14 Abstracts

as the guinea pig maximisation test and the murine local lymph node assay, can be used both for hazard identification of potential contact sensitisers as well as to obtain a reasonably accurate assessment of risk. At this time, methods for identifying potential contact sensitisers are often being used as general hazard identification methods in the hope of screening out compounds that could induce any type of drug allergy. The usefulness of these methods for hazard identification of potential respiratory and systemic allergens is controversial (Choquet-Kastylevsky and Descotes, 1998). Animal models of anaphylaxis and other systemic hypersensitivity reactions do not appear to be sufficiently reliable for routine hazard identifica- tion in drug development. There are some animal models that might prove to be useful but need to be extensively evaluated before any recommendations can be made. At this time, the popliteal lymph node assay and adaptations of the local lymph node assay appear to be the most promising (Gutting et al., 1999; Manetz and Meade, 1999). Development of sensitive and specific methods for hazard identification of poten-tially allergenic and/or autoimmunogenic drugs is an important goal for both drug companies and regulatory agencies.

Choquet-Kastylevsky, G., Descotes, J., 1998. Toxicology 129, 27-35. Gutting, B.W., Schomaker, S.J., Kaplan, A.H., Amacher, D.E., 1999. Toxicological Sciences 5 1, 71-79. Manetz, T.S., Meade, B.J., 1999. Toxicological Science, 48, 206-217.

Glucocorticoids and Fetal Programming

Jonathan R. Seckl Molecular Medicine Centre, University of Edinburgh, Western Gen. Hosp., Edinburgh EH4 2XU, UK

Epidemiological studies in more than 20 distinct human populations have associated low weight or thinness at birth with a substantially increased risk of cardiovascular and metabolic disorders, including hypertension, insulin resistance/type 2 diabetes and ischaemic heart disease, in adult life (Barker, 1994). The concept of fetal ‘programming’ has been advanced to explain this phenomenon, but the underlying mechanism(s) are unclear. It has been suggested that fetal overexposure to glucocorticoids may, in part, explain the observations (Edwards et al., 1993). In support of this contention, prenatal glucocorticoid therapy reduces birth weight in humans. Moreover, steroids including glucocorticoids alter organ maturation and exert long-term organisational effects during specific ‘windows’ of development.

Recent experimental studies in rats (Seckl, 1998) have shown that birth weight is reduced following prenatal exposure to the synthetic glucocorticoid dexamethasone, which readily crosses the placenta, or to carbenoxolone, which inhibits 11/3-hydroxysteroid dehydrogenase type 2 (1 I p-HSD2), the physiological feto-placental ‘barrier’ to the higher levels of glucocorticoids in the maternal circulation. Whilst the offspring regain the weight deficit by weaning, as adults they exhibit permanent hypertension, hypergly caemia and increased hypothalamic-pituitary-adrenal (HPA) axis activity. Indeed, physiological varia- tions in placental 1 lb-HSD2 activity near term correlate directly with fetal weight. Finally, maternal protein restriction, which also ‘programmes’ offspring hypertension and hyperglycaemia, itself attenuates selectively placental 11 P-HSD2 activity. The critical ‘window’ for glucocorticoid program- ming appears to be the last third of gestation in the rat. In humans, 11 fi-HSD2 gene mutations cause low birth weight and some studies show reduced placental 11 B-HSD2 activity in association with intrauterine growth retardation. Moreover, low birth weight babies have higher plasma cortisol levels throughout adult life, indicating HPA axis programming also occurs in humans.

The molecular mechanisms are beginning to be unravelled. Prenatal dexamethasone permanently increases hepatic expression of PEPCK, the rate-limiting enzyme of gluconeogenesis (Nyirenda et al., 1998). This may be due to permanent changes in the expression of specific transcription factors in the hepatocyte. Key amongst these nuclear proteins may be the glucocorticoid receptor (GR) itself which

Abstructs 15

shows increased expression in the liver. Moreover, prenatal dexamethasone permanently reduces GR expression in the hippocampus, a critical brain region for glucocorticoid feedback upon the HPA axis, reducing feedback sensitivity and thus elevating circulating glucocorticoid levels. The different directions of programming of GR in different tissues appear to reflect specific effects upon one or more of the multiple tissue-specific alternate first exons/promoters of the GR gene (McCormick et al., 2000). Transcriptional programming of GR and other genes also appears to underlie the hypertension other behavioural aspects of the prenatal glucocorticoid exposure phenotype. Overall, the data suggest that both pharmacological and physiological exposure to excess glucocorticoids in the prenatal period can programme cardiovascular, metabolic and neuroendocrine pathologies in adult life. Glucocorticoid programming may explain. in part, the association between fetal events and disorders in adult life.

Barker, D.J.P., 1994. Mothers, babies and disease in later life. BMJ Publishing Group, London. Edwards, C.R.W., Benediktsson, R., Lindsay, R., Seckl, J.R., 1993. Lancet 341, 355.. 7. McCormick, J., Lyons, V., Jacobson, M., Diorio, J., Nyirenda, M., Weaver, S.. Yau, J.L.W., Ester, W., Meaney, M.J., Seckl, J.R., Chapman, K.E., 2000. Molec Endocrinol (in press). Nyirenda, M.J., Lindsay, R.S., Kenyon, C.J., Burchell, A., Seckl, J.R., 1998. J. Clin. Invest. 101, 2174-8 1. Seckl, J.R., 1998. Clinics Perinatol 25, 939-64.

Neuroendocrine Actions of Hormonal Disruptors

Nigel Brooks Molecular Endocrinology Group, Zeneca Central Toxicology Laboratory, Alderley Park, Macclesfield, Cheshire SK10 4TJ, UK

The neuroendocrine system controls and integrates many of the physiological processes of the body, including the reproductive system. A major goal of this laboratory is to understand how chemicals interact with hormone receptors to have adverse effects on reproductive health. In particular we are investigating how exposure to steroid hormones (and chemicals which mimic the actions of steroids --~ so called environmental estrogens) during critical windows in development, can have permanent organisational effects on the structure and function of the neuroendocrine system. These are effects which often do not manifest themselves until adulthood when they are reflected as altered patterns of reproductive hormones and characteristics of sexual behaviour. We have demonstrated that exposure to the synthetic estrogen, diethylstilbestrol, during early neonatal development, results in structural changes in the sexually dimorphic nucleus of the pre-optic area in the hypothalamus and altered basal and GnRH stimulated LH secretion from the pituitary gland. We have also highlighted an important role for specific neurotrophic factors which mediate the effects of steroids on neural growth and survival in those hypothalamic regions which are responsible for these changes.

The effects of estrogens are mediated by estrogen receptors which belong to a superfamily of ligand-activated transcription factors. Estrogen receptors are promiscuous. They allow many hundreds of different chemicals to bind to them. Until recently, the effects of estrogens were thought to be mediated by a single receptor, ERm. However, the discovery of a second estrogen receptor, ERl3 has prompted a re-evaluation of the molecular basis for estrogen action. Our objective has been to characterise the molecular basis for diverse responses to natural and xenobiotic estrogens, and in particular the selectivity of chemicals for agonism and antagonism at different forms of the estrogen receptor and the differential interaction of these receptors with responses elements on estrogen-responsive genes. Structural differences in these estrogen receptors leads to differences in relative binding affinities of xenobiotic estrogens. Furthermore, ERB has a unique tissue distribution compared with ERc(, which could lead to tissue-specific biological responses.