Acquired Protective Immunity in Atlantic Salmon Salmo salar against the Myxozoan Kudoa thyrsites Involves Induction of MHIIβ+ CD83+ Antigen-Presenting Cells (original) (raw)

ABSTRACT

The histozoic myxozoan parasite Kudoa thyrsites causes postmortem myoliquefaction and is responsible for economic losses to salmon aquaculture in the Pacific Northwest. Despite its importance, little is known about the host-parasite relationship, including the host response to infection. The present work sought to characterize the immune response in Atlantic salmon during infection, recovery, and reexposure to K. thyrsites. After exposure to infective seawater, infected and uninfected smolts were sampled three times over 4,275 degree-days. Histological analysis revealed infection severity decreased over time in exposed fish, while in controls there was no evidence of infection. Following a secondary exposure of all fish, severity of infection in the controls was similar to that measured in exposed fish at the first sampling time but was significantly reduced in reexposed fish, suggesting the acquisition of protective immunity. Using immunohistochemistry, we detected a population of MHIIβ+ cells in infected muscle that followed a pattern of abundance concordant with parasite prevalence. Infiltration of these cells into infected myocytes preceded destruction of the plasmodium and dissemination of myxospores. Dual labeling indicated a majority of these cells were CD83+/MHIIβ+. Using reverse transcription-quantitative PCR, we detected significant induction of cellular effectors, including macrophage/dendritic cells (mhii/cd83/mcsf), B cells (igm/igt), and cytotoxic T cells (cd8/nkl), in the musculature of infected fish. These data support a role for cellular effectors such as antigen-presenting cells (monocyte/macrophage and dendritic cells) along with B and T cells in the acquired protective immune response of Atlantic salmon against K. thyrsites.

INTRODUCTION

The Myxozoa are obligate endoparasitic metazoans comprising over 2,000 species (1, 2) and are common parasites of fish, invertebrates, and occasionally vertebrates in fresh and marine environments. Infections with some myxozoa cause serious economic impact to fisheries and aquaculture, such as in whirling disease (Myxobolus cerebralis) and soft flesh syndrome (Kudoa thyrsites) of salmonids (3, 4). The genus Kudoa is comprised of 44 described species distributed throughout marine and estuarine fishes worldwide, within which they infect a range of tissues, including ovary, brain, kidney, and muscle (3). Kudoa thyrsites completes development with the formation of myxospores within plasmodia in the muscle of over 35 species of marine fish, and clinical signs are not observed. However, after 38 to 56 h postmortem, parasite-derived proteases digest the surrounding muscle fibers and produce a pH-dependent myoliquefaction that is proportional to the severity of infection (5–7). The resulting degraded fillet quality causes economic losses to both wild fisheries globally (6) as well as finfish aquaculture in Western Canada (8), and infections have been observed in Ireland (9), Chile (9), and Australia (10). Notably, in British Columbia, Canada, K. thyrsites presents a significant challenge to the salmonid aquaculture industry (7), with losses reaching Can$6 million in 2015 (Marine Harvest Canada, personal communication).

Most identified myxozoans are known or inferred to possess a biphasic life cycle involving an invertebrate definitive host and a vertebrate intermediate host (11), which becomes infected by exposure to the actinospore stage. Indeed, neither the life cycle of K. thyrsites nor other aspects of its host-parasite relationship are well understood. Plasmodia occur in Atlantic salmon myocytes after 9 weeks of K. thyrsites exposure (12), and infections typically resolve between 26 and 52 weeks postexposure (13).

There is no available treatment or prophylactic to prevent infection by K. thyrsites. In commercial aquaculture, efforts to minimize impacts of infection rely on early detection (8) or improving site selection and fish husbandry (14). Some orally administered compounds have been successful in reducing myxozoan infections in teleosts, including K. thyrsites in Atlantic salmon (Salmo salar) (15, 16); however, adverse effects have limited commercial applications. Thus, the lack of available options to prevent and/or treat K. thyrsites infections necessitates a more thorough understanding of the host-parasite relationship, particularly the host response.

Myxozoan parasites of fish elicit a range of host-specific responses. Some cause little or no host cellular response, some occupy immunoprivileged sites to escape host detection, and others elicit inflammation or seroconversion in the host (17, 18). There is an apparent variability in host susceptibility to K. thyrsites. For example, among Pacific salmon (Oncorhynchus spp.), the parasite has been observed in farmed and wild Coho (O. kisutch) salmon (19) but not in farmed Chinook (O. tshawytscha) salmon (20). In returning adult wild Pacific salmonids, the parasite was present in the cardiac muscle of Coho, Chinook, and pink (O. gorbuscha) salmon but not in chum (O. keta) and sockeye (O. nerka) salmon (21). Differential susceptibility to another myxozoan, Ceratonova shasta, is well documented among Pacific salmon species and among stocks within salmon species (reviewed in reference 22). Similarly, inter- and intraspecific differences in susceptibility to infection have been reported in brown trout (S. trutta) and rainbow trout (Oncorhynchus mykiss) to the causative agent of proliferative kidney disease, Tetracapsuloides bryosalmonae (23), common carp (Cyprinus carpio) to Thelohanellus nikolskii (24), and gilthead sea bream (Sparus aurata) and turbot (Scophthalmus maximus) to Enteromyxum scophthalmi (25).

Although not well studied, there are multiple examples of cellular and acquired immunity to this group of parasites. For example, Cuesta et al. (26) demonstrated strong cellular responses in the intestinal mucosa of gilthead sea bream to Enteromyxum spp., while Davey et al. (27) showed interferon (IFN)-stimulated and major histocompatibility complex (MHC) class II genes were important at the local level. In rainbow trout, a large accumulation of IgT+ cells in the intestinal lamina propria has been demonstrated in fish surviving infection with C. shasta (28).

Pathological responses associated with K. thyrsites infections occur in mahi mahi, Coryphaena hippurus (29), and Pacific hake, Merluccius productus (30), and inflammation coincided with resolution of infection in Atlantic salmon (31). In Atlantic salmon, sexual maturation increases the likelihood of high infection severity (14); however, the innate mechanisms of resistance, including host recognition, cellular targeting, and/or destruction of parasite stages, remain unknown. Experiments conducted by Jones et al. demonstrated that following resolution of K. thyrsites infections, Atlantic salmon are protected against subsequent infections (32). To investigate the mechanisms responsible for resolution and the ensuing protective immune response, previously unexamined histological samples from the latter study were probed either with histochemical stains or with monoclonal antibodies targeting cellular effectors, and flash-frozen samples were screened de novo for immune-related transcript expression. The objectives for the current study were to characterize the cellular immune response of Atlantic salmon during resolution of the initial infection and subsequent protection against a secondary exposure.

RESULTS

K. thyrsites infection resolves over time, and surviving fish are protected against reinfection.

The severity of K. thyrsites infection was measured in control and infected groups at T1 (1,985 degree-days [dd]), T2 (3,500 dd), T3 (4,275 dd), and T4 (6,225 dd) (Fig. 1A). The severity of infection (mean number of plasmodia per square millimeter ± standard deviations [SD]) at T1, T2, and T3 was 1.42 ± 2.29, 0.93 ± 1.66, and 0.12 ± 0.19 and 1.81 ± 1.68, 1.58 ± 1.88, and 0.08 ± 0.11 following brief and long exposures, respectively. There was no significant difference in mean severity between brief and long exposure times at any time point; therefore, the two groups were pooled at each time for subsequent analysis. We did not detect plasmodia in naive control fish until after secondary exposure (T4). At this time, the mean severity in controls was comparable to that of infected fish at T1 and significantly different from that of reexposed infected fish at T4 (1.22 ± 1.86; P < 0.001). There was no evidence of reinfection in previously exposed fish at T4 (Fig. 1A). We observed a decreasing trend over time (T1→T4) in the abundance of plasmodia in infected fish despite a secondary exposure with raw seawater (RSW) at 4,275 dd (P < 0.001).

FIG 1

FIG 1 (A) Infection severity index of K. thyrsites measured by plasmodia/mm2 in Atlantic salmon muscle samples at 1,985 dd (T1), 3,500 dd (T2), and 4,275 dd (T3), and after secondary exposure at 6,300 dd (T4), in control (Cnt; red) and infected (Inf; blue) fish. (B) RT-qPCR of K. thyrsites 18S rRNA in control and infected fish. Bars represent the mean CT value (±SD). Differences among groups were determined using two-way ANOVA followed by a post hoc Tukey's HSD, with a cutoff P value of <0.01. Significant differences between groups are denoted by lowercase letters. The arrow represents the secondary exposure, where both control and infected fish were exposed to infective salt water for ∼550 dd.

The presence of K. thyrsites was also quantified by measuring the abundance of r18S (kt18S) transcript in control and infected salmon muscle by reverse transcription-quantitative PCR (RT-qPCR) (Fig. 1B). There was significantly higher abundance in infected fish at T1 (P < 0.0001) and T2 (P < 0.0001) compared to that of control fish. At T2, 3 fish in the control group showed the presence of K. thyrsites 18S mRNA (data not shown). We excluded these animals from subsequent analysis. At T4, kt18S transcript abundance in the control fish (naive to infection) was significantly elevated (P < 0.0001) and reached levels comparable to those observed in the infected groups at T1. In contrast, there was a significant decrease in kt18S expression at T4 in the infected group (reexposed group) compared to the control group (P < 0.0001). Moreover, there was a significant decrease in kt18S expression in infected fish over time (T1→T4; P < 0.001).

Antiparasitic response involves activation of MHIIβ+, CD8+, CD83+, and IgM+ cells.

The monoclonal antibody Sasa MH class II β F1-4 (α-MHIIβ) detected cells that were large, with an amorphous nucleus, diffuse cytoplasm, and dendritic-like projections, or cells that had morphological features more similar to those of lymphocytes. In the musculature of K. thyrsites-infected salmon, these cells were associated with infected myocytes in a pattern divisible into four major stages (Fig. 2): initial detection of the infected myocyte, which was completely surrounded by MHIIβ+ cells (Fig. 2A, stage 1); infiltration of the infected myocyte by MHIIβ+ cells (Fig. 2A, stage 2); complete infiltration and degradation of the myocyte by MHIIβ+ cells (Fig. 2A, stage 3); and complete degradation of the plasmodia and engulfment of myxospores by MHIIβ+ cells in the immediate area or distal to the myocyte (Fig. 2A, stage 4). MHIIβ+ cells were also scattered throughout the musculature of naive control fish. In these fish at T1 to T3, we did not observe stages of myocyte infiltration by MHIIβ+ cells; however, after exposure of control fish to RSW, a pattern of MHIIβ+ staining and infiltration was observed that was similar to that of infected fish at T1.

FIG 2

FIG 2 Immune detection of K. thyrsites by MHIIβ+ cells. (A) Photomicrographs of stages of cellular response. In stages 1 and 2, MHIIβ+ cells surround and infiltrate an infected myocyte. In stage 3, MHIIβ+ cells are recruited to the infected myocyte and surround the plasmodia (p) while disintegrating the myocyte. In stage 4, degraded plasmodia (star) and released spores are engulfed by MHIIβ+ cells (red arrowhead). (B) Proportion of stages of detection by MHIIβ+ cells in infected fish at T1 to T4 (InfT1, InfT2, InfT3, and InfT4) and in control fish at T4 (CntT4). (C) Number of MHIIβ+ cells in the musculature of infected salmon over time. The arrow represents the secondary exposure, where both control and infected fish were exposed to infective seawater for ∼550 dd. Significance was determined using a two-way ANOVA followed by post hoc Tukey's HSD with P < 0.05. Lowercase letters denote differences between groups at each sampling time, while an asterisk denotes differences over time in each group.

Plasmodia were classified as positive (+plas) for immune detection if a myocyte was surrounded entirely by or infiltrated with MHIIβ+-stained cells or if the myocyte was disintegrated with observable myxospores. If a myocyte contained plasmodia but no MHIIβ+ cells, then it was considered negative or not recognized by the immune system (−plas). The ratio of +plas to −plas was calculated for each group and served as a proxy for the host response to K. thyrsites. There was a shift in the ratio of +plas to −plas over time (Fig. 2B).

There were significantly more MHIIβ+ cells in the musculature of infected fish than in control fish from T1 to T3 (P < 0.001) (Fig. 2C). After secondary exposure, the number of labeled cells in control fish increased significantly to levels comparable to those of infected fish at T1 (P < 0.001). In contrast, the number of labeled cells did not increase after secondary exposure in previously infected fish (P = 0.695) and was not significantly different from that of controls prior to exposure (P = 0.959). There was a significant positive correlation between severity of infection (plasmodia per square millimeter) and abundance of MHIIβ+ cells (P = 0.002; Pearson's r = 0.30) (see Fig. S1 in the supplemental material).

The monoclonal antibody Sasa CD8α (α-CD8α) detected cells associated with infected myocytes in all stages of infection. The CD8a+ cells either surrounded or infiltrated infected myocytes or were present in lesions of cellular aggregates (Fig. 3A). These cells presented variable morphology; some were small and round as previously described (33), while others were larger and diffuse, with large nuclei and a dendritic-like morphology (Fig. 3A). CD8α+ cells were not observed in the musculature of noninfected fish (data not shown).

FIG 3

FIG 3 Photomicrographs of K. thyrsites-infected muscle tissue (T2; 3,500 dd) probed with the monoclonal antibody Sasa CD8α. (A) CD8α+ cells (brown) associated with myocytes containing intact plasmodia (p) as well as with disintegrated myocytes. Positive cells were observed with engulfed myxospores (light blue; arrowhead) (a), infiltrating infected myocytes (arrowhead) (b), or associated with fibrinolytic lesions (asterisk) (c). There appeared to be two morphologies of CD8α+ cells, dendritic-like (d) and lymphocyte-like (e). (B) Dual staining with α-CD8α (red) and α-MHIIβ (brown). Dual-labeled cells were associated with fibrinolytic lesions (a, asterisk) or infiltrating infected myocytes (a and b, arrowheads).

Dual labeling of tissue sections with α-MHIIβ revealed a CD8α+/MHIIβ+ population of cells in inflammatory lesions associated with infected myocytes (Fig. 3B).

The monoclonal antibody Omy CD83 14-55-12 (α-CD83) detected a population of cells associated with later stages of infection and predominantly in areas of disintegrated myocytes. These cells were large, with a diffuse and irregularly shaped nucleus, and often contained myxospores (Fig. 4A). When colabeled with α-MHIIβ, it was seen that CD83+/MHIIβ+ cells were involved in all 4 stages of detection (see above) and were present in inflammatory lesions (Fig. 4B).

FIG 4

FIG 4 Photomicrographs of K. thyrsites-infected muscle tissue (T2; 3,500 dd) probed with monoclonal antibody Omy CD83 (red). (A, a and b) Positive cells were observed associated with later stages of infection after plasmodia were disintegrated (asterisk) and often with engulfed myxospores. (B) Dual labeling of CD83 (a; red) and MHIIβ (b; brown) revealed most cells were CD83+/MHIIβ+ (c to e; red and brown). Dual-labeled cells were observed at high densities within stage 4 lesions and associated with free myxospores or with engulfed myxospores (arrowheads).

The monoclonal Sasa IgM (α-IgM) detected cells in the musculature of uninfected and infected Atlantic salmon. Positive cells resembled lymphocytes with a round and compact nucleus. More of these cells were counted in infected fish (data not shown), were observed between myocytes in K. thyrsites-infected tissue, and were associated with lesions in later stages of infection (Fig. S2), but they were never observed infiltrating infected myocytes.

Kudoa thyrsites infection induces gene expression of several cellular effector molecules.

Transcript abundance of several cellular effectors and immune response markers were characterized in control and infected fish muscle at T1, T2, and T4. We first explored the data set using an unsupervised partitioning clustering analysis (R package cluster) with expression profiles of all sample-gene combinations to identify patterns of similar expression profiles. Despite significant variability in the data, _k_-means clustering analysis revealed 3 distinct expression profiles grouped according to infection status (Fig. 5). This analysis indicated that naive fish had expression profiles similar to those of infected fish that were protected against secondary infection, while infected fish early in the challenge were similar in gene expression to control fish after initial exposure. The final cluster was comprised of infected fish later in infection and reexposed infected fish.

FIG 5

FIG 5 Cluster analysis of all sample-gene combinations. (A) The optimal number of clusters was calculated by the k means gap statistic (k). (B) Cluster plot of all samples using k. This analysis showed 73.4% of the variability in the data set is explained by the time of infection with K. thyrsites. Cluster 1 is comprised of uninfected controls (27.4% and 22.5% for T2 and T1, respectively), reexposed infected fish (34.3%; T4), and infected fish (15.7%; T2). Cluster 2 was comprised of infected fish later in the infection (48.4%; T2) and infected fish after reexposure to K. thyrsites (19.4%; T4), control fish after secondary exposure (16.1%; T4), and infected fish early after primary infection (16.1%; T1). Cluster 3 was comprised of fish early after primary exposure (T1 infected fish, 37.5%; T4 control fish, 25%) and infected fish after reexposure (14.1%; T4).

We then performed hierarchical clustering and calculated Pearson's correlation coefficients (r) for all genes to understand the relationship in expression of the different cellular markers. There was significant correlation among the expression profiles of macrophage/dendritic cell (DC) markers (mhiib, mcsf, and cd83), T cell markers (cd8 and nkl), and B-cell markers (igt and igm) (Fig. 6A and B and Table S1).

FIG 6

FIG 6 (A) Correlational matrix of expression profiles for all genes of interest. The colored scale shows degree of correlation ranging from r = −1.00 (orange) to r = 1.00 (purple). The size of the colored circle indicates significance. See Table S1 in the supplemental material for associated r and P values. (B) Hierarchical clustering of log2 CNRQs for cellular markers with high positive correlation. Gene names are labeled on the base of the figure, while hierarchical clustering of the columns based on individual samples shows similar expression profiles of genes associated with particular cell types.

Macrophage/dendritic cell markers highly expressed during K. thyrsites infection.

In concordance with protein abundance of MHIIβ determined by immunohistochemistry (IHC), transcript abundance of _mhii_β was significantly higher in infected fish than in controls T2 (P < 0.001), and this expression decreased over time (T2→T4) (Fig. 7). After secondary exposure (T4), expression of _mhii_β was significantly higher in control fish than in reinfected fish. In the latter group, expression was comparable to that of uninfected controls.

FIG 7

FIG 7 Expression profiles of macrophage/dendritic cell markers (mhii, mcsf, and cd83) in control and infected salmon at T1 (1,985 dd), T2 (3,500 dd), and T4 (6,225 dd). Prior to sampling at T4, both control and infected fish were reexposed (dotted line, arrowhead). Expression differences are represented by log2 calibration-normalized relative quantities (CNRQ), with boxplots and whiskers showing the median expression and 95% confidence intervals, respectively. Statistical differences were detected with a two-way ANOVA followed by a post hoc Tukey's HSD for pairwise comparisons (P < 0.05). Differences between groups are denoted by lowercase letters, while differences over time are denoted by an asterisk.

Similar to _mhii_β, transcript abundance of cd83 and mcsf was higher in infected fish at T2 (P < 0.0001) and in control fish at T4 following the secondary exposure (Fig. 7). In the reexposed fish, there was a significant decrease in expression of _mcsf_ and _cd83_ (_P_ values of <0.005 and <0.001, respectively). There was a significant correlation (_r_ > 0.80, P < 0.0001) among the expression levels of mhii, cd83, and mcsf, indicating coexpression of these transcripts (Fig. 6B).

There was a significant difference in transcript abundance of c-type lectin clec4m in infected fish compared to that of control fish at T2 (Fig. S3) (P < 0.001), but this response was not observed after reexposure. There was no increase in clec4m expression following exposure of control fish.

Evidence for a cytotoxic T cell response to K. thyrsites.

Expression of cd8 and nkl was significantly induced in infected fish compared to controls at T1 (P = 0.026 and P = 0.002, respectively) and at T2 (P < 0.05 and P < 0.001, respectively) (Fig. 8). However, we failed to detect significant differential expression of cd4 (data not shown). After secondary exposure, expression of tcr, cd8, and nkl increased in controls, while in reexposed infected fish there was sustained expression of nkl only. In contrast, transcript abundance of tcr and cd8 decreased after reexposure (P < 0.01 and P < 0.05, respectively) and was lower than that of controls (P < 0.01). The expression of both nkl and tcr was positively correlated with cd8 (r = 0.88 and r = 0.80, respectively; P < 0.0001) (Fig. 6B), as was the expression of tcr and cd83 (r = 0.79, P < 0.0001).

FIG 8

FIG 8 Expression profiles of T cell markers (cd8, tcr, and nkl) in control and infected salmon at T1 (1,985 dd), T2 (3,500 dd), and T4 (6,225 dd). For interpretation, see the description in the legend to Fig. 7.

B cell markers activated in response to K. thyrsites.

There was significant induction of both igt and igm expression in infected fish compared to that of controls at T2 (P < 0.00001) (Fig. 9). After secondary exposure, the abundance of both igm and igt was significantly induced in control fish (P < 0.001), with sustained expression in reinfected fish. There was a strong positive correlation between igt and igm (r = 0.86, P < 0.0001) (Fig. 5C), indicating coregulation of these transcripts.

FIG 9

FIG 9 Expression profiles of B cell markers (igm and igt) in control and infected salmon at T1 (1,985 dd), T2 (3,500 dd), and T4 (6,225 dd). For interpretation, see the description in the legend to Fig. 7.

Interleukins.

There was significant but transient upregulation of il12 in infected fish compared to that of control fish at T1 (P < 0.001) (Fig. 10). After secondary exposure, transcript abundance was significantly induced in control fish at T4 (P = 0.007) to levels comparable to those of infected fish observed at T1. We failed to detect reliable expression of il4 (data not shown).

FIG 10

FIG 10 Expression profiles of cytokines (il4 and il12) and marker of cellular proliferation (pcna) in control and infected salmon at T1 (1,985 dd), T2 (3,500 dd), and T4 (6,225 dd). For interpretation, see the description in the legend to Fig. 7.

Cellular proliferation.

There was a significant induction of pcna over time (T1→T2) in infected fish (P < 0.00001), and at T2 expression in infected fish was higher than that of control fish (P < 0.001). After secondary exposure, pcna expression increased in control fish (P < 0.0001) and was significantly higher than that of reexposed infected fish (P < 0.01) (Fig. 10).

DISCUSSION

The present study demonstrated that the response of Atlantic salmon to myocytes infected with K. thyrsites involves infiltration of the lesion with MHIIβ+, MHIIβ+/CD83+, MHIIβ+/CD8α+, CD83+, and IgM+ cells. This response is further characterized by upregulation of Mϕ/DC (mhii, cd83, and mcsf), cytotoxic T cell (_cd8_α and tcr), and B cell (igm and igt) genetic markers. Furthermore, we provide evidence for cytotoxic activity in the significant expression of nkl and its upstream inducer, il12. The expression of these markers was positively correlated with both the abundance of K. thyrsites 18S rRNA and infection severity. Taken together, these results indicate a cell-mediated immune response is involved in the resolution of K. thyrsites infection in Atlantic salmon, and further that CD8α+ cytotoxic T cell killing is a candidate mechanism for protection upon reexposure to the parasite.

We observed a population of MHIIβ+ cells whose abundance was proportional to infection severity and which progressively infiltrated infected myocytes and appeared to be associated with disintegration of the infected cell. MHIIβ+ cells were first observed surrounding intact infected myocytes, while infiltration of the infected myocyte by these cells was observed later. As the myocyte became degraded, the plasmodium gradually lost integrity and myxospores eventually were released and subsequently ingested by phagocytes, including MHIIβ+ cells. The presence of macrophages (Mϕ) within infected salmon myocytes has been previously reported at a low level (34). Here, infiltrating MHIIβ+ cells were frequently observed within plasmodia, and as the infection progressed, most plasmodia had deteriorated and MHIIβ+ cells were observed with engulfed myxospores. We propose that this population of cells is directly involved in parasite detection and clearance. Additionally, the persistence of these cells in fish demonstrating protection against reinfection suggests they also play a role in the acquired protective immune response. This work challenges an earlier paradigm in which stages of the infection prior to myocyte rupture went undetected by the host (3, 34).

Dual labeling of infected muscle showed that many of the cells involved in detection of K. thyrsites were MHIIβ+/CD83+. Thus, concordant upregulation of _mhii_β with cd83, a hallmark marker for mature DCs (35, 36), together with immunohistochemical data, confirms a large proportion of these cells are DCs. A recent study described a population of cells in rainbow trout possessing all of the characteristics of DCs (37), and subsequent studies have identified DCs in zebrafish (38), salmon (39), and trout (40). In the latter study, the authors describe a CD8α+/MHIIβ+ population of DCs in the skin of trout with potent phagocytic and cross-presentation abilities, while Haugland et al. described a population of phagocytic cells expressing cd83 and mh class II in Atlantic salmon as progenitor DCs (39). Thus, current evidence supports the existence of DCs in teleosts, which is intuitive given the importance of DCs in vertebrate innate immunity (41) and the repertoire of mammalian cellular equivalents already confirmed in teleosts (42–46).

Our data suggested a heterogeneous population of leukocytes is present in the musculature of K. thyrsites-infected salmon, which act in concert to eliminate the parasite. For example, in addition to _mhii_β and cd83, we also observed concordant upregulation of mcsf, a macrophage/monocyte marker, and igm and igt, surface immunoglobulins of teleost B cells which also possess antigen-processing capabilities (43). IgM+ cells were also in the same lesions as the MHIIβ+, CD8α+, and CD83+ cells. Lymphocyte-mediated responses have been described as important determinants of immunity in other myxozoan infections of fish. For example, a strong IgM+ B cell response was observed in turbot infected with E. scophthalmi (47) and gilthead sea bream infected with E. leei (48). In the present study, upregulation of both igm and igt in response to infection with K. thyrsites and IgM+ cells was associated with later stages of myocyte detection. Generally, upregulation of IgM is associated with a switch from Th1- to Th2-like cytokine profiles which is regulated by anti-inflammatory cytokines, including interleukin-4 (49). Paradoxically, despite a significant induction of igm in the muscle of K. thyrsites-infected salmon, we did not detect differential expression of the Th2 marker interleukin-4 at any time point throughout the infection, indicating that either the typical Th2-associated pathways are not induced during infection with this parasite or that transient transcription of this marker was missed due to sampling design.

In contrast, upregulation of il-12 indicated a Th1-type response in K. thyrsites-infected muscle. Interleukin-12 is produced by activated phagocytes (e.g., Mϕ, DCs, and B cells) in response to intracellular parasites in early phases of detection (50, 51). In mice, resistance to acute protozoan infections is characterized by early synthesis of IL-12 by CD8α+ DCs (52), which permits host survival in low-level chronic infections (53). IL-12 acts as a growth factor for activated T and NK cells and enhances cytotoxic T lymphocyte and NK cell killing by activating transcription of cytolytic molecules, including perforin and granzymes (54). In the present study, il-12 expression declined over time, suggesting a positive correlation between severity of K. thyrsites and il-12 production. Resistance to Ceratonova shasta infections in Chinook salmon is associated with overexpression of IFN-γ (49). Although the latter study did not assess levels of IL-12, the IL-12/IFN-γ regulatory pathway has been thoroughly characterized in vertebrates, including teleosts (55, 56). The absence of il-12 induction after reexposure of salmon to K. thyrsites suggests protection against subsequent infection cannot be explained by the presence of IL-12, as has been described for other intracellular parasites (51, 57, 58). Instead, the data suggest an important role for IL-12 early in the host response to K. thyrsites infections. Based on the importance of this cytokine during infection with other intracellular parasites, future studies should investigate a comparable mechanism during K. thyrsites infections in Atlantic salmon.

CD8+ DCs play a pivotal role in antigen presentation and T-cell priming against intracellular pathogens, including Listeria monocytogenes (59), Salmonella enterica serovar Typhimurium (60), and Plasmodium spp. (61). Upregulation of cd83 transcription was observed in a CD8α+/MHIIβ+ population of leukocytes in the skin of rainbow trout following poly(I·C) stimulation, further substantiating this population as a DC subset in teleosts (40). Here, we show significant upregulation of _mhii_β, cd8a, and cd83 in the Atlantic salmon in response to K. thyrsites. Furthermore, application of salmon-specific markers revealed a population of spore-phagocytic leukocytes in skeletal muscle coexpressing MHIIβ and CD8α or CD83 and having dendritic-like morphology. We propose this cell population is similar to the dendritic-like cells described by Granja et al. (40).

It is likely that some of the CD8α+ population of cells observed in this study were cytotoxic T lymphocytes (CD8α+/CD83−/MHIIβ-/+). Cytotoxic T lymphocytes use both secretory and nonsecretory killing mechanisms to protect the host against infected and transformed cells (42), and a subset of CD8+ T cells (known as effector memory cells) migrate to sites of infection and display immediate effector function (62). Activation of CD8+ T cells or adoptive transfer of these cells can induce protective immunity against protozoan parasites (63). In mice, persistent activation of CD8+ T cells following infection with viral, bacterial, or parasitic infection has been associated with increased protection against subsequent infection of Trypanosoma cruzi (64). Furthermore, NK-lysin released from activated CD8+ T cells has been shown to directly lyse T. cruzi-infected cells (65).

We observed a protective effect in Atlantic salmon after secondary exposure to K. thyrsites concomitant with persistent upregulation of _cd8_α and the presence of CD8α+ cells, suggestive of a comparable mechanism in fish. The presence of NK-lysins has been confirmed in teleosts (66–70), and recent work has shown that rainbow trout (O. mykiss) CD8α+ T cells express granulysin/NK-lysin and are the dominant cytolytic cell population in teleost fish (71). Interestingly, we detected significant and persistent upregulation of nk-lysin in response to K. thyrsites despite decreasing parasite severity over time. Furthermore, the pattern of nk-lysin expression was highly correlated with that of _cd8_α, implying coregulation of these two genes. Thus, infection with K. thyrsites in Atlantic salmon appears to elicit a cytotoxic T cell response characterized by upregulation of il12, cd8, and nkl.

Although our data indicate an activated immune response in the musculature of infected Atlantic salmon, the parasite stage targeted during reexposure is unknown. A putative actinospore stage may interact with a resident Mϕ or DC at the proposed infection site of K. thyrsites in the mucosa (72, 73) prior to establishing infection in striated muscle. Migratory (extrasporogonic) stages within teleost circulatory systems have been described in several myxozoans (74, 75). Indeed, in Atlantic salmon infected with K. thyrsites, weakly infective circulating extrasporogonic stages have been detected (13, 34), which may interact with antigen-presenting cells (APCs) to activate the host immune response. There is evidence of stage-specific surface antigens in K. thyrsites (76), raising the possibility of differential stimulation of defense responses during early and late stages of parasite development.

The current study demonstrated significant transcriptional induction of a c-type lectin (c-type lectin family m) associated with later stages of K. thyrsites infection. Either secreted as soluble proteins or expressed on the surface of DCs and Mϕ, c-type lectins are known to play an important role in host-parasite interactions (18, 77), including fish parasites (78–82). More importantly, the presence of carbohydrate terminals specifically detected by lectins on the surface of the myxozoans Tetracapsuloides bryosalmonae, Myxobolus cerebralis, and E. scophthalmi suggests a role in host-parasite interactions (83–85). Activation of this gene during the response to K. thyrsites may represent a comparable host-parasite interaction. However, clec4m expression was associated with later stages of infection only after a large proportion of infected myocytes had been disintegrated by the host response. Thus, it is unlikely that this molecule is involved in the initial host-parasite interaction but may instead be important during the acquired protective response reported here.

Conclusions.

In summary, this study provides evidence for an acquired cell-mediated immune response in Atlantic salmon infected with the myxozoan histozoic parasite K. thyrsites. The proteomic and transcriptomic data indicate a heterogeneous population of leukocytes is involved in the processing and protective response of Atlantic salmon to K. thyrsites. Specifically, an APC-mediated (Mϕ and DCs) cytotoxic T cell response is involved in the resolution of infection and in the protective response against subsequent infection.

MATERIALS AND METHODS

Fish husbandry.

Juvenile Atlantic salmon S. salar from a single commercial hatchery stock were maintained in a research aquarium at the Pacific Biological Station, Nanaimo, British Columbia, as described previously (32). Temperature ranged from 8.2 to 13.6°C, and mean salinity was 29.2 practical salinity units. The fish were smoltified in 6,500-liter flowthrough tanks supplied with UV-treated seawater (UVSW) and acclimated for a minimum of 2 weeks prior to experimentation. Salmon were fed a commercial pelleted diet (EWOS) at a daily rate of 1% biomass. Husbandry protocols followed guidelines of the Canadian Council on Animal Care. The fish were exposed to K. thyrsites by holding them in raw seawater (RSW; no UV treatment) as described below and in Jones et al. (32). The duration of all exposure events was expressed in degree-days (dd), which are the sum of daily water temperatures.

Experimental design and sampling procedure.

Fish (85 g) were randomly allocated into 6 2,500-liter tanks (n = 150/tank) with flow rates of 40 liters UVSW/min. Duplicate tanks were assigned to each of 3 treatment groups: UVSW only (naive controls), RSW (infected) for 440 dd (brief), and RSW for 950 dd (long). All fish were maintained in UVSW following the RSW exposure. Muscle sample collection was performed as described previously (32). Samples were collected from 20 fish/tank at 1,985 dd (T1) and 3,500 dd (T2) following the onset of exposure and from 13 to 20 fish/tank at 4,275 dd (T3). Immediately following T3 sampling, all remaining fish were exposed to RSW for 530 dd and then maintained in UVSW until 6,225 dd (T4), when final samples were collected (n = 25 to 30/tank). For sampling, fish were sedated in 0.25 mg liter−1 Aquacalm (Syndel Laboratories, Canada) and euthanized in 250 mg liter−1 MS-222 (Syndel Laboratories). Three skeletal muscle samples were dissected from the same locations on each fish immediately after euthanasia and fixed in 10% neutral buffered formalin (NBF) for 24 to 48 h, and then they were transferred to 70% isopropanol. A subset of samples (n = 20 from T1, T2, and T4) was flash-frozen in liquid nitrogen (LN2) and stored at −80°C until RNA extraction.

Histology and immunohistochemistry (IHC).

NBF-fixed samples were processed for histology as previously described (32). Serial sections (5 μm) were obtained (n = 15 fish per treatment, per time) using a Leica RM2135 microtome (Leica Microsystems, Germany) and placed on SuperFrost UltraPlus (Menzel-Gläser) positively charged glass slides to dry overnight at 40°C. After deparaffinization in xylene and rehydration in graded isopropanol, the sections were either stained with hematoxylin and eosin (H&E) for routine histopathology or probed with monoclonal antibodies against immune cell markers (Table 1).

TABLE 1

TABLE 1 Monoclonal antibodies used in immunohistochemistry indicating dilutions, antigen retrieval methods, and the original source

Antigen retrieval and immunolabeling was performed as previously described (87). After deparaffinization and rehydration, sections were heated to 100°C in antigen retrieval buffer (Table 1), cooled to room temperature for 10 min in phosphate-buffered saline (PBS), and then washed twice in Tris-buffered saline plus 0.2% Tween 20 (TBS-T; pH 8.0) for 5 min with gentle agitation. For MHIIβ and mIgM (membrane-bound IgM) detection, sections were blocked in protein blocker for 10 min before gentle rinsing with TBS-T. The sections were incubated with primary antibody in TBS-T and 1% bovine serum albumin (BSA; Sigma) overnight at 4°C in a humid chamber. After incubation, the sections were washed in TBS-T (two times for 5 min each) and incubated in a mouse-specifying reagent (Expose mouse/rabbit specific horseradish peroxidase [HRP]/3,3′-diaminobenzidine [DAB] kit; Abcam) for 10 min, followed by a 10-min incubation in hydrogen peroxide blocker. Labeled cells were detected after a 15-min incubation in a goat anti-rabbit HRP conjugate followed by 10 min with DAB in PBS with 0.015% H2O2. For detection of CD83 and CD8α, an alkaline phosphatase (AP) detection kit (Expose rabbit/mouse specific AP kit; Abcam) was used according to the manufacturer's instructions. Blocking, incubations with primary antibodies, and washings were as described above. Sections were incubated in biotinylated goat anti-rabbit for 15 min, washed in TBS-T (two times for 5 min each), and then incubated in streptavidin AP for 15 min. After a final wash (two times for 5 min each), the sections were developed using the StayRed chromogen by following the manufacturer's instructions (Abcam).

All sections were counterstained in 1% Alcian blue (3 min) and Mayer's hematoxylin (diluted 1/20, 30 s), dehydrated in graded ethanol, cleared in xylene, and cover slipped (Permount). Sections treated with irrelevant antibodies served as negative controls, while sections known to contain the target molecules served as positive controls.

Positively labeled cells were quantified by counting the total number of cells in 10 fields of view (FoV) in an “S” pattern in each section at ×400 magnification (FoV of 0.25 mm2). If an FoV included a plasmodium, it was disregarded and the next FoV was considered to avoid bias due to exaggerated cellular abundance in these areas.

Infection severity for each time point was determined from H&E-stained sections by quantifying the number of plasmodia and the total area of each muscle section (ImageJ V2.0.0-rc-43/1.51 h). An infection severity index for each fish (n = 15 fish per treatment, per time) was calculated as the arithmetic average number of K. thyrsites plasmodia per square millimeter from the three muscle sections.

Antibody preparation and validation.

Antibodies against trout Oncorhynchus mykiss CD83 were raised in mice (NMRI) by immunizing with a synthetic peptide representing the C-terminal end of the amino acid residues from 141 to 154 (C-LESTDQSEERDTI) of CD83. The antigen was mixed with GERBU adjuvant (GERBU Biotechnik GmbH) in accordance with the manufacturer's procedure. Mice received 2 subcutaneous injections at least 14 days apart, with 25 μg of synthetic peptide coupled to diphtheria toxoid via the end cysteine (Statens Serum Institut, Copenhagen, Denmark). Immunized mice received an intravenous boost with 25 μg of the antigen, administered with adrenalin, 14 days later. The fusion of spleen cells and selection was done 3 days after the boost as described by Kohler and Milstein (88); however, the SP2/0-AG14 myeloma cell line was used as a fusion partner. Positive clones were selected by screening against the immunized peptide coupled to BSA in enzyme-linked immunosorbent assay (ELISA). Cloning was performed by limited dilution, and single clones were grown in culture flasks in RPMI plus 10% fetal calf serum (FCS).

Immunohistochemical characterization of the antibodies was done comparing the staining pattern in various tissues. Here another trout CD83 antibody, made against a different determinant (the N-terminal part of CD83 Ig-like domain), was used as the reference antibody. The CD83 MAb 14-54-04 was considered CD83 specific when it produced the same pattern of stained cells as the reference.

RNA extraction and qPCR.

Total RNA was isolated from frozen muscle samples (n = 20 fish per treatment per time) using a modified phenol-chloroform extraction method as previously described (89), followed by an on-column purification (RNeasy RNA purification kit; Qiagen). Potential contaminating genomic DNA was eliminated using Turbo DNase (Ambion) by following manufacturer's instructions for a routine digestion. Resulting RNA was quality checked using automated electrophoresis (Experion HighSens kit; Bio-Rad) and an RNA quality indicator cutoff of >7.5. High-quality RNA (2 μg) was synthesized into cDNA using the iScript reverse transcription supermix (Bio-Rad) with a mix of random hexamers and oligo(dT) primers by following manufacturer's instructions in 40-μl reactions.

Following cDNA synthesis, an aliquot from every sample was pooled prior to being diluted 3-fold. Primer efficiencies were determined from a 5-fold, 6-point serial dilution of the pooled cDNA sample. All assay efficiencies were between 90 and 105%, and specificity was determined by melting-point analysis and sequencing of PCR amplicons (Table 2).

TABLE 2

TABLE 2 Primers used for RT-qPCR, indicating the forward and reverse sequences, expected amplicon size, and original source

qPCR amplification was performed using an Sso-Fast advanced qPCR kit (Bio-Rad) per the manufacturer's instructions for the CFX thermal cycler (Bio-Rad) and using the following thermal regimen: 95°C for 30 s (1 cycle) and 95°C for 15 s, followed by 60°C for 30 s (40 cycles) and then a melt curve (65°C to 95°C reading fluorescence at 0.5-s increments). Samples were plated in triplicate using the Aurora automated plate dispenser run by VersaWare, and technical replicates were accepted only with a deviation of <0.5 threshold cycles (CT). No-template (NTC) and no-reverse transcriptase (NRT) controls were run for every gene.

Raw expression profiles were imported into qBASE+ (Biogazelle), and calibrated normalized relative quantities (CNRQs) were calculated. Elongation factor 1-α (_ef1_α), eukaryotic initiation factor-3 subunit-6 (eif3,6), ribosomal protein 40S (rps40), β-actin, and glyceraldehyde phosphate dehydrogenase (gapdh) were selected as candidate normalizer genes. Of these, _ef1_α, eif3,6, and β-actin showed the highest stability (geNorm M value and coefficient of variation of 0.592 and 0.235, respectively) and therefore were chosen for CNRQ analysis (95).

Resulting log2-transformed CNRQs (log2CNRQ) were tested for significance by two-way analysis of variance (ANOVA) and post hoc Tukey honestly significant difference (HSD) in R (v3.3.13; R Development Core Team 2012) with time and infection status as the two explanatory variables. Distance matrix analysis, correlational matrices, hierarchical clustering, and clustering analysis were also performed in R using the ggplot2 package and log2CNRQ values of all sample-condition combinations (n = 247).

ACKNOWLEDGMENTS

We thank Carter Van Iderstine, Heather Wotton, Kaitlin Fitzpatrick, and Brittany Ng for their excellent help with sample preparation. Holly Hicklin is gratefully acknowledged for fish care.

This research was funded in part by the Fisheries and Oceans Canada Aquaculture Collaborative Research and Development Program, Marine Harvest Canada, Cermaq Canada, and Novartis/Elanco Animal Health. L.M.B. was funded by an NSERC postdoctoral fellowship during this study.

We are grateful to Jordan Poley for valuable comments on an earlier version of the manuscript and to two anonymous reviewers for their excellent insight and comments that greatly improved the final manuscript.

We have no conflicts of interest to declare.

Supplemental Material

File (zii012172244s1.pdf)

• Download
• 873.23 KB

REFERENCES

Wolf K, Markiw ME. 1984. Biology contravenes taxonomy in the Myxozoa: new discoveries show alternation of invertebrate and vertebrate hosts. Science 225:1449–1452.

Chang ES, Neuhof M, Rubinstein ND, Diamant A, Philippe H, Huchon D, Cartwright P. 2015. Genomic insights into the evolutionary origin of Myxozoa within Cnidaria. Proc Natl Acad Sci U S A 112:14912–14917.

Moran JDW, Whitaker DJ, Kent ML. 1999. A review of the myxosporean genus Kudoa Meglitsch, 1947, and its impact on the international aquaculture industry and commercial fisheries. Aquaculture 172:163–196.

Hedrick RP, El-Matbouli M, Adkison MA, MacConnell E. 1998. Whirling disease: re-emergence among wild trout. Immunol Rev 166:365–376.

Dawson-Coates JA, Chase JC, Funk V, Booy MH, Haines LR, Falkenberg CL, Whitaker DJ, Olafson RW, Pearson TW. 2003. The relationship between flesh quality and numbers of Kudoa thyrsites plasmodia and spores in farmed Atlantic salmon, Salmo salar L. J Fish Dis 26:451–459.

Henning SS, Hoffman LC, Manley M. 2013. A review of _Kudoa_-induced myoliquefaction of marine fish species in South Africa and other countries. S Afr J Sci 109:1–5.

Funk VA, Olafson RW, Raap M, Smith D, Aitken L, Haddow JD, Wang D, Dawson-Coates JA, Burke RD, Miller KM. 2008. Identification, characterization and deduced amino acid sequence of the dominant protease from Kudoa paniformis and K. thyrsites: a unique cytoplasmic cysteine protease. Comp Biochem Physiol B Biochem Mol Biol 149:477–489.

Kent ML, Morgolis L, Whitaker DJ, Hoskins GE, McDonald TE. 1994. Review of myxosporea of importance in salmonid fisheries and aquaculture in British Columbia. Folia Parasitol (Praha) 41:27–37.

Whipps CM, Kent ML. 2006. Phylogeography of the cosmopolitan marine parasite Kudoa thyrsites (Myxozoa: Myxosporea). J Eukaryot Microbiol 53:364–373.

Munday BL, Su XQ, Harshbarger JC. 1998. A survey of product defects in Tasmanian Atlantic salmon (Salmo salar). Aquaculture 169:297–302.

Fontes I, Hallett SL, Mo TA. 2015. Comparative epidemiology of myxozoan diseases, p 317–342. In Okamura B, Gruhl A, Bartholomew JL (ed), Myxozoan evolution, ecology and development. Springer International Publishing, Cham, Switzerland.

Moran JDW, Margolis L, Webster JM, Kent ML. 1999. Development of Kudoa thyrsites (Myxozoa: Myxosporea) in netpen-reared Atlantic salmon determined by light microscopy and a polymerase chain reaction test. Dis Aquat Organ 37:185–193.

Moran JDW, Whitaker DJ, Kent ML. 1999. Natural and laboratory transmission of the marine myxozoan parasite Kudoa thyrsites to Atlantic salmon. J Aquat Anim Health 11:110–115.

St-Hilaire S, Ribble C, Whitaker DJ, Kent M. 1998. Prevalence of Kudoa thyrsites in sexually mature and immature pen-reared Atlantic Salmon (Salmo salar) in British Columbia, Canada. Aquaculture 162:69–77.

Jones SRM, Forster I, Liao X, Ikonomou MG. 2012. Dietary nicarbazin reduces prevalence and severity of Kudoa thyrsites (Myxosporea: Multivalvulida) in Atlantic salmon Salmo salar post-smolts. Aquaculture 342-343:1–6.

Athanassopoulou F, Karagouni E, Dotsika E, Ragias V, Tavla J, Christofilloyanis P, Vatsos I. 2004. Efficacy and toxicity of orally administrated anti-coccidial drugs for innovative treatments of Myxobolus sp. infection in Puntazzo puntazzo. Dis Aquat Organ 62:217–226.

Gómez D, Bartholomew J, Sunyer JO. 2014. Biology and mucosal immunity to myxozoans. Dev Comp Immunol 43:243–256.

Alvarez-Pellitero P. 2008. Fish immunity and parasite infections: from innate immunity to immunoprophylactic prospects. Vet Immunol Immunopathol 126:171–198.

Kabata Z, Whitaker DJ, Bagshaw JW. 1986. Kudoa thyrsites (Gilchrist) (Myxosporea: Multivalvulida) in Coho salmon, Oncorhynchus kisutch (Walbum). Can J Zool 64:1038–1040.

Kent ML, Andree KB, Bartholomew JL, El-Matbouli M, Desser SS, Devlin RH, Feist SW, Hedrick RP, Hoffmann RW, Khattra J, Hallett SL, Lester RJ, Longshaw M, Palenzeula O, Siddall ME, Xiao C. 2001. Recent advances in our knowledge of the myxozoa. J Eukaryot Microbiol 48:395–413.

Kabata Z, Whitaker DJ. 1989. Kudoa thyrsites (Gilchrist, 1924) (Myxozoa) in the cardiac muscle of Pacific salmon (Oncorhynchus spp.) and steelhead trout (Salmo gairdneri). Can J Zool 67:341–342.

Sitjà-Bobadilla A, Estensoro I, Pérez-Sánchez J. 2016. Immunity to gastrointestinal microparasites of fish. Dev Comp Immunol 64:187–201.

Grabner DS, El-Matbouli M. 2009. Comparison of the susceptibility of brown trout (Salmo trutta) and four rainbow trout (Oncorhynchus mykiss) strains to the myxozoan Tetracapsuloides bryosalmonae, the causative agent of proliferative kidney disease (PKD). Vet Parasitol 165:200–206.

Molnár K. 2002. Site preference of myxosporean spp. on the fins of some Hungarian fish species. Dis Aquat Organ 52:123–128.

Sitjà-Bobadilla A, Redondo MJ, Bermúdez R, Palenzuela O, Ferreiro I, Riaza A, Quiroga I, Nieto JM, Alvarez-Pellitero P. 2006. Innate and adaptive immune responses of turbot, Scophthalmus maximus (L.), following experimental infection with Enteromyxum scophthalmi (Myxosporea: Myxozoa). Fish Shellfish Immunol 21:485–500.

Cuesta A, Muñoz P, Rodríguez A, Salinas I, Sitjà-Bobadilla A, Alvarez-Pellitero P, Esteban MA, Meseguer J. 2006. Gilthead seabream (Sparus aurata L.) innate defence against the parasite Enteromyxum leei (Myxozoa). Parasitology 132:95–104.

Davey GC, Calduch-Giner JA, Houeix B, Talbot A, Sitjà-Bobadilla A, Prunet P, Pérez-Sánchez J, Cairns MT. 2011. Molecular profiling of the gilthead sea bream (Sparus aurata L.) response to chronic exposure to the myxosporean parasite Enteromyxum leei. Mol Immunol 48:2102–2112.

Zhang Y-A, Salinas I, Li J, Parra D, Bjork S, Xu Z, LaPatra SE, Bartholomew J, Sunyer JO. 2010. IgT, a primitive immunoglobulin class specialized in mucosal immunity. Nat Immunol 11:827–835.

Langdon JS. 1991. Myoliquefaction post-mortem (“milky flesh”) due to Kudoa thyrsites (Gilchrist) (Myxosporea: Multivalvulida) in mahi mahi, Coryphaena hippurus L. J Fish Dis 14:45–54.

Morado JF, Sparks A. 1986. Observations on the host-parasite relations of the Pacific whiting, Merluccius productus (Ayres), and two myxosporean parasites, Kudoa thyrsitis (Gilchrist, 1924) and K. paniformis Kabata & Whitaker, 1981. J Fish Dis 9:445–455.

Moran JDW, Kent ML. 1999. Kudoa thyrsites (Myxozoa: Myxosporea) infections in pen-reared Atlantic salmon in the northeast Pacific Ocean with a survey of potential nonsalmonid reservoir hosts. J Aquat Anim Health 11:101–109.

Jones SRM, Cho S, Nguyen J, Mahony A. 2016. Acquired resistance to Kudoa thyrsites in Atlantic salmon Salmo salar following recovery from a primary infection with the parasite. Aquaculture 451:457–462.

Hetland DL, Jørgensen SM, Skjødt K, Dale OB, Falk K, Xu C, Mikalsen AB, Grimholt U, Gjøen T, Press CM. 2010. In situ localisation of major histocompatibility complex class I and class II and CD8 positive cells in infectious salmon anaemia virus (ISAV)-infected Atlantic salmon. Fish Shellfish Immunol 28:30–39.

Marshall WL, Sitjà-Bobadilla A, Brown HM, MacWilliam T, Richmond Z, Lamson H, Morrison DB, Afonso LOB. 2016. Long-term epidemiological survey of Kudoa thyrsites (Myxozoa) in Atlantic salmon (Salmo salar L.) from commercial aquaculture farms. J Fish Dis 39:929–946.

Nestle FO, Di Meglio P, Qin J-Z, Nickoloff BJ. 2009. Skin immune sentinels in health and disease. Nat Rev Immunol 9:679–691.

Hellman P, Eriksson H. 2007. Early activation markers of human peripheral dendritic cells. Hum Immunol 68:324–333.

Bassity E, Clark TG. 2012. Functional identification of dendritic cells in the teleost model, rainbow trout (Oncorhynchus mykiss). PLoS One 7:e33196.

Shao T, Zhu L, Nie L, Shi W, Dong W, Xiang L. 2015. Characterization of surface phenotypic molecules of teleost dendritic cells. Dev Comp Immunol 49:38–43.

Haugland GT, Jordal AEO, Wergeland HI. 2012. Characterization of small, mononuclear blood cells from salmon having high phagocytic capacity and ability to differentiate into dendritic like cells. PLoS One 7:e49260.

Granja AG, Leal E, Pignatelli J, Castro R, Abós B, Kato G, Fischer U, Tafalla C. 2015. Identification of teleost skin CD8α+ dendritic-like cells, representing a potential common ancestor for mammalian cross-presenting dendritic cells. J Immunol 195:1825–1837.

Banchereau J, Steinman RM. 1998. Dendritic cells and the control of immunity. Nature 392:245–252.

Nakanishi T, Fischer U, Dijkstra JM, Hasegawa S, Somamoto T, Okamoto N, Ototake M. 2002. Cytotoxic T cell function in fish. Dev Comp Immunol 26:131–139.

Li J, Barreda DR, Zhang Y-A, Boshra H, Gelman AE, Lapatra S, Tort L, Sunyer JO. 2006. B lymphocytes from early vertebrates have potent phagocytic and microbicidal abilities. Nat Immunol 7:1116–1124.

Reite OB, Evensen Ø. 2006. Inflammatory cells of teleostean fish: a review focusing on mast cells/eosinophilic granule cells and rodlet cells. Fish Shellfish Immunol 20:192–208.

Fischer U, Koppang EO, Nakanishi T. 2013. Teleost T and NK cell immunity. Fish Shellfish Immunol 35:197–206.

Whyte SK. 2007. The innate immune response of finfish–a review of current knowledge. Fish Shellfish Immunol 23:1127–1151.

Bermúdez R, Vigliano F, Marcaccini A, Sitjà-Bobadilla A, Quiroga MI, Nieto JM. 2006. Response of Ig-positive cells to Enteromyxum scophthalmi (Myxozoa) experimental infection in turbot, Scophthalmus maximus (L): a histopathological and immunohistochemical study. Fish Shellfish Immunol 21:501–512.

Estensoro I, Calduch-Giner JA, Kaushik S, Pérez-Sánchez J, Sitjà-Bobadilla A. 2012. Modulation of the IgM gene expression and IgM immunoreactive cell distribution by the nutritional background in gilthead sea bream (Sparus aurata) challenged with Enteromyxum leei (Myxozoa). Fish Shellfish Immunol 33:401–410.

Bjork SJ, Zhang Y-A, Hurst CN, Alonso-Naveiro ME, Alexander JD, Sunyer JO, Bartholomew JL. 2014. Defenses of susceptible and resistant Chinook salmon (Onchorhynchus tshawytscha) against the myxozoan parasite Ceratomyxa shasta. Fish Shellfish Immunol 37:87–95.

Trinchieri G. 1995. Interleukin-12: a proinflammatory cytokine with immuneregulatory functions that bridge innate resistance and antigen-specific adaptive immunity. Annu Rev Immunol 13:251–176.

Trinchieri G. 2003. Interleukin-12 and the regulation of innate resistance and adaptive immunity. Nat Rev Immunol 3:133–146.

Tosh KW, Mittereder L, Bonne-Annee S, Hieny S, Nutman TB, Singer SM, Sher A, Jankovic D. 2016. The IL-12 response of primary human dendritic cells and monocytes to Toxoplasma gondii is stimulated by phagocytosis of live parasite rather than host cell invasion. J Immunol 196:345–356.

Gazzinelli RT, Hieny S, Wynn T, Wolf S, Sher A. 1993. Interleukin 12 is required for the T-lymphocyte-independent induction of interferon gamma by an intracellular parasite and induces resistance in T-cell-deficient hosts. Immunology 90:6115–6119.

Kobayashi BYM, Fitz L, Ryan M, Hewick RM, Clark SC, Chan S, Loudon R, Sherman F, Perussia B, Trinchieri G. 1989. Identification and purification of natural killer cell stimulatory factor (NKSF), a cytokine with multiple biologic effects on human lymphocytes. J Exp Med 170:827–845.

Secombes CJ, Wang T, Hong S, Peddie S, Crampe M, Laing KJ, Cunningham C, Zou J. 2001. Cytokines and innate immunity of fish. Dev Comp Immunol 25:713–723.

Rosenzweig SD, Holland SM. 2005. Defects in the interferon-gamma and interleukin-12 pathways. Immunol Rev 203:38–47.

Aliberti JCS, Cardoso MAG, Martins GA, Gazzinelli RT, Vieira LQ, Silva JS. 1996. Interleukin-12 mediates resistance to Trypanosoma cruzi in mice and is produced by murine macrophages in response to live trypomastigotes. Infect Immun 64:1961–1967.

Wynn T, Eltoum I, Oswald IP, Cheever W, Sher A. 1994. Endogenous interleukin 12 (IL-12) regulates granuloma formation induced by eggs of Schistosoma mansoni and exogenous IL-12 both inhibits and prophylactically immunizes against egg pathology. J Exp Med 179:1551–1561.

Belz GT, Shortman K, Bevan MJ, Heath WR. 2005. CD8+ dendritic cells selectively present MHC class I-restricted noncytolytic viral and intracellular bacterial antigens in vivo. J Immunol 175:196–200.

Yrlid U, Wick MJ. 2002. Antigen presentation capacity and cytokine production by murine splenic dendritic cell subsets upon Salmonella encounter. J Immunol 169:108–116.

Lundie RJ, de Koning-Ward TF, Davey GM, Nie CQ, Hansen DS, Lau LS, Mintern JD, Belz GT, Schofield L, Carbone FR, Villadangos JA, Crabb BS, Heath WR. 2008. Blood-stage Plasmodium infection induces CD8+ T lymphocytes to parasite-expressed antigens, largely regulated by CD8alpha+ dendritic cells. Proc Natl Acad Sci U S A 105:14509–14514.

Lanzavecchia A, Sallusto F. 2000. Dynamics of T lymphocyte responses: intermediates, effectors, and memory cells. Science 290:92–97.

Sallusto F, Geginat J, Lanzavecchia A. 2004. Central memory and effector memory T cell subsets: function, generation, and maintenance. Annu Rev Immunol 22:745–763.

Martin MD, Kim MT, Shan Q, Sompallae R, Xue HH, Harty JT, Badovinac VP. 2015. Phenotypic and functional alterations in circulating memory CD8 T cells with time after primary infection. PLoS Pathog 11:1–37.

Kumar S, Tarleton RL. 1998. The relative contribution of antibody production and CD8+ T cell function to immune control of Trypanosoma cruzi. Parasite Immunol 20:207–216.

Pereiro P, Varela M, Diaz-Rosales P, Romero A, Dios S, Figueras A, Novoa B. 2015. Zebrafish Nk-lysins: first insights about their cellular and functional diversification. Dev Comp Immunol 51:148–159.

Hirono I, Kondo H, Koyama T, Arma NR, Hwang JY, Nozaki R, Midorikawa N, Aoki T. 2007. Characterization of Japanese flounder (Paralichthys olivaceus) NK-lysin, an antimicrobial peptide. Fish Shellfish Immunol 22:567–575.

Wang Q, Bao B, Wang Y, Peatman E, Liu Z. 2006. Characterization of a NK-lysin antimicrobial peptide gene from channel catfish. Fish Shellfish Immunol 20:419–426.

Toda H, Araki K, Moritomo T, Nakanishi T. 2011. Perforin-dependent cytotoxic mechanism in killing by CD8 positive T cells in ginbuna crucian carp, Carassius auratus langsdorfii. Dev Comp Immunol 35:88–93.

Praveen K, Leary JH, Evans DL, Jaso-Friedmann L. 2006. Nonspecific cytotoxic cells of teleosts are armed with multiple granzymes and other components of the granule exocytosis pathway. Mol Immunol 43:1152–1162.

Takizawa F, Dijkstra JM, Kotterba P, Korytář T, Kock H, Köllner B, Jaureguiberry B, Nakanishi T, Fischer U. 2011. The expression of CD8alpha discriminates distinct T cell subsets in teleost fish. Dev Comp Immunol 35:752–763.

Gomez D, Sunyer JO, Salinas I. 2013. The mucosal immune system of fish: the evolution of tolerating commensals while fighting pathogens. Fish Shellfish Immunol 35:1729–1739.

Xu Z, Parra D, Gómez D, Salinas I, Zhang Y-A, Jørgensen LVG, Heinecke RD, Buchmann K, LaPatra S, Sunyer JO. 2013. Teleost skin, an ancient mucosal surface that elicits gut-like immune responses. Proc Natl Acad Sci U S A 110:13097–13102.

Holzer A, Sommerville C, Wootten R. 2003. Tracing the route of Sphaerospora truttae from the entry locus to the target organ of the host, Salmon salar L, using an optimized and specific in situ hybridization technique. J Fish Dis 26:647–655.

Bjork SJ, Bartholomew JL. 2009. Effects of Ceratomyxa shasta dose on a susceptible strain of rainbow trout and comparatively resistant Chinook and Coho salmon. Dis Aquat Organ 86:29–37.

Young CA, Jones SRM. 2005. Epitopes associated with mature spores not recognized on Kudoa thyrsites from recently infected Atlantic salmon smolts. Dis Aquat Organ 63:267–271.

Figdor CG, Kooyk Van Y, Adema GJ. 2002. C-type lectin receptors on dendritic cells and langerhans cells. Nat Rev Immunol 2:77–84.

Kim JH, Ogawa K, Wakabayashi H. 1999. Lectin-reactive components of the microsporidian Glugea plecoglossi and their relation to spore phagocytosis by head kidney macrophages of ayu Plecoglossus altivelis. Dis Aquat Organ 39:59–63.

Xu D, Klesius PH, Shoemaker CA. 2001. Effect of lectins on the invasion of Ichthyophthirius theront to channel catfish tissue. Dis Aquat Organ 45:115–120.

Buchmann K, Sigh J, Nielsen C, Dalgaard M. 2001. Host responses against the fish parasitizing ciliate Ichthyophthirius multifiliis. Vet Parasitol 100:105–116.

Tsutsui S, Tasumi S, Suetake H, Suzuki Y. 2003. Lectins homologous to those of monocotyledonous plants in the skin mucus and intestine of pufferfish, Fugu rubripes. J Biol Chem 278:20882–20889.

Woo PTK. 2003. Cryptobia (Trypanoplasma) salmositica and salmonid cryptobiosis. J Fish Dis 26:627–646.

Morris DJ, Adams A. 2004. Localization of carbohydrate terminals on Tetracapsuloides bryosalmonae using lectin histochemistry and immunogold electron microscopy. J Fish Dis 27:37–45.

Knaus M, El-Matbouli M. 2005. Characterisation of carbohydrate-binding sites in developmental stages of Myxobolus cerebralis. Parasitol Res 97:505–514.

Redondo MJ, Cortadellas N, Palenzuela O, Alvarez-Pellitero P. 2008. Detection of carbohydrate terminals in the enteric parasite Enteromyxum scophthalmi (Myxozoa) and possible interactions with its fish host Psetta maxima. Parasitol Res 102:1257–1267.

Pettersen FE, Fyllingen I, Kavlie A, Maaseide NP, Glette J, Endresen C, Wergeland HI. 1995. Monoclonal antibodies reactive with serum IgM and leukocytes from Atlantic salmon (Salmo salar L.). Fish Shellfish Immunol 5:275–287.

Braden LM, Koop BF, Jones SRM. 2015. Signatures of resistance to Lepeophtheirus salmonis include a Th2-type response at the louse-salmon interface. Dev Comp Immunol 48:178–191.

Kohler G, Milstein C. 1975. Continuous cultures of fused cells secreting antibody of predefined specificity. Nature 256:495–497.

Poley JD, Braden LM, Messmer AM, Whyte SK, Koop BF, Fast MD. 2016. Cypermethrin exposure induces metabolic and stress-related gene expression in copepodid salmon lice (Lepeophtheirus salmonis). Comp Biochem Physiol D Genomics Proteomics 20:74–84.

Skugor S, Glover KAA, Nilsen F, Krasnov A. 2008. Local and systemic gene expression responses of Atlantic salmon (Salmo salar L.) to infection with the salmon louse (Lepeophtheirus salmonis). BMC Genomics 9:498.

Tadiso TM, Lie KK, Hordvik I. 2011. Molecular cloning of IgT from Atlantic salmon, and analysis of the relative expression of tau, mu and delta in different tissues. Vet Immunol Immunopathol 139:17–26.

Fast MDD, Muise DMM, Easy REE, Ross NWW, Johnson SCC. 2006. The effects of Lepeophtheirus salmonis infections on the stress response and immunological status of Atlantic salmon (Salmo salar). Fish Shellfish Immunol 21:228–241.

Sutherland BJG, Hanson KC, Jantzen JR, Koop BF, Smith CT. 2014. Divergent immunity and energetic programs in the gills of migratory and resident Oncorhynchus mykiss. Mol Ecol 23:1952–1964.

Olsvik PA, Lie KK, Jordal A-EO, Nilsen TO, Hordvik I. 2005. Evaluation of potential reference genes in real-time RT-PCR studies of Atlantic salmon. BMC Mol Biol 6:21.

Vandesompele J, De Preter K, Pattyn F, Poppe B, Van Roy N, De Paepe A, Speleman F. 2002. Accurate normalization of real-time quantitative RT-PCR data by geometric averaging of multiple internal control genes. Genome Biol 3:RESEARCH0034.