Expression of genes involved in inflammation and growth – does sampling site in human full-term placenta matter?

Subjects

We included 12 women in this study. For estimation of the sample size needed, we reviewed previous publications on the subject [10], [11]. One study found differences in gene expression according to sampling site using 10 placentas [10] and another using six placentas [11]. The women in this study gave birth between May 2013 and April 2015, all with Nordic ethnicity. To be included in the study, the woman had to be healthy, have normal weight [body mass index (BMI) 18.5–25 kg/m²] and gain weight during the pregnancy within the recommendation of the Institute of Medicine (11–16 kg) [16]. Furthermore, the pregnancy had to be uncomplicated and singleton. The delivery had to be vaginal as placental gene expression seems to be affected by delivery mood [15], [17]. It also had to be uneventful and spontaneous at full term (38–42 weeks of gestation). The newborn baby had to be healthy with an Apgar score above 7 at 5 min of age and without any signs of malformation. The infant’s birth weight was not restricted in the study. The newborns’ birth weight and length z-scores were calculated using a Swedish reference population [18]. We included an equal number of female and male placentas. Women who smoked during the pregnancy were not eligible for the study. Written informed consent was obtained from all participating women, and the study was approved by the Ethical Review Board in Uppsala, Sweden (2012/531).

Placental tissue sample collection

Only placentas with a centrally inserted umbilical cord (i.e. located within the inner third of the placental disk radius) and placentas without any macroscopic abnormalities were included in the study. The same person (MA) collected all samples. She was present at the delivery of all children included in the study to avoid delayed collection of tissue samples as this may impact gene expression [15]. Thus, all samples were taken immediately after delivery. The samples were taken from nine well-defined sites in each placenta as described previously by Wyatt et al. [11] (Figure 1). To obtain these samples, a triangular segment of each placenta was excised with its base located 2 cm from the lateral edge and its apex near the cord insertion. The decidual layer along with the basal plate, as well as the chorionic surface and membranes, were removed. The samples, approximately 0.5 g each, were rinsed in saline (Gibco, Life Technologies, Stockholm, Sweden) to remove blood and stored at −80°C in RNAlater (Ambion, Stockholm, Sweden) until analysis.

Figure 1:

Locations of sampling sites 1A–3C within the placenta.

A=mean of 1A, 2A, 3A; B=mean of 1B, 2B, 3B; C=mean of 1C, 2C, 3C. 1=mean of 1C, 1B, 1A; 2=mean of 2C, 2B, 2A; 3=mean of 3C, 3B, 3A. Original artwork by Wyatt et al. [11] reproduced with permission.

RNA extraction and cDNA synthesis

The procedures for RNA preparation and cDNA synthesis have been described previously [19]. Fifteen to 20 mg of placental tissue was homogenized in batches using TissueLyser LT (Qiagen, Nordic, Sollentuna, Sweden) for 4 min at 30 Hz and then placed in −80°C until RNA isolation with Allprep DNA/RNA/protein Mini Kit, (Qiagen) was completed according to the manufacturer’s instructions. The RNA concentrations were measured by a Thermo Scientific NanoDrop 2000 (Fisher Scientific, Göteborg, Sweden). cDNA was synthesized from 100 ng of RNA using a High Capacity cDNA Reverse Transcription Kit (Applied Biosystems, Foster City, CA, USA) according to the manufacturer’s instructions. For each sample, four cDNA reactions were performed. They were pooled and stored at −20°C until further analysis.

Quantitative real-time PCR and choice of reference gene

Quantitative real-time PCR (qPCR) was performed using a 7900HT Fast Real-Time polymerase chain reaction (PCR) System (Applied Biosystems, Foster City, CA, USA). The probes used and their assay IDs are presented in Table 1. The PCR reagents and the well plates were obtained from Applied Biosystems. Samples from the same placenta were run in duplicate with a TaqMan^® Fast Universal PCR Master Mix (2×), no AmpErase^® UNG (Applied Biosystems™) in 384-well plates. Crossing threshold (Ct) values were calculated by 7900HT Fast Real-Time PCR System software using the second-derivative maximum method. Transcription of reference genes may differ between clinical groups [21]. Cytochrome c-1 (CYC1), glyceraldehyde-3-phosphate dehydrogenase (GAPDH), tyrosine 3-monooxygenase/tryptophan 5-monooxygenase activation protein, zeta polypeptide (YWHAZ), and topoisomerase 1 (TOP1) were selected as possible reference genes. The expression stability value (Table 1) was calculated using the “NormFinder” algorithm [20], and CYC1 was chosen for normalization of the genes of interest as it showed the best stability value in our samples. The ΔCt method was used to normalize the expression level of the gene of interest to the reference gene CYC1 [22]. The ΔΔCt method was used to determine fold change in gene expression normalized to CYC1 and relative to a control site [22]. Ct values reported as >35 or as “not detected” were considered to be negative, according to the manufacturer’s recommendation.

Statistical analysis

Data are given as the mean±standard error of the mean (SEM) or as proportions (%). To reduce the number of comparisons needed, the nine sampling sites were collected into three variables (r₁, r₂ and r₃). This accomplished in two steps. First, the mean values of the expression at the horizontal and the vertical axes were calculated. This gave rise to three horizontal mean values (sites A, B and C) and three vertical mean values (sites 1, 2 and 3), where site A represents the mean basal expression, site B represents the mean middle horizontal expression and site C represents the mean chorionic expression. Site 1 represents the mean central expression, site 2 represents the mean middle vertical expression, and site 3 represents the mean lateral expression (see Figure 1). Three transverse values also exist: the expression at sites 1A, 2B and 3C (see Figure 1). Second, the r value was calculated using Pearson’s correlation test. r₁ is the correlation between the values at sites 1, 2 and 3 (vertical location). r₂ is the correlation between the values at sites A, B and C (horizontal location) and r₃ is the correlation between the expression value at 1A, 2B and 3C (transverse location). This was performed for every normalized gene of interest in each placenta. Finally, r₁ was compared with r₂, r₂ was compared with r₃, and r₁ was compared with r₃ using a Wilcoxon signed ranks test. Statistical analyses were performed with SPSS version 22 (IBM Corporation, Armonk, NY, USA) and the level of statistical significance was set at a two-sided P<0.05.

Clinical characteristics

The clinical characteristics of the study subjects, divided into female and male newborns, are summarized in Table 2. The mean birth weight was approximately 200 g higher in girls than in boys, and a trend towards higher placental weight was seen in the female group of children. However, these differences did not reach statistical significance. Neither birth weight nor birth weight z-score correlated with placental weight (r=0.063, P=0.845 and r=0.093, P=0.778, respectively).

The gene expression of inflammatory substances and growth factors according to sample site

The expression of all genes was successfully analyzed and no Ct values exceeded 35. The coefficients of variation (CVs) were below 11% for all transcripts. High CVs were obtained for leptin expression, with a mean CV of 7%. The mean CVs for the other transcripts were 2% or below. The expression of the leptin and the IR gene were high at sites 2C and 3C and differed significantly by sampling site (for leptin: r₁ vs. r₂, P=0.028, and for IR: r₂ vs. r₃, P=0.041). The comparison between r₂ and r₃ almost reached statistical significance for leptin (P=0.050) (Figure 2). In one of the placentas, the leptin expression was very high at site 2C (Figure 3). However, the difference between r₁ and r₂ in leptin expression was still statistically significant when that outlier was excluded from the calculations (P=0.028). Furthermore, the difference between r₂ and r₃ in leptin expression reached statistical significance when the outlier was omitted (P=0.019).

Figure 2:

The gene expression of inflammatory and growth mediators at nine different placental sampling sites (1A–3C according to Figure 1, n=12).

The expression at each site is given as the mean fold change (±SEM) in relation to the expression at site 1A. r₁, the trend of gene expression from the placental medial to lateral sites; r₂, the trend of gene expression from the placental basal to chorionic sites; r₃, the trend of placental transversal gene expression at the sampling sites 1A, 2B, and 3C. The comparisons between the trends (r-values) were performed with the Wilcoxon signed ranks test. *P<0.05.

Figure 3:

Heatmaps visualizing the expression of leptin and IR genes at nine sampling sites (1A–3C) in 12 different placentas.

Rows, different placentas. Columns, different sites. F, female, M, male. Birth weights in grams. Birth lengths in centimetres. Placental weights in grams. Colour: red, higher gene expression, green, lower gene expression. Gene expression related to the reference gene cytochrome c-1 (CYC1).

When separating data by sex, gene expression of leptin showed significant difference between r₁ and r₂ only in female placentas (P=0.028). When the outlier with high leptin expression at site 2C was excluded, a significant difference in leptin expression was seen also between r₂ and r₃ in female placentas (P=0.046), but not in male placentas. Gene expression of IR showed a trend towards different expression when comparing r₂ vs. r₃ in male placentas (P=0.075). No significant difference in gene expression of IR between r₁, r₂ or r₃ was seen in female placentas.

The expression of the other genes did not differ significantly by site (Figure 2). The heat maps of the expression profiles of leptin and IR showed that in addition to differing expression by site, low or high expression generally occurred in different placentas (Figure 3).