Statistics from Altmetric.com
Editor—Severe, isolated, and generalised deficiency of complex IV (cytochrome c oxidase, COX) can result in Leigh syndrome (LS) (MIM 256000), an early onset mitochondrial disorder characterised by rapidly progressive, symmetrical degeneration of the brain stem, diencephalon, and basal ganglia.1 2 SURF-1, a gene located on chromosome 9q34, has recently been identified as the gene responsible for numerous cases of LSCOX.3 4
SURF-1 associated LSCOX is usually inherited as an autosomal recessive trait. We report here a homozygous loss of function mutation ofSURF-1 in two monozygotic LSCOXfemale twins, owing to uniparental disomy of two almost identical maternal chromosomes 9.
The probands were born to non-consanguineous parents at 33 weeks of gestation by caesarean section. The mother was 46 years old. The pregnancy was uneventful until the 24th week, when persistent uterine contractions ensued. Two older sibs of the probands are alive and healthy. The family history was negative for neurological or metabolic disorders. Birth weight and body length were <3rd centile and the Apgar score was 1-8 in both patients. Growth rate and psychomotor development were regular during the first 8 months of life. During the following months the patients developed a rapidly progressive clinical syndrome characterised by failure to thrive, psychomotor regression, hypotonia, ophthalmoparesis, mild bilateral optic atrophy, and ataxia. At 18 months both patients had mild lactic acidosis. MRI showed symmetrical paramedian lesions in the mesencephalon and brain stem, as typically found in LS. Both patients died of respiratory failure in the third year of life.
Needle muscle biopsies performed at 24 months of age showed a diffuse reduction of the histoenzymatic reaction to COX. Biochemically, COX activity in muscle homogenates was 12.1 nmol/min/mg in one patient and 3.6 nmol/min/mg in the second (normal values 68 ± 20), while the activities of the other respiratory complexes were all normal. The COX defect was also detected in cultured fibroblasts of one patient (0.4 nmol/min/mg, normal value 25 ± 11), but this assay was not performed in the second patient. Specific activities of the respiratory complexes in a muscle homogenate of the mother were all normal.
Automated sequence analysis of the nine exons of theSURF-1 gene in the probands showed the presence of a previously reported3 homozygous frameshift mutation (751C>T). This mutation destroys aBsiWI restriction site, which is present in the wild type gene. BsiWI RFLP analysis showed a heterozygous mutation in the mother, while no mutation was detected in the probands' father, sister, or brother (fig 1). A de novo mutation in the paternal chromosome 9 identical to the mutation carried by the mother was considered unlikely. Non-paternity was excluded by linkage analysis with numerous microsatellite markers. To test the hypothesis of chromosome 9 specific paternal non-contribution, we then analysed three STSs (D9S1831, D9S1826, and D9S158), flanking the SURF-1 locus at 9q34. All three markers showed the presence of one maternal allele only, while the paternal allele was consistently absent. To verify whether the paternal non-contribution was the result of a microdeletion at 9q34, the cosmid P117B6, which contains the SURF-1gene,5 was used as a probe in FISH experiments3 6 on metaphases from one proband. The probe detected two comparable signals on both chromosome 9 homologues (fig2). These results excluded the presence of a deletion in a paternal chromosome, suggesting instead a mechanism of uniparental disomy (UPD) of two maternal chromosomes. To test this hypothesis, additional microsatellites distributed along the whole of chromosome 9 were analysed for a total of 22 markers (fig 1). With the exception of two small regions (D9S288-D9S286, and D9S167-D9S283-D9S287, see fig 1) the alleles were all homozygous; in 15 instances the obligate contribution of the paternal allele was unequivocally missing. In particular, homozygosity was detected for 10/10 markers encompassing theSURF-1 locus, in the interval defined by markers D9S1831-D9S158. We conclude that loss of the contribution of a second normal SURF-1 allele has led to the manifestation of LS in our patients.
UPD is defined as the exceptional inheritance of a pair of chromosomes from one parent only, as the result of gamete complementation, chromosome loss in trisomy, or duplication in monosomy. In isodisomy, the uniparental pair is a duplicate of the same chromosome DNA template, and causes an increased risk of a recessive disorder by reduction to homozygosity.7 In our patients the presence of two small heterozygous regions can be explained as the result of two crossing over events in otherwise identical maternal chromosomes 9. These data indicate that the double maternal contribution was the result of a non-disjunction which occurred at the second meiotic division, with maintenance of euploidy in the zygote by elimination of the paternal contribution. The age of the mother (46 years) could have favoured the non-disjunction event in our patients, as it is known that the risk of such an event increases with maternal age.
In addition, the haplotype reconstruction showed evidence of two recombinant events in the probands' sibs, close to theSURF-1 locus (fig 1). The brother shares with the probands the maternal allele for D9S186, but not that for D9S1861; the sister shares with the probands the maternal alleles for D9S1826 and D9S158, but not that for D9S1818. Since both sibs are homozygous wild type for SURF-1, these recombination events indicate that the disease locus is contained within the interval between markers D9S1861 and D9S1826.3
UPD may also cause functional balance disruption of imprinted genes.7 The existence of imprinted genes on chromosome 9 is controversial, but it seems unlikely.8-10 Our patients did not show gross dysmorphic features or malformations apart from LS. With the limitations because of the brief survival and severe phenotype, this observation suggests that chromosome 9 does not contain maternally imprinted genes crucial for embryonic development.
We are indebted to Ms Barbara Geehan for revising the manuscript and to Mr Franco Carrara and Mr Matteo Granatiero for technical assistance. This work was supported by Fondazione Telethon-Italy (grant No 767 and grant No E672), Ministero dell'Università e della Ricerca Scientifica e Tecnologica, Associazione Italiana Ricerca sul Cancro (AIRC), Ricerca Finalizzata Ministero della Sanità No ICS 120.2/RF96.348, and European Union Human Capital and Mobility network grant “Mitochondrial Biogenesis in Development and Disease”.
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.