Mammalian Toxicology, Session 13

Adequacy of models for developmental toxicity; C 10, 29; Steroid disruptors: estrogen, androgen, progestin, corticoid; C 20, 21, 29; Thyroid, retinoid and other disruptors; C 10, 20, 21; Take Home Exam Questions Distributed

Adequacy of models for developmental toxicity

The best existing models for developmental toxicity use multiple generations to evaluate impacts on both the first and second generations following dosage of the parental generation (usually of either the male or female parent only so as to optimize the identification of the target of the actual insult).  Evaluations of the first generation indicate direct effects of toxic insult on development.  This may involve germline mutations in the parental generation if exposure took place in the parents prior to mating. It may involve genotoxic impacts involving the germline or embryonic F₁ tissues.  It may involve teratogenic insults derived from either mutational or epigenetic influences within the F₁ embryo.  More subtle impacts on the F₁ generation may appear later in development or during aging, e.g., increased incidences of particular tumor types.  Evaluations of the F₂ generation primarily serve to demonstrate the integrity of the F₁ generation's germinal and reproductive tract.  The model organisms used might include common test species, but they may also include more rapidly reproducing animals such as zebra fish, a model for exploration of development within the laboratory.  Explorations may also include evaluation of mRNA transcription or protein synthesis within specific tissues or cells within the developing test organism.

Note however that disruptions of many of the epigenetic influences on development (hormonal, nutritional) are not ascertained, or are observed only indirectly, in these tests.  Tests that look at shorter test segments or isolated tissues are further blinded to the complexities of the developmental process which is actually a three body problem (father, mother, offspring).  As environmental exposures of populations become more important with respect to developmental insults (see anti-estrogens and anti-androgens below) it will probably become important to include test designs that explore a variety of non-protein as well as protein hormones during the course of gestation and development.  Moreover, it will also be important to explore the impacts of exposure of both parents on the developmental outcomes of the offspring.

Endocrine Disruption

As a topic of active interest currently and in the recent past there are many references that can be pointed to:

Silent Spring. R. Carson. Fawcett Crest: New York, NY. 1970. [Many other editions are available.]

Our Stolen Future. T. Colborn, D. Dumanoski, J.P. Myers. Penguin Books: New York, NY. 1997.

Out Stolen Future - an activist website

Hormonally Active Agents in the Environment. Committee on Hormonally Active Agents in the Environment, Board on Environmental Studies and Toxicology, Commission on Life Sciences, National Research Council. National Academy Press: Washington, D.C. 1999.

Endocrine and Hormonal Toxicology. P.W. Harvey, K.C. Rush, A. Cockburn. John Wiley & Sons Ltd.: Chichester, UK. 1999.

Generations at Risk: Reproductive Health and the Environment. T. Schettler, G. Solomon, M. Valenti, A. Huddle. The MIT Press: Cambridge, MA. 1999.

Hormonal Chaos: The Scientific and Social Origins of the Environmental Endocrine Hypothesis. S. Krimsky. Johns Hopkins University Press: Baltimore, MD. 2000.

Note that many of the considerations we mentioned in discussing reproductive toxicity again arise when examining the overall endocrine axes: lipid uptake by steroidogenic tissues, actions of steroidogenic P₄₅₀ enzymes on xenobiotics to produce proximal toxicants, interferences with enterohepatic breakdown of lipoid hormones causing disruptions of feedback loops, and modification of the molecular loads carried by hormone binding globulins in circulation.

There are few big surprises here, mainly logical outcomes of the physical chemical and physiological characteristics that make up the endocrine system itself.

We have already covered the physiology of the reproductive systems of the male and female including the multiple levels of intercellular communication necessary for normal function to occur.  Note that this provides many targets for potential disruption by pharmacological or environmental agents.  Regulation by the neuroendocrine components of the hypothalamus provide a target for a variety of neurotoxic agents that may ultimately be expressed as reproductive disruption and, therefore, toxicity. Secondary controls on hypothalamic control such as the stress axis (adrenal axis) and leptin regulation provide additional areas that may be disrupted by external agents yet express themselves as reproductive problems.  Disruption of the feedback regulation exerted by gonadal steroids on the hypothalamus and hypophysis provide yet another target for disruption of reproduction at the level of the brain and pituitary.   But potential targets also exist with respect to the binding and actions of the gonadotropins at their primary target cells in the gonads, the cells supporting gamete production.

Steroid disruptors: estrogen, androgen, progestin, corticoid

Environmental Estrogens and Other Hormones

Leydig cells in the testis and granulosa, thecal, stromal, and luteal cells in the ovary are all steroidogenic.  They make use of cholesterol and cholesterol ester stores sequestered as lipid droplets to generate the gonadal steroids that both support gonadal and reproductive tract function and provide the systemic feedback to regulate hypothalamic and hypophysial control of reproduction.  Thus, these cells attract and concentrate lipophilic compounds that are capable of physically partitioning into fat. Brain tissue also tends to be vulnerable in this way because of the concentration of lipids in myelin.  Oocytes themselves tend to accumulate such materials because of their yolk content.  Sperm are spared due to their comparatively low lipid content.

The Sertoli cells and male gametes are vulnerable to other agents.  Those capable of disrupting tight junction formation may break down the junctional complexes between Sertoli cells and developing spermatogenic cells, disrupting the blood testis barrier that is needed to maintain integrity of the semeniferous tubule compartments and keep spermatogenesis well orchestrated.  In contrast, the granulosa cells (especially of the cumulus layer) and the oocyte are vulnerable to agents that disrupt gap junctional complexes since these are needed to coordinate the functions of the ovarian follicle and the growth of the oocyte.

While the spermatogenic cells are particularly sensitive to toxins affecting rapidly growing cells, mammalian oocytes are sensitive to agents that interfere with both meiosis and apoptosis (the normal manner in which oocytes begin their decline toward atresia). Agents that speed the loss of oocytes can lead to premature depletion of the oocyte pool because that pool is physiologically limited in size at mammalian birth.  Spermatogenesis on the other hand seems relatively robust but can still be disrupted by any agent capable of modifying the sequence of events leading to formation of normal spermatozoa or their transport to the female tract.

The female tract itself is a sensitive target because of the cyclical proliferation, differentiation, and regression that takes place in the tissues of the endometrium, and, to a lesser extent, in the vagina.  Steroid mimics, agents that disrupt cellular proliferation (or regression), or those that alter paracrine signal transmission among the tissues of the female tract can change the timing or movement of gametes and/or zygotes and blastulae through the female system.  Changes in ciliary beat in the Fallopian tube, changes in fimbrial activity, alterations in cervical mucus secretion or biochemistry, alterations of female tract contractility or contractile responses to seminal prostaglandins can all affect the success of gamete movement, fertilization, and/or implantation and gestation.

Finally, any agents that disrupt enterohepatic metabolism of steroids or other reproductive tract signaling agents will have potential reproductive toxicity effects.  Either they will cause steroids to be cleared too rapidly and therefore will alter the normal conversation between the gonads and the hypothalamus/pituitary or they will cause steroids to be cleared too slowly, similarly disrupting this set of feedback equilibria. Such impacts would include alterations of hepatic production of serum binding globulins. Sex-hormone binding globulin/testosterone-estradiol binding globulin (SHBG/TeBG), corticoid binding globulin (CBG), thyroid-binding globulin (TBG), and thyroid-binding prealbumin/transthyretin (a binder for both TBG and retinoic acid derivatives) levels all respond to circulating hormone levels.  SHBG, CBG, and TBG all increase in response to bioavailable estradiol levels; they decrease in response to androgens (or to the decline in estradiol to androgen ratio).  SHBG and TBG also seem to increase with circulating thyroid hormone levels.  Finally, CBG binds both progesterone and corticoids efficiently while it SHBG binds preferentially to androgens.  Thus, endocrine steroid mimics not only change the dynamics of steroid binding to receptors in target cells, they also may alter bioavailable endogenous hormone levels by changing serum buffering capacity for these hormones and/or competing for that capacity.

In testing for impacts on the endocrine system we again run into the problem seen with genetic and developmental toxicity.  To determine if the toxin is effecting the parents, offspring must be generated.  But because the offspring may themselves be carriers of the toxic defect imposed by exposure of the parents to the toxic agent, the offspring must be evaluated as well as the parents for anomalies of structure and function. And function includes reproductive function.  Thus, reproductive toxicity tests, as they exist at the present time, include evaluation through three generations (parental, F₁, and F₂) to ascertain what the impacts on reproduction might be.

A couple of additional articles also help to demonstrate the scope of a small part of the endocrine disruption literature, these are linked to the posted session.

[An article concerning disruption of sexual development in birds on the West coast as a result of exposure to chlorinated hydrocarbons.]

[An article from the Ecologist regarding possible lindane contamination of chocolate from Ghana as a result of economically driven agricultural practices.]

These two articles remind us that the legacy of synthetic polychlorinated hydrocarbons is still with us.  Even 40 years after the publication of Silent Spring and the governmental actions in setting up the EPA, NIEHS, and strengthening the FDA, these pesticides and their relatives remain a persistent problem.  The impact of chlorinated aromatics including DDT and PCBs has continued to unfold since their development in the 1930s and their association with the decline of predatory bird species described in the 1960s and early 1970s.  Dr. Jeremy Hatch of UMB/Biology works on a population of roseate terns who have a decreased fertility and marked decrease in the proportion of males due to contamination of food sources near Cape Cod with chlorinated hydrocarbons.  The Wisconsin Raptor Center and the WARF Institute played a major role in describing the decline of predatory bird species as a result of egg-shell thinning, calcium metabolism problems, and vitality reduction associated with accumulation of excessive levels of these compounds in bird fat and egg yolk.  The compounds accumulate up the food chain, concentrate in fatty tissues, inhibit the carbonic anhydrase necessary for egg shell formation, and tend to be "off-loaded" in the fatty yolk of the eggs. Thus, young are exposed to even higher levels than are adults during crucial times of development.  Although PCBs were in use when DDT was being identified with bird population problems, it wasn't until the mid- to late 1970s when it became fully apparent that this series of widely used electrically nonconductive, fire retardant, heat dispersing agents was leaking into the environment in significant quantities and generating many of the same kinds of problems the chlorinated hydrocarbon pesticides had.

The efficacy of DDT as an insecticide was legend in the period following World War II.  But by the late 1960s most areas using it to suppress malaria bearing mosquitoes noted a marked decline in its impact due to development of DDT-resistant strains of Anopheles and Aedes species. Interestingly, a report on 5/4/03 on National Public Radio described the recent, apparently successful, house-spraying program now being used in South Africa to decrease malarial infections.  Because this program is seeing success it is being extended.  One must only wonder how long it can remain successful and how severe an impact it will have on native non-target species as well as polar species where these compounds tend to ultimately accumulate.

All of these compounds are resistant to degradation but gradually give way to the actions of mixed function oxidases and/or exposure to intense UV irradiation or to high heat.  They are slowly oxidized to related ketones, carboxylic acids, and alcohols that are themselves slowly broken down by P₄₅₀ type enzymes and/or conjugated to glucuronic acid, sulfate, or amino acids to yield more water soluble compounds.  Along the way, however, these compounds and some of their metabolites may mimic estrogens or thyroid hormones.

Estrogen:

The classic example of a synthetic entering the environment and food chain and having a marked toxicological
impact is provided by the actions of diethylstilbestrol.  This compound is a potent synthetic estrogen that acts via the same mechanisms as estradiol (but is not buffered by binding to the sex hormone binding globulin that modulates free levels of estradiol).  It binds to intracellular estrogen receptors with an affinity approximately 100 times higher tha n the endogenous ligand 17-estradiol.  When given during early to mid-gestation to prevent early miscarriage this compound proved a developmental toxicant.  Female children of treated women demonstrated a range of reproductive tract anomalies up to and including altered cervical cell lineages that eventually gave rise to cervical cancers.  Male children also demonstrated impacts ranging up to hypospadiasis (incomplete ventral fusion of the tissue folds forming the penis).  Given the multistage nature of the sexual differentiation process [genetic sex -- being positive for SRY > gonadal sex -- gonadal morphogenesis from the indifferent gonad medulla in males or cortex in females > differentia tion in the early testis of Sertoli cells that produce Anti-Mullerian Hormone and Leydig cells that produce testosterone > internal phenotypic sex  -- suppression of Mullerian ductal derivatives including Fallopian tubes and uterus in the 1^st trimester male fetus combined with the androgenic support of the further differentiation of the Wolffian ductal derivatives to produce the interior components of the male reproductive tract including the epididymis, vas deferens, and seminal vesicles & external phenotypic sex - prenatal testosterone and 5-dihydrotestosterone, generated within target tissues, masculinize external genitalia, causing formation of the prostate, penis, and scrotum, as well as internal organs such as liver and kidney, while peripubertal androgens and estrogens drive morphological maturation > brain sex - appears to be largely defined by estrogen generated in the target cells during the natal/perinatal period in both females and males while later behaviors seem to respond preferentially to estrogens in females and androgens in males in mammals it is perhaps only surprising that more marked impacts failed to occur in male offspring.

And while pharmacological or environmental exposures to synthetic estrogens are described above, a variety of reports now discuss phytoestrogens as one natural source of ecological exposure that may help explain different reproductive and breast cancer rates across animal populations including humans.  Soy extracts are being sold as a natural alternative source of estrogenic substances that can help alleviate the undesirable symptoms associated with menopause, e.g., hot flashes, drying and thinning of skin and vaginal linings.  Meanwhile a variety of plant oils and extracts are being used in skin lotions to combat the hormonal impacts of aging.

Note that estrogenic impacts also include a wide range of alterations of metabolic parameters including circulating lipid profiles (decrease LDL) and bone calcium deposition (suppressed by estrogens).  Since androgen frequently antagonizes these changes it is often difficult to distinguish in whole animal models between the anti-estrogenic effect of a compound and its androgenic effects.  Likewise, it is similarly difficult to distinguish between an anti-androgenic effect and an estrogenic effect.  It is largely due to this ambiguity that there is a much more limited literature dealing with androgenic or anti-androgenic effects of xenobiotics.  This is also compounded by the fact that several different androgenic steroids can bind to intracellular androgen receptors with differing, but often sufficient, affinity to induce transcriptional and translational changes, testosterone, 5-dihydrotestosterone, and androst-4,5-ene-3,17-dione can all interact with the androgen receptor.

The calculated impacts of cumulative exposures to environmental estrogens cannot explain some of the information marshaled as evidence for endocrine disruption in males.  The downward secular trend in human male sperm counts has been suggested as one such index (though the possibility that better techniques have provided better, but lower counts cannot be totally discarded even now).  Another is the inability of endogenous androgens to block nipple development or to support internal and external sex phenotypic development in rats (as it normally does) in animals dosed with suspect compounds.  On the basis of this evidence, p,p'-DDE, metabolites of vinclozolin, and di-n-butyl phthalate have all been designated anti-androgens.

Progestins & Corticoids:

Note that progestins and glucocorticoids share rather similar chemical structures and solubility properties.  Their receptor proteins (progesterone receptor, PR, and glucocorticoid receptor, GR) are likewise quite similar; both sets of compounds are capable of binding, with lower affinity than the major ligand, to each other's receptors.  Moreover, that binding can trigger the complementary biological activity.  Thus, for example, given high enough concentrations of progesterone, immune suppression can be induced via activation of the glucocorticoid receptors. In addition, both steroids can bind to the mineralocorticoid receptor (MR) that normally binds aldosterone.  The usual physiological barrier to this happening in all MR-containing tissues is the presence of the enzyme 11-hydroxysteroid dehydrogenase that metabolizes cortisol to cortisone.  Finally, both glucocorticoids and progestins bind rather effectively to CBG and, in those species such as guinea pig that have it, to PBG, progesterone binding globulin.  So they can displace or replace one another in a number of biological situations.

One class of pharmaceuticals is know to act as anti-progestins/anti-glucocorticoids, e.g., RU486, mifepristone.  Another is used in oral contraception/postmenopausal hormonal replacement to replace intrinsically generated progesterone (medroxyprogesterone, norethindrone, norgestrel, norethynodrel, norgestimate) and seems to counteract the positive effects that estrogen has on cardiovascular parameters.  A third class comprises synthetic glucocorticoids: prednisone, prednisolone, dexamethasone, triamcinolone) that are immunosuppressive and promote cardiac hypertrophy and fibrosis at toxic levels.  Most of the clinical side-effects among the agonistic drugs derive from occupation of non-targeted receptors, PR and MR for glucocorticoids, GR and PR for progestins.  Their impacts on organisms exposed to high levels of the drugs in effluents or untreated sewage streams derive from both their targeted and their untargeted agonistic actions.  The antagonist functions by blocking both PR and GR.  Fortunately most of these compounds contain enough functional groupings to allow them to be catabolized and excreted rather efficiently.

Arsenic has also been described as an endocrine disruptor.   The issue of arsenic as an endocrine disrupting agent is somewhat puzzling.  Still, arsenic's position in the periodic table suggests it may interact and/or interfere with metabolic processes involving phosphorylation.  As this is common during many transduction processes including the movement of glucocorticoid receptors to the nucleus with the concomitant dissociation of a series of heat shock proteins, it may not be so surprising that arsenic may have some specificity in its actions in such systems.  It may also be possible that arsenate's resemblance to vanadate plays a role because vanadate is commonly used during in vitro experimentation to stabilize isolated glucocorticoid receptors.  Such stabilization reduces hormone-receptor turnover and may lock the expression of certain genes either "on" or "off."

Thyroid, retinoid and other disruptors

Thyroid:

Exposures to PCBs (and the structurally related polybrominated biphenyls, PBBs) can potentially alter thyroid function in part because of the structural similarities of some of the PCB isomers and thyroxine or triiodothyronine.  About 50% of thyroxine is carried in circulation by thyroid binding globulin, TBG, while 45⁺% is carried by albumin or transthyretin.  Much of thyroxine enters cells by diffusion, but a substantial quantity also enters via the aromatic amino acid transport proteins in cell membranes.  After cell entry, thyroxine is de-iodinated to triiodothyronine which then binds to receptors that usually reside on DNA binding sites within the nucleus; binding often results in release of the hormone receptor complex from DNA and in a change in DNA structure and transcriptional activity.  What are the obvious possible targets for PCB toxic effects?  If we include molecular clearance pathways, are there other targets?  What if the toxicant is a slightly acidic derivative of a PCB?  What about an amine derivative of a PCB?

Note the molecules needed to synthesize thyroxine, transport it to its somatic targets, activate it in target cells by removing an iodine atom, allow it to activate genes via binding to a receptor sitting on a thyroid recognition element within the DNA, and terminate its actions via oxidation, conjugation and an increased water solubility and urinary elimination.  If we see a structural similarity of a compound to a known hormone or physiological compound, the obvious place to look for potential molecular targets are those molecules normally involved in that compound's physiology and actions.  In the absence of a structural similarity, but with a known set of end effects, the targets may lie on the paths leading to normal activations of the ultimately affected tissue.  As this often involves hormones, hormonal pathways may well be fruitfully explored.

Retinoid & Other:

Although most attention has been directed to the issue of environmental estrogens and/or environmental anti-androgens, thought should also be given to compounds that have potential to disrupt the thyroid axis, vitamin D metabolism, retinoid growth regulation, and the adrenal axis.

Coverage in C&D pays particular attention to tumorigenesis and induction of neoplasias in the endocrine tissues.  It also notes, however, that many of these effects are the indirect result of disruptions of normal homeostatic feedback loops with much of the hyperplastic induction being due to chronic elevations in trophic hormone production and target tissue stimulation.  While impacts that produce neoplasias are definitely dramatic, they are really the extreme manifestations of functionally more important disruptions of endocrine homeostasis.  Considerable note was taken of the apparently higher susceptibility of rodents, particularly some of the favored toxicity testing strains such as Fischer 344 rats, to formation of endocrine tumors relative to what is seen in other rodent strains, or other species including humans.  Is it worth considering how functional tests of the endocrine axis might serve as more sensitive biomarkers for toxic insults to these systems than waiting for rats to develop visible tumors?

Further, although healthy adult reproductive function is required for species health, it is not a requirement of individual health.  On the other hand, pancreatic, parathyroid, and kidney/adrenal glomerular function are required for individual well-being.  In the absence of the internal homeostasis governed by these systems no individual would be capable of surviving to reproduction much less successfully reproducing.  Shouldn't more emphasis be placed on these systems vis-a-vis endocrine disruption if the promotion of health of the population are among our goals?

Take Home Exam Questions (Due Next Week)