. 3
( 51 .)


prostaglandin concentrations via production of 15-hydroxy prostoglandin dehy-
drogenase, and thus may contribute to myometrial quiescence (not stimulation)
during most of the pregnancies (McKeown and Challis, 2003).
From the comparative and evolutionary perspective, CRH remains a prime can-
didate for research into the regulation of human pregnancy and fetal development.
6 M. L. Power and J. Schulkin


Rate of increase in maternal serum CRH 0.32








Preterm Term
Figure I.4 The rate of increase in maternal serum CRH concentration is greater in pregnancies
destined to deliver preterm (adapted from Leung et al., 2001, with permission). Data are
the means and 95% con¬dence intervals

Anthropoid primates are the only species known to produce placental CRH during
pregnancy (Robinson et al., 1989; Bowman et al., 2001). Understanding this appar-
ently unique anthropoid primate adaptation may be key to understanding the nor-
mal course of human pregnancy, and metabolic disruptions of pregnancy. This
will likely require the further development of nonhuman primate models of
human pregnancy, fetal development, and placental function.

Origins of adult-onset disease

That events in utero affect pregnancy outcome is a fact. Preterm birth and IUGR
are the most obvious, and possibly the most signi¬cant, examples of events in
utero leading to post natal morbidity and mortality. What is new is the evidence
that birth outcomes heretofore considered successful might lead to poor health
outcomes in adult life. Epidemiologic studies have indicated that the effects of
birth size on latter disease extend into the normal birth weight range, and thus are
not restricted to the serious effects of IUGR or premature birth (Barker, 1991;
Barker et al., 1993; Curhan et al., 1996a, b).
This realization has led to a concerted search for mechanisms. Much of this
search has centered on excessive or inappropriate activation of the HPA axis, both
maternal and fetal. Activation of the fetal hypothalamic“pituitary“adrenal (HPA)
axis is a common characteristic across species that results in increased output of
fetal glucocorticoids, which contribute to mechanisms associated with the onset
7 Introduction: brain and placenta, birth and behavior, health and disease



Odds ratio




Neonatal death RDS IVH Chronic lung disease
Figure I.5 A single course of antenatal steroids signi¬cantly decreases the risk of neonatal death,
RDS, and IVH, but does not decrease the incidence of chronic lung disease. Data
(means and 95% con¬dence intervals) are from Dudley et al. (2003)

of parturition and to the normal maturation of fetal organ systems. The fetus
responds to an adverse intrauterine environment with precocious HPA activation,
and premature upregulation of critical genes at each level along the axis. Thus in
compromised pregnancies the fetus may be exposed inappropriately to sustained
elevations of glucocorticoids.
An important theme in the chapters by Debra Sloboda and colleagues (Chapter 4)
and Jonathon Seckl and colleagues (Chapter 5) is that glucocorticoids are potent
steroids that have organizing effects on fetal organs. Key targets for in utero pro-
gramming of physiology include glucocorticoid receptor gene expression and the
CRH system. Sloboda and colleagues review data from animal models concerning
the effects of exogenous glucocorticoids on pregnancy and fetal development.
The use of glucocorticoids to mature fetal lung tissue prior to preterm birth
has had a signi¬cant positive effect on neonatal morbidity and mortality. A single
course of antenatal corticosteroids signi¬cantly reduces the risk of respiratory dis-
tress syndrome (RDS), intraventricular hemorrhage (IVH), and neonatal mortal-
ity, although it does not reduce the overall incidence of chronic lung disease
(Dudley et al., 2003; Figure I.5). However, animal studies, such as the one described
in Chapters 4 and 5, have demonstrated that glucocorticoid administration in late
gestation can result in IUGR and signi¬cant alterations in metabolic and HPA axis
function and regulation. This raises cautionary warnings concerning both the use
of multiple doses of glucocorticoids to mature fetal lung tissue in pregnancies at
risk for preterm birth, and the accuracy with which pregnancies at risk for preterm
8 M. L. Power and J. Schulkin

birth can be predicted. The administration of glucocorticoids to a fetus that is car-
ried to term may not be benign.
Seckl and colleagues review the evidence (epidemiologic and physiologic) con-
cerning the programming of fetal physiology in utero. They present the case that
glucocorticoids play important roles in both appropriate and inappropriate pro-
gramming. They discuss the placental enzyme 11 -hydroxysteroid dehydrogenase
type 2, which acts as a barrier to glucocorticoids. Regulation of this enzyme may
serve to increase or decrease fetal exposure to maternal glucocorticoids. They dis-
cuss programming of the cardiovascular system, liver, pancreas, and brain by gluco-
corticoids and the subsequent increased vulnerability to adult onset diseases such
programming can engender.
Elysia Davis and colleagues continue the theme of stress, HPA activation, gluco-
corticoids and their effects on pregnancy. Their focus is on human behavior and
human data. They discuss a neurobiologic model in which maternal psychosocial
stress in¬‚uences developmental outcomes that are mediated, in part, via maternal“
placental“fetal neuroendocrine mechanisms. They present data on the conse-
quences of stress during pregnancy on neuroendocrine processes and fetal and
infant development. They also note the uniqueness of placental CRH in anthropoid
primates, and that placental CRH and cortisol may contribute to the organization
of the fetal central nervous system (Sandman et al., 1997; Florio and Petraglia,

Feed-forward regulation of CRH by glucocorticoids

Until recently, it was the received view that glucocorticoids restrained CRH pro-
duction. The model system was the HPA axis, wherein hypothalamic CRH stimu-
lated pituitary adrenocorticotrophic hormone (ACTH) production, which in turn
stimulated cortisol production by the adrenals. Cortisol crossed the blood“brain
barrier and exerted negative feedback on CRH neurons in the hypothalamic para-
ventricular nucleus (PVN), restraining the system. It is a curious fact that inde-
pendent groups of researchers, working on different CRH producing organs (the
brain and the placenta) found at roughly the same time that glucocorticoids can
also stimulate CRH production. Glucocorticoid added to cultured human placen-
tal tissue resulted in the upregulation of CRH gene expression (Robinson et al.,
1988; Jones et al., 1989; Figure I.6). In several regions of the brain (e.g. amygdala
and bed nucleus of the stria terminalis, and areas of the paraventricular region of
the hypothalamus that project to the brainstem) CRH messenger ribonucleic acid
(mRNA) expression similarly is upregulated by glucocorticoids (Swanson and
Simmons, 1989; Makino et al., 1994; Watts and Sanchez-Watts, 1995; Figure I.7).
9 Introduction: brain and placenta, birth and behavior, health and disease



Sulfatase Cortisol Sulfate

Placenta Fetus

d5 5
CRH peptide pg/108 cells/h
CRH mRNA (arbitrary units)

* 4

* 3



0 0
(b) Dexamethasone
Figure I.6 (a) A positive feedback loop is established between the fetus and the placenta, where
cortisol (either maternal or from the fetal) adrenal upregulates placental CRH messenger
ribonucleic acid (mRNA) expression and CRH peptide content. Placental CRH stimulates the
fetal HPA axis to produce more cortisol. (b) Dexamethasone increases CRH mRNA and CRH
peptide concentration in cultured placental cells. From Robinson et al. (1988), with permission

The majority of CRH neurons within the PVN are clustered in the parvicellular
division. Other regions with predominant CRH-containing neurons are the lateral
bed nucleus of the stria terminalis and the central region of the central nucleus
of the amygdala (CeA). To a smaller degree, there are CRH cells in the lateral
10 M. L. Power and J. Schulkin

Figure I.7 Corticosterone decreases CRH expression in the rat PVN but increases CRH expression in
rat central nucleus of the amygdala (CeA). From Makino et al. (1994), with permission

hypothalamus, prefrontal and cingulate cortex. In brainstem regions, CRH cells are
clustered near the locus coeruleus (Barringtons™ nucleus) (Valentino et al., 1995),
parabrachial region and regions of the solitary nucleus (Figure I.8).
In this volume, Watts reviews neural regulation of CRH axons. He emphasizes that
there is cell speci¬city in how CRH and the CRH gene is regulated. Glucocorticoids
repress CRH gene expression in the hypothalamic paraventricular nucleus (the
familiar negative feedback system of the HPA axis), but in other regions (e.g. CeA)
glucocorticoids stimulate CRH gene expression, and in others glucocorticoids have
no effect at all. Even within the PVN, basal levels of glucocorticoids appear neces-
sary to sustain CRH gene expression. Adrenalectomized rats show a suppressed
CRH response in the PVN to hypovolemia rather than an exaggerated response
(Tanimura and Watts, 2000). It turns out that the ˜usual™ negative restraint of CRH
by glucocorticoids has actually only been seen in one (admittedly important) set
of CRH expressing neurons. Thus the increase in human placental CRH mRNA
expression when exposed to glucocorticoids does not appear to represent an
unusual circumstance. The current state of knowledge supports the idea that
glucocorticoids have variable effects on CRH regulation depending on cell type,
and intracellular and extracellular factors. The original idea of glucocorticoids
functioning as a negative feedback response molecule has been expanded to a more
¬‚exible, context-oriented understanding of regulation.
11 Introduction: brain and placenta, birth and behavior, health and disease



. 3
( 51 .)