Novel Identification of Dermacentor variabilis Arp2/3 Complex and Its Role in Rickettsial Infection of the Arthropod Vector

Tick-borne spotted fever group (SFG) Rickettsia species must be able to infect both vertebrate and arthropod host cells. The host actin-related protein 2/3 (Arp2/3) complex is important in the invasion process and actin-based motility for several intracellular bacteria, including SFG Rickettsia in Drosophila and mammalian cells. To investigate the role of the tick Arp2/3 complex in tick-Rickettsia interactions, open reading frames of all subunits of the protein including Arp2, Arp3, ARPC1, ARPC2, ARPC3, ARPC4, and ARPC5 were identified from Dermacentor variabilis. Amino acid sequence analysis showed variation (ranging from 25–88%) in percent identity compared to the corresponding subunits of the complex from Drosophila melanogaster, Mus musculus, Homo sapiens, and Saccharomyces cerevisiae. Potential ATP binding sites were identified in D. variabilis (Dv) Arp2 and Arp3 subunits as well as five putative WD (Trp-Asp) motifs which were observed in DvARPC1. Transcriptional profiles of all subunits of the DvArp2/3 complex revealed greater mRNA expression in both Rickettsia-infected and -uninfected ovary compared to midgut and salivary glands. In response to R. montanensis infection of the tick ovary, the mRNA level of only DvARPC4 was significantly upregulated compared to uninfected tissues. Arp2/3 complex inhibition bioassays resulted in a decrease in the ability of R. montanensis to invade tick tissues with a significant difference in the tick ovary, indicating a role for the Arp2/3 complex in rickettsial invasion of tick cells. Characterization of tick-derived molecules associated with rickettsial infection is imperative in order to better comprehend the ecology of tick-borne rickettsial diseases.


Introduction
Ticks serve as both the transmission vectors and reservoir hosts for members of the obligate intracellular spotted fever group (SFG) Rickettsia. In tick populations, horizontal transmission of rickettsiae between ticks can occur via vertebrate hosts; vertical transmission between life cycle stages and transovarial transmission facilitates maintenance of the infection in tick hosts. Some tick species such as the American dog tick, Dermacentor variabilis, are associated with horizontal transmission of pathogenic SFG Rickettsia (e.g. Rickettsia rickettsii) [1], as well as vertical transmission of organisms with limited or no pathogenicity to humans (e.g. Rickettsia montanensis) [2][3][4][5][6]. The route of transmission correlates to the pathogenicity of the SFG Rickettsia and the impact on tick fitness, yet the influence of constitutive and introduced rickettsial infections in tick populations is understudied.
Towards understanding the process of SFG Rickettsia infection of ticks, several studies have surveyed the tick biological or immune response to infection [7][8][9][10][11][12]; however, at the molecular level little is known about the process of tick cell infection by SFG Rickettsia. It is recognized that rickettsiae enter host cells through receptormediated endocytosis [13][14]. Tick-derived histone H2B was demonstrated to play a role in tick cell infection by a non-SFG species, R. felis, in a tick-derived cell line [15], corroborating findings of a role for nuclear proteins in SFG Rickettsia mammalian cell invasion [16]. More recently, dysregulation of tick-derived acatenin [17] and vacuolar-ATPase [18] were associated with rickettsial infection of tick-derived cell lines and whole organs. The host-derived molecules critical to cell infection by SFG Rickettsia have been examined in mammalian and Drosophila cells [16,[19][20][21][22]. Despite differences between host molecules associated with rickettsial entry in vertebrate and invertebrate hosts, the actinrelated protein 2/3 (Arp2/3) complex is recognized as a central molecule stimulated during the internalization of SFG Rickettsia into host cells, independent of cell origin.
The recognized central role for Arp2/3 complex in invasion for several bacterial pathogens compelled our examination of the molecular characteristics of the tick Arp2/3 complex to determine the role of the protein in SFG Rickettsia invasion of the natural tick host. Novel gene sequences for all seven subunits of the Arp2/3 complex from D. variabilis were isolated and compared to other species. Also, transcriptional profiles of the Arp2/3 complex subunits in unexposed and R. montanensis-exposed tick tissues (midgut, ovary, and salivary glands) were investigated. Additionally, to test the hypothesis that the Arp2/3 complex is important in rickettsial invasion of tick cells, biochemical inhibition assays were conducted ex vivo. The functional study of the tick Arp2/3 complex at the tissue level provides insight into the molecular mechanisms of SFG Rickettsia infection in natural vector hosts.

Ethics Statement
The animal care and use performed during the following experiments was approved by the Louisiana State University Institutional Animal Care and Use Committee (Protocol Number: 10-035).

Ticks and Tissue Recovery
Rickettsia-free D. variabilis colonies were maintained on vertebrate hosts at Louisiana State University, School of Veterinary Medicine as previously described [41]. For all bioassays, unfed or partiallyfed (4 days) unmated female ticks were washed with 1% bleach (5 min), 70% ethanol (2 min), and 1% benzalkonium chloride (5 min). The ticks were rinsed once with sterile water between each wash and rinsed three times after the final wash. After airdrying, tick midgut, ovary, and salivary glands were excised and washed in sterile phosphate buffered saline (PBS, pH 7.4). For RNA extraction, buffer RLT (QIAGEN, Germantown, MD) or TRIzol reagent (Invitrogen, Carlsbad, CA) was added; tissues were passed through 27G needles or homogenized by grinding with plastic pestles for several minutes. The lysates were immediately used or stored at 280uC. For invasion assays, each tissue was placed individually into 1.7 ml microcentrifuge tubes containing 200 ml of L15C medium supplemented with 10% fetal bovine serum (Hyclone, Waltham, MA), 5% tryptose phosphate broth (Difco, Sparks, MD), 0.1% lipoprotein-cholesterol concentrate (LPC, MP Biomedicals, Santa Ana, CA), 0.6% HEPES solution (1 M, Sigma, St. Louis, MO), and 1.2% sodium bicarbonate solution (5%, Sigma). The samples were kept on ice until used in bioassays on the same day.

Rickettsia Propagation and Tick Infection Procedures
Rickettsia rickettsii isolate Sheila Smith [42] and R. montanensis isolate M5/6 [43] were propagated in an African green monkey kidney cell line (Vero E6) cells cultured in Dulbecco's modified Eagle's medium (DMEM) high glucose (Invitrogen) containing 5% fetal bovine serum (Hyclone) and maintained in a humidified 5% CO 2 incubator at 34uC. To generate a cDNA library, ticks were infected with R. rickettsii by needle inoculation. Briefly, frozen stock of R. rickettsii infected Vero cells (,95% of the monolayer was infected) were thawed and centrifuged at 160006g for 10 min. The cell pellet was reconstituted in 500 ml PBS and an equal aliquot was used to inject five unfed female ticks at the area between Coxa I and basis capituli. The injected ticks were kept at room temperature for 1 h prior to tissue removal. For organ specific invasion assays, R. montanensis was semi-purified from host cells using a modified protocol of Weiss et al. [44] as previously described [18]. The number of rickettsiae was enumerated by counting Rickettsia stained with a LIVE/DEAD BacLight Bacterial Viability Kit (Molecular Probes, Carlsbad, CA) in a Petroff-Hausser bacterial counting chamber (Hausser Scientific, Horsham, PA) and examined with a Leica microscope (Buffalo Grove, IL) [45].

Cloning of the Tick Arp2/3 Complex Subunit Full-length cDNAs
The full-length cDNA for all seven subunits of Arp2/3 complex were identified in cDNA libraries generated from unfed (R. rickettsii-infected) or partially-fed (uninfected) D. variabilis using the SMARTer RACE cDNA Amplification Kit (Clontech, Mountain View, CA) or the GeneRacer Kit (Invitrogen), respectively, according to the manufacturers' instructions. RACE-ready cDNAs were synthesized from total or mRNA using iScript reverse transcription kit (Bio-Rad, Hercules, CA) or SuperScript III Reverse Transcriptase (Invitrogen). Both 59-and 39-end fragments of the Arp2/3 complex subunits were amplified utilizing primers as shown in Table S1. Amplicons were cloned into pCR4-TOPO vector and transformed into TOP10 E. coli (Invitrogen). The plasmids were isolated and sequenced at Louisiana State University, School of Veterinary Medicine. Sequence of DNA was analyzed using BioEdit software and similarity comparison was carried out against protein database in GenBank using BlastX. Amino acid sequence analyses were conducted using web-based software suits. Multiple sequence comparison by log-expectation (MUSCLE, http://www.ebi.ac.uk/Tools/msa/muscle/) was used to create sequence alignment files and to calculate the percent identity matrix (created by Clustal2.1). The alignment output was created using GeneDoc software. ATP binding sites were predicted using NsitePred web server [46] and the conserved regions in proteins were identified by using the Simple Modular Architecture Research Tool (SMART, http://smart.emblheidelberg.de/).

Transcriptional Analysis during Rickettsia Infection
To determine the transcriptional profiles of the Arp2/3 complex subunit genes (all subunits) in dissected D. variabilis tissues from unfed females during Rickettsia infection, tick tissues (midgut, ovary, and salivary glands) were excised and exposed to R. montanensis (8610 7 per tissue) or complete L15C medium (uninfected groups). The samples were centrifuged at 4uC, 7006g for 2 min to facilitate the binding between Rickettsia and tick tissues. Rickettsiae were allowed to infect the tissues at 32uC for 1 h. The samples were then washed twice with 1 ml PBS and collected by centrifugation at 4uC, 2756g for 4 min. While using dissecting microscope, the supernatant was removed, leaving each tissue in each tube. Three samples of the same tissues were pooled and placed in 800 ml TRIzol reagent for RNA and DNA extraction as described in the manufacturer's protocol. First-strand cDNA was then synthesized from 75 ng of DNase-treated total RNA using iScript reverse transcription kit (Bio-Rad) according to manufacturer's instruction. Quantitative PCRs (qPCRs) were then performed using gene-specific primers (Table S2) for each subunit of the DvArp2/3 complex and the housekeeping gene, glyceraldehyde-3-phosphate dehydrogenase (GAPDH). All qPCR reactions were prepared in 96-well plates in a 35 ml volume composed of 0.1 mM each forward and reverse primers, DNase/RNase-free water, 2 ml of cDNA (sample) or water (negative control) and 2X LightCycler 480 SYBR Green I Master (Roche, Indianapolis, IN). The mixtures were aliquoted in triplicate 10 ml reactions onto 384-well plates and run on LightCycler 480 system II (Roche). Quantitative PCR assay conditions consisted of a 95uC pre-incubation for 10 min, 35 amplification cycles of 95uC for 15 sec, 60uC for 30 sec, and 72uC for 5 sec followed by a melting curve step of 95uC for 5 sec and 65uC for 1 min. A no RT reaction (water was added instead of reverse transcriptase) was performed to confirm an absence of genomic DNA (gDNA). Analyses of the crossing point (Cp) ratio of target (DvArp2, DvArp3, DvARPC1, DvARPC2, DvARPC3, DvARPC4, and DvARPC5) and reference (GAPDH) gene values were conducted with LightCycler 480 (1.5.0) software (Roche) using Basic Relative Quantification analysis (DDCT-Method; Roche). Data are presented as the ratio of a target cDNA sequence to a reference cDNA sequence. To confirm the infection of tissues in the assays, DNA was extracted from the same samples after RNA isolation. Copies of rickettsial outer membrane protein B gene (RmOmpB) were quantified using qPCR as previously described [18]. The infection experiments were performed twice independently.

Arp2/3 Complex Inhibition Assays
A whole organ infection bioassay was developed based on a modified protocol of Bell [47]. Briefly, tick tissues including midgut, ovary, and salivary glands, placed individually in 1.7 ml microcentrifuge tubes, were treated with 500 mM CK-666 (EMD Millipore, Billerica, MA), an Arp2/3 complex inhibitor that binds between Arp2 and Arp3 subunits to prevent the complex's ability to nucleate actin, and incubated at 32uC. After 3 h, R. montanensis (8610 7 per tissue) was used to infect tick tissues for 1 h. To remove excess extracellular rickettsiae, the tissues were then washed twice with 1 ml PBS and collected by low-speed centrifugation as described above. Genomic DNA was then extracted from the samples using the DNeasy Blood & Tissue Kit (QIAGEN) and eluted with 35 ml DNase/RNase free water. The numbers of rickettsiae and tick cells were then quantified using probe-based qPCR as previously described [18]. The experiments were performed in quadruplicate for each treatment group and the results were the combination of the three independent experiments.

Statistical Analysis
Analysis of Variance (ANOVA) was conducted using the SAS statistical package (Version 9.3) GLM procedure. For transcriptional analysis, relative gene expression was analyzed using a twoway interaction (rickettsial infection and tick tissues). Pairwise t tests of least-squares means were used to examine the interaction effects of relative mRNA expression of each subunit of the DvArp2/3 complex between Rickettsia-exposed and -unexposed tissues or between tissues (midgut, ovary, salivary gland). For biochemical inhibition assays, the same tests were used to study a role of DvArp2/3 complex during rickettsial invasion of tick tissues. P-values of #0.05 were considered significantly different.

Cloning and Sequence Analysis of DvArp2/3 Complex Subunits
Full-length cDNA clones corresponding to the transcript of DvArp2/3 complex subunit genes (DvArp2, DvArp3, DvARPC1, DvARPC2, DvARPC3, DvARPC4, and DvARPC5) from D. variabilis were isolated. The GenBank accession numbers, open reading frame (ORF) lengths, number of deduced amino acid sequences, and estimated molecular weights (MW) of each of the DvArp2/3 complex subunits are shown in Table 1.
Amino acid sequence analyses of DvArp2/3 complex subunits were performed using a web-based multiple sequence alignment (MUSCLE) and the percent identity compared to the corresponding subunits of the Arp2/3 complex from Drosophila melanogaster, Mus musculus, Homo sapiens, and Saccharomyces cerevisiae are shown in Table 2. For each subunit the similarity ranged from 25-88%. Because Arp2 and Arp3 bind to ATP, the proteins were analyzed for ATP binding sites using NsitePred web server. Putative ATPbinding sites were identified for both Arp2 (Figure 1, underlined) and Arp3 (Figure 2, underlined) molecules, suggesting conserved activity among homologs. As shown in Figure 3, five putative WD (Trp-Asp) motifs which are conserved domains in ARPC1 protein [48], were also identified in the ARPC1 subunit from D. variabilis. Alignments for the remaining subunits, DvARPC1, DvARPC2, DvARPC3, DvARPC4, and DvARPC5 are provided in Figures S1-S5.

Expression of DvArp2/3 Complex Subunit mRNAs in Tick Tissues Infected Ex vivo
To define the transcriptional profiles of the DvArp2/3 complex (all subunits) in D. variabilis tissues (midgut, ovary, and salivary glands) in response to R. montanensis infection, tick tissues were dissected out of the ticks and exposed to rickettsiae. Transcriptional activity of DvArp2, DvArp3, DvARPC1, DvARPC2, DvARPC3, DvARPC4, and DvARPC5 mRNA were measured by quantitative reverse-transcription (qRT)-PCR. The mRNA of all DvArp2/3 complex subunits was detectable in all tick tissues, and in both R. montanensis-exposed and -unexposed tissues ( Figure 4). Interestingly, the mRNA levels were expressed at a greater level in the ovary compared to the midgut and salivary glands with significant differences for DvArp3 (P = 0.0496 in uninfected ovary compared to midgut; P = 0.0031 and 0.0105 in infected ovary compared to midgut and salivary glands, respectively), DvARPC4 (P = 0.0217 and 0.0270 in uninfected ovary compared to midgut and salivary glands, respectively; P,0.0001 and P = 0.0012 in infected ovary compared to midgut and salivary glands, respectively), and DvARPC5 (P,0.0001 in uninfected ovary compared to both midgut and salivary glands; P,0.0001 in infected ovary compared to both midgut and salivary glands). The transcription of DvARPC4 was significantly (P = 0.0311) upregulated in response to R. montanensis infection in the ovary, compared to uninfected tissues.
To confirm the infection of tick tissues in the assays, DNA was extracted from the same samples after RNA isolation and the copies of the rickettsial gene (RmOmpB) in infected tissues were quantified by qPCR. The average numbers of invading Rickettsia from two independent experiments are 1.56610 4 , 1.09610 4 , and 1.93610 4 , in midgut, ovary, and salivary glands, respectively.

DvArp2/3 Complex Inhibition Assay
To further characterize the Arp2/3 complex in rickettsial infection of a vector host, an inhibition assay was performed in tick tissues. Midgut, ovary, and salivary glands were recovered from unfed female ticks and treated with 500 mM CK-666, an Arp2/3 complex inhibitor, for 3 h. R. montanensis was then used to infect the tissues (8610 7 per tissue) for 1 h, and the tissues were washed twice with PBS to remove extracellular rickettsiae. Genomic DNA was then extracted from the samples and the number of invading rickettsiae and tick cells were quantified by qPCR. Compared to inhibitor vehicle alone, the presence of CK-666 influenced rickettsial invasion by decreasing the number of rickettsiae entering the cells by as much as 70%. As shown in Figure 5, inhibition of DvArp2/3 complex resulted in a decrease in R. montanensis invasion of all tissues with significant difference (P = 0.0477) in the ovary.

Discussion
The Arp2/3 complex is a seven-subunit protein actin nucleator widely expressed in eukaryotic cells. In order to invade host cells, several bacterial pathogens, including SFG Rickettsia [16,21,[49][50], exploit the host Arp2/3 complex. For these tick-borne bacteria, this interactive process has been examined primarily in vitro with model systems without assessing the utility of the Arp2/3 complex in SFG Rickettsia infection in ticks. The current study provides the first molecular description of host machinery and utilization by rickettsiae in a competent vector host.
The current study provides the molecular and functional characterization of the Arp2/3 complex from D. variabilis, a competent vector of SFG Rickettsia. Full-length cDNAs encoding all seven subunits of the protein (DvArp2, DvArp3, DvARPC1, DvARPC2, DvARPC3, DvARPC4, and DvARPC5) were isolated, and multiple sequence alignments showed variation in percent identity compared to the corresponding subunits of the complex from D. melanogaster, M. musculus, H. sapiens, and S. cerevisiae. Although DvARPC1 is one of the more divergent subunits, conserved putative WD domains of ARPC1 [48] were observed in ARPC1 isolated from D. variabilis. The WD repeat, also known as the Trp-Asp or WD40 motif, is involved in a wide variety of cellular processes such as RNA processing, signal transduction, cytoskeleton assembly, and macromolecular protein complex formation [51][52]. Welch and colleagues [48] suggested the ARPC1 subunit influences assembly and maintenance of the Arp2/3 complex structure correlating with the capability of WD motif containing proteins in the coordination of multiprotein complexes. It was also postulated that ARPC1 facilitated the binding of the Arp2/3 complex with proteins that regulate its functions [48]. Furthermore, amino acid sequence analysis of DvArp2 and DvArp3 revealed putative ATP binding sites consistent with studies demonstrating that ATP binding on Arp2 and Arp3, as well as ATP hydrolysis on Arp2, were required for Arp2/3 complex-mediated actin cytoskeleton rearrangement [25][26][27][28][29]. Identification of the Arp2/3 complex subunits and the conserved nature of active subunits suggests ticks have a viable Arp2/3 complex. Toward functional characterization in ticks, transcriptional profiles of DvArp2/3 complex subunits were examined in both Rickettsia-infected and -uninfected tick tissues. The results indicate mRNAs of all subunits are expressed at greater levels in the tick ovary (both in Rickettsia-infected and -uninfected ovary) than in midgut and salivary glands with significant difference for DvArp3 (in uninfected ovary compared to midgut only and in infected ovary compared to both midgut and salivary glands), DvARPC4, and DvARPC5. The abundant expression of DvArp2/3 complex transcripts implies an important role of this molecule in the tick ovary. In Drosophila, the Arp2/3 complex is essential for oogenesis; Hudson and Cooley [53] demonstrated arp3 and arpc1 mutants inhibit germ line nurse cells from transporting the cytoplasmic contents to the oocytes. The increased activity in the tick ovary is intriguing as SFG Rickettsia are vertically maintained in tick populations via transovarial transmission. If infection of the ovary results in increased Arp2/3 complex activity and, similar to Drosophila, Arp2/3 activity is associated with successful oogenesis, then a possible beneficial fitness effect might be associated with rickettsial infection in the tick system. The role of the Arp2/3 complex in the tick ovary relative to oogenesis requires thorough examination.
The Arp2/3 complex is important in the regulation of actin polymerization, a key process exploited by SFG Rickettsia to invade host cells, including Drosophila and mammalian cells [16,21,[49][50]. During rickettsial invasion, the host cell Arp2/3 complex is activated by RickA, a nucleation-promoting factor mimic expressed by Rickettsia [50,54]. Recently, transformed Rickettsia bellii that overexpressed Rickettsia monacensis rickA were internalized quicker than negative control transformants [55] suggesting a role for the Arp2/3 complex in rickettsial entry. Of the seven subunits of the Arp2/3 complex, studies have demonstrated that ARPC4 is the most crucial subunit for the complex assembly in vertebrates and S. cerevisiae [56][57]. In the current study, the mRNA level of DvARPC4 was shown to be significantly upregulated in response to the early stage of R. montanensis infection of the tick ovary implying the importance of this subunit during host cell invasion. However functions of the individual Arp2/3 complex subunits need to be confirmed at protein level. Because tick tissues including midgut, ovary, and salivary glands are essential for both horizontal and vertical transmission of SFG Rickettsia [58], the role    The asterisk denotes a significant difference between treatment groups (unexposed-or Rickettsia-exposed group) in the same tissue. For each subunit, different letters above bars represents significance differences between tissues. doi:10.1371/journal.pone.0093768.g004 of the DvArp2/3 complex was further studied at the protein level during R. montanensis infection of D. variabilis. Using an ex vivo bioassay, a decrease in percent relative rickettsial invasion was observed in all tick tissues treated with CK-666, a specific chemical inhibitor of the Arp2/3 complex [59]. When compared to untreated, control tissues, a significant decrease was realized in the tick ovary. The lack of complete abolition of invasion was not observed in CK-666-treated cells likely due to multiple factors including the inability for the inhibitor to reach every cell in the organ explants or, possibly, the rickettsiae use an alternate mechanism for infection. Compared to other studies using CK-666, inhibition of rickettsial infection of host cells is typically not 100% [21]. Thus, both transcriptional dysregulation and protein function suggest an essential role for the Arp2/3 complex during rickettsial invasion of tick tissues. As a multifunctional protein, the Arp2/3 complex is also found to be important in actin-based motility of intracellular pathogens. For example, L. monocytogenes and S. flexneri express surface proteins that either mimic or activate host nucleation-promoting factors leading to the stimulation of the Arp2/3 complex and subsequent actin tail assembly and organization at the bacterial surface [40]. However, the importance of the complex in Rickettsia movement has been debated in the last decade [14,50,[54][55][60][61][62][63][64]. For example, in vitro studies utilizing Rickettsia conorii [50] and R. rickettsii [54] demonstrated that the activation of Arp2/3 complex by RickA facilitated actin nucleation and the organization of Ybranched actin networks. The roles for Arp2/3 complex in actin nucleation and Y-branched filament formation were proposed to be involved in an early stage of rickettsial movement [54]. In contrast, a knock-down of Arp2/3 complex subunits in a nonvector Drosophila cell model had only moderately impacted the length of R. parkeri actin tail formation, suggesting a non-essential role of the molecule in actin-based motility in Drosophila [64]. Further studies to investigate the role of the Arp2/3 complex in SFG Rickettsia movement in a vector host are required.
In summary, the present study provides the first description of all seven subunits of the tick-derived Arp2/3 complex and assigns a novel role for the protein in facilitating the uptake of Rickettsia into specific tick tissues. The current study also highlights several findings of importance; the mRNA level of the individual Arp2/3 complex subunits was expressed at a greater level in the ovary (both in Rickettsia-infected and -uninfected ovary) compared to the midgut and salivary glands. Likewise, DvARPC4 mRNA was significantly upregulated in response to rickettsial invasion of the tick ovary, and inhibition of the DvArp2/3 complex significantly decreased the entry of Rickettsia into the tick ovary. Further characterization of tick Arp2/3 complex is required for better understanding the precise mechanisms of the complex in rickettsial infection of arthropod vectors. Alternate inhibitions assays using CK-548, an Arp2/3 complex inhibitor specifically acting on the Arp3 subunit, or siRNA of individual subunits will allow a detailed analysis of the role and function of individual subunits of the Arp2/3 complex in the arthropod vector. Building upon the findings of the current study, the interaction between the Arp2/3 complex and SFG Rickettsia in regards to transmission by ticks requires further study.