Introduction
Amyotrophic lateral sclerosis (ALS) is a degenerative disease of the nervous system that primarily affects upper and lower motor neurons. Reference Siddique, Siddique, Adam, Ardinger and Pagon1 The diagnosis of ALS is supported by clinical findings and electrophysiological studies (revised El Escorial criteria and Awaji-Shima criteria). Reference Carvalho and Swash2–Reference De Carvalho, Dengler, Eisen and at4 Approximately 90% of ALS cases are sporadic. Reference Millecamps and V.A.P.B.5 ALS type 8 (ALS8) is one of the familial ALS types, first described in Brazil in 2004 in a large family of Portuguese–Brazilian ancestry, with a founding event 23 generations ago (Portuguese colonization in Brazil). ALS8 is linked to a mutation in the vesicle-associated membrane protein/synaptobrevin-associated membrane protein B (VAPB) gene (20q13.33, c.166C>T, p. Pro56Ser) Reference Nishimura, Mitne-Neto and Silva6–Reference Nishimura, Al-Chalabi and Zatz7 and has autosomal dominant inheritance, similar presentation in both sexes, adult onset, and slow progression. The phenotypes described in Brazilian families were spinal muscular atrophy, typical ALS symptoms, and atypical ALS symptoms (lower motor neuron signs with predominantly proximal weakness, postural tremor preceding motor neuron symptoms and signs, severe cramps, abdominal weakness, kyphoscoliosis, and laryngeal spasms). Reference Nishimura, Al-Chalabi and Zatz7–Reference Nishimura, Mitne-Neto and Silva8 In 2010, Funke described the first German patient with the P56S mutation who did not have Brazilian or Portuguese ancestry but exhibited clinical features similar to those described in Brazil. Reference Funke, Esserb and Krüttgen9 Although the clinical features of ALS8 are unique and apparently recognizable, this disease is accepted as a solely familial form of ALS, reported primarily in Brazilian and Japanese patients. Reference Siddique, Siddique, Adam, Ardinger and Pagon1
Sensory abnormalities are found in 7 to 91% of ALS patients, depending on the method used (clinical evaluation, electrophysiological study, sural nerve, or skin biopsy). Reference Hammad, Silva, Glass, Sladky and Benatar10–Reference Mondelli, Rossi, Passero and Guazzi13 Skin biopsy enables quantification of intraepidermal nerve fiber (IENF) density in peripheral neuropathies such as small fiber neuropathy, hereditary sensory and motor neuropathy, and sensory and autonomic neuropathy, Reference Lauria, Cornblath and Johansson14–Reference Lauria, Hsiehb and Johanssonc19 as IENFs are sensory and autonomic nerve terminals. Reference Weis, Katona and Mulle-Newen11,Reference Vicková-Moravcová, Bednarík and Dusek20–Reference McCarthy, Hsieh and Stocks21 The skin biopsy technique is easy to perform and identifies IENFs with immunohistochemistry using an antibody against the protein gene product 9.5 (PGP 9.5). Reference McCarthy, Hsieh and Stocks21 In 2011, Weis described, for the first time, lower IENF density in patients with ALS than in controls, suggesting the presence of distal axonal alterations in the early stages of ALS. Reference Weis, Katona and Mulle-Newen11
The objective of this study was to investigate the sensory abnormalities detected by clinical evaluation, electrophysiological study, and skin biopsy in patients with ALS8 compared to those in patients with sporadic ALS (sALS) and healthy controls.
Methods
The project was approved by the Ethics Committee of the Federal University of São Paulo (protocol -149/07) and carried out in accordance with the Code of Ethics of the Declaration of Helsinki. All participants signed an informed consent form, agreeing to the investigation and the publication of anonymised results in scientific journals.
Patients
Forty consecutive ALS8 patients (ALS8 group) and 10 sALS patients (sALS group) were selected from the Neuromuscular Diseases Service. Twelve healthy individuals formed the control group, which comprised doctors, resident doctors, undergraduate/graduate students, and university employees. The selected sALS patients and controls were matched with ALS8 patients by sex and age.
The ALS8 diagnosis was established by family history, neurological evaluation, electrophysiological study, and the presence of the p. Pro56Ser mutation in exon 2 of the VAPB gene. Reference Nishimura, Al-Chalabi and Zatz7 The sALS diagnosis was based on clinical and electrophysiological signs of ALS in agreement with the Revised El Escorial and Awaji-Shima criteria. Reference Carvalho and Swash2–Reference De Carvalho, Dengler, Eisen and at4 We included sALS patients with: possible, probable, or definite disease; no family history of any neurological disease; and negative genetic testing results for ALS8. The exclusion criteria established for all groups were as follows: A) alcoholism or use of illicit drugs; B) exposure to toxic agents such as solvents, mercury, lead, or gold; C) diagnosis of diabetes mellitus or glucose intolerance, hypothyroidism, cancer, or HIV/AIDS; D) neuropathies or peripheral vascular disease; and E) immobility or cachexia.
Clinical and Neurophysiological Evaluation
Sensory evaluation included testing for pain, touch, temperature, and vibration sense. For the pain evaluation, we used a nonreusable pin, starting from the proximal to distal parts of the body (hands and feet). Tactile perception was tested with a piece of dry cotton in the same way that pain was evaluated. Thermal sense was tested with a cold water tube (4°C). The evaluation of the vibration sense was made with a 128-Hz tuning fork placed over the bone prominences of the hallux, the lateral malleolus, the terminal phalanx of the thumb, and the styloid apophysis of the radius bilaterally. The neurophysiological examination included bilateral needle electromyography (EMG) and nerve conduction studies, with sensory sural nerve action potential amplitude and conduction velocity recordings.
Skin Biopsy
A skin biopsy was performed for all patients and controls using a 3-mm punch (Kolplast®, Brazil) after local anesthesia with 2-ml 2% lidocaine (Aristo-Brazil® Chemical Industries). The biopsy was performed at the distal portion of the right leg, 8 cm above the external malleolus, within the sural nerve territory, in a region free of recent bruises or atrophic/hypertrophic scars. Skin fragments were immediately fixed in Zamboni solution and processed according to previously established protocols. Reference Vicková-Moravcová, Bednarík and Dusek20,Reference Sommer22,Reference Sommer and Lauria23 Five sections of 15 mm were obtained from each biopsy with a cryostat at −25°C (IEC-Minotome Plus, GMI, USA). The sections were fixed on electrically charged slides (Biogen®, USA) and incubated with the primary polyclonal PGP 9.5 antibody (Chemicon AB1761, Sigma–Aldrich/MERCK, Germany). After exposure to the secondary antibody, a rabbit IgG conjugated to fluorescein (F0205 DAKO, USA), a mounting medium for fluorescence was used (Vectashield Mounting Medium is Fluorescence H1000, Vector, USA). The sections were observed under a conventional epifluorescence microscope (BX60 Olympus, Japan) at 200× and 400× magnifications. The quantitative analysis of IENF per millimeter of epidermis (linear density) was obtained by dividing the total number of fibers crossing the basal membrane by the length of epidermis analyzed (in millimeters), as previously described. Reference Hsiung-Fei, To-Jung and Whei-Min24 We used the cutoff point of 3.3 fibers per millimeter of skin for the normal linear density of IENF, and any value below this cutoff was considered abnormal. Reference Moura, Oliveira, Zanoteli, Cardoso, Schmidt and Gabbai25
Statistical Analysis
The sample size was calculated based on a 10% frequency of IENF density decrease in the general population and a 79% decrease in the sALS population. Reference Weis, Katona and Mulle-Newen11 To detect a difference of at least 69% in the frequency of IENF density decrease between the ALS8 and control groups, with a detection power of 80% and an alpha level of 0.05, at least eight ALS patients and eight control subjects were needed. Then, the 40 ALS8 patients were compared with 10 sALS patients. A subset of 14 ALS8 patients was selected (based on similar sex and age range to the control group) and compared with the 12 controls. Data were tabulated and expressed as central tendency and dispersion values. Categorical data were described as absolute and relative frequencies (n(%)). Noncategorical data were expressed as the mean ± standard deviations or median (quartiles 25%–75%).To analyze differences between/among groups, we used the chi-square test for categorical data, unpaired t test or Anova for parametric data, and Mann–Whitney test or Kruskall–Wallis for nonparametric data. Pearson’s or Spearman’s correlation analyses were used for parametric or nonparametric data, respectively. To better verify the relationship between IENF density and other variables that showed a significant correlation, multiple regression analysis was performed. We used GraphPad Prism Software (version 5.03) for statistical analysis, and a p value <0.05 was considered significant.
Results
At the time of the biopsy, the sex and mean age of the ALS8 patients did not differ from those of sALS patients or controls (Tables 1 and 2). However, ALS8 patients were significantly younger at the onset of muscle weakness than patients with sALS and had a significantly longer disease progression (Table 1).
Clinical and demographic data: ALS8 and sporadic ALS
ALS8: amyotrophic lateral sclerosis type 8; sALS: sporadic amyotrophic lateral sclerosis; M: male; F: female; n: number; X 2: chi-square test; ns: not significant; IENF: intraepidermal nerve fiber; mm: millimeter; BIPAP: bilevel positive airway pressure. * Mann–Whitney test. ** Unpaired t test.
Demographic and clinical data: ALS8 and control groups
ALS8: amyotrophic lateral sclerosis type 8; M: male; F: female; n: number; X 2: chi-square test; ns: not significant; IENF: intraepidermal nerve fiber.
Fourteen of the ALS8 patients (35%) and three sALS patients (30%) had sensory abnormalities in the neurological examination, with no significant difference between these groups (Table 1). The control group subjects had no sensory complaints or sensory abnormalities in the neurological examination. In both ALS groups, vibration sense abnormalities were more frequent than thermal or pinprick sensory abnormalities, but there was no significant difference between the two groups regarding the frequency of these abnormalities (Table 1). All ALS patients underwent nerve conduction studies that showed normal sensory sural nerve action potential amplitudes and sensory conduction velocities.
There was no significant difference in the number of patients among patient subgroups defined by Revised El Escorial classification criteria (Table 1). There was no significant difference in the number of patients using BIPAP (bilevel positive airway pressure) between the two ALS groups (Table 1).
IENFs were observed in the biopsies of all patients and controls (Figure 1) and were located in the epidermis and reticular/papillary dermis. There was no significant difference between the sALS and ALS8 groups in the density of IENFs or in the number of patients with an IENF density under the cutoff point of 3.3/mm (Table 1). There was no significant difference in the IENF density of patients using BIPAP between the two ALS groups (Table 1). However, the ALS8 group presented an IENF density significantly lower than that of the controls and a greater number of patients with an IENF density under the cutoff point of 3.3/mm (Table 2). There was no correlation between epidermal thickness and IENF density in the three groups (Pearson’s test, p: ns).
Longitudinal frozen sections from skin biopsy samples immunostained with polyclonal PGP9.5 antibody (immunofluorescence, Olympus BX60 microscope). Left figure: arrow indicates intraepidermal nerve fibres (IENF) from a control subject skin biopsy (400x). Right figure: indicates reduced density of IENF from an ALS8 patient skin biopsy (200x).
ALS8 Group
The IENF density was significantly lower in female patients than in male patients (1.9 ± 1.3 vs. 3.2 ± 2, p < 0.05, unpaired t test). Furthermore, IENF density was significantly lower in those with vibration sense abnormalities than in those with normal sensory results (1.3 ± 0.6/mm vs. 2.9 ± 1.9/mm, unpaired t test, p < 0.05). There was no significant correlation between IENF density and age at biopsy, age at weakness onset or disease duration (p: ns, Spearman’s correlation analysis). There was no difference in IENF density among the El Escorial subgroups (p: ns; Kruskal–Wallis) or between the following subgroups: patients with or without sensory complaints or thermal/pain sense abnormalities (unpaired t test; p: ns). In the ALS8 group (n = 40), there were no differences in the IENF density between the groups with/without bulbar dysfunction (n = 9/31; 3.2 ± 0.7/mm vs. 2.3 ± 0.2/mm; p ns, unpaired t test), or with independent walk/assisted walk/wheelchair (n = 17/7/15; 2.4 ± 1.5/mm vs. 4.1 ± 2.6/mm vs. 2.2 ± 1.5/mm, p ns, Anova). One patient was restricted to the bed and had an IENF density of 0.3/mm. Multiple regression analysis was performed to verify the relationship between IENF density and the variables that showed a significant correlation (sex and reduced vibration sense) beyond age and reduced thermal/pain sense (Table 3). This analysis confirmed that female sex and reduced vibration sense could predict a reduced IENF density in this group. Females presented a lower IENF density than that of males (regression coefficient/standard deviation: −1.2/0.5, R 2 0.2, p < 0.05), and patients with reduced vibration sensation presented a lower IENF density than that of patients with normal vibratory sensation (regression coefficient/standard deviation: −1.5/0.5, R 2 0.2, p < 0.01).
Multiple regression estimations for IENF density in the ALS8 group
ALS8: amyotrophic lateral sclerosis type 8; IENF: intraepidermal nerve fiber. *p < 0.1. **p < 0.05. ***p < 0.01.
Discussion
Sensory complaints and sensory abnormalities in clinical evaluation were present in approximately one-third of ALS patients in both the sALS and ALS8 groups, which is similar to the 32% frequency found by Hammad et al., 2007. Reference Hammad, Silva, Glass, Sladky and Benatar10 An IENF density below 3.3/mm was found in two-thirds of the patients in both groups, which is in contrast to the 79% reported in sALS by Weiss et al., 2011. Reference Weis, Katona and Mulle-Newen11 The reduced IENF density in ALS patients suggests that the pathophysiological process of injury is not restricted to motor neurons but could involve sensory neurons and/or their axonal projections. Reference Hammad, Silva, Glass, Sladky and Benatar10
The IENF density in the control group was similar to that previously described in Brazilian controls of 5.2 ± 1.8 Reference Moura, Oliveira, Zanoteli, Cardoso, Schmidt and Gabbai25 ; other groups reported IENF densities from 11.3 ± 3 to 17.8 ± 3. Reference Vicková-Moravcová, Bednarík and Dusek20,Reference McCarthy, Hsieh and Stocks21,Reference Lauria, Holland and Hauer26 Moura et al., 2004, emphasize that this variability could be related to different techniques, slice thicknesses, and microscopy methods. Reference Moura, Oliveira, Zanoteli, Cardoso, Schmidt and Gabbai25 There is controversy about decreases in IENF density with age, especially in individuals over 60 years of age. Reference Vicková-Moravcová, Bednarík and Dusek20,Reference Sommer and Lauria23–Reference Lauria, Holland and Hauer26 McArthur et al., 1998, reported a peak in IENF density in the second decade of life and a decrease in the third decade, followed by a stable density. Reference McArthur, Stocks, Hauer, Cornblath and Griffin27 The results of studies by McCarthy et al., 1995, and Lauria et al., 1999, did not support this finding. Reference McCarthy, Hsieh and Stocks21,Reference Lauria, Holland and Hauer26 Others reported a decrease of 0.6 to 1.8 fibers/mm per decade in healthy subjects. Reference Goransson, Mellgren, Lindal and Omdal28 As ALS patients are usually older, they could be in the period of relative IENF density stability. Reference McArthur, Stocks, Hauer, Cornblath and Griffin27 In sALS, a lower IENF density could represent the synergy of two processes: the physiological loss and the loss associated with ALS. Reference Goransson, Mellgren, Lindal and Omdal28–Reference El-Salem, Khassawneh, Alrefai, Dwairy and Rawashdeh31
Weis et al., 2011, conducted the first study that also evaluated IENF density by skin biopsy in patients with sALS. Reference Weis, Katona and Mulle-Newen11 Similar to our ALS8 data, their data showed a lower IENF density in ALS8 patients than in controls, but they also found an inverse correlation with increasing age that was not found in our study. Reference Weis, Katona and Mulle-Newen11 A lower IENF density in patients with sensory complaints was also reported. Reference Weis, Katona and Mulle-Newen11 Interestingly, this lower IENF density was not correlated with objective sensory abnormalities after both clinical evaluation and electrophysiological studies. Reference Weis, Katona and Mulle-Newen11
In our ALS8 group, there was a lower IENF density among women than among men, for which we found no explanation except for a predisposition for IENF disease in females. Studies by McCarthy et al., 1995, and Lauria et al., 1999, have not shown differences in IENF density between sexes. Reference McCarthy, Hsieh and Stocks21,Reference Lauria, Holland and Hauer26 Goransson et al., 2004 and Umapathi et al., 2006 found a slightly higher IENF density in healthy women than in healthy men. Reference Goransson, Mellgren, Lindal and Omdal28,Reference Umapathi, Tan, Tan and Chan32 Dalla Bella et al., 2016, reported that IENF loss did not correlate with the onset, phenotype, course, or disease severity of their ALS patients. Reference Dalla Bella, Lombardi and Porretta-Serapiglia33 The peculiar aspect of the reduced IENF density among women in our ALS8 sample and the absence of correlations between age/disease duration and IENF density raised the question of whether sensory abnormalities are a particular feature of ALS8.
A lower IENF density in ALS8 patients with abnormalities in vibration sense was previously described by Weis et al., 2011, in sALS patients and by Lauria et al., 1999, in patients with large fiber neuropathies. Reference Weis, Katona and Mulle-Newen11,Reference Lauria, Holland and Hauer26 This finding may be related to the concomitant involvement of multiple systems. Reference Weis, Katona and Mulle-Newen11,Reference Lauria, Cornblath and Johansson14 Previous studies on ALS have shown more abnormalities in large-calibre myelinated fibers, which could result from a neuronopathy or a dying-back axonopathy, such as lesions in Clarke’s nuclei and in posterior columns of the spinal cord. Reference Hammad, Silva, Glass, Sladky and Benatar10,Reference Mondelli, Rossi, Passero and Guazzi13,Reference Lauria, Cornblath and Johansson14 Hammad et al., 2007, showed that large-calibre myelinated fibers were reduced in 73% of sALS patients, and our findings suggest that the same could occur in ALS8 patients. Reference Hammad, Silva, Glass, Sladky and Benatar10 The findings of normal sensory sural nerve action potential amplitude and sensory conduction velocity in our sample are in agreement with the findings of low frequency of abnormalities reported previously in sporadic and familial ALS by Pugdahl et al., 2007. Reference Pugdahl, Fuglsang-Frederiksen and de Carvalho34
The sensory abnormalities and reduction in IENF density in the sALS and ALS8 groups were similar, despite different ages at disease onset; that is, the ALS8 group had longer disease progression than that of the ALS group. Future studies could address the reason for this similar degree of reduction in IENF density, despite different disease durations. It remains to be determined whether this discrepancy results from the activity of different processes of IENF lesions in each state (ALS8 vs. sALS) or different disease severities between ALS8 and sALS. This last hypothesis agrees with the finding of slowly progressive defects in the animal model of ALS8, the VAPB/ALS8 knock-in mice. Reference Larroquette, Seto and Gaub35 Additionally, a broader sensory evaluation with quantitative sensory testing and peripheral nerve biopsy could provide more information about sensory changes in ALS8 patients.
In conclusion, this study showed that ALS8 patients presented with sensory complaints, abnormalities in sensory evaluation, and reductions in IENF density. These aspects can be considered in the clinical characterization of this subtype of ALS.
Acknowledgements
We acknowledge all patients and families who agreed to participate in this study.
Authors’ Disclosures
MMN was a PhD student at the time that he contributed to the study and is currently an employee of Grupo Fleury, a private diagnostic company in Brazil.
Statement of Authorship
All authors revised the manuscript for intellectual content, gave final approval of the version to be published, and agreed to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.
MCMC, LHDC, and LMA: neurological evaluation, skin biopsies, intraepidermal nerve fiber density, and manuscript draft.
MMN and MZ: molecular study.
ASBO and HCAS: study design and conceptualization.
Funding
This work was supported by the Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPQ), Coordenação de Aperfeiçoamento de Pessoal de Nível Superior – Brasil (CAPES) under Grant number 001.
Conflict of Interest
The authors report no conflict of interest.
