Cloning of Toll-like Receptor 3 Gene from Schizothorax prenanti (SpTLR3), and Expressions of Seven SpTLRs and SpMyD88 after Lipopolysaccharide Induction

Toll-like receptor 3 (SpTLR3) from Schizothorax prenanti (S. prenanti) was cloned and identified, and the tissue distribution of the SpTLR3 gene was examined in this study. Moreover, the relative mRNA expression levels of myeloid differentiation factor 88 gene (SpMyD88) and seven TLR genes (SpTLR2, SpTLR3, SpTLR4, SpTLR18, SpTLR22-1, SpTLR22-2 and SpTLR22-3) from S. prenanti after lipopolysaccharide (LPS) challenge were analyzed through quantitative real-time polymerase chain reaction (qRT-PCR). The full length of SpTLR3 gene is 3097 bp, and complete coding sequence (CDS) is 2715 bp, which encodes 904 amino acids. The SpTLR3 amino acid sequence shared 43.94–100% identity with TLR3 sequences from other vertebrates; SpTLR3 was expressed in all eight tissues examined; and the highest level appeared in the liver, which was significantly higher than in all other tissues (p < 0.05), followed by the levels in the heart and muscles. LPS significantly up-regulated all eight genes in the S. prenanti tissues at 12 or 24 h (p < 0.05). Compared with the PBS control group, no significant transcripts changes were found in SpTLR2 or SpTLR3 at 12 h after LPS induction, but they were significantly up-regulated at 24 h (p < 0.001). The most abundant transcripts were found in the head kidney SpTLR22 genes after 24 h LPS induction, with high to low levels, which were SpTLR22-1 (564-fold), SpTLR22-3 (508-fold) and SpTLR22-2 (351-fold). Among these eight genes, the expression level of SpTLR4 was the least up-regulated. Overall, SpTLR4 in the head kidney was involved in the antibacterial immune response earlier, and the level was increased at 12 h with extreme significance after LPS stimulation (p < 0.001), while the other seven genes were the most significantly up-regulated at 24 h post injection. Taken together, the results suggest that SpMyD88, SpTLR2, SpTLR3, SpTLR4, SpTLR18, SpTLR22-1, SpTLR22-2 and SpTLR22-3 participate in an innate immune response stimulated by LPS, and the response intensity of the genes was organ-specific, with differing kinetics. Our findings will contribute to a more complete understanding of the roles of these TLR genes in antibacterial immunity.


Introduction
The vertebrate immune system includes the innate immune system and the adaptive immune system. In mammals, the adaptive immune system plays a major role in the eventual elimination of pathogens. However, the difference is that fish belong to poikilotherm, and body temperature is not suitable for the development of acquired immune function. Therefore, the innate immune system plays a "leading role" in fish immunity [1]. Similar to mammals, the innate immunity of fish is mainly due to a series of evolutionarily highly conserved pattern recognition receptors (PRRs). The PRRs recognize pathogen-associated molecular patterns (PAMPs), which are conserved on the surface of microorganisms but not present in the host body. The invasion signal is transmitted to SpTLR5-2) [28] and SpTLR25 [29]. We treated S. prenanti with LPS intraperitoneal injection in this study, and examined the relative mRNA expression of MyD88, TLR2, TLR3, TLR4, TLR18, TLR22-1, TLR22-2 and TLR22-3 genes from S. prenanti in different immune organs (hepatopantras, head kidney, hindgut and spleen) at different infection times, which will help to further clarify the roles of SpTLRs and SpMyD88 genes involved in the antibacterial immunity mechanisms of fish.

Animal Treatment
We obtained healthy cultured S. prenanti (109.3 ± 27.1 g) in November 2020 from a professional breeding cooperative (Qunfu Yang-Fish, Hanzhong, China), which is a Chinese national aquatic germplasm resources breeding farm and commercial breeding aqua-farm. The fish were maintained in glass tanks with a size of 60 × 40 × 30 cm 3 and with aerated water at a temperature of 19 ± 1 • C; The tank filter was cleaned and about a quarter of the aerated tap water was replaced daily. The fish were fed with a commercial feed (Floating compound feed with crude protein ≥ 32%, crude fat ≥ 3%) at a rate of 2% of their body weight twice daily.
The fish were anesthetized with eugenol at a concentration of 80 mg/L for 3 min before dissected. The heart, liver, head kidney, hindgut, muscle, intraperitoneal fat, spleen and gill were collected from eugenol anaesthetized fish to detect the tissue distribution of SpTLR3 mRNA in unstressed conditions. To understand the expression of SpMyD88, SpTLR2, SpTLR3, SpTLR4, SpTLR18, and SpTLR22s (SpTLR22-1, SpTLR22-2, SpTLR22-3) in response to LPS stimulation, the fish were challenged with an intraperitoneal injection of LPS (10 mg/kg), and the control group was injected with phosphate-buffered saline (PBS) at the same volume. The liver, head kidney, hindgut and spleen were collected from four S.prenanti of PBS-injection and LPS treatment at 12 and 24 h, respectively. The dissected tissues and preserved in liquid nitrogen for RNA extraction.

RNA Extraction and cDNA Synthesis
Tissue total RNA was extracted using the TaKaRa MiniBEST Universal RNA Extraction Kit (TaKaRa, Dalian, China). The total RNA concentration and purity were determined by RNA electrophoresis and the optical density absorption ratio (A260/280) in the Nanodrop One spectrophotometer (Thermo Fisher Scientific, Wilmington, NC, USA). First-stand cDNA for SpTLR3 partial sequence amplification was synthesized using the RevertAid First Strand cDNA Synthesis Kit (Thermo Scientific, Vilnius, Lithuania), according to the method recommended by the manufacturer. The SpTLR3 5 -rapid amplification of cDNA ends (RACE) cDNA was prepared using the SMARTer ® RACE 5 /3 kit (TaKaRa, Dalian, China).

Tissue Distribution of TLR3 mRNA in Unstressed Conditions
The measured tissues in control group included heart, liver, spleen, head kidney, muscle, intraperitoneal fat, hindgut and gill. Total RNA was extracted and cDNA was synthesized, as described previously ("RNA extraction and cDNA synthesis"). The qRT-PCR analysis was performed using the applied Biosystems Step One Plus (Life Technologies, Foster City, CA, USA). The gene-specific primers are listed in Table 1. S. prenanti-specific β-actin primers were used to normalize the cDNA quantity for each tissue sample. Quantification of SpTLR3 and β-actin was performed in triplicate on all samples using FastStart Essential DNA Green Master (Roche Diagnostics, Risch-Rotkreuz, Switzerland), according to the manufacturer's instructions. The qRT-PCR data were calculated according to the 2 −∆∆CT method [32].

Detection of the Expression Patterns Induced by LPS
For the LPS challenge, 8 fish were injected intraperitoneally from the base of the pectoral fin with LPS (Sigma, L2880). The dose of injection was 10 mg/kg LPS per fish. Another 8 fish were injected intraperitoneally PBS at the same amount, as a control. 4 fish samples were taken at 12 and 24 h after LPS injection respectively. In generally, head kidney, liver, and hindgut of fish are regarded as immune organs and central to the immune responses [33], the three tissues plus spleen, another vital immune organ were collected from each fish and used to isolate total RNA. To detect the expression of SpMyD88, SpTLR2, SpTLR 3, SpTLR4, SpTLR18, SpTLR22-1, SpTLR22-2 and SpTLR22-3 changes after LPS induction, total RNA extraction, cDNA synthesis, and qRT-PCR were performed, as described previously (Tissue distribution of SpTLR3 mRNA in unstressed conditions). S. prenantispecific β-actin primers were used to normalize the cDNA quantity for each tissue sample. The quantitative primers of SpTLR2, SpTLR4, SpMyD88 were designed according to the transcriptome sequencing of S. prenanti. The quantitative primers of SpTLR22-1, SpTLR22-2 and SpTLR22-3 were designed according to the gene sequences unloaded to GenBank (Gen-Bank accesion no. MN082612, MN082613, MN082614). Quantitative primers of SpTLR18 refer to the article published by Li et al. [34]. The quantitative primers used are listed in Table 1.

Statistical Analysis
SPSS 22.0 and Graphpad Prism 5.0 software was used for data analysis and histogram, respectively. The mRNA expression levels were analyzed by using One-way ANOVA method. All data are presented as the mean ± standard error (n = 4), and the statistically significant differences between PBS control and LPS treatment groups at each time point are expressed with asterisks, * p < 0.05, ** p < 0.01, *** p < 0.001 vs. corresponding control group at the time points.

Identification, Structural, and Phylogenetic Analysis of SpTLR3
The  Figure 3a shows the distribution of strands, helixes, and coils, and Figure 3b shows the predicted 3D structure of the SpTLR3 protein. The SpTLR3 amino acid sequence was the most similar to those of other fish and was the closest to snow trout (Schizothorax richardsonii) TLR3, with 100% identity. We analyzed the phylogeny of the SpTLR3 amino acid sequences to determine the evolutionary relationships between SpTLR3 and TLR3 from other vertebrates, based on sequences in the GenBank database ( Figure 4). The results revealed a high amino acid sequence identity between SpTLR3 and the TLR3 of Cyprinids. Moreover, SpTLR3 and the TLR3 of S. richardsonii have the same amino acid sequence identity as the two fish belong to the same genus Schizothorax (Teleostei: Cyprinidae).

Tissue Distribution of SpTLR3 Expression in S. prenanti
We quantified SpTLR3 mRNA expression in the heart, liver, spleen, head kidney, muscle, intraperitoneal fat, hindgut and gill tissue from four fish using qRT-PCR to determine SpTLR3 transcript expression. The loading control for normalization was β-actin. Figure 5 shows the ubiquitous, but variable expression of SpTLR3 transcripts in all eight tissues. The highest expression level of SpTLR3 appeared in the liver, and the level was significantly higher than in all other tissues (p < 0.05). Furthermore, the expression of

Tissue Distribution of SpTLR3 Expression in S. prenanti
We quantified SpTLR3 mRNA expression in the heart, liver, spleen, head kidney, muscle, intraperitoneal fat, hindgut and gill tissue from four fish using qRT-PCR to determine SpTLR3 transcript expression. The loading control for normalization was β-actin. Figure 5 shows the ubiquitous, but variable expression of SpTLR3 transcripts in all eight tissues. The highest expression level of SpTLR3 appeared in the liver, and the level was significantly higher than in all other tissues (p < 0.05). Furthermore, the expression of SpTLR3 in the heart and muscle was more pronounced than that in the spleen, head kidney, intraperitoneal fat, hindgut and gill (p < 0.05). The expression SpTLR3 did not significantly differ in the spleen, head kidney, intraperitoneal fat, hindgut and gill tissues.
SpTLR3 in the heart and muscle was more pronounced than that in the spleen, head kidney, intraperitoneal fat, hindgut and gill (p < 0.05). The expression SpTLR3 did not significantly differ in the spleen, head kidney, intraperitoneal fat, hindgut and gill tissues.

Expression of SpMyD88 and SpTLRs after Injection of LPS
To determine the change in MyD88, TLR2, TLR3, TLR4, TLR18, TLR22-1, TLR22-2 and TLR22-3 from S. prenanti after LPS infection at 12 and 24 h, the mRNA levels of the genes in the liver, head kidney, spleen and hindgut were quantified using qRT-PCR.

Expression of SpMyD88 after Injection of LPS
The expression of SpMyD88 in the four tissues examined showed no significant change after the treatment of S. prenanti with LPS 12 h later (Figure 6a). However, at 24 h of LPS infection, the gene was significantly up-regulated in all four tissues, especially in the head kidney and spleen tissues, which showed an exponentially up-regulated trend compared with the PBS control group and the 12 h infection group (p < 0.001). Moreover, in the liver and hindgut, SpMyD88 of the 24 h infection group was significantly higher than that in the PBS control group and 12 h infection group (p < 0.01).

Expression of SpMyD88 and SpTLRs after Injection of LPS
To determine the change in MyD88, TLR2, TLR3, TLR4, TLR18, TLR22-1, TLR22-2 and TLR22-3 from S. prenanti after LPS infection at 12 and 24 h, the mRNA levels of the genes in the liver, head kidney, spleen and hindgut were quantified using qRT-PCR.

Expression of SpMyD88 after Injection of LPS
The expression of SpMyD88 in the four tissues examined showed no significant change after the treatment of S. prenanti with LPS 12 h later (Figure 6a). However, at 24 h of LPS infection, the gene was significantly up-regulated in all four tissues, especially in the head kidney and spleen tissues, which showed an exponentially up-regulated trend compared with the PBS control group and the 12 h infection group (p < 0.001). Moreover, in the liver and hindgut, SpMyD88 of the 24 h infection group was significantly higher than that in the PBS control group and 12 h infection group (p < 0.01).

Expression of SpTLR2 after Injection with LPS
At 12 h after LPS infection, SpTLR2 expression in all four tissues examined showed no significant change, which was similar to the expression pattern of SpMyD88. However, at 24 h after LPS challenge, the expression level of SpTLR2 in the liver was significantly increased (p < 0.05), whereas the other three tissues were extremely significant up-regulated (p < 0.001). In particular, the expression level of the gene was extremely high in the head kidney, followed by the hindgut and the liver, showing that the expression levels of the SpTLR2 gene in the tissues after infection were extremely different (Figure 6b). SpTLR18 (e), SpTLR22-1 (f), SpTLR22-2 (g) and SpTLR22-3 (h) transcripts in liver, head kidney, hindgut, and spleen of S. prenanti at 12 and 24 h after LPS injection determined by qRT-PCR. Values were normalized using β-actin. Data are expressed as the means ± standard error (n = 4). Statistically significant differences between the groups are indicated by asterisks (* p < 0.05, ** p < 0.01, *** p < 0.001). PBS, phosphate-buffered saline.

Expression of SpTLR2 after Injection with LPS
At 12 h after LPS infection, SpTLR2 expression in all four tissues examined showed no significant change, which was similar to the expression pattern of SpMyD88. However, at 24 h after LPS challenge, the expression level of SpTLR2 in the liver was significantly increased (p < 0.05), whereas the other three tissues were extremely significant up-regulated (p < 0.001). In particular, the expression level of the gene was extremely high in the head kidney, followed by the hindgut and the liver, showing that the expression levels of the SpTLR2 gene in the tissues after infection were extremely different (Figure 6b).

Expression Levels of SpTLR3 after Injection with LPS
After LPS treatment, the expression pattern of SpTLR3 in both the 12 h infection group and the 24 h infection group was similar to the pattern of SpTLR2; that is, compared with the PBS control group, the expression level of SpTLR3 in the LPS 12 h infection group showed no significant change in the four tissues, but the level in the 24 h infection group was significantly higher than that in the PBS and 12 h infection group (p < 0.001), and the highest expression level was found in the head kidney, followed by the spleen and hindgut (Figure 6c).

Expression of SpTLR4 after Injection with LPS
After LPS infection, SpTLR4 expression was different from that of the above genes. Compared with the PBS group, the SpTLR4 of the liver and hindgut in the 12 h infection group showed no significant change, while that of the head kidney tissue was significantly up-regulated (p < 0.001), and the high expression level persisted until 24 h after infection. The levels of the 24 h infection group were as follows: the gene expression of SpTLR4 in the liver was exponentially up-regulated and more strongly significant than that of the PBS group and 12 h infection group (p < 0.001). The level in the head kidney remained high in the 24 h infection group, which was significantly higher than that in the PBS group (p < 0.001). The level in the hindgut was significantly higher than that in the PBS control group and 12 h infection group (p < 0.05). The difference is that the expression level in the spleen was downregulated, with a highly significant difference in both the 12 h and 24 h infection groups compared to in the PBS control group (p < 0.05) (Figure 6d). SpTLR18 (e), SpTLR22-1 (f), SpTLR22-2 (g) and SpTLR22-3 (h) transcripts in liver, head kidney, hindgut, and spleen of S. prenanti at 12 and 24 h after LPS injection determined by qRT-PCR. Values were normalized using β-actin. Data are expressed as the means ± standard error (n = 4). Statistically significant differences between the groups are indicated by asterisks (* p < 0.05, ** p < 0.01, *** p < 0.001). PBS, phosphate-buffered saline.

Expression Levels of SpTLR3 after Injection with LPS
After LPS treatment, the expression pattern of SpTLR3 in both the 12 h infection group and the 24 h infection group was similar to the pattern of SpTLR2; that is, compared with the PBS control group, the expression level of SpTLR3 in the LPS 12 h infection group showed no significant change in the four tissues, but the level in the 24 h infection group was significantly higher than that in the PBS and 12 h infection group (p < 0.001), and the highest expression level was found in the head kidney, followed by the spleen and hindgut (Figure 6c).

Expression of SpTLR4 after Injection with LPS
After LPS infection, SpTLR4 expression was different from that of the above genes. Compared with the PBS group, the SpTLR4 of the liver and hindgut in the 12 h infection group showed no significant change, while that of the head kidney tissue was significantly up-regulated (p < 0.001), and the high expression level persisted until 24 h after infection. The levels of the 24 h infection group were as follows: the gene expression of SpTLR4 in the liver was exponentially up-regulated and more strongly significant than that of the PBS group and 12 h infection group (p < 0.001). The level in the head kidney remained high in the 24 h infection group, which was significantly higher than that in the PBS group (p < 0.001). The level in the hindgut was significantly higher than that in the PBS control group and 12 h infection group (p < 0.05). The difference is that the expression level in the spleen was downregulated, with a highly significant difference in both the 12 h and 24 h infection groups compared to in the PBS control group (p < 0.05) (Figure 6d).

Expression of SpTLR18 after Injection with LPS
The temporal expression pattern of SpTLR18 was similar to that of SpTLR2, SpTLR3 and SpMyD88; that is, no significant difference was found between the 12 h infection group and the PBS group. The SpTLR18 gene was overexpressed in the liver, head kidney and hindgut, with significant levels of p < 0.001. The level of SpTLR18 in the spleen was relatively low, but showed a significance level of p < 0.01 (Figure 6e).

Expression of SpTLR22s after Injection with LPS
As shown in Figure 6f-h, the temporal expression patterns of SpTLR22 genes (SpTLR22-1, SpTLR22-2 and SpTLR22-3) in the four organs examined after LPS stimulation were consistent. After 12 h, in the fish challenged with LPS, no significant changes were found compared with the PBS-injection group; however, the levels of SpTLR22-3 in the spleen were significantly up-regulated (p < 0.05). In contrast, the highest transcripts were found in the liver post-24 h LPS injection for all three SpTLR22s, followed by the levels of SpTLR22-1 and SpTLR22-2 in the hindgut, spleen and liver, and then by the levels of SpTLR22-3 in the liver, hindgut and spleen. Furthermore, the expression levels of all SpTLR22s in the four organs spiked after 24 h LPS stimulation, and the level of SpTLR22-1 in liver was significantly higher than that in the PBS control and 12 h groups (p < 0.01), whereas the levels of other SpTLR22s in different organs were much higher than those in the PBS and 12 h groups (p < 0.001).

Discussion
In this study, we cloned the SpTLR3 gene for the first time, The predicted SpTLR3 amino acid sequence included the conserved typical structure of TLR protein family: 17 LRR domain, Transmembrane domain and TIR domain. The previous studies have demonstrated that the LRR domain was related to recognizing the pathogen components and the number of LRR domain differs in different animals [17,35] For instance, 16 LRR domains were found in Japanese flounder [17]. In the present study, there were three conserved regions: Box 1, Box 2 and Box 3, indicating that the TIR domain is relatively conserved throughout TLR3 in fish. Box1 and Box2 play an important role in mediating receptor binding to downstream signaling molecules, while Box3 directly controls the expression and localization of receptor molecules [36]. Multiple sequence alignment showed that the SpTLR3 protein is moderately conserved, and that its amino acid sequence is highly similar to that of Cyprinids. The results of the SpTLR3 amino acid sequencing and phylogenetic tree analysis revealed that S. prenanti is more closely related to S. richardsonii. This was consistent with previous findings of SpTLR5-1 [28], whereas SpTLR5-2 [28] and SpTLR22s (SpTLR22-1, SpTLR22-2, and SpTLR22-3) [37] were more closely related to common carp (Cyprinus carpio). Additionally, S. prenanti apelin receptor (APJ) [38] and protein nucleobindin-2A (NUCB2A) [39] were more closely related to zebrafish (Danio rerio) and goldfish (Carassius auratus).
MyD88 is an important transmitter of the TLR signaling pathway, which is highly conserved and plays an indispensable role in both TLR signaling pathway and (Interleukin-1) IL-1 receptor family. It goes through a series of signal transductions after the MyD88depended pathway is activated by TLRs recognition, that lead to transcription factors such as Nuclear factor-κB (NF-κB), activator protein-1 (AP-1) and interferon regulatory factors (IRFs), eventually triggering inflammatory responses [14,40]. To date, few studies on fish MyD88 have been conducted. Takano et al. [41] found that the number of MyD88-positive cells in the kidney and spleen increased three days after Japanese flounder (Paralichthys olivaceus) was infected with Edwardsiella tarda. The yellow drum (Nibea albiflora) was treated with Pseudomonas plecoglossicida, then, the NaMyD88 in the head kidney was rapidly upregulated and remained significantly higher than that in the PBS control group from 2 to 24 h after infection, indicating that the gene plays an important role in the innate immunity of the fish [42]. These results are similar to that of SpMyD88 in our study. Moreover, when the yellow croaker (Pseudosciaena crocea) was infected with Vibrio parahaemolyticus, the expression level of spleen PcMyD88 was significantly higher at 24 and 48 h than that of the control group, and the highest expression level appeared at 48 h post-injection [43]. It can be concluded that MyD88 plays a crucial role in the innate immunity of fishes, whereas the time of immune response varies with different pathogens in different species.
TLR3 is the only member of the fish TLR3 subfamily [17], mainly involved in mediating antiviral immunity [18][19][20]. However, fish TLR3 can also senses bacterial stimulation. It was reported that TLR3 mRNA levels in the liver, kidney and spleen of channel catfish were significantly up-regulated at different time points after infection with Edwardsiella ictaluri [21]. However, there was no significant change in the TLR3 of zebrafish infected with Mycobacterium marinum [48], although the expression of TLR3 was significantly downregulated in Rainbow trout infected with Yersinia reuteri [54]. In the current study, the expression level of SpTLR3 increased exponentially in the head kidney, spleen and hindgut after 24 h LPS injection. In conclusion, the response of fish TLR3 to PAMPs may depend on many factors, such as time, host species and pathogens. Similar to TLR3, TLR4 is also the only member of the TLR4 subfamily of fish, and in mammals, TLR4 is the direct receptor of bacterial LPS. The biggest difference between TLR signaling pathway in fish and that in mammals lies in the TLR4-mediated signaling pathway [54]. Moreover, TLR4 is only present in a few fish [49]. Zebrafish TLR4 has been demonstrated not only to fail to recognize LPS challenge in vitro, but also have no responsive to heat-killed Escherichia coli and Legionella pneumophila [55,56]. Moreover, TLR4 negatively regulated NF-κB activation; therefore, the LPS recognition system in fish is quite different from that in mammals [46,56]. In contrast, it was reported that both LPS and A. hydrophila (a Gram-negative fish pathogen) infection could significantly up-regulate the expression of TLR4 from rohu (Labeo rohita) [57]. Our study confirmed the following three conclusions. First, the TLR4 in S. prenanti is present. Second, the SpTLR4 is involved in anti-LPS immunity, which is consistent with the conclusion of Samanta et al. [57], but different from that of Sullivan et al. [55] and Sepulcre et al. [56]. Finally, the liver is the most sensitive organ for SpTLR4 detection. The expression level of SpTLR4 was the least up-regulated among the eight genes examined, which was in fact the most different SpTLRs in our study. It was the only one that up-regulated at 12 h in head kidney. Those most different results verified the specificity of fish TLR4, and the immune interpretation need further study.
In general, the expression levels of the SpTLR22s were the highest in the seven induced SpTLRs and SpMyD88 examined in the present study. TLR22, which was also one of the TLRs specific to fish, belonging to the TLR11 subfamily, first discovered in goldfish in 2003, The TLR22 of goldfish macrophages was highly up-regulated by LPS, heat-killed Aeromonas salmonicida, and live Mycobacterium chelonei [58]. Subsequently, TLR22 has been cloned and reported in nearly 20 fish species, such as zebrafish [49], Japanese flounder [59], Atlantic salmon [60], yellowtail (Seriola lalandi) [61], and common carp [62]. To date, most of the studies reporting on TLR22 function have focused on bacteria, viruses, parasite infection and PAMP stimulation. It has been found that TLR22 is a multi-functional immune receptor that participates in almost all the defensive immune responses of pathogenic microorganisms. The results of SpTLR22s in this study also confirmed these conclusions. Moreover, two subtypes of TLR22 (namely, TLR22-1 and TLR22-2) were first discovered in Rainbow trout in 2007, which have very similar functions and were called 'twin' TLRs, and the expression of the two OmTLR22s in peripheral leukocytes was significantly upregulated after stimulation by A. salmonicida [63]. Subsequently, three copies of the TLR22 gene (namely, TLR22-1, TLR22-2 and TLR22-3) were found in Atlantic salmon (GenBank accession numbers: AM233509, FM206383 and BT045774), and even up to twelve TLR22 gene copies (namely, TLR22a~l) have been found in Atlantic cod (Gadus morhua) [64]. Du et al. reported that both SpTLR22-1 and SpTLR22-3 in the head kidney and spleen respond to the stimulation of LPS and A. hydrophila, but not SpTLR22-2, in the two immune tissues [14]; this was different from our findings, as we found that the SpTLR22-2 levels in the four organs were significantly up-regulated at 24 h after LPS stimulation. Moreover, SpTLR22-1 in the head kidney was significantly up-regulated at 12 h after LPS challenge, but no pronounced significance was found at the 24 h time point [14], showing a different temporal expression pattern to our findings, in addition to the expression in the spleen. Notably, the temporal expressions of SpTLR22s in our study showed similar kinetics to one another in different immune organs, indicating that the functions of the subtypes of SpTLR22s are highly similar. Our results may indicate that both the SpTLR22s jointly mediate the recognition of LPS and participate in the immune response.

Conclusions
In this study, SpTLR3 was cloned and identified for the first time, and the secondary structure composition, the 3D structural models of the SpTLR3 protein, as well as the binding site were predicted. Furthermore, the SpTLR3 gene was expressed in all the tissues examined and mainly expressed in immune-related tissues. Phylogenetic analysis showed that the SpTLR3 protein is most closely related to TLR3 from snow trout. Multiple sequence alignment showed that the SpTLR3 is moderately conserved. Among the seven immune tissues we examined, the distribution of SpTLR3 was the most abundant in the liver, followed by the kidney and spleen. Moreover, the LPS organ specifically induced SpTLR2, SpTLR3, SpTLR4, SpTLR18, SpTLR22s (SpTLR22-1, SpTLR22-2, and SpTLR22-3) and SpMyD88 at 12 or 24 h, which are involved in antibacterial immunity. It is worth noting that the SpTLR22s were the most sensitive to LPS induction.