Susceptibility of Pimephales promelas and Carassius auratus to a strain of koi herpesvirus isolated from wild Cyprinus carpio in North America

Collection of wild carp from a CyHV-3-exposed population

This study was carried out in accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health. All protocols for sampling, procedures and experimental endpoints involving live fish conducted in this study were approved by the Institutional Animal Care & Use Committee (IACUC), University of Minnesota (St. Paul, Minnesota, USA), under the approval numbers IACUC-1806-36036A and 1808-36276A. Experiments were performed in compliance with the ARRIVE guidelines on animal research³².

Wild carp were sampled from Lake Elysian (Waseca County, Minnesota, Coordinates: 44.178144, − 93.69066) by boat electrofishing from September 3 to 9, 2019 (Fig. 1a). This lake was expected to have a CyHV-3-exposed carp population following a confirmed outbreak in 2017³. Captured wild adult carp (n = 116) were euthanized by immersion in a solution of ~ 3 mL/L pure clove oil (90% Eugenol; Velona, Elk Grove Village, IL, USA) for 15 min and transported on ice to the University of Minnesota for necropsy. Brain, gill and kidney tissues from up to three carp were pooled in a 1:5 (weight:volume) dilution of Hank’s Balanced Salt Solution (HBSS; Cellgro, Lincoln, NE, USA) containing 100 IU/mL of Penicillin and Streptomycin and maintained at a pH of 7.4 at 4 °C for 24 h prior to preparation for qPCR and cell culture screening for CyHV-3 (described below). Gill tissues from ten freshly-dead carp obtained from a shallow bay in the Southern portion of the lake were also obtained and pooled by five individuals for a total of two sample pools.

An additional 17 wild carp collected as part of the previously described sampling event were placed in an aerated live well and transferred to the Minnesota Aquatic Invasive Species Research Center’s Containment Laboratory (MCL). These carp were housed in a ~ 1400 L tank with flow through well water at 20 °C and treated with 0.6% aquarium salt once per day. Carp were acclimated for 1 day and then anesthetized via immersion in a solution of 100 µL/L of clove oil and uniquely marked using colored injectable elastomer (Northwest Marine Technology, Anacortes, WA, USA). Additionally, a small portion (~ 0.2 cm²) of each carp’s gills were sampled for qPCR screening for CyHV-3 and tested immediately. Carp determined to be CyHV-3 negative (n = 12) were euthanized following testing. Carp determined to be CyHV-3-positive (n = 5) by specific qPCR were held for a total of 5 days, during which, water temperature was gradually increased to 26 °C in order to increase viral shedding. CyHV-3-positive carp gill biopsies were again sampled and screened on the fifth day to identify carp with high qPCR copy numbers. All CyHV-3-positive carp were then euthanized, and the brain, gill and kidney tissues were removed as previously described. Pooled tissues from two wild carp with clinical signs consistent with KHVD (Fig. 1b,c) and with high qPCR copy numbers, were subjected to cell culture immediately following necropsy. In addition, a 10 g portion of this pooled tissue was processed and used to challenge naive carp in the in-vivo infection model. Tissues were homogenized in a 1:5 volume of HBSS containing 100 IU/mL Penicillin and Streptomycin (pH = 7.4). The sample was centrifuged at 2360 × g at 25 °C for 10 min, then the supernatant was passed through a 0.45 µm syringe filter.

In-vivo infection trial

To increase the potential of obtaining an isolate of CyHV-3, naïve carp previously determined to be CyHV-3 negative by qPCR, were challenged with CyHV-3-positive tissue homogenates obtained from wild carp. Two naïve carp, purchased from Osage Catfisheries (Osage Beach, MO, USA), were pair housed in a 60 L aquarium with flow through well water (flow rate = 3–4 tank volumes/h) at 21–22 °C. Aquaria were set up with a standpipe drain covered by a cylindrical wire screen filter of approximately 15 cm in length and 4.4 cm in diameter. Additionally, a PVC pipe section of 15 cm in length and 10 cm in diameter was added to each tank for shelter. Each carp was exposed to 0.5 mL of CyHV-3-positive tissue homogenate by IP-injection and monitored for signs of disease for 6 days and then euthanized. Pooled samples of brain, gill and kidney tissue were subjected to qPCR and cell culture analysis. Following cell culture analysis (below) a second infection trial was performed to satisfy River’s postulates (i.e. that CyHV-3 isolated from wild diseased carp would cause similar disease in naïve carp)³³. Two additional naïve carp purchased from Osage Catfisheries were IP-injected with 0.5 mL of CyHV-3-positive (qPCR and cell culture positive) cell culture supernatant. Carp were housed and observed for disease signs as previously described for 11 days and then sacrificed. Pooled samples of brain, gill, and kidney then were tested by CyHV-3-specific qPCR to confirm the presence of CyHV-3.

Cell culture analysis

CCB cells were maintained in Eagle’s Minimum Essential Medium (EMEM) containing Eagles’s salts (Sigma, St. Louis, MO, USA), 10% fetal bovine serum (FBS), 1% non-essential amino acids (NEAA, Sigma), 2 mM l-glutamine and glucose (Sigma) up to 4.5 g/L. The KF-1 cells were cultured in EMEM containing Eagles’s salts (Sigma), 10% FBS and 0.025 M HEPES. Penicillin 100 U/L and streptomycin 0.1 mg/L (Sigma) were used as an anti-bacterial agent in both cell culture media and the cells were maintained at 25 °C.

Cell culture methods to isolate CyHV-3 were performed according to the US Fish and Wildlife Service and American Fisheries Society-Fish Health Section Blue Book³⁴. Briefly, pooled tissues were homogenized in Hank’s Balanced Salt Solution (HBSS; Cellgro) and centrifuged at 2360 × g for 15 min. The inoculum was added to the 24-well plates with 80% confluent cell cultures in two dilutions, (1/10 and 1/100) and incubated at 25 °C for 14 days. A blind passage was performed for an additional 14 days if no cytopathic effects (CPE) were observed on the first passage. If CPE was observed during the first passage, then the second passage was performed in a 25 cm² flask. The virus was harvested when CPE reached 70–80% of the monolayer. The infected cultures were exposed to two freeze/thaw cycles at − 80 °C, and then centrifuged at 3800 × g for 15 min at 4 °C. The clarified supernatants and pellets were collected and stored at − 80 °C.

Whole-genome sequencing and sequence analysis

Whole-genome sequencing was performed at the University of Minnesota Veterinary Diagnostic Laboratory for genetic characterization of the CyHV-3 isolate (KHV/Elysian/2019) obtained from wild carp. In brief, after CCB cells, infected with wild carp tissues, reached 80% CPE, the supernatant was collected and stored at − 80 °C. The frozen supernatant was freeze-thawed three times, and centrifuged at 2896 × g for 25 min at 4 °C. Nucleic acid purification of CCB cell culture supernatant was done using a QIAamp MinElute Virus Spin Kit (Qiagen, Hilden, Germany) following manufacturer instructions. The extracted nucleic acids were subjected to library preparation using Nextera Flex DNA library kit (Illumina, San Diego, CA, USA) following manufacturer instructions. The library was normalized according to the median fragment size measured by Tape Station 2.0 (Agilent, Santa Clara, CA, USA) and library concentration measured by Qubit. The library was submitted to the University of Minnesota Genomic Center (UMGC) for sequencing via MiSeq V3 (2X75-bp) paired end chemistry.

Raw fastq files were trimmed to remove Illumina adapters using Trimmomatic (v 0.39) with a minimum quality score of 20. Then, bowtie2 (v 2.3.5) was used to remove host contamination and unmapped reads were used for assembly with SPAdes (v3.13.0) with k-mer values of 29, 33 and 55 with the options “careful with a minimum coverage of 5 reads per contig”. Then contigs were searched into the RefSeq viral and non-redundant protein reference databases using Diamond BLASTx with an e-value of 1e − 5 for significant hits. Taxon assignments were made with MEGAN6 Community Edition according to the lowest-common-ancestor algorithm. ORFs prediction and genome annotation were done using Prokka (v1.14.5). The resulting alignment (GenBank accession no. MT914509) was aligned with 19 other CyHV-3 genomes available on NCBI using Mafft (v7) with the FFT-NS-2 alignment strategy and the following parameters: scoring matrix BLOUSUM62, gap open penalty 1.53, offset value 0. Model selection, maximum likelihood (ML) tree construction, and calculation of bootstrap values were done with R 4.0 (R Software) using phangorn (v2.5.5). ML trees were constructed using the top scoring model (GTR + G + I) and 100 bootstrap replicates were generated to assess the reliability of clades obtained in the tree. Additionally, this genome assembly was compared with the previously reported thymidine kinase gene sequence obtained from carp sampled during a large mortality event in Lake Elysian in 2017 (F36, GenBank accession no. MK987089).

Investigation of species specificity

To investigate the host range of KHV/Elysian/2019, six carp purchased from Osage Catfisheries, previously determined to be CyHV-3-negative by qPCR, were intraperitoneally (IP) injected with 0.5 mL of the filtered tissue homogenate material (Fig. 2a). The IP-injected carp (IP-carp) were housed as previously described for 9 days prior to their use in the cohabitation trial (Fig. 2b). The IP-carp were monitored twice daily for signs of disease. After 9 days the gills, skin and vent of each IP-carp was swabbed aseptically with a single sterile cotton swab (Dynarex, Orangeburg, NY, USA) for determination of viral load by qPCR. FHM and goldfish were challenged with CyHV-3 via cohabitation. One cohabitation tank (tank A) contained ten naïve FHM, five naïve sentinel carp (S-carp) and three IP-carp (Fig. 2a). One cohabitation tank (tank B) contained ten naïve goldfish, five naive S-carp and three of the IP-carp. S-carp were included in each tank setup to act as a positive control for within-tank transmission of CyHV-3. Two additional negative control tanks with the same stocking density and conditions contained ten naïve FHM (tank D) and ten naïve goldfish (tank E), as well as eight naïve carp (confirmed to be CyHV-3-negative by specific qPCR). Average standard length and weight for fishes used in these experiments was 13 cm and 64 g for carp, 7 cm and 13 g for FHM, and 10 cm and 38 g for goldfish. All tanks consisted of ~ 60 L aquaria with flow-through well water as previously described. Fishes were fed a commercial feed (Skretting classic trout, Skretting, Tooele, UT, USA) daily and monitored twice daily to observe changes to fish health. IP-carp that died during the trial were allowed to remain in the tank for 24 h prior to removal for necropsy, but any morbidity or mortality of other experimental groups were immediately removed and necropsied.

At 0, 3, 6, 9, 12, and 15 days post exposure (dpe) by cohabitation, five FHM, five goldfish, and all IP-carp and S-carp from each tank were anesthetized by immersion in a buffered solution of 100 mg/L of MS-222 and the gills, skin and vent of each fish was swabbed with a sterile swab for determination of viral load by qPCR (Fig. 2c). For FHM and goldfish, the five individuals were randomly sampled at each time-point. Additionally, the wire screen filter of the outflow standpipe was swabbed at the same intervals during the course of the trial to evaluate loading of CyHV-3 DNA in the environment. All swabs were stored at − 20 °C in individual plastic bags until nucleic acid extraction could be performed. At 11 dpe, half of the FHM and goldfish from cohabitation tanks were euthanized by immersion in a buffered solution of 3 g/L of MS-222 and necropsied (Fig. 2c). The remaining FHM and goldfish were maintained until 20 dpe and then euthanized and necropsied. To visually record the presence of gross pathology, representative IP carp, and fish from cohabitation groups (S-carp, FHM, and goldfish) were randomly selected and photographed at 0 and 6 dpe in a small glass aquarium (Fig. 3).

For each necropsied fish, wet mounts of gill and skin scrapes were viewed at 40× magnification to identify potential parasitic infections. Then the skin of each necropsied fish was rinsed briefly with 70% ETOH and clean water. Brain, gill, kidney and skin tissue were collected individually for each fish and split into two duplicate samples. The first sample duplicates were placed in Whirl–Pak sample bags (Nasco, Fort Atkinson, WI, USA) and preserved at − 20 °C until nucleic acid extraction and screening for CyHV-3 DNA was performed. The second sample duplicates were placed in 1 mL of RNAlater solution (Ambion) in 1.5 mL microcentrifuge tubes (Globe Scientific, Mahwah, NJ, USA) and frozen at − 20 °C. An individual FHM and goldfish from each time-point (11- and 20-dpe) was preserved in 10% NBF (TissuePro, Gainesville, FL, USA) for histological analysis. Individual representatives of each species from control tanks and moribund S-carps from each experimental tank were also preserved for histological analysis.

Due to the detection of CyHV-3 DNA in a single FHM in tank A, a second trial with FHM (tank C) was performed as described previously (Fig. 2a). Brain, gill, kidney, and skin tissue from two S-carp exposed in the first trial with disease signs and positive qPCR test for CyHV-3 (tank A) were pooled, homogenized and filtered as previously described. Three new carp purchased from Osage Catfisheries were IP injected with 0.5 mL of this tissue homogenate and maintained as previously described for 9 days prior to screening for CyHV-3 by qPCR and used in the cohabitation trial. All other conditions and procedures were done as described for the first cohabitation trial with the following exceptions. In the second trial, portions of brain, gill, kidney and skin tissues obtained from a moribund S-carp at 5 dpe and four FHM at 11 dpe, respectively, were pooled as previously described and subjected to cell culture. Additionally, duplicate swabs from the tank C outflow standpipe filter were obtained and preserved in 1 mL of RNAlater solution (Sigma) as previously described for tissue samples.

Nucleic acid purification using chelex resin and detection of CyHV-3 by qPCR

For nucleic acid purification, chelex resin (Sigma) was used as described by Zida et al.³⁵ and briefly summarized here. For pooled tissue samples, approximately 100 mg of each tissue was homogenized in 1 mL of nuclease free water (NFW) and then centrifuged, with 50 μL of the resulting supernatant later used as starting material. For swabs, the cotton end was cut off and vortexed, then centrifuged and finally the cotton was removed leaving the supernatant. For each sample type, 150 μL of chilled 80% ETOH was added, then centrifuged and the supernatant removed. Samples were allowed to air dry for 10 min to remove residual ETOH. 150 μL of 20% Chelex was added to each sample and vortexed. Samples were then incubated at 90 °C for 20 min and centrifuged and immediately used for qPCR.

A Taqman probe-based qPCR was used for the detection of CyHV-3 DNA targeting the ORF89 gene³⁶ using a StepOnePlus thermocycler with default settings (Applied Biosystems). Nucleic acid purifications from all samples were screened for CyHV-3 using a PrimeTime gene expression master mix kit (Integrated DNA Technologies, Coralville, IA, USA), with each reaction containing 400 nM of primers (KHV-86f: GAC-GCC-GGA-GAC-CTT-GTG, KHV-163r: CGG-GTT-GTT-ATT-TTT-GTC-CTT-GTT) and 250 nM of the probe (KHV-109p: [TAMRA] CTT-CCT-CTG-CTC-GGC-GAG-CAC-G-[IBRQ]. The reaction mix was subjected to an initial denaturation at 95 °C for 3 min, followed by 40 cycles of denaturation at 95 °C for five sec and annealing at 60 °C for 30 s. A threshold cycle of 38 was used as a cut off. The standard curve for quantification of CyHV-3 genomes was performed using a laboratory synthesized DNA fragment containing the ORF89 sequence as previously described by Padhi et al.³. The results for virus load are presented as the number of viral copies per mL of tissue supernatant. All samples obtained from FHM and goldfish were tested in triplicate with the exception of samples that had positive qPCR Ct values, which were re-tested up to six times.

RNA purification and reverse transcription polymerase chain reaction (RT-PCR)

Individual tissues of preserved brain, gill, kidney, and skin from one representative S-carp from each experimental tank (A, B and C) were selected as positive controls for CyHV-3 mRNA detection (total of 12 tissue samples). All preserved tissue samples from FHM or goldfish which had at least one positive qPCR test were also screened for CyHV-3 mRNA to determine if replicating virus was present (total of eight tissue samples). Additionally, preserved swabs of the outflow standpipe filter were also screened. For RNA purification, RNA was extracted from tissues using the RNeasy Mini Kit (Qiagen) according to the manufacturer instructions for animal tissues, using ~ 30 mg tissue samples preserved in RNAlater. For swabs, cotton was cut from the end of the swab and used as the starting material. CyHV-3 mRNA was detected using the RT-PCR developed by Yuasa et al.²⁹ with the primers, (KHV RT F3: GCC-ATC-GAC-ATC-ATG-GTG-CA, KHV RT R1: AAT-GCC-GCT-GGA-AGC-CAG-GT). The RT-PCR was performed using a One-step RT-PCR kit (Qiagen) according to the manufacturer instructions. The reaction mix was subjected to a single step of reverse transcription at 50 °C for 30 min and denaturation at 95 °C for 15 min, followed by 40 cycles of: 94 °C for 30 s, 65 °C for 30 s, 72 °C for one minute and a final extension step was 72 °C for 10 min. PCR products were separated and visualized on 2% agarose gels containing 0.75 μg/mL ethidium bromide (Genesee Scientific, San Diego, CA, USA). PCR products for carp, FHM and goldfish templates (clear band at the 219 bp location) were cut from gels and purified by precipitation with a 20% PEG, 2.5 M NaCl solution. Purified RT-PCR products were subjected to Sanger sequencing at the University of Minnesota Genomics Center (UMGC). Sequences were trimmed and analyzed using 4 peaks (v1.8) and consensus sequences were generated using BioEdit (v7.2.1). Sequence identities were compared with available reference sequences by BLASTn analysis of the National Center of Biotechnology sequence database.

Histology

Histology was used to demonstrate the presence or absence of lesions in cohabitation disease trial specimens. Histological samples of gill tissue were prepared from formalin-fixed samples of representative fishes of each species from trial and control tanks. Gill samples were dissected from formalin-fixed specimens and decalcified in 10% ethylenediaminetetraacetic acid (EDTA) for 10 days. Following decalcification, samples were dehydrated in an ethanol series to 100% ethanol, infiltrated with toluene, and subsequently embedded in paraffin. Paraffin sections were cut at 6 µm thickness using a Leica Jung 820 Histocut Rotary Microtome and mounted on slides. Sections were stained with Hematoxylin and Eosin using a protocol modified from Humasson³⁷.

Statistical analysis

R 4.0 (R Software) was used for statistical analysis and data presentation. CyHV-3 qPCR copy numbers are presented as averages of all positive tests for samples with duplicate tests and were Log transformed prior to statistical testing. Significant differences (p < 0.05) in virus load of IP injected carp and cohabitated fish were determined using a 1-way ANOVA with subsequent pairwise multiple comparisons using the Holm-Sidak method and data were presented as box plots of 25–75% (+ minimum and maximum values) with an indication of mean and median. Bivariate associations were measured with odds ratios and 95% confidence intervals (Cis).

Source: Ecology - nature.com