|
|
||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
ANIMAL NUTRITION |



* Center for Nutrition and Pregnancy, Department of Animal Science, North Dakota State University, Fargo 58108-6050; and
Rowett Research Institute, Bucksburn, Aberdeen AB21 9SB, United Kingdom
| Abstract |
|---|
|
|
|---|
0.01) throughout gestation in High-fed ewes (58.3, 68.8, and 81.1 ± 1.6 kg, respectively). Relative jejunum mass (g/kg of maternal BW) was greater (P = 0.003) in Control-fed ewes compared with High-fed ewes and tended (P = 0.11) to decrease from d 50 to 130. There were diet x stage of gestation interactions (P
0.08) for ileum and small intestinal total and relative weights. Ileum mass (g/kg of maternal BW) in Control-fed ewes was less (P = 0.07) compared with High-fed ewes at d 50, was equal (P = 0.19) to High-fed ewes at d 90, and was greater (P = 0.02) than High-fed ewes at d 130. Small intestine mass (g/kg of maternal BW) was similar between Control- and High-fed ewes at d 50 and 90, but Control-fed ewes had greater (P = 0.01) mass at d 130. Jejunal RNA and protein concentrations were less (P
0.07) and DNA was unaffected (P = 0.43) in Control-fed compared with High-fed ewes. Stage of gestation did not affect jejunal RNA, DNA (mg/g), or protein concentrations. Jejunal cellular proliferation was not affected by diet or stage of gestation. In jejunal mucosal tissue, capillary number decreased, whereas capillary surface density and area per capillary increased (P = 0.01) with advancing pregnancy (d 50 vs. d 130), but were independent of diet. Data indicated that intestinal mass as a proportion of maternal BW declined in overnourished, gestating ewe lambs. This response was more pronounced during late gestation and is likely explained by the increasing maternal BW and adiposity rather than by the changing cellularity or cell proliferation.
Key Words: cellularity intestine nutrition pregnancy vascularity
| INTRODUCTION |
|---|
|
|
|---|
Previous studies have shown that the normal hierarchy of nutrient partitioning during pregnancy can be dramatically altered in young, growing females (Wallace et al., 1996
, 1999b
, 2001
). When young, pregnant ewes are overnourished to promote rapid tissue gain, both placental growth and fetal growth are compromised and mammary gland function immediately after parturition is impaired (Wallace et al., 1996
, 1999b
, 2002b
, 2006a
,b
). The impacts of overnutrition in pregnant ewes on maternal intestinal growth, cellularity, and vascularity have not been characterized and may similarly be responsive to altered nutrient supply or partitioning. Therefore, we hypothesize that previously observed alterations in nutrient partitioning to the gravid uterus and mammary gland may be partially mediated by dietary-induced changes in maternal intestinal growth, cellularity, and vascular development. In addition, we hypothesized that overnutrition would predispose the maternal gut to more dramatic changes in growth indices during advancing pregnancy.
| MATERIALS AND METHODS |
|---|
|
|
|---|
Animals, Experimental Design, and Treatments
Animals.
Embryos from superovulated adult ewes (Border Leicester x Scottish Blackface), inseminated by a single sire, were recovered on d 4 after estrus and transferred synchronously in singleton into the uterus of recipient ewe lambs (Dorset Horn x Greyface), exactly as described previously (Wallace et al., 1997
). Donors were multiparous, between 3 and 4 yr of age, weighed 82.5 ± 2.0 kg, and had a BCS of 2.3 ± 0.05 (on a 5-point scale, with 1 being emaciated and 5 obese) at the time of embryo recovery. This protocol ensured that placental and fetal growth was not influenced by varying fetal number or partial embryo loss. In addition, the use of a single sire and a limited number of embryo donors maximized the homogeneity of the resulting fetuses. Embryo transfer was carried out during the midbreeding season, and 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, recipient ewe lambs were peripubertal (approximately 7.5 mo of age) and had a mean BW of 44.3 ± 1.1 kg, a BSC of 2.3 ± 0.1, and an ovulation rate of 2.2 ± 0.21.
Treatments.
Recipient ewe lambs were initially allocated to 1 of 2 dietary treatments. Treatments were a control (Control) or high (High) quantity of the same complete diet. The dietary amount in the Control group was calculated to maintain normal maternal adiposity throughout gestation and to provide the estimated ME and protein requirement of ewe lambs carrying a singleton fetus according to the stage of pregnancy (based on Agricultural and Food Research Council; AFRC, 1993
). Previous studies have shown that this diet, as fed to ewes in the Control group, does not retard placental and fetal growth in this genotype (Wallace et al., 2004a
). In contrast, the High or ad libitum intakes were equivalent to approximately twice the estimated ME requirements and were calculated to promote rapid maternal BW gain. The complete diet was offered in 2 equal feedings at 0800 and 1600 h daily and contained 2.84 Mcal of ME/kg and 15.9% CP on a DM basis. On a fresh basis, the diet contained 30% (wt/wt) coarsely chopped grass hay, 42.25% rolled barley, 10% molasses, 16.75% soybean meal, 0.35% salt, 0.5% dicalcium phosphate, and 0.15% of a mineral supplement and had an average calculated DM of 86%. The mineral supplement contained 18% Ca, 10% Mg, 0.23% Co, 4.6% Mn, 3.9% Zn, 0.13% I, and 0.01% Se. The amount of feed offered ewes in the High-intake group was gradually increased over a 2-wk period until the amount of daily refusal was approximately 15% of the total offered (equivalent to ad libitum intakes). The amount of feed offered was reviewed 3 times weekly and adjusted, on an individual basis and when appropriate, on the basis of BW change data (recorded weekly) and the amount of feed refusal (recorded daily). Maternal body condition was subjectively assessed on a 5-point scale (1 = emaciated, 5 = obese; Russel et al., 1969
) every 2 wk by the same experienced technician. Pregnancy status was determined by transabdominal ultrasonography at approximately 45 d of gestation (gestation length = 145 d). Pregnancy was established and maintained in 22 and 18 ewes in the Control and High groups, respectively. Ewes were slaughtered at d 50, 90, or 130 of gestation.
Experimental Design. At embryo transfer, pregnant peripubertal ewe lambs were initially allocated evenly to either the Control or High group on the basis of their current BW, BCS, ovulation rate, and, where possible, donor source. Similarly, after being confirmed pregnant, ewe lambs were allocated to either the d 50, 90, or 130 group on the basis of initial BW gain and again were balanced for donor source where possible. The resulting treatment arrangement was a 2 x 3 factorial with diet (Control vs. High) and stage of pregnancy (50, 90, and 130 d) being combined into 6 individual treatments.
Necropsy and Tissue Harvesting Procedures
One hour before slaughter, ewes were weighed to obtain a final BW and were injected via a temporary jugular catheter with 5-bromo-2-deoxy-uridine (BrdU; 5 mg/kg of BW). Ewes were killed by administration of an overdose of sodium pentobarbitone i.v. (20 mL of Euthesate; 200 mg of pentobarbitone/mL; Willows Francis Veterinary, Crawley, UK) and were exsanguinated by severing the main blood vessels of the neck. Maternal blood was collected in a 57-L plastic container and weighed. The gravid uterus was quickly removed, weighed, and opened. At 90 and 130 d of gestation, fetuses were killed by immediate intracardiac administration of a sodium pentobarbitone overdose (5 mL of Euthesate, Willows Francis Veterinary). Each fetus was toweled dry and weighed, and the crown-rump length, girth at the umbilicus, biparietal head diameter, and sex were recorded. The gravid uterus was immediately removed and dissected from the vagina at the cervix and weighed. The ewe was eviscerated and the full viscera were weighed.
The liver, spleen, and pancreas were dissected out of the visceral tissues and weighed. The stomach complex was divided from the esophagus at the cardia and from the intestine at the pyloric valve. Digesta and fat were removed and the stomach complex was weighed. Intestinal tissues were located and the demarcations of duodenum, jejunum, ileum, and large intestine were made (Scheaffer et al., 2004a
; Soto-Navarro et al., 2004
; Neville et al., 2008
). A 150-cm section of jejunum was immediately removed for vascular perfusion (Soto-Navarro et al., 2004
; Reed et al., 2007
) and fixed with Carnoys solution (6:3:1 parts of absolute ethanol:chloroform:glacial acetic acid). After specific intestinal regions were identified, the mesentery was dissected away from the tissue; digesta were gently stripped, and the segments were weighed. Digesta from the stomach complex and intestines were combined and weighed. Perirenal, omental, and mesenteric fat was dissected and weighed.
Subsamples of maternal jejunal tissues (5 to 10 cm) were collected from the proximal end of the 150-cm section of jejunum before vascular perfusion and briefly washed in sterile PBS to remove digesta before being either frozen or immersion fixed. Samples were later analyzed for DNA, RNA, protein, and cellular proliferation. These samples were collected at a site 15 cm caudally down the mesenteric vein from the mesentericileocecal vein junction and then along the mesenteric arcade to the point of intestinal intersection. Samples were wrapped in foil, snap-frozen in supercooled isopentane (submerged in liquid N), and stored at –80°C.
Cellularity Estimates
Freshly thawed tissue samples (0.1 g) were homogenized with an UltraTurrax T-8 homogenizer (IKA Works, Staufen, Germany) in 3 mL of 2% ice-cold perchloric acid. After centrifugation for 10 min at 1,000 x g at 40°C, the supernatant was decanted and the pellet was dissolved in 5 mL of 0.3 M NaOH, and 0.5 mL was taken for the protein assay (Lowry et al., 1951
). The remaining 4.5 mL was mixed with 1 mL of 20% perchloric acid and centrifuged as described above. The supernatant was taken for spectrophotometric determination of RNA and the pellet was mixed with 3 mL of 2% perchloric acid on ice and centrifuged as described above, and the supernatant was taken for DNA analysis via the Burton method (Bonis et al., 1991
). The concentration of DNA was used as an index of hyperplasia, with protein:DNA and RNA:DNA ratios used as indexes of hypertrophy (Swanson et al., 2000
; Scheaffer et al., 2003
; Soto-Navarro et al., 2004
).
Jejunal Cell Proliferation.
For histological estimates of tissue cellular proliferation, BrdU was used (Jablonka-Shariff et al., 1993
; Jin et al., 1994
). Fresh jejunal tissue sections were immersed in Carnoys fixative for 6 h, followed by 70% ethanol, which was changed once after 24 h. Jejunal samples were embedded in paraffin wax blocks, sectioned (4 µm), and 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), followed by the staining method described previously for determination of the cellular proliferation index (Reynolds et al., 1992
). Cellular proliferation was quantified in 20 fields per ewe by using Image Pro Plus 5.0 analysis software (MediaCybernetics Inc., Silver Spring, MD).
Small Intestine Vascularity.
Cross-sections of perfused intestinal tissue were processed as described above, and 4-µm tissue sections were stained by using periodic acid-Schiff staining procedures (Luna, 1968
) to contrast the vascular tissue. Capillaries were circumscribed and mean capillary area, capillary number, and capillary circumference measurements, along with area of tissue analyzed (Reed et al., 2007
), were made in the intestinal mucosa in 6 fields per ewe with Image Pro Plus 5.0 analysis software (MediaCybernetics Inc.).
Calculations
Empty BW was calculated as BW minus total digesta weight. Maternal BW was calculated as empty BW minus gravid uterine weight. To express organ mass on a maternal BW-specific basis, fresh organ mass (g) was divided by maternal BW (kg).
Total DNA, RNA, and protein contents were calculated by multiplying DNA, RNA, and protein concentrations by fresh tissue weights (Swanson et al., 2000
). Percentage of proliferating cells was estimated by dividing the number of BrdU-stained nuclei by the total number of nuclei present within the area of tissue analyzed. The number of proliferating cells was calculated by dividing total tissue DNA (mg) by 6.6 x 10–12 g (the average amount of DNA per nucleus; Baserga, 1985
), then multiplying that value by the percentage of cell proliferation (Zheng et al., 1994
).
Capillary area density was determined by dividing the total capillary area (µm2) by the area of tissue analyzed (µm2) and multiplying by 100 to express vascularity as a percentage (Scheaffer et al., 2004a
; Soto-Navarro et al., 2004
). Capillary number density was calculated by dividing the total number of vessels counted by tissue area (µm2) and then multiplying by 1,000,000 to express the data as capillaries per millimeter squared. To estimate the capillary surface density (total capillary circumference per unit of tissue area), mean capillary perimeter (circumference; µm) was divided by tissue area (µm2). Although capillary surface density actually represents the circumference of the capillary cross-sections, it is nevertheless proportional to their surface area (Borowicz et al., 2007
). Finally, area per capillary was determined by dividing the total capillary area by the capillary number and expressing the area per capillary (µm2). Total vascularity (mL) was calculated by multiplying the percentage of capillary area density by tissue mass (assuming a constant density).
Statistics
Data were analyzed as a completely randomized design with a 2 x 3 factorial arrangement of treatments by using PROC GLM (SAS Inst. Inc., Cary, NC). The model contained effects for diet (Control vs. High), stage of gestation (50, 90, and 130 d), and diet x stage interactions. When interactions were present (P < 0.10), means were separated by LSD and the most conservative SEM was reported. Main effects were considered significant when P
0.10.
| RESULTS |
|---|
|
|
|---|
In the ewes that conceived after embryo transfer, initial ewe BW and BCS were not affected by diet or stage of gestation. Diet x stage of gestation interactions (P < 0.10) were present for final BW, ADG, empty BW, maternal BW, gravid uterus, fetal BW, final BCS, and BSC change. Therefore, interaction means are presented in Table 1
. Final ewe BW, BCS, and BCS change at necropsy were greater (P = 0.01) in High-fed ewes compared with Control-fed ewes at d 50, 90, and 130 of gestation. In Control-fed ewes, final BW at necropsy was greater (P < 0.10) at d 130 compared with d 50 and 90 of gestation, which did not differ. In ewes fed the High diets, final BW was least (P
0.01) at d 50, intermediate at d 90, and greatest at d 130 of gestation.
|
0.01) at d 50, intermediate at d 90, and greatest at d 130.
Consistent with our experimental design, ewe ADG was greater (P = 0.01) in High-fed ewes compared with Control-fed ewes at each stage of gestation measured. Within Control-fed ewes, ADG was least (P
0.01) at d 50 (22.9 g), intermediate at d 90 (62.9 g), and greatest at d 130 (110.9 g). In High-fed ewes, ADG was similar among days of gestation and averaged 292.5 ± 11.2 g. Empty BW (final BW – digesta) and maternal BW (empty BW – gravid uterine weight) were less (P = 0.01) in Control-fed ewes compared with High-fed ewes at d 50, 90, and 130 of gestation. Gravid uterine weight was similar in Control- and High-fed ewes at d 50 and 90 of gestation; however, at d 130 of gestation, ewes fed the High diet had reduced (P < 0.10) gravid uterine weights compared with ewes fed the Control diet. Fetal BW was not altered by diet at d 50 and 90 of gestation and was less (P = 0.01) in High-fed ewes compared with Control-fed ewes at d 130.
Nongastrointestinal Tissues
Because of diet x stage of gestation interactions (P < 0.08) for liver, spleen, pancreas, blood, and internal fat masses, interaction means are presented in Table 2
. Liver mass (g) was greater (P = 0.01) in ewes fed High compared with Control diets. Liver mass (g) was greater (P = 0.01) at d 130 compared with d 50 and 90 of gestation. When expressed as grams per kilogram of maternal BW, liver mass was greater in Control-fed ewes at d 130 compared with d 50 of gestation. Conversely, in ewes fed High diets, liver mass (g/kg of maternal BW) was less (P
0.10) at d 130 compared with d 50. This response reflects a proportionally more rapid increase in maternal BW than in maternal liver from d 50 to 130 of gestation.
|
0.07). As expected, perirenal fat mass (g) was greater (P = 0.01) in ewes fed High diets compared with ewes fed Control diets at all stages of gestation measured. In Control-fed ewes, perirenal fat mass (g) was similar at d 50, 90, and 130 of gestation; however, in High-fed ewes, perirenal fat was greater (P
0.10) at d 90 and 130 compared with d 50 of gestation. When expressed as grams per kilogram of maternal BW, perirenal fat was greater (P
0.10) in High-fed ewes compared with Control-fed ewes at d 130 of gestation. Total internal fat was greater (P
0.10) in High-fed ewes compared with Control-fed ewes at d 90 and 130, but not at d 50 of gestation.
No diet x stage of gestation interactions were present in spleen mass data. Ewes fed High diets had greater (P
0.01) spleen mass (g) compared with Control-fed ewes; however, when expressed on a basis of grams per kilogram of maternal BW, spleen mass was similar in Control- and High-fed ewes. No differences were noted in spleen mass with advancing stage of gestation.
Diet x stage of gestation interactions were present (P = 0.01) in pancreas mass data. Ewes fed High diets had greater (P
0.10) pancreatic mass (g) compared with ewes fed Control diets at d 50, 90, and 130 of gestation. However, when expressed as grams per kilogram of maternal BW, Control- and High-fed ewes had similar pancreatic masses at d 50 of gestation, whereas at d 90 and 130, Control-fed ewes had a greater (P
0.01) proportional pancreatic mass compared with those fed High diets. In Control-fed ewes, pancreas mass (g) was greater (P
0.10) at d 130 compared with d 50 and 90 of gestation. In High-fed ewes, pancreatic mass (g) was not affected by stage of gestation. When pancreatic mass was expressed as grams per kilogram of maternal BW, it was greater (P
0.10) at d 130 compared with d 50 and 90 of gestation in Control-fed ewes and was less (P
0.10) at d 130 compared with d 50 and 90 of gestation in High-fed ewes. As a result, proportional pancreatic mass (g/kg of maternal BW; Table 2
) in High-fed ewes at d 130 was 47% that of Control-fed ewes. In addition, at d 130 of gestation, pancreatic mass expressed both as grams and as grams per kilogram of maternal BW was less (P
0.10) in High-fed ewes compared with Control-fed ewes.
Diet x stage of gestation interactions (P
0.05) were present for blood mass data. Blood mass (g) was greater (P = 0.01) in ewes fed High compared with Control diets at d 50 and 90 of gestation. At d 130, blood mass was similar in Control- and High-fed ewes. In Control-fed ewes, blood mass was greater (P
0.01) at d 130 compared with d 50 and 90 of gestation. Conversely, blood mass was similar in High-fed ewes at all stages of gestation measured. When expressed as grams per kilogram of maternal BW, blood mass was less (P
0.10) in High-fed ewes compared with Control-fed ewes at d 130 of gestation. In Control-fed ewes, blood mass was greater (P
0.10) at d 130 compared with d 50 of gestation, whereas in High-fed ewes, blood mass was less at d 130 compared with d 50 of gestation. These data indicate that in ewe lambs fed High diets, blood mass in proportion to maternal BW declined with advancing stage of gestation. In contrast, in Control-fed ewes, blood mass (g/kg of maternal BW) increased late in gestation. As a result, ewes fed High diets had 70% of the blood mass per unit of maternal BW at d 130 compared with Control-fed ewes.
Gastrointestinal Tissues
Main effects of diet and stage of gestation on gastrointestinal tract (GIT) mass data are presented in Table 3
. When present, the diet x stage of gestation interactions (P
0.10) are presented in Table 3
. Total empty GIT mass (g) was greater (P
0.01) in ewes fed High diets compared with ewes fed Control diets; however, when GIT mass was expressed as grams per kilogram of maternal BW, GIT mass of ewes fed High diets was less (P = 0.01) compared with ewes fed Control diets. Stage of gestation did not alter GIT mass (g); however, advancing stage of gestation did result in less (P = 0.01) proportional GIT mass (g/kg of maternal BW), with d 130 being less than d 50 of gestation.
|
Within the small intestine, duodenal mass (g and g/kg of maternal BW) was not altered by diet; however, duodenal mass was less (P
0.01) at d 90 and 130 compared with d 50. Jejunal mass (g) was not affected by diet; however, when expressed as grams per kilogram of maternal BW, jejunal mass was less (P = 0.01) in ewes fed High diets compared with ewes fed Control diets. As a result, High-fed ewes had 71% of the proportional jejunal mass (g/kg of maternal BW) of Control-fed ewes. Additionally, ewes necropsied at d 130 tended (P = 0.11) to have reduced jejunal mass (g/kg of BW) compared with d 50 of gestation. Diet x stage of gestation interactions (P
0.02) were present in ileal mass when expressed both as grams and as grams per kilogram of maternal BW. Ewes fed High diets had greater (P = 0.10) ileal mass (g) at d 50 and 90 of gestation than ewes fed Control diets. Ileal mass was similar among dietary treatments at d 130 of gestation. When expressed as grams per kilogram of maternal BW, ewes fed High diets had greater (P = 0.01) ileal mass compared with ewes fed Control diets at d 50 of gestation and had reduced (P = 0.01) ileal mass compared with ewes fed Control diets at d 130. Within Control-fed ewes, ileal mass was less (P
0.10) at d 90 compared with d 130 of gestation. Conversely, in High-fed ewes, ileal mass was less (P = 0.01) at d 130 compared with d 50 of gestation.
For total small intestinal mass (duodenal + jejunum + ileum), diet x stage of gestation interactions were present (P
0.08) when data were expressed either as grams or as grams per kilogram of maternal BW. Small intestinal mass (g) was greater (P
0.06) in High-fed ewes compared with Control-fed ewes at d 50 and 90 of gestation and were not different at d 130. When expressed as grams per kilogram of maternal BW, small intestinal mass was less (P = 0.01) in High-fed ewes compared with Control-fed ewes at d 130 and was not influenced by diet at d 50 and 90 of gestation. As a result, at d 130 High-fed ewes had 63% of the proportional small intestinal mass (g/kg of maternal BW) of Control-fed ewes. In Control-fed ewes, small intestinal mass (g) was less (P
0.10) at d 90 compared with d 50 and d 130 of gestation. Conversely, in High-fed ewes, small intestinal mass was less (P
0.10) at d 130 compared with d 50 and 90. When expressed as grams per kilogram of maternal BW, small intestinal mass was less (P
0.10) in Control-fed ewes at d 90 and 130 compared with d 50 of gestation. In High-fed ewes, small intestinal mass (g/kg of maternal BW) was greatest at d 50, intermediate at d 90, and least at d 130 of gestation (P
0.10).
Large intestinal mass (g) was greater (P = 0.01) in High-fed ewes compared with Control-fed ewes and was greater (P = 0.08) at d 90 and 130 compared with d 50 of gestation. When expressed as grams per kilogram of maternal BW, large intestinal mass was less (P = 0.01) in High-fed ewes compared with Control-fed ewes and was not altered (P = 0.60) by stage of gestation.
Jejunal Cellularity
Jejunal DNA concentrations (mg/g) and contents (g) were not affected by diet or stage of gestation (Table 4
). These data indicated that cell numbers were similar in Control- and High-fed ewes and across the stages of gestation measured. In contrast to DNA data, RNA concentrations and contents were greater (P
0.08) in High-fed ewes compared with Control-fed ewes. Protein concentrations were increased (P = 0.07) in High-fed ewes compared with Control-fed ewes. For jejunal cellularity data, only RNA:DNA yielded a diet x stage of gestation interaction (P = 0.03). Therefore, interaction means for RNA:DNA are presented in Table 4
. In ewes fed High diets, RNA:DNA ratios were increased (P
0.01) at d 50 and 90 compared with Control-fed ewes, indicating increased translational capacity and possibly increased cell size. Ratios of RNA:DNA were not affected (P = 0.72) by dietary treatment at d 130 of gestation. Protein:DNA ratios were not affected by diet or stage of gestation.
|
No diet x stage of gestation interactions were noted in jejunal crypt cell proliferation or mucosal vascularity estimates; therefore, the main effects of diet and stage of gestation are presented in Table 5
. Percentage of crypt cell proliferation and total proliferating nuclei were not affected by either diet or stage of gestation (P > 0.49). Capillary area density, capillary number density, capillary surface density, area per capillary, and total jejunal vascularity were not altered (P > 0.30) by dietary treatments. Capillary area density was less at d 90 compared with d 50 and 130 of gestation. Capillary number density was greatest (P
0.10) at d 50, intermediate at d 90, and least at d 130. Capillary surface density and area per capillary were greater (P
0.10) at d 130 compared with d 50 and 90 of gestation (Table 5
and Figure 1
). Total jejunal vascularity was greater (P
0.01) at d 130 compared with d 90, whereas both were less compared with d 50 of gestation.
|
|
| DISCUSSION |
|---|
|
|
|---|
The overfed first-parity ewe lamb model used in this study is unique (Wallace et al., 2004a
,b
, 2006a
; Luther et al., 2005
), robust, and highly relevant for investigating developmental programming (Reynolds et al. 2005a
,b
, 2006
; Wu et al., 2006
). It has been shown that overfeeding singleton-bearing, first-parity ewe lambs results in rapid maternal growth at the expense of the nutrient demands of the gravid uterus (Wallace et al., 1996
, 1999a
,b
, 2001
, 2006b
). This model also results in major placental growth restriction (30 to 40%), leading to the premature delivery of lambs with low BW at birth (25 to 30% reduction in BW at birth; Wallace et al., 2001
) when compared with moderately fed Control ewe lambs. In addition, this model is characterized by a major reduction in colostrum yield and quality at parturition (Wallace et al., 2001
, 2006c
). Therefore, in the growing first-parity ewe lamb, dietary intake and the resulting alteration in nutrient partitioning during gestation have a major influence on pregnancy outcome (Wallace et al., 2001
; Wu et al., 2006
; Caton et al., 2007
); however, mechanisms underlying the responses observed have not been clearly defined.
In the present study, by design, Control-fed ewes performed within expected ranges, with gains of 23, 63, and 111 g/d for d 50, 90, and 130, respectively. Conversely, High-fed ewes gained 299, 291, and 288 g/d for d 50, 90, and 130, respectively. As expected, High-fed ewes were heavier and had greater BCS compared with controls. Fetal BW was less at d 130 in High-fed ewes compared with Control-fed ewes. These maternal and fetal responses are consistent with other published data using this overnutrition model in young, first-parity ewes (Wallace et al., 1999b
, 2002b
) and indicate that model objectives were being met by nutritional treatments.
Visceral Organs and Internal Fat
As expected, ewes fed High diets had greater perirenal and internal fat compared with ewes fed Control diets. This result is consistent with other published data evaluating the effects of high planes of nutrition fed to gestating first-parity ewe lambs (Wallace et al., 1999b
, 2002a
,b
). Likewise, maternal liver mass was greater in ewes fed High diets compared with ewes fed Control diets. These findings agree with the results of others working with sheep (Wester et al., 1995
; Fluharty and McClure, 1997
; Swanson et al., 2000
) who have reported increased liver masses in response to increased dietary intakes. Liver mass increased in both Control- and High-fed ewes as pregnancy advanced. In Control-fed ewes, proportional liver mass was greater at d 130 compared with d 50 of gestation, which is consistent with observations in mature ewes (Scheaffer et al., 2004b
) at d 130 compared with d 90 or nonpregnant ewes. In contrast, ewes in the present study fed High diets had less proportional liver mass at d 130 compared with d 50 of gestation. These data demonstrate that relative liver mass as a proportion of maternal BW was responding differently across treatments as pregnancy advanced, with High-fed ewes presenting declining and Control-fed ewes presenting increasing proportional liver mass. This response was also observed in other tissues, as discussed below, and implies differing patterns of proportional growth in internal organs in relation to whole-animal maternal BW changes. These responses are partially explained by large increases in maternal BW in ewes fed High diets, which are driven partly by increasing body fat (Wallace et al., 1999b
, 2004a
, 2006b
), but also likely by increasing lean tissue growth.
An interesting new finding in this study was that ewes fed High diets had 70% of the proportional blood mass at d 130 of gestation compared with controls. Although total blood mass was not influenced by dietary treatment, this significant proportional decline in blood volume (measured as mass) could have relevant physiological effects during the last one-third of pregnancy, when rapidly growing ewe lambs are experiencing high metabolic demand from both rapid maternal and conceptus growth. In humans, normal pregnancy is associated with a 40% increase in blood volume from early to late gestation (Picciano, 2003
). A similar change in absolute blood mass was recorded here between d 50 and 130 of gestation in pregnancies of Control-fed ewes, whereas blood mass did not change in High-fed ewes across stage of pregnancy. In humans, failure of this normal pregnancy-associated increase in blood volume is associated with a variety of negative outcomes in offspring, including small size for the gestational age, low birth weight, and preterm delivery (Scholl, 2005
), which are mirrored in the current data with High-fed ewe lambs. The results of this study together with an earlier study in nutrient-restricted ewe lambs (Reed et al., 2007
) suggest that dietary intake is a powerful modulator of this normal physiological response to pregnancy.
Pancreatic mass (g) increased in Control-fed ewes from d 50 to 130 of pregnancy, but was similar throughout pregnancy in those fed High diets. In fact, although pancreatic mass in Control-fed ewes increased during pregnancy (54.3 vs. 73.9 ± 3.7 g), in High-fed ewes (69.8 vs. 58.3 ± 3.7 g) it appeared to be in decline (although not significantly). Consequently, when pancreatic mass was expressed in grams per kilogram of maternal BW, differing growth trajectories were readily apparent and were highly significant, with High-fed ewes having only 47% of the proportional pancreatic mass of controls at d 130 of gestation. These large differences in pancreatic mass per unit of maternal BW raise questions regarding both endocrine and exocrine functions of the pancreas during late pregnancy in ewes fed High diets. However, circulating maternal insulin concentrations are elevated from early in gestation in high-intake vs. control-intake ewe lambs, and there is no evidence of reduced insulin concentrations during late gestation in the former group (Wallace et al., 1997
, 1999b
).
The decrease in fetal BW in late gestation observed in this and earlier studies is closely associated with a corresponding reduction in placental mass. Although in sheep the majority of placental growth occurs before d 90 of gestation, the mass of the placenta is not perturbed by maternal overfeeding until sometime during the final one-third of pregnancy (Wallace et al., 2001
, 2006a
). This relatively late-onset placental growth restriction is, however, preceded in midgestation by reduced proliferative activity within the fetal trophectoderm (Lea et al., 2005
) and decreased placental expression of several angiogenic growth factors (Redmer et al., 2005
). Although the precise mechanisms linking maternal nutrition with these changes in placental growth and development are currently being investigated, the data herein suggest that significant whole-animal and metabolic changes are already measurable by as early as d 50 of gestation.
Gastrointestinal Tissues
As indicated previously, maternal visceral tissues use a disproportional amount of energy (and other nutrient resources) in relation to their contribution to overall body mass (Ferrell, 1988
; Reeds et al., 1999
). This fact alone makes visceral tissues an important consideration when evaluating whole-animal responses to dietary treatments. In normally fed ruminants, the liver and gut consume approximately 40% of maintenance energy demands (Huntington et al., 1990
; Reynolds et al., 1991
; Caton et al., 2000
). The GIT consumes approximately 20% of maintenance energy (Webster, 1989
; Eisemann and Nienaber, 1990
) through various processes, including ion transport, protein turnover, enzyme secretion, and active transport (Gregg and Milligan, 1982
; Baldwin, 1995
; Caton et al., 2000
). In addition to the high rate of metabolic activity of the maternal gut, and hence maintenance energy use, it is one of the key nutrient transferring tissues working in concert with the placenta, fetal gut, and mammary gland to provide sustenance to the developing fetus and neonate. Therefore, from a whole-animal standpoint, the maternal gut is a heavy nutrient user as well as an essential nutrient provider and stands at a pivotal crossroads in net nutrient delivery to the developing conceptus and neonate. Consequently, nutrient utilization by the gut is dynamic in terms of blood flow and nutrient flux, being responsive to diet quality (Reynolds et al., 1991
; Seal and Reynolds, 1993
; Goetsch, 1998
) and quantity (Huntington et al., 1990
; Scheaffer et al., 2004b
; Reed et al., 2007
). Data indicate that gut tissues are highly competitive with other tissues for nutrients (MacRae et al., 1997
) and are responsive to the physiological changes associated with lactation (Reynolds et al., 2001
) and pregnancy (Freetly and Ferrell, 1997
, 1998
; Scheaffer et al., 2004b
). In addition, these responses are associated with changes in tissue mass and cellularity (Jin et al., 1994
; Schaeffer et al., 2003, 2004a) and, as recently reported from our laboratory, vascularization (Reed et al., 2007
) and angiogenic factor profiles (Neville et al., 2007
).
In the present study, mass (g) of the GIT was greater in High-fed ewes compared with Control-fed ewes. This response was driven by an increase in total mass of stomach, small intestine, and large intestine in High-fed ewes compared with Control-fed ewes. Conversely, when expressed as grams per kilogram of maternal BW, gastrointestinal mass was less in ewes fed High diets compared with Control diets, indicating asynchronous growth of visceral organs in relation to the rapid maternal BW gains being experienced by High-fed ewes. Thus, although total gut mass, and likely energy use, increased in High-fed ewes, there was less available gut tissue per unit of maternal BW. These changes with advancing gestation likely are not due to changing intake. The gut is very responsive to intake amounts; however, in this study intake in High-fed ewes ranged from 2.0 to 2.1 kg of DM/ewe daily throughout gestation, with no decline by d 130. In contrast, intakes of Control-fed ewes ranged from 0.7 to 0.8 kg of DM/d up to d 90 of gestation, at which time they increased slowly to meet the demands for gravid uterine growth and approached 1.0 kg of DM/ewe daily at d 130 of gestation.
Within the small intestine, total mass and proportional duodenal mass were not affected by treatment and declined with advancing pregnancy. Recent evidence suggests that duodenal tissues of sheep increase in mass in response to elevated estrogen (ONeil et al., 2005
). In addition, ONeil et al. (2006)
reported that estrogen increased jejunal crypt cell proliferation after 24 h of administration. Estrogen receptors are expressed within the intestine (Kawano et al., 2004
) and are functional because they induce physiological changes in the small intestine (Díaz et al., 2004
; Chen et al., 2005
). Circulating estrogen (estradiol) is catabolized by the liver, and in the present study, total liver mass increased with advancing pregnancy. Therefore, declining duodenal mass with advancing pregnancy may be due to reduced estrogen concentrations resulting from hepatic metabolism or reduced placental production resulting from reduced placental mass, or both. Interestingly, total small intestinal mass in High-fed ewes also declined with advancing pregnancy, whereas liver mass increased. We speculate that these changes may be related to estrogen concentrations and hepatic metabolism, placental production, or both. In support, recent assessments of maternal estrogen concentrations in this model suggest a reduction from midpregnancy onward in High vs. Control pregnancies (Wallace et al., 2008
). Additional research in this direction would likely provide unique insight into the relationships of intestinal growth and estrogen metabolism.
Differing growth trajectories were evident for small intestinal tissue, with total mass similar from d 50 to 130 in Control-fed ewes and declining in High-fed ewes. Similar trends were noted when ileal mass was expressed as either grams or grams per kilogram of maternal BW. Proportional small intestinal mass in High-fed ewes at d 130 was 63% of d 130 Control-fed ewes and 58% of d 50 High-fed ewes.
Others have shown with mature (Scheaffer et al., 2004b
) and young, first-parity (Reed et al., 2007
) ewes that nutrient restriction during pregnancy reduces absolute maternal jejunal and total small intestinal mass by 17 and 20%, respectively. When expressing data as grams per kilogram of maternal BW, differences in small intestinal mass are retained in young (Reed et al., 2007
), but not mature (Scheaffer et al., 2004b
), ewes. Taken together, these data indicate that the maternal small intestine is very responsive to either dietary excess or restriction and that responses likely are more prone to be asymmetrical in young ewes than in mature ewes.
Jejunal Cellularity, Crypt Cell Proliferation, and Mucosal Vascularity
Jejunal DNA concentration and contents were unresponsive to either dietary treatment or advancing gestation. These data indicated that cell numbers were similar in Control- and High-fed ewes and across the stages of gestation measured. Others using first-parity ewe lambs have shown that nutrient restriction reduces jejunal DNA content (Reed et al., 2007
). In addition, jejunal DNA concentration has been reported to increase in response to pregnancy in beef heifers (Scheaffer et al., 2003
). In mature ewes, plane of nutrition x stage of gestation interactions were reported in jejunal DNA concentrations, with restricted ewes having less at d 90 and more at d 130 of gestation compared with maintenance-fed ewes. Effects of overfeeding on jejunal DNA concentrations in pregnant ewe lambs have not been reported previously. However, Swanson et al. (2000)
reported that jejunal DNA was not affected by increased dietary intake in growing wethers, and Burrin et al. (1992)
reported that ad libitum intakes in wether lambs decreased jejunal DNA when compared with those fed maintenance amounts of intake.
In contrast to DNA data, jejunal RNA concentrations and contents were greater in High-fed ewes compared with Control-fed ewes, perhaps indicating a greater amount of translational activity, which would seem reasonable with the increased transport demand associated with high intakes. Diet x stage of gestation interactions were noted for RNA:DNA, which resulted from High-fed ewes having greater RNA:DNA at d 50 and 90, but having similar ratios at d 130 compared with Control-fed ewes. In addition, Reed et al. (2007)
reported that maternal RNA:DNA were greater in restricted-fed ewes compared with control-fed ewes, also suggesting a need for greater translation capacity. Other reports from mature ewes (Scheaffer et al., 2004a
) and beef heifers (Scheaffer et al., 2003
) have indicated little impact of advancing gestation on jejunal RNA:DNA.
Jejunal crypt cell proliferation was not affected by either diet or stage of gestation. Work with mature or young, first-parity ewes fed maintenance or restricted diets indicated no differences in jejunal crypt cell proliferation (Scheaffer et al., 2004a
; Reed et al., 2007
). In addition, ad libitum intakes in growing wethers had little impact on jejunal crypt cell proliferation, as measured by proliferating cell nuclear antigen and immunohistochemistry, when compared with controls fed 40% less (Swanson et al., 2000
). In contrast, Scheaffer et al. (2003)
reported that jejunal crypt cell proliferation in pregnant beef heifers was less at d 120 and 200 of gestation compared with nonpregnant controls. Many of the researchers cited above reported changes in jejunal or intestinal mass in response to dietary treatment or stage of pregnancy; however, few reported changes in jejunal crypt cell proliferation. This likely is due to the timing of measurement. Mammalian intestinal epithelial cell turnover usually occurs every 24 to 96 h (Goke and Podolsky, 1996
), and most treatments discussed above lasted several weeks during gestation before single assessments of jejunal crypt cell proliferation. Therefore, it is likely that changes in proliferation rate that resulted in mass changes were transitory until steady state was reached, at which time proliferation rates stabilized and differences became more difficult to detect.
Measures of jejunal vascularity were not affected by dietary treatment. Nutrient restriction has been shown to increase percentage of jejunal vascularity in mature pregnant ewes (Scheaffer et al., 2004a
) and to decrease it in first-parity ewe lambs (Reed et al., 2007
) at d 130 of gestation. In the present study, capillary area density and total jejunal vascularity declined from d 50 to 90 and then increased from d 90 to 130 of pregnancy. Reasons for reduced jejunal microvascular volume at d 90 compared with other times measured are unclear, but may be related to pregnant ewes transitioning from an anabolic to a catabolic state with advancing pregnancy (Stock and Metcalfe, 1994
). Scheaffer et al. (2004a)
similarly reported increased jejunal vascularity from d 90 to 130 in mature ewes, with responses being more pronounced in restricted-fed compared with control-fed ewes. It appears from these 2 studies that increased jejunal vascularity during the last one-third of pregnancy in ewes may partially underlie how maternal intestinal tissues cope with the rapidly increasing nutrient demands of the conceptus associated with the last one-third of gestation.
In the present study, capillary number density declined with advancing stage of pregnancy, whereas capillary surface density and area per capillary increased. These data indicate that jejunal vascularity was altered with advancing pregnancy and resulted from a smaller number of vessels, but larger ones, that had increased the surface area. These responses likely are due to increasing nutrient demand and concomitant changes in whole-organ blood flow; however, mechanisms underlying these vascular responses are unknown. Insight into the underlying mechanisms may be gleaned from recent data indicating that nutrient restriction in pregnant ewe lambs upregulated mRNA expression of several vascular growth factors and receptors (including the vascular endothelial growth factor system) in maternal jejunal mucosal scrapes (Neville et al., 2007
) when compared with controls.
In summary, overfeeding first-parity ewe lambs increased maternal BW, BCS, ADG, and visceral adiposity. In support of our hypothesis, changes in gastrointestinal and other visceral tissues were evident in response to overfeeding and stage of gestation. Although these responses may partially explain previously observed changes in nutrient partitioning to the uterus and mammary gland and resulting reductions in BW at birth, other underlying mechanisms are likely present. Pregnant ewe lambs consuming High intakes had proportionally less pancreatic and intestinal mass and blood volume when compared with Control-fed ewes. Although the precise functional relevance of these changes remains to be determined, overfeeding was associated with fetal growth restriction in late gestation, as reported previously. In this study, jejunal cellularity, crypt cell proliferation, and vascularity measures were largely unresponsive to dietary excess. Measures of jejunal mucosal vascularity were responsive to advancing stage of gestation, indicating intestinal vascular plasticity in response to physiological processes associated with advancing pregnancy.
| Footnotes |
|---|
2 Corresponding author: Joel.Caton{at}ndsu.edu
Received for publication March 17, 2008. Accepted for publication August 26, 2008.
| LITERATURE CITED |
|---|
|
|
|---|
- and β-positive cells in adult male and female mouse intestine at various estrogen levels. Histochem. Cell Biol. 121:399–405.[CrossRef][Medline]
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |