Abstract

To the Editor: The Leber hereditary optic neuropathy (LHON) phenotype was first defined, more than 125 years ago, as a maternally inherited optic neuropathy that primarily affects young adult men (Leber 1871). Loss of central vision may be acute or subacute, and peripheral vision is preserved. Slow delayed recovery can occur, but relative central scotomata usually remain (Johns et al. 1993). The chronic stage of LHON is characterized by optic atrophy (MIM 535000). LHON is associated with three primary mtDNA mutations, all of which occur in genes coding for subunits of complex I of the mitochondrial respiratory chain: G3460A in ND1, G11778A in ND4, and T14484C in ND6 (MIM 516006.0001) (Howell et al. 1995). All three of these mutations alter evolutionarily conserved amino acids and are not found in control individuals. The relative frequency of the three primary LHON mtDNA mutations varies considerably in different populations, although G11778A is the most common worldwide. We have recently found that T14484C is by far the most common mutation in French-Canadian families with LHON (Macmillan et al. 1998). The results of previous studies of mutation profiles of several metabolic disorders have demonstrated founder effects, including phenylketonuria (Rozen et al. 1994) and familial hyperchylomicronemia (De Braekeleer et al. 1991), in the French-Canadian population. To test the hypothesis that the predominance of the T14484C mutation in French-Canadian families with LHON is caused by a founder effect, we sequenced a segment of the mtDNA displacement (D) loop and a segment of the control region from French-Canadian families with the T14484C mutation. Variation in these noncoding regions has been used extensively to study the evolution of modern populations. The regions sequenced included hypervariable regions (HVR) I and II (Vigilant et al. 1989). The D-loop region extends from nucleotide (nt) 189 backward through 0 to nt 16024 of the 16,569-bp mtDNA. The control region extends from nt 189 forward, toward the tRNAs. These two regions contain the fastest-evolving regions of mtDNA (Upholt and Dawid 1977), which have an estimated rate of evolution that is 2.8–5 times that of the remainder of the mitochondrial genome (Aquadro and Greenberg 1983; Cann et al. 1984). The average nucleotide-sequence variation in this region has been calculated to be 1.7% (Aquadro and Greenberg 1983). We reviewed the records of patients with suspected LHON who were independently referred, for molecular diagnosis, to the Montreal Neurological Hospital DNA Diagnostic Laboratory. By cycle sequencing with the use of a New England Biolabs kit (manual) or dye-labeled dideoxy terminators on an ABI system (automated), we sequenced regions that included HVR I and HVR II in the following individuals: one member of each French-Canadian family with the T14484C mutation, one member of a French-Canadian family with the G11778A mutation, and one member of a French-Canadian family without a family history of LHON and with none of the three primary LHON mutations. Sequencing was done in two reactions, by use of the following primer pairs: L15996 and {type:entrez-nucleotide,attrs:{text:H16401,term_id:881221,term_text:H16401}}H16401 (which includes HVR I) and L29 and H408 (which includes HVR II; Vigilant et al. 1989). “L” refers to “light” strand and “H” refers to “heavy” strand, and the numbers refer to the Cambridge sequence (Anderson et al. 1981). The institutional review boards of the Montreal Neurological Hospital and the University of Illinois at Chicago approved this study. We analyzed 27 independently referred French-Canadian families with LHON and the T14484C mutation, all of which were homoplasmic for the T14484C mutation. We found eight homoplasmic transition mutations (C16069T, T16126C, G16213A, A73G, G185A, G228A, A263G, and C295T; table 1), compared with the Cambridge sequence, in the families with the T14484C mutation. Of the 27 families analyzed, 26 shared identical substitutions at all eight sites that were different from the Cambridge sequence. Six of these mutations were found only in families with the T14484C mutation and not in either the family with the G11778A family or the family without LHON (table 1). The mutated sites were distributed throughout the D-loop and control region: one was located in the large central conserved sequence block (CSB; A73G), one (G228A) was in CSB 1, two (T16126C and G16213A) were in HVR I, three (G185A, A263G, and C295T) were in HVR II, and one (C16069T) was just outside HVR I. Table 1 Variable Sites in the D-Loop and Control Regions of mtDNA from French-Canadian Families with LHON and the T14484C Mutation (27 Pedigrees), with LHON and the G11778A Mutation, and without LHON In addition, 22/27 families with LHON and the T14484C mutation had a C insertion in the homopolymeric stretch of C's before the T at position 310. All 27 families with the T14484C mutation, the family with the G11778A mutation, and the family without LHON had a C insertion in the homopolymeric stretch of C's after the T at position 310. Finally, there were some sites at which the sequence appeared to be heteroplasmic; however, these results require independent confirmation. These data demonstrate that French-Canadian families with LHON and the T14484C mutation likely share the same maternal lineage and suggest that they may all have been derived from a single founder woman. All the observed sequence variants—with the exception of the insertions in the homopolymeric tract of C—were transitions, as compared with the Cambridge sequence; the transition-to-transversion ratio of nucleotide substitutions has been reported to be very high in this region (Aquadro and Greenberg 1983). All our patients were positive for mutations at positions 3394, 4216, and 13708, relative to the standard sequence, and it is likely that they belong to the same haplogroup as does the patient described by Brown et al. (1992). This haplogroup is clearly related to white haplogroup J, which is characterized by mutations at positions 4216 and 13708 and which is where T14484C mutations cluster in the European population (Brown et al. 1997; Torroni et al. 1997). The most-similar published haplotype is that of the haplogroup J T14484C “TAS2” individual, in which seven of the mutations (C16069T, T16126C, A73G, G185A, G228A, A263G, and C295T) were common (Howell et al. 1995). There were four additional TAS2 mutations that were not found and three additional TAS2 mutations that were not evaluated in the French-Canadian families. The French-Canadian mutation array is even less similar to those reported for five other haplogroup J T14484C individuals, seven G11778A individuals, and three G3460A individuals (Howell et al. 1995; Hofmann et al. 1997). Furthermore, the French-Canadian array is different from that reported for haplogroup J individuals without LHON (Hofmann et al. 1997). Although the T16213A mutation has been reported in several populations (Horai and Hayasaka 1990; Lum et al. 1994; Mountain et al. 1995), this is the first report of its association with a family with LHON. The single family (pedigree 4) that did not share this mutation may have undergone a reversion mutation to the original sequence; the representative of this family also had a C→A transversion at site 295. The small changes in the D-loop–sequence homopolymeric tracts of C starting at position 303 are variants of the published sequence first reported by Greenberg et al. (1983). The C insertion after the T at position 310 was present in all French-Canadian families, and it was also found in the family with TAS2. However, the C insertion before the T at position 310 was not universally present in the French-Canadian families with the T14484C mutation. These insertions are very unstable and may have arisen after the migration of the common female ancestor to New France. These French-Canadian families with LHON and the T14484C mutation represent a unique genetic resource in which to evaluate the rate of accumulation of sequence variants and the resolution of heteroplasmy in mtDNA on a time scale of a few hundred years.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call