| Human Molecular Genetics | Pages |
Constitutive skipping of alternatively spliced exon 10 in the ATP7A gene abolishes Golgi localization of the Menkes protein and produces the occipital horn syndrome
Introduction
Results
Splice-donor mutation in the ATP7A gene
The exon 10-skipped alternatively spliced variant has a different subcellular distribution
Discussion
Materials And Methods
Subjects
RT-PCR direct sequencing analysis of ATP7A mRNA
Genomic DNA amplification and sequence analysis
Detection of the IVS10A+3[rarr]T mutation in the pedigree
Immunocytochemical studies
Acknowledgements
References
Constitutive skipping of alternatively spliced exon 10 in the ATP7A gene abolishes Golgi localization of the Menkes protein and produces the occipital horn syndrome
The ATP7A gene encodes a copper-transporting ATPase. Mutations in this gene result in two clinically distinct X-linked inherited disorders: Menkes disease and occipital horn syndrome (OHS). We identified a single exon skipping in the ATP7A transcript in cells from the affected proband, affected cousins and obligate carriers in a family with OHS. Genomic sequencing identified an A[rarr]T transversion at the +3 position in the splice donor site of intron 10 (gtaaagt[rarr]gttaagt) in all affected individuals and the obligate female carriers. This mutation results in the constitutive skipping of exon 10 and creates an in-frame deletion of transmembrane domains 3 and 4 (78 amino acids) in the mature transcript. The exon 10-skipped transcript is present in low amounts as an alternatively spliced product in normal individuals. Immunocytochemical assay shows that these two protein products have different subcellular distributions: the major form is concentrated in the perinuclear Golgi system while the minor form (as the only form in this family with OHS) is co-localized with the endoplasmic reticulum-resident BiP protein (GRP78). These findings indicate that endoplasmic reticulum localization only of a variant ATP7A protein is insufficient to effect normal copper transport.
INTRODUCTION
Occipital horn syndrome (OHS), previously known as X-linked cutis laxa and Ehlers Danlos syndrome (EDS) type IX, is a connective tissue disorder characterized by mild skin laxity, a characteristic facies, hernias, bladder diverticulae, varicosities and skeletal abnormalities that include occipital exostoses which give rise to the syndrome's name (1-4). Previous studies in several members of one family with OHS demonstrated low serum copper and ceruloplasmin levels, and reduced activity of lysyl oxidase, a copper-dependent enzyme which initiates cross-linking of collagen molecules and of elastin (2). The findings were similar to those in Menkes disease, an X-linked lethal neurodegenerative disorder of abnormal copper transport, and implied that these two disorders might be allelic. Recent investigations have shown that both conditions result from mutations in the same copper-transporting ATPase gene, ATP7A, which has been mapped to Xq13.3 (5-9). Menkes disease usually results from nonsense mutations, frameshifts, large deletions and mutations at the invariant splice sites that lead to frameshifts, most of which result in very low levels of the ATP7A mRNA (10,11). In contrast, most previously reported mutations in OHS permit the production of a small amount of the normal ATP7A mRNA with, in some instances, an abnormal product. The low amount of normal protein encoded by the small amount of normal mRNA could effect transfer of the cation to some copper-dependent enzymes (12-14).
We have now determined that the mutation in the ATP7A gene in the expanded four-generation family with OHS (2) results in constitutive skipping of the alternatively spliced exon 10. The mRNA lacks the sequence of exon 10 that encodes 78 amino acids of two transmenbrane domains, and the resultant protein remains in the endoplasmic reticulum (ER) rather than progressing to the Golgi. This suggests that the mutation either interferes with folding and export or deletes a Golgi `localization domain' so that most copper transfer to the apoproteins does not occur. This study now demonstrates that a mutant Menkes protein is synthesized by cells from an individual with OHS and suggests that failure to transport the protein to the Golgi contributes to the phenotype.
RESULTS
Splice-donor mutation in the ATP7A gene
The entire coding sequence of the ATP7A gene was amplified in four overlapping fragments. Three products were identical to control samples. Amplification of a cDNA fragment with primers in exons 9 and 15 generated a single band of ~600 bp from the affected proband. In contrast, the cDNA derived from control cells generated two bands, a major one of ~850 bp and a minor one of ~600 bp (Fig. 1). The sequence of the shorter fragment from the control and of the single band from the proband indicated that the entire 234 bp exon 10 was missing in each (data not shown).
Figure
No mutation was detected in the intron 9 splice acceptor site in the proband genomic DNA. There was an A[rarr]T transversion at the +3 position (gtaaagt[rarr]gttaagt) in the intron 10 donor site (data not shown). This mutation generated an MseI restriction site (Fig. 2). When a genomic DNA fragment that included this site was amplified and digested with MseI, the sample from the proband had a 231 bp band, while the two obligate carriers, the mother and the grandmother of the proband, had the 255 bp wild-type band and the 231 bp band, and normal controls had only the 255 bp wild-type band (Fig. 2). The mutation in the affected proband was identified in two affected male cousins and the expected obligated carriers, the maternal grandmother and two additional female relatives (Fig. 2). The A[rarr]T transversion at the +3 position was not seen in genomic DNA from 50 unrelated phenotypically normal females (100 copies of the ATP7A gene).
Figure
When cDNA synthesized from cells of eight unrelated males and six unrelated females was examined, each had some of the exon 10-deleted product, but it was, in all instances, the minor product (Fig. 3).
Figure
The ATP7A protein is thought to be a transmembrane copper-transporting ATPase that recently was localized in the trans-Golgi network (15-17). To determine whether the exon 10-deleted mutation product was localized in the same region, we stained dermal fibroblasts from the proband and controls with the specific antibody to the ATP7A protein derived by Dierick et al. (15). A Golgi-concentrated staining pattern was observed in the wild-type cells, similar to the previous reports (15-17); cells from the proband in this family stained diffusely across the cells, with the ATP7A protein distributed in the ER where it co-localized with a known ER-resident protein, BiP (GRP78) (Fig. 4).
Figure
The mutation in the ATP7A gene in this family abolishes the major wild-type splicing and results in constitutive skipping of exon 10 to yield the product which is normally the minor spliced mRNA. The protein product, which lacks 78 amino acids and two transmembrane domains, remains in the rough ER and is not translocated to the Golgi, the site of major accumulation of the full-length product (15-17). In normal cells, the major transcript contains exon 10 but, in all cells we tested, the minor product constitutes 10-15% of the transcript and lacks exon 10. In cells from virtually all individuals with Menkes disease and OHS, the amount of mRNA from the ATP7A gene is very low. In this instance, the mRNA is abundant but the protein is synthesized and mislocalized. It has long been hypothesized that Golgi localization requires a positive retention signal [reviewed in (18,19)]. Putative signals have been mapped to cytoplasmic and, in most cases, transmembrane domains in some Golgi-resident proteins. Although comparison of their primary amino acid sequences has not identified a consensus sequence, these type II transmembrane Golgi-resident proteins do show similar domain structures in transmembrane sequences. Interestingly, the amino acid sequence of the exon 10 of ATP7A protein contains two transmembrane domains with considerable similarity to the transmembrane domain of the [beta]1,4-galactosyltansferase (GalT) which contains a Golgi retension signal (20). The ATP7A protein, in contrast to most Golgi proteins, contains several transmembrane domains so the role of sequence or structure in one such domain is uncertain. Many proteins that are destined for export from the rough ER are retained if improperly folded. With the ATP7A protein, it is not clear if there is a `positive' Golgi localization signal that operates within the ER prior to processing or if the exon-deleted material is misfolded and so does not exit the rough ER/membrane compartment. It is not clear if the short protein is capable of transporting copper but, if so, it could move copper to the lumen of the ER. Some of the genes involved in copper transport, as well as some of the proteins that depend on copper for their activity, have been identified in yeast or mammalian cells, or both. At the cell surface, there are at least two copper-transporting proteins, in yeast designated as CTR1 (21) and FRE1 (22), one of which, CTR1, has been identified in mammalian cells (23). Copper transport into the Golgi is presumably a function of the ATP7A gene product which, as we suggest here, may also function to transport copper to the ER. A protein recently has been identified which can transport copper across the mitochondrial membrane (24), but no transporter to the peroxisome has been identified yet. Different extracellular and subcellular compartments contain various amount of copper and unique copper proteins (25,26) which serve distinct specific physiological functions. Copper-dependent proteins include the extracellular protein lysyl oxidase [cross-links collagen and elastin (2)], proteins involved in neurotransmitter function (tyrosine oxidase and dopamine [beta]-hydroxylase), proteins involved in energy metabolism [cytochrome oxidase, located in the mitochondria (27)], the free radical scavenger proteins [superoxide dismutase located in cytosol and peroxisomes (28)] and metallothioneins in cytosol, nucleus and lysosomes (29,30). The severe Menkes disease phenotype presumably represents defects in distribution of copper to all copper-dependent enzymes that transit the Golgi, whereas the milder OHS phenotypes reflect a more limited effect on copper proteins. With rare exception, Menkes disease results from deletion of part or all of the ATP7A gene or from premature termination codons which result in very low levels of mRNA (10,11). Thus little or no protein product is produced, and all Golgi-transported copper proteins should be affected. In contrast, OHS and the milder mouse mutations which result from missense mutations or exon-skipping mutations that produce some normal ATP7A mRNA may affect only some copper-dependent enzymes, especially those sensitive to copper concentration (12-14). Direct evidence from further in vitro functional assays is necessary to determine if the protein product of the ATP7A exon 10-skipped mRNA functions in copper transport. The proband, a 29-year-old Caucasian male with typical manifestations of OHS, and his four-generation kindred (2) were studied (see Fig. 3A). Skin biopsies and/or peripheral blood samples were taken from each subject with appropriate consent. Growth and maintenance of dermal fibroblasts were performed as described elsewhere (2). Lymphoblastoid lines were prepared by infection with Epstein-Barr virus. The proband (III-2) and his affected family were described in a previous publication (2). When re-examined at the age of 29 years, he had typical occipital exostoses, which developed during aldolescence, and recurrent hernias. He had developed significant orthostatic hypotension in late childhood, had lifelong frequent stools (5-10 per day) without malabsorption, and had an atonic bladder that required catheterization to void. The second boy (III-5 in Fig. 2A) was 9 years old when first seen at the clinic (2). He, too, developed chronic diarrhea, orthostatic hypotension and an atonic bladder. He died at 24 years of age in a car accident. The third boy (III-1 in Fig. 2A), a new patient in the family, was 13 years old when first seen at the clinic. Serum copper and ceruloplasmin levels had been normal at birth but dropped to borderline levels by 8 months of age. He had moderate joint laxity and soft skin, learning difficulty, easy fatiguability and frequent excercise-induced cramping of his legs and feet. He, too, had chronic diarrhea and orthostatic hypotension, but maintained bladder function. His mother died of a malignant brain tumor. The mothers (II-3 and II-5 in Fig. 2A) of the two origional patients had been mildly hyperextensible as children, but by puberty joint mobility was normal. Physical examination of these two women and the two girls (IV-1 and IV-2 in Fig. 2A) was normal. Another male, brother of the mothers (II-2), is said to have been similar in appearance to the boys. He died at ~9 months of age, apparently of pneumonia. RNA was prepared from fibroblasts using the Qiagen RNeasy mini kit (Qiagen, Chatsworth, CA), and synthesis of cDNA was performed using the BRL SuperScriptT cDNA kit (Life Technologies, Rockville, MD). Preliminary analysis of four overlapping segments of the ATP7A cDNA was performed as described elsewhere (10) except that the cDNA fragment that contained exon 10 was amplified with primers A and B (Table 1). PCR products were purified using Magic Kit (Promega, Madison, WI) and directly sequenced using SequenaseE version 2.0 DNA sequencing kits (United States Biochemical Company, Cleveland, OH).
DNA was prepared from fibroblasts or white blood cells using the QIAamp tissue kit (Qiagen, Chatsworth, CA) and was used as a substrate for amplification by PCR. The intron 9 splice acceptor site was amplified from genomic DNA with primers C and D, and the intron 10 splice donor site was amplified from genomic DNA with primers E and F. The amplified DNAs were purified, and sequenced as described above. The mutation creates an MseI restriction site (ttaa). Genomic DNA from the family members was amplified by PCR with primers E and G, digested in a 25 µl reaction mixture that contained 21.5 µl of the PCR product, 2.5 µl of the 10× NEB2 buffer and 4 U of the restriction enzyme MseI (New England Biolabs, Beverly, MA). The fragments were fractionated in an 8% polyacrylamide gel. Immunocytochemistry was carried out essentially as described previously (15,31). Cells were cultured in chamber slides (Corning Costar, Cambridge, MA) to ~70% confluence and fixed by the addition of cold (4°C) 2% paraformaldehyde (Sigma, St Louis, MO) followed by incubation at room temperature for 15 min. The cells were incubated for 1 h at room temperature in blocking solution [2% goat serum, 0.1% Tween-20 in phosphate-buffered saline (PBS)]. The previously characterized ATP7A protein polyclonal antibody (15) was diluted 1:200, and an anti-BiP monoclonal antibody (Stressgen, Victoria, Canada) was diluted 1:400 and added separately or together, and incubated for 3-5 h at room temperature. The second antibodies, fluorescein isothiocyanate (FITC)-conjugated goat anti-rabbit IgG and Texas red-conjugated donkey anti-mouse IgG (Vector Laboratories, Burlingame, CA) were added at a 1:300 and 1:200 dilution, respectively, for 40 min in the dark at room temperature. The slides were mounted using Vectashield (Vector Laboratories, Burlingame, CA) and examined with a Nikon photo microscope. We thank Drs S. Das and J. Gitschier for providing parts of the primary screening primers. We are also deeply indebted to Drs H. Dierick and T. Glover for providing the polyclonal antibody against ATP7A protein, and to Drs Lynne Smith and Robert Underwood for their support in immunocytochemical experiments and microscopy. This work was supported by NIH grant AR21557, and grants from Children's Brittle Bone Foundation and Osteogenesis Imperfecta Foundation (M.Q.).
The exon 10-skipped alternatively spliced variant has a different subcellular distribution
DISCUSSION
MATERIALS AND METHODS
Subjects
Name
Sequence
Orientation
cDNA or genomic coordinate
A
5[prime]-TATTGTGTGTACCTGTACAG-3[prime]
Sense
2300
B
5[prime]-GGAAAATCGTCTTTCAGGAAATGG-3[prime]
Antisense
3147
C
5[prime]-AGCCATCGGCCAGTGCAAT-3[prime]
Antisense
2535
D
5[prime]-TATTGTGTGTACCTGTACAG-3[prime]
Sense
2300
E
5[prime]-CGGAGGCTGGTACTTCTACA-3[prime]
Sense
2320
F
5[prime]-GAGCCTCTGATGTTTTGCCC-3[prime]
Antisense
2570
G
5[prime]-GGAAAATCGTCTTTCAGGAAATGG-3[prime]
Antisense
Intron 10 125
RT-PCR direct sequencing analysis of ATP7A mRNA
Genomic DNA amplification and sequence analysis
Detection of the IVS10A+3[rarr]T mutation in the pedigree
Immunocytochemical studies
ACKNOWLEDGEMENTS
REFERENCES
*To whom correspondence should be addressed. Tel: +1 206 543 4206; Fax: +1 206 616 1899; Email: pbyers@u.washington.edu
This page is run by Oxford University Press, Great Clarendon Street, Oxford OX2 6DP, as part of the OUP Journals
Comments and feedback: www-admin{at}oup.co.uk
Last modification: 14 Feb 1998
Copyright© Oxford University Press, 1998.
This article has been cited by other articles:
![]() |
P de Bie, P Muller, C Wijmenga, and L W J Klomp Molecular pathogenesis of Wilson and Menkes disease: correlation of mutations with molecular defects and disease phenotypes J. Med. Genet., November 1, 2007; 44(11): 673 - 688. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. Lutsenko, N. L. Barnes, M. Y. Bartee, and O. Y. Dmitriev Function and Regulation of Human Copper-Transporting ATPases Physiol Rev, July 1, 2007; 87(3): 1011 - 1046. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. Fu, Y. Jin, and L. J. Arend Smac3, a Novel Smac/DIABLO Splicing Variant, Attenuates the Stability and Apoptosis-inhibiting Activity of X-linked Inhibitor of Apoptosis Protein J. Biol. Chem., December 26, 2003; 278(52): 52660 - 52672. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. A. Palmer and A. K. Percy Neuropathology of Occipital Horn Syndrome J Child Neurol, October 1, 2001; 16(10): 764 - 766. [Abstract] [PDF] |
||||
![]() |
J. R. Forbes and D. W. Cox Copper-dependent trafficking of Wilson disease mutant ATP7B proteins Hum. Mol. Genet., August 12, 2000; 9(13): 1927 - 1935. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Schaefer and J. D. Gitlin IV. Wilson's disease and Menkes disease Am J Physiol Gastrointest Liver Physiol, February 1, 1999; 276(2): G311 - G314. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. S. Payne, E. J. Kelly, and J. D. Gitlin Functional expression of the Wilson disease protein reveals mislocalization and impaired copper-dependent trafficking of the common H1069Q mutation PNAS, September 1, 1998; 95(18): 10854 - 10859. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||






