Aspartate Aminotransferase Macroenzyme ... - Clinical Chemistry

CLIN. CHEM. 40/7, 1340-1343 (1994)

Aspartate Aminotransferase Macroenzyme Complex in Serum Identified and Characterized Marie-Jos#{233} Stasia,”4 Fran#{231}oise Morel’

Anny Surla,’

Jean-Charles

Renversez,2

Macromolecular aspartate aminotransferase was found in the serum of an apparently healthy patient. This complex was composed of aspartate aminotransferase (AST; EC 2.6.1.1) and immunoglobulin. Electrophoresis of the patient’s serum showed an abnormal band migrating between mitochondrial (m) and cytosolic (s) AST. The macromolecular complex was purified by gel filtration on Sephacryl S300. The molecular mass of the complex was estimated to be 250 kDa, suggesting that the complex probably consists of one immunoglobulinmolecule associated with one AST molecule. By immunoelectrophoresis, the immunoglobulinwas found to be an lgG with K-A type light chain. When we used polyclonal antibodies against human mAST or sAST, the sAST antibodies strongly inhibited the AST activity of the macrocomplex, whereas the mAST antibodies had no effect. Thus the AST molecule of the macrocomplex is an sAST type. Indexing Terms: immunoglobuliri-enzyme complexes/isoenzymes/ chromatography, gel filtration/immunoelectrophoresis

Abnormal enzyme activities in a patient’s serum are generally associated with disease. Sometimes persistently increased enzyme activity is due to the formation of a macrocomplex between immunoglobulin and (e.g.,) amylase, nikaline phosphatase, creatine kinase, or glucose-6-phosphate dehydrogenase [see review by llamaley and Wilding (1)]. Aspartate aminotransferase (AST; EC 2.6.1.1) exists in two forms, one cytosolic (sAST), the other of mitochondrial origin (mAST).5 Recently, several individuals with serum containing AST complexed with immunoglobulin have been reported (2-6). In these instances, the presence of complexes was not always associated with disease, although the presence of mAST in serum indicates mitochondrial damage and is usually associated with serious disease (3). We report here a patient with serum containing complexes between sAST and IgG.

Case Report The patient was a 68-year-oldwoman whose serum AST activity was persistently above normal. She had no complaints, and the results of the physical examination 1orathim

d’Enzymologie

and 2Laboratoire

de Biochimie

A,

Universitaire, 38043 Grenoble Cedex 09, France. 3Laboratoire d’Analyses M#{233}dicales, Pene-Morel, 2 Bd. Joseph Vallier, Grenoble, France. 4Author for correspondence. Fax mt +33-76-76-56-08. 6Nonstandard abbreviations: AST, aspartate aminotransferase; m, mitochondrial; s, cytosolic; SDS, sodium dodecyl sulfate; and PAGE, polyaciylamide gel electrophoresis. Received December 14, 1993; accepted April 1, 1994. Centre Hospitalier

1340

CLINICALCHEMISTRY,Vol.40, No. 7, 1994

Fran#{231}oise Pene,3 Arlette

Morel-Femelez,3

and

were normal. She lived at home and was in a stable condition. Venous blood was drawn with her informed consent for laboratory examinations. High AST activity was the only abnormal finding from assays of several serum enzymes. No clinical explanation was found for the persistently abnormal concentration of AST. Materials and Methods To separate the AST isoenzymes, we used the electrophoresis method described by SakRkibara et al. (7), with the followingmodifications: 9 mmo]/L NaC1 was added to the migrating buffer, electrophoresis was performed for 60 min at room temperature, a constant current of 4 mA was applied for each band, and the staining medium was dissolvedin agarose solution. To detect the macroenzyme complex, we performed immunoprecipitation with an automatic nephelometer (BNA; Behringwerke, Marburg, Germany), using antisera from Atlantic Antibodies (Incstar, Stillwater, MN) and a Bebring ORTG standard in the presence of Polyethylene Glycol 6000, 6 mmolJL (8). The heavy and the light chains of the immunoglobulins were identified immunochemically by immunofixation on agarose gel (Hydragel; Sebia, Issy lea Moulineaux, France) (9). Prior to subjecting the protein to electrophoresis, we dissolved it in 0.06 molIL Tris-HC1, pH 6.8, containing 150 mL of glycerol, 50 mL of /3-mercaptoethanol, 23 g of sodium dodecyl sulfate (SDS), and 10 mg of bromphenol blue per liter. After incubation at 100#{176}C for 5 mm, the solubilized samples were subjected to SDS-polyacrylamide gel electrophoresis (PAGE) as described by Laemmu and Favre (10), with a 4% stacking gel and a 13% resolving gel. Gels were stained with Coomassie Blue R250.

Human mAST isoenzyme was purified from human neonatal liver tissue according to Leung and Henderson (11). The purification procedure took six steps: homogenization, heat treatment, (NH4)2S04 fractionation, DEAE-Sephacel chromatography, Sepharose 4B aspartate chromatography, and isoelectric focusing. As shown by SDS-PAGE (Fig. 1), the chromatographically punfled mAST was contaminated with several other proteins, whereas the material after electrofocusing contained pure mAST. Human sAST isoenzyme was purified from human erythrocytes according to Rej et al. (12). We measured total AST activity by the method of the French Society of Clinical Biology (13), using an Hitachi 705 automated analyzer (Boehringer Mannheim, Mannheim, Germany) at 30#{176}C in the presence of 0.1 mmol/L pyridoxal phosphate, without a blank. Refer-

with 5 mg of Blue Dextran (106 kDa), 0.8 mg of lactate dehydrogenase (139 kfla), and 0.3 mg of hexokinase (99 kDa) in 1 mL of 100 mmol/L phosphate buffer, pH 7.4. The volume of serum applied to the column was 1 mL. Protein elution was detected by absorbance at 280 nm. AST activity was measured as described above.

mAST

I

*Oi ;IA

Results

I

rnarIrs

Fig. 1. SDS-PAGE

bM att Iectrofoising

of purified mAST stained with Coomassie

Blue.

Samples before and after electrofocusing were subjected to SOS-PAGE as described in text. The molecular-mass standards (kDa) used were bovine serum albumin (66), egg albumin(45), glyceraldehyde-3-phosphate dehydrogenase (36), carbonIc anhydrase (29), soybean trypsin inhibitor (20), and a-lactalbumin (14).

ence values for AST activity in apparently healthy subjects were between 0 and 14 UIL (n = 125). Proteins were routinely determined in serum by the method of Bradford (14). Serum for use as control samples was collected from rabbits before they were immunized. Polyclonal antibodies were raised in the rabbits by making three subcutaneous injections, at 2-week intervals, of 100 g of human liver mAST or of human erythrocyte sAST dissolved in 500 L of phosphate-buffered saline and 500 jL of Freund’s adjuvant.For the firstinjectionwe used complete Freund’s adjuvant; for the two others, we used incomplete Freund’s adjuvant. Samples obtained by bleeding the rabbits 3 weeks afterward were immunoblotted by the method of Towbin et al. (15). Protein samples were subjected to SDS-PAGE according to Laemmli and Favre (10) as described above. After electrophoretic transfer of the resulting bands from slab gels onto 0.2-pm-thick nitrocellulose sheets and incubation with anti-mAST and anti-sAST antibodies, preimmune serum, and anti-human IgG (Sigma Chemical Co., St. Louis, MO) diluted 200-fold, the immunoreactive complex was detected with an enhanced chemiluminescence technique from Amersham (Amersham, Bucks, UK). The chemiluminescence was detected on x-ray ifim routinely exposed for 1 min. For gel permeation chromatography, we used Sephacryl S300 obtained from Pharmacia Fine Chemicals (Uppsala, Sweden). The macrocomplex was fractionated on a 59 x 1.6 cm glass-jacketed column of Sephacryl S300 eluted with 100 mmol/L phosphate buffer, pH 7.4, at 4#{176}C. The flow rate was 4 mL/15 mm; 2.2-mL fractions were collected. The gel filtration column was calibrated

Electrophoresis of the patient’s serum (Fig. 2, lanes 2 and 3) and of the isolated macrocomplex with AST activity (lane 5) revealed an abnormally migrating AST that differed from purified human eiythrocytes sAST (lane 1) and purified human liver mAST (lane 4), and from serum of a patient with acute hepatitis containing both isoenzymes (lane 6). Fig. 3 shows the Sephacryl S300 gel filtration elution profiles of the patient’s serum with the abnormally migrating AST and of the serum from a patient with high AST activity (2610 UIL). The AST activity of the patient’s serum containing macroenzyme complex eluted in one peak, with a molecular mass of 250 kDa. Uncomplexed AST activity (Fig. 3B) eluted more slowly. Apparently the first patient’s serum contained a macrocomplex composed of one molecule of immunoglobulin and one molecule of AST. We confirmed the presence of immunoglobulin in the macrocomplex with high AST activity by two experiments. Immunoprecipitation of the Sephacryl S300 gel filtration-purified macrocomplex (4.4 g/L) with antisera to IgG, IgA, or 1gM revealed the presence of IgG (1.6 g/L), IgA (0.2 gIL), and 1gM (0.06 g/L). Immunoelectrophoresis (Fig. 4) confirmed that the major immunoglob-

Fig. 2. Electrophoresis of serum from the patient and from control subjects: lane 1, AST-staining purified human liver sAST (total AST actMty 490 UIL); lane 4, AST-staining purified human ilver mAST (total AST activity 204 U/L); lanes 2 and 3, abnormal AST activity migration from patient’s serum (total AST activity 452 U/L); lane 5, migrationof AST macrocomplex separated from Sephacryl S300 gel filtration (total AST activity 112 U/L); and lane 6, AST activity after electrophoresis of serum from a patient with acute hepatitis (total AST activity 670 UIL).

CLINICALCHEMISTRY,Vol. 40, No. 7, 1994

1341

A

control

AST actMty Abso,banc. at 2SOnm

ig G

eAST

mAST

._-_-_. ----

..43KDa

SMCG 10

15

40

Blus Dsztran

LDH

Hexokinase

U C

a

.0 0

a 1.0.0 ,(

5

10

15

20

25

30

35

SMCG

Purified human liver sAST (laneS), 22 g; purified human liver mAST (lane M), 5 g; pailialtypurified macrocompiex withAST activity (lane C), 136 jig; and humanlgG fromSigma(lane G), 42 g, were subjected to SDS-PAGE. The separated proteins were electrotransferredontonitrocellulose and immunoreacted with:preimmuneserumdiluted200-fold(control); human lgG antiserum(fromSigma)diluted1000-fold(lgG); 5AST antiserum diluted 200-fold (sAST); and mAST antiserum diluted 200-fold (mAST). Standard markers were included (M).

a,

0

SMCG

Fig. 5. Immunoidentification of the AST isoenzyme type in the macrocomplex separated from the patient’s serum.

45

number

B

SMCG

40

45

Fraction number Fig. 3. Sephacryl S300 gel filtration elution patternof AST activities from patient’s serum containing AST-immunoglobulin complex: (A) 1 mL of the patient’s serum (total AST activity 452 U/L); (B) 1 mL of serum with high AST activity (2160 U/L). The column was calIbrated with Blue Dextran, lactate dehydrogenase, and hexokinase as described in the text.

-J

Fig. 4. Immunoelectrophoresls of the partially purified macrocomplex from the patient’s serum. The heavy and light chains of the immunogiobulins were Identifiedby immunofixatlon on agarose gel (Hydragel; Sebla) as described in the text

ulin present in the macrocomplex was IgG with A and K light chains. The AST isoenzyme specificity associated with the macrocomplex proteins was studied by immunoblotting (Fig. 5). Purified human erythrocytes sAST and purified human liver mAST, partially purified macrocomplex protein with AST activity, and human IgG (Sigma) were 1342 CUNICAL CHEMISTRY,Vol. 40, No. 7, 1994

subjected to SDS-PAGE and electrotransferred onto nitrocellulose. Preimmune serum diluted 200-fold reacted only with a 46-kDa protein in the partially purified macrocomplex and with human IgG, not with purified sAST or mAST, which suggests that this 46-kDa protein is a part of the IgG molecule. We confirmed this hypothesis by finding that 1000-fold-diluted antibodies raised against human IgG strongly reacted with a 46-kDa protein and with several others in the macnocomplex proteins and IgG proteins. Those antibodies reacted faintly with purified sAST and mAST, probably because of traces of IgG in the sample. Unfortunately, sAST has the same molecular mass as the 46-kDa protein that makes up part of the IgG molecule. sAST antiserum diluted 200-fold did not cross-react with purified mAST but did induce a strong immunoreaction with purified sAST and the macrocomplex at 46-kDa. Antibodies raised against purified mAST and diluted 200-fold reacted with purified mAST (43 kDa), not with purified 46-kfla sAST protein or with a 43-kDa mAST in the macrocomplex. We concluded that the AST isoenzyme in the macrocomplex was most probably a cytosolic form. To confirm that sAST was the isoenzyme implicated in the formation of the macrocomplex, we used immunoinhibition. Increasing concentrations of sAST antiserum (0-600 j.g)inhibited AST activity from the patient’s serum (Fig. 6). Fifty percent of the AST activity from our patient’s serum was inhibited with 100 pg of sAST antiserum, whereas no inhibition was observed with any concentration of mAST antiserum used (data not shown). DiscussIon Persistently abnormal activities of serum AST in patients have been reported previously (2-6). Konttinen et al. (2) reported the two first individuals in whom the macromolecular complexes involved AST. In the first case, the serum AST was bound by IgG; in the second,

3 .c C

0

50

100

150

Anti

Fig. 6. Inhibition of the AST activity of the patient’s serum by sAST and mAST antiserum. We Incubated 150 MLof the patient’sserum for 1 h at room temperature with increasing concentrations ofsAST antiserum (1.04 g/L) or mAST antiserum (5.8 g/L). Inhibiting50% of theAST activItyfromthe patients serum took 100 g of 5AST antiserum. No inhibitionwasobtainedfrommAST antiserum with anyconcentration used (not shown).

the binding protein could not be identified. The nature of the AST implicated in the formation of the macrocomplex was not identified, but the authors thought it might be a cytosolic form because the two individualswere in excellent health and because the mitochondrial isoenzyme makes a significant contribution only in cases of severe tissue damage. In 1983 Weidner et al. (4) described two patients, in good health, with immunoglobulin-complexed serum AST; the AST specificity was not identified, but the binding protein was again a serunt IgG. In the same year, Nagamine and Okochi (3) reported a patient with metastatic liver cancer in whom AST was complexed with IgG and IgA. They also did not identify the nature of the AST, but mAST was probably implicated because of the severity of liver damage in cancer. Moriyama et al. (5) described a patient with benign disease whose serum contained a macrocomplex between mitochondrial AST and K-type IgG. Therefore, the conditions associated with the incidence of immunoglobulin-complexed AST have not yet been clearlydetermined. Nevertheless, a more recent study by Moriyama et al. (6) examining a population with an increased ratio of AST/alanine aminotransferase activities found that IgA-complexed AST was often associated with liver malignancies. More data from patients showing an association of AST-immunoglobulin complexes with liver disease are needed before the clinical significance of such complexes can be understood. The type of immunoglobulin associated with the complexes was always determined (2-6), but the nature of the AST isoenzyme implicated in the formation of the macrocomplexes was often deduced instead of identified. Here we have described the presence of IgG-complexed sAST in the serum of a woman without clinical symp-

toms. She had never had autoinmiune diseases or hepatitis, and the results of her hematological and immunological tests were normal. The IgG complexed with the AST enzyme was a K-Atype. To our knowledge, this is the first time that the specificity of AST enzyme in the macrocomplex has been determined by using polyclonal antibodies to the two AST isoforms. Anti-sAST abolished the AST activity from the patient’s serum and antimAST had no effect. Study of the biochemical nature of the AST macrocomplex by Sephacryl S300 gel filtration suggests that the complex may consist of one immunoglobulin molecule associated with one AST molecule. As with our patient, most of the previously published reports (2-5) concerned healthy persons whose above-normal AST was found unexpectedly during laboratory evaluations. Rapid identification of AST complexes should prevent such patients from undergoing needless testing and invasive procedures. We thank L. Laval for assistance in preparing and E. Rooney for helpful discussion.

the manuscript

References 1. Remaley AT, Wilding P. Macroenzymes: biochemical characterization, clinical significance and laboratory detection [Review]. Olin Chem 1989;35:2261-70. 2. Konttinen A, Murros sJ, Ojala K, Salaspuro M, Somer H, Rfisanen J. A new cause of increased serum aspartate aminotransferase activity. Olin Chim Acta 1978;84:145-7. 3. Nagamine M, Okochi K. Oomplexesof immunoglobulins A and G with aspartate ammotransferase isoenzymes in serum. Olin Ohem 1983;29:379-81. 4. Weidner N, Lott JA, Yale VD, Wahi RL, Little RA. Immunoglobulin-complexed aspartate aminotransferase. Clin Ohem 1983; 29:382-4. 5. Moriyama T, Ashii T, Kikuiri K, Nishiyama Y, Ito Y, Nobuoka M, et al. Mitochondrial aspartate aminotransferase linked to imniunoglobulin G of the ic-Atype: report of a case. Olin Chim Acta 1986;160:297-305. 6. Moriyama T, Nobuoka M, Makino M. Incidence and properties of aspartate anunotransferase-immunoglobulin complexes in patients with a high serum aspartate to alanine aniinotransferase ratio. Olin Ohim Acta 1990;190:47-56. 7. Sakakibara S, Shiomi K, Kobayashi S, Ikeda T, Inai S, Kaganuyama H. A convenient and sensitive method for the determi-

nation of serum aspartate aminotransferase isozymes after electrophoresis. Olin Chini Acta 1983;133:119-23. 8. Sheidegger JJ. Une microm#{233}thode de l’immunoelectrophor#{232}se. mt Arch Allergy 1955;7:103-7. 9. Mancini 0, Oarbonara AO, Heremans JH. Immunochemical quantitation of antigens by radial immunodiffusion. Immunochemistry 1965;2:235-54. 10. Laemmli UK, Favre M. Maturation of the head of bacteriophage T4. I. DNA packaging events. J Mol Biol 1973;80:575-99. 11. Leung FY, Henderson AR. Isolation and purification of aspartate aminotransferase isoenzymes from human liver by chromatography and isoelectric focusing. Olin Chem 1981;27:232-8. 12. Rej R, Vanderlinde RE, Fasce OJ Jr. An L-aspartate:2-oxoglutarate aminotransferase Reference Material from human erythrocytes: preparation and characterization. Olin Ohem 1972;18:374-

83. 13. Bergmeyer HU, Scheibe P, Wahiefeld AW. Optimization of methods for aspartate aminotransferase and alanine aminotransferase. Olin Ohem 1978;24:58-73. 14. Bradford MM. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 1976;72:248-54. 15. Towbin H, Staehelin T, Gordon J. Electrophoretic transfer of protein from polyacrylainide gels to nitrocellulose sheets: procedure and some applications. Proc Nati Acad Sci USA 1979;76: 4350-4. CLINICAL CHEMISTRY, Vol. 40, No. 7, 1994

1343