A Sustained Immune Response Supports Long-Term Antiviral Immune Priming in the Pacific Oyster, Crassostrea gigas

Abstract

Over the last decade, innate immune priming has been evidenced in many invertebrate phyla. If mechanistic models have been proposed, molecular studies aiming to substantiate these models have remained scarce. We reveal here the transcriptional signature associated with immune priming in the oyster Crassostrea gigas Oysters were fully protected against Ostreid herpesvirus 1 (OsHV-1), a major oyster pathogen, after priming with poly(I·C), which mimics viral double-stranded RNA. Global analysis through RNA sequencing of oyster and viral genes after immune priming and viral infection revealed that poly(I·C) induces a strong antiviral response that impairs OsHV-1 replication. Protection is based on a sustained upregulation of immune genes, notably genes involved in the interferon pathway and apoptosis, which control subsequent viral infection. This persistent antiviral alert state remains active over 4 months and supports antiviral protection in the long term. This acquired resistance mechanism reinforces the molecular foundations of the sustained response model of immune priming. It further opens the way to applications (pseudovaccination) to cope with a recurrent disease that causes dramatic economic losses in the shellfish farming industry worldwide.IMPORTANCE In the last decade, important discoveries have shown that resistance to reinfection can be achieved without a functional adaptive immune system, introducing the concept of innate immune memory in invertebrates. However, this field has been constrained by the limited number of molecular mechanisms evidenced to support these phenomena. Taking advantage of an invertebrate species, the Pacific oyster (Crassostrea gigas), in which we evidenced one of the longest and most effective periods of protection against viral infection observed in an invertebrate, we provide the first comprehensive transcriptomic analysis of antiviral innate immune priming. We show that priming with poly(I·C) induced a massive upregulation of immune-related genes, which control subsequent viral infection, and it was maintained for over 4 months after priming. This acquired resistant mechanism reinforces the molecular foundations of the sustained response model of immune priming. It opens the way to pseudovaccination to prevent the recurrent diseases that currently afflict economically or ecologically important invertebrates.

Keywords: OsHV-1; POMS; antiviral response; immune memory; innate immunity; interferon; oyster; poly(I·C); priming; transcriptomic.

FIG 1

Experimental design used to identify molecular basis underlying poly(I·C) priming. Specific-pathogen-free (SPF) oysters, highly susceptible to Pacific oyster mortality syndrome, were anesthetized before being primed with poly(I·C) or filtered seawater (FSW). Ten days postpriming (10 DPP), oysters from each condition were challenged with OsHV-1 inoculum (1.32 × 10⁸ copies of DP gene per oyster) or OsHV-1-free inoculum (control). A part of the initial batch of oysters was kept untreated during the experiment. Survival of oysters was monitored for 10 days postchallenge. Three pools of 15 oysters for each condition were sampled postpriming (T0, 0.5 day, 1 day, and 10 days) and postchallenge (DPC) (0.5 day and 1 day) for viral load analyses or RNA sequencing.

FIG 2

Poly(I·C) priming induces a phenotypical reversal upon OsHV-1 challenge. (A) Kaplan-Meier survival curves of oysters were primed by injection with poly(I·C) (19 μg g⁻¹ of oyster) or filtered seawater as a control or left nontreated before being challenged by injection, 10 days postpriming, with an OsHV-1 inoculum (1.32 × 10⁸ copies of DP gene per oyster; solid lines) or with a control inoculum (OsHV-1-free, dotted lines) or left nontreated. All conditions except FSW plus OsHV-1 demonstrated 100% survival rates and appear hidden and merged with other ones reaching the same survival rate. Survival in each group of 60 oysters was monitored for 10 days after challenge. ***, P value < 0.0001, log rank test; n = 60. (B) Graph representing the final survival rates (10 days postchallenge) of oyster batches not treated (NT) or injected with poly(I·C) or FSW.

FIG 3

Poly(I·C) stimulation restrains OsHV-1 replication in the oyster. The OsHV-1 DNA load was quantified by quantitative PCR and expressed as viral genomic units per nanogram of total oyster DNA (A); viral replication was estimated by the total number of RNA-seq reads mapped on the OsHV-1μVarA genome (B). Virus dynamics was followed at different times after pathogenic (OsHV-1 inoculum) or nonpathogenic (control inoculum) challenges. Challenges were realized 10 days after priming with FSW or poly(I·C). Different letters above bars indicate statistically significant differences for each time compared to 10 DPP using one-way ANOVA (a, P value < 0.05; b and c, P value < 0.001; n = 3 pools of 15 oysters).

FIG 4

Poly(I·C) stimulates a strong innate immune response in oysters. (A) Comparisons of transcriptomes were performed between poly (I·C) and FSW-treated oysters at 0.5, 1, and 10 DPP using the DESeq2 package (FDR < 0.05). The table indicates the number of differentially expressed genes (DEG) between conditions and the total number of nonredundant DEG across all conditions (4,577). (B) Venn diagram representing the 4,577 DEG in each comparison. (C) Heat map of 148 enriched gene ontology categories found in postpriming conditions. GO enrichment analyses were done using a rank-based statistical test on log₂ fold change of differentially expressed genes at each time compared to the control. A category was considered enriched under an FDR of <0.01. The intensity of the enrichment is expressed as the ratio between the number of genes that were significantly up (yellow) or down (blue) regulated in the category/total number of genes in the category. GO categories were clustered (1 to 6) according to the Pearson uncentered correlation (MeV_4_9_0 software). (D) Heat map focusing on the two clusters containing 27 immune-related GO categories (underlined in yellow) that were significantly enriched at 0.5 DPP, 1 DPP, or 1 DPC (FDR > 0.01, biological processes).

FIG 5

Poly(I·C) induces a massive upregulation of genes. The heat map shows expression kinetics of the 1,587 transcripts of oyster primed with poly(I·C) or injected with FSW as a control and challenged after 10 days with OsHV-1. The color scale indicates the log₂ fold change calculated by DEseq2, of the 1,587 transcripts identified by RBGOA in comparing [poly(I·C) or FSW] 0.5, 1, and 10 DPP to T0 (FDR < 0.05) and 0.5 and 1 DPC to 10 DPP [poly(I·C) or FSW] (FDR < 0.05). The same transcripts are indicated in the same position for each condition. Heat maps for the poly(I·C) condition are shown in panels A and B which correspond to the postpriming and postchallenge time points, respectively. Heat maps for the control (FSW) condition are shown in panels C and D which correspond to the postpriming and postchallenge time points, respectively.

FIG 6

Regulation profile of transcripts from primed oysters after OsHV-1 challenge. (A) The expression profile followed by the 1,587 DEG could be categorized into 5 expression patterns: (i) “not regulated,” when genes are not regulated either postpriming injection or postchallenge compared to the reference times; (ii) “priming specific,” when genes are regulated after priming but not after challenge; (iii) “challenge specific,” when genes are regulated only after challenge; “recalled” (iv) and “recalled/opposite” (v), when genes are regulated in early times postpriming followed by a return to a basal state and regulated postchallenge, in the same or opposite way as postpriming; and (vi) “sustained,” when gene expression is maintained through priming and challenge (no return to a basal state). (B) Pie chart of the percentage of genes following the 5 different expression profiles regrouped into 4 (not regulated, regulated only after priming, regulated after challenge regrouping challenge specific and recalled/opposite patterns, and sustained) in poly(I·C)-primed oysters after OsHV-1 challenge. (C) Pie chart of the percentage of genes from the poly(I·C)-primed condition with a sustained expression profile (without genes also sustained in nonprimed oysters) that have been categorized as associated with an antibacterial (B), antiviral (V), or antimicrobial (VB) or not associated with any treatment (other). (D) Chart of the percentage of genes from the V and VB categories categorized associated with immune pathways. ND, not determined (genes could not be associated with a known immune pathway); AMP, antimicrobial peptide.

FIG 7

Gene expression pattern of a long-term priming experiment. The expression pattern of 11 candidate genes was followed in oysters primed with poly(I·C) or injected with FSW as a control and challenged after 126 days with OsHV-1. (A) Heat map generated from the log₂ fold change (FC) of the 11 genes at 1, 14, 56, and 126 days postpriming and 0.5, 1, and 2 days postchallenge. Log₂ FC was calculated by the 2^−ΔΔCq method in comparing each time postpriming and postchallenge to T0. Genes from primed oysters with FC significatively different from the control (FSW) at 126 DPP are indicated by asterisks (multiple t test; *, P < 0.1; **, P < 0.5; ***, P < 0.001). (B) Graph plotting the log₂ FC of the 11 genes over the kinetic of priming and challenge. Asterisks indicate significant differences between transcripts’ expression at 14 or 126 days postpriming and time zero (multiple t test; *, P < 0.1; **, P < 0.5; ***, P < 0.001; ****, P < 0.0001).