Yamama Naciri-Graven

Selecting the flat oyster Ostrea edulis (L.) for survival when infected with the parasite Bonamia ostreae

Abstract In the early 1970s, the appearance of the protozoan parasite Marteilia refringens restricted the production of the European flat oyster Ostrea edulis to subtidal areas. Ten years later a second protozoan parasite, Bonamia ostreae, caused a further serious decline in flat oyster farming in subtidal areas. Prophylactic measures were taken to sustain oyster farming, but the continuing presence of the two parasites led to the initiation of a programme to select for resistant oysters. Efforts were concentrated on B. ostreae and two selected strains were obtained in 1985 (S85) and 1989 (S89). The programme was based on mass selection and parasite inoculation or natural infections. This paper focuses on the survival and weights of the third generation of S85 (S85-G3), the second generation of S89 (S89-G2) and a cross between S85-G2 and S89-G1. The % survival of S85-G3 oysters was more than 4-times higher than the control group after a 20 month experiment in the wild, and significant differences in parasite prevalence were recorded. No significant differences between S89-G2 and the control group were observed for either survival or parasite prevalence. This was attributed either to a less powerful experiment or to the effect of genetic load clearing as a result of inbreeding. The % survival of the cross between the two strains was more than twice as high as the control group and significant differences were recorded for parasite prevalence. Good evidence for additivity of the resistance was provided by both the intermediate behaviour of a cross between S85-G2 and the controls and the rapid improvement that was obtained from the first generations. The selected strains showed a tendency toward higher weights and higher weight variances when compared to controls. These populations are suspected to have undergone bottlenecks, which would explain the increase in phenotypic variance. No measurable natural resistance has arisen in the wild. This could be due first to the relatively short time that has elapsed since the first infections, secondly to the early sale of resistant oysters as soon as they reach a marketable size, and thirdly to the slow kinetics of B. ostreae infection that allows both susceptible and potentially resistant oysters to participate in recruitment. Improved resistance seems to be related to the delayed mortality of selected oysters. This last result is discussed in the light of a previous study conducted in the laboratory which showed that resistant and susceptible oysters have different haemograms.

Population bottleneck and effective size in Bonamia ostreae-resistant populations of Ostrea edulis as inferred by microsatellite markers

Genetic variability at five microsatellite loci was analysed in three hatchery-propagated populations of the flat oyster, Ostrea edulis. These populations were part of a selection programme for resistance to the protozoan parasite Bonamia ostreae and were produced by mass spawns, without control of the genealogy. Evidence for population bottlenecks and inbreeding was sought. A reduction in the number of alleles, mainly due to the loss of rare alleles, was observed in all selected populations, relative to the natural population from which they were derived. Heterozygote excesses were observed in two populations, and were attributed to substructuring of the population into a small number of families. Pedigree reconstruction showed that these two populations were produced by at most two spawning events involving a limited number of parents. Most individuals within these populations are half or full-sib, as shown by relatedness coefficients. The occurrence of population bottlenecks was supported by estimates of effective number of breeders derived by three methods: temporal variance in allelic frequencies, heterozygote excess, and a new method based on reduction in the number of alleles. The estimates from the different methods were consistent. The evidence for bottleneck and small effective number of breeders are expected to lead to increasing inbreeding, and have important consequences for the future management of the three O. edulis selected populations.

Influences of triploidy, parentage and genetic diversity on growth of the Pacific oyster Crassostrea gigas reared in contrasting natural environments.

An increasing number of hypotheses are being proposed to explain the faster growth potential of triploids in molluscs, including their partial sterility or their higher heterozygosity compared to diploids. Triploid advantage however, remains controversial for poorer sites, because of a potential trade‐off with survival. These questions were addressed in Crassostrea gigas by deploying meiosis II triploids and their diploid siblings from a single mass spawning of three males and seven females, in two contrasting locations for their trophic resources. One hundred and fifty individuals were sampled at each site after nine months, measured for weight and biochemical composition, and genotyped using three microsatellite and seven allozyme loci. Higher performance was observed at the fast‐growing site for all traits except shell weight, and triploids had greater weights and biochemical contents than diploids at harvest. Triploids also grew faster at the poorer site, and showed similar survival rates to diploids at both sites. Triploids had significantly higher average allozyme and microsatellite diversity. However, they performed better for a wide range of individual heterozygosity values, arguing for an advantage of the triploid state per se, that could be due to positive effects on growth of both sterility of triploids with subsequent resource re‐allocation and possible faster transcription with three copies of each gene. Despite evidence of very low or no inbreeding in the diploid sample, positive associations between individual allozyme diversity and growth were detected, which explained little but significant amounts of phenotypic variation. These associations were interpreted as direct effects of allozymes, either alone or including epistatic interactions with other loci. In addition, measures of individual distance (mean‐d2) specific to microsatellites, were negatively correlated with growth in diploids, indicating possible effects of outbreeding depression between more distant genomes of parents from distinct populations.

Separate effects of triploidy, parentage and genomic diversity upon feeding behaviour, metabolic efficiency and net energy balance in the Pacific oyster Crassostrea gigas.

Triploid oysters were induced using cytochalasin B upon retention of either the first (meiosis I triploids) or the second (meiosis II triploids) polar body in embryos from a single cohort derived from mixed parentage. Allozyme and microsatellite assays enabled the confirmation of both parentage and triploidy status in each oyster. Comparison of meiosis I triploids, meiosis II triploids and diploid siblings established that improved physiological performance in triploids was associated with increased allelic variation, rather than with the quantitative dosage effects of ploidy status. An unidentified maternal influence also interacted with genotype. Among full sibs, allelic variation measured as multi-locus enzyme heterozygosity accounted for up to 42% of the variance in physiological performance; significant positive influences were identified upon feeding rate, absorption efficiency, net energy balance and growth efficiency (= net energy balance divided by energy absorbed). Whilst allelic variation was greater in both meiosis I and meiosis II triploids than in diploid siblings, both allelic variation and net energy balance were highest in triploids induced at meiosis I. This suggests that it may be preferable to induce triploidy by blocking meiosis I, rather than meiosis II as has traditionally been undertaken during commercial breeding programmes.

Comparative growth of Bonamia ostreae resistant and wild flat oyster Ostrea edulis in an intensive system : II. Second year of the experiment

This paper reports the comparative growth and mortality of two crosses involving S85 and S89, two Ostrea edulis strains selected by IFREMER for their resistance to Bonamia ostreae. A first year of monitoring had revealed that the two crosses grew significantly better than two control populations from Quiberon Bay (Brittany) and Palavas (Mediterranean Sea). In a second and final year experiment which lasted from April to November 1995, the cross between the first generation of S89 and wild oysters (G0G1) performed better for most growth characteristics (whole weight, shell weight, length and width), than both the cross between the first generation of S89 and the second generation of S85 (G1G2) and the Quiberon control group, which were not significantly different from each other. Mortality was quite high during the experiment but was lower for the G0G1 cross (61%) than for both G1G2 (67%) and the Quiberon controls (76%). A low level of B. ostreae infection was detected during the experiments indicating early infections. These results are consistent with the previous one showing a direct or indirect effect of selection to B. ostreae resistance on growth.

Comparison of genetic variability and parentage in different ploidy classes of the Japanese oyster Crassostrea gigas.

Chemical treatments with cytochalasin B were used to induce triploidy in the progeny of a mass fertilization of 3 male and 7 female Crassostrea gigas parents. Triploids were produced either by retention of the first (meiosis I (MI) triploids) or the second (meiosis II (MII) triploids) polar bodies. These animals, together with their diploid siblings, were divided for two experiments. One set was used to compare physiological performance, and the other set deployed to compare growth in two different natural environments. For both experiments, genetic variability in different ploidy classes was estimated using three microsatellite loci and eight allozyme loci. The microsatellite loci were highly polymorphic, allowing independent confirmation of ploidy status and the unambiguous identification of parentage for each oyster. Significant differences in parentage were found between ploidy classes, despite the fact they originated from the same mass fertilization. This indicates that the assumptions of a common genetic background among random samples of animals taken from the same mass fertilization may not be generally valid. Knowledge of parentage also allowed the more accurate scoring of allozyme loci. As expected, triploids were found to be significantly more polymorphic than diploids. However, MI triploids were not significantly more polymorphic than MII triploids. MII triploid genotypes were used to estimate recombination rates between loci and their centromeres. These rates varied between 0.29 and 0.71, indicating only moderate chiasma interference.