ORIGINAL RESEARCH published: 22 September 2021 doi: 10.3389/fneur.2021.693309 Gene Therapy Overexpressing Neuregulin 1 Type I in Combination With Neuregulin 1 Type III Promotes Functional Improvement in the SOD1G93A ALS Mice Guillem Mòdol-Caballero 1,2† , Mireia Herrando-Grabulosa 1,2† , Sergi Verdés 3,4 , Edited by: Belén García-Lareu 2,3 , Neus Hernández 1,2 , Isaac Francos-Quijorna 1,2‡ , Alberto Dubrovsky, Rubén López-Vales 1,2 , Assumpció Bosch 2,3,4 and Xavier Navarro 1,2* Fundacion Favaloro Neurosciences Institute, Argentina 1 Department of Cell Biology, Physiology and Immunology, Institute of Neurosciences, Universitat Autonoma de Barcelona, Reviewed by: Barcelona, Spain, 2 Centro de Investigación Biomédica en Red Sobre Enfermedades Neurodegenerativas (CIBERNED), Elena Abati, Barcelona, Spain, 3 Department of Biochemistry and Molecular Biology, Institute of Neurosciences, Universitat Autònoma De University of Milan, Italy Barcelona, Barcelona, Spain, 4 Unitat Mixta UAB-VHIR, Vall d’Hebron Institut de Recerca (VHIR), Barcelona, Spain Zhiyv Neal Niu, Mayo Clinic, United States Amyotrophic lateral sclerosis (ALS) is a fatal neurodegenerative disease affecting the *Correspondence: Xavier Navarro neuromuscular system for which currently there is no effective therapy. Motoneuron (MN) xavier.navarro@uab.cat degeneration involves several complex mechanisms, including surrounding glial cells and † These authors share first authorship skeletal muscle contributions. Neuregulin 1 (NRG1) is a trophic factor present particularly ‡ Present address: in MNs and neuromuscular junctions. Our previous studies revealed that gene therapy Isaac Francos-Quijorna, overexpressing the isoform I (NRG1-I) in skeletal muscles as well as overexpressing the Regeneration Group, The Wolfson Centre for Age-Related Diseases, isoform III (NRG1-III) directly in the central nervous system are both effective in preserving Institute of Psychiatry, Psychology and MNs in the spinal cord of ALS mice, opening novel therapeutic approaches. In this Neuroscience (IoPPN), King’s College study, we combined administration of both viral vectors overexpressing NRG1-I in skeletal London, London, United Kingdom muscles and NRG1-III in spinal cord of the SOD1G93A mice in order to obtain a synergistic Specialty section: effect. The results showed that the combinatorial gene therapy increased preservation of This article was submitted to MNs and of innervated neuromuscular junctions and reduced glial reactivity in the spinal Neuromuscular Disorders and Peripheral Neuropathies, cord of the treated SOD1G93A mice. Moreover, NRG1 isoforms overexpression improved a section of the journal motor function of hindlimb muscles and delayed the onset of clinical disease. However, Frontiers in Neurology this combinatory gene therapy did not produce a synergic effect compared with single Received: 10 April 2021 therapies, suggesting an overlap between NRG1-I and NRG1-III activated pathways and Accepted: 26 August 2021 Published: 22 September 2021 their beneficial effects. Citation: Keywords: amyotrophic lateral sclerosis, neuregulin 1, ErbB receptors, motoneuron, neuromuscular junction, Mòdol-Caballero G, spinal cord Herrando-Grabulosa M, Verdés S, García-Lareu B, Hernández N, Francos-Quijorna I, López-Vales R, INTRODUCTION Bosch A and Navarro X (2021) Gene Therapy Overexpressing Neuregulin 1 Amyotrophic lateral sclerosis (ALS) is a fatal neurodegenerative disorder characterized by the Type I in Combination With Neuregulin 1 Type III Promotes Functional loss of upper and lower motoneurons (MNs) leading to progressive paresis, muscle atrophy and Improvement in the SOD1G93A ALS death, often due to respiratory failure (1–3). In most patients the cause of ALS is unknown Mice. Front. Neurol. 12:693309. or sporadic (sALS), although some inherit familiar forms of the disease (fALS), associated with doi: 10.3389/fneur.2021.693309 several alterations in genes (4), such as superoxide dismutase 1 (SOD1) (5), hexanucleotide repeat Frontiers in Neurology | www.frontiersin.org 1 September 2021 | Volume 12 | Article 693309 Mòdol-Caballero et al. Neuregulin Gene Therapy for ALS expansions in chromosome 9 open reading frame 72 (C9orf72) in SOD1G93A mice (40). Importantly, NRG1-I expressed by the (6, 7), TAR-DNA binding protein (TDP-43) (8, 9), and DNAJC7 Schwann cells has a crucial role in axonal regeneration and (10). During the last decades, progress has been made on remyelination (41), in development of NMJs (31) and in muscle identifying the pathophysiological mechanisms underlying the reinnervation (42). Thus, overexpression of NRG1-I causes development of ALS (2). It is likely that the disease has neuromuscular functional improvement, maintaining muscle a multifactorial cause, contributed by complex interactions innervation, and also increasing MN survival in SOD1G93A between genetic and molecular pathways (11). Among these, mice (43). Therefore, considering these recent findings, in the mutant SOD1 has been described to form aggregates leading present study our aim was to combine these two gene therapies to an impaired protein degradation and autophagy, exerting overexpressing NRG1-III in the spinal cord to preserve MNs and toxic effects, and causing MN degeneration (12). Furthermore, NRG1-I in muscles to maintain their innervation, searching for a evidence has also shown that neighboring cells, such as microglia, synergic effect. astrocytes and interneurons participate in the SOD1-related ALS pathogenesis (13–16). Previous studies have shown that MATERIALS AND METHODS disruption of neuromuscular junctions (NMJs) and axonal degeneration precedes MN death (17, 18), whereas a toxic role Animals and Study Design of skeletal muscle fibers was also reported (19). Transgenic B6SJL-Tg(SOD1∗ G93A)1Gur mice expressing the Among the animal models carrying different ALS-related mutation G93A in the human SOD1 gene were obtained from the mutations, the most extensively used is the transgenic mouse with Jackson Laboratory (Bar Harbor, ME). Non-transgenic wild-type the human mutation of a glycine to alanine conversion at the 93rd (WT) littermates were used as controls. Mice were maintained amino acid of the SOD1 gene (SOD1G93A ) (20, 21). These mice by crossing transgenic males with F1 hybrid females originally show various important clinical and pathological characteristics obtained from Janvier Laboratories (France). Transgenic mice of both the familial and the sporadic forms of the human ALS were identified by analyzing tissue samples from the tail. (18, 21, 22). In addition, further relevance of this model comes Animals were maintained under standard conditions and from the finding that accumulation of wild-type SOD1 protein handled in accordance with the guidelines of the European may also cause MN disease in mice (23). Union Council (Directive 2010/63/EU) and Spanish regulations Regrettably there are no effective treatments for ALS, and the on the use of laboratory animals. The Ethics Committee of the current standard of care is limited to symptomatic and palliative Universitat Autònoma de Barcelona approved the corresponding therapies (24). Only two drugs have been approved for ALS, experimental procedures. riluzole, that extends patients survival for a few months, and In this study we included three groups of B6xSJL female mice, edaravone, proposed to function by reducing oxidative injury in one of WT mice and two of SOD1G93A mice, administered at 7 central neurons. weeks of age with AAV coding either for NRG1-III and NRG1-I Interestingly, the ALS field has entered a thrilling era of or the respective mock vectors. The following mice groups were therapeutic development as shown by the potential of gene used for functional analysis: WT mice (n = 6 females), SOD therapies in the recent years (3, 25, 26). This approach Mock mice (n = 18 females), SOD NRG1-I + NRG1-III FL mice allows to specifically deliver a gene of interest overcoming the (n = 11 females). Then, mice were distributed in two subgroups limitation of large proteins crossing the blood-brain barrier (27). at 16 weeks, one for analysis of mRNA expression: WT (n = 3 Particularly, adeno-associated vectors (AAV) have the ability to females), SOD Mock (n = 4 females), SOD NRG1-I + NRG1-III provide sustained gene expression without risk for insertional (n = 5 females), and another for histological studies: WT (n = 3 mutagenesis. Moreover, viral transgene expression may target females), SOD Mock (n = 5 females), SOD NRG1-I + NRG1-III specific cell types by using viral vectors with particular tropism (n = 6 females). and employing promoters that induce selective and defined gene A time-chart showing the AAV administration and the follow- expression (28). up made during progression of the disease in the SOD1G93A mice Notably, spliced forms of Neuregulin 1 (NRG1) yield can be found in the Supplementary Figure 1. neurotrophic factors that play essential roles in promoting MN survival, supporting axonal, and neuromuscular development Virus Production and Administration and maintenance (29–35). NRG1 is expressed by spinal MNs, The AAV vectors used have been previously described (40, 43). with particular localization at the subsurface cisterns close to The full length sequence of the NRG1-III or the extracellular postsynaptic sites of cholinergic C synapses (36, 37). Strikingly, domain of the NRG1-I isoforms containing a HA-tag were cloned alterations in the NRG1 and ErbB receptors have been linked to between AAV2 ITRs under the regulation of the ubiquitous CMV ALS since loss-of-function mutations of ErbB4 is the cause of a promoter or the human muscle desmin promoter, respectively. form of autosomal-dominant ALS (38). In this sense, previous To stabilize mRNA expression, the woodchuck hepatitis virus ′ studies indicated alterations of the expression of NRG1 isoforms responsive element (WPRE) was added at 3 of the constructs. in both ALS patients and SOD1G93A mice, with downregulation AAV8 and AAVrh10 vectors were generated using the triple of NRG1 type III and upregulation of NRG1 type I in the spinal transfection system in HEK293-AAV cells of expression and cord (39, 40). In a recent study, we showed that treatment with capsid plasmids (Rep2Cap8 or Rep2Caprh10, provided by J.M. an AAV coding for the full-length form of NRG1-III improves Wilson, University of Pennsylvania, Philadelphia, USA) and motor function of hindlimb muscles and increases MN survival adenoviral helper sequences contained in pXX6 (44). Viral Frontiers in Neurology | www.frontiersin.org 2 September 2021 | Volume 12 | Article 693309 Mòdol-Caballero et al. Neuregulin Gene Therapy for ALS particles were purified in an iodixanol gradient, and titred using for quantification relative to controls using the 36B4 gene for the Picogreen (Invitrogen) system (45) by the Viral Production mRNA or Cyclophilin for DNA (47). Unit of UAB-VHIR (http://www.viralvector.eu). AAV8 vector Spinal cord and GM muscle DNA was extracted with coding mock or NRG1-I, and AAVrh10 virus coding mock or proteinase K (0.1 mg/ml; Sigma), followed by phenol/chloroform NRG1 type III were produced as in previous studies (40, 43). extraction. Real time primers for the housekeeping gene For the combined therapy, intravenous administration cyclophilin B, Desmin-NRG1-I or CAG-Nrg1-III constructs, of 1x1012 vg of AAV8-desmin-NRG1-I and intrathecal with the forward primer anneals with the promoter sequence and administration of AAVrh10-CMV-NRG1-III were performed the reverse primer with each Nrg1 isoform, are the following: in 7 weeks old mice. For intrathecal injection, under anesthesia mCycloB Fwd TCAACCTCTCCTCTCCTGCC; mCyclo Rev: with ketamine/xylazine (100/10 mg/kg i.p.), the spine was GGTTTCTCCACTTCGATCTTGC; Desmin GGGTTTGGG surgically exposed and 10 µl of viral vector (1 × 1011 vg of GTTCTGAATGTG; Nrg1-I Rev CGGTCCTTCTTCTTGCCC AAVrh10-CMV-Nrg1-III vector or mock vector) were delivered TT; CMV AGAACCCACTGCTTACTGGC; Nrg1-III CGG between lumbar vertebrae L3 and L4 into the cerebrospinal fluid TCCTTCTTCTTGCCCTT. Viral genome copies/cell were (CSF) with a 30G needle attached to a Hamilton syringe, as calculated using a standard curve generated from known previously described (46). Suitable injection into the intrathecal amounts of a plasmid DNA containing a CMV-Nrg1-III or space was confirmed by the animal’s tail flick. Des-Nrg1-I sequences or a 500 bp cyclophilin PCR product (CyclophilinB-Fwd5617: CATGCCTATGGTCCTAGCTT and Electrophysiological Tests CyclophilinB-Rv6141) purified by Geneclean (Q-Biogene) in Motor nerve conduction tests were performed as previously 10 ng per ml of salmon’s sperm DNA (Sigma) and assuming described (18, 43) by stimulating the sciatic nerve and recording that 1 mg of mouse genomic DNA contains 3x105 haploid the compound muscle action potential (CMAP) from tibialis genomes (46). anterior (TA), gastrocnemius (GM), and plantar interossei (PL) muscles at 9, 12, 14, and 16 weeks of age. Motor unit Histological Analysis number estimation (MUNE) was calculated following previously Other subgroups of mice were perfused with 4% described protocol using an incremental technique (18, 43). paraformaldehyde in PBS also at 16 weeks of age. Lumbar Motor evoked potentials (MEP) recorded from TA and GM spinal cord and the GM muscle were obtained. Spinal cords muscles were studies to determine the central descending were post-fixed and cryopreserved in 30% sucrose in PBS. GM pathways as previously reported (18, 43). muscles were directly cryopreserved in 30% sucrose in PBS. For spinal MN counting, transverse sections were cut at Locomotion Tests 20 µm thickness with a cryotome (Leica, Germany) and serially To evaluate motor coordination and strength of the animals the collected. Slides containing L4-L5 spinal cord sections separated rotarod test was conducted by placing mice onto the rod turning 100 µm were stained with cresyl violet. Motoneurons were at a speed of 14 rpm, weekly from 9 to 16 weeks of age (43). identified in the lateral side of the ventral horn and counted if fulfilling the criteria of having diameter larger than 20 µm, Nucleic Acids Extraction and Real Time polygonal shape and prominent nucleolus. PCR For astrocyte and microglia labeling, other sections of L4-L5 Mice were sacrificed by decapitation after deep anesthesia at 16 segments were immune-labeled with primary antibodies anti- weeks of age. L4-L5 spinal cord segments and gastrocnemius ionized calcium binding adapter molecule 1 (Iba-1, 1:1,000; muscles were harvested and frozen. Total RNA was extracted 019-19741, Wako, Japan), and anti-glial fibrillary acidic protein with QIAzol Lysis Reagent (QIAGEN) and 1 µg per sample (GFAP, 1:1,000; 130300, Invitrogen, USA), respectively (40, 43). was reverse transcribed using the iScriptTM Reverse transcriptase Images were visualized under a fluorescence microscope (Nikon and Reverse Transcription Supermix (Bio-Rad). The reverse Eclipse NI, Japan). The integrated density of Iba1 and GFAP was transcription cycle conditions were 25◦ C for 5 min, 42◦ C for measured by centering the ventral horn using ImageJ software. 20 min and 95◦ C for 1 min. We analyzed mRNA expression For NMJ labeling, longitudinal sections of the GM muscle of Nrg1-I and Nrg1-III, by means of specific primer sets: were cut at 60 µm thickness and collected in sequential series Nrg1-I Fwd TGGGAACGAGCTGAACCGCA; Nrg1-I Rev in Olmos solution. Sections were processed for immunolabeling TCCAGAGTCAGCCAGGGATG; Nrg1-III Fwd TTCCCTTCT with primary antibodies anti-synaptophysin (1:500; AB130436, CCAGCTCGGACC; Nrg1-III Rev GTCCCAGTCGTGGATGTA Abcam, UK) and anti-neurofilament 200 (NF200, 1:1,000; GATG. Mouse 36B4 was used to normalize the expression AB5539, Millipore, USA) as previously reported (43). Then, levels (m36B4 Fwd ATGGGTACAAGCGCGTCCTG; m36B4 incubated with Alexa 594-conjugated secondary antibody Rev AGCCGCAAATGCAGATGGAT. Messenger RNA analyses (1:200; A11042, Invitrogen, USA) and Alexa 488 conjugated were performed by real-time PCR (Bio-Rad CFX384 real-time alfa-bungarotoxin (1:200; B13422, Life Technologies, USA). PCR system). The thermal cycling conditions started with 5 min Images were visualized and captured under scanning confocal polymerase activation at 95◦ C, followed by 45 cycles of 15s at microscopy (LSM 700 Axio Observer, Carl Zeiss, Germany). 95◦ C, 30s at 60◦ C, 30s at 72◦ C and 5s at 65◦ C to 95◦ C (increasing NMJs were classified depending on whether the endplate 0.5◦ C every 5s). Melting curves were analyzed by monitoring the was fully occupied (innervated), partially occupied or vacant SYBR Green decrease after extension. The Pfaffl method was used (denervated). At least 100 endplates were analyzed per muscle. Frontiers in Neurology | www.frontiersin.org 3 September 2021 | Volume 12 | Article 693309 Mòdol-Caballero et al. Neuregulin Gene Therapy for ALS FIGURE 1 | Combined-gene therapy upregulates NRG1-I in the GM muscle and NRG1-III in the spinal cord. (A) mRNA expression analysis indicated an overexpression of NRG1-I in the GM muscle upon treatment in the SOD1G93A mice. (B) In the lumbar spinal cord, NRG1-III overexpression rescues NRG1-III mRNA levels similar to WT group, whereas it is downregulated in the SOD1G93A mock group. (C,D) Viral genome analysis reveals that intravenous and intrathecal AAV administration were effective in targeting both GM muscle (C) and Spinal Cord (D). [(n = 3 WT, 3-4 SOD Mock, 4-5 SOD NRG1-I + NRG1-III mice per group, one-way ANOVA, ***p < 0.001, **p < 0.01, *p < 0.05 vs. SOD Mock mice]. Data are shown as mean ± SEM. Immunoblot Analysis ECL substrate (CYANAGEN). Image density was analyzed with Snap-frozen lumbar spinal cord samples were sonicated Image LabTM software (Bio-Rad). At least 3 animals per group and homogenized in RIPA lysis buffer plus a mixture of and treatment and 3 different blots were quantified for each protease inhibitors (Millipore). Total protein concentration protein analyzed. was determined by BCA Protein Assay Kit (ThermoScientific). Protein separation was performed on 10% SDS-polyacrylamide Statistical Analysis electrophoresis gel (VWR Life Science), transferred to PVDF To avoid bias, experiments were performed by researchers membranes (GE Healthcare) and immunoblotted. The following blinded regarding the treatment received by each group, and primary antibodies were used: rabbit anti-phospho Erk1/2 mice were randomly distributed in groups. Sample size was (1:500, #9101Thr202/Tyr204, Cell Signaling), total Erk (1:500; selected according to previous observations in our lab. Data are #9102, Cell Signaling), rabbit anti-phospho Akt (1:500, Ser473 shown as mean ± SEM. Results of motor nerve conduction #9271, Cell Signaling), total Akt (1:500; #9272 Cell Signaling), and locomotion tests were analyzed with repeated measurements and rabbit anti-GAPDH (1:1,000; #14C10, Cell Signaling). ANOVA, followed by Bonferroni post-hoc test when necessary. The secondary antibodies were anti-rabbit HRP-conjugated Results of MUNEs and MEPs were analyzed with t-Student test. secondary antibody (1:10,000; Dako) and Westar Eta C Ultra 2.0 For clinical disease onset the Log-rank (Mantel-Cox) test was Frontiers in Neurology | www.frontiersin.org 4 September 2021 | Volume 12 | Article 693309 Mòdol-Caballero et al. Neuregulin Gene Therapy for ALS applied. Results of histological and molecular biology analyses ± 0.5), and higher number of innervated NMJs (74.7 ± 3.7) were compared using t-Student or ANOVA with Tukey’s post-hoc than in SOD1G93A mock mice (43.0 ± 4.7) (Figures 3A,B). The test. A difference was considered significant if p ≤ 0.05. combined gene therapy also produced a significant decrease of astrocyte and microglial activation (6.38 × 107 ± 7.18 × 106 and 3.54 × 107 ± 9.42 × 106 , respectively) whereas the untreated RESULTS mice showed marked glial reactivity in the ventral horn (5.05 × NRG1-I and NRG1-III Overexpression in 108 ± 1.99 × 108 and 2.27 × 108 ± 8.86 × 107 , respectively) Skeletal Muscles and Spinal Cord (Figures 3C,D). Taking into account the positive results obtained from the application of either NRG1-I or NRG1-III gene therapies (40, 43), NRG1 Type I and Type III Isoforms we aimed to test whether a combined therapy overexpressing Signaling Pathways both isoforms at different sites of the motor unit could produce We then evaluated whether the overexpression of NRG1 isoforms a synergic effect on the SOD1G93A mice. Vector injections were activated downstream cell survival signaling pathways. We performed at the same intervention, previous to disease onset found that treatment with the combined gene therapy strongly (at 7 weeks of age). mRNA analysis at the advanced-stage of the upregulated Akt phosphorylation in the spinal cord compared to disease (16 weeks) showed that NRG1-I expression was increased SOD1G93A untreated mice (Figure 4A). In contrast, activation of in the GM muscle (38.77 ± 11.18) of treated in comparison Erk was significantly decreased in the SOD1G93A treated mice, in with the control SOD1G93A mice (0.76 ± 0.41) (Figure 1A), accordance with our previous reports (40, 43). and NRG1-III was increased in the spinal cord (1.12 ± 0.04) In correspondence with the above results, we found that compared to the respective SOD1G93A control mice (0.68 ± 0.05) NRG1-I and NRG1-III overexpression produced a global (Figure 1B). In agreement with these results, we detected 0.18 ± functional improvement as shown in the rotarod test and delayed 0.05 viral genomes (vg)/cell of Desmin-NRG1-I in the GM and the clinical disease onset (Figures 4B,C). 14.21 ± 4.58 vg/cell of CMV-NRG1-III in the spinal cord of the treated SOD1G93A mice (Figures 1C,D). As expected, qPCR did Combined NRG1 Gene Therapy Does Not not detect changes in NRG1-I mRNA in the spinal cord of treated Produce a Synergic Effect on SOD1G93A animals, since AAV8 does not cross the BBB and the Neuromuscular Function in SOD1G93A Mice desmin promoter is not expressed in spinal cord. Low NRG1- Finally, we compared the functional and histological results III expression was detected in GM muscles of treated SOD1G93A obtained by the combined gene therapy with those previously that can be explained by CSF drainage to the periphery, although reported with NRG1-III (40) or NRG1-I (43) single gene therapy. five times lower than by intravenous administration of the muscle The electrophysiological results showed that the combined specific promoter vector (Supplementary Figure 2). therapy did not increase the CMAP amplitude of PL, TA, and GM muscles when compared to the gene therapies overexpressing Combined NRG1 Gene Therapy Preserves only NRG1-III or NRG1-I (Figures 5A–C). Consistent with Motor Function in SOD1G93A Mice these results, a similar delayed onset and improvement in the The results of motor nerve conduction tests showed a significant rotarod test performance was found with the three therapies preservation of the amplitude of PL, GM and TA CMAPs in (Figures 5H,I). Furthermore, the histological analysis showed treated SOD1G93A compared to mock SOD1G93A control mice that the number of spinal MNs, and the astrocyte and microglia (Figures 2A–C). This was further corroborated by the increase activation were similar comparing the different SOD1G93A in number and amplitude of remaining motor units, tested in treated mice groups (Figures 5D,F,G). Interestingly, we found both PL and TA muscles at the last evaluation at 16 weeks of that gene therapy overexpressing NRG1-I in muscles alone and age (Figures 2D–F). Moreover, the plot of motor units grouped combined with NRG1-III significantly increased the number by their action potential amplitude showed a higher number of of occupied NMJ whereas the NRG1-III therapy alone did the treated group in most of the intervals, reflecting a general not (Figure 5E), which reinforces the theory of the dying-back effect on all types of motor units (Figures 2D–F). The MEPs mechanism in ALS and suggests the need to a muscle-directed recorded on GM and TA muscles hand also a significantly therapy, as we previously suggested with other neurotrophic higher amplitude in the treated than in the control SOD1G93A factors (48). group, indicating preserved connections with the upper MNs (Figure 2G). DISCUSSION Combined Gene Therapy Preserves the The results of the present study indicate that generalized Number of MN and Reduces Glial increased expression of NRG1-I in body muscles as well Reactivity as overexpression of NRG1-III in the spinal cord promote The histological analysis performed at 16 weeks of age confirmed neuromuscular functional improvement and increase MN the improvement in neuromuscular function. There was higher survival while reducing glial cell reactivity around spinal MNs in number of surviving spinal MNs in the SOD1G93A treated mice the SOD1G93A murine model of ALS. However, these positive (11.9 ± 0.4) compared to the control SOD1G93A mice (6.9 effects did not show a synergic effect in the transgenic SOD1G93A Frontiers in Neurology | www.frontiersin.org 5 September 2021 | Volume 12 | Article 693309 Mòdol-Caballero et al. Neuregulin Gene Therapy for ALS FIGURE 2 | Combined-gene therapy produces functional improvement of the SOD1G93A mice. Electrophysiological tests demonstrate that AAV-NRG1-I and AAV-NRG1-III FL injections promoted a significant preservation of the CMAP amplitude of plantar (A), tibialis anterior (B), and gastrocnemius (C) muscles in the SOD1G93A mice (n = 6 WT, 16-18 SOD Mock, 7-9 SOD NRG1-I + NRG1-III mice per group, *p < 0.05 vs. SOD Mock mice; #### p < 0.0001, ### p < 0.001, # p < 0.05 vs. WT mice). Electrophysiological estimation of motor unit number (MUNE) and mean amplitude of single motor unit potential (SMUA) of plantar (D) and tibialis anterior (E) muscles show a general preservation of motor units induced by the combined therapy, confirmed by the frequency distribution (F). (G) The combined therapy increased the amplitude of the MEPs, indicating enhanced connection between upper and lower MNs. (n = 14-21 SOD Mock, 5-9 SOD NRG1-I + NRG1-III mice per group, ***p < 0.001, **p < 0.01, *p < 0.05 vs. SOD Mock mice). Data are shown as mean ± SEM. mice compared to treatment with gene therapy overexpressing our group in organotypic cultures of the spinal cord subject either NRG1-I or NRG1-III alone. to excitotoxicity (52), and that overexpression of NRG1-III in NRG1 isoforms are essential trophic factors due to their the spinal cord or of NRG1-I in the skeletal muscle of the several functions in the development and maintenance of the SOD1G93A mice both improve motor function and maintains nervous system (49). Importantly, NRG1/ErbB system alterations the number of surviving spinal MNs (40, 43). In light of these have been reported in ALS human patients (38, 50). It has latter results, we were encouraged to combine the two gene been described in ALS patients as well as in the SOD1G93A therapies for overexpressing NRG1-I and NRG1-III to produce mice, that the expression of NRG1-III is reduced in the spinal a synergic effect on the SOD1G93A mice. Our hypothesis was cord in correlation with MN loss (39, 51, 52). In this sense, that increasing production of NRG1-III at the spinal cord (by restoration of C-boutons number and slight improvement of means of intrathecal administration of AAVrh10-CMV-NRG1- survival of the SOD1G93A mice was shown after viral vector III) would activate protective pathways in the MNs, whereas inducing NRG1-III administration in the CNS (51). In addition, overexpression of NRG1-I at the skeletal muscles (by means of neuroprotective effects on MNs have been previously found by systemic injection of AAV8-desmin-NRG1-I) would maintain Frontiers in Neurology | www.frontiersin.org 6 September 2021 | Volume 12 | Article 693309 Mòdol-Caballero et al. Neuregulin Gene Therapy for ALS FIGURE 3 | Overexpression of NRG1-I and NRG1-III preserves spinal motoneurons and reduces glial reactivity. (A) Representative images of L4 spinal cord and GM neuromuscular junctions of wild type and SOD1G93A mice, treated with NRG1-I and NRG1-III or with mock vector (scale bar = 100 µm). (B) Histological analysis indicated a higher preservation of the number of spinal MNs, as well as an increased proportion of innervated NMJs in the GM muscle of the treated SOD1G93A mice (n = 3 WT, 5 SOD Mock, 6 SOD NRG1-I + NRG1-III mice per group, ****p < 0.0001, ***p < 0.001, *p < 0.05 vs. SOD Mock mice). (C) Representative confocal images of astrocytes labeled with GFAP, and microglia labeled with Iba-1, in the spinal cord ventral horn of SOD1G93A mice (scale bar = 100 µm). (D) The viral-mediated delivery of both NRG1-I and NRG1-III decreased astrocyte and microglia reactivity in the spinal cord of the SOD1G93A mice (n = 3-4 SOD Mock, 5-6 SOD NRG1-I + NRG1-III mice per group, *p < 0.05). Data are shown as mean ± SEM. the NMJ and promote re-innervation of the already denervated the NMJ. Although published data show that dual AAV therapies muscle (42). Thus, the combined gene therapy was aimed to injected in murine models can be successful (53, 54), this issue target the two main sites of the ALS pathogenesis, the MN and still presents some concerns. Administration of high-titer AAV Frontiers in Neurology | www.frontiersin.org 7 September 2021 | Volume 12 | Article 693309 Mòdol-Caballero et al. Neuregulin Gene Therapy for ALS FIGURE 4 | Combined gene therapy overexpressing NRG1-I and III activates survival pathways in spinal cord and produces functional improvement. (A) NRG1-I and combined with type III overexpression increases AKT phosphorylation and diminishes Erk2 activation in the spinal cord of the treated mice, as demonstrated by western blot. Quantifications show relative phosphorylation compared to total protein, normalized by GAPDH and are represented by fold-change compared to WT animals; (n = 3–5 WT mice, n = 6-9 SOD Mock, n = 4-7 SOD NRG1-I mice, n = 4-6 SOD NRG1-I + NRG1-III mice per group, ****p < 0.0001, ***p < 0.001, **p < 0.01 vs. SOD Mock mice). (B) The clinical onset of disease was significantly delayed by the combined therapy (n = 14 SOD Mock, 5 SOD NRG1-I + NRG1-III mice per group, Mantel-Cox test, **p < 0.01, *p < 0.05 vs. SOD Mock mice). (C) NRG1-I and III overexpression produced an improvement in the Rotarod performance of the treated SOD1G93A mice from 14 to 16 weeks (n = 14 SOD Mock, 5 SOD NRG1-I + NRG1-III mice per group, **p < 0.01, *p < 0.05 vs. SOD Mock mice). Data are shown as mean ± SEM. vectors induce the generation of antibodies that could neutralize in terms of MN survival and inflammation when applied in the the effect of a second treatment; to avoid this, we performed peripheral and central nervous system, respectively. However, the both vector injections on the same day. Indeed, the injection beneficial actions of these isoforms might overlap somehow in the of both viral vectors was effective, producing increased mRNA pathways activated in the MNs, since the combined gene therapy expression of NRG1-III in the spinal cord and of NRG1-I in the did not show any further positive result compared with both muscles. We previously demonstrated that intravenous injection gene therapies alone in the mice. Certainly, NRG1-I activates the of an AAV8 vector with the desmin promoter is able to transduce PI3K/AKT and ERK pathways when overexpressed in the GM most skeletal muscles and the heart in SOD1G93A mice (48). muscle of the SOD1G93A mice (42), whereas NRG1 treatment Furthermore, the results showed that the combined-gene therapy also activates the same pathways in the spinal cord to promote promoted functional improvement of the SOD1G93A mice, MN survival (52, 56). Consequently, the maximum effect that although it did not achieve the expected synergic effect. One of can be reached by the overexpression of NRG1 isoforms may be the possible explanations could be regarding the limitations of the limited. We cannot discard, though, that a synergistic effect in used model. Indeed, very few studies have reported synergistic spinal cord and muscle is already exerted by NRG1-III through effects after combinatory therapies (55). This might reflect that drainage to the periphery of AAVrh10-NRG1-III vector as part of there is an endogenous limitation for the beneficial outcomes that the CSF circulation, and since we used a ubiquitous promoter for can be achieved using the SOD1G93A mouse model. In view of this approach. Indeed, we detected slight increase of NRG1-III the past and recent results, NRG1-I and NRG1-III may differ in expression in GM muscle, although 4–5 times lower than when their functions in the nervous system but share positive effects using the muscle specific promoter. Frontiers in Neurology | www.frontiersin.org 8 September 2021 | Volume 12 | Article 693309 Mòdol-Caballero et al. Neuregulin Gene Therapy for ALS FIGURE 5 | Combined gene therapy overexpressing NRG1-I and III does not promote a synergic effect. (A) Electrophysiological tests show that the combined gene therapy overexpressing NRG1 isoforms I and III does not improve the preservation of the CMAP amplitude of PL (A), TA (B) and GM (C) compared to both gene therapies performed alone in SOD1G93A mice (18-31 SOD Mock, 7-9 SOD NRG1-I + NRG1-III, 10-11 SOD NRG1-I and 20 SOD NRG1-III mice per group, ****p < 0.0001, ***p < 0.001, *p < 0.05 vs. SOD NRG1-III mice; ### p < 0.001, # p < 0.05 vs. SOD NRG1-I mice; $$ p < 0.01, $ p < 0.05 vs. SOD NRG1-I + NRG1-III mice). (D) Histological analysis shows similar numbers of spinal MNs in the different treated SOD1G93A mice, (E) but a significant increase on the proportion of innervated NMJs in the GM muscle for the combined NRG1 therapy and the NRG1-I single therapy compared to the NRG1-III alone (n = 6-7 WT, 8 SOD Mock, 6 SOD NRG1-I + NRG1-III, 7 SOD NRG1-I and 5-7 SOD NRG1-III mice per group, ****p < 0.0001 *p < 0.05 vs. SOD Mock mice; ## p < 0.01, # p < 0.05 vs. SOD NRG1-III mice). (F) The viral-mediated delivery of the different gene therapies combining NRG1 isoforms decreased astrocyte (G) and microglia reactivity in the spinal cord of the SOD1G93A mice (n = 6–8 SOD Mock, 8 SOD Mock, 5-6 SOD NRG1-I + NRG1-III, 6 SOD NRG1-I and 8 SOD NRG1-III mice per group, T-student, *p < 0.05 vs. SOD Mock mice). (H) The clinical onset of disease was significantly delayed by the gene therapies targeting NRG1 (n = 23 SOD Mock, 5 SOD NRG1-I + NRG1-III, 9 SOD NRG1-I and 10 SOD NRG1-III mice per group; Mantel-Cox test, **p < 0.01, *p < 0.05 vs. SOD Mock mice). (I) NRG1-I and III overexpression produced a similar improvement in the Rotarod performance of the treated SOD1G93A mice compared to the two gene therapies alone (n = 23 SOD Mock, 5 SOD NRG1-I + NRG1-III, 7 SOD NRG1-I and 10 SOD NRG1-III mice per group, ****p < 0.0001, *p < 0.05 vs. NRG1-III mice; # p < 0.05 vs. SOD NRG1-I mice; $$ p < 0.01, $ p < 0.05 vs. SOD NRG1-I + NRG1-III mice). Data are shown as mean ± SEM. Regarding sex differences, it is worth to note that in our other hand, this mouse model shows more severe symptoms previous studies we showed that gene therapy targeting NRG1- and earlier manifestations in males than in females (59). I produced positive effects in both male and female mice, Consequently, therapeutic treatments are often more successful but in contrast, gene therapy for NRG1-III in the spinal cord in females than in males (60). only produced beneficial effects in the female SOD1G93A mice In summary, the combined NRG1-I and III overexpression (40, 43). Therefore, for the combined NRG1 gene therapy we significantly improved motor function preservation and used female mice, corroborating the improvements induced promoted neuro-protection, accompanied by reduction of by NRG1. Progesterone may participate in modulating NRG1- neuro-inflammatory reaction. However, we did not see an III expression in the spinal cord and play a role on the improved effect compared to the NRG1-I or III overexpression neuroprotective effects observed in female mice (57, 58). On the alone. Further studies evaluating the long-term effect of NRG1 Frontiers in Neurology | www.frontiersin.org 9 September 2021 | Volume 12 | Article 693309 Mòdol-Caballero et al. Neuregulin Gene Therapy for ALS gene therapy may help to determine its consideration for the Telethon, cooperative project 2015-01 from CIBERNED, translation to treat ALS patients. PID2020-116735RB-I00 from MICINN and TERCEL (RD16/0011/0014) funds from the Instituto de Salud Carlos III DATA AVAILABILITY STATEMENT of Spain. The original contributions presented in the study are included in the article/Supplementary Material, further inquiries can be ACKNOWLEDGMENTS directed to the corresponding authors. We acknowledge the excellent technical help of Monica Espejo, Jessica Jaramillo, and Israel Blasco. ETHICS STATEMENT The animal study was reviewed and approved by Ethics SUPPLEMENTARY MATERIAL Committee of the Universitat Autònoma de Barcelona. The Supplementary Material for this article can be found AUTHOR CONTRIBUTIONS online at: https://www.frontiersin.org/articles/10.3389/fneur. 2021.693309/full#supplementary-material XN and AB conceived the funded projects and the studies. Supplementary Figure 1 | Time-chart of the gene therapy administration and the GM-C, MH-G, and NH performed the in vivo experiments follow-up of the SOD1G93A mice along the course of the disease. Elements of the and histological analyses. BG-L, IF-Q, and RL-V contributed figure were modified from SMART (Servier Medical Art), licensed under a Creative to interventions and injections of viral vectors. SV and BG- Common Attribution 3.0 Generic License http://smart.servier.com/. L performed molecular analyses. GM-C, MH-G, AB, and XN Supplementary Figure 2 | (A) NRG1-III mRNA expression analysis in the GM contributed to the analyses and discussion of the results. GM-C muscle. NRG1-III mRNA is also increased, by 6-fold the WT levels, probably due and MH-G drafted the first version of the paper. All the authors to virus drainage to circulation from the CSF. However, note that NRG1-I contributed to writing and correcting the manuscript. expression driven from the desmin promoter induces a 40-times increase of NRG1-I in the GM muscle (see Figure 1A). (B) On the contrary, no statistically significant changes in NRG1-I isoform are detected in the spinal cord between FUNDING treated or non-treated SOD1G93A animals, as expected, since neither AAV8 crosses the BBB nor the desmin promoter expresses in spinal cord tissue. N = 3 This work was funded by grant TV3201428-10 of WT, 3-4 SOD Mock, 4-5 SOD NRG1-I + NRG1-III mice per group. One-way Fundació La Marato-TV3, grant #20289 of AFM- ANOVA, ∗ p < 0.05 vs. SOD Mock group. REFERENCES 9. Yokoseki A, Shiga A, Tan CF, Tagawa A, Kaneko H, Koyama A, et al. TDP- 43 mutation in familial amyotrophic lateral sclerosis. Ann Neurol. (2008) 1. Taylor JP, Brown, Jr RH, Cleveland DW. Decoding ALS: from genes to 63:538–42. doi: 10.1002/ana.21392 mechanism. Nature. (2016) 10:197–206. doi: 10.1038/nature20413 10. Farhan SMK, Howrigan DP, Abbott LE, Klim JR, Topp SD, Byrnes AE, et al. 2. Hardiman O, Al-Chalabi A, Chio A, Corr EM, Logroscino G, Robberecht Exome sequencing in amyotrophic lateral sclerosis implicates a novel gene, W, et al. Amyotrophic lateral sclerosis. Nat Rev Dis Primers. (2017) 3:17071. DNAJC7, encoding a heat-shock protein. Nat Neurosci. (2019) 22:1966–74. doi: 10.1038/nrdp.2017.71 doi: 10.1038/s41593-019-0530-0 3. Kim G, Gautier O, Tassoni-Tsuchida E, Ma XR, Gitler AD. ALS genetics: 11. Mancuso R, Navarro X. Amyotrophic lateral sclerosis: current perspectives Gains, losses, and implications for future therapies. Neuron. (2020) 108:822– from basic research to the clinic. Progr Neurobiol. (2015) 133:1–26. 42. doi: 10.1016/j.neuron.2020.08.022 doi: 10.1016/j.pneurobio.2015.07.004 4. van der Spek RAA, van Rheenen W, Pulit SL, Kenna KP, van den 12. Robberecht W, Philips T. The changing scene of amyotrophic lateral sclerosis. Berg LH, Veldink JH, et al. The project MinE databrowser: bringing Nat Rev Neurosci. (2013) 14:248–64. doi: 10.1038/nrn3430 large-scale whole-genome sequencing in ALS to researchers and the 13. Zhao W, Beers DR, Henkel JS, Zhang W, Urushitani M, Julien JP, et al. public. Amyotroph Lateral Scler Frontotemporal Degener. (2019) 20:432–40. Extracellular mutant SOD1 induces microglial-mediated motoneuron injury. doi: 10.1080/21678421.2019.1606244 Glia. (2010) 58:231–43. doi: 10.1002/glia.20919 5. Rosen DR, Siddique T, Patterson D, Figlewicz DA, Sapp P, Hentati A, 14. Brites D, Vaz AR. Microglia centered pathogenesis in ALS: et al. Mutations in Cu/Zn superoxide dismutase gene are associated with insights in cell interconnectivity. Front Cell Neurosci. (2014) 8:117. familial amyotrophic lateral sclerosis. Nature. (1993) 364:362. doi: 10.1038/36 doi: 10.3389/fncel.2014.00117 4362c0 15. Allodi I, Montañana-Rosell R, Selvan R, Löw P, Kiehn O. Locomotor 6. DeJesus-Hernandez M, Mackenzie IR, Boeve BF, Boxer AL, Baker deficits in a mouse model of ALS are paralleled by loss of V1-interneuron M, Rutherford NJ, et al. Expanded GGGGCC hexanucleotide repeat connections onto fast motor neurons. Nat Commun. (2021) 12:3251. in noncoding region of C9ORF72 causes chromosome 9p-linked doi: 10.1038/s41467-021-23224-7 FTD and ALS. Neuron. (2011) 72:245–56. doi: 10.1016/j.neuron.2011. 16. Zhao W, Beers DR, Appel S. H. Immune-mediated mechanisms in the 09.011 pathoprogression of amyotrophic lateral sclerosis. J Neuroimmune Pharmacol. 7. Renton AE, Majounie E, Waite A, Simón-Sánchez J, Rollinson S, Gibbs (2013) 8:888–99. doi: 10.1007/s11481-013-9489-x JR, et al. A hexanucleotide repeat expansion in C9ORF72 is the 17. Fischer LR, Culver DG, Tennant P, Davis AA, Wang M, Castellano-Sanchez A, cause of chromosome 9p21-linked ALS-FTD. Neuron. (2011) 72:257–68. et al. Amyotrophic lateral sclerosis is a distal axonopathy: evidence in mice and doi: 10.1016/j.neuron.2011.09.010 man. Exp Neurol. (2004) 185:232–40. doi: 10.1016/j.expneurol.2003.10.004 8. Kabashi E, Valdmanis PN, Dion P, Spiegelman D, McConkey BJ, Vande 18. Mancuso R, Santos-Nogueira E, Osta R, Navarro X. Electrophysiological Velde C, et al. TARDBP mutations in individuals with sporadic and familial analysis of a murine model of motoneuron disease. Clin Neurophysiol. (2011) amyotrophic lateral sclerosis. Nat Genet. (2008) 40:572–4. doi: 10.1038/ng.132 122:1660–70. doi: 10.1016/j.clinph.2011.01.045 Frontiers in Neurology | www.frontiersin.org 10 September 2021 | Volume 12 | Article 693309 Mòdol-Caballero et al. Neuregulin Gene Therapy for ALS 19. Moloney EB, de Winter F, Verhaagen J. ALS as a distal axonopathy: 39. Song F, Chiang P, Wang J, Ravits J, Loeb JA. Aberrant neuregulin 1 signaling molecular mechanisms affecting neuromuscular junction stability in the in amyotrophic lateral sclerosis. J Neuropathol Exp Neurol. (2012) 71:104. presymptomatic stages of the disease. Front Neurosci. (2014) 8:252. doi: 10.1097/NEN.0b013e3182423c43 doi: 10.3389/fnins.2014.00252 40. Mòdol-Caballero G, García-Lareu B, Verdés S, Ariza L, Sánchez-Brualla 20. Gurney ME, Pu H, Chiu AY, Dal Canto MC, Polchow CY, Alexander I, Brocard F, et al. Therapeutic role of neuregulin 1 type III in SOD1- DD, et al. Motor neuron degeneration in mice that express a human linked amyotrophic lateral sclerosis. Neurotherapeutics. (2020) 17:1048–60. Cu,Zn superoxide dismutase mutation. Science. (1994) 264:1772–5. doi: 10.1007/s13311-019-00811-7 doi: 10.1126/science.8209258 41. Stassart RM, Fledrich R, Velanac V, Brinkmann BG, Schwab MH, Meijer D, 21. Ripps ME, Huntley GW, Hof PR, Morrison JH, Gordon JW. Transgenic et al. A role for Schwann cell-derived neuregulin-1 in remyelination. Nat mice expressing an altered murine superoxide dismutase gene provide an Neurosci. (2013) 16:48–54. doi: 10.1038/nn.3281 animal model of amyotrophic lateral sclerosis. Proc Natl Acad Sci USA. (1995) 42. Mancuso R, Martínez-Muriana A, Leiva T, Gregorio D, Ariza L, Morell M, 92:689–3. doi: 10.1073/pnas.92.3.689 et al. Neuregulin-1 promotes functional improvement by enhancing collateral 22. Bosco DA, Landers JE. Genetic determinants of amyotrophic lateral sclerosis sprouting in SOD1(G93A) ALS mice and after partial muscle denervation. as therapeutic targets. CNS Neurol Disord Drug Targets. (2010) 9:779–90. Neurobiol. Dis. (2016) 95:168–78. doi: 10.1016/j.nbd.2016.07.023 doi: 10.2174/187152710793237494 43. Mòdol-Caballero G, Herrando-Grabulosa M, García-Lareu B, Solanes N, 23. Graffmo KS, Forsberg K, Bergh J, Birve A, Zetterström P, Andersen PM, Verdés S, Osta R, et al. Gene therapy for overexpressing Neuregulin 1 type I et al. Expression of wild-type human superoxide dismutase-1 in mice in skeletal muscles promotes functional improvement in the SOD1G93A ALS causes amyotrophic lateral sclerosis. Hum Mol Genet. (2013) 22:51–60. mice. Neurobiol Dis. (2020) 137:104793. doi: 10.1016/j.nbd.2020.104793 doi: 10.1093/hmg/dds399 44. Zolotukhin S, Byrne BJ, Mason E, Zolotukhin I, Potter M, Chesnut 24. Pasinelli P, Brown RH. Molecular biology of amyotrophic lateral K, et al. Recombinant adeno-associated virus purification using novel sclerosis: insights from genetics. Nat Rev Neurosci. (2006) 7:710–23. methods improves infectious titer and yield. Gene Ther. (1999) 6:973–85. doi: 10.1038/nrn1971 doi: 10.1038/sj.gt.3300938 25. Cappella M, Ciotti C, Cohen-Tannoudji M, Biferi MG. Gene therapy for 45. Piedra J, Ontiveros M, Miravet S, Penalva C, Monfar M, Chillon M. ALS-A perspective. Int J Mol Sci. (2019) 20:4388. doi: 10.3390/ijms20 Development of a rapid, robust, and universal picogreen-based method to 184388 titer adeno-associated vectors. Hum Gene Ther Methods. (2015) 26:35-42. 26. Abati E, Bresolin N, Comi G, Corti S. Silence superoxide dismutase 1 (SOD1): doi: 10.1089/hgtb.2014.120 a promising therapeutic target for amyotrophic lateral sclerosis (ALS). 46. Homs J, Pagès G, Ariza L, Casas C, Chillón M, Navarro X, et al. Intrathecal Exp Opin Ther Targets. (2020) 24:295–310. doi: 10.1080/14728222.2020. administration of IGF-I by AAVrh10 improves sensory and motor deficits in 1738390 a mouse model of diabetic neuropathy. Mol Ther Methods Clin Dev. (2014) 27. Federici T, Boulis NM. Gene-based treatment of motor neuron diseases. 15:1–7. doi: 10.1038/mtm.2013.7 Muscle Nerve. (2006) 33:302–23. doi: 10.1002/mus.20439 47. Pfaffl MW. A new mathematical model for relative quantification in real-time 28. Kügler S. Tissue-specific promoters in the CNS. Methods Mol Biol. (2016) RT-PCR. Nucleic Acids Res. (2001) 29:e45. doi: 10.1093/nar/29.9.e45 1382:81–91. doi: 10.1007/978-1-4939-3271-9_6 48. Mòdol-Caballero G, García-Lareu B, Herrando-Grabulosa M, Verdés 29. Syroid DE, Maycox PR, Burrola PG, Liu N, Wen D, Lee KF, et al. Cell death in S, López-Vales R, Pagès G, et al. Specific expression of glial-derived the Schwann cell lineage and its regulation by neuregulin. Proc Natl Acad Sci neurotrophic factor in muscles as gene therapy strategy for amyotrophic USA. (1996) 93:9229–34. doi: 10.1073/pnas.93.17.9229 lateral sclerosis. Neurotherapeutics. (2021). doi: 10.1007/s13311-021-0 30. Sandrock, Jr AW, Dryer SE, Rosen KM, Gozani SN, Kramer R, 1025-6. [Epub ahead of print]. et al. Maintenance of acetylcholine receptor number by neuregulins 49. Esper RM, Pankonin MS, Loeb JA. Neuregulins: versatile growth and at the neuromuscular junction in vivo. Science. (1997) 276:599–603. differentiation factors in nervous system development and human doi: 10.1126/science.276.5312.599 disease. Brain Res Rev. (2006) 51:161–75. doi: 10.1016/j.brainresrev.200 31. Loeb JA, Khurana TS, Robbins JT, Yee AG, Fischbach GD. Expression patterns 5.11.006 of transmembrane and released forms of neuregulin during spinal cord 50. Lopez-Font I, Sogorb-Esteve A, Javier-Torrent M, Brinkmalm G, Herrando- and neuromuscular synapse development. Development. (1999) 126:781–91. Grabulosa M, García-Lareu B, et al. Decreased circulating ErbB4 ectodomain doi: 10.1242/dev.126.4.781 fragments as a read-out of impaired signaling function in amyotrophic 32. Wolpowitz D, Mason TB, Dietrich P, Mendelsohn M, Talmage DA, lateral sclerosis. Neurobiol Dis. (2019) 124:428–38. doi: 10.1016/j.nbd.201 Role LW. Cysteine-rich domain isoforms of the neuregulin-1 gene are 8.12.021 required for maintenance of peripheral synapses. Neuron. (2000) 25:79–91. 51. Lasiene J, Komine O, Fujimori-Tonou N, Powers B, Endo F, Watanabe S, doi: 10.1016/S0896-6273(00)80873-9 et al. Neuregulin 1 confers neuroprotection in SOD1-linked amyotrophic 33. Michailov GV, Sereda MW, Brinkmann BG, Fischer TM, Haug B, Birchmeier lateral sclerosis mice via restoration of C-boutons of spinal motor C, et al. Axonal neuregulin-1 regulates myelin sheath thickness. Science. neurons. Acta Neuropathol. Commun. (2016) 4:15. doi: 10.1186/s40478-016-0 (2004) 304:700–3. doi: 10.1126/science.1095862 286-7 34. Nave KA, Salzer JL. Axonal regulation of myelination by neuregulin 1. Curr 52. Mòdol-Caballero G, Santos D, Navarro X, Herrando-Grabulosa M. Opin Neurobiol. (2006) 16:492–500. doi: 10.1016/j.conb.2006.08.008 Neuregulin 1 reduces motoneuron cell death and promotes neurite growth in 35. Velanac V, Unterbarnscheidt T, Hinrichs W, Gummert MN, Fischer TM, an in vitro model of motoneuron degeneration. Front Cell Neurosci. (2018) Rossner MJ, et al. Bace1 processing of NRG1 type III produces a myelin- 11:431. doi: 10.3389/fncel.2017.00431 inducing signal but is not essential for the stimulation of myelination. Glia. 53. Yang Y, Wang L, Bell P, McMenamin D, He Z, White J, et al. A dual AAV (2012) 60:203–17. doi: 10.1002/glia.21255 system enables the Cas9 mediated correction of a metabolic liver disease in 36. Issa AN, Zhan WZ, Sieck GC, Mantilla CB. Neuregulin-1 at synapses newborn mice. Nat Biotechnol. (2016) 34:334–8. doi: 10.1038/nbt.3469 on phrenic motoneurons. J Comp Neurol. (2010) 518:4213–25. 54. Bengtsson NE, Hall JK, Odom GL, Phelps MP, Andrus CR, Hawkins RD, doi: 10.1002/cne.22449 et al. Muscle-specific CRISPR/Cas9 dystrophin gene editing ameliorates 37. Gallart-Palau X, Tarabal O, Casanovas A, Sábado J, Correa FJ, Hereu M, et al. pathophysiology in a mouse model for Duchenne muscular dystrophy. Nat Neuregulin-1 is concentrated in the postsynaptic subsurface cistern of C- Commun. (2017) 8:14454. doi: 10.1038/ncomms14454 bouton inputs to α-motoneurons and altered during motoneuron diseases. 55. Mancuso R, Del Valle J, Morell M, Pallás M, Osta R, Navarro X. Lack of FASEB J. (2014) 28:3618–32. doi: 10.1096/fj.13-248583 synergistic effect of resveratrol and sigma-1 receptor agonist (PRE-084) in 38. Takahashi Y, Fukuda Y, Yoshimura J, Toyoda A, Kurppa K, Moritoyo H, SOD1G93A ALS mice: overlapping effects or limited therapeutic opportunity? et al. ErbB4 mutations that disrupt the neuregulin-ErbB4 pathway cause Orphanet J Rare Dis. (2014) 9:78. doi: 10.1186/1750-1172-9-78 amyotrophic lateral sclerosis type 19. Am J Hum Genet. (2013) 93:900–5. 56. Chen S, Hou Y, Zhao Z, Luo Y, Lv S, Wang Q, et al. Neuregulin-1 doi: 10.1016/j.ajhg.2013.09.008 accelerates functional motor recovery by improving motoneuron survival Frontiers in Neurology | www.frontiersin.org 11 September 2021 | Volume 12 | Article 693309 Mòdol-Caballero et al. Neuregulin Gene Therapy for ALS after brachial plexus root avulsion in mice. Neuroscience. (2019) 404:510–8. Conflict of Interest: The authors declare that the research was conducted in the doi: 10.1016/j.neuroscience.2019.01.054 absence of any commercial or financial relationships that could be construed as a 57. Gonzalez-Deniselle MC, López-Costa JJ, Saavedra JP, Pietranera L, Gonzalez potential conflict of interest. SL, Garay L, et al. Progesterone neuroprotection in the Wobbler mouse, a genetic model of spinal cord motor neuron disease. Neurobiol Dis. (2002) Publisher’s Note: All claims expressed in this article are solely those of the authors 11:457–68. doi: 10.1006/nbdi.2002.0564 and do not necessarily represent those of their affiliated organizations, or those of 58. Lacroix-Fralish ML, Tawfik VL, Nutile-McMenemy N, Deleo JA. Neuregulin the publisher, the editors and the reviewers. Any product that may be evaluated in 1 is a pronociceptive cytokine that is regulated by progesterone in the spinal this article, or claim that may be made by its manufacturer, is not guaranteed or cord: implications for sex specific pain modulation. Eur J Pain. (2008) 12:94– endorsed by the publisher. 103. doi: 10.1016/j.ejpain.2007.03.010 59. Alves CJ, de Santana LP, dos Santos AJ, de Oliveira GP, Duobles T, Scorisa JM, Copyright © 2021 Mòdol-Caballero, Herrando-Grabulosa, Verdés, García-Lareu, et al. Early motor and electrophysiological changes in transgenic mouse model Hernández, Francos-Quijorna, López-Vales, Bosch and Navarro. This is an open- of amyotrophic lateral sclerosis and gender differences on clinical outcome. access article distributed under the terms of the Creative Commons Attribution Brain Res. (2011) 1394:90–104. doi: 10.1016/j.brainres.2011.02.060 License (CC BY). The use, distribution or reproduction in other forums is permitted, 60. Liu J, Allender E, Wang J, Simpson EH, Loeb JA, Song F. Slowing provided the original author(s) and the copyright owner(s) are credited and that the disease progression in the SOD1 mouse model of ALS by blocking original publication in this journal is cited, in accordance with accepted academic neuregulin-induced microglial activation. Neurobiol. Dis. (2018) 111:118–26. practice. No use, distribution or reproduction is permitted which does not comply doi: 10.1016/j.nbd.2017.12.012 with these terms. Frontiers in Neurology | www.frontiersin.org 12 September 2021 | Volume 12 | Article 693309