Two-year-old oysters were acquired from an aquaculture farm in Dalian and acclimatized for one week in the laboratory. Oysters were cultured at 15-20 °C in aerated seawater filtered through a 300−mesh nylon sieve. A 50% daily water exchange was performed, and oysters were fed with spirulina powder thoroughly dissolved in seawater. The V. splendidus strain maintained in our laboratory was cultured by inoculating into 2216E liquid medium. All procedures followed ethical guidelines approved by the Dalian Ocean University Ethics Committee.

To elucidate the potential molecular mechanisms for immune priming, which is characterized by elevated haemocyte counts, miRNA transcriptome sequencing of haemocytes was performed. Based on our previous study, 36 oysters were randomly assigned into four groups designated as the SS, SV, VS and VV groups (Figure 1A). In the SS and SV groups, the oysters received a first injection with 100 µL of sterile seawater, and a secondary injection with 100 µL of sterile seawater (SS) or a diluted suspension of live V. splendidus (4.0×10⁸ CFU mL⁻¹) (SV) at 7 d after the first injection, respectively. In the VS and VV groups, the oysters were first stimulated with 100 µL of heat-killed V. splendidus (4.0×10⁸ CFU mL⁻¹), and then treated with 100 µL of sterile seawater (VS) or 100 µL of live V. splendidus (VV) as the second stimulation at 7 d after the first injection, respectively. To account for the stress sensitivity of haemocytes, two times of injection with 100 µL of sterile seawater in SS group, and one injection with 100 µL of sterile seawater in SV or VS were designed. The use of inactivated bacteria, analogous to inactivated vaccines in vertebrates (8), enables safe delivery of the complete antigenic profile and induces an initial immune response. The time span of seven days was employed based on the previous reports in both vertebrates (21) and invertebrates (22, 23). For each group, haemocytes from three oysters were pooled at 6 h following the second V. splendidus injection. All samples were stored at -80 °C for subsequent RNA extraction and library preparation.

Following extraction with TRIzol reagent, total RNA was quantified and qualified by using a NanoDrop 2000 spectrophotometer (Thermo Fisher Scientific, America). Small RNA libraries were constructed and subsequently sequenced on the Illumina sequencing platform by Genedenovo Biotechnology (Guangzhou, China). The raw sequencing data were processed through a stringent filtering pipeline to obtain high-quality small RNA data. First, reads without 3’ adapters (3’ adapter-null) were removed. Subsequently, low-quality reads were filtered out by eliminating those containing more than one base with quality scores below 20 or any ambiguous ‘N’ bases, resulting in high-quality reads. Further filtration removed reads without insert fragments or with insert lengths shorter than 18 nucleotides. Additionally, reads containing poly A sequences (>70% of adenines bases) were excluded. The remaining high-confidence small RNA (sRNAs) sequences, termed clean sRNAs, were then used for downstream analyses. Sequence filtering was performed using in−house Perl scripts provided by Genedenovo Biotechnology (https://github.com/apanhui/miRNA_analysis_2). To remove known non-coding RNAs (rRNA, snRNA, snoRNA, tRNA), clean tags were filtered by alignment against the GenBank (Release 209.0) and Rfam (Release 211.0) databases using blastall (v2.2.25). Reads were mapped to the C. gigas genome (GCF_000297895.1) and aligned to MirGeneDB (https://mirgenedb.org) using Bowtie v1.1.2 with parameters -v 0 --best --strata -a for known miRNA identification, and novel miRNAs were predicted via hairpin structure analysis using MiRdeep2 (v2.0.0.7). Data obtained from the present study were deposited in the NCBI Sequence Read Archive under the accession number PRJNA1392860. edgeR (v3.12.1) was used to identify DE miRNAs, with a threshold of fold change ≥ 2 and p < 0.05.

To identify clusters of miRNAs exhibiting similar expression pattern, the expression patterns of all miRNAs yielded by high-throughput sequencing were analyzed using Mfuzz (http://www.omicsmart.com).

The prediction of miRNA target genes was carried out using miRanda (v2.1.2), TargetScan (v7.0) and RNAhybrid (v3.3a) (24). The online platform Omicsmart (http://www.omicsmart.com) was used to perform KEGG enrichment analysis (Release 101) on the target genes of the screened DE miRNAs.

Cgi-miR-1175-P6/P7-y was identified as a key miRNA for immune priming. To elucidate the function of Cgi-miR-1175-P6/P7-y in immune priming in oysters, gain-/loss-of-function assay were performed in vivo as previously described (25). A total of 45 oysters were selected and randomly assigned to five groups, designated as the sterile seawater group (SW), the mimics-treated group (Cgi-miR-1175-P6/P7-y mimics + VS), the mimics negative control group (mimics NC + VS), the inhibitor-treated group (Cgi-miR-1175-P6/P7-y inhibitor + VS) and the inhibitor negative control group (inhibitor NC + VS). The oysters in SW group received an injecton with 100 μL PBS, whereas other groups received injections of specific transfection complexes, each prepared by mixing 10 μL of mimic NC, mimic, inhibitor NC, or inhibitor with 10 μL of Lipo6000 and 80 μL of PBS. After 12 h, the SW group was stimulated with sterile seawater, while the other four groups were stimulated with V. splendidus. At 24 h post-stimulation, the hemolymph was collected using a 10 mL syringe and centrifuged at 700 × g for 12 minutes (min) at 4 °C. To assess haemocyte proliferation rates and the expression levels of Cgi-miR-1175-P6/P7-y and proliferation-related genes, haemocytes samples were collected from nine oysters per group and pooled into three biological replicates with three oysters per replicate. The sequences of Cgi-miR-1175-P6/P7-y mimics, mimics NC, Cgi-miR-1175-P6/P7-y inhibitor and inhibitor NC are synthesized by Genepharma (Suzhou, China) (Table 1).

5-Ethynyl-2’-deoxyuridine (EdU) labeling was performed using Alexa Fluor 488 (Beyotime, China) (26). Haemocyte samples were fixed with 4% paraformaldehyde (Beyotime, China) for 15 min and then treated with a permeabilization solution (0.3% Triton X-100 in PBS, Beyotime, China) for 10 min. A portion of the sample was left unstained as a negative control. The remaining samples were stained with the Alexa Fluor 488 kit (C0071S, Beyotime, China) according to the manufacturer’s instructions. EdU-labeled circulating haemocytes were analyzed using a BD FACSAria II flow cytometer (BD Biosciences, USA), and data were processed with FlowJo software.

According to the manufacturer’s instructions, the miRNA First Strand cDNA Synthesis (Sangon Biotech, China) was used to reverse transcribe miRNAs and the TransScript^® Uni All-in-One SuperMix (TransGene, China) was used to synthesize cDNA from 800 ng of total RNA. For reverse transcription in the Cgi-miR-1175-P6/P7-y functional experiments, the miRNA 1st strand cDNA synthesis kit (Accurate Biology, China) was employed. Quantitative real-time PCR (qRT-PCR) was performed with PerfectStart^® Green qPCR SuperMix (TransGene, China) to examine the miRNA expression levels of Cgi-miR-1175-P6/P7-y, Cgi-miR-1993_3p, Cgi-miR-1990_3p, novel-m0115-5p, Cgi-miR-375-P3_3p, novel-m0275-3p, novel-m0095-3p, novel-m0147-5p in four groups (SS, SV, VS and VV), as well as the expression levels of Cgi-miR-1175-P6/P7-y, cyclin-dependent kinase 6 (CgCDK6), CgCDK14, scavenger receptor class F member 2 (CgSCARF2), CgMyc-A, CgRunx and Interleukin 17-1 (CgIL17-1) in mimics- and inhibitor-treated groups, respectively. The amplification reaction was performed in a total volume of 10 μL containing 5 μL of 2×PerfectStart^® Green qPCR SuperMix (AQ602, TransGene, China), 2 μL of cDNA, 0.2 μL of 10 μM each primer, and 2.6 μL of ddH₂O. qRT-PCR was performed on a QuantStudio™ 6 Flex instrument under the following cycling conditions: 94 °C for 30 s, followed by 40 cycles of 94 °C for 5 s and annealing for 32 s. The relative mRNA or miRNA expression was calculated using the 2^-ΔΔCt method, with Elongation factor (EF, NM_001305313) or U6 small nuclear 2 (U6, XR_004597400.1) as internal references. Detailed information on all the primers was listed in Table 1.

Statistical analysis of haemocyte proliferation rate and all qRT-PCR data was performed using SPSS 22.0. Differences between groups were assessed for significance using t-tests and One-Way ANOVA.

To investigate the C. gigas miRNA expression profiles in response to immune priming, the miRNA transcriptome sequencing of haemocytes was conducted. Twelve cDNA libraries were constructed using samples of SS 1-3, SV 1-3, VS 1-3, VV 1-3. A total of 115.13 M high-quality reads (SS = 27.89 M, SV = 30.14 M, VS = 28.24 M and VV = 28.87 M) were successfully yielded (Table 2). The four libraries yielded 8.02 to 11.35 million clean sRNAs, with unique mapping rates ranging from 70.33% to 82.75% (Table 2). After pooling and removing duplicates, a total of 694 miRNAs (51 known C. gigas miRNAs, 360 MirGeneDB miRNAs and 283 novel miRNAs) were identified as differentially expressed across all pairwise comparisons among the SS, SV, VS, and VV groups. A significant U-bias at the first nucleotide was observed across all known, existent, and novel miRNAs (Figure 1B), which is consistent with the previous study (27). The expression levels of the 694 miRNAs, quantified as transcripts per million (TPM), were clustered and visualized by heatmap (Figure 1C). Mfuzz clustering analysis further grouped these miRNAs into 12 distinct clusters based on their coordinated expression patterns across the four groups (Figure 1D).

The DE miRNAs were screened between SS-SV and VS-VV comparisons. The SS-SV comparison represents the response of non-primed oysters to a primary live V. splendidus stimulation and serves as the control for evaluating priming-specific effects in the VS-VV group. A total of 470 DE miRNAs were identified, with 115 up-regulated and 212 down-regulated miRNAs after the primary stimulation (SS-SV) (Figure 2A), and 107 up-regulated and 103 down-regulated miRNAs after the secondary stimulation (VS-VV) (Figure 2B, Supplementary Table 1). Notably, 67 miRNAs were found to be differentially expressed in both the primary and the secondary stimulations (Figure 2C). These 67 DE miRNAs exhibited distinct expression patterns and were divided into six modules (modules 1 to 6), containing 13, 35, 9, 5, 3 and 2 DE miRNAs, respectively (Figures 3A–F). After the primary stimulus, the DE miRNAs in module 2 and module 3 exhibited sustained low (Figure 3B) or high (Figure 3C) expression levels. Additionally, three DE miRNAs in module 5 (Cgi-miR-1175-P6-y, Cgi-miR-1175-P7-y, and novel-0095-3p) showed lower expression upon secondary stimulation compared to the primary response (Figure 3E), while novel-m0074-5p and novel-m0270-5p showed higher expression upon secondary stimulation (Figure 3F). The ‘-P6-y’ and ‘-P7-y’ suffixes in Cgi-miR-1175-P6-y and Cgi-miR-1175-P7-y reflect their relative positions within the gene clusters in the eutherian last common ancestor, while sharing identical mature sequences. This study collectively designates them as Cgi-miR-1175-P6/P7-y for simplicity. DE miRNAs in modules 5 and 6 have proposed to enhance immune responses. Meanwhile, DE miRNAs in modules 2 and 3, which maintained consistently low or high expression after the primary response, may contribute to immune memory formation or support energy provision for secondary stimulation. Eight of these 67 DE miRNAs were validated by qRT-PCR (Figure 3G), confirming the high reliability of the sequencing data.

KEGG enrichment analysis was performed on the target genes of the putative immune enhancing DE miRNAs (module 5 and module 6) and the potential immune memory-related DE miRNAs (module 2 and module 3). The target genes of DE miRNAs in module 5 showed significant enrichment in cell proliferation-related signaling pathways (e.g., ErbB and wnt signaling) and metabolic reprogramming pathways (e.g., carbohydrate digestion and absorption, glycolysis/gluconeogenesis and insulin resistance). Module 5 was also enriched in immune-related signaling pathways, such as the AGE-RAGE, chemokine signaling pathway and Fc epsilon RI signaling pathways (Figure 4A). In contrast, module 6 miRNAs were primarily enriched in neuroendocrine- and neurotransmitter-related pathways, including serotonergic synapse, cortisol synthesis and secretion, thyroid hormone synthesis and estrogen signaling pathway (Figure 4B). Interestingly, the potential immune memory-related DE miRNAs in modules 2 and 3 were also associated with neuroendocrine (serotonergic synapse, cortisol synthesis and secretion, glutamatergic synapse and cushing syndrome) as well as intercellular communication(e.g. long-term depression and gap junction pathways) (Supplementary Figure 1). Together, these findings reveal a division of labor among miRNA modules, with module 5 driving immune activation and modules 2, 3, and 6 supporting systemic immune homeostasis.

Based on the KEGG enrichment results, target genes of miRNAs in modules 5 and 6 were further investigated. The target genes of five putative immune enhancing DE miRNAs in modules 5 and module 6 were further screened, respectively. The expression level of Cgi-miR-1175-P6/P7-y, which was significantly decreased upon secondary stimulation, may enhance immune responses or promote cell proliferation by releasing post-transcriptional repression of multiple pattern recognition receptors (PRRs), cell cycle-related genes and glycolytic-related genes, including SCARF1/2, CDK6/14, hexokinase 2 (HK2) and pyruvate kinase PKM (PKM). Meanwhile, novel-miR-0095-3p may target mannose receptor and cell survival-related genes, such as macrophage mannose receptor 1 (MMR1), C-type mannose receptor 2 (MRC2), inhibitor of apoptosis (IAP) and telomerase protein component (TEP) (Figure 5). Additionally, target gene analysis revealed that the novel-m0074-5p and novel-m0270-5p modulate various neuroendocrine receptors, such as 5-Hydroxytryptamine receptor (5-HTR), ecdysone receptor (EcR), neuronal acetylcholine receptor (nAChR), D (2) dopamine receptor (DRD2), neuropeptide FF receptor 2(NPFFR2), gonadotropin-releasing hormone receptor (GnRHR), thyrotropin-releasing hormone receptor (TRHR), prostaglandin E2 receptor EP3/4 (EP3/4), as well as neurotransmitter transporters (e.g., GABA transporters, GAT), thereby maintaining homeostasis of the immune micro-environment.

Since Cgi-miR-1175-P6/P7-y in module 5 was predicted to target multiple cell proliferation-related genes, its biological function in oysters during immune priming was further validated. CgCDK6, CgCDK14 and CgSCARF2 were identified as the putative target gene of Cgi-miR-1175-P6/P7-y. Binding site between Cgi-miR-1175-P6/P7-y and CgCDK6, CgCDK14 and CgSCARF2 was shown in Figures 6A–C.

Gain/loss-of-function assay were performed to elucidate the regulatory role of Cgi-miR-1175-P6/P7-y in oyster haemocyte proliferation. The mimics-treated group exhibited a 4.37-fold up-regulation of Cgi-miR-1175-P6/P7-y relative to the NC group (p < 0.0001, Figure 6D). Furthermore, the mRNA expression levels of CgCDK6, CgCDK14 and CgSCARF2 were lower than that of in the mimics NC group (Figures 6E-G). The FITC threshold to identify EdU-positive cells was determined using unstained cell controls (Figure 7A). Following V. splendidus stimulation, treatment with Cgi-miR-1175-P6/P7-y mimics reduced the percentage of EdU⁺ haemocytes to 0.72-fold that of the mimics NC group (Figure 7B) and decreased the mRNA expression levels of CgMyc-A, CgRunx, and CgIL17–1 to 0.59, 0.75, and 0.48-fold, respectively (p < 0.05, Figures 7C–E). Conversely, the inhibitor-treated oysters exhibited a 0.60-fold down-regulation of Cgi-miR-1175-P6/P7-y relative to the inhibitor NC group (p < 0.0001, Figure 8A). Furthermore, the mRNA expression levels of CgCDK6, CgCDK14 and CgSCARF2 in the inhibitor-treated group were higher than those in the inhibitor NC group (Figures 8B–D). After treatment with Cgi-miR-1175-P6/P7-y inhibitor the percentage of EdU⁺ haemocytes was increased to 1.56-fold that of the inhibitor NC group following V. splendidus stimulation (Figures 8E, F), and the mRNA expression levels of CgMyc-A, CgIL17-1, and CgRunx also increased to 1.41, 2.03, and 3.48-fold, respectively (Figures 8G–I, p < 0.05). Collectively, these reciprocal results demonstrate that Cgi-miR-1175-P6/P7-y acts as a key negative regulator of haemocyte proliferation in oysters.