r, a Histone-Like Protein H1 (H-NS) Family Protein Encoded the IncP-7 Plasmid pCAR1, Is a Key Global Regulator
Histone-like protein H1 ( H-NS ) family proteins are nucleoid-associated proteins ( NAPs ) conserved among many bacterial species . 
The IncP-7 plasmid pCAR1 is transmissible among various Pseudomonas strains and carries a gene encoding the H-NS family protein , Pmr . 
Pseudomonas putida KT2440 is a host of pCAR1 , which harbors five genes encoding the H-NS family proteins PP_1366 ( TurA ) , PP_3765 ( TurB ) , PP_0017 ( TurC ) , PP_3693 ( TurD ) , and PP_2947 ( TurE ) . 
Quantitative reverse transcription-PCR ( qRT-PCR ) demonstrated that the presence of pCAR1 does not affect the transcription of these five genes and that only pmr , turA , and turB were primarily transcribed in KT2440 ( pCAR1 ) . 
In vitro pull-down assays revealed that Pmr strongly interacted with itself and with TurA , TurB , and TurE . 
Transcriptome comparisons of the pmr disruptant , KT2440 , and KT2440 ( pCAR1 ) strains indicated that pmr disruption had greater effects on the host transcriptome than did pCAR1 carriage . 
The transcriptional levels of some genes that increased with pCAR1 carriage , such as the mexEF-oprN efflux pump genes and parI , reverted with pmr disruption to levels in pCAR1-free KT2440 . 
Transcriptional levels of putative horizontally acquired host genes were not altered by pCAR1 carriage but were altered by pmr disruption . 
Identification of genome-wide Pmr binding sites by ChAP-chip ( chromatin affinity purification coupled with high-density tiling chip ) analysis demonstrated that Pmr preferentially binds to horizontally acquired DNA regions . 
The Pmr binding sites overlapped well with the location of the genes differentially transcribed following pmr disruption on both the plasmid and the chromosome . 
Our findings indicate that Pmr is a key factor in optimizing gene transcription on pCAR1 and the host chromosome . 
Nucleoid-associated proteins ( NAPs ) have architectural and regulatory functions in bacterial cells . 
Bacterial chromosomal DNA is folded into a compact nucleoid body by NAPs ( 9 , 11 ) . 
Because of their DNA-binding ability , NAPs can also influence the expression of genes ( 9 , 11 ) . 
Histone-like protein H1 ( H-NS ) , a NAP family member , is an oligomeric DNA-binding protein identified in Escherichia coli because of its effect on transcription in vitro ( 13 , 16 ) . 
H-NS acts as a global repressor and binds to horizontally acquired DNA regions ( 28 ) . 
Plasmid-encoded H-NS can function as a `` stealth '' protein to switch off gene expression on chromosomes or plasmids and to maintain host cell fitness ( 15 ) . 
H-NS also interacts with paralogous proteins , such as StpA and Hfp in E. coli , or other NAPs ( 12 , 16 , 27 ) . 
* Corresponding author . 
Mailing address : Biotechnology Research Center , University of Tokyo , 1-1-1 Yayoi , Bunkyo-ku , Tokyo 113-8657 , Japan . 
Phone : 81-3-5841-3064 . 
Fax : 81-3-5841-8030 . 
E-mail : anojiri @mail . 
ecc.u-tokyo . 
ac.jp . 
† C.-S.Y. and C.S. contributed equally to this work . 
‡ Present address : Department of Environmental Life Sciences , Graduate School of Life Sciences , Tohoku University , Sendai 980-8577 , Japan . 
§ Present address : Japan Collection of Microorganisms , Microbe Division , RIKEN BioResource Center , 2-1 Hirosawa , Wako , Saitama 351-0198 , Japan . 
¶ Supplemental material for this article may be found at http://jb . 
asm.org / . 
Published ahead of print on 16 July 2010 . 
Tendeng et al. ( 39 ) suggested that conserved MvaT proteins from Pseudomonas bacteria belong to the H-NS family , despite their limited sequence similarity with H-NS . 
Recently MvaT and MvaU from Pseudomonas aeruginosa PAO1 , functional homologous H-NS proteins from Pseudomonas bacteria , were shown to interact with each other ( 44 ) . 
Castang et al. ( 5 ) reported that these two H-NS family proteins bind to the same chromosomal regions and that they function coordinately . 
Interestingly , P. putida KT2440 has five genes encoding H-NS family proteins , and recently Renzi et al. ( 30 ) named them as follows : PP_1366 ( turA ) , PP_3765 ( turB ) , PP_0017 ( turC ) , PP_3693 ( turD ) , and PP_2947 ( turE ) . 
TurA and TurB were copurified as the TOL plasmid ( pWW0 ) upper operon repressors A and B , respectively , and both bound to the Pu promoter ( a 54-dependent promoter of the operon encoding enzymes for the upper pathway of toluene degradation in pWW0 ) , suggesting that these two proteins could interact with each other ( 31 ) . 
Renzi et al. ( 30 ) proposed that TurA and TurB belonged to groups I and II , respectively , and that these groups contained orthologous H-NS family proteins present in all Pseudo-monadaceae species . 
Conversely , TurC , TurD , and TurE belonged to group III , which contained species-specific H-NS family proteins ( 30 ) . 
The self-transmissible pCAR1 , an IncP-7 archetypal plasmid , endows the host strain with carbazole-degrading ability ( 23 , 36 , 38 ) . 
pCAR1 carries the pmr gene , encoding the H-NS family protein designated Pmr ( plasmid-encoded MvaT-like regulator ) ( 25 ) and belonging to the above-mentioned group III . 
The effect of plasmid carriage on host strains may change in different hosts , and therefore , we performed transcriptome comparisons between pCAR1-free and pCAR1-containing KT2440 strains ( 25 , 35 ) . 
Based on the comparisons , pCAR1 carriage affected the iron acquisition system of the host KT2440 strain , enhanced resistance to chloramphenicol by inducing the mexEF-oprN operon , and induced the transcription of PP_3700 ( parI ) ( 35 ) . 
We also discovered that pmr was transcribed in four distinct Pseudomonas host bacterial strains ( 26 , 35 ) . 
These data suggest that Pmr could interact with other H-NS family proteins , such as TurA , TurB , TurC , TurD , and TurE , encoded on the KT2440 chromosome . 
In the present study , we assessed the in vivo transcriptional profiles of genes encoding H-NS family proteins on both pCAR1 and the KT2440 chromosome . 
Additionally , we investigated the in vitro interaction of Pmr with itself and with other H-NS family proteins . 
Furthermore , we assessed the effect of pmr disruption on the transcriptome of the host strain and identified genome-wide Pmr-binding sites . 
Taken together , we clarified the role of Pmr as a horizontally acquired H-NS family protein . 
MATERIALS AND METHODS
Bacterial strains and plasmids . 
The bacterial strains and plasmids used in this study are listed in Table 1 . 
E. coli strains for cloning and expression of genes were grown in L broth ( LB ) ( 32 ) at 37 °C or 25 °C , and the Pseudomonas strains were cultivated with LB at 30 °C . 
Ampicillin ( Ap ) ( 50 g/ml ) , chloramphenicol ( Cm ) ( 30 g/ml ) , kanamycin ( Km ) ( 50 g/ml ) , gentamicin ( Gm ) ( 120 g/ml ) , rifampin ( Rif ) ( 250 g/ml ) , streptomycin ( Sm ) ( 450 g/ml ) , or tetracycline ( Tc ) ( 12.5 g/ml ) was added to the selective medium . 
For plate cultures , the above media were solidified with 1.6 % agar ( wt/vol ) . 
DNA manipulations . 
Plasmid DNA extraction from E. coli was performed using the alkaline lysis method ( 32 ) , and total DNA from Pseudomonas strains was extracted using hexadecyltrimethylammonium bromide as described previously ( 1 ) . 
Restriction enzymes ( New England Biolabs , Ipswich , MA ; Toyobo , Tokyo , Japan ) and the Ligation High reagent ( Toyobo ) were used according to the manufacturers ' instructions . 
DNA fragments were extracted from agarose gels using the Ezna gel extraction kit ( Omega Bio-Tek , Norcross , GA ) according to the manufacturer 's instructions . 
PCR was performed with Ex Taq Hot Start polymerase ( Takara Bio , Shiga , Japan ) according to the manufacturer 's instructions . 
All other experiments were performed according to standard methods ( 32 ) . 
All primers used are presented in Table S1 in the supplemental material . 
RNA extraction . 
RNA extractions from strain KT2440 , KT2440 ( pCAR1 ) , or KT2440 ( pCAR1 pmr ) were performed as follows : an overnight culture of each strain in LB was washed and transferred into 100 ml NMM-4 buffer ( 37 ) supplemented with 0.1 % succinate by adjusting the turbidity to 0.05 at 600 nm and then incubated at 30 °C in a rotating shaker at 120 rpm . 
At early log phase growth ( turbidity of 0.15 to 0.20 at 600 nm ) , we used the RNAprotect bacterial reagent ( Qiagen , Valencia , CA ) to stabilize the total RNA in the bacterial cultures , and subsequently , RNA extraction was performed using the RNeasy Midi kit ( Qiagen ) or Nucleospin RNA II ( Macherey-Nagel GmbH & Co. . 
KG , Düren , Germany ) according to the manufacturers ' instructions . 
The eluted RNA was treated with RQ1 RNase-free DNase ( Promega , Fitchburg , WI ) at 37 °C for 30 min . 
Following inactivation of the DNase by the addition of the stop reagent and subsequent incubation at 65 °C for 10 min , RNA samples were repurified with th 
RNeasy Mini column ( Qiagen ) or Nucleospin RNA binding column ( MachereyNagel ) according to each manufacturer 's RNA cleanup protocol . 
Primer extension and pmr disruption . 
We identified the transcription start point ( tsp ) of pmr by primer extension analysis , performed as described previously ( 25 ) . 
We used the IRD800-labeled primer PMR-R ( Aloka Co. , Ltd. , Tokyo , Japan ) ( see Table S1 in the supplemental material ) , which anneals to the coding region of pmr from 218 to 237 ( 77782 to 77763 on pCAR1 ; see Table S1 ) . 
The extension reaction was performed with 4 l of 5 First Strand buffer containing 10 g of total RNA , 2 pmol of the labeled primer , 100 U of SuperScript III reverse transcriptase ( Invitrogen , Carlsbad , CA ) , 40 U of RNaseOUT ( Invitrogen ) , 10 mM dithiothreitol ( DTT ) , and 0.5 mM deoxynucleoside triphosphates ( dNTPs ) ( Toyobo ) . 
After denaturation of the RNA and the labeled primer at 65 °C for 5 min , the remaining reagents were added , and then the mixture was incubated at 50 °C for 30 min . 
The extended product was purified by phenol-chloroform extraction and ethanol precipitation and then dissolved in 2 l of H2O and 1 l of IR2 stop solution ( Li-Cor Inc. , Lincoln , NE ) . 
The solution was then denatured at 95 °C for 2 min and subjected to electrophoresis using a Li-Cor model 4200L-2 automated DNA sequencer ( Li-Cor ) . 
A sequence ladder was obtained using the same primer and the template plasmid pUB11 ( Table 1 ) . 
pmr disruption in pCAR1 was designed by removing the region containing the tsp ( 77486 to 77909 ) . 
The 3.8-kb EcoRI-PstI fragment from 75681 to 79457 in pCAR1 ( GenBank/EMBL/DDBJ accession number AB0088420 ) was inserted into pK19mobsacB ( 33 ) , and then the SmaI fragment containing the nonpolar Gm resistance cassette of pSJ12 ( 21 ) was inserted into blunt-ended SalI-SacI sites from 77486 to 77909 in the opposite direction to yield pK19mobsacBpmrGm . 
Using a method described previously ( 29 ) , pK19mobsacBpmrGm was introduced into KT2440 ( pCAR1 ) by filter mating with E. coli S17-1 ( pir ) transformants , and subsequently , double-crossover recombinants were screened . 
Quantitative RT-PCR . 
Quantitative reverse transcription-PCR ( qRT-PCR ) was performed using the ABI 7300 real-time PCR system ( Applied Biosystems , Foster City , CA ) as described previously ( 25 ) . 
The primers used for qRT-PCR are shown in Table S1 in the supplemental material , and all of the products were between 100 and 150 bp in length . 
16S rRNA was used as an internal normalization standard . 
All of the reactions were carried out at least in triplicate , and the data were normalized using the average of the internal standard . 
Preparation of a KT2440 ( pCAR1 ) derivative containing a gene encoding the His-tagged Pmr protein . 
The construction of the KT2440 ( pCAR1 ) derivative strain expressing Pmr containing six histidine ( His ) residues at the C terminus was performed using a homologous recombination-based gene replacement system with suicide vectors , antibiotic resistance selection , and sucrose counterselection ( 33 ) . 
The preparation of the DNA region to replace the pmr gene with a modified gene that expresses His-tagged Pmr was performed by overlap extension PCR as described by Choi and Schweizer ( 6 ) . 
Briefly , the primers Pmr-His01 and Pmr-His02 were used to amplify the His-tagged pmr gene . 
The prim-ers Pmr-His03 and Pmr-His04 were used to amplify the downstream region of the untagged pmr gene . 
Simultaneously , the primers Gm-F and Gm-R were used to amplify the Gm cassette flanked by flippase recognition target ( FRT ) sites from pPS856 ( 20 ) . 
The primers used are listed in Table S1 in the supplemental material . 
These three partially overlapping DNA fragments were amplified and then spliced together by in vitro overlap extension PCR . 
The resulting DNA fragment was cloned into the pT7Blue T vector . 
After verification of the inserted sequence , the fragment was excised and then recombined into the suicide vector pK19mobsacB to yield pK19mobsacBpmrHis . 
The pK19mobsacBpmrHis construct was introduced into KT2440 by filter mating with E. coli S17-1 ( pir ) transformants , and double-crossover recombinants were subsequently screened by sucrose counterselection to yield the KT2440 ( pCAR1 ) derivative , replacing the pmr gene with a gene encoding the His-tagged Pmr protein . 
Finally , the Gm resistance gene was removed by site-specific recombination of FRT sites with Flp recombinase supplied from E. coli S17-1 ( pir ) transformants containing pFLP2Km . 
Then , pFLP2Km was constructed by insertion of the EcoRV fragment of pTKm ( 47 ) containing the Km resistance gene cassette into the ScaI site of pFLP2 ( 20 ) . 
PCR analyses were performed to confirm the final construction of the derivative strain . 
Western blot analysis for growth phase-dependent expression of Pmr . 
Cell lysates for Western blot analyses were prepared using the B-Per reagent ( Pierce Biotechnology , Inc. , Rockford , IL ) according to the manufacturer 's instructions . 
The protein samples were quantified using the bicinchoninic acid ( BCA ) protein assay reagent kit ( Pierce ) , and 40 g of protein sample for Pmr or 5 g of protein sample for the RNA polymerase subunit was loaded in each lane . 
Proteins were separated on a 15 % SDS-polyacrylamide gel and transferred to a Sequi-Blot polyvinylidene difluoride ( PVDF ) membrane ( Bio-Rad , Foster City , CA ) . 
Anti-His antibody ( GE Healthcare Bio-Sciences , Piscataway , NJ ) or anti-RNA polymerase subunit ( NeoClone , Madison , WI ) was used as the primary antibody , and enhanced chemiluminescence ( ECL ) peroxidase-labeled anti-mouse antibody ( GE Healthcare Bio-Sciences ) was used as the secondary antibody . 
Proteins were detected using the Immobilon Western chemiluminescent horse-radish peroxidase ( HRP ) substrate ( Millipore , Billerica , MA ) , and LAS1000 plus ( Fujifilm , Tokyo , Japan ) was used for imaging analyses . 
Overexpression of Pmr and other H-NS family proteins in E. coli cells . 
To construct the C-terminal-His-tagged Pmr expression plasmid , the pET-26b ( ) vector ( Novagen , San Mateo , CA ) was used . 
The insert was amplified by PCR using the pCAR1-covered clone pUB11 as template DNA and the primer set with artificial NdeI and XhoI sites at the 5 and 3 ends of the pmr gene . 
The nucleotide sequence of the insert was confirmed , and the resultant expression plasmid was designated pET-C-His-pmr . 
To express each C-terminal-FLAG-tagged H-NS family protein ( Pmr , PP_0017 [ TurC ] , PP_1366 [ TurA ] , PP_2947 [ TurE ] , PP_3693 [ TurD ] , PP_3765 [ TurB ] ) , pFLAG-CTC ( Sigma-Aldrich , St. Louis , MO ) was used as a vector . 
Each insert was amplified by PCR using the primer set with artificial NdeI and SalI sites at the 5 and 3 ends of each gene and pUB11 ( for pmr ) or total DNA of the P. putida strain KT2440 ( for others ) as a template . 
The resulting expression plasmids were designated pFLAGpmr , pFLAG0017 , pFLAG1366 , pFLAG2947 , pFLAG3693 , pFLAG3765 , and expressed FLAG-tagged forms of Pmr , PP_0017 ( TurC ) , PP_1366 ( TurA ) , PP_2947 ( TurE ) , PP_3693 ( TurD ) , and PP_3765 ( TurB ) , respectively . 
Transformed E. coli BL21 ( DE3 ) harboring each expression plasmid of H-NS family proteins was grown at 25 °C to a cell turbidity at 600 nm of 0.6 to 0.8 and was induced overnight by the addition of isopropyl - D-thiogalactoside ( IPTG ) at a final concentration of 0.5 mM . 
The expression level of each protein was con-firmed by Tricine-SDS-PAGE ( 34 ) . 
Pull-down assays . 
Pull-down assays were performed using the MagneHis protein purification system ( Promega ) . 
Cells expressing His-tagged or FLAG-tagged H-NS family proteins were harvested by centrifugation and washed twice with 25 mM Tris-HCl ( pH 8.0 , 4 °C ) containing 2 mM EDTA and 10 % glycerol . 
Cells were then resuspended in 700 l of MagneHis binding/wash buffer and broken by ultrasonication , and crude extracts were obtained by centrifugation ( 17,000 g , 15 min , 4 °C ) . 
Protein concentrations were estimated with the Bio-Rad protein assay reagent ( Bio-Rad ) according to the manufacturer 's instructions . 
Crude extract ( 200 g ) containing His-tagged Pmr was mixed with the following amounts of crude extract containing FLAG-tagged proteins , according to each protein expression level : Pmr , 225 g ; TurC , 900 g ; TurA , 450 g ; TurE , 225 g ; TurD , 1350 g ; and TurB , 225 g . 
After the addition of 30 l of MagneHis Ni particles ( Promega ) , the protein mixture was incubated at 4 °C and centrifuged ( 10 rpm , 1 h ) . 
Elution of His-tagged Pmr was done according to the manufacturer 's instructions . 
Each protein sample was separated by Tricine-SDS-PAGE and transferred to a PVDF membrane ( iBlot gel transfer stack , PVDF , regular ; Invitrogen ) using the iBlot gel transfer system ( Invitrogen ) according to the manufacturer 's instructions . 
Anti-His antibody ( GE Healthcare Bio-Sciences ) or monoclonal anti-FLAG M2 antibody ( Sigma-Aldrich ) was used as the primary antibody , and ECL peroxidase-labeled anti-mouse antibody ( GE Healthcare Bio-Sciences ) was used as the secondary antibody . 
Detection of the proteins was performed similarly to that described above for Western blot analyses . 
Phenotype MicroArray ( PM ) analyses . 
Phenotypic differences between KT2440 ( pCAR1 ) and KT2440 ( pCAR1 pmr ) in carbon metabolism were compared for cell respiration of each strain using 96-well plate microarrays ( Biolog PM1 and PM2 ; Biolog , Hayward , CA ) ( 4 ) . 
Each plate well contained defined medium with a unique carbon compound plus indicator dye for cell respiration , and each medium was made at Biolog . 
Excluding carbon-free wells ( negative controls ) , the PM1 and PM2 Biolog assays can assess the ability to use 190 carbon compounds as the sole carbon source . 
Experiments were performed in duplicate , according to the manufacturer 's instructions , except that the strains were precultured on R2A plates ( 1.5 % agar ) and data collection was performed manually using the Biolog MicroLog MicroStation system . 
Tiling array transcriptome analyses of pCAR1 and the KT2440 chromosome . 
Transcriptome analyses with our custom-made tiling arrays were performed as described previously ( 26 , 35 ) . 
Briefly , total RNA was extracted in parallel from samples of each host culture ( 1 109 cells from two exponential-phase cultures [ the turbidity of each culture was 0.15 to 0.20 at 600 nm ] derived from two independent precultures ) . 
cDNAs reverse transcribed from these RNAs were hybridized individually with each microarray chip using the GeneChip hybrid-ization oven 640 ( Affymetrix , Inc. , Santa Clara , CA ) at 60 rpm and at 50 °C for 16 h with the KT2440 chromosomal tiling array or at 45 °C for 16 h with the pCAR1 tiling array . 
After washing , staining , and scanning of the chips , the signal intensities for each probe were computed using the Affymetrix Tiling Analysis Software program , v. 1.1 ( TAS ) . 
We used the median signal intensities of the probes located within each gene as an indicator of the expression level . 
Comparisons between two conditions were performed using each of the biologicall duplicated data , and we identified upregulated and downregulated open reading frames ( ORFs ) with fold changes of 1.5 in the four data comparisons ( between replicate 1 of KT2440 ( pCAR1 ) and replicate 1 of KT2440 ( pCAR1 pmr ) and between replicate 1 of KT2440 ( pCAR1 ) and replicate 2 of KT2440 ( pCAR1 pmr ) ; see Tables S2 to S4 in the supplemental material ) . 
The data were visualized using the IGB software package ( Affymetrix ) . 
The fold change of pmr expression levels between KT2440 ( pCAR1 ) and KT2440 ( pCAR1 pmr ) was only 4.5 to 5.5 ( see Table S2 ) because the Gm resistance gene introduced into the pmr gene was transcribed in the counterdirection to the pmr gene , and the read through from the Gm resistance gene was detected ( see Fig . 
S2 ) . 
Chromatin affinity purification coupled with high density tiling chip ( ChAP-chip ) analysis . 
An overnight culture of the KT2440 ( pCAR1 ) derivative expressing 6-His-tagged Pmr in LB at 30 °C was inoculated into 200 ml NMM-4 supplemented with 0.1 % ( wt/vol ) succinate to obtain an initial turbidity at 600 nm of 0.05 and then incubated at 30 °C in a rotating shaker at 120 rpm for 4 h to a turbidity at 600 nm of 0.20 to 0.30 . 
The His-tagged Pmr and DNA in the cells were in vivo cross-linked by the addition of formaldehyde to a final concentration of 1 % for 15 min with shaking at 30 °C . 
The cross-linking reaction was quenched by the addition of glycine to a final concentration of 125 mM for 5 min , and then the cells were washed twice with chilled Tris-EDTA ( TE ) buffer ( pH 8.0 ) . 
The resulting harvested cells were disrupted by sonication on ice in 2.4 ml of QuickPick Imac wash buffer ( Bio-Nobile , Turku , Finland ) . 
After centrifugation ( 17,000 g , 20 min ) , the supernatant was affinity purified using the QuickPick Imac metal affinity kit ( Bio-Nobile ) according to the manufacturer 's instructions to yield 6-His-tagged Pmr . 
Cross-links were dissociated by heating at 65 °C for 4 h , and the resulting DNA was purified using the Qiaquick kit ( Qiagen ) according to the manufacturer 's instructions . 
Terminal labeling of the purified DNA fragments and hybridization to the pCAR1 and KT2440 chromosomal tiling arrays were performed as described above . 
Signal intensities of DNA hybridization on the arrays were computed to identify protein-binding sites using TAS , which uses nonparametric quantile normalization and a Hodges-Lehmann estimator for fold enrichment ( Affymetrix Tiling Array Software v1 .1 User 's Guide ) with the biologically duplicated affinity-purified fractions ( treatment DNA ) and those of DNA isolated from the biologically duplicated wholecell extract fractions before purification ( control DNA ) . 
Microarray data accession number . 
The array data reported in this article have been deposited in the Gene Expression Omnibus ( GEO ) of the National Center for Biotechnology Information ( NCBI ) ( GEO ; http://www.ncbi.nlm.nih . 
gov/geo / ) under the GEO Series accession no . 
GSE21968 . 
RESULTS AND DISCUSSION
Transcriptional profiles of pmr and other H-NS family genes in P. putida KT2440 . 
Transcriptional levels of H-NS family proteins change in the presence or absence of other H-NS homologous proteins , and they are not always transcribed under the same growth condition ( 7 , 27 , 44 ) . 
First , we determined the transcription start point ( tsp ) of pmr to construct a pmr disruptant strain by extinguishing its transcription ( see Materials and Methods ) . 
The tsp of pmr ( 1 ) was located 69 bp upstream of the annotated start codon of Pmr ( nucleo-tide at 77571 of pCAR1 ; see Fig . 
S1 in the supplemental material ) , corroborating our findings using a previous tiling array analysis ( 26 ) . 
To clarify the transcriptional profiles of the H-NS family genes , qRT-PCR analyses were performed for KT2440 , KT2440 ( pCAR1 ) , and KT2440 ( pCAR1 pmr ) , along their growth curves . 
As demonstrated in Fig. 1A and B , the transcriptional levels of turA ( PP_1366 ) , turC ( PP_0017 ) , and turD ( PP_3693 ) in early log-phase growth were higher than those in the stationary phase , whereas turB ( PP_3765 ) and turE ( PP_2947 ) were transcribed in the late log and stationary growth phases , compared with the early log phase growth in KT2440 , confirming a previous report ( 48 ) . 
In KT2440 ( pCAR1 ) , pmr was transcribed in early log phase growth ( the cell turbidity was about 0.18 at 600 nm ) , and its transcription was reduced in the stationary phase ( the cell turbidity was 0.56 ) ( Fig. 1A and B ) . 
The transcriptional profiles of other H-NS-encoding genes did not change with pCAR1 carriage or with pmr disruption ( Fig. 1A and B ) . 
Similar results were also obtained by transcriptome analysis using tiling arrays with these three strains in early log phase growth : the signal intensities of the H-NS family proteins did not change with pCAR1 carriage or with pmr disruption ( Fig. 1C ) . 
Taken together with the results of the transcriptional profiles of pmr , turA , turB , turC , turD , and turE , pmr and turA were the primary transcribed genes in the early log phase growth , whereas turB was transcribed in the late log and stationary growth phases in KT2440 ( pCAR1 ) ( Fig. 1B and C ) . 
Translational profiles of Pmr in P. putida KT2440 ( pCAR1 ) . 
Because previous reports indicated that the translational pro-files of some H-NS family proteins were different from their transcriptional profiles ( 7 , 44 ) , we confirmed the translational profiles of Pmr . 
Western blot analysis was performed with the crude extract from KT2440 ( pCAR1 ) cells in the growth phase that expressed C-terminal-6-His-tagged Pmr . 
Pmr signals in KT2440 ( pCAR1 ) were detected throughout the growth phase , and translational levels of Pmr were higher in the late log and stationary growth phases than in early log phase growth ( Fig. 1A and D ) . 
Notably , the translational profile of Pmr differed from the transcriptional profile ( Fig. 1B and D ) and from those of other previously reported H-NS family proteins ( 7 , 44 ) . 
Currently , we could not explain the physiological meaning ( s ) of the discrepancy between pmr transcription and translation . 
Reciprocal transcription and translation of a gene encoding an H-NS-like protein , Sfh of pSF-R27 , have been investigated in detail before ( 14 ) . 
Those authors showed that a blockade of sfh mRNA translation occurred in early exponential growth and was relieved at the onset of stationary phase , responsible for the expression pattern of Sfh ( 14 ) . 
They proposed that con-finement of Sfh expression may ensure that the conjugative plasmid pSF-R27 carrying sfh minimizes the disruption on the physiology of the host cell ( 14 ) . 
It is therefore possible that Pmr translation may have been regulated in a similar manner to reduce effects on the host cell ; however , further investigations are still necessary to clearly explain the Pmr translation mechanism . 
Pmr interacts with itself and with three other H-NS family proteins . 
Many reports have indicated that H-NS family proteins can interact with themselves and with paralogous proteins , such as StpA , Hfp , or MvaU ( 7 , 22 , 27 , 44 ) . 
Thus , Pmr may interact with itself or other H-NS family proteins expressed from the host chromosome . 
To assess this possibility , we performed pull-down assays followed by Western blot analyses to clarify whether Pmr interacted with itself and/or other H-NS family proteins . 
As revealed in Fig. 2B ( lane 1 of each sample ) , we detected anti-FLAG signals from each crude extract , indicating that each H-NS family protein was expressed in E. coli . 
Anti-His signals were also detected in each eluant after the pull-down assays , indicating that His-tagged Pmr existed in each eluant ( data not shown ) . 
In contrast , anti-FLAG signals in the eluants were detected only in the mixtures of His-tagged Pmr with FLAG-tagged Pmr , TurA , TurB , and TurE , whereas those with FLAG-tagged TurC and TurD were not detected ( Fig. 2B , lane 2 of each sample ) . 
This result indicates that the strength of the interactions between Pmr and itself or between Pmr and TurA , TurB , or TurE is higher than those between Pmr and TurC or TurD . 
One important featur of H-NS family proteins is their modular structure ( 10 ) . 
Additionally , KT2440 proteins have putative structures similar to that of H-NS : a well-conserved amino-terminal oligomerization domain ( see Fig . 
S3A , blue box , in the supplemental material ) , a conserved carboxyl-terminal nucleic acid-binding domain ( see Fig . 
S3A , red box ) , and a poorly conserved flexible linker that connects the two aforementioned domains ( see Fig . 
S3A ) . 
When the amino acid sequences of the H-NS family proteins of KT2440 were aligned , their putative oligomerization domains at the N-terminal regions were well conserved ( see Fig . 
S3A ) , although the identity between H-NS family proteins of KT2440 , including Pmr and the H-NS protein of E. coli , was low ( see Fig . 
S3B ) . 
Although it was difficult to predict why Pmr could have heteromeric interactions with three H-NS family proteins but not with two other H-NS family proteins , some residues from the latter may be important for the interaction . 
Notably , the homologous proteins of TurA and TurB are conserved in all Pseudomonadaceae species , but TurC , TurD , and TurE are species-specific proteins ( 30 ) encoded in the putative horizontally acquired DNA region ( 24 ) . 
Taken together with the result that turA and turB were transcribed primarily in the early log and late log growth phases , respectively , Pmr may primarily interact with TurA and TurB , although the functional significance of TurE is presently unclear . 
Considering the reciprocal transcription and translation of Pmr ( Fig. 1B and D ) , it is necessary to analyze the translational levels of Tur proteins in vivo . 
Phenotypic alteration by pmr disruption . 
To assess the effects of pmr disruption on the phenotypes of KT2440 ( pCAR1 ) , comparisons of the catabolic abilities of KT2440 ( pCAR1 ) and KT2440 ( pCAR1 pmr ) were performed using Biolog PM analyses by measuring the absorbance of colored cultures derived from a tetrazolium dye used as a reporter of cell respiration . 
From the comparisons for each of the 190 substrates as a sole carbon source , reproducible reductions of the maximum absor-bance of the color were observed in the KT2440 ( pCAR1 pmr ) culture , compared with results for the KT2440 ( pCAR1 ) culture , with nine compounds ( D-fructose , L-serine , L-valine , saccharic acid , D-malic acid , pyruvic acid , methyl pyruvate , D-ribono-1 ,4 - lactone , and inosine ; see Fig . 
S4 in the supplemental material ) . 
We did not detect any difference between the two strains in the culture using the other carbon sources ( for ex ample , the result with D-glucose was shown in Fig . 
S4 ) . 
These results indicated that pmr disruption affected the catabolic abilities of KT2440 ( pCAR1 ) with several carbon sources , suggesting that Pmr may function as a global regulator of many genes . 
Transcriptome alteration by pmr disruption . 
To confirm the effects of pmr disruption on the host cells , we performed transcriptome comparisons between KT2440 ( pCAR1 ) and KT2440 ( pCAR1 pmr ) using custom-made tiling arrays of genome sequences of pCAR1 and the KT2440 chromosome ( 26 , 35 ) . 
To evaluate the transcriptional and translational profiles of pmr ( Fig. 1 ) , transcriptome comparisons were performed for cells in early log phase growth . 
Overview . 
We found that the transcription of 31 genes on pCAR1 and 159 genes on the KT2440 chromosome were altered by pmr disruption , with a fold change of 1.5 ( see Materials and Methods ; see also Tables S2 and S3 in the supplemental material ) . 
We identified 2 and 19 upregulated genes on pCAR1 and the KT2440 chromosome , respectively , and 29 and 140 downregulated genes on pCAR1 and the KT2440 chromosome , respectively . 
Based on our previous study ( 35 ) , we identified 112 genes altered by pCAR1 carriage with a fold change of 1.5 in both of the duplicate data ( see Table S3 ) . 
Notably , the number of downregulated genes following pmr disruption was larger than that with pCAR1 carriage ( see Fig . 
S5 ) , suggesting that Pmr may play an important role in mediating the transcription of the chromosomal genes of the host KT2440 by pCAR1 carriage . 
The comparison of the transcriptome changes with pCAR1 carriage with those with pmr disruption enabled us to classify 5,398 genes of KT2440 ( after rRNA and tRNA removal ) based on their transcriptional patterns . 
First , the transcription of 5,146 genes was not affected by pCAR1 carriage or by pmr disruption . 
Among the remaining 252 genes , 43 ( group A ) or 50 ( group B ) were upregulated or downregulated by pCAR1 carriage , respectively , but neither of their transcription levels was affected by pmr disruption ( Fig. 3 ; see also Table S4 ) . 
Only one gene ( group D ) was downregulated by both pCAR1 carriage and pmr disruption ( no gene was classified into group C ) ( Fig. 3 and Table 2 ; see also Table S4 ) . 
Seventeen genes ( group E ) were upregulated by pCAR1 carriage but downregulated by pmr disruption , and one gene ( group F ) was the reverse ( Fig. 3 ; see also Table S4 ) . 
In total , 122 genes ( group G ) or 18 genes ( group H ) were upregulated or downregulated by pmr disruption , respectively , but neither group was affected by pCAR1 carriage ( Fig. 3 ; see also Table S4 ) . 
Doyle et al. ( 15 ) proposed that these H-NS proteins encoded on plasmids have `` stealth '' functions to minimize the effect on host strain fitness , comparing the number of genes whose transcriptional levels were altered in the presence or absence of the H-NS protein . 
Regarding KT2440 ( pCAR1 ) , the number of differentially transcribed genes with pmr disruption ( 159 genes on the chromosome ) was larger than that with pCAR1 carriage ( 112 genes ) . 
Additionally , 88 % of them ( belonging to group G or H ) were altered only by the absence of Pmr , suggesting that Pmr had a `` stealth '' function , as mentioned above . 
The transcription levels of only 12 % of the differentially transcribed genes with pmr disruption ( 18 genes ) reverted t levels similar to those in pCAR1-free KT2440 ( groups E and F in Fig. 3 ; see also Table S4 ) , suggesting that these genes were regulated primarily by Pmr itself , directly or indirectly . 
These results suggest that Pmr is a key global regulator of many genes , both on pCAR1 and on the host chromosome . 
Martins dos Santos et al. ( 24 ) demonstrated that KT2440 had many putative horizontally acquired DNA regions . 
These regions include 1,105 ORFs , corresponding to about 20 % of the total ORFs in KT2440 . 
Because H-NS family proteins bind to horizontally acquired DNA regions ( 16 , 28 ) , we calculated the ratio of ORFs in the regions in the above differentially transcribed genes ( Fig. 3 ) . 
Of the 112 genes ( Fig. 3 , groups A to F ) differentially transcribed by pCAR1 carriage , 23 ( 21 % ) were located in the putative horizontally acquired DNA region ( Table 2 ; see also Table S4 in the supplemental material ) . 
Conversely , 56 ( 35 % ) of 159 genes differentially transcribed by pmr disruption ( Fig. 3 , groups C to H ) were in this region ( Table 2 ; see also Table S4 ) . 
Notably , the proportions of groups B , G , and H were high : 28 % , 39 % , and 28 % , respectively ( Table 2 ; see also Table S4 ) . 
The average G C content of pCAR1 and the KT2440 chromosome is 56.3 % and 61.6 % , respectively . 
We then calculated the G C content in the 500 bp upstream of each ORF . 
The G C content of pCAR1-borne genes differentially transcribed by pmr disruption was significantly below the average : for most upstream regions of 30 among 31 affected ORFs ( Table 3 ) , it was below 61.6 % , and for those of 27 ORFs , including the car or parAB genes ( see Table S2 ) , it was even below 56.3 % ( Table 3 ) . 
Concerning the ORFs on the KT2440 chromosome , the G C content of the upstream regions of 92 ( 58 % ) among 159 ORFs was below 61.6 % , and that for 37 ORFs was below 56.3 % ( Table 3 ) . 
As revealed in Table 3 , the ratio of these ORFs to the total affected ORFs was higher ( 87 % in pCAR1 and 23 % in the KT2440 chromosome ) than the ratio of the ORFs whose upstream regions were low in G C content ( below 56.3 % ) to the total ORFs ( 64 % in pCAR1 and 17 % in the KT2440 chromosome ) . 
Notably , the ORFs with a ratio of 56.3 % in the upstream region ( 16 ORFs ) among the ORFs affected by pCAR1 carriage ( 112 ORFs ) was 14 % ( Table 3 ) . 
Thus , some ORFs with low-G C regions may be specifically regulated by Pmr . 
Downregulated genes on pCAR1 with pmr disruption . 
The transcription levels of the genes on the car operon , involved in carbazole degradation , were downregulated ( Fig. 4A ; see also Table S2 in the supplemental material ) . 
When KT2440 ( pCAR1 ) is grown with succinate , the car operon is constitutively transcribed from the PcarAa promoter ( 26 , 35 ) , and it is induced by anthranilate , an intermediate of the carbazole deg-radation pathway , from the Pant promoter , further upstream ( 42 ) . 
Thus , the constitutively expressed carbazole-degrading enzymes will be required to produce anthranilate . 
This suggests that the downregulation of the constitutive transcription levels of car genes may have caused the growth delay with carba-zole . 
In fact , the growth rate of KT2440 ( pCAR1 pmr ) was delayed compared with that of KT2440 ( pCAR1 ) in NMM-4 buffer with carbazole as a sole carbon source ( data not shown ) . 
The transcriptional levels of the parAB genes were also reduced in the pmr disruptants ( Fig. 4B ; see also Table S2 ) . 
The parAB genes are required for the stable maintenance of pCAR1 in the host strain ( 36 ) , and thus , the downregulation of these genes may cause instability of pCAR1 . 
However , we did not detect changes in the stability of pCAR1 or pCAR1 pmr in KT2440 cells ( data not shown ) , suggesting that the effects of the downregulation of the parAB genes on plasmid stability may be insignificant . 
It is also possible that the chromosomally encoded ParAB system ( ParABKT2440 ) for the partition of the KT2440 chromosome may have been involved in plasmid partition ; however , the transcriptional levels of these genes were unaltered in the pmr disruptant ( data not shown ) . 
Additionally , the cis-acting centromere-like parS sequence is indispensable for the function of ParABKT2440 ; however , the 16-nucleotide ( nt ) parS sequence of P. putida KT2440 ( 5 - TGTTNCACGT GAAACA-3 ) ( 3 , 18 ) was not found in the pCAR1 sequence ( data not shown ) . 
The reason pCAR1 pmr was stable in the host strain was not clear . 
Notably , the transcriptional levels of the car and parAB genes were altered in different host strains ( 26 , 35 ) , and the transcription of these genes may be related to the Pmr concentration . 
pmr disruption alters chromosomal gene transcription that is upregulated by pCAR1 carriage . 
The mexEF-oprN operon , encoding the efflux pump , was upregulated in KT2440 ( pCAR1 ) and downregulated in KT2440 ( pCAR1 pmr ) ( Fig. 5A ; see also group E of Table S4 in the supplemental material ) . 
In our previous study , these gene products enhanced the chloramphenicol ( Cm ) resistance of the host strain ; KT2440 ( pCAR1 ) showed resistance to concentrations of Cm higher than 300 g/ml , although KT2440 was not able to grow with that concentration . 
( 35 ) . 
Therefore , we assessed the Cm resistance ( 300 g/ml ) of the pmr disruptants . 
Cm resistance reverted to the levels of pCAR1-free KT2440 , indicating that the downregulation of the mexEF-oprN operon may occur with the loss of resistance at that concentration . 
Westfall et al. ( 45 ) reported that the transcription of mexEF-oprN orthologous genes in P. aeruginosa PAO1 ( normally untranscribed ) was induced in the mvaT ( PA4315 ) mutant on the PAO1 chromosome . 
Thus , H-NS family proteins may be involved in the transcriptional regulation of these genes in PAO1 . 
Although our case contrasted with the PAO1 case , i.e. , pmr disruption caused the downregulation of the mexEF-oprN operon , Pmr may have contributed to the transcriptional regulation of these genes . 
Herrera et al. ( 19 ) recently reported that PhhR ( PP_4489 ) , a transcriptional regulator of phenylalanine hydroxylase phhAB genes , modulates the level of expression of mexEF-oprN together with MexT ( PP_2826 ) . 
Notably , the transcriptional levels of both phhR and mexT were not changed by pCAR1 carriage or by pmr disruption ( data not shown ) , suggesting that Pmr may be the third element for the regulation of the mexEF-oprN operon . 
The parI gene encodes a putative ParA-like ATPase containing an N-terminal DNA-binding motif , and its transcription was upregulated in KT2440 ( pCAR1 ) but downregulated in KT2440 ( pCAR1 pmr ) ( Fig. 5B ; see also group E of Table S4 in the supplemental material ) . 
This corroborated our previous results that the parI promoter was activated in the presence of pCAR1 because of the parA product from pCAR1 ( 25 ) . 
Therefore , the decrease in the parI transcriptional level in KT2440 ( pCAR1 pmr ) was caused by the reduced parA transcription ( Fig. 4B ; see also Table S2 ) , although the reasons for the parA gene downregulation in KT2440 ( pCAR1 pmr ) remain unclear . 
Pmr preferentially binds to foreign DNA and low-G C regions of the host chromosome . 
Because the transcription of many genes was affected by pCAR1 carriage and by pmr disruption , we identified genome-wide Pmr-binding DNA regions on both pCAR1 and the KT2440 chromosome . 
We performed ChAP-chip analyses to identify the Pmr-binding sites on the KT2440 chromosome and in pCAR1 in early log phase growing cells , as well as transcriptome analyses , although the translational levels of Pmr were higher in the late log and stationary growth phases than in early log phase growth ( Fig. 1D ) . 
Consequently , 241 and 26 Pmr-binding sites were detected ( with a P value of 0.01 ) on the KT2440 chromosome and in pCAR1 , respectively ( see Table S5 in the supplemental material ) . 
First , we calculated the G C content of the regions identified . 
The average G C content of the 241 Pmr-binding regions in the KT2440 chromosome was significantly lower ( 52.5 % ) than that of the entire KT2440 chromosome ( 61.6 % ) . 
The 26 Pmr-binding regions in pCAR1 also demonstrated an average G C content ( 52.5 % ) that was lower than that of the entire pCAR1 plasmid ( 56.3 % ) . 
Indeed , a high association was found between the Pmr-binding sites on the KT2440 chromosome and the putative foreign DNA region ( Fig. 6A ) . 
Notably , 73 % of the Pmr-binding sites in the KT2440 chromosome were located in foreign DNA regions . 
Interestingly , many Pmr binding sites in pCAR1 overlapped with the localization of the differentially transcribed genes with pmr disruption ( Fig. 6B ) . 
Similarly , many binding sites in the KT2440 chromosome were also found near regions where the differentially transcribed genes localized , although not every gene near a binding-site region was affected by pmr disruption ( Fig. 6A ) . 
These data indicate that Pmr regulates the transcription of many genes by binding to intergenic or intragenic regions of target genes . 
To determine the relative positions of the Pmr-binding sites to each intergenic or intragenic region of the ORFs , distribution analyses were performed for the ChAP chip analysis data ( Fig. 7 ) . 
The Pmr binding site number peaked at around 200 bp upstream from the translational start point and at around 300 bp downstream from the translational endpoint ( Fig. 7 ) , which was similar in the ChAP-chip analysis when different P-value thresholds were used ( Fig. 7 ) . 
This analysis indicated that Pmr may bind preferentially to intergenic regions rather than to intragenic regions of ORFs , con-firming many reports that H-NS family proteins regulate gene expression by binding to target promoter regions ( 10 , 16 ) . 
Our ChAP-chip analysis demonstrated that Pmr bound preferentially to DNA with a low G C content in KT2440 ( pCAR1 ) and that Pmr bound to intergenic regions and regulated the transcription of genes in the flanking regions of the binding sites . 
We also performed ChAP-chip analysis to identify the binding sites of the TurA and TurB proteins , which are encoded on the KT2440 chromosome . 
However , the detected TurA - and TurB-binding sites were almost identical to those of Pmr , and most of them were DNA regions with low G C content ( data not shown ) . 
These results were similar to those observed with E. coli or P. aeruginosa PAO1 , in which the two H-NS family proteins ( H-NS and StpA or MvaT and MvaU ) bound to the same regions of the chromosome ( 5 , 43 ) . 
Recently Dillon et al. ( 8 ) reported that the DNA binding sites of plasmid-encoded Sfh of Salmonella overlapped with those of H-NS . 
Sfh does not bind uniquely to any site , and the number of binding sites in Sfh is smaller than that in H-NS . 
Although Sfh binding sites are located within H-NS , the DNA binding sites greatly expand in the absence of H-NS , suggesting that Sfh may play a `` backup '' role for H-NS ( 8 ) . 
These facts suggest that the three protein 
Pmr , TurA , and TurB may function coordinately as global regulators in the cells and that Pmr may also perform `` backup '' functions for the other proteins , different from those of H-NS and Sfh , because the binding sites of Pmr , TurA , and TurB are identical . 
However , previous genome-wide analyses of the binding sites of H-NS family proteins , including ours , did not necessarily take into account the in vivo protein-protein interaction ( s ) . 
In other words , the detected sites are not necessarily showing how they bind to the DNA sequences by forming the homo - or heteromultimer of the H-NS family proteins in vivo . 
Considering the coordinate functions for DNA binding and transcriptional regulation by Pmr and TurA to TurE , analyses from protein structure viewpoints will be necessary to understand how they compose the homo - or heteromultimer in vivo in the presence or absence of target DNA . 
Conclusions . 
In this study , we demonstrated that the plas-mid-encoded H-NS family protein Pmr forms homomeric and heteromeric oligomers in vitro and that pmr , turA , and turB are the primary transcribed genes at different growth phases . 
We also revealed that pmr disruption affected the carbon catabo-lism of KT2440 ( pCAR1 ) and that Pmr is a key factor that regulates the transcription of genes on both pCAR1 and the host chromosome in two ways : ( i ) Pmr may alter the transcriptional levels of genes in group E or F , such as mexEF-oprN and parI , and ( ii ) Pmr may minimize the effect of the transcription of many genes in group G or H , such as those in the putative foreign DNA regions . 
The identification of genome-wide binding sites in Pmr by ChAP-chip analysis indicated that Pmr binds to putative foreign DNA regions with low G C content . 
Additionally , Pmr binds preferentially to intergenic regions and may regulate many genes in the flanking regions of the binding sites . 
These findings indicate that Pmr is involved in the regulation of the expression of many genes , directly or indirectly , and that this regulation may be closely related to its DNA-binding regions and its interaction with other H-NS family proteins , primarily TurA and TurB . 
Recently three H-NS family proteins in pathogenic E. coli , the endogenous hns and stpA genes and the horizontally acquired hfp gene , were shown to be differentially transcribed at distinct temperatures ( 27 ) . 
Thus , we must further analyze the transcriptional and translational levels of Pmr , TurA , TurB , TurC , TurD , and TurE under conditions other than those we have used to date . 
Notably , the pmr gene is conserved in another IncP-7 plasmid , pWW53 ( 46 ) , indicating that Pmr is an important protein for IncP-7 plasmids . 
Moreover , H-NS family proteins are expressed from other plasmids , such as H-NS from R27 ( IncHI ) ( 17 ) , Sfh from pSf-R27 ( IncHI ) ( 15 ) , Orf4 from R446 ( IncM ) ( 41 ) , or an undeposited ORF from pQBR103 ( IncP-3 ) ( 40 ) . 
A key function of H-NS expressed from mobile genetic elements is maintaining host cell fitness ( 15 ) . 
H-NS is a member of the NAP family , and its coordinate functions with other NAPs are also important for host cell fitness maintenance ( 9 , 11 ) . 
In the case of the R27 studies , the Hha-like protein , a protein-protein modulator of H-NS activity , is encoded on the plasmid ( 17 ) ; however , no candidates for Pmr modulator-encoding genes are found on pCAR1 . 
Interestingly , pCAR1 harbors two other genes encoding putative NAPs other than Pmr , although their transcriptional levels were unaltered by pmr disruption . 
Analyses of the function ( s ) of these gene products are necessary to understand how these H-NS family proteins behave when pCAR1 is introduced into the host cell by conjugative transfer . 
Such information would help to explain the adaptive and evolutionary mechanisms of bacteria acquiring foreign genes by horizontal gene transfer . 
ACKNOWLEDGMENTS
We thank Akira Yokota of the Institute of Molecular and Cellular Biosciences , the University of Tokyo , for use of his Biolog MicroLog MicroStation system . 
This study was supported by the Program for Promotion of Basic Research Activities for Innovative Biosciences ( PROBRAIN ) in Ja-pan . 
Y.T. was supported by research fellowships from the Japan Society for the Promotion of Science ( JSPS ) for Young Scientists .