Biosynthesis in Ectothiorhodospira shaposhnikovii - American

Biomacromolecules 2004, 5, 40-48

40

Poly(hydroxyalkanoic acid) Biosynthesis in Ectothiorhodospira shaposhnikovii: Characterization and Reactivity of a Type III PHA Synthase Shiming Zhang,† Steven Kolvek,† Steve Goodwin,*,† and Robert W. Lenz‡ Department of Microbiology and Polymer Science and Engineering Department, University of Massachusetts, Amherst, Massachusetts 01003 Received June 3, 2003; Revised Manuscript Received October 3, 2003

Ectothiorhodospira shaposhnikoVii is able to accumulate polyhydroxybutyrate (PHB) photoautotrophically during nitrogen-limited growth. The activity of polyhydroxyalkanoate (PHA) synthase in the cells correlates with PHB accumulation. PHA synthase samples collected during the light period do not show a lag phase during in vitro polymerization. Synthase samples collected in the dark period displays a significant lag phase during in vitro polymerization. The lag phase can be eliminated by reacting the PHA synthase with the monomer, 3-hydroxybutyryl-CoA (3HBCoA). The PHA synthase genes (phaC and phaE) were cloned by screening a genomic library for PHA accumulation in E. coli cells. The PHA synthase expressed in the recombinant E. coli cells was purified to homogeneity. Both sequence analysis and biochemical studies indicated that this PHA synthase consists of two subunits, PhaE and PhaC and, therefore, belongs to the type III PHA synthases. Two major complexes were identified in preparations of purified PHA synthase. The large complex appears to be composed of 12 PhaC subunits and 12 PhaE subunits (dodecamer), whereas the small complex appears to be composed of 6 PhaC and 6 PhaE subunits (hexamer). In dilute aqueous solution, the synthase is predominantly composed of hexamer and has low activity accompanied with a significant lag period at the initial stage of reaction. The percentage of dodecameric complex increases with increasing salt concentration. The dodecameric complex has a greatly increased specific activity for the polymerization of 3HBCoA and a negligible lag period. The results from in vitro polymerizations of 3HBCoA suggest that the PHA synthase from E. shaposhnikoVii may catalyze a living polymerization and demonstrate that two PhaC and two PhaE subunits comprise a single catalytic site in the synthase complex. Introduction Poly(hydroxyalkanoates) (PHAs) are a class of polyesters produced by a wide range of bacteria as carbon and energy storage in the form of intracellular inclusion bodies. Poly(3-hydroxybutryric acid) (PHB) was first described as a storage compound in Bacillus megaterium.16 Within the cells, PHA synthases initiate and catalyze the polymerization of coenzyme A thioester derivatives of hydroxyalkanoic acids (HACoAs) to produce PHAs while releasing free CoA. The properties of PHA synthases have significant influence on PHA production during both in vivo and in vitro synthesis.1,11,13,19,23,33 Of the different types of PHA synthases investigated to date, three types have been described.20 Types I and II, which are represented by the synthases of Ralstonia eutropha and Pseudomonas oleoVorans, respectively, consist of a single subunit (PhaC).9,12,25,26,31,33 The type I synthases preferentially polymerize monomers with 3-5 carbons,10 whereas the type II synthases preferentially polymerize monomers with more than 5 carbons.4,5 Type III synthases, as represented by the synthase of Chromatium Vinosum, consist of two different * To whom correspondence should be addressed. Phone: 413-545-4204. Fax: 413-545-1977. E-mail: [email protected]. † Department of Microbiology. ‡ Polymer Science and Engineering Department.

subunits, PhaC and PhaE.19-20 Type III synthases have a similar preference for monomer chain length as that of the type I synthases (3-5 carbons). The genes encoding type III synthases have been cloned from several purple sulfur photosynthetic bacteria, including C. Vinosum,22 Thiocystis Violacea,21 Thiocapsa pfennigii,18 and from the cyanobacterium Synechocystis sp. PCC6803.11,15 Biochemical studies have been carried out mainly on the PHA synthase from C. Vinosum.13,20,24,37 While having a similar preference for monomer size, type III PHA synthases have been reported to differ from type I synthases in many aspects. The purified PHA synthase from Ralstonia eutropha (PhaCRe), a representative of type I synthases, contains two physically different forms, monomer and dimer in an equilibrium,9 and the dimer has been shown to be the active form.40,41 As a result of this equilibrium, the polymerization reaction catalyzed by PhaCRe displays a significant lag phase.9 Similarly, the PHA synthase from Chromatium Vinosum (PhaECCv), a representative of type III PHA synthases, was also found to form oligomeric complexes,13,20 but it has not been found to have a significant lag phase.13,20,24 Despite the differences noted above, there may be close similarities between type I and type III synthases with respect to the mechanisms of polymerization. Mutagenesis and protein chemistry studies revealed that the cysteine residue,

10.1021/bm034171i CCC: $27.50 © 2004 American Chemical Society Published on Web 12/03/2003

PHA Synthase from E. shaposhnikovii

C319, in PhaCRe is an essential residue involved in the covalent catalysis of the polymerization reaction,9,36 and a corresponding cysteine, C149, in PhaECCv was also shown to be involved in the covalent catalysis with that PHA synthase.24 The present investigation was focused on the PHA synthase from the anoxygenic photosynthetic bacterium, Ectothiorhodospira shaposhnikoVii, which is a nearly obligatory phototrophic bacterium that is able to grow vigorously under photolithoautotrophic conditions.2,38 This bacterium is known to accumulate large amounts of PHA, producing up to 29% of the dry cell mass as PHA when grown with acetate as a carbon source,17 and the PHA synthase of this bacterium is a type III synthase. In the present study, the synthase of E. shaposhnikoVii was obtained from a recombinant strain of E. coli containing the genes coding for the two proteins that form the enzyme. The synthase was purified and evaluated for the in vitro polymerization of the 3HBCoA monomers in aqueous solution to determine the active form of the enzyme and the characteristics of the polymerization reaction, including the extent of the lag period, the kinetic parameters for the polymerization reaction, and the relationship between molecular weight of the polymer formed and the molar ratio of monomer to enzyme complex. One purpose of the molecular weight study was to determine whether the in vitro reaction is a “living polymerization” as has been reported for the polymerization of HBCoA by the synthase of R. eutropha.35,37-41 Materials and Methods Culture Conditions for E. shaposhnikoWii. E. shaposhnikoVii was cultured in the medium described by Bognar and Newman with sodium bicarbonate as the sole carbon source.2 To study the accumulation of PHB, 1 L of culture was first grown anaerobically in the light in medium containing 20 mM NH4Cl at 35 °C to an A660 of 0.4. The cells were spun down and washed once with the nitrogen-limited medium (1 mM NH4Cl). The cells were then allowed to continue growing anaerobically in the light in the nitrogen-limited medium. After 11.5 h, the culture was shifted to the dark. Samples of 100-150 mL were collected at various time points over the light and the dark periods, and the cells were pelleted and stored at -80 °C for later analyses of PHA accumulation and PHA synthase activity as detailed below. Preparation of Cell Free Extract. Cell free extract of E. shaposhnikoVii was prepared by sonication on a Sonic Dismembrator 550 (Fisher Scientific) followed by centrifugation at 20 000 g for 30 min to obtain clear supernatant for the activity assay. PHA Analysis. Lyophilized cell pellets were heated at 110 °C for 1 h in the presence of 2.5 N NaOH. After the addition of an equal amount of 2.5 N HCl, samples were centrifuged at 20 000 g for 10 min at 4 °C. The supernatant was then filtered through a 0.22 µm filter and analyzed by high performance liquid chromatography (HPLC) (Shimadzu LC6A, Kyoto, Japan) using a UV detector. Poly(3-hydroxybutyrate-3-hydroxyvalerate) copolymer (Aldrich Chemical) was used as a standard (17% HV content). Cloning of the E. shaposhnikoWii PHA Biosynthetic Genes. The plasmid pBluescript II SK+ (Stratagene La Jolla,

Biomacromolecules, Vol. 5, No. 1, 2004 41

CA) was used as a cloning and expression vector. E. coli Top 10 (Invitrogen, Carlsbad, CA) was used as the host for cloning and expression in all experiments. Genomic DNA was isolated using the method of Zhang et al.42 Recovery and purification of DNA fragments from agarose gels were performed using a QIAquick Kit, and large-scale preparation of plasmid DNA was conducted using a Qiagen Maxi Kit (Qiagen, Valencia, Ca.) following the vendor’s instructions. Unless otherwise noted, all other molecular cloning techniques were performed according to Sambrook et al.30 Genomic DNA of E. shaposhnikoVii was partially digested with ApaI (New England Biolabs), and fragments of 2-20 kb were ligated into the ApaI cloning site of pBluescript II SK+. Ligation products were used to transform E. coli Top10 cells via electroporation on Gibco BRL Cell-Porator according to the manufacturer’s instructions. The transformants were plated onto PHA accumulating medium and allowed to grow for 36 h. The colonies accumulating PHA were identified visually under UV light. E. coli cultures were routinely grown and maintained in Luria-Bertani (LB) medium. PHA accumulating medium consisted of LB medium supplemented with 20 g/L of glucose. For the selection of recombinants, ampicillin was used at a final concentration of 100 µg/mL. Nile Red (0.5 µg/mL) was added to the agar media to screen for PHA production. All cultures were incubated at 35 °C. Broth cultures were incubated on a horizontal shaker set at 225300 rpm. For PHA accumulation experiments, triplicate 50 mL cultures of Top10 (pSK64) cells were grown in 250 mL Erlenmeyer flasks with LB medium with shaking at 250 rpm for 36 h at 35 °C. Glucose was added to a final concentration of 0.5% (w/v) after filter sterilization (0.22 µm Millipore Millex). Nucleotide Sequence Analysis. Sequencing was done on a PE ABI 377 automated sequencer (Foster City, CA). Sequence analysis was done using the programs in the National Center for Biotechnology Information and the programs in the GCG package (Wisconsin package, version 10). The sequence reported in this study is available from GenBank under accession number AF307334. Purification of the PHA Synthase. Top10(pSK64) cells were grown in 1.6 L of LB medium in a 4-liter Erlenmeyer flask shaking at 225 rpm at 35 °C for 12-16 h (O.D.600 1-1.5). Cells were harvested at 6000 g at 4 °C for 5 min. The cell pellet from 250 mL of culture was resuspended in 15 mL of 20 mM Tris-Cl pH 8.0 (buffer A) and digested with lysozyme (100 µg/mL) for 10 min at room temperature. The suspension was then sonicated in an ice-water bath using a Sonic Dismembrator 550 (Fisher Scientific) with the following cycle: 0.8 s on and 0.2 s off for 1 min. The cells were allowed to cool for 3 min, and then the sequence was repeated for another minute. The sample was then centrifuged at 20 000 g for 30 min at 4 °C and the supernatant was filtered through a low protein binding membrane with 0.45 µm pore size (Acrodisc, Pall Corporation). This solution was termed the cell free extract (CFE). Chromatography was performed at room temperature at a developing flow rate of 2 mL min-1 and fractions of 2 mL. The CFE was loaded onto an Econo-Pac High Q column

42

Biomacromolecules, Vol. 5, No. 1, 2004

(Bio-Rad, 5 mL bed volume) that had been equilibrated with 20 mM Tris-Cl pH 8.0 (buffer A). The column was then washed with 4 bed volumes of buffer A, and the PHA synthase was eluted with a 60 mL linear gradient of 0-40% buffer B (20 mM Tris-Cl, pH 8.0, 1.5 M NaCl). Fractions containing high activity were pooled and loaded directly onto a HiTrap Blue column (Pharmacia Biotech, 5 mL bed volume) which had been equilibrated with 15% buffer B. The column was washed with 6 bed volumes of 15% buffer B and then with a 60 mL linear gradient of 15-100% buffer B. The enzyme was eluted at 80-100% buffer B. Fractions containing high activity were pooled and diluted 10-fold with buffer C (buffer A adjusted to pH 6.0) before reloading onto the buffer C preequilibrated Econo-Pac Q column. The column was run exactly as the previous column except at a pH of 6.0. Fractions of high activity were pooled and stored at -20 °C. PHA Synthase Activity Assay. PHA synthase activity was assayed spectrophotometrically by measuring the absorption decrease at 236 nm due to the breakage of the thioester bond in 3-hydroxybutryl-CoA according to Fukui.7 The absorption change in the first 30 s was taken to calculate the initial reaction velocity. The reaction was carried out in 150 mM Tris-Cl pH 8.0 in a total volume of 1 mL. The synthase was added last to initiate the reaction. One unit of activity is defined as the amount required to catalyze the conversion of 1 µmol of substrate in 1 min. Protein concentration was determined according to Bradford 3. Synthesis of (R)-3-Hydroxybutyryl CoA. R-3HBCoA was synthesized according to the method described by Simon and Shemen,32 using (R)-β-butyrolactone as the starting material. The lactone was added in 10:1 molar ratio into coenzyme A in 0.2 M K2CO3 solution, and the reaction was set on an ice-water bath with stirring. The formation of thioester was monitored by measuring the absorption increase at 236 nm. Meanwhile, the free CoA remaining in the reaction mixture was monitored by reacting with 5,5′dithiobis(2-nitrobenzoic acid), DTNB, in 100 mM Tris-Cl, pH 8.0. The completed reaction mixture, usually in 5-10 min, was adjusted to neutral pH and extracted with ethyl ether three times to remove any lactone. The aqueous phase was evaporated to remove any ether and lyophilized to obtain 3HBCoA that was used for PHA synthase activity assay and in vitro polymerization without further purification. In Vitro Polymerization of PHB and Molecular Weight Determination. The reactions were carried out in 20 mL of 100 mM Tris buffer solution at pH 8.0, containing 100 µg/ mL BSA, 58 µmol of 3HBCoA monomer and varied amounts of PHA synthase to obtain various ratios of monomer/synthase. The reaction progress was followed by measuring the released free CoA. To do this, a 20 µL aliquot of reaction solution at different time points was added to 20 µL of 5% trichloroacetic acid solution followed by addition of 1 mL of 1 mM DTNB in 100 mM Tris-Cl (pH 8.0) solution. Readings at 412 nm were then recorded to calculate the CoA concentration using a molar 412 of 13 700. After completion, the reaction mixture was extracted three times with warm chloroform (∼50 °C). The volume of the extract was reduced by evaporation under reduced pressure before

Zhang et al.

Figure 1. Relationship between PHB accumulation and the in vitro PHB synthase activity in the E. shaposhnikovii cells after transfer into nitrogen limited medium, the transition of the culture from the light to the dark is marked by the “light off” arrow.

precipitating the PHB polymer with methanol. The polymer pellet was dissolved in chloroform and reprecipitated with methanol. The final polymer pellet was dried under reduced pressure. The molecular weight of the polymers was determined by gel permeation chromatography (GPC) on a Shodex GPC system 11 (Showa Denko K. K., Tokyo, Japan) equipped with RI detector. The columns used were AC-80P (Guard column) and K-806M. Polystyrene standards were used as references. The absolute molecular weight determined by the scattering method has been found to be 0.7 of the value determined by the GPC method,6 and accordingly, the molecular weights of the polymers obtained by GPC were adjusted by a factor of 0.7. Results Relationship between PHB Accumulation and PHA Synthase Activity. PHB accumulation and PHA synthase activity in E. shaposhnikoVii were examined during growth both in the light and in the dark after the cells were transferred into nitrogen limited medium as shown in Figure 1. The activity of the synthase was determined by measuring the decrease in monomer concentration as indicated by the decrease in absorption of the thioester bond at 236 nm. There was no detectable amount of PHB in the cells grown in the normal growth medium. The cells were then transferred into the nitrogen-limited medium (time zero point). After 7 h, the cells began to accumulate PHB, and the PHB content increased rapidly during the next 3 h, until the cells were transferred to the dark regimen. The cells stopped growing in the dark, as expected2,38 and the PHB content dropped briefly followed by a moderate increase as shown in Figure 1. As shown in the figure, the specific activities of PHA synthase in these samples in vitro correlated well with PHB accumulation. The activity of the synthase was determined in vitro by a detailed kinetic study of cell-free extracts collected during the period of maximum PHB accumulation. The synthase did not show a lag phase in these in vitro polymerization reactions as shown in Figure 2a which is a plot of the rate of the reaction of 3HBCoA monomer with the synthase. In contrast, the samples from the dark period displayed a significant lag phase in the reaction as shown by the “first run” in 2b, but as expected, this lag phase was absent after


Figure 2. Polymerization reactions of 3HBCoA catalyzed by the crude extract of E. shaposhnikovii cells: (a) sample collected at the beginning of the dark period; (b) sample collected 40 h into the dark period, and after completion of the first run of the reaction, fresh 3HBCoA was added to the reaction mixture for the second run reaction.

the enzyme had reacted with the monomer in the “second run” in Figure 2b. The specific activity of the primed synthase from the dark period was found to be similar to the specific activity of the synthase collected during the period of maximum PHB accumulation. Cloning of the PHA Biosynthetic Genes of E. shaposhnikoWii. Several strategies have been used in cloning PHA synthetic genes from a variety of bacteria.28 Recently, Speikermann et al.34 reported the use of Nile Red to stain PHA granules in living cells. This approach was adopted here to screen for clones of the PHA synthetic genes from E. shaposhnikoVii. After transformation with a genomic library of E. shaposhnikoVii DNA, one E. coli transformant was identified as a putative PHA producer by visualizing the fluorescence emission under UV light in the presence of Nile Red. From this clone, the plasmid DNA was extracted and used to retransform E. coli cells. In this case, all transformants were fluorescent under UV illumination. This plasmid, pSK64, was found to contain a 6.4 kb insert. To verify that the 6.4 kb insert was from E. shaposhnikoVii DNA, a Southern hybridization was performed using a BglII fragment (∼500 bp) of this insert to hybridize to E. shaposhnikoVii genomic DNA that had been digested with ApaI. On the autoradiograph, a hybridization band appeared at the same position as the control signal of 6.4 kb insert (data not shown). The successful cloning of PHA biosynthetic


genes was further confirmed by examining the PHA accumulation in the recombinant E. coli cells. In the presence of 20 mM glucose in LB medium, PHB accumulated to 24% (w/w dry cell). Sequence analysis identified seven open reading frames in the 6.4 kb insert (Figure 3). Five of them are biosynthesis relevant genes, phaB, phaP, phaA, phaE, and phaC. The latter two, phaE and phaC, encode the two subunits of the PHA synthase. The primary structure of the phaE gene product shows significant homology to the PhaE proteins of C. Vinosum and T. Violacea (39% and 37% identities, respectively). The predicted Mr of the product (42061) is slightly greater than the Mr of the PhaE proteins of C. Vinosum (40525) and T. Violacea (41450). This gene is designated as phaEEs. The primary structure of the phaC gene product shows strong similarity to the PHA synthase subunit C (PhaC) from C. Vinosum (68%) and from T. Violacea (67%). The predicted Mr of the gene product (39405) is in good agreement with the PhaC proteins of C. Vinosum (39730) and T. Violacea (39550). This gene is designated as phaCEs. Purification of the PHA Synthase of E. shaposhnikoWii. It was observed that when grown in LB medium, the Top10(pSK64) cells expressed significant amounts of PHA synthase (around 5%), so further subcloning was not pursued. The purification procedure consisted of two rounds of ion exchange chromatography on a Q column at pH 8.0 and 6.0, with an intervening round of affinity chromatography on a HiTrap Blue column. The PHA synthase was purified to greater than 90% homogeneity with an overall yield of 41% (Table 1). The PHA synthase exists as large oligomeric complex(es), as indicated by size exclusion chromatography, similar to that of PhaECCv.20 SDS-PAGE analysis of the purified PHA synthase revealed two subunits in a nearly 1:1 ratio (Figure 4), which is consistent with the stoichiometry of PhaE and PhaC in PhaECCv.24 The apparent Mr of the upper band is 41.7 kd, which is close to the predicted Mr of the phaEEs gene product (42061). The lower band has an apparent Mr of 40.6 kD, which agrees with the predicted Mr of the phaCEs gene product (39405). Both the gene information and the SDS-PAGE results demonstrated that PHA synthase from E. shaposhnikoVii is a complex of two types of subunits, PhaC and PhaE, and that it therefore belongs to the type III PHA synthases. By analogy to the PHA synthase from C. Vinosum (PhaECCv), the PHA synthase from E. shaposhnikoVii is designated as PhaECEs in this report. In Vitro Polymerization of 3HBCoA with PhaECEs. The purified PHA synthase from the recombinant E. coli cells did not show a significant lag phase during in vitro polymerization reactions under the conditions used. However, the activity of PhaECEs was strongly affected by buffer concentration as shown in Figure 5a. The optimal concentration was found to be around 150 mM Tris, with lower concentrations affecting activity more severely. This result is believed to be an ionic strength effect, because a similar phenomenon was observed using various concentrations of NaCl solution. Determination of the optimal pH was therefore done in 20 mM Tris buffer solution complemented with 100

44


Zhang et al.

Figure 3. Restriction map and organization of the PHA biosynthesis genes in the 6.4 kb insert. Table 1. Purification of the PHA Synthase of E. Shaposhnikovii

a

samples

volume (mL)

total protein (mg)

total activity (units)

specific activity (units/mg Pro.)

purification (fold)

yield (%)

CFEa Econo Q (pH 8.0) HiTrap Blue Econo Q (pH 6.0)

15 17 24 10

168 63 5.78 3.94

265 185 117 109

1.57 2.93 20.2 27.6

1.86 12.8 17.5

100 69.8 44.1 41.1

Cell free extract.

Figure 4. SDS-PAGE analysis of the purified PHA synthase. Lane 1 to 5, samples were run on a mini-gel (4-20% gradient). Lane 1, molecular weight standards; lane 2, cell free extract; lane 3, first Econo Pac Q elutent (pH8.0); lane 4, HiTrap Blue elutent; lane 5, second Econo Pac Q elutent (pH6.0); lane 6, the same sample as in the lane 5 run on 12% gel concentration with a slab of 15 × 15 cm dimension.

mM NaCl to maintain the ionic strength nearly constant at different pH values. By doing so, the optimal pH for the PHA synthase activity was estimated to be 8.0 as shown in Figure 5b. The in vitro polymerization of 3HBCoA was performed at three monomer-to-synthase ratios using a fixed amount of monomer and varying amounts of PHA synthase. The reaction did not show a lag phase and went to completion in 2-4 min with 91% monomer conversion. A good linear relationship between the polymer molecular weight and the monomer/synthase ratio was observed as shown in Figure 6, indicating that the in vitro polymerization of 3HBCoA by PhaECEs may be a living polymerization; that is, a polymerization reaction which does not have a chain transfer or termination reaction and in which each molecule of initiator polymerizes one polymer chain and remains constantly bound to that chain.

Kinetic properties of PhaECEs were investigated using three monomers: 3-HBCoA, 3-hydroxyvaleryl CoA (3-HVCoA), and 4- hydroxybutyryl CoA (4HBCoA). The Km and Vmax values were obtained by double-reciprocal plots of the initial reaction velocities as a function of the substrate concentrations. The kcat values were calculated from Vmax values based on the assumption that the two PhaC and two PhaE comprise a single catalytic site as explained below. The results are summarized in Table 2. Lag Phase in the Polymerization Reaction and Its Elimination. While developing the purification protocol, it was noticed that after loading the sample onto HiTrap Blue column at 20 mM Tris solution containing 0.125 M NaCl, part of the PHA synthase activity could be washed out by 20 mM Tris buffer solution in the absence of NaCl, whereas the remaining activity had to be eluted at around 1.4 M NaCl concentration. These two PHA synthase preparations showed similar patterns with respect to the subunit compositions on SDS-PAGE (data not shown), but they displayed different kinetic properties. Based on this observation, the purified PHA synthase sample was passed through the HiTrap Blue column once more and the PHA synthase activities that washed out at either low salt concentration or at high salt concentration were collected separately for analysis. Although the PHA synthase preparation that was eluted at high salt concentration did not show a significant lag phase in the polymerization reaction (data not shown), the PHA synthase preparation that was washed out with 20 mM Tris buffer solution in the absence of NaCl demonstrated a significant lag phase as shown in Figure 7 for the “first run” of the reaction. The lag phase disappeared in the “second run” of the reaction, after fresh monomer was added, and a large increase of the specific activity was observed. The above observations suggest that addition of salt into the PHA synthase sample may be able to overcome the lag


Biomacromolecules, Vol. 5, No. 1, 2004 45 Table 2. Kinetic Parameters of the Polymerization Reactions Initiated and Catalyzed by PhaECEs substrates [R]-3HBCoA 4HBCoA [R]-3HVCoA

Km (M) 10-5

6.5 × 8.5 × 10-5 3.0 × 10-5

kcat (s-1)a

kcat/Km (M-1 s-1)

320 297 122

4.9 × 106 3.5 × 106 4.1 × 106

a Based on the assumption that 2 PhaC and 2 PhaE subunits comprise one catalytic site.

Figure 5. Characterization on PhaECEs: (a) effect of buffer concentration and (b) effect of pH on the enzyme activity.

Figure 6. Relationship between the molecular weight of PHB and the molar ratio of 3HBCoA-to-synthase.

phase and increase the activity in the PhaECEs sample. To test this possibility, sodium chloride was added to the PhaECEs sample that had been eluted in 20 mM Tris buffer solution in the absence of NaCl. As seen in Figure 8, addition of NaCl shortened the lag time in the reaction and increased the specific activity of the PHA synthase. The PHA synthase sample in the absence of NaCl possessed a specific activity of 17 units/mg protein. When NaCl was added to the above PHA synthase sample to a final concentration of 0.125 M,

Figure 7. Lag phase in the polymerization reaction with PhaECEs and its elimination by priming reaction: (2) first run of the reaction; (b) the second run of the reaction; additional monomer was added to the mixture after the first run of the reaction was completed.

Figure 8. Effect of salt concentration on the kinetic properties of PhaECEs: (() synthase sample in 20 mM Tris buffer solution; (2) in 20mM Tris buffer solution containing 0.125 M NaCl; (b) in 20 mM Tris buffer solution containing 1.5 M NaCl.

the specific activity increased to 52 units/mg protein. Further increasing the concentration of NaCl to a final concentration of 1.5 M led to an increase of the specific activity to 83 units/mg protein. These results clearly demonstrate that increasing the salt concentration in the PHA synthase sample efficiently shortened the lag phase and greatly increased the PHA synthase specific activity, in a manner similar to the priming reaction. PhaECEs Activity Directly Related to Different Forms of Enzyme Complexes. To determine whether the different

46


Zhang et al.

Figure 9. Size exclusion chromatography of PhaECEs samples in different salt concentrations or after partial priming, numbers above the peaks indicate molecular weight in Daltons.

specific activities in the samples had a direct relationship with different PHA synthase complexes, the samples discussed above and a partially primed sample were subjected to size exclusion chromatography. For each sample, three runs of chromatography were performed separately with PHA synthase sample and with molecular weight standards to determine the size of the PHA synthase complexes. The chromatography conditions were the same for all three samples. As shown in Figure 9, three complexes can be distinguished corresponding to sizes of 151 000, 473 000, and 955 000 Daltons, respectively, with the latter two being the dominant. Based on the molecular weight of the C and E subunits and their 1:1 stoichiometry in the PHA synthase complex, the 473 000 complex should be a hexamer 6(PhaECEs), and the 955 000 complex should be a dodecamer 12(PhaECEs). The very minor 151 000 complex is likely a dimer 2(PhaECEs), but its activity cannot be separated from the tail activity of the 473 000 complex. The relative amounts of the two complexes (hexamer and dodecamer) changed greatly, depending upon the salt concentration in the enzyme solution. The PHA synthase in 20 mM Tris buffer in the absence of NaCl predominantly contained the hexameric complex with only small amounts of dodecameric complex. In the sample containing 0.125 M NaCl, the dodecamer increased from the conversion of hexamer. When the concentration of NaCl was increased to 1.5 M, most of the hexamer was converted into dodecamer. The activity profile of the fractions shows a pattern similar to the PHA synthase protein profile (data not shown). The conversion of hexamer to dodecamer was also observed in the partially primed PHA synthase sample (Figure 9). These results clearly demonstrate that different specific activities

Table 3. Comparison of the Measured and Predicted Molecular Weights of PHB

Mna sample

M/Eb

measured

predictedc

Mwd

Mw/Mn

1 2 3

5× 1 × 104 2 × 104

354 000 811 000 1 603 000

391 000 783 000 1 610 000

478 000 1 039 000 2 338 000

1.35 1.28 1.46

103

a Number average molecular weight. b Molar ratio of 3HBCoA (M) to the synthase (E), based on an assumption that the equivalent molecular weight of a single catalytic site in the oligomeric synthase complex is 163 000 Daltons. c Based on 91% conversion and the assumption in 1. d Weight average molecular weight.

are directly related to different forms of the PHA synthase complexes. Dodecamer is a much more active complex than hexamer, although it cannot be concluded that the hexamer is completely inactive. Two PhaCEs and PhaEEs Subunits Is Equivalent to a Single Catalytic Site. Because the dominant physical form of the PhaECEs is an oligomeric complex, there is a question as to what is the basic catalytic unit in the complex. If the polymerization of 3HBCoA with this PHA synthase has properties of a living polymerization as suggested above, then this question can be addressed by comparing the molecular weight of polymer produced at various ratios of substrate to enzyme. Table 3 summarizes the measured molecular weight of synthesized PHB samples at three molar ratios of 3HBCoA to PHA synthase. As shown in Table 3, the number average molecular weights (Mn) of the polymers, at 3HBCoA/ synthase molar ratios of 5000, 10 000, and 20 000 were 354 000, 811 000, and 6 103 000 Daltons, respectively. If two PhaC and two PhaE subunits comprise a single catalytic site in the PHA synthase complex, the predicted Mn values are 391 300, 782 600, and 1 610 000 Daltons. Therefore,


based on the definition of a living polymerization given above, the results obtained strongly support the assumption that two PhaCEs and two PhaEEs subunits comprise a single initiation and catalytic site in the PHA synthase complex structure. That is, the active site serves to both initiate and catalyze a chain growth polymerization reaction for the conversion of HBCoA to PHB. Discussion Although the polymerization reactions have the characteristics of a “living polymerization” based on the relationship between Mn and the monomer-to-initiator ratio, the molecular weight distributions are fairly broad, approximately 1.31.5 Mw/Mn, because the initiation reaction is slow compared to propagation. The initiation process involves the continuous formation of the active catalyst-initiator complex from individual protein molecules in solution, so the initiation reaction occurs over a relatively long period of time compared to the propagation reaction. Despite the fact that over 30 PHA synthase genes have been cloned, only four PHA synthases (at least one of each type) have been purified for further studies.9,19,20,27,29,37 This study describes the properties of an additional purified type III PHA synthase. E. shaposhnikoVii produced only PHB homopolymers in vivo, but the purified synthase (PhaECEs) was shown to polymerize two other HACoA substrates in vitro including 3-hydroxyvalerylCoA (3HVCoA) and 4-hydroxybutyrylCoA (4HBCoA). The results are similar to those obtained with the PhaCRe, a Type I synthase.39,41 Of the three effective monomers, 3HBCoA was the most efficient substrate for polymerization with PhaECEs, whereas 3HVCoA was intermediate and 4HBCoA was the least efficient. A detailed analysis of these kinetic data showed that, compared with 3HBCoA, the increase in the chain size in 3HVCoA had more effect on the catalytic rate than on the binding affinity of the enzyme as indicated by the much smaller value of kcat and also the smaller value of Km. However, moving the hydroxyl group from the β carbon to the γ carbon strongly affected the substrate binding affinity of the enzyme but had little effect on the catalytic rate as shown by the slightly smaller value of kcat and much higher value of Km. These features are also observed in PhaCRe toward the same three substrates.41 The similarity in kinetics and substrate specificities of these two types of synthases suggest that, even though they are very different in overall structure, they may share some attributes with respect to binding and catalytic sites. The present results demonstrate that two major oligomeric complexes, hexamer and dodecamer, can exist for PHA synthase from E. shaposhnikoVii in aqueous solution and that their ratio is dependent upon the salt concentration in the enzyme solution. The dodecamer formed at high salt concentration has a much higher specific activity than the hexamer that is dominant at low salt concentration. The hexamer also shows a significant lag phase during the polymerization reaction, as expected, and the lag phase of the hexameric synthase was eliminated after priming by reaction with 3HBCoA. Priming was accompanied by the


conversion of hexamer to dodecamer and a large increase of specific activity. The significant lag phase and low catalytic rate during the initial stage of the polymerization reaction of the hexamer is comparable to the results obtained with the purified monomer of PhaCRe during in vitro polymerization.9 Multihydroxyl compounds, such as fructose, have also been shown to eliminate the lag phase with PhaCRe,41 but they are not able to efficiently eliminate the lag phase in PhaECEs, nor are salts able to eliminate the lag phase with PhaCRe (data not shown). These observations suggest that there are differences in the complexing characteristics of the two synthases. Dimerization of PhaCRe subunits may be promoted by hydrophobic interactions. Large amounts of hydroxyl groups in the solution could serve to weaken the interactions between the protein and water and lead to dimerization of the monomers by hydrophobic interactions. For PhaECEs, dimerization of the hexamer to dodecamer may be prevented by the charged surface of the complexes. Introduction of salt into the solution could serve to neutralize the charge at the complex surface and facilitate the dimerization of hexameric complex to dodecamer. The function of salt in regulating the activity of PhaECEs seems to match the halophilicity of E. shaposhnikoVii. Despite the different responses to the concentrations of salts and multihydroxyl compounds between the two types of synthases, the priming reaction greatly improves the catalytic performance of both types. In fact, for all three types of PHA synthases a lag phase has been found to exist and in each case the lag phase can be eliminated by priming the synthase with monomer [refs 9 and 27 and this study]. The suggestion that the polymerization of 3HBCoA by the PHA synthase from E. shaposhnikoVii may be a living polymerization is consistent with the behavior of PhaCRe,8,35 but it is very different from the type of reactivity reported for PhaECCv.14 PHB synthesized in vitro with PhaECCv was found to be of much lower molecular weight than would be expected for a living polymerization.14 In the present case, the relationship between the molecular weight of PHB and the ratio of the 3HBCoA-to-synthase for PhaECEs indicated that the basic catalytic unit in the large synthase complex is equivalent to two PhaC and two PhaE subunits. That is, a comparison of the polymer size as a function of the ratio of 3HBCoA-to-synthase indicates that the equivalent enzyme size representing a single catalytic site was 160 kD. Based on 1:1 stoichiometry of PhaC and PhaE in PhaECEs, this size fits the combination of two PhaC and two PhaE subunits which have a formula weight of 163 kD. From the earlier observation that a homodimer of PhaCRe produces a single polymer chain,41 a revised mechanistic model has been proposed for the polymerization reaction catalyzed by type I synthases.41 This model emphasizes that two identical C319 residues from each subunit form a single catalytic site. For type III synthase PhaCECv, a similar model has also been proposed,13 and C149 in the active site is known to be involved in covalent catalysis in that synthase.24 This residue is conserved in PhaECEs, which suggests that two C149 residues are also required for catalysis by this synthase, analogous to the situation in PhaCRe.

48


As noted above, the two classes of PHA synthases have many similarities including that both exist in two different physical forms. However, there must also be some fundamental differences in the mechanism of polymerization between PhaECES and PhaCRe. For PhaCRe, dimerization is required to provide the two C319 residues for the catalytic site.36,41 On the other hand, the two C149 residues needed to form a single catalytic site in PhaECEs are already included in a hexamer, which contains three pairs of 2(PhaECEs). The reasons for the differences in reaction kinetics between the hexamer and the dodecamer remain to be explained by further investigation. A detailed understanding of the in vitro polymerization with PHA synthases will ultimately provide greater insights into the control of substrate specificity and reaction kinetics within PHA accumulating cells. Acknowledgment. This work was supported by a grant to R.W.L. and S.G. from the New Energy and Industrial Technology Development Organization, Japan. The authors are grateful to Takagi Yasuo for technical assistance. References and Notes (1) Anderson, A. J.; Dawes, E. A. Microbiol. ReV. 1990, 54, 450-472. (2) Bognar, A.; Desrosiers, L.; Libman, M.; Newman, E. B. J. Bacteriol. 1982, 152, 706-713. (3) Bradford, M. M. Anal.Biochem. 1976, 72, 248-254. (4) Brandl, H.; Gross, R. A.; Lenz, R. W.; Fuller, R. C. Appl. EnViron. Microbiol. 1988, 54, 1977-1982. (5) de Smet, M. J.; Eggink, G.; Witholt, B.; Kingma, J.; Wynberg, H. J. Bacteriol. 1983, 154, 870-878. (6) Doi, Y.; Kawaguchi, Y.; Koyama, N.; Nakamura, S.; Hiramitsu, M.; Yoshida, Y.; Kimura, H. FEMS Microbiol. ReV. 1992, 103, 103108. (7) Fukui, T.; Yoshimoto, A.; Matsumoto, M.; Hosokawa, S.; Saito, T.; Nishikawa, H.; Tomita, K. Arch. Microbiol. 1976, 110, 149-156. (8) Gerngross, T. U.; Martin, D. P. Proc. Natl. Acad. Sci. U.S.A. 1995, 92, 6279-6283. (9) Gerngross, T. U.; Snell, K. D.; Peoples, O. P.; Sinskey, A. J.; Csuhai, E.; Masamune, S.; Stubbe, J. Biochemistry 1994, 33, 9311-9320. (10) Haywood, G. W.; Anderson, A. J.; Dawes, E. A. FEMS Microbiol. Lett. 1989, 57, 1-6. (11) Hein, S.; Tran, H.; Steinbu¨chel, A. Arch. Microbiol. 1998, 170, 162170. (12) Huisman, G. W.; Wonink, E.; Meima, R.; Kazemier, B.; Terpstra, P.; Witholt, B. J. Biol. Chem. 1991, 266, 2191-2198. (13) Jia, Y.; Kappock, J.; Frick, T.; Sinskey, A. J.; Stubbe, J. Biochemistry 2000, 39, 3927-3936. (14) Jossek, R.; Reichelt, R.; Steinbu¨chel, A. Appl. Microbiol. Biotechnol. 1998, 49, 258-266. (15) Kaneko, T.; Sato, S.; Kotani, H.; Tanaka, A.; Asamizu, E.; Nakamura, Y.; Miyajima, N.; Hirosawa, M.; Sugiura, M.; Sasamoto, S.; Kimura, T. H.; Matsuno, T. A.; Muraki, A.; Nakazaki, N.; Naruo, K.; Okumura, S.; Shimpo, S.; Takeuchi, C.; Wada, T.; Watanabe, A.;

Zhang et al. Yamada, M.; Tabata, S. DNA Res. 1996, 30, 185-209. (16) Lemoigne, M. Bull. Soc. Chim. Biol. (Paris) 1926, 8, 770-782. (17) Liebergesell, M.; Hustede, E.; Timm, A.; Steinbu¨chel, A.; Fuller, R. C.; Lenz, R. W.; Schlegel, H. G. Arch. Microbiol. 1991, 155, 415421. (18) Liebergesell, M.; Mayer, F.; Steinbu¨chel, A. Appl. Microbiol. Biotechnol. 1993, 40, 292-300. (19) Liebergesell, M.; Rahalkar, S.; Steinbu¨chel, A. Appl. Microbiol. Biotechnol. 2000, 54, 186-194. (20) Liebergesell, M.; Sonomoto, K.; Madkour, M.; Mayer, F.; Steinbu¨chel, A. Eur. J. Biochem. 1994, 226, 71-80. (21) Liebergesell, M.; Steinbu¨chel, M. A. Appl. Microbiol. Biotechnol. 1993, 38, 493-501. (22) Liebergesell, M.; Steinbu¨chel, A. Eur. J. Biochem. 1992, 209, 135150. (23) McCool, G. J.; Cannon, M. C. J. Bacteriol. 1999, 181, 585-592. (24) Mu¨h, U.; Sinskey, A. J.; Kirby, D. P.; Lane, W. S.; Stubbe, J. Biochemistry 1999, 38, 826-837. (25) Peoples, O. P.; Sinskey, A. J. J. Biol. Chem. 1989, 264, 1529815303. (26) Prieto, M. A.; Bu¨hler, B.; Jing, K.; Witholt, B.; Kessler, B. J. Bacteriol. 1999, 181, 858-868. (27) Qi, Q.; Steinbu¨chel, A.; Rehm, B. H. A. Appl. Microbiol. Biotechnol. 2000, 54, 37-43. (28) Rehm, B. H. A.; Steinbu¨chel, A. Int. J. Biol. Macromol. 1999, 25, 3-19. (29) Ren, Q.; de Roo, G.; Kessler, B.; Witholt, B. Biochem. J. 2000, 349, 599-604. (30) Sambrook, J.; Fritsch, E. F.; Maniatis, T. Molecular cloning a laboratory manual, 2nd ed.; Cold Spring Harbor Laboratory Press, Plainview, NY, 1989. (31) Schubert, P.; Steinbu¨chel, A.; Schlegel, H. G. J. Bacteriol. 1988, 170, 5837-5847. (32) Simon, E. J.; Shemin, D. J. A J. Am. Chem. Soc. 1953, 72, 149. (33) Slater, S. C.; Voige, W. H.; Dennis, D. E. J. Bacteriol. 1988, 170, 4431-4436. (34) Speikermann, P.; Rehm, B.; Kalscheuer, R.; Baumeister, D.; Steinbu¨chel, A. Arch. Microbiol. 1999, 171, 73-80. (35) Su, L.; Lenz, R. W.; Takagi, Y.; Zhang, S.; Goodwin, S.; Zhong, L.; Martin, D. P. Macromolecules 2000, 33, 229-231. (36) Wodzinska, J.; Snell, K. D.; Rhomberg, A.; Sinkey, A. J.; Biemann, K.; Stubbe, J. J. Am. Chem. Soc. 1996, 118, 6319-6320. (37) Yuan, W.; Jia, Y.; Tian, J. M.; Snell, K. D.; Muh, U.; Sinskey, A. J.; Lambalot, R. H.; Walsh, C. T.; Stubbe, J. Arch. Biochem. Biophys. 2001, 394, 87-98. (38) Zhang, S. Ph.D. Thesis, University of Montreal, 1997. (39) Zhang, S.; Kamachi, M.; Takagi, Y.; Lenz, R. W.; Goodwin, S. Appl. Microbiol. Biotechnol. 2001, 56, 131-136. (40) Zhang, S.; Kolvek, S.; Lenz, R. W.; Goodwin S. Biomacromolecules 2003, 4, 504-509. (41) Zhang, S.; Yasuo, T.; Lenz, R. W.; Goodwin, S. Biomacromolecules 2000, 1, 244-251. (42) Zhang, W.; Reading, C.; Deisseroth, R. B. Trends Genet. 1992, 8, 332.

BM034171I

Biosynthesis in Ectothiorhodospira shaposhnikovii - American

Recommend Documents