Cotinine in human placenta predicts induction of gene expression in fetal tissues

Abstract

Maternal cigarette smoking during pregnancy is associated with increased risk of perinatal morbidity and mortality. However, the mechanisms underlying adverse birth outcomes following prenatal exposure to cigarette smoke remain unknown due, in part, to the absence or unreliability of information regarding maternal cigarette smoke exposure during pregnancy. Our goal was to determine if placental cotinine could be a reliable biomarker of fetal cigarette smoke exposure during pregnancy. Cotinine levels were determined in placentas from 47 women who reported smoking during pregnancy and from 10 women who denied cigarette smoke exposure. Cotinine levels were significantly higher in placentas from women reporting cigarette smoking (median = 27.2 ng/g) versus women who reported no smoke exposure (2.3 ng/g, P < 0.001). Receiver operating characteristic curve analysis identified an optimal cut point of 7.5 ng/g (sensitivity = 78.7%, specificity = 100%) to classify placenta samples from mothers who smoked versus those from mothers who did not. Among 415 placentas for which maternal cigarette smoking status was unavailable, 167 had cotinine levels > 7.5 ng/g and would be considered positive for cigarette smoke exposure. Data from quantitative reverse-transcription polymerase chain reaction analyses demonstrated that in utero cigarette smoke exposure predicted by cotinine in placenta is associated with changes in the expression of xenobiotic-metabolizing enzymes in fetal tissues. CYP1A1 mRNA in fetal lung and liver tissue and CYP1B1 mRNA in fetal lung tissue were significantly induced when cotinine was detected in placenta. These findings indicate that cotinine in placenta is a reliable biomarker for fetal exposure and response to maternal cigarette smoking during pregnancy.

Fig. 1.

Cotinine levels (ng/g) in human placentas from mothers reporting no cigarette smoking versus mothers who reported smoking during their pregnancy. Cotinine concentrations in whole cell extracts were determined (as described in Materials and Methods) from placentas from women who denied smoking during their pregnancy (n = 10) and from those who reported cigarette smoking during pregnancy (n = 47).

Fig. 2.

Receiver operating characteristics (ROC) curve for placenta cotinine concentration and maternal smoking status.

Fig. 3.

Frequency distribution of placenta cotinine concentration (ng/g placenta) in 415 placentas from mothers for whom information regarding cigarette smoking was unavailable. The three peaks are consistent with unexposed, low/passively exposed, and actively intrauterine smoke exposed fetuses. The cut point of 7.5 ng/g determined by ROC analysis is shown. In addition, arrows indicate cotinine levels for seven mothers who reported the number of cigarettes consumed per day. Numbers above the arrows are the reported cigarettes per day.

Fig. 4.

Expression of xenobioticmetabolizing enzymes in prenatal lung and liver in response to maternal cigarette smoking as predicted by placental cotinine concentrations. Quantitative RT-PCR was performed as described on RNA isolated from fetal liver or lung from pregnancies with (n = 23) or without (n = 23) maternal cigarette smoke exposure as determined by cotinine levels in the placenta. CYP1A1 mRNA expression was induced in fetal liver and lung tissue, and CYP1B1 expression was induced in fetal lung in response to maternal cigarette smoking. * P < 0.05.

Fig. 5.

(A) Correlation between serum and placenta cotinine levels determined in a mouse model of maternal cigarette smoking. (B) Expression of Cyp1a1 and Cyp1b1 mRNA are induced in postnatal lung tissue after maternal cigarette smoke exposure during pregnancy. Quantitative RT-PCR was performed as described on RNA isolated from lungs of pups on postnatal day 5 that were from dams either unexposed (n = 8) or exposed (n = 8) to cigarette smoke during pregnancy (n = 8). * P < 0.05.