Brief Clinical Report

Identification of a Missense Mutation in aFriedreich’s Ataxia Patient: Implications for Diagnosisand Carrier Studies

Claire Bartolo,1 Jerry R. Mendell,2 and Thomas W. Prior1*1Department of Pathology, The Ohio State University, Columbus, Ohio2Department of Neurology, The Ohio State University, Columbus, Ohio

Approximately 95% of all Friedreich’s ataxia(FA) patients are homozygous for a largeGAA triplet-repeat expansion in the first in-tron of the Friedreich’s ataxia gene (FRDA).The remaining cases are expected to be com-pound heterozygous with a GAA expansionon one allele and a point mutation on theother. Generally, the clinical diagnostic pro-file in this group of patients is indistinguish-able from that in classic FA patients withhomozygous expansions. This study de-scribes a mildly affected patient whopresents with only one expanded allele bySouthern blot analysis. Point mutationscreening shows a single base change inFRDA exon 3 resulting in a nonconservativeamino acid replacement in the N-terminalportion of the frataxin protein. Extendedfamily studies show that two of the patient’ssibs are carriers of the expanded allele andone is a carrier of the missense mutation.This case study demonstrates the benefits ofimplementing a combined Southern blotand point mutation diagnostic protocol forcompound heterozygous patients. By identi-fying both mutations, this procedure con-firms the diagnosis of FA in patients with anatypical disease course and allows for morecomplete family studies. Am. J. Med. Gen.79:396–399, 1998. © 1998 Wiley-Liss, Inc.

KEY WORDS: Friedreich’s ataxia; com-pound heterozygote; triplet-repeat; missense mutation;carrier study

INTRODUCTION

Friedreich’s ataxia (FA) is the most common of theinherited ataxias and is transmitted in an autosomalrecessive manner. It occurs with a prevalence of 1 in50,000 and has an estimated carrier frequency of 1 in110 in European populations. FA is a neurodegenera-tive disorder characterized by a progressive gait andlimb ataxia with onset of symptoms before 25 years ofage [Harding, 1981]. Disease manifestations includeloss of vibration and position sense, areflexia, dysar-thria, and skeletal abnormalities [Geoffrey et al., 1976;Harding, 1981]. In addition, many patients exhibit hy-pertrophic cardiomyopathy that often leads to prema-ture death [Harding, 1993].

FA is caused by defects in the FRDA gene, whichencodes a 210-amino acid protein called frataxin. Ap-proximately 95% of affected individuals are homozy-gous for an unstable GAA trinucleotide expansion inthe first intron of the gene [Campuzano et al., 1996].Normal chromosomes contain 7–22 triplet repeats, andmutated chromosomes have more than 120 repeats[Durr et al., 1996]. Bidichandani et al. [1998] recentlydemonstrated that the large GAA intronic expansionsinhibit transcription elongation, possibly by formingstable intermolecular triplex structures, resulting inreduced FRDA transcript levels in FA patients. Theamount of transcript produced appears to be inverselycorrelated to the size of the expanded allele [Campu-zano et al., 1996; Bidichandani et al., 1997; Filla et al.,1996]. It is also directly related to an increase in clini-cal severity, especially disease onset age, rate of dis-ease progression, and the occurrence of cardiac abnor-malities [Durr et al., 1996; Filla et al., 1996; Monter-mini et al., 1997].

The FRDA gene is homozygously expanded in 95% ofFA cases, and the remaining patients are compoundheterozygotes with a GAA expansion on one allele anda point mutation on the other. The point mutationsidentified to date are private mutations, except for theI154F mutation, which has been described in morethan one patient [Campuzano et al., 1996; Filla et al.,1996]. Since these alterations are distributed randomlyacross the coding region, their detection requiresscreening the entire gene. However, due to the small

*Correspondence to: Thomas W. Prior, Ph.D., Department ofPathology, 121 Hamilton Hall, 1645 Neil Ave., Columbus, OH43210.

Received 29 April 1998; Accepted 30 June 1998

American Journal of Medical Genetics 79:396–399 (1998)

© 1998 Wiley-Liss, Inc.

size of the FRDA gene, when compared with larger andmore complex genes such as DMD and Factor VIII, thisbecomes a more manageable task. The elucidation ofboth mutant alleles not only confirms the clinical diag-nosis of FA in compound heterozygous patients, butalso allows for more accurate family studies. This isclearly illustrated by the case study we present in thisreport.

MATERIALS AND METHODSClinical Study

The patient, a 34-year-old woman, presented ini-tially at age 14 years with an unsteady gait and aninability to maintain balance. An electromyogram in-dicated a mild sensory neuropathy. Disease progres-sion was slow, and at 23 years motor examination in-dicated no atrophy or fasciculations but a somewhatdiminished muscle tone in all four limbs. Strength wasnormal except for mild bilateral ankle dorsiflexorweakness. Sensory examination demonstrated a mildsensory loss, diminished proprioception and vibrationat the toes, absence of muscle stretch reflexes, andflexor plantar responses.

Currently, the patient has a mild end gaze horizontalnystagmus but no nystagmus in vertical gaze or in theprimary position. Fundi and visual fields are normalbut a mild tremor and scanning quality of the voice hasbeen noted. On cerebellar testing, there is head andtrunk titubation while seated, and mildly impaired co-ordination is evident. The patient exhibits a wide-based gait with severe truncal instability, especiallywith feet placed together and eyes closed.

Presently the patient is 34 years old. She still exhib-its flexor plantar responses and continues to work dailyas a computer analyst and uses a scooter or a walkeronly when traveling long distances. She never devel-oped symptoms of cardiac disease and there is no evi-dence of congestive heart failure. The remaining rela-tives are normal.

Southern Blotting

Genomic DNA was isolated from whole blood leuko-cytes and 15 mg were digested for 5 hours at 65°C withBsiHKAI enzyme (New England Biolabs). The digestedDNA was size fractionated by 0.8% agarose gel electro-phoresis. The gels were blotted onto nylon membranes,as described by Southern, and hybridized with a 32P-radiolabeled 463-bp genomic fragment containing exon1 [Durr et al., 1996]. Blots were initially washed in 2×saline sodium citrate (SSC) and 0.5% sodium dodecylsulfate (SDS) at 42°C for 30 minutes, followed by asecond 30 minute, 42°C wash in 1 × SSC and 0.5% SDS.Finally, blots were washed twice at 60°C in a solutionof 0.5 × SSC and 0.4% SDS. Autoradiographs with in-tensifying screens were processed at −70°C for 1–5days. The BsiHKAI enzyme produces a 2.4 kb targetfragment.

Point Mutation Analysis

The individual exons of the FRDA gene werescreened to scan for point mutations in the coding re-

gion of this gene. Amplification was accomplished inthe presence of 100 ng of each oligonucleotide primer[Campuzano et al., 1996] and 2.5 U of Perkin-ElmerTaq polymerase in a final volume of 50 mL of the fol-lowing solution: 0.5 mM of each dNTP, 2 mM MgCl2, 67mM Tris (pH 8.8), 6.7 mM EDTA, and 10 mM b-mercaptoethanol. The amplification reactions con-sisted of 30 cycles of 55°C annealing temperature for 2min, 68°C extension temperature for 3 min, and 95°Cdenaturation temperature for 1 min in a thermocycler(Ericomp). After polymerase chain reaction (PCR)product confirmation on a 2.5% agarose gel, thesamples were heated to 95°C for 5 min, followed by a 30min incubation at 37°C to allow heteroduplex forma-tion. 15 mL of this product was mixed with 3 mL of 6×gel loading buffer (FMC) and electrophoresed on a 50-cm Hydrolink-MDE™ gel (FMC) at 13,000 VH. The gelconsists of 1× MDE™ gel solution (FMC) in 0.6× TBEand 15% urea. Products were visualized by ethidiumbromide staining.

Sequence analysis of the heteroduplex positive prod-uct was performed using the double-stranded DNACycle Sequencing System (Gibco BRL). Primers wereend-labeled with g[32P]-ATP. Product from the se-quencing reaction was analyzed using a 5% denaturingpolyacrylamide gel.

RESULTSMutation Analysis

Southern blot analysis (Fig. 1B) showed that theproposita (II-1) had one normal-sized allele and one ex-panded allele, a result also confirmed by long-PCRanalysis (data not shown). The expanded repeat lengthwas determined by comparison with a 1-kb DNA ladder(Gibco BRL) and was estimated at 840 ± 20 repeats.Since the clinical presentation of this patient was con-sistent with FA, a point mutation analysis was under-taken to determine whether the nonexpanded alleleharbored a point mutation. The entire frataxin codingsequence was screened by heteroduplex analysis andan aberrantly migrating heteroduplex band was ob-served in the exon 3 PCR product (Fig. 1A). Dideoxy-nucleotide sequencing demonstrated a T- to C- transi-tion at nucleotide 842 of the frataxin gene resulting inthe replacement of the TTA codon for leucine by theTCA codon for serine (L106S) in the N-terminal helicaldomain of the frataxin protein (Fig. 2). This is a non-conservative amino acid replacement, substituting apolar amino acid with lower helical forming tendencies.The Leu106 residue lies within a region that is con-served throughout evolution occurring in the Mus mus-culus frataxin protein (Genbank U95736) and also intwo homologous sequences from the nematode Cae-norhabditis elegans and the yeast Saccharomyces cer-evisia [Campuzano et al., 1996]. Isoleucine and valineoccur at this position in the CyaY genes in Escherichiacoli (Genbank P27838), Erwinia chrysanthemi (Gen-bank P40128), and Yersinia pestis (Genbank P46356).However, these amino acids have a-helical forming ten-dencies comparable to leucine. Moreover, the singlebase change was not observed on 100 normal chromo-somes screened and was absent in all patients with

Friedreich’s Ataxia 397

homozygous GAA trinucleotide repeats. No other mu-tations were detected in this individual after screeningthe entire gene for point mutations.

Family Studies

The remaining relatives elected to be tested for car-rier status, which was performed using the Southernblot procedure and heteroduplex analysis (Fig. 1). Af-fected individual (II-1) inherited the paternal ex-panded allele, whereas the missense mutation was in-herited from the mother. Two sibs (II-2 and II-3) wereheterozygous for the expanded allele, one (II-6) washeterozygous for the missense mutation, and the re-maining two (II-4 and II-5) were negative for both mu-tations. It is unlikely that the point mutation is a poly-morphism given the evolutionary conservation of theleucine residue and the lack of any other point muta-tion in the individual described here. In addition, thesole individual who inherited both types of mutation isclinically affected and no other unaffected relatives in-herited both mutations.

DISCUSSION

We describe a FA patient with a milder phenotypethan predicted for classic FA. This patient presentedwith a slow rate of disease progression, slightly dimin-ished lower limb muscle weakness and wasting, and noevidence of cardiac abnormalities or diabetes. Southernblot analysis demonstrated one expanded allele with arepeat size of approximately 840 repeats, and pointmutation screening showed an exon 3 T- to C- substi-tution, resulting in a nonconservative leucine to serineamino-acid replacement. This established the com-pound heterozygous state in the patient and excludedthe diagnosis of other ataxic syndromes.

The expansion mutation has been shown to suppresstranscription and result in reduced frataxin transcriptlevels [Campuzano et al., 1996]. On the other hand, theexon 3 missense mutation (L106S) is expected to act ata posttranscriptional level [Durr et al., 1996; Bidichan-dani et al., 1997] and allow the synthesis of a partiallyfunctional protein. A similar case involving an exon 4missense mutation, G130V in another mildly affectedFA patient was shown to have mRNA levels equivalentto clinically normal carriers [Bidichandani et al., 1997].It is likely that the same posttranscriptional mecha-nism operates in both patients and results in themilder phenotype observed. In contrast, a nonsensemutation, L106X, located at the same codon as our pa-tient (II-1) resulted in the total deficiency of frataxinand a more severe course of the disease [Campuzano etal., 1996].

The detection of missense mutations can contributeto a better understanding of the structural and func-tional domains of this protein. This is unlike expandedalleles and nonsense mutations that act by a loss-of-function mechanism and consequently can providelittle or no information on frataxin protein function.The L106S mutation identified in this study and theG130V mutation reported previously [Bidichandani etal., 1997] appear to be located outside the highly con-served motifs of the frataxin protein secondary struc-ture [Gibson et al., 1996]. This possibly increases thelikelihood of producing a partially functional protein

Fig. 2. Identification of the missense mutation in exon 3 of the gene.Sequencing of the sense strands is shown. Unaffected control shows thenormal sequence while the patient shows a T- to C- transition producing aL106S substitution.

Fig. 1. Mutation analysis in a FA family. A: Heteroduplex analysis ofFA exon 3. Individuals I-2, II-1, and II-6 demonstrate the heteroduplexband (*) caused by the missense mutation. B: Southern blot analysis dem-onstrates one expanded allele in individuals I-1, II-1, II-6.

398 Bartolo et al.

and the milder phenotype. On the other hand, theI154F mutation occurs in the most conserved frataxinprotein motif I [Gibson et al., 1996] and is associatedwith a severe disease expression [Campuzano et al.,1996].

The identification of both the expansion and pointmutations in the affected individual (II-1) also enabledaccurate carrier studies for all at-risk relatives. In thisstudy, Southern analysis demonstrated that II-4, II-5,and II-6 did not carry the expansion (Fig. 1B); however,their carrier risk was still 50% due to the possibility ofinheriting the unknown mutation. Point mutationanalysis provided a definitive diagnosis by establishingthat II-6 is a carrier of the missense mutation and II-4and II-5 are noncarriers (Fig. 1A).

This study demonstrates the clinical impact of pointmutation analysis in compound heterozygotes. Weshow that a combined Southern blot and point muta-tion diagnostic protocol can confirm the diagnosis of FAin compound heterozygous patients with an atypicaldisease course. It also provides for more accurate car-rier testing of all at-risk relatives. Since FRDA is asmall gene, point mutation screening is not labor-intensive or costly. We therefore suggest that it shouldbe considered for patients who have a single expandedallele and an unusual disease presentation.

ACKNOWLEDGMENTS

We are indebted to the family members for their co-operation and participation in this study. We also ac-knowledge the photography assistance of ArthurWeeks.

REFERENCESBidichandani SI, Ashizawa T, Patel PI (1998): The GAA triplet-repeat

expansion in Friedreich ataxia interferes with transcription and maybe associated with an unusual DNA structure. Am J Hum Genet 62:111–121.

Bidichandani SI, Ashizawa T, Patel PI (1997): Atypical Friedreich ataxiacaused by homozygosity for a novel missense mutation and the GAAtriplet-repeat expansion. Am J Hum Genet 60:1251–1256.

Campuzano V, Montermini L, Molte MD, Pianese L, Cossee M, CavalcantiF, Monros E, Rodius F, Duclos F, Monticelli A, Zara F, Canizares J,Koutnikova H, Bidichandani SI, Gellera C, Brice A, Trouillas P, DeMichele G, Filla A, De Frutos R, Palau F, Patel PI, Di Donato S, MandelJL, Cocozza S, Koenig M, Pandolfo M (1996): Friedreich’s ataxia: Au-tosomal recessive disease caused by an intronic GAA triplet repeatexpansion. Science 271:1423–1427.

Durr A, Cossee M, Agid Y, Campuzano V, Mignard C, Penet C, Mandel JL,Brice A, Koenig M (1996): Clinical and genetic abnormalities in pa-tients with Friedreich’s ataxia. N Engl J Med 335:1169–1175.

Filla A, De Michele G, Cavalcanti F, Pianese L, Monticelli A, CampanellaG, Cocozza S (1996): The relationship between trinucleotide (GAA) re-peat length and clinical features in Friedreich ataxia. Am J Hum Genet59:554–560.

Geoffroy G, Barbeau A, Breton G, Lemieux B, Aube M, Leger C, BouchardJB (1976): Clinical description and roentgenologic evaluation of pa-tients with Friedreich’s ataxia. Can J Neurol Sci 3:279–286.

Gibson TJ, Koonin EV, Musco G, Pastore A, Bork P (1996): Friedreich’sataxia protein: Phylogenetic evidence for mitochondrial dysfunction.TINS 19:465–468.

Harding AE (1981): Friedreich’s ataxia: A clinical and genetic study of 90families with an analysis of early diagnostic criteria and intrafamilialclustering of clinical features. Brain 104:589–620.

Harding AE (1993): Clinical features and classification of inherited atax-ias. Adv Neurol 61:1–14.

Montermini L, Richter A, Morgan K, Justice CM, Julien D, Castellotti B,Mercier J, Poirier J, Capozzoli F, Bouchard JP, Lemieux B, Mathieu J,Vanasse M, Seni MH, Graham G, Andermann F, Andermann E, Mel-ancon SB, Keats BJ, Di Donato S, Pandolfo M (1997): Phenotypic vari-ability in Friedreich ataxia: Role of the associated GAA triplet repeatexpansion. Ann Neurol 41:675–682.

Friedreich’s Ataxia 399