The study was approved by the Human Research Committee at the University of Texas Medical Branch (UTMB) in Galveston, Texas and at Meharry Medical College, Nashville Tennessee. Written informed consent was obtained from all 159 recruited women. A case was defined as a pregnant woman who presented at the labor and delivery ward and was diagnosed by a physician as being in idiopathic preterm labor. The clinical criteria for preterm labor were those used by the American College of Obstetricians and included regular contractions, cervical dilation of 2 cm and/or cervical effacement. Exclusion criteria included maternal illness, anemia, uterine malformations, placental abruption, placenta previa, and steroid use. All women in preterm labor were clinically evaluated for symptoms of chorioamnionitis, bacterial vaginosis (BV) and urinary tract infection (UTI). BV was diagnosed based on clinical symptoms and by the evaluation of vaginal discharge including the presence of 20% clue cells. Women diagnosed with UTI, BV or chorioamnionitis were excluded. However, women with idiopathic PTL who developed clinical infection during their stay in L&D were included. Pregnant control patients were evaluated in a similar fashion during a prenatal clinical visit. In addition, exclusion criteria included patients who exhibited recurrent PTL, patients with a high risk of PTL, and patients who admitted to drug use. Controls were pregnant women not in PTL, who presented at the same hospital and were also subjected to the same exclusion criteria as the PTL cases. Both case and control populations were 18 years old or older. Neither cases nor controls were offered financial compensation for participation in the study. Cases and controls were limited to the following racial groups: white (non-Hispanic), black, and Hispanic. Both hospitals are state funded, have a primary population of the under-served, and provide free access to prenatal care.
Because the major goal of this investigation was to evaluate the TLR4 levels on peripheral blood, we only collected blood samples. The following clinical groups were evaluated: pregnant controls without preterm labor (n = 76) and women with PTL (weeks 24-36) (n = 63). The present analysis is restricted to the 41 controls and 41 cases who were evaluated for TLR4 gene expression after 23 weeks gestation; 25 controls and 12 PTL were tested for TLR4 receptor density analysis, and 10 controls and 10 PTL were tested for WBC count analysis.
To better understand the basic principles controlling TLR4 levels, we also evaluated 12 pregnant control subjects at term who were not in labor (weeks 39-40) and another 8 pregnant control patients at term who were in labor. These 20 patients presented at the same hospitals and were subject to the same exclusion criteria as the PTL and pregnant control groups.
The major goal of this study was to evaluate the association of TLR4 levels and PTL. In order to conduct the study, blood samples were collected from PTL cases and pregnant controls. Blood samples were collected from PTL cases prior to any treatments. Since PTL status and TLR4 levels were measured simultaneously, we do not imply causality and/or temporality and can only report PTL prevalence. This study is a cross-sectional analysis evaluating statistically significant differences of TLR4 expression levels between PTL cases and non-PTL pregnant controls.
White blood cell and RNA isolation
Peripheral venous blood (5 mL) was drawn once into heparinized vacutainers from each case prior to treatment of PTL and from controls during a scheduled prenatal clinic visit. White blood cells were separated from erythrocytes by dextran sedimentation and pelleted by centrifugation; total RNA was isolated using Tri-Reagent (Sigma, St. Louis, Mo). Isolated RNA was quantified by optical density readings at 260 nm, and the purity was estimated by the ratio of 260/280 nm.
Reverse transcriptase-polymerase chain reaction
The Dual Gene Quantitative (Maxim Biotech) and iQ SYBR Green Real Time PCR (Bio-Rad) methods were used to determine TLR4 mRNA levels. Isolated RNA was treated with RNAse-free DNase (Ambion, Austin, Tx) to ensure there was no contamination with genomic DNA. When the Dual Gene kit was used, 1 μg RNA was reverse transcribed using Moloney-Murine Leukemia Virus reverse transcriptase (RT) and random decamer primers using the manufacturer's protocol (RETROscript Kit, Ambion). Dual quantitative polymerase chain reaction (PCR) amplification was performed using 5 μL of cDNA (~0.25 μg), 1.5 U of Taq polymerase (Life Technologies, Carlsbad, Calif.), and PCR thermal cycler (Thermo Hybrid, Franklin, Mass.). The TLR4 primer sequences were TLR4 forward-5'-GGA AGT TGA ACG AAT GGA ATG TG-3' and TLR4 reverse 5'-TCT GAG TCG TCT CCA GAA GIT GTG-3'. The primer sequences for the 18S rRNA gene were not disclosed by the company. The optimized amplification protocol (96°C for 1 minute, 36 cycles of 94°C for 1 minute, 60°C for 90 seconds followed by a final extension at 72°C for 10 minutes) allowed good linear range amplification of TLR4. The observed differences between replicates from the same women were less than 10% and on average varied from 2% to 5%. Of several housekeeping genes evaluated, we selected the 18S rRNA gene since its expression remains constant during pregnancy. As the number of 18S copies greatly exceeds that of TLR4, the amount of 18S was determined using a 1:500 dilution of the rtRNA with linear range amplification at cycle 14.
Real time RT-PCR was performed using the iScript cDNA synthesis and iQ SYBR green reaction mixes and analyzed using the Bio-Rad iCycler apparatus. For this, 600 ng of RNA was reverse transcribed using the iScript cDNA synthesis kit in a total volume of 25 ul. For TLR4, 2 ul of cDNA reaction mix (48 ng) was subjected to PCR analysis in a total volume of 25 ul: 2 ul of a 1:500 diluted sample was used to determine the number of 18S rRNA copies. TLR4-specific oligos hybridized within exons 2 and 3 were mRNA-specific, generated a 236 bp fragment, and exhibited a 94.5% PCR efficiency. The sequences of the TLR4 forward and reverse oligo's were 5'-AAA ATC CCC GAC AAC CTC CC and 5'-AGT CCA GAA AAG GCT CCC AGG, respectively. The TLR4 mRNA pool sizes were normalized using 18S rRNA as the internal standard using oligonucleotides 18Sf 5'-CGG ACA GGA TTG ACA GAT TGA TAG C and 18Sr 5'-TGC CAG AGT CTC GTT CGT TAT CG. The 18S amplicon was an 118 bp fragment and exhibited an 81% PCR efficiency. During the course of our analysis, the specificity of the amplification was determined by melt curve analysis and by visualization using agarose gel fractionation.
Quantitation of RT-PCR products
For the Dual Gene kit, the intensities of the PCR products were digitally captured and quantitated using an AlphaImager HP image-scanning system (Alpha Innotech Corporation, San Fernando, Calif.). The normalized expression of the TLR4 gene was calculated as follows: X/Y = standardized TLR4 gene expression, where X = TLR4 gene expression level, and Y = 18S ribosomal gene (18S) expression level.
For the iCycler real time analysis, the changes in TLR4 expression were determined using the absolute standard curve method normalized using 18S rRNA. TLR4 and 18S copy numbers were determined by extrapolation from a standard curve generated using linearized plasmid DNA (pCR2.1) that contained the TLR4 and 18S rRNA amplicon, respectively. The correlation coefficients of the standard curves were >0.99 and encompassed the entire range of experimental copy numbers. The Ct threshold values were in the 200 - 300 range.
Flow cytometry & quantitation of TLR4 receptor molecules expression in patients WBC's
White blood cells isolated from pregnant women were resuspended at 5 × 106 cells/mL in the staining buffer (phosphate buffered saline without Ca2+ and Mg2+, 1% BSA, 0.05% sodium azide). 50 μL aliquots were stained with anti-TLR4 (IgG2a) monoclonal antibody conjugated with phycoerithrin (PE, 80 μg/mL; Santa Cruz Biotechnology, Santa Cruz, Calif.) and anti-CDI4 IgG2a conjugated with fluorescein (BD PharMingen, San Diego, Calif.), and incubated on ice for 30 minutes in the dark. Isotype-matched irrelevant antibody controls were used to detect nonspecific staining. Cells were washed three times, pelleted by centrifugation (1,000 × g for 2 minutes at 4°C), and resuspended in staining buffer. Staining was performed in the presence of 100 μg/mL of human IgG to block nonspecific binding to Fc_ receptors. Cells were analyzed by flow cytometry on a Becton Dickinson FACScan microfluorometer (Becton Dickinson & Co., Franklin Lakes, NJ). The percentage of TLR4+ granulocytes and monocytes was determined per 15,000 events analyzed using forward (FSC) and side (SSC) scattering or FSC, SSC and CD14 staining respectively. Data were analyzed with CellQuest
software (BD Bioscience, San Jose, Calif.). The quantitation of the number of monocyte TLR4 receptors was conducted using anti-TLR4 mAb-PE conjugates and QuantiBRITE PE beads (Becton Dickinson & Co) as recommended by the manufacturer.
In order to estimate the required sample size, a priori power analysis was conducted with G*Power software version 3 , using a two-tailed t-test with an alpha error probability of 0.05 and an effect size of 0.5. The results were that 26 patients (13 cases and 13 controls) would be required to achieve 80% power and 42 patients (21 cases and 21 controls) would be required in order to achieve 95% power.
Demographic variables are presented as medians and interquartile ranges (IQR) or as numbers and percents. Differences in demographic characteristics of PTL and pregnant controls were assessed using Mann-Whitney tests and Kruskal-Wallis tests. Multiple linear regression was used to compare the mean TLR4 gene expression of PTL and pregnant controls while adjusting for confounding. According to a normal probability plot TLR4 gene expression was not normally distributed, therefore we performed Box-Cox regression to determine the most appropriate transformation. A reciprocal transformation of TLR4 gene expression was used for the regression analysis, and the estimated means were then retransformed for presentation. Statistical analysis was conducted with STATA Software, (Stata Corp, College Station, Texas). A P-value of less than 0.05 was considered significant.