Aristotle University Of Thessaloniki (AUTH)
Genetic and genomic research to improve the productivity and robustness of farmed fish
Abstract
dc:descriptionNowadays, seafood comes both from wild and farmed populations, however aremarkable increase in the total production from the farmed populations is evident.In the southern European countries, species such as the European seabass and thegilthead seabream have been established as the leading Mediterranean species.However, given the wide consumer demand for different seafood, new species arebecoming more popular. The increasing production of aquaculture necessitatesfurther research into the genetic architecture such us growth performance and diseaseresistance against pathogens. For this purpose, the European seabass as well as twonew species with, emerging importance for aquaculture the sharpsnout seabream andthe common dentex have been studied in this dissertation using a variety of geneticand genomic approaches. Our research was focused on the growth performance aswell as the welfare and health traits of those species.The European seabass shows a higher stress response not only when being handled,but also when it is exposed to transfer etc. compared to other Mediterranean marinespecies. This higher stress response influences the growth and welfare of the fish.Apart from stress response, the European seabass is vulnerable to some Vibriopathogen species which, in some cases, could lead to high mortality after infection.To investigate the heritability of the stress response, body weight and resistanceagainst Vibrio infection and to identify QTL affecting the above traits in Europeanseabass, a commercial group of fish (Named by company as batch 10, 620 fish) wasselected. In this group, a stress experiment was performed in order to collectphenotypic measurements for stress indicators like cortisol, lysozyme and glucoselevels as well as their weight which was recorded in 4 different growth time points.Furthermore, a subgroup of batch 10 (332 fish) was selected to participate in theVibrio challenge in order to investigate evidence of significant genetic control fordisease resistance. Two multitrait animal models combined the stress indicators withbody weight and the disease resistance were analyzed. The heritability from all theabove models and traits ranged from 0.23 to 0.55, while the genetic/phenotypiccorrelations between traits reveal a negative genetic correlation between body weightand cortisol levels (-0.43) along with a positive genetic correlation between cortisollevels and resistance against Vibrio (0.33). The fish were genotyped using 35microsatellite markers and a linkage map was constructed providing six linkagegroups. A Quantitative Traits Loci (QTL) analysis for the stress indicators, the bodyweight and the resistance against Vibrio was performed. QTL affecting the stressindicators and the body weight were detected in all the linkage groups but noevidence of QTL for the disease resistance was identified. In conclusion, the studiedstress indicators can be considered as inheritable traits in European seabass and incombination with the detected QTL for those traits, candidates can be selected basedon their QTL genotype providing offspring with lower stress response, using aMarker Assisted Selection (MAS) approach. QTL affecting the body weight werealso detected, which can be accounted into the selection candidates’ process, in orderto improve the accuracy of selection.Since QTL were detected and moderate heritability was estimated for the above traitsin the population, further research was performed to deepen into the geneticknowledge of those traits and to create a useful tool of selection of breeders. Thefirst group of fish (batch 10, 620 offspring) was genotyped using a 57K SNP arrayas well as a second group of commercial fish (batch 13, 330 offspring) which wasselected and participated in a new stress experiment (under the same conditions), inorder to collect the phenotypic measurements. In the studied population (batch 10and batch 13), the Linkage Disequilibrium was calculated around 0.036, andhaplotype blocks were constructed including only 7.37% of the total SNPs. GWASfor all the above measurements was performed and two QTL associated with lactatelevels explaining 5% of the phenotypic variation were detected. One QTL wasdiscovered to affect the lysozyme levels, explaining 2.2% of the phenotypicvariation. Moreover, a putative QTL affecting growth performance was detectedusing all offspring, while two QTL associated with body weight were detected whenonly the first group (batch 10) was analyzed. Haplotype block based RegionalHeritability Mapping and a Regional Heritability mapping was performed on a sizeregion of 20 consecutive SNPs, but no evidence of statistically significant regions orblocks was detected. The last step was to estimate the genetic gain per generation forthe lactate levels using two different approaches, traditional polygenic (i.e., pedigreebased) selection and MAS. Thus, EBVs were estimated using only polygenicinfinitesimal models and a MAS selection was performed using the EBVs frompolygenic infinitesimal models combined with the QTL effect. The fish wereselected using the EBVs from the above approaches and compared, revealing ahigher decrease on the lactate levels using the QTL effect through MAS comparedto polygenic (i.e., pedigree based) selection. In conclusion, further research is neededin order to study the putative QTL affecting growth and to clarify its effect inproduction. Focusing on the stress response, lactate levels can be used beneficiallyas stress indicator for the European seabass. The use of the QTL affecting lactatelevels, can be helpful to the candidate’s selection into a breeding program since ahigher decrease of lactate levels was estimated in the candidates which were selectedusing it through MAS selection.Since only few QTL were detected using Genome Wide Association (GWAS)analysis, Genomic Selection and traditional polygenic (pedigree based) approachwere performed in order to compare them and select the most competitive strategyof the studied traits in the European seabass. The fish from the two batches wereused in order to estimate the heritability for the stress indicators as well as for thebody weight using Maximum likelihood. The genetic variance was estimated usingtwo different relationship matrices, the pedigree and the genomic, both were fittedin the model, one at a time. Starting with stress indicators, the heritability wasestimated using polygenic (pedigree based) infinitesimal models and it ranged from0.31 to 0.63. When using the genomic relationship matrix (i.e., genomic heritability),it ranged from 0.43 to 0.64. The body weight’s, heritability was also estimated usingpolygenic (pedigree based) infinitesimal models and it ranged from 0.54 to 0.75while using the genomic relationship matrix (i.e., genomic heritability), it rangedfrom 0.54 to 0.61. The breeding values were estimated using Best Linear UnbiasedPrediction (BLUP) and GBLUP (Genomic BLUP) approaches and they werecompared by estimating the correlations between EBVs and GEBVs. The correlationranged from 0.78 to 0.85 for stress response and from 0.85 to 0.91 for growth. Theaccuracy of selection (0.78 to 0.99) as well as the accuracy of estimation (0.32 to0.78) were estimated for all phenotypes, both provided better estimates when thegenomic relationship matrix was used. Selecting the 45% of the population asselection candidates, the genetic gain per generation was estimated for each trait andrevealed a better performance for the Genomic Selection (i.e., genomic relationshipmatrix model) (higher body weight at late growth stages and a better decrease instress bioindicator traits). Based on the genetic gain results, selection candidates forbreeders through Genomic Selection provided lower levels in all stress indicatortraits as well as higher body weight at the last two stages of growth compared totraditional approaches. The use of Genomic Selection indicates that a fasterimprovement per generation can be achieved, and all stress indicator traits can beused in order to improve the stress response in European seabass.Apart from the European seabass, limited information on growth performance existsfor new aquaculture species in the industry. Since new species are being used inaquaculture, genetic information which can be used in the development of a breedingstrategy could be beneficial in order to improve their production efficiency. Thus, aselected number of fish for the sharpsnout seabream (129 offspring) and the commondentex (131 offspring) were genotyped using a Next Generation stochasticgenotyping method (ddRAD) and phenotypes related to growth performance at thejuvenile stage were collected. Genetic data generated by the ddRAD were analyzedusing STACKS to discover SNP markers. The discovered SNPs were combined, andthe first linkage map supporting 24 linkage groups was constructed. Furthermore, apilot QTL analysis was performed for both species suggesting putative QTLaffecting growth at early stages. Even though, the QTL were not statisticallysignificant, a good trailing of the LOD score was identified using two differentmodels (with and without the effect of the polygenic component). When we searchfor the coordinates of those sequences of loci on reference genomes of other speciesof interest, they were found to be close to genes and other QTL reported to affectgrowth in other species. Two comparative analyses were performed between thestudied species and two species of interest respectively, gilthead seabream andEuropean seabass, in order to identify the genetic similarity between the species andto validate the quality of the new genetic maps. The genetic similarity between thestudied species and the gilthead seabream was higher to that of the respectivesimilarity with the European seabass. That was expected since the studied speciesare phylogenetically more related to the gilthead seabream. Finally, the quality ofthe two linkage maps was verified suggesting that high-quality results wereproduced. In conclusion, the genetic information which was found through the studyof the sharpsnout seabream and the common dentex provided evidence for QTLaffecting growth at early stages which can be utilized by the industry for theestablishment of a breeding program for those species. Moreover, the high similaritybetween gilthead seabream can be helpful for the further investigation of thosespecies since QTL that could be detected in gilthead seabream could be verified andutilized in future breeding programs in those new aquaculture species.
Degree
thesis:*- Grantor dc:publisher
- Aristotle University Of Thessaloniki (AUTH)
- Year dc:date
- 2021
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Oikonomou, Stavroula
- Οικονόμου, Σταυρούλα
Subjects
dc:subject × 27- Ευρωπαϊκό λαβράκι
- Μυτάκι
- Συναγρίδα
- Μικροδορυφορικοί δείκτες
- Mικροσυστοιχία μονοκλεοτιδικών πολυμορφισμών SNP
- Ανίχνευση QTL
- Ανάλυση ανίχνευσης συσχέτισης
- Απλότυπος
- Σωματικό βάρος
- Aντοχή στην καταπόνηση
- European seabass
- Sharpsnout seabream
- Common dentex
- Microsatellites
- SNP array
- QTL analysis
- GWAS
- Haplotype blocks
- Body weight
- Stress response
- Βιολογία
- Φυσικές Επιστήμες
- Γενετική και Κληρονομικότητα
- Biological Sciences
- Natural Sciences
- Biology
- Genetics and Heredity
Rights
- Language dc:language
- eng
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/51224