Skip to main content
NCBI home page
Infection and Immunity logoLink to Infection and Immunity
. 2016 Feb 24;84(3):790–797. doi: 10.1128/IAI.01205-15

Nonselective Persistence of a Rickettsia conorii Extrachromosomal Plasmid during Mammalian Infection

Sean P Riley a,b, Abigail I Fish a,b, Daniel A Garza a,b, Kaikhushroo H Banajee a,b, Emma K Harris a,b, Fabio del Piero b, Juan J Martinez a,b,✉
Editor: C R Roy
PMCID: PMC4771360  PMID: 26755154

Abstract

Scientific analysis of the genus Rickettsia is undergoing a rapid period of change with the emergence of viable genetic tools. The development of these tools for the mutagenesis of pathogenic bacteria will permit forward genetic analysis of Rickettsia pathogenesis. Despite these advances, uncertainty still remains regarding the use of plasmids to study these bacteria in in vivo mammalian models of infection, namely, the potential for virulence changes associated with the presence of extrachromosomal DNA and nonselective persistence of plasmids in mammalian models of infection. Here, we describe the transformation of Rickettsia conorii Malish 7 with the plasmid pRam18dRGA[AmTrCh]. Transformed R. conorii stably maintains this plasmid in infected cell cultures, expresses the encoded fluorescent proteins, and exhibits growth kinetics in cell culture similar to those of nontransformed R. conorii. Using a well-established murine model of fatal Mediterranean spotted fever, we demonstrate that R. conorii(pRam18dRGA[AmTrCh]) elicits the same fatal outcomes in animals as its untransformed counterpart and, importantly, maintains the plasmid throughout infection in the absence of selective antibiotic pressure. Interestingly, plasmid-transformed R. conorii was readily observed both in endothelial cells and within circulating leukocytes. Together, our data demonstrate that the presence of an extrachromosomal DNA element in a pathogenic rickettsial species does not affect either in vitro proliferation or in vivo infectivity in models of disease and that plasmids such as pRam18dRGA[AmTrCh] are valuable tools for the further genetic manipulation of pathogenic rickettsiae.

INTRODUCTION

Rickettsia conorii is a pathogenic member of the spotted fever group (SFG) of the genus Rickettsia. Multiple tick-borne SFG Rickettsia species can cause a wide range of human and zoonotic diseases, with untreated cases of Rocky Mountain spotted fever (R. rickettsii) disease resulting in 20 to 25% mortality rates (1–3). A similar disease in Eurasia, Mediterranean spotted fever (MSF), is caused by R. conorii and results in 2 to 32% mortality rates in reported human cases (4). Symptoms of these diseases are manifested 2 to 14 days following arthropod transmission and are the result of widespread bacterial infection of the endothelium with concurrent inflammation (5, 6). While broad-spectrum antibiotic treatment can significantly reduce morbidity and mortality, untreated infections can proceed to more severe disease manifestations, including pulmonary edema, interstitial pneumonia, and multiorgan failure (7).

Historically, analysis of the genetic determinants of SFG Rickettsia disease has been hampered by a lack of viable tools for the genetic manipulation of these bacteria. This genetic recalcitrance was largely a consequence of a lack of both an axenic growth medium and selectable markers. However, the development of tools for chromosomal mutagenesis and selectively maintained extrachromosomal DNA in rickettsial species has been demonstrated (8). Plasmids isolated from nonpathogenic Rickettsia bacteria have been adapted for replication and selection in both Rickettsia and Escherichia species (9). These plasmids can potentially be useful for heterologous gene expression, reporter gene expression, and in trans complementation in diverse Rickettsia species.

One such plasmid, pRam18dRGA[AmTrCh], is derived from the 18-kbp plasmid from “Candidatus Rickettsia amblyommii” (10, 11). The nonessential portion of the naturally occurring plasmid was removed, and the remaining portion was inserted into an Escherichia coli vector. The gfpuv and rpaar2 genes and a multiple cloning site (MCS) were introduced to produce the plasmid pRam18dRGA[MCS] (9, 12, 13). The 948-bp AmTrCh cassette encoding mCherry was subsequently added to create pRam18dRGA[AmTrCh] (9, 14). This plasmid has been successfully transformed into five pathogenic and nonpathogenic Rickettsia species (9, 15).

Here, we utilized this plasmid to transform pathogenic R. conorii to examine the in vivo persistence of an extrachromosomal DNA element in the absence of antibiotic selection and to determine whether the presence of the plasmid affects the virulence of transformed R. conorii Malish 7 in a murine model of MSF. These analyses subsequently yielded the identification of new host cell types that are parasitized by R. conorii in vivo.

MATERIALS AND METHODS

Bacterial transformation.

R. conorii strain Malish 7 was electroporated with pRam18dRGA[AmTrCh] in a manner similar to that of Rachek et al. (16). Briefly, approximately 4 × 108 R. conorii bacteria were purified by sucrose density centrifugation (17). We made the bacteria electrocompetent by washing them four times with ice-cold 250 mM sucrose. A 10-μl volume of 1 mg/ml pRam18dRGA[AmTrCh] was added to 4 × 107 electrocompetent R. conorii cells in an electroporation cuvette with a 0.1-cm gap. After 30 min of incubation on ice, the sample was electroporated at 1,700 V, 150 Ω, and 50 μF with a resulting time constant of approximately 6.5 ms. This mixture was immediately transferred to 1 ml of 8 × 106 Vero cells in Hanks' balanced salt solution. After 1 h of incubation at 34°C and 5% CO2 with rotation, the mixture was transferred to a 75-cm flask containing 15 ml of Dulbecco's modified Eagle's medium (DMEM) with 10% bovine serum albumin and 1% nonessential amino acids. After 24 h of incubation at 34°C and 5% CO2, the medium was changed to complete DMEM with 200 ng/ml rifampin. The medium was replaced every 3 days until bacillary growth was observed by Diff-Quik staining.

Sorting and recovery of transformants.

Single-bacterial-cell sorting was performed as described in reference 18. Briefly, the peridinin chlorophyll protein (PerCP)/Cy5.5 fluorochrome was conjugated to the anti-R. conorii antibody RcPFA (19) with the Lightning-Link PerCP/Cy5.5 conjugation kit (Innova Biosciences). R. conorii(pRam18dRGA[AmTrCh]) was stained with anti-R. conorii antibody RcPFA-PerCP/Cy5.5 at 10 μg/ml. Labeled R. conorii was applied to a FACSJazz cell sorter with forward scatter (FSC) threshold/trigger, side scatter (SSC) versus FSC graph, and SSC versus log 488-nm excitation 692/30-nm emission graph. Lysed Vero cells and nontransformed R. conorii were used as controls to set negative and positive sort gates. pRam18dRGA[AmTrCh]-transformed R. conorii was applied to the sorter and sorted with 1 sort event into 32 wells, 5 sort events into 32 wells, and 50 sort events into 32 wells of a 96-well plate containing Vero cells in complete DMEM containing 200 ng/ml rifampin. After 5 days of culture, the infected Vero cells were lifted with trypsin-EDTA and expanded into three identical 96-well plates containing Vero cells in complete DMEM with 200 ng/ml rifampin. Flow cytometric analyses and cell sorting of pathogenic R. conorii were performed in a CDC-approved biosafety level 3 (BSL3) laboratory in the Department of Pathobiological Sciences at the Louisiana State University (LSU) School of Veterinary Medicine (SVM) by utilizing protocols approved by the LSU Institutional Biological and Recombinant DNA Safety Committee (IBRDSC).

At 10 total days of growth after sorting, one of the three identical plates was fixed with 4% paraformaldehyde. This plate was stained with the anti-R. conorii antibody RcPFA (19) and an anti-rabbit–Alexa Fluor 546 antibody and observed under ×10 magnification in an Olympus IX71 inverted microscope for the presence of R. conorii (RcPFA) and GFPuv-mediated fluorescence. Subsequently, DNA was extracted as described below from wells containing R. conorii and demonstrating GFPuv-mediated fluorescence. Whole DNA was extracted with the PureLink Genomic DNA kit (Life Technologies) according to the manufacturer's instructions. The extracted DNA was subjected to PCR with gfpuv_F and gfpuv_R (see Table S1 in the supplemental material), followed by agarose gel electrophoresis to ensure the presence of pRam18dRGA[AmTrCh]. The clone utilized in this study came from a single-event sort into 1 well of a 96-well plate.

Immunofluorescence microscopy.

R. conorii and R. conorii(pRam18dRGA[AmTrCh]) were cultured in Vero cells. Both types of bacteria were purified by sucrose density centrifugation (17) and mounted onto a single glass slide with a Cytopro Dual sample chamber with a Cytospin centrifuge. The bacteria were fixed with 4% paraformaldehyde and stained with RcPFA with anti-rabbit–Alexa Fluor 350. All microscopy was captured with a Leica DM4000B equipped with a 100× oil objective, an EvolutionOEi camera, with ImagePro Plus software, and green fluorescent protein (GFP) (excitation wavelength, 488/40 nm; emission wavelength, 527/30 nm), tetramethyl rhodamine isocyanate (TRITC) (excitation wavelength, 515 to 560 nm; emission wavelength, LP590 nm), and 4′,6-diamidino-2-phenylindole (DAPI) (excitation wavelength, 360/40 nm; emission wavelength, 470/40 nm) filters. Analysis of fluorescence intensity was performed with the RGB profile plot plugin within the NIH ImageJ freeware package (http://rsb.info.nih.gov/ij/).

Flow cytometric analysis.

Sucrose density gradient-purified bacteria were analyzed on a BD FACSJazz equipped with aerosol containment and placed within a Baker class IIA biosafety cabinet within the BSL3 laboratory suite at the LSU SVM. R. conorii(pRam18dRGA[AmTrCh]) was stained with anti-R. conorii antibody RcPFA-PerCP/Cy5.5 at 10 μg/ml. All cytometric events were initially gated for FSC and SSC to eliminate large events. PerCP/Cy5.5-mediated fluorescence was measured at 488-nm excitation and 692/40-nm emission wavelengths. GFPuv-mediated fluorescence was measured at 488-nm excitation and 530/40-nm emission wavelengths. This fluorescence combination did not produce cross-channel background fluorescence, so compensation was not required. Each heat map represents at least 50,000 FSC/SSC gated events.

Growth in culture.

R. conorii and R. conorii(pRam18dRGA[AmTrCh]) were used to infect Vero cells at a multiplicity of infection (MOI) of 1 bacterium/host cell. Bacteria were allowed to grow at 34°C and 5% CO2 for a total of 5 days in DMEM–10% FBS supplemented with 200 ng/ml rifampin for the transformed bacteria. At days 1, 3, 5, 7, and 9 postinfection, the infected Vero cells were scraped from the culture plate and DNA was extracted as described above for host, R. conorii, and plasmid quantification.

R. conorii quantification.

The chromosomal ratio of mouse or Vero (actin) to R. conorii (sca1) DNA was queried by quantitative PCR (qPCR) with TaqMan master mix with a 5-min incubation at 95°C, followed by 55 cycles of 95°C for 15 s, 53°C for 15 s, and 68°C for 30 s. Mouse actin was amplified with actin_Ms_F, actin_Ms_R, and actin_Ms_Max(Vic). Vero actin was amplified with actin_Vero_F, actin_Vero_R, and actin_V_Hex(Vic). R. conorii sca1 was amplified with sca1_Rc_F, sca1_Rc_R, and sca1_Rc/Rr_Fam (see Table S1 in the supplemental material). All unknowns were quantified by ΔΔCT compared to molar standards.

Plasmid quantification.

DNA from the above-described growth curve was used to quantify the presence of the pRam18dRGA[AmTrCh] plasmid during growth in culture. The ratio of plasmid (gfpuv) to R. conorii (sca1) was determined by qPCR with SYBR master mix (Roche) by preincubation at 95°C for 5 min, followed by 50 cycles of 95°C for 15 s, 55°C for 15 s, and 68°C for 30 s. Specificity of amplification was validated by melting curve analysis. gfpuv was amplified with gfpuv_F and gfpuv_R. R. conorii sca1 was amplified with sca1_Rc_F and sca1_Rc_R (see Table S1 in the supplemental material).

Animal infection.

Five- to 7-week-old male C3H/HeN mice were inoculated by retro-orbital injection with 6 × 106 (1.5 50% lethal doses [LD50]) of R. conorii (n = 5) or R. conorii(pRam18dRGA[AmTrCh]) (n = 11) or phosphate-buffered saline (PBS) (n = 5) (19). Infected mice were monitored twice daily for signs of disease and daily for weight change. Animals that exhibited symptoms of severe disease that were consistent with not recovering from the infection were removed from the study and scored as succumbing to the infection as previously described (19). These symptoms included ruffled fur, hunched posture, shallow respiration, immobility when touched, and loss of at least 15% of the initial body weight. Animal experiments were conducted in accordance with protocols approved by the IBRDSC and Institutional Animal Care and Use Committee (IACUC) at the LSU SVM. Two predesignated mice were sacrificed at 1, 3, and 5 days postinfection. Spleens, kidneys, livers, hearts, lungs, brains, and blood were aseptically extracted to determine bacterial loads and plasmid copy numbers and assess pathological lesions. Each organ was split into two sections for use in PCR analyses as described above and pathological examination as described below.

Indirect IHC analysis.

Murine tissues, including those from the lung, liver, and spleen, were collected immediately after euthanasia and fixed in 10% buffered formalin (1:10 tissue-to-formalin ratio). Samples were routinely processed and embedded in paraffin, and 5-μm sections were cut for staining with hematoxylin and eosin (H&E). To localize R. conorii within the infected animals, isolated tissues were additionally subjected to immunohistochemical (IHC) examination with an anti-RcPFA antibody that recognizes several SFG rickettsial species (19) or with an anti-Iba1 macrophage-specific antibody. Primary antibody staining was followed by a biotinylated anti-rabbit IgG secondary antibody (Vector BA 1000 at 1:1,000; Vector Laboratories) and exposure to the detection reagent (Vectastain ABC Rabbit IgG kit, Vector PK 6102; Vector Laboratories). Slides were analyzed, micrographs were captured, and pathological changes were recorded by a Diplomate of the American College of Veterinary Pathologists (DACVP).

Identification of R. conorii in leukocytes.

Whole mouse blood was recovered by cardiac puncture immediately after euthanasia. Blood was immediately anticoagulated in a BD Microtainer with K2EDTA diluted 1:1 with PBS. Leukocytes were purified with Ficoll-Paque Premium (GE Healthcare) as directed by the manufacturer. Leukocyte preparations were fixed with 4% paraformaldehyde, permeabilized with 0.1% Triton X-100, and blocked with 2% bovine serum albumin. For initial identification of R. conorii infection, host nuclei were labeled with 1 ng/ml DAPI and phalloidin-Texas Red (1:200; Life Sciences). These cells were visualized at a magnification of ×100 as described above. Images were created by the identification of fields containing GFPuv-positive bacilli, followed by capture of Texas Red and DAPI images. Micrographs of the host cells without the corresponding GFPuv-positive R. conorii image were scored for cell lineage by a DACVP. Only after blind identification of the host cell types were the location and number of GFPuv-positive R. conorii bacteria attributed to each cell type.

Statistics.

Bacterial growth in culture was assessed by nonlinear-fit analysis of the slope and intercept. Plasmid contents in culture and in mice were analyzed by one-way analysis of variance. Differences in survival between R. conorii and R. conorii(pRam18dRGA[AmTrCh]) were analyzed by comparison of survival curves via log rank (Mantel-Cox) test in five mice per experimental group. All statistical analyses were performed with the tests indicated within the Prism software package (GraphPad Software Inc.).

RESULTS

Transformation and isolation of pRam18dRGA[AmTrCh]-transformed R. conorii in cell culture.

Previous reports had demonstrated the feasibility of using plasmids, including pRam18dRGA[AmTrCh], to transform both recognized human-pathogenic and nonpathogenic members of the genus Rickettsia (9-12, 15). Although successful, these strategies have technical limitations when used to obtain clonal transformed populations. Use of either limiting dilution or plaque assays to isolate Rickettsia clones are time and labor intensive, and repeated passage of these bacteria outside a relevant animal model can potentially contribute to attenuation (20). We sought to determine whether transformation of R. conorii is achievable with this plasmid and whether high-throughput fluorescence-activated cell sorting (FACS) would be useful in isolating clonal transformed R. conorii populations.

The plasmid pRam18dRGA[AmTrCh] (9) was introduced into purified R. conorii Malish 7 by electroporation, and transformed bacteria were subsequently subjected to rifampin selection for 10 days. After outgrowth of Rickettsia bacilli was observed via Diff-Quik staining and microscopic analysis, the infected host cells were lysed by needle passage and unbroken cells were removed with a 2-μm filter. The liberated bacteria were subsequently labeled with a PerCP/Cy5.5-conjugated anti-R. conorii antibody (RcPFA) (19) and subjected to flow cytometric analysis. Rifampin-resistant, fluorescently labeled R. conorii bacilli were individually sorted by extremely stringent single-event sorting with a FACSJazz cell sorter (18). After outgrowth from the single parent bacterium, clonal populations were screened for the presence of GFPuv-positive R. conorii by PCR with primers specific for the gfpuv gene on the pRam18dRGA[AmTrCh] plasmid (data not shown). A single R. conorii(pRam18dRGA[AmTrCh]) clone was propagated and purified by sucrose density centrifugation for further analysis.

The pRam18dRGA[AmTrCh] plasmid encodes two fluorescent proteins, GFPuv and mCherry. mCherry expression is driven by the Anaplasma marginale tr promoter, and the encoded protein fluoresces into the TRITC filter set (21, 22). gfpuv expression is driven by the R. prowazekii ompA promoter, and GFPuv fluoresces adequately into the FITC filter set (13). Accordingly, we hypothesized that transformed R. conorii(pRam18dRGA[AmTrCh]) should readily fluoresce into the FITC (GFPuv) and TRITC (mCherry) channels as has been previously observed in other Rickettsia species (9, 15). As shown in Fig. 1A, pRam18dRGA[AmTrCh]-transformed R. conorii bacteria labeled with the RcPFA antibody (19) are observable by fluorescence microscopy under TRITC and FITC filter sets, while untransformed bacteria are not visible. Analysis of the fluorescence signal intensity across a portion of an R. conorii(pRam18dRGA[AmTrCh]) photomicrograph indicates that individual bacteria elaborate both GFPuv and mCherry signals (Fig. 1B).

FIG 1.

FIG 1

Open in a new tab

In vitro fluorescent properties of R. conorii(pRam18dRGA[AmTrCh]). (A) R. conorii and R. conorii(pRam18dRGA[AmTrCh]) were purified and stained with the anti-Rickettsia antibody RcPFA and a corresponding anti-rabbit–Alexa Fluor 350 antibody (blue). Surface staining of the bacteria is apparent in both anti-RcPFA panels, but only R. conorii(pRam18dRGA[AmTrCh]) demonstrates GFPuv (green)- and mCherry (red)-mediated fluorescence. (B) Unstained R. conorii(pRam18dRGA[AmTrCh]) was analyzed for GFPuv- and mCherry-mediated fluorescence. Analysis of fluorescence intensity along the magenta line indicates that individual bacteria elaborate both GFPuv- and mCherry-mediated fluorescence, as demonstrated by overlapping peaks. However, the mCherry fluorescence is not always above the limit of detection.

Flow cytometric analysis of GFPuv fluorescence.

Since GFPuv can be excited under a 488-nm laser and can emit into common filter sets (23), we hypothesized that GFPuv-mediated fluorescence should be observable by flow cytometry in R. conorii(pRam18dRGA[AmTrCh]). First, we conjugated the anti-R. conorii antibody RcPFA with the PerCP/Cy5.5 tandem fluorophore. We used RcPFA-PerCP/Cy5.5 to identify R. conorii when excited at 488 nm and detected with a 692/40-nm emission filter. Indeed, purified R. conorii are readily detectible by flow cytometry with RcPFA-PerCP/Cy5.5 (Fig. 2A versus B). Additionally, GFPuv-mediated fluorescence from R. conorii(pRam18dRGA[AmTrCh]) can be detected with FITC filter sets (488-nm excitation and 530/40-nm emission) (Fig. 2A versus C). This is a novel way to identify Rickettsia-specific events and to identify bacteria transformed to GFPuv fluorescence (Fig. 2D).

FIG 2.

FIG 2

Open in a new tab

Flow cytometric analysis of GFPuv-mediated fluorescence. Nonlabeled R. conorii (A), RcPFA-PerCP/Cy5.5-labeled R. conorii (B), nonlabeled R. conorii(pRam18dRGA[AmTrCh]) (C), and RcPFA-PerCP/Cy5.5-labeled R. conorii(pRam18dRGA[AmTrCh]) (D) were analyzed by flow cytometry for identification of R. conorii-specific events (RcPFA) and GFPuv fluorescence. RcPFA-PerCP/Cy5.5 labels both R. conorii (A versus B) and R. conorii(pRam18dRGA[AmTrCh]) (C versus D). R. conorii(pRam18dRGA[AmTrCh]) demonstrates significant intrinsic GFPuv-mediated fluorescence (A versus C and B versus D).

In vitro growth of transformed R. conorii in cell culture.

We next sought to determine if the presence of an extrachromosomal DNA element in R. conorii has a deleterious effect on fitness in cell culture. R. conorii and R. conorii(pRam18dRGA[AmTrCh]) were inoculated into Vero cell cultures at an MOI of 1. Samples were removed from these cultures 1, 3, 5, 7, and 9 days postinoculation, and genomic DNA was extracted from both the host and the bacteria. qPCR analysis of the ratio of R. conorii (sca1) to Vero (actin) DNA contents indicates that transformed and nontransformed bacteria proliferate similarly over 9 days in Vero cell culture (Fig. 3).

FIG 3.

FIG 3

Open in a new tab

Presence of pRam18dRGA[AmTrCh] does not alter R. conorii fitness in culture. R. conorii(pRam18dRGA[AmTrCh]) (dashed line) and plasmid-free R. conorii (solid line) proliferate similarly in Vero cell culture. qPCR data are expressed as the ratios of R. conorii sca1 to Vero actin DNA after 3, 5, 7, and 9 days of culture in Vero cells. Differences in slope or intercept are not significant.

We then determined whether the plasmid was stably maintained in vitro without selective antibiotic pressure. The copy number of pRam18dRGA[AmTrCh] was determined by qPCR as the ratio of plasmid (gfpuv) to R. conorii chromosomal (sca1) DNA. The pRam18dRGA[AmTrCh] copy number over the time course of culture indicates that the plasmid is maintained without antibiotic selection (see Table 2). The plasmid copy numbers determined over time are similar to those documented in other transformed Rickettsia species (9, 11, 15).

TABLE 2.

Cellular locations of GFPuv-positive bacilli in an ex vivo leukocyte preparation

Parameter No. counted
All bacteria 207
Bacteria present in lymphocytes 110
Bacteria in monocytes 67
Bacteria in granulocytes 14
Bacteria not associated with a host cell 16
Open in a new tab

R. conorii(pRam18dRGA[AmTrCh]) experimental mammalian infection.

An important question pertaining to the future of rickettsial genetics is the persistence of extrachromosomal DNA in animal models of infection when lacking selective pressure. Indeed, pRam18dRGA can be maintained in R. monacensis for five passages in tissue culture without antibiotic selection (9), but this phenotype has not been investigated in an in vivo model. Additionally, future uses of extrachromosomal DNA will be dependent on establishing that the existence of the plasmid does not, in itself, modulate the infectivity of the bacterium. To address these questions, we utilized a well-established murine model of fatal MSF to examine the pathogenesis of R. conorii(pRam18dRGA[AmTrCh]) (19, 24). Susceptible C3H/HeN mice were intravenously infected with 4 × 106 infectious units of R. conorii or R. conorii(pRam18dRGA[AmTrCh]). This dose is 1.5 LD50 for nontransformed R. conorii (19). Mice were monitored for signs of disease, including: ruffled fur, hunched posture, shallow respiration, immobility, and weight loss. R. conorii- and R. conorii(pRam18dRGA[AmTrCh])-infected mice demonstrated progressive morbidity, as demonstrated by persistent weight loss (Fig. 4A). As demonstrated in Fig. 4B, infected mice succumbed to infection during the normal course of disease with death at days 4 to 5.5 postinfection with a slightly later time of death than the wild type. Nevertheless, this finding establishes that R. conorii(pRam18dRGA[AmTrCh]) retains infectivity in a mouse model of infection.

FIG 4.

FIG 4

Open in a new tab

R. conorii(pRam18dRGA[AmTrCh]) retains virulence in a mouse model of fatal infection. (A) Mice were infected with R. conorii (dashed line) or R. conorii(pRam18dRGA[AmTrCh]) (dotted line) or treated with PBS (solid line) and then monitored for morbidity as determined by weight change over the course of infection. (B) The same infected mice were monitored for survival. R. conorii harboring pRam18dRGA[AmTrCh] is capable of producing fatal disease. Times to euthanasia differed significantly (P < 0.05).

To confirm that R. conorii(pRam18dRGA[AmTrCh])-infected mice experienced the normal disease progression, we quantified the bacterial burdens in major organs. Bacterial DNA could be detected and quantified in the mouse spleen and liver after 1 day of infection (Fig. 5A). The total bacterial burden increased after 3 days of infection (Fig. 5B), and this ultimately resulted in rickettsial organ tropism and quantity similar to those previously observed in fatal R. conorii infection (Fig. 5C) (19). Most importantly, pRam18dRGA[AmTrCh] was readily detected at all of the time points tested and the plasmid quantity did not decrease over the course of infection (Fig. 5D). Together, these data demonstrate that the pRam18dRGA[AmTrCh] extrachromosomal DNA does not affect R. conorii pathogenesis and that the plasmid is stably maintained by R. conorii in an animal model of infection.

FIG 5.

FIG 5

Open in a new tab

R. conorii(pRam18dRGA[AmTrCh]) proliferates within infected organs and maintains the extrachromosomal plasmid. R. conorii(pRam18dRGA[AmTrCh]) dissemination into mouse organs after 1 (A), 3 (B), or 5 (C) days of infection is shown. Chromosomal equivalents of R. conorii sca1 and mouse actin were determined by qPCR. (D) Nonselective persistence of pRam18dRGA[AmTrCh] throughout the mammalian infection as determined by the qPCR ratio of chromosomal R. conorii sca1 and the plasmid gene gfpuv. The change in plasmid copy number is not significant.

Pathological examination of animals after 5 days of infection demonstrated R. conorii(pRam18dRGA[AmTrCh]) infection of endothelial cells in the heart and lungs (see Fig. S1 in the supplemental material), as has been previously reported (24). The mouse liver additionally demonstrated multifocal to coalescing coagulative hepatic necrosis characterized by the presence of numerous neutrophils and leukocytic debris (Fig. 6A). IHC staining of this lesion revealed the presence of numerous bacteria within the cytoplasm of macrophages and neutrophils (Fig. 6B). Additionally, bacteria were noted to be infecting cells within vessels (Fig. 6C). In the mouse liver, the inflammatory cell population within vessels, sinusoids, and microabscesses is composed mainly of round cells with macrophage morphology expressing the Iba1 macrophage marker (Fig. 6D). There is no evidence of sloughed endothelial cells within the lumen of blood vessels. These data demonstrate a normal distribution of plasmid-transformed R. conorii in the infected animal along with the presence of these bacteria within macrophages in the liver.

FIG 6.

FIG 6

Open in a new tab

Examination of mouse liver pathology. (A) H&E staining of mouse liver demonstrates microabscess development when the animal is succumbing to infection. (B) IHC staining with anti-R. conorii antibody and hematoxylin counterstain demonstrates the presence of Rickettsia antigen-positive intact bacilli (brown) within liver microabscess. (C) Rickettsia antigen-positive intact bacilli (brown) within cells circulating in a liver capillary. (D) Iba1-positive macrophages (brown) within both a liver venule (lower right) and microabscesses (arrowheads). All images were captured at ×60.

Observation of ex vivo GFPuv-positive bacilli parasitizing leukocytes.

We have demonstrated that R. conorii(pRam18dRGA[AmTrCh]) is present within the endothelium of the infected mammalian host (see Fig. S1 in the supplemental material) but also within several different cell types in infected target tissues and in the circulation (Fig. 6). In addition, we demonstrated that the pRam18dRGA[AmTrCh] plasmid is stably maintained within infected cell cultures and in vivo (Fig. 5D). We therefore hypothesized that transformed R. conorii expresses genes contained in the plasmid and are readily detectable by endogenous GFPuv expression in vivo. To this end, we sacrificed mice that were succumbing to fatal R. conorii infection and isolated the total leukocytes from their blood. After fixation and labeling of the nucleus (DAPI) and host cytoplasm (phalloidin), GFPuv-fluorescent bacilli were readily observed within granulocytic, monocytic, and lymphocytic cells (Fig. 7).

FIG 7.

FIG 7

Open in a new tab

Ex vivo fluorescence of R. conorii(pRam18dRGA[AmTrCh]). Microscopic identification of R. conorii(pRam18dRGA[AmTrCh])-infected cells from a white blood cell preparation. Cells were stained with DAPI (blue) to identify host nuclei and with phalloidin (red) to illuminate the cytoplasm. Green fluorescence results from the intrinsic GFPuv-mediated fluorescence encoded by pRam18dRGA[AmTrCh]. Infected host cells demonstrate intact GFPuv-positive bacilli. White bars, 10 μm. All images were captured at ×100.

We next sought to elucidate the different cell types present within the mammalian bloodstream that are infected by R. conorii by determining the locations of GFPuv-positive R. conorii within the ex vivo white blood cell preparation. GFPuv-positive bacilli were located by fluorescence microscopy without observing the location of the bacteria relative to host cells. After detection of the bacteria, images of the host cells in that microscopic field were acquired. A board-certified veterinary clinical pathologist identified the types of leukocytes in the microscopic field without knowledge of the location of the bacteria. Only after identification of host cells were the locations of the bacteria assigned. Of the 207 GFP-positive R. conorii bacteria that could be assigned to a host location, 110 were present in lymphocytes, 67 were in monocytes, 14 were in granulocytic cells, and 16 were not associated with any host cell (see Table 2). Thus, these data suggest that in addition to endothelial cells, cells of nonendothelial origin support R. conorii parasitism during fatal mammalian infection.

DISCUSSION

Here, we have demonstrated the ability of R. conorii to be transformed with the plasmid pRam18dRGA[AmTrCh] (9). R. conorii stably maintains this plasmid in culture with no apparent effect on fitness but with the addition of GFPuv- and mCherry-mediated fluorescence phenotypes (Fig. 1 to 3; Table 1). By utilizing a well-characterized mouse model of fatal MSF, we were able to demonstrate that R. conorii(pRam18dRGA[AmTrCh]) retains infectivity in a mouse model of fatal infection (Fig. 4). A fatal R. conorii(pRam18dRGA[AmTrCh]) challenge results in disseminated infection and, importantly, retention of the extrachromosomal DNA (Fig. 5). Microscopic examination of R. conorii(pRam18dRGA[AmTrCh]) within the mouse demonstrated the well-characterized endothelial-cell-targeted infection (see Fig. S1 in the supplemental material), with newly identified infection of mouse macrophages within the liver and infection of nonendothelial cells within the microvascular circulation (Fig. 6). Further microscopic examination showed the presence of GFPuv-positive bacilli within leukocytes (Fig. 7), and pathological examination was used to identify the nature of this host infection (Table 2). Taken together, these data indicate that the use of pRam18dRGA plasmids will add depth to our analysis of mammalian models of Rickettsia infection. The GFPuv-mediated fluorescence phenotype will permit a more complete analysis of the nature of the pathogenic process and allow us to identify host cells parasitized by pathogenic Rickettsia species.

TABLE 1.

Plasmid copy number during growth in Vero cells

Culture time (days) Avg plasmid copy no. ± SDa
3 3.09 ± 0.52
5 2.82 ± 2.03
7 0.52 ± 0.30
9 0.55 ± 0.14
Open in a new tab
a

Changes in copy number are not significant.

In vivo maintenance of pRam18dRGA plasmids (Fig. 5 and 7) and the adaptable nature of the pRam18dRGA[MCS] plasmid allow additional ways to examine Rickettsia pathogenesis. The pRam18dRGA[MCS] plasmid contains a multiple cloning site for the insertion of any DNA (9). This design feature makes this plasmid a true shuttle vector, because the additional DNA will be maintained within the bacteria during in vivo Rickettsia infection. First and foremost among potential uses of shuttle vectors is proper and complete examination of gene deletion mutants, because of the essential need for in trans gene complementation (25). pRam18dRGA[MCS] derivatives can also be used to monitor bacterial gene expression through the insertion of appropriate reporter genes, for analysis of natural conjugation systems, or for the heterologous expression of many different bacterial genes (9, 26).

Of particular importance is our ex vivo microscopic observation of fluorescent R. conorii directly in the infected animal. Analysis of in vivo or ex vivo fluorescent bacilli will promote a more complete examination of Rickettsia pathogenesis, as has been done with other bacteria (27). This microscopic technique can also be readily used to examine cell types infected or parasitized by pathogenic Rickettsia, observe arthropod infection in real time, and monitor the kinetics of rickettsemia in infected mammals.

SFG rickettsiae have long been known to infect the endothelium of the infected animal (24, 28, 29). In this report, we provide evidence of nonendothelial parasitism by R. conorii (Fig. 6 and 7; Table 2). We propose that infection of nonendothelial cells has been overlooked during prior examinations of SFG Rickettsia pathogenesis and that our newly developed techniques have allowed a more complete analysis of infection. This idea is supported by several in vitro and ex vivo studies that have demonstrated that rickettsial species can infect and grow within macrophages and hepatocytes (30–33), suggesting that the interaction with cells other than those of the endothelium may be an important and underappreciated aspect of the pathogenesis of rickettsial diseases. Additionally, IHC and immunofluorescence microscopy both display intact bacilli, and nonendothelial cells frequently contain multiple plasmid-transformed R. conorii bacteria (Fig. 6 and 7), suggesting that these bacteria may not be readily susceptible to immune-mediated killing. A more complete examination of the viability of bacilli, infection kinetics, and host tropism of R. conorii parasitism is necessary.

In conclusion, we have utilized R. conorii(pRam18dRGA[AmTrCh]) to demonstrate the value of plasmids in the analysis of in vivo Rickettsia pathogenesis. The techniques described here have promoted our identification of novel host cell types parasitized by pathogenic R. conorii in a fatal model of MSF. Therefore, the use of pRam18dRGA[AmTrCh] and related plasmids will strongly influence and enhance the course of future rickettsial research.

Supplementary Material

Supplemental material
supp_84_3_790__index.html (975B, html)

ACKNOWLEDGMENTS

We thank Kevin Macaluso (LSU SVM) for helpful comments on experimental design and manuscript construction, David Wood (University of South Alabama) for dissemination of techniques for electrotransformation, Ulrike Munderloh (University of Minnesota) for distribution of reagents, and Rebecca Christofferson (LSU SVM) for aid in the statistical analysis of data.

The content of this report is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

Footnotes

Supplemental material for this article may be found at http://dx.doi.org/10.1128/IAI.01205-15.

REFERENCES

  • 1.Childs JE, Paddock CD. 2002. Passive surveillance as an instrument to identify risk factors for fatal Rocky Mountain spotted fever: is there more to learn? Am J Trop Med Hyg 66:450–457. [DOI] [PubMed] [Google Scholar]
  • 2.Drexler NA, Dahlgren FS, Heitman KN, Massung RF, Paddock CD, Behravesh CB. 31 August 2015. National surveillance of spotted fever group rickettsioses in the United States, 2008–2012. Am J Trop Med Hyg doi: 10.4269/ajtmh.15-0472. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Smadel JE. 1959. Status of the rickettsioses in the United States. Ann Intern Med 51:421–435. doi: 10.7326/0003-4819-51-3-421. [DOI] [PubMed] [Google Scholar]
  • 4.de Sousa R, Nobrega SD, Bacellar F, Torgal J. 2003. Mediterranean spotted fever in Portugal: risk factors for fatal outcome in 105 hospitalized patients. Ann N Y Acad Sci 990:285–294. doi: 10.1111/j.1749-6632.2003.tb07378.x. [DOI] [PubMed] [Google Scholar]
  • 5.Hand WL, Miller JB, Reinarz JA, Sanford JP. 1970. Rocky Mountain spotted fever. A vascular disease. Arch Intern Med 125:879–882. [PubMed] [Google Scholar]
  • 6.Walker DH, Occhino C, Tringali GR, Di Rosa S, Mansueto S. 1988. Pathogenesis of rickettsial eschars: the tache noire of boutonneuse fever. Hum Pathol 19:1449–1454. doi: 10.1016/S0046-8177(88)80238-7. [DOI] [PubMed] [Google Scholar]
  • 7.Walker DH, Gear JH. 1985. Correlation of the distribution of Rickettsia conorii, microscopic lesions, and clinical features in South African tick bite fever. Am J Trop Med Hyg 34:361–371. [DOI] [PubMed] [Google Scholar]
  • 8.Beare PA, Sandoz KM, Omsland A, Rockey DD, Heinzen RA. 2011. Advances in genetic manipulation of obligate intracellular bacterial pathogens. Front Microbiol 2:97. doi: 10.3389/fmicb.2011.00097. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Burkhardt NY, Baldridge GD, Williamson PC, Billingsley PM, Heu CC, Felsheim RF, Kurtti TJ, Munderloh UG. 2011. Development of shuttle vectors for transformation of diverse Rickettsia species. PLoS One 6:e29511. doi: 10.1371/journal.pone.0029511. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Baldridge GD, Burkhardt NY, Felsheim RF, Kurtti TJ, Munderloh UG. 2008. Plasmids of the pRM/pRF family occur in diverse Rickettsia species. Appl Environ Microbiol 74:645–652. doi: 10.1128/AEM.02262-07. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Baldridge GD, Burkhardt NY, Labruna MB, Pacheco RC, Paddock CD, Williamson PC, Billingsley PM, Felsheim RF, Kurtti TJ, Munderloh UG. 2010. Wide dispersal and possible multiple origins of low-copy-number plasmids in Rickettsia species associated with blood-feeding arthropods. Appl Environ Microbiol 76:1718–1731. doi: 10.1128/AEM.02988-09. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Baldridge GD, Burkhardt NY, Oliva AS, Kurtti TJ, Munderloh UG. 2010. Rickettsial ompB promoter regulated expression of GFPuv in transformed Rickettsia montanensis. PLoS One 5:e8965. doi: 10.1371/journal.pone.0008965. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Liu ZM, Tucker AM, Driskell LO, Wood DO. 2007. Mariner-based transposon mutagenesis of Rickettsia prowazekii. Appl Environ Microbiol 73:6644–6649. doi: 10.1128/AEM.01727-07. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Felsheim RF, Herron MJ, Nelson CM, Burkhardt NY, Barbet AF, Kurtti TJ, Munderloh UG. 2006. Transformation of Anaplasma phagocytophilum. BMC Biotechnol 6:42. doi: 10.1186/1472-6750-6-42. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Wood DO, Hines A, Tucker AM, Woodard A, Driskell LO, Burkhardt NY, Kurtti TJ, Baldridge GD, Munderloh UG. 2012. Establishment of a replicating plasmid in Rickettsia prowazekii. PLoS One 7:e34715. doi: 10.1371/journal.pone.0034715. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Rachek LI, Tucker AM, Winkler HH, Wood DO. 1998. Transformation of Rickettsia prowazekii to rifampin resistance. J Bacteriol 180:2118–2124. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Ammerman NC, Beier-Sexton M, Azad AF. 2008. Laboratory maintenance of Rickettsia rickettsii. Curr Protoc Microbiol Chapter 3:Unit 3A.5. doi: 10.1002/9780471729259.mc03a05s11. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Riley SP, Macaluso KR, Martinez JJ. 2015. Electrotransformation and clonal isolation of Rickettsia species. Curr Protoc Microbiol 39:3A.6.1–3A.6.20. doi: 10.1002/9780471729259.mc03a06s39. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Chan YG, Riley SP, Chen E, Martinez JJ. 2011. Molecular basis of immunity to rickettsial infection conferred through outer membrane protein B. Infect Immun 79:2303–2313. doi: 10.1128/IAI.01324-10. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Perez Gallardo F, Fox JP. 1948. Infection and immunization of laboratory animals with Rickettsia prowazekii of reduced pathogenicity, strain E. Am J Hyg 48:6–21. [DOI] [PubMed] [Google Scholar]
  • 21.Barbet AF, Agnes JT, Moreland AL, Lundgren AM, Alleman AR, Noh SM, Brayton KA, Munderloh UG, Palmer GH. 2005. Identification of functional promoters in the msp2 expression loci of Anaplasma marginale and Anaplasma phagocytophilum. Gene 353:89–97. doi: 10.1016/j.gene.2005.03.036. [DOI] [PubMed] [Google Scholar]
  • 22.Shaner NC, Campbell RE, Steinbach PA, Giepmans BN, Palmer AE, Tsien RY. 2004. Improved monomeric red, orange and yellow fluorescent proteins derived from Discosoma sp. red fluorescent protein. Nat Biotechnol 22:1567–1572. doi: 10.1038/nbt1037. [DOI] [PubMed] [Google Scholar]
  • 23.Crameri A, Whitehorn EA, Tate E, Stemmer WP. 1996. Improved green fluorescent protein by molecular evolution using DNA shuffling. Nat Biotechnol 14:315–319. doi: 10.1038/nbt0396-315. [DOI] [PubMed] [Google Scholar]
  • 24.Walker DH, Popov VL, Wen J, Feng HM. 1994. Rickettsia conorii infection of C3H/HeN mice. A model of endothelial-target rickettsiosis. Lab Invest 70:358–368. [PubMed] [Google Scholar]
  • 25.Falkow S. 1988. Molecular Koch's postulates applied to microbial pathogenicity. Rev Infect Dis 10(Suppl 2):S274–S276. doi: 10.1093/cid/10.Supplement_2.S274. [DOI] [PubMed] [Google Scholar]
  • 26.Lee SH, Camilli A. 2000. Novel approaches to monitor bacterial gene expression in infected tissue and host. Curr Opin Microbiol 3:97–101. doi: 10.1016/S1369-5274(99)00058-2. [DOI] [PubMed] [Google Scholar]
  • 27.Konjufca V, Miller MJ. 2009. Imaging Listeria monocytogenes infection in vivo. Curr Top Microbiol Immunol 334:199–226. doi: 10.1007/978-3-540-93864-4_9. [DOI] [PubMed] [Google Scholar]
  • 28.Feng HM, Wen J, Walker DH. 1993. Rickettsia australis infection: a murine model of a highly invasive vasculopathic rickettsiosis. Am J Pathol 142:1471–1482. [PMC free article] [PubMed] [Google Scholar]
  • 29.Walker DH, Crawford CG, Cain BG. 1980. Rickettsial infection of the pulmonary microcirculation: the basis for interstitial pneumonitis in Rocky Mountain spotted fever. Hum Pathol 11:263–272. doi: 10.1016/S0046-8177(80)80008-6. [DOI] [PubMed] [Google Scholar]
  • 30.Walker DH, Popov VL, Feng HM. 2000. Establishment of a novel endothelial target mouse model of a typhus group rickettsiosis: evidence for critical roles for gamma interferon and CD8 T lymphocytes. Lab Invest 80:1361–1372. doi: 10.1038/labinvest.3780144. [DOI] [PubMed] [Google Scholar]
  • 31.Gambrill MR, Wisseman CL Jr. 1973. Mechanisms of immunity in typhus infections. 3. Influence of human immune serum and complement on the fate of Rickettsia mooseri within the human macrophages. Infect Immun 8:631–640. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Oster CN, Burke DS, Kenyon RH, Ascher MS, Harber P, Pedersen CE Jr. 1977. Laboratory-acquired Rocky Mountain spotted fever. The hazard of aerosol transmission. N Engl J Med 297:859–863. [DOI] [PubMed] [Google Scholar]
  • 33.Radulovic S, Price PW, Beier MS, Gaywee J, Macaluso JA, Azad A. 2002. Rickettsia-macrophage interactions: host cell responses to Rickettsia akari and Rickettsia typhi. Infect Immun 70:2576–2582. doi: 10.1128/IAI.70.5.2576-2582.2002. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplemental material
supp_84_3_790__index.html (975B, html)
IAI.01205-15_zii999091612so1.pdf (13.4MB, pdf)

Articles from Infection and Immunity are provided here courtesy of American Society for Microbiology (ASM)

RESOURCES