Branched-chain amino acid aminotransferase activity decreases during development in skeletal muscles of sheep

Nutrient Metabolism

Branched-Chain Amino Acid Aminotransferase Activity Decreases during

Development in Skeletal Muscles of Sheep

Magali Faure, Franc¸oise Glomot and Isabelle Papet

Unite´ d’Etude du Me´tabolisme Azote´, Institut National de la Recherche Agronomique, Centre de Recherches de Clermont-Ferrand-Theix, France

ABSTRACT The catabolism of branched-chain amino acid (BCAA) differs between sheep and monogastric animals. The transamination of BCAA seems to be affected by development of the sheep. We studied the developmental changes in the activity and expression of the BCAA aminotransferase (BCAT) isoenzymes in skeletal muscle of sheep. Five muscles were taken from fetus, newborn, preruminant and ruminant lambs. BCAT specific activity and the contribution of each BCAT isoenzyme [mitochondrial and cytosolic (BCATm and BCATc, respec-tively)] were quantified using radioenzymatic and immunoprecipitation assays. BCATm and BCATc mRNAs were assessed by real-time reverse transcription–polymerase chain reaction. BCAT specific activities were 62% (dia-phragma) to 83% (longissimus dorsi) lower in the ruminant lamb than in the fetal sheep. BCATm and BCATc were both expressed in sheep skeletal muscle at all developmental stages. BCATc was mainly responsible for the developmental decrease in BCAT specific activity. BCATc specific activities were 77% (diaphragma) to 92% (longissimus dorsi) lower in the ruminant lamb than in the fetal sheep, whereas BCATm specific activities were only 36% (semimembranosus) to 56% (longissimus dorsi) lower. BCATc and BCATm mRNAs in the longissimus dorsi were not affected by development of the sheep. The developmental decrease in BCATc activity, and to a lesser extent in BCATm activity, probably involves posttranscriptional mechanisms in sheep. The present results are consistent with lower in vivo metabolism of BCAA in ruminant than in the fetal sheep. J. Nutr. 131: 1528 –1534, 2001.

KEY WORDS: ● branched-chain amino acid aminotransferase ● development ● sheep ● skeletal muscle The first step in the catabolic pathway of the

branched-chain amino acids (BCAA),2leucine, isoleucine and valine, is the reversible transamination catalyzed by BCAA aminotrans-ferase (BCAT, EC 2.6.1.42). The products of transamination are branched-chain ␣-ketoacids (BCKA), ␣-ketoisocaproate (KIC), ␣-keto-␤-methylvalerate (KMV) and ␣-ketoisovaler-ate (KIV), respectively. We recently cloned the sheep cDNA of mitochondrial BCAT (BCATm) (Faure et al. 1999) and cytosolic BCAT (BCATc) (Bonfils et al. 2000). The second step is irreversible oxidative decarboxylation of BCKA cata-lyzed by the mitochondrial BCKA dehydrogenase enzyme complex (EC 1.2.4.4; Harris et al. 1997) forming the branched-chain acyl-CoA derivatives. This step commits the BCAA carbon skeleton to the degradation pathway. A unique feature of BCAA metabolism in mammals is its tissue speci-ficity; most indispensable amino acids are degraded in the liver, whereas BCAA are metabolized extensively in extrahe-patic tissues (Harper et al. 1984). BCAA contribute to the

synthesis of dispensable amino acids (alanine, glutamine) that participate in interorgan exchanges of nitrogen within the body (Harper et al. 1984).

For the past four decades, the regulation of BCAA catabolic enzymes has been studied extensively in monogastric animals and in humans. Investigations of BCAA catabolism in rumi-nants are scarce, but there is converging evidence suggesting that BCAA catabolism is unique in ruminants. Indeed, several indirect measures of evidence, including arteriovenous differ-ences in blood amino acid concentrations, whole body amino acid fluxes, in vitro BCAA oxidation measurements and en-zymatic assays, support the hypothesis that BCAA catabolism is low in ruminant muscle (Papet et al. 1992, Teleni 1993). The reported values for BCAT specific activity in ruminant tissues were lower than those in rats (Bergen et al. 1988, Goodwin et al. 1987, Papet et al. 1988, Wijayasinghe et al. 1983). We recently showed that the expression of BCAT (mRNA and activity) in skeletal muscle, which is the primary site of BCAA transamination (Papet et al. 1988), was much lower in sheep than in rats (Faure et al. 1999). Studies carried out at several developmental stages (fetus, growing lambs, adults and pregnant ewes) suggest that BCAT activity de-creases during development (Bergen et al. 1988, Faure et al. 1999, Goodwin et al. 1987, Liechty et al. 1987, Papet et al. 1988). Such a phenomenon does not occur in rats (Cappuc-cino et al. 1978, Kadowaki and Knox 1982). The present study

1_{To whom correspondence should be addressed at. E-mail: papet@}

clermont.inra.fr

2_{Abbreviations used: BCAA, chain amino acid; BCAT,}

branched-chain amino acid aminotransferase; BCATc, cytosolic BCAT; BCATm, mitochon-drial BCAT; BCKA, branched-chain␣-ketoacid; DIA, diaphragma; KIC, ␣-ketoiso-caproate; KIV,␣-ketoisovalerate; KMV, ␣-keto-␤-methylvalerate; LD, longissimus dorsi; PLP, pyridoxal-5⬘-phosphate; SM, semimembranosus; ST, semitendino-sus; TFL, tensor fascia latae.

0022-3166/01 $3.00 © 2001 American Society for Nutritional Sciences.

Manuscript received 15 August 2000. Initial review completed 24 September 2000. Revision accepted 30 January 2001.

1528

was undertaken to determine the biochemical basis responsible for the decrease in BCAT activity in skeletal muscle during the development of sheep.

MATERIALS AND METHODS

Sheep and collection of blood and tissues. Male Limousine ⫻ Romanov ⫻ Ile de France lambs from a herd at Institut National de la Recherche Agronomique Clermont-Ferrand-Theix Center were studied at different developmental stages: fetus, newborn, prerumi-nant and rumiprerumi-nant (Table 1). Three pregprerumi-nant ewes were slaughtered at 141 d of gestation (term is 147 d), and the fetuses were removed immediately from the uteruses. Newborn, preruminant and ruminant lambs were born through noninduced births and were allowed to stay with their dams for 1 d. Six newborn lambs were killed the day after birth. Lambs included in the preruminant group were transferred into individual pens in the animal care facility of the center. They were fed twice daily and allowed free access to food. The diet was an artificial diet containing 60% milk (Agnodor; Sanders, Athis Mons, France). The composition of powdered diet (96 g/100 g dry matter) was 240 g crude protein, 240 g fat, 5 g cellulose and 75 g ash per kg, and the gross energy was 22 MJ/kg. Lambs included in the ruminant group were initially reared in the barn, where they had free access to the commercial milk, a commercial concentrate and hay. When they were⬃38 d old (12 kg), they were transferred to individual pens in the animal care facility of the center and weaned through the com-plete withdrawal of milk. They were fed only hay (7.5 MJ metabolized energy and 77 g crude protein per kg) and a commercial concentrate (11.1 MJ metabolized energy and 175 g protein per kg, Ucabec; Lapeyrouse, France) for⬃3 wk. Experiments complied with the Guide for the Care and Use of Laboratory Animals (National Research Coun-cil 1985). All sheep were anesthetized with pentobarbital injection and then exsanguinated. Blood was collected from each sheep and the plasma was immediately separated through centrifugation. Aliquots were stored at⫺20°C for amino acid, nonesterified fatty acid, glucose, lactate and thyroid hormone triiodothyronine (T3) and thyroxin (T4) assays and at ⫺80°C for␣-ketoacid, insulin, cortisol and glu-cagon assays. Five skeletal muscles [longissimus dorsi (LD), tensor fascia latae (TFL), semimembranosus (SM), semitendinosus (ST) and diaphragma (DIA)] were removed. DIA, which is composed of slow twitch-oxidative (slow red) and fast twitch-oxidative glycolytic (fast red) fibers, is the most oxidative. TFL, which is composed of fast twitch-glycolytic (fast white) fibers, is the most glycolytic. ST is classified as a fast white muscle and LD and SM as fast red muscles (Briand et al. 1981, Finkelstein et al. 1992, Lacourt and Arnal 1974). Aliquots of each sample were immediately frozen and stored at ⫺80°C until BCAT enzymatic assay or frozen in liquid nitrogen and stored at_{⫺80°C for RNA analysis.}

Tissue preparation. Frozen skeletal muscles (5 g) were homog-enized in 20 mL of 10 mmol potassium phosphate/L buffer (pH 8) as previously described (Faure et al. 1999). Homogenates were centri-fuged for 20 min at 10,000_{⫻ g, and supernatants were used for BCAT} activity and immunoprecipitation assays.

BCAT activity assay. BCAT activity assays were performed as previously described (Faure et al. 1999). Substrates used in the present enzymatic assay were isoleucine (12 mmol/L) and [1-14_C]KIC

(0.5 mmol/L). The␣-ketoacid specific radioactivity was ⬃350 dpm/ nmol. Linearity with respect to incubation time and amount of the extract used in the assay was established for each type of muscle sample. Protein was determined using the Bio-Rad protein assay (Richmond, CA) with bovine serum albumin as a standard. A unit of BCAT activity was defined as 1 nmol of BCAA formed/min at 37°C. BCAT specific activity was expressed as nmol/(min䡠 mg protein).

Immunoprecipitation experiment. An antiserum raised against the sheep BCATm isoenzyme was used (Faure et al. 1999). Sheep tissue homogenates were adjusted to 0.5 mol NaCl/L, and immuno-precipitation was conducted as described elsewhere (Hall et al. 1993). The volume of antiserum needed to quantitatively neutralize BCATm was determined to be 25␮L per 200 ␮L of homogenate (see Results). The controls used preimmune serum. The BCATm specific activity was calculated from the percentage of immunoprecipitated activity for each sample. The BCATc specific activity was obtained by the difference between BCAT and BCATm specific activities. The accuracy of this methodology was confirmed by comparing these results with quantification of BCATc and BCATm activity after their separation using ion exchange chromatography (see Results).

Ion exchange chromatography. A 100,000 ⫻ g supernatant obtained from 2.5 g of fetal longissimus dorsi was applied to a Fractogel EMD DEAE 650 (S) ion exchange column (1⫻ 10 cm; Merck, Darmstadt, Germany) equilibrated with 10 mmol potassium phosphate/L buffer (pH 7.5) containing 1 mmol dithiothreitol and 5 mmol benzamidine per L (Faure et al. 1999). After the column had been washed with the equilibrium buffer, BCAT activity was eluted by application of 0 –1 mmol/L NaCl linear gradient in the latter buffer. BCAT activity was quantified in the chromatographic frac-tions as described above. Chromatography was carried out at 4°C. The column was attached to an L-6620 Intelligent Pump HPLC system equipped with an L-4200 UV Detector and an L-5200 Frac-tion Collector (Merck).

RNA isolation and real-time quantitative reverse transcription– polymerase chain reaction (PCR). Total RNA was isolated by extrac-tion with guanidinium isothiocyanate (Chomczynski and Sacchi 1987). Single-stranded cDNAs were obtained through reverse transcription of 5 ␮g total RNA using the Superscript preamplification system (Life Tech-nologies, Gaithersburg, MD). The levels of BCATm and BCATc mRNAs were quantified by real-time PCR using the fluorescent TaqMan methodology and a 7700 Sequence Detector System (PE Biosystems, Courtaboeuf, France) (Heid et al. 1996, Holland et al. 1991). The amplification and product-reporting system used is based on the 5⬘ to 3⬘ exonuclease activity of the Taq DNA polymerase. In addition to the two amplification primers, as in conventional PCR, a dual-labeled fluoro-genic hybridization probe is also included. One fluorescent dye serves as a reporter, and its emission is quenched by a second fluorescent dye. During the extension phase of PCR, the reporter is cleaved from the probe thus released from the quencher, resulting in an increase in fluorescent emission. The intensity of the fluorescent emission is plotted versus PCR cycle number to generate an amplification curve for each sample. The fractional cycle number at which the fluorescence is higher than a fixed threshold is defined as the threshold cycle (CT). The

standard procedure was performed with 5␮L cDNA diluted 6.7 times. Amplification primers and TaqMan probes were designed based on the 3⬘ end regions of sheep BCATm and BCATc cDNA sequences [Fig. 1; GenBank accession numbers AF050173 (Faure et al. 1999) and AF184916 (Bonfils et al. 2000), respectively] using Primer Express Oligo Design (version 1.0; PE Biosystems). The sense and antisense primers for amplification of BCATm were 5 ⬘-TGCAATATGAAGTAA-GAAGCGGG-3⬘ and 5⬘-GGCCAAGAAGCCTGGGTC-3⬘, respec-tively (Oligo Express, Paris, France), and the TaqMan BCATm probe was FAM-5⬘-AGCCGGGACGCCTGGATCTCC-3⬘-TAMRA (Eu-rogentec, Lie`ge, Belgium). The sense and antisense primers for amplifi-cation of BCATc were 5_{⬘-GGACAATCACGGTAGCCTGAG-3⬘} and 5⬘-CTAAAATCAGGTAGCCAAAGACATTTC-3⬘, respec-tively (Oligo Express), and the TaqMan BCATc probe was VIC-5 ⬘-TGTGTGCGCTTGGGACAGACTGTG-3⬘-TAMRA (PE Biosys-tems). The cycling conditions included 2 min at 50°C, 10 min at 95°C

TABLE 1

Characteristics of the sheep1

Fetus Newborn Preruminant Ruminant

n 6 6 5 5

Age, d ⫺6.0 ⫾ 0.42 1.0⫾ 0.0 16.0⫾ 0.4 58.8⫾ 1.9 Weight, kg 4.2⫾ 0.2 3.3⫾ 0.2 7.2⫾ 0.1 18.7⫾ 1.6 Growth

rate, g/d NM3 NM 300 ⫾ 18 287 ⫾ 24

1Values are means⫾SE, n⫽ 5 or 6.

2The age of the fetus is the difference between the term and the day of the collection of fetuses.

3NM, not measured.

and 40 cycles at 95°C for 15 s and at 60°C for 1 min. Checks, based on serial dilutions of each cDNA, proved that the efficiencies of the two PCR were 100% and that the above amplification primers and TaqMan probes were highly specific to their templates. For each amplicon, three PCR reactions were carried out on two independent reverse transcrip-tions, and each assay was performed in triplicate. The intra-assay and interassay coefficients of variations of the CTwere⬍1 and 2%,

respec-tively.

Metabolites and hormone analyses. Plasma BCAA concentra-tions were quantified by liquid chromatography with an automatic amino acid analyzer (Biotronic LC 3000; Roucaire, Ve´lizy, France) using norleucine as an internal standard. Plasma (2 mL) was depro-teinized with 10% trichloroacetic acid. Then amino acids contained in the acid-soluble fraction were purified using the cationic resin Dowex AG 50 (100 –200 mesh).

Plasma BCKA concentrations were determined using reserved-phase HPLC after precolumn derivatization of_{␣-ketoacids with} o-phenylene-diamine as described elsewhere (Koike and Koike 1984, Livesey and Edwards 1985). The internal standard was␣-ketovalerate.

Plasma glucose and lactate concentrations were determined using the colorimetric enzymatic GOD-PAP and MPR3 methods, respec-tively (Boehringer-Mannheim Biochemicals, Indianapolis, IN) and an autoanalyzer (Cobas Mira; Roche Diagnostic Systems, Neuilly-sur-Seine, France). Nonesterified fatty acids were assayed enzymatically using the Nefac C kit (Wako Chemicals, Neuss, Germany). Radio-immunoassays were performed to quantify insulin (ERIA Diagnostics Pasteur, Marnes la Coquette, France), glucagon (Biodata SpA, Rome, Italy) and cortisol (CORT-CT2, Oris Group, Gif-sur-Yvette, France). Plasma free T3 and free T4 were assayed by electrochemilumines-cence (Elecsys; Boehringer-Mannheim).

Statistical analysis. Statistical analyses were performed using Stat-View (version 5.0; SAS Institute, Cary, NC). The level of significance was set at P_{⬍ 0.05. Log-transformed BCAT activity data, which had} homogeneous variances, were analyzed by one- or two-way (muscle type ⫻ developmental stage) ANOVA. When a significant effect was de-tected, differences among groups were compared by the protected least significant difference Fisher test. The effect of the developmental stage on plasma levels and CTdata were evaluated with the nonparametric

Kruskal-Wallis test, and multiple comparison of means was performed with the nonparametric Mann-Whitney U test.

RESULTS

Skeletal muscle chromatography. All BCAT activity found in the 100,000 ⫻ g supernatant obtained from sheep longissimus dorsi was retained by the ion exchange Fractogel EMD DEAE column. Two peaks of BCAT activity eluted from the column at 150 and 400 mmol NaCl/L, indicating that the two BCAT isoenzymes are expressed in skeletal muscle from sheep (Fig. 2). The peak eluting at the lowest salt

concentra-tion corresponded to BCATm, because the sheep placental BCATm activity eluted at 150 mmol/L when the same pro-tocol was performed (Faure et al. 1999). In agreement with the chromatographic data obtained for rat BCAT isoenzymes (Hall et al. 1993, Wallin et al. 1990), sheep BCATc activity eluted at the highest salt concentration.

Immunoprecipitation validation. The antiserum used to immunoprecipitate BCATm was raised against purified sheep BCATm (Faure et al. 1999). As shown in Fig. 3, the addition of increasing volumes of the antiserum to 200 ␮L sheep placenta homogenate, which exhibits only BCATm, resulted in a gradual neutralization of the enzyme. Complete neutral-ization was obtained with 20␮L of antiserum. Then, 25 ␮L of antiserum were used in all further assays to ensure quantitative immunoprecipitation of BCATm. Using homogenates of sheep longissimus dorsi in which both isoenzymes are ex-pressed, we verified that the antiserum immunoprecipitated BCATm specifically and accurately. Indeed, the percentage of BCAT activity neutralized by the antiserum (25.8 ⫾ 3.4, n ⫽ 3) was the same as the percentage of BCAT activity recovered in the BCATm peak from DEAE ion exchange chromatography (24.7 ⫾ 2.1, n ⫽ 3, Fig. 2). Thus, when BCAT isoenzymes are in the native conformation, the anti-serum is specific for BCATm, and the immunoprecipitation method accurately determines the relative contribution of the two BCAT isoenzymes.

BCAT specific activity. Muscle type and developmental stage affected the specific activity of BCAT with a significant interaction between the two variables (Fig. 4). The ranking

FIGURE 1 Partial nucleotide sequence of sheep mitochondrial and cytosolic branched-chain amino acid aminotransferase [BCATm and BCATc, respectively (shown as bcatc and bcatm)] cDNA and positioning of the amplification primers and the TaqMan probes used for the real-time polymerase chain reaction assays. The sequence accession numbers are AF050173 and AF184916 for BCATm and BCATc, respectively. Amplification primers are underlined, and TaqMan probes are double underlined.

FIGURE 2 Chromatographic separation of mitochondrial and cy-tosolic branched-chain amino acid aminotransferase (BCATm and BCATc, respectively) from sheep skeletal muscle. The fetal longissimus dorsi supernatant was loaded onto the Fractogel EMD DEAE 650 (S) column. BCATm and BCATc activities were eluted at 150 and 400 mmol NaCl/L, respectively.

FIGURE 3 Antiserum neutralization of mitochondrial branched-chain amino acid aminotransferase in sheep placenta homogenate. The 10,000⫻ g sheep placenta supernatant was incubated with increasing volumes of mitochondrial branched-chain amino acid aminotransferase antiserum as described in Materials and Methods. The residual enzyme activity was measured in the supernatant obtained after immunopre-cipitation.

order of the muscles, starting with the highest BCAT specific activity, was roughly as follows: DIA, LD, SM, TFL and ST. DIA exhibited a significantly higher activity than SM, TFL and ST at all developmental stages, whereas differences be-tween DIA and LD were significant only in preruminant and ruminant sheep. BCAT specific activity was higher in LD than in ST at all stages of development; whereas SM and TFL activity levels did not differ, except in ruminants. In all mus-cles, BCAT specific activity decreased during development. Differences among the four developmental stages were signif-icant in all muscles studied, except that BCAT specific activ-ity levels did not differ between fetal and newborn sheep in DIA, SM, TFL or ST. From fetal to ruminant stages, BCAT specific activities decreased by 83, 78, 68, 67 and 62% in LD, TFL, ST, SM and DIA, respectively.

Muscle type and developmental stage also affected the

specific activity of both BCAT isoenzymes, with a significant interaction between the two variables (Fig. 4). Differences between muscles observed for BCATc and BCATm specific activities were less dramatic than those for the total activity. For both isoenzymes, the ST activities were significantly lower than the DIA activities at all developmental stages. The specific activity of both BCATc and BCATm varied signifi-cantly during the development of sheep in all studied muscles, except BCATm specific activity in ST. However, BCATc and BCATm specific activity variations were not parallel with the developmental decrease being more gradual and greater for BCATc than for BCATm. Differences in BCATc specific activity were significant among the four developmental stages, except that the DIA and ST data from the newborns did not differ from their respective values in the fetal and preruminant sheep. Further ST data from the newborn and fetal sheep did not differ. From fetal to ruminant stages, BCATc specific activity decreased significantly by 92, 88, 83, 80 and 77% in LD, TFL, SM, ST and DIA, respectively. BCATm specific activities of LD, TFL, ST, DIA and SM from the ruminant sheep were 56, 46, 46, 42 and 36% lower than that in respec-tive muscles from the fetal sheep.

Muscle BCAT capacity. The BCAT capacity of each muscle [nmol/(min䡠 100 g body)] was evaluated by taking into account the contribution of the weight of each muscle to the body weight and the BCAT specific activity (Fig. 5). Devel-opmental stage affected the BCAT capacity for all muscles studied, except ST. DIA, LD, SM and TFL capacities were significantly lower in ruminant sheep than in sheep at other developmental stages. TFL capacity was significantly lower in the preruminant than in the fetal and newborn sheep. Sur-prisingly, LD capacity was lower in the newborn than in the fetal and preruminant sheep. The sum of the capacities of the five muscles was also affected by the developmental stage; it was⬃60% lower in the ruminant sheep than in the younger developmental stages.

BCATc and BCATm mRNAs. The quantification of

BCATc and BCATm mRNAs was carried out for LD because it exhibited the greatest decrease in specific activity of both BCAT isoenzymes during development. The levels of BCATc and BCATm mRNA did not change in LD during the devel-opment of sheep (Table 2). In addition, there was no corre-lation between the activity of each BCAT isoenzyme and its respective mRNA level. The mRNA measurements were con-ducted using the same conditions for BCATc and BCATm, and the PCR efficiencies were the same. The fact that the CT

FIGURE 4 Branched-chain amino acid aminotransferase (BCAT), mitochondrial BCAT (BCATm) and cytosolic BCAT (BCATc) specific activities in skeletal muscle during development in sheep. Bars are means⫾SE, n⫽ 6 (fetus and newborn) or 5 (preruminant and ruminant). For each activity, values not sharing a common letter are significantly different, P⬍ 0.05. DIA, diaphragma; LD, longissimus dorsi; SM, semi-membranosus; TFL, tensor fascia latae; ST, semitendinosus.

FIGURE 5 Branched-chain amino acid aminotransferase (BCAT) capacity in skeletal muscle during development in sheep. Data were calculated from the BCAT specific activity expressed by g of muscle and the percent muscle weight per total body weight (g/100 g body). For each muscle, areas are means andSE, n⫽ 6 (fetus and newborn) or 5 (preruminant and ruminant). Total areas not sharing a letter are significantly different, P ⬍ 0.05. DIA, diaphragma; LD, longissimus dorsi; SM, semimembranosus; ST, semitendinosus; TFL, tensor fascia latae.

values were similar for both isoenzymes indicated that the levels of the two mRNAs were in the same range.

Plasma BCAA and BCKA. With the exception of

leucine, developmental stage affected plasma BCAA and BCKA concentrations (Fig. 6). Plasma concentrations of va-line, leucine and isoleucine were 140, 100 (not significant due to high variability, P⫽ 0.06) and 50% (P ⫽ 0.26) higher in the newborn than in the fetal sheep, respectively, and returned to their respective fetal values in growing sheep. Similarly, plasma concentrations of KIV, KIC and KMV were 240, 246 and 126% (P⬍ 0.05) higher in the newborn than in the fetal sheep, respectively, and returned to their respective fetal levels in both preruminant and ruminant sheep. However, plasma valine and KMV concentrations were significantly higher in the ruminant than in the preruminant sheep.

Plasma metabolites and hormones. Developmental stage affected plasma concentrations of glucose, lactate, nonesteri-fied fatty acids, insulin, cortisol and T3 (Table 3). Plasma concentration of glucose was low in fetal sheep, was 261% greater in the newborn and was maintained at this level in the preruminant and ruminant sheep. In contrast, plasma lactate level was high in the fetal sheep and was 62% lower in the newborn and older sheep. Plasma free nonesterified fatty acid concentration increased by 13-fold at birth and then declined progressively to reach the fetal level in the ruminant sheep. Plasma concentrations of insulin, cortisol and T3 were 310 (not significant due to high variability, P ⫽ 0.08), 231 and 487% higher in newborn compared with fetal sheep, respec-tively. Insulin levels were maintained at a high level in the preruminant and ruminant sheep, whereas cortisol and T3 decreased. Variations in plasma metabolites and hormones were consistent with the nutritional and physiological modi-fications occurring in ruminant species, but there were no correlations between plasma concentrations of any metabolite or hormone, and BCAT activities.

DISCUSSION

Here we present the first in-depth investigation of the developmental changes in BCAT activity in five sheep skel-etal muscles of different metabolic types. The activity of BCAT was high in the most oxidative muscle (DIA), inter-mediate in LD and low in SM, ST and TFL (the most glyco-lytic muscle). This is consistent with data obtained in rats (Yang and Birkhahn 1997), but the reason for such a differ-ence is unknown. In all muscle types, the specific activity of BCAT decreased during the development of sheep, as previ-ously reported (Bergen et al. 1988, Goodwin et al. 1987).

Our results obtained using complementary techniques

es-tablished that sheep differ from rats and humans. Indeed, sheep skeletal muscle expresses both BCAT isoenzymes, whereas BCATm is the sole isoenzyme expressed in skeletal muscle of rats and humans (Hutson et al. 1992, Suryawan et al. 1998). The DEAE chromatography of a homogenate of sheep muscle clearly showed two peaks corresponding to the two BCAT isoenzymes. The simultaneous expression of these BCAT isoenzymes was confirmed by the presence of BCATc and BCATm mRNA in sheep LD. The reason for such a species difference is unknown, but differences between mono-gastric and ruminant promoters and/or tissue specific transcrip-tion factors may be involved. The pattern of expression of BCAT isoenzymes in sheep muscle raises questions concerning their physiological roles. BCKA formed in the cytosol could either be released from the muscle or be transferred to and then decarboxylated in the mitochondria. BCKA formed in the mitochondria could be directly decarboxylated or trans-ported out of the mitochondria.

It is important to note that the two BCAT isoenzymes do not change exactly the same during the development of sheep. That means that the decrease in their activities is not due to a passive dilution due to increases in the size and protein content of muscle fibers but rather due to specific regulatory mechanisms. Variation in BCAT activity reflects variation in BCAT protein (no regulation of activity at the protein level has been ever described), so activity is dependent on the rates of synthesis and degradation of the protein.

The specific activity of BCATc decreases at each develop-mental step and is almost quantitatively responsible for the decrease of BCAT activity. The mechanism involved appears to be independent of its mRNA level (the results obtained for LD are likely valid for the other muscles). The translation of BCATc may be decreased and/or its degradation may be in-creased in skeletal muscles during the development of sheep.

FIGURE 6 Plasma branched-chain amino acid (BCAA) and branched-chain␣-ketoacid (BCKA) concentrations during development in sheep. Bars are means ⫾ SE, n ⫽ 6 (fetus and newborn) or 5 (preruminant and ruminant). For each metabolite, values not sharing a letter are significantly different, P⬍ 0.05. VAL, valine; LEU, leucine; ILE, isoleucine; KIV,␣-ketoisovalerate; KIC, ␣-ketoisocaproate; KMV, ␣-ke-to-␤-methylvalerate.

TABLE 2

Real-time reverse transcription—polymerase chain reaction for BCATm and BCATc mRNA in longissimus dorsi at

different developmental stages in sheep1

Fetus Newborn Preruminant Ruminant BCATm2 27.84⫾ 0.15 27.86 ⫾ 0.22 27.97 ⫾ 0.11 27.86 ⫾ 0.14 BCATc 29.54_{⫾ 0.46 28.71 ⫾ 0.81 28.46 ⫾ 0.21 29.32 ⫾ 0.20}

1Values are means⫾SE, n⫽ 4–6, of CT(threshold cycle, defined

as the fractional cycle number at which the fluorescence exceeded a threshold limit).

2BCATm and BCATc, mitochondrial and cytosolic branched-chain amino acid aminotransferase, respectively.

Mechanisms involved in these putative regulatory points are unknown. The metabolic importance of the observed reduc-tion of BCATc activity during the development of sheep is unclear. In rats and humans, BCATc is mainly expressed in brain. BCATc is involved in BCAA catabolism, but it may be involved also in other cellular processes related to cell cycle and apoptosis (Eden and Benvenisty 1999, Kholodilov et al. 2000). The existence of such a function for sheep BCATc in muscle fibers is questionable.

Based on the results obtained for LD, developmental vari-ations in the specific activity of BCATm probably are not due to BCATm mRNA variations. Therefore, the points of regu-lation of BCATm likely are posttranscriptional as for BCATc. However, the mechanisms are probably specific to each isoen-zyme because the variations in BCATm and BCATc activity follow different patterns. We have previously shown that in sheep, the high BCAT activity observed in placenta is asso-ciated with a higher level of BCATm mRNA than in muscle (Faure et al. 1999). Rats adapted to a 50% casein diet exhibit higher BCAT activity in muscle, kidney and brain than rats fed 6, 18 or 35% casein diets, but only muscle had a signifi-cantly higher level of BCATm mRNA (Torres et al. 1998). Lactation induces an increase in the expression of the BCATm gene in the mammary gland of rats (DeSantiago et al. 1998). Taken together, these results indicate that the regula-tion of both rat and sheep BCATm is tissue specific and may involve transcriptional and posttranscriptional mechanisms. The signaling is not known, and in the present study, none of the metabolites and hormones measured seemed to be in-volved.

Despite a gradual decrease in the specific activity during the development of sheep, the BCAT capacity of the muscles studied decreased only at the ruminant stage. Considering that all the skeletal muscles combined account for⬃40% of body weight, the whole body muscle BCAT capacity would be⬃4.8 ␮mol/(min 䡠 100 g body) in the newborns through prerumi-nants and only 1.8␮mol/(min 䡠 100 g body) in ruminants. In vivo, both deamination and reamination occur, but this futile cycle is less intense in the ruminant, except at the fetal stage, than in monogastric animals. This is consistent with the fact that in food-deprived sheep, the amount of alanine released from the hindlimb muscle is much lower than values reported for humans (see Teleni 1993). Only 34 – 42% of metabolized leucine is transaminated to KIC and 6 –18.5% of metabolized KIC is reaminated to leucine in preruminant or ruminant sheep (Nissen and Ostaszewski 1985, Oddy and Lindsay 1986, Pell et al. 1986), whereas these values reach 50 – 80% and 60 –90%, respectively in humans, pigs or dogs (Helland et al.

1986, Matthews et al. 1981, Nissen and Haymond 1981). In sheep fetus, which catabolizes a large fraction of leucine that enters the placenta (Kennaugh et al. 1987, van Veen et al. 1987), 68% of metabolized leucine is transaminated to KIC (Lietchy et al. 1992), leading to an absolute rate of leucine deamination of 1.2␮mol/(min 䡠 100 g body). For comparison, this rate is 0.3␮mol/(min 䡠 100 g tissue) when measured in the intact hindlimb of the preruminant (Oddy and Lindsay 1986) and only 0.01– 0.03␮mol/(min 䡠 100 g body weight) (Nissen and Ostaszewski 1985, Pell et al. 1986) in the ruminant. It is clear that the in vivo rate of BCAA transamination depends on the concentrations of BCAA, BCKA, glutamate and ␣-ke-toglutarate, as well as the rate of product removal for each tissue. However, the variation of BCAT activity in skeletal muscle, which is the main in vivo site of BCAA transamina-tion, probably contributes significantly to the decrease in the in vivo rate of BCAA transamination. It is more difficult to determine the consequences of the modification of the pattern of expression of BCAT isoenzymes. The lower plasma, and probably muscle, BCKA concentrations in sheep compared with those in monogastric animals could contribute to the lower rate of reamination of BCKA. Sheep skeletal muscle has a much lower ability to oxidize BCAA to CO₂than rat muscle (Coward and Buttery 1979). More of the BCAA, at least for leucine, may be used for lipogenesis because this amino acid is more lipogenic in cattle muscle than in rat or pig muscles (Vovk and Yanovich 1990).

In conclusion, sheep skeletal muscle exhibits the unique expression of both BCATc and BCATm at all developmental stages. BCATc is mainly responsible for the developmental decrease in BCAT activity. The expression of both enzymes appears to be regulated at the posttranscriptional level during the development, but the mechanisms involved are unknown. The present results are consistent with a lower in vivo BCAA oxidation rate in ruminant than in the fetal sheep.

LITERATURE CITED

Bergen, W. G., Busboom, J. R. & Merkel, R. A. (1988) Leucine degradation in sheep. Br. J. Nutr. 59: 323–333.

Bonfils, J., Faure, M., Gibrat, J.-F., Glomot, F. & Papet, I. (2000) Sheep cytosolic branched-chain amino acid aminotransferase: cDNA cloning, pri-mary structure and molecular modelling and its unique expression in muscles. Biochim. Biophys. Acta 1494: 129 –136.

Briand, M., Talmant, A., Briand, Y., Monin, G. & Durand, R. (1981) Metabolic types of muscle in the sheep: I. myosin ATPase, glycolytic, and mitochondrial enzyme activities. Eur. J. Appl. Physiol. 46: 347–358.

Cappuccino, C. C., Kadowaki, H. & Knox, W. E. (1978) Assay of leucine aminotransferase in rat tissues and tumors. Enzyme 23: 328 –338. Chomczynski, P. & Sacchi, N. (1987) Single-step method of RNA isolation by

TABLE 3

Plasma metabolite and hormone concentrations at different developmental stages in sheep1

Fetus Newborn Preruminant Ruminant

Glucose, mmol/L 1.6⫾ 0.5c 5.9⫾ 1.4ab 6.7⫾ 0.2a 5.6⫾ 0.2b Lactate, mmol/L 11.7_{⫾ 0.9}a 4.5_{⫾ 0.6}b 3.6_{⫾ 0.2}b 2.8_{⫾ 0.4}b NEFA,2_␮mol/L 40 ⫾ 4c 574 ⫾ 63a 256 ⫾ 64b 64 ⫾ 9c Insulin, pmol/L 93 _{⫾ 11}b 383 _{⫾ 136}ab 417 _{⫾ 175}a 243 _{⫾ 48}a Glucagon, pmol/L 91 ⫾ 6 134 ⫾ 18 130 ⫾ 22 98 ⫾ 12 Cortisol, nmol/L 49 _{⫾ 7}b 163 _{⫾ 59}a 45 _{⫾ 9}bc 25 _{⫾ 7}c T3, pmol/L 4.5⫾ 0.5c 26.3⫾ 3.7a 19.5⫾ 0.9a 7.8⫾ 0.7b T4, pmol/L 38.1_{⫾ 4.6} 40.7_{⫾ 5.8} 36.6_{⫾ 2.8} 30.3_{⫾ 0.6}

1Values are means⫾SE, n⫽ 6 (fetus and newborn) or 5 (preruminant and ruminant). Values in a row not sharing a superscript letter are different,

P_{⬍ 0.05.}

acid guanidinium thiocyanate-phenol-chloroform extraction. Anal. Biochem. 162: 156 –159.

Coward, B. J. & Buttery, P. J. (1979) Catabolism of branched chain amino acids by ruminant muscle. Proc. Nutr. Soc. 38: 139A.

DeSantiago, S., Torres, N., Suryawan, A., Tovar, A. R. & Hutson, S. M. (1998) Regulation of branched-chain amino acid metabolism in the lactating rat. J. Nutr. 128: 1165–1171.

Eden, A. & Benvenisty, N. (1999) Involvement of branched-chain amino acid aminotransferase (Bcat1/Eca39) in apoptosis. FEBS Lett. 457: 255–261. Faure, M., Glomot, F., Bledsoe, R., Hutson, S. & Papet, I. (1999) Purification

and cloning of the mitochondrial branched-chain amino acid aminotransfer-ase from sheep placenta. Eur. J. Biochem. 259: 104 –111.

Finkelstein, D. I., Andrianakis, P., Luff, A. R. & Walker, D. W. (1992) Develop-mental changes in hindlimb muscles and diaphragm of sheep. Am. J. Physiol. 263: R900 –R908.

Goodwin, G. W., Gibboney, W., Paxton, R., Harris, R. A. & Lemons, J. A. (1987) Activities of chain amino acid aminotransferase and branched-chain 2-oxo acid dehydrogenase complex in tissues of maternal and fetal sheep. Biochem. J. 242: 305–308.

Hall, T. R., Wallin, R., Reinhart, G. D. & Hutson, S. M. (1993) Branched chain aminotransferase isoenzymes: purification and characterization of the rat brain isoenzyme. J. Biol. Chem. 268: 3092–3098.

Harper, A. E., Miller, R. H. & Block, K. P. (1984) Branched-chain amino acid metabolism. Annu. Rev. Nutr. 4: 409 – 454.

Harris, R. A., Hawes J. W., Popov, K. M., Zhao, Y., Shimomura, Y., Sato, J., Jaskiewicz, J. & Hurley, T. D. (1997) Studies on the regulation of the mitochondrial alpha-ketoacid dehydrogenase complexes and their kinases. Adv. Enzyme Regul. 37: 271–293.

Heid, C. A., Stevens, J., Livak, K. J. & Williams, P. M. (1996) Real time quantitative PCR. Genome Res. 6: 986 –994.

Helland, S. J., Ewan, R. C., Trenkle, A. & Nissen, S. (1986) In vivo metabolism of leucine and alpha-ketoisocaproate in the pig: influence of dietary glucose or sucrose. J. Nutr. 116: 1902–1909.

Holland, P. M., Abramson, R. D., Watson, R. & Gelfand, D. H. (1991) Detection of specific polymerase chain reaction product by utilizing the 5⬘—3⬘ exonu-clease activity of Thermus aquaticus DNA polymerase. Proc. Natl. Acad. Sci. USA 88: 7276 –7280.

Hutson, S. M., Wallin, R. & Hall, T. R. (1992) Identification of mitochondrial branched chain aminotransferase and its isoforms in rat tissues. J. Biol. Chem. 267: 15681–15686.

Kadowaki, H. & Knox, W. E. (1982) Cytosolic and mitochondrial isoenzymes of branched-chain amino acid aminotransferase during development of the rat. Biochem. J. 202: 777–783.

Kennaugh, J. M., Bell, A. W., Teng, C., Meschia, G. & Battaglia, F. C. (1987) Ontogenetic changes in the rates of protein synthesis and leucine oxidation during fetal life. Pediatr. Res. 22: 688 – 692.

Kholodilov, N. G., Neystat, M., Oo, T. F., Hutson, S. M. & Burke, R. E. (2000) Upregulation of cytosolic branched chain aminotransferase in substantia nigra following developmental striatal target injury. Brain Res. Mol. Brain. Res 75: 281–286.

Koike, K. & Koike, M. (1984) Fluorescent analysis of alpha-keto acids in serum and urine by high-performance liquid chromatography. Anal. Biochem. 141: 481– 487.

Lacourt, A. & Arnal, M. (1974) Evolution en fonction de l’aˆge des

caracte´ris-tiques me´taboliques de muscles d’agneaux en relation avec la synthe`se prote´ique in vivo. Proceedings of the 20th European Meeting of Meat Re-search Workers, pp 226 –229. 1974, Dublin, Ireland.

Liechty, E. A., Barone, S. & Nutt, M. (1987) Effect of maternal fasting on ovine fetal and maternal branched-chain amino acid transaminase activities. Biol. Neonate 52: 166 –173.

Livesey, G. & Edwards, W.T.E. (1985) Quantification of branched-chain alpha-keto acids as quinoxalinols: importance of excluding oxygen during derivati-zation. J. Chromatogr. 337: 98 –102.

Matthews, D. E., Bier, D. M., Rennie, M. J., Edwards, R. H., Halliday, D., Millward, D. J. & Clugston, G. A. (1981) Regulation of leucine metabolism in man: a stable isotope study. Science 214: 1129 –1131.

National Research Council (1985) Guide for the Care and Use of Laboratory Animals. Publication no. 85–23 (rev), NIH, Washington, D.C.

Nissen, S. & Haymond, M. W. (1981) Effects of fasting on flux and intercon-version of leucine and alpha-ketoisocaproate in vivo. Am. J. Physiol. 241: E72–E75.

Nissen, S. & Ostaszewski, P. (1985) Effects of supplemental dietary energy on leucine metabolism in sheep. Br. J. Nutr. 54: 705–712.

Oddy, V. H. & Lindsay, D. B. (1986) Determination of rates of protein synthesis, gain and degradation in intact hind-limb muscle of lambs. Biochem. J. 233: 417– 425.

Papet, I., Grizard, J., Bonin, D. & Arnal, M. (1992) Regulation of branched chain amino acid metabolism in ruminants. Diabetes Metab. 18: 122–128. Papet, I., Lezebot, N., Barre, F., Arnal, M. & Harper, A. E. (1988) Influence of

dietary leucine content on the activities of branched-chain amino acid ami-notransferase (EC 2.6.1.42) and branched-chain alpha-keto acid dehydroge-nase (EC 1.2.4.4) complex in tissues of preruminant lambs. Br. J. Nutr. 59: 475– 483.

Pell, J. M., Caldarone, E. M. & Bergman, E. N. (1986) Leucine and alpha-ketoisocaproate metabolism and interconversions in fed and fasted sheep. Metabolism 35: 1005–1016.

Suryawan, A., Hawes, J. W., Harris, R. A., Shimomura, Y., Jenkins, A. E. & Hutson, S. M. (1998) A molecular model of human branched-chain amino acid metabolism. Am. J. Clin. Nutr. 68: 72– 81.

Teleni, E. (1993) Catabolism and synthesis of amino acids in skeletal muscle: their significance in monogastric mammals and ruminants. Aust. J. Agric. Res. 44: 443– 461.

Torres, N., Lopez, G., De Santiago, S., Hutson, S. M. & Tovar, A. R. (1998) Dietary protein level regulates expression of the mitochondrial branched-chain aminotransferase in rats. J. Nutr. 128: 1368 –1375.

van Veen, L. C., Teng, C., Hay, W. W., Jr., Meschia, G. & Battaglia, F. C. (1987) Leucine disposal and oxidation rates in the fetal lamb. Metabolism 36: 48 –53. Vovk, S. I. & Ianovich, V. G. (1990) The lipogenic role of leucine in animal

skeletal muscles. Biokhimiia 55: 2090 –2094.

Wallin, R., Hall, T. R. & Hutson, S. M. (1990) Purification of branched chain aminotransferase from rat heart mitochondria. J. Biol. Chem. 265: 6019 – 6024.

Wijayasinghe, M. S., Milligan, L. P. & Thompson, J. R. (1983) In vitro degra-dation of leucine in muscle, adipose tissue, liver, and kidney of fed and starved sheep. Biosci. Rep. 3: 1133–1140.

Yang, Q. & Birkhahn, R. H. (1997) Branched-chain transaminase and keto acid dehydrogenase activities in burned rats: evidence for a differential adaptation according to sex. Nutrition 13: 640 – 645.