Menkes Disease (MD) is a rare X-linked recessive fatal neurodegenerative disorder caused by mutations in the ATP7A gene, and most patients are males. Female carriers are mosaics of wild-type and mutant cells due to the random X inactivation, and they are rarely affected. In the largest cohort of MD patients reported so far which consists of 517 families we identified 9 neurologically affected carriers with normal karyotypes.
We investigated at-risk females for mutations in the ATP7A gene by sequencing or by multiplex ligation-dependent probe amplification (MLPA). We analyzed the X-inactivation pattern in affected female carriers, unaffected female carriers and non-carrier females as controls, using the human androgen-receptor gene methylation assay (HUMAR).
The clinical symptoms of affected females are generally milder than those of affected boys with the same mutations. While a skewed inactivation of the X-chromosome which harbours the mutation was observed in 94% of 49 investigated unaffected carriers, a more varied pattern was observed in the affected carriers. Of 9 investigated affected females, preferential silencing of the normal X-chromosome was observed in 4, preferential X-inactivation of the mutant X chromosome in 2, an even X-inactivation pattern in 1, and an inconclusive pattern in 2. The X-inactivation pattern correlates with the degree of mental retardation in the affected females. Eighty-one percent of 32 investigated females in the control group had moderately skewed or an even X-inactivation pattern.
The X- inactivation pattern alone cannot be used to predict the phenotypic outcome in female carriers, as even those with skewed X-inactivation of the X-chromosome harbouring the mutation might have neurological symptoms.
Menkes Disease (MD [MIM 309400]) is a rare X-linked recessive disorder, with an incidence of about 1:298.000  and is caused by mutations in the gene ATP7A [MIM 300011]. This gene encodes a copper transporting P-type ATPase, essential for the release of dietary copper from the intestine to the body, including the brain. The clinical features of MD are consequences of an impaired copper distribution and the malfunction of a large number of copper-requiring enzymes . Classical MD is characterized by mental retardation, hypothermia, seizures, cutis laxa, hypo-pigmentation, abnormal hair (kinky hair or pili torti), and decreased serum ceruloplasmin levels .
The majority of Menkes patients are males and, to our knowledge, only 9 affected females have been described so far [4-10]. Five of these patients had X-autosomal translocations [5-7,9,10] one had 45X/46XX mosaicism , one had an unknown karyotype , and two patients had normal karyotypes . Skewed X-inactivation has been reported previously in unaffected female carriers , but to our knowledge, X-inactivation has not previously been investigated in females who are affected with MD, but have a normal karyotype. This report includes the clinical description and molecular characterization of 9 affected females with normal karyotypes. Furthermore, we compare the X-inactivation pattern in the disease-manifesting carriers with the pattern in 49 asymptomatic MD carriers and the pattern in 32 normal females with no ATP7A mutation.
In a cohort of 517 unrelated families referred to the Kennedy Center in Denmark for genetic confirmation of MD, we found 10 manifesting carriers. One patient with 45X/46XX mosaicism has been described previously  and will not be described further here. Nine of the affected carriers, F1-F9, have normal karyotypes. The females were born in Europe (United Kingdom, France or Germany) or United States from 1969 to 2002. The clinical phenotypes of patients F1 and F2 have been described previously [4,12].
- Patient F1 (family, 9529)
- Patient F2 (family, 93232)
- Patient F3 (family, 9228)
- Patient F4 (family, 9229).
- Patient F5 (family, 91284).
- Patient F6 (family, 92209)
- Patient F7 (family, 92283)
- Patient F8 (family, 92292)
- Patient F9 (family, 94228)
Investigation of carrier status
The carrier status of the female family members was determined by testing for the identified family-specific mutation in ATP7A by either sequencing or multiplex ligation-dependent probe amplification analysis (MLPA) (Salsa P104 kit, MRC, Amsterdam, Holland), depending on the type of mutation.
DNA and RNA isolation and cDNA synthesis
Genomic DNA was prepared from peripheral blood lymphocytes, Epstein-Barr virus transformed lymphocytes, or cultured fibroblasts with the NaCl extraction method . RNAeasy was used for RNA isolation (QIAgen, Bothell, WA) and single-stranded cDNA was synthesized using the High-Capacity cDNA Archive Kit (Applied Biosystems, Foster, CA).
The methylation status of the human androgen receptor (HUMARA) gene was used as an indicator for X-chromosome inactivation [14,15]. The amplified products were analysed using an ABI 3130XL Genetic Analyzer, and the ratio between the signal intensities of the two alleles was calculated using GeneMapper 3.0 software (Applied Biosystems, Foster, CA). For each female, the ratio between signals from the two alleles in the undigested sample was used as a correction factor. DNA from an affected male from the same family was included whenever possible in order to reveal the allelic AR variant on the mutant X-chromosome.
Quantitation of ATP7A transcript
Real-time PCR amplification of ATP7A transcripts from fibroblasts from the patient F2 was carried out on cDNA with an ABI7300 Genetic Analyzer in accordance with the manufacturer's instructions (Applied Biosystems, Foster, CA) as described previously . We used FAM-labelled Taq-Man probes and primers that anneal to the junction between exon 22/exon 23 and exon 1/exon 2 in ATP7A, respectively (Applied Biosystems assay numbers: Hs00921963_m1 and Hs00921966_m1). A VIC-probe and primers to human GADPH (Applied Biosystems assay numbers: 4326317E) were used as an endogenous control.
Copper uptake and retention in amniotic cells and fibroblasts was performed as described previously .
The Menkes gene is controlled by X-inactivation
In males, fibroblasts with a mutation in the ATP7A gene accumulate and retain more 64Cu than normal cells . Heterozygous females are expected to be mosaic for ATP7A gene expression with one population of cells expressing the wild-type ATP7A gene and the other population expressing the mutated ATP7A gene due to random X-chromosome inactivation . This has previously been shown biochemically by isolating two different cell populations (clones) from a pool of fibroblasts from a heterozygous female; one cell population accumulated a large amount of copper, just like cells from affected males, while the other cell population behaved like normal control cells . We confirmed this inactivation pattern by investigating the expressed allele from two such isolated clones. Using RT-PCR analysis, we found that the cell clone with the increased copper uptake expressed mutant ATP7A transcripts, whereas the cell clone with the normal copper uptake expressed wild-type transcripts (Figure 1).
Figure 1. ATP7A transcripts in fibroblasts from a heterozygous female with the mutation c.1554-?_2172 +?del (Ex6_9del). Clone 43 which is characterized by increased copper accumulation and retention (63.4 ng 64Cu/mg protein/20 hours and 61.1% retention) expresses only the mutant transcripts (1365 bp, missing exon 6-9). In contrast clone 29, characterized by normal copper uptake and retention (11.2ng 64Cu/mg protein/20 hours and 11.3% retention) expresses only wild-type transcripts (1880 bp). In uncloned cell populations with normal copper uptake and retention (23.1 ng 64Cu/mg protein/20 hours and 11.3% retention) from the same female, only normal wild-type transcripts were detected. PCR amplification of a cDNA fragment from exon 4 to exon 10 was performed by nested PCR with the primer-pair (5'-caaaagcagcccaagtacctc-3'/5'-ggtggttgccagcacaatcagt-3') followed by the primer-pair (5'-cagaagggtcacagcaaagg-3/'5'-ggtggttgccagcacaatcagt-3). I: index (D98-34147H), affected male with the mutation Ex6_9del in ATP7A. C: a control sample (D98-35541H). Pool: uncloned pool from the heterozygous female (D98-35540H). Cl43: Clone 43. Cl29: Clone 29. The PCR products were separated on a 2% agarose gel and visualized with ethidium bromide.
Identification of mutations in affected female carriers
ATP7A mutations were identified in all 9 females, except for F2 (Table 1). The coding exons of ATP7A in F2 were screened for mutations on genomic DNA, but no mutations could be identified. In this patient the ATP7A transcript was below the detection limit, in contrast to the normal amount of the endogenous control GADPH transcript (not shown).
Table 1. Summary of clinical, biochemical and genetic findings in manifesting carriers
Based on the clinical phenotypes and the carrier status we suggest that all 9 clinically affected females are suffering from MD (F1-F9). This is supported by the fact that the copper uptake and/or retention were increased in all affected females (not measured in patient F5). Serum copper and ceruloplasmin were not reduced in all 9 patients, but the levels can be normal - also in affected males with mild symptoms (Table 1).
X-inactivation patterns in affected female carriers
A possible association between skewed X-inactivation and disease severity was investigated by measuring the X-inactivation patterns in the 9 affected females (Table 2). One female (F9) had a highly skewed inactivation of the mutant X-chromosome (ratio = 100:0), and three females (F6, F7, and probably F2) had highly skewed inactivation of the normal X-chromosome (ratio ≤ 4:96). Two females had moderately skewed inactivation of either the mutant X-chromosome (F4) (ratio = 80:20) or the normal X-chromosome (F1) (ratio = 24:76), and one female had an even X-inactivation pattern (F5). The X-inactivation assay was inconclusive for two females (F3, F8) due to homozygosity of the AR polymorphism.
Table 2. X-inactivation pattern in affected females and family members
The clinical phenotypes, biochemical and molecular findings of the 9 clinically affected carriers (F1-F9) are summarized in Table 1.
X-inactivation patterns of unaffected female carriers
For comparison, the X-inactivation patterns of 49 unaffected female carriers were measured. The first sub-group was composed of 15 women from the same families as the manifesting carriers. (Table 2). The second sub-group consisted of 34 unaffected carriers from other Menkes families (Table 3). A very high fraction of the unaffected carriers (69%, 34 females out of 49) had highly skewed inactivation of the mutant X-chromosome (ratio ≥ 90:10), whereas a minor fraction (24%, 12 females out of 49) had a more moderate inactivation (90:10 ≥ ratio ≥ 70:30) of the mutant X-chromosome, and only three females exhibited random X-inactivation. These results seem to be independent of both the DNA-source (lymphocytes, transformed lymphocytes or fibroblasts), family and of the type of mutation in the ATP7A gene (Table 2 and 3).
Table 3. X-inactivation status of members of other families affected with MD
X-inactivation patterns of normal non-carrier females
The X-inactivation patterns of 32 non-carrier females from the same Menkes families revealed that 6 individuals had highly skewed (ratio ≥ 90:10) and 11 individuals had moderately skewed (90:10 ≥ ratio ≥ 70:30) X-inactivation patterns. About half of the non-carriers (47%, 15 of 32) had an almost even X-inactivation pattern (70:30 ≥ ratio ≥ 50:50) (Table 2 and 3).
Affected carriers of MD are rare. In a total of 517 families we know of 9 clearly affected females. However, we know also of mildly affected females who have not been characterized as manifesting carriers. These mildly affected females had pili torti, hypopigmentation of the skin or other mild symptoms of MD. The total number of females with any symptoms of MD is unknown.
Due to random X-inactivation, females are mosaic for X-linked gene expression. On average, approximately 50% of the cells express genes from the paternal X-chromosome and the other 50% express genes from the maternal X-chromosome . Three different mechanisms for skewed X-inactivation patterns have been suggested : 1) Skewed X-inactivation by chance. 2) A genetic defect in the X-inactivation process. 3) Post-inactivation growth advantage/disadvantage for cells with a mutation on the X-chromosome. It has been shown that several X-linked mental retardation disorders have a strong association with skewed X-chromosome inactivation in carrier females. The mechanism is however unknown .
The observation that most unaffected carriers have an inactivation of the mutant X-chromosome concurs with the recent paper by Desai et al. , in which complete inactivation of the mutant X-chromosome was observed in one Menkes family with 6 heterozygous females. We also observed skewed inactivation of the mutant X chromosme in most unaffected carriers. We found highly skewed inactivation (≥ 90:10) of the mutant X-chromosome in 69% of the asymptomatic carriers, and moderately skewed inactivation of the mutant X-chromosome in 24% of the carriers. Only 6% exhibited an even inactivation and none of the female exhibited skewed inactivation of the normal X-chromosome. Investigation of the manifesting carriers revealed a large variation in the observed X-inactivation pattern. Although more than half of the affected females (57%, 4 out of 7) had skewed inactivation of the normal X-chromosome, complete inactivation of the mutated X-chromosome was also observed in one female, whereas an even X-inactivation pattern was found in another. Finally, about 19% of the normal females without any mutation in the ATP7A gene exhibited highly skewed X-inactivation. In conclusion, skewed X-inactivation of the mutant X chromosome was noted in the majority of asymptomatic carriers whereas skewed X-inactivation of the normal X-chromosome was noted in the majority of the affected females. Thus although a clear tendency was observed, a number of exceptions from this tendency were observed.
X-inactivation is influenced by several factors and the X-inactivation in DNA from blood samples, cultured EBV transformed lymphocytes or fibroblast cultures may not reflect the pattern in the rest of the body. In addition, the risk of a recombination between the ATP7A locus and the AR locus during meiosis could lead to a faulty interpretation. The distance between the two loci corresponds to a recombination frequency of about 12%. Furthermore, the pattern obtained from a cell culture might depend on the number of passages. Long-term culture can affect the X-inactivation pattern (own unpublished observation).
The symptoms of MD were milder in all 9 analyzed manifesting carriers than in affected males with the same mutations. Whereas affected males from the same families had the classical form of MD (except the brother of F5 who had a slightly milder form of MD with prolonged survival), the affected females survive longer, and in general they exhibit milder symptoms. Five female cases of MD carrying X-autosome translocations have been described previously. Although the clinical symptoms of these females are generally more severe than the symptoms of the affected females with normal karyotypes described here, the clinical features were, at least also for the majority of these female patients, milder than those of the affected boys [9,10]. Studies of late replicating X performed in three of the five patients with X-autosome translocations revealed that the normal X-chromosome was inactive, at least in the majority of the analyzed cells [7,9,10].
The milder phenotypes observed in females with normal karyotypes could be attributed to higher expression levels of the normal ATP7A gene. The degree of X-inactivation of the normal X-chromosome correlated with the degree of mental retardation. Highly skewed inactivation of the normal X-chromosome was observed in affected females with severe mental retardation (F6, F7, and probably F2), whereas highly skewed inactivation of the mutant X-chromosome was only observed in one female with mild mental retardation (F9).
It has been shown that certain X-linked genes, among them the ATP7A gene, escape X-inactivation to a certain degree . This phenomenon might affect the phenotype in both group of affected females, those with a balanced X-autosomal translocation and those with a normal karyotype, in which the normal X-chromosome is inactivated. The milder phenotypes, compared to those of affected males might be attributed to expression of normal ATP7A transcript from the inactivated normal X-chromosome.
Few females are affected with MD, and their clinical symptoms are generally milder than those of affected boys with the same mutations. Although the X-inactivation pattern correlates somewhat with presence or absence of MD signs in female carriers, it does not sufficiently explain the observed phenotypes in all of them, as even those with a skewed inactivation of the X-chromosome that harboured the mutation, might have neurological symptoms. Thus, the X-inactivation pattern alone can not be used to predict the phenotypic outcome in female carriers.
List of abbreviations
MD: Menkes disease; MLPA: Multiplex ligation-dependent probe amplification analysis; HUMARA: human androgen receptor.
The authors declare that they have no competing interests.
Generation and analysis of the clinical data: MTZ, AJ, LB, CB, DR, RF, SJ, SM, MA. Mutation identification: LBM, ZT. X-inactivation analysis: LBM. Study concept: LBM. Manuscript draft: LBM. Substantial participation in the design of the project and writing the paper: TGJ, ML, NH, MTZ, AJ, LB, SM. Copper uptake and retention: NH, LBM. All authors participated in the writing of this paper in the context of their individual expertise, and all have read and approved the final version of the article.
We thank Pr R. Touraine, Genetic, CHU St Etienne France for clinical information (F9). We thank Jette Bune Rasmussen for her photographic help with the figures and Susan Peters for valuable comments on the manuscript. We also thank Pia Hougaard for technical assistance. Patients and family members and also colleagues who referred the blood samples and skin biopsies to our laboratory for routine diagnosis are acknowledged.
Human genetics 1991, 86:408-410. PubMed Abstract
Pediatrics 1962, 29:764-779. PubMed Abstract
Clin Genet 1990, 38:452-459. PubMed Abstract
Am J of med Genet 1987, 26:503-510. Publisher Full Text
Clin Genet 1994, 46:295-298. PubMed Abstract
Acta Pathol Jpn 1978, 28:585-594. PubMed Abstract
Sirleto P, Surace C, Santos H, Bertini E, Tomaiuolo AC, Lombardo A, Boenzi S, Bevivino E, Dionisi-Vici C, Angioni A: Lyonization effects of the t(X; 16) translocation on the phenotypic expression in a rare female with Menkes disease.
J Inher Metab Dis 1983, 6:89. PubMed Abstract
Allen RC, Zoghbi HY, Moseley AB, Rosenblatt HM, Belmont JW: Methylation of HpaII and HhaI sites near the polymorphic CAG repeat in the human androgen-receptor gene correlates with X chromosome inactivation.
J Inherit Metab Dis 1989, 12(Suppl 1):207-14. PubMed Abstract