Skip Navigation


Human Molecular Genetics Advance Access originally published online on August 2, 2005
Human Molecular Genetics 2005 14(17):2595-2605; doi:10.1093/hmg/ddi294
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
14/17/2595    most recent
ddi294v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
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 Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (28)
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Naukkarinen, J.
Right arrow Articles by Peltonen, L.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Naukkarinen, J.
Right arrow Articles by Peltonen, L.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

© The Author 2005. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oupjournals.org

USF1 and dyslipidemias: converging evidence for a functional intronic variant

Jussi Naukkarinen1,2, Massimiliano Gentile1,5, Aino Soro-Paavonen3, Janna Saarela1,2, Heikki A. Koistinen3, Päivi Pajukanta4, Marja-Riitta Taskinen3 and Leena Peltonen1,2,4,*

1Department of Molecular Medicine, National Public Health Institute, Finland, 2Department of Medical Genetics, University of Helsinki, Biomedicum, 00290 Helsinki, Finland, 3Department of Medicine, Helsinki University Central Hospital, Helsinki, Finland, 4Department of Human Genetics, David Geffen School of Medicine at UCLA, University of California, Los Angeles, CA 90095-7088, USA and 5Biomedicum Bioinformatics Unit, University of Helsinki, Finland

* To whom correspondence should be addressed at: Biomedicum Helsinki, Haartmaninkatu 8, 00290 Helsinki, Finland. Tel: +358 947448393; Fax: +358 947448480; Email: leena.peltonen{at}ktl.fi

Received May 23, 2005; Accepted July 21, 2005


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 
Upstream transcription factor 1 (USF1), the first gene associated with familial combined hyperlipidemia (FCHL), regulates numerous genes of glucose and lipid metabolism. Phenotypic overlap between FCHL, type 2 diabetes and the metabolic syndrome makes this gene an intriguing candidate in the disease process of these traits as well. As no disease-associated mutations in the coding region of USF1 have been identified, we addressed the functional role of intronic single nucleotide polymorphisms (SNPs) which define the FCHL-risk alleles of USF1, and identified that a 20 bp DNA sequence, containing the critical intronic SNP, binds nuclear protein(s), representing a likely transcriptional regulatory element. This functional role is further supported by the differential expression of USF1-regulated genes in fat biopsy between individuals carrying different allelic variants of USF1. Importantly, apolipoprotein E (APOE) is the most downregulated gene in the risk individuals, linking the potential risk alleles of USF1 with the impaired APOE-dependent catabolism of atherogenic lipoprotein particles.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 
Familial combined hyperlipidemia (FCHL) is the most common genetic dyslipidemia and contributes significantly to the risk of premature coronary heart disease (1Go,2Go). The genetic background of this complex trait is polygenic and multiple genetic loci for FCHL have been established in linkage analyses (3Go–5Go). We recently reported evidence for the first FCHL gene by demonstrating that specific alleles of upstream stimulatory factor 1 (USF1) defined by the intragenic single nucleotide polymorphisms (SNPs) showed evidence for association with FCHL as well as several of its component traits (6Go). Allele-specific effects of USF1 in the response to glucose load as well as in modulating the correlation between body mass index (BMI) and cholesterol levels have since then been shown in a large multicenter study sample of 822 healthy young males from 11 different European populations (7Go). Association has been replicated either with the strongest associated SNPs of the original study (usf1s1=rs3737787 and usf1s2=rs2073658) (6Go) or with SNPs in strong linkage disequilibrium (LD) with them (7Go), implying the critical role of this particular region covering USF1. All of the disease-associated SNPs represent variants of the non-coding regions of USF1 and their relationship to the actual functional defect has remained unknown.

USF1 is a transcription factor of the basic helix–loop–helix leucine zipper family and is a key regulator of numerous genes involved in lipid and glucose metabolism, such as several apolipoproteins, enzymes and transporters. USF1 mediates their regulation in response to metabolic cues, such as glucose and insulin, by binding specific promoter sequences called E-boxes with the consensus sequence CACGTG either as a homodimer (8Go,9Go) or more often as a heterodimer with the related USF2 transcription factor coded by a gene on 19q (10Go). The central role of the USF1 transcription factor in coordinating various players in these intricate metabolic pathways makes it an attractive candidate gene for several complex diseases with defective glucose and/or lipid metabolism, including type 2 diabetes and the metabolic syndrome, which co-localize with the 1q21 locus containing USF1. In addition to the genetic overlap, extensive phenotypic overlap exists between the FCHL and the metabolic syndrome, exemplified by a recent report that nearly two-thirds of individuals with FCHL in a US study also fit the diagnostic criteria of the metabolic syndrome (11Go). This overlap also suggests that USF1 may contribute to the molecular pathology of these disorders.

For numerous common diseases, the associated variants do not represent obvious defects in the coded polypeptide (6Go,12Go,13Go). Rather, the reported variants are located in the putative regulatory regions, such as in the promoter or conserved intronic sequences. Nor have we identified any variants in the coding region of the USF1 gene that could explain the functional defect contributing to the complex FCHL phenotype. Subsequent sequencing of USF1 in other populations has also failed to detect any such coding variants (7Go) and has led to the question whether some of the disease-associated polymorphisms are themselves functional or rather just serve as markers for an extended at-risk haplotype.

Abnormal regulation of the transcript levels, typically resulting in abnormal levels of the polypeptide in relevant tissues, could potentially explain the allelic differences in the function of USF1, contributing to the disease phenotype. Very little is so far known about the transcriptional regulation of the USF1 gene itself. Some initial evidence for a regulatory element located in intron 7 came from our earlier in vitro studies indicating that a 268 bp segment, including the critical SNP usf1s2, enhanced the transcription efficiency of a reporter gene (6Go).

A hypothesis was recently put forward suggesting that an allele of usf1s2 may activate an alternative promoter within USF1, resulting in the start of transcription from one of the two AUGs in exon 8 and thus a shorter protein product lacking the trans-activation domain (14Go). Such truncated USFs have been shown to act as trans-dominant inhibitors in vitro (15Go,16Go) and could impact the normal functioning of USF1.

Here, we addressed the possible functional role of the sequence around the USF1 SNPs associated with dyslipidemias to explain how allelic variations in this region could contribute to the dyslipidemic phenotype.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 
Critical intronic sequence binds nuclear protein
Among the nine genotyped intragenic USF1 SNPs, two represent synonymous variants in the coding region, whereas seven were located in the introns (Fig. 1). The strongest evidence for association in FCHL families was initially observed with two SNPs: usf1s1 in the 3'-untranslated region (UTR) and usf1s2 in intron 7, located 1.24 kb apart and essentially in complete LD (R2=0.93, D'=0.98). We analyzed the sequence environment of all 17 intronic SNPs identified in the original sequencing (2Go) across species to monitor for phylogenetic conservation that would provide clues of their functional importance. The strongest associating SNP, usf1s2 in intron 7, was located in a DNA stretch fully conserved from human to chimp, dog, mouse, and rat, within a genomic region otherwise rich in non-conserved nucleotides (Fig. 2). Only two other SNPs were located in such a conserved sequence stretch, namely rs1556259 and rs1556260 both in intron 1, but because they revealed no association with FCHL or its component traits, we did not pursue them further. The regional conservation of this sequence containing usf1s2 encouraged us to study whether it harbored some elements functionally important to the dynamics of USF1 transcription.



View larger version (13K):
[in this window]
[in a new window]
  		Figure 1. Schematic view of the 6.7 kb USF1 gene. Exons are depicted as thick boxes, UTRs as thinner boxes and introns as lines. Genotyped USF1 SNPs are marked above the gene with associating SNPs indicated with asterisks. A segment of intron 7 is amplified to show the location of the sequence (black bar), used to generate the 20mer probe used in the EMSA. Nearby SNPs are indicated with larger font and arrows.

 


View larger version (20K):
[in this window]
[in a new window]
  		Figure 2. Cross-species conservation and EMSA probes. Two probes were constructed, which were capable of producing a shift in the EMSA, one of length 34 bp and the other 20 bp. The 34mer probe contained all three SNPs from this intron 7 region, whereas the 20mer probe only contained the critical usf1s2 SNP. The cross-species sequence conservation and the consensus sequence are shown below. Y stands for pyrimidine and R for purine. Notably, the nucleotide at usf1s2 itself is fully conserved, the risk allele representing the ancestral allele.

 
We first determined whether the region of usf1s2 represents a binding site for DNA binding proteins. Two 34mer probes (Fig. 2) containing SNPs usf1s2–4 were constructed and allowed to vary for the two alleles of usf1s2. After incubation with nuclear extract proteins of HeLa cells, both critical sequence variants produced an electrophoretic mobility shift (EMS) on a polyacrylamide gel. To further restrict the potentially functional sequence motif, we performed the EMS analyses using a shorter, 20mer probe pair that shared with the 34 bp probe, the critical most conserved nucleotide sequence. This probe produced a mobility shift, comparable with the 34 bp shift, whereas a similar 20 bp probe representing the sequence containing the other strongly associated SNP usf1s1 located in the 3'-UTR of USF1 did not produce a shift (Fig. 3A). The binding of the probes to nuclear proteins could be competed using unlabeled specific probe, but not with a non-specific probe (Fig. 3B).



View larger version (90K):
[in this window]
[in a new window]
  		Figure 3. (A) EMSA results show that the sequence around usf1s2 binds nuclear protein(s) from HeLa cell extract. A comparable gel shift was obtained with both the 34 and 20 bp probes. The different usf1s2 allelic variants both produce a gel shift, marked by an arrow. Conversely, neither variant of the 20 bp probe representing the sequence around the other associated SNP usf1s1 in the 3'-UTR is capable of producing a gel shift. (B) The specificity of the binding of nuclear protein(s). The 34 bp probe representing the sequence around usf1s2 produces a strong gel shift, which can be gradually competed with the addition of increasing molar concentrations of unlabeled probe.

 
Carriers of USF1 risk allele show differential expression of downstream genes in fat
A qualitative or quantitative functional change of a transcription factor, such as USF1, would be expected to be reflected in the expression efficiency or pattern of the genes under its control. We hypothesized that if the usf1s2 polymorphism was either functional itself or served as a marker for an unknown functional element in the vicinity, we should be able to see a difference in the transcriptional profile of USF1-regulated genes in fat biopsies of individuals carrying either the ‘risk’ or the ‘non-risk’ allele. This would represent an eloquent in vivo approach to address the function of the potential susceptibility polymorphism. Through a query of a transcription factor database (Transfac) and published literature, a total of 40 USF1-controlled genes were identified and selected for further analysis, regardless of the knowledge over biological pathway or tissue specificity (Table 1).


View this table:
[in this window]
[in a new window]
  		Table 1. Genes with reported involvement of USF1 in their regulation
 
To study the possible effects of allelic variants of USF1 on the transcriptional profiles, we obtained fat biopsies from 19 individuals from our cohort of dyslipidemic families (FCHL and low HDL-C). They included seven individuals homozygous for the rare 2–2 genotype of usf1s2 (marking the ‘non-risk’ haplotype) and 12 individuals carrying the common one allele (marking the ‘risk’ haplotype) in either heterozygous (8Go) or homozygous (4Go) form. Out of 40 listed USF1-controlled genes, 29 were represented on the Affymetrix U133A chips used in this study, some genes by multiple probe sets. After rigorous statistical filtering and analysis (see Materials and Methods), we found that 13 genes, represented by a total of 19 probe sets, were expressed in the adipose tissue at a sufficiently high level to produce reliable signals and were included in the study (Table 1). Several highly relevant genes of lipid and glucose metabolism were on this list, as well as a few genes whose relevancy is not immediately obvious. After normalization, three genes (represented by a total of six probe sets all in agreement) differed significantly (P≤0.05) in their expression between the two haplotype groups of USF1, as evaluated using a two-sample t-test with no assumption of equal variance. All three genes, differentially expressed between individuals carrying either the ‘risk’ or the ‘non-risk’ haplotype of USF1, were highly relevant to the phenotype: the ATP-binding cassette subfamily A (ABCA1) (17Go), angiotensinogen (AGT) (18Go) and apolipoprotein E (APOE) (19Go) (Fig. 4).



View larger version (44K):
[in this window]
[in a new window]
  		Figure 4. Schematic representation of the identification of the significantly differentially regulated USF1-controlled genes. The initial list of 40 genes was narrowed down to the 13 that were expressed in the fat biopsies. Of these, three important metabolic genes were differentially expressed at steady state between individuals carrying the risk or non-risk haplotype of USF1. P-values are from a two-sample t-test with no assumption of equal variance.

 
Differential response of ACACA to insulin
Signals such as serum insulin and glucose are critical in the regulation of various metabolic genes. Insulin is known to influence the ability of USF1 to bind the E-box sequence and thus participates in the regulation of gene expression in response to metabolic changes (20Go). To evaluate the possible contribution of these factors on the expression of the USF1-controlled genes, analyses of co-variance (ANCOVA) models were fitted to the data. We further extended the models to also test for possible effects of BMI, triglycerides and homeostatic model assessment (HOMA), a measure of insulin resistance based on the values for fasting serum insulin and glucose (21Go). For all but one of the genes tested, no significant contribution was observed from the various co-variates, hence resulting in test statistics essentially the same as those of the simple, two-sample t-test. However, in agreement with earlier findings (22Go), we observed a detectable effect of the insulin level on the expression of acetyl-CoA carboxylase alpha (ACACA) (P=0.05). This relationship was closer scrutinized using linear regression, which demonstrated a moderately strong negative correlation (R2=0.453) between the steady-state transcript level of ACACA and fasting levels of insulin. Partial regression for the haplotype groups additionally demonstrated that this correlation was in essence much stronger in the individuals with the 2–2 ‘non-risk’ haplotype (R2=0.956) than in individuals carrying the ‘risk’ haplotype (R2=0.093) of USF1 (Fig. 5).



View larger version (17K):
[in this window]
[in a new window]
  		Figure 5. Correlation of ACACA expression and insulin levels. The transcript levels of ACACA, as measured from adipose tissue, are significantly negatively correlated with serum insulin (P=0.05). This correlation is much stronger in individuals with the ‘non-risk’ allele of the SNP usf1s2. Two outliers have been removed from this representation.

 
We also tested whether any effect of parameters such as sex, study cohort (FCHL or low HDL) or familiality should be taken into account in our analyses by performing an unsupervised clustering of individual expression levels. No effect could be detected for any measures looked at, as evidenced by the random clustering of individuals with respect to these variables (data not shown).

Changes in APOE stand out in whole genome transcript profile
In addition to the analyses of known USF1-regulated genes, we tested the whole microarray data for altered transcript levels of genes between carriers of the different USF1 haplotypes. Approaches of this kind have been successfully used to identify the pathways and collections of co-regulated genes in different sets (23Go). This has most often been done when comparing groups with a clear phenotypic difference, such as diabetic versus non-diabetic (23Go) or cancer tissue versus non-cancerous tissue (24Go). In our study, changes in which the expression differences were ≥1.5-fold and that reached our limit of statistical significance (P≤0.05) in the two-sample t-test were defined as significant. This approach identified 15 genes, among which 10 were upregulated and five were downregulated in individuals with the non-risk haplotype (Fig. 6). Again, the top gene on the list of downregulated genes in the risk individuals was APOE. The expression of APOE in the adipose tissue of individuals with the risk haplotype of USF1 was twice as low as expression in those carrying the non-risk haplotype. Other potentially interesting genes on the list included CYP4B1, involved in fatty acid metabolism, and VEGF, involved in angiogenesis, hypertension and it is an essential mediator in angiotensin II-induced vascular inflammation (25Go). Experimental data are needed to verify whether USF1 plays a role in the regulation of these genes as well.



View larger version (33K):
[in this window]
[in a new window]
  		Figure 6. Global gene expression differences between carriers and non-carriers of ‘risk’ allele of USF1 as defined by the SNP usf1s2. Log2-transformed fold changes of gene expression for all genes that passed the stringent filtering steps are plotted as a function of the log of the P-values from the statistical test for differential expression. The vast majority of genes (colored gray) show no significant expression changes, whereas a few genes exhibit ≥1.5-fold changes in expression and a t-test P-value ≤0.05 (illustrated by the intersecting dotted lines overlaid on the plot). They are colored green (for downregulated) or red (for upregulated) in carriers of the ‘risk’ allele. Notably, APOE (represented by two probe sets) is the most downregulated transcript among all genes on the array. In addition, the two other USF1-target genes with statistically significant expression changes but that did not meet our most stringent fold-change criteria, namely ABCA1 and AGT, are shown.

 
No strong effect of critical SNP on expression of regional genes
Finally to investigate whether the putative regulatory element in intron 7 could represent a strong cis-regulatory element and exert its control on the expression of other genes in the vicinity of USF1, we studied the expression levels of 10 flanking genes from the 5' CD244 gene all the way to APOA2, a stretch of 392 kb. Of these 10 genes, six are transcribed from the same DNA strand as USF1 and four from the opposite strand. The only probe set whose expression level differed significantly depending on an individual's allele at usf1s2 was one for the adjacent platelet F11 receptor (F11R, also known as JAM1) gene (P=0.013). This was interesting as the critical chromosomal interval showing an association in FCHL families extended into the F11R gene in alleles of high-triglyceride men (6Go). On the U133A array, two probe sets represent F11R; however, only one showed significant difference between the two USF1 haplotype groups. Upon closer examination of the representative sequence in the genome, we noted that the probe set which showed differential expression did not actually represent the F11R gene, but rather a short expressed sequence tag (EST) (AW995043 [GenBank] ) immediately adjacent to it, 43.5 kb 3' from the USF1 gene.

Intronic SNP does not activate an alternative promoter in exon 8
We designed two primer pairs to amplify the 3' and the 5' ends of the USF1 cDNA, respectively, and measured the relative amounts of each product with quantitative real-time PCR. If one allele of the SNP activated an internal promoter causing the transcription of a truncated mRNA, as suggested earlier (14Go), the quantities of the 5' and 3' products, amplified from an invididual's cDNA, would deviate from the 1:1 ratio expected of a full-length mRNA. Comparing the relative quantities of these two products in cDNA from five individuals homozygous for the risk allele and five individuals homozygous for non-risk alleles of usf1s2, we observed no such deviation from the 1:1 ratio (two-tailed t-test, P=0.35). This result would imply that there is no allelic difference in the length of USF1 transcript produced. To compare expression levels of USF1 transcript in these samples, GAPDH expression was used to normalize the data. No between-group differences existed in the expression of USF1.


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
MATERIALS AND METHODS
 REFERENCES
 
Here, we investigated the non-coding SNPs reported to characterize the alleles associated with FCHL and component traits of the metabolic syndrome (6Go,7Go). We observed that the DNA sequence containing the strongest associating SNP usf1s2 was conserved across species and binds protein(s) of nuclear extract, as shown by its ability to produce a mobility shift in an electrophoretic mobility shift assay (EMSA) experiment. In addition to this in vitro evidence, we were able to see differential expression of downstream genes of USF1 in the adipose tissue of 19 individuals depending on whether they carried either the risk or the non-risk allele of the SNP usf1s2. Our data are in accordance with the hypothesis that functional variants in complex disease are most likely located in non-coding regions of genes, such as the promoter or the enhancers located in introns. Here, we show that the disease-associating intron 7 SNP (usf1s2), defining the risk allele of USF1 in FCHL families, shows evidence for its functional importance both in vitro and in vivo.

Transcription factors bind to very specific nucleotide sequences characterized by a short core-sequence of ~4–6 bp flanked by a variable number of degenerate nucleotides. The sequence around usf1s2 in intron 7 agrees well with these criteria showing the perfect cross-species conservation of 5 bp. Our EMSA results lend strong evidence supporting the hypothesis that the sequence surrounding usf1s2 truly represents a functional element. We earlier reported that a 268 bp segment that included this conserved DNA motif enhanced expression of a reporter gene and only in the correct orientation (6Go). This speaks strongly for the cis-regulatory role of this intronic sequence, but does not pinpoint the functional element within it. This to our knowledge is the first demonstration of a possible regulatory element of the USF1 gene. However, as we reported earlier for a small number of fat biopsies (6Go), we observed no differences in the steady-state levels of USF1 transcripts in fat between individuals carrying different alleles of usf1s2, not too surprising for a gene known to be highly dependent on the local variations of the insulin content. Just as in the reporter gene expression assay which failed to detect any difference in expression efficiency between segments carrying different variants of usf1s2, we could not unequivocally show a quantitatively verified difference in the binding capacity of the two SNP variants in the EMSA. We recognize the obvious limitations of the functional assays applied here. The EMSA is a purely in vitro assay, in which the DNA sequence under study is in essence naked and is tested in the absence of its normal cellular environment with all its transcriptional machinery and host of other regulatory elements. Some of these interacting elements can be found at a significant distance and would not be present in the probe used for the EMSA. Any tissue-specific effects would also be abolished in the in vitro assay. However, our data from the expression profiles of USF1-regulated genes in fat would indicate an allele-specific difference in the expression pattern of these genes and would imply an allele-specific difference in the function of USF1.

We analyzed the known downstream genes of USF1 for possible changes in expression. As the transcriptional regulation of genes is usually the fine-tuned result of a concert of various transcription factors and enhancers/repressors that depend on the tissue and different hormonal/environmental cues, it is not expected that a change in any single factor would have a dramatic effect. However, we found the USF1-regulated genes APOE (19Go), ABCA1 (17Go) and AGT (18Go) being significantly differentially regulated depending on the specific allele at the SNP usf1s2. All three genes are highly relevant to the dyslipidemic phenotype. ABCA1 is involved in the first step of the reverse transport of cholesterol by mediating the efflux of phospholipids and cholesterol from macrophages to the nascent HDL particles (26Go). Loss of function alleles of ABCA1 have been shown to result in Tangier's disease and familial hypoalphalipoproteinemia (27Go), characterized by very low HDL levels. AGT is an essential component in the control of blood pressure and volume by regulating the amount of water absorption by the kidneys, among other things. APOE facilitates the removal of chylomicron and VLDL remnants from the circulation via the LDL receptor-related protein (LRP)-mediated endocytosis in the liver (28Go–30Go). APOE has a high affinity to the LDL receptor and an over-expression of APOE results in marked reduction in plasma LDL (31Go). A reduction in APOE thus leads to an accumulation and increased residence time of cholesterol-rich chylomicron and VLDL remnants in circulation—a highly atherogenic phenotype (28Go,32Go). Defects in APOE have also been shown to result in familial dysbetalipoproteinemia with impaired clearance of cholesterol and triglycerides from plasma (33Go,34Go). Recent evidence suggests that APOE has also a critical role in intracellular lipid metabolism. The recycling of APOE from triglyceride-rich lipoproteins is critical for HDL metabolism and cholesterol efflux (35Go). The apparent unfavorable effect of the usf1s2 risk allele on APOE expression shown here follows fittingly from our earlier findings of the association of USF1 with FHCL and component traits (6Go).

Our data agree well with recent findings by Putt et al. (7Go) They showed that in their sample of 822 healthy young men, the marker 475C>T (rs2073655), which is in full LD with usf1s2, was by itself associated with plasma APOE levels in a dose-dependent manner (P=0.018) and showed an interactive effect on APOE levels with a SNP in APOCIII (P=0.0012). Putt et al. also reported an association of the risk haplotype of USF1 (again, defined by SNPs in LD with usf1s1 and usf1s2, essentially representing the same segment of DNA as our risk haplotype) with an unfavorable response of peak glucose and triglyceride response to fat load, as measured by an oral glucose tolerance test (OGTT) and an oral fat tolerance test, respectively. Only a weak trend of association was seen with the TG trait in the EARSII study, but as the authors stated, it only highlights the difference in study subjects. We originally saw the strongest association in men with triglycerides above the 90th population percentile, whereas the EARSII study consisted of healthy, young males with comparatively low TG levels. Further, only a borderline association with steady-state measurements of fasting glucose was identified, which would suggest that the USF1 polymorphism has its strongest effect on the control of the post-prandial response of relevant genes to metabolic stress. Considering that in western countries where three meals are consumed daily, most of the time is spent in a post-prandial state, making any defects there highly influential. It remains to be seen what allele-specific differences arise among the expression of various USF1-regulated genes when the body responds to such metabolic stress as imposed by an OGTT or an insulin clamp.

The correlation of the ACACA expression with insulin levels replicated the earlier findings (22Go), but additionally revealed an important difference in the extent of this correlation between the two USF1 allelic haplotypes. The correlation was especially strong within the protective haplotype group. This differential transcriptional response to insulin is very interesting, given the known role of USF1 in mediating the response of metabolic genes to changes in insulin and glucose levels (20Go). ACACA occupies a key position in overall lipid metabolism as the enzyme catalyzing the rate-limiting step in the biosynthesis of long-chain fatty acids (36Go). These findings suggest a role for USF1 in the complex molecular pathway resulting in a well-established insulin resistance in tissues of patients with FCHL and the metabolic syndrome.

An investigation of the USF1 regional genes did not show any influence of the usf1s2 alleles over their expression, suggesting that the effects are contained to the USF1 gene. However, a small unknown EST (AW995043 [GenBank] ) immediately 3' of F11R was expressed differently between the groups carrying different alleles at usf1s2. This EST was expressed at a very low level, approaching the detection limit of the arrays, but the data were further validated by closer scrutiny of the signals observed for the individual probe pairs. Several different analysis approaches consistently reproduced the results found at the probe set level, demonstrating that ‘risk’ allele carriers exhibited higher expression for this particular EST. ESTs usually represent fragments of transcribed genes, but as AW995043 [GenBank] is transcribed from the opposite strand compared with F11R and has no overlap with any known splice variant, it does not seem to be a part of it. The differential expression of this EST may be an anomaly, or it could represent a small regulatory RNA molecule with an as of yet unknown function.

Small changes in the level of expression of individual genes are not likely to result in a pathological condition, but it has been suggested that complex metabolic diseases can be a result of such small changes occurring in several genes in the same metabolic pathway. These could occur as the result of several individual changes in the regulatory regions of different genes in the same pathway or conversely (and perhaps more likely) as the result of a change in a transcription factor common to the regulation of these genes. Although recognizing the limitations of conclusions drawn from a limited number of patient samples and the need to characterize in vitro the exact molecular mechanisms involved, converging genetic and functional evidence for the role of the variants identified is encouraging. Along these lines, the allelic variants of USF1 carrying different functional features with apparent influence on the expression levels of several downstream genes would affect several metabolic pathways, critical for the molecular pathogenesis of dyslipidemias and potentially the metabolic syndrome (Fig. 7).



View larger version (27K):
[in this window]
[in a new window]
  		Figure 7. Schematic representation of the mechanism of allele-specific regulation of the USF1 transcript levels and probable consequences of the variations in the amount of USF1 protein. Protein(s) bind a regulatory sequence in intron 7 of USF1 and affect the level of transcription. USF1 dimerizes (most often with USF2) and binds an E-box sequence in the promoter of numerous genes to activate their transcription in response to signals such as glucose and dietary carbohydrates. Post-translational control of USF1 activity is mediated by phosphorylation of the dimer, which precludes its binding to the E-box motif (20Go). The observed decrease in the transcript level of downstream genes, if reflected at the polypeptide level, would result in changes highly relevant for dyslipidemias and the metabolic syndrome.

 

   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
REFERENCES
 
Electrophoretic mobility shift assay
DNA probes representing both strands of the regions of interest were ordered from Proligo and 5' end labeled with [{gamma}-32P]ATP using T4 polynucleotide kinase. Excess un-incorporated label was removed using the QIAquick kit (Qiagen), according to manufacturer's instructions. Nuclear extracts were incubated for 30 min at room temperature in binding buffer [50 mM Tris–HCl (pH 7.5), 5 mM MgCl2, 2.5 mM EDTA, 2.5 mM dithiothreitol, 2.5 mM NaCl, 0.25 µg/µl poly(dI–dC)·poly(dI–dC) and 20% glycerol] and then electrophoresed on a 6% polyacrylamide gel containing 0.5 M TBE buffer. Gels were autoradiographed at –70°C. In order to test for specificity of binding, the extracts were run with an increasing concentration of unlabeled ‘cold’ ds-probe as well as non-specific probe representing the sequence around the 3'-UTR SNP usf1s1, which did not produce a gel shift.

Expression array analysis
Collection of study subjects from the FCHL (6Go) and low HDL-C families (37Go) has been described earlier. Each subject gave a written informed consent prior to participating in the study. The Ethics Committee of the Helsinki University Central Hospital approved the study design and all the samples were collected in accordance with the Helsinki declaration. We selected 19 individuals for fat biopsy based on their USF1 haplotype, defined by the SNP usf1s2. They included 12 carriers of the risk allele of the critical SNP usf1s2 and seven individuals homozygous for the non-risk allele. Nine of these had been included in our original report (6Go). The average age in both groups was 49 and the gender distribution was close to even (seven females and five males in the risk group versus four females and three males in the non-risk group). Fat biopsies were collected, RNA was extracted and quantified as described previously (6Go). RNA labeling, array processing and scanning were done according to the standard protocol by Affymetrix with minor modifications, as described previously (6Go).

Scanned images were analyzed with Affymetrix Microarray Suite 5 (Affymetrix, Santa Clara, CA, USA) software employing the Statistical Expression Algorithm. Global scaling to a target intensity of 100 was applied to all arrays, after which further data processing was carried out using GeneSpring 6.1 data analysis software (Silicon Genetics, Redwood City, CA, USA). For each probe array, a per gene normalization was applied so that signal intensities were divided by the median intensity calculated using all 19 probe arrays, effectively centering the data around unity.

To identify differentially expressed genes between the two haplotypes, a strategy consisting of two filtering steps, in combination with a statistical analysis, was adopted. First, unreliable or inconsistent data were removed using the Affymetrix detection calls, requiring genes to be scored as present in >50% of the samples in each haplotype group. To avoid losing potentially interesting data pertaining to genes whose expression was ‘turned-off’ in one group but ‘turned-on’ in the other, also genes scoring absent calls in 100% of samples in one group and at least 50% present calls in the other were included. Normalized values were then averaged over samples in each haplotype group and ratios of these were calculated. The distribution of the ratios was evaluated and a cut-off limit of 1.5-fold was selected to focus the attention on the most prominent and reliable expression changes. Significant changes were determined by applying a two-sample t-test, allowing for unequal variances across groups, where a two-sided P-value of 0.05 or lower was considered statistically significant. For the genes represented by more than one probe set on the array, the measurements associated with the more conservative P-value were used.

Statistical analyses
The effect of haplotype on gene expression for selected genes was evaluated using a two-sample t-test, with no assumption of equal variances. Two-sided significance values were calculated and a type I error probability of ≤5% was used to determine the statistical significance. To control for possible confounding contribution from clinically relevant parameters on the observed differences between haplotype groups, ANCOVA were performed. BMI, levels of insulin and triglycerides and HOMA index were included as co-variates to the factor determined by haplotype group, and separate models for each co-variate were evaluated for main and interaction effects. Again, type I errors at a probability of ≤5% were considered statistically significant. Closer scrutiny of haplotype effects on the relationship between gene expression and co-variates was done by linear regression analysis. The linear models were evaluated by studying R, R2 and the F statistic.

Unsupervised hierarchical clustering of samples with respect to patterns of gene expression for selected genes was performed employing an agglomerative algorithm using unweighted pair-group average linkage amalgamation rules. Cluster similarity was determined with Pearson's correlation. Possible associations between branching pattern and gender, affection status (FCHL or low HDL) and familial relationships were analyzed by overlaying status information on the dendrogram and visually assessing potential clusters.

Quantitative real-time PCR
Quantitative real-time PCR of the 5' and 3' ends of the USF1 transcript, as expressed in adipose tissue, was done using the SYBR-Green assay (Applied Biosystems). We carried out two-step RT–PCR using TaqMan RT–PCR kit. Primer sequences are available upon request. We carried out all reactions in triplicate using the ABI Prism 7900 HT Sequence Detection System and analyzed the data using Sequence Detector version 2.2 software.


   ACKNOWLEDGEMENTS
 
We thank Lea Puhakka and Päivi Tainola for expert technical assistance and Drs Matti Jauhiainen, Juha Saharinen, Christian Ehnholm and Oscar Puig for discussions and comments on the manuscript. We thank all the families for their participation in this study. This work was supported by the Center of Excellence of the Academy of Finland, Biocentrum Helsinki, EVO grant from the Helsinki University Hospital, by the 1RO1HL70150-01A1 grant from NIH and by the QLG2-CT-2002-01254 of the European Community.

Conflict of Interest statement. One author has something to declare. Marja Riitta Taskinen has received honoraria and consulting fees from Merck Sharp and Dome, Pfizer Labs Glaxo Smith and Kline, Bristol-Myers Squibb and Takeda and Sanofi Aventis and research support from Laboratories Fournier and Eli Lilly and Novartis.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 

  1. Goldstein, J.L., Schrott, H.G., Hazzard, W.R., Bierman, E.L. and Motulsky, A.G. (1973) Hyperlipidemia in coronary heart disease. II. Genetic analysis of lipid levels in 176 families and delineation of a new inherited disorder, combined hyperlipidemia. J. Clin. Invest., 52, 1544–1568.[ISI][Medline]

  2. Nikkila, E.A. and Aro, A. (1973) Family study of serum lipids and lipoproteins in coronary heart-disease. Lancet, 1, 954–959.[Medline]

  3. Wojciechowski, A.P., Farrall, M., Cullen, P., Wilson, T.M., Bayliss, J.D., Farren, B., Griffin, B.A., Caslake, M.J., Packard, C.J., Shepherd, J. et al. (1991) Familial combined hyperlipidaemia linked to the apolipoprotein AI-CII-AIV gene cluster on chromosome 11q23–q24. Nature, 349, 161–164.[CrossRef][Medline]

  4. Aouizerat, B.E., Allayee, H., Cantor, R.M., Dallinga-Thie, G.M., Lanning, C.D., de Bruin, T.W., Lusis, A.J. and Rotter, J.I. (1999) Linkage of a candidate gene locus to familial combined hyperlipidemia: lecithin: cholesterol acyltransferase on 16q. Arterioscler. Thromb. Vasc. Biol., 19, 2730–2736.[Abstract/Free Full Text]

  5. Pajukanta, P., Terwilliger, J.D., Perola, M., Hiekkalinna, T., Nuotio, I., Ellonen, P., Parkkonen, M., Hartiala, J., Ylitalo, K., Pihlajamaki, J. et al. (1999) Genomewide scan for familial combined hyperlipidemia genes in finnish families, suggesting multiple susceptibility loci influencing triglyceride, cholesterol, and apolipoprotein B levels. Am. J. Hum. Genet., 64, 1453–1463.[CrossRef][ISI][Medline]

  6. Pajukanta, P., Lilja, H.E., Sinsheimer, J.S., Cantor, R.M., Lusis, A.J., Gentile, M., Duan, X.J., Soro-Paavonen, A., Naukkarinen, J., Saarela, J. et al. (2004) Familial combined hyperlipidemia is associated with upstream transcription factor 1 (USF1). Nat. Genet., 36, 371–376.[CrossRef][ISI][Medline]

  7. Putt, W., Palmen, J., Nicaud, V., Tregouet, D.A., Tahri-Daizadeh, N., Flavell, D.M., Humphries, S.E. and Talmud, P.J. (2004) Variation in USF1 shows haplotype effects, gene:gene and gene: environment associations with glucose and lipid parameters in the European Atherosclerosis Research Study II. Hum. Mol. Genet., 13, 1587–1597.[Abstract/Free Full Text]

  8. Casado, M., Vallet, V.S., Kahn, A. and Vaulont, S. (1999) Essential role in vivo of upstream stimulatory factors for a normal dietary response of the fatty acid synthase gene in the liver. J. Biol. Chem., 274, 2009–2013.[Abstract/Free Full Text]

  9. Ribeiro, A., Pastier, D., Kardassis, D., Chambaz, J. and Cardot, P. (1999) Cooperative binding of upstream stimulatory factor and hepatic nuclear factor 4 drives the transcription of the human apolipoprotein A-II gene. J. Biol. Chem., 274, 1216–1225.[Abstract/Free Full Text]

  10. Groenen, P.M., Garcia, E., Debeer, P., Devriendt, K., Fryns, J.P. and Van de Ven, W.J. (1996) Structure, sequence, and chromosome 19 localization of human USF2 and its rearrangement in a patient with multicystic renal dysplasia. Genomics, 38, 141–148.[CrossRef][ISI][Medline]

  11. Hopkins, P.N., Heiss, G., Ellison, R.C., Province, M.A., Pankow, J.S., Eckfeldt, J.H. and Hunt, S.C. (2003) Coronary artery disease risk in familial combined hyperlipidemia and familial hypertriglyceridemia: a case–control comparison from the National Heart, Lung, and Blood Institute Family Heart Study. Circulation, 108, 519–523.[Abstract/Free Full Text]

  12. Horikawa, Y., Oda, N., Cox, N.J., Li, X., Orho-Melander, M., Hara, M., Hinokio, Y., Lindner, T.H., Mashima, H., Schwarz, P.E. et al. (2000) Genetic variation in the gene encoding calpain-10 is associated with type 2 diabetes mellitus. Nat. Genet., 26, 163–175.[CrossRef][ISI][Medline]

  13. Rioux, J.D., Daly, M.J., Silverberg, M.S., Lindblad, K., Steinhart, H., Cohen, Z., Delmonte, T., Kocher, K., Miller, K., Guschwan, S. et al. (2001) Genetic variation in the 5q31 cytokine gene cluster confers susceptibility to Crohn disease. Nat. Genet., 29, 223–228.[CrossRef][ISI][Medline]

  14. Shoulders, C.C. (2004) USF1 on trial. Nat. Genet., 36, 322–323.[CrossRef][ISI][Medline]

  15. Lefrancois-Martinez, A.M., Martinez, A., Antoine, B., Raymondjean, M. and Kahn, A. (1995) Upstream stimulatory factor proteins are major components of the glucose response complex of the L-type pyruvate kinase gene promoter. J. Biol. Chem., 270, 2640–2643.[Abstract/Free Full Text]

  16. Viollet, B., Lefrancois-Martinez, A.M., Henrion, A., Kahn, A., Raymondjean, M. and Martinez, A. (1996) Immunochemical characterization and transacting properties of upstream stimulatory factor isoforms. J. Biol. Chem., 271, 1405–1415.[Abstract/Free Full Text]

  17. Yang, X.P., Freeman, L.A., Knapper, C.L., Amar, M.J., Remaley, A., Brewer, H.B., Jr and Santamarina-Fojo, S. (2002) The E-box motif in the proximal ABCA1 promoter mediates transcriptional repression of the ABCA1 gene. J. Lipid. Res., 43, 297–306.[Abstract/Free Full Text]

  18. Yanai, K., Saito, T., Hirota, K., Kobayashi, H., Murakami, K. and Fukamizu, A. (1997) Molecular variation of the human angiotensinogen core promoter element located between the TATA box and transcription initiation site affects its transcriptional activity. J. Biol. Chem., 272, 30558–30562.[Abstract/Free Full Text]

  19. Salero, E., Gimenez, C. and Zafra, F. (2003) Identification of a non-canonical E-box motif as a regulatory element in the proximal promoter region of the apolipoprotein E gene. Biochem. J., 370, 979–986.[CrossRef][ISI][Medline]

  20. Nowak, M., Helleboid-Chapman, A., Jakel, H., Martin, G., Duran-Sandoval, D., Staels, B., Rubin, E.M., Pennacchio, L.A., Taskinen, M.R., Fruchart-Najib, J. et al. (2005) Insulin-mediated down-regulation of apolipoprotein A5 gene expression through the phosphatidylinositol 3-kinase pathway: role of upstream stimulatory factor. Mol. Cell. Biol., 25, 1537–1548.[Abstract/Free Full Text]

  21. Wallace, T.M., Levy, J.C. and Matthews, D.R. (2004) Use and abuse of HOMA modeling. Diabetes Care, 27, 1487–1495.[Abstract/Free Full Text]

  22. Lopez-Casillas, F., Ponce-Castaneda, M.V. and Kim, K.H. (1991) In vivo regulation of the activity of the two promoters of the rat acetyl coenzyme-A carboxylase gene. Endocrinology, 129, 1049–1058.[Abstract]

  23. Mootha, V.K., Lindgren, C.M., Eriksson, K.F., Subramanian, A., Sihag, S., Lehar, J., Puigserver, P., Carlsson, E., Ridderstrale, M., Laurila, E. et al. (2003) PGC-1alpha-responsive genes involved in oxidative phosphorylation are coordinately downregulated in human diabetes. Nat. Genet., 34, 267–273.[CrossRef][ISI][Medline]

  24. Dyrskjot, L., Thykjaer, T., Kruhoffer, M., Jensen, J.L., Marcussen, N., Hamilton-Dutoit, S., Wolf, H. and Orntoft, T.F. (2003) Identifying distinct classes of bladder carcinoma using microarrays. Nat. Genet., 33, 90–96.[CrossRef][ISI][Medline]

  25. Zhao, Q., Ishibashi, M., Hiasa, K., Tan, C., Takeshita, A. and Egashira, K. (2004) Essential role of vascular endothelial growth factor in angiotensin II-induced vascular inflammation and remodeling. Hypertension, 44, 264–270.[Abstract/Free Full Text]

  26. Oram, J.F. (2002) ATP-binding cassette transporter A1 and cholesterol trafficking. Curr. Opin. Lipidol., 13, 373–381.[CrossRef][ISI][Medline]

  27. Brooks-Wilson, A., Marcil, M., Clee, S.M., Zhang, L.H., Roomp, K., van Dam, M., Yu, L., Brewer, C., Collins, J.A., Molhuizen, H.O. et al. (1999) Mutations in ABC1 in Tangier disease and familial high-density lipoprotein deficiency. Nat. Genet., 22, 336–345.[CrossRef][ISI][Medline]

  28. Zhang, S.H., Reddick, R.L., Piedrahita, J.A. and Maeda, N. (1992) Spontaneous hypercholesterolemia and arterial lesions in mice lacking apolipoprotein E. Science, 258, 468–471.[Abstract/Free Full Text]

  29. Beisiegel, U., Weber, W. and Bengtsson-Olivecrona, G. (1991) Lipoprotein lipase enhances the binding of chylomicrons to low density lipoprotein receptor-related protein. Proc. Natl Acad. Sci. USA, 88, 8342–8346.[Abstract/Free Full Text]

  30. Mahley, R.W. (1988) Apolipoprotein E: cholesterol transport protein with expanding role in cell biology. Science, 240, 622–630.[Abstract/Free Full Text]

  31. Shimano, H., Yamada, N., Katsuki, M., Shimada, M., Gotoda, T., Harada, K., Murase, T., Fukazawa, C., Takaku, F. and Yazaki, Y. (1992) Overexpression of apolipoprotein E in transgenic mice: marked reduction in plasma lipoproteins except high density lipoprotein and resistance against diet-induced hypercholesterolemia. Proc. Natl Acad. Sci. USA, 89, 1750–1754.[Abstract/Free Full Text]

  32. Wilhelm, M.G. and Cooper, A.D. (2003) Induction of atherosclerosis by human chylomicron remnants: a hypothesis. J. Atheroscler. Thromb., 10, 132–139.[Medline]

  33. Kypreos, K.E., Li, X., van Dijk, K.W., Havekes, L.M. and Zannis, V.I. (2003) Molecular mechanisms of type III hyperlipoproteinemia: the contribution of the carboxy-terminal domain of ApoE can account for the dyslipidemia that is associated with the E2/E2 phenotype. Biochemistry, 42, 9841–9853.[CrossRef][Medline]

  34. Rall, S.C., Jr and Mahley, R.W. (1992) The role of apolipoprotein E genetic variants in lipoprotein disorders. J. Intern. Med., 231, 653–659.[ISI][Medline]

  35. Heeren, J., Grewal, T., Laatsch, A., Becker, N., Rinninger, F., Rye, K.A. and Beisiegel, U. (2004) Impaired recycling of apolipoprotein E4 is associated with intracellular cholesterol accumulation. J. Biol. Chem., 279, 55483–55492.[Abstract/Free Full Text]

  36. Ha, J., Daniel, S., Kong, I.S., Park, C.K., Tae, H.J. and Kim, K.H. (1994) Cloning of human acetyl-CoA carboxylase cDNA. Eur. J. Biochem., 219, 297–306.[ISI][Medline]

  37. Soro, A., Pajukanta, P., Lilja, H.E., Ylitalo, K., Hiekkalinna, T., Perola, M., Cantor, R.M., Viikari, J.S., Taskinen, M.R. and Peltonen, L. (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][ISI][Medline]


Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?


This article has been cited by other articles:


Home page
 Nucleic Acids ResHome page
G. Tuteja, S. T. Jensen, P. White, and K. H. Kaestner
Cis-regulatory modules in the mammalian liver: composition depends on strength of Foxa2 consensus site
Nucleic Acids Res., June 13, 2008; (2008) gkn366v1.
[Abstract] [Full Text] [PDF]

Home page
 Arterioscler. Thromb. Vasc. Bio.Home page
K. Kristiansson, E. Ilveskoski, T. Lehtimaki, L. Peltonen, M. Perola, and P. J. Karhunen
Association Analysis of Allelic Variants of USF1 in Coronary Atherosclerosis
Arterioscler. Thromb. Vasc. Biol., May 1, 2008; 28(5): 983 - 989.
[Abstract] [Full Text] [PDF]

Home page
 StrokeHome page
S.-M. Brand-Herrmann
Where Do We Go for Atherothrombotic Disease Genetics?
Stroke, April 1, 2008; 39(4): 1070 - 1075.
[Full Text] [PDF]

Home page
 Arterioscler. Thromb. Vasc. Bio.Home page
A. P. Reiner, C. S. Carlson, N. S. Jenny, J. P. Durda, D. S. Siscovick, D. A. Nickerson, and R. P. Tracy
USF1 Gene Variants, Cardiovascular Risk, and Mortality in European Americans: Analysis of Two US Cohort Studies
Arterioscler. Thromb. Vasc. Biol., December 1, 2007; 27(12): 2736 - 2742.
[Abstract] [Full Text] [PDF]

Home page
 Eur Heart JHome page
Authors/Task Force Members, I. Graham, D. Atar, K. Borch-Johnsen, G. Boysen, G. Burell, R. Cifkova, J. Dallongeville, G. De Backer, S. Ebrahim, et al.
European guidelines on cardiovascular disease prevention in clinical practice: executive summary: Fourth Joint Task Force of the European Society of Cardiology and Other Societies on Cardiovascular Disease Prevention in Clinical Practice (Constituted by representatives of nine societies and by invited experts)
Eur. Heart J., October 1, 2007; 28(19): 2375 - 2414.
[Full Text] [PDF]

Home page
 HypertensionHome page
M. E. Dickson, M. B. Zimmerman, K. Rahmouni, and C. D. Sigmund
The -20 and -217 Promoter Variants Dominate Differential Angiotensinogen Haplotype Regulation in Angiotensinogen-Expressing Cells
Hypertension, March 1, 2007; 49(3): 631 - 639.
[Abstract] [Full Text] [PDF]

Home page
 J. Lipid Res.Home page
G. M. van der Vleuten, A. Isaacs, A. Hijmans, C. M. van Duijn, A. F. H. Stalenhoef, and J. de Graaf
The involvement of upstream stimulatory factor 1 in Dutch patients with familial combined hyperlipidemia
J. Lipid Res., January 1, 2007; 48(1): 193 - 200.
[Abstract] [Full Text] [PDF]

Home page
 J. Lipid Res.Home page
M. C. G. J. Brouwers, R. M. Cantor, N. Kono, J. l. Yoon, C. J. H. van der Kallen, M. A. L. Bilderbeek-Beckers, M. M. J. van Greevenbroek, A. J. Lusis, and T. W. A. de Bruin
Heritability and genetic loci of fatty liver in familial combined hyperlipidemia
J. Lipid Res., December 1, 2006; 47(12): 2799 - 2807.
[Abstract]