REPRODUCTION
RESEARCH
Nutritional manipulation between early to mid-gestation: effects
on uncoupling protein-2, glucocorticoid sensitivity, IGF-I
receptor and cell proliferation but not apoptosis in the ovine
placenta
M G Gnanalingham, P Williams, V Wilson, J Bispham, M A Hyatt, A Pellicano, H Budge,
T Stephenson and M E Symonds
Institute of Clinical Research, Centre for Reproduction and Early Life, University of Nottingham, Nottingham
NG7 2UH, UK
Correspondence should be addressed to M E Symonds at Academic Division of Child Health, School of Human Development,
University Hospital, Nottingham NG7 2UH, UK; Email: michael.symonds@nottingham.ac.uk
Abstract
In sheep, modest maternal nutrient restriction (NR) over the period of rapid placental growth restricts placentome growth and results in
offspring in which glucocorticoid action is enhanced. Therefore, this study investigated the placental effects of early to mid-gestational NR on
glucocorticoid receptor (GR), 11b-hydroxysteroid dehydrogenase type 2 (11bHSD2), uncoupling protein-2 (UCP2), and IGF type-I receptor
(IGF-IR) mRNA abundance together with cell proliferation and apoptosis as determined histologically, and the mitochondrial proteins
voltage-dependent anion channel and cytochrome c that are involved in apoptosis. Placenta was sampled at 80 and 140 days gestation (dGA;
term w147 dGA). NR was imposed between 28 and 80 days gestation when control and nutrient-restricted groups consumed 150 or 60%
respectively of their total metabolizable energy requirements. All mothers were then fed to requirements up to term. Total fetal placentome
weights were decreased by NR at 80 dGA but were heavier at 140 dGA following 60 days of nutritional rehabilitation. GR and UCP2 mRNA
abundance increased whilst 11bHSD2 mRNA decreased with gestational age. NR persistently up-regulated GR and UCP2 mRNA abundance.
11bHSD2 mRNA was reduced by NR at 80 dGA but increased near to term. IGF-IR mRNA abundance was only decreased at 80 dGA. Placental
apoptosis and mitochondrial protein abundance were unaffected by NR, whereas cell proliferation was markedly reduced. In conclusion,
placental UCP2 and local glucocorticoid action are affected by the gestational nutritional status and may result in the offspring showing
enhanced glucocorticoid sensitivity, thereby predisposing them to disease in later life.
Reproduction (2007) 134 615–623
Introduction et al. 1998, Heasman et al. 1998) and causes a reduction
in the mean weight of individual placentomes (Clarke
Maternal nutrition is a major factor determining
et al. 1998), but the mechanism mediating this response
placental growth and development which in turn can
has not been established. These adaptations within the
have a strong influence on nutrient supply to the fetus
placenta occur in conjunction with a reduced potential
(McMillen & Robinson 2005). Importantly, the impact of
a reduction in maternal food intake at defined stages of capacity to inactivate maternal cortisol through the
pregnancy is not confined to the fetal period but can enzyme 11b-hydroxysteroid dehydrogenase (11bHSD)
extend well into later life (Barker 2001, McMillen & type 2 (Whorwood et al. 2001), which itself may be in
Robinson 2005, Symonds et al. 2005). In the sheep, like response to a decrease in maternal plasma cortisol
the human, the period of pregnancy in which placental (Bispham et al. 2003). Local tissue glucocorticoid
growth is greatest is from the time of implantation (or hormone action is, however, regulated by expression of
attachment in sheep) to around mid-gestation (Ehrhardt the glucocorticoid receptor (GR, type 2) and isoforms of
& Bell 1995). Ovine placental mass peaks at w80 days 11bHSD at the level of gene transcription (Bamberger
gestation (Ehrhardt & Bell 1995) and is followed by et al. 1996). 11bHSD type 1 (11bHSD1) acts as an
marked change in its structural properties and confor- 11-oxoreductase to catalyze the conversion of cortisone
mation (Stegmann 1974). During this period, maternal to bioactive cortisol, thereby amplifying glucocorticoid
nutrient restriction (NR; i.e. w50%) has significant action. Conversely, 11bHSD type 2 (11bHSD2) acts as
effects on placental growth and morphology (Clarke an 11-dehydrogenase, catalyzing the inactivation of
q 2007 Society for Reproduction and Fertility DOI: 10.1530/REP-06-0369
ISSN 1470–1626 (paper) 1741–7899 (online) Online version via www.reproduction-online.org
616 M G Gnanalingham and others
cortisol to cortisone, maintaining the specificity of the A) 80 days gestation, coincident with the peak in
mineralocorticoid receptor for aldosterone (Stewart & placental weight and following w50 days of
Krozowski 1999). One aim of our study was, therefore, maternal NR.
to determine whether glucocorticoid action within the B) 140 days gestation, near to term (w147 days),
placenta may be permanently reset following maternal following w60 days of nutritional rehabilitation.
NR. Glucocorticoids also have an important role in
regulating uncoupling protein (UCP) abundance in the
mitochondria of the fetus and newborn (Gnanalingham
et al. 2006) but the extent to which this may extend to the Results
placenta is currently unknown. Total weight of the fetal component of the placenta was
UCP2 has the most widespread tissue abundance and decreased by early to mid-gestational NR at 80 days
is present in the uterus of mice (Pecquer et al. 2001, gestation compared with controls (C 507.8G64.4; NR,
Rousset et al. 2003). Its function remains a subject of 326.2G20. 3 g (P!0.05)), but were heavier at 140 days
intense debate (Stuart et al. 2001), but the abundance of gestation following 60 days of nutritional rehabilitation
UCP2 mRNA and/or protein has been shown to be (C 183.6G9.6; NR, 364.4G21. 3 g (P!0.05)). Fetal
developmentally regulated in both the lung and adipose weights were not different between groups at either
tissue of sheep where one postulated role includes the gestational age, but near-term NR fetuses possessed more
regulation of apoptosis (Voehringer et al. 2000). In both perirenal adipose tissue than controls (C 19.2G1.9;
these tissues, the abundance of UCP2 mRNA and protein NR 23.2G1.4 g (P!0.05)) and had larger kidneys
is strongly influenced by the current and the past (C 16.5G2.1; NR 20.3G1.2 g (P!0.05)). There were no
nutritional states (Mostyn et al. 2003, Gnanalingham differences in weights between groups with respect to all
et al. 2005a, 2005c). It is currently not known whether the other major organs sampled (data not shown).
UCP2 is present in the placenta nor whether it is UCP2, GR, and 11bHSD2 mRNA were all detected in
developmentally controlled or nutritionally responsive the fetal placentome at 80 and 140 days gestation
in utero. Therefore, an extension of our study was to (Fig. 1A–C). UCP2 and GR mRNA abundance increased
determine whether UCP2 is present in the ovine placenta (P!0.01) with gestational age and were further raised,
and if it is developmentally regulated. Other important compared with controls (P!0.01), by early to mid-
mitochondrial proteins whose function is related to gestational NR (Fig. 1A and B). Interestingly, 11bHSD2
UCP2 include voltage-dependent anion channel (VDAC; mRNA abundance was lower (P!0.01) in the placenta
Voehringer et al. 2000) located in the outer mito- of NR compared with control mothers at 80 days
chondrial membrane (Colombini 1979). VDAC along gestation (Fig. 1C). It then only decreased with
with UCP2 (Voehringer et al. 2000) could be responsible gestational age in controls with the result that by
for the release of cytochrome c from the intermembrane 140 days gestation 11bHSD2 mRNA abundance was
space, a process that has been implicated in the chain of significantly greater in placenta sampled from NR
events culminating in apoptosis (Crompton 1999). mothers compared with controls. Finally, irrespective
Placenta of clinically compromised pregnancies shows of maternal diet and gestational age UCP2 and GR
increased rates of apoptosis (Huppertz & Herrler 2005), mRNA abundance were positively correlated (Fig. 2).
but the extent to which apoptosis is enhanced by IGF-IR mRNA abundance was decreased at 80 days
maternal NR has not been examined. We therefore gestation by NR, although by 140 days gestation this
aimed to determine not only the effect of maternal food effect was negated by refeeding (Fig. 3).
intake on apoptosis but also the cellular location of Expression patterns of PCNA were similar between
VDAC and cytochrome c in the placenta. groups being specifically localized within the villi and
Maternal dietary manipulation in early pregnancy glandular epithelium (Fig. 4a and b). PCNA expression
results in a number of endocrine adaptations that may was, however, markedly decreased in the placenta of
potentially impact on placental function. These include nutrient-restricted mothers in which it was weakly
a reduction in the plasma concentrations of a range of expressed in all samples compared with moderate or
maternal metabolic hormones including cortisol, thyroid strong expression in all control animals. Apoptosis,
hormones, insulin-like growth factor (IGF)-I, and insulin as assessed by either TUNEL staining or the detection
(Bispham et al. 2003, Symonds et al. 2007). A reduction of the cleaved form of caspase 3, was unaffected by
in placental IGF type-I receptor (IGF-IR) is associated NR (C 3.6G0.2; NR 3.5G0.3 positive cells per
with intrauterine growth retardation (Reid et al. 2002, 200!magnification field). Both VDAC and cytochrome
Laviola et al. 2005). The extent to which the same effect c were located within the maternal uterine syncytium
on the IGF-IR may result from changes in maternal region of the placenta (see representative example for
nutrition is unknown and was a further aim of the present cytochrome c in Fig. 5). The abundance of neither
study. In order to achieve the aims of the study, placenta mitochondrial protein was influenced by NR or age and
was sampled from control and nutritionally manipulated mRNA expression of VDAC was similarly unaffected
singleton-bearing sheep at: (data not shown).
Reproduction (2007) 134 615–623 www.reproduction-online.org
Nutritional manipulation in the placenta 617
Figure 2 Positive relationship between uncoupling protein (UCP)-2
mRNA and glucocorticoid receptor (GR) mRNA abundance (R2Z0.77,
P!0.0001, where yZ0.80xC0.56) in all sampled fetal placentome tissue
(nZ20), irrespective of gestational age or nutritional group. Controls
represented by open symbols and nutrient restricted by closed symbols;
placenta sampled at 80 days gestation represented by diamonds and those
sampled at 140 days gestation represented by squares.
Discussion
Our study demonstrates the complexity of placental
adaptations following exposure of the mother to a period
of NR extending over more than one-third of gestational
length. As such, a global reduction in maternal food
intake from the time of uterine attachment and
continued throughout the time of maximal placental
growth restricts placental size in conjunction with
increased glucocorticoid action. The magnitude of
these changes then alters following the restoration of
the maternal diet from 80 days gestation when the
normal decrease in placental mass does not occur whilst
Figure 1 Effect of early to mid-gestational maternal nutrient restriction
on the abundance of (A) uncoupling protein-2 (UCP2), (B) glucocorti-
coid receptor (GR), and (C) 11b-hydroxysteroid dehydrogenase type 2
(11bHSD2) mRNA in fetal placentomes at 80 and 140 days gestation Figure 3 Effect of early to mid-gestational maternal nutrient restriction on
(dGA; term 147 dGA) from sheep that consumed 60% (nutrient the abundance of insulin-like growth factor type I receptor (IGF-IR) mRNA
restricted, NR) or 150% (control) of their metabolizable energy in fetal placentomes at 80 and 140 days gestation (dGA; term 147 dGA),
requirements between 28 and 80 dGA. Examples of mRNA expression from sheep that consumed 60% (nutrient restricted, NR) or 150% (control)
are included. Values are mean with their standard errors (nZ5 per of their metabolizable energy requirements between 28 and 80 dGA.
group). **P!0.01, mean value significantly different from control Examples of mRNA expression are included. Values are means with their
group at the same gestational age. Significant differences with standard errors (nZ5 per group). *P!0.05, mean value significantly
gestational age are indicated by adjoining lines. different from control group at the same gestational age.
www.reproduction-online.org Reproduction (2007) 134 615–623
618 M G Gnanalingham and others
Figure 4 Representative image showing reduced expression Figure 5 Representative image showing immunocytochemical detec-
of proliferating cell nuclear antigen (PCNA) in placenta following tion of cytochrome c in the ovine sheep placenta at mid-gestation, i.e.
maternal nutrient restriction between 28 and 80 days gestation. within the maternal uterine syncytium at a magnification of 400!.
Example from either (a) control fed or (b) nutrient-restricted mother Example of staining (a) with (i.e. positive) or (b) without (i.e. negative)
with PCNA being present within the villi and glandular epithelium at a inclusion of cytochrome c antibody.
magnification of 200!.
mRNA abundance for both the GR and 11bHSD2 are Gnanalingham et al. 2005a), which are the two organs
increased. These placental responses appear to be whose growth is increased in previously nutrient-
mediated in part by a reduction in cell proliferation restricted fetuses.
and are likely to have a pronounced effect on both It is known that for adipose tissue during the period of
nutrient supply to the fetus as well as its endocrine exponential growth there is a strong correlation between
environment. Indeed, the enhanced placental gluco- total fat mass and both local glucocorticoid action and
corticoid action with NR is likely to be the mechanism UCP2 expression (Gnanalingham et al. 2005a). In the
by which cell proliferation is reduced (Rogatsky et al. present study on the ovine placenta, we found a positive
1997, Demasi et al. 2007). As a consequence, tissue relationship between UCP2 and GR mRNA with
endocrine sensitivity of the fetus and offspring are reset, gestational age that is clearly not related to placental
particularly with respect to glucocorticoids (Whorwood mass per se. The exact mechanisms by which both GR
et al. 2001, Gnanalingham et al. 2005a, 2005c), thereby and UCP2 mRNA are up-regulated in the placenta
placing them at a potentially increased risk of cardio- remain to be established. Endocrine regulation of UCP2
vascular or metabolic compromise in later life. Impor- during development has been most widely studied in
tantly, these adaptations are greatest in the perirenal fetal adipose tissue in which cortisol and the biologically
adipose tissue (which constitutes w80% of fetal adipose active thyroid hormone, triiodothyronine (T3), both
tissue) and the kidney (Whorwood et al. 2001, appear to be involved (Gnanalingham et al. 2005d ).
Reproduction (2007) 134 615–623 www.reproduction-online.org
Nutritional manipulation in the placenta 619
These hormones can affect local glucocorticoid action (Whorwood et al. 2001). The time course of this
by specifically influencing the expression of GR and adaptation corresponds to the period in which
11bHSD isoforms. Indeed, regulation of fetal UCP2 maternal plasma cortisol adapts from being reduced
mRNA by cortisol is in accord with the effects of during the period of NR to subsequently rising as the
umbilical cord occlusion which results in a precocious maternal diet is restored to the same level as controls
rise in UCP2 mRNA in the fetal lung and adipose tissue (Bispham et al. 2003). Taken together, these findings
(Gnanalingham et al. 2005e). Under these adverse indicate that it is not only the prevailing maternal
hypoxic conditions, however, it is cortisol, rather than cortisol that influences placental and, thus, fetal
T3, which regulates UCP2 abundance. In the present exposure to cortisol but it is the magnitude of change
study, however, the endocrine mechanisms may be in the maternal circulation throughout pregnancy. As
maternally driven as we have previously established that such, a transient rise in maternal cortisol in late
both maternal plasma cortisol and thyroid hormones are gestation has no effect on fetal cortisol (Edwards &
decreased over the period of NR (Bispham et al. 2003). McMillen 2001). Indeed, the increase in placental
It remains to be established whether these directly 11bHSD2 activity with gestation that only occurred in
impact on placental glucocorticoid action and in the nutrient-restricted group would be predicted to
particular increase its sensitivity to cortisol. reduce fetal cortisol exposure (Langley-Evans et al.
Interestingly, the increase in mitochondrial mRNA 1996). This may therefore be the mechanism by which
abundance for UCP2 was not accompanied by any fetal sensitivity is raised in these offspring, which is in
change in VDAC or cytochrome c protein. This contrasts direct contrast to the response of adipose tissue in
with adipose tissue in which nutritional programming of offspring born to mothers nutrient restricted in late
all of these proteins is observed (Mostyn et al. 2003) gestation (Gnanalingham et al. 2005a).
and emphasizes their tissue-specific regulation In contrast to the GR, mRNA abundance for the IGF-IR
(Gnanalingham et al. 2005b, 2006). Unfortunately, we was decreased in placenta of NR mothers at 80, but not
were unable to confirm whether UCP2 protein was 140, days gestation. This coincided with the stage at which
similarly affected because of the current unavailability of mean placentome mass was reduced and is in accord with
specific antibodies for ovine UCP2 (Gnanalingham et al.
findings in rats and humans in which intrauterine growth
2005f). The functional consequence of an increase in
retardation is accompanied with reduced placental IGF-IR
placental UCP2 remains to be established. One putative
(Reid et al. 2002, Laviola et al. 2005). It should be noted,
role for UCP2 is in apoptosis (Voehringer et al. 2000),
however, that the placenta is markedly different between
which, in the placenta, is determined in part by
sheep, rats, and humans which have discoid, hemochorial
glucocorticoid sensitivity (Waddell et al. 2000).
placenta, whereas in sheep placenta are cotyledonary
Apoptosis was, however, unaffected by NR in our
synepitheliochorial that may represent an evolutionary
study. Interestingly, we have shown for the first time
that both VDAC and cytochrome c are located within the development and can limit the transport of some
syncytial region of the placenta, which is the major site molecules from the mother to fetus (Carter & Mess 2007).
of glucose transport across the placenta (Dandrea et al. Therefore, in the sheep, restoration of the maternal diet
2001). This process obviously requires appreciable together with the concomitant rise in maternal plasma
amounts of energy, which will need to be met within IGF-I (Bispham et al. 2003) acts to restore placental
the mitochondria, and does not appear to be impaired by responsiveness. At the same time, this adaptation is
reduced maternal food intake and the reduction in accompanied by a maintenance of placental mass,
placental cell proliferation. increased fetal length at term, and resetting of the
The persistent increase in placental GR mRNA relationship between fetal plasma IGF-I and body
abundance with maternal NR was not accompanied dimensions (Heasman et al. 2000).
by a similar change in 11bHSD2 mRNA, which was In conclusion, we have shown that maternal NR
transiently decreased at 80 days, but then increased targeted between the time of conceptus implantation
near to term. These data extend previous findings in throughout peak placental growth results in a
which no effect of early to mid-gestational NR pronounced change in local glucocorticoid action
was found on 11bHSD1 mRNA at term and a within the placenta, one consequence of which is
reduction in 11bHSD2 was only present in mid- increased UCP2 mRNA abundance. These adaptations
gestation (Whorwood et al. 2001). Unlike the present occur in conjunction with a reduction in placental cell
study that utilized RT-PCR, previous work may have proliferation but in the absence of any affect on
been unable to detect this gene in term placenta apoptosis. Changes in glucocorticoid action could then
because the less sensitive technique of Northern contribute to changes in cortisol exposure of the fetus,
blotting was used (Whorwood et al. 2001). By term, thereby, causing both immediate (Whorwood et al.
the placentae of NR mothers demonstrate a marked 2001) and long-term changes in local glucocorticoid
up-regulation in 11bHSD2 mRNA that is likely to be action and UCP2 abundance within specific offspring
paralleled by an increase in 11bHSD2 enzyme activity tissues (Gnanalingham et al. 2005a, 2005c).
www.reproduction-online.org Reproduction (2007) 134 615–623
620 M G Gnanalingham and others
Materials and Methods required Home Office approval as designated by the Animals
(Scientific Procedures) Act (1986).
Animals and experimental design
Twenty singleton-bearing Welsh Mountain sheep of similar age
Laboratory analyses
(median 3 years) and weight (36.1G0.9 kg (meanGS.E.M.))
were employed for the studies and individually housed at mRNA detection
28 days gestation as described previously (Bispham et al. 2003,
Total RNA was isolated from fetal placentome tissue using Tri-
Gnanalingham et al. 2005d). Animals were allocated to one of
Reagent (Sigma) and the expression of UCP2, GR (type 2),
two nutritional groups using stratified randomization with
11bHSD2, and IGF-IR mRNA determined by reverse tran-
respect to body weight. The sheep consumed either 60% (i.e.
scriptase-PCR (RT-PCR) as described previously (Bispham et al.
nutrient restricted, NR (nZ10)) or 150% (controls (nZ10) of
2003, Gnanalingham et al. 2005c). The analysis used
their calculated metabolizable energy (ME) requirements for oligonucleotide cDNA primers to UCP2, GR (type 2),
both the maternal maintenance and the growth of the 11bHSD2, and IGF-IR genes as published previously (Bispham
conceptus on the basis of producing a 4.5 kg newborn at et al. 2003, Gnanalingham et al. 2005c), whilst VDAC
term (Agricultural Research Council 1980). Feed intakes were expression was determined using the primers of Cesar &
measured daily and animals weighed once every 2 weeks. NR Wilson (2004), all generating specific intron-spanning
animals consumed all of the feed offered, whereas those fed to products. Standard curves for each gene were initially
appetite consumed 150% of ME requirements as not all of the established in order to optimize the amount of cDNA required
hay was eaten. Food consumption between 28 and 80 days for each subsequent analysis. Agarose gel electrophoresis and
gestation was 3.2–3.8 MJ/day of ME in the NR group (w60% of ethidium bromide staining confirmed the presence of both the
ME requirements) and 8.7–9.9 MJ/day of ME in the control product and the ribosomal 18S, and densitometric analysis was
group, which was fed to appetite (and consumed w 150% of performed using a Fujifilm LAS-1000 cooled charge-coupled
ME requirements). The amount of feed given to each mother device (CCD) camera. Consistency of lane loading for each
was increased at by w5–10% at 43 and 61 days gestation to sample was verified and all results expressed as a ratio of a
meet the higher energy requirements associated with growth of reference sample to r18S abundance. All analyses and gels
the conceptus (Agricultural Research Council 1980). The diet were conducted in duplicate.
comprised chopped hay (ME content: 7.91 MJ/kg dry matter,
crude protein content (nitrogen!6.25): 69 g/kg dry matter) and Protein detection
barley-based concentrate (ME content: 11.6 MJ/kg dry matter,
crude protein content: 162 g/kg dry matter). The proportion of Mitochondria and plasma membranes were prepared from w1 g
of fetal placentome (Budge et al. 2000) and protein content
hay to concentrate fed was w3:1 dry weight. All diets
determined by the Lowry method (Lowry et al. 1951). Western
contained adequate minerals and vitamins. These were added
blotting was utilized to measure the abundance of VDAC and
separately to the diet with equal amounts provided to all sheep
cytochrome c mitochondrial proteins (Mostyn et al. 2003).
and thus were sufficient to fully meet their requirements. After
Identical amounts of placental protein were loaded (i.e. 10 mg)
80 days gestation, all animals were offered sufficient feed to
onto each gel for each sample. Following electroblotting of the
meet 100% of the ME requirements. These animals consumed
polyacrylamide gel onto a nitrocellulose membrane, Ponceau red
between 6.5 and 7.5 MJ/day of ME and the amount of feed
staining was used to visually confirm that similar amounts of
provided was increased by w10% at 100 and 120 days
protein had been transferred before subjecting the membranes to
gestation to meet the increased ME requirements that immunodetection (Mostyn et al. 2003). Abundance of cyto-
accompany the increase in fetal weight with gestation. chrome c was determined using a specific antibody (sc-7159;
In order to determine the effect of early to mid-gestational Santa Cruz, Biotechnology Inc., Santa Cruz, USA) at a dilution of
maternal NR on fetal placentome development, five sheep 1 in 1000. VDAC abundance was determined using an antibody
within each nutrition group were randomized to tissue raised in rabbits to ovine VDAC1, purified from the kidney of a
sampling at either 80 or 140 days gestation. Each animal was newborn sheep (Mostyn et al. 2003) at a dilution of 1 in 2000.
humanely euthanized following i.v. administration of Densitometric analysis was performed using AIDA software (Aida
200 mg/kg pentobarbital sodium (Euthatal: RMB Animal version 2.0; raytest Isotopenmeßgerate GmBH Straubenhardt,
¨
Health, Dagenham, UK). The entire uterus was removed from Germany) on each membrane following image detection using a
each animal and a number of randomly chosen A type Fujifilm LAS-1000 cooled CCD camera (Fuji Photo Film Co. Ltd,
placentomes as determined by their visual appearance (Vatnick Tokyo, Japan). All values were expressed in densitometric units.
et al. 1991) were sampled. These represent the majority of Specificity of detection was confirmed using nonimmune rabbit
placentomes in the sheep (Clarke et al. 1998, Heasman et al. serum. All gels were run in duplicate and a reference sample
1998) and were immediately dissected and either put into 10% (placental mitochondria from a control sheep sampled at
(v/v) formalin and embedded in paraffin wax for subsequent 140 days gestation) was included on each to allow comparison
histological analysis, or separated into maternal and fetal between gels.
components and immediately placed in liquid nitrogen and In addition, further Western blots were performed to confirm
stored at K80 8C for later analysis. In addition, total placental the effects of maternal nutrition on GR and IGF-IR protein
and fetal weights were recorded together with all major organs. abundance. This was undertaken using polyclonal antibodies
All operative procedures and experimental protocols had the for each protein purchased from Santa Cruz (catalogue
Reproduction (2007) 134 615–623 www.reproduction-online.org
Nutritional manipulation in the placenta 621
numbers SC 8992 and 713 respectively). Each antibody was were performed using the same assessor who was blinded to
tested at a range of dilutions from 1:150 to 1:1000 under both nutritional grouping using reference slides to check consistency
mild reducing and nonreducing conditions using up to 80 mg of grading assessment. To assess the relative incidence of
protein. Unfortunately, neither antibody yielded a concise apoptosis between groups, slides were analyzed using
signal (in any of the animals) that was in accord with its fluorescence microscopy with FITC and u.v. filters to visualize
predicted molecular mass (i.e. signals were nonspecific). green (fragmented DNA) and a semi-quantitative apoptosis
Nonspecificity of detected bands was confirmed through index (0–4; Lepault et al. 2005) used. This procedure was
regression analysis of molecular weight markers to determine undertaken on 30 randomly selected sections from each
exact size and through incubation with nonimmune rabbit nutritional group at a magnification of 200!. In addition, as
serum. All Western blots were run in duplicate and included a the TUNEL reaction is unable to discriminate apoptotic from
range of molecular weight markers and a positive reference necrotic cells, caspase-3, a marker of early apoptosis (Gown &
sample (plasma membranes isolated from 1-day-old sheep). Willingham 2002) was localized using a rabbit anti-caspase 3
polyclonal antibody (Abcam plc, Cambridge, UK) diluted 1:50.
Immunohistochemistry and assessment of apoptosis Immunohistochemistry was performed using the Bond auto-
To establish the cellular location of VDAC and cytochrome c mated system with a bond polymer refine detection kit (Vision
within the placenta, immunohistochemistry was performed Biosystems Limited). A negative control was performed for
using the antibodies described above at dilutions ranging from each test section by omitting incubation in the primary
1:100 to 1:1000 (Dandrea et al. 2001). Following incubation antibody. Tonsil, in which apoptosis is pronounced, was used
with enzyme-conjugated second antibody and chromogen as the positive control (Krajewska et al. 1997). For each section,
substrate, sections were examined by light microscopy. The the number of positive cells in five randomly selected fields was
specificity of the procedure was confirmed by the absence of used to calculate the mean positive cell count G S.E.M per
binding when adjacent sections were incubated with rabbit 200! field.
serum from an unimmunized rabbit in place of rabbit anti-
ovine primary antibody. It was not possible to perform the same
Statistical analysis
analyses with respect to the location of UCP2 due to the lack of
appropriately validated antibodies (Gnanalingham et al. All data are presented as the meansGS.E.M. Significant
2005f). The same analyses were undertaken for the GR and differences (P!0.05) between gestational ages and nutritional
IGF-IR, but in accord with our failure to detect protein at the groups were analyzed by ANOVA. Significant correlations
correct molecular weight by Western blotting neither antibody between molecular parameters were assessed by Spearman’s
showed specificity of binding to the placenta. This procedure Rank Order Test (SPSS v11.0; SPSS Inc., Chicago, Illinois, USA).
was undertaken on 20–30 randomly selected sections from
each nutritional group.
Measurement of cell proliferation within the placenta was
also undertaken by determining the amount of immunoreactive Acknowledgements
proliferating cell nuclear antigen (PCNA) expression (Waseem
This work was funded by the Special Trustees of Nottingham
& Lane 1990). This was undertaken using a mouse MAB to
University Hospitals, the British Heart Foundation Lectureship,
PCNA (PC10) (Alexis Biochemicals, Bingham, Nottingham) at
and the European Union Sixth Framework Programme for
a 1:200 dilution. Relative staining intensity for immunoreactive
Research and Technical Development of the European
nuclear PCNA expression was visually assessed in a blinded Community – The Early Nutrition Programming Project
manner and scored as either absent (K; no nuclear staining), (FOOD-CT-2005-007036). The authors declare that there is
weak (C), moderate (CC), or strong (CCC) (Taylor et al. no conflict of interest that would prejudice the impartiality of
2000). Immunohistochemistry was performed on a computer- this scientific work.
ized Bond automated immunohistochemistry system (Vision
Biosystems Limited, Newcastle-upon-Tyne, UK) using a bond
polymer refine detection kit. A negative control was performed
for each test section by omitting incubation in the primary
antibody. Tonsil, in which PCNA staining is pronounced, was References
used as the positive control (Waseem & Lane 1990). Sections Agricultural Research Council 1980 Requirements for energy, The
were dehydrated in ascending concentrations of alcohol and Nutritional Requirements of Ruminant Livestock, Slough, UK: Common-
xylene before coverslips were mounted. wealth Agricultural Bureau.
Bamberger CM, Schulte HM & Chrousos GP 1996 Molecular determinants
These sections were also used to assess the relative incidence
of glucocorticoid receptor function and tissue sensitivity to glucocorti-
of apoptosis by the TUNEL assay (Roche Diagnostics; Gavrieli coids. Endocrinology Reviews 17 245–261.
et al. 1992). Sections were stored at 4 8C in the dark and the Barker DJP 2001 The malnourished baby and infant. British Medical
magnitude of staining analyzed within 48 h. Appropriate Bulletin 60 69–88.
negative controls were included to assess both nonspecific Bispham J, Gopalakrishnan GS, Dandrea J, Wilson V, Budge H, Keisler DH,
Broughton Pipkin F, Stephenson T & Symonds ME 2003 Maternal
binding and the extent of autofluorescence. Positive controls
endocrine adaptation throughout pregnancy to nutritional manipulation:
were treated with DNase type I to produce fragmentation of consequences for maternal plasma leptin and cortisol and the
chromosomal DNA and washed separately to prevent any programming of fetal adipose tissue development. Endocrinology 144
residual DNase activity affecting other sections. All analyses 3575–3585.
www.reproduction-online.org Reproduction (2007) 134 615–623
622 M G Gnanalingham and others
Budge H, Bispham J, Dandrea J, Evans E, Heasman L, Ingleton PM, Heasman L, Brameld JM, Mostyn A, Budge H, Dawson J, Buttery PJ,
Sullivan C, Wilson V, Stephenson T & Symonds ME 2000 Effect of Stephenson T & Symonds ME 2000 Maternal nutrient restriction during
maternal nutrition on brown adipose tissue and its prolactin receptor early to mid gestation alters the relationship between IGF-I and body size
status in the fetal lamb. Pediatric Research 47 781–786. at term in fetal sheep. Reproduction, Fertility and Development 12
Carter AM & Mess A 2007 Evolution of the placenta in Eutherian mammals. 345–350.
Placenta 28 295–262. Huppertz B & Herrler A 2005 Regulation of proliferation and apoptosis
Cesar MC & Wilson JE 2004 All three isoforms of the voltage-dependent during development of the preimplantation embryo and the placenta.
anion channel (VDAC1, VDAC2, and VDAC3) are present in mito- Birth Defects Research. Part C, Embryo Today: Reviews 75 249–261.
chondria from bovine, rabbit, and rat brain. Archives of Biochemistry and Krajewska M, Wang HG, Krajewski S, Zapata JM, Shabaik A, Gascoyne R &
Biophysics 422 191–196. Reed JC 1997 Immunohistochemical analysis of in vivo patterns of
Clarke L, Heasman L, Juniper DT & Symonds ME 1998 Maternal nutrition in expression of CPP32 (Caspase-3), a cell death protease. Cancer Research
early-mid gestation and placental size in sheep. British Journal of 57 1605–1613.
Nutrition 79 359–364. Langley-Evans SC, Phillips GJ, Benediktsson R, Gardner DS, Edwards CR,
Colombini M 1979 A candidate for the permeability pathway of the outer Jackson AA & Seckl JR 1996 Protein intake in pregnancy, placental
mitochondrial membrane. Nature 279 643–645. glucocorticoid metabolism and the programming of hypertension in the rat.
Crompton M 1999 The mitochondrial permeability transition pore and its Placenta 17 169–172.
role in cell death. Biochemical Journal 341 233–249. Laviola L, Perrini S, Belsanti G, Natalicchio A, Montrone C, Leonardini A,
Dandrea J, Wilson V, Gopalakrishnan G, Heasman L, Budge H, Vimercati A, Scioscia M, Selvaggi L, Giorgino R, et al. 2005 Intrauterine
Stephenson T & Symonds ME 2001 Maternal nutritional manipulation growth restriction in humans is associated with abnormalities in
of placental growth and glucose transporter-1 abundance in sheep. placental insulin-like growth factor signaling. Endocrinology 146
Reproduction 122 793–800. 1498–1505.
Demasi MA, Montor WR, Ferreira GB, Pimenta DC, Labriola L & Lepault E, Celeste C, Dore M, Martineau D & Theoret CL 2005
Sogayar MC 2007 Differential proteomic analysis of the anti-proliferative Comparative study on microvascular occlusion and apoptosis in body
effect of glucocorticoid hormones in ST1 rat glioma cells. Journal of and limb wounds in the horse. Wound Repair and Regeneration 13
Steroid Biochemistry Molecular Biology 103 137–148. 520–529.
Edwards LJ & McMillen IC 2001 Maternal undernutrition increases arterial Lowry OH, Rosenbrough NJ, Farr AL & Randall RJ 1951 Protein
blood pressure in the sheep fetus during late gestation. Journal of measurement with the Folin phenol reagent. Journal of Biological
Physiology 533 561–570. Chemistry 193 265–275.
Ehrhardt RA & Bell AW 1995 Growth and metabolism of the ovine placenta McMillen IC & Robinson JS 2005 Developmental origins of the metabolic
during mid-gestation. Placenta 16 727–741. syndrome: prediction, plasticity, and programming. Physiological
Reviews 85 571–633.
Gavrieli Y, Sherman Y & Ben-Sasson SA 1992 Identification of programmed
Mostyn A, Wilson V, Dandrea J, Yakubu DP, Budge H, Alves-Guerra MC,
cell death in situ via specific labeling of nuclear DNA fragmentation.
Pecqueur C, Miroux B, Symonds ME & Stephenson T 2003 Ontogeny
Journal of Cell Biology 119 493–501.
and nutritional manipulation of mitochondrial protein abundance in
Gnanalingham MG, Mostyn A, Symonds ME & Stephenson T 2005a
adipose tissue and the lungs of postnatal sheep. British Journal of
Ontogeny and nutritional programming of adiposity: potential role of
Nutrition 90 323–328.
glucocorticoid sensitivity and uncoupling protein-2. American Journal of
´
Pecquer C, Alves-Guerra M-C, Gelly C, Levi-Meyrueis C, Couplan E,
Physiology 289 R1407–R1415.
Collins S, Ricquier D, Bouillaud F & Miroux B 2001 Uncoupling
Gnanalingham MG, Mostyn A, Gardner DS, Stephenson T & Symonds ME
protein-2: in vivo distribution, induction upon oxidative stress and
2005b Developmental regulation of adipose tissue: nutritional manipu-
evidence for translational regulation. Journal of Biological Chemistry 276
lation of local glucocorticoid action and uncoupling protein 2.
8705–8712.
Adipocytes 1 221–228.
Reid GJ, Flozak AS & Simmons RA 2002 Placental expression of
Gnanalingham MG, Mostyn A, Dandrea J, Yakubu DP, Symonds ME &
insulin-like growth factor receptor-1 and insulin receptor in the
Stephenson T 2005c Ontogeny and nutritional programming of
growth-restricted fetal rat. Journal of the Society for Gynecologic
uncoupling protein-2 and glucocorticoid receptor mRNA in the ovine Investigations 9 210–214.
lung. Journal of Physiology 565 159–169. Rogatsky I, Trowbridge JM & Garabedian MJ 1997 Glucocorticoid
Gnanalingham MG, Mostyn A, Forehead AJ, Fowden AL, Symonds ME & receptor-mediated cell cycle arrest is achieved through distinct cell-
Stephenson T 2005d Increased uncoupling protein-2 mRNA abundance specific transcriptional regulatory mechanisms. Molecular Cell Biology
and glucocorticoid sensitivity in adipose tissue in the sheep fetus during 17 3181–3193.
late gestation is dependent on plasma cortisol and triiodothyronine. Rousset S, Alves-Guerra M-C, Ouadghiri-Bencherif S, Kozak LP, Miroux B,
Journal of Physiology 567 283–292. Richard D, Bouillaud F, Ricquier D & Cassard-Doulcier A-M 2003 UCP2 is
Gnanalingham MG, Giussani DA, Sivathondan P, Forehead AJ, expressed in the female mice reproductive tract whereas UCP1 is not.
Stephenson T, Symonds ME & Gardner DS 2005e Chronic umbilical Journal of Biological Chemistry 278 45843–45847.
cord compression results in premature maturation of lung and brown Stegmann JHJ 1974 Placental development in sheep. Bijdragen tot de
adipose tissue in the late gestation ovine fetus. American Journal of Dierkunde 44 4–72.
Physiology 289 E456–E465. Stewart PM & Krozowski ZS 1999 11b-Hydroxysteroid dehydrogenase.
Gnanalingham MG, Mostyn A, Webb R, Keisler DH, Raver N, Alves- Vitamins and Hormones 57 249–324.
Guerra MC, Pecqueur C, Miroux B, Symonds ME & Stephenson T 2005f Stuart JA, Harper JA, Brindle KM, Jekabsons MB & Brand MD 2001
Differential effects of leptin administration on the abundance of Physiological levels of mammalian uncoupling protein 2 do not
uncoupling protein-2 and glucocorticoid action during neonatal uncouple yeast mitochondria. Journal of Biological Chemistry 276
development. American Journal of Physiology 289 E1093–E1100. 18633–18639.
Gnanalingham MG, Mostyn A, Gardner DS, Stephenson T & Symonds ME Symonds ME, Budge H, Stephenson T & Gardner DS 2005 Leptin,
2006 Developmental regulation of the lung in preparation for life after birth: fetal nutrition and long term outcomes for adult hypertension.
nutritional manipulation of local glucocorticoid action and uncoupling Endothelium 12 73–79.
protein 2. Journal of Endocrinology 188 375–386. Symonds ME, Budge H, Mostyn A, Stephenson T & Gardner DS 2007
Gown AM & Willingham MC 2002 Improved detection of apoptotic cells in Maternal diet through pregnancy – the key to future good health of the
archival paraffin sections: immunohistochemistry using antibodies to next generation. In Reducing Diets: New research, Ed. SV Watkins. New
cleaved caspase 3. Journal of Histochemistry and Cytochemistry 50 York: Nova Science Publishers, Inc.
449–454. Taylor KM, Gray CA, Joyce MM, Stewart MD, Bazer FW & Spencer TE
Heasman L, Clarke L, Firth K, Stephenson T & Symonds ME 1998 Influence 2000 Neonatal ovine uterine development involves alterations in
of restricted maternal nutrition in early to mid gestation on placental expression of receptors for estrogen, progesterone, and prolactin.
and fetal development at term in sheep. Pediatric Research 44 546–551. Biology of Reproduction 63 1192–1204.
Reproduction (2007) 134 615–623 www.reproduction-online.org
Nutritional manipulation in the placenta 623
Vatnick I, Schoknecht PA, Darrigrand R & Bell AW 1991 Growth and Whorwood CB, Firth KM, Budge H & Symonds ME 2001 Maternal
metabolism of the placenta after unilateral fetectomy in twin pregnant undernutrition during early- to mid-gestation programmes tissue-specific
ewes. Journal of Developmental Physiology 15 351–356. alterations in the expression of the glucocorticoid receptor, 11b-
Voehringer DW, Hirschberg DL, Xiao J, Lu Q, Roederer M, Lock CB, hydroxysteroid dehydrogenase isoforms and type 1 angiotensin II
Herzenberg LA, Steinman L & Herzenberg LA 2000 Gene microarray receptor in neonatal sheep. Endocrinology 142 2854–2864.
identification of redox and mitochondrial elements that control
resistance or sensitivity to apoptosis. PNAS 97 2680–2685.
Waddell BJ, Hisheh S, Dharmarajan AM & Burton PJ 2000 Apoptosis in rat
placenta is zone-dependent and stimulated by glucocorticoids. Biology
Received 14 December 2006
of Reproduction 63 1913–1917.
Waseem NH & Lane DP 1990 Monoclonal antibody analysis of the First decision 26 January 2007
proliferating cell nuclear antigen (PCNA). Structural conservation and the Revised manuscript received 26 June 2007
detection of a nucleolar form. Journal of Cell Science 96 121–129. Accepted 13 July 2007
www.reproduction-online.org Reproduction (2007) 134 615–623