Variaciones genéticas en los trastornos del metabolismo lipídico

Résumé

Cardiovascular diseases are the main cause of death, disability and consumption of health resources both in Spain and in most of the developed world. Cardiovascular diseases are, mostly of vascular origin and atherosclerosis is the main cause of their development. The etiology of the atheroma plaque is multifactorial. Risk factors include age, body mass index, arterial hypertension, smoking, diabetes and dyslipidaemia. Primary hyperlipemias are a heterogeneous group of lipid metabolism disorders characterized by a high cardiovascular risk. The most important clinical forms are accompanied by increases in the plasma concentration of cholesterol, triglycerides or both. Although the concentration of cholesterol and triglycerides depends on multiple genetic and environmental mechanisms, subjects who have extremes of hypercholesterolemia or hypertriglyceridemia usually have a higher genetic risk. In spite of the important advances in the field of genetics that have allowed the identification of numerous loci responsible for different dyslipidaemias, there is still a high percentage of dyslipidaemias of unknown genetic origin. This thesis is made by a compendium of 5 different publications, which main objective is to discover novel genes and understand their role on cholesterol and triglyceride metabolism in the pathogenesis of hereditary dyslipidaemia. The first study aimed to identify genetic variants in LMF1 gene in subjects with triglyceride (TG) values above 500 mg/dL and without secondary causes recruited at the Lipid Unit at the Miguel Servet Hospital. The promoter, exons and exon-intron regions of the LMF1 gene were sequenced in 112 patients with severe primary hypertriglyceridemia (HTG). Five patients (4.46%) were carriers of 4 rare variants in LMF1 previously associated with HTG. In addition, 2 common variants were identified with an allelic frequency different from that observed in the general population: c.194-28 T> G and c.729 + 18C> G. A bioinformatic analysis of the variants found was carried out, identifying the variants p. (Arg364Gln), p. (Arg451Trp), p. (Pro562Arg) and p. (Leu85Leu) as potentially harmful. Therefore, we conclude that our results suggest that the LMF1 gene contributes to the etiology of severe primary HTG in a significant percentage of patients, with a combination of mutations with moderate to aggressive effects and polymorphisms classically associated with this dyslipidaemia. In the second study, our objective was to identify the prevalence of HTG andd the frequency of rare mutations causing HTG in a whole non-selected population. A total of 23,310 subjects, aged over 18 years, from a primary care-district in a middle-class area of Zaragoza (Spain) with TG >500 mg/dL were selected to establish HTG prevalence. Those diagnosed as primary HTG were considered for further genetic analysis. The promoters, coding regions and exon-intron boundaries of LPL, LMF1, APOC2, APOA5, APOE and GPIHBP1 genes were sequenced. The frequency of rare variants identified was studied in 90 controls. We defined rare variants those their frequencies were less than 1 % in general population obtained from the 1000 Genomes project. Our results show that one hundred ninety-four subjects (1.04 %) had HTG and 90 subjects (46.4 %) met the inclusion criteria for primary HTG. In this subgroup, nine patients (12.3 %) were carriers of 7 rare variants in LPL, LMF1, APOA5, GPIHBP1 or APOE genes. Three of these mutations are described for the first time in this work. The presence of a rare pathogenic mutation did not confer a differential phenotype or a higher family history of HTG. We concluded that the prevalence of rare mutations in candidate genes in subjects with primary HTG is low. The low frequency of rare mutations, the absence of a more severe phenotype or a dominant transmission of HTG indicates the complexity and heterogeneity of this disease. In the third study, our objective was to determine the contribution to low-density lipoprotein cholesterol (LDL-C) of the single nucleotide variants associated with polygenic hypercholesterolemia in probands with genetic hypercholesterolemia without mutations in candidate genes (FH-) and the genetic score in cascade screening in their family members. We recruited 49 nonfamilial hypercholesterolemia genetic hypercholesterolemia families (294 participants) and calculated cholesterol gene scores, derived from single nucleotide variants in SORT1, APOB, ABCG8, APOE and LDLR and lipoprotein(a) plasma concentration. Our results showed that risk alleles in SORT1, ABCG8, APOE, and LDLR showed a statistically significant higher frequency in blood relatives than in the 1000 Genomes Project. However, there were no differences between affected and nonaffected members in each of the single nucleotide variants. Although, the contribution of the cholesterol gene score to LDL-C was significantly higher in affected than in nonaffected participants (p = 0.048). The percentage of the LDL-C variation explained by the score was 3.1%, and this percentage increased to 6.9% in those families with the highest genetic score in the proband. In this work we concluded that nonfamilial hypercholesterolemia genetic hypercholesterolemia families concentrate risk alleles for high LDL-C. Their contribution varies greatly among families, indicating the complexity and heterogeneity of these forms of hypercholesterolemias. The gene score explains a small percentage of LDL-C, which limits its use in diagnosis. In the fourth study, we analysed the prevalence the of ABCG5/G8 genetic variants in FH- patients, as defects in these genes relate to intestinal hyperabsorption of cholesterol and thus ABCG5/G8 variants could explain in part the mechanism of hypercholesterolemia. We sequenced the ABCG5/G8 genes in 214 FH- and 97 controls and quantified surrogate markers of cholesterol absorption (5α-cholestanol, β-sitosterol, campesterol, stigmasterol, and sitostanol) by high-performance liquid chromatography–tandem mass spectrometry in both studied groups. We found 8 FH- patients (3.73%) with a pathogenic mutation in ABCG5/G8 genes. We observed significantly higher concentration of surrogate markers of cholesterol absorption in FH- than in controls. In addition, we found significantly higher concentrations of cholesterol absorption markers in FH- with ABCG5/G8 defects than FH- patients without mutation in ABCG5/G8 genes. A gene score created from 27 common single nucleotide variants in ABCG5/G8 associated with hypercholesterolemia was significantly higher in cases than in controls (p= 0.032). Subjects with a gene score above the mean had significantly higher 5α-cholestanol and stigmasterol than those with a lower gene score. We concluded that FH- subjects accumulate an excess of rare and common gene variations in ABCG5/G8 genes. This variation is associated with increased intestinal absorption of cholesterol, as determined by surrogate makers, suggesting that these loci contribute to hypercholesterolemia by enhancing intestinal cholesterol absorption. Finally, we concluded this thesis with one study still pending publication in the Scientific reports journal, in which we aim to investigate the role of genetic factors in weight loss in a large cohort of subjects with overweight under a homogeneous dietary intervention. We retrospectively recruited all subjects (N= 788) aged over 18 with a BMI between 25-40 kg/m2 and a control group of subjects (168 patients) with normal BMI (18.5-25 kg/m2) from a Lipid Unit with at least one-year of follow-up from 2008 to 2016. All cases received counselling by a nutritionist which included healthy diet and increase of physical activity. We genotyped 25 single nucleotide variants (SNV) in 25 genes previously associated with obesity and calculated weighted genetic score derived from 5 SNV (FLJ35779, KCTD15, NRXN3, RPL27A and SH2B1 genes). Our results showed the risk allele in CADM2 showed a higher frequency in overweight and obese subjects than in controls (p=0.007). The mean follow-up was 5.58 ± 2.68 years. Subjects with lower genetic score showed higher weight loss along follow-up. The genetic score was the variable that explained best weight change variance, after baseline weight, explaining 1.7% weight change variance at one year and 1.5% weight change variance at the end of follow-up (p<0.001 and p=0.001, respectively) adjusting by baseline weight, sex, age and years of follow-up.