Characterization of a Coxiella burnetii ftsZ Mutant Generated by Himar1 Transposon Mutagenesis^{▿ †}

Bacterial strains and plasmids.

The bacterial strains and plasmids used in the present study are listed in Table 1. E. coli was grown in Luria-Bertani (LB) broth (33). E. coli transformants were selected on LB agar plates containing either 50 μg of kanamycin (Km)/ml or 10 μg of chloramphenicol (Cm)/ml. C. burnetii Nine Mile phase II (clone 4, RSA439) was propagated in African green monkey kidney (Vero) fibroblasts (CCL-81; American Type Culture Collection) grown in RPMI medium (Invitrogen, Carlsbad, CA) supplemented with 2% fetal bovine serum (FBS) at 37°C in a 5% CO₂ atmosphere. This strain is exempt from U.S. Centers for Disease Control and Prevention select agent regulations (http://www.cdc.gov/od/sap/sap/exclusion.htm) and is suited for work at biosafety level 2 (13).

Construction of Himar1 C9 transposase and transposon plasmids.

The restriction enzymes used in the present study were obtained from New England Biolabs (Ipswich, MA). PCR was performed by using Accuprime Pfx or Accuprime Taq (Invitrogen). PCR primers were obtained from Integrated DNA Technologies (San Diego, CA) or Operon Biotechnologies, Inc. (Huntsville, AL), and their sequences are listed in Table 2. All ligations were carried out by using Ligate-it (USB, Cleveland, OH).

Figure 1A depicts the Himar1 C9 transposase expression plasmid pCR2.1-P1169-Himar1C9, where transposase expression is driven by the C. burnetii CBU1169 promoter (P1169). Himar1C9 from pBVS2-Himar1C9 (generously provided by P. Stewart, Rocky Mountain Laboratories) was amplified by using Accuprime Taq DNA polymerase and the primers Himar1C9-NdeI-F and Himar1C9-HindIII-R and then cloned into pCR2.1-Topo. P1169 was amplified from C. burnetii genomic DNA (gDNA) by using Accuprime Taq and the primers P1169-PstI-F and P1169-NdeI-R and then cloned into the pCR2.1-topo vector, creating pCR2.1-P1169. A PstI-NdeI fragment from pCR2.1-P1169 containing P1169 was subsequently ligated into PstI-NdeI-cut pCR2.1-Himar1C9 to generate pCR2.1-P1169-Himar1C9.

Figure 1B depicts the Himar1 transposon plasmid p1898-Tn. BamHI and BspHI restriction sites were incorporated into the pITR-pflgB-GENT (generously provided by P. Stewart) by PCR using the primers pITR-pflgB-GENT-BamHIF and pITR-pflgB-GENT-BspHIR. The subsequent PCR product was ligated together to create pITR-pflgB-GENT-BB. P1169 was amplified from C. burnetii gDNA using P1169-AvaI-F and P1169-NdeI-R, and the resulting PCR fragment was cloned into pCR-BLUNTII-Topo to create pBLUNTII-P1169. A PCR fragment containing the Cm acetyltransferase (CAT) gene was amplified from pEXP1-DEST (Invitrogen) using the primers CAT-NdeI-F and CAT-EcoRV-R and cloned into pCR-BLUNTII-Topo (Invitrogen) to create pBLUNTII-CAT. The NdeI-EcoRV fragment containing CAT was excised and ligated into NdeI-EcoRV cut pBLUNTII-P1169 to create pBLUNTII-P1169-CAT. The AvaI-EcoRV fragment containing P1169-CAT was excised and ligated into AvaI-EcoRV-cut pITR-pflgB-GENT-BB, resulting in the replacement of the flgB promoter and gentamicin gene with P1169-CAT, to produce pITR-P1169-CAT. The mCherry gene was PCR amplified from pcDNA6.2-C-mCherry-DEST (generously provided by S. Grieshaber, University of Florida) using mCherry-BspHI-F and mCherry-BamHI-R primers and cloned into pCR2.1-Topo to create pCR2.1-mCherry. A BamHI fragment containing mCherry was excised and ligated into BamHI cut pITR-P1169-CAT to create pTn. A promoterless version of CBU1898, encoding a C. burnetii IS1111A transposase with 21 copies in the C. burnetii genome, was amplified using CBU1898NotI-F and CBU1898NotI-R and then cloned into pCR-BLUNTII-Topo to create pBLUNTII-1898. A NotI fragment from pBLUNTII-1898 containing CBU1898 was cloned into NotI-cut pTn to create p1898-Tn. Ampicillin, Km, and Cm are not used in the clinical treatment of Q fever (10). Therefore, use of genes that confer resistance to these antibiotics is appropriate for C. burnetii transformation studies, and their use was approved by the Rocky Mountain Laboratories Institutional Biosafety Committee.

Electroporation of C. burnetii and cultivation of transformed bacteria.

Large-scale p1898-Tn and pCR2.1-P1169-Himar1C9 plasmid purifications were conducted by using a GenElute HP endotoxin-free plasmid maxiprep kit (Sigma-Aldrich, St. Louis, MO) and concentrated by using a Montage PCR filter unit (Fisher Scientific, Pittsburgh, PA). C. burnetii was purified from infected cells by renografin density gradient centrifugation as previously described (5), washed twice in Coxiella transformation buffer (CTM; 272 mM sucrose, 10% glycerol), and then resuspended in CTM at approximately 2.5 × 10¹¹ genome equivalents (GE) per ml. C. burnetii GE were enumerated via TaqMan quantitative PCR (qPCR) as previously described (6). To 50 μl of a C. burnetii suspension, 3 μl (12 μg) of p1898-Tn and 4 μl (14 μg) of pCR2.1-P1169-Himar1C9 were added. The cell suspension was placed in a 0.1-cm gap electroporation cuvette on ice and electroporated with a field strength of 16 or 20 kV/cm, a resistance of 500 Ω, and a capacitance of 25 μF using an ECM630 Electro-Cell manipulator (BTX, Holliston, MA). As controls, C. burnetii was electroporated without DNA or mock electroporated with DNA. For first-round electroporations, 950 μl of RPMI was added directly to the cuvette, and 100 μl of the mixture was used to infect confluent Vero cells in one well of a six-well plate. For the second-round electroporations, the entire 1 ml was used to infect confluent Vero cells in a T-75 flask. Infected Vero cell cultures were incubated for 2 h at room temperature with gentle rocking, and then RPMI supplemented with 2% FBS was added to cell cultures, which were incubated for an additional 22 h. Cm was then added to the media at a final concentration of 5 μg/ml. After 1 week, the cell monolayers were harvested by scraping and disrupted by sonication. The sonicate was centrifuged at 1,000 × g for 5 min to pellet host nuclei and large cell debris, and the supernatants were used to infect new Vero cell monolayers. This process was repeated every 1 to 2 weeks. During all cell culture, the medium was removed every 3 to 4 days and replaced with fresh medium containing antibiotic.

Isolation and whole-genome amplification of gDNA.

C. burnetii gDNA was isolated by using either a PowerMicrobial Midi or a Ultraclean Microbial DNA isolation kit (Mo Bio, Carlsbad, CA) with an additional heating step (85°C for 30 min) prior to the physical disruption of the cells. Plasmid DNA was isolated by using a Qiaprep Spin miniprep kit (Qiagen, Valencia, CA) or a GenElute HP endotoxin-free plasmid maxiprep kit. gDNA was whole-genome amplified using 8 μl of purified gDNA as a template and the Illustra GenomiPhi V2 DNA amplification kit (GE Healthcare, Piscataway, NJ).

PCR and Southern blot detection of integrated transposons.

Detection of the Tn was conducted by amplifying CAT, mCherry, and CAT-mCherry genes using the primer pairs CAT-NdeI-F/CAT-EcoRV-R, mCherry-BspHI-F/mCherry-BamHI-R and CAT-NdeI-F/mCherry-BamHI-R (Table 2), respectively. For Southern blots, gDNA or whole-genome amplified DNA was digested with BamHI and BsaHI or with BsaHI alone and then separated on a 0.8% agarose gel by electrophoresis. Digested gDNA was transferred by blotting to Hybond N+ membranes (GE Healthcare) as described by Sambrook et al. (33), except that the transfer medium used was 0.4 M NaOH. Probe DNA specific to CAT or mCherry genes was generated by PCR using the primer pairs CAT-NdeI-F/CAT-EcoRV-R and mCherry-BspHI-F/mCherry-BamHI-R, respectively. Probe DNA specific to ftsZ (CBU0141) was generated using the primer pair CBU0141-F/CBU0141-R. A probe specific to the 1-kb Plus DNA marker (Invitrogen) was also used. Probe DNA (200 ng) labeling and subsequent blot hybridizations were conducted according to the instructions and using the reagents provided for a Gene Images AlkPhos direct labeling and detection kit (GE Healthcare).

Rescue cloning and sequencing of Tn integration sites.

gDNA was isolated from transformed C. burnetii and digested with ClaI or PsiI for 4 h at 37°C. Digested gDNA was heated at 65°C for 20 min and ligated together. E. coli Top10 cells (Invitrogen) were transformed with ligated DNA and plated on LB agar containing 10 μg of Cm/ml. Colonies containing rescued plasmid with a ColE1 origin of replication were grown overnight in LB broth containing 10 μg of Cm/ml. Plasmid DNA was isolated and sequenced using the sequencing primers ColE1-3′-out and Cm-5′-out (Table 2) to obtain the genomic Tn integration site. Sequences were analyzed by using VectorNti 10 advance (Invitrogen) and the BLAST algorithm (http://blast.ncbi.nlm.nih.gov/Blast.cgi).

Clonal isolation of a C. burnetii Tn mutant.

Vero cells cultivated on 12-mm glass coverslips in a 24-well tissue culture plate were infected with Tn mutagenized C. burnetii. At 5 days postinfection, coverslips were transferred to a 90-mm-diameter petri dish containing approximately 10 ml of RPMI medium, and individual parasitophorous vacuoles were extracted by micromanipulation as previously described (3). The harvested content from individual vacuoles was mixed with 200 μl of RPMI supplemented with 2% FBS, and the suspension was used to infect Vero cells in one well of a six-well tissue culture plate. C. burnetii transformant clones were subsequently expanded in Vero cells in the presence of Cm (5 μg/ml) and purified as described above.

Reverse transcription-PCR (RT-PCR) detection of CAT and mCherry transcripts.

Confluent Vero monolayers in six-well plates were infected with C. burnetii and cells lysed for RNA extraction at 3 and 7 days postinfection. Infected cells were washed twice with phosphate buffered saline (PBS; 1 mM KH₂PO₄, 155 mM NaCl, 3 mM Na₂HPO₄ [pH 7.4]) and lysed with 1 ml of TRIzol (Invitrogen), and then lysates were homogenized twice for 40 s at setting 5 in a FastPrep FP120A instrument using FastRNA Pro Blue vials (MP Biomedicals, Irvine, CA). To each sample, 200 μl of 1-bromo-3-chloropropane (Sigma-Aldrich) was added, and the mixture vortexed for 15 s, followed by heating for 10 min at 65°C. Lysates were subsequently centrifuged for 15 min at 12,000 × g at 4°C, the aqueous phase was removed, and RNA was extracted by using an RNeasy kit (Qiagen). The RNA yield was determined by spectrophotometry (A₂₆₀/A₂₈₀), and the integrity was verified by using Agilent 2100 Bioanalyzer (Agilent Technologies, Santa Clara, CA). A total of 50 ng of each RNA was treated with DNase I, reverse transcribed into cDNA, and amplified by using a CellsDirect One-Step qRT-PCR kit (Invitrogen) according to the manufacturer's protocol. A minus reverse transcriptase control was conducted by using Platinum Taq DNA polymerase (Invitrogen) in place of the Superscript III-Platinum Taq mix in a CellsDirect One-Step qRT-PCR kit. The expression of the CAT and mCherry genes was quantified by qPCR using primers and probes (Table 2) specific for these genes.

Light and immunofluorescence microscopy.

For immunofluorescence labeling, Vero cells on 12-mm glass coverslips in a 24-well tissue culture plate were infected with C. burnetii for 5 days followed by fixation in 100% methanol for 5 min. The C. burnetii parasitophorous vacuole membrane was stained using a monoclonal antibody directed against human CD63 (BD Pharmingen, San Jose, CA) and an Alexa Fluor 488 goat anti-mouse immunoglobulin G (IgG) (Invitrogen) antibody. For live-cell imaging, Vero cells on glass-bottom 35-mm petri dishes were infected for 5 days and then viewed by differential interference contrast (DIC) and confocal fluorescence microscopy. Microscopy was conducted with a modified Perkin-Elmer UltraView spinning disc confocal system connected to a Nikon Eclipse Ti microscope equipped with a Photometrics Cascade:512F digital camera (Roper Scientific, Tucson, AZ). Confocal fluorescent (0.2-μm sections) and DIC images were acquired by using Metamorph software (Universal Imaging, Dowingtown, PA). All images were processed by using ImageJ software (written by W. S. Rasband at the U.S. National Institutes of Health, Bethesda, MD [http://rsb.info.nih.gov/ij/]) and Adobe Photoshop (Adobe Systems, San Jose, CA).

Quantification of C. burnetii infectivity and replication.

C. burnetii infectivity for Vero cells was quantified by using a fluorescent focus-forming unit assay as previously described (6). The kinetics of C. burnetii replication in Vero cells were quantified as previously described (6).

SEM.

Scanning electron microscopy (SEM) was performed on C. burnetii that were purified from Vero cell monolayers at 7 days postinfection. Bacteria were fixed on silica chips (Ted Pella, Inc., Redding, CA) overnight at 4°C with 2.5% glutaraldehyde, 4% paraformaldehyde, and 0.05% sucrose in a 0.1 M sodium cacodylate buffer (pH 6.8). Cells were then postfixed in 0.5% reduced osmium using a Pelco Biowave microwave (Ted Pella, Inc.) at 250 W 2X (2 min on, 2 min off, 2 min on) under 15 in Hg vacuum. The samples were ethanol dehydrated for 45 s in a microwave under vacuum at 250 W and critical point dried in a Bal-Tec cpd 030 drier (Balzers, Pell City, AL). Chips were then coated with 75 Å of iridium in an IBS ion beam sputterer (South Bay Technology, Inc., San Clemente, CA) and imaged by using a Hitachi S-5200 In-Lens SEM (Hitachi, Pleasanton, CA).

Expression of recombinant FtsZ, FtsZ::Tn, and FtsA.

Full-length C. burnetii ftsZ and ftsA were amplified by PCR using the primer pairs CBU0141-F/CBU0141-R and CBU0140-F/CBU0140-R, respectively. ftsZ::Tn was amplified using the primer pair CBU0141-F/CBU0141-B2-R (Table 2). PCR products were cloned into pENTR-d-Topo (Invitrogen), and the ftsZ and ftsZ::Tn plasmid clones were subsequently used in a LR clonase II reaction with pEXP1-DEST (Invitrogen) to generate plasmids encoding N-terminally Xpress-tagged protein (Table 1). The ftsA plasmid clone was used in an LR clonase II reaction with pDEST-15 (Invitrogen) to generate a plasmid encoding an N-terminally glutathione S-transferase (GST)-tagged protein (Table 1). The pEXP1-ftsZ, pEXP1-ftsZ::Tn, and pDEST-15-ftsA plasmids were used as templates in in vitro transcription/translation (IVTT) reactions using the RTS 100 E. coli HY kit (Roche, Indianapolis, IN) to achieve expression of cell-free recombinant protein. Total protein from IVTT reactions was separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) on a 10% gel and transferred to a nitrocellulose membrane (Bio-Rad, Hercules, CA). After transfer, membranes were blocked for 1 h at room temperature in Tris-buffered saline (TBS; 150 mM NaCl, 100 mM Tris-HCl [pH 7.6]) containing 0.1% Tween 20 and 5% nonfat milk. Membranes were then incubated for 1 h at room temperature in TBS-Tween 20 containing a mouse monoclonal anti-Xpress (Invitrogen) or a goat polyclonal anti-GST antibody (GE Healthcare). Membranes were washed and incubated for 1 h at room temperature in TBS-Tween 20 containing anti-mouse or anti-goat immunoglobulin G secondary antibody conjugated to horseradish peroxidase (Pierce, Rockford, IL). Reacting proteins were detected via enhanced chemiluminescence using SuperSignal West Pico reagent (Pierce).

FtsZ pull-down assays.

An IVTT reaction containing GST-FtsA was mixed with an IVTT reaction containing equal amounts of either Xpress-FtsZ or Xpress-FtsZ::Tn, and the mixtures were incubated for 2 h at 37°C. Glutathione-Sepharose 4B (GE Healthcare) was added to IVTT reaction mixtures, which were then incubated at room temperature for 1 h. Glutathione-Sepharose was pelleted by centrifugation at 500 × g for 5 min and washed five times in TBS containing 1% Triton X-100. The glutathione-Sepharose was resuspended in SDS-PAGE loading buffer and boiled for 10 min. Proteins were separated by SDS-PAGE on a 10% gel and transferred to a nitrocellulose membrane (Bio-Rad). Immunoblotting was conducted as described above using anti-Xpress to detect the presence of Xpress-FtsZ or Xpress-FtsZ::Tn.

Construction of a C. burnetii-optimized Himar1 transposon system.

Chromosomal transposition of the Himar1 transposon occurs in E. coli when the transposon and unregulated Himar1 transposase are placed on the same plasmid. This results in low yields of the original construct (P. Stewart, unpublished data). Therefore, we placed these elements on separate plasmids and used a two-plasmid approach to transform C. burnetii. The promoter for the CBU1169 gene (P1169), which encodes the small heat shock protein Hsp20, was chosen to drive expression of both the Himar1 C9 transposase variant and the CAT and mCherry genes within the Himar1 transposon. Previous microarray and qPCR studies in our lab showed that the CBU1169 promoter is active throughout the C. burnetii infectious cycle (data not shown). The Himar1 C9 transposase plasmid pCR2.1-P1169-Himar1C9 contains the CBU1169 promoter immediately upstream of the Himar1 C9 transposase gene (Fig. 1A). The Himar1 transposon plasmid p1898-Tn contains a ColE1 origin of replication, CAT and mCherry red fluorescent protein-coding genes flanked by the transposon ITR elements (Fig. 1B). The CAT and mCherry genes both have the CBU1169 RBS directly upstream of their respective start codons, which allows expression of both genes as a single transcriptional unit from P1169 upstream from the CAT gene. Resistance to Cm was chosen as a method of positive selection since the antibiotic is efficacious in cell culture models of C. burnetii infection (31) and is not used in the clinical treatment of Q fever (10). mCherry was chosen as a fluorescent marker instead of green fluorescent protein because C. burnetii autofluoresces green when excited with light in the 488-nm range (unpublished data). A promoterless copy of CBU1898, encoding an IS1111A transposase with 21 copies in the C. burnetii genome, was inserted into a NotI site outside of the Himar1 transposon to promote homologous recombination of the transposon plasmid into the C. burnetii chromosome in the event that transposition was unsuccessful.

Himar1 transposon mutagenesis.

Prior to initiating electroporation experiments, the concentration of Cm required to inhibit C. burnetii growth in Vero cells was established. Vero cells were infected with C. burnetii in the presence of 0.5, 2, 5, or 10 μg of Cm/ml. At 5 days postinfection, focus-forming units were not detected in cell cultures treated with 5 or 10 μg of Cm/ml (data not shown). Thus, 5 μg/ml was chosen as the concentration of Cm for the selection of transformants. C. burnetii was electroporated with p1898-Tn and pCR2.1-P1169-Himar1C9 using a field strength of 16 kV/cm as described in Materials and Methods. Ten percent of the electroporated C. burnetii, termed B2, was used to infect one well of a six-well plate containing approximately 10⁶ Vero cells. Controls included C. burnetii exposed to the same amount of both plasmids, but not electroporated, and C. burnetii electroporated in the absence of DNA. At 24 h postinfection Cm was added to the growth media for selection. Every 1 to 2 weeks infected Vero cells were harvested, disrupted gently by sonication, and released C. burnetii used to infect new monolayers under constant Cm selection. After five C. burnetii passages in Vero cells, organism-containing parasitophorous vacuoles were clearly visible by phase-contrast light microscopy only in cells infected with organisms electroporated in the presence of both plasmids. gDNA was isolated from cell cultures and PCR conducted to detect the presence of full-length genes encoding CAT (606 bp) and mCherry (708 bp), and the CAT-mCherry region (1,472 bp) of p1898-Tn. Predicted PCR products were observed only in DNA extracted from cultures infected with B2 (data not shown). To determine whether the p1898-Tn DNA of B2 transformants had integrated by homologous recombination via CBU1898 or was autonomously replicating, PCR of the region from CBU1898 to the CAT gene (using primers CBU1898-NotI-F and CAT-EcoRV-R; Table 2) was conducted. A predicted PCR product of ∼2.3 kb was not detected (data not shown), indicating that transposition of the Himar1 transposon had occurred.

B2 transformants were then expanded, and gDNA was isolated from purified bacteria. Rescue cloning of the Tn ColE1 origin of replication was conducted by cutting B2 gDNA with ClaI, an enzyme that does not cut the Himar1 Tn but has 1,121 cut sites throughout the C. burnetii genome. ClaI-digested B2 gDNA was self-ligated and used to transform E. coli. Plasmid DNA was isolated from 36 Cm-resistant colonies and sequenced to establish the Tn integration site. All clones were identical, with a Tn integration in the 3′ end of CBU0141 that encodes the cell division protein FtsZ (Fig. 2A).

The identification of 36 identical ftsZ transposon insertions by rescue cloning did not necessarily preclude the existence of other Tn insertions within B2 C. burnetii. Therefore, we derived a clonal population of C. burnetii, termed B2c, from B2 using micromanipulation (3). PCR of B2 gDNA using ftsZ-specific primers resulted in only two bands: an ∼3.7-kb band representing ftsZ with the integrated Tn and an ∼1.2-kb band representing full-length ftsZ (Fig. 2B). Tn-disrupted ftsZ was also detected in B2c gDNA, but full-length ftsZ was not. These data indicated B2c was clonal for ftsZ::Tn and that C. burnetii transformants harboring other Tn insertions were absent. The presence of full-length ftsZ in the B2 sample indicated that organisms containing additional Tn insertions were present and/or that B2 cell cultures still contained carry over nontransformed C. burnetii. Transposition of the Tn in ftsZ of B2 and B2c was also confirmed by Southern blotting (Fig. 2C).

Optimization of Himar1 transposon mutagenesis.

We identified a single Himar1 transposon insertion in our initial experiment. Therefore, we tested a higher electroporation field strength (20 kV/cm) to determine whether transformation efficiencies could be improved. In addition, we increased the number of Vero cells used in the initial C. burnetii infection to 8 × 10⁶ cells and, in most cases, infected cells with the entire electroporated sample. Seven electroporation experiments were conducted with C. burnetii electroporation experiments 1, 2, 3, 6, and 7 using a field strength of 16 kV/cm and electroporation experiments 4 and 5 using a field strength of 20 kV/cm. T-75 flasks containing Vero cell monolayers were infected with all electroporated C. burnetii with the exception of electroporation experiment 7, where two T-75 flasks were each infected with one-half of electroporated organisms (experiments 7a and 7b). At 24 h postinfection, 5 μg of Cm/ml was added to each flask for selection. Infected Vero cell monolayers were disrupted, and new monolayers were infected every 1 to 2 weeks under constant antibiotic selection. Consistent with the previous experiment, parasitophorous vacuoles containing Cm-resistant C. burnetii were clearly visible following the fifth passage of C. burnetii transformants in Vero cells. All flasks were positive by PCR for CAT, mCherry, and CAT-mCherry regions (data not shown). Total gDNA (host and bacterial) was extracted from infected cells corresponding to each electroporation experiment and whole genome amplified. Amplified DNA was digested with ClaI or PsiI, enzymes that do not cut within the transposon, and the digested DNA was self-ligated to allow rescue cloning of the Tn ColE1 origin of replication. A total of 94 Cm-resistant colonies were obtained, with 55 and 39 colonies recovered from PsiI- and ClaI-digested DNA, respectively. Table 3 lists the 34 unique Tn integration sites revealed by sequencing rescue-cloned plasmid DNA from each electroporation experiment. The sizes of the plasmids ranged from 3,416 bp to more than 13,000 bp. Twenty-nine Tn insertions were within coding regions, while five Tn insertions were intergenic. Two insertions were found in the QpH1 plasmid. Disrupted coding regions included 16 encoding proteins with predicted functions and 13 encoding hypothetical proteins. Insertions into seven genomic sites (CBU1745, CBU2021, CBU0921, CBU1430, CBU1701, and two intergenic regions) were common to both ClaI- and PsiI-derived rescue clones, indicating restriction sites of both enzymes reside close to the Tn integration site (Table 3). No common Tn insertion events were observed between the seven different electroporation experiments. Electroporation experiments 4 and 5, where organisms were electroporated at the higher field strength (20 kV/cm), yielded the highest number of rescue clones (6 and 13, respectively).

Figure 3 shows a Southern blot of amplified gDNA isolated from C. burnetii associated with each electroporation experiment. DNA was digested with BsaHI, an enzyme that does not cut within the Tn, and BamHI, which has two cut sites within the Tn generating a single 775-bp band containing the mCherry gene. The 775-bp band was evident in all lanes containing DNA from electroporated C. burnetii. Bands greater than ∼1.2 kb were also observed and corresponded to BamHI/BamHI or BamHI/BsaHI fragments containing the ITR-p1169-CAT region of p1898-Tn (∼1.2 kb) plus flanking DNA. Based on the C. burnetii Nine Mile genome sequence, the expected sizes for the BamHI/BamHI or BamHI/BsaHI fragments detected by the CAT probe for each of the 34 mapped Tn insertions were established (see Table S1 in the supplemental material). In general, a band on the Southern blot could be associated with the expected size of the disrupted restriction fragment that resulted from each identified insertion event. For example, in electroporation experiment 1, Tn insertions disrupted CBU0573, CBU0964, CBU1196, CBU1745, and CBU2020 and bands of the expected sizes (1,614, 1,768, 1,729, 2,961, and 1,631 bp, respectively) were evident on the Southern blot (Fig. 3C). The Southern blot also revealed band sizes that could not be associated with a Tn insertion identified by rescue cloning. These bands likely correspond to Tn insertions too far from a ClaI or PsiI restriction site to be efficiently rescue cloned or represent fragments that are toxic in E. coli.

Characterization of the B2c ftsZ::Tn mutant.

The clonal isolation of the B2c transformant allowed analysis of CAT and mCherry gene expression during the C. burnetii infectious cycle and characterization of the mutant's phenotype. Vero cells were infected with B2c and RNA extracted at 3 and 7 days postinfection. By quantitative RT-PCR, CAT gene expression of B2c was determined to be ∼3-fold higher than mCherry gene expression at each time point (data not shown). The lower expression of the mCherry gene may be due to its downstream location relative to the CAT gene on the transcriptional unit contained on the Tn. Nonetheless, transcription of the mCherry gene correlated with red fluorescent organisms in live (Fig. 4A) and fixed (Fig. 4B) Vero cells at 5 days postinfection.

FtsZ is highly conserved in eubacteria and plays an essential regulatory and structural role in bacterial cell division (41). Therefore, we examined B2c for potential growth defects. Growth of B2c and wild-type C. burnetii in Vero cells was quantified by qPCR over 7 days in the absence of Cm. B2c had a generation time during exponential phase of ∼19.8 h with a net 1.63-log increase in GE over 7 days. This contrasted with a generation time of 11.7 h and a net 2.51-log increase in GE for wild-type organisms (Fig. 5). B2c growth kinetics were similar in cultures containing 5 μg of Cm/ml (data not shown). The slower growth of B2c relative to wild-type organisms correlated with a higher percentage of B2c organisms containing a division septa (45.6% versus 8.2%, respectively) (Fig. 6). Six percent of B2c cells also contained more than one division septae, a morphology not seen with wild-type organisms. The occurrence of filamentous forms of B2c with incomplete septae and presumably containing more than one genome was associated with a lower infectivity of B2c for Vero cells than control organisms on a per genome basis. Approximately 2.5 more B2c than wild-type C. burnetii GE were required to generate a single infectious foci in Vero cells (data not shown). Collectively, these data indicate that the Tn insertion in B2c disrupts FtsZ function and consequently cell division.

Binding properties of B2c FtsZ::Tn.

The Tn insertion in the ftsZ gene of B2c results in substitution of the FtsZ C-terminal 25 amino acids (aa) with 17 Tn-encoded amino acids (Fig. 7). The C terminus of native FtsZ contains a motif consisting of 8 aa [L(D/E)(I/V)PX(F/Y)(L/I)(R/K)] that is essential for binding of the membrane-associated protein FtsA and localization of FtsZ polymers to the plasma membrane (41). Therefore, we examined the ability of wild-type C. burnetii FtsZ and B2c FtsZ containing the 17 Tn-encoded amino acids to bind FtsA. Fusion proteins corresponding to FtsZ, FtsZ::Tn and FtsA with predicted molecular masses of 47, 45, and 73 kDa, respectively, were produced by IVTT (Fig. 8A). IVTT lysates were mixed together, and GST pull-down assays were performed. As shown by immunoblotting, GST-FtsA only bound wild-type FtsZ (Fig. 8B). Thus, the growth defects of B2c are associated with the inability of FtsZ::Tn to bind FtsA.