different antigenic reactivities of bovine brain glutamate dehydrogenase isoproteins

Journal of Neurochemisfv Lippincott Williams & Wilkins, Inc., Philadelphia 0 1999 International Society for Neurochemistry

Different Antigenic Reactivities of Bovine Brain Glutamate Dehydrogenase Isoproteins

So0 Young Choi, Joung Woo Hong, Min-Sun Song, Seong Gyu Jeon, Jae Hoon Bahn, Byung Ryong Lee, Jee-Yin Ahn, and *Sung-Woo Cho

Department of Genetic Engineering, Division of Life Sciences, Hallym University, Chunchon; and *Department of Biochemistry, University of Ulsan, College of Medicine, Seoul, Korea

Abstract: The structural differences between two types of glutamate dehydrogenase (GDH) isoproteins (GDH I and GDH II), homogeneously isolated from bovine brain, were investigated using a biosensor technology and monoclonal antibodies. A total of seven monoclonal antibodies raised against GDH II were produced, and the antibodies recognized a single protein band that comi- grates with purified GDH II on sodium dodecyl sulfate- polyacrylamide gel electrophoresis and immunoblot. Of seven anti-GDH II monoclonal antibodies tested in the immunoblot analysis, all seven antibodies interacted with GDH II, whereas only four antibodies recognized the protein band of the other GDH isoprotein, GDH 1. When inhibition tests of the GDH isoproteins were performed with the seven anti-GDH II monoclonal antibodies, three antibodies inhibited GDH II activity, whereas only one antibody inhibited GDH I activity. The binding affinity of anti-GDH II monoclonal antibodies for GDH II (K, = 1.0 nM) determined using a biosensor technology (Pharma- cia BIAcore) was fivefold higher than for GDH I (K, = 5.3 nM). These results, together with epitope mapping analysis, suggest that there may be structural differences between the two GDH isoproteins, in addition to their different biochemical properties. Using the anti-GDH II antibodies as probes, we also investigated the cross- reactivities of brain GDHs from some mammalian and an avian species, showing that the mammalian brain GDH enzymes are related immunologically to each other. Key Words: Glutamate dehydrogenase isoproteins-Mono- clonal antibodies-Protein-protein interaction. J. Neurochem. 72, 21 62-21 69 (1 999).

malian sources, crystallization of bovine liver GDH was reported (Peterson et al., 1997). However, remarkably little is known about the detailed function and structure of mammalian GDH, especially the brain enzymes. As the pathology of the disorders associated with GDH defects is restricted to the brain, the enzyme may be of particular importance in the biology of the nervous system. The importance of the pathophysiological nature of the GDH-deficient neurological disorders has attracted considerable interest (Hussain et al., 1989). The enzyme isolated from a patient with a variant form of multisys- tem atrophy displayed a marked reduction of one of the GDH isoproteins (Plaitakis et al., 1993). Although the origin of the GDH polymorphism is not known, it was reported that multiple differently sized mRNAs and multiple gene copies for GDH are present in the human brain (Mavrothalassitis et al., 1988). A novel cDNA encoded by an X chromosome-linked intronless gene also has been isolated from human retina (Shashidharan et al., 1994). However, it is not known whether the distinct properties of the GDH isoproteins are essential for the regulation of glutamate metabolism. Therefore, it is essential to have a detailed structural and functional de- scription of the various types of brain GDH to elucidate the pathophysiological nature of the GDH-deficient neurological disorders.

We previously have isolated two types of GDH isoproteins (designated GDH I and GDH 11) from bovine

Glutamate dehydrogenase (GDH; EC 1.4.1.2-4) is a family of enzymes that catalyze the reversible deamina- tion of L-glutamate to 2-oxoglutarate using NAD+, NADP+, or both as coenzymes (Fisher, 1985). Mamma- lian GDH is ccxnposed of six identical subunits, and the regulation of GDH is very complex (Fisher, 1985). It is only in recent years that the three-dimensional structure of GDH from microorganisms has been available (Baker et al., 1992; Yip et al., 1995). There is, however, rela- tively low identity between the microbial and mammalian GDHs. Very recently, for the first time from mam-

Received September 18, 1998; revised manuscript received Decem- ber 31, 1998; accepted December 31, 1998.

Address correspondence and reprint requests to Dr. S. Y. Choi at Department of Genetic Engineering, Division of Life Sciences, Hallym University, 1 - 1 Okchon-Dong, Chunchon 200-702, Korea.

The present address of M . 3 . Song is Department of Biochemistry, Biophysics and Genetics, School of Medicine, University of Colorado, Denver, CO 80262, U.S.A.

The present address of J. W. Hong is Department of Biochemistry, Ohio State University, Columbus, OH 43210, U.S.A.

Abbreviations used: DME, Dulbecco’s modified Eagle’s medium; GDH, glutamate dehydrogenase; HAT, hypoxanthine-aminopterin- thymidine; HT, hypoxanthine-thymidine; SDS, sodium dodecyl sulfate; TBS, Tris-buffered saline.

2162

CROSS-REACTIVITIES OF BRAIN GDH ISOPROTEINS 2163

brain (Cho et al., 1995), and the regulatory properties of the GDH isoproteins have been reported (Cho and Lee, 1996; Cho et al., 1996, 1998a,b; Kim et al., 1997). Both GDH I and GDH I1 were readily solubilized and showed the same molecular size of 57 kDa. Our work also led to the finding that GDH is present in bovine brain in “heat- labile” (GDH I) and “heat-stable’’ (GDH 11) forms (Cho et al., 1995). It has been reported that the activities of the GDH isotypes differ in their relative resistance to thermal inactivation, detergent extractability, and allosteric regulation characteristics (Colon et al., 1986; Shashidha- ran et al., 1997). Similar results were reported by other investigators showing that reduction in GDH activity in patients with neurodegenerative disorders was largely limited to the heat-labile form (Plaitakis et al., 1984). Very recently, Stanley et al. (1998) have reported that the hyperinsulinism- hyperammonemia syndrome is caused by mutations in the GDH gene that affects enzyme sen- sitivity to GTP-induced inhibition. The mutations identified in the patients with hyperinsulinism and hyperammonemia (Stanley et al., 1998) lie exactly within a sequence of 15 amino acids that we previously suggested to contain the GTP binding site of the brain GDH isoproteins (Cho et al., 1996). To our knowledge, a comparison of the detailed structures and functions of any GDH isoproteins rarely has been reported.

In the present studies, the purified GDH I1 isoprotein was injected as an immunogen into BALBk mice, and several monoclonal antibodies to the protein were produced from the fusion experiments. The monoclonal antibodies, which specifically recognized GDH on western blots, were characterized and used as probes for a cross-reactivity study of the bovine brain GDH isoproteins. The interactions between specific anti-GDH I1 monoclonal antibodies and the GDH isoproteins were investigated further with a BIAcore analyzer. The BIA- core system (Pharmacia Biosensor) allows a quantitative analysis of molecular interactions in real time. Therefore, association and dissociation rate constants can be readily calculated. Our results suggest the possibility that there are structural differences in their epitopes between the two GDH isoproteins.

MATERIALS AND METHODS

Materials Bovine brains were obtained from Majang Slaughterhouse

(Seoul, Korea). NADH, bovine serum albumin, L-glutamate, EDTA, phenylmethylsulfonyl fluoride, 2-oxoglutarate, ADP, ammonium acetate, and 2-mercaptoethanol were purchased from Sigma Chemical Co. (St. Louis, MO, U.S.A.). Dulbecco’s modified Eagle’s medium (DME), penicillirdstreptomycin antibiotics, and hypoxanthine-aminopterin-thymidine (HAT) were purchased from GibcoBRL (Grand Island, NY, U.S.A.). Goat anti-mouse IgG conjugated with alkaline phosphatase was obtained from Jackson Immuno Research (West Grove, PA, U.S.A.). P20 detergent was purchased from Pharmacia Biosen- sor Ltd. (Uppsala, Sweden).

Purification of GDH isoproteins and enzyme assay GDH isoproteins were purified from bovine brains according

to a procedure developed in our laboratory (Cho et al., 1995). Enzyme activity was measured spectrophotometrically in the direction of reductive amination of 2-oxoglutarate by following the decrease in absorbance at 340 nm. All assays were performed in duplicate, and initial velocity data were correlated with a standard assay mixture containing 50 mM triethanol- amine, pH 8.0, 100 mM ammonium acetate, 0.1 mM NADH, 2.6 mM EDTA, and 1 mM ADP at 25°C. GDH concentrations were adjusted to give a measured rate of <0.04 absorbance units per minute. The reaction was started with the addition of 2-oxoglutarate to 10 mM final concentration. One unit of enzyme was defined as the amount of enzyme required to oxidize 1 pmol of NADH per minute at 25°C.

Production of anti-GDH I1 monoclonal antibodies For injection, the purified GDH I1 (50 p g in a volume of 150

pl) was mixed with an equal volume of Freund’s complete adjuvant by sonication for three 15-s bursts. The antigen- adjuvant mixture was injected into a female BALB/c mouse (6-8 weeks old). The first injection was followed by three booster injections at 3-4-week intervals. The final injection was given 3 days before the cell fusion without adjuvant. The feeder layer cells were prepared 1 day before fusion. A 12-18- week-old BALB/c mouse was killed by cervical dislocation, and its abdominal skin was removed carefully. Five milliliters of ice-cold 11.6% sucrose solution was injected into the peritoneal cavity, -3 ml of the injected solution was pulled out, and peritoneal cells were collected by centrifugation for 5 min at 650 g.

The fusion experiments were performed as follows (Choi et al., 1995). In brief, spleen cells released by tearing the removed spleen with fine forceps were collected in a 15-ml centrifuge tube. Then the prepared spleen cells and SP2/o-Ag-14 mouse myeloma cells (Shulman et al., 1978) were combined, and 1 ml of 50% polyethylene glycol 1500 in DME (serum-free) was added slowly. The fusion process was allowed to continue for 90 s at 37°C and stopped by adding DME. To avoid an osmotic shock, 1 ml of DME was added slowly for the first 1 min, and 2 ml was added for the next 1 min. For a period of 10 min, a total of 20 ml of DME was added. Cells were collected by centrifugation for 1 min at 650 g, suspended in 20 ml of selective HAT medium (DME supplemented by 20% fetal bovine serum, antibiotics, and HAT) carefully by swirling, and centrifuged for 1 min at 650 g. The cells were resuspended in 120 ml of HAT medium, and 1 ml of cell suspension was transferred into each well of five 24-well plates. About 2 weeks after the fusion, culture supernatants were collected and first screened by immunodot-blot analysis with purified enzyme as an antigen and then by western blot analysis. Positive clones selected by the screening methods were transferred to six-well plates, grown in tissue culture flasks (75 cm’), and frozen in a liquid nitrogen tank. All positive clones were frozen first and cloned by limiting dilution after thawing. For cloning of a single specific antibody-secreting cell, aliquots of cultured cells were diluted in fresh DME and counted using a hemocytome- ter. Samples to be cloned were diluted in hypoxanthine-thymidine (HT) medium to 15 cells/ml. Seventy microliters of well suspended sample was plated in each well of a 96-well plate, and 140 pl of fresh HT medium was added. It was fed at days 5 and 12 with two drops of medium. The cells of each well were expanded and reassayed by western blot analysis.

J. Neurochem., Vol. 72, No. 5, 1999

2164 S. Y. CHOI ET AL.

Immunoblot analysis For immunodot-blotting, small squares (1 X 1 cm) were

drawn on a sheet of nitrocellulose paper (10 X 10 cm) and marked by numbering. One microliter of antigen solution (0.5 mg/ml) was applied onto each square and air-dried. The blots were incubated for 1 h in Blotto [2% nonfat dry milk in Tris-buffered saline (TBS)], rinsed briefly with TBS, and air- dried. The blots were processed by the procedures described for western blotting. For western blotting, proteins separated by sodium dodecyl sulfate (SDS)-polyacrylamide gel electrophoresis were transferred to nitrocellulose membranes (Towbin et al., 1979), and the membranes were rinsed briefly in distilled water and air-dried. The blots were blocked with Blotto for 1 h. After rinsing with TBS, the blots were incubated in culture supernatants for 1 h and washed three times in TBS containing Tween 20 at 5-min intervals. The blots were treated with alkaline phosphatase-conjugated goat anti-mouse IgG for 1 h and washed three times at 5-min intervals with TBS containing Tween 20. Following the final rinse for 5 min with an alkaline phosphatase buffer (100 mM Tris-HC1 and 5 mM MgCl,, pH 9.5), color reaction was started by incubating the blots in alkaline phosphatase buffer containing nitroblue tetrazolium and bromochloroindolyl phosphate. For 10 ml of solution, 60 p l of nitroblue tetrazolium (50 mg/ml in 70% dimethylformamide) and 30 p l of bromochloroindolyl phosphate (50 mg/ml in 100% dimethylformamide) were added to the 10 ml of alkaline phosphatase buffer. When the color reaction had reached the desired intensity, the reaction was stopped by rinsing the membranes with several changes of distilled water. The blots were photographed while still moist.

Purification of monoclonal antibodies For purification of monoclonal antibodies, 100 ml of the

culture supernatant was centrifuged for 30 min at 15,000 g to clarify cells and insoluble aggregates and applied onto 1 ml of a protein A-agarose column (Sigma). The column was washed with phosphate-buffered saline until the absorbance of unbound proteins came down to the background level, and antibodies were eluted with 0.1 M glycine-HC1, pH 2.5. The eluted antibodies were neutralized with the addition of 1 M Tris and dialyzed against phosphate-buffered saline.

Cell culture Human neuroblastoma SK-N-SH (ATCC no. HTB11) and

human glioblastoma U-373-MG (ATCC no. HTB17) were obtained from Korean Cell Line Bank. All cell lines were grown in DME containing 10% fetal bovine serum according to the supplier’s instruction.

Cross-reactivities among mammalian brains Several animal brains from dog, cat, cow, pig, rabbit, and

chicken were removed and homogenized in 10 mM potassium phosphate containing 0.1 mM EDTA, 1 mM 2-mercaptoethanol, and 1 mM phenylmethylsulfonyl fluoride. The individual 25% (wt/vol) homogenates were centrifuged at 10,000 g for 1 h, and 5 p1 of each supernatant was mixed with an equal volume of 2X SDS sample buffer and boiled for 3 min. The cooled samples were subjected to SDS-polyacrylamide gel electrophoresis and transferred to nitrocellulose membranes. The blots were processed by the procedures described above for western blotting. In the case of human brain, total proteins were prepared by homogenizing a small fragment of cerebral cortex removed from a 45-year-old man who had surgery after an accident.

Epitope mapping One-dimensional epitope mapping was carried out according

to a procedure previously described (Choi et al., 1995). Ten micrograms of purified GDH isoproteins in SDS sample buffer was mixed with an equal volume of Staphylococcus aureus V-8 protease solution (0.5 p g in SDS sample buffer). The mixtures were applied onto an SDS-polyacrylamide gel, and the separated peptides were transferred for immunoblotting analysis as described above.

Immobilization and analysis of proteins on BIAcore Protein-protein interaction between monoclonal antibodies

and GDH isoproteins was performed using a Pharmacia Bio- sensor BIAcore instrument. CM5 research-grade sensor chips (Pharmacia Biosensor) were used for all experiments. The indirectly oriented immobilization of antibodies on the CM5 sensor chip was carried out as follows. First, rabbit anti-mouse IgG Fc (ramfc) was coupled to the chip by injecting 100 ng of ramfc in 10 mM sodium acetate, pH 4.5, at a flow rate of 5 pl/min at 20°C. The carboxyl-methyl dextran matrix of the sensor chip was activated using a 30-pl (6 min) injection of a mixture of 0.2 M 1-ethyl-3-[(3-dimethylamino)propyl]carbodi- imide and 0.05 M N-hydroxysuccinimide in water to convert the carboxyl group of the sensor chip matrix to an N-hydroxysuccinimide ester. This ester is susceptible to nucleophilic attack by amino groups of proteins, resulting in an amide linkage of the protein to the sensor chip. Under these conditions, typically 3,700 resonance units of ramfc were immobilized on the CM5 chip. The interactions of GDH isoproteins with monoclonal antibodies were measured by two subsequent injections; the monoclonal antibodies were captured by ramfc, and followed by GDH isoproteins. Protein-protein interaction studies were carried out in HEPES-buffered saline (10 mM HEPES/KOH, pH 7.5, 150 mM NaCI, 3.4 mM EDTA, 0.005% surfactant P20). Kinetic rate constants (Icon and it,,,) and the equilibrium dissociation constants ( K D = Icoff/kon) were determined using the BIAlogue Kinetics Evaluation Software.

RESULTS

Production and characterization of monoclonal antibodies

GDH isoproteins were purified according to a procedure developed in our laboratory (Cho e t al., 1995) and exhibited a single protein band on an SDS-polyacrylamide gel. To enhance the immunogenicity of the protein and obtain antibodies with a better reactivity on western blot, purified enzyme was denatured in the presence of SDS and injected into animals. From two fusion experiments, 46 positive clones were initially screened by immunodot-blot analysis. Because goat anti-mouse IgG antibody was used as a second antibody, all monoclonal antibodies screened by the procedure are IgG classes. Among the hybridomas, some clones lost the ability to produce monoclonal antibodies continually or produced monoclonal antibodies that reacted weakly with the protein on western blots and thus were discarded. Eighteen hybridomas of the 46 clones were finally selected for further study. Seven representative monoclonal antibodies purified by using a protein A-affinity column are shown in Fig. 1. To check the specificity of the anti-GDH I1 monoclonal antibodies, total proteins of bovine brain were extracted, separated by SDS-polyacrylamide gel


CROSS-REACTIVITIES OF BRAIN GDH ISOPROTEINS

1 2 3 4 5 6 1 8

f 55 kDa

57kDa -?)

f 29kDa f 24kDa

FIG. 1. SDS-polyacrylamide gel electrophoresis of purified anti- GDH II monoclonal antibodies. A total of 46 anti-GDH II monoclonal antibodies was initially selected by immunodot-blot analysis, and seven representative antibodies purified by a protein A-agarose affinity column are shown here. Lane 1, gdhmAbl; lane 2, gdhmAb2; lane 3, gdhmAb3; lane 4, gdhmAb4; lane 5, gdhmAb5; lane 6, gdhmAb6; lane 7, gdhmAb7; lane 8, molecular mass marker proteins.

electrophoresis, and immunoblotted with the monoclonal antibodies. The antibodies specifically recognized a protein band corresponding to the position of purified GDH on an SDS-polyacrylamide gel (data not shown).

Cross-reactivity of the anti-GDH I1 monoclonal antibodies with GDH from other mammalian and avian species

To examine the cross-reactivity of the anti-GDH II monoclonal antibodies with other mammalian and avian GDH enzymes, brains from a dog, cat, cow, pig, rabbit, human, and chicken were removed, and total proteins of the brain homogenates were separated, transferred, and probed with the seven monoclonal antibodies. All seven monoclonal antibodies recognized the GDH in the same manner in the animal species tested, and the result of one immunoblot with gdhIImAb1 is shown in Fig. 2. Be- cause the monoclonal antibodies recognized GDH from several mammalian (including human) and avian species, it was of interest to examine the immunoreactivity of the

(A) (B) 1 2 3 4 5 6 7 1 2 3 4 5 6 7

57kDa-c

FIG. 2. Cross-reactivities of GDH from some mammalian and an avian species with anti-GDH II monoclonal antibodies. Animal brains were removed, and total proteins of the brain homogenates were immunoblotted with the anti-GDH II monoclonal antibodies. All of the seven monoclonal antibodies recognized the GDHs in the same manner in the animal species tested, and only one immunoblot with gdhllmAbl is shown. Lane 1, cow; lane 2, human; lane 3, pig; lane 4, chicken; lane 5, rabbit; lane 6, cat; lane 7, dog. A SDS-polyacrylamide gel electrophoresis of total proteins of each brain homogenate. B Corresponding immunoblot probed with monoclonal antibody (gdhllmAb1).

2165

1 2 3

FIG. 3. lmmunoblots of total proteins from human neuronal and glial cells with anti-GDH II antibodies. The cells were grown in DME containing 10% fetal bovine serum, and total proteins were extracted and immunoblotted with monoclonal antibody, gdhll- mAbl . Lane 1, bovine brain homogenate as a control; lane 2, human neuroblastorna SK-N-SH; lane 3, human glioblastoma U-373-MG.

anti-GDH 11 monoclonal antibodies with GDH from the other human cell lines. Total proteins from human neuroblastoma and glioblastoma cells were extracted and immunoblotted with the monoclonal antibodies. As shown in Fig. 3, the monoclonal antibodies also recognized the same protein band in the immunoblots of total proteins of the human cell lines. These results of the cross-reactivities of brain GDH from various species suggest that the GDH enzymes are related immunologically to each other.

Recognition of the anti-GDH I1 monoclonal antibodies to the GDH I isoprotein

As the structural information about GDH isoproteins is not yet available, it was of interest to compare the immunoreactivities of the anti-GDH II monoclonal antibodies with the GDH isoproteins. To determine if the anti-GDH I1 monoclonal antibodies interact with GDH I and GDH I1 differently, samples of the purified enzymes were incubated with the purified anti-GDH I1 monoclonal antibodies. The results of immunoblotting analysis showed that the immunoreactivities of the anti-GDH I1 monoclonal antibodies with GDH I were different from those with GDH II. All seven anti-GDH 11 monoclonal antibodies recognized GDH II on the immunoblot (Fig. 4A). Those differences were analyzed further with an inhibition test by measuring remaining activities of the two isoproteins after a 1-h incubation with each antibody in 0.1 M potassium phosphate, pH 7.2, at room temperature. Among the seven anti-GDH I1 monoclonal antibodies tested, three monoclonal antibodies (gdhIImAb4, gdhIImAb5, and gdhIImAb7) inhibited GDH I1 activity, and the extent of the inhibition was a maximum of 18% (gdhIImAb7), whereas no inhibition was observed following treatment by the other four monoclonal antibodies as shown in Fig. 4B. Unlike the interaction of the antibodies with GDH II, three (gdhIImAb3, gdhIImAb5, and gdhIImAb6) of seven anti-GDH II monoclonal antibodies did not recognize the protein band on the immunoblot of GDH I as shown in Fig. 5A. When the

J. Neurochem., Vol. 72, No, 5, 1999


57kDa- 50IrDa-

40 kDa - 25 kDa-

1 2 3 4 5 6 7 Monoclonal antibadies

FIG. 4. lmmunoblots and inhibition of GDH II probed with anti- GDH II monoclonal antibodies. A lmmunoblot analysis of GDH II probed with representative anti-GDH I1 monoclonal antibodies. Lane 1, gdhmAbl; lane 2, gdhmAb2; lane 3, gdhmAb3; lane 4, gdhmAb4; lane 5, gdhmAb5; lane 6, gdhmAb6; lane 7, gdhmAb7. 6 Inhibition of GDH II activity by anti-GDH II monoclonal antibodies. Samples (1 pg in 10 pl of 0.1 M potassium phosphate, pH 7.2) of purified GDH II were incubated with anti- GDH II monoclonal antibodies (10 pg in 90 pI in phosphate- buffered saline) for 1 h at room temperature. Then the remaining activities of the samples were determined by adding standard assay mixtures. As a control, fresh culture supernatant was used.

inhibition tests of GDH I with the seven anti-GDH I1 monoclonal antibodies were performed, only one monoclonal antibody (gdhIImAb7) inhibited GDH I activity as shown in Fig. 5B, in contrast to the result that three anti-GDH I1 monoclonal antibodies inhibited the GDH I1 activity (Fig. 4B). These results show striking differences both between inhibition patterns for the two isoproteins and between the patterns of inhibition and of recognition in the immunoblot analysis.

The different immunoreactivities of the anti-GDH I1 monoclonal antibodies with GDH isoproteins were examined further by epitope mapping analysis with V-8 protease. GDH I and GDH I1 were digested with V-8 protease and immunoblotted with anti-GDH I1 monoclonal antibodies. The results in Fig. 6A show that two subgroups recognizing different peptide fragments of

57 kDa - 1 2 3 4 5 6 7

Monoclonsl antibadies

FIG. 5. lmmunoblots and inhibition of GDH I probed with anti- GDH II monoclonal antibodies. A lmmunoblot analysis of GDH I probed with representative anti-GDH II monoclonal antibodies. Lane 1, gdhmAbl; lane 2, gdhmAb2; lane 3, gdhmAb3; lane 4, gdhmAb4; lane 5, gdhmAb5; lane 6, gdhmAb6; lane 7, gdhmAb7. B: Inhibition of GDH I activity by anti-GDH II monoclonal antibodies. Inhibition assays of GDH I with seven anti- GDH II monoclonal antibodies were performed as described in Fig. 48.

(A) (B) I 2 3 4 5 6 7 1 2 3 4 5 6 7

FIG. 6. lmmunoreactivities of anti-GDH II monoclonal antibodies with GDH isoproteins digested with V-8 protease. The purified GDH I (A) and GDH II (B) were digested with V-8 protease and separated on 10-20% gradient SDS-polyacrylamide gel. The separated peptides were transferred and immunoblotted with anti-GDH I1 monoclonal antibodies. Lane 1, gdhmAb1; lane 2, gdhmAb2; lane 3, gdhmAb3; lane 4, gdhrnAb4; lane 5, gdhmAb5; lane 6, gdhmAb6; lane 7, gdhmAb7.

GDH I1 were identified among the antibodies tested. The monoclonal antibodies of group I (gdhIImAb1 and gdhII- mAb2) showed two bands at 50 and 25 kDa, indicating that the epitopes recognized by the two antibodies are located near one another or on the same site. Unlike the monoclonal antibodies of group I, the monoclonal antibodies of group I1 (gdhIImAb3-7) showed an extra band at 40 kDa with a diminishing band at 25 kDa (Fig. 6A). When the epitope mapping of the anti-GDH I1 monoclonal antibodies with GDH I was performed under the same experimental conditions, the results with GDH I were quite distinct from those with GDH I1 (Fig. 6B). The monoclonal antibodies of group I (gdhIImAb1 and gdhIImAb2) showed a major band at 25 kDa, whereas the band at 50 kDa was barely visible. Among the monoclonal antibodies of group I1 (gdhIImAb3-7), only gdhIImAb4 and gdhIImAb7 reacted with GDH I, showing two bands at 40 and 25 kDa (Fig. 6B). Therefore, whereas all seven anti-GDH I1 antibodies reacted with GDH 11, only four antibodies (gdhIImAb1, gdhIImAb2, gdhIImAb4, and gdhIImAb7) showed immunoreactivities with GDH I and three antibodies (gdhIImAb3, gdhII- mAb5, and gdhIImAb6) did not interact with GDH I. These results are consistent with those shown in Fig. 5 and suggest that some epitopes of GDH I are different from those of GDH 11.

Protein-protein interactions of anti-GDH I1 monoclonal antibodies with GDH I and GDH I1

To compare further the different antigenic reactivities of anti-GDH I1 monoclonal antibodies with GDH I and GDH I1 quantitatively, we examined the interactions directly in the Pharmacia BIAcore. The monoclonal antibody used for this study was gdhIImAb5, which showed the characteristic difference in immunoreactivity between the two GDH isoproteins (Figs. 4-6). By using the methods described above, k,, and k,, values were calculated for GDH I and GDH 11. Each measurement was done at least twice and up to four times on different surfaces. The results of kinetic experiments are summa- rized in Table 1. The binding affinity of anti-GDH I1 monoclonal antibody for GDH I1 (KD = 1.0 nM) was

J. Neurochem., Voi. 72, No. 5, 1999


TABLE 1. Interaction of anti-GDH II monoclonal antibody with GDH I and GDH II

Equilibrium Association rate Dissociation rate dissociation constant (k,,,,) constant (kofF) constant

GDH isotvne (M-' s-') (S-') (KrJ (nM)

GDH I 1.1 x 104 5.8 x 10-5 5.3 5 0.63 GDH I1 5.2 x 104 5.0 X lops 1.0 2 0.15

Results are the averages of two separate experiments, with the error expressed as the range of the two data sets.

fivefold higher than for GDH I (KD = 5.3 nM). Both GDH I and GDH I1 showed a very slow dissociation, almost close to the limit of the machine (-1 X lop5 s-'), and there were no differences in dissociation rate constant (koff) between the GDH isoproteins. The difference between GDH I and GDH I1 in their binding affinity for gdhIImAb5 was caused mainly by their association rate constants (Icon), 1.1 1 x lo4 M-' s-l and5.21 X lo4 M-' s-' for GDH I and GDH 11, respectively. These results indicate that the molecular recognition processes of gdhIImAb4 to GDH I and to GDH I1 are different. It is therefore reasonable to suggest that the conformation of protein epitope surface on GDH I is different from that on GDH 11.

DISCUSSION

Partial deficiency of GDH isoproteins has been reported in some patients with cerebellar degeneration, suggesting that the enzymes are important in brain function (Hussain et al., 1989; Plaitakis et al., 1993). At present, the functional significance of GDH in nerve tissue remains uncertain. The observations that one of the GDH isoproteins is reduced in patients with multisys- temic neurological disorders have been obtained by many (Konagaya et al., 1986; Abe et al., 1992), but not all (Aubby et al., 1988; Duvoisin et al., 1988) investigators. The existence of the hyperinsulinism- hyperam- monemia syndrome also highlights the importance of GDH in the regulation of insulin secretion and indicates that GDH has an important role in regulating hepatic ureagenesis (Fahien et al., 1988; Bryla et al., 1994; Stanley et al., 1998). Further studies of GDH and its isotypes may provide explanations for the absence of CNS symptoms due to hyperammonemia (Zammarchi et al., 1996; Weinzimer et al., 1997).

Molecular biological studies revealed that multiple GDH-specific genes are present in the human and at least two of these genes are functional (Mavrothalassitis et al., 1988; Shashidharan et al., 1994). Similar studies also showed the presence of two GDH activities in rat brain differing in their relative resistance to thermal inactivation, solubility, and allosteric regulation characteristics (Colon et al., 1986). Recent studies in our laboratory have shown that two different GDH isoproteins (GDH I and GDH 11) are also present in bovine brain (Cho et al.,

1995). Unlike most previous reports, which present a soluble and a particulate form of GDH (Plaitakis et al., 1993; Rajas and Rouset, 1993), both GDH I and GDH I1 were readily solubilized and no detergents were required for the initial extraction step (Cho et al., 1995). The purified GDH I and GDH I1 showed different characteristics in their heat stability, sensitivities to the action of ADP, and remaining activities after limited proteolysis with trypsin.

Although the origin of the GDH polymorphism is not understood, recent studies in our laboratory have suggested that the two GDH isoproteins are different gene products rather than the results of posttranslational mod- ifications based on their differences in amino acid sequences of N-terminal amino acid sequences (Cho et al., 1995) and GTP-binding site (Cho et al., 1996). This possibility is supported by previous studies suggesting that human GDH is encoded by a multigene family (Michaelidis et al., 1993), as well as by the demonstra- tion of two loci for human GDH on chromosomes 10 and X, respectively (Shashidharan et al., 1994). However, remarkably little is known about the structural and functional differences between the GDH isoproteins due to the lack of detailed information for the three-dimensional structure of any mammalian GDHs, although a very recent study has reported the crystallization of bovine liver GDH for the first time from mammalian sources (Peterson et al., 1997). In the present study, we have produced a library of monoclonal antibodies raised against bovine brain GDH I1 and present the first use of the monoclonal antibodies to examine the structural re- lationship of the GDH isoproteins.

From the immunoblotting analysis of the cross-reactivities test with anti-GDH I1 monoclonal antibodies to GDH I, it is of interest that GDH I is immunologically different from GDH 11. Whereas all seven anti-GDH I1 monoclonal antibodies recognized GDH I1 on the immunoblots of GDH I1 (Fig. 4A), only four antibodies recognized the protein band in the immunoblot of GDH I (Fig. 5A). When the inhibition tests of the GDH isoproteins with the seven anti-GDH I1 monoclonal antibodies were performed, only one monoclonal antibody inhibited GDH I activity (Fig. 5B) in contrast to the result that three anti-GDH I1 monoclonal antibodies inhibited the GDH I1 activity (Fig. 4B). The results from the epitope mapping analysis (Fig. 6) also support the possibility that GDH I could be different from GDH I1 either in amino acid sequences or in protein structure. To obtain more information from the epitope mapping analysis, N- terminal sequence analysis of the proteins obtained from well separated bands is in progress in our laboratory.

To understand further the interactions of anti-GDH I1 monoclonal antibodies with GDH I and GDH 11, we examined the interactions directly in the Pharmacia BIA- core. The equilibrium binding constant between antigen and antibody can be measured in a variety of ways, as long as the complex can be separated from free ligand once the reaction has reached equilibrium. However, few methods allow the analysis of the interaction in real time



and thus the determination of kinetic rate constants. The biosensor technology uses the optical phenomenon of surface plasmon resonance to monitor the interaction of an immobilized ligand to a protein in the flow solution that is passed over it (Fagerstam et al., 1992; Malmqvist, 1993). The binding affinity of the anti-GDH I1 monoclonal antibody for GDH I1 was fivefold higher than for GDH I (Table 1). There were big differences between the GDH isoproteins in their association rate constant (ken), whereas no significant differences were observed in their dissociation rate constant (koff). These results indicate that the molecular recognition processes of anti-GDH I1 monoclonal antibodies to GDH I and GDH I1 are different. It is therefore quite possible that there are differences between GDH I and GDH I1 in their tertiary or quarter- nary structure of this epitope, although GDHs are known to have very high sequence similarities among various species (Michaelidis et al., 1993).

On the basis of the specificity of monoclonal antibodies characterized by western blot analysis, inhibition studies, epitope mapping, and biosensor technology, the results in the present work suggest that there are structural differences in their epitopes between the two GDH isoproteins from bovine brain. Taken together with the differences in their amino acid sequences and in their regulatory properties (Cho et al., 1995, 1996; Shashidha- ran et al., 1997), the present findings support the possibility that different types of GDHs may function differently in a biological system, as many proteins have functions distinct from those for which they were origi- nally identified. Actually, other roles of GDHs have been reported. For instance, a membrane-bound form of GDH possesses a microtubule-binding activity (Rajas et al., 1996), and GDH reacts as an RNA-binding protein and shows a possible role in the regulation of transcription (Preiss et al., 1993; McDaniel, 1995; Bringaud et al., 1997). Recently, Cavallaro et al. (1997) have identified GDH as one of the late memory-related genes in the hippocampus, and Frattini et al. (1997) have identified GDH as a new member of the ring finger gene family in

It would appear that we have just begun to unravel the mystery of GDHs and their role in a biological system. To our knowledge, comparison of the detailed structure and function of the GDH isoproteins rarely has been reported. Therefore, further studies are required to elucidate the physiological roles of various types of GDH isoproteins. An issue not addressed in this work is whether our GDH isoproteins are the bovine counterparts of the human brain pair (Shash- idharan et al., 1994, 1997). Further sequence characterization is in progress.

Xq24-25.

Acknowledgment: This work was supported by a Korea Science and Engineering Foundation grant (95-0403- 12) and by the Ministry of Science and Technology (97-Nl-02-03-A- 09) to S. Y. Choi.

REFERENCES

Abe T., Ishiguro S. I., Saito H., Kiyosawa M., and Tamai M. (1992) Partially deficient glutamate dehydrogenase activity and attenu- ated oscillatory potentials in patient with spinocerebellar degeneration. Invest. Ophthalmol. Vis. Sci. 33, 447-452.

Aubby D., Saggu H. K., Jenner P., Quinn N. P., Harding A. E., and Marsden C. D. (1988) Leukocyte glutamate dehydrogenase activity in patients with degenerative neurological disorders. J. Neurol. Neurosurg. Psychiatry 51, 893-902.

Baker P. J., Britton K. L., Rice D. W., Rob A,, and Stillman T. J. (1992) Structural consequences of sequence patterns in the fingerprint region of the nucleotide binding fold. J. Mol. B i d . 228, 662-671.

Bringaud F., Stripecke R., French G. C., Freedland S., Turck C., Byme E. M., and Simpson L. (1997) Mitochondria1 glutamate dehydrogenase from Leishmania tarentolae is a guide RNA-binding protein. Mol. Cell. Biol. 17, 3915-3923.

Bryla J., Michalik M., Nelson J., and Ereciriska M. (1994) Regulation of the glutamate dehydrogenase activity in rat islets of Langerhans and its consequence on insulin release. Metabolism 43, 11 87- 1195.

Cavallaro S., Meiri N., Yi C. L., Musco S., Ma W., Goldherg J., and Alkon D. L. (1997) Late memory-related genes in the hippocampus revealed by RNA fingerprinting. Proc. Natl. Acad. Sci. USA 94, 9669-9673.

Cho S.-W. and Lee J. E. (1996) Modification of brain glutamate dehydrogenase isoproteins with pyridoxal 5’-phosphate. Bio- chimie 78, 817-821.

Cho S.-W., Lee J., and Choi S. Y. (1995) Two soluble forms of glutamate dehydrogenase isoproteins from bovine brain. Eur. J. Biochem. 233, 340-346.

Cho S.-W., Ahn J.-Y., Lee J., and Choi S. Y. (1996) Identification of a peptide of the guanosine triphosphate binding site within brain glutamate dehydrogenase isoproteins using 8-azidoguauosiue triphosphate. Biochemistry 35, 13907-1391 3.

Cho S.-W., Cho E. H., and Choi S. Y. (199th) Activation of two types of brain glutamate dehydrogenase isoproteins by gabapentin.

Cho S.-W., Yoon H.-Y., Ahn J.-Y., Choi S. Y., and Kim T. U. (19986) Identification of an NAD+ binding site of brain glutamate dehydrogenase isoproteins by pbotoaffinity labeling. J. Biol. Chem. 273, 31125-31130.

Choi E. Y., Park S. Y., Jang S. H., Song M.-S., Cho S.-W., and Choi S. Y. (1995) Production and characterization of monoclonal antibodies to bovine brain succinic semialdehyde reductase. J. Neu- rochem. 64, 371-377.

Colon A. D., Plaitakis A., Perakis A., Berl S., and Clarke D. D. (1986) Purification and characterization of a soluble and a particulate glutamate dehydrogenase from rat brain. J . Neurochem. 46,181 1- 1819.

Duvoisin R. C., Nicklas W. J., Ritchie V., Sage I., and Chokroverty S. (1988) Glutamate dehydrogenase deficiency in patients with olivopontocerebellar atrophy. J. Neurol. Neurosurg. Psychiatry 33, 1322-1326.

Fagerstam L. G., Karlsson A. F., Karlsson R., Persson B., and Ronn- berg I. (1992) Biospecific interaction analysis using surface plasmon resonance detection applied to kinetic, binding site and concentration analysis. J. Chromatogr. 597, 397-410.

Fahien L. A,, MacDonald M. J., Kmiotek E. H., Mertz R. J., and Fahien C. M. (1988) Glutamate dehydrogenase. IV. Studies on enzyme inactivation and coenzyme binding. J . Biol. Chem. 263, 13610- 13614.

Fisher H. F. (1985) L-Glutamate dehydrogenase from bovine liver. Methods Enzymol. 113, 16-27.

Frattini A,, Faranda S., Bagnasco L., Patrosso C., Nulli P., Zucchi I., and Vezzoni P. (1 997) Identification of a new member (ZNF183) of the ring finger gene family in Xq24-25. Gene 192, 291-298.

Hussaiu M. H., Zannis V. I., and Plaitakis A. (1989) Characterization of glutamate dehydrogenase isoproteins purified from the cerebel- lum of normal subjects and patients with degenerative neurological disorders, and from human neoplastic cell lines. J. Biol. Chem. 264, 20730-20735.

FEBS Lett. 426, 196-200.

J. Neurochem., Vol. 72, No. 5. 1999


Kim S. W., Lee J., Song M.-S., Choi S. Y., and Cho S.-W. (1997) Essential active-site lysine of brain glutamate dehydrogenase isoproteins. J. Neurochem. 69, 418-422.

Konagaya Y., Konagaya M., and Takayanagi T. (1986) Glutamate dehydrogenase and its isoenzyme activity in olivopontocerebellar atrophy. J. Neurol. Sci. 74, 231-236.

Malmqvist M. (1993) Biospecific interaction analysis using biosensor technology. Nature 361, 186-187.

Mavrothalassitis G., Tzimagiorgis G., Mitsialis A,, Zannis V. I., Plaitakis A,, Papamatheakis J., and Moschonas N. K. (1988) Isolation and characterization of cDNA clones encoding human liver glutamate dehydrogenase: evidence for a small gene family. Proc. Natl. Acad. Sci. USA 85, 3494-3498.

McDaniel H. G. (1995) Comparison of the primary structure of nuclear and mitochondria1 glutamate dehydrogenase from bovine liver. Arch. Biochem. Biophys. 319, 3 16-321.

Michaelidis T. M., Tzimagiorgis G., Moschonas N., and Papamatheakis J. (1 993) The human glutamate dehydrogenase gene family: gene organization and structural characterization. Genomics 16, 150- 160.

Peterson P. E., Pierce J., and Smith T. J. (1997) Crystallization and characterization of bovine liver glutamate dehydrogenase. J . Struct. Biol. 120, 73-77.

Plaitakis A,, Berl S., and Yahr M. D. (1984) Neurological disorders associated with deficiency of glutamate dehydrogenase. Ann. Neu- rol. 15, 144-153.

Plaitakis A,, Flessas P., Natsiou A. B., and Shashidharan P. (1993) Glutamate dehydrogenase deficiency in cerebellar degenerations: clinical, biochemical and molecular genetic aspects. Can. J. Neu- rol. Sci. 3 (Suppl.), S109-S116.

Preiss T., Hall A. G., and Lightowlers R. N. (1993) Identification of bovine glutamate dehydrogenase as an FWA-binding protein. J. B id . Chem. 268, 24523-24526.

Rajas F. and Rouset B. (1993) A membrane-bound form of glutamate dehydrogenase possesses an ATP-dependent high-affinity microtubule-binding activity. Biochem. J. 295, 447-455.

Rajas F., Gire V., and Rouset B. (1996) Involvement of a membrane- bound form of glutamate dehydrogenase in the association

of lysosomes to microtuhules. J. Biol. Chem. 271, 29882- 29890.

Shashidharan P., Michaelidis T. M., Robakis N. K., Kresovali A,, Papamatheakis J., and Plaitakis A. (1994) Novel human glutamate dehydrogenase expressed in neural and testicular tissues and encoded by an X-linked intronless gene. J . Biol. Chem. 269, 16971- 16976.

Shashidharan P., Clarke D. D., Ahmed N., Moschonas N., and Plaitakis A. (1997) Nerve tissue-specific human glutamate dehydrogenase that is thermolabile and highly regulated by ADP. J. Neurochem.

Shulman M., Wilde C. D., and Kohler G. (1978) A better cell line for making hybridomas secreting specific antibodies. Nature 276,

Stanley C. A,, Lieu Y. K., Hsu B. Y. L., Burlina A. B., Greenberg C . R., Hopwood N. J., Perlman K., Rich B. H., Zammarchi E., and Poncz M. (1998) Hyperinsulinism and hyperammonemia in in- fants with regulatory mutations of the glutamate dehydrogenase gene. N. Engl. J. Med. 338, 1353-1357.

Towbin H., Staehelin T., and Gordon J. (1979) Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheet: procedure and some applications. Proc. Natl. Acad. Sci. USA 76, 4350-4354.

Weinzimer S. A., Stanley C. A,, Berry G. T., Yudkoff M., Tuchman M., and Thornton P. S. (1997) A syndrome of congenital hyperinsulinism and hyperammonemia. J . Pediarr. 130, 661- 664.

Yip K. S. P., Stillman T. J., Britton K. L., Artymiuk P. J., Baker P. J., Sedelnikova S. E., Engel P. C., Pasquo A,, Chiaraluce R., Con- salvi V., Scandurra R., and Rice D. W. (1995) The structure of Pyrococcus furiosus glutamate dehydrogenase reveals a key role for ion-pair networks in maintaining enzyme stability at extreme temperatures. Structure 3, 1147-1 158.

Zammarchi E., Filippi L., Novembre E., and Donati M. A. (1996) Biochemical evaluation of a patient with a familial form of leucine- sensitive hypoglycemia and concomitant hyperammonemia. Me- tabolism 45, 957-960.

68, 1804-1811.

269-270.


different antigenic reactivities of bovine brain glutamate dehydrogenase isoproteins

Documents