1. BackgroundThe genus Thogotovirus of the family Orthomyxoviridae comprises a small group of primarily tick-associated viruses that are globally distributed. In general, Thogotoviruses have a negative-sense, RNA genome of six segments and are distantly related to the archetype virus of the family, influenza virus. Until recently, only two members of the genus that are found in the Eastern Hemisphere, Thogoto and Dhori viruses, have been associated with human infection and disease [1,2]. Prior to 2014, only one Thogotovirus, Aransas Bay virus had been identified in the United States [3,4]. Aransas Bay virus has not been associated with human disease.
In June of 2014, a previously healthy adult male from Kansas died from complications related to an illness marked by fever, thrombocytopenia and leukopenia [5]. Prior to becoming unwell, the man reported finding an engorged tick on his shoulder. An isolate was derived from the blood of this patient during the course of diagnostic testing [5]. This isolate was subsequently identified as a novel orthomyxovirus of the genus Thogotovirus by partial-genomic high throughput sequencing (HTS) and microscopy [5]. The virus was named Bourbon virus (BRBV) after the patient’s county of residence [5]. To the best of our knowledge, BRBV represents the first human pathogen of the genus Thogotovirus to be identified in the Western Hemisphere.
3. Study design3.1. BRBV isolation and purificationAn isolate of BRBV was derived directly from 200 μl of a 1:10 dilution of patient blood that was taken on day nine post onset of illness and inoculated onto confluent Vero cells in a T25 flask. The inoculated flask was then incubated at 37 °C and reviewed for cytopathic effect daily. Substantial cytopathic effect was observed at day 3 post-inoculation. The supernatant from infected cells was then harvested for further evaluations. For protein analysis, BRBV was grown in multiple flasks of confluent Vero cells and purified on a glycerol tartrate gradient as described by others [6].
3.2. High throughput sequencing and analysisBRBV RNA was extracted and purified from 0.1 mL of gradient-purified virus using the QIAGEN QIAamp kit (Qiagen) and resuspended in 0.1 mL of RNAse-free dH20. HTS was conducted using the Ion Torrent PGM system (Life Technologies) and methods described elsewhere [7]. The 5′ and 3′ terminal ends of BRBV genomic segments were determined using the Invitrogen 5′/3′ RACE System kits (Life Technologies) according to the manufacturer’s protocol, followed by capillary sequencing on an ABI 3130 instrument (Life Technologies). Sequence data generated by all methods, comprising greater than 4.5 million reads and generated from repeated runs, were analyzed using the CLC Genomics Workbench 7.5.1 (CLCbio) and NGen 4 (DNAstar) software programs, employing both de novo and templated assembly parameters. The total approximated average genome coverage across all genomic segments is 1000X. ORF determination was conducted using the EditSeq function of the Lasergene 9 package (DNAstar), and basic analyses of both nucleotide and amino acid sequences were conducted using Mega version 5 software [8].
3.3. Real-time fluorescent probe based RT-PCRReal-time primers and FAM-labeled probes were designed to detect the PB1 polymerase subunit (segment 2) and NP nucleoprotein (segment 5) ORFs using the Primer3Plus website at (http://www.bioinformatics.nl/cgi-bin/primer3plus/primer3plus.cgi) and newly derived BRBV sequence data (Table 1). Real-time assays were performed to confirm the presence of BRBV RNA in the isolate and purified BRBV preparation (derived as described above) using the QIAGEN QuantiTect Probe RT-PCR kit (Qiagen).
3.4. Phylogenetic analysesPhylogenetic analyses were conducted on BRBV and diverse Thogotoviruses that are represented in GenBank. Alignments of the deduced amino acid sequences of all major Thogotovirus ORFs were generated using the Clustal W function of MEGA version 5 software [8–10]. Maximum parsimony and neighbor-joining (NJ) trees were generated and analyzed with 2000 replicates for bootstrap testing [10,8] with highly similar topologies generated by all methods. Generated NJ trees were chosen as representative of trees that were generated by all methods (Fig. 1).
3.5. Preparation of immune seraCD-1 mice were inoculated with BRBV preparations ranging in titer from 0.1–1,000 PFU/0.1 mL inoculation using both intraperitoneal (IP) and intracranial (IC) routes of injection. All methods were conducted in accordance with institutional standards for the care and use of laboratory animals. After inoculation, mice were monitored daily for signs of morbidity and mortality. On day 24 post-inoculation (dpi), surviving mice [n = 25 (IP) and n = 24 (IC)] were bled by cardiac puncture. Resultant sera were screened by PRNT using a 90% plaque reduction threshold to confirm anti-BRBV activity and stored frozen at −80 °C for further analyses.
3.6. ImmunoblottingBRBV proteins from purified virus were separated by SDS-PAGE on a reduced 4–12% Bis/Tris gel (Invitrogen). All procedures were performed at room temperature. Proteins were blotted electrophoretically from the gels onto nitrocellulose membranes and washed for 15 min in PBS/0.1% Tween wash buffer. Non-specific binding sites were blocked with 10% goat serum/PBS for one hour while rocking. Pooled mouse immune serum, derived as described above, was diluted 1:500 in PBS and incubated with the membrane for 1 hour with gentle rocking. Membranes were washed again in PBS/0.1% Tween wash buffer three times for 5 min each. Goat anti-mouse antibody conjugated to alkaline phosphatase (Jackson Immunoresearch) was diluted 1:200 in PBS and incubated on the membrane for 1 hour with gentle rocking. Membranes were washed as described above, and BCIP/NBT phosphatase substrate (KPL) was added to the membrane until a color change appeared. The reaction was stopped by the addition of dH20.
3.7. Two-way plaque reduction neutralization assaysFor PRNT analyses, approximately 100 PFU of BRBV or reference Thogotovirus strain was incubated with serial dilutions of BRBV mouse immune serum or reference antibodies using methods described by others [11]. Incubated virus/antibody mixtures were then inoculated onto Vero cells which were subsequently overlaid with agar and incubated at 37 °C using methods described elsewhere [11]. Plaques were counted on day 4 using a 90% plaque reduction threshold. Thogotovirus reference strains and antibodies were provided by the Arbovirus Diseases Branch of the Division of Vector-Borne Diseases, Centers for Disease Control and Prevention (CDC). Species-level identities of the reference strains were confirmed by serological analyses prior to this study.
3.8. Cell culture and BRBV growth kineticsTo evaluate growth kinetics in culture, a variety of vertebrate and invertebrate cell lines were inoculated in duplicate at approximately 0.1 MOI of BRBV in T25 flasks. These cell lines included kidney cells of monkey (Vero, Vero E6 and LLC-MK2) and hamster (BHK21Cl-15) origin, human hepatocarcinoma and cervical carcinoma (HUH-7 and HeLa), duck embryo (DE), frog (XLK-WG), Aedes, Anopheles and Culex spp. mosquito (CCL-125, C6/36, C7/10, AP-61, Culex quinquefasciatus) and Dermacentor (DAE15, [12]; ANE58, [13], Amblyomma, Rh ipicephalus, and Hyalomma (AVL/CTVM17, [14]; RAE/CTVM1, [14]; HAE/CTVM9, [15] spp. tick cells. Inoculated flasks were then incubated under optimal, cell line-specific conditions. Supernatants were collected from infected cells on a daily basis ranging from day 0 to day 15 and were subjected to plaque assay [11] to determine BRBV titers expressed as PFU/mL.
4. Results4.1. HTS and phylogenetic analysesAnalyzed data from HTS reveal six RNA segments of the BRBV genome and the major proteins, generally shared among all Thogotoviruses, encoded therein (Table 2). Major ORFs include those that encode a putative PB2 polymerase subunit (segment 1), PBI polymerase subunit (segment 2), PA polymerase subunit (segment 3), GP glycoprotein (segment 4), NP nucleoprotein (segment 5), and M matrix protein (segment 6). When compared to the amino acid sequences of previously described Thogotoviruses, the translated ORFs of BRBV share percentage identities across all coding regions that range from ~24 to 82%. The percentage identities that BRBV shares with Dhori virus and its subtype, Batken virus, the two members of the genus with which BRBV shares the highest percentage identities, range from ~59% identity in the putative GP (most divergent) ORF to ~82% identity to in the putative PBI subunit (relatively conserved) ORF. Comprehensive phylogenetic analyses (Fig. 1) generally support these findings, with BRBV grouping with strong support in all analyzed segments with Dhori and Batken viruses.
4.2. Evaluation of immune seraFifty five percent (27/49) serum samples collected from mice inoculated with BRBV were determined to possess specific anti-BRBV activity by PRNT analyses. This includes most mice (85%, 17/20) that were inoculated with relatively high titers of between 100 and 1000 PFU/inoculation of BRBV by both IC and IP routes. IP inoculations resulted in a higher proportion of mice generating anti-BRBV antibodies (76%, 19/25) than those receiving IC inoculation (33%, 8/24) (P < 0.01). Of interest, mice inoculated by both routes showed no signs of morbidity (i.e. tremors, sick rodent posture, ruffled fur, dehydration, extended hunched posture, recumbency, lack of response to stimuli, signs of paralyses or limb weakness, unresponsiveness) or mortality during the 24 day post-inoculation period.
4.3. Protein and serological analysesPurified BRBV was separated as described to determine immunogenicity of the viral structural proteins. Pooled sera from mice inoculated with BRBV reacted with all three structural proteins: glycoprotein (GP) (approximately 60 KDa), nucleocapsid (NP) protein (approximately 52 KDa), and matrix (M) protein (approximately 30 KDa; Fig. 2). These results generally confirm the masses calculated from the translated putative structural protein ORFs represented in Table 2.
Results from 2-way PRNT analyses reveal that for all represented Thogotoviruses, a high degree (≥160 PRNT titer) of neutralizing activity is only seen with mouse immune serum or reference antibodies for the specific virus strain against which they were generated (Table 3). These findings indicate a low level of serological cross-reactivity among Thogotoviruses. Low levels of one-way cross-neutralization were observed for BRBV with anti-Dhori virus mouse serum (PRNT titer of 20; Table 3), which supports the phylogenetic determination of a degree of distant relatedness between these two viruses (Fig. 1).
4.4. Virus growth kineticsComprehensive analyses of the growth kinetics of BRBV are presented (Fig. 3). Of note, the highest titers of between ~107̂ and 109̂ PFU/mL of BRBV are documented in supernatants taken from hamster (CHK-Cl-15) and monkey (Vero) cell lines over the 15 day time course (Fig. 3). Also, sustained growth of BRBV in several tick cell lines is documented (Fig. 3). Specifically, BRBV was shown to replicate to high titers of ~105̂ and 107̂ PFU/mL in Hyalomma and Rhipicephalus cell lines (RAE/CTVM1 and HAE/CTVM9) with intermediate titers of ~104̂ –106̂ PFU/mL achieved in the Amblyomma cell line, AVL/CTVM17 over the 14 day time course (Fig. 3). Less robust growth of ~101̂ –105̂ PFU/.1 mL was observed in the Dermacentor cell lines, DAE15 and ANE58 (Fig. 3). Finally, weak and declining BRBV titers were derived from all evaluated mosquito cell lines (Fig. 3).
5. DiscussionTo support both research and diagnostic aims, we have conducted basic analyses of BRBV at the molecular and serological levels. Findings from these analyses definitively identify BRBV as a distinct member of the genus Thogotovirus. Phylogenetically, BRBV is most closely related to Dhori virus and its subtype, Batken virus (Fig. 1) that together have been known to occur in regions throughout Africa, Asia and Europe [16–18,1,19]. Dhori and Batken viruses have been isolated from Hyalomma ticks and antibodies to Dhori virus have been identified in camels, goats, horses, cattle and humans [16–18,1,19]. Of interest, Batken virus has also been isolated from several mosquito species [19].
Robust growth in a variety of invertebrate and vertebrate cell lines suggests a probable association between BRBV and mammalian and tick hosts (Fig. 3). In addition, relatively high titers derived from all mammalian cells are compatible with the known susceptibility of the human host to BRBV infection and disease (Fig. 3). Of note, the high titers of BRBV derived over a sustained time course from hamster cells is possibly reflective of a small mammal serving as an amplification host (Fig. 3). Certainly, the susceptibility of CD-1 mice to BRBV infection as represented by seroconversion in the absence of morbidity and mortality (see Results) is consistent with this hypothesis. As an aside, the lack of morbidity and mortality in BRBV-infected CD-1 mice indicates that they are unlikely to be useful in the development of a murine model for BRBV-related human disease. Although, more comprehensive investigations, including analyses of viral load in infected mice, are recommended to better understand the replication and pathology of BRBV in the murine host.
Growth of BRBV in cell lines derived from the hard ticks Hyalomma, Rhipicephalus and Amblyomma suggests a likely association of BRBV with a tick vector (Fig. 3). Furthermore, the geographic location of the BRBV infection discussed here is coincident with geographic distribution of a number of tick species, including Amblyomma americanum that has been associated with the human pathogens Ehrlichia chaffeensis and Heartland virus [20,21]. Although, differential speculation on species-level tick host associations is not made at this time and it is noted that no soft tick cell lines were included in this study. Of interest, Thogoto virus, the type species of the genus Thogotovirus, also grows in the HAE/CTVM9 and RAE/CTVM1 cell lines used in the present study [22]. Finally, weak and declining titers of BRBV derived from mosquito cell lines are reflective of a lack of replication and an unlikely association between BRBV and a mosquito vector (Fig. 3).
While not discussed here, BRBV was discovered as part of an epidemiological study of the relatively newly discovered, tick-associated Heartland virus [4]. The isolation of BRBV as part of the diagnostic component of this study highlights the probability that there are other, as of yet undiscovered, tick-associated pathogens within the United States and elsewhere. Taken together, the data presented herein suggest a relatively distant ancestral divergence of BRBV from other Thogotoviruses. Such divergence is suggestive of the long-term wide dispersal of Thogotoviruses globally and the plausible isolated evolution of BRBV in the Western Hemisphere. Furthermore, these findings are consistent with the hypothesis that BRBV has been circulating in the United States, undetected, for many years. Finally, it is important to note that while an association between BRBV and a tick vector has been suggested through clinical history [4], phylogenetics and in vitro characterization, the host associations and transmission dynamics of BRBV have yet to be conclusively determined through more comprehensive investigations.