MOLECULAR ANALYSIS OF APOB, SAR1B, ANGPTL3, AND MTTP IN PATIENTS WITH PRIMARY HYPOBETALIPOPROTEINEMIA IN A CLINICAL LABORATORY SETTING: EVIDENCE SUPPORTING POLYGENICITY IN MUTATION-NEGATIVE PATIENTS 1,2,3 ¥ 2,4 ¥ Francisco Blanco-Vaca *, Jesús M. Martin-Campos , Ángel Beteta-Vicente 1, Marina Canyelles 1,3, Susana Martínez 1, Rosa Roig 1, Núria Farré 4, Josep Julve 2,4, Mireia Tondo 1* 1 Hospital de la Santa Creu i Sant Pau, Servei de Bioquímica – IIB Sant Pau, Barcelona 2 CIBER de Diabetes y Enfermedades Metabólicas Asociadas (CIBERDEM), Barcelona 3 Departament de Bioquímica i Biologia Molecular, Universitat Autònoma de Barcelona 4 Institut de Recerca de l’Hospital de la Santa Creu i Sant Pau- IIB Sant Pau, Barcelona ¥ These authors contributed equally to this study. * Address correspondence to: Dra. Mireia Tondo or Dr. Francisco Blanco-Vaca, Servei de Bioquímica, Hospital de la Santa Creu i Sant Pau, C/ Sant Quintí 89, 08041 Barcelona, Spain. Tel.: +34-93-5537358, fax: +34-93-5537287, e-mail: mtondo@santpau.cat; fblancova@santpau.cat. 1 Abstract Background and aims: Primary hypobetalipoproteinemia is generally considered a heterogenic group of monogenic, inherited lipoprotein disorders characterized by low concentrations of LDL cholesterol and apolipoprotein B in plasma. Lipoprotein disorders include abetalipoproteinemia, familial hypobetalipoproteinemia, chylomicron retention disease, and familial combined hypolipidemia. Our aim was to review and analyze the results of the molecular analysis of hypolipidemic patients studied in our laboratory over the last 15 years. Methods: The study included 44 patients with clinical and biochemical data. Genomic studies were performed and genetic variants were characterized by bioinformatics analysis. A weighted LDL cholesterol gene score was constructed in order to evaluate common variants associated with impaired lipid concentrations and their distribution among patients. Results: Twenty-three patients were genetically confirmed as affected by primary hypobetalipoproteinemia. In this group of patients, the most prevalent mutated genes were APOB (in 17 patients, with eight novel mutations identified), SAR1B (in 3 patients, with one novel mutation identified), ANGPTL3 (in 2 patients), and MTTP (in 1 patient). The other 21 patients could not be genetically diagnosed of hypobetalipoproteinemia despite presenting suggestive clinical and biochemical features. In these patients, two APOB genetic variants associated with lower LDLc cholesterol were more frequent than in controls. Moreover, the LDL cholesterol gene score, calculated with 11 SNPs, was significantly lower in mutation- negative patients. Conclusion: Around half of the patients could be genetically diagnosed. The results suggest that in at least some of the patients without an identified mutation, primary hypobetalipoproteinemia may in fact have a polygenic origin. 2 Keywords: Combined hypolipidemia; Familial hypobetalipoproteinemia; Chylomicron retention disease; Abetalipoproteinemia; Polygenic hypocholesterolemia. Highlights: Primary hypobetalipoproteinemia has been considered monogenic; however, our findings suggest that in some cases it may also have a polygenic origin. 3 Abbreviations HBL: hypobetalipoproteinemia LDLc: LDL cholesterol ApoB: apolipoprotein B MTTP: microsomal triglyceride transfer protein PCSK9: proprotein convertase subtilisin/kexin type 9 SAR1B : secretion associated Ras related GTPase ANGPTL3 : angiopoietin-like 3 MYLIP: myosin regulatory light chain interacting protein DGAT1: diacylglycerol O-acyltransferase NPC1L1: Niemann-Pick C1-like 1 cholesterol transporter 1 FHBL: familial hypobetalipoproteinemia TG: triglyceride CMRD: chylomicron retention disease CM: chylomicron VLDL: very low density lipoprotein ABL: abetalipoproteinemia FHBL1: familial hypobetalipoproteinemia type 1 FHBL2: familial hypobetalipoproteinemia type 2 HDLc: HDL cholesterol TC: total cholesterol SNP: single nucleotide polymorphism HBL/M-: hypobetalipoproteinemia mutation-negative patients HBL/M+: hypobetalipoproteinemia mutation-positive patients 4 INTRODUCTION Primary hypobetalipoproteinemia (HBL) is a heterogenic group of inherited lipoprotein disorders characterized by low concentrations of low-density lipoprotein (LDL) cholesterol (LDLc) and apolipoprotein B (ApoB) in plasma1. HBL may manifest as a fat malabsorption syndrome commonly accompanied by growth failure, fat-soluble vitamin deficiency, hepatic steatosis, abdominal distension, diarrhea, vomiting, and neurological and ophtalmological disorders in the context of hypolipidemia1. Primary HBL can be classified according to various criteria, including clinical and biochemical phenotype, the gene involved, and the mode of inheritance of the condition. Familial hypobetalipoproteinemia (FHBL) type 1 (FHBL1) (OMIM 615558) is a codominant disorder caused by mutations in the APOB gene frequently giving rise to a truncated ApoB protein2. Affected heterozygotes are asymptomatic or present with hepatic steatosis, and low levels of LDLc and ApoB3. Homozygotes may be clinically and biochemically indistinguishable from abetalipoproteinemic patients2, 3. FHBL1 can also be caused by loss- of-function mutations in the proprotein convertase subtilisin/kexin type 9 (PCSK9) gene which encodes a protein that targets the LDL receptor for lysosomal degradation. APOB and PCSK9 genes are also associated to familial hypercholesterolemia. Familial combined hypolipidemia (OMIM 605019), also known as familial hypobetalipoproteinemia type 2 (FHBL2), is a rare recessive disorder caused by mutations in the angiopoietin-like 3 (ANGPTL3) gene4. Loss-of-function mutation increases lipoprotein lipase activity and may decrease high-density lipoprotein (HDL) cholesterol (HDLc) by increasing endothelial lipase, two enzymes that in physiological conditions are catabolized by the action of ANGPTL3. Patients typically present with no clear phenotype other than hypolipidemia2. Chylomicron retention disease (CMRD) or Anderson’s disease (OMIM 246700) is a rare recessive disorder caused by mutations in the secretion associated Ras Related GTPase 5 (SAR1B) gene characterized by the accumulation of lipid droplets within the enterocytes and the selective absence of ApoB-48-containing particles from plasma5. Biochemical manifestations include reduced LDLc levels and absence of chylomicrons (CM)5. Abetalipoproteinemia (ABL) (OMIM 200100) is a rare recessive disorder caused by mutations in the microsomal triglyceride transfer protein (MTTP) gene. The gene encodes for a molecular chaperone critical for the formation of triglyceride-rich lipoproteins such as very low-density lipoproteins (VLDL) and CM. Patients typically present with an absence of LDLc and CM, low TG levels, and very low vitamin E levels3. Most of these disorders are very rare.. Estimated prevalence for familial hypobetalipoproteinemia is 1 in 1000-3000 individuals and less than 1 in 1 million individuals for the rest of types of primary HBL3, 6. Studies have reported variants in other genes associated very infrequently with primary HBL, such as the APOC3 gene7, the miosin regulatory light chain interacting protein (MYLIP) gene8, and the diacylglycerol O- acyltransferase 1 (DGAT1) gene9. It is noteworthy that extreme cholesterol phenotypes can result from the accumulation of common small-effect LDLc affecting alleles11, 12. Therefore, the objective of this study was to review, analyze, and evaluate genetic variants in 44 hypolipidemic patients recruited to our laboratory over the last 15 years. This included the search for a potentially polygenic basis for the disorder in patients who were negative for mutations in the well-established genes causing primary HBL. 6 MATERIALS AND METHODS Subjects Subjects were recruited over a 15-year period and were referred to our Metabolism Unit for genetic analysis based on biochemical and/or clinical suspicion of primary HBL, according to the requesting practitioners. Only two patients were followed at our hospital; thus, clinical and/or biochemical data were provided by the medical centers of origin (mainly from Spain) in summarized form, and as a result, the clinical and/or biochemical information obtained was sometimes heterogeneous or incomplete. Primary HBL is usually defined as persistent low total cholesterol with plasma LDLc and ApoB concentrations below the fifth percentile in the Spanish population, 130 mg/dL, 62 mg/dL and 63 mg/dL respectively13, 14. In order to avoid molecular analysis of secondary HBL, the exclusion criteria included severe chronic liver disease, chronic pancreatitis, cystic fibrosis, end-stage renal disease, hyperthyroidism, cachexia, and malabsorption. In some cases, relatives were also submitted for study. Genetic studies were performed to confirm the diagnosis. All blood samples were obtained in accordance with the Helsinki Declaration of 1964, as revised in 2000. The Ethics Committee of the Hospital de la Santa Creu i Sant Pau reviewed and approved the study protocol and all individuals provided written informed consent. To study the polygenic basis, a sample of 503 European subjects from the 1000 Genomes Project (www.1000genomes.org) was used as the control group and gene score distribution compared. Gene amplification and resequencing Genomic DNA was isolated from whole blood (usually EDTA or citrate blood) using the QIAamp DNA blood minikit (Qiagen, Hilden, Germany). All the exons and exon-flanking intronic regions of the MTTP, APOB, SAR1B, PCSK9, ANGPTL3, APOC3, MYLIP, and DGAT1 genes were amplified by PCR. The resulting products were purified using GFX PCR DNA and a Gel Band Purification Kit (GE Healthcare, Buckinghamshire, UK) and sequenced using a Big Dye Terminator cycle sequencing kit v.3.1 (Applied Biosystems, Foster, CA, 7 USA) on an ABI3130XL automated analyzer (Applied Biosystems). The resulting chromatograms were analyzed with the Staden package program15. The primers used for gene amplification and Sanger sequencing are available upon request. A sequential analysis was performed to optimize the diagnostic process. The APOB gene was sequenced first, and if it was negative, the SAR1B, ANGPTL3, and MTTP genes were sequenced. Finally, if no putative pathogenic variants were detected, the rest of the genes were also sequenced. Characterization of variants and bioinformatics analysis The nomenclature of the allelic variants follows the recommendations of the Human Molecular Genome Variation Society (http://www.hgvs.org). All variants were checked against the Human Gene Mutation Database (HGMD, www.hgmd.cf.ac.uk), 1000 Genome Project (phase3browser.1000genomes.org), and ClinVar (www.ncbi.nlm.nih.gov/clinvar) databases in order to characterize them. No functional studies were performed hence bioinformatics functional analysis was used. The impact of point mutations on the structure and function of the protein was assessed with the following software: SIFT (sift.bii.a- star.edu.sg)16, PolyPhen2 (genetics.bwh.harvard.edu/pph2/index.shtml)17, Panther (www.pantherdb.org)18, Provean (provean.jcvi.org)19, i-Mutant (gpcr.biocomp.unibo.it/cgi/predictors/I-Mutant3.0)20, SNPs3D (www.snps3d.org)21, PMut (mmb.pcb.ub.es/pmut2017)22, Mutation Taster (www.mutationtaster.org)23, and Mutation Assessor (mutationassessor.org)24. Variants affecting introns were analyzed using Human Splicing Finder v.3.0 software (www.umd.be/HSF3/HSF.html)25. According to the American College of Medical Genetics and Genomics guidelines26, point mutations causing premature stop codons, small insertions, or deletions causing a frameshift and a premature stop codon, large rearrangements, and mutations affecting intron donor or acceptor splice sites (positions +1, +2, -2, or -1) were considered directly pathogenic. The remaining variants (synonymous mutations, missense mutations, small in-frame insertions and deletions, mutations affecting a promoter, 5’UTR or 3’UTR, and intronic variants) were considered pathogenic depending on the existence of functional analysis previously reported in the literature, identification as 8 pathogenic or likely pathogenic in databases such as ClinVar, or in the absence of previous information, when the programs used in the bioinformatics analysis gave as a result a probable alteration in protein function, protein expression, or gene regulation. Frequency distribution of variants and LDLc gene score We used contingency tables to evaluate the frequency distribution of the observed nonpathogenic variants in patients that tested negative for a pathogenic mutation versus 503 European controls from the 1000 Genomes Project. We also calculated a weighted LDLc gene score11 (the weighted sum of the estimated per-allele effect in LDLc changes, where weights are the effect sizes calculated from genome-wide association studies, Supplementary Table 1) and compared its distribution among patients with HBL with no confirmed pathogenic variant, patients with an identified pathogenic mutation, and 503 European controls from the 1000 Genomes Project. Using TaqMan probes (Thermo Fisher Scientific, USA), 11 SNPs influencing LDLc were genotyped: PCSK9 (rs2479409), CELSR2- SORT1 (rs629301), APOB (rs1367117), ABCG8 (rs4299376), SLC22A1 (rs1564348), HFE (rs1800562), MYLIP (rs3757354), NYNRIN (rs8017377), LDLR (rs6511720), and APOE (rs429358 and rs7412)11. This LDLc-specific gene score was calculated using the per-allele beta coefficients reported by the Global Lipid Genetics Consortium27. Statistical analysis A nonparametric sample t-test and Fisher’s exact test were performed with SPSS v13.0 (IBM). 9 RESULTS The study included 44 patients, 27 males and 17 females aged 0.5 to 76 years (mean age 15.4 ± 14.0 years). All the patients were Spanish, except for two Italians and one German. All available demographic, clinical, and biochemical information is presented in Supplementary Table 2. Among the 44 patients studied, 23 (52.3%) were considered genetically diagnosed, 17 presented with mutations in the APOB gene (73.9%), three in the SAR1B gene (13.0%), two in the ANGPTL3 gene (8.7%), and one in the MTTP gene (4.4%). The results for all genetically diagnosed patients are presented in Table 1. Twenty-one patients (47.7%) did not present with a pathogenic mutation; however, most of them presented with genetic variants potentially associated with changes in lipid metabolism. APOB mutations: Of the 17 patients with APOB mutation, total cholesterol (TC) was provided in 12, with values ranging from 25 to 111 mg/dL, average ± SD 72.8 ± 30.9. LDLc was provided in 10 patients, and the values ranged from 1 to 46 mg/dl, average ± SD 19.9 ± 13.7. ApoB values were referred in nine patients and ranged from 4 to 35 mg/dl, average 24.1 ± 9.3. HDLc was provided in eight patients and ranged from 21 to 89 mg/dl, average ± SD 44.8 ± 18.5. TG was provided in 10 patients, nine of whom presented with decreased values and one of whom presented with values that fell within the reference interval. TG- concentration values ranged from 7 to 96 mg/dL, average ± SD 29.2 ± 25.5. Notably, patient 4 was homozygote for the novel APOB mutation c.2726C>A (p.Ser909*). All the in silico prediction programs classified this mutation as damaging. The mutation generates an estimated 80% size reduction of the ApoB protein. The parents of the proband were studied and in both cases presented with the mutation c.2726C>A (p.Ser909*) in heterozygosity. No clinical data were provided The mother presented with decreased TC (93 mg/dL), LDLc (25 mg/dL), apoB (<22 mg/dL), TG (23.7 mg/dL), and vitamin A (0.23 mg/L) and a normal HDLc (63.3 mg/dL). The father also presented with decreased TC (87 mg/dL), 10 LDLc (34 mg/dL), ApoB (30.9 mg/dL), and HDLc (25.9 mg/dL) but with a normal TG concentration (133.3 mg/dL). Patient 8 was a compound heterozygote for the variants c.3G>C (p.Met1Ile) and c.6627_6628delAA (p.Leu2212*). We described both mutations for the first time, and their pathogenicity was predicted by all the in silico prediction programs. We estimated that the second mutation generates a 50% size reduction of the ApoB protein. Data and DNA only from the father were sent, testing positive for the c.3G>C mutation (p.Met1Ile) and negative for c.6627_6628delAA (p.Leu2212*) and presenting with decreased LDLc (39 mg/dl) and ApoB (21 mg/dL). One case reported in the literature exhibited an interaction between a heterozygous ApoB mutation and an ApoE2/E2 genotype28. In this case, both father and proband presented with an ApoE3/E3 genotype. Other novel mutations described in the APOB gene in this study were c.133C>T (p.Arg45*); c.1407C>A (p.Tyr469*); c.11728G>T (p.Glu3910*); c.2781delT (p.Ser928Profs*25); and c.10430_10440delTTGACCACAAG (p.Val3477Alafs*2), all of which cause a truncated protein. All the consulted in silico programs predicted their pathogenicity. Some previously reported mutations were also found in our cohort and have been included in Table 1 together with a designation of the ApoB truncated forms. Of the submitted relatives, five of them presented with the familial mutation in the APOB gene and therefore were genetically diagnosed with FHBL1. In addition, some variants previously reported to be associated with more subtle changes in lipid or lipoprotein plasma levels were also identified, see Table 2. These reported variants were found in both genetically diagnosed and non-genetically-confirmed patients; of these variants, only three were found exclusively in negative patients: c.2188G>A (p.Val730Ile), c.3383G>A (p.Arg1128His), and c.3427C>T (p.Pro1143Ser). SAR1B mutations: Three patients were genetically diagnosed for SAR1B pathogenic variants, all in homozygous form. We found one novel variant, c.83_84delTG 11 (p.Leu28Argfs*7), and two previously reported. One of the patients, a 26-year-old male, was referred with only biochemical data and was found to be homozygous for p.Asp137Asn. Two relatives were also diagnosed, one presented the variant c.409G>A (p.Asp137Asn), and the other one presented with c.1-4482_58+1406del5946insGCATGATGGCGGGT, both of them in homozygosity. ANGPTL3 mutations: As previously reported by our group, two patients were genetically confirmed as homozygous MTTP mutations: Only one patient was found to be homozygous for a previously described pathogenic variant Moreover, variants associated with subtle changes in lipid profile were found, see Table 2. Results of PCSK9, MYLIP, APOC3 and DGAT1 sequencing: Some variants potentially associated with plasma lipid level variability were found for PCSK9, MYLIP and APOC3 genes, see Table 2. No variants previously associated with primary HBL were detected for the DGAT1 gene. Contingency tables to evaluate the frequency distribution of three common genetic variants of the APOB gene: Several common single nucleotide polymorphisms (SNP) associated with impaired lipid concentrations27 were identified in those patients that tested negative for primary HBL. After performing a Fisher’s exact test, three SNP presented with a statistically significantly different frequency of distribution when compared to 503 European subjects from the 1000 Genomes Project (Table 3). The APOB variants with statistical significance were the following: rs1367117, allele T, associated with high LDLc (p=0.00052); rs1042031, allele A, associated with low LDLc (p=0.021); and rs1042034, allele A, associated with high LDLc (p=0.0014). Concerning the 21 patients who could not be genetically diagnosed with primary HBL, TC was provided in 12 of the patients, with values ranging from 57 to 114 mg/dL, average ± SD of 85.7 ± 19.9. LDLc was provided in 12 patients, all of whom had decreased values except 12 for one patient. Values ranged from 13 to 52 mg/dl, average ± SD 29.9 ± 12.9. ApoB values were referred in 11 patients and ranged from 16 to 50 mg/dl, average ± SD 30.3 ± 10.0. HDLc was provided in 6 patients, two of whom presented with decreased values and four of whom presented with values that fell within the reference interval. The values ranged from 17 to 64 mg/dl, average ± SD 42.5 ± 16.0. TG was provided in 9 patients, six of whom presented with decreased values and three of whom were normal. The values ranged from 12 to 75 mg/dL, average ± SD 38.8 ± 21.1. Analysis of the weighted LDLc gene score: Regarding a possible polygenic basis of the hypolipidemia of these patients, LDLc weighted scores expressed as mean (SD) were as follows: 0.8868 (0.212) for the control population, 503 subjects from the European population in the 1000 Genome Project; 0.8033 (0.217) for 23 primary HBL patients with identified mutation (HBL/M+); 0.7366 (0.327) for 20 primary HBL mutation-negative patients (HBL/M-). The gene score using the 11 SNPs was significantly different between HBL/M- and controls (p=0.043), whereas no significant differences were found between HBL/M+ and controls (p=0.07) or between HBL/M- and HBL/M+ (p=0.618) (Figure 1). Six of the HBL/M- subjects (29%) presented a weighted LDLc gene score below the 10th percentile (0.6088) of the control population. The results for the LDLc scores of all patients are presented in Tables 1 and 2. 13 DISCUSSION The present study presents the outcomes of our molecular diagnosis of hypolipidemic patients in a clinical laboratory setting. Although the study actually resulted from an unprogrammed clinical practice setting, it constitutes the largest experience published in relation to the genetic diagnosis of primary HBL in Spain. The most recent genetic studies in Spanish patients presenting with low LDLc described the clinical and genetic heterogeneity of a total of five patients with FHBL133, 34 . The benefits of knowing the patient genotype include identification of additional family members as well as appropriate clinical follow-up and treatment, which includes adherence to a low-fat diet and supplementation with essential fatty acids and fat soluble vitamins. Out of 44 referred patients, 23 were genetically confirmed for the studied genes as affected by monogenic causes of primary HBL phenotype. Among them, the most prevalent mutated genes found in our study were APOB (17 of 23 patients), SAR1B, ANGPTL3, and MTTP, with substantially fewer mutations found in these other genes. Indeed, this should be the recommended sequence of actions with which to screen for primary HBL in our media in case that Sanger sequencing method is used. Despite their important role in lipid metabolism, Sanger sequencing did not allow identifying pathogenic variants in PCSK9, APOC3, MYLIP, or DGAT1. The inclusion of relatives in the study revealed seven further cases that were genetically confirmed. Interestingly, two of the relatives with APOB mutations did not present any sign of disease, demonstrating that screening for mutations in relatives should be strongly recommended because some cases might be underdiagnosed. Thus, early diagnosis is important for preventing or minimizing clinical sequelae. There was a clear overlap between clinical and biochemical data and the affected genes involved in primary HBL. The latter further revealed that it may not be a feasible strategy to predict whether or not a patient will present with a monogenic detectable mutation by means of clinical and biochemical phenotype. However, steatosis or liposoluble vitamin deficiency would rule out mutations in ANGPTL3 as a monogenic cause of HBL2. Additionally, the 14 pattern of Mendelian inheritance in a family may direct the analysis towards APOB (co- dominant) or ANGPTL3, SAR1B and MTTP (recessive). Because most of our HBL/M+ patients presented with APOB mutations, it is worth recalling that the clinical symptoms could include hepatic steatosis, liposoluble vitamin deficiency, diarrhea, growth failure, cognitive impairment, and multineuritis. Previously published data described APOB heterozygous mutation carriers individuals as asymptomatic, or have increased susceptibility to diseases that may go underdiagnosed for producing few symptoms, such as liver steatosis 1. The symptomatology observed in some of our patients may be a consequence of a decreased intestinal capacity for chylomicron synthesis/secretion (INTRODUIR REFERÈNCIA: Hooper AJ et al. J Clin Endocrinol Metab 2007; 92: 1474-1478) which is largely dependent on the conservation of the intestinal ApoB-48 sequence,. Protein truncation will always affect the complete liver-specific ApoB-100, indicating that the APOB gene will not form the full protein needed for VLDL, intermediate-density lipoprotein (IDL), LDL and Lipoprotein (a) structure and function35. There were 21 index patients which could not be genetically diagnosed. It is well known that up to 50% of interindividual variation in plasma LDLc is due to genetic variation by an accumulation of small-effect LDLc-impairing alleles11, 12 . In that sense, we observed a statistically significantly different frequency distribution of three APOB common genetic variants associated with LDLc concentration when comparing 503 subjects from the 1000 Genome Project with negative patients (HBL/M-) (Table 3). Two of the APOB genetic variants could be contributing to the observed phenotype, because the allele associated with decreased levels of LDLc was more frequent in HBL/M- than in the control population. These findings led us to test the recently proposed LDLc gene score to distinguish polygenic and monogenic forms of familial hypercholesterolemia11, 12 and primary HBL36 through a panel of SNPs that are known to influence LDLc levels. The primary HBL group without known mutation (HBL/M-) had a significantly lower mean weighted LDLc gene score when compared to the European 1000 Genome Project participants suggests that some HBL/M- 15 patients may present a polygenic inheritance pattern It is noteworthy that a recent work carried out with 121 women with LDLc below the first percentile, around 50% of the cases with primary HBL could be explained by polygenic predisposition36. Recent studies showed a polygenic basis for some cases previously identified as monogenic familial hypercholesterolemia11, 12, we did not find differences in the weighted LDLc gene of HBL/M+ and HBL/M-, suggesting that polygenicity influences phenotype expression in both forms of the disease. This study has several limitations mainly due to the retrospective manner in which it was conducted from a clinical laboratory. One is the heterogeneous degree of clinical information available according to that supplied by the clinician who requested the molecular analysis. The requests for molecular analyses were accepted although the information received was sometimes incomplete. Despite this, the results (identification of pathogenic gene variants in 52% of the cases) are similar or better to that obtained by us and others in the case of Familial Hypercholesterolemia (Martín-Campos JM et al. Autosomal dominant hypercholesterolemia in Catalonia: Correspondence between clinical-biochemical and genetic diagnostics in 967 patients studied in a multicenter clinical setting. J Clin Lipidol 2018; in press), which argues in favor of a solid suspicion of primary hypobetalipoproteinemia by the practitioners that requested the genetic study. Another limitation is that large rearrangements in the studied genes were not evaluated. According to HGMD, large rearrangements in these genes are scarce. Lastly, the idea of finding negative patients harboring these or other pathogenic variants only detectable through whole exome or genome sequencing should not be ruled out since this study does not exclude the possibility that other pathogenic and low- frequency mutations in other genes which could also cause primary HBL. In conclusion, our report expands the existing molecular information relating to primary HBL, confirming the heterogenic nature and presentation of this syndrome. Although this condition has been considered a monogenic disorder, our findings support the idea that some cases may also have a polygenic origin as recently suggested36. The concept of polygenicity, 16 demonstrated previously in other dyslipidemias11, 12, 37, 38 , should be further examined in patients with clinical and biochemical data of primary HBL without monogenic disorders. Conflicts of interest None Financial support This study was funded by Instituto de Salud Carlos III (ISCIII) and FEDER “Una manera de hacer Europa,” grants PI14/01648 (to FBV and JMMC) and CP13/00070 (to JJ), and Fundació la Marató de TV3 grant 20152431 (to FBV). JJ is a Miguel Servet researcher, funded by ISCIII (CP13/00070). CIBER de Diabetes y Enfermedades Metabólicas Asociadas (CIBERDEM) is a project of the ISCIII. Author contributions FBV, JMMC, JJ and MT devised the study. JMMC and MT did the data analysis. ABV, MC, SM, RR, and NF contributed to the genotyping. FBV, JMMC and MT interpreted the data and wrote the manuscript. JJ also discussed the results and contributed to the final version of the manuscript. 17 Figure 1. Weighted LDLc score as mean (SD) was 0.8868 (0.212) for EUR controls, 0.8033 (0.217) for HBL/M+ patients, and 0.7366 (0.327) for HBL/M- patients. The significance of gene score between HBL/M- and controls (A), HBL/M+ and controls (B), and HBL/M- and HBL/M+ (C) was p = 0.043, p = 0.07, and p = 0.618, respectively. 18 Table 1. Genetic data from patients affected with one or two pathogenic variants potentially causing primary HBL. LDLc Patient Gene DNA (Pathogenic Variants)a Proteinb ApoB % IDc Probandd score 8 APOB c.3G>C p.Met1Ile new Hz 0.972 c.6630_6631delAA p.Leu2212* ApoB-48 new Hz 7 APOB c.133C>T p.Arg45* ApoB-0 new Hz 1.007 13 APOB c.1124G>A p.Ser375Lysfs*93 ApoB-10 CS07214239 Hz 0.700 12 APOB c.1315C>T p.Arg439* ApoB-9 CM02148740 Hz 0.783 1 APOB c.1407C>A p.Tyr469* ApoB-10 new Hz 0.813 9 APOB c.1407C>A p.Tyr469* ApoB-10 new Hz 1.044 2 APOB c.2726C>A p.Ser909* ApoB-19 new Hz 0.576 4 APOB c.2726C>A p.Ser909* ApoB-19 new Hm 0.707 14 APOB c.2726C>A p.Ser909* ApoB-19 new Hz 0.130 11 APOB c.2781delT p.Ser928Profs*25 ApoB-20 new Hz 0.881 5 APOB c.3843-2A>G NA CS07213941 Hz 0.958 15 APOB c.3997C>T p.Arg1333* ApoB-29 rs121918383/CM88000642 Hz 0.946 10 APOB c.6034C>T p.Arg2012* ApoB-44 rs147863759/CM96263443 Hz 0.655 19 3 APOB c.7600C>T p.Arg2534* ApoB-55 CM05100544 Hz 0.915 17 APOB c.7600C>T P.Arg2534* ApoB-55 CM05100544 Hz 0.655 16 APOB c.10430_10440delTTGACCACAAG p.val3477Alafs*2 ApoB-76 new Hz 0.835 6 APOB c.11728G>T p.Glu3910* ApoB-86 new Hz 0.546 20 SAR1B c.1-4482_58+1406del5946InsGCATGATGGCGGGT NA - CG08107929 Hm 0.965 18 SAR1B c.83_84delTG p.Leu28Argfs*7 - new Hm 0.566 19 SAR1B c.409G>A p.Asp137Asn - rs28942109/CM03559730 Hm 0.903 21 ANGPTL3 c.363_367delCTCAA p.Asn121Leufs*3 - rs398122988/CD111013331 Hm 1.070 22 ANGPTL3 c.363_367delCTCAA p.Asn121Leufs*3 - rs398122988/CD111013331 Hm 1.018 23 MTTP c.1691T>C p.Ile564Thr - CM06832732 Hm 0.835 HBL, hypobetalipoproteinemia. NA: not applicable; a, positions are given according to the reference sequences: NM_000384 (APOB), NM_016103 (SAR1B), NM_014495 (ANGPTL3), and NM_000253 (MTTP). b, positions are given according to the UniProt sequences: P04114 (APOB), Q9Y6B6 (SAR1B), Q9Y5C1 (ANGPTL3), and P55157 (MTTP). c, in dbSNP (rs) and/or Human Gene Mutation Database (HGMD). d, Hm, homozigous; Hz, heterozigous. 20 Table 2. Variants of unknown significance associated to genetically and non-genetically diagnosed patients. Patient Gene DNAa Proteinb IDc Associated tod Probande LDLc score 1 APOB c.1853C>T p.Ala618Val rs679899/CM984191 ApoB45 Hz 0.813 c.11248C>T p.His3750Tyr new Hz 2 APOB c.293C>T p.Thr98Ile rs1367117/CM020623 Cholesterol46, 47 Hz 0.576 3 APOB c.13013G>A p.Ser4338Asn rs1042034/CM133911 Total cholesterol48 Hm 0.915 5 APOB c.1853C>T p.Ala618Val rs679899/CM984191 ApoB45 Hz 0.958 c.8216C>T p.Pro2739Leu rs676210/CM980093 Hypocholesterolemia49, 50 Hz 6 APOB c.8216C>T p.Pro2739Leu rs676210/CM980093 Hypocholesterolemia49, 50 Hz 0.546 c.13013G>A p.Ser4338Asn rs1042034/CM133911 Total cholesterol48 Hz 7 APOB c.293C>T p.Thr98Ile rs1367117/CM020623 Cholesterol46, 47 Hz 1.007 c.1853C>T p.Ala618Val rs679899/CM984191 ApoB45 Hz c.8216C>T p.Pro2739Leu rs676210/CM980093 Hypocholesterolemia49, 50 Hz c.13013G>A p.Ser4338Asn rs1042034/CM133911 Total cholesterol48 Hz 8 APOB c.1853C>T p.Ala618Val rs679899/CM984191 ApoB45 Hm 0.972 c.1254G>A p.Glu4181Lys rs1042031/CM068651 Ischemia and stroke51, 52 Hz c.13013G>A p.Ser4338Asn rs1042034/CM133911 Total cholesterol48 Hm 21 9 APOB c.1853C>T p.Ala618Val rs679899/CM984191 ApoB45 Hz 1.044 c.3740A>G p.Tyr1247Cys rs61741164 Possibly Damaging Hz c.11245C>T p.His3750Tyr new Hz c.13013G>A p.Ser4338Asn rs1042034/CM133911 Total cholesterol48 Hm c.13451C>T p.Thr4484Met rs12713450/CM104784 Hypertrigliceridemia53 Hz 15 APOB c.1853C>T p.Ala618Val rs679899/CM984191 ApoB45 Hz 0.946 16 APOB c.1853C>T p.Ala618Val rs679899/CM984191 ApoB457 Hz 0.835 c.11248C>T p.His3750Tyr new Hz 17 APOB c.1853C>T p.Ala618Val rs679899/CM984191 ApoB45 Hz 0.655 c.13013G>A p.Ser4338Asn rs1042034/CM133911 Total cholesterol48 Hz 19 APOB c.1853C>T p.Ala618Val rs679899/CM984191 ApoB45 Hz 0.903 21 APOB c.1853C>T p.Ala618Val rs679899/CM984191 ApoB45 Hz 1.070 c.13013G>A p.Ser4338Asn rs1042034/CM133911 Total cholesterol48 Hm 22 APOB c.13013G>A p.Ser4338Asn rs1042034/CM133911 Total cholesterol48 Hm 1.018 23 APOB c.1853C>T p.Ala618Val rs679899/CM984191 ApoB45 Hz 0.835 24 APOB c.1853C>T p.Ala618Val rs679899/CM984191 ApoB45 Hz 1.138 c.13013G>A p.Ser4338Asn rs1042034/CM133911 Total cholesterol48 Hm MYLIP c.1025A>G p.Asn342Ser rs9370867/CM116474 High degrad. of LDLR56 Hz No effect57, 58 APOC3 c.*71G>T NA rs4225/CR1611071 ↓TG concentration59 Hm 22 25 APOB c.1853C>T p.Ala618Val rs679899/CM984191 ApoB45 Hz 0.450 c.8216C>T p.Pro2739Leu rs676210/CM980093 Hypocholesterolemia49, 50 Hz c.13013G>A p.Ser4338 Asn rs1042034/CM133911 Total cholesterol48 Hz PCSK9 c.43_44insCTG p.Leu23_Gly24insLeu rs35574083/CI102605/CE062370 Low LDLc60, 61 Hz c.137G>T p.Arg46Leu rs11591147/CM061168 Low LDLc, ↓risk CHD62 Hz c.158C>T p.Ala53Val rs115836807CM102603 Hypocholesterolemia60 Hz c.1420G>A p.Val474Ile rs562556/CM045990 LDLc54, 55 Hz MTTP c-101-478G>T Not applicable rs1800591/CR982417 Low LDLc, SNAFLD63, 64 Hz c.891G>C p.Gln297His rs2306985/CM1510847 Low LDLc63 Hz c.383T>C p.Ile128Thr rs3816873 Hz APOC3 c.*71G>T NA rs4225/CR1611071 ↓TG concentration59 Hm 26 APOB c.1853C>T p.Ala618Val rs679899/CM984191 ApoB45 Hz 0.819 c.8216C>T p.Pro2739Leu rs676210/CM980093 Hypocholesterolemia49, 50 Hz c.13013G>A p.Ser4338Asn rs1042034/CM133911 Total cholesterol48 Hz APOC3 c.*71G>T NA rs4225/CR1611071 ↓TG concentration59 Hm 27 APOB c.1853C>T p.Ala618Val rs679899/CM 984191 ApoB45 Hz 0.837 c.8216C>T p.Pro2739Leu rs676210/CM980093 Hypocholesterolemia49, 50 Hz c.13013G>A p.Ser4338 Asn rs1042034/CM133911 Total cholesterol48 Hz PCSK9 c.43_44insCTG p.Leu23_Gly24insLeu rs35574083/CI102605/CE062370 Low LDLc60, 61 Hz c.158C>T p.Ala53Val rs11583680/CM102603 Hypocholesterolemia60 Hz 23 c.1420G>A p.Val474Ile rs562556/CM045990 LDLc54, 55 Hz 28 APOB c.1853C>T p.Ala618Val rs679899/CM984191 ApoB45 Hz 0.281 c.13013G>A p.Ser4338Asn rs1042034/CM133911 Total cholesterol48 Hm MTTP c.891G>C p.Gln297His rs2306985/CM1510847 Low LDLc63 Hz c.1151A>C p.Asp384Ala rs17029215/CM057721 Hz 29 APOB c.1853C>T p.Ala618Val rs679899/CM984191 ApoB45 Hz 0.867 c.8216C>T p.Pro2739Leu rs676210/CM980093 Hypocholesterolemia49, 50 Hz c.13013G>A p.Ser4338Asn rs1042034/CM133911 Total cholesterol48 Hz APOC3 c.102T>C p.Gly34= rs4520/CM942309 ↓ plasma TG65 Hz c.*40G>C NA rs5128/041556 ↑TG concentrations59 Hz c.*71G>T NA rs4225/CR1611071 ↓TG concentration59 Hz 30 APOB c.13013G>A p.Ser4338Asn rs1042034/CM133911 Total cholesterol48 Hz - PCSK9 c.1420G>A p.Val474Ile rs562556/CM045990 LDLc54, 55 Hz 31 APOB c.1853C>T p.Ala618Val rs679899/CM984191 ApoB45 Hz 0.935 c.8216C>T p.Pro2739Leu rs676210/CM980093 Hypocholesterolemia49, 50 Hz c.13013G>A p.Ser4338 Asn rs1042034/CM133911 Total cholesterol48 Hz MTTP c.891G>C p.Gln297His rs2306985/CM1510847 Low LDLc63 Hm 32 APOB c.3383G>A p.Arg1128His rs12713843/CM040371 “Recessive” HBL66 Hz 1.136 c.8216C>T p.Pro2739Leu rs676210/CM980093 Hypocholesterolemia49, 50 Hz c.13013G>A p.Ser4338Asn rs1042034/CM133911 Total cholesterol48 Hz 24 PCSK9 c.43_44insCTG p.Leu23_Gly24insLeu rs35574083/CI102605/CE062370 Low LDLc60, 61 Hz c.137G>T p.Arg46Leu rs11591147/CM061168 Low LDLc, ↓risk CHD62 Hz c.158C>T p.Ala53Val rs11583680/CM102603 Hypocholesterolemia60 Hz c.1420G>A p.Val474Ile rs562556/CM045990 LDLc54, 55 Hz APOC3 c.*71G>T NA rs4225/CR1611071 ↓TG concentration59 Hm MTTP c.891G>C p.Gln297His rs2306985/CM1510847 Low LDLc63 Hz c.1151A>C p.Asp384Ala rs17029215/CM057721 Hz 33 APOB c.1853C>T p.Ala618Val rs679899/CM984191 ApoB45 Hz 1.018 c.8216C>T p.Pro2739Leu rs676210/CM980093 Hypocholesterolemia49, 50 Hz c.13013G>A p.Ser4338 Asn rs1042034/CM133911 Total cholesterol48 Hz PCSK9 c.43_44insCTG p.Leu23_Gly24insLeu rs35574083/CI102605/CE062370 Low LDLc60, 61 Hz c.158C>T p.Ala53Val rs11583680/CM102603 Hypocholesterolemia60 Hz APOC3 c.102T>C p.Gly34= rs4520 /CM942309 ↓ plasma TG65 Hz c.*40G>C NA rs5128/041556 ↑TG concentrations67 Hz c.*71G>T NA rs4225/CR1611071 ↓TG concentration59 Hz 34 APOC3 c.*71G>T NA rs4225/CR1611071 ↓TG concentration59 Hz 0.963 35 APOB c.2188G>A p.Val730Ile rs12691202/CM981492 Impairment of HBL68 Hz 0.573 APOC3 c.102T>C p.Gly34= rs4520/CM942309 ↓ plasma TG65 Hm 36 APOC3 c.102T>C p.Gly34= rs4520/CM942309 ↓ plasma TG65 Hz 1.023 c.*71G>T NA rs4225/CR1611071 ↓TG concentration59 Hz 25 37 APOB c.13013G>A p.Ser4338Asn rs1042034/CM133911 Total cholesterol48 Hz 0.776 APOC3 c.*71G>T NA rs4225/R1611071 ↓TG concentration59 Hz 38 APOB c.1853C>T p.Ala618Val rs679899/CM984191 ApoB45 Hz -0.187 c.8216C>T p.Pro2739Leu rs676210/CM980093 Hypocholesterolemia49, 50 Hz c.13013G>A p.Ser4338Asn rs1042034/CM133911 Total cholesterol48 Hz APOC3 c.*71G>T NA rs4225/CR1611071 ↓TG concentration59 Hz 39 APOB c.1853C>T p.Ala618Val rs679899/CM984191 ApoB45 Hz 0.651 APOC3 c.102T>C p.Gly34= rs4520/CM942309 ↓ plasma TG65 Hm 40 APOB c.3427C>T p.Pro1143Ser rs72653077/CM087422 Hypocholesterolemia35 Hz 0.506 41 APOB c.1853C>T p.Ala618Val rs679899/CM984191 ApoB45 Hm 0.831 c.8216C>T p.Pro2739Leu rs676210/CM980093 Hypocholesterolemia49, 50 Hz c.1254G>A p.Glu4181Lys rs1042031/CM068651 Ischemia and stroke51, 52 Hz c.13013G>A p.Ser4338Asn rs1042034/CM133911 Total cholesterol48 Hz 42 APOB c.13013G>A p.Ser4338Asn rs1042034/CM133911 Total cholesterol48 Hm 1.027 c.293C>T p.Thr98Ile rs1367117/CM020623 Cholesterol46, 47 Hz 43 APOB c.293C>T p.Thr98Ile rs1367117/CM020623 Cholesterol46, 47 Hz 0.659 c.13013G>A p.Ser4338Asn rs1042034/CM133911 Total cholesterol48 Hm c.13451C>T p.Thr4484Met rs12713450/CM104784 Hypertrigliceridemia53 Hz 44 APOB c.293C>T p.Thr98Ile rs1367117/CM020623 Cholesterol46, 47 Hz 0.427 26 a , positions are given according to the reference sequences: NM_000384 (APOB), NM_016103 (SAR1B), NM_014495 (ANGPTL3), NM_000253 (MTTP), NM_174936 (PCSK9), NM_013262 (MYLIP), and NM_000040 (APOC3). b, positions are given according to the UniProt sequences: P04114 (APOB), Q9Y6B6 (SAR1B), Q9Y5C1 (ANGPTL3), P55157 (MTTP), Q8NBP7 (PCSK9), Q8WY64 (MYLIP), and P02656 (APOC3). c, in dbSNP (rs) and/or Human Gene Mutation Database (HGMD). d, CHD: Coronary heart disease; SNAFLD: Susceptibility to non alcoholic fatty liver disease. e, Hm, homozigous; Hz, heterozigous. 27 Table 3. APOB common variants associated with plasma LDLc concentrations. Freqg Freqh SNP ID HGMDa Locationb DNAc Proteind pe MAf EUR HBL/M- Impacti Teslovich et al. (2010)j rs1367117 CM020623 exon 4 c.293C>T p.Thr98Ile 0.00052 T 0.298 0.028 moderate Allele T associated with high LDLc (p=4.5E-18) rs1042031 CM068651 exon 29 c.12541G>A p.Glu4181Lys 0.021 A 0.213 0.028 moderate Allele A associated with low LDLc (p=1.83E-20) rs1042034 CM133911 exon 29 c.13013G>A p.Ser4338Asn 0.0014 G 0.218 0.417 moderate Allele A associated with high LDLc (p=8.32E-22) a Identification in Human Gene Mutation Database. b According to reference sequence NM_000384. c According to UniProt sequence P04114. d Affected protein position. e Signification in Fisher’s exact test. f Minor allele (MA). g Frequency in the European population of the 1000 Genomes database. h Frequency in hypobetalipoproteinemia mutation negative patients (HBL/M-). i Result of the in silico analysis 28 regarding the impact on the functionality of the protein. j Association in the GWAS of Teslovich 27. 29 REFERENCES 1 Welty, F. K., Hypobetalipoproteinemia and abetalipoproteinemia, Curr Opin Lipidol, 2014, 25: 161-168. 2 Ramasamy, I., Update on the molecular biology of dyslipidemias, Clin Chim Acta, 2016, 454: 143-185. 3 Lee, J. and Hegele, R. A., Abetalipoproteinemia and homozygous hypobetalipoproteinemia: a framework for diagnosis and management, J Inherit Metab Dis, 2014, 37: 333-339. 4 Tarugi, P., Bertolini, S. and Calandra, S., Angiopoietin-like protein 3 (ANGPTL3) deficiency and familial combined hypolipidemia., J Biomed Res, 2018, [Epub ahead of print]: doi: 10.7555/JBR.7532.20170114. 5 Ferreira, H., Ramos, R. N., Quan, C. F., Ferreiro, S. R., Ruiz, V. C., et al., Chylomicron retention disease: a description of a new mutation in a very rare disease., Pediatr Gastroenterol Hepatol Nutr, 2018, 21: 134-140. 6 Hooper, A. J. and Burnett, J. R., Update on primary hypobetalipoproteinemia, Curr Atheroscler Rep, 2014, 16: 423. 7 Jørgensen, A. B., Frikke-Schmidt, R., Nordestgaard, B. G. and Tybjærg-Hansen, A., Loss-of-Function Mutations in APOC3 and Risk of Ischemic Vascular Disease, New England Journal of Medicine, 2014, 371: 32-41. 8 Sorrentino, V., Fouchier, S. W., Motazacker, M. M., Nelson, J. K., Defesche, J. C., et al., Identification of a loss-of-function inducible degrader of the low-density lipoprotein receptor variant in individuals with low circulating low-density lipoprotein, Eur Heart J, 2013, 34: 1292-1297. 9 Chandak, P. G., Obrowsky, S., Radovic, B., Doddapattar, P., Aflaki, E., et al., Lack of acyl-CoA:diacylglycerol acyltransferase 1 reduces intestinal cholesterol absorption and attenuates atherosclerosis in apolipoprotein E knockout mice, Biochim Biophys Acta, 2011, 1811: 1011-1020. 10 Stitziel, N. O., Won, H. H., Morrison, A. C., Peloso, G. M., Do, R., et al., Inactivating Mutations in NPC1L1 and Protection from Coronary Heart Disease., New England Journal of Medicine, 2014, 371: 2072-2082. 11 Talmud, P. J., Shah, S., Whittall, R., Futema, M., Howard, P., et al., Use of low- density lipoprotein cholesterol gene score to distinguish patients with polygenic and monogenic familial hypercholesterolaemia: a case-control study, Lancet, 2013, 381: 1293- 1301. 12 Futema, M., Shah, S., Cooper, J. A., Li, K., Whittall, R., et al., Refinement of variant selection for the LDL cholesterol genetic risk score in the diagnosis of the polygenic form of clinical familial hypercholesterolemia and replication in samples from 6 countries., Clin Chem, 2015, 61: 231-238. 13 Gómez-Gerique, J., Gutiérrez-Fuentes, J., Montoya, M., Porres, A., Rueda, A., et al., [The lipid profile of spanish population: the DRECE study (Diet and Risk of Cardiovascular Disease in Spain)], Med Clin-Barcelona, 1999, 113: 730-735. 14 Villar-Serrano, J., Server-Huertas, R. and Bretó-Gilabert, M., [Serum levels of apolipoproteins A-I and B in a population sample of the Comunidad Valenciana: utility for assesment of cardiovascular risk.], Clin Invest Arterioscl, 2001, 13: 195-199. 15 Staden, R., Beal, K. and Bonfield, J., The Staden package, 1998, Methods Mol Biol, 2000, 132: 115-130. 16 Vaser, R., Adusumalli, S., Leng, S. N., Sikic, M. and Ng, P. C., SIFT missense predictions for genomes, Nat Protoc, 2016, 11: 1-9. 17 Adzhubei, I. A., Schmidt, S., Peshkin, L., Ramensky, V. E., Gerasimova, A., et al., A method and server for predicting damaging missense mutations, Nat Methods, 2010, 7: 248- 249. 30 18 Mi, H., Muruganujan, A. and Thomas, P. D., Panther in 2013: modeling the evolution of gene function, and other gene attributes, in the context of phylogenetic trees., Nucleic Acid Research, 2013, 41: D377-D386. 19 Choi, Y., Sims, G. E., Murphy, S., Miller, J. R. and Chan, A. P., Predicting the functional effect of aminoacid substitutions and indels., PLoS ONE, 2012, 7: e46688. 20 Capriotti, E., Calabrese, R. and Casadio, R., Predicting the insurgence of human genetic diseases associated to single point protein mutations with Support Vector Machines and evolutionary information., Bioinformatics, 2006, 22: 2729-2734. 21 Yue, P., Melamud, E. and Mould, J., SNPs3D: candidate gene and SNP selection for association studies., BMC Bioinformatics, 2006, 7: 166. 22 Ferrer-Costa, C., Gelpí, J. L., Zamakola, L., Parraga, I., de la Cruz, X., et al., PMUT: a web-based tool for the annotation of pathological mutations on proteins., Bioinformatics, 2005, 21: 3176-3178. 23 Schwarz, J. M., Cooper, D. N., Schuelke, M. and Seelow, D., MutationTaster2: mutation prediction for the deep-sequencing age., Nat Methods, 2014, 11: 361-362. 24 Reva, B. A., Antipin, Y. A. and Sander, C., Predicting the functional impact of protein mutations: application to cancer genomics., Nucleic Acid Res, 2011, 39: e118. 25 Desmet, F. O., Hamroun, D., Lalande, M., Collod-Beround, G., Claustres, M., et al., Human Splicing Finder: an online bioinformatics tool to predict splicing signals., Nucleic Acid Res, 2009, 37: e67. 26 Richards, S., Aziz, N., Bale, S., Bick, D., Das, S., et al., Standards and guidelines for the interpretation of sequence variants: a joint consensus recommendation of the American College of Medical Genetics and Genomics and the Association for Molecular Pathology., Genet Med, 2015, 17: 405-424. 27 Teslovich, T. M., Musunuru, K., Smith, A. V., Edmondson, A. C., Stylianou, I. M., et al., Biological, clinical and population relevance of 95 loci for blood lipids, Nature, 2010, 466: 707-713. 28 Burnett, J. R. and Hooper, A. J., Common and rare gene variants affecting plasma LDL cholesterol, Clin Biochem Rev, 2008, 29: 11-26. 29 Charcosset, M., Sassolas, A., Peretti, N., Roy, C. C., Deslandres, C., et al., Anderson or chylomicron retention disease: molecular impact of five mutations in the SAR1B gene on the structure and the functionality of Sar1b protein., Mol Genet Metab, 2008, 93: 74-84. 30 Jones, B., Jones, E., Bonney, S., Patel, H., Mensenkamp, A., et al., Mutations in a Sar1 GTPase of COPII vesicles are associated with lipid adsorption disorders., Nat Genet, 2003, 34: 29-31. 31 Martín-Campos, J. M., Roig, R., Mayoral, C., Martinez, S., Martí, G., et al., Identification of a novel mutation in the ANGPTL3 gene in two families diagnosed of familial hypobetalipoproteinemia without APOB mutation., Clin Chim Acta, 2012, 413: 552-555. 32 Sakamoto, O., Abukawa, D., Takeyama, J., Arai, N., Nagano, M., et al., An atypical case of abetalipoproteinaemia with severe fatty liver in the absence of steatorrhoea or acanthocytosis., Eur J Pediatr, 2006, 165: 68-70. 33 Pocovi, M. and Civeira, F., [Clinical and genetic heterogeneity in hypobetalipoproteinemia], Med Clin-Barcelona, 2009, 133: 61-62. 34 Martin-Morales, R., Garcia-Diaz, J. D., Tarugi, P., Gonzalez-Santos, P., Saavedra- Vallejo, P., et al., Familial hypobetalipoproteinemia: analysis of three Spanish cases with two new mutations in the APOB gene, Gene, 2013, 531: 92-96. 35 Leren, T. P. and Berge, K. E., Identification of mutations in the apolipoprotein B-100 gene and in the PCSK9 gene as the cause of hypocholesterolemia, Clin Chim Acta, 2008, 397: 92-95. 36 Balder, J.-W., Rimbert, A., Zhang, X., Kanninga, R., Viel, M., et al., Genetics, lifestyle, and low-density lipoprotein cholesterol in young and apparently healthy womwn, Circulation, 2018, 137: 820-831. 37 Vakkilainen, J., Pajukanta, P., Cantor, R. M., Nuotio, I. O., Lahdenperä, S., et al., Genetic influences contributing to LDL particle size in familial combined hyperlipidaemia., Eur J Hum Genet, 2002, 10: 547-552. 31 38 Motazacker, M. M., Peter, J., Treskes, M., Shoulders, C. C., Kuivenhoven, J. A., et al., Evidence of a polygenic origin of extreme high-density lipoprotein cholesterol levels., Arterioscler Thromb Vasc Biol, 2013, 33: 1521-1528. 39 Homer, V. M. and George, P. M., Novel Mutation (c.G1124A) in Exon 9 of the APOB Gene Causes Aberrant Splicing and Familial Hypobetalipoproteinemia, Clinical Chemistry, 2007, 53: 1165. 40 Huang, L. S., Kayden, H., Sokol, R. J. and Breslow, J. L., ApoB gene nonsense and splicing mutations in a compound heterozygote for familial hypobetalipoproteinemia, Journal of Lipid Research, 1991, 32: 1341-1348. 41 Di Leo, E., Magnolo, L., Lancellotti, S., Crocè, L., Visintin, L., et al., Abnormal apolipoprotein B pre-mRNA splicing in patients with familial hypobetalipoproteinaemia, Journal of Medical Genetics, 2007, 44: 219. 42 Collins, D. R., Knott, T. J., Pease, R. J., Powell, L. M., Wallis, S. C., et al., Truncated variants of apolipoprotein B cause hypobetalipoproteinaemia, Nucleic Acids Research, 1988, 16: 8361-8375. 43 Srivastava, N., Noto, D., Averna, M., Pulai, J., Srivastava, R. A. K., et al., A new apolipoprotein B truncation (apo B-43.7) in familial hypobetalipoproteinemia: Genetic and metabolic studies, Metabolism, 1996, 45: 1296-1304. 44 Fouchier, S. W., Sankatsing, R. R., Peter, J., Castillo, S., Pocovi, M., et al., High frequency of APOB gene mutations causing familial hypobetalipoproteinaemia in patients of Dutch and Spanish descent, Journal of Medical Genetics, 2005, 42: e23. 45 Ilmonen, M., Knudsen, P., Taskinen, M. R. and Tikkanen, M. J., Genetic variation in the amino-terminal part of apolipoprotein B: studies in hyperlipidemic patients., Atherosclerosis, 1998, 138: 367-374. 46 Bentzen, J., Jorgensen, T. and Fenger, M., The effect of six polymorphisms in the Apolipoprotein B gene on parameters of lipid metabolism in a Danish population, Clin Genet, 2002, 61: 126-134. 47 Hochner, H., Allard, C., Granot-Hershkovitz, E., Chen, J., Sitlani, C. M., et al., Parent- of-Origin Effects of the APOB Gene on Adiposity in Young Adults, PLoS Genet, 2015, 11: e1005573. 48 Kim, D. S., Burt, A. A., Ranchalis, J. E., Jarvik, E. R., Rosenthal, E. A., et al., Noven gene-by-environment interactions: APOB and NPC1L1 variants affect the relationship between dietary and total plasma cholesterol., J Lipid Res, 2013, 54: 1512-1520. 49 Gutiérrez-Cirlos, C., Ordóñez-Sánchez, M. L., Tusie-Luna, M. T., Patterson, B. W., Schonfeld, G., et al., Familial hypobetalipoproteinemia in a hospital survey: genetics, metabolism and non-alcoholic fatty liver disease., Ann Hepatol, 2011, 10: 155-164. 50 Leren, T. P., Bakken, K. S., Hoel, V., Hiermann, I. and Berg, K., screening for mutations of the apolipoprotein B gene causing hypocholesterolemia., Hum Genet, 1998, 102: 44-49. 51 Gaudreault, N., Ducharme, V., Lamontagne, M., Guauque-Olarte, S., Mathieu, P., et al., Replication of genetic association studies in aortic stenosis in adults, Am J Cardiol, 2011, 108: 1305-1310. 52 Mustafina, O. E., Novikova, L. B., Nasibullin, T. R., Kolchina, E. M. and Tuktarova, I. A., [An analysis of association between the apolipoprotein B gene EcoR1 polymorphism and ischemic stroke], Zh Nevrol Psikhiatr Im S S Korsakova, 2006, Suppl 17: 66-70. 53 Johansen, C. T., Wang, J., Lanktree, M. B., Cao, H., McIntyre, A. D., et al., Excess of rare variants in genes identified by genome-wide association study of hypertriglyceridemia, Nat Genet, 2010, 42: 684-687. 54 Al-Waili, K., Al-Zidi, W. A., Al-Abri, A. R., Al-Rasadi, K., Al-Sabti, H. A., et al., Mutation in the PCSK9 gene in Omani Arab subjects with autosomal dominant hypercholesterolemia and its effect on PCSK9 protein structure., Oman Med J, 2013, 28: 48-52. 55 Shioji, K., Mannami, T., Kokubo, Y., Inamoto, N., Takagi, S., et al., Genetic variants in PCSK9 affect the colesterol level in Japanese, J Human Genet, 2004, 49: 109-114. 32 56 Weissglas-Volkov, D., Calkin, A. C., Tusie-Luna, T., Sinsheimer, J. S., Zelcer, N., et al., The N342S MYLIP polymorphism is associated with high total cholesterol and increased LDL receptor degradation in humans., J Clin Invest, 2011, 121: 3062-3071. 57 Dhyani, A., Tibolla, G., Baragetti, A., Garlaschelli, K., Pellegatta, F., et al., IDOL N342S variant, atherosclerosis progression and cardiovascular disorders in the Italian general population., PLoS ONE, 2015, 10: e0122414. 58 Santos, P. C. J. L., Oliveira, T. G. M., Lemos, P. A., Mill, J. G., Krieger, J. E., et al., MYLIP p.N342S polymorphism is not associated with lipid profile in the Brazilian population., Lipids Health Dis, 2012, 11: 83-89. 59 Hu, S. L., Cui, G. L., Huang, J., Jiang, J. G. and Wang, D. W., An APOC3 3'UTR variant associated with plasma triglycerides levels and coronary heart disease by creating a functional miR-4271 binding dite., Scientific Report, 2016, 6: 32700. 60 Mayne, J., Ooi, T. C., Raymond, A., Cousins, M., Bernier, L., et al., Differential effects of PCSK9 loss of function variants on serum lipid and PCSK9 levels in Caucasian and African Canadian populations., Lipids Health Dis, 2013, 12: 70-81. 61 Yue, P., Averna, M., Lin, X. and Schonfeld, G., The c.43_44insCTG variation in PCSK9 is associated with low plasma LDL-cholesterol in a caucasian population., Human Mutation, 2006, 27: 460-466. 62 Cohen, J. C., Boerwinkle, E., Mosley, T. H. J. and Hobbs, H. H., Sequence variations in PCSK9, low LDL, and protection against coronary heart disease., N Engl J Med, 2006, 354: 1264-1272. 63 Hsiao, P. J., Lee, M. Y., Wang, Y. T., Jiang, H. J., Lin, P. C., et al., MTTP-297H polymorphism reduced serum cholesterol but increased risk of non-alcoholic fatty liver disease-a cross-sectional study, BMC Med Genet, 2015, 16: 93. 64 Zheng, W., Wang, L., Su.X. and Hu, H. F., MTP -493G>T polymorphism and susceptibility to nonalcoholic fatty liver disease: a meta-analysis., DNA Cell Biol, 2014, 33: 361-369. 65 Xu, C. F., Talmud, P., Schuster, H., Houlston, R., Miller, G., et al., Association between genetic variation at the APO AI-CIII-AIV gene cluster and familial combined hyperlipidaemia., Clin Genet, 1994, 46: 385-397. 66 Lancellotti, S., Di Leo, E., Penacchioni, J. Y., Balli, F., Viola, L., et al., Hypobetalipoproteinemia with an apparently recessive inheritamce due to a "de novo" mutation of apolipoprotein B., Biochim Biophys Acta, 2004, 1688: 61-67. 67 Newman, D. L., Hoffjan, S., Bourgain, C., Abney, M., Nicolae, R. I., et al., Are common disease susceptibility alleles the same in outbred and founder populations?, Eur J Hum Genet, 2004, 12: 584-590. 68 Lam, M. C. W., Singham, J., Hegele, R. A., Riazy, M., Hiob, M. A., et al., Familial hypobetalipoproteinemia-induced nonalcoholic steatohepatitis., Case Rep Gastroenterol, 2012, 6: 429-437. 33