Genomic architecture of sickle cell disease clinical variation in children from West Africa : a case-control study design

Contexte : L’anémie falciforme ou drépanocytose est un problème de santé important, particulièrement pour les patients d’origine africaine. La variation phénotypique de l’anémie falciforme est problématique pour le suivi et le traitement des patients. L’architecture génomique responsable de cette variabilité est peu connue. Principe : Mieux saisir la contribution génétique de la variation clinique de cette maladie facilitera l’identification des patients à risque de développer des phénotypes sévères, ainsi que l’adaptation des soins. Objectifs : L’objectif général de cette thèse est de combler les lacunes relatives aux connaissances sur l’épidémiologie génomique de l’anémie falciforme à l’aide d’une cohorte issue au Bénin. Les objectifs spécifiques sont les suivants : 1) caractériser les profils d’expressions génomiques associés à la sévérité de l’anémie falciforme ; 2) identifier des biomarqueurs de la sévérité de l’anémie falciforme ; 3) identifier la régulation génétique des variations transcriptionelles ; 4) identifier des interactions statistiques entre le génotype et le niveau de sévérité associé à l’expression ; 5) identifier des cibles de médicaments pour améliorer l’état des patients atteints d’anémie falciforme. Méthode : Une étude cas-témoins de 250 patients et 61 frères et soeurs non-atteints a été menée au Centre de Prise en charge Médical Intégré du Nourrisson et de la Femme Enceinte atteints de Drépanocytose, au Bénin entre février et décembre 2010. Résultats : Notre analyse a montré que des profils d’expressions sont associés avec la sévérité de l’anémie falciforme. Ces profils sont enrichis de génes des voies biologiques qui contribuent à la progression de la maladie : l’activation plaquettaire, les lymphocytes B, le stress, l’inflammation et la prolifération cellulaire. Des biomarqueurs transcriptionnels ont permis de distinguer les patients ayant des niveaux de sévérité clinique différents. La régulation génétique de la variation de l’expression des gènes a été démontrée et des interactions ont été identifiées. Sur la base de ces résultats génétiques, des cibles de médicaments sont proposées. Conclusion: Ce travail de thèse permet de mieux comprendre l’impact de la génomique sur la sévérité de l’anémie falciforme et ouvre des perspectives de développement de traitements ciblés pour améliorer les soins offerts aux patients.

Identification of Genomic Variants Associated with Adolescent Idiopathic Scoliosis (AIS) in French-Canadian Population

La scoliose idiopathique est une déformation tridimensionnelle de la colonne vertébrale dont la pathogenèse reste obscure. Cette maladie affecte 2-4% des adolescents de 10-18 ans parmi les garçons et les filles. Il est à noter que les filles sont plus sévèrement affectées et ce en plus grand nombre que les garçons. Les études de jumeaux ont montré que les facteurs génétiques jouent un rôle important dans la scoliose idiopathique de l'adolescent (SIA). Depuis 2010, les études d'association pan génomiques ont été multipliées dans les recherches, visant à trouver des gènes candidats impliqués dans la SIA à travers des examens des polymorphismes nucléotidiques (SNPs). Un test génétique nommé "ScoliScore" a été publié pour essayer de prédire la progression de courbure dans la population caucasienne. Cependant, l'association n'a pas été reproduite dans une grande étude japonaise, soulignant l'importance d'une étude de réplication dans une population caucasienne indépendante. Dans ce contexte, mon projet de maîtrise a permis de génotyper plus de 1,4 millions de SNPs dans une cohorte canadienne-française dans le but: 1) de valider l'association de ScoliScoreTM; et 2) d’identifier les variants génomiques associées à la SIA dans la population québécoise. Notre étude a montré qu’aucun des variants constituant le test ScoliScoreTM n’était associé à la SIA. Ceci suggère que l'absence d'association dans une cohorte japonaise n'est pas due à l'appartenance ethnique. Aussi, nous avons identifié des variants génomiques associés significativement à l’initiation et/ou la progression de SIA dans la population québécoise, suggérant des gènes candidats impliqués dans la pathogenèse de SIA.

Development of a Fully Automated Image Analysis Method for High Density cDNA and array CGH Microarray Based Genomic Studies

In this thesis, different techniques for image analysis of high density microarrays have been investigated. Most of the existing image analysis techniques require prior knowledge of image specific parameters and direct user intervention for microarray image quantification. The objective of this research work was to develop of a fully automated image analysis method capable of accurately quantifying the intensity information from high density microarrays images. The method should be robust against noise and contaminations that commonly occur in different stages of microarray development.

Enlarging a training set for genomic selction by imputation of un-genotyped animals in populations of varying genetic architecture

Background: The most common application of imputation is to infer genotypes of a high-density panel of markers on animals that are genotyped for a low-density panel. However, the increase in accuracy of genomic predictions resulting from an increase in the number of markers tends to reach a plateau beyond a certain density. Another application of imputation is to increase the size of the training set with un-genotyped animals. This strategy can be particularly successful when a set of closely related individuals are genotyped. ----- Methods: Imputation on completely un-genotyped dams was performed using known genotypes from the sire of each dam, one offspring and the offspring’s sire. Two methods were applied based on either allele or haplotype frequencies to infer genotypes at ambiguous loci. Results of these methods and of two available software packages were compared. Quality of imputation under different population structures was assessed. The impact of using imputed dams to enlarge training sets on the accuracy of genomic predictions was evaluated for different populations, heritabilities and sizes of training sets. ----- Results: Imputation accuracy ranged from 0.52 to 0.93 depending on the population structure and the method used. The method that used allele frequencies performed better than the method based on haplotype frequencies. Accuracy of imputation was higher for populations with higher levels of linkage disequilibrium and with larger proportions of markers with more extreme allele frequencies. Inclusion of imputed dams in the training set increased the accuracy of genomic predictions. Gains in accuracy ranged from close to zero to 37.14%, depending on the simulated scenario. Generally, the larger the accuracy already obtained with the genotyped training set, the lower the increase in accuracy achieved by adding imputed dams. ----- Conclusions: Whenever a reference population resembling the family configuration considered here is available, imputation can be used to achieve an extra increase in accuracy of genomic predictions by enlarging the training set with completely un-genotyped dams. This strategy was shown to be particularly useful for populations with lower levels of linkage disequilibrium, for genomic selection on traits with low heritability, and for species or breeds for which the size of the reference population is limited.

Genomic architecture of adaptive color pattern divergence and convergence in Heliconius butterflies

Identifying the genetic changes driving adaptive variation in natural populations is key to understanding the origins of biodiversity. The mosaic of mimetic wing patterns in Heliconius butterflies makes an excellent system for exploring adaptive variation using next-generation sequencing. In this study, we use a combination of techniques to annotate the genomic interval modulating red color pattern variation, identify a narrow region responsible for adaptive divergence and convergence in Heliconius wing color patterns, and explore the evolutionary history of these adaptive alleles. We use whole genome resequencing from four hybrid zones between divergent color pattern races of Heliconius erato and two hybrid zones of the co-mimic Heliconius melpomene to examine genetic variation across 2.2 Mb of a partial reference sequence. In the intergenic region near optix, the gene previously shown to be responsible for the complex red pattern variation in Heliconius, population genetic analyses identify a shared 65-kb region of divergence that includes several sites perfectly associated with phenotype within each species. This region likely contains multiple cis-regulatory elements that control discrete expression domains of optix. The parallel signatures of genetic differentiation in H. erato and H. melpomene support a shared genetic architecture between the two distantly related co-mimics; however, phylogenetic analysis suggests mimetic patterns in each species evolved independently. Using a combination of next-generation sequencing analyses, we have refined our understanding of the genetic architecture of wing pattern variation in Heliconius and gained important insights into the evolution of novel adaptive phenotypes in natural populations.

Bacterial evolution by genomic island transfer occurs via DNA transformation in planta

Our understanding of the evolution of microbial pathogens has been advanced by the discovery of "islands" of DNA that differ from core genomes and contain determinants of virulence [1, 2]. The acquisition of genomic islands (GIs) by horizontal gene transfer (HGT) is thought to have played a major role in microbial evolution. There are, however, few practical demonstrations of the acquisition of genes that control virulence, and, significantly, all have been achieved outside the animal or plant host. Loss of a GI from the bean pathogen Pseudomonas syringae pv. phaseolicola (Pph) is driven by exposure to the stress imposed by the plant's resistance response [3]. Here, we show that the complete episomal island, which carries pathogenicity genes including the effector avrPphB, transfers between strains of Pph by transformation in planta and inserts at a specific att site in the genome of the recipient. Our results show that the evolution of bacterial pathogens by HGT may be achieved via transformation, the simplest mechanism of DNA exchange. This process is activated by exposure to plant defenses, when the pathogen is in greatest need of acquiring new genetic traits to alleviate the antimicrobial stress imposed by plant innate immunity [4].

Genomic and genetic analyses of diversity and plant interactions of Pseudomonas fluorescens

Background: Pseudomonas fluorescens are common soil bacteria that can improve plant health through nutrient cycling, pathogen antagonism and induction of plant defenses. The genome sequences of strains SBW25 and Pf0-1 were determined and compared to each other and with P. fluorescens Pf-5. A functional genomic in vivo expression technology (IVET) screen provided insight into genes used by P. fluorescens in its natural environment and an improved understanding of the ecological significance of diversity within this species. Results: Comparisons of three P. fluorescens genomes (SBW25, Pf0-1, Pf-5) revealed considerable divergence: 61% of genes are shared, the majority located near the replication origin. Phylogenetic and average amino acid identity analyses showed a low overall relationship. A functional screen of SBW25 defined 125 plant-induced genes including a range of functions specific to the plant environment. Orthologues of 83 of these exist in Pf0-1 and Pf-5, with 73 shared by both strains. The P. fluorescens genomes carry numerous complex repetitive DNA sequences, some resembling Miniature Inverted-repeat Transposable Elements (MITEs). In SBW25, repeat density and distribution revealed 'repeat deserts' lacking repeats, covering approximately 40% of the genome. Conclusions: P. fluorescens genomes are highly diverse. Strain-specific regions around the replication terminus suggest genome compartmentalization. The genomic heterogeneity among the three strains is reminiscent of a species complex rather than a single species. That 42% of plant-inducible genes were not shared by all strains reinforces this conclusion and shows that ecological success requires specialized and core functions. The diversity also indicates the significant size of genetic information within the Pseudomonas pan genome.

Non-genomic signalling of the retinoic X receptor through inhibition of Gq signalling in human platelets

Retinoid X receptors (RXRs) are important transcriptional nuclear hormone receptors, acting as either homodimers or the binding partner for at least one fourth of all the known human nuclear receptors. Functional nongenomic effects of nuclear receptors are poorly understood; however, recently peroxisome proliferator-activated receptor (PPAR) gamma, PPARbeta, and the glucocorticoid receptor have all been found active in human platelets. Human platelets express RXRalpha and RXRbeta. RXR ligands inhibit platelet aggregation and TXA(2) release to ADP and the TXA(2) receptors, but only weakly to collagen. ADP and TXA(2) both signal via the G protein, Gq. RXR rapidly binds Gq but not Gi/z/o/t/gust in a ligand-dependent manner and inhibits Gq-induced Rac activation and intracellular calcium release. We propose that RXR ligands may have beneficial clinical actions through inhibition of platelet activation. Furthermore, our results demonstrate a novel nongenomic mode for nuclear receptor action and a functional cross-talk between G-protein and nuclear receptor signaling families.

Genomic analysis of recombinant Sabin clinical isolates

Recombination in Poliovirus vaccine strains is a very frequent phenomenon. In this report 23 polio/Sabin strains isolated from healthy vaccinees or from VAPP patients after OPV administration, were investigated in order to identify recombination sites from 2C to 3D regions of the poliovirus genome. RT-PCR, followed by Restriction Fragment Length Polymorphism (RFLP) screening analysis were applied in four distant genomic regions (5' UTR, VP1, 2C and 3C-3D) in order to detect any putative recombinant. The detected recombinants were sequenced from 2C to the end of the genome (3' UTR) and the exact recombination sites were determined with computational analysis. Five of the 23 isolated strains were recombinant in one genomic region, two of them in 2C, isolates EP16:S3/S2, EP23:S3/S1, two in 3D isolates EP6:S2/S1, EP12:S2/S1 and one in 3A isolate EP9:S2/Sl. Point mutations were found in strains EP3, EP6, EP9 and EP12. Recombination specific types and sites re-occurrence along with point mutations are discussed concerning the polioviruses evolution.

On the structural differences between markers and genomic AC microsatellites

AC microsatellites have proved particularly useful as genetic markers. For some purposes, such as in population biology, the inferences drawn depend on the quantitative values of their mutation rates. This, together with intrinsic biological interest, has led to widespread study of microsatellite mutational mechanisms. Now, however, inconsistencies are appearing in the results of marker-based versus non-marker-based studies of mutational mechanisms. The reasons for this have not been investigated, but one possibility, pursued here, is that the differences result from structural differences between markers and genomic microsatellites. Here we report a comparison between the CEPH AC marker microsatellites and the global population of AC microsatellites in the human genome. AC marker microsatellites are longer than the global average. Controlling for length, marker microsatellites contain on average fewer interruptions, and have longer segments, than their genomic counterparts. Related to this, marker microsatellites show a greater tendency to concentrate the majority of their repeats into one segment. These differences plausibly result from scientists selecting markers for their high polymorphism. In addition to the structural differences, there are differences in the base composition of flanking sequences, marker flanking regions being richer in C and G and poorer in A and T. Our results indicate that there are profound differences between marker and genomic microsatellites that almost certainly affect their mutation rates. There is a need for a unified model of mutational mechanisms that accounts for both marker-derived and genomic observations. A suggestion is made as to how this might be done.

The chromosomal assessment of salt tolerant substituted tritipyrum using genomic fluorescent in situ hybridiization (FISH)

Wheat, although moderately tolerant to salt, can not be cultivated in many areas. However, in the triticeae tribe, some of the wild wheat relatives are highly tolerant, e.g. Thinopyrum bessarabicum, which grows on the sea shore. Eight primary hexaploid tritipyrum lines, amphiploids between Triticum durum and Thinopyrum bessarabicum have been produced which can set seed in at least 250 mM NaCl. These tritipyrums (2n=6x=42, AABBEbEb) due to reasons such as brittle rachis, continuous production of tillers, late maturity, tall stature and meiotic instability will not fulfill the requirements of a successful commercial salt tolerant crop. To overcome such problems the substituted tritipyrum, in which selected Eb chromosomes are replaced by D genome chromosomes of 6x wheat, was produced from 6x tritipyrum x 6x wheat hybrids (F1: 2n=6x=42, AABBDEb) followed by selfing and backcrossing with 6x tritipyrum. The fertile plants among the above progenies were screened by the genomic fluorescent in situ hybridization technique to identify their Eb and D chromosome constitution. This study showed that producing tritiprum with variable numbers of Eb and D genome chromosomes is feasible and that FISH is a useful technique for determining the number of Eb chromosomes present.

Confocal imaging of pseudomonas syringae pv. phaseolicola colony development in bean reveals reduced multiplication of strains containing the genomic island PPHGI-1.

Pseudomonas syringae pv. phaseolicola is the seed borne causative agent of halo blight in the common bean Phaseolus vulgaris. Pseudomonas syringae pv. phaseolicola race 4 strain 1302A contains the avirulence gene hopAR1 (located on a 106-kb genomic island, PPHGI-1, and earlier named avrPphB), which matches resistance gene R3 in P. vulgaris cultivar Tendergreen (TG) and causes a rapid hypersensitive reaction (HR). Here, we have fluorescently labeled selected Pseudomonas syringae pv. phaseolicola 1302A and 1448A strains (with and without PPHGI-1) to enable confocal imaging of in-planta colony formation within the apoplast of resistant (TG) and susceptible (Canadian Wonder [CW]) P. vulgaris leaves. Temporal quantification of fluorescent Pseudomonas syringae pv. phaseolicola colony development correlated with in-planta bacterial multiplication (measured as CFU/ml) and is, therefore, an effective means of monitoring Pseudomonas syringae pv. phaseolicola endophytic colonization and survival in P. vulgaris. We present advances in the application of confocal microscopy for in-planta visualization of Pseudomonas syringae pv. phaseolicola colony development in the leaf mesophyll to show how the HR defense response greatly affects colony morphology and bacterial survival. Unexpectedly, the presence of PPHGI-1 was found to cause a reduction of colony development in susceptible P. vulgaris CW leaf tissue. We discuss the evolutionary consequences that the acquisition and retention of PPHGI-1 brings to Pseudomonas syringae pv. phaseolicola in planta.