Jump to
Abstract
Rationale: Autosomal-dominant hypercholesterolemia (ADH) is characterized by elevated low-density lipoprotein cholesterol levels and increased risk for coronary vascular disease. ADH is caused by mutations in the low-density lipoprotein receptor, apolipoprotein B, or proprotein convertase subtilisin/kexin 9. A number of patients, however, suffer from familial hypercholesterolemia 4 (FH4), defined as ADH in absence of mutations in these genes and thereafter use the abbreviation FH4.
Objective: To identify a fourth locus associated with ADH.
Methods and Results: Parametric linkage analysis combined with exome sequencing in a FH4 family resulted in the identification of the variant p.Glu97Asp in signal transducing adaptor family member 1 (STAP1), encoding signal transducing adaptor family member 1. Sanger sequencing of STAP1 in 400 additional unrelated FH4 probands identified a second p.Glu97Asp carrier and 3 additional missense variants, p.Leu69Ser, p.Ile71Thr, and p.Asp207Asn. STAP1 carriers (n=40) showed significantly higher plasma total cholesterol and low-density lipoprotein cholesterol levels compared with nonaffected relatives (n=91).
Conclusions: We mapped a novel ADH locus at 4p13 and identified 4 variants in STAP1 that associate with ADH.
Introduction
Autosomal-dominant hypercholesterolemia (ADH) is characterized by elevated low-density lipoprotein cholesterol (LDL-C) levels and increased risk for coronary vascular disease. ADH is a common disease with a prevalence of ≈1 in 200 individuals in most Western countries.1 To date, mutations in 3 genes, LDL receptor (LDLR), specific domains of apolipoprotein B (APOB), and proprotein convertase subtilisin/kexin 9 (PCSK9), have been shown to cause ADH.2–5 Although mutations in these genes can be found in a large proportion of clinically diagnosed ADH patients, the underlying molecular determinant remains unknown in a substantial fraction of patients.3 Unraveling these molecular determinants is a major focus of cardiovascular research, yet identification of novel ADH genes is largely challenged by the phenotypic and genetic heterogeneity of this disease.6,7
We describe STAP1, encoding signal transducing adaptor family member 1, as a fourth gene associated with ADH.
Editorial, see p 534
In This Issue, see p 533
Methods
A detailed description of the experimental procedure and statistical analysis is provided in the Online Data Supplement.
Briefly, we identified a family affected with ADH in which we first excluded the presence of causal LDLR, APOB, and PCSK9 mutations in the proband (Figure; familial hypercholesterolemia 4 [FH4]-0102-III-12). We then obtained genetic mapping data to identify shared haplotypes, phenocopies and nonpenetrants, and selected the most distantly related and informative individuals for exome sequencing. Exome sequencing was performed for 3 affected family members (FH4-0102-III-12, FH4-0102-IV-4, and FH4-0102-IV-11).
Family FH4-0102. Half-blackened symbols indicate affected members (>95th percentile), quarter-blackened symbols indicate possibly affected members (between 75th and 95th percentile), and white symbols indicate unaffected members (<75th percentile). A dot indicates carriers of the p.Glu97Asp variant. The triangle represents the proband. Age (in years) at lipid measurement, total cholesterol (TC), low-density lipoprotein cholesterol (LDL-C), high-density lipoprotein cholesterol (HDL-C), triglycerides (TG) in mmol/L, and history and age onset of coronary vascular disease are given. Percentiles matched for age and sex, pTC, pLDL, pHDL, and pTG are given between brackets, a percentile indicated by 99 resembles a >95th percentile. Individuals with mapping data are marked with +. For individuals with exome data symbols are marked with brackets. AMI indicates acute myocardial infarction; CABG, coronary artery bypass graft; CVD, coronary vascular disease; and UPN, unique personal number.
Results
Parametric linkage analysis revealed the highest possible logarithm of odds scores when individual FH4-0102-III-7, FH4-0102-III-8, and FH4-0102-III-11 were considered to be phenocopies, resulting in highly suggestive LOD scores of 3.0 (θ=0.0) at chromosomes 4p15.1-q13.3, 11p13-q14.1, and 13q14.13-q32.1 (Online Figure I). Exome sequencing was performed for 3 affected family members with shared haplotypes at the 3 identified linkage intervals. Several discrete filtering steps (Online Table I) of the exome sequence data led to the identification of a variant in STAP1, p.Glu97Asp (c.291G>C). This variant, which was verified by Sanger sequencing, affected a highly conserved amino acid and was predicted to be pathogenic by SIFT, Polyphen-2, and MutationTaster. The variant was absent in 400 controls of matched ancestry, the Genome of the Netherlands (http://www.nlgenome.nl/), the 1000G population, and the National Heart, Lung, and Blood Institute-Exome Sequencing Project data set (http://evs.gs.washington.edu/EVS/), suggesting that it is not a polymorphism. Testing of remaining family members by Sanger sequencing revealed 11 affected carriers (total cholesterol or LDL-C levels >95th percentile), 4 possibly affected carriers (between 75th and 95th percentile) and 1 unaffected carrier (<75th percentile), reflecting a disease penetrance of 0.94 (Figure).
We subsequently sequenced all coding regions of STAP1 in 400 unrelated FH4 probands (Online Table II) and identified a second p.Glu97Asp carrier and 3 additional missense variants, p.Leu69Ser (c.206T>C), p.Ile71Thr (c.212T>C), and p.Asp207Asn (c.619G>A). All variants were located in exon 3, except for p.Asp207Asn located in exon 6, and all variants altered highly conserved amino acid residues. SIFT, Polyphen-2, and MutationTaster predicted the p.Leu69Ser variant to be pathogenic, whereas p.Ile71Thr and p.Asp207Asn were predicted to possibly affect protein function. None of these variants were detected in 400 controls, the 1000G population, or in the Genome of the Netherlands. In the National Heart, Lung, and Blood Institute-Exome Sequencing Project data set, p.Ile71Thr was found once and p.Asp207Asn was found 3×. However, variants at low frequency associated with an hypercholesterolemic phenotype can be expected, considering the substantial number of patients at increased cardiovascular risk included in National Heart, Lung, and Blood Institute-Exome Sequencing Project.
The frequency of rare STAP1 variants predicted to be pathogenic or possibly pathogenic in our FH4 cohort was 1.3% (5 of 400 individuals). This was 3× and 6× higher than the rate observed in National Heart, Lung, and Blood Institute-Exome Sequencing Project (0.4%, 17 of 4300 European American individuals; P=0.035) and the Genome of the Netherlands (0.2%, 1 of 499 individuals; P=0.067), respectively (Online Table III).
The STAP1 variants found in the 4 additional probands were further evaluated in the relatives of the replicate families. In total, 45 carriers of the rare STAP1 variants and 91 noncarrier relatives were identified. Individuals carrying a rare STAP1 variant had significantly higher total cholesterol and LDL-C levels compared with relatives without the STAP1 variants (Table). High-density lipoprotein cholesterol levels were not significantly different, whereas triglycerides levels were slightly increased. Similar results were observed when the carriers of the discovery family were excluded (Online Table IV).
Clinical Characteristics of Carriers and Noncarriers
We additionally compared the ADH phenotype of the STAP1 carriers with the ADH phenotype observed in a large cohort of LDLR and APOB mutation carriers. Total cholesterol and LDL-C levels were significantly lower in STAP1 carriers when compared with patients molecularly diagnosed with familial hypercholesterolemia (FH, OMIM 606945), caused by mutations in LDLR. However, cholesterol levels in STAP1 carriers were comparable with those observed in patients molecularly diagnosed with familial defective apolipoproteinemia B (OMIM 107730), caused by mutations in APOB.
In studies looking at ADH, individuals are considered affected if their plasma cholesterol levels are >95th percentile adjusted for age and sex, and in ADH-affected families a disease penetrance of 0.9 is usually assumed. Nevertheless, the ADH phenotype of patients affected by well-established pathogenic mutations in the LDLR and APOB varies enormously from severe to moderate, as recently demonstrated.8,9 In the 5 STAP1 families identified here, such variation was also observed, most evident in family FH4-0150 (Online Figure IIA), which displayed only a moderate penetrance of 0.76. The variants p.Ile71Thr in family FH4-0247 (Online Figure IIB) and p.Glu97Asp in family FH4-317 (Online Figure IIC) displayed complete penetrance. Further expansion of family FH4-0356 was unfortunately not possible (Online Figure IID). Overall, the disease penetrance of the missense STAP1 variants in all 5 families is 0.94±0.10 with a phenocopy rate of 0.19±0.11.
Discussion
We report here for the first time association of rare STAP1 variants with increased total cholesterol and LDL-C levels. The lipid phenotype of STAP1 carriers is not as severe as generally observed in FH patients, but resembles the minor lipid and coronary vascular disease phenotype observed in familial defective apolipoproteinemia B patients.4 It is of note that in genome-wide association studies, single nucleotide polymorphism in the STAP1 locus have not been shown to be associated with cholesterol levels. However, single nucleotide polymorphisms in the STAP1 locus have been associated with the risk for Parkinson disease.10 Although the exact biological explanation for this association remains elusive, increased cholesterol levels might play a role because high total serum cholesterol levels have been associated with a slow clinical progression of Parkinson disease.11
The function of STAP1, also known as BRDG1 (BCR downstream signaling protein 1) or stem cell adaptor protein 1, is largely unknown. STAP1 contains a Pleckstrin homology domain, a Src homology 2 domain, and several tyrosine phosphorylation sites.12 It has been suggested that the Pleckstrin homology domain of STAP1 functions as a phosphoinositide-binding domain and facilitates the association of STAP1 with membranes.12 Intriguingly, the p.Leu69Ser, p.Ile71Thr, and p.Glu97Asp alterations all occur within the Pleckstrin homology domain of STAP1 and may thus affect its interaction with membranes or membrane proteins (Online Figure III). The STAP1 p.Asp207Asn variant is located in the Src homology 2 domain, a domain implicated in propagation of signal transduction emanating from upstream receptor tyrosine kinase pathways.13 Indeed, STAP1 has been suggested to act downstream of the receptor tyrosine kinases c-kit and c-fms.12 In line with STAP1 functioning downstream of c-kit to control systemic cholesterol levels, it is interesting to point out that W/Wv mice, which harbor a loss-of-function mutation in c-kit, show increased plasma cholesterol levels.14,15 Similarly, gain-of-function mutations in c-kit cause leukemia,16 and leukemic patients often present with hypocholesterolemia.17–21 Intriguingly, therapeutic administration of receptor tyrosine kinase inhibitors to these patients is associated with increases in plasma cholesterol.22,23 This raises the possibility that STAP1 and potentially other downstream genes in this signaling pathway have an effect on circulating levels of cholesterol.
Our study provides strong genetic evidence for involvement of STAP1 in controlling cholesterol homeostasis, yet the molecular mechanism behind this is presently unknown. Expression of STAP1 in the liver, the main organ maintaining whole-body cholesterol homeostasis, is low (not shown). Studies aimed at identifying the extrahepatic mechanism underlying the potential role of STAP1 in cholesterol metabolism are required and are expected to yield new insight into the pathogenesis of ADH.
Acknowledgments
We thank individuals and families for their participation in this study and referring doctors, nurses, and other staff for their assistance. We thank Sekar Kathiresan and Nathan Stitziel for providing us with the exome sequence data.
Sources of Funding
J.J.P. Kastelein is a recipient of the Lifetime Achievement Award of the Dutch Heart Foundation (2010T082). N. Zelcer is supported by a European Research Council consolidator grant (617376) and holder of a Vidi grant (17106355) from the Netherlands Organisation for Scientific Research (NWO). G.K. Hovingh is holder of a Veni grant (91612122) from the NWO. This work is supported by CardioVascular Research Initiative (CVON2011-19; Genius) and the European Union (Resolve: FP7-305707 and TransCard: FP7-603091-2).
Disclosures
None.
Footnotes
In June 2014, the average time from submission to first decision for all original research papers submitted to Circulation Research was 15 days.
Brief UltraRapid Communications are designed to be a format for manuscripts that are of outstanding interest to the readership, report definitive observations, but have a relatively narrow scope. Less comprehensive than Regular Articles but still scientifically rigorous, BURCs present seminal findings that have the potential to open up new avenues of research. A decision on BURCs is rendered within 7 days of submission.
The online-only Data Supplement is available with this article at http://circres.ahajournals.org/lookup/suppl/doi:10.1161/CIRCRESAHA.115.304660/-/DC1.
- Nonstandard Abbreviations and Acronyms
- ADH
- autosomal-dominant hypercholesterolemia
- APOB
- apolipoprotein B
- FH
- familial hypercholesterolemia
- FH4
- ADH in absence of mutations in LDLR, APOB, or PCSK9
- LDL-C
- low-density lipoprotein cholesterol
- LDLR
- low-density lipoprotein receptor
- PCSK9
- proprotein convertase subtilisin/kexin 9
- STAP1
- signal transducing adaptor family member 1
- Received June 22, 2014.
- Revision received July 15, 2014.
- Accepted July 16, 2014.
- © 2014 American Heart Association, Inc.
References
Novelty and Significance
What Is Known?
Autosomal-dominant hypercholesterolemia (ADH) leads to an increase in total and low-density lipoprotein cholesterol and accelerated atherosclerotic disease.
Mutations in the genes encoding low-density lipoprotein receptor, apolipoprotein B, or proprotein convertase subtilisin/kexin 9 are established causes of ADH.
A substantial proportion of clinically diagnosed hypercholesterolemic patients do not carry mutations in the 3 established causal genes for ADH.
The identification of novel ADH genes is essential to identifying additional targets to lower low-density lipoprotein cholesterol to reduce coronary vascular disease risk.
What New Information Does This Article Contribute?
We identified signal transducing adaptor family member 1 (STAP1) as a novel gene associated with ADH.
This finding reveals a previously unrecognized signaling pathway that contributes to the pathogenesis of ADH.
Identification of STAP1 as an ADH gene might expand current genetic strategies for the molecular diagnosis of ADH patients and lead to new therapeutic targets for ADH.
The prevalence of ADH is ≈1:200 individuals in most Western countries. The diagnosis is usually made on the basis of clinical symptoms. However, because these symptoms develop late in life, establishing the diagnosis in younger patients is often difficult. Genetic analysis, that is, the demonstration of a causative mutation, provides unequivocal diagnosis. A proportion of the patients with a clinical diagnosis of ADH does not carry mutations in the established ADH genes but might carry mutations in genes that have not been associated previously with the hypercholesterolemic phenotype. Here, we identify STAP1 as a novel ADH gene, a finding of prognostic value that could also yield new insight into the pathogenesis of ADH.
This Issue
Jump to
Article Tools
Share this Article
- Mutations in STAP1 Are Associated With Autosomal Dominant HypercholesterolemiaNovelty and Significance
