HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

J. Biol. Chem., Vol. 277, Issue 17, 14747-14756, April 26, 2002

This Article Abstract Full Text (PDF) All Versions of this Article: 277/17/14747    most recent M100766200v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Motoki, A. Articles by Katoh, S. Search for Related Content PubMed PubMed Citation Articles by Motoki, A. Articles by Katoh, S. Social Bookmarking                      What's this?

A Domain of the Manganese-stabilizing Protein from Synechococcus elongatus Involved in Functional Binding to Photosystem II^*

Akihiro Motoki^§, Mina Usui, Tsuneo Shimazu, Masahiko Hirano, and Sakae Katoh^¶

From the  Biological Sciences Department, Toray Research Center Inc., Kamakura 248-8555, Japan and the ^¶ Department of Biology, Faculty of Sciences, Toho University, Funabashi 274-8510, Japan

Received for publication, January 26, 2001, and in revised form, January 22, 2002

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Site-directed mutagenesis was performed to investigate whether the two protease-sensitive sequences Phe¹⁵⁶-Gly¹⁶³ and Arg¹⁸⁴-Ser¹⁹¹, of the manganese-stabilizing protein (MSP) from a thermophilic cyanobacterium, Synechococcus elongatus (Motoki, A., Shimazu, T., Hirano, M., and Katoh, S. (1998) Biochim. Biophys. Acta 1365, 492-502), are involved in functional interaction with photosystem II (PSII). The ability of MSP to bind to its functional site on the PSII complex and to reactivate oxygen evolution was dramatically reduced by the substitution of Arg¹⁵², Asp¹⁵⁸, Lys¹⁶⁰, or Arg¹⁶² with uncharged residues, by insertion of a single residue between Phe¹⁵⁶ and Leu¹⁵⁷, or by deletion of Leu¹⁵⁷. Substitution of each of the four charged residues with an identically charged residue showed that the charges at Asp¹⁵⁸, and possibly Lys¹⁶⁰, are important for the electrostatic interaction with PSII. The reactivating ability was also strongly affected by the alteration of Phe¹⁵⁶ to Leu. Replacement of Lys¹⁸⁸, the only strictly conserved charged residue in the Arg¹⁸⁴-Ser¹⁹¹ sequence, by Gln had only a marginal effect on the function of MSP. High affinity binding of MSP to PSII was also affected significantly by mutation at Arg¹⁵², which is located in a region (Val¹⁴⁸-Arg¹⁵²) strictly conserved among the 14 sequences so far reported. These results imply that the Val¹⁴⁸-Gly¹⁶³ sequence, which is well conserved among MSPs from cyanobacteria to higher plants, is a domain of MSP for functional interaction with PSII.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Photosynthetic oxygen evolution is mediated by PSII,¹ a multiprotein complex carrying chlorophylls, carotenoids, and a variety of redox cofactors (for reviews see Refs. 1 and 2). Among the protein components of PSII, the extrinsic 33-kDa protein is also called the manganese-stabilizing protein (MSP) and is associated with the lumenal surface of the PSII complex together with two or three other extrinsic proteins, i.e. the 23-kDa and 17-kDa proteins in higher plants and green algae (3), cytochrome c₅₅₀ and a 12-kDa protein in cyanobacteria (4), and cytochrome c₅₅₀, the 12-kDa protein, and a 20-kDa protein in red algae (5). MSP stabilizes the manganese cluster, which catalyzes oxidation of water to molecular oxygen. Dissociation of the protein by washing with 1 M CaCl₂ or MgCl₂ (6) or 2.6 M urea plus 0.2 M NaCl (7) leads to a gradual release of two of four Mn²⁺ ions present in the cluster. Oxygen evolution is strongly suppressed by the release of MSP, but the lost activity can be restored by rebinding of the protein (8, 9). Because no cofactor is associated with MSP, it is considered that binding of the protein itself is responsible for optimal activity of oxygen evolution.

Cross-linking experiments with various bifunctional reagents have shown that MSP is associated with or in close proximity to essentially all of the major intrinsic proteins of the PSII complex (10-15). Various attempts have been made to locate binding sites for PSII or identify amino acid residues on MSP that are involved in binding to its functional site on the PSII complex. The N-terminal sequence of spinach MSP was suggested to have a binding site to PSII because removal of 16 or 18 amino acid residues from its N terminus by protease digestion resulted in total loss of the protein binding (16). Recently, evidence was presented indicating that the N-terminal sequence is necessary for maintaining the binding ability of the protein to PSII but might not be involved in the intermolecular binding itself (17). It was also suggested that Asp⁹, the only conserved, charged residue in the N-terminal 18-amino acid sequence, might engage in both intra- and intermolecular interactions (18). Cross-linking with EDC, which covalently links amino and carboxyl groups in van der Waals contact, indicated that MSP is bound directly to CP47, an intrinsic chlorophyll-carrying protein of the PSII complex through electrostatic interaction (10-12). Analysis of EDC-cross-linked products showed that a charge-pair interaction exists between a charged residue in the Asp¹-Lys⁷⁶ sequence of the spinach MSP and an oppositely charged residue in the Phe³⁶⁴-Asp⁴⁴⁰ sequence of CP47 (13).

Chemical modification with specific reagents showed that six conserved lysyl residues distributed over the entire sequence of spinach MSP are accessible to at least one of the reagents when the protein is free in solution but not when the protein is associated with PSII membranes (19, 20). This suggests that these lysyl residues are located in domains of the protein that are in contact with the PSII complex and might participate in intermolecular interaction. Cross-linking of the Lys¹⁵⁹-Lys²³⁶ domain of spinach MSP to CP47 with a reagent that cross-links lysyl residues within a 1.2-nm radius has been reported (21). Four arginyl residues of the spinach protein may also be involved in the binding to PSII because they were modified with a specific reagent only when the protein was free in solution (20).

Involvement of negatively charged amino acid residues on MSP in binding to PSII was also suggested by EPR studies with a free radical-relaxing agent dysprosium, which showed that the binding site for MSP on the PSII complex is positively charged (22). Cross-linking experiments in which carboxyl groups of spinach MSP were activated for cross-linking prior to reconstitution with PSII membranes or vice versa, showed that all carboxyl groups involved in intermolecular cross-linking with EDC are located on MSP (23). The binding affinity of spinach MSP was strongly reduced by chemical modification of carboxyl groups of aspartyl and glutamyl residues with glycine methyl ester when it was free in solution but not when it was bound to PSII membrane (24). The reduced binding affinity was related to modification of two or three acidic amino acid residues located in domains Asp¹⁵⁷-Asp¹⁶⁸ and Glu²¹²-Gln²⁴⁷, because these residues were modified only when the protein was free in solution. Mutation of Asp¹⁵⁹ to other residues in Synechocystis MSP affected oxygen evolution significantly (25). On the other hand, in another study, chemical modification of carboxyl groups with glycine methyl ester failed to affect the ability of the spinach protein to reactivate oxygen evolution upon reconstitution at a high protein/PSII ratio (20). Site-directed mutagenesis of completely or partially conserved aspartyl and glutamyl residues on spinach MSP was performed, but none of the mutations had a dramatic effect (26, 27). All of the mutant proteins were able to bind to PSII membranes, and the ability of the protein to reactivate oxygen evolution was only marginally affected by alteration of several residues. Thus, amino acid residues on MSP that are involved in functional interaction with PSII still remain to be identified.

We showed previously that cleavage of MSP from the thermophilic cyanobacterium Synechococcus elongatus at a single site between Phe¹⁵⁶ and Gly¹⁶³, or between Agr¹⁸⁴ and Ser¹⁹¹, by trypsin, chymotrypsin, or lysylendopeptidase resulted in total loss of the ability of the protein to bind to PSII and to restore the oxygen-evolving activity (28). Introduction of a nick between Phe¹⁵⁶ and Leu¹⁵⁷, a cleavage site specific to chymotrypsin, by site-directed mutagenesis also led to loss of the binding and reactivating ability of MSP. The cause of the inactivation of MSP may, however, not be related to a local change at the protease-cleavage site, because the mutation was accompanied by a significant change in the secondary conformation of the protein. Insertion of a methionyl residue between Phe¹⁵⁶ and Leu¹⁵⁷ abolished functional binding of MSP and unexpectedly created an appreciable binding affinity for nonspecific sites of the PSII complex. Because the protein conformation was little affected by insertion of the residue, the observed changes in the protein binding were ascribed to a small structural change(s) in the mutated polypeptide. Based on these observations, we proposed that the two protease-sensitive regions, Phe¹⁵⁶-Gly¹⁶³ and Arg¹⁸⁴-Ser¹⁹¹, are possible candidates for interaction with the PSII complex.

In the present study, site-directed mutagenesis was performed to identify the amino acid residues involved in effective binding of Synechococcus MSP to PSII. Conserved, charged residues in the first protease-sensitive sequences were substituted with an uncharged residue, and the ability of the mutant proteins to bind to urea/NaCl-washed PSII and to restore the oxygen evolution was determined. Charge-preserving substitution with an identically charged residue was also performed to examine whether the residues are involved in electrostatic interaction. For comparison, conserved, uncharged residues in and near the first protease-sensitive sequence Phe¹⁵⁶-Gly¹⁶³ and conserved, charged residues in other domains of the protein were altered. The results obtained indicate that Arg¹⁵², Phe¹⁵⁶, Asp¹⁵⁸, Lys¹⁶⁰, and Arg¹⁶² located in or near the first protease-sensitive sequence are essential for binding of MSP to its functional site. Mutant proteins were also constructed by inserting a residue with a side group smaller than that of Met between Phe¹⁵⁶ and Leu¹⁵⁷ or by deleting Leu¹⁵⁷, in order to elucidate the structural configuration of MSP around Phe¹⁵⁶ and Leu¹⁵⁷, which are required for the functional binding of MSP.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Site-directed Mutagenesis of the psbO Gene-- Site-directed mutagenesis was performed by the method described previously (28). The pET8c expression vector, which contains the wild-type psbO gene of S. elongatus, was constructed as described (29). The XbaI/BamHI fragment of the vector containing the whole psbO gene was cloned into the multiple cloning site of pBluescriptII phagemid vector (Stratagene) and subcloned into SacI/HindIII site of the M13mp18 phagemid vector. The Sculpture^TM in vitro mutagenesis system (Amersham Biosciences) was used to introduce a desired mutagenesis into a M13mp18 template (30-32). The synthesized, mutagenic oligonucleotides employed are shown in Table I. The mutation was checked by sequencing the single-stranded DNA of each mutated psbO gene. The mutated psbO genes were expressed in Escherichia coli cells, and the mutant MSPs were purified by the same method employed for isolation of the wild-type MSP (29). The yields of purified protein were 5-10 mg of protein/liter of culture medium.

Preparation of Oxygen-evolving PSII Complexes-- Oxygen-evolving PSII complexes were isolated from S. elongatus as described by Ichimura et al. (33). Thylakoid membranes suspended in solution A (50 mM MES/NaOH (pH 6.0), 10 mM NaCl, 5 mM MgCl₂) containing 1 M sucrose (1 mg of chlorophyll/ml) were treated with 0.20-0.25% sucrose monolaurate for 1 h at 20 °C in the dark. After the addition of an equal volume of solution A to reduce the sucrose concentration to 0.5 M, 2 volumes of the suspension was placed on 1 volume of solution A containing 1 M sucrose and centrifuged at 300,000 × g for 1 h. A green band that appeared at the interface between 0.5 and 1 M sucrose contained PSII complexes. The complexes were collected by centrifugation at 370,000 × g for 40 min and suspended in solution A containing 25% glycerol.

Binding of MSP-- PSII complexes were treated with 2.6 M urea and 0.2 M NaCl for 30 min at 0 °C to extract the three extrinsic proteins, precipitated by centrifugation, and suspended in solution A containing 25% glycerol. Washed PSII complexes were incubated with each mutant protein in 20 mM MES/NaOH (pH 6.0), 10 mM CaCl₂, and 10% glycerol at 0 °C at an indicated protein/PSII ratio. A stoichiometry of 45 chlorophylls/PSII was assumed for the PSII complexes (33). After incubation for 1 h, the PSII complexes were precipitated by centrifugation at 370,000 × g for 40 min and then washed once with and resuspended in 50 mM MES/NaOH (pH 6.0), 10 mM MgCl₂, and 25% glycerol. Wild-type and mutant MSPs bound to PSII were analyzed by SDS-PAGE. Samples were denatured with 5% SDS and 60 mM dithiothreitol for 30 min, then applied to polyacrylamide gels. The acrylamide concentration was 4.5% for the stacking gel and 12.5% for the resolving gel, and both gels contained 0.1% SDS and 6 M urea. After electrophoresis, gels were stained with Coomassie Brilliant Blue R-250 and scanned at 560 nm with an ATTO AE-6900 densitometer. Amounts of MSPs bound to PSII were estimated by measuring the peak area of the protein with that of CP47 as reference.

Determination of Oxygen Evolution-- Oxygen evolution was determined with a Clark-type oxygen electrode at 40 °C as described (33). White light of a saturating intensity was provided from a halogen lamp (Watanabe Shoko R&D Corp). The reaction medium contained 50 mM MES/NaOH (pH 6.0), 10 mM NaCl, 5 mM MgCl₂, 10 mM CaCl₂, 1 M sucrose, and 0.4 mM 2,6-dichloro-p-benzoquinone.

Circular Dichroism Spectrometry-- Far-UV and near-UV CD spectra were determined with a JASCO J-820 spectrometer. Data were collected every 0.1 nm with a bandwidth of 1 or 2 nm and a 0.5-s time constant. For determination of the far-UV CD spectra, proteins were dissolved in 20 mM potassium phosphate (pH 6.5) at a protein concentration of 80 µg/ml, which was determined by amino acid analysis. Eight repetitive scans were collected and averaged; the secondary structure elements were analyzed with the program VARSLC (34, 35). Near-UV CD spectra were determined with proteins dissolved in 20 mM MES/NaOH (pH 6.5). Measurement was repeated 6-16 times depending on the protein concentrations (see Fig. 4), and the signals obtained were averaged.

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Substitution of Charged Amino Acid Residues-- There are several lines of evidence indicating that binding of MSP to its functional site involves electrostatic interactions between charged amino acid residues on the protein and oppositely charged residues on intrinsic proteins of PSII (6, 7, 36, 37). Single-amino acid substitution experiments were, therefore, performed for conserved, charged residues located in the two protease-sensitive sequences. The first protease-sensitive sequence, Phe¹⁵⁶-Gly¹⁶³, contains two positively charged amino acid residues, Lys¹⁶⁰ and Arg¹⁶², and a negatively charged residue, Asp¹⁵⁸, which are completely conserved among 14 sequences of MSPs from cyanobacteria, algae, and higher plants (Fig. 1). Mutant proteins were constructed by replacing Lys¹⁶⁰ and Arg¹⁶² with Gln (K160Q,R162Q), and by replacing Asp¹⁵⁸ with Asn (D158N). Binding affinities of the mutant proteins to PSII were determined by measuring the amount of each mutant protein bound to urea/NaCl-washed PSII at different protein/PSII ratios. Reconstituted PSII complexes were washed once to remove loosely bound MSP. As shown in Fig. 2, binding of wild-type MSP was saturated at a protein/PSII ratio of two. A further increase in the protein/PSII ratio led to only a small increase in the amount of the protein bound, reflecting the ability of the protein to bind specifically and stoichiometrically to its functional site on the PSII complex. On the other hand, binding of the protein to PSII complexes was dramatically altered by the mutations introduced. The binding curves of the K160Q and R162Q mutant proteins were sigmoidal, and the amounts of the two mutant proteins bound to PSII at low protein/PSII ratios were much lower than that of the wild-type protein, indicating that alteration of Lys¹⁶⁰ and Arg¹⁶² leads to a large decrease in the binding affinity for the functional site. At protein/PSII ratios above four, the amount of mutant proteins bound increased linearly with an increase in protein concentration; this increase is apparently steeper than that of the wild-type protein. This is a feature of nonspecific binding that has been described first for mutant proteins with Met inserted between Phe¹⁵⁶ and Leu¹⁵⁷ and with Leu¹⁵⁷ replaced by Met (28). Nonspecific binding was particularly enhanced upon substitution of Asp¹⁵⁸ with Asn so that the amount of the D158N mutant protein bound to PSII far exceeded that of the wild-type protein at a protein/PSII ratio of six. Nevertheless, the high-affinity binding of the protein to PSII was appreciably reduced by the mutation. The results show that all three of the conserved, charged residues located in the Phe¹⁵⁶-Gly¹⁶³ sequence are required for high affinity binding of the protein to PSII, consistent with the notion that the first protease-sensitive sequence is a domain for interaction with PSII. The high affinity binding, however, was also severely affected by alteration of Arg¹⁵² to Gln, which is located three residues upstream of the first protease-sensitive sequence. This indicates that the binding domain is larger than the Phe¹⁵⁶-Gly¹⁶³ sequence and possibly involves the Val¹⁴⁸-Arg¹⁵² sequence, which is strictly conserved in all the sequences of MSP shown in Fig. 1.

View larger version (60K): [in this window] [in a new window]   Fig. 1.   Alignment of amino acid sequences of MSP from 14 oxygenic photosynthetic organisms. Amino acid residues substituted or deleted by site-directed mutagenesis in the present study are indicated by triangles. Dashes indicate amino acid residues identical to those in S. elongatus; dots indicate vacant position in the sequence; stars indicate residues completely conserved in the 14 sequences. The two protease-sensitive sequences are boxed. 1, Synechococcus elongatus; 2, Anacystis nidulans R2; 3, Synechocystis sp. PCC6803; 4, Anabaena sp. PCC7120; 5, Chlamydomonas reinhardtii; 6, Euglena gracilis Z; 7, garden pea; 8, Arabidopsis thaliana; 9, wheat; 10, potato; 11, tomato; 12, tobacco; 13, rice; 14, spinach.

View larger version (22K): [in this window] [in a new window]   Fig. 2.   Effects of substitution of Arg152, Asp158, Lys160, and Arg162 with an uncharged residue on the binding affinity of MSP to PSII. Urea/NaCl-washed PSII complexes were reconstituted with each mutant protein at the indicated protein/PSII ratios. The amount of wild-type protein bound to PSII at the protein/PSII ratio of three was taken as 100%. , wild-type MSP; , D158N; , R152Q; , K160Q; ×, R162Q.

The effects of the substitution mutations on the ability of MSP to restore the oxygen-evolving activity were also investigated. As shown in Table II, oxygen evolution was strongly suppressed by urea/NaCl washing, but the activity was partially restored by the addition of wild-type MSP. This reflects the normal function of the protein because the small magnitude of reactivation (about 30% of activity in untreated PSII complexes) can be ascribed to the following two reasons. First, the full activity of oxygen evolution of cyanobacterial or red algal PSII requires three or four extrinsic proteins (38-40); MSP only partially supports the activity. Second, the smaller reactivating effect of MSP may be because a larger proportion of PSII preparations had been irreversibly damaged during protein extraction. This, however, will in principle not affect our reconstitution results as far as the comparison of the relative effects of wild-type and mutant MSPs is concerned. The mutations that strongly affected the high affinity binding of the protein had a severe impact on the reactivation ability of MSP. The levels of oxygen evolution restored by the mutant MSPs were only less than 20% of that restored by wild-type protein even at a protein/PSII ratio of five. The results confirm that the four charged residues are essential for functional interaction with PSII and that the nonspecific binding to PSII is nonfunctional. Table II also shows that not all of the strictly conserved, charged residues are required for the function of MSP. Substitution of Lys¹⁸⁸, the only strictly conserved, charged residue in the second protease-sensitive sequence, Arg¹⁸⁴-Ser¹⁹¹ with Gln, had only a marginal effect on the reactivation ability of MSP. This casts a doubt on the notion that the second protease-sensitive sequence is a region for interaction with PSII. The alteration of Lys⁵⁹, Arg⁷³, Lys¹²³, and Lys²³⁴ to Gln had no effect on the function of MSP. Thus, inactivation of MSP is specific to the substitution of a charged residue in the Val¹⁴⁸-Gly¹⁶³ sequence. The present study, however, does not exclude the existence of a binding site for PSII in other regions of the protein because there are still eight strictly conserved, charged residues that remain to be investigated (see Fig. 1).

CD spectroscopy was performed to investigate the effect of the substitution mutations on the secondary and tertiary conformations of MSP. The secondary conformation of the protein has previously been estimated from the UV CD spectrum in the 197-250 nm region as well as Fourier-transform infrared spectrum (41). Because CD data provide more accurate estimates of the proportions of secondary structure elements when the CD spectrum is extended to shorter wavelengths (35), the CD spectra of wild-type and mutant proteins were determined from 181 to 260 nm. The far-UV CD spectrum of wild-type protein showed a strong positive band at 197 nm, a broad negative band between 205 and 225 nm, and a zero-line crossover point at 192 nm (Fig. 3). Substitution of Arg¹⁵², Asp¹⁵⁸, Lys¹⁶⁰, and Arg¹⁶² had no significant effect on the spectrum (Fig. 3 and data not presented). The relative abundance of secondary structure elements estimated from the far-UV CD data is shown in Table III. The wild-type protein contains -sheet severalfold more abundantly than -helix. The values are similar to those of spinach MSP (42-44) but different from the previous estimate from the CD spectrum in the 197-250 nm region, which predicted a higher proportion of -helix and a lower proportion of -sheet than those shown in Table III (41). Although not shown here, Fourier-transform infrared spectroscopy with an improved precision also indicated a larger abundance of -sheets than shown previously (41). The mutations that significantly decreased the binding and reactivating abilities of MSP had no significant effect on the secondary conformation of the protein; small variations in the proportion of the secondary structure elements observed are in the range of experimental error. Thus, loss of the functional binding to PSII is not a result of distortion of the secondary conformation of the protein.

View larger version (21K): [in this window] [in a new window]   Fig. 3.   Far-UV CD spectra of wild-type and mutant MSPs with Arg152 replaced by Gln and Asp158 replaced by Asn. Thick line, wild-type MSP; thin line, R152Q; dotted line, D158N.

The near-UV CD spectra were determined to examine whether tertiary structure of the protein is affected by the substitution mutations. The near-UV CD spectrum of spinach MSP shows two prominent bands, a tryptophan band at 294 nm and a tyrosine band at 285 nm. These bands disappeared when the tertiary structure that contributes to the local environment of the aromatic amino acids was altered by cleavage of the sulfhydryl bond (45), acidification (43), or heat treatment (44) of the protein. Synechococcus MSP lacks tryptophan, and accordingly no band at 294 nm was observed (Fig. 4). Moreover, and unexpectedly, no conspicuous tyrosine band was observed at 285 nm. A simple explanation would be that the 285 nm band originates from one (or more) of the three tyrosine residues that are present in the spinach protein but not in the Synechococcus protein (see Fig. 1). Fig. 4, however, shows four positive bands at 261, 266, 274, and 282 nm, which can be ascribed to the fine-structure bands of phenylalanine and tyrosine (46). A more likely explanation, therefore, is that a peak at 290 nm corresponds to a tyrosine band that is overlapped with a sharp negative band at shorter wavelengths. In any case, these spectral features remained essentially unaltered upon substitution of any of the four charged residues, indicating that inactivation of the mutant MSPs was not associated with a significant change in the tertiary structure of the protein, either (Fig. 4 and data not presented).

View larger version (22K): [in this window] [in a new window]   Fig. 4.   Near-UV CD spectra of wild-type and mutant MSPs with Arg152 or Arg162 replaced by Gln. Solid line, wild-type MSP (985 µg/ml); dashed line, R152Q (827 µg/ml); dotted line, R162Q (1760 µg/ml).

If the four charged residues contribute to the effective binding of MSP through electrostatic interaction with oppositely charged residues on the PSII complex, substitution of each residue with an identically charged residue might have little impact on the function of the protein. This was found to be the case with Asp¹⁵⁸. Alteration of the aspartyl residue to a glutamyl residue had no effect on the binding of the protein to PSII (Fig. 5). The D158E mutant protein was able to reconstitute oxygen evolution activity as effectively as wild-type protein (Table IV). Thus, a negative charge at Asp¹⁵⁸ is critically important for the functional binding of MSP. In contrast, charge-preserving alteration of Arg¹⁵² to Lys, Lys¹⁶⁰ to Arg, or Arg¹⁶² to Lys resulted in a significant decrease in the high affinity binding and an increase in nonspecific binding to PSII (Fig. 5). The amounts of the R152K, K160R, and R162K mutant proteins bound to PSII at low protein/PSII ratios appear to be larger than those of the corresponding charge-deleted mutant proteins. The difference should be ascribed to the difference in nonspecific binding, because the charge-preserving mutations affected the ability of the protein to reconstitute oxygen evolution as severely as the corresponding charge-deleting mutations. Only the K160R mutant protein restored oxygen evolution somewhat more effectively than the K160Q mutant protein, suggesting participation of this residue in electrostatic interaction (Table IV).

View larger version (22K): [in this window] [in a new window]   Fig. 5.   Effects of substitution of Arg152, Asp158, Lys160, and Arg162 with an identically charged residue on the binding affinity of MSP to PSII. Experimental conditions were the same as in Fig. 2. , wild-type MSP; , D158E; , R152K; , K160R; ×, R162K.

Substitution of Uncharged Residues-- Five uncharged residues, Phe¹⁵⁶, Leu¹⁵⁷, Pro¹⁵⁹, Gly¹⁶¹, and Gly¹⁶³, which are located in the Phe¹⁵⁶-Gly¹⁶³ sequence, are strictly conserved among MSPs from cyanobacteria to higher plants, except for Gly¹⁶¹, which is conservatively replaced by Ala in two cyanobacteria (see Fig. 1). Substitution of Phe¹⁵⁶ with Leu resulted in a nearly total loss of the reactivating ability of MSP (Table V). Thus, Phe¹⁵⁶ is required for effective binding of the protein, although the effects of the mutation on binding of the protein to its functional site could not be determined accurately because of an extremely strong affinity the mutant protein exhibited for nonspecific sites (not shown). Alteration of Phe¹⁵⁶ to Tyr had no effect on the function of the protein. This indicates that the structure of the side group is crucial for effective interaction with PSII. Substitution of Gly¹⁶¹ or Gly¹⁶³ with Gln led to a small decline in the reactivating ability of MSP, whereas alteration of Pro¹⁵⁹ to Ala had no effect.

The Thr¹⁵³-Asn¹⁵⁵ and Leu¹⁶⁴-Ser¹⁶⁶ sequences, which are located adjacent to the first protease-sensitive sequence, are not conserved even in cyanobacterial MSPs. Replacement of a single residue in these two sequences or the entire three-amino acid sequences had no effect on the ability of MSP to restore the oxygen-evolving activity (not shown). Although Arg¹⁵² is required for the functional binding of MSP, Val¹⁴⁸, Pro¹⁴⁹, and Tyr¹⁵¹, located in the conserved Val¹⁴⁸-Arg¹⁵² sequence, could be replaced by Thr, Ala, and Phe, respectively, without affecting the reactivating ability of MSP. No attempt was made to alter Ser¹⁵⁰. Gly¹⁶⁷ and Tyr¹⁶⁸ are also completely conserved in the 14 sequences of MSP. The reactivating ability of MSP was moderately affected by substitution of Tyr¹⁶⁸ with Phe, whereas replacement of Gly¹⁶⁷ by Gln had no influence on the function of MSP.

Insertion of a Residue between Phe¹⁵⁶ and Leu¹⁵⁷ and deletion of Leu¹⁵⁷-- We showed previously (28) that MSP becomes unable to bind to its functional site and reactivate oxygen evolution when a methionyl residue is inserted between Phe¹⁵⁶ and Leu¹⁵⁷. The mutant protein, however, showed a strong affinity for nonspecific binding. Extension of this experiment provided important information on both effective and nonspecific interaction of MSP with PSII. Because the protein conformation was little affected by the insertion of Met, loss of the functional binding of the protein is ascribed to a small structural change in the protease-sensitive sequence. If this interpretation was correct, the inactivation of MSP can be ascribed to an effect of either the side group of the methionyl residue inserted or a small increase in the polypeptide length between Phe¹⁵⁶ and Leu¹⁵⁷. To examine the effect of the side group, mutant MSPs were constructed by inserting an amino acid residue that has a side group smaller than that of Met between Phe¹⁵⁶ and Leu¹⁵⁷. The three mutant MSPs, which were constructed by inserting Gly, Ala, and Val, are called 156+G+157, 156+A+157 and 156+V+157, respectively. Leu¹⁵⁷ is not essential for the functional binding of MSP to PSII because the residue could be substituted with Met without affecting the reactivating ability of the protein (28). In the fourth mutant protein (157L), therefore, Leu¹⁵⁷ was deleted to examine whether the protein binding is affected by a change in the polypeptide length. As shown in Table VI, the protein became totally ineffective in reactivation of oxygen evolution when Val, Ala, or even Gly, which has no side chain, was inserted between Phe¹⁵⁶ and Leu¹⁵⁷, indicating that inactivation of the protein is independent of the size and structure of side chain of the residue inserted. Deletion of Leu¹⁵⁷ also led to a large decline in the reactivating ability of the protein. None of the four mutant proteins bound to PSII to an appreciable amount at the protein/PSII ratio of five, indicating that the mutations abolished high affinity binding without enhancing nonspecific binding of the protein to PSII (data not shown). These results indicate that binding of MSP to its functional site is extremely sensitive to a small change in the polypeptide length in the region between Phe¹⁵⁶ and Leu¹⁵⁷. The observation that nonspecific binding appeared upon insertion of Met but not upon insertion of Gly, Ala, or Val implies that the structural factor that confers an affinity for nonfunctional sites on the protein is the side group of Met inserted. This is consistent with the previous observation that nonfunctional binding became appreciable upon replacement of Leu¹⁵⁷ by Met (28).

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

We have constructed more than 30 variants of Synechococcus MSP by substituting, inserting, or deleting an amino acid residue to investigate whether the protease-sensitive Phe¹⁵⁶-Gly¹⁶³ and Arg¹⁸⁴-Ser¹⁹¹ sequences are regions for functional binding to PSII. The cyanobacterial MSP served as an excellent tool for investigation of the protein binding by means of site-directed mutagenesis and in vitro reconstitution, allowing us to identify several charged and uncharged amino acid residues that are essential for binding of MSP to its functional site. The mutations that effectively altered the ability of MSP to reactivate oxygen evolution are illustrated in Fig. 6.

View larger version (16K): [in this window] [in a new window]   Fig. 6.   Mutations that affected the function of MSP. The large and small upward arrows indicate mutations that affected the ability of Synechococcus MSP to reactivate oxygen evolution strongly and weakly, respectively. Downward arrows indicate mutations that had no effect on the reactivating ability of the protein. The first protease-sensitive sequence is boxed. Asterisks indicate residues completely conserved in the 14 sequences of MSP.

The first protease-sensitive sequence contains a highly conserved acidic residue, Asp¹⁵⁸. Substitution of this residue with Asn resulted in a large decrease in the ability of MSP to bind to functional site and to restore the oxygen-evolving activity, whereas its charge-preserving replacement by Glu had no inhibitory effect, indicating that the negative charge at Asp¹⁵⁸ is required for the protein binding. Thus, Asp¹⁵⁸ is involved either in charge-pair interaction with an oppositely charged residue on an intrinsic protein of PSII or in intramolecular charge-pair interaction necessary for maintenance of a functional structure of MSP. Far- and near-UV CD spectroscopy demonstrated that inactivation of MSP by mutation of Asp¹⁵⁸ to Asn was not accompanied by a significant change in the secondary and tertiary conformation of the protein, respectively. Substitution of the corresponding aspartyl residue (Asp¹⁵⁷) in spinach MSP was also shown to have no influence on intramolecular salt bridges formed upon cross-linking with EDC (17). In addition, carboxyl groups in the domain Asp¹⁵⁷-Asp¹⁶⁸ of the spinach protein were shown to participate in the binding of spinach MSP to PSII (24). These observations favor the involvement of Asp¹⁵⁸ in an intermolecular charge-pair interaction between MSP and PSII. Substitution of the corresponding aspartyl residue on MSPs from other photosynthetic organisms had weaker effects, a 5-15% decline in the reactivating ability of spinach MSP (27) and a 35-40% reduction in the rate of oxygen evolution in Synechocystis cells (25). These differences may be related to differences in protein binding between higher plants and cyanobacteria or in the flexibility of proteins between mesophiles and thermophiles.

Six conserved lysyl residues, Lys⁶⁶, Lys⁷⁶, Lys¹³⁰, Lys¹⁵⁹, Lys¹⁸⁶, and Lys²³⁶, of spinach MSP were suggested to be located in regions of the protein in contact with PSII because they were modified with amino group-specific reagents when MSP was free in solution but not when the protein was bound to PSII membranes (19, 20). The present study shows that not all of the conserved lysyl residues are required for functional binding to PSII. Substitution of Lys⁵⁹, Lys¹²³, Lys¹⁸⁸, and Lys²³⁴ of Synechococcus MSP (which correspond to Lys⁶⁶, Lys¹³⁰, Lys¹⁹⁰, and Lys²³⁶ of the spinach protein, respectively) with Gln had no influence on the function of the protein. Alteration of Lys¹⁶⁰ (which corresponds to Lys¹⁵⁹ of the spinach protein) to Gln had, however, a severe impact on the ability of the protein to effectively bind to PSII. Loss of protein binding is not related to a change in the secondary or tertiary structure of the protein. The effect of the charge-preserving substitution of Lys¹⁶⁰ with Arg on the reactivating ability of the protein was somewhat less than that of the charge-deleting substitution of the residue with Gln. This can be explained by assuming that Lys¹⁶⁰ participates in the electrostatic interaction essential for the function of the protein and that Arg acts only imperfectly for Lys¹⁶⁰ due to the difference in the side group. The corresponding lysyl residue (Lys¹⁵⁹) on spinach MSP was unable to participate in an intramolecular charge-pair interaction because the residue was accessible to amino group modifiers when the protein was free in solution (20). It is highly likely, therefore, that Lys¹⁶⁰ contributes to binding of MSP to its functional site by electrostatically interacting with a negatively charged residue on an intrinsic protein of the PSII complex.

The contribution of arginyl residues to binding of spinach MSP has been suggested by chemical modification of the guanidino group of arginyl residues with 2,3-butanedione (20). None of the six arginyl residues present in MSP reacted with the reagent when the protein was bound to PSII membranes, whereas treatment of the protein in solution led to modification of four arginyl residues, which was accompanied by loss of the ability of the protein to bind to PSII and to reactivate oxygen evolution. Modified residues have not been identified yet because of the instability of the modified products (20). Site-directed mutagenesis allowed us to successfully identify arginyl residues that are essential for the protein binding and also provided important information as to a domain of the protein for effective binding to PSII. There are three arginyl residues, Arg⁷³, Arg¹⁵², and Arg¹⁶², which are completely conserved among MSPs from cyanobacteria, algae, and higher plants (Fig. 1). Arg⁷³ is not essential because the residue could be substituted with Gln without affecting the function of the protein. Mutations at either Arg¹⁵² or Arg¹⁶² affected the function of MSP significantly. Because these mutations did not induce any significant changes in the protein conformation, the results indicate that the two arginyl residues directly participate in effective interaction with PSII. No direct evidence was obtained indicating that Arg¹⁵² and Arg¹⁶² are involved in charge-pair interaction; the reactivating ability of MSP was nearly equally suppressed by charge-deleting and charge-preserving substitutions of each residue. A possibility remains, however, that the two residues participate in charge-pair interaction but that a positive charge each residue carries becomes incapable of interacting with a negative charge on the partner residue when Arg is altered to Lys. Because Arg¹⁵² is located three residues apart from the first protease-sensitive sequence, the result also shows that the binding domain of MSP is larger than suggested by the proteolytic experiments (28) and may involve the well conserved Val¹⁴⁸-Arg¹⁵² sequence.

Substitution of uncharged residue in the first protease-sensitive sequence also affected the function of MSP. Substitution of Phe¹⁵⁶ with Leu resulted in an almost complete loss of the ability of the protein to restore the oxygen-evolving activity, whereas replacement of the same residue by Tyr had no effect on the function of MSP. Thus, the aromatic side group of Phe¹⁵⁶ is required for the functional binding of the protein. The reactivating ability of MSP was moderately affected by substitution of either Gly¹⁶¹ or Gly¹⁶³ with Gln. This suggests that the side chain of Gln, which replaced Gly sterically, interferes with the protein-protein interaction. Alternatively, because Gly has no side chain, it may confer a greater conformational flexibility, possibly required in the first protease-sensitive region for the effective association of the protein with its functional site. The protein binding, however, was not appreciably affected by substitution of Pro¹⁵⁹ with Ala; this may give another constraint on the polypeptide conformation.

Conserved, uncharged residues located near the Phe¹⁵⁶-Gly¹⁶³ sequence were also examined. As stated above, Arg¹⁵² is located in the completely conserved five-amino acid sequence. Although substitution of three uncharged residues, Val¹⁴⁸, Pro¹⁴⁹, and Tyr¹⁵¹, in the sequence failed to affect the function of MSP, the result should be interpreted with caution because Val¹⁴⁸ and Tyr¹⁵¹ were replaced by residues with relatively small differences in the structure of the side chain. In particular, because Tyr¹⁵¹ was substituted with Phe, a possibility remains that an aromatic group at position 151 plays a role in the functional binding of the protein, which will be addressed in future studies. Furthermore, amino acid residues in the Gly¹⁶⁷-Asp¹⁶⁹ sequence, which is located three residues after the first protease-sensitive sequence, are either strictly conserved or conservatively replaced. Possible contribution of this sequence to the protein binding cannot be excluded because substitution of Tyr¹⁶⁸ with Phe weakly affected the function of MSP, and single amino acid substitution has not been performed yet for Asp¹⁶⁹.

Extension of the previous experiments with the mutant protein having Met inserted (28) between Phe¹⁵⁶ and Leu¹⁵⁷ showed that binding of MSP to its functional site is extremely sensitive to a small change in the polypeptide length caused by insertion of an amino acid residue between Phe¹⁵⁶ and Leu¹⁵⁷ or deletion of Leu¹⁵⁷. This is consistent with the above described results that residues in or near the first protease-sensitive region participate in the interaction of MSP with PSII and further suggests that the relative location of these residues is critical for effective binding to PSII. We suggest that binding of MSP to its functional site involves multiple charged residues in the Val¹⁴⁸-Gly¹⁶³ region, which interact electrostatically with oppositely charged residues on intrinsic proteins of the PSII complex, although contribution of other interactions cannot be excluded. Such a protein-complex interaction is possible only when the charged residues on the surface of MSP are arranged spatially so as to become in van der Waals contact with respective partner residues on the PSII complex. As such, the protein binding will be weakened or abolished by insertion or deletion of a residue, which alters the location of the charged residues in the Val¹⁴⁸-Gly¹⁶³ domain.

The Val¹⁴⁸-Gly¹⁶³ domain becomes a region for nonfunctional binding to PSII upon modification. Replacement of a single essential residue in the domain conferred on the protein the ability to nonspecifically bind to the PSII complex. A comparison of the insertion mutants constructed in the present and previous experiments (28) showed that the structure of the side chain of an inserted residue is critical for the nonspecific binding of the protein. It is concluded, therefore, that the conserved amino acid residues in the Val¹⁴⁸-Gly¹⁶³ domain are also important for the protein to minimize nonspecific interactions with PSII. Several mutant proteins exhibited an extremely strong affinity for nonspecific binding. This implies caution in using binding of MSP to PSII as an index of the functional integrity of the protein when reconstitution is performed at a low protein/PSII protein ratio.

In contrast to the first protease-sensitive sequence where all the eight residues are completely conserved except for Gly¹⁶¹, which is conservatively replaced, the second protease-sensitive Arg¹⁸⁴-Ser¹⁹¹ sequence has only two strictly conserved residues, Asn¹⁸⁶ and Lys¹⁸⁸. No evidence was obtained to support the proposition that a binding site for PSII is located in the Arg¹⁸⁴-Ser¹⁹¹ sequence. Substitution of Lys¹⁸⁸, the sole strictly conserved charged residue in the sequence, with Gln had only a marginal effect on the reactivating ability of MSP. This is consistent with the results with spinach MSP, where the corresponding lysyl residue (Lys¹⁹⁰) was accessible to the water-soluble reagents even when the protein was bound to PSII membranes (19, 20). There is another charged residue, Arg¹⁸⁴, in the second protease-sensitive sequence, which is conserved in two other cyanobacteria and conservatively replaced by Lys in other photosynthetic organisms. Because the lysyl residue (Lys¹⁸⁶) of spinach MSP was modified with specific reagents only when the protein was free in solution (19, 20), a possibility remains that Arg¹⁸⁴ participates in the functional binding of Synechococcus MSP.

	ACKNOWLEDGEMENTS

We are grateful to Dr. T. Takakuwa, Nihon Bunko Co., for measurement and analysis of CD spectra. We thank Dr. Jian-Ren Shen of the Institute of Physical and Chemical Research (Riken) for reading the manuscript.

	FOOTNOTES

^* This work was performed under the management of the Research Association for Biotechnology as a part of the R&D Project of Basic Technology for Future Industries, supported by the New Energy and Industrial Technology Development Organization.The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

^§ To whom correspondence should be addressed: Biological Sciences Dept., Toray Research Center Inc., Kamakura 248-8555, Japan. Tel.: 81-467-32-9963; Fax: 81-467-32-0414; E-mail: akihiro_motoki@trc. toray.co.jp.

Published, JBC Papers in Press, January 23, 2002, DOI 10.1074/jbc.M100766200

	ABBREVIATIONS

The abbreviations used are: PSII, photosystem II; MSP, extrinsic 33-kDa manganese-stabilizing protein; EDC, 1-ethyl-3-(3-dimethylaminopropyl)-carbodiimide; MES, 2-(N-morpholino)ethanesulfonic acid.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

CiteULike

Complore

Connotea

Del.icio.us

Digg

R