Studies on Causes of Multiple Sclerosis : From Genes to Transcriptome

Multiple sclerosis (MS) is the most common cause of neurological disability in young adults, affecting more than two million people worldwide. It manifests as a chronic inflammation in the central nervous system (CNS) and causes demyelination and neurodegeneration. Depending on the location of the demyelinated plaques and axonal loss, a variety of symptoms can be observed including deficits in vision, coordination, balance and movement. With a typical age of onset at 20-40 years, the social and economic impacts of MS on lives of the patients and their families are considerable. Unfortunately the current treatments are relatively inefficient and the development of more effective treatments has been impeded by our limited understanding of the causes and pathogenesis of MS. Risk of MS is higher in biological relatives of MS patients than in the general population. Twin and adoption studies have shown that familial clustering of MS is explained by shared genetic factors rather than by shared familial environment. While the involvement of the human leukocyte antigen (HLA) genes was first discovered four decades ago, additional genetic risk factors have only recently been identified through genome-wide association studies (GWAS). Current evidence suggests that MS is a highly polygenic disease with perhaps hundreds of common variants with relatively modest effects contributing to susceptibility. Despite extensive research, the majority of these risk factors still remain to be identified. In this thesis the aim was to identify novel genes and pathways involved in MS. Using genome-wide microarray technology, gene expression levels in peripheral blood mononuclear cells (PBMC) from 12 MS patients and 15 controls were profiled and more than 600 genes with altered expression in MS were identified. Three of five selected findings, DEFA1A3, LILRA4 and TNFRSF25, were successfully replicated in an independent sample. Increased expression of DEFA1A3 in MS is a particularly interesting observation, because its elevated levels have previously been reported also in several other autoimmune diseases. A systematic review of seven microarray studies was then performed leading to identification of 229 genes, in which either decreased or increased expression in MS had been reported in at least two studies. In general there was relatively little overlap across the experiments: 11 of the 229 genes had been reported in three studies and only HSPA1A in four studies. Nevertheless, these 229 genes were associated with several immunological pathways including interleukin pathways related to type 2 and type 17 helper T cells and regulatory T cells. However, whether these pathways are involved in causing MS or related to secondary processes activated after disease onset remains to be investigated. The 229 genes were also compared with loci identified in published MS GWASs. Single nucleotide polymorphisms (SNP) in 17 of the 229 loci had been reported to be associated with MS with P-value less than 0.0001 including variants in CXCR4 and SAPS2, which were the only loci where evidence for correlation between the associated variant and gene expression was found. The CXCR4 variant was further tested for association with MS in a large case-control sample and the previously reported suggestive association was replicated (P-value is 0.0004). Finally, common genetic variants in candidate genes, which had been selected on the basis of showing association with other autoimmune diseases (MYO9B) or showing differential expression in MS in our study (DEFA1A3, LILRA4 and TNFRSF25), were tested for association with MS, but no evidence of association was found. In conclusion, through a systematic review of genome-wide expression studies in MS we have identified several promising candidate genes and pathways for future studies. In addition, we have replicated a previously suggested association of a SNP variant upstream of CXCR4 with MS. Keywords: autoimmune disease, common variant, CXCR4, DEFA1A3, HSPA1A,gene expression, genetic association, GWAS, MS, multiple sclerosis, systematic review

HDL subspecies : association with low HDL-C, obesity, and metabolic syndrome

AIMS An independent, powerful coronary heart disease (CHD) predictor is a low level of high-density lipoprotein cholesterol (HDL-C). Discoidal preβ-HDL particles and large HDL2 particles are the primary cholesterol acceptors in reverse cholesterol transport, a key anti-atherogenic HDL mechanism. The quality of HDL subspecies may provide better markers of HDL functionality than does HDL-C alone. We aimed I) to study whether alterations in the HDL subspecies profile exist in low-HDL-C subjects II) to explore the relationship of any changes in HDL subspecies profile in relation to atherosclerosis and metabolic syndrome; III) to elucidate the impact of genetics and acquired obesity on HDL subspecies distribution. SUBJECTS The study consisted of 3 cohorts: A) Finnish families with low HDL-C and premature CHD (Study I: 67 subjects with familial low HDL-C and 64 controls; Study II: 83 subjects with familial low HDL-C, 65 family members with normal HDL-C, and 133 controls); B) a cohort of 113 low- and 133 high-HDL-C subjects from the Health 2000 Health Examination Survey carried out in Finland (Study III); and C) a Finnish cohort of healthy young adult twins (52 monozygotic and 89 dizygotic pairs) (Study IV). RESULTS AND CONCLUSIONS The subjects with familial low HDL-C had a lower preβ-HDL concentration than did controls, and the low-HDL-C subjects displayed a dramatic reduction (50-70%) in the proportion of large HDL2b particles. The subjects with familial low HDL-C had increased carotid atherosclerosis measured as intima-media-thickness (IMT), and HDL2b particles correlated negatively with IMT. The reduction in both key cholesterol acceptors, preβ-HDL and HDL2 particles, supports the concept of impaired reverse cholesterol transport contributing to the higher CHD risk in low-HDL-C subjects. The family members with normal HDL-C and the young adult twins with acquired obesity showed a reduction in large HDL2 particles and an increase in small HDL3 particles, which may be the first changes leading to the lowering of HDL-C. The low-HDL-C subjects had a higher serum apolipoprotein E (apoE) concentration, which correlated positively with the metabolic syndrome components (waist circumference, TG, and glucose), highlighting the need for a better understanding of apoE metabolism in human atherosclerosis. In the twin study, the increase in small HDL3b particles was associated with obesity independent of genetic effects. The heritability estimate, of 73% for HDL-C and 46 to 63% for HDL subspecies, however, demonstrated a strong genetic influence. These results suggest that the relationship between obesity and lipoproteins depends on different elements in each subject. Finally, instead of merely elevating HDL-C, large HDL2 particles and discoidal preβ-HDL particles may provide beneficial targets for HDL-targeted therapy.

LIN28B, LIN28A, KISS1, and KISS1R in idiopathic central precocious puberty

Abstract Background Pubertal timing is a strongly heritable trait, but no single puberty gene has been identified. Thus, the genetic background of idiopathic central precocious puberty (ICPP) is poorly understood. Overall, the genetic modulation of pubertal onset most likely arises from the additive effect of multiple genes, but also monogenic causes of ICPP probably exist, as cases of familial ICPP have been reported. Mutations in KISS1 and KISSR, coding for kisspeptin and its receptor, involved in GnRH secretion and puberty onset, have been suggested causative for monogenic ICPP. Variation in LIN28B was associated with timing of puberty in genome-wide association (GWA) studies. LIN28B is a human ortholog of the gene that controls, through microRNAs, developmental timing in C. elegans. In addition, Lin28a transgenic mice manifest the puberty phenotypes identified in the human GWAS. Thus, both LIN28B and LIN28A may have a role in pubertal development and are good candidate genes for monogenic ICPP. Methods Thirty girls with ICPP were included in the study. ICPP was defined by pubertal onset before 8 yrs of age, and a pubertal LH response to GnRH testing. The coding regions of LIN28B, LIN28A, KISS1, and KISS1R were sequenced. The missense change in LIN28B was also screened in 132 control subjects. Results No rare variants were detected in KISS1 or KISS1R in the 30 subjects with ICPP. In LIN28B, one missense change, His199Arg, was found in one subject with ICPP. However, this variant was also detected in one of the 132 controls. No variation in LIN28A was found. Conclusions We did not find any evidence that mutations in LIN28B or LIN28A would underlie ICPP. In addition, we confirmed that mutations in KISS1 and KISS1R are not a common cause for ICPP.

A linear sequence of extant families and genera of lycophytes and ferns

Throughout the history of the classification of extant ferns (monilophytes) and lycophytes, familial and generic concepts have been in great flux. For the organisation of lycophytes and ferns in herbaria, books, checklists, indices and spore banks and on the internet, this poses a problem, and a standardized linear sequence of these plants is therefore in great need. We provide here a linear classification to the extant lycophytes and ferns based on current phylogenetic knowledge; this provides a standardized guide for organisation of fern collections into a more natural sequence. Two new families, Diplaziopsidaceae and Rhachidosoraceae, are here introduced.

Molecular diagnosis of developmental disorders

ABSTRACT Idiopathic developmental disorders (DDs) affect ~1% of the population worldwide. This being a considerable amount, efforts are being made to elucidate the disease mechanisms. One or several genetic factors cause 30-40% of DDs, and only 10% are caused by environmental factors. The remaining 50% of DD patients go undiagnosed, mostly due to a lack of diagnostic techniques. The cause in most undiagnosed cases is though to be a genetic factor or a combination of genetic and environmental factors. Despite the surge of new technologies entering the market, their implementation into diagnostic laboratories is hampered by costs, lack of information about the expected diagnostic yield, and the wide range of selection. This study evaluates new microarray methods in diagnosing idiopathic DDs, providing information about their added diagnostic value. Study I analysed 150 patients by array comparative genomic hybridization (array CGH, 44K and 244K), with a subsequent 18% diagnostic yield. These results are supported by other studies, indicating an enourmous added diagnostic value of array CGH, compared with conventional cytogenetic analysis. Nevertheless, 80% of the patients remained undiagnosed in Study I. In an effort to diagnose more patients, in Study IV the resolution was increased from 8.9 Kb of the 244K CGH array to 0.7 Kb, by using a single-nucleotide polymorphism (SNP) array. However, no additional pathogenic changes were detected in the 35 patients assessed, and thus, for diagnostic purposes, an array platform with ca 9 Kb resolution appears adequate. The recent vast increase in reports of detected aberrations and associated phenotypes has enabled characterization of several new syndromes first based on a common aberration and thereafter by delineation of common clinical characteristics. In Study II, a familial deletion at 9q22.2q22.32 with variable penetrance was described. Despite several reports of aberrations in the adjacent area at 9q associated with Gorlin syndrome, the patients in this family had a unique phenotype and did not present with the syndrome. In Study III, a familial duplication of chromosome 6p22.2 was described. The duplication caused increased expression of an important enzyme of the γ-aminobutyric acid (GABA) degradation pathway, causing oxidative stress of the brain, and thus, very likely, the mild mental retardation of these patients. These two case studies attempted to pinpoint candidate genes and to resolve the pathogenic mechanism causing the clinical characteristics of the patients. Presenting rare genetic and clinical findings to the international science and medical community enables interpretation of similar findings in other patients. The added value of molecular karyotyping in patients with idiopathic DD is evident. As a first line of testing, arrays with a median resolution of at least 9 Kb should be considered and further characterization of detected aberrations undertaken when possible. Diagnostic whole-exome sequencing may be the best option for patients who remain undiagnosed after high-resolution array analysis.