| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
* Department of Animal Nutrition and Physiology, Danish Institute of Agricultural Sciences, Foulum, DK-8830 Tjele, Denmark;
and
Institute of Experimental Clinical Research, Aarhus University Hospital, DK-8000 Aarhus, Denmark;
and
Institute of Anatomy and Physiology, Royal Veterinary and Agricultural University, DK-1870 Frederiksberg C, Denmark; and
and
The Hannah Research Institute, Ayr, KA6 5HL, U.K.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterial and MethodsResultsDiscussionImplicationsLiterature Cited |
|---|
Key Words: Binding Proteins • Biochemical Receptors • Heifers • Insulin-Like Growth Factor • Plane of Nutrition • Somatotropin
|
Introduction |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsDiscussionImplicationsLiterature Cited |
|---|
; Vestergaard et al., 1995; Radcliff et al., 1997) and bull (Groenewegen et al., 1990) calves. Growth hormone exerts some of its action on skeletal muscle growth via IGF-I (Florini and Ewton, 1995) and consistent with this, the concentration of IGF-I in blood is generally increased by GH treatment (Sandles and Peel, 1987; Breier et al., 1988; Vestergaard et al., 1995). The biological effects of IGF-I is modulated by specific IGFBP (Florini and Ewton, 1995; Jones and Clemmons, 1995), and the effects of long-term GH-treatment of growing cattle include elevated IGFBP-3 and reduced IGFBP-2 levels in blood (Hodkingson et al., 1991; Vestergaard et al., 1995; Rausch et al., 2002).
GH effects are greater in well-fed vs. underfed animals (Peters, 1986; Houseknecht et al., 1992; Rausch et al., 2002). It is also seen that a high vs. low feeding level produces a muscle-dependent anabolic response (Vestergaard et al., 2000; Therkildsen et al., 2002). However, the interaction between feeding level and exogenous GH on skeletal muscle IGF receptors and muscle growth has not been thoroughly studied in growing cattle. We hypothesize that the muscle-dependent response to nutritional manipulation and the feeding level-dependent response to GH could be due to changes in the sensitivity to the anabolic signal of IGF-I.
Thus, the objective of the present study was to investigate if changes in hormones, binding proteins, and receptors of the GH-IGF axis could explain the responses in muscle growth due to GH treatment and/or changes in feeding level in prepubertal heifer calves.
|
Material and Methods |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsDiscussionImplicationsLiterature Cited |
|---|
). In short, 24 prepubertal Friesian heifer calves (approximately 150 kg of BW) were allocated to six blocks according to sire and BW. Within each block, animals were assigned to one of four treatments in a randomized way but assuring that the mean initial BW was similar for the four treatments. The four treatments were a low feeding level with or without daily GH treatment or a high feeding level with or without GH treatment arranged in a 2 x 2 factorial design with six blocks. The experimental period was 8 wk (3-wk pretreatment and 5-wk treatment period).
The low feeding level, which was also employed for all animals before start of the treatment period, consisted of ad libitum access to a roughage-based mixture (Table 1). The high feeding level consisted of ad libitum access to a concentrate mixture (Table 1) and of limited access (2 kg/d) to the roughage-based mixture. Feed was supplied twice daily (0800 and 1600) and feed refusals were weighed every morning before new feed was offered. Animals had free access to water. Samples for feedstuff analysis were analyzed for chemical composition and the energy content was calculated (Table 1).
|
Health, Care and Use of Experimental Animals
Blood sampling, GH and placebo injections, and housing and rearing of animals were in compliance with Danish laws and regulations for the humane care and use of animals in research (The Danish Ministry of Justice, Animal Testing Act [Consolidation Act No. 726 of September 9, 1993 as amended by Act No. 1081 of December 20, 1995]). Furthermore, the Danish Animal Experimentation Inspectorate approved the study protocols and supervised the experiment. The health of the animals was monitored, but there were no serious illnesses, and the few veterinary treatments were unrelated to the experimental treatments.
Blood Sampling
Blood samples taken by venipuncture from vena jugularis between 1100 and 1200 were collected weekly starting 3 wk before treatments. Blood was collected in Na-heparinized tubes, cooled, centrifuged (2,000 x g at 4°C for 20 min), and plasma stored (-20°C or -80°C) until assayed. Weekly plasma samples were assayed for GH, insulin, and total IGF-I. Samples obtained in the week before and at the end of the 5-wk treatment period were assayed for GH-binding protein (GHBP), free IGF-I, IGFBP, and IGF-I distribution on 35- and 150-kDa IGFBP-fractions.
Hormone Assays
Free IGF-I and insulin were determined using noncompetitive time-resolved immunofluorometric assays (TR-IFMA) of the sandwich type as previously described (Frystyk et al., 1995; Løvendahl and Purup, 2002). Free IGF-I was measured after ultrafiltration by centrifugation according to Frystyk et al. (1994). For the weekly samples, total extractable IGF-I and GH were assayed by competitive RIA (Etherton et al., 1987; Sejrsen and Foldager, 1992). The weekly sampling protocol was not optimal to specifically address the effects on GH concentrations due to the pulsatile nature of its secretion. However, GH was also measured in order to document that GH was injected properly to all GH-treated animals.
Neutral Size-Exclusion Chromatography
Plasma was separated in 35- and 150-kDa IGFBP fractions and a 7-kDa (i.e., "unbound") fraction by neutral size-exclusion chromatography using HPLC before IGF-I analyses (Hodgkinson et al., 1991). In short, a superose 12 HR column (Pharmacia Biotech AB, Uppsala, Sweden) was loaded with 50 µL of plasma and eluted with a 0.05 M phosphate buffer (pH 7.4). The flow rate was 0.4 mL/min, and fractions of three drops were collected between 23 and 40 min of loading. The relationship between fraction number and molecular weight was established using a standard protein solution (Bio Rad Laboratories, Hercules, CA). The IGF-I in the 35- and 150-kDa IGFBP fractions was measured by a commercial RIA-kit (Nichols Inst. Diagnostics, San Juan, Capistrano, CA, catalog No. 40-2100).
Growth Hormone Binding Protein Assay
Recombinant bovine GHBP (rbGHBP), expressed in Escherichia coli (a gift from R. J. Collier, Monsanto Co., MO), was used for immunization and in the RIA. An antiserum to rbGHBP was raised in rabbit by repeated injection of 50 to 200 µg of GHBP in incomplete Freund’s adjuvant. Samples (100 µL) or GHBP standards were incubated with 100 µL of antiserum to rbGHBP diluted 1:5,000 in RIA buffer (38 mM NaH2PO4, pH 7.4, 154 mM NaCl, 5 g/L of RIA-grade BSA and 1 g/L of thiomersal) containing 2% (vol/vol) normal rabbit serum and incubated overnight at 20°C. The rbGHBP was iodinated with sodium [125I]iodide to a specific activity of approximately 130 MBq/µmol by means of the iodogen-coated tube method (Fraker and Speck, 1978). One hundred microliters of [125I]GHBP (20,000 cpm; 6.5 fmol) in RIA buffer was added and incubated for a further 24 h at 20°C. Then, 300 µL of second antibody mixture, containing 16% (wt/vol) PEG 6000, RIA buffer, and goat anti-serum to rabbit IgG in the proportions 5:3:2, was added and incubated for 4 to 6 h at 20°C. The mixture was then centrifuged at 1,500 x g for 30 min, the supernatant decanted, and the radioactivity in the pellets determined by 
-counting. All 48 samples (wk -1 and 5) were analyzed in a single assay and the intraassay CV was 8%. Sensitivity of the assay was below 0.2 µg/L.
Western Ligand Blotting for IGF-Binding Protein
Levels of IGFBP in plasma were evaluated by Western ligand blotting (Hossenlopp et al., 1986). Briefly, plasma samples (2 µL) were dissolved in nonreducing SDS-polyacrylamide gel buffer, heated to 95°C for 5 min, and separated overnight by SDS-PAGE (10% SDS) at constant current and temperature (4°C). Samples from two blocks of four animals were run on the same gel. A prestained molecular weight marker (Amersham Pharmacia Biotech, Uppsala, Sweden) was run in parallel lanes, and a heifer plasma sample was used as internal standard. Following transfer of proteins onto nitrocellulose membranes by electroblotting, the blots were blocked with 1% BSA and then incubated with [125I]IGF-I (Novo Nordisk A/S, Bagsværd, Denmark) and washed. Autoradiographs (Kodak X-Omat, xar-5 film, Eastman Kodak, Rochester, NY) from the blots were exposed at -70°C for 7 to 14 d and evaluated by laser scanning densitometry (Shimadzu CS-9001PC, Shimadzu, Duisberg, Germany). Abundance of IGFBP was expressed in arbitrary units.
Five bands of IGFBP were identified in plasma: the 40- to 43-kDa IGFBP (identified as IGFBP-3 using Western immunoblotting), a 34-kDa IGFBP (identified as IGFBP-2 using Western immunoblotting), a 28-kDa IGFBP (not identified, but most likely representing IGFBP-5, IGFBP-1 and glycosylated IGFBP-4), and a 24-kDa IGFBP (most likely nonglycosylated IGFBP-4).
Hormone Receptor Assays
Frozen liver or muscle tissue obtained at slaughter (see paragraph below) was used to prepare crude cell membranes by homogenization and differential centrifugation (Purup et al., 1995). Protein content of membrane preparations was determined by the bicinchoninic acid assay (Pierce, Life Technologies, Roskilde, Denmark). Ligand-binding assays were performed essentially as described previously (Purup et al., 1995).
Binding of bovine GH (bGH) to liver membranes was quantified using [125I]bGH as iodinated ligand and bGH as the unlabelled competitor, and previous studies (Purup et al., 1995) have shown that such assays are specific for bGH receptors (GH-R). Likewise, binding of IGF-I to liver and muscle membranes was quantified using [125I]IGF-I as the iodinated ligand and IGF-I as the unlabelled competitor. Using IGF-I, IGF-II, and insulin as competitors, we (Purup et al., 1995) have shown that such assays are specific for type-1 IGF-receptors (IGF-R).
With the liver membranes, only the specific binding of [125I]IGF-I and [125I]bGH was estimated. These assays were run in triplicate and used 400 µg of membrane protein, 1 µg of bGH or 0.1 µg of IGF-I as unlabelled hormones, 20,000 to 30,000 cpm (0.4 ng) of [125I]bGH or 40,000 to 50,000 cpm (0.12 ng) of [125I]IGF-I as labeled hormones. Incubation was at 20°C for 24 h (GH-R assay) or 4°C for 48 h (IGF-R assay). Specific binding was calculated as the difference in cpm bound in the presence (nonspecific binding) and absence (total binding) of excess unlabelled hormone. Data are expressed as percent bound (specific binding/total binding).
For muscle tissue, IGF-I saturation curves and Scatchard analysis were conducted to obtain a more complete picture of binding characteristics. Assays were conducted in which [125I]IGF-I (0.013 pmol/tube) was competed with graded doses (nine concentrations) of cold IGF-I (0.10 to 26.3 pmol/tube) and run in triplicate with 400 µg of membrane protein per tube and using incubation conditions similar to those described above. The binding data were analyzed by the EBDA program (G. A. McPherson, 1983; version 3.0). The obtained Scatchard plots were linear and they were used (by including nine, eight or seven of the concentration points from the curves) to estimate the concentration of type-1 IGF-R (i.e., receptor density, Bmax) and the receptor affinity (Kd). In the two muscles, IGF binding (i.e., percentage bound) was 4 to 8%.
Slaughter Procedure, Tissue Sampling and Carcass Quality Recordings
One block of animals was slaughtered on a given day. The control and GH-treated animals from the low and high feeding levels were slaughtered in alternate order 16 h after the last meal was offered. Animals were loaded and transported (300 m) to the slaughter unit at Research Centre Foulum and stunned with captive bolt before exsanguination. The liver was removed within 7 ± 0.4 min of exsanguination, and samples to be used for receptor assays (40 to 100 g) were snap-frozen in liquid nitrogen. Samples to be used for IGF-R analysis from right-side supraspinatus (SS) and longissimus (LM) muscles (40 to 50 g) were removed within 11 ± 0.4 and 14 ± 0.4 min of exsanguination, respectively. Hot carcass, liver, heart, kidneys, and kidney fat were weighed. Twenty-four hours after slaughter, the remaining left side LM (filet section) and SS muscles were excised, weighed, and, after freezing, analyzed for intramuscular fat (IMF) (Soxtec) and protein (N x 6.25) content. After sampling and recording of carcasses, all carcass and noncarcass components of GH-treated animals were destructed.
Statistical Analyses
The statistical analysis was carried out as a randomized complete block design with 2 x 2 treatments using the GLM procedures of SAS (SAS Inst., Inc., Cary, NC). The model included the fixed effects of block (6) and treatments (GH, feed, and the GH x feed interaction) as main effects. To adjust for small differences in initial BW between treatment groups (Table 2), initial BW was included as a covariate in most analyses. Results are presented as least squares means ± SEM for the 4 treatment groups although very few interactions were present. Plasma concentrations of GH were loge-transformed before analysis to reduce heterogeneity of variance, and results are expressed as the back-transformed means with the true 95% confidence intervals. For the various hormonal traits and muscle weights, correlations were calculated using the MANOVA procedure of SAS on residuals after adjusting for block and initial BW effects in the model.
|
|
Results |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsDiscussionImplicationsLiterature Cited |
|---|
). There was an 18% higher ADG (P < 0.13) with GH treatment. The GH treatment increased carcass weight (P < 0.02), dressing percentage (P < 0.06), and weights of LM (+7%, P < 0.05) and SS (+7%, P < 0.06) muscles with no effects on protein or IMF content of these muscles (data not shown). Weights of liver and heart (and kidneys) were increased by GH treatment only at the high feeding level (data not shown). For all other performance and carcass quality traits, there was an additive effect of GH and feeding level. The 81% higher ME intake and the 140% higher digestible CP intake at the high vs. low feeding level resulted in a 115% higher ADG (P < 0. 001). Furthermore, high feeding level substantially increased carcass weight, dressing percentage, and weights of LM (+22%) and SS (+12%) muscles (all P < 0.001). Kidney fat (1.9 vs. 0.8 kg, P < 0.001), LM IMF (1.1 vs. 0.66%, P < 0.001), LM protein (21.4 vs. 20.8%, P < 0.05), SS IMF (1.3 vs. 0.93%, P < 0.001), and SS protein (20.1 vs. 19.2%, P < 0.003) were all higher for the high vs. low feeding level.
Concentrations of Hormones, Metabolites, and Receptors
The high vs. low feeding level increased the GH binding to GH-R in the liver by 92% (P < 0.001) and also increased the concentration of GHBP in plasma by 70% (P < 0.001) (Table 3). The IGF-I binding to liver membranes was 45% lower with the high vs. low feeding level, and GH treatment further reduced this binding (P < 0.05).
|
). A similar trend for increased Bmax with the high feeding level was seen in SS muscle. Receptor affinities (Kd-values) were not changed in either muscle with GH treatment or changed feeding level.
The high feeding level gradually increased total plasma IGF-I (Figure 1) up to 96% (P < 0.001) and free IGF-I almost 15 times (P < 0.001) vs. the low feeding level (Table 4). The GH treatment further increased plasma IGF-I (total and free), but mainly at the high feeding level. The difference seen in IGF-I at the low feeding level (CON: 123 vs. GH: 162 µg/L) was mainly due to differences present before the start of treatment (127 vs. 153 µg/L, see Figure 1 and footnote to Table 4). However, this pretreatment difference was the only one seen among the various hormone and IGFBP concentrations and receptor binding characteristics. For plasma GH and insulin, there were no interactions between GH treatment and feeding level. The GH treatment increased plasma GH considerably, whereas reduced plasma GH was detected with the high vs. low feeding level. High feeding level increased plasma insulin by 225%.
|
|
). The levels of the 28-kDa (putatively, IGFBP-5, -1, and glycosylated -4) and the 24-kDa (putatively IGFBP-4) IGFBP were slightly higher in GH-treated animals (P < 0.10 and P < 0.06, respectively). However, the high feeding level increased both these IGFBP (P < 0.001).
Both the high feeding level and GH treatment increased the amount of IGF-I bound in the 150-kDa complex, and the high feeding level also increased the amount of IGF-I bound in the 35-kDa classes (Table 4). Because the proportion of total IGF-I that is considered "unbound" after neutral size-exclusion chromatography highly overestimates the real amount of free IGF-I (Frystyk et al., 1994), and because heparin in the plasma may also release some of the IGF-I from the IGFBP, the 35- and 150-kDa fractions both underestimate the amount of IGF-I actually bound to the IGFBP. In fact, the proportion of IGF-I recovered in the 35- and 150-kDa IGFBP fractions varied from 58 to 79% of the total IGF-I measured in the samples before HPLC.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsDiscussionImplicationsLiterature Cited |
|---|
; Vestergaard et al., 1995; Radcliff et al., 1997). Elevation of the feeding level (100 vs. 55% of the ad libitum intake of ME) was considerably more efficient than daily GH administration in stimulating ADG and carcass weight.
In more long-term experiments, a different response to GH treatment in different muscles was seen (Eisemann et al., 1989). We achieved a similar anabolic response to 5-wk of GH treatment in both LM and SS muscles, whereas the high feeding level resulted in a larger response in LM than in SS muscle. Similar muscle type-dependent responses to nutritional manipulation in growing cattle have been reported earlier (Vestergaard et al., 2000; Therkildsen et al., 2002).
Therefore, we hypothesized that a muscle-dependent response to nutritional manipulation—but not to GH treatment—could be due to a changed sensitivity to the anabolic signal of IGF-I. In agreement with this hypothesis, GH treatment did not affect type-1 IGF-R binding characteristics in either LM or SS muscle, in accordance with previous findings in growing heifers (Boge et al., 1994) and lambs (Oldham et al., 1993).
In contrast to GH treatment, a high feeding level markedly increased the type-1 IGF-R density (Bmax) in muscle membranes of LM but did not change the receptor affinity (Kd). The effects were similar but smaller in SS in agreement with the lower anabolic response in this muscle. The Bmax was higher in SS than in LM muscle, and such a muscle-type dependent difference in type-1 IGF-R density was also seen in another heifer study (Boge et al., 1995). Very little data are available on chronic feeding level-effects on functional IGF-I receptors in muscles. Oldham et al. (1993) found that in 8-mo-old wether lambs, low or high nutrition did not affect IGF-I binding in the muscles, and that the type-1 IGF-R were localized on the connective tissue, not the muscle fibers. Although not significant, the IGF-I binding to the connective tissue was slightly higher in the high vs. low nutrition group (Oldham et al., 1993). Comparing fasted and fed 6-mo-old female lambs showed that type-1 IGF-R were present in both muscle fibers and connective tissue, but only receptors of the latter tissue were sensitive to fasting (i.e., increased binding) (Oldham et al., 1993).
The most interesting results in relation to the present findings come from a study where rats were fed at three different protein levels (Dardevet et al., 1991). Going from the medium to the high protein level, there was a slightly higher growth rate, higher plasma IGF-I, and increased binding density of high-affinity, type-1 IGF-R (1.5 vs. 1.1 pmol/mg of protein) in crude hind-leg skeletal muscle microsomal membranes (Dardevet et al., 1991). Although not directly comparable, this finding in rats is supportive to the present feeding (i.e., energy and protein) level-effect on IGF binding (1.4 vs. 0.8 pmol/mg of protein) in muscles from prepubertal heifers. Thus, our findings suggest that an increased density of type-1 IGF-R is involved in the increased muscle growth seen at high feeding level, whereas changes in receptor density and/or affinity are not necessarily involved in GH-stimulated muscle growth of prepubertal heifers.
Therefore, the muscle anabolic effect of GH seems to rely on other effects. Among the endocrine factors measured in the present experiment, the supraphysiological increased GH concentration could be involved. The high concentration with daily GH treatment might have affected local (i.e., muscle, connective, and adipose tissue) growth factor production to act in an autocrine or a paracrine way on the muscles. This growth factor is probably not IGF-I, since GH treatment does not seem to affect muscle IGF-I mRNA levels in pigs (Combes et al., 1997). However, we have no data for further elaborating this possible mode of action.
The high feeding level was associated with reduced plasma GH in this and other studies (Sejrsen et al., 1983; Rausch et al., 2002). Plasma GH in GH-treated calves was also lower at the high feeding level compared with the low feeding level, indicating a more rapid turnover of GH at the high feeding level (Breier, 1999; Rausch et al., 2002). The high feeding level also increased hepatic GH binding (GH-R), and at the same time, increased plasma GHBP. Accordingly, we observed a strong correlation between hepatic GH binding and plasma GHBP (r = 0.80; P < 0.001) as also seen in other studies (Baumann, 1994; Ketelslegers et al., 1996). This indicates increased density (i.e., Bmax) of GH-R (Breier, 1999) and a higher receptor turnover and a higher concentration of external GH-R measured as a high-affinity GHBP in the blood (Baumann, 1994). Both the increased GH binding and the higher insulin level that probably will increase liver IGF-I mRNA (Pell et al., 1993), will stimulate hepatic IGF-I production.
In contrast to the high feeding level, GH administration appears to reduce hepatic GH binding as seen on the high feeding level (-20%), as well as in another GH study (-27%) with prepubertal heifers on a moderate feeding level (Purup et al., 1993). Concomitantly, plasma GHBP was slightly reduced with GH treatment. Others have found little or no effect of GH treatment on GHBP in blood (Baumann, 1994; Combes et al., 1997).
The effects of both high feeding level and GH treatment on endocrine measurements of the IGF axis (i.e., total and free IGF-I, IGFBP-3, and -2, and IGF-I distribution) all go in the same direction and agree with previous findings (Hodkingson et al., 1991; Armstrong et al., 1993; Purup et al., 1993; Vestergaard et al., 1995; Rausch et al., 2002). The high feeding level resulted in a large increase in plasma IGF-I, a result which most likely is a combined effect of protein and energy intake, because the protein level is known to influence the basal IGF-I level (Breier, 1999) as well as the IGF-I response to varying energy levels (Elsasser et al., 1989). Generally, there were interactions between the two treatments, as seen in other experiments with cattle (Breier et al., 1986; Houseknecht et al., 1992; Rausch et al., 2002) and sheep (Hodgkinson et al., 1991; Hua et al., 1993). The generally larger response in the IGF-axis to feeding level compared with GH treatment is in agreement with the largest growth rate, and muscle anabolism is seen with the nutritional manipulation. Although these changes give no direct link to muscle growth, they seem to support description of the animal’s general anabolism (Jones and Clemmons, 1995). However, these endocrine changes give no contribution to the understanding of the individual responses of various muscle types to nutritional changes. The explanation for the muscle-type dependent response is more likely related to the net effect on muscle protein synthesis and degradation, and on the actual rate of muscle protein turnover in individual muscles. Although we have no data to directly characterize the two muscles used in this experiment, our own results (Therkildsen et al., 2002) suggest that such differences exist between LM and SS muscles.
The muscle anabolic effect of GH in the present study was independent of the feeding level, which is in contrast to other findings (Peters, 1986; Houseknecht et al., 1992; Rausch et al., 2002). The changes in the IGF axis hormones with GH treatment, especially at the high feeding level, included more total IGF-I, much more free IGF-I, more IGFBP-3, less IGFBP-2, and relatively more IGF-I bound in the 150-kDa complex. In fact, the distribution of IGF-I within the ternary 150-kDa complex with acid labile subunit (ALS) and IGFBP-3 and the binary complexes with the individual IGFBP is important for the regulation of IGF-I activity (Jones and Clemmons, 1995; Hossner et al., 1997). Insulin-like growth factor-I is bound more tightly in the ternary complex, which prolongs the half-life of IGF-I, and most of the interchange between bound and free IGF-I seems to occur with IGF-I bound in the 35-kDa classes (Jones and Clemmons, 1995). Neutral size-exclusion HPLC showed that a greater proportion of IGF-I was carried within the ternary complex, when animals were at the high feeding level. This seems to be a consequence of less IGFBP-3 forming binary complexes with IGF-I, and with a reduced IGFBP-2 level (Hossner et al., 1997). So even though GH treatment at the low feeding level had little effect on total IGFBP (and thus on IGFBP-3 levels as estimated by Western ligand blotting), the overall result indicates a shift in distribution toward the ternary complex with GH treatment. In the present study, a similar but more pronounced effect of GH treatment (i.e., shift in distribution towards the ternary complex) was seen at the high feeding level, in agreement with the overall increases in IGF-I and IGFBP-3. Together with findings seen in lambs (Hodgkinson et al., 1991), this supports that IGFBP-3 is IGF-I-dependent and ALS is GH-dependent (Jones and Clemmons, 1995).
Free IGF-I was also increased at the high feeding level, especially with GH treatment. An increase in free IGF-I, estimated from size-exclusion chromatography, was also seen in GH-treated lambs at a high feeding level (Hodgkinson et al., 1991). Free IGF-I was, like total IGF-I, highly correlated with LM and SS muscle weight (r = 0.70 to 0.80, P < 0.001). However, the concentration of free IGF-I was 10 to 100 times lower than the Kd of the type-1 IGF-R in the muscles. The low concentration in relation to Kd and the short half-life of free IGF-I (i.e., less than 20 min; Hossner et al., 1997) question whether free IGF-I in the systemic circulation is an important factor for regulating skeletal muscle growth. However, free IGF-I may play a role in a paracrine or autocrine mode of action of skeletal muscle growth.
The reason why the larger response in these hormones and binding proteins with GH treatment at the high vs. low feeding level did not result in increased carcass and muscle growth could be related to the rather short experimental period used in the present experiment. What is usually seen in more long-term GH-experiments is, like the endocrine changes, that the anabolic response is also larger at a high feeding level. In the present experiment, however, only heart, liver, and kidney weights showed this interaction between feeding level and GH treatment. Therefore, it is possible that the more favorable endocrine changes in the IGF axis with GH treatment at the high feeding level would also result in increased muscle deposition if the treatment period was longer. The developmental stage of these animals may also be partly responsible for the fact that the larger changes in the IGF axis are not reflected in increased muscle growth. The very lean dairy prepubertal heifer and the short treatment period were chosen in order to be able to focus on muscle growth and to be able to exclude possible interference from fat deposition and fat mobilization. In fact, the possibility for GH to act via lipolytic and/or anti-lipogenic pathways (Etherton et al., 1995) to direct NEFA to be utilized in muscle tissue, for example, is almost nonexistent at this stage of development. In support of this, our data show that 5 wk of GH treatment did not reduce kidney or intramuscular fat and had only a marginal effect on plasma NEFA levels (i.e., +13%, data not shown). This is in contrast to findings in some more longterm experiments with GH supplementation (Groenewegen et al., 1990; Vestergaard et al., 1993), but is in agreement with others (Kirchgessner et al., 1987; Ono et al., 1996; Rausch et al., 2002).
In conclusion, some of the effects of exogenously administered GH on tissue metabolism, including skeletal muscle, are mediated through the GH-IGF axis, and the hormonal changes seen in the present study are consistent with the changes in performance, carcass, and muscle characteristics. Thus, GH treatment, especially in well-fed animals, stimulates total and free IGF-I and IGFBP-3 levels, reduces IGFBP-2 levels, and increases the proportion of IGF-I bound in the ternary 150-kDa complex. The changes in plasma free-IGF-I, IGFBP-2, and IGFBP-3 follow the change in plasma IGF-I independent of whether these changes are due to exogenous GH or altered feeding level. In contrast to GH treatment, a high feeding level increases GH binding and more markedly reduces IGF-I binding in liver. The GH treatment and high feeding level both stimulate daily gain and muscle growth. Muscle growth of LM and SS muscles is equally affected by 5 wk of GH treatment, whereas high feeding level has a larger effect in LM muscle with "mixed" fiber types than in SS muscle with more "red" fibers. The GH treatment does not affect type-1 IGF-R characteristics in muscles, whereas high feeding level increases type-1 IGF-R density in LM muscle. The effects are similar but smaller in SS muscle. Despite interactions in hormone concentrations, the anabolic effects of GH treatment and change in feeding level seem additive in skeletal muscle.
|
Implications |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsDiscussionImplicationsLiterature Cited |
|---|
|
Footnotes |
|---|
2 Correspondence: Blichers Allé 1, Research Centre Foulum (phone: +45-89-99-15-07; fax: +45-89-99-15-25; E-mail: mv{at}agrsci.dk).
Received for publication September 26, 2002. Accepted for publication May 14, 2003.
|
Literature Cited |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsDiscussionImplicationsLiterature Cited |
|---|
This article has been cited by other articles:
|
|
 |
|
  R. L. Larson Effect of cattle disease on carcass traits J Anim Sci, June 1, 2005; 83(13_suppl): E37 - 43. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
