Innate Immune Responses of Skin Mucosa in Common Carp (Cyprinus Carpio) Fed a Diet Supplemented with Galactooligosaccharides

Simple Summary The skin mucosa in fish is equipped with innate immune mechanisms, which constitute the first line of defense against potentially harmful factors in the aquatic environment. Prebiotics, such as galactooligosaccharides (GOS), mediate modulation of the immune responses. Dietary prebiotics promote the development of intestinal microbiota, which interacts with the immune system. In this study, we analyzed the effects of the GOS prebiotic applied as a feed additive on molecular responses in the skin mucosa of the common carp. The genes analyzed encode various proteins associated with the innate immune response in skin mucosa in fish, such as mucin secretion, antimicrobial peptides, and enzymes. Modulated expression of those genes at the mRNA level regulates the defense mechanisms in the skin mucosa. In this study, supplementation with GOS increased the mRNA level of interferon and lysozyme, which are involved in fighting infection. At the same time, GOS decreased the mRNA level of CRP, which is a marker of inflammation. We conclude that supplementation with GOS modulates innate immune processes in the skin mucosa of common carp. Further studies could be focused on analyzing the effects of GOS on the microbiota composition of the skin mucosa and the mitigating effects against environmental stress. Abstract Galactooligosaccharides (GOS) are well-known immunomodulatory prebiotics. We hypothesize that GOS supplemented in feed modulates innate immune responses in the skin-associated lymphoid tissue (SALT) of common carp. The aim of this study was to determine the impact of GOS on mRNA expression of the immune-related genes in skin mucosa. During the feeding trial, the juvenile fish (bodyweight 180 ± 5 g) were fed two types of diet for 50 days: control and supplemented with 2% GOS. At the end of the trial, a subset of fish was euthanized (n = 8). Skin mucosa was collected, and RNA was extracted. Gene expression analysis was performed with RT-qPCR to determine the mRNA abundance of the genes associated with innate immune responses in SALT, i.e., acute-phase protein (CRP), antimicrobial proteins (His2Av and GGGT5L), cytokines (IL1β, IL4, IL8, IL10, and IFNγ), lectin (CLEC4M), lyzosymes (LyzC and LyzG), mucin (M5ACL), peroxidase (MPO), proteases (CTSB and CTSD), and oxidoreductase (TXNL). The geometric mean of 40s s11 and ACTB was used to normalize the data. Relative quantification of the gene expression was calculated with ∆∆Ct. GOS upregulated INFγ (p ≤ 0.05) and LyzG (p ≤ 0.05), and downregulated CRP (p ≤ 0.01). We conclude that GOS modulates innate immune responses in the skin mucosa of common carp.


Fish, Feeds, and Experimental Design
The experiment was conducted at the Experimental Station of Feed Production Technology and Aquaculture (affiliated with Poznań University of Life Sciences) in Muchocin (Poland). Three hundred one-year-old fish of the common carp (Cyprinus carpio), with a mean bodyweight of 180 g (±5 g), were placed in 12 concrete tanks (60 m 3 ). The distribution of the fish was 25 individuals per tank. The tanks were individually supplied with water from the Struga Dormowska river, in an open system with a mechanical prefiltration chamber. Construction of the tanks allowed the maintenance of the maximum water level with constant water flow. Each tank was equipped with an automatic belt feeder allowing permanent access to the feed for 12 h a day.
The daily diet ration was calculated based on the carp feeding key developed by Miyatake [32]. Water temperature and the current fish weight was accounted for. The dietary formulation and proximate composition of the feeds are shown in Table 1. The feeds were processed using a single-screw warm extruder (Metalchem S-60, Gliwice, Poland). The extrusion conditions were as follows: cylinder temperature under stress of increasing pressure 90 • C, 100 • C in the high-pressure zone, 110 • C in the head, with a screw diameter of 6 mm and a speed of 52 rpm. The fish were divided into two groups: control group (CON), which received a diet without supplements, and experimental group (GOS), fed a diet supplemented with 2% GOS (Bi 2 tos ® , Clasado Biosciences Ltd., Jersey, UK). The feeding trial lasted 50 days, from 30.04. to 19.06.2018. Every ten days, individual weights of all fish in each tank were measured, and the feed ration and rearing indices were calculated. Table 1. Dietary formulation and proximate composition of the feed.

Tissue Collection and RNA Isolation
Samples of the skin mucosa were collected from randomly selected individuals (n = 8) with an average body weight of 500 g (±10 g). The mucus was gently scraped from the skin using a sterile glass slide and stabilized in 3 ml of RNAlater (Invitrogen, Waltham, MA, USA). The samples were stored at −80 • C until total RNA isolation. Prior to total RNA isolation, the samples of skin mucosa were homogenized with the TissueRuptor homogenizer (Qiagen GmbH, Hilden, Germany) in TRIzol®LS Reagent (Ambion/Thermo Fisher Scientific, Waltham, MA, USA). The lysate was processed using a EURx Universal RNA Purification Kit (EURx, Gdansk, Poland). The RNA quality and quantity were determined by gel electrophoresis using 2% agarose gel (to verify the integrity of 18S and 28S rRNA) and NanoDrop 2000 (to measure the absorbance at 260/280 nm) (Scientific Nanodrop Products, Wilmington, DE, USA). RNA was frozen at -80 • C prior to downstream analyses.

Reverse Transcription-Quantitative PCR (RT-qPCR)
Reverse transcription (RT) was performed using a Maxima First Strand cDNA Synthesis Kit for RT-qPCR (Thermo Scientific/Fermentas, Vilnius, Lithuania), following the manufacturer's recommendations. Obtained cDNA was diluted to 70 ng/µL and stored at −20 • C. RT-qPCR reactions were conducted with a total volume of 10 µL. The reaction mixture contained 1x Maxima SYBR Green qPCR Master Mix (Thermo Scientific/Fermentas, Vilnius, Lithuania), 1 µM of each primer (Sigma-Aldrich, Germany), and 2 µl of diluted cDNA. Thermal cycling was performed in a LightCycler II 480 (Roche Diagnostics, Basel, Switzerland). The qPCR amplification comprised an initial denaturation step for 15 min at 95 • C, followed by 40 cycles of denaturation (10 s at 95 • C), annealing (15 s at 58 • C), and extension (30 s at 72 • C). Fluorescence was measured at the end of each extension step. The thermal program was completed by the melting curve, which was generated by increasing the temperature in small increments up to 98 • C and measuring the fluorescence of the melting amplicon. Each RT-qPCR reaction was conducted in triplicates (reference genes) or duplicates (target genes). Oligonucleotide primers were synthesized based on sequences from literature or in-house designed. The selection of reference and target genes and primer details are described in the Section 2.4. "Gene selection".

Reference Genes
Reference genes for the relative gene expression analysis were selected based on a two-step selection process. First, the related literature was studied to pinpoint the relevant panel of reference genes for carp [24,[33][34][35]. Second, the RT-qPCR analysis was performed on a full set of cDNA samples to determine the reference genes' quality and stability in the samples from skin mucosa. The RT-qPCR for reference genes was performed based on the methodology described in Section 2.3. "Reverse Transcription Quantitative PCR (RT-qPCR)". Table 2 presents the list of the reference genes and the respective oligonucleotide primers. Ct values from CON and GOS groups were analyzed using RefFinder [36]. RefFinder integrates different algorithms that are commonly used in reference genes analysis, including BestKeeper [37], NormFinder [38], geNorm [39], and the comparative delta-Ct method [40]. Analysis of the panel of reference genes allowed for selecting the best combination of the reference genes for relative expression of the target genes.

Target Genes
The selection of the target genes was based on the literature on the immune responses generated in SALT [5,[41][42][43][44]. First, a list of immunological processes that occur in SALT to protect fish from external factors was determined. These processes include defense against Gram-positive bacteria, Gram-negative bacteria, viruses, fungi, and yeast; hydrolyzing peptide and glycosidic bonds of the cell walls; promoting phagocytosis; activating complement pathways; turning off opsonization; and stress response. Next, proteins involved in these processes were pinpointed, and the underlying genes were considered target genes for this study. In effect, a comprehensive panel of the genes expressed in the fish skin mucosa was selected. A DNA sequence of the respective target genes was derived from a gene-related section of the NCBI database (https://www.ncbi.nlm.nih.gov/gene). The RT-qPCR primers were designed using Primer-BLAST [45], which is a primer designing tool implemented in the NCBI database (https://www.ncbi.nlm.nih.gov/tools/primer-blast/index.cgi). The RT-qPCR primers were designed based on the following criteria: amplicon size between 70 and 200 bp, span of an exon-exon junction (does not apply to CRP gene due to the presence of only one exon), optimal melting temperatures around 60 • C, the 3' end of primers contains a C or G residue (if possible), CG content around 40%-60%, and exclusion of primer-dimer formation. The list of the analyzed genes, including their biological function and primer sequences, is presented in Table 3. Table 3. Immune-related genes and primer sequences for RT-qPCR analysis of the skin mucosa in common carp.

Name
Gene Gene ID Function 1 Primer Sequences (5'→3') Ref. 2 Acute-phase protein C-reactive protein CRP 109083752 Host defense: it promotes agglutination, bacterial capsular swelling, phagocytosis, and complement fixation through its calcium-dependent binding to phosphorylcholine.  Actins are highly conserved proteins that are involved in cell motility, structure, integrity, and intercellular signaling F:ATCCGTAAAGACCTGTATGCCA R:GGGGAGCAATGATCTTGATCTTCA [24] 40S ribosomal protein S11 40s s11 109061205 Relation with viral mRNA translation and activation of the mRNA pathways upon binding of the cap-binding complex and eIFs, and subsequent binding to 43S F:CCGTGGGTGACATCGTTACA R:TCAGGACATTGAACCTCACTGTCT [33]

Relative Quantification of Gene Expression and Statistical Analysis
Normalization of the expression levels (Ct values) of the target genes was performed with two selected reference genes (ACTB and 40s s11). The geometric mean of Ct values between reference genes was used to calculate ∆Ct, according to the formula: ∆Ct = Ct target -Ct reference. Relative gene expression was calculated with the ∆∆Ct algorithm, in which CON was considered a calibrator. The fold change in the target genes in GOS was calculated as 2 -∆∆Ct [48]. To compare the data between GOS and CON, a Student's t-test was performed using the SAS Enterprise Guide 9.4 program (SAS Institute, Cary, NC, USA). The difference between GOS and CON was considered significant when p < 0.05. Figures were prepared using GraphPad Prism 7 (GraphPad, La Jolla, CA, USA).

Reference Genes
The results of the reference genes used for relative gene expression analysis in skin mucosa of common carp are presented in Figure 1. All computational methods (the comparative delta-Ct method, BestKeeper, NormFinder, and GeNorm) showed that the two genes that are most stable in the carp skin mucosa are: 40s s11 and ACTB. In addition, GeNorm indicated that the best set of the candidate genes for normalization of the experiment would be the geometric mean of the ACTB/40s s11 genes. In the following calculations of the relative immune-related gene expression analysis, the geometric mean of those two reference genes was used.

Relative Quantification of Gene Expression and Statistical Analysis
Normalization of the expression levels (Ct values) of the target genes was performed with two selected reference genes (ACTB and 40s s11). The geometric mean of Ct values between reference genes was used to calculate ∆Ct, according to the formula: ∆Ct = Ct target -Ct reference. Relative gene expression was calculated with the ∆∆Ct algorithm, in which CON was considered a calibrator. The fold change in the target genes in GOS was calculated as 2 -∆∆Ct [48]. To compare the data between GOS and CON, a Student's t-test was performed using the SAS Enterprise Guide 9.4 program (SAS Institute, Cary, NC, USA). The difference between GOS and CON was considered significant when p < 0.05. Figures were prepared using GraphPad Prism 7 (GraphPad, La Jolla, CA, USA).

Reference Genes
The results of the reference genes used for relative gene expression analysis in skin mucosa of common carp are presented in Figure 1. All computational methods (the comparative delta-Ct method, BestKeeper, NormFinder, and GeNorm) showed that the two genes that are most stable in the carp skin mucosa are: 40s s11 and ACTB. In addition, GeNorm indicated that the best set of the candidate genes for normalization of the experiment would be the geometric mean of the ACTB/40s s11 genes. In the following calculations of the relative immune-related gene expression analysis, the geometric mean of those two reference genes was used.

Immune-Related Gene Expression
The relative gene expression analysis for immune-related genes in the carp skin mucosa is shown in Figure 2. The majority of the genes were upregulated in GOS vs. CON. Only CRP and LyzC were downregulated. Statistically significant differences at the mRNA level were demonstrated for INFγ and LyzG genes, which were upregulated in GOS (p ≤ 0.05). The CRP gene was significantly downregulated in GOS (p ≤ 0.01). Furthermore, a suggestive statistical trend was found for the MPO gene, which was upregulated in GOS compared to CON (0.05 < p <0.1).
ribosomal RNA (18s rRNA), and 40s ribosomal protein s11 (40s s11). Dataset was generated for GOSsupplemented and control animals (n = 8) using RT-qPCR. qPCR reactions were performed in triplicates. RefFinder was used to calculate the gene stability values. 40s s11 and ACTB (labeled green) were selected as the most stable pair of reference genes for skin mucosa study in carp. Figures were prepared using GraphPad Prism 7 (GraphPad, La Jolla, CA, USA).

Immune-Related Gene Expression
The relative gene expression analysis for immune-related genes in the carp skin mucosa is shown in Figure 2. The majority of the genes were upregulated in GOS vs. CON. Only CRP and LyzC were downregulated. Statistically significant differences at the mRNA level were demonstrated for INFγ and LyzG genes, which were upregulated in GOS (p ≤ 0.05). The CRP gene was significantly downregulated in GOS (p ≤ 0.01). Furthermore, a suggestive statistical trend was found for the MPO gene, which was upregulated in GOS compared to CON (0.05 < p <0.1).

Discussion
Expression of the immune-related genes gives insight into the mechanisms of the innate immune responses. In aquaculture, replacing antibiotics with prebiotics, probiotics, or synbiotics, increases immunological competence and resistance to diseases in an environmentally friendly way [49]. The immunomodulatory role of prebiotics results from direct interactions with the innate immune system and/or indirectly, by selectively stimulating the growth of commensals inhabiting the host's mucosa [50,51]. This way, prebiotics can stimulate intestinal epithelial cells to release cytokines that modulate the spectrum of the mucosal immune system, including dendritic cells, T cells, and B cells, which, in turn, trigger the transcription of immune-related genes (e.g., tumor necrosis factor α, or lysozyme). This process leads to an increase in innate immune responses [52,53].

mRNA Expression Stability of the Reference Genes
One of the most important factors that can skew the results of the relative gene expression at the mRNA level is the selection of the most stable reference genes for a given tissue. To our knowledge, the information regarding the most suitable genes for normalizing the RT-qPCR data in the skin mucosa of common carp was lacking. In fish, the most common reference genes are ACTB [54], B2M, 18s rRNA, EF1α, and GAPDH [55][56][57][58]. This diversity may be due to the species characteristics, individual tissues, age, and type of experiment. Thus, it is necessary to test and compare different housekeeping genes in all experimental conditions [59]. In this study, five candidate genes were evaluated for the normalization of RT-qPCR in the skin mucosa of the common carp. The highest expression stability in skin mucosa was found for ACTB and 40s s11 genes. Gene ACTB encodes actin. All actins are highly conserved and involved in cell motility, structure, integrity, and intercellular signaling. ACTB is a protein found in most vertebrate cells as components of the cytoskeleton (https://www.genecards.org/). The protein encoded by the 40s s11 gene is a member of the S17P family of ribosomal proteins. This family is a component of the small ribosome subunit (40S). The main function of this gene is binding RNA and rRNA. Protein 40s s11 is also a structural component of the ribosome (https://www.genecards.org/). In our experiment, the mRNA expression of ACTB and 40s s11 genes was not affected by the experimental factor (GOS). For this reason, they are good candidates for internal control genes in RT-qPCR experiments.

Immune-Related Gene Expression in Skin Mucosa
The obtained results of the immune-related gene expression in response to GOS supplementation showed a significant increase in the genes associated with antiviral (IFNγ) and antimicrobial (LyzG) immune responses. On the other side, the CRP gene representing the acute phase response, was decreased in the GOS-supplemented group. IFNγ is an antiviral and immunoregulatory cytokine that is necessary for cellular defense. It is produced by T cells and natural killer cells as a dimerized soluble glycoprotein [60]. Lysozymes are an important element of innate immunity. They are able to catalyze hydrolysis of the bacterial glycosidic bonds. Several types of lysozymes have been described, such as lysozyme C (chicken), lysozyme G (goose), phage, bacterial and plant lysozymes [61]. In fish, lysozyme is found in mucus, serum, and ova [62] and is found in two forms, C and G. Fish lysozymes are thought to have a much more bactericidal effect than those produced by the higher vertebrates [61]. CRP belongs to the family of soluble proteins that are involved in the acute phase reaction (APR) to injury, damage, or infection. CRP is able to bind to phosphorylcholine, pneumococcal C-polysaccharide, and phospholipids, as well as to autogenic compounds, such as apoptotic nuclear components and other intracellular components released after cell death [63]. It also binds harmful molecules, such as mercury, increases phagocytic clearance, and triggers complement activation via the classical pathway [64].

Effects of GOS in Fish
Several papers have reported improvement in the immunological properties of the fish skin mucosa after application of the GOS prebiotic. They refer to the increase in the activity of immunological factors at the protein level in the skin mucosa. Hoseinifar et al. [65] compared the effects of three prebiotics (FOS, GOS, and inulin) in common carp. A significant increase in dermal lysozyme activity in the experimental group fed GOS was demonstrated, which is in line with the present study. Hoseinifar et al. [23] demonstrated similar effects in rainbow trout (Oncorhynchus mykiss) fed diets supplemented with 2% GOS. The supplementation of GOS (1%-2%) in the nutrition of the white Caspian fish (Rutilus frisii kutum) for eight weeks also increased the lysozyme activity and total immunoglobulin level [66]. The introduction of XOS in diets fed to the white Caspian fish also increased bactericidal activity in the skin mucus [67]. In goldfish (Carassius auratus auratus), 1% and 2% GOS significantly improved the immune parameters of the skin mucus (lysozyme and total protein) compared to the control and to the reduced proportion of GOS in the mix (0.5%) [68]. These studies suggest that dietary prebiotics (including GOS) indirectly support mucosal immunity. However, further research is needed on total IgA initiation by prebiotics in fish to fully understand the effect of immunomodulation [69].
The effects of GOS supplemented to the fish also improved intestinal function and muscle structure in common carp (Cyprinus carpio). The feeding trial reported in this paper lead to the discovery that GOS provided in the diet improved the histological picture of the intestines, including the height and thickness of intestinal villi. Such morphological changes in the fish guts increased the absorptive surface of the small intestine. The elevated ratio of villi height to intestinal crypt depth suggests improved maturity of the intestinal mucosa of the GOS-supplemented carp diet (Ziółkowska et al., submitted). Regarding the muscle structure, the addition of 2% GOS increased the diameter and density of the white muscle fibers responsible for the marbling of the fish meat. Along with that, the percentage of muscle fiber atrophy decreased (Ziółkowska, personal communication). In conclusion, GOS supplementation in carp improved intestinal and muscular morphology.

Immunomodulatory Role of GOS
GOS used in this study was produced from lactose by galactosyltransferases from Bifidobacterium bifidum NCIMB 41,171 isolated from a human stool sample [70]. The immunomodulatory effects of this particular compound have been well-established in human and poultry. The introduction of GOS to human diets increased fecal bifidobacteria abundance while reducing less desirable strains [71][72][73]. Dietary GOS significantly increased fagocytosis, stimulated NK activity, increased levels of anti-inflammatory cytokine (IL-10) and decreased levels of pro-inflammatory cytokines (IL-6, IL-1, and TNF-α) in elderly people [72]. Supplementation of GOS in diets of overweight adults led to an increased abundance of colonic Bifidobactera, as well as increased production of fecal secretory IgA [71]. GOS has also been reported to alleviate the syndromes of irritable bowel syndrome [74], prevented the symptoms of traveler's diarrhea [75], and exerted positive effects on GI symptoms, including bloating, abdominal pain, and flatulence [76].
The immunomodulatory effects of GOS used for early stimulation of the chicken microbiota through in ovo technology have been widely documented [77][78][79]. Transcriptomic analysis revealed that GOS delivered in ovo modulated genes associated with lymphocyte proliferation, activation and differentiation, as well as cytokine production in the caecal tonsils of broiler chickens [78]. It was also found that GOS delivered in ovo increased expression of the genes related to mucosal immune response, intestinal barrier function, and nutrient sensing in the chicken gastrointestinal tract [77]. Particularly beneficial effects of GOS delivered in ovo were determined during heat stress in broiler chickens. Even a short-term increase in ambient temperatures resulted in elevated expression of the genes associated with immune response and stress response [80]. These effects were alleviated by GOS delivered in ovo, most likely due to maintaining intestinal eubiosis [80]. Furthermore, GOS delivered in ovo mitigated harmful effects of chronic heat stress on the performance and welfare of broiler chickens [81], as well as meat composition and quality [82].

Conclusions
Supplementation of the standard diets with GOS modulated innate immune responses of common carp. In this study, we found that dietary GOS exerted immunomodulatory effects on skin mucosa, which was manifested by mRNA expression of the genes involved in cytokine, lysozyme, and acute-phase protein production. In conclusion, GOS activated immunomodulatory pathways leading to gene expression modulation in SALT of common carp.