Genetic variation at a splicing branch point in intron 9 of the low density lipoprotein (LDL)-receptor gene: a rare mutation that disrupts mRNA splicing in a patient with familial hypercholesterolaemia and a common polymorphism
Genetic variation at a splicing branch point in intron 9 of the low density lipoprotein (LDL)-receptor gene: a rare mutation that disrupts mRNA splicing in a patient with familial hypercholesterolaemia and a common polymorphism Julie C. Webb, Dilip D. Patel, Carol C. Shoulders1, Brian L. Knight and Anne K. Soutar*
MRC Lipoprotein Team and 1Molecular Medicine Group, Clinical Sciences Centre, Royal Postgraduate Medical School, Hammersmith Hospital, Du Cane Road, London W12 ONN, UK
Received May 7, 1996;Revised and Accepted June 13, 1996
Mutations in the coding sequence, splice junctions or promoter of the gene for the low density lipoprotein (LDL) receptor are known to be the underlying cause of familial hypercholesterolaemia (FH), but mutations of this type cannot be identified in all patients with a clinical diagnosis of FH. We show here that minor sequence changes elsewhere in introns can be deleterious. A minor rearrangement 30 bp upstream from the junction of intron 9 with exon 10 was detected as a heteroduplex in amplified genomic DNA from one out of 300 heterozygous FH patients. The mutation destroys the only consensus sequence for a splicing branch point in intron 9 and analysis of mRNA from cells from the patient showed that it causes retention of intron 9 or, more rarely, in the use of cryptic splice sites in exon 10. The effect of the mutation on mRNA splicing was confirmed by analysis of mRNA in cells transfected with LDL-receptor mini-gene constructs expressing exons 9 and 10, together with the normal or mutant intron 9. A common C/T polymorphism within this branch point in intron 9 of the LDL-receptor gene does not affect mRNA splicing in vitro and is not associated with significant differences in mean plasma cholesterol concentration in a healthy population.
Familial hypercholesterolaemia (FH) is caused by defects in the gene for the low density lipoprotein (LDL) receptor that affect its function and give rise to a well-characterised clinical phenotype (1 ). Attempts to identify the underlying mutation in the LDL-receptor gene in groups of patients frequently leave a number of individuals with a clear diagnosis of FH in whom there is no detectable mutation in the coding sequence, in the intron:exon junctions or in the proximal promoter region of the gene that is believed to contain all the information necessary for sterol-regulated transcription of the gene. In at least some of these cases it is probable that the defect does lie somewhere in the LDL-receptor gene, rather than in a different gene that influences its regulation or in an unrelated gene such as that for apolipoprotein B (2 ), because a particular allele of the LDL-receptor gene co-segregates with hypercholesterolaemia in the patient's family (3 ,4 ). Thus it is likely that minor deleterious sequence variations occur elsewhere in the LDL-receptor gene in FH patients, possibly in intronic regions that are either required for efficient mRNA splicing (5 ) or that are important in maintaining mRNA stability (6 ). Paucity of nucleotide sequence data for these regions has probably impeded the identification and characterization of such mutations in the LDL-receptor gene but, on the other hand, remarkably few minor mutations that lie outside the immediate intron:exon junctions in the introns of any genes have been described that cause any inherited human disorders (5 ). In this paper we describe a minor rearrangement in intron 9 of the LDL-receptor gene that results in defective mRNA splicing because it destroys the putative branch point (7 ).
In addition to rare defects due to the many different deleterious mutations in the LDL-receptor gene, it is widely believed that more common genetic variation in the gene in apparently normolipaemic individuals might affect LDL-receptor function sufficiently to influence plasma cholesterol concentration but not to result in the severe clinical manifestations of FH. For example, in some populations it has been possible to show an association between plasma cholesterol concentration and inheritance of the PvuII polymorphism in intron 15 of the LDL-receptor gene, suggesting that this polymorphism might be in linkage disequilibrium with a functional variant (8 ). The characterisation of the mutation described in this paper demonstrates that a common polymorphism in intron 9 of the LDL-receptor gene lies within the consensus sequence for a splicing branch point (9 ) and therefore we investigated whether it is associated with differences in plasma cholesterol concentration in the population.
During screening of genomic DNA from a group of 300 patients with a diagnosis of familial hypercholesterolaemia for known mutations in the LDL-receptor gene, the PCR product of exons 9 and 10 amplified together was analysed by polyacrylamide gel electrophoresis to detect heteroduplexes caused by a 4 bp deletion in exon 9 (10 ). One sample was found to have an unusual pattern (Fig. 1 A) and nucleotide sequencing of cloned PCR products revealed that the patient was heterozygous for a minor rearrangement in intron 9 approximately 25 bp upstream from the junction of intron 9 with exon 10, in which 9 bp are replaced with five apparently unrelated bases (Fig. 1 B). The remainder of the sequence of intron 9 and of both intron:exon junctions in the mutant allele was the same as that in normolipaemic individuals [Fig. 1 C and (11 )]. Comparison of the normal and mutant sequences for intron 9 with the known consensus sequence for eukaryotic splicing branch points (9 ) suggested that the mutation involved the only putative branch point sequence present in intron 9 and might, therefore, affect mRNA splicing (Fig. 1 D). The mutation also encompasses the site of a common polymorphism, a C to T transition at position -30 from the start of exon 10 that is detected as the presence (C) or absence (T) of a cutting site for the restriction enzyme HhaI (Fig. 1 C and D). We had already identified the polymorphism in individuals of different ethnic origin, including European, Chinese, Afro-Caribbean and Asian Indian (Webb, Sun and Soutar, unpublished observations) and it has recently been described to occur with a frequency of 0.44 for the T allele in 92 unrelated individuals in Denmark (12 ). The normal allele in our patient with the branch point mutation was the slightly less common T allele.
To determine whether the mutation affected LDL-receptor mRNA splicing, mRNA was isolated from cultured Epstein-Barr-Virus (EBV)-transformed lymphoblasts from the patient and from normolipaemic individuals. A segment of the LDL-receptor mRNA from mid exon 9 to mid exon 11 was amplified by RT-PCR with 32P-labelled primers and the products analysed by denaturing PAGE (8%), as shown in Figure 2 A. As expected, a single major product of 503 bp was present in cells from the normolipaemic individual, but there were two major bands of similar intensity present in the cells from the patient heterozygous for the mutation in intron 9. One of the bands corresponded to the normal fragment, while the other fragment was larger by 81 bp, suggesting that the mutant intron 9 had been retained in the mRNA. Minor bands of 449, 348 and greater than 600 bp were also visible when the gel was over-exposed.
To confirm that the base changes in the branch point in the intron were the underlying cause of defective splicing, mammalian expression vectors were constructed in plasmid pcDNA-3 (Invitrogen) that comprised exon 9, intron 9 and exon 10 from the normal and mutant alleles of the LDL-receptor gene from the patient and from an individual who was homozygous for the presence of the HhaI cutting site, under the transcriptional control of the promoter from the immediate early gene of the human cytomegalovirus (CMV) and employing the bovine growth hormone termination and polyadenylation signals. The constructs also contained the neomycin resistance (NeoR) gene under the control of the SV40 early promoter, which provided an internal control for the efficiency of transfection and expression. A diagram of the constructs is shown in Figure 3 A.
Although there was no obvious effect of the HhaI polymorphism in intron 9 on mRNA splicing in vitro, we investigated whether or not it might have any detectable effect on LDL-receptor function in vivo by examining the mean plasma cholesterol concentration in individuals of different HhaI genotype in a group of 261 healthy English men who had attended a routine health screening programme. Exons 9 and 10 were amplified from genomic DNA, digested with HhaI and the products analysed by agarose gel electrophoresis (Fig. 4 ). The frequency of the T allele was 0.45, and the distribution of alleles was in Hardy-Weinberg equilibrium. As shown in Table 1 , no significant differences in age- and body mass index (BMI)-adjusted fasting total cholesterol, LDL-cholesterol or high density lipoprotein (HDL)-cholesterol concentration or in plasma triglyceride concentrations were observed between heterozygous individuals and those homo- zygous for either allele.
There is little doubt that the minor rearrangement in intron 9 of the LDL receptor gene is the cause of defective mRNA splicing seen in the patient's cells and of FH in the patient, presumably because the mutation abolishes the branch point consensus sequence. This mutant allele is rare in the population of FH patients in the London area, as it was found in a single individual in the group of 300 that we have been screening routinely for newly-identified mutations. Eukaryotic mRNA splicing is normally initiated by simultaneous cleavage at the 5' (donor) splice site and formation of a 2',5' phosphodiester bond between an invariant adenine at the branch point and the guanosine residue at the 5' end of the intron to form a lariat structure, and is followed by cleavage at the 3' (acceptor) splice site and ligation of the exons (7 ). The destruction of the branch point in the LDL-receptor gene in this patient, and particularly in the loss of the adenine base, appears to result in failure to splice out the intron, although small amounts of products in which alternative cryptic splice sites in exon 10 are used were also detected. In cells from the heterozygous patient, the amount of mRNA containing the mutant intron was only slightly less than the amount of mRNA from the normal allele, suggesting that retention of intron 9 had little effect on the stability of the mRNA.
Mutations at a branch point have not been commonly described as a cause of human disease, and this is one of the first to be characterised fully. This may be partly because there is considerable flexibility in the eukaryotic branch point sequence (9 ), and because sequence analysis of many genes has not been extended sufficiently far into the introns to identify a putative branch point that may lie as much as 60 bases into the intron. Deletion of a branch point in the androgen receptor gene has been suggested to be the probable cause of a case of X-linked androgen insensitivity syndrome. The mutation was a large deletion in an intron that left 18 bp of normal sequence intact at the 3' end of the intron, including the intron:exon junction, and approximately 3 kb at the 5' end of the intron. Analysis of mRNA by RT-PCR showed the presence of two products, a major one (92%) in which the downstream exon was skipped and a minor one that was spliced normally, apparently through use of a cryptic branch point (14 ). On the other hand, a point mutation in a putative branch point of an intron in the gene for the neural cell adhesion molecule L1 that gives rise to X-linked hydrocephalus gives rise to a defectively-spliced mRNA in which an alternative splice site upstream from the normal site is used (15 ). A recent abstract (16 ) describes a point mutation (T to C) in an intron of the gene for lecithin:cholesterol acyltransferase (LCAT) at a position 21 bp upstream from the intron: exon junction that results in retention of the intron. Thus the precise defect in mRNA splicing resulting from a branch point mutation varies considerably. It is known that only a small number of defined intermediates are formed during pre-mRNA processing (17 ), suggesting that there is a preferred order in which the introns in a gene are removed. However, this order is not necessarily numerical and we tentatively suggest that retention of the intron in the mRNA in a gene with a branch point mutation, as in the case of intron 9 of the LDL-receptor gene or intron 4 of the LCAT gene, occurs when the mutant intron is spliced later than other neighbouring introns and if there are no alternative branch points or cryptic splice sites present in the defective intron. This is probably more likely when the intron is small, as in the case of the retained intron in the LDL-receptor (81 bp) or LCAT mRNA (83 bp) (18 ).
Two other genetic disorders involving mutations in introns at some distance from the intron:exon junction have been described in which single base substitutions that lie in the polypyrimidine tract between the branch point and the splice junction result in defective splicing (19 ,20 ). Clearly, the number of genetic diseases known to be caused by mutations in introns other than at the exact intron:exon junction will increase as more information about the nucleotide sequence of introns becomes available, and strategies for screening for disease-causing mutations will need to take this into account.
The observation that the mutation in this patient interfered with splicing raised the interesting possibility that a common C to T polymorphism in the LDL-receptor gene that lay within the branch point consensus sequence might affect mRNA splicing sufficiently to have a small but significant effect on LDL receptor activity. However, we failed to detect any affect of the polymorphism on splicing in vitro or on plasma lipid concentrations in the population. On the one hand, this was not surprising because the consensus sequence for a mammalian branch point contains either C or T in the relevant position (see Fig. 1 ). However, the mutation in intron 4 of the LCAT gene that results in retention of intron 4 was also reported to be a C to T transition, at position -21 from the start of exon 5. This places it in the central position of the 7 bp consensus, changing the sequence from CCCTGAC to CCCCGAC, and this site can also be either C or T in the consensus.
The index patient is a 38 year old white female with a diagnosis of heterozygous familial hypercholesterolaemia based on a total plasma cholesterol concentration before treatment of 11.1 mmol/l and plasma triglyceride of 1.22 mmol/l; her HDL-cholesterol concentration was 1.30 mmol/l. She has tendon xanthomas but no arcus or overt CHD. Her father is hypercholesterolaemic and has CHD; a paternal uncle suffered a fatal myocardial infarction aged 35 years, a paternal aunt has had CHD since the age of 50 years and her paternal grandfather died aged 50 years. Blood samples were not obtainable from these or other family members.
The control group comprised 261 healthy white male volunteers aged 21 to 60 years (mean age 44.93 +- 0.57) of English parentage and grandparentage who attended a BUPA health screening programme in London during 1991. Following consent, fasting blood samples were obtained for isolation of DNA and for routine plasma lipid analysis. Total cholesterol, LDL-cholesterol and HDL-cholesterol values were adjusted for age and BMI. Association between the HhaI polymorphism and plasma lipid levels was evaluated by the Student's t-test.
Genomic DNA was isolated from whole blood from the index patient and a fragment encompassing exons 9 and 10 of the LDL receptor gene was amplified with primers 537 and 538 and analysed by PAGE as described in detail elsewhere (10 ). Genomic DNA from the control group was isolated as described previously (21 ). Amplification, enzyme digestion and analysis of the PCR products from the control group was carried out in microtiter plates as described before (10 ).
Total RNA was isolated from EBV-transformed lymphoblasts that had been pre-incubated with lipoprotein-deficient serum for 48 h and LDL-receptor mRNA was amplified by RT-PCR (22 ). A fragment of the cDNA from mid exon 9 to mid exon 11 was amplified with primers G and H (5'-CCTGAGGAACGTGGTCGCTCT and 5'-CCCCCATTGACATCGATGCTT); PCR primers were end-labelled with 32P (23 ) where indicated in the text.
PCR products comprising exons 9 and 10, with intron 9, were cloned into a plasmid T-vector and then the insert was excised as a SphI/SacI fragment and cloned into the polylinker of a mammalian expression vector, plasmid pcDNA3 (Invitrogen). The complete nucleotide sequence of each insert in the large scale preparation of each pcDNA3 construct was confirmed. The plasmid constructs were transfected as DEAE-dextran complexes (24 ) into COS cells as described previously (25 ). Total RNA was isolated from the cells by the method of Okayama et al., (26 ) to minimize plasmid DNA contamination and treated with RNAse-free DNAse, (Promega RQ1) as recommended by the supplier, to remove any remaining traces of DNA. The mRNA transcribed from the LDL-receptor gene insert was amplified by RT-PCR with primers 537 and 538 as described above. The mRNA from the NeoR gene product was amplified with primers AKS-249 and AKS-250 (5'-AAGCGGGAAGGGACTGGC and 5'-AGGCGATAGAAGGCGATGC). Control reactions in which RT was omitted from the first step were included to confirm the absence of plasmid DNA.
We are indebted to Professor G. R. Thompson for access to clinical material, and are grateful to Sister C. Neuwirth and the late Sister S. N. McCarthy for their help in obtaining blood samples and information from FH patients. We are also grateful to Dr Ann Hale and her colleagues at BUPA for blood samples and blood lipid values from the healthy control population.
BMI, body mass index; CMV, cytomegalovirus; EBV, Epstein-Barr-Virus; FH, familial hypercholesterolaemia; HDL, high density lipoprotein; LCAT, lecithin:cholesterol acyltransferase; LDL, low density lipoprotein.
1 Goldstein, J.L., Hobbs, H.H. and Brown, M.S. (1995) Familial Hypercholesterolaemia. In Scriver, C.R., Beaudet, A.L., Sly, W.S. and Valle, D. (eds), The Metabolic and Molecular Bases of Inherited Disease, McGraw-Hill, New York, Vol. II, pp. 1981-2030.
2 Tybjaerg, H.A., Gallagher, J., Vincent, J., Houlston, R., Talmud, P., Dunning, A.M., Seed, M., Hamsten, A., Humphries, S.E. and Myant, N.B. (1990)Familial defective apolipoprotein B-100: detection in the United Kingdom and Scandinavia, and clinical characteristics of ten cases. Atherosclerosis, 80, 235-242.
3 Fischer, H.J., Schuster, H., Keller, C., Wolfram, G. and Zollner, N. (1991)Identification of a 76-year-old patient with compound heterozygous familial hypercholesterolemia by haplotype analysis of the LDL receptor gene. Klin. Wochenschr., 69, 842-846.MEDLINE Abstract
4 Schuster, H., Rauh, G., Gerl, C., Keller, C., Wolfram, G. and Zollner, N. (1991)Use of DNA haplotype analysis in diagnosis of familial hypercholesterolaemia in 31 German families. J. Med. Genet., 28, 865-870.MEDLINE Abstract
5 Cooper, D.N. and Krawczak, M. (1993) Splice site mutations. In Human Gene Mutation. BIOS Scientific Publishers Ltd, Oxford, pp. 239-260.
6 Ross, J. (1988)Messenger RNA turnover in eukaryotic cells. Mol. Biol. Med., 5, 1-14.MEDLINE Abstract
8 Pedersen, J.C. and Berg, K. (1990)Gene-gene interaction between the low density lipoprotein receptor and apolipoprotein E loci affects lipid levels. Clin. Genet., 38, 287-294.MEDLINE Abstract
9 Krainer, A.R. and Maniatis, T. (1988) RNA splicing. In Hames, B.D. and D.M. Glover, D.M. (eds), Transcription and Splicing. IRL Press, Oxford, pp. 131-206.
10 Webb, J.C., Sun, X.-M., McCarthy, S.N., Neuwirth, C., Thompson, G.R., Knight, B.L. and Soutar, A.K. (1996)Characterization of mutations in the low density lipoprotein (LDL)-receptor gene in patients with familial hypercholesterolaemia (FH) and frequency of these mutations in FH patients in the United Kingdom. J. Lipid Res., 37, 368-381.MEDLINE Abstract
11 Sudhof, T.C., Goldstein, J.L., Brown, M.S. and Russell, D.W. (1985)The LDL receptor gene: a mosaic of exons shared with different proteins. Science, 228, 815-822.MEDLINE Abstract
12 Nissen, H., Hansen, A.B.B. and Jensen, H.K. (1995)A frequent HhaI polymorphism in intron 9 of the low density lipoprotein receptor gene detected by the denaturing gradient gel electrophoresis technique. Clin. Genet., 48, 221-222.MEDLINE Abstract
13 Hobbs, H.H., Russell, D.W., Brown, M.S. and Goldstein, J.L. (1990)The LDL receptor locus in familial hypercholesterolemia: mutational analysis of a membrane protein. Annu. Rev. Genet.,24, 133-170.MEDLINE Abstract
14 Ris-Stalpers, C., Verleun-Mooijman, M.C., de-Blaeij, T.J., Degenhart, H.J., Trapman, J. and Brinkmann, A.O. (1994)Differential splicing of a human androgen receptor pre-mRNA in X-linked Reifenstein syndrome, because of a deletion involving a putative branch site. Am. J. Hum. Genet., 54, 609-617.MEDLINE Abstract
15 Rosenthal, A., Jouet, M. and Kenwrick, S. (1992)Aberrant splicing of neural cell adhesion molecule L1 mRNA in a family with X-linked hydrocephalus. Nature Genet., 2, 107-112.MEDLINE Abstract
16 Kuivenhoven, J.A., Wiebush, H., van Tol, A., Senz, J., Pritchard, H., Funke, H., Assmann, G. and Kastelein, J.P. (1995)A non-splice site intronic point mutation interferes with correct mRNA processing. A genetic basis for Fish-eye disease. Circulation, 92, Suppl 1, I-426, Abstract 2033.
17 Nordstrom, J.L., Roop, D.R., Tsai, M.-J. and O'Malley, B.W. (1979)Identification of potential ovomucoid mRNA precursors in chick oviduct nuclei. Nature, London, 278, 328-331.
18 McLean, J., Wion, K., Drayna, D., Fielding, C.J. and Lawn, R.L. (1987)Human lecithin-cholesterol acyltransferase gene: complete sequence and sites of expression. Nucleic Acids Res., 14, 9397-9406.
19 Higashi, Y., Tanae, A., Inoue, H., Hiromasa, T. and Fujii-Kuriyama, Y. (1988)Aberrant splicing and missense mutations cause steroid 21-hydroxylase. Proc. Natl Acad. Sci. USA, 85, 7486-7490.MEDLINE Abstract
20 Tee, M.-K., Lin, D., Sugawara, T., Holt, J.A., Guigen, Y., Buckingham, B., Strauss, J.F.I. and Miller, W.L. (1995)T to A transversion 11 bp from a splice acceptor site in the human gene for steroidogenic acute regulatory protein causes congenital lipoid adrenal hyperplasia. Hum. Mol. Genet., 4, 2299-2305.MEDLINE Abstract
21 Shoulders, C.C., Narcisi, T.M.E., Jarmuz, A., Brett, D.J., Bayliss, J.D. and Scott, J. (1993)Characterization of genetic markers in the 5'-flanking region of the apoAI gene. Hum. Genet., 91, 197-198.MEDLINE Abstract
22 Sun, X.M., Patel, D.D., Bhatnagar, D., Knight, B.L. and Soutar, A.K. (1995)Characterization of a splice-site mutation in the gene for the LDL receptor associated with an unpredictably severe clinical phenotype in English patients with heterozygous FH. Arterioscl. Thromb. Vasc. Biol.,15, 219-227.
23 Maniatis, T. (1989) Labelling of synthetic oligonucleotides by phosphorylation with bacteriphage T4 polynucleotide kinase. In J. Sambrook, J., Fritsch, E.F. and Maniatis, T. (eds), Molecular Cloning: A Laboratory Manual, 2nd edition. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y., pp. 11.31 - 11.33.
24 Levinson, A.D. (1990)Expression of heterologous genes in mammalian cells. Methods Enzymol., 185, 507-508.
25 Sun, X.-M., Patel, D.D., Webb, J.C., Knight, B.L., Fan, L.-M., Cai, H.-J. and Soutar, A.K. (1994)Familial hypercholesterolemia in China: Identification of mutations in the LDL receptor gene that result in a receptor negative phenotype. Arterioscl. Thromb., 14, 85-94.
26 Okayama, H., Kawaichi, M., Brownstein, M., Lee, F., Yokota, T. and Arai, K. (1987)High-efficiency cloning of full-length cDNA; construction and screening of cDNA expression libraries for mammalian cells. Methods Enzymol., 154, 3-28.MEDLINE Abstract
27 Friedewald, W.T., Levy, R.I. and Fredrickson, D.S. (1972)Estimation of the concentration of low density lipoprotein in plasma without use of the preparative ultracentrifuge. Clin. Chem., 18, 499-502.MEDLINE Abstract
*To whom correspondence should be addressed
This page is maintained by OUP admin. Last updated Thu Oct 31 15:27:05 GMT 1996. Part of the OUP Journals World Wide Web service.Copyright Oxford University Press, 1996
