Epidemiological Studies of Phthalates
Diesters of phthalic acid, commonly referred to as phthalates, are widely used in industry and commerce, including in personal care products (such as makeup, shampoo and soaps), plastics, medical tubing and medication coatings, paints and some pesticide formulations. Polyvinyl chloride (PVC) plastic cannot be made without adding phthalate plasticizers, which act as softeners. Phthalates are not chemically bonded to PVC, so the phthalates readily migrate out of the PVC (e.g., when a baby sucks on a PVC toy), resulting in human exposure to phthalates.
Despite the growing body of literature on phthalate reproductive toxicity and data demonstrating extensive human exposure, few studies have examined the effects of these chemicals on human reproductive development. Duty and colleagues reported dose-response relationships between tertiles of mono-butyl phthalate (MBP) and sperm motility and concentration as well as between tertiles of monobenzyl phthalate (MBzP) and sperm concentration. They also reported inverse dose-response relationships between monoethyl phthalate (MEP) and increased sperm DNA damage measured using the neutral single cell gel electrophoresis (Comet) assay. In this population of men attending an infertility clinic, increased urinary concentration of MBzP was also associated with decreased follicle stimulating hormone (FSH), while increases in MBP were marginally associated with increased inhibin-B .
Over the past 5 years, increasingly detailed rodent studies have demonstrated that there is a significant reduction in anogenital distance (AGD), which is the tissue that forms into the scrotum, in Sprague-Dawley rats following prenatal exposure at high doses to BBzP, DBP, and DEHP. This demasculinizing effect is due to disruption of the androgen-signaling pathway, resulting in interference with the synthesis of testosterone in the testes of male rat fetuses. The consequence of a decrease in testosterone during the fetal period of masculinization is that normal mascu-linization, which occurs in response to testosterone, does not occur. Virtually every aspect of development of a masculine phenotype is disrupted by these phtha-lates: in addition to shortened AGD, the penile and urethral development are abnormal, resulting in an increased frequency of hypospadias (a genital malformation in which the urethral opening is improperly located), and incomplete descent of the testes (cryptorchidism), testes are abnormally small, and other reproductive organs are abnormal. Three phthalates (DEHP, DBP, and BzBP) have been shown to down-regulate expression of the insl3 (insulin-like hormone 3) gene, which plays a critical role in gubernacular development and thus normal testicular descent. The cluster of morphological and functional changes induced by these phthalates has been referred to as the phthalate syndrome. Research described below has begun to relate prenatal phthalate exposure in humans to similar reproductive and developmental effects.
In a recent study, Swan and colleagues examined the concentration of urinary phthalate metabolites in mother's prenatal urine samples in relation to the AGD and other measures of male genital development. This analysis includes physical measurements on 134 boys, obtained on average at 12.6 months of age, whose mothers had participated in a multi-center pregnancy cohort study. The AGD in millimeters was measured and regression analysis used to obtain predicted AGD, controlling for age and weight at examination. Urinary phthalate metabolites were obtained in 85 prenatal urine samples provided by mothers of these boys at 28.3 weeks gestation, on average. In regression analyses the concentration of four phthalate metabolites that are found in urine (MEP, MBP, MBzP, and MiBP) were significantly, and inversely related to age-adjusted AGD. While MEHP was not related to AGD, the secondary metabolites of DEHP (MEOHP and MEHHP) were related, though not statistically significantly at p = 0.05. The authors considered mother's ethnicity and smoking, time of day and season of sample collection as covariates but none were retained in final models. The odds ratio for short AGD (less than 25% of expected for age and weight) were 10.2 (95% CI 2.5 to 42.2) for MBP in the highest quartile compared to the lowest quartile). For highest compared to lowest MEP, MBzP, and MiBP the odds ratios were 4.7, 3.8, and 7.3, respectively (all p values <0.05). While none of these boys exhibited frank malformations, short AGD was associated with an increased proportion of boys with incomplete testicular descent as well as with smaller penile volume.
The cut-offs defining 'high' phthalate exposure were quite similar to the 75th percentiles ofphthalate distribution in a national sample (CDC 3rd Report). A comparison of these levels to doses used in most phthalate rodent studies suggests that humans are at least an order of magnitude more sensitive to these metabolites than rodents. However, until recently, phthalates have been tested singly. Recent mixture studies have shown that these exposures are at least dose-additive, so exposure to multiple phthalates at low levels, which is the typical human exposure pattern, may convey considerably greater risk than predicted by the typical single exposure rodent study. This is supported from results reported by Swan and colleagues. Boys whose mothers had high levels of multiple phthalate metabolites were at particularly high risk of short AGD. Thus, these data suggest that prenatal exposure to phthalates at current environmental levels, may produce a cluster of genital changes consistent with the phthalate syndrome previously identified in rodents.
While Swan et al. did not obtain infant hormone levels, a recent Danish study examined serum hormone levels in 96 boys at three months of age in relation phthalate metabolite levels in breast milk samples collected during postnatal months 1-3. MEP and MBP were significantly correlated with (sex hormone binding globulin) SHBG, which is the plasma-binding protein for both testosterone and estradiol (r = 0.323, p =-0.002 and r = 0.272, p = 0.01, for MEP and MBP, respectively); MMP, MEP, and MBP were positively correlated with LH/testoster-one, and MBP was inversely correlated with free testosterone (all p values <0.05). (Approximately 50% of these boys were cryptorchid; these associations did not differ between normal and cryptorchid boys.) This study suggests an antiandrogenic (testosterone inhibiting) effect of perinatal phthalate exposure in humans, consistent with rodent studies.
Antiandrogenic effects have also been associated with exposure to pesticides. For example, the fungicide vin-clozolin that is commonly sprayed on grapes in vineyards also has antiandrogenic activity. This antiandrogen action is quite different from phthalates however, in that vinclozolin binds to and thus blocks endogenous testosterone from binding to androgen receptors. Other pesticides, such as the insecticides DDT and its replacement after it was banned, the insecticide methoxychlor, have been shown to be converted in vivo to antiandrogenic metabolites; these compounds also bind to androgen receptors but, similar to vinclozolin, do not activate them and are thus antiandrogens. However, both DDT and methoxychlor are examples of products that produce complex effects by interacting with more than one aspect of the endocrine system. Commercial DDT is a mixture of a contaminant (o,p'DDT), which binds to estrogen receptors and activates responses similar to estradiol, while the active insecticide p,p'DDT is metabolized to the highly persistent compound p,p'DDE, which is an antiandrogen. Since o,p'DDT has a shorter half life than p,p'DDE, the initial consequence of DDT exposure is activation of estrogen and inhibition of androgen response systems, while the long-term consequence is persistent antiandrogenic effects. The in vivo metabolite of methoxychlor is similar to bisphenol A in that it has both estrogenic and antiandrogenic activity.
A recent finding that has generated concern is that administration of vinclozolin or methoxychlor to male rats during fetal life via administration to the pregnant mother resulted in male offspring with decreased sperm numbers and decreased fertility. This effect of exposure of Fx generation males to these chemicals during fetal life (when genetic programming is occurring during cellular differentiation) resulted in the same phenotype occurring in F2, F3, and F4 generations transmitted in each generation through the male offspring. Importantly, in each subsequent generation, the frequency of this phenotype was constant at about 90% of males. Thus, exposure of only the Fx generation led to a 'programmed' change in the genome that was transmitted at least through 3 more generations. This phenotype was associated with epigenetic changes (covalent methyla-tion of cytosine bases) in the sperm that was transmitted across generations. These findings raise the possibility that adverse effects caused by environmental chemical exposure during fetal life may not be just limited to the exposed generation, but through nonclassical (epige-netic) modifications of DNA, may become stably fixed into the genome of the germ line and transmitted across generations.
Was this article helpful?