Pathogens transmitted by ticks frequently require a period of reactivation and/or replication prior to being transmitted from an infected tick to a naive host. This phenomenon is an adaptation by tick-borne agents to the many months that may pass between tick blood meals,1 and has been well documented for the agents of Rocky Mountain spotted fever (Rickettsia rickettsii), Lyme disease (Borrelia burgdorferi), human babesiosis (Babesia microti), and human granulocytic ehrlichiosis (Anaplasma phagocytophilum).1–4 In response to the physiologic changes that accompany prolonged tick feeding, genes responsible for pathogen replication and/or migration through the hemocoel and into the salivary glands are differentially expressed,5,6 resulting in the transition of a pathogen from a dormant state to a fully infectious form. As a result, a variable period exists (generally 12–48 hours) between tick attachment and pathogen transmission. Tick-borne viruses lack the complex genetic and physiologic machinery that typically allow tick-borne bacterial and parasitic agents to enter into, and emerge from, a period of metabolic inactivity between tick blood meals. They may therefore differ from more complex organisms in the fundamental mechanisms of transmission, including the duration of the reactivation period, if any, required for transmission to occur. Powassan (POW) virus (Flaviviridae, Flavivirus) strains belonging to Lineage II7 (also referred to as deer tick virus8–10), appear to circulate in nature in an enzootic cycle with deer ticks (Ixodes scapularis) serving as the main vector.8 Human residents of and visitors to sites infested with deer ticks may thus be exposed to POW virus-infected ticks. We therefore determined the duration of tick attachment required for POW virus transmission using nymphal deer ticks and a mouse model of POW virus infection.
The POW virus strain DTV-SPO was isolated from adult deer ticks collected near Spooner, Wisconsin in 1997,9 and assigned to Lineage II on the basis of nucleotide and protein sequence data.7 The virus stock had been subjected to one suckling mouse passage and two passages on African green monkey kidney (Vero) cells. The POW virus-free deer ticks used in these experiments were derived from adults collected near Albany, New York in 2001. Infected nymphs were generated by feeding larval ticks on six-week old, female BALB/c mice that were inoculated by the intraperitoneal route with 105 plaque-forming units (PFU) of DTV-SPO 24 hours following tick attachment. Engorged larvae were collected following detachment, a sample was assayed for the presence of POW viral RNA and viral load, and the remainder was held at 22°C and a relative humidity of 95% within the Biosafety Level 3 insectary at the Wadsworth Center Arbovirus Laboratories. The RNA was isolated from individual ticks using RNeasy spin columns (Qiagen, Valencia CA) as directed by the manufacturer. Viral RNA was detected using a quantitative real-time (TaqMan assay; Applied Biosystems, Foster City, CA) reverse transcriptase–polymerase chain reaction (RT-PCR) using primers (forward: 5′-gtgactggctatttgagcacctt-3′), reverse: 5′-tggatctaaccttcgctatgaattc-3′) designed to amplify an 83-basepair region of the POW virus nonstructural protein 5 coding region, and a probe (5′-6FAM-cgagccagggtga-MGBNFQ-3′) designed to specifically bind all known linage II POW virus strains. Viral load was estimated by the TaqMan assay as described earlier using an equation derived from a standard curve of samples with known virus titer (100.3, 102.3, 104.3, and 106.3 PFU/0.1 mL). Eighteen of 20 (90%) engorged larvae detaching from viremic BALB/c mice had detectable POW viral RNA. The mean estimated titer in engorged larvae at the time of detachment was 101.07 PFU per tick (SD = 0.25) (Figure 1A). To determine the optimal number of putatively infected nymphal ticks to use in each transmission experiment, tick homogenates were screened for the presence of viral RNA. Forty-six (20%) of 235 nymphs that had fed on viremic mice as larvae contained detectable POW viral RNA, and the mean estimated titer in these ticks was 103.15 PFU per tick (SD = 1.17) (Figure 1A). Nymphal ticks were used in transmission experiments 4–8 weeks after molting.
To perform transmission experiments, eight putatively infected nymphal deer ticks were placed on six-week-old female BALB/c mice held within a plexiglas restraining cage (PlasLabs, Lansing, MI) and allowed to attach and feed for 15, 30, 60, or 180 minutes. At stated time intervals, mice were removed from restraining cages and all attached ticks were removed to individual tubes. The number of infected ticks that had fed on each mouse was determined and viral load in these ticks was estimated by the TaqMan assay as described earlier. Unattached ticks were discarded and in all cases every tick that had been placed on a mouse was accounted for. Infection in mice was monitored using clinical, molecular, and serologic criteria. Visual inspection of mice for clinical signs (ruffled fur, dramatic weight loss, weakness, ataxia, and hind limb paralysis) was performed three times a day. Mice that became moribund, ataxic, or developed hind limb paralysis were killed and brains and spleens were assayed for POW viral RNA using the TaqMan assay as described earlier. Symptom onset was variable, ranging from 5 to 11 days following tick attachment. Euthanized mice with POW viral RNA in the brains and/or spleens were considered infected. Mice that did not meet criteria for euthanizating were held for 14 days and assayed for POW virus-specific antibodies using a plaque-reduction neutralization test with rhesus monkey kidney (LLC-MK2) cells and POW virus strain 64-70627 essentially as described.11 Mice with a four-fold or greater increase from pre-exposure to post-exposure in 90% neutralization titers were considered to have become infected.
The infection status of any particular tick was unknown at the time transmission experiments were conducted. To account for greater than one (or zero) infected ticks attaching to a single mouse, a minimum transmission index was computed by dividing the number of mice that became infected by the total number of infected ticks that had attached. A mouse was considered exposed if at least one infected nymph attached. Statistical analyses were preformed using GraphPad Prism version 4 (GraphPad Software, Inc., San Diego, CA). Twenty-three mice were exposed to the bites of 31 POW virus-infected ticks, and 22 of these animals (96%) became infected. The animal that did not become infected was fed upon for 15 minutes by two infected ticks with titers of 103.59 and 103.92. It displayed no clinical signs and had not developed POW virus-specific neutralizing antibodies at the time of euthanizating, 14 days post-attachment. Powassan virus was efficiently transmitted by nymphal ticks that fed for as few as 15 minutes (Table 1). All exposed mice that were fed upon by infected ticks for at least 30 minutes became infected. Infections in mice to which only one infected tick had attached permitted examination of the relationship between tick viral load and transmission (Figure 1B). Viral load in these ticks ranged from 101.57 to 104.14 and were generally representative of the range of titers observed in all infected nymphs (Figure 1). To determine whether viral titer increased in response to host attachment, POW viral titers in nymphs that had attached to mice were also estimated. Viral titers were variable in ticks that had been allowed to attach to mice for 15, 30, 60, and 180 minutes, ranging from 101.34 through 104.14, but the mean titer and ranges at each time point were similar (P = 0.1445, by Kruskal-Wallis test) (Figure 1B).
Comparatively little work exists in the published literature on the basic interactions that occur between tick-borne encephalitis (TBE) complex flaviviruses and their tick vectors. Published studies have tended to use model systems that may not be representative of natural transmission cycles. In one study, adult ticks were inoculated with TBE virus, circumventing the requirements for infection of the midgut epithelium and transstadial transmission.12 In another study,13 the tick species examined (Dermacentor andersoni) is of secondary importance to POW virus maintenance and transmission to humans.14–16 Since the first demonstration of POW virus transmission by deer ticks,17 it has been recognized that 1) this virus exists in North America as a genetically diverse population consisting of two major lineages,7,18,19 and 2) one of these lineages perpetuates in nature in a deer tick-white footed mouse (Peromyscus leucopus) cycle.8 Furthermore, it has been shown that the separation of POW virus into two lineages has occurred as a result of natural selection, presumably due to reliance on taxonomically divergent tick hosts:7 POW virus Lineage I, which includes the prototype LB strain, is maintained by Ix. (Pholeoixodes) cookei, while Lineage II is maintained by Ix. scapularis. Arthropod species– and virus strain–dependent differences in the dynamics of intra-arthropod infection with several flaviviruses are well described.20–22 Accordingly, these trials focused on the ecologically homologous virus-vector pair of POW virus Lineage II and deer ticks.
We observed dynamics of virus infection within the tick that differ from previous reports. Ninety percent of larvae engorging on viremic mice became infected with POW virus, and the mean titer was approximately 10 PFU per engorged larva. Costero and Grayson17 observed that a smaller proportion (10%) of larval ticks became orally infected following feeding on a viremic animal, but titers in these larvae were approximately three log10 PFU higher than those observed here. The differences in efficiency of oral infection may be due to more sensitive methods applied in this study (RT-PCR versus suckling mouse inoculation), the ability of the RT-PCR to detect nonviable RNA in engorged larval ticks, or to the existence of POW virus in the tick blood meal that failed to initiate a productive infection. The strikingly lower titers observed in engorged larvae in our study are more difficult to explain, but may be due to differences in the viral strain and/or tick population used.
Transstadial transmission was relatively inefficient in our system: 20% of newly ecdysed nymphs that had fed as larvae on viremic mice retained infection, indicating an approximately 22% efficiency of transstadial transmission. This is lower than what has been reported in studies of Lineage I POW virus,17 but consistent with estimates obtained in preliminary experiments performed in conjunction with the original description of Lineage II POW virus.10 The mean titer observed in flat nymphs was approximately one log10 PFU lower than observed by Costero and Greyson,17 but similar to that observed by investigators using a POW virus Lineage I-Dermacentor andersoni system.13 Overall, these observations indicate that Lineage II POW virus efficiently infects larval deer ticks and is transmitted transstadially to approximately 20% of the resulting nymphs. Viral replication appears to occur during the metamorphosis that occurs between larval and nymphal stages of these ticks, resulting in an approximately two log10 PFU increase in virus titer between larval detachment and nymph emergence. Attachment to a vertebrate host does not appear to result in an increase in virus titer within a time frame relevant for virus transmission, although it may be that POW virus replication would occur in nymphs that had attached for longer periods of time.
The requirement for a period of reactivation prior to transmission of several tick-borne agents has been described as an adaptation to persistence within tick vectors for the several months that may pass between episodes of tick feeding.1 This phenomenon is of public health importance in that infected ticks that have attached to humans may be removed within 12–48 hours with minimal risk of transmission. In studies using TBE virus and Ix. persulcatus, TBE virus was deposited in the cement plug of these ticks within one hour of attachment,12 suggesting that a similar phenomenon may not exist for tick-borne viruses. Similarly, POW virus is transmitted efficiently within 15 minutes of attachment. It may be that virus transmission occurred much earlier: 15 minutes represents a maximum duration of attachment since the timer was started once all experimental ticks were placed on the mouse’s fur. Consequently, no comparable grace period exists between the time of tick attachment and virus transmission in this model transmission system. The public health implications of this finding are unclear at present: although the prevalence of POW virus in deer ticks is high in certain locations,9 human infections are apparently rare: four cases occurring over a two-year period have been described as an outbreak.23 Although initial reports speculated that Lineage II POW virus strains may be less pathogenic than Lineage I strains,9 subsequent work indicates that Lineage II POW virus may cause illness in humans.18 The prevalence of exposure to POW virus among human residents of deer tick-infested areas, and the relationship between viral inoculum and POW viral pathogenesis remain poorly described. It may be that a large viral inoculum, delivered over several hours or days, is required to produce illness in humans. Powassan virus infection of tick salivary glands has been described13 and along with the findings reported here suggest that infectious POW virus is present in tick salivary secretions inoculated during the earliest stages of feeding, and may be immediately inoculated.
Minimum transmission indices of ticks following attachment to mice
|Duration of attachment (minutes)||No. of mice exposed||No. of infected mice||No. of infected ticks||Minimum transmission|
Authors’ addresses: Gregory D. Ebel and Laura D. Kramer, Arbovirus Laboratories, Wadsworth Center, New York State Department of Health, 5668 State Farm Road, Slingerlands, NY 12159 and Department of Biomedical Sciences, School of Public Health, The University at Albany, State University of New York, Albany, NY, 12208, Telephone: 518-862-4311, Fax: 518-869-4530, E-mail:
Acknowledgment: We thank Sam Telford for thoughtful comments on the manuscript. The animal experiments were performed in accordance with the Wadsworth Center Institutional Animal Care and Use Committee protocol #00-355.
Financial support: This work was supported by the Wadsworth Center, New York State Department of Health.
Katavolos P, Armstrong PM, Dawson JE, Telford SR III, 1998. Duration of tick attachment required for transmission of granulocytic ehrlichiosis. J Infect Dis 177 :1422–1425.
Spencer RR, Parker RR, 1923. Rocky Mountain spotted fever: infectivity of fasting and recently fed ticks. Public Health Rep 38 :333–339.
Piesman J, Mather TN, Sinsky RJ, Spielman A, 1987. Duration of tick attachment and Borrelia burgdorferi transmission. J Clin Microbiol 25 :557–558.
Piesman J, Spielman A, 1980. Human babesiosis on Nantucket Island: prevalence of Babesia microti in ticks. Am J Trop Med Hyg 29 :742–746.
Karakashian SJ, Rudzinska MA, Spielman A, Lewengrub S, Piesman J, Shoukrey N, 1983. Ultrastructural studies on sporogony of Babesia microti in salivary gland cells of the tick Ixodes dammini.Cell Tissue Res 231 :275–287.
Ribeiro JM, Mather TN, Piesman J, Spielman A, 1987. Dissemination and salivary delivery of Lyme disease spirochetes in vector ticks (Acari: Ixodidae). J Med Entomol 24 :201–205.
Ebel GD, Campbell E, Goethert HK, Spielman A, Telford SR, 2000. Enzootic transmission of deer tick virus in New England and Wisconsin sites. Am J Trop Med Hyg 63 :36–42.
Ebel GD, Foppa I, Spielman A, Telford SR, 1999. A focus of deer tick virus transmission in the northcentral United States. Emerg Infect Dis 5 :570–574.
Telford SR III, Armstrong PM, Katavolos P, Foppa I, Garcia AS, Wilson ML, Spielman A, 1997. A new tick-borne encephalitis-like virus infecting New England deer ticks, Ixodes dammini.Emerg Infect Dis 3 :165–170.
Lindsey HS, Calisher CH, Matthews JH, 1976. Serum dilution neutralization test for California group virus identification and serology. J Clin Microbiol 4 :503–510.
Alekseev AN, Burenkova LA, Vasilieva IS, Dubinina HV, Chunikhin SP, 1996. Preliminary studies on virus and spirochete accumulation in the cement plug of ixodid ticks. Exp Appl Acarol 20 :713–723.
Chernesky MA, McLean DM, 1969. Localization of Powassan virus in Dermacentor andersoni ticks by immunofluorescence. Can J Microbiol 15 :1399–1408.
Artsob H, Spence L, Surgeoner G, McCreadie J, Thorsen J, Th’ng C, Lampotang V, 1984. Isolation of Francisella tularensis and Powassan virus from ticks (Acari: Ixodidae) in Ontario, Canada. J Med Entomol 21 :165–168.
Artsob H, 1989. Powassan encephalitis. Monath T, ed. The Arboviruses: Epidemiology and Ecology. Boca Raton, FL: CRC Press, 29–49.
Costero A, Grayson MA, 1996. Experimental transmission of Powassan virus (Flaviviridae) by Ixodes scapularis ticks (Acari:Ixodidae). Am J Trop Med Hyg 55 :536–546.
Kuno G, Artsob H, Karabatsos N, Tsuchiya KR, Chang GJ, 2001. Genomic sequencing of deer tick virus and phylogeny of Powassan-related viruses of North America. Am J Trop Med Hyg 65 :671–676.
Beasley DW, Suderman MT, Holbrook MR, Barrett AD, 2001. Nucleotide sequencing and serological evidence that the recently recognized deer tick virus is a genotype of Powassan virus. Virus Res 79 :81–89.
Sardelis MR, Turell MJ, Dohm DJ, O’Guinn ML, 2001. Vector competence of selected North American Culex and Coquillettidia mosquitoes for West Nile virus. Emerg Infect Dis 7 :1018–1022.
Armstrong PM, Rico-Hesse R, 2001. Differential susceptibility of Aedes aegypti to infection by the American and Southeast Asian genotypes of dengue type 2 virus. Vector Borne Zoonotic Dis 1 :159–168.
Boromisa RD, Rai KS, Grimstad PR, 1987. Variation in the vector competence of geographic strains of Aedes albopictus for dengue 1 virus. J Am Mosq Control Assoc 3 :378–386.