OUP user menu

Phenotypic and genotypic heterogeneity in hereditary motor neuronopathy type V
A clinical, electrophysiological and genetic study

Michaela Auer-Grumbach, Wolfgang N. Löscher, Klaus Wagner, Erwin Petek, Eva Körner, Hans Offenbacher, Hans-Peter Hartung
DOI: http://dx.doi.org/10.1093/brain/123.8.1612 1612-1623 First published online: 1 August 2000

Summary

We report on a large four-generation Austrian family with autosomal dominant distal hereditary motor neuronopathy type V (distal HMN V). Forty-seven at-risk family members, of whom 21 were definitely affected, underwent detailed clinical, electrophysiological and genetic studies. The age at onset was in the second decade of life in most affected individuals, but clinical presentation was rather variable. While the majority of patients were primarily disabled by progressive asymmetrical wasting of the thenar and the first dorsal interosseus muscles, others had marked foot deformity and gait disturbance with the occasional absence of hand involvement. Sensation sense was normal except for the reduced response to vibration. Many individuals showed brisk tendon reflexes and some elevated muscle tone in the lower limbs, but extensor plantar responses were rarely observed. Electrophysiological evaluation revealed normal or reduced motor nerve conduction velocities, normal or prolonged distal motor latencies, and low compound motor action potentials, depending on the degree of muscle wasting. Sensory nerve studies were usually within the normal range or slightly to moderately abnormal in older or severely affected persons. Electromyography showed high-amplitude motor unit potentials and reduced recruitment compatible with anterior horn cell degeneration. Central motor conduction times were prolonged in two-thirds of the patients. Molecular genetic studies excluded Charcot–Marie–Tooth 1A syndrome and proximal spinal muscular atrophy linked to chromosome 5q as well as the known gene loci for distal HMN II on chromosome 12q, HMN V on chromosome 7p and juvenile amyotrophic lateral sclerosis on chromosome 9q. The findings in this family thus provide detailed clinical and electrophysiological information on HMN V and demonstrate broad phenotypic variability in this disorder. Hallmark features are discussed that appear to be most reliable to differentiate this type of HMN V from other variants of hereditary neuropathies, and a set of diagnostic criteria is proposed. Furthermore, this is the first report of prolonged central motor conduction times in HMN V, which indicates additional involvement of the central motor pathways in this disease. Finally, molecular genetic studies demonstrate genetic heterogeneity, suggesting the existence of at least a second genetic subtype in HMN V.

  • hereditary motor neuronopathy
  • HMN V
  • distal spinal muscular atrophy
  • Charcot-Marie-Tooth
  • chromosome 7p
  • CMAP = compound motor action potential
  • CMCT = central motor conduction time
  • FDI = first dorsal interosseus muscle
  • HMN = hereditary motor neuronopathy
  • HMSN = hereditary motor and sensory neuropathy
  • NCV = nerve conduction velocity

Introduction

Charcot–Marie–Tooth syndrome (hereditary motor and sensory neuropathy, HMSN) is a genetically determined common disorder of the peripheral nervous system with an estimated incidence of 40/100 000 (Skre, 1974). Based on clinical, electrophysiological and morphological criteria, this disorder can be divided into three main subtypes: a demyelinating, an axonal and a spinal form. Charcot–Marie–Tooth type 1, the demyelinating form, and Charcot–Marie–Tooth type 2, the axonal variant, are characterized clinically by distal muscle weakness and wasting of the upper and lower limbs, foot deformity, gait disturbance and reduced or absent tendon reflexes (Harding and Thomas, 1980). A variable degree of sensory loss is usually present and the disease is occasionally accompanied by additional features (Kwon et al., 1995; Thomas et al., 1997; Auer-Grumbach et al., 1998). Until molecular genetic testing became available for individual Charcot–Marie–Tooth 1 cases, the two subtypes could be distinguished only by electrophysiological and histopathological methods (Harding and Thomas, 1980). The spinal form of Charcot–Marie–Tooth syndrome [also called distal spinal muscular atrophy; in the recent gene mapping literature the preferred term is distal hereditary motor neuronopathy (HMN)] is phenotypically different from classical Charcot–Marie–Tooth types 1 and 2 as there is no clinically apparent sensory loss and the sensory nerves are electrophysiologically and morphologically normal (Harding, 1993). It has been emphasized that distal HMN is caused by degeneration of the spinal motor neurons (Harding, 1993). Based on the wide spectrum of clinical variation and different modes of inheritance, distal HMN has been divided into seven subtypes (distal HMN I–VII) (Harding, 1993). When muscle weakness and wasting are confined predominantly to the hands, the disease is termed HMN type V. Both sporadic and familial cases of HMN V have been observed (Meadows and Marsden, 1969; McLeod and Prineas, 1971; O'Sullivan and McLeod, 1978). To date, however, only a few families have been described in the literature (Silver, 1966; Lander et al., 1976; van Gent et al., 1985; de Visser et al., 1988; Christodoulou et al., 1995; Gross et al., 1998).

While some of these kinships demonstrated symmetrical wasting of all small hand muscles, others presented with a striking, asymmetrical distribution, usually affecting the thenar and first dorsal interosseus (FDI) muscles predominantly. Clinical sensory loss was rarely observed and was only a minor feature when it was present, but foot deformity was a frequent sign. In addition, mild spastic paraplegia was noted in some families and was sometimes prominent enough for the disorder to be classified as a hereditary spastic paraplegia rather than an HMN (Silver, 1966; de Visser et al., 1988).

Recent molecular genetic studies have led to the definition of two gene loci in distal HMN: HMN II on chromosome 12q (Timmerman et al., 1996) and HMN V on chromosome 7p (Christodoulou et al., 1995). The latter was first reported in a large Bulgarian kinship in which affected individuals were afflicted with prominent wasting of the thenar and FDI muscles. Lower limb involvement was observed in 40% of cases. Brisk tendon reflexes and mild pyramidal tract involvement were found in only one branch of the family. Sensation was clinically and electrophysiologically normal except mildly reduced vibration. Interestingly, in 1996 Ionasescu and colleagues described a similar phenotype in a family from Iowa, and the gene locus also mapped to chromosome 7p (Ionasescu et al., 1996). In contrast to the Bulgarian family, in affected individuals of this family tendon reflexes were reduced and sensory loss was common. Consequently, the disorder was classified as an axonal Charcot–Marie–Tooth type 2 and was subcategorized genetically as Charcot–Marie–Tooth type 2D. Subsequently, a disease locus on chromosome 7p was confirmed in a large Mongolian Charcot–Marie–Tooth 2D kinship in which patients with and without sensory loss were observed within the same family. Genetic linkage studies in this family suggested that Charcot–Marie–Tooth 2D and HMN V are caused by mutations in a single gene (Sambuughin et al., 1998). A refined genetic analysis narrowed the critical region for Charcot–Marie–Tooth 2D/HMN V to ~1.5 cM (centiMorgan) and a BAC/PAC contig map has been constructed containing the whole of the region of interest (Ellsworth et al., 1999).

We report here a large Austrian family presenting with signs and symptoms of HMN V and provide detailed clinical and electrophysiological data in 21 affected and 26 at-risk individuals. Genetic linkage analysis did not confirm the known HMN V gene locus on chromosome 7p in our family, therefore suggesting genetic heterogeneity of this disorder, and we attempted to establish diagnostic criteria for different genetic subtypes. The large number of definitely affected persons in this family allowed us to demonstrate the broad spectrum of phenotypic variation of this disorder. We also investigated whether additional electrophysiological findings contribute to the diagnosis of the disease in subclinical cases and whether they aid in distinguishing different genetic subtypes.

Patients and methods

Family data and clinical assessment

A part of the pedigree of the family comprising 134 members in four generations is shown in Fig. 1. Several male-to-male transmissions suggest autosomal dominant inheritance. In total, 47 at-risk individuals were recruited after obtaining informed consent for clinical, electrophysiological and genetic studies, and a further eight spouses participated in the genetic linkage studies. A full neurological and detailed electrophysiological examination was performed on all at-risk family members (aged 12–68 years, mean 35 years) by one of us (M.A.-G.). Information on deceased family members was obtained independently from several older relatives, when available.

Fig. 1

Partial pedigree of family with distal HMN type V. Definitely affected family members are shown in black; at-risk individuals and spouses are marked as normal.

Motor strength was assessed using the standard MRC (Medical Research Council) scale (grades 0–5) and reflexes were quantified as absent = 0, hypoactive = +1, normal = +2, brisk = +3 or +4, following the NINDS (National Institute of Neurological Disorders and Stroke) scale. Touch sensation was tested with a monofilament and vibration sense was quantitated with a graduated Rydel–Seiffer tuning fork (grades 0–8). The ability to recognize written numbers in the distal parts of the upper and lower limbs was assessed in every individual.

As there was considerable phenotypic variation in the expression of the disease and the family was ascertained for genetic linkage studies, strict diagnostic criteria were used for the phenotypic classification of the patients and at-risk individuals. According to their medical history and physical and electrophysiological evaluation, at-risk individuals were diagnosed as being either definitely affected or probably affected. None was classified as not affected, because the penetrance of the disease is not known. A person was classified as definitely affected when all of the following criteria were met: (i) age at onset before the fourth decade of life; (ii) unilateral or bilateral weakness and wasting of thenar and/or FDI muscles and/or marked foot deformity; (iii) clinical absence of sensory loss except for impaired vibration sense; (iv) pathological motor nerve conduction speeds in at least two nerves but normal or only slightly abnormal sensory nerve conduction speeds. Individuals with polyneuropathy of known cause were excluded. At-risk individuals were defined as probably affected if they were normal on clinical and electrophysiological examination or had only mild foot deformity and/or brisk tendon reflexes and/or sweating disturbance or pathological sensory nerve conduction speeds or prolonged central motor conduction times.

Neurophysiology

Electromyography and nerve conduction velocity studies

Nerve conduction velocity (NCV) studies and EMG followed standard techniques (Aminoff, 1998) using the electromyograph MS60 (Medelec, Old Woking, UK) or a Myohandy portable EMG (Micromed Neurodata, Mogliano Veneto, Italy). Responses for motor nerve conduction studies were recorded from distal muscles using surface electrodes. Sensory nerve conduction studies were performed antidromically with the use of surface or ring electrodes. Multiple motor nerves, i.e. median, ulnar, peroneal and tibial nerves, were measured bilaterally in most individuals and sensory nerve conduction studies were performed on the median and sural nerve on at least one side. Semiquantitative EMG was performed with concentric needle electrodes on distal muscles of the upper and lower limbs.

Magnetic evoked potentials

Transcranial and nerve root magnetic stimuli were delivered through a Magstim 200 stimulator (Magstim, Whitland, Dyfed, UK). A standard round coil with an inner diameter of 9 cm was used. The motor cortex of the hand and foot area and the spinal nerve roots C8 and L5 were stimulated and magnetic evoked motor potentials were recorded from the abductor digiti minimi and the tibialis anterior muscles bilaterally using surface electrodes and a commercial EMG amplifier (Toennies–Jäger, Würzburg, Germany). Motor cortex stimulations were performed with underlying contraction and stimulator output was increased to yield the maximum response amplitude. At least two reproducible responses were recorded from each stimulation site. Cortical and spinal latencies were determined visually and the central motor conduction times were calculated.

Statistics

The distribution of electrophysiological data was analysed using the Shapiro–Wilks W-test and non–parametric statistics were used accordingly. As no side differences were found for nerve conduction studies (Wilcoxon matched pair test), data for both sides were pooled to compare definitely and probably affected groups. Differences between groups were analysed using the Mann–Whitney U-test.

Molecular genetics

DNA marker analysis

Peripheral blood samples were obtained from patients and relatives after they had given informed consent. DNA isolation from leucocytes was performed according to standard methods.

After determination of the DNA concentration, ~100 ng genomic DNA was used for PCR (polymerase chain reaction) to amplify short tandem repeats for the genotype analysis. One oligonucleotide primer was labelled with a fluorochrome (IRD700, IRD800; MWG–Biotech, Ebersberg, Germany) and the resulting amplification products were analysed using a DNA sequencer with a dual laser system (DNA sequencer 4000; LI–COR, Lincoln, Nebr., USA). For the exclusion of distal HMN II on chromosome 12, the DNA markers D12S1282, D12S1349, D12S2079, D12S340, D12S378 and D12S86 were used; for the exclusion of HMN V on chromosome 7 the markers D7S435, D7S1514, D7S2492, D7S632 were used (Christodoulou et al., 1995; Timmerman et al., 1996; Sambuughin et al.,1998). Juvenile amyotrophic lateral sclerosis, linked to chromosome 9q, was excluded using the markers D9S1830 and D9S1863 (Chance et al., 1998), Charcot–Marie–Tooth 2B was excluded by the use of the markers D3S1551 and DS31744, and markers D9S197 and D9S910 were used to exclude HSN I (Kwon et al., 1995; Nicholson et al., 1996). PCR was performed in a 15 μl reaction containing 1 pmol of each primer, 0.15 U DYNAZyme (Finzyme, Espoo, Finland) in a UNO II thermocycler (Biometra, Göttingen, Germany). Before gel analysis, 1 μl of the PCR product was mixed with 4 μl loading dye, denatured for 3 min at 70°C and loaded on a 4% polyacrylamide sequencing gel for the LI–COR sequencer. Data collection and analysis were performed using BaseImagIR Data Collection V4.00 and RFLPScan V3.00 as supplied by LI–COR.

Charcot–Marie–Tooth 1A caused by a duplication on chromosome 17p11.2 and proximal spinal muscular atrophy linked to chromosome 5q were tested by routine methods (Reiter et al., 1996) in one representative family member (IV/58).

Linkage analysis

Two-point linkage studies were performed using the LINKAGE computer package version 5.1 obtained from the Laboratory of Statistical Genetics at Rockefeller University, USA (http://linkage.rockefeller.edu). We used five age-dependent penetrance classes for performing linkage analysis in our family to account for the age at onset of the disease (Ott et al., 1991). Mean age at onset was 15.4 ± 3.42 years (n = 18). According to the classification described above, family members with an uncertain phenotype were assigned an unknown genetic status in our linkage calculations. For the calculations, we assumed equal female and male recombination fractions and a disease frequency of 10–4. Allele numbers and frequencies were obtained from the Genome Data Base (http://gdbwww.gdb.org).

Results

Clinical findings

Classification of family members

Forty-seven at-risk individuals underwent complete neurological examination. Mean age at investigation was 35 years. Eighteen individuals expressed the disease fully at the time of investigation and were aware of being affected. The remaining 29 at-risk individuals were screened systematically for HMN V by clinical and electrophysiological testing and two further definitely affected individuals were detected. One individual (III/51), who was clinically and electrophysiologically normal but gave birth to an affected child, was also classified as affected, although she did not fulfil the diagnostic criteria mentioned above. The asymptomatic status of this person might be the result of incomplete penetrance. Alternatively, the affected child IV/56 may even represent a phenocopy. Thus, a total of 21 definitely affected and 26 probably affected family members were identified. Histories of the deceased individuals of generation II and III revealed that nine further persons had been definitely affected.

Age at onset and presenting features in the 21 definitely affected family members

The age at onset was taken to be the time when the patient first noticed weakness in the hands and/or gait disturbance. Although this varied widely, from 10 to 38 years (mean = 15.4 ± 3.42 years), most patients (17/21) developed symptoms during the second decade of life. Three individuals were unable to define the time when the disease started. In 12 cases, atrophy of the thenar and FDI muscles was the initial and most prominent symptom, although on further questioning most of them remembered that foot deformity had been present, but so far unnoticed, to a variable degree since early childhood. In others, amyotrophy of the hands appeared later (3/9) or even remained absent (6/9). Interestingly, muscle atrophy in the hands was frequently asymmetrical in distribution (12/15), the right side being more severely affected than the left in most cases (9/12), while the lower limbs were always involved to an equal degree. Eight patients were brought to medical attention primarily because of marked foot deformity and gait problems. All patients except the clinically normal disease carrier had mild to very severe foot deformity, presenting as pes cavus, pes planus and/or hammer-toes. Less frequently, atrophy of the ankle extensors occurred. None of the patients complained about sensory loss and only two individuals noticed paraesthesia, and, in addition, muscle cramps in the legs. Eight persons reported prominent hyperhidrosis in the hands and feet, and five frequently had cold feet. Additional features such as hypacusis, depression and hip dysplasia were observed rarely. The disease was slowly progressive in all patients, but none was severely handicapped. Figures 2 and 3 demonstrate the variability of hand muscle involvement and foot deformity.

Fig. 2

Foot deformity and hand muscle wasting in patient III/37. This 47-year-old patient remembered having had foot deformity since early childhood. At the age of 18 years he first noticed wasting of the thenar and FDI muscles of the right hand. At that time he became aware of being afflicted with a hereditary disease. Foot deformity progressed to severe gait disturbance.

Fig. 3

Mild foot deformity and bilateral but asymmetrical hand muscle wasting in patient IV/56. This 16-year-old male first noticed wasting of the right thenar at the age of 14 years. This was slowly progressive, and started in the left hand 2 years later. He reported having had high arches as long as he could remember. Although foot deformity is only mild, motor NCV in the lower limbs was severely abnormal (peroneal NCV 29 m/s, distal motor latency 12.1 ms).

Neurological examination: results for the 21 definitely affected individuals

Proximal muscle strength was normal in both the upper and the lower limbs in all definitely affected individuals. In 15 persons muscle power in the small hand muscles was reduced depending on the degree of muscle wasting (MRC grade 0 to 4+). Fourteen individuals revealed diminished muscle power in the distal lower limbs, which occurred particularly in the toe and ankle extensors (MRC grade 1 to 4+). The tendon reflexes in the upper and lower limbs were normal (NINDS +2), diminished (NINDS +1) or brisk (NINDS +3 or +4) or rarely absent (NINDS = 0). Muscle tone was usually normal in the upper limbs and increased in the lower limbs in three patients only. A Babinski sign was present in only one patient. No individual had a deformity of the spinal column or an abnormality of the cranial nerves. Clinical and neurological findings for all 21 definitely affected family members are summarized in Table 1.

View this table:
Table 1

Clinical features in 21 patients with definite disease

Patient/pedigree/sexAge at examination (years)Age at onset of symptomatic disease (years)Foot deformity Peroneal muscle atrophyHand muscle wasting right/leftVibration UE/LETendon reflexes (Radial/ knee/ankle)Hyper-hidrosis
Reflex grading according to NINDS scale. UE = upper extremities; LE = lower extremities; + = mild; ++ = moderate; +++ = severe; – = not present; ? = unknown.
III/2/F6115+++++/++8/72/2/1+
III/3/M5912++++–/–7/63/4/3+
III/18/M4017+++++++/+7/63/4/3
III/19/F4014+++/+6/42/4/3
III/28/M5510+–/–7/52/0/0++
III/34/M5214++++++++/+6/42/3/2+
III/35/F5115+++–/–7/52/4/3+
III/37/M4718+++++++++/+6/53/3/0++
III/43/M68?++–/–6/62/3/0
III/45/F4338+++/+7/63/3/3
III/47/M50?+++++++/++7/42/3/2+
III/51/F40?–/–8/72/2/2
III/53/F3910+++++/+7/72/3/1
IV/4/F3215+++/+6/62/3/3
IV/6/M2818+++++/++8/71/3/1
IV/26/M1915++/+8/72/3/3
IV/ 27/M1212++++/++8/72/3/3
IV/33/M2814++++–/–7/52/3/3++
IV/37/F1217++/+++6/41/2/2
IV/56/M1614++++/++7/41/2/2
IV/58/M1910++++++/++7/52/3/1

Main findings in 26 probably affected family members

Nine individuals had mild foot deformity, 20 had brisk tendon reflexes in the upper or lower limbs and 14 complained about excessive sweating. None of them showed signs of muscle weakness or wasting or other additional features. One 18-year-old individual was clinically normal but had pathological median motor nerve conduction speeds. A further 52-year-old person was clinically normal but had marked prolonged central motor conduction times to the lower limbs, as did others.

Main findings in nine definitely affected deceased family members

Hand muscle involvement had been observed in six patients, although this was the first sign of the disease in only two of them. Seven individuals had had marked foot deformity and gait problems. One of them had been unable to walk at advanced age; all others remained ambulatory. One person had been asymptomatic but had children who were definitely affected.

Neurophysiological results

Nerve conduction velocity studies

NCV studies were performed in 21 definitely affected and 26 at-risk individuals. In general the median and the peroneal nerves were more severely impaired than the ulnar and tibial nerves. The values of motor NCV, distal motor latencies and compound motor action potentials (CMAPs) in the definitely affected individuals varied widely from normal to severely reduced, and this was strongly correlated with the degree of muscle wasting. Conduction block was never observed, but pronounced dispersion of CMAPs was frequently present. In only one case was it impossible to obtain a CMAP from the extensor digitorum brevis muscle. Median and sural sensory NCVs and sensory nerve action potentials were usually within the normal range, but occasionally sensory NCVs were slightly to moderately slowed and sensory nerve action potentials were reduced, corresponding to a severe clinical phenotype or a long duration of the disease. In 14/26 probably affected individuals, NCVs were slightly abnormal and frequently consisted of prolonged median distal motor latencies, whereas the NCVs of the ulnar, peroneal and tibial nerves were within the normal range in most cases, as were those of the sensory median and the sural nerves. In the majority of these people, pathological values were obtained in only one motor nerve and therefore cannot be associated unequivocally with the disease. Table 2 summarizes the results of motor and sensory nerve conduction studies.

View this table:
Table 2

Results of electrophysiological studies

Probably affected individualsDefinitely affected individualsP
Mean ± SD (n)95% confidence limitsMean ± SD (n)95% confidence limits
Family members were classified as definitely affected or probably affected as described in Patients and methods. As there was no side difference for any parameter, data for the two sides are pooled. M = median nerve; U = ulnar nerve; P = peroneal nerve; T = tibial nerve; SU = sural nerve; DML = distal motor latency; CV = conduction velocity; CMAP = compound motor action potential; S = sensory; A = amplitude.
M–DML 3.9 ± 0.66 (49) 3.7–4.1 4.5 ± 0.80 (40) 4.2–4.80.001
M–CV57.7 ± 7.42 (49)55.5–59.853.5 ± 5.70 (40)51.7–55.30.005
M–CMAP10.4 ± 3.46 (49) 9.4–11.4 5.9 ± 4.55 (40) 4.5–7.40.000
U–DML 3.1 ± 0.74 (23) 2.8–3.5 3.4 ± 0.77 (20) 3.1–3.80.209
U–CV61.1 ± 3.79 (19)59.2–62.955.3 ± 12.87 (18)48.9–61.70.070
U–CMAP14.2 ± 3.63 (23)12.6–15.712.2 ± 4.23 (19)10.2–14.30.121
P–DML 4.6 ± 0.79 (47) 4.4–4.8 7.3 ± 3.68 (35)60.0–8.50.000
P–CV49.7 ± 4.28 (47)48.4–51.038.3 ± 13.31 (35)33.8–42.90.000
P–CMAP 9.9 ± 4.14 (47)8.6–11.1 3.2 ± 4.65 (36) 1.6–4.70.000
1.6–4.4
T–DML 4.8 ± 0.73 (47) 4.6–5.0 6.5 ± 2.34 (36) 5.7–7.30.000
T–CV50.5 ± 3.88 (47)49.4–51.742.4 ± 5.94 (35)40.4–44.50.000
T–CMAP22.9 ± 8.17 (47)20.5–25.3 9.1 ± 8.75 (35) 6.1–12.10.000
M–S–CV55.4 ± 7.31 (24)52.3–58.548.4 ± 6.50 (27)45.9–51.00.001
M–S–A42.5 ± 23.84 (23)32.2–52.922.6 ± 16.86 (27)16.0–29.30.001
SU–CV45.7 ± 4.38 (32)44.2–47.340.0 ± 9.39 (27)36.3–43.70.003
SU–A21.6 ± 12.15 (32)17.3–26.011.3 ± 7.88 (15) 6.9–15.70.004

EMG

EMG revealed severely reduced recruitment, indicating pronounced loss of motor neurons. Individual motor unit potentials were often polyphasic and very high in amplitude (up to 15 mV). Spontaneous activity was rarely observed and, when present, consisted of low-frequency fibrillation or fasciculation potentials. These findings are consistent with chronic neurogenic disorder.

Magnetic evoked potentials

Motor evoked potentials were recorded in 15 definitely and 15 probably affected subjects and prolonged central motor conduction times (CMCTs) were observed in 9/15 definitely affected and 6/15 probably affected (χ2 test, n.s.). However, when comparing upper and lower extremities separately, prolonged CMCTs were more frequently observed in the definitely than the probably affected group (Fisher's exact test, P < 0.05). Thus, CMCTs were prolonged in 2/30 upper and 9/29 lower limbs in definitely affected patients while only 3/30 upper and 5/28 lower limbs showed prolonged CMCTs in probably affected persons. The means and standard deviations of CMCTs are shown in Table 3.

View this table:
Table 3

Central motor conduction times

Definitely affected individualsProbably affected individualsUpper limit of normal (ms)
Prolonged CMCT (ms)Normal CMCT (ms)Prolonged CMCT (ms)Normal CMCT (ms)
Means and standard deviations of central motor conduction times are shown for upper and lower limbs in patients classified as definitely and probably affected by HMN V. Numbers in brackets denote the frequencies of normal and prolonged CMCT within each subgroup. The upper limits of normal are the reference values established in our laboratory, representing the mean + 2.5 SD derived from a normal population of 40 healthy subjects. Recordings in the upper limb were taken from the abductor digiti minimi and in the lower limbs from the tibialis anterior muscle.
Upper limbs10.4 ± 0.3 6.6 ± 0.9 9.9 ± 1.1 6.2 ± 1.2 8.3
(n = 2/30)(n = 28/30)(n = 3/30)(n = 27/30)
Lower limbs20.4 ± 1.515.6 ± 2.322.8 ± 1.815.7 ± 1.318.5
(n = 10/29)(n = 19/29)(n = 5/28)(n = 23/28)

Results of molecular genetic studies

Genotyping of the definitely affected and probably affected individuals revealed recombinants for the HMN II and HMN V loci on chromosomes 12q and 7p, respectively. Two-point linkage studies resulted in significant negative LOD (log10 odds ratio) scores for almost all markers analysed (Table 4) except for the markers D7S2492 and D12S2079, which yielded inconclusive LOD scores (–2 < z < +3) because of uninformative alleles in the family. These results indicate that the disease in our family is not linked to the two hitherto described HMN loci on chromosomes 7p (distal HMN V) and 12q (distal HMN II). Furthermore, juvenile amyotrophic lateral sclerosis on chromosome 9q, Charcot–Marie–Tooth 2B and HSNI on chromosomes 3q and 9q were excluded by our linkage studies (data not shown). In addition, we calculated two-point LOD scores with individuals III/51 and IV/56, who had been classified as unknown because the clinically normal mother had an affected child and a phenocopy could not be ruled out. As a result of this modification, the LOD scores in our family changed only marginally and no significant differences were obtained.

View this table:
Table 4

Two-point LOD scores between HMN II and HMN V and various short tandem repeat markers

Recombination fractions
00.010.050.10.20.30.4
Chromosome 7
D7S435–3.65–2.80–1.66–0.93–0.220.020.05
D7S2492–0.76–0.72–0.290.070.300.260.14
D7S632–3.16–1.53–0.64–0.110.350.410.27
D7S1514–9.16–4.76–2.75–1.90–0.94–0.36–0.06
Chromosome 12
D12S86–2.43–1.04–0.46–0.250.110.250.18
D12S2079–0.35–0.34–0.170.010.130.100.02
D12S1282–2.37–1.90–0.510.190.580.490.25
D12S1349–3.09–1.64–0.93–0.430.110.190.13
D12S378–3.72–2.16–1.42–0.89–0.26–0.040.03
D12S340–3.47–1.76–0.66–0.210.050.160.14

Charcot–Marie–Tooth 1A and proximal spinal muscular atrophy linked to chromosome 5q were excluded in one severely affected patient (IV/58).

Discussion

The present study provides the largest series of HMN V patients reported so far and focuses on the clinical presentation, the neurophysiological characteristics and the genetic background of this disorder. In principle, it shows the enormous heterogeneity in the presenting signs, the distribution of wasting and upper motor neuron involvement. The diagnosis of HMN type V in this family was based primarily on the findings of predominant wasting and weakness restricted to the hands, which was the leading sign in >50% of the patients, while gait disturbance was usually less pronounced and sensory abnormalities were absent or only minimally present. In 1966, Silver described two families afflicted with a neuropathy of autosomal dominant inheritance, in which wasting of the hands was the first and most prominent manifestation of the disease, although differences were clearly present (Silver, 1966). While in the first family (family K) hand muscle involvement was frequently distributed asymmetrically to the thumb and the FDI muscles, patients of the second family (family A) invariably had bilateral symmetrical wasting of all small hand muscles, which developed at a later stage, but was more severe and disabling. Similar phenotypes were observed in the families reported by Lander and colleagues (Lander et al.,1976), van Gent and colleagues (van Gent et al., 1985) and de Visser and colleagues (de Visser et al.,1988), in which hand muscle involvement particularly resembled that of Silver's family K, as was the case in our family. In contrast to these families, however, all members of Gross's kinship (Gross et al., 1998) presented with symmetrical weakness of the distal muscles of the upper limbs in the distribution of the radial, ulnar and median nerves, similarly to Silver's family A. It is unclear whether this striking asymmetrical atrophy of the thenar and FDI muscles, noted in some families with HMN V, reflects the existence of a distinct genetic entity. Unfortunately, genetic linkage studies in the above-mentioned families have not yet been reported. In our family the disease was not associated with a gene locus on chromosome 7p, as found in the Bulgarian HMN V kinship (Christodoulou et al., 1995), the family from Iowa with Charcot–Marie–Tooth 2D (Ionasescu et al., 1996) and the Mongolian HMN V/Charcot–Marie–Tooth 2D family (Sambuughin et al., 1998), in which asymmetrical wasting and weakness of the thenar and FDI muscles were not reported.

A further representative and common feature in our HMN V patients was foot deformity of a variable degree, which occurred frequently in all the families with HMN V reported to date. However, special attention should be paid to those individuals in whom the disease starts in the feet and progresses to weakness and wasting of the ankle extensors, whereas the hands are involved later or might even be spared, as found in eight of our patients. On the one hand, such cases demonstrate the broad clinical variability of the disease. On the other hand, in these individuals the disorder cannot be distinguished from peroneal muscular atrophy with pyramidal features on purely clinical grounds (Harding and Thomas, 1984). This clearly emphasizes the importance of evaluating the family history carefully and the examination of further family members to arrive at a correct clinical diagnosis.

Hyper-reflexia was most frequently associated with predominant wasting of the small hand muscles, and can therefore be considered as a typical, albeit inconstant, finding in HMN V. Occasionally, additional increased muscle tone, extensor plantar responses and a spastic gait indicate involvement of the pyramidal tract in the disease. In the families reported by Lander and colleagues (Lander et al., 1976), van Gent and colleagues (van Gent et al., 1985) and Gross and colleagues (Gross et al., 1998) and in our family, pyramidal signs were usually only slight, whereas they were absent or barely present in the chromosome 7p-linked families (Christodoulou et al., 1995; Sambuughin et al., 1998). However, in some of the affected family members of Silver (Silver, 1966) and de Visser and colleagues (de Visser et al., 1988), pyramidal tract involvement was so prominent that the disorder was classified as hereditary spastic paraplegia. Since both authors observed individuals with hand muscle involvement or gait and pyramidal disturbance only, it has been suggested that both lower and upper motor neuron signs were segregating as independent autosomal dominant traits. This seems very unlikely in our family, as we found individuals displaying clinical features that resembled peroneal muscular atrophy with mild pyramidal features without hand muscle involvement, who, surprisingly had an affected child with predominant hand muscle wasting. Thus, we hypothesize that both features are caused by only one mutated gene, which, however, can give rise to marked phenotypic differences.

The absence or presence of sensory abnormalities has been used as the most important feature to distinguish HMN from HMSN. Nevertheless, it has been demonstrated already that both phenotypes, with or without sensory loss, can be observed within one family (van Gent et al., 1985; Sambuughin et al., 1998). This, again, has implications for the classification within this disease group. We, too, found impaired vibration and mildly to moderately slowed sensory NCV and reduced sensory nerve amplitudes in older and severely affected patients, indicating that the sensory nerves may become involved additionally with advanced disease. Therefore, with regard to the families of van Gent and colleagues (van Gent et al., 1985) and Sambuughin and colleagues (Sambuughin et al., 1998) and our family, we believe that distinction between HMN and HMSN cannot be based exclusively on the involvement of the sensory nerves. Of more reliable value might be the relation between the pathology of a motor nerve and its analogous sensory nerve; in HMN, motor nerves can be severely impaired or CMAPs might even not be recordable, while the sensory nerves remain normal for a long time, whereas in HMSN both motor and sensory nerves are affected simultaneously.

We have not undertaken studies to determine autonomic nervous system involvement in our family. Hyperhidrosis, which was frequently reported by our patients and individuals who were probably affected, may be a feature of additional involvement of the autonomic nervous system.

Other findings in our kindred, such as depression, hip dysplasia and hypacusis, seem to occur by chance as they have been observed only in single patients.

The results of NCV and EMG studies were consistent with a predominant chronic axonal motor neuropathy in the majority of the definitely affected patients investigated. The CMAPs were largely reduced in all but the ulnar nerves and the EMG showed high-amplitude potentials and reduced recruitment, whereas distal motor latencies and motor NCVs were normal or mildly to severely abnormal. In the upper limbs the median nerve was significantly more severely damaged than the ulnar nerve, suggesting that the median nerve is most useful for screening of this type of HMN V. Our observation in one 18-year-old female with a normal phenotype but mild abnormalities of the median motor NCV might emphasize that the disease is sometimes recognized only electrophysiologically, and underlines the importance of electrophysiological assessment in detecting gene carriers.

The sensory nerves were mildly to moderately involved in advanced disease and the changes were predominantly axonal. At later stages of the disease, the electrophysiological features may be indistinguishable from the findings in HMSN II (Chad, 1989; Emeryk–Szajewska et al., 1998; Paraskevas et al., 1998), but the clinical characteristics are distinctive enough to exclude the latter disease.

Transcranial magnetic stimulation revealed prolonged CMCTs in 50% of the family members studied, indicating additional CNS involvement in HMN V. Prolonged CMCTs have been reported in hereditary neuropathies, but the results so far remain conflicting. In HMSN I some authors have found central motor pathway involvement independently of clinical pyramidal signs (Mano et al., 1993; Sartucci et al., 1997), and in 1990 Claus and colleagues observed prolonged CMCTs in the presence of only pyramidal signs (Claus et al., 1990). In the latter study, the slowing of central motor conduction was less pronounced in patients with the axonal variant, HMSN II, suggesting a different pathology. Disease-related changes in motor neuron excitability might also account for the slight slowing of CMCT noted in our patients. In support of a report by Schnider and colleagues, who observed prolonged CMCTs in clinically unaffected members of a HMSN V family (Schnider et al., 1991), we occasionally recorded prolonged CMCTs in probably affected individuals, even in the absence of clinical signs of the disease. Whether this reflects predominant involvement of the corticospinal tract in HMN V or earlier expression of disease-related changes in long tracts cannot be decided at present.

Two loci for HMN have been reported to date. HMN II has been mapped to chromosome 12q and HMN V to chromosome 7p (Christodoulou et al., 1995; Timmerman et al., 1996). Linkage and segregation analysis with six chromosome 12 and four chromosome 7 markers yielded conclusively negative LOD scores, excluding these types of HMN in our family. This strongly implies the existence of a further locus responsible for HMN and particularly highlights the genetic heterogeneity of HMN V.

In conclusion, the findings in this family with HMN V (the largest such family ever reported) demonstrate clearly that this disorder represents a phenotypically and genetically heterogeneous disease with autosomal dominant inheritance. At least a second genetic subtype of HMN V must exist, which may be distinguished from HMN V linked to chromosome 7p by the presence of asymmetrical weakness and wasting of the thenar and FDI muscles and pyramidal involvement, sometimes evidenced by brisk tendon reflexes and/or prolonged central motor conduction times. Foot deformity and sensory disturbance are less reliable in the differential diagnosis of both types of HMN V. Due to its broad clinical variability, the disorder can mimic peroneal muscular atrophy with pyramidal features in individual cases. Electrophysiological studies might occasionally identify subclinical involvement in clinically normal family members. Genetic linkage studies of further HMN V families are needed to expand our knowledge of phenotype–genotype correlations in different types of HMN V and to confirm the differential diagnostic considerations based on observations in our HMN V patients. A genome-wide search in this large family is warranted to elucidate the second gene locus responsible for HMN V.

Acknowledgments

We gratefully acknowledge the cooperation of the family members and thank Mrs Margit Schuster and Mrs Gerda Zmugg for their excellent technical assistance. This work was supported by the Austrian Science Fund (FWF), grant P13563–BIO. We thank Professor K. V. Toyka, Würzburg, Germany, for critical reading of the manuscript.

References

View Abstract