Pseudodominance in Friedreich Ataxia—Impact of High Prevalence of Carriers and Intrafamilial Clinical Variation

ABSTRACT

Background

Friedreich ataxia (FA) is the most common form of autosomal recessive (AR) ataxia. It is a rare disease, but carriers are frequent (1/100). Pseudodominance in FA has seldomly been reported; it may pose additional challenges for diagnosis.

Cases

A family with two consecutive generations affected by FA is described. The proband and two younger siblings had typical FA, characterized by infantile‐onset ataxia, hyporeflexia, Babinski sign, cardiomyopathy, and loss of ambulation in the second decade of life. Another female sibling had delayed‐onset (>25 years old), with mild cerebellar and sensitive ataxia since her mid‐30s. Their father presented very late‐onset FA (>40 years old), with sensitive axonal neuropathy. All five patients had biallelic (GAA)_n expansion in FXN. The first three had larger expansions (>800 repeats), while the latter two had one shorter expanded allele (~90 repeats).

Literature Review

Pseudodominant inheritance has been described in 13 neurological disorders. Seven are movement disorders, of which three were associated with high frequency of carriers (FA, Wilson's disease and PRKN‐related parkinsonism).

Conclusions

Clinicians should be aware of the possibility of pseudodominance when facing an apparent autosomal dominant pedigree, particularly in disorders with high frequency of carriers and variable expression. Otherwise, genetic diagnoses may be delayed.

Friedreich ataxia (FA) (OMIM# 229300) is the commonest autosomal recessive (AR) ataxia in Caucasians, with an estimated prevalence of 0.5–1:100,000. ¹ , ² Most frequently, it is caused by a biallelic intronic (GAA)_n expansion in FXN (OMIM × 606829), resulting in decreased expression of frataxin, a protein involved in mitochondrial iron metabolism. ¹ Typical FA starts before age 25, and involves the peripheral nervous system, posterior columns, cerebellum and pyramidal tracts. It is characterized by dysarthria, square wave jerks, muscle weakness, impaired proprioception and vibration, abolished deep tendon reflexes (DTR), Babinski sign and progressive gait and limb ataxia. ¹ Non‐neurological features include skeletal anomalies (scoliosis, pes cavus), neurosensorial deafness, hypertrophic cardiomyopathy, optic atrophy and diabetes. ¹ Atypical FA is rare (about 15% of patients) and comprises late‐onset FA (LOFA) and very LOFA (vLOFA), when starting after 25 and 40 years of age, respectively. These atypical forms are characterized by milder symptoms, slower progression and paucity of extra‐neurological involvement. ³ FA is highly challenging at a genetic level as, (1) in a minority of patients it may be due to a compound heterozygous GAA repeat expansion with a point mutation; ¹ (2) the range of pathogenic expanded alleles is wide (66–1700 GAAs), with expansions ranging from 44 to 66 being unstable and conferring high risk for disease; ⁴ and (3) the frequency of heterozygotes in the general population is estimated at 1/100. ¹ , ² , ⁵ Two main clinical consequences emerge from this genotypic variability: (1) inter and intra‐familial phenotypic heterogeneity, in that disease severity correlates partially and inversely with the size of expanded alleles (particularly the shorter one); ⁵ and (2) a high population frequency of carriers may lead to a pseudodominant transmission, challenging its diagnosis. While the clinical heterogeneity of FA is well known, descriptions of pseudodominance are scarce. ⁶ , ⁷ , ⁸ , ⁹ , ¹⁰ , ¹¹

Here, we describe a FA family, genetically confirmed, exhibiting phenotypic variability and pseudodominant inheritance, and review pseudodominance in neurological disorders.

Patients and Methods

A Portuguese family of five affected members with FA in two consecutive generations, has been followed at the movement disorders outpatient clinic for 30 years. Information was retrieved from medical records, according to a structured protocol including demographic, clinical data and complementary investigation (MRI, nerve conduction study, echocardiogram, ECG and genetic analysis, when available).

Family Study

Both parents were originally from the same small village (around 500 inhabitants), but there was no known consanguinity (Fig. 1). Three offspring had early‐onset and one late‐onset FA; the only healthy sibling was found to have been adopted. The fact that all four biological sibs were affected, unusual for an AR condition, prompted our formal examination of both parents. The clinical description and FXN genotype of the affected cases are as follows:

FIG. 1.

The family's pedigree shows a pseudo‐dominant pattern of transmission. Filled symbols represent affected individuals: black, classical FA; dark gray, LOFA; and light gray, vLOFA; half‐filled symbol indicates heterozygous carrier; symbol in bracket means adopted child; and “*” refers to individuals who have been observed and performed genetic analysis of FXN. Size of normal and expanded alleles is indicated, together with current age, age‐of‐onset (o.) and age at death (d.).

Case Series

Case 1 (Patient III‐1)

The proband, a male patient, homozygous for a ~ 850 (GAA)_n expansion, had a normal psychomotor development and was otherwise healthy until the age of 5 years, when he started experiencing walking difficulties. Neurologic examination (by age 9) showed absent DTR, bilateral Babinski sign, decreased vibration in the halluces and positive Romberg test. In the following years, he developed paraparesis, limb dysmetria and cerebellar dysarthria, being on a wheelchair by age 15 years. Scoliosis, pes cavus, hypertrophic cardiomyopathy and atrial fibrillation were also present. He died at the age of 23, shortly after a cerebellar embolic stroke.

Case 2 (Patient III‐2)

A 41‐year‐old male sib, homozygous for a ~ 923 (GAA)_n allele, was first observed when 8 years‐old. Although asymptomatic, neurologic examination showed absent DTR in lower limbs (LL) and positive Romberg test. Four years later, gait unsteadiness and slurred speech were noticed. These gradually progressed and, by age 20, he was wheelchair‐bound. Neurological examination at age 33 years included severe dysarthria, hypotonic‐areflexic tetraparesis, bilateral Babinski sign, cervical and hands dystonia, stocking‐and‐glove hypoesthesia, decreased vibration in the halluces and marked axial and appendicular ataxia, orthostatism being impossible (video 1, segment 1). He also had thoracic scoliosis and Raynaud syndrome. Audiogram and ophthalmologic evaluation ruled out neurosensorial hypoacusis and optic atrophy. Hypertrophic cardiomyopathy was known since pediatric age and, by mid‐30s, atrial fibrillation and diabetes were diagnosed.

Video 1.

Typical Friedreich ataxia. Segment 1. Patient III‐2 is in a wheelchair; we may hear the cerebellar dysarthria and observe head titubation and mild torticollis, distal atrophies of the hands, tetraparesis and upper limbs dysmetria. Segment 2. Patient III‐4, also in a wheelchair has mildly fragmented ocular pursuit with saccadic intrusions, tetraparesis and upper limbs dysmetria and dysdiadochokinesia. Lower limbs coordination could not be evaluated due to motor deficit. Pes cavus, Raynaud phenomenon and hand muscles atrophy are also observed.

Case 3 (Patient III‐4)

A 32‐year‐old sister, had a history similar to III‐2. Her genotype was ~840/~940 GAAs at the FXN locus. She was observed while asymptomatic, at 4 years of age: neurologic examination showed only absent DTR in LL. Gait ataxia occurred by age 15, and she was in a wheelchair 5 years later. Cervical/ brain MRI at that time, did not reveal spinal or cerebellar atrophy. On examination, at 27 years of age, she had moderate dysarthria, hypermetric saccades, hypotonia and muscle weakness, with distal atrophy of the hands, Raynaud syndrome and bilateral pes cavus; DTR were diminished in the upper limbs (UL) and absent in LL, vibration sense was reduced distally and Babinski sign was bilateral. She presented cervical and hands dystonia, limbs and truncal ataxia, being unable to stand or walk (video 1, segment 2). She had an intra‐auricular communication (IAC), surgically corrected in childhood, hypertrophic cardiomyopathy and atrial fibrillation. In her late twenties she suffered a right medial cerebral artery embolic stroke and is currently bedridden.

Case 4 (Patient III‐3)

Their 38‐year‐old sister had a normal examination at age four. Her genotype at FXN was ~90/~874. At age 33, she was still asymptomatic, but on neurological evaluation there was a mild cervical dystonia, mild left limbs dysmetria; decreased vibration sense and abolished DTR in LL. Like her sister, she had a congenital IAC, surgically corrected in childhood, with a normal echocardiogram in adulthood and an ECG showing abnormal right ventricular conduction. On MRI there was no cerebellum or cervical atrophy. She currently maintains mild left limbs dysmetria, with abolished LL DTR and diminished vibration distally (video 2).

Video 2.

Late‐onset Friedreich ataxia. Patient III‐3 exhibits mild torticollis to the left and laterocollis to the right, normal ocular movements, mild left limbs dysmetria on finger chasing and mild left dysdiadochokinesia. DTR are normal in the upper limbs and abolished on lower limbs. She has negative Romberg test and normal tandem gait.

Case 5 (Patient II‐1)

The 68‐year‐old father (~90/~973 GAAs), had history of alcohol abuse, but was healthy until age 45 years, when he first noticed pain and tingling in the LL, which worsened progressively. He was called for examination at age 65 years, and presented diminished DTR, stocking‐and‐glove hypoesthesia, decreased vibration sense in the halluces, mild left dysmetria on heel‐to‐knee and mildly broad‐based gait (video 3). Nerve conduction studies confirmed a sensitive axonal neuropathy. Myocardiopathy, neurosensorial hypoacusis and diabetes were excluded. Brain and cervical MRI displayed only mild cerebellar atrophy (involving mainly the superior vermis).

Video 3.

Very late‐onset Friedreich ataxia. Patient II‐1 has normal ocular movements and no upper limbs dysmetria. He has mild upper limbs bradykinesia, left dysmetria on heel–knee test and reduced lower limb DTR. Romberg test is negative, but gait is mildly broad‐based.

Case 6 (Individual II‐2)

The 63‐year‐old mother had a normal neurological examination, and was a carrier of one expanded allele at the FXN locus (8/~900 GAAs).

Literature Review

Pseudodominance may result from couples where one partner is homozygote or compound heterozygote and the other harbors (at least) one pathogenic variant in the same gene (and at least one partner is affected) for an AR disease. ¹² This may happen (1) with declared or “hidden” consanguinity, or in endogamic populations, often with successive consanguineous mattings (offspring should be homozygotes); or (2) just due to the high frequency of carriers in the population (offspring may be either homozygotes or compound heterozygotes). That should be distinguished from cases when an allele for an AR disease is expressed due to deletion of the other allele (a hemizygous state, the equivalent of a female affected by a X‐linked disease). Also, in some diseases, heterozygotes may occasionally show an intermediate (milder and sometimes different) phenotype from the homozygotes, ¹² a situation called manifesting heterozygotes.

We conducted a PubMed search, without time limit, using the terms: “pseudo‐dominance,” “pseudo‐dominant,” “pseudodominance” OR “pseudodominant”. All abstracts, written in English, were reviewed and those related to neurological conditions were selected: a total of 44 articles were identified and assessed; 20 were excluded: (1) when only the abstract was available (n = 2); (2) no description of family studies (n = 6); (3) families wrongfully classified as “pseudodominant” (ie, families with AR disorders and homozygotes in one generation and heterozygotes in the other generation) (n = 6); and (4) no genetic testing or insufficient clinical and genealogical information, making pseudodominance questionable (n = 6). Six articles were additionally included, after references screening (Fig. 2). Out of the 30 articles considered, 24 were single or small‐series case reports, five were large‐scale series and one was a population‐based study (supplementary table).

FIG. 2.

Flowchart of selected articles for analysis.

FA, Wilson's disease (WD) and PRKN‐related Parkinsonism were the diseases more often associated with pseudodominance: eight WD families, ¹³ , ¹⁴ , ¹⁵ six FA families, ⁶ , ⁷ , ⁸ , ⁹ , ¹⁰ , ¹¹ and six families with PRKN‐related Parkinsonism. ¹⁶ , ¹⁷ , ¹⁸ , ¹⁹ , ²⁰ Ten other neurological disorders with pseudodominant inheritance were identified, relying (with few exceptions) on individual family reports: spinal muscular atrophy (five); ²¹ , ²² , ²³ McArdle disease (three); ²⁴ , ²⁵ , ²⁶ ocular motor apraxia type 1 (one) and type 2 (two); ²⁷ , ²⁸ , ²⁹ pseudoxanthoma elasticum with cerebrovascular disease (two); ³⁰ , ³¹ and one family each with PINK1‐related Parkinsonism, ³² neuronal ceroid lipofuscinosis due to CTSF variants, ³³ Charcot–Marie‐Tooth type 2 due to GDAP1 variants, ³⁴ Charlevoix‐Saguenay ataxia, ³⁵ and Brown‐Vialetto‐Van Laere syndrome (due to SLC52A3 variants). ³⁶ In FA, WD and PRKN‐related Parkinsonism there were families in whom pseudodominance was related to a population high frequency of heterozygotes (1/90 in WD; 3.2–4.6% in PRKN‐related Parkinsonism), ⁷ , ¹⁰ , ¹⁴ , ³⁷ while in both FA and PRKN‐related Parkinsonism there were also reports of consanguinity. ⁸ , ¹¹ , ¹⁶ , ¹⁸ , ²⁰ Of note, analysis of pseudodominance related to PRKN may be complex, since single heterozygous variants alone are a risk factor for Parkinson disease. ³⁷ There are no specific studies to evaluate frequency of pseudodominance in families with FA, WD or PRKN‐related Parkinsonism, but it seems to be a rare event. As for the remaining neurological disorders, prevalence of carriers has rarely been determined. As they are rare (or very rare) in general population, pseudodominance occurred either due to consanguinity or chance homozygote‐heterozygote mating.

Discussion

Phenotypic Spectra and GAA Transmission

We report a very interesting family with FA, with no declared consanguinity, in which two consecutive generations were affected and the three clinical forms of FA identified: (1) typical FA (patients III‐1, III‐2 and III‐5), with infantile‐onset ataxia, hyporeflexia, Babinski sign, cardiomyopathy, and loss of ambulation in the second decade; (2) LOFA in patient III‐4, with mild cerebellar and sensitive ataxia; and (3) vLOFA in patient II‐1, who developed sensitive neuropathy in middle age, with no extra‐neurological involvement. As expected, patients with typical FA (and hence severe disease) had two longer expanded alleles (>800 repeats), while both atypical patients had one short expansion (~90 repeats), a genotype–phenotype correlation and intra‐familial heterogeneity in line with previous reports. ¹ , ²

The asymptomatic mother carried one expanded allele on FXN (~900) and the father one short (~90) and one larger expanded allele (~973 GAAs). There was unstable transmission of the GAA repeat to the offspring (Fig. 1), as the expanded alleles slightly increased or decreased in size, compared to their parents’. Contraction of the maternal expansion occurred on III‐1 and III‐3 (shorter allele most probably inherited from the father), whereas it expanded further on III‐2. The larger expansion from the father contracted upon transmission to III‐1 and III‐2. In III‐4, the paternal or maternal origin of each allele is more difficult to ascertain, though both a maternal and a paternal contraction seem likely. As in other triplet repeat expansion disorders, the expanded GAA in FXN has been described as a “dynamic mutation,” meaning that during parent‐offspring transmission there is instability (changes in size) of the expanded repeat. ⁵ In autosomal dominant trinucleotide‐repeat disorders, the expanded allele tends to increase (creating genetic anticipation). On the contrary, in FA further expansion or contraction of expanded alleles is difficult to predict in offspring. ⁵ , ⁶ Two factors have been identified to influence intergenerational variation of the repeat size in FA: (1) parental sex and size of the transmitted allele: paternal alleles tend to contract in offspring, whereas maternal transmission may either lead to further expansion or contraction (with a strong tendency for smaller expansions to expand further and for larger expansions to contract); ⁵ and (2) genetic status: size of the expanded allele in heterozygotes tends to remain stable throughout generations, which may account for the maintenance of FA expansions in the population. ⁵ Importantly, not only repeat size and intergenerational instability explains age‐of‐onset and severity in individual FA patients. Indeed, the role of epigenetic modifiers and (GAA)_n somatic mosaicism across different tissues (heart, pancreas, cerebellum, spinal cord) ³⁸ have been the focus of recent studies.

Pseudodominance

The pattern of pseudodominance in the described family, with two consecutive generations affected and intra‐familial heterogeneity, has seldomly been described. ⁶ , ⁷ , ⁸ , ⁹ , ¹⁰ , ¹¹ The clinical impact of such extremely rare pedigrees justifies the importance of this report: the father of this family was only diagnosed after all his offspring were found to have FA. In all seven pseudodominant FA families (including the present report) the affected parent had late or very‐late onset, contributing to reproduction and transmission of the disease to the next generation. ⁶ , ⁷ , ⁸ , ⁹ , ¹⁰ , ¹¹ Disease onset was quite variable in the second generations, ranging from 5 to 48 years, ⁶ , ⁷ , ⁸ , ⁹ , ¹⁰ , ¹¹ with two cases being still asymptomatic in their twenties. ⁹

The carrier rate of FA in Portugal is 1/106 (similar to other Caucasian populations), prevalence of the disease being 1.0 per 100,000. ² , ³⁹ The family here reported originated from a very small and geographically isolated village, in a region where prevalence of FA is the triple of the national mean. Still, we could not be certain whether the pseudodominant pattern was due to the high frequency of carriers or to “hidden” consanguinity. Haplotyping could have helped us to clarify this aspect, but was beyond the scope of the present work. Among the further six families described with FA in two generations, there was also no known consanguinity in three, ⁷ , ⁹ , ¹⁰ although two (like ours) were from the same village. ⁷ , ¹⁰ This highlights the need for carrier screening and genetic counseling programs for more common recessive diseases, as well as in isolated areas and regions of known higher prevalence.

With this report we wish to raise awareness about (1) atypical FA phenotypes and (2) families with consecutive affected generations in AR disorders (pseudodominance), when there is a high prevalence of carriers or in endogamic populations. Both may delay clinical and genetic diagnosis, and impact on early treatments and clinical management, including proper genetic counseling.

Author Roles

(1) Research project: A. Conception, B. Organization, C. Execution; (2) Data Analysis: A. Design, B. Execution, C. Review and Critique; (3) Manuscript Preparation: A. Writing of the first draft, B. Review and Critique.

M.J.M.: 1A, 1B, 1C, 2B, 3A.

J.O.: 1C, 2C, 3B.

M.S.: 1A, 2C, 3B.

A.F.B.: 2B, 3B.

A.S.: 2B, 3B.

J.S.: 1A, 2C, 3B.

J.B.: 1A, 2C, 3B.

J.D.: 1A, 1B, 1C, 2A, 2C, 3B.

Disclosures

Ethical Compliance Statement: This study was approved by the ethics committee of Centro Hospitalar Universitário do Porto and informed consent was obtained from family members. The authors confirm that they have read the Journal's position on issues involved in ethical publication and affirm that this work is consistent with those guidelines.

Funding Sources and Conflicts of Interest: No specific funding was received for this work. The authors declare that there are no conflicts of interest relevant to this work.

Financial Disclosures for the Previous 12 Months: The authors have nothing to declare.

Supporting information

Table 1. Literature review.