We report on deletion mapping and X inactivation analysis of a gene for X linked non-specific mental retardation (MRX) at Xp21.3–Xp22.11, on the basis of molecular studies in two families with Xp microdeletions involving the DAX-1 gene. In family A, mental retardation (MR) was profound in the older brother with an episode of adrenal crisis, severe in the younger brother with no episode of adrenal crisis, and mild to moderate in the sister and the mother with no signs of adrenal hypoplasia. In family B, MR was absent in the male patient with adrenal hypoplasia. Polymerase chain reaction for 16 loci in the middle of Xp showed that the brothers of family A had a small Xp deletion between DXS7182 and DXS1022, and that the patient of family B had a tiny Xp deletion between DXS319 and DXS1022. Microsatellite analysis for tetranucleotide repeats in the promoter region of the DAX-1 gene and Southern blotting for DAX-1 and DXS28 showed that the sister and the mother of family A were heterozygous for the interstitial deletion. X inactivation analysis for the methylation status of the AR gene and the HPRT gene indicated that the normal X and the deleted X chromosome underwent random X inactivation in both the sister and the mother. The results imply that an MRX gene subject to X inactivation is present in a roughly 4 Mb region between DXS7182 and DAX-1, and that reduced expression of the normal MRX gene caused by random X inactivation results in MR in carrier females.
- MRX gene
- chromosomal localisation
- X inactivation
Statistics from Altmetric.com
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.
The X chromosome carries a number of genes for X linked non-specific mental retardation (MRX).1 2 To date, more than 60 MRX families have been reported with a peak lod score of ⩾2.0, and various X chromosome aberrations involving at least eight different regions have been identified in association with mental retardation (MR). Although significant overlaps exist among the intervals of assignment, it has been suggested that at least 10-12 different MRX genes are required to explain all the regional localisations of MRX genes indicated by linkage studies of MRX families and by molecular analyses of the aberrant X chromosomes.1 2 However, with the exception of FRAXE,3 OPHN1 for MRX60,4 and GDI1 for MRX41 and MRX48,5 MRX genes have not been cloned and, thus, further studies are necessary to isolate the MRX genes.
MR is often manifested by obligate carrier females for a mutant MRX gene.1 This implies that such an MRX gene is subject to X inactivation, and that reduced expression of the normal MRX gene caused by random X inactivation results in MR in carrier females. In support of this, it has been suggested that carrier females with haploinsufficiency of an MRX gene escaping X inactivation are free from MR, because of expression of the normal MRX gene from inactive as well as active X chromosomes.6 However, X inactivation status has not been studied in carrier females with MR.
Here, we report on deletion mapping and X inactivation analysis of an MRX gene at Xp21.3-Xp22.11, on the basis of molecular studies in two families with Xp microdeletions involving the DAX-1 gene7for adrenal hypoplasia congenita (AHC) and hypogonadotrophic hypogonadism (HHG).8
This family consisted of two boys, one girl, and unrelated parents. The older boy (patient 1) was born at 40 weeks of gestation after an uncomplicated pregnancy. At birth, the length was 52.8 cm (+1.8 SD), weight 4500 g (+3.4 SD), and head circumference 35.2 cm (+1.2 SD). At 39 days of age, he was admitted to our hospital because of vomiting, lethargy, failure to thrive (weight 4.4 kg), apnoea, and generalised pigmentation. He was diagnosed as having acute adrenal failure resulting from AHC by biochemical and endocrine studies and was placed on adrenal steroid supplementation therapy. There were no dysmorphic features. Subsequent development was severely retarded with generalised spasticity. At 7 years of age, a CT scan showed brain atrophy and an EEG showed non-specific diffuse slow waves. At 9 years 2 months of age, he was unable to speak single words or stand without support and exhibited growth failure with a length of 105 cm (–2.8 SD), a weight of 21.5 kg (–1.4 SD), and a head circumference of 49.5 cm (–2.2 SD).
The younger brother (patient 2) was the product of an uncomplicated term pregnancy. At birth, length was 54.6 cm (+2.7 SD), weight 4300 g (+2.7 SD), and head circumference 35.0 cm (+1.0 SD). From birth, he was managed in the neonatal intensive care unit because of the AHC of the older brother. Physical examination at birth showed a vigorous infant with mild pigmentation of the lips and nipples, and endocrine studies in the early neonatal period indicated AHC. Thus, adrenal steroid supplementation therapy was successfully started with no episode of adrenal failure and, thereafter, he had no adrenal crisis until 6 years 2 months of age. Nevertheless, he showed definite developmental retardation; he had head control at 6 months old, spoke single words at 2 years 6 months, walked without support at 3.2 years, and spoke two word sentences at 4 years 2 months. At 5 years of age, physical examination showed no distinctive dysmorphic features, and neurological investigation showed no abnormalities, such as spasticity, incoordination, or hypotonia. He responded well to small sounds coming from behind him. Height was 107.9 cm (+0.2 SD), weight 20.2 kg (+1.3 SD), and head circumference 51.0 cm (+0.2 SD). A CT scan and an EEG were grossly normal. He was hyperactive with a pleasant personality, but was not aggressive. He fed himself fairly well, but was unable to dress himself, have meaningful conversations, or follow oral instructions. His developmental age appeared to be less than 2 years, but precise developmental assessment was refused by the parents. He was enrolled in a special school for mentally delayed children. At 6 years 2 months of age, he developed symptoms of adrenal crisis such as lethargy and vomiting, developed bronchopneumonia, and died suddenly at home. Abdominal necropsy showed severe adrenal hypoplasia, but brain necropsy was refused by the parents.
The younger sister was delivered after an uncomplicated term pregnancy, with a length of 51.5 cm (+1.2 SD), weight of 3200 g (+1.0 SD), and head circumference of 34.5 cm (+0.9 SD). She was free from AHC but had moderate MR. At 5 years 1 month of age, clinical assessment was carried out and showed no dysmorphic features or neurological abnormalities. She responded well to small sounds coming from behind her. Her height was 102.3 cm (–1.0 SD), weight 17.2 kg (–0.1 SD), and head circumference 50.6 cm (+0.4 SD). She was somewhat hyperactive with a pleasant personality. She fed and dressed herself, but had obvious difficulty in having meaningful conversations or in following oral instructions. Her developmental quotient was estimated as 63 by the Enjoji method, with verbal development being more severely affected than motor development.
The 32 year old mother also had mild to moderate MR. She had obvious difficulty in daily conversation, although she performed housework and reared her children. There were no minor anomalies, neurological abnormalities, or hearing difficulties. Her height was 162.5 cm (+1.1 SD), weight 57.5 kg (+0.8 SD), and head circumference 56.1 cm (+1.0 SD). She had a submissive personality and was not hyperactive. Intellectual assessment was refused.
The 35 year old father was mentally normal. His height was 170.5 cm (+0.2 SD).
This family has been reported previously.9 In brief, the older brother (patient 3), who was described at 17 years old, is now 26 years of age and has AHC and HHG. The younger maternal half brother, who was documented at 8 years old, is now 18 years of age and also has AHC and HHG. Both patients are mentally normal with average records at standard schools.
Chromosome analysis was performed on 50 peripheral lymphocytes of patients 1-3 and the sister and mother of family A by standard and high resolution G banding.
Genomic DNA was extracted from peripheral leucocytes of patients 1-3 and the sister and mother of family A and was examined by polymerase chain reaction (PCR), microsatellite analysis, and Southern blotting. For PCR analysis, 1 μg of genomic DNA was amplified with the primers defining 16 loci at Xp21.3-Xp22.11 (fig 1). For microsatellite analysis, 1 μg of genomic DNA was amplified with the primers flanking the GGAA tetranucleotide repeat located in the promoter region of the DAX-1 gene,11 and the PCR products were loaded onto a 3% NusieveTM gel (FMC BioProducts) mixed with a standard agarose gel (3:1) to get better resolution. The primer sequences and annealing temperature for the PCR and microsatellite analyses were as reported in Zarianaet al 7 (1AF/1AR), Guoet al 11 (15/3140), and in the Genome Database. For Southern blot analysis, 10 μg of genomic DNA was digested with EcoRI and hybridised with the PCR product of DAX-1 and the C7 probe for DXS28,12together with the probe for the autosomal TK gene13 used as an internal band intensity control.
X INACTIVATION ANALYSIS
The X inactivation pattern was analysed in the sister and the mother of family A, for the methylation status of the AR gene14 and the HPRT gene.15 For the methylation analysis of the AR gene, 1 μg of genomic DNA was amplified with a fluorescently labelled forward primer and an unlabelled reverse primer flanking the polymorphic CAG repeat and two methylation sensitive HpaII sites before and after HpaII digestion. The PCR products were loaded onto an autosequencer (ABI PRISM 310TM) and examined for marker size and peak height by GeneScanTM. The primer sequences and the annealing temperature were as described previously.14 For the methylation analysis of the HPRT gene, 10 μg of genomic DNA was double digested withPvuII/BamHI, with or without subsequent HpaII digestion, and hybridised with the probe pB1.715 by Southern blotting.
The karyotype was apparently normal 46,XY in patients 1-3 and 46,XX in the sister and the mother of family A. However, the middle part of Xp appeared to be somewhat shortened in the X chromosomes of patients 1 and 2 and in one of the two X chromosomes of the sister and the mother.
Representative results are shown in fig 2 and the PCR data of patients 1-3 are summarised in fig 1. Patients 1 and 2 had an interstitial deletion between DXS7182 and DXS1022, and patient 3 had an interstitial deletion between DXS319 and DXS1022 (fig 2, above). Tetranucleotide repeat analysis showed the absence of a band common to the sister and the mother of family A, confirming heterozygosity for the interstitial deletion of the two females (fig 2, below). Southern blot analysis showed that DAX-1 was absent in patients 1-3, and that DXS28 was absent in patients 1 and 2 and present in patient 3 (data not shown). Comparisons of band intensity between DAX-1 and TK and between DXS28 and TK indicated that both DAX-1 and DXS28 were present in a single copy in the sister and the mother of family A (data not shown).
X INACTIVATION ANALYSIS
Representative results are shown in fig 3. The sister was heterozygous for the AR gene polymorphism, with the 276 bp marker of non-maternal (paternal) origin and the 279 bp marker of maternal origin. The peak height ratio between the two markers was 1.3 and 2.8 before and after HpaII digestion, respectively, so that after the compensation for peak height difference contributed by slippage, the ratio of inactive X chromosomes was calculated as 68% for paternally derived X chromosomes and 32% for maternally derived X chromosomes (fig 3, above). The mother was heterozygous for the HPRT gene polymorphism, with the 18 kb band inherited by the sister and the 12 kb band not transmitted to the sister. The intensity of the two bands was similar before and afterHpaII digestion, indicating the occurrence of random X inactivation (fig 3, below).
The deletion analysis showed that patients 1 and 2 with definite non-specific MR had an interstitial Xp deletion between DXS7182 and DXS1022, whereas patient 3 without MR had an interstitial Xp deletion between DXS319 and DXS1022 (fig 1). According to the genetic map reported by Ferrero et al,10the deletional size is estimated to be roughly 4.5 Mb in patients 1 and 2 and roughly 0.5 Mb in patient 3. Although the profound MR in patient 1 could more or less be ascribed to the adrenal crisis in early infancy, severe MR in patient 2 occurred in the absence of adrenal crisis. It is unlikely that DAX-1 deletion results in MR independently of adrenal crisis, because patient 3 was free from MR with DAX-1 deletion. In addition, MR has not been described in patients with DAX-1 mutations, at least in the absence of severe adrenal crisis.8 16 Thus, our results imply the presence of an MRX gene in the approximately 4 Mb critical region between DXS7182 and DAX-1, although it might be possible that an MRX gene resides between DXS1017 and DXS1022 and is deleted in patients 1 and 2 but is preserved in patient 3 (fig 1). The sister and the mother of family A with mild to moderate MR were heterozygous for the microdeletion, which was associated with random X inactivation. Although the X inactivation studies were carried out on peripheral blood cells that are apparently irrelevant to the development of MR (the critical organ would be the central nervous system), the results indicate that the MR gene postulated between DXS7182 and DAX-1 is subject to X inactivation, and that reduced expression of the MRX gene caused by inactivation of the normal X chromosomes results in mild to moderate MR. Since MR was milder in the female carriers than in the male patients in family A, it appears that the degree of MR is primarily determined by the quantity of expression of the MRX gene.
The association of MR with microdeletions around Xp21 has been reported frequently. In particular, Billuart et al 17 identified a roughly 4 Mb deletion between DXS6764 (distal to DXS1202) and DXS704 (fig 1) in a male patient with non-specific MR alone. The result, in conjunction with our results, may further localise the MRX gene to an approximately 2.5 Mb region between DXS7182 and DXS704. MR has also been documented in male patients with various types of contiguous gene deletion syndrome involving the loci for Duchenne muscular dystrophy (DMD), glycerol kinase (GK) deficiency, or AHC.18 MR in most, if not all, such patients may be explained by concomitant loss of the MRX gene proposed here, although MR could be the result of pleiotropic effects of loss of the DMD or GK gene19 20 or of adrenal crisis resulting from AHC. Furthermore, MR has also been reported in female patients heterozygous for interstitial Xp deletions involving the loci for DMD, GK, and AHC.21-23 MR in such female patients may also be ascribed to reduced expression of the MRX gene proposed here, because heterozygosity for DMD, GK, or AHC is unlikely to cause MR. In this context, it is noteworthy that the female patient KC missing the critical region for the MRX gene proposed here has MR under random inactivation of the normal X and the deleted X chromosome.21 22
Linkage studies in MRX families are also consistent with the presence of an MRX gene(s) in the middle of Xp.1 2 In particular, the critical region defined by the present study partially overlaps with the location of MRX10 and is totally included in the locations of eight different MRX loci (table 1). Interestingly, a cryptic deletion around DXS1218 has been identified for MRX34,29 and DXS1218 is shared in common by the regional locations of the nine MRX numbers and is lost from the X chromosome with molecularly defined Xp microdeletions.17 22 Thus, if an MRX gene exists around DXS1218, MR in most, if not all, of the nine MRX families and Xp microdeletion patients may be accounted for by a mutation or deletion of the MRX gene. Although obligate carrier females are free from MR in MRX29, MRX33, and MRX38 (table 1), it is uncertain at this time whether the lack of MR in such carrier females is explained by skewed X inactivation leading to preferential expression of the normal MRX gene proposed here or by impairment of a separate MRX gene(s) with a different character.
In summary, the present study indicates that an MRX gene subject to X inactivation is present between DXS7182 and DAX-1, and that reduced expression of the normal MRX gene caused by random X inactivation results in MR in carrier females. Further studies of similarly affected patients will permit definition of a more precise localisation and characterisation of the MRX gene.
We would like to thank Drs T Yokota and Y Koizumi for helpful comments. This work was supported in part by a grant in aid from the Ministry of Education, Science, Sports and Culture, and a grant for Paediatric Research from the Ministry of Health and Welfare.