ABSTRACT
Kraft pulp (KP), an intermediate product obtained when wood chips are converted to paper, contains highly digestible fiber. This study evaluated the effect of KP inclusion in calf starters on growth performance, health, and plasma glucagon-like peptide 2 (GLP-2) concentration in calves. Twenty-five Holstein heifer calves were raised on a high plane of nutrition program using milk replacer containing 29% crude protein and 18% fat until 49 d after birth, and were fed calf starters containing KP at 0 (CON; n = 14) or 12% (KPS; n = 11) on a dry matter basis. All calves were fed the treatment calf starters and timothy hay ad libitum. Blood was collected at 4, 14, 21, 35, 49, 70, and 91 d after birth. Dry matter intake (DMI) of milk replacer and hay was not affected by treatment, whereas calf starter DMI was lower for KPS (0.93 kg/d) than for CON (1.03 kg/d). Higher neutral detergent fiber (NDF) content in KPS (31.7%) than in the CON starter (22.1%) resulted in higher NDF intake for KPS (0.55 kg/d) than for CON (0.47 kg/d). However, the consumption of starch was lower for KPS (0.29 kg/d) than for CON (0.33 kg/d). Despite the lower starter intake for KPS, body weight and average daily gain did not differ between treatments. No significant difference was observed in the plasma concentrations of metabolites, except for ß-hydroxybutyrate (BHB); BHB concentration was lower for KPS (216 µmol/L) than for CON (257 µmol/L). The area under the curve for plasma GLP-2 concentration was higher for KPS (54.1 ng/mL × d) than for CON (36.0 ng/mL × d). Additionally, the fecal score postweaning (1.19 and 1.48 for KPS and CON, respectively) and the number of days that calves developed diarrhea throughout the experimental period (2.50 d and 8.10 d for KPS and CON, respectively) were lower for KPS than for CON. These results indicate that feeding KP reduces the severity and frequency of diarrhea without adversely affecting growth performance. This could be attributed to the increased plasma GLP-2 concentration induced by higher NDF intake.
Subject(s)
Diet , Glucagon-Like Peptide 2 , Animals , Cattle , Female , Weaning , Diet/veterinary , Animal Feed/analysis , Body Weight , Diarrhea/veterinary , 3-Hydroxybutyric AcidABSTRACT
In our previous studies, we revealed the effect of lactose inclusion in calf starters on the growth performance and gut development of calves. We conducted the present study as a follow-up study to identify the shift in rumen microbiota and its relation to rumen fermentation when calves are fed a lactose-containing starter. Thirty Holstein bull calves were divided into 2 calf starter treatment groups: texturized calf starter (i.e., control; n = 15) or calf starter in which starch was replaced with lactose at 10% (i.e., LAC10; n = 15) on a dry matter basis. All calves were fed their respective treatment calf starter ad libitum from d 7, and kleingrass hay from d 35. Rumen digesta were collected on d 80 (i.e., 3 wk after weaning) and used to analyze rumen microbiota and fermentation products. There was no apparent effect of lactose feeding on the α-diversity and overall composition of rumen microbiota. Amplicon sequencing and real-time PCR quantification of the 16S rRNA gene confirmed that the abundance of butyrate-producing bacteria (i.e., Butyrivibrio group and Megasphaera elsdenii) did not differ between the control and LAC10 groups. Conversely, the relative abundance of Mitsuokella spp., which produce lactate, succinate, and acetate, was significantly higher in the rumen of calves that were fed lactose, whereas the lactate concentration did not differ between the control and LAC10 groups. These findings suggest that the lactate production can be elevated by an increase of Mitsuokella spp. and then converted into butyrate, not propionate, since the proportion of propionate was lower in lactose-fed calves. In addition, we observed a higher abundance of Coriobacteriaceae and Pseudoramibacter-Eubacterium in the LAC10 group. Both these bacterial taxa include acetate-producing bacteria, and a positive correlation between the acetate-to-propionate ratio and the abundance of Pseudoramibacter-Eubacterium was observed. Therefore, the higher abundance of Coriobacteriaceae, Mitsuokella spp., and Pseudoramibacter-Eubacterium in the rumen of lactose-fed calves partially explains the increase in the proportion of rumen acetate that was observed in our previous study.
Subject(s)
Microbiota , Rumen , Animal Feed/analysis , Animals , Body Weight , Cattle , Diet/veterinary , Fermentation , Follow-Up Studies , Lactose/metabolism , Male , RNA, Ribosomal, 16S/genetics , RNA, Ribosomal, 16S/metabolism , Rumen/metabolism , WeaningABSTRACT
The objective of this study was to evaluate the effects of the starch content of pre- and postpartum diets on productivity, plasma energy metabolites, and serum markers of inflammation of dairy cows during the calving transition period. Eighty-eight primiparous and multiparous cows were randomly assigned to pre- and postpartum dietary treatments balanced for parity and pretrial body condition score at d 28 ± 3 before expected calving date. Cows were fed either a control [Control; 14.0% starch, dry matter (DM) basis] or high-starch (High; 26.1% starch, DM basis) prepartum diet commencing 28 ± 3 d before expected calving date. Following calving, cows were fed either a high-fiber (HF; 33.8% neutral detergent fiber, 25.1% starch, DM basis) or high-starch (HS; 27.2% neutral detergent fiber, 32.8% starch, DM basis) postpartum diet for the first 20 ± 2 d following calving. Cows fed the High prepartum diet had greater DM intake (12.4 vs. 10.2 kg/d), plasma concentrations of insulin (1.72 vs. 14.2 ng/mL), glucose (68.1 vs. 65.0 mg/dL), and glucagon-like peptide-2 (0.41 vs. 0.32 ng/mL) before parturition, but increased plasma free fatty acid concentration (452 vs. 363 µEq/L) and milk fat yield (1.64 vs. 1.48 kg/d) after parturition. Cows fed the HS postpartum diet had lower plasma free fatty acid (372 vs. 442 µEq/L) and serum haptoglobin (0.46 vs. 0.70 mg/mL) concentrations over a 3-wk period after calving. In addition, there was a tendency for interaction between prepartum and postpartum diets for milk yield, where feeding the HS postpartum diet increased milk yield compared with the HF diet for cows fed the Control prepartum diet (40.8 vs. 37.9 kg/d) but not for cows fed the High prepartum diet. These results suggest that management efforts to minimize the change in diet fermentability during the calving transition by feeding the High prepartum diet, the HF postpartum diet, or both did not increase productivity of dairy cows but increased fat mobilization after calving. Our findings also suggest that feeding high-starch postpartum diets can decrease fat mobilization and serum indicators of systemic inflammation and increase milk production even with the transition from a low-starch prepartum diet.
Subject(s)
Cattle Diseases , Lactation , Animals , Cattle , Diet/veterinary , Female , Inflammation/veterinary , Milk , Plasma , Postpartum Period , Pregnancy , StarchABSTRACT
In this study, we investigated the effects of feeding a moderate- or high-energy close-up diet to close-up cows on response of newborn calves to intravenously (i.v.) injected glucagon-like peptide 1 (GLP-1). Newborn Holstein heifer calves (n = 37) from cows fed with a moderate-energy [M, 1.54 Mcal/kg of dry matter (DM) NEl; 14% starch; n = 17] or high-energy (H, 1.63 Mcal/kg of DM NEl; 26% starch; n = 20) diet in the last 28 d prepartum were assigned to one of two treatment groups, which were i.v. injected with saline (MC and HC, n = 9 and 10, respectively) or GLP-1 solution at 1.0 µg/kg BW (MG and HG, n = 8 and 10, respectively) immediately after milk replacer (MR; 26% CP, 16% fat) feeding. Blood samples were obtained through a jugular vein catheter at -10, 0, 10, 20, 30, 40, 50, 60, 90, and 120 min relative to MR feeding at 2, 10, and 20 d after birth, and plasma glucose, insulin, and GLP-1 concentrations were measured. Plasma GLP-1 concentration tended to increase starting from 30 min after MR feeding in the MC relative to the HC group at 10 (0.77 ng/mL vs 0.69 ng/mL for MC and HC, respectively; P = 0.10) and 20 d after birth (0.47 ng/mL vs 0.35 ng/mL for MC and HC, respectively; P = 0.07). Plasma glucose and insulin concentrations after MR feeding did not differ between MC and HC groups at 2 and 20 d after birth but were higher (P < 0.05) in MC (158 mg/dL and 3.64 ng/mL for glucose and insulin, respectively) than in HC (143 mg/dL and 2.46 ng/mL for glucose and insulin, respectively) calves at 10 d after birth. The elevation in plasma glucose concentration after MR feeding was suppressed by direct glucose-lowering action of i.v. injected GLP-1 at 2, 10, and 20 d after birth in M and H calves; this direct glucose-lowering action by GLP-1 was greater (P < 0.05) for H than for M calves at 20 d after birth. These results indicate that feeding a high-energy close-up diet to cows affects glucose status in their female offspring via suppression of postprandial plasma concentrations of GLP-1 and insulin as well as the alteration in the glucose-lowering action of GLP-1 after feeding depending on the day after birth.
Subject(s)
Animal Feed/analysis , Diet/veterinary , Energy Intake , Glucagon-Like Peptide 1/pharmacology , Prenatal Nutritional Physiological Phenomena , Animal Nutritional Physiological Phenomena , Animals , Animals, Newborn , Blood Glucose , Cattle , Female , Glucagon-Like Peptide 1/administration & dosage , Glucagon-Like Peptide 1/blood , Insulin/blood , Milk Substitutes , PregnancyABSTRACT
The objective of this study was to evaluate the effects of starch concentration of close-up diets on plasma concentrations of energy metabolites and rumen pH of dairy cows after calving. Eighteen multiparous Holstein dairy cows (mean parity = 2.78; mean body weight = 708 kg; mean body condition score = 3.08) fitted with ruminal cannulas were assigned to treatment balanced for parity, body condition score, and expected calving date. Cows were enrolled in the study at d 28 ± 3 before the expected calving date and fed either a low-starch (LS; 14.0% starch) or high-starch (HS; 26.1% starch) diet until parturition. All cows were fed a common diet after calving (25.1% starch). A grain challenge was performed on d 7 ± 2 and 21 ± 2 after calving by dosing 6.35 kg (dry matter basis) of finely ground barley and wheat grain (1:1) into the rumen via cannula. Feeding the HS diet before calving increased the duration (369 vs. 49 min/d) and area of pH below 5.8 (85.1 vs. 5.2 pH × min/d) during d -10 to -8. In addition, even though all cows were fed a common diet after calving, HS cows tended to have longer duration (177 vs. 76 min/6 h) and greater area of pH below 5.8 (67.8 vs. 20.3 pH × min/6 h) during a grain challenge on d 7. Plasma concentration of insulin tended to be greater in cows fed the HS diet (1.40 vs. 1.09 ng/mL), whereas plasma free fatty acid concentration was not different between treatments during the grain challenge on d 7. During the grain challenge on d 21, neither rumen pH nor blood metabolites were different between the HS and LS cows. These findings suggested that feeding an HS diet during the close-up period does not mitigate rumen pH depression but may exacerbate it after calving compared with feeding an LS diet.
Subject(s)
Cattle/metabolism , Diet/veterinary , Edible Grain , Energy Metabolism/physiology , Rumen/chemistry , Starch/administration & dosage , Animals , Body Weight , Edible Grain/metabolism , Female , Hordeum/metabolism , Hydrogen-Ion Concentration , Insulin/blood , Lactation/physiology , Parity , Parturition/metabolism , Pregnancy , Rumen/metabolism , Triticum/metabolismABSTRACT
This study evaluated how feeding colostrum- or a colostrum-milk mixture for 3 d postnatal affects plasma glucagon-like peptide-2 (GLP-2), serum insulin-like growth factor-1 (IGF-1), and small intestinal histomorphology in calves. Holstein bulls (n = 24) were fed colostrum at 2 h postnatal and randomly assigned to receive either colostrum (COL), whole milk (WM), or a 1:1 COL:WM mixture (MIX) every 12 h from 12 to 72 h. A jugular venous catheter was placed at 1 h postnatal to sample blood frequently for the duration of the experiment. Samples were collected at 1, 2, 3, 6, 9, 11, and 12 h. Following the 12-h meal, blood was collected at half-hour intervals until 16 h and then at 1-h intervals from 16 to 24 h. A 27-h sample was taken, then blood was sampled every 6 h from 30 to 60 h. Again, blood was taken at half-intervals from 60 to 64 h, then at 65 and 66 h, following which, a 2-h sampling interval was used until 72 h. Plasma GLP-2 (all time points) and serum IGF-1 (at time points: 1, 6, 12, 18, 24, 36, 48, and 72 h) were both analyzed. Duodenal, jejunal, and ileal tissues were collected at 75 h of age to assess histomorphology and cellular proliferation. Feeding COL, rather than WM, increased plasma GLP-2 by 60% for 2 h and tended to increase GLP-2 by 49.4% for 4 h after the 60-h meal. Insulin-like growth factor-1 area under the curve (from 12 to 72 h) tended to be 27% greater for COL than WM calves but was otherwise unaffected by treatment. Ileal crypts tended to proliferate more with MIX than WM, whereas ileal crypt proliferation did not differ for COL compared with MIX or WM and was not different between treatments in the proximal jejunum. Villi height was increased 1.8 and 1.5× (COL and MIX vs. WM) in the proximal and distal jejunum, respectively, whereas MIX duodenal and ileal villi height tended to be 1.5 and 1.4× that of WM. Crypt depth did not differ in any region. Surface area of the gastrointestinal tract was reduced for WM by 60 and 58% (proximal jejunum) and 38 and 52% (ileum) relative to COL and MIX and was 54% less than MIX in the distal jejunum. Overall, extended COL feeding minimally increased plasma GLP-2 and serum IGF-1 compared with WM feeding. As COL and MIX similarly promoted small intestinal maturation, feeding calves transition milk to promote intestinal development could be a strategy for producers.
Subject(s)
Animal Feed , Cattle , Colostrum , Glucagon-Like Peptide 2/blood , Insulin-Like Growth Factor I/metabolism , Intestine, Small/growth & development , Milk , Animals , Animals, Newborn , Cattle/blood , Diet/veterinary , Ileum/growth & development , Ileum/metabolism , Intestinal Mucosa/metabolism , Jejunum/growth & development , Jejunum/metabolism , MaleABSTRACT
The objective of this study was to evaluate the effect of feeding a prepartum diet with a high or moderate starch content on growth and insulin sensitivity of female offspring early in life. Thirty-eight Holstein heifer calves were born to dams fed either a high-starch (26% starch on a DM basis, HI; n = 20) or moderate-starch (14% starch on a DM basis, MOD; n = 18) prepartum diet commencing at 28 ± 3 d before expected parturition date. Following birth, all calves were housed individually and fed three 2-L meals of colostrum within the first 24 h of life and offered 10 L/d of milk replacer (26% CP, 18% fat, mixed to 130 g/L). Body weight of calves was measured at birth and on d 2 (after colostrum feeding but before milk feeding), 10 ± 2, and 20 ± 2. A glucose tolerance test was performed at a minimum of 6 h after their last colostrum or milk meal to evaluate insulin sensitivity on d 2, 10 ± 2 and 20 ± 2. Body weight did not differ throughout between HI and MOD calves; however, calves born to primiparous dams were smaller compared with those born to multiparous dams. Glucose or insulin concentrations were not different before the glucose tolerance test. Following the glucose tolerance test, maximum glucose concentrations were not different between treatments at any time point. However, HI calves had greater insulin area under the curve, and HI calves had greater maximum insulin concentrations on d 2. Glucose or insulin clearance rates were not different nor was the calculated insulin sensitivity index between treatments. These findings suggest that feeding a HI prepartum diet may reduce some insulin sensitivity indicators of female offspring early in life.
Subject(s)
Animal Feed/analysis , Cattle/physiology , Colostrum/metabolism , Starch/administration & dosage , Animals , Animals, Newborn , Blood Glucose/analysis , Body Weight , Cattle/growth & development , Dairying , Diet/veterinary , Female , Glucose Tolerance Test/veterinary , Insulin/blood , Insulin Resistance , Milk/metabolism , Milk Substitutes/metabolism , Parturition , PregnancyABSTRACT
Glucagon-like peptide-1 (GLP-1) plays a role in the regulation of glucose homeostasis via the stimulation of insulin secretion. The objective of this study was to evaluate the effect of extended colostrum feeding on plasma concentration of GLP-1. Holstein bull calves (n = 27) were fed pooled colostrum at 7.5% of birth body weight at 2 h after birth and then fed mature milk (M), a 50:50 mixture of pooled colostrum and milk (CM), or pooled colostrum (C; n = 9 for each treatment) at 5% of birth body weight at 12 h after birth and every 12 h thereafter until 72 h after birth. Blood samples were obtained before (1 and 2 h after birth) and after (until 72 h after birth; 42 time points) the first colostrum feeding, and plasma concentrations of glucose, insulin, and GLP-1 were measured. Data were analyzed by ANOVA of JMP 13 (SAS Institute Inc., Cary, NC) with treatment, time, and treatment × time interaction as fixed effects. Treatment × time interaction was observed for plasma insulin and glucose concentrations, which were mainly the result of lower concentrations from 1 to 2 d after birth for C compared with M. Conversely, on d 3 after birth, the difference between treatments was not observed for insulin and glucose. For the entire experimental period, plasma GLP-1 concentration was higher for C (2.25 ng/mL) compared with M (1.41 ng/mL) and tended to be higher compared with CM (1.58 ng/mL). A treatment × time interaction was observed for GLP-1, but unlike glucose and insulin, this was mainly the result of higher concentrations from 54 to 72 h after birth (on d 3 after birth) for C compared with M or CM. Postprandial plasma concentration of glucose was not correlated with that of GLP-1 but was positively correlated with that of insulin for the 4-h period after feeding on d 1 (r = 0.30) and d 3 after birth (r = 0.33). Postprandial plasma concentration of GLP-1 was positively correlated with that of insulin for the 4-h period after feeding on d 3 after birth (r = 0.20). These results indicate that extended colostrum feeding may increase plasma GLP-1 concentrations, especially 3 d after birth, but further study is necessary to determine the effect on plasma insulin and glucose concentrations.
Subject(s)
Blood Glucose , Colostrum/physiology , Glucagon-Like Peptide 1/blood , Insulin/blood , Milk/physiology , Animals , Animals, Newborn , Cattle , Diet/veterinary , Female , Male , Parturition , Postprandial Period , PregnancyABSTRACT
Glucagon-like peptide (GLP)-1 is involved in glucose homeostasis via its role in stimulating insulin secretion, whereas GLP-2 increases mucosal growth of the small intestine. To our knowledge, the effect of delayed colostrum feeding on plasma GLP-1 and GLP-2 in neonatal calves has not been evaluated. To investigate the effect of delayed colostrum feeding on plasma concentrations of GLP-1 and GLP-2 in newborn calves, we randomly assigned 27 Holstein bull calves to 1 of 3 treatment groups: those fed colostrum within 1 h after birth (control), 6 h after birth (6H), and 12 h after birth (12H; n = 9 for each treatment). Blood samples were obtained before the colostrum feeding and every 3 h after each colostrum feeding for a 36-h period, and plasma concentrations of GLP-1, GLP-2, insulin, and glucose were measured. Plasma GLP-1 concentration at 12 h after colostrum feeding was lower in 12H than in control calves. In addition, plasma insulin concentration was lower in the 6H and 12H calves than in the controls. Plasma glucose and GLP-2 concentrations were, however, not affected by treatment. These results indicate that delayed colostrum feeding can decrease plasma GLP-1 and insulin concentrations without affecting glucose or GLP-2 concentration.
Subject(s)
Animals, Newborn/blood , Colostrum/physiology , Glucagon-Like Peptide 1/blood , Glucagon-Like Peptide 2/blood , Animals , Cattle , Diet , Female , Insulin , Male , PregnancyABSTRACT
The objective of this study was to evaluate the effects of lactose inclusion in calf starters on plasma glucagon-like peptide (GLP)-1 and GLP-2 concentrations and gastrointestinal tract development in calves. Holstein bull calves (n = 45) were raised on an intensified nursing program using milk replacer containing 28.0% CP and 15.0% fat, and were fed a texturized calf starter containing 0 (control), 5.0 (LAC5), or 10.0% (LAC10; n = 15 for each treatment) lactose on a DM basis. Lactose was included in the starter by partially replacing dry ground corn in pelleted portion of the starter. All calf starters were formulated with 23.1% CP. The ethanol-soluble carbohydrate concentrations of the control, LAC5, and LAC10 starters were 7.3, 12.3, and 16.8% on a DM basis, respectively. Starch concentrations of the control, LAC5, and LAC10 starters were 29.7, 27.0, and 21.4% on a DM basis, respectively. All calves were fed treatment calf starters ad libitum. Blood samples were obtained weekly from 1 to 11 wk of age, and used to measure plasma GLP-1, GLP-2, and insulin concentrations, serum ß-hydroxybutyrate (BHB) concentration, and blood glucose concentration. At 80 d of age, calves were euthanized, and weights of the reticulorumen, omasum, abomasum, small intestine, and large intestine tissue were measured. Serum BHB concentration was higher for calves fed the LAC10 (171 µmol/L) starter than for those fed the control (151 µmol/L) and LAC5 (145 µmol/L) starters. Plasma GLP-1 and GLP-2 concentrations did not differ between treatments. However, relative to the baseline (1 wk of age), the plasma GLP-1 concentration was higher for the LAC10 (125.9%) than for the LAC5 (68.2%) and control (36.8%), and for the LAC5 than for the control (36.8%). Moreover, similar differences between treatments were observed for GLP-2 concentration relative to the baseline (88.2, 76.9, and 74.9% for LAC10, LAC5, and control treatments, respectively). The serum BHB concentration was positively correlated with the plasma GLP-1 concentration (r = 0.428). Furthermore, the plasma GLP-1 concentration was positively correlated with the insulin concentration (r = 0.793). The weights of the reticulorumen, omasum, abomasum, small intestine, and large intestine were not affected by the treatments. In conclusion, inclusion of lactose in calf starters resulted in higher plasma GLP-1 and GLP-2 concentrations, and BHB might be associated with higher plasma GLP-1 concentration.
Subject(s)
Animal Feed/analysis , Glucagon-Like Peptide 1/blood , Glucagon-Like Peptide 2/blood , Lactose/pharmacology , 3-Hydroxybutyric Acid/blood , Animals , Cattle , Diet/veterinary , Insulin/blood , Lactose/chemistry , MaleABSTRACT
The objective of this study was to evaluate effects of partial replacement of dry ground corn with lactose in calf starters on dry matter intake, growth rate, ruminal pH, and volatile fatty acid profile. Sixty Holstein bull calves were raised on a high plane of nutrition program until 55 d of age. Calves were fed texturized calf starters containing 30.1% steam-flaked grains and lactose at 0 (control), 5, or 10% (n = 20 for each treatment) on a dry matter basis. All calves were fed treatment calf starters ad libitum from d 7 and kleingrass hay from d 35. Ruminal pH was measured continuously immediately after weaning (d 55-62) for 15 calves (n = 5 per treatment), and 3 wk after weaning (d 77 to 80) for the other 45 calves (n = 15 per treatment). Dry matter intake, growth performance, and ruminal pH variables were not affected by treatment. However, according to Spearman's correlation coefficient (rs) analyses, lactose intake was positively correlated with dairy minimum ruminal pH (rs = 0.306) for the data collected from d 77 to 80. Similarly, hay intake was not affected by treatment, but positively correlated with daily mean (rs = 0.338) and maximum ruminal pH (rs = 0.408) and negatively correlated with duration pH <5.8 (rs = -0.329) and area pH <5.8 (rs = -0.325), indicating that the variation in hay intake among animals might have masked treatment effects on ruminal pH. Ruminal molar ratio of acetate was higher (45.2 vs. 40.6%), and that of propionate was lower in 10% lactose than control (35.3 vs. 40.2%) for ruminal fluid collected on d 80; however, molar ratio of butyrate was not affected by treatment. These results indicate that lactose inclusion in calf starters up to 10% of dry matter might not affect dry matter intake and growth performance of calves, but that greater lactose and hay intake might be associated with higher ruminal pH.
Subject(s)
Animals, Newborn/growth & development , Cattle , Fermentation , Rumen/metabolism , Animal Feed , Animal Nutritional Physiological Phenomena , Animals , Cattle/growth & development , Cattle/metabolism , Diet , Lactose , Male , Zea maysABSTRACT
Glucagon-like peptide-2 (GLP-2), a gut peptide secreted by enteroendocrine L cells, has recently been identified as a key regulator of intestinal growth and absorptive function in ruminants. However, reports on GLP-2 secretion are few, and more information regarding its secretion dynamics is needed. In this study, two experiments were conducted to elucidate the daily rhythm of GLP-2 secretion in response to feeding regimen and to investigate the effect of volatile fatty acids (VFA) on GLP-2 release in sheep. In experiment 1, blood samples were collected over 3 d from 4 Suffolk mature wethers adapted to a maintenance diet fed once daily; day 1 sampling was preceded by 24 h of fasting to reach steady state. On days 1 and 3, samples were collected every 10 min from 11:00 to 14:00 on both days and then every 1 h until 00:00 on day 1 only; feed was offered at 12:00. On day 2, feed was withheld, and sampling was performed every hour from 01:00 to 00:00. In experiment 2, 5 Suffolk mature wethers were assigned to 5 treatment groups of intraruminal administration of saline, acetate, propionate, butyrate, or VFA mix (acetate, propionate, and butyrate in a ratio of 65:20:15) in a 5 × 5 Latin square design. Blood samples were collected at 0, 1.5, 3, 6, 9, 12, 15, 20, 25, 30, 40, 50, 60, 90, and 120 min relative to the beginning of administration at 12:00. In both experiments, plasma GLP-2, glucagon-like peptide-1 (GLP-1), glucose, insulin, and ß-hydroxy butyric acid (BHBA) levels were measured. In experiment 1, incremental area under the curve was greater (P < 0.05) post-feeding than pre-feeding on days 1 and 3 for GLP-2 and tended to be greater (P < 0.1) on day 1 for GLP-1. Plasma insulin, glucose, and BHBA levels increased (P < 0.05) on day 1 post-feeding. Plasma GLP-2 was poorly correlated with GLP-1 but positively correlated with insulin, glucose, and BHBA. In experiment 2, administration of butyrate and VFA mix remarkably increased plasma GLP-2 (P = 0.05) and BHBA (P < 0.0001) levels compared with those in other treatments. Plasma GLP-1 levels were higher with butyrate administration compared with those in the saline, acetate, and VFA mix (P = 0.019). Propionate administration increased plasma glucose (P = 0.013) and insulin (P = 0.053) levels. Thus, our data confirmed that GLP-2 release is responsive to feeding and might be promoted by BHBA produced by the rumen epithelial metabolism of butyrate. Further molecular- and cellular-level studies are needed to determine the role of butyrate as a signaling molecule for GLP-2 release.