Genetic modification of the effect of maternal household air pollutionexposure on birth weight in Guatemalan newborns
Abstract
Low birth weight is associated with exposure to air pollution during pregnancy. The purpose of thisstudy was to evaluate whether null polymorphisms of Glutathione S-transferases (GSTs), specificallyGSTM1 and GSTT1 genes in infants or mothers, modify the association between high exposures to house-hold air pollution (HAP) from cooking fires and birth weight. Pregnant women in rural Guatemala wererandomized to receive a chimney stove or continue to use open fires for cooking. Newborns were mea-sured within 48 h of birth. 132 mother–infant pairs provided infant genotypes (n = 130) and/or maternalgenotypes (n = 116). Maternal null GSTM1 was associated with a 144 g (95% CI, −291, 1) and combinedmaternal/infant null GSTT1 was associated with a 155 g (95% CI, −303, −8) decrease in birth weight.Although there was a trend toward higher birth weights with increasing number of expressed GST genes,the effect modification by chimney stove use was not demonstrated.
1.Introduction
Low birth weight, a significant cause of infant morbidity andmortality, is associated with maternal exposure to air pollutionduring pregnancy. Evidence of this association is drawn from atleast two dozen studies in developed countries [1–3] where air pollutant concentrations are relatively low and low birth weight(LBW, newborn weight <2500 g) is relatively rare. More than 95%of all LBW infants are born in low income countries [4]. In thesecountries, approximately 80% of rural households are exposed to amajor source of air pollution, smoke inside the home generated bycooking and heating with solid fuels (e.g. wood) [5].
Under these conditions, pregnant women are highly exposedto toxic by-products of combustion from cooking fires, and thesematernal exposures impact newborn birth weight. A systematicreview with a meta-analysis of five studies that examined the rela-tionship between maternal exposure to solid fuels and birth weight[6–10] found a 38% increased risk of LBW (OR 1.38, 95% CI, 1.25,1.52) among those exposed to smoke from solid cooking fuel [11].Two of these studies were conducted in Guatemala [8,10], where57% of all households [12] and 85% of rural households use woodfuel [5].∗Corresponding author. Tel.: +1 415 502 5628; fax: +1 415 753 2161.E-mail address: lisa.thompson@nursing.ucsf.edu (L.M. Thompson).
Not all pregnant women exposed to air pollution have low birthweight infants. Thereason for this may be differential genetic sus-ceptibility. Maternal genetic polymorphisms were found to modifythe associations between birth weight and tobacco smoke [13–23]and birth weight and outdoor air pollution exposures [22,24]; sev-eral studies even found independent gene effects on birth outcomes[25–27]. Two isoforms of the Glutathione S-transferase super genefamily, GSTM1 and GSTT1, have been proposed as candidate genesfor susceptibility to inhaled oxidants from air pollutants. The mod-ifying effect that maternal and infant genotypes of GSTM1 andGSTT1 have on exposures to solid fuel smoke and the resultingimpact on newborn birth weight are unknown.
Three major air pollutants from solid fuels burned in inefficient,poorly ventilated cookstoves are particulate matter (PM), car-bon monoxide (CO) and polycyclic aromatic hydrocarbons (PAHs)[28,29]. The toxicity of PM, a composite mixture of liquid drops andsolid particles suspended in air, may be related to the physical andchemical properties of the particles, such as particle size and sur-face composition [30,31]. For health effects, PM is most commonlyclassified by the mean aerodynamic diameter of the particle, withsmaller particles exerting larger effects. In Guatemala, 48-h kitchenconcentrations of PM2.5(PM with aerodynamic diameter <2.5 _m)averaged 636 _g/m3(Standard deviation (SD): 402 _g/m3, n = 50)in homes using open fires for cooking and 69 _g/m3(SD: 89 _g/m3,n = 49) in homes using well-maintained chimney stoves [32]. How-ever, even the kitchen concentrations where chimney stoves were used exceed the World Health Organization Air Quality Guidelinesof 25 _g/m3for PM2.5averaged over a 24-h period [33].
Carbon monoxide (CO) is a product of incomplete combustionreleased when organic solid fuels, such as wood, are burned withinsufficient oxygen supply. Mean CO exposures over a 24-h periodhave been measured at 2–50 parts per million (ppm) among house-holds that burn solid fuel, and often exceed the 9 ppm standard (inan 8-h period) set by the WHO [34,35]. CO levels in homes usingsolid fuels are sometimes high enough to result in blood carboxy-hemoglobin (COHb) levels between 2.5% and 16%, with upper limitscomparable to levels measured in heavy smokers [36–38].
Polycyclic aromatic hydrocarbons (PAHs) are a large class oforganic compounds consisting of hydrogen and carbon organizedin fused ring structures. PAHs are found on combustion-sourcedparticles as well as in the vapor phase of combustion emissions.Tobacco smoke is a major source of exposure to PAHs, but cookingand heating with solid fuels is also a major contributor to outdoorand indoor exposures in low resource countries [39–41].
High exposures to combustion by-products may impact fetalgrowth by several mechanisms, including interference withtransplacental delivery of oxygen, which may cause fetal growthretardation, a form of LBW. Due to physical characteristics (e.g.large surface area) and chemical composition (e.g. PAHs adheringto surface of particle), small particles (≤PM2.5) can induce oxidativestress, which causes local inflammation in the maternal pulmonarysystem, leading to both short-term and chronic damage to the lung[42]. Lung damage decreases maternal oxygen exchange, ultimatelyaffecting oxygen transport to the fetus. In addition, the smallest par-ticles can be transported from the lung across the alveolar-capillarymembrane into the blood circulation, where they may exert effectson the cardiovascular system [43]. PAHs are transported by dif-fusion across membranes in the respiratory and gastrointestinaltract and are absorbed into the bloodstream. PAHs are capableof crossing the placenta [44,45]. Umbilical cord blood PAH-DNAadducts have been associated with decreased fetal growth, afteradjusting for maternal environmental tobacco smoke exposuresin China [46], Krakow, and New York [47,48]. Carbon monoxideis a potent fetotoxicant that binds with both maternal and fetalhemoglobin, forming carboxyhemoglobin and making oxygen lessavailable for oxygenating tissue (such as the placenta). Exposuresto low, constant CO in utero, as measured by maternal carboxyhe-moglobin, produce large decrements in oxygen tension in the fetalblood stream [49]. In animal models, maternal exposure to ambientconcentrations of elevated CO has been associated with poor fetaldevelopment including reduced birth weight [50–52].
Air pollutants are damaging xenobiotic substances in and ofthemselves, but they also induce harmful endogenous by-productsof oxidative stress [53,54]. Genes that modulate oxidative stressare good candidates for investigating the interaction between airpollution and adverse human health. Glutathione S-transferases(GSTs) are a family of detoxifying enzymes that play an impor-tant role in protecting cells from reactive oxygen species (ROS),which can cause oxidative stress. GST enzymes conjugate ROSwith glutathione, thus allowing the detoxification and excretion ofharmful substances, such as ROS. A recent systematic review foundsuggestion of gene-environment interactions between outdoor airpollution and GST polymorphisms on respiratory lung function, butevidence for specific pollutants acting in concert with specific geneswas not conclusively determined [55]. The complete deletion ofboth alleles (null polymorphism) confers absence of an importantdetoxifying enzyme, thus potentially increasing individual suscep-tibility to air pollutants.
Our aim was to evaluate whether the null polymorphismsof GSTM1 and GSTT1 genes in infants or mothers would mod-ify the association between exposure to household air pollution(HAP) from wood fuel use and newborn birth weight. Our a priorihypothesis was that between the low and high HAP exposuregroups, we would find differences in birth weight in mothers and/orinfants with null genotypes compared to those with non-null geno-types. This is a sub-study of a larger birth cohort reported elsewhere[10], which found that rural, Guatemalan mothers randomized toreceive a wellmaintained chimney stove during pregnancy hada 39% reduction in personal exposures to carbon monoxide andhad infants who weighed 89 g more than infants whose mothersused open fires during pregnancy. Here we report on the cohort ofpregnant women and children who participated in two studies thatmeasured the impact of a randomized chimney stove on child pneu-monia and pulmonary growth: the RESPIRE (Randomized ExposureStudy of Pollution Indoors and Respiratory Effects) study [56] andthe follow-up study CRECER (Chronic Respiratory Effects of EarlyChildhood Exposure to Respirable Particulate Matter) [57].
2. Materials and methods
2.1. Study populationThe study population, located in 23 rural communities in theWestern highlands of Guatemala, consists of primarily indigenousMam-speaking women and children. The majority of the house-holds in these communities used wood fuel for cooking during thestudy period. Between October 2002 and May 2003, 266 pregnantwomen who used open fires were enrolled into the RESPIRE study;half were randomized to receive a vented chimney stove. Amongwomen recruited during pregnancy, 254 singleton, healthy new-borns were born (five miscarriages, four stillbirths, two pregnancieswith multiple gestations, and one child with Down syndrome wereexcluded). We were able to measure birth weight on 224 (88%)newborns within 1 week and 190 (75%) within 48 h of a homedelivery. Between January 2007 a
การแปล กรุณารอสักครู่..
