JOURNAL OF BACTERIOLOGY,0021-9193/01/$04.0010 DOI: 10.1128/JB.183.5.1505–1510.2001

Mar. 2001, p. 1505–1510 Vol. 183, No. 5

Copyright © 2001, American Society for Microbiology. All Rights Reserved.

A Eukaryotic-Type Protein Kinase, SpkA, Is Required forNormal Motility of the Unicellular Cyanobacterium

Synechocystis sp. Strain PCC 6803AYAKO KAMEI, TAKASHI YUASA, KUMI ORIKAWA, XIAO XING GENG, AND MASAHIKO IKEUCHI*

Department of Life Sciences (Biology), The University of Tokyo, Meguro, Tokyo 153-8902, Japan

Received 9 August 2000/Accepted 30 November 2000

The genome of the unicellular cyanobacterium Synechocystis sp. strain PCC 6803 comprises many openreading frames (ORFs) which putatively encode eukaryotic-type protein kinase and protein phosphatase.Based on gene disruption analysis, a region of the hypothetical ORF sll1575, which retained a part of theprotein kinase motif, was found to be required for normal motility in the original isolate of strain PCC 6803.Sequence determination revealed that in this strain sll1575 was part of a gene (designated spkA) whichharbored an entire eukaryotic-type Ser/Thr protein kinase motif. Strain ATCC 27184 and a glucose-tolerantstrain derived from the same isolate as the PCC strain had a frameshift mutation dividing spkA into ORFssll1574 and sll1575. The structural integrity of spkA agreed well with the motility phenotype, determined bycolony morphology on agar plates. The spkA gene was expressed in Escherichia coli as a His-tagged protein,which was purified by Ni21 affinity chromatography. With [g-32P]ATP, SpkA was autophosphorylated andtransferred the phosphate group to casein, myelin basic protein, and histone. SpkA also phosphorylatedseveral proteins in the membrane fraction of Synechocystis cells. These results suggest that SpkA is a eukary-otic-type Ser/Thr protein kinase and regulates cellular motility via phosphorylation of the membrane proteinsin Synechocystis.

Protein phosphorylation-dephosphorylation is a mechanismwidely used to regulate proteins. In prokaryotes, phospho-transfer of the protein His kinase to the Asp residue in theresponse regulator is predominant in various signal transduc-tion pathways (15). However, determination of the completegenome of the unicellular cyanobacterium Synechocystis sp.strain PCC 6803 (10) revealed a number of open readingframes (ORFs) that are homologous to the eukaryotic-typeprotein kinase and protein phosphatase. Recent progress ingenome analysis has further shown that bacteria and archaeauniversally have several types of Ser/Thr protein kinase andprotein phosphatase, which were originally believed to be spe-cific to eukaryotes (3, 11, 13, 21). These findings strongly sug-gested that prokaryotes have signal transduction systems inaddition to the well-known sensor His kinase and responseregulator systems.

Bacteria lacking flagella such as cyanobacteria can move bygliding motility. Photoresponsive gliding motility is unique tocyanobacteria and has been studied mainly in filamentous or-ganisms for many decades (6, 7). Despite extensive studies,very little has been established with respect to the regulatorymechanism of motility in cyanobacteria. On the other hand,twitching or swimming motility in unicellular cyanobacteria,though not as conspicuous as in filamentous cyanobacteria, hasbeen described (5, 14, 17). The motility of Synechocystis strainsp. PCC 6803, though described as sporadic and very slow (14),seems to be a feasible target for molecular analysis, since the

complete genome has been determined (10). It was recentlysuggested that an alternative sigma factor, SigF, and putativepilin subunit gene, sll1694, are essential for the motility of thiscyanobacterium (1).

In an earlier study, we showed that a putative Ser/Thr pro-tein phosphatase gene, slr2031, plays a crucial role in motilityof Synechocystis cells (9). To extend these findings, we evalu-ated the counteracting protein kinase as a regulator of motilityby means of targeted disruption of genes with a Ser/Thr pro-tein kinase motif. Here we reported that the protein kinaseSpkA is required for the normal motility of Synechocystis cells.The spkA gene was not listed in the original annotation of theSynechocystis genome (10) because of a frameshift mutation inthe sequenced strain.

MATERIALS AND METHODS

Strains and culture conditions. Strains PCC 6803 and ATCC 27184 wereobtained as the unicellular cyanobacterium Synechocystis sp. strain PCC 6803from the Pasteur Culture Collection and American Type Culture Collection,respectively; both were independently deposited by R. Kunisawa as isolates fromthe same strain (Berkeley strain number 6803) (12, 14). The glucose-tolerantstrain, isolated by Williams (18), was a kind gift from W. Vermaas (Arizona StateUniversity). Standard strains and mutants were grown in BG11 medium bufferedwith N-tris(hydroxymethyl)methyl-2-aminoethanesulfonic acid-KOH (pH 7.8) at31°C at a light intensity of 20 mE m22 s21. Solid medium was supplemented with0.8% (wt/vol) agar and 0.3% (wt/vol) sodium thiosulfate and used for examina-tion of motility by colony morphology. Kanamycin (20 mg/ml) was added tomaintain gene-disrupted mutants; antibiotics were not included for character-ization of the mutant phenotype. For cloning and subcloning of plasmids inEscherichia coli, strains XL10 and JM109 were used; BL21 (DE3)pLysS was usedfor expression with pET28a.

Construction of spkA disruption mutant. Disruption of spkA was achieved asdisruption of sll1575. A part of sll1575 was amplified by PCR using primer 1(59-GGGTCAAGTCTACCGAGC-39), primer 2 (59-ATCCGACTAGGCATGGGC-39), and Taq polymerase (Ampli-Taq; PE Applied Biosystems, Foster City,Calif.) and cloned into pT7Blue-T vector (Novagen, Madison, Wis.). sll1575 was

* Corresponding author. Mailing address: Department of Life Sci-ences (Biology), The University of Tokyo, Komaba 3-8-1, Meguro,Tokyo 153-8902, Japan. Phone: 81-3-5454-6641. Fax: 81-3-5454-4337.E-mail: mikeuchi@ims.u-tokyo.ac.jp.

1505

on June 3, 2018 by guesthttp://jb.asm

.org/D

ownloaded from

interrupted at the MscI site by insertion of the Tn5-derived kanamycin resistancecassette in the same direction as sll1575. Although the cassette allows expressionof downstream genes due to lack of transcription termination, the map (Fig. 1A)suggests that insertion of the cassette does not affect expression of the flankingORFs. Mutants were generated by transformation of Synechocystis cells with thisDNA and selected on BG11 plates containing kanamycin (20 mg/ml). Completesegregation was confirmed by PCR with the primers described above (notshown).

DNA sequence analysis. The full-length DNA of spkA of the PCC and glucose-tolerant strains and a region around the border of sll1574 and sll1575 of theATCC strain were determined by the BigDye terminator fluorescence detectionmethod (PE Applied Biosystems), using a capillary sequencer (ABI PRISM 310Genetic Analyzer; PE Applied Biosystems).

Cloning of spkA. The coding region of spkA was amplified from the genomicDNA of the motile PCC strain by PCR with primer 3 (59-GATGCTAGCGCTATGACCCCTG-39) and primer 4 (59-ATGAGCTCACAATCCTGAAACCT-39), containing NheI and SacI sites, respectively. PCR was performed with PfuDNA polymerase (Stratagene, La Jolla, Calif.) according to the manufacturer’sinstructions. Following initial denaturation at 95°C for 1 min, each sample wassubjected to 35 cycles consisting of denaturation at 95°C for 30 s, annealing at57°C for 1.5 min, and elongation at 72°C for 4 min. The PCR product was clonedinto pPCR-Script (Stratagene) according to the manufacturer’s instructions andpropagated in E. coli XL10 (Epicurian Coli XL10-Gold Kan; Stratagene). Thecloned spkA was sequenced, excised with NheI and SacI (New England Biolabs,Beverly, Mass.) and then inserted into pET28a (Novagen) as a fusion with theN-terminal His tag.

Isolation of total RNA and Northern blotting. Total RNA was isolated byusing an RNeasy Midi kit (QiaGEN, Hilden, Germany). The standard protocolfor breakage of cells was modified as follows. Synechocystis cells collected froma 100-ml culture (A730 5 0.6 to 1.0) were disrupted with a Mini-Bead Beater(Biospec, Bartlesville, Okla.) and zircon beads (100 mm in diameter; Biospec) forthree pulses of 50 s at 4°C in 0.9 ml of buffer provided in the kit. After removalof the beads by brief centrifugation, the volume of cell lysate and ethanolconcentration were adjusted and subjected to a spin column chromatographyaccording to the manufacturer’s instructions. Total RNA (10 mg) thus isolatedwas fractionated in a 1.2% denaturing agarose gel and blotted onto a Hy-bond-N1 membrane (Amersham Pharmacia, Uppsala, Sweden). As a probe,pilA1 (sll1694) was amplified with primer 5 (59-CACATATGGCTAGTAATTTTAAATTC-39) and primer 6 (59-GGCACGTGTTTAATTACTTCAGCACC-39). Labeling of the probe and detection were done by using ECL (enhancedchemiluminecence) direct nucleic acid labeling and detection systems (Amer-sham Pharmacia) according to the manufacturer’s instructions.

Expression and purification of SpkA. pET28a carrying spkA was introducedinto E. coli BL21(DE3)pLysS. Cells were grown at 37°C in 250 ml of Luria brothmedium containing kanamycin (20 mg/ml) and chloramphenicol (37 mg/ml) to anA600 of about 0.5. Then isopropyl-b-D-thiogalactoside (IPTG) was added to afinal concentration of 0.5 mM, and the cultures were incubated for 2 h at 25°C.The cells were harvested by centrifugation, washed with 50 mM Tris-HCl (pH7.5) containing 100 mM NaCl and 1 mM phenylmethylsulfonyl fluoride, andresuspended in 25 ml of the same medium plus 10% (wt/vol) glycerol. The cellsuspension was once frozen at 285°C for 15 min, thawed on ice, and thensonicated at 4°C for 9 min (three cycles of 3-min bursts with a cooling period) ina sonicator (model 200M; Kubota Co., Tokyo, Japan). The cell extract wascentrifuged at 16,000 3 g for 30 min, and the supernatant was subjected to Ni21

affinity chromatography.A Hi-Trap chelating column (Amersham Pharmacia) charged with Ni21 was

equilibrated with 50 mM Tris-HCl (pH 7.5) containing 100 mM NaCl, 10%(wt/vol) glycerol, and 5 mM imidazole (buffer A). The column was loaded withthe cell extract and washed with buffer A; then His-tagged SpkA protein waseluted using linear gradient from 5 to 500 mM imidazole. Proteins were sepa-rated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) followed by staining with Coomassie brilliant blue R-250. Alternatively,proteins resolved in the SDS-gel were blotted to a polyvinylidene difluoridemembrane (Immobilon; Millipore, Bedford, Mass.) and the His tag portion wasvisualized with His-Probe (Pierce, Rockford, Ill.) as instructed by the manufac-turer.

Assay of protein kinase activity. Autophosphorylation of SpkA and phosphor-ylation of myelin basic protein (MBP), histone, or casein were assayed in vitrowith [g-32P]ATP. About 0.5 mg of the purified SpkA protein was added to 10 mlof phosphorylation buffer containing 20 mM 2-morpholinoethanesulfonic acid(pH 6.5), 10 mM MgCl2, and 0.1 mM [g-32P]ATP (3,000 Ci mol21) with orwithout 2.5 mg of bovine MBP (Sigma, St. Louis, Mo), 2.5 mg of histone (typeIIIS, calf thymus; Sigma), or 6.25 mg of bovine casein (partially dephosphory-

lated; Sigma) and incubated for 15 min at 30°C. Control phosphorylation exper-iments were done with a crude extract from E. coli before induction. SDS (finalconcentration, 1%) and dithiothreitol (final concentration, 60 mM) were addedto stop the reaction. After boiling for 5 min, proteins were resolved by SDS-PAGE. The gels were stained with Coomassie brilliant blue R-250, dried, andthen subjected to autoradiography with X-ray film (X-Omat Blue XB-1; EastmanKodak, Rochester, N.Y.).

In vitro phosphorylation of cell extracts. Synechocystis cells were harvestedfrom a 50-ml culture (A730 5 0.5 to 0.8) by centrifugation at 8,000 3 g for 5 min,washed with 20 mM Tris-HCl (pH 7.5) containing 100 mM NaCl, and resus-pended in 0.9 ml of the same buffer. The cells were disrupted with the zirconbeads in a Mini-Bead Beater for three pulses of 50 s at 4°C. After removal of thebeads by brief centrifugation, cell extracts were fractionated into soluble andmembrane fractions by centrifugation at 541,000 3 g for 30 min at 4°C. Themembranes were resuspended in the original volume of the buffer. Soluble andmembrane fractions, both derived from 1.25 mg of chlorophyll, were incubatedwith [g-32P]ATP in 20 ml of the phosphorylation buffer described above. Whenstated, 1 mg of the purified SpkA protein was included.

FIG. 1. Characterization of the sll1575-disrupted mutant. (A) Genemap showing the relative positions of sll1573, sll1574, and sll1575 andinsertion of the kanamycin resistance (KmR) cassette. (B) Colonymorphology of wild-type (WT) and mutant cells grown under lateralillumination at 20 mE m22 s21 (arrow). Cells were grown as singlecolonies on 0.8% agar–BG11 medium for 5 days at 31°C. (C) Expres-sion of pilA1 (sll1694) as revealed by Northern hybridization.

1506 KAMEI ET AL. J. BACTERIOL.

on June 3, 2018 by guesthttp://jb.asm

.org/D

ownloaded from

RESULTS

Novel gene is required for motility. The Synechocystis ge-nome contains genes encoding seven putative Ser/Thr proteinkinases which show similarity to typical eukaryotic-type proteinkinases (10). Among them, sll1575 is unique; it encodes onlythe C-terminal part of the conserved protein kinase motif,while the upstream gene sll1574, originally annotated as ahypothetical gene, encodes the remainder. To determinewhether sll1575 is functional, we constructed a gene disruptionmutant (Fig. 1A) from the motile wild type, which was ob-tained as PCC strain 6803 (denoted the PCC strain). Aftercomplete segregation, it was found that the sll1575-disruptedmutant formed domed, round-shaped colonies on agar plates,while the parent strain formed flat sheet-like colonies, indica-tive of loss of motility in the mutant. Figure 1B shows typicalnonmotile colonies of the spkA mutant in contrast with activemovement of the wild type toward the light source (positivephototaxis). This also indicates that the sequence of sll1575 isfunctional in the PCC strain.

To confirm that the split gene comprising sll1574 and sll1575

encodes a functional protein kinase, we cloned it from the PCCstrain and determined the nucleotide sequence. Unexpectedly,the gene was no longer split but consisted of a single ORF dueto frameshifting as a result of lack of a single A in the lastcodon for Asn in sll1574 (Fig. 2). The new ORF encodes aprotein of 521 amino acid residues. We designated this novelORF spkA (for Synechocystis protein kinase). The deducedprotein of spkA has the entire motif (subdomains I to XI)common to a number of eukaryotic-type Ser/Thr protein ki-nases (8). We also determined the nucleotide sequence of spkAin derivatives of this Synechocystis strain and found that ATCC27184 and the widely used glucose-tolerant strain (12, 18)carried the same frameshift mutation as the sequenced strain(not shown). In accordance with this finding, these substrainswere nonmotile on agar plates.

Recently it was shown that motility of Synechocystis sp. strainPCC 6803 requires a type IV-like pilus structure, which issupported by a number of subunits and biogenesis factors (1, 2,19). To determine whether the biogenesis of pili was regulatedby the protein kinase SpkA, we examined the mRNA level of

FIG. 2. Gene and protein sequences of spkA. (A) Sequence variation in spkA. Part of the nucleotide and deduced amino acid sequences of spkAin the PCC, ATCC, and GT strains. The putative initiation codon (GTG) of sll1575 is underlined. One base pair insertion together with theframeshifted codons in the ATCC and glucose-tolerant (GT) strains are shown in reverse type. (B) Amino acid sequence alignment of SynechocystisSpkA (S.6803 spkA) with ORF520 from Anabaena sp. strain PCC 7120 (A.7120 ORF520), ORF517 from Nostoc punctiforme ATCC 29133 (N.p.ORF517), and Pkn2 from Myxococcus xanthus (M.x. pkn2). Residues conserved with those in SpkA are shown in reverse type. Subdomains I toXI typical to eukaryotic-type protein kinases are shown above the alignment, and highly conserved residues are indicated with asterisks (see textfor details). Note that only the N-terminal kinase domain of Pkn2 is presented.

VOL. 183, 2001 EUKARYOTIC-TYPE PROTEIN KINASE SpkA IN SYNECHOCYSTIS 1507

on June 3, 2018 by guesthttp://jb.asm

.org/D

ownloaded from

pilA1 (sll1694), which encodes a major pilin subunit of the pili.Northern blot analysis revealed no significant difference inpilA1 mRNA levels of the wild-type and the spkA mutantstrains (Fig. 1C). We observed cell surface architecture byelectron microscopy and found that both the wild type and thespkA mutant had the two types of pili, thick and thin (S.Yoshihara, A. Kamei, and M. Ikeuchi, unpublished results),the latter of which is known to be essential for motility (2, 19).These observations suggest that spkA is not an essential factorbut regulates motility via an unidentified signal transductionpathway.

Protein kinase activity of SpkA. We tried to express thefunctional spkA gene with an N-terminal His tag in E. coliunder control of the T7 promoter. However, the dye-stainedprofile was not changed after induction, even though expres-sion was induced at 25°C (Fig. 3). We could detect the His-tagged SpkA protein only by Western blotting with His-Probe.Not only the major 63-kDa band but also many bands in thelow-molecular-mass region were visualized with His-Probe, in-dicative of rapid degradation in cells. Under the conditionsused, very little SpkA was recovered in the inclusion body (notshown). The soluble fraction was subjected to Ni21 affinitycolumn chromatography. Although a large part of the 63-kDaprotein and many degraded proteins were not adsorbed, asmall part of the His-tagged SpkA was purified to homogeneity(Fig. 3, lane 3). This may suggest that improperly folded SpkAprotein is not stable and is rapidly degraded in the cytoplasmof E. coli.

We examined protein kinase activity of SpkA with[g-32P]ATP (Fig. 4). We could detect weak but significantautophosphorylation activity of SpkA and strong phosphoryla-tion of histone, MBP, and casein, which are general substratesof typical Ser/Thr protein kinases. Among these, MBP wasphosphorylated to the greatest extent. A very similar phos-phorylation pattern was obtained with the crude soluble extractbefore chromatography (not shown). As a negative control, wedetected no phosphorylation in the crude extract of E. colibefore induction (Fig. 4, lane 1). Thus, we conclude that SpkA

of Synechocystis is a Ser/Thr protein kinase which belongs tothe large family of protein kinases in eukaryotes.

In vitro phosphorylation of cyanobacterial proteins. To de-termine the intrinsic substrate of SpkA, we performed in vitrophosphorylation in crude extracts from wild-type and spkAmutant cells in the presence or absence of His-tagged SpkA(Fig. 5). We detected SpkA-dependent phosphorylation in a90-kDa band and a doublet band around 30 kDa in the mem-

FIG. 3. Expression and purification of SpkA. Proteins resolved on12% polyacrylamide gels were visualized by staining with Coomassiebrilliant blue (A) and Western blotting with His-Probe (B). Lane 1, cellextract of E. coli after induction with IPTG; lanes 2 and 3, flowthroughfraction and His-tagged SpkA-enriched fraction in Ni21 affinity chro-matography, respectively. Positions of molecular size markers areshown in kilodaltons at the left (phosphorylase b, 94 kDa; bovineserum albumin, 67 kDab; ovalbumin, 43 kDa; carbonic anhydrase, 30kDa; trypsin inhibitor, 20.1 kDa). The arrow head shows the 63-kDaSpkA band.

FIG. 4. Detection of protein kinase activity. Phosphorylated pro-teins were resolved on 15% polyacrylamide gels and visualized bystaining with Coomassie brilliant blue (A) and autoradiography (B).Lane 1, cell extract of E. coli before induction with IPTG; lanes 2 to 5,Ni21 affinity-purified SpkA protein without (lane 2) or with histone,MBP, and casein, as indicated. Positions of molecular size markers areshown in Kilodaltons at the left; the arrow head shows autophosphor-ylation of the 63-kDa SpkA band.

FIG. 5. In vitro phosphorylation of Synechocystis proteins withSpkA. Proteins of soluble (S) and membrane fractions (M) from Syn-echocystis strain PCC and DspkA cells were incubated with (1) orwithout (2) purified SpkA and resolved on an SDS–15% polyacryl-amide gel. (A) Dye-stained gel; (B) autoradiogram of the same gel.Lane 1, Ni21 affinity-purified SpkA without cell extracts. Purified SpkAprotein was extraneously added in lanes 3, 5, 7, and 9. Positions ofmolecular size markers are shown in Kilodaltons at the left; positionsof the major phosphorylated polypeptides are indicated with the ar-rowhead and asterisk on the right.

1508 KAMEI ET AL. J. BACTERIOL.

on June 3, 2018 by guesthttp://jb.asm

.org/D

ownloaded from

brane fractions in addition to autophosphorylation of the ex-ogenous SpkA (Fig. 5, lanes 7 and 9). These bands were notdetected in the soluble fractions or did not correspond to thedye-stained bands in the membrane fractions. These findingssuggest that minor membrane-associated proteins of about 90and 30 kDa are the intrinsic substrates of SpkA. On the otherhand, there were heavily labeled bands around 17.5 kDa and inthe low-molecular-mass region of the soluble fractions even inthe absence of SpkA (Fig. 5, lane 4). The 17.5-kDa band waslocated just below the major stained band of phycocyanin. Theprominent label in the low-molecular-mass region was not dueto the unreacted [g-32P]ATP. Curiously, labeling of the latterwas very weak in the wild type. Disruption of spkA may havechanged the expression of other protein kinases. Anyway phos-phorylation of these bands was independent of SpkA, sugges-tive of an intrinsic protein kinase which is yet to be identified.

DISCUSSION

In this study, we demonstrated experimentally that a novelgene, spkA, encodes a Ser/Thr-type protein kinase resemblingtypical eukaryotic enzymes. Clearly, SpkA phosphorylated thegeneral substrate proteins, intrinsic membrane proteins, anditself. Before our work, it was assumed that a protein kinasewas encoded by a split gene comprising sll1574 and sll1575 (10,11, 21). We created sll1575-disrupted mutants from both thenonmotile glucose-tolerant strain and the motile PCC strain.As a result, the mutant from the PCC strain lost motility, whilethat from the glucose-tolerant strain showed no defect. Thisled us to further confirm the gene sequence in both PCC andglucose-tolerant strains. In fact, the gene in the PCC strain wasuninterrupted and expressed an active product in E. coli. Onthe other hand, the gene in the glucose-tolerant strain was splitas in the sequenced Kazusa strain. The glucose-tolerant strainhas been widely used as standard in many laboratories becauseof its application to genetic engineering of photosynthetic ap-paratus (18). Thus, especially in gene analysis we must use awild type as much as possible, although it is rather difficult toensure this beforehand. Spontaneous inactivation and its dom-ination in a culture stock such as ATCC may reflect the phys-iological significance of motility-related phenomena of thiscyanobacterium in photosynthetic propagation.

Homology search of the protein database with the deducedSpkA protein revealed that it is a member of Pkn2 family inbacteria, which belongs to the eukaryotic Ser/Thr protein ki-nase superfamily (13). Although there was no obvious homologof SpkA in the database, we could identify homologs in on-going genome projects, namely, ORF520 in Anabaena sp.strain PCC 7120 (http://www.kazusa.or.jp/cyano/anabaena/)and ORF517 in Nostoc punctiforme ATCC 29133 (http://www.jgi.doe.gov/JGI_microbial/html/nostoc_homepage.html). On theother hand, we could not detect the homolog in the genome ofthe marine cyanobacterium Prochlorococcus marinus MED4(http://www.jgi.doe.gov/JGI_microbial/html/prochlorococcus_homepage.html). Since the P. marinus genome contain nogenes involved in motility such as the pilM cluster (19), theymay not retain the ability to regulate the motility. Sequencealignment of the N-terminal halves of the three SpkA ho-mologs and the typical Pkn2 from Myxococcus xanthus (16)revealed the following common features (Fig. 2):

GXGXXGXV motif in subdomain I for ATP binding; Lysresidue in subdomain II, necessary for phosphotransfer (4);Glu residue in subdomain III; DXKPXN motif in subdomainVI as a Ser/Thr-specific feature; triplet DFG in subdomainVII; Asp residue in subdomain IX; and Arg residue in subdo-main XI (8). Recent genome analysis revealed that the Pkn2family has many components in cyanobacteria (7 genes in Syn-echocystis sp. strain PCC 6803) and mycobacteria (11 genes inMycobacterium tuberculosis) but not many in other bacteria(11). We expressed in E. coli the other six proteins of the Pkn2family in Synechocystis and detected phosphorylation activity inmost of them (A. Kamei, and M. Ikeuchi, unpublished data).Thus, it is now clear that SpkA as well as other proteins haveSer/Thr protein kinase activities in Synechocystis.

The C-terminal half of SpkA was also conserved inAnabaena ORF520 and Nostoc ORF517 (Fig. 2). Notably,there is a variable region between the N-terminal kinase motifand the C-terminal conserved domain. This region may simplyconnect the two domains. Homology search of the databasewith the C-terminal conserved domain revealed no homologyto known proteins or any motif. At the moment, we assumethat the C-terminal part of SpkA is important for determina-tion of the substrate specificity or regulation of the kinaseactivity. In vitro experiments suggested that SpkA regulatesmotility via phosphorylation of 90- and 30-kDa proteins in themembrane fraction in situ, although their identities are notknown (Fig. 5).

The nonmotile phenotype of the spkA-disrupted mutantstrongly suggests that protein phosphorylation regulates motil-ity by a molecular mechanism that remains to be clarified. Wealso recently identified many genes which are essential for themotility of Synechocystis, such as the pilM cluster (19). In thepilM disruptant, motility and transformation competency wereabolished, along with loss of the thick pili on the cell surface.By contrast, the spkA mutant was nearly nonmotile on agarplates (Fig. 1B), although it retained both thick and thin pili.The mutant also expressed mRNA of the major pilin genepilA1 (sll1694) at a level comparable to the wild-type level (Fig.1C). On the other hand, the mutant showed slight motility onsoft agar, as judged by occasional formation of a small fringe ofa single-cell layer surrounding the domed colonies (not shownin Fig. 1B). Movement of the mutant cells on soft agar was veryweak, and it was difficult to determine whether phototacticproperties were also affected. Anyway, these findings suggestthat spkA is not essential for motility or biogenesis of the thickpili but stimulates motility by phosphorylation of some yetunidentified component(s) of the motility apparatus or signaltransduction pathway to regulate it. In this context, it is of notethat the putative protein phosphatase gene slr2031 was also aregulatory factor for motility (9). In theory, it is not impossiblethat both SpkA kinase and Slr2031 phosphatase attack thesame target protein, which is involved in motility. Understand-ing of the complicated processes of motility in cyanobacteriarequires determination of the target protein(s) for these en-zymes.

ACKNOWLEDGMENTS

This work was supported by a Research Fellowship for Young Sci-entists from the Japan Society of the Promotion of Science (to A.K.),Grants-in-Aid for Scientific Research on Priority Areas C “Genome

VOL. 183, 2001 EUKARYOTIC-TYPE PROTEIN KINASE SpkA IN SYNECHOCYSTIS 1509

on June 3, 2018 by guesthttp://jb.asm

.org/D

ownloaded from

Biology” (12206002) (to M.I.) and for Scientific Research C(08836002) and B (11554035, 09NP1501) (to M.I.) from the Ministry ofEducation, Science and Culture, Japan, and a grant for ScientificResearch from the Human Frontier Science program (to M.I.).

REFERENCES

1. Bhaya, D., N. Watanabe, T. Ogawa, and A. R. Grossman. 1999. The role ofan alternative sigma factor in motility and pilus formation in the cyanobac-terium Synechocystis sp. strain PCC6803. Proc. Natl. Acad. Sci. USA 96:3188–3193.

2. Bhaya, D., N. R. Bianco, D. A. Bryant, and A. R. Grossman. 2000. Type IVpilus biogenesis and motility in the cyanobacterium Synechocystis sp.PCC6803. Mol. Microbiol. 37:941–951.

3. Bork, P., N. P. Brown, H. Hegyi, and J. Schultz. 1996. The protein phospha-tase 2C (PP2C) superfamily: detection of bacterial homologues. Protein Sci.5:1421–1425.

4. Carrera, A. C., K. Alexandrov, and T. M. Roberts. 1993. The conservedlysine of the catalytic domain of protein kinases is actively involved in thephosphotransfer reaction and not required for anchoring ATP. Proc. Natl.Acad. Sci. USA 90:442–446.

5. Castenholz, R. W. 1973. Movements, p. 320–339. In N. G. Carr and B. A.Whitton (ed.), The biology of blue-green algae. Blackwell Scientific Publi-cations Ltd., Oxford, United Kingdom.

6. Diehn, B., M. E. Feinleib, W. Haupt, E. Hildebrand, F. Lenci, and W.Nultchh. 1979. Terminology of behavioral responses of motile microorgan-isms. Photochem. Photobiol. 26:559–560.

7. Hader, D.-P. 1987. Photosensory behavior in procaryotes. Microbiol. Rev.51:1–21.

8. Hanks, S. K., A. M. Quinn, and T. Hunter. 1988. The protein kinase family:conserved features and deduced phylogeny of the catalytic domains. Science241:42–52.

9. Kamei, A., T. Ogawa, and M. Ikeuchi. 1998. Identification of a novel gene(slr2031) involved in high-light resistance in the cyanobacterium Synechocys-tis sp. PCC 6803, p. 2901–2905. In G. Garab (ed.), Photosynthesis: mecha-nism and effects. Kluwer Academic Publishers, Dordrecht, The Netherlands.

10. Kaneko, T., S. Sato, H. Kotani, A. Tanaka, E. Asamizu, Y. Nakamura, N.Miyajima, M. Hirosawa, M. Sugiura, S. Sasamoto, T. Kimura, T. Hosouchi,A. Matsuno, A. Muraki, N. Nakazaki, K. Naruo, S. Okumura, S. Shimpo, C.

Takeuchi, T. Wada, A. Watanabe, M. Yamada, M. Yasuda, and S. Tabata.1996. Sequence analysis of the genome of the unicellular cyanobacteriumSynechocystis sp. strain PCC 6803. II. Sequence determination of the entiregenome and assignment of potential protein-coding regions. DNA Res.3:109–136.

11. Leonard, C. J., L. Aravind, and E. V. Koonin. 1998. Novel families ofputative protein kinases in bacteria and archaea: evolution of the “eukary-otic” protein kinase superfamily. Genome Res. 8:1038–1047.

12. Rippka, R., J. Deruelles, J. B. Waterbury, M. Herdman, and R. Y. Stanier.1979. Generic assignments, strain histories and properties of pure cultures ofcyanobacteria. J. Gen. Microbiol. 111:1–61.

13. Shi, L., M. Potts, and P. J. Kennelly. 1998. The serine, threonine, and/ortyrosine-specific protein kinases and protein phosphatases of prokaryoticorganisms. A family portrait. FEMS Microbiol. Rev. 22:229–253.

14. Stanier, R. Y., R. Kunisawa, M. Mandel, and G. Cohen-Bazire. 1971. Puri-fication and properties of unicellular blue-green algae (order Chroococcales).Bacteriol. Rev. 35:171–205.

15. Stock, J. B., A. M. Stock, and J. M. Motten. 1990. Signal transduction inbacteria. Nature 344:395–400.

16. Udo, H., J. Munoz-Dorado, M. Inouye, and S. Inouye. 1995. Myxococcusxanthus, a Gram-negative bacterium, contains a transmembrane proteinserine/threonine kinase that blocks the secretion of b-lactamase by phos-phorylation. Genes Dev. 9:972–983.

17. Waterbury, J. B., J. M. Willey, D. G. Franks, F. W. Valois, and S. W. Watson.1985. A cyanobacterium capable of swimming motility. Science 230:74–76.

18. Williams, J. G. K. 1988. Construction of specific mutations in photosystem IIphotosynthetic reaction center by genetic engineering methods in Synecho-cystis sp. PCC 6803. Methods Enzymol. 167:766–778.

19. Yoshihara, S., X. X. Geng, S. Okamoto, K. Yura, T. Murata, M. Go, M.Ohmori, and M. Ikeuchi. Mutational analysis of genes involved in pilusstructure, motility and transformation competency in the unicellular motilecyanobacterium Synechocystis sp. PCC 6803. Plant Cell Physiol., in press.

20. Yoshihara, S., F. Suzuki, H. Fujita, X. X. Geng, and M. Ikeuchi. 2000. Novelputative photoreceptor and regulatory genes required for the positive pho-totactic movement of the unicellular motile cyanobacterium Synechocystis sp.PCC 6803. Plant Cell Physiol. 41:1299–1304.

21. Zhang, C. C., L. Gonzalez, and V. Phalip. 1998. Survey, analysis and geneticorganization of genes encoding eukaryotic-like signaling proteins on a cya-nobacterial genome. Nucleic Acids Res. 26:3619–3625.

1510 KAMEI ET AL. J. BACTERIOL.

on June 3, 2018 by guesthttp://jb.asm

.org/D

ownloaded from