All experiments involving mice were performed in accordance with the animal use protocol approved by the State Veterinary and Food Administration of the Slovak Republic (permit number 1335/12-221).

Five-week-old female BALB/c mice were purchased from Dobra Voda Breeding Station, Institute of Experimental Pharmacology and Toxicology, Slovak Academy of Sciences (SAS). The animals were housed at the Institute of Virology, Biomedical Research Center, SAS (Bratislava, Slovakia) under standard conditions. Food and water were provided ad libitum. Upon arrival, mice were allowed to adapt to the local environment before being incorporated into the study. Mice were 6 weeks old at the start of the study. At the end of the experiment animals were euthanized by cervical dislocation under anesthesia induced by carbon dioxide.

I. ricinus ticks were obtained from a laboratory colony maintained at the Institute of Zoology SAS (Bratislava, Slovakia). The F1 generation of laboratory-bred I. ricinus females was used for virus inoculation. TBEV (Hypr strain prepared as a 10% mouse brain suspension of 1.1 × 10⁹ PFU/ml in Leibovitz's L-15 medium) was provided by the Institute of Virology, Biomedical Research Center, SAS. Fasting I. ricinus females were inoculated into the haemocoel with TBEV (5.5 x 10⁴ PFU per tick) through the coxal plate of the second pair of legs using a digital microinjector TM system (MINJ-D-CE; Tritech Research, Inc., USA) (Kazimírová et al., 2012). By this procedure, ~100% of the ticks were found to acquire TBEV infection (Slovák et al., 2014). Inoculated ticks were incubated at room temperature and 85% relative humidity in a desiccator for 21 days prior to the infestation experiments.

Two groups of mice (n = 6 per group) were infested with either TBEV-infected or uninfected (control) I. ricinus females. Ticks were placed in small neoprene capsules glued on the shaved backs of the mice (two capsules per mouse, four tick females per capsule) (Kazimírová et al., 2012; Hermance and Thangamani, 2014). Ticks in each capsule were allowed to feed for either 1 or 3 h. After the allotted feeding time, skin biopsies were taken from euthanized mice, at each tick feeding site using a Premier Uni-Punch (Premier Products Co., Plymouth Meeting, PA). For immunohistochemical analysis, skin biopsies (n = 3) were harvested with attached ticks and placed in 4% formaldehyde. For RNA extraction, the attached ticks were removed from the skin biopsies and the biopsies (n = 3) were placed in RNALater (Ambion, Life Technologies, Carlsbad, CA). Control biopsies were taken from the shaved skin of naïve tick-free mice and stored in either 4% formaldehyde or RNALater.

Total RNA was extracted as previously described (Heinze et al., 2012; Hermance and Thangamani, 2014) and RNAseq analysis (Illumina deep sequencing / NGS) was performed on these samples. Briefly, 1 μg of total RNA from mouse skin biopsies (n = 3) was poly A+ selected and fragmented using divalent cations and heat (94°C, 8 min). Illumina TruSeq v2 sample preparation kits (Illumina Inc., San Diego, CA) were used for the RNA-Seq library construction. Each sample library was uniquely indexed to allow combining libraries during sequencing and subsequent separation post-sequencing. NGS was performed at the NGS core facility, Sealy Center for Molecular Medicine, The University of Texas Medical Branch (UTMB). Sample libraries were analyzed by the Illumina HiSeq 1500 using a 2 × 50 base paired end run protocol, with TruSeq v3 sequencing-by-synthesis chemistry. Reads were aligned to the mouse mm10 reference genome using TopHat version v2.0.4. Cuffdiff version 2.0.2 was used to estimate differential gene expression between TBEV-infected and uninfected feeding sites after 1 or 3 h of tick feeding. The total dataset of 23,000 genes was filtered for p-values ≤ 0.05 and a fold change ≤ −1.5 or ≥ +1.5. The Log₂(fold change) and p-value data for each gene expression comparison (Supplement Table 1) were then uploaded to Ingenuity Pathway Analysis (IPA) software (Ingenuity Systems, Redwood City, CA) for further transcriptional analysis of the early cutaneous immune response, as previously described (Heinze et al., 2012; Hermance and Thangamani, 2014).

Real-time PCR was used to validate the NGS data (Supplement Table 1). Skin biopsies were harvested from the feeding sites of TBEV-infected and uninfected ticks as described above. At each time point, biopsies from three mice were used for the real-time PCR validation. Fifteen gene targets were selected for real-time PCR analysis. We selected these genes based our previous studies with POWV where these genes were shown to be modulated genes of interest during the host anti-tick response to POWV infection (Hermance and Thangamani, 2014, 2015). This list also includes genes that were not differentially modulated as per RNASeq analysis. Primers were purchased from Integrated DNA Technologies and the primer sequences are provided in Supplement Table 2. Primers were mixed with IQ SYBR green supermix (Bio-Rad) and loaded into iCycler IQ PCR 96-well plates (Bio-Rad) to create customized PCR arrays where each gene was measured in triplicate. For each PCR plate, 1 μg of total RNA extracted from skin biopsies was converted into cDNA using the RT² First Strand kit (Qiagen). cDNA was loaded onto the 96-well PCR plates, which were run on an iCycler iQ5 real-time PCR instrument (Bio-Rad) with the following cycling protocol: 10 min at 95°C; 15 s at 95°C, 1 min 60°C for 40 cycles, and an 80-cycle (+0.5°C /cycle) 55–95°C melt curve. Every array included GAPDH as an endogenous control gene and a no-template control. The iCycler's software was used to calculate the threshold cycle (C_T) values for all analyzed genes. The delta-delta C_T method was used to calculate fold-changes in gene expression between TBEV-infected and uninfected tick feeding sites. Data normalization was achieved by correcting all C_T values to the average C_T values of the GAPDH housekeeping gene. Statistically significant differences in gene expression between the 1 vs. 3 h tick feeding time points were determined by the Student's t-test. P-values less than 0.05 were considered significant. SPSS statistical software was used.

The skin biopsies with attached ticks were formalin-fixed for a minimum of 48 h in 4% formaldehyde. These biopsy samples were treated with Decal (Decal Chemical Corp, Tallman, NY) for 2 h, and then paraffin-embedded (Heinze et al., 2012). Five micron paraffin sections were taken from each sample and adhered to glass slides. The slides were deparaffinized in xylene and then rehydrated in decreasing concentrations of ethanol (Hermance et al., 2016). For antigen retrieval, the slides were treated with 10 mM Tris Base + 0.05% Tween 20 (pH 10) for 20 min with microwave heating. Upon returning to room temperature, endogenous peroxidase quenching was performed by incubating the slides in 4% H₂O₂ for 30 min. The primary antibody for TBEV detection used in this study was a Hyper Mouse Immune Ascitic Fluid (HMIAF) antibody against TBEV; therefore, the Mouse-On-Mouse kit (MOM, Vector labs, Burlingame, CA) was used in order to reduce the endogenous mouse Ig background staining generated when mouse primary antibodies are used on mouse tissue (Santos et al., 2016). Slides were incubated for 1 h at room temperature with the MOM mouse Ig blocking reagent. The HMIAF primary antibody against TBEV was diluted 1:250 in the MOM diluent and incubated for 30 min at room temperature. Secondary antibody consisted of MOM biotinylated horse anti-mouse IgG reagent, which was incubated for 30 min at room temperature. The biotinylated secondary antibody was detected with a streptavidin-peroxidase ultrasensitive polymer (Sigma-Aldrich, St. Louis, MO) followed by staining with the NovaRED HRP substrate kit (Vector Laboratories Inc., Burlingame, CA). Slides were counterstained with Harris hematoxylin and coverslips were mounted using Permount (ThermoFisher Scientific, Waltham, MA). Uninfected skin and tick biopsy sections generated from uninfected tick feeding sites were used as negative controls to verify the specificity of the MIAF anti-TBEV primary antibody. Secondary antibody only (no primary antibody) controls were used to confirm that the MOM biotinylated anti-mouse IgG reagent did not bind non-specifically to cellular components.

In total, 1,548 genes were analyzed by Ingenuity Pathway Analysis (IPA) software (Supplement Table 1). The focus of the present study is on the cutaneous immune response observed at the TBEV-infected vs. the uninfected tick feeding sites after 1 and 3 h of tick feeding. The networks generated from IPA analysis illustrate the interrelationships between genes and the temporal changes in gene modulation. The top IPA-generated networks are shown for the TBEV-infected vs. uninfected tick feeding sites after 1 h of tick feeding (Figure 2A), the TBEV-infected vs. uninfected tick feeding sites after 3 h of tick feeding (Figure 2B), and the TBEV-infected tick feeding site after 3 h of tick feeding vs. the TBEV-infected tick feeding site after 1 h of feeding (Figure 2C). The three bio-functions most associated with the networks representing the TBEV-infected vs. uninfected comparisons (Figures 2A,B) are cellular movement, organismal injury and abnormalities, and hematological system development and function. After 1 h of TBEV-infected tick feeding, 19 host genes shown in the Figure 2A network were downregulated and 16 host genes were upregulated. After 3 h of TBEV-infected tick feeding, 22 host genes in the Figure 2B network were downregulated, 12 were upregulated, and one gene (Ngp) had a fold change of zero. Thus, as 1 h progressed to 3 h of TBEV-infected tick feeding, the network displayed a slight shift toward cutaneous gene downregulation (Figure 2). The IPA-generated network for the TBEV-infected 3 h vs. the TBEV-infected 1 h tick feeding site (Figure 2C) is unique from the Figures 2A,B networks, as it lacks Ccl2, KLF2, and ZBTB16, but instead includes 20 unique genes which contribute to this network's association with bio-functions such as molecular transport, organismal injury and abnormalities, and cellular movement.

In both the 1 and 3 h comparisons between TBEV-infected vs. uninfected tick feeding sites, inflammatory response was the primary predicted host response. Based on the IPA transcriptional immunoprofiling, the inflammatory response was projected to be activated in the 1 h comparison of the TBEV-infected vs. uninfected tick feeding site (activation z-score = 1.549), and inhibited in the 3 h comparison (activation z-score = −1.26). Temporal changes in gene expression for all genes predicted to have direct correlative relationships with the inflammatory response were plotted after 1 or 3 h of TBEV-infected vs. uninfected tick feeding (Figure 3). Transcriptional levels of cytokines Ccl2, Ccl12, Cxcl1, Cxcl2, Cxcl5, IL6, and IL10 were all upregulated after 1 h of TBEV-infected tick feeding, thus contributing to the overall activation of the inflammatory response (Figure 3). At the 3 h time point, the majority of these cytokine transcriptional level were still upregulated, with the exception of Cxcl5 and IL10, which were both slightly downregulated. Real-time PCR validation of a few selected immune genes followed the general pattern observed in the Illumina NGS analysis (Figure 1E). All of the enzymes projected to have a direct, correlative relationship with the inflammatory response were downregulated at 3 h, further contributing to the predicted inhibition of the inflammatory response at the feeding site of a TBEV-infected tick (Figure 3). Additionally, transcript expression of several receptors (CCR1, CCR5, and Sell) were all significantly upregulated after 1 h of TBEV-infected tick feeding (Figure 3), likely contributing to the overall accumulation of immune cells to the tick attachment site. In summary, the cutaneous gene expression analysis indicates that the inflammatory response is activated after only 1 h of TBEV-infected tick feeding, and that increased recruitment and accumulation of immune cells is expected at the 1 h time point. However, after 3 h of TBEV-infected tick feeding, the cutaneous inflammatory response is predicted to undergo slight inhibition, as demonstrated by the downregulated expression of over half of the genes shown to have a correlative relationship with the inflammatory response (Figure 3).

An IPA core comparison analysis was used to analyze changes in host cutaneous gene expression in response to TBEV-infected tick feeding vs. uninfected tick feeding observed across the two experimental time points (1 and 3 h). Figure 4 shows activation status prediction networks for the top five immune responses or bio-functions generated from the IPA core comparison analysis. “Maintenance of leukocytes” was the top predicted bio-function in the core comparison analysis (Figure 4). This IPA category moniker refers to bio-functions associated with the normal cellular activities that maintain cellular homeostasis, including engulfment, phagocytosis, regulation, and stasis of cells. The maintenance of leukocytes was predicted to be inhibited at both the 1 and 3 h comparisons of TBEV-infected vs. uninfected tick feeding (activation z-score = −2 for both time points). At both the 1 and 3 h time points, GATA3 transcription was downregulated (Figure 4). GATA3 is a transcription factor required for both the maintenance and development of Th2 cells (Pai et al., 2004). Furthermore, IL34, which is highly expressed by keratinocytes in the epidermis and plays an important role in the maintenance and development of Langerhans cells (Greter et al., 2012), was also downregulated after 1 and 3 h of TBEV-infected tick feeding (Figure 4). CCR4 knockout (CCR4^−/−) bone marrow-derived dendritic cells are less efficient in the maintenance of Th17 responses compared to wild type dendritic cells (Poppensieker et al., 2012); therefore, the downregulation of CCR4 transcription in the present study contributes to the inhibition of leukocyte maintenance at the 1 and 3 h time points. Mouse integrin subunit beta 7 (ITGB7) is necessary for the maintenance of memory T lymphocytes expressing the CD4 protein (Yang et al., 2011). Since ITGB7 is downregulated in the present transcriptional immunoprofiling study, it likely contributes to the inhibited maintenance of memory CD4+ T cells after 1 and 3 h of TBEV-infected tick feeding.

“Mobilization of phagocytes” was the second predicted bio-function in the core comparison analysis. After 1 and 3 h of TBEV-infected vs. uninfected tick feeding, the mobilization of phagocytes was predicted to be activated (activation z-score = 1.982 for both time points) (Figure 4). Monocytes, macrophages, neutrophils, dendritic cells, and mast cells are several of the professional phagocytic cells. In mice, there are two Ccl2 homologs, Ccl2 and Ccl12, both of which are potent phagocyte chemoattractants. The transcript levels of these two chemokines are upregulated at both the 1 and 3 h time points, thereby increasing the mobilization and chemotaxis of phagocytes at the TBEV-infected tick feeding site (Figures 3, 4). CCR5 is known to regulate accumulation of activated macrophages in the West Nile virus-infected mouse brain (Glass et al., 2005), and upregulated transcription of this chemokine receptor in the present study likely contributes to the activated mobilization of phagocytes. Mast cells and tissue resident macrophages release the potent neutrophil chemoattractants, Cxcl1 and Cxcl2 (De Filippo et al., 2008, 2013). In response to tissue inflammation, mast cells positioned near the vasculature initiate neutrophil infiltration from the circulation by releasing Cxcl1 and Cxcl2, while tissue resident macrophages also release Cxcl1 and Cxcl2, recruiting neutrophils deeper into the inflamed tissue (De Filippo et al., 2013). In the present study transcript levels of both these chemokines are both upregulated during TBEV-infected tick feeding, thus contributing to the activated mobilization of neutrophils that infiltrate the tick feeding site very early (Figures 3, 4).

The third predicted bio-function generated in the IPA core comparison analysis was “accumulation of helper T lymphocytes.” The activation z-score for this bio-function decreased from 2.19 at 1 h post-TBEV-infected tick feeding to 1.452 at the 3 h time point. Transcript levels of CCR1 and Cxcl2 are upregulated at both time points in the present study and these genes are associated with the recruitment and accumulation of helper T lymphocytes at the cutaneous site of TBEV-infected tick feeding (Figure 4). Selectin L (SELL) is a cellular adhesion molecule found on the surface of most lymphocytes that mediates lymphocyte capture and rolling at sites of inflammation (Tedder et al., 1995). In the present study, SELL transcription was upregulated at the 1 h time point, contributing to the IPA prediction that helper T lymphocytes are accumulating, but the downregulated transcription of SELL at 3 h is inconsistent with the predicted activation pattern, as depicted by the yellow arrow in Figure 4. TLR9, a receptor that preferentially binds unmethylated CpG sequences in bacterial and viral DNA, was downregulated in the present study (Figure 4). Here, the downregulation of TLR9 transcription is predicted to activate the accumulation of T lymphocytes at the skin feeding site of a TBEV-infected tick. This prediction is supported by a recently developed mouse model of cutaneous lupus that is dependent on the expression of TLR7 and the loss of TLR9, which ultimately results in extensive accumulation of IFNγ-producing T lymphocytes (Mande et al., 2017). Distal-less 3 (DLX3) is a transcription factor involved in the terminal differentiation of keratinocytes. In a mouse model with epidermal ablation of DLX3, there was an increase of IL17-producing CD4⁺ T cells, CD8⁺ T cells, and γδ T cells in the skin and draining lymph nodes (Hwang et al., 2011); thus, the downregulated transcription of DLX3 in the skin biopsies from the present study likely contributes to the accumulation of helper T lymphocytes at the cutaneous feeding site of infected ticks (Figure 4).

“Maintenance of cells” was the fourth predicted bio-function in the IPA core comparison analysis. This bio-function was predicted to be inhibited at both experimental time points (1 h activation z-score = −1.342, 3 h activation z-score = −2.236). The “maintenance of cells” activation network is very similar to the top predicted network, “maintenance of leukocytes”; however, the only difference is that Sonic hedgehog (SHH) is present in the “maintenance of cells” network (Figure 4). The touch dome, a highly specialized mechanosensory epidermal structure composed of distinct keratinocytes in close association with innervated Merkel cells, is maintained by signaling with SHH to tissue-specific stem cells (Xiao et al., 2015). In the present study SHH transcription is upregulated after 1 h of TBEV-infected tick feeding, a finding that is inconsistent with the predicted pattern of inhibited cell maintenance (depicted by the yellow arrow in Figure 4); however, at the 3 h time point, SHH transcription was downregulated, contributing to the predicted inhibition of touch dome cell maintenance (Figure 4). Some damage to the perineural skin microenvironment is likely to occur because of the mechanical injury induced by tick feeding, and this could ultimately impact SHH signaling and the touch dome homeostasis.

The “activation of neutrophils” was the fifth predicted bio-function in the core comparison analysis. After 1 h of TBEV-infected vs. uninfected tick feeding, neutrophil activation was predicted to be activated (activation z-score = 1.342); however, after 3 h, neutrophils were predicted to be inhibited (activation z-score = −2.236) (Figure 4). The upregulated transcription of Cxcl5 and TREML2 at the 1 h time point, followed by the downregulation of these genes at the 3 h time point contribute to this phenomenon because Cxcl5 is a chemokine involved in the activation of neutrophils (Proost et al., 1993) and TREML2 potentiates neutrophil activation in response to G protein-coupled receptor signaling (Halpert et al., 2011). Likewise, Lipocalin-2 (LCN2) transcription was upregulated at 1 h post-TBEV-infected tick feeding and downregulated at 3 h. LCN2 is expressed in neutrophils and is a potent inducer of their chemotaxis and migration to sites of inflammation (Schroll et al., 2012); thus, its upregulation at the 1 h time point further contributes to the activation of neutrophils at the feeding site of TBEV-infected ticks. VTCN1 belongs to the B7 family of costimulatory proteins. VTCN1 inhibits the expansion of neutrophils from their progenitors, as demonstrated by VTCN1-deficient mice that display enhanced neutrophil-mediated innate immunity (Zhu et al., 2009); therefore, the downregulation of VTCN1 transcription after 1 h, and its upregulation after 3 h, contributes to the activation of neutrophils at the cutaneous feeding site of a TBEV-infected tick (Figure 4).

Immunohistochemical analysis was performed at the feeding sites of TBEV-infected and uninfected I. ricinus to determine if the gene expression data could be correlated to the tissue morphology and inflammation. The tick feeding site is characterized by extravasated erythrocytes and leucocytes, and with a steady influx of inflammatory cells. As early as 1 h post-TBEV-infected tick feeding, neutrophils were observed at both infected and uninfected tick feeding sites, with a marked increase at the infected tick feeding site. After 3 h of TBEV-infected tick feeding, increased recruitment of inflammatory cells was observed at the infected tick feeding site compared to the uninfected tick feeding site. Among these, TBEV antigens were localized in fibroblasts and mononuclear cells, but not in neutrophils (Figure 5). These immunohistochemistry observations support the cutaneous immune gene expression data, showing an inflammatory micro-environment at the feeding site of TBEV-infected ticks. The recruitment of inflammatory cells appears to be much more pronounced in the TBEV-infected tick feeding site than the uninfected tick feeding site.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.