Udine Special Section

Genetics in restless legs syndrome

Luigi Ferini-Strambi*, Maria Teresa Bonati, Alessandro Oldani,Paolo Aridon, Marco Zucconi, Giorgio Casari

Sleep Disorders Center, University Vita-Salute San Raffaele and Human Molecular Genetics Unit,

Dibit-San Raffaele Hospital, Via Stamina d’Ancona 20, 20122 Milano, Italy

Received 1 November 2002; accepted 15 October 2003

Abstract

Several studies on Restless legs syndrome (RLS) have suggested a substantial genetic contribution in the etiology of this sleep disorder.

Clinical surveys of idiopathic RLS patients have shown that up to 60% report a positive family history. Investigations of single families with

RLS have suggested an autosomal dominant mode of inheritance with variable expressivity, and some families show possible anticipation. At

present, only one twin study is available, showing a high concordance rate (83.3%) between identical twins.

Despite several reports suggesting a genetic contribution to the etiology of idiopathic RLS, few molecular genetic studies have been

carried out attempting to identify genes that can predispose to this disorder. In particular, genes encoding for the GABA A receptor subunits,

the gene for the alpha1 subunit of the glycine receptor, and genes involved in dopaminergic transmission and metabolism have been

analyzed, but no significant findings have been reported.

Genome-wide studies have been conducted to map genes that play a role in vulnerability to RLS. In a single French–Canadian family

significant linkage was established on chromosome 12q. The susceptibility locus on chromosome 12q was not confirmed in two South

Tyrolean families, or in our two Italian families. However, the efforts toward the identification of RLS genes must continue in order to obtain

a better characterization of the syndrome and to identify new therapeutic strategies.

q 2004 Elsevier B.V. All rights reserved.

Keywords: Genetics; Restless legs syndrome; Genome-wide

1. Familial aggregation

Several studies on Restless legs syndrome (RLS) have

suggested a substantial genetic contribution in the aetiology

of this sleep disorder. Familial aggregation has been

frequently reported since Ekbom’s original clinical

description in 1944 [1]. Ekbom estimated the frequency of

hereditary RLS compared to sporadic cases as ‘one-third’.

Some clinical surveys of idiopathic RLS patients have since

shown that up to 60% report a positive family history [2–4].

Recently, Winkelman et al. performed a study to assess the

clinical characteristics of hereditary RLS patients in

comparison to those of non-hereditary RLS [5]. The authors

analysed the clinical data of 300 RLS patients diagnosed

according to the criteria of the International RLS

Study Group [6] using a standard questionnaire covering

demographic data, family history, clinical symptoms,

subjective sleep disturbances and course of the disease.

Family history was rated as definitely positive when at least

one first-degree relative was directly examined and

classified with RLS: 42.3% of the patients with idiopathic

RLS and 11.7% of those with secondary RLS due to uremia

had definite, positive hereditary RLS. Hereditary and non-

hereditary RLS patients had similar clinical characteristics,

except for two aspects: (a) patients with the hereditary form

were significantly younger at the age of onset (35.45 vs.

47.17, P , 0:05Þ; (b) women with hereditary RLS more

frequently experienced a worsening of symptoms during

pregnancy (19.1% vs. 2.6%, P , 0:05). Although it may be

possible that the onset of the disorder is recognized earlier in

subjects with hereditary RLS because of the increased

awareness of symptoms, the finding of Winkelman et al. is

in agreement with other reports [2–4,7]. Labuda [8] found a

risk of 19.9% for first-degree relatives among RLS patients

compared to 3.5% for first-degree relatives of control

1389-9457/$ - see front matter q 2004 Elsevier B.V. All rights reserved.

doi:10.1016/j.sleep.2004.01.005

Sleep Medicine 5 (2004) 301–304

www.elsevier.com/locate/sleep

* Corresponding author. Tel.: þ39-02-2643-3363; fax: þ39-02-2643-

3394.

E-mail address: ferinistrambi.luigi@hsr.it (L. Ferini-Strambi).

subjects, indicating an estimated relative risk in first-degree

relatives (lambda) of 5.7.

2. Pattern of inheritance, penetrance, anticipation

Investigations of single families with RLS have

suggested an autosomal dominant mode of inheritance

with variable expressivity [9–12]. Godbout et al. [13],

evaluating 209 cases from six families with RLS, found

44.5% to have familial RLS symptoms spanning three to

five generations. The smallest number of affected members

in any family was 6 and the largest number was 44.

Lazzarini et al. [12] examined five pedigrees with 81

affected members. One-factor analysis of variance of ages

of onset between generations and segregation

ratios calculated for each generation showed an autoso-

mal-dominant mode of inheritance and a male:female ratio

of 1:1.4.

A recent study of a series of consecutive, unselected RLS

patients provided strong evidence for a difference in the

mode of inheritance between families with early- versus

late-age onset; 238 patients and 537 first-degree relatives

were classified as RLS-affected or non-affected. Assess-

ments were based on direct, personal standardized diag-

nostic interviews. Complex segregation analysis stratified

the families according to the mean age at onset of the

disorder within a family, resulting in two groups. In Group

A (mean age at onset up to 30 years of age), segregation

analysis strongly favored a single, autosomal dominant gene

acting with a multifactorial component, but in Group B

(older than 30 years) there was no evidence for a major gene

effect. The authors concluded that, “An age at onset-related

differentiation of probands provides a new approach for

future linkage studies that may shed more light on the

genetics and the pathophysiology of this disorder” [14].

In 1996, Trenkwalder et al. [11], in a large German RLS

pedigree, confirmed the variable expressivity first described

by Walters et al. [10]; in the German pedigree an age-

dependent penetrance of as much as 56% up to age 40, with

full penetrance by age 60, was found. In one study that

excluded the secondary causes of RLS, the presence of a

family history increased from 64 to 92% when all first-

degree family members were directly interviewed [2],

suggesting that RLS is highly penetrant. Moreover,

Lazzarini et al. [12], who investigated 5 pedigrees, observed

full penetrance in 4 pedigrees and 86% overall penetrance in

one family.

It is known that monozygotic twin studies provide

significant information regarding both genetic and environ-

mental factors that may affect disease phenotype. Ondo

et al., in a study of 12 monozygotic twin pairs, found that 10

of the 12 pairs were concordant for RLS symptoms. Despite

the high concordance rate (83%), the RLS severity and age

at onset often varied between twins. However, this study

suggests that a significant portion of the familial aggregation

of RLS may be due to genetic factors [15].

Concerning the anticipation phenomenon, Trenkwalder

et al. [11] found that the mean age of RLS onset fell from the

second to the fourth generation in a large German family.

More recently, Lazzarini et al. [12] found anticipation in

only 2 of the 5 analyzed pedigrees; the authors conclude that

“The variation in anticipation between our pedigrees and

that of Trenkwalder et al. could suggest heterogeneity at the

molecular level.”

3. Molecular genetic studies

Despite several reports suggesting a genetic contribution

to the etiology of idiopathic RLS, few molecular genetic

studies have been carried out to identify genes that

predispose to RLS [16–18]. In particular, genes encoding

for the GABA A receptor subunits (alpha1, 2 and 4–6;

beta1–3, gamma 1–3) and the gene for the alpha1 subunit

of the glycine receptor have been analyzed, but no

significant findings have been reported.

Several studies have suggested the involvement of the

dopaminergic (DA) system in the etiology of RLS.

Functional imaging studies with SPECT and PET tech-

niques, as well as the beneficial effects of various

dopaminergic agents in the treatment of RLS, support the

hypothesis that the central DA system, particularly the

striatonigral system, is implicated in the disease [19]. A

recent study examined eight relevant candidate genes

involved in dopaminergic transmission and metabolism:

DA-receptors D1 to D5, dopamine transporter, tyrosine

hydroxylase and dopamine beta-hydroxylase. The study was

performed on 92 RLS patients and 182 controls matched for

ethnic background; no significant difference was found in

the genotypic or allelic distribution between the two groups.

The authors concluded that, “The importance of DA in the

pathogenesis of RLS cannot be ruled out as other loci

implicated in the dopaminergic system have not yet been

examined. Moreover, the possibility remains that other

functional polymorphisms that are not in linkage disequili-

brium with those investigated might affect the occurrence or

the outcome of the syndrome.” [18].

4. Genome-wide studies

Genome-wide studies have been conducted to map genes

that play a role in the vulnerability to RLS.

In the study by Desautels et al. [20], DNA from 25

members of a single French–Canadian family was sub-

jected to a genome-wide scan of markers for genetic linkage

and haplotype analyses. Genotyping was performed using

380 genetic markers at approximately 10 centi-Morgan

intervals across the human genome. Logarithm of odds

(LOD) scores were maximized over different modes of

L. Ferini-Strambi et al. / Sleep Medicine 5 (2004) 301–304302

inheritance with the assumption that the mode of RLS

inheritance is unknown. Significant linkage was established

on chromosome 12q, for a series of adjacent microsatellite

markers with a maximum two-point LOD score of 3.42

(autosomal recessive mode of inheritance). Haplotype

analysis refined the genetic interval, positioning the RLS-

predisposing gene in a 14.71-cM region between D12S1044

and D12S78. Several candidate genes have been mapped

within the region of interest. Among these is the gene

encoding the tridecapeptide neurotensin (NTS) that is

reported to act as a neuromodulator of DA transmission

[21]. Moreover, autoradiographic analyses showed a

dense localization of NTS receptors on DA-containing

neurons [22].

The susceptibility locus for RLS on chromosome 12q

was not confirmed by Kock et al. [23] in a study of two large

South Tyrolean families. Genomic DNA was isolated from

51 family members, and for each family both a dominant

model and a recessive model (adopting the model

parameters used by Desautels et al. [20]) were considered.

Moreover, these authors questioned certain parameters used

in the Desautels recessive model, which requires, for

example, a disease-allele frequency on the order of a

common polymorphism and further specifies a genotype-

specific penetrance value of 0.80 for f0 (representing the

probability that homozygous normal individuals are

affected, or, more simply, a phenocopy rate of 80% that

far exceeds the population prevalence of RLS (5–10%)).

Brooks has recently characterized as ‘unexpected’ the

suggestion by Desautels et al. of an autosomal recessive and

pseudodominant pattern of inheritance in RLS [24].

Pseudodominant inheritance requires a high prevalence of

subjects in the population who carry a defective gene that

acts in a recessive manner. Desautels et al. assumed that

25% of their population carry the defective gene; in such a

population, a patient would have a 1 in 4 chance of marrying

a carrier, and each of their children a 50% chance of

inheriting two defective genes (the same probability as of

Fig. 1. Pedigrees of two Italian RLS families.

Fig. 2. Parametric multipoint exclusions of RLS locus on 12q in two Italian RLS families.

L. Ferini-Strambi et al. / Sleep Medicine 5 (2004) 301–304 303

inheriting a dominant trait). Brooks concluded that, “These

conditions likely restrict the extent to which the finding of

this genetic linkage will be applicable to other popula-

tions…We have witnessed the first publication of RLS gene

mapping; more are sure to come.” On the other hand, the

genetic model recently provided by Winkelman et al. [14],

clearly consistent with dominant inheritance, should be

taken into account in the next linkage studies of RLS.

We have recently investigated two Northern Italian

families affected by idiopathic RLS. The detailed family

pedigrees (family A and B) are shown in Fig. 1. The mean

age-at-onset was 38 ^ 6 years for family A and 34 ^ 8 for

family B. Exclusion mapping of chromosome 12q for

families A and B was performed with an affected-only

approach, considering an autosomal dominant inheritance.

Multipoint LOD scores were calculated by GENEHUNTER

2.1 [25]. The parametric multipoint exclusions of RLS locus

on 12q are shown in Fig. 2.

Our data support the hypothesis of wide genetic

heterogeneity for RLS. However, the efforts toward the

identification of RLS genes must continue in order to obtain

a better clinical characterization of the syndrome, to suggest

the development of new diagnostic tests, and to identify new

therapeutic approaches.

Note added in proof

During the revision process a new RLS locus has been

mapped on chromosome 14q13–21 (Brain 126: 1485–92,

2003).

References

[1] Ekbom KA. Astenia crurum paraesthetica (irritable legs). Acta Med

Scand 1944;118:197.

[2] Ondo W, Jankovic J. Restless legs syndrome: Clinicoetiologic

correlates. Neurology 1996;47:1435–41.

[3] Walters A, Hickey K, Maltzman J, Joseph D, Hening W, Wilson V,

Chokroverty S. Aquestionnaire study of 138 patients with restless legs

syndrome: the night-walkers survey. Neurology 1996;46:92–5.

[4] Montplaisir J, Boucher S, Poirier G, Lavigne G, Lapierre O,

Lesperance P. Clinical, polysomnographic, and genetic characteristics

of restless legs syndrome: a study of 133 patients diagnosed with new

standard criteria. Mov Disord 1997;12:61–5.

[5] Winkelman J, Wetter TC, Collado-Seidel V, Gasser T, Dichgans M,

Yassouridis A, Trenkwalder C. Clinical characteristics and frequency

of the hereditary restless legs syndrome in a population of 300

patients. Sleep 2000;23:597–602.

[6] Walters AS. The International Restless Legs Syndrome Study Group.

Toward a better definition of the restless legs syndrome. Mov Disord

1995;10:634–42.

[7] Allen RP, Earley CJ. Definig the phenotype of the restless legs

syndrome (RLS) using age-of-symptom onset. Sleep Med 2000;1:

1–19.

[8] Labuda MC. Epidemiology. Family patterns and genetic of restless

legs syndrome. Washington, DC: RLS Foundation Conference; 1997.

[9] Montplaisir J, Godbout R, Boghen D, et al. Familial restless legs with

periodic movements in sleep: electrophysiologic, biochemical, and

pharmacological study. Neurology 1985;35:130–4.

[10] Walters AS, Picchietti D, Hening W, Lazzarini A. Variable

expressivity in familial restless legs syndrome. Arch Neurol 1990;

47:1219–20.

[11] Trenkwalder C, Collado-Seidel V, Gasser T, Oertel WH. Clinical

symptoms and possible anticipation in a large kindred of familial

restless legs syndrome. Mov Disord 1996;11:389–94.

[12] Lazzarini A, Walters AS, Hickey K, Coccagna G, Lugaresi E,

Ehrenberg BL, Picchietti DL, Brin M, Stenroos ES, Verrico T,

Johnson WG. Studies of penetrance and anticipation in five

autosomal-dominant restless legs syndrome pedigrees. Mov Disord

1999;46:92–5.

[13] Godbout R, Montplaisir J, Poirier G. Epidemiological data in familial

restless legs syndrome. Sleep Res 1987;16:338.

[14] Winkelmann J, Muller-Myhsok B, Wittchen HU, Hock B, Prager M,

Pfister H, Strhole A, Eisensehr I, Dichgans M, Gasser T, Trenkwalder

C. Complex segregation analysis of restless legs syndrome provides

evidence for an autosomal dominant mode of inheritance in early age

at onset families. Ann Neurol 2002;52:297–302.

[15] Ondo WG, Dat Vuong K, Wang Q. Restless legs syndrome in

monozygotic twins: clinical correlates. Neurology 2000;55:1404–7.

[16] Dichgans M, Walther E, Collado-Seigel V, Trenkwalder C, Muller-

Myhsok B, Oertel W, Gasser T. Autosomal dominant restless legs

syndrome: genetics model and evaluation of 22 candidate genes. Mov

Disord 1996;11:87.

[17] Zai C, Wigg KG, Barr CL. Genetics and sleep disorders. Semin Clin

Neuropsychiatry 2000;5:33–43.

[18] Desautels A, Turecki G, Montplaisir J, Ftouhi-Paquin N, Michaud M,

Chouinard VA, Rouleau GA. Dopaminergic neurotransmission and

restless legs syndrome: a genetic association analysis. Neurology

2001;57:1304–6.

[19] Chaudhuri KR, Appiah-Kubi LS, Trenkwalder C. Restless legs

syndrome (Editorial). J Neurol Neurosurg Psychiatry 2001;71:143–6.

[20] Desautels A, Turecki G, Montplaisir J, Sequeira A, Verner A,

Rouleau GA. Identification of a major susceptibility locus for

restless legs syndrome on chromosome 12q. Am J Hum Genet

2001;69:1266–70.

[21] Ervin GN, Birkemo LS, Nemeroff CB, Prange AJ. Neurotensin blocks

certain amphetamine-induced behaviours. Nature 1981;291:73–6.

[22] Quirion R, Chiueh CC, Everist HD, Pert A. Comparative localization

of neurotensin receptors on nigrostriatal and mesolimbic dopamin-

ergic terminals. Brain Res 1985;327:85–389.

[23] Kock N, Culjkovic B, Maniak S, Schilling K, Muller B, Zuhlke C,

Ozelius L, Klein C, Pramstaller P, Kramer PL. Mode of inheritance

and susceptibility locus for restless legs syndrome, on chromosome

12q. Am J Hum Genet 2002;71:205–8.

[24] Brooks DG. The first genetic locus mapped for RLS is unlikely to be

the last. Article reviewed: ‘Identification of a major susceptibility

locus for restless legs syndrome on chromosome 12q’. Sleep Med

2002;3:285–6.

[25] Kruglyak L, Lander ES. Faster multipoint linkage analysis using

Fourier transforms. J Comput Biol 1998;5:1–7.

L. Ferini-Strambi et al. / Sleep Medicine 5 (2004) 301–304304