Branchedchain amino acid catabolism: unique segregation of pathway enzymes in organ systems and peripheral nerves

Branched-Chain Amino Acid Catabolism: Unique Segregation of Pathway Enzymes in Organ Systems and Peripheral Nerves

Andrew J. Sweatt1

Mac Wood1

Agus Suryawan1

Reidar Wallin2

Mark C Willingham3

Susan M. Hutson1

1 Department of Biochemistry Wake Forest University School of Medicine Medical Center Blvd. Winston-Salem NC 27157

2 Department of Internal Medicine Section on Rheumatology

Wake Forest University School of Medicine Medical Center Blvd. Winston-Salem NC 27157

3 Department of Pathology Wake Forest University School of Medicine

Medical Center Blvd., Winston-Salem NC 27157

Running Title: Branched-Chain Amino Acid Catabolic Enzymes

Contact Information: Andrew J. Sweatt Department of Biochemistry Wake Forest University School of Medicine Medical Center Blvd. Winston-Salem NC 27157

Tel 336-713-7201 FAX 336-716-7671

e-mail [email protected]

Copyright (c) 2003 by the American Physiological Society.

Articles in PresS. Am J Physiol Endocrinol Metab (September 9, 2003). 10.1152/ajpendo.00276.2003

2

ABSTRACT

We have examined the localization of the first two enzymes in the branched-chain

amino acid catabolic pathway – the branched-chain aminotransferase (BCAT) isozymes

(mitochondrial BCATm and cytosolic BCATc) and the branched-chain -keto-acid

dehydrogenase (BCKD) enzyme complex. Antibodies specific for BCATm or BCATc were

used to immunolocalize the respective isozymes in cryosections of rat tissues. BCATm was

expressed in secretory epithelia throughout the digestive tract, with the most intense

expression in the stomach. BCATm was also strongly expressed in secretory cells of the

exocrine pancreas, uterus, and testis, as well as in the transporting epithelium of convoluted

tubules in kidney. In muscle, BCATm was located in myofibrils. Liver, as predicted, was not

immunoreactive for BCATm. Unexpectedly, BCATc was localized in elements of the

autonomic innervation of the digestive tract, as well as in axons in the sciatic nerve. The

distributions of BCATc and BCATm did not overlap. BCATm-expressing cells also

expressed the second enzyme of the BCAA catabolic pathway, BCKD. In selected monkey

and human tissues examined by immunoblot and/or immunohistochemistry, BCATm and

BCATc were distributed in patterns very similar to those found in the rat. The results show

that BCATm is in a position to regulate BCAA availability as protein precursors and anabolic

signals in secretory portions of the digestive and other organ systems. The unique expression

of BCATc in neurons of the peripheral nervous system, without co-expression of BCKD,

raises new questions about the physiological function of this BCAT isozyme.

Keywords not in title: digestive system, human, leucine, monkey, rat

1

INTRODUCTION

In the body, the nutritionally indispensable branched chain amino acids (BCAAs)

serve a number of important metabolic functions. BCAAs are key nitrogen donors for the

synthesis of the metabolically significant dispensable amino acids glutamine and alanine.

Glutamine is an important energy substrate for the gastrointestinal tract (38). Glutamine and

alanine are also the major carriers of nitrogen from amino acid oxidation in skeletal muscle

to the liver (7, 20, 33, 48, 56). In the central nervous system, BCAAs are thought to

participate in an intercellular shuttle between neurons and astroglia that provides nitrogen for

synthesis of the excitatory amino acid glutamate (3, 4, 31, 39, 40, 64). In addition to the role

of BCAAs in nitrogen metabolism, the BCAA leucine serves as an anabolic nutritional

signal. Leucine stimulates protein synthesis in selected tissues via activation of the

ribosomal protein S6 kinase-1 (12, 19, 42, 61). Furthermore, high physiologic concentrations

of leucine stimulate secretion of insulin, and it has been postulated that this effect occurs in

part via activation of glutamate dehydrogenase (43, 52).

The initial reaction in the degradation of most indispensable amino acids is essentially

irreversible, with excess amino acids being oxidized primarily in the liver. Thus, the major

fate of indispensable amino acids in peripheral tissues is synthesis into tissue proteins.

Catabolism of the BCAAs differs markedly from that of the other indispensable amino acids,

in that the first step is reversible. In addition, BCAA catabolic enzymes appear to be

distributed throughout the body, including in tissues of the digestive tract (18, 28, 36). The

physiological significance of BCAA metabolism in tissues other than skeletal muscle and

brain is not understood.

2

The first step in breakdown of the BCAAs is the reversible transfer of the -amino

group to -ketoglutarate to form glutamate and the respective branched chain -keto-acids in

a reaction catalyzed by the branched chain aminotransferase (BCAT) isozymes

(mitochondrial BCATm and cytosolic BCATc; reviewed in 23, 35). The next and first

irreversible step in BCAA catabolism is the oxidative decarboxylation of the branched chain

-keto-acid products of the transamination reaction. This step is catalyzed by the

mitochondrial branched chain -keto-acid dehydrogenase multi-enzyme complex (BCKD),

which contains multiple copies of three enzymes: a branched chain -keto-acid

decarboxylase (E1); a dihydrolipoyl transacylase (E2); and a dihydrolipoyl dehydrogenase

(E3). Activity of BCKD is regulated by phosphorylation/dephosphorylation of the E1

subunit, (21, 49). Based on observed tissue-specific differences in the activity of BCAT and

the BCKD complex, it is thought that oxidation of BCAA involves extensive movement of

metabolites between tissues (10, 24, 28, 30, 34, 53, 56).

The distribution of the BCAT isozymes in different tissues has been determined from

measurements of enzyme activity and by Western blot analysis of BCAT proteins. In the rat,

BCATm is found in most tissues, with very high BCAT activity found in the stomach,

pancreas, and salivary glands (36, 56). BCATc appears to have more limited expression than

BCATm. BCATc activity has been identified in rat brain, ovary, and placenta (18, 35). In

the rat, BCKD activity is found in most organs, with highest activities occurring in the liver

and kidney (56). In other organs, including the stomach, intestine and brain, BCKD activities

are up to two orders of magnitude lower than for liver and kidney (56).

The localization of the BCAA catabolic enzymes to particular cell types within a

tissue has been investigated only for cells of the central nervous system. Immunostaining of

cell cultures derived from rat brain revealed that BCAT isozymes are differentially

3

expressed, with BCATm expressed in astroglia, and BCATc expressed in neurons (4, 31, 40).

In contrast, BCKD appeared to be expressed in both cell types (3). In this study, we report

on the immunolocalization of BCAA catabolic enzymes to specific cell types in tissues

known to have BCAA catabolic activities or to express mRNA coding for the catabolic

enzymes. Particular attention was focused on tissues of the digestive tract, in which we

demonstrate expression of BCATc in peripheral nerves, without concomitant expression of

other BCAA catabolic enzymes.

4

MATERIALS AND METHODS

Tissues. Rat tissues (salivary, esophagus, stomach, pancreas, duodenum, jejunum,

ileum, colon, skeletal muscle, kidney, pancreas, ovary, uterus, testis, liver) were removed

from male and female Sprague Dawley or Long-Evans rats (225-250 g) and either flash

frozen in liquid nitrogen and stored at -80oC or cryo-embedded in OCT Compound (Sakura

Finetek, Torrance, CA) by freezing in liquid nitrogen and stored at -80oC. Monkey tissues

(brain, skeletal muscle, adipose, liver, kidney, pancreas, jejunum, stomach) were obtained

from monkeys involved in a study of the effect of dietary cholesterol on lipoprotein

metabolism (50). Portions of the monkey tissues had been used previously to measure tissue

BCAT and BCKD activities (56). Experimental procedures involving animals were

approved by the Institutional Animal Care and Use Committee of the Wake Forest University

School of Medicine. Human tissues (brain, muscle, adipose, liver, kidney, pancreas,

jejunum, heart) were collected as described in a previously published study (56). Specimens

were obtained from patients undergoing surgical procedures. All samples were obtained

from tissues that would have been discarded following pathological examination during

surgery. Stomach was obtained from a single organ donor who had given consent for use of

the tissue for research purposes. All tissues were flash frozen and stored in liquid nitrogen

until analyzed. The protocol for human tissues was approved by the Institutional Review

Board at Wake Forest University School of Medicine.

Antibodies. For identification of BCATm in rat tissues, a polyclonal antibody raised

in rabbits against purified human recombinant BCATm was used. Characterization and

affinity purification of this antibody have been described previously (40). With the exception

of rat stomach, where three different BCATc antibodies were tested to verify the localization

5

of BCATc (see Fig. 4), an antibody raised in rabbits against purified human recombinant

BCATc was used for the immmunolocalization of BCATc in rat tissues. Affinity purification

of this antibody is described below. The other two BCATc antibodies, used with rat

stomach, were an immunoaffinity purified rabbit anti-rat BCATc peptide antibody, directed

against the first 50 amino acids of the rat enzyme (40), and an IgG fraction of an antiserum

that was raised in rabbits against purified rat brain BCATc protein (18). These two BCATc

antibodies and the BCATm antibody have been used previously to identify BCATc and

BCATm in rat tissues or rat brain primary cell cultures by immunoblotting and/or by

immunohistochemistry (3, 4, 18, 31, 40). Antiserum for localization of BCKD was generated

against the purified E2 subunit of the rat liver BCKD complex, and was the gift of Dr.

Yoshiharu Shimomura (Nagoya Institute of Technology, Nagoya, Japan). An IgG fraction of

the E2 antiserum has been used previously to identify BCKD in primary cell cultures derived

from rat brain (3, 25).

For preparation of the affinity-purified human BCATc and E2 antibodies, the antisera

were made 50% saturated with ammonium sulfate, and centrifuged for 10 min at 10,000 xg.

The supernatants were discarded, and the pellets were washed in PBS. Subsequently, the

protein pellets were dissolved in PBS, followed by dialysis against PBS, and loaded onto an

affinity resin having human BCATc or rat liver BCKD complex as a ligand. The columns

were washed with PBS and the fraction of antigen-specific antibodies was eluted in 0.1 M

sodium acetate buffer (pH 4.0) containing 4 M urea and 0.5 M NaCl. The affinity purified

antibodies were dialyzed overnight against 50% glycerol/water at 4oC. The dialyzed

antibodies were aliquoted and stored at –80oC. Human BCATc-Sepharose and rat liver

BCKD-Sepharose were prepared by coupling the purified human recombinant BCATc or

6

purified rat liver BCKD complex to Affigel 10 support (Bio-Rad, Richmond, CA) according

to the manufacturer’s directions.

Immunoblotting. Tissues for SDS-PAGE/immunoblot were pulverized with mortar

and pestle while submerged in liquid nitrogen. Proteins were extracted from tissue powders

by three rounds of freeze-thaw-sonication in 25 mM HEPES (pH 7.4) containing 0.4%

CHAPS, 1 mM EDTA, 1 mM EGTA, 1 mM DTT, 10 ug/ml leupeptin, 5 mM benzamidine,

and 1 mM diisopropyl fluorophosphate. Proteins in aliquots of tissue extracts (10–60 µg

protein) were separated by SDS-PAGE using 10% gels. Purified recombinant human

BCATc and BCATm proteins (11) were included as standards. Proteins were transferred to

Immobilon P membranes. Membranes were blocked with 5% nonfat milk/PBS or 1%

BSA/PBS and incubated with immunoaffinity-purified rabbit anti-human BCATm (0.3-0.5

µg/ml), anti-rat peptide BCATc (0.4-0.6 µg/ml), or anti-human BCATc (0.4-0.6 µg/ml)

antibodies. The immunoreactive protein bands were visualized using the Enhanced

Chemiluminescence (ECL; for monkey and human tissue blots) or ECL Plus (for rat tissue

blots) detection system according to the manufacturer’s instructions (Amersham Biosciences,

Piscataway NJ), and detected on X-ray film (Amersham Biosciences). Immunoreactive

band intensities were analyzed for film exposures producing signals below saturation (bands

were translucent). Band intensities were quantified in scanned images of the film

(ImageQuant Software, Amersham Biosciences) and are reported as arbitrary units/µg protein

loaded. For quantification of BCATm in monkey tissues, band intensities for individual

samples from each tissue were compared to intensities of a series of purified recombinant

human BCATm or human BCATc standards (2-8 ng purified BCAT). Extract protein

concentrations were adjusted so that band intensities were within the linear range of the

BCAT standards.

7

Immunohistochemistry. Frozen sections, 6-8 µm in thickness, were collected

directly on slides or collected on adhesive tape and transferred to adhesive-coated slides

using a UV-crosslinking system (Instrumedics CryoJane System, Hackensack, NJ). Sections

were fixed by immersion in acetone (10 min), followed by lyophilization and storage at 25oC.

Sections were rehydrated by immersion in PBS for 10 min. Non-specific binding sites were

blocked by treatment with 1% BSA/PBS for 15 min. Sections were incubated with the

primary antibodies diluted to 5-10 µg/ml in BSA/PBS for 30-60 min, and rinsed three times

with PBS before incubation with horseradish peroxidase- or FITC-conjugated secondary

antibodies (Jackson ImmunoResearch Laboratories, West Grove, PA). After washing with

PBS, color was developed for HRP-conjugates using diaminobenzidine/H2O2. Controls

consisted of incubations of sections with secondary antibodies only, or with primary

antibodies that had been preincubated overnight with a 10-fold excess of competing antigen,

followed by secondary antibodies. In some experiments, immunolabeled sections were

counterstained with hematoxylin, a nuclear stain, to aid in the identification of cell types. To

visualize myelin in sciatic nerve, cryosections were fixed with 4% formaldehyde in PBS, and

stained with Oil Red O by standard methods (22). Tissues were viewed with a Zeiss

AxioPlan 2 microscope, and images were obtained using an AxioCam digital camera and

AxioVision imaging software (Carl Zeiss USA, Thornwood, NY). Images were adjusted and

assembled using Adobe Photoshop 6.0 (Adobe Systems Incorporated, San Jose CA).

RNA Extraction and RT PCR. RT-PCR was used to determine if BCATc mRNA was

present in tissues outside the brain that exhibited BCATc-specific immunostaining in nervous

elements. Flash-frozen tissue that had been powdered under liquid nitrogen (100-300 mg)

and stored at -80 oC was used for preparation of total RNA. Total RNA was extracted from

the frozen tissue powder using TriZol reagent according to manufacturer’s instructions (Life

8

Technologies, Rockville, MD). First-strand cDNA was synthesized using oligo(dT) primer

and SuperScript II reverse transcriptase (Gibco BRL, Gaithersburg MD) with 10µg RNA per

sample. The integrity of the cDNA was confirmed by PCR using primers for

glyceraldehyde-3-phosphate-dehydrogenase (G3PDH): sense primer: 5’-

CCTTCATTGACCTCAACTACATGG-3’; antisense primer: 5’-

TCCACCACCCTGTTGCTGTAGC-3’. Rat BCATc was amplified with the following

sequences: sense primer 5’-TCATGGCCTACTTGTCCCGG-3’; antisense primer 5’-

CCATTAGGGCAACTCCAGTGT-3’. The predicted PCR product of 1241 bp includes all

1236 nucleotides of the BCATc coding sequence - ATG start through TAA stop (26), plus

five nucleotides derived from the ends of the primers.

For PCR, cDNA (1.5 µl) was added to each reaction mixture (50 µl total volume)

containing 10 mM Tris-HCl (pH 9.0), 50 mM KCl, 1.5 mM MgCl2 , 0.2 mM deoxynucleotide

triphosphates , and 0.2 µM of each primer. After an initial heat-denaturation step, 0.5 µl Taq

polymerase (5 unit/µl) was added to each reaction. An amplification program of denaturation

(94 C, 1 min), annealing (60 C, 2 min) and extension (72 C, 1 min) was used for 34 cycles

followed by a final elongation step at 72 C for 10 min. A second round of PCR was

conducted by adding 1.5 µl of the first PCR reaction to a fresh reaction mixture. Aliquots of

each reaction mixture (10 µl) were analyzed on a 2% agarose gel followed by staining with

ethidium bromide.

9

RESULTS

Expression of BCATm and BCATc in rat tissues. Immunoblotting with affinity-

purified hBCATm antibodies showed ubiquitous but variable expression of BCATm protein

in rat digestive tissues. To provide an estimate of the relative amounts of BCATm in

digestive tissues, it was necessary to load different amounts of extract protein for each tissue

on gels used for Western blotting (see Fig. 1A). Immunoreactive band intensities were then

converted to intensity per µg protein for each tissue, and the value for pancreas was taken as

100%. As shown in Fig. 1B, there was a 20-fold range of BCATm protein concentrations in

the digestive system tissues. Consistent with previous reports (18, 24, 35), the highest

relative concentrations of BCATm were found in pancreas and stomach. The other parts of

the upper digestive tract (esophagus, duodenum) and the salivary glands had BCATm

concentrations that were 24-40% of the pancreas. The lowest concentrations of BCATm

were found in jejunum, ileum, and colon which had 5-8% of the pancreatic levels. In tissues

outside the digestive system and as shown previously (24, 56), heart and kidney had the

highest BCATm concentrations, and all other tissues examined contained BCATm, including

testis, spleen, uterus, lung, kidney, and thymus (data not shown).

In the rat, BCATc protein has been detected by immunological methods in brain,

ovary, and placenta (18). As shown in Fig. 1C, BCATc protein was not detected by

immunoblotting in whole tissue extracts of the BCATm-expressing tissues of the digestive

system. Similar results were obtained with other BCATm-expressing tissues (data not

shown). Thus, the immunoblot results are consistent with the current view that, in the rat,

BCATc is not expressed outside the brain and female reproductive tissues. In other species,

it has been reported that BCATc is expressed in a wider range of tissues (5, 17). Therefore,

10

we used RT-PCR to determine if BCATc mRNA was detectable in tissues of the digestive

system. After a single round of PCR (34 cycles), a 1241 bp band corresponding to the

BCATc mRNA was observed in several digestive tissues (ileum, jejunum, colon; data not

shown). As shown in Fig. 1D, when an aliquot of the first round PCR reaction was used in a

second round of PCR amplification, a BCATc mRNA-derived band was observed in all

tissues of the digestive system, although at almost undetectable levels in the esophagus. Low

concentrations of BCATc mRNA were also found in testis, kidney, spleen, thymus, heart,

lung, and liver (data not shown). The results raised the possibility that BCATc could be

expressed at low levels in rat tissues or in a selected cell type within a tissue.

Localization of BCATm protein in rat tissues. BCATm immunoreactivity was found

in epithelial cells in all portions of the digestive tract (Fig. 2). Controls for stomach and

colon (no secondary antibody) are shown in Fig. 2G and H. In the submandibular salivary

gland, labeling was strongest in the serous-secreting components of the secretory epithelium;

mucus-secreting epithelia were not stained (Fig. 2A). Throughout the stomach, the heaviest

immunolabeling for BCATm was observed in the epithelial cells of the middle and deep

zones of the gastric mucosa (Fig. 2B). Parietal cells, which have abundant peripherally

located mitochondria, were stained intensely for BCATm (Fig. 2B and inset). Chief cells

were also labeled. The superficial zone of the mucosa, including much of the mucus–

secreting epithelium, was not labeled (not shown). The longitudinal and circular smooth

muscle layers, as well as the thin muscularis mucosae underlying the gastric mucosa showed

light immunolabeling. Light BCATm staining of the smooth muscle was seen throughout the

digestive tract.

In the small intestine, labeling of the duodenum was principally of cells in the deep

zone of the mucosa, in particular, those lining the crypts (Fig. 2C). Labeling in the jejunum

11

and ileum was diffuse and much lighter than in the duodenum, with the highest

concentrations of label again found in the crypts (Fig. 2D & E). Cell types in the small

intestinal crypts include lysozyme-secreting Paneth cells, though some endocrine and stem

cells are also present and may be immunolabeled. Immunoreactivity for BCATm was not

seen in the absorptive cells lying in the more superficial zone of the small intestine mucosa.

The colon was labeled in much the same pattern as the duodenum, with the crypt epithelium

being most heavily labeled (Fig. 2F). The epithelial cells of the crypt are precursor cells for

the absorptive, mucus, and enteroendocrine cells of the more superficial mucosal epithelium.

BCATm also showed cell-specific localization in tissues outside the digestive system

(Fig. 3). In Fig. 3, BCATm-specific immunostaining is seen as a brown color and the cell

nuclei are stained with hematoxylin (blue/purple color). In skeletal muscle (Fig. 3A) and

heart (data not shown) BCATm was found in myofibrils. In kidney, BCATm was located in

the cortex. Intense labeling with the BCATm antibody was seen in epithelial cells lining

short tubule segments in the cortex (Fig. 3B). The smaller lumina of these tubules are

indicative of distal convoluted tubules, as are the more apical positions of the nuclei of the

immunoreactive epithelial cells. Less intense staining was observed in cells of the

glomerulus. Immunoreactive cells were seen in interstitial capillaries (Fig. 3B). These cells,

which may be endothelial cells, were not labeled in the controls (data not shown). BCATm-

specific immunostaining was not evident in proximal tubules and collecting ducts. In the

pancreas, there was intense immunolabeling for BCATm in acinar cells of the exocrine

pancreas, whereas immunolabeling was weak in the Islets of Langerhans and in the cells

lining the intercalated ducts (Fig. 3C). Light staining was observed in cells at the periphery

and in cells in the interior of the islets. BCATm-specific immunostaining was not observed

in the immunoadsorbed control shown in Fig. 3I. BCATm was also found in the lung, where

12

it was localized to the bronchiolar and alveolar epithelium (data not shown). Consistent with

earlier reports (32), BCATm was not seen in sections of liver (compare Fig. 3G & H).

All tissues of the reproductive tract of male and female rats that were examined

expressed BCATm. In ovarian follicles, labeling for BCATm was seen in the secretory cells

that make up the theca interna (arrows in Fig. 3D). The follicular epithelial cells

(granulosum) and stromal fibroblasts (theca externa) were unlabeled. BCATm was localized

in the uterus to secretory epithelial cells in the deep portions of endometrial glands (Fig. 3E).

Surrounding smooth muscle was not labeled. In the testis, BCATm was restricted to the

interstitial tissues, with the most intense labeling in Leydig cells (Fig. 3F). Seminiferous

tubules did not show appreciable BCATm-specific immunoreactivity.

Localization of BCATc in rat tissues. The observation that low levels of BCATc

mRNA were present in rat tissues raised the possibility that BCATc protein might also be

present in these tissues, albeit at levels too low to be detected easily by immunoblotting of

whole tissue extracts. Therefore, several different BCATc-specific antibodies were used to

look for BCATc-specific staining in rat stomach (see Methods). Rat stomach was chosen

because it expresses the BCATc mRNA, and the high concentration of BCATm in this tissue

would provide a good test for the specificity of the BCATc antibodies. All three anti-BCATc

antibodies exhibited a similar staining pattern. Cells and processes of the neural elements of

the stomach exhibited intense staining for BCATc (Fig. 4B-D). Intense BCATc-specific

labeling was found in the myenteric (Auerbach’s) plexus lying between the circular and

longitudinal smooth muscle layers. Immunolabeling was also observed in fine processes of

the submucosal (Meissner’s) plexus (not shown). Because the affinity purified anti-human

BCATc antibody showed high specificity and low background (Fig. 4D), it was used to

localize BCATc in other rat tissues.

13

BCATc-specific immunostaining was found in neural elements of all gastrointestinal

tract tissues that were examined (Fig. 5). In the salivary gland (Fig. 5A), bundles and finer

processes of immunoreactive neurons were seen in the vascular/connective tissue beds.

BCATc-specific staining was also seen in nerves in the pancreas (Fig. 5B). From the

stomach to the colon, BCATc was immunolocalized in the nerves and ganglia of the

myenteric nerve plexus (Fig. 5C-G). It appeared that cell bodies as well as neuronal

processes were immunolabeled. In these and all other tissues, BCATc immunostaining was

not evident in non-neuronal cells. Salivary gland and jejunum controls are shown in Fig. 5H

and 5I, respectively.

To determine if BCATc was also present in nerves outside the digestive tract, rat

sciatic nerve was examined for BCATc-specific immunoreactivity. Nerve-specific

localization of BCATc was confirmed in cross-sections of the sciatic nerve (Fig. 6).

Labeling within the cross-sectioned nerve appeared as well-defined small circular and

polygonal profiles (Fig. 6A). Companion sections of the sciatic nerve were stained with the

lipid stain Oil Red O to visualize the myelin sheath surrounding the nerve axons (Fig. 6B).

Comparison of Fig. 6A and B suggests that BCATc immunoreactivity is restricted to the

nerve axons and not found in the myelin sheath. As the nerve enters the muscle

compartment, immunoreactivity for BCATc could also be seen in tangentially and cross-

sectioned axons (Fig. 6C).

Localization of BCKD in rat tissues. In general, the distribution of the E2 subunit of

BCKD in rat tissues paralleled that of BCATm (Fig. 7). BCKD E2-specific staining was

found in the epithelial cells of the middle and deep zones of the gastric mucosa (compare Fig.

7A and Fig 2B). As seen with BCATm, cells with parietal cell morphology exhibited intense

E2 immunostaining. There was no specific labeling for BCKD E2 in the myenteric nerve

14

plexus. E2 was present in the exocrine and endocrine pancreas, though there was less

difference in intensity of immunolabeling between acinar cells and islets than was observed

with BCATm. (compare Fig. 7B & Fig. 3C). BCKD immunoreactivity was the same as

observed with BCATm in skeletal muscle and heart, whereas in kidney BCKD E2-specific

staining was found in the proximal as well as in the distal convoluted tubules (data not

shown). Consistent with reports showing that rat liver has the highest concentration of

BCKD activity and protein (21, 56, 59), intense BCKD E2-specific staining was found in

liver hepatocytes (Fig. 7C). Immunoadsorbed liver and pancreas controls are shown in Fig

7G and 7H, respectively.

In the female reproductive tract, immunoreactivity for BCKD E2 was seen in cells

composing the granulosa and theca interna in the ovary (Fig 7D) and in the epithelial cells of

the uterine endometrium (Fig. 7E). As observed with BCATm, BCKD E2 was found in the

Leydig cells of the testis and not in the seminiferous tubule (Fig. 7F). Thus, the distribution

of BCKD is very similar to that of BCATm.

Expression of BCAT isozymes in selected human and nonhuman primate tissues.

Immunoblotting of African Green monkey and human tissues with isozyme-specific affinity-

purified anti-BCAT antibodies confirmed the ubiquitous expression of BCATm and

expression of BCATc in monkey and human brain tissue (Fig. 8). BCATm protein levels

varied considerably both in monkey and human tissues. The human recombinant proteins

were used as standards to estimate the level of the BCAT proteins in the monkey tissues.

The results indicated that the highest concentration of BCATm protein is found in pancreas

and kidney followed by muscle, stomach, and jejunum. Some degradation of BCATm can be

observed in the pancreas extracts. Significant concentrations of BCATc protein were only

observed in brain tissue. In monkey brain BCATc represented about 80% of total BCAT

15

protein (BCATm ~20%) which is similar to what has been reported for rat brain (24, 35, 36).

BCATm was near the limits of detection in liver and subcutaneous fat.

The human BCATm and rat BCKD antibodies were used to examine the cellular

localization of these proteins in available monkey tissues. In most tissues the localization of

these enzymes was similar to what was observed in rat tissues. An exception was monkey

liver, which like human liver, has measurable BCAT activity (56). As shown in Fig. 9A,

BCATm immunoreactivity was not found in hepatocytes. Immunoreactivity appeared to be

localized in cells that have the location and morphology of Kupffer cells. As in the rat,

BCKD E2 immunoreactivity was found in hepatocytes. For other tissues (stomach, pancreas,

kidney, skeletal muscle, and heart), the localization of BCATm immunoreactivity paralleled

the localization of this protein in the rat (data not shown). As in the rat, distal convoluted

tubules of monkey kidney were labeled, though the glomerulus was not immunoreactive for

BCATm. For BCKD, the labeling pattern in liver, pancreas, skeletal and heart muscle was

the same as in the corresponding rat tissues (data not shown).

DISCUSSION

A novel finding of the present work is the consistent localization of BCAA catabolic

enzymes in secretory epithelial cells. This is particularly marked for BCATm, which appears

along the course of the digestive tract in cells that secrete either salivary components, acid,

enzymes, or enzyme precursors. BCATm was not found in absorptive cells or in goblet cells,

which secrete mucus components. In other organ systems, BCATm is found in protein- and

steroid- secreting cells (e.g., ovarian endometrium, testicular Leydig cells, respectively) as

16

well as in transporting epithelia (kidney distal convoluted tubules). The BCATm-containing

cells are variously derived from ecto- endo-and mesoderm. For the most part, these cells

have in common only their position at an interface with the lumen of an organ (i.e.,

epithelial). Second, our results show that with a few exceptions, BCATm is expressed with

BCKD, the second enzyme of the BCAA catabolic pathway.

The preferential localization of BCATm in secretory epithelial cells raises interesting

questions about the function(s) of the BCAA in these cells. For example, high concentrations

of BCATm and BCKD are found in the pancreatic acini that synthesize and secrete proteins,

rather than in the islet, where leucine and -ketoisocaproate (KIC, the transamination product

of leucine) are known to stimulate insulin secretion (43, 60, 61). The high concentrations of

BCATm in the pancreatic acini could function to provide KIC as a signaling molecule to the

islet. Alternatively, BCATm activity in the acini may serve to dampen leucine signaling in

the islet. KIC release from pancreas has not been measured directly. However, even though

differences did not reach statistical significance, portal KIC concentrations were 6.4 M

higher than arterial KIC concentrations in fed dogs (40.6 versus 34.2 M; reference 63). The

portal vein receives blood from the stomach, pancreas, and intestine. Leucine stimulates

protein synthesis in several tissues (15, 42, 46, 60). If leucine or KIC is a nutrient signal in

the exocrine pancreas, then the high concentrations of BCATm may regulate the availability

of the active metabolite.

In the gastrointestinal tract, BCAA catabolic enzymes are found in the secretory

epithelial cells rather than in the absorptive epithelial cells of the intestinal mucosa.

Therefore, only a small fraction of enterally fed BCAAs would be expected to be oxidized by

the intestinal epithelium during the absorptive process. This interpretation is consistent with

results from studies using leucine and KIC tracers in which it has been shown that rates of

17

nitrogen transfer (BCAA transamination) are higher than rates of oxidation (BCKD step) and

that oxidation by the gut is limited (2, 44, 45, 62, 63). For example, Matthews and

coworkers (44, 45) calculated that <2% of nasogastrically delivered leucine and about 5% of

KIC were oxidized on the first pass in humans. Yu et al. (63) measured metabolism of

arterially delivered leucine across the midgut (duodenum to transverse colon) and liver in

dogs. The gut accounted for only 4% of whole body leucine oxidation in fasted dogs and for

13% in fed dogs.

Across species, stomach has one of the highest specific activities of BCAT (35, 56).

While a number of studies have examined leucine metabolism across the splanchnic bed,

where amino acids derived from hydrolysis in the small intestine are plentiful (2, 13, 16, 44,

45, 47, 62, 63), information on the stomach is limited. Amino acid transporters that transport

BCAA have been identified in the stomach. Sobrevia et al (55) have reported that an “L”

type Na+-independent transport system is present in the basolateral surface of the oxyntic

glands, which contain both parietal and chief cells. This transporter is likely to mediate

BCAA uptake by the gastric epithelium from the arterial circulation. Another amino acid

transporter, the ATB0+ protein, is located on the luminal surface of lung epithelial cells, and

is also expressed in mouse stomach (54). This indicates that free BCAA, if available in the

stomach, can also be taken up at the luminal surface of the gastric epithelium. On the other

hand, mRNA encoding a recently cloned peptide transport protein was expressed in the small

intestine but not in stomach or colon in several animal species (8). Thus, it is likely that most

stomach BCAA metabolism is derived from arterial BCAAs. However, since stomach

metabolism of lumen-derived free BCAAs is possible, it could affect conclusions about

BCAA requirements and the proportion of BCAA metabolism that occurs in the digestive

system.

18

Generally, liver is the primary site of catabolism of indispensable amino acids. It is

also likely that liver is the primary site of BCAA oxidation in the rat (28, 53, 56). What sets

BCAAs apart from other indispensable amino acids is that nitrogen transfer via

transamination is largely extrahepatic. For example, in the rat BCATm is not found in liver,

but liver has the highest concentration and activity of BCKD (20,21,56). On the other hand,

nonhuman primates and humans have measurable liver BCAT activity, but it is still less than

6% of total body capacity (17, 56), and BCATm protein is near the limits of detection in both

human and monkey liver (Fig. 8). In the present study, BCATm is found in monkey liver in

what appears to be the phagocytic Kupffer cells, which make up a small proportion of the

cells in the liver. In contrast, BCKD is found in the far more numerous liver hepatocytes in

monkey and rat (Fig 7C). Thus, monkey liver is an example of a tissue where BCKD is

expressed in a cell without BCATm. Whether or not this is true in human liver remains to be

determined. Nevertheless, it is still likely that, as in the rat, under most conditions liver is a

primary site of BCAA oxidation in humans and nonhuman primates. This hypothesis is

supported by a case report showing that after transplantation of a normal liver to a patient

with Maple Syrup Urine Disease (BCKD defect), the patient was able to tolerate a diet with a

normal protein content (58).

For BCATc, the brain is the primary site of expression in human, monkey, sheep, and

rat (14, 56; see also Fig. 8), and in rat, BCATc is found in neurons in culture (3, 25).

However, the expression of BCATc in rat peripheral nerves was unexpected. There is

evidence that BCATc has a more widespread distribution in other animals. By separating

BCATc from BCATm activity using DEAE-cellulose chromatography of crude tissue

homogenates, Goto et al (17) concluded that BCATc activity represents a variable but

significant proportion of total BCAT activity in most human tissues. In a more recent study

19

involving, BCATc-specific antibodies were used to immunoprecipitate BCATc from sheep

tissue homogenates (5). Significant activity attributed to BCATc was found in sheep muscle,

along with BCATc mRNA (5). If the rat is representative of other animals, then the BCATc

measured in these studies may be localized in nervous elements of peripheral tissues of these

animals as well.

Although we have not yet investigated the types of nerves that express BCATc, in the

rat, this isozyme is found in neural elements supplying the salivary gland and pancreas, as

well as in Auerbach’s and Meissner’s plexuses in the gut wall. These nerves are all parts of

the autonomic nervous system, supplying motor neurons that innervate glands and smooth

muscle in the digestive system. In the stomach and intestine, some of the BCATc

immunolabeling may be associated with the interstitial cells of Cajal. These cells are closely

associated with enteric ganglia and smooth muscle cells, and have some neuronal

characteristics (51). In the sciatic nerve, the nonuniform distribution of BCATc

immunoreaction in the axons (see Fig. 6) may reflect the association of BCATc with a

particular class of neurons or with a specific neuronal component that is unevenly distributed

in the axoplasm. BCATc is a target of the neuroactive drug gabapentin (25), which is used

widely to treat neuropathic pain (1). We have also found BCATc in spinal cord neurons (AJ

Sweatt, MA Garcia-Espinosa, R Wallin, SM Huston, unpublished observations). It is

possible that inhibition of BCATc by gabapentin may contribute to its efficacy in the

treatment of neuropathic pain.

In the central nervous system, the BCAT isozymes are thought to participate in a

shuttle that provides amino nitrogen for de novo synthesis of excitatory neurotransmitter

glutamate in brain and in the retina (25, 40, 64). In this scheme, nitrogen is shuttled between

BCATm in astroglia and BCATc in glutamatergic and/or -aminobutyric acid (GABAergic)

20

neurons. In the peripheral nervous system, including the autonomic division innervating the

gastrointestinal tract, the principal neurotransmitters are acetylcholine, the catecholamines

(norepinephrine and epinephrine), and neuroactive peptides. For these neurotransmitter

systems, the function of a nitrogen shuttle operating between neurons and surrounding glia is

not clear, although -aminobutyric acid is synthesized in tissues outside the brain. An

alternative function for BCATc might be to regulate (or attenuate) the anabolic signal

provided by leucine (12, 19, 42, 61). The apparent absence of BCKD in the peripheral

nerves indicates that KIC produced from leucine by BCATc in neurons could not be

metabolized, and is likely released. Finally, it should be noted that for both BCATm and

BCATc, non-enzymatic roles in intracellular signaling may be possible. Recent work has

shown that a splice variant of BCATm with an internal deletion of 12 amino acids is

expressed in colon carcinoma cells (41). This isoform of BCATm binds to the thyroid

hormone receptor and enhances its effects on nuclear transcription activity (41).

Furthermore, BCATm has a redox-active CXXC center (9) and has been found to associate

with BCKD and other proteins (27). Identifying the constituents of neurons with which

BCATc might interact may shed light on its role in the peripheral nervous system.

21

ACKNOWLEDGEMENTS

The work reported here was supported by grants DK 34738 and NS 38641 from the U.S.

National Institutes of Health and grant 98-35200-6067 from the U.S. Department of

Agriculture (SMH).

Current Address for Agus Suryawan:

U.S. Department of Agriculture/ Agriculture Research Service Children’s Nutrition Research Center Department of Pediatrics Baylor College of Medicine Houston TX 77030

22

DISCLOSURE STATEMENT

The authors of the paper, “Branched-Chain Amino Acid Catabolism: Unique Segregation of

Pathway Enzymes in Organ Systems and Peripheral Nerves,” have no potential conflicts of

interest in submitting this paper to the APS for consideration for publication.

23

REFERENCES

1. Backonja M, and Glanzman RL. Gabapentin dosing for neuropathic pain: evidence from

randomized, placebo-controlled clinical trials. Clin Ther 25: 81-104., 2003.

2. Biolo G, and Tessari P. Splanchnic versus whole-body production of alpha-ketoisocaproate

from leucine in the fed state. Metabolism 46: 164-167, 1997.

3. Bixel M, Shimomura Y, Hutson S, and Hamprecht B. Distribution of key enzymes of

branched-chain amino acid metabolism in glial and neuronal cells in culture. J Histochem

Cytochem 49: 407-418, 2001.

4. Bixel MG, Hutson SM, and Hamprecht B. Cellular distribution of branched-chain amino

acid aminotransferase isoenzymes among rat brain glial cells in culture. J Histochem

Cytochem 45: 685-694, 1997.

5. Bonfils J, Faure M, Gibrat JF, Glomot F, and Papet I. Sheep cytosolic branched-chain

amino acid aminotransferase: cDNA cloning, primary structure and molecular modelling and

its unique expression in muscles. Biochim Biophys Acta 1494: 129-136, 2000.

6. Burrin DG, Stoll B, Chang X, van Goudever JB, Fujii H, Hutson S, and Reeds PJ.

Parenteral nutrition results in impaired lactose digestion and hexose absorption when enteral

feeding is initiated in infant pigs. Am J Clin Nutr : In Press, 2003.

7. Chang TW, and Goldberg AL. The metabolic fates of amino acids and the formation of

glutamine in skeletal muscle. J Biol Chem 253: 3685-3693, 1978.

8. Chen H, Wong EA, and Webb KE, Jr. Tissue distribution of a peptide transporter mRNA

in sheep, dairy cows, pigs, and chickens. J Anim Sci 77: 1277-1283, 1999.

24

9. Conway ME, Yennawar N, Wallin R, Poole LB, and Hutson SM. Identification of a

peroxide-sensitive redox switch at the CXXC motif in the human mitochondrial branched

chain aminotransferase. Biochemistry 41: 9070-9078, 2002.

10. Cree TC, Hutson SM, and Harper AE. Gas-liquid chromatography of alpha-keto acids:

quantification of the branched-chain-alpha-keto acids from physiological sources. Anal

Biochem 92: 159-163, 1979.

11. Davoodi J, Drown PM, Bledsoe RK, Wallin R, Reinhart GD, and Hutson SM.

Overexpression and characterization of the human mitochondrial and cytosolic branched-

chain aminotransferases. J Biol Chem 273: 4982-4989, 1998.

12. Dumont FJ, and Su Q. Mechanism of action of the immunosuppressant rapamycin. Life Sci

58: 373-395, 1996.

13. Elango R, Pencharz PB, and Ball RO. The branched-chain amino acid requirement of

parenterally fed neonatal piglets is less than the enteral requirement. J Nutr 132: 3123-3129,

2002.

14. Faure M, Hayes H, Bledsoe RK, Hutson SM, and Papet I. Assignment of the gene of

mitochondrial branched chain aminotransferase (BCAT2) to sheep chromosome band 14q24

and to cattle and goat chromosome bands 18q24 by in situ hybridization. Cytogenet Cell

Genet 83: 96-97, 1998.

15. Flaim KE, Peavy DE, Everson WV, and Jefferson LS. The role of amino acids in the

regulation of protein synthesis in perfused rat liver. I. Reduction in rates of synthesis

resulting from amino acid deprivation and recovery during flow-through perfusion. J Biol

Chem 257: 2932-2938, 1982.

25

16. Gelfand RA, Glickman MG, Jacob R, Sherwin RS, and DeFronzo RA. Removal of

infused amino acids by splanchnic and leg tissues in humans. Am J Physiol 250: E407-413,

1986.

17. Goto M, Shinno H, and Ichihara A. Isozyme patterns of branched-chain amino acid

transaminase in human tissues and tumors. Gann 68: 663-667, 1977.

18. Hall TR, Wallin R, Reinhart GD, and Hutson SM. Branched chain aminotransferase

isoenzymes. Purification and characterization of the rat brain isoenzyme. J Biol Chem 268:

3092-3098, 1993.

19. Hara K, Yonezawa K, Weng QP, Kozlowski MT, Belham C, and Avruch J. Amino acid

sufficiency and mTOR regulate p70 S6 kinase and eIF-4E BP1 through a common effector

mechanism. J Biol Chem 273: 14484-14494, 1998.

20. Harper AE. Thoughts on the role of branched-chain alpha-keto acid dehydrogenase complex

in nitrogen metabolism. Ann N Y Acad Sci 573: 267-273, 1989.

21. Harris RA, Hawes JW, Popov KM, Zhao Y, Shimomura Y, Sato J, Jaskiewicz J, and

Hurley TD. Studies on the regulation of the mitochondrial alpha-ketoacid dehydrogenase

complexes and their kinases. Adv Enzyme Regul 37: 271-293, 1997.

22. Humason GL. Animal Tissue Techniques. San Francisco: W. H. Freeman and Company,

1979.

23. Hutson S. Structure and function of branched chain aminotransferases. Prog Nucleic Acid

Res Mol Biol 70: 175-206, 2001.

24. Hutson SM. Subcellular distribution of branched-chain aminotransferase activity in rat

tissues. J Nutr 118: 1475-1481, 1988.

26

25. Hutson SM, Berkich D, Drown P, Xu B, Aschner M, and LaNoue KF. Role of branched-

chain aminotransferase isoenzymes and gabapentin in neurotransmitter metabolism. J

Neurochem 71: 863-874, 1998.

26. Hutson SM, Bledsoe RK, Hall TR, and Dawson PA. Cloning and expression of the

mammalian cytosolic branched chain aminotransferase isoenzyme. J Biol Chem 270: 30344-

30352, 1995.

27. Hutson SM, Conway ME, Fujii H, and Wallin R. Discovery of a regulated metabolon

involving key enzymes of the leucine catabolic pathway. Experimental Biology 2003 (FASEB

Annual Meeting), San Diego CA, 2003. Abstract 524.8.

28. Hutson SM, Cree TC, and Harper AE. Regulation of leucine and alpha-ketoisocaproate

metabolism in skeletal muscle. J Biol Chem 253: 8126-8133, 1978.

29. Hutson SM, and Hall TR. Identification of the mitochondrial branched chain

aminotransferase as a branched chain alpha-keto acid transport protein. J Biol Chem 268:

3084-3091, 1993.

30. Hutson SM, and Harper AE. Blood and tissue branched-chain amino and alpha-keto acid

concentrations: effect of diet, starvation, and disease. Am J Clin Nutr 34: 173-183, 1981.

31. Hutson SM, Lieth E, and LaNoue KF. Function of leucine in excitatory neurotransmitter

metabolism in the central nervous system. J Nutr 131: 846S-850S, 2001.

32. Hutson SM, Wallin R, and Hall TR. Identification of mitochondrial branched chain

aminotransferase and its isoforms in rat tissues. J Biol Chem 267: 15681-15686, 1992.

33. Hutson SM, and Zapalowski C. Relationship of branched-chain amino acids to skeletal

muscle gluconeogenic amino acids. In: Metabolism and clinical implications of branched-

chain amino and keto-acids, edited by M. Walser and J. R. Williamson. New York:

Elsevier/North Holland, 1981, p. 245-250.

27

34. Hutson SM, Zapalowski C, Cree TC, and Harper AE. Regulation of leucine and alpha-

ketoisocaproic acid metabolism in skeletal muscle. Effects of starvation and insulin. J Biol

Chem 255: 2418-2426, 1980.

35. Ichihara A. Aminotransferases of branched-chain amino acids. In: Transaminases, edited by

P. Christen and D. E. Metzler. New York: John Wiley & Sons, 1985, p. 430-438.

36. Ichihara A, Noda C, and Goto M. Transaminase of brainched chain amino acids. X. High

activity in stomach and pancreas. Biochem Biophys Res Commun 67: 1313-1318, 1975.

37. Kholodilov NG, Neystat M, Oo TF, Hutson SM, and Burke RE. Upregulation of cytosolic

branched chain aminotransferase in substantia nigra following developmental striatal target

injury. Brain Res Mol Brain Res 75: 281-286, 2000.

38. Labow BI, and Souba WW. Glutamine. World J Surg 24: 1503-1513, 2000.

39. LaNoue KF, Berkich DA, Conway M, Barber AJ, Hu LY, Taylor C, and Hutson S. Role

of specific aminotransferases in de novo glutamate synthesis and redox shuttling in the retina.

J Neurosci Res 66: 914-922, 2001.

40. Lieth E, LaNoue KF, Berkich DA, Xu B, Ratz M, Taylor C, and Hutson SM. Nitrogen

shuttling between neurons and glial cells during glutamate synthesis. J Neurochem 76: 1712-

1723, 2001.

41. Lin HM, Kaneshige M, Zhao L, Zhang X, Hanover JA, and Cheng SY. An isoform of

branched-chain aminotransferase is a novel co-repressor for thyroid hormone nuclear

receptors. J Biol Chem 276: 48196-48205, 2001.

42. Lynch CJ, Patson BJ, Anthony J, Vaval A, Jefferson LS, and Vary TC. Leucine is a

direct-acting nutrient signal that regulates protein synthesis in adipose tissue. Am J Physiol

Endocrinol Metab 283: E503-513, 2002.

28

43. Malaisse WJ, Hutton JC, Carpinelli AR, Herchuelz A, and Sener A. The stimulus-

secretion coupling of amino acid-induced insulin release: metabolism and cationic effects of

leucine. Diabetes 29: 431-437, 1980.

44. Matthews DE, Harkin R, Battezzati A, and Brillon DJ. Splanchnic bed utilization of

enteral alpha-ketoisocaproate in humans. Metabolism 48: 1555-1563, 1999.

45. Matthews DE, Marano MA, and Campbell RG. Splanchnic bed utilization of leucine and

phenylalanine in humans. Am J Physiol 264: E109-118, 1993.

46. May ME, and Buse MG. Effects of branched-chain amino acids on protein turnover.

Diabetes Metab Rev 5: 227-245, 1989.

47. Metges CC, El-Khoury AE, Selvaraj AB, Tsay RH, Atkinson A, Regan MM, Bequette

BJ, and Young VR. Kinetics of L-[1-(13)C]leucine when ingested with free amino acids,

unlabeled or intrinsically labeled casein. Am J Physiol Endocrinol Metab 278: E1000-1009,

2000.

48. Odessey R, Khairallah EA, and Goldberg AL. Origin and possible significance of alanine

production by skeletal muscle. J Biol Chem 249: 7623-7629, 1974.

49. Reed LJ, Damuni Z, and Merryfield ML. Regulation of mammalian pyruvate and

branched-chain alpha-keto acid dehydrogenase complexes by phosphorylation-

dephosphorylation. Curr Top Cell Regul 27: 41-49, 1985.

50. Rudel L, Deckelman C, Wilson M, Scobey M, and Anderson R. Dietary cholesterol and

downregulation of cholesterol 7 alpha-hydroxylase and cholesterol absorption in African

green monkeys. J Clin Invest 93: 2463-2472, 1994.

51. Sanders KM. A case for interstitial cells of Cajal as pacemakers and mediators of

neurotransmission in the gastrointestinal tract. Gastroenterology 111: 492-515, 1996.

29

52. Sener A, Somers G, Devis G, and Malaisse WJ. The stimulus-secretion coupling of amino

acid-induced insulin release. Biosynthetic and secretory responses of rat pancreatic islet to L-

leucine and L-glutamine. Diabetologia 21: 135-142, 1981.

53. Shinnick FL, and Harper AE. Branched-chain amino acid oxidation by isolated rat tissue

preparations. Biochim Biophys Acta 437: 477-486, 1976.

54. Sloan JL, Grubb BR, and Mager S. Expression of the amino acid transporter ATB 0+ in

lung: possible role in luminal protein removal. Am J Physiol Lung Cell Mol Physiol 284:

L39-49, 2003.

55. Sobrevia L, Medina V, Reinicke K, and Bravo I. Uptake of L-leucine and L-phenylalanine

across the basolateral cell surface in isolated oxyntic glands. Biochim Biophys Acta 1106:

257-263, 1992.

56. Suryawan A, Hawes JW, Harris RA, Shimomura Y, Jenkins AE, and Hutson SM. A

molecular model of human branched-chain amino acid metabolism. Am J Clin Nutr 68: 72-

81, 1998.

57. Wallin R, Hall TR, and Hutson SM. Purification of branched chain aminotransferase from

rat heart mitochondria. J Biol Chem 265: 6019-6024, 1990.

58. Wendel U, Saudubray JM, Bodner A, and Schadewaldt P. Liver transplantation in maple

syrup urine disease. Eur J Pediatr 158 Suppl 2: S60-64, 1999.

59. Wohlhueter RM, and Harper AE. Coinduction of rat liver branched chain alpha-keto acid

dehydrogenase activities. J Biol Chem 245: 2391-2401, 1970.

60. Xu G, Kwon G, Cruz WS, Marshall CA, and McDaniel ML. Metabolic regulation by

leucine of translation initiation through the mTOR-signaling pathway by pancreatic beta-

cells. Diabetes 50: 353-360, 2001.

30

61. Xu G, Kwon G, Marshall CA, Lin TA, Lawrence JC, Jr., and McDaniel ML. Branched-

chain amino acids are essential in the regulation of PHAS-I and p70 S6 kinase by pancreatic

beta-cells. A possible role in protein translation and mitogenic signaling. J Biol Chem 273:

28178-28184, 1998.

62. Yu YM, Burke JF, Vogt JA, Chambers L, and Young VR. Splanchnic and whole body L-

[1-13C,15N]leucine kinetics in relation to enteral and parenteral amino acid supply. Am J

Physiol 262: E687-694, 1992.

63. Yu YM, Wagner DA, Tredget EE, Walaszewski JA, Burke JF, and Young VR.

Quantitative role of splanchnic region in leucine metabolism: L-[1-13C,15N]leucine and

substrate balance studies. Am J Physiol 259: E36-51, 1990.

64. Yudkoff M, Daikhin Y, Grunstein L, Nissim I, Stern J, and Pleasure D. Astrocyte

leucine metabolism: significance of branched-chain amino acid transamination. J Neurochem

66: 378-385, 1996.

65. Zielke HR, Huang Y, Baab PJ, Collins RM, Jr., Zielke CL, and Tildon JT. Effect of

alpha-ketoisocaproate and leucine on the in vivo oxidation of glutamate and glutamine in the

rat brain. Neurochem Res 22: 1159-1164, 1997.

31

FIGURE LEGENDS

Figure 1. BCATm is detectable by immunoblot throughout the rat gastrointestinal tract,

while BCATc is detectable only by RT-PCR. A: Immunoblot of BCATm. Amounts of

tissue extract loaded onto the gel ranged from 10-60 µg protein/lane. Recombinant human

BCATm (8 ng) was used as a standard . B: Relative tissue concentrations of BCATm.

Immunoreactive band intensity per µg protein loaded was calculated for each tissue, and is

presented as a percentage of the value for pancreas, which had the highest concentration

(taken as 100%). C: Immunoblot for BCATc. Forty µg of tissue extract protein was loaded

on the gel, except for brain, where 20 µg was used. D: RT-PCR for BCATc mRNA. cDNA

was reverse-transcribed from mRNA from rat gastrointestinal tissues, used as the template

for PCR with BCATc-specific primers, and reamplified in second round of PCR (see

Methods). Lanes are in the same order as for the immunoblot, except for the positive control,

for which a BCATc-encoding plasmid was used as the PCR template.

Figure 2. BCATm is present in epithelial cells in all portions of the rat gastrointestinal tract.

A: Salivary gland. B: Stomach. C: Duodenum. D: Jejunum. E: Ileum. F: Colon. G: Stomach

control (secondary antibody only). H: Colon control. Images in panels B-H are oriented

such that the mucosa is at the top, while the smooth muscle layers are at the bottom of each

panel. In all tissues, eosinophils display non-immunologically based reactivity (small

arrows). Labeling for BCATm is most prominent over serous-secreting portions of the

salivary gland, and over the middle (neck) and deeper zones, including crypts, of the gastric

mucosa (large arrows). In panel B, inset shows immunofluorescent localization of BCATm

in the deep zone of the gastric epithelium. Glands are seen in longitudinal (*) and cross-

32

section (L indicates lumen of gland). Immunolabel is strongest in the peripheral cytoplasm

of the epithelial cells. Magnification for all panels is X250. Scale bar: 100 µm.

Figure 3. BCATm is present in specific cell types in rat tissues. In this figure and other

color figures, cell nuclei were counterstained with hematoxylin (blue/purple) to allow

identification of cell types in the tissues. Specific immunolabel appears brown.

A: Skeletal muscle, X600. B: Kidney, X440. Arrows indicate immunoreactive distal

convoluted tubules. Occasional immunoreactive cells are seen in the interstitial capillaries.

(glom: glomerulus) C: Pancreas, X280 (islet: Islet of Langerhans). D: Ovary, X170.

Arrows indicate immunoreactive theca interna of an antral follicle. E: Uterus, X1040. F:

Testis, X200. Arrows indicate immunoreactive interstitial Leydig cells (tubule: seminiferous

tubule). G: Liver, X560. H: Liver control, X520. Primary antibody (5-10 ug/ml) was

preadsorbed with recombinant BCATm. I: Pancreas, X440. Immunoadsorbed control. Scale

bar is 50 um for panel C, 100 um for panel F, and 25 µm for all other panels.

Figure 4. BCATc antibodies identify BCATc in nervous elements in rat stomach. A:

Control (secondary antibody only); B: Anti-rat BCATc N-terminal peptide antibody. C:

Anti-rat BCATc, IgG fraction. D: Anti-recombinant human BCATc antibody. Deep zone of

gastric mucosa is at top of each panel, while smooth muscle layers are at the bottom of each

panel. All three antibodies react with elements lying between the inner circular and outer

longitudinal muscle layers. The immunoreactive structures are the ganglia and neural fibers

of the intrinsic innervation of the gut (Auerbach’s plexus, large arrows). In panel D,

additional labeling of fine innervating fibers is seen within the circular muscle layer. Small

33

arrows indicate eosinophils showing non-immunological background activity. Magnification

for all panels is X140. Scale bar is 100 µm.

Figure 5. BCATc is localized in nerves of the rat gastrointestinal tract tissues. Cryosections

were probed with affinity-purified anti-recombinant human BCATc, followed by secondary

antibody and peroxidase substrate as described in the Methods. A: Salivary gland. B:

Pancreas. C: Stomach. D: Duodenum. E: Jejunum. F: Ileum. G: Colon. H: Salivary gland

control (secondary antibody only). I: Jejunum control. Labeling for BCATc is restricted to

nervous elements lying between lobes of the salivary gland and to nerves and ganglia in

Auerbach’s plexus in the gastrointestinal tract (large arrows). Small arrows indicate

eosinophil background activity. Magnification is X500 for panels A and B, and X250 for all

other panels. Scale bar is 50 µm for panels A and B, and 100 µm for all other panels.

Figure 6. BCATc is expressed in axons in the sciatic nerve. A: Rat sciatic nerve was

immunolabeled with affinity-purified anti-recombinant human BCATc. Arrows indicate

immunoreactive axons. Arrowheads correspond to structures stained in Panel. B. B: Sciatic

nerve was stained with Oil Red O to reveal lipid-rich elements (myelin, indicated by

arrowheads). C: Rat gastrocnemius was labeled with anti-recombinant human BCATc. N-

nerve; m – muscle. Magnification for all panels is X920. Scale bar is 25 µm.

Figure 7. Branched chain -keto acid dehydrogenase (BCKD) is expressed ubiquitously in

rat tissues. The E2 subunit of branched chain -keto acid dehydrogenase (BCKD) was

immunolocalized in selected rat tissues. In all tissues except stomach, cell nuclei were

counterstained with hematoxylin. A: Stomach. Large arrows indicate immunoreactive

34

epithelial cells in the neck region of the gastric mucosa. Eosinophils display background

activity (small arrows). B: Pancreas. Circle shows approximate boundary of an Islet of

Langerhans. C: Liver. D: Ovary. Arrow indicates immunoreactive cells of theca interna.

E: Uterus. Arrow indicates immunoreactive epithelial cells lining endometrial cavity and

asterisk indicates lumen of an endometrial gland with immunoreactive epithelium. F: Testis.

Arrow indicates immunoreactive interstitial Leydig cells. G, H: Respective liver and

pancreas controls (primary antibody immunoadsorbed with purified BCKD complex).

Magnification in panels A and E is X250, in panel D is X125, and in all other panels is X500.

Scale bar is 100 µm for A and E, 200 µm for D, and 50 µm for all other panels.

Figure 8. In the monkey and the human, BCAT isozymes show tissue distributions similar to

those in the rat. Extracts of monkey and human tissues were analyzed by immunoblot with

antibodies directed against recombinant human BCATm or BCATc. Recombinant human

BCATm and BCATc (20 ng each) were used as standards. Human tissues are in the same

order as monkey tissues, except for heart, which was used instead of stomach. Tissue

extracts were loaded at 30 µg protein/lane. Inset table shows concentrations of BCATc and

BCATm in monkey tissues, as determined by densitometry in separate blots (see Methods).

Figure 9. In the monkey, BCATm is localized to Kupffer cells in liver and to distal

convoluted tubules in kidney. In these sections, nuclei were counterstained with

hematoxylin. A: Liver. Arrows indicate immunoreactive Kupffer cells in the liver sinusoidal

lining. Inset shows control section exposed to immunoadsorbed antibody. B: Kidney.

Arrows indicate immunoreactive epithelium of convoluted tubules. Inset shows control

section. Magnification for both panels is X230. Scale bar is 100 µm.

35

Figure 1

36

Figure 2

37

Figure 3

38

Figure 4

39

Figure 5

40

Figure 6

41

Figure 7

42

Figure 8

43

Figure 9