| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Letters to the Editor |
Molecular Pathology Unit, Canterbury Health Laboratories, Christchurch, New Zealand
aAddress correspondence to this author at: Canterbury Health Laboratories, Cnr Tuam Street and Hagley Avenue, Christchurch 8001, New Zealand. Fax 0064 33640 545; e-mail: Vivienne.homer{at}cdhb.govt.nz.
To the Editor:
Familial hypobetalipoproteinemia (FHBL) is commonly caused by mutations in the apolipoprotein B gene (APOB). The APOB gene encodes 2 proteins, apolipoprotein (apo) B-48 and apo B-100. Apo B-48 is formed in the intestine and is essential for the formation and recognition of dietary derived chylomicrons, and apo B-100 is found in VLDLs and LDLs of hepatic origin and is involved in the endogenous transport of triglycerides, cholesterol, and fat-soluble vitamins. A number of abnormally truncated apo B proteins have been described, and by convention are referred to by a centile system reflecting their apparent Mr in relation to apo B-100 (1).
Truncations shorter than apo B-27 are not expressed in lipoproteins, and those shorter than apo B-75 are underrepresented in LDL (2)(3)(4). Consequently homozygous mutations in the N-terminal third of APOB result in the virtual absence of both apo B-48 and apo B-100 and their corresponding lipoproteins, and thus very low concentrations of plasma triglycerides, cholesterol, and the fat-soluble vitamins. This condition is known as FHBL and is characterized clinically by failure to thrive, steatorrhea, and eventually both central and peripheral neurological abnormalities (1). Heterozygotes usually experience a milder phenotype or are asymptomatic.
We report a novel APOB mutation, identified in a family with low total cholesterol and apo B concentrations in plasma. The proband, a 64-year-old man, had an LDL cholesterol concentration of 1.4 mmol/L and an apo B concentration of 0.39 g/L, and his 2 daughters both had LDL cholesterol concentrations 
0.5 mmol/L, and apo B concentrations <0.35 g/L. In the mother, the concentrations of these analytes were within reference intervals. Western blotting of plasma from all 4 individuals showed no apo B truncations. DNA sequencing of the exons and exon/intron boundaries of the APOB gene revealed a novel heterozygous c.G1124A mutation in the proband and his 2 daughters, which was not present in the mother. No other APOB gene mutations were identified.
The c.G1124A mutation predicts a p.Ser348Asn substitution in the ß
1 domain, which is essential for lipoprotein assembly. The p.Ser348Asn substitution may affect the structure or function of this domain but is predicted to be benign, according to Polyphen (http://www.polyphen.com), with a position-specific independent counts difference score of 0.675. Alternatively the mutation at the ultimate nucleotide of exon 9 could affect splicing at the adjacent intron 9 donor splice site, with various potential splicing outcomes (Fig. 1A
).
|
Indeed, the programs SpliceView (http://l25.itba.mi.cnr.it/
webgene/wwwspliceview.html) and NNSplice (http://www.fruitfly.org/seq_tools/splice.html) predicted that the G
A mutation would abolish splicing at the normal donor splice site of intron 9, and activate a cryptic donor site 40 bp into the intron. Gene Splicer (http://www.tigr.org/tdb/GeneSplicer/gene_spl.html) also predicted abolishment of the normal donor site but did not predict the usage of a cryptic splice site.
To confirm these in silico predictions we performed minigene expression studies. A minigene construct spanning exons 8–11 and the intervening sequences was cloned into the pcDNA3.1/V5-His TOPO TA vector and then transfected into COS-7 cells. After 48 h the mRNA was isolated and reverse transcription PCR was performed. The cDNA was amplified using primers within exon 8 and exon 11. The expected 650-bp product was visualized in the wild-type, and a larger product of 690 bp was observed in the mutant (Fig. 1B
). DNA sequencing of the 690-bp and 650-bp bands revealed that the increase in size of the mutant product reflected the inclusion of the first 40 bp of intron 9. A cryptic donor splice site between c.1124 + 40 and c.1124 + 41 was activated in the mutant construct, and the normal intron 9 acceptor site was used. Predictably, this message results in a frame shift in the translated protein, a substitution of serine 348 to lysine, and the insertion of 92 new amino acids before a premature stop is encountered at residue 440 (Ser348LysfsX93). The resulting mutant protein, a truncated apo B-9.7, would not be viable for lipoprotein formation. In vivo the majority of the transcripts would be expected to use the cryptic splice site in intron 9, creating the truncated apo B-9.7 and causing the observed FHBL.
From this analysis, we have shown that the novel c.G1124A mutation causes FHBL by disrupting splicing. We identified 3 family members who were heterozygous for this mutation but were largely asymptomatic because each still had 1 normal APOB allele.
This case highlights the difficulty of interpreting novel mutations identified in diagnostic laboratories and the need for a clear strategy to determine their significance. If sufficient family members are not available, linkage analysis may be uninformative and functional analysis is essential.
Acknowledgments
Grant/funding support: National Heart Foundation of New Zealand and Foundation for Research, Science and Technology Contract grant CNTX0501 (V.M.H.).
Financial disclosures: None declared.
References
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
