Article Text

Download PDFPDF

A common ancestral haplotype in carrier chromosomes from different ethnic backgrounds in vacuolating megalencephalic leucoencephalopathy with subcortical cysts
  1. Y Shinar1,*,
  2. B Ben-Zeev2,*,
  3. N Brand2,
  4. H Lahat3,
  5. V Gross-Zur4,
  6. D MacGregor5,
  7. T Bahan3,
  8. D L Kastner6,
  9. E Pras3
  1. 1Heller Institute of Medical Science, Sheba Medical Centre, Tel Hashomer 52621, Israel
  2. 2Paediatric Neurology Unit, Sheba Medical Centre, Tel Hashomer 52621, Israel
  3. 3Danek Gartner Institute of Human Genetics, Sheba Medical Centre, Tel Hashomer 52621, Israel
  4. 4Paediatric Neurology Unit, Shearey Zedek Medical Centre, Jerusalem 81031, Israel
  5. 5Division of Neurology, Hospital for Sick Children, Toronto, Canada M5G1X8
  6. 6Arthritis and Rheumatism Branch, National Institute of Arthritis and Musculoskeletal and Skin Diseases, National Institutes of Health, Bethesda, Maryland 20892, USA
  1. Correspondence to:
 Dr Pras, Danek Gartner Institute of Human Genetics, Sheba Medical Centre, Tel Hashomer 52621, Israel;

Statistics from

Request Permissions

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.

Vacuolating megalencephalic leucoencephalopathy with subcortical cysts (VL) is a newly described, inherited leucodystrophy (MIM 604004). Clinically, the disease is characterised by accelerated head growth beginning in the first year of life and resulting in extreme macrocephaly and mild delays in gross motor milestones. In most cases, these early manifestations are followed by evolution of pyramidal symptoms and signs, cerebellar ataxia, epilepsy, and in older patients dystonia and athetosis.1 Cognitive function is relatively spared in most patients. Brain imaging with computed tomography (CT) or magnetic resonance imaging (MRI) shows diffuse cerebral white matter swelling with progressive cystic-like changes, prominent in the frontotemporal regions, with preservation of grey matter structures.1,2 Pathological specimens from VL patients showed splitting of the myelin sheaths between the lamellae consistent with an oedematous process, with sparing of the exons.3 Although progressive in nature, VL is characterised by a relatively mild clinical course compared to the severity of the neuroradiological findings.4,5 About 70 cases of the disease have been described in different ethnic groups. The molecular basis of this disorder remains unknown. The inheritance is autosomal recessive and the disease gene was recently mapped to a 3 cM interval between D22S1161 and the telomere of chromosome 22q.6 Linkage was established in a group of 13 Turkish families all originating from rural areas of central and south eastern Anatolia. No shared alleles or shared haplotypes were detected between the Turkish families.


Six of the seven families included in this study (fig 1) have been described in detail by Ben Zeev et al.7 Families 1, 2, 4, and 6 are of Libyan Jewish origin and family 3 is of Turkish Jewish origin. Family 7 is of non-Jewish Indian origin and family 5 is of mixed ancestry. The father is of Libyan Jewish origin and the mother is Ashkenazi. The parents in families 2, 3, and 4 are first cousins, while the parents in the other families are unrelated. The study was approved by the Helsinki Committee at the Sheba Medical Centre and participants gave informed consent. Computed screening of the full chromosome 22 sequence, telomeric to D22S922, showed two new CA repeats located in clones WI14811 and STS28616. These polymorphic repeats were amplified with the primers 5′-GGAGAAT CACTTAAACTCAG-3′ and 5′-TTCAGCAGTTTTTCTGTCCC-3′, 5′-TGGAAGAAAGAAATCTCAAA-3′ and 5′-TGAACTCAAGGT TTGCTAAG-3′, respectively. The markers N66C4 and ARSA6 were amplified with primers 5′-TGTACATCCTTACTGCTCG-3′ and 5′-ACGGCAGTGGGGAAACACAA-3′, 5′-CCGGCCAAA AATGACTTTTA-3′ and 5′-CTGGAAAGAGCAAGACCCTG-3′, respectively. Amplification was carried out as described elsewhere.8 Lod scores were calculated with the LINKAGE (version 5.1) package of programs, assuming recessive inheritance and a disease allele frequency of 0.004. Haplotypes were constructed so as to minimise recombinants. In the consanguineous families, a single carrier chromosome was counted, while two such chromosomes were counted in the non-consanguineous families. DNA samples from 25 normal Libyan Jews were also typed for these markers and, together with the non-carrier chromosomes from the VL families, were used as controls (overall 58 control chromosomes).

Figure 1

Typing for five markers on chromosome 22q (from top to bottom): D22S922, WI14811, STS28616, ARSA, and N66C4. Carrier chromosomes in the parents and affected subjects are shown in boxes.


A schematic physical map of the region stretching between D22S922 and the telomere is presented in fig 2. In order to place the markers accurately on the map, genomic clones containing these markers were compared to the full sequence of chromosome 22q. Two point lod scores between the disease and five chromosome 22q markers are presented in table 1. A maximal two point lod score of 5.93 was obtained with STS28616 at 𝛉=0.00. Four of the five markers yielded peak lod scores >3.00 at 𝛉=0.00. Haplotype analysis presented in fig 1 is consistent with linkage to 22q in all of the families. A recombination event in subject 2-05 for D22S922 defines this marker as the centromeric boundary. A haplotype constructed with the four distal markers used in this study, WI14811, STS28616, ARSA, and N66C4 (alleles 2, 2, 3, and 4, respectively), was present in seven of eight Libyan Jewish carrier chromosomes but in none of the 61 control Libyan Jewish chromosomes (p<0.0001). Even more interesting is the fact that 10 of the 11 carrier chromosomes in this study share the same alleles of the markers STS28616, ARSA, and N66C4 (alleles 2, 3, and 4, respectively), while none of the 68 control chromosomes share these alleles (p<0.0001). The only carrier chromosome that does not share these three alleles is that of the Ashkenazi mother in family 5 (5-02). This chromosome, however, does share with all the other carrier chromosomes the alleles of the two most telomeric markers, which appear only in two of 68 control chromosomes (p<0.0001).

Table 1

Two point lod scores between VL and chromosome 22q markers

Figure 2

Schematic map of the interval containing the VL gene on chromosome 22q. Numbers under the horizontal line represent the distance in MB from the top of the chromosome.


Topcu et al6 mapped the gene causing VL to a 3 cM interval between D22S1161 and the telomere, a region spanning 2 Mb. In this report we show that seven VL families from four different ethnic origins all map to the same region on chromosome 22q, thus providing evidence for genetic homogeneity. The gene causing the disease is located very close to the telomere of 22q, a region which is not well covered by the polymorphic markers provided in genome screening kits. Indeed, only by using markers developed by Topcu et al6 and by ourselves were we able to detect linkage in our families. A haplotype constructed with the four telomeric markers was found in most of the Libyan Jewish carrier chromosomes but in none of the control Libyan Jewish chromosomes, suggesting a common founder. When only the three most telomeric markers were included, the haplotype extended to include all but one of the carrier chromosomes tested in this study, but none of the control chromosomes. This may reflect an ancient common mutation segregating in these families. If this is the case, the deviation from the historical haplotype at marker WI14811 in affected members from families 1, 3, and 7 may have resulted from recombinants that have occurred in past generations in these families. When attempting to narrow an interval that contains a candidate gene, analysis of recombinants is usually the first stage. A fairly large interval may be left after all recombinant events are used, especially when dealing with a small panel of families. In contrast, analysis of historical recombinants is a very powerful tool, owing to the fact that it takes advantage of recombination events that have occurred over multiple generations and not only the last one. The size of an interval retaining remnants of an ancestral chromosome is inversely proportional to the number of generations that have elapsed since a given mutation was introduced into a population and to the recombination rate. If the assumption that all the carrier chromosomes sharing alleles for N66C4, ARSA, and STS2861 have a common founder is correct, the deviation from the ancestral chromosome observed at WI14811 results from a historical recombination that has occurred between WI14811 and STS28616. This confines the disease gene to a 1 000 000 bp region between WI14811 and the telomere. When the haplotype is constructed only from the two most telomeric markers, N66C4 and ARSA, the haplotype formed is common to all carrier chromosomes. This theoretically could reflect an ancestral recombination between STS28616 and ARSA, thus reducing the gene interval to <400 000 bp. However, since this haplotype comprises only two markers, the data should be interpreted cautiously. A search for additional polymorphisms from this region, especially from the interval between ARSA and N66C4, may substantiate the reduction of the interval to less than 0.4 Mb.

The region between WI14811 and the telomere contains 12 known genes and five ESTs. All of the 12 genes and one of the ESTs are expressed in the CNS, as well as in many other tissues. Sequencing of these genes and other expressed sequences from the region will eventually identify the gene causing this disease.


We would like to thank Mr Etgar Levy-Nissenbaum for his help. This study was supported by the Israeli Ministry of Health, grant No 4352, the Ministry of Science, grant No 6279, and by the Benjamin and Rebecca Bernard Memorial Fund, Sackler Faculty of Medicine, Tel Aviv University.



  • * These authors contributed equally to this work.