Naïve common carp fry of Amur wild carp (Amur sazan, AS) and koi strains were obtained from the University of South Bohemia in Ceske Budejovice, Faculty of Fisheries and Protection of Waters, located in Vodnany, Czech Republic. Fish were raised in a recirculation system filled with tap water at 20°C and fed a commercial feed (Perla Plus, Skretting Norway) at a rate of 1% body weight per day. Prior to their use in infection experiments, fish had been confirmed to be free of ectoparasites by means of fresh skin and gill surface smears and examination using light microscopy. They were also confirmed to be free of specific viruses by means of RT-qPCR assays as described earlier (5, 31). These assays included the following viruses that infect common carp: CEV, cyprinid herpesvirus 3 (CyHV-3), spring viremia of carp virus (SVCV), and common carp paramyxovirus (CCPV).

All experiments were carried out in accordance with national and international regulations for experimentation with animals, under approval of the 2nd Local Ethics Committee in Krakow, Poland (no. 355/2021).

Amur sazan and koi carp (approximately 150 g) were kept at the same tank at 20°C for 3 weeks. Prolonged restraint stress (24 h) was induced by immobilizing fish in nets as described previously (32). Briefly, fish (n = 6 from each strain) were stressed by restraining each fish in a net and placing them in the plastic tank (120 L) with aeration at 20°C. Fish restrained in the net were in full contact with the water for 24 h. During the stress challenge, fish were not fed. After 24 h of the restraint stress, fish were immediately transferred all at once to a tank with an overdose of a buffered solution of MS-222 in water (0.4 g/L) and were euthanized prior to blood and tissues collection. Control fish were sampled following rapid netting and euthanized prior to blood and tissues collection immediately before sampling of fish from the stress group.

Amur sazan and koi carp (approximately 150 g) were infected with CEV by cohabitation with clinically affected, virus-shedding donor fish, since propagation of this virus in vitro has not been established yet (12). Infection was performed with CEV from genogroup IIa at two temperatures: 12°C and 18°C. Prior to infection, fish were divided into six groups (n = 10 fish in each group): (i) control koi (Ctr Koi), (ii) CEV-infected koi (CEV Koi), (iii) salt-treated control koi (Ctr Koi NaCl), (iv) salt-treated CEV-infected koi (CEV Koi NaCl), (v) control Amur sazan (Ctr Amur), and (vi) CEV-infected Amur sazan (CEV Amur). Fish from each group were placed in plastic tanks (120 L) with aeration at 20°C. Before infection, the water temperature was lowered from 20°C to 18°C by 1°C per day in the case of experiments performed at 18°C, and from 20°C to 12°C by 1°C per day in the case of experiments performed at 12°C. Fish from each group were acclimatized to the final water temperatures for at least 2 weeks. CEV infection was performed by adding to each tank KSD-affected virus-shedding donor fish (n = 3 fish per infected tank) or healthy, non-infected fish (n = 3 fish per control tank). In the salt-treated koi groups, NaCl (0.6% w/v) was added to the water on day 3 post exposure. At 6 days post exposure, fish from each group were immediately transferred all at once to a tank with an overdose of a buffered solution of MS-222 in water (0.4 g/L) and were euthanized prior to blood and tissues collection.

Blood was collected from the caudal vein using S-Monovette (Sarstedt, Germany). Blood, 100 µl, was used for blood parameter analysis, while the rest of the blood was centrifuged at 2,400 × g for 10 min at 4°C. Plasma was collected and kept at −80°C for further analysis. The brain was removed, and the hypothalamus containing the nucleus preopticus (NPO) was carefully separated from the brain. The pituitary gland (PIT), located at the base of the brain, was collected after removing the brain. Head kidney (HK) and gills were also collected. All tissues were placed into RNAlater and kept at −80°C for further analysis.

Blood, 100 µl, was collected into a glass capillary and loaded into an OPTI CCA-TS blood gas analyzer (OPTI Medical Systems, USA) equipped with OPTI sensor cassettes E-Ca for measuring blood pH, carbon dioxide partial pressure (pCO₂), oxygen partial pressure (pO₂), base excess (BE), total carbon dioxide (tCO₂), blood hydrogen carbonate (HCO₃), standard hydrogen carbonate (CO₂-independent) (stHCO₃), sodium (Na⁺), potassium (K⁺), and calcium (Ca²⁺) levels, as well as the total hemoglobin (tHb) content and hematocrit (Hct). Blood parameters were measured as described previously (14).

Blood plasma was used to assess glucose and cortisol concentration. Glucose concentration was measured using glucose test strips and glucometer iXell® OLED (Genexo, Poland). Cortisol concentration was measured using immunoenzymatic assay according to the manufacturer’s protocol with commercial kit Cortisol ELISA (Neogen, USA).

From each fish, a gill arch was collected immediately after killing and fixed with 4% phosphate-buffered formaldehyde solution for subsequent histological analysis. Sections (3 μm) of paraffin-embedded gills were cut and stained with alcian blue and periodic acid Schiff (AB-PAS). Pathomorphological changes of the gill tissue caused by CEV infection were assessed using the methodology described earlier (33). The following changes were recorded: hyperplasia of epithelial cells, lifting of the epithelium, proliferation of the interlamellar cellular mass, the presence of cells with edema or foamy contents, and the presence of infiltrating granulocytes.

Gill tissues fixed with 4% phosphate-buffered formaldehyde and paraffin embedded were sectioned at 4 µm and placed on adhesion slides Superfrost Plus Gold (Epredia, UK) and stained as described previously (34). Briefly, prepared slides were submitted to deparaffinization, antigen retrieval with sodium citrate buffer (10 mM of sodium citrate, 0.05% Tween 20, pH = 6), and blocking with SuperBlock Blocking Buffer (Thermo Fisher Scientific, USA). Next, slides were incubated with a primary antibody: anti-ZAP70 rabbit monoclonal antibody (Zap-70 (99F2) (35), #2705, Cell Signaling Technology, Inc., USA) at 1:300 in Pierce Immunostain Enhancer (Thermo Fisher Scientific, USA). After washing, slides were incubated with a secondary antibody: Alexa Fluor 555-labeled donkey anti-rabbit antibody (Thermo Fisher Scientific, USA) at 1:1,000 in Pierce Immunostain Enhancer (Thermo Fisher Scientific, USA). Slides were washed and mounted with Roti-Mount with DAPI (ROTH, Germany). Images were captured on KEYENCE BZ-X810 (Keyence, Japan) (×20 objective magnification) and analyzed using ImageJ. Only ZAP70 positive cells associated with the secondary lamella were counted using digitally magnified images. The following procedure was used for counting ZAP70-positive cells: (i) counting the secondary lamellae, (ii) counting the positive cells on previously counted lamellae, and (iii) counting the ratio of ZAP70-positive cells to the number of secondary lamellae.

DNA was isolated from gill tissue stored in RNAlater. First, the tissue was mechanically lysed in a Tissuelyser II (Qiagen, Germany), and DNA was extracted using the QIAamp DNA Mini Kit (Qiagen, Germany) according to the manufacturer’s protocol. After isolation, samples were diluted to 50 ng μl⁻¹ and stored at −80°C.

Total RNA isolation and synthesis of cDNA was performed as previously described (5, 36). In order to verify the purity of the RNA, selected samples without the addition of reverse transcriptase (no-RT) were used to check for genomic DNA contamination. cDNA samples were diluted 40× with nuclease-free water (Thermo Fisher Scientific, Germany).

Statistical analysis was carried out in a SigmaPlot 12.5 software (Systat Software). Data from viral load and gene expression studies were transformed using a Log10(x) transformation before statistical analysis. Significant differences (p ≤ 0.05) in viral load and in gene expression, cortisol, and glucose concentration were assessed using two-way ANOVA, with subsequent pairwise multiple comparisons using the Holm–Sidak test.

To compare the stress response between koi carp and Amur sazan, fish were subjected to the prolonged restraint stress protocol, and the cortisol and glucose levels in blood plasma as well as the expression of genes involved in the activation of the HPI axis were studied. As expected, restraint stress induced significant increase in the plasma levels of cortisol (from 134 to 575 ng/ml in koi and from 4 to 407 ng/ml in Amur sazan) and glucose (from 104 to 742 mg/dl in koi and from 62 to 763 mg/dl in Amur sazan). Interestingly, basal levels of cortisol and glucose in blood plasma were significantly higher in control koi than in control Amur sazan ( Figures 1A, B ).

Upon restraint stress, the gene expression study revealed an upregulated expression of il-1β in NPO of both koi and Amur sazan and in PIT of koi carp. Moreover, gr1 expression was downregulated in NPO of Amur sazan ( Figures 1C, D ; Supplementary Table S2 ). In HK of both carp strains, the restraint stress induced expression of the 11βhsd2 gene (encoding 1β-hydroxysteroid dehydrogenase type 2 converting active cortisol into inactive cortisone), while in HK of Amur sazan, the restraint stress downregulated the expression of the cyp11b1 gene (encoding 1β-hydroxylase involved in conversion of cortisol from 11-deoxycortisol) in HK ( Figure 1E ; Supplementary Table S2 ).

No statistically significant differences were observed between unstressed koi and unstressed Amur ( Figures 1C–E ; Supplementary Table S2 ). The only genes that differed significantly in expression level between stressed koi carp and Amur sazan were gr1 and mr in NPO. The expression of both genes was significantly higher in stressed koi compared to stressed Amur sazan ( Figure 1C ; Supplementary Table S2 ). Thus, our study demonstrated similar responses of koi and Amur sazan to restraint stress, although the basal levels of plasma cortisol and glucose were higher in koi carp.

There were clear differences in KSD development between 12°C and 18°C. At 12°C, only CEV-infected koi showed mild clinical signs of infection (slight lethargy) at day 6 post exposure to CEV-positive donors, while CEV-infected salt-treated koi and Amur sazan did not display any clinical signs of disease. At 18°C, CEV-infected koi started to demonstrate clinical signs of KSD at 5 days post exposure. At 6 days post exposure, fish showed severe lethargy, were laying at the bottom of the tank on the side of their body, and were not responsive to stimuli. CEV-infected salt-treated koi and Amur sazan as well as uninfected control fish did not display any clinical signs of KSD.

Gill tissues collected at 6 days post exposure were studied for histopathological changes. At 12°C, gills of CEV-infected koi showed partial occlusion of the intralamellar space, most likely related to hyperplasia of epithelial cells of the gills, and low infiltration of eosinophilic cells were noted. CEV-infected salt-treated koi showed similar histopathological changes, while Amur sazan and control uninfected fish did not show any obvious gill pathology. At 18°C, nearly complete occlusion of the intralamellar space with the accumulation of cellular debris and an infiltration of eosinophilic cells were noted in the gills of CEV-infected koi at 6 days post exposure. Despite showing no visible clinical signs of KSD, salt-treated CEV-infected koi showed almost as severe histopathological changes as CEV-infected koi. In Amur sazan and uninfected controls, no histopathological changes were noted in the gills ( Figure 2 ).

Quantification of CEV at 6 days post exposure revealed significantly higher viral load in the gills of koi and Amur sazan kept at 18°C than in fish kept at 12°C ( Figure 3 ). The same trend was also observed in salt-treated koi, although the differences were not statistically significant. At 18°C, the highest viral load was demonstrated in koi carp (mean 1.77 × 10⁶ copies) followed by salt-treated koi (mean 1.64 × 10⁵ copies), while the lowest viral load was measured in Amur sazan (mean 2.9 × 10¹) ( Figure 3 ). These results demonstrate the effect of temperature and carp genetic background on the viral load.

Chemical parameters were analyzed in the blood of uninfected controls and CEV-infected (at 6 days post exposure to KSD-affected donor koi) koi, salt-treated koi, and Amur sazan, at both 12°C and 18°C. Blood parameters of koi and salt-treated koi were strongly affected by CEV infection at both 12°C ( Table 1 ) and 18°C ( Table 2 ). At 12°C CEV-infected koi showed increased levels of pH, BE, tCO₂, HCO₃, and stHCO₃ and decrease in pO₂, Na⁺, and Ca²⁺. In turn, in salt-treated koi, CEV elevated the pH, pCO₂, BE, tCO₂, HCO₃, and stHCO₃, and decreased pO₂, Na⁺, K⁺, and Ca²⁺ were measured. In the blood of CEV-infected Amur sazan kept at 12°C, the level of the analyzed parameters was not significantly different compared to uninfected fish, except for the pCO₂, which was lower in the CEV-infected fish than in uninfected controls ( Table 1 ).

CEV infection at 18°C resulted in increased pH, BE, tCO₂, HCO₃, stHCO₃, K⁺, tHB, and Hct(c) and decreased Na⁺ and Ca²⁺ concentrations in koi. Salt-treated koi infected with CEV showed higher levels of pH, pCO₂, BE, tCO₂, HCO₃, and stHCO₃ and a decrease in pO₂ and Na⁺ than the uninfected control group. Increased levels of tHB and Hct(c) were the only changes in blood parameters of infected Amur sazan ( Table 2 ).

To study the effect of CEV infection on stress response in koi carp and Amur sazan, cortisol and glucose levels were measured in blood plasma. In koi carp kept at 12°C, infection with CEV resulted in significantly increased cortisol concentration (from 11 to 160 ng/ml in koi and from 16 to 97 ng/ml in salt-treated koi) and in increased glucose concentration (from 61 to 434 mg/dl in koi, and from 69 to 198 mg/dl in salt-treated koi). In contrast, there were no significant differences in the cortisol and glucose concentrations between CEV-infected and control Amur sazan ( Figures 4A, C ). At 18°C, cortisol and glucose levels were significantly higher in all infected groups of fish compared to uninfected control fish, with the exception of glucose level measured in the blood of Amur sazan. In all strains/groups of fish kept at 18°C, infection with CEV resulted in increased cortisol concentration (from 57 to 573 ng/ml in koi, from 6 to 72 ng/ml in salt-treated koi, and from 7 to 13 ng/ml in Amur sazan) and in increased glucose concentration (from 92 to 442 mg/dl in koi and from 57 to 180 mg/dl in salt-treated koi) ( Figures 4B, D ).

In the CEV-infected fish, significantly higher cortisol and glucose concentrations were observed in koi carp compared to salt-treated koi and Amur sazan at both temperatures, except for cortisol level at 12°C, where no significant differences were observed between koi and salt-treated koi carp. At 18°C, but not at 12°C, the basal cortisol level in uninfected fish was significantly higher in koi compared to salt-treated koi and Amur sazan ( Figure 4B ). There were no significant differences in the basal concentration of glucose between uninfected fish from different strains/treatment at both temperatures ( Figures 4C, D ).

The CEV-induced stress response was also studied by analyzing the expression of genes involved in the stress response in the HPI axis organs and in the gills of fish kept at 12°C and 18°C. At 12°C, no significant changes in gene expression after CEV infection were observed in NPO and PIT in koi, salt-treated koi, and Amur sazan, with the exception of gr1, which was downregulated in PIT of CEV-infected salt treated koi, compared to uninfected control salt-treated koi ( Figures 5A, B ; Supplementary Table S3 ). In HK, CEV infection downregulated mr and gr1 expression in koi, upregulated il-1β expression in salt-treated koi, and upregulated mr and downregulated il-1β expression in Amur sazan at 12°C ( Figure 5C ; Supplementary Table S3 ). In the gills, CEV infection induced il-1β and reduced mr expression in koi and salt-treated koi at 12°C ( Figure 5D ; Supplementary Table S3 ).

At 18°C, CEV infection induced significantly the expression of all examined stress-related genes in NPO of koi and salt-treated koi (with the exceptions of pomc and il-1β in the salt-treated group), whereas only a slight upregulation of crh, crhbp, and gr1 expression was observed in Amur sazan ( Figure 5E ; Supplementary Table S4 ). In PIT of fish kept at 18°C, CEV infection induced significant upregulation of the expression of crhbp and il-1β and downregulation of mr and gr1 in the case of koi and downregulation of gr1 in salt-treated koi. No significant changes in gene expression between control and infected fish were observed in Amur sazan ( Figure 5F ; Supplementary Table S4 ). In HK, upregulation of il-1β and downregulation of mr and gr1 expression was observed in CEV-infected koi at 18°C. In salt-treated koi, CEV infection downregulated expression of the star gene (encoding steroidogenic acute regulatory protein involved in cholesterol transport to mitochondria) and cyp11b1 (encoding cytochrome P450 family 11 subfamily B member 1 protein involved in steroid synthesis). No significant changes in gene expression were observed in HK between control and infected Amur sazan kept at 18°C ( Figure 5G ; Supplementary Table S4 ). In the gills, CEV infection in fish kept at 18°C reduced gr1 and gr2 expression and induction of il-1β expression in koi and salt-treated koi. Moreover, downregulation of mr expression was observed in infected salt-treated koi. No significant changes in gene expression were observed between control and infected Amur sazan kept at 18°C ( Figure 5H ; Supplementary Table S4 ). Together, our results clearly demonstrated that CEV infection induces stress response, which is more pronounced in susceptible koi fish at 18°C, and this also correlates with the results of the viral load analysis.

Multiplex RT-qPCR enabled the analysis of the expression of genes involved in the oxysterol pathway, antiviral response, adaptive immune response, inflammation, apoptosis, and cell stress in CEV-infected (at 6 days post exposure of fish to CEV-infected donors) and control fish. The results presented in a heatmap ( Figures 6 , 7 ) were later confirmed using RT-qPCR with selected genes involved in antiviral and adaptive immune response.

In koi and salt-treated koi, CEV infection at 12°C induced the expression of genes involved in the oxysterol pathway (ch25h, cyp7b1, fdps, and gpr183). Expression of antiviral response genes (gig, ifna2, vig, pkz, irf7, and mx2) was upregulated in CEV-infected koi and infected salt-treated koi. Expression of some of these antiviral genes (gig, irf7, and mx2) was also upregulated in Amur sazan. CEV infection induced upregulation of the expression of selected adaptive-immune genes in koi (tcra1, cd8b1, and cd8b2) and in salt-treated koi (cd4, cd8a1, cd8b1, and cd8b2). Expression of genes involved in inflammatory response was also upregulated in infected koi (il-1β, il-10, and mpo) and infected salt-treated koi (il-1β and il-10). In Amur sazan, no upregulation of the expression of adaptive and inflammatory response genes was observed. Moreover, in koi and salt-treated koi, CEV infection upregulated the expression of genes encoding proteins involved in apoptosis (casp6, tp53, afap1, and iap/birc). Interestingly, casp9 expression was upregulated upon CEV infection in koi and downregulated in Amur sazan. Expression of genes related to cell stress (hsp70, serpinb1, and fel) was also upregulated in infected koi, while only hsp70 expression was upregulated in infected salt-treated koi. In Amur sazan, no changes in the expression of cell stress-related genes were demonstrated upon CEV infection ( Figure 6 ; Supplementary Table S5 ).

In koi and salt-treated koi, CEV infection performed at 18°C resulted in the upregulation of the expression of genes involved in the oxysterol pathway (ch25h, ch25-like, and fdps). Expression of antiviral genes (gig1, ifna2, vig1, irf7, mx2, and trim21) was upregulated in infected koi and infected salt-treated koi, while the expression of gig1, irf7, mx2, and trim21 was elevated in infected Amur. Moreover, the expression of pkz was upregulated in salt-treated CEV-infected koi, but was downregulated in CEV-infected koi. Interestingly, CEV infection of koi at 18°C resulted in the reduced expression of genes involved in adaptive immune response (tcra1, tcra2, cd4, cd8a1, cd8b1, and cd8b2). Expression of pro-inflammatory il-1β was upregulated in koi upon CEV infection, but expression of gene excoding myeloperoxidase (mpo) was downregulated in these fish. Expression of apoptotic genes was differently altered by the CEV infection in koi, where expression of casp6 and casp9 was downregulated, while iap/birc expression was upregulated. During CEV infection at 18°C, hsp70 expression was significantly upregulated in koi, while expression of fel was downregulated in koi and salt-treated koi ( Figure 7 ; Supplementary Table S6 ).

RT-qPCR analysis of selected genes was performed to validate the multiplex RT-qPCR results. Our study confirmed the upregulation of the expression of antiviral genes (mx2) and downregulation of the expression of adaptive response genes (tcra2, cd4, and cd8b1) in CEV-infected koi at 18°C ( Figures 8A–D ). Similar to the results of multiplex RT-qPCR, expression of mx2 was also upregulated in CEV-infected salt-treated koi compared to that in control fish ( Figure 8A ), whereas expression of tcra2, cd4, and cd8b1 was not changed in CEV-infected salt-treated koi and Amur sazan compared to that in control fish. ( Figures 8B–D ).

Expression analysis of cd3 and zap70 genes combined with immunofluorescent staining using anti-ZAP70 antibodies (35) in the gills of CEV-infected and control koi, both kept at 18°C, was also performed to assess the effect of CEV infection on T cells in the gills ( Figures 9A–F ). Our results demonstrated a lower expression of cd3 and zap70 in the gills of CEV-infected koi compared to that in control uninfected koi ( Figures 9A, B ). However, immunofluorescent staining revealed a higher number of ZAP70-positive cells in the gills of CEV-infected koi ( Figures 9C–F ).

Our transcriptional analysis showed that CEV may affect immune responses depending on temperature, which correlates with viral load and KSD development. At both temperatures, we observed an upregulation of antiviral immune response genes in koi and salt-treated koi upon CEV infection, but it was only in infected koi at 18°C that we found clear CEV-induced downregulation of the expression of adaptive immune response genes. Whether this is accompanied by suppression of the immune response or is indicative of a redistribution of leukocytes following infection requires further investigation.