J. Lipid Res. Neurobiology of Lipids (ISSN1683-5506)
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Originally published In Press as doi:10.1194/jlr.M700078-JLR200 on September 23, 2007

Papers In Press, published online ahead of print December 1, 2007
J. Lipid Res., doi:10.1194/jlr.M700078-JLR200
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Supplemental Data
Right arrow All Versions of this Article:
M700078-JLR200v1
48/12/2632    most recent
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Sherva, R.
Right arrow Articles by Neuman, R. J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Sherva, R.
Right arrow Articles by Neuman, R. J.
Social Bookmarking
Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?

Journal of Lipid Research, Vol. 48, 2632-2639, December 2007
Copyright © 2007 by American Society for Biochemistry and Molecular Biology

Evidence for a quantitative trait locus affecting low levels of apolipoprotein B and low density lipoprotein on chromosome 10 in Caucasian familiesboxs

Richard Sherva, Pin Yue, Gustav Schonfeld and Rosalind J. Neuman1

Washington University School of Medicine, St. Louis, MO 63110

boxs The online version of this article (available at http://www.jlr.org) contains supplementary data in the form of one table. Back

Published, JLR Papers in Press, September 23, 2007.

1 To whom correspondence should be addressed. e-mail: roz{at}psychiatry.wustl.edu


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
High plasma apolipoprotein B (apoB) and LDL cholesterol levels increase cardiovascular disease risk. These highly correlated measures may be partially controlled by common genetic polymorphisms. To identify chromosomal regions that contain genes causing low plasma levels of one or both parameters in Caucasian families ascertained for familial hypobetalipoproteinemia (FHBL), we conducted a whole-genome scan using 443 microsatellite markers typed in nine multigenerational families with at least two members with FHBL. Both variance components and regression-based linkage methods were used to identify regions of interest. Common linkage regions were identified for both measures on chromosomes 10q25.1-10q26.11 [maximum log of the odds (LOD) = 4.2 for LDL and 3.5 for apoB] and 6q24.3 (maximum LOD = 1.46 for LDL and 1.84 for apoB). There was also evidence for linkage to apoB on chromosome 13q13.2 (LOD = 1.97) and to LDL on chromosome 3p14.1 at 94 centimorgan (LOD = 1.52). Bivariate linkage analysis provided further evidence for loci contributing to both traits (6q24.3, LOD = 1.43; 10q25.1, LOD = 1.74). We evaluated single nucleotide polymorphisms (SNPs) in genes within our linkage regions to identify variants associated with apoB or LDL levels. The most significant finding was for rs2277205 in the 5' untranslated region of acyl-coenzyme A dehydrogenase short/branched chain and LDL (P = 10–7). Three additional SNPs were associated with apoB and/or LDL (P < 0.01). Although only the linkage signal on chromosome 10 reached genome-wide statistical significance, there are likely multiple chromosomal regions with variants that contribute to low levels of apoB and LDL and that may protect against coronary heart disease.

Supplementary key words genetics • cardiovascular disease • lipids • genome-wide linkage analysis

Abbreviations: ACADSB, acyl-coenzyme A dehydrogenase short/branched chain; apoB, apolipoprotein B; BMI, body mass index; cM, centimorgan; FHBL, familial hypobetalipoproteinemia; FOXP1, forkhead box protein P1; h2, narrow sense heritabilities; LOD, log of the odds; PCSK9, proprotein convertase subtilisin/kexin type 9; PNLIP, pancreatic triacylglycerol lipase precursor; PNLIPRP, pancreatic lipase-related protein precursor; PPAPDC1A, phosphatidic acid phosphatase type 2 domain-containing 1A; SCD, stearoyl-coenzyme A desaturase; SLC25A26, solute carrier family 25 member 26; SNP, single nucleotide polymorphism; UTR, untranslated region


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
High plasma apolipoprotein B (apoB) and LDL cholesterol concentrations are equally and independently associated with an increased long-term relative risk of coronary heart disease (1, 2). Patients in the highest tertile of apoB and LDL have an ~2-fold increase in coronary heart disease risk over those in the lower tertiles after multivariate adjustment. Other studies suggest that increased apoB levels may be associated with an even greater risk of coronary heart disease than high LDL cholesterol concentrations (3, 4).

Data from twin studies show that there is a strong heritability for apoB and LDL levels (5, 6), and numerous genome scans have attempted to identify loci linked to lipid levels. A recent summary of the results from various lipid genome scans lists significant or suggestive evidence for linkage on chromosomes 1, 2, 3, 6, and 11 for apoB and on chromosomes 1, 3, 4, 5, 6, 10, 11, 13, 15, 17, 19, and X for LDL (7). In addition, linkage scans for loci influencing both traits have identified regions on chromosomes 2, 11, and 18 (7).

In this study, we performed genome-wide linkage analysis in a cohort of Caucasian families ascertained for familial hypobetalipoproteinemia (FHBL). FHBL is a disorder in which subjects have extremely low levels of apoB and LDL, generally in the lower sex- and age-adjusted 5% of the population (8), a condition that may provide a possible protective advantage against coronary problems. Mutations in the apoB gene (APOB) on chromosome 2p23-24 have been shown to produce truncations in the apoB protein, resulting in low apoB levels, but only in a minority of FHBL families (9). Previously, we reported strong linkage signals to a region of chromosome 3p21 in several FHBL families without the truncated form of the apoB protein (10). We have also shown that these families differ significantly in fatty liver content from FHBL families, with truncation producing mutations in the apoB gene (11). The FHBL families analyzed here do not have truncated forms of the apoB protein, nor do they show linkage to chromosome 3p21. We followed up our linkage analyses by testing for an association between apoB and LDL levels and single nucleotide polymorphisms (SNPs) in potential susceptibility genes in the linked regions.


   METHODS
TOP
 ABSTRACT
 INTRODUCTION
 METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Recruitment
Potential FHBL families were identified by screening volunteers for total plasma cholesterol levels < 150 mg/dl at various locations in the metropolitan area of St. Louis, MO, including shopping malls, the Red Cross blood bank, and our Lipid Research Clinic. Probands provided information on their health status and on the structure of their families. Families were invited to participate in our study based on the presence of two or more affected individuals in two or more generations, the absence of truncated forms of apoB in plasma on immunoblotting (12), the absence of diabetes mellitus, thyroid, liver, and kidney diseases, and a willingness to participate. None of the participants was taking medications known to affect lipid metabolism. The Human Studies Committee of the Washington University Medical Center approved all protocols, and informed consent was obtained from all participants.

Lipid analyses, DNA isolation, and genotyping
Methods for the determination of lipid levels and blood extraction have been described (13). Microsatellite genotyping was performed by the Mammalian Genotyping Service (Marshfield, WI) (14) using Screening Set 14 (http://research.marshfieldclinic.org/genetics/home/index.asp). SNP genotyping was conducted at the Genotyping Core in the Department of Human Genetics, Washington University School of Medicine, using the high-performance Sequenom MassARRAY system. Briefly, the technology involves PCR amplification of the region containing the SNP of interest, an optimized primer extension reaction to generate allele-specific DNA products, and chip-based mass spectrometry for the separation and analysis of the DNA. Primers of selected SNPs were designed to fit them into four multiplexes of 384-well format. The overall successful rate of this genotyping project was 96%. All pedigrees and genotypes were checked for errors with PEDCHECK (15), PREST (16), and MERLIN-ERROR (17), and unlikely genotypes were set to missing using the PEDWIPE feature in MERLIN.

Statistical methods
We used two fundamentally different methods of quantitative linkage analysis, a recently introduced regression model, MERLIN-REGRESS implemented in the MERLIN software package (17), and a variance components model implemented in SOLAR version 2.0.3 (18). The regression-based linkage analysis uses trait-squared sums and differences to predict identity by descent allele sharing between relative pairs in pedigrees. This reversal of the independent and dependent variables used in Haseman-Elston methodologies for analyzing quantitative traits in general pedigrees (19) has the benefits of robustness for nonnormality of quantitative traits and misspecification of the required input parameters, the population trait's mean, variance, and heritability (17). We estimated the means and variances of apoB and LDL from the founders in families in the current study plus additional FHBL families used in our previous analyses (10, 13, 20) (n = 58). Heritability for this regression method was estimated using SOLAR. Because variance components analyses require normal trait distributions, the phenotypes were defined to be the square root of the subject's serum apoB or LDL level. Age, gender, and body mass index (BMI) were included as covariates. Because LDL and apoB levels are highly correlated (21), a bivariate model was also specified in SOLAR to test for loci jointly contributing to the variance of apoB and LDL. Bivariate models estimate variance parameters and regression parameters for each individual trait as in univariate models but also include a parameter estimating the linkage correlations between two traits. This can increase the power to detect loci with pleiotropic effects (22). Five thousand replicates of our kindreds were simulated for the regions showing the most evidence for linkage under the hypothesis of no linkage between trait and marker. Empirical log of the odds (LOD) score distributions and P values were computed for both the regression-based and variance components genome scans using the simulated data.

To narrow the linkage regions, we chose one tag SNP for genotyping (using an r2 threshold of 0.8) in coding regions or untranslated regions (UTRs) of purported candidate genes located within a 1 LOD score interval of our peak linkage regions identified on chromosomes 3, 6, 10, and 13 (see Table 2 below). For a more even distribution, a few additional tag SNPs were selected on genes whose function was unknown. Although we preferred common SNPs, some SNPs with low minor allele frequencies (<=0.05) were retained because of the possibility that rare variants may prove to be important for our quantitative traits (23). We tested for an association of these SNPs with apoB and LDL using linear regression models with covariance corrected for correlation among relatives (the GENMOD procedure in SAS version 9.1). To increase our power, three additional families in which linkage to the APOB gene had been ruled out were included in this analysis. To reduce the skewness in the distributions, measured apoB and LDL levels were log-transformed. Genetic effects were modeled as the additive effect of the number of minor alleles, and age, gender, and BMI were included as covariates. Additive coding has been shown to be robust for the detection of nonadditive effects when the minor allele frequency is >20% (24).


View this table:
[in this window]
[in a new window]

  		TABLE 2. Chromosomal regions suggesting linkage to apoB, LDL, or both

 

   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
One hundred samples from 10 families were genotyped by the Mammalian Genotyping Services using Screening Set 14. However, one nine member family had severe Mendelian errors that may have resulted from mixing of blood samples and could not be corrected. Therefore, this family was dropped from further analysis, leaving 91 individuals in nine pedigrees for genetic analysis. Table 1 displays the characteristics of the nine families on variables of interest. The average apoB and LDL levels in the affected pedigree members were ~40% and 60% lower, respectively, than the average levels in those subjects considered normal (8). Nonparametric Kruskal-Wallis tests showed significant differences in the means of apoB (P < 0.0001), LDL (P < 0.0001), and BMI (P < 0.0001) between FHBL and normal individuals, whereas the difference in mean age between these groups was marginally statistically significant (P = 0.06).


View this table:
[in this window]
[in a new window]

  		TABLE 1. Baseline characteristics of study participants by family and affection status (n = 94)

 
Regression-based analysis
The mean serum apoB and LDL levels obtained within the founders in our data, 104 and 126 mg/dl, respectively, correspond very well with National Health and Nutrition Examination Survey III survey estimates of 99 and 123 mg/dl, respectively (21, 25). Because population variances were not reported for the National Health and Nutrition Examination Survey III data and the mean trait levels so closely matched our estimates, we opted to use the internally calculated means and variances for analysis. Subsequent sensitivity analyses (see Discussion for details) led us to believe that the location and magnitude of our linkage peaks would not be significantly affected by such small differences between estimates. In the regression-based analyses, the highest multipoint LOD score for apoB was located at 140 centimorgan (cM) on chromosome 10 (LOD = 3.50; P = 0.002). Likewise, the highest LOD score for LDL was at the same location on chromosome 10 (LOD = 4.22; P = 0.0006). The next highest LOD score for apoB occurred on chromosome 13q13.2 (LOD = 1.20; P =0.009). In Figs. 1A , 2A , and 3A , we plotted the multipoint LOD scores from the regression-based genome scans for apoB and LDL on chromosomes 10, 13, and 6, respectively. Based on the change in the whole-sample LOD scores observed when families were excluded from analysis one at a time, we observed strong evidence for linkage on chromosome 10 in three of the families (families 1, 3, and 9 in Table 1), suggesting the presence of genetic heterogeneity in the sample. All of the families had positive LOD scores at this location, however.


Figure 1
Figure 1
View larger version (18K):
[in this window]
[in a new window]

  		Fig. 1. A: Results of univariate multipoint regression-based linkage analyses for serum apolipoprotein B (apoB) and LDL on chromosome 10. B: Results of univariate and bivariate multipoint variance components linkage analyses for serum apoB and LDL on chromosome 10. cM, centimorgan; LOD, log of the odds.

 

Figure 2
Figure 2
View larger version (16K):
[in this window]
[in a new window]

  		Fig. 2. A: Results of univariate multipoint regression-based linkage analyses for serum apoB and LDL on chromosome 13. B: Results of univariate and bivariate multipoint variance components linkage analyses for serum apoB and LDL on chromosome 13.

 

Figure 3
Figure 3
View larger version (16K):
[in this window]
[in a new window]

  		Fig. 3. A: Results of univariate multipoint regression-based linkage analyses for serum apoB and LDL on chromosome 6. B: Results of univariate and bivariate multipoint variance components linkage analyses for serum apoB and LDL on chromosome 6.

 
Variance components analysis
SOLAR variance components genome scans for apoB and LDL were adjusted for gender, age, and BMI. Gender was not a significant predictor of either apoB (P = 0.48) or LDL (P = 0.11). Age was a significant predictor of apoB (P = 0.02) but not for LDL (P = 0.11), whereas BMI was significant for both traits (P = 0.0002 and 0.003, respectively). SOLAR estimated the narrow sense heritabilities (h2) to be 61% for apoB and 21% for LDL. The maximum multipoint LOD scores for LDL and apoB were on chromosome 10 at 128 cM (LOD = 2.20; P = 0.0002) and 137 cM (LOD = 1.92; P = 0.002), respectively. The highest LOD score for apoB on chromosome 13 occurred at the same location identified by MERLIN-REGRESS, 13q13.2 at 26 cM (LOD = 1.97; P =0.001). Figures 1B, 2B, and 3B display the chromosome-wide LOD scores for apoB, LDL, and the bivariate variance components genome scans on chromosomes 10, 13, and 6, respectively. Other LOD scores > 1.5 were observed for LDL on chromosome 3p14.1 (LOD = 1.52; P = 0.0009) and for LDL and apoB at 146 cM on chromosome 6q24.3 for both traits (LOD = 1.84 and 1.46, respectively).

The maximum LOD scores in the apoB-LDL bivariate analyses occurred on chromosomes 6q24.3 (LOD = 1.43; P = 0.01) and 10q25.1 (LOD = 1.74; P = 0.004) at the same locations as the signals observed for each individual trait. Table 2 summarizes the locations and significance of the highest LOD scores from regression and variance components analyses.

Candidate gene association analysis
We tested for genetic association between 49 SNPs on chromosomes 3, 6, 10, and 13 in potential candidate genes (see supplementary Table I) and the log of the individuals' measured serum apoB and LDL. Our sample size increased to 153 genotyped subjects when we added the three additional families. Table 3 displays the most significant (unadjusted for multiple testing) association findings by gene and chromosome, along with the direction of the effect on apoB and LDL and the location and type of the polymorphisms. Table 3 shows that the most significant association was between rs2277250 and LDL in the gene acyl-coenzyme A dehydrogenase short/branched chain (ACADSB) on chromosome 10 (Z = 5.33; P = 0.000001, unadjusted for multiple testing). This finding is still significant after using the Bonferroni correction for 49 tests. (Our SNPs are all in different genes, and apoB and LDL are very highly correlated traits.) There was an associated increase in apoB and LDL levels of 10% and 14%, respectively, for each minor allele carried at this locus. Overall, five SNPs were associated with LDL, and four were associated with apoB, at P < 0.05. Four of these were associated with both apoB and LDL.


View this table:
[in this window]
[in a new window]

  		TABLE 3. SNPs associated with apoB and/or LDL

 
We sequenced the coding regions of a promising (based on position and known functionality) candidate gene, stearoyl-coenzyme A desaturase (SCD; including 5' UTRs and 3' UTRs, exon/intron junctions, and all exons) using one affected individual from three of the original nine families and three unrelated controls. However, we did not find any variations that cosegregate with FHBL status in these families. The SCD SNP (rs2234970) that was typed in our total sample was not associated with apoB (P = 0.3) or LDL (P = 0.46).

Determining which SNPs influence linkage signals
In an attempt to determine whether the most significantly trait-associated SNPs were responsible for the initially observed linkage signals, we repeated the regression-based scans of chromosome 10 on apoB and LDL after regressing out the effects of the most significant SNPs in the region (rs2277250 and rs2463147). The LOD score for LDL decreased from 4.22 to 0.58 at 140 cM after adjustment for rs2277250 and to 0.88 after adjustment for rs2463147. Likewise, the LOD score for apoB in the region decreased from 3.5 to 0.63 after adjustment for rs2277250 and to 0.9 after adjusting for rs2463147. The LOD score on chromosome 3 in the variance components analysis of LDL (LOD = 1.52) also decreased after adjustment for the most significant SNPs in the region (to 0.95 after adjusting for rs9819895 and to 0.58 after adjusting for rs2311298), indicating that each of the SNPs we identified accounts for a significant proportion of the allele sharing among phenotypically similar relatives in the linkage regions.


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
Our strongest linkage signals for both apoB and LDL were located on chromosome 10 using two distinctly different analytical methods. The maximum LOD scores for LDL from regression-based and variance components scans were located 12 cM apart at 140 cM (122.3 Mb) and 128 cM (107.1 Mb), respectively. The finding in this region replicates a previously reported linkage region for LDL (LOD = 2.5; P = 0.0001) found by Coon et al. (26) in a sample of 500 Caucasian families. Both methods of analysis produced significant LOD scores for apoB 3 cM apart at the same chromosome 10 location, 10q26.11. The peak on chromosome 6q24.3 (149 Mb) for LDL in our SOLAR analysis is in the same region as a previous LDL signal (LOD = 2.9; P = 0.00003) (27). To our knowledge, this is the first study to show linkage to apoB on chromosome 10, although a SNP in the region was significantly associated with several metabolic traits, including reduced serum apoB (28). Previous reports of linkage on chromosome 6 (7) are ~17 Mb upstream from our signal. The chromosome 13q13.2 (32.9 Mb) peak for apoB is in the same region as two previous linkage findings for low levels of serum HDL cholesterol (29, 30), and our peak on chromosome 6 is <20 cM downstream from a previously reported HDL3 linkage region (31).

Candidate genes
We expanded our search for potential variants associated with FHBL or low LDL levels by identifying SNPs in genes within our most significant linkage regions. Our approach may be considered a "discovery" phase because the function of some of these genes with respect to these lipid measures is unknown. However, several genes located near the peak on chromosome 10 have potential links to lipid metabolism. For example, pancreatic triacylglycerol lipase precursor (PNLIP), a 56 kDa protein secreted by the pancreas that is essential for the hydrolysis and absorption of long-chain triglyceride fatty acids in the intestine is located ~1 Mb upstream from the chromosome 10 peak (32). As mentioned previously, a SNP in PNLIP (1359 C/T) was significantly associated with decreased serum apoB, apoA-I, lipoprotein [a], and total cholesterol in 185 neonates of African descent (29). However, the polymorphism was not present in a sample of 100 Caucasians selected from the general population. Therefore, because pancreatic lipase-related protein 3 precursor (PNLIPRP3) and PNLIPRP1 interact with and flank PNLIP (33), we used our resources to sequence SNPs in those genes.

The fatty acid CoA ligase ACSL5, also located on 10q25.1, catalyzes the formation of acyl-CoA from fatty acid, ATP, and CoA, a reaction essential to mammalian fatty acid metabolism. ACSL5, in conjunction with fatty acid transporter protein, also mediates the cellular uptake of fatty acids (34). One SNP was typed in this gene but it was not significantly associated with apoB or LDL.

Another candidate on chromosome 10q24.31, SCD, catalyzes a rate-limiting step in the synthesis of unsaturated fatty acids. It has been shown in animal models to play an important role in cholesterol and lipoprotein homeostasis (35). It is thought to be an intermediary between leptin and obesity, as leptin represses RNA levels and the enzymatic activity of SCD. In another experiment, mice without Scd1 were lean and hypermetabolic, and in genetically obese (ob/ob) mice, those with mutations in Scd1 were significantly less obese than ob/ob mice without Scd1 mutations. The mice with mutations in Scd1 also had significantly reduced hepatic triglyceride storage and VLDL production (36).

Although no SNPs subsequently typed in the initially hypothesized candidate genes appear to be associated with apoB or LDL, several other SNPs in these and other regions may be. At the 0.05 level, three SNPs on chromosome 10 showed association with both traits: two located ~2 Mb apart in the genes ACADSB and phosphatidic acid phosphatase type 2 domain-containing 1A (PPAPDC1A). ACADSB is a mitochondrial enzyme that oxidizes straight-chain or branched-chain acyl-CoAs in the metabolism of fatty acids or branched-chain amino acids (37). In this sample, minor alleles of rs2277250 are more common in those with higher apoB and LDL, and the association is stronger for LDL. The SNP is located in the gene's 5' UTR and is a known tag SNP. The second significant SNP on chromosome 10, rs2463147, is in the gene PPAPDC1A, which is likely a membrane-associated phosphatidic acid phosphatase, based on domain homology (38).

Another candidate, solute carrier family 25 member 26 (SLC25A26), located under the peak on 3p14.1, is involved in the mitochondrial transport of S-adenosylmethionine, the methyl group donor for almost all biological methylation reactions, and is necessary for the biosynthesis of lipoic acid and ubiquinone, which are in turn cofactors in enzyme complexes necessary for aerobic metabolism (39). Finally, SNPs were chosen in forkhead box protein P1 (FOXP1) based on its proximity to the linkage peak on chromosome 3 and the fact that it is a transcription factor with significant sequence homology with the DNA binding domains of the hepatocyte nuclear factor-3/forkhead protein family (40) and is expressed at high levels in gastrointestinal tissue (41).

One of our goals was to better characterize the pathways causing FHBL besides the common truncation mutations in the apoB gene. One possibility is that these families harbor one of the several apoB mutations not detectable by immunoblotting. This seems unlikely, as we have the ability to detect truncated forms as small as apoB-30 and extensive fine-mapping on chromosome 2 reveals no evidence for linkage to the apoB region in any of these families. Another possible source of variation in apoB or LDL levels is variation in the apoE gene, as we have shown previously that {epsilon}4 alleles are associated with increased serum apoB and LDL (42). We did not type apoE variants in the current study, although it is unlikely that differences at this locus could fully account for the magnitude of the differences in apoB and LDL observed between FHBL-affected and unaffected family members. To adjust for apoE effects, we will likely genotype this locus in the future. Another gene of substantial interest for decreasing LDL levels is proprotein convertase subtilisin/kexin type 9 (PCSK9) on chromosome 1p32.3. Increasing interest in PCSK9 has produced several new human and animal studies of this gene. PCSK9 promotes the degradation of the LDL receptor in hepatocytes, apparently both intracellularly and by being a secreted protein that can bind the LDL receptor and be internalized. It has emerged as a promising therapeutic target (43).

The candidate genes with evidence for an association with apoB and LDL act in pathways related to the absorption of lipids in the gut (PNLIP), system-wide (PPAPDC1A) or mitochondrial (ACADSB and SLC25A26) lipid metabolism, or transcription factors active in the liver and gastrointestinal tract with unknown consequences (FOXP1). The observation that certain individuals with FHBL exhibit steatorrhea and diarrhea supports the hypothesis that a subset of FHBL cases are attributable to reduced intestinal lipid absorption. The lower BMI observed in FHBL cases also supports this hypothesis, as does the possibility that there is an overall increase in the rate at which these individuals metabolize lipids.

Sensitivity analysis
Because of the ascertainment of probands with FHBL, the mean apoB and LDL levels in this population were significantly lower than the means measured in the founders in the pedigrees and from previous population estimates. Although the LOD scores from regression-based methods are purported not to be sensitive to the input h2 and population means (19), we tested the effects of varying these estimates for both traits within a biologically plausible range. For both traits, reducing the h2 estimate by one-third (to 40% for apoB and 14% for LDL) decreased the maximum LOD score on chromosome 10. Increasing the h2 of apoB to 80% for apoB and 27% for LDL increased the magnitude of the chromosome 10 LOD score for LDL but decreased the LOD score for apoB. Increasing the input population mean LDL and apoB by 20 mg/dl reduced both peak LOD scores, and reducing the population mean by the same amount increased both LOD scores. Although the magnitude of the peak LOD scores on chromosome 10 changed after systematically changing the input population means, variances, or h2 estimates, the location of the peak LOD score on chromosome 10 did not change, indicating that the location of linkage peaks may be robust to misspecification of the model parameters.

Strengths and limitations
One limitation of this analysis is the lack of data on covariates known to affect apoB and LDL levels. Although some key factors were measured, potentially important lifestyle covariates such as physical activity, dietary composition, and smoking status were not. Adjustment for BMI may act as a surrogate for lifestyle factors and reduce the potential for confounding as a result of differences in diet and exercise.

The results presented are based on a relatively small sample size of nine three-generation pedigrees containing 91 genotyped subjects. Nevertheless, the smallest P values on chromosome 10 provide strong evidence for suggestive linkage after correction for multiple tests according to the guidelines of Lander and Kruglyak (44), increasing our confidence that the signal is real. In addition, the agreement between the results from linkage methods with totally different mathematical underpinnings also increases our confidence in the results. Furthermore, the fact that there was evidence for linkage to both apoB and LDL on chromosome 10 in both the regression and variance components methods, along with the corroborating results from the bivariate scan, increases the chance that we are observing a locus affecting both traits rather than an artifact of testing correlated traits. Perhaps the most intriguing aspect of the signal on chromosome 10 is its proximity to the SNP in PNLIP that showed an association with lower serum apoB in African neonates (28). If this SNP is present in these families, it may explain why they exhibit FHBL without the apoB truncation mutation.

Finally, the SNP genotyping conducted here was in no way comprehensive; instead, it was meant to narrow the search for candidate genes in linkage regions. Subsequent genotyping will provide better coverage of the linked regions and attempt to locate causal polymorphisms for FHBL in the genes we did identify.


   ACKNOWLEDGMENTS
 
This research was supported by National Institutes of Health Grant HL59515 and National Institutes of Health Training Grant T32 MH-014677. The authors thank the Minnesota Supercomputing Institute for use of their resources.

Manuscript received February 12, 2007 and in revised form May 30, 2007 and in re-revised form September 18, 2007.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
  1. St-Pierre, A. C., B. Cantin, G. R. Dagenais, J. P. Despres, and B. Lamarche. 2006. Apolipoprotein-B, low-density lipoprotein cholesterol, and the long-term risk of coronary heart disease in men. Am. J. Cardiol. 97: 997–1001.[CrossRef][Medline]

  2. Jiang, R., M. B. Schulze, T. Li, N. Rifai, M. J. Stampfer, E. B. Rimm, and F. B. Hu. 2004. Non-HDL cholesterol and apolipoprotein B predict cardiovascular disease events among men with type 2 diabetes. Diabetes Care. 27: 1991–1997.[Abstract/Free Full Text]

  3. Moss, A. J., R. E. Goldstein, V. J. Marder, C. E. Sparks, D. Oakes, H. Greenberg, H. J. Weiss, W. Zareba, M. W. Brown, C. S. Liang, et al. 1999. Thrombogenic factors and recurrent coronary events. Circulation. 99: 2517–2522.[Abstract/Free Full Text]

  4. Talmud, P. J., E. Hawe, G. J. Miller, and S. E. Humphries. 2002. Nonfasting apolipoprotein B and triglyceride levels as a useful predictor of coronary heart disease risk in middle-aged UK men. Arterioscler. Thromb. Vasc. Biol. 22: 1918–1923.[Abstract/Free Full Text]

  5. Lamon-Fava, S., D. Jimenez, J. C. Christian, R. R. Fabsitz, T. Reed, D. Carmelli, W. P. Castelli, J. M. Ordovas, P. W. Wilson, and E. J. Schaefer. 1991. The NHLBI Twin Study: heritability of apolipoprotein A-I, B, and low density lipoprotein subclasses and concordance for lipoprotein(a). Atherosclerosis. 91: 97–106.[CrossRef][Medline]

  6. Fenger, M., K. Schousboe, T. I. Sorensen, and K. O. Kyvik. 2006. Variance decomposition of apolipoproteins and lipids in Danish twins. Atherosclerosis. 191: 40–47.[Medline]

  7. Bosse, Y., Y. C. Chagnon, J. P. Despres, T. Rice, D. C. Rao, C. Bouchard, L. Perusse, and M. C. Vohl. 2004. Compendium of genome-wide scans of lipid-related phenotypes: adding a new genome-wide search of apolipoprotein levels. J. Lipid Res. 45: 2174–2184.[Abstract/Free Full Text]

  8. Contois, J. H., J. R. McNamara, C. J. Lammi-Keefe, P. W. Wilson, T. Massov, and E. J. Schaefer. 1996. Reference intervals for plasma apolipoprotein B determined with a standardized commercial immunoturbidimetric assay: results from the Framingham Offspring Study. Clin. Chem. 42: 515–523.[Abstract/Free Full Text]

  9. Wu, J., J. Kim, Q. Li, P. Y. Kwok, T. G. Cole, B. Cefalu, M. Averna, and G. Schonfeld. 1999. Known mutations of apoB account for only a small minority of hypobetalipoproteinemia. J. Lipid Res. 40: 955–959.[Abstract/Free Full Text]

  10. Yuan, B., R. Neuman, S. H. Duan, J. L. Weber, P. Y. Kwok, N. L. Saccone, J. S. Wu, K. Y. Liu, and G. Schonfeld. 2000. Linkage of a gene for familial hypobetalipoproteinemia to chromosome 3p21.1-22. Am. J. Hum. Genet. 66: 1699–1704.[CrossRef][Medline]

  11. Yue, P., T. Tanoli, O. Wilhelm, B. Patterson, D. Yablonskiy, and G. Schonfeld. 2005. Absence of fatty liver in familial hypobetalipoproteinemia linked to chromosome 3p21. Metabolism. 54: 682–688.[CrossRef][Medline]

  12. Aguilar-Salinas, C. A., P. H. Barrett, K. G. Parhofer, S. G. Young, D. Tessereau, J. Bateman, C. Quinn, and G. Schonfeld. 1995. Apoprotein B-100 production is decreased in subjects heterozygous for truncations of apoprotein B. Arterioscler. Thromb. Vasc. Biol. 15: 71–80.[Abstract/Free Full Text]

  13. Yue, P., B. Yuan, D. S. Gerhard, R. J. Neuman, W. L. Isley, W. S. Harris, and G. Schonfeld. 2002. Novel mutations of APOB cause apoB truncations undetectable in plasma and familial hypobetalipoproteinemia. Hum. Mutat. 20: 110–116.[CrossRef][Medline]

  14. Broman, K. W., J. C. Murray, V. C. Sheffield, R. L. White, and J. L. Weber. 1998. Comprehensive human genetic maps: individual and sex-specific variation in recombination. Am. J. Hum. Genet. 63: 861–869.[CrossRef][Medline]

  15. O'Connell, J. R., and D. E. Weeks. 1998. PedCheck: a program for identification of genotype incompatibilities in linkage analysis. Am. J. Hum. Genet. 63: 259–266.[CrossRef][Medline]

  16. McPeek, M. S., and L. Sun. 2000. Statistical tests for detection of misspecified relationships by use of genome-screen data. Am. J. Hum. Genet. 66: 1076–1094.[CrossRef][Medline]

  17. Abecasis, G. R., S. S. Cherny, W. O. Cookson, and L. R. Cardon. 2002. Merlin—rapid analysis of dense genetic maps using sparse gene flow trees. Nat. Genet. 30: 97–101.[CrossRef][Medline]

  18. Almasy, L., and J. Blangero. 1998. Multipoint quantitative-trait linkage analysis in general pedigrees. Am. J. Hum. Genet. 62: 1198–1211.[CrossRef][Medline]

  19. Sham, P. C., S. Purcell, S. S. Cherny, and G. R. Abecasis. 2002. Powerful regression-based quantitative-trait linkage analysis of general pedigrees. Am. J. Hum. Genet. 71: 238–253.[CrossRef][Medline]

  20. Pulai, J. I., R. J. Neuman, A. W. Groenewegen, J. Wu, and G. Schonfeld. 1998. Genetic heterogeneity in familial hypobetalipoproteinemia: linkage and non-linkage to the apoB gene in Caucasian families. Am. J. Med. Genet. 76: 79–86.[CrossRef][Medline]

  21. Bachorik, P. S., K. L. Lovejoy, M. D. Carroll, and C. L. Johnson. 1997. Apolipoprotein B and AI distributions in the United States, 1988–1991: results of the National Health and Nutrition Examination Survey III (NHANES III). Clin. Chem. 43: 2364–2378.[Abstract/Free Full Text]

  22. Todorov, A. A., G. P. Vogler, C. Gu, M. A. Province, Z. Li, A. C. Heath, and D. C. Rao. 1998. Testing causal hypotheses in multivariate linkage analysis of quantitative traits: general formulation and application to sibpair data. Genet. Epidemiol. 15: 263–278.[CrossRef][Medline]

  23. Blangero, J. 2004. Localization and identification of human quantitative trait loci: king harvest has surely come. Curr. Opin. Genet. Dev. 14: 233–240.[CrossRef][Medline]

  24. Lettre, G., C. Lange, and J. N. Hirschhorn. 2007. Genetic model testing and statistical power in population-based association studies of quantitative traits. Genet. Epidemiol. 31: 358–362.[CrossRef][Medline]

  25. Carroll, M. D., D. A. Lacher, P. D. Sorlie, J. I. Cleeman, D. J. Gordon, M. Wolz, S. M. Grundy, and C. L. Johnson. 2005. Trends in serum lipids and lipoproteins of adults, 1960–2002. J. Am. Med. Assoc. 294: 1773–1781.[Abstract/Free Full Text]

  26. Coon, H., J. H. Eckfeldt, M. F. Leppert, R. H. Myers, D. K. Arnett, G. Heiss, M. A. Province, and S. C. Hunt. 2002. A genome-wide screen reveals evidence for a locus on chromosome 11 influencing variation in LDL cholesterol in the NHLBI Family Heart Study. Hum. Genet. 111: 263–269.[CrossRef][Medline]

  27. Rainwater, D. L., L. Almasy, J. Blangero, S. A. Cole, J. L. VandeBerg, J. W. MacCluer, and J. E. Hixson. 1999. A genome search identifies major quantitative trait loci on human chromosomes 3 and 4 that influence cholesterol concentrations in small LDL particles. Arterioscler. Thromb. Vasc. Biol. 19: 777–783.[Abstract/Free Full Text]

  28. Hegele, R. A., D. D. Ramdath, M. R. Ban, M. N. Carruthers, C. V. Carrington, and H. Cao. 2001. Polymorphisms in PNLIP, encoding pancreatic lipase, and associations with metabolic traits. J. Hum. Genet. 46: 320–324.[CrossRef][Medline]

  29. Elbein, S. C., and S. J. Hasstedt. 2002. Quantitative trait linkage analysis of lipid-related traits in familial type 2 diabetes: evidence for linkage of triglyceride levels to chromosome 19q. Diabetes. 51: 528–535.[CrossRef][Medline]

  30. Soro, A., P. Pajukanta, H. E. Lilja, K. Ylitalo, T. Hiekkalinna, M. Perola, R. M. Cantor, J. S. Viikari, M. R. Taskinen, and L. Peltonen. 2002. Genome scans provide evidence for low-HDL-C loci on chromosomes 8q23, 16q24.1-24.2, and 20q13.11 in Finnish families. Am. J. Hum. Genet. 70: 1333–1340.[CrossRef][Medline]

  31. Yang, Q., C. Q. Lai, L. Parnell, L. A. Cupples, X. Adiconis, Y. Zhu, P. W. Wilson, D. E. Housman, A. M. Shearman, R. B. D'Agostino, et al. 2005. Genome-wide linkage analyses and candidate gene fine mapping for HDL3 cholesterol: the Framingham Study. J. Lipid Res. 46: 1416–1425.[Abstract/Free Full Text]

  32. Davis, R. C., A. Diep, W. Hunziker, I. Klisak, T. Mohandas, M. C. Schotz, R. S. Sparkes, and A. J. Lusis. 1991. Assignment of human pancreatic lipase gene (PNLIP) to chromosome 10q24-q26. Genomics. 11: 1164–1166.[CrossRef][Medline]

  33. Giller, T., P. Buchwald, D. Blum-Kaelin, and W. Hunziker. 1992. Two novel human pancreatic lipase related proteins, hPLRP1 and hPLRP2. Differences in colipase dependence and in lipase activity. J. Biol. Chem. 267: 16509–16516.[Abstract/Free Full Text]

  34. Oikawa, E., H. Iijima, T. Suzuki, H. Sasano, H. Sato, A. Kamataki, H. Nagura, M. J. Kang, T. Fujino, H. Suzuki, et al. 1998. A novel acyl-CoA synthetase, ACS5, expressed in intestinal epithelial cells and proliferating preadipocytes. J. Biochem. (Tokyo). 124: 679–685.[Abstract/Free Full Text]

  35. Miyazaki, M., Y. C. Kim, M. P. Gray-Keller, A. D. Attie, and J. M. Ntambi. 2000. The biosynthesis of hepatic cholesterol esters and triglycerides is impaired in mice with a disruption of the gene for stearoyl-CoA desaturase 1. J. Biol. Chem. 275: 30132–30138.[Abstract/Free Full Text]

  36. Cohen, P., M. Miyazaki, N. D. Socci, A. Hagge-Greenberg, W. Liedtke, A. A. Soukas, R. Sharma, L. C. Hudgins, J. M. Ntambi, and J. M. Friedman. 2002. Role for stearoyl-CoA desaturase-1 in leptin-mediated weight loss. Science. 297: 240–243.[Abstract/Free Full Text]

  37. Rozen, R., J. Vockley, L. Zhou, R. Milos, J. Willard, K. Fu, C. Vicanek, L. Low-Nang, E. Torban, and B. Fournier. 1994. Isolation and expression of a cDNA encoding the precursor for a novel member (ACADSB) of the acyl-CoA dehydrogenase gene family. Genomics. 24: 280–287.[CrossRef][Medline]

  38. Strausberg, R. L., E. A. Feingold, L. H. Grouse, J. G. Derge, R. D. Klausner, F. S. Collins, L. Wagner, C. M. Shenmen, G. D. Schuler, S. F. Altschul, et al. Mammalian Gene Collection Program Team. 2002. Generation and initial analysis of more than 15,000 full-length human and mouse cDNA sequences. Proc. Natl. Acad. Sci. USA. 99: 16899–16903.[Abstract/Free Full Text]

  39. Agrimi, G., M. A. Di Noia, C. M. Marobbio, G. Fiermonte, F. M. Lasorsa, and F. Palmieri. 2004. Identification of the human mitochondrial S-adenosylmethionine transporter: bacterial expression, reconstitution, functional characterization and tissue distribution. Biochem. J. 379: 183–190.[CrossRef][Medline]

  40. Li, C., and P. W. Tucker. 1993. DNA-binding properties and secondary structural model of the hepatocyte nuclear factor 3/fork head domain. Proc. Natl. Acad. Sci. USA. 90: 11583–11587.[Abstract/Free Full Text]

  41. Banham, A. H., N. Beasley, E. Campo, P. L. Fernandez, C. Fidler, K. Gatter, M. Jones, D. Y. Mason, J. E. Prime, P. Trougouboff, et al. 2001. The FOXP1 winged helix transcription factor is a novel candidate tumor suppressor gene on chromosome 3p. Cancer Res. 61: 8820–8829.[Abstract/Free Full Text]

  42. Yue, P., W. L. Isley, W. S. Harris, S. Rosipal, C. D. Akin, and G. Schonfeld. 2005. Genetic variants of ApoE account for variability of plasma low-density lipoprotein and apolipoprotein B levels in FHBL. Atherosclerosis. 178: 107–113.[CrossRef][Medline]

  43. Lambert, G. 2007. Unravelling the functional significance of PCSK9. Curr. Opin. Lipidol. 18: 304–309.[Medline]

  44. Lander, E., and L. Kruglyak. 1995. Genetic dissection of complex traits: guidelines for interpreting and reporting linkage results. Nat. Genet. 11: 241–247.[CrossRef][Medline]


Add to CiteULike CiteULike   Add to Complore Complore   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us   Add to Digg Digg   Add to Reddit Reddit   Add to Technorati Technorati    What's this?



This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Supplemental Data
Right arrow All Versions of this Article:
M700078-JLR200v1
48/12/2632    most recent
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Sherva, R.
Right arrow Articles by Neuman, R. J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Sherva, R.
Right arrow Articles by Neuman, R. J.
Social Bookmarking
Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
 All ASBMB Journals   Journal of Biological Chemistry 
 Molecular and Cellular Proteomics   ASBMB Today