A Dithiothreitol-Sensitive Tetrameric Protease from Spinach Thylakoids Has Polyphenol Oxidase Activity

Plant Cell Physiol. 38(2): 179-187 (1997) JSPP © 1997

A Dithiothreitol-Sensitive Tetrameric Protease from Spinach Thylakoids Has Polyphenol Oxidase Activity Tomohiko Kuwabara1'2, Tetsuya Masuda2 and Sachiko Aizawa 2 1 2

Institute of Biological Science, University of Tsukuba, Tsukuba, Ibaraki, 305 Japan Master's Program in Biosystem Studies, University of Tsukuba, Tsukuba, Ibaraki, 305 Japan

Key words: Active oxygen — Copper — Dithiothreitol-sensitive tetrameric protease — Photosystem II — Polyphenol oxidase (EC 1.10.3.1) — Proteolysis.

Plant polyphenol oxidase (EC 1.10.3.1) is an enzyme whose true physiological function remains to be determined (Mayer 1987, Vaughn et al. 1988). Cytochemical and immunocytochemical studies have shown that the enzyme is located in various types of plastid and is associated with thylakoid membranes in chloroplasts (Vaughn and Duke 1981, 1984). Some researchers have expressed doubt about the localization of the enzyme, because there are no phenolic substrates in chloroplasts. In the early 1990s, cDNAs for PPO were cloned from various plants (Shahar

et al. 1992, Cary et al. 1992, Hunt et al. 1993, Newman et al. 1993); they revealed that the protein was translated as a precursor with a transit peptide that could direct the transport of the precursor into the thylakoid lumen. Recently, Sommer et al. (1994) and Sokolenko et al. (1995) showed that the precursor to PPO, synthesized in vitro, was imported into intact chloroplasts and then to the thylakoid lumen. These molecular biological studies, together with cytochemical and immunocytochemical studies, established the lumenal localization of PPO, which suggests that the physiological function of this protein is unlikely to be the oxidation of phenolic compounds. The difficulties associated with biochemical studies of PPO stem from the nature of the protein, which readily degrades during purification. Spinach PPO was first characterized as a 42.5-kDa protein that was reversibly converted into a 158-kDa tetramer (Golbeck and Cammarata 1981). Until the middle 1980s, the molecular mass of PPO was believed to be around 40 kDa. However, studies with antibody against PPO revealed the presence of cross-reactive 60- to 63-kDa polypeptides (Lanker et al. 1988). Robinson and Dry (1992) purified the 60-kDa PPO from Viciafaba and suggested that the 41- to 43-kDa species might be a product of degradation generated during purification. It seems probable that the previously purified PPO from spinach (Golbeck and Cammarata 1981) was a similar degradation product. Amino acid sequences deduced from cDNAs for PPO also indicate that the molecular masses of mature PPOs should be around 60 kDa (Shahar et al. 1992, Cary et al. 1992, Hunt et al. 1993, Newman et al. 1993, Hind et al. 1995, Sokolenko et al. 1995). We have been interested in chloroplast proteases that might be involved in turnover of proteins for photosynthesis. We purified a DTT-sensitive protease of 39 kDa from spinach PSII membranes which was reversibly convertible to a 156-kDa tetramer (Kuwabara and Hashimoto 1990). Recently, this 39-kDa protease was found to derive from a 60-kDa protein (Kuwabara 1995). The purified 60-kDa protein degraded the extrinsic 23-kDa protein of PSII to yield a fragment of 20 kDa. As indicated above, the structural characteristics of PPO and the DTT-sensitive tetrameric protease (DSTP) are very similar. Furthermore, the amino-terminal amino acid sequences of the two proteins are identical up to at least the tenth residue (Kuwabara 1995). The circumstan-

Abbreviations: DOPA, /?-(3,4-dihydroxyphenyl) alanine; DSTP, dithiothreitol-sensitive tetrameric protease; PPO, polyphenol oxidase. 1 To whom correspondence should be addressed. 179

Downloaded from pcp.oxfordjournals.org by guest on October 17, 2011

Polyclonal antibody raised against a dithiothreitol-sensitive tetrameric protease (DSTP) from PSII membranes specifically inhibited the polyphenol oxidase (PPO) activity of spinach thylakoids. DSTP was copurified with PPO activity on an affinity column prepared with antibody against DSTP. These results suggest that DSTP and PPO are the same protein. During purification of DSTP, Tween 20 was essential for stabilization of the protein, which was degraded in the absence of the detergent. Gel-filtration chromatography of the purified DSTP revealed the presence of 230kDa (tetramer) and 60-kDa (monomer) species. The copper content of monomer species was determined to be 0.4 Cu atom per protein molecule, when the molecular weight of the protein was calculated to be 62,243, which is the value reported for spinach PPO [Hind et al. (1995) Biochemistry 34: 8157]. Purified DSTP caused the degradation as well as the dimerization of the extrinsic 23-kDa protein of PSII. The degradation of the protein was suppressed under anaerobic conditions induced by the presence of glucose oxidase and glucose together. This fact suggests that oxygen molecules are involved in the proteolytic reaction and that the proteolytic activity and PPO activity may be correlated with each other.

180

Protease with polyphenol oxidase activity

tial evidence indicates that these proteins are identical. With respect to enzymatic activity, by contrast, there is no experimental evidence to validate this hypothesis, because studies of PPO and DSTP have proceeded independently. In this study, we purified DSTP from spinach thylakoids using an affinity column with conjugated antibodies against DSTP. The purified protein had the activities of both a protease and PPO.

Materials and Methods

Downloaded from pcp.oxfordjournals.org by guest on October 17, 2011

Materials—Spinach plants (Spinacia oleracea L. cv. Viroflay) were grown under natural light at 25 ±0.5°C during the day and at 20±0.5°C during the night, at a relative humidity of 70±5% in a greenhouse for about 8 weeks. IgG against DSTP was purified from antiserum (Kuwabara 1995) by precipitation with ammonium sulfate (50% saturation) and anion-exchange column chromatography on DE-52 (Whatman). An affinity column was prepared from 26 mg of IgG against DSTP and 10 ml of a wet gel of cyanogen bromide-activated Sepharose 4B (Pharmacia), as described by the manufacturer. Glucose oxidase was obtained from Sigma. Purification of DSTP—Preparation of spinach chloroplasts and extraction of DSTP (PPO) from thylakoid membranes were performed essentially by the methods of Golbeck and Cammarata (1981) with minor modifications. In brief, chloroplasts prepared with 50 mM Tris-HCl (pH 7.5)/0.4 M sucrose/1 mM MgCl2/5 mM sodium ascorbate were suspended in 25 mM MES-NaOH (pH 6.5)/0.3 M sucrose/10 mM NaCl/30% (w/v) ethylene glycol and stored in liquid nitrogen until use. The suspension was thawed, diluted with 50 mM Tris-HCl (pH7.5)/l mM MgCl2/5 mM sodium ascorbate, and centrifuged at 14,500 xg for lOmin to remove ethylene glycol. The thylakoid membranes were washed twice with the same buffer by resuspension and centrifugation. The membranes were suspended in the same buffer at a Chi concentration of 2 mg ml"' and sonicated for 10 x 10 s with 10-s intervals (UD-201; TOMY, Tokyo). The suspension was centrifuged at 360,000 x g for 60min. The supernatant was passed through a cellulose acetate filter with a pore size of 0.45 nm (DISMIC-25cs; ADVANTEC, Tokyo). The filtrate was supplemented with 0.1% (w/v) Tween 20 for stabilization of DSTP. Forty ml of extract were loaded on HiTrap Q-Sepharose FF (two 5-ml pre-packed columns connected in tandem; Pharmacia) that had been equilibrated with 20 mM sodium phosphate (pH 6.5)/0.1% Tween 20. The column was washed with 100 ml of the equilibration buffer and then with 30 ml of the same buffer that contained 50 mM NaCl. DSTP was eluted with 30 ml of 20 mM sodium phosphate (pH 6.5)/300 mM NaCl/0.1% Tween 20 (buffer A) at a flow rate of 2 ml min" 1 . Elution was monitored by absorbance at 280 nm. The peak fraction, designated the HiTrap Q preparation (20 ml), was loaded on a column (1.5 cm i.d.x4.7 cm) of the antibody-conjugated Sepharose 4B equilibrated with buffer A at 15°C at a flow rate of 1 ml min~', with 5-min pauses between loading of each 6-ml aliquot to facilitate antigen-antibody interactions. The column was washed with 40 ml of buffer A at 15°C. DSTP was eluted at room temperature (25°C) with 48 ml of buffer A that contained 50% (w/v) ethylene glycol. When substantial PPO activity was found in the washings from the column, the affinity chromatography was repeated with this fraction. The resultant preparation was designated affinity-purified DSTP. Concentration and gel-filtration chromatography of affinity-

purified DSTP—Affinity-purified DSTP was loaded on a column of hydroxyapatite (1.6 cm i.d. x 5 cm; Seikagaku Kogyo, Tokyo) that had been equilibrated with 10 mM sodium phosphate (pH 6.5)/0A% Tween 20 at a flow rate of 1 ml min" 1 . After washing of the column with 20 ml of the equilibration buffer, DSTP was eluted with 40 ml of 100 mM sodium phosphate (pH 6.5). The first 10 ml of the eluate were discarded, and the next 30 ml were collected. This chromatographic step removed the ethylene glycol and the concentration of Tween 20 was reduced (note that the resultant preparation contained a trace of Tween 20 that had been carried over from the washing step). The preparation was concentrated with Centriprep-10 (Amicon) to about 0.6 ml. The concentrated DSTP was loaded on a column of Superose 12 (1 cm i.d. x 30 cm; Pharmacia) that had been equilibrated with 20 mM sodium phosphate (pH 6.5)/0.1 M NaCI/0.1% Tween 20. The column was eluted with the same buffer at a flow rate of 0.5 ml min"'. Fractions of 0.5 ml were collected. Assay of PPO activity—PPO activity was assayed under oxygen-saturated conditions by measuring the uptake of O2 with 1 ml of reaction mixture that contained 50 mM HEPES-NaOH (pH 7.0)/12.5mM DL-DOPA/0.3 mM linolenic acid (Golbeck and Cammarata 1981) at 25°C with a Clark-type oxygen electrode (DW-1; HANSATECH). Usually, 10-20/il of sample were used for a measurement. One unit of activity was defined as the amount of the enzyme that consumed 1 /umo\ O2 min" 1 under the above conditions. The apparent Km was measured as described above with DL-DOPA as substrate, in the concentration range from 0.6 to 12.5 mM. Three measurements were made at each concentration. A double-reciprocal plot of the data gave a straight line with a correlation coefficient of 0.99. Reaction of DSTP with the 23-kDa protein—An aliquot of the preparation of DSTP was mixed with the purified 23-kDa protein in a total volume of 50^*1. The mixture was dialyzed against 10 mM HEPES-NaOH (pH 7.0) at 7°C for 4 h and then incubated at 30°C for 4h, unless otherwise stated. Polypeptides generated by the reaction were analyzed by SDS-PAGE in the presence of 4 M urea (Kuwabara and Hashimoto 1990) and by immunostaining with the antiserum against the 23-kDa protein and peroxidase-conjugated secondary antibody. Quantitation of copper and protein—Copper was quantitated by measuring atomic absorption at 326 nm with a flameless atomic absorption spectrometer (model 603 equipped with an HGA-2200 system; PERKIN-ELMER). Tips of micropipettes and sample tubes were washed with 0.1 M HC1, rinsed with pure water, and dried in a clean oven before use. In our hands, accurate injection of a 20-//1 sample into the instrument was difficult when the sample contained 0.1% Tween 20. Therefore, DSTP-T and DSTP-M were prepared by chromatography on Superose 12 as described above, but without use of Tween 20 in the elution buffer. The samples of DSTP were not desalted to avoid degradation of DSTP under low-salt conditions (Kuwabara and Hashimoto 1990). The standard addition method was adopted for the determination. Thirty-five nl of the sample of DSTP were mixed with an equal volume of a series of dilutions of a standard solution that had been prepared by diluting a standard solution of Cu(NO3)2 (Wako, Osaka) with 0.1 M HNO3 (analytical grade; Wako). Twenty/il of the mixture were injected into a vessel of carbon graphite, dried at 100°C for 30 s, charred at 900°C for 30 s, and atomized at 2,700 c C for 5.5 s. Three measurements were performed for each experimental point. The elution buffer used for chromatography on Superose 12 was also analyzed in the same manner for correction of the results. The extent of atomic absorption and the amount of standard Cu added were linearly related with a correla-

Protease with polyphenol oxidase activity tion coefficient over 0.99 in each sample. Protein was quantitated by the method of Bradford (1976) with a protein assay kit (Bio-Rad) and bovine IgG as the standard. We noticed that 0.1% Tween 20 and 50% (w/v) ethylene glycol in a sample both reacted with the dye in the kit when the microassay procedure was used. Therefore, a standard curve was made for every preparation of DSTP by adding the buffer used for that preparation to a series of dilutions of the standard solution. Other procedures—-Chi was quantitated as described by Arnon (1949). Polypeptides in preparations of DSTP were analyzed by SDS-PAGE on 12% polyacrylamide gels without urea (Kuwabara and Hashimoto 1990). The extrinsic 23-kDa protein was purified from PSII membranes of spinach and its concentration was calculated from absorption coefficient of 22 mM" 1 cm"' at 277 nm (Kuwabara and Suzuki 1994). Polyclonal antibody against the 23-kDa protein was raised by the standard method in rabbits, with the purified 23-kDa protein as antigen, as described previously (Murata et al. 1984).

Results

inhibitory effect at the same concentration, suggesting that the antibodies against DSTP specifically recognized PPO. Copurification of DSTP protein and PPO activityTaking advantage of the cross-reactivity, we purified PPO activity using the affinity column prepared with antibody against DSTP. The PPO activity was eluted from the column with a solution at pH 6.5 that contained 50% (w/v) ethylene glycol (see Materials and Methods). The use of the near-neutral solution was essential for the stabilization of spinach PPO (Golbeck and Cammarata 1981); we first tried using acidic and basic solutions for the elution, but we found that the protein was eluted with greatly reduced PPO activity. Conditions that allowed us to elute DSTP from the affinity column are compared in Table 1. Using the affinity column, we completed the purification in only two days. Another essential feature of the purification procedure was the inclusion of Tween 20 in all the solutions used; this significantly inhibited degradation of the protein during purification (Kuwabara 1995). Figure 2 shows electrophoretic profiles of preparations during the course of the purification. The purified preparation contained a polypeptide of 62 kDa, the reduced form of the 60-kDa DSTP (Kuwabara 1995). Table 2 shows a summary of the purification in terms of PPO activity. Table 1 Effects on PPO activity of conditions for elution of DSTP from the affinity column Activity' %

Condition 50% (w/v) Ethylene glycol (pH 6.5)'

96

c

13

pH2.5

pH 11.0"

IgG/Chl (w/w) Fig. 1 Effects of antibodies against DSTP on the PPO activity of thylakoid membranes. Spinach thylakoids were sonicated and treated with 0.2% (w/v) Triton X-100 at 2 mg Chi ml" 1 for 1 min. Antibody (IgG) against DSTP or control antibody was added to the thylakoids at designated concentrations; the final concentrations of Chi and Triton X-100 were 1 mg ml"' and 0.1%, respectively. After incubation at room temperature for 10 min, the PPO activity was measured in 10//1 of the suspension. The points are means of two measurements. The control activity was 13 units (mgChl)" 1 .

54

The HiTrap Q preparation was brought to pH 6.5 in 50% ethylene glycol, to pH 2.5, or to pH 11.0. After the samples had been incubated at room temperature for 3 min, the PPO activity was measured. ° Values are expressed as percentages of respective controls. * The HiTrap Q preparation (pH 6.5) was mixed with an equal volume of 100% ethylene glycol. In the control, ethylene glycol was replaced by water. The activity of the control was 13.9 units (ml extract)" 1 . ' The HiTrap Q preparation (12.2 ml) was mixed with 0.1 M glycine-HCl (pH 2.0), 0.35 M NaCl, and 0.05% Tween 20 (10 ml), which yielded a pH of 2.5. After the 3-min incubation, the mixture was neutralized with 2.2 ml of 0.3 M Na3PO4. In the control, the above two solutions were pre-mixed and added to the HiTrap Q preparation. The activity of the control was 11.4 units (ml extract)" 1 . d The HiTrap Q preparation (10 ml) was mixed with 0.1 M 3cyclohexylaminopropanesulfonic acid (CAPS)-NaOH (pH 11.5), 0.35 M NaCl, and 0.05% Tween 20 (2.9 ml), which yielded a pH of 11.0. After the 3-min incubation, the mixture was neutralized with 3 ml of 0.1 M MES (unbuffered). In the control, the above two solutions were pre-mixed and added to the HiTrap Q preparation. The activity of the control was 8.9 units (ml extract)" 1 .

Downloaded from pcp.oxfordjournals.org by guest on October 17, 2011

Effects of antibodies against DSTP on PPO activity of thylakoid membranes—In a previous study (Kuwabara 1995), we showed that DSTP and PPO are structurally related with respect to their amino-terminal amino acid sequence and their ability to form tetramers. To examine the possible identity of the two proteins, we raised polyclonal antibodies against DSTP and examined their effects on the PPO activity of thylakoid membranes (Fig. 1). The antibody inhibited the PPO activity by 90% at a concentration of 2.0 mg IgG (mg Chi)" 1 ; control IgG had essentially no

181

182

Protease with polyphenol oxidase activity

kDa _ 94 - 67 "

43

- 30 - 20.1 - 14.4

From thylakoid membranes that contained 80 mg chlorophyll, we copurified about 1 mg of protein with PPO activity with a specific activity of 420 units (mg protein)" 1 . During chromatography on HiTrap Q, PPO was activated about 1.4-fold, probably as a result of aging (see below) and/or removal of inhibitory material(s) in the extract (Augustin et al. 1985). It is possible that the activation during purification contributed to the apparent high recovery of PPO activity. When the purified preparation was stored on ice for a week, PPO was further activated to a level 1.5fold higher than the value shown in Table 2. The apparent Km for D L - D O P A , measured under saturating O2 with the stored preparation, was 7.3 mM. This value is very close to 8.3 mM, the apparent Km obtained by Golbeck and Cammarata (1981) with spinach 42.5-kDa PPO. Our preparation had no monophenolase activity with tyrosine as the substrate, again resembling the preparation of Golbeck and Cammarata (1981). The copurification of DSTP pro-

Table 2 Purification of PPO activity during the purification of DSTP Total volume (ml)

Total activity (units)*

Total protein (mg)

Specific activity (units mg"1)

Recovery' %

Extract HiTrap Q preparation

40

520

20

710

17.8 8.55

29 83

100 137

Affinity-purified DSTP

98

410

0.97

420

79

Fraction

The preparations of DSTP were stored on ice until PPO activity was measured two days after the day of extraction. ° Total activity of the extract was taken as 100%. * One unit of PPO activity was defined as the amount that consumed 1 /imol O2 min"1 in 50 mM HEPES-NaOH (pH 7.5)/12.5 mM DLDOPA/0.3 mM linolenic acid at 25°C.

Downloaded from pcp.oxfordjournals.org by guest on October 17, 2011

Fig. 2 SDS-PAGE of preparations of DSTP. Lane 1, extract; lane 2, HiTrap Q preparation; lane 3, affinity-purified DSTP; and lane 4, molecular mass markers [from the top, phosphorylase b (94 kDa), bovine serum albumin (67 kDa), ovalbumin (43 kDa) carbonic anhydrase (30 kDa), trypsin inhibitor (20.1 kDa), and alactalbumin (14.4 kDa)]. Polypeptides were stained with Coomassie brilliant blue.

tein and PPO activity indicated that DSTP and PPO are the same protein. Tetrameric structure of DSTP—To study the oligomeric structure of DSTP, we subjected the purified DSTP to gel-filtration chromatography on Superose 12 after concentration. Two peaks of proteins were eluted, accompanied by PPO activity (Fig. 3A) as well as proteolytic activity (Fig. 3C: see Discussion for the apparent one broad peak of the proteolytic activity). The polypeptide composition of these peaks was similar (Fig. 3B); both peaks contained a 62-kDa polypeptide and faintly stained 41-kDa and 21-kDa products of degradation, which were generated during the course of the purification (Robinson and Dry 1992, Kuwabara 1995). The eluent for the gel-filtration chromatography contained 0.1% Tween 20 for stabilization of DSTP. Under these conditions, the elution of metal-containing proteins was retarded from that of some non-metal proteins; for example, ovalbumin (43 kDa) eluted before the second peak of DSTP (Fig. 4). For the estimation of the molecular mass of DSTP, we referred to the straight line that represented elution of metal-containing proteins, because copper was detected in our preparation (see below). The molecular masses estimated for the proteins in the two peaks were 230 and 60 kDa (Fig. 4). The value of 60 kDa for the latter peak was consistent with that of the 62-kDa polypeptide (Fig. 3B) and seemed to justify our approach. Based upon these molecular masses, the two peaks appeared to represent the tetrameric and monomeric forms of the protein; they were designated DSTP-T and DSTP-M, respectively. The copper content of DSTP—To determine the copper content of DSTP-T and DSTP-M, we performed chromatography on Superose 12 without Tween 20 in the elution buffer. The absence of the detergent might be expected to facilitate degradation of DSTP; thus, this manipulation was a compromise that we introduced for the accurate determination of the copper content. Under these conditions, we obtained essentially the same elution profile as we did when Tween 20 was included in the elution buffer, but the height of the peak of DSTP-T was about half of that of the peak of DSTP-M (data not shown). Table 3 summa-

Protease with polyphenol oxidase activity

B

183

Fraction number 21 22 23 24 25 26 27 28 29 30 31 32 M

kDa - 94 - 67 - 43

—

- 30

- 20.1

Fraction number

-

4 >

•a

I

-

x

25 Fraction number

2

I

23 kDa 20 kDa

30

Fig. 3 Gel-filtration chromatography of purified DSTP on Superose 12. Affinity-purified DSTP was concentrated and subjected to chromatography on Superose 12 as described in Materials and Methods. (A), Elution profiles in terms of /428o and PPO activity. The points are means of three measurements. Fractions shown by bars were collected as DSTP-T and DSTP-M. (B), SDS-PAGE of each fraction. Polypeptides were stained with silver. Lane M contains the same molecular mass markers as does lane 4 in Fig. 2. (C), Elution profile in terms of the proteolytic activity that degraded the 23-kDa protein. Immunoblot with antiserum against the 23-kDa protein is shown.

rizes the P P O activities and the concentrations of copper and protein in the preparations of DSTP-T and DSTP-M.

The copper content of DSTP-M was determined to be 0.4 atom Cu per enzyme molecule, calculated on the assump-

Table 3 PPO activity and levels of copper and protein in preparations of DSTP-T and DSTP-M Preparation

PPO activity (units ml"')

Cu concentration (nM)

DSTP-T

6.7

64

DSTP-M

11.4

191

Protein concentration

PPO activity Cu (units nmol"1)

PPO activity Protein (units mg~')

Cu/Protein" (mol/mol)

21'

100

320 c

0.2 c

340

0,4

Chromatography of affinity-purified DSTP on Superose 12 was performed as described in Materials and Methods except that the elution buffer contained no Tween 20. The preparations of DSTP-T and DSTP-M were analyzed for PPO activity and the concentrations of copper and protein. " The molecular weight of the protein was assumed to be 62,243 (Hind et al. 1995). * Probably an overestimate due to the overlapping elution of micelles of Tween 20 during chromatography on Superose 12 (see text). c Probably an underestimate due to the overestimation of the concentration of protein, as mentioned above.

Downloaded from pcp.oxfordjournals.org by guest on October 17, 2011

21 22 23 24 25 26 27 28 29 30 31 32

s

184

Protease with polyphenol oxidase activity 1,000

Additive DSTP-T (230 kDa)

DSTP-M (60 kDa)

Glucose oxidase Glucose

I

3 100

23 kDa 20 k D a 10 20

25

30

Fraction number

tion that the molecular weight of the protein was 62,243 (Hind et al. 1995). The copper content of DSTP-T was estimated to be 0.2 Cu atom per monomer unit, but this value could be an underestimate caused by overestimation of protein content introduced by the co-elution of micelles of Tween 20 with DSTP-T (by eluting Tween 20 only, we confirmed that the detergent eluted at the position of the shoulder that can be seen at fraction 25 in Fig. 3A). Involvement of oxygen in the proteolytic reaction of DSTP—To examine possible involvement of oxygen in the proteolytic reaction of DSTP, purified DSTP was treated with glucose oxidase plus glucose prior to the reaction with the 23-kDa protein in order to induce anaerobic conditions. DSTP thus treated did not degrade the 23-kDa protein, whereas DSTP treated with either glucose oxidase or glucose alone degraded the substrate as well as the untreated DSTP did (Fig. 5). These results suggest that oxygen molecules are essential for the proteolysis of the 23-kDa protein by DSTP. One may suspect instead that hydrogen peroxide generated by the reaction of glucose oxidase might have inhibited DSTP. However, when DSTP was treated with 1 mM hydrogen peroxide, the proteolytic reaction was not affected at all (data not shown). Thus, hydrogen peroxide does not seem to be involved in the inhibition of the proteolysis. Degradation and dimerization of the 23-kDa protein by free Cu2* ions—Copper has the potential to oxidize amino acid residues (Gutteridge and Wilkins 1983, Stadt-

Fig. 5 Effects of glucose oxidase plus glucose on the proteolytic activity of DSTP. Affinity-purified DSTP (0.6//g), which had been dialyzed against 10 mM HEPES-NaOH (pH 7.0) at 7°C for 4 h, was treated with 1 n% of glucose oxidase and 44 mM glucose in a total volume of 48 (il at 30°C for 10 min. After addition of 2 fig of the 23-kDa protein, the mixture was incubated at 30°C for 4 h for the proteolytic reaction. For controls, glucose and/or glucose oxidase were omitted from the reaction mixture. Immunoblot with antiserum against the 23-kDa protein is shown.

man 1993). We wondered if Cu 2+ ions liberated from DSTP during incubation might degrade the 23-kDa proCone, of Cu ions, M -8 .7 .6

.5 -4 .3

10 10 10 10 10 10 F D

23 kDa 20 kDa

Fig. 6 Effects of Cu 2+ ions on the electrophoretic profile of the 23-kDa protein. The protein (1.5 ^tg) was dialyzed against 10 mM HEPES-NaOH (pH 7.0) at 7°C for 4 h and then incubated with designated concentrations of Cu(NO3)2 (10~8 to 10~5 M; the standard solution for atomic absorption analysis was used) or CuCl2 (10~4 to 10"3 M) at 37°C for 4 h. As controls, the 23-kDa protein was mixed with free Cu 2+ ions [10//M Cu(NO3)2, lane F] or with DSTP-M that contained 6.7 pmol of endogenous copper (lane D) in a total volume of 50^1; then the solution was dialyzed and incubated as above. Immunoblot with antiserum against the 23-kDa protein is shown. The asterisk indicates polypeptides of 41-43 kDa that were probably dimers of the 23-kDa protein.

Downloaded from pcp.oxfordjournals.org by guest on October 17, 2011

Fig. 4 Estimation of molecular masses, of DSTP-T and DSTPM. Molecular masses of DSTP-T and DSTP-M were estimated from the results of Figure 3 with metal-containing proteins (shown by open circles: ferritin, 440 kDa; catalase, 232 kDa; carbonic anhydrase B, 29 kDa; and Cyt c, 12.4 kDa) as standards. Note that results for elution of non-metal proteins (shown by open triangles: aldolase, 158 kDa; bovine serum albumin, 67 kDa; ovalbumin, 43 kDa; trypsin inhibitor, 20.1 kDa; and RNase A, 13.7 kDa) did not fall on the same straight line as those for the metal-containing proteins under the present conditions.

Protease with polyphenol oxidase activity

tein. To examine this possibility, the 23-kDa protein was incubated with various concentrations of Cu 2+ ions (Fig. 6). Cu 2+ ions at concentrations up to 10~7 M caused no detectable changes in the electrophoretic profile of the 23-kDa protein under the conditions used. However, more concentrated Cu 2+ ions caused dimerization and further oligomerization of the protein, and 10~3 M Cu2+ ions degraded the protein to fragments of 19.5 kDa and smaller, concomitant with prominent oligomerization. Nevertheless, the 20-kDa fragment characteristic of the reaction with DSTP was not generated at any of the tested concentrations of Cu 2+ ions. By contrast, the reaction with DSTP, with endogenous copper present at 1.3 x 10~7 M, clearly generated the 20-kDa fragment as well as the dimer polypeptides that appeared in the reaction with free Cu 2+ ions (Fig. 6, lane D). These results suggest that DSTP and free Cu 2+ ions affected the 23-kDa protein similarly in some respects, but that the 20-kDa fragment was characteristic of the reaction with DSTP.

In the present study, DSTP was copurified with PPO activity from spinach thylakoids. This observation, together with the structural similarities between DSTP and PPO (Kuwabara 1995), suggests that the two proteins may be identical. It is known that the precursor to PPO, synthesized in vitro, can be imported into the thylakoid lumen (Sommer et al. 1994, Sokolenko et al. 1995), which is the location of the 23-kDa protein (Murata and Miyao 1985). Therefore, this protein could degrade the 23-kDa protein in vivo. Since no o-diphenol substrates for PPO activity have been found in the thylakoid lumen (Mayer 1987), it seems possible that this protein might function as a protease at that location. DSTP was previously purified from PSII membranes, but its level in these membranes was variable (Kuwabara and Hashimoto 1990, Kuwabara 1995). By contrast, the yield of the protein from intact thylakoids was fairly constant (Table 2). In the previous purification, some of the protein might have been lost during treatment of thylakoids with Triton X-100, which was used for the preparation of PSII membranes. The factor(s) that facilitates the binding of DSTP to the thylakoid membranes remains to be identified. It is possible that active photosynthesis might influence the binding, as is the case for other soluble thylakoid-lumenal proteins such as plastocyanin and violaxanthin de-epoxidase (Hager and Holocher 1994). The present study revealed that the 60-kDa DSTP formed a tetramer (Fig. 4). This tetramer-forming capacity must reside in the amino-terminal 41-kDa moiety because such fragments were able to form tetramers (Kuwabara and Hashimoto 1990, Golbeck and Cammarata 1981). In plants other than spinach, Anosike and Ayaebene (1982) reported

that the 31-kDa PPO of Dioscorea bulbifera formed a tetramer, and Wichers et al. (1984) reported that the 42kDa tyrosinase (an enzyme related to PPO) of Mucuna pruriens formed a dimer. However, such oligomeric structures have not been reported for many other PPOs. It remains to be determined whether the oligomeric structure represents the native state in vivo or just the structure of the protein extracted from thylakoid membranes, although the former seems more likely. At the start of this study, we regarded the 20-kDa fragment of the 23-kDa protein as the product of primary degradation by DSTP (Kuwabara 1995). In the course of the study, however, we found two other products of degradation, just below the band of the undigested 23-kDa protein (Fig. 3C). This finding suggests that the larger one of them, not the 20-kDa fragment, is the product of primary degradation. Unfortunately, the resolution of our SDS-PAGE was not always detailed in the 23-kDa region, and we could not quantitatively analyze the proteolytic activity by quantitating the product of primary degradation. The question of how the proteolytic activity is correlated with the PPO activity remains be clarified in the future through improvement in the resolution of the SDS-PAGE. In Fig. 3C, the intensity of staining of the 20-kDa fragment as well as of the undigested 23-kDa protein was very similar in lanes 22-29, and the peaks of DSTP-T and DSTP-M were not apparent for the proteolytic activity. The only remarkable difference between these two peaks was that one of the two degradation products described in the preceding paragraph accumulated in the reaction with DSTP-M, but not in the reaction with DSTP-T (Fig. 3C). This result suggests that DSTP-T rapidly degraded that fragment, whereas the similar degradation by DSTP-M was rather slow. Another possibility is that only the tetramer has the proteolytic activity and that the activity of DSTPM can be ascribed to the tetramer that was generated from the monomer by the equilibrium between the two molecular species (Kuwabara and Hashimoto 1990) after the separation of the two peaks. The copper content of plant PPO has not yet been established, but the amino acid sequences of PPOs from various plants have revealed two Cu-binding sites in each protein molecule (Shahar et al. 1992, Cary et al. 1992, Hunt et al. 1993, Newman et al. 1993, Hind et al. 1995, Sokolenko et al. 1995). One reason for this absence of information may be that PPO tends to degrade during purification (Mayer 1987) and the purification of the mature 60-kDa PPO was difficult. In the present study, the degradation was minimized by rapid purification on an affinity column and the use of 0.1% Tween 20 as a protein stabilizer. We determined that the monomeric species, DSTP-M, contained 0.4 Cu atom per protein molecule. Since PPO activity was enhanced by aging of the purified protein, the fact that the purified DSTP had substantial

Downloaded from pcp.oxfordjournals.org by guest on October 17, 2011

Discussion

185

186

Protease with polyphenol oxidase activity

The authors are grateful to Dr. N. Nakajima for providing the opportunity to use spinach plants grown at the National Institute for Environmental Studies, Tsukuba. This research was supported by a Grant-in-Aid for Cooperative Research (no. 04304004) from the Ministry of Education, Science, and Culture of Japan and by the Project Research Fund of the University of Tsukuba. References Anosike, E.O. and Ayaebene, A.O. (1982) Properties of polyphenol oxidase from tubers of the yam Dioscorea bulbifera. Phytochemistry 21: 1889-1893. Arnon, D.I. (1949) Copper enzymes in isolated chloroplasts. Polyphenoloxidase in Beta vulgaris. Plant Physiol. 24: 1-15.

Augustin, M.A., Ghazali, H.M. and Hashim, H. (1985) Polyphenoloxidase from guava (Psidium guajava L.). J. Sci. Food. Agric. 36: 12591265. Bradford, M.M. (1976) A rapid and sensitive method for the quantitation of protein utilizing the principle of protein-dye binding. Anal. Biochem. 72: 248-258. Cary, J.W., Lax, A.R. and Flurkey, W.H. (1992) Cloning and characterization of cDNAs for Vicia faba polyphenol oxidase. Plant Mol. Biol. 20: 245-253. Golbeck, J.H. and Cammarata, K.V. (1981) Spinach thylakoid polyphenol oxidase. Isolation, activation, and properties of the native chloroplast enzyme. Plant Physiol. 67: 977-984. Gutteridge, J.M.C. and Wilkins, S. (1983) Copper salt-dependent hydroxyl radical formation. Damage to proteins acting as antioxidants. Biochim.

Biophys. Ada 759: 38-41. Hager, A. and Holocher, K. (1994) Localization of the xanthophyll-cycle enzyme violaxanthin de-epoxidase within the thylakoid lumen and abolition of its mobility by a (light-dependent) pH decrease. Planta 192: 581589. Hind, G., Marshak, D.R. and Coughlan, S.J. (1995) Spinach thylakoid polyphenol oxidase: cloning, characterization, and relation to a putative protein kinase. Biochemistry 34: 8157-8164. Hunt, M.D., Eannetta, N.T., Yu, H., Newman, S.M. and Steffens, J.C. (1993) cDNA cloning and expression of potato polyphenol oxidase. Plant Mol. Biol. 21: 59-68. Kuwabara, T. (1995) The 60-kDa precursor to the dithiothreitol-sensitive protease of spinach thylakoids: structural similarities between the protease and polyphenol oxidase. FEBS Lett. 371: 195-198. Kuwabara, T. and Hashimoto, Y. (1990) Purification of a dithiothreitolsensitive tetrameric protease from spinach PSII membranes. Plant Cell Physiol. 31: 581-589. Kuwabara, T. and Suzuki, K. (1994) A prolyl endoproteinase that acts specifically on the extrinsic 18-kDa protein of photosystem II: purification and further characterization. Plant Cell Physiol. 35: 665-675. Kuwabara, T. and Suzuki, K. (1995) Reversible changes in conformation of the 23-kDa protein of photosystem II and their relationship to the susceptibility of the protein to a proteinase from photosystem II membranes. Plant Cell Physiol. 36: 495-504. Lanker, T., Flurkey, W.H. and Hughes, J.P. (1988) Cross-reactivity of polyclonal and monoclonal antibodies to polyphenoloxidase in higher plants. Phytochemistry 27: 3731-3734. Lerch, K. (1981) Copper monooxygenases: tyrosinase and dopamine /?monooxygenase. In Metal Ions in Biological Systems. Edited by Sigel, H. Vol. 13. pp. 143-186. Dekker, New York. Mayer, A.M. (1987) Polyphenol oxidase in plants-Recent progress. Phytochemistry 26: 11-20. Murata, N. and Miyao, M. (1985) Extrinsic membrane proteins in the photosynthetic oxygen-evolving complex. Trends Biochem. 10: 122-124. Murata, N., Miyao, M., Omata, T , Matsunami, H. and Kuwabara, T. (1984) Stoichiometry of components in the photosynthetic oxygen evolution system of photosystem II particles prepared with Triton X-100 from spinach chloroplasts. Biochim. Biophys. Ada 765: 363-369. Newman, S.M., Eannetta, N.T., Yu, H., Prince, J.P., Carmen de Vicente, M., Tanksley, S.D. and Steffens, J.C. (1993) Organization of the tomato polyphenol oxidase gene family. Plant Mol. Biol. 21: 1035-1051. Robinson, S.P. and Dry, I.B. (1992) Broad bean leaf polyphenol oxidase is a 60-kilodalton protein susceptible to proteolytic cleavage. Plant Physiol. 99: 317-323. Shahar, T., Hennig, N., Gutfinger, T., Hareven, D. and Lifschitz, E. (1992) The tomato 66.3-kDa polyphenoloxidase gene: molecular identification and developmental expression. Plant Cell A: 135-147. Sokolenko, A., Fulgosi, H., Gal, A., Altschmied, L., Ohad, I. and Herrmann, R.G. (1995) The 64 kDa polypeptide of spinach may not be the LHCII kinase, but a lumen-located polyphenol oxidase. FEBS Lett. 371: 176-180. Sommer, A., Ne'eman, E., Steffens, J . C , Mayer, A.M. and Harel, E. (1994) Import, targeting, and processing of a plant polyphenol oxidase. Plant Physiol. 105: 1301-1311. Stadtman, E.R. (1993) Oxidation of free amino acids and amino acid residues in proteins by radiolysis and by metal-catalyzed reactions. Annu. Downloaded from pcp.oxfordjournals.org by guest on October 17, 2011

PPO activity suggests that it might not have been in the fully native form. The number of Cu atoms in the native protein remains to be determined. The results depicted in Fig. 5 suggest that oxygen molecules are essential for the degradation of the 23-kDa protein by DSTP. Since DSTP can function as PPO, it is possible that some reactive complex is formed by the reaction center copper of DSTP and oxygen, as in some oxygenases that contain copper (Lerch 1981). It seems likely that such active species caused the proteolysis. It should be noted that the valence of the reaction center copper is not necessarily + 2 when it is functioning. Since the 23-kDa protein has only one cysteine residue and its sulfhydryl group is expected to be exposed on the surface of the protein molecule (Kuwabara and Suzuki 1995), it is possible that the sulfhydryl group can reduce the reaction center copper. The valence of the reaction center copper as well as the structure formed by the copper and oxygen needs to be clarified to unveil the mechanism by which the proteolysis takes place. Free Cu 2+ ions not only oligomerized but also degraded the 23-kDa protein, although they did not generate the 20-kDa fragment characteristic of proteolysis by DSTP (Fig. 6). The oligomerization of the 23-kDa protein may not have been due to the formation of intermolecular disulfide bridge because the samples illustrated in Fig. 6 had been treated with 5 mM DTT prior to SDS-PAGE. The disulfide-linked dimer of the 23-kDa protein that occurs upon storage of the protein under non-reducing conditions can be eliminated by such treatment with DTT. As described above, the reduction of Cu 2+ ions by the sulfhydryl group of the 23-kDa protein may lead to the formation of Cu + ions. If oxygen molecules were substantially reduced by the sulfhydryl group and/or Cu + ions, hydrogen peroxide would be generated. This situation would allow hydroxyl radicals to be generated through the Fenton reaction (Gutteridge and Wilkins 1983, Stadtman 1993), causing the oligomerization and degradation of the 23-kDa protein. This possible mechanism of the structural alterations of the 23-kDa protein remains to be experimentally proven.

Protease with polyphenol oxidase activity Rev. Biochem. 62: 797-821. Vaughn, K.C. and Duke, S.O. (1981) Tissue localization of polyphenol oxidase in Sorghum. Protoplasma 108: 319-327. Vaughn, K.C. and Duke, S.O. (1984) Tentoxin stops the processing of polyphenol oxidase into an active protein. Physiol. Plant. 60: 257-261. Vaughn, K.C, Lax, A.R. and Duke, S.O. (1988) Polyphenol oxidase: the

187

chloroplast oxidase with no established function. Physiol. Plant. 72: 659-665. Wieners, H.J., Peetsma, O.J., Malingre, T.M. and Huizing, H.J. (1984) Purification and properties of a phenol oxidase derived from suspension cultures of Mucuna pruriens. Planta 162: 334-341.

(Received September 27, 1996; Accepted December 4, 1996)

Downloaded from pcp.oxfordjournals.org by guest on October 17, 2011