Abstract
Two strains of viral hemorrhagic septicemia virus (VHSV) with known different virulence characteristics in vivo were studied (by a time course approach) for their abilities to infect and translocate across a primary culture of gill epithelial cells (GEC) of rainbow trout (RBT; Oncorhynchus mykiss). The strains included one low-virulence marine VHSV (ma-VHSV) strain, ma-1p8, and a highly pathogenic freshwater VHSV (fw-VHSV) strain, fw-DK-3592B. Infectivities toward trout head kidney macrophages were also studied (by a time course method), and differences in in vivo virulence were reconfirmed, the aim being to determine any correlation between in vivo virulence and in vitro infectivity. The in vitro studies showed that the fw-VHSV isolate infected and caused a cytotoxic effect in monolayers of GEC (demonstrating virulence) at an early time point (2 h postinoculation) and that the same virus strain had translocated over a confluent, polarized GEC layer by 2 h postinoculation. The marine isolate did not infect monolayers of GEC, and delayed translocation across polarized GEC was seen by 48 h postinoculation. Primary cultures of head kidney macrophages were also infected with fw-VHSV, with a maximum of 9.5% virus-positive cells by 3 days postinfection, while for the ma-VHSV strain, only 0.5% of the macrophages were positive after 3 days of culture. In vivo studies showed that the fw-VHSV strain was highly virulent for RBT fry and caused high mortality, with classical features of viral hemorrhagic septicemia. The ma-VHSV showed a very low level of virulence (only one pool of samples from the dead fish was VHSV positive). This study has shown that the differences in virulence between marine and freshwater strains of VHSV following the in vivo infection of RBT correlate with in vitro abilities to infect primary cultures of GEC and head kidney macrophages of the same species.
Viral hemorrhagic septicemia virus (VHSV) belongs to the Novirhabdovirus genus of the family Rhabdoviridae. European VHSV of freshwater origin causes disease primarily in rainbow trout (RBT; Oncorhynchus mykiss), with high mortality. Outbreaks of viral hemorrhagic septicemia (VHS) in marine fish species like turbot (Scophthalmus maximus) (17, 22, 23) and Japanese flounder (Paralichthys olivaceus) (10) have also been recorded previously. VHSV has been implicated in the mortality of Pacific herring (Clupea pallasi) (15), and in later studies it was shown that this species of fish is highly susceptible to experimental challenge (11). Infection trials have shown previously that marine VHSV (ma-VHSV) isolates originating from marine fish have low levels of pathogenicity in RBT compared with freshwater VHSV (fw-VHSV) strains (24). However, no studies have been carried out to determine the mechanisms for these differences at the level of host-pathogen interactions.
Fw-VHSV strains infect through waterborne exposure, and the prime port of entry is suggested to be the skin and/or gills. This supposition is based on previous observations in which early virus replication in gill epithelial cells (GEC) of RBT in situ has been demonstrated (1, 4, 18). Further, studies have shown that epithelial cells from skin and gills are capable of supporting the replication of VHSV (31) and that the viral replication in excised fins correlates with resistance to waterborne challenge (21). The progression of an infectious hematopoietic necrosis virus (IHNV) infection was suggested to occur from the gills into the circulation and/or from the oral region into the gastrointestinal tract, with subsequent distribution into the circulation (5). The replication of IHNV in internal organs at early stages was proposed previously to take place in the kidney (5). This suggestion is concordant with findings reported for both VHSV and IHNV in RBT (1) and for ma-VHSV in turbot (2) based on the results of immunohistochemistry analyses (2, 7). In these previous studies, we showed that VHSV and IHNV were detected in macrophages and melanomacrophages at early times postinfection (p.i.), indicating that these cells support virus replication in vivo. The replication of VHSV and IHNV in macrophages cultured in vitro has also been demonstrated previously (3, 6). In a recent study, the fin base was identified as a possible port of entry for IHNV (9).
In studies by Skall et al. (24), 139 ma-VHSV isolates from wild marine fish and from farmed turbot did not cause mortality in RBT after challenge by immersion, but some isolates caused up to 60% mortality after injection. In this study, we have investigated if the low level of virulence of ma-VHSV in RBT can be related to the ability of ma-VHSV to translocate over a confluent gill epithelium. Pärt and colleagues (20) have developed a method for the in vitro culturing of GEC on filters. The GEC in filter cultures establish a monolayer firmly attached to glass and plastic supports. These cells have the appearance of a differentiated epithelium, and tight junctions are established (20). This method has been used in studies of ion transport and acid-base regulations in RBT (30). In this experiment, we used this in vitro model to study the translocation of VHSV through a confluent gill epithelium and we compared an ma-VHSV isolate (ma-1p8, genotype 1b) to a virulent fw-VHSV reference strain (fw-DK-3592B, genotype 1a). The marine strain was selected on the basis of being nonlethal to RBT following immersion challenge while causing around 40% mortality when injected intraperitoneally into RBT. The same isolates were also tested for their abilities to infect isolated head kidney macrophages in vitro, and the virulence characteristics of the isolates used were again confirmed by an experimental challenge of RBT fry.
MATERIALS AND METHODS
Preparation of GEC.
RBT GEC were isolated according to the method described by Pärt et al. (20). RBT of approximately 25 to 100 g cultured in freshwater were killed by a blow to the head and decapitated. The gill arches were dissected, transferred into petri dishes, and rinsed twice with 10 ml of phosphate-buffered saline (PBS; Ca2+ and Mg2+ free) containing 200 μg of penicillin-streptomycin (PEST; Gibco)/ml, 400 μg of gentamicin (Gibco)/ml, and 200-μg/ml amphotericin B (prepared from 250-μg/ml Fungizone; Gibco).
Coagulated blood was gently removed from the gills, and the filaments were excised from the arches and washed twice in 10 ml of the PBS solution. The filaments were thereafter transported to the lab on ice, transferred into 5 ml of trypsin (a 2.5% solution from Gibco), and incubated on a shaker (with shaking at 120 to 200 rpm) for 20 min.
The cell suspension was aspirated from the tubes and filtrated through a 100-μm nylon filter into a stopping solution, PBS containing 10% Gibco fetal bovine serum (FBS). Remaining filaments were trypsinized in 5 ml of 2.5% trypsin solution for an additional 20 min and filtered into the same stopping solution. The cell suspension was then centrifuged at 200 × g for 10 min at 4°C, and the cell pellet was resuspended in culture medium (Leibowitz L-15 medium supplemented with 2 mM l-glutamine, 5% FBS, 100 μg of PEST/ml, and 200 μg of gentamicin/ml). The cells were then transferred into 75-cm2 Falcon culture flasks with GEC culture medium (Leibowitz L-15 medium with 2 mM glutamine, 5% FBS, 100 μg of PEST/ml, and 200 μg of gentamicin/ml), and the flasks were incubated at 20°C. After 24 h of incubation, the flasks were rinsed twice with PBS to remove unattached cells. Culture medium was changed every second day until the experiment started.
Virus propagation.
The virulent freshwater strain fw-DK-3592B (genotype Ia) of VHSV, isolated from RBT in Denmark (24), or the marine strain ma-1p8 (genotype Ib), isolated from herring (Clupea harengus) in the Baltic Sea (16, 24), was grown on cells of the bluegill fry caudal trunk cell line BF-2 in Eagle's minimum essential medium with Earle's balanced salt solution supplemented with 16.4 mM Tris buffer, 5.3 mM NaHCO3, 10% FBS, 4 mM l-glutamine, and 50 μg of gentamicin ml−1. A stock of fw-DK-3592B with a dose of 8.66 log 50% tissue culture infective doses (TCID50) ml−1 and a stock of ma-1p8 with a dose of 7.37 log TCID50 ml−1 were frozen in vials at −80°C.
Infection assay of GEC.
Adherent GEC were trypsinized 5 to 7 days after isolation and transferred at concentrations of 400,000 to 500,000 cells/well into 24-well plates covered with 13-mm-diameter plastic coverslips. The GEC were cultured for 2 to 3 days. At the time of infection, frozen vials with virus were thawed and GEC were inoculated with 105 TCID50 of the fw-DK-3592B strain or the ma-1p8 strain/ml.
On each day from days 1 to 6 postinoculation, three wells of inoculated cells per virus strain were fixed by the addition of 1 ml of 80% acetone and left for 2 min. Fixed cell layers were then washed with Tris-buffered saline (TBS) and incubated with the monoclonal antibody (MAb) IP5B11 (12) against VHSV N protein for 1 h at 37°C. Coverslips were subsequently washed with TBS and incubated for 30 min with rabbit anti-mouse immunoglobulin G (Dako), followed by an immunocomplex of alkaline phosphatase and mouse monoclonal anti-alkaline phosphatase (APAAP; Dako) for 30 min. After washing, fast red (1 g/liter; Sigma, St. Louis, MO) and naphthol AS-MX-phosphate (0.2 g/liter; Sigma) with 1 mM levamisole (Sigma) in 0.1 M TBS were added for development for 20 min. Samples were counterstained with hematoxylin and mounted with Aquamount (BDH Laboratory). The numbers of infected cells were estimated by evaluating 200 counted cells on each of three parallel coverslips at 1, 2, 3, and 6 days postinoculation with virus.
GEC viability assay.
After 3 to 7 days of culturing, the adherent GEC were trypsinized and transferred into 96-well plates at a concentration of 40,000 to 50,000 cells/well in 200 μl of GEC culture medium. The GEC were subcultured for an additional 2 to 3 days before the removal of the medium and the addition of 100 μl of a 106-TCID50/ml preparation of fw-DK-3592B or ma-1p8. Zero to 7 days after inoculation, 20 μl of CellTiter AQueous One solution reagent (Promega) was added to each well containing 100 μl of medium. The cells were incubated for an additional 24 h at 15°C. The quantity of formazan product in each well was measured at 490 nm by using a 96-well absorbance plate reader and was directly proportional to the number of living cells in the culture well.
Transepithelial passage in GEC.
GEC were grown in culture flasks for about 5 to 9 days, after which the cells were trypsinized and resuspended in medium. A sample was counted by using KOVA Glasstic slide 10 with grids. A quantity of 200,000 cells was added to each filter (Falcon 0.4-μm-pore-size polyethylene terephtalate track-etched membrane) of a cell culture insert. The cells were cultured on cell culture chamber inserts with porous-bottomed dishes providing virus access to pass through the membrane surface (Fig. 1). The medium was frequently changed, ensuring good cell growth. The transepithelial resistance was measured daily by using a Millicell-ERS meter with chopstick electrodes (Millipore Co., Bedford, MA). Results were expressed as kilo-ohm-square centimeters. The infection of cells was initiated when resistance exceeded 1 kΩ·cm2 (usually after 24 to 48 h of incubation). GEC inserts were inoculated with 105 TCID50 of ma-1p8 or fw-DK-3592B/ml. At 2, 4, 8, 24, and 48 h, a sample of 100 μl from the basal side of the filter in each of three replicate cell culture chambers was collected into a medium vial (Fig. 1). The samples were transferred at a 1:100 dilution onto a subconfluent layer of BF-2 cells (20 to 48 h old) cultured on 24-well plates. The inoculated cells were incubated at 15°C and inspected for up to 7 days after inoculation for the occurrence of cytopathic effects (CPE) following primary incubation. After 7 days of culture, cell culture medium (supernatant) was passed onto fresh (20- to 48-h-old) BF-2 cells cultured on 24-well plates (second passage). Inoculated cells were inspected for the occurrence of CPE during the secondary incubation period, lasting up to 7 days after inoculation. CPE consisted of cell rounding of focal appearance, followed by the detachment of cells from the plastic surface. Virus was identified by using a standard enzyme-linked immunosorbent assay (ELISA) technique with MAb IP5B11 as the primary antibody (19).
FIG. 1.
Layout of the filters on which the GECs were grown. The setup mirrors the water-to-blood barrier of the gills of RBT.
In vitro infection of RBT macrophages.
Disease-free RBT (with no history of disease during the life cycle of the fish) of 30 to 80 g cultured in freshwater were anesthetized with chlorobutanol (1 g/liter) and killed by decapitation. The head kidney was dissected out and transferred into incubation buffer (saline solution buffered with 20 mM HEPES [pH 7.4]). The head kidney was torn into pieces with two sterile needles and gently pressed through a nylon mesh (pore size, 0.3 mm) with a glass rod. The cell fraction was concentrated by centrifugation and resuspended in incubation buffer. Homogenous cell fractions were obtained by discontinuous Percoll (Pharmacia) gradient density centrifugation (2,060 × g). Cells concentrated in the bands between the fraction densities of 37 and 54% Percoll were diluted in L-15 medium containing penicillin at 100 U/ml, streptomycin at 100 μg/ml, and kanamycin at 100 μg/ml (all from Gibco) to give a final cell concentration of 106 cells/ml. Aliquots of 1 ml of this solution were seeded onto 12-mm Ø coverslips in culture wells, and the coverslips were incubated at 14°C in air. The cell medium was changed, and the wells were washed twice with PBS to remove the nonadherent cell fraction. The following day (after 24 h of incubation), the wells were inoculated with identical concentrations, 3.9 × 107 infective particles, of the fw-DK-3592B strain or the ma-1p8 strain, and the number of infected cells was estimated by evaluating 200 counted cells on two parallel coverslips incubated for 1, 2, 3, or 6 days p.i. The cell layers were fixed in 80% acetone, washed with TBS, and incubated with MAb IP5B11 for 1 h at 37°C. Coverslips were then washed with TBS and incubated for 30 min with rabbit anti-mouse immunoglobulin G (Dako), followed by APAAP (Dako) for 30 min. After washing, fast red (1 g/liter; Sigma, St. Louis, MO) and naphthol AS-MX-phosphate (0.2 g/liter; Sigma) with 1 mM levamisole (Sigma) in 0.1 M TBS were added for development for 20 min. Samples were counterstained with hematoxylin and mounted with an aqueous mounting medium (Aquamount; BDH Laboratory).
Macrophage viability assay.
A 200-μl mixture of 8 μM ethidium homodimer (which fluoresces red to visualize dead cells) and 2 μM calcein AM (which fluoresces green to visualize live cells; both from Molecular Probes) was added to each well and diluted in PBS. Incubation was for 30 to 120 min before the coverslips were evaluated in an inverted fluorescence microscope (Leica). At 6 days postinoculation, the numbers of live and dead cells were estimated by assessing 100 counted cells on eight parallel coverslips.
In vivo virulence study with RBT.
The fw-DK-3592B and ma-1p8 strains were tested for their in vivo virulence in RBT fry. Both isolates were propagated and titrated on the BF-2 bluegill fry caudal trunk cell line (28) according to standard procedures (12). Both isolates used had low passage numbers (maximum, five passages). Specific-pathogen-free RBT reared at the laboratory, with an average weight of approximately 0.2 g (determined by weighing 15 fish), were used in the trial. Approximately 200 fish per tank were transferred into small tanks numbered 1 through 3, each containing 8 liters of softened tap water, with the temperature maintained at 9 to 11°C throughout the experiment. The challenge dose was 1.6 × 105 TCID50/ml for ma-1p8 (tank 3) and 2.9 × 103 TCID50/ml for fw-DK-3592B (tank 2), and virus-free medium was used in the negative control (tank 1). The fish were challenged for 2 h in aerated water by immersion, after which water flow was resumed. Mortality was recorded daily for up to 42 days p.i. Samples were obtained from fish that died during the trial and from the survivors, including those in the negative control group, and used for virus isolation. Samples from fish in the negative control group and the group infected with ma-1p8 that died on the same day were all pooled (for up to seven fish per pool). The pooled sample was frozen at −25 or −80°C until examination. At termination, samples from 25 fish from each group were combined into three pools and examined for virus as described previously (16).
Virus distribution and quantification.
For tanks 1 to 4, sequential samples from six fish from each group were collected on day 0 (before the onset of infection) and 1, 2, 3, 5, 7, 10, 14, and 21 p.i. Fish were culled for immunohistochemistry analysis (three fish) and virus quantification (three fish). Immunohistochemistry analysis was performed with slides of paraffin-embedded specimens (14), which were examined as described previously (8). Fish were thawed in cold water for virus quantification and weighed, and Eagle's minimum essential medium was added to give a dilution of 1:10 (wt/vol). The fish were grounded with a mortar and pestle and sterile sand. Hexamicin was added, and the samples were kept at 4°C overnight and titrated on BF-2 cells in 96-well plates (NUNC), with the calculation of the titer after 7 days. Samples that did not produce CPE upon the first passage were subcultured by one additional passage.
RESULTS
Virus infectivity and CPE in GEC.
The infectivity of strain ma-1p8 in GEC was assessed by immunohistochemistry analysis, and any CPE were assessed on the basis of morphological changes as defined previously. GEC were infected with a dose of 105 TCID50/ml in each well, and over a period of 7 days, cells were shown to be negative by immunocytochemistry and no CPE were observed. For fw-DK-3592B, 23% of the GEC were found to be infected by day 1 p.i. by immunocytochemical staining (Fig. 2) and 64% were infected by day 2. On days 3 and 4, counting was not possible as no cells were found to be attached to the coverslips because of virus-induced CPE.
FIG. 2.
GECs infected with fw-DK-3592B, shown on day 1 p.i. Single cells were detected to be virus positive in the cytoplasm by alkaline phosphatase staining (red coloring) using MAb IP5B11 and APAAP/fast red. The slide is counterstained with Mayer's hematoxylin (blue nuclei). Bar, 10 μm.
GEC viability assay.
The virulence of the ma-1p8 isolate was also evaluated by assessing the viability of ma-1p8-infected GEC by a CellTiter 96 AQueous One solution cell proliferation assay, used for determining the numbers of viable cells in cytotoxicity assays. As this assay is considered to be a more sensitive method than morphological assessment, it was included to document the cytotoxicity (CPE) induced by ma-1p8 infection. The viability of GEC inoculated with isolate ma-1p8 was not different from that of uninfected cells, while GEC infected with fw-DK-3592B showed a rapid decline in viability, and at 3 days p.i., the viability of these cells was close to zero, indicating full cytolysis (Fig. 3).
FIG. 3.
GEC cultures were infected with strain fw-DK-3592B or ma-1p8 and incubated for 7 days. One parallel culture was left uninfected. Mean absorbance values at 490 nm for GECs incubated with CellTiter AQueous One solution reagent are given, and each value is based on the means of results for 11 wells. The background absorbance level was set at zero.
Transepithelial passage of VHSV in GEC.
In an attempt to mimic in vivo virus invasion across the gill epithelial barrier (Fig. 1), GEC were grown on filters, on which a polarization of the cells is established (19, 29), and the filter insert cultures were subsequently inoculated on the apical side with ma-1p8 (Fig. 1). Samples collected into medium vials from the basal side of three replicate chambers of filter cultures of GEC inoculated with ma-1p8 were virus negative by culture on BF-2 cells and by ELISA at 2, 4, 8, and 24 h p.i. By 48 h p.i., two of three replicate medium vials were virus positive by culture on BF-2 (CPE), and this result was confirmed by ELISA (Table 1). In contrast, fw-DK-3592B-infected GEC in the basal medium were virus positive (one of three replicates) at 2 h after virus was added to the apical medium. Two of three replicates were positive by 4, 8, and 24 h p.i., and by 48 h p.i., cells in the basal medium in all three replicates were virus positive (Table 1). These findings indicate a rapid translocation of the fw-DK-3592B strain across the polarized epithelial cells.
TABLE 1.
Presence of virus in BF-2 cellsa
| Hours p.i. | No. of wells positive for:
|
|
|---|---|---|
| fw-DK-3592B | ma-1p8 | |
| 2 | 1 | 0 |
| 4 | 2 | 0 |
| 8 | 2 | 0 |
| 24 | 2 | 0 |
| 48 | 3 | 2 |
GEC filter insert cultures were infected with fw-DK-3592B and ma-1p8 on the apical side, and samples were collected on the basal side (see Fig. 1 for details) and examined for the presence of virus in BF-2 cells. The numbers of positive wells (of three per time point) over an incubation period of 48 h after virus was added to the apical side are indicated.
In vitro infection of RBT macrophages.
As findings in previous studies (6, 28) have indicated that macrophages possibly play role in the primary replication of VHSV after the virus bypasses the primary barriers (skin and mucus), head kidney macrophages isolated by gradient centrifugation were infected with the ma-VHSV and fw-VHSV isolates. Interestingly, few macrophages were found to be infected with the fw-DK-3592B strain, with the proportions of infected macrophages in two parallel wells averaging 4% by day 1 and peaking at 8.75% positive cells by culture day 3 (Table 2). By day 6 of culture, the percentage of positive cells had declined to 3.5%. Positive staining for virus was seen in confined structures in the cytoplasm of infected macrophages (Fig. 4). By this time and later, a few macrophages (see the description of the viability assessment below) were found to have a condensed nucleus or to be in lysis, possibly as a result of virus replication. In macrophage cultures infected with the ma-1p8 strain, an average of 0.5% virus-positive cells were found by day 1 and there was no increase in the number of infected macrophages over the culture period (Table 2). By day 6, no cells were found to be positive.
TABLE 2.
Detection of VHSV in cultured macrophages isolated from head kidneys of RBT and subsequently infected with isolate fw-DK-3592B or isolate ma-1p8a
| Days p.i. | % of positive cells infected with fw-DK-3592B on:
|
% of positive cells infected with ma-1p8 on:
|
||
|---|---|---|---|---|
| C1 | C2 | C1 | C2 | |
| 1 | 3.0 | 5.0 | 0 | 1.0 |
| 2 | 4.0 | 8.0 | 0 | 0.25 |
| 3 | 8.0 | 9.5 | 0.5 | 0.5 |
| 6 | 3.5 | 3.5 | 0 | 0 |
The number of infected cells was estimated by assessing 200 counted cells on each of two parallel coverslips (C1 and C2) after the staining of virus-positive cells by using MAb IP5B11 and development with APAAP.
FIG. 4.
Macrophages infected with fw-DK-3592B, shown on day 3 p.i. after cytoplasmic staining (red coloration). Virus antigen was detected by alkaline phosphatase-coupled antibodies visualized by fast red. The slide was counterstained with Mayer's hematoxylin (blue nuclei). Bar, 10 μm.
Viability assays of macrophages.
Since the detection of VHSV by immunoenzyme methods has limited sensitivity (8) and there is a possibility of underestimating the number of virus-infected cells by this method, an evaluation of the viability of macrophages kept in culture was also included, with both infected and uninfected parallels. The viability of macrophages was assessed by staining the cultures with a combination of ethidium homodimer (red fluorescence, indicating dead cells) and calcein AM (green fluorescence, indicating live cells) to visualize live versus dead macrophages in culture.
The viability assay (counting of 100 cells on eight parallel coverslips) showed that 78% of uninfected macrophages were viable by day 6 in culture. This result was not different from that found for cultures infected with strain ma-1p8, while in cultures infected with strain fw-DK-3592B, only 55% of the cells were viable by day 6 p.i. (Fig. 5). By visual examination with a fluorescence microscope, the condensed cells were identified as dead cells (Fig. 6).
FIG. 5.
Mean percentage (± standard error of the mean) of live macrophage cells infected with the fw-DK-3592B or ma-1p8 isolate. Results from 6 days p.i. are given. Ethidium homodimer and calcein were used to stain dead and live cells, respectively. The numbers are based on the results of counting 100 cells in each of eight parallel wells.
FIG. 6.
Viability assay results demonstrated by visualizing dead macrophages with ethidium homodimer (red fluorescence) and live macrophages with calcein AM (green fluorescence). Bar, 10 μm.
In vivo challenge study with RBT.
The in vivo virulence characteristics of isolates fw-DK-3592B and ma-1p8 have been tested previously (24), but since these isolates had been subjected to additional passages in cell culture, a challenge trial with RBT fry was performed to confirm previous results. For isolate fw-DK-3592B, the cumulative mortality was 100% by day 13 and fish died with classical VHS symptoms (Fig. 7). Samples collected from dead fish on days 7 and 12 were all found to be VHSV positive by culture and ELISA. For isolate ma-1p8, the cumulative mortality by day 24 postchallenge was 26%, higher than that for the uninfected controls (15% by day 24). No fish in the ma-1p8 challenge group or the controls died between days 24 and 42.
FIG. 7.
The percentages of cumulative mortality following VHSV challenge with fw-DK-3592B and ma-1p8 strains are shown. Each group was challenged in a separate tank. A separate control tank with nonchallenged fish was also included. Results for fish from which samples were obtained for studies of viral distribution are not included in this figure.
VHS was confirmed by histology and immunohistochemistry analyses, with classical histological changes in target organs such as the kidney in the fw-DK-3592-B group by day 3 p.i. (data not shown). For ma-1p8-infected fish, histopathological changes were not observed at any sample collection times and no virus was detected by immunohistochemistry analysis at any time points postchallenge. By culture, fish were positive for fw-DK-3592B at a low titer (7.1 × 102 TCID50/g of tissue) by day 2 postchallenge, and the titer reached a peak (1.3 × 108 TCID50/g of tissue) by day 7. In the ma-1p8 group, 10 pools of dead fish (pools of 1 to 7 fish, with 31 fish total) were all negative for VHSV by culture. On day 42, three pools of eight, eight, and nine fish were examined and one pool (of eight fish) was virus positive. In the controls, all pools examined were negative for VHSV. Together, these findings indicate high nonspecific mortality in the ma-1p8 and control groups.
DISCUSSION
In previous studies, it has been well-documented that marine isolates of VHSV are nonpathogenic to RBT following immersion challenge, while various rates of mortality (up to 60%) have been found after injection challenge (24). The underlying factors have not been elucidated in any detail. In this study, we confirmed published results that a marine isolate of VHSV (ma-1p8) is very inefficient at infecting fry of RBT following immersion challenge. These findings correlate with a weak ability of the virus to infect and translocate over polarized, primary GEC cultures and the low level of in vitro infectivity of isolate ma-1p8 in primary cultures of head kidney macrophages. These results contrast with the findings for the highly virulent freshwater isolate of VHSV fw-DK-3592B.
The portal of entry for VHSV infection of RBT has not been conclusively proven. We have previously shown that fw-VHSV can be detected in GEC by immunohistochemistry at early time points following an experimental infection; however, virus positivity is limited to a few epithelial cells (1). Similar observations have been made by others (3, 4, 18), but these studies can only serve as indication as to how the virus gains access to the systemic circulation in trout. An interesting observation is the finding that a strain of VHSV (Makah) with a low degree of virulence toward RBT showed low levels of replication in fin tissue and none at all in gill tissue ex vivo (31). Our findings are in concert with this observation.
Several studies have suggested that kidney and spleen macrophages have an important role as target cells for the initial replication of VHSV (4), and previous studies have shown that kidney and spleen macrophages are infected in vivo (1, 7, 8). In this study, we found that the strain fw-DK-3592B infected head kidney macrophages in culture (Fig. 4), also leading to reduced cell viability, i.e., the virus strain was virulent (able to induce cell damage) in isolated head kidney macrophages. In contrast, a very small percentage of the cultured macrophages were infected with ma-1p8 and the number of viable cells was not different from that in the uninfected control (Fig. 5), indicating a lack of virulence toward macrophages. Also, the proportions of cells that were infected and supported viral replication were limited, below 10% for fw-DK-3592B and 0.5% for ma-1p8. There is also a possibility that we have underestimated the number of virus-positive cells, since infected cells are more easily washed off from the slide during incubation for immunohistochemical staining and viability staining. However, our findings are concordant with those of a previous study (28) in which it was shown that in primary cultures of macrophages from RBT, 8% of the cells supported VHSV replication, and in those of macrophages from turbot (Scophthalmus maximus), 1.7% of the cells were found to be virus positive (a virus strain of RBT origin was studied). No significant CPE was observed in experiments performed by Tafalla et al. (28), in contrast to our findings, while Estepa et al. (6) showed that VHSV lyses macrophages from RBT in vitro, which is more in line with what we observed. It thus seems that macrophages can be infected with virulent strains of VHSV (of RBT origin). However, the initial VHSV replication occurs in endothelial cells and, to a lesser degree, in macrophages. Further, there is also a relatively low level of infectivity in macrophages in vitro, and given these findings together, it remains to be documented that macrophages are the most important target cells for VHSV in the early stages of infection.
Concordant with previous findings, ma-1p8 infected fry of RBT at a very low prevalence following in vivo challenge, while the freshwater strain was highly virulent (4). The mortality developed rapidly among the fw-DK-3592B-challenged fish from days 4 to 10 and reached 100% by day 12 p.i. The background mortality in the controls was higher than normal and was possibly associated with the use of fry for challenge. Fish the size of RBT fry are more susceptible to handling and sorting than larger fish, which can account for some of the background mortality. Fry were used since they are considered to be more susceptible to challenge than larger fish, which would provide a sensitive biological test system. Overall, from the results of histological examination, immunohistochemistry analysis, and virus reisolation, there is good documentation provided to state that isolate ma-1p8 was avirulent, with no VHS-related mortality occurring over the course of the experiment. At termination, three sample pools were examined and one of the pools was found to be positive for VHSV by culture, showing that isolate ma-1p8 can infect RBT but without clinical signs or pathology in internal organs, corroborating previous findings by Skall et al. (24).
Over the last 20 years, VHSV strains have been isolated from several marine fish species (25) and these findings have created a concern that marine strains of VHSV may be a potential source of infection for farmed RBT (25). The results of studies performed by Skall et al. (24) demonstrate very clearly that RBT has a very low level of susceptibility to several VHSV strains originating from marine fish when tested by immersion challenge. A total of 139 ma-VHSV isolates from wild marine fish and from farmed turbot did not cause mortality in RBT challenged by immersion, but some isolates caused mortality after injection (24). The low level of virulence of ma-VHSV strains in RBT can be related to their limited ability to pass a confluent gill epithelium or the external barriers to infection. However, one should not rule out the possibility that ma-VHSV can cause disease in RBT in a commercial setting, and the results of previous studies (26, 27) show that the ma-VHSV isolates are genetically closely related to fw-VHSV isolates. Recently, the first known disease outbreak in sea-farmed RBT caused by genotype 3 of VHSV was documented in Norway (13). Immersion challenge trials of this isolate in RBT showed high mortality and provided the first registration of an ma-VHSV of genotype 3 being virulent for RBT (13). This finding is an indication of the possible adaptation of marine isolates of VHSV to a new host. Future studies including investigations of the infectivity and virulence of this new variant of ma-VHSV isolates toward RBT GEC would be interesting endeavors and should be pursued.
Footnotes
Published ahead of print on 27 August 2008.
REFERENCES
- 1.Brudeseth, B. E., J. Castric, and Ø. Evensen. 2002. Studies on pathogenesis following single and double infection with viral hemorrhagic septicamia virus and infectious hematopoietic necrosis virus in rainbow trout (Oncorhynchus mykiss). Vet. Pathol. 39180-189. [DOI] [PubMed] [Google Scholar]
- 2.Brudeseth, B. E., R. S. Raynard, J. A. King, and Ø. Evensen. 2005. Sequential pathology after experimental infection with marine viral hemorrhagic septicemia virus isolates of low and high virulence in turbot (Scophthalmus maximus L). Vet. Pathol. 429-18. [DOI] [PubMed] [Google Scholar]
- 3.Chilmonczyk, S., I. Voccia, and D. Monge. 1995. Pathogenesis of viral haemorrhagic septicaemia virus: cellular aspects. Vet. Res. 26505-511. [PubMed] [Google Scholar]
- 4.de Kinkelin, P., S. Chilmonczyk, M. Dorson, M. Le Berre, and A. M. Baudaouy. 1979. Some pathogenic facets of rhabdoviral infection of salmonid fish, p. 357-375. In P. A. Bachmann (ed.), Proceedings of the 4th Munich Symposium on Microbiology: mechanisms of viral pathogenesis and virulence. WHO, Munich, Germany.
- 5.Drolet, B. S., J. S. Rohovec, and J. A. Leong. 2002. The route of entry and progression of infectious haematopoietic necrosis virus in Oncorhynchus mykiss (Walbaum): a sequential immunohistochemical study. J. Fish Dis. 17337-347. [Google Scholar]
- 6.Estepa, A., D. Frias, and J. M. Coll. 1992. Susceptibility of trout kidney macrophages to viral hemorrhagic septicemia virus. Viral Immunol. 5283-292. [DOI] [PubMed] [Google Scholar]
- 7.Evensen, Ø., W. Meier, T. Wahli, N. J. Olesen, P. E. V. Jørgensen, and T. Håstein. 1994. Comparison of immunohistochemistry and virus cultivation for identification of viral haemorrhagic septicaemia virus in experimentally infected rainbow trout (Oncorhynchus mykiss). Dis. Aquat. Organ. 20101-109. [Google Scholar]
- 8.Evensen, Ø., and N. J. Olesen. 1997. Immunohistochemical detection of VHS virus in paraffin-embedded specimens of rainbow trout (Oncorhynchus mykiss): the influence of primary antibody, fixative and antigen unmasking on method sensitivity. Vet. Pathol. 34253-261. [DOI] [PubMed] [Google Scholar]
- 9.Harmache, A., M. LeBerre, S. Droineau, M. Giovannini, and M. Brémont. 2006. Bioluminescence imaging of live infected salmonids reveals that the fin bases are the major portal of entry for Novirhabdovirus. J. Virol. 803655-3659. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Isshik, T., T. Nishizawa, T. Kobayashi, T. Nagano, and T. Miyazaki. 2001. An outbreak of VHSV (viral haemorrhagic septicaemia virus) infection in farmed Japanese flounder Paralichthys olivaceus in Japan. Dis. Aquat. Organ. 4787-99. [DOI] [PubMed] [Google Scholar]
- 11.Kocan, R., M. Bradley, N. Elder, T. R. Meyers, W. N. Batts, and J. R. Winton. 1997. North American strain of viral hemorrhagic septicemia virus is highly pathogenic for laboratory-reared Pacific herring. J. Aquat. Anim. Health 9279-290. [Google Scholar]
- 12.Lorenzen, N., N. J. Olesen, and P. E. V. Jørgensen. 1988. Production and characterization of monoclonal antibodies to four Egtved virus structural proteins. Dis. Aquat. Organ. 435-42. [Google Scholar]
- 13.Lyngstad, T. M., H. R. Høgåsen, I. Ørpetveit, H. Hellberg, O. B. Dale, and A. Lillehaug. 17 March 2008, posting date. National Veterinary Institute's report series, no. 3. Scientific evaluation of the eradication of viral hemorrhagic septicemia in Storfjorden. National Veterinary Institute, Oslo, Norway. [In Norwegian.] http://www.vetinst.no/nor/Forskning/Rapporter/Rapportserie/Rapportserie-2008/3-2008-Bekjempelse-av-VHS-i-Storfjorden-endelig-versjon.
- 14.Meier, W., and K. Pfister. 1981. Viral hemorrhagic septicemia (VHS virus) in pike (Esox lucius L.). Clinical, macroscopic, histological and electron-microscopical findings; direct visualisation of Egtved virus. Schweiz. Arch. Tierheilkd. 12337-49. [PubMed] [Google Scholar]
- 15.Meyers, T. R., and J. R. Winton. 1995. Viral hemorrhagic septicemia virus in North America. Annu. Rev. Fish Dis. 53-24. [Google Scholar]
- 16.Mortensen, H. F., O. E. Heuer, N. Lorenzen, L. Otte, and N. J. Olesen. 1999. Isolation of viral haemorrhagic septicaemia virus (VHSV) from wild marine fish species in the Baltic Sea, Kattegat, Skagerrak and the North Sea. Vir. Res. 6395-106. [DOI] [PubMed] [Google Scholar]
- 17.Munro, A. S. 1996. Scottish aquaculture research report, no. 3. Report on the first recorded outbreak of viral haemorrhagic septicaemia (VHS) in Great Britain and subsequent actions to contain, eradicate and investigate the origins of the infection. Environment and Fisheries Department, Scottish Office of Agriculture, Aberdeen, United Kingdom.
- 18.Neukirch, M., and B. Glass. 1984. Some aspects of virus shedding by rainbow trout (Salmo gairdneri Rich.) after waterborne infection with viral haemorrhagic septicaemia (VHS) virus. Zentralbl. Bakteriol. Mikrobiol. Hyg. A 257433-438. [PubMed] [Google Scholar]
- 19.Olesen, N. J., and P. E. V. Jørgensen. 1991. Rapid detection of viral haemorrhagic septicaemia virus in fish by ELISA. J. Appl. Ichthyol. 7183-186. [Google Scholar]
- 20.Pärt, P., L. Norrgran, E. Bergström, and P. Sjöberg. 1993. Primary cultures of epithelial cells from rainbow trout gills. J. Exp. Biol. 175219-232. [Google Scholar]
- 21.Quillet, E., M. Dorson, G. Aubard, and C. Torhy. 2001. In vitro viral haemorrhagic septicaemia virus replication in excised fins of rainbow trout: correlation with resistance to waterborne challenge and genetic variation. Dis. Aquat. Organ. 45171-182. [DOI] [PubMed] [Google Scholar]
- 22.Ross, K., U. McCarthy, P. Huntly, B. Wood, D. Stuart, E. Rough, D. Smail, and D. Bruno. 1994. An outbreak of viral haemorrhagic septicaemia (VHS) in turbot (Scophthalmus maximus L.) in Scotland. Bull. Eur. Assoc. Fish Pathol. 14213-214. [Google Scholar]
- 23.Schlotfeldt, H., W. Ahne, P. Vestergård Jørgensen, and W. Glende. 1991. Occurrence of viral haemorrhagic septicaemia in turbot (Schopthalmus maximus L.): a natural outbreak. Bull. Eur. Assoc. Fish Pathol. 11105-107. [Google Scholar]
- 24.Skall, H. F., W. J. Slierendrecht, J. A. King, and N. J. Olesen. 2004. Experimental infection of rainbow trout Oncorhynchus mykiss with viral haemorrhagic septicaemia virus isolates from European marine and farmed fishes. Dis. Aquat. Organ. 5899-110. [DOI] [PubMed] [Google Scholar]
- 25.Skall, H. F., N. J. Olesen, and S. Mellergaard. 2005. Viral haemorrhagic septicaemia virus in marine fish and its implications for fish farming. J. Fish Dis. 28509-529. [DOI] [PubMed] [Google Scholar]
- 26.Snow, M., C. O. Cunningham, W. T. Melvin, and G. Kurath. 1999. Analysis of the nucleoprotein gene identifies distinct lineages of viral haemorrhagic septicaemia virus (VHSV) within the European marine environment. Vir. Res. 6335-44. [DOI] [PubMed] [Google Scholar]
- 27.Snow, M., N. Bain, J. Black, V. Taupin, C. O. Cunningham, J. A. King, H. F. Skall, and R. S. Raynard. 2004. Genetic population structure of marine viral haemorrhagic septicaemia virus (VHSV). Dis. Aquat. Organ. 6111-21. [DOI] [PubMed] [Google Scholar]
- 28.Tafalla, C., B. Figueras, and B. Novoa. 1998. In vitro interaction of viral haemorrhagic septicaemia virus and leukocytes from trout (Oncorhynchus mykiss) and turbot (Scophthalmus maximus). Vet. Immunol. Immunopathol. 62359-366. [DOI] [PubMed] [Google Scholar]
- 29.Wolf, K., M. Gravell, and R. G. Malsberger. 1966. Lymphocystis virus: isolation and propagation in centrarchid fish cell lines. Science 1511004-1005. [DOI] [PubMed] [Google Scholar]
- 30.Wood, C. M., and P. Pärt. 2000. Intracellular pH regulation and buffer capacity in CO2/HCO3-buffered media in cultured epithelial cells from rainbow trout gills. J. Comp. Physiol. B 170175-184. [DOI] [PubMed] [Google Scholar]
- 31.Yamamoto, T., W. N. Batts, and J. R. Winton. 1992. In vitro infection of salmonid epidermal tissues by infectious hematopoietic necrosis virus and viral hemorrhagic septicemia virus. J. Aquat. Anim. Health 4231-239. [Google Scholar]