Piscine orthoreovirus (PRV) replicates in Atlantic salmon (Salmo salar L.) erythrocytes ex vivo

Blood samples

Heparinized blood (0.5 mL) was collected from the caudal vein of six naïve Atlantic
salmon (F1-F6). The fish had an average weight of 40 g and originated from a naïve
population kept in a fresh water tank at VESO aquatic research facility (Vikan, Norway).
The fish were anesthetized prior to sampling by bath immersion (2–5 min) in benzocaine
chloride (0.5 g/10 L water) and euthanized after sampling using benzocaine chloride
(1 g/5 L water) for 5 min. The RBC were isolated from the blood samples as described
below and used as recipient erythrocytes in the first passage. Three weeks later heparinized
blood samples were collected from six fish (F11-F16) of the same fish population.
RBC were isolated and used as recipient erythrocytes in the second passage.

Isolation of naïve RBC

We diluted the heparinized blood 1:15 in phosphate-buffered saline (PBS) and isolated
the RBC using a Percoll gradient as previously described 12]. The cells were counted using Countess (Invitrogen, Eugene, Oregon, USA) and resuspended
to a concentration of 3?×?10⁷ cells/mL in Leibovitz’s L15 medium (Life Technologies, Carlsbad, CA, USA) supplemented
with fetal calf serum (2%) and gentamicin (50 ?g/mL). All recipient RBCs (first passage
F1-F6; second passage F11-F16) were confirmed PRV-negative by RT-qPCR and flow cytometry
before initiation of the study (described below).

Preparation of inoculums

The PRV inoculum was prepared from a batch of pooled heparinized blood samples collected
6 weeks post challenge (wpc) from cohabitant fish of a PRV challenge experiment (VESO,
Vikan, Norway). We centrifuged the pooled blood sample at 2000?g for 10 min at 4 °C, removed the plasma and diluted the blood pellet 1:10 in L15 with
gentamicin (50 ?g/mL). The diluted blood was then sonicated on ice by 8?×?10 s pulses
at 25 Hz with 30 s rest in between, and the lysed blood was centrifuged at 2000?g for 5 min at 4 °C to remove cellular debris. The resulting supernatant was used as
PRV inoculum in the first passage. Similarly, a negative control inoculum was prepared
as described above, using heparinized blood collected from naïve fish. The control
inoculum was confirmed PRV negative by RT-qPCR. The PRV inoculum used had a Ct-value
of 22.3.

Ex vivo setup; 1^st passage

In the first PRV passage, RBC isolated from six naïve fish were inoculated and cultured
ex vivo for 21 days. The culture experiment was performed in 24-deepwell microtiter
plates with pyramidal bottom (AB Ninolab, Stockholm, Sweden) to allow easy centrifugation
and washing of the cells. The plates were closed using Mikro-Flask sandwich cover
and clamp system by Duetz (Applikon, Foster City, CA, USA) to minimize evaporation
and prevent well to well contamination. Incubation was performed at 15 °C in an Ecotron
incubation shaker (Infors HT, Basel Switzerland) shaken at 225 rpm to ensure that
the RBC was kept in a homogenous suspension.

The cultivation setup included four identical PRV infected plates and four control
plates that were sequentially harvested at 1, 7, 14 and 21 days post infection (dpi).
On each plate, RBC isolated from fish F1-F6 was plated at 3?×?10⁷ RBC per well in 1 mL medium. The wells were subsequently inoculated with 100 ?L of
either the PRV-infected inoculate or the control inoculate described above. Following
24 h of incubation the plates were centrifuged at 500?g for 5 min at 4 °C to remove the supernatant and washed in Leibovitz’s L15 medium
(Life Technologies). Finally, the RBC was resuspended to a final concentration of
10?×?10⁶ RBC/mL in 3 mL Leibovitz’s L15 medium (Life Technologies) supplemented with fetal
calf serum (2%) and gentamicin (50 ?g/mL). The 1 dpi samples were sampled immediately
after resuspension and the remaining plates were incubated until harvested at 7, 14
and 21 dpi respectively. The 0 dpi samples were collected prior to inoculation.

At each time point a sample of 2 mL (20?×?10⁶ RBC) was collected from each well for RNA isolation and subsequent RT-qPCR analysis
to assess the viral load in the RBC and the supernatant. In addition, the immune response
of infected RBC was assayed. The remaining 1 mL (10?×?10⁶ RBC) was used for PRV detection by flow cytometry analysis and immunofluorescence/confocal
microscopy to study viral inclusions. The number of cells were counted using Countess
(Invitrogen) in order to determine the amount of cell lysis. Details for all analyses
are described below.

Ex vivo setup; 2^nd passage

In the second PRV passage, RBC isolated from six fish were infected with supernatant
from the first passage and cultured ex vivo for 21 days. A pool of supernatant was
prepared from the 21 dpi samples of the first passage after centrifugation at 1000?g for 5 min at 4 °C and used as inoculum.

The cultivation setup included two identical PRV infected plates that were sequentially
harvested at 14 and 21 dpi. On each plate, RBC isolated from fish F11-F16 was plated
at 3?×?10⁷ RBC per well in 1 mL media as for the first passage. The cells were subsequently
inoculated with 100 ?L of a 10-fold dilution of the supernatant from the first passage.
The RBC were incubated for 24 h, centrifuged, washed and resuspended in growth medium
to a final concentration of 10?×?10⁶ RBC/mL. The plates were cultured under identical conditions as those of the first
passage and harvested at 14 and 21 dpi. The 0 dpi samples had been collected prior
to inoculation.

At each time point a sample of 2 mL (20?×?10⁶ RBC) was collected from each well for RNA isolation and RT-qPCR analysis to detect
PRV in the RBC fraction and 1 mL was used for flow cytometry analysis and immunofluorescence
microscopy. Details for the analyses are described below.

RNA isolation

The 2?×?10⁶ cells (2 mL) harvested from each culture were pelleted at 1000?g for 5 min at 4 °C and total RNA was isolated from the RBC pellet and the supernatant.
The RBC pellet was homogenized in Isol-RNA Lysis Reagent (5 PRIME, Hilden, Germany)
using 5 mm steel beads and TissueLyser II (Qiagen) for 2?×?5 min at 25 Hz. After addition
of chloroform and centrifugation, the aqueous phase was collected and proceeded with
automated RNA isolation using RNeasy Mini QIAcube Kit (Qiagen) as described by the
manufacturer. For the supernatant samples, a combination of Trizol LS (Invitrogen)
and RNeasy Mini spin column (Qiagen) was used. Briefly, 100 ?L supernatant was mixed
and incubated with Trizol LS before adding chloroform, separating the phases by centrifugation,
and then proceeding with the RNeasy Mini spin column as recommended by the manufacturer,
eluting isolated RNA in 50 ?L RNase-free water. RNA was quantified using a NanoDrop
ND-1000 spectrophotometer (Thermo Fisher Scientific, Wilmington, DE, USA).

RT-qPCR; PRV detection

The Qiagen OneStep kit (Qiagen) was used for RT-qPCR. The input was 100 ng (5 ?L of
20 ng/uL) total RNA from RBC per reaction and 5 ?L purified RNA solution from the
cell free samples, i.e. supernatants and inoculums. RT-qPCR targeted PRV segment S1
using the following conditions: 400 nM primer, 300 nM probe, 400 nM dNTPs, 1.26 mM
MgCl2, 1:100 RNase Out (Invitrogen) and 1?×?ROX reference dye with the following cycle
parameters: 30 min at 50 °C, 15 min at 94 °C, 35 cycles of 94 °C/15 s, 54 °C/30 s
and 72 °C/15 s in a Mx3005P (Stratagene, La Jolla, CA, USA). The samples were run
in duplicate, and a sample was defined as positive if both parallel samples had a
Ct??35. Primers and probe are listed in Table 1.

Table 1. Primers and probe used in this study

qPCR; Antiviral response

Reverse transcription was performed using 200 ng RNA per sample, and a mixed input
from representative samples was used to prepare a seven point concentration standard
curve. The RNA was denaturated at 95 °C for 5 min prior to cDNA synthesis, using the
Quantitect reverse transcription kit with integrated genomic DNA removal (Qiagen).
Quantitative PCR (qPCR) was run in duplicate on cDNA corresponding to 10 ng RNA input.
The antiviral response gene expression analysis was performed using 2?×?Maxima SYBR
Green qPCR Master Mix (Thermo Fisher scientific, Waltham, MA, USA) and 500 nM specific
primers targeting Atlantic salmon IFN?, myxovirus resistance (Mx), RIG-I and PKR respectively.
The assays were run 40 cycles of 94 °C/15 s, 60 °C/30 s in a Mx3005P device (Stratagene).
Primers used are listed in Table 1.

Flow cytometry

RBC sampled during the first and second passage were screened for PRV by flow cytometry.
The RBC were stained intracellularly using a rabbit polyclonal antibody against PRV
putative outer capsid protein ?1 (Anti-?1, #K275) as previously described 12]. The corresponding zero serum (Anti-?1 Zero #K275) 14] was used as negative control serum. The cells were analyzed on a Gallios Flow Cytometer
(Beckman Coulter, Miami, FL, USA) counting 30 000 cells per sample. The data were
analyzed using the Kaluza software (Becton Dickinson). The forward/side gate was set
to adjust for variation in background staining. The arithmetic mean fluorescence intensity
(MFI) was calculated for all PRV infected (green) and control (grey) samples at each
time point, in addition to the ?MFI. The differences were calculated statistically
using Wilcoxon matched pairs signed rank test due to the small sample size (n?=?6).

Immunofluorescence microscopy

The PRV ?1 stained RBC used in flow cytometry were also assayed by immunofluorescence
microscopy. The nuclei were stained with Hoechst 33342 (Invitrogen) before they were
transferred to Countess chamber slides (Invitrogen). Photographs were taken by an
inverted fluorescence microscope (Olympus IX81), at 20× and 40× magnification.

Confocal microscopy

Smears of RBC were dried before fixation for 5 min in ice cold methanol. Slides were
rehydrated in PBS and blocked in PBS with 5% skimmed milk before sequentially staining
with a mouse monoclonal antibody against dsRNA (J2; 1:200) (Scicons, Hungary) and
rabbit anti-?1 (1:750) with washing in-between. Secondary antibody; goat anti-mouse
Alexa Fluor 488 and goat anti-rabbit Alexa Fluor 594 (1:1000) were obtained from Molecular
Probes (Invitrogen). The nucleus was counterstained with Hoechst 33342 (Invitrogen)
and coverslips were mounted with Fluoroshield (Sigma-Aldrich). The images were generated
on an inverted Zeiss LSM 710 confocal laser scanning microscope (CLSM) (Carl Zeiss,
Oberkochen, Germany), using lasers of 405 nm, 488 nm and 594 nm to excite the respective
stains. In post-production, the original color of dsRNA (Alexa Flour 488) and PRV-?1
protein (Alexa Flour 594) was switched to match the immunofluorescence pictures, in
which PRV-?1 protein was colored green.

Data analysis

The PRV RT-qPCR results from 1 dpi were compared to paired samples collected at 7,
14 and 21 dpi in the first. For the immune gene analysis of IFN?, Mx, RIG-I and PKR
the fold increase at 1, 7, 14 and 21 dpi was calculated, assigning the qPCR results
from the uninfected paired samples as 1. In the second passage, the results from the
undiluted and the 10 fold diluted inoculum was compared at each time point. The differences
were analyzed statistically using Wilcoxon matched pairs signed rank test due to the
small sample size (n?=?6). All statistical analysis described were performed with GraphPad Prism (GraphPad
Software inc., USA) and p-values of p???0.05 were considered as significant.