Study design and participants

This study was based on a prospective birth cohort among Chinese Han population, which was carried out at Fuxing Hospital affiliated to Capital Medical University in Beijing, China. In this cohort, 1050 healthy mother-infant pairs were recruited from Jan 2018 to Dec 2018 at gestational 6–8 weeks when they established their pregnancy files according to the stringent inclusion and exclusion criteria. Specifically, the eligible women were healthy at 20–40 years old and intended to deliver in this hospital without smoking and drinking history. Conversely, the pregnant women with gestational hypertension, hypothyroidism, diabetes, heart diseases, hepatitis, cirrhosis, severe fatty liver, nephritis, severe anemia, leukemia, malignant tumor of immune system and hemorrhagic diseases, medical history of using antibiotics and thyroxine related drugs, the individuals without colostrum after delivery, and their infants with birth defects, genetic and metabolic diseases, ischemia, and hypoxia were all excluded from this study. Besides, the participants who were intermittently given exogenous DHA-rich n-3 PUFAs supplementation were also rejected to ensure the consistency of the data.

This study was approved by the Ethics Committee of Beijing Pediatric Research Institution, Beijing Children's Hospital affiliated to Capital Medical University (No: 2016–08), which was also registered at the website of http://www.chictr.org.cn/show proj.aspx?proj = 4673 (No: ChiCTR-OCH-14004900). Meanwhile, all subjects were granted both the written informed consents and Health Insurance Portability and Accountability Act Authorization after they were clearly informed the significance of this survey by the trained investigators.

Basic information and food frequency questionnaires

The basic information questionnaires, including maternal age, height, pre-pregnancy weight, pre-pregnancy body mass index (BMI), prenatal weight, prenatal BMI, gestational age, mode of delivery and infant sex, were respectively performed among the participants by the trained investigators through face to face method when they were involved in this study. Meanwhile, the anthropometric measurements (lenght and weight) were collected at birth by the pediatricians using the nearest millimeter (HW-1000HW-2000, China) on a digital board with three repeated samples to calculate the mean values when they were delivered. Then the head circumference was measured by the tapes.

The food frequency questionnaires (FFQ) were used to determine the total DHA intake both from the dietary and exogenous DHA-rich n-3 PUFAs supplementation at gestational 12 (early pregnancy), 27 (middle pregnancy) and 36 week (late pregnancy). Exactly, the pregnant women were asked to obtain the dietary specified averagely amounts and types of foods with DHA-rich n-3 PUFAs, which were mainly included the deep-sea fish (yellow croaker, codfish, trichiurus haumela, sardine, sockeye salmon, tynny, saury, muraenesox, cinereus and braised flatfish head), algae, nuts, egg yolk and various seed oils (flax seed oil, peril-la seed oil and walnut oil). In this process, nine possible frequency categories were ranged from never/almost never to ≥ 6 times per week to calculate the total dietary DHA intake using the SY-2 Nutrition Analysis Software. Meanwhile, the open-ended questions were asked to collect and calculate the exogenous DHA-rich n-3 PUFAs supplementation, including their usual time, brands, doses and types according to the commercial specifications. Therefore, total DHA intake during the pregnancy was the sum from the dietary and exogenous DHA-rich n-3 PUFAs supplementation. Then, the subjects were respectively divided into four groups according to the firstly gestational weeks of exogenous DHA-rich n-3 PUFAs supplementation, which were exogenous DHA-rich n-3 PUFAs supplementation groups from the early (0–12 week, S1 group), middle (13–27 week, S2 group) and late pregnancy (> 27 week, S3 group), with non-exogenous DHA-rich n-3 PUFAs supplementation during the whole pregnancy as the control group.

Blood and colostrum collection

1 mL venous blood was obtained from the pregnant women for DNA extraction when they were recruited into this study. Meanwhile, they were asked to collect 5 mL colostrum within postpartum 3–5 days. All biological samples were immediately transported to the laboratory using the dry ice and stored at -80℃ refrigerator until the usage.

Profiles of PUFAs in the colostrum

The profiles of PUFAs were measured by the gas chromatography (GC) after they were extracted from 100 μL colostrum with 2 mL mixture of chloroform and methanol (1:9), containing 0.001% butylated hydroxytoluene (BHT) and 2μL pentadecanoic acid standard (C15:0, cat. no. P6125, Sigma-Aldrich Chemie GmbH, Germany, 0.15 g/mL ethanol). The above mixture was heated at 100 °C for 1 h, while 5 mL 6% K2CO3 solution and 200 μL n-hexane were added to these tubes, mixed on a vortex, and centrifuged at 3000 rpm/min for 15 min (10℃). Then the clear n-hexane top layer was transferred to the GC auto sampler vials to determine the profiles of PUFAs using the Agilent 6890 N GC with the flame ionization detection (P/N 19091J-433, HP-5 capillary column was 30 m × 0.32 mm × 0.25 μm). All measurements were done in the duplicate and reported as the averages, in which the profiles of PUFAs were expressed as the proportions of fatty acid/all fatty acids.

DNA extraction from the venous blood

The whole DNA was extracted from 200 mL venous blood using the DNA extraction kits according to manufacturer’s instructions (cat. no. DP348-03, Tiangen Biotech (Beijing) Co. Ltd, China). The quantity and quality of the purified DNA samples were respectively evaluated by the ND-1000 spectrophotometer (NanoDrop 2000C, Thermo Fisher Scientific, CN) and agarose gel electrophoresis.

Selection of the SNPs and their genotypes

All 26 SNPs were selected in the genes of Fads1, Fads2, Fads3, Elvol2 and Elvol5, which were sited at the potential 5′ or 3′ regulatory regions in the strong linkage disequilibrium (LD) to enable their genotypes (r2 > 0.7) [9, 19,20,21,22,23,24,25,26,27,28,29,30]. As shown in the Additional file 1: Tables S1, S2, the genotypes of rs174448 (A/G, 11:61,872,101), rs174537 (G/T, 11:61,785,208), rs174550 (T/C, 11:61,804,006) and rs174553 (G/A, 11:61,807,686) in the Fads1, rs174598 (A/T, 11:61,853,722), rs174602 (C/T, 11:61,856,942), rs174609 (T/C, 11:61,860,339), rs174619 (G/A, 11:61,862,194), rs498793 (C/T, 11:61,857,233) and rs3168072 (A/T, 11:61,864,038) in the Fads2, rs174455 (C/T, 11:61,888,710), rs174464 (G/A, 11:61,890,454) and rs76996928 (T/C, 11:61,883,454) in the Fads3, rs1323739 (G/C, 6:11,004,328), rs2295602 (C/T, 6:11,005,609), rs2180725 (T/C, 6:11,025,187), rs3798710 (G/C, 6:11,002,550) and rs78793420 (T/C, 6:11,039,168) in the Elvol2, and rs209512 (A/G, 6:53,338,779), rs2281274 (T/C, 6:53,278,756), rs2294852 (C/G, 6:53,292,416), rs2397142 (C/G, 6:53,335,501), rs6909592 (G/C, 6:53,298,292), rs9349665 (T/C, 6:53,341,701), rs9395858 (C/T, 6:53,313,689) and rs12207094 (A/T, 6:53,339,377) in the Elvol5 were associated with the strongest GWAS signals for up to 20% of their variations using the International HapMap Project SNP database and NCBI database (http://www.ncbi. nlm.nih.gov/snp).

All above 26 SNPs were measured by the Matrix-Assisted Laser Desorption Ionization Time of Flight Mass Spectrometry (MALDI-TOF-MS, Sequenom) with the validated primers in the Additional file 1: Table S1, in which the genotyping success rate was more than 97% and respectively presented in the Chromosome 11 (Fads1, Fads2 and Fads3) and 6 (Elovl2 and Elovl5). Their minor allele frequencies (MAF) were ranged from 0.138 to 0.441. Then their genotype analysis was performed using the graphical Java interface of THESIAS software package (http://ecgene.net/ genecanvas). Meanwhile, the program was based on the maximum likelihood model linked to the SEM algorithm using the statistically reconstruct genotypes in the unrelated individuals and then investigated the effects of covariate adjusted genotypes as well as their interactions between maternal DHA-rich n-3 PUFAs supplementation and the related significant genotypes of 26 SNPs on the profiles of PUFAs among all 1050 participants according to the Hapmap database (http://hapmap.ncbi.nlm.nih.gov).

Statistical analysis

The statistical analysis was carried out using the SPSS 21.0, in which P < 0.05 was recognized as the significant importance. Exactly, the normal distribution of outcome variables was evaluated both by the Kolmogorov–Smirnov test and distributional Q-Q plot. All data was expressed as mean (geometric mean) ± standard error (SE) and mean changes with respect to the baseline. The differences of the general characteristics were detested with one-way analysis of variance (ANOVA) and χ2 test. The distributions of all 26 SNPs were determined for the deviation from Hardy–Weinberg equilibrium using the χ2 test. Inter-locus LD based on the observed numbers of the genotypes in the Fads1, Fads2, Fads3, Elvol2 and Elvol5 was established using the CubeX software (http://www.oege.org/software/cubex/). Then the genetic data was analyzed for each SNP separately and categorized as the homozygous for the major allele by changing to the numerical variables. Meanwhile, the linear regression was used to assess the known 26 SNPs and maternal DHA-rich n-3 PUFAs supplementation on the profiles of PUFAs. Furthermore, the interactions between maternal DHA-rich n-3 PUFAs supplementation and the differently related genotypes of SNPs on the profiles of PUFAs were explored using the partial regression analysis by adjusting the maternal age, gestational week, mode of delivery, infant sex and BMI at birth.