Role and Function of LitR, an Adenosyl B12-Bound Light-SensitiveRegulator of Bacillus megaterium QM B1551, in Regulation ofCarotenoid Production

Hideaki Takano, Kou Mise, Kenta Hagiwara, Naoya Hirata, Shoko Watanabe, Minami Toriyabe, Hatsumi Shiratori-Takano, Kenji Ueda

Life Science Research Center, College of Bioresource Sciences, Nihon University, Kanagawa, Japan

ABSTRACT

The LitR/CarH family of proteins is a light-sensitive MerR family of transcriptional regulators that contain an adenosyl B12 (co-enzyme B12 or AdoB12)-binding domain at the C terminus. The genes encoding these proteins are found in phylogenetically di-verse bacterial genera; however, the biochemical properties of these proteins from Gram-positive bacteria remain poorly under-stood. We performed genetic and biochemical analyses of a homolog of the LitR protein from Bacillus megaterium QM B1551, aGram-positive endospore-forming soil bacterium. Carotenoid production was induced by illumination in this bacterium. Invivo analysis demonstrated that LitR plays a central role in light-inducible carotenoid production and serves as a negative regula-tor of the light-inducible transcription of crt and litR itself. Biochemical evidence showed that LitR in complex with AdoB12

binds to the promoter regions of litR and the crt operon in a light-sensitive manner. In vitro transcription experiments demon-strated that AdoB12-LitR inhibited the specific transcription of the crt promoter generated by a �A-containing RNA polymeraseholoenzyme under dark conditions. Collectively, these data indicate that the AdoB12-LitR complex serves as a photoreceptorwith DNA-binding activity in B. megaterium QM B1551 and that its function as a transcriptional repressor is fundamental to thelight-induced carotenoid production.

IMPORTANCE

Members of the LitR/CarH family are AdoB12-based photosensors involved in light-inducible carotenoid production in nonpho-totrophic Gram-negative bacteria. Our study revealed that Bacillus LitR in complex with AdoB12 also serves as a transcriptionalregulator with a photosensory function, which indicates that the LitR/CarH family is generally involved in the light-induciblecarotenoid production of nonphototrophic bacteria.

The LitR (light-induced transcription regulator)/CarH family isa MerR family of transcriptional regulatory proteins that con-

tain a cobalamin (Cbl; synonym, vitamin B12)-binding domain atthe C terminus (1–4). The first observation regarding the involve-ment of vitamin B12 in the light-inducible carotenoid (Crt) pro-duction of the Gram-negative gliding bacterium Myxococcusxanthus was presented by Cervantes and Murillo (5). In nonpho-totrophic bacteria, the main function of Crt is to protect cells fromphotooxidative damage by scavenging harmful agents, such as sin-glet and triplet molecular species, produced upon illumination (6,7). Induction of Crt production by light reasonably saves energy inthe dark. Genetic and biochemical studies demonstrated thatCarH is a central regulator of light-inducible Crt production in M.xanthus (3).

Our previous study revealed that the Crt production of Strep-tomyces coelicolor A3(2), a high-GC, Gram-positive bacterium, ismarkedly increased by illumination, and a genetic study of itsregulatory mechanism revealed that LitR of S. coelicolor, a ho-molog of CarH, is a central regulator of crt biosynthesis gene ex-pression (8). The phenotype conferred by the knockout of litRsuggests that LitR serves as a positive regulator of the photodepen-dent expression of the extracytoplasmic function (ECF)-typesigma factor LitS, whose gene is located adjacent to litR. The pre-cise function of LitS was demonstrated using an in vitro transcrip-tion assay; RNA polymerase (RNAP) containing �LitS directs thetranscription of crt biosynthesis genes. We recently reported thatcobalamin biosynthesis genes are required for light-inducible Crtproduction (9).

Our previous study also showed that a LitR homolog encodedin the genome of Thermus thermophilus HB27, a high-GC, Gram-negative thermophilic bacterium, plays a central role as a negativeregulator of light-inducible carotenoid production (4). litR com-prises an operon with ldrP, whose product belongs to the CRP(cyclic AMP receptor protein)/FNR family (10). In vivo and invitro studies showed that LitR directly binds to the intergenic pro-moter region of litR and crtB to repress bidirectional transcriptionunder dark conditions. Light irradiation causes the inactivation ofLitR, which in turn allows the expression of LdrP, directly activat-ing the transcription of the crt biosynthesis gene cluster and thegenes adjacent to litR (2). Interestingly, litR, cobalamin biosynthe-

Received 4 December 2014 Accepted 23 April 2015

Accepted manuscript posted online 27 April 2015

Citation Takano H, Mise K, Hagiwara K, Hirata N, Watanabe S, Toriyabe M, Shiratori-Takano H, Ueda K. 2015. Role and function of LitR, an adenosyl B12-bound light-sensitive regulator of Bacillus megaterium QM B1551, in regulation of carotenoidproduction. J Bacteriol 197:2301–2315. doi:10.1128/JB.02528-14.

Editor: V. J. DiRita

Address correspondence to Hideaki Takano, takano.hideaki@nihon-u.ac.jp.

Supplemental material for this article may be found at http://dx.doi.org/10.1128/JB.02528-14.

Copyright © 2015, American Society for Microbiology. All Rights Reserved.

doi:10.1128/JB.02528-14

July 2015 Volume 197 Number 14 jb.asm.org 2301Journal of Bacteriology

on Decem

ber 24, 2020 by guesthttp://jb.asm

.org/D

ownloaded from

sis, and other light-inducible genes, including the crt genes, wereall found to be located on the large plasmid of this organism (2).

Ortiz-Guerrero et al. (1) showed that CTt2, a chimeric recom-binant protein composed of the N-terminal DNA-binding do-main of CarH and the C-terminal Cbl-binding domain of T. ther-mophilus HB8 LitR, is an adenosyl B12 (AdoB12)-binding proteinwith light-sensitive DNA-binding activity. The CoOC bond ofAdoB12 is chemically cleaved by light, which converts AdoB12 tohydroxocobalamin (OHB12). Therefore, light causes photolysis ofAdoB12 bound to CTt2, causing the conversion of AdoB12 toOHB12, which abolishes its DNA-binding activity. The photolysisof AdoB12 also induces a dramatic conformational change, i.e.,subunit dissociation of CTt2 from a tetramer to a monomer. Thefull-length native CarH protein of T. thermophilus HB8 (TtCarH),an ortholog of LitR with a single-residue substitution, has beenreported to exhibit light-dependent DNA-binding behavior (11).

One of the intriguing features of LitR is the wide distribution ofits homologs in phylogenetically diverse genera of nonpho-totrophic bacteria (1, 12, 13); the genes for these homologs arefrequently flanked by genes for crt biosynthesis and DNA pho-tolyase. In the current study, we examine the role and function ofa LitR homolog of Bacillus megaterium QM B1551, a low-GC,Gram-positive, endospore-forming soil bacterium. Althoughmembers of the LitR/CarH family are widely distributed in bothGram-negative and Gram-positive bacteria, biochemical analysesof this family have been limited to proteins from Gram-negativebacteria (1, 4). Therefore, we believe that a biochemical study of aLitR protein derived from a Gram-positive bacterium providesinsight into the general role of LitR/CarH family members. Theresults obtained in this study indicate that B. megaterium LitR, incomplex with AdoB12, serves as a photoreceptor protein directlyregulating carotenoid production in a light-inducible manner.

MATERIALS AND METHODSBacterial strains, plasmids, and culture media. The wild-type (WT)strains B. megaterium QM B1551, DSM319, and NBRC15308 used in thisstudy were obtained from the Bacillus Genetic Stock Center (BGSC)(Ohio State University; http://www.bgsc.org/), DSMZ (Braunschweig,Germany), and the Biological Resource Center, NITE (NBRC), Kazusa,Japan, respectively. Bacillus subtilis 168 and its �yvqK, �yvrA, �yvrB,and �yvrC mutants were obtained from the National BioResourceProject (NIG), Japan—B. subtilis. Escherichia coli HST08 andRosetta2(DE3)pLysS (TaKaRa Bio Inc., Shiga, Japan) were used as hostsfor DNA manipulation and protein expression, respectively. pUC19 andpUC118 (TaKaRa Bio) were used for general DNA manipulation in E. coli.pT7Blue and pMD19 (TaKaRa Bio) were used for TA cloning of PCR-generated DNA fragments. pGEX-6P-2 (GE Healthcare UK Ltd., Buck-inghamshire, England) and pET51-b(�) (TaKaRa Bio) were used for theoverexpression of B. megaterium QM B1551 LitR and SigA, respectively.pUCTV2, an E. coli-Bacillus temperature-sensitive plasmid shuttle vector(14) carrying a tetracycline resistance gene, was obtained from F. Mein-hardt and used for gene disruption and complementation analysis of B.megaterium. pDG1661 (carrying a chloramphenicol resistance gene andamyE for homologous recombination) was obtained from the BGSC andused as a chromosomal integration plasmid for B. subtilis. Enzymes usedfor DNA manipulation were purchased from TaKaRa Bio and New Eng-land BioLabs (Ipswich, MA). The conditions for culture and genetic ma-nipulation of E. coli have been described previously (15). B. megaterium,B. subtilis, and E. coli were grown at 37°C in Luria-Bertani (LB) medium(15), and 1.0 to 1.5% agar (Kokusan, Tokyo, Japan) was added to preparesolid media. Cells were illuminated in an illuminating incubator (BR-180LF; Taitech, Saitama, Japan) equipped with a white light fluorescent

lamp (20 W; Toshiba, Tokyo, Japan) if required. To enable the selection ofE. coli transformants, 50 �g ml�1 ampicillin, 50 �g ml�1 kanamycin, and10 �g ml�1 tetracycline were added. For selection of Bacillus, 10 �g ml�1

tetracycline, 5 �g ml�1 chloramphenicol, and 1 �g ml�1 erythromycinwere added.

Taxonomic characterization of light-dependent isolates. Eightystrains that exhibited photodependent phenotypes were isolated from soilby cultivation at 28°C or 37°C on LB solid medium. An illuminatingincubator (BR-180LF) equipped with white fluorescent lamps was used.To taxonomically characterize the isolates, total DNAs were extractedfrom the isolates by use of a PurElute bacterial genomic kit (Edge BioSys-tems, Gaithersburg, MD) and used for PCR amplification of the 16S rRNAgene by using the B8F/B1500R primer pair (the oligonucleotide primersused are shown in Table S1 in the supplemental material) (16). The puri-fied PCR amplicon was TA cloned into pMD19. The nucleotide sequenceof the 16S rRNA gene clone was determined with a BigDye Terminatorv3.1 cycle sequencing kit on an ABI3100 automated DNA sequencer(Thermo Fisher Scientific Inc., Waltham, MA). The BLASTN program(http://www.ncbi.nlm.nih.gov/BLAST/) and the SEQUENCE_MATCHprogram from the Ribosomal Database Project (17) were used to comparethe obtained nucleotide sequences of the 16S rRNA gene with those in theGenBank/EMBL/DDBJ nucleotide sequence databases. The analysis re-vealed that 24 isolates exhibited �99.5% sequence identity with Bacillusspp. Of these, 5 strains, affiliated with B. megaterium (strains 164 and 423),Bacillus pumilis (strains 393), and Bacillus sphaericus (strains 294 and319), were used in this study.

Extraction of carotenoids. To examine light-dependent carotenoidproduction, B. megaterium was cultured at 28°C for 2 days in liquid LBmedium under light and dark conditions. Extraction of carotenoids wasperformed as described previously (4). A BR-180LF illuminating incuba-tor equipped with white, blue, green, and red fluorescent lamps (20 W;Toshiba, Tokyo, Japan) was used. Light intensity was measured using amodel LI-250 light meter (Li-Cor Inc., Lincoln, NE). A model U-2800AUV spectrometer (Hitachi High-Tech Science Corp., Tokyo, Japan) and aNanoDrop 2000 spectrophotometer (Thermo Fisher Scientific) were usedto record the absorption spectra of the carotenoid fractions.

Quantification of Cbl. The Cbl content was determined by cultivatingLactobacillus leichmannii ATCC 7830 (ATCC, Manassas, VA) as an indi-cator strain in Difco B12 assay medium (Becton, Dickinson and Co.) ac-cording to the instructions supplied by the manufacturer.

Gene disruption. The temperature-sensitive E. coli-Bacillus shuttlevector pUCTV2 was used for construction of markerless mutants of litR,cobS (BMQ_1998), and the crt operon (BMQ_0654 to BMQ_0656) of B.megaterium QM B1551. pUCTV2 is replicated stably in Bacillus spp. at30°C; at the nonpermissive temperature of 42°C, plasmid replication isdisturbed, eventually leading to loss of the vector (14). To construct eachdisruption plasmid, the upstream region and the downstream region wereamplified by PCR, using the primer sets DRF/DRMR and DRMF/DRDR(litR), DSF/DSMR and DSMF/DSR (cobS), and DCF/DCMR and DCMF/DCR (crt operon) (see Table S1 in the supplemental material). The twoamplified fragments were digested by SphI for litR and BMQ_0654 toBMQ_0656 and by KpnI for cobS, and after purification, the digestionproducts were ligated. The ligated products were then amplified by PCRswith the primer sets DRF/DRDR (litR), DSF/DSR (BMQ_1998), andDCF/DCR (BMQ_0654 to BMQ_0656). The amplified fragments werecloned into pUC118, and the nucleotide sequences were verified by se-quencing using an ABI3100 genetic analyzer or by sequencing performedby Eurofins Genomics K.K. (Tokyo, Japan). Each resulting plasmid wasdigested with BamHI and inserted into the same site of pUCTV2 to gen-erate a disruption plasmid. The resultant plasmid was introduced into B.megaterium QM B1551 by polyethylene glycol (PEG)-mediated proto-plast transformation (18). Subsequently, tetracycline-resistant transfor-mants were grown at 42°C in LB medium containing tetracycline to selectthe single-crossover (plasmid-integrated) colonies. Among these trans-formants, single-crossover strains were selected by PCRs using the appro-

Takano et al.

2302 jb.asm.org July 2015 Volume 197 Number 14Journal of Bacteriology

on Decem

ber 24, 2020 by guesthttp://jb.asm

.org/D

ownloaded from

priate primers. A single selected strain was serially grown in LB mediumwithout tetracycline to induce a second-crossover recombination. Finally,tetracycline-sensitive colonies were selected as double-crossover mutants,yielding the �litR, �crt, and �BMQ_1998 strains. The expected double-crossover-mediated homologous recombination of such derivatives wasverified by PCRs using the appropriate primer sets.

Plasmids for genetic complementation. For genetic complementa-tion, pUCTV2thrC, an integration vector directing the thrC region of B.megaterium, was constructed as follows. The thrC locus is generally used asan integration site for integrative vectors of B. subtilis. The 1.1-kb internalregion of thrC was amplified by a PCR using the primers thrCF and thrCR(see Table S1 in the supplemental material), recovered as an MfeI-digestedfragment, and inserted between the EcoRI sites of pUCTV2, giving rise topUCTV2thr. The intact litR gene with its own promoter region was am-plified by a PCR using the primer set RcF/RcR (see Table S1) and clonedinto the BamHI site of pUCTV2thrC. The resulting plasmid was intro-duced into the �litR strain, and whole-plasmid-integrated strains withsingle crossovers were selected and designated �litR/litR strains. Theseintegrated strains were obtained using gene disruption methods similar tothose described above. In all cases, proper integration was verified by PCR.

Functional analysis of LitR in B. subtilis 168 as a heterologous host.Because the B. subtilis 168 chromosome appeared not to contain a litRhomolog, we used this strain as a heterologous host to investigate thefunction of litR. An amyE-integrative reporter assay vector, pDG1661(19), was used to measure the activity of the light-induced promoter andrelated promoters in B. subtilis in vivo via �-galactosidase activity. TheDNA fragments containing the promoter regions of litR and rpoB(BMQ_0128) were amplified by PCR, using the RZF/RZR primers for litRand the BZF/BZR primers for rpoB (control) (see Table S1 in the supple-mental material), recovered as BamHI/EcoRI-digested fragments, and in-serted between the same sites of pDG1661. In a similar manner, litR withits own promoter was amplified by a PCR using the RZF/RPZR primers.For mutation of the histidine residue at position 190 to alanine, the entirelitR gene was generated by overlap extension PCR, using the primer pairsRZF/R190ZMR and R190ZMF/RZR. The resulting plasmids carryingeach promoter sequence were integrated into the amyE locus of B. subtilis168 and the �yvqK (accession no. BSU33150), �yvrA (accession no.BSU33160), �yvrB (accession no. BSU33170), and �yvrC (accession no.BSU33180) strains by single-crossover recombination. A B. subtilis strainharboring a reporter plasmid was grown in LB liquid medium under lightand dark conditions at 37°C. A �-galactosidase assay was performed onthe bacterial cultures according to the standard method (15), using o-ni-trophenyl-�-D-galactopyranoside (ONPG) as the substrate.

Mutational analysis of the LitR-binding site in the litR promoter.Base substitutions in the litR promoter were generated by a two-stage PCRprocedure using complementary mutagenic primers. The regions up-stream and downstream of a mutation site were amplified by PCR, usingthe primers RZF/MP1R and MP1F/RZR for pM1, RZF/MP2R and MP2F/RZR for pM2, RZF/MP3R and MP3F/RZR for pM3, RZF/MP4R andMP4F/RZR for pM4, RZF/MP5R and MP5F/RZR for pM5, RZF/MP6Rand MP6F/RZR for pM6, RZF/MP7R and MP7F/RZR for pM7, RZF/MP8R and MP8F/RZR for pM8, RZF/MP9R and MP9F/RZR for pM9,RZF/MP10R and MP10F/RZR for pM10, RZF/MP11R and MP11F/RZRfor pM11, and RZF/MP12R and MP12F/RZR for pM12 (see Table S1 inthe supplemental material). The two DNA fragments prepared in thefirst-round PCRs were used as templates in second-round PCRs with theRZF/RZR primers. The fragments amplified by the second-round PCRswere phosphorylated by T4 polynucleotide kinase and then cloned intothe HincII site of pUC118. After the mutations were confirmed by nucle-otide sequencing, the litR promoter sequences containing various muta-tions were excised by use of EcoRI and BamHI and ligated between thesame sites of pDG1661, resulting in plasmids pM1 through pM12. Theplasmids were integrated into the amyE locus of the B. subtilis 168 chro-mosome to measure the �-galactosidase activity.

5=-RACE for determination of transcriptional start points. To deter-mine the transcriptional start sites of the litR, crtI1, and polA promoters,the assignment of the mRNA 5= end was performed by 5= rapid amplifi-cation of cDNA ends (5=-RACE), using a 5=-Full RACE core set (TaKaRaBio), or by directed amplification of transcription start sites (DMTSS)(20). The 5=-Full RACE core set was used according to the manufacturer’sinstructions. DMTSS is a method for determining transcriptional startsites in which cDNA is labeled with a homopolynucleotide tail. ThecDNAs were labeled at the 3=-terminal ends with a homoadenine se-quence [poly(A)] by incubation with terminal deoxynucleotidyltrans-ferase (TdT; TaKaRa Bio) at 37°C for 30 min. The TdT enzyme was inac-tivated by heating to 70°C for 10 min. The first-round PCR was performedwith A1, a specific primer for the target gene, and the DMTSS-1 primer,which contains a homothymidine [poly(T)] sequence. The following re-action program was used: denaturation and amplification for 30 cycles of95°C for 30 s, 55°C for 30 s, and 72°C for 45 s, followed by a step at 72°C for3 min. The second-round PCR was performed using the A2 and DMTSS-4primers. The amplified DNA fragment was cloned into the T-vectorpMD19. The oligonucleotide primers used for this analysis are summa-rized in Table S1 in the supplemental material. The inserted DNA se-quences were sequenced using an ABI3100 genetic analyzer, or sequenc-ing was performed by Eurofins Genomics.

RNA preparation and semiquantitative reverse transcription-PCR(RT-PCR) analysis. B. megaterium strains were grown in liquid LB me-dium at 37°C. Total RNAs of the B. megaterium strains were extractedusing a minikit (Qiagen GmbH, Hilden, Germany) according to the man-ufacturer’s instructions. cDNA synthesis was performed with the Super-Script III first-strand synthesis system (Thermo Fisher Scientific), using 2�g purified total RNA, according to the manufacturer’s protocol. Thedouble-stranded DNA was amplified with GO Taq master mix (PromegaCorp., Madison, WI) and analyzed by agarose gel electrophoresis withethidium bromide staining. The following PCR program was used: initialdenaturation at 95°C for 180 s and amplification for 25 cycles of 95°C for30 s, 55°C for 30 s, and 72°C for 45 s, followed by a step at 72°C for 3 min.The oligonucleotide primers used are listed in Table S1 in the supplemen-tal material.

Preparation of AdoB12-bound LitR and SigA proteins. For the prep-aration of soluble LitR and LitRH190A recombinant proteins, each proteinwas expressed as a protein fused with glutathione S-transferase (GST).The N-terminal GST fusion to LitR was constructed using the PCR clon-ing primers RexF and RexR (see Table S1 in the supplemental material).These primers contain restriction sites for BamHI and EcoRI, respectively.The amplicon was cloned into the same site of the expression plasmidpGEX-6P-2 (GE Healthcare), giving rise to the plasmid pGEXLitR. Togenerate LitRH190A, a LitR protein in which histidine is replaced withalanine at amino acid position 190, the entire litR gene was generated byoverlap extension PCR, using the primer pairs RexF/R190ZMR andR190ZMF/RexR. The two DNA fragments amplified in the first-roundPCR were used as the template in a second-round PCR with the primer setRexF/RexR. The amplified fragment was digested with EcoRI and BamHIand ligated between the same sites of the pGEX-6P-2 plasmid, resulting inpGEXLitRH190A. pETSigA, an expression vector for SigA, was constructedbased on pET-51b(�). sigA was amplified with the primer set AexF/AexRand cloned into the NcoI and SacI sites of pET-51b(�). SigA contained aHis10 tag at the C terminus. All genes cloned above were under the regu-lation of an isopropyl �-D-thiogalactoside (IPTG)-inducible promoter.

E. coli Rosetta2(DE3)pLysS cells harboring pGEXLitR,pGEXLitRH190A, or pETSigA were first cultured overnight at 28°C in LBliquid medium. Subsequently, each seed culture was inoculated at 1% into100 ml of LB medium prepared in a 500-ml baffled Erlenmeyer flask. After3 h of cultivation at 28°C with rotary shaking (135 rpm), IPTG was addedto a final concentration of 0.1 mM, with continued incubation at 28°C for4 h. The E. coli cells thus obtained were harvested by centrifugation, sus-pended in phosphate-buffered saline (PBS; 140 mM NaCl, 2.7 mM KCl,10 mM Na2PO4, and 1.8 mM KH2PO4), and disrupted with a sonicator

LitR of B. megaterium

July 2015 Volume 197 Number 14 jb.asm.org 2303Journal of Bacteriology

on Decem

ber 24, 2020 by guesthttp://jb.asm

.org/D

ownloaded from

(Astrason XL2020; Misonix, Farmingdale, NY) or an APV Gaulin 1000mechanical cell presser (APV Homogenisers, As, Denmark). The cell ex-tract was centrifuged at 70,000 g for 30 min, and the resultant superna-tant was used for affinity chromatography. The GST tag was removed bytreatment with PreScission protease (GE Healthcare). The following stepwas performed under conditions of dim light or under dark conditions.To prepare the LitR protein associated with AdoB12, purified LitR wasincubated in the presence of a 10-fold molar excess of AdoB12 (Sigma-Aldrich Corp., St. Louis, MO) for 1 h at 37°C. Dialysis against PBS wasperformed to remove unbound AdoB12 from the mixture. A U-2800A UVspectrometer or a NanoDrop 2000 spectrophotometer was used to recordthe absorption spectra of the resultant AdoB12-LitR complexes. The pro-tein concentration was measured with a protein assay kit (Bio-Rad Labo-ratories, Hercules, CA).

Gel filtration column chromatography. The molecular size of LitRwas estimated by gel filtration of 1.1 mg of the purified recombinantprotein. Gel filtration was performed using a Superdex 200 HR 10/30column on an ÄKTA fast-performance liquid chromatography (FPLC)system (GE Healthcare) according to the manufacturer’s recommenda-tions. The column was developed with 1 PBS (140 mM NaCl, 2.7 mMKCl, 10 mM Na2PO4, and 1.8 mM KH2PO4) at a flow rate of 0.25 mlmin�1. Molecular size standards (ferritin, conalbumin, carbonic anhy-drase, and RNase, with sizes of 440, 75, 29, and 13.7 kDa, respectively)included in a gel filtration calibration kit (GE Healthcare) were used todetermine the void volume (V0), and vitamin B12 (1.35 kDa) was used toestimate the total bed volume (Vt).

Analytical ultracentrifugation. Analytical ultracentrifugation-sedi-mentation velocity (AUC-SV) experiments were carried out with an Op-tima XL-1 analytical ultracentrifuge (Beckman Coulter, Inc., Pasadena,CA) with a four-hole An60Ti rotor. The nonilluminated and illuminatedAdoB12-LitR proteins (A280 0.6) were centrifuged at 40,000 rpm and20°C. Protein concentrations were monitored at 280 nm. Data were ana-lyzed with a continuous c(s) distribution model by using the softwareprogram SEDFIT (version 11.8).

Gel shift assay. DNA binding was determined using a gel shift assay asdescribed previously (4). Probe DNA fragments were generated by PCR,using the primer sets RFPF/RFPR (PlitR), CFPF/CFPR (PcrtI1), and PAF/PAR (PsigA) (see Table S1 in the supplemental material), with the chro-mosome of a QM B1551 WT strain as the template. The litR promoterwith the mutated LitR-binding site was amplified using the primer setRFPF/RFPR, with pM1 to pM12, constructed as described above, as tem-plates. These probe DNA fragments were labeled at the 5= end with[�-32P]ATP (PerkinElmer, Inc., Boston, MA) by using T4 polynucleotidekinase. A total of 0.5 to 5.0 ng of 32P-labeled probe (10,000 to 20,000 cpm)was mixed with 0 to 32 pmol of recombinant LitR, prepared as describedabove; the mixture was then incubated at 37°C for 30 min in 50 �l ofbinding buffer [containing 10 mM Tris-HCl (pH 7.0), 50 mM KCl, 1 mMEDTA, 1 mM dithiothreitol, 10% (vol/vol) glycerol, 1 �g poly(dI-dC),and 50 �g ml�1 of bovine serum albumin]. Under light conditions, thesample was illuminated with white light at approximately 42 �mol s�1

m�2. Following 10 min of exposure to white light, reaction mixes wereincubated at 37°C for 30 min, and specific DNA-protein complexes wereseparated from free probe in a nondenaturing polyacrylamide gel con-taining 6% acrylamide. The gels were dried, radioactive signals were de-tected by exposing dried gels to a Fuji imaging plate (Fuji Film, Tokyo,Japan), and images were scanned with a Typhoon 9410 or a Typhoon FLA9500 image analyzer (GE Healthcare).

DNase I footprint analysis. The LitR-binding sites in the PlitR andPcrt promoter regions were determined using DNase I footprint analysisas described previously (4). 32P-labeled DNA fragments were prepared byPCR, using the RFPF/RFPR primers for PlitR and the CFPF/CFPR prim-ers for Pcrt (see Table S1 in the supplemental material). Each reactionmixture (50 �l) contained 10 kcpm 32P-labeled DNA probe; 0 to 8 pmolLitR; 20 mM Tris-HCl, pH 7.2; 1.0 mM EDTA-NaOH, pH 8.0; 50 mMNaCl; and 10% glycerol. After sample incubation at 25°C for 30 min,

DNase I was added to a final concentration of 20 �g ml�1, and the mixturewas further incubated for 1 min. The reaction was terminated by adding100 �l of stop solution (containing 100 mM Tris-HCl, pH 8.0; 100 mMNaCl; 1% sodium N-lauroyl sarcosinate; 10 mM EDTA-NaOH, pH 8.0;and 25 mg ml�1 salmon sperm DNA) and 300 �l of phenol:chloroform(1:1). After ethanol precipitation, the pellet was washed with 80% ethanol,dissolved in a 6-�l formamide-dye mixture, and run in a 6% polyacryl-amide gel. To determine the LitR-binding site by high-resolution analysis,Maxam-Gilbert sequencing ladders (A�G reactions) generated from the32P-labeled probe DNA fragment were used as a reference.

In vitro runoff transcription assay. The in vitro runoff transcriptionassay was performed as described previously (4, 21, 22). DNA templatescontaining the transcriptional start sites of PlitR and Pcrt were generatedby PCR, using the primers PLF/PLR for the PlitR, M7, and M12 templates(233 bp), PCF/PCR for the Pcrt template (306 bp), and PPF/PPR for thePpolA template (302 bp) (see Table S1 in the supplemental material). Atotal of 0.5 pmol of template DNA was mixed with 2 pmol commercialRNA polymerase core enzyme of E. coli (AR Brown, Tokyo, Japan), 100nmol of each ribonucleotide, including [�-32P]CTP (PerkinElmer), and 0to 8 pmol of LitR. The RNAP holoenzyme was reconstituted by combin-ing the commercial RNAP core enzyme and the B. megaterium �A recom-binant protein purified from E. coli. Transcripts were analyzed by poly-acrylamide gel electrophoresis. Marker 10 (pBR322 MspI digest) (NipponGene Co., Ltd., Tokyo, Japan) labeled with [�-32P]ATP was used as astandard. The LitR recombinant proteins were irradiated at approxi-mately 42 �mol s�1 m�2 with blue light ( max 450 nm), green light( max 530 nm), and red light ( max 660 nm). The irradiance providedby the various light treatments was measured using a model LI-250 lightmeter.

RESULTSLight-induced carotenoid production and the distribution ofthe LitR/CarH family in Bacillus spp. The ability of Bacillus spp.to produce carotenoids has been reported previously (23–27);however, the correlation between carotenoid production and lighthas not been investigated in this genus. In our screen of light-responsive bacteria (see Materials and Methods), the strains iso-lated from soil were identified as B. megaterium, B. sphaericus, andB. pumilus by 16S rRNA gene-based phylogenetic analysis. Thesebacteria produced a pale or strong yellow pigment under lightconditions; in contrast, the nonilluminated colonies of this strainappeared white or pale yellow (Fig. 1). This suggests that a light-responsive regulatory mechanism also exists in this group of bac-teria.

To examine whether the yellow pigment was composed ofcarotenoids, we selected B. megaterium QM B1551, a strain whosegenome has been sequenced completely (28), as a model. Thisbacterium exhibited photodependent carotenoid production(Fig. 1). Carotenoid production was induced by blue and greenlight but not by red light (Fig. 2; see Fig. S1 in the supplementalmaterial). The UV-visible absorption spectrum of a methanol ex-tract of the illuminated cells showed a typical carotenoid profile,exhibiting multiple absorption peaks near 450 nm (Fig. 2). Thisabsorption spectrum was not observed for the extract from aknockout mutant for the carotenoid biosynthesis gene (see be-low). This confirmed that the spectrum observed here is that ofcarotenoids. We tried to purify the carotenoids produced by B.megaterium QM B1551, but their high photosensitivity createddifficulties in further purification and identification. We also con-firmed that B. megaterium DSM319 (28), another strain whosegenome has been sequenced, also exhibited photodependent car-otenoid production (Fig. 1 and 2).

Takano et al.

2304 jb.asm.org July 2015 Volume 197 Number 14Journal of Bacteriology

on Decem

ber 24, 2020 by guesthttp://jb.asm

.org/D

ownloaded from

In accordance with the observation that B. megaterium QMB1551 undergoes light-dependent carotenoid production, thisbacterium was found to possess two putative crt biosynthesisoperons (see Fig. S2A in the supplemental material). TheBMQ_0654-BMQ_0656 operon encodes a putative phytoene de-saturase (BMQ_0654; designated CrtI1), a phytoene desaturase(BMQ_0655; CrtI2), and a phytoene synthase (BMQ_0656;CrtB), whereas the BMQ_2952-BMQ_2949 operon encodes aconserved hypothetical protein (BMQ_2952), a putative phos-pholipid/glycerol acyltransferase (BMQ_2951), a putative glyco-syltransferase (BMQ_2950), and a putative phytoene desaturase(BMQ_2949; CrtI3). It is probable that CrtB and CrtI are involvedin lycopene biosynthesis. BMQ_2950 may be involved in the gly-cosylation of carotenoids, because it contains a glycosyltransferasedomain. B. megaterium DSM319 also harbors the crt operon in thesame gene organization (data not shown). In contrast, the ge-nomes of closely related Bacillales species (including Bacillusclausii, Bacillus cellulosilyticus, Bacillus thuringiensis, Bacillus an-thracis, Bacillus cereus, and Bacillus pseudofirmus) retain single-locus crt operons (see Fig. S2A).

B. megaterium retains a litR homolog (BMQ_4356 for QM

B1551 [see Fig. S2B in the supplemental material] and BMD_4342for DSM319). Unlike those of T. thermophilus and M. xanthus, thelitR homolog of B. megaterium is located at a locus different fromthat of the putative carotenoid biosynthesis genes. The LitR pro-tein of B. megaterium QM B1551 consists of 303 amino acid resi-dues, exhibiting 28% amino acid sequence identity to that of T.thermophilus (see Fig. S3). The N-terminal (corresponding to N8to Y45) and C-terminal (K94 to I167 and K179 to F268) regionsexhibit distinct similarities to the helix-turn-helix domain of aMerR family regulatory protein (E value 2e�08) and the B12-binding domain (E value 1.3e�15 and 6e�14, respectively)(see Fig. S3). The gene organization at the litR locus is conserved inthe genomes of some Bacillus spp., including B. clausii, B. cellulosi-lyticus, B. thuringiensis, B. anthracis, B. cereus, and B. pseudofirmus(see Fig. S2B).

Knockout of the crt, litR, and cbl genes in B. megaterium QMB1551. To verify whether the crtI1-crtI2-crtB (BMQ_0654 toBMQ_0656) genes are involved in crt biosynthesis in B. megate-rium QM B1551, a markerless mutant lacking these three codingsequences (�crt) was generated. As shown in Fig. 2, the �crt mu-tant was defective in carotenoid production under both dark and

FIG 1 Light-inducible carotenoid production of Bacillus spp. Colonies of B. megaterium, B. pumilus, and B. sphaericus, originally isolated from soil, and of B.megaterium NBRC 15308, QM B1551, and DSM319 were grown at 37°C on solid LB medium under dark and light conditions. Colonies producing carotenoidsappear yellow, whereas nonproducing colonies appear white or cream.

LitR of B. megaterium

July 2015 Volume 197 Number 14 jb.asm.org 2305Journal of Bacteriology

on Decem

ber 24, 2020 by guesthttp://jb.asm

.org/D

ownloaded from

light conditions. This indicates that the gene products of crtI1-crtI2-crtB (here designated the crt operon) are responsible for car-otenoid biosynthesis.

To investigate the role of litR, a markerless mutant of litR (des-ignated �litR) was constructed. The resultant �litR mutant pro-duced large amounts of carotenoids under both dark and lightconditions (Fig. 2). To generate a genetically complementedstrain, we originally constructed a chromosomal integration vec-tor, pUCTV2thr (see Materials and Methods). The geneticallycomplemented strain partially restored the light-inducible caro-tenoid production. This was probably due to the introduction ofthe litR gene into a genomic region different from the originalcoding region. These results indicate that LitR is involved in thelight-dependent genetic control of carotenoid production, basedon its function as a negative regulator.

Concordant with the fact that B. megaterium was previouslyused for the commercial production of cobalamin, B. megateriumhas a complete anaerobic pathway for cobalamin biosynthesis(29), including a large (cbiWHXJCDETLFGA-cysG-cbiY-btuR)and a small (cbiB-cobDUSC) operon (30, 31). Although Cbl pro-duction by B. megaterium has been well documented, the correla-tion between vitamin B12- and light-dependent carotenoid pro-duction has not yet been studied in this bacterium. B. megateriumQM B1551 encodes two proteins involved in the final step of the

oxygen-independent pathway of AdoB12 biosynthesis: (i) adeno-sylcobinamide-GDP ribosyltransferase (BMQ_1998) and (ii) AT-P:cob(I)alamin adenosyltransferase (BMQ_2001). We con-structed a markerless null mutant of BMQ_1998. As shown in Fig.2, the �BMQ_1998 strain showed weak carotenoid production,irrespective of illumination. This result indicates that BMQ_1998serves as a main biosynthesis enzyme to produce AdoB12 and sug-gests that AdoB12 is involved in the light-inducible carotenoidproduction.

Transcriptional analysis. To examine whether photodepen-dent carotenoid production is regulated at the transcriptionallevel, we performed a transcriptional analysis. Prior to the analy-sis, transcriptional start sites were determined by 5=-RACE withrespect to the promoter regions preceding litR and crtI1 (see Ma-terials and Methods). We also determined the transcriptional startsite of polA (BMQ_4764; gene encoding DNA polymerase I),which was used as a control promoter in the in vitro runoff tran-scriptional assay (see below). Using total RNA isolated from illu-minated cells, 5=-RACE assigned a single transcriptional start site40 bp upstream of the translational initiation codon (ATG) of litR(Fig. 3A), 32 bp upstream for crtI1 (Fig. 3A), and 34 bp upstreamfor polA. The potential �35 and �10 sequences of litR(TTTACAN17TATAAT), crtI (TTGTATN17TATAAT), and polA(TTGTCCN18TAAAAT) were similar to promoter sequences rec-

FIG 2 Light-induced carotenoid production in Bacillus megaterium. UV-visible spectra are shown with regard to the crude carotenoid fraction extracted fromcells of B. megaterium QM B1551 WT, �litR, �litR/litR, �BMQ_1998, and �crt strains and B. megaterium DSM319 grown at 37°C for 15 h under dark and lightconditions.

Takano et al.

2306 jb.asm.org July 2015 Volume 197 Number 14Journal of Bacteriology

on Decem

ber 24, 2020 by guesthttp://jb.asm

.org/D

ownloaded from

ognized by the B. subtilis �A-RNA polymerase (Fig. 3A), for whichthe consensus sequence is TTGACAN16 –18TATAAT (32). Thesepromoters are designated PlitR, PcrtI1, and PpolA, respectively.

We examined the mRNA levels of litR and crtI1 under dark andlight conditions by semiquantitative RT-PCR analyses. An essen-tial sigma factor gene, sigA, was used as the control, because illu-mination did not affect the growth of B. megaterium (data notshown), which suggests that transcription of sigA is not affected byillumination.

As shown in Fig. 4A, the transcription of litR and crtI1 in theWT strain increased in response to illumination. In contrast, crtI1transcription occurred in the �litR strain under both conditions.We were unable to evaluate the transcriptional activity of litR inthis mutational background, because the coding sequence hadbeen deleted, from the initiation codon to the stop codon. Thetranscription of litR and crtI1 in the complemented strains waspartially induced by light. This complementation analysis sug-gested that mutational effects were due to gene disruption.

The BMQ_2952-BMQ_2949 operon (see Fig. S2A in the sup-plemental material) encodes the putative enzymes involved in themodification of carotenoids, such as glycosylation and acylation.The transcription level of BMQ_2949 (crtI3) was constitutive

(data not shown), indicating that LitR regulates only BMQ_0654to BMQ_0656 and thus is involved in the early step of carotenoidbiosynthesis.

Functional analysis of LitR in a heterologous host. In order toinvestigate whether LitR serves as a photosensitive regulator andwhether an additional protein is required for light-inducible tran-scription, we examined the in vivo function of litR in B. subtilis168, a heterologous host. Since B. subtilis 168 does not retain ahomolog of the LitR/CarH family (see Fig. S2B in the supplemen-tal material), it was suitable for the analysis. A chromosomal inte-gration plasmid, pDG1661, carrying each gene or promoter fusedto the lacZ gene, was used to monitor light-dependent transcrip-tion via �-galactosidase activity (see Materials and Methods).

The PlitR-litR-lacZ-containing B. subtilis transformantshowed a higher transcriptional activity under light conditionsthan under dark conditions (Fig. 4B). In contrast, B. subtilis 168strains carrying PlitR-lacZ (the construct lacking the LitR codingsequence) and PrpoB-lacZ did not exhibit significant differencesin transcriptional activity between light and dark conditions. Thisresult suggests that LitR expressed in the heterologous host servesas a photodependent transcriptional regulator, similarly to thecase in its original host, B. megaterium QM B1551.

FIG 3 Promoter regions of litR and crtI1. (A) Nucleotide sequences of the promoter regions preceding litR and crtI1. The C-terminal amino acid sequence ofBMQ_4357 and the N-terminal amino acid sequences of LitR and CrtI1 are also shown. Possible �10 and �35 hexamer sequences of the litR and crtI1 promotersare indicated by lowercase letters. The transcriptional start sites (designated �1), determined by 5=-RACE, are indicated by bent arrows. (B) Alignment ofLitR-binding sequences. The consensus sequence is also shown.

LitR of B. megaterium

July 2015 Volume 197 Number 14 jb.asm.org 2307Journal of Bacteriology

on Decem

ber 24, 2020 by guesthttp://jb.asm

.org/D

ownloaded from

We also constructed a 168 strain harboring litRH190A, whichencodes a protein in which the histidine at residue 190 (190His) isreplaced with alanine (see Fig. S3 in the supplemental material).190His of LitR is a conserved amino acid residue in the Cbl-bindingdomain predicted to be involved in the association of LitR withAdoB12. As shown in Fig. 4B, the transformant harboring thelitRH190A-lacZ fusion showed constitutive transcriptional activityunder both dark and light conditions. The results indicate that litRis used exclusively for light-responsive transcription and that190His is essential for the function of LitR.

The addition of Cbl derivatives to LB medium was not requiredfor the light-inducible transcription in B. subtilis described above.Since B. subtilis retains yvrA, yvrB, and yvrC, encoding a putativeB12 uptake transporter, and yvqK, encoding a homolog of ATP:cob(I)alamin adenosyltransferase, which is involved in the finalstep of AdoB12 biosynthesis (33), we assumed that an intermediateof Cbl may be taken up from LB medium and used for the synthe-sis of AdoB12. Actually, our bioassay using L. leichmannii as anindicator strain demonstrated that LB medium contains 0.2 ng/mlCbl. On the other hand, fresh LB medium in which B. subtilis wascultured overnight contained 0.1 ng/ml Cbl. This indicates thatabout half of the Cbl contained in LB was utilized by this bacte-rium. We also studied the light-inducible transcription in the mu-tants of the yvr and yvq genes and confirmed that it did not occurin those mutants (see Fig. S4 in the supplemental material). Thisled us to the assumption that an inactive cobalamin in LB me-

dium, such as cyanocobalamin (CNB12), is taken up by theYvrABC transporter and converted into AdoB12 by YvqK.

Association of a LitR recombinant protein with AdoB12. Toexamine the biochemical properties of LitR and LitRH190A, therecombinant proteins were prepared in an E. coli expression sys-tem. Both proteins were successfully obtained as soluble proteinsin the E. coli expression system (see Materials and Methods).

The interaction of AdoB12 with the prepared LitR proteins wasanalyzed based on their absorption spectra, using dialysis equili-bration and a spectrophotometer (Fig. 5). Unexpectedly, the UV-visible absorption spectrum of the AdoB12-treated LitR proteinshowed a typical OHB12 profile, exhibiting multiple absorptionpeaks near 350 nm and 550 nm. Illumination did not cause anyspectrum change in the AdoB12-LitR complex. In contrast,LitRH190A treated with AdoB12 did not show the absorption spec-trum of either AdoB12 or OHB12. Henceforth, we refer to the com-plex between LitR and AdoB12 as AdoB12-LitR.

Light-induced subunit dissociation of the AdoB12-LitR pro-tein complex. To determine their subunit structures based onestimated relative molecular masses, the illuminated and dark-incubated AdoB12-LitR proteins were subjected to analytical gelfiltration chromatography (see Materials and Methods). The LitRprotein, purified to near homogeneity, migrated as a ca.36,500-Mr protein in an SDS-polyacrylamide gel, which was al-most identical to the calculated molecular mass (35.88 kDa) (datanot shown). The dark-incubated and illuminated AdoB12-LitR

FIG 4 Transcriptional analysis of litR and crt promoters in B. megaterium QM B1551 (A) and in a heterologous host, B. subtilis 168 (B). (A) Quantification oftranscripts using semiquantitative RT-PCR. The amounts of litR, crtI1, and sigA (control) transcripts produced in the wild-type strain (WT), the litR mutant(�litR), and the genetically complemented litR mutant (�litR/litR) were estimated. cDNA synthesis was performed in the presence (�) or absence (�) of reversetranscriptase (RT). ND, not detected due to elimination of the primer annealing site. (B) A chromosomal integration plasmid, pDG1661, was used to monitorthe transcriptional activities of promoters preceding litR via measuring the �-galactosidase activity in B. subtilis 168. The promoter preceding rpoB (encoding the�-subunit of RNA polymerase) (PrpoB) was used as a control. The transformants were grown in liquid LB medium under dark conditions (circles) or lightconditions (squares).

Takano et al.

2308 jb.asm.org July 2015 Volume 197 Number 14Journal of Bacteriology

on Decem

ber 24, 2020 by guesthttp://jb.asm

.org/D

ownloaded from

proteins were eluted from the gel filtration column at positionscorresponding to Mr values of 139,000 and 89,000 (Fig. 6A). Thelight-dependent change of the molecular size was also observed byanalytical ultracentrifugation (Fig. 6B); the molecular sizes of thedark-incubated and illuminated AdoB12-LitR proteins were esti-mated to correspond to Mr values of 109,000 and Mr values of60,700 and 64,200, respectively. We speculate that AdoB12-LitR inthe dark or in the light forms a tetramer or a dimer, respectively, insolution. We also estimated the relative molecular sizes of theB12-free Apo-LitR proteins by using gel filtration chromatogra-phy. Both the dark-incubated and illuminated Apo-LitR proteinswere eluted at positions corresponding to an Mr of 123,000 (seeFig. S5 in the supplemental material). This result indicates thatAdoB12-LitR of B. megaterium undergoes subunit dissociation inresponse to illumination and that AdoB12 is essential for the light-induced dissociation.

Light-sensitive DNA-binding activity of AdoB12-LitR. Wenext performed a gel shift assay to examine the interaction ofAdoB12-LitR with the light-inducible promoters PlitR and PcrtI1.The results showed that AdoB12-LitR prepared under dark condi-tions caused a specific gel shift of a DNA probe containing PlitRand PcrtI1 (Fig. 7A). In contrast, irradiation of AdoB12-LitR priorto its addition to the reaction mixture did not retard the DNA

FIG 5 Absorption spectra of LitR recombinant proteins. Spectra are shownfor free AdoB12 (A), free OHB12 (B), AdoB12-treated LitRH190A (C), andAdoB12-LitR (D) under dark or illuminated conditions.

FIG 6 Light-inducible subunit dissociation of AdoB12-LitR as analyzed by gelfiltration (A) and analytical ultracentrifugation (B). (A) To estimate the rela-tive molecular mass of the AdoB12-LitR protein, the purified recombinantproteins were analyzed by gel filtration. The elution volumes of AdoB12-LitRon a Superdex 200 HR 10/30 column are indicated. The panel shows the elu-tion profiles for AdoB12-LitR under dark and illuminated conditions. Eachprotein examined eluted at positions corresponding to the tetrameric (Mr,13,900) and dimeric (Mr, 8,900) structures. (B) Sedimentation velocity exper-iments were performed with nonilluminated and illuminated AdoB12-LitR.The sedimentation coefficients [c(s)], i.e., 4.35S and 6.45S for nonilluminatedAdoB12-LitR and 4.26S for illuminated AdoB12-LitR, are shown. The molecu-lar sizes of the proteins were estimated to be Mr of 60,700 and 109,000 and of64,200 on the basis of the conversion of c(s) to c(M).

LitR of B. megaterium

July 2015 Volume 197 Number 14 jb.asm.org 2309Journal of Bacteriology

on Decem

ber 24, 2020 by guesthttp://jb.asm

.org/D

ownloaded from

probe, or it produced a smeared signal. Similarly, the AdoB12-unbound LitR protein did not cause a band shift (data not shown).The gel shift induced by the AdoB12-LitR complex was not de-tected with a DNA probe containing the sigA promoter region. Wealso examined the loss of the DNA-binding activity of AdoB12-LitR with increasing illumination times (Fig. 7B). At least 3 min ofexposure to light at 100 �mol s�1 m�2 was required for the inac-tivation of AdoB12-LitR.

We also examined the DNA-binding activity of the LitRH190A

mutant by using this assay (Fig. 7A). No marked gel shift wasdetected using the LitRH190A protein, indicating that the 190Hisresidue is essential for the AdoB12 interaction and is involved inthe DNA-binding activity. Collectively, these results showed thatAdoB12-LitR is a photosensitive DNA-binding protein and thatAdoB12 is required for the light-sensitive DNA-binding activity ofLitR.

Determination of LitR-binding sites. DNase I footprint anal-ysis was performed to identify each of the AdoB12-LitR bindingsites in PlitR and PcrtI1. Figure 8 shows the region protected bybinding of AdoB12-LitR prepared under dark conditions. ForPlitR, nucleotide positions �36 to �67 of the sense strand andpositions �27 to �69 of the antisense strand, with respect to thetranscriptional start site, were protected from DNase I digestion(also see Fig. 3A). For PcrtI1, nucleotide positions �13 to �22 ofthe sense strand and positions �15 to �18 of the antisense strandwere protected (Fig. 8 and 3A).

Alignment of two LitR-binding sequences yielded a definitiveconsensus sequence, 5=-TG(T/A)A(C/T)A-N16-T(G/A)TA(C/T)A-3= (Fig. 3B). To confirm whether the predicted consensussequence is recognized by LitR, we performed a reporter assayusing B. subtilis 168 harboring litR-lacZ with respect to the varioustypes of mutated promoter. The mutated nucleotide sequences are

FIG 7 Gel shift assay of AdoB12-LitR. (A) The amounts of AdoB12-LitR and AdoB12-LitRH190A used in lanes 1, 2, 3, 4, and 5 were 0, 4, 8, 16, and 32 pmol,respectively. Purified AdoB12-LitR or LitRH190A was mixed with the probes for PlitR (135 bp), PcrtI1 (146 bp), and PsigA (230 bp) and applied to a nondenaturingpolyacrylamide gel. Open and closed triangles indicate the probe and the protein-DNA complex, respectively. (B) AdoB12-LitR (4 pmol) was incubated in thedark (9 min) or under white light conditions (0 to 9 min) prior to the addition of a 32P-labeled probe for PlitR. Only the shifted band (the LitR-AdoB12–DNAcomplex) is shown.

Takano et al.

2310 jb.asm.org July 2015 Volume 197 Number 14Journal of Bacteriology

on Decem

ber 24, 2020 by guesthttp://jb.asm

.org/D

ownloaded from

summarized in Table 1. A reporter assay with M7 (5=-TGACAC-N16-TATACA-3=), M9 (5=-TGAACA-N14-TATACA-3=), M11(5=-TGAACA-N16-TATACC-3=), and M12 (5=-TGAACA-N16-TATCAC-3=) showed identical transcription levels under both darkand light conditions (see Fig. S6 in the supplemental material).Consistent with the results of the reporter assay, the recombinantAdoB12-LitR protein was unable to bind to the promoter region ofM7, M9, or M12 in the gel shift assay (see Fig. S7). These resultsindicate that the consensus sequence as well as the length of thespacer is important for the recognition of LitR. On the other hand,AdoB12-LitR was able to bind M11. The reason for this inconsis-tency is not known, but the mutated sequence may allow weak andconstitutive binding of LitR, hence resulting in low but light-in-dependent transcription.

In vitro runoff transcription analysis. To reproduce the light-dependent transcription observed in vivo, we performed an invitro transcriptional analysis with the following components: theAdoB12-LitR complex, the C-terminally histidine-tagged recom-binant �A protein of B. megaterium (see Materials and Methods),

and a commercially available E. coli RNA polymerase core en-zyme. Concordant with the fact that the �10 and �35 regions ofPlitR, PcrtI1, and PpolA showed high similarity to a consensus se-quence recognized by B. subtilis 168 �A, the RNA polymerase coreenzyme containing �A (E�A) generated transcripts of the expectedmolecular sizes according to the relative positions of the transcrip-tional start sites (87 bases for PlitR, 87 bases for PcrtI1, and 91bases for PpolA) (Fig. 9A). E�A also generated a specific transcriptfor PpolA, irrespective of the presence of the AdoB12-LitR complex(Fig. 9A), indicating that AdoB12-LitR does not act on PpolA.

We then examined the light-dependent repressor activity ofAdoB12-LitR. As shown in Fig. 9A, the amounts of litR and crtI1transcripts generated by E�A were markedly reduced as theamount of AdoB12-LitR protein (0 to 8 pmol) added to the reac-tion mixture increased under dark conditions. Meanwhile, theaddition of illuminated AdoB12-LitR did not inhibit their tran-scription. The AdoB12-LitR-dependent repression in the dark wasnot observed when the mutated M7 (5=-TGACAC-N16-TATACA-3=) and M12 (5=-TGAACA-N16-TATCAC-3=) promoters were

FIG 8 DNase I footprint analysis to determine the LitR-binding sites in the litR and crtI1 promoters. The assay was performed on the sense (�) and antisense(�) strands. The amounts of recombinant AdoB12-LitR used were 0 pmol (lanes 1 and 7), 0.5 pmol (lanes 2), 1 pmol (lanes 3), 2 pmol (lanes 4), 4 pmol (lanes5), and 8 pmol (lanes 6) for the litR promoter and 0 pmol (lanes 1 and 8), 0.25 pmol (lanes 2), 0.5 pmol (lanes 3), 1 pmol (lanes 4), 2 pmol (lanes 5), 4 pmol (lanes6), and 8 pmol (lanes 7) for the crtI1 promoter. Positional numbering is based on the transcriptional start point of each promoter, numbered as �1. The DNaseI digests were run with chemically cleaved probes (G�A lanes for both strands).

LitR of B. megaterium

July 2015 Volume 197 Number 14 jb.asm.org 2311Journal of Bacteriology

on Decem

ber 24, 2020 by guesthttp://jb.asm

.org/D

ownloaded from

used as templates. These results indicate that the AdoB12-LitRcomplex serves as an active repressor of the transcriptional initia-tion at PlitR and PcrtI1 and that the repression is abolished byillumination. The light-dependent transcriptional repression wasnot observed with respect to the AdoB12-treated LitRH190A mutant

protein (Fig. 9B), while the weak transcriptional repression ofPcrtI1 by AdoB12 plus LitRH190A was still observed. The reason forthis phenomenon is not clear. The mutation may confer a confor-mational change that causes a light-independent repression spe-cific to the crtI promoter.

TABLE 1 Nucleotide sequences of mutated LitR-binding sites and DNA-binding activities of AdoB12-LitR to the mutated LitR-binding sites

Nucleotide LitR-binding activitya Mutation site in litR promoterb

WT �� AGCCTTGAACAAAATCTTTACATAGCATATACAAATTTTM1 �� AGCCTCGAACAAAATCTTTACATAGCATATACAAATTTTM2 �� AGCCTTTAACAAAATCTTTACATAGCATATACAAATTTTM3 �� AGCCTTGCACAAAATCTTTACATAGCATATACAAATTTTM4 � AGCCTTGACCAAAATCTTTACATAGCATATACAAATTTTM5 � AGCCTTGAAAAAAATCTTTACATAGCATATACAAATTTTM6 �� AGCCTTGAACCAAATCTTTACATAGCATATACAAATTTTM7 � AGCCTTGACACAAATCTTTACATAGCATATACAAATTTTM8 �� AGCCTGTAACAAAATCTTTACATAGCATATACAAATTTTM9 � AGCCTTGAACAAAATCTT--CATAGCATATACAAATTTTM10 �� AGCCTTGAACAAAATCTTTACATAGCAGATACAAATTTTM11 �� AGCCTTGAACAAAATCTTTACATAGCATATACCAATTTTM12 � AGCCTTGAACAAAATCTTTACATAGCATATCACAATTTTa �, low; ��, high; �, none.b The predicted consensus sequence for LitR binding is shown in bold. Mutated and deleted residues are indicated by underlining and dashes, respectively.

FIG 9 In vitro runoff transcription assay. (A) The indicated amounts of the E. coli RNA polymerase core enzyme containing B. megaterium �A (E�A) and theAdoB12-LitR recombinant protein were added to the reaction mixture with the promoter DNA fragments. Transcripts of the predicted lengths were detected: 87bp for PlitR, M7, and M12; 91 bp for PcrtI1; and 87 bp for PpolA. (B) The AdoB12-LitRH190A recombinant protein was added to the reaction mixture. (C) Theeffects of AdoB12, OHB12 (hydroxocobalamin), CNB12 (cyanocobalamin), and MeB12 (methylcobalamin) were examined using AdoB12-LitR and the DNAfragments containing PlitR. (D) The effects of different light wavelengths, i.e., white light, 365 nm, 450 nm, 530 nm, and 633 nm, were examined using theAdoB12-LitR protein and PlitR. D and L denote dark and white light, respectively.

Takano et al.

2312 jb.asm.org July 2015 Volume 197 Number 14Journal of Bacteriology

on Decem

ber 24, 2020 by guesthttp://jb.asm

.org/D

ownloaded from

We examined the type of Cbl that confers LitR repressor activ-ity (Fig. 9C). AdoB12 was the only Cbl that decreased the PlitRtranscription level under dark conditions. We also determinedwhich light wavelengths affect LitR activity (Fig. 9D). Illuminationwith white light and light wavelengths of 365 nm, 450 nm, and 530nm induced transcription, whereas illumination with 633-nmlight and dark conditions did not affect the transcription level. Thelight wavelengths that inhibited AdoB12-LitR activity were consis-tent with the absorption spectrum of AdoB12. Collectively, theseresults demonstrate that the AdoB12-LitR complex serves as alight-sensitive repressor of PlitR and PcrtI1.

DISCUSSION

LitR/CarH homologs are found in phylogenetically diverse bacte-rial genera; however, the biochemical properties of Gram-positivebacteria in this family remain poorly understood. In this study, weshowed that the LitR protein derived from B. megaterium, aGram-positive bacterium, serves as a light-sensitive repressor pro-tein that stimulates the light-inducible transcription of crt genes.We successfully reproduced the light-inducible transcription ofcrt genes in vitro. The evidence obtained in the present and previ-ous studies suggests that the AdoB12-LitR-mediated regulatorymechanism is common to light-induced carotenoid production inboth Gram-positive and Gram-negative bacteria.

Figure 10 illustrates the molecular mechanism underlyinglight-inducible carotenoid production in B. megaterium QM

B1551, based on the genetic and biochemical data obtained in thisstudy. The AdoB12-LitR recombinant protein binds specifically tothe promoter regions of litR itself and the crt operon, and it re-presses the expression of these genes. The binding sites are locatedupstream of (positions �36 to �67) and near (positions �13 to�22) the transcriptional start sites of the promoters preceding litRand crt, respectively. Illumination caused the subunit dissociationof AdoB12-LitR and weakened the DNA-binding activity, which inturn allowed the �A-containing RNA polymerase holoenzyme tobind to the promoters to generate the transcripts for both genes.The LitR protein was also expressed under illumination (Fig. 5A)and displayed a weak DNA-binding activity (Fig. 7A), which maycontribute to preventing carotenoid overproduction. Conse-quently, the carotenoid production level by wild-type B. megate-rium under illumination was lower than that of the litR null mu-tant under the same conditions (Fig. 2). The �BMQ_1998 strainexhibited a moderate level of carotenoid production compared tothe wild-type strain under dark and illumined conditions (Fig. 2).The underlying reason for this result is unclear at present. Apo-LitR may have a regulatory activity that is different from that ofAdoB12-LitR.

The molecular mechanism of light-inducible transcription inB. megaterium QM B1551 appears to be simple compared to thatin M. xanthus and T. thermophilus. This is based on two pieces ofevidence. First, disruption of litR caused overproduction of caro-tenoids, indicating that LitR is the only or the major negative

FIG 10 Working model for light-inducible transcriptional control of the litR and crt genes by AdoB12-LitR. LitR is bound by AdoB12 and associates with thepromoter regions of litR and crtI1, repressing the transcription of both genes under dark conditions. Under light conditions, the absorption of light by thetetrameric AdoB12-LitR complex causes its dissociation into a dimeric form, probably due to the photolysis of AdoB12. The dissociation inactivates LitR andallows the RNA polymerase (RNAP) holoenzyme to initiate mRNA synthesis of litR and the crt operon, consisting of crtI1-crtI2-crtB.

LitR of B. megaterium

July 2015 Volume 197 Number 14 jb.asm.org 2313Journal of Bacteriology

on Decem

ber 24, 2020 by guesthttp://jb.asm

.org/D

ownloaded from

regulator of the photoresponse. Second, the light-induced tran-scription was demonstrated in vitro by adding only two proteins,AdoB12-LitR and RNAP, indicating that any other regulator is notrequired. In contrast, the mechanism known for M. xanthus ishighly complex. The two photodependent signaling pathways, i.e.,the singlet oxygen (1O2)-dependent and B12-dependent pathways,involve multiple regulators, including two MerR family regulators(CarA and CarH), a sigma factor, an anti-sigma factor, an antire-pressor, etc. (1, 34). Although T. thermophilus appears to haveonly the B12-dependent pathway, like B. megaterium, light-induc-ible transcription in this thermophile requires two transcriptionalregulators, namely, LitR and LdrP (4).

YtvA, originally found in B. subtilis, is an LOV domain-con-taining, blue light-responsive photoreceptor. This protein func-tions as a blue light-inducible molecular switch (35–37). By ab-sorbing blue light, YtvA activates �B, a stress response sigmafactor, hence initiating the phosphorylation cascade downstreamof the �B-dependent general stress response. We suppose that asimilar photodependent mechanism exists in B. megaterium, be-cause it retains a ytvA homolog (BMQ_3060). The induction ofthe �B-dependent general stress response by light in B. subtilisrequires additional salt stress (37). This means that the YtvA-me-diated response is not specific to light. In contrast, the LitR systemenables a quick response to illumination.

The in vitro transcription experiment clearly demonstratedthat AdoB12-LitR and �A-containing RNA polymerase activatetranscription in a light-dependent manner (Fig. 9). This assay alsodemonstrated that AdoB12 confers a repressor activity on LitR thatis sensitive to the specific light wavelength. In our study, we per-formed the same assay with the LitR protein from T. thermophilusunder various conditions, but we failed to reproduce the light-dependent transcription. Similarly, in vitro reconstruction has notyet been reported for M. xanthus, probably due to the difficulty inpreparation of the native recombinant protein. The data pre-sented in this study provide the first example of in vitro reproduc-tion of light-inducible transcription by a member of the LitR/CarH protein family.

The involvement of AdoB12 in the proper activity of LitR wasverified by a gel shift assay (Fig. 7A) and an in vitro runoff tran-scription assay (Fig. 9C). However, AdoB12-treated LitR from B.megaterium showed a spectrum similar to that of OHB12 underboth dark and light conditions (Fig. 5). In contrast, T. thermophi-lus LitR showed a spectrum similar to that of AdoB12 under darkconditions and one similar to that of OHB12 under light condi-tions (1, 11). At present, the reason for this inconsistency is un-clear. However, one possibility is that the adenosyl moiety ofAdoB12 is located in the internal region of the protein, which couldcause LitR to exhibit a spectrum similar to that of OHB12 underboth conditions.

Gel filtration chromatography and analytical ultracentrifuga-tion demonstrated that B. megaterium AdoB12-LitR undergoessubunit dissociation in response to illumination (Fig. 6). T. ther-mophilus AdoB12-LitR comprises a tetramer which undergoessubunit dissociation in response to illumination (1, 11). Thus, ourresult indicates that the biochemical properties of B. megateriumLitR differ slightly from those of T. thermophilus LitR in terms ofthe subunit structure. We expect that a detailed structural charac-terization of AdoB12-LitR will provide critical information regard-ing the light-induced conformational changes in LitR and theirsignificance in transcriptional control in B. megaterium.

ACKNOWLEDGMENTS

We thank Kei Asai, Saori Kosono, Akira Nakamura, Fumio Arisaka, Shoi-chi Amano, Yoshihiro Agari, and Akeo Shinkai for helpful discussions.

This study was supported by a Grant-in-Aid for Scientific Research(C) (grant 25450113) to H.T. and by the High-Tech Research CenterProject of the Ministry of Education, Culture, Sports, Science and Tech-nology, Japan, the Noda Institute for Scientific Research, the NAGASEScience and Technology Development Foundation, and the CharitableTrust of the Araki Medical and Biochemical Research Memorial Fund.

REFERENCES1. Ortiz-Guerrero JM, Polanco MC, Murillo FJ, Padmanabhan S, Elías-

Arnanz M. 2011. Light-dependent gene regulation by a coenzyme B12-based photoreceptor. Proc Natl Acad Sci U S A 108:7565–7570. http://dx.doi.org/10.1073/pnas.1018972108.

2. Takano H, Agari Y, Hagiwara K, Watanabe R, Yamazaki R, Beppu T,Shinkai A, Ueda K. 2014. LdrP, a cAMP receptor protein/FNR familytranscriptional regulator, serves as a positive regulator for the light-inducible gene cluster in a megaplasmid of Thermus thermophilus. Micro-biology 160:2650 –2660. http://dx.doi.org/10.1099/mic.0.082263-0.

3. Pérez-Marín MC, Padmanabhan S, Polanco MC, Murillo FJ, Elías-Arnanz M. 2008. Vitamin B12 partners the CarH repressor to downregu-late a photoinducible promoter in Myxococcus xanthus. Mol Microbiol67:804 – 819. http://dx.doi.org/10.1111/j.1365-2958.2007.06086.x.

4. Takano H, Kondo M, Usui N, Usui T, Ohzeki H, Yamazaki R, Wash-ioka M, Nakamura A, Hoshino T, Hakamata W, Beppu T, Ueda K.2011. Involvement of CarA/LitR and CRP/FNR family transcriptional reg-ulators in light-induced carotenoid production in Thermus thermophilus.J Bacteriol 193:2451–2459. http://dx.doi.org/10.1128/JB.01125-10.

5. Cervantes M, Murillo FJ. 2002. Role for vitamin B(12) in light inductionof gene expression in the bacterium Myxococcus xanthus. J Bacteriol 184:2215–2224. http://dx.doi.org/10.1128/JB.184.8.2215-2224.2002.

6. Edge R, McGarvey DJ, Truscott TG. 1997. The carotenoids as anti-oxidants—a review. J Photochem Photobiol B 41:189 –200. http://dx.doi.org/10.1016/S1011-1344(97)00092-4.

7. Armstrong GA. 1997. Genetics of eubacterial carotenoid biosynthesis: acolorful tale. Annu Rev Microbiol 51:629 – 659. http://dx.doi.org/10.1146/annurev.micro.51.1.629.

8. Takano H, Obitsu S, Beppu T, Ueda K. 2005. Light-induced caroteno-genesis in Streptomyces coelicolor A3(2): identification of an extracytoplas-mic function sigma factor that directs photodependent transcription ofthe carotenoid biosynthesis gene cluster. J Bacteriol 187:1825–1832. http://dx.doi.org/10.1128/JB.187.5.1825-1832.2005.

9. Takano H, Hagiwara K, Ueda K. 2015. Fundamental role of cobalaminbiosynthesis in the developmental growth of Streptomyces coelicolor A3(2).Appl Microbiol Biotechnol 99:2329 –2337. http://dx.doi.org/10.1007/s00253-014-6325-z.

10. Agari Y, Kuramitsu S, Shinkai A. 2012. X-ray crystal structure ofTTHB099, a CRP/FNR superfamily transcriptional regulator from Ther-mus thermophilus HB8, reveals a DNA-binding protein with no requiredallosteric effector molecule. Proteins 80:1490 –1494. http://dx.doi.org/10.1002/prot.24049.

11. Díez AI, Ortiz-Guerrero JM, Ortega A, Elías-Arnanz M, PadmanabhanS, García de la Torre J. 2013. Analytical ultracentrifugation studies ofoligomerization and DNA-binding of TtCarH, a Thermus thermophiluscoenzyme B12-based photosensory regulator. Eur Biophys J 42:463– 476.http://dx.doi.org/10.1007/s00249-013-0897-x.

12. Takano H, Beppu T, Ueda K. 2006. The CarA/LitR-family transcrip-tional regulator: its possible role as a photosensor and wide distribution innon-phototrophic bacteria. Biosci Biotechnol Biochem 70:2320 –2324.http://dx.doi.org/10.1271/bbb.60230.

13. Takano H, Asker D, Beppu T, Ueda K. 2006. Genetic control forlight-induced carotenoid production in non-phototrophic bacteria. JInd Microbiol Biotechnol 33:88 –93. http://dx.doi.org/10.1007/s10295-005-0005-z.

14. Wittchen KD, Meinhardt F. 1995. Inactivation of the major extracel-lular protease from Bacillus megaterium DSM319 by gene replacement.Appl Microbiol Biotechnol 42:871– 877. http://dx.doi.org/10.1007/BF00191184.

15. Maniatis T, Fritsch EF, Sambrook J. 1982. Molecular cloning: a

Takano et al.

2314 jb.asm.org July 2015 Volume 197 Number 14Journal of Bacteriology

on Decem

ber 24, 2020 by guesthttp://jb.asm

.org/D

ownloaded from

laboratory manual. Cold Spring Harbor Laboratory Press, Cold SpringHarbor, NY.

16. Shiratori H, Ikeno H, Ayame S, Kataoka N, Miya A, Hosono K, BeppuT, Ueda K. 2006. Isolation and characterization of a new Clostridium sp.that performs effective cellulosic waste digestion in a thermophilic metha-nogenic bioreactor. Appl Environ Microbiol 72:3702–3709. http://dx.doi.org/10.1128/AEM.72.5.3702-3709.2006.

17. Cole JR, Chai B, Farris RJ, Wang Q, Kulam SA, McGarrell DM, GarrityGM, Tiedje JM. 2005. The Ribosomal Database Project (RDP-II): se-quences and tools for high-throughput rRNA analysis. Nucleic Acids Res33:D294 –D296. http://dx.doi.org/10.1093/nar/gki038.

18. Brown BJ, Carlton BC. 1980. Plasmid-mediated transformation in Ba-cillus megaterium. J Bacteriol 142:508 –512.

19. Guérout-Fleury AM, Frandsen N, Stragier P. 1996. Plasmids for ectopicintegration in Bacillus subtilis. Gene 180:57– 61. http://dx.doi.org/10.1016/S0378-1119(96)00404-0.

20. Mendoza-Vargas A, Olvera L, Olvera M, Grande R, Vega-Alvarado L,Taboada B, Jimenez-Jacinto V, Salgado H, Juárez K, Contreras-MoreiraB, Huerta AM, Collado-Vides J, Morett E. 2009. Genome-wide identi-fication of transcription start sites, promoters and transcription factorbinding sites in E. coli. PLoS One 4:e7526. http://dx.doi.org/10.1371/journal.pone.0007526.

21. Nishiyama M, Kobashi N, Tanaka K, Takahashi H, Tanokura M. 1999.Cloning and characterization in Escherichia coli of the gene encoding theprincipal sigma factor of an extreme thermophile, Thermus thermophilus.FEMS Microbiol Lett 172:179 –186. http://dx.doi.org/10.1111/j.1574-6968.1999.tb13467.x.

22. Shinkai A, Kira S, Nakagawa N, Kashihara A, Kuramitsu S, YokoyamaS. 2007. Transcription activation mediated by a cyclic AMP receptor pro-tein from Thermus thermophilus HB8. J Bacteriol 189:3891–3901. http://dx.doi.org/10.1128/JB.01739-06.

23. Pane L, Radin L, Franconi G, Carli A. 1996. The carotenoid pigments ofa marine Bacillus firmus strain. Boll Soc Ital Biol Sper 72:303–308.

24. Mitchell C, Iyer S, Skomurski JF, Vary JC. 1986. Red pigment in Bacillusmegaterium spores. Appl Environ Microbiol 52:64 – 67.

25. Steiger S, Perez-Fons L, Fraser PD, Sandmann G. 2012. Biosynthesis ofa novel C30 carotenoid in Bacillus firmus isolates. J Appl Microbiol 113:888 – 895. http://dx.doi.org/10.1111/j.1365-2672.2012.05377.x.

26. Khaneja R, Perez-Fons L, Fakhry S, Baccigalupi L, Steiger S, To E,Sandmann G, Dong TC, Ricca E, Fraser PD, Cutting SM. 2010. Caro-tenoids found in Bacillus. J Appl Microbiol 108:1889 –1902. http://dx.doi.org/10.1111/j.1365-2672.2009.04590.x.

27. Duc LH, Fraser PD, Tam NK, Cutting SM. 2006. Carotenoids present inhalotolerant Bacillus spore formers. FEMS Microbiol Lett 255:215–224.http://dx.doi.org/10.1111/j.1574-6968.2005.00091.x.

28. Eppinger M, Bunk B, Johns MA, Edirisinghe JN, Kutumbaka KK,Koenig SS, Creasy HH, Rosovitz MJ, Riley DR, Daugherty S, Martin M,Elbourne LD, Paulsen I, Biedendieck R, Braun C, Grayburn S, DhingraS, Lukyanchuk V, Ball B, Ul-Qamar R, Seibel J, Bremer E, Jahn D, RavelJ, Vary PS. 2011. Genome sequences of the biotechnologically importantBacillus megaterium strains QM B1551 and DSM319. J Bacteriol 193:4199 – 4213. http://dx.doi.org/10.1128/JB.00449-11.

29. Moore SJ, Warren MJ. 2012. The anaerobic biosynthesis of vitamin B12.Biochem Soc Trans 40:581–586. http://dx.doi.org/10.1042/BST20120066.

30. Biedendieck R, Malten M, Barg H, Bunk B, Martens JH, Deery E, LeechH, Warren MJ, Jahn D. 2010. Metabolic engineering of cobalamin (vi-tamin B12) production in Bacillus megaterium. Microb Biotechnol 3:24 –37. http://dx.doi.org/10.1111/j.1751-7915.2009.00125.x.

31. Raux E, Lanois A, Warren MJ, Rambach A, Thermes C. 1998. Cobala-min (vitamin B12) biosynthesis: identification and characterization of aBacillus megaterium cobI operon. Biochem J 335:159 –166.

32. Haldenwang WG. 1995. The sigma factors of Bacillus subtilis. MicrobiolRev 59:1–30.

33. Forouhar F, Kuzin A, Seetharaman J, Lee I, Zhou W, Abashidze M,Chen Y, Yong W, Janjua H, Fang Y, Wang D, Cunningham K, Xiao R,Acton TB, Pichersky E, Klessig DF, Porter CW, Montelione GT, TongL. 2007. Functional insights from structural genomics. J Struct FunctGenomics 8:37– 44. http://dx.doi.org/10.1007/s10969-007-9018-3.

34. Galbis-Martínez M, Padmanabhan S, Murillo FJ, Elías-Arnanz M. 2012.CarF mediates signaling by singlet oxygen, generated via photoexcitedprotoporphyrin IX, in Myxococcus xanthus light-induced carotenogenesis.J Bacteriol 194:1427–1436. http://dx.doi.org/10.1128/JB.06662-11.

35. Gaidenko TA, Kim TJ, Weigel AL, Brody MS, Price CW. 2006. Theblue-light receptor YtvA acts in the environmental stress signaling path-way of Bacillus subtilis. J Bacteriol 188:6387– 6395. http://dx.doi.org/10.1128/JB.00691-06.

36. Avila-Pérez M, Hellingwerf KJ, Kort R. 2006. Blue light activates thesigmaB-dependent stress response of Bacillus subtilis via YtvA. J Bacteriol188:6411– 6414. http://dx.doi.org/10.1128/JB.00716-06.

37. Suzuki N, Takaya N, Hoshino T, Nakamura A. 2007. Enhancement of asigma(B)-dependent stress response in Bacillus subtilis by light via YtvAphotoreceptor. J Gen Appl Microbiol 53:81– 88. http://dx.doi.org/10.2323/jgam.53.81.

LitR of B. megaterium

July 2015 Volume 197 Number 14 jb.asm.org 2315Journal of Bacteriology

on Decem

ber 24, 2020 by guesthttp://jb.asm

.org/D

ownloaded from