3.1 The quality inspection results of all samples
The scRNA-Seq data of different samples of hemocytes from H. discus Hanna were subjected to quality control and preliminary statistical analysis using Cell Ranger(Table S2). The V_group, N_group, NN_group, NV_group and VV_group libraries obtained 396,266,876, 414,300,518, 447,684,335, 421,542,139 and 368,963,729 raw reads respectively. All Q30 values were above 90.50%, indicating good quality of the sequencing data. BioProject accession number PRJNA979786 has been assigned to all raw reads submitted to the NCBI Short Read Archive database.
Owing to the deficient quality of the published abalone genome assembly, this investigation employed the full-length transcriptome of H. discus hannai as a reference sequence for aligning and annotating the obtained sequences 58. Ultimately, the number of high-quality cells captured in the V_group, N_group, NN_group, NV_group, and VV_group samples, which can be utilized for subsequent analyses, were 9,483, 9,547, 8,702, 8,947, and 6,200, respectively (Table S3, FigureS1). The number of genes detected in each treatment group surpassed 30,000, and according to the results of full-length transcriptome data alignment, the alignment rate of the data extracted from the five samples was higher than 66.7% (Table S3).
3.2 Cluster analysis of hemocytes of abalone H. discus hannai
The Seurat package was utilized to amalgamate and scrutinize multiple filtered datasets. Following the implementation of UMAP dimensionality visualization analysis and t-SNE clustering, a total of fifteen clusters (cluster_0–14) (Fig. 2A-B) have been identified. Each dot on the graph signifies a unique cell, distinguished by color based on subgroup classification. Moreover, clustering outcomes indicate considerable differences in cell quantities between different clusters, which suggests that functional variations exist among each cluster (Fig. 2C).
In addition, in order to further scrutinize the functional disparities among different clusters of resting hemocytes in H. discus hannai, we have conducted comparative analyses on the differential genes between various hemocyte clusters in the N_group sample. The results have revealed that crucial immune-regulating genes such as allograft inflammatory factor 1 (AIF1), an inhibitor of NF-κB (NFKBIA), CD63, and some members of the caspase gene family alongside certain heat shock proteins (HSPs), are significantly upregulated in different hemocyte clusters (Fig. 2D).
3.3 Identification of cell clusters
Due to the late start of research on marine invertebrates, insufficient information on marker genes for different cell types, coupled with poor genome assembly quality, resulting in a large number of omissions 65, it is difficult to identify different blood cell clusters with marker genes of H. discus hannai. Notwithstanding this, there are similarities in the expression patterns of differential genes within the different hemocyte clusters, hinting at them being discrete components of a certain type of cell. In light of this, we undertook KEGG enrichment analysis on the upregulated genes in distinct cell clusters (Fig. 3A). By combining the outcomes of functional enrichment analysis of diverse clusters and consulting existing studies on the functions of different hemocyte types 45,51, we further classified the original 15 clusters into three cell types: hyalinocytes (cluster_1, cluster_4, cluster_7), semi-granulocytes (cluster_2, cluster_5, cluster_6, cluster_8, cluster_9, cluster_13, cluster_14), and granulocytes (cluster_0, cluster_3, cluster_10, cluster_11, cluster_12). Granulocytes constitute a proportion of 60.12% (in which GRCs amount to 28.13% of the total cells and SGRCs account for 31.99%)(Fig. 3B). Hemocytes (HCs) make up roughly 39.88%, which aligns with previously documented findings. The distribution of diverse cellular types of hemocytes in H. diversicolor was identified through flow cytometry 49,50. Furthermore, the marker gene was identified in conjunction with varying gene expression patterns among different types of cells. The results revealed that allograft inflammatory factor 1 (AIF1), Cell Division Cycle 42 (CDC42), matrix metalloproteinase-18 (MMP18), and CD63 exhibited significantly high expressions in GRCs cells. In contrast, Thioredoxin-2 (TRX2), glutathione s-transferase 7 (GST7), and caspase-3 (CASP3) showed substantial overexpression in SGRCs cells. For HCs, genes such as complement C3-like (C3), proliferating cell nuclear antigen (PCNA), and histone H2A-beta (HIS2A) were highly expressed (Fig. 3C-D).
3.4 Functional analysis of marker gene in different cell clusters
Fifteen hemocyte clusters from all samples were re-clustered using the aforementioned classification method. Statistical results suggest that there was no significant alteration in the overall clustering results of three cell types (GRCs, SGRCs and HCs) among different treatment groups of V. parahaemolyticus. Still, the number of cells of different types varied (Fig. 4A). After exposure to distinct treatments with V. parahaemolyticus, the number of granulocytes in V_group, NV_group and VV_group samples changed to varying degrees compared to 60.12% in the N_group, with the most substantial increase observed in V_group, where the proportion of granulocytes rose to 74.53%. In contrast, proportions in NV_group and VV_group increased to 67.88% and 69.61%, respectively. Moreover, the proportion of HCs in the three treatment groups decreased relative to the control group (Fig. 4A).
The identification of marker genes was performed on various samples after re-clustering. In V_group, significantly upregulated GRCs marker genes such as AIF1, AAH2, IP6K1, CPNE8 and ALOX5 were identified; SGRCs marker genes included TRX2, GIMAP9, GST7, CASP3 and PRDX6; and HCs marker genes comprised of NIP7, C3, DKC1, FKBP46 and DDX51(Fig. 4B). In NV_group, significantly upregulated GRCs marker genes included AIF1, CDC42, SQSTM1, PSTPIP1 and NOCT; SGRCs marker genes comprised of TRX2, LRP1, GST7, FSTL3 and PFE; and HCs marker genes included DDX21, C3, SLC6A9, Nop2 and HIS2A(Fig. 4C). In VV_group, significantly upregulated GRCs marker genes were AIF1, CDC42, ELF3, NFKB1 and TLR4; SGRCs marker genes included LCP1, PDIK1B, GS2, FSTL3 and PFE; and HCs marker genes included FKBP46, C3, SLC6A9, RELN and PCNA (Fig. 4D). Hemocytes represent the primary immune regulatory organs in invertebrates. Various hemocyte types also respond differently to external stimuli. Marker genes of diverse hemocyte types function as labels for identifying them, characterized by stable high expression characteristics that should not vary substantially with changes in the external environment. Therefore, we comprehensively analyzed relevant genes identified in the aforementioned different samples and coupled them with results obtained in the above study from marker gene screening under non-stress conditions. We determined some marker genes that can be stably expressed at a high level in various cell types in different treatment samples, including AIF1 and CDC42 in GRCs, TRX2 and GST7 in SGRCs, and C3 in HCs. The expression heatmap of these genes is depicted in Fig. 4E. Additionally, the t-SNE graph represents their expression patterns in different cells (Fig. 4F).
3.5 Analysis of weighted gene co-expression network
In order to further investigate the important regulatory mechanisms of GRCs in the immune modulation process of H. discus hannai in a resting state, we conducted WGCNA on 5 clusters of GRCs belonging to N_group and ultimately identified 8 clustering modules with significant differences in gene expression patterns that were allocated to different clusters comprising GRCs (cluster_0, cluster_3, cluster_10, cluster_11, and cluster_12). Among these modules, the turquoise and blue modules exhibited a higher level of gene enrichment, which consisted of 6,475 and 4,341 genes, respectively (Fig. 5A-B). In addition, a heatmap of gene clustering was generated for different modules to visualize the expression matrix of related genes between modules (Fig. 5C). The correlation analysis results showed a significant positive correlation (p < 0.05) between the brown, blue, and yellow modules (Fig. 5D), with high expression levels observed in cluster_11 (Fig. 5E).
The brown, blue, yellow, and red modules were selected as target modules to explore the functional differences of GRCs further, and KEGG enrichment analysis was carried out on the genes grouped in these modules. The results indicated that immune-related signaling pathways such as NF-κB signaling pathway, Endocytosis, Toll and Imd signaling pathway, NLR signaling pathway, CLR signaling pathway, and Fc gamma R-mediated phagocytosis were significantly enriched in the brown module. Energy metabolism-related signaling pathways, such as Oxidative phosphorylation and Thermogenesis, were significantly enriched in the red module. Moreover, some immune-related signaling pathways, such as the PI3K-Akt signaling pathway, IL-17 signaling pathway, and Phagosome, also showed significant enrichment (p < 0.05) in the red module (Fig. 5F).
Hub genes are pivotal genes with significant roles in biological processes and can be identified within different modules based on their K.in values that indicate the level of connectivity and regulatory function of the gene within the module. The higher the K.in value of a gene, the greater its connectivity and the more central its regulatory function. Based on this theory, 5–6 genes with high K.in values were selected as hub genes for each target module, and a molecular interaction network was constructed using Cytoscape to visualize the relationship between these hub genes and their associated genes (Fig. 6A-D).
The results revealed that calcium-binding protein genes (CALM), which play roles in signal transduction and transcriptional regulation, ubiquitin-conjugating enzyme genes (EFF) responsible for ubiquitination regulation, and toll-like receptor genes (TLR3) were identified as hub genes in the blue module(Fig. 6A). In the related network of the brown module, the core transcription factor NF-κB of the NF-κB signaling pathway and immune-related genes such as GST, Tollo, and Perlucin were identified as hub genes (Fig. 6B). For the yellow module, signal transduction-related genes such as CAP-Gly domain-containing linker protein 1-like (CLIP1) and phosphatidylinositol 3-kinase regulatory subunit alpha (PIK3R1), as well as genes involved in intracellular engulfment such as SMURF2 were identified as hub genes (Fig. 6C). Key immunoregulatory factors such as CD109, MMP18, and HSP90 were also identified as hub genes in the red module (Fig. 6D).
3.6 Response of different cell types to V. parahaemolyticusinfection
The results of KEGG enrichment analysis of upregulated genes of different hemocyte types in different treatment groups after V parahaemolyticus infection showed that in the V_group, signaling pathways associated with energy metabolism, such as oxidative phosphorylation, were notably enriched in both SGRCs and HCs; however, HCs possessed a higher number of enriched genes. Conversely, signaling pathways involved in energy metabolism, such as thermogenesis and protein processing in the endoplasmic reticulum, were solely significantly enriched in HCs. In addition, the phagocytosis-related signaling pathway, Phagosome, was significantly enriched in both GRCs and HCs. Contrarily, the commonly enriched signaling pathways in SGRCs and GRCs were primarily related to immunity, such as the NLR signaling pathway, consistent with the biological functions of granulocytes. Meanwhile, signaling pathways involved in immune regulation, such as Endocytosis and IL-17 signaling pathways, were exclusively enriched in GRCs (Fig. 7A). In the NV_group, signaling pathways related to energy metabolism and phagocytosis, such as oxidation phosphorylation and Phagosome, were significantly enriched in SGRCs and HCs. Certain signaling pathways involved in transcription, translation, folding, and degradation processes, such as Ribosome, Proteasome, and Spliceosome, were specifically enriched in HCs. In SCRGs, signaling pathways involved in signal transduction, such as the Hippo signaling pathway and Rap1 signaling pathway, were specifically enriched.
In contrast, signaling pathways specifically enriched in GRCs, those involved in immune regulation, such as IL-17 signaling pathway, NLR signaling pathway, RLR signaling pathway, TLR signaling pathway, Toll and Imd signaling pathway, were all prominently present (Fig. 7B). Compared to the previously mentioned two groups, there were no signal pathways commonly enriched in HCs of VV_group with the other two cell types—the significantly enriched pathways in HCs primarily involved translation, folding, and degradation processes.
Additionally, signal pathways associated with energy metabolism, such as oxidative phosphorylation and Thermogenesis, were specifically enriched in HCs. Signaling pathways involved in the degradation and metabolism of exogenous substances, such as the Metabolism of xenobiotics by cytochrome P450 and Phagosome, were significantly enriched in SGRCs. GRCs displayed a significant increase in the number of enriched signal pathways compared to HCs and SGRCs, with most of the genes significantly enriched in signaling pathways involved in signal transduction such as MAPK signaling pathway, NF-κB signaling pathway, TNF signaling pathway, signal pathways involved in degradation such as Endocytosis and Phagosome, and signal pathways involved in immune regulation such as NLR signaling pathway, RLR signaling pathway, TLR signaling pathway, CLR signaling pathway, Toll and Imd signaling pathway, IL-17 signaling pathway, and Fc gamma R-mediated phagocytosis (Fig. 7C).
3.7 Response of GRCs to V. parahaemolyticusinfection
In the previous study, we have functionally annotated and defined the gene sets that may be involved in the immune regulation response of abalone hemocytes to V. parahaemolyticus infection. These gene sets include immune response genes (IRGs), potential immune-enhancing genes (PEGs), immune-enhancing regulatory genes (ERGs), and essential immune-enhancing genes (EEGs) 56. Our analysis found that GRCs from different treatment groups were enriched with numerous immune-related signaling pathways, suggesting their critical regulatory role in abalone's response to V. parahaemolyticus infection. Therefore, we employed the previous gene set screening approach to identify DEGs in GRCs from different comparison groups and performed KEGG functional enrichment analysis. The results indicated that 801 DEGs in GRCs responded to both V. parahaemolyticus infections simultaneously, defining them as co-response genes of GRCs (CRGs-b) (Fig. 8A), whereas 489 common DEGs were obtained from the comparison groups of NN_group vs NV_group, V_group vs VV_group, and N_group vs V_group, which might relate to GRCs' faster immune response after secondary V. parahaemolyticus infection and were designated as immune response genes of GRCs (IRGs-b) (Fig. 8B). Moreover, by comparing 801 CRGs-b with the putative immune-enhancing gene set, we identified 361 common DEGs that were classified as potential immune-enhancing genes of GRCs (PEGs-b) (Fig. 8C). Finally, 246 immune-enhancing genes of GRCs (ERGs-b) were identified via comparison of IRGs-b and PEGs-b using the Venn diagram, which may possess specific immune memory regulatory functions. The KEGG enrichment results revealed that 801 CRGs-b mainly enriched immune-related signaling pathways such as the NF-κB signaling pathway, IL-17 signaling pathway, TLR signaling pathway, and NLR signaling pathway (Fig. 8A), and 361 PEGs-b were primarily enriched in phagocytosis-related pathways, such as Fc gamma R-mediated phagocytosis and Phagosome (Fig. 8B). The 489 IRGs-b were significantly enriched in immune-related pathways, including the TLR signaling pathway, NF-κB signaling pathway, and Fc gamma R-mediated phagocytosis (Fig. 8C). In addition, the 246 ERGs-b were significantly enriched in phagocytosis-related pathways such as Fc gamma R-mediated phagocytosis and Phagosome, as well as apoptosis-related pathways like Apoptosis (Fig. 8D).
To reveal the molecular interactions among the DEGs in the significantly enriched pathways, we used Cytoscape software to construct network diagrams of some key signaling pathways in abalone GRCs (Fig. 9). As shown in the figure, as important signaling pathways involved in abalone innate immune regulation, NF-κB signaling pathway, TLR signaling pathway and NLR signaling pathway have complex molecular interactions among them, and among them, IRAK4 and NFKB1 act as core factors and participate in the regulation of these three signaling pathways (Fig. 9). At the same time, there is also an interaction between NF-κB signaling pathway and Fc gamma R-mediated phagocytosis (Fig. 9).
3.8 Pseudo-temporal analysis of GRCs stimulated by V. parahaemolyticus
Monocle2 constructed a developmental trajectory in pseudo-time to study the differentiation pattern of GRCs in samples stimulated by V. parahaemolyticus and control samples. The results indicate that the hemocytes of H. discus hannai infected by V. parahaemolyticus (V_group, NV_group, and VV_group) and those treated with normal saline (N_group and NN_group) were evidently differentiating along distinct branches of the pseudo-time trajectory, exhibiting three distinct differentiation states (Fig. 10A-C). Based on cell trajectory analysis, we defined the concentrated distribution of state_3 in the GRCs of H. discus hannai hemocytes treated with normal saline as the starting point of differentiation. After infection with V. parahaemolyticus, these cells could differentiate into state_1 and state_2, respectively (Fig. 10B). Furthermore, all three V_group, NV_group, and VV_group samples were observed to be distributed among three different differentiation states, mainly concentrated in state_1 and state_2 (Fig. 10C).
The expression heat map of differentially differentiated genes revealed an increase in the expression of granulocyte marker genes such as AIF1, MMP18, and TRX2 in Gene_cluster2 with deepening cell differentiation. However, highly expressed genes in granulocytes, such as CD63 and CDC42, gradually decreased in Gene_cluster1 with the deepening of cell differentiation (Fig. 10D). As state_1 and state_2 were two directions of granulocyte differentiation with obvious differentiation boundaries with state_3, responding to different stages of V. parahaemolyticus infection, we analyzed the upregulated genes in state_3 and state 1_2, respectively. The results showed that granulocyte high-expression genes such as AIF1, GST7, CDC42, and TRX2 were upregulated in state1_2. The important regulatory factors of immune signaling pathways, such as MAPK14 and MyD88, were also significantly upregulated (Fig. 10A-E). KEGG enrichment results of these upregulated genes showed significant enrichment of energy metabolism-related pathways, such as Oxidative phosphorylation and Thermogenesis, in state_3. At the same time, the IL-17 signaling pathway, NLR signaling pathway, TLR signaling pathway, and other pathways involved in innate immune regulation were significantly enriched in state1_2. Moreover, Fc gamma R-mediated phagocytosis pathways involved in phagocytosis were significantly enriched by differential genes in state1_2 and state_3, respectively (Fig. 10F).
3.9 Re-clustering of GRCs
The outcomes of pseudo-temporal analysis disclose that despite being the same type of cells, there exist dissimilarities in functional differentiation and also that the differentiation status of the identical cell type differs in diverse physiological states. The KEGG enrichment results revealed significant distinctions in signaling pathways of notable enrichment in varying differentiation conditions. All of these findings indicate that GRCs themselves are somewhat heterogeneous. It has been previously reported in bivalve-related research that there are three different developmental conditions in GRCs 66. Through re-clustering analysis of GRCs from H. discus hannai hemocytes, further examination of the heterogeneity of the same cell type and the similarities and variances of cellular functions at the molecular level is anticipated. In this study, using the subgrouping technique for GRCs from H. discus hannai, three separate subgroups (Sub-cluster_0–2) displaying elevated cell segregation between GRCs were identified (Fig. 11A). Analyzing the existing state of every subgroup in dissimilar treatment groups showed that two subsets, Sub-cluster_0 and Sub-cluster_1, coexisted in different treatment groups simultaneously. However, Sub-cluster_0 mainly existed in V. parahaemolyticus stimulated samples and had a greater number of cells in it, while Sub-cluster_1 predominantly contained more cells in the control group and accounted for less in different experimental groups. Notably, Sub-cluster_2 primarily existed in samples without infection of V. parahaemolyticus (Fig. 11B). The findings of differential gene analysis demonstrated that 1,784 DEGs were significantly upregulated in all three subclusters (p < 0.05, | log2FC | ≥ 0.585). The heatmap of gene expression selected from each subpopulation indicates a significant difference, as depicted in Fig. 11C. The key marker genes of GRCs, namely AIF1 and CDC42, and immune-regulation-related genes, such as MyD88 and IRAK4, were notably upregulated in Sub-cluster_0, while significant genes like CD109, MMP18, CASP3, and A2ML1 were upregulated in Sub-cluster_1 and Sub-cluster_2, respectively, which aligns with the overall expression pattern. There is a certain intersection between Sub-cluster_1 and Sub-cluster_2, implying some similarities in the biological functions of the two subclusters (Fig. 11C). The KEGG enrichment results of each subcluster showed that there are several mutual signaling pathways in Sub-cluster_1 and Sub-cluster_2, including oxidative phosphorylation involved in energy metabolism and Fc gamma R-mediated phagocytosis involved in cell phagocytosis, further indicating the functional similarity between the two subclusters (Fig. 11D). In Sub-cluster_0, most of the signaling pathways are specifically enriched, including various immune-related signaling pathways such as the IL-17 signaling pathway, NLR signaling pathway, RLR signaling pathway, TLR signaling pathway, and Toll and Imd signaling pathway (Fig. 11D).
Finally, we investigated the quantity and function of differential genes belonging to the same sub-cluster under varying V. parahaemolyticus treatments. Sub-cluster_2 was absent in some treatment groups, so our analysis only focused on Sub-cluster_0 and Sub-cluster_1. Within Sub-cluster_0, a total of 811 DEGs were upregulated when stimulated with high dosages of V. parahaemolyticus (V_group vs NV_group). These genes were primarily associated with immune signaling pathways such as NLR signaling pathway, RLR signaling pathway, TLR signaling pathway, CLR signaling pathway, and NF- κB signaling pathway, as well as phagocytosis-related pathways including Fc gamma R-mediated phagocytosis, and energy metabolism-related pathways such as Oxidative phosphorylation and Thermogenesis (Fig. S2 A). In the hypothetical immune enhancement gene cluster (VV_group vs NV_group), we identified 660 upregulated DEGs that were mainly involved in phagocytosis-related pathways, such as Phagosome and Fc gamma R-mediated phagocytosis, and energy metabolism-related pathways such as Oxidative phosphorylation and Thermogenesis (Fig. S2 B). Additionally, upon re-infection with V. parahaemolyticus (V_group vs VV_group), 660 DEGs were detected, showing significant upregulation; these genes were primarily linked to immune signaling pathways such as NLR signaling pathway, NF-κB signaling pathway, RLR signaling pathway, TLR signaling pathway, and IL-17 signaling pathway, phagocytosis-related pathways such as Fc gamma R-mediated phagocytosis and Autophagy, and energy metabolism-related pathways such as Oxidative phosphorylation and Thermogenesis (Fig. S2 C).
In Sub-cluster_1, a total of 245 DEGs showed up-regulation upon infection with high dosages of V. parahaemolyticus (V_group vs NV_group), and these genes were primarily associated with phagocytosis-related pathways, including Endocytosis, as well as the synthesis of amino acids, such as Biosynthesis of amino acids (Fig. S3 A). Moreover, in the hypothetically assumed immune enhancement gene cluster (VV_group vs NV_group), we detected 468 upregulated DEGs that were primarily enriched in phagocytosis-related pathways, such as Endocytosis, apoptosis-related pathways like Apoptosis, and immune-related signaling pathways, including NLR signaling pathway (Fig. S3B). Finally, upon further infection of the organism with V. parahaemolyticus (V_group vs VV_group), 613 DEGs were found to be significantly upregulated, and these genes were mainly enriched in immune signaling pathways such as Toll and Imd signaling pathway, NF-κB signaling pathway, NLR signaling pathway, RLR signaling pathway, TLR signaling pathway, and IL-17 signaling pathway, phagocytosis-related pathways such as Fc gamma R-mediated phagocytosis, as well as apoptosis-related pathways such as Apoptosis (Fig. S3C).
3.10 Cell communication of different clusters
By selecting receptors and ligands from various cell clusters across multiple samples, we created molecular interaction networks among these cell clusters in diverse samples via the utilization of Cytoscape software, ultimately depicting intricate interactions between these cell clusters (Fig. 12A). Although three different cell types were involved, overall cellular communication between clusters 5, 6, 11, 12, and 14 was found to be stronger compared to other clusters (Fig. 12A-D). The expression abundance heatmap of ligand-receptor pairs among subpopulations in different treatment groups is illustrated in Fig. 11E-H. Ultimately, 74 interacting receptor-ligand pairs were identified and screened from the 15 hemocyte clusters detected in disparate samples (Table S4).
Upon analysis of all ligand and receptor genes, it was revealed that 15 correlated genes were present in each sample. Additionally, the maximum number of ligand genes were observed to be screened in VV_group(Fig. 12I). The GO enrichment analysis of all receptor-ligand genes showcased that transmembrane receptor activity, signaling receptor activity, and receptor activity were meaningfully enriched across different samples, with receptor binding only displaying significant enrichment in VV_group. Notably, the enrichment degree of cell communication in the NV_group was lower in comparison to other groups, whereas the enrichment degree of receptor agonist activity, receptor activator activity, and receptor regulator activity was higher (Fig. 12J). KEGG enrichment results illustrated that the primary signaling pathways relevant to H. discus hannai hemocyte communication included Cytokine-cytokine receptor interaction, Axon guidance, Wnt signaling pathway, mTOR signaling pathway, PI3K-Akt signaling pathway, and Melanogenesis (Fig. 12K).
To analyze the communication relationship between GRCs and SGRCs, cluster_11, possessing the strongest correlation among all cell clusters from the cell interaction network diagram presented in Fig. 12, was utilized as the ligand signaling cell, while cluster_14, having the strongest communication relationship with it, acted as the receptor signaling cell. Four classic signaling pathways were selected: PI3K-Akt signaling pathway and NF-κB signaling pathway, which relate to immune signaling transmission and cell-based regulation, and NLR signaling pathway and TLR signaling pathway, found to be enriched multiple times in prior analysis and involved in innate immune regulation. Ligand activity assays were subsequently performed on ligand-signaling cells using specific gene sets in receptor-signaling cells. There were some differences in the number of detected ligands within different treatment groups, with the smallest quantity observed in N_group and the greatest in VV_group; overall, the ligand activity in VV_group was generally robust (Fig. S4). Under various treatments, ligands regulating the PI3K-Akt signaling pathway gene set were highly active, while those regulating the NF-κB signaling pathway gene set had a low activity rate in N_group and V_group. Following exposure to high dosages of V. parahaemolyticus, viability was enhanced, generally displaying higher viability within VV_group(Fig. S4). In addition, ligands regulating the NLR signaling pathway and TLR signaling pathway gene sets displayed the most activity in VV_group(Fig. S4).
In addition, the NicheNet software was utilized to screen ligand-target gene pairs in different target cell pairs for the aforementioned four signaling pathway gene sets and score the regulatory potential of the ligand-target genes. Based on the score, a ligand-target gene regulatory potential heatmap was created (Fig. S5). Results displayed that under identical screening conditions, the number of ligand-regulated target genes was relatively small in both N_group and NV_group, while the largest number of ligand-regulated target genes occurred within the VV_group(Fig. S5). In terms of regulatory potential, HSP90 exhibited the strongest regulatory activity on TLR2, which existed across all four treatment groups. Apart from HSP90's potent regulatory effect on TLR2, the regulatory activities of all ligands on target genes within the PI3K-Akt signaling pathway were weak as a whole, while the regulatory activities on target genes in the NF-κB signaling pathway, NLR signaling pathway, and TLR signaling pathway were comparative across various treatments (Fig. S5).
3.11 Effect of dsRNA exposure assay for NFκB and TLR2 gene expression
The expression of genes associated with the TLR signaling pathway and NF-κB signaling pathway in hemocytes after TLR2 silencing by dsRNA was assessed by qPCR. The results indicated that TLR2 gene expression in the experimental group was substantially lowered at all time points relative to the control group (p < 0.05) (Fig. 13A). Similarly, other genes in the pathway: HIF1A, NFκB, EIF4E, FADD, TRAF6, IRAK4, MyD88, CASP8, Akirin2, 14-3-3ζ, MKK4, RIP1, and MAPK14 demonstrated significant downregulation at different time points in the experimental group compared with the control group (p < 0.05) (Fig. 13A). In particular, TRAF6, IRAK4, NFκB, FADD and MyD88 showed robust interference effects and were persistently downregulated at each time point of the experiment (p < 0.05) (Fig. 13A). The expression of genes related to the two target pathways in blood lymphocytes after NFκB was interfered with by dsRNA was further detected by qPCR. The results showed that compared with the control group, the expression level of the NFκB gene in the experimental group was significantly down-regulated at different time points (p < 0.05) (Fig. 13B). At the same time, other related genes in the pathway: HIF1A, EIF4E, Akirin2, 14-3-3ζ, MKK4, and MAPK14 were also significantly down-regulated to varying degrees at different time points in the experimental group compared with the control group (p < 0.05) (Fig. 13B), while the expression patterns of TLR2, FADD, TRAF6, IRAK4, MyD88, CASP8, and RIP1 did not change due to NFκB gene interference (Fig. 13B).