| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Research Article |
Rowett Research Institute,3 Bucksburn, Aberdeen AB21 9SB, United Kingdom
Department of Animal and Grassland Research,4 National Agricultural Research Center for Kyushu Okinawa Region, Kumamoto 861-1192, Japan
ABSTRACT
In the overnourished adolescent sheep, maternal tissue synthesis is promoted at the expense of placental growth and leads to a major decrease in lamb birth weight at term. Maternal growth hormone (GH) concentrations are attenuated in these pregnancies, and it was recently demonstrated that exogenous GH administration throughout the period of placental proliferation stimulates uteroplacental and fetal development by Day 81 of gestation. The present study aimed to determine whether these effects persist to term and to establish whether GH affects fetal growth and body composition by increasing placental size or by altering maternal metabolism. Adolescent recipient ewes were implanted with singleton embryos on Day 4 postestrus. Three groups of ewes offered a high dietary intake were injected twice daily with recombinant bovine GH from Days 35 to 65 of gestation (high intake plus early GH) or from Days 95 to 125 of gestation (high intake plus late GH) or remained untreated (high intake only). A fourth moderate-intake group acted as optimally nourished controls. Pregnancies were terminated at Day 130 of gestation (6 per group) or were allowed to progress to term (8–10 per group). GH administration elevated maternal plasma concentrations of GH, insulin, glucose, and nonesterified fatty acids during the defined treatment windows, while urea concentrations were decreased. At Day 130, GH treatment had reduced the maternal adiposity score, percentage of fat in the carcass, and internal fat depots and leptin concentrations, predominantly in the high-intake plus late GH group. Placental weight was lower in high-intake vs. control dams but independent of GH treatment. In contrast, fetal weight was elevated by late GH treatment, and these fetuses had higher relative carcass fat content, perirenal fat mass, and liver glycogen concentrations than all other groups. Expression of leptin mRNA in fetal perirenal fat and fetal plasma leptin concentrations were not significantly altered by maternal nutritional intake or GH. In pregnancies proceeding to term, the duration of gestation, fetal placental mass, and lamb birth weight were reduced in high-intake compared with control dams but were not significantly affected by GH treatment. In conclusion, exogenous GH has profound effects on maternal endocrinology, metabolism, and body composition when administered during early and late pregnancy. Treatment during late pregnancy has a modest effect on fetal growth independent of placental size and a profound effect on fetal adiposity, which may have implications beyond the fetal period.
adolescent pregnancy, conceptus, fetal adiposity, growth hormone, mechanisms of hormone action, nutrient partitioning, placenta, pregnancy
INTRODUCTION
When pregnancy coincides with the continued growth of the mother, the normal hierarchy of nutrient partitioning may be altered at the expense of fetal nutrient supply. In the human, up to 50% of adolescents continue to grow while pregnant, and despite larger pregnancy weight gains and increased fat stores, these girls deliver smaller babies compared with nongrowing adolescent mothers [1]. Similarly, it has consistently been shown that overnourishing the singleton-bearing adolescent sheep throughout pregnancy promotes maternal growth (primarily of adipose tissue) at the expense of the gravid uterus [2–4]. Consequently, the growth of the placenta is impaired, resulting in the premature delivery of low-birth-weight lambs relative to moderately fed control adolescents of equivalent gynecological age. Rapid maternal growth is also associated with higher rates of spontaneous abortion in late gestation [3]. A sheep model replicates the key features of adverse pregnancy outcome in human adolescents, namely, an increased risk of abortion, preterm delivery, and low birth weight [5–7].
In the overnourished adolescent model, maternal growth hormone (GH) concentrations are attenuated throughout gestation [8], and the actively maturing somatotrophic axis may play a key role in coordinating nutrient use during pregnancy. Although maternal plasma GH concentrations during human adolescent pregnancy have not been reported, to our knowledge, obese adolescent girls who have reached puberty have blunted GH secretion [9]. Because inadequate placental growth is central to the reduction in fetal growth and birth weight observed in the sheep model [2, 10], a recent study investigated the therapeutic effect of maternal GH administration throughout the period of rapid placental proliferation (namely, Days 35–80 of gestation) [11]. Based on necropsy data obtained on Day 81 of pregnancy, the growth of the uteroplacenta was stimulated, and the transplacental glucose gradient was increased 2-fold. Furthermore, fetal insulin and glucose levels were elevated, and the mass of the fetal liver and kidneys was increased. At this stage of gestation, the fetus has achieved only 8% of its final birth weight, and we clearly need to establish whether birth weight would be restored to normal levels if the pregnancies progressed to term. Moreover, the mechanisms underlying this major GH-mediated alteration in nutrient partitioning in favor of uteroplacental and fetal growth are not completely clear. While the initial results suggest that maternal GH may be operating directly to enhance the growth of the placenta, the higher anabolic state in the fetus may simply reflect the increased availability of glucose in the maternal circulation and be independent of placental size or transport effects. GH does not cross the placenta [12], but elevated maternal GH concentrations may improve nutrient supply to the gravid uterus by inhibiting the lipogenic actions of insulin and by increasing glucose concentrations in the maternal and fetal circulation [13]. The present study aimed to clarify these putative mechanisms by administering GH to overnourished adolescent dams during the period of rapid placental growth (Days 35–65 of gestation) or after the growth of the placenta is complete and fetal nutrient demand is high (Days 95–125). The impact on fetal growth and on maternal and fetal body composition was determined in a subset of ewes at Day 130 of gestation, while the remaining ewes were allowed to spontaneously deliver at term.
MATERIALS AND METHODS
Animals and Experimental Design
All procedures were licensed under the U.K. Animals (Scientific Procedures) Act of 1986 and by the Rowett Research Institute's Ethical Review Committee. Embryos from superovulated adult ewes (Border Leicester x Scottish Blackface), inseminated by a single sire, were recovered on Day 4 postestrus and were transferred synchronously in singleton into the uterus of recipient ewe lambs (Dorset Horn x Mule) as described previously by Wallace et al. [8]. Donor ewes (n = 10) were multiparous, between 3 and 4 yr of age, and had a body condition score of 2.40 ± 0.03 U at the time of embryo recovery. Embryo transfer was carried out during the midbreeding season, and the animals were housed in individual pens under natural lighting conditions at the Rowett Research Institute (57°N, 2°W). At the time of embryo transfer, the recipient ewe lambs were pubertal (age, 
8.5 mo), with a mean live weight of 43.1 ± 0.35 kg and a body condition score of 2.30 ± 0.01 U. Immediately following embryo transfer, recipients were initially allocated to one of two nutritional treatments on the basis of live weight, body condition score, and ovulation rate at the time of transfer. Where possible, care was also taken to randomize for the maternity of the embryo. Recipients were individually offered a moderate (n = 19) or high (n = 56) quantity of the same complete diet. The dietary level in the moderate group was in fact a control intake level calculated to maintain normal maternal adiposity throughout gestation and to meet the estimated metabolizable energy requirements for optimum conceptus growth and pregnancy outcome in this genotype. In adolescent dams, this was achieved by allowing a moderate maternal weight gain (50 g/day) during the first two thirds of gestation, followed by stepwise increases in maternal intake during the final third of gestation to meet the increasing demands of the developing fetus. In contrast, the high-intake dams received approximately twice maintenance that was calculated to promote rapid maternal growth leading to obesity [4]. The complete diet supplied 12 MJ of metabolizable energy and 140 g of crude protein per kilogram of body weight and was offered in two equal feeds at 0800 and 1600 h daily. The diet contained 30% (w/w) coarsely milled hay, 42.25% barley, 10% molasses, 16.75% soybean meal, 0.35% salt, 0.5% dicalcium phosphate, and 0.15% vitamin-mineral supplement and had an average dry matter of 86%. Animals offered moderate intakes received their entire ration immediately, while those offered high intakes had the level of feed gradually increased during a 2-wk period until the level of daily feed refusal was approximately 15% of the total offered (equivalent to ad libitum intakes). The level of feed offered was reviewed three times weekly and was adjusted on an individual basis as appropriate on the basis of body weight change data and the level of feed refused (recorded daily). Maternal body condition or adiposity score was subjectively assessed on a five-point scale (1 = emaciated, 5 = obese) [14] by one member of the team at approximately fortnightly intervals.
Pregnancy rate was initially estimated by measurement of plasma progesterone concentrations at Day 18 of the estrous cycle. High-intake dams were then allocated to receive GH from Days 35 to 65 of gestation (high intake plus early GH) or from Days 95 to 125 of gestation (high intake plus late GH) or remained untreated (high intake only). All moderate-intake control dams remained untreated throughout pregnancy. Recombinant bovine GH was administered twice daily (at 0800 and 1800 h) at a dose rate of 0.14 mg/kg of body weight per day. The GH was solubilized in 0.035 mM sodium bicarbonate buffer (pH 9.5) and was administered subcutaneously, taking care to alternate the injection sites on either side of the neck. Individual dose volumes (<2 ml) were adjusted weekly on the basis of current live weight throughout the treatment period. All ewes (including the control group) not receiving GH during the defined treatment windows were temporarily restrained twice daily, and all ewes rapidly adapted to being handled. Transabdominal ultrasound at approximately Day 45 of gestation confirmed viable fetuses in 16, 14, 15, and 15 ewes in the control, high-intake, high-intake plus early GH, and high-intake plus late GH groups, respectively. Further ultrasound examinations of all ewes at least once 5 to 7 days after treatment had commenced during the early and late GH treatment windows were used to subjectively assess fetal fluid levels and to check fetal viability. At Day 130 of gestation (5 days after the late GH treatment ceased), six randomly selected pregnancies per group were terminated as described herein. The remaining ewes were allowed to spontaneously deliver at term.
Measurements at Necropsy
On Day 130 of gestation, selected ewes received their morning feed, were weighed and scored for body condition score, and were then killed by i.v. injection of an overdose of sodium pentobarbitone (20 ml of Euthesate; 200 mg/ml of pentobarbitone; Willows Francis Veterinary, Crawley, U.K.) 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 of Euthesate) to kill the fetus. After clamping the cord, the fetus was removed, dried, and weighed. All major fetal organs were dissected and weighed. Selected organs, including the liver and perirenal fat, were sampled and snap frozen in liquid nitrogen-chilled isopentane and stored at –80°C until analysis. Twelve intact placentomes (representative of size and morphology of the remaining placentomes and from both uterine horns) were removed, weighed, and snap frozen as described or were fixed for analyses not reported herein. For the remaining placental tissues, the intact placentomes were dissected, and their number and weight were recorded (represents 85% of the tissue originally available). The chorioallantoic membranes and residual uterus were weighed, and the volume of the combined amniotic and allantoic fluids was recorded.
The maternal liver and perirenal fat were dissected and weighed. The omental and mesenteric fat was stripped from the gastrointestinal tract and together with the perirenal fat weight provided an index of the maternal internal fat depot. The maternal body was divided into carcass and noncarcass components, and the chemical composition of the carcass was determined after removing the pelt, head, and feet. The composition of the fetal carcass was determined after removing the head only. The carcasses were individually homogenized in a Wolfking (Slagelse, Denmark) mincer [15]. Three samples of each homogenate were freeze-dried, ground in a laboratory mincer, and subsampled for analyses. The residual moisture content of the carcass remaining after freeze-drying was obtained by drying to constant weight at 100°C. The fat content was determined by the chloroform-methanol technique [16] and the nitrogen content by the Kjeldahl method [17]. The ash content of the samples was determined by ashing at 600°C to constant weight. To determine fetal liver glycogen stores, glycogen was extracted and hydrolyzed to release glucose, and then glucose was measured using glucose oxidase and glucose peroxidase [18].
Pregnancy Outcome at Term
For the remaining ewes, pregnancy outcome was determined after spontaneous vaginal delivery at term. All ewes were supervised throughout the delivery period. Immediately after parturition, the lamb was towel dried, weighed, and measured (girth and crown rump length). After the placenta (fetal cotyledons plus membranes) was delivered, it was washed, blotted, and weighed. Fetal cotyledons were dissected and counted, and their total weight was recorded.
Blood Sampling and Biochemical Analysis
Blood samples were collected from all ewes by jugular venipuncture at approximately 10-day intervals from Day 38 of gestation onward. These samples were analyzed for GH, insulin, glucose, urea, and nonesterified fatty acids (NEFAs). An additional maternal blood sample collected on the morning of necropsy was also analyzed for leptin. Fetal blood samples were analyzed for insulin and leptin. GH, insulin, and leptin concentrations were measured in duplicate by radioimmunoassays described previously [8, 19, 20]. The limit of detection was 1 ng of GH, 4 µU of insulin, and 0.4 ng of leptin per milliliter. The intraassay and interassay coefficients of variation were 4.2% and 4.4%, respectively, for GH and 3.3% and 6.9%, respectively, for insulin. The intraassay coefficient of variation for leptin was 10.2%. Plasma glucose concentrations were measured in duplicate with a Yellow Springs Instruments (YSI) (Yellow Springs, OH) dual biochemistry analyzer (model 2700). The YSI analyzer was calibrated with known standards after every fourth determination. Plasma NEFA and urea concentrations were determined as originally described [21, 22].
Quantitative Real-Time RT-PCR
The mRNA levels for fetal perirenal fat leptin were determined using quantitative real-time RT-PCR. Total RNA was extracted from fetal fat (100 mg) using an RNeasy lipid tissue kit (QIAGEN GmBH; Hilden, Germany). The quality and quantity of total RNA were determined via capillary electrophoresis using an Agilent 2100 Bioanalyzer (Agilent Technologies, Wilmington, DE). All reagents and procedures used for the real-time RT-PCR were purchased from and used as described by Applied Biosystems [23]. For each sample, approximately 30 ng of total RNA was reverse transcribed in triplicate using TaqMan Reverse Transcription Reagents and MultiScribe Reverse Transcriptase. A Taqman probe and primers sets were designed from species-specific sequences for leptin using Primer Express Software (Applied Biosystems, Warrington, Cheshire, U.K.). The sequences of cDNA forward and reverse primers and the Taqman probe for leptin were as follows: 5'-GCTCCACCCTCTCCTGAGTTT-3' (132–152, GenBank accession number U84247), 5'-ACCAACAGATCCTCGCCAGT-3' (182–201), and 5'-TCCAAGATGGACCAGACATTGGCAATC-3' (154–180), respectively. Polymerization and amplification reactions for each RT were carried out in duplicate in 96-well PCR plates in a final volume of 10 µl using the Applied Biosystems ABI PRISM 7000 Sequence Detector. Hybridization and polymerization were carried out at 60°C, generally for 40 cycles. Quantification was determined using a relative standard curve method using different doses of a reference standard cDNA generated from RNA pooled from Day 130 fetal perirenal fat samples. Individual tissue mRNA was expressed relative to the samples' internal 18S RNA using 18S PDAR kit reagents from Applied Biosystems.
Data Analysis
Data were analyzed by one-way ANOVA (Minitab 13; Minitab Inc., State College, PA). If significant differences were indicated (P < 0.05), this was followed post hoc by Fisher least significant difference procedure to determine which treatments differed. Regression analyses were also used to combine specific group comparisons for the Day 130 data set only. For the data reported, this primarily involved comparing the moderate-intake (control) group with all three high-intake groups, or comparing the high-intake plus late GH group with the high-intake and the high-intake plus early GH groups. The latter approach has been clearly indicated in the text. While maternal hormone or metabolite concentrations were measured at approximately 10-day intervals, individual means were calculated for the periods spanning Days 38 to 58 (early pregnancy), Days 68 to 88 (midpregnancy), and Days 98 to 119 (late pregnancy), and these individual means were analyzed by one-way ANOVA as already described. Within treatment groups, maternal data for pregnancies terminated at Day 130 were initially compared with those progressing to term. Because no differences were detected in any of the parameters measured, the data were combined. For the body condition score data, the change in adiposity score during defined periods of gestation was compared using paired Student t-test. Correlation analyses were by Pearson product moment correlation test.
RESULTS
Three high-intake plus late GH treated ewes displayed persistent inappetance within 5 days of commencing GH treatment. Major polyhydramnios was evident at ultrasonography, and the ewes and fetuses were withdrawn from the study and humanely killed to comply with the U.K. Home Office regulations. A fourth ewe displayed similar but less severe symptoms that immediately resolved when GH treatment ceased. This ewe gave birth to a viable lamb. All data pertaining to these 4 pregnancies were excluded from further analyses.
Maternal Dietary Intakes, Live Weight and Body Condition Score Changes, and Body Composition at Necropsy
The mean weekly maternal food intakes from embryo transfer to Day 130 of gestation are shown in Figure 1. Maternal dietary intakes were elevated in high-intake dams compared with moderate-intake (control) dams throughout gestation (P < 0.001). Voluntary food intakes decreased during GH treatment in the early and late GH treatment groups but remained elevated relative to the control group throughout gestation (P < 0.001). During the period spanning Days 35 to 65 of gestation, food intake in the high-intake plus early GH group was 14.3% lower than that in the high-intake group (P < 0.05) and was 11.3% lower than that in the high-intake plus late GH group (P > 0.05). During the late pregnancy treatment window (Days 95–125 of gestation), food intake in the high-intake plus late GH group was reduced by 31.0% and 33.6% compared with that in the high-intake group and the high-intake plus early GH group, respectively (P < 0.001). Within the three high-intake groups, food intake was equivalent during the intervening period (Days 66–94 of gestation) when none of the animals were receiving exogenous GH.
|
The changes in maternal weight and adiposity score in response to the high dietary intakes for all ewes combined are shown in Figure 1. From an equivalent live weight and adiposity score at embryo transfer, maternal live weight changes in the high-intake groups throughout the study were similar and largely unaffected by GH treatment (maternal weight gains of 305, 299, and 280 g/day for the high-intake, high-intake plus early GH, and high-intake plus late GH groups, respectively). The external assessment of body condition revealed that the moderate dietary intakes maintained maternal adiposity at the desired initial control level throughout gestation in all ewes. The high-intake groups became increasingly obese as pregnancy proceeded, but adiposity scores plateaued during the early and late GH treatment periods. During the period spanning Days 35 to 65 of gestation, the percentage increase in adiposity score for the high-intake plus early GH dams was 3.6% compared with 11% and 12% in the high-intake dams and the high-intake plus late GH dams, respectively (P < 0.001). Similarly, during late pregnancy the percentage increase in adiposity score in the high-intake plus late GH dams was 0.8% compared with 6.2% and 7.9% in the high-intake dams and the high-intake plus early GH dams, respectively (P < 0.002).
More details of the effect of GH on maternal tissue deposition were obtained by dissection and body composition analyses in a subset of ewes at Day 130 of gestation (Table 1). Within the three high-intake groups, final live weight corrected for the weight of the gravid uterus was equivalent, but the corrected live weight gain calculated throughout gestation was lowest in the high-intake plus late GH group. Treatment of high-intake dams with GH reduced the body condition score, the percentage of fat in the maternal carcass, the relative mass of the internal fat depots, and the leptin concentration in the maternal circulation. These indices of maternal adiposity were lowest in the high-intake plus late GH group and were significantly different compared with those in the high-intake and control groups. Irrespective of treatment group, the percentage of fat in the maternal carcass was positively associated with the external body score (r = 0.786, n = 24, P < 0.001), the relative mass of the internal fat depots (r = 0.793), and the maternal leptin concentrations (r = 0.840). This contrasts with the inverse relationship between the percentage of fat and the percentage of protein in the maternal carcass (r = –0.920, P < 0.001).
|
Pregnancy Outcome at Necropsy
Details of placental and fetal morphometry at necropsy on Day 130 of gestation are given in Table 2. Late but not early gestation GH treatment of high-intake dams restored gravid uterine weight to that of control dams. Total and mean placentome weight was lower in high-intake dams than in moderate-intake control dams (P < 0.004, regression analyses) but was not affected by GH treatment during early or late gestation. Placentome number was equivalent in all four groups. In contrast, within the high-intake groups, fetal weight was elevated by late pregnancy GH treatment (P < 0.03, regression analyses) and was intermediate between the moderate-intake and other high-intake groups. These changes in fetal weight were closely mirrored by significant differences in the absolute weight of all fetal organs except the brain and adrenal glands (data not shown). When expressed relative to fetal weight, the brain, kidneys, and gut were larger in the high-intake and high-intake plus early GH groups compared with the control group, while the high-intake plus late GH group relative organ weights were intermediate (Table 2).
|
Exogenous GH treatment during early and late pregnancy was associated with varying degrees of polyhydramnios (excess fetal fluid volume) in all ewes, as seen at ultrasound while treatment was ongoing. There was no evidence of polyhydramnios in any of the high-intake or control ewes at any stage, or in the high-intake plus early GH group when examined at approximately Day 97 of gestation. At autopsy, the combined amniotic and allantochorionic fluid volume was equivalent in the high-intake group compared with the high-intake plus early GH group but remained elevated in the high-intake plus GH group (Table 2).
The chemical composition of the fetal body and other indices of adiposity at necropsy are given in Table 3. Fetal empty carcass weight was equivalent in the control and high-intake plus late GH groups and higher than in the remaining high-intake groups (P < 0.05). There was no difference in the relative carcass ash or crude protein content between treatments. In marked contrast, the absolute and relative fetal carcass fat content and perirenal fat mass were significantly increased by late GH treatment relative to all other groups (P < 0.05). This increase in adiposity was mirrored by a similar increase in fetal liver glycogen concentration and total liver glycogen stores. Irrespective of treatment group, the percentage of fat in the fetal carcass was positively associated with the relative perirenal fat mass (r = 0.703, n = 24, P < 0.001).
|
Maternal Metabolic and Endocrine Status
Maternal plasma GH concentrations were significantly elevated in control compared with high-intake dams throughout the study (P < 0.01) (Fig. 2a). Exogenous GH administration stimulated a pharmacological increase in maternal GH concentrations during the treatment window in the early and late GH treatment groups (Fig. 2a). Maternal insulin concentrations were lower in control compared with high-intake dams throughout gestation (P < 0.01) (Fig. 2b). Comparison of the high-intake groups revealed that exogenous GH treatment stimulated a 3-fold to 4-fold increase in maternal insulin concentrations during the early and late GH treatment windows. Similarly, maternal glucose concentrations were elevated 2-fold to 3-fold during the GH treatment periods (Fig. 3a), while maternal NEFA concentrations increased by 3-fold and 5-fold during the early and late GH treatment windows, respectively (Fig. 3b). Maternal plasma urea concentrations were lower in control vs. high-intake ewes throughout the study (P < 0.01). During the early gestation window, urea concentrations in the high-intake plus early GH group were equivalent to those in the control group, whereas during the late gestation window, GH treatment decreased urea concentrations to a level significantly below that of the control group (P < 0.05) (Fig. 3c). The limited sampling frequency suggests that these changes in maternal endocrinology and metabolic status were largely limited to the defined treatment periods in that concentrations had returned to normal for the high-intake group by the first sampling point after GH treatment had ceased.
|
|
Fetal Endocrine Status and Perirenal Fat Leptin Expression at Necropsy
Fetal plasma glucose analysis was compromised at the plasma separation stage; therefore, fetal plasma glucose data at necropsy are unavailable. Fetal insulin concentrations were lower in fetuses from high-intake vs. moderate-intake dams (P < 0.05) but were unaffected by GH treatment (Table 4). Maternal nutrition and GH treatment did not significantly affect relative fetal perirenal fat leptin mRNA expression or circulating leptin concentrations in the fetal plasma.
|
Pregnancy Outcome at Term
All pregnancies allowed to progress to term resulted in the birth of viable lambs. Again, early and late GH treatment significantly reduced the maternal adiposity score at term (P < 0.05). The duration of gestation, fetal placental mass, and lamb birth weight were significantly reduced in high-intake vs. moderate-intake dams (P < 0.001, by regression analyses) (Table 5) but were not affected by GH treatment during early or late gestation. Lamb girth in the high-intake plus late GH group was intermediate between that of the control group and the remaining high-intake groups.
|
DISCUSSION
The key features of the overnourished adolescent paradigm have been reproduced in the present study, namely, increased maternal growth and adiposity leading to impaired placental development and fetal growth restriction [2, 4]. Exogenous GH had a profound effect on maternal endocrinology and metabolism during the early and late pregnancy treatment windows and was associated with a significant decrease in maternal adiposity relative to the high-intake only group. Treatment with GH during late but not early pregnancy modestly stimulated fetal growth and had a major effect on fetal body composition as assessed at Day 130 of gestation. Placental weight was not affected by GH administration during either treatment period, suggesting that exogenous GH primarily alters the fetal growth trajectory in these adolescent pregnancies by affecting maternal metabolism and the availability of glucose in the maternal circulation. The timing and duration of GH administration also appear to be crucial in that treatment during early pregnancy had no persistent effect on fetal size and body composition near term. This is probably because the necropsy stage (Day 130) was too far away from the early GH treatment window (Days 35–65) for any stimulating effects on fetal growth or body composition to be maintained. Similarly in the pig, treatment with porcine GH during early to mid gestation did not improve birth weight at term, whereas treatment during most of pregnancy or during late pregnancy only enhanced piglet weight at term [24–26].
The changes in maternal endocrinology and metabolic status assessed at 10-day intervals throughout gestation were similar to those reported in detail during the defined treatment window in an earlier study [11]. Administration of GH induced a state of insulin resistance with significantly elevated maternal insulin, glucose, and fatty acid concentrations during the specific treatment periods. It is well established that GH inhibits tissue responsiveness to insulin, resulting in decreased lipogenesis [27]. In the growing nonpregnant animal, this would favor redirecting the glucose destined for maternal lipid synthesis to protein (muscle) deposition, but in the pregnant animal nutrients can also be partitioned to the gravid uterus. Indeed, glucose uptake by the gravid uterus should be favored because placental glucose uptake and transfer are insensitive to variations in maternal insulin [28]. Glucose is the primary fetal fuel [29], and fetal demand is obviously highest during the final third of pregnancy, when the fetus completes 80% of its growth. In the overnourished adolescent model, the placenta does not significantly compromise fetal nutrient supply until the final third of gestation [10, 30]. The results of the present study suggest that GH treatment during late pregnancy improves fetal nutrient supply in these putatively growth-restricted pregnancies and thereby partially ameliorates fetal growth restriction in the overnourished dams. Fetal glucose levels were not measured in the present study because of a sample processing error. However, it is axiomatic that the 3-fold elevation in maternal glucose would produce a similar increase in the transplacental glucose gradient, resulting in high fetal glucose concentrations, as reported previously [11]. Therefore, although placental growth was stunted in response to the high-intake diet, this was countered in part by an excess of substrate for fetal growth during the GH treatment window.
A major increase in fetal adiposity (namely, relative perirenal fat mass and carcass fat content) and in fetal liver glycogen concentrations was measured in the high-intake plus late GH group. Similar increases in the absolute and relative mass of the perirenal and subcutaneous fat depots have previously been reported in normally growing sheep fetuses directly infused with glucose for approximately 27 days before spontaneous delivery [31], indicating a major role for glucose in fetal lipid deposition. Body lipid accounts for 60% to 70% of the energy available to the newborn lamb, and as much as 90% of liver glycogen stores can be used during the first day after birth [32]. Therefore, on the face of it, this increase in fetal lipid deposition may be beneficial during the neonatal period, protecting the newborn from hypothermia. However, the increase in fetal adiposity may have consequences beyond the neonatal period. Offspring of human pregnancies complicated by maternal diabetes mellitus, gestational diabetes, or maternal glucose intolerance during fetal life have a higher risk of becoming obese or glucose intolerant as adults [33–38]. Therefore, naturally occurring conditions and experimental treatments associated with excess nutrient supply to the fetus may result in permanent changes within the appetite regulatory system or the adipocyte, resulting in increased adiposity postnatally [39]. For example, diabetes in pregnant rats leads to malorganization of hypothalamic neuropeptidergic neurons in their offspring, and these acquired alterations are preventable by treatment of maternal diabetes [40]. Further support for this concept comes from a recent study in sheep [41]. Intrafetal infusion of glucose for only 10 days in late pregnancy was associated with an increase in size of the lipid locules in the fetal perirenal fat depot and with altered hypothalamic gene expression of neuropeptides known to regulate appetite and energy balance in adult life. Clearly, follow-up studies are required to document the long-term consequences of the major changes in fetal adiposity observed in response to maternal GH treatment in the present study.
The increase in fetal adiposity in the high-intake plus late GH group was not paralleled by statistically significant increases in fetal perirenal fat leptin mRNA expression or circulating fetal leptin concentrations. Necropsies were carried out five days after treatment had ceased, by which stage maternal glucose concentrations had returned to normal for the high dietary intake group. This normalization of placental and fetal glucose supply resulted in normal fetal insulin concentrations, and recent findings indicate that insulin rather than glucose is the major regulator of leptin expression in ovine fetal adipose tissue, with a 2-fold increase in mRNA being measured after 24 h of selective hyperinsulinemia with euglycemia [42]. The failure to observe changes in leptin gene expression or in circulating fetal concentrations in the present study is consistent with this observation.
The altered body composition of the fetuses in the high-intake plus late GH group may also partially reflect an anabolic effect of exogenous GH on protein metabolism in the dams. GH increases amino acid uptake and protein synthesis while reducing urea synthesis [43]. This results in low blood urea concentrations, as reported in the control vs. high-intake dams (high vs. low endogenous GH) in this study, and was further exacerbated in the high-intake plus late GH group during exogenous GH treatment. Maternal carcass crude protein percentage was also modestly increased in the high-intake plus late GH relative to the high-intake group. Although glucose is the major substrate source for the fetus, amino acids are required for tissue accretion. Hence, maternal GH treatment, by slightly increasing maternal tissue amino acid uptake, may have reduced the availability of some specific amino acids for placental transfer to the fetus. Individual amino acid accretion rates into the fetal carcass were not measured in the present study, but such a scenario could have limited amino acid incorporation into fetal protein at a time when excess glucose availability was promoting adipose tissue deposition.
We did not demonstrate significant changes in birth weight in the small number of high-intake plus late GH pregnancies (n = 5) that proceeded to term in the present study, but this does not preclude an effect on neonatal body composition as reported herein for other species. In this regard, the modest increase in lamb girth relative to the other high-intake groups is intriguing and may be indicative of an effect on central adiposity in these neonates. The number of animals studied in the term group was compromised by four ewes having to be withdrawn from the study because of severe inappetance. This metabolic disturbance in response to exogenous GH was probably due to the abrupt increase in fat mobilization in an already fat animal, resulting in ketosis. Transient and less severe decreases in intake were also observed in the early GH group and in the animals in a previous study [11], but these animals had a lower body fat proportion when treatment commenced on Day 35 of gestation. In retrospect, reducing the dose administered to allow for the major differences in maternal adiposity may have avoided these problems. Previous ovine studies using an equivalent dose of GH during late pregnancy in adult ewes have not reported a major decrease in maternal intake or any similar metabolic crisis [44–46], but ewes were generally being fed to meet maintenance requirements rather than the twice maintenance rations used herein. However, in the pregnant pig the fetal growth response to exogenous porcine GH is highly dependent on maternal nutrient intake and on the stage and duration of the treatment window [25]. Indeed, no fetal endocrine or growth response to exogenous GH in moderate-intake control ewes was previously detected [11].
In the earlier study, GH treatment for 45 days resulted in a 4-fold increase in fetal fluid volume at necropsy, which was strongly correlated with the elevated maternal glucose and fetal glucose and lactate concentrations [11]. This was of concern because of the known clinical association between polyhydramnios and fetal death in late pregnancy and at term in humans [47, 48]. In this study, the occurrence of polyhydramnios was monitored by ultrasound and was evident in all ewes within one week of commencing GH treatment. The polyhydramnios mirrored the GH-mediated changes in maternal metabolic indices and largely resolved once treatment ceased. Moreover, in GH-treated ewes proceeding to term, parturition was normal, and all GH-treated fetuses were viable at birth. Nevertheless, the occurrence of polyhydramnios, albeit transient, in combination with the marked increase in fetal adiposity and frequently observed severe maternal inappetance may limit the therapeutic value of attempting to use exogenous maternal GH treatment to ameliorate fetal growth restriction.
In conclusion, exogenous GH profoundly alters maternal endocrinology when administered during early and late gestation windows, and it reduced maternal fat deposition in the overnourished adolescent dam. In late pregnancy when fetal nutrient demand is high, this alteration in nutrient partitioning and the associated increase in maternal glucose availability have a modest positive effect on fetal growth that is independent of the growth of the placenta. This is associated with a major increase in fetal adiposity, which may have implications beyond the fetal period.
ACKNOWLEDGMENTS
We thank Terry Atkinson for iodinating insulin, GH, and leptin; Patricia Findlay for performing the leptin assay; and Graham Horgan for statistical advice. The GH was kindly donated by Monsanto Company (St. Louis, MO).
FOOTNOTES
101 Supported by the Scottish Executive Environment and Rural Affairs Department as part of the grant-in-aid to the Rowett Research Institute. 
1 Correspondence: FAX: 01224 716622; Jacqueline.Wallace{at}rri.sari.ac.uk
Received: 20 March 2006.
First decision: 20 April 2006.
Accepted: 30 April 2006.
REFERENCES
This article has been cited by other articles:
|
|
 |
|
  C. D Wright, R. J Orbus, T. R H Regnault, and R. V Anthony Effects of early gestation GH administration on placental and fetal development in sheep J. Endocrinol., July 1, 2008; 198(1): 91 - 99. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. G Lea, P. Wooding, I. Stewart, L. T Hannah, S. Morton, K. Wallace, R. P Aitken, J. S Milne, T. R Regnault, R. V Anthony, et al. The expression of ovine placental lactogen, StAR and progesterone-associated steroidogenic enzymes in placentae of overnourished growing adolescent ewes Reproduction, April 1, 2007; 133(4): 785 - 796. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
