| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Rowett Research Institute,3 Bucksburn, Aberdeen AB21 9SB, United Kingdom
Department of Animal and Range Sciences,4 North Dakota State University, Fargo, North Dakota 58105
Hirosaki University,5 Hirosaki 036-8561, Japan
ABSTRACT
Limiting maternal nutrient intake during ovine adolescent pregnancy progressively depleted maternal body reserves, impaired fetal nutrient supply, and slowed fetal soft tissue growth. The present study examined placental growth, angiogenic gene expression, and vascular development in this undernourished adolescent model at Days 90 and 130 of gestation. Singleton pregnancies were established, and ewes were offered an optimal control (C; n = 14) or low (L [0.7 x C]; n = 21) dietary intake. Seven ewes receiving L intakes were switched to C intakes on Day 90 of gestation (L-C). Fetal body weight (P < 0.01) and glucose concentrations (P < 0.03) were reduced in L versus C pregnancies by Day 130, whereas L-C group values were intermediate. Placental cellular proliferation, gross morphology, and mass were independent of maternal nutrition at both Day 90 and 130. In contrast, capillary area density in the maternal caruncular portion of the placentome was reduced by 20% (P < 0.001) at both stages of gestation in L compared with C groups. Caruncular capillary area density was equivalent in the L and L-C groups at Day 130. Placental mRNA expression of five key angiogenic ligands or receptors increased (P < 0.001) between Days 90 and 130 of gestation. VEGFA mRNA expression was higher (P < 0.04) in L compared with C and L-C pregnancies at Day 130, but otherwise gene expression of the remaining angiogenic factors and receptors analyzed was unaffected by maternal intake. Undernourishing the pregnant adolescent dam restricts fetal growth independently of changes in placental mass. Alterations in maternal placental vascular development may, however, play a role in mediating the previously reported reduction in maternal and hence fetal nutrient supply.
adolescent pregnancy,, angiogenesis,, female reproductive tract,, fetus,, placenta,, placental vascular development,, pregnancy,, progesterone,, undernutrition
Becoming pregnant during adolescent life is a major risk factor for premature delivery, low birth weight, neonatal mortality, and maternal death [1]. The risk of these adverse pregnancy outcomes is considered to be particularly acute in young girls who have not achieved their adult height at the time of conception, and hence still have the potential to grow while pregnant [2–4]. Suboptimal dietary intakes are commonplace in sections of the general adolescent population [5] and, consequently, many adolescents may be in danger of becoming pregnant with poor nutrient stores [6] and/or subsequently experiencing inadequate gestational weight gains. Indeed, several studies have reported an association between low or insufficient pregnancy weight gains in human adolescent mothers (a proxy indicator of maternal undernutrition) and the incidence of low birth weight [7–11]. The putative role of the placenta in mediating these effects has not been studied.
We recently replicated part of this human problem by relatively undernourishing singleton-bearing young adolescent sheep to maintain maternal body weight at the level determined at conception. This effectively prevented further maternal growth and progressively depleted maternal body reserves relative to control-fed sheep that were nourished to maintain initial adiposity throughout gestation. By late pregnancy, maternal and fetal glucose concentrations were suppressed and fetal weight was reduced due to a slowing of soft tissue growth in the nutrient-restricted pregnancies [12]. This companion paper examines the role of the placenta in mediating these effects in undernourished compared with control intake pregnancies. Thus, we examined placental proliferation, angiogenic gene expression, and vascular development at two key stages of gestation: Day 90, which equates to the apex of placental growth [13] and exponential vascular development of the fetal cotyledon [14, 15], and Day 130, which is coincident with maximum placental vascularization, blood flow, and nutrient transfer function [14–16]. Placental parameters also were determined in a further group of undernourished adolescents whose nutrient intakes were switched to match those of control pregnancies from Days 90–130 of gestation.
Animals and Experimental Design
All procedures were licensed under the United Kingdom Animals (Scientific Procedures) Act of 1986 and approved by the Rowett Research Institute's Ethics Review Committee. Embryos from superovulated adult ewes (Border Leicester x Scottish Blackface) inseminated by a single sire were recovered on Day 4 after estrus and were transferred synchronously in singleton into the uterus of recipient ewe lambs (Dorset Horn x Greyface), exactly as described previously [17]. Details of the age, weight, and adiposity of the donor and recipient animals used in this study together with full details of the experimental design and composition of the diet used are presented in detail in the companion paper [12]. Briefly, adolescent recipients of equivalent age, weight, and adiposity were individually offered an optimal control (C; n = 18) or low (L; n = 28) quantity of the same complete diet immediately following embryo transfer. The dietary level in the control group was calculated to maintain normal maternal adiposity throughout gestation (achieved by promoting a live weight gain of 
50 g/day) and to provide 100% of the estimated metabolizable energy and protein requirement of the adolescent sheep carrying a singleton fetus according to stage of pregnancy [18]. In contrast, the dietary intake in the low group was calculated to maintain maternal live weight at embryo transfer, and hence gradually deplete maternal tissue reserves throughout gestation. At Day 90 of gestation, a representative group of seven low-intake ewes were switched to the control intake (L-C). The level of feed offered was reviewed three times weekly and adjusted on an individual basis on the basis of body weight change data (determined weekly) as and when appropriate. Maternal body condition was subjectively assessed on a five-point scale (1 = emaciated, 5 = obese) every 2 wk [19]. Pregnancy status was determined by transabdominal ultrasonography at approximately Day 45 of gestation (gestation length = 145 days) and revealed viable fetuses in 14 and 21 ewes in the C and L groups, respectively.
Blood Sample Collection and Necropsy Procedures
Weekly blood samples were collected from Day 7 of gestation onward by jugular venepuncture 3 h after the morning feed and, in addition to the endocrine and metabolic data previously presented in the companion paper [12], were analyzed for progesterone concentrations measured directly in duplicate by a double-antibody radioimmunoassay [20]. The intraassay and interassay coefficients of variation for progesterone were 8.7% and 8.8%, respectively.
Prior to necropsy on either Day 90 or Day 130 of gestation and to determine placental cellular proliferation (labeling index), dams were infused with bromodeoxyuridine (BrDU, 5 mg/kg body weight, dissolved in PBS to form a saturated solution of 16.7 mg BrdU/ml, pH 7.0), via a catheter inserted in the jugular vein exactly 1 hour prior to lethal injection [21]. Killing was achieved by intravenous administration of an overdose of sodium pentobarbitone (20 ml Euthesate; 200 mg pentobarbitone/ml; Willows Francis Veterinary, Crawley, UK) and exsanguination (by severing the main vessels of the neck). The gravid uterus was removed, weighed, and opened, and a fetal blood sample was collected by cardiac puncture immediately before intracardiac administration of sodium pentobarbitone (3 ml Euthesate). Fetal plasma glucose concentrations were determined in duplicate using a Yellow Springs Instruments dual biochemistry analyzer (model 2700; Yellow Springs, OH) as previously described [22]. After clamping the cord, the fetus was removed, dried, weighed, and dissected into its component organs. Eight representative placentomes from the gravid horn were immediately removed, their gross morphology classified per Vatnick et al. [23], and weighed. Four of these placentomes were snap frozen in liquid nitrogen-cooled isopentane and stored at –80°C for subsequent gene expression studies. The other four placentomes were sliced into 5-mm cross-sections and immersion fixed in Carnoy's fixative for 6 h followed by 70% ethanol, changed once after 24 h (for BrDU immunohistochemistry).
Perfusion and Quantification of Placental Vascularity
Maternal caruncle and fetal cotyledon tissues of the placenta were perfused after catheterizing branches of the uterine and umbilical arteries, respectively, as previously described [24]. Briefly, a catheter was inserted into the artery, secured with suture, and perfused in the following sequence: 1) 10 ml PBS; 2) 5 ml of 1% Evans Blue; 3) 10 ml PBS; 4) 10 ml Carnoy's solution; 5) 10 ml PBS; 6) 10.8 ml of a plastic resin casting mixture (Mercox; Ladd Research Industries Inc., Burlington, VT). Perfused placentomes were then removed from the uterus, weighed, sliced as above, and immersion fixed in Carnoy's solution. All remaining placentomes were dissected and weighed and their gross morphology recorded. Mercox-perfused placentome tissues (one sample each from maternal caruncle and fetal cotyledon perfusions) were embedded in paraffin wax blocks, sectioned (5 µm), and stained with periodic acid and Schiff's reagent as previously described [25].
Thereafter, placentome vascularity was quantified according to Borowicz et al. [24]. Briefly, placentome sections were visualized with a Leica DMR microscope (Leica Microsystems, Welzlar, Germany) at 20x magnification, and 10 digital images per section were captured with a Hamamatsu camera (C5810; Leica Microsystems). The images were analyzed using Image-Pro Plus version 4.5.1 software (Media Cybernetics Inc., Silver Spring, MD). Total caruncle area and shrinkage area was determined, and each caruncle or cotyledon vessel was individually traced to quantify the total vessel number, perimeter, and area. Based on these measurements, total cotyledon area [total image area – (caruncle area + shrinkage area)] was determined. In addition, for each perfused image of cotyledon and caruncle, vessel density (%; vessel area:tissue area), vessel number density (vessel number:tissue area), vessel surface density (vessel perimeter:tissue area), and area per vessel (total vessel area:vessel number) were determined. Lastly, for each image the caruncle-cotyledon area (%; caruncle area:cotyledon area) was calculated.
BrDU was immunolocalized and quantified in placental tissue that had been immersion fixed in Carnoy's solution and embedded in paraffin wax blocks exactly as described previously [26]. Sections were incubated with a monoclonal antibody to BrDU (100 µg/ml; Roche Diagnostics Ltd., East Essex, UK) or normal mouse immunoglobulin G (2 µg/ml, negative control). BrDU staining was quantified using the image analysis system described above in approximately 20 fields of view transversely across the maternal-fetal interface at the central part of the placentome section. Positively stained proliferating nuclei were measured and expressed as a percentage of the total nuclear area stained with hematoxylin.
Messenger RNA levels for key angiogenic growth factors (vascular endothelial growth factor A [VEGFA], fibroblast growth factor 2 [FGF2], angiopoietin 1 [ANGPT1]) and their receptors (kinase insert domain receptor [KDR, a VEGF receptor], fms-related tyrosine kinase 1 [FLT1, a VEGF receptor]) were determined using quantitative real-time RT-PCR. Probes and primer sequences and all methodology were exactly as previously described [27]. Reference-standard cDNA was generated from RNA pooled from Day 90 and Day 130 placentomes. Individual placentome mRNA for each gene of interest was expressed relative to the sample's internal 18S RNA.
Data are presented as means ± SEM. Differences between treatments for data collected weekly or at Day 90 or Day 130 of gestation were analyzed by a one-way ANOVA (General Linear Models, Minitab 13; Minitab Inc., State College, PA), and then the least significant difference was determined between individual treatment means. A similar approach was used to analyze weekly maternal progesterone concentrations after calculating a mean individual value spanning the first (from Day 4 to Day 49), second (from Day 50 to Day 90), and final third (Day 91 to Day 130) of gestation. Weekly maternal progesterone concentrations are presented with all animals in each necropsy group up to Day 90 of gestation, as there were no significant differences in any of the parameters measured between pregnancies terminated at Day 90 versus Day 130 within groups up to this point. Maternal and fetal necropsy, placental vascularity, and angiogenic growth factor expression data collected at Day 90 or Day 130 of gestation from the C versus L groups were analyzed by two-way ANOVA to determine the main effects of nutrition and stage of gestation and the interaction between these factors. Differences of P < 0.05 were considered significant.
The dietary-induced changes in maternal live weight, adiposity, and body composition in relation to fetal growth, organ allometry, and body composition have been presented in detail in the companion paper [12]. Key parameters essential only to the interpretation of the current study have been reproduced in Table 1, together with placental parameters.
|
As per the experimental design, maternal body weight (that is, live weight at necropsy minus gravid uterus weight) was maintained throughout pregnancy in the L group, whereas the control intake facilitated a modest maternal body weight gain of approximately 50 g/day (Table 1). Consequently, the external adiposity score progressively declined in the nutrient-restricted dams as pregnancy developed, whereas adiposity was maintained at the level observed at conception in the control group. On average, the difference in adiposity score between C and L groups was 0.4 units at Day 90 (P < 0.01) and 0.8 units at Day 130 (P < 0.001). Maternal adiposity in the L-C group during late gestation was intermediate and greater (P < 0.001) than in the L group but lower (P < 0.005) than in the C group.
Fetoplacental Growth and Vascular Morphology
At Day 90 of gestation, total placentome weight, mean placentome weight, placentome number, gross morphology, and cellular proliferation rate were not influenced by maternal diet, and the fetus was unperturbed, as indicated by similarities in fetal weight and glucose concentrations between C and L groups (Table 1). By Day 130 of gestation, maternal nutrient restriction (L group) led to a modest yet highly significant reduction (P < 0.01) in fetal weight compared with the C group. The smaller fetuses had lower plasma glucose concentrations (P < 0.03) and a greater brain:liver weight ratio (P < 0.005), but these fetal effects were unrelated to placental mass, placentome number, gross morphology, cellular proliferation rate, and the fetal:placental weight ratio, all of which were equivalent between L and C groups. Fetal weight in the L-C group was intermediate and similarly unrelated to the placental parameters presented in Table 1. As expected, total placentome weight (P < 0.002) and placental cellular proliferation rate (P < 0.001) decreased between Days 90 and 130 of gestation, and was accompanied by a change in gross placental morphology (P < 0.001) from predominately A-B types to a mixed A-D population. Cotyledonary capillary changes in relation to maternal dietary intake and stage of gestation are detailed in Table 2. Capillary surface density was greater (P < 0.05) at Day 90 in L compared with C groups, but otherwise capillary morphology was unaffected by maternal diet. In contrast, highly significant increases (P < 0.001) in cotyledon vessel area and surface and number densities were observed as gestation advanced.
|
Capillary changes in the maternal caruncular portion of the placentome in relation to maternal dietary intake and stage of gestation are similarly detailed in Table 2. At both Days 90 and 130 of gestation, an 18%–20% reduction (P < 0.001) in maternal caruncle capillary area density was observed in the L versus the C group (Fig. 1). Relative to the C group, a similar reduction (18%; P < 0.002) in capillary area density was observed in the L-C group at Day 130 of gestation. At Day 90, but not Day 130, of gestation, a 12% decrease (P < 0.04) in capillary surface density was observed in the maternal caruncle of the nutrient-restricted dams when compared to the control dams. However, maternal dietary intake did not influence caruncle capillary number density or mean capillary area at either stage of gestation. Comparison of vascular parameters at Day 90 compared with Day 130 revealed that capillary area density (P < 0.001) and mean capillary area (P < 0.001) increased, whereas capillary number density decreased (P < 0.02), with advancing gestation.
|
Placental Angiogenic Growth Factor and Receptor Expression
Angiogenic factor mRNA expression was determined in the whole placentome at Day 90 and Day 130 (Fig. 2). A 2- to 4-fold increase in mRNA expression (P < 0.001) was observed for all of the angiogenic growth factors and receptors (GOI: 18S) between Days 90 and 130 of gestation. Relative to the C and L-C groups, VEGFA mRNA expression (ratio to 18S) was greater (P < 0.04) in the L group at Day 130 of gestation. The two VEGF receptors, namely, FLT1 and KDR, and the other angiogenic growth factors, ANGPT1 and FGF2, were unaltered by maternal diet at either necropsy stage.
|
Maternal Progesterone Concentrations
Maternal progesterone concentrations throughout gestation are illustrated in Figure 3. Progesterone concentrations were independent of maternal dietary intake during the first third of gestation and were higher (P < 0.05) in L versus C groups during the second third of gestation. Despite similarities in placental mass at Days 90 and 130 of gestation, maternal progesterone concentrations during the last third of gestation (Days 91–130) were inversely related to maternal intake and thus were greater in the L versus C (P < 0.01) and L-C groups (P < 0.03).
|
We have recently reported that limiting maternal intake during adolescent pregnancy progressively depleted maternal body reserves, impaired fetal nutrient supply, and slowed fetal soft tissue growth [12]. As the resulting reduction in fetal weight in late pregnancy was independent of changes in placental mass per se, we suggested that fetal growth restriction was directly due to reduced nutrient availability in the maternal circulation. However, the data presented herein further suggest that alterations in maternal placental vascular development may play a role in mediating the reduction in nutrient availability to the gravid uterus. In spite of the fact that placental cellular proliferation rate, mass, and gross morphology at necropsy were not influenced by maternal dietary intake, a highly significant reduction in maternal caruncle capillary area density was observed at both Day 90 and Day 130 of gestation in the nutrient-restricted pregnancies. On Day 130, this reduction in maternal caruncle capillary area density was, in fact, greater than the decrease in fetal weight (approximately 18% to 20% vs. 17% to 10% for L and L-C groups, respectively).
As cross-sectional capillary area determines to a large extent the "blood flow capacity" of a vascular bed [28], the aforementioned reductions in capillary area density of the caruncle imply that blood flow to the maternal placenta (i.e., uterine blood flow) may have been compromised in the underfed dams. It is well established that the degree of vascularization within the placenta determines the rate of exchange of nutrients, gases, and wastes between the mother and her fetus [15] and that blood flow to the uterus is a determinant of fetal growth during the last half of pregnancy in sheep [29]. Furthermore, in the human, increased uterine vascular resistance and reduced uterine blood flow are used as predictors of pregnancies at risk of intrauterine growth restriction [30–33] and may similarly be indicative of changes in placental vascular development [11, 34]. Furthermore, this novel finding supports the recent hypothesis that maternal undernutrition during pregnancy impairs placental angiogenesis and vascular development in humans [35].
Uterine blood flow has not yet been measured in these undernourished adolescent ewes but has been shown to decrease in nutrient-restricted adult ewes during midgestation [36] and late gestation [37]. Chandler et al. [37] imposed a severe maternal nutrient restriction (30%–40% of maintenance requirements) in singleton-bearing adult ewes for 21 days during late gestation and reported a subsequent 30% decrease in uterine blood flow. As in the present adolescent study, maternal nutrient restriction in these adult ewes was associated with maternal and fetal hypoglycemia and reduced fetal mass but no change in placental mass per se. Similarly, Kelly [36] severely restricted maternal intake in mature ewes from Days 28 to 108 of gestation and reported a significant decrease in uterine blood flow (assessed by Doppler arterial waveform analysis) at Day 100 of pregnancy. This observation was also associated with maternal hypoglycemia and decreased lamb birth weight. Although Kelly [36] did not determine placental mass per se, mean placentome diameter measured via ultrasound was reduced by 10% in nutrient-restricted ewes.
The mechanism whereby reduced maternal dietary intake compromised placental vascular development and, putatively, uterine blood flow is unknown but may in part be secondary to maternal anemia. The progressive induction of chronic maternal anemia (50% decrease in hematocrit) in adult sheep by repetitive exchange transfusions prevents the normal gestational increase in uterine blood flow and is associated with a 40% reduction in fetal weight in late pregnancy [38]. Although less severe, preliminary observations in undernourished adolescent dams indicate that low dietary intakes are associated with a 26% decrease (P < 0.001) in maternal hematocrit relative to optimally nourished controls at Day 130 of gestation (26% ± 0.5% versus 35% ± 0.4%, respectively, n = 16 per group; Wallace et al., unpublished data). Similarly, in human adolescents, anemia during midpregnancy and late pregnancy is associated with an increased risk of preterm delivery, low birth weight, and maternal and infant mortality [39–41].
The reduction in caruncle capillary area density in nutrient-restricted adolescent ewes realimented to control intakes between Days 90 and 130 of gestation was equivalent to that of the continuously nutrient-restricted adolescent ewes. This suggests that the developmental trajectory of the vasculature in the maternal caruncle was established in the first half of pregnancy, during the period of rapid cellular proliferation in the sheep placenta [13], and cannot be reversed by subsequent nutritional manipulation. Despite the persistent reduction in caruncle capillary area density in the L-C group, fetal growth was partially restored after realimentation, suggesting that nutrient availability in the maternal circulation is the predominant limitation to fetal growth in this undernourished paradigm.
Despite decreases in caruncle capillary area density in the nutrient-restricted group at Day 90 of gestation, the expression of ANGPT1, FGF2, and VEGFA and its receptors (FLT1 and KDR) in whole-placentome tissue was not affected by maternal diet. As changes in gene expression may be expected to precede alterations in vascular structure, the timing of the Day 90 necropsy may simply have been too late for nutritionally mediated changes in these selected angiogenic factors to be detected. A significant increase in whole-placentome VEGFA mRNA expression was detected at Day 130 of gestation, but only in continuously undernourished animals, suggesting that this occurred in response to the current nutritional intake and the associated changes in maternal hormone status and metabolism.
The expression of VEGFA is variously upregulated by hypoxia [42], hypoglycemia [43] and various steroids [44, 45]. Maternal progesterone concentrations during pregnancy in sheep are known to be inversely related to maternal intake [22, 46] and similarly were elevated during the final third of gestation in the continuously nutrient restricted dams in the present study. Greater circulating steroid concentrations during nutrient restriction are largely due to a decreased rate of metabolic clearance [47], which may have impacted placental VEGFA expression in this study. In addition, treatment of ovariectomized mice with progesterone stimulates endometrial angiogenesis and is in part mediated by increased VEGFA mRNA expression [45]. This suggests that the increased VEGFA mRNA expression in the whole placentome of undernourished adolescent dams may have been directly mediated by progesterone. In addition, the increase in late pregnancy placental VEGFA expression could have been in response to the increasingly hypoglycemic conditions in the underfed dams (see maternal glucose data presented in Luther et al. [12] or as a consequence of maternal anemia and/or attenuated uterine blood flow, leading to reduced oxygen delivery and hence hypoxia at the level of the gravid uterus.
Although proliferative growth of the ovine placenta is largely confined to the first half of gestation, the current study confirms that major changes in vascular structure are programmed to occur during the second half of pregnancy to accommodate exponential growth of the fetus. Irrespective of maternal dietary intake in these adolescent pregnancies, capillary area density increased from Day 90 to 130 of gestation in both the maternal caruncle (1.5-fold) and fetal cotyledon (2-fold), and cotyledon capillary surface and number density increased 2-fold as gestation progressed. These changes mirrored those recently reported during normal pregnancy in adult ewes, using identical procedures [24]. Furthermore, both sets of data are similar to the observations of Stegeman [14] both in pattern and magnitude of the increase in placental vascularization. In addition, both Borowicz et al. [24] and Teasdale [48] showed an increase in the number of capillaries or capillary endothelial cells in the fetal cotyledon during the last third of gestation in sheep. Thus, although cellular proliferation of the entire placentome may decline, selected cell types, and especially endothelial cells, appear to increase in number during the last half of gestation; the mechanism of this increase remains to be determined.
Using a microscopic, morphometric point-counting method, Stegeman [14] demonstrated that vascularization of the caruncle increased by 1.5-fold from approximately Day 80 to Day 136 of gestation in adult ewes carrying twin fetuses. Moreover, these increases in placentome capillary area density and in cotyledon capillary number and surface density from midgestation to late gestation are in broad agreement with previously documented changes in uterine (maternal) and umbilical (fetal) blood flow in vivo. Measurements of blood flow at midgestation and late gestation reveal a 2- to 3-fold increase in uterine flow [49] and a 3- to 4-fold increase in umbilical flow [16]. The relatively greater increase in umbilical flow is commensurate with higher cotyledon capillary surface density, and hence greater surface area for transcapillary exchange [14, 24, 50].
As gestation advanced from Day 90 to Day 130, the relative expression of the angiogenic factors and receptors increased substantially (2- to 4-fold) within the whole placentome, commensurate with the changes in vascular development discussed above. The increase in angiogenic gene expression:18S ratios observed were due to both a decrease in the expression of 18S rRNA and an increase in the expression of the angiogenic factor/receptor mRNA itself. The decline in 18S rRNA expression may represent the gestational decrease in placental cellular proliferation. The overall rate of placental cellular proliferation declined from Day 90 to Day 130 of gestation in the current study and reflected the decrease in placental weight.
The current findings provide mechanistic insights into the relationship between inadequate maternal weight gain during adolescent pregnancy and birth weight in our undernourished ovine paradigm. The alterations in maternal placental vascular development reported herein may play a role in mediating reductions in maternal, and hence fetal, nutrient supply, which in turn underlie the previously reported reduction in fetal soft tissue growth [12]. Furthermore, the failure to reverse these changes in maternal placental vascularity after realimentation to control dietary intakes may have relevance to human adolescent pregnancies, as inadequate gestational weight gains, particularly during the first two thirds of gestation (when the human placenta also undergoes its main growth phase), have been shown to be a predictor of reduced infant birth weight [7–10].
FOOTNOTES
1Supported by the Scottish Executive Environment and Rural Affairs Department and the National Institutes of Health. 
Correspondence: 2FAX: 44 1224 716686; e-mail: Jacqueline.Wallace{at}rowett.ac.uk
Received: 11 March 2007.
First decision: 7 April 2007.
Accepted: 23 April 2007.
REFERENCES
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
