Crystal structure of phosphoserine aminotransferase from Escherichia coli at 2.3 A resolution: comparison of the unligated enzyme and a complex with alpha-methyl-l-glutamate

Phosphoserine aminotransferase (PSAT; EC 2.6.1.52), a member of subgroup IV of the aminotransferases, catalyses the conversion of 3-phosphohydroxypyruvate to l-phosphoserine. The crystal structure of PSAT from Escherichia coli has been solved in space group P212121 using MIRAS phases in combination with density modification and was refined to an R-factor of 17.5% (Rfree=20.1 %) at 2.3 A resolution. In addition, the structure of PSAT in complex with alpha-methyl-l-glutamate (AMG) has been refined to an R-factor of 18.5% (Rfree=25.1%) at 2.8 A resolution. Each subunit (361 residues) of the PSAT homodimer is composed of a large pyridoxal-5'-phosphate binding domain (residues 16-268), consisting of a seven-stranded mainly parallel beta-sheet, two additional beta-strands and seven alpha-helices, and a small C-terminal domain, which incorporates a five-stranded beta-sheet and two alpha-helices. A three-dimensional structural comparison to four other vitamin B6-dependent enzymes reveals that three alpha-helices of the large domain, as well as an N-terminal domain (subgroup II) or subdomain (subgroup I) are absent in PSAT. Its only 15 N-terminal residues form a single beta-strand, which participates in the beta-sheet of the C-terminal domain. The cofactor is bound through an aldimine linkage to Lys198 in the active site. In the PSAT-AMG complex Ser9 and Arg335 bind the AMG alpha-carboxylate group while His41, Arg42 and His328 are involved in binding the AMG side-chain. Arg77 binds the AMG side-chain indirectly through a solvent molecule and is expected to position itself during catalysis between the PLP phosphate group and the substrate side-chain. Comparison of the active sites of PSAT and aspartate aminotransferase suggests a similar catalytic mechanism, except for the transaldimination step, since in PSAT the Schiff base is protonated. Correlation of the PSAT crystal structure to a published profile sequence analysis of all subgroup IV members allows active site modelling of nifs and the proposal of a likely molecular reaction mechanism.     

Purification and characterization of Thermotoga maritima endonuclease IV, a thermostable apurinic/apyrimidinic endonuclease and 3'-repair diesterase

An endonuclease IV homolog was identified as the product of a conceptual open reading frame in the genome of the hyperthermophilic bacterium Thermotoga maritima. The T. maritima endonuclease IV gene encodes a 287-amino-acid protein with 32% sequence identity to Escherichia coli endonuclease IV. The gene was cloned, and the expressed protein was purified and shown to have enzymatic activities that are characteristic of the endonuclease IV family of DNA repair enzymes, including apurinic/apyrimidinic endonuclease activity and repair activities on 3'-phosphates, 3'-phosphoglycolates, and 3'-trans-4-hydroxy-2-pentenal-5-phosphates. The T. maritima enzyme exhibits enzyme activity at both low and high temperatures. Circular dichroism spectroscopy indicates that T. maritima endonuclease IV has secondary structure similar to that of E. coli endonuclease IV and that the T. maritima endonuclease IV structure is more stable than E. coli endonuclease IV by almost 20 degrees C, beginning to rapidly denature only at temperatures approaching 90 degrees C. The presence of this enzyme, which is part of the DNA base excision repair pathway, suggests that thermophiles use a mechanism similar to that used by mesophiles to deal with the large number of abasic sites that arise in their chromosomes due to the increased rates of DNA damage at elevated temperatures.     

Two bacterial collagenolytic serine proteases have different topological specificities

From among 2000 soil isolates, we purified a secreted serine protease from Streptomyces omiyaensis (SOT), which has extremely high gelatinolytic activity. Using sequence analysis, the primary structure of SOT showed 77% identity with that of S. griseus trypsin (SGT). We constructed recombinants SOT and SGT using S. lividans. They indicated similar properties on optimum pH and temperature, thermostability, and substrate preference using fluorescence energy transfer combinatorial libraries. SOT greatly hydrolyzed both type I and type IV collagens, but SGT has poor ability to hydrolyze type IV collagen. Furthermore, SOT exhibits higher hydrolytic activities toward other protein substrate such as gelatin and casein than SGT. These results suggest that these two enzymes have different topological specificities in spite of their similar primary structures. We also constructed chimeras between SOT and SGT to investigate which domain is associated with differences in their substrate specificity. In comparison to substrate specificities of chimeras, we found that the N-terminal domain contributes to the determination of topological specificity.     

Sheep cytosolic branched-chain amino acid aminotransferase: cDNA cloning, primary structure and molecular modelling and its unique expression in muscles

This paper presents the cloning and the molecular modelling of the cytosolic branched-chain amino acid aminotransferase (BCATc) from sheep brain. The sheep BCATc cDNA (3 kb) encodes a mature polypeptide of 385 amino acids with a calculated molecular mass of 43072.93 Da. The sequence of the sheep BCATc cDNA is more similar to other mammalian BCATc cDNAs (53-87% identical) than to the sheep mitochondrial branched-chain amino acid aminotransferase (52%). Sheep BCATc belongs to the IV Folding class of pyridoxal-5'-phosphate-depending enzymes. Based on the known structure of the branched-chain amino acid aminotransferase (BCAT) from Escherichia coli, a molecular model of sheep BCATc (amino acid residues 62-385) was built. This is the first three-dimensional model of any mammalian BCAT. It suggests that the enzymatic mechanism of sheep BCATc and likely of all mammalian BCAT is very similar to the mechanism of the E. coli BCAT and confirms the hypotheses regarding to the substrate binding sites of E. coli BCAT. Sheep skeletal muscle, which is the main in vivo site for transamination of branched-chain amino acids, exhibits an unique expression of BCATc.     

The Rapid L- and D-Aspartate Uptake in Cultured Astrocytes

Astrocytes in primary culture possess a rapid L-aspartate saturable transport system (K_m = 93 M; V_max = 81 nmol/min/mg protein), which shows certain stereospecificity since V_max was 36% lower for D-aspartate uptake. These are values obtained at short incubation time (15 seconds), to obtain approximate initial rate conditions. Metabolic energy inhibitors, rotenone and iodoacetate very potently inhibited the L- and D-aspartate uptake processes, indicating that the transport process is an active one. However, the accumulation of L-aspartate was "enhanced by inhibitors of L-aspartate metabolism, such as the aspartate aminotransferase inhibitor, aminooxyacetate and L-methionine sulfoximine, an inhibitor of glutamine synthetase, whereas D-aspartate (a non-metabolizable analog of L-aspartate) uptake was not affected. The accumulated levels of L-aspartate in the presence of aminooxyacetate were similar to the levels of D-aspartate. These effects of L-aspartate metabolic inhibitors, suggest that due to metabolism of the the L-aspartate, short incubation time (eg., 15 seconds) is necessary to measure the initial rate of L-aspartate uptake, in order to obtain the "true kinetic parameters.     

The primary structure of the variable region of an immunoglobin IV light-chain amyloid-fibril protein (AL GIL).

The primary structure of the variable region of an amyloid-fibril protein GIL of immunoglobulin lambda-light-chain origin (AL) was determined. The AL protein obtained from the fibrils in the spleen of a 54-year-old man with primary systemic amyloidosis could be assigned to subgroup IV of the lambda variable-region sequence. About 50% of the protein was found to be truncated in the N-terminus and lacked the first six amino acid residues. The polypeptides consisted of about 146 amino acid residues and contained traces of carbohydrate. An acceptor site for N-glycosylation was found in positions 90-93, but no glycopeptide could be isolated. Comparison of the amino acid sequence of AL protein GIL with that of the only Bence-Jones protein of subgroup IV previously studied revealed a sequence homology of 89%. A similar comparison made with other AL proteins gave sequence homologies below 66%.     

Screw-sense inversion characteristic of alpha-helical poly(beta-p-chlorobenzyl L-aspartate) and comparison with other related polyaspartates

This is one of a series of studies on the reversal of the helix sense of polyaspartates originated from the pioneering work of Goodman and his associates in 1960s. Poly(beta-p-chlorobenzyl L-aspartate) (PClBLA) is one of the well-studied polyaspartate derivatives in both solution and the solid state. The chemical structure of PClBLA differs from those of poly(beta-benzyl L-aspartate) (PBLA) and poly(beta-phenethyl L-aspartate) (PPLA) only at the terminal of the relatively long side chain. PBLA takes a left-handed form (L) in conventional helicoidal solvents and does not exhibit any screw-sense inversion. In contrast to PBLA, both PClBLA and PPLA form a right-handed helix (R) in chlorinated alkane solvents and exhibits a reversal of alpha-helix sense at higher temperatures. Yet the transition behaviors in the presence of denaturant acid are quite different between these two polymers. While PPLA exhibits transitions such as R --> L --> coil by lowering temperature, PClBLA directly goes into the coil state without showing the reentrant L form. The cause of these phenomenological differences among these polymers has been investigated by constructing the phase diagram.     

Three-dimensional structure of kynureninase from Pseudomonas fluorescens

Kynureninase [E.C. 3.7.1.3] is a pyridoxal-5'-phosphate (PLP)-dependent enzyme that catalyzes the hydrolytic cleavage of l-kynurenine to anthranilic acid and l-alanine. Sequence alignment with other PLP-dependent enzymes indicated that kynureninase is in subgroup IVa of the aminotransferases, along with nifS, CsdB, and serine-pyruvate aminotransferase, which suggests that kynureninase has an aminotransferase fold. Crystals of Pseudomonas fluorescens kynureninase were obtained, and the structure was solved by molecular replacement using the CsdB coordinates combined with multiple isomorphous heavy atom replacement. The coordinates were deposited in the PDB (ID code 1QZ9). The structure, refined to an R factor of 15.5% to 1.85 A resolution, is dimeric and has the aminotransferase fold. The structure also confirms the prediction from sequence alignment that Lys-227 is the PLP-binding residue in P. fluorescens kynureninase. The conserved Asp-201, expected for an aminotransferase fold, is located near the PLP nitrogen, but Asp-132 is also strictly conserved and at a similar distance from the pyridinium nitrogen. Mutagenesis of both conserved aspartic acids shows that both contribute equally to PLP binding, but Asp-201 has a greater role in catalysis. The structure shows that Tyr-226 donates a hydrogen bond to the phosphate of PLP. Unusual among PLP-dependent enzymes, Trp-256, which is also strictly conserved in kynureninases from bacteria to humans, donates a hydrogen bond to the phosphate through the indole N1-hydrogen.     

Cloning, expression and characterization of a new aspartate aminotransferase from Bacillus subtilis B3

In the present study, we report the identification of a new gene from the Bacillus subtilis B3 strain (aatB3), which comprises 1308 bp encoding a 436 amino acid protein with a monomer molecular weight of 49.1 kDa. Phylogenetic analyses suggested that this enzyme is a member of the Ib subgroup of aspartate aminotransferases (AATs; EC 2.6.1.1), although it also has conserved active residues and thermostability characteristic of Ia-type AATs. The Asp232, Lys270 and Arg403 residues of AATB3 play a key role in transamination. The enzyme showed maximal activity at pH 8.0 and 45 °C, had relatively high activity over an alkaline pH range (pH 7.0-9.0) and was stable up to 50 °C. AATB3 catalyzed the transamination of five amino acids, with L-aspartate being the optimal substrate. The K(m) values were determined to be 6.7 mM for L-aspartate, 0.3 mM for α-ketoglutarate, 8.0 mM for L-glutamate and 0.6 mM for oxaloacetate. A 32-residue N-terminal amino acid sequence of this enzyme has 53% identity with that of Bacillus circulans AAT, although it is absent in all other AATs from different organisms. Further studies on AATB3 may confirm that it is potentially beneficial in basic research as well as various industrial applications.     

Rat cytosolic aspartate aminotransferase: molecular cloning of cDNA and expression in Escherichia coli

cDNA clones for rat cytosolic aspartate aminotransferase (cAspAT, L-aspartate:2-oxoglutarate aminotransferase) [EC 2.6.1.1] were isolated from a rat cDNA library, and the primary structure of the gene for cAspAT was deduced from its cDNA sequence. Rat cAspAT consists of 412 amino acids and its molecular weight is 46,295. The deduced amino acid sequence of rat cAspAT was compared with the sequences of AspATs from other species. The degree of sequence identities of rat/mouse cAspAT, rat/pig cAspAT, rat/chicken cAspAT, rat/pig mAspAT, and rat/Escherichia coli AspAT were 97.1, 89.6, 81.7, 48.1, and 41.2%, respectively. A coding region of rat cAspAT cDNA was inserted into E. coli expression vector pUC9, and enzymatically active cAspAT was expressed as a beta-galactosidase-cAspAT hybrid protein. This hybrid protein represented about 18% of the soluble proteins in E. coli and its kinetic properties were comparable with those of cAspAT preparations purified from rat liver.     

A multispecific quintet of aromatic aminotransferases that overlap different biochemical pathways in Pseudomonas aeruginosa

Pseudomonas aeruginosa possesses dual enzymatic sequences to both L-phenylalanine and L-tyrosine, a biosynthetic arrangement further complicated by the presence of five aromatic aminotransferases. Each aminotransferase is capable of transamination in vitro with any of the three keto acid intermediates in the aromatic pathway (phenylpyruvate, 4-hydroxyphenylpyruvate, or prephenate). The fractional contribution of these aminotransferases to particular transamination reactions in vivo can best be approached through the systematic and sequential elimination of individual aminotransferase activities by mutation. A program of sequential mutagenesis has produced two aminotransferase-deficient mutations. The first mutation imposed a phenotype of bradytrophy for L-phenylalanine (doubling time of 2.4 h in minimal salts/glucose medium compared to a 1.0-h doubling time for wild type). This mutant completely lacked an enzyme denoted aminotransferase AT-2. A genetic background of aminotransferase AT-2 deficiency was used to select for a second mutation which produced a phenotype of multiple auxotrophy for L-phenylalanine, L-aspartate, and L-glutamate. The double mutant completely lacked activity for aromatic aminotransferase AT-1 in addition to the missing aminotransferase AT-2. Enzymes AT-1 (Mr = 64,000) and AT-2 (Mr = 50,000) were readily separated from one another by gel filtration and were individually characterized for pH optima, freeze-thaw stability, heat lability, and molecular weight. The phenotypic and enzymological characterizations of the aminotransferase mutants strongly support the primary in vivo role of enzyme AT-2 in L-phenylalanine and L-tyrosine biosynthesis, while enzyme AT-1 must primarily be engaged in L-aspartate and L-glutamate synthesis. The substrate specificities and possible in vivo functions for AT-3, AT-4, and AT-5 are also considered.     

Three-dimensional structure of the elastase of Pseudomonas aeruginosa at 1.5-A resolution

Pseudomonas aeruginosa elastase (PAE) is a zinc metalloprotease with 301 amino acids. We have crystallized and solved the three-dimensional structure of PAE, using data to 1.5-A resolution, and have refined the native molecular structure to R = 0.188. The overall tertiary structure of the PAE molecule is similar to that of thermolysin, with which it shares 28% amino acid sequence identity. Nearly all of the active site residues that might potentially interact with substrates are identical in the two proteins. However, the active site cleft is significantly more "open" in PAE than in thermolysin.     

Identification and characterization of SppA, a novel light-inducible chloroplast protease complex associated with thylakoid membranes.

A new component of the chloroplast proteolytic machinery from Arabidopsis thaliana was identified as a SppA-type protease. The sequence of the mature protein, deduced from a full-length cDNA, displays 22% identity to the serine-type protease IV (SppA) from Escherichia coli and 27% identity to Synechocystis SppA1 (sll1703) but lacks the putative transmembrane spanning segments predicted from the E. coli sequence. The N-terminal sequence exhibits typical features of a cleavable chloroplast stroma-targeting sequence. The chloroplast localization of SppA was confirmed by in organello import experiments using an in vitro expression system and by immunodetection with antigen-specific antisera. Subfractionation of intact chloroplasts demonstrated that SppA is associated exclusively with thylakoid membranes, predominantly stroma lamellae, and is a part of some high molecular mass complex of about 270 kDa that exhibits proteolytic activity. Treatments with chaotropic salts and proteases showed that SppA is largely exposed to the stroma but that it behaves as an intrinsic membrane protein that may have an unusual monotopic arrangement in the thylakoids. We demonstrate that SppA is a light-inducible protease and discuss its possible involvement in the light-dependent degradation of antenna and photosystem II complexes that both involve serine-type proteases.     

Catalytic efficiency of subunits of the dimeric aminotransferase molecule in solution and in crystal

The catalytic efficiency of each of the two subunits of aspartate aminotransferase [EC2.6.1.1] from the chicken heart cytosol was determined for two physical states of the enzyme, in solution and in crystal. One subunit exhibits a higher catalytic efficiency in both states. The efficiency of the other is about 60% that of the first subunit for 2-oxoglutarate and 30–37% for L-aspartate.     

Novel features of prephenate aminotransferase from cell cultures of Nicotiana silvestris

A prephenate aminotransferase enzyme that produces L-arogenate was demonstrated in extracts from cultured-cell populations of Nicotiana silvestris. The enzyme was very active with low concentrations of prephenate, but required high concentrations of phenylpyruvate or 4-hydroxyphenylpyruvate to produce activity levels that were detectable. It is the most specific prephenate aminotransferase described to date from any source. Only L-glutamate and L-aspartate were effective amino-donor substrates. Prephenate concentrations greater than 1 mM produced substrate inhibition, an effect antagonized by increasing concentrations of L-glutamate cosubstrate. The enzyme was stable to storage for at least a month in the presence of pyridoxal 5'-phosphate, EDTA, and glycerol, and exhibited an unusually high temperature optimum of 70 degrees C. The identity of L-arogenate formed during catalysis was verified by high-performance liquid chromatography. DEAE-cellulose chromatography revealed two aromatic aminotransferase activities that were distinct from prephenate aminotransferase and which did not require the three protectants for stability. The aromatic aminotransferases were active with phenylpyruvate or 4-hydroxyphenylpyruvate as substrates, but not with prephenate. Both of the latter enzymes were similar in substrate specificity, and each exhibited a temperature optimum of 50 degrees C for catalysis. The primary in vivo function of the two aromatic aminotransferases is probably to transaminate between the aspartate/2-ketoglutarate and glutamate/oxaloacetate couples, since activities with the latter substrate combinations were an order of magnitude greater than with aromatic substrates. The demonstrated existence of a specific prephenate aminotransferase in N. silvestris meshes with other evidence supporting an important role for L-arogenate in tyrosine and phenylalanine biosynthesis in higher plants.     

Conversion of aspartate aminotransferase into an L-aspartate beta-decarboxylase by a triple active-site mutation.

The conjoint substitution of three active-site residues in aspartate aminotransferase (AspAT) of Escherichia coli (Y225R/R292K/R386A) increases the ratio of L-aspartate beta-decarboxylase activity to transaminase activity >25 million-fold. This result was achieved by combining an arginine shift mutation (Y225R/R386A) with a conservative substitution of a substrate-binding residue (R292K). In the wild-type enzyme, Arg(386) interacts with the alpha-carboxylate group of the substrate and is one of the four residues that are invariant in all aminotransferases; Tyr(225) is in its vicinity, forming a hydrogen bond with O-3' of the cofactor; and Arg(292) interacts with the distal carboxylate group of the substrate. In the triple-mutant enzyme, k(cat)' for beta-decarboxylation of L-aspartate was 0.08 s(-1), whereas k(cat)' for transamination was decreased to 0.01 s(-1). AspAT was thus converted into an L-aspartate beta-decarboxylase that catalyzes transamination as a side reaction. The major pathway of beta-decarboxylation directly produces L-alanine without intermediary formation of pyruvate. The various single- or double-mutant AspATs corresponding to the triple-mutant enzyme showed, with the exception of AspAT Y225R/R386A, no measurable or only very low beta-decarboxylase activity. The arginine shift mutation Y225R/R386A elicits beta-decarboxylase activity, whereas the R292K substitution suppresses transaminase activity. The reaction specificity of the triple-mutant enzyme is thus achieved in the same way as that of wild-type pyridoxal 5'-phosphate-dependent enzymes in general and possibly of many other enzymes, i.e. by accelerating the specific reaction and suppressing potential side reactions.     

The primary structure of mitochondrial aspartate aminotransferase from human heart

The complete amino acid sequence of the mitochondrial aspartate aminotransferase (L-aspartate:2-oxoglutarate aminotransferase, EC 2.6.1.1) from human heart has been determined based mainly on analysis of peptides obtained by digestion with trypsin and by chemical cleavage with cyanogen bromide. Comparison of the sequence with those of the isotopic isoenzymes from pig, rat and chicken showed 27, 29 and 55 differences, respectively, out of a total of 401 amino acid residues. Evidence for structural microheterogeneity at position 317 has also been obtained.     

Purification and cDNA-derived sequence of adenylosuccinate synthetase from Dictyostelium discoideum

Adenylosuccinate synthetase (IMP:L-aspartate ligase (GDP), EC 6.3.4.4) plays an important role in purine biosynthesis catalyzing the GTP-dependent conversion of IMP to AMP. The enzyme was purified from the cytosol of Dictyostelium discoideum using GTP-agarose chromatography as the critical step. It has an apparent molecular mass of 44 kDa. Monoclonal antibodies identified several forms of the enzyme with pI values between 8.1 and 9.0. Michaelis-Menten constants (Km) were low for the nucleotide substrates IMP (Km = 30 microM) and GTP (Km = 35 microM) as compared with the value for aspartic acid (Km = 440 microM). These values are in good agreement with constants reported from other organisms. Immunological studies indicated that the protein is predominantly localized in the cytosol and only partially associated with particulate fractions. The enzyme is present throughout the developmental cycle of D. discoideum. Using monoclonal antibodies, the gene was cloned from a lambda gt11 expression library. The complete sequence represents the first reported primary structure of an eucaryotic adenylosuccinate synthetase. Southern blots hybridized with a cDNA probe demonstrate that adenylosuccinate synthetase is encoded by a single gene and contains at least one intron. The deduced amino acid sequence shows 43% identity to adenylosuccinate synthetase from Escherichia coli. Homologous regions include short sequence motifs, such as the glycine-rich loop which is typical for GTP-binding proteins.     

Crystal structure of Trypanosoma cruzi tyrosine aminotransferase: substrate specificity is influenced by cofactor binding mode

The crystal structure of tyrosine aminotransferase (TAT) from the parasitic protozoan Trypanosoma cruzi, which belongs to the aminotransferase subfamily Igamma, has been determined at 2.5 A resolution with the R-value R = 15.1%. T. cruzi TAT shares less than 15% sequence identity with aminotransferases of subfamily Ialpha but shows only two larger topological differences to the aspartate aminotransferases (AspATs). First, TAT contains a loop protruding from the enzyme surface in the larger cofactor-binding domain, where the AspATs have a kinked alpha-helix. Second, in the smaller substrate-binding domain, TAT has a four-stranded antiparallel beta-sheet instead of the two-stranded beta-sheet in the AspATs. The position of the aromatic ring of the pyridoxal-5'-phosphate cofactor is very similar to the AspATs but the phosphate group, in contrast, is closer to the substrate-binding site with one of its oxygen atoms pointing toward the substrate. Differences in substrate specificities of T. cruzi TAT and subfamily Ialpha aminotransferases can be attributed by modeling of substrate complexes mainly to this different position of the cofactor-phosphate group. Absence of the arginine, which in the AspATs fixes the substrate side-chain carboxylate group by a salt bridge, contributes to the inability of T. cruzi TAT to transaminate acidic amino acids. The preference of TAT for tyrosine is probably related to the ability of Asn17 in TAT to form a hydrogen bond to the tyrosine side-chain hydroxyl group.     

Cloning and nucleotide sequence of regA, a putative response regulator gene of Rhodobacter sphaeroides

Abstract A 0.9 kb DNA fragment carrying the Rhodobacter capsulatus reg A gene, which encodes an oxygen-dependent, positively-acting response regulator of photosynthetic gene expression, was used as a probe in Southern hybridisation experiments to determine whether a similar gene occurs in R. sphaeroides . A strongly hybridising DNA fragment isolated from a R. sphaeroides plasmid gene bank was isolated, sequenced and found to contain an open reading frame which exhibits 75% identity with the R. capsulatus reg A gene. The deduced amino acid sequence of 184 residues shows 81% identity and 89% similarity with the R. capsulatus RegA protein, and significant similarities with other response regulators of the two component sensor-regulator type. Introduction of the R. sphaeroides gene into a R. capsulatus reg A mutant, which exhibits abnormally low levels of membrane-bound photosynthetic complexes, resulted in a 22–33-fold increase in these complexes to approximately 62–65% of wild-type levels. This is the first study to identify a putative response regulator in R. sphaeroides and to complement a regulatory mutation in R. capsulatus with a gene from another species. Further studies of associated genes may identify the different mechanisms by which the regulation of photosynthesis complex formation occurs in response to environmental stimuli in R. sphaeroides and R. capsulatus .