Alteration of Substrate Chain-Length Specificity of ... - ACS Publications

Ken'ichiro Matsumoto , Chiaki Hori , Ryunosuke Fujii , Masahiro Takaya , Takashi .... T. Nomura, Tomoyo Tanaka, Zhihua Gan, Kazuhiro Kuwabara, Hideki ...
0 downloads 0 Views 117KB Size
Download PDF
Biomacromolecules 2004, 5, 480-485

480

Alteration of Substrate Chain-Length Specificity of Type II Synthase for Polyhydroxyalkanoate Biosynthesis by in Vitro Evolution: in Vivo and in Vitro Enzyme Assays Kazuma Takase,† Ken’ichiro Matsumoto,† Seiichi Taguchi,*,†,‡ and Yoshiharu Doi†,§ Polymer Chemistry Laboratory, RIKEN Institute, 2-1 Hirosawa, Wako-shi, Saitama, 351-0198, Japan, Division of Molecular Chemistry, Graduate School of Engineering, Hokkaido University, Sapporo 060-8628, Japan, and Department of Innovative and Engineered Materials, Tokyo Institute of Technology, 4259 Nagatsuta, Midori-ku, Yokohama 226-8502, Japan Received August 28, 2003; Revised Manuscript Received December 5, 2003

In our previous study, in vitro evolution of type II polyhydroxyalkanoate (PHA) synthase (PhaC1Ps) from Pseudomonas sp. 61-3 yielded eleven mutant enzymes capable of synthesizing homopolymer of (R)-3hydroxybutyrate [P(3HB)] in recombinant Escherichia coli JM109. These recombinant strains were capable of accumulating up to approximately 400-fold more P(3HB) than strains expressing the wild-type enzyme. These mutations enhanced the ability of the enzyme to specifically incorporate the 3HB-coenzyme A (3HBCoA) substrate or improved catalytic efficiency toward the various monomer substrates of C4 to C12 (R)3-hydroxyacyl-CoAs which can intrinsically be channeled by PhaC1Ps into P(3HB-co-3HA) copolymerization. In this study, beneficial amino acid substitutions of PhaC1Ps were analyzed based on the accumulation level and the monomer composition of P(3HB-co-3HA) copolymers generated by E. coli LS5218 [fadR601 atoC(Con)] harboring the monomer supplying enzyme genes. Substitutions of Ser by Thr(Cys) at position 325 were found to lead to an increase in the total amount of P(3HB-co-3HA) accmumulated, whereas 3HB fractions in the P(3HB-co-3HA) copolymer were enriched by substitutions of Gln by Lys(Arg, Met) at position 481. This strongly suggests that amino acid substitutions at positions 325 and 481 are responsible for synthase activity and/or substrate chain-length specificity of PhaC1Ps. These in vivo results were supported by the in vitro results obtained from synthase activity assays using representative single and double mutants and synthetic substrates, (R,S)-3HB-CoA and (R,S)-3-hydroxydecanoyl-CoA. Notably, the position 481 was found to be a determinant for substrate chain-length specificity of PhaC1Ps. Introduction Polyhydroxyalkanoates (PHAs) are natural, biodegradable, thermoplastic polyesters that may be used as replacements for petrochemical polymers with a wide variety of applications.1 Of all of the PHAs, poly(3-hydroxybutyrate) [P(3HB)] has been studied most extensively, mainly because it was the first to be discovered. After extraction of amorphous P(3HB) from the cell with organic solvents, it becomes highly crystalline and is a stiff but brittle material in this state.2 This physical property limits the ability the P(3HB) homopolymer to be processed. Most recently, cold-drawing and two-step drawing methods have successfully been adopted to overcome this problem by our group.3,4 Furthermore, the usefulness of incorporation of secondary monomer units into the 3HB-based copolymers has been demonstrated to improve the physical deficiencies of P(3HB).5 Hence, a combination of advanced methodology for biosynthesis of 3HB-based copolymers with cold-drawing * To whom correspondence should be addressed. Tel: +81-(0)11-7066610. Fax: +81-(0)11-706-6610. E-mail: [email protected]. † RIKEN Institute. ‡ Hokkaido University. § Tokyo Institute of Technology.

would be an attractive project to produce high performance PHA materials which are applicable for expanded practical uses. Pseudomonas sp. 61-3 PHA synthase PhaC1Ps is considered to be a useful target for enzyme evolution, because it can synthesize random 3HB-based copolymers that are more ductile, easier to mold, and tougher than the P(3HB) homopolymer. However, PhaC1Ps possesses extremely weak in vitro substrate specificity toward 3HB-CoA as compared to other longer monomer substrates [3-hydroxyhexanoyl-CoA (C6) to 3-hydroxydodecanoyl-CoA (C12)].29 In vitro evolution6,7 serves as a powerful methodology for approaching the above purpose. In fact, type I PHA synthases from Ralstonia eutropha and Aeromonas caViae (at present termed Aeromonas punctata) were successfully improved in terms of enhancement of PHA synthesis or expansion of variation in the monomer compositions of PHA copolymers,8-13 by use of our developed in vitro enzyme evolution strategies. In addition, drastically enhanced synthesis of P(3HB) in recombinant Escherichia coli has been achieved by artificial activity improvement of type II Pseudomonas sp. 61-3 synthase (PhaC1Ps) intrinsically capable of generating 3HB-based copolymers,14 based on the systematic evolution program including error-prone PCR mutagenesis,

10.1021/bm034323+ CCC: $27.50 © 2004 American Chemical Society Published on Web 02/14/2004

Substrate Specificity Altered PHA Synthase

Biomacromolecules, Vol. 5, No. 2, 2004 481

site-specific saturation mutagenesis, and recombination of beneficial mutations.15 In this study, to address the individual responsibilities of amino acid substitutions that positively contribute to enhancement of P(3HB) synthesis, in vivo and in vitro enzyme assays were carried out using evolved mutants of PhaC1Ps and synthetic substrates. Experimental Section Bacterial Strains and Culture Conditions. E. coli JM10916 was used for all standard genetic engineering. For accumulation of P(3HB-co-3HA) copolymers from dodecanoate, recombinant E. coli LS5218 [fadR601, atoC(Con)]17 strains were grown on M9 medium18 containing 0.3% sodium dodecanoate as a sole carbon source with 0.4% (vol./vol.) Brij-3519 for 72 h at 37 °C. Brij-35 is a detergent (formally termed polyoxyethylene dodecyl ether) for solubilizing dodecanoate. When needed, ampicillin (100 µg/mL) was added to the medium. DNA Manipulation and Plasmid Construction. Standard recombinant DNA manipulation18 was used for isolation of plasmid DNA. All of the restriction endonucleases and modification enzymes for genetic engineering were purchased from TaKaRa Shuzo Co., Ltd. (Kyoto, Japan) and used under conditions recommended by the supplier. All other chemicals were of analytical grade for biochemical use and were used without further purification. The plasmid vector pGEM′′C1ABJ4 was constructed for biosynthesis of P(3HB-co-3HA) in E. coli LS5218 strain, as illustrated in Figure 1. First, the plasmid vector pGEM′′ABex15 was digested by NdeI and then ligated with a synthetic NdeI-NheI-FbaI-MunI-NdeI linker (nucleotide sequences are also shown in Figure 1). The resultant vector was named pGEM′′ABexII. Next, pUCJ421 was digested with XbaI and EcoRI, and then the 0.5 kb of phaJ4Pa fragment was purified and ligated with NheI and MunI digested pGEM′′ABexII. The resultant plasmid was termed pGEM′′ABJ4ex. Finally, pETphaC1Ps15 was digested with XbaI and BamHI, and then the 1.7 kb of phaC1Ps fragment was purified and ligated with XbaI and BglII digested pGEM′′ABJ4ex. The resultant plasmid was termed pGEM′′C1ABJ4. DNA Sequencing. DNA sequencing for confirmation of new plasmid constructs was carried out by the dideoxy chain termination method with the Prism 377 DNA sequencer (Applied Biosystems) and the CEQ2000XL DNA Analysis System (Beckman Coulter) using the BigDye terminator cycle sequencing ready reaction kit (Applied Biosystems) and Dye Terminator Cycle Sequencing with Quick Start Kit (Beckman Coulter), respectively. Analysis of the Content and Composition of P(3HBco-3HA) Produced by Recombinant E. coli LS5218. E. coli LS5218 recombinants were cultivated on 100 mL M9 medium containing sodium dodecanoate (0.3wt %/vol.) and ampicillin (100 µg/mL) at 37 °C for 72 h.22 To determine the cellular P(3HB-co-3HA) content and composition, ca. 30 mg of dry cells was subjected to methanolysis with a solution consisting of 1.7 mL of methanol, 0.3 mL of 98% sulfuric acid, and 2.0 mL of chloroform at 100 °C for 140

Figure 1. Construction of the plasmid vector, pGEM′′C1ABJ4, used for synthesizing P(3HB-co-3HA) copolyester from dodecanoate in recombinant Escherichia coli LS5218 strain. PhbARe and PhbBRe encode β-ketothiolase and NADPH-dependent acetoacetyl-CoA reductase gene derived from Ralstonia eutropha, respectively; PRe and TRe denote the promoter and terminator regions of the PhbCABRe operon in R. eutropha, respectively; phaJ4Pa encode (R)-specific enoyl-CoA hydratase gene derived from Pseudomonas aeruginosa; phaC1Ps encode PHA synthase gene derived from Pseudomonas sp. 61-3.

min to convert the constituents to their methyl esters. The addition of 1.0 mL of water to the reaction mixture induced phase separation. The lower chloroform layer was used for gas chromatography (GC) analysis on a Shimadzu GC-17A system equipped with Neutra Bond-1 capillary column (30 m by 0.25 mm) and a flame ionization detector.23 In Vitro Enzyme Assay. The wild and recombinant strains of E. coli JM109 were inoculated into 100 mL of LB medium, containing 50 µg/mL ampicillin for recombinant strains, and cultivated for 12 h at 30 °C. Cells were harvested and resuspended in 2 mL of ice cooled 40 mM potassium phosphate buffer (pH 7.5). Subsequently, cells were disrupted by sonication (UD-200, TOMY) for 4 min. After broken cells were centrifuged (12,000 × g, 10 min, 4 °C), the supernatant was moved to a new microtube and diluted to 2 µg of protein/ µL by the same phosphate buffer. (R,S)-3-Hydroxybutyryl-CoA (3HB-CoA) was purchased from Sigma. The chemical synthesis of (R,S)-3-hydroxydecanoyl-CoA (3HD-CoA) was performed using a modified method of Schubert et al.24 Synthesized 3HD-CoA was

482

Biomacromolecules, Vol. 5, No. 2, 2004

purified by reverse phase column (µBondapak C18, Waters) using HPLC as described by Kraak et al.25 PHA synthase activities were determined by a modified spectroscopic assay based on the method described by Roo et al.26 Reaction mixtures contained 2 mM 3HB-CoA or 3HD-CoA and 1 µg/µL bovine serum albumin (BSA) in 40 mM phosphate buffer (pH 7.5). To start the reactions, 60 µL of this reaction mixture was added to 5 µL of crude extract, and the materials were incubated at 30 °C. To terminate the reaction, 60 µL of 1% (wt./vol.) trichloroacetic acid (TCA) was added at defined time points shown in Figure 6, parts A and B. After centrifugation (1,200 × g, 3 min, 4 °C), the remaining solid phase was removed by filtration with a cellulose acetate filter (pore size 0.45 µm, Whatman). 90 µL of the filtered reaction solution was transferred to a microtiter plate, and 10 µL 5,5′-dithio-bis(2-nitrobenzoic acid) (DTNB) (10 mM) was added. The absorbance was measured at 405 nm using a microplate reader (ARVO SX, Wallac). One unit of enzyme activity is defined as the amount required to catalyze the transformation of 1 µmol substrate in 1 min. The concentration of total cellular proteins was determined by the method of Bradford27 using Bio-Rad Protein Assay Kit (Hercules, CA) and BSA as the standard. Western Blotting. Rabbit antisera against PhaC1Ps was prepared by injection of a synthetic 17-mer oligo-peptide (CSGKLKKSPTSLGNKAY: near the COOH terminus of PhaC1Ps). Whole-cell extracts of the recombinant E. coli strains were prepared by sonication,8 and soluble fractions were obtained by centrifugation (18 000 × g, 10 min, 4 °C). The concentration of total cellular proteins was determined by using a Bio-Rad Protein Assay Kit (Hercules, CA) with BSA as the standard. 10 µg of each soluble protein was subjected to SDS-PAGE on 12.5% gel and electroblotted to a PVDF membrane using a Criterion Blotter (Bio-Rad). Western blotting was performed as described previously,28 and protein bands were visualized by using goat anti-rabbit IgG conjugated to alkaline phosphatase as the secondary antibody. Molecular Weight Analysis of the P(3HB-co-3HA) Copolymers. The copolymers accumulated in the cells were extracted with chloroform for 72 h and purified by reprecipitation with hexane. Molecular mass data of copolymers were obtained by gel permeation chromatography using a Shimadzu 10A system with a RID-10A refractive-index detector with serial columns of Shodex K802 and K806M. Polystyrene standards with low polydispersity were used to construct a calibration curve.33 Results Construction of Exogeneous PHA Biosynthetic Pathways in Escherichia coli LS5218. Previously, we identified beneficial amino acid substitutions in PhaC1Ps that positively contribute to enhanced P(3HB) accumulation in Escherichia coli JM109.15 Five single mutants [S325C(T) and Q481K(M, R)] and six double mutants [S325C(T)/Q481K(M, R)] were acquired as evolvants exhibiting higher P(3HB) contents, up to 38-fold (for single mutants) and 400-fold (for double mutants), than the wild-type enzyme. It was of great

Takase et al.

Figure 2. Proposed model of P(3HB-co-3HA) production in recombinant E. coli LS5218 from dodecanoate by two metabolic pathways. The monomers necessary for P(3HB-co-3HA) synthesis are provided as 3HB-CoA from acetyl-CoA via acetyl-CoA dimerization pathway consisting of PhbARe (β-ketothiolase) and PhbBRe (NADPH-dependent acetoacetyl-CoA reductase) from R. eutropha, and 3HA-CoAs from β-oxidation intermediates, trans-2-enoyl-CoAs, via (R)-specific hydration catalyzed by PhaJ4Pa [(R)-specific enoyl-CoA hydratase] from P. aeruginosa.

interest to address which acquired factor, either an increase in the whole PhaC1Ps synthase activity independent of the chain-length of C4-C12 monomer substrates or an increase in the activity specific for 3HB-CoA, was responsible for the significant enhancement of P(3HB) accumulation in E. coli JM109. To examine the changes in the activity and substrate chainlength specificity of the mutants of PhaC1Ps, the two defined metabolic pathways for PHA synthesis were established in E. coli LS5218, as illustrated in Figure 2. One is the acetylCoA dimerization pathway consisting of the 3HB-CoA monomer supplying enzymes, PhbARe (β-ketothiolase) and PhbBRe (NADPH-dependent acetoacetyl-CoA reductase), with PhaC1Ps. The other is the combination of PhaC1Ps with the fatty acid β-oxidation pathway which can channel 3HACoA monomer units (C4-C12) into PHA copolymer, P(3HBco-3HA), via mediation of (R)-specific enoyl-CoA hydratase (PhaJ4Pa) from P. aeruginosa.21 The plasmid pGEM′′C1ABJ4 harboring these pathway genes (Figure 1) was used to assay the ability of the PhaC1Ps mutants to accumulate PHA. In Vivo Assay of PhaC1Ps Mutants. Twelve recombinant E. coli LS5218 strains with pGEM′′C1ABJ4 carrying wildtype or each mutant phaC1Ps were cultivated at 37 °C for 72 h in M9 medium containing sodium dodecanoate, and then subjected to GC analysis. The P(3HB-co-3HA) contents of E. coli recombinants are presented in Figure 3, and all values are averages in triplicate tests. The recombinant harboring wild-type phaC1Ps accumulated 13 wt % of P(3HB-co-3HA) in dry cells. With the exception of the Q481R mutant, most of the recombinants harboring individual single mutant PhaC1Ps enzymes exhibited comparable or higher P(3HB-co-3HA) contents (13 to 22 wt %), compared to the recombinant harboring the wild-type phaC1Ps. Between the two single mutants substituted at position 325, the S325T mutant accumulated higher content

Substrate Specificity Altered PHA Synthase

Figure 3. Cellular P(3HB-co-3HA) contents of recombinant E. coli LS5218 strains harboring single or double mutants of the Pseudomonas sp. 61-3 PHA synthase. All cells were cultivated on M9 medium containing 0.3% (wt./vol.) sodium dodecanoate for 72 h at 37 °C. The cellular P(3HB-co-3HA) content and composition were determined by gas chromatography after methanolysis of lyophilized cells in the presence of 15% sulfuric acid. All values are averages in triplicate tests. SC/QK, S325C/Q481K double mutant; SC/QM, S325C/Q481M double mutant; SC/QR, S325C/Q481R double mutant; ST/QK, S325T/ Q481K double mutant; ST/QM, S325T/Q481M double mutant; ST/ QR, S325T/Q481R double mutant.

of P(3HB-co-3HA) (22 wt %) than the S325C mutant, and the Q481M mutant exhibited the highest content (20 wt %) of P(3HB-co-3HA) among the Gln481 mutants. These results were in agreement with those obtained for P(3HB) accumulation in E. coli JM109.15 Moreover, as expected, all of the Ser325/Gln481 double mutants exhibited higher P(3HB-co-3HA) contents (28 to 35 wt %) compared to the parental single mutants (7 to 22 wt %). This result showed that a combination of the Ser325 mutation with the Gln481 mutation is cooperatively functional for much higher P(3HB-co-3HA) accumulation via the β-oxidation and acetyl-CoA dimerization pathways (Figure 2) in E. coli LS5218. This result is similar to our previous finding where P(3HB) was accumulated via the acetyl-CoA dimerization pathway in E. coli JM109.15 Western blot analysis of soluble fractions of recombinant E. coli LS5218 cells revealed that the amount of PHA synthase was indistinguishable among wild-type and mutant enzymes as shown in Figure 4, parts A and B. The mutations in PHA synthase also affected the monomer composition of the P(3HB-co-3HA) copolymer accumulated by recombinant E. coli LS5218. The recombinants expressing the single mutant (S325C or S325T) substituted at position 325 exhibited similar PHA compositions, with the main fraction consisting of 3HO similar to that of the wild-type enzyme (Figure 5). On the other hand, increased 3HB fractions (2.3- to 2.9-fold of wild-type) were observed for all of the single mutants (Q481K, Q481M, and Q481R) substituted at position 481. The Q481K mutant exhibited the highest 3HB fraction (40 mol %). Furthermore, these mutants exhibited increases in the mol % of the 3HHx fraction. However, there were decreases in the mol % fractions of 3HO to 3HDD, when compared to the P(3HB-co-3HA) copolymer synthesized by the wild-type PhaC1Ps. Basically, all of the Ser325/Gln481 double mutants exhibited a tendency

Biomacromolecules, Vol. 5, No. 2, 2004 483

Figure 4. Western blot analysis of a soluble fraction of recombinant E. coli LS5218 harboring wild type and mutants phaC1Ps genes by anti-PhaC1Ps antiserum. (A) Single mutants. Control plasmid (pGEM′′ABex) (lane 1), wild-type PhaC1Ps (lane 2), S325C mutant (lane3), S325T mutant (lane 4), Q481K mutant (lane 5), Q481M mutant (lane 6) and Q481R mutant (lane 7). (B) Double mutants. Control plasmid (pGEM′′ABex) (lane 1), wild-type PhaC1Ps (lane 2), S325C/Q481K mutant (lane 3), S325C/Q481M mutant (lane 4), S325C/Q481R mutant (lane 5), S325T/Q481K mutant (lane 6), S325T/Q481M mutant (lane 7), S325T/Q481R mutant (lane 8).

Figure 5. Composition of cellular P(3HB-co-3HA) of recombinant E. coli LS5218 strains harboring single or double mutants of the Pseudomonas sp. 61-3 PHA synthase. 3HB, 3-hydroxybutyrate; 3HHx, 3-hydroxyhexanoate; 3HO, 3-hydroxyoctanoate; 3HD, 3-hydroxydecanoate; 3-HDD, 3-hydroxydodecanoate. Others are the same as described in the legend of Figure 3.

to shift the substrate specificity/monomer composition toward C4 and C6 as did the Gln481 single mutants. These results strongly suggest that substitution at position 481 contributed to substrate specificity alteration and that the major fraction of the PHA copolymer was shifted from 3HO to 3HB. Table 1 summarizes the results of molecular weight analysis. For four copolymers extracted from E. coli recombinants (producing wild-type enzyme, S325T, Q481K, and S325T/Q481K mutants), their number-average molecular weights were determined to be 6.9 × 104 to 11.0 × 104 by gel permeation chromatographic analysis. From these results, S325T mutation tends to generate a polymer with lower molecular weight than Q481K mutant. In Vitro Assay of PhaC1Ps Mutants. Recombinant E. coli strains harboring the wild-type and mutant phaC1Ps genes accumulated P(3HB-co-3HA) copolymers with various monomer compositions. This divergence of monomer composition

484

Biomacromolecules, Vol. 5, No. 2, 2004

Takase et al.

Table 1. Molecular Weights of P(3HB-co-3HA) Copolymer Samples molecular weighta

Mn wild-type S325T mutant Q481K mutant S325T/Q481K mutant

(104)

10.8 8.7 11.0 6.9

Mw (104)

Mw/Mn

16.2 11.8 15.0 9.8

1.5 1.4 1.4 1.4

a M , number-average molecular weight; M , weight-average molecular n w weight; Mw/Mn, polydispersity.

Figure 6. In vitro assay of recombinant E. coli harboring the wildtype and mutant PHA synthase genes. A: The time course of the released CoA from (R,S)-3-hydroxybutyryl-CoA (3HB-CoA) by the crude extract of E. coli harboring no plasmid (diamond), wild-type PHA synthase gene (square), S325T mutant (triangle), Q481K mutant (cross), and S325T/Q481K mutant (circle). B: The time course of the released CoA from (R,S)-3-hydroxydecanoyl-CoA (3HD-CoA). Symbols are the same as A. C: Enzyme activities of the wild-type and mutant PHA synthase toward 3HB-CoA (black bar) and 3HDCoA (white bar). All values are averages ( S.D of at least five independent experiments.

in the copolymers was likely caused by differences in the substrate specificities of the PHA synthase mutants. However, the altered substrate specificities of the mutant enzymes could not clearly be determined in vivo, because in vivo monomer composition is affected not only by the substrate specificity of the PHA synthase but also by the distribution and amount of monomers supplied in the cells. To eliminate

the monomer supply and distribution in the cells as factors that could effect PHA monomer composition, the substrate specificities of the wild-type and mutant PHA synthases were characterized by an in vitro assay system using synthetic substrates. The specific activities of the enzymes toward two substrates, 3HB-CoA and 3HD-CoA, were assayed to biochemically characterize the PhaC1Ps enzymes. 3HB-CoA was chosen as a substrate because our initial aim was to create an enzyme with enhanced ability to incorporate the C4 monomer into P(3HB-co-3HA) compared to the wild-type enzyme. The 3HD-CoA substrate was used as another substrate because previous studies showed that the wild-type PhaC1Ps synthase had the highest activity toward 3HD-CoA among C4-C12 substrates. By measuring the activities of wild-type and mutant PhaC1Ps enzymes for these two substrates, we can deduce the effects of the individual point mutations on the substrate specificity and reactivity of the enzyme. To analyze mutational effects on the substrate specificity and reactivity of PhaC1Ps, two single mutants (S325T and Q481K) were subjected to in vitro assay as representative samples along with the double mutant S325T/ Q481K. These mutants were chosen because they represent enzymes with the highest PHA production (for S325T) and the production of the polymer with the highest 3HB composition (for Q481K) among the single mutants. Figure 6, parts A and B, shows the time course of the CoA-release from 3HB-CoA and 3HD-CoA catalyzed by the PHA synthases. In all conditions tested, CoA was released at nearly a constant rate without a significant lag-phase. Reaction rates were determined from the linear region (020 min). Activities shown in Figure 6C were calculated by subtracting the reaction rate of the crude extract of E. coli harboring no plasmid (control sample) from those of the recombinant crude extracts harboring the wild-type and mutant PHA synthase genes. For S325T, synthase activities were clearly increased toward both 3HB-CoA (2.7-fold) and 3HD-CoA (2.8-fold) compared with the wild-type PhaC1Ps activity. The in vivo PHA content and composition of 3HB were also increased by the S325T mutant, indicating that the S325T substitution increased the total catalytic activity of the enzyme. In contrast, the specific activity of the Q481K mutant toward 3HD-CoA was lower than that of the wildtype enzyme, whereas the activity toward 3HB-CoA was slightly reinforced. Thus, the relative activity toward the C4 substrate (C4/C10 ratio) by the Q481K mutant was remarkably increased compared to that of the wild-type. This alteration of substrate specificity was in good agreement with the change in monomer composition in the P(3HB-co-3HA) copolymer generated by the Q481K mutant in vivo. Although the Q481K mutation resulted in little increase in the activity toward 3HB-CoA, the combination of S325T and Q481K showed a strong synergistic effect on the activity toward 3HB-CoA (6.7-fold of wild-type). Thus, the S325T/Q481K double mutant has the highest activity toward C4 substrate among the PhaC1Pss tested in vitro. For 3HD-CoA, the specific activity of the double mutant enzyme was between those of each single mutant enzyme. The cooperative effect exerted by both mutations could account for the enhanced

Substrate Specificity Altered PHA Synthase

Biomacromolecules, Vol. 5, No. 2, 2004 485

production of P(3HB) homopolymer15 and of P(3HB-co3HA) copolymer.

tions related to substrate specificity and catalytic efficiency of PhaC1Ps and other types of PHA synthases.

Discussion

Acknowledgment. The authors thank Ms. Y. Ichikawa and Ms. R. Nakazawa for DNA sequencing (Bioarchitect Research Group, RIKEN Institute). This work was supported by Solution Oriented Research for Science and Technology (SORST) of the Japan Science and Technology Corporation (JST) and by the Special Postdoctoral Research Program of RIKEN Institute (to K.T.).

Until now, there has been no report on the successful alteration of the substrate chain-length specificity of PHA synthase from a type I enzyme to a type II counterpart or vice versa. In the previous study,15 we acquired several candidates of evolved PhaC1Ps enzymes with altered substrate specificity and a shift in activity toward the shortest chainlength substrate (3HB-CoA), based on the ability for these enzymes to accumulate a much higher amount of P(3HB) homopolymer in recombinant E. coli. The in vivo and in vitro assays in this study have demonstrated that the position 481 is a determinant for the substrate specificity of PhaC1Ps. This position is located adjacent to His479 which forms a putative catalytic diad inferred by sequence alignment. His479 has an equivalent role in other PHA synthases, for example, His508 (Ralstonia eutropha)30 and His331 (Allochromatium Vinosum).31 These His residues were shown to be the general base catalysts to generate the Cys thiolate required for covalent catalysis by site-directed mutagenesis. At present, it is unclear why substitutions by basic amino acids at position 481 result in fine-tuning PhaC1Ps to favor the 3HB monomer. Interestingly, the Q481M mutation gave rise to improvements in both activity and substrate specificity alteration. These beneficial mutations can scarcely be obtained by natural or artificial point mutations within a restricted genetic code.32 On the other hand, mutation of S325T contributes to increased catalytic efficiency. Ser325 is located near the Cys296 residue which is the active center of PhaC1Ps. Throughout the sequence alignment of all PHA synthases (type I to type III), Ser is strictly conserved at position 325, for all of type II PHA synthases including PhaC1Ps, whereas Thr is highly conserved at aligned positions of types I and III PHA synthases, both of which prefer a short chain-length 3HB substrate. Thr and Cys, preferable amino acids for activity increase, are polar residues most similar to Ser in chemical structure. X-ray crystallographic analysis would clarify the feasibility that enhanced accumulation of P(3HB) homopolymer and P(3HB-co-3HA) copolymer is achieved by these minor substitutions in bulkiness and polarity of amino acid. Investigating a functional connectivity between beneficial substitutions at positions 325 and 481 provides useful insights on the structure-function relationship of PhaC1Ps. Recently, we have obtained evidence that the Q481R mutation was gained from the first S325T mutation by a second errorprone PCR and the opposite mutation also occurred at similar low frequency (Takase, K. et al., unpublished data). This suggests the strong functional connectivity between both mutations that had been acquired for drastically enhanced synthesis of P(3HB).15 A switch between type II PHA synthase and its counterparts would be an attractive project from both academic and practical viewpoints. For this purpose, we are in the progress of further exploration of other beneficial amino acid substitu-

References and Notes (1) Madison, L. L.; Huisman, G. W. Microbiol. Mol. Biol. ReV. 1999, 63, 21. (2) Doi, Y. Macromol. Symp. 1995, 98, 585. (3) Aoyagi, Y.; Doi, Y.; Iwata, T. Polym. Degrad. Stab. 2003, 79, 209. (4) Iwata, T.; Tsunoda, K.; Aoyagi, Y.; Kusaka, S.; Yonezawa, N.; Doi, Y. Polym. Degrad. Stab. 2003, 79, 217. (5) Sudesh, K.; Abe, H.; Doi, Y. Prog. Polym. Sci. 2000, 25, 1503. (6) Arnold, F. H. Acc. Chem. Res. 1998, 31, 125. (7) Taguchi, S.; Ozaki, A.; Momose, H. Appl. EnViron. Microbiol. 1998, 64, 4652. (8) Taguchi, S.; Maehara, A.; Takase, K.; Nakahara, M.; Doi, Y. FEMS Microbiol. Lett. 2001, 198, 65. (9) Taguchi, S.; Maehara, A.; Takase, K.; Nakahara, M.; Nakamura, H.; Doi, Y. RIKEN ReV. 2001, 42, 3. (10) Taguchi, S.; Nakamura, H.; Hiraishi, T.; Yamato, I.; Doi, Y. J. Biochem. 2002, 131, 801. (11) Taguchi, S.; Matsusaki, H.; Matsumoto, K.; Takase, K.; Taguchi, K.; Doi, Y. Polymer Int. 2002, 51, 899. (12) Taguchi, S.; Nakamura, H.; Kichise, T.; Tsuge, T.; Yamato, I.; Doi, Y. Biochem. Eng. J. 2003, 16, 107. (13) Kichise, T.; Taguchi, S.; Doi, Y. Appl. EnViron. Microbiol. 2002, 68, 2411. (14) Matsusaki, H.; Manji, K.; Taguchi, K.; Kato, M.; Fukui, T.; Doi, Y. J. Bacteriol. 1998, 180, 6459. (15) Takase, K.; Taguchi, S.; Doi, Y. J. Biochem. 2003, 133, 139. (16) Yanisch-Perron, C.; Vieira, C.; Messing, J. Gene 1985, 33, 103. (17) Spratt, S. K.; Ginsburgh, C. L.; Nunn, W. D. J. Bacteriol. 1981, 146, 1166. (18) Sambrook, J.; Fritsch, E. F.; Maniatis, T. Molecular Cloning, 2nd ed.; Cold Spring Harbor Laboratory Press: Plainview, NY, 1989. (19) Salanitro, J. P.; Wegener, W. S. J. Bacteriol. 1971, 108, 885. (20) Matsusaki, H.; Abe, H.; Taguchi, K.; Fukui, T.; Doi, Y. Appl. Microbiol. Biotechnol. 2000, 53, 401. (21) Tsuge, T.; Taguchi, K.; Taguchi, S.; Doi, Y. Int. J. Biol. Macromol. 2003, 31, 195. (22) Fukui, T.; Yokomizo, S.; Kobayashi, G.; Doi, Y. FEMS Microbiol. Lett. 1999, 170, 69. (23) Kato, M., Bao, H. J.; Kang, C. K.; Fukui, T.; Doi, Y. Appl. Microbiol. Biotechnol. 1996, 45, 363. (24) Schubert, P.; Steinbu¨chel, A.; Schlegel, H. G. J. Bacteriol. 1988, 170, 5837. (25) Kraak, M. N.; Kessler, B.; Witholt, B. Eur. J. Biochem. 1997, 250, 432. (26) de Roo, G.; Ren, Q.; Witholt, B.; Kessler, B. J. Microbiol. Methods 2000, 41, 1. (27) Bradford, M. M. Anal. Biochem. 1976, 72, 248. (28) Murata, T.; Takase, K.; Yamato, I.; Igarashi, K.; Kakinuma, Y. J. Biol. Chem. 1997, 272, 24885. (29) Matsumoto, K.; Matsusaki, H.; Taguchi, K.; Seki, M.; Doi, Y. Biomacromolecules 2002, 3, 787. (30) Jia, Y.; Yuan, W.; Wodzinska, J.; Park, C.; Sinskey, A. J.; Stubbe, J. Biochemistry 2001, 40, 1011. (31) Jia, Y.; Kappock, T. J.; Frick, T.; Sinskey, A. J.; Stubbe, J. Biochemistry 2000, 39, 3917. (32) Taguchi, S.; Komada, S.; Momose, H. Appl. EnViron. Microbiol. 2000, 66, 1410. (33) Kusaka, S.; Abe, H.; Lee, S. Y.; Doi, Y. Appl. Microbiol. Biotechnol. 1997, 47, 140. (34) Gerngross, T. U.; Snell, K. D.; Peoples, O. P.; Sinskey, A. J. Biochemistry 1994, 33, 9311.

BM034323+