10.1186/s13148-016-0173-x), and regressed the mean difference and the average of EPIC and pyrosequencing values to asses if there were proportional bias. We have now added some lines to the methods and results (lines 276-281 and 334-347).

Correlations indeed have some limitations in statistical testing of agreement, especially when there is a small range of values/small variation as we have for 3 of 4 CpG sites, https://doi.org/10.1159/000337798), while Bland-Altman plots illustrate agreement better (https://doi.org/10.1038/ki.2008.306). We have added a few lines to the Strengths and limitations section discussing the results (lines 361-367).

Line 309-310: There are a number of other studies studying epigenetics perinatally (eg members of the PACE consortium https://www.niehs.nih.gov/research/atniehs/labs/epi/pi/genetics/pace/index.cfm) and cohorts who also have South Asian and European ancestry maternal and offspring samples eg Born in Bradford: https://borninbradford.nhs.uk/). I would argue that in terms of sample size EPIPREG is relatively small compared to some of these other cohorts.

Reply= We have changed the wording to moderate throughout the paper.

Figure 2: There appears to be some population stratification which is particularly noticeable in the South Asian group (two groups are evident on both the 1st and 2nd PCs). It would be useful to comment if this can be explored further.

Reply=Thanks, for pointing this out! We are aware and have added a few lines mentioning the reason for the stratification (lines 329-331).

Figure 3: This figure isn’t very informative given that the CpG sites tested have very different distributions. A 4 panel figure showing the R2 for each of the 4 CpG sites would show the relationship between the two methods tested much more clearly. This plot could also be improved if the scales were labelled the same (ie both 0-1 or 0-100) and the labels were descriptive (eg “% methylation measured by bisulphite pyrosequencing” instead of “BSP”).

Reply=We have added a four-panel figure showing the agreement between the chip and pyrosequencing using Bland-Altman plots.

Submitted filename: Response to Reviewers.docx

Click here for additional data file.

2 Aug 2021

Cohort Profile: Epigenetics in Pregnancy (EPIPREG) – population-based sample of European and South Asian pregnant women with epigenome-wide DNA methylation (850k) in peripheral blood leukocytes.

PONE-D-21-08932R1

Dear Dr. Sommer,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Lee-Ling Lim

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #2: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #2: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #2: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #2: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #2: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #2: The authors have made revisions which satisfy the comments/questions raised in my original review. I have no further comments and recommend this paper for publication.

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #2: No

5 Aug 2021

PONE-D-21-08932R1

Cohort Profile: Epigenetics in Pregnancy (EPIPREG) – population-based sample of European and South Asian pregnant women with epigenome-wide DNA methylation (850k) in peripheral blood leukocytes

Dear Dr. Sommer:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Lee-Ling Lim

Academic Editor

PLOS ONE