©2016 Dustri-Verlag Dr. K. Feistle ISSN 0722-5091

DOI 10.5414/NP300889e-pub: November 20, 2015

Received June 3, 2015; accepted in revised form July 9, 2015

Correspondence to Elisabetta Tasca, PhD Neuromuscular Laboratory, Fondazione Ospedale San Camillo IRCCS, via Alberoni 70, 30100 Lido Venice, Italy elisabetta.tasca@ospedalesancamillo.net

Key wordsALS – miRNA – bio-markers – myostatin – muscle atrophy

Circulating microRNAs as biomarkers of muscle differentiation and atrophy in ALSElisabetta Tasca, Valentina Pegoraro, Antonio Merico, and Corrado Angelini

Fondazione Ospedale San Camillo IRCCS, Lido Venice, Italy

Abstract. Aims: The identification of circulating biomarkers is needed to facili-tate diagnosis and prognosis of amyotrophic lateral sclerosis (ALS) and to offer indica-tors of therapeutic response in clinical tri-als. We aimed to investigate the levels of muscle-specific microRNAs in serum of ALS patients subdivided according to bul-bar or spinal onset. Methods: In 14 ALS pa-tients (10 spinal, 4 bulbar) we measured the serum levels of muscle-specific miR-206, miR-1, miR-133a/b, miR-27a, and the ex-pression of myostatin and follistatin, which are negative regulators of muscle growth. Morphometric analysis of muscle fiber size was used to correlate muscle atrophy with biochemical-molecular parameters. Results: In ALS patients the expression of miR-206 and miR-133 was significantly increased and that of miR-27a was significantly reduced as compared to controls, and also between spi-nal vs. bulbar ALS. Myostatin/follistatin ratio was significantly higher in ALS than in con-trols and in bulbar versus spinal ALS. Bulbar ALS patients present higher degree of mus-cle atrophy than spinal ALS, as documented by our muscle fiber morphometric analysis. Conclusions: Muscle mass regulators are particularly down-expressed in bulbar ALS, suggesting a more rapid and diffuse atrophic process. These biomarkers may be considered as useful biochemical and molecular indica-tors involved both in neuromuscular junction maintenance and reinnervation process.

Introduction

Amyotrophic lateral sclerosis (ALS) is a neurodegenerative disease, caused by primary degeneration of anterior horn neurons. About two-thirds of ALS patients have a spinal form of the disease (spinal ALS), which involves the limbs at onset. Conversely, patients with bulbar onset ALS (bulbar ALS) usually pres-ent with dysarthria and dysphagia and rapid evolution [1, 2]. The weakness of the limbs

develops in most cases within 1 – 2 years. While in spinal ALS death occurs on average 5 years after onset [1], bulbar ALS is associ-ated with a more rapid decline [2].

While the etiology of isolated ALS re-mains largely unknown, the death of motor neurons occurs as the result of a complex ge-netic-environmental interaction and activa-tion of pathogenetic mechanisms including oxidative stress, dysregulation of RNA pro-cessing, mitochondrial dysfunction, protein aggregation, excitotoxicity, abnormal axonal transport, and inflammation (as reviewed in [3, 4]). The pathogenetic processes involve motor neurons and also non-neuronal cells, such as astrocytes, microglia, and possibly skeletal muscle. Muscle atrophy in ALS is associated with a perturbation in the mo-lecular network controlling the consequent muscle organization and function, including autophagy, mitochondrial homeostasis, and muscle regeneration [3, 4]. Therefore, struc-tural and metabolic alterations in skeletal muscle may exacerbate the disease outcome.

MicroRNAs (miRNAs) are small non-coding RNAs that modulate a wide range of biological functions [5]. MiRNAs regulate post-transcriptional mRNA expression, typi-cally by binding to the 3’-untranslated region of the mRNA sequence; their upregulation results in a translational repression and gene silencing and in a decreased expression of the corresponding protein product.

MiRNA are frequently altered in sev-eral pathological conditions, including can-cer, cardiovascular and metabolic disease, Parkinson’s disease and ALS. MiRNAs are known to be secreted by various cell types, and, unlike most mRNAs, they are markedly stable in circulating body fluids due to pro-tein protection from ribonucleases. Because

Clinical Neuropathology, DOI 10.5414/NP300889

Tasca, Pegoraro, Merico, and Angelini 2

of these properties, miRNAs have gained attention for their potential as minimally in-vasive and cost-effective disease biomarkers. They can be reliably detected even at low concentration and used not only as markers of disease, but also of disease staging, and possibly to measure the efficacy of novel drugs.

In ALS, the role of biomarkers related to exocitotoxicity, oxidative stress, inflamma-tion, angiogenesis, and neurodegeneration has been investigated [6]. In particular, miR-27a, miR-155, miR-142-5p, miR-223, and miR532-3p were found to be highly expressed in ALS patients, and miR-27b, miR-146a and miR-532-3p were commonly elevated in ce-rebrospinal fluid samples of ALS patients [7]. While miR-155 promotes pro-inflammatory pathways, other miRNAs are more expressed in mature muscle fibers and proliferating myo-blasts. These miRNAs (miR-206, miR-133a, miR-133b, miR-1) are called “myo-miRNA” and are considered markers of muscle regen-eration, myogenesis, fiber type differentiation, degeneration, and injury, and might represent indicators of residual muscle mass consequent to a chronic atrophy of muscle [8, 9].

Besides their role as biomarkers of the disease in ALS, the levels of some miRNA, such as miR-338-3p, were found to be di-rectly correlated with disease duration [10].

It is conceivable that different molecular mechanisms underlie the different clinical manifestations of ALS at onset. In the pres-ent study we compared, in bulbar ALS and spinal ALS, the expression of a panel of se-rum miRNA involved in muscle regeneration and atrophy as well as in inflammatory and in angiogenetic processes, and compared the levels of miRNA-27a with the plasma levels of myostatin and follistatin protein, which are involved in muscle atrophy and hyper-trophy. Furthermore, we compared these molecular-biochemical data with the degree of muscle fiber atrophy obtained by morpho-metric analysis in muscle biopsies of ALS patients.

Materials and methods

Patient cohort

For the purpose of the study we collected serum and plasma samples from 14 sporad-

ic ALS patients (mean age 64 years) and 8 healthy subjects used as controls (mean age 55 years). ALS patients were clinically ex-amined at the IRCCS San Camillo Hospital and diagnosed according to the “El Escorial” criteria [1].

In order to disclose a possible differen-tial expression of myo-miRNAs among ALS patients with different clinical presentation at onset, we subdivided the 14 ALS patients into two groups: 10 patients presenting with spinal ALS at onset (spinal ALS group) and 4 presenting with bulbar ALS at onset (bul-bar ALS group). Their level of physical dis-ability was assessed using the revised ALS functional rating scale (ALS-FRS-R).

Plasma and serum sampling and processing

Written informed consent was obtained from patients and controls to collect periph-eral blood samples. Serum and plasma sam-ples were stored frozen at –80 °C until use.

Muscle fiber morphometric analysis

Twelve of the 14 ALS patients (4 bulbar ALS, 8 spinal ALS) and 7 controls underwent quadriceps femoris muscle biopsy as part of the diagnostic procedures, obtained after written informed consent. The morphometric analysis of muscle fibers was conducted on cross sec-tions stained for hematoxylin-eosin, which have been used to digitalize 5 – 7 non-over-lapping random microscope fields. ImageJ software (v.1.34) was used to measure fiber cross-sectional area, fiber diameter, the coef-ficient of fiber size variability (normal val-ues: 0 – 250 units), the fiber atrophy factor and the fiber hypertrophy factor which have been developed to express the proportion of abnormally small or large fibers in the sec-tion (normal values: 0 – 200 units) [11].

RNA extraction and quantitative real-time PCR

Serum miRNAs were extracted by miR-Neasy extraction kit (Qiagen, Hilden, Ger-many), following the manufacturer’s specifi-

Biomarkers and miRNA in ALS 3

cations. RNA was transcribed using TaqMan MicroRNA Reverse Transcription Kit (Ap-plied Biosystems, Foster City, CA, USA). qRT-PCR was performed using TaqMan Uni-versal Master Mix and specific probes (Ap-plied Biosystems). The expression levels of miRNAs were quantified by real-time PCR using individual miRNA-specific probes (Ap-plied Biosystems), obtained in duplicated ex-periments, and calculated utilizing the ΔΔCt method. Values were normalized to spiked-in miR-39-3p of Caenorhabolis elegans and ex-pressed as “relative expression of control”.

ELISA quantification of myostatin and follistatin

Elisa quantification test was used to de-termine the plasma concentration of myo-statin (K1012, Immunodiagnostik Bens-heim, Germany) and follistatin (ab113319, Abcam Cambridge, MA, USA), according to the manufacturer’s instructions [12]. 20 μL of each plasma were diluted 1 : 10 in sample buffer solution; 200 μL of biotinylated myostatin tracer were added to 200 μL of each diluted sample of standard and control, followed by 100 μL of streptavidin-labeled peroxidase complex. For follistatin analysis, 25 μL of each plasma were diluted 1 : 10 in assay solution. 100 μL of biotinylated follistatin detection antibody were followed by 100 μL of peroxidase-streptavidin solution and 100 μL of tetra-methyl-benzidine substrate. Stop solution was added to stop the reaction. Within 15 minutes, the absorbance was measured at 450 nm using a microtiter plate reader. A 4-parameter algorithm was used to calculate the standard curve against which the concentration of myostatin samples was determined. The follistatin concentration was determined drawing the best-fit straight line though the standard points. The concentration of each protein was obtained in duplicated ex-periments, and the values were expressed as the ratio between myostatin and follistatin, as previously reported [12], since myostatin can be found in the blood in an inactive state when it is bound to follistatin.

Statistical analysis

Values are expressed as mean ± standard deviation. Statistical analysis was performed

by Student’s t-test, where values were con-sidered statistically significant at p < 0.05.

Results

Patients

The ALS patients involved in the study were 6 females and 8 males with an average of 64 years. In the bulbar ALS subgroup we included 4 patients (1 male, 3 females) with similar age at study as the spinal ALS patients who included 3 females and 7 males (Table 1).

The clinical disability score ranged from 26 to 43 points (corresponding to mild-mod-erate severity) and it was not significantly different in the two groups. In particular, all patients were able to walk during the period of our study (1 – 3 years). Despite the rela-tively mild functional disability score, 2 of 4 bulbar ALS patients and 1 of 10 spinal ALS patients underwent percutaneous endoscopic gastrostomy and tracheostomy during study, and another 1 bulbar ALS patient underwent percutaneous endoscopic gastrostomy. After the end of the study, 2 bulbar ALS patients have died at age 71 and 78 years, 4 – 5 years after the onset of the disease (Table 1). To our knowledge, all the other patients are alive at this time.

Morphometric analysis

In ALS patients the coefficient of fiber size variability and the fiber atrophy fac-tor were significantly increased, and the cross sectional area and fiber diameter were significantly reduced as compared to con-trols (Table 2). All muscle biopsies showed prominent muscle fiber atrophy, but the bul-bar ALS group showed a higher degree of fiber atrophy parameters than the spinal ALS group of patients (Figure 1). No muscle tis-sue was available for molecular analysis.

Expression of myo-miRNA

The expression levels of miRNA-206, miRNA-133a, and miRNA-133b resulted to be higher in ALS patients than in con-trols; the difference was significant for miRNA-206, miRNA133b (p < 0.005), and

Tasca, Pegoraro, Merico, and Angelini 4

miRNA133a (p < 0.05) (Table 3) (Figure 2). Interestingly, when comparing the val-ues in the two subgroups of ALS patients, we observed that all 4 myo-miRNA had a higher mean value in spinal ALS than in bulbar ALS, and the difference was signifi-cant for miRNA-206, miRNA-133a, and miRNA-133b (p < 0.005). When compared to the control group, the spinal ALS group showed a significant higher mean value for miRNA-206 and miRNA-133b (p < 0.005).

Expression of inflammatory and angiogenic miRNA

The analysis of the expression levels of inflammatory miRNA showed that miRNA-146a and miRNA-149* were significantly reduced in ALS patients as compared to con-trols (p < 0.005), whereas miRNA-221 and miRNA-155 were not significantly different between patients and controls (Table 3) (Fig-ure 3). Significantly higher levels of miR-NA-155 (p < 0.005) were observed in bulbar ALS patients than in spinal ALS patients. Furthermore, when compared to controls, spinal ALS showed a significantly lower mean value of miRNA-146a, miRNA-155 (p < 0.05), and miRNA-149* (p < 0.005),

and bulbar ALS showed a significantly lower mean value of miRNA-146a and miR-NA-149* (p < 0.05).

Expression of miRNA-27

The analysis of miRNA-27a showed that there was a significant down-regulation in all ALS patients as compared to controls (p < 0.005) (Table 3) (Figure 4). Further-more, when compared to controls, spinal ALS and bulbar ALS showed a significantly lower value (p < 0.05). The levels did not dif-fer in the two clinical groups.

Quantitative protein expression of myostatin and follistatin

The ratio between the concentrations of myostatin/follistatin was on average higher in ALS patients than in controls (Table 3) (Figure 4). When analyzing the two clinical ALS groups, the spinal ALS group showed significantly lower values than the bulbar ALS group (p < 0.005), and the bulbar ALS group showed significantly higher values than controls (p < 0.05).

Table 1. Clinical data.

Phenotype at onset (n. of patients)

Gender Age at study (years)

Functional rating score at study (ALS-FRS-R)

Clinical features; death

Spinal ALS F 65 42 –Spinal ALS F 58 26 Non invasive ventilatorSpinal ALS F 77 28 Non invasive ventilatorSpinal ALS M 73 33 –Spinal ALS M 59 30 PEG, tracheostomySpinal ALS M 67 32 –Spinal ALS M 56 30 –Spinal ALS M 75 26 –Spinal ALS M 54 43 –Spinal ALS M 52 28 –Total Spinal ALS (10) 64 32Bulbar ALS F 75 30 PEG, tracheostomy;

died 4 years after onsetBulbar ALS F 45 28 PEGBulbar ALS F 68 26 –Bulbar ALS M 70 28 PEG, tracheostomy;

died 5 years after onsetTotal Bulbar ALS (4) 64 28Total ALS (14) 64 30

Biomarkers and miRNA in ALS 5

Discussion

Several studies have investigated the use-fulness of circulating biomarkers that might indicate the disease progression in ALS. Among these, fibronectin, interleukin-6, lipid peroxidation, growth factors, and especially angiogenin and neurofilaments [2, 13] were those of main interest because of different

levels in spinal-onset vs. bulbar-onset ALS patients.

As found by other authors [3], our study demonstrated that in serum of ALS patients the levels of miR-206 are higher than in con-trols. In miR-206 knockout mice, miR-206 is the major miRNA that regulates the repair of neuromuscular junctions (NMJ) following nerve injury [14]: indeed, miR-206 over-ex-

Figure 1. Comparison of muscle fiber morphology in biopsies from spinal ALS (A) and bulbar ALS (B). In bulbar ALS there are more atrophic fibers than in spinal ALS. Hematoxylin-eosin stain (200× magnifica-tion). Histograms showing the comparison between the various morphometric parameters of fibers size in total ALS, spinal ALS, bulbar ALS and controls. Bulbar ALS shows higher degree of fiber atrophy than spinal ALS. *p < 0.001; **p < 0.005; §p< 0.05.

Table 2. Muscle fibers morphometric data.

Disease Cross-sectional area (μm2)

Coefficient fiber size variability (units)

Fiber diameter (μm)

Fiber atrophy factor (units)

Fiber hypertrophy factor (units)

Spinal ALS 2,227 ± 901§ 484.5 ± 213.0* 36.2 ± 8.2** 1,099.1 ± 612.3** 28.2 ± 46.6Bulbar ALS 1,805 ± 601** 435.3 ± 91.3** 33.8 ± 6.7** 777.5 ± 117.3** 1.9 ± 3.9Total ALS 2,087 ± 811* 468.4 ± 177.9** 35.4 ± 7.5** 991.9 ± 517.1** 19.5 ± 39.4Controls 3,302 ± 568 161.9 ± 18.9 50.8 ± 4.4 119.5 ± 89.2 0

**p < 0.001,*p < 0.005, §p < 0.05 patient vs. control.

Tasca, Pegoraro, Merico, and Angelini 6

pression promotes the reinnervation process after denervation by regulating the histone deacetylase 4 (HDAC4) and FGF pathway [9, 15, 16]. According to Bruneteau et al. [16] there is a correlation between the potential role of reinnervation ability and ALS disease progression; they investigated also the miR-206-HDAC4 pathway in motor-point muscle biopsies of 8 ALS patients, demonstrating that muscle reinnervation is higher in long-term ALS survivors. In ALS the levels and function of RNA and RNA-binding proteins are abnormal; aggregates of proteins and RNA are detected both in motor neurones

and non-neuronal cells [17]. Furthermore, in almost all cases, some disturbances in neuro-nal cytoskeletal architecture and NMJ func-tion is a primary consequence [17].

One novel and interesting finding from our study, is the significantly higher values of miR-133a, miR-133b, and miR-206 in spinal ALS than in bulbar ALS. This dif-ferential expression may be attributed to the slower progression of spinal ALS as com-pared to bulbar ALS. MiR-206 might delay the onset and progression of ALS by promot-ing the regeneration of NMJ [3, 18]. The ex-pression of miR-206 increases significantly

Figure 2. Histograms showing the comparison between the values of muscle-specific miRNA in total ALS, spinal ALS (Spi-ALS), bulbar ALS (B-ALS), and controls. *p < 0.05; **p < 0.005.

Table 3. Expression levels of mirnas, and plasma concentration of myostatin/follistatin.

Biomarker Spinal ALS Bulbar ALS Total ALS ControlsmiRNA-1 1.18 ± 0.50 0.89 ± 0.38 1.03 ± 0.44 1.00 ± 0.42miRNA-133a 1.35 ± 0.45 0.36 ± 0.18 0.74 ± 0.24 1.00 ± 0.37miRNA-133b 3.09 ± 0.34 0.70 ± 0.19 1.56 ± 0.21 1.00 ± 0.57miRNA-206 6.86 ± 0.48 1.53 ± 0.17 3.45 ± 0.23 1.00 ± 0.60miRNA-146a 0.41 ± 0.56 0.36 ± 0.44 0.39 ± 0.50 1.00 ± 0.60miRNA-149* 0.14 ± 0.37 0.44 ± 0.36 0.23 ± 0.31 1.00 ± 0.60miRNA-155 0.44 ± 0.58 1.39 ± 0.44 0.75 ± 0.42 1.00 ± 0.57miRNA-221 0.72 ± 0.49 0.73 ± 0.48 0.72 ± 0.49 1.00 ± 0.53miRNA-27a 0.38 ± 0.40 0.37 ± 0.55 0.37 ± 0.46 1.00 ± 0.63Myostatin/follistatin ratio 0.55 ± 0.12 0.85 ± 0.31 0.69 ± 0.26 0.53 ± 0.30

Biomarkers and miRNA in ALS 7

in the subsynaptic region of muscle fiber in ALS mice, since it is capable of sensing damage or loss of motor neurons and to pro-mote the regeneration of functional synapses [19]. When miR-206 was inactivated in ALS mice, the time from disease onset to death was shortened, indicating that miR-206 is protective in ALS [18]. Indeed, the levels of miR-206 were higher in ALS long-term sur-vivor patients than in patients with rapidly progressive disease [16].

Besides the acquired notion that high lev-els of miR-206 are found in ALS, we dem-onstrated that also the levels of miRNA-133a and miRNA-133b are differentially expressed in bulbar-onset ALS vs. spinal-onset ALS. These data suggest that miR-133b may also have a specific protective function at the syn-apse in ALS, where denervation occurs fol-lowing motor neuron cell death.

Earlier studies demonstrated that in-creased levels of miR-155 were found in

Figure 3. Histograms showing the comparison between the values of inflammatory and angiogenic miR-NA in total ALS, spinal ALS (Spi-ALS), bulbar ALS (B-ALS), and controls. *p < 0.05; **p < 0.005.

Figure 4. Left panel: Histogram showing the comparison between the values of miRNA-27a in total ALS, spinal ALS (Spi-ALS), bulbar ALS (B-ALS), and controls. Right panel: Histogram showing the comparison between the ratio of plasma myostatin/follistatin concentration in the various clinical groups of ALS patients and controls. *p < 0.05; **p < 0.005.

Tasca, Pegoraro, Merico, and Angelini 8

peripheral monocytes [7] and in spinal cord tissue [20] from ALS patients, where miR-155 promotes pro-inflammatory pathways. Although the serum levels of miR-155 and miR-221 in our series of ALS patients did not differ from controls, we found higher val-ues of miR-155 in bulbar ALS than in spinal ALS. This result might suggest a major role of such miRNAs in bulbar ALS subgroup of patients, but the mechanisms by which it would be exerted are unclear.

In addition, miR-146a and miR-149* were down-regulated in our series of ALS patients as compared to controls, and also between bulbar ALS and spinal ALS. Pre-vious studies on miR-146a in ALS patients have shown increased levels as compared to controls, but these investigations were con-ducted on cerebrospinal fluid [7] or spinal cord tissue [21], possibly explaining this ap-parently discordant finding.

Another novel and interesting result from our study is the demonstration that myo-statin/follistatin levels were increased in ALS patients as compared to controls, and that this increase is especially observed in bulbar ALS. Our results would suggest that bulbar ALS patients are more likely to suf-fer from diffuse muscle atrophy than spinal ALS. This data agrees with the notion that myostatin can be used as a marker of muscle atrophy, and it is consistent with the muscle fiber morphometric data, where bulbar ALS showed more prominent fiber atrophy than spinal ALS. Indeed, myostatin deficient mice show larger muscles [22], and myostatin pathway suppresses muscle growth and in-duces muscle atrophy. Follistatin was studied for its role in regulation of muscle growth in mice, as an antagonist to myostatin, which inhibits excessive muscle growth [12].

In this study we also investigated the ex-pression of miRNA-27a, which is involved in muscle growth, and promotes myoblasts proliferation by reducing the expression of myostatin, a critical inhibitor of skeletal muscle myogenesis and hypertrophy [23]. In the animal model, the levels of myostatin have been found to be inversely correlated with miR-27a [24]. The observation that we found significantly lower levels of miR-27a in ALS patients than in controls agrees with the notion that increased myostatin corre-lates with down-regulation of miR-27a, and

with the consequent high degree of muscle atrophy that characterizes ALS muscle.

According to the literature, bulbar-onset ALS patients have a more rapid decline than spinal-onset ALS patients [2] and similar re-sults are observed in our patients’ series. Our study presents a cross-sectional evaluation in a series of ALS patients, and, although the levels of the biomarkers investigated cannot be used as a prognostic factor of disease pro-gression, the differences observed in the two subclinical groups of ALS patients might suggest a possible differential role in the pathogenetic events.

The results from our study demonstrated that myo-miRNAs can be used as non-inva-sive circulating biomarkers of ALS, in rela-tion with the molecular basis of denervation and reinnervation in the different clinical subtypes of the disease. Although larger-scale investigations may be needed to con-firm our results, this study offers promising findings in this disorder.

Acknowledgments

This work was supported by research grants from the Association Française contre les Myopathies (15696 to CA), the Comi-tato Telethon Fondazione Onlus (GTB12001 to CA), the EuroBioBank network and the BBMRI network (BBMRNR).

Disclosures

The Authors declare no conflict of inter-est.

References[1] Brooks BR, Miller RG, Swash M, Munsat TL;

World Federation of Neurology Research Group on Motor Neuron Diseases. El Escorial revisited: revised criteria for the diagnosis of amyotrophic lateral sclerosis. Amyotroph Lateral Scler Other Motor Neuron Disord. 2000; 1: 293-299. Cross-Ref PubMed

[2] Turner MR, Kiernan MC, Leigh PN, Talbot K. Biomarkers in amyotrophic lateral sclerosis. Lan-cet Neurol. 2009; 8: 94-109. CrossRef PubMed

[3] Toivonen JM, Manzano R, Olivan S, Zaragoza P, Garcia-Redondo A, Osta R. MicroRNA-206: a potential circulating biomarker candidate for

Biomarkers and miRNA in ALS 9

amyotrophic lateral sclerosis. PLos One. 2014; 9: e89065.PubMed

[4] Scaramozza A, Marchese V, Papa V, Salaroli R, Sorarù G, Angelini C, Cenacchi G. Skeletal mus-cle satellite cells in amyotrophic lateral sclerosis. Ultrastruct Pathol. 2014; 38: 295-302. CrossRef PubMed

[5] Volontè C, Apolloni S, Parisi C. MicroRNAs: newcomers into the ALS picture. CNS Neurol Disord Drug Targets. 2015; 14: 194-207. Cross-Ref PubMed

[6] Cloutier F, Marrero A, O’Connell C, Morin P Jr. MicroRNAs as potential circulating biomarkers for amyotrophic lateral sclerosis. J Mol Neurosci. 2015; 56: 102-112. PubMed

[7] Butovsky O, Siddiqui S, Gabriely G, Lanser AJ, Dake B, Murugaiyan G, Doykan CE, Wu PM, Gali RR, Iyer LK, Lawson R, Berry J, Krichevsky AM, Cudkowicz ME, Weiner HL. Modulating in-flammatory monocytes with a unique microRNA gene signature ameliorates murine ALS. J Clin Invest. 2012; 122: 3063-3087. CrossRef PubMed

[8] Matsuzaka Y, Kishi S, Aoki Y, Komaki H, Oya Y, Takeda S, Hashido K. Three novel serum bio-markers, miR-1, miR-133a, and miR-206 for Limb-girdle muscular dystrophy, Facioscapulo-humeral muscular dystrophy, and Becker muscu-lar dystrophy. Environ Health Prev Med. 2014; 19: 452-458. CrossRef PubMed

[9] Williams AH, Valdez G, Moresi V, Qi X, McAnally J, Elliott JL, Bassel-Duby R, Sanes JR, Olson EN. MicroRNA-206 delays ALS progression and pro-motes regeneration of neuromuscular synapses in mice. Science. 2009; 326: 1549-1554. CrossRef PubMed

[10] De Felice B, Annunziata A, Fiorentino G, Borra M, Biffali E, Coppola C, Cotrufo R, Brettschneider J, Giordana ML, Dalmay T, Wheeler G, D’Ales-sandro R. miR-338-3p is over-expressed in blood, CFS, serum and spinal cord from sporadic amyo-trophic lateral sclerosis patients. Neurogenetics. 2014; 15: 243-253. CrossRef PubMed

[11] Dubowitz V, Sewry C. Muscle biopsy. A practical approach. IIIrd Edition. Saunders Elsevier. 2007.

[12] Wu RH, Wang P, Yang L, Li Y, Liu Y, Liu M. A po-tential indicator of denervated muscle atrophy: the ratio of myostatin to follistatin in peripheral blood. Genet Mol Res. 2011; 10: 3914-3923. CrossRef PubMed

[13] Boylan KB, Glass JD, Crook JE, Yang C, Thomas CS, Desaro P, Johnston A, Overstreet K, Kelly C, Polak M, Shaw G. Phosphorylated neurofilament heavy subunit (pNF-H) in peripheral blood and CSF as a potential prognostic biomarker in amyo-trophic lateral sclerosis. J Neurol Neurosurg Psy-chiatry. 2013; 84: 467-472. CrossRef PubMed

[14] Valdez G, Heyer MP, Feng G, Sanes JR. The role of muscle microRNAs in repairing the neuromus-cular junction. PLos One. 2014; 9: e93140.PubMed

[15] Fox MA, Sanes JR, Borza DB, Eswarakumar VP, Fässler R, Hudson BG, John SW, Ninomiya Y, Pedchenko V, Pfaff SL, Rheault MN, Sado Y, Segal Y, Werle MJ, Umemori H. Distinct target-derived signals organize formation, maturation, and main-tenance of motor nerve terminals. Cell. 2007; 129: 179-193. CrossRef PubMed

[16] Bruneteau G, Simonet T, Bauché S, Mandjee N, Malfatti E, Girard E, Tanguy ML, Behin A, Khiami F, Sariali E, Hell-Remy C, Salachas F, Pradat PF, Fournier E, Lacomblez L, Koenig J, Romero NB, Fontaine B, Meininger V, Schaeffer L, et al. Mus-cle histone deacetylase 4 upregulation in amyo-trophic lateral sclerosis: potential role in reinner-vation ability and disease progression. Brain. 2013; 136: 2359-2368. CrossRef PubMed

[17] Peters OM, Ghasemi M, Brown RH Jr. Emerging mechanisms of molecular pathology in ALS. J Clin Invest. 2015; 125: 1767-1779. CrossRef PubMed

[18] Brown RH. Medicine. A reinnervating microR-NA. Science. 2009; 326: 1494-1495. CrossRef PubMed

[19] Ma G, Wang Y, Li Y, Cui L, Zhao Y, Zhao B, Li K. MiR-206, a key modulator of skeletal muscle de-velopment and disease. Int J Biol Sci. 2015; 11: 345-352. CrossRef PubMed

[20] Koval ED, Shaner C, Zhang P, du Maine X, Fischer K, Tay J, Chau BN, Wu GF, Miller TM. Method for widespread microRNA-155 inhibition pro-longs survival in ALS-model mice. Hum Mol Genet. 2013; 22: 4127-4135. CrossRef PubMed

[21] Campos-Melo D, Dropplemann CA, He Z, Volken-ing K, Strong MJ. Altered microRNA expression profile in amyotrophic lateral sclerosis: a role in the regulation of NFL mRNA levels. Mol Brain. 2013; 6: 26. PubMed

[22] Whittemore LA, Song K, Li X, Aghajanian J, Davies M, Girgenrath S, Hill JJ, Jalenak M, Kelley P, Knight A, Maylor R, O’Hara D, Pearson A, Quazi A, Ryerson S, Tan XY, Tomkinson KN, Veldman GM, Widom A, Wright JF, et al. Inhibition of myo-statin in adult mice increases skeletal muscle mass and strength. Biochem Biophys Res Commun. 2003; 300: 965-971. CrossRef PubMed

[23] Huang Z, Chen X, Yu B, He J, Chen D. MicroRNA-27a promotes myoblast proliferation by targeting myostatin. Biochem Biophys Res Commun. 2012; 423: 265-269. CrossRef PubMed

[24] Yang T, Chen XL, Huang ZQ, Wen WX, Xu M, Chen DW, Yu B, He J, Luo JQ, Yu J, Mao XB, Zheng P. MicroRNA-27a promotes porcine myo-blast proliferation by downregulating myostatin expression. Animal. 2014; 8: 1867-1872. Cross-Ref PubMed