CA2108927C - Wilson disease gene - Google Patents

Abstract

Wilson disease (hepatolenticular degeneration) is an autosomal recessive disorder of copper transport, resulting in copper accumulation and toxicity to the liver and brain. The gene (locus WND) has been mapped to chromosome 13 band q14.3. On three overlapping yeast artificial chromosomes (YACs) from this region, a sequence similar to the proposed copper binding motifs of the putative ATPase (MNK) defective in Menkes disease was identified. It was shown that this sequence forms part of a P-type ATPase gene (Wc1) that is very similar to MNK, with at least six putative metal binding domains homologous to those found in prokaryotic heavy metal transporters. This gene lies within a 300 kb region that has been identified as a likely location for WND. The gene is expressed in the liver and kidney and is a candidate for Wilson disease.

Description

~ 2108927 BACKGROUND OF THE INVENTION

Copper is an essential trace metal for prokaryotes and eukaryotes, and is a required component for a variety of enzymes, including cytochrome oxidase and other electron transport pro-teins. Dietary intake of copper generally far exceeds the trace amounts required, and organisms have evolved effective means for the elimination of the excess. Toxicity of copper is believed to act predominantly through the formation of highly reactive hydrox-ide radicals, which can damage cell membranes, mitochondria, pro-teins and DNA1.

Copper homeostasis requires appropriate mechanisms for copper absorption, cellular transport, incorporation into protein, storage, and excretion. In mammalian systems, various proteins or peptides have been recognized for these functions2: albumin (and copper histidine) for copper transport in the blood, ceruloplasmin as a possible copper donor to tissues and enzymes3, and metallo-thionein for intracellular copper storage4. The mechanism of copper efflux from tissues has remained an enigma.

Menkes disease and Wilson disease are both caused by a disruption in copper transport (see review5). However, these two diseases affect different tissues. In X-linked Menkes disease, copper export is defective in many tissues5 but is normal in the liver6. Copper enters into the intestinal cells, but is not transported further, resulting in severe copper deficiency. In contrast, Wilson disease is characterized by failure to incor-porate copperinto ceruloplasmin in the liver, and failure to excrete copper from the liver into bile. This results in toxic accumulation of copper particul.arly in the liver, and also in kidney, brain, and cornea. The resulting liver cirrhosis and/or progressive neurological damage has an age of onset from childhood to early adulthood. Consequently, there is a real need to identi-fy the gene responsible for Wilson disease in order t:o develop new diagnostic and therapeutic strategies useful in the detection and treatment of the disease.

SUMMARY OF THE INVENTION

The Wilson disease gene (WND) has been assigned to a single locus by genetic linkage first to esterase D7, then to a cluster of polymorphic markers on chromosome 13 band q14.38,9,10.
Multipoint linkage analysis indicates that WND is flanked proxi-mally by marker D13S31 and distally by D13S59 at distances of 0.4 cM and 1.2 cM respectively11. Marker D13S31 is sufficiently close to show allelic association with the disease in two different populations12,13.

The inventors have isolated new markers between D13S59 and D13S31 and have used them to construct a long range restric-tion map of the WND region14,15. Three CA repeat markers, D13S314, D13S133 and D13S316, have been positioned in a 300 kb region within this map16. These markers show high allelic association with WND and allowed the identification of specific Wilson disease haplotypes in the region16. In addition D13S314 was used to define the proximal boundary of the Wilson disease region using a recombination event that is present in one of the Wilson disease families tested.

To try and isolate the Wilson disease gene, a probe from the proposed copper binding region of Menkes (MNK) was hybridized at low stringency to 19 YACs isolated from a 1-1.5 Mb region immediately distal to marker D13S31.

This strategy has lead to the isolation of a gene with all the characteristics required for a copper transporting ATPase, which is predicted to be the gene for Wilson disease.
Accordingly, the present invention provides a DNA
sequence containing the gene for Wilson disease.

More specifically, the present invention provides a nucleic acid molecule comprising the DNA sequence as illustrated in Figure 10, the complementary sequence thereof or an allelic variant thereof.

The complete sequence of the metal binding ATPase defective in Wilson disease forms part of the present invention. The DNA sequence includes six copper binding domains, a phosphate domain, a transduction, transmembrane, phosphorylation and ATP binding domains.
The sequence of each of these domains forms part of the invention. The 5' and 3' untranslated regions and stated intron sequences are included in the application.

The present invention provides a nucleic acid molecule comprising the underlined DNA sequence designated as Cul in Figure 10.
The present invention provides a nucleic acid molecule comprising the underlined DNA sequence designated as Cu2 in Figure 10.
The present invention provides a nucleic acid molecule comprising the underlined DNA sequence designated as Cu3 in Figure 10.

The present invention provides a nucleic acid molecule comprising the underlined DNA sequence designated as Cu4 in Figure 10.
The present invention provides a nucleic acid molecule comprising the underlined DNA sequence designated as Cu5 in Figure 10.
The present invention provides a nucleic acid molecule comprising the underlined DNA sequence designated as Cu6 in Figure 10.
The present invention provides a nucleic acid molecule comprising the underlined DNA sequence designated as Pt/T in Figure 10.
The present invention provides a nucleic acid molecule comprising the underlined DNA sequence designated as Tm in Figure 10.
The present invention provides a nucleic acid molecule comprising the dotted underlined DNA sequence designated as Ph in Figure 10.
The present invention provides a nucleic acid molecule comprising the underlined DNA sequence designated as ATP-hinge in Figure 10.
The present invention also provides use of a nucleic acid molecule comprising the DNA sequence for Wilson disease, the complementary sequence thereof or an allelic variant thereof (including the above-mentioned domains) to detect Wilson disease.
The present invention further includes the use of nucleic acid molecule comprising the DNA sequence for Wilson disease or an allelic variant or any fragment of the given DNA
sequence, derived in any way, including amplification by the polymerase chain reaction, for the diagnosis of Wilson disease (hepatolenticular degeneration) or the heterozygous state, or for the identification of mutations. This might include such methods as direct sequencing of PCR amplified fragments, or the examination of differences in small amplified fragments using such methods as single strand conformation polymorphism (SSCP) or heteroduplex analysis.
Thus, the present invention provides a use of a primer to detect a mutation in the Wilson disease gene in a Wilson disease patient, wherein said primer comprises 5' TGT AAT CCA
GGT GAC AAG CG 3' or 51 CAC AGC ATG GAA GGG AGA G 3'.
The present invention also includes the use of a nucleic acid molecule comprising the DNA sequence for Wilson disease or an allelic variant or any fragment of the given DNA
sequence, derived in any way, including amplification by the polymerase chain reaction, for use in plasmids or any other vector for therapy of Wilson disease or Menkes disease (another disorder of copper transport). This includes short term use of plasmid containing any part of the sequence in this application as initial therapy for rapid removal of copper, in the early phases of treatment, when patients are at risk from hemolysis and other complications from rapid release of copper. The application also includes use for enhancing heavy metal transport in humans or any other organism.
Thus, the present invention provides use of a nucleic acid molecule comprising the DNA sequence for Wilson disease, the complementary sequence thereof, or an allelic variant thereof to treat Wilson disease.
The present invention further includes the use of a nucleic acid molecule comprising the DNA sequence for Wilson disease or an allelic variant or any fragment of the given DNA sequence, derived in any way, including amplification by the polymerase chain reaction, to obtain portions of the Wilson disease gene for deriving the homologous gene from other organisms. Specifically, this includes the use of the described nucleic acid molecule to obtain the homologous gene in the rat, for the study of the Long-Evans Cinnamon (LEC) mutant, and the mouse gene, for the study of the toxic milk mouse, both of the above as potential models for Wilson disease.
The present invention yet also includes the use of a nucleic acid molecule comprising the DNA sequence for Wilson disease or an allelic variant or fragment thereof, derived by any means such as cloning or PCR amplification to obtain the corresponding gene from any other species.
Thus, the present invention provides use of a nucleic acid molecule comprising the DNA sequence for Wilson disease, the complementary sequence thereof, or an allelic variant thereof to isolate the Wilson disease gene, or fragment thereof, from a mammal.
The present invention further includes the use of a nucleic acid molecule comprising the DNA sequence for Wilson disease or an allelic variant or fragment thereof in therapy to remove heavy metal from an organ.
The present invention yet also includes the use of a nucleic acid molecule comprising the DNA sequence for Wilson disease, or an allelic variant or fragment thereof in animal breeding, for example to enhance the excretion of copper and other heavy metals.
The present invention provides a use of a nucleic acid molecule comprising the DNA sequence of Figure 10, the complementary sequence thereof or an allelic variant thereof to reduce metal toxicity in an animal.
The invention also includes all of the DNA markers associated with the Wilson disease that the inventors have developed.
In particular, the present invention includes -5a-DNA markers D13S314, D13S315 and D13S316 as well as any other markers that detect the same dinucleotide repeat polymorphisms as these three markers.
The present invention provides a DNA marker associated with the gene for Wilson disease characterized in that it detects the same dinucleotide repeat polymorphism as DNA
marker D13S314, wherein said marker can be amplified using primers comprising sequences 5' GAG TGG AGG AGG AGA AAA GA
3' and 5' GTG TGA CTG GAT GGA TGT GA 3'; a DNA marker associated with the gene for Wilson disease characterized in that it detects the same dinucleotide repeat polymorphism as DNA marker D13S315, wherein said marker can be amplified using primers comprising the sequences 5' GCC ATC CAG AGT TAA ACC A 3' and 5' TTA TAG CTT TTC TCA
TGC ATT C 3'; and a DNA marker associated with the gene for Wilson disease characterized in that it detects the same dinucleotide repeat polymorphism as DNA marker D13S316, wherein said marker can be amplified using primers comprising the sequences 5' GCA GCA ATG CTT TGT
GCA TAA 3' and 5' TGT TTC CCA CCA ATC TTA CCG 3'.

The invention further includes the use of the above-described DNA makers to detect Wilson disease.

The invention further provides a diagnostic kit for detecting Wilson disease comprising at least one pair of primers selected from the group consisting of a) 5' GAG TGG AGG AGG AGA AAA GA 3' and 5' GTG TGA CTG GAT GGA TGT GA 3';
b) 5' GCC ATC CAG AGT TAA ACC A 3' and 5' TTA TAG CTT TTC TCA TGC ATT C 3'; and c) 5' GCA GCA ATG CTT TGT GCA TAA 3' and 5' TGT TTC CCA CCA ATC TTA CCG 3'; and instructions for detecting Wilson disease.

- 5b -~ 2108927 BRIEF DESCRIPTION OF THE DRAWINGS

Fig. 1 Schematic map of the WND candidate region. The positions of the markers relative to the YACs are indicated.
D13S314 and D13S315 were derived from cosmids identified by end-clones of 27D8 and 235H9 respectively. D13S316 was derived from a cosmid identified by D13S196. The established flanking markers D13S31 and D13S59 are also shown. A physical map of this region has been constructed (Bull and Cox, 1993). YAC isolation and characterization has been described elsewhere (Bull et al. sub-mitted).

Fig. 2 Statistically significant allele distributions.
The number of chromosomes carrying a specific allele is shown in white for normal and black for WND chromosomes.

Fig. 3 Isolation of Wcl.a) Restriction maps. At the top is a long range restriction map around the proximal flanking marker for WND, D13S31. Wcl and markers D13F71S1, D13S196 and D13S31 map to the intervals shown with thick black bars. The location of three microsatellite markers are shown with asterisks.
At the bottom are restriction maps of cosmids cosL and cosJ. cDNA

clones were mapped to the intervals shown with thick bars.
Restriction sites are : M=Mlu I; R=Nru I; N=Not I; H=HindIII;
E=EcoRI; X=XbaI. MluI and NotI sites within YAC 27D8 are shown above and below the line respectively. Three other YACs mentioned in the text (95C3, 53C12 and 68F4) were isolated using PCR primers specific to the right (proximal) end of 27D8. b) Hybridization of probes Mcl.a and cosL.d to YACs from the Wilson disease region.
YACs 232H4 and 68F4 are overloaded.

Fig. 4 Chromosome mapping of cDNA fragments. a) Each cDNA was hybridized at high stringency to EcoRI and HindIII
digests of genomic DNA from Human, hybrid ICD, YAC 95C3 and Ham-ster. The pattern obtained by hybridization of cDNA clone Wcl.i7 to EcoRI digests of these samples is shown. This clone cross hybridized to a single hamster fragment (indicated by an asterisk). b) Linkup of clone Wcl.i7 with marker D13S31. M1uI
and Nru I digests of the chromosome 13 hybrid ICD were separated with a CHEF PFGE system (Biorad) through 1% agarose with a 300-2000s ramped pulse at 50V for 150h. Sizes (in kb) of detected fragments are indicated on the left. Cross hybridization of Wc(i7) to the hamster component of ICD is indicated with an asterisk.

Fig. 5 Partial DNA sequence of Wcl. Alignment of the amino acid sequences of Wcl with MNK is indicated, amino acid identities in MNK are indicated with a period. insertions and deletions in MNK compared to Wcl are indicated as is the position of a splice site that occurred in one of the cDNA clones. The amino acid numbers of MNK are shown.

Fig. 6 Alignment of cDNA fragments with Mc1. The coding region of the Mc1 cDNA is represented at the top of the figure by a thick stippled line. The location of functional elements are indicated. The relative positions of probes Mcl.a and Mcl.b are shown below with narrow stippled lines. The rela-tive positions of cDNA clones from Wcl are indicated with thin black lines. The Thick black line shows the part of the Wcl transcript that has been sequenced. Clone Wcl.f8 contains an unspliced intron. This is indicated with a dashed line.

Fig. 7 Alignment of nucleotide and derived amino acid sequence of the metal binding domains of Mc1 and Wcl. The nucleo-tide sequence of each domain of Wcl is shown. A translation of the domains and the corresponding domains of Mcl are shown direct-ly below. Nucleotides and amino acids that are conserved in Mc1 are underlined. Residues that are also conserved in bacterial metal binding domains (see Fig. 6) are indicated at the bottom of the figure with an asterisk. Nucleotides that form part of an unspliced intron within the cDNA clone containing the Cu5 domain are shown in italics.

Fig. 8 Alignment of derived amino acid sequence with other genes. Derived amino acid sequences of Wc1 were aligned with Mc1 and other proteins from bacteria. Residues found in both Mc1 and Wcl that are also conserved in one or more of the other proteins are underlined. Those that are invariant in the aligned sequences are shown in bold type. Abbreviations are E. hirae Cu (Enterococcus hirae copper transporting ATPase26) E. 'coli Hg (Escherichia coli mercuric transport protein merP27) S. aureus Cd(Staphylococcus aureus cadmium efflux ATPase28) R. meliloti Fix(Rhizobium meliloti nitrogen fixation protein Fix138).

Fig. 9 Northern blot hybridization. Probe Wcl.c8 was hybridized to a Northern blot (Clontech) containing 2 mg of polyA+
RNA from a selection of tissues.

Fig. 10 Complete DNA sequence of the Wilson Disease gene.

DETAILED DESCRIPTION OF THE INVENTION

Summary In 51 families with Wilson disease, we have studied DNA
haplotypes of dinucleotide repeat polymorphisms (CA repeats) in the 13q14.3 region to examine these markers for association with the Wilson disease gene (WND). In addition to a previously described marker (D13S133), we have developed three new highly polymorphic markers (D13S314, D13S315, D13S316) close to the WND

locus. We have examined the distribution of marker alleles at the loci studied and have found that D13S314, D13S133, and D13S316 each show non-random distribution on chromosomes carrying the WND
mutation. We have studied haplotypes of these three markers and have found that there are highly significant differences between WND and normal haplotypes in Northern European families. These findings have important implications for mutation detection and molecular diagnosis in Wilson disease families.

Materials and Methods Families Studied Peripheral blood was collected from 28 Canadian families, consisting of 22 of Northern European, three of Southern European, two of Oriental and one of Indian origin, and 23 families from the United Kingdom, consisting of nine of Northern European, four of Southern European, four of Indian, three of Sardinian and three of Middle Eastern origin. 21 of these families have been described elsewhere12,13. The remaining 30 kindreds consisted of 56 parents, 47 patients and 21 unaffected individuals. The ethnic origin of the parents of each patient was determined where possible. Diagnosis of Wilson disease was originally established by clinical symptomatology, slit lamp examination for Kaiser-Fleischer rings of the cornea, and biochemical tests (plasma copper, ceruloplasmin concentrations and urinary copper excretion, and in most cases, measurement of liver copper levels). In some cases, diagnosis was confirmed by radioactive copper studies39 DNA Analysis DNA was extracted from whole blood collected in EDTA by a salt precipitation method40.
The markers used in this study are listed in Table 1. All three new markers were derived from cosmid clones isolated from a flow-sorted chromosome 13 library. D13S316 was derived from a cosmid identified by D13S196, an anonymous marker derived by Alu-PCR of a hybrid containing the upper half of chromosome 1314, D13S314 and D13S315 were derived from cosmids identified by endclones of YACs 27D8 (identified by D13S196) and 235H9 (identified by D13S31) respectively14. Endclones were obtained by an inverse PCR method36 The probes were labelled using a T7 Quickprime' random labelling kit (Pharmacia), hybridized to filter lifts of the cosmid library as described previously1zand exposed to film. Primary positive signals were picked and replated for secondary and if necessary, tertiary screening. DNA was isolated from the clones, digested with three to five of enzymes, electrophoresed in 1% agarose and transferred to nylon membrane (Hybond N+, Amersham). The blot was probed with poly-dCdA (Pharmacia) to identify bands containing potentially poly-morphic repeat units. Positive bands were subcloned and sequenced to determine the length of the repeat. Primers were designed using the OLIGOTM software package (Research Genetics).
All markers were typed by amplification of the poly-dCdA tract by the polymerase chain reaction. The reactions were carried out in 10 l volumes containing 50 mM KC1, 10 mM Tris, pH 8.0, 10 mg/ml BSA, 1.5 mM MgC12, 200 M each of dCTP, dGTP and dTTP, 25 M dATP, 0.2 Ci [a 35S] -dATP and 0.5 units of AmplitaqTM (Perkin Elmer). Amplification was performed in an MJ research PTC-100-96V Programmable Thermal ControllerTM with 30 cycles of 30 seconds denaturation at 94 C, 30 seconds annealing at the appropriate temperature (Table 1), and 30 seconds extension at 72 C. The samples were then electrophoresed through 6% denaturing polyacrylamide gels, which were dried and exposed to film at room temperature for 1 to 5 days.
Allele sizes were determined by amplification of DNA from the original cosmid clone from which the marker was derived. Consistency in allele size determination was checked by including reactions performed on the DNA of 2 to 3 samples of known genotype on each gel.
Gels were read independently by two individuals and ambiguous results were repeated.

Results All four markers were typed in the 51 Wilson disease families. The data are shown in Tables 3 and 4. The relative positions of the markers used in this study are summarized in Figure 1. Also shown is the location of a candidate for WND gene 0 c~
~+1089c r~
~ ~ 75415-2 (Wcl) which we have recently identified as discussed in Experi-ment 2 and in reference 43. D13S314 was derived from the endclone of the 27D8 YAC and anchors the proximal end of the candidate region through a recombination event we have described12. All markers centromeric to this, including D13S315, are also recombin-ant and the event does not include D13S133 or D13S316. The loca-tion of D13S133 relative to the other three markers is based on the pattern of amplification of YACs in the region. Rare recom-bination events were observed for two markers. One obligate crossover occurred in 68 meioses informative for D13S314 and in 100 meioses informative for D13S315, no crossovers were found in 44 meioses for D13S133 and 124 meioses for D13S316.

Alleles associated with the normal and disease chromo-somes were determined. Statistically significant deviations between normal and WND chromosomes were seen in the Northern European families for D13S133, D13S314 and D13S316. These distributions are shown graphically in Figure 2. Tests of signi-ficance were done by the x2 method for large contingency tables with the following results: D13S133 gave a value of 1,9.94 (11 d.f., p < 0.05), D13S314 yielded a value of 22.58 (9 d.f., p< 0.01), and D13S316 resulted in values of 26.62 (8 d.f., p 0.001). Haplotypes of the three markers which gave signi-ficant evidence of association were constructed for each family.
Table 5 summarizes the haplotypes present on Northern European WND
chromosomes and their corresponding frequencies on normal chromo-somes. It has been shown that allele differences of single repeat units in microsatellite markers arise more frequently than larger deviations41. Therefore, because of the variability of CA repeat markers due to single allele slippage and the possibility of mis-assignment of alleles, haplotypes which differ by no more than two bp at a single marker were grouped. The data are significant at p 0.0001 (X2 = 51.44, 8 d.f.). While D13S315 does not show significant amounts of association, several of the haplotypes can be seen to extend to this marker. Haplotype C is exclusively associated with allele 5 at D135315 and allele 4 is present on all chromosomes carrying haplotypes D and E. A specific allele of this marker does not segregate exclusively with the other disease haplotypes. There is the possibility that the variant haplotypes are in fact different mutations present on a similar haplotype, and the technique of grouping similar haplotypes should be used with caution. However, in the case of haplotypes A and B, only one variant is present on WND
chromosomes and the several normal haplotypes have been grouped, therefore any error in classifying haplotypes results in a more conservative estimate of the relative frequency of these WND
chromosomes. In the case of haplotypes C, D, and E, the very low frequency of alleles 5 and 6 on normal chromosomes (0 of 44 haplotypes) as well as the exclusive association with particular alleles of D13S315, make it likely that these groupings are justified.
The validity of this technique will be determined once mutations have been identified in the WND gene.
Northern European families were further subdivided into more specific ethnic groups in order to determine if any haplotypes were characteristic of a particular area. Of the 10 0 2108,927 75415-2 chromosomes carrying haplotype A, six are of British, one French, one Dutch, one German and one of unknown origin. Haplotype B
chromosomes include seven of British and one of Jewish origin.

The seven haplotype C disease chromosomes include two German, two Polish, one Dutch, one French and one of British origin. Chromo-somes with haplotype D consist of three British, three French and one German, and the haplotype 5-11-11, found on a chromosome of French origin is also likely related because it shares an extended haplotype, including D13S315, D13S228 and both RFLPs at D13S31, with other group D disease chromosomes. Haplotype E is found on three chromosomes of German and two of British origin.
Chromosomes from other geographical/ethnic groups show no significant differences from those present on normal chromo-somes due to small sample size.

Discussion We have studied four highly polymorphic dinucleotide repeat polymorphisms near the WND locus and have found that three of them exhibit significant levels of allelic association with the dis-ease. The high degree of association seen between the disease and the markers D13S314 and D13S316 provides strong support for a candidate gene (Wcl) which we have identified on this YAC as dis-cussed in Experiment 2 and in reference 43. These three markers also form haplotypes around Wcl which can be shown to be present on disease chromosomes but not on the normal chromosomes in the same population.

One marker (D13S315) failed to detect significant levels of association with Wilson disease despite its location between Wcl and two markers (D13S31 and D13S228) which have previously been demonstrated to be in disequilibrium with the WND locus13 and its specific association with three common Northern European haplotypes (C,D and E). This marker is furthest from the disease gene and is likely too mutable to detect association over this large distance. Another explanation is that the association previously seen is due to the chance association of alleles with WND chromosomes. However, other studies have found similar patterns of high and low disequilibrium across a disease region42.

The existence of haplotypes found commonly on WND
chromosomes also provides clues as to the number of possible muta-tions present in the Northern European population. While haplo-tvnP.q A_ R_ C. and E are similar, the larae number of chromosomes -1r----= -= -= ---- , ~

of German, Dutch and Polish origin carrying haplotypes C and E
suggest that this is a separate mutation from that present on haplotypes A and B, which are found predominantly on chromosomes of British origin. Haplotypes D and G are very different and almost certainly represent separate mutation events. Evidence of common origins or admixture in the European population can be seen in the existence of the common German haplotypes on three chromo-somes of British and one of French origin. Haplotype D is present on three of six French haplotypes while three British and one German family also carry this haplotype. It is also possible that this represents different mutations that have occurred on the same haplotype in these populations. The additional five haplotypes observed on single WND chromosomes may represent separate muta-tional events or rearrangements of more common haplotypes by an ancestral recombination event, as is likely the case with haplo-type 5-11-11 which is present in a French family and is likely related to haplotype D (5-11-5), and haplotype 6-17-12 which is similar to haplotypes C and E (6-17-10/11). We conclude that there are likely to be at least three common Northern European mutations; one found on haplotypes A and B, another on haplotypes C and E, and a third on haplotype D as well as four or more rare mutations on haplotype G and three of the singly represented haplotypes.

The number of patients in other geographical/ethnic groups are fewer in number it is not possible to get statistically significant data regarding association and haplotypes in most cases. The haplotypes present in our two Sardinian families are interesting because they appear to define three distinctly different haplotypes (4-4-7, 4-7-10, and 7-17-11), unexpected in this island population. This suggests the possibility of at least three different mutations in the Sardinian population. The 7-17-11 haplotype appears on three of six disease chromosomes of Italian origin, in a Jewish family, and is the most common haplo-type present on WND chromosomes in the British families. This haplotype may indicate the presence of a common widespread muta-tion instead of different mutations on the same haplotype.

The rarity of observed recombination events make these markers ideal for DNA diagnosis of families in which an affected child is available for testing. The presence of two highly poly-morphic loci on either side of the candidate gene ensures that any family will most likely be informative for presymptomatic diagnosis of sibs of patients. Occasionally, diagnosis of Wilson disease is difficult to establish and haplotype analysis at the present time could provide information in some cases. The presence of haplotypes A or B would not provide information while the presence of haplotypes C through G could provide support for presence of Wilson disease, but would not be definitive. With the identification of a candidate gene, specific mutations may be defined to provide a more definitive diagnosis.

Summary New markers for Wilson Disease have been isolated in the region of the gene on chromosome 13q14.3 as described in Experi-ment 1. The markers were used to construct a long range restriction map and to obtain 19 YACs in the region. Using the copper-binding motif of the ATPase defective in Menkes disease, a homologous region was identified on three overlapping YACs and on cosmids from a chromosome 13 library. Cosmids were used to iso-late cDNA clones by a direct PCR-based cDNA selection strategy.

The sequence of the isolated gene shows considerable homology with the Menkes ATPase throughout all its functional domains, including at least 6 copper-binding domains, trans-duction, phosphorylation and ATP-binding domains. The gene is expressed in the liver where there is no expression of the MNK
gene. This is compatible with the defect in copper transport in the liver observed in patients with Wilson disease.

Methodology General methods. Southern blotting, and PCR were performed as described in14. Pulsed-field gel electrophoresis (PFGE) is described previously in15. DNA Sequencing was done using a T7 sequencing kit (USB).

Markers. Marker D13S31 is an established marker for Wilson disease with alleles that exhibit strong allelic association with WND12. D13S196 and D13F71S1 were isolated by Alu-PCR and mapped to the region of WND as described in14. The three markers were used in the construction of a long range restriction map of the WND region15. Probe EHR4 was rescued from the distal end of YAC 235H9 (see Table 6).

Cell lines. ICD is a human-hamster somatic cell hybrid containing the proximal half of chromosome 13 as the only human componentl4.
YACs. YACs were identified from pooled YAC DNA and then isolated from the CEPH YAC library by D. LePaslier using the primers shown in Table 6. All of them are located between the two established markers for Wilson disease D13S31 and D13S5911. Geno~nic DNA was isolated as described in35 and sizes were determined by pulsed-field gel electrophoresis (PFGE). YAC 27D8 was characterized in detail. To confirm that it was non-chimeric, probes (27L and 27R) from the left and right ends of the YAC were amplified by inverse PCR36 and hybridized to Mlu I, Nru I and Not I digests of ICD as described in15, enabling the YAC to be aligned within the long range restriction map. Further verification was achieved through restriction analysis of the YAC by complete and partial digests with Mlu I and Not I. Partial digestion was achieved using 1-5 mm incubation in the presence lU restriction enzyme. DNA was separated using a CHEF DR II PFGETM system (Biorad) (using a 4-40s pulse for 16h at 200V) and hybridized to probes specific to the right or left arms of the pYAC4 vector. YACs 53C12, 95C3 and 68F4, isolated using primers for probe 27R, have not been fully characterized. YAC
232 H4, used as a negative control, does not hybridize to any of the probes in the Wilson disease region. Isolation of Menkes (MNK) cDNA
probes. To isolate probes (Mcl.a and Mcl.b) for the MNK gene, reverse transcription was performed with a M-MTV reverse transcription kit (BRL) using 0.5 mg total RNA from cultured human myotubes as template and primer Mc4062 (5' GC(A/G)TCATTGAT(T/G)CC(A/G)TC(C/T)CC 3') corresponding to position 4062 of the MNK cDNA17) in the presence of 40U RNase inhibitor (Pharmacia). Probe Mcl.a was amplified from the reverse transcribed template by PCR14using primers within the putative copper binding region of MNK: Mc967 (5' CAA TGA TTC AAC AGC CAC TT3 ') and Mcl965(5' TTA ATA TGT GCT TTG TTG GTT G 3'). Thirty five cycles were performed using an annealing temperature of 60 C. An additional probe Mci.b was amplified using primers Mc2942 (5' TTT GCA GAC AAA CTC AGT GG 3') and Mc3835 (5' GTC TGC AAT GGC TAT CAA GC 3'). PCR products were directly subcloned into a T-tailed vector (Promega). The relative location of the probes within the MNK cDNA is shown in Fig. 5.

Low stringency hybridization.

Probe Mcl.a was hybridized14at 50 C to HINDIII digests of genomic DNA isolated from YACs in the Wilson disease region (Table 6).
Filters were washed once in 2XSSC and once in 0.2XSSC at room temperature and exposed over-night. Similar hybridization conditions were used to screen 100,000 cosmids from a chromosome 13 specific library (Los Alamos) Isolation of cDNA fragments.
cDNA fragments for Wcl were isolated by a direct selection strategy22,23 using purified insert from cosmids cosL and cosJ. The DNA
was immobilized on filters and incubated with a combination of primary cDNA pools made from adult and fetal liver. Following two rounds of hybridization, cDNAs were subcloned into BluescriptTM vector (Stratagene). Two hundred colonies were picked at random and arrayed.
The colonies were screened at low stringency with probes Mcl.a and Mcl.b. To check their localization, positively hybridizing clones were hybridized, under normal conditions of stringency, firstly to EcoRI digests of cosL and cosJ and secondly to HindIII and EcoRI
digests of YAC 95C3, human and hamster genomic DNA and the human-hamster somatic cell hybrid ICD. Having confirmed that the clones mapped to the correct region of chromosome 13, they were sequenced and placed into contigs based on an overlap region of at least 50bp.
Sequencing data was used to align the clones with each another and with the known sequence of MNK. Clone Wcl.f3 was found to be the most 3' fragment. To isolate the 3' end of the cDNA, fragment Wcl.f3 was labelled to a specific activity of 1x108 cpm/ g and used as a probe to screen four human cDNA libraries.
2x106 plaques were screened from an adult liver libraries (Stratagene), 1x106 from a second adult liver library (Clontech) and 1x106 from an human hepatoma library and 1x106 from an adult kidney library (Clontech). From a positively hybridizing clone isolated from the Kidney library, a cDNA fragment (Wcl.bl-1) was amplified using an upper primer (350U: GTG GCT AGC ATT CAC CTT
TCC) developed from the 31 end of close Wcl.f3 and a lower primer developed from an arm of the cloning vector. An additional cDNA
fragment (Wcl.87-90) was isolated by RT-PCR (described above).

The first strand was extended on l g of poly A+ fetal liver RNA
(Clontech) from primer F3L(5'ATGCGTATCCTTCGGACAGT3'). Forty cycles of PCR were performed using primers 1009U
(5'GGCACATGCAGTACCACTCT3') and 1662L (5'TCTGTCTGGGAGATGTGCTT3') with an annealing temperature of 670C.

Cosmid mapping. Cosmids cosJ and cosL were digested to completion with Not I and then partially digested with XbaI, EcoRI or HindIII. Southern blots of the digested DNA were hybridized to probes derived from the arms of the cosmid vector. Fragments detected were used to construct restriction maps of the two cosmids.

Alignment of derived amino acid sequences. Alignment of the cDNA
clones to MNK and other proteins was carried out at NCBI using the BLAST network service.

Northern blot hybridization. 10 g of total RNA isolated from each of the following tissues: brain, lung, spleen, heart, stomach, esophagus, muscle, liver and lymphoblasts, was separated using a 1% agarose gel containing formamide37. The RNA was trans-ferred onto nylon filters(HybondN+, Amersham). cDNA Wcl.c8 was labelled to a specific activity of 1x108 cpm/ l and hybridized for 20h to the filter in a solution containing 50% formamide, 6xSSC, 0.1% SDS. The filter was washed once in 2xSSC, and once in = 2108927 . 75415-2 0.2xSSC, 0.1% SDS at 65 C. The filter was exposed to X-Ray film (Kodak) for 6 days. In addition, a Northern blot containing polyA+ RNA from heart, brain, placenta, lung, liver, muscle, kidney and pancreas (Clontech) was probed with Wcl.c8 using conditions recommended by the manufacturer.

Isolation of Wc1 Probe Mcl.a from the proposed copper binding region of MNK
(nucleotides 965-196517) was hybridized at low stringency to the 19 YACs listed in Table 6. Above the background hybridization, shown by all YACs (represented in Fig. 3b by YACs 232H4 and 68F4) additional bands were observed in three overlapping YACs: 27D8, 95C3 and 53C12 (Fig. 3b). Two fragments (2.5 kb and 8.9 kb) were detected only in these YACs. The location of YAC 27D8 with respect to the established marker D13S31 is shown in Fig. 3a. To isolate the cross hybridizing sequence, probe Mcl.a was used to screen a chromosome 13 specific cosmid library under the same conditions of low stringency. Two overlapping cosmids (cosJ and cosL) were isolated. From cosL, a non-repetitive probe (cosL.d) was isolated that was also found to cross hybridize tb Mcl.a. To check the localization of cosL.d, it was hybridized under normal conditions of stringency to YACs 27D8, 53C12 and 95C3 (Fig. 3b).
The same 2.5 kb and 8.9 kb fragments were detected.

The cosmids cosL and cosJ were used to isolate expressed sequences from liver using a direct PCR based cDNA selection strategy22,23. To isolate clones from regions of Wcl that were similar to MNK, 200 selected cDNA clones were arrayed and screened at low stringency with probe Mcl.a and a probe (Mcl.b) more = (~ ~ ~ ~ ~ ~ ' towards the 3' end of the MNK cDNA in the ATP-binding region (nucleotides 2940-383017). Thirteen individual cDNA clones of 500-1000 bp in length were isolated, of which eight were char-acterized in detail. To check that they were all located within the correct region of chromosome 13, each cDNA was hybridized to EcoRI and HindIII digests of cosJ and cosL, YAC 95C3, and a chromosome 13 hybrid, ICD. A representative result is shown in Fig. 4a. All fragments detected by the clones used for further analysis mapped only within cosJ or cosL with no other homologous regions on hybrid ICD. In addition, one of the cDNA fragments (Wcli7) was hybridized to Mlu I and Not I digests of DNA from hybrid ICD that had been separated by pulsed field gel electro-phoresis (PFGE). The probe detected a 2200 kb Nru I fragment and 2100 kb and 1200 kb Mlu I fragments that are all common to the established marker D13S3115 (Fig. 4b).

In an attempt to isolate larger cDNA fragments, a total of 4x106 colonies from three liver cDNA libraries were screened with clone Wcl.f3. No further cDNAs were obtained.

The order of the cDNA clones was established by mapping each clone on cosmids cosL and cosJ digested with several restric-tion enzymes (Fig. 3a). The gene covers a region of at least 20 kb.

DNA sequence analysis DNA sequence was obtained for all the clones shown in Fig.3a. Sequence analysis of the cDNA clones revealed that Wcl is very similar to MNK. This enabled the isolated clones to be aligned with the MNK cDNA as shown in Figs. 5, 6 and 10. This alignment agreed with the position of the clones on the cosmid map.

Translation of the nucleotide sequence revealed six putative heavy metal binding domains very similar to the six do-mains found at the 5' end of MNK17,18,19. Alignment of these domains with the corresponding domains in MNK is shown in Fig. 7.
The six Wcl copper domains in the figure show a mean amino acid identity of 65 percent with the corresponding copper domains one to six of MNK. One of the clones Wcl.f8 seems to be unspliced message since it contains a splice donor site. The site is also present in genomic DNA (not shown). The cDNA selection method we used occasionally selects such products22.

Both MNK and Wcl also contain highly conserved domains characteristic of the P-type family of cation transporting ATPases. This family includes magnesium, calcium, potassium, sodium and proton pumps from various organisms. Members of the family contain a highly conserved region containing the motif Asp-Lys-Thr-Gly-Thr (DKTGT), that includes an aspartate residue which forms a phosphorylated intermediate during the cation trans-port cycle. Forty three residues N-terminal to this aspartate is a proline residue thought to be involved in transduction of the energy from ATP hydrolysis to cation transport24. C-terminal to the transduction and phosphorylation domains is a highly conserved ATP-binding domain including a Gly-Asp-Gly (GDG) motif. Align-ments of MNK and Wcl around these three domains are shown in Fig. 8. The identity between MNK and Wcl is 86 perceht throughout the transduction/phosphorylation domains and 79 percent throughout the ATP-binding domain.

Also shown in Fig. 7 is the alignment and homology of the functional domains of Wcl with various heavy metal transport-ing ATPases from bacteria (for a review see25). As has previously been demonstrated for MNK17,18,19, the functional domains of Wcl are more closely related to these prokaryotic genes than to any characterized eukaryotic gene, except MNK. The most closely related gene is copA from the gram positive bacteria Enterococcus hirae a gene involved in copper transport26. Alignments are also shown with a mercuric transporting plasmid encoded protein merP
from Escherichia coli27, a cadmium exporting ATPase from Staphyl-ococcus aureus28 and a protein involved in nitrogen fixation (FixI) from the symbiotic bacterium Rhizobium meli.loti26. In addition to the N-terminal metal binding domains characteristic of this sub-group of ATPases, three other conserved residues are present that are not a general feature of P-type ATPases. These are, two cysteine residues, flanking the invariant proline in the transduction domain and a proline situated 8 residues C-terminal to it. These residues may be involved in conferring metal speci-ficity to the proteins17. DNA sequence is being submitted to genbank.

Expression To determine the tissue distribution of the Wcl message, clone Wcl.C8 was hybridized to Northern blots containing RNA from a variety of tissues. Total RNA was analyzed from brain, lung, spleen, heart, esophagus, muscle, liver and lymphoblasts. Tran-~ 2108927 75415-2 script was detected only in the liver, and in relatively low abundance, only a small fraction based on the actin control (data not shown). Poly A+ RNA was analyzed from a number of tissues (Fig. 9). Transcript of 7.5 kb was detected at an almost equal abundance in the liver and kidney. A slight trace of message of a similar size was also detected in heart, brain, lung, muscle, placenta and pancreas. The transcript appeared to be slightly smaller than the MNK transcript which is approximately 8.0-8.5 kb17,18,19.

The placenta appeared to have an additional transcript of about 7 kb.

Discussion There is strong evidence that the Wcl gene encodes a copper transporting protein. The gene shows high homology with MNK, which is proposed to be involved in transporting copper from intestinal and other cells. Sequence identity is observed in functionally important regions: the energy transduction, phos-phorylation and ATP binding domains are 79% identical or greater.
In comparing the metal-binding and transduction domains of Wcl, MNK, and the copper-resistant bacteria E. hirae, there are certain conserved residues that may be specific for copper transport (Fig. 8).

Wcl is predicted to be the Wilson disease gene because it lies within a region of chromosome 13 that is known to contain WND. A cluster of three highly polymorphic markers D13S133, D13S314 and D13S316, all located within YAC 27D8 and spanning a region of about 300 kb, show strong allelic association with WND

and together define a good candidate region for the gene. Wcl is flanked proximally by D13S314 and distally by D13S133 and D13S31616. No other homologous copper-binding doma.in,s, transduc-tion, phosphorylation, or ATP-binding domains were found within the Wilson disease region.

The expression patterns of Wcl and MNK are very differ-ent. MNK is expressed in lung, skeletal muscle and heart, but is scarcely detectable in the liver or kidney. In contrast, Wcl is expressed mainly in the liver and kidney. This tissue expression is appropriate for Wilson disease. A key feature in Wilson dis-ease is accumulation of copper in the liver. The expression in kidney is consistent with the occurrence of kidney damage, believed to be due to copper toxicity, in many Wilson disease patients. Abnormalities of renal tubular function include amino-aciduria, proteinuria, uricosuria, hypercalciuria, defective urinary acidification, renal stones, and occasionally full blown Fanconi syndrome29,1.

The two main biochemical characteristics of Wilson disease are the disruption of incorporation of copper into cerulo-plasmin in the liver and a severe reduction of copper excretion from the liver into the bile5. Any candidate gene must have potential for involvement in both processes. Ceruloplasmin defi-ciency, almost always associated with Wilson disease30 has been recognized as being very closely related to the basic defect. The localization of the ceruloplasmin locus to chromosome 331 showed that a defect in the ceruloplasmin molecule could not be the basic defect in Wilson disease. However, the deficiency is present in = 2108927 75415-2 patients in early life, before high levels of copper accumulate in the liver. Ceruloplasmin is a 132 kDa glycoprotein containing six atoms of tightly bound copper per molecule, synthesized in hepato-cytes32, and a possible donor of copper to tissues and enzymes3.
Copper is incorporated during the biosynthesis of ceruloplasmin which is then secreted from the hepatocytes into the plasma32.

The two processes, copper incorporation and ceruloplasmin secre-tion, appear to be independent of one another32,33. Incorporation of copper into apoceruloplasmin in vitro can only be achieved under reducing conditions32. It is therefore interesting to note that Wcl contains CXXC motifs in each of its metal binding domains, together with one CXC motif in the transduction domain.
Similar motifs are characteristic of many transition metal binding proteins34. The motifs are abundant in metallothionein and bind copper in the reduced (CuI) state4. Incorporation of copper into ceruloplasmin might require close proximity of the two molecules, and some affinity of ceruloplasmin to the membrane ATPase might be predicted. The pathway involved in copper excretion into bile may be similarly sensitive to the redox state of copper. Wcl there-fore has the potential to play a direct role in copper incorpora-tion into ceruloplasmin, and in copper excretion, by maintaining the metal ion in the correct redox state (Cu I).

Although much is known about the role of copper in many essential enzymes, and about its transport in the blood, the mech-anism of copper transport between tissues has remained unclear.
The isolation of a second human gene for a putative copper trans-porting ATPase, with contrasting tissue distribution, helps to ~ 2108927 reveal exciting new directions in the study of copper transport in health and disease. Wcl and MNK are the only such metal trans-porters isolated to date from eukaryotes, but the high degree of homology preserved between the toxic metal binding ATPases of organisms as evolutionarily divergent as bacteria and humans indicates the fundamental importance of this type of molecule.

EXAMPLES OF APPLICATIONS OF THE INVENTION

The identification of the gene responsible for Wilson disease as well as markers associated with the disease has important implications for the development of new diagnostic and therapeutic strategies for the disease. Below are some examples of the use of the present invention in the diagnosis and treatment of Wilson disease.

A. Diagnosis of presymptomatic sibs After the first individual in a family is diagnosed with Wilson disease, there is frequently difficulty in determining whether other sibs, who have a one in four change of being affect-ed, are actually patients. We have demonstrated in some families (Cox, D.W. and Billingsley, G.D., The application of DNA markers to the diagnosis of presymptomatic Wilson disease. Proceedings of: Genetics of Psychiatric Diseases Wenner-Gren International Symposium, et. L. Wetterberg, Stockholm, pp. 167,988; Houwen et al., H. Hepatol. 17:269, 1993) that an incorrect diagnosis can be made, even when all possible biochemical and radioactive studies are carried out. Reliable diagnosis can be made with the markers we have developed. While other markers have been developed by others in this region, ours are particularly useful in that they ~ 2108927 75415-2 are within about 200 kb of the Wilson disease gene, are very highly polymorphic, and the combination of these alleles, or haplotypes, have been studied both in our patients and in normal individuals (see Experiment 1). The markers we have found particularly useful are as follows:

D13S314 - 12 alleles D13S315 - 9 alleles D13S316 - 9 alleles In addition, we have used, in our haplotypes, D13S133, a marker which was developed by others, which we have identified as being very close to the Wilson disease locus.

Our own markers can be used to reliably diagnose Wilson disease in sibs, and because of the high variability are most likely to be informative in all families. We have already successfully carried out presymptomatic diagnosis in at least six families.

B. Diagnosis of patients Diagnosis of Wilson disease is particularly difficult for those with liver disease, since copper accumulation, charac-teristic of Wilson disease, also occurs in other liver diseases which have a biliary obstructive component. Every abnormal bio-chemical test in Wilson disease can be found to be abnormal in some other type of liver disease. For example, in addition to high liver copper, ceruloplasmin typically decreased in Wilson disease, may be elevated into the normal range.

1) Determination of haplotypes In some cases, the haplotypes we have developed with our DNA markers, along with D13S133, can be used to increase the certainty of a diagnosis of WND that a patient has Wilson disease.
This is because some of the haplotypes which occur in patients are rare in the general population. If a patient has one,of these haplotypes, the chances of having a Wilson disease mutation are high. In combination with biochemical data, positive support for a diagnosis of Wilson disease could be obtained and treatment initiated immediately. Examples of haplotypes which are consid-erably more common in Wilson disease, and have not been found in the normal population are as follows: (refer to Experiment 1 for further description of haplotypes). These haplotypes are comprised of the following markers:

Haplotype C: 6 - 17 - 10 - 5 (particularly in German patients) Haplotype D: 5 - 11 - (5 or 4) - 5 (particularly in French patients) Haplotype E: 6 - 17 - 11 - 4 (particularly in German and British patients) Among our patients of Northern European origin, these haplotypes represent 40% of a series of 47 random patients. This suggests that the haplotype approach could be useful in a rela-tively large proportion of cases.

This approach is useful even when the mutation is not known. However, direct analysis of the mutation will of course be more reliable. Typically, detection of all mutations for disease takes a considerable length of time, and may not be complete for years.
2) Mutation analysis The proposed sequence can be used for the analysis of specific mutations in patients with Wilson disease. The direct analysis of such mutations has important implications for diagnosis. All regions of the sequence can be analyzed by methods such as the polymerase chain reaction, with primer sequences from within the cDNA region as given, or from intron sequences not presented as part of the present sequence. Any of the sequence which is amplified is included in the invention, whether amplified from sequences given or from sequences lying immediately adjacent (in introns). The amplified portions of the sequence also include similar sequences which may have one or a few nucleotides altered, with the end result being amplification of the sequence given. Regions of 250 to 300 base pairs can be analyzed through mutation analysis by direct sequencing. Another method for detecting mutations is through the examination of fragments of 200 to 300 base pairs, which are then analyzed by single strand polymorphism confirmation (SSCP) analysis or by heteroduplex analysis. Either of these methods can detect differences from the normal sequence. The exact mutation can then be confirmed by sequencing. However, once mutations are established, such a survey will be useful for direct mutation detection.
We have used specific primers, as shown below, to amplify a 275 base pair portion of the WND gene, followed by single strand conformation polymorphism (SSCP) and heteroduplex = 2108927 analysis: Two patients have been identified to date with this specific mutation.
B8.3a, 21-mer, 5' TGT AAT CCA GGT GAC AAG CAG 3' B8.3b, 19-mer, 5' CAC AGC ATG GAA GGG AGA G 3' The same approach can be used to identify other mutations throughout the 4120 base pair sequence of the gene.

a) Detection of point mutations The sequence we have obtained is useful for the direct detection of mutations. Based on this sequence, we have developed PCR primers to amplify the functional motifs of the protein:
copper binding, energy transduction, phosphorylation, and ATP
binding. From our sequence, we have developed sequencing primers to sequence PCR products to identify mutations.

b) Detection of deletions or duplications We have also developed primers which will be useful for the detection of deletions. We expect that a large proportion of the mutations in the Wilson disease gene will involve deletion (or duplications), particularly of the copper binding regions.

Because there are six very similar motifs in the copper binding region, as we know from studies of the immunoglobulin heavy chain region carried out in our laboratory, deletions and duplications tend to occur frequently in the present of repeated sequences. In fact this has been demonstrated for Menkes disease17. There are about 16% of patients with Menkes disease who have deletions in the copper binding region. The PCR primers we have developed can be used directly to identify such deletions. All of these primer sequences lie within the region we have submitted in this application.
For additional mutations, the intron exon boundaries we have sequenced will provide a useful source for PCR primers to amplify exons of the gene for the further search for mutations.
Therapy Therapy for Wilson disease at the present time involves chelation of excess copper through the use of a chelating agent such as penicillamine, a potent agent which binds copper through its cysteine residues. But there are problems with the use of this agent, and the neurological symptoms can be worsened on initial treatment as copper is released from to the liver and transfers to other tissues, for example the brain. In addition, about 15% of the patients experience side effects from the therapy, including depression of the immune system, and reduction in the number of white and red blood cells.
Zinc therapy is being used in some cases, but tends to cause intestinal irritation and is not tolerated well by some patients.
The new basis of therapy would involve introduction of the Wilson disease gene in a plasmid. We have discussed the copper and mercury containing plasmids in our publications (Bull et al. Nature Genetics (1993) 5(4): 327-37; Bull and Cox Trends in Genetics (1994)10:246-252). Copper is used extensively in agriculture as a fungicide and bactericide. Certain bacteria have adapted to survive high copper conditions by replicating a high copy number of a plasmid which contains a sequence to encode an ATPase with a copper binding domain, very similar to the Wilson disease gene.

Creation of a construct similar to that found in the copper resistant bacteria therefore appears to be a possible approach.

Such a construct would then have to enter into the liver. This experiment has already been shown to be successful in the Watanabe rabbit, which has heritable hyperlipidemia, and demonstrated that allogenic hepatocytes can be transplanted into affected rabbits to ameliorate hypercholesterolemia (Wilson et al.
PNAS 85:4421, 1988). In this rabbit, the gene was introduced in a plasmid construct attached to an asialoglycoprotein r'eceptor, which targets to the liver cell. A similar approach is therefor feasible for a plasmid containing the Wilson disease gene. Human hepatocytes, cultured in vitro, can be transfected with an adeno-virus containing the Wilson disease gene, and returned to the affected donor into the peripheral circulation. This model has already been tested in rats with the alphal-antitrypsin gene (Jaffe et al. Nature Genetics 1, 1992). Partial hepatectomy can improve the stability of targeted DNA (Wilson et al. J. Biochem.
267:963, 1992). It is of interest that in these studies, DNA in the hepatocytes was present in stabilized plasmids, which do not self-replicate. The Wilson disease therefore could be used directly as a plasmid to be added to cultured hepatocytes, or perhaps to be administered directly through the portal system.
Episomes which contained alphal-antitrypsin were found to remain relatively stable and produced the product (alphal-antitrypsin) for at least four months. Since even a low production of alphal-antitrypsin product should avoid the copper accumulation which takes place, this approach is technically feasible.

Other potential therapies We have outlined in Example 2 that the Wilson disease gene is similar to genes on cadmium resistance and mercury resistance plasmids in bacteria. The similarity exists through all of the functional domains; metal binding, transduction phosphorylation and ATP binding. The Wilson disease gene could therefore be used, if incorporated into a plasmid construct, to remove excess cadmium or mercury from tissues. As expressed above, this is feasible for removal from the liver. Cadmium is particularly carcinogenic in the kidney, and it is of'interest that the Wilson disease gene is expressed in kidney (Experi-ment 2). Targeting of the gene to the kidney could alleviate cadmium toxicity in those who have been inadvertently exposed.
The metal binding regions are very similar for the Wilson disease gene, and for the mercury and cadmium resistance plasmids. It is very likely that this sequence will be found to bind the other heavy metals. The differences outlined in Experiment 2, Figure 8, may suggest that slight alteration in the copper binding region could increase the specific binding for mercury and cadmium.

A construct containing the Wilson disease gene could potentially be used to overcome the defect in Menkes disease, since the copper binding region is very similar. A new process of targeting tissues with DNA-coated gold pellets (Yang et al. PNAS
87:9568, 1993) suggest that the intestinal cells, the site of the defect in Menkes disease, could be induced to incorporate Wilson disease DNA to allow transport of copper out of that tissue.

! 210$927 75415-2 Introduction of the plasmid into the intestinal epithelial cells seems also to be feasible.

Another approach, for both Wilson and Menkes disease would be to induce overexpression of the defective gene, which may be possible if there is residual activity of the gene product. We have found from our haplotype studies that most patients with Wilson disease appear to be genetic compounds, that is they prob-ably carry two different mutations. At least one of these may have residual activity.

Non-human applications The Wilson disease gene could be targeted into the germ line of organisms for which the accumulation of toxic metals is a problem. For example, the targeting of the Wilson disease or of similar sequence into a plasmid into the germ line of fish stocks could increase the ability of such stocks to eliminate heavy metals, in regions which have naturally-occurring or pollution induced metal contamination.

Copper toxicity has been noted as a problem in sheep, as may also be a problem in other domestic species. It is possible that this toxicity in sheep is due to particularly low levels of expression of the homologous gene to the P-type ATPase described in this application for WND. The sequence presented may therefore have some application in therapy for toxicity in sheep, or in other animal species, or could be used in breeding to produce sheep, or other species which are more copper resistant. The sheep is given as only one example of an animal sensitive to copper toxicity. Other uses are also envisioned for the removal ~ 2108927 75415-2 of copper or other toxic metals not only from sheep, but a variety of other organisms, including the removal of inercury from fish or any other species.

The DNA sequence of the present invention can be used to obtain the equivalent gene from the rat, to study the homologous gene. The human sequence in this application could be used to facilitate obtaining the sequence for the homologous gene in the Long-Evans Cinnamon (LEC) rat, an inbred strain of mutant rat, in which the defect in copper metabolism may be similar to that of Wilson disease. Any use of the human sequence or a portion of it to be used for study of the LEC rat and its normal counterpart are included in this application.

The DNA sequence of the present invention can be used to obtain the equivalent gene from the mouse, to study the homologous gene. The human sequence in this application could be used to facilitate obtaining the sequence for the homologous gene in the toxic milk mouse, an inbred strain of mutant mouse, the defect in copper metabolism which may be identical to that of Wilson dis-ease. Any use of the human sequence or a portion of it to be used for study of the toxic milk mouse and its normal counterpart are included in this application.

While the above refer to specific applications of the present invention, it is to be appreciated that other uses, that are conceivable by one skilled in the art, are also within the scope of the present invention.

~ 2108927 75415-2 References 1. Brewer, G. J. & Yuzbasiyan-Gurkan, V. Wilson Disease.
Medicine 71, 139-164 (1992).

2. Sarkar, B. in Metal ions in biological systems (ed Sigel, H.) 233-281 (Marcel Dekker, Inc., New York, 1981).

3. Orena, S. J., Goode, C. A. & Linder, M. C. Binding and uptake of copper from ceruloplasmin. Biochem. Biophys. Res. Comm.
139, 822-829 (1986) 4. Kagi, J. H. R. & Schaffer, A. Biochemistry of metallothionein. Biochem 27, 8509-8515 (1988).

5. Danks, D. M. in Metabolic Basis of Inherited Disease (eds Beaudet, A.L., Sly, W.S. & Valle, D.) Vol. 6, 1411-1431 (McGraw-Hill, New York, 1989).

6. Darwish, H. M., Hoke, J. E. & Ettinger, M. J. Kinetics of Cu(II) transport and accumulation by hepatocytes from copper-deficient mice and the brindled mouse model of Menkes disease. J. Biol. Chem. 258, 13621-13626 (1983).

7. Frydman, M., Bonne-Tamir, B., Farrer, L. A., et al. Assign-ment of the gene for Wilson disease to chromosome 13. Proc Natl Acad Sci USA 82, 1819-1821 (1985).

8. Bowcock, A. M., Farrer, L. A., Cavalli-Sforza, L. L., et al.
Mapping the Wilson disease locus to a cluster of linked polymorphic markers on chromosome 13. Am J Hum Genet 41, 27-35 (1987).

9. Bowcock, A. M., Farrer, L. A., Hebert, J. M., et al. Eight closely linked loci place the Wilson disease locus within 13q14-q21. Am J Hum Genet 43, 664-674 (1988).

~ 2108927 10. Yuzbasiyan-Gurkan, V., Brewer, G. J., Boerwinkle, E. & Venta, P. J. Linkage of the Wilson disease gene to chromosome 13 in North-American pedigrees. Am J Hum Genet 42, 825-829 (1988).

11. Farrer, L. A., Bowcock, A. M., Hebert, J. M., et al.

Predictive testing for Wilson's disease using tightly linked and flanking DNA markers. Neurology 41, 992-999 (1991).

12. Thomas, G. R., Roberts, E. A., Rosales, T. 0., et al. Allelic association and linkage studies in Wilson disease. Hum. Mol.
Genet. (1993) . (in press) 13. Houwen, R. H. J., Berger, R., Cox, D. W. & Buys, C. H. C. M.
Allelic association for Wilson disease-D13S31. J Hepatol 16, S15 (1992).

14. Bull, P. C., Barwell, J. A., Hannah, H. T-L., et al.
Isolation of new probes in the region of the Wilson disease locus, 13q14.2-14.3. Cytogenet Cell Genet 64, 12-17 (1993).

15. Bull, P. C. & Cox, D. W. Long range restriction mapping of 13q14.3 focused on the Wilson disease region. Genomics 16, 593-598 (1993).

16. Thomas, G. R., Bull, P. C., Roberts, E. A., Walshe, J. R. &
Cox, D. W. Haplotype studies in wilson disease. Am J Hum Genet (1993) . (in press) 17. Vulpe, C., Levinson, B., Whitney, S., Packman, S. & Git-schier, J. Isolation of a candidate gene for Menkes disease and evidence that it encodes a copper-transporting ATPase.
Nature Genet. 3, 7-13 (1993).

= 2108927 18. Mercer, J. F. B., Livingstone, J., Hall, B., et al. Isolation of a partial candidate gene for Menkes disease by positional cloning. Nature Genet. 3, 20-25 (1993).

19. Chelly, J., Tumer, Z., Tonnesen, T., et al. Isolation of a candidate gene for Menkes disease that encodes a potential heavy metal binding protein. Nature Genet. 3, 14-19 (1993).

20. Sass-Kortsak, A. Copper metabolism. Adv. Clin. Chem. 8, 1-67 (1965).

21. Owen, C. A. Jr. Metabolism of radiocopper (64Cu) in the rat.
Am. J. Physiol. 209, 900-904 (1965).

22. Rommens, J. M., Lin, B., Hutchinson, G. B., et al. A tran-scription map of the region containing the Huntington disease gene. Hum. Mol. Genet. 2, 901-907 (1993).

23. Goldberg, Y. P., Lin, B. -Y., Andrew, S. E., et al. Cloning and mapping of the a-adducin gene close to D4S95, and assess-ment of its relationship to Huntington disease. Hum. Mol.
Genet. 1, 669-675 (1992).

24. Vilsen, B., Andersen, J. P., Clarke, D. M. & MacLennan, D. H.
Functional consequences of proline mutations in the cytoplas-mic and transmembrane sectors of Ca++-ATPase of sarcoplasmic reticulum. J. Biol. Chem. 264, 21024-21030 (1989).

25. Silver, S., Nucifora, G., Chu, L. & Misra, T. K. Bacterial resistance ATPases: primary pumps for exporting toxic cations and anions. Trends Biochem Sci 14, 76-80 (1989).

26. Odermatt, A., Suter, H., Krapf, R. & Solioz, M. Primary structure of two P-type ATPases involved in copper ~ 2108927 homeostasis in Enterococcus hirae. J. Biol. Chem. 268, 12775-12779 (1993).

27. Griffin, H. G., Foster, T. J., Silver, S. & Misra, T. K. Cloning and DNA sequence of the mercuric and organo-mercurial resistance determinants of plasmid pDU1358. Proc Natl Acad Sci USA 84, 3112-3116 (1987).

28. Nucifora, G., Chu, L., Misra, T. K. & Silver, S. Cadmium resistance from Staphylococcus aureus plasmid p1258 cadA gene results from a cadmium-efflux ATPase. Proc Natl Acad Sci USA
86, 3544-3548 (1989).

29. Yarze, J. C., Martin, P., Munoz, S. J. & Friedman, L. S.
Wilson Disease: current status. Am. J. Med. 92, 643-654 (1992).

30. Sternlieb, I. Perspectives on Wilson Disease. Hepatology 12, 1234-1239 (1990).

31. Yang, F., Naylor, S. L., Lum, J. B., et al. Characterization, mapping, and expression of the human ceruloplasmin gene. Proc Nat1 Acad Sci USA 83, 3257-3261 (1986).

32. Sato, M. & Gitlin, J. D. Mechanisms of copper incorporation during the biosythesis of human ceruloplasmin. J. Biol. Chem.
266, 5128-5134 (1991).

33. Gitlin, J. D., Schroeder, J. J., Lee-Ambrose, L. M & Cousins, R. J. Mechanisms of ceruloplasmin biosynthesis in normal and copper-deficient rats. Biochem. J. 282, 835-839 (1992).

34. O'Halloran, T. V. Transition metals in control of gene expression. Science 261, 715-725 (1993).

~ 2108927 35. Scherer, S. & Tsui, L-T. in Advanced techniques in chromosome research (ed Adolph, K.W.) 33-72 (Marcel Dekker, Inc., New York, Basel, Hong kong, 1991).

36. Triglia, T., Peterson, M. G. & Kemp, D. J. A procedure for in vitro amplification of DNA segments that lie outside the boundaries of known sequences. Nucl Acids Res 16, 8186-8180 (1988).

37. Sambrook, J. Fritsch, E. F. & Maniatis, T. Molecular cloning, a laboratory manual (Cold Spring Harbor Laboratory Press, Cold Spring Harbor, 1989).

38. Kahn, D., David, M., Domergue, 0., et al. Rhizob.ium melitoli fixGHI sequence predicts involvement of a specific cation pump in symbiotic nitrogen fixation. J bacteriol 171, 929-939 (1989).

39. Cox DW, Fraser FC and Sass-Kortsak A (1972) A genetic study of Wilson's disease: evidence for heterogeneity. Am J Hum Genet 24:646-666 40. Miller SA, Dykes DD and Polesky HF (1988) A simple salting out procedure for extracting DNA from human nucleated cells.
Nucl Acids Res 16:1215 41. Oudet C, Mornet E, Serre JL, Thomas F, Lentes-Zengerling S, Kretz C, Deluchat C, et al (1993) Linkage disequilibrium between the fragile X mutation and two closely linked CA
repeats suggests that fragile X chromosomes are derived from a small number of founder chromosomes. Am J Hum Genet 52:297-304 42. MacDonald ME, Lin C, Srinidhi L, Bates G, Altherr M, Whaley WL, Lehrach H, et al (1991) Complex pattern of linkage dis-equilibrium in the Huntington disease region. Am J Hum Genet 49:723-734 43. Bull PC, Thomas GR, Rommens JM, Forbes JR and Cox DW (1993) Identification of a putative copper binding P--type ATPase similar to the Menkes (MNK) gene: a candidate for the Wilson disease gene. (submitted) WtT' ~hlel 2108927 Markers tisecl in this sttidy Annealing No.
Lcx us Primers Temr Alleles PIC 133201 133202 D13S314 GAGTOGAGGAGGAGAAAAGA 62' 12 0.76 141 / 141 145/ 141 GTGTGACTGGATGGATGTGA

D13S315 GCCATCCAGAGTTAAACCA 58' 8 0.45 164 / 1G2 1G4 / 162 TTATAGCTTi I'CTCATGCATTC

D 1 1S 116 GCAGCAATGCTTTGTGCATAA 02' 9 0.73 1401140 1401140 TGTTTCCCACCAATCTTACCG

D 11S 133 See Ref. (Petrukhin et al. 1993) Rcfercnce genotylies froin CEPH f<unily 1332. Numhcrs arc allele sizes in hasc pairs.

,_._.

Table 2 Ailele size detinitiuns Aliete I)135316 D13S133 D13S314 D13S315 11) 169 145 Values rerresenl allele sizes in base pairs (bp) ~.~.~
ApU1e3 2108927 Marker distrihutivns on 'ilson disease family chrvmosomes NE" SE' Sarc)" mE" Or" IP
Mlrker A11e)e N W N W N W N W N W N W
D13S316 1 2 l) t) O O O O O 0 0 0 0 t) O 0 Q 0 0 0 0 R 4 2 1 ( 0 0 0 0 0 0 3 0 .......... 1...... ct.......... a...... o........c~...... ~......... n.....
j)......... Q ...... o........ ....... o Total 56 60 14 14 5 6 5 6 4 4 10 11 2 t) 0 0 U l 0 I t) 0 0 0 0 9 3 0 1 0 0 0 t) 0 0 0 0 0 R ........ 7 ...... I........... 0...... !1......... 0...... 5?.........
0....... 1......... 0...... 1.... ....(1....... 5) Tcital 52 53 13 14 4 4 14 17 1 2 10 11 " Families were grouped according to geographical origin. NE = Nortliern European, SE = Soutliern Euro)xau-, Sard =
Surdinian, ME = Middle Eastern, Or = Oriental, IP = Indian/Pakistani #I)le 4 Marker (listr-ilmutiems oti ZVilsuti (lisease familv chrumesomes NO SE'i S.vc!'t ME11 Or'r iP'r Marker nllele N W N W N W N W N W N W
1)1 3S 1 33 I U U 0 U 0 O I 0 0 0 0 2 0 I l 0 0 U 1 0 0 0 0 0 0 l 2 0 1 0 0 0 0 0 2 5 7 6 O U 0 0 1 0 l 1 2 0 0 9 2 0 0 I 0 0 0 t U 0 1 1 t I 0 0 0 0 1 0 0 0 0 0 1( 5 7 0 0 0 0 0 0 0 0 1 0 13 U 0 t 0 0 0 0 0 0 0 0 0 16 2 0 0 0 l 0 l) 0 U U 1 O
.17...... _.22....40...........5...... . 8 .........
0......2........Ø.....:}.........3......2........2.......1 Total 49 58 12 12 4 4 5 6 4 4 10 11 7 1 l 0 0 0 ( 0 2 1 0 0 ~
K......... ~)...... a........... (?...... I......... (t...... 0.........
(t...... 0......... (?...... U........ 51....... 0 Total 57 58 14 14 6 6 5 6 4 4 10 11 n See notes for table 2 We 5 Iiahlutype distrihutiun un cl)rumusumes of Nurthern I?urupean oril;in WND Normal GrcniF Flsihlotp}~cs No. Freq No. Frcq A 7-17-1() 9 3 0.21 4 0.09 8 0.17 12 0.27 C 6-17-10 7 0.15 U 0.00 7 0.15 0 0.00 E 6-17-1l 5 0.11 u (LW

3 0.06 0 0.00 2 0.04 u 0.00 5 others 5 0.11 4 0.09 IS otlrers 0 0.00 24 0.55 ....................................................=--..........................
"Tutsil 47 44 F{<qplutypes are given in the order: D 13S316 - D 13S 133 - D 13S314 Table 6 YACs in the Wilson disease region D Number Probe Primers YACs D13F71S1/2 pB32.3 CCGGGTATCTTAATTGGTGT 11G2;102F4;296G5 D13S196 pB40.3 GCAAAGTTCATAGGAAACCAGG 27D8;86A3;90H11;
ACATTTTGGTCAGACACTGGC 220A9;298H2 27R ATTGGGCATCTCTTGCTGTT 9B2;53C12;68F3 TGCAGGAATTCACTGTGTGA 95C3;407F11;117E9 EHR4 GGCCAGAATGACAAAATTCA 378B12;215B5;407F3 GGCTTCATGAGTGTGGTCCT

Claims (23)

1. A nucleic acid molecule comprising the DNA sequence as illustrated in Figure 10, the complementary sequence thereof or an allelic variant thereof.

2. The nucleic acid molecule according to claim 1 which comprises the DNA sequence as illustrated in Figure 5, the complementary sequence thereof or an allolic variant thereof.

3. The nucleic acid molecule according to claim 1 consisting of the DNA sequence as illustrated in Figure 10, the complementary sequence thereof or an allelic variant thereof.

4. A nucleic acid molecule comprising the underlined DNA
sequence designated as Cul in Figure 10.

5. A nucleic acid molecule comprising the underlined DNA
sequence designated as Cu2 in Figure 10.

6. A nucleic acid molecule comprising the underlined DNA
sequence designated as Cu3 in Figure 10.

7. A nucleic acid molecule comprising the underlined DNA
sequence designated as Cu4 in Figure 10.

8. A nucleic acid molecule comprising the underlined DNA
sequence designated as Cu5 in Figure 10.

9. A nucleic acid molecule comprising the underlined DNA
sequence designated as Cu6 in Figure 10.

10. A nucleic acid molecule comprising the underlined DNA
sequence designated as Pt/T in Figure 10.

11. A nucleic acid molecule comprising the underlined DNA
sequence designated as Tm in Figure 10.

12. A nucleic acid molecule comprising the dotted underlined DNA sequence designated as Ph in Figure 10.

13. A nucleic acid molecule comprising the underlined DNA
sequence designated as ATP-hinge in Figure 10.

14. Use of a nucleic acid molecule according to any one of claims 1 to 13, the complementary sequence thereof, or an allelic variant thereof to detect Wilson disease.

15. A DNA marker associated with the gene for Wilson disease characterized in that it detects the same dinucleotide repeat polymorphism as DNA marker D13S314, wherein said marker can be amplified using primers comprising sequences 5' GAG TGG AGG AGG AGA AAA GA 3' and 5' GTG TGA CTG GAT
GGA TGT GA 3'.

16. A DNA marker associated with the gene for Wilson disease characterized in that it detects the same dinucleotide repeat polymorphism as DNA marker D13S315, wherein said marker can be amplified using primers comprising the sequences 5' GCC ATC CAG AGT TAA ACC A 3' and 5' TTA TAG CTT TTC TCA
TGC ATT C 3'.

17. A DNA marker associated with the gene for Wilson disease characterized in that it detects the same dinucleotide repeat polymorphism as DNA marker D13S316, wherein said marker can be amplified using primers comprising the sequences 5' GCA GCA ATG CTT TGT GCA TAA 3' and 5' TGT TTC CCA CCA ATC TTA CCG 3'.

18. Use of a DNA marker according to any one of claims 15 to 17 to detect Wilson disease.

19. A kit comprising at least one pair of primers selected from the group consisting of a) 5' GAG TGG AGG AGG AGA AAA GA 3' and 5' GTG TGA CTG GAT GGA TGT GA 3';

b) 5' GCC ATC CAG AGT TAA ACC A 3' and 5' TTA TAG CTT TTC TCA TGC ATT C 3'; and c) 5' GCA GCA ATG CTT TGT GCA TAA 3' and 5' TGT TTC CCA CCA ATC TTA CCG 3';

and instructions for detecting Wilson disease.

20. Use of a nucleic acid molecule according to any one of claims 1 to 13, the complementary sequence thereof, or an allelic variant thereof to treat Wilson disease.

21. Use of a nucleic acid molecule according to any one of claims 1 to 13, the complementary sequence thereof, or an allelic variant thereof to isolate the Wilson disease gene, or fragment thereof, from a mammal.

22. Use of a primer to detect a mutation in the Wilson disease gene in a Wilson disease patient, wherein said primer comprises 5' TGT AAT CCA GGT GAC AAG CG 3' or 5' CAC AGC ATG GAA GGG
AGA G 3'.

23. Use of a nucleic acid molecule according to any one of claims 1 to 13, the complementary sequence thereof or an allelic variant thereof to reduce metal toxicity in an animal.