Mtdna Sequence Polymorphism in CADASIL

Neurogenetics (2006) 7: 185-194 (corrected version)

DOI 10.1007/s10048-006-0049-x

Johanna Annunen-Rasila · Saara Finnilä · Kati Mykkänen · Jukka S. Moilanen · Johanna Veijola · Minna Pöyhönen · Matti Viitanen · Hannu Kalimo · Kari Majamaa

Mitochondrial DNA sequence variation and mutation rate in patients with CADASIL

Abstract Mutations in the NOTCH3 gene cause cerebral autosomal dominant arteriopathy with subcortical infarcts and leukoencephalopathy (CADASIL), which is clinically characterized by recurrent ischemic strokes, migraine with aura, psychiatric symptoms, cognitive decline and dementia. We have previously described a patient with CADASIL caused by a R133C mutation in the NOTCH3 gene and with a concomitant myopathy caused by a 5650G>A mutation in the MTTA gene in mitochondrial DNA (mtDNA). We assume that the co-occurrence of the two mutations is not coincidental and that mutations in the NOTCH3 gene may predispose the mtDNA to mutations. We therefore examined the nucleotide variation in the mtDNA coding region sequences in 20 CADASIL pedigrees with 77 affected patients by conformation-sensitive gel electrophoresis and sequencing. The sequence variation in mtDNA was then compared with that among 192 healthy Finns. A total of 179 mtDNA coding region sequence differences were found relative to the revised Cambridge reference sequence, including four novel synonymous substitutions, two novel nonsynonymous substitutions and one novel tRNA substitution. We found that maternal relatives in two pedigrees differed from each other in their mtDNA. Furthermore, the average number of pairwise differences in sequences from the 41 unrelated maternal lineages with CADASIL was higher than that expected among haplogroup-matched controls. The numbers of polymorphic sites and polymorphisms that were present in only one sequence were also higher among the CADASIL sequences than among the control sequences. Our results show that mtDNA sequence variation is increased within CADASIL pedigrees. These findings suggest a relationship between NOTCH3 and mtDNA.

Keywords CADASIL · mtDNA · DNA sequence analysis · Genetic variation · NOTCH3

J. Annunen-Rasila · S. Finnilä · J. S. Moilanen · J. Veijola · K. Majamaa

Department of Neurology, University of Oulu, Oulu, Finland

J.Annunen-Rasila · S. Finnilä · J. Veijola · K. Majamaa

Clinical Research Center, Oulu University Hospital, Oulu, Finland

K. Mykkänen

Department of Medical Genetics, University of Turku, Turku, Finland

J. S. Moilanen

Department of Clinical Genetics, Oulu University Hospital, Oulu, Finland

J. S. Moilanen

Institute of Medical Technology, University of Tampere, Tampere, Finland

M. Pöyhönen

Department of Medical Genetics, University of Helsinki, Helsinki, Finland

M. Pöyhönen

Department of Clinical Genetics, Helsinki University Central Hospital, Helsinki, Finland

M. Viitanen

Department of Geriatric Medicine, University of Turku, Turku, Finland

M. Viitanen

Division of Clinical Geriatrics, Karolinska Institute, Karolinska University Hospital, Stockholm, Sweden

H. Kalimo

Department of Pathology, University of Helsinki, Helsinki, Finland

H. Kalimo

Department of Pathology, Helsinki University Central Hospital, Helsinki, Finland

H. Kalimo

Department of Pathology, University of Turku, Turku, Finland

K. Majamaa ()

Department of Neurology, University of Turku, 20014 Turku, Finland

e-mail:

Tel: +358-2-3131732

Fax: +358-2-3132737

Introduction

Cerebral autosomal dominant arteriopathy with subcortical infarcts and leukoencephalopathy (CADASIL) (MIM 125310) is an adult-onset disease that is characterized clinically by recurrent ischaemic strokes, migraine with aura, psychiatric symptoms, cognitive decline and dementia [1-3]. CADASIL is caused by mutations in the NOTCH3 gene, which codes for a transmembrane receptor molecule [4]. More than 130 mutations have been described, all of which give rise to an unpaired number of cysteines within the epidermal growth factor repeat domain of the receptor and lead to an abnormal secondary structure of the protein [5, 6]. Only three pedigrees with CADASIL are known to carry a NOTCH3 mutation that does not involve a cysteine residue [7-9].

Recent studies have suggested a mitochondrial aberration in CADASIL patients. We have described a patient with CADASIL caused by the R133C mutation in NOTCH3 and with concomitant myopathy caused by a 5650G>A mutation in the MTTA gene in mitochondrial DNA (mtDNA) [10]. Complex I activity of the mitochondrial respiratory chain has also been found to be decreased in muscle from patients with CADASIL [11]. Furthermore, histochemical studies have revealed ragged-red fibres with decreased cytochrome c oxidase (COX) staining in one patient [11] and ultrastructural studies have revealed core-like lesions and mitochondrial abnormalities in three others [12]. No clinical symptoms of myopathy were detected in these patients, however. Finally, subsarcolemmal aggregation of muscle mitochondria has been found in a CADASIL patient, who showed increased parenchymal brain lactate in magnetic resonance spectroscopy [5].

mtDNA is a maternally inherited 16.6-kbp haploid genome that codes for 13 subunits of the mitochondrial respiratory chain and also the 12S and 16S rRNAs and 22 tRNAs required for intramitochondrial protein synthesis [13]. Transcription and replication of mtDNA rely entirely on proteins encoded by nuclear genes, and indeed, recent proteomic studies have shown that about 25% of all mitochondrial proteins are involved in expressing or maintaining mtDNA [14, 15]. Based on our previous finding of a patient with concomitant mutations in the NOTCH3 and MTTA genes, we hypothesized that NOTCH3 may contribute to mtDNA maintenance and that mutations in NOTCH3 may increase the frequency of mutations in mtDNA. We therefore set out to determine mtDNA coding region sequences in 77 Finnish CADASIL patients and compare the observed sequence variation with that in controls.

Materials and methods

Patients and controls

We studied 77 Finnish CADASIL patients belonging to 20 unrelated pedigrees and 41 maternal lineages (Table 1) [16]. Pedigree F harboured the R182C mutation, while all the remaining pedigrees carried the R133C mutation. Seventeen pedigrees originated from the provinces of Central Ostrobothnia or Northern Savo. The controls consisted of 480 healthy anonymous blood donors recruited at Finnish Red Cross offices in these two provinces and in the neighbouring provinces of Northern Ostrobothnia and Kainuu [17]. The research protocol was approved by the ethical committees of Oulu University Hospital, Turku University Hospital and the Finnish Red Cross. The samples were studied after obtaining informed consent from the patients.

Molecular Methods

Total DNA was extracted from blood with a Nucleon BACC3 kit (Amersham Biosciences, UK) or QIAamp Blood Kit (Qiagen, Hilden, Germany). The coding region of mtDNA spanning between nucleotides (nts) 577-16023 was amplified using 63 pairs of oligonucleotide primers with a standard polymerase chain reaction (PCR) protocol [18]. The amplified fragments were then subjected to conformation-sensitive gel electrophoresis (CSGE) as described previously [18]. Fragments differing in mobility in CSGE from the homologous reference fragment were sequenced (ABI PRISM 377 Sequencer using DYEnamic ET Terminator Cycle Sequencing Kit, Amersham Biosciences, Amersham, UK). The primers used for sequencing were the same as those used in the amplification reactions. Sequence variants were identified by comparison with the revised Cambridge reference sequence [19].

The sequence data were imported into the previously constructed phylogenetic network based on a series of 192 Finns [17]. This enabled an easy identification of discrepancies suggesting sequencing errors and visualization of sequence differences between the cases and controls. All the sequence differences identified were then searched for in MITOMAP (a human mitochondrial genome database, http://www.mitomap.org, 2005), the mtDB-human mitochondrial genome database (http://www.genpat.uu.se/mtDB/), the mtSNP database (a database of human mitochondrial genome polymorphisms, http://www.giib.or.jp/mtsnp/index_e.shtml) and GenBank (http://www.ncbi.nlm.nih.gov/Genbank/index.html) and in previous reports [20, 21]. All variants that had not been reported previously were confirmed by sequencing one strand twice or both strands once, or by restriction fragment analysis to show AciI site loss caused by 3954C>T, HaeIII site loss caused by 8997C>A, Hsp92 site gain caused by 9338A>G, Hpy99I site loss caused by 9777G>A or BsmFI site gain caused by 12100A>G. All novel substitutions were also evaluated with respect to heteroplasmy by using results from CSGE, sequencing and restriction fragment analysis. The consequences of all the novel substitutions for the secondary structures of proteins or tRNA molecules were examined using information available in databases (compilation of mammalian mitochondrial tRNA genes, http://mamit-trna.u-strasbg.fr) [22-24].

Table 1 Patients with CADASIL and with a mutation in NOTCH3

Pedigreea

Patient

Relationship to propositus

Age

Haplogroup

A1

1 M, propositus

45 U

2 Sister

60 U

B2

3 F, propositus

Distant relative

M, propositus

Brother

Distant maternal relative

M, propositus

Mother

Brother

F, propositus

Sister

F, propositus

Daughter

Sister

Second-degree maternal relative

I10

F, propositus

J11

M, propositus

Sister

J12

Distant relative

K13

M, propositus

K14

Daughter

K15

Distant relative

La16

F, propositus

Brother

La17

Father

Lb18

M, propositus

Lb19

Son

Lc20

Ld21

F, propositus

Sister

Le22

M23

M, propositus

M24

Son

M25

Distant relative

Na26

M, propositus

Brother

Na27

Son

Na28

Second-degree relative

Nb29

F, propositus

Son

Nc30

F, propositus

Sister

Brother

Sister

Brother

Nc31

Second-degree relative

Table 1 (continued)

Nd32

F, propositus

Sister

Ne33

F, propositus

Brother

Distant maternal relative

O34

F, propositus

Sister

Brother

Daughter

Son

O35

Distant relative

O36

Distant relative

P37

F, propositus

Mother

Q38

M, propositus

Brother

R39

F, propositus

S40

M, propositus

T41

M, propositus

M male, F female

aLetters denote nuclear pedigrees; numbers denote matrilineal pedigrees

The discrepancies in mtDNA coding region sequences between maternal relatives were examined by sequencing or by restriction fragment analysis, which was used to analyse 3954C>T, causing AciI site loss, 12630G>A causing DpnII site gain, and 13928G>C, causing BccI site gain. In addition, allele status at nt 13827 was determined by allele-specific amplification of both wild-type and mutant DNA in the presence of locked nucleic acid (LNA) oligonucleotides (Proligo LCC, Paris, France). The population samples harbouring 12630G>A, 13827A>G or 13928G>C and samples harbouring the wild-type alleles were used as controls in all restriction fragment analyses and allele-specific amplifications. The presence of 3954C>T (patients 47 and 48 in pedigree Nb) and the presence of 12630G>A and 13928G>C (patients 15 and 16 in pedigree H) in blood DNA were also verified by cloning an amplified fragment spanning nts 3706-4055, nts 12500-12854 or nts 13696-14007, respectively, of mtDNA into a pCR2.1-TOPO vector (TOPO TA Cloning Kit, Invitrogen, Leek, Netherlands) [10]. Positive colonies (n=50) were cultured overnight and the variants were then detected as described above and by sequencing some of the clones using insert-specific forward primers. In addition, DNA from two healthy controls harboring the 12630G allele or the 12630A allele and from two other controls with the 13928G allele or the 13928C allele were cloned and analyzed in a similar fashion. It would have been informative to verify by cloning the allele status of positions 12630 and 13928 in the other available mother-offspring pair in family H (patient 18 and patient 19 or patient 20), but the lack of sample from patient 18 prevented these analyses.

The 77 Finnish CADASIL sequences have been deposited in the EMBL database (accession numbers AM260558–AM260634).

Comparison of nucleotide diversity between CADASIL pedigrees and controls