Ontogeny and subcellular localization of rat liver mitochondrial
branched chain amino-acid aminotransferase
Nimbe Torres
1
, Carolina Vargas
1
, Rogelio Herna
´
ndez-Pando
2
,He
´
ctor Orozco
2
, Susan M. Hutson
3
and
Armando R. Tovar
1
1
Departamento de Fisiologı
´
a de la Nutricio
´
n, Instituto Nacional de Ciencias Me
´
dicas y Nutricio
´
n ‘Salvador Zubira
´
n’, Me

of hepatic cells and blood cell precursors; in contrast, adult
liver showed mild immunoreactivity located only in the
mitochondria of hepatocytes. BCAT activity in isolated fetal
liver nuclei was 0.64 mU
:
mg
21
protein whereas it was
undetectable in adult liver nuclei. By Western blot analysis
the BCATm antibody recognized a 41-kDa protein in fetal
liver nuclei, and proteins of 41 and 43 kDa in fetal liver
supernatant. In adult rat liver supernatant, the BCATm
antibody recognized only a 43-kDa protein; however,
neither protein was detected in adult rat liver nuclei. The
appearance of the 41-kDa protein was associated with the
presence of the highly active form of BCATm. These results
suggest the existence of active and inactive forms of BCAT
in rat liver.
Keywords: branched-chain amino acids; mitochondria;
nuclei; ontogeny.
The branched-chain amino acids (BCAA) leucine, iso-
leucine, and valine are required mainly for body protein
synthesis. The initial enzymes for catabolism of the BCAA
are regulated differently from other amino-acid degrading
enzymes. The first step in the degradation of these amino
acids is a reversible transamination catalyzed by the
branched chain amino-acid aminotransferase (BCAT;
EC 2.6.1.42). The products of this reaction are the
corresponding branched chain 2-oxo acids that can be
reaminated to their corresponding amino acids [1], or

fetal hepatocytes cultured for 2 days showed BCAT activity,
indicating the possibility that not only the hematopoietic
cells were responsible for BCAT activity but that fetal
hepatocytes may contribute also to the enzyme activity [12].
In the present study, we measured the BCAT activity,
amount of protein and BCATm mRNA expression pattern as
well as the immunolocalization at cellular and subcellular
Correspondence to A. R. Tovar. Departamento de Fisiologı
´
adela
Nutricio
´
n, Instituto Nacional de Ciencias Me
´
dicas Nutricio
´
n ‘Salvador
Zubira
´
n’, Me
´
xico DF 14300, Me
´
xico. Fax: 1 525 6551076,
Tel.: 1 525 5731200 ext. 2801/2802,
E-mail:
Enzymes: branched chain amino acid aminotransferase (BCAT, EC
2.6.1.42).
(Received 25 June 2001, revised 27 September 2001, accepted
28 September 2001)

A sample of liver, kidney or heart was suspended in buffer
(4 mL extraction buffer per gram tissue) containing 225 m
M
mannitol, 75 mM sucrose, 0.1 mM EDTA, 5 mM Mops and a
mix of protease inhibitors including 1 m
M EDTA, 1 mM
EGTA, 1 mM diisopropylfluorophosphate, 5 mM benzami-
dine, 5 m
M dithiothreitol, 10 mg
:
mL
21
leupeptin and 1%
Triton X-100. Supernatant fraction of fetal liver or heart was
obtained from a pool of 19–24 fetuses. The tissue
suspension was centrifuged at 30 000 g for 60 min at
4 8C. The supernatant was assayed for BCAT activity.
Isolation of nuclei
Nuclei were isolated as described [13]. Liver was
homogenated in 10 m
M Tris/HCl pH 7.5 containing 0.3 M
sucrose, 5 mM dithiothreitol and 0.05% triton X-100. After
centrifugation at 83 000 g for 45 min with an 70 Ti rotor at
4 8C through a cushion of 2.3
M sucrose, 2 mM MgCl
2
,
10 m
M Tris/HCl pH 7.5, nuclei were counted and suspended
at a concentration of 2 Â 10

14
C]isocaproate
was 3.3 Bq
:
nmol
21
. After 5 min the reaction was stopped
by addition of 500 mLof2
M sodium acetate pH 3.4. The
remaining 2-oxo[1-
14
C]isocaproate not transaminated was
chemically decarboxylated by adding 250 mL of 30%
hydrogen peroxide. A sample of 250 mL of the reaction
mixture was added to a scintillation vial. Then 10 mL of
liquid scintillation cocktail (BCS, Amersham) was added
and samples were counted (Wallac, Turku, Finland). Each
assay was performed in duplicate. A unit of activity was
defined as 1 mmol [1-
14
C]leucine formed per min at 37 8C.
BCAT specific activity was expressed as mU
:
mg protein
21
.
SDS/PAGE
SDS/PAGE was carried out according to Ausubel et al. [13]
in 10% gels using 40 mg of protein. Prior to electrophoresis,
all samples were boiled for 5 min in the presence of 4%

transferred to a nylon membrane filter Hybond-N
1
(Amersham) and cross-linked with a UV crosslinker
(Amersham). The probe was a 900-bp Pst1–Eco R1
fragment of rat BCATm cDNA cloned in pT7 Bluescript
[6] and labeled with deoxycytidine 5
0
[a-
32
P]dCTP
(3000 Ci
:
mmol
21
, Amersham) using the rediprime DNA
labeling system (Amersham). Filters were prehybridized
with rapid-hyb buffer (Amersham) at 65 8C for 45 min, and
then hybridized with the labeled probe for 2.5 h at 65 8C.
Membranes were washed once with 2 Â NaCl/Cit, 0.1%
SDS at room temperature for 20 min and then washed twice
with 0.1 Â NaCl/Cit, 0.1% SDS at 65 8C for 15 min each.
Digitized images were prepared and quantitation of
radioactivity in the bands was carried out by using the
Instant Imager electronic autoradiography system (Packard
Instruments). Membranes were also exposed to Extascan
film (Kodak) at 270 8C with an intensifying screen.
q FEBS 2001 Nuclear BCAT in fetal liver (Eur. J. Biochem. 268) 6133
RT/PCR
Reverse transcription (RT)/PCR was performed with 5 mg
RNA from rat fetal or adult liver, and kidney. Total RNA

RACE was carried out according to
manufacturer instructions (Life Technologies). Gene
specific primers, including nested primers, were designed
based on the RT/PCR product of BCATm amplified from
adult rat liver. The external and nested gene specific reverse
primers for the 5
0
RACE amplification were: 5
0
-GGCGTA
CCTGCTTGTCTCTGC-3
0
and 5
0
-CAAAGAGCTGCAAT
GAGTAGT-3
0
corresponding to nucleotides 339–359 and
297–317 of rat heart BCATm cDNA, respectively. The
external and nested gene specific forward primers for the 3
0
RACE amplification were: 5
0
-CAGAAGGAGTTGAAGG
CTATT-3
0
and 5
0
-ACGGAACCAGTGCCCACGATT-3
0

primary antibody, the endogenous peroxidase activity was
quenched with 0.03% H
2
O
2
in absolute methanol; liver
sections were then incubated with the primary antibody
diluted 1 : 500 in NaCl/P
i
overnight at 4 8C. Bound
antibodies were detected with goat anti-(rabbit IgG) Ig
labeled with peroxidase diluted 1 : 100 in NaCl/P
i
and
diaminobenzidine. For negative controls tissue was
incubated with primary antibody previously pre-adsorbed
with purified enzyme.
For immunoelectron microscopy studies, small tissue
fragments of fetal and adult liver were fixed by immersion in
4% paraformaldehyde dissolved in Sorensen’s buffer pH 7.4
for 2 h at 4 8C. After rinsing, free aldehyde groups were
blocked in 0.5
M NH
4
Cl in NaCl/P
i
for 1 h. Tissue samples
were dehydrated in graded ethyl alcohols and embedded in
LR-White hydrosoluble resin. The same fixation and
embedding procedure was used for nuclei isolated from

commercial sources and were at least reagent grade.
RESULTS
Developmental pattern of liver BCAT activity
Maximal BCAT activity, 7.28 mU
:
mg protein
21
, occurred in
fetal liver at 17 days’ gestation. BCAT activity decreased
significantly after birth: by 68% and 94% at birth and 21
days after birth, respectively. Mean BCAT activity in adult
rat liver was 0.38 ^ 0.05 mU
:
mg protein
21
, approximately
2% of that in heart. After day 20 postnatal, liver BCAT
specific activity remained low, similar to the levels reported
in the literature (Fig. 1A).
Fig. 3. Subcellular localization of BCATm in fetal and adult liver
by immunoelectron microscopy. (A) Fetal hepatocytes showed
immunolabeling in mitochondria (m) and chromatin (c) associated to
the nuclear membrane (nm) (Â 50 000). (B) The same pattern of
nuclear immunolabeling was seen in nuclei isolated from fetal liver by
differential ultracentrifugation at 32 000 g. (C) At the structural level,
adult hepatocytes showed immunolabeling in mitochondria (m), and
occasional gold particles were found in cytoplasm (Â 50 000).
Bar ¼ 0.5 mm.
q FEBS 2001 Nuclear BCAT in fetal liver (Eur. J. Biochem. 268) 6135
Immunohistochemical localization of BCATm in fetal rat

fetal liver and adult liver are shown in Fig. 4. Fetal liver
nuclear BCAT specific activity was 0.65 ^ 0.08 mU
:
mg
protein
21
whereas it was undetectable in adult liver nuclei.
These results indicate that approximately 10% of the BCAT
activity in fetal liver is associated with the nuclei. BCAT
specific activity in fetal liver was 19-fold higher than in the
adult liver supernatant, indicating that liver has a high
capacity for transamination of BCAA during the fetal stage,
but that this is lost after birth. Furthermore, BCAT specific
activity in fetal liver is 35% of that found in adult heart
which is considered to be one of the organs with high BCAT
activity.
Western blot analysis of BCATm in fetal and adult rat liver
When equal amounts of protein (40 mg) were subjected to
SDS/PAGE and immunoblotting, a 41-kDa protein corre-
sponding to the active form of BCATm was detected in fetal
liver nuclei, fetal liver supernatant, heart supernatant and
kidney mitochondria. However, this protein was not found in
adult liver nuclei or adult liver supernatant, indicating that
the protein found in fetal liver nuclei was the same of that
found in kidney and heart. The BCATm antiserum
recognized a protein of < 43 kDa in addition to the
41-kDa protein in fetal liver supernatant. In adult liver
supernatant only a faint 43-kDa band was seen (Fig. 4).
Thus, the appearance of the 41-kDa protein on immunoblots
was always associated with the presence of the highly active

obtained by reverse transcription, and BCATm cDNA was amplified by
PCR using primers specific for rat BCATm. The size of the product
obtained was 979 bp. b-actin was used as standard for mRNA integrity.
6136 N. Torres et al. (Eur. J. Biochem. 268) q FEBS 2001
100% homology with the sequence of BCATm heart cDNA.
Furthermore, 5
0
and 3
0
RACE amplified the end terminals of
rat liver cDNA, and a single band for each amplification was
obtained. The sequence of the products of both amplifica-
tions also showed 100% homology with rat heart BCATm
cDNA. These results suggest that the low abundance mRNA
for the 43-kDa protein is possibly derived from the same
gene that produces the 41-kDa protein.
Developmental pattern of heart BCAT
BCAT activity in heart showed a different developmental
pattern than that observed in liver. BCAT activity increased
significantly (P , 0.01) up to day 21 after birth, and it was
25% higher with respect to the activity at birth. On day 21,
the BCAT activity reached the values reported for this organ
in adults rats. Western blot and Northern blot analysis
followed a similar pattern (Fig. 6A, and C).
DISCUSSION
The activity of several hepatic amino-acid degrading
enzymes is absent or low during fetal life, increases rapidly
at birth, and reaches the activity level found in adults from
12 h to several days after birth [17–20]. On the contrary, the
activity of hepatic BCAT followed a different develop-

short stretch of cationic amino acids containing four to six
residues of lysine or arginine [26]. An examination of the
mature BCATm protein showed that it does not contain a
typical consensus sequence for its import to the nuclei.
However this protein contains two cationic rich stretches,
located between amino acids 80 and 90 (KAYKGR
DKQVR) and 290 and 299 (RKVTMKELKR) that may
contribute to the nuclear localization of the enzyme. Perhaps
BCATm protein is transported to the nuclei by a specific
importin [27] that is present only during fetal life.
The high expression of BCATm during fetal life and the
very low branched-chain 2-oxo acid dehydrogenase
complex activity in liver and heart [20] reduce the oxidation
Fig. 6. Developmental pattern of BCAT activity, amount of BCATm
protein and BCATm mRNA levels in rat heart. (A) BCAT activity in
fetal and postnatal heart. The results are expressed as mean ^ SEM,
n ¼ 4 –24. (B) Western blot analysis of BCATm using anti-BCATm.
(C) Northern blot analysis of BCATm mRNA. All lanes contained heart
total RNA from at least four different rats.
Fig. 7. Northern blot analysis of BCATc and BCATm in rat
placenta. Total RNA was isolated from placenta of rats on day 17 and
19 of gestation as described in Materials and methods. Blots were
hybridized with the 900 bp Pst1 Eco R1 fragment of rat BCATm cDNA
or the 1400 bp Eco R1 fragment of rat BCATc cDNA [35].
q FEBS 2001 Nuclear BCAT in fetal liver (Eur. J. Biochem. 268) 6137
of BCAA: there is no need life for the disposal of these
amino acids during fetal. Thus, transamination of branched
chain 2-oxo acids by BCATm may play a specific role in
BCAA conservation which can then be used in protein
synthesis [28] during gestation.

liver is the same as that of heart BCATm cDNA [6]. It
is possible that some step in the processing of the
BCATm protein is inactive in the adult liver, but is
active in fetal liver. Studies in our laboratory are in
progress to elucidate the mechanism of regulation of the
two forms in liver. At the present time, we cannot rule
out the possibility that the 43-kDa protein is responsible
for the low BCAT activity in liver, although it has been
proposed that BCAA are transaminated by the
asparagine aminotransferase in liver.
Although BCATm expression is unresponsive to dietary
protein or hormones (hydrocortisone and glucagon) in
extrahepatic tissues [4], conditions related to cell growth as
in fetal liver [11], growth of hepatocytes in culture [32], and
lactating mammary gland tissue [8,9] stimulates BCATm
activity and expression. There is evidence to support the role
of BCAT in cell growth. Two yeast proteins have been
shown to function as BCAT [33,34]; mutation of one of
these BCAT homologs produces a short G1 stage indicating
that this protein is involved in cell cycle regulation. On
the other hand, the mouse BCATc gene is highly
expressed early in embryogenesis and in several c-myc
based tumors. Thus, BCAT may play additional roles in
situations where high cell proliferation takes place and
perhaps the nuclear localization of BCATm in fetal liver
is involved in one of them. Further studies are required
to clarify the possible role of BCAT in situations of
accelerated cell proliferation.
ACKNOWLEDGEMENT
Financial support was from CONACYT 25637M (to A. R. T.), Me

´
ndez-Pando, R., Lo
´
pez, G.,
Suryawan, A., DeSantiago, S., Hutson, S.M. & Torres, N. (2001)
Localization and expression of BCAT during pregnancy and
lactation in the rat mammary gland. Am. J. Physiol. Endocrinol.
Metab. 280, E480–E488.
10. Ichihara, A. (1975) Isozyme patterns of branched-chain amino acid
transaminase during cellular differentiation and carcinogenesis. In
Carcinofetal Proteins: Biology and Chemistry (Hirai, H. &Alpert,
E., eds), pp. 347– 359. New York Academy of Sciences, New York.
11. 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.
12. Hutson, S., Wallin, R. & Hall, T.R. (1992) Identification of
mitochondrial branched chain aminotransferase and its isoforms in
rat tissues. J. Biol. Chem. 267, 15681–15686.
13. Ausubel, F.M., Brent, R., Kingston, R.E., Moore, D.D., Seidman,
J.G., Smith, J.A. & Struhl, K. (1994). Current Protocols in
Molecular Biology. Wiley, New York.
14. Hutson, S.M. (1988) Subcellular distribution of branched-chain
aminotransferase activity in rat tissues. J. Nutr. 118, 1475–1481.
15. Chomczynski, P. & Sacchi, N. (1987) Single-step method of RNA
isolation by acid guanidinium thiocyanate-phenol-chloroform
extraction. Anal. Biochem. 162, 156–159.
16. Ru
¨
diger, H.W., Langenbeck, U. & Goedde, H.W. (1972) A
simplified method for the preparation of

ase GSTT1-1 in the liver and lung of the mouse. Biochem. J. 329,
431–432.
25. Rose, A.M., Joyce, P.B., Hopper, A.K. & Martin, N.C. (1992)
Separate information required for nuclear and subcellular
localization: additional complexity in localizing an enzyme shared
by mitochondria and nuclei. Mol. Cell. Biol. 12, 5652 –5658.
26. Chelsky, D., Ralph, R. & Jonak, G. (1989) Sequence requirements
for synthetic peptide-mediated translocation to the nucleus. Mol.
Cell. Biol. 9, 2487–2492.
27. Gorlich, D. & Kutay, U. (1999) Transport between the cell nucleus
and the cytoplasm. Annu. Rev. Cell. Dev. Biol. 15, 607–660.
28. Shiota, T., Yagi, M. & Walser, M. (1989) Utilization for protein
synthesis in individual rat organs of extracellular 2 ketoisocaproate
relative to utilization of extracellular leucine. Metabolism 38,
612–618.
29. Palou, A., Arola, L. & Alemany, M. (1977) Plasma amino acid
concentrations in pregnant rats and in 21-day foetuses. Biochem. J.
166, 49– 55.
30. Matthews, J.C., Beveridge, M.J., Malandro, M.S., Rothstein, J.D.,
Campbell-Thompson, M., Verlander, J.W., Kilberg, M.S. & Novak,
D.A. (1998) Activity and protein localization of multiple glutamate
transporters in gestation day 14 vs. day 20 rat placenta. Am.
J. Physiol. 274, C603– C614.
31. Harper, A.E., Miller, R.H. & Block, K.P. (1984) Branched-chain
amino acid metabolism. Ann. Rev. Nutr. 4, 409–454.
32. Ichihara, A., Noda, C. & Tanaka, K. (1981) Oxidation of branched
chain amino acids with special reference to their transaminase. In
Metabolism and Clinical Implications of Branched Chain Amino
and Ketoacids (Walser, M. &Williamson, J.R., eds), pp. 227–231.
Elsevier/North Holland, Amsterdam, the Netherlands.