The Syringate O-Demethylase Gene of Sphingobium sp. Strain SYK-6 Is Regulated by DesX, while Other Vanillate and Syringate Catabolism Genes Are Regulated by DesR

Syringate is a major degradation product in the microbial and chemical degradation of syringyl lignin. Along with other low-molecular-weight aromatic compounds, syringate is produced by chemical lignin depolymerization. Converting this mixture into value-added chemicals using bacterial metabolism (i.e., biological funneling) is a promising option for lignin valorization. To construct an efficient microbial lignin conversion system, it is necessary to identify and characterize the genes involved in the uptake and catabolism of lignin-derived aromatic compounds and to elucidate their transcriptional regulation. In this study, we found that the transcription of desA, encoding syringate O-demethylase in SYK-6, is regulated by an IclR-type transcriptional regulator, DesX. The findings of this study, combined with our previous results on desR (encoding a MarR transcriptional regulator that controls the transcription of ligM and desB), provide an overall picture of the transcriptional-regulatory systems for syringate and vanillate catabolism in SYK-6.

ABSTRACT

Syringate and vanillate are the major metabolites of lignin biodegradation. In Sphingobium sp. strain SYK-6, syringate is O demethylated to gallate by consecutive reactions catalyzed by DesA and LigM, and vanillate is O demethylated to protocatechuate by a reaction catalyzed by LigM. The gallate ring is cleaved by DesB, and protocatechuate is catabolized via the protocatechuate 4,5-cleavage pathway. The transcriptions of desA, ligM, and desB are induced by syringate and vanillate, while those of ligM and desB are negatively regulated by the MarR-type transcriptional regulator DesR, which is not involved in desA regulation. Here, we clarified the regulatory system for desA transcription by analyzing the IclR-type transcriptional regulator desX, located downstream of desA. Quantitative reverse transcription (RT)-PCR analyses of a desX mutant indicated that the transcription of desA was negatively regulated by DesX. In contrast, DesX was not involved in the regulation of ligM and desB. The ferulate catabolism genes (ferBA), under the control of a MarR-type transcriptional regulator, FerC, are located upstream of desA. RT-PCR analyses suggested that the ferB-ferA-SLG_25010-desA gene cluster consists of the ferBA operon and the SLG_25010-desA operon. Promoter assays revealed that a syringate- and vanillate-inducible promoter is located upstream of SLG_25010. Purified DesX bound to this promoter region, which overlaps an 18-bp inverted-repeat sequence that appears to be essential for the DNA binding of DesX. Syringate and vanillate inhibited the DNA binding of DesX, indicating that the compounds are effector molecules of DesX.

IMPORTANCE Syringate is a major degradation product in the microbial and chemical degradation of syringyl lignin. Along with other low-molecular-weight aromatic compounds, syringate is produced by chemical lignin depolymerization. Converting this mixture into value-added chemicals using bacterial metabolism (i.e., biological funneling) is a promising option for lignin valorization. To construct an efficient microbial lignin conversion system, it is necessary to identify and characterize the genes involved in the uptake and catabolism of lignin-derived aromatic compounds and to elucidate their transcriptional regulation. In this study, we found that the transcription of desA, encoding syringate O-demethylase in SYK-6, is regulated by an IclR-type transcriptional regulator, DesX. The findings of this study, combined with our previous results on desR (encoding a MarR transcriptional regulator that controls the transcription of ligM and desB), provide an overall picture of the transcriptional-regulatory systems for syringate and vanillate catabolism in SYK-6.

INTRODUCTION

Lignin, a major component of plant cell walls, is the most abundant aromatic polymer in nature; thus, its decomposition is essential for the Earth’s carbon cycle (1). Lignin is generated by the oxidative coupling of three types of p-hydroxyphenylpropanoids, called monolignols: coniferyl alcohol, sinapyl alcohol, and p-coumaryl alcohol (2). Lignin structure differs depending on the type of plant; gymnosperm (softwood) lignins mainly contain guaiacyl (G) units derived from coniferyl alcohol; angiosperm (hardwood) lignins mainly contain syringyl (S) units derived from sinapyl alcohol and G units; and monocot (grass) lignins contain G units, S units, and H units derived from p-coumaryl alcohol (3, 4). Because lignin is abundant and renewable, it holds great potential as a bioresource. One option to realize this potential is through biological funneling, a process that uses bacterial catabolic functions to transform heterogeneous mixtures of low-molecular-weight aromatic compounds produced by chemical lignin depolymerization into valuable platform chemicals (5–7).

Sphingobium sp. strain SYK-6 is an aerobic alphaproteobacterium with the best-characterized catabolic systems for lignin-derived aromatic compounds (7, 8). As more details on the SYK-6 catabolic system and its regulation emerge, our understanding of bacterial lignin catabolism is enhanced, and such information can also be used to transform lignin through biological funneling (7–11). SYK-6 is capable of utilizing various types of lignin-derived biaryls (e.g., β-aryl ether, biphenyl, phenylcoumaran, and diarylpropane) and monoaryls (e.g., ferulate, vanillin, and syringaldehyde) as its sole carbon and energy sources (7, 8). These aromatic compounds, derived from G and S units, are degraded through vanillate (VA) and syringate (SA), respectively. Both are common intermediate metabolites in the biodegradation of lignin. VA is then converted to protocatechuate (PCA) by tetrahydrofolate (H₄folate)-dependent VA/3-O-methyl gallate (3MGA) O-demethylase (LigM). The resulting PCA is further metabolized via the PCA 4,5-cleavage pathway that includes a step catalyzed by PCA 4,5-dioxygenase (LigAB) (Fig. 1) (12). On the other hand, SA is converted to 3MGA by H₄folate-dependent SA O-demethylase (DesA), whose sequence is 49% similar to that of LigM in SYK-6 (Fig. 1). The resulting 3MGA is metabolized by branching into three pathways; among them, the gallate (GA) cleavage pathway, which is involved crucially in the process, includes O demethylation by LigM and ring cleavage by GA dioxygenase (DesB) (Fig. 1) (13).

FIG 1.

Transcriptional regulation of catabolism of vanillate and syringate in Sphingobium sp. SYK-6. The transcriptional regulation of ligM and desB by DesR (27) and the regulation of the PCA 4,5-cleavage genes by LigR (26) are highlighted in blue and gray, respectively. The transcriptional regulation of desA by DesX investigated in this study is highlighted in red. The enzymes were as follows: LigM, vanillate/3MGA O-demethylase; LigA and LigB, small and large subunits, respectively, of PCA 4,5-dioxygenase; LigC, CHMS (4-carboxy-2-hydroxymuconate-6-semialdehyde) dehydrogenase; LigI, PDC (2-pyrone-4,6-dicarboxylate) hydrolase; LigU, OMA (4-oxalomesaconate) delta-isomerase; LigJ, KCH (2-keto-4-carboxy-3-hexenedioate) hydratase; LigK, CHA (4-carboxy-4-hydroxy-2-oxoadipate) aldolase; DesA, syringate O-demethylase; DesZ, 3MGA 3,4-dioxygenase; DesB, gallate dioxygenase. Transcriptional regulators were as follows: DesX, IclR-type regulator; DesR, MarR-type regulator; LigR, LysR-type regulator. TCA, tricarboxylic acid.

The bacterial degradation of VA through the pathway composed of an oxygenase-type VA O-demethylase (VanAB) and the PCA 3,4-cleavage pathway enzymes has been mainly investigated in Pseudomonas strains (14, 15), Acinetobacter baylyi ADP1 (16), Caulobacter crescentus (17, 18), Corynebacterium glutamicum ATCC 13032 (19), and Rhodococcus jostii RHA1 (20). In A. baylyi ADP1, C. crescentus, and C. glutamicum ATCC 13032, the transcriptional regulation of vanAB is negatively regulated by a GntR-type, a GntR-type, and a PadR-like transcriptional regulator, respectively, all of which are called VanR (17, 21, 22). VA was determined to be an effector molecule for the last two systems (17, 22). The degradation of SA by bacteria other than SYK-6 has recently been reported in Novosphingobium aromaticivorans DSM 12444 (23), Microbacterium sp. strain RG1 (24), and Pseudomonas sp. strain NGC7 (25). While the SA catabolic-pathway genes have been identified in DSM 12444 and predicted in RG1, the regulatory system of SA catabolism has not been studied in any bacterium.

In regard to the transcriptional regulation involved in VA and SA catabolism in SYK-6, we have reported that the PCA 4,5-cleavage pathway genes are positively regulated by LigR, a LysR-type transcriptional regulator that recognizes PCA and GA as effectors (Fig. 1) (26). Moreover, we have shown that a MarR-type transcriptional regulator, DesR, negatively regulates the transcription of ligM and desB, and VA and SA are effectors that release repression by DesR (Fig. 1) (27). Although the transcription of desA is also induced by VA and SA, DesR does not participate in the transcriptional regulation of desA; thus, its regulatory system remains unknown.

In this study, we clarified the regulatory system of SA catabolism in SYK-6 by identifying and characterizing an IclR-type transcriptional regulator, DesX, which regulates desA transcription.

RESULTS

Search for genes involved in transcriptional regulation of desA.

To determine the presence of a transcriptional regulator of desA in Sphingobium sp. SYK-6, we performed an electrophoretic mobility shift assay (EMSA) using SYK-6 cell extracts and desAp1 to desAp5 probes containing sequences upstream of and within desA, as shown in Fig. 2A. SYK-6 was grown in Wx minimal medium containing 10 mM sucrose, 10 mM glutamate, 0.13 mM methionine, and 10 mM proline (Wx-SEMP) (28) (noninducing conditions) and in Wx-SEMP containing 5 mM SA or VA (SA/VA-inducing conditions). The cell extracts from these cultures were incubated with each of the desAp1 to desAp5 probes. A band shift was observed only with desAp2, regardless of the cell extracts used (Fig. 2B). These results indicate that an unknown SYK-6 protein binds to the DNA region spanning from 53 bp upstream to 229 bp downstream of the initiation codon of SLG_25010 (which is just upstream of desA).

FIG 2.

Identification of an SYK-6 protein that binds to the upstream region of desA. (A) Gene organization around desA. ferC, MarR-type transcriptional-regulator gene; ferB, feruloyl-CoA hydratase/lyase gene; ferA, feruloyl-CoA synthetase gene; SLG_25010, putative hydrolase gene; SLG_24970 (desX), IclR-type transcriptional-regulator gene; vceA, vanilloyl acetic acid/3-(4-hydroxy-3,5-dimethoxyphenyl)-3-oxopropanoic acid-converting enzyme gene (42). The horizontal black bars under the map show the DNA fragments used for EMSA (desAp1 to desAp5 probes). (B) EMSAs of SYK-6 cell extracts using the desAp1 to desAp5 probes. Digoxigenin-labeled probes (500 pM) were incubated in the presence and absence of extracts (0.4 μg protein/μl) of wild-type cells grown in Wx-SEMP, Wx-SEMP plus VA, and Wx-SEMP plus SA. CP, DNA-protein complex; F, free probe. (C) EMSAs of ΔferC and Δ24970 mutant cell extracts using the desAp2 probe. The desAp2 probe (500 pM) was incubated in the presence (+) and absence (−) of the extracts (0.4 μg protein/μl) of ΔferC and Δ24970 mutant cells grown in Wx-SEMP.

Upstream of desA (SLG_25000) are ferC (SLG_25040) and the ferBA (SLG_25030-SLG_25020) operon involved in ferulate (FA) catabolism. The ferC gene encodes a MarR-type transcriptional regulator that negatively regulates ferBA (Fig. 2A). Downstream of desA, there is SLG_24970, which is similar to the IclR-type transcriptional regulator (Fig. 2A; see Table S1 in the supplemental material). To clarify whether the gene product of ferC or SLG_24970 is involved in the binding of the desAp2 probe, we conducted EMSAs using cell extracts of a ferC mutant (ΔferC) and an SLG_24970 mutant (Δ24970). The ΔferC mutant was obtained in our previous study (28), while the Δ24970 mutant was constructed through homologous recombination in this study (see Fig. S1 in the supplemental material). EMSAs using a cell extract of the ΔferC mutant grown in Wx-SEMP showed a band shift similar to that of the wild type; however, no such band shift occurred in EMSA using a cell extract of the Δ24970 mutant grown in the same medium (Fig. 2C). These results strongly suggest that the band shift was due to the binding of the SLG_24970 gene product to the desAp2 region. Thus, we designated SLG_24970 desX.

To determine whether the disruption of ferC and desX affects SA and VA catabolism in SYK-6, we measured the growth of the ΔferC and ΔdesX mutants on 5 mM SA and VA. There was no significant difference in the growth of the ΔferC and ΔdesX mutants and the wild type on SA and VA, but the ΔdesX mutant tended to grow just a little faster (see Fig. S2 in the supplemental material).

Transcriptional analysis of the SA and VA catabolism genes in a desX mutant.

To determine whether desX is involved in the transcriptional regulation of the SA and VA catabolism genes desA, ligM, and desB in SYK-6, quantitative reverse transcription-PCR (qRT-PCR) analyses of these genes were performed using total RNAs isolated from the wild-type and ΔdesX mutant cells grown under noninducing and SA/VA-inducing conditions. The transcription levels of desA, ligM, and desB in the wild type under SA/VA-inducing conditions increased 28- to 50-, 5.7- to 14-, and 4.4- to 7.6-fold, respectively, compared to noninducing conditions (Fig. 3A to C). In ΔdesX mutant cells under noninducing conditions, the transcription levels of ligM and desB were similar to those of the wild type under the same conditions. Additionally, the transcription of ligM and desB was activated in ΔdesX mutant cells, as well as in the wild type, under SA/VA-inducing conditions (Fig. 3B and C). In contrast, the transcription levels of desA in ΔdesX mutant cells under noninducing conditions increased 48-fold compared to that of the wild type under the same conditions and were comparable to that of the wild type under inducing conditions (Fig. 3A). Therefore, it is evident that the transcription of desA is negatively regulated by DesX. The desA transcription levels were close to those of ΔdesX mutant cells under both noninducing and SA/VA-inducing conditions. These results suggest that the regulation of desA transcription is governed by DesX and that DesX is not involved in the transcriptional regulation of ligM and desB.

FIG 3.

qRT-PCR analysis of the expression of desA, ligM, desB, ferB, SLG_25010, and desX. Total RNAs were isolated from the cells of SYK-6 and a desX mutant (ΔdesX) grown in Wx-SEMP, Wx-5 mM SA, and Wx-5 mM VA. The relative mRNA amounts of desA (A), ligM (B), desB (C), ferB (D), SLG_25010 (E), and desX (measured only in the wild type) (F) indicate the fold increases relative to the amount of mRNA in SYK-6 cells grown in Wx-SEMP (a level of 1.0). The value for each amount of mRNA was normalized to the level of 16S rRNA. Each value is the average ± standard deviation (error bars) of three independent experiments. Statistical differences were determined by Student's t test. The asterisks indicate statistically significant differences between the values linked by brackets (ns, P > 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001).

Upstream of desA are SLG_25010, which encodes a putative hydrolase, and the ferBA operon, both of which are transcribed in the same direction as desA (Fig. 2A). qRT-PCR analyses showed that the transcription levels of ferB in the wild-type and ΔdesX mutant cells were almost the same under both noninducing and SA/VA-inducing conditions (Fig. 3D), indicating that DesX does not participate in the regulation of the ferBA operon. In contrast, the transcription of SLG_25010 in the wild type increased 11- to 39-fold under SA/VA-inducing conditions compared to noninducing conditions (Fig. 3E). In ΔdesX mutant cells under noninducing conditions, the transcription level of SLG_25010 increased 59-fold compared to that of the wild type grown under the same conditions. This level is equivalent to that in ΔdesX mutant cells under SA/VA-inducing conditions (44- to 51-fold) (Fig. 3E). Therefore, in the transcriptional regulation of SLG_25010, DesX acts as a repressor, while SA and VA appear to be inducers of SLG_25010, along with desA. In addition, the transcription level of desX in the wild type remained nearly constant regardless of the culture conditions (Fig. 3F), indicating that desX is constitutively expressed.

Operon structure of the ferB-ferA-SLG_25010-desA gene cluster.

In our previous study, the ferB-ferA-SLG_25010-desA gene cluster formed an operon during growth in the presence of SA (29). In contrast, although ferB and ferA formed an operon during growth in the presence of FA, SLG_25010 was not cotranscribed with ferBA (28). In this study, qRT-PCR analysis indicated that the transcriptional inducibilities of ferBA and SLG_25010-desA in SYK-6 cells grown on SA and VA were completely different (Fig. 3). We therefore reevaluated the operon structure of the ferB-ferA-SLG_25010-desA gene cluster by reverse transcription (RT)-PCR analysis using total RNA isolated from SYK-6 grown in Wx-SEMP medium (noninducing conditions) and Wx medium containing 5 mM SA, VA, or FA (SA/VA/FA-inducing conditions) (Fig. 4A). Under noninducing conditions, only the ferB-ferA region was amplified (Fig. 4B). In contrast, under SA/VA/FA-inducing conditions, the ferA-SLG_25010 and SLG_25010-desA regions were amplified, in addition to the ferB-ferA region, although amplification of the ferA-desA region was not observed (Fig. 4C to E). These results strongly suggest that the ferB-ferA-SLG_25010-desA gene cluster consists of the ferBA operon and the SLG_25010-desA operon. The amplification of ferA-SLG_25010 is likely a consequence of the read-through from the ferB promoter. The reason the ferA-SLG_25010 region was amplified under SA/VA-inducing conditions that did not induce the ferBA operon, but not under noninducing conditions, is discussed below.

FIG 4.

RT-PCR analysis of the ferB-ferA-SLG_25010-desA gene cluster in SYK-6. (A) Organization of the ferB-ferA-SLG_25010-desA gene cluster. The bars with numbers below the map indicate the regions to be amplified and correspond to the numbering in panels B to E. (B to E) Agarose gel electrophoresis of RT-PCR assays with primers targeting ferA-SLG_25010 (lanes 1; expected size, 932 bp), SLG_25010-desA (lanes 2 and 3; 1,155 bp and 1,132 bp, respectively), ferA-desA (lanes 4; 1,987 bp), and ferB-ferA (lanes 5; 955 bp). Total RNAs isolated from SYK-6 cells grown in Wx-SEMP (B), Wx-5 mM SA (C), Wx-5 mM VA (D), and Wx-5 mM FA (E) were used as templates for cDNA synthesis. Lanes: M, molecular size markers; G, control PCR with the SYK-6 genomic DNA; + and −, RT-PCR with or without reverse transcriptase, respectively.

Determination of the promoter region of the SLG_25010-desA operon.

Promoter assays using lacZ as the reporter gene were performed to determine the promoter region of the SLG_25010-desA operon. pSDA1 to pSDA4 were constructed by cloning DNA fragments containing various regions from ferA to desA into the region upstream of lacZ in a promoter-probe vector, pSEVA225 (Fig. 5A). Promoter activities of SYK-6 cells harboring each pSDA1 to pSDA4 plasmid grown under noninducing and SA/VA-inducing conditions were then measured as β-galactosidase activities (Fig. 5A). Prominent promoter activities were observed only in SYK-6 cells harboring pSDA2 under SA/VA-inducing conditions that elicited activities that were 12 times higher than those under noninducing conditions. These results suggest that an SA/VA-inducible promoter is located in the intergenic region between ferA and SLG_25010.

FIG 5.

Identification of the promoter region of the SLG_25010-desA operon. (A) Promoter analysis of the upstream regions of desA. (Left) DNA fragments used for promoter analysis. The transcription start site of the SLG_25010-desA operon is shown by a bent arrow. Putative −35 and −10 sequences and positions of IR-DA are shown by violet and red boxes, respectively. (Right) β-Galactosidase activities of SYK-6 cells harboring each reporter plasmid grown in the presence and absence of 5 mM SA or VA. Each value is the average ± standard deviation (error bars) of the results of three independent experiments. Statistical differences were determined by one-way ANOVA with Dunnett’s multiple-comparison test. The asterisks indicate statistically significant differences between the values linked by brackets (ns, P > 0.05; ****, P < 0.0001). (B) Determination of the transcription start site of the SLG_25010-desA operon by primer extension. The transcription start site of the SLG_25010-desA operon was determined using total RNA isolated from SYK-6 cells grown in Wx-5 mM SA and the PE25010 primer. (Left) Analysis of extended products using an ABI 3730xl DNA analyzer. AU, arbitrary units. (Right) Nucleotide sequence of the SLG_25010 promoter region. The initiation codon of SLG_25010 is shaded black. The transcription start site is shown by a bent arrow (+1). Putative −35 and −10 sequences are underlined. The inverted-repeat sequence (IR-DA) is indicated by convergent arrows. The putative Shine-Dalgarno (SD) sequence is double underlined. The dashed arrow indicates the position of the PE25010 primer.

The transcription start site of the SLG_25010-desA operon was determined by primer extension analysis using a fluorescently labeled oligonucleotide primer and total RNA isolated from SYK-6 cells grown on SA. This analysis yielded a 76-nucleotide extension product that maps the transcription start site of the SLG_25010-desA operon to the T residue located 10 nucleotides upstream of the initiation codon of SLG_25010 (Fig. 5B). The putative −35 and −10 sequences, bearing similarities to the conserved sequences of the Escherichia coli σ⁷⁰-dependent promoter, were found upstream of the transcription start site of the SLG_25010-desA operon (Fig. 5B). Additionally, we found an 18-bp inverted-repeat sequence, named IR-DA (5′-TCTTCGTATATACGAAGA-3′) at positions −21 to −4 (overlapping the putative −10 sequence) (Fig. 5B).

To determine whether the putative −35 and −10 sequences are necessary for the transcription of the SLG_25010-desA operon, deletions were introduced in the putative promoter region in pSDA2 (Fig. 5A). Like SYK-6 cells harboring pSDA2, SYK-6 cells harboring pSDA2a (containing putative −35 and −10 sequences) exhibited promoter activity that was 9.4-fold higher under SA/VA-inducing conditions than activity under noninducing conditions (Fig. 5A). However, the promoter activity was not observed in SYK-6 cells harboring pSDA2b (containing only a putative −10 sequence) or pSDA2c (lacking putative −35 and −10 sequences) (Fig. 5A). These results suggest that the putative −35 and −10 sequences function as the essential promoter of the SLG_25010-desA operon.

Binding of DesX to the promoter region of the SLG_25010-desA operon.

desX fused with a His tag at the 5′ terminus was expressed under the control of the T7 promoter in E. coli BL21(DE3) harboring p16bdX (Table 1). Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) analysis showed the His tag-fused DesX as an approximately 29-kDa protein, which is close to the theoretical molecular weight of 28,968 (see Fig. S3 in the supplemental material). DesX was purified to near homogeneity using Ni affinity chromatography (see Fig. S3).

TABLE 1.

Strains and plasmids used in this study

Strain or plasmid	Relevant characteristic(s)a	Reference or source
Sphingobium sp.
SYK-6	Wild type; Nalr Smr	43
ΔferC (SME043)	SYK-6 derivative; ferC::kan; Nalr Smr Kmr	28
ΔdesX (SME085)	SYK-6 derivative; SLG_24970 (desX)::kan; Nalr Smr Kmr	This study
E. coli
NEB 10-beta	Δ(ara-leu)7697 araD139 fhuA ΔlacX74 galK16 galE15 e14-ϕ80dlacZΔM15 recA1 relA1 endA1 nupG rpsL (Smr) rph spoT1 Δ(mrr-hsdRMS-mcrBC)	New England Biolabs
BL21(DE3)	F− ompT hsdSB(rB− mB−) gal dcm (DE3); T7 RNA polymerase gene under control of the lacUV5 promoter	44
Plasmids
pUC18	Cloning vector; Apr	45
pBluescript II KS(+)	Cloning vector; Apr	46
pK19mobsacB	oriT sacB; Kmr	47
pIK03	pBluescript II KS(+) with a 1.3-kb EcoRV fragment carrying kan of pUC4K; Apr Kmr	29
pJB866	RK2 broad-host-range expression vector; Tcr Pm xylS	48
pSEVA225	RK2 ori lacZ promoter-probe broad-host-range vector; Kmr	49
pET-16b	Expression vector; T7 promoter; Apr	Novagen
pUC401	pUC18 with a 4.8-kb SalI fragment carrying SLG_24970 (desX)	29
pKPEV	pBluescript II KS(+) with a 1.0-kb PstI-EcoRV fragment of pUC401	This study
pKPEVK	pKPEV with a 1.3-kb EcoRV fragment of pIK03 carrying kan	This study
pMPEVK	pK19mobsacB with a 2.3-kb PstI-SalI fragment of pKPEVK	This study
pKNIF	pBluescript II KS(+) with a 1.2-kb NruI fragment of pUC401	This study
pMPEVKNI	pMPEVK with a 1.2-kb SalI-EcoRI fragment of pKNIF	This study
pJB24970	pJB866 with a 1.1-kb PCR-amplified HindIII-BamHI fragment carrying desX	This study
pSDA1	pSEVA225 with a 396-bp fragment carrying the sequence between positions −548 and −153 relative to the SLG_25010 initiation codon	This study
pSDA2	pSEVA225 with a 378-bp fragment carrying the sequence between positions −243 and +135 relative to the SLG_25010 initiation codon	This study
pSDA2a	pSEVA225 with a 238-bp fragment carrying the sequence between positions −103 and +135 relative to the SLG_25010 initiation codon	This study
pSDA2b	pSEVA225 with a 162-bp fragment carrying the sequence between positions −27 and +135 relative to the SLG_25010 initiation codon	This study
pSDA2c	pSEVA225 with a 144-bp fragment carrying the sequence between positions −9 and +135 relative to the SLG_25010 initiation codon	This study
pSDA3	pSEVA225 with a 441-bp fragment carrying the sequence between positions +48 and +488 relative to the SLG_25010 initiation codon	This study
pSDA4	pSEVA225 with a 387-bp fragment carrying the sequence between positions +430 and +816 relative to the SLG_25010 initiation codon	This study
p16bdX	pET-16b with a 750-bp NdeI-BamHI fragment carrying desX	This study

^aNal^r, Sm^r, Km^r, Ap^r, and Tc^r, resistance to nalidixic acid, streptomycin, kanamycin, ampicillin, and tetracycline, respectively.

EMSAs were performed to determine whether DesX binds to the promoter region of the SLG_25010-desA operon using DNA probes containing various regions around the SLG_25010-desA operon (desAp1 to desAp5 probes) (Fig. 6A). The probes were incubated with purified DesX, resulting in a band shift that showed the formation of a DNA-DesX complex with the desAp2 probe that contains the promoter region of the SLG_25010-desA operon (Fig. 6B). The other DNA probes generated no other band shifts (Fig. 6B). Incubating the desAp2 probe with different concentrations of DesX resulted in one band shift for all the concentrations (Fig. 6C), implying that there is one binding site for DesX in the desAp2 region. To determine if IR-DA is required for DNA binding of DesX, we performed EMSAs using the desAp6 to desAp8 probes shown in Fig. 6A. A band shift was observed only for the desAp7 probe, which contains IR-DA (Fig. 6B). We further examined the binding of DesX to DNA probes containing mutated IR-DA. EMSA using the desAp9 probe containing IR-DA at the 3′ end showed the formation of a DNA-DesX complex (Fig. 6A and D). Using the desAp9mL and desAp9mR probes containing a 9-base mutation in the left and right halves of the desAp9 probe, respectively, resulted in significantly reduced band shifting (Fig. 6A and D). All these results make clear that IR-DA is essential for the DNA binding of DesX. Additionally, we performed EMSAs using DNA probes containing the ferB, ligM, and desB promoter regions, which include the binding sites of FerC, DesR, and DesR, respectively (see Fig. S4 in the supplemental material). DesX did not bind to any of these DNA probes, regardless of the presence or absence of SA or VA. This supports the results of qRT-PCR analyses, indicating that DesX is not involved in regulating the ferBA operon, ligM, and desB.

FIG 6.

Determination of the binding region of DesX. (A) DNA fragments used for EMSA. The transcription start site of the SLG_25010-desA operon is shown by a bent arrow. Putative −35 and −10 sequences and positions of IR-DA are shown by violet and red boxes, respectively. Mutated sequences of IR-DA in the desAp9mL and desAp9mR probes are shown. (B) EMSAs of the binding of DesX to the desAp1 to desAp8 probes. The DNA probes (400 pM) were incubated in the presence (+) and absence (−) of purified DesX (8 ng protein/μl). (C) EMSAs of the binding of various concentrations of DesX. Purified DesX (0.5 to 16 ng protein/μl) was incubated with the desAp2 probe (400 pM). (D) EMSAs of the binding of DesX to the desAp9 and the IR-DA mutated probes (desAp9mL and desAp9mR). Each probe (400 pM) was incubated in the presence (+) and absence (−) of purified DesX (8 ng protein/μl).

Identification of effector molecules of DesX.

Previously, we demonstrated that SA and VA are inducers of desA (27). Here, we also show that the transcription of desA and SLG_25010 is significantly induced in the presence of SA or VA (Fig. 3A and E). To examine the effects of SA and VA on the DNA binding of DesX, we performed EMSAs using DesX and the desAp2 probe in the presence of either SA or VA (Fig. 7). In the presence of SA, band shifts decreased in a concentration-dependent manner and disappeared entirely at 5 mM SA. In the presence of VA, a similar decrease in band shifts was observed with the desAp2 probe. Therefore, both SA and VA are effector molecules for DesX, and SA and VA have different affinities for DesX, since SA affected DNA binding of DesX at lower concentrations (5 mM SA versus 50 mM VA) (Fig. 7).

FIG 7.

Identification of the effector molecules of DesX. Shown are EMSAs of the binding of DesX to the desAp2 probe containing IR-DA in the presence of SA or VA. Purified DesX (4 ng protein/μl) and the desAp2 probe (400 pM) were incubated in the presence (0.05, 0.5, 5, or 50 mM) or absence of SA or VA.

DISCUSSION

This study reveals that the ferB-ferA-SLG_25010-desA gene cluster consists of the ferBA operon and the SLG_25010-desA operon and that the transcription start site of the SLG_25010-desA operon is located 10 bp upstream of the SLG_25010 initiation codon (Fig. 4 and 5B). An IclR-type transcriptional regulator, DesX, is the only regulator that controls the transcription of the SLG_25010-desA operon, at least under the culture conditions used (Fig. 3A and E); it is not involved in the transcriptional regulation of ligM and desB (Fig. 3B and C; see Fig. S4). The IclR-type transcriptional regulators are known to include both activators and repressors (30). Many IclR-type transcriptional regulators that have been reported as activators are thought to bind to the upstream region of the −35 sequence on the target promoter, thus recruiting RNA polymerase (31, 32). In contrast, the binding sequences of the repressors HmgR and IphR are located downstream of the −10 sequence on the target promoter (33, 34). EMSAs showed DesX binding to IR-DA, an inverted-repeat sequence that overlaps the −10 sequence of the SLG_25010 promoter (Fig. 6). Therefore, DesX appears to repress the transcription of the SLG_25010-desA operon by preventing binding of RNA polymerase to the SLG_25010 promoter.

DesX recognizes both SA and VA as effector molecules and has higher affinity for SA than for VA (Fig. 7). Based on these findings and those in our previous study (26, 27), the catabolism of SA in SYK-6 is as follows (Fig. 1). The binding of SA to DesX first triggers the derepression of desA, enabling the conversion of SA to 3MGA by DesA. The binding of SA to DesR also derepresses ligM and desB (27), allowing LigM to O demethylate 3MGA to GA. The aromatic ring of GA is then cleaved by DesB. Because GA is an effector molecule of LigR (26), GA binding to LigR activates the transcription of the PCA 4,5-cleavage pathway genes involved in the catabolism of the ring cleavage product of GA.

Both DesX and DesR recognize VA and SA as effector molecules. This is reasonable for DesR, because it is involved in the regulation of ligM, which is involved in the O demethylation of both VA and 3MGA (an intermediate metabolite of SA). Mutant analysis indicated that DesA, which is mainly involved in the O demethylation of SA, is also partially involved in the conversion of VA (29, 35). Accordingly, DesX recognizes VA as an effector molecule, in addition to SA, which may enable more efficient conversion of VA. Although DesX has higher affinity for SA than for VA in vitro (Fig. 7), desA is transcribed at a higher rate in vivo under VA-inducing conditions than under SA-inducing conditions (Fig. 3A). The reason for this discrepancy is still unclear; however, the in vivo results may not completely reflect in vitro results, as the former is affected by other factors, such as the level of substrate uptake. It is also unclear why the millimolar concentration of effector is necessary for the dissociation of DesX from DNA in EMSA. The need for high concentrations of effector was similar for DesR (27).

The transcription start site of the SLG_25010-desA operon is located 10 bp upstream from the initiation codon of SLG_25010, and a weak Shine-Dalgarno sequence is 8 to 6 bp upstream of the initiation codon (Fig. 5B). According to the ribosome binding site (RBS) calculator (36), the translation initiation rate of SLG_25010 from the putative mRNA sequence of the SLG_25010-desA transcript is 1.77, a rate that is markedly lower than those for desA (2,395), ferB (2,577), ferA (230), ligM (177), and desB (6,023). Therefore, SLG_25010 may be poorly translated from this mRNA. Read-through from the ferB promoter was observed in the intergenic region between ferA and SLG_25010 under SA-, VA-, or FA-inducing conditions (Fig. 4C to E). We speculate that when SYK-6 cells were incubated under noninducing conditions, DesX binding to the promoter region of the SLG_25010-desA operon probably inhibited transcription from the ferB promoter. In contrast, when SYK-6 cells were incubated under SA/VA (FA)-inducing conditions, DesX was released from the promoter region, thus allowing transcription from the ferB promoter to proceed. The mRNA generated from this read-through may be required for the translation of SLG_25010 (translation initiation rate, 82). The function of SLG_25010, currently annotated as a hydrolase, has not been clarified. In SYK-6 cells, a metabolite in the SA catabolism, 4-carboxy-2-hydroxy-6-methoxy-6-oxohexa-2,4-dienoate (CHMOD), is spontaneously converted to 2-pyrone-4,6-dicarboxylate (PDC). Simultaneously, CHMOD is converted to 4-oxalomesaconate (OMA) by a hydrolase whose gene has not yet been identified (Fig. 1) (37). Recently, the methylesterase (DesC) and the cis-trans isomerase (DesD) genes were reported to be involved in the conversion of CHMOD to OMA during SA catabolism in N. aromaticivorans DSM 12444 (23). In the SYK-6 genome, the SLG_12720 and SLG_07230 amino acid sequences are 45% and 40% similar to those of desC and desD, respectively (see Fig. S5 in the supplemental material). However, there is no SLG_25010 ortholog in the DSM 12444 genome. It will be necessary to investigate the involvement of these genes in the conversion of CHMOD in SYK-6 in the future.

In N. aromaticivorans DSM 12444, SA is converted to 3MGA by the Saro_2404 gene product (DesA_NA), whose amino acid sequence is 71% similar to that of SYK-6 DesA (see Fig. S5). The resulting 3MGA is metabolized via CHMOD as described above (23). In the DSM 12444 genome, a BLAST search revealed Saro_2407, encoding a product with 51% amino acid sequence identity with DesX. In the intergenic region between Saro_2403 and desA_NA, there is an incomplete 18-bp inverted-repeat sequence (5′-TTTTCAAGCACGCGAAAA-3′ [where underlining indicates inverted repeats]) that matched 10 of the 18 nucleotides of IR-DA (see Fig. S5). Because Saro_2404 (desA_NA) and Saro_2407 (desX ortholog) are close to each other, and an inverted-repeat sequence similar to IR-DA is present upstream of desA_NA (see Fig. S5), desA_NA is probably regulated by the Saro_2407 gene product. In DSM 12444, VA is converted to PCA by the Saro_2861 gene product (LigM_NA), which is 78% similar to the SYK-6 LigM (see Fig. S5), and PCA is metabolized via the PCA 4,5-cleavage pathway (23). DSM 12444 also possesses Saro_0803, which is 51% similar to SYK-6 DesR (see Fig. S5), implying that Saro_2861 (ligM_NA) is regulated by the Saro_0803 gene product (DesR ortholog) in DSM 12444.

In conclusion, we have clarified the transcriptional-regulatory system for the SA O-demethylase gene desA in SYK-6, a model bacterial degrader of lignin-derived aromatic compounds. Our present results, combined with our previous findings, have enabled us to provide an overall picture of the regulatory systems for SA and VA catabolism. This information is essential for creating engineered bacteria that can efficiently produce value-added chemicals from lignin.

MATERIALS AND METHODS

Bacterial strains, plasmids, culture conditions, primers, and chemicals.

The bacterial strains and plasmids used in this study are listed in Table 1, and the PCR primers are listed in Table 2. Sphingobium sp. SYK-6 and its mutants were grown at 30°C with shaking (160 rpm) in lysogeny broth (LB); Wx medium (see Table S2 in the supplemental material) (28) containing SEMP and 5 mM SA, VA, or FA; and Wx-SEMP containing 5 mM SA or VA. Whenever necessary, the media for SYK-6 and its mutants and transformants were supplemented with 50 mg of kanamycin/liter, 12.5 mg of tetracycline/liter, or 12.5 mg of nalidixic acid/liter. E. coli NEB 10-beta was used for the cloning experiments. E. coli BL21(DE3) was used for the expression of desX. E. coli strains were grown at 30°C or 37°C with shaking (160 rpm) in LB. The media for E. coli transformants were supplemented with 100 mg of ampicillin/liter, 25 mg of kanamycin/liter, or 12.5 mg of tetracycline/liter. SA, VA, and FA were purchased from Tokyo Chemical Industry Co. Ltd.

TABLE 2.

Primers used in this study

Purpose	Primer	Sequence (5' to 3')
qRT-PCR analysis	16S_qF	GCGCAGAACCTTACCAACGT
	16S_qR	AGCCATGCAGCACCTGTCA
	desA_qF	GCCTTCGCCTTCCTCAACTA
	desA_qR	CACCGGAACCCACTGCTT
	ligM_qF	GCTCTCCGACACGATGATCA
	ligM_qR	ACGTACTGCTTCGCCTTGTTG
	desB_qF	TTTCGAGCATTATTCGCATTTC
	desB_qR	TCCGCAGGCGAATATTCCT
	ferB_qF	CCGGTGGAACGGGAAGA
	ferB_qR	CCACGCCACGTTGTTCAC
	25010_qF	GACATGCTGTGGCAGATGTG
	25010_qR	CGCATCTGCCGCTCATAC
	desX_qF	CAGAAGGTGGACTCGTCGT
	desX_qR	AGGATATCGAGCGTGCG
RT-PCR analysis	RTferB_F	TGACGTACGACAATGCGGAA
	RTferA_F	TGACGCCTCCATCATTCTCG
	RTferA_R	ATGATCCGCGTCTTCTCGTC
	RT25010_F1	ATGAGTCACTCGCCTTCCA
	RT25010_F2	TCAGCACCGGCATTCACTT
	RT25010_R	GCATCGATGAGGGCATCCAT
	RTdesA_R1	CCGACGAGGTTGAACTGGTT
	RTdesA_R2	TCATAATCCGCCCAGGGAC
Gene complementation	pJB24970_F	GACGTCACCATGGGAAGCTTGACACGATCTACCTGCGCA
	pJB24970_R	CCTGCAGGATATCTGGATCCTCCTCGTGGGACTGGTCAT
Primer extension	PE25010a	ACGGTCATCGCAGATCAG
Promoter assay	pSDA1_F	ACCTGCAGGCATGCAAGCTTTCCATCATTCTCGACGGCG
	pSDA1_R	ATGTTTTTCCTCCTAAGCTTTCAGTCCACCAGCATCAGG
	pSDA2_F	ACCTGCAGGCATGCAAGCTTGCGCATCCAGGAACTCGAT
	pSDA2a_F	ACCTGCAGGCATGCAAGCTTGCATGACCTTTCAATTGTGCG
	pSDA2b_F	ACCTGCAGGCATGCAAGCTTCGTATATACGAAGAACATCGG
	pSDA2c_F	ACCTGCAGGCATGCAAGCTTCGGATTTCCATGAGTCACTCG
	pSDA2_R	ATGTTTTTCCTCCTAAGCTTTTCCCCATAGCCCGCAA
	pSDA3_F	ACCTGCAGGCATGCAAGCTTCCTGATCTGCGATGACCGT
	pSDA3_R	ATGTTTTTCCTCCTAAGCTTCGCATCTGCCGCTCATAC
	pSDA4_F	ACCTGCAGGCATGCAAGCTTGACATGCTGTGGCAGATGTG
	pSDA4_R	ATGTTTTTCCTCCTAAGCTTTTCCGGCATTGTCCAGCA
Protein expression	p16bdX_F	TATCGAAGGTCGTCATATGATCCAGAAGGTGGACTC
	p16bdX_R	CTTTGTTAGCAGCCGGATCCTCAGCGGTCGCCAAG
EMSAs	desAp1_F	CTGCAGGATGTGCGCC
	desAp1_R	GTGAATGCCAGCCCCAAA
	desAp2_F	GCATGACCTTTCAATTGTGCG
	desAp2_R	AAGTGAATGCCGGTGCTGAT
	desAp3_F	AGATCATTCGCCGCGCA
	desAp3_R	ATGGCTTCATGCTGCACC
	desAp4_F	ACCGGCGCCGATGATGC
	desAp4_R	TGGTTGAAGAGCACGGCGG
	desAp5_F	AACTGGCGCAACGAGCA
	desAp5_R	CTGCAGCTGGAAGCGATAGT
	desAp6_F	ATGAGTCACTCGCCTTCCA
	desAp7_F	TTTCCCTCTGCACGACGT
	desAp7_R	GTGCCCAGATACGGTCATC
	desAp8_F	GCGCATCCAGGAACTCGAT
	desAp8_R	TTTCATCTACTCGAAGACCGG
	IR-DA_R	AATCCGATGTTCTTCGTATATACGAAGATTTCATCT
	IR-DA_mL_R	AATCCGATGTTCTTCGTATGCGTAGGAGTTTCATCT
	IR-DA_mR_R	AATCCGATGTCTCCTACGCATACGAAGATTTCATCT
	ferBp_F	ATATTGTCGAGCGGGCTG
	ferBp_R	ATCTGGCGGTTGAGCTTG
	desBp_F	GCCGTTCCCCTCTCAGGC
	desBp_R	AAACCGCCGATGATCTTTGC
	ligMp_F	GTTTCCGCCTGGTCTGAA
	ligMp_R	AGGTGCCGACATCAGCTA

^aPrimer labeled with 6-carboxyfluorescein at the 5′ end.

EMSA.

DNA fragments were amplified from the SYK-6 genomic DNA by PCR using the specific primers listed in Table 2. EMSAs were performed on cell extracts using a DIG gel shift kit (2nd generation; Roche). DNA fragments were labeled at their 3′ ends with digoxigenin-11-ddUTP using terminal transferase. Cell extracts of SYK-6 and mutants were prepared as described previously (27). The DNA-protein binding reactions were conducted at 20°C for 20 min in a final volume of 10 μl. EMSA reaction mixtures contained cell extracts (0.4 μg protein/μl), 5 fmol digoxigenin-labeled probe, 1 μg poly(dI-dC), and binding buffer [20 mM HEPES, 1 mM EDTA, 10 mM (NH₄)₂SO₄, 1 mM dithiothreitol, 0.2% (wt/vol) Tween 20, 30 mM KCl, pH 7.6]. EMSA reaction mixtures using purified DesX contained purified DesX (0.5 to 16 ng protein/μl), 4 fmol nonlabeled probe, and binding buffer. To examine the association of DesX with effector molecules, 1 μl of a solution of SA or VA (final concentration, 0.05, 0.5, 5, or 50 mM) was added to the reaction mixture. After incubation, 2.5 μl loading buffer (60% [vol/vol] 0.25× TBE buffer [89 mM Tris, 89 mM boric acid, and 20 mM EDTA, pH 8.0], 40% [vol/vol] glycerol, and 0.2% [vol/vol] bromophenol blue) was added, and samples were separated on a 5% native polyacrylamide gel in 0.5× TBE buffer. After electrophoresis, the labeled DNAs were detected using a CSPD-based chemiluminescence detection system (Roche) (26), and the non-labeled DNAs in the gel were stained with SYBR Gold nucleic gel stain (Invitrogen) and photographed under a 470-nm blue LED.

Construction of the SLG_24970 mutant.

The 1.0-kb PstI-EcoRV fragment carrying the upstream region of SLG_24970 from pUC401 was inserted into the corresponding sites of pBluescript II KS(+) to obtain pKPEV. The 1.3-kb EcoRV fragment carrying the kanamycin resistance gene (kan) of pIK03 was inserted into the EcoRV site of pKPEV to form pKPEVK. The 2.3-kb PstI-SalI fragment of pKPEVK was then inserted into the corresponding sites of pK19mobsacB to obtain pMPEVK. The 1.2-kb NruI fragment carrying the downstream region of SLG_24970 from pUC401 was cloned into the EcoRV site of pBluescript II KS(+), and the 1.2-kb SalI-EcoRI fragment (including the 1.2-kb NruI fragment) from the resulting plasmid (pKNIF) was inserted into the corresponding site of pMPEVK, generating an SLG_24970 disruption plasmid, pMPEVKNI. pMPEVKNI was introduced into SYK-6 cells by electroporation, and the mutant was selected as previously described (38, 39). Disruption of the gene was confirmed by Southern hybridization analysis using the total DNA of mutant candidates digested with SalI and the digoxigenin-labeled probes (Roche), the 1.4-kb ApaI fragment carrying SLG_24970, and the 1.3-kb EcoRV fragment carrying kan.

The complementary plasmid, pJB24970, was constructed by amplifying a 1.1-kb fragment carrying SLG_24970 from SYK-6 genomic DNA using the primer pair listed in Table 2. The resulting fragment was inserted into the HindIII-BamHI sites of pJB866 by an NEBuilder HiFi DNA assembly cloning kit (New England Biolabs). The nucleotide sequence of the insert was confirmed by sequencing. pJB866 and pJB24970 were independently introduced into Δ24970 mutant and SYK-6 cells by electroporation, and the growth of transformants was measured as described below.

Bacterial growth measurement.

Cells of SYK-6, its mutants, and complemented strains were grown in LB for 24 h, harvested by centrifugation (5,000 × g; 5 min; 4°C), washed twice with Wx medium (without solution II [see Table S2]), and resuspended in 1 ml of the same medium. Cells were then inoculated into 5 ml of Wx medium containing 5 mM SA or VA to an optical density at 660 nm (OD₆₆₀) of 0.2. The cells were incubated at 30°C with shaking (60 rpm), and cell growth was periodically monitored by measuring the OD₆₆₀ using a TVS062CA biophotorecorder (Advantec Co., Ltd.). Complemented strains of the Δ24970 mutant were analyzed by growing cells in Wx medium containing tetracycline and 1 mM m-toluate, an inducer of the P_m promoter in pJB866.

Isolation of total RNA.

Cells of SYK-6 and the ΔdesX mutant were grown in LB for 24 h, harvested by centrifugation (5,000 × g; 5 min; 4°C), washed twice with Wx medium (without solution II), and resuspended in the same medium. The resulting cells were inoculated into 10 ml of Wx-SEMP, Wx-5 mM SA, Wx-5 mM VA, and Wx-5 mM FA to an OD₆₀₀ of 0.2 and incubated until the OD₆₀₀ of the culture reached 0.5 to 0.6. Total RNAs were isolated as described previously (27).

qRT-PCR and RT-PCR.

cDNAs were synthesized by reverse transcription using a PrimeScript II 1st-strand cDNA synthesis kit (TaKaRa Bio). Total RNA (1 μg) was reverse transcribed using PrimeScript reverse transcriptase with random primers (6-mers). Reactions for each sample included a reverse-transcriptase-negative control to verify no genomic DNA contamination had occurred. The synthesized cDNA was purified using a NucleoSpin gel and PCR clean-up kit (TaKaRa Bio) and eluted with 30 μl of 5 mM Tris-HCl buffer (pH 8.5). qRT-PCRs used a Thunderbird SYBR qPCR mixture (TOYOBO) and were amplified on a LightCycler 480 System II (Roche). The specific primer pairs used for qRT-PCR analyses were designed using the Primer Express version 2.0 software program (Applied Biosystems) (Table 2). qRT-PCR contained 2 μl of cDNA sample, gene-specific primers (10 pmol), and Thunderbird SYBR qPCR mixture (10 μl) in a total reaction volume of 20 μl. Thermal-cycling conditions were as follows: 20 s at 95°C, followed by 40 repeats of 3 s at 95°C and 30 s at 60°C. The LightCycler 480 System II detected and analyzed changes in the fluorescence emission and performed a melting curve analysis at the end of qRT-PCR to verify the specificity of the amplification. The amount of RNA in each sample was normalized relative to 16S rRNA, which was used as an internal standard. Each measurement was conducted in triplicate, and the means and standard deviations were calculated. RT-PCR was performed using 1 μl of cDNA sample, specific primers (Table 2), and Q5 Hot Start high-fidelity DNA polymerase (New England Biolabs). The PCR products were electrophoresed on a 0.8% agarose gel.

Promoter assay.

Reporter plasmids were constructed using DNA fragments containing desA and its upstream region, which were PCR amplified from SYK-6 genomic DNA and the primer pairs listed in Table 2. The PCR products were inserted into the HindIII site upstream of the promoterless lacZ of pSEVA225. The nucleotide sequence of the insert was confirmed by sequencing. Each plasmid was introduced into SYK-6 cells by electroporation, and SYK-6 transformants were grown for 24 h in LB containing kanamycin, harvested by centrifugation, washed twice with Wx medium (without solution II), and resuspended in the same medium. The resulting cells were inoculated into 10 ml Wx-SEMP, Wx-5 mM SA, and Wx-5 mM VA containing kanamycin to an OD₆₀₀ of 0.2 and incubated until the OD₆₀₀ of the culture reached 0.5 to 0.6. Cells were harvested by centrifugation, washed twice with Wx medium (without solution II), and resuspended in the same medium to an OD₆₀₀ of 2.0. The β-galactosidase activity of the cells was measured using 2-nitrophenyl-β-d-galactopyranoside according to a modified Miller assay [https://openwetware.org/wiki/Beta-Galactosidase_Assay_(A_better_Miller)] and expressed in Miller units as previously described (40).

Primer extension.

Total RNA was isolated from SYK-6 cells grown in Wx-5 mM SA. cDNA was synthesized from total RNA (5 μg) using a PrimeScript II 1st-strand cDNA synthesis kit and a 6-carboxyfluorescein-labeled PE25010 primer (2 pmol). The extended products were purified with a NucleoSpin gel and PCR clean-up kit and then analyzed using an ABI 3730xl DNA analyzer (Applied Biosystems) at Macrogen.

Expression of desX in E. coli and purification of DesX.

A DNA fragment carrying desX was amplified by PCR using SYK-6 genomic DNA and the primer pair listed in Table 2. The PCR product was cloned into the NdeI-BamHI sites of pET-16b with an NEBuilder HiFi DNA assembly cloning kit, and the nucleotide sequence of the insert of the resulting p16bdX was confirmed by sequencing. E. coli BL21(DE3) cells harboring p16bdX were grown for 12 h at 37°C in LB containing ampicillin, and the culture was inoculated into the same fresh medium (final concentration, 1%). When the OD₆₀₀ of the culture reached 0.5 to 0.6, the expression of desX with an N-terminal His tag was induced for 4 h at 30°C by adding 1 mM isopropyl-β-d-thiogalactopyranoside. The cells were harvested by centrifugation (5,000 × g; 5 min; 4°C), washed twice with 50 mM Tris-HCl buffer (pH 7.5) containing 300 mM NaCl and 100 mM imidazole (buffer A), resuspended in buffer A, and then disrupted using an ultrasonic disintegrator. The supernatant obtained by centrifugation (19,000 × g; 15 min; 4°C) was applied to a His SpinTrap column (GE Healthcare) previously equilibrated with buffer A. After centrifugation (100 × g; 1 min; 4°C), samples were washed three times with buffer A, and then His tag-fused DesX was eluted with 50 mM Tris-HCl buffer (pH 7.5) containing 300 mM NaCl and 500 mM imidazole. Purified DesX was desalted and concentrated by centrifugal filtration using an Amicon Ultra centrifugal filter unit (10-kDa cutoff; Merck Millipore). Before the assay, insoluble aggregates in the DesX solution were removed by centrifugal filtration using an Ultrafree-MC filter (Merck Millipore). The protein concentration was determined by the Bradford method (41) with bovine serum albumin as the standard (Nacalai Tesque). The gene expression and purity of the preparation were examined by SDS-12% PAGE, and the protein bands in gels were stained with Coomassie brilliant blue.

Statistical analysis.

Statistical tests were performed using GraphPad Prism 8 (GraphPad Software). One-way analysis of variance (ANOVA) with Dunnett’s test and Student's t test were used for multiple and pairwise comparisons, respectively. A P value of <0.05 was considered statistically significant.