Niakha virus: A novel member of the family Rhabdoviridae isolated from phlebotomine sandflies in Senegal

Growth in cell culture

NIAV produced marked cytopathic effect (CPE) in BHK (baby hamster kidney), Vero E6 (African green monkey kidney) and C6/36 (Aedes albopictus) cells, 4, 8 and 7 days after inoculation, respectively, but it did not produce CPE in PP-9 (Phlebotomus papatasi) cells. Ultrathin-section electron micrographs of the virus in infected BHK and PP-9 cells are shown in Figures 1 A-D and E-F, respectively.

Virus morphology

Virus particles were regularly found packed in stacks in expansions of perinuclear space and in cisterns of granular endoplasmic reticulum of BHK cells (Figure 1, A-C). They were ~45 nm in diameter and 180-200 nm in length. The accumulation of NIAV particles in intracellular vesicles is not unusual for rhabdoviruses. In fact, particles stacked in intracellular vesicles were reported in the first description of VSV morphogenesis in 1968 (Hackett et al., 1968). While mature virions of animal rhabdoviruses exit the cell by a budding process through the cell membrane, it is possible that the observed stacks of NIAV will be transported through these membrane bound vesicles (e.g endoplasmic reticulum) to the cell surface for budding. Virions observed budding from the BHK cell surfaces (Figure 1, D), exhibited a two-rounded end bullet-like particle morphology of ~40 nm in diameter, rather than the classic one-rounded end bullet-like rhabdovirus morphology. A similar particle morphology was observed with Oak Vale virus (OVRV)(Gubala, 2009), and Tupaia virus (Kurz et al., 1986). In ultrathin sections of PP-9 cells, virus particles ~50 nm in diameter and up to 350 nm long were observed mostly at the cell surface, where they could be seen budding (Figure 1, E). Aggregates of tightly packed virions were also observed in the extracellular space of the sandfly cells (Figure 1, F).

Antigenic relationships

Complement fixation (CF) tests were conducted using NIAV mouse brain antigen and hyperimmune ascitic fluid (MIAF) and antisera against 23 other rhabdoviruses, including 19 viruses isolated previously in Africa (Table 1). NIAV MIAF reacted only with the homologous (NIAV) antigen at a dilution of 1:512. No CF cross-reactions were detected with any of the other 23 rhabdoviruses.

Phylogenetic analysis

The ML tree of 45 rhabdovirus L protein sequences is shown in Figure 2. Members of the Cytorhabdovirus, Novirhabdovirus and Nucleorhabdovirus genera are highly divergent from all other rhabdoviruses including NIAV and hence were excluded from this analysis to maximize phylogenetic resolution with respect to NIAV. Notably, in this analysis NIAV falls as a distinct lineage within the major group of animal rhabdoviruses, which includes a number of recognized and proposed viral genera. While NIAV is most closely related to Oak Vale rhabdovirus (OVRV) in this phylogeny, and with relatively good bootstrap support (75%), these viruses are still relatively divergent, indicative of a long evolutionary separation. Although a lack of bootstrap support at the base of this part of the tree prevents fine-scale resolution of the evolutionary history of NIAV, its divergent phylogenetic position, combined with its lack of cross-reaction to other rhabdoviruses in CF tests (see above), strongly suggests that it represents a novel virus species. Indeed, NIAV is clearly more phylogenetically distinct than a number of recognized species within the Rhabdoviridae.

Genome organization and assembly

A sequencing library was prepared from the sample RNA extracted from BHK infected cells, using an Illumina TruSeq RNA v2 kit following the manufacturer’s protocol. The sample was sequenced on a HiSeq 1000 using the 2×50 paired-end protocol. Reads in fastq format were quality-filtered, and any adapter sequences were removed, using Trimmomatic (Lohse et al., 2012) software. The de novo assembly program ABySS (Simpson et al., 2009) was used to assemble the reads into contigs, using several different sets of reads, and k values from 20 to 40. A nearly full-length contig was obtained from a 250,000 read random subset of the total reads that included non-viral and viral reads and a k value of 25. Reads were mapped back to the contig using bowtie 2 (Langmead and Salzberg, 2012), and visualized with the Integrated Genomics Viewer (Robinson et al., 2011) to verify that the assembled contig was correct. About 5% of the reads in the sample mapped to the viral contig, resulting in about 350,000 reads mapped out of about 7.6 million total.

The 11,124 nt NIAV genome comprises partially complementary 3′- and 5′-terminal non-coding regions (approximately 57 nt and 54 nt, respectively) and five coding regions (genes), each bounded by putative transcriptional regulatory sequences (Figure 3A-C). The transcription termination/polyadenylation (TTP) signals ([G/U]UAC[U]₇) are relatively conserved and similar to those of most other animal rhabdoviruses (Figure 3B). However, transcription initiation (TI) sequences, although also similar to those of many other rhabdoviruses, are more cryptic and apparently quite variable, with only the first three nucleotides (UUG) fully conserved and alternative initiation sites appear to be available in the first three coding regions. Uniquely among those rhabdoviruses sequenced to date, the gene 2 TI sequence appears to be located upstream of the gene 1 TTP sequence to create an overlap that is more commonly associated with rhabdovirus L genes. Gene 1 contains a single ORF encoding a 426 amino acid (aa) polypeptide (47.4 kDa) with identifiable sequence identity to the N proteins of other rhabdoviruses. Gene 2 also contains a single ORF encoding a 297-aa acidic polypeptide (33.6 kDa; pI 4.78) that exhibits the highest sequence similarity to the TUPV P protein and a similar size and charge distribution to the P proteins of other animal rhabdoviruses. Gene 3 contains two consecutive ORFs: the first ORF encodes a 187-aa polypeptide (20 kDa) that aligns with the M proteins of other animal rhabdoviruses; and the second ORF (U1) encodes an 81-aa polypeptide (9.4 kDa) that shows no significant sequence identity with any other known protein and is of unknown function. The termination codon of the M ORF and the initiation codon of the U1 ORF overlap by a single nucleotide, resulting in a -1 shift in reading frames. Gene 4 encodes the 511-aa NIAV G protein. It displays the common features of rhabdovirus G proteins which are class I transmembrane glycoproteins containing an N-terminal signal peptide, a C-terminal transmembrane domain, multiple N-glycosylation sites and a set of highly conserved cysteine residues (Figure 4). Multiple sequence alignments indicated that the NIAV G protein is most similar to those of TUPV and DURV with which it shares ten of the twelve cysteine residues (C₁ to C_XII) that are highly conserved amongst animal rhabdoviruses (Walker and Kongsuwan, 1999) (Figure 4A). Unlike those rhabdoviruses analyzed to date, the NIAV contains an alternative ORF in the N-terminal region of the G gene. ORF U2 encodes a small 97-aa hydrophobic protein (10.8 kDa) with two predicted transmembrane domains and the predicted topology of a class IVb transmembrane protein with a C-terminal ectodomain and no predicted N-glycosylation sites (Figure 4B). The function of the NIAV putative U2 is presently unknown.

Significantly, NIAV shared with DURV and TUPV (but not OVRV) 10 of the 12 cysteine residues that are highly conserved among animal rhabdoviruses. Each of these viruses lacks the C_VIII-C_X disulphide bridge which, in the VSIV G protein, links the second segment of pleckstrin homology domain (domain III) to helix C which sits just before the junction with the fusion domain (domain IV)(Roche et al., 2006; Roche et al., 2007). The NIAV G protein does have an additional pair of cysteine residues downstream of C_XII in the lateral domain (domain I). These cysteine residues, which are likely to form a disulphide bridge, are absent from the tupavirus G proteins. They are in a region that appears to correspond to the [s’t] loop of VSINV which is exposed on the G protein surface and is the location of an antigenic site (Vandepol et al., 1986). NAIV gene 5 encodes the L protein (RNA-dependent RNA polymerase). Also unique among known rhabdoviruses is that an alternative ORF (U3) overlaps the N-terminal region of the L ORF by 142 nucleotides. ORF U3 encodes a highly basic 51-aa polypeptide (6.4 kDa; pI 11.65) of unknown function. U3 is the first ORF that would be detected by ribosomal scanning on the L mRNA suggesting that it may attenuate L ORF expression as appears to have been achieved in some other rhabdoviruses by an overlap with the gene upstream (Tordo et al., 1986) (reviewed in (Walker et al., 2011)).

Immunodetection of NIAV proteins

Western immunoblotting was conducted using NIAV MIAF and control pre-immune serum on NIAV- and mock-infected Vero cells. At 4 days post infection (dpi), infected whole-cell extracts contained seven proteins that were not present in mock-infected cell extracts (Figure 5). Identification of the putative viral proteins encoded by the genomic nucleotide sequence was based on their predicted sizes. Since the bands are not individually identified (e.g. against protein specific antisera), they are referred to below as putative proteins. The largest of these proteins is presumed to be the L protein, which was predicted to have mass of 243.2 kDa. The second largest is the G protein with a calculated size of 59.9 kDa (55.9 with the signal peptide removed and unglycosylated). This protein contains two predicted N-glycosylation sites each adding approximately 3.5 kDa, which corresponds to the observed size of approximately 63 kDa (Figure 5). The third largest protein is the N protein with a calculated size of 47.4 kDa. The P protein, with a calculated size of 33.6 kDa is the fourth largest protein. This protein is usually very immunogenic and reacts strongly with MIAF. However, its strong charge characteristics and alternating acidic and basic domains and high proportion of proline residues render its electrophoretic characteristics in SDS-gels very inaccurate. Thus, conclusive identification of the P band is difficult using MIAF in the absence of other information, as either of the two strong bands migrating at approximately 35 kDa and 45 kDa could be P (Figure 5). The other may be a digested fragment of a larger protein (probably G). The M protein with a calculated size of 18kDa is the fifth largest protein. Resolution of the length of the M protein may be unreliable to determine by SDS-PAGE alone for which size estimations can be influenced by structural features. Furthermore, in this MIAF-probed immunoblot there is no evidence for expression of the putative U1, U2, or U3 ORFS (Figure 3A, B), and such evidence will require additional work. Lastly, the absence of any identifiable bands in the mock-infected cell lysates probed with NIAV-raised MIAF or the mock- and NIAV-infected cell lysates probed with a control pre-immune mouse serum, suggests that the additional bands beyond the 5 that match the predicted sizes of the canonical ORFs (N-P-N-G-L) in the NIAV genome sequence are due to non-specific reactivity of the NIAV-MIAF.

NIAV pathology in mice

Three litters of newborn (1-2 day old) ICR mice with an average size of 10 pups were inoculated intracerebrally (i.c.)(15-20μL each), intraperitoneally (i.p.)(100μL each) or subcutaneously (s.c.)(100μL each) with a stock of Vero-grown NIAV (DakArD 88909) virus containing ~10^6.5 TCID₅₀/mL. The animals were observed daily for illness and were euthanized when they became severely ill. By day 3 post-inoculation, all members of the i.c.-inoculated group were dead or dying. The groups inoculated i.p.- or s.c. developed signs of illness (lethargy, loss of balance, spastic movements) more slowly, but all were dead or very sick by day 7. In contrast, adult ICR mice injected i.p. with a 10% brain suspension from infected newborn mice survived and did show signs of illness They subsequently developed high levels of CF antibody. Collectively, these observations suggest that the susceptibility of mice to NIAV disease is age related.

Sequentially cut thin sections of tissues (brain, lung, liver, spleen and kidney), collected from three sick i.p.-infected mice, were stained with hematoxylin and eosin (H&E) for histopathology and for immunohistochemistry, using a NIAV-specific polyclonal antibody (mouse immune ascitic fluid). Sections of the brain showed numerous apoptotic bodies in the granular layer of the cerebellum (Figure 6A, black arrow). Purkinje cells in many sections were degenerating with eosinophilic nuclei and vacuolated cytoplasm (Figure 6A, blue arrow). There was perivascular cuffing (Figure 6B) and meningitis (Figure 6C). Purkinje cells demonstrated lightly esosinophilic areas in the cytoplasm most consistent with viral inclusions (Figure 6D). Immunohistochemistry showed NIAV antigen - positive neurons in the cerebrum (Figure 6E) and in the cerebellum including Purkinje cells (Figure 6F-G).

Sections of the heart, lung and kidney showed no significant histopathologic changes. Extramedullary hematopoiesis was present in the liver and spleen, but this was likely due to the young age of the animals. No abnormalities were identified in the liver or spleen.

Virus source

Niakha virus (DakArD 88909) was isolated from a single pool of 200 mixed sandfly species (Phlebotomus duboscqi and Sergentomyia sp.) collected in light traps near a ground pool at Niakha in the district of Barkedji in the Ferlo region (15°17′ N; 14°53′ W) of Senegal in March 1992 (Fontenille et al., 1994). The original isolation was made at the Institut Pasteur de Dakar by intracerebral inoculation of newborn mice with a homogenate of the sandfly pool. Initial serological studies at the Institut Pasteur de Dakar failed to demonstrate an antigenic relationship of DakArD88909 with 80 other African arboviruses (Fontenille et al., 1994); a sample of the virus of unknown passage history was subsequently sent by Dr. Jean Pierre Digoutte to the World Reference Center for Emerging Viruses and Arboviruses (WRCEVA) for further study. Stocks were prepared after 3 passages in Vero cells. Because NIAV plaques poorly in Vero or BHK cells, quantification was based on its cytopathic effect (CPE) in serial 10-fold dilutions of Vero cells (TCID₅₀).

Cells

Four different cell lines [BHK-21 (baby hamster kidney), Vero E6 (African green monkey kidney), C6/36 (Aedes albopictus) and PP-9 (Phlebotomus papatasi)] were used in this study. Each was originally obtained from the American Type Culture Collection (ATCC) (Manassas, VA). The PP-9 cell line was initiated by Robert Tesh and Govind Modi from embryonated eggs of Phlebotomus papatasi, using techniques described previously (Tesh and Modi, 1983). Monolayer cultures of vertebrate (BHK, Vero E6) or insect (PP-9) cells in T-25 flasks were inoculated with 100 μl of ~10^5.5 TCID₅₀ of a Vero stock. After 2 hours of incubation at 37°C (vertebrate cells) or 28°C (insect cells), maintenance medium (Vasilakis et al., 2013) was added and the culture was maintained at either 37°C or 28°C, respectively, and examined daily for evidence of viral cytopathic effect (CPE).

Transmission electron microscopy

For ultrastructural analysis, ultrathin sections of infected cells were fixed for at least 1 h in a mixture of 2.5% formaldehyde prepared from paraformaldehyde powder, and 0.1% glutaraldehyde in 0.05 M cacodylate buffer pH 7.3 to which 0.03% picric acid and 0.03% CaCl₂ were added. The monolayers were washed in 0.1 M cacodylate buffer, cells were scraped off and processed further as a pellet. The pellets were post-fixed in 1% OsO4 in 0.1 M cacodylate buffer pH 7.3 for 1 h, washed with distilled water and en bloc stained with 2% aqueous uranyl acetate for 20 min at 60° C. The pellets were dehydrated in ethanol, processed through propylene oxide and embedded in Poly/Bed 812 (Polysciences, Warrington, PA). Ultrathin sections were cut on a Leica EM UC7 ultramicrotome (Leica Microsystems, Buffalo Grove, IL), stained with lead citrate and examined in a Philips 201 transmission electron microscope at 60 kV.

Preparation for vRNA extraction

BHK cell monolayers, grown to 90% confluence in a T150 flask, were inoculated with approximately 4 ml of DakArD 88909 virus harvested from a T25 flask of BHK cells (as described above). Infected cells were incubated for 1 h with periodic gentle rocking to facilitate virus adsorption at 37°C. Following incubation, the viral inocula were removed, cell monolayers were washed thrice with PBS to remove unadsorbed virus, 12 ml of culture medium (DMEM-high glucose supplemented with 10% FBS, and 50mg/ml penicillin/streptomycin) (Invitrogen, Grand Island, NY) was then added, and the cells were incubated at 37°C. Supernatants were harvested approximately 4 days later when marked CPE was developed and clarified by low speed centrifugation (2,000 X g, 10 min at 4°C). To remove remaining cellular debris, the supernatant was filtered through a 0.45 μM filter (EMD Millipore, Billerica, MA). The clarified supernatant was transferred to a 50 ml tube and treated with a cocktail of DNases [14 U Turbo DNase (Ambion, Austin, TX), 20 U Benzonase (EMD Millipore, Billerica, MA) and 20 U RNase One (Promega, Madison, WI)] for 1 h at 37°C. To further reduce the volume, 24 ml of supernatant was loaded on top of 8 ml of 30% sucrose (in TEN, pH 7.4) located on the bottom of a SW28 Beckman UltraClear tube, and centrifuged at 150,000 x g for 4 h at 4°C. The pellet was resuspended in 250 μl RNase/DNase and protease-free water (Ambion, Austin, TX). vRNA was then extracted through the Trizol method and resuspended in 50 μl RNase/DNase and protease-free water (Ambion, Austin, TX).

Next generation sequencing

Genomic analysis

BLASTX was used to identify sequence homologies with other viruses. Open reading frames were identified using MacVector® version 12.0 (Accelrys, Cary, NC). The presence of conserved protein domains was determined using the SMART webserver (Letunic et al., 2012; Schultz et al., 1998)

Nucleotide sequence accession numbers

The complete genome sequence of NIAV was determined in this study and assigned GenBank accession number KC585008. The GenBank accession numbers for the genome sequences of other rhabdoviruses used in the phylogenetic analyses are as follows: ARV, Adelaide River virus (AFR23540); ABLV, Australian bat lyssavirus (NP478343); ARAV, Aravan virus (ABV03822); BEFV, bovine ephemeral fever virus (NP065409); CHPV, Chandipura virus (AED98393); COCV, Cocal virus (ACB47438); CPV, coastal plains virus (ADG86364); DMelSV, Drosophila megalomaster sigmavirus (YP003126913); DObSV, Drosophila obscura sigmavirus (ACU65444); DURV, Durham virus (ADB88761); DUVV, Duvenhage virus (AFK93192); EVEX, Eel virus European X (AFX58972); EBLV1, European bat lyssavirus 1 (YP001285392); FLAV, Flanders virus (AAN73288); HIRV, Hirame rhabdovirus (NP919035); IKOV, Ikoma lyssavirus (AFQ26098); IHNV, infectious hematopoietic necrosis virus (NP042681); IRKV, Irkut virus (ABV03823); ISFV, Isfahan virus (CAH17548); JURV, Jurona virus (AEG25349); KHUV, Khujand virus (ABV03824); KIMV, Kimberley virus (AFR67096); KOTV, Kotonkan virus (YP006202628); LBV, Lagos bat virus (AFW16650); MARAV, Maraba virus (AEI52253); MOKV, Mokola virus (YP142354); MOUV, Moussa virus (ACZ81407); NGAV, Ngaingan virus (YP003518294); OVRV, Oak Vale virus (AEJ07650); OBOV, Obodhiang virus (YP0062000965); OZEV, Ozernoe virus (ACS70797); PRV, Perinet virus (AEG25355); PFRV, Pike fry rhabdovirus (ACP28002); RABV, rabies virus (ABN11300); SBV, Shimoni bat virus (ADD84511); SCRV, Siniperca chuatsi rhabdovirus (YP802942); SHRV, snakehead virus (NP050585); SVCV, spring viraemia of carp virus (NP116748); TIBV, Tibrogargan virus (ADG86355); TUPV, Tupaia virus (YP238534); VSAV, Vesicular stomatitis Alagoas virus (ACB47443); VHSV, viral hemorrhagic septicemia virus (BAM29126); VSIV, Vesicular Stomatitis virus, Indiana (NP041716); VSNJV, Vesicular stomatitis New Jersey virus (AAA48442); WCBV, West Caucasian bat virus (ABV03821); and WONV, Wongabel virus (YP002333280). The ICTV has not yet recognized NIAV as a member species in the family Rhabdoviridae. As a consequence, its abbreviation should be considered tentative and subject to approval.

Phylogenetic analysis

The L protein sequence of NIAV was compared with those of 44 other rhabdoviruses downloaded from GenBank. Because of their excessive sequence divergence to the other rhabdoviruses including NIAV, members of the Cytorhabdovirus, Novirhabdovirus and Nucleorhabdovirus genera were excluded from this analysis; their inclusion resulted in reduced phylogenetic resolution at internal nodes of the tree, including those relating to NIAV (not shown). All protein sequences were aligned using MUSCLE (Edgar, 2004) under default settings. Ambiguously aligned regions, which may negatively effect phylogenetic analysis, were removed using the Gblocks program (Talavera and Castresana, 2007); this resulted in a final sequence alignment of 1110 amino acid residues. The phylogenetic relationships among these sequences were determined using the maximum likelihood (ML) method available in the PhyML 3.0 program (Guindon et al., 2010), employing the WAG+Γ model of amino acid substitution and subtree pruning and regrafting (SPR) branch-swapping. The support for each node was determined using 1,000 bootstrap replicates.

Antigens and immune reagents

Antigens used in complement fixation (CF) tests and for immunizing animals were prepared from infected newborn mouse brains. CF antigens were prepared by the sucrose/acetone extraction method (Clarke and Casals, 1958). Specific hyperimmune mouse ascitic fluids were prepared against each of the 24 rhabdoviruses listed in Table 1. The immunization schedule consisted of four intraperitoneal injections given at weekly intervals. Immunogens consisted of 10% suspensions of homogenized infected mouse brain in PBS mixed with Freund’s adjuvant just prior to inoculation. Sarcoma 180 cells were also given intraperitoneally after the final immunization to induce ascites formation. All animal experiments were carried out under an animal protocol approved by the University of Texas Medical Branch IACUC committee.

Serologic testing

Complement fixation tests were conducted using a microassay described previously (Tesh et al., 1983), using 2 full units of guinea-pig complement. Titers were recorded as the highest antibody dilutions giving 3+ or 4+ fixation of complement on a scale of 0–4+.

Immunoblotting

BHK monolayers were grown to 90% confluency in 25 cm² tissue culture flasks and were infected with NIAV or were mock infected as described above. After 2 h, the virus inoculum was aspirated from the cells, unabsorbed virus was removed by thrice washing, and fresh media (DMEM) (Invitrogen) supplemented with 5% fetal bovine serum (FBS) (Atlanta Biologicals, GA) and 1% penicillin/streptomycin (Invitrogen, Grand Island, NY) was added. Four days after infeciton (p.i.) and incubation at 37°C, when marked CPE had developed, the cells were harvested into cell lysis buffer (50mM Tris-HCl, pH 8.0, 150 mM NaCl, 1% Triton X-100). The cell preparations were then treated according to the Biorad instructions for MiniPROTEAN TGX gel analysis of proteins. Briefly, 20 μl sample loading buffer containing β-mercaptoethanol were added to 20 l aliquots of the cell preparations. Samples were heated at 95°C for 10 min, and centrifuged for 1 min at 500 x g at 4°C. Proteins were separated under reducing conditions on MiniPROTEAN TGX 10-20% Tris-Glycine SDS-PAGE gels, and were transferred onto Hybond-P PVDF membrane (GE Healthcare Biosciences, Pittsburgh, PA)) with Transfer Buffer (Biorad, San Diego, CA) according to the Biorad western blotting transfer protocol. Membranes were processed by the WesternBreeze© Chromogenic Immunodetection system (Invitrogen, Grand Island, NY) following the manufacturer’s protocol. The control pre-immune mouse serum and polyclonal MIAF raised to NIAV were used at 1:5,000 dilution, whereas the secondary, goat anti-mouse alkaline phosphatase-conjugated antibody was used as was pre-diluted by the manufacturer’s WesternBreeze© Chromogenic Immunodetection system (Invitrogen, Grand Island, NY).

Animal studies

All animal studies were carried out under a protocol approved by the UTMB Institutional Animal Care and Use Committee. Approximately 10 pups from each of three litters of newborn (1-2 day old) ICR mice were inoculated intracerebrally (i.c.), intraperitoneally (i.p.) or subcutaneously (s.c.) with a Vero stock of NIAV containing ~10^6.5 TCID₅₀/mL. Each of the i.c.-inoculated pups received 15-20μl of virus, whereas 100μl of virus was injected in each of the i.p. and s.c.-inoculated pups. The animals were observed daily for illness and were euthanized when they became seriously ill. Tissues (brain, lung, liver, spleen and kidney) were collected from three of the sick i.p.-infected mice for histopathology.

Tissues were fixed in 10% neutral buffered formalin (RICCA Chemical Company, Arlington, TX) for 24 hours and transferred to 70% ethanol for storage and subsequent embedding in paraffin. Sections (5μm) were cut and stained by the hematoxylin and eosin method.

Immunohistochemistry

Slides were deparaffinized in Xylene for 15 min, followed by rehydration in ethanol and ethanol/water mixtures as follows: 100% ethanol for 9 min, 95% ethanol/5% deionized water for 3 min, 80% ethanol/20% deionized water for 5 min. Antigen retrieval was facilitated by Protease XXIV (BioGenex, Fremont, CA) incubation for 5 min at RT, followed by three Tris-buffered saline containing 0.1% Tween 20 (TBST) washes. Further processing of the slides was according to the VECTASTAIN® ABC AP Kit (Burlingame, CA) protocol. Incubation with a NIAV-specific hyperimmune mouse ascitic fluid was performed overnight at 4°C. Developed slides were counterstained by the hematoxylin and eosin method and mounted with DePex (EMS, Hatfield, PA).