Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Open Access

Peer-reviewed

Research Article

Transcriptional Analysis of the Conjugal Transfer Genes of Rickettsia bellii RML 369-C

  • Chan C. Heu ,

    heuxx012@umn.edu

    Affiliation Department of Entomology, University of Minnesota, 219 Hodson Hall, 1980 Folwell Ave, St. Paul, Minnesota, United States of America

    ⨯
  • Timothy J. Kurtti,

    Affiliation Department of Entomology, University of Minnesota, 219 Hodson Hall, 1980 Folwell Ave, St. Paul, Minnesota, United States of America

    ⨯
  • Curtis M. Nelson,

    Affiliation Department of Entomology, University of Minnesota, 219 Hodson Hall, 1980 Folwell Ave, St. Paul, Minnesota, United States of America

    ⨯
  • Ulrike G. Munderloh

    Affiliation Department of Entomology, University of Minnesota, 219 Hodson Hall, 1980 Folwell Ave, St. Paul, Minnesota, United States of America

    ⨯

Abstract

Rickettsia bellii is an obligate intracellular bacterium that is one of the few rickettsiae that encode a complete set of conjugative transfer (tra) genes involved in bacterial conjugation and has been shown to exhibit pili-like structures. The reductive genomes of rickettsiae beg the question whether the tra genes are nonfunctional or functioning to enhance the genetic plasticity and biology of rickettsiae. We characterized the transcriptional dynamics of R. bellii tra genes in comparison to genes transcribed stably and above the background level to understand when and at what levels the tra genes are active or whether the tra genes are degenerative. We determined that the best reference genes, out of 10 tested, were methionyl tRNA ligase (metG) or a combination of metG and ribonucleoside diphosphate reductase 2 subunit beta (nrdF), using statistical algorithms from two different programs: Normfinder and BestKeeper. To validate the use of metG with other rickettsial genes exhibiting variable transcriptional patterns we examined its use with sca2 and rickA, genes involved in actin based motility. Both were shown to be up-regulated at different times of replication in Vero cells, showing variable and stable transcription levels of rickA and sca2, respectively. traATi was up-regulated at 72 hours post inoculation in the tick cell line ISE6, but showed no apparent changes in the monkey cell line Vero and mouse cell line L929. The transcription of tra genes was positively correlated with one another and up-regulated from 12 to 72 hours post inoculation (HPI) when compared to RBE_0422 (an inactivated transposase-derivative found within the tra cluster). Thus, the up-regulation of the tra genes indicated that the integrity and activity of each gene were intact and may facilitate the search for the optimal conditions necessary to demonstrate conjugation in rickettsiae.

Citation: Heu CC, Kurtti TJ, Nelson CM, Munderloh UG (2015) Transcriptional Analysis of the Conjugal Transfer Genes of Rickettsia bellii RML 369-C. PLoS ONE 10(9): e0137214. https://doi.org/10.1371/journal.pone.0137214

Editor: Kelly A. Brayton, Washington State University, UNITED STATES

Received: June 10, 2015; Accepted: August 13, 2015; Published: September 9, 2015

Copyright: © 2015 Heu et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited

Data Availability: All relevant data are within the paper and its Supporting Information files.

Funding: This work was supported by UGM, National Institutes of Health, funding number 5RO1AI081690.

Competing interests: The authors have declared that no competing interests exist.

Introduction

Rickettsiae are obligate-intracellular, coccobacillary, Gram-negative bacteria that include invertebrate symbionts. Genetic analysis of rickettsiae has revealed that plasmids are widely distributed within the genus [1,2] and tra genes are common [3], suggesting that rickettsiae can transfer mobile genetic elements, i.e. plasmids, to receptive bacteria via conjugation. Rickettsia bellii strain RML 369-C [4], Rickettsia massiliae MTU5 [5], Rickettsia felis strain LSU-lb [6], and the rickettsial endosymbiont of Ixodes scapularis (REIS, i.e. Rickettsia buchneri) [7,8] possess a set of tra genes that encodes an apparently complete bacterial conjugation system, but it is not known whether they are functional. In contrast, several other rickettsiae, e.g. Rickettsia felis strain California 2, only encode a few of the tra genes, suggestive of an incomplete conjugation system [2]. Nevertheless, pili-like structures have been observed in R. felis as well as R. bellii [2,4,9]. Furthermore, the tra clusters of R. bellii RML 369-C, R. felis LSU-lb, and R. massiliae MTU are similar at the nucleotide sequence level, gene order, and gene orientation to at least one of the tra clusters of R. buchneri, suggesting multiple lateral gene transfer events between rickettsiae [7,10]. Despite this, bacterial conjugation has not been documented in any rickettsiae.

The initiation of conjugation in Escherichia coli, the model organism, begins with pilus contact with a recipient cell followed by retraction of the pilus and the formation of a stable mating pair [11]. The timing and signal to process the transfer DNA is unknown; nevertheless, the mechanism of generating the transfer DNA has been established. Briefly, TraI is the main processing enzyme with nickase and helicase activity [12,13] that covalently binds to the transfer DNA with the help of accessory proteins [14–16]. Subsequently, TraI-bound DNA docks with TraD [17], an inner membrane coupling protein that initiates the helicase activity of TraI to generate a single-stranded transfer DNA. TraD then opens the pore for transport of the nucleoprotein complex to the recipient thus completing conjugation. E. coli has been shown to express TraI until late stationary phase; however, mating success was highest from mid-log phase to early stationary phase [18]. TraI shows homology to TraATi in R. bellii at the protein level, sharing similar domains with presumably similar functions (Fig 1) as well as belonging to the same class of enzyme, relaxases [19]. As seen in E. coli [18], certain growth phases of rickettsiae may correlate with transcription of traATi and other tra genes, suggesting that conjugation occurs around the time of highest transcription.

thumbnail
Download:
Fig 1. Graphical display of the domains of R. bellii TraATi and its related protein.

MobA/L: this domain is responsible for specific DNA strand transfer in the donor bacterium; ABC ATPase: functions in ATP and GTP binding and hydrolysis, which is a characteristic of conjugative proteins; UvrD C: this domain is found at the carboxyl terminal end of helicases suggesting their role in helicase activity; Toprim: a domain found in DnaG-type primases; TrwC: this domain signifies the relaxase in conjugative DNA transfer during cell-to-cell contact; Conjugal TraA/I: a protein segment involved in DNA binding, nicking, and helicase activity; Primase: a protein segment involved in priming of DNA. The function of each domain was assigned by the Conserved Domain Database. Rbe = R. bellii, Rbu = R. buchneri, Atu = A. tumefaciens, Eco = E. coli, N = amino terminal end of the protein, and C = carboxyl terminal end of the protein.

https://doi.org/10.1371/journal.pone.0137214.g001

The aim of this research was to follow the transcriptional patterns of traATi in R. bellii in comparison with stably transcribed reference gene(s) to facilitate the assessment of the function of the tra genes. We determined the best reference gene at different times of growth in ISE6 (I. scapularis tick embryonic cells), Vero (African green monkey kidney cells) and L929 (mouse fibroblast cells), all of which support continuous R. bellii replication, and may be taken to represent the rickettsial life cycle alternating between the arthropod and the mammalian host. A two-step quantitative reverse transcription PCR (qRT-PCR) and two specialized statistical programs, Normfinder [20] and BestKeeper [21], were used to screen 10 housekeeping genes in R. bellii: 16S rRNA, atpB, dnaK, gltA, gyrA, infB, metG, nrdF, rpoB, and tlc5 (S1 Table). Of these, metG was found to be the most stably transcribed throughout the infection cycles. We validated this as our reference gene by using it to demonstrate the differential transcription patterns of rickA and sca2 as reported [22]. We then used metG to analyze the transcription of the components of R. bellii tra cluster, traATi, traDTi, traL, traE, traB, traV, traC, traW, traU, trbC, traN, traF, traH, traG, traDF, and RBE_0422 during growth in ISE6 cells, and we used metG, nrdF, and a combination of metG and nrdF for a comparative analysis of transcription activity of traATi during replication in different host cells in vitro.

Materials and Methods

Determining the domains of TraATi and related proteins

The location and functions of the domains of TraATi of R. bellii RML 369-C (YP_537591.1), TraI of R. buchneri (KDO03561.1), TraA of Agrobacterium tumefaciens (WP_035228570.1), and TraI of E. coli (WP_038999217.1) were identified using Batch CDD search [23] on the Conserved Domain Database site (http://www.ncbi.nlm.nih.gov/Structure/cdd/cdd.shtml). The graphical representations of the domains of each protein were drawn in Adobe Photoshop to illustrate the location and overall function of each segment of the proteins (Fig 1).

Host cells

Cell line ISE6 [24] was cultured at 34°C using a modified Leibovitz’s L15 medium (L15C-300) supplemented with 5% fetal bovine serum (FBS), 5% tryptose phosphate broth, and 0.1% lipoprotein concentrate as described [25,26]. Vero (ATCC CCL-81) and L929 (ATCC CCL-1) were cultured at 37°C using RPMI+GlutaMAX 1640 medium (Gibco, Grand Island, NY) supplemented with 10% FBS and 2 mM L-glutamine.

Preparation of cell-free R. bellii stabilate for inoculation of host cell cultures

R. bellii RML 369-C was grown in ISE6, Vero and L929 cells in 25 cm2 tissue culture flasks (BD Falcon, NJ) at 34°C in L15-C300 supplemented with 10% FBS for at least 2 subcultures within the respective host cells before preparing and freezing cell-free rickettsia stocks. In brief, infected cells were scraped off the flask, suspended in L15C-300 medium, transferred to 1.5 ml microcentrifuge tubes containing 60–90 grit silicone carbide (Lortone, Inc., Mukilteo, WA), and vortexed at maximum speed for at least 20 seconds to rupture host cells and obtain cell-free R. bellii. Liberated R. bellii were filter-purified using a Whatman 2.0 μm filter (GE Healthcare Life Science, NJ) as described in Oliver et al. (2014), collected by centrifugation at 13,000x rcf for 5 minutes at 4°C and resuspended in freezing medium (L15C-300 with 50% FBS and 10% DMSO), then aliquoted into cryogenic vials and frozen at a controlled rate of -1°C/min for storage in liquid nitrogen.

Inoculation of host cell cultures

Uninfected cells were seeded into 12.5 cm2 tissue culture flasks (BD Falcon, NJ) with 3 flasks for each time point and cells were allowed to attach overnight. Before inoculation, the multiplicity of infection was determined by counting rickettsiae and host cells. Rickettsiae were visualized using LIVE/DEAD BacLight Viability Kit (Molecular Probes, OR) and quantified using a Petroff-Hauser counting chamber. The number of host cells in triplicate cultures was quantified using an Improved Neubauer counting chamber. The number of live rickettsia divided by the host cell counts yielded the multiplicity of infection. Subsequently, the rickettsial stabilates were diluted to the point where inoculating host cell layers with a total volume of 200 μl resulted in 10–50 rickettsiae per host cell. Immediately following inoculation, cultures were transferred to 4°C for 1 hour to allow binding of rickettsiae to cells and to synchronize the infection. Cell cultures were shifted from 4°C to 34°C and cultures incubated at 34°C for 1 hour to facilitate endocytosis of bound rickettsiae. Cell layers were washed twice with 1x phosphate buffered saline containing calcium and magnesium to remove unbound rickettsiae. Fresh L15C-300 with 10% FBS was added to infected cell layers and cultures incubated at 34°C for 12, 24, 36, 48, and 72 HPI for infected ISE6 cultures and 12, 24, 36, 48, 60, 72, 84, and 96 HPI for infected Vero and L929 cultures.

Quantification of R. bellii

Quantification methods were validated by comparing purified cell-free R. bellii (rickettsiae prepared, RP) with yields from host cells infected with R. bellii (whole cell, WC) at 12 and 72 HPI. The total number of R. bellii obtained from each extraction methods was determined by qPCR, using a standard curve based on quantification of a single copy gene, gltA. Statistical significance was measured using a two-sided two sample Student’s t-test with significance set at an alpha of 0.05.

Rickettsial DNA and RNA were extracted from each culture. Whole cell lysis of rickettsiae and host cells was initially used for preparing total DNA and RNA. However, first strand cDNA synthesis using random hexamer yielded low rickettsial cDNA at earlier times post inoculation of rickettsiae; thus, cell-free rickettsial preparation were used. Cell cultures were harvested and RP were prepared with silicone grit as stated previously and stored at -80°C. Pellets were thawed and immediately lysed for DNA and RNA extraction using ZR-Duet DNA/RNA MiniPrep kit (Zymo Research, CA) according to the manufacturer’s protocols. Extracted DNA and RNA were stored at -20°C and -80°C, respectively. qPCR was used to quantify the gltA copy numbers from purified DNA of liberated R. bellii grown in ISE6, Vero, and L929 [1,27]. The pCR4 TOPO (Invitrogen, NY) plasmid fused with the PCR product of Rbellii qCSF/R primers (S1 Table) and the respective primers were used to generate the standard curve for quantifying absolute copy numbers of R. bellii for each time point using Brilliant II SYBR Green qPCR Low Rox Master Mix (Agilent, CA) at 1x with a final concentration of 0.08 μM of each primer, and 5 μl of the total eluted R. bellii DNA. The cycling parameters were 95°C for 10 minutes, followed by 40 cycles of 95°C for 30 seconds, 55°C for 1 minute, and 72°C for 30 seconds. A dissociation curve was used to confirm a single PCR product with a cycle starting at 95°C for 1 minute, 55°C for 30 seconds, and 95°C for 30 seconds immediately after the quantification cycles.

Growth of R. bellii in tick and mammalian cell cultures

Cell cultures infected with R. bellii were extracted for DNA at 12, 24, 36, 48, and 72 HPI for ISE6 and additional times of 60, 84, and 96 HPI for Vero and L929. Rickettsiae numbers were quantified as previously mentioned. The values from triplicate cultures at each time point were averaged and graphed using SigmaPlot 13 (Systat Software, Inc., CA).

Selection, quantification, and validation of R. bellii reference gene transcripts

To identify stable reference genes to analyze the transcriptional responses of the tra genes in R. bellii grown for different times (12, 24, 36, 48, and 72 HPI) and in different host cells (48 HPI only), a two-step qRT-PCR was used employing sample maximization [28]. Reference genes (16S rRNA, atpB, dnaK, gltA, gyrA, infB, metG, nrdF, rpoB, and tlc5) were selected based on their predicted molecular function in different metabolic classes and their property as housekeeping genes (S1 Table). In addition, five out of the ten reference genes were selected to show transcription levels above background in Rickettsia rickettsii Sheila Smith based on tiling array data (S1 Text, S1 Fig, S2 Table).

First strand cDNA was synthesized from total RNA using random hexamer primers (Integrated DNA Technology, IA) and Superscript II Reverse Transcriptase (Life Technologies, NY) following the manufacturer’s protocol and stored at -20°C. All cDNA was diluted 1:20 with water and 5 μl of the dilution was added to each qPCR reaction. Master mix and cycling parameters were similar to the conditions used to determine gltA copy numbers, except the primer hybridization temperature for traV detection was lowered to 50°C based on the GC content of the primer set used. The transcript numbers were normalized using the ratio of gltA genomic copy numbers detected at all time points to 12 HPI.

Normfinder and BestKeeper (http://moma.dk/normfinder-software and http://www.gene-quantification.com/bestkeeper.html, respectively) were used to rank and identify the best reference gene or combination of reference genes to use for comparative analysis. Normfinder uses a model based approach to determine the most stable gene [20] and can handle high variations from low copy numbers of transcript from earlier time points because of lower number of rickettsiae and high copy numbers at later times. The ranking of our genes using BestKeeper was based on the highest coefficient of correlation for the best reference gene to the lowest coefficient of correlation for the least favorable reference gene [21]. BPROM, a promoter prediction program (http://linux1.softberry.com/berry.phtml?topic=bprom&group=programs&subgroup=gfindb), was used to identify potentially shared promoters upstream of the best reference genes identified by BestKeeper to discount similarly regulated genes for use as reference genes.

Transcription of rickA and sca2 in Vero cells to validate metG as a reference gene

Validation of metG as the best reference gene was accomplished using primers to determine the transcription pattern of rickA and sca2. The differences within and between the transcription patterns of rickA and sca2 were statistically analyzed using multiple comparisons with an ANOVA test and a Holm-Sidak correction with an alpha value set at 0.05 in SigmaPlot 13.

Relative transcription of traATi to reference genes

First strand cDNA synthesis was generated, as mentioned previously, from RNA of R. bellii isolated from infected ISE6, Vero, and L929 cells at all the time points. Two-step qRT-PCR was done using primers for traATi and for the reference genes metG and a combination of metG and nrdF. Template consisted of 1st strand cDNA synthesized at all time points in all cell lines. The 2−ΔΔCT method was used to quantify fold changes of traATi:metG at all time points relative to 12 HPI [29]. SigmaPlot 13 was used to generate line graphs.

Transcriptional analysis of the tra from 12 to 72 HPI

Transcriptional analysis of traATi, traDTi, traL, traE, traB, traV, traC, traW, trbC, traU, traF, traH, traG, traDF, and RBE_0422 was carried out under similar conditions and cycling parameters, as previously stated, using cDNA from 12 and 72 HPI of R. bellii grown in ISE6 as templates together with respective primer sets (S1 Table).

Results

Domains of R. bellii TraATi are related to conjugative relaxases of other bacterial species

The domains of R. bellii TraATi are depicted in Fig 1. The MobA/L, ABC ATPase, and UvrD C domains of R. bellii TraATi and R. buchneri TraI constitute the conjugal segment of TraA. This conjugal segment is predicted to have similar function as the conjugal segment of A. tumefaciens TraA and E. coli TraI (underlined in Fig 1) such as DNA binding, nicking, and unwinding [12,13,30] when generating an ssDNA for DNA transfer, which are all qualities of relaxases although the amino acid sequences may vary between the families of relaxases [31] In addition to the conjugal segment of TraATi of R. bellii, a primase was identified with a Toprim domain at the carboxyl terminal end. The overall domains of R. bellii TraATi are similar to the well characterized conjugal relaxases of other bacteria; thus, we suspect that R. bellii TraATi expression may be a good indicator of the time of induction for potential bacterial conjugation.

Growth of R. bellii in tick and mammalian cell cultures

To verify that there were no differences in rickettsial numbers determined from RP vs. WC, R. bellii DNA was prepared and the gltA genomic copy numbers in each preparation were compared (Table 1). According to the two-sided two sample Student’s t-test with alpha set to 0.05, no significant differences were observed between the two isolation procedures. Thus, the process of preparing cell-free rickettsial preparations did not cause a significant loss of rickettsiae when compared with whole cell lysate.

thumbnail
Download:
Table 1. A comparison of gltA copy numbers of RP and WC R. bellii.

https://doi.org/10.1371/journal.pone.0137214.t001

A two-sided two sample Student’s t-test with an alpha value at 0.05 was used to compare the number of detected gltA copy numbers in rickettsiae prepared (RP) by semi-purification and whole cell (WC) lysis of R. bellii RML 369-C grown in ISE6 at 34°C. Results show no significant difference between the two methods.

Subsequently, gltA copy numbers were quantified using the RP method and qPCR for all time points and host cells. The growth curves for R. bellii in different cell lines (Fig 2) shared similar growth phases and a doubling time of approximately 8 hours during the period of 36 to 60 HPI similar to times reported for Rickettsia prowazekii and Rickettsia rickettsii [32–34]. The initial higher copy numbers of gltA from R. bellii isolated from L929 could have resulted from inefficient washing or highly efficient phagocytosis and killing of rickettsiae, but slower digestion of rickettsial DNA by L929. We concluded that the replication of R. bellii was similar in the different host cells using an MOI of 10–50 followed by a wash step.

thumbnail
Download:
Fig 2. Growth curve of R. bellii in ISE6, Vero, and L929.

qPCR detected a single copy gene of gltA for triplicates of culture at each time point. Standard error bars are shown. Growth trials using Vero and L929 were done simultaneously while ISE6 was done separately.

https://doi.org/10.1371/journal.pone.0137214.g002

Identifying the best reference gene and best combination of genes

Ten R. bellii housekeeping genes were selected based on their different functional classes, i.e. translation, DNA replication: 16S rRNA, atpB, dnaA, gltA, gyrA, infB, metG, nrdF, rpoB, and tlc5 (S1 Table). Results for R. rickettsii microarray analysis showed transcription levels above background for each of the selected genes (S1 Fig). Normfinder and BestKeeper were used to analyze the transcript data of the R. bellii reference genes (Table 2). Normfinder ranked metG as the most stably transcribed gene throughout a 72 HPI growth period in ISE6 and at 48 HPI between the different host cell lines. Normfinder identified a combination of metG and nrdF for transcriptional analysis throughout a 72 HPI in ISE6, while gyrA and metG were selected by Normfinder when comparing between host cells at 48 HPI. In comparison, BestKeeper ranked gyrA and metG as the top two genes to use for a time point analysis and for comparing between host cells based on the coefficient of correlation (Table 2). A caveat of BestKeeper is that it can identify reference genes that are transcriptionally co-regulated, which is not ideal when identifying the best reference gene. To avoid co-regulated transcription, the selected genes were chosen based on their different molecular functions that belong to different metabolic classes, i.e. gyrA and metG functions in DNA replication and initiation of translation, respectively, suggesting that the genes tested are not co-regulated. Analysis of the genetic sequences upstream of metG and gyrA using BPROM identified a promoter upstream of gyrA at position 879453 with a linear discriminant function score of 9.77, and located a -10 box at 879468 with a score of 61, and a -35 box at 879492. Also, metG has a promoter at position 282683 with linear discriminant function score of 9.18, a -10 box at 282668 with score of 63, and a -35 box at 282649 with a score of 47. Both promoters were shown to have several transcription factor binding sites, but share only a leucine-responsive regulatory protein binding site. This transcription factor is involved in controlling the cell response to nutritional stress, particularly to leucine levels [35]. However, R. bellii does not encode a leucine-responsive regulatory protein-like protein according to the annotated R. bellii RML 369-C genome, indicating that metG and gyrA are not similarly regulated. Overall, metG is the best reference gene selected by Normfinder and BestKeeper either as the best or second best gene to use as a reference when comparing rickettsiae replicating in different host cells.

thumbnail
Download:
Table 2. The rankings of each candidate gene by Normfinder and BestKeeper.

https://doi.org/10.1371/journal.pone.0137214.t002

Transcription of rickA and sca2 is induced at different growth phases in Vero cells, initially validating metG as a reference gene

R. bellii uses actin-based motility similar to other intracellular bacteria [25]. Thus far, rickA and sca2 were found to be involved in actin tail polymerization [36,37], and their products, RickA and Sca2, were shown to be expressed during distinct phases of intracellular growth [22]. Therefore, we were interested in defining the transcriptional patterns of rickA and sca2 to validate our reference gene. Relative transcription of rickA and sca2 was determined as described using metG as the reference gene. Up-regulation of sca2 was observed from 12 to 24 HPI (Fig 3A), which correlated with the lag phase of the growth curve (Fig 2). rickA showed a trend of up-regulation between 48 to 96 HPI with significant differences in transcription levels at 72 and 96 HPI that coincided with the mid-log phase to the stationary phase. sca2 showed a significant difference in relative transcription level between 12 and 24 HPI that coincided with the lag phase and between 24 and 72 HPI that coincided with the beginning of the log phase and stationary phase, respectively (Fig 3A). The transcription patterns of sca2 and rickA were modulated by distinct phases of infection and growth. Analyses of rickA and sca2 transcription showed significant differences between each other at 24 HPI (Fig 3B). Furthermore, rickA and sca2 transcript levels displayed a trend similar to the growth curve (Figs 1 and 3B). Thus, sca2 may play an important role during the lag phase of invasion especially during the beginning of the log phase, while rickA may have an important role in the exponential phase of R. bellii replication.

thumbnail
Download:
Fig 3. Initial validation of reference genes.

rickA and sca2 of R. bellii grown in Vero cells were selected to demonstrate differential gene transcription using metG as a reference. (A) The line graph depicts the fold changes of rickA and sca2 relative to metG at all time points to 12 HPI. Letters indicate significant difference within each gene at different times (lower case = rickA, upper case = sca2) (B) The graph shows the transcription level of rickA and sca2 at each time point and (*) indicates significant difference. An ANOVA with Holm-Sidak correction was used to determine significant differences with alpha set at 0.05. Standard error bars are shown.

https://doi.org/10.1371/journal.pone.0137214.g003

Transcription of traATi in R. bellii is induced in ISE6 at 72 HPI

Two-step qRT-PCR was used to quantify traATi mRNA through time in different host cells using metG and a combination of metG and nrdF as reference genes. At 24, 36, and 48 HPI traATi in R. bellii grown in ISE6 did not show any apparent differences in transcription when compared to 12 HPI using metG only, nrdF only, or metG and nrdF as reference genes, but at 72 HPI traATi was up-regulated compared to 12 HPI (Fig 4). In addition, traATi was also up-regulated in R. bellii grown in ISE6 in comparison to R. bellii grown in Vero and L929 at 72 HPI (Fig 4). As a result of the induction of traATi observed at 72 HPI, R. bellii was grown in ISE6 for 12 and 72 HPI and examined for the transcriptional patterns of the other annotated tra genes. All examined tra genes were found to be up-regulated at 72 HPI (Table 3) as indicated by the lower CT values where the threshold is reached with lower number of cycles when starting with a greater quantity of template.

thumbnail
Download:
Fig 4. Fold change of traATi in different host cells at different hours post infection.

qRT-PCR was used to determine the traATi transcription pattern using metG only, nrdF only, and a combination of metG and nrdF. Standard error bars are shown.

https://doi.org/10.1371/journal.pone.0137214.g004

thumbnail
Download:
Table 3. CT values of tra genes and metG at 12 and 72 HPI.

https://doi.org/10.1371/journal.pone.0137214.t003

Discussion

Several studies have analyzed relative transcription patterns in rickettsiae, e.g. using 16S rDNA, metG, and gltA [25,38,39]. However, the reference genes chosen were not rigorously evaluated and validated as the most stably transcribed genes under experimental conditions using statistical programs. In this study, we evaluated 10 different genes and identified and validated metG or metG and nrdF as the best reference genes to use for transcriptional analysis of the tra genes during R. bellii growth in ISE6 cell cultures and in different mammalian cell lines using NormFinder. BestKeeper identified gyrA and metG as the top two reference genes. Based on the promoters (predicted by BPROM) and the differences in molecular functions between metG and gyrA, we surmised that they were unlikely to be transcriptionally co-regulated. metG and nrdF were also shown to be constitutively transcribed in R. bellii (S2 Fig). Tiling array data from R. rickettsii also showed that transcription levels of metG, nrdF, gyrA, gltA, and 16S rDNA were above background (S1 Fig). Both results corroborated the stability of the transcription of metG, nrdF, and gyrA indicating that all three may be good reference genes. However, metG consistently ranked as the best reference gene to be used for relative transcriptional analysis. metG was also used to determine the transcriptional response of R. rickettsii to a shift in temperature suggesting that both pathogenic and nonpathogenic rickettsiae may transcribe metG at stable levels under different treatment conditions.

We further validated our reference genes by tracking transcriptional patterns of sca2 and rickA. We observed a differential transcription pattern of sca2 and rickA, with a distinctive up-regulation of sca2 during the lag phase of rickettsial growth and a trend of increasing variance of the transcription of rickA during exponential phase. Sca2 up-regulation at 24 HPI suggested that it played an important role when R. bellii was invading and adapting to the host cells until it was ready for the exponential phase (Fig 3A). In contrast, rickA demonstrated high variability of transcription levels during the mid-exponential growth phase to the stationary phase suggesting discontinuous and variable motility of R. bellii (Fig 3A). This variability of rickA transcript may be attributed to the first asynchronous infection of neighboring cells near 48 HPI when rickettsiae begin to spread and infect new cells [33,40]. Both rickA and sca2 had general transcription patterns that trended with the growth curve of R. bellii except at the beginning of the exponential phase (24 HPI) where sca2 transcription was significantly different from rickA (Fig 3B). Sca2 is found at the surface [41] and has been shown to be involved in adhesion to mammalian cells [42]. Thus, sca2 is probably transcribed significantly more than rickA at the initial part of the exponential phase to express enough Sca2 to ensure that they can continue to adhere to and invade host cells, as well as to enable them to use actin-base motility for later stages of cell-to-cell spread as indicated by Reed et al. 2014 [22]. Expression patterns of RickA and Sca2 in Rickettsia parkeri were significantly different at early and late infections correlating with actin-tail morphology and were observed along with rickA or sca2 knockouts to determine the differences in cell-to-cell spread [21]. In contrast, our results indicated that rickA and sca2 were differentially transcribed at different times post infection [22] to highlighting the importance of rickA and sca2 at different growth phases of R. bellii. Furthermore, Sca2 of R. bellii lacks the formin homology 2 domain, but maintains the formin homology 1 domain, and has been shown to sufficiently accelerate actin polymerization indicating it functions as a tandem monomer-binding nucleator [43]. These differences may result in different regulation of expression and function of sca2 and indirectly affecting rickA. Nonetheless, we demonstrated that a significant difference in the transcription levels of rickA and sca2 of R. bellii existed during the transition of lag to exponential growth phase. Thus, this is the initial validation of metG as a suitable reference gene.

TraATi was chosen as the indicator of the conditions to examine transcription of the other tra genes because it was shown to reflect responses of the entire tra cluster. TraATi has domains that suggest it is functionally similar to TraI (Fig 1), indicative of its role in transfer DNA synthesis [12,13]. Specifically, TraATi of R. bellii has a MobA/L domain that probably has nickase activity similarly to the TrwC domain in E. coli, and the UrvD domain is shared by both R. bellii and E. coli. The main difference is that R. bellii TraATi has a primase domain not present in TraI of E. coli (Fig 1). A possible role of primase in R. bellii TraATi may be the initiation of DNA replication immediately following the generation of transfer ssDNA in the donor, or initiation of replication of the transfer DNA in the recipient. Thus far, the primase domain is unique to the relaxases of rickettsiae and insight into the function of this domain may identify a different mechanism in conjugation.

The conditions that stimulated high transcriptional activity of traATi using metG as the reference gene were met when R. bellii was growing in ISE6 at 34°C for 72 hours (Fig 4), near the beginning of the stationary phase (Fig 1). In contrast, R. bellii did not show any changes in traATi transcriptional activity when grown in mammalian cells. Possibly, cellular and metabolic processes of mammalian cells inhibited or indirectly altered the transcription of traATi. In nature, R. bellii has been observed to primarily infect arthropods, particularly ticks, so the increase in relative transcription found when grown in a tick cell line in the early stationary phase (Fig 2, 72 HPI) suggested the importance of the tick host in possible bacterial conjugation. Timing of transcription during the early stationary phase may prepare the bacteria for conjugal activity during this phase. This is in contrast to E. coli in which mating efficiency is maximal during rapid replication (mid-log phase) when tra transcripts are still detected although Tra proteins that are components of the transfer structure remain stable into stationary phase when TraM and TraI are lost [18].

The transcription of the other annotated tra genes analyzed was positively correlated with traATi. Although traDF showed lower transcription activity overall, it still correlated positively with the other tra genes (Table 3), while transcription of RBE_0422 showed no correlation with the other tra genes despite the prediction that RBE_0422 and traATi were part of an operon (S3 Table, [44–46]). All the tra genes were simultaneously up-regulated at early stationary phase or 72 HPI, further validating metG as the best reference gene (Table 3). traF mRNA was not detectable during the lag phase or at 12 HPI, but was present at the early stationary phase indicating that traF may not be important during the lag phase, but plays a role near the early stationary phase. Also, this shows that not all tra genes were transcribed together and may have different transcriptional regulation. A mutation in traF in E. coli eliminated formation of the pilus but did not affect replication, suggesting that pilus formation is not important for growth in E. coli [47], and this may also be true for R. bellii traF. Rickettsia bellii traG showed highest relative transcription near the early stationary phase indicating the importance of traG at this time and life stage. In E. coli, traG is multifunctional and participates in pilus assembly, mating pair stabilization, and entry exclusion, which further highlights this time point and stage of growth as a potential period to target for testing conjugation in R. bellii [11,48–50]. Taken together, our results suggest that testing for conjugation in R. bellii should be done in tick cell culture at later stages of infection.

Determining if the tra genes are functional as operons may be useful for understanding the regulation of these genes. There was evidence indicating that traB, -V, and -C may form an operon as suggested by overlapping of the start and stop codon of adjacent genes and similar degree of transcriptional up-regulation. Further evidence from the Database for prOkaryotic OperRons support traB, -V, and -C as an operon that also includes traL and-E (S3 Table). The Database for prOkaryotic OpeRons identified another tra operon that included traU, trbC, traN, -F, and -H (S3 Table) and genetic evidence show overlapping start and stop codons from trbC, traN, -F, and -H supporting these tra genes as belonging to an operon. The CT values and fold increase of traU, trbC, traN, -F, and -H differed from each other despite evidence of an operon. However, differential and “segregational” decay of polycistronic transcripts are common and may explain the transcriptional pattern observed. The stability of transcripts can be affected by the secondary structure of RNA and/or the frequency of translation [51,52] highlighting the importance of proteomic analysis to support the regulation of transcription.

We considered RBE_0435 of R. bellii RML 369-C (S3 Fig) to be a possible candidate for the putative gene encoding the propilin (or pilin) subunits of the pilus. Searching for this protein homolog is difficult considering this coding gene is poorly conserved [53] although key characters such as a signal peptide and two transmembrane-spanning domain can help to identify potential homologs as seen with the homologous protein VirB2 [54]. Interestingly, the proteins involved in directing pilin to the periplasm (TraQ in E. coli) and the enzyme responsible for N-acetylation of pilin (TraX in E. coli) have not been found in R. bellii (S3 Table). This suggests that this potential pilin protein either does not require the modification and/or that it may not be involved in forming pili. Alternatively, the pili-like structures that have been observed in R. bellii [4] could originate from a different system, i.e. the VirB/D system. Contrarily, RBE_0435 locus has been illustrated to be the carboxyl-terminal end of traK [7]. The pili aspect of the tra system remains elusive.

If rickettsiae can engage in lateral gene transfer, then this could contribute new genes to the rickettsial genome, buffer the rate of reductive genome evolution [5] and enhance rickettsial fitness in new environments. The tra genes can facilitate import of new genes through recombination and insertion at a particular hot spot between traATi and traDF, as suggested by Blanc et al. [10]. Evidence of this phenomenon is illustrated by the tra genes of R. buchneri where genes typical of Gram-positive aminoglycoside antibiotic biosynthesis gene clusters were found in a similar hot spot suggesting that the tra cluster can support piggybacking genes into the rickettsial genome to increase fitness and slow down reductive genome evolution [7]. Weinert et al. (2009) have pointed out that there is no evidence for the preferential transfer of genes between closely related species of rickettsiae and speculate that instead they may engage in “regenerative horizontal” DNA exchange further combatting a reductive genome evolution. This could be achieved by recombination of foreign DNA into the rickettsial genome, but recombination events have been argued to be generally rare in Rickettsia [55], although they seem more common in some species, e.g. R. bellii [56]. The possibility that the tra gene cluster can mediate acquisition of new genes and act as a hot spot for insertion to assist in the repair of damaged DNA in R. bellii through natural transformation is intriguing.

There is, however, ample evidence that some rickettsiae, especially those symbiotically associated with arthropods, harbor horizontally acquired genetic material on their plasmids that may have been acquired via bacterial conjugation. For example, Rickettsia peacockii maintains five genes in the pRPR plasmid that are distantly related to the Pseudomonas aeruginosa glycosylation island and may function in the uptake of dihydroxyacetone phosphate for phospholipid biosynthesis [57]. Furthermore, R. buchneri encodes a complete biotin operon on pREIS2 that shares homology with biotin operons recently acquired by horizontal gene transfer in Neorickettsia risticii, N. sennetsu and Lawsonia intracellularis [7]. Recently, an RTX type I secretion system (TISS) was found in R. felis plasmid pLSU-lb that was closely related to the TISS genes of Cardinium endosymbiont cBtQ1 of Bemisia tabaci, in which the RTX TISS genes were thought to originate through lateral gene transfer from Vibrio species [6]. In addition, other R. bellii isolates contain plasmids that have yet to be sequenced [1], and that may contribute to the dissemination of operons that mediate symbiosis. There are other mobile genetic elements such as the integrated genetic element found on the R. buchneri chromosome that was supported by bioinformatics to be closely related to and include the conjugation genes of R. bellii [7]. The acquisition of the tra genes may have been via a plasmid based on the recent discovery that R. felis LSU-lb harbors a plasmid containing the tra cluster closely related to pREIS3 of R. buchneri. The mechanism of transfer of an integrative conjugative element requires a functional integrase or recombinase that will excise the sequence from the chromosome followed by transfer strand synthesis and transport into the recipient either by encoding its own conjugation system or parasitizing another present in the donor [58,59]. Other strains of R. bellii have been documented to have a native plasmid [1] that could act as a vehicle for the spread of the tra genes, and it is conceivable that R. bellii RML 369-C may have lost its native plasmid through long term passage in cell culture [60]. Therefore, horizontally acquired functional genes and operons in rickettsiae may have been acquired in more than one way, and while having a native plasmid possibly increases the frequency at which genetic material is spread, it is likely not the only mechanism used. Thus far, the tra genes have not been demonstrated to function in bacterial conjugation and may be remnants of a once functional system; however, up-regulation of the tra genes in this research implies a potentially, functional conjugative and gene transfer system.

Conclusion

We have demonstrated that the best reference gene out of the ten selected were metG or metG and nrdF. The same reference gene(s) can be used to standardize future relative transcriptional analysis of rickettsiae with similar treatments or to form a basis of potential reference genes with different treatments. Using the identified reference gene(s), we show that the transcription of the tra genes of R. bellii is highly upregulated at 72 HPI relative to 12 HPI only in tick cell culture, indicating that these genes are very active and contribute to the biology of R. bellii. Future research should explore the functions of the tra genes guided by bioinformatics to obtain detailed documentation of the mechanism of DNA mobility to and from the chromosome and plasmid, and the demonstration of the process of rickettsial conjugation.

Supporting Information

S1 Fig. Screenshots of Artemis of Rickettsia rickettsii Sheila Smith showing transcription levels of 5 genes: metG, nrdF, gyrA, gltA, and 16S rRNA.

The order from left to right is of metG (A1G_05810), nrdF (A1G_03685), gyrA (A1G_01555), gltA (A1G_07170), and 16s rRNA gene (A1G_r07597). Each vertical black bar represents the hybridization level of mRNA from one probe on the array. The orientation and length of the genes are shown below the transcription peaks.

https://doi.org/10.1371/journal.pone.0137214.s001

(TIF)

S2 Fig. The relative transcription pattern of metG and nrdF to traATi.

Relative transcription of metG and nrdF was analyzed using traATi as a reference gene to show constitutive, but slightly varying transcription of both genes. Standard error bars are shown.

https://doi.org/10.1371/journal.pone.0137214.s002

(TIF)

S3 Fig. Amino acid alignment of TraA pilin.

Sequences include TraA of Escherichia coli K-12 (CAA73225.1), TraA of Enterobacter cloacae (AKN35276.1), TraA of Klebsiella pneumoniae (CDO11547.1), TraA of Salmonella enterica subsp. enterica serovar Typhimurium (YP 009077414.1), RBE_0435 of Rickettsia bellii RML 369-C (WP 008579911.1), and RMA_0719 of Rickettsia massiliae MTU5 (ABV84869.1). Vertical lines indicate cleavage site of the signal peptide predicted by Phobius except for R. bellii and R. massiliae, which was predicted using Signal-BLAST. Underlined sequences show the transmembrane spanning membrane domain predicted using Phobius. “.” = same amino acid above it and “-” = space.

https://doi.org/10.1371/journal.pone.0137214.s003

(TIF)

S1 Table. Information about the genes investigated and the primer pairs.

https://doi.org/10.1371/journal.pone.0137214.s004

(XLSX)

S2 Table. "Area under the curve" from RrSS tiling microarray for 5 genes selected.

https://doi.org/10.1371/journal.pone.0137214.s005

(XLSX)

S3 Table. A comparison of predicted tra operons in E. coli, R. bellii, and R. massiliae.

https://doi.org/10.1371/journal.pone.0137214.s006

(XLSX)

S1 Text. Supporting information methods.

https://doi.org/10.1371/journal.pone.0137214.s007

(DOCX)

Acknowledgments

We would like to thank Nicole Y. Burkhardt for contributing to our work including plasmids and advice.

Author Contributions

Conceived and designed the experiments: CCH. Performed the experiments: CCH CMN. Analyzed the data: CCH CMN TJK UGM. Contributed reagents/materials/analysis tools: CCH CMN TJK UGM. Wrote the paper: CCH CMN TJK UGM.

References

  1. 1. Baldridge GD, Burkhardt NY, Labruna MB, Pacheco RC, Paddock CD, Williamson PC, et al. Wide dispersal and possible multiple origins of low-copy-number plasmids in rickettsia species associated with blood-feeding arthropods. Appl Environ Microbiol. 2010;76: 1718–1731. pmid:20097813
  2. 2. Ogata H, Renesto P, Audic S, Robert C, Blanc G, Fournier PE, et al. The genome sequence of Rickettsia felis identifies the first putative conjugative plasmid in an obligate intracellular parasite. PLoS Biol. 2005;3: e248. 05-PLBI-RA-0181R2 [pii]. pmid:15984913
  3. 3. Weinert LA, Welch JJ, Jiggins FM. Conjugation genes are common throughout the genus Rickettsia and are transmitted horizontally. Proc Biol Sci. 2009;276: 3619–3627. pmid:19608593
  4. 4. Ogata H, La Scola B, Audic S, Renesto P, Blanc G, Robert C, et al. Genome sequence of Rickettsia bellii illuminates the role of amoebae in gene exchanges between intracellular pathogens. PLoS Genet. 2006;2: e76. pmid:16703114
  5. 5. Blanc G, Ogata H, Robert C, Audic S, Suhre K, Vestris G, et al. Reductive genome evolution from the mother of Rickettsia. PLoS Genet. 2007;3: e14. 06-PLGE-RA-0186R4 [pii]. pmid:17238289
  6. 6. Gillespie JJ, Driscoll TP, Verhoeve VI, Utsuki T, Husseneder C, Chouljenko VN, et al. Genomic diversification in strains of Rickettsia felis Isolated from different arthropods. Genome Biol Evol. 2014;7: 35–56. pmid:25477419
  7. 7. Gillespie JJ, Joardar V, Williams KP, Driscoll T, Hostetler JB, Nordberg E, et al. A Rickettsia genome overrun by mobile genetic elements provides insight into the acquisition of genes characteristic of an obligate intracellular lifestyle. J Bacteriol. 2012;194: 376–394. pmid:22056929
  8. 8. Kurtti TJ, Felsheim RF, Burkhardt NY, Oliver JD, Heu CC, Munderloh UG. Rickettsia buchneri sp. nov.: A rickettsial endosymbiont of the blacklegged tick Ixodes scapularis. Int J Syst Evol Microbiol. 2015. ijs.0.000047 [pii].
  9. 9. Ogata H, Robert C, Audic S, Robineau S, Blanc G, Fournier PE, et al. Rickettsia felis, from culture to genome sequencing. Ann N Y Acad Sci. 2005;1063: 26–34. 1063/1/26 [pii]. pmid:16481487
  10. 10. Blanc G, Ogata H, Robert C, Audic S, Claverie JM, Raoult D. Lateral gene transfer between obligate intracellular bacteria: evidence from the Rickettsia massiliae genome. Genome Res. 2007;17: 1657–1664. gr.6742107 [pii]. pmid:17916642
  11. 11. Firth N, Skurray R. Characterization of the F plasmid bifunctional conjugation gene, traG. Mol Gen Genet. 1992;232: 145–153. pmid:1348105
  12. 12. Matson SW, Sampson JK, Byrd DR. F plasmid conjugative DNA transfer: the TraI helicase activity is essential for DNA strand transfer. J Biol Chem. 2001;276: 2372–2379. pmid:11054423
  13. 13. Matson SW, Ragonese H. The F-plasmid TraI protein contains three functional domains required for conjugative DNA strand transfer. J Bacteriol. 2005;187: 697–706. 187/2/697 [pii]. pmid:15629940
  14. 14. Howard MT, Nelson WC, Matson SW. Stepwise assembly of a relaxosome at the F plasmid origin of transfer. J Biol Chem. 1995;270: 28381–28386. pmid:7499340
  15. 15. Nelson WC, Howard MT, Sherman JA, Matson SW. The traY gene product and integration host factor stimulate Escherichia coli DNA helicase I-catalyzed nicking at the F plasmid oriT. J Biol Chem. 1995;270: 28374–28380. pmid:7499339
  16. 16. Ragonese H, Haisch D, Villareal E, Choi JH, Matson SW. The F plasmid-encoded TraM protein stimulates relaxosome-mediated cleavage at oriT through an interaction with TraI. Mol Microbiol. 2007;63: 1173–1184. MMI5576 [pii]. pmid:17238924
  17. 17. Lang S, Kirchberger PC, Gruber CJ, Redzej A, Raffl S, Zellnig G, et al. An activation domain of plasmid R1 TraI protein delineates stages of gene transfer initiation. Mol Microbiol. 2011;82: 1071–1085. pmid:22066957
  18. 18. Frost LS, Manchak J. F- phenocopies: characterization of expression of the F transfer region in stationary phase. Microbiology. 1998;144: 2579–2587. pmid:9782507
  19. 19. Francia MV, Varsaki A, Garcillan-Barcia MP, Latorre A, Drainas C, de la Cruz F. A classification scheme for mobilization regions of bacterial plasmids. FEMS Microbiol Rev. 2004;28: 79–100. pmid:14975531
  20. 20. Andersen CL, Jensen JL, Orntoft TF. Normalization of real-time quantitative reverse transcription-PCR data: a model-based variance estimation approach to identify genes suited for normalization, applied to bladder and colon cancer data sets. Cancer Res. 2004;64: 5245–5250. pmid:15289330
  21. 21. Pfaffl MW, Tichopad A, Prgomet C, Neuvians TP. Determination of stable housekeeping genes, differentially regulated target genes and sample integrity: BestKeeper—Excel-based tool using pair-wise correlations. Biotechnol Lett. 2004;26: 509–515. pmid:15127793
  22. 22. Reed SC, Lamason RL, Risca VI, Abernathy E, Welch MD. Rickettsia actin-based motility occurs in distinct phases mediated by different actin nucleators. Curr Biol. 2014;24: 98–103. pmid:24361066
  23. 23. Marchler-Bauer A, Bryant SH. CD-Search: protein domain annotations on the fly. Nucleic Acids Res. 2004;32: W327–31. pmid:15215404
  24. 24. Munderloh UG, Jauron SD, Fingerle V, Leitritz L, Hayes SF, Hautman JM, et al. Invasion and intracellular development of the human granulocytic ehrlichiosis agent in tick cell culture. J Clin Microbiol. 1999;37: 2518–2524. pmid:10405394
  25. 25. Oliver JD, Burkhardt NY, Felsheim RF, Kurtti TJ, Munderloh UG. Motility characteristics are altered for Rickettsia bellii transformed to overexpress a heterologous rickA. Applied and Environmental Microbiology. 2013.
  26. 26. Munderloh UG, Kurtti TJ. Formulation of medium for tick cell culture. Exp Appl Acarol. 1989;7: 219–229. pmid:2766897
  27. 27. Eremeeva ME, Dasch GA, Silverman DJ. Evaluation of a PCR assay for quantitation of Rickettsia rickettsii and closely related spotted fever group rickettsiae. J Clin Microbiol. 2003;41: 5466–5472. pmid:14662926
  28. 28. Hellemans J, Mortier G, De Paepe A, Speleman F, Vandesompele J. qBase relative quantification framework and software for management and automated analysis of real-time quantitative PCR data. Genome Biol. 2007;8: R19. gb-2007-8-2-r19 [pii]. pmid:17291332
  29. 29. Livak KJ, Schmittgen TD. Analysis of Relative Gene Expression Data Using Real-Time Quantitative PCR and the 2−ΔΔCT Method. Methods. 2001;25: 402–408. http://dx.doi.org/10.1006/meth.2001.1262. pmid:11846609
  30. 30. Alt-Morbe J, Stryker JL, Fuqua C, Li PL, Farrand SK, Winans SC. The conjugal transfer system of Agrobacterium tumefaciens octopine-type Ti plasmids is closely related to the transfer system of an IncP plasmid and distantly related to Ti plasmid vir genes. J Bacteriol. 1996;178: 4248–4257. pmid:8763954
  31. 31. Garcillan-Barcia MP, Francia MV, de la Cruz F. The diversity of conjugative relaxases and its application in plasmid classification. FEMS Microbiol Rev. 2009;33: 657–687. pmid:19396961
  32. 32. Hopps HE, Jackson EB, Danauskas JX, Smadel JE. Study on the growth of rickettsiae. III. Influence of extracellular environment on the growth of Rickettsia tsutsugamushi in tissue culture cells. J Immunol. 1959;82: 161–171. pmid:13631246
  33. 33. Wisseman CL Jr, Waddell AD. In vitro studies on rickettsia-host cell interactions: intracellular growth cycle of virulent and attenuated Rickettsia prowazeki in chicken embryo cells in slide chamber cultures. Infect Immun. 1975;11: 1391–1404. pmid:806528
  34. 34. Wisseman CL Jr, Edlinger EA, Waddell AD, Jones MR. Infection cycle of Rickettsia rickettsii in chicken embryo and L-929 cells in culture. Infect Immun. 1976;14: 1052–1064. pmid:825463
  35. 35. Cho BK, Barrett CL, Knight EM, Park YS, Palsson BO. Genome-scale reconstruction of the Lrp regulatory network in Escherichia coli. Proc Natl Acad Sci U S A. 2008;105: 19462–19467. pmid:19052235
  36. 36. Haglund CM, Choe JE, Skau CT, Kovar DR, Welch MD. Rickettsia Sca2 is a bacterial formin-like mediator of actin-based motility. Nat Cell Biol. 2010;12: 1057–1063. pmid:20972427
  37. 37. Gouin E, Egile C, Dehoux P, Villiers V, Adams J, Gertler F, et al. The RickA protein of Rickettsia conorii activates the Arp2/3 complex. Nature. 2004;427: 457–461. pmid:14749835
  38. 38. Gaywee J, Radulovic S, Higgins JA, Azad AF. Transcriptional analysis of Rickettsia prowazekii invasion gene homolog (invA) during host cell infection. Infect Immun. 2002;70: 6346–6354. pmid:12379714
  39. 39. Galletti MF, Fujita A, Nishiyama MY Jr, Malossi CD, Pinter A, Soares JF, et al. Natural blood feeding and temperature shift modulate the global transcriptional profile of Rickettsia rickettsii infecting its tick vector. PLoS One. 2013;8: e77388. pmid:24155949
  40. 40. Silverman DJ, Wisseman CL Jr, Waddell A. In vitro studies of Rickettsia-host cell interactions: ultrastructural study of Rickettsia prowazekii-infected chicken embryo fibroblasts. Infect Immun. 1980;29: 778–790. pmid:6783534
  41. 41. Sears KT, Ceraul SM, Gillespie JJ, Allen ED Jr, Popov VL, Ammerman NC, et al. Surface proteome analysis and characterization of surface cell antigen (Sca) or autotransporter family of Rickettsia typhi. PLoS Pathog. 2012;8: e1002856. pmid:22912578
  42. 42. Cardwell MM, Martinez JJ. The Sca2 autotransporter protein from Rickettsia conorii is sufficient to mediate adherence to and invasion of cultured mammalian cells. Infect Immun. 2009;77: 5272–5280. pmid:19805531
  43. 43. Haglund CM. Rickettsia Sca2 is a bacterial formin-like mediator of actin-based motility. University of California Berkeley. 2011.
    • 44. Mao X, Ma Q, Zhou C, Chen X, Zhang H, Yang J, et al. DOOR 2.0: presenting operons and their functions through dynamic and integrated views. Nucleic Acids Res. 2014;42: D654–9. pmid:24214966
    • 45. Mao F, Dam P, Chou J, Olman V, Xu Y. DOOR: a database for prokaryotic operons. Nucleic Acids Res. 2009;37: D459–63. pmid:18988623
    • 46. Dam P, Olman V, Harris K, Su Z, Xu Y. Operon prediction using both genome-specific and general genomic information. Nucleic Acids Res. 2007;35: 288–298. gkl1018 [pii]. pmid:17170009
    • 47. Anthony KG, Klimke WA, Manchak J, Frost LS. Comparison of proteins involved in pilus synthesis and mating pair stabilization from the related plasmids F and R100-1: insights into the mechanism of conjugation. J Bacteriol. 1999;181: 5149–5159. pmid:10464182
    • 48. Audette GF, Manchak J, Beatty P, Klimke WA, Frost LS. Entry exclusion in F-like plasmids requires intact TraG in the donor that recognizes its cognate TraS in the recipient. Microbiology. 2007;153: 442–451. 153/2/442 [pii]. pmid:17259615
    • 49. Manning PA, Morelli G, Achtman M. traG protein of the F sex factor of Escherichia coli K-12 and its role in conjugation. Proc Natl Acad Sci U S A. 1981;78: 7487–7491. pmid:7038680
    • 50. Marrero J, Waldor MK. Interactions between inner membrane proteins in donor and recipient cells limit conjugal DNA transfer. Dev Cell. 2005;8: 963–970. S1534-5807(05)00173-5 [pii]. pmid:15935784
    • 51. Koraimann G, Teferle K, Mitteregger R, Wagner S, Hogenauer G. Differential mRNA decay within the transfer operon of plasmid R1: identification and analysis of an intracistronic mRNA stabilizer. Mol Gen Genet. 1996;250: 466–476. pmid:8602164
    • 52. Kristoffersen SM, Haase C, Weil MR, Passalacqua KD, Niazi F, Hutchison SK, et al. Global mRNA decay analysis at single nucleotide resolution reveals segmental and positional degradation patterns in a Gram-positive bacterium. Genome Biol. 2012;13: R30. pmid:22537947
    • 53. Lawley TD, Klimke WA, Gubbins MJ, Frost LS. F factor conjugation is a true type IV secretion system. FEMS Microbiol Lett. 2003;224: 1–15. S0378109703004300 [pii]. pmid:12855161
    • 54. Gillespie JJ, Ammerman NC, Dreher-Lesnick SM, Rahman MS, Worley MJ, Setubal JC, et al. An anomalous type IV secretion system in Rickettsia is evolutionarily conserved. PLoS One. 2009;4: e4833. pmid:19279686
    • 55. Weinert LA, Werren JH, Aebi A, Stone GN, Jiggins FM. Evolution and diversity of Rickettsia bacteria. BMC Biol. 2009;7: 6. pmid:19187530
    • 56. Hernandez-Lopez A, Chabrol O, Royer-Carenzi M, Merhej V, Pontarotti P, Raoult D. To tree or not to tree? Genome-wide quantification of recombination and reticulate evolution during the diversification of strict intracellular bacteria. Genome Biol Evol. 2013;5: 2305–2317. pmid:24259310
    • 57. Felsheim RF, Kurtti TJ, Munderloh UG. Genome sequence of the endosymbiont Rickettsia peacockii and comparison with virulent Rickettsia rickettsii: identification of virulence factors. PLoS One. 2009;4: e8361. pmid:20027221
    • 58. Wozniak RA, Waldor MK. Integrative and conjugative elements: mosaic mobile genetic elements enabling dynamic lateral gene flow. Nat Rev Microbiol. 2010;8: 552–563. pmid:20601965
    • 59. Bellanger X, Payot S, Leblond-Bourget N, Guedon G. Conjugative and mobilizable genomic islands in bacteria: evolution and diversity. FEMS Microbiol Rev. 2014;38: 720–760. pmid:24372381
    • 60. Baldridge GD, Burkhardt NY, Felsheim RF, Kurtti TJ, Munderloh UG. Plasmids of the pRM/pRF family occur in diverse Rickettsia species. Appl Environ Microbiol. 2008;74: 645–652. pmid:18065613