Colostrum contains carbohydrates, fats, proteins, peptides, minerals, vitamins, peptide and nonpeptide hormones, cytokines, enzymes, polyamines and nucleotides in different amounts than mature milk (Grosvenor et al. 1993, Koldovsky 1989). These substances are important for cellular growth and differentiation and influence gastrointestinal (GI)⁵ tract functions, metabolism, immune reactions and endocrine systems in newborn animals (Koldovsky 1989). Colostral antibodies in neonatal calves are essential for passive immunity. Colostrum intake in neonatal calves influences GI tract function, GI regulatory peptide production and secretion, and exerts metabolic and endocrine changes (Baumrucker and Blum 1994, Baumrucker et al. 1994a and 1994b, Brugère 1989, Demigné and Rémésy 1984, Grütter and Blum 1991a, and 1991b, Guilloteau et al. 1995, Lee et al. 1995). Colostrum-borne hormones, growth factors and cytokines have pre- or postabsorptive effects (Burrin et al. 1992, Grosvenor et al. 1993), but only limited studies exist in calves, such as on insulin-like growth factor I (IGF-I) (Baumrucker and Blum 1994, Baumrucker et al. 1994b).

Colostral concentrations of proteins and of peptides, including peptide hormones and growth factors (Grütter and Blum 1991a, and 1991b, Ronge and Blum 1988) rapidly decrease after the onset of lactation in cows. Furthermore, in neonatal calves, absorption in the small gut of colostral proteins such as -globulins and -glutamyltransferase is restricted largely to d 1 of life (Baumrucker et al. 1994a, Michanek et al. 1989, Stott et al. 1979a, 1979b and 1979c). Only trace amounts of certain peptides such as IGF-I are absorbed (Vacher et al. 1995), and for insulin there is evidence against absorption (Grütter and Blum 1991a). In addition, the efficiency of absorption of -globulins is modified by the time point of first colostrum intake after birth (Stott et al. 1979b) and by oral administration of glucose and/or by plasma glucose levels (Tyler and Ramsay 1993). Endocrine factors such as insulin, thyroid hormones and cortisol can also influence the time point of gut closure for protein and peptide absorption (Brugère 1989). Amounts of absorbed colostral substances and postprandial changes are expectedly variable. In addition to early and transient effects, there may be long-term modifications by nutrition.

On the basis of these premises, we tested the hypothesis that feeding calves with glucose or water instead of colostrum on d 1 of life not only influences the efficiency of absorption of proteins and peptides, but that differences in nutrition on d 1 of life exert priming effects on gastrointestinal, hematologic, metabolic and endocrine traits and exert lasting effects on health status.

MATERIAL AND METHODS

A health index was formed as described by Moser et al. (1994) on the basis of daily measurements or scoring of the following clinical traits: rectal temperature, heart rate, respiratory rate, pulmonary sounds, coughing, nasal discharge, eye discharge, fecal consistency, feed intake and behavior.

Experimental protocols were approved by the cantonal and federal committees for animal experimentation. Calves were randomly assigned to three groups (C, G and W), each consisting of seven animals with an initial BW of 43.0 ± 2.6, 43.6 ± 1.8 and 44.5 ± 2.2 kg, respectively. On d 1, groups C, G and W were first fed at 7, 5 and 5 h, respectively, after birth and for the second time 8 h later. On d 2, groups C, G and W were first fed at 30, 27 and 27 h, respectively, after birth and again 8 h later. From d 3 on, calves were fed at 0800 and 1600 h. Group C received colostrum twice daily on d 1 (1.5 L of milkings 1 and 2, respectively) and on d 2 (2.0 L of milkings 3 and 4, respectively) and then they were fed mature milk twice daily (5% of BW) up to d 7. Group G on d 1 received 1.5 L of glucose monohydrate (2 g/kg BW, dissolved in tapwater) twice daily, and group W received 1.5 L tapwater twice daily. Calves of groups G and W were fed colostrum twice on d 2 (1.5 L of milkings 1 and 2, respectively) and on d 3 (milkings 3 and 4, respectively) and then they were fed mature milk twice daily (5% of BW) up to d 7. 

Cows were milked twice daily. The first four milkings were collected separately in plastic bottles for d 1 and 2, respectively, and either fed directly to the calves of group C or stored at 4°C and warmed to 40°C immediatly before being given to the calf. On the basis of the composition of the colostrum and milk pool (Table 1), dry matter, crude fat, lactose, crude protein, IGF-I and IGF-II intakes were calculated per kilogram BW (Table 2).

Table 1. Composition of colostrum and milk fed to all calves during the first 7 d of life [View Table]

Table 2. Intake of crude fat, lactose, crude protein and insulin-like growth factors I and and II by all calves during the first 7 d of life1 [View Table]

On d 1 of life, calves were prophylactically treated against bacterial infections with s.c. Baytril 5% (Bayer AG, Leverkusen, Germany) for the first 3 d [2 mL/(40 kg BW·d)]. Furthermore, 20 mL of -globulins (Serimmun plus; Biokema SA, Crissier, Switzerland), corresponding to 2 g -globulins, was injected intravenously at the end of d 2. 

The first blood samples in groups C, G and W were taken between 5 and 7 h after birth. Blood samples were taken on d 1 and 2 at 5 min before and 0.5, 1, 2, 4 and 7 h after the first four feedings, and on d 7 at 5 min before and 0.5, 1, 2, 4 and 7 h after the morning feeding by jugular puncture using evacuated tubes. Tubes containing dipotassium-EDTA (1.8 g/L blood) were used to collect blood for the determination of hematologic traits, total protein, albumin, total bilirubin, bile acids, urea, creatinine, glucose, nonesterified fatty acids (NEFA), insulin, IGF-I, IGF-II, growth hormone (GH), 3,5,3-triiodothyronine (T₃), thyroxine (T₄) and cortisol. Tubes containing dipotassium-EDTA (1.8 g/L blood) and the protease inhibitor aprotinin (10 IU/L blood; Sigma Chemical, St. Louis, MO) were used to collect blood for the determination of gastrin and glucose-dependent insulinotropic polypeptide (GIP). Tubes containing dipotassium-EDTA (1.8 g/L blood) and sodium fluoride (3 g/L blood) were used to collect blood for the determination of lactic acid. Tubes (10 mL) without anticoagulants were used for the determination of prolactin (PRL) in serum. Of the first sampling on d 1 and 7, EDTA-containing blood was used for hematologic analyses. Samples were cooled on ice and centrifuged at 1000 × g for 20 min. Tubes without anticoagulants were left at room temperature for 15-30 min and were then centrifuged. Supernatants (plasma or serum) were stored at 20°C for later analyses.

Concentrations of GH, PRL, IGF-I, insulin, GIP, T₃, T₄ and cortisol were measured by RIA as described by Kinsbergen et al. (1994), Ceppi et al. (1994), Hammon and Blum (1997) and Hugi et al. (1997) with the following modifications: for cortisol determinations, the antibody (anti-cortisol-21-thyroglobulin-serum), raised in a rabbit, was from BioMakor, Rehovot, Israel; for IGF-II determinations, the standards were purchased from Amano International Enzymes, Troy, VA; for IGF-II determination, the antiserum against recombinant human IGF-II was from GroPep, Adelaide, Australia (cross-reaction <2% with IGF-I).

Gastrin was measured by RIA. The antiserum against synthetic human gastrin was raised in a rabbit (final dilution of 1:750,000). Bovine plasma containing high amounts of gastrin was serially diluted and paralleled the standard curve, indicating that the antibodies cross-reacted similarly with human and bovine gastrin. Synthetic human gastrin (Fluka, Buchs, Switzerland) was used for iodination, and synthetic human gastrin (Medical Research Council, London, England) was used for standards. The sensitivity of the assay was 16 pmol/L; 50% inhibition of binding (ID₅₀) was at 140 pmol/L. The intra- and interassay variabilities were <10 and 15%, respectively.

The concentrations of total protein, albumin, urea, creatinine, NEFA, glucose, total bilirubin, gastrin, GIP, GH, PRL, insulin, IGF-I, IGF-II and cortisol were measured in all blood samples on d 1, 2 and 7. Lactic acid, bile acid, T₃ and T₄ were determined in the first blood sample of d 1, 2 and 7, and in the first blood sample on d 7.

Analyses in colostrum and milk. Dry matter and concentrations of crude fat and crude protein were determined by Weender analysis using standard procedures at the Federal Research Station of Animal Production, CH-Posieux as decscribed (Hugi et al. 1997, Moser et al. 1994). Lactose was measured enzymatically as described by Hugi et al. (1997) and casein-deprived protein according to Grütter and Blum (1991b). IGF-I and IGF-II were measured in casein-deprived colostrum, obtained by treatment with rennin (Ronge and Blum 1988), by RIA as in plasma.

Areas under the concentration curves from which prefeeding values were substracted and served as measures of incremental or decremental changes (_{0-7 h}) to evaluate net effects of feeding. Furthermore, mean concentrations in samples between 0 and 7 h after each feeding on d 1, 2 and 7 were calculated.

Within groups, differences between prefeeding (basal) values, differences () between postfeeding peak or nadir values and prefeeding (preprandial or basal) values, differences between total incremental or decremental changes (_{0-7 h}) after feedings on d 1, 2 and 7, as well as differences between mean postfeeding concentrations, were compared.

Between groups, preprandial (basal) values, maximal incremental and decremental changes (), total incremental or decremental differences (_{0-7 h}), mean concentrations and values at 7 and 15 h after the first feeding on d 1, 2 and 7 were compared.

Data were analyzed by ANOVA using the GLM procedure of the SAS System for Windows (SAS Institute 1993). The model used was Y_ijk=µ + treatment_i + time_j + (treatment_i × time_j) + e_ijk, where Y_ijk = measured value; µ = general mean; treatment_i = feeding (water, glucose or colostrum); time_j = age at the time of blood sampling; treatment_i × time_j = interaction between time and treatment; and e_ijk = residual error. Paired t test was used to evaluate differences of values within groups, and Student's t test was used to localize differences (P < 0.05) between groups.

RESULTS

Body weight (Fig. 1A) decreased (P < 0.01) in groups G and W during the first 24 h. From d 2 to 7, BW increased (P < 0.001) in group W, but remained lower (P < 0.01) than on d 1. Mean BW (from d 1 to 7) did not differ among groups C, G and W (42.5 ± 2.4, 42.0 ± 1.7 and 42.5 ± 2.3 kg, respectively).

Heart rate (Fig. 1B) on d 1 in group W was lower (P < 0.05) than in groups C and G. Heart rate decreased from d 1 to 7 (P < 0.001) in group G. Heart rate on d 2 was higher (P < 0.05) in group C than in group W. In group C, heart rate on d 7 was higher (P < 0.01) than in groups G and W. Mean heart rate (from d 1 to 7) was higher (P < 0.05) in group C (134 ± 4 beats/min) than in groups G and W (118 ± 5 and 111 ± 3 beats/min, respectively).

Rectal temperature (Fig. 1C) increased (P < 0.01) within 24 h in groups C and G and continued to rise until d 4 in all groups. Rectal temperature on d 4 was higher (P < 0.05) in group C than in groups G and W. Mean rectal temperature (from d 1 to 7) in group C (39.2 ± 0.1°C) was higher (P < 0.05) than in groups G and W (39.0 ± 0.1 and 39.0 ± 0.1°C, respectively).

Respiratory rate (Fig. 1D) of group G on d 1 was lower (P < 0.01) than in group C. Daily mean respiratory rate did not change significantly within groups, but mean respiratory rate in group C on d 7 was higher (P < 0.05) than in groups G and W. Mean respiratory rate (from d 1 to 7) was higher (P < 0.05) in group C (54 ± 3 respirations/min) than in groups G and W (42 ± 1 and 45 ± 2 respirations/min, respectively).

Frequency of loose feces was lower (P < 0.05) in group C than in groups G and W, but there were no other significant differences in the health index among the three groups (data not shown).

Table 3. Plasma concentrations of albumin, globulin, urea and creatinine in calves of group C, G and W on d 1, 2 and 7 of life1 [View Table]

Plasma globulin concentration (Table 3), calculated as the difference between concentrations of total protein (not shown) and albumin, continuously increased (P < 0.001 and P < 0.05 at 7 h after the first and second meal, respectively) in group C on d 1 and then tended to decrease (P < 0.1) up to d 7. Globulin concentration in group G decreased (P < 0.05) after the first glucose intake and tended to decrease (P < 0.1) after water intake on d 1, but increased (P < 0.01 at 7 h after first meal in group W) on d 2 and then remained elevated up to d 7. On d 7, globulin concentrations before the first meal in groups G and W were higher (P < 0.01) than those on d 1 and 2. Incremental changes (_{0-7 h}) after the first meal on d 2 were smaller in groups G and W (P < 0.05) than in group C on d 1 (not shown). Concentrations 7 h after intake of the first meal on d 1, before and at 7 h after the second meal on d 1, and before the first and second meal on d 2 were higher (P < 0.05) in group C than in groups G and W. At 7 h after the second meal on d 2, the concentration in group C tended to be higher (P < 0.1) than in group W, which in turn tended to be higher (P < 0.1) than in group G. On d 7, the concentration was higher (P < 0.05) in group C than in groups G and W and higher (P < 0.05) in group W than in group G.

Plasma urea concentration (Table 3) in group C decreased (P < 0.05) on d 1 after the second feeding. In group G, the concentration decreased (P < 0.05) on d 1 after the first and second feeding, was lowest preprandially on d 2 and then increased to levels which on d 7 were higher preprandially (P < 0.01) than on d 1 and 2. In group W, mean concentration (not shown) on d 7 was higher (P < 0.05) than on d 1 and 2. On d 2, concentrations before the first meal in group G were lower (P < 0.05) than in groups C and W and were lower (P < 0.05) before the second meal in groups C and G than in group W. On d 7, prefeeding concentration and mean concentration (not shown) in group W was higher (P < 0.01) than in group C and tended to be higher (P < 0.1) than in group G.

Plasma (preprandial) creatinine concentration (Table 3) decreased (P < 0.001) in all three groups up to d 7, and concentrations decreased irregularly (P < 0.05) postprandially. There were no significant group differences.

Plasma glucose concentration (Fig. 2) in group C increased transiently (P < 0.05) within 2 h after each feeding on d 1 and 2, but not on d 7. Glucose concentration in group G increased markedly (P < 0.01) up to 2 h after glucose intake on d 1, increased postprandially (P < 0.001) on d 2 after intake of colostrum, but less (P < 0.01) than on d 1 and did not change significantly on d 7. Preprandial concentrations in group G on d 7 were lower (P < 0.01) than before the first feeding on d 2. Concentrations in group W on d 1 decreased transiently (P < 0.05) up to 2 h after water intake and then slowly increased. Glucose concentrations in group W increased only slightly (P < 0.05) on d 2 after colostrum intake and on d 7 after milk feeding. Levels before the first feeding on d 1 were lower (P < 0.05) in group W than in group G and tended to be lower (P < 0.1) than in group C. On d 1, incremental changes (_{0-7 h}) after each meal in group G were greater than in groups C (P < 0.01) and W (P < 0.001), and after the second meal, postprandial increments in the group C were greater (P < 0.05) than in group W. During d 1, mean glucose concentration in group G was higher (P < 0.01) than in group C and higher (P < 0.05) in group C than in group W; 15 h after the first feeding, glucose concentration in group W was lower than in group C (P < 0.01) and group G (P < 0.05). On d 2, mean glucose concentrations and levels at 7 h after the meals were higher (P < 0.01) in group C than in groups G and W. Preprandial levels on d 7 were higher (P < 0.01) in group C than in group W and tended to be higher (P < 0.1) than in group G. 

Preprandial plasma lactic acid concentration decreased from d 1 to 7 (means of group C, G and W on d 1, 2 and 3: 3.7 ± 1.0, 2.1 ± 0.3 and 1.2 ± 0.1 mmol/L, respectively). There were no group differences.

Plasma NEFA concentration (Fig. 3) in group C was highest immediately after birth and decreased transiently after feedings (P < 0.01 at 4 h after the first feeding and 7 h after fourth feeding); preprandial and mean levels on d 7 were lower (P < 0.05) than values before the first feeding on d 1. NEFA concentrations in group G also decreased reversibly (P < 0.01) between up to 4 h after the first meals and on d 7 at 7 h after feeding and reached particularly low levels on d 1 after the second glucose ingestion; preprandial levels on d 7 were lower than values before the first feeding on d 1 and 2 (P < 0.05). NEFA concentrations in group W increased transiently (P < 0.05) after water intakes on d 1, but decreased transiently (P < 0.05) on d 2 and on d 7; preprandial values on d 7 were lower (P < 0.05) than values before the first feedings on d 1 and 2. On d 1, concentrations in group W at 7 and 15 h after the first meal were higher (P < 0.05) than in groups C and G. At 15 h after the first meal on d 2, the concentrations in group C were lower (P < 0.05) than in the other groups. On d 7, there were no group differences.

Total plasma bilirubin concentrations (Fig. 4) in group C did not significantly change on d 1, increased transiently (P < 0.05) after the first feeding on d 2, then decreased; on d 7, preprandial and mean concentrations were lower (P < 0.05) than values before first feedings on d 1 and 2. Concentrations in group G on d 1 decreased transiently (P < 0.05) after the second glucose intake, were higher (P < 0.05) before feeding on d 2 than on d 1, then decreased to preprandial levels, which were lower on d 7 (P < 0.05) than values before the first feedings on d 1 and 2. Concentrations in group W on d 1 increased (P < 0.05) at 7 h after each feeding. Prefeeding levels in group W were higher (P < 0.05) on d 2 than on d 1, then decreased and were lower on d 7 (P < 0.05) than values before the first feedings on d 1 and 2. Concentrations at 7 h after the first and second feeding on d 1 in group W were higher (P < 0.01) than in groups C and G.

Preprandial plasma bile acid concentrations in groups C, G and W (d 1: 13.7 ± 2.2, 15.9 ± 1.4 and 14.2 ± 3.3 µmol/L, respectively; d 2: 16.0 ± 1.9, 10.6 ± 2.0 and 17.1 ± 4.7 µmol/L, respectively; d 7: 19.9 ± 4.7, 17.5 ± 3.9 and10.9 ± 1.8 µmol/L, respectively) were not different. Bile acid concentrations decreased transiently at 7 h after the first meal (P < 0.05) in group G. 

Plasma GIP concentration (Fig. 6) in group C increased (P < 0.05) after feedings on d 1 and 2, but did not rise significantly on d 7. Highest GIP concentrations of all groups were reached in group C on d 1 at 2 h after the second feeding. Postprandial increments (_{0-7 h}) on d 1 and 2 in group C became progressively smaller after intake from the first to the fourth meal. Concentrations in group C before the first feeding on d 2 were higher (P < 0.05) than on d 1 and tended to be higher (P < 0.1) than on d 7. Mean concentrations were higher (P < 0.01) on d 1 after the second feeding than after feedings on d 2 and 7 and higher (P < 0.05) on d 2 than on d 7. On d 1, increments (_{0-7 h}) after the first meal in group C were greater (P < 0.05) than after the second meal. Concentrations in group G increased (P < 0.05) on d 1, 2 and 7 after each meal. Concentrations in group G before first feeding on d 2 were lower (P < 0.01) than on d 1 and 7. GIP concentrations in group W decreased (P < 0.05) after the first water intake, but increased transiently (P < 0.05) after the second water intake on d 1, increased transiently (P < 0.05) on d 2 after first colostrum intake but did not change significantly on d 7. Preprandial concentrations of group W on d 7 were similar to those before the first feeding on d 1 and tended to be higher (P < 0.1) than on d 2. Concentrations at 7 h after intakes of first and second colostrum on d 1 and preprandial concentrations on d 2 were higher (P < 0.001) in group C than in groups G and W. Mean concentrations on d 1 in group C were higher (P < 0.05) than in groups G and W. On d 1, increments (_{0-7 h}) in group C after the first feeding were greater (P < 0.001) than in group G, in which the rise was greater (P < 0.05) than in group W.

Plasma insulin concentrations (Fig. 7) in group C increased transiently (P < 0.05) after feedings on d 1 and 2, but did not change significantly on d 7. Postprandial increments (_{0-7 h}) tended to be greater (P < 0.1) after the first meal than after the other meals. In group G, insulin markedly increased (P < 0.01) postprandially on d 1, was lower preprandially on d 2 (P < 0.05) than on d 1, tended to increase (P < 0.1) after colostrum intakes on d 2, but did not change after feed intake on d 7, i.e., postprandial increments (_{0-7 h}) on d 2 were lower (P < 0.05) than those of d 1. In group W, insulin concentration decreased after the first water intake, but increased transiently (P < 0.05) after the second water intake on d 1, tended to increase (P < 0.1) on d 2 after intake of colostrum, but did not change postprandially on d 7; on d 2 incremental changes (_{0-7 h}) were greater (P < 0.05) than on d 1. Postprandial increments (_{0-7 h}) on d 1 were greater (P < 0.01) in groups C and G than in group W. Insulin concentrations at 7 h after feed intakes on d 1 and at the beginning of d 2 were higher (P < 0.05) in group C than in groups G and W. Mean concentrations on d 7 in group C were higher (P < 0.05) than in the other groups.

Plasma IGF-I concentration (Fig. 8) in group C increased slowly on d 1 (P < 0.05) up to 7 h after second colostrum intake, but did not rise postprandially on d 2 and 7. Concentrations in group C before the first feeding on d 2 were similar to those of d 1, but were lower on d 7 (P < 0.05) than on d 1 and 2. Concentrations in groups G and W on d 1 decreased (P < 0.05) up to 15 h after the first glucose or water intake, but did not change significantly on d 2 and 7. Concentrations in groups G and W before the first feedings on d 2 were lower (P < 0.05) than on d 1 and were lower (P < 0.05) on d 7 than on d 1 and 2. Preprandial concentrations before the first meal on d 2 were higher (P < 0.05) in group C than in groups G and W. Mean concentrations in group C on d 1 after the second feeding and mean levels on d 2 were higher (P < 0.05) and tended to be higher on d 7 (P < 0.1) in group C than in groups G and W. 

Plasma IGF-II concentrations (Table 4) in group C on d 2 before first feeding were lower (P < 0.05) than on d 1, whereas concentrations before the first feeding in groups G and W on d 1, 2 and 7 were comparable. There were no consistent postprandial changes and no significant group differences.

Table 4. Concentrations of plasma insulin-like growth factor II (IGF-II), plasma growth hormone (GH), serum prolactin (PRL) and plasma cortisol in newborn calves of groups C, G and W on d 1, 2 and 7 of life1 [View Table]

Preprandial and postprandial plasma GH concentrations (Table 4) did not change in a consistent manner. There were no consistent group differences.

Preprandial serum prolactin concentrations (Table 4) were higher (P < 0.05) on d 2 than on d 1 in groups C and G, but not in group W. Preprandial concentrations on d 7 in groups C and G were higher (P < 0.05) than on d 1 and were lower (P < 0.05) in group W than on d 2. PRL did not change in a predictable manner after feed intake. Concentrations before the second feeding on d 1 were lower (P < 0.05) in group G than in group C, but there were no other significant group differences.

Preprandial plasma cortisol concentrations (Table 4) in groups C and G were lower (P < 0.05) on d 2 than on d 1, were lower (P < 0.05) in all groups on d 7 than on d 1 and were lower (P < 0.05) in groups G and W than on d 2. Cortisol concentrations decreased in all groups on d 1 after both feedings, on d 2 after the first feedings, but not in all groups after the second feedings on d 2 and 7. There were no significant group differences.

Preprandial plasma T₃ and T₄ concentrations (Table 5) in groups C and W were lower on d 2 (P < 0.05) than on d 1. On d 7, concentrations were lower (P < 0.05) in all groups than on d 1 and 2. Concentrations on d 1 were lower (P < 0.05) in group W than in groups C and G, but concentrations did not differ among groups on d 2 and 7. 

Table 5. Concentrations of plasma 3,5,3-triiodothyronine and plasma thyroxine in newborn calves on d 1, 2 and 7 of life1 [View Table]

DISCUSSION

Higher albumin concentrations and lower plasma urea concentrations on d 7 in group C suggest that fewer amino acids were deaminated or that more protein was synthesized than in groups G and W. Because creatinine concentrations in the three groups were similar, differences in renal function were unlikely and cannot explain differences of plasma urea levels.

Hepatic glycogen mobilization, which is marked in newborns (Girard 1986), probably allowed maintenance of near normal glucose levels in group W on d 1. Glucose is an important energy substrate for the newborn, especially for the intestine (Girard 1986). Oral glucose administration probably covered some of the intestinal glucose needs. However, with glucose in group G, much less energy was ingested than in calves of group C, although, on the basis of other studies (Hugi et al. 1997), a more marked rise of glucose on d 1 in group G than in groups C and W was expected. Nevertheless, relatively low glucose levels on d 2 in group G were surprising. Higher glucose concentrations and postprandial responses in group C on d 2 and 7 compared with groups G and W were probably the result of several factors. Thus intake of colostrum may have stimulated small intestinal lactase activity, hence lactose digestion and absorption of glucose and galactose, as in newborn pigs (Tivey et al. 1994). In addition, early postnatal intake of colostrum, providing high amounts of gluconeogenic substrates, may have enhanced gluconeogenesis (Girard 1986). Moreover, glucose sparing as a consequence of greater fatty acid oxidation (Girard 1986) possibly contributed to higher glucose levels from d 2 on in group C than in groups G and W. 

Higher levels of NEFA in group W than in the other groups on d 1 were expected. Although lactate reduces fat mobilization during energy deficiency (Demigné and Rémésy 1984), lactate levels were apparently not sufficiently high to reduce NEFA responses to energy deficiency in water-fed calves. The marked fall of NEFA especially after the second glucose intake on d 1 indicates reduced fat mobilization, in part as a consequence of enhanced insulin release. The lack of difference in the levels on d 2 and 7 in the three groups suggests that energy intakes were similar. Bilirubin increases during food deprivation in mature ruminants, as was the case in group W on d 1 and d 2. The increased NEFA in group W on d 1 likely competed with plasma albumin binding and hepatic handling of bilirubin.

On d 1, a rise of insulin in response to colostrum and glucose intakes and low levels in the group provided only water were expected. Increments in groups C and G on d 1 were similar, although glucose increased much more in group G than in group C. Colostral insulin may have been absorbed and/or factors other than glucose, such as gastrointestinal hormones, stimulated insulin secretion. Because absorption of colostral insulin is unlikely (Grütter and Blum 1991a), the increase of plasma insulin in group C after the first meal was primarily the consequence of enhanced secretion. GIP increased much more in group C than in groups G and W, but there is no evidence to date that GIP is insulinogenic in calves (Guilloteau et al. 1995). Higher basal insulin levels on d 2, a tendency for a more marked rise of insulin in group C on d 2 and on d 7 than in groups W and G and a complete lack of insulin response to milk intake in groups G and W on d 7 may have in part been in association with differences in glucose levels. The data suggest that differences in feeding on d 1 may have prolonged effects on postprandial insulin secretion.

Because various macromolecules can be transported through the neonatal gut of calves (Michanek et al. 1989), a more marked rise in IGF-I levels on d 1 in calves fed colostrum instead of mature milk may lead to the conclusion that colostrum-borne IGF-I was absorbed by our neonatal calves. However, only small amounts of IGF-I or Long-R³-IGF-I are absorbed by newborn calves (Hammon and Blum 1997, Vacher et al. 1995). The sluggish rise of IGF-I in group C and the fall of IGF-I levels during the first 24 h of life in groups G and W in this study therefore probably mirrored increased and decreased endogenous production, respectively. Differences on d 1 in concentrations of insulin, which stimulates IGF-I production in cattle, cannot explain the higher IGF-I levels in group C than group G because insulin levels on d 1 were similar. Concentrations of GH, T₄ and T₃ , which can influence IGF-I production, were not significantly different among groups and thus also cannot explain higher levels of IGF-I in group C than groups G and W. Furthermore, GH enhances IGF-I production only slightly in the neonatal calf (Breier et al. 1988, Hammon and Blum 1997). IGF-I poduction was possibly enhanced in group C by nutrient factors, especially fatty acids and amino acids. IGF binding protein patterns, hence metabolic clearance rates of IGF-I, can be altered by nutrition in neonatal calves (Hammon and Blum 1997). A reduced metabolic clearance rate of IGF-I in group C compared with the other groups can therefore not be excluded. The different behavior of IGF-II in colostrum and in blood plasma compared with IGF-I indicates that in the neonatal calf the nutritional effects on IGF-II levels are different than those of IGF-I.

The inconsistent postprandial changes and group differences of GH in this study indicate that this hormone was not affected by different feeding on d 1. As in mature ruminants, it apparently takes more than 1 d until GH increases during fasting in neonatal calves. NEFA and triglycerides can modify GH secretion (Coxam et al. 1989). However, although there were group differences in triglycerides (Blum et al. 1997) and NEFA, effects on GH levels were not detectable.

An increase in PRL in colostrum-fed calves on d 1 was also seen previously (Baumrucker and Blum 1994). Importantly, this study also shows that differences in feeding on d 1 had no significant influence on PRL responses. In addition, absorption of PRL, which is present at high concentration in bovine colostrum, could be excluded as a cause of the postnatal rise in PRL.

Levels of cortisol were high at birth and then decreased, in accordance with Mao et al. (1994). Because cortisol concentrations were similar in the three groups, different feeding on d 1 had no influence, and stress effects, if any, were comparable.

Concentrations of T₄ and T₃ were very high at birth and then decreased to relatively low concentrations on d 7, in accordance with Vermorel et al. (1989). As in mature ruminants, T₄ and T₃ levels decrease during energy deficiency in young calves (Grongnet et al. 1985, Kinsbergen et al. 1994), but the 24-h feed restriction in group W on d 1 had no significant effects. High feeding intensity and energy intake in cattle increase plasma T₄ and T₃ levels, and greater amounts of colostrum fed in the study of Grongnet et al. (1985) than in our trial may have caused a relatively enhanced production and/or a slower postnatal fall of T₄ and T₃ concentrations than in our study. T₃ is present in cows milk in considerably higher amounts than T₄ (Ronge and Blum 1988). Because T₄ and T₃ levels in the three calf groups were similar, differences in the thryroid status were not responsible for differences in metabolism in our calves. Thyroid hormones are important for the maturation of small intestinal function (Brugère 1989). Because levels of T₄ and T₃ on d 2 in groups C, G and W were similar, thyroid hormones were not likely responsible for obviously different digestive and absorptive capacities for proteins and peptides.

This study shows that intake of colostrum instead of glucose or water on d 1 of life had variable effects on the different traits determined in this study. Thus, health status, body weight, hematologic traits and concentrations of creatinine, lactate, bile acids, GH, PRL, cortisol, T₄ and T₃ were not significantly affected. However, other traits (rectal temperature, heart rate, respiratory rate, fecal consistency, albumin, globulin, urea, glucose, total bilirubin, NEFA, gastrin, GIP, insulin and IGF-I) were influenced by different feeding on d 1 of life. Differences were seen only on d 1 and/or d 2 (NEFA, bilirubin, gastrin, GIP) or lasted up to d 7 (globulin, glucose, insulin, IGF-I) or appeared only on d 7 (albumin, urea). Although it is well documented that early colostrum feeding in calves and other species is important for passive immunity, this study demonstrates additionally that differences in commencement of colostrum feeding and in glucose status have important effects on clinical, metabolic and endocrine traits.

ACKNOWLEDGMENTS