Thesis (PhD)--Stellenbosch University, 2004. / ENGLISH ABSTRACT: Progressive familial heart block I (PFHBI) is an inherited autosomal dominant cardiac conduction disorder which segregates in a large South African (SA) pedigree, two smaller SA families and a Lebanese family. It specifically affects conduction in the ventricles and is of unknown cause. Clinically, PFHBI is detected on electrocardiogram (ECG) by evidence of bundle-branch disease, i.e., as right bundle branch block, left anterior or posterior hemiblock, or complete heart block with broad QRS complexes. The PFHBI-causative gene was mapped to a lOcM region on chromosome 19ql3.3 using linkage analysis, and the locus was subsequently reduced to 7cM by genetic fine mapping. The present study involved a multi-strategy approach to search for the PFHBI gene. The objectives were the further reduction of the PFHBI locus by genetic fine mapping using published and novel markers, searching for short gene transcripts from publicly available databases and the generation of an integrated map of the locus to which genes were mapped. Prioritised genes were screened for PFHBI-causing mutations and, in addition, the PFHBI locus was searched for the presence of a G protein-encoding gene (PI 15- RhoGEF), a connexin (Cx) gene and any genes containing a CTG repeat expansion motif, since these genes are plausible PFHBI candidate genes. Genotyping and fine genetic mapping using known and novel polymorphic dinucleotide (CA)n and novel tetranucleotide (A3G)n repeat markers across the PFHBI locus were performed. Publicly available databases, such as LLNL (Livermore, USA), and GENEMAP (NCBI) were searched for ESTs which, in turn, were extended using clustering programmes, such as UNIGENE (NCBI) and STACK (SANBI), and the resulting consensus sequences were subsequently BLAST-searched against the protein databases. Using the available data, an integrated physical and genetic map of the PFHBI locus was generated and, as the HGP progressed, a number of novel genes were placed thereon. Subsequently, genes were prioritised on the basis of position, function and expression profile. Genetic fine mapping reduced the PFHBI locus from 7cM to 4cM. The EST approach yielded 38 ESTs, of which 24 ESTs matched proteins, such as activating transcription factor 5 (ATF5), actin-binding protein (KPTN) and zinc finger protein 473 (ZFP473) (May 2003). All the map data generated experimentally and computationally were placed on the PFHBI map. The PI 15-RhoGEF was excluded as a PFHBI candidate gene and although homologous sequences to connexin 37 (Cx37) was located on both chromosome 19 radiation hybrid clones (RHG12 and ORIM-7), it was not identified on the DNA clones spanning the PFHBI locus. No evidence of an expansion of a CTG repeat motif sequence in PFHBI-affected individuals was found. Five highly prioritised candidate genes, namely, 5CZ2-associated X protein (BAX), potassium voltage-gated channel Shaker-related subfamily member 7 (KCNA7’), potassium inwardly-rectifying channel, subfamily J, member 14 (KIR2.4), lin-7 homolog B {LIN-7B) and glycogen synthase 1 (GSYI) were selected for mutation screening. No disease associated mutations were identified in the exonic and flanking intronic regions of these genes. In summary, this study reduced the PFHBI locus substantially and generated a detailed map of the region. A number of attractive candidate genes were excluded from causing PFHBI; however, several plausible candidate genes are still present at this gene-rich locus and remain to be screened. Identifying the PFHBI-causative gene and associated mutation will provide a platform for further studies to understand the pathophysiology, not only of PFHBI, but also of other more commonly occurring conduction disturbances. / AFRIKAANSE OPSOMMING: Progressiewe familiele hartblok I (PFHBI) is ‘n autosomaal dominant oorerflike kardiale geleidingstoomis wat in ‘n groot Suid-Afrikaanse (SA) familie, twee kleiner SA families en ‘n Lebanese familie segregeer. Dit affekteer hoofsaaklik die geleiding in die ventrikels en die oorsaak daarvan is onbekend. Klinies word PFHBI op elektrokardiogram (EKG) geidentifiseer as a bondeltak-siekte, naamlik, as regter bondeltakblok, linker anterior of posterior hemiblok, of volledige hartblok met wye QRS komplekse. Die PFHBI-veroorsakende geen is voorheen deur koppelingsanalise tot ‘n lOcM gebied op chromosoom 19ql3.3 gekarteer, en daaropvolgens is die lokus verklein tot 7cM deur genetiese fyn kartering. Die huidige studie behels ‘n veelvuldige-strategie benadering in die soektog na die PFHBI geen. Die doel van die studie was die verdere verkleining van die PFHBI lokus deur gebruik te maak van beide gepubliseerde en nuwe genetiese merkers, die identifisering van kort geentranskripte (ESTs) uit publieke databanke en die generasie van ‘n geintegreerde kaart van die lokus. Geprioritiseerde gene is geanaliseer vir die PFHBI-veroorsakende mutasie en, daarby, is die PFHBI lokus deursoek vir die teenwoordigheid van ‘n G proteien-enkodeeringsgeen (PIJ5-RhoGEF), ‘n konneksien (Kx) geen en enige gene wat ‘n uitgebreide CTG-herhalingsmotief bevat, aangesien hierdie gene as sterk PFHBI kandidaatgene geag is. Genotipering en fynkartering deur die gebruik van bekende asook nuwe polimorfiese dinukleotied- [(CA)n] en nuwe tertranukleotied- [(A3G)n] herhalingsmerkers wat die PFHBI lokus oorbrug, is uitgevoer. Publieke databanke, soos LLNL (Livermore, USA), en GENEMAP (NCBI) is ondersoek vir ESTs wat vervolgens verleng is deur gebruik te maak van groeperende programme soos UNIGENE (NCBI) en STACK (SANBI) en die gevolglike konsensus volgordes is daama met behulp van BLAST geanaliseer teen die protei'endatabanke. Die bekomde data is vervolgens gebruik om ‘n geintegreerde fisiese en genetiese kaart van die PFHBI lokus te produseer en, soos die mens genoomprojek gevorder het, is nuwe gene daarop geplaas. Daarna is gene geprioritiseer vir mutasie analise gebaseer op posisie, funksie en uitdrukkingsprofiele. Genetiese fynkartering het die PFHBI lokus van 7cM tot 4cM verklein. Die EST benadering het 38 ESTs gei'dentifiseer, waarvan 24 ESTs proteien gelyke gehad het, bv aktiverende transkripsie faktor 5 (ATF5), aktien-verbindingsprotei'en (KPTN) en sink-vingerproteien 473 (ZFP473) (Mei 2003). A1 die karterings data wat eksperimenteel en rekenaar-gewys gegenereer is, is op die PFHBI kaart geposisioneer. Die P115-RhoGEF is uitgeskakel as ‘n PFHBI kandidaatgeen en alhoewel ’n volgorde met homologie aan konneksien37 (Kx37) gevind is op albei chromosoom 19 radiasiehibried klone (RGH12 and ORIM-7), is dit nie gei'dentifiseer in die DNS klone wat die PFHBI lokus oorbrug nie. Geen bewyse van uitbreiding van CTG herhalingsmotiewe is gevind in PFHBIaangetasde persone nie. Vyf hoogs-geprioritiseerde kandidaat gene, naamlik, BCL2-geassosieerde X proteien (BAX), kalium spanningsbeheerde kanaal, subfamilie J, lid 14 (KIR2.4), lin-7 homoloog B (LIN-7b) en glikogeen sintase 1 (GYS1), is geselekteer vir mutasie-analise. Geen siekteveroorsakende mutasie is egter gei'dentifiseer in die eksoniese of die naasliggende introniese gebiede van hierdie gene nie. Ter opsomming, hierdie studie het die PFHBI lokus verklein en het ‘n omvattende kaart van die gebied gegenereer. Verskillende kandidaat gene is uitgesluit as die oorsaak van PFHBI, alhoewel daar nog heelwat goeie kandidaat gene in hierdie geen-ryke lokus is wat geanaliseer behoort te word. Die identifiseering van die PFHBI-veroorsakende mutasie sal ‘n platform bied vir verdere studies om die patofisiologie van nie alleen PFHBI nie, maar ook meer algemene geleidingstoomisse, te verstaan.

Dissertation (PhD)--University of Stellenbosch, 2004. / ENGLISH ABSTRACT: Heart failure due to cardiomyopathy or cardiac conduction disease is a major cause of mortality and morbidity in both developed and developing countries. Although defined as separate clinical entities, inherited forms of cardiomyopathies and cardiac conduction disorders have been identified that present with overlapping clinical features and/or have common molecular aetiologies. The objective of the present study was to identify the molecular cause of progressive familial heart block type II (PFHBII), an inherited cardiac conduction disorder that segregates in a South African Caucasian Afrikaner family (Brink and Torrington, 1977). The availability of family data tracing the segregation of PFHBII meant that linkage analysis could be employed to identify the chromosomal location of the disease-causative gene. Human Genome Project (HGP) databases have provided additional resources to facilitate the identification of positional candidate genes. Clinical examinations were performed on individuals of the PFHBII-affected family, and, where available, clinical records of subjects examined in a previous study by Brink and Torrington (1977) were re-assessed. Retrospective data suggested redefining the classification of PFHBII. Subsequently, linkage analysis was used to test described dilated cardiomyopathy (DCM), hypertrophic cardiomyopathy (HCM) and cardiac conduction-causative loci on chromosomes 1, 2, 3, 6, 7, 9, 11, 14, 15 and 19 for their involvement in the development of PFHBII. Once a locus was mapped, bioinformatics tools were applied to identify and prioritise positional candidate genes for mutation screening. The retrospective and prospective clinical study redefined PFHBII as a cardiac conduction and DCM-associated disorder and simultaneously allowed more family members to be traced.Fortuitously, candidate loci linkage analysis mapped the PFHBII locus to chromosome 1q32, to a region that overlapped a previously described DCM-associated disorder (CMD1D), by the generation of a maximum pairwise lod score of 3.13 at D1S3753 (theta [θ]=0.0) and a maximum multipoint lod score of 3.7 between D1S3753 and D1S414. However, genetic fine mapping and haplotype analysis placed the PFHBII-causative locus distal to the CMD1D locus, within a 3.9 centimorgan (cM) interval on chromosome 1q32.2-q32.3, telomeric of D1S70 and centromeric of D1S505. Bioinformatics analyses prioritised seven candidate genes for mutation analysis, namely, a gene encoding a potassium channel (KCNH1), an extracellular matrix protein (LAMB3), a protein phosphatase (PPP2R5A), an adapter protein that interacts with a cytoskeletal protein (T3JAM), a putative acyltransferase (KIAA0205) and two genes encoding proteins possibly involved in energy homeostasis (RAMP and VWS59). The PFHBII-causative mutation was not identified, although single sequence variations were identified in four of the seven candidate genes that were screened. Although the molecular aetiology was not established, the present study defined the underlying involvement of DCM in the pathogenesis of PFHBII. The new clinical classification of PFHBII has been published (Fernandez et al., 2004) and should lead to tracing more affected individuals in South Africa or elsewhere. The identification of a novel disease-causative locus may point toward the future identification of a new DCM-associated aetiology, which, in turn, might provide insights towards understanding the associated molecular pathophysiologies of heart failure. / AFRIKAANSE OPSOMMING: Hartversaking as gevolg van kardiomiopatie of kardiale geleidingsiekte is ‘n hoof-oorsaak van mortaliteit and morbiditeit in beide ontwikkelde en ontwikkelende lande. Alhoewel gedefinieer as verskillende kliniese entiteite is oorerflike vorms van kardiomiopatie en kardiale geleidingsstoornisse geïdentifiseer met oorvleuelende kliniese eienskappe en/of molukulêre oorsake. Die doelwit van hierdie studie was om die molukulêre oorsaak van progressiewe familiële hartblok tipe II (PFHBII), ‘n oorerflike kardiale geleidingsstoornis, wat in ‘n Suid-Afrikaanse Kaukasiër familie segregeer (Brink en Torrington, 1977), te identifiseer. Die beskikbaarheid van familie data, beteken dat koppelingsanalise gebruik kan word om die chromosomale posisie van die siekte-veroorsakende geen te identifiseer. Menslike Genoom Projek (MGP) databanke het addisionele hulpbronne beskikbaar gestel om die identifikasie van posisionele kandidaat gene te vergemaklik. Kliniese ondersoeke is uitgevoer op PFHBII-geaffekteerde familielede, en waar beskikbaar is kliniese rekords van persone, wat in ‘n vorige studie deur Brink en Torrington (1977) geassesseer was, herontleed. Retrospektiewe data-analise het die kliniese herdefinisie van PFHBII voorgestel. Daarna is koppelingsanalise gebruik om dilateerde kardiomiopatie (DKM), hipertrofiese kardiomiopatie (HKM) en kardiale geleidingssiekte-veroorsakende loki op chromosoom 1, 2, 3, 6, 7, 9, 11, 14, 15 en 19 te ondersoek vir hul moontlike bydrae tot die ontwikkeling van PFHBII. Toe die lokus gekarteer was, is bioinformatiese ondersoeke gebruik om posisionele kandidaat gene te identifiseer en prioritiseer vir mutasie analise. Die retrospektiewe en prospektiewe kliniese ondersoek het PFHBII herdefinieer as ‘n geleidingsstoornis en DKM-verbonde siekte, en terselfde tyd het dit gelei tot die opsporingvan nog familielede. Toevallig het kandidaat loki-analise die PFHBII lokus op chromosoom 1q32 gekarteer, na ‘n gebied wat met ‘n voorheen-beskyfde DKM-verbonde stoornis (CMD1D) oorvleuel, met die opwekking van ‘n makisimum paargewyse lod-getal van 3.13 by D1S3753 (theta [θ] = 0.0) en ‘n maksimum multipunt lod-getal van 3.7 tussen D1S3753 en D1S414. Genetiese fynkartering en haplotipe-analise het die PFHBII-veroorsakende lokus afwaards van die CMD1D lokus geplaas, in ‘n 3.9 centimorgan (cM) gebied op chromosoom 1q32.2-q32.3, telomeries van D1S70 en sentromeries van D1S505. Bioinformatiese analise het daarnatoe gelei dat sewe kandidaat gene vir mutasie analise geprioritiseerd is, naamlik, gene wat onderskeidelik ‘n kalium kanaal (KCNH1), ‘n ekstrasellulêre matriksproteïen (LAMB3), ‘n proteïen fosfatase (PPP2R5A), ‘n aansluiter proteïen wat met ‘n sitoskilet proteïen bind (T3JAM), ‘n asieltansferase (KIAA0205) en twee gene moontlik betrokke in energie homeostase (RAMP en VWS59) enkodeer. Die PFHBII-veroorsakende geen is nie geïdentifiseer nie, alhoewel enkele volgorde-wisselings geïdentifiseer is in vier van die sewe geanaliseerde kandidaat gene. Alhowel die molekulêre oorsaak van die siekte nie vasgestel is nie, het die huidige studie die onderliggende betrokkenheid van DKM in die pathogenese van PFHBII gedefinieer. Die nuwe kliniese klassifikasie van PFHBII is gepubiliseer (Fernandez et al., 2004) en sal lei tot die identifisering van nog geaffekteerde persone in Suid Afrika of in ander lande. Die identifikasie van ‘n nuwe siekte-verbonde lokus mag lei tot die toekomstige identifikasie van ‘n nuwe DKM-verbonde genetiese oorsaak wat, opsig self, dalk insig kan gee in die molekulêre patofisiologie van hartversaking.