Flavivirus Infection of Ixodes scapularis (Black-Legged Tick) Ex Vivo Organotypic Cultures and Applications for Disease Control

ABSTRACT Ixodes scapularis ticks transmit many infectious agents that cause disease, including tick-borne flaviviruses (TBFVs). TBFV infections cause thousands of human encephalitis cases worldwide annually. In the United States, human TBFV infections with Powassan virus (POWV) are increasing and have a fatality rate of 10 to 30%. Additionally, Langat virus (LGTV) is a TBFV of low neurovirulence and is used as a model TBFV. TBFV replication and dissemination within I. scapularis organs are poorly characterized, and a deeper understanding of virus biology in this vector may inform effective countermeasures to reduce TBFV transmission. Here, we describe short-term, I. scapularis organ culture models of TBFV infection. Ex vivo organs were metabolically active for 9 to 10 days and were permissive to LGTV and POWV replication. Imaging and videography demonstrated replication and spread of green fluorescent protein-expressing LGTV in the organs. Immunohistochemical staining confirmed LGTV envelope and POWV protein synthesis within the infected organs. LGTV- and POWV-infected organs produced infectious LGTV and POWV; thus, the ex vivo cultures were suitable for study of virus replication in individual organs. LGTV- and POWV-infected midgut and salivary glands were subjected to double-stranded RNA (dsRNA) transfection with dsRNA to the LGTV 3′ untranslated region (UTR), which reduced infectious LGTV and POWV replication, providing a proof-of-concept use of RNA interference in I. scapularis organ cultures to study the effects on TBFV replication. The results contribute important information on TBFV localization within ex vivo I. scapularis organs and provide a significant translational tool for evaluating recombinant, live vaccine candidates and potential tick transcripts and proteins for possible therapeutic use and vaccine development to reduce TBFV transmission.

IMPORTANCE Tick-borne flavivirus (TBFV) infections cause neurological and/or hemorrhagic disease in humans worldwide. There are currently no licensed therapeutics or vaccines against Powassan virus (POWV), the only TBFV known to circulate in North America. Evaluating tick vector targets for antitick vaccines directed at reducing TBFV infection within the arthropod vector is a critical step in identifying efficient approaches to controlling TBFV transmission. This study characterized infection of female Ixodes scapularis tick organ cultures of midgut, salivary glands, and synganglion with the low-neurovirulence Langat virus (LGTV) and the more pathogenic POWV. Cell types of specific organs were susceptible to TBFV infection, and a difference in LGTV and POWV replication was noted in TBFV-infected organs. This tick organ culture model of TBFV infection will be useful for various applications, Another research area in which tick organ cultures might be valuable tools is the knockdown of tick gene expression by RNA interference (RNAi). For example, soaking Amblyomma americanum midgut sections and salivary glands with double-stranded RNA (dsRNA) that targets endogenous gene transcripts reduces transcript expression (41)(42)(43)(44)(45). Because of the importance of I. scapularis in a variety of infections, the effect of dsRNA treatment on TBFV replication and organ viability is important to evaluate. Although specific Ixodes spp. transcripts have been implicated in TBFV infection in tick cell cultures (46)(47)(48), ex vivo organ cultures can provide a means to more precisely probe tissue-specific vector interactions.
In this study, we dissected midgut, salivary glands, and synganglion from female I. scapularis ticks and confirmed that the organs were viable in ex vivo culture for over a week. Furthermore, we infected the organ cultures with LGTV and POWV, demonstrated virus dissemination, and identified production of infectious virus. We also transfected dsRNA targeting the LGTV 3= untranslated region (UTR) into organ cultures of TBFV-infected midgut and salivary gland tissues and observed a reduction in TBFV replication. Thus, we developed a tick organ culture model that can be applied as a functional genomics tool for studying TBFV biology and screening for vaccine or drug candidates. The organ culture model may also prove useful for high-containment studies where working with live ticks and select agents may be challenging.

RESULTS
Viability of dissected I. scapularis organs. The overarching goal of our study was to develop an organotypic culture model of isolated organs from the adult I. scapularis female tick to study the biology of TBFVs. The initial step was to determine the viability of these organs in culture, using a resazurin salt-based viability assay (48,49). Dissected midgut and salivary glands were viable for up to 240 h (10 days), whereas synganglion specimens were viable for up to 216 h (9 days) ( Fig. 1A to C). Thus, these three organs were metabolically active for a time sufficient to study ex vivo virus infection.
Infection of tick organ cultures with a GFP-expressing LGTV. The next step was to analyze whether the dissected tick organ cultures would support infection and spread of TBFV. We infected midgut, salivary gland, and synganglion cultures with LGTV expressing green fluorescent protein (LGTV GFP ) and observed the cultures by live fluorescence microscopy, using the GFP signal as an indicator of virus infection and gene expression. The growth of recombinant flaviviruses may differ from parental viruses (50,51); however, the potential difference in growth between LGTV GFP and its wild-type LGTV TP21 strain in these organs has not been determined.
GFP expression was observed in midgut ( Fig. 2; see also Movie S1 and S3 in the supplemental material), salivary gland (Fig. 3), and synganglion ( Fig. 4) cultures at selected time points postinfection. Nonspecific autofluorescence was observed in both LGTV GFP -infected and mock-infected organs and was especially high in the synganglion, but background fluorescence was distinguishable from GFP expression from LGTV GFP at different time points in different organs. LGTV GFP replication was first noted in the midgut tissue (MGT) at 72 hours postinfection (hpi) ( Fig. 2; Movie S1) and in salivary gland tissue at 108 hpi (Fig. 3), but not until 180 hpi in synganglion tissue (Fig. 4) in one imaging experiment. In addition, midgut cultures exhibited peristaltic contractions (Movie S1, S2, S3 and S4), a finding typical for this specific organ type, and these were consistently seen at multiple time points. Evidence of virus spread, manifested by spreading GFP expression, was observed from earlier to later time points (Fig. 2, 3, and 4; Movie S1 and S3).
In the midgut, LGTV GFP replication, observed via GFP expression, was seen diffusely throughout the diverticula of the midgut ( Fig. 2; Movie S1 and S3). This suggested that the epithelial cells (consisting of reserve, secretory, and digestive cells) (52,53) were likely targeted by the virus, although the specific infected cell type was not identified. GFP expression in the salivary gland appeared as puncta, primarily along lobular ducts and within acini (Fig. 3); however, we were unable to identify the specific type of acinus (54)(55)(56) in our experiment. GFP expression in synganglion localized in puncta in residual MGT as well as in the retrocerebral area (RCA) of the synganglion (Fig. 4).
Detection of LGTV and POWV protein synthesis in infected organ cultures. The experiments with LGTV GFP indicated that the virus infected the midgut, salivary glands, and synganglion. Spread of virus infection was noted in the midgut and salivary glands. In order to confirm these results with an independent method and to compare infection of LGTV with that by a pathogenic TBFV, we infected cultures with either LGTV or POWV and prepared samples for immunohistochemical staining every 48 h between 48 hpi and 192 hpi.
The infected organ sections were stained either with an anti-LGTV envelope (E) antibody or with a mouse polyclonal antiserum that recognizes multiple POWV proteins. Staining specific for LGTV or POWV was observed in infected organs ( Fig. 5 and  6; Fig. S1 and S2), but not in mock-infected organs (Fig. S3). In addition, spread of LGTV and POWV proteins was observed in infected midgut and salivary gland cells from earlier to later time points, but spread of TBFV proteins within synganglion (other than at the time of first observation of proteins) was not clearly shown ( Fig. 5 and 6). LGTV E and POWV protein synthesis in the midgut were initially observed at 48 hpi (Fig. 5) and 96 hpi (Fig. 6), respectively. Epithelial cells of the midgut diverticulum were infected, as indicated by the presence of viral proteins (Fig. S1), but again, specific infected cell types were not identified. LGTV E and POWV protein synthesis in the salivary gland were also initially observed at 48 hpi (Fig. 5) and 96 hpi (Fig. 6), respectively. Primary areas of viral protein synthesis in salivary glands were localized to cells lining the lobular ducts and granular acinus types II and/or III (54)(55)(56). LGTV E and LGTV expressed GFP, shown in green within the organ. A mock-infected organ was used for comparison, and small amounts of autofluorescence were observed. Serial imaging of the same LGTV GFPinfected and mock-infected midgut organ was performed. The asterisk at 72 hpi denotes the first time point where GFP expression within the LGTV GFP -infected midgut was distinguishable from autofluorescence in the mock-infected midgut.
Virus Infection of Black-Legged Tick Organ Cultures ® POWV protein synthesis in the synganglion was initially observed at 144 hpi (Fig. 5) and 96 hpi (Fig. 6), respectively. LGTV E and POWV protein synthesis in the synganglion localized both in the MGT and in the periganglionic sinus/sheath (PGS/H). POWV protein synthesis was also observed in the RCA (Fig. 6) of synganglion.
Infectious LGTV and POWV replication in organ cultures. In the previous sections, we utilized two different methods to demonstrate TBFV infection and gene LGTV expressed GFP, shown in green within the organ. Mock-infected organs were used for comparison, and small amounts of autofluorescence were observed within the lobular duct (ld; denoted by magenta arrow). Serial imaging of the same LGTV GFP -infected and mock-infected midgut organ was completed. White arrows denote GFP puncta within LGTV GFP -infected salivary glands. A great amount of GFP puncta were observed at 180 hpi (white arrows were not used). The asterisk at 108 hpi denotes the first time point where GFP expression within LGTV GFP -infected salivary glands was distinguishable from autofluorescence in mock-infected salivary glands.
Grabowski et al. ® expression within dissected tick organ cultures. To determine if infectious virus was being produced, we next employed an immunofocus assay (57)(58)(59) to evaluate if virus was generated in these cultures.
Organ cultures were infected with 5 ϫ 10 5 focus-forming units (FFU) of either LGTV or POWV, a sufficient concentration to establish TBFV infection in Ixodes spp. ticks (30,60,61). Infectious LGTV and POWV were produced from all three infected tick organ cultures ( Fig. 7 and 8). All infected organ types produced greater infectious virus replication than the initial inoculum control (with no organs) at multiple time points LGTV expressing GFP is denoted by green within the organ. A mock-infected organ was used for comparison, and high autofluorescence was observed. Serial imaging of the same LGTV GFP -infected and mock-infected midgut organ was performed. The white box denotes GFP puncta within LGTV GFPinfected synganglion in midgut tissue and in the retrocerebral area. The asterisk at 180 hpi denotes the time point where puncta of GFP expression within LGTV GFP -infected synganglion was distinguishable from autofluorescence in mock-infected synganglion.
Virus Infection of Black-Legged Tick Organ Cultures ® postinfection. At these collection time points, peak infectious LGTV and POWV production levels in the midgut cultures were observed at 108 hpi ( Fig. 7A and 8A). Peak infectious LGTV and POWV levels in the salivary gland cultures were observed at 180 hpi ( Fig. 7B and 8B). Similarly, peak infectious LGTV replication in the synganglion cultures was observed at 180 hpi ( Fig. 7C), but infectious POWV replication did not peak during these time points (Fig. 8C). Taken together, these results provide evidence that short-term cultures of individual tick organs are sufficient for TBFV replication. In addition, greater levels of infectious LGTV replication, in comparison with infectious POWV replication, were observed from infected midguts at 36 to 180 hpi, from salivary glands at 108 to 180 hpi, and from synganglia at 180 hpi.
Reduction of infectious TBFV replication by dsRNA transfection targeting the LGTV 3=-UTR. In the previous sections, we demonstrated that dissected organs from female I. scapularis were viable in culture and that the cultures supported replication of the low-neurovirulence LGTV and the pathogenic POWV, thus indicating that the organ cultures were suitable for the detailed study of TBFV biology in a highly controlled arthropod system. The next aim in our long-term project was to examine the role of specific tick proteins in the various steps of the TBFV life cycle by targeting specific gene transcripts using dsRNA-mediated RNAi. The effect of transfecting dsRNA targeting transcripts of tick genes in organs as a way to disrupt TBFV replication has not been ). md, midgut diverticulum; ld, lobular duct; a.c., acinus; pg/h, periganglionic sinus/sheath; n.p., neuropile; rca, retrocerebral area; mgt, midgut tissue. LGTV envelope protein within tissue is denoted by the prominent purple color. Mock-infected organs at 96 hpi were processed and stained in the same manner with LGTV antibody to identify any potential nonspecific staining background (see Fig. S3).
reported previously. As proof of principle, we targeted the 3= UTR of the viral RNA with dsRNA, reasoning that if viral transcript knockdown was successful, virus replication would be impaired. Therefore, we transfected LGTV-and POWV-infected midguts and salivary glands with dsRNA targeting the LGTV 3=-UTR (48) and measured infectious virus at 108 and 180 hpi, respectively. Levels of infectious LGTV and POWV replication in both the midgut ( Fig. 9A and B) and the salivary glands ( Fig. 9C and D) were suppressed at least 10-fold ( Fig. S4A to D). Importantly, LGTV 3= UTR dsRNA transfection did not have a detrimental effect on organ viability ( Fig. S5A and B). These results indicated that the tick organ cultures are a suitable system in which to study the effects of RNAi-mediated gene expression modification on TBFV replication.

DISCUSSION
The detailed study of virus biology in ticks has been hampered by a dearth of convenient systems. This is particularly true for work with highly pathogenic viruses, because arthropods infected with biosafety level 3 (BSL-3) or BSL-4 pathogens require laborious and cumbersome containment procedures during pathogenesis studies (62). Here, we have described methods that increase safety while investigating infection of arthropod tissues. Specifically, our study used dissected tick organs for ex vivo TBFV infection in order to (i) culture organs for periods of up to 9 to 10 days, (ii) assess the viability of individual organs while being cultured, (iii) dramatically reduce the handi- ). md, midgut diverticulum; ld, lobular duct; a.c., acinus; pg/h, periganglionic sinus/sheath; n.p., neuropile; rca, retrocerebral area; mgt, midgut tissue. POWV proteins within tissue are denoted by the prominent purple color. Mock-infected organs at 192 hpi were processed and stained in the same manner, using POWV antibody to identify any potential nonspecific staining background (see Fig. S3).
Virus Infection of Black-Legged Tick Organ Cultures ® caps of working with tick tissues infected with high-containment TBFV infection, and (iv) minimize the possibility of accidental sharps injuries during dissection of virusinfected ticks.
I. scapularis is a competent vector for both LGTV and POWV, and POWV specifically infects cells of midgut and salivary glands of the hard tick D. andersoni when infected in vivo (63). Based on these findings, we hypothesized that TBFVs would infect and replicate in ex vivo I. scapularis organs, even though the method deviated from the natural infection route in vivo. We found that both LGTV (Fig. 2, 3, 4, and 5; Fig. S1 and S2 and Movie S1 and S3) and POWV ( Fig. 6; Fig. S1 and S2) infected midgut, salivary glands, and synganglion. Furthermore, there was clear evidence of virus spread in midgut (Fig. 2, 5, and 6; Movie S1 and S3) and salivary glands (Fig. 3, 5, and 6), but the evidence was uncertain in synganglion ( Fig. 4, 5, and 6). Localization of both LGTV and POWV proteins in these organs was similar.
We used both LGTV GFP localization and TBFV immunohistochemical staining of salivary glands (Fig. 3, 5, and 6; Fig. S2) to show that TBFV replication occurred both in the lobular ducts and in granular acini (acinus types II and/or III). Future confirmation and identification of the acinus type infected in this organ culture system by using LGTV replication from infected organ cultures. Immunofocus assays were completed with supernatants collected from wells with LGTV-infected organs and control wells without organs (initial LGTV inoculum). Supernatants were collected from either one midgut (A), one salivary gland pair (B), or one synganglion (C) at each time point. Error bars represent standard errors of the means, and data are representative of two technical replicates for each sample for the immunofocus assays of two biological replicates. electron and fluorescence microscopy may provide insight into how the virus is released into the saliva. It is known that TBFV transmission can occur within the first few hours of tick attachment (29,64). This is likely due to the presence of virus in the salivary glands of unfed ticks and highlights the importance of tick salivary glands in POWV transmission. Furthermore, it is important to identify which type of acinus is susceptible to TBFV infection, since agranular (type I) and granular (type II and III) acini have distinct physiological functions (55,56). As a result, the virus may be using and/or altering the function of the acini, and the pattern of secretion may differ between infected and uninfected acini (37). Future identification of protein expression and/or the secretome of TBFV-infected acini would be highly beneficial to provide insight into salivary gland biology in the context of TBFV transmission.
The synganglion typically serves as an organ of interest for tick control measures due to its neurological and motor control relevance. We were uncertain whether infectious LGTV or POWV replication measured from synganglion supernatants was produced from infected actual synganglion cell types or from the small part of the MGT that was attached following synganglion dissection (Fig. 4, 5, and 6; Table S1). However, LGTV GFP was identified within the RCA of the synganglion (Fig. 4), and immunohisto-  (65). These findings suggest that certain viruses may have a tropism for the PGS/H in the tick synganglion. Possibly related is some evidence that TBFV infection in Ixodes spp. ticks is phenotypically associated with greater tick activity and aggression (35,36), but the mechanisms for this behavior and whether it is associated with TBFV infection of the synganglion are unknown.
All three dissected organ types produced infectious LGTV (Fig. 7A to C) and POWV ( Fig. 8A to C). Interestingly, all organ types produced at least 10-fold more infectious LGTV than POWV. These differences in infectious virus replication levels may reflect I. scapularis as a natural vector for POWV but not for LGTV. Also, POWV infections in this study were carried out with POWV lineage I, which may replicate differently than POWV lineage II (DTV) in I. scapularis organs. The latter virus has been isolated from naturally infected populations of I. scapularis (21,22). Detailed growth comparisons between both POWV lineages in tick organ cultures may provide further insight. These findings suggest that the ex vivo organ culture system may also prove useful in examining genes responsible for arthropod host range determinants for specific tick species.
The development of new recombinant, live vaccine candidates against TBFVs would be facilitated by viruses with restricted replication. Incorporation of microRNA sequences into these vaccine candidates (66,67) has provided a possible approach to increase the effectiveness of these live vaccines. Confirmation of restricted replication of these vaccine candidates in arthropod systems is also advisable. Thus, the ex vivo tick organ model offers a simplified screen for specific organs, such as the midgut and salivary glands. LGTV-infected salivary gland pair (C), and POWV-infected salivary gland pair (D) with 10 ng pGEM dsRNA and LGTV 3= UTR dsRNA. Transfections of midguts and salivary glands were completed for 108 and 180 h, respectively, before supernatants were collected for immunofocus assays, which were normalized to the pGEM dsRNA negative-control response. An unpaired t test was completed to compare FFU per milliliter of the LGTV 3= UTR dsRNA to that for the pGEM negative control. Error bars represent standard errors of the means, and data are representative of two technical replicates for each sample for the immunofocus assays of two biological replicates. **, P Յ 0.01; ***, P Յ 0.001; ****, P Յ 0.0001.
Transfection of the LGTV 3= UTR dsRNA in LGTV-and POWV-infected tick organs led to a decrease in not only infectious LGTV replication but also infectious POWV replication by at least 10-fold in both a midgut ( Fig. S4A and B) and a salivary gland pair ( Fig. S4C and D). This suggests that RNAi-mediated gene silencing is functional in the dissected, I. scapularis midgut and salivary glands, providing an RNAi-mediated antiviral response. Reduction of infectious POWV replication with LGTV 3= UTR dsRNA transfection suggests that the nucleotide conservation of the 3= UTR of both LGTV and POWV is high enough to stimulate a similar antiviral RNAi response corresponding to the POWV 3= UTR. Little is known about endogenous dsRNA soaking of dissected I. scapularis midgut and salivary glands (44), although microinjection of I. scapularis with endogenous dsRNA (44,(68)(69)(70)(71)(72)(73) provided successful transcript knockdown of endogenous gene expression. A proficient dsRNA-mediated transcript knockdown may also occur with I. scapularis organ culture; however, this is a subject that requires confirmation in future studies. Nonetheless, our study provides evidence that irrelevant pGEM and LGTV 3= UTR dsRNA may be used for suitable negative and positive controls for future endogenous dsRNA transfection studies using LGTV-and POWV-infected ex vivo tick organs.
The genome of I. scapularis has been annotated (74-77), and I. scapularis transcripts have been targeted in many functional RNAi-mediated silencing studies. I. scapularis nymph transcript expression changes in association with TBFV infection have also been observed (37). In addition, hundreds of I. scapularis tick cell proteins have been identified to have expression changes in association with TBFV infection (47,49). The identification of I. scapularis transcripts/proteins involved in TBFV infection needs to be explored in vitro and in vivo. Tick organ culture provides a convenient platform for these areas of investigation.
This study provides a foundation for using individual tick organs in cultures to study TBFV biology. However, variables exist that may affect organ viability and TBFV replication in organs, such as bloodfeeding and the virus concentration used for infection. Unfed female tick organs were utilized for this study (as in the study by Sunyakumthorn et al. [78]); thus, results may reflect a bias to a specific adult tick sex (or tick life stage) and physiological state. In vivo studies with TBEV in Ixodes ricinus ticks have shown higher levels of TBEV replication in blood-fed versus unfed ticks and, in particular, greater replication in salivary glands (60). We hypothesize that a similar trend may occur with TBFV replication in ex vivo organs from blood-fed versus unfed I. scapularis ticks; thus, use of organs from blood-fed ticks may be an option for future studies. In addition, different concentrations of the initial TBFV inoculum may also affect TBFV replication kinetics in these organs. Mitzel et al. showed that different concentrations for the LGTV inoculum affect the establishment of LGTV genome replication in I. scapularis larvae (30). Similarly, we suspect that different concentrations of the initial TBFV inoculum in female I. scapularis organ cultures could reveal varied levels of TBFV replication at different time points postinfection.
In summary, we have characterized the viability and extent of flavivirus infection of I. scapularis ex vivo organs, specifically showing a difference in replication between a low-neurovirulence flavivirus and a more pathogenic flavivirus. We highlighted cell types of specific tick organs that may be susceptible to flavivirus infection and utilized individual organs for dsRNA-mediated RNAi assays to show subsequent reductions in flavivirus replication. Overall, we have created a framework in which tick organ cultures can be utilized for further flavivirus pathogenesis studies using the tick system.

MATERIALS AND METHODS
I. scapularis tick organ dissection and preparation. Unfed, female I. scapularis ticks were obtained from a laboratory colony maintained at the Oklahoma State University Tick Rearing Facility. Ticks were maintained as described by Mitzel et al. (30). Ticks were surface sterilized (79) by soaking them in 3% H 2 O 2 for 15 s followed by soaking in 70% ethanol for 15 s. They were then wicked free of excess fluid on sterile Whatman paper and subsequently dissected in sterile phosphate-buffered saline (PBS) on a sterile glass microscope slide. The midgut was excised first and carefully removed from the cavity, followed by the salivary glands and synganglion. Tracheal tubes were removed carefully from the organs before experimental use. L15C-300 medium (80, 81) was used for organ culture and was not supplemented with antibiotics. Organ viability assay. To determine the viability of dissected individual organs, separate organs were transferred, one at a time using one tip of the forceps, into complete L15C-300 medium with alamarBlue (a resazurin salt-based cell viability reagent; AbD Serotec, Raleigh, NC) at a 1:10 dilution in wells of a 96-well plate (49). The other tip of the forceps was dipped into a separate control well with medium supplemented the same way and served as a no-organ control.

Virus Infection of Black-Legged Tick Organ Cultures
Organs and their control wells were incubated at 34°C without CO 2 for 12 h, and absorbance was measured at a wavelength of 570 nm with a Molecular Devices SpectraMax Plus 384 plate reader and SoftMax Pro v6.5 software. Two machine replicates were completed with one organ per well and a corresponding control well that served as a biological replicate for each organ type. Three separate experiments were performed for each organ type (n ϭ 3).
LGTV and POWV stocks were created using African Green monkey kidney cells (Vero; ATCC).
LGTV wild-type strain TP21, passage 5 stock was amplified in Vero cells (multiplicity of infection [MOI], 0.01) (82) and grown as described above, except with no FBS, to passage 6 in order to provide a working stock for experimental infections. POWV LB prototype strain (originally obtained from Robert Tesh, University of Texas Medical Branch) (83) stock was also amplified in Vero cells at the same MOI and grown in the same manner as LGTV to provide a working stock for experimental infections.
LGTV and POWV stock titers were determined in immunofocus assays as described by Offerdahl et al. and Mlera et al. (57)(58)(59). The same method was used to determine infectious LGTV and POWV titers in experimental samples, with two technical replicates per sample. In parallel, a mock stock was prepared from mock-infected Vero cells.
LGTV capable of enhanced GFP (eGFP) gene expression (LGTV GFP ) was generated using a strategy described previously (50). A PCR-amplified sequence of the eGFP gene was fused at the 3= end with the codon-optimized sequence of the E/NS1 stem-anchor region (nucleotide positions 2171 to 2488 of the LGTV genome) of the RepC58 plasmid (66). This sequence was inserted into the infectious clone of the TP21 strain of LGTV (described by Tsetsarkin et al. [66]) at nucleotide position 2489 of the LGTV genome. The plasmid was propagated in Escherichia coli (strain MC1061) and purified using the Endo Free Plasmid Maxi kit (Qiagen, Valencia, CA). The sequence of the plasmid carrying the LGTV GFP gene is available from the authors upon request.
LGTV GFP DNA plasmid transfection was carried out in Vero cells as described by Tsetsarkin et al. (66,84), except that the X-tremeGene siRNA transfection reagent (Roche, Basel, Switzerland) was used per the manufacturer's instructions. Transfected cells were incubated for 7 days before supernatant was collected for infectious LGTV GFP stock (LGTV GFP passage 1). The titer for LGTV GFP passage 1 stock was determined in immunofocus assays as described above with the LGTV TP21 strain stock. To create a mock stock for a control, supernatant was collected from transfected cells without the LGTV GFP DNA plasmid and prepared in the same manner.

Virus infection of organs for LGTV GFP expression, immunohistochemical staining, and infectious virus replication.
LGTV GFP infection of organs to image GFP expression was completed as follows. Organs were dissected as described above and placed into complete L15C-300 medium in an 8-well Labtek dish (Nunc, Rochester, NY). A total of 5 ϫ 10 5 FFU of LGTV GFP was placed into the infection wells, and mock stock was placed into the mock infection wells. One midgut, one salivary gland pair, and one synganglion were dissected and placed into separate LGTV GFP -infected or mock-infected wells. One hour of adsorption with rocking at 34°C with no CO 2 was completed. Adsorption medium was carefully removed from the wells without removing the organs. Fresh complete L15C-300 medium was placed into the infection wells and cultures were incubated at 34°C with no CO 2 . Serial imaging of LGTV GFP -infected and mock-infected organs at 3, 36, 72, 108, 144, and 180 hpi was performed as described below.
To infect cultures with LGTV and POWV for immunohistochemical staining, organs were dissected as described above and placed into complete L15C-300 medium in wells of a 96-well plate. A total of 5 ϫ 10 5 FFU of LGTV or POWV was placed into the infection wells, and mock stock was placed into mock infection wells. One hour of adsorption with rocking at 34°C with no CO 2 was completed. Infected organs were incubated at 34°C with no CO 2 . Organs were collected at 48, 96, 144, and 192 hpi, with processing and staining of organs completed as described below.
LGTV and POWV infection of organs to determine levels of infectious virus replication were completed as follows. Sections (3 by 3 mm in size) of sterile gelfoam adsorbable gelatin sponge (Pharmacia and Upjohn, Kalamazoo, MI) were cut and submerged into complete L15C-300 medium in wells of a 96-well plate. Using sterile forceps, dissected organs were placed onto the soaked gelfoam and placed back into the original well. A total of 5 ϫ 10 5 FFU of LGTV or POWV was placed into the well with the organ and into a well with no organ but just gelfoam (initial virus inoculum control; this control was used to separate the initial infectious virus inoculum from the infectious virus being produced from the organs). One hour of adsorption with rocking at 34°C with no CO 2 was completed. Adsorption medium was carefully removed without removing organs, and gelfoam with organs was carefully rinsed three times with sterile 1ϫ PBS. After rinsing with PBS, complete L15C-300 medium was placed into each well, followed by incubation at 34°C with no CO 2 . Wells with organs and control wells with no organs were processed in the same manner. Supernatants were collected from wells with organs and wells with no organs at 3, 36, 72, 108, 144, and 180 hpi, and virus titers were determined in immunofocus assays as described above, with two separate experiments for each organ type (n ϭ 2).
Analysis of LGTV GFP replication.
LGTV GFP -infected and mock-infected organs contained in the 8-well Labtek dish were viewed on a Zeiss model AxioVert.A1 microscope, and serial images and videos of the LGTV GFP -infected and mock-infected organs were captured using a Zeiss model AxioCam 503 mono camera. Fluorescence excitation was determined using the 89 North model PhotoFluor L.M. 75 images. Images and videos were processed with Zeiss ZEN 2 (Blue edition) software using autoexposure settings. Green fluorescent images and black and white videos were obtained with 10ϫ magnification, with green and white denoting GFP expression in images and videos, respectively. Mock-infected organs were completed in parallel with LGTV GFP -infected organs in order to identify potential autofluorescence that occurred with fluorescent imaging of tick organs (40,63).
Paraffin processing and sectioning of dissected organs. Dissected tick organs consisting of midgut, salivary glands, and the synganglion that were infected with LGTV or POWV (as described above) for 48, 96, 144, and 192 hpi were fixed in 10% neutral buffered formalin (Cancer Diagnostics, Durham, NC) for a minimum of 24 h and then transferred to 70% ethanol. In parallel, mock-infected organs for 96 and 192 hpi were treated and collected in the same manner to provide negative controls for LGTV and POWV protein synthesis, respectively. Individual organs (midgut, salivary gland pair, or synganglion) were rinsed in PBS and covered in a thin layer of Histogel matrix (Thermo Scientific, Kalamazoo, MI). The Histogel matrix-encased organs were allowed to solidify at 4°C, placed in tissue cassettes, and processed using a VIP-6 Tissue Tek tissue system (Sakura Finetek, USA). Tick organs were embedded in Ultraffin paraffin polymer (Cancer Diagnostics, Durham, NC) and serially sectioned at 5 m, and selected air-dried slides were stained with hematoxylin and eosin.
Immunohistochemical staining of viral proteins and imaging. Anti-POWV and anti-LGTV virus immunoreactivities were detected using POWV ascites fluid (which detects multiple viral proteins; ATCC) (59) at a 1:2,000 dilution and anti-LGTV-E mouse monoclonal antibody (detects viral E protein; 11H12) at a 1:50 dilution, respectively. The secondary antibody was a ready-to-use Discovery OmniMap anti-mouse horseradish peroxidase conjugate (prediluted; Roche). The tissues were then processed for immunohistochemical staining using the Discovery Ultra automated processor (Ventana Medical Systems, Tucson, AZ) with a Discovery Purple kit (Ventana Medical Systems). Stained cross-sections of infected and mock-infected organs were viewed on an Olympus model BX51 microscrope, and images were collected using an Olympus model DP80 camera. Images were processed with Olympus cellSens Dimension v1.13 software. Images were captured at 20ϫ, 40ϫ, and 100ϫmagnification (indicated in the figures) with LGTV or POWV protein staining denoted by a prominent purple color. Mock-infected organs were processed in parallel with infected organs at 96 hpi with antibody for LGTV and at 192 hpi with antibody for POWV in order to identify any nonspecific staining.
Extraction of RNA from LGTV-infected ISE6 cells and cDNA synthesis. ISE6 cells grown to ϳ2 ϫ 10 6 cells in a T25 flask were infected with LGTV at an MOI of 10. After 36 hpi, medium was removed and RLT lysis buffer (from the Qiagen RNeasy minikit) was directly applied to lyse the attached cells. Further processing of LGTV-infected ISE6 RNA was carried out according to the manufacturer's instructions for the RNeasy minikit (Qiagen).
An iScript cDNA synthesis kit (Bio-Rad, Hercules, CA) and corresponding instructions were used to synthesize cDNA from RNA samples. The following thermocycler conditions were used for cDNA synthesis: 25°C for 5 min, 42°C for 50 min, and 85°C for 5 min (48).
Transfection of infected tick organs with dsRNA.
LGTV and POWV infection and incubation of midgut and salivary glands for posttreatment with pGEM or LGTV 3= UTR dsRNA were completed as described above for infecting organs to determine infectious virus replication, except 100 l complete L15C-300 medium was placed into each well followed by 50 l dsRNA transfection mix prepared according to methods described in references 48 and 85 with 10 ng dsRNA (150-l total volume in the well). With the used initial virus inoculum concentration (5ϫ 10 5 FFU), midgut and salivary glands were transfected for 108 and 180 h, respectively, since peak infectious LGTV and POWV replication levels were observed for the midgut and salivary glands at these time points. Supernatants were collected from these infected, dsRNA-transfected organs and subjected to immunofocus assays (as described above) to determine infectious virus replication levels.
The effects of LGTV 3= UTR dsRNA transfection on midgut and salivary gland viability were assessed as described above for transfecting organs, except organs were not infected and were placed directly into 96 wells with the transfection-medium mix with no gelfoam. After the specified transfection time, midgut and salivary glands were removed and placed into alamarBlue-infused medium. Organs were incubated for 12 h, and viability readings were collected as described above.