Isolation of “Candidatus Rickettsia andeanae” (Rickettsiales: Rickettsiaceae) in Embryonic Cells of Naturally Infected Amblyomma maculatum (Ixodida: Ixodidae) Author(s): F.A.G. Ferrari, J. Goddard, G. M. Moraru, W.E.C. Smith, and A. S. Varela-Stokes Source: Journal of Medical Entomology, 50(5):1118-1125. Published By: Entomological Society of America URL: http://www.bioone.org/doi/full/10.1603/ME13010

BioOne (www.bioone.org) is a nonprofit, online aggregation of core research in the biological, ecological, and environmental sciences. BioOne provides a sustainable online platform for over 170 journals and books published by nonprofit societies, associations, museums, institutions, and presses. Your use of this PDF, the BioOne Web site, and all posted and associated content indicates your acceptance of BioOne’s Terms of Use, available at www.bioone.org/page/terms_of_use. Usage of BioOne content is strictly limited to personal, educational, and non-commercial use. Commercial inquiries or rights and permissions requests should be directed to the individual publisher as copyright holder.

BioOne sees sustainable scholarly publishing as an inherently collaborative enterprise connecting authors, nonprofit publishers, academic institutions, research libraries, and research funders in the common goal of maximizing access to critical research.

VECTOR/PATHOGEN/HOST INTERACTION, TRANSMISSION

Isolation of “Candidatus Rickettsia andeanae” (Rickettsiales: Rickettsiaceae) in Embryonic Cells of Naturally Infected Amblyomma maculatum (Ixodida: Ixodidae) F.A.G. FERRARI,1 J. GODDARD,2 G. M. MORARU,1 W.E.C. SMITH,1

AND

A. S. VARELA-STOKES1,3

J. Med. Entomol. 50(5): 1118Ð1125 (2013); DOI: http://dx.doi.org/10.1603/ME13010

ABSTRACT The Gulf Coast tick, Amblyomma maculatum Koch, has become increasingly important in public health for its role as a vector of the recently recognized human pathogen, Rickettsia parkeri. More recently, these ticks were also found to harbor a novel spotted fever group rickettsia, “Candidatus Rickettsia andeanae.” First identiÞed in Peru, and subsequently reported in ticks collected in the United States, Chile, and Argentina, “Ca. R. andeanae” remains largely uncharacterized, in part because of the lack of a stable isolate. Although the isolation of “Ca. R. andeanae” was recently described in DH82, Vero, and Drosophila S2 cells, its stability in these cell lines was not shown. To evaluate “Ca. R. andeanae” transmission and pathogenicity in vertebrates, as well as further describe biological characteristics of this candidate species to fulÞll criteria for its establishment as a new species, availability of a stable isolate is essential. Here we describe the propagation of “Ca. R. andeanae” by using a primary culture derived from naturally infected A. maculatum embryos. Subsequent passage of the “Ca. R. andeanae” isolate to ISE6 (Ixodes scapularis embryonic) and Vero (African green monkey kidney epithelial) cell lines demonstrated limited propagation of the rickettsiae. Treatment of the infected primary cells with tetracycline resulted in cultures negative for “Ca. R. andeanae” by polymerase chain reaction and microscopy. Establishment of an isolate of “Ca. R. andeanae” will promote further investigation into the signiÞcance of this tick-associated rickettsia, including its role in spotted fever and interactions with the sympatric species, R. parkeri in A. maculatum. KEY WORDS Candidatus Rickettsia andeanae, Amblyomma maculatum, Gulf Coast tick, spotted fever group rickettsiae, culture

Rickettsiae are gram-negative, obligate, intracellular bacteria, some of which can be pathogenic to animals. Most bacteria in the spotted fever group of rickettsiae (SFGR) are maintained in nature within ticks that function as hosts for the bacteria as well as vectors for transmission to vertebrates during a bloodmeal (Dumler and Walker 2005). In vitro cultivation of SFGR is possible by using a variety of cell lines, including Vero, L-929, HEL, and MRC5 cells (Dumler and Walker 2005). However, several embryonic tick cell lines have been used to aid in isolating fastidious bacteria, including Rickettsia spp., that fail to grow in mammalian cell lines (BellÐSakyi et al. 2007). Tick cell lines have also been valuable for proteomics and genomics studies as well as the production of vaccines (BellÐSakyi et al. 2007). Finally, the ability to establish tick cell lines offers an approach to isolating, characterizing, and 1 Department of Basic Sciences, College of Veterinary Medicine, Mississippi State University, 240 Wise Center Dr., Mississippi State, MS 39762. 2 Department of Biochemistry, Molecular Biology, Entomology & Plant Pathology, Mississippi State University, 100 Old Hwy. 12, Mississippi State, MS 39762. 3 Corresponding author, e-mail: [email protected].

investigating hostÐpathogen relationships for novel rickettsiae in the intracellular environment of their natural tick host (BellÐSakyi et al. 2007). This approach has been successful in the propagation of Rickettsia peacockii (Niebylski et al. 1997), a fastidious endosymbiont that was eventually isolated in Dermacentor andersoni cells (Simser et al. 2001). In 2004, a novel SFGR was identiÞed in Amblyomma maculatum Koch and Ixodes boliviensis Neumann ticks in the Peruvian Andes (Blair et al. 2004). “Candidatus Rickettsia andeanae” has now been reported from A. maculatum (commonly known as the Gulf Coast tick) collected in Mississippi, Florida, Georgia, Kansas, Oklahoma, and Virginia (Sumner et al. 2007, Paddock et al. 2010, Fornadel et al. 2011, Jiang et al. 2011, LuceÐFedrow et al. 2011, VarelaÐStokes et al. 2011, Wright et al. 2011). Current knowledge of “Ca. R. andeanae” has been gained mainly from surveys detecting the bacterium in ticks by polymerase chain reaction (PCR) assays. In addition to harboring “Ca. R. andeanae,” A. maculatum is the known vector of Rickettsia parkeri, an SFGR Þrst isolated in 1939 and considered nonpathogenic until the Þrst human infection was diagnosed in 2002 (Parker et al. 1939, Paddock et

0022-2585/13/1118Ð1125$04.00/0 䉷 2013 Entomological Society of America

September 2013

FERRARI ET AL.: “Candidatus RICKETTSIA ANDEANAE” AND A. maculatum

al. 2004). Although no human (or vertebrate) infections with “Ca. R. andeanae” have been reported to date, thorough investigations to evaluate potential infectivity and pathogenicity of this rickettsia will require establishing a stable isolate to be used in experimental animal infections. Recently, “Ca. R. andeanae” was propagated in three different cell lines (DH82, VERO, and Drosophila S2) (LuceÐFedrow et al. 2011). However, as none of these cultures are from organisms naturally found infected with this novel SFGR, they may not reßect the natural system, and their usefulness for characterizing “Ca. R. andeanae” further is unknown. Tick-borne pathogens respond to the host cell environment by changing gene expression (Nelson et al. 2008). For example, Ehrlichia chaffeensis propagated in tick cells expressed a different set of proteins than the same Ehrlichia species grown in macrophages (Singu et al. 2006); similarly, transcription levels for the spoT3 gene differed for Rickettsia conorii grown in mammalian compared with cells of Aedes albopictus (Skuse) (Rovery et al. 2005). In addition, previous attempts to grow “Ca. R. andeanae” in Vero cells, in Ae. albopictus cell line (C6/36), and in Ixodes scapularis Say tick cell line, ISE6, were unsuccessful (Paddock et al. 2010). Thus, the viability of this novel rickettsia may not be optimal in cells other than those of its natural host. In this study, we sought to develop an embryonic cell line from the natural tick host species, A. maculatum, to use for isolation of “Ca. R. andeanae.” Our rationale was that such a cell line would allow for more efÞcient and stable growth of this rickettsia. Consequently, future investigations of “Ca. R. andeanae” transmission and pathogenicity, as well as further genomic and proteomic characterization studies, would be possible. In addition, investigations of other organisms found in A. maculatum, including the human pathogen, R. parkeri, would beneÞt from use of the Gulf Coast tick host cell line. Materials and Methods Source of Tick Embryos. To establish Gulf Coast tick cell lines, both naturally infected with “Ca. R. andeanae” and not infected with this rickettsia, we used embryonated eggs produced by engorged female A. maculatum. We purchased eight fully engorged gravid A. maculatum females that were reared in laboratory colonies at the Oklahoma State University (OSU; n ⫽ 4) and the Texas A&M University (TAMU; n ⫽ 4). These tick sources were chosen because we previously detected presence and absence of “Ca. R. andeanae” DNA by PCR assay in A. maculatum purchased from OSU colonies and TAMU colonies, respectively; both colonies were also negative for R. parkeri by PCR (G.M.M. and A.S.V.-S., unpublished data). On arrival, engorged females were physically wiped with 70% ethanol and each placed in an individual well of a sterile multi-well plate for oviposition. Plates were kept in a chamber maintained at ⬇92% humidity by using saturated potassium nitrate solution. Portions of each egg mass were collected and

1119

tested by PCR for evidence of SFGR by using groupwide primers, and then for “Ca. Rickettsia andeanae” and R. parkeri by using speciÞc primers, as described later in the text. Isolation of “Ca. R. andeanae” in Embryonic A. maculatum Cells. After tick embryos had completed ⬇75% of their development (determined by microscopic observation), portions of ⬇100 mg of eggs (or approximately one-tenth of the egg mass) from individual females were removed for establishment of cultures. Each portion (three to six per egg mass) was processed separately. Engorged TAMU ticks arrived 3 d after embryos from OSU egg masses had been processed. TAMU egg masses were similarly processed at about 4 wk after oviposition, which coincided with embryos reaching ⬇75% development. We surface-disinfected egg samples inside a biosafety cabinet by using 3-min washes on a rocker as follows: 0.5% household bleach containing a drop of Tween 80, 70% ethanol, twice in sterile phosphate-buffered saline (pH 7.4), and the last wash in modiÞed LeibovitzÕs medium (L-15B300) (Munderloh and Kurtti 1989, Munderloh et al. 1999). When processing the egg masses from TAMU ticks, we included an initial disinfection step with 0.1% benzalkonium chloride before disinfection with bleach and ethanol, and the rinses in phosphate-buffered saline and medium, as previously described. After egg samples were disinfected, we added 5 ml of L-15B300 medium supplemented with 20% heat-inactivated fetal bovine serum, 10% tryptose phosphate broth, antibiotics (100 U/ml of penicillin and 100 mg/ml of streptomycin sulfate, SigmaÐAldrich, St. Louis, MO), and antifungal (10 mg/ml of amphotericin, SigmaÐAldrich). Eggs were crushed in this medium by using a sterile glass rod and then centrifuged at 100 ⫻ g for 1 min to separate embryos from egg shells. For each portion of OSU egg mass processed, we seeded one 12.5-cm2 plug-seal cap ßask (BD Falcon, BD Biosciences, San Jose, CA) with 5 ml of supernatant, and two 12.5-cm2 ßasks with the remaining pellet resuspended in 10 ml of medium. For TAMU egg mass portions, the supernatant was divided into two ßat-bottom tubes (Nunc, Thermo Fisher, Rochester, NY), whereas the remaining pellet was resuspended and transferred to one ßat-bottom tube. Flasks and tubes were incubated at 33⬚C for tick cell growth and then moved to 28.5⬚C to favor rickettsial growth after 3 mo, when tick cell monolayers became established (U. G. Munderloh, personal communication). Antibiotics initially added to primary cultures were discontinued after the third week, whereas amphotericin was continued. Cell medium was fully replaced for uninfected cells from the TAMU source weekly and twice a week in infected cells established from the OSU source; passages were made every other week or as needed. When conßuent, 10 Ð50% of the cells were transferred to other ßasks by scraping cells off the ßask and transferring to a ßask with new medium for a Þnal volume of 5 ml. For some ßasks, a minimum of at least half of cells in that ßask were transferred to cryovials with medium containing 10% dimethyl sulfoxide. Cry-

1120

JOURNAL OF MEDICAL ENTOMOLOGY

ovials were frozen by using a Nalgene Mr. Frosty freezing container (SigmaÐAldrich) at ⫺80⬚C overnight and then stored in an ultra-low (⫺150⬚C) freezer. Resuscitated cells from individual cryovials were Þrst diluted in 5 ml L-15B300 medium, incubated for 5 min at room temperature, and then centrifuged at 1,250 ⫻ g. Pelleted cells were resuspended in L-15B300 supplemented, as previously described, and then transferred to a 12.5-cm2 ßask. Cultures were observed for conßuency weekly by using an inverted microscope. To evaluate cultures for rickettsial infection, samples of spent medium, collected approximately every week for up to 5 mo, were centrifuged onto glass slides by using a Statspin Cytofuge (Iris Sample Processing, Westwood, MA), and “cytospin” preparations were stained with acridine orange or Diff-Quik (Dade Behring, Newark, DE). Culture samples were also tested approximately bi-monthly by PCR and quantitative PCR (QPCR) assays and once by immunoßuorescence assay following a protocol using human anti-Rickettsia rickettsii antibodies as previously described (Edwards et al. 2011) for evidence of rickettsiae. In an attempt to establish an uninfected cell line by using the OSU source of cells, two ßasks of cells were treated with tetracycline (initially 10 mg/ml and then increased to 40 mg/ml, SigmaÐAldrich) for 40 d to eliminate rickettsial infection (Simser et al. 2001). We tested for rickettsial infection within a week after initiation of tetracycline and once a week for 1 mo by stained cytospin preparations and PCR and QPCR assays of treated ßasks. Passage of “Ca. R. andeanae” to ISE6 and Vero Cell Lines. The cell line ISE6 (derived from I. scapularis embryonated eggs) (Kurtti et al. 1996; provided by U. G. Munderloh, University of Minnesota) was maintained at 33⬚C for the Þrst month and at 28.5⬚C thereafter in L15B300 medium supplemented with heatinactivated 20% fetal bovine serum and 10% tryptose phosphate broth. Vero cells were maintained at 37⬚C, 5% CO2 in MEM plus 10% fetal bovine serum. Vero and ISE6 cell cultures were challenged with ⬇0.2 ml of pelleted medium and cells acquired by centrifuging (1,250 ⫻ g at room temperature) 5 ml spent medium and scraped cells from OSU primary cultures. When challenged, both ISE6 and Vero cultures received amphotericin, at a concentration previously described for A. maculatum cultures, during the Þrst month cultures were maintained. Challenged cultures were sampled approximately weekly as previously, for up to 3 mo, to monitor infection. Molecular Analyses. For genomic DNA extraction from egg masses and cultures at various time points, we used an Illustra Tissue & Cells genomicPrep Mini Spin Kit (GE Healthcare, Piscataway, NJ). Approximately 50 mg samples of eggs were homogenized by using a sterile pestle per sample in proteinase K and lysis buffer from the kit, following the manufacturerÕs instructions for tissues. For the culture time points, at least 10 ml of spent medium per cell line were collected in 15-ml centrifuge tubes and then centrifuged at 1,250 ⫻ g for 5 min. After the supernatant was

Vol. 50, no. 5

discarded, we followed the manufacturerÕs protocol for remaining cells. For egg masses, we Þrst screened for evidence of any SFGR DNA by using a nested PCR protocol with SFGR-wide primers targeting the rickettsial outer membrane protein A (rompA) gene (Paddock et al. 2004). Positive samples were subsequently tested by PCR assay by using species-speciÞc rompA primers for “Ca. R. andeanae” and R. parkeri (Paddock et al. 2010, VarelaÐStokes et al. 2011). For culture samples taken at speciÞc time points, we only ran the species-speciÞc rompA assays. All PCR assays included a positive control of DNA extracted from cultured R. parkeri (TateÕs Hell strain) or “Ca. R. andeanae” infected Gulf coast ticks, conÞrmed previously by PCR and sequencing. Water was used as a negative control for all assays. All PCR products were electrophoresed in a 2% agarose gel; ethidium bromide was Þrst incorporated into gels before electrophoresis of products. PCR products selected for sequencing were puriÞed by using the Zymo Clean and Concentration kit (Zymo Research Corporation, Irvine, CA) and sequenced by EuroÞns MWG Operon (Huntsville, AL). Consensus sequences for “Ca. R. andeanae” were generated by ClustalX2 alignment and identiÞcation conÞrmed by using GenBank BLAST searches (http:// blast.ncbi.nlm.nih.gov/Blast.cgi). Quantitative Real-Time PCR. We evaluated relative growth of “Ca. R. andeanae” over time from samples of spent media collected from naturally infected established A. maculatum cell cultures, tetracyclinetreated A. maculatum cell cultures, ISE6 cell cultures, and Vero cell cultures. Primers Rx-190-F and Rx190-R, previously designed for detection of “Ca. R. andeanae” by QPCR (Paddock et al. 2010), were used at 300 nM concentrations in a similar Brilliant II SYBR Green (Agilent Technologies, Santa Clara, CA) assay. A template volume of 2 ␮l was used in a Þnal reaction volume of 20 ␮l. To take into account variability in rickettsial numbers in samples because of differences in number of host cells collected, we determined relative rickettsial levels as a ratio of rickettsial DNA (“CaRa”) to host gene (“HG”) DNA. For HGs, we selected primers 16S⫹2 and 16S-1 to amplify a 303-bp portion of the tick 16S mitochondrial rRNA gene (Black and Piesman 1994) in A. maculatum and ISE6 cell cultures. For Vero cell culture samples, we used primers 17 F and 106R (Tennant et al. 2011) to amplify a 90-bp portion of the glyceraldehyde-3-phosphate dehydrogenase gene (GAPDH). Ten-fold serial dilutions by using DNA template from known positive samples were included in all reactions to generate standard curves. For QPCR performed on samples from each set of cell culturesÑA. maculatum cells, ISE6 cells, and Vero cellsÑall culture samples as well as standards for host cell and “Ca. R. andeanae” DNA were set up in duplicate. The standard curve was used to evaluate reaction efÞciency and calculate relative values for the amount of rickettsial and host genes present to calculate CaRa/HG ratios. Thermal cycling parameters were adjusted slightly from Paddock et al. (2010) by empirically determining that an annealing temperature of 54⬚C was suitable to efÞciently amplify

September 2013

FERRARI ET AL.: “Candidatus RICKETTSIA ANDEANAE” AND A. maculatum

1121

Fig. 1. Cell culture cytospin preparations stained with acridine orange. Top row shows rickettsiae (tip of arrows) from “Ca. R. andeanae” infected (a) A. maculatum embryonic cells, (b) ISE6 cells, and (c) Vero cells at day 89. Bottom row represents uninfected (d) A. maculatum embryonic cells, (e) ISE6 cells, and (f) Vero cells at days 89 (A. maculatum), 196 (ISE6), and 96 (Vero). MagniÞcation 1,000⫻ under oil immersion. Scale bar is equivalent to 5 ␮m. Images were captured by using an Olympus BX41 compound microscope and Nikon DS-Fi1 5-Megapixel CCD Color Camera. (Online Þgure in color.)

rickettsial DNA at the same time as host cell DNA. Thus, ampliÞcation of “Ca. R. andeanae” occurred concurrently with ampliÞcation of the speciÞc host cell target, whether this was targeting 16S mitochondrial rRNA or GAPDH genes. A standard dissociation curve was included in the thermal cycle program, and all assays were performed by using a Stratagene MX3005P QPCR system (Agilent Technologies, Santa Clara, CA). Results Most eggs deposited by A. maculatum females contained embryos by day 25Ð30 postoviposition, as observed by daily inspections under a dissection microscope. Primary cultures from the OSU source grew slowly and took 2 mo to form a cell monolayer. Of 57 total ßasks with primary cultures from OSU source embryos or eggs shells, seven were lost to contamination (primarily fungal) within the Þrst week. After complete monolayers were observed in two ßasks, ⬃2 wk after their initiation, we were able to successfully passage cells to two additional new ßasks each. By the end of the second month of culturing OSU source cells, which included pooling selected ßasks that contained few cells, and losing three additional ßasks, 29 ßasks remained and were maintained for treatment with tetracycline, cryopreservation, and propagation of rickettsiae. Cryopreserved cells established from the supernatant of one egg mass and frozen for 2 mo could be resuscitated and were then split into two ßasks after 17 d in culture; however, another vial of cryopreserved cells did not survive. Cells from the TAMU source processed by using a protocol similar to OSU tick cells did not form a conßuent monolayer

during 3 mo nor did they appear to be growing based on microscopic observations. Because they did not thrive, tubes with TAMU cells were discarded. There was no evidence of “Ca. R. andeanae” in cultures treated with tetracycline by stained cytospins, PCR, or QPCR by 2 wk after initiation of tetracycline and for the additional three weekly time points. Regarding the isolation of “Ca. R. andeanae” from naturally infected A. maculatum ticks purchased from OSU, no bacteria were observed in the Þrst biweekly cytospin preparations that were started 2 wk after onset of the experiment. We used Diff-Quik, acridine orange staining, and indirect immunoßuorescence assay of cytospin preparations to demonstrate growth of “Ca. R. andeanae.” We Þrst identiÞed bacteria 5 wk after the initiation of cultures, when weakly stained bacteria were observed in cytospin preparations from ßasks containing cells established from egg masses of one of the OSU ticks. After 2 mo from the initiation of cultures, when rickettsia-like bacteria could be observed in cytospin preparations from the A. maculatum cells, we challenged ISE6 and Vero cell cultures. Cytospin preparations were stained weekly by using either acridine orange (Fig. 1) or Diff-Quik (Fig. 2) for almost 3 mo, alternating the staining method each week for the three cell lines. Rickettsia-like bacteria were seen intracellularly as well as extracellularly; no bacterium could be identiÞed intranuclearly. Bacteria were not consistently observed in all cytospin preparations of cell culture ßasks, and were particularly difÞcult to Þnd in cytospin preparations from ISE6 and Vero cells. Even when observed, rickettsiae were not present in high numbers (as compared with R. parkeri). Immunoßuorescence was performed once for cells derived from the OSU tick colony ⬇7 wk after

1122

JOURNAL OF MEDICAL ENTOMOLOGY

Vol. 50, no. 5

Fig. 2. Cell culture cytospin preparations stained with Diff-Quik. Top row shows rickettsiae (tip of arrows) from “Ca. R. andeanae” infected (a) A. maculatum embryonic cells, (b) ISE6 cells, and (c) Vero cells at days 47 (A. maculatum cells), 196 (ISE6), and 83 (Vero). Bottom row represents uninfected (d) A. maculatum embryonic cells, (e) ISE6 cells, and (f) Vero cells at day 196. MagniÞcation 1,000⫻ under oil immersion. Scale bar is equivalent to 5 ␮m. Images were captured by using an Olympus BX41 compound microscope and Nikon DS-Fi1 5-Megapixel CCD Color Camera. (Online Þgure in color.)

initiation of cultures. Most (6/10) OSU source cultures that were tested demonstrated ßuorescence staining characteristic of rickettsiae when compared with positive control (R. parkeri); background ßuorescence was observed for negative control cells (Fig. 3). We challenged ISE6 and Vero cells with 0.2 ml of spent medium in addition to a sample of cells that was removed by pipetting or scraping from the “Ca. R. andeanae” infected A. maculatum source ßask. After the initial challenge of ISE6 and Vero cells, we tested all three cell lines by using cytospin preparations coupled with PCRs and QPCRs. Challenges were repeated on the same cultures three (ISE6 cells) to four (Vero cells) times over 3 mo and occurred following time points when samples taken for PCR tests became

negative for rickettsiae. PCR and QPCR assays were performed for 11 and 8 time points, respectively, spread over a 3-mo period (Table 1). At least two ßasks of ISE6 and Vero cells each were kept unchallenged to use as negative controls for cytospin preparations, PCR, and QPCR. These remained negative in the cytological and molecular analyses when tested at the 11 time points. We used QPCR to compare the “Ca. R. andeanae” growth from samples of spent media collected at eight time points from naturally infected established A. maculatum cell cultures, tetracycline-treated A. maculatum cell cultures, ISE6 cell cultures, and Vero cell cultures. As shown in Fig. 4a, pooled samples from naturally infected A. maculatum cell culture ßasks were consistently positive throughout the time points.

Fig. 3. Immunoßuorescence antibody assay by using R. rickettsii antibodies and ßuorescein isothiocyanate-labeled secondary antibodies. (a) Extracellular rickettsiae (tip of arrow) in A. maculatum embryonic cells; (b) R. parkeri grown in ISE6 cells as a positive control (successful culture in ISE6 cells described by Burkhardt et al. 2011); (c) uninfected A. maculatum embryonic cell at day 56. MagniÞcation 1,000⫻, oil immersion. Scale bar is equivalent to 5 ␮m. Images were captured by using an Olympus BX41 compound microscope and Nikon DS-Fi1 5-Megapixel CCD Color Camera. (Online Þgure in color.)

FERRARI ET AL.: “Candidatus RICKETTSIA ANDEANAE” AND A. maculatum

September 2013

1123

Table 1. PCR results for “Ca. R. andeanae” species-specific rompA assay for several time points in A. maculatum cells initiated from OSU A. maculatum ticks, and from challenged ISE6 and Vero cells No. days after Þrst A. maculatum culture was initiated

Naturally infected A. maculatum cells

ISE6 cells

Vero cells

89 93 96 107 121 125 132 139 146 153

Positive Positive N/a Positive Positive Positive Positive Positive Positive Positive

Positive Positive Negative Negative Negative Positive Negative Positive Positive Negative

Positive Positive Negative Negative Positive Positive Negative Negative Negative Positive

Samples from ongoing A. maculatum cultures were used to challenge ISE6 and Vero cells on days 79, 111, and 118 following initiation of A. maculatum cultures, as well as on day 149 for Vero cells only.

Shortly after onset of treatment, antibiotic-treated A. maculatum cells were positive for rickettsial DNA but became negative at later time points. Samples tested from the single inoculated ISE6 (Fig. 4b) and Vero cell culture ßasks (Fig. 4c) differed from samples of A. maculatum cells. Whereas QPCR assays on untreated A. maculatum cultures demonstrated ßuctuations in relative rickettsial DNA during the time culture were maintained, relative rickettsial DNA levels in ISE6 and Vero cell cultures were highest at time points occurring directly after challenge. Exceptions occurred on days 139 and 146, where challenged ISE6 cells were positive by PCR and QPCR (day 146 only) despite not having been challenged since day 118 and having a negative PCR result on day 132. Thus, rickettsiae may have begun to multiply at this point. Results of QPCR and PCR assays were generally comparable. Reaction efÞciencies were 91.5 and 157 for 16S rRNA housekeeping gene targets in ISE6 and A. maculatum cultures, respectively, and 115% for the GAPDH gene target in Vero cell cultures. The efÞciencies ⬎110% suggest presence of inhibitors in the reaction. Reaction efÞciencies for detection of “Ca. R. andeanae” were 90.5, 80.3, and 84.8% in A. maculatum, ISE6, and Vero cells, respectively. R2 values were ⬎0.99 for all standard curves, with the exception of the 16S rRNA gene target of A. maculatum (0.941) and GAPDH gene target for the Vero cells (0.969). Tick DNA from two A. maculatum culture time points (days 146 and 153) was ampliÞed by PCR by using primers for a fragment of the 16 rRNA gene as described by Black and Piesman (1994) and subsequently sequenced. A BLAST search conÞrmed 100% identity with 100% coverage of our sample with A. maculatum 16S rRNA sequences deposited in the NCBI GenBank (http://blast.ncbi.nlm.nih.gov/Blast. cgi). By using “Ca. R. andeanae” rompA gene primers, we ampliÞed rickettsial DNA from one time point (day 125 after initiation of A. maculatum cultures) in the original A. maculatum cultures as well as chal-

Fig. 4. “Ca. R. andeanae” relative infection levels determined by calculating the ratio of rickettsial DNA versus host DNA by using QPCR of three different cell lines. (a) Naturally infected (diagonal lines) and tetracycline-treated (solid black) A. maculatum embryonic cells; (b) challenged (diagonal lines) and uninfected (solid black) ISE6 cells; (c) challenged (diagonal lines) and uninfected (solid black) Vero cells. Note the difference in y-axis scale for growth in Vero cells. Samples were not available for all time points. Samples from ongoing A. maculatum cultures were used to challenge ISE6 and Vero cells on days 79, 111, and 118 following initiation of A. maculatum embryonic cell cultures, as well as on day 149 for Vero cells only.

lenged ISE6 and Vero cell cultures; these amplicons were submitted for sequencing as well. By using a BLAST search, we conÞrmed that the rickettsial DNA sequences in the three cell lines were 100% identical to the “Ca. R. andeanae” rompA sequence in NCBI GenBank. Discussion In this study, we were able to propagate “Ca. R. andeanae” primary cultures of embryonic cells from naturally infected female A. maculatum. “Ca. Rickettsia andeanae” appeared to show better growth within naturally infected A. maculatum cells, as opposed to growth of passages made to ISE6 and Vero cells. Given that the source of embryonic cells was the natural tick host for “Ca. R. andeanae,” A. maculatum, and source

1124

JOURNAL OF MEDICAL ENTOMOLOGY

ticks demonstrated evidence of “Ca. R. andeanae” infection, as opposed to the Vero and ISE6 cells that usually required additional challenges based on PCR and QPCR result; these data suggest that “Ca. R. andeanae” is more stable in the A. maculatum cells. Previous attempts to isolate “Ca. R. andeanae” in Vero E6, ISE6, and C6/36 (mosquito) cells by using A. maculatum naturally infected with “Ca. R. andeanae” were unsuccessful (Paddock et al. 2010). Similarly, after several failed attempts to isolate R. peacockii, the Þrst successful isolate was obtained in D. andersoni tick cells (Simser et al. 2001). Recently, “Ca. R. andeanae” was propagated in mammalian cells (Vero and DH82) and S2 cells from Drosophila melanogaster (LuceÐFedrow et al. 2011). A stable rickettsial isolate propagated in cells from a host in which it is naturally found may provide a more realistic model for future studies. Rickettsial proteins may be differentially expressed depending on the environment in which they are grown, particularly when grown in mammalian compared with arthropod cell lines (Rovery et al. 2005, Singu et al. 2006, Nelson et al. 2008). In this study, “Ca. R. andeanae” was isolated and multiplied in cells from A. maculatum, where they were consistently detected by PCR and QPCR assays and more routinely detected by cytospin preparations. In comparison, in Vero cell cultures initially inoculated with “Ca. R. andeanae,” challenges were repeated when cytospin and PCR (and QPCR) assays failed to demonstrate rickettsiae and rickettsial DNA, suggesting that the rickettsiae were not thriving. ISE6 cells appeared to act similarly; however, they showed evidence of rickettsial DNA 3 and 4 wk after the Þnal challenge, suggesting a tick cell line, while not from the natural host, may be an appropriate co-culture for “Ca. R. andeanae.” At one time point, ⬃2 mo after initiation of naturally infected A. maculatum cultures, a peak in the ratio of “Ca. R. andeanae” to tick 16S mitochondrial DNA was seen in the QPCR assay. This peak may have been because of differences in sample collection or in the growth status of the rickettsia relative to the newly established cells. As this was early in the establishment of the primary culture, the embryonic cells may have been more tightly adhered in the monolayer and may not have detached or been lysed by replicating rickettsiae. Thus, more infected cells would be retained in the cell layer and the amount of tick 16S mitochondrial rDNA would be lower in supernatant medium, allowing the ratio to be higher than at later time points when more of the infected cells would have detached. Similarly, Vero cells are tightly adherent and rarely seen ßoating in medium as compared with ISE6 cells. Thus, low numbers of Vero cells in sampled media may have accounted for the markedly higher ratio of rickettsial DNA to host DNA in Vero cultures. Overall, we observed low numbers of “Ca. R. andeanae” in cytospin preparations and slow growth, consistent with results reported by Luce-Fedrow et al. (LuceÐFedrow et al. 2011). In addition, these authors observed “Ca. R. andeanae” in the cell cytoplasm and extracellularly, with no intranuclear bacteria, which is

Vol. 50, no. 5

comparable with our Þndings (LuceÐFedrow et al. 2011, F.A.G.F. and A.S.V.-S., unpublished data). Interestingly, the use of egg masses from infected female A. maculatum to establish the “Ca. R. andeanae” isolate supports transovarial transmission of this rickettsia. Further investigation is needed to conÞrm vertical transmission. Although we aspired to concurrently establish an A. maculatum embryonic cell line and isolate the rickettsiae naturally infecting these cells, we have not yet been able to demonstrate the former. Recently, Rickettsia raoultii was isolated from infected Dermacentor reticulatus embryos by using a similar approach, and primary cell cultures allowed ampliÞcation of the pathogenic rickettsiae, which were subsequently subinoculated into other tick cell lines (Alberdi et al. 2012). The authors observed a decrease in D. reticulatus cell viability (concurrent with an increase in R. raoultii density) over time and variable susceptibility to the rickettsiae among cell lines that were subinoculated (Alberdi et al. 2012). It may be that our system supported growth of “Ca. R. andeanae” from naturally infected embryos, but conditions were not optimal to promote propagation of the cells. One reason our A. maculatum cells (either the TAMU cells or the tetracycline-treated OSU cells) did not show evidence of thriving may have been that they were not maintained for a sufÞcient length of time to allow enough cells to multiply and attach to the ßask. The timing of embryonic development for initiation of TAMU cells, in particular, may not have been optimal. In addition, a sufÞcient amount of spent medium may not have remained in the ßasks when fresh medium was replaced weekly, thus allowing any suspended cells to be lost. These observations were noted by Moulton (1978), who was successful in preparing primary cultures of A. maculatum from preimaginal bodies of fed nymphs, and was also able to subculture these cells. A future objective is to use these tactics on cryopreserved A. maculatum cells (both tetracyclinetreated and “Ca. R. andeanae” infected) that were initiated through this study. In addition, since the completion of this study, we have observed these rickettsiae within inoculated embryonic cells of Amblyomma americanum (AAE2 cells provided by U. G. Munderloh, University of Minnesota). Thus, another objective is to determine the suitability of AAE2 cells, which share the same genus as the natural tick host for “Ca. R. andeanae,” for long-term propagation of this rickettsia. We suspect that cultivation of “Ca. R. andeanae” will be optimal in tick cells that are taxonomically closest to their natural host. The availability of “Ca. R. andeanae” isolates now provides a source of this microorganism for more in-depth studies investigating its biology, infectivity, and pathogenicity to vertebrates and future genetic characterization to aid in the species status designation. Acknowledgments We thank Ulrike Munderloh (University of Minnesota) and Christopher Paddock (Centers for Disease Control and

September 2013

FERRARI ET AL.: “Candidatus RICKETTSIA ANDEANAE” AND A. maculatum

Prevention) for intellectual input in this and other rickettsial research in our laboratory, and for critical review of this manuscript. This work was supported by intramural funding through Mississippi State University, College of Veterinary Medicine.

References Cited Alberdi, M. P., A. M. Nijhof, F. Jongejan, and L. Bell–Sakyi. 2012. Tick cell culture isolation and growth of Rickettsia raoultii from Dutch Dermacentor reticulatus ticks. Ticks Tick Borne Dis. 3: 349 Ð354. Bell–Sakyi, L., E. Zweygarth, E. F. Blouin, E. A. Gould, and F. Jongejan. 2007. Tick cell lines: tools for tick and tickborne disease research. Trends Parasitol. 23: 450 Ð 457. Black, W. C., and J. Piesman. 1994. Phylogeny of hard- and soft-tick taxa (Acari: Ixodida) based on mitochondrial 16S rDNA sequences. Proc. Natl. Acad. Sci. U.S.A. 91: 10034 Ð 10038. Blair, P. J., J. Jiang, G. B. Schoeler, C. Moron, E. Anaya, M. Cespedes, C. Cruz, V. Felices, C. Guevara, L. Mendoza, et al. 2004. Characterization of spotted fever group rickettsiae in ßea and tick specimens from northern Peru. J. Clin. Microbiol. 42: 4961Ð 4967. Burkhardt, N. Y., G. D. Baldridge, P. C. Williamson, P. M. Billingsley, C. C. Heu, R. F. Felsheim, T. J. Kurtti, and U. G. Munderloh. 2011. Development of shuttle vectors for transformation of diverse Rickettsia species. PLoS ONE 6: e29511. Dumler, J., and D. Walker. 2005. Order II. Rickettsiales, pp. 96 Ð145. In D. R. Boone, G. M. Garrity, and R. W. Castenholz (eds.), BergeyÕs Manual of Systematic Bacteriology, 2nd Part C ed. Springer, East Lansing, MI. Edwards, K. T., J. Goddard, T. L. Jones, C. D. Paddock, and A. S. Varela–Stokes. 2011. Cattle and the natural history of Rickettsia parkeri in Mississippi. Vector Borne Zoonotic Dis. 11: 485Ð 491. Fornadel, C. M., X. Zhang, J. D. Smith, C. D. Paddock, J. R. Arias, and D. E. Norris. 2011. High Rates of Rickettsia parkeri infection in gulf coast ticks (Amblyomma maculatum) and identiÞcation of “Candidatus Rickettsia Andeanae” from Fairfax County, Virginia. Vector Borne Zoonotic Dis. 11: 1535Ð1539. Jiang, J., E. Y. Stromdahl, and A. L. Richards. 2011. Detection of Rickettsia parkeri and Candidatus Rickettsia andeanae in Amblyomma maculatum gulf coast ticks collected from humans in the United States. Vector Borne Zoonotic Dis. 12: 175Ð182. Kurtti, T. J., U. G. Munderloh, T. G. Andreadis, L. A. Magnarelli, and T. N. Mather. 1996. Tick cell culture isolation of an intracellular prokaryote from the tick Ixodes scapularis. J. Invertebr. Pathol. 67: 318 Ð321. Luce–Fedrow, A., C. Wright, H. D. Gaff, D. E. Sonenshine, W. L. Hynes, and A. L. Richards. 2011. In vitro propagation of Candidatus Rickettsia andeanae isolated from Amblyomma maculatum. FEMS Immunol. Med. Microbiol. 64: 74 Ð 81. Moulton, J. E. 1978. Preparation of primary cultures of tick cells. Am. J. Vet. Res. 39: 1558 Ð1564. Munderloh, U. G., and T. J. Kurtti. 1989. Formulation of medium for tick cell culture. Exp. Appl. Acarol. 7: 219 Ð 229.

1125

Munderloh, U. G., S. D. Jauron, V. Fingerle, L. Leitritz, S. F. Hayes, J. M. Hautman, C. M. Nelson, B. W. Huberty, T. J. Kurtti, G. G. Ahlstrand, et al. 1999. Invasion and intracellular development of the human granulocytic ehrlichiosis agent in tick cell culture. J. Clin. Microbiol. 37: 2518 Ð2524. Nelson, C. M., M. J. Herron, R. F. Felsheim, B. R. Schloeder, S. M. Grindle, A. O. Chavez, T. J. Kurtti, and U. G. Munderloh. 2008. Whole genome transcription proÞling of Anaplasma phagocytophilum in human and tick host cells by tiling array analysis. BMC Genomics 9: 364. Niebylski, M. L., M. E. Schrumpf, W. Burgdorfer, E. R. Fischer, K. L. Gage, and T. G. Schwan. 1997. Rickettsia peacockii sp. nov., a new species infecting wood ticks, Dermacentor andersoni, in western Montana. Int. J. Syst. Bacteriol. 47: 446 Ð 452. Paddock, C. D., P. E. Fournier, J. W. Sumner, J. Goddard, Y. Elshenawy, M. G. Metcalfe, A. D. Loftis, and A. Varela– Stokes. 2010. Isolation of Rickettsia parkeri and identiÞcation of a novel spotted fever group Rickettsia sp. from Gulf Coast ticks (Amblyomma maculatum) in the United States. Appl. Environ. Microbiol. 76: 2689 Ð2696. Paddock, C. D., J. W. Sumner, J. A. Comer, S. R. Zaki, C. S. Goldsmith, J. Goddard, S. L. McLellan, C. L. Tamminga, and C. A. Ohl. 2004. Rickettsia parkeri: a newly recognized cause of spotted fever rickettsiosis in the United States. Clin. Infect. Dis. 38: 805Ð 811. Parker, R., G. Kohls, G. Cox, and G. Davis. 1939. Observations on an infectious agent from Amblyomma maculatum. Public Health Rep. 54: 1482Ð1484. Rovery, C., P. Renesto, N. Crapoulet, K. Matsumoto, P. Parola, H. Ogata, and D. Raoult. 2005. Transcriptional response of Rickettsia conorii exposed to temperature variation and stress starvation. Res. Microbiol. 156: 211Ð218. Simser, J. A., A. T. Palmer, U. G. Munderloh, and T. J. Kurtti. 2001. Isolation of a spotted fever group Rickettsia, Rickettsia peacockii, in a Rocky Mountain wood tick, Dermacentor andersoni, cell line. Appl. Environ. Microbiol. 67: 546 Ð552. Singu, V., L. Peddireddi, K. R. Sirigireddy, C. Cheng, U. Munderloh, and R. R. Ganta. 2006. Unique macrophage and tick cell-speciÞc protein expression from the p28/ p30-outer membrane protein multigene locus in Ehrlichia chaffeensis and Ehrlichia canis. Cell. Microbiol. 8: 1475Ð 1487. Sumner, J. W., L. A. Durden, J. Goddard, E. Y. Stromdahl, K. L. Clark, W. K. Reeves, and C. D. Paddock. 2007. Gulf Coast ticks (Amblyomma maculatum) and Rickettsia parkeri, United States. Emerg. Infect. Dis. 13: 751Ð753. Tennant, K. V., E. N. Barker, Z. Polizopoulou, C. R. Helps, and S. Tasker. 2011. Real-time quantitative polymerase chain reaction detection of haemoplasmas in healthy and unhealthy dogs from Central Macedonia, Greece. J. Small Anim. Pract. 52: 645Ð 649. Varela-Stokes, A. S., C. D. Paddock, B. Engber, and M. Toliver. 2011. Rickettsia parkeri in Amblyomma maculatum ticks, North Carolina, USA, 2009 Ð2010. Emerg. Infect. Dis. 17: 2350 Ð2353. Wright, C. L., R. M. Nadolny, J. Jiang, A. L. Richards, D. E. Sonenshine, H. D. Gaff, and W. L. Hynes. 2011. Rickettsia parkeri in gulf coast ticks, southeastern Virginia, USA. Emerg. Infect. Dis. 17: 896 Ð 898. Received 9 January 2013; accepted 10 July 2013.