|
|
||||||||
1From the Netherlands Ophthalmic Research Institute, Royal Netherlands Academy of Art and Sciences (KNAW), Amsterdam, The Netherlands; the 2Department of Clinical Genetics, Academic Medical Center, Amsterdam, The Netherlands; the 3Department of Pathology, Free University, Amsterdam, The Netherlands; the 4Department of Ophthalmology, University of Leuven, Leuven, Belgium; the 5Department of Ophthalmology, University of Amsterdam, Amsterdam, The Netherlands; the 6Institute of Epidemiology and Biostatistics, Erasmus Medical Center, Rotterdam, The Netherlands.
| Abstract |
|---|
|
|
|---|
METHODS. A combination of single-strand conformational polymorphism (SSCP), PCR, sequencing, and Southern blot analysis was used to identify mutations in the ABCC6 gene in 62 patients. Haplotypes of 16 patients with the R1141X mutation were determined with eight polymorphic markers spanning the ABCC6 locus. The effect of the R1141X mutation on the expression of ABCC6 was studied in leukocytes and cultured dermal fibroblasts from affected skin in patients heterozygous or homozygous for the R1141X mutation. ABCC6 expression was analyzed by RT-PCR and immunocytochemistry with ABCC6-specific monoclonal antibodies.
RESULTS. The ABCC6 R1141X mutation was found on 19 alleles in 16 patients with PXE and occurred in heterozygous, homozygous, or compound heterozygous form. All R1141X alleles were associated with a common haplotype, covering at least three intragenic ABCC6 markers. None of the patients or healthy control subjects had a similar ABCC6 haplotype. Furthermore, the results showed that the expression of the normal allele in R1141X heterozygotes was predominant, whereas no detectable, or very low, ABCC6 mRNA levels were found in R1141X homozygotes. Immunocytochemical staining of cultured dermal fibroblasts with ABCC6-specific monoclonal antibodies showed no evidence of the presence of a truncated protein in patients with PXE who were homozygous for R1141X.
CONCLUSIONS. A specific founder effect for the R1141X mutation exists in Dutch patients with PXE. The R1141X mutation induces instability of the aberrant mRNA. Functional haploinsufficiency or loss of function of ABCC6 caused by mechanisms, such as nonsense-mediated decay (NMD), may be involved in the PXE phenotype.
Ocular signs eventually develop in most patients with PXE. Initially, eye abnormalities consist of peau dorange or mottled hyperpigmentation of the retina. Subsequently, cometlike streaks, pinpoint white lesions of the choroid, sometimes with a hypopigmented tail in the retinal pigment epithelium, also called comets, and angioid streaks appear. Angioid streaks are cracks in the aging Bruchs membrane that most often radiate from the disc in a manner mimicking blood vessels and frequently a site of subretinal neovascular growth from the choroid. Eventually, a disciform macular degeneration develops that is frequently devastating to central vision.
Eye, skin, and cardiovascular abnormalities all appear to result from mineralization and calcification of elastic fibers in connective tissue of the affected tissues and organs, including the internal elastic lamina of Bruchs membrane.1 2
We and others localized the PXE gene to chromosome 16, location p13.13 4 5 and implicated ABCC6 gene mutations in all genetic forms of PXE.6 7 8 The ABCC6 gene, formerly called multidrug resistance protein 6 (MRP6), is a member of the adenosine triphosphate (ATP)binding cassette (ABC) transporter superfamily C and encodes a 1503-amino-acid transmembrane protein.9 The protein is composed of three hydrophobic membrane-spanning domains, 17 transmembrane spanning helices and two evolutionary conserved nucleotide binding folds (NBF1 and NBF2).10 In humans, ABCC6 is mainly expressed in liver and kidney.6 11 12 Using semiquantitative RT-PCR, we have found low ABCC6 expression levels in the retina, as well as in other tissues usually affected by PXE.6 Recently, we have developed several monoclonal antibodies against ABCC6 and localized the protein to the basolateral side of human hepatocytes and renal epithelial cells.12 Evidence was obtained that ABCC6 transports glutathione conjugates, including leukotrien-C4 (LTC4) and N-ethylmaleimide S-glutathione (NEM-GS).13 Loss of ABCC6 function associated with three mutations was found to be involved in PXE.13 However, the functional relationship between specific substrates transported by ABCC6 and the accumulation of abnormal elastic fibers in PXE remains to be elucidated.
At least 57 distinct ABCC6 mutations associated with PXE have been observed in different populations. These include nonsense, missense, and putative splice site mutations, as well as deletions and an insertion.6 7 8 14 15 Most mutations were located toward the carboxy terminal end of the protein, and formed three clusters: in the NBF1 domain, in the 8th cytoplasmic loop between the 15th and 16th transmembrane regions, and in NBF2. The frequency of mutations in the eighth cytoplasmic loop was higher than those in the NBF1 and NBF2 domains,15 which suggests that this domain is critical for normal protein function. Although the autosomal recessive form of PXE (arPXE) is the predominant form of the disease, different molecular mechanisms may be involved in various types of PXE. No clear phenotypegenotype correlation has been established for the ABCC6 gene mutations so far. The mechanism of calcification of elastic fibers in Bruchs membrane and skin is not known. Therefore, it is important to study specific ABCC6 gene mutations in more detail at both the RNA and protein levels. Such studies may contribute to the understanding of the pathologic molecular cause underlying PXE and the clarifying of the functionally important regions of the protein.
In this study, the recurrence of the R1141X mutation in 16 patients with PXE and the level of expression of ABCC6 mRNA with the R1141X mutation in (PXE) leukocytes and (PXE) fibroblasts were analyzed. In addition, the ABCC6 protein in cultured dermal fibroblasts from a patient homozygous for R1141X was studied with ABCC6-specific monoclonal antibodies.
| Materials and Methods |
|---|
|
|
|---|
The diagnosis of PXE in individuals was based on the results of dermatological and ophthalmologic examinations. Most patients had obvious skin lesions. In six patients without any obvious skin lesions, skin biopsy specimens were taken from the neck, and sections were stained by the von Kossa method. In 15 patients, recent ophthalmologic examination included visual acuity measurement, slit lamp examination, ophthalmoscopy, and often fluorescein angiography. In one patient, medical records indicated the presence of angioid streaks. In six patients, cardiovascular examinations were performed that included electrocardiograms (ECGs). Control subjects for mutational analyses were spouses of (PXE and other) patients of The Netherlands Ophthalmic Research Institute and the Academic Medical Center in Amsterdam. Definition of ethnic origin of each subject was based on the country of birth in four generations. All investigations adhered to the tenets set forth by the Declaration of Helsinki.
Mutation Analysis
Genomic DNAs were prepared from patients peripheral blood lymphocytes according to standard procedures. Primers used for polymerase chain reactions were selected from the published sequence of the human chromosome 16 BAC clone A-962B4 (GenBank Accession No. U91318; http://www.ncbi.nlm.nih.gov/Genbank; provided in the public domain by the National Center for Biotechnology Information, Bethesda, MD) and the TIGR database (http://www.tigr.org; provided in the public domain by The Institute for Genomic Research [TIGR], Rockville, MD). Fragments with a mobility shift in a single-strand conformational polymorphism (SSCP) assay were sequenced. The PCR, SSCP analysis, and cycle sequencing were essentially performed as we described previously.6 Potential intragenic deletions of genomic DNA were confirmed either by Southern blot analysis, familial segregation of CA repeats, or the consistent lack of amplification of the relevant exons. All patients with one mutant allele were analyzed further by hybridization on EcoRI-digested and PstI-digested genomic DNA using the amplified ABCC6 exons from genomic DNA as probes. In addition, the ABCC6 promoter region was sequenced up to 1 kb upstream of the first ATG. All coding sequences were analyzed in more than 100 healthy individuals, to distinguish between disease-causing mutations and polymorphic variants.
Restriction Analysis of the R1141X Mutation and Polymorphisms
After identification of the R1141X mutation and detection of other intragenic polymorphisms, optimized protocols were designed by using PCR and restriction analyses. The R1141X mutation leads to the loss of a BsiYI restriction site, which was confirmed by a restriction fragment length polymorphism assay. The BsiYI enzyme cleaves the wild-type sequence but not the mutant sequence. Similarly, the intragenic polymorphisms, 1896 C
A, 2490 C
G, and 3803 G
A were determined by the digestion of PCR fragments with enzyme HpyCH4III, HeaIII, and BstNI, respectively. The digested products were separated on 3% agarose gels. Allele frequencies for identified mutations and polymorphisms were determined both in our cohort of patients with PXE and in more than 100 control chromosomes from our healthy study participants.
Haplotype Analysis and Assessment of Founder Effect
We constructed haplotypes with eight polymorphic markers, which span
1.5 cM around the ABCC6 gene on chromosome 16 at location p13.1. These included the ABCC6 flanking markers D16S3060, 972AAAG1, 962CA2, and D16S764 as well as the intragenic markers 3803 G
A, 2490 C
G, 1896 C
A, and CA(18). Markers were analyzed on 6% polyacrylamide gels or on 3% agarose gels after enzyme digestion. We constructed (phase-known) haplotypes of all informative alleles from patients with the R1141X mutation. Control haplotypes were determined on the basis of the allele segregation within individual families. Assessments of founder haplotypes were based on
2 analyses.16
Fibroblast Cultures
For functional studies, primary fibroblast cultures were established from dermal biopsy specimens of affected skin of patients with PXE and healthy control subjects. Cells were grown in RPMI (Life Technologies, Gaithersburg, MD), containing 10% fetal calf serum, and were used between the second and fourth passages. Total RNA was isolated from subconfluent cultures.
RT-PCR Analysis of the ABCC6 Transcript
Total RNA was prepared from cultured dermal fibroblasts and peripheral blood leukocytes of patients with PXE and healthy control subjects, using the reagent (RNAzol; Cinna/Biotech Laboratories, Houston, TX). For RT-PCR analysis, 4 µg of total RNA was reverse-transcribed in the presence of oligo(dT)12-18 and 200 U reverse transcriptase (Superscript II RT; Life Technologies). Part of the RT reaction product was used as a template in a PCR reaction with ABCC6 cDNA-specific primers spanning exon 24.6 RT-PCR of ß-actin6 served as a control. In heterozygotes, the ratio of expression between the wild-type and the mutant ABCC6 allele in fibroblasts or in leukocytes was determined by cloning a mutation-specific RT-PCR product into a vector (pGEMTeasy; Promega, Madison, WI) followed by transformation of Escherichia coli strain DH5
. The presence or absence of the R1141X mutation in individual clones was checked with digestion with BsiYI, and, accordingly, the inserts of individual clones were assigned to be expression products from the mutant or wild-type allele.
Immunocytochemistry
Dermal fibroblasts were seeded onto sterile glass coverslips and cultured for an additional 24 hours. Cells were washed with PBS, fixed for 7 minutes in acetone at room temperature, incubated with primary rat monoclonal antibodies (1:10) for 1 hour (in the presence of 2% normal rabbit serum), washed, and further incubated with a Cy3-labeled goat anti-rat secondary antibody (1:400, Jackson ImmunoResearch, West Grove, PA). The slides were examined under a fluorescence microscope (Leica DMRB, Heidelberg, Germany). The monoclonal antibodies for immunocytochemical staining were M6II-7, MRPr1, and M5I-1 reactive to ABCC6, ABCC1, and ABCC5, respectively.11 12 17
| Results |
|---|
|
|
|---|
T substitution at nucleotide 3421 in the putative eighth intracellular loop. This mutation produces a stop codon at 1141 instead of an arginine residue and results in a shorter mRNA and, theoretically, in a C-terminaltruncated protein that lacks part of the transmembrane domain and the second ATP-binding domain. Previously, we found that the R1141X mutation was associated with a strong increase in the prevalence of coronary artery disease.18 In our PXE cohort, R1141X was found in 19 of 29 nondeletion alleles in DNA of 16 patients with PXE. A summary of the results is presented in Table 1 . In three patients, the R1141X mutation was observed in a homozygous form; in seven, it was in heterozygous form; and in six, it occurred in combination with other mutations in the second allele in compound heterozygous form.
|
In the three other compound heterozygotes, the combination R1141X with a del22 bp in exon 16, a 3375delT, or Q749X was identified. In family P26007, the proband showed a 22-bp deletion in exon 16, which created a premature stop codon at position 688, and which resulted in the loss of eight amino acids (Arg-Ile-Asn-Leu-Thr-Val-Pro-Glu [648-655]) of putative protein sequence. In family P26273, the proband inherited a maternal deletion of a T at cDNA position 3775 in exon 27, which resulted in a frameshift at codon 1259 and premature termination at codon 1272. The last patient, P26240, inherited the R1141X mutation and another nonsense mutation, Q749X. The latter mutation occurred in NBF1 of the ABCC6 protein.
Founder Effect for the R1141X Mutation
To determine whether the R1141X mutation originates from recurrent de novo mutational events or from founder effects, ABCC6-associated haplotypes of all patients carrying the R1141X mutation and of control subjects were constructed. Characteristics and frequency of eight markers, which spanned approximately 1.5 cM of DNA, are summarized in Table 2 .
|
A in exon 15 predicting an H632Q substitution, (2) 2490 C
G in exon 19 predicting an A830G substitution, and (3) 3803 G
A in exon 27 predicting a R1268Q substitution. These dimorphisms could easily be detected by PCR and restriction analysis (Fig. 1) . The C
A change at nucleotide 1896 in exon 15 occurred in 24 of 32 alleles (75.0%) in the patients and 80 of 204 alleles (39.2%) in the control subjects. Statistical analysis showed a significant difference between the patients and the control subjects (
2 = 32.1, P < 0.001). The second dimorphism (2490 C
G) occurred frequently in the patient group (86.2%) but rarely in the control group (14.2%). Highly significant differences were found between patients and control subjects for this dimorphism (
2 = 135.3, P < 0.0001). The dimorphism (3803 G
A) revealed no differences between the patient and control groups. Both groups had 30% wild-type alleles.
|
A)-G(2490C
G)-A(1896C
A) (Table 2) . None of 10 control subjects carried this haplotype. In 17 patients, this common haplotype (G-G-A) was extended to at least one extra marker allele, allele 4 (962CA2), distal to ABCC6. Thirteen of the latter group shared at least one intragenic marker allele, CA(18), located close to the 5' end of the ABCC6 gene, or even larger regions around the disease-related gene.
Effect of R1141X on ABCC6 Expression in Dermal Fibroblasts
The ABCC6 mRNA with the R1141X mutation encodes a C-terminally truncated ABCC6 protein that lacks part of one of the transmembrane domains and one of the ATP-binding cassette domains. To determine the effect of this mutation on the expression of the ABCC6 gene, we analyzed dermal fibroblasts from individuals homozygous or heterozygous for R1141X and healthy control subjects. The presence of the mutations was confirmed by PCR amplification of exon 24 containing the R1141X mutation followed by restriction fragment length polymorphism analysis (Fig. 2) and direct sequencing (not shown). Next, we analyzed the amount of mRNA from alleles with the R1141X mutation. RT-PCR analyses of cultured PXE fibroblasts with the R1141X in homozygous form did not contain detectable ABCC6 mRNA. Fibroblasts from those heterozygous for the R1141X mutation appeared to have a reduced level of ABCC6 mRNA compared with the healthy control subjects (Fig. 3) . The latter result indicates that the R1141X mutation affects the abundance of the mutant mRNA. To examine this in more detail, we determined the ratio of steady state transcript levels between wild-type and mutated mRNA in mononuclear blood cells and fibroblasts of patients heterozygous for the R1141X mutation. As a control for the latter analysis, we also determined this ratio in blood cells from a healthy donor and from a patient with PXE heterozygous for a missense mutation R1459C in exon 30 of ABCC6. In fibroblasts of patients heterozygous for the R1141X mutation, no mRNA containing the mutation was found, whereas in blood cells, only 5% of the ABCC6 mRNA was from the mutated allele (Table 3) . In contrast, the relative amount of mRNA carrying the R1459C mutation (48%) was very similar to that of the wild-type mRNA (52%; Table 3 ), whereas the cells of a healthy donor contained only wild-type mRNA.
|
|
|
|
In summary, in all patients homozygous or compound heterozygous for the R1141X mutation, we observed ocular and skin abnormalities and, less frequently, cardiovascular problems. However, because the expression of the disease in these tissues is highly variable among our patients, we could not correlate a distinct phenotype with the R1141X mutation.
| Discussion |
|---|
|
|
|---|
Founder Effect for the R1141X Mutation
Mutation analysis of the ABCC6 gene in patients with PXE has yielded 57 different ABCC6 mutations to date.15 The R1141X mutation was reported to be the most common mutation by us and others, especially in European patients.14 15 Recently, we also found that R1141X may be associated with a strong increase in the prevalence of coronary artery disease.18 The association between its high frequency and the geographical distribution could reflect a founder effect from a common ancestor. To test this hypothesis, we analyzed the R1141X mutation in more detail in this study.
The majority of our R1141X mutant alleles (17/19) shared a common haplotype spanning at least one ABCC6 flanking marker. Our results and statistical analysis suggested that a founder effect exists in the Dutch PXE group. In only two patients did partial aberrations of the consensus haplotype occur. These could be due to (ancient) recombination events including CA(18), 972AAAG1, and D16S764.
Identification of founder effects in the local population, as presented in this study, can greatly simplify genetic analysis of the disease, because, initially, the founder mutation can be rapidly screened in all patients. Associated clinical studies may provide further accurate information for genetic counseling and prenatal diagnosis.
Predominant Expression of the Normal ABCC6 Allele in Patients with PXE Heterozygous for the R1141X Mutation
For several mammalian mRNAs, it has been shown that a nonsense mutation or a frameshift mutation that generates a nonsense codon may greatly influence the abundance of these transcripts. A specific mechanism called nonsense-mediated mRNA decay (NMD) accelerates decay of transcripts coding for truncated proteins and thus minimizes potential metabolic damage.19 20 We found no detectable ABCC6 mRNA in patients with PXE who were homozygous for the R1141X mutation. Consistent with this observation, no ABCC6 protein was detected in cultured dermal fibroblasts of a patient homozygous for R1141X. Using a more quantitative approach, we found that in cultured dermal fibroblasts of a R1141X heterozygote, only transcripts from the wild-type allele were detected. In mononuclear blood cells of a R1141X heterozygote the mutated transcript was detected, but the abundance was reduced to 5% of total ABCC6 mRNA.
Our results suggest that the R1141X mutation induces instability of the aberrant ABCC6 mRNA, which leads to a reduced abundance of the corresponding transcript due to alterations in RNA processing by NMD. The latter mechanism may in part be an explanation of the obvious variability in the expression of the disease. The possibility that NMD contributes to a particular phenotype has also been suggested for other genes, such as fibrillin 1 in Marfan syndrome21 and ß-globin in beta zero-thalassemia.22 23 So far, because of the small size of the families, we have not established a clear correlation between the level of ABCC6 mRNA and the patients phenotype. However, it is reasonable to assume that dosage-dependent severity caused by the presence of NMD of mRNA may be involved in PXE. Complete loss of ABCC6 function causes PXE in homozygotes or compound heterozygotes, whereas partial loss of ABCC6 function in heterozygotes may result in a variable phenotype ranging from no signs at all to the complete PXE phenotype.
Features of the Phenotype
The clinical variability in PXE was demonstrated previously2 24 and also occurred in our R1141X patient cohort. In this study, we could not firmly predict the phenotype from the genotype, or vice versa. The correlation between genotype and phenotype may be obscured by several factors. The small size of our cohort limited the evaluation of the genotype and phenotype correlation. In addition, additional unknown environmental, metabolic, or genetic determinants may modify the phenotype. In future studies, we have to investigate the PXE phenotype in a thorough prospective way to obtain any significant clues for genotypephenotype relationships.
| Conclusion |
|---|
|
|
|---|
| Footnotes |
|---|
Submitted for publication September 24, 2002; accepted November 7, 2002.
Disclosure: X. Hu, None; R. Peek, None; A. Plomp, None; J. ten Brink, None; G. Scheffer, None; S. van Soest, None; A. Leys, None; P.T.V.M. de Jong, None; A.A.B. Bergen, None
The publication costs of this article were defrayed in part by page charge payment. This article must therefore be marked "advertisement" in accordance with 18 U.S.C.
1734 solely to indicate this fact.
Corresponding author: Arthur A. B. Bergen, Department of Ophthalmogenetics, The Netherlands Ophthalmic Research Institute, Meibergdreef 47, 1105 BA Amsterdam, The Netherlands; a.bergen{at}ioi.knaw.nl.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
Y. Shi, S. F. Terry, P. F. Terry, L. G. Bercovitch, and G. F. Gerard Development of a Rapid, Reliable Genetic Test for Pseudoxanthoma Elasticum J. Mol. Diagn., February 1, 2007; 9(1): 105 - 112. [Abstract] [Full Text] [PDF] |
||||
![]() |
N Chassaing, L Martin, P Calvas, M Le Bert, and A Hovnanian Pseudoxanthoma elasticum: a clinical, pathophysiological and genetic update including 11 novel ABCC6 mutations J. Med. Genet., December 1, 2005; 42(12): 881 - 892. [Abstract] [Full Text] [PDF] |
||||
![]() |
T. G.M.F. Gorgels, X. Hu, G. L. Scheffer, A. C. van der Wal, J. Toonstra, P. T.V.M. de Jong, T. H. van Kuppevelt, C. N. Levelt, A. de Wolf, W. J.P. Loves, et al. Disruption of Abcc6 in the mouse: novel insight in the pathogenesis of pseudoxanthoma elasticum Hum. Mol. Genet., July 1, 2005; 14(13): 1763 - 1773. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |