Isolation of lumenal proteins from spinach thylakoid membranes by triton X-114 phase partitioning.

The proteins present in the thylakoid lumen of higher plant chloroplasts have not been rigorously examined. In this communication we present a simple and rapid procedure for the isolation of the soluble proteins and extrinsic membrane proteins present in the thylakoid lumen from spinach. Our procedure involves extensive washing of the thylakoid membranes followed by Triton X-114 phase partitioning. When analyzed by one-dimensional polyacrylamide gel electrophoresis (PAGE), we obtain results which are very similar to those obtained by Kieselbach et al. using more classical methods [T. Kieselbach, A. Hagman, B. Andersson, W.P. Schroder, J. Biol. Chem. 273 (1998) 6710-6716]. About 25 major proteins are observed upon Coomassie blue staining. Upon two-dimensional isoelectric focusing-sodium dodecyl sulfate-PAGE and either Coomassie blue or silver staining, however, numerous other protein components are resolved. Our findings indicate that the total number of proteins (soluble and extrinsic membrane) present in the lumen may exceed 150.

Isolation of a highly active photosystem II preparation from Synechocystis 6803 using a histidine-tagged mutant of CP 47.

Site-directed mutagenesis was used to produce a Synechocystis mutant containing a histidine tag at the C terminus of the CP 47 protein of Photosystem II. This mutant cell line, designated HT-3, exhibited slightly above normal rates of oxygen evolution and appeared to accumulate somewhat more Photosystem II reaction centers than a control strain. A rapidly isolatable (<7 h) oxygen-evolving Photosystem II preparation was prepared from HT-3 using dodecyl-beta-d-maltoside solubilization and Co2+ metal affinity chromatography. This histidine-tagged Photosystem II preparation stably evolved oxygen at a high rate (2440 micromol O2 (mg chl)-1 h-1), exhibited an alpha-band absorption maximum at 674 nm, and was highly enriched in a number of Photosystem II components including cytochrome c550. Fluorescence yield analysis using water or hydroxylamine as an electron donor to the Photosystem II preparation indicated that virtually all of the Photosystem II reaction centers were capable of evolving oxygen. Proteins associated with Photosystem II were highly enriched in this preparation. 3,3',5, 5'-Tetramethylbenzidine staining indicated that the histidine-tagged preparation was enriched in cytochromes c550 and b559 and depleted of cytochrome f. This result was confirmed by optical difference spectroscopy. This histidine-tagged Photosystem II preparation may be very useful for the isolation of Photosystem II preparations from mutants containing lesions in other Photosystem II proteins.

Epitope mapping of the monoclonal antibody FAC2 on the apoprotein of Cpa-1 in photosystem II.

Using a combination of cyanogen bromide cleavage and endoproteinase digestion we have shown that the putative epitope for the monoclonal antibody FAC2 lies in the region 360Pro(-391)Ser on the apoprotein of CPa-1. This region lies entirely within the large extrinsic loop of this protein. We have shown previously that the epitope of FAC2 becomes exposed in oxygen-evolving membranes upon treatment with alkaline Tris which releases all four of the manganese associated with the oxygen-evolving site of photosystem II. The epitope is not exposed, however, after CaCl(2) treatment and exposure to low concentrations of chloride, conditions which lead to the release of two of the four manganeses associated with the oxygen-evolving site. These results suggest that, upon release of the chloride-insensitive manganese from photosystem II membranes, a conformational change occurs which leads to the exposure of 360Pro(-391)Ser on CPa-1 to the monoclonal antibody FAC2.

Interaction of CPa-1 with the manganese-stabilizing protein of photosystem II: identification of domains on CPa-1 which are shielded from N-hydroxysuccinimide biotinylation by the manganese-stabilizing protein.

The structural organization of photosystem II proteins has been investigated by use of the amino group-labeling reagent N-hydroxysuccinimidobiotin (NHS-biotin) and calcium chloride-washed photosystem II membranes. We have previously shown that the presence of the extrinsic, manganese-stabilizing protein on photosystem II membranes prevents the modification of lysyl residues located on the chlorophyll protein CPa-1 (CP-47) by NHS-biotin [Bricker, T. M., Odom, W. R., & Queirolo, C. B. (1988) FEBS Lett. 231, 111-117]. Upon removal of the manganese-stabilizing protein by calcium chloride-washing, CPa-1 can be specifically modified by treatment with NHS-biotin. Preparative quantities of biotinylated CPa-1 were subjected to chemical cleavage with cyanogen bromide. Two major biotinylated peptides were identified with apparent molecular masses of 11.8 and 15.7 kDa. N-terminal sequence analysis of these peptides indicated that the 11.8-kDa peptide was 232G-330M and that the 15.7-kDa peptide was 360P-508V. The 15.7-kDa CNBr peptide was subjected to limited tryptic digestion. The two smallest tryptic fragments identified migrated at apparent molecular masses of 9.1 (nonbiotinylated) and 7.5 kDa (biotinylated). N-terminal sequence analysis and examination of the predicted amino acid sequences of these peptides suggest that the 9.1-kDa fragment was 422R-508V and that the 7.5-kDa fragment was 360P-421A. These results strongly suggest that two NHS-biotinylated domains, 304K-321K and 389K-419K, become exposed on CPa-1 when the manganese-stabilizing protein is removed by CaCl2 treatment. Both of these domains lie in the large extrinsic loop E of CPa-1.

Use of a monoclonal antibody in structural investigations of the 49-kDa polypeptide of photosystem II.

A monoclonal antibody, FAC2, was isolated by immunization of mice with a Photosystem II core preparation followed by splenic fusion and standard monoclonal antibody screening and production techniques. This antibody recognizes the 49-kDa polypeptide of Photosystem II which is the apoprotein of CPal. The antigenic determinant recognized by this antibody lies on a cyanogen bromide fragment which appears as a doublet with an apparent molecular mass of 14.5 kDa. FAC2 was used to follow the effects of trypsin on the 49-kDa polypeptide in a membrane environment. Our results indicate that the extrinsic polypeptides of Photosystem II which are known to be involved in oxygen evolution protect the 49-kDa polypeptide from tryptic attack. Additionally, Photosystem II membranes which are treated with alkaline Tris exhibit a large increase in the ability to bind FAC2. This increase is not observed with membranes treated with calcium chloride or sodium chloride. These results indicate that the 49-kDa polypeptide may be at least structurally associated with the component(s) responsible for oxygen evolution.

Interaction of the 33-kDa extrinsic protein with photosystem II: identification of domains on the 33-kDa protein that are shielded from NHS-biotinylation by photosystem II.

The structural association of the spinach 33-kDa extrinsic protein of photosystem II with the membrane-bound components of the photosystem was investigated by labeling the 33-kDa extrinsic protein with the amino group-specific reagent N-hydroxysuccinimidobiotin both on NaCl-washed photosystem II membranes and free in solution. After quenching of the labeling reagent and isolation of the biotinylated molecules, the biotinylation sites were identified by Staphylococcus V8 protease digestion and analysis of the resultant peptide fragment mixture by matrix-assisted laser desorption/ionization mass spectrometry. When the 33-kDa extrinsic protein was modified on PS II membranes, three domains were biotinylated: 14K, 41K-76K, and 190K-236K. When the 33-kDa extrinsic protein was modified in solution, four additional domains were biotinylated: 1E-4K, 20K, 101K-105K, and 159K-186K. These additional modified domains reside in portions of the 33-kDa protein that are not accessible to the bulk solvent when the protein is associated with PS II and may define regions of interaction with the photosystem.

Carboxylate groups on the manganese-stabilizing protein are required for its efficient binding to photosystem II.

The effects of the modification of carboxylate groups on the manganese-stabilizing protein of photosystem II were investigated. Carboxylate groups (including possibly the C-terminus) on the manganese-stabilizing protein were modified with glycine methyl ester in a reaction facilitated by 1-ethyl-3-[3-(dimethylamino)propyl]carbodiimide. The manganese-stabilizing protein that was modified while associated with NaCl-washed photosystem II membranes contained 1-2 modified carboxylates, whereas the protein that was modified while free in solution contained 4 modified carboxylates. Both types of modified protein could reconstitute oxygen evolution at high manganese-stabilizing protein to photosystem II reaction center ratios. However, the protein that had been modified in solution exhibited a dramatically altered binding affinity for photosystem II. No such alteration in binding affinity was observed for the protein that had been modified while associated with the photosystem. Mapping of the sites of modification was carried out by trypsin and Staphylococcus V8 protease digestion of the modified proteins and analysis by matrix-assisted laser desorption/ionization mass spectrometry. These studies indicated that the domains (157)D-(168)D and (212)E-(247)Q (C-terminus) are labeled only when the manganese-stabilizing protein is modified in solution. Modified carboxylates in these domains are responsible for the altered binding affinity of this protein for the photosystem.

Alterations of the oxygen-evolving apparatus in a (448)Arg --> (448)S mutant in the CP47 protein of photosystem II under normal and low chloride conditions.

We have shown previously that a mutant which contained the alteration (448)R --> (448)S (R448S) in the CP47 protein of photosystem II exhibited a defect in its ability to grow and assemble functional photosystem II reaction centers under chloride-limiting conditions [Wu, J., Masri, N., Lee, W., Frankel, L. K., and Bricker, T. M. (1999) Plant Mol. Biol. 39, 381-386]. In this paper we have examined the function of the oxygen-evolving complex under chloride-sufficient (480 microM) and chloride-limiting (< 20 microM) conditions. When placed under chloride-limiting conditions, both the control strain K3 and R448S cells exhibit a loss of steady-state oxygen evolution, with t(1/2) of 16 and 17 min, respectively. Upon the addition of chloride, both recover their oxygen-evolving capacity relatively rapidly. However, R448S exhibits a much slower reactivation of oxygen evolution than does K3 (t(1/2) of 308 and 50 s, respectively). This may indicate a defect at the low-affinity, rapidly exchanging chloride-binding site [Lindberg, K., and Andréasson, L.-E. (1996) Biochemistry 35, 14259-14267]. Additionally, alterations in the distribution of S states and S-state lifetimes were observed. Under chloride-sufficient conditions, the R448S mutant exhibits a significant increase in the proportion of reaction centers in the S(3) state and a greatly increased lifetime of the S(3) state. Under chloride-limiting conditions, the proportion of reaction centers in both the S(2) and S(3) states increases significantly, and there is a marked increase in the lifetime of the S(2) state. These alterations are not observed in the control strain K3. Our observations support the hypothesis that (448)R of CP47 may participate in the formation of the binding domain for chloride in photosystem II and/or in the functional interaction with the 33 kDa protein with the photosystem.

Hydrodynamic studies on the manganese-stabilizing protein of photosystem II.

The solution conformation of the manganese-stabilizing protein of photosystem II was examined by analytical ultracentrifugation. Sedimentation velocity and sedimentation equilibrium studies were performed. These experiments yielded values for of 2.26 S with a diffusion constant, D, of 7.7 x 10(-)7 cm2 s-1. This s value is significantly lower than the apparent s value of 2.6 S previously reported [Miyao, M., and Murata, N. (1989) Biochim. Biophys. Acta 977, 315-321]. The molecular mass of the protein, 26.531 kDa, was verified by MALDI mass spectrometry. The diffusion coefficient was also determined by dynamic light scattering. The z-weighted average of D was 6.8 x 10(-)7 cm2 s-1. This result was somewhat lower than that observed by analytical ultracentrifugation due to the presence of slowly diffusing components in the sample. A two-component exponential fit of the dynamic light scattering data, however, gave D = 7.52 x 10(-)7 cm2 s-1 for the major component of the sample, which is in excellent agreement with the value determined by analytical ultracentrifugation. The value of s, the apparent sedimentation coefficient, was found to depend on the concentration of the protein and varied about 4% per milligram of protein. This is a feature of proteins which are asymmetric in solution. This asymmetry was examined using both the v-bar and Teller methods. Both methods indicated a significant degree of asymmetry for the manganese-stabilizing protein. Our findings indicate that the prolate ellipsoid model for the manganese-stabilizing protein is elongated in solution, with approximate dimensions of about 12.6 nm x 3.0 nm, yielding an axial ratio of 4.2.

Random mutagenesis in the large extrinsic loop E and transmembrane alpha-helix VI of the CP 47 protein of Photosystem II.

The intrinsic chlorophyll-protein CP 47 is a component of Photosystem II which functions in both light-harvesting and oxygen evolution. Using the Escherichia coli mutator strain XL-1 Red, we introduced mutations at 14 sites in the large extrinsic loop E of CP 47 and its adjacent transmembrane alpha-helix VI. Four mutant cell lines were recovered in which the histidyl residues 455H, 466H and 469H were altered. The cell lines H455T, H455Y, H469Y, and the double mutant F432L,H466R exhibited phenotypes that supported the identification of the histidyl residues 455H, 466H and 469H as chlorophyll ligands. Four additional mutant cell lines were recovered which contained mutations at positions 448R in the large extrinsic loop of CP 47. These mutants, R448K, R448Q, R448S, and R448W, exhibited variable phenotypes ranging from moderate alteration of photoautotrophic growth and oxygen evolution rates to a complete inhibition of these parameters. Those mutants exhibiting photoautotrophic growth and oxygen evolution capability under standard conditions were unable to grow photoautotrophically or evolve oxygen when grown at low chloride concentrations. Finally, a mutant cell line exhibiting a substitution at position 342G was recovered. The mutant G342D exhibited moderate alterations of photoautotrophic growth and oxygen evolution. In addition to these alterations, mutants were recovered in which deletions and insertions (leading to frame shifts) and stop codons were introduced. These mutants uniformly lacked the ability to either grow photoautotrophically or evolve oxygen.