| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Unidad de Lípidos and Laboratorio de Investigación Molecular (A.L.G.-O., D.R., A.C., F.C.), Hospital Universitario Miguel Servet, Instituto Aragonés de Ciencias de la Salud, 50009 Zaragoza, Spain; Unidad de Lípidos (M.C., M.J., E.R.), Servicio de Endocrinología y Nutrición, Institut dInvestigacions Biomediques August Pi i Sunyer, Hospital Clínic, Barcelona and Ciber CB06/03 Fisiopatología Obesidad y Nutrición, Instituto de Salud Carlos III, 08036 Barcelona, Spain; and Departamento de Bioquímica (M.P.), Biología Molecular y Celular, Universidad de Zaragoza, 50009 Zaragoza, Spain
Address all correspondence and requests for reprints to: Angel-Luis García-Otín, Ph.D., Laboratorio de Investigación Molecular, Hospital Universitario Miguel Servet, Instituto Aragonés de Ciencias de la Salud (I+CS), P° Isabel la Católica, 1-3, 50009 Zaragoza, Spain. E-mail: algarcia.iacs{at}aragon.es.
| Abstract |
|---|
|
|
|---|
Objective: Our objective was to test the hypothesis that increased intestinal cholesterol absorption might play a role in the lipid abnormalities of subjects with ADH without identified genetic defects.
Design and Setting: This is a cross-sectional study of consecutive subjects with primary hyperlipidemia identified during an 18-month period in two lipid clinics.
Study Subjects: A total of 52 subjects with a clinical diagnosis of ADH were examined for molecular defects in LDLR and APOB. No APOB defects were found. Functional LDLR mutations occurred in 31 (60%) subjects, who received a diagnosis of familial hypercholesterolemia (FH). Those for whom no mutations could be identified were labeled as non-FH ADH. In addition, 38 subjects with familial combined hyperlipidemia (FCH) and 45 normolipidemic control subjects were studied.
Interventions: Interventions were diagnostic.
Main Outcome Measures: Serum noncholesterol sterols were used as markers for the efficiency of intestinal cholesterol absorption.
Results: Adjusted campesterol to cholesterol ratios increased in the order non-FH ADH more than FH more than controls more than FCH, with mean values (95% confidence interval) in 102 mmol/mol cholesterol of 505 (424–600), 397 (345–458), 335 (294–382), and 284 (247–328), respectively. Thus, cholesterol absorption was lowest in FCH and highest in non-FH ADH.
Conclusions: Increased intestinal cholesterol absorption may partially explain the high cholesterol levels of non-FH ADH subjects. Serum noncholesterol sterols are a useful tool for the differential diagnosis of genetic hypercholesterolemias, especially FCH and ADH unrelated to LDLR or APOB defects.
| Introduction |
|---|
|
|
|---|
Heterozygosity for functional mutations of LDLR is by far the most frequent and best characterized cause of ADH (4), but no LDLR or APOB defects are detected in up to 40% of patients with a clinical diagnosis of definite FH submitted to molecular testing (5, 6, 7). When performed, screening for PCSK9 mutations did not improve the proportion of ADH subjects with a molecular diagnosis (5, 8), which suggests that the genetic background of ADH is complex and not limited to these three genes.
Hypercholesterolemia may occur from the deregulation of one or more pathways in whole-body cholesterol homeostasis. Intestinal cholesterol hyperabsorption leading to increased uptake of exogenous cholesterol by the liver, a compensatory reduction in both hepatic cholesterol synthesis and LDLR expression, and an ensuing reduction in low-density lipoprotein (LDL) clearance, is a potential cause of hypercholesterolemia (9). We hypothesized that some ADH subjects with no known genetic defect in classical candidate loci would have increased intestinal cholesterol absorption as an important contributor to elevated serum cholesterol levels.
Normal serum contains small but detectable amounts of noncholesterol sterols, including plant sterols and cholesterol precursors, and their ratios to cholesterol are accepted as surrogate markers for the efficiency of cholesterol absorption and cholesterol synthesis, respectively (10). Therefore, we determined serum noncholesterol sterols in normolipidemic control subjects and in well-phenotyped patients with familial dyslipidemias, including ADH with and without known genetic defects and FCH. The results suggest that intestinal cholesterol hyperabsorption plays a causal role in ADH with no LDLR or APOB defects.
| Subjects and Methods |
|---|
|
|
|---|
Consecutive, unrelated subjects older than 18 yr of age attending two Lipid Clinics in Spain (Hospital Universitario Miguel Servet in Zaragoza, and Hospital Clinic in Barcelona) with clinical diagnoses of ADH (n = 52) or FCH (n = 38) were recruited from January 2004 to June 2005. Secondary hyperlipidemia was excluded by standard methods. A diagnosis of ADH was made when subjects had off-treatment LDL-cholesterol levels above the age- and sex-specific 95th percentile of a Spanish reference population (11), triglyceride (TG) levels less than 200 mg/dl, and at least two first-degree relatives with a similar lipid phenotype. FCH was diagnosed when both total cholesterol and TG levels were above the 90th percentiles of the reference population, apolipoprotein B serum concentration was over 1.2 g/liter, and at least two first-degree relatives disclosed a similar lipid phenotype. Apolipoprotein E (APOE) genotyping in subjects with an FCH phenotype allowed exclusion of type III hyperlipoproteinemia. To avoid misclassification of patients with a clinical diagnosis of FCH, two specific test were performed. First, Achilles tendon sonograms were obtained in all FCH patients following a standard protocol (12), and assessed for both thickness and abnormal echostructure diagnostic of xanthoma eventually undetected by physical examination. Second, FCH patients with LDL cholesterol more than 250 mg/dl (n = 9) were submitted to DNA testing for identification of mutations in the LDLR and APOB genes (see Genetic analyses). All patients in the study had been advised to follow a lipid-lowering diet according to the Adult Treatment Panel III guidelines (13). In addition, we examined a control group of 45 normolipidemic subjects recruited from family physicians lists and hospital staff. All participants gave written informed consent to a protocol approved by the local review boards.
Laboratory methods
To obtain a baseline lipid profile, overnight fasting blood was drawn after at least 4 wk without hypolipidemic drug treatment. Cholesterol and TGs were determined by standard enzymatic methods. High-density lipoprotein (HDL)-cholesterol was quantified after precipitation with phosphotungstic acid and magnesium chloride. LDL-cholesterol was estimated with the Friedewald equation except in subjects with TGs more than 300 mg/dl, when it was measured by ultracentrifugation techniques, as described (14). Apolipoproteins A1 and B and lipoprotein(a) were determined using turbidimetric techniques.
Serum noncholesterol sterol concentrations were analyzed by gas chromatography using a modification of a method previously described by Heinemann et al. (15). Briefly, epi-coprostanol (2 µg) was added to serum (0.1 ml) as internal standard. After alkaline hydrolysis, extraction, and derivatization to trimethylsilyl ethers, the sterols were quantified by gas chromatography on a 30-m nonpolar capillary column (TRB-Esterol; Teknokroma, Barcelona, Spain) with a Perkin-Elmer GC Autosystem (Perkin-Elmer, Norwalk, CT). Each run quantified lanosterol, lathosterol, campesterol, sitosterol, and stigmasterol. Noncholesterol sterols are expressed as ratios to cholesterol (102 mmol/mol cholesterol).
Genetic analyses
DNA was isolated from EDTA blood samples following standard protocols. APOE genotyping was performed in all study subjects as previously described (16). The DNA of individuals with a clinical diagnosis of ADH was screened for LDLR and APOB mutations using Lipochip version 4.0 (Progenika Biopharma S. A., Bilboa, Spain), which is a microarray designed to detect the 204 more prevalent LDLR mutations in the Spanish population (7). When no mutations were detected, the LDLR gene coding sequences, exon-intron boundaries, and short proximal intronic sequences were sequenced to search for mutations not included in the microarray design. Large rearrangements in the LDLR gene were analyzed using a method based on quantitative fluorescent multiplex PCR (17). The presence of mutations within the putative receptor-binding region of the APOB gene was screened as described (18).
Statistical analysis
Comparison of lipid variables between non-FH ADH and FH groups was performed using the Students t test for data normally distributed and Mann-Whitney U test for skewed data. Noncholesterol sterol to cholesterol ratios were log transformed to achieve variance homogeneity. To detect group differences in noncholesterol sterol ratios, an analysis of covariance was performed considering log-transformed values as dependent variables, and sex, age, and body mass index (BMI) as confounding variables (covariates). The homogeneity of regression coefficients was verified for each covariate. When the omnibus test gave a significant result, the Tukeys least-significant difference post hoc test was used to detect significant differences in pairwise group comparisons. APOE genotype influence in noncholesterol sterol ratios was specifically investigated with similar statistical tests. Analyses were performed with SPSS software (version 11.0; SPSS, Inc., Chicago, IL), with statistical significance set at P < 0.05.
| Results |
|---|
|
|
|---|
Table 1
shows the characteristics of the study groups. Women predominated among ADH subjects, whereas the converse occurred in the FCH and control groups. Age and BMI were similar among groups, except that FCH subjects were slightly older than FH and control subjects and weighed more than subjects in other groups. Tendon xanthomas were present more frequently in FH subjects than in those with non-FH ADH. FH subjects also showed a more severe lipid profile than those with non-FH ADH. This was due in part because women in this group had a higher HDL-cholesterol level.
|
Specific comparison between the non-FH ADH and FH groups showed differences for total cholesterol and LDL-cholesterol levels, which were higher in FH. HDL-cholesterol and TG levels were significantly different only in women because women in the FH group had higher TG and lower HDL-cholesterol concentrations.
The serum sex-, age-, BMI-, and APOE genotype-adjusted noncholesterol sterol to cholesterol ratios in the study groups are shown in Table 2
. Lanosterol and lathosterol ratios are markers of cholesterol synthesis rate, whereas campesterol, sitosterol, and stigmasterol ratios correlate positively with intestinal cholesterol absorption efficiency. No differences in sterol ratios were observed between genders. However, there were important differences among study groups. Non-FH ADH subjects showed the highest campesterol, sitosterol, and stigmasterol ratios, and the lowest lathosterol ratios of all study groups. No differences between hyperlipidemic groups were observed for the lanosterol ratio, but all of them had a decreased ratio compared with the control group (P < 0.05). Conversely, FCH subjects disclosed the lowest campesterol and sitosterol ratios, whereas their lathosterol ratios were similar to control values. Furthermore, the noncholesterol sterol to cholesterol ratios of these patients were similar between FCH patients with higher and lower serum cholesterol concentrations (data not shown), indicating that this group was homogeneous regarding cholesterol absorption and synthesis markers.
|
Dyslipidemic subjects were categorized into tertiles taking as reference the control group values. Figure 1
shows the subjects tertile distribution according to major noncholesterol sterol ratios in the three hypercholesterolemic groups. Fully two thirds of non-FH ADH subjects ranked both in the lowest tertile of cholesterol synthesis and the highest tertile of cholesterol absorption. The distribution of FCH subjects showed the converse pattern, whereas FH subjects were evenly distributed among tertiles.
|
| Discussion |
|---|
|
|
|---|
A limitation of our study is that we did not investigate PCSK9 mutations as a potential cause of ADH. However, recent reports indicate that the frequency of PCSK9 defects is less than 2% in non-FH ADH (5, 8, 19), which makes improbable that any such cases were present in our series. On the other hand, our study has the strengths of a cross-sectional design in consecutive subjects with familial dyslipidemia defined by strict clinical criteria that underwent thorough molecular testing for common defects that cause an ADH phenotype.
A critical point to support our interpretation of the data is the accuracy of the diagnosis of FH. It should be noted that if some of our non-FH ADH patients were actually FH patients with an undetected LDLR mutation, the values of cholesterol absorption markers would increase globally in the FH group, and the between-group differences would be blunted. The FH diagnostic accuracy relies on a powerful methodology to detect LDLR mutations. We used the Lipochip microarray as a first screening tool for the detection of common mutations, followed by LDLR sequencing and analysis of large rearrangements when the first test was negative, the whole procedure thus guaranteeing a very high detection rate of causal mutations. Reports from other authors (20) show that FH-causing LDLR mutations are located in intronic positions with a nonnegligible frequency, which would not have been detected using LDLR analysis strategies different from that actually used (7). Because 56 of the 57 LDLR intronic mutations described in the public Human Gene Mutation Database (http://www.hgmd.cf.ac.uk) can be easily detected by our technique (only the IVS4 + 25 C more than T mutation would be skipped), we are confident that the patients classification into FH and non-FH ADH as described is correct with a very small margin for error.
The criteria that we used to define FCH were based on total cholesterol and TG levels above the 95th percentiles of reference values, but some FCH subjects disclosed LDL-cholesterol levels more than 260 mg/dl, which is a proposed threshold for the clinical diagnosis of FH, regardless of TG levels. Because no LDLR gene defects were detected in these subjects, they could also be considered as belonging to the non-FH ADH group. However, even if their lipid phenotype was ambiguous, the noncholesterol sterols to cholesterol ratios indicated high-cholesterol synthesis and low absorption, and were similar to those of the overall FCH group, supporting the validity of our classification.
In our study, subjects with non-FH ADH had milder hypercholesterolemia, lower TGs, and higher HDL-cholesterol levels than FH subjects. In contrast with FCH, they also showed normal TGs, normal to high HDL cholesterol, and a low BMI. Moreover, compared with FH, FCH, and normal control subjects, non-FH ADH subjects had higher serum campesterol to cholesterol and sitosterol to cholesterol ratios, which are validated markers of intestinal sterol absorption efficiency (10). This would suggest that high-cholesterol concentrations in ADH subjects without common genetic defects would be due, at least in part, to increased intestinal cholesterol absorption and increased cholesterol delivery to the liver via chylomicron remnants, with ensuing down-regulation of hepatic LDLR and attendant decrease in LDL uptake (9). It is well known that intestinal cholesterol absorption has a high interindividual variability (21). Bosner et al. (22) used stable isotopic tracers to study cholesterol absorption in healthy men and women, and reported absorption ranges from 29–80%. The variation in intestinal sterol absorption is genetically determined (23), with a reported heritability of 80% for campesterol and 73% for sitosterol levels in a Dutch twin study (24). Such between-subject variability in sterol absorption has been implicated in the variance of total cholesterol and LDL-cholesterol concentrations in the Finnish population (25) and Dutch families (26). However, no prior studies have suggested that cholesterol hyperabsorption could be a major cause of hypercholesterolemia in the population.
Gylling and Miettinen (23) studied noncholesterol sterols in patients with CHD and in siblings of selected probands with the lowest and highest sterol absorption ratios. They reported that high absorbers, both probands and siblings, had a lower BMI, lower TGs, and higher HDL cholesterol than low absorbers. Similar characteristics were present in the non-FH ADH individuals reported here.
The sterol ratios of intestinal absorption and hepatic synthesis were negatively related to each other in the 4S study (27), and low absorbers/high synthesizers usually had insulin resistance and other components of the metabolic syndrome (28). In our study we also found a low-absorption/high-synthesis pattern in FCH subjects, which is concordant with the proposed mechanism of FCH and its close relationship with insulin resistance (29). Moreover, our finding of enhanced cholesterol synthesis rate in FCH subjects agrees with previous reports in mixed hyperlipidemia patients by using serum cholesterol precursor analysis (30).
FCH can be difficult to differentiate from FH on clinical grounds because it is also associated with elevated LDL-cholesterol concentrations and a high risk of early-onset CHD. In addition, there is no unequivocal diagnostic test or group of clinical criteria for FCH (31). Therefore, it has been speculated that some subjects with a clinical diagnosis of ADH could in fact be FCH bearers, especially in studies in which the diagnosis of genetic hypercholesterolemia was based on clinical features, and family history of early-onset CHD was an inclusion criterion (32). Our results support the use of serum noncholesterol sterols as markers of FCH that could help in the differential diagnosis with other genetic dyslipidemias, such as FH subjects showing high TG concentrations.
Important advances in the understanding of the mechanisms for sterol absorption in humans have recently occurred (33). The ABCG5/ABCG8 heterodimer plays a crucial role in the transport of sterols from the intestinal mucosa and hepatocytes to the intestinal lumen and bile, respectively (34, 35), and mutations in these two loci are responsible for the rare disease, sitosterolemia. More recently, the Niemann-Pick C1-Like 1 (NPC1L1) protein has been identified as a specific sterol transporter in the brush border of the small intestine (36), and it has been demonstrated to be the target of ezetimibe, a specific inhibitor of cholesterol absorption. Furthermore, Cohen et al. (37) have recently reported that rare nonsynonymous NPC1L1allele variants are collectively associated with efficiency of cholesterol absorption and LDL-cholesterol levels, especially in African-American subjects, for whom a low absorber phenotype with low LDL cholesterol is associated with an increased cumulative frequency of nonsynonymous sequence variants of NPC1L1, potentially leading to only mildly deleterious effects on protein function but enough to attain serum LDL-cholesterol decreases of approximately 10%.
These findings provide new insights into the genetic control of sterol absorption. In fact, variation in these genes has been associated with sterol absorption, insulin resistance, and LDL-cholesterol levels (38), and with the cholesterol-lowering effect of ezetimibe (39). If these loci or other unknown genes related to sterol absorption are implicated in non-FH ADH needs to be determined. Nevertheless, intestinal cholesterol absorption is a complex process (40), and recent quantitative trait loci detection studies in mice suggest that it can be controlled by multiple loci (41).
In summary, high serum plant sterol to cholesterol ratios, which are good markers of intestinal sterol absorption, characterize subjects with a clinical diagnosis of ADH and no defects in the LDLR or APOB genes. This suggests that intestinal cholesterol hyperabsorption could explain in part the high serum cholesterol levels of some individuals or families with ADH. Furthermore, serum noncholesterol sterols appear to be a useful tool for the differential diagnosis of genetic hypercholesterolemias, especially FCH and ADH.
| Acknowledgments |
|---|
| Footnotes |
|---|
Disclosure Statement: The authors have nothing to disclose.
First Published Online June 12, 2007
Abbreviations: ADH, Autosomal dominant hypercholesterolemia; APOB, apolipoprotein B-100; APOE, apolipoprotein E; BMI, body mass index; CHD, coronary heart disease; FCH, familial combined hyperlipidemia; FDB, familial defective APOB; FH, familial hypercholesterolemia; HDL, high-density lipoprotein; LDL, low-density lipoprotein; LDLR, low-density lipoprotein receptor; NPC1L1, Niemann-Pick C1-Like 1; PCSK9, protein convertase subtilisin/kexin type 9; TG, triglyceride.
Received November 22, 2006.
Accepted June 5, 2007.
| References |
|---|
|
|
|---|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |