Maintenance of the Cell Morphology by MinC in Helicobacter pylori

In the model organism Escherichia coli, Min proteins are involved in regulating the division of septa formation. The computational genome analysis of Helicobacter pylori, a gram-negative microaerophilic bacterium causing gastritis and peptic ulceration, also identified MinC, MinD, and MinE. However, MinC (HP1053) shares a low identity with those of other bacteria and its function in H. pylori remains unclear. In this study, we used morphological and genetic approaches to examine the molecular role of MinC. The results were shown that an H. pylori mutant lacking MinC forms filamentous cells, while the wild-type strain retains the shape of short rods. In addition, a minC mutant regains the short rods when complemented with an intact minCHp gene. The overexpression of MinCHp in E. coli did not affect the growth and cell morphology. Immunofluorescence microscopy revealed that MinCHp forms helix-form structures in H. pylori, whereas MinCHp localizes at cell poles and pole of new daughter cell in E. coli. In addition, co-immunoprecipitation showed MinC can interact with MinD but not with FtsZ during mid-exponential stage of H. pylori. Altogether, our results show that MinCHp plays a key role in maintaining proper cell morphology and its function differs from those of MinCEc.


Introduction
Helicobacter pylori, the etiologic agent of human gastritis, peptic ulceration, and gastric carcinoma, infects at least half of the world's population with the organism being highly restricted to the gastric mucosa of humans [1]. During infection, the major actively replicating forms of H. pylori cells are spiral-shaped, but they can convert to cocci under environmental stresses, such as starvation and antibiotic treatment. The coccoid form is viable, but not culturable in vitro. It is less virulent than the spiral form; however, it is thought to be crucial in disease transmission and insensitive to antibiotic treatment [2]. Therefore, cell shape is an important pathogenicity factor for H. pylori. So far, the maintenance and establishment of the spiral structure in H. pylori is known to occur through peptidoglycan relaxation and an intracellular scaffold [3,4,5,6]. While cell division accuracy is crucial for maintaining the shape of some bacteria [7], little is known for H. pylori.
The process of cell division involves the spatial and temporal regulation of the septum formation by the cytoskeletal proteins [7]. In Escherichia coli, precise formation of septum between the two newly segregated sister chromosomes is initiated by FtsZ, which assembles into a ring and recruits a number of proteins, such as FtsA, ZipA, and ZapA, to form the septal ring [8]. In addition, Min proteins are required for the correct division site selection, which prevent polar divisions by blocking the Z ring assembly at cell poles [8]. The functions of FtsZ and min operon have been characterized in many bacteria [9,10,11,12,13]. Although the task of these proteins, which are involved in cell division, is almost identical in reported bacteria, the components and precise regulation mechanisms in preventing polar division of Min systems appears to be different among prokaryotic cells. For example, the Min system, which consists of 3 proteins, MinC, MinD, and MinE, and which is composed of an operon in E. coli, can oscillate periodically between the 2 poles of a cell, and MinE destabilized, but only 2 of them, MinC and MinD, are present in Bacillus subtilis. The function of MinE is substituted by an unrelated DivIVA protein, which targets MinCD in division sites and retains them at the cell poles. However, we found that the homologs of minC, D, and E are present, but not contained, in an operon in the sequenced genomes of H. pylori. Moreover, the amino acid sequences of MinC show a low identity with the corresponding proteins of reported bacteria. So far, whether minC plays any role in cell division of H. pylori remains unclear. Therefore, studying MinC's functions is vital for understanding the cell division and shape-determining factors of H. pylori. In this study, we generated a minC mutant to study its biological characters. Our results show that minC plays a crucial role in maintaining the cell morphology and the movement capabilities of H. pylori.

Bacterial Strains, Plasmids and Growth Conditions
Bacterial strains and plasmids used in this study are listed in Table 1. Strains of E. coli were cultivated in Luria-Bertani (LB) (Difco Laboratories, Detroit, MI) solid and liquid media at 37uC. H. pylori NCTC 11637 was used to construct mutants. H. pylori strains were grown microaerobically at 37uC on blood agar plate (BAP) containing Columbia agar base (Becton Dikinson, Franklin Lakes, NJ, USA) and 5% horse blood or in Brucella broth (Becton Dikinson) containing 5% fetal bovine serum (FBS). Bacterial growth was measured by monitoring OD 600 , while live cells were determined by viable count on BAPs. When required, antibiotics were supplemented: ampicillin (Ap, 100 mg/mL), chloramphenicol (Cm, 30 mg/mL), and kanamycin (Kan, 50 mg/mL).

DNA Techniques
The methods described by Sambrook et al [14] were used for preparation of chromosomal DNAs, restriction digestion, DNA ligation and E. coli transformations. Plasmids were isolated by using High-Speed Plasmid Mini Kit (Geneaid, Taipei, Taiwan). Natural transformation of H. pylori was performed as described elsewhere [15,16].
Cell Length Determination, Immunostaining, and Image Acquisition H. pylori from overnight liquid cultures was inoculated into fresh Brucella broth to obtain an initial OD 600 of 0.05 and grown to an OD 600 of 0.6 to 0.8. Cells were examined microscopically on poly L-lysine-treated slides with a thin layer of 1% agarose in LB. Cell length was measured as the axis length from one pole to the other of the cells captured in microscope, using ImageJ version 1.46 (http://rbs.info.nih.gov/ij/). Average cell length was determined using at least 2 independent measurements, each on 200 cells. DNA was stained with 49, 6-diamidino-2-phenylindole (DAPI; Sigma, St. Louis, MO) at a final concentration of 1 mg/mL and membrane was stained with FM4-64 (Molecular Probes/Invitrogen) at a concentration of 1 mg/mL. Bacterial viability was determined by staining the cells with SYTO9/propidium iodine (PI) (LIVE/DEAD BacLight kit, Molecular Probes/Invitrogen) at 24, 48, and 72 h, followed by fluorescence microscopic observation of the stained cells. Images were obtained with a Nikon E800 microscope using a 1006Objective with A = 1.45 and processed using Adobe Photoshop CS3. The subcellular localization of MinC Hp was carried out using immunofluorescence (IF) microscopy [17]. Bacteria were spread on a clean glass slide and allowed to dry briefly. Bacteria on the glass slides were fixed with methanol at room temperature for 15 min, followed by incubation with 0.1% Triton X-100 in PBS for 1 h. The bacteria were treated with 100 mg/mL of lysozyme and 5 mM EDTA in PBS for 1 h at room temperature. Prior to IF staining, bacteria were incubated with 10% (w/v) bovine serum albumin (BSA) in PBS for 30 min at 37uC to block nonspecific binding. Three PBS washes were performed following each incubation or treatment. After incubation for 1 h with anti-MinC Hp (1:200), the slides were washed 5 times with PBS containing 0.05% Tween 20 (PBST). Incubation using FITCconjugated anti-rabbit IgG (1:500) (Santa Cruz, CA, USA) diluted in blocking buffer was carried out for 30 min at 37uC. The cells were washed 3 times with PBST. The nucleoids were stained with DAPI at a final concentration of 0.5 mg/mL in H 2 O. The cells were washed once in H 2 O. The images of the bacteria were subsequently visualized with a Nikon E800 microscopy.

Sequencing and Identification of the minC Gene
The oligonucleotide primers used in this study are listed in Table 2. Primers HP1054-F and HP1052-R for a PCR corresponded to the nucleotide (nt) 2924 to 2946, relative to the hp1054 start codon, and nt 2269 to 2248, relative to the termination codon of hp1052, respectively. A PCR was performed to amplify the fragment, using the H. pylori NCTC 11637 genomic DNA as the template. The amplicon was purified using the Gel/ PCR DNA Fragments Extraction Kit (Geneaid, Taipei, Taiwan) and directly sequenced using a 3730 DNA analyzer (Applied Biosystems, CA, USA). The sequence analysis was performed using NCBI packages.

Construction and Complementation of a H. pylori minC Mutant
The H. pylori minC was a PCR amplified from the strain NCTC 11637 genomic DNA using primers minCN and minCC. The PCR product was cloned into the SphI site of pUC18 to generate pCPY001. To create an insertional mutant of minC in H. pylori, the chloramphenicol acetyltransferase (cat) cassette form pAV35 (kindly provided by J. M. Ketley) [18] was digested with PvuII and inserted into the unique SphI restriction site 27 bp from the start site of minC Hp in pCPY001 to generate pCPY002. The chromosomal minC locus was disrupted through a homologous recombination upon transforming strain NCTC 11637 with pCPY002. A transformant, selected on BAPs supplemented with Cm, was designated H. pylori PY1. The resulting PY1 was confirmed by a PCR analysis using the same primers of minCN and minCC.
To complement H. pylori PY1, plasmid pCHL2 was constructed in several steps to be used as a vector. (1) The hp0405 gene, amplified using PCR with the primers 0405-F and 0405-R, was cloned into the EcoRV site of pOC10 [19] and the fragment between StuI-HincII restriction sites was subsequently removed, which resulted in the vector pOC0405. (2) The H. pylori flagella promoter, P flaA , was amplified using a PCR with primers pflaF and pflaR and ligated with a pTZ57R/t vector to generate pTZ-PflaA.
(3) The kan cassette of pJMK30 (provided by J. M. Ketley) [18] was cloned into the XbaI site of pTZ-PflaA to generate pTZ-PflaAKm. (4) P flaA and kan were subsequently cloned into the EcoRI site of pOC0405, generating plasmid pCHL2. The obtained pCHL2 vector possessed a unique StuI for cloning genes of interest between P flaA and kan.
Regarding complementation, minC Hp was PCR-amplified from the H. pylori NCTC 11637 genomic DNA, using the primer pair minCN/minCC and was cloned into the unique StuI restriction site of pCHL2 to generate the construct pCPY003. The plasmids were then transformed into mutant PY1 and NCTC 11637 through natural transformation and were selected on BAPs supplemented with Kan. The ectopic integration of the cloned minC in the strain PY2 (PY1-complemented strain) was verified with the PCR using the primers minCN and minCC, in which an amplicon of 1.3 kb (the minC gene plus a cat gene) and 0.6 kb (the minC gene only) were obtained. The ectopic integration of cloned minC Hp in the strain NCTC 11637 was the designated H. pylori PY3 and was verified with PCR analysis using a pair of primers pflaF-minCC. minC Ec was PCR-amplified from pCPY005 using the primer pair NHis-F/minCec-R and cloned into the unique StuI restriction site of pCHL2 to generate the construct pCPY006. Subsequently, the plasmids were introduced into strain PY1 and NCTC 11637 through natural transformation (described above) and transformants were designated H. pylori PY2-5 and PY3-1, respectively. Finally, the complemented strains were verified with the PCR analysis using a pair of primers, pflaF-minCec-R.

Plasmids Construction
The minC Hp and ftsZ gene were amplified by PCR using the genomic DNA of NCTC 11637 as the template, with the primers minCN/minCC and FtsZP-F/FtsZP-R as the primers, respectively. The products were digested with EcoRI and XhoI and cloned into pET30a cleaved with the same enzymes to yield pCPY004 and pCPY007, respectively. The minD gene was amplified by the PCR with the primers PminD1-F/PminD2-R and the amplicon was digested with SacI and HindIII. The SacI-HindIII fragment was cloned into pET30a cleaved with the same enzymes to yield pCPY008. Purified MinC Hp , FtsZ or MinD proteins from E. coli strain BL21(DE3) carrying pCPY004, pCPY007, or pCPY008  were used to raise rabbit anti-MinC Hp , anti-FtsZ, or anti-MinD polyclonal antiserum, respectively (Protech Technology Enterprise, Taipei, Taiwan). The minC Ec gene was amplified by the PCR from E. coli MG1655 genomic DNA using the primers minCec-F and minCec-R. The PCR products were digested with BamHI and SalI and ligated with BamHI-SalI-cleaved pET30a to generate pCPY005. The XbaI-XhoI fragments of pCPY004 and pCPY005 were cloned into XbaI-SalI-cleaved pBAD33 to yield pCPY009 and pCPY010, respectively.

Western Blot Analysis
Bacteria cell liquid cultures were centrifuged at 13,0006g for 1 min and cell pellets were washed twice in PBS. Pellets were resuspended in sterile water. The bacterial suspension was sonicated using an ultrasonicator (Model XL, Misonix, Farmingdale, NY) to break the bacteria. Total protein concentrations were determined by using the Bio-Rad Dc protein assay kit on samples diluted 20-fold in water and on BSA standards in the same diluted buffer. The equal amounts of cell protein per lane were mixed with sample buffer (62.5 mM Tris-HCl, pH 6.8 containing 5% 2mercaptoethanol, 2% SDS, 10% glycerol, and 0.01% bromophenol blue) and heated in a boiling water bath for 10 min. The samples were subjected to polyacrylamide (12%) gel electrophoresis; the protein bands were subsequently transferred onto the polyvinylidene difluoride (PVDF) membranes and probed with rabbit anti-MinC Hp antibodies. A peroxidase-conjugated goat affinity-purified antibody against rabbit immunoglobulin G was used as the secondary antibody (Cell Signaling). Immunoreactive bands were detected using enhanced chemiluminescence (Millipore) and X-ray film. Band intensities on blots were measured using ImageJ version 1.46.

Motility Assay
Cells of H. pylori grown in liquid cultures were stabbed into a soft agar plate containing 0.35% Bacto-agar in Brucella Broth medium and 5% FBS, followed by the incubation of the culture under microaerobic conditions for 3 to 5 days.

Cell Morphology Analysis
To test the MinC sensitivity of E. coli, the plasmid pCPY009 or pCPY010 was introduced into E. coli MG1655. Exponentially growing of strain was serially diluted by 10. Then 3 ml cultures from each dilution was spotted on a plate with and without arabinose (Sigma) and incubated overnight at 37uC. To study the effects of overexpression, MinC Hp , cells harboring pCPY009 were grown overnight in LB medium at 30uC. An overnight culture was diluted 100-fold in an LB medium supplemented with Cm and grown for 2 h at 30uC. The cell cultures were added at various concentrations (0, 0.002, 0.02, or 0.2%) of arabinose and grown at 30uC for an additional 2 h. The cells were spun down, resuspended in LB medium, mixed 1:1 with 2% LB agarose, spotted onto a coverslip, and observed under a light microscope.

Immunoprecipitation
Liquid cultures of H. pylori were grown to mid-log phase (OD 600 = 0.6 -0.8) and harvested by centrifugation at 60006g. Cell-free extracts were prepared by suspending cells in a sonication buffer (20 mM Tris-HCl, pH 8.0 containing 300 mM NaCl) and sonicated for 1 min with an ultrasonicator. The resultant suspensions were then centrifuged at 13,0006g for 20 min at 4uC and the supernatants containing proteins were collected. The supernatants were mixed with 30 ml of 20% (w/v) Protein A Sepharose CL-4B (GE Healthcare) for 1 h to remove nonspecifically bound proteins. An aliquot (50 ml) of supernatant fraction was used as a control for total protein levels prior to immunopre-

Nucleotide Sequence Accession Number
The nucleotide sequence of the H. pylori NCTC 11637 minC has been deposited in the GenBank under accession number KC896795.

Results
min Genes are Present in H. pylori, but minC is not Clustered with minD and minE, and Shares a Low Similarity to other Reported Bacteria As of March 2013, over 40 strains of H. pylori genome have been sequenced and registered in database. Our database search revealed that minC is separated from minD-minE in all sequenced H. pylori strains, different from other bacterial genera in which the corresponding genes are clustered [12,20]. Also, the same genome organization of min occurs in all sequenced Helicobacter spp. Furthermore, genes flanking minC and minDE are conserved in all sequenced Helicobacter spp.: nlpD and lpxC flanking minC gene and ilvC and dprA flanking minDE ( Figure 1A). These findings indicate that these DNA regions are stable.
The minC, minD, and minE genes of H. pylori encode proteins of 195 aa (22.38 kDa), 268 aa (29.31 kDa), and 77 aa (8.92 kDa), respectively. MinD and MinE show high degrees of identity in amino acid sequence to those from other bacteria (48.5% and 43.3% identical to E. coli and B. subtilis MinD, and 29.9% and 14.3% identical to E. coli and B. subtilis MinE/DivIVA, respectively). However, MinC shares low identity with E. coli and B. subtilis MinC proteins, 17.4% and 6.2%, respectively. Therefore, this study further investigates the role of MinC in H. pylori.
Because of the genetic variability of H. pylori, we focused our work on the reference strain NCTC 11637. To investigate the H. pylori minC, the gene was cloned from strain NCTC 11637 by a PCR-based strategy using primers HP1054-F and HP1052-R designed according to the minC flanking sequences of H. pylori strain 26695 (Table 2) for DNA amplification. Sequencing of the amplicon revealed 1,182 bp containing ORF with 195 aa which shared 87% to 95% identity to the MinC from other sequenced H. pylori strains and shared 20%, 13%, 13%, and 14% identity with the homologs from E. coli, B. subtilis, Neisseria gonorrhoeae, and Thermotoga maritime, respectively.
It has been shown that four conserved glycine residues at the MinC C-terminus are essential for MinC functionality as a cell division inhibitor and for the interaction of MinC with other Min proteins in E. coli [21]. Four glycine residues (G 104 , G 122 , G 129 , and G 139 ) were also found at C-terminus of the H. pylori MinC proteins. In addition, a lysine residue at the N-terminus (K 3 ) and an arginine residue (R 102 ) at the C-terminus were also conserved among all known MinC proteins ( Figure 1B). These findings suggest that H. pylori MinC is a factor involved its cell division, functioning to interact with MinD.

Mutation of H. pylori minC causes Cell Filamentation and Growth and Motility Defects
To investigate the role of MinC, a minC mutant of NCTC 11637 was constructed by a marker exchange and designated as PY1. Light microscopic observation of PY1 after gram-staining showed cells with different sizes. As shown in Figure 2, the cell length ranged from 1.6 to 25.7 mm with 64.5% of them falling between 5-10 mm. Comparing to that of the wild-type (2.2260.75 mm), cell elongation of PY1 was obvious. PY1 exhibited a growth rate similar to that of the wild-type, with a generation time of about 6 h ( Figure 3A). The OD 600 of the wild-type reached the maximum (1.34) at 24 h, while that of the mutant increased to a maximum of 1.88 at 36 h. The OD 600 of both strains declined gradually following cessation of growth. However, the number of viable cells of the wild-type was about 10 times more than that of PY1 at 36 h (5.8610 8 versus 4.5610 7 CFU/ml), and the amount of cell protein   To compare the cell morphology and membrane integrity, cells of NCTC 11637 and PY1 grown to 24, 48, and 72 h were stained with LIVE/DEADH kit and examined by fluorescence microscopy. All the wild-type cells were in rod form till 24 h, a portion of them (ca. 6%) became coccoid form at 48 h, and then almost all cells were in cocoid form at 72 h. In contrast, all PY1 cells maintained the elongated form throughout the experiment ( Figure 3). Both strains appeared alive at 24 h; a large portion of the PY1 cells were dead (75%) at 48 h and almost all of them were dead at 72 h. Compared to the mutant cells, larger portions of the wild-type kept alive at 48 h (63%) and 72 h (48%). Furthermore, PY1 contained clearly segregated nucleoids (Figure 4), indicating that mutation in minC caused no defects in chromosome replication or segregation.
It is known that cell morphology affects the cell motility in numerous bacteria [22]. As motility of H. pylori is crucial in colonizing the gastric mucosa [23], we have evaluated the effects of minC mutation on the motility in this study. Tests were performed on a soft agar plate, comparing the area of spreading zones between the minC mutant and its parental strain. The results showed that the motility activity is reduced by half in mutant strains. As shown in Figure 5, growth of the wild-type cells resulted in a spreading zone of 15 mm in diameter after 72 h ( Figure 5), while that formed by PY1 was about 7 mm in diameter ( Figure 5). Since cell division related proteins are not involved in flagellar biosynthesis, it appears that the cellular elongation has reduced the cell motility.

Reversal of mutant minC Phenotype in H. pylori by Complementation
To perform the complementation test, we constructed a vector, pCHL2 (Table 1), that allowed for ectopic integration of the plasmid into H. pylori chromosome. The integration, targeted at the locus of hp0405 of H. pylori, was shown to cause no detectable  effects on the physiology or morphology of H. pylori [24,25]. A complemented strain, PY2, with the minC Hp gene integrated into the locus of hp0405 of PY1 was constructed. The expression of MinC Hp in the complemented strain was confirmed by Western blot analysis ( Figure 2E). A densitometry analysis indicated that the expression of MinC Hp in PY2 was 1.6 times greater than that of the wild-type NCTC 11637. In addition, about 99% of PY2 cells regained normal cell morphology, exhibited a normal cell length distribution ( Figure 2D), and restored their motility ( Figure 5). This result suggested that elongation of PY1 cells was significant because of the minC mutation and it was not a polar effect on downstream gene expression.

Cellular Localization of MinC in H. pylori
In E. coli and B. subtilis, MinC is an effector of the Min system responsible for antagonizing cell division and for preventing the sedimentation of FtsZ [11]. However, consequence of minC mutation may not be the same for H. pylori, because mutation in minC gene causes the cell to elongate instead of mini-cell formation observed in E. coli and B. subtilis.
To detect the cellular location of MinC, IF microscopy was performed using antibodies against MinC and a secondary antibody tagged to FITC. The results showed that MinC in the mid-log cells assembled into helix-form structures and located mainly in poles ( Figure 6).

MinC Hp Interacts with MinD but not with FtsZ during Mid-exponential Stage of H. pylori
To examine whether MinC interacts with MinD and FtsZ in H. pylori, co-IP was performed using antibodies prepared against MinC, FtsZ, or MinD separately, followed by detection of the proteins co-precipitated by Western blotting (Figure 7). Unexpectedly, the MinD protein was precipitated with MinC, but FtsZ was not (Figure 7, lanes 2 and 3).

The Effects of MinC Ec in H. pylori
To test whether MinC of E. coli can complement the deficiency in MinC in H. pylori, the E. coli minC gene was cloned in pCHL2 and introduced into PY1 (forming strain PY2-5) for complementation. Results showed that 81% of the cells had a length shorter than 5 mm with an average of 3.24 mm, demonstrating that MinC Ec could complement the deficiency in MinC Hp in H. pylori.
To inspect the effects of MinC Hp and MinC Ec on H. pylori cell division, minC Hp and minC Ec were cloned and inserted into the hp0405 locus of NCTC 11637, resulting in strains PY3 and PY3-1, respectively ( Table 1). Cells of PY3 had an average length similar to that of the wild-type (2.9961.22 mm) and about 7.1% of them were longer than 5 mm ( Figure 8A). In contrast, the average cell length of PY3-1 increased to 5.3163.36 mm (Table 3), and the cells shorter than 2 mm decreased to 7% ( Table 3).

The Effects of MinC Hp in E. coli
To inspect the effects of MinC Hp and MinC Ec on E. coli cell division, minC Hp and minC Ec were cloned in pBAD33 and introduced into the MG1655, resulting in strains MG1655(pCPY009) and MG1655(pCPY010), respectively. As shown in Figure 9A, growth of the MG1655(pCPY010) cells containing minC Ec gene under control of arabinose-inducible promoter was inhibited by the presence of 0.2% arabinose. But MG1655(pBAD33), carrying the cloning vector only, and MG1655(pCPY009) grew well in the presence of 0.2% arabinose, suggesting that overproduction of MinC Hp was not lethal in E. coli.
Light microscopy showed that MG1655 carrying cloning vector only or containing minC Hp was similar to the wild-type in morphology ( Figure 9B), but MG1655(pCPY010) formed filaments in the presence of 0.002% arabinose. In immunoblotting with anti-MinC Hp serum, it was shown that MinC Hp levels were elevated with increased concentrations of arabinose ( Figure 9C).
IF microscopy showed that MinC Hp localized at both poles of the E. coli cells before septum formation ( Figure 9D). Upon septation, the majority of the cells contained intense fluorescence at septum, while some of them still retained small amounts of fluorescence at the poles ( Figure 9D), suggesting that MinC Hp also localized at the poles during the late stage of septation in E. coli.

Discussion
In many bacteria, Min proteins are involved in regulation of cell division. It is known that not all three min genes are ubiquitously present in all microorganisms and the entire minCDE cluster appears to be present only in Gram negative bacteria [26]. In this study, we reveal that H. pylori possesses homologs of minC and minDE, except that they are in two loci. Our sequence analysis here shows that residues conserved in other bacteria are all present in MinC Hp ( Figure 1B).
MinC is required for inhibition of septation by FtsZ in many bacteria and deficiency in MinC results in over septation that in turn causes mini-cell formation [8]. In contrast, a MinC deficient mutant of H. pylori was found to form elongated cells in this study. Our observations suggest that MinC of H. pylori is involved in normal septation that is required for normal cell division. To our knowledge, this is the first report that MinC is required for normal septation instead of inhibiting septation.
In E. coli, MinE imparts topological specificity by stimulating MinCD oscillation, thereby ensuring that the concentration of  MinCD is highest at the poles [27]. In B. subtilis, MinCDJ are localized at the poles or the site of division through polar targeting by DivIVA [28]. In this study, IF microscopy revealed that MinC Hp in the mid-log cells assembled into helix-form structures and located mainly in poles, but do not interact with FtsZ, suggesting that MinC Hp -FtsZ interaction is not required for mediation of cell division. It is possible that MinC Hp interacts with other proteins during different stages of cell division in H. pylori.
Several studies have shown that Min proteins can function in heterologous background, for examples, the chloroplasts are enlarged when minC of E. coli is introduced into Arabidopsis thaliana [29], cells of B. subtilis transformed with minC of E. coli are elongated [10], E. coli cells transformed with minC and minD of N. gonorrhoeae are elongated [13]. In this study, MinC Ec provided in trans resulted in elongation of the wild-type cells and was able to restore the wild-type length to the mutant PY1 (Table 3). It is possible that expression of MinC Ec may prevent the polymerization of FtsZ Hp in H. pylori, thereby inhibiting cell division and resulting in cell elongation [26]. In contrast, expression of MinC Hp in E. coli did not cause detectable effects on cell morphology. Since the amino acid sequence of FtsZ Hp share high degree of similarity with FtsZ Ec (70%) and MinC Hp exhibited no co-IP reaction with FtsZ Hp , it appears reasonable to predict that MinC Hp cannot interact with FtsZ Ec . Consequently, MinC Hp could not inhibit the Z-ring formation in E. coli. In addition, based on the observations that i) H. pylori and E. coli FtsZ have different architecture of filaments, ii) the FtsZ-ring positions at both central and acentric regions in H. pylori, and iii) daughter cells show considerably different sizes owing to the asymmetrical division of the cells, Specht et al. [9] suggest that FtsZ of H. pylori possesses a unique intrinsic characteristic different from that of E. coli and the cell cycle of H. pylori is clearly dissimilar to that of E. coli. Thus, the present observations that i) minC mutation causes cell elongation instead of mini-cell formation, ii) MinC does not interact with FtsZ, and iii) MinC Hp causes no effects on cell division when expressed in E. coli have confirmed and extended the previous findings in H. pylori cell division.