Progress in genetic counselling and prenatal diagnosis of maternally inherited mtDNA diseases

Neuromuscular Disorders 10 (2000) 484±487

www.elsevier.com/locate/nmd

Progress in genetic counselling and prenatal diagnosis of maternally inherited mtDNA diseases Joanna Poulton*, David R. Marchington Department of Paediatrics, University of Oxford, John Radcliffe Hospital, Oxford OX3 9DU, UK Received 21 December 1999; received in revised form 24 February 2000; accepted 26 February 2000

Abstract Mitochondrial DNA is almost entirely maternally inherited. Thousands of copies of mitochondrial DNA are present in every nucleated cell and in most normal individuals these are virtually identical (homoplasmy). Mitochondrial DNA diseases may be caused by mutations in either mitochondrial (Nature 1988;331:717±719) or nuclear genes (Nature 1989;339(6222):309±311; Br J Hosp Med 1996;55:712±716) and hence give rise to maternal or autosomal patterns of inheritance. Antenatal diagnosis of mitochondrial diseases based on chorionic villus sampling is available for Mendelian disorders and the syndromes caused by mutations at bp 8993 (associated with both Leigh's syndrome or neurogenic weakness ataxia and retinitis pigmentosa). However, prenatal diagnosis of many other maternally inherited mitochondrial DNA diseases is less reliable because it is not possible to predict the way in which heteroplasmic mitochondrial DNA mutations segregate within tissues with con®dence. This review focuses on the substantial progress that has been made recently, and on the applicability of prenatal diagnosis to genetic counselling in this ®eld. q 2000 Elsevier Science B.V. All rights reserved. Keywords: Genetic counselling; Prenatal diagnosis; Mitochondrial DNA

1. Introduction Despite substantial advances in mitochondrial genetics, prenatal diagnosis has not been available for the majority of mtDNA diseases. Unaffected female relatives of patients with a maternally inherited mtDNA point mutation who have a low risk of transmitting the disorder may therefore choose not to have children because of their perception of the recurrence risks. However, this situation is changing.

2. The biology of mtDNA Mitochondrial DNA (mtDNA) encodes polypeptides involved in electron transport [1] and is maternally inherited. Unlike nuclear DNA, where there are only two copies of each gene per cell, thousands of copies of mtDNA are present in every nucleated cell. Normal individuals are homoplasmic (that is, virtually all of their mtDNAs are identical). Heteroplasmy (the presence of both normal and mutant mtDNA in a single individual) is present in the vast majority of mtDNA diseases, so that the proportion of mutant mtDNA in any cell or tissue may vary from 0±100%. The polypeptides encoded by mtDNA are all subunits of * Corresponding author.

the respiratory chain. This is a highly complex array of multimeric enzymes which generate ATP. The majority of these are encoded in the nucleus, as are very many mitochondrial proteins. Hence there are potentially a large number of diseases with a Mendelian pattern of inheritance. 3. Prenatal diagnosis in Mendelian mitochondrial diseases Over the past year there have been a number of publications on the cloning of nuclear genes for mitochondrial diseases [2] (Table 1). Prenatal diagnosis is possible for these diseases, as with so many other Mendelian disorders when the mutation can be identi®ed. Surf 1 is numerically the most important. A number of groups have shown that mutations in this gene underlie 75% of cytochrome oxidase de®cient Leigh's syndrome, that is about 10±20% of all Leigh's syndrome [3]. In addition mutations have been identi®ed in various subunits of complex 1, in proteins involved in assembly [4] or transport of macromolecules into the mitochondrion [5], and in one involved in turnover of nucleotides [6]. The genes for the other better known Mendelian disorders affecting mtDNA such as autosomal dominant proximal external ophthalmoplegia (AdPEO) and mtDNA depletion have not yet been identi®ed, so that

0960-8966/00/$ - see front matter q 2000 Elsevier Science B.V. All rights reserved. PII: S09 60-8966(00)0013 0-9

J. Poulton, D.R. Marchington / Neuromuscular Disorders 10 (2000) 484±487

485

Table 1 Nuclear gene mutations causing mitochondrial disease

mutant mtDNA in different tissues, different cells within a tissue and perhaps even different mitochondria.

Complex 1 de®ciency Flavoproteins: NDUFV1 Iron sulphur protein: NDUFS4 Other electron transfer proteins NDUFS7 NDUFS8

5. Studies of the mitochondrial bottleneck

Complex II de®cient Leigh syndrome Flavoprotein SDHFp Complex IV de®ciency Leigh syndrome: Surf-1 Cardioencephalomyopathy: Sco 2 [4] Autosomal recessive MNGIE [6] Thymidine phosphorylase Autosomal recessive spastic paraparesis Paraplegin: metalloprotease [26] Friedreich's ataxia [27] Frataxin a X-linked deafness dystonia syndrome [8]: DDP a Frataxin is mutated in Friedreich's ataxia, resulting in an iron transport defect and secondary respiratory chain dysfunction that may contribute to the cardiac pathology. It is included at the request of the editor.

prenatal diagnosis is not available. The remainder of this article will focus on genetic advice to individuals with maternally inherited mtDNA disease.

4. Heteroplasmy and progressive mtDNA disease Heteroplasmy with variation in the level of mutant mtDNA between tissues is common in mitochondrial diseases. There appears to be a threshold effect and symptoms arise in tissues with a high level of mutant. Accumulation of mutant mtDNAs in affected tissues appears to explain the progressive nature of these disorders [7,8]. For instance, in one patient the proportion of a mutant mtDNA (i) rose successively in sequential muscle biopsies (ii) was ,0 in cultured cells (iii) fell after local muscle damage, because the mutant load in satellite cells was lower than in the mature muscle ®bres. The vast majority of mtNA is maternally inherited. When there is a point mutation difference between a mother and her offspring, there may be complete switching of mtDNA type in a single generation: that is, each was homoplasmic with regard to that base. Because oocytes contain approximately 100 000 mtDNAs and yet the mutation probably only occurs once, there must be a restriction followed by an ampli®cation in numbers of mtDNAs whereby the mutant mtDNA becomes the mitochondrial founder for the child. We refer to this as a genetic `bottleneck'. After birth there may also be segregation of mtDNAs so that affected individuals have different levels of

We investigated the bottleneck in normal oocytes from couples in our IVF clinic referred for male infertility. We used naturally occurring length variation in the large noncoding region of mtDNA. This region of mtDNA deviates from the rule that homoplasmy is the norm, as some individuals are heteroplasmic for different length variants. We showed that there is little or no difference in the frequency distribution of length variants between several different tissues from any normal individual. We studied a heteroplasmic length variant in oocytes from controls and from a patient with a pathogenic mtDNA mutation and showed that segregation of founder mtDNA molecules has probably occurred by the time the oocytes are mature [9]. We demonstrated that no such segregation occurred in multiple samples of placenta. Further studies are essential as the apparent bottleneck size may depend on the mtDNA mutation. For instance, segregation was very marked in a family carrying the mtDNA mutation at position T8993G [10] compared with a patient with the mtDNA rearrangement [11]. Four groups have recently constructed heteroplasmic mouse models of mtDNA segregation by introducing donor cytoplasm into a fertilized recipient mouse egg [12± 15]. Analysis of developing female germ cells demonstrated that the major component of the bottleneck occurs between the primordial germ cell and primary oocyte stage. These data imply that the major component of the bottleneck has occurred by the time oocytes are mature. If mutant mtDNA remains uniformly distributed among individual cells of the embryo it should be possible to assess the level of mutant mtDNA prenatally, by sampling either chorionic villus (CVS) or pre-implantation embryos. In a recent study, every blastomere in pre-implantation embryos derived from heteroplasmic oocytes in mouse [13,16] contained very similar levels of mutant mtDNA. A comparable result was obtained when blastomeres of a preimplantation embryo from a woman with the A3243G mutation were analyzed, Jansen et al. (private communication) Few carrier/affected fetuses have been analyzed for load of mutant mtDNA in different tissues. However, the limited existing data from studies on human fetuses with pathogenic mtDNA mutations also suggest that mutant mtDNAs do not segregate much during embryogenesis [17±19]. In two of the animal models of heteroplasmic mtDNA segregation the proportion of each mtDNA variant was uniform in all tissues of the fetuses analyzed. Taken together, these studies suggest that a major bottleneck occurs during oogenesis and that mtDNA does not segregate much during embryogenesis. However, in two of these experiments there was tissuespeci®c, directional selection for different mtDNA geno-

486

J. Poulton, D.R. Marchington / Neuromuscular Disorders 10 (2000) 484±487

types in the same animal [15,20]. These studies used nonpathogenic mtDNA variants but are consistent with the limited studies which have been performed on pathogenic variants. As yet there are no animal models using detrimental mtDNA mutations to further our understanding of these processes. 6. Prenatal diagnosis of mtDNA disease DNA based CVS is useful for prenatal diagnosis of Mendelian disorders such as Surf 1 in cytochrome oxidase de®cient Leigh's syndrome where the mutations have been identi®ed. Precise recommendations regarding prenatal diagnosis for maternally inherited mtDNA diseases have been formulated at a recent ENMC workshop and will be published shortly [21]. These recommendations depend on the particular mutation and hence the recommendation of the workshop was that specialist advice be sought in counselling these patients. Currently the options open to women with mtDNA disease are: 1. Oocyte donation: in practice, there is a limited supply of donors and maternal relatives such as sisters are high risk. 2. Preimplantation diagnosis: current data suggest that the varied tissue distribution of mtDNA mutants, which is found postnatally, has not developed in the preimplantation embryo in which heteroplasmic mtDNA is uniformly distributed between blastomeres [13,16]. 3. Chorionic villus sampling: little is known regarding the tissue distribution of mtDNA mutants in the developing fetus. Such evidence as exists suggests that the mutant load in extra-embryonic tissues probably re¯ects that of the fetus [22]. 4. There may be many asymptomatic maternal relatives who feel unable to risk having children because of these uncertainties. Estimations of recurrence risks based on blood levels of mutant mtDNA are reasonable in some types of mutation but may be very inaccurate in others [23]. Oocyte sampling might be able to estimate recurrence risks more accurately particularly in `private mutations' where there are minimal historical data. 7. Requirements for mtDNA prenatal diagnosis A major problem for genetic counselling is that the correlation between phenotypic severity and level of mutant is poor in many mtDNA diseases. Prenatal diagnosis would be easy if and only if there were (1) a close correlation between load of mutant mtDNA and disease severity, (2) uniform distribution of mutant in all tissues, and (3) no change in mutant load with time. These are ful®lled in a families with mutations at bp 8993 but not in the majority of mtDNA disorders [22±25].

Recurrence risks for a severe phenotype based on maternal level of mutant have been estimated in a handful of disorders, most accurately for mutations at bp 8993 [22]. Such data are limited by the accuracy with which the level of mutant mtDNA in oocytes can be predicted from the tissues available for analysis. In the case of mitochondrial encephalomyopathy lactic acidosis and stroke- like episodes (MELAS) due to mutations at bp 3243, blood levels of mutant mtDNA may be very misleading. We suggest that there are situations where preconceptual genetic counselling based on oocyte sampling could be useful. It is probable that this should be restricted to women where prenatal diagnosis can subsequently be carried out. Pre-implantation diagnosis has some theoretical advantages but is not widely available in the UK. As with all IVF procedures the rate of achieving pregnancy is substantially lower than natural conception. It is therefore most applicable to women with a very high recurrence risk. CVS has been successfully carried out in several women carrying the NARP mutation. In all cases the mutant load was either high or low, enabling accurate predictions. However, caution is needed for other mtDNA disorders in which the correlation between severity and mutant load is less precise. As with all of the options, there is a risk that CVS may be unhelpful where the mutant load has an intermediate level. We anticipate that our understanding of the transmission genetics and segregation of mtDNA mutants will be revolutionized by new ®ndings based on the animal models of mtDNA disease currently being developed and that prenatal diagnosis will become routine thereafter.

References [1] Attardi G, Schatz G. Biogenesis of mitochondria. Annu Rev Cell Biol 1988;4:289±333. [2] Smeitink vdHL J. Human mitochondrial complex I in health and disease. Am J Hum Genet 1999;64:1505±1510. [3] Tiranti V, Jaksch M, Hofmann S, et al. Loss-of-function mutations of SURF-1 are speci®cally associated with Leigh syndrome with cytochrome c oxidase de®ciency. Ann Neurol 1999;46(2):161±166. [4] Papadopoulou LCSC, Davidson MM, Tanji K, et al. Fatal infantile cardioencephalomyopathy with COX de®ciency and mutations in SCO2, a COX assembly gene. Nat Genet 1999;23:333±337. [5] Koehler CMLD, Merchant S, Renold A, Junne T, Schatz G. Human deafness dystonia syndrome is a mitochondrial disease. Proc Natl Acad Sci USA 1999;96:2141±2146. [6] Nishino I, Spinazzola A, Hirano M. Thymidine phosphorylase gene mutations in MNGIE, a human mitochondrial disorder. Science 1999;283(5402):689±692. [7] Poulton J, O'Rahilly S, Morten K, Clark A, Mitochondrial DNA. diabetes and pancreatic pathology in Kearns-Sayre syndrome. Diabetologia 1995;38:868±871. [8] Weber K, Wilson J, Taylor L, Brierley E, Johnson M, Turnbull D. A new mtDNA mutation showing accumulation with time and restriction to skeletal muscle. Am J Hum Genet 1997;60:373±380. [9] Marchington D, Hartshorne G, Barlow D, Poulton J. Homoploymeric tract heteroplasmy in mtDNA from tissues and single oocytes: support for a genetic bottleneck. Am J Hum Genet 1997;60:408±416.

J. Poulton, D.R. Marchington / Neuromuscular Disorders 10 (2000) 484±487 [10] Blok R, Cook D, Thorburn D, Dahl H. Skewed segregation of the mtDNA nt 8993 (T ! G) mutation in human oocytes. Am J Hum Genet 1997;60(6):1495±1501. [11] Marchington D, Hartshorne G, Barlow D, Poulton J. Evidence from human oocytes for a genetic bottleneck in a mitochondrial DNA disease. Am J Hum Genet 1998;63:769±775. [12] Laipis P. Construction of heteroplasmic mice containing two mitochondrial DNA genotyoes by micromanipulation of single-cell embryos. Methods Enzymol 1996;264:345±357. [13] Jenuth J, Peterson A, Fu K, Shoubridge E. Random genetic drift in the female germline explains the rapid segregation of mammalian mitochondrial DNA. Nat Genet 1996;14(2):146±151. [14] Meirelles F, Smith L. Mitochondrial genotype in a mouse heteroplasmic lineage produced by embryonic karyoplast transplantation. Genetics 1997;145:445±451. [15] White S. Molecular mechanisms of mitochondrial disorders (PhD). Melbourne, Australia: University of Melbourne, 1999. [16] Molnar M, Shoubridge E. Preimplantation diagnosis for mitochondrial disorders. Neuromusc Disord 1999;9(6-7):521. [17] Suomalainen A, Majander A, Pihko H, Peltonen L, Syvanen AC. Quanti®cation of tRNA3243(Leu) point mutation of mitochondrial DNA in MELAS patients and its effects on mitochondrial transcription. Hum Mol Genet 1993;2(5):525±534. [18] Harding AE, Holt IJ, Sweeney MG, Brockington M, Davis MB. Prenatal diagnosis of mitochondrial DNA8993 T±±G disease. Am J Hum Genet 1992;50(3):629±633. [19] Matthews PM, Hopkin J, Brown RM, Stephenson JB, Hilton-Jones D, Brown GK. Comparison of the relative levels of the 3243 (A ! G)

[20] [21] [22] [23]

[24]

[25] [26]

[27]

487

mtDNA mutation in heteroplasmic adult and fetal tissues. J Med Genet 1994;31(1):41±44. Jenuth J, Peterson A, Shoubridge E. Tissue-speci®c selection for different mtDNA genotypes in heteroplasmic mice. Nat Genet 1997;16:93±95. Poulton J, Turnbull D. 74th ENMC International Workshop: Mitochondrial Diseases. Neuromusc Disord 2000;10:460±462. White S, Collins V, Wolfe R, et al. Genetic counseling and prenatal diagnosis for the mitochondrial DNA mutations at nucleotide 8993. Am J Hum Genet 1999;65:474±482. Chinnery PF, Howell N, Lightowlers R, DM T. The inheritance of MELAS and MERRF: the relationship between maternal mutation load and the frequency of affected offspring. Brain 1998;121:1889± 1894. White SL, Shanske S, McGill JJ, Mountain H, Geraghty MT, DiMauro S, Dahl HH, Thorburn DR. Mitochondrial DNA mutations at nucleotide 8993 show a lack of tissue- or age-related variation. J Inher Metab Dis 1999;22(8):899±914. Chinnery PFHN, Lightowlers R, Turnbull DM. Molecular pathology of MELAS and MERRF: the relationship between mutation load and clinical phenotype. Brain 1997;120:1713±1721. Casari G, De Fusco M, Ciarmatori S, et al. Spastic paraplegia and OXPHOS impairment caused by mutations in paraplegin, a nuclearencoded mitochondrial metalloprotease. Cell 1998;93(6):973± 983. Rotig ADLP, Chretien D, Foury F, et al. Aconitase and mitochondrial iron-sulphur protein de®ciency in Friedreich ataxia. Nat Genet 1997;17:215±217.