Human subjects

Women with pregnancies complicated by preeclampsia (n = 43) and women with normal pregnancies (n = 59) were recruited at St. Olav’s University Hospital (Trondheim, Norway) and Haukeland University Hospital (Bergen, Norway) from 2002 through 2006. Preeclampsia was defined as persistent hypertension (blood pressure of ≥140/90 mm Hg) plus proteinuria (≥0.3 g/L or ≥1+ by dipstick) developing >20 weeks of pregnancy. Due to tissue sampling procedures, only pregnancies delivered by cesarean section were included. Women with preeclamptic pregnancies had cesarean section performed for medical indications, whereas women with normal pregnancies underwent cesarean section for reasons considered irrelevant to the aim of the study (eg, breech presentation, cephalopelvic disproportion in previous delivery, and fear of vaginal delivery). None of the included mothers were in labor prior to cesarean section. Exclusively healthy women with no history of preeclampsia were accepted in the normal pregnancy group. Multiple pregnancies, pregnancies with chromosomal aberrations, fetal and placental structural abnormalities, or suspected perinatal infections were excluded from both study groups. The study was approved by the Norwegian Regional Committee for Medical Research Ethics. Informed consent was obtained from all participants prior to collection of decidual samples.

Decidual tissue collection

Samples of decidua basalis tissue were obtained by vacuum suction of the placental bed, a procedure that allows the collection of tissue from the whole placental bed. Collected samples were flushed with saline solution to remove excessive blood. The decidual tissue was immediately submerged in RNA-later (Ambion, Austin, TX).

Total RNA isolation

Total RNA was isolated using a TRIzol (Invitrogen, Carlsbad, CA) extraction protocol with chloroform interphase separation, isopropanol precipitation, and ethanol wash steps. Precipitated total RNA was resuspended in Rnase-free water and purified with an RNeasy Mini Kit using spin technology (Qiagen, Valencia, CA). Spectrophotometric determination of purified total RNA yield (μg) was performed using the NanoDrop ND-1000 (Thermo Scientific, Wilmington, DE). Total RNA quality was measured using RNA 6000 Nano Series II Kit on a BioAnalyzer 2100 (Agilent Technologies, Santa Clara, CA). Ethical approval for total RNA processing and decidua expression analysis was obtained from the institutional review board at the University of Texas Health Science Center in San Antonio.

Synthesis, amplification, and purification of antisense RNA

Antisense RNA (aRNA) was synthesized, amplified, and purified using the Illumina TotalPrep RNA Amplification Kit according to manufacturer’s instructions (Ambion, Austin, TX). Synthesis of aRNA was performed using a T7 Oligo(dT) primer, and the amplification underwent in vitro transcription with a T7 RNA polymerase to generate multiple copies of biotinylated aRNA from a double-stranded complementary DNA (cDNA) template. Purified aRNA yield was determined spectrophotometrically using the NanoDrop ND-1000.

Microarray data

Purified aRNA was hybridized to Illumina’s HumanWG-6 v2 Expression BeadChip (Illumina Inc, San Diego, CA). Washing, blocking, and transcript signal detection (streptavidin-Cy3) was performed using Illumina’s 6 × 2 BeadChip protocol. Samples were scanned on the Illumina BeadArray 500GX Reader using Illumina BeadScan image data acquisition software (version 2.3.0.13). Illumina’s BeadStudio Gene Expression software module (version 3.2.7) was used to subtract background noise signals and generate an output file for statistical analysis.

Real-time quantitative polymerase chain reaction

We performed a verification of the microarray experiment with quantitative real-time (RT)-polymerase chain reaction (PCR) on 6 of the most differentially expressed transcripts using a 7900HT Fast RT-PCR instrument (Applied Biosystems, Foster City, CA). The 6 genes were prioritized for RT-PCR based on beta values, false discovery rate (FDR) P values, and manual literature searches. RT quantitative PCR was run with 93 samples. Two of the total collection of 95 samples were excluded due to shortage of biological material. Preoptimized TaqMan Gene Expression Assays (Applied Biosystems) were run, in triplicate, to measure messenger RNA expression levels relative to the reference genes, TATA box binding protein and glyceraldehyde-3-phosphate dehydrogenase. Reverse transcription and PCR amplification was performed in a 2-step procedure, following Applied Biosystems High-Capacity cDNA ReverseTranscription Kit Protocol and TaqMan Gene Expression Master Mix Protocol. Negative controls were run, in triplicate, without RT enzyme or no cDNA template.

Statistical analysis

Transcript data for each sample were preprocessed and analyzed using our Sequential Oligogenic Linkage Analysis Routines (SOLAR) statistical analysis software program, as previously described. To evaluate the magnitude of differential gene expression the displacement of each detected transcript’s mean expression value was measured between the 2 groups. A standard regression analysis was performed on the preeclamptic group to test whether the mean transcription level differed from that of the normal pregnancy group.

The messenger RNA expression levels were calculated by the Comparative threshold cycle (CT) method, as described elsewhere. For each target gene, the mean CT value for each sample was used for analysis, after exclusion of outliers. Outliers were determined as values >2SD from the mean. Delta CT (ΔCT) values were computed as the difference between the given mean value for a target gene and the mean of the CT values for the 2 reference genes. Fold change values were calculated, based on the differences in ΔCT values between tissue from preeclamptic women and women with normal pregnancy (2 ^−ΔΔCT ). A t test statistic (SPSS, version 16; SPSS, Inc, Chicago, IL) evaluated the difference between the ΔCT values of the preeclamptic pregnancies, compared with the normal pregnancy group. Analyzing for the 2 reference genes separately did not change the results.

Canonical pathway and network identification

Differentially expressed transcripts in the preeclamptic group ( P < .05; FDR, P < .1) were imported into Ingenuity Pathways Analysis (IPA, v7.5; Ingenuity Systems, Redwood City, CA). Transcripts’ gene identifiers were mapped to their corresponding gene object in the Ingenuity Pathways Knowledge Base. IPA was used to bioinformatically identify canonical (ie, cell signaling and metabolic) pathways and gene-gene interaction networks potentially involved in preeclampsia within our dataset. IPA gene-gene networks were constructed from the published literature, and they diagrammatically represent molecular relationships between gene-gene products.

Significant IPA pathways were further analyzed with Rotation Gene Set Enrichment Analysis (ROMER; Fred Hutchinson Cancer Research Center, Seattle, WA) pathway analysis, using the limma package, available via the Bioconductor Project (Fred Hutchinson Cancer Research Center).