The mycobacterial binuclear iron monooxygenases require a specific chaperonin-like protein for functional expression in a heterologous host

Authors


Correspondence

K. Kino, Department of Applied Chemistry, Faculty of Science and Engineering, Waseda University, 3-4-1 Ohkubo, Shinjuku-ku, Tokyo 169-8555, Japan

Fax: +81 3 3232 3889

Tel: +81 3 5286 3211

E-mail: kkino@waseda.jp

Abstract

The mimABCD gene clusters in Mycobacterium smegmatis strain mc2155 and Mycobacterium goodii strain 12523 encode binuclear iron monooxygenases that oxidize propane and phenol. In this study, we attempted to express each mimABCD gene cluster in a heterologous host. The actinomycetous strain Rhodococcus opacus B-4, which is phylogenetically close to Mycobacterium, was selected as the host. Each mimABCD gene cluster was cloned into the RhodococcusEscherichia coli shuttle vector, pTip-QC2, and then introduced into R. opacus cells. Although whole-cell assays were performed with phenol as a substrate, the transformed R. opacus cells did not oxidize this substrate. SDS/PAGE analysis revealed that the oxygenase large subunit MimA was expressed in the insoluble fraction of R. opacus cells. We found that a gene designated mimG, which lies downstream of mimABCD, exhibits similarity in the amino acid sequence of its product with the products of genes encoding the chaperonin GroEL. When the mimG gene was cloned and coexpressed with each mimABCD gene cluster in R. opacus strain B-4, this host successfully acquired oxidation activity towards phenol. SDS/PAGE and western blotting analyses demonstrated that MimA was clearly soluble when in the presence of MimG. These results indicated that MimG played essential roles in the productive folding of MimA, and that the resulting soluble MimA protein led to the active expression of MimABCD.

Abbreviations
Bmo

butane monooxygenase

Mmo

methane monooxygenase

Prm

propane monooxygenase

Introduction

Binuclear iron monooxygenases constitute a family of proteins that contain a binuclear iron center at the active site; these enzymes introduce one oxygen atom, which is derived from molecular oxygen, into organic molecules, including alkanes, alkenes, and aromatics [1, 2]. These monooxygenases are widely distributed throughout several prokaryotic taxa, such as α-proteobacteria and γ-proteobacteria. This type of monooxygenase has only recently been found in actinobacteria, and these actinomycetous enzymes constitute a new defined subfamily within the family of binuclear iron monooxygenases. These actinomycetous monooxygenases include an alkene monooxygenase from Nocardia corallina (Rhodococcus rhodochrous) strain B-276 [3], a propene monooxygenase from Mycobacterium sp. strain M156 [4], two propane monooxygenases (Prms), one from Gordonia sp. strain TY-5 and another from Rhodococcus sp. strain RHA1 [5, 6], and a tetrahydrofuran monooxygenase from Pseudonocardia sp. strain K1 [7]. The monooxygenase gene clusters in the actinomycetes each contain four genes, which encode an oxygenase large subunit, an oxygenase small subunit, a reductase, and a coupling protein (Fig. 1). The oxygenase component, which comprises the catalytic large subunit and the structural small subunit, activates molecular oxygen by using electrons that are transferred from NAD(P)H by the reductase component. The coupling protein interacts with the oxygenase component, and is essential for full oxidation activity [8, 9]. Notably, the oxygenase component in these actinomycetous enzymes comprises two subunits in an αβ or an α2β2 quaternary structure, whereas this component in the enzymes of other bacteria, including methanotrophs and pseudomonads, comprises three subunits in an α2β2γ2 quaternary structure [1, 2].

Figure 1.

Genetic organization of the binuclear iron monooxygenase gene clusters. ORF number, gene name or both are shown. The GenBank accession numbers of the sequences are as follows: Mycobacterium smegmatis strain mc2155, NC_008596; Rhodococcus sp. strain RHA1, NC_008268; Gordonia sp. strain TY-5, AB112920; Bradyrhizobium japonicum strain USDA 110, NC_004463; Rhodobacter sphaeroides strain 2.4.1, NC_007493.

More recently, we identified the binuclear iron monooxygenase gene clusters in Mycobacterium smegmatis strain mc2155 and Mycobacterium goodii strain 12523 [10]. The gene clusters each comprise four genes – mimA, mimB, mimC, and mimD – which encode an oxygenase large subunit, a reductase, an oxygenase small subunit, and a coupling protein, respectively (Fig. 1). These gene clusters were found to play essential roles in propane and acetone metabolism in these mycobacterial strains. Interestingly, these gene clusters are also responsible for the regioselective oxidation of phenol to hydroquinone, which is of biotechnological importance [10]. The multicomponent monooxygenases encoded by these mimABCD gene clusters exhibit high amino acid similarities with those encoded by the prmABCD gene clusters of Rhodococcus sp. strain RHA1 and Gordonia sp. strain TY-5 [10]. PrmABCD of strain TY-5 was also found to oxidize phenol in addition to propane [5, 10]. The functions of the Mim monooxygenases and Prms were identified by gene deletion analysis; however, heterologous expression of these monooxygenases has not been reported.

Here, we attempted to reconstitute MimABCD activity in heterologous hosts. We selected a Rhodococcus strain as the host, because attempts to express either mimABCD gene cluster in Escherichia coli were unsuccessful. However, introduction of either mimABCD gene cluster alone into the Rhodococcus strain did not endow this host with monooxygenase activity. We reasoned that another factor was required for the functional expression of MimABCD; therefore, we searched for a candidate gene within the genome sequence of M. smegmatis strain mc2155. These attempts led to the identification of a specific chaperonin-like protein that was essential for the expression of functional mycobacterial binuclear iron monooxygenases in this heterologous host.

Results

Expression of mimABCD in E. coli

We first attempted to express the mimABCD gene clusters in E. coli. For convenience, we named the gene cluster from M. smegmatis strain mc2155 mimABCDsm, and that from M. goodii strain 12523 mimABCDgo [10]. Each mimABCD gene cluster was cloned into the pET21a vector. Within the resulting constructs (pETmimABCDsm and pETmimABCDgo; Table 1), the mimA gene had a ribosome binding site derived from the vector, whereas the other three genes had their natural ribosome binding sites. pETmimABCDsm and pETmimABCDgo were separately introduced into E. coli BL21(DE3) cells. The expression of the mimABCD gene clusters was induced with isopropyl thio-β-d-galactoside under the control of the T7 promoter. Although whole-cell assays were performed with phenol as a substrate, the transformed E. coli cells did not oxidize this substrate (data not shown). SDS/PAGE analysis revealed the presence of a band corresponding to MimA (63 kDa) in the whole-cell sample from each transformed E. coli strain, but this band was not evident in the soluble fraction (Fig. 2). It was reported that the oxygenase large subunit of several binuclear iron monooxygenases was also insoluble and inactive in E. coli hosts [11-13]. Moreover, no band corresponding to MimB (39 kDa), MimC (42 kDa) or MimD (13 kDa) was detected (Fig. 2). These results indicated difficulties in actively expressing mycobacterial MimABCD in the Gram-negative E. coli, which is phylogenetically distant from the Gram-positive actinomycetous Mycobacterium.

Table 1. Bacterial strains and plasmids used in this study
Strain or plasmidCharacteristicsReference or source
Strains
Mycobacterium smegmatis mc2155Wild type, ATCC 700084ATCC
Mycobacterium goodii 12523Wild type [10]
Rhodococcus opacus B-4Wild type, NBRC 108011, host used for expression [15]
Escherichia coli DH5αHost used for cloningTakara Bio
E. coli BL21(DE3)Host used for expressionNovagen
Plasmids
pETmimABCDsmpET21a containing mimABCDsm of M. smegmatis strain mc2155 [10]
pETmimABCDgopET21a containing mimABCDgo of M. goodii strain 12523 [10]
pTip-QC2RhodococcusE. coli shuttle vector, Chlr [14]
pTip-RT2RhodococcusE. coli shuttle vector, Tetr [14]
pTipmimABCDsmpTip-QC2 containing mimABCDsm under the control of the tipA promoterThis study
pTipmimABCDgopTip-QC2 containing mimABCDgo under the control of the tipA promoterThis study
pTipmimGsmpTip-RT2 containing mimGsm under the control of the tipA promoterThis study
pTipmimAsmflagpTip-QC2 containing mimAsm with an in-frame C-terminal FLAG-tag sequenceThis study
pTipmimBsmflagpTip-QC2 containing mimBsm with an in-frame C-terminal FLAG-tag sequenceThis study
pTipmimCsmflagpTip-QC2 containing mimCsm with an in-frame C-terminal FLAG-tag sequenceThis study
pTipmimDsmflagpTip-QC2 containing mimDsm with an in-frame C-terminal FLAG-tag sequenceThis study
Figure 2.

SDS/PAGE analysis of expression of mimABCD in Escherichia coli. The samples that were prepared from E. coli BL21(DE3) cells carrying pETmimABCDsm (mimABCDsm), pETmimABCDgo (mimABCDgo) or pET21a (empty vector) were loaded onto a polyacrylamide gel (15%). S, soluble-fraction sample; W, whole-cell sample. The molecular mass corresponding to MimA is indicated by an arrow.

Expression of mimABCD in Rhodococcus opacus

We then attempted to express the mimABCD gene clusters in a Gram-positive actinomycete, R. opacus, which is phylogenetically close to Mycobacterium. Each mimABCD gene cluster was cloned into the RhodococcusE. coli shuttle vector, pTip-QC2 [14], which belongs to the pAL500 family of plasmids, which replicate by a θ-type mechanism. The resulting plasmids, pTipmimABCDsm for M. smegmatis strain mc2155 and pTipmimABCDgo for M. goodii strain 12523 (Table 1), were separately introduced into R. opacus strain B-4 cells [15]. In this Rhodococcus host, expression of the mimABCD gene clusters was under the control of the tipA promoter and induced by thiostrepton. The transformed R. opacus cells, however, again did not oxidize the substrate phenol (Fig. 3). SDS/PAGE analysis revealed that each MimA protein was insoluble even in these R. opacus cells (Fig. 4). These observations indicated that the absence of oxidation activity was at least partly caused by the insolubility of the MimA component. Although bands corresponding to MimB (39 kDa), MimC (42 kDa) and MimD (13 kDa) were not evident in SDS/PAGE analysis (Fig. 3), western blotting analysis proved that these proteins were expressed in the soluble fraction, as described later (see Fig. 5).

Figure 3.

Monooxygenase activity of transformed R. opacus cells towards phenol. Whole cells of Rhodococcus opacus strain B-4 carrying pTipmimABCDsm alone (mimABCDsm), pTipmimABCDgo alone (mimABCDgo), pTipmimABCDsm and pTipmimGsm (mimABCDsm + mimGsm), pTipmimABCDgo and pTipmimGsm (mimABCDgo + mimGsm) or pTipmimGsm alone (mimGsm) were reacted with phenol for 24 h, and the monooxygenation product hydroquinone was quantified with HPLC. Bars represent the averages of three independent experiments, and error bars represent the standard deviations from the means.

Figure 4.

SDS/PAGE analysis of expression of mimABCD in Rhodococcus opacus. The samples that were prepared from R. opacus strain B-4 cells carrying pTipmimABCDsm alone (mimABCDsm), pTipmimABCDgo alone (mimABCDgo), pTipmimABCDsm and pTipmimGsm (mimABCDsm + mimGsm), pTipmimABCDgo and pTipmimGsm (mimABCDgo + mimGsm) or pTipmimGsm alone (mimGsm) were loaded onto a polyacrylamide gel (7.5%). S, soluble-fraction sample; W, whole-cell sample. The molecular masses corresponding to MimA and MimG are indicated by arrows.

Figure 5.

Western blotting analysis of expression of mimABCD in Rhodococcus opacus. (A) The samples prepared from R. opacus strain B-4 cells expressing MimAsm–FLAG alone (mimA), MimAsm–FLAG and MimGsm (mimA + mimG) or MimGsm alone (mimG) were subjected to SDS/PAGE; the resulting western blots were treated with antibody against FLAG. (B–D) With a similar technique, expression of mimBsm (B) mimCsm (C) and mimDsm (D) was analyzed. The concentration of acrylamide was adjusted to 7.5% for the analysis of MimA, to 10% for MimB and MimC, and to 20% for MimD. Protein profiles of SDS/PAGE and western blotting are shown in the left panel and the right panel, respectively. S, soluble-fraction sample; W, whole-cell sample.

Analysis of the genome sequence of M. smegmatis strain mc2155

We reasoned that another factor was required for the enzymatic function of the mimABCD gene products; therefore, we searched for a candidate gene within the genome sequence of M. smegmatis strain mc2155. Previously, we identified a regulator gene, mimRsm, in the region upstream of mimABCDsm [16], and we demonstrated that the mimRsm gene product is an acetone-responsive positive regulator of mimABCDsm gene cluster transcription. Moreover, four ORFs – Msmeg_1975, Msmeg_1976, Msmeg_1977, and Msmeg_1978 – lie downstream of the mimABCDsm gene cluster (Fig. 1). Msmeg_1975 is annotated as an amidohydrolase, and Msmeg_1976 is annotated as a hypothetical protein. Msmeg_1977 has significant amino acid similarity (74%) with the dehydrogenase (Adh1) that converts 2-propanol to acetone in Gordonia sp. strain TY-5 [5]. Furthermore, we found that Msmeg_1978 has appreciable amino acid similarity with the chaperonin GroEL. In general, GroEL is involved in the productive folding of proteins. The genome sequence of M. smegmatis strain mc2155 has two other copies of the groEL gene, Msmeg_0880 and Msmeg_1583 [17]. Msmeg_1978 shares 52% and 50% amino acid identity with Msmeg_0880 and Msmeg_1583, respectively. Msmeg_0880 shares 65% amino acid identity with Msmeg_1583. The gene product corresponding to Msmeg_0880 is reportedly essential for growth [18], whereas that corresponding to Msmeg_1583 was required for biofilm formation [19]. In contrast, the function of the Msmeg_1978 ORF has not been determined. We hypothesized that Msmeg_1978, which lies immediately downstream of mimABCDsm, might be involved in the productive folding of MimA, and thus in the formation of the active MimABCD complex.

Coexpression of mimABCD with mimG in R. opacus

We designated the gene corresponding to Msmeg_1978 mimGsm, and coexpressed mimGsm along with each mimABCD gene cluster in R. opacus cells. The mimGsm gene was cloned into the pTip-RT2 vector [14], which belongs to the pIJ101/pJV1 family of plasmids, which replicate by a rolling cycle-type mechanism; this vector is compatible with the pTip-QC2 vector into which each mimABCD gene cluster had been cloned. The resulting plasmids pTipmimGsm (Table 1) and pTipmimABCDsm or pTipmimABCDgo were simultaneously introduced into R. opacus cells. When whole-cell assays were performed with phenol as a substrate, the transformed R. opacus cells did oxidize this substrate (Fig. 3). Within 24 h of the reaction, cells carrying pTipmimABCDsm and pTipmimGsm produced 1.6 mm hydroquinone from 10 mm phenol, and cells carrying pTipmimABCDgo and pTipmimGsm produced 2.3 mm hydroquinone. The initial rates of hydroquinone production by cells carrying pTipmimABCDsm and pTipmimGsm and cells carrying pTipmimABCDgo and pTipmimGsm were estimated to be 0.55 μmol and 0.80 μmol, respectively, per gram dry cell weight per minute. SDS/PAGE analysis revealed that the relative intensity of a band corresponding to MimGsm (58 kDa) was strong in the soluble fraction of cells carrying pTipmimABCDsm and pTipmimGsm and cells carrying pTipmimABCDgo and pTipmimGsm, although this band overlapped with a band that was derived from the host cells (Fig. 4). Furthermore, SDS/PAGE analysis indicated that coexpression of the mimGsm gene caused MimA to be expressed in the soluble fraction of R. opacus cells (Fig. 4).

Western blotting analysis of MimG effects on MimABCD expression

The effects of MimG on MimABCD expression were examined in detail by western blotting analysis. The genes mimAsm, mimBsm, mimCsm and mimDsm were individually cloned, along with an in-frame C-terminal FLAG-tag sequence, into the pTip-QC2 vector (Table 1). Each resulting fusion gene was coexpressed with mimGsm in R. opacus cells. An antibody against FLAG was used to probe western blots, and this analysis revealed that, in the absence of MimGsm, MimAsm was expressed exclusively as the insoluble form in R. opacus cells (Fig. 5A). In contrast, MimAsm was clearly soluble when in the presence of MimGsm (Fig. 5A). Moreover, we confirmed that MimBsm, MimCsm and MimDsm were evident in the soluble fraction of R. opacus cells in the absence and the presence of MimGsm (Fig. 5B–D). These results indicated that MimG played essential roles in the productive folding of MimA, and that the resulting soluble MimA led to the active expression of MimABCD.

Discussion

Here, we succeeded in expressing functional mycobacterial binuclear iron monooxygenases in a heterologous system. Introduction of only the mimABCD gene cluster into the Rhodococcus strain was insufficient to reconstitute MimABCD activity in this host. Fortunately, our initial difficulty in reconstituting functional monooxygenase activity in this host led to the identification of a novel chaperonin-like protein, MimGsm, which is encoded in the region downstream of mimABCDsm. To our knowledge, this study is the first to demonstrate that the active expression of binuclear iron monooxygenases in a heterologous host is dependent on the coexpression of a specific chaperonin-like protein.

We experimentally demonstrated that MimG plays essential roles in the productive folding of MimA, and that this correct folding led to the successful functional expression of MimABCD (Fig. 3). MimGsm (Msmeg_1978) of M. smegmatis strain mc2155 shares 49% amino acid identity with a typical chaperonin, GroEL of E. coli strain K-12 [20]. We also found that a homolog of mimGsm exists in M. goodii strain 12523 by PCR with the same oligonucleotide primers as those used for amplification of the mimGsm gene (Table 2, and data not shown). The genome sequence of the host strain R. opacus B-4 (GenBank accession number: NC_008268) contains two copies of the groEL gene, ROP_18630 and ROP_62500, whose gene products share 53% and 51% amino acid identity with MimGsm, respectively. These low similarities suggest that the endogenous GroELs in the R. opacus host did not function as a substitute for MimGsm in the heterologous expression system. GroELs generally require a cochaperonin, GroES, to assist in the folding of target proteins. The genome sequence of M. smegmatis strain mc2155 has two other copies of the groEL gene, Msmeg_0880 and Msmeg_1583, in addition to mimGsm, whereas this M. smegmatis genome sequence contains only one copy of the groES gene, Msmeg_1582 [17]. Conceivably, this groES gene product cooperates with MimGsm to correctly fold MimA in M. smegmatis strain mc2155. Furthermore, we found that the genome sequence of the host strain R. opacus B-4 also contains one copy of the groES gene, ROP_62490, whose gene product shares 90% amino acid identity with the mycobacterial GroES; this R. opacus groES gene product might enable MimGsm to function in the heterologous host. SDS/PAGE and western blotting analyses of the mimABCD gene products in R. opacus cells revealed that only the oxygenase large subunit, MimA, was intrinsically insoluble, and that coexpression of MimGsm caused MimA to be expressed as the soluble form (Figs 4 and 5A). Interestingly, the oxygenase small subunit, MimC, was soluble irrespective of the absence or the presence of MimGsm (Fig. 5C), although the large and small subunits are known to be paralogous proteins and to be similar in quaternary structure [1]. Only the large subunits of the monooxygenases contain a binuclear iron center where molecular oxygen is activated. The two closely spaced iron ions might generate electrostatic repulsion, as was suggested for binuclear copper ions in tyrosinases [21]. This electrostatic repulsion might cause MimA to be unstable in the maturation process of the active center; MimG may support MimA in this process, enabling it to escape inactivation.

Table 2. Oligonucleotide primers used in this study. Restriction sites are underlined
PrimerSequence (5′- to 3′)Restriction site
mimG-F TTCCATATGGCCAAAGAACTGCGGTT NdeI
mimG-R CGGAATTCTCAGTAGATGTTGGACGGGCGGA EcoRI
mimA-F TTCCATATGAGCAGACAAAGCCTGAC NdeI
mimAflag-R CGGAATTCACTTGTCGTCATCGTCTTTGTAATCGGCCGGGACCCCGCCGCGGCGGGGTC EcoRI
mimB-F TTCCATATG GCGGATTCGCACAAGAT NdeI
mimBflag-R CGGAATTCACTTGTCGTCATCGTCTTTGTAATCCTGGTCGAAGATGGGGCTGGTGAAGCTG EcoRI
mimC-F TTCCATATGTCAGCCCCTGAGAAGCCAA NdeI
mimCflag-R CGGAATTCACTTGTCGTCATCGTCTTTGTAATCTTTGTTCAACTCCTCGGGGATGTCCAGAT EcoRI
mimD-F TTCCATATGAGCAGCATGCAGTTCGG NdeI
mimDflag-R CGGAATTCACTTGTCGTCATCGTCTTTGTAATCCTTCGCGGTGAGGTCGAAGCCGAT EcoRI

The mimABCD gene clusters of the Mycobacterium strains exhibit high sequence similarity with the prmABCD gene clusters of Rhodococcus sp. strain RHA1 and Gordonia sp. strain TY-5. We also found that the mimGsm homologs RHA1_ro00448 and orf3 lie downstream of the prmABCD gene clusters of Rhodococcus sp. strain RHA1 and Gordonia sp. strain TY-5, respectively (Fig. 1). MimGsm shares 84% and 74% amino acid identity with the gene products corresponding to RHA1_ro00448 and orf3, respectively. These facts strongly suggest that the MimGsm homologs also play essential roles in the formation of the active binuclear iron monooxygenases in these actinomycetes. Furthermore, analysis of published genome sequences revealed that homologous gene clusters encoding binuclear iron monooxygenase exist not only in several actinobacteria, but also in some proteobacteria, including Bradyrhizobium japonicum and Rhodobacter sphaeroides, and that homologs of mimGsm also lie downstream of these monooxygenase gene clusters (Fig. 1). We reasoned that the monooxygenase gene clusters might be transferred, together with the groEL genes, by horizontal transmission to the genomes of actinomycetes and other bacteria. However, when the amoABCD gene cluster of the actinomycetous strain N. corallina B-276 was introduced into Streptomyces lividans, this host acquired AmoABCD activity without coexpression of a chaperonin gene [13]. The amoABCD gene cluster exhibits low similarity in amino acid sequence (~ 40%) with the mimABCD gene clusters. It is possible that AmoABCD did not require a specific chaperonin for functional expression, or that the chaperonin was supplied to this monooxygenase by the host strain.

Besides the reports on actinomycetes, only a few reports focused on a Thauera strain and methanotrophs have indicated that chaperonins are involved in the function of monooxygenases within this family. In a Thauera butanovora strain, previously referred to as Pseudomonas butanovora, the bmoXYBZDC gene cluster encoding a butane monooxygenase (Bmo) lies upstream of the bmoG gene, which encodes a GroEL-like protein; Kurth et al. [22] reported that a bmoG-deficient mutant lost the ability to grow on C2–C8 n-alkanes. In Methylosinus trichosporium strain OB3b, the mmoXYBZDC gene cluster, which encodes a methane monooxygenase (Mmo), is preceded by the mmoG gene; Stafford et al. [23] reported that a mmoG-deficient mutant could not express this monooxygenase. The Mmo has been functionally expressed in homologous hosts [24, 25], as well as in heterologous hosts, including Pseudomonas, Agrobacterium, and Rhizobium strains; however, only relatively low oxidation activities were obtained when this monooxygenase was expressed in the heterologous hosts [26, 27]. BmoG may be involved in a process after Bmo gene transcription [22], whereas MmoG has been shown to play a role in Mmo gene transcription [23, 28]. The function of the GroEL-like proteins in methanotrophs may be relatively complex. In contrast, several binuclear iron monooxygeases, including phenol and toluene monooxygenases from Gram-negative bacteria, were readily expressed as functional proteins without coexpression of a chaperonin, even in E. coli [29-32]. Conceivably, distinct and sophisticated systems of monooxygenase expression have evolved within separate lineages of microorganism, so that the microorganisms can adapt to variable habitats.

Finally, our approach has provided novel insights into the molecular mechanisms that underlie the expression system of the mycobacterial binuclear iron monooxygenases. The heterologous expression system presented here may facilitate detailed biochemical characterization of these monooxygenases, and lead to practical biocatalytic applications for these enzymes, which catalyze interesting and biotechnologically important reactions. Furthermore, the findings of this study might be helpful in studies concerning the expression of other binuclear iron monooxygenases, especially the heterologous expression of Mmos, which is also technically challenging [33, 34].

Experimental procedures

Bacterial strains, plasmids, and cultivation media

The bacterial strains and plasmids that were used or constructed in this study are listed in Table 1. The bacteria were grown in LB medium, which contained (per liter) Bacto tryptone (10 g), Bacto yeast extract (5 g), and NaCl (10 g) (pH 7.0).

Construction of mimABCD and mimG expression plasmids

The plasmids used for expression of the mimABCD gene clusters in R. opacus cells were constructed by use of the pTip-QC2 vector (Table 1). The mimABCDsm and mimABCDgo gene clusters were cut from pETmimABCDsm and pETmimABCDgo (Table 1), respectively, by digestion with NdeI and EcoRI. Each gene cluster was subsequently inserted into the pTip-QC2 vector, which had been digested with NdeI and EcoRI. The resulting plasmids, pTipmimABCDsm and pTipmimABCDgo, were amplified in E. coli DH5α cells.

The pTip-RT2 (Table 1) vector was used to construct the mimG gene expression plasmid. Two oligonucleotide primers, mimG-F and mimG-R (Table 2), were designed to amplify the mimGsm gene that corresponds to the Msmeg_1978 ORF, based on the genome sequence of M. smegmatis strain mc2155 (GenBank accession number: NC_008596). PCR was used to amplify the region between the two oligonucleotide primers from genomic DNA of M. smegmatis strain mc2155. This amplified DNA fragment was digested with NdeI and EcoRI, and then inserted into the pTip-RT2 vector. The resulting plasmid, pTipmimGsm, was amplified in E. coli DH5α cells.

After the correct generation of the pTip plasmids had been confirmed by sequencing, these plasmids were introduced into R. opacus cells by electroporation.

Construction of expression plasmids for individual Mim components

The genes mimAsm, mimBsm, mimCsm and mimDsm were individually cloned, along with an in-frame C-terminal FLAG-tag sequence, into the pTip-QC2 vector (Table 1). Two oligonucleotide primers, mimA-F and mimAflag-R (Table 2), were designed to amplify the mimAsm gene. The sequence encoding the FLAG-tag (Asp-Tyr-Lys-Asp-Asp-Asp-Asp-Lys) was included in the C-terminal primer mimAflag-R. The region between the two oligonucleotide primers was amplified from the pETmimABCDsm plasmid by PCR. This amplified DNA fragment was digested with NdeI and EcoRI, and then inserted into the pTip-QC2 vector. The resulting plasmid, pTipmimAsmflag, was amplified in E. coli DH5α cells. With a similar technique, the plasmids pTipmimBsmflag, pTipmimCsmflag and pTipmimDsmflag were constructed by use of the oligonucleotide primers listed in Table 2.

Preparation of whole cells

The transformed E. coli BL21(DE3) cells carrying pETmimABCDsm or pETmimABCDgo were cultivated at 25 °C in LB medium supplemented with ampicillin (100 μg·mL−1). After cultivation for 12 h, isopropyl thio-β-d-galactoside (1 mm) was added to the medium, and cultivation was continued for an additional 12 h. Cells were harvested by centrifugation at 15 000 g for 10 min, washed with potassium phosphate buffer (50 mm, pH 7.5) containing glycerol (10%, v/v), and stored at − 80 °C until use.

The transformed R. opacus strain B-4 cells were cultivated at 30 °C in LB medium supplemented with chloramphenicol (17 μg·mL−1) and/or tetracycline (8 μg·mL−1). When the cell growth reached a D600 nm of 0.6–0.8, thiostrepton (1 μg·mL−1) was added to the medium, and cultivation was continued for an additional 24 h. Cells were harvested by centrifugation at 15 000 g for 10 min, washed with potassium phosphate buffer (50 mm, pH 7.5) containing glycerol (10%, v/v), and stored at − 80 °C until use.

Reaction with whole cells

In whole-cell assays with transformed E. coli cells, the reaction mixture (250 μL) contained cells of the E. coli strain (10 g of dry cell weight per liter), the substrate phenol (1 mm), ethanol (1%, v/v) and potassium phosphate buffer (50 mm, pH 7.5) containing glycerol (10%, v/v). In whole-cell assays with transformed R. opacus cells, the reaction mixture (250 μL) contained cells of the R. opacus strain (2 g of dry cell weight per liter), the substrate phenol (10 mm), ethanol (1%, v/v) and aqueous basal medium [35] containing glucose (5 g·L−1).

Product analysis

HPLC analysis was performed with an HPLC system (1100 series; Agilent, Palo Alto, CA, USA) with an XTerra MS C18 IS column (4.6 × 20 mm; particle size, 3.5 μm; Waters, Milford, MA, USA), as described previously [36]. Following the reaction, methanol (250 μL) was added to the reaction mixture. The resulting sample (10 μL) was then injected into the HPLC system. Mobile phases A and B were composed of acetonitrile/methanol/potassium phosphate buffer (10 mm, pH 2.7) at a ratio of 2.5 : 2.5 : 95, and of methanol, respectively. The samples were eluted with 0% phase B for 3 min, and then with a linear gradient of 0–80% phase B for 6 min at a flow rate of 1 mL·min−1. The reaction product, hydroquinone, was detected spectrophotometrically at a wavelength of 290 nm.

SDS/PAGE and western blotting analyses

The expression levels of Mim proteins were examined with SDS/PAGE and western blotting analyses. Frozen cells were suspended in potassium phosphate buffer (50 mm, pH 7.5) containing glycerol (10%, v/v), and were disrupted with an ultraoscillator. The disrupted cells were used for the preparation of whole-cell samples that included both soluble and insoluble proteins. After centrifugation at 15 000 g for 30 min at 4 °C, the resulting supernatant was used for the preparation of soluble-fraction samples. The protein concentration was measured with a Coomassie protein assay kit (Pierce, Rockford, IL, USA) with a BSA standard [37]. The samples (2.5–5 μg of protein) were treated with SDS, and then loaded onto a polyacrylamide gel. The concentration of acrylamide was adjusted to 7.5–20%, depending on the molecular masses of Mim proteins. In western blotting analysis, proteins in the gel after SDS/PAGE were transferred onto a poly(vinylidene difluoride) membrane via electroblotting. This membrane was treated with mouse anti-FLAG M2 Ig (Sigma-Aldrich, St Louis, MO, USA) at a dilution of 1 : 5000 as the primary antibody, and alkaline phosphatase-conjugated goat anti-(mouse IgG) (Promega, Madison, WI, USA) at a dilution of 1 : 5000 as the secondary antibody. The membrane was subsequently stained with Western Blue (Promega), which is a mixture of Nitro Blue tetrazolium and 5-bromo-4-chloro-3-indolylphosphate.

Acknowledgements

We thank H. Ohtake and K. Honda (Osaka University) for the gift of R. opacus strain B-4. We also thank T. Tamura (National Institute of Advanced Industrial Science and Technology) for the gift of the pTip vectors.

Ancillary

Advertisement