OUP user menu

Cortical hyperexcitability may precede the onset of familial amyotrophic lateral sclerosis

Steve Vucic, Garth A. Nicholson, Matthew C. Kiernan
DOI: http://dx.doi.org/10.1093/brain/awn071 1540-1550 First published online: 9 May 2008

This article has a correction. Please see:

Summary

Familial amyotrophic lateral sclerosis (FALS) is an inherited neurodegenerative disorder of the motor neurons. While 10–15% of cases are caused by mutations in the copper/zinc superoxide-dismutase-1 (SOD-1) gene, the dying-forward hypothesis, in which corticomotoneurons induce anterograde excitotoxic motoneuron degeneration, has been proposed as a potential mechanism. The present study applied novel threshold tracking transcranial magnetic stimulation techniques to investigate the mechanisms underlying neurodegeneration in FALS. Studies were undertaken in 14 asymptomatic and 3 pre-symptomatic SOD-1 mutation carriers, followed longitudinally for up to 3-years. The pre-symptomatic subjects were asymptomatic at the time of their initial study but developed symptoms during the follow-up period. Results were compared to 7 SOD-1 FALS patients, 50 sporadic ALS patients and 55 normal controls. Short-interval intracortical inhibition (SICI) was significantly reduced in SOD-1 FALS (1.2 ± 0.6%) and sporadic ALS patients (0.7 ± 0.3%) compared to asymptomatic SOD-1 mutation carriers (9.8 ± 1.5%, P<0.00001) and normal controls (8.5 ± 1.0%, P<0.00001). SICI reduction was accompanied by increases in intracortical facilitation, motor evoked potential amplitudes and the slope of the magnetic stimulus-response curve. In two pre-symptomatic SOD-1 mutation carriers SICI was completely absent (SICI patient 1, 3.2%; patients 2, 1.3%), while in one subject there was a 32% reduction in SICI prior to symptom onset. These three individuals subsequently developed clinical features of ALS. Simultaneous investigation of central and peripheral excitability has established that cortical hyperexcitability develops in clinically affected SOD-1 FALS patients, similar to that seen in sporadic ALS patients, thereby suggesting that a similar pathophysiological process in evident in both familial and sporadic ALS patients. In addition, the present study has established that cortical hyperexcitability precedes the development of clinical symptoms in pre-symptomatic carriers of the SOD1 mutation, thereby suggesting that cortical hyperexcitability underlies neurodegeneration in FALS.

  • familial ALS
  • cortical hyperexcitability
  • SOD-1

Introduction

Amyotrophic lateral sclerosis (ALS) is a rapidly progressive neurodegenerative disorder of the motor neurons in the cortex, brainstem and spinal cord (Borasio and Miller, 2001; Rowland and Shneider, 2001). Degeneration of motor neurons results in progressive paresis of limb, bulbar and respiratory muscles leading to death typically 3–5 years after symptom onset (Cudkowicz et al., 2004).

Ten percent of all ALS cases are familial (FALS), in which two or more family members are clinically affected (Andersen, 2006a). Autosomal dominant and recessive modes of inheritance have been reported and eight genetic loci have been identified in FALS to date (Andersen, 2006a). Mutations in one of these genes, the copper/zinc superoxide-dismutase-1 gene (SOD-1), result in a typical ALS phenotype, called ALS 1 (Rosen et al., 1993). The SOD-1 gene is located on chromosome 21 q22.1, spans 11 kilobases of genomic DNA and comprises 5 exons and 4 introns (Levanon et al., 1985). To date over 122 different SOD-1 mutations have been reported in FALS, the majority being missense mutations with an autosomal-dominant mode of inheritance evident in most mutations (Gros-Louis et al., 2006; Andersen, 2006a). The penetrance in SOD-1 FALS, which refers to the proportion of individuals with a particular genotype expressing the disease, is age dependent, with ∼ 50% of patients expressing the disease by age 43 years and >90% by 70 years (Cudkowicz et al., 1997; Robberecht, 2002; Aggarwal and Nicholson, 2005). Only 17 mutations are associated with this high degree of penetrance (Andersen, 2006a). Some mutations, such as the I113T may be transmitted asymptomatically from grandparents to grandchildren (Suthers et al., 1994; Jones et al., 1995).

Mutations in the SOD-1 gene result in acquisition of new cytotoxic activity by the SOD-1 protein, referred to as a ‘toxic gain of function’ (Andersen, 2006b). Although the mechanisms by which the SOD-1 protein results in neurodegeneration remain as yet undefined, from the transgenic SOD-1 mouse model there is increasing evidence of glutamate-mediated excitotoxicity (Van Den Bosch et al., 2006). Specifically, impaired glutamate clearance and increased glutamate concentration within the cortex have been reported in the SOD-1 (G93A) mice (Roy et al., 1998). The presence of SOD-1 mutations also increases motor neuron sensitivity to glutamatergic excitotoxicity by post-synaptic calcium-dependent mechanisms (Van Den Bosch et al., 2006). Of further relevance, SOD-1 mutations may result in mitochondrial dysfunction, further increasing the susceptibility of motor neurons to glutamate-mediated cell death (Hand and Rouleau, 2002).

Cortical hyperexcitability is mediated by glutamate, the principle excitatory neurotransmitter in the brain (Hand and Rouleau, 2002; Stys, 2005), which causes neurotoxicity by activating Ca2+-dependent enzyme systems (Stys, 2005). Of clinical relevance, cortical excitability may be assessed by transcranial magnetic stimulation (TMS) techniques (Kujirai et al., 1993; Vucic et al., 2006). TMS studies in sporadic ALS patients, have provided evidence for increased excitability early in the course of disease (Eisen et al., 1993; Prout and Eisen, 1994; Mills and Nithi, 1997; Vucic and Kiernan, 2006; Ziemann et al., 1997b). Conversely, in SOD-1 FALS patients the TMS findings have been contradictory, with some reporting cortical hyperexcitability (Weber et al., 2000; Turner et al., 2005b), while others have reported normal cortical excitability (Stewart et al., 2006). This discrepancy may be explained in part by marked variability in the motor evoked potential (MEP) amplitude with consecutive stimuli resulting from spontaneous fluctuations in the resting threshold of cortical neurons (Kiers et al., 1993). In order to clarify these discordant findings in FALS, and to further investigate the site of disease onset in ALS, novel threshold tracking TMS techniques were applied longitudinally in asymptomatic SOD-1 mutation carriers with the results compared to FALS patients manifesting the disease and to sporadic ALS patients.

Methods

Patients

Subjects were recruited from the SOD-1 database at the Molecular Genetic Laboratory, Concord Hospital and from the multidisciplinary motor neuron disease clinic at Prince of Wales Hospital (Sydney, Australia). In total 17 asymptomatic and pre-symptomatic SOD-1 mutation carriers (7 males, 10 females: age range 23–60 years, mean age: 40 years) were followed longitudinally for up 3 years (2.2 ± 0.2 years), in addition to seven clinically definite familial ALS patients as per the revised El Escorial criteria (Brooks et al., 2000) who also expressed SOD-1 mutation (4 males, 3 females: age range 25–70 years, mean age 47 years). The pre-symptomatic subjects were also asymptomatic at the time of their initial study but developed symptoms during the follow-up period. Families with three different SOD-1 mutations were studied including: valine to glycine mutation in exon 5 at codon position 148 (V148G); isoleucine to threonine mutation in exon 4 at codon position 113 (I113T); glutamic acid to glycine mutation in exon 4 at codon position 100 (E100G). The pattern of inheritance was autosomal dominant and all subjects were heterozygous for the SOD-1 mutation. Fifty sporadic ALS patients were also studied (32 males, 18 females: age range 26–78 years, mean age 58 years).

ALS patients were clinically reviewed on a regular basis through the multidisciplinary MND clinic at Prince of Wales Hospital. This incorporated assessment using the amyotrophic lateral sclerosis function rating scale revised (ALSFRS-R) (Cedarbaum et al., 1999). Strength was assessed using the Medical Research Council (MRC) rating scale (Medical Research Council, 1976), and a separate hand function score was also recorded (Triggs et al., 1999). Asymptomatic SOD-1 mutation carriers also underwent clinical examination at the time of formal neurophysiological assessment. The results of cortical excitability for familial and sporadic ALS patients were compared to 55 healthy controls (28 men; 27 women, aged 23–83 years, mean: 47 years). All subjects gave informed consent to the procedures, which were approved by the South East Sydney Area Health Service Human Research Ethics Committee.

Peripheral nerve testing

The median nerve was stimulated electrically at the wrist in all studies. The resultant compound muscle action potential (CMAP) was recorded from the abductor pollicis brevis monolaterally in most cases, and was measured from baseline to negative peak. From peripheral nerve conduction studies, the neurophysiological index (NI), a marker of ALS progression, was derived according to a previously reported formula (de Carvalho and Swash, 2000): Embedded Image where F-wave frequency was expressed as the number of F responses recorded in 20 trials.

Cortical excitability

Subsequently, cortical excitability was assessed by applying TMS to the motor cortex by means of a 90 mm circular coil oriented to induce current flow in a posterior–anterior direction. The coil was adjusted until the optimal position for an evoked response was obtained from the abductor pollicis brevis muscle monolaterally. Currents were generated by two high-power magnetic stimulators which were connected via a BiStim (Magstim Co., Whitland, South West Wales, UK) such that conditioning and test stimuli could be independently set and delivered through the one coil.

Threshold tracking paired-pulse TMS was performed according to a previously reported method (Vucic et al., 2006). The threshold tracking strategy used a target response of 0.2 mV (± 20%), located in the middle of the established linear relationship between the logarithm of the MEP amplitude and the stimulus intensity (Vucic et al., 2006). Resting motor threshold (RMT) was defined as the stimulus intensity required to produce and maintain this target MEP response.

Initially, the stimulus response curve for cortical stimulation was determined by increasing the intensity of the magnetic stimulus to the following levels: 60, 80, 90, 100, 110, 120, 130, 140 and 150% RMT. Three stimuli were delivered at each level of stimulus intensity. The maximum MEP amplitude (mV) and MEP onset latency (ms) were recorded. Central motor conduction time (CMCT, ms) was calculated according to the F-wave method (Cros et al., 1990). Subsequently, the cortical silent period (CSP) induced by single-pulse TMS was recorded while ALS patients performed a voluntary contraction, set to ∼30% of maximal voluntary contraction (Vucic et al., 2006). The level of contraction was monitored by recording the surface EMG of the abductor pollicis brevis. While performing the task, patients were provided with visual feedback information. As for the stimulus response curves, CSP duration curves were determined by increasing the intensity of the magnetic stimulus to the following levels: 60, 80, 90, 100, 110, 120, 130, 140 and 150% RMT. The CSP duration was measured from the beginning of MEP to the return of EMG activity (Cantello et al., 1992).

Short-interval intracortical inhibition (SICI) and intracortical fascilitation (ICF) were determined by using subthreshold conditioning stimuli (70% RMT) at increasing interstimulus intervals (ISIs) delivered in a fixed order as follows: 1, 1.5, 2, 2.5, 3, 3.5, 4, 5, 7, 10, 15, 20 and 30 ms (Vucic et al., 2006). Stimuli were delivered sequentially as a series of three channels: channel 1 tracked the stimulus intensity required to produce the unconditioned test response (i.e. RMT); channel 2 monitored the response to the subthreshold conditioning stimulus; and channel 3 tracked the stimulus required to produce the target MEP when conditioned by a stimulus equal in intensity to that on channel 2. Stimuli were delivered every 5–10 s and the computer advanced to the next ISI only when tracking was stable.

SICI was measured as the increase in the test stimulus intensity required to evoke the target MEP. Inhibition was calculated using the following formula: Embedded Image Facilitation was measured as the decrease in the conditioned test stimulus intensity required to evoke a target MEP.

Recordings of CMAP and MEP were amplified and filtered (3 Hz–3 kHz) using a GRASS ICP511 AC amplifier (Grass-Telefactor, Astro-Med Inc., West Warwick, RI, USA) and sampled at 10 kHz using a 16-bit data acquisition card (National Instruments PCI-MIO-16E-4). Data acquisition and stimulation delivery (both electrical and magnetic) were controlled by QTRACS software (© Professor Hugh Bostock, Institute of Neurology, Queen Square, London, UK).

Statistical analysis

Student's t-test was used to compare mean differences between groups for averaged SICI (1–7 ms), ICF (10–30 ms) and RMT. Analysis of variance (ANOVA) was used for multiple comparisons between magnetic stimulus response and CSP curves generated by increasing the test stimulus intensity from 60% to 150% RMT. A probability (P) value of <0.05 was considered statistically significant. Results are expressed as mean ± standard error of the mean.

Results

The clinical and genetic features for 17 asymptomatic SOD-1 mutation carriers and 7 clinically affected FALS patients are summarized in Table 1. The asymptomatic SOD-1 mutation carriers underwent annual TMS testing at which time they were clinically reviewed by the same examiner. The physical examination in all the asymptomatic SOD-1 mutation carriers, including the three pre-symptomatic carriers was normal. Specifically, there were no upper motor neuron features, such as increased muscle tone, hyper-reflexia, extensor plantar responses, positive Hoffman sign or the presence of a jaw-jerk in any of the subjects at the time of testing. Three SOD-1 mutation carrier subjects were pre-symptomatic, i.e. asymptomatic at the time of TMS, and their case histories are outlined.

View this table:
Table 1

Clinical details for the 17 asymptomatic superoxide dismutase-1 (SOD-1) mutation carriers and 7 clinically affected familial amyotrophic lateral sclerosis (FALS) with SOD-1 mutation

Asymptomatic SOD-1Age (years)/ SexSOD-1 mutationFollow-up period (years)ALS onsetTime to ALS onset (months)ALSFRS-RTriggs hand scoreMRC
1a43/MV148G3UL, LL3(46) 4805
2a34/ME100G2UL, LL3(47) 4805
3a41/FV148G3UL, LL8(44) 4805
425/FV148G34805
560/FI113T14805
647/FV148G34805
737/MV148G34805
825/FI113T34805
944/FV148G34805
1043/FV148G34805
1149/MV148G14805
1223/FV148G24805
1345/FE100G14805
1457/MI113T24805
1531/MV148G34805
1654MI113T24805
1731/FV148G14805
Mean402.24.74805
SEM3.10.21.700
Symptomatic SOD1 patientsAge (years)/ SexSOD-1 mutationALS onsetDisease duration (months)ALSFRS-RTriggs hand scoreMRC
1870/FI113TUL94124
1935/MV148BLL334204
2049/FE100GLL143714
2144/MV148GLL64505
2253/MI113TUL74614
2350/MV148GLL133805
24b25/FV148GLL1.54805
Mean4711.942.40.64.4
SEM5.43.91.60.30.2
Sporadic ALS patients (N = 50)
Mean58.721.638.21.24.2
SEM1.83.40.90.10.1
  • aAlthough the ALSFRS-R scores in three pre-symptomatic SOD-1 mutation carriers were normal at the time of cortical excitability testing, the ALSFRS-R reduced with the onset of clinical features of ALS some time after cortical excitability testing (ALSFRS-R for case#1, 46; case#2, 47; case#3, 44). In addition, for case 3, the MRC score was normal at the time of cortical excitability testing in May 2005 and April 2006. However, at last testing, in 2007, the MRC score was reduced as measured from the weak left abductor pollicis brevis muscle (see case history #3 in Results). bPatient 24 reported difficulty with running and endurance training. Physical examination disclosed a generalized hyperreflexia and neurophysiological testing revealed neurogenic changes, consistent with ALS. Data for 50 sporadic ALS patients (32 males and 18 females) is also presented for comparison. Twenty-nine ALS patients reported upper limb onset, 12 lower limb, 8 bulbar and 1 respiratory onset of disease.

  • Three different SOD-1 mutations were present in the patients studied; GTA to GGA sequence change resulting in valine to glycine substitution in exon 5 at the 148 position (V148G), GAA to GGA sequence change resulting in a glutamic acid to glycine substitution in exon 4 at the 100 position (E100G), and ATT to ACT sequence change resulting in an isoleucine to threonine substitution in exon 4 at the 113 position (I113T). The follow-up period refers to the time period between first assessment at time of cortical excitability testing to last review, measured in years. The site of disease onset was classified as either upper limb (UL), lower limb (LL) or bulbar (B). Disease duration refers to the period from symptom onset to date of testing, while time to ALS onset refers to time to development of clinical features of ALS after the last testing. The patients were clinically graded using the amyotrophic lateral sclerosis functional rating scale revised (ALSFRS-R), with a maximum score of 48 when there is no disability. Muscle strength was clinically assessed using the Medical Research Council (MRC) for the abductor pollicis brevis, as this muscle was utilized for excitability testing.

Case histories

Case 1 (Table 1)

A 43-year-old male who underwent initial threshold tracking TMS in June 2005. At the time of testing, the subject was asymptomatic and formal neurological examination was normal. Furthermore, the CMAP amplitude (9.7 mV) and neurophysiological index (2.2) were within normal limits. Although surface EMG, including testing over the target muscle was unremarkable, formal needle EMG testing was not undertaken at time of cortical excitability testing. In September of 2005, this subject developed fasciculations, muscle cramps and weakness of the right foot and hip. Subsequently, he developed wasting of the muscles in the right leg, including calf, anterior leg and thigh muscles. By December 2006, he developed fasciculations in the proximal upper limb muscles along with weakness of the shoulder girdle muscles. Neurophysiological testing in September 2005 established features of denervation in the lower limbs accompanied by fasciculations.

Case 2 (Table 1)

A 34-year-old male who underwent threshold tracking TMS testing in June 2006. Three months later, he reported generalized fasciculations and difficulty running. Although the neurological examination was normal at the time of TMS testing, re-assessment at the time of symptom onset disclosed widespread fasciculations with a generalized hyperreflexia. At the time of cortical excitability testing, the CMAP amplitude (9.1 mV) and NI (2.3) were unremarkable. Formal needle EMG testing was not undertaken at the time of cortical excitability testing. However, subsequent needle EMG testing disclosed fasciculations of complex morphology, with associated neurogenic abnormalities.

Case 3 (Table 1)

A 41-year-old female who underwent TMS testing in May 2005 and April 2006. Peripheral neurophysiological testing, in the form of CMAP amplitude and NI, in May 2005 (CMAP amplitude, 13 mV; NI, 3.3) and April 2006 (CMAP amplitude, 12.8 mV; NI, 3.0) were normal, as was surface EMG sampling. In December 2006, the subject reported lower-limb weakness manifesting as difficulty with walking and standing on her toes. By late January 2007, the subject developed exertional dyspnoea, left-hand weakness (MRC grade 4, see Table 1) and generalized fasciculations. Neurophysiological testing, nerve conduction studies and needle electromyography (EMG), disclosed widespread neurogenic changes consistent with ALS. The subject underwent further TMS testing in April 2007. Currently, the patient is wheelchair-dependent, with global weakness, and requires nocturnal pressure support ventilation. The SOD-1 mutation (V148G) in this family results in an aggressive ALS phenotype with median survival from disease onset being typically <12 months.

Peripheral nerve studies

The CMAP amplitude (symptomatic FALS 6.1 ± 1.6 mV; sporadic ALS 5.9 ± 0.6 mV; controls 10.3 ± 0.5 mV, P < 0.001) and neurophysiological index (symptomatic FALS 1.3 ± 0.4; sporadic ALS, 0.8 ± 0.1; controls 2.6 ± 0.2, P < 0.005), both quantitative markers of peripheral disease burden, were significantly reduced in symptomatic FALS and sporadic ALS patients compared to controls. There were no differences in the CMAP amplitude (asymptomatic SOD-1 mutation carriers, 9.9 ± 0.8 mV; controls 10.3 ± 0.5 mV) and neurophysiological index (asymptomatic SOD-1 mutation carriers, 2.3 ± 0.3; controls 2.6 ± 0.2) between asymptomatic SOD-1 mutation carriers and normal controls. Although the MEP onset latency was longer in FALS and sporadic ALS patients (symptomatic FALS 22.6 ± 1.1 ms; ALS 22.1 ± 0.3 ms; asymptomatic SOD-1 mutation carriers 20.9 ± 0.7 ms; controls 20.4 ± 0.3 ms, P < 0.01), there was no significant difference in the CMCT between FALS and sporadic ALS patients compared to asymptomatic SOD-1 mutation carriers and controls (symptomatic FALS 5.0 ± 0.5 ms; ALS 5.1 ± 0.3 ms; asymptomatic SOD-1 mutation carriers 5.2 ± 0.4 ms; controls 5.1 ± 0.2 ms). The MEP onset latency in the three pre-symptomatic subjects (case #1, 19.2 ms; case #2, 20.3 ms; case #3, 20.3 ms) and the CMCT (case #1, 5.7 ms; case #2, 5.3 ms; case #3, 5.3 ms) were unremarkable.

Cortical excitability

The full sequence of cortical excitability was well tolerated and successfully completed in all asymptomatic SOD-1 mutation carriers, FALS and sporadic ALS patients. Short-interval intracortical inhibition, as reflected by an increase in the conditioned stimulus intensity required to track a constant target MEP of 0.2 mV, was significantly reduced in symptomatic FALS (−1.2 ± 0.6%) and sporadic ALS (0.8 ± 1.2%) patients compared to asymptomatic SOD-1 mutation carriers (9.8 ± 1.5%, N = 14) and controls (8.5 ± 1.1%, P < 0.001, Fig. 1A and B). In two pre-symptomatic SOD-1 mutation carriers who developed clinical features of ALS within 3-months of assessment of cortical excitability, SICI was absent (averaged SIC 1–7 ms case #1, −3.2%; case #2, −1.3%, Table 1, Fig. 2), in sharp contrast to the remaining asymptomatic SOD-1 mutation carriers (Fig. 2). In addition, peak SICI at ISIs <1 and 3 ms were reduced in these two pre-symptomatic SOD-1 mutation carriers (case#1 SICI ISI <1 ms = −4.8%, ISI 3 ms = −3.8%; case #2 SICI ISI <1 ms = −8.4%, ISI 3 ms = 1.4%) when compared to asymptomatic SOD-1 mutation carriers (SICI ISI <1 ms = 6.8 ± 0.7%, ISI 3 ms = 11.8 ± 0.6%) and controls (SICI ISI <1 ms = 7.0 ± 0.9%, ISI 3 ms = 13.7 ± 1.3%) In the third pre-symptomatic SOD-1 mutation carrier (case #3), there was a 32% reduction in averaged SICI prior to the development of clinical features of ALS and SICI was reduced in the motor cortex contralateral to side of development of muscle weakness (Fig. 3A and B). In addition, there was a marked reduction in SICI at ISI <1 (study 1 SICI <1 ms = 4.5%, study 2 SICI <1 ms = 0%, study 3 SICI <1 ms = 0%) and at ISI 3 ms (study 1 SICI 3 ms = 18.6%, study 2 SICI 3 ms = 8.7%, study 3 SICI 3 ms = 6.7%) prior to development of symptoms. As part of a subsequent diagnostic work-up, this subject underwent a muscle biopsy which revealed the presence of groups of atrophic muscle fibres with mixed-type fibres, a finding distinctive of motor neuron disease (Fig. 3C) (Baloh et al., 2007). In contrast to these three subjects, in the remaining asymptomatic SOD-1 mutation carriers, there was a non-significant increase of SICI over the follow-up period (averaged SICI year 2, 11.0 ± 1.5%, P = 0.3, N = 10; year 3, 11.9 ± 2.7%, P = 0.24, N = 7).

Fig. 1

(A) SICI, defined as the stimulus intensity required to maintain a target output of 0.2 mV (see ‘Methods’ section) was reduced in clinically affected familial amyotrophic lateral sclerosis (FALS) and sporadic ALS patients when compared to normal controls (P < 0.001). (B) Averaged SICI, between interstimulus interval (ISI) 1–7 ms was also reduced in FALS and sporadic ALS patients compared to asymptomatic superoxide dismutase-1 (SOD-1) mutation carriers and controls (P < 0.001). The error bars represent standard error of the mean. The asymptomatic SOD-1 mutation carrier data point does not include values from pre-symptomatic patients.

Fig. 2

(A) Averaged SICI, between 1 and 7 ms was reduced in a pre-symptomatic SOD-1 mutation carrier (case #3, see Table 1) prior to the development of ALS. SICI in the follow-up studies is expressed as a fraction of the SICI value measured at the first study, that is the SICI values were normalized. (B) In case #3, SICI was reduced in the right motor cortex contralateral to the side of muscle weakness. Stimulation of both cortices was undertaken in only one subject (case #3) and performed at the time of study 3 in 2007. (C) Biopsy of the left vastus lateralis muscle, obtained in March 2007 and stained with ATPase pH 9.4. Type I fibres are light and type II fibres are dark. The biopsy demonstrates grouped atrophic fibres with both type I and type II fibres (mixed-type fibres, illustrated by red arrow, magnification × 10). * The asymptomatic SOD-1 mutation carrier data point does not include values from case #1–3. The error bars represent standard error of the mean.

Fig. 3

Averaged SICI is depicted for three pre-symptomatic subjects (case #1–3, from Table 1) and 14 asymptomatic superoxide dismutase-1 (SOD-1) mutation carriers (subjects 4–17, Table 1). Averaged SICI was absent in two pre-symptomatic subjects, but was present in all the asymptomatic SOD-1 mutation carriers. The numbers on the x-axis, correspond to the number ascribed to each subject in Table 1. Although subject #3 appeared to have normal SICI, there was reduction of 32% between studies 1 and 2, as depicted in Fig. 2A, suggesting that a relative reduction in SICI may herald cortical hyperexcitability.

Short-interval intracortical inhibition is followed by a period of ICF marked by a decrease in the test stimulus intensity required to maintain the target MEP of 0.2 mV. ICF was increased in symptomatic FALS (−2.5 ± 0.7%) and sporadic ALS patients (−3.0 ± 1.2%) when compared to asymptomatic SOD-1 mutation carriers (−0.1 ± 0.5%) and controls (−1 ± 0.3%) over a period of testing from 10 to 30 ms (Fig. 4A, P < 0.05). ICF increased in two pre-symptomatic SOD-1 mutation carriers prior to the development of ALS (case #1, −9.3%; case #2, −6.1%) compared to remaining asymptomatic SOD-1 mutation carriers and controls (Fig. 4B). In the third pre-symptomatic SOD-1 mutation carrier (case #3), ICF again increased in the motor cortex contralateral to the side of muscle weakness (Fig. 4C and D).

Fig. 4

(A) Intracortical facilitation (ICF), which develops at interstimulus intervals (ISI) of 10–30 ms, was increased in clinically affected familial amyotrophic lateral sclerosis (FALS) and sporadic amyotrophic lateral sclerosis (SALS) patients compared to asymptomatic SOD-1 mutation carriers and controls. (B) In two pre-symptomatic superoxide dismutase-1 (SOD-1) mutation carriers (case #1 and #2), ICF was increased compared to the remaining asymptomatic superoxide dismutase-1 (SOD-1) mutation carriers and controls prior to development of clinical features of ALS. (C) In a further pre-symptomatic SOD-1 mutation carrier (case #3), ICF increased prior to the development of ALS. (D) Intracortical facilitation was increased in the right motor cortex, which was contralateral to the side of development of muscle weakness. Stimulation of both cortices was undertaken in only one subject (case #3). The error bars represent standard error of the mean.

Another measure of cortical excitability, the MEP amplitude which is expressed as a percentage of the CMAP response, reflects the density of corticospinal projection onto the anterior horn cells. The MEP amplitude was increased in both clinically affected FALS patients and sporadic ALS patients compared to asymptomatic SOD-1 mutation carriers and controls (symptomatic FALS 42.8 ± 13.0%; ALS 37.7 ± 4.5%; asymptomatic SOD-1 mutation carriers 20.6 ± 4.2%; controls 26.2 ± 2.4%, P < 0.05). Interestingly, in the three pre-symptomatic subjects, the MEP amplitude was not increased (case #1, 18.8%; case #2, 25%; case #3, 13%). The absolute MEP amplitude in clinically affected FALS was 1.5 ± 0.3 mV, in sporadic ALS 2.1 ± 0.3 mV, asymptomatic SOD-1 mutation carriers 2.1 ± 0.6 mV and controls 2.6 ± 0.7 mV. In the three pre-symptomatic subjects the absolute MEP amplitude was as follows: case #1, 1.8 mV; case #2, 0.5 mV; case #3, 1.6 mV. The slope of the tangent for the steepest point of the stimulus-response curve was greater in symptomatic FALS patients compared to asymptomatic SOD-1 mutation carriers and controls (ANOVA P < 0.05, Fig. 5).

Fig. 5

The motor evoked potentials (MEP) amplitude and the gradient of the stimulus-response curve was increased in clinically affected familial amyotrophic lateral sclerosis (FALS) and sporadic ALS (SALS) patients when compared to asymptomatic superoxide dismutase-1 (SOD-1) mutation carriers and normal controls (ANOVA, P < 0.05). The error bars represent standard error of the mean.

RMT, defined as the unconditioned stimulus intensity required to produce and maintain a target MEP response, was not significantly different between the groups (symptomatic FALS 59.2 ± 5.1%; sporadic ALS 57.9 ± 1.6%; asymptomatic SOD-1 mutation carriers, 59.4 ± 2.2%; controls, 60.6 ± 1.3%). However, the RMT was significantly reduced in sporadic ALS patients with limb-onset disease (sporadic ALS limb-onset 56.1 ± 1.7%; controls, 60.6 ± 1.3%, P < 0.05). Furthermore, the RMT in one pre-symptomatic patient was reduced when compared with normal controls (case #1, 56%, 95% confidence interval 58–63.2%), while it was at the lower limit of normal in a second pre-symptomatic patient (case #2, 58%). In the third pre-symptomatic patient the RMT was comparable to that evident in normal controls (case #3, 60%).

Changes in the CSP, defined as a period of electrical silence that interferes with ongoing EMG activity following an MEP in a contracting muscle, was also measured. In the present series, the increase in the CSP recruitment curve were non-linear in FALS patients similar to the relationship established previously for control subjects (Vucic et al., 2006). The mean CSP duration increased from 0 to 199.6 ± 6.2 ms in affected FALS and from 0 to 208.9 ± 4.2 ms in asymptomatic SOD-1 mutation carriers as stimulus intensity increased from 60 to 150% RMT, and was not significantly different when compared to controls (0 to 209.0 ± 4.7 ms). However in three pre-symptomatic SOD-1 mutation carriers, CSP duration was reduced compared to controls and other asymptomatic SOD-1 mutation carriers, but comparable to sporadic ALS patients, prior to the development of clinical features of ALS (case #1 CSP duration, 0–180 ms; case #2, 0–192 ms; case #3, 0–141 ms; sporadic ALS 0–177 ± 8.5 ms, 95% confidence interval normal controls 199.6–218.4 ms). In sporadic ALS patients, the mean CSP duration increased from 0 to 176.7 ± 8.5 ms, and was significantly reduced when compared to controls (P < 0.01).

Discussion

The present longitudinal studies utilize novel threshold tracking TMS techniques to explore cortical excitability in familial SOD-1 positive ALS and sporadic ALS patients, and have established the presence of cortical hyperexcitability in both symptomatic FALS and sporadic ALS patients. Reduction in short-interval intracortical inhibition, associated with increases in intracortical facilitation and cortical stimulus-response curve gradient were all indicative of cortical hyperexcitability. In contrast, in asymptomatic SOD-1 mutation carriers cortical excitability was normal. However, an increase in cortical excitability in three pre-symptomatic SOD-1 mutation carriers preceded the development of clinical features of ALS. Together, these findings confirm that cortical hyperexcitability develops prior to clinical symptoms in FALS.

Where does ALS begin?

Although the findings in the present study do not directly support a pathogenic mechanism in ALS, namely excitotoxicity, they are compatible with the ‘dying forward’ hypothesis that proposes that the initial abnormality in ALS occurs within the corticomotorneurons or the motor cortex, with anterograde excitotoxicity underlying subsequent anterior horn cell degeneration. However, given that formal needle EMG was not performed at the time of cortical excitability studies, it remains possible that subtle, undetected changes in anterior horn cells may have preceded or concurrently developed with cortical hyperexcitability. The findings in the present study, however, are in keeping with flumazenil PET studies that demonstrated cortical abnormalities in two pre-symptomatic SOD1 patients (Turner et al., 2005a), and with studies in a SOD-1 (G93A) mouse model suggesting that degeneration of spinal cord motor neurons is secondary to dysfunction within CNS motor pathways, i.e. within the primary motor cortex, corticospinal tract and bulbospinal pathways (Browne et al., 2006).

Evidence for cortical hyperexcitability in the present study is provided by a significant reduction of SICI in both clinically affected SOD-1 and sporadic ALS patients, and in longitudinal studies in three pre-symptomatic SOD-1 mutation carriers who developed ALS within months of developing cortical hyperexcitability. Interestingly, although case #3 exhibited SICI prior to and at the time of development of ALS, there was a marked reduction of SICI 8 months prior to development of ALS, illustrating the point that a relative reduction in SICI may be indicative of the development of cortical hyperexcitability. The mechanisms underlying reduction of SICI are complex and involve both degeneration of GABA secreting inhibitory cortical interneurons, both in the primary motor cortex and extramotor cortical areas, and glutamate-mediated down-regulation of SICI (Nihei et al., 1993; Stefan et al., 2001; Maekawa et al., 2004). In addition, dysfunction of GABAergic synapses, whereby they become excitatory, may be another mechanism underlying the reduction of SICI (Ben-Ari et al., 2004). Specifically, in the neonatal brain, GABAergic synapses are excitatory. Conversion from excitatory to inhibitory transmission is dependent on the establishment of the chloride electrochemical gradient by expression of a specific K+/Cl co transporter (KCC2) on post-synaptic membranes (Ben-Ari et al., 2004). Studies assessing the function of KCC2 in FALS patients may yet confirm this potential mechanism of SICI reduction.

Although the reduction of SICI in the present study was interpreted as reflecting cortical hyperexcitability, other interpretations remain possible. Given the fact that SICI is reduced in other neurological and psychiatric disorders (Ridding et al., 1995a, b; Abbruzzese et al., 1997; Chen et al., 1997; Ziemann et al., 1997a, b; Hanajima et al., 1998; Fitzgerald et al., 2002; Zanette et al., 2002; Rosenkranz et al., 2005) it could be interpreted that SICI reduction reflects a compensatory phenomenon related to brain injury or a ‘bystander’ phenomena.

Alternatively, it may continue to be valid to interpret the reduction of SICI in other neurological and psychiatric disorders as cortical hyperexcitability, with the absence of motor neuron degeneration in these other conditions explained by a greater resistance of the motor neurons to the stresses of cortical hyperexcitability. Specifically, a number of cell-specific molecular features of motor neurons in ALS patients render them vulnerable to cortical hyperexcitability, mediated via glutamate excitotoxicity. First, motor neurons in ALS patients preferentially express an AMPA receptor lacking the functional GluR2 subunit, thereby rendering the motor neurons more permeable to Ca2+ and thereby to activation of Ca2+-dependent enzymatic pathways that mediate neuronal death (Van Damme et al., 2002, 2005; Kawahara et al., 2004; Kwak and Kawahara, 2005). Secondly, motor neurons vulnerable to degeneration lack the intracellular expression of Ca2+-binding proteins parvalbumin and calbindin D28k, required to buffer intracellular Ca2+ (Ince et al., 1993; Alexianu et al., 1994). More recently, increased expression of the inositol 1,4,5-triphosphate receptor 2 (ITPR2) gene was reported in ALS (van Es et al., 2007). ITPR2 is involved in glutamate-mediated neurotransmission, whereby stimulation of glutamate receptors results in binding of inositol 1,4,5-triphosphate to ITPR2, which subsequently increases intracellular calcium (Choe and Ehrlich, 2006; van Es et al., 2007). Aberrant activity of ITPR2 results in higher intracellular concentration of Ca2+ and ultimately apoptosis (Gutstein and Marks, 1997).

Another possible explanation for the SICI findings in the pre-symptomatic SOD-1 mutation carriers is random fluctuation with a regression to the mean. Although the findings depicted in Fig. 3 would seem to argue against random fluctuations, a larger cohort of pre-symptomatic SOD-1 mutation carriers would be required to definitively exclude this possibility.

Although there was no significant reduction of SICI in asymptomatic SOD-1 mutation carriers when compared to controls, it remains possible that subtle abnormalities of SICI may have gone undetected. Specifically, given that the expression of SICI is stronger with higher test stimulus intensities which generate larger MEP amplitudes (Daskalakis et al., 2002), possibly due to recruitment of later I-waves that are susceptible to inhibition by the conditioning stimulus (Di Lazzaro et al., 1998), it remains possible that selecting a tracking target of 0.2 mV may have been suboptimal for identifying subtle deficient SICI in the asymptomatic SOD-1 mutation carriers. In addition, measurement of SICI using a conditioning stimulus intensity of 70% RMT, which usually results in maximum SICI (Kujirai et al., 1993), may have resulted in saturated inhibition (floor effect). In order to detect subtle reduction of SICI, lower conditioning stimuli intensities may be required (MacKinnon et al., 2005). Given the fact that conditioning stimulus intensity and the tracking target MEP amplitude were not varied in the present study, subtle deficiencies of SICI in asymtpomatic SOD-1 mutation carriers may have passed undetected. Interestingly, in two asymptomatic SOD-1 mutation carriers (case #6 and case#10), who continue to remain asymptomatic, SICI appeared to be reduced. While this finding may raise issues regarding the specificity of the link between cortical hyperexcitability and neurodegeneration, longer-term follow-up will be essential to clarify whether these subjects develop the clinical features of ALS.

Previous cortical excitability studies in familial SOD-1 positive ALS patients have reported contradictory findings (Weber et al., 2000; Turner et al., 2005b; Stewart et al., 2006). While normal cortical excitability was reported in patients with the D90A (Weber et al., 2000; Turner et al., 2005b) and I113T mutations (Stewart et al., 2006), abnormalities of corticomotoneuronal function have been recently reported in patients with the A4V mutations (Stewart et al., 2006). The present findings establish the presence of cortical hyperexcitability in patients with the V148G, E100G and I113T SOD-1 mutations, in keeping with the findings in the A4V mutation. The homozygous D90A mutation is associated with a slowly progressive phenotype, probably related to co-inheritance of protective or modifying genes with the D90A alleles (Simpson and Al-Chalabi, 2006). The inheritance of these protective or modifying genes may have been responsible for normal cortical excitability in the D90A SOD-1 positive patients.

In addition to reduction of SICI, ICF was increased in clinically affected FALS, sporadic ALS and the three pre-symptomatic FALS subjects. The mechanisms underlying this increase of ICF have not been elucidated. Studies describing the relationship of descending corticospinal tract volleys to ICF have established that the conditioning stimulus does not alter the I-wave amplitude or number in the descending corticospinal tract volley (Di Lazzaro et al., 2006). Rather, it was postulated that ICF occurred as a result of either (i) the conditioning stimulus increasing excitability at the spinal cord level, such that motoneurons are more likely to be activated by a descending corticospinal volley or (ii) the conditioning stimulus altering the composition of the descending corticospinal volley such that a greater proportion of descending corticospinal activity was destined for the target motor neurons. Either of these two mechanisms could have been responsible for the increased ICF in evident in the present series. Interestingly the ICF in normal controls, was less than that established by the constant stimulus method (Kujirai et al., 1993). The reasons for this discrepancy remain unclear, and may be related to technical factors, including less variability in the MEP amplitude with the threshold tracking technique.

In conjunction with reduction of SICI and increase in ICF, the resting motor threshold was reduced in sporadic ALS patients, although this reduction was only significant in patients with limb onset-disease, as previously established (Vucic and Kiernan, 2006). Interestingly, although the RMT was reduced in clinically affected FALS patients, this reduction was not significant. While it remains possible that cortical hyperexcitability is less prominent in clinically affected FALS patients, more likely the lack of a significant difference may reflect the smaller sample size in the clinically affected FALS group.

In conclusion, the results from the present study complement previous findings in sporadic ALS patients, where cortical hyperexcitability developed as an early feature (Eisen and Weber, 2000; Vucic and Kiernan, 2006). Ultimately, if earlier cortical abnormalities could be reliably identified, institution of neuroprotective agents at that stage, including anti-glutamatergic agents such as riluzole, may result in a slower disease progression and prolonged survival.

Acknowledgements

S.V. was awarded the Clinical Fellowship of Motor Neuron Disease Research Institute of Australia, with funding provided by the Motor Neuron Disease Association of NSW. Grant support was also received from the NSW Ministry for Science and Medical Research Spinal Cord Injury & Related Neurological Conditions Research Grant Program and from the National Health and Medical Research Council of Australia. The authors gratefully acknowledge the assistance of Roger Pamphlett with the muscle biopsy findings. Input from Professor Hugh Bostock, particularly with threshold tracking software, is again gratefully acknowledged.

Footnotes

  • Abbreviations:
    Abbreviations:
    ALSFRS-R
    amyotrophic lateral sclerosis function rating scale revised
    ANOVA
    analysis of variance
    CMAP
    compound muscle action potentials
    CMCT
    central motor conduction time
    CSP
    cortical silent period
    FALS
    familial amyotrophic lateral sclerosis
    ICF
    intracortical facilitation
    ISI
    interstimulus interval
    MEP
    motor evoked potential
    MRC
    Medical Research Council
    NI
    neurophysiological index
    RMT
    resting motor threshold
    SICI
    short-interval intracortical inhibition
    SOD-1
    copper/zinc superoxide-dismutase-1 gene
    TMS
    transcranial magnetic stimulation

References

View Abstract