Martínez-Muriana et al. Stem Cell Research & Therapy (2020) 11:53 https://doi.org/10.1186/s13287-020-1573-6 RESEARCH Open Access Combined intramuscular and intraspinal transplant of bone marrow cells improves neuromuscular function in the SOD1G93A mice Anna Martínez-Muriana1,2, Diego Pastor3,4, Renzo Mancuso1,5, Amaya Rando6, Rosario Osta6, Salvador Martínez4, Rubèn López-Vales1,2 and Xavier Navarro1,2,7* Abstract Background: The simultaneous contribution of several etiopathogenic disturbances makes amyotrophic lateral sclerosis (ALS) a fatal and challenging disease. Here, we studied two different cell therapy protocols to protect both central and peripheral nervous system in a murine model of ALS. Methods: Since ALS begins with a distal axonopathy, in a first assay, we performed injection of bone marrow cells into two hindlimb muscles of transgenic SOD1G93A mice. In a second study, we combined intramuscular and intraspinal injection of bone marrow cells. Fluorescence-activated cell sorting was used to assess the survival of the transplanted cells into the injected tissues. The mice were assessed from 8 to 16 weeks of age by means of locomotion and electrophysiological tests. After follow-up, the spinal cord was processed for analysis of motoneuron survival and glial cell reactivity. Results: We found that, after intramuscular injection, bone marrow cells were able to engraft within the muscle. However, bone marrow cell intramuscular injection failed to promote a general therapeutic effect. In the second approach, we found that bone marrow cells had limited survival in the spinal cord, but this strategy significantly improved motor outcomes. Moreover, we also found that the dual cell therapy tended to preserve spinal motoneurons at late stages of the disease and to reduce microgliosis, although this did not prolong mice survival. Conclusion: Overall, our findings suggest that targeting more than one affected area of the motor system at once with bone marrow cell therapy results in a valuable therapeutic intervention for ALS. Keywords: ALS, SOD1, Bone marrow, Stem cells, Motoneuron disease Introduction sporadic; however, several mutations have been identified Amyotrophic lateral sclerosis (ALS) is an adult-onset in familial cases of ALS (fALS). The most prevalent forms neurodegenerative disease characterized by a progressive include mutations encoded in the TARDBP (TDP43) gene, and selective death of upper and lower motoneurons the hexanucleotide repeat expansions in chromosome-9 (MNs) [1]. Resulting from neuronal loss, patients suffer open reading frame 72 (C9ORF72) or in the superoxide progressive atrophy and eventual muscle paralysis that dismutase gene (SOD1) [3]. Indeed, the most used experi- leads to their death a few years after the onset [2]. As in mental model of ALS is a transgenic mouse that overex- other neurodegenerative disorders, most ALS cases are presses a high-copy number of the mutated form of the human SOD1 gene [4]. This transgenic ALS model mimics * Correspondence: xavier.navarro@uab.cat the human pathology, being the most used preclinical tool 1 Institute of Neurosciences and Department of Cell Biology, Physiology and Immunology, Universitat Autònoma de Barcelona, Barcelona, Spain to study the disorder and test novel therapies [5]. The 2 Centro de Investigación Biomédica en Red sobre Enfermedades pathogenic mechanisms underlying MN death involve Neurodegenerativas (CIBERNED), Bellaterra, Spain alterations of multiple molecular mechanisms, including Full list of author information is available at the end of the article © The Author(s). 2020 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Martínez-Muriana et al. Stem Cell Research & Therapy (2020) 11:53 Page 2 of 11 glutamate-mediated excitotoxicity, defective axonal trans- 12:12-h light–dark cycle. Endpoint criterium was consid- port, or activation of neighboring glial cells [6–8]. The ered when animals lost the righting reflex for longer than simultaneous disruption of several molecular mechanisms 30 s. At 8 weeks of age (prior to BMC injection), animals and cell types raises the difficulty to find an effective treat- were electrophysiologically tested to obtain baseline levels. ment. In fact, novel therapies that only target a single SOD1G93A females and males were then distributed among factor have largely failed when translated into human the different experimental groups according to their pro- clinical trials [9]. These failures suggest that more effective genitors, weight, and electrophysiology baseline values in therapeutic approaches should act simultaneously on sev- balanced groups, receiving either BMCs or vehicle. eral targets in order to mediate global neuroprotection. In this way, cell therapy has emerged as a promising way to Bone marrow cells target several cells and mechanisms involved in ALS. BMCs were collected from female C57Bl6/J mice or GFP Bone marrow cells (BMCs) are easy and fast to isolate (green fluorescent protein) mutant mice (C57Bl/6-Tg and do not need further manipulation before their trans- [ACTB-EGFP]1Osb/J) [24]. Using the same procedures plant, thus being a very convenient source for cell therapy already described by our group [14, 15], femurs of these applications. BMCs are composed by several cell subsets mice were harvested and the attached muscles and con- that include hematopoietic progenitors and a heteroge- nective tissues were removed. The epiphyseal ends of neous population of mesenchymal stromal cells (MSCs) the bones were clipped and the marrow was extracted [10]. Several studies demonstrate that BMC transplantation using intraosseous perfusion with 2 ml of Dulbecco’s modi- ameliorates the progression of diverse neurodegenerative fied Eagle medium (DMEM, Gibco). Immediately, BMCs diseases, including ALS [11–13]. Similarly, previous reports were centrifuged at 1000 rpm at 4 °C for 5 min and concen- from our group have revealed that intramuscular injections trated to 35,000 cells/μl. For the identification of the cells of BMCs in mdf/ocd and SOD1G93A mice induce protec- after injection, we used GFP-labeled BMCs for long-term tion of spinal MNs and improve motor function [14, 15]. studies or labeled with PKH67 (Sigma), a cell membrane The exact molecular mechanisms underlying the beneficial labeling dye, for the short-term survival experiments. effects of BMC transplantation are still under debate; how- For transplantation, 8-week-old, prior to MN death and ever, they likely involve the immunomodulatory actions of when axonal degeneration starts [25], SOD1G93A mice BMCs and the release of neurotrophic factors [13]. Studies were anesthetized with ketamine (90 mg/kg, i.p.) and xyla- using BMCs as factor carriers have shown neuroprotection zine (10 mg/kg, i.p.). As described previously by our group when transplanted into the spinal cord of SOD1G93A mice [14, 15], for intramuscular injections, a total volume of and human patients [12, 16–19]. However, none of these 14 μl green-labeled BMC suspension (0.5 × 106 BMCs) treatments preserved neuromuscular function and stability was injected into quadriceps femoris (QF) and 7 μl of neuromuscular junctions (NMJs) at suitable levels [20]. (0.25 × 106 BMC) into tibialis anterior (TA) muscles Previous studies have attempted to protect both NMJs [15] bilaterally using a 27-G needle attached to a Hamilton and MNs in SOD1G93A mice [12, 21, 22] separately, show- syringe. For intraspinal injections, L4-L5 spinal cord ing transient mild effects on disease outcomes. Since ALS segments were exposed by means of a small laminectomy begins with a distal axonopathy that precedes MN soma and a volume of 5 μl was injected using a glass needle degeneration [23], in this work, we have attempted to (30 μm i.d.) coupled to a 10-μl Hamilton syringe. Injec- protect both MN soma, as well as, their terminal axons in tions were made at a perfusion speed of 2 μl/min con- skeletal muscle. Thus, we investigated the effects of the trolled by an automatic injector (KDS 310 Plus; KD combined injection of BMCs into hindlimb muscles Scientific), and the tip of the needle was maintained inside and in the spinal cord to protect NMJs and spinal cord the tissue 3 min after each injection to avoid liquid reflux. MNs, respectively, in SOD1G93A mice. Sham-treated SOD1G93A mice received the same volume injections of vehicle (DMEM). Materials and methods Animals The animals used in this work were transgenic mice carry- Functional evaluation ing the G93A human SOD1 mutation (B6SJL-Tg [SOD1- Locomotion, coordination, and muscle strength was G93A]1Gur) obtained from the Jackson Laboratory (Bar evaluated by rotarod test. Prior to cell injections, all Harbor, ME, USA) and provided from the colony main- animals were trained twice per week on the rotating rod tained at the University of Zaragoza. For phenotyping, at a constant speed of 14 rpm. All animals used in this DNA was extracted from tail samples and further analyzed study reached baseline levels (an arbitrary time of 180 s). by PCR to identify hemizygous transgenic mice and wild- After injections, BMC- or sham-treated SOD1G93A mice type littermates. Mice were maintained with food and were evaluated in the rotarod once per week until 16 water ad libitum at room temperature of 22 ± 2 °C under a weeks of age. The time each animal was able to remain Martínez-Muriana et al. Stem Cell Research & Therapy (2020) 11:53 Page 3 of 11 on the rotarod turning at a constant speed of 14 rpm For immunostainings, slides were placed over a hotplate was measured. at 37 °C for 30 min. Then, samples were rehydrated in Locomotion was further analyzed using a speed- PBS for 5 min and further blocked with a blocking solu- controlled treadmill (Digigait, Mouse Specifics). To tion of 5% NDS in PBS-T 0.1% at room temperature. evaluate the maximum speed that SOD1G93A mice Once blocked, sections were incubated overnight at were able to run at the end-stage of the disease, 4 °C with anti-Iba1 (Wako, 1:1000) and anti-GFAP animals were placed on the treadmill belt and their (Dako, 1:1000). Sections were then washed in PBS-T capacity to run with increasing velocities (5, 10, 15, and further incubated with a specific secondary anti- 20, 25, and 30 cm/s) was recorded [26]. body bound to an Alexa-594 or -488 fluorocrom (1: Electrophysiological tests were used to evaluate motor 500, Invitrogen) for 1 h. function every 2 weeks from 8 to 16 weeks of age. Nerve conduction test was performed by stimulating the sciatic Flow cytometry nerve (square pulses of 20-μs duration, Grass S88) through Fluorescence-activated cell sorting (FACS) was used a pair of needles located percutaneously in the sciatic to characterize BMCs and to assess the survival of notch. Compound muscle action potentials (CMAPs) were the transplanted cells into the target tissues. For the recorded in sciatic nerve muscle targets: TA and gastrocne- immunophenotyping, BMCs were obtained from fe- mius medialis muscle (GM) with microneedle electrodes murs of C57Bl6 mice as indicated above. Then, red placed on the muscle [25]. All recordings were amplified blood cells (RBC) were lysed using RBC lysis Buffer and visualized in a digital oscilloscope (Tektronix 450S). (10x) following the manufacturer’s instructions (Biole- Measurements of the CMAP amplitude were obtained gend, 420301). Cell suspension was further washed from the baseline to the maximal negative peak and the with FACS buffer (DMEM + 10% FBS) and centri- latency from stimulus to the onset of the first negative fuged twice at 300×g for 10 min at 4 °C. Samples were deflection. The recording needles were placed under a further split in several tubes and immunostained. Pri- microscope, guided by anatomical landmarks, to ensure mary antibody labeling was performed for 1 h at 4 °C reproducibility of needle location on all animals. During in FACS buffer with the following conjugated anti- the tests, the mouse body temperature was maintained bodies: mouse lineage antibody cocktail (Lin)-PerCP- using a controlled heating pad. Cy5.5 (1:200, 561317, BD Biosciences), Sca-1-PE-Cy7 (1:200, 561021, BD Biosciences), c-Kit-APC (1:200, Histology 561074, BD Biosciences), and CD34 (1:200, 560238, At 16 weeks of age, BMC or SHAM SOD1G93A and BD Biosciences). For the gating strategy, we set up wildtype littermate mice were euthanized with an the cut off based on the corresponding isotypes con- overdose of sodium pentobarbital. Immediately, ani- trol expression (Additional file 1: Figure S1A). For the mals were transcardially perfused with a solution of analysis, cells were first gated based on size (FSC) 4% paraformaldehyde (PFA) in 0.1 M phosphate buffer and complexity (SSC) to remove all the debris. Cells (PB) and tissue samples were harvested. Treated and were further gated based on Lin expression to dis- control muscles were cryopreserved and maintained criminate between progenitors (Lin−) and lineage ma- in a solution of 30% sucrose with azide in 0.1 M PB ture cells (Lin+). The percentage of mesenchymal at 4 °C. Spinal cord lumbar segments were harvested stromal cells (MSC) was calculated by counting all and post-fixed in 4% PFA for 2 h, and then cryopre- the Lin−, Sca-1+, c-Kit− and CD34− cells [27]. served and maintained at 4 °C with 30% sucrose and For the in vivo evaluation of BMC engraftment, ani- azide in 0.1 M PB. mals were sacrificed and transcardially perfused with 60 Lumbar spinal cords were serially cut in transverse ml of 0.9% sterile saline solution, and spinal cords and sections of 40-μm thickness using a cryostat (Leica). hindlimb muscles were rapidly collected and kept in Free-floating series of 10 sections were sequentially col- DMEM. Samples were further mechanically broken up lected and kept in Olmos solution at − 20 °C. To assess using a syringe and passed through a 40-μm cell strainer MN survival, one series was rehydrated and stained for (BD2 Falcon) with DMEM to disaggregate the cells. To 3 h with a 3.1 mM Cresyl violet solution. Slices were eliminate debris, the obtained cell suspension was centri- then washed, dehydrated, and mounted with DPX fuged twice at 300×g for 10 min at 4 °C. Centrifuged cells (Fluka). L4-L5 MNs were localized by morphology and were further washed, and the obtained pellet was sus- size. Only MNs that were polygonal-shaped, larger than pended and fixed in a 1% PFA solution and immediately 20 μm in diameter and with well-stained nucleoli were analyzed in a FACSCanto flow cytometer (BD Biosci- counted. MNs present in the lateral side of both ven- ences). Grafted cells were distinguished from tissue resi- tral horns were quantified in four serial sections of dent cells using the FITC channel where both GFP and the L4-L5 segment [25]. PKH67 dyes are detected. Martínez-Muriana et al. Stem Cell Research & Therapy (2020) 11:53 Page 4 of 11 Statistics and CD34−), which contain an heterogenous population Data are represented as mean ± SEM and analyzed using of progenitors and stem cells [10], only represented ~ 4% the GraphPad Prism 6 software package (GraphPad Soft- of the whole bone marrow content (Fig. 1a, b). ware). Locomotion and electrophysiological results were analyzed using two-way repeated measurements ANOVA BMCs do not improve neuromuscular function in grafted with Tukey post-hoc test for multiple comparisons. For muscle of SOD1G93A mice histological data, two-tailed t-Student test was used. Sur- Prior to assess the therapeutic effect of bone marrow cell vival and forced locomotion results were analyzed using injection, we evaluated the survival of the cell grafting the Mantel–Cox test. Differences were considered statisti- within the muscle (Fig. 2a). We injected bone marrow cally significant at p < 0.05. green-labeled cells in hindlimbs muscles of SOD1G93A mice and performed flow cytometry at 7 and 21 days Results post-injection (dpi). FACS analysis revealed that green- Bone marrow cells immunophenotypic characterization labeled BMCs were engrafted within the TA muscle for Adult mouse bone marrow is composed by two main stem at least 3 weeks post-injection (Fig. 2b, c). cell lineage populations: hematopoietic stem cells (HSCs) Once determined that BMCs were able to survive in and mesenchymal stromal cells (MSCs). Previous reports the muscle, we tested their therapeutic effects in the have demonstrated MSCs therapeutic properties in neuro- SOD1G93A mouse model of ALS. Bilateral injections of degenerative diseases, including ALS [13]. However, the BMCs or vehicle were performed in the TA and QF MSC population in the mouse bone marrow is quite lim- muscles of SOD1G93A mice at 8 weeks of age (Fig. 3a). ited. Here, using FACS immunophenotyping, we found Electrophysiological tests showed that BMC injection that mouse bone marrow is mainly composed by mature did not improve the CMAP amplitude compared to the lineage hematopoietic cells (~ 34%, Lin+) and progenitor DMEM-injected animals, neither in grafted muscles cells (~ 55%, Lin−) (Fig. 1a, b). MSCs (Lin−, Sca-1+, c-Kit− (TA) nor in non-injected gastrocnemius medialis (GM) Fig. 1 Bone marrow cells FACS characterization. a Representative FACS plots showing the gating strategy to quantify the amount of hematopoietic lineage cells (Lin+), the precursor cells (Lin−) and the MSCs (Lin−, Sca-1+, c-Kit−, CD34−). b Bar plot showing the percentage of each population within the mouse bone marrow Martínez-Muriana et al. Stem Cell Research & Therapy (2020) 11:53 Page 5 of 11 Fig. 2 Bone marrow cells survival after transplantation in hindlimb muscles of SOD1G93A mice. a Representative drawing showing the first study design. b, c Representative FACS plots showing green-labeled cells in muscle at 7 days (b) and 21 days (c) after BMC injection in the tibialis anterior (TA) muscle muscles, used as internal control. Besides, bilateral BMC (Fig. 5b), but they were not present in the spinal cord injections were not enough to improve the overall motor parenchyma at 3 weeks post-injection (Fig. 5c). outcome, neither in the rotarod nor in the treadmill Despite having a limited survival, we found that the tests (Fig. 3b). Then, we aimed to know whether BMC combined intramuscular and intraspinal BMC injection intramuscular injections were able to protect spinal significantly preserved the CMAP amplitude of both MNs, as reported in a previous study [14]. We found treated and untreated muscles, which are innervated by that BMC-treated animals had 10% higher number of spinal cord-grafted segments (Fig. 6a). Indeed, BMC injec- surviving MNs compared to SHAM mice, although tion into the spinal cord and muscles further improved this difference did not reach statistical significance motor outcome in SOD1G93A mice as revealed by better (p = 0.0692) (Fig. 4a–d). performance in the rotarod and forced locomotion tests (Fig. 6b, c). Due to the differences in disease progression Intraspinal and intramuscular BMC injections preserve between genders [25], we also performed the dual injec- lower motoneuron function in SOD1G93A mice tion of BMCs in SOD1G93A male mice. Similar to female To further protect spinal MNs, we injected BMCs into animals, BMC-injected male mice also showed improved both L4-L5 lumbar spinal cord and hindlimb muscles CMAP amplitude of TA muscles, although it was not sta- (Fig. 5a). Firstly, we assessed the survival of the cells after tistically significant, probably due to the small sample size spinal injection. Similar to the muscle injection, BMCs (Fig. 6d). Finally, we assessed the survival of SOD1G93A were successfully detected for 1 week after transplantation mice after the combined BMC injection. Since there were Martínez-Muriana et al. Stem Cell Research & Therapy (2020) 11:53 Page 6 of 11 Fig. 3 Effects of bone marrow cells injection in hindlimb muscles of SOD1G93A mice. a Electrophysiological analysis of BMC injection effect in treated muscles (tibialis anterior) and non-treated muscles (Gastrocnemius medialis) (n = 16 SHAM SOD1G93A vs. 17 BMC-treated SOD1G93A). b Gait and locomotion analysis by rotarod (n = 15 SHAM SOD1G93A vs. 16 BMCs-treated SOD1G93A) and forced locomotion test (n = 8 SHAM SOD1G93A vs. 11 BMCs-treated SOD1G93A) after injection of BMCs in hindlimb muscles. Error bars indicate SEM Fig. 4 Analysis of MN survival after injection of BMCs in hindlimb muscles of SOD1G93A mice. a–c Representative micrographs showing lumbar spinal MNs in a WT, b SHAM SOD1, or c BMC-treated SOD1 mice. Scale bar = 50 μm. d Quantification of MN sparing after treatment with BMCs (n = 7 SHAM SOD1G93A vs. 9 BMCs-treated SOD1G93A). Dotted line indicates the average number of MNs in WT mice. Error bars indicate SEM Martínez-Muriana et al. Stem Cell Research & Therapy (2020) 11:53 Page 7 of 11 Fig. 5 Bone marrow cells survival after injection into the spinal cord of SOD1G93A mice. a Illustrative image representing the second therapeutic approach. b, c Representative dotted FACS plots displaying green-labeled BMCs in the lumbar spinal cord at 7 dpi (b) and 21 dpi (c) in SOD1G93A mice no significant differences between genders, we pooled the intramuscular injections) but these results were not sta- lifespan of males and females. Even though the combined tistically significant. Since non-neuronal cells actively BMCs-treated animals showed greater preservation of contribute to MN degeneration, we analyzed the reactiv- muscular function at early stages of the disease, the treat- ity of astrocytes and microglial cells in the ventral horn ment failed to increase mice lifespan (Fig. 6e). of lumbar cord segments. Immunolabeling revealed that the combined BMC injections tend to reduce microglial BMC intraspinal and intramuscular injections tend to but not astroglial immunoreactivity in the ventral horn preserve spinal MNs and to reduce microglial reactivity of treated mice (Fig. 7e–j). We then investigated whether the combined BMC injec- tion preserved L4-L5 spinal MNs (Fig. 7a–d). We found Discussion that the dual injection of BMCs increased the percentage The complex pathophysiology of ALS implies a huge dif- of MNs spared compared to intramuscular injection ficulty for finding successful therapies. Indeed, most pre- (52% intramuscular injection vs. 67% intraspinal and clinical therapies have failed to translate into human Martínez-Muriana et al. Stem Cell Research & Therapy (2020) 11:53 Page 8 of 11 Fig. 6 Evaluation of motor outcomes after intraspinal and intramuscular BMC injection in SOD1G93A mice. a Electrophysiological evaluation of the compound muscle action potential (CMAP) in BMC-treated muscle (tibialis anterior) and non-treated muscle (Gastrocnemius medialis) in females (n = 8 SHAM SOD1G93A vs. 12 BMCs-treated SOD1G93A). b Evaluation of the locomotion by Rotarod test and c forced locomotion in the Digigait (n = 10 SHAM SOD1G93A vs. 12 BMC-treated SOD1G93A). d TA CMAP amplitude in male SOD1G93A mice after BMC treatment (n = 5 per group). e Evaluation of the maximum lifespan of SOD1G93A mice after combined intraspinal and intramuscular BMC injection (n = 16 SHAM SOD1G93A vs. 18 BMC-treated SOD1G93A). *p < 0.05 vs. SHAM SOD1G93A mice. Error bars indicate SEM trials due to the wide variety of factors involved in this and a tendency to preserve spinal MNs but did not pro- neurodegenerative disease [9]. For this reason, ap- long mice survival. proaches that simultaneously act on different targets in- ALS is a complex neurodegenerative disease where dis- volved in ALS pathophysiology may provide better ruption of multiple cellular functions leads to progressive chances for an effective treatment for patients. There- loss of MN. At the central level, MNs die due to the alter- fore, in this work, we aimed to study different protocols ation of molecular pathways such as autophagy, aggregation of BMC transplant in SOD1G93A mice. We found that of misfolded proteins, or glutamate-mediated excitotoxicity BMCs transplanted into two SOD1G93A hindlimb mus- [8]. However, MN intrinsic pathways are not the only pro- cles failed to induce general motor improvement and cesses involved in MN degeneration. Indeed, different re- protection of spinal MNs. However, when these BMC ports point to the contribution of non-neuronal cells, transplants were combined with injection of BMCs into highlighting the importance of glial and immune responses the lumbar spinal cord, the therapeutic effects of BMCs as direct players in the death of MNs [28]. In this sense, were boosted, resulting in enhanced locomotor skills peripheral nervous system neuroimmune alterations are Martínez-Muriana et al. Stem Cell Research & Therapy (2020) 11:53 Page 9 of 11 Fig. 7 Evaluation of MN preservation and glial reactivity in the lumbar spinal cord after combined BMC injection. a, b, c. Representative photographs of lumbar spinal cord sections showing L4 MNs of a WT, b BMC-treated SOD1G93A mice, and c SHAM SOD1G93A mice. d Quantification of MN survival after injection of BMCs in spinal cord and hindlimb muscles. Dotted line indicates the mean number of MNs in wild type mice. e, f, h, i Illustrative micrographs showing immunoreactivity against Iba-1 (green, microglia) or GFAP (red, astrocytes) at 16 weeks of age in BMCs-treated SOD1G93A mice and SHAM SOD1G93A mice. Quantification of g microglia and j astrocytes after injection of BMCs (n = 6) or vehicle (n = 4) in SOD1G93A mice. Error bars indicate SEM also found in ALS [29, 30]. In fact, at the peripheral level, periods. Although in a previous study we showed that and prior to neuronal degeneration, spinal MNs disconnect some cells were still grafted after 2 months post- from their target muscles in a process known as “dying- injections, the number of cells at that stage was scarce back” [23, 30]. As a result, the maintenance of the NMJs is [14]. Therefore, it is expected that greater beneficial ef- impaired and peripheral axons retract and progressively fects could be achieved by doing multiple BCM injec- degenerate. Due to the complexity of ALS disease, which tions over disease progression, as well as, by increasing involves events that occur in the central and peripheral the number of grafted muscles. systems, as well as in multiple cell types, it is necessary to Despite the limited functional improvement observed develop treatments that can target more than one of these after intramuscular injection of BMCs, when quantifying pathological episodes. the number of MNs in the lumbar spinal cord, we found Cell therapy has emerged as a promising approach to a tendency to have higher number of surviving neurons treat multiple disorders due to their potent and diverse in BMC-injected mice. In previous reports, we revealed beneficial effects [11–13]. BMC transplantation has that BMCs are able to produce GDNF, which can be shown promising results in several animal models of retrogradely transported to the neuronal soma and may ALS [12, 14, 15, 21]. Since ALS begins with a distal account for the 10% increased MN survival observed axonopathy that rapidly progresses retrogradely to the after muscle grafting of BMC [14, 15]. However, this neuronal soma, we therefore aimed to first protect the beneficial effect was insufficient to induce extensive NMJs by transplanting BMCs into the hindlimb muscles neuronal sparing. Indeed, the functional improvement at pre-symptomatic stages of the disease. Here we found in the previous study in SOD1G93A mice was mild showed that BMCs survived for at least 3 weeks after the [15], and we have not been able to reproduce it under injection, which is in accordance with previous reports slightly different conditions, such as age at injection, revealing the presence of BMCs in the muscle tissue be- housing laboratory, and animals colony. A recent report tween 1 and 2 months after injection [14, 15]. Similar to showed that, whereas BMCs injected intramuscularly or a recent study [21], we found that intramuscular injec- intravenously did not produce significant improvements tion of BMCs did not preserve neuromuscular integrity in the SOD1G93A mice, simultaneous injections in the nor was translated into general improvement of loco- muscle and intravenously delayed the onset of disease motor outcomes. This lack of therapeutic effect may be and decreased microgliosis although this did not pre- explained because of several factors. First, we only serve lumbar ventral horn MNs [21], For these reasons, treated QF and TA muscles, which are not the only we decided to inject BMCs in the hindlimb muscles and effectors involved in locomotion. Second, the limited also into the lumbar spinal cord of SOD1G93A mice. We grafting of the cells could restrict the therapeutic effects found that, in contrast to the skeletal muscles, the sur- to the first weeks after injection. We found that BMCs vival of the BMCs within the spinal cord was limited to remained inside the muscle at least for 3 weeks after in- just a few days. These results are consistent with previ- jections, but they are unlikely to survive for much longer ous reports indicating that survival of grafted cells in the Martínez-Muriana et al. Stem Cell Research & Therapy (2020) 11:53 Page 10 of 11 spinal cord is strongly compromised after 1 week [12, that this could be achieved by repeating the cell injec- 22, 31]. This is likely due to the immune reaction found tion at different time-points over the disease progres- within the nervous system of the ALS mice [28]. Since sion, in multiple muscles and areas of the CNS. Thus, microglial cells become activated at pre-symptomatic our work suggests that combining muscle and spinal stages of the disease, they may rapidly phagocytose the BMC injections represents an interesting cell therapy injected cells [29, 32]. Indeed, this immune response approach for treating ALS. may account for the difference in the survival rate of the grafted BMCs in the spinal cord as compared to the Supplementary information skeletal muscle. At peripheral level, infiltration of circu- Supplementary information accompanies this paper at https://doi.org/10. lating monocytes occurs even prior to microgliosis [29]. 1186/s13287-020-1573-6. However, they accumulate within the nerve bundles but Additional file 1: Figure S1. FACS plots gating and isotype controls. not at the NMJs or muscle fibers [30]. This less promin- Representative FACS plots showing the gating cut off for each primary ent immune reaction in the muscle could increase the antibody. survival of BMCs. Despite BMCs having a short-time survival in the spinal cord, injection of BMCs into the Abbreviations lumbar cord and hindlimb muscles improved the global ALS: Amyotrophic lateral sclerosis; BMC: Bone marrow cells; motor function of SOD1G93A mice. This new approach CMAP: Compound muscle action potential; DMEM: Dulbecco’s modified Eagle medium; dpi: Days post-injection; FACS: Fluorescent-activated cell led to ~ 15% greater preservation of spinal MNs at 16 sorting; fALS: Familial ALS; GFP: Green fluorescent protein; weeks of age as compared to control mice and doubled GM: Gastrocnemius medialis muscle; MN: Motoneuron; NMJ: Neuromuscular the beneficial actions observed after transplantation of junction; PB: Phosphate buffer; PFA: Paraformaldehyde; QF: Quadriceps femoris; SEM: Standard error of the mean; SOD1: Superoxide dismutase 1; BMCs into the hindlimb muscles. Even though we can- TA: Tibialis anterior not rule out the possibility that the beneficial effects found when combining both injections rely only in the Authors’ contributions AMM, DP, AR, and RM performed experimental work in the study. RO intraspinal approach, previous reports support that the provided the experimental animals. RLV, SM, and XN designed and combination of different BMC administration routes is supervised the study. AMM drafted the first version of the paper. DP, SM, needed to reach therapeutic effects in SOD1G93A mice RLV, RO, RM, and XN revised and drafted the final version of the article. All authors read and approved the final manuscript. [21]. In line with these findings, other studies have re- vealed that intraspinal BMC injection in SOD1G93A mice Funding leads to a mild and transitory delay in disease progres- This study was supported by the Spanish State Research Agency “Severo sion [12, 22], strongly suggesting that our findings are Ochoa” Program for Centers of Excellence in R&D (SEV-2013-0317) and Instituto de Salud Carlos III Red de Terapia Celular, TERCEL, co-funded by the likely due to the concomitant effects of the intramuscu- European Union (ERDF/ESF, “Investing in your future”). lar and intraspinal injections. This overall therapeutic ef- fect might be explained by the production of growth Availability of data and materials factors, as well as, by the immunomodulatory effects of The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request. the BMCs [21, 33]. Indeed, we found a marked reduction in microgliosis in the spinal cord of SOD1 mice after Ethics approval and consent to participate intraspinal injection of BMCs, even though the results All the experimental procedures were approved by the Ethical Committee of did not reach statistical significance. Previous reports the Universitat Autònoma de Barcelona and followed the European Communities Council Directive 2010/63/EU, and the methods for each have revealed that microgliosis is a key factor contribut- procedure were performed in accordance with the approved guidelines and ing to MN death [29, 34]. Therefore, it is not surprising regulations. to observe greater MN survival by reducing microgliosis. Despite the beneficial actions of this dual transplant- Consent for publication Not applicable ation, the survival of mice was not prolonged, which is likely due to the protective effects of the BMCs at the Competing interests site of the transplant (lumbar region) but not in those The authors declare that they have no competing interests. CNS areas where they were not grafted. Author details 1 Institute of Neurosciences and Department of Cell Biology, Physiology and Conclusion Immunology, Universitat Autònoma de Barcelona, Barcelona, Spain. 2Centro Overall, our study indicates that combining intramus- de Investigación Biomédica en Red sobre Enfermedades Neurodegenerativas (CIBERNED), Bellaterra, Spain. 3Centro de Investigación Deportiva, Universidad cular injection of BMCs with BMC transplants in the Miguel Hernández, Elche, Spain. 4Instituto de Neurociencias, UMH-CSIC, San spinal cord enhanced the therapeutic actions of the Juan de Alicante, Spain. 5VIB Center for Brain and Disease Research, KU cells and improved motor function and preservation Leuven, Leuven, Belgium. 6Laboratory of Genetic Biochemistry (LAGENBIO), Health Research Institute of Aragón, Universidad de Zaragoza, Zaragoza, of spinal MNs. Although this injection protocol is in- Spain. 7Faculty of Medicine, Universitat Autònoma de Barcelona, 08193 sufficient to increase mice survival, we do not discard Bellaterra, Spain. Martínez-Muriana et al. Stem Cell Research & Therapy (2020) 11:53 Page 11 of 11 Received: 22 October 2019 Revised: 13 January 2020 a mouse model of amyotrophic lateral sclerosis: Functional outcomes from Accepted: 27 January 2020 different administration routes. Brain Res. 2019;1712:73–81. 22. Bursch F, Rath KJ, Sarikidi A, Böselt S, Kefalakes E, Osmanovic A, et al. Analysis of the therapeutic potential of different administration routes and References frequencies of human mesenchymal stromal cells in the SOD1 G93A mouse 1. Bruijn LI, Miller TM, Cleveland DW. Unraveling the mechanisms involved in model of amyotrophic lateral sclerosis. J Tissue Eng Regen Med. 2019;13: motor neuron degeneration in ALS. Annu Rev Neurosci. 2004;27:723–49. 649–63. 2. Wijesekera LC, Leigh PN. Amyotrophic lateral sclerosis. Orphanet J Rare Dis. 23. Dadon-Nachum M, Melamed E, Offen D. The “dying-back” phenomenon of 2009;4:3. motor neurons in ALS. J Mol Neurosci. 2011;43:470–7. 3. Renton AE, Chio A, Traynor BJ. State of play in amyotrophic lateral sclerosis 24. Okabe M, Ikawa M, Kominami K, Nakanishi T, Nishimune Y. “Green mice” as genetics. Nat Neurosci. 2014;17:17–23. a source of ubiquitous green cells. FEBS Lett. 1997;407:313–9. 4. Gurney M, Pu H, Chiu A, Dal Canto M, Polchow C, Alexander D, et al. Motor 25. Mancuso R, Santos-Nogueira E, Osta R, Navarro X. Electrophysiological neuron degeneration in mice that express a human Cu,Zn superoxide analysis of a murine model of motoneuron disease. Clin Neurophysiol. 2011; dismutase mutation. Science. 1994;264:1772–5. 122:1660–70. 5. Ripps ME, Huntley GW, Hof PR, Morrison JH, Gordon JW. Transgenic mice 26. Mancuso R, Olivan S, Osta R, Navarro X. Evolution of gait abnormalities in expressing an altered murine superoxide dismutase gene provide an animal SOD1(G93A) transgenic mice. Brain Res. 2011;1406:65–73. model of amyotrophic lateral sclerosis. Proc Natl Acad Sci U S A. 1995;92: 27. Boxall SA, Jones E. Markers for characterization of bone marrow 689–93. multipotential stromal cells. Stem Cells Int. 2012;2012:975871. 6. Murray LM, Talbot K, Gillingwater TH. Review: Neuromuscular synaptic 28. Philips T, Robberecht W. Neuroinflammation in amyotrophic lateral sclerosis: vulnerability in motor neurone disease: Amyotrophic lateral sclerosis and Role of glial activation in motor neuron disease. Lancet Neurol. 2011;10: spinal muscular atrophy. Neuropathol Appl Neurobiol. 2010;36:133–56. 253–63. 7. Ferraiuolo L, Kirby J, Grierson AJ, Sendtner M, Shaw PJ. Molecular pathways 29. Martínez-Muriana A, Mancuso R, Francos-Quijorna I, Olmos-Alonso A, Osta R, of motor neuron injury in amyotrophic lateral sclerosis. Nat Rev Neurol. Perry VH, et al. CSF1R blockade slows the progression of amyotrophic lateral 2011;7:616–30. sclerosis by reducing microgliosis and invasion of macrophages into 8. Mancuso R, Navarro X. Amyotrophic lateral sclerosis: Current perspectives peripheral nerves. Sci Rep. 2016;6:1–13. from basic research to the clinic. Prog Neurobiol. 2015;133:1–26. 30. Chiu IM, Phatnani H, Kuligowski M, Tapia JC, Carrasco MA, Zhang M, et al. 9. Benatar M. Lost in translation: Treatment trials in the SOD1 mouse and in Activation of innate and humoral immunity in the peripheral nervous human ALS. Neurobiol Dis. 2007;26:1–13. system of ALS transgenic mice. Proc Natl Acad Sci U S A. 2009;106:20960–5. 10. Squillaro T, Peluso G, Galderisi U. Clinical trials with mesenchymal stem cells 31. Torres-Espín A, Redondo-Castro E, Hernández J, Navarro X. Bone marrow : An update. Cell Transplant. 2016;25:829–48. mesenchymal stromal cells and olfactory ensheathing cells transplantation 11. Zaverucha-do-Valle C, Mesentier-Louro L, Gubert F, Mortari N, Padilha AB, after spinal cord injury–a morphological and functional comparison in rats. Paredes BD, et al. Sustained effect of bone marrow mononuclear cell Eur J Neurosci. 2014;39:1704–17. therapy in axonal regeneration in a model of optic nerve crush. Brain Res. 32. Gravel M, Beland L-C, Soucy G, Abdelhamid E, Rahimian R, Gravel C, et al. IL- 2014;1587:54–68. 10 controls early microglial phenotypes and disease onset in ALS caused by 12. Gubert F, Decotelli AB, Bonacossa-Pereira I, Figueiredo FR, Zaverucha-do- misfolded superoxide dismutase 1. J Neurosci. 2016;36:1031–48. Valle C, Tovar-Moll F, et al. Intraspinal bone-marrow cell therapy at pre- and 33. Gao F, Chiu SM, Motan DAL, Zhang Z, Chen L, Ji H-L, et al. Mesenchymal symptomatic phases in a mouse model of amyotrophic lateral sclerosis. stem cells and immunomodulation: Current status and future prospects. Stem Cell Res Ther. 2016;7:41. Cell Death Dis. 2016;7:e2062. 13. Volkman R, Offen D. Concise review: Mesenchymal stem cells in 34. Frakes AE, Ferraiuolo L, Haidet-Phillips AM, Schmelzer L, Braun L, Miranda CJ, neurodegenerative diseases. Stem Cells. 2017;35:1867–80. et al. Microglia induce motor neuron death via the classical NF-kappaB 14. Pastor D, Viso-Leon MC, Botella-Lopez A, Jaramillo-Merchan J, Moraleda JM, pathway in amyotrophic lateral sclerosis. Neuron. 2014;81:1009–23. Jones J, et al. Bone marrow transplantation in hindlimb muscles of motoneuron degenerative mice reduces neuronal death and improves motor function. Stem Cells Dev. 2013;22:1633–44. Publisher’s Note 15. Rando A, Pastor D, Viso-león MC, Martínez A, Manzano R, Navarro X, et al. Springer Nature remains neutral with regard to jurisdictional claims in Intramuscular transplantation of bone marrow cells prolongs the lifespan of published maps and institutional affiliations. SOD1 G93A mice and modulates expression of prognosis biomarkers of the disease. Stem Cell Res Ther. 2018;9(1):90. 16. Pastor D, Viso-Leon MC, Jones J, Jaramillo-Merchan J, Toledo-Aral JJ, Moraleda JM, et al. Comparative effects between bone marrow and mesenchymal stem cell transplantation in GDNF expression and motor function recovery in a motorneuron degenerative mouse model. Stem Cell Rev. 2012;8:445–58. 17. Blanquer M, Moraleda JM, Iniesta F, Gómez-Espuch J, Meca-Lallana J, Villaverde R, et al. Neurotrophic bone marrow cellular nests prevent spinal motoneuron degeneration in amyotrophic lateral sclerosis patients: A pilot safety study. Stem Cells. 2012;30:1277–85. 18. Ruiz-Lopez FJ, Guardiola J, Izura V, Gomez-Espuch J, Iniesta F, Blanquer M, et al. Breathing pattern in a phase I clinical trial of intraspinal injection of autologous bone marrow mononuclear cells in patients with amyotrophic lateral sclerosis. Respir Physiol Neurobiol. 2016;221:54–8. 19. García Santos JM, Blanquer M, Torres del Río S, Iniesta F, Espuch JG, Pérez- Espejo MÁ, et al. Acute and chronic MRI changes in the spine and spinal cord after surgical stem cell grafting in patients with definite amyotrophic lateral sclerosis: Post-infusion injuries are unrelated with clinical impairment. Magn Reson Imaging. 2013;31:1298–308. 20. Suzuki M, McHugh J, Tork C, Shelley B, Klein SM, Aebischer P, et al. GDNF secreting human neural progenitor cells protect dying motor neurons, but not their projection to muscle, in a rat model of familial ALS. PLoS One. 2007;2:e689. 21. Gubert F, Bonacossa-Pereira I, Decotelli AB, Furtado M, Vasconcelos-dos- Santos A, Mendez-Otero R, et al. Bone-marrow mononuclear cell therapy in