Comparative Study on Recombinant Chloroplastic and Cytosolic Ascorbate Peroxidase Isozymes of Spinach

ARCHIVES OF BIOCHEMISTRY AND BIOPHYSICS

Vol. 353, No. 1, May 1, pp. 55– 63, 1998 Article No. BB970612

Comparative Study on Recombinant Chloroplastic and Cytosolic Ascorbate Peroxidase Isozymes of Spinach1 Kazuya Yoshimura, Takahiro Ishikawa,* Yoshihiro Nakamura,† Masahiro Tamoi, Toru Takeda, Toshiji Tada,† Keiichiro Nishimura,† and Shigeru Shigeoka2 Department of Food and Nutrition, Faculty of Agriculture, Kinki University, 3327-204 Nakamachi, Nara 631-8505, Japan; *Department of Biochemistry, Wakayama Medical College, 27 KyuBancho, Wakayama 640-8155, Japan; and †Research Institute for Advanced Science and Technology, Osaka Prefecture University, Gakuen-cho 1-2, Sakai, Osaka 599-8570, Japan

Received October 24, 1997, and in revised form January 16, 1998

The spinach stromal, thylakoid-bound, and cytosolic ascorbate peroxidase isozymes (EC 1.11.1.11) were overexpressed in Escherichia coli, and their enzymatic properties were compared with the respective native isozymes. The purification of the recombinant stromal and cytosolic ascorbate peroxidases using the conventional column chromatography yielded 0.73 and 2.2 mg of protein/liter of bacteria culture with enzyme activities of 800 and 486 mmol min21 mg protein21, respectively. In every respect, the recombinant stromal, thylakoid-bound, and cytosolic ascorbate peroxidase isozymes exhibited identical enzymatic properties with each native isozyme. Specifically, the recombinant stromal and thylakoid-bound ascorbate peroxidase isozymes showed high utilization of ascorbate as an electron donor and had a very short lifetime in ascorbate-depleted medium. Polyclonal antibodies raised against both purified recombinant stromal and cytosolic ascorbate peroxidase isozymes were prepared. Both antibodies showed a cross-reaction with the recombinant and native ascorbate peroxidase isozymes. © 1998 Academic Press Key Words: ascorbate peroxidase; isozyme; recombinant enzyme; chloroplasts; cytosol; spinach.

1 The nucleotide sequence data reported have been submitted to the DDBJ, EMBL, and GenBank nucleotide sequence databases with the following accession numbers: D83669 for cDNA of stromal AsAP, D77997 for thylakoid-bound ascorbate peroxidase, and D85864 for cytosolic ascorbate peroxidase. 2 To whom correspondence should be addressed. Fax: 181-742-432252. E-mail: [email protected].

0003-9861/98 $25.00 Copyright © 1998 by Academic Press All rights of reproduction in any form reserved.

Ascorbate peroxidase (AsAP3; EC 1.11.1.11) is known to play a key role in scavenging excess hydrogen peroxide (H2O2) generated by many diverse cellular reactions (1, 2). Higher plants have developed chloroplastic isozymes which exist as stromal soluble (sAsAP) and thylakoidbound (tAsAP) forms (1–4). AsAP isozymes have also been detected in a cytosolic soluble form (cAsAP) and glyoxysome-bound form (gAsAP) (5–11). One of the characteristic properties of AsAP isozymes is that they are very labile in AsA-depleted medium; especially, chloroplastic isozymes (sAsAP and tAsAP) lost their activities within several minutes compared to the cytosolic forms (2). These facts indicate one of the reasons why it is difficult to obtain large amounts of the highly purified chloroplastic AsAP isozymes. Chloroplastic AsAP isozymes have therefore been purified only from spinach and tea leaves (5, 12). On the other hand, the cAsAP has been purified from many plant species (5– 8, 13). cDNAs for cAsAP encoded by nuclear genes have been isolated and characterized from several plant sources including spinach (14 –16). To date, the recombinant pea cAsAP has been used for structural study, and the crystal structure of its recombinant protein has been refined to an R 5 0.19 for data between 8.0 and 2.2 Å resolution (17, 18). In a previous study, we demonstrated the first complete cloning of cDNAs and the nuclear gene encoding sAsAP and tAsAP from spinach leaves. The molecular characterization indicates 3 Abbreviations used: AsA, ascorbate; AsAP, ascorbate peroxidase; cAsAP, cytosolic ascorbate peroxidase; gAsAP, glyoxysome-bound ascorbate peroxidase; IPTG, isopropyl b-D-thiogalactopyranoside; LB, Luria-Bertani broth; sAsAP, stromal ascorbate peroxidase; tAsAP, thylakoid-bound ascorbate peroxidase; PBS, phosphate-buffered saline; BSA, bovine serum albumin; PMSF, phenylmethylsulfonyl fluoride.

55

56

YOSHIMURA ET AL.

that both chloroplastic AsAP isozymes arise from a common pre-mRNA by alternative splicing of two 39terminal exons (4, 19). Advances in the expression of the recombinant AsAP isozymes utilizing these cDNA clones, as shown in Fig. 1, would therefore provide new approaches for characterization of the structure and function of each AsAP isozyme of higher plants. In this paper, we have introduced the cDNA clones encoding spinach chloroplastic and cAsAP isozymes into a pET expression vector to overexpress these proteins in Escherichia coli and have obtained the purified recombinant sAsAP and cAsAP by conventional column chromatography. The recombinant AsAP isozymes obtained were used for comparative studies on the native AsAP isozymes from spinach leaves. MATERIALS AND METHODS Materials. The cDNAs encoding spinach AsAP isozymes were originally cloned into plasmid pBluescript SK(1) (4, 16). Plasmid pET-3a and its companion production E. coli strain BL21(DE3)pLysS were obtained from Novagen, Inc. (Madison, WI). The E. coli strain DH5a was from Gibco (Gaithersburg, MD). The molecular biology reagents and enzymes were of analytical grade and were purchased from commercial sources. Construction of spinach AsAP isozymes expression plasmid. The complete coding sequences of spinach AsAP isozymes were derived from a cDNA library constructed from greening cotyledons (4, 16). For the construction of plasmid to express the AsAP isozymes, the DNA fragments encoding the mature form were amplified by PCR. The oligonucleotide primers used for the amplification of the spinach AsAP isozyme gene into the pET3a vector are shown below (Fig. 1): P-1: P-2: P-3: P-4: P-5:

59-TTTAGCACGCATATGTACGCTTCTGATCC-39, 59-TGGATCCTTTAATCCTTGTTAGATG-39, 59-ACTCATGGATCCACAAATCAATTTCCCGC-39, 59-AGACTTTAACATATGGGAAAGAGC-39, and 59-GGATCCTTCTATCAGTCTTTCTCC-39.

The sequence was homologous to the spinach AsAP isozyme cDNA except for the replacement of the original nucleotides, which introduced the desired restriction sites (bold sequences). The DNA fragments coding the mature form of sAsAP and tAsAP were amplified using oligonucleotides P-1/P-2 and P-1/P-3, respectively. The DNA fragment encoding cAsAP was amplified using oligonucleotides P-4 and P-5. PCR amplification was carried out in a 100-ml reaction mixture containing 10 ml of 103 PCR Buffer (Mg21 free), 1.5 mM MgCl2, 200 mM each dNTP, 2.5 units of recombinant Taq DNA polymerase (Takara, Japan), 1.0 mM each oligonucleotide, and 5 ng of template cDNA. The conditions of the PCR were as follows: 30 cycles at 94°C for 1 min, 60°C for 1 min, and 72°C for 2 min. The DNA fragments generated by PCR were purified by gel electrophoresis and ligated into pT7Blue T-vector (Novagen, WI). Following transformation of E. coli strain DH5a cells, clones of candidate mutants were isolated and sequenced across the region of interest by the dideoxy chain-primer method in order to establish the fidelity of all AsAP isozyme construction. From these plasmids digested with NdeI and BamHI, a 0.90-kbp DNA fragment encoding the mature form of sAsAP, a 1.05-kbp DNA fragment encoding the mature form of tAsAP, or a 0.78-kbp DNA fragment encoding cAsAP was isolated and integrated into the pET-3a expression vector treated with the same restriction enzymes and then introduced into the E. coli strain DH5a. Plasmid DNA was prepared from the ampicillin-resistant transformants and verified by digestion with the restriction enzymes NdeI and BamHI. The resulting constructs, designated pET/sAsAP,

pET/tAsAP, and pET/cAsAP, were introduced into the E. coli strain BL21(DE3)pLysS to test each one’s ability to direct the synthesis of recombinant spinach AsAP isozymes (Fig. 2). Production of spinach recombinant AsAP isozymes in E. coli. Competent E. coli BL21(DE3)pLysS transformed pET/sAsAP, pET/ tAsAP, and pET/cAsAP were grown in 50 ml of LB medium supplemented with ampicillin at 37°C overnight. The cultures were then inoculated with the final 1.0 liter LB culture. When the culture reached an absorbance of 0.6 at 600 nm, 400 mM IPTG was added, and the bacteria were grown for a further 6 h at 37°C. Cells were harvested by centrifugation at 6000g for 10 min, and the pellets were kept frozen at 220°C for analysis of the accumulation by SDS–PAGE and for the purification of the enzymes. Enzyme assays. The AsAP activity was assayed spectrophotometrically according to Shigeoka et al. (20). The assay mixture (2 ml) contained 50 mM potassium phosphate buffer, pH 7.0, 0.4 mM AsA, 0.1 mM H2O2, and the enzyme at 36°C. The assay was initiated by the addition of the enzyme, and the oxidation of AsA was followed by a decrease in the absorbance at 290 nm (2.8 mM21 cm21). Oxidation of alternate electron donors was measured in the same assay mixture as that used for AsA, but AsA was replaced by 20 mM pyrogallol (430 nm, 2.47 mM21 cm21), 10 mM guaiacol (470 nm, 22.6 mM21 cm21), 0.4 mM D-iso-AsA (290 nm, 3.3 mM21 cm21), 0.15 mM NAD(P)H (340 nm, 6.22 mM21 cm21), and 40 mM reduced Cyt c (550 nm, 19 mM21 cm21). The activity of glutathione peroxidase was determined spectrophotometrically by following the oxidation of NADPH in the presence of glutathione reductase (20). The protein was determined with Coomassie brilliant blue G-250 using bovine serum albumin (BSA) as the standard according to Bradford (21). Purification of recombinant AsAP isozymes. All purification steps were carried out at 4°C. The recombinant E. coli cells (3.8 g wet wt) transformed with pET/sAsAP were resuspended in 20 ml of 10 mM potassium phosphate buffer, pH 7.0, containing 20% (w/v) sorbitol, 1 mM AsA, 0.1% PMSF, and 1 mM EDTA (buffer A) and disintegrated by passing them through a cooled French pressure cell at 15,000 psi (SLM Aminco, Inc.). This lysate was centrifuged at 15,000g for 20 min. The supernatant, designated the crude extract, was subjected to ultracentrifugation at 100,000g for 30 min. The obtained supernatant was then loaded onto a DEAE-Sephacel column (1.6 3 12 cm) equilibrated with buffer A. The column was washed with 50 ml of buffer A and eluted with a 400-ml linear gradient of NaCl (0 – 0.3 M) at a flow rate of 0.7 ml min21. The activities of sAsAP were eluted as a single peak at 0.08 M NaCl. The active fractions were subjected to (NH4)2SO4 precipitation at 30% saturation. The precipitate was removed by centrifugation, and the supernatant was loaded onto a HiLoad 16/10 Phenyl-Sepharose column equilibrated with 30% saturated (NH4)2SO4 in buffer A. The enzyme activity was eluted with a descending gradient of (NH4)2SO4 from 30 to 0% saturation in 144 ml of buffer A at an elution rate of 0.8 ml min21. Active fractions were combined and fractionated with (NH4)2SO4, and the pellet precipitating at 70% saturation was dissolved in 2 ml of buffer A. The enzyme solution was loaded onto a HiLoad 16/60 Superdex 200 column equilibrated with buffer A containing 0.15 M NaCl. The active fractions were combined and stored at 220°C. The recombinant E. coli cells (5.2g wet wt) transformed with pET/cAsAP were resuspended in 20 ml of buffer A and disintegrated by passing them through a cooled French pressure cell. This lysate was centrifuged at 15,000g for 20 min. The crude extract was subjected to ultracentrifugation at 100,000g for 30 min. The obtained supernatant was then loaded onto a DEAE-Sephacel column (1.6 3 12 cm) equilibrated with buffer A. The column was washed with 50 ml of buffer A and eluted with a 400-ml linear gradient of NaCl (0 – 0.3 M) at a flow rate of 0.7 ml min21. The activities of cAsAP were eluted as a single peak at 0.12 M NaCl. Active fractions were combined and fractionated with (NH4)2SO4, and the pellet precipitating at 70% saturation was dissolved in 2 ml of buffer A. The enzyme

RECOMBINANT ASCORBATE PEROXIDASE ISOZYMES OF SPINACH solution was loaded onto a HiLoad 16/60 Superdex 200 column equilibrated with buffer A containing 0.15 M NaCl. The active fractions were combined and adjusted to 30% saturation with (NH4)2SO4. The precipitate was removed by centrifugation, and the supernatant was chromatographed on a 5/5 Phenyl-Superose column equilibrated with 30% saturated (NH4)2SO4 in buffer A. The column was eluted with an 11-ml linear gradient of 30 – 0% (NH4)2SO4 at a flow rate of 0.15 ml min21. The active fractions were combined and stored. Partial purification of native AsAP isozymes from spinach leaves. Intact chloroplasts were isolated from spinach leaves (Spinacia oleracea L.) by the method of Takeda et al. (22). Isolated chloroplasts were suspended in buffer A and disrupted by ultrasonic treatment (10 kHz) for 20 s. The lysate was centrifuged by at 100,000g for 30 min. The precipitate was resuspended in buffer A to determine the tAsAP activity. The supernatant was loaded onto a DEAE-Sephacel column (1.6 3 12 cm) equilibrated with buffer A. The column was washed with 50 ml of buffer A and eluted with a 400-ml linear gradient of NaCl (0 – 0.3 M) at a flow rate of 0.7 ml min21. The activities of sAsAP were eluted as a single peak at 0.08 M NaCl. The supernatant to remove intact chloroplasts from crude homogenate was used for the partial purification of cAsAP. It was then centrifuged at 100,000g for 30 min and the obtained supernatant was subjected to the same purification procedure by a DEAE-Sephacel column as that for the recombinant cAsAP. Antibody production. Polyclonal antibodies raised against the purified recombinant spinach AsAP isozymes were prepared as described (23). Mice were injected with 50 mg purified recombinant cAsAP and sAsAP emulsified with Freund’s complete adjuvant followed by three subcutaneous injections. After bleedings, the antisera were separated from the blood and cleaned by saturation with a crude extract from the E. coli strain BL21(DE3)pLysS. SDS–PAGE and immunoblotting. Electrophoresis was performed as previously described (24), using a 12.5% resolving gel. Following SDS–PAGE, the enzyme was visualized using either Coomassie blue staining or immunoblotting. Coomassie staining of proteins was accomplished by an incubation gel fixative (20% methanol, 7% acetic acid) containing 0.2% Coomassie brilliant blue R-250 for 30 min and proteins were destained overnight in fixative. For immunoblotting, proteins separated by SDS–PAGE were electrotransferred from the gel to an Immobilon-P transfer membrane (Millipore, Bedford, MA) using a semidry electroblotting system according to the manufacturer’s procedures (Bio-Rad). After transfer, the membrane was blocked for 1 h in 3% dry milk in phosphate-buffered saline (PBS). The monoclonal antibody to Euglena AsAP (EAP1) (25) and the polyclonal antibodies to the recombinant sAsAP and cAsAP were diluted 1:2000 in TBS (10 mM Tris, pH 8.0, and 150 mM NaCl) containing 0.1% BSA, and the membrane was incubated at room temperature for 1 h. The membrane was washed in TPBS (PBS containing 0.05% Tween 20), followed by incubation for 30 min in anti-mouse-Igs peroxidase-conjugated secondary antibody (Boehringer, Germany) diluted 1:5000 in TBS containing 0.1% BSA. The membrane again was washed in TPBS, and the immunoreactive proteins were visualized using 4-methoxy-1-naphthol (Aldrich, WI). N-terminal sequence analysis. The purified recombinant proteins were separated using SDS–PAGE (12.5% gels) with 2-mercaptoethanol according to Laemmli (26) and transferred to an Immobilon P transfer membrane (Millipore). The membrane was extensively washed with water, stained with 0.06% Coomassie R-250 in 50% methanol for 5 min, and destained with 30% aqueous methanol, 7% acetic acid for 10 min. The portion of the membrane containing the desired protein band was cut out, and the N-terminal sequence was determined using an ABI Model 492 gas-phase protein sequencer.

57

RESULTS AND DISCUSSION

Construction of the expression plasmid for spinach AsAP isozymes. We previously isolated two cDNAs encoding chloroplastic AsAP isozymes from spinach greening cotyledons (4). The tAsAP isozyme was identical to sAsAP through the coding region, except for the C-terminal amino acid, position 365. The remainder of the C-terminal coding region of tAsAP was substituted by a different sequence that encoded 50 amino acids, which was the putative transmembrane segment, resulting in an open reading frame of 415 amino acids (Fig. 1). Thus, the isolation and organization of the gene (ApxII) indicated that mRNAs for both tAsAP and sAsAP are produced from only one gene by alternative splicing of the 39-terminal exons (19). The availability of two full-length cDNAs encoding sAsAP and tAsAP isozymes enabled us to amplify, by PCR, fragments of these clones for the production of the mature form of proteins in E. coli. Because it was certain from both native mature sAsAP and tAsAP where the spinach chloroplastic AsAPs begin, Tyr-71 was chosen as the first amino acid for the recombinant sAsAP and tAsAP (Fig. 1). A Met was added to the immediate upstream region of the chosen Tyr by the ATG initiation codon, present in the NdeI restriction site. The NdeI and BamHI sites included in the PCR products were used to clone the fragment into the expression vector pET-3a (Fig. 2). The resulting constructions designated pET/sAsAP and pET/tAsAP were used to transform E. coli strain BL21(DE3)pLysS. Concurrently, we constructed the expression system using pET-3a for the cDNA encoding the spinach cAsAP isozyme that was designated pET/cAsAP. In this case, the initiation codon was the authentic ATG codon. Expression of the recombinant AsAP isozymes in E. coli. The pET expression system turned out to be a good system for the high-level expression of active chloroplastic AsAP and cAsAP isozymes. Overexpression of the recombinant sAsAP and cAsAP was confirmed by assay of the enzyme activity and by SDS–PAGE analysis of the enzyme protein in soluble fractions obtained from crude extracts of recombinant E. coli cells. As shown in Fig. 3A, the protein bands corresponding to the recombinant sAsAP and cAsAP, which correlated with the molecular mass (33 and 28 kDa, respectively) calculated from the deduced amino acid sequence of each cloned AsAP, were observed. Both recombinant sAsAP and cAsAP accounted for nearly 30% of the total protein in the E. coli cells. The soluble fractions of recombinant sAsAP and cAsAP showed enzyme activities of 21.1 and 13.7 mmol min21 mg protein21: enzyme activities were not found in the insoluble fractions. Dalton et al. (27) have recently reported the heterologous expression of cAsAP from soybean in E. coli using the expression vector of pQE-30 with the additional six His residues to the authentic N-termi-

58

YOSHIMURA ET AL.

FIG. 1. Comparison of amino acid sequences between AsAP isozymes from spinach. The sequence alignment is shown for the stromal AsAP (sAsAP), the thylakoid-bound AsAP (tAsAP), and the cytosolic AsAP (cAsAP) from spinach. The gaps are introduced to optimize the alignment. Residues found at the same position as sAsAP are shown as white letters on black. The arrowhead indicates the processing site of chloroplastic AsAPs (sAsAP and tAsAP).

nus. Most of the recombinant AsAP isozyme was present in the apo-form, and the addition of hemin to the AsAP solution resulted in a significant increase in the specific activity. In our case, the addition of hemin to the cultures of E. coli transformed with pET/sAsAP and pET/cAsAP did not cause the respective AsAP activity to increase. In addition, in vitro reconstitution with hemin of the recombinant AsAP according to the method described previously (27) had no effect on the increase in AsAP activity. The theoretical Rz values (A403/A280 or A401/A280) of recombinant sAsAP and cAsAP were 1.20 and 1.38, respectively, which were determined with the absorption spectra of the purified recombinant sAsAP and cAsAP described afterward. These results and the detection of a distinct red color in homogeneous solutions of each enzyme suggest that most of the recombinant AsAP isozymes exist as the holo-form and the heme production can keep up with the protein synthesis in E. coli. Immu-

noblot analysis using Euglena EAP1, which reacted with chloroplastic AsAPs and cAsAP from higher plants (25), was performed. Immunoreactive bands of both recombinant sAsAP and cAsAP were observed at the position corresponding to the major IPTG-induced products in gels stained with Coomassie blue (Fig. 3B). These crossreactivities correlated well with the previous report (25). The EAP1 reacted with a 38-kDa band corresponding to the predicted size of the recombinant tAsAP. The total content of the recombinant tAsAP expressed in the crude extract of E. coli was almost equal to that of the recombinant sAsAP according to SDS–PAGE and immunoblot analysis (Figs. 3A and 3B). Recombinant tAsAP accounted for nearly 30% of the total extracts from the E. coli cells. The total activity of recombinant tAsAP in the insoluble fraction was almost equal to that in the soluble fraction (data not shown). The insoluble recombinant tAsAP activity was 32.6 mmol

RECOMBINANT ASCORBATE PEROXIDASE ISOZYMES OF SPINACH

59

FIG. 2. Construction of plasmid pET/AsAP for the expression of spinach AsAP isozymes. Shown are the steps performed to ligate the mature sAsAP-encoding PCR-amplified fragment into the expression vector pET-3a, yielding the pET/sAsAP plasmid. The details are described under Materials and Methods. Ampr, ampicillin-resistance gene; ori, E. coli replication origin; f10s10, T7 promoter and transcription initiation site; tf, T7 transcription terminator.

min21 mg protein21. It seems likely that the recovery of the recombinant tAsAP as a membrane-bound form in the insoluble fraction of E. coli is due to the Cterminal region which is rich in hydrophobic amino acid residues and constructs a hydrophobic domain. Similarly, more than 95% of the recombinant gAsAP, whose C-terminus had a hydrophobic microbody-membrane binding domain, was recovered in the insoluble membrane fraction as an active form in E. coli (11). Purification of recombinant sAsAP and cAsAP. The recombinant spinach sAsAP was purified from the E. coli

by conventional chromatographic steps, giving a large quantity of highly pure protein, approximately 0.73 mg liter21 of bacterial culture (Table I). The purified recombinant sAsAP was homogeneous according to SDS– PAGE and showed the same subunit molecular mass as the native sAsAP (Fig. 4). The enzyme was found to be a monomeric form by gel filtration on a HiLoad 16/60 Superdex 200 column, supporting the fact that chloroplastic AsAP isozymes exist as a monomeric form (3, 5, 12). The specific activity of the purified recombinant sAsAP was 800 mmol min21 mg protein21 (Table I), which was a

60

YOSHIMURA ET AL.

FIG. 3. Expression of spinach AsAP isozymes in E. coli cells. Protein samples (corresponding to 50 mg of total protein) prepared from cell lysates with 0.1 mM IPTG induction for 4 h were analyzed by 12.5% SDS–PAGE and immunoblots. (A) Coomassie brilliant blue staining; (B) immunoblots using monoclonal antibody raised against Euglena AsAP. Lane 1, molecular mass standards (Pharmacia); lanes 2 and 6, pET-3a-transformed E. coli; lanes 3 and 7, pET/cAsAPtransformed E. coli; lanes 4 and 8, pET/sAsAP-transformed E. coli; lanes 5 and 9, pET/tAsAP-transformed E. coli. Experimental conditions are described under Materials and Methods.

value almost similar to those of the native spinach and tea sAsAPs (5, 12). The N-terminal amino acid sequence of the recombinant sAsAP was MYASDPAQLKNARED, which was in agreement with that predicted from the cDNA sequence. The purification of recombinant cAsAP was also carried out (Table I). The specific activity of the purified recombinant cAsAP was 486 mmol min21 mg protein21, which was similar to that of the native cAsAP and was approximately 15-fold greater than the purified recombinant cAsAP of soybean (27). In the case of soybean recombinant cAsAP, it expressed as the heterologous products of apo- and holoenzyme in E. coli (27). SDS– PAGE analysis of the purified recombinant cAsAP

FIG. 4. SDS–PAGE of the purified recombinant sAsAP and cAsAP. Proteins were visualized by Coomassie brilliant blue staining. Lane 1, molecular mass standards (Pharmacia); lane 2, purified recombinant sAsAP (1.0 mg of protein); lane 3, purified recombinant cAsAP (1.0 mg of protein).

showed the homologous 28-kDa protein (Fig. 4). The enzyme was found to be a monomeric form by gelfiltration analysis. cAsAP isozymes may be grouped into two classes in terms of the structure of the subunit. One class exists as a monomeric form in the cAsAPs of spinach (7), potato tuber (13), Brassica rapa (8), and Euglena (25). The other class is a homodimer in the cAsAPs of pea (6) and legume root nodules (28), in which each subunit fails to associate via disulfide bonds. The recombinant pea and soybean cAsAP isozymes also exist as homodimers (17, 27). Characterization of recombinant sAsAP, tAsAP, and cAsAP. Table II shows a comparison of the enzymatic properties of both recombinant and native AsAP isozymes. The properties of recombinant sAsAP, tAsAP, and cAsAP produced in E. coli were indistinguishable from those of the respective native isozymes. The recombinant chloroplastic AsAP isozymes utilized

TABLE I

Purification Scheme for Recombinant Spinach sAsAP and cAsAP from E. coli

Recombinant sAsAP Crude extract Ultracentrifugation DEAE-Sephacel 30% (NH4)2SO4 Phenyl-Sepharose 70% (NH4)2SO4 Superdex 200 Recombinant cAsAP Crude extract Ultracentrifugation DEAE-Sephacel 70% (NH4)2SO4 Superdex 200 30% (NH4)2SO4 Phenyl-Superose

Total protein (mg)

Total activity (mmol min21)

Specific activity (mmol min21 mg protein21)

Yield (%)

176.9 164.9 13.5 9.0 3.9 3.8 0.73

3732 3632 2997 2882 1476 1454 584.0

21.1 22.0 222 320 378 383 800

100 97.3 80.3 77.2 39.5 39.0 15.6

280.8 255.6 75.6 69.3 6.0 5.7 2.2

3844 3664 3559 3541 2098 1979 1070

13.7 14.3 47.1 51.1 350 347 486

100 95.3 92.6 92.1 54.6 51.5 27.8

61

RECOMBINANT ASCORBATE PEROXIDASE ISOZYMES OF SPINACH TABLE II

Comparison of Some Enzymatic Properties of Recombinant sAsAP, tAsAP, and cAsAP Isozymes with Those of Spinach Native Isozymes Recombinant

Molecular mass (kDa) Gel filtration SDS–PAGE Donor specificity (%)a AsA Iso-AsA GSH Cyt c NAD(P)H Pyrogaroll Guaiacol Km (mM) AsA H2O2 Optimum pH

Native

sAsAP

tAsAP

cAsAP

sAsAP

tAsAP

cAsAP

33 33 (33,239 Da)b

n.d. 38 (37,710 Da)

28 28 (27,560 Da)

30 6 1c 30c

40 6 2d 40d

31e 31e

100 58 0 0 0 19 0.6

100 65 0 0 0 15 0.3

100 32 0 0 0 350 3.9

100 59 0 0 0 25 1.4

100 60 0 0 0 13 0.1

0.33 0.04 7.0

0.35 0.04 7.0

0.52 0.01 7.0

0.30c 0.03c 7.0

0.57d 0.08d 7.0

100 38 0 0 0 368 4.3 0.60 0.01 7.0

Note. nd, not determined. a The peroxidase activity for AsA was shown as 100%. b Data in parentheses are calculated from deduced amino acid sequences (4, 16). c Reference 12. d Reference 3. e Reference 7.

AsA at a higher rate as an electron donor and showed no or low activity for electron donors such as GSH, NAD(P)H, Cyt c, guaiacol, and pyrogallol. The relative activities among the electron donors of recombinant chloroplastic AsAP isozymes were similar to those of the native chloroplastic enzymes. The recombinant cAsAP showed a distinct preference for pyrogallol as electron donor, and the value was similar to that of the native cAsAP. The Km values of both recombinant sAsAP and cAsAP for AsA (0.33 and 0.52 mM) and H2O2 (0.04 and 0.01 mM) were similar to those of each native isozyme (3, 12). The optimum pH of both the recombinant sAsAP and cAsAP was 7.0, which corresponded well to each value of the native enzymes. Miyake et al. (3) have reported that the properties of the native tAsAP are very similar to those of sAsAP with respect to high specificity for AsA and Km values for H2O2 and AsA. The sole difference in terms of properties between tAsAP and sAsAP is the higher molecular weight of the membrane-bound enzyme compared to the sAsAP; the tAsAP is bound to thylakoid membranes in such a form that the active site of the enzyme is exposed to the stroma for access of the substrate. We could demonstrate that not only the native sAsAP and tAsAP isozymes but also the recombinant enzymes possess the similar characterizations. AsAP has been known as a labile enzyme in an AsA-depleted medium. Compared with the cAsAP

isozyme, the chloroplastic AsAP isozymes have a halftime of within a minute in that medium (1). When the purified recombinant sAsAP was diluted with the AsAdepleted medium, the enzyme showed rapid inactivation, whose half-time was approximately 15 s. The half-inactivation time of the purified recombinant cAsAP was approximately 60 min, a value very similar to that reported for the native cAsAP (1). We have found that gAsAP, which was highly homologous to cAsAP, was also relatively stable in the AsA-depleted medium (11). It is an interesting problem that in spite of no detectable amount of AsA in E. coli cells, the recombinant AsAP isozymes are overexpressed as an active form. Miyake and Asada (29) have reported that the inactivation of tAsAP isozyme in an AsA-depleted medium is caused by the instability of Compound I to H2O2 when AsA is not available for Compound I and that no inactivation of sAsAP occurs under anaerobic conditions. One likely explanation for this expression, therefore, is that E. coli cells maintain highly reduced and anaerobic conditions in vivo. As shown in Figs. 5A and 5B, the absorption spectra of the purified recombinant sAsAP and cAsAP closely resembled those of native enzymes from several sources (3, 5, 11, 28). Characteristic absorption maxima of a Soret band at 403 and 401 nm were found for the recombinant sAsAP and cAsAP, respectively. The Soret peaks of the recombinant sAsAP and cAsAP were

62

YOSHIMURA ET AL.

FIG. 5. Absorption spectrum of purified recombinant sAsAP (A) and cAsAP (B). The sample cuvette contained 115 mg of the purified sAsAP or 130 mg of the purified cAsAP in 1 ml of 50 mM potassium phosphate, pH 7.0, 20% (w/v) sorbitol, 0.1% PMSF, and 1 mM EDTA. —, native form; - - -, reduced form; – z –, CN form.

shifted to 418 and 403 nm, respectively, by reduction with dithionite, and respective additional peaks appeared at 537 and 532 nm. The absorbance coefficients of the oxidized forms of sAsAP (403 nm) and cAsAP (401 nm) were 7.2 3 104 M21 cm21 and 6.7 3 104 M21 cm21, respectively. The cyanide complex of the oxidized sAsAP and cAsAP gave peaks at 420 and 537 nm (Fig. 5A) and 419 and 534 nm (Fig. 5B), respectively. Immunological cross-reactivity of the recombinant AsAP with the native AsAP isozymes from spinach. On immunoblot analysis, polyclonal antibodies against the recombinant sAsAP and cAsAP isozymes reacted with the respective recombinant AsAP isozyme proteins (Figs. 6A and 6B). The polyclonal antibody against sAsAP from tea leaves cross-reacted with both sAsAP and cAsAP isozymes (3, 5). In contrast, the polyclonal antibody against the native pea cAsAP and the monoclonal antibody against the native spinach cAsAP failed to cross-react with the respective chloroplastic isozymes (6, 30). Polyclonal antibodies against the recombinant spinach sAsAP and cAsAP isozymes could efficiently cross-react with both chloroplastic AsAPs and cAsAP. This result is not surprising because of the existence of highly conserved domains between chloroplastic AsAPs and cAsAP (Fig. 1). Recently, we have defined that the Euglena monoclonal antibody (EAP1), which cross-reacts with both chloroplastic AsAPs and cAsAPs from higher plants, recognizes the common epitope at the site around the proximal His residue in AsAP isozymes (Yoshimura et al., unpublished data). CONCLUSIONS

These results presented here demonstrate the fact that chloroplastic AsAPs and cAsAPs can be distinguished by their specificity for the electron donor and

stability in AsA-depleted medium. It is an interesting problem how the chloroplastic AsAP isozymes provide a possible characteristic structure allowing the specific binding of AsA and their lability. These questions raise a structural problem, which can be solved only by obtaining the three-dimensional structure of the enzymes. Jespersen et al. (31) have proposed that the Trp-175 residue near the proximal His residue is a strong candidate related to the specificity of chloroplastic AsAP isozymes toward AsA. The pET/sAsAP plasmid is an excellent expression system for the spinach chloroplastic AsAP isozyme. The yields of the recombinant enzyme are relatively high, and its enzymatic properties are identical to those of the native sAsAP isolated from spinach chloroplasts. In our laboratory we are progressing toward facilitating a crystallographic and site-directed mutagenesis study of spinach sAsAP.

FIG. 6. Production and cross-reactivity of anti-recombinant sAsAP (A) and cAsAP (B) polyclonal antisera. Purified recombinant cAsAP (lanes 1 and 4), purified recombinant sAsAP (lanes 2 and 5), and solubilized recombinant tAsAP (lanes 3 and 6) were separated by SDS–PAGE and then subjected to immunoblot analysis using antirecombinant cAsAP and sAsAP serum, respectively, as described under Materials and Methods. Relative molecular mass standards are indicated.

RECOMBINANT ASCORBATE PEROXIDASE ISOZYMES OF SPINACH

ACKNOWLEDGMENTS This work was supported by Kansai Research Foundation for Technology Promotion and by a grant from NEDO/RITE’s International Joint Research Program.

REFERENCES 1. Asada, K. (1992) Physiol. Plant. 85, 235–241. 2. Asada, K. (1997) in Oxidative Stress and the Molecular Biology of Antioxidant Defenses (Scandalios, J. G., Ed.), pp. 715–735, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY. 3. Miyake, C., Cao, W.-H., and Asada, K. (1993) Plant Cell Physiol. 34, 881– 889. 4. Ishikawa, T., Sakai, K., Yoshimura, K., Takeda, T., and Shigeoka, S. (1996) FEBS Lett. 384, 289 –293. 5. Chen, G.-X., and Asada, K. (1989) Plant Cell Physiol. 30, 987– 998. 6. Mittler, R., and Zilinskas, B. A. (1991) Plant Physiol. 97, 962– 968. 7. Tanaka, K., Takeuchi, E., Kubo, A., Sakaki, T., Haraguchi, K., and Kawamura, Y. (1991) Arch. Biochem. Biophys. 286, 371–375. 8. Ishikawa, T., Takeda, T., and Shigeoka, S. (1996) Plant Sci. 120, 11–18. 9. Yamaguchi, K., Mori, H., and Nishimura, M. (1995) Plant Cell Physiol. 36, 1157–1162. 10. Bunkelmann, J. R., and Trelease, R. N. (1996) Plant Physiol. 110, 589 –598. 11. Ishikawa, T., Yoshimura, K., Sakai, K., Tamoi, M., Takeda, T., and Shigeoka, S. (1998) Plant Cell Physiol. 39, 23–34. 12. Nakano, Y., and Asada, K. (1987) Plant Cell Physiol. 28, 131–140. 13. Elia, M. R., Borraccino, G., and Dipierro, S. (1992) Plant Sci. 85, 17–21.

63

14. Mittler, R., and Zilinskas, B. A. (1991) FEBS Lett. 289, 257–259. 15. Kubo, A., Saji, H., Tanaka, K., Tanaka, K., and Kondo, N. (1992) Plant Mol. Biol. 18, 691–701. 16. Ishikawa, T., Sakai, K., Takeda, T., and Shigeoka, S. (1995) FEBS Lett. 367, 28 –32. 17. Patterson, W. R., and Poulos, T. L. (1994) J. Biol. Chem. 269, 17020 –17024. 18. Patterson, W. R., and Poulos, T. L. (1995) Biochemistry 34, 4331– 4341. 19. Ishikawa, T., Yoshimura, K., Tamoi, M., Takeda, T., and Shigeoka, S. (1997) Biochem. J. 328, 895–900. 20. Shigeoka, S., Nakano, Y., and Kitaoka, S. (1980) Arch. Biochem. Biophys. 201, 121–127. 21. Bradford, M. (1976) Anal. Biochem. 72, 248 –254. 22. Takeda, T., Yokota, A., and Shigeoka, S. (1995) Plant Cell Physiol. 36, 1089 –1095. 23. Tamoi, M., Ishikawa, T., Takeda, T., and Shigeoka, S. (1996) Arch. Biochem. Biophys. 334, 27–36. 24. Shigeoka, S., and Nakano, Y. (1993) Biochem. J. 292, 463– 467. 25. Ishikawa, T., Takeda, T., Kohno, H., and Shigeoka, S. (1996) Biochim. Biophys. Acta 1290, 69 –75. 26. Laemmli, U. K. (1970) Nature 227, 680 – 685. 27. Dalton, D. A., Diaz del Castillo, L., Kahn, M. L., Joyner, S. L., and Chatfield, J. M. (1996) Arch. Biochem. Biophys. 328, 1– 8. 28. Dalton, D. A., Hanus, F. J., Russell, S. A., and Evans, H. J. (1987) Plant Physiol. 83, 789 –794. 29. Miyake, C., and Asada, K. (1996) Plant Cell Physiol. 37, 423– 430. 30. Saji, H., Tanaka, K., and Kondo, N. (1990) Plant Sci. 69, 1–9. 31. Jespersen, H. M., Kjaersgård, I. V. H., Østergaard, L., and Welinder, K. G. (1997) Biochem. J. 326, 305–310.