_116n11_science_1439-1575

Meeting Report: Moving Upstream—Evaluating Adverse Upstream End Points
for Improved Risk Assessment and Decision-Making
Tracey J. Woodruff,1 Lauren Zeise,2 Daniel A. Axelrad,3 Kathryn Z. Guyton,4 Sarah Janssen,5,6 Mark Miller,2,7
Gregory G. Miller,3 Jackie M. Schwartz,1 George Alexeeff,2 Henry Anderson,8 Linda Birnbaum,9 Frederic Bois,10
Vincent James Cogliano,11 Kevin Crofton,9 Susan Y. Euling,4 Paul M.D. Foster,12 Dori R. Germolec,12 Earl Gray,9
Dale B. Hattis,13 Amy D. Kyle,14 Robert W. Luebke,9 Michael I. Luster,15 Chris Portier,12 Deborah C. Rice,16
Gina Solomon,5 John Vandenberg,4 and R. Thomas Zoeller 17
1Program on Reproductive Health and the Environment, Department of Obstetrics and Gynecology, University of California, San Francisco,San Francisco, California, USA; 2Office of Environmental Health Hazard Assessment, California Environmental Protection Agency, Oakland,California, USA; 3Office of Policy, Economics and Innovation, and 4National Center for Environmental Assessment, U.S. EnvironmentalProtection Agency, Washington, DC, USA; 5Department of Medicine, University of California, San Francisco, San Francisco, California,USA; 6National Resource Defense Council, San Francisco, California, USA; 7Pediatric Environmental Health Specialty Unit, University ofCalifornia, San Francisco, California, USA; 8Wisconsin Division of Public Health, Madison, Wisconsin, USA; 9National Health andEnvironmental Effects Research Laboratory, U.S. Environmental Protection Agency, Research Triangle Park, North Carolina, USA; 10InstitutNational de l’Environnement Industriel et des Risques, Verneuil-en-Halatte, France; 11International Agency for Research on Cancer, Lyon,France; 12National Institute of Environmental Health Sciences, National Institutes of Health, Department of Health and Human Services,Research Triangle Park, North Carolina, USA; 13George Perkins Marsh Institute, Clark University, Worcester, Massachusetts, USA; 14Schoolof Public Health, University of California, Berkeley, Berkeley, California; 15National Institute for Occupational Safety and Health, Atlanta,Georgia, USA; 16Environmental and Occupational Health Program, Maine Center for Disease Control and Prevention, Augusta, Maine, USA;17Department of Biology, University of Massachusetts, Amherst, Amherst, Massachusetts, USA defines an adverse effect as “a biochemical BACKGROUND: Assessing adverse effects from environmental chemical exposure is integral to public
change, functional impairment, or pathologic health policies. Toxicology assays identifying early biological changes from chemical exposure are
lesion that affects the performance of the increasing our ability to evaluate links between early biological disturbances and subsequent overt
whole organism, or reduces an organism’s abil- downstream effects. A workshop was held to consider how the resulting data inform consideration
ity to respond to an additional environmental of an “adverse effect” in the context of hazard identification and risk assessment.
challenge” (U.S. EPA 2007a) and, for exam- OBJECTIVES: Our objective here is to review what is known about the relationships between chemical
ple, considers such end points as alterations in exposure, early biological effects (upstream events), and later overt effects (downstream events)
circulating levels of sex hormones to be an through three case studies (thyroid hormone disruption, antiandrogen effects, immune system disrup-
adverse effect (U.S. EPA 1996). Identifying an tion) and to consider how to evaluate hazard and risk when early biological effect data are available.
adverse effect forms the basis for hazard identi- DISCUSSION: Each case study presents data on the toxicity pathways linking early biological pertur-
fication and for defining the critical effect for bations with downstream overt effects. Case studies also emphasize several factors that can influ-
ence risk of overt disease as a result from early biological perturbations, including background
chemical exposures, underlying individual biological processes, and disease susceptibility. Certain
overt disease to elucidating toxicologic path- effects resulting from exposure during periods of sensitivity may be irreversible. A chemical can act
ways has been recognized and endorsed by the through multiple modes of action, resulting in similar or different overt effects.
CONCLUSIONS: For certain classes of early perturbations, sufficient information on the disease
process is known, so hazard and quantitative risk assessment can proceed using information on
upstream biological perturbations. Upstream data will support improved approaches for consider-
University of California-San Francisco, Suite 1100,1330 Broadway St., Oakland, CA 94612, USA.
ing developmental stage, background exposures, disease status, and other factors important to
Telephone: (510) 986-8942; fax: (510) 986-8960.
assessing hazard and risk for the whole population.
KEY WORDS: adverse health effects, androgen antagonists, hazard identification, immunotoxicants,
This work was supported by California Environ- risk assessment, science policy, thyroid hormone, toxicologic assessments. Environ Health Perspect
mental Protection Agency, Office of Environmental 116:1568–1575 (2008). doi:10.1289/ehp.11516 available via http://dx.doi.org/ [Online 10 July 2008]
and Health Hazard Assessment, contract OEHHA-06-S34; the U.S. Environmental Protection Agency,Office of Policy, Economics, and Innovation, To evaluate the potential of environmental hazard and risk assessments have often been National Center for Environmental Economics and chemicals to cause harm, to estimate the risks the more overt diseases or defects, rather than National Center for Environmental Assessment, that chemical exposures pose to the popula- events that occur earlier in the disease process.
contract EP07H001060; the Intramural ResearchProgram of the National Institute of Environmental tion, and to identify opportunities for preven- Increasingly, toxicology assays are providing Health Sciences (NIEHS), National Institutes of tion and intervention, the type and extent of more information on how chemicals can inter- Health (NIH); University of California, Berkeley, adverse effects associated with exposure to a fere with cellular signaling or metabolism, dis- NIEHS Superfund Program at Berkeley, NIH grant chemical must be elucidated. To date, hazard P42 ES04705; and University of California, and risk assessments have relied largely on data expression, or otherwise play a role early in dis- from traditional toxicologic studies, such as the ease processes. As scientific understanding of This report has been reviewed by the National Institute of Occupational Safety and Health and the 2-year, chronic toxicology and carcinogenesis the mechanisms through which chemical expo- U.S. Environmental Protection Agency’s Office of studies or the two-generation reproductive tox- sures advance pathologic processes resulting in Research and Development, and approved for publica- icity assay. A primary goal of these studies is to disease increases, so too does the opportunity tion. Approval does not signify that the contents neces- identify whether chemical exposures cause for effective and efficient hazard identification sarily reflect the views and policies of the agency. The overt disease outcomes, such as birth defects and risk assessment. A necessary step in incor- findings and conclusions in this report are those of the and neoplasia. These studies also provide data porating data on early biological perturbations authors and do not necessarily represent the views ofinstitutions affiliated with the authors.
on biological events that precede these overt is to consider how these early events relate to The authors declare they have no competing disease outcomes, often referred to as precursor the concept of “adverse effects.” The U.S.
effects. Adverse effects identified in existing Received 27 March 2008; accepted 9 July 2008.
VOLUME 116 | NUMBER 11 | November 2008 • Environmental Health Perspectives • Were the steps from the upstream event(s) to deficits of TH and is diagnosed primarily by agencies (Collins et al. 2008; National Research the overt downstream effect(s) identified? comparing an individual’s level of TH and TSH Council 2007). These organizations note that How does our understanding of these steps to population reference ranges. Untreated con- moving toward a focus on perturbations along inform the use of the upstream event as a basis genital (neonatal) hypothyroidism can have the disease pathway should result in assays that for risk assessment, particularly in situations severe consequences on neurologic develop- can test more chemicals with reduced cost and when we only have data on upstream events? time and fewer animals, and improve the scien- • How does an understanding of variability in hypothyroidism are treated with T4, and even tific understanding of the relationships between background biological status (e.g., susceptibil- small doses (e.g., 2 µg/kg/day) significantly ity or sensitivity, genetic or otherwise) affect improve later cognitive performance, demon- strating the sensitivity of the developing brain to sary to consider how data on early biological • How might each end point be influenced by TH insufficiency (Oerbeck et al. 2003; Selva changes relate to the concept of an adverse effect and how these data might best be inte- • In considering an upstream biological end grated into hazard and risk assessment. To point that is measured as a continuous variable ing pregnancy can cause lasting developmental [e.g., thyroid-stimulating hormone (TSH)], deficits in the child. Decrements in human what approach should be taken in deciding maternal T4 during the early fetal period are whether a certain degree of upstream change and Risk Assessment, was held 16–17 May will lead to the overt downstream outcomes reduced IQ scores, even for small T4 deficits that do not constitute maternal hypothyroidism should changes in upstream indicators within Workshop Summary
“normal” ranges be considered in evaluating Three case studies were presented at the work- potential for overt downstream outcomes in shop that described available data on toxico- by a complex interplay of dynamic processes, logic pathways connecting chemical exposures • Can the association between upstream events including dietary intake of iodine (necessary to early upstream biological events and then to and downstream effects for chemicals consid- for synthesis of TH); the transport of iodine subsequent overt effects, which are considered ered in the case study be generalized to other into the thyroid; its synthesis and storage in “downstream”: a) thyroid hormone disruption chemicals that have the same upstream effects? the thyroid gland; its release and transport and related toxicities; b) antiandrogen-mediated • What are the obstacles to the use of the through circulation; the deiodination of T4 to male reproductive effects; and c) upstream upstream event in risk assessment (hazard T3 in peripheral tissues; and the degradation indicators of immunosuppression. To frame identification; dose–response assessment)? discussions in terms of implications for hazard is regulated through a negative feedback loop identification and risk assessment processes, the involving the hypothalamus and pituitary.
case studies explored the following questions: Case Study Synopsis:
• What are the precursor effects or other early Thyroid Hormone Disruption
signals the pituitary gland to release TSH, biological changes linked to the case study and Related Toxicities
which stimulates the thyroid to increase TH.
Thyroid hormone overview. The thyroid hor-
When levels rebound, the hypothalamus sig- • Is there sufficient evidence to associate the mones (TH) thyroxine (T4) and triiodothyrine nals the pituitary to decrease TSH. Finally, upstream events with the overt downstream (T3) are essential to neurologic development, TH action is mediated through TH receptors skeletal growth, and normal function of the pul- • What related information would or does monary system, metabolism, and cardiovascular In hazard identification and risk assess- enhance our understanding of the relation- system. This case study emphasized the effects of ment, TH levels could serve as an upstream ship between the upstream event and down- thyroid disruption on neurologic development.
indicator of adverse effects on neurologic Hypothyroidism is a condition of persistent development. Table 1 provides an overview of Table 1. Classes, mechanisms of action, and effects of thyroid-disrupting chemicals on thyroid hormone homeostasis.
Decreased thyroidal synthesis of T3 and T4 Perchlorate, chlorate, bromate, nitrates, thiocyanate Decreased thyroidal synthesis of T3 and T4 Methimazole, propylthiourea, amitrole, mancozeb, soy isoflavones, benzophenone 2, 1-methyl-3-propyl-imidazole-2-thione Increased biliary elimination of T3 and T4 Acetochlor, phenobarbital, 3-methylcholanthrene, PCBs, 1-methyl-3-propyl-imidazole-2-thione Increased biliary elimination of T3 and T4 TCPOBOP, pregnenolone-16α-carbonitrile, TCDD, Hydroxlyated PCBs, triclosan, pentachlorophenol FD&C Red dye no. 3, propylthiouracil, PCBs, Abbreviations: AhR, aryl hydrocarbon receptor; CAR: constitutive androstane receptor; FD&C, Federal Food, Drug, and Cosmetic Act (1938); MCT, monocarboxylate transporter; OATPs,organic anion-transporting polypeptides; PXR, pregnane X receptor; TCDD, 2,3,7,8-tetrachlorodibenzo-p-dioxin; TCPOBOP, 1,4-bis[2-(3,5-dichloropyridyloxy)]benzene; TH, thyroidhormone; TR, thyroid receptor; TRE, thyroid hormone response elements. Data from Crofton (2008). Environmental Health Perspectives • VOLUME 116 | NUMBER 11 | November 2008 classes, mechanisms of action, and effects for among individuals in the level of TH that rep- multiple mechanisms of action. There are also chemicals that disrupt TH homeostasis and resents homeostasis, and the normal range of numerous uncertainties in extrapolating thy- action. For example, perchlorate inhibits the fluctuations for an individual is narrower than roid data from laboratory animals to estimate uptake of iodine, resulting in decreased synthe- the normal range for a population (Figure 1).
effects in humans, including species differ- sis of TH. Polychlorinated biphenyls (PCBs) Consequently, a TH value within the popula- ences in the expression or structure of specific and the pesticide acetochlor activate enzymes tion reference range is not necessarily normal functional proteins that may affect the toxicity in the liver that increase excretion of TH, thus or healthy for the individual (Andersen et al.
reducing circulating levels (Crofton 2008).
TH levels influence a spectrum of health
Factors influencing effects of thyroid-
population range may therefore be associated outcomes and symptoms, in addition to neuro-
disrupting chemicals. Susceptibility varies
with adverse effects for some individuals.
logic development. Lower T4 and higher TSH
markedly with life stage. The fetus, infant, Assessments considering thyroid-disrupting
levels are correlated with adverse effects on the and child appear the most susceptible to neu- chemicals (TDCs) in isolation are likely to
rologic disruption. The permanent effects of underestimate the potential disruption of TH
1991) and cardiovascular system, including TH insufficiency on neurologic development levels or action by real-world chemical mixtures.
increased blood pressure and less favorable were discussed above. Infants are likely less blood lipid profiles (Asvold et al. 2007a, resistant to fluctuations in TH synthesis than 2007b). A meta-analysis of 14 epidemiologic TDCs, including dioxins, PCBs, perchlorate, studies found an overall increase in risk of months’ supply of TH and the half-life of T4 brominated flame retardants, bisphenol A, and coronary heart disease of > 65% in those with in serum is 7–10 days (Greer et al. 2002; several pesticides (Centers for Disease Control subclinical hypothyroidism (elevated in TSH Vulsma et al. 1989). In contrast, the newborn and Prevention 2008). A mixture of 18 TDCs with normal T4) (Rodondi et al. 2006).
thyroid stores a 1-day supply of TH and the (dioxins, dibenzofurans and PCBs) was tested Conclusions regarding thyroid data. Many
half life of T4 in serum is 3 days (Vulsma at doses comparable to human exposure levels environmental chemicals are capable of dis- et al. 1989). Therefore, effects on TH synthe- for effects on serum T4 in rats. Components of rupting thyroid hormone levels, and many of this mixture affect T4 through two different these have the effect of decreasing circulating mechanisms of action: The dioxins, dibenzo- levels of T4. Compensatory mechanisms may Recognizing subpopulations at risk is a key
furans, and dioxin-like PCBs in the mixture not be sufficient to counteract the potential consideration in evaluating effects on TH levels
activate one set of liver enzymes, and the non- downstream consequences of these T4 decre- or action as adverse. Hypothyroidism is preva-
dioxin-like PCBs activate a separate set of liver ments. First, findings in both animals and lent in the U.S. population. Between 1999 and humans indicate that even small maternal T4 2002, 7.3% of the U.S. population ≥ 12 years effect on T4 at environmentally relevant doses of age reported having thyroid disease or taking and a 2- to 3-fold greater than dose-additive range during pregnancy can have adverse neu- thyroid medication. Hypothyroidism is fre- effect on T4 at higher doses (Crofton et al.
rodevelopmental consequences, such as reduc- quently undetected and therefore untreated.
2005), demonstrating that exposures to chemi- tion in IQ, in the developing child. Second, This condition often persists even in those who cals acting on different pathways can have fetuses and infants do not have stored thyroid cumulative effects on an upstream marker.
hormone, and thus have limited capacity to Assessments considering single TDCs in isola- respond to thyroid hormone decrements dur- tion are therefore likely to underestimate the ing critical stages of development. Third, there is a substantial prevalence of thyroid hormone causes an increased demand on the thyroid exposure to chemical mixtures. It is appropri- insufficiency in the population of U.S. women, gland, and hypothyroidism is twice as com- ate to presume cumulative effects unless there indicating that compensatory processes are mon during pregnancy (Aoki et al. 2007).
is evidence to the contrary, and it is important already compromised for many individuals.
The current method for assessing thyroid
for risk assessments to consider real-life Therefore, for hazard identification pur- health may not be predictive of adverse down-
stream effects. Thyroid health is evaluated pri-
Challenges for assessing risks of exposure to
TDC exposures that would result in reduced marily by comparing an individual’s level of TDCs. Currently available mathematical mod-
T4 in a population should be considered an TH and TSH to population reference ranges.
els may not be able to accurately predict the adverse effect. Further-upstream disruptions However, there is a substantial variability that result in lowered T4 levels, such as inhi- bition of iodine uptake, should similarly be considered adverse when their potential con-sequences include alteration of T4 levels.
Because additivity or synergy of TDCs withdifferent modes of action has been demon-strated, and background exposure to many TDCs is common, risk assessments shouldconsider simultaneous exposures to multiple Frequency
agents and account for these interactions.
demand on the thyroid gland, pregnant womenare likely to have heightened sensitivity to thy- roid toxicants, particularly if they have low dietary iodine or thyroid peroxidase antibodies.
4 (nmol/L)
These sensitive subpopulations are likely to be Figure 1. Distribution of 12 monthly measurements of total T4 in 15 healthy men (white bars) and one indi-
vidual (black bars). The distribution in one individual is about half the width of the distribution in the group.
sizable; for example, 38% of 126 pregnant Frequency represents number of measurements. Adapted from Andersen et al. (2002) with permission. women in the National Health and Nutrition VOLUME 116 | NUMBER 11 | November 2008 • Environmental Health Perspectives Examination Survey (NHANES), 2001–2002, antiandrogens in that they bind antagonisti- had low iodine intake (< 100 µg/L urinary cally to the androgen receptor (AR), interfering semen quality in adult males (Hauser et al.
iodine) (Caldwell et al. 2005). Because thyroid with endogenous testosterone and dihydro- 2006). The exposure levels in most subjects in hormone levels affect cardiovascular risk fac- these studies were comparable to the general tors, the potential impact of TDCs on the decreasing the expression of androgen-depen- health of adults in the general population also dent genes. In fetal rat studies, exposure to Antiandrogen effects as adverse end points
is large. Therefore, effects on thyroid represent these chemicals has been consistently associated in hazard identification and risk assessment.
a risk to the general population and not just with retained nipples, decreased sperm counts, An antiandrogenic agent, therefore—one that decreased anogenital distance, hypospadias, and decreased size or agenesis of the accessory action (e.g., binding to the androgen receptor, Case Study Synopsis:
sex glands (Gray et al. 2006). The severity of or via interference with androgen synthesis)— Antiandrogen-Mediated Male
can be predictably associated with a series of Reproductive Effects
adverse end points in animals and likely also Overview of mammalian male reproductive
industrial chemicals with widespread exposure in humans. Several additional scientific find- development. Early in fetal development,
in the general population—that do not bind ings need also be considered when consider- male sexual differentiation is determined ing antiandrogen effects as upstream adverse by expression of a Y chromosome gene, the activity by inhibiting testosterone synthesis.
sex-determining region Y (SRY). Appropriately Specifically, phthalates with four to six carbon Specific periods during development
timed SRY expression results in differentiation side-chains interfere with testosterone produc- are uniquely susceptible to perturbations.
of Sertoli cells in the testes, initiating a cascade tion by inhibiting cholesterol uptake into the Experiments in rats have demonstrated that of events that result in development and differ- mitochondria by steroidogenic acute regulatory both the dose and the timing of exposure are entiation of male reproductive structures and protein and by inhibiting some, but not all, of important for development of phthalate syn- regression of female structures. The interstitial drome. Gestation days 12–19 have been iden- cells of the testes, Leydig cells, produce two (Barlow et al. 2003). These phthalates have tified as a sensitive period in rats, during been most consistently shown to reduce testos- which exposure to dibutyl phthalate induces development: the steroid sex hormone testos- terone production in animal research to date, phthalate syndrome, with some irreversible terone and the peptide hormone insulin-like-3 effects (Carruthers and Foster 2005).
(INSL-3). Testosterone is necessary for differ- (Gray et al. 2000; Liu et al. 2005).
entiation of the Wolffian duct into the epi- bioactive metabolites responsible for phthalate didymis and secondary sex organs such as the phthalate-induced decrease in testosterone toxicity. Excretion of phthalate monoesters in seminal vesicles. Both testosterone and INSL-3 results in a syndrome of anti-androgenic repro- urine is enhanced by glucuronide conjugation.
are necessary for testicular descent in mam- ductive abnormalities, referred to as the mals: Testosterone mediates transabdominal “phthalate syndrome,” which is characterized curonidation pathway is immature and ineffi- descent of the testes (Klonisch et al. 2004), by malformations of the epididymis, vas defer- cient during the period of susceptibility for whereas INSL-3 is necessary for gubernacular ens, seminal vesicles, and prostate; hypospa- phthalate syndrome. This relatively low level of dias, cryptorchidism, and testicular injury; glucuronidation compared to adults could ren- permanent changes (feminization) in the reten- der the fetus much more susceptible to the bio- descent of the testes into the scrotum (Wilson tion of nipples and areolae (sexually dimorphic logically active phthalate metabolites during a et al. 2004). Disruption of testosterone and/or structures in rodents); and demasculinization critical period of sexual development. Exposure INSL-3 production or action can result in of the growth of the perineum, resulting in a to rats during this period of gestation has cryptorchidism (undescended testes), one of reduced anogenital distance (Mylchreest et al.
demonstrated the proportion of free phthalate the most common birth defects in humans.
2000). As with other antiandrogenic chemicals, monoesters relative to glucuronidated phtha- Exposure to antiandrogens during fetal
the severity of effects increases with dose.
late monoesters in amniotic fluid is much life. During fetal life, there is a transient peak
The carboximide pesticides and four to six higher than levels in the urine of the pregnant in testosterone levels necessary for proper side-chain carbon phthalates have different development of male reproductive tissues.
modes of action resulting in the same spectrum Exposure to mixtures of AR antagonists
Disruption in androgen action during this and androgen synthesis disruptors. More than
critical time window results in a number of development. In the developing male repro- 95% of the population from 6 to > 65 years of abnormalities that consistently develop in lab- ductive tissues, disruption of either of these age are exposed to at least five phthalates on a oratory animals (e.g., rats and rabbits), includ- two parts of the androgen action pathway con- regular basis (Silva et al. 2004). Recent studies ing retained female structures, such as nipples verges into a final common pathway: a decrease show that exposure to mixtures of chemicals in male rodents, and malformations of male in androgen-dependent gene expression.
that interfere with androgen action results in reproductive structures, such as hypospadias The spectrum of effects seen in phthalate dose-additive effects. Rats exposed to a mixture (an abnormal location of the urethral opening) syndrome parallels a spectrum of human dis- of AR antagonists, vinclozolin, procymidone, eases called “testicular dysgenesis syndrome”: and flutamide, all acting through the same infertility, cryptorchidism, hypospadias, and mode of action, at doses that would not have androgen activity, including a decrease in testicular cancer. Several epidemiologic stud- caused hypospadias alone, resulted in > 50% of testosterone production or interference with ies have shown associations between certain the animals having hypospadias (Christiansen phthalate metabolites and male developmen- et al. 2008). Rider et al. (2008) found that pre- modes of action has been well described in tal reproductive outcomes, including short- natal exposure to a mixture of seven phthalates the literature for two different classes of and pesticides with differing modes of action chemicals with antiandrogenic effects.
prenatally exposed to phthalates (Swan et al.
(i.e., AR antagonist or inhibition of androgen First, carboximide pesticides such as lin- 2005), changes in reproductive hormone lev- synthesis) produced cumulative, dose-additive uron, vinclozolin, and procymidone are classic els in male infants exposed to phthalates in outcomes in the androgen-dependent tissues.
Environmental Health Perspectives • VOLUME 116 | NUMBER 11 | November 2008 Exposures to phthalates can result in
mediator production, and generation of long- Effects on the immune system during periods
adverse effects from modes of action other
lived memory cells that respond rapidly on sub- of susceptibility. In a review of select develop-
than disrupting testosterone synthesis. In
sequent exposures to the same or closely related mental immunotoxicology literature, Luebke addition to inhibiting the production of testos- antigens. The acquired immune response is et al. (2006) concluded that exposure to envi- terone, phthalates also interfere with pro- mediated by two types of lymphocytes, T cells ronmental chemicals during key developmental duction of INSL-3, which is necessary for and B cells. T cells are a source of soluble media- periods can affect immune function in labora- gubernaculum development (Wilson et al.
tors known as cytokines that stimulate other tory animals. Developmental exposure sup- 2004) and subsequent descent of the testis into immune system cells, including B cells, and act presses T-cell function through adolescence in the scrotum. Unique to phthalate exposure is as effector cells with cytotoxic activity. B cells mice and throughout adulthood in rats.
an absence of the gubernacular ligament in mature into plasma cells, which serve to produce Gestational/neonatal exposure to diazepam males exposed in utero. Thus, phthalates can immunoglobulins (Ig), or antibodies, of various (DZP) and diethylstilbestrol (DES) in rodents act through two different modes of action to subclasses, including IgM, IgG, IgE, IgA, and can cause long-lasting (up to lifetime) effects on IgD. Each of these antibody subclasses serves a the immune system, including suppression of Conclusion regarding antiandrogen data.
unique function in the immune response.
IgM and IgG antibody responses and nonspe- The prenatal exposure of males to antiandro- Factors influencing immune function.
cific lymphocyte proliferation (DES) at doses genic chemicals illustrates several important Numerous factors influence the outcome of that cause only short-term immunotoxicity in points when considering upstream indicators an encounter with an infectious agent. Innate adults. Developmental exposure to DZP, lead, of adverse effects: First, it is necessary for expo- defenses, critical in the early phase of resis- or tributyltin oxide caused immunotoxicity at sure to occur during the critical window, the tance, may be overcome by large numbers of lower doses in young than in mature animals, period of reproductive organ development, in and developmental effects were persistent.
order for certain developmental effects to be pathogen may also prevent effective innate Clinical experience indicates that adult immune observed. Second, perturbations early in the function typically recovers soon after therapeutic development of the male reproductive system microorganisms before infection ensues. Host immunosuppressive treatment ends. However, predictably result in a wide array of adverse attributes such as age, sex, genotype, lifestyle, the developmental data suggest that screening outcomes that are permanent and irreversible.
and disease status all influence immunocom- chemicals exclusively in adult animals may fail Third, exposures to different chemicals with petence, and each may influence the inci- to detect persistent effects or those effects that different modes of action can result in the same occur at lower doses (Luebke et al. 1999).
outcomes due to a deficiency in androgen- Immune function also declines with age, as mediated gene expression. Finally, exposure to at birth and are thus at increased risk of infec- do the normal physiologic processes that limit tion. Protective antibodies are transferred from microbial invasion. Although data are limited, action with effects on the same end point. In mother to fetus across the placenta and to the this suggests that relatively small changes in newborn in breast milk, although passive pro- cause adverse effects in upstream indicators, tection decreases rapidly as these proteins are exposure may have more severe consequences if including a reduction in fetal testosterone lev- catabolized. Average IgM and IgG antibody combined with normal immunosenescence.
els or androgen receptor binding, which can levels do not reach 50% of adult levels until Several epidemiologic studies have described increase the risk of a constellation of down- 7–12 months, and IgA reaches adult levels by associations between early-life chemical expo- 3–5 years. A preponderance of naive T cells sure, altered immune end points, and frequency hypospadias, and, later in life, infertility.
of infections. Increased incidence of otitis Case Study Synopsis: Immune
reduced cell-mediated immunity in the young.
media and respiratory infections were reported in children exposed to PCBs (Dallaire et al.
Function, Immunotoxicity, and
advanced age; however, few studies have evalu- 2006; Dewailly et al. 2000; Nakanishi et al.
Resistance to Infection and
ated the effects of immunotoxicants in older 1985; Weisglas-Kuperus et al. 2000; Yu et al.
Neoplasia: The Downstream
1998). For example, at 3 months of age, multi- Impacts of Unintended
ple upstream indicators of impaired immune Immunosuppression
have established that extrinsic factors, including function were detected in breast-fed infants Immune system overview. The immune system
with higher PCB exposure levels, including consists of a complex system of tissues, cells, and agents, and psychological factors may influence reduced numbers of white blood cells and lym- soluble mediators that protects the body against the course of infection. Furthermore, un- phocytes, and lower serum IgA levels at 7 and foreign substances, including infectious agents intended immunosuppression resulting from 12 months of age (Dewailly et al. 2000). These and some types of tumor cells. Immune cells are exposure to environmental chemicals alone or studies also reported changes in immune system located throughout the body, either in discrete in combination with other intrinsic and extrin- biomarkers, such as changes in blood cell organs, such as the spleen, thymus, and lymph sic factors can shift the distribution of normal counts and T-cell subsets or decreased serum Ig nodes, or in diffuse accumulations of lymphoid immune response, resulting in an increase in levels with increasing PCB exposures (Karmaus and myeloid cells, as are found in association individuals classified as “immunosuppressed.” et al. 2001; Weisglas-Kuperus et al. 2000).
with the skin, lungs, and gastrointestinal tract.
Immunosuppressed individuals might express Antibody responses to tetanus toxoid vaccine Destruction of infectious agents relies on both alterations of immune function such as reduced were significantly decreased after early postnatal innate and adaptive immune responses. Innate antibody production, decreased immune cell PCB exposure (Heilmann et al. 2006).
responses are rapid, do not require clonal expan- counts, or ineffective cell signaling. Recent data sion, and are stimulated by recognition of also suggest that, in addition to suppression, pathogen-associated molecules by macrophages, developmental exposure (i.e., exposure from natural killer cells, and granulocytes. Adaptive, birth through puberty) to certain chemicals quently develop opportunistic infections.
or acquired antigen-specific responses, rely on may shift the pattern of cytokine production, Xenobiotic agents are likely to cause consider- antigen recognition and subsequent events that leading to a greater incidence of allergy and ably more subtle immunologic effects. The culminate in cell proliferation and maturation, interaction among host genetics, pathogen VOLUME 116 | NUMBER 11 | November 2008 • Environmental Health Perspectives virulence, and pathogen dose plays a role in issues that should be addressed in existing haz- greater than additive response (“synergism”) determining the frequency and severity of ill- ard identification and risk assessment practices.
are possibilities, the case studies support appli- ness. For individuals exhibiting mild immuno- Chemical and biological background. Each
cation of dose additivity as a default, as recom- of the case studies illustrated the importance of mended in the guidelines. Considering the xenobiotic chemical exposure, the infectious considering preexisting or continuous exposure influence of multiple chemical exposures on dose of a given pathogen may be lower than to environmental chemicals as well as preexist- upstream events can facilitate a more realistic that which would cause disease in an individual ing biological or disease susceptibilities that characterization of the risks of downstream contribute independently to risk of overt dis- effects when assessing a single chemical that is mally. Limited studies in populations with mild ease. Preexisting exposures or biological vulner- an additional increment to the background forms of immunosuppression (e.g., under psy- abilities can increase the baseline risk of the chological stress or administered immuno- population or enhance already initiated disease Biological background. Background
suppressive therapies) have provided qualitative processes (Figure 2). Consequently, slight per- health status, as influenced by age, preexisting turbations in upstream biological indicators are disease, genetics, and other factors, can influ- function can also lead to an increased incidence more likely to increase risk of subsequent of infectious or neoplastic diseases (Cohen et al.
downstream events given an already more acti- upstream and downstream adverse end points.
vated state among segments of the population.
For example, the implications of a chemical’s Conclusions regarding immunosuppression
Chemical background and exposure to
interference with iodine uptake into the thy- data. In summary, exposures to environmen-
mixtures. NHANES data indicate that the
roid are substantially greater for an infant than tal contaminants have the potential to affect entire U.S. population has measurable levels of for an adult because of differences in thyroid upstream immune indicators, including anti- multiple environmental contaminants in their hormone storage and circulating half life.
body synthesis, T-cell function, and other Likewise, the implications are greater for the Prevention 2008). Other studies of exposure 38% of pregnant women in the United States result in compromised downstream resistance pathways show the population is in constant who are deficient in iodine (Aoki et al. 2007) to infection. However, the impact of back- contact with xenobiotics in air, food, water, and who simultaneously have greater thyroid ground levels of xenobiotics on the burden of disease has not been clearly established. If a Working Group 2005, 2007; National Library development of a fetus. Individuals with pre- positive correlation between increasing expo- of Medicine 2008; U.S. EPA 2007b; U.S.
existing immune suppression, such as organ sure and disease is assumed, then even small Food and Drug Administration 2007). U.S.
transplant patients, those who are HIV posi- changes in immune function will represent an EPA guidelines for mixtures risk assessment tive, and those at early or late life stages, might increased risk for developing disease.
recommend default assumptions of dose addi- experience disproportionately higher disease tivity for mixtures of chemicals with similar Common Themes
toxicologic activity and response additivity for Periods of susceptibility. Exposure to envi-
The case studies demonstrate that certain toxi- mixtures of chemicals that act independently ronmental chemicals during periods of suscep- city pathways are understood well enough that (U.S. EPA 2000a). The case studies showed tibility can pose a unique risk of subsequent screening assays to identify early biological that these preexisting and concurrent exposures downstream effects, both in the short and long changes could be used for hazard identification can increase the effect a given chemical expo- term, as well as diminished capacity for recov- and risk assessment. For example, if a chemical sure has on disease risk, consistent with the ery from decrements to physiological systems.
is found to suppress thyroid function, the asso- dose-additivity default. For example, separate For example, exposure to thyroid-disrupting ciated data could be used to assess risk for studies of antiandrogenic chemicals and thy- chemicals such as perchlorate during the fetal developmental neurotoxicity, because the latter roid-disrupting chemicals found that mixtures effect can be predicted on the basis of the of chemicals acting on common systems via adverse effects on neurologic development by mechanism of action alone. Similarly, when a different modes of action had dose-additive decreasing TH levels; the same mild perturba- chemical is found to be antiandrogenic, the effects. Although a less-than-additive response current science would support using the data [referred to in the U.S. EPA mixtures guide- adverse challenge to a healthy adult. Similarly, to assess increased risk of developmental lines as “antagonism” (U.S. EPA 2000a)], or a to impair male reproductive development in effects. Data from whole-animal tests of thedownstream outcomes would not be requiredfor risk assessment of these outcomes. The findings of this workshop support the recentNational Academy of Sciences recommenda- tion that toxicity testing move toward the direct assessment of upstream events along tox-icity pathways and gradually move away from the current whole-animal end point–based assays (National Research Council 2007).
challenges that arise when using upstream Percent of population
events to assess the adverse effects of chemicalexposures during hazard identification or riskassessment. Several common themes emergedfrom the case study discussions. Some of thesethemes are not unique to the use of upstream Value of physiological parameter
indicators as the basis of hazard identification Figure 2. Distribution of a typical physiological parameter within the population and how that may vary
and risk assessment. Rather, they underscore depending on the influence of chemical and biologic background.
Environmental Health Perspectives • VOLUME 116 | NUMBER 11 | November 2008 rodents, exposure to phthalates must occur process of considering continuous upstream while development of appropriate principles during gestation days 12–19. When exposure and practices for assessing the emerging science occurs during periods of developmental sus- Reversibility. Biological perturbations
for timely public health–based decision-making ceptibility, the dose required to produce an resulting from chemical exposure are often effect will typically be lower than the dose Identifying classes of upstream events. The
required to produce effects in fully developed, reversible. For example, exposure to chemicals case studies of thyroid perturbation, anti- can result in immune system suppression and androgen activity, and certain types of immune Multiple or complex modes of action.
increased opportunity for infections. A sup- changes each represent a class of perturbations Exposure to a chemical can influence disease whose links to downstream events were deter- processes through multiple modes of action and chemical exposure is stopped, and the risk of mined by workshop participants to be suffi- also increase the risk of more than one down- infectious disease returns to original levels.
ciently characterized by the data to move stream, overt effect. Hence, measurements that However, for assessments of a constant chronic toward considering issues of implementation.
focus on a particular pathway or individual or lifetime exposure, recovery from shorter A suggested next step is to evaluate chemicals downstream event may provide an incomplete duration exposures should not be considered for their ability to cause the upstream pertur- picture of the range of effects produced by the when evaluating whether the observed response bation identified in the case studies. For these chemical. Phthalates work through multiple is adverse. Whether a perturbation is reversible chemicals, hazard and risk assessment could be pathways to decrease androgen production but can depend on intrinsic factors, such as age or initiated with data on the relationship between also interfere with the production of INSL-3.
disease status, and extrinsic factors, such as pre- exposures and early perturbations. Fewer data The resulting deficits in these hormones during existing or co-exposures to other contaminants are needed for subsequent downstream events, male fetal development can increase the risk of or the timing of exposure. For example, gesta- given what is already known on risks linking several different downstream effects, both man- tional and neonatal exposure to DZP and DES the pathway from early to late events, and ifested early in life (e.g., cryptorchidism) and in rodents can cause long-lasting effects on with the presumption that downstream conse- later in life (e.g., effects on sperm). Data inter- immune function; in contrast, adult rodents quences are likely to occur when the upstream pretations therefore should consider the possi- exposed to the same dose experience reversible bility of a complex set of modes of action that Participants noted other types of early bio- produce adverse effects on the whole system.
Population variability and defining the
logical perturbations that were not discussed in Focusing on one single aspect of these complex normal range. Basing the definition of “nor-
the case studies but that are also likely suffi- mal” function for a biologic parameter on ciently characterized as to be defined as a class.
characterization of system response.
population reference ranges may not reflect Early biological perturbations that appeared to Continuous versus discrete events. Overt
an individual’s normal range or variability be sufficiently characterized include certain downstream effects, such as cancer, birth and may fail to detect an adverse effect. For high affinity and persistent interactions with the defects, and infectious illnesses are typically dis- Ah receptor, changes in hormonal responses crete events. In contrast, upstream indicators of adverse decline in thyroid hormone but still [which are identified in reproductive toxicity adverse effects, such as changes in hormone remain within the normal population refer- risk assessment guidelines as an adverse event levels or cellular markers of immune function, ence range (Figure 1). For situations in which (U.S. EPA 1996)], and cholinesterase inhibi- are often measured on a continuous scale. Risk the interindividual variability in a continuous tion [which has been identified as an indicator assessment practices recognize differences in upstream indicator is high, alternative methods of adverse effects (U.S. EPA 2000b)]. It would frequency of discrete events as an indication of to detect adverse changes in the biologic also be beneficial to expand the list to include an adverse effect. Characterization of changes parameter will need to be developed.
biological perturbations that appear common to in continuous outcomes in risk assessment many environmental chemicals, such as effects Conclusions and
Recommendations
change constituting adversity. Should any Workshop participants concluded that for the information on upstream indicators does not chemical-induced increase or decrease be con- toxicity pathways represented in these case necessarily reduce uncertainty, although it does sidered adverse? Should there be a focus on studies, it may be feasible to move toward add more information about the exposure– particular cut points in the distribution of using direct evaluation of upstream mechanis- disease continuum and increases the predictive population or individual baseline levels? These tic indicators as the basis for risk assessment power of risk assessment and can demonstrate questions should be considered in the context greater variability across study subjects than of the issues of background exposures and age- The case studies illustrate the complexities previously recognized. Hence, using upstream related susceptibilities noted above. One possi- indicators does not necessarily mean reducing bility is to consider that deviations from exposures to environmental contaminants.
applicable uncertainty or adjustment factors.
expected baseline (e.g., for a given individual, Capturing and translating the complexities of Next steps. The case studies suggest
with all its characteristics, susceptibilities, age, the science is an ongoing challenge in the regu- sex, physiologic state) will increase the proba- latory and policy arena. While science pursues upstream biological markers have utility for bility (risk) of downstream adverse effects. For new areas of inquiry, decision making requires hazard identification and risk assessment. The example, changes in biological function dur- timely answers to questions about risks and haz- approaches to using these upstream indicators ing susceptible periods, such as during devel- ards to public health in order to mitigate future should consider factors such as biological back- or current potential harm. The regulatory con- ground, chemical background (and the poten- text also requires a different sufficiency of evi- tial for dose additivity or synergism), periods of function changes of the same magnitude at dence whereby regulatory decisions can be other life stages. Whether the perturbation is made based on evidence that a chemical is likely modes of action that can increase the risk of reversible and whether it is possible to use to cause a particular outcome. The value of a subsequent downstream events. The case stud- population reference ranges to detect adversity parallel-tracks approach was noted, in which ies also illustrated the need to better integrate are two key questions that may arise in the scientific investigation continues in one track the nonzero baseline concept into hazard and VOLUME 116 | NUMBER 11 | November 2008 • Environmental Health Perspectives risk assessment—in other words, background Crofton KM, Craft ES, Hedge JM, Gennings C, Simmons JE, reproductive development in rats exposed to di(n-butyl) exposures and processes can confer a greater Carchman RA, et al. 2005. Thyroid-hormone-disrupting phthalate during late gestation. Toxicol Sci 55:143–151.
potential for effects than when exposure to a chemicals: evidence for dose-dependent additivity or syn- Nakanishi Y, Shigematsu N, Kurita Y, Matsuba K, Kanegae H, ergism. Environ Health Perspect 113:1549–1554.
Ishimaru S, et al. 1985. Respiratory involvement and chemical is considered in isolation. An impor- Dallaire F, Dewailly É, Vezina C, Muckle G, Weber JP, Bruneau immune status in yusho patients. Environ Health Perspect tant implication of this concept is that the cur- S, et al. 2006. Effect of prenatal exposure to polychlori- nated biphenyls on incidence of acute respiratory infec- National Library of Medicine. 2008. Household Products tions in preschool Inuit children. Environ Health Perspect Database. Available: http://householdproducts.nlm.nih.
noncancer health effects no longer fits with gov/index.htm [accessed March 13 2008].
what we know about the science and needs of Dewailly E, Ayotte P, Bruneau S, Gingras S, Belles-Isles M, Roy National Research Council. 2007. Toxicity Testing in the 21st the decision-maker for current understanding R. 2000. Susceptibility to infections and immune status in Century: A Vision and Strategy. Washington, DC: National Inuit infants exposed to organochlorines. Environ Health of potential risk. Disease status, susceptibility, Oerbeck B, Sundet K, Kase BF, Heyerdahl S. 2003. Congenital Environmental Working Group. 2005. National Tap Water hypothyroidism: influence of disease severity and L-thyroxine consider when assessing the implications of Quality Database. Available: http://www.ewg.org/sites/ treatment on intellectual, motor, and school-associated outcomes in young adults. Pediatrics 112:923–930.
exposure, as they can put a portion of the Environmental Working Group. 2007. Skin Deep Cosmetic Pop VJ, Brouwers EP, Vader HL, Vulsma T, van Baar AL, de Safety Database. Available: http://www.ewg.org/node/ Vijlder JJ. 2003. Maternal hypothyroxinaemia during early incremental exposures can increase risk of pregnancy and subsequent child development: A 3-year Federal Food, Drug, and Cosmetic Act. 1938. Public Law 75-717. follow-up study. Clin Endocrinol (Oxf) 59:282–288.
downstream events. This suggests that the Gray LE Jr, Ostby J, Furr J, Price M, Veeramachaneni DN, Rider CV, Furr J, Wilson VS, Gray LE. 2008. A mixture of seven assumption of a threshold dose level—below Parks L. 2000. Perinatal exposure to the phthalates DEHP, antiandrogens induces reproductive malformations in rats.
which no deleterious effects occur (U.S. EPA BBP, and DINP, but not DEP, DMP, or DOTP, alters sexual International J Androl 31:249–262.
differentiation of the male rat. Toxicol Sci 58:350–365.
Rodondi N, Aujesky D, Vittinghoff E, Cornuz J, Bauer DC. 2006.
Gray LE Jr, Wilson VS, Stoker T, Lambright C, Furr J, Noriega N, Subclinical hypothyroidism and the risk of coronary heart concepts traditionally used for nonthreshold et al. 2006. Adverse effects of environmental antiandro- disease: a meta-analysis. Am J Med 119:541–551.
gens and androgens on reproductive development in Selva KA, Harper A, Downs A, Blasco PA, Lafranchi SH. 2005.
mammals. Int J Androl 29:96–104; discussion 105–108.
Neurodevelopmental outcomes in congenital hypo- Greer MA, Goodman G, Pleus RC, Greer SE. 2002. Health EFERENCES
effects assessment for environmental perchlorate conta- target T4 and TSH. J Pediatr 147:775–780.
mination: the dose response for inhibition of thyroidal Silva MJ, Barr DB, Reidy JA, Malek NA, Hodge CC, Caudill SP, Andersen S, Pedersen KM, Bruun NH, Laurberg P. 2002.
radioiodine uptake in humans. Environ Health Perspect et al. 2004. Urinary levels of seven phthalate metabolites in the U.S. population from the National Health and Nutrition subjects: a clue to the understanding of subclinical thyroid Haddow JE, Palomaki GE, Allan WC, Williams JR, Knight GJ, Examination Survey (NHANES) 1999–2000. Environ Health disease. J Clin Endocrinol Metab 87:1068–1072.
Gagnon J, et al. 1999. Maternal thyroid deficiency during Aoki Y, Belin RM, Clickner R, Jeffries R, Phillips L, Mahaffey KR.
pregnancy and subsequent neuropsychological develop- Swan SH, Main KM, Liu F, Stewart SL, Kruse RL, Calafat AM, ment of the child. N Engl J Med 341:549–555.
et al. 2005. Decrease in anogenital distance among male tion and their association with participant characteristics: Hauser R, Meeker JD, Duty S, Silva MJ, Calafat AM. 2006.
infants with prenatal phthalate exposure. Environ Health National Health and Nutrition Examination Survey Altered semen quality in relation to urinary concentrations (NHANES 1999–2002). Thyroid 17:1211–1223.
of phthalate monoester and oxidative metabolites.
U.S. EPA. 1996. Guidelines for Reproductive Toxicity Risk Asvold BO, Bjoro T, Nilsen TI, Vatten LJ. 2007a. Association Assessment. Washington, DC:Risk Assessment Forum, between blood pressure and serum thyroid-stimulating Heilmann C, Grandjean P, Weihe P, Nielsen F, Budtz- U.S. Environmental Protection Agency.
hormone concentration within the reference range: a Jorgensen E. 2006. Reduced antibody responses to vacci- U.S. EPA. 2000a. Supplementary Guidance for Conducting Health population-based study. J Clin Endocrinol Metab 92:841–845.
nations in children exposed to polychlorinated biphenyls.
Risk Assessment of Chemical Mixtures EPA/630/R-00/002.
Asvold BO, Vatten LJ, Nilsen TI, Bjoro T. 2007b. The association Washington, DC: U.S. Environmental Protection Agency.
between TSH within the reference range and serum lipid Karmaus W, Kuehr J, Kruse H. 2001. Infections and atopic dis- U.S. EPA. 2000b. The Use of Data on Cholinesterase Inhibition concentrations in a population-based study. The HUNT orders in childhood and organochlorine exposure. Arch for Risk Assessments of Organophosphorous and Study. Eur J Endocrinol 156:181–186.
Carbamate Pesticides. Washington, DC:Office of Pesticide Barlow NJ, Phillips SL, Wallace DG, Sar M, Gaido KW, Foster Kiecolt-Glaser JK, Robles TF, Heffner KL, Loving TJ, Glaser R.
Programs, U.S. Environmental Protection Agency.
PM. 2003. Quantitative changes in gene expression in fetal 2002. Psycho-oncology and cancer: psychoneuroimmunol- U.S. EPA. 2002. A Review of The Reference Dose and Reference rat testes following exposure to di(n-butyl) phthalate.
ogy and cancer. Ann Oncol 13 (suppl 4):165–169.
Concentration Processes. EPA/630/P-02/002F. Washington, Klonisch T, Fowler PA, Hombach-Klonisch S. 2004. Molecular DC:Risk Assessment Forum, U.S. Environmental Protection Calafat AM, Brock JW, Silva MJ, Gray LE Jr. Reidy JA, Barr and genetic regulation of testis descent and external geni- DB, et al. 2006. Urinary and amniotic fluid levels of phtha- talia development. Dev Biol 270:1–18.
U.S. EPA. 2007a. Integrated Risk Information System (IRIS): late monoesters in rats after the oral administration of Liu K, Lehmann KP, Sar M, Young SS, Gaido KW. 2005. Gene Glossary of IRIS Terms. Available: http://www.epa.gov/ di(2-ethylhexyl) phthalate and di-n-butyl phthalate.
expression profiling following in utero exposure to phtha- ncea/iris/help_gloss.htm [accessed 10 May 2007].
late esters reveals new gene targets in the etiology of tes- U.S. EPA. 2007b. Where You Live. Available: http://www.epa.
Caldwell KL, Jones R, Hollowell JG. 2005. Urinary iodine con- ticular dysgenesis. Biol Reprod 73:180–192.
gov/air/where.html [accessed 13 March 2008].
centration: United States National Health And Nutrition Luebke RW, Chen DH, Dietert R, Yang Y, King M, Luster MI.
U.S. Food and Drug Administration. 2007. Pesticides, Metals, Examination Survey 2001–2002. Thyroid 15:692–699.
2006. The comparative immunotoxicity of five selected Chemical Contaminants and Natural Toxins. Available: Carruthers CM, Foster PM. 2005. Critical window of male repro- compounds following developmental or adult exposure.
http://www.cfsan.fda.gov/~lrd/pestadd.html [accessed ductive tract development in rats following gestational J Toxicol Environ Health B Crit Rev 9:1–26.
exposure to di-n-butyl phthalate. Birth Defects Res Luebke RW, Copeland CB, Andrews DL. 1999. Effects of aging Vulsma T, Gons MH, de Vijlder JJ. 1989. Maternal-fetal transfer on resistance to Trichinella spiralis infection in rodents of thyroxine in congenital hypothyroidism due to a total Centers for Disease Control and Prevention. 2008. National exposed to 2,3,7,8-tetrachlorodibenzo-p-dioxin. Toxicology organification defect or thyroid agenesis. N Engl J Med Report on Human Exposure to Environmental Chemicals.
Available: http://www.cdc.gov/exposurereport/ [accessed Main KM, Mortensen GK, Kaleva MM, Boisen KA, Damgaard Weisglas-Kuperus N, Patandin S, Berbers GA, Sas TC, Mulder IN, Chellakooty M, et al. 2006. Human breast milk contami- PG, Sauer PJ, et al. 2000. Immunologic effects of back- Christiansen S, Scholze M, Axelstad M, Boberg J, Kortenkamp nation with phthalates and alterations of endogenous ground exposure to polychlorinated biphenyls and dioxins A, Hass U. 2008. Combined exposure to anti-androgens reproductive hormones in infants 3 months of age. Environ in Dutch preschool children. Environ Health Perspect causes markedly increased frequencies of hypospadias in the rat. Intl J Androl 31:241–248.
Mendelson CR, Boggaram V. 1991. Hormonal control of the sur- Wilson VS, Lambright C, Furr J, Ostby J, Wood C, Held G, et al.
Cohen S, Tyrrell DA, Smith AP. 1991. Psychological stress and factant system in fetal lung. Annu Rev Physiol 53:415–440.
2004. Phthalate ester-induced gubernacular lesions are susceptibility to the common cold. N Engl J Med Morreale de Escobar G, Obregón MJ, Escobar del Rey F. 2000.
associated with reduced INSL3 gene expression in the Is neuropsychological development related to maternal fetal rat testis. Toxicol Lett 146:207–215.
Collins FS, Gray GM, Bucher JR. 2008. Toxicology. Transforming hypothyroidism or to maternal hypothyroxinemia? J Clin Yu ML, Hsin JW, Hsu CC, Chan WC, Guo YL. 1998. The immuno- environmental health protection. Science 319:906–907.
logic evaluation of the Yucheng children. Chemosphere Crofton KM. 2008. Thyroid disrupting chemicals: mechanisms Mylchreest E, Wallace DG, Cattley RC, Foster PM. 2000.
and mixtures. Int J Androl 31:209–223.
Dose-dependent alterations in androgen-regulated male Environmental Health Perspectives • VOLUME 116 | NUMBER 11 | November 2008

Source: http://superfund.berkeley.edu/pdf/87.pdf