Oyster hemolymph is a complex and dynamic ecosystem hosting bacteria, protists and viruses.

BACKGROUND: The impact of the microbiota on host fitness has so far mainly been demonstrated for the bacterial microbiome. We know much less about host-associated protist and viral communities, largely due to technical issues. However, all microorganisms within a microbiome potentially interact with each other as well as with the host and the environment, therefore likely affecting the host health. RESULTS: We set out to explore how environmental and host factors shape the composition and diversity of bacterial, protist and viral microbial communities in the Pacific oyster hemolymph, both in health and disease. To do so, five oyster families differing in susceptibility to the Pacific oyster mortality syndrome were reared in hatchery and transplanted into a natural environment either before or during a disease outbreak. Using metabarcoding and shotgun metagenomics, we demonstrate that hemolymph can be considered as an ecological niche hosting bacterial, protist and viral communities, each of them shaped by different factors and distinct from the corresponding communities in the surrounding seawater. Overall, we found that hemolymph microbiota is more strongly shaped by the environment than by host genetic background. Co-occurrence network analyses suggest a disruption of the microbial network after transplantation into natural environment during both non-infectious and infectious periods. Whereas we could not identify a common microbial community signature for healthy animals, OsHV-1 µVar virus dominated the hemolymph virome during the disease outbreak, without significant modifications of other microbiota components. CONCLUSION: Our study shows that oyster hemolymph is a complex ecosystem containing diverse bacteria, protists and viruses, whose composition and dynamics are primarily determined by the environment. However, all of these are also shaped by oyster genetic backgrounds, indicating they indeed interact with the oyster host and are therefore not only of transient character. Although it seems that the three microbiome components respond independently to environmental conditions, better characterization of hemolymph-associated viruses could change this picture.

Cecropins as a marker of Spodoptera frugiperda immunosuppression during entomopathogenic bacterial challenge.

An antimicrobial peptide (AMP) of the cecropin family was isolated by HPLC from plasma of the insect pest, Spodoptera frugiperda. Its molecular mass is 3910.9 Da as determined by mass spectrometry. Thanks to the EST database Spodobase, we were able to describe 13 cDNAs encoding six different cecropins which belong to the sub-families CecA, CecB, CecC and CecD. The purified peptide identified as CecB1 was chemically synthesized (syCecB1). It was shown to be active against Gram-positive and Gram-negative bacteria as well as fungi. Two closely related entomopathogenic bacteria, Xenorhabdus nematophila F1 and Xenorhabdus mauleonii VC01(T) showed different susceptibility to syCecB1. Indeed, X. nematophila was sensitive to syCecB1 whereas X. mauleonii had a minimal inhibitory concentration (MIC) eight times higher. Interestingly, injection of live X. nematophila into insects did not induce the expression of AMPs in hemolymph. This effect was not observed when this bacterium was heat-killed before injection. On the opposite, both live and heat-killed X. mauleonii induced the expression of AMPs in the hemolymph of S. frugiperda. The same phenomenon was observed for another immune-related protein lacking antimicrobial activity. Altogether, our data suggest that Xenorhabdus strains have developed different strategies to supplant the humoral defense mechanisms of S. frugiperda, either by increasing their resistance to AMPs or by preventing their expression during such host-pathogen interaction.

The dlt operon of Bacillus cereus is required for resistance to cationic antimicrobial peptides and for virulence in insects.

The dlt operon encodes proteins that alanylate teichoic acids, the major components of cell walls of gram-positive bacteria. This generates a net positive charge on bacterial cell walls, repulsing positively charged molecules and conferring resistance to animal and human cationic antimicrobial peptides (AMPs) in gram-positive pathogenic bacteria. AMPs damage the bacterial membrane and are the most effective components of the humoral immune response against bacteria. We investigated the role of the dlt operon in insect virulence by inactivating this operon in Bacillus cereus, which is both an opportunistic human pathogen and an insect pathogen. The Delta dlt(Bc) mutant displayed several morphological alterations but grew at a rate similar to that for the wild-type strain. This mutant was less resistant to protamine and several bacterial cationic AMPs, such as nisin, polymyxin B, and colistin, in vitro. It was also less resistant to molecules from the insect humoral immune system, lysozyme, and cationic AMP cecropin B from Spodoptera frugiperda. Delta dlt(Bc) was as pathogenic as the wild-type strain in oral infections of Galleria mellonella but much less virulent when injected into the hemocoels of G. mellonella and Spodoptera littoralis. We detected the dlt operon in three gram-negative genera: Erwinia (Erwinia carotovora), Bordetella (Bordetella pertussis, Bordetella parapertussis, and Bordetella bronchiseptica), and Photorhabdus (the entomopathogenic bacterium Photorhabdus luminescens TT01, the dlt operon of which did not restore cationic AMP resistance in Delta dlt(Bc)). We suggest that the dlt operon protects B. cereus against insect humoral immune mediators, including hemolymph cationic AMPs, and may be critical for the establishment of lethal septicemia in insects and in nosocomial infections in humans.

Cg-IkappaB, a new member of the IkappaB protein family characterized in the pacific oyster Crassostrea gigas.

Inhibitors of NF-kappaB (IkappaBs) have been implicated as major components of the Rel/NF-kappaB signaling pathway, which is an important mediator of immune responses throughout much of the animal kingdom. Here we report the first characterization of a bivalve mollusc cDNA that encodes a novel IkappaB member called Cg-IkappaB showing the conservation of most IkappaB protein characteristics. Sequences as well as phylogenetic analyses reveal a high level of identity between not only Cg-IkappaB and other mollusc and insect IkappaB-like proteins but also similarities to vertebrate IkappaB-alpha and -epsilon isoforms. Expression analyses demonstrated that the transcript is widely expressed in all oyster tissues. This work is consistent with our previous discovery of several members of an NF-kappaB pathway in Crassostrea gigas and further sustains the hypothesis of a conserved scheme of immune gene regulation through most of the metazoans.

First evidence of the activation of Cg-timp, an immune response component of Pacific oysters, through a damage-associated molecular pattern pathway.

In a previous work, we characterized a Crassostrea gigas cDNA (Cg-timp) encoding a protein which presents all the features of vertebrate tissue inhibitor of metalloproteinase (TIMP). The expression pattern of this gene led us to propose that Cg-timp is an important factor in oyster wound healing and defense mechanisms. Here we describe the analysis of Cg-timp expression in oysters challenged by live or dead bacteria as well as by bacterial secretory/excretory products and metalloproteinase. Surprisingly, bacterial secretory/excretory products activate Cg-timp gene expression whereas heat-inactivated ones do not. To address the question of the signal transduction pathway involved in Cg-timp gene activation, we isolated and sequenced Cg-timp promoter and upstream region. A 1-kb genomic DNA fragment flanking the 5'-end of the gene contains several regulatory elements and notably three NF-kappaB binding sites. The potential involvement of these motifs in Cg-timp gene regulation is discussed.

Cg-Rel, the first Rel/NF-kappaB homolog characterized in a mollusk, the Pacific oyster Crassostrea gigas.

We report here the identification and functional characterization of Cg-Rel, a gene encoding the Crassostrea gigas homolog of Rel/NF-kappaB transcription factors found in insects and mammals. Sequence and phylogenetic analysis showed that Cg-Rel shares the structural organization of Rel/NF-kappaB transcription factors of class II. It includes a Rel homology domain as well as a C-terminal transactivation domain (TD). Overexpression of Cg-Rel in the Drosophila S2 cell line activated the expression of a NF-kappaB-dependent reporter gene, whereas transfection with a Cg-Rel construct containing a C-terminal deletion of the TD or using a reporter gene with mutated kappaB binding sites failed to activate expression. These results suggest that Cg-Rel is a functional member of the Rel family of transcription factors, making this the sixth structurally homologous component of the Rel/NF-kappaB pathway characterized in C. gigas. Based on homology to other invertebrates' Rel/NF-kappaB cascade, the function of the oyster pathway may serve to regulate genes involved in innate defense and/or development. These findings serve to highlight a potentially important regulatory pathway to the study of oyster immunology, hence allowing comparison of the immune system in vertebrates and invertebrates, an important key issue to understand its evolution.

Cg-TIMP, an inducible tissue inhibitor of metalloproteinase from the Pacific oyster Crassostrea gigas with a potential role in wound healing and defense mechanisms(1).

We have cloned and characterized a cDNA encoding Cg-TIMP, the first tissue inhibitor of metalloproteinase identified in mollusks. The isolated cDNA encodes a protein of 221 residues that has a domain organization similar to that of vertebrate TIMPs including a signal sequence, and the 12 cysteines characteristic of the TIMP signature. Analysis of Cg-TIMP expression in adult oyster tissues, by Northern blot and in situ hybridization, indicates that Cg-TIMP was only expressed in hemocytes which are the key components of defense mechanisms in mollusks. We also observed that Cg-TIMP mRNA accumulated during shell damage and bacterial challenge. This pattern of expression suggests that Cg-TIMP may be an important factor in wound healing and defense mechanisms.

Characterization of a cDNA encoding a 72 kDa heat shock cognate protein (Hsc72) from the Pacific oyster, Crassostrea gigas.

A full-length cDNA encoding a 72 kDa heat shock cognate protein (Hsc72) was isolated from a Crassostrea gigas hemocyte library. This cDNA is 75% identical to a human hsc70 cDNA. C. gigas cDNA contains a 659 amino acid open reading frame encoding a 72 kDa protein which is 87% identical to a human Hsc70 protein. Northern blotting indicated that even if the hsc72 gene was constitutively expressed in oyster hemocytes, it could be stimulated by heat shock in vitro as well as in vivo. Homologies observed both at the cDNA and protein levels, as well as the expression pattern of the hsc72 mRNA, support our conclusion that the isolated cDNA clone corresponds to an oyster heat shock cognate gene.

Oyster IKK-like protein shares structural and functional properties with its mammalian homologues.

In our search for genes involved in oyster immunity we isolated a cDNA encoding a polypeptide closely related to the mammalian IkappaB kinase (IKK) family. IKK proteins play a central role in cell signaling by regulating nuclear factor-kappaB (NF-kappaB) activation. We report here the cloning of an oyster IKK-like protein (oIKK) which possesses the characteristic organization of the mammalian IKK proteins, namely an amino-terminal kinase domain followed by a leucine zipper region and a carboxyl-terminal helix-loop-helix motif. When transfected into human cell lines, oIKK activated the expression of NF-kappaB-controlled reporter gene, whereas transfections with mutants of oIKK deleted within the kinase domain or within the helix-loop-helix motif respectively abolished and greatly reduced reporter gene activation. These results indicate that oIKK can replace the hIKK-alpha in catalyzing NF-kappaB nuclear translocation, and in triggering gene expression. Our results sustain the concept of an evolutionarily conserved signaling machinery in which IKK plays a major role.

Light intensity regulation of cab gene transcription is signaled by the redox state of the plastoquinone pool.

The eukaryotic green alga Dunaliella tertiolecta acclimates to decreased growth irradiance by increasing cellular levels of light-harvesting chlorophyll protein complex apoproteins associated with photosystem II (LHCIIs), whereas increased growth irradiance elicits the opposite response. Nuclear run-on transcription assays and measurements of cab mRNA stability established that light intensity-dependent changes in LHCII are controlled at the level of transcription. cab gene transcription in high-intensity light was partially enhanced by reducing plastoquinone with 3-(3,4-dichlorophenyl)-1,1-dimethyl urea (DCMU), whereas it was repressed in low-intensity light by partially inhibiting the oxidation of plastoquinol with 2,5-dibromo-3-methyl-6-isopropyl-p-benzoquinone (DBMIB). Uncouplers of photosynthetic electron transport and inhibition of water splitting had no effect on LHCII levels. These results strongly implicate the redox state of the plastoquinone pool in the chloroplast as a photon-sensing system that is coupled to the light-intensity regulation of nuclear-encoded cab gene transcription. The accumulation of cellular chlorophyll at low-intensity light can be blocked with cytoplasmically directed phosphatase inhibitors, such as okadaic acid, microcystin L-R, and tautomycin. Gel mobility-shift assays revealed that cells grown in high-intensity light contained proteins that bind to the promoter region of a cab gene carrying sequences homologous to higher plant light-responsive elements. On the basis of these experimental results, we propose a model for a light intensity signaling system where cab gene expression is reversibly repressed by a phosphorylated factor coupled to the redox status of plastoquinone through a chloroplast protein kinase.

Is the IS1 transposase, InsAB', the only IS1-encoded protein required for efficient transposition?

The transposase of the bacterial insertion sequence IS1 is normally expressed by inefficient translational frameshifting between an upstream reading frame which itself specifies a transposition inhibitor, InsA, and a second consecutive reading frame located immediately downstream. A fused-frame mutant which carries an additional base pair inserted at the point of frameshifting was constructed. This mutant exhibits high transposition activity and should express the transposase, InsAB', constitutively without frameshifting. Unexpectedly, a second protein species was observed to be expressed from this mutant. We demonstrate here that this protein, InsA*, results from continued frameshifting on the modified frameshift motif. The protein retains the activities of the repressor InsA. Its elimination, by further modification of the frameshift motif, results in a further increase in various transposition activities of IS1. These results support the hypothesis that a single IS1-encoded protein, InsAB', is necessary for transposition.

Translational control of transposition activity of the bacterial insertion sequence IS1.

The experiments reported here provide strong evidence indicating that the transposition frequency of the bacterial insertion sequence IS1 is determined principally by two IS1-specified proteins. The first, InsA, was previously shown to bind to the ends of the element and to act as a repressor. We present both physical and genetic evidence which reveals that the second, the InsAB' transposase, is a fusion of InsA with the product of a downstream reading frame, InsB'. Synthesis of this protein occurs by a -1 frameshift between the insA and insB' frames. It requires the presence of an intact retroviral-like frameshift signal composed of an A6C motif and a downstream region able to form several alternative secondary structures. In vivo studies show that IS1 transposition activity depends on the relative rather than on the absolute levels of InsA and InsAB'. The ratio is determined primarily at the translational level by frameshifting and appears to be relatively insensitive to large variations in levels of transcription. This novel homeostatic control could therefore protect IS1 from activation as a consequence of insertion into active transcription units.

Optimized conditions for electrotransformation of bacteria are related to the extent of electropermeabilization.

Electropulsation is a simple and efficient way to introduce cloned genes into a variety of cell types, even with walled species. In the case of bacteria, we observed no direct correlation between survival rate and transformation yield. In the present work, we show that the yield of transformation is directly related to the level of the electric-field induced level of cell permeabilization. From experiments on Escherichia coli, it was confirmed that the extent of associated ATP leakage was a reliable assay. This approach was extended to other strains, such as Salmonella typhimurium, which to date had not been electrotransformed by plasmids.

The regulatory role of the IS1-encoded InsA protein in transposition.

We show here that the protein InsA, which is encoded by IS1 and binds specifically to the terminal inverted repeats of this insertion sequence, negatively regulates IS1 transposition activity. We demonstrate that it inhibits both IS1-mediated cointegrate formation and transposition of a synthetic IS1-based transposon ('omegon'; omega-on). These results also indicate that the omega-on which does not itself encode IS1 transposition functions can be complemented in trans, presumably by the copies of IS1 resident in the Escherichia coli chromosome. Using insA-lacZ gene fusions, we show that at least part of this effect can be explained by the ability of InsA to repress expression of IS1-encoded genes both in cis or in trans. The experiments involving omega-on transposition raise the possibility that InsA inhibits transposition directly by competition with the transposase for their cognate site within the ends of IS1.