Duck Interleukin-22: Identification and Expression Analysis in Riemerella anatipestifer Infection

Riemerella anatipestifer is one of the most devastating pathogens affecting the global duck farms. Infection is involved in secretion of proinflammatory cytokines, including interleukin- (IL-) 17A. During the immune response to infection, IL-22 and IL-17A are often produced concurrently and at high levels in inflamed tissues. Little is known about duck IL-22 (duIL-22) during R. anatipestifer infection. We describe the characterization of duIL-22 and its mRNA expression analysis in splenic lymphocytes and macrophages treated with heat-killed R. anatipestifer and in the spleens and livers of R. anatipestifer-infected ducks. Full-length cDNA of duIL-22 encoded 197 amino acids. The deduced amino acid sequence of duIL-22 shared a 30.4–40.5% similarity with piscine counterparts, 57.4–60.1% with mammalian homologs, and 93.4% similarity to the chicken. Duck IL-22 mRNA expression level was relatively high in the skin of normal ducks. It was increased in mitogen-stimulated splenic lymphocytes and in killed R. anatipestifer-activated splenic lymphocytes and macrophages. Compared with healthy ducks, IL-22 transcript expression was significantly upregulated in the livers and spleens on days 1 and 4 postinfection, but not on day 7. IL-17A was significantly increased in the spleens only on day 4 postinfection and in the livers at all time points. When splenic lymphocytes were stimulated with heat-killed R. anatipestifer, CD4+ cells predominantly produced IL-22 while IL-17A was expressed both by CD4+ and CD4− cells. These results suggested that IL-22 and IL-17A are likely expressed in different cell types during R. anatipestifer infection.

IL-22 is well-identified and characterized in mammals and in many lower vertebrates, such as fugu [7], zebra fish [8], western clawed frog [9], Atlantic cod and haddock [10], rainbow trout [11], turbot [12], and so-iuy mullet [13]. In avian species, a putative sequence was found in turkey [14] and identified in chickens [15]. In chickens, costimulation of chicken embryo kidney cells with LPS and recombinant IL-22 enhances secretion of proinflammatory cytokines, chemokines, and AMPs [15]. There is little information about the function of IL-22 during the avian immune response.
Riemerella anatipestifer is a Gram-negative, nonmotile, and non-spore-forming bacterium belonging to the Flavobacteriaceae family. It induces acute and chronic septicemia manifested by airsacculitis, perihepatitis, fibrinous pericarditis, and meningitis [16]. Clinical disease primarily affects most bird species, including domestic ducks, turkeys, geese, chickens, and other wild birds [16][17][18]. At least 21 serotypes of R. anatipestifer that differ in virulence both between and sometimes within a given serotype have been identified, with less or no significant cross-protection among serotypes. Infection depending on strain-associated virulence causes mortality rate in a 5-75% range [16,19,20]. Although R. anatipestifer is an infectious disease that presents a significant threat of economic loss in the duck farms over the world [16], very little is known about the mechanisms of protective immunity underlying the pathogenesis of R. anatipestifer.
Our comparative expression analyses of immuneassociated genes in chickens and ducks found that IL-17A is significantly upregulated during R. anatipestifer infection in ducks and in splenic lymphocytes stimulated with heat-killed R. anatipestifer [21][22][23]. IL-22 and IL-17A are cytokines coexpressed by Th17 cells in response to IL-23 [24,25]. IL-22 and IL-17A exert similar protective function in controlling extracellular bacterial infection by inducing strong mucosal immunity [26] via induction of AMPs, recruitment of neutrophils to the site of bacterial invasion, and enhancement of mucosal barrier maintenance by stimulation of epithelial cell proliferation and tight junction protein production [24,27,28]. On the contrary, coexpression of IL-22 and IL-17A has been implicated in the development of proinflammatory processes during inflammatory and autoimmune diseases, such as psoriasis and Crohn's disease [25,26,29].
The goal of this study was to identify and analyze gene expression of IL-22 in ducks. We also examined the potential association between IL-22 and IL-17A in R. antipestifer-infected ducks. We, here, describe the full-length cDNA encoding duck IL-22 (duIL-22) protein, together with gene expression analyses in normal tissues and mitogen-treated splenic lymphocytes. This study found that IL-22 was upregulated during R. anatipestifer infection and was predominantly produced by CD4 + cells. IL-17A was produced by CD4 + cells and other cell types. These results suggested that there are differences in IL-17A and IL-22 production during R. anatipestifer infection.

Materials and Methods
2.1. Animals and Infection. Pekin ducklings were purchased (Joowon ASTA Ducks, Korea). Throughout the experiment, they were kept in wire cages in temperature-controlled conditions and provided ad libitum access to water and food. Sixty ducks were injected via the intramuscular route with 5 × 10 7 CFU of R. anatipestifer serotype 7 suspended in 200 μl PBS. On days 1, 4, and 7 after infection, the spleens and livers were harvested from five ducks at each time point for gene expression analysis. Similarly, control birds were injected with 200 μl phosphate-buffered saline (PBS), and the spleens were collected and analyzed time-matched to the infected group. All experiments were maintained in strict accordance with Gyeongsang National University (GNU) Guide to the Care and Use of Experimental Animals and were approved by IACUC (GNU-180206-M0009). The bacteria were grown first in sheep blood agar plates (Asan Pharmaceutical, Korea) in a 37°C incubator for 48 h with 5% CO 2 . A single selected colony was inoculated into tryptic soy broth (TSB; Difco, USA) and incubated in a shaking incubator at 37°C until the logarithmic growth phase. Serial dilutions (1 : 10) of the bacterial stock were plated onto sheep blood agar plates to determine the final concentration of the stock to be used as the inoculum.

Molecular Cloning and Sequence Analysis of DuIL-22.
Specific primers (Table 1) for reverse transcriptionpolymerase chain reaction (RT-PCR) were designed from the predicted sequence to obtain a partial sequence of the internal regions of duIL-22 cDNA from duck splenic lymphocytes stimulated with ConA. With a partial sequence information obtained, full-length duIL-22 cDNA was generated by applying 5 ′ /3 ′ rapid amplification of cDNA ends (RACE) with the specific primers (Table 1), high-fidelity DNA polymerase (Bioneer, Korea), and a 5 ′ /3 ′ RACE kit (2 nd Generation; Roche Applied Sciences, Germany), following the manufacturer's protocol. RT-PCR products were then cloned to TA Vector (RBC, Taiwan) and sequenced (Macrogen, Korea). RT-PCR was performed on a thermocycler (Bio-Rad, Hercules, CA, USA) using 5 min at 95°C; 35 cycles of 1 min at 95°C, 1 min at 55°C, and 1 min at 72°C; and a final 5 min extension at 72°C. The cloned duIL-22 cDNA sequence was submitted to GenBank (accession number MT360382).
The cloned sequence was analyzed for similarity and percent identity with other known homologue sequences using EMBOSS Needle pairwise alignment (http://www.ebi.ac.uk/ Tools/psa/emboss_needle). Protein identification was performed using the Expert Protein Analysis System (ExPASy; http://www.expasy.org/tools/). Amino acid sequence comparison was performed using Clustal Omega multiple sequence alignment (http://www.ebi.ac.uk/Tools/msa/clustalo/). The signal peptide sequence, N-glycosylation site, and the theoretical pI value and molecular weight were predicted using the SignalP 4.1 Server (http://www.cbs.dtu.dk/services/SignalP), NetNGlyc 1.0 Server (http://www.cbs.dtu.dk/services/NetNGlyc/), and Compute pI/Mw (http://www.expasy.org/compute_pi), respectively. A phylogenetic tree was generated based on multiple alignments of the protein sequences of IL-10 cytokine family members using the CLUSTAL X Multiple Sequence Alignment Program and neighbor-joining method in the MEGA7 program with a bootstrap value of 10,000 for the evolutionary analysis [30].
2.3. Cell Culture. Lymphocytes and macrophages were obtained from the spleens of two-week-old healthy ducks [31,32]. The cells were grown in Dulbecco's modified Eagle's medium (DMEM) (Gibco Life Technologies, USA) containing 2.5. Western Blotting. As previously described [23], cell lysis, sample preparation, SDS-PAGE, and Western blot analysis were carried out. Blotted membranes were incubated with monoclonal anti-MYC mouse antibody (Cell Signaling Technology, Danvers, MA, USA) and horseradish peroxidase-conjugated goat anti-mouse IgG antibody (Promega). The membranes were then washed, incubated with EZ West Lumi plus (ATTO, Japan), and were visualized using a ChemiDoc Imaging System (Bio-Rad, USA).
2.6. Splenic CD4 + Cell Isolation. Single-cell suspensions were collected from the spleens of two-week-old normal ducks [32]. The lymphocyte was isolated by using Ficoll-Paque PLUS ((GE Healthcare Bio-Sciences, Sweden)). After isolation, the cells were treated using R. anatipestifer killed by boiling (1 × 10 6 CFU/ml) and were sorted by using an anti-PE multisort kit (Miltenyi Biotech, Auburn, CA), following the manufacturer's protocol. Briefly, the cells were incubated with PE-conjugated anti-chicken CD4 and MACS buffer (PBS buffer supplemented with 0.5% BSA and 2 mM EDTA). After incubation, unbound antibodies were removed using centrifugation, and the pellet was resuspended and incubated with anti-PE multisort beads and MACS buffer. Unbound beads were removed using centrifugation, and the CD4 + cells were collected using positive selection in a LS column (Miltenyi Biotech, USA).

Quantitative Real-Time RT-PCR (qRT-PCR).
Total RNA from cells and tissues of R. anatipestifer-infected ducks and the time-matched untreated/uninfected control ducks were isolated using RiboEx (GeneAll, Korea). They were purified using an RNeasy mini kit (Qiagen, Germany). RNA was subjected to reverse transcriptase using a QuantiTect reverse transcription kit (Qiagen, Germany) with a random hexamer primer. cDNA was subjected to qRT-PCR analysis with SYBR Green (Bioneer, Korea) and gene-specific primers (Table 1) by using a CFX 96 real-time PCR system (Bio-Rad). Relative expression levels of each transcript were        Journal of Immunology Research with InStat® software (GraphPad, USA) and were tested using Student's t -tests or one-way analysis of variance (ANOVA), followed by Dunnett's multiple comparison tests. A P value < 0.05 was considered to be a statistically significant result. The protein was predicted to have a 28-aa signal peptide sequence, a 169-aa mature peptide, and one putative Nlinked glycosylation area (Asn-X-Ser/Thr) at position 60 ( Figure 1(a)). Multiple sequence alignment of duIL-22 to its homologs on other vertebrates revealed five conserved cysteine residues and the presence of a relatively well conserved IL-10 family motif (Figure 1(b)). Comparison of the duIL-22 nucleotide sequence to its piscine, mammalian, and chicken IL-22 homologs revealed identities of 9.  Table 2). The relationships between the identified duIL-22 and other members of the IL-10 family were further investigated by constructing a phylogenetic tree with the neighbor-joining method in the MEGA7 program. Analysis of the tree revealed that the duIL-22 protein was more closely related to its bird counterparts than to its piscine and mammalian counterparts. Likewise, IL-22 formed a distinct branch from other members of the IL-10 cytokine family (Figure 1(c)). 7 Journal of Immunology Research mRNA expression levels were high in the skin; moderate levels were detected in the liver, heart, lungs, and brain. Low levels of IL-22 transcripts were found in the intestines and lymphoid organs (Figure 2(b)). DuIL-22 mRNA transcripts on mitogen-stimulated lymphocytes such as LPS, PolyI:C, and ConA were generally upregulated at all times, compared with unstimulated cultured splenic lymphocytes (Figure 2(c)).

IL-22 and IL-17A
Are Upregulated during R. anatipestifer Stimulation and Infection. Gene expression analyses of Th17-associated cytokines from our previous studies showed that IL-17A and IL-17F expression was increased in the spleens and livers of R. anatipestifer-infected ducks and in splenic lymphocytes treated with R. anatipestifer killed by boiling [22,33]. Both IL-22 and IL-17A promoted induction of AMP; they are implicated together in the inflammatory processes involved in cases of psoriasis, inflammatory bowel diseases, and rheumatoid arthritis [24,34,35]. These findings suggested that both cytokines have similar roles. Thus, the mRNA expression patterns of IL-17A and IL-22 were monitored in duck splenic macrophages and lymphocytes activated with heat-inactivated R. anatipestifer ( Figure 3) and in R. anatipestifer-infected ducks (Figure 4). Compared with the untreated cultured controls, IL-17A and IL-22 expression was significantly increased at all time points in the splenic lymphocytes and macrophages treated with heat-inactivated R. anatipestifer. Expression levels in the stimulated splenic lymphocytes showed changes of 7.7-127.8-fold and 6.7-131.2-fold for IL-22 and IL-17A, respectively. Similarly, when compared to their unstimulated culture controls, we found an upregulated expression of a 3.9-73.2-fold change for IL-22 and a 6.7-139.5-fold change for IL-17A on splenic macrophages stimulated with heatinactivated R. anatipestifer. These findings suggested that duck IL-22 and IL-17A were upregulated in vitro during stimulation with R. anatipestifer.
As both cytokines were upregulated in the splenic macrophages and lymphocytes stimulated with heat-inactivated R. anatipestifer, the expression profiles of these cytokines were monitored in the spleens and livers of infected ducks

IL-22 Is
Mainly Produced by CD4 + Cells. IL-22 and IL-17A are cytokines coexpressed by Th17 cells in response to IL-23 [24,25]. Recently, duck IL-23p19 was found to be unimportant in the IL-17A response during the early stages of R. anatipestifer infection in ducks [23]. Thus, it would be interesting to know what cells might be involved in the production of IL-22 during the infection, and more so if IL-22 is somewhat produced in the same manner or maybe regulated differently with IL-17A during R. anatipestifer infection. Splenic lymphocytes of ducks were activated with heat-inactivated R. anatipestifer, and CD4 + cells were separated. The CD4 + predominantly produced IL-22, while IL-17A was expressed both by CD4 + and other cells of the spleen ( Figure 5).

Discussion
Clinical R. anatipestifer infection in ducks is considered one of the most economically critical diseases in the duck farms.
In duck populations, infection can result in a mortality rate as high as 75%, depending on strain virulence [16,20]. Our previous studies are associated with upregulated expression of IL-17A and IL-17F mRNA in infected ducks and in activated splenic lymphocytes [22,33]. The association of IL-17 with infection is further confirmed when increased survival rates and decreased bacterial burdens in ducks infected with R. anatipestifer occur after expression levels of IL-17A and IL-17F cytokines are downregulated using berberine treatment [33]. IL-17A is the signature cytokine of Th17 cells and is often associated with various autoimmune and proinflammatory diseases, such as organ-specific inflammation [36] and psoriasis [37]. Th17 cells also express IL-22. IL-22 is a novel IL-10 family member that can induce production of AMPs (β-defensins and S100 proteins) [35] and acute phase proteins (serum amyloid A and LPSbinding protein) [38]. Similar to IL-17A, IL-22 is linked to the pathogenesis of various proinflammatory processes. Therefore, this study was performed to investigate whether there was a relationship between IL-22 and IL-17A, or whether they were regulated differently during R. anatipestifer infection. To answer this question, duIL-22 was first identified from mitogen (ConA)-treated splenic lymphocytes. We then characterized its molecular features and expression profiles of transcripts in normal tissues of ducks, in mitogen-stimulated cells, and in the spleens and livers of R. anatipestifer-infected ducks. We also monitored expression levels of IL-22 and IL-17A in CD4 + and CD4cells separated from heat-killed R. anatipestifer-stimulated splenic lymphocytes. This study was the first to perform molecular cloning and characterization of IL  [13]. Five cysteine residues were also conserved between mammalian and avian species, compared with their piscine counterparts, who had three or four  Journal of Immunology Research conserved cysteine residues [10,13]. Comparison of nucleotide sequence and amino acid sequence of duIL-22 revealed higher identities and similarities to the poultry species, compared with their piscine and mammalian counterparts. Phylogenetic tree analysis of the IL-10 family found that duIL-22 was more closely related to its poultry counterparts than its mammalian or piscine homologs. Moreover, IL-22 was generally more evolutionarily closer to IL-10 and human IL-26 cytokines (Figure 1(c)).
Sequence analysis revealed that duIL-22 contained a 594 bp ORF predicted to encode 197 amino acids with a signal peptide (amino acid 1-28) and a single N-linked glycosylation site. A one putative N-glycosylation area in duck IL-22 protein is consistent with the presence of a single putative N-glycosylation area in its chicken counterpart. However, mouse and human IL-22 have three sites and rainbow trout have no putative glycosylation sites [11,15]. Interestingly, when supernatants and lysates of COS-7 cells expressing duIL-22-MYC construct were treated with PNGase-F, Western blot analysis revealed a 19 kDa protein on the supernatants and lysates, whereas a 22.8 kDa protein was detected on the untreated supernatants and lysates. This result suggested that duIL-22 had one N-linked glycosylation (Figure 2(a)).
IL-22 tissue distribution experiments found that it was expressed in a wide variety of tissues, primarily in the skin, followed by the liver, heart, lung, brain, and kidney. This result was likely because IL-22R is expressed on stromal and epithelial cells of these tissues [1]. This result differed from distribution in the chicken, in which high expression is found in the thymus, spleen, and intestine (jejunum and ileum) [15]; duIL-22 expression levels were relatively low in these tissues (Figure 2(b)). DuIL-22 distribution was also different from IL-22 distribution in teleost species, in which IL-22 is highly expressed in the intestines, tail fins, gills, and gonads [8,10,11]. Although primarily produced by lymphoid tissues, IL-22 is also produced in nonlymphoid sources, such as macrophages in the lungs of humans and mice, in response to lung injury. Experimental models found that IL-22 is produced in neutrophils during colitis and in fibroblasts during rheumatoid arthritis [1].
Expression of duIL-22 was significantly upregulated in mitogen-stimulated splenic lymphocytes at all time points. These results were consistent with results indicating that IL-22 expression is induced in rainbow trout splenocyte primary cultures stimulated with phorbol myristate acetate [11], a potent enhancing factor for T cell colony growth [39]. IL-22 expression is induced in turbot head kidney, spleen, and intestine cells stimulated with phorbol myristate acetate and phytohemagglutinin [12]. Similarly, expression of IL-22 and IL-17A transcripts in splenic lymphocytes and macrophages was stimulated with heat-killed R. anatipestifer at all times. Thus, to study a possible link between IL-22 and IL-17A in R. anatipestifer infection, ducks were injected with R. anatipestifer via the intramuscular route, and the spleens and livers were collected and analyzed at days 1, 4, and 7 postinfection. Unlike our in vitro findings that both cytokines which were expressed at all times are expressed with a similar pattern, only IL-17A was significantly induced in the liver at all time points; IL-22 was induced on days 1 and 4, but not on day 7. In the spleen, IL-22 was significantly expressed on days 1 and 4 after infection, while significant IL-17A expression was found only day 4. Although, to a certain extent, IL-22 and IL-17A expression was overlapped in the spleens of ducks infected with R. anatipestifer, these results implied a possibility that IL-22 and IL-17A are expressed in different cell types of R. anatipestifer-infected ducks. It is interesting to note that distinct expression of IL-17A and IL-23p19, unlike in vitro expression patterns analyzed with R. anatipestifer-treated splenic lymphocytes, was identified in vivo analyzed with the spleens of ducks infected with R. anatipestifer [23]. Differential expression of IL-22 and IL-17A was confirmed in peritoneal lavage fluids of Salmonella Enteritidis-infected mice; IL-22 was induced on day 1, but not on day 14, whereas IL-17A was induced on day 14, but not on day 1 [40]. Thus, CD4 + cell expression levels were analyzed for further insight on IL-22 and IL-17A production during R. anatipestifer infection. IL-22 was detected at significant levels on CD4 + cells, while IL-17A was present on CD4 + cells and other splenocytes. These findings suggested that IL-22 and IL-17A were expressed in different cell types during R. anatipestifer infection.
In conclusion, duIL-22 was cloned to report IL-22 expression profiles on duck tissues and immune cells. Furthermore, we aimed to determine whether there was a connection between IL-22 and IL-17A during R. anatipestifer infection because both cytokines are coexpressed by certain cell subsets for protection or during disease progression. Unlike the in vitro results, we found that IL-22 and IL-17A were not expressed in the same manner, especially in the spleens of infected ducks. Moreover, while CD4 + cells expressed both IL-22 and IL-17A, IL-22 was primarily produced only from CD4 + cells. Further investigation is needed to accurately identify subsets of cells that express IL-22.

Data Availability
Data will be available from authors on request.