<p>The domestication of the pig began about 9 000 years ago and many of the existing domestic breeds have been selected for phenotypic traits like lean meat and fast growth. Domestic pigs are phenotypically very different from the ancestral wild boar that has adapted to survive in their natural environment. Because of their divergence, crosses between domestic pigs and wild boars are suitable for constructing genetic maps and Quantitative trait locus (QTL) analyses. A cross between the Large White and the European wild boar was thus initiated in the late 1980s. A major QTL for fat deposition and growth, denoted <i>FAT1</i>, was found on chromosome 4. The aim of this thesis was to further characterise the <i>FAT1</i> locus and to identify the causative gene(s) and mutation(s). We have identified new markers and constructed a high-resolution linkage and RH map of the <i>FAT1</i> QTL interval. We also performed comparative mapping to the human genome and showed that the pig chromosome 4 is homologous to human chromosomes 1 and 8. The gene order is very well conserved between the two species. In parallel we have narrowed down the <i>FAT1</i> QTL interval by repeated backcrossing to the domestic Large White breed for six generations. The QTL could be confirmed for fatness but not for growth. Furthermore, the data strongly suggested that there might be more than one gene underlying the <i>FAT1</i> QTL. Depending on which hypothesis to consider, the one- or two-loci model, the <i>FAT1</i> interval can be reduced to 3,3 or 20 centiMorgan (cM), respectively, based on the backcross experiments. In the last study we confirm the two-loci model with one locus primarily effecting abdominal fat and another locus primarily effecting subcutaneous fat. We have identified a missense mutation in the <i>RXRG</i> gene which is in strong association with the abdominal fat QTL and the mutation is a potential candidate for that locus.</p><p>Brown adipose tissue (BAT) is a specific type of fat essential for non-shivering thermogenesis in mammals. Piglets appear to lack BAT and rely on shivering as the main mechanism for thermoregulation. Uncoupling protein 1 (<i>UCP1</i>) gene is exclusively expressed in BAT and its physiological role is to generate heat by uncoupling oxidative phosphorylation. We show that the <i>UCP1</i> gene has been disrupted in the pig lineage about 20 years ago. The inactivation of <i>UCP1</i> provides a genetic explanation for the poor thermoregulation in piglets. </p>

The domestication of the pig began about 9 000 years ago and many of the existing domestic breeds have been selected for phenotypic traits like lean meat and fast growth. Domestic pigs are phenotypically very different from the ancestral wild boar that has adapted to survive in their natural environment. Because of their divergence, crosses between domestic pigs and wild boars are suitable for constructing genetic maps and Quantitative trait locus (QTL) analyses. A cross between the Large White and the European wild boar was thus initiated in the late 1980s. A major QTL for fat deposition and growth, denoted FAT1, was found on chromosome 4. The aim of this thesis was to further characterise the FAT1 locus and to identify the causative gene(s) and mutation(s). We have identified new markers and constructed a high-resolution linkage and RH map of the FAT1 QTL interval. We also performed comparative mapping to the human genome and showed that the pig chromosome 4 is homologous to human chromosomes 1 and 8. The gene order is very well conserved between the two species. In parallel we have narrowed down the FAT1 QTL interval by repeated backcrossing to the domestic Large White breed for six generations. The QTL could be confirmed for fatness but not for growth. Furthermore, the data strongly suggested that there might be more than one gene underlying the FAT1 QTL. Depending on which hypothesis to consider, the one- or two-loci model, the FAT1 interval can be reduced to 3,3 or 20 centiMorgan (cM), respectively, based on the backcross experiments. In the last study we confirm the two-loci model with one locus primarily effecting abdominal fat and another locus primarily effecting subcutaneous fat. We have identified a missense mutation in the RXRG gene which is in strong association with the abdominal fat QTL and the mutation is a potential candidate for that locus. Brown adipose tissue (BAT) is a specific type of fat essential for non-shivering thermogenesis in mammals. Piglets appear to lack BAT and rely on shivering as the main mechanism for thermoregulation. Uncoupling protein 1 (UCP1) gene is exclusively expressed in BAT and its physiological role is to generate heat by uncoupling oxidative phosphorylation. We show that the UCP1 gene has been disrupted in the pig lineage about 20 years ago. The inactivation of UCP1 provides a genetic explanation for the poor thermoregulation in piglets.

As espécies selvagens relacionadas ao tomateiro (Lycopersicon esculentum Mill. Syn. Solanum lycopersicum L) em latitudes do Sul do Equador ao norte do Chile. Estas condições ambientais contrastantes possibilitaram o aparecimento de grande diversidade genética dentre estas espécies. Lycopersicon cheesmanii f .minor (Hook.f.) C. H. Hill. Syn. Solanum galapagense S. Darwin & Peralta é endêmica das Ilhas Galápagos e possui características peculiares, como porte reduzido, dormência de sementes e folhas bastante recortadas. Estas características também estão presentes em mutantes de tomateiro com deficiência no hormônio giberelina (GA). Mutações em GA como gib1, gib2, gib3 (deficientes) e procera (resposta constitutiva) são amplamente conhecidas em tomateiro. Já os alelos Sp (Self Pruning) e Pts (Petroselinum), presentes em S. galapagense, alteram altura e recorte foliar, respectivamente. Visando entender a natureza destas variações genéticas naturais, foram feitos cruzamentos e retrocruzamentos sucessivos de S. galapagense com a cultivar miniatura de tomateiro Micro-Tom (MT), onde tentou se isolar plantas segregando para características presentes no parental selvagem (dormência, nanismo e maior recorte foliar). Na geração BC1F2 foram selecionados indivíduos com porte menor que MT e folhas com bordos muito recortados. Sementes BC1F3 apresentaram taxa de germinação de 47,6 %, contrastando com o valor 94,5 % apresentado por MT. Entretanto, após aplicação de 100 µM de GA3 a taxa de germinação de BC1F3 foi elevada para 72 %. Em análise de curvas de dose-resposta a GA, as plantas BC1F3 apresentaram menor porte que MT, sendo que este nanismo não foi completamente revertido aplicando-se GA. Em gerações avançadas de retrocruzamentos, constatou-se que a dormência de sementes foi perdida durante as introgressões. Tal constatação leva a crer que o principal componente da dormência de S. galapagense não está ligado ou é efeito pleiotrópico dos genes que respondem pelo seu nanismo ou recorte foliar. A análise de segregação de 278 plantas BC4F2 mostrou que o principal componente do nanismo de S. galapagense segrega na proporção 3:1, sendo a mutação recessiva presente em S. galapagense denominada galapagos dwarf (gdw). Plantas quase isogênicas a MT (geração BC6Fn) foram obtidas carregando os alelos Pts, Sp e gdw. Experimentos comparando-as com MT confirmaram que o alelo Pts aumenta pronunciadamente o recorte foliar de tomateiro e diminui ligeiramente a germinação e o porte das plantas. Tal observação está de acordo com a descoberta recente de que PTS codifica para um gene da classe KNOX, que podem estar envolvidos com GA. Em cominação com Pts, esse alelo parece ter efeito discreto no recorte foliar, mas somente em combinação com Pts. Surpreendentemente, sementes Sp tiveram germinação precoce, comparadas com MT (sp). Já o alelo gdw não mostrou ter efeito na germinação, mas confirmou afetar a altura e o recorte foliar. Esses resultados evidenciam que porte reduzido e folhas bastante recortadas de S. galapagense podem ser atribuídos principalmente a gdw e Pts. A dormência parece ser controlada por outro(s) gene(s) ainda desconhecido(s). Como o novo gene descoberto, GDW, não afeta a germinação, é pouco provável que esteja ligado a GA, podendo ser uma nova classe de genes controlando nanismo. / The tomato (Lycopersicon esculentum Mill. Syn. Solanum lycopersicum L) related wild species evolved into a wide range of latitudes, from the southern of Ecuador to the northern of Chile. These contrasting environment conditions allowed the emergence of great genetic diversity among these species. Lycopersicon cheesmanii f. minor (Hook.f.) C. H. Hill. Syn. Solanum galapagense S. Darwin & Peralta is endemic in the Galapagos Islands and has characteristics such as small plant size, dormant seeds and profusely divided leaves, being these characteristics also common in tomato gibberellin (GA) mutants. GA mutants such as gib1, gib2, gib3 (deficient), and procera (constitutive) are widely known in tomato. In addition to the afore mentioned mutations, the alleles Sp (Self Pruning) and Pts (Petroselinum), present S. galapagense, lead to changes in plant height and leaf architecture, respectively. Aiming at the understanding of such natural genetic variations, successive crosses and backcrosses of S.galapagense with the miniature tomato cultivar Micro-Tom (MT) were made, attempting to isolate plants segregating characteristics of the parental wild species (dormancy, dwarfism, and increased leaf indentation). In the BC1F2 generation were selected individuals shorter than MT, and with more serrated leaf margins or with more leaflets when compared to MT. BC1F3 seeds showed germination rates of only 46.7 %, contrasting with the 94.5% of MTs germination. However, after the application of 100 µM GA, the rate of germination of BC1F3 was increased to 72 %. Analysis GA dose-responses showed that the BC1F3 plants displayed smaller sizes than the MT plants, and this dwarfism was not completely reversed by GA applications. In advanced backcross generations, it was found that the dormancy of seeds was lost during the process of introgression, since the selection was not for this trait, but only for plant with small size and/or more divided leaves. This finding suggests that the main component of dormancy of S.galapagense is not connected to or is not a pleiotropic effect of the genes for dwarfism and leaf shape. The analyzes of the segregation of 278 BC4F2 plants, already harboring the alleles sp and pts from MT, showed that the main component of the dwarfism of S. galapagense segregates in the proportion 3:1, and then, the recessive mutation present in this specie was named galapagos dwarf (gdw). Plants nearly isogenic to MT (BC6Fn generation) were obtained carrying the alleles Pts, Sp and gdw. Experiments comparing these plants with MT confirmed that the allele Pts produces more divided leaf and leaflets, and showed that this allele has a slight effect in reducing seed germination and plant size. This observation is consistent with the recent discovery that Pts codes for a gene from the KNOX class, which may be involved with the GA hormone. The Sp allele had no direct effect on plant height, but only indirectly due to its indeterminate growth. This allele seems to have a slight effect on leaf shape, but only in combination with Pts. Surprisingly, Sp seeds had an early germination when compared to MT (sp). On the other hand, the allele gdw did not show any effect on germination, but confirmed to affect plant height and leaf architecture. Taken together, these results showed that the reduced plant size and the profusely divided leaves of S. galapagense can be attributed mainly to the effect of Pts and gdw alleles. Regarding seed dormancy, this trait appears to be controlled by other(s) gene(s), yet unknown, although Pts also contributes to this characteristic. Since the new gene that we discovered, GDW, does not affect the germination, it is unlikely that it can be linked the GA hormone, but it may represent a new class of genes controlling an important agronomic trait, the dwarfism.

Les allèles en ségrégation des deux génomes Elaeis ont été tracés dans un pseudo backcross interspécifique Elaeis (E. oleifera x E. guineensis) x E. guineensis. Dix-neuf QTL de proportion d'acides gras de l'huile de palme par rapport à 16 QTL des mêmes caractères d'un croisement intraspécifique E. guineensis ont mis en évidence des QTL communs ou spécifiques aux deux espèces. Aucune corrélation n'existe entre proportions d'acides gras et le rendement en huile de palme. Le déterminisme génétique et les principaux gènes associés aux acides gras du mésocarpe ont été étudiés, à l'aide de banques d'ADNc pleine longueur et des extraits d'ARNm de mésocarpe de fruits en développement des espèces Elaeis chacune représentée par quatre fonds génétiques. L'expression de 113 gènes Elaeis clés ou de facteurs putatifs de régulation des voies de synthèse de novo des acides gras et des triacylglycérols fut caractérisée au niveau de l'espèce par analyse in silico de profils de transcription et par analyse Real Time RT-qPCR. Les gènes ont été cartographiés sur le pseudo backcross avec 180 marqueurs SNP intra-géniques. Quarante-huit gènes étaient exprimés différemment entre espèces dont 94% plus exprimés chez E. guineensis. Le facteur de transcription EgAP2-2 (WRI1-like) influence le début de la synthèsede novo des acide gras par action principale sur les gènes MAT et KAS III, parait indépendant de l'expression d'autres gènes et non différentiel entre espèces. Les variations en C16:0 et C18:0 entre espèces s'expliquent par niveau d'expression allélique et l'activité enzymatique relative des isoformes présentes dans le génome de FATB (2), KAS II (1) et SAD (3). / An Elaeis interspecific pseudo-backcross of first generation (E. oleifera x E. guineensis) x E. guineensis allowed tracing segregating alleles from both Elaeis genomes. Nineteen quantitative trait loci (QTLs) associated to palm oil fatty acid proportions compared to 16 QTLs of same traits in an intra-specific oil palm cross evidenced common or specific QTLs in E. guineensis and E. oleifera. No correlation was found in oil palm between mesocarp fatty acid proportions and yieldtraits. The genetic determinism of main fatty acid proportions was confirmed. Genes related to palm oil fatty acids were investigated using full-length cDNA libraries and mRNA extracts from the mesocarp of developing fruits in each Elaeis species represented by four genetic pools. Expression of 113 key Elaeis genes or putative regulation factors of de novo fatty acid and triacylglycerol pathways were characterized at species level by in silico transcript profiling and Real Time RT-qPCR analysis. Genes were mapped on the pseudo-backcross using 180 intra-gene SNP markers. Forty-eight genes were differentially expressed between Elaeis species, with 94% over expressed in E. guineensis. The EgAP2-2 (WRI1-like) transcription factor might influence the start of the de novo fatty acid pathway by main action only on the MAT and KAS III genes, as independent in expression from other genes and not differential between species. Between species variations in C16:0 and C18:0 can be deciphered by relative expression levels and enzyme activities of the isoforms in the genome of FATB (2), KAS II (1) and SAD (3). An oil palm producing more oleic acid in proportion than pure E. oleifera is feasible.

Sunflower (Helianthus annuus L.) high oleic acid content lines differ from conventional sunflower by an increase in oleic acid (C18:1) content of more than 60%. The current sunflower cultivars under production in South Africa are standard sunflower with high levels of linoleic acid (C18:2). The aim of this study was to improve the quality of oil produced by local sunflower germplasm with respect to oleic acid through employing a marker-assisted breeding technique to facilitate and speed up the recovery of the high oleic acid allele into the background of the recurrent parent genome. Eleven sunflower breeding genotypes with high and low oleic acid traits were obtained from the Agricultural Research Council-Grain Crops Institute (ARC-GCI) in Potchefstroom. The breeding genotypes were phenotypically characterised based on their oleic and linoleic acid levels using gas chromatography. Results demonstrated that the average mean of oleic and linoleic acid contents in high oleic acid genotypes were 72% and 17% respectively, while the average mean of oleic acid and linoleic acid contents in wild type lines were 33.5 % and 54 % respectively. These results indicated a perfect negative correlation between the amount of oleic and linoleic acids possessed in high and low oleic acid genotypes (R2 = -99.16%). Sequence characterised amplified region (SCAR) markers were tested to ascertain if any of the ten available dominant FAD2-1 markers was segregating with the high oleic acid allele. Four dominant SCAR markers (FAD2-1F4/R1; FAD2-1F4/R2; FAD2-1F13/R1; FAD2-1F14/R2) were strongly associated with the high oleic acid trait (P< 0.001). With regard to the inheritance of the high oleic acid trait, 143 plants of the F2 segregating population derived from a cross between the high oleic acid parent (AP901-95-3-4-1) and low oleic acid parent (H55-9-2-1-1) were genotyped with the four SCAR markers to determine the genetic state concerning the high oleic acid gene (Ol). Results from a Chi square analysis of the observed frequencies of each dominant FAD2-1 marker locus in 143 F2 individuals indicated that the deviation from the expected ratio of 3:1 (high to low oleic acid) was not statistically significant (P< 0.95) from the observed segregation ratio. These results were consistent with the previous finding that an incomplete dominant gene governs sunflower high oleic acid. A multiplex assay of 78 Simple sequence repeat (SSR) markers was optimised and evaluated on 143 plants of the F2 population to determine suitable SSR markers that can be used in a marker-assisted background selection. Only 14 markers were suitable for marker-assisted background selection based on their high polymorphic information content, allele frequency and maximum allele numbers. In conclusion, this study demonstrated the potential of using SSR and SCAR marker systems as a breeding tool to characterise and speed up the selection process in marker-assisted backcross breeding. / Thesis (MSc (Botany))--North-West University, Potchefstroom Campus, 2013.

Sunflower (Helianthus annuus L.) high oleic acid content lines differ from conventional sunflower by an increase in oleic acid (C18:1) content of more than 60%. The current sunflower cultivars under production in South Africa are standard sunflower with high levels of linoleic acid (C18:2). The aim of this study was to improve the quality of oil produced by local sunflower germplasm with respect to oleic acid through employing a marker-assisted breeding technique to facilitate and speed up the recovery of the high oleic acid allele into the background of the recurrent parent genome. Eleven sunflower breeding genotypes with high and low oleic acid traits were obtained from the Agricultural Research Council-Grain Crops Institute (ARC-GCI) in Potchefstroom. The breeding genotypes were phenotypically characterised based on their oleic and linoleic acid levels using gas chromatography. Results demonstrated that the average mean of oleic and linoleic acid contents in high oleic acid genotypes were 72% and 17% respectively, while the average mean of oleic acid and linoleic acid contents in wild type lines were 33.5 % and 54 % respectively. These results indicated a perfect negative correlation between the amount of oleic and linoleic acids possessed in high and low oleic acid genotypes (R2 = -99.16%). Sequence characterised amplified region (SCAR) markers were tested to ascertain if any of the ten available dominant FAD2-1 markers was segregating with the high oleic acid allele. Four dominant SCAR markers (FAD2-1F4/R1; FAD2-1F4/R2; FAD2-1F13/R1; FAD2-1F14/R2) were strongly associated with the high oleic acid trait (P< 0.001). With regard to the inheritance of the high oleic acid trait, 143 plants of the F2 segregating population derived from a cross between the high oleic acid parent (AP901-95-3-4-1) and low oleic acid parent (H55-9-2-1-1) were genotyped with the four SCAR markers to determine the genetic state concerning the high oleic acid gene (Ol). Results from a Chi square analysis of the observed frequencies of each dominant FAD2-1 marker locus in 143 F2 individuals indicated that the deviation from the expected ratio of 3:1 (high to low oleic acid) was not statistically significant (P< 0.95) from the observed segregation ratio. These results were consistent with the previous finding that an incomplete dominant gene governs sunflower high oleic acid. A multiplex assay of 78 Simple sequence repeat (SSR) markers was optimised and evaluated on 143 plants of the F2 population to determine suitable SSR markers that can be used in a marker-assisted background selection. Only 14 markers were suitable for marker-assisted background selection based on their high polymorphic information content, allele frequency and maximum allele numbers. In conclusion, this study demonstrated the potential of using SSR and SCAR marker systems as a breeding tool to characterise and speed up the selection process in marker-assisted backcross breeding. / Thesis (MSc (Botany))--North-West University, Potchefstroom Campus, 2013.