. 7
( 51 .)


(Roberts and Cooper, 2001). It is associated with abnormal placentation, due to the
29 Placental expression of neurohormones and other neuroactive molecules

altered cytotrophoblast proliferation and invasion of endometrium, causing a
reduced placental perfusion, the impairment of placental angiogenesis with the
insuf¬ciency and failure of spiral arteries remodeling (Roberts and Cooper, 2001).
The reduced and/or low perfusion of placenta and the fetus is consequently the
main cause of fetal growth restriction (FGR), a PE complication. Maternal con-
centrations of CRH are greatly increased in PE (Laatikainen et al., 1991; Petraglia
et al., 1996a), in presence of plasma CRH-BP levels signi¬cantly lower than in
healthy controls (Perkins et al., 1995; Petraglia et al., 1996a). In addition, also cord
venous plasma CRH concentrations are signi¬cantly higher in patients with PE
and higher than in cord arterial plasma, indicating the secretion of CRH from the
placenta into the fetal circulation (Laatikainen et al., 1991). In addition to CRH,
also the remaining hormones with vasodilatory actions and involved in the stress
response, such as ACTH and cortisol, are increased in the fetuses from PE preg-
nancies (Goland et al., 1995) as well as in the FGR fetuses (Goland et al., 1993).
Concentrations of CRH in the fetal circulation are signi¬cantly increased in
pregnancies complicated by abnormal umbilical artery ¬‚ow velocity waveforms,
thus representing a stress-responsive compensatory mechanism in the human pla-
centa (Giles et al., 1996). It is not known whether this deranged secretion is part of
the primary pathophysiology of these conditions or occurs as a secondary response
to the increased vascular resistance in abnormal pregnancies. The concentration of
CRH in the fetal circulation is signi¬cantly increased in pregnancies complicated
by abnormal umbilical artery ¬‚ow velocity waveforms, thus representing a stress-
responsive compensatory mechanism in the human placenta.

Peptide signaling and fetal/maternal endocrinology

Neurohormones produced by human placenta, decidua and fetal membranes are
secreted into maternal and fetal circulation, and amniotic ¬‚uid. In these compart-
ments, levels may increase from early to term pregnancy, or just at term. However,
the role of these changes in the regulation of maternal and fetal endocrinology may
be of some relevance (Reis and Petraglia, 2001), as well as the putative role of the
placenta as the central organ in this bidirectional system. A typical example is the
modulation of the HPA axis activity and hormone secretion in pregnancy.
The activity of the maternal HPA axis is increased in pregnant women, and high
levels of free and bound cortisol circulate in pregnant women (Challis et al., 2000;
Florio et al., 2002d). Indeed, hypercortisolemia is characteristic of pregnancy, and
the correlation between plasma CRH and salivary or urinary free cortisol levels
would suggest that placental CRH is responsible for these alterations, even though
other factors may act in modulating maternal HPA axis function in pregnancy
(Goland et al., 1994; Challis et al., 2000). However, some discrepancies occur
30 F. Petraglia et al.

between CRH and ACTH. In fact, although plasma ACTH levels increase through-
out pregnancy, they remain within the normal range of non-pregnant women
(Barbieri, 1994). This is probably because CRH-BP counteracts the secretory
action of CRH on both maternal pituitary and placental ACTH (Potter et al., 1992;
Petraglia et al., 1993a; 1996b). Furthermore, injecting pregnant women with
exogenous CRH does not induce an increase of circulating ACTH, suggesting that
high cortisol levels may desensitize maternal pituitary corticotrophs (Schulte and
Healy, 1987; Sasaki et al., 1989; Schulte et al., 1990).
Thus, some discrepancies exist in the HPA axis regulation between pregnant and
non-pregnant women. In fact, the administration of exogenous glucocorticoid to
pregnant women may increase maternal plasma and placental levels of immuno-
reactive CRH (Marinoni et al., 1998), decreasing cortisol (Tropper et al., 1987;
Marinoni et al., 1998) and ACTH levels (Marinoni et al., 1998). To date, it is
unclear whether maternal plasma ACTH originates from the maternal pituitary,
placenta, or both. The diurnal rhythm for plasma ACTH, cortisol and -END is
maintained in pregnant women; however, CRH does not have a circadian rhythm
(Chan et al., 1993; Petraglia et al., 1994a). These ¬ndings and the fact that the
changes of plasma CRH do not correlate with those of ACTH or cortisol through-
out normal pregnancy or out of the time of labor (Chan et al., 1993; Florio et al.,
2002d) underlie the differences in HPA regulation in pregnancy and support the
following statements:
(1) pituitary ACTH release is regulated centrally;
(2) placental CRH is not the only regulator of maternal ACTH and cortisol
levels (Florio et al., 2002d).
Placental CRH secreted into the fetal circulation may stimulate the production
of pituitary ACTH as well as of adrenal hormones (Figure 1.4). The effect of CRH
on fetal pituitary ACTH release is potentiated by arginine vasopressin and possibly
mediated by cAMP, and may be antagonized by dexamethasone (Vale et al., 1993).
Recent studies revealed a direct effect of CRH on dehydroepiandrosterone sulfate
(DHEA-S) release from cultured fetal adrenal cells (Smith et al., 1998). Expression
of mRNA encoding type 1 CRH receptor was identi¬ed in mid-gestation human
fetal adrenals (Smith et al., 1998) suggesting that the fetal adrenal cortex may be
directly responsive to CRH (Figure 1.4). Placenta of humans and higher primates
uses DHEA-S supplied by the fetal adrenals as the main substrate for estrogen syn-
thesis, and estrogens produced by the placenta play a pivotal role in the endocrine
control of pregnancy and induce many of the key changes involved at parturition
(Challis et al., 2000).
Human CRH increased DHEA-S production by cultured human fetal adrenal
cortical cells in a dose-dependent fashion, being as effective as ACTH at stimulating
31 Placental expression of neurohormones and other neuroactive molecules

Mother Placenta Fetus



Pituitary Pituitary


Adrenals Adrenals

Placental Cortisol
aromatization DHEA-S

Fetal lung
Uterine contractility

Figure 1.4 In vitro evidences for a role of CRF on placental, maternal and fetal endocrinology.
Placental CRF at the end of pregnancy stimulates fetal pituitary ACTH secretion, which
in turn stimulates fetal adrenal cortisol and DHEA-S production. The increasing
concentrations of cortisol, in addition to maturating enzymes in organs critical for
postnatal existence, further stimulate production of placental CRF by a feed-forward
mechanism. The increasing production of DHEA-S provides additional substrate for
placental aromatization to estrogen, which triggers the cascade leading to labor and
delivery. In addition, CRF modulates directly the myometrial contractility, and indirectly by
stimulating the release of uterotonic substances (prostaglandins, oxytocin)

DHEA-S production, although it was considerably less potent than ACTH in stim-
ulating cortisol synthesis (Smith et al., 1998). CRH did not alter cell number, indi-
cating that it is not mitogenic for fetal adrenal cortical cells. Therefore, placental
CRH production, which rises exponentially during human pregnancy, may play a
key role in promoting DHEA-S production by the fetal adrenals, which could lead
to an increase in placental estrogen synthesis (Smith et al., 1998; Challis et al., 2000;
Florio et al., 2002d).
With respect to neurohormones and fetal adrenal, several ¬ndings suggest a role
for chromogranin A (CgA; Box 1.3). In fact, during pregnancy, the highest CgA
levels found in the umbilical cord blood and mainly at parturition, are most probably
of fetal adrenal origin and may have a role in preparing the fetus for the extra-uterine
life (Florio et al., 2002c). In fact, CgA is costored and coreleased with catecholamine
(Taupenot et al., 2003), and both increase signi¬cantly in cord blood during the
32 F. Petraglia et al.

Box 1.3 Chromogranin A

The CgA is a 49-kDa glycoprotein of 439 amino acids, belonging to the granin
family of regulated secretory proteins initially described in the core of the adre-
nal medullary chromaf¬n granules, but subsequently identi¬ed in secretory
granules throughout the neuroendocrine system and in a variety of neurons,
both central and peripheral (Taupenot et al., 2003). Immunohistochemical
studies have shown a widespread distribution of CgA immunoreactivity in neu-
roendocrine cells and in tumors originating from these cells and serum levels
are raised in patients with neuroendocrine tumors (Taupenot et al., 2003). The
CgA is mainly costored within the granules with catecholamines, ENK, vasoin-
testinal peptide (VIP), substance P and NPY, and coreleased by exocytosis with
catecholamines and NPY from storage vesicules (Taupenot et al., 2003). In plasma,
CgA levels are increased in response to large-amplitude stressful events, that is
hypoxia, physical exercise or other stressful events (Taupenot et al., 2003).

Expression and localization
Syversen et al. (1992) showed the presence of CgA mRNA and peptide in
intrauterine tissues as placental trophoblast, decidua and fetal membranes. The
CgA immunoreactivity was demonstrated by immuno¬‚uorescence studies of
isolated trophoblasts and decidual cells from term placentas. Double immuno-
¬‚uorescence of isolated trophoblasts showed colocalization of CgA with hPL
and hCG. Since syncytiotrophoblasts are the placental source of hPL, that indi-
cates that this cell is one site of CgA production. By Northern blotting, a distinct
band corresponding to CgA mRNA was demonstrated in the human placental
cell line (TPA-30-1), whereas in placental homogenates an mRNA band of a
slightly larger size was found (Syversen et al., 1992).

Biological ¬‚uids
CgA is measurable in maternal and fetal plasma, in umbilical cord blood and in
amniotic ¬‚uid (Syversen et al., 1992; Moftaquir-Handaj et al., 1995; Florio et al.,
2002c). No signi¬cant differences were found in maternal CgA levels during
pregnancy compared with levels out of pregnancy, even if median CgA level in
maternal sera at term tended to be higher than at 6“11 weeks or in sera from
non-pregnant women (Syversen et al., 1992). In umbilical cord sera median
CgA level was signi¬cantly higher than in term sera, whilst in amniotic ¬‚uid
median CgA value was signi¬cantly higher at term than in second trimester
(Syversen et al., 1992).
33 Placental expression of neurohormones and other neuroactive molecules

With respect to labor, umbilical cord plasma and amniotic ¬‚uid levels of CgA
were higher in women who had spontaneous VD than in those delivered by ECS
(Moftaquir-Handaj et al., 1995; Florio et al., 2002c), suggesting a fetal origin,
whilst no change was detected in maternal circulation (Florio et al., 2002c).

stress of delivery (Padbury and Martinez, 1988; Moftaquir-Handaj et al., 1995).
The sympathoadrenergic system is activated to withstand the stress of birth, and
several other neuropeptides, other than CgA, are secreted by the adrenal medulla
into cord blood at the time of delivery, including catecholamines, enkephalins
(ENK) and NPY (Poyner et al., 2002; Taupenot et al., 2003). The ability of the
sympathoadrenal system to develop a response is essential for fetal life (during
which the fetus grows under a state of relatively low oxygen tension), as well as for
survival during parturition, when compression of the umbilical cord and the placen-
tal circulation occurs because of uterine contractions that intermittently deprive
the infant of oxygen (Padbury and Martinez, 1988). The CgA levels in the fetal cir-
culation at birth are associated with high levels of NPY (Lundberg et al., 1986) and
catecholamines (Wang et al., 1999) so that CgA could inhibit the excessive cate-
cholamine release to counteract the vasoconstrictive effects of catecholamines and
NPY (Poyner et al., 2002; Taupenot et al., 2003). As CgA also possesses some func-
tion related to vasodilation (Poyner et al., 2002; Taupenot et al., 2003), umbilical
cord CgA release could help to regulate and prevent the vascular constrictive
effects of catecholamines and NPY when they are secreted in excess.

Peptide signaling and the control of myometrial contractility

For parturition to occur, the cervical connective tissue and smooth muscle must be
capable of dilation to allow the passage of the fetus from the uterus, but the uterus
itself must be converted from a quiescent structure with dysynchronous contrac-
tions to an active coordinately contracting organ. On this regard, the entire preg-
nancy may be viewed as the result of the constrant equilibrium between factors
activating and others inhibiting myometrial contractility, so that the term or
preterm labor is the consequence of shift-forward activating (uterotonic) factors,
with the decrease of the role of quiescence (inhibitors) neurohormones (Figure 1.5).

(A) Role of OT
The major uterotonic factors triggering uterine contractility are OT and PGs
(Petraglia et al., 1996d; Challis et al., 2000). Historically, OT was assumed to be the
initiating factor of parturition because clinical administration initiates labor
which is indistinguishable from spontaneous labor. After this ˜proof of concept™
experiment, the role of OT was extensively investigated in many animal species.
34 F. Petraglia et al.



PGF2 and PGE2 release DHEA-S

Influx of Ca2

Human myometrium

Cell membrane hyper polarization

Adenylyl cyclase

NO formation


Figure 1.5 Human pregnancy may be viewed as the result of a constant equilibrium between
activators and inhibitors of myometrial contractility. In particular, OT, PGs, CRF and NPY are
able to stimulate, whilst CGRP and PTHrP inhibit the activity of the prenant myometrium
activing on the uterine contractile machine


. 7
( 51 .)