Pathogen-free rainbow trout (90 days-old, mean length 4.02 ± 0.26 cm, mean weight 0.6 ± 0.15 g) were distributed in tanks (n = 3) receiving 15°C water and were fed (1% BW/day) daily with floating trout pellets (Aqua Garant, Austria). The North American strain (TL) rainbow trout with known susceptibility to both myxozoans was used for the exposure experiment (Rucker and El-Matbouli, 2007; El-Matbouli et al., 2009). Prior to the exposure trail, fish (n = 10) were randomly sampled and tested for M. cerebralis and T. bryosalmonae by qPCR (Kotob et al., 2018). The fish were observed for signs of morbidity (uncoordinated swimming and/or lethargy) three times a day for timely removal and euthanizing of such fish where needed.

The fish were initially divided into three groups ( Figure 1 ). The first group (group 1) fish (n = 36) were exposed to M. cerebralis triactinomyxons (TAMs) (2000 TAMs/fish), while fish (n = 36) in the second group (group 2) were exposed to free T. bryosalmonae spores released from 22 mature parasite sacs at 16°C (Kotob et al., 2017). The TAMs and the spores were produced as described in Saleh et al. (2019a) and Kotob et al. (2017). Thirty days after infection, half of the fish from group 1 and group 2 (single infection groups) were reciprocally co-infected with the other parasite, separately [group 3 (n = 18) and group 4 (n = 18)]. The last group (group 5) fish (n = 27) were subjected to mock exposure to specific pathogen free water and used as a negative control. At 1 day and 4 day as well as 31 days post exposure (dpe) (day 1 post co-infection (dpc)), fish (n = 9) from each group were euthanized and sampled. Caudal fins (portal of entry of M. cerebralis) and gills (portal of entry of T. bryosalmonae) were washed with sterile phosphate buffer and stored at -80°C until further analysis. Before, at zero time point and at all time points during the experiment, fish (n = 9) from each group were screened for parasites by microscopic examination of gill and skin scrapings and kidney swabs were taken to assess for bacterial agents by culture on blood agar plates as previously described (Anacker & Ordal, 1959). No infectious agents were detected prior to or during the experiment. In addition, before and at all time points during the experiment, homogenates of brain, entire viscera, spleen and kidney were subjected to viral screenings using BF-2 and EPC cell lines according to standard cell culture methods (Manual of Diagnostic Tests for Aquatic Animals, 2021). No virus was detected before or throughout the experiment. During the in vivo experiments, the parasite load of M. cerebralis and T. bryosalmonae was confirmed by qPCR at all-time points as previously described (Kotob et al., 2018).

Total RNA was extracted from the caudal fin and gill samples at each time point using RNeasy Mini Kit (Qiagen, Hilden, Germany) according to the manufacturer’s instructions. To remove any residual DNA contamination, an on-column DNase digestion step was performed. RNA concentration and quality control were determined using a Nanodrop 2000c spectrophotometer and by agarose gel electrophoresis (Thermo Fisher Scientific, Wilmington, USA). The isolated RNA was used to synthesize cDNA, utilizing iScript cDNA Synthesis Kit (Bio-Rad, Munich, Germany) according to the manufacturer’s instructions. The synthesized cDNA was diluted with nuclease-free water, aliquoted and stored at −20°C until further analysis.

Parasite load determination for M. cerebralis (using the forward primer Myx18-909 F CTTTGACTGAATGTTATTCAGTTACAGCA and the reverse primer Myx18-996 R GCGGTCTGGGCAAATGC) and T. bryosalmonae (using the forward primer RPL18 F GTAAACGGGGACAAAAAGA and the reverse primer RPL18 R GGAGCAGCACCAAAATAC), was performed using samples from caudal fin and gills as described previously (Kelley et al, 2004; Gorgoglione et al., 2013). The PCR reaction of 20 μl final volume contained 4 μl of 1:10-fold diluted cDNA, 1× SsoAdvanced Universal SYBR Green Supermix (Bio-Rad), 0.4 μM of each primer, and DEPC-treated sterile distilled water (Bio-Rad). The PCR reaction consisted of an initial 5 min of cDNA denaturation at 95°C, followed by 35 cycles of 95°C for 30 s, 57–62°C for 30 s and 72°C for 30 s. The detection thresholds for Myx18 and RPL18 were 36.43 and 37.61, respectively. A melting-point curve was measured, starting from 55°C with an increase of 0.5°C at every 10 s up to 95°C, for detecting non-specific binding. The quantity of gene expression level was assessed using CFX96 Touch Real-Time PCR detection system (Bio-Rad, Munich, Germay). Elongation factor 1 alpha was used as a reference gene for normalizing the expression of targeted genes (Kumar et al., 2014). Calculation of relative gene expression was performed using CFX manager software version 3.1 (Bio-Rad, Munich, Germany).

Control and infected fish tissue samples were processed for proteomic analysis. Caudal fin and gill tissues were solubilized using 400 µl pre-cooled denaturing lysis buffer (7M urea, 2M thiourea, 4% CHAPS and 1% DTT) containing mammalian protease inhibitor cocktail (Sigma Aldrich, Vienna, Austria). Sample suspensions were disrupted by sonication. The lysates were incubated overnight at 4°C, vortexed and centrifuged at 18000 x g for 30 min at 4°C and the supernatants collected. The total protein concentration of each lysate was determined colorimetrically with a NanoDrop 2000c spectrophotometer and using the Pierce 660 nm Protein Assay according to the manufacturer’s instructions (Thermo Scientific, Vienna, Austria).

Sample preparation was performed using a Filter-Aided-Sample preparation-Protocol (FASP) according to Wiśniewski et al. (2009) and Wiśniewski et al. (2016) as described in Ramires et al. (2022). In brief, 30 µg of protein were digested on a 10 kDa Pall Nanosep centrifugal device (Cytiva, MA, USA).

After reduction with dithiothreitol and alkylation with iodoacetamide, proteins were digested using Trypsin-LysC-Mix from Promega (WI, USA). In order to stop the trypsin digestion, 5% trifluoroacetic acid was added. Peptides were extracted in three rounds each of 50 µL 50 mM Tris followed by centrifugation. Peptides were acidified using trifluoroacetic acid to a pH below 2. After C18 clean-up with spin columns (Pierce, Thermo Fisher), peptides were separated on a nanoRSLC system equipped with a 25 cm Acclaim PepMap C18 column (Thermo Fisher) and analysed on a high-resolution Q Exactive HF Orbitrap mass spectrometer as described in Gutiérrez et al. (2019).

The database search was performed using the Proteome Discoverer Software 2.4.1.15 (Thermo Fisher Scientific). The protein databases were downloaded from the UniProt homepage (http://www.uniprot.org) for rainbow trout (taxonomy ID 8022), and myxozoa (taxonomy ID 35581). Search settings were as follows: 10 ppm precursor mass tolerance and 0.02 Da fragment mass tolerance; dynamic modifications allowed were oxidation of methionine as well as the N-terminal protein modifications acetylation, methionine loss and the combination of both, and static modification of carbamidomethylation on cysteine. For protein identification, only proteins with at least two identified peptides were reported. Intensity-based label free quantification was used in order to compare protein abundance in the experiments. Protein abundance was extracted from mass spec raw data in Proteome Discoverer software followed by normalization to total area sums. Exported normalized abundance values were filtered in Excel to remove proteins with missing values before subsequent statistical analysis using R programming language according to R Core Team (2021).

To determine differentially expressed proteins in the trout tissue samples, statistical evaluation was performed in the R programming language. The differential expression of proteins was evaluated using one-way analysis of variance (ANOVA) for each protein. To compensate for multiple testing, the method of Benjamini and Hochberg (1995) was used to control the false discovery rate (FDR). The differences were considered significant if FDR-adjusted p values were smaller than the significance level of α = 0.01. For such proteins, the honest significant difference method of Tukey was applied as a post hoc test to assess the significance of the pairwise comparisons. Protein expression was considered differential if the adjusted p-value was below α and the absolute fold change at least two (fold change <−2 or >+2). The mass spectrometry proteomics data have been deposited to the ProteomeXchange (Deutsch et al., 2023) Consortium via the PRIDE (Perez-Riverol et al., 2022) partner repository (Perez-Riverol et al., 2016) with the dataset identifier PXD050412. The biological processes, cellular components and molecular functions of the modulated proteins were assigned. Kyoto Encyclopedia of Genes and Genomes (KEGG) PATHWAY database was employed to predict involved pathways. To determine the protein-protein functional interaction network of the differentially regulated proteins, amino acid sequences were assessed using STRING v11.0 (Szklarczyk et al., 2017). The representation of the protein-protein network was analysed in the database, experiment, text mining, co-expression, neighborhood, gene fusion and co-occurrence.

In the caudal fin and the gill tissues, the relative expression of the 18S rRNA of M. cerebralis and the 60S ribosomal protein L18 of T. bryosalmonae (RPL18) genes is shown in single and co-infections ( Figure 2 ). The highest load of M. cerebralis was observed in the Mc group on day 4. The lowest parasite count was observed in the Mc group, while the Mc+ group had a higher load than the Mc on day 31 (1 dpc) ( Figure 2A ).

The highest load of T. bryosalmonae was observed in the Tb group on day 1. The T. bryosalmonae load decreased over time and the lowest parasite count was observed in the Tb+ group on day 31 (1 dpc) ( Figure 2B ).

The proteomic analysis of caudal fin tissues showed 16 different proteins functioning in signal transduction, immune protection, metabolism and tissue repair were significantly modulated after 1 and 4 dpe to M. cerebralis ( Table 1 ).

Among the differentially regulated proteins in caudal fins, a Ly6 (lymphocyte antigen-6)/uPAR (urokinase-type plasminogen activator receptor) superfamily member was significantly upregulated following exposure to M. cerebralis. Additionally, an SH3-domain-containing protein was extremely reduced at 4 dpe to M. cerebralis.

Clathrin heavy chain protein, a further differentially regulated protein identified in caudal fins, was on the other hand significantly induced at 4 dpe to M. cerebralis.

Of importance, two interferon IFN induced GTP-binding Mx proteins with functions in host immunity and leucocyte activation showed elevated protein abundance, increasing with time from day 1 to day 4 followed by proteins involved in tissue repair at day 4.

Rainbow trout fibulin-1 was highly upregulated in the current investigation at day 1 and day 4 after exposure to M. cerebralis as well as at day 4 in control fish.

The abundance of the laminin EGF-like domain-containing protein was also highly increased in the caudal fin tissues at day 1 and day 4 after exposure to M. cerebralis, as well as at day 4 in control fish.

In this proteome analysis, the abundance of two poly polymerase (PARP) catalytic domain-containing proteins were induced at 4 dpe to M. cerebralis.

Further, in this analysis, the abundance of a slingshot-1 protein was significantly induced at 1 and 4 dpe to M. cerebralis. The slingshot-1 protein showed higher abundance at 4 than at 1 dpe, as compared to the control group. Furthermore, the level of derlin was highly increased at 1 and 4 dpe to M. cerebralis as compared to the control group.

In the current study, the abundance of the anti-oxidative and anti-inflammatory TLDc domain-containing protein was highly induced in the caudal fin tissues at 4 dpe to M. cerebralis as compared to 1 dpe infected fish and control.

In this investigation, the abundance of cytidine monophosphate kinase 2 (CMPK2) was induced in rainbow trout caudal fin tissues at 1 dpe and showed significant upregulation at day 4 as compared to control and 1 dpe to M. cerebralis.

Following exposure to M. cerebralis, the abundance of succinyl-CoA:3-ketoacid-coenzyme A transferase (SCOT) was elevated at day 4 in this study.

Another protein significantly increased, at day 4, in caudal fins after M. cerebralis exposure was the barrier to autointegration factor (BAF). Similarly, the eukaryotic translation initiation factor 2B (EIF2B) was increased at day 4 in infected rainbow trout following exposure to M. cerebralis.

Based on protein domain characterization of the annotated proteins that were significantly modulated in the caudal fin tissues, various molecular functions appear to be regulated, including leucocyte activation, host protection and tissue repair ( Figure 3 ).

After 31 days of single-infection and 1 day of co-infection, caudal fin tissues had 4 proteins that were differentially regulated between the groups ( Table 2 ). The complement factor H protein showed the highest abundance in the co-infection groups compared to single infection groups.

The ras related GTPase protein, which is involved in parasite recognition, showed highest abundance in Tb+ groups (initially infected with T. bryosalmonae then after 30 days subsequently infected with M. cerebralis). On the other hand, a decreased abundance (specifically in Mc+ fish) was observed for a serine/threonine protein phosphatase, which has a role in the regulation of the stress response, perhaps indicating the failure of coping with stress caused by the parasites. Nevertheless, the abundance of the TPR-region domain containing protein was decreased specifically in Tb+ fish.

The pathways and biological processes assigned by the KEGG orthology provided important functional information about the proteomic changes at the portals of entry of M. cerebralis and T. bryosalmonae ( Figure 4 ). In our study, in addition to the identified pathways ribosome, RNA transport and lysosome, the endocytosis pathway was assigned for rainbow trout caudal fin proteins.

In the gills, 27 proteins were differentially modulated post exposure to T. bryosalmonae ( Table 3 ). The proteomic analysis reveals different proteins involved in multiple roles such as parasite recognition, signal transduction and immune protection that were significantly upregulated after 1 day and/or 4 days. Among the upregulated proteins, two mucins, two DNA-(apurinic or apyrimidinic site) lyases, two Sodium/potassium-transporting ATPases, two C2 domain-containing proteins, B-cell receptor (CD22), and an immunoglobulin (Ig) domain containing protein are involved in host protection and immunity ( Figure 3 ). In addition, many proteins involved in parasite recognition (synaptopodin, caldesmon 1b, proteases), signal transduction (protein kinase C, creatine kinase, Ryanodine receptor 1), and metabolism, repair of oxidative damage and oxidative stress (DNA-lyase, serotransferrin, splicing factor, proline-and glutamine-rich) were increased at day 1 and/or day 4 when compared with control fish or sampling time.

At day 31 post T. bryosalmonae single infection and one day post co-infection, eleven proteins including seven host proteins involved in signal transduction and host immunity, were differentially modulated ( Table 4 ). This included integrin beta which is involved in parasite recognition and signal transduction, which exhibited increased abundance (≥ 100) in the co-infection groups compared to the Mc group. The MHC class II antigen alpha chain showed elevated abundance (≥ 100) in the Tb+ group compared to Mc+ and the single infection groups Tb and Mc. Conversely, the abundance of the Ig-like domain-containing protein was extremely decreased (≤ 100) in the Tb+ group compared to Mc+ and the single infection groups Tb and Mc. Other proteins which showed down regulation (≤ 100) in the co-infection groups were EF-1_beta_acid domain-containing protein and the apoptotic chromatin condensation inducer 1. Fibulin 7 showed elevated abundance (≥ 100) in the Mc+ but reduction in the Tb+ group. The abundance of the all-trans-retinol 13,14-reductase was elevated (≥ 100) in the Mc+ group and Tb+ as compared to control group.

In addition to host proteins, four myxozoan proteins were significantly upregulated (≥ 100) in gills in the co-infection Mc+ group, and included two hsp70 proteins, calmodulin and a histone ( Table 5 ). The differentially regulated parasite proteins are likely involved in parasite development and virulence ( Figure 5 ).

Among the identified pathways in the gills; ribosome, mRNA surveillance pathway (to ensure quality of mRNA needed for correct translation), RNA transport (essential for gene expression), and focal adhesion are present ( Figure 5 ). The protein-protein interaction network of gill proteins shows that the parasite proteins (HSP70, calmodulin and histone) are connected with the host proteins ( Figure 6 ).

The highest load of M. cerebralis was observed on 4 dpe, indicating the suppressive effect of M. cerebralis. During the course of the single infection with M. cerebralis, the spores successfully parasitize, multiply in the caudal fin and suppress the host immune responses. The parasite load decreases considerably on 31 dpe (1 dpc), as the parasites leave the infection site to reach the target organ (head cartilage) (Saleh et al., 2019a). The increased M. cerebralis load in the caudal fin indicates that the fish are not able to maintain proper tissue hemostasis, immunity or limit parasite burden during WD as previously reported (Severin and El-Matbouli, 2007; Severin et al., 2010; Baerwald, 2013; Kotob et al., 2018; Saleh et al., 2020a; 2020b). On the other hand, the T. bryosalmonae burden decreased over time showing and the lowest parasite count was observed in the Tb+ group on 1 dpc. The reduced load in the gills implies that the fish are aimed at maintaining proper tissue hemostasis, immunity and limit parasite burden (Gorgoglione et al., 2013; Kotob et al., 2018; Bailey et al., 2020).

Ly6/uPAR domain-containing protein was one of the significantly increased proteins in caudal fin in the Mc infected fish. It is a known biomarker of inflammation, present in most metazoans and plays important roles in different biological processes including host immunity, cellular adhesion, and cell signalling (Guo et al., 2015). The increased level of this protein at day 4 in the Mc infected fish indicates an involvement in the immune response to M. cerebralis and in signal transduction during WD. Indeed, Ly6/uPAR is known to be involved in disease pathology in severe malaria, viral infection and virus-host interactions, and can enhance virus infection by targeting viral entry (Plewes et al, 2014; Mar et al., 2018; Yu et al., 2019). Similarly, Ly6/uPAR might be involved in M. cerebralis infection and fish-myxozoan interactions. During WD, uncontrolled inflammation augments tissue damage triggered by M. cerebralis, giving a possible means for Ly6/uPAR to enhance parasite entry and increase parasitic burden (Saleh et al., 2019a; 2020a).

Another interesting finding was in relation to the SH3 domain containing glutathione-S-transferase (GST)-fusion protein that strongly inhibits the formation of endocytic clathrin-coated vesicle (CCV) in permeabilized cells (Simpson et al., 1999). Clathrin-mediated endocytosis (CME) is a major route of entry into eukaryotic cells. In fact, exceptionally fast CME turnover aids Trypanosoma brucei evasion of the host immune system (Manna et al., 2015). At day 4 post exposure to M. cerebralis, the reduced abundance of the SH3-domain-containing protein points to a non-restricted increase of endocytic CCV formation. This likely occurs as the abundance of the clathrin heavy chain protein increases, as seen at day 4 in the Mc infected fish. Taken together, this seems to be part of the parasite invasion strategy through endocytic CCV formation. Apparently, M. cerebralis uses the CME to facilitate its entry into the host and evade the fish immune system. The endocytosis pathway was also reported to have an important role for T. bryosalmonae nutrition acquisition and virulence (Morris and Adams, 2008), recently confirmed by transcriptomic analysis of PKD infected kidney tissue (Ahmad et al., 2021; Faber et al., 2021). After exposure to M. cerebralis, within minutes the sporoplasm (contains 64 internal cells) migrates to deeper layers of the epidermis, then within hours of initial infection, aggregates and single cells from the sporoplasm begin mitotic replication alternating between inter and intracellular locations, which is a critical phase in the early stages of M. cerebralis infection, invasion and replication in trout (El-Matbouli et al., 1995; Baerwald et al., 2008). This pathway is used as an evasion strategy by infectious hematopoietic necrosis virus (IHNV) in rainbow trout (Liu et al., 2011). The CME pathway is also used by Piscirickettsia salmonis for host entry (Ramírez et al., 2015), which suggests that the CME mechanisms are involved in the transport of large particles and pathogens in fish. Hence, this pathway might be also used by M. cerebralis as an evasion strategy for speedy parasite uptake. However, functional assays are needed to prove whether M. cerebralis uses this pathway as an evasion strategy.

Among the differentially regulated proteins identified in common carp skin mucus after exposure to the deleterious parasite Ichthyophthirius multifiliis, three interferon (IFN)-induced GTP-binding Mx-like proteins were upregulated (Saleh et al., 2019b). These proteins are implicated in complement activation, which leads to opsonization, leucocyte stimulation, and direct pathogen killing (Das et al., 2019). These proteins are released to prevent pathogen attack and regulate cellular activities such as endocytosis and trafficking of nucleoproteins into the nucleus (Haller and Kochs, 2002). They also play an important role in parasite recognition and signal transduction (Timmerhaus et al., 2011; Dawar et al., 2016; Saleh et al., 2019b). In fish, the Mx gene(s) is typically induced by Type-I IFN, the main cytokine mediating the innate immune function (Das et al., 2019). However, some salmonid Mx genes induced by IFN-γ (Type-II IFN) have been reported recently (Robertsen et al., 2019; Wang et al., 2019). Since the expression of trout IFN-γ is known to be increased after exposure to M. cerebralis (Baerwald et al., 2008; Saleh et al., 2020b), this suggests a link to the current findings where the two IFN-induced GTP-binding Mx proteins may have been activated by IFN-γ-mediated innate immunity. The abundance of the two GTP-binding Mx proteins was differentially increased for one protein at 4 dpe and for the second at 1 and 4 dpe. The results suggest their involvement in diverse activities including parasite recognition, signal transduction and host protection during infection by M. cerebralis.

Fibulin-1 and laminin were also highly increased post-infection. Fibulin-1 has a key role in remodelling the extracellular matrix (ECM) of the fin and is a main constituent of the ECM. It can mediate cell signal transduction by binding to ECM proteins such as laminin and fibronectin (Martin et al., 1988). Zebrafish fibulin-1 was also found to be expressed in fins as well as numerous other tissues (Feitosa et al., 2012). The increased abundance of fibulin-1 in M. cerebralis infected rainbow trout (as well as in unexposed fish) points to its involvement in fin development as well as in parasite recognition and signal transduction.

The EGF-like domains in laminin and in other ECM proteins, are indicators for cellular growth and differentiation. These domains stimulate adjacent cells in a specific and precise manner (Engel, 1989). The upregulation of laminin in a similar manner as fibulin-1 points to an association of these two proteins in the rainbow trout ECM and their involvement in cellular development, growth promotion, tissue repair and signal transduction.

PARP was also increased in single infected caudal fins, although not to the same extent as fibulin-1 and laminin. PARPs are involved in chromatin regulation, transcription, RNA biology, and DNA repair. PARP, as a prominent marker of apoptosis was upregulated in rainbow trout, revealing apoptosis induction under endoplasmic reticulum stress (Séité et al., 2019). PARP acts on both host and viral proteins through binding to deltex E3 ubiquitin ligase 3L (DTX3L) to mediate efficient immune responses to pathogens (Gupte et al., 2017). The catalytically inactive PARP-9 associates with DTX3L to recruit immune responses to viral pathogens (Zhang et al., 2015). This binding stimulates E3 ubiquitin ligase activity and promotes IFN-stimulated gene expression. During WD, ubiquitin like protein 1 was upregulated by more than 100-fold and IFN regulating factor 1 (IRF 1) was upregulated by more than 15 fold after exposure to M. cerebralis (Baerwald et al., 2008). In the current study, the rainbow trout 40S protein S30, which belongs to the ribosomal ubiquitin like protein eS30 fusion protein (UBIM) was predicted to be a functional partner and is the central node of the protein–protein interaction network of caudal fin proteins ( Figure 4 ). IFN activates its downstream transcription factor STAT1 and successive binding of STAT1 to genomic loci to stimulate proinflammatory gene expression. In WD, several studies described IFN as a common link in the response to M. cerebralis (Baerwald et al., 2008; Saleh et al., 2020b). In fact, the expression of many proinflammatory genes was upregulated in response to M. cerebralis (Rucker and El-Matbouli, 2007; Severin and El-Matbouli, 2007, Severin et al., 2010; Baerwald, 2013; Saleh et al., 2020a; 2020b). The increased expression of STAT1 and several proinflammatory molecules including IFNγ, IRF 1, iNOS, Nramp α, Nramp β, IL-1β, STAT3 and IL-17 during WD suggests that PARP is also involved in the activation of the fish immune response against M. cerebralis.

In this study, the slingshot protein of rainbow trout was apparently increased in the Mc group at 1 and 4 dpe, presumably to retain the cytoskeleton structure and cope with tissue disruption due to parasite proliferation. In Drosophila, the slingshot protein family was shown to play a pivotal role in actin dynamics by reactivating cofilin in vivo (Niwa et al., 2002). Derlin is a component in the endoplasmic reticulum degradation (ERAD) pathway and is required for the dislocation and degradation of certain misfolded proteins from the ER to the cytosol (Lilley and Ploegh, 2004). The increased abundance of derlin heavy chain after exposure to M. cerebralis indicates a possible involvement in degradation of misfolded proteins due to stress caused by parasite proliferation during WD. In zebrafish, the TLDc domain-containing protein (oxidative resistance gene 1) is crucial for defence against oxidative stress (Xu et al., 2020). TLDc domain-containing proteins were shown recently to be modulated in the fish skin proteome after parasitic infection (Saleh et al., 2019b). The increased abundance of the TLDc domain-containing protein here indicates that the host attempts to reduce oxidative stress caused by M. cerebralis as a means to protect rainbow trout against infection.

Carp cytidine monophosphate kinase 2 (CMPK2) is upregulated by bacterial challenge and has a protective effect on the gut barrier, suggesting that teleost CMPK2 is involved in innate immune responses against bacterial and viral infections (Liu et al., 2019; Feng et al., 2021). CMPK2 was also upregulated in Atlantic salmon in response to infection with the monogenean parasite Gyrodactylus salaris and after injection with the immunostimulants LPS and Poly (I:C) (Collins et al., 2007). So far these are the only sources available that signify its potential antiparasitic and antibacterial function in fish. The increased abundance of this protein in M. cerebralis infected fish indicates that CMPK2 might have a similar protective function in rainbow trout immunity against parasites as in Atlantic salmon. The succinyl-CoA:3-ketoacid-coenzyme A transferase (SCOT) catalyzes the conversion of CoA from succinyl-CoA to acetoacetate to produce acetoacetyl-CoA and succinate. Acetoacetyl-CoA is further converted into acetyl-CoA, which passes in the citric acid cycle to either deliver energy or be stored as a fatty acid (Zhang et al., 2013). The increased SCOT level at day 4 in this study indicates that parasite propagation [highest parasite load was observed at day 4 post exposure to M. cerebralis ( Figure 2A )] likely influences metabolism in the fish host.

The barrier to autointegration factor (BAF) is known to be used by viruses to improve infectivity and protein biosynthesis. In fact, BAF expression is increased in Atlantic salmon following exposure to infectious salmon anemia virus (Workenhe et al., 2009). In addition, Atlantic salmon BAF is upregulated in response to IHNV, which is thought to facilitate viral protein transcription within the host DNA (Miller et al., 2007). So far these are the only sources available that signify its potential antimicrobial function in fish. BAF is likely involved in the fish response to myxozoan parasites and the increased abundance at day 4 suggests that M. cerebralis uses BAF to increase infectivity and propagation in rainbow trout similar to IHNV in Atlantic salmon. The abundance of the eukaryotic translation initiation factor 2B (EIF2B) is involved in chromatin organization and represses gene expression in the early response of Atlantic salmon to infection (Krasnov et al., 2012). This protein was downregulated in rainbow trout following infection with Aeromonas salmonicida, which likely decreases translation generally but at the same time can increase the translation of stress response mRNAs (Causey et al., 2018). The increased abundance of EIF2B at day1 and day 4 might suggest its involvement in the rainbow trout early immune response against M. cerebralis.

After coinfection, the relatively low number of the differentially regulated protein identified was actually anticipated and can be attributed to the waning of the initial response of the single infection as the parasites leave the skin and gills after 4 day and subsequently most of the preliminary local responses are no longer present at day 31.

Complement factor H is a multidomain, multifunctional protein, which can bind to a range of ligands and protect host cells against damage mediated by disturbance of complement system (Wei et al., 2022). This inhibitory regulator of the alternative pathway of the complement system prevents the synthesis of C3 convertase to suppress complement activation (Du Clos and Mold, 2008). In Nile tilapia, the complement factor H protein has a robust binding capacity to pathogenic bacteria and is involved in the maintenance of homeostasis within the host and pathogen (Wei et al., 2022). The increased abundance in the co-infection groups indicates that the complement factor H protein likely contributes to tissue haemostasis and the immune response of rainbow trout against myxozoan co-infections. On the other hand, the down regulation of serine/threonine protein phosphatase could be a part of the immune evasion strategy of T. bryosalmonae to suppress host immunity and support subsequent co-infections. TPR-region domain containing protein, which controls protein organization and haemostasis was differentially modulated in the co-infections group. Indeed, during T. bryosalmonae infection a general immunosuppression of the fish host is observed as a parasitic strategy to evade the immune system like that of protozoan parasites (Bailey et al., 2020).The abundance of the TPR-region domain containing protein was extremely reduced in the Tb+ group, although the parasite load was decreased compared to the other groups, signifying that the primary suppressive action after T. bryosalmonae exposure was endorsed at 31 dpe (1 dpc) apparently to limit inflammatory responses, tissue destruction and subsequent invasion with M. cerebralis (seen by the decreased parasitic load) and to maintain tissue haemostasis properly.

Several proteins were differentially regulated in the gills after exposure to T. bryosalmonae. The host protection and immunity proteins serotransferrin, mucin, B-cell receptor (CD22), and Ig domain containing protein were increased at day 1 and/or day 4. The highest load of T. bryosalmonae observed in the Tb group at day 1, was decreased at day 4 and the lowest parasite count was observed in the Tb+ group at day 31 (day 1 after co-infection) ( Figure 2B ).

Indeed, previous studies have shown that M. cerebralis and T. bryosalmonae can induce B- and T-cell associated molecules (Saleh et al., 2019a; Bailey et al., 2020). Moreover, several proteins including synaptopodin, caldesmon 1b and proteases likely involved in parasite recognition were differentially modulated. Other proteins such as protein kinase C, creatine kinase and ryanodine receptor 1 may have roles in signal transduction. Furthermore, proteins involved in transport, metabolism, repair of oxidative damage and oxidative stress including sodium/potassium-transporting ATPase subunit alpha and beta, DNA-lyase and splicing factor (proline-and glutamine-rich) were differentially abundant at day 1 and/or day 4. Indeed, multiple proteins with similar functions were previously reported modulated in response to parasitie infections in fish (Saleh et al., 2019b). In the current study, the abundance of the proteins involved in host immunity and protection was mostly induced at day 1 compared to control fish, whereas they were reduced at 4 compared to 1 dpe to T. bryosalmonae. On the other hand, the abundance of the proteins involved in metabolism and repair of oxidative damage was increased at day 1 compared to control group and reduced at 4 compared to 1 dpe, indicating that rainbow trout is able to maintain proper haemostasis, limit inflammatory reactions and provide adequate immune responses against T. bryosalmonae.

Integrin was identified among the differentially regulated proteins after co-infection in the gills. In mammals, it is involved in cellular processes such as focal adhesion, signal transduction, interactions, cell migration, growth and differentiation, cell-cell and cell-pathogens (Yamada and Danen, 2000). Integrins are cell receptors that can bind the protein components of viral proteins, and play crucial roles in fish immune responses to viral infection by inducing Type I IFN (Xiong et al., 2022). This protein had highest abundance in the co-infection groups compared to the M. cerebralis single infection fish, which may indicate that T. bryosalmonae induces rainbow trout integrin to limit a secondary infection from other myxozoans (in this case to limit M. cerebralis). A further protein showing elevated abundance in the co-infection groups, as compared to control fish was the all-trans-retinol 13,14-reductase, also known as retinol saturase. In mammals, it is implicated in glucose and lipid metabolism, macrophage function, vision, and the generation of reactive oxygen species (Weber et al., 2020). In the current study, increased abundance in Mc+ and Tb+ fish suggests its involvement in the host stress response to myxozoan co-infections.

Focal adhesion was identified among multiple pathways by the protein-protein interaction network analysis of gill proteins in this study. The focal adhesion pathway acts as a signal sensor to mediate signal transduction and the immune response to stimuli. Indeed, a previous study has shown that parasitic infection (I. multifiliis) stimulated the focal adhesion pathways in rainbow trout (Roh et al., 2021).

Lastly, the upregulation of B-cell markers (MHC II, B-cell receptor CD22, Ig-domain containing protein) was also seen in the gills, mainly in the Tb single and/or co-infection groups. This may link to previous studies that have shown that M. cerebralis and T. bryosalmonae can induce B- and T-cells (CD8+ and CD4+ cells) (Saleh et al., 2019a; Bailey et al., 2020), potentially here as part of the local mucosal response at this site. Indeed, the importance of IgT⁺ B cell responses in trout gills infected with other types of parasites (I. multifiliis) is well known (Xu et al., 2016).

In addition to the modulated host proteins, four myxozoan proteins were also found to be significantly increased in Mc+ fish gills. These included two hsp70, calmodulin and histone. These myxozoan proteins apparently have a role in parasite virulence and growth. Specifically, hsp70 has a role in cell structure, cell regulation, protein assembly, folding, and translocation during parasitic infections in fish (Abernathy et al., 2011; Saleh et al., 2021). In a recent transcriptome study, Hsp60, 70, and 90 T. bryosalmonae homologues were found to be abundantly expressed in bryozoa and fish suggesting potential contribution to parasite survival, temperature-driven development and host invasion (Faber et al., 2021; Shivam et al., 2023). Interestingly, as a vaccine adjuvant, hsp70 gives high protection against the fish parasite Cryptocaryon irritans in orange spotted grouper (Josepriya et al., 2015). Hence, it is worth to functionally investigate whether and how myxozoan hsp70 could exert such protection in fish. The M. cerebralis histone was found connected to host proteins in the protein–protein interaction network. This protein is likely essential for accelerated cell division and myxozoan stage differentiation, as seen in a previous study of Myxobolus bejeranoi where differential expression of histones were found to be associated with parasitic invasion (Maor-Landaw et al., 2023). Calmodulin was also involved in the gill protein-protein interaction network, and is likely a key virulence factor that contributes to parasite motility and invasion of myxozoan parasites, as reported for other parasites (Long et al., 2017).