Assessment of Structural and Functional Divergence Far from the Large Subunit Active Site of Ribulose-1,5-bisphosphate Carboxylase/Oxygenase
2003; Elsevier BV; Volume: 278; Issue: 49 Linguagem: Inglês
10.1074/jbc.m309993200
ISSN1083-351X
AutoresYuchun Du, Srinivasa R. Peddi, Robert J. Spreitzer,
Tópico(s)Porphyrin Metabolism and Disorders
ResumoDespite conservation of three-dimensional structure and active-site residues, ribulose-1,5-bisphosphate carboxylase/oxygenase (Rubisco, EC 4.1.1.39) enzymes from divergent species differ with respect to catalytic efficiency and CO2/O2 specificity. A deeper understanding of the structural basis for these differences may provide a rationale for engineering an improved enzyme, thereby leading to an increase in photosynthetic CO2 fixation and agricultural productivity. By comparing 500 active-site large subunit sequences from flowering plants with that of the green alga Chlamydomonas reinhardtii, a small number of residues were found to differ in regions previously shown by mutant screening to influence CO2/O2 specificity. When directed mutagenesis and chloroplast transformation were used to change Chlamydomonas Met-42 and Cys-53 to land plant Val-42 and Ala-53 in the large subunit N-terminal domain, little or no change in Rubisco catalytic properties was observed. However, changing Chlamydomonas methyl-Cys-256, Lys-258, and Ile-265 to land plant Phe-256, Arg-258, and Val-265 at the bottom of the α/β-barrel active site caused a 10% decrease in CO2/O2 specificity, largely due to an 85% decrease in carboxylation catalytic efficiency (Vmax/Km). Because land plant Rubisco enzymes have greater CO2/O2 specificity than the Chlamydomonas enzyme, this group of residues must be complemented by other residues that differ between Chlamydomonas and land plants. The Rubisco x-ray crystal structures indicate that these residues may reside in a variable loop of the nuclear-encoded small subunit, more than 20 Å away from the active site. Despite conservation of three-dimensional structure and active-site residues, ribulose-1,5-bisphosphate carboxylase/oxygenase (Rubisco, EC 4.1.1.39) enzymes from divergent species differ with respect to catalytic efficiency and CO2/O2 specificity. A deeper understanding of the structural basis for these differences may provide a rationale for engineering an improved enzyme, thereby leading to an increase in photosynthetic CO2 fixation and agricultural productivity. By comparing 500 active-site large subunit sequences from flowering plants with that of the green alga Chlamydomonas reinhardtii, a small number of residues were found to differ in regions previously shown by mutant screening to influence CO2/O2 specificity. When directed mutagenesis and chloroplast transformation were used to change Chlamydomonas Met-42 and Cys-53 to land plant Val-42 and Ala-53 in the large subunit N-terminal domain, little or no change in Rubisco catalytic properties was observed. However, changing Chlamydomonas methyl-Cys-256, Lys-258, and Ile-265 to land plant Phe-256, Arg-258, and Val-265 at the bottom of the α/β-barrel active site caused a 10% decrease in CO2/O2 specificity, largely due to an 85% decrease in carboxylation catalytic efficiency (Vmax/Km). Because land plant Rubisco enzymes have greater CO2/O2 specificity than the Chlamydomonas enzyme, this group of residues must be complemented by other residues that differ between Chlamydomonas and land plants. The Rubisco x-ray crystal structures indicate that these residues may reside in a variable loop of the nuclear-encoded small subunit, more than 20 Å away from the active site. The rbcL gene is one of the most sequenced genes in nature, and it is the most sequenced gene in the chloroplasts of eukaryotes. There are more than 2000 rbcL entries in the National Center for Biotechnology Information Entrez Proteins data base. Because rbcL encodes the catalytic large subunit of the ratelimiting photosynthetic enzyme ribulose-1,5-bisphosphate carboxylase/oxygenase (Rubisco, 1The abbreviations used are: Rubiscoribulose-1,5-bisphosphate carboxylase/oxygenaseBicineN,N-bis(2-hydroxyethyl)glycineCABP2-carboxy-d-arabinitol 1,5-bisphosphateRuBPribulose 1,5-bisphosphateΩCO2/O2 specificity factor. EC 4.1.1.39) (reviewed in Refs. 1.Spreitzer R.J. Salvucci M.E. Annu. Rev. Plant Biol. 2002; 53: 449-475Crossref PubMed Scopus (693) Google Scholar and 2.Andersson I. Taylor T.C. Arch. Biochem. Biophys. 2003; 414: 130-140Crossref PubMed Scopus (80) Google Scholar), it may be possible to exploit the vast sequence data base to gain a deeper understanding of the structure-function relationships of this enzyme. In particular, the values of catalytic efficiency and CO2/O2 specificity vary among Rubisco enzymes from divergent species (3.Jordan D.B. Ogren W.L. Nature. 1981; 291: 513-515Crossref Scopus (372) Google Scholar, 4.Read B.A. Tabita F.R. Arch. Biochem. Biophys. 1994; 312: 210-218Crossref PubMed Scopus (111) Google Scholar, 5.Uemura K. Miyachi Anwaruzzaman S. Yokota A. Biochem. Biophys. Res. Commun. 1997; 233: 568-571Crossref PubMed Scopus (137) Google Scholar), but residues directly involved in catalysis at the α/β-barrel active site are nearly 100% conserved. Thus, divergent residues, relatively far from the active site, must account for the differences in kinetic constants. The potential functional significance of these residues may be deduced by comparing the numerous Rubisco x-ray crystal structures that reside in the Protein Data Bank (6.Berman H.M. Westbrook J. Feng Z. Gilliland G. Bhat T.N. Weissig H. Shindyalov I.N. Bourne P.E. Nucleic Acids Res. 2000; 28: 235-242Crossref PubMed Scopus (27938) Google Scholar). Structures have been solved for Rubisco enzymes from evolutionarily distant prokaryotes, algae, and land plants (e.g. Refs. 7.Andersson I. Knight S. Schneider G. Lindqvist Y. Lundqvist T. Branden C.I. Lorimer G.H. Nature. 1989; 337: 229-234Crossref Scopus (187) Google Scholar, 8.Schreuder H.A. Knight S. Curmi P.M.G. Andersson I. Cascio D. Sweet R.M. Branden C.I. Eisenberg D. Protein Sci. 1993; 2: 1136-1146Crossref PubMed Scopus (48) Google Scholar, 9.Newman J. Gutteridge S. J. Biol. Chem. 1993; 268: 25876-25886Abstract Full Text PDF PubMed Google Scholar, 10.Andersson I. J. Mol. Biol. 1996; 259: 160-174Crossref PubMed Scopus (145) Google Scholar, 11.Sugawara H. Yamamoto H. Shibata N. Inoue T. Okada S. Miyake C. Yokota A. Kai Y. J. Biol. Chem. 1999; 274: 15655-15661Abstract Full Text Full Text PDF PubMed Scopus (56) Google Scholar, 12.Maeda N. Kitano K. Fukui T. Ezaki S. Atomi H. Miki K. Imanaka T. J. Mol. Biol. 1999; 293: 57-66Crossref PubMed Scopus (43) Google Scholar, 13.Taylor T.C. Backlund A. Bjorhall K. Spreitzer R.J. Andersson I. J. Biol. Chem. 2001; 276: 48159-48164Abstract Full Text Full Text PDF PubMed Scopus (82) Google Scholar). ribulose-1,5-bisphosphate carboxylase/oxygenase N,N-bis(2-hydroxyethyl)glycine 2-carboxy-d-arabinitol 1,5-bisphosphate ribulose 1,5-bisphosphate CO2/O2 specificity factor. Rubisco is a bifunctional enzyme that catalyzes either the carboxylation or oxygenation of RuBP, thereby initiating the Calvin cycle of photosynthesis or the photorespiratory pathway that leads to the loss of fixed carbon. The competition of CO2 and O2 at the rate-determining partial reaction (14.Pierce J. Andrews T.J. Lorimer G.H. J. Biol. Chem. 1986; 261: 10248-10256Abstract Full Text PDF PubMed Google Scholar) is defined by the CO2/O2 specificity factor (Ω = VcKo/VoKc, where Vc and Vo are the Vmax values for carboxylation and oxygenation, and Kc and Ko are the Km values for CO2 and O2, respectively) (15.Laing W.A. Ogren W.L. Hageman R.H. Plant Physiol. 1974; 54: 678-685Crossref PubMed Google Scholar, 16.Chen Z. Spreitzer R.J. Photosynth. Res. 1992; 31: 157-164Crossref PubMed Scopus (60) Google Scholar). However, net carboxylation is defined by the difference between the velocities of carboxylation and oxygenation (15.Laing W.A. Ogren W.L. Hageman R.H. Plant Physiol. 1974; 54: 678-685Crossref PubMed Google Scholar). If one understood the structural basis for differences in the catalytic properties of different Rubisco enzymes, it might be possible to design an improved enzyme as a means for increasing the agricultural production of food, fiber, and renewable energy (reviewed in Refs. 1.Spreitzer R.J. Salvucci M.E. Annu. Rev. Plant Biol. 2002; 53: 449-475Crossref PubMed Scopus (693) Google Scholar and 17.Spreitzer R.J. Photosynth. Res. 1999; 60: 29-42Crossref Scopus (95) Google Scholar). As in land plants, the Rubisco holoenzyme of the green alga Chlamydomonas reinhardtii is composed of eight ∼55-kDa large subunits (coded by the chloroplast rbcL gene) and eight ∼16-kDa small subunits (coded by a family of two nearly identical rbcS genes in the nucleus) (reviewed in Refs. 1.Spreitzer R.J. Salvucci M.E. Annu. Rev. Plant Biol. 2002; 53: 449-475Crossref PubMed Scopus (693) Google Scholar and 18.Spreitzer R.J. Rochaix J.D. Goldschmidt-Clermont M. Merchant S. The Molecular Biology of Chloroplasts and Mitochondria in Chlamydomonas. Kluwer Academic Publishers, Dordrecht, Netherlands1998: 515-527Google Scholar). However, in contrast to land plants, Chlamydomonas can survive in the absence of photosynthesis when provided with acetate as an alternative carbon source, and its rbcL and rbcS genes can both be eliminated or replaced with mutant gene copies via genetic transformation of the chloroplast and nucleus, respectively (18.Spreitzer R.J. Rochaix J.D. Goldschmidt-Clermont M. Merchant S. The Molecular Biology of Chloroplasts and Mitochondria in Chlamydomonas. Kluwer Academic Publishers, Dordrecht, Netherlands1998: 515-527Google Scholar, 19.Spreitzer R.J. Esquivel M.G. Du Y.C. McLaughlin P.D. Biochemistry. 2001; 40: 5615-5621Crossref PubMed Scopus (35) Google Scholar). Despite the fact that Chlamydomonas and land plant large subunits are ∼90% identical in amino acid sequence, Chlamydomonas Rubisco has an Ω value (Ω = 60) at least 20% lower than the Ω values of land plant enzymes (Ω = 80–100) (3.Jordan D.B. Ogren W.L. Nature. 1981; 291: 513-515Crossref Scopus (372) Google Scholar, 17.Spreitzer R.J. Photosynth. Res. 1999; 60: 29-42Crossref Scopus (95) Google Scholar). When compared with a representative collection of ∼500 flowering-plant large subunit sequences, there are only 34 residues in the Chlamydomonas large subunit (of a total of 475 residues) that differ from those characteristic of land plants (see supplemental table). Residues identical to those of Chlamydomonas at each of the 34 positions are found in less than 5% of the flowering plant sequences. Various groups of these residues are most likely responsible for the differences in kinetic properties between Chlamydomonas and land plant Rubisco enzymes (3.Jordan D.B. Ogren W.L. Nature. 1981; 291: 513-515Crossref Scopus (372) Google Scholar, 17.Spreitzer R.J. Photosynth. Res. 1999; 60: 29-42Crossref Scopus (95) Google Scholar). However, to change each of the residues to the predominant land plant residue, one at a time and in all possible combinations, would require the creation of ∼234 mutant enzymes. One way to further narrow the search for the structural determinants of differences in catalytic constants is to limit analysis to those regions of the large subunit that are already known to influence the rate-limiting step of catalysis. Mutant screening and selection in Chlamydomonas have previously identified four large subunit regions in which amino acid substitutions cause changes in Ω (reviewed in Ref. 18.Spreitzer R.J. Rochaix J.D. Goldschmidt-Clermont M. Merchant S. The Molecular Biology of Chloroplasts and Mitochondria in Chlamydomonas. Kluwer Academic Publishers, Dordrecht, Netherlands1998: 515-527Google Scholar). In the present study, the two large subunit regions farthest from the active site have been investigated by directed mutagenesis and chloroplast transformation. One of these regions was previously defined by pseudoreversion of a photosynthesis-deficient G54D null mutant (Fig. 1A). A G54V substitution in the large subunit N-terminal domain causes a 17% decrease in Ω (20.Spreitzer R.J. Thow G. Zhu G. Plant Physiol. 1995; 109: 681-685Crossref PubMed Scopus (22) Google Scholar). The other was defined by intragenic suppression in which either an A222T or V262L substitution in the large subunit (21.Hong S. Spreitzer R.J. J. Biol. Chem. 1997; 272: 11114-11117Abstract Full Text Full Text PDF PubMed Scopus (31) Google Scholar, 22.Du Y.C. Spreitzer R.J. J. Biol. Chem. 2000; 275: 19844-19847Abstract Full Text Full Text PDF PubMed Scopus (32) Google Scholar) or N54S or A57V substitution in the small subunit (23.Du Y.C. Hong S. Spreitzer R.J. Proc. Natl. Acad. Sci. U. S. A. 2000; 97: 14206-14211Crossref PubMed Scopus (29) Google Scholar) complements the low Ω and thermal instability of an L290F mutant enzyme (Fig. 1B) (24.Chen Z. Chastain C.J. Al-Abed S.R. Chollet R. Spreitzer R.J. Proc. Natl. Acad. Sci. U. S. A. 1988; 85: 4696-4699Crossref PubMed Scopus (78) Google Scholar, 25.Chen Z. Hong S. Spreitzer R.J. Plant Physiol. 1993; 101: 1189-1194Crossref PubMed Scopus (28) Google Scholar). At the restrictive temperature of 35 °C, the L290F mutant strain completely lacks photosynthesis due to the loss of Rubisco holoenzyme (24.Chen Z. Chastain C.J. Al-Abed S.R. Chollet R. Spreitzer R.J. Proc. Natl. Acad. Sci. U. S. A. 1988; 85: 4696-4699Crossref PubMed Scopus (78) Google Scholar). Large subunit residues 222, 262, and 290 reside in the C-terminal domain at the bottom of the α/β-barrel, whereas small subunit residues 54 and 57 reside in the loop between β-strands A and B that shields large subunit residues from solvent (Fig. 1B). All of these residues are between 16 and 30 Å away from the transition-state analog CABP bound in the active site (13.Taylor T.C. Backlund A. Bjorhall K. Spreitzer R.J. Andersson I. J. Biol. Chem. 2001; 276: 48159-48164Abstract Full Text Full Text PDF PubMed Scopus (82) Google Scholar). Changing two Chlamydomonas residues to those characteristic of land plants in the large subunit N-terminal domain had little effect on catalysis, but changing three at the bottom of the α/β-barrel domain caused a dramatic decrease in Ω. These latter residues must be complemented by other residues in land plant Rubisco that differ from those characteristic of Chlamydomonas Rubisco. Examination of the Chlamydomonas Rubisco x-ray crystal structure (13.Taylor T.C. Backlund A. Bjorhall K. Spreitzer R.J. Andersson I. J. Biol. Chem. 2001; 276: 48159-48164Abstract Full Text Full Text PDF PubMed Scopus (82) Google Scholar) indicates that these other residues may reside in or near the Rubisco small subunit. Strains and Culture Conditions—C. reinhardtii 2137 mt+ is the wild-type strain (26.Spreitzer R.J. Mets L. Plant Physiol. 1981; 67: 565-569Crossref PubMed Google Scholar). Photosynthesis-deficient, acetate-requiring mutants 18-7G mt+ and 25B1 mt+ were used as hosts for chloroplast transformation. Mutant 18-7G results from an rbcL UAG nonsense mutation that terminates large subunit translation after residue Thr-65 (27.Spreitzer R.J. Goldschmidt-Clermont M. Rahire M. Rochaix J.D. Proc. Natl. Acad. Sci. U. S. A. 1985; 82: 5460-5464Crossref PubMed Google Scholar). Mutant 25B1 was created by inserting 480 bp of yeast DNA at a PstI site in the 3′ coding region of rbcL (28.Newman S.M. Gillham N.W. Harris E.H. Johnson A.M. Boynton J.E. Mol. Gen. Genet. 1991; 230: 65-74Crossref PubMed Scopus (49) Google Scholar). Both mutant strains fail to accumulate Rubisco holoenzyme (27.Spreitzer R.J. Goldschmidt-Clermont M. Rahire M. Rochaix J.D. Proc. Natl. Acad. Sci. U. S. A. 1985; 82: 5460-5464Crossref PubMed Google Scholar, 28.Newman S.M. Gillham N.W. Harris E.H. Johnson A.M. Boynton J.E. Mol. Gen. Genet. 1991; 230: 65-74Crossref PubMed Scopus (49) Google Scholar). All strains were maintained at 25 °C in darkness with 10 mm acetate medium containing 1.5% Bacto-agar (26.Spreitzer R.J. Mets L. Plant Physiol. 1981; 67: 565-569Crossref PubMed Google Scholar). For biochemical analysis, cells were grown with 250–500 ml of liquid acetate medium at 25 °C on a rotary shaker (220 rpm) in darkness. Directed Mutagenesis and Chloroplast Transformation—Using a plasmid containing the Chlamydomonas rbcL gene (22.Du Y.C. Spreitzer R.J. J. Biol. Chem. 2000; 275: 19844-19847Abstract Full Text Full Text PDF PubMed Scopus (32) Google Scholar), directed mutagenesis was performed with synthetic oligonucleotides and a kit from Amersham Biosciences (29.Deng W.P. Nickoloff J.A. Anal. Biochem. 1992; 200: 81-88Crossref PubMed Scopus (1079) Google Scholar). Mutations were created to produce the single amino acid substitutions M42V (ATG → GTA), C53A (TGT → GCT), C256F (TGT → TTC), K258R (AAA → CGT), and I265V (ATT → GTA). Directed mutant plasmids were then used to create double mutant (M42V/C53A, C256F/K258R, C256F/I265V, K258R/I265V) and triple mutant (C256F/K258R/I265V) substitutions by making the same rbcL base changes as for the single mutant substitutions. The mutant plasmids were transformed into the chloroplast of mutant 18-7G mt+ (for changes at residues 42 and 53) or 25B1 mt+ (for changes at residues 256, 258, and 265) by using a helium-driven biolistic device (30.Finer J.J. Vain P. Jones M.W. McMullen M.D. Plant Cell Rep. 1992; 11: 323-328Crossref PubMed Scopus (409) Google Scholar, 31.Zhu G. Spreitzer R.J. J. Biol. Chem. 1994; 269: 3952-3956Abstract Full Text PDF PubMed Google Scholar). In all cases, homologous gene recombination gave rise to photosynthesis-competent colonies on minimal medium in the light (80 microeinsteins/m2/s). Colonies were cloned out and screened via DNA purification, PCR amplification, restriction enzyme analysis, and/or DNA sequencing (22.Du Y.C. Spreitzer R.J. J. Biol. Chem. 2000; 275: 19844-19847Abstract Full Text Full Text PDF PubMed Scopus (32) Google Scholar, 31.Zhu G. Spreitzer R.J. J. Biol. Chem. 1994; 269: 3952-3956Abstract Full Text PDF PubMed Google Scholar, 32.Zhu G. Spreitzer R.J. J. Biol. Chem. 1996; 271: 18494-18498Abstract Full Text Full Text PDF PubMed Scopus (40) Google Scholar) to ensure that the polyploid chloroplast genome was homoplasmic for the rbcL mutations (33.Spreitzer R.J. Chastain C.J. Curr. Genet. 1987; 11: 611-616Crossref Scopus (40) Google Scholar). The rbcL gene from each mutant strain was then PCR amplified and completely sequenced to confirm that only the expected mutations were present. Biochemical Analysis—Dark-grown cells were sonicated at 0 °C for 3 min in 50 mm Bicine (pH 8.0), 10 mm NaHCO3, 10 mm MgCl2, and 1 mm dithiothreitol. Cell debris was removed by centrifugation at 30,000 × g for 15 min, and the amount of protein in the supernatant (cell extract) was quantified (34.Bradford M.M. Anal. Biochem. 1976; 72: 248-254Crossref PubMed Scopus (217547) Google Scholar). Samples were prepared (35.Thow G. Zhu G. Spreitzer R.J. Biochemistry. 1994; 33: 5109-5114Crossref PubMed Scopus (33) Google Scholar) and subjected to SDS-polyacrylamide gel electrophoresis with a 7.5–15% polyacrylamide gradient in the running gel (36.Chua N.H. Methods Enzymol. 1980; 69: 434-446Crossref Scopus (331) Google Scholar). Proteins were transferred from the gel to nitrocellulose, probed with rabbit anti-tobacco Rubisco immunoglobulin G (0.5 μg/ml), and detected via enhanced chemiluminescence (Amersham Biosciences) (35.Thow G. Zhu G. Spreitzer R.J. Biochemistry. 1994; 33: 5109-5114Crossref PubMed Scopus (33) Google Scholar). Rubisco holoenzyme was purified from cell extract by sucrose gradient centrifugation in 50 mm Bicine (pH 8.0), 10 mm NaHCO3, 10 mm MgCl2, and 1 mm dithiothreitol (33.Spreitzer R.J. Chastain C.J. Curr. Genet. 1987; 11: 611-616Crossref Scopus (40) Google Scholar). RuBP carboxylase and oxygenase activities were measured via the incorporation of acid-stable 14C from NaH14CO3 (24.Chen Z. Chastain C.J. Al-Abed S.R. Chollet R. Spreitzer R.J. Proc. Natl. Acad. Sci. U. S. A. 1988; 85: 4696-4699Crossref PubMed Scopus (78) Google Scholar). The Ω value of purified and activated Rubisco (20 μg/reaction) was determined by assaying carboxylase and oxygenase activities simultaneously in 130 μm [1-3H]RuBP (9.7 Ci/mol) and 2 mm NaH14CO3 (5.0 Ci/mol) in 30-min reactions at 25 °C (37.Jordan D.B. Ogren W.L. Plant Physiol. 1981; 67: 237-245Crossref PubMed Google Scholar, 38.Spreitzer R.J. Jordan D.B. Ogren W.L. FEBS Lett. 1982; 148: 117-121Crossref Scopus (42) Google Scholar). [1-3H]RuBP and phosphoglycolate phosphatase were synthesized/purified by standard methods (37.Jordan D.B. Ogren W.L. Plant Physiol. 1981; 67: 237-245Crossref PubMed Google Scholar, 39.Kuehn G.D. Hsu T.C. Biochem. J. 1978; 175: 909-912Crossref PubMed Scopus (39) Google Scholar). Rubisco thermal stability was assayed by incubating purified enzymes (5 μg) in 0.5 ml of 50 mm Bicine (pH 8.0), 10 mm NaH14CO3 (2 Ci/mol), and 10 mm MgCl2 at various temperatures for 10 min (22.Du Y.C. Spreitzer R.J. J. Biol. Chem. 2000; 275: 19844-19847Abstract Full Text Full Text PDF PubMed Scopus (32) Google Scholar, 25.Chen Z. Hong S. Spreitzer R.J. Plant Physiol. 1993; 101: 1189-1194Crossref PubMed Scopus (28) Google Scholar). The samples were cooled on ice for 5 min, and carboxylase activity was initiated at 25 °C by adding 20 μl of 10 mm RuBP. Reactions were terminated after 1 min with 0.5 ml of 3 m formic acid in methanol. Mutagenesis, Transformation, and Recovery of Mutants—Based on the x-ray crystal structure of Chlamydomonas Rubisco (13.Taylor T.C. Backlund A. Bjorhall K. Spreitzer R.J. Andersson I. J. Biol. Chem. 2001; 276: 48159-48164Abstract Full Text Full Text PDF PubMed Scopus (82) Google Scholar), there are only two large subunit residues that are substantially different from those characteristic of land plants in the area surrounding Gly-54. Whereas Met-42 and Cys-53 are in van der Waals contact in the Chlamydomonas enzyme (Fig. 1A), land plants most often have Val-42 and Ala-53 (see supplemental table). In the area around Leu-290, there are three closely associated large subunit residues that differ from those of land plants (Fig. 1B). The Chlamydomonas large subunit contains methyl-Cys-256, Lys-258, and Ile-265, whereas land plants contain Phe-256, Arg-258, and Val-265 (see supplemental table). Methylation of Cys-256 is apparent in the Chlamydomonas crystal structure (13.Taylor T.C. Backlund A. Bjorhall K. Spreitzer R.J. Andersson I. J. Biol. Chem. 2001; 276: 48159-48164Abstract Full Text Full Text PDF PubMed Scopus (82) Google Scholar), but this posttranslational modification has not yet been observed in the crystal structures of Rubisco from any other species (7.Andersson I. Knight S. Schneider G. Lindqvist Y. Lundqvist T. Branden C.I. Lorimer G.H. Nature. 1989; 337: 229-234Crossref Scopus (187) Google Scholar, 8.Schreuder H.A. Knight S. Curmi P.M.G. Andersson I. Cascio D. Sweet R.M. Branden C.I. Eisenberg D. Protein Sci. 1993; 2: 1136-1146Crossref PubMed Scopus (48) Google Scholar, 9.Newman J. Gutteridge S. J. Biol. Chem. 1993; 268: 25876-25886Abstract Full Text PDF PubMed Google Scholar, 10.Andersson I. J. Mol. Biol. 1996; 259: 160-174Crossref PubMed Scopus (145) Google Scholar, 11.Sugawara H. Yamamoto H. Shibata N. Inoue T. Okada S. Miyake C. Yokota A. Kai Y. J. Biol. Chem. 1999; 274: 15655-15661Abstract Full Text Full Text PDF PubMed Scopus (56) Google Scholar, 12.Maeda N. Kitano K. Fukui T. Ezaki S. Atomi H. Miki K. Imanaka T. J. Mol. Biol. 1999; 293: 57-66Crossref PubMed Scopus (43) Google Scholar). Mutations were created to change the Chlamydomonas large subunit residues (singularly and in groups) to those characteristic of land plants. When the mutant rbcL genes were transformed into the Rubisco-deficient host strains, photosynthesiscompetent transformants were recovered on minimal medium in the light at frequencies comparable with that observed with wild-type rbcL (22.Du Y.C. Spreitzer R.J. J. Biol. Chem. 2000; 275: 19844-19847Abstract Full Text Full Text PDF PubMed Scopus (32) Google Scholar, 32.Zhu G. Spreitzer R.J. J. Biol. Chem. 1996; 271: 18494-18498Abstract Full Text Full Text PDF PubMed Scopus (40) Google Scholar). Except for the K258R/I265V double and C256F/K258R/I265V triple mutants, which grew moderately slower than wild type on minimal medium in the light at 25 °C, all of the mutant strains had a wild-type phenotype when growth was compared in spot tests (26.Spreitzer R.J. Mets L. Plant Physiol. 1981; 67: 565-569Crossref PubMed Google Scholar) on minimal and acetate medium (in light or darkness). When equal amounts of total soluble proteins were subjected to Western blotting, the N-terminal domain single (M42V and C53A) and double (M42V/C53A) mutants had levels of Rubisco holoenzyme similar to that of wild type (data not shown). With respect to the bottom of the α/β-barrel, only the K258R/I265V double and C256F/K258R/I265V triple mutants had somewhat less Rubisco holoenzyme than wild type (Fig. 2). Because the L290F substitution causes a decrease in holoenzyme thermal stability, and its suppressor substitutions (large subunit A222T or V262L, or small subunit N54S or A57V) improve Rubisco thermal stability (21.Hong S. Spreitzer R.J. J. Biol. Chem. 1997; 272: 11114-11117Abstract Full Text Full Text PDF PubMed Scopus (31) Google Scholar, 22.Du Y.C. Spreitzer R.J. J. Biol. Chem. 2000; 275: 19844-19847Abstract Full Text Full Text PDF PubMed Scopus (32) Google Scholar, 23.Du Y.C. Hong S. Spreitzer R.J. Proc. Natl. Acad. Sci. U. S. A. 2000; 97: 14206-14211Crossref PubMed Scopus (29) Google Scholar), it seemed possible that the K258R/I265V double and C256F/K258R/I265V triple substitutions in this region might also affect holoenzyme thermal stability. However, the triple mutant substitutions (Fig. 3), or any of the other substitutions (data not shown), had no effect on the thermal stability of purified holoenzyme analyzed in vitro. The structures of the K258R/I265V double and C256F/K258R/I265V triple mutant enzymes may be altered in ways that make them somewhat more susceptible to proteolysis in vivo, thereby causing a small decrease in the amount of holoenzyme (Fig. 2). Nonetheless, substantial amounts of stable Rubisco could be purified from all the mutant strains.Fig. 3Thermal inactivation of purified Rubisco from wild type (○) and the triple large subunit mutant C256F/K258R/I265V (•). Rubisco was incubated at each temperature for 10 min, cooled on ice, and assayed for RuBP carboxylase activity at 25 °C. Activities were normalized to the specific activities measured after the 35 °C incubation (wild type, 1.1 μmol/min/mg; mutant C256F/K258R/I265V, 0.2 μmol/min/mg).View Large Image Figure ViewerDownload Hi-res image Download (PPT) Catalytic Efficiency of the N-terminal Domain Mutant Enzymes—The N-terminal domain single (M42V and C53A) and double (M42V/C53A) mutant enzymes were first analyzed with respect to carboxylase activity at limiting CO2 (0.98 mm NaHCO3 in the absence of O2) and the inhibition of this activity by O2. The ratio (R) of these two activities is defined by Kc and Ko according to the relationship R = 1 + Kc[O2]/Ko(Kc + [CO2]) (24.Chen Z. Chastain C.J. Al-Abed S.R. Chollet R. Spreitzer R.J. Proc. Natl. Acad. Sci. U. S. A. 1988; 85: 4696-4699Crossref PubMed Scopus (78) Google Scholar). As shown in Table I, all of the mutant enzymes had a normal ratio of carboxylase activities, indicating that the Ko/Kc values are also normal. Only the M42V enzyme had a modest decrease in carboxylase-specific activity when assayed with saturating CO2 (12.4 mm NaHCO3 in the absence of O2), which appeared to be compensated for by the C53A substitution in the M42V/C53A double mutant enzyme (Table I). The M42V/C53A double mutant enzyme must also have a slightly greater level of oxygenase activity to account for its small but significant decrease in Ω (Table I). However, because these changes are all quite small, it is difficult to propose that the phylogenetic differences in the hydrophobic core of the large subunit N-terminal domain (Fig. 1A) make a major contribution to the differences in catalytic constants between Chlamydomonas and land plant Rubisco.Table IKinetic properties and oxygen inhibition of Rubisco purified from wild type, large subunit single mutants M42V and C53A, and double mutant M42V/C53AEnzymesΩaValues are the means ± S.D. (n — 1) of three separate enzyme preparationsRuBP carboxylase activitybValues are the simple average of two enzyme preparations assayed in duplicateRatio (A/B)100% N2 12.4 mm NaHCO3100% N2 0.98 mm NaHCO3 (A)100% O2 0.98 mm NaHCO3 (B)VcKo/VoKcμmol CO2/h/mg proteinWild type60 ± 187259.62.6M42V59 ± 152186.92.6C53A59 ± 193269.62.7M42V/C53A57 ± 1892810.32.7a Values are the means ± S.D. (n — 1) of three separate enzyme preparationsb Values are the simple average of two enzyme preparations assayed in duplicate Open table in a new tab Catalytic Efficiency of the Bottom-of-the-barrel Mutant Enzymes—Mutant enzymes resulting from single substitutions at the bottom of the α/β-barrel were found to have small decreases in Vc and, with respect to the C256F mutant enzyme, a small increase in Kc (Table II). However, when combined to form the C256F/K258R/I265V triple mutant enzyme, these substitutions caused a 10% decrease in Ω, largely due to an 85% decrease in carboxylation catalytic efficiency (Vc/Kc) (Table II). The double mutant enzymes had greater decreases in Vc/Kc than those of the single mutant enzymes (compare Tables II and III), and the C256F/I265V and K258R/I265V double mutant enzymes had small decreases in Ω (Table III). Thus, as the Chlamydomonas enzyme was changed to be more like that of land plants, both its catalytic efficiency (of the single and double mutant enzymes) and then specificity (of the double and triple mutant enzymes) decreased. Because these effects on catalysis are substantial, although negative, one must conclude that the phylogenetic differences at the bottom of the α/β-barrel domain (Fig. 1B) may play a major role in determining the differences in catalytic constants between Chlamydomonas and land plant Rubisco.Table IIKinetic properties of Rubisco purified from wild type, large subunit single mutants C256F, K258R, and I265V, and triple mutant C256F/K258R/I265VEnzymesΩaValues are the means ± S.D. (n — 1) of three separate enzyme preparationsVcaValues are the means ± S.D. (n — 1) of three separate enzyme preparationsKcaValues are the means ± S.D. (n — 1) of three separate enzyme preparationsKoaValues are the means ± S.D. (n — 1) of three separate enzyme preparationsVobCalculated valuesVc/KcbCalculated valuesKo/KcbCalculated valuesVc/VobCalculated valuesVcKo/VoKcμmol/h/mgμm CO2μm O2μmol/h/mgWild type61 ± 1143 ± 928 ± 1508 ± 79435.1183.4C256F59 ± 191 ± 834 ± 2602 ± 186272.7183.3K258R58 ± 1107 ± 127 ± 1515 ± 23354.0193.0I265V61 ± 1122 ± 1529 ± 1643 ± 107444.2222.8C256F/K258R/I265V55 ± 143 ± 964 ± 11292 ± 105160.7202.7a Values are the means ± S.D. (n — 1) of three separate enzyme preparationsb Calculated values Open table in a new tab Table IIIKinetic properties of Rubisco purified from wild type and large subunit double mutants C256F/K258R, C256F/I265V, and K258R/I265VEnzymesΩaValues are the means ± S.D. (n — 1) of three separate enzyme preparationsVcaValues are the means ± S.D. (n — 1) of three separate enzyme preparationsKcaValues are the means ± S.D. (n — 1) of three separate enzyme preparationsKoaValues are the means ± S.D. (n — 1) of three separate enzyme preparationsVobCalculated valuesVc/KcbCalculated valuesKo/KcbCalculated valuesVc/VobCalculated valuesVcKo/VoKcμmol/h/mgμmO2μmO2μmol/h/mgWild type61 ± 1154 ± 826 ± 1535 ± 63525.9213.0C256F/K258R58 ± 265 ± 743 ± 3891 ± 149231.5212.8C256F/I265V57 ± 189 ± 536 ± 2771 ± 31332.5212.7K258R/I265V57 ± 152 ± 531 ± 1780 ± 13231.7252.3a Values are the means ± S.D. (n — 1) of three separate enzyme preparationsb Calculated values Open table in a new tab The loops between β-strands and α-helixes of the C-terminal α/β-barrel domain contribute the majority of catalytically essential residues to the active site, but several active-site residues also reside in the N-terminal domain of an adjacent large subunit (reviewed in Refs. 1.Spreitzer R.J. Salvucci M.E. Annu. Rev. Plant Biol. 2002; 53: 449-475Crossref PubMed Scopus (693) Google Scholar and 2.Andersson I. Taylor T.C. Arch. Biochem. Biophys. 2003; 414: 130-140Crossref PubMed Scopus (80) Google Scholar). In particular, Glu-60 at the end of α-helix B interacts with the C-2 carboxyl group of the transition-state analog CABP (10.Andersson I. J. Mol. Biol. 1996; 259: 160-174Crossref PubMed Scopus (145) Google Scholar, 13.Taylor T.C. Backlund A. Bjorhall K. Spreitzer R.J. Andersson I. J. Biol. Chem. 2001; 276: 48159-48164Abstract Full Text Full Text PDF PubMed Scopus (82) Google Scholar). Despite the fact that a G54V substitution in the same α-helix caused a 17% decrease in the Ω value of Chlamydomonas Rubisco (20.Spreitzer R.J. Thow G. Zhu G. Plant Physiol. 1995; 109: 681-685Crossref PubMed Scopus (22) Google Scholar), changing Met-42 and Cys-53 to the Val-42 and Ala-53 residues characteristic of land plants had only a minor effect on catalysis (Table I). Because the only other residues that differ between Chlamydomonas and land plants in this region are relatively far away and exposed to solvent (Fig. 1A, residues 86 and 105), it is difficult to conclude that the hydrophobic core of the N-terminal domain contributes substantially to the differences in catalytic properties between Chlamydomonas and land plant Rubisco enzymes (3.Jordan D.B. Ogren W.L. Nature. 1981; 291: 513-515Crossref Scopus (372) Google Scholar, 17.Spreitzer R.J. Photosynth. Res. 1999; 60: 29-42Crossref Scopus (95) Google Scholar). In contrast, changing methyl-Cys-256, Lys-258, and Ile-265 to the Phe-256, Arg-258, and Val-265 residues characteristic of land plants at the bottom of the α/β-barrel caused a significant decrease in Ω and catalytic efficiency (Table II). Considering that land plant Rubisco has an Ω value greater than that of Chlamydomonas Rubisco (3.Jordan D.B. Ogren W.L. Nature. 1981; 291: 513-515Crossref Scopus (372) Google Scholar, 17.Spreitzer R.J. Photosynth. Res. 1999; 60: 29-42Crossref Scopus (95) Google Scholar), these land plant residues must be complemented by other residues in the land plant holoenzyme, and these residues must be different from those of Chlamydomonas. There are only three other large subunit residues that differ between Chlamydomonas and land plants in the region surrounding residues 256, 258, and 265 (Fig 1A). Chlamydomonas has Val-221, Val-235, and Ile-282, but land plants have Cys-221, Ile-235, and His-282 (see supplemental table). Although these residues are relatively far from 256, 258, and 265, there may be structural relationships that mediate communication between them (10.Andersson I. J. Mol. Biol. 1996; 259: 160-174Crossref PubMed Scopus (145) Google Scholar, 13.Taylor T.C. Backlund A. Bjorhall K. Spreitzer R.J. Andersson I. J. Biol. Chem. 2001; 276: 48159-48164Abstract Full Text Full Text PDF PubMed Scopus (82) Google Scholar). For example, Leu-219 is in van der Waals contact with both 221 and 256, Leu-240 is in contact with 221 and 265, Tyr-283 is in contact with 258 and 282, and Cys-284 is in contact with 265 and 282. However, the most obvious differences between Chlamydomonas and land plant Rubisco enzymes in the region surrounding residues 256, 258, and 265 derive from interactions with the loop between β-strands A and B of the small subunit (Fig. 1B). Because Chlamydomonas, and green algae in general, has five more residues in the βA–βB loop than do land plants (reviewed in Ref. 40.Spreitzer R.J. Arch. Biochem. Biophys. 2003; 414: 141-149Crossref PubMed Scopus (163) Google Scholar), small subunit residues appear to displace large subunit residues characteristic of land plants in this region (Fig. 4). In Chlamydomonas, methyl-Cys-256 and Lys-258 are in van der Waals contact with Val-63 residues from different small subunits. Lys-258 is also in van der Waals contact with small subunit Ser-62 and forms a hydrogen bond with large subunit Asn-287 (Fig. 4A). In spinach, Phe-256 is in van der Waals contact with Arg-258 from a neighboring large subunit, as well as with His-56 in the small subunit βA–βB loop (Fig. 4B). Arg-258 forms an ionic bond with Glu-259 from a neighboring large subunit (Fig. 4B). Chlamydomonas and spinach small subunit residues in structurally similar locations are also in van der Waals contact with large subunit residue 258, but the identities of the interacting small subunit residues are different (Val-63 and Cys-65 in Chlamydomonas, Ser-58 and Pro-59 in spinach) (Fig. 4). It seems likely that changes in residues 256, 258, and 265 would alter the interactions between the large subunit and small subunit βA–βB loop, and such alterations may account for the decrease in catalytic efficiency and specificity of the triple mutant enzyme. One generally considers a neutral theory of evolution to account for the majority of amino acid differences in the sequences of phylogenetically divergent proteins (e.g. Ref. 41.Kimura M. The Neutral Theory of Molecular Evolution. Cambridge University Press, Cambridge1983Crossref Google Scholar). In apparent agreement with this view of evolution, single amino acid substitutions had little effect on Rubisco function (Table I) despite presumed substantial alterations in residue interactions at the small/large subunit interface (Fig. 4). The significance of the unique posttranslational modifications observed in Chlamydomonas Rubisco (13.Taylor T.C. Backlund A. Bjorhall K. Spreitzer R.J. Andersson I. J. Biol. Chem. 2001; 276: 48159-48164Abstract Full Text Full Text PDF PubMed Scopus (82) Google Scholar) are not yet understood, but the substitution of methyl-Cys-256 with Phe also had no substantial effect on Rubisco function (Table I). However, when double (Table III) and triple (Table II) substitutions were created, significant deleterious changes were observed in catalytic efficiency and specificity. When considered as a group, the differences in the identities of these residues between species are, in fact, not neutral. Although a number of possible mechanisms have been proposed by which differences in the small subunit may influence large subunit catalysis (reviewed in Ref. 40.Spreitzer R.J. Arch. Biochem. Biophys. 2003; 414: 141-149Crossref PubMed Scopus (163) Google Scholar), one must assume that a cascade of structural alterations ultimately influences the position of active-site residues. Whether such differences can be observed by x-ray crystallography or deduced from the effects of additional mutants or suppressors remains to be determined. Nonetheless, because the C256F/K258R/I265V triple mutant enzyme has significant alterations in kinetics, this engineered region, along with some additional, complementing changes, may convert the catalytic properties of the Chlamydomonas enzyme to those of land plant enzymes. Perhaps substituting a few additional large subunit residues or the entire small subunit βA–βB loop (Fig. 1B), which may be possible in Chlamydomonas (18.Spreitzer R.J. Rochaix J.D. Goldschmidt-Clermont M. Merchant S. The Molecular Biology of Chloroplasts and Mitochondria in Chlamydomonas. Kluwer Academic Publishers, Dordrecht, Netherlands1998: 515-527Google Scholar), will achieve this goal, thereby identifying a region of focus for the design of a better enzyme. We thank Vijay Chandrasekaran for assistance in the preparation of Figs. 1 and 4 and Sriram Satagopan for help with phylogenetic analysis. Download .pdf (.02 MB) Help with pdf files
Referência(s)