Blog

By Núria Gaja-Capdevila, Neus Hernández, Xavier Navarro, and Mireia Herrando-Grabulosa

Excerpt from the article published in Frontiers in Pharmacology, 10 December 2021 | https://doi.org/10.3389/fphar.2021.780588

Editor’s Highlights

• Amyotrophic lateral sclerosis (ALS) is a fatal neurodegenerative disorder characterized by the progressive loss of upper and lower motoneurons (MNs), causing muscle paralysis and early death. 
• The etiology underlying the development of ALS remains poorly understood, with evidence supporting the presence of autoimmune mechanisms, but abnormal protein aggregation and altered proteostasis are common features of sporadic and familial ALS forms.
• Several lines of evidence indicate that protein aggregates are not just a consequence of neuronal degeneration, but rather that disturbed proteostasis plays a fundamental role early in the pathogenesis of proteinopathies.
• Enhancement of autophagy, an intracellular lysosome degradation system responsible for the clearance of cytoplasmic components and organelles, has been reported in ALS with alterations in several steps.
• The unfolded protein response (UPR) and the autophagy pathway have been linked; thus, XBP1 conditional deletion in the nervous system ameliorated pathogenesis.
• Sig-1R agonist was able to preserve neuromuscular function of the hindlimbs and increased the number of surviving MNs in the treated female SOD1^G93A mice. 
• Sig-1R ligands are promising tools for ALS treatment, although more research is needed to ascertain their mechanisms of action.

Abstract

Amyotrophic Lateral Sclerosis (ALS) is a neurodegenerative disorder characterized by the death of motoneurons (MNs) with a poor prognosis. There is no available cure, thus, novel therapeutic targets are urgently needed. Sigma-1 receptor (Sig-1R) has been reported as a target to treat experimental models of degenerative diseases and, importantly, mutations in the Sig-1R gene cause several types of motoneuron disease (MND). In this study we compared the potential therapeutic effect of three Sig-1R ligands, the agonists PRE-084 and SA4503 and the antagonist BD1063, in the SOD1^G93A mouse model of ALS. Pharmacological administration was from 8 to 16 weeks of age, and the neuromuscular function and disease progression were evaluated using nerve conduction and rotarod tests. At the end of follow up (16 weeks), samples were harvested for histological and molecular analyses. The results showed that PRE-084, as well as BD1063 treatment was able to preserve neuromuscular function of the hindlimbs and increased the number of surviving MNs in the treated female SOD1^G93A mice. SA4503 tended to improve motor function and preserved neuromuscular junctions (NMJ), but did not improve MN survival. Western blot analyses revealed that the autophagic flux and the endoplasmic reticulum stress, two pathways implicated in the physiopathology of ALS, were not modified with Sig-1R treatments in SOD1^G93A mice. In conclusion, Sig-1R ligands are promising tools for ALS treatment, although more research is needed to ascertain their mechanisms of action.

Introduction

Amyotrophic lateral sclerosis (ALS) is a fatal neurodegenerative disorder characterized by the progressive loss of upper and lower motoneurons (MNs), causing muscle paralysis and early death. Despite persistent efforts to develop treatments for this disease, no effective cure is available for ALS patients. Riluzole and edaravone are the only drugs approved by the FDA, but they have limited therapeutic benefits increasing the lifespan a few months (Edaravone (MCI-186) ALS 19 Study Group, 2017; Ludolph and Jesse, 2009). The exact pathophysiological mechanisms contributing to MN degeneration in ALS remain unclear. Nevertheless, the generation of transgenic animal models carrying ALS-related mutations has accelerated the research on physiopathology and preclinical therapeutic assays for ALS. Since the first mutation identified in familiar ALS was in the SOD1 gene (Rosen et al., 1993), the SOD1^G93A mouse is the most widely used ALS model, which develops the main clinical, electrophysiological and histopathological features of both familial and sporadic forms of the disease (Turner and Talbot, 2008; Mancuso et al., 2011b). Despite mutations in SOD1 gene are present in 20% of familiar ALS cases and only 2% of all ALS cases (Ragagnin et al., 2019), the interest of the SOD1 transgenic models is increased because alterations in SOD1 protein also occur in sporadic ALS cases (Bosco and Landers, 2010), and accumulation of wild-type SOD1 causes ALS (Graffmo et al., 2013).

Sigma-1 receptor (Sig-1R) is a protein ubiquitously expressed in the central nervous system (CNS) (Langa et al., 2003) and particularly enriched in the MNs. It is located in the endoplasmic reticulum (ER) cisternae at postsynaptic sites of C-terminals and at the mitochondria associated-endoplasmic reticulum membrane (MAM), an active and dynamic site in which there is crosstalk between mitochondria and ER (Hayashi and Su, 2007; Mavlyutov et al., 2012). Sig-1R is involved in numerous cellular processes, such as ion channel modulation, protein and lipid transport, ER stress response and mitochondria function (Penke et al., 2018; Herrando-Grabulosa et al., 2021). In recent years, genetic analyses revealed Sig-1R gene mutations involved in a juvenile form of ALS (Al-Saif et al., 2011; Watanabe et al., 2016) and in forms of motor neuropathies (Almendra et al., 2018; Ververis et al., 2019). Moreover, either accumulation of mutant SOD1 or absence of Sig-1R induced MAM disruption and mitochondrial dysfunction (Bernard-Marissal et al., 2015; Watanabe et al., 2016).

Nowadays, several Sig-1R ligands have interest as potential therapeutic agents against CNS disorders, including chronic neurological conditions such as pain, stroke, Huntington disease, among others (Bruna et al., 2018; Reilmann et al., 2019; Urfer et al., 2014). Regarding motoneuron diseases (MND), the Sig-1R agonist PRE-084 has shown positive effects reducing the MN death in vitro in organotypic culture of spinal cord subjected to excitotoxic damage (Guzmán-Lenis et al., 2009) and in vivo in the SOD1^G93A murine model of ALS (Mancuso et al., 2012), in the wobbler mouse model of spontaneous MN degeneration (Peviani et al., 2014), and after spinal nerve injury in adult mice and rats (Gaja-Capdevila et al., 2021; Penas et al., 2011). Studies testing two other Sig-1R agonists, SA4503 and pridopidine, also showed that treatment ameliorates ALS pathology (Ionescu et al., 2019; Ono et al., 2014), and pridopidine is being tested in a clinical trial for ALS (ClinicalTrials.govNCT04615923). However, Sig-1R ligands may act differently, and even contrarily, on neuroprotective mechanisms by modulating calcium homeostasis (Tadić et al., 2017). Thus, the Sig1-R appears as a promising target to promote MN protection, but more studies are needed to establish the role of Sig-1R ligands in MND models, and the type of ligand that may be most effective. Considering the recent data, the aim of the work reported here was to comparatively evaluate the therapeutic efficacy of three Sig-1R ligands in an in vitro model of MN degeneration and in the SOD1^G93A mouse. Moreover, it was investigated whether the administration of Sig-1R ligands could promote the modulation of glial reactivity. Finally, regarding the importance of ER stress and autophagy in the ALS pathogenesis (Hetz and Saxena, 2017; Medinas et al., 2017; Nguyen et al., 2019), we investigated whether these Sig-1R ligands modulate these molecular pathways.

…

Results

Sig-1R Ligands Exert Neuroprotection in SCOCs Under Chronic Excitotoxicity

Addition of THA to the SCOC induced a significant reduction of about 40% in the number of SMI-32 labeled MNs in the ventral horn after 21 DIV, compared to control slices (Figure 1). Slices treated with THA and PRE-084, BD1063 or SA4503 showed significant preservation of MNs at the two doses tested. The Sig-1R agonists PRE-084 and SA4503 significantly reduced MN death in a dose-dependent manner, with maximal protective effect at 30 μM (Figures 1A,B). The antagonist BD1063 also prevented MN death, with a highest effect at 3 μM (Figure 1C). Furthermore, the positive control against excitotoxicity Riluzole presented significant MN protection at similar levels than the Sig-1R ligands tested.

Treatment With Sig-1R Ligands Preserves Neuromuscular Function in SOD1^G93A Mice

SOD1^G93A mice treated with Sig-1R ligands for 8 weeks maintained a gain of body weight throughout the study, and did not present any secondary effect, indicating lack of general toxicity of these ligands (Figure 2A).

In order to assess the effect of Sig-1R ligands on neuromuscular function of the SOD1^G93A mice, we performed motor nerve conduction tests during the follow-up. Results showed that SOD1^G93A mice treated with PRE-084 and with BD1063 had a significantly higher TA and GM CMAP amplitude compared to saline administered SOD1^G93A mice at 13–16 weeks of age (Figure 2B and Supplementary Figure S1A). CMAP amplitude differences did not reach statistical significance in SA4503 treated group compared with untreated mice at the end of the follow up. Male SOD1^G93A treated with PRE-084 had a significant preservation of TA and GM CMAP amplitude throughout the follow-up compared to the saline group, while BD1063 treatment has lower effect (Supplementary Figure S1B, C). The PL CMAP amplitude showed mild preservation for groups treated with PRE-084 and SA4503 at both doses, although it was only significant for PRE-084 (Figure 2C).

The rotarod test revealed that SOD1^G93A mice treated with SA4503 at dose of 0.25 mg/kg and PRE-084 significantly improved the functional outcome compared with the untreated group, but animals treated with BD1063 and SA4503 1 mg/kg did not have any improvement (Figure 2D). Furthermore, the disease onset was delayed by 1 week (14 weeks of age) in mice treated with SA4503 (0.25 and 1 mg/kg) and PRE-084 versus the saline group (13 weeks) (Figure 2E).

The innervation of NMJ of the GM muscle was evaluated at 16 weeks of age. All the SOD1^G93A mice treated with Sig-1R ligands had a significantly higher number of innervated endplates compared with the saline group, supporting the preservation of CMAP amplitude observed in the nerve conduction tests (Figures 2E,F).

Sig-1R Ligands Promote MN Survival and Reduce Microglial Reactivity in SOD1^G93A Mice

The quantification of α-MNs in the ventral horn of lumbar spinal cord sections stained with cresyl violet at 16 weeks of age, revealed that treatment with PRE-084 and BD1063 mildly but significantly prevented the death of spinal MNs at the end-stage of the disease in comparison with untreated SOD1^G93A mice, whereas SA4503 at both doses assessed did not have a noticeable effect (Figures 3A,B).

Microglial and astroglial response was found markedly increased in SOD1^G93A compared to WT mice. Treatment with Sig-1R ligands reduced microglial reactivity, although only PRE-084 and SA4503 at 0.25 mg/kg caused a significant decrease, whereas the other treatments did not reach statistical significance (Figures 3C,D). Regarding astroglial immunoreactivity in the spinal cord ventral horn, we found that it was not modified by administration of the Sig-1R ligands in female SOD1^G93A mice (Figures 3E,F).

Analyses of Autophagic Flux and ER Stress During the Progression of ALS With Sig-1R Ligands

Considering that PRE-084 and BD1603 significantly preserved spinal MNs, we performed WB analyses of lumbar spinal cord lysates from WT and SOD1^G93A saline, PRE-084 and BD1063 treated mice at 8 and 16 weeks of age to evaluate two main molecular pathways implicated in ALS pathogenesis. Three markers of autophagy flux (Figure 4) were analyzed and we found that there was no differences in Beclin-1 protein levels between WT and SOD1^G93A mice groups at the two time points evaluated. A progressive increase of LC3-II levels was observed in SOD1^G93A saline group, observing higher levels at 16 weeks of age compared with WT mice group. In contrast, at 8 weeks of age, higher LC3-II levels were detected in PRE-084 group that were maintained at 16 weeks. No differences regarding LC3-II protein levels were detected in BD1063 group (Figures 4A–D). Immunofluorescence analyses of lumbar spinal cord sections revealed a significant accumulation of p62/SQSTM1 immunoreactive dots in the MNs and glial cells, mainly astroglia, of SOD1^G93A animals at 16 weeks of age (Figure 4E–G). In summary, treatment with Sig-1R ligands PRE-084 and BD1063 did not markedly modify the protein levels of autophagic markers in comparison with saline SOD1^G93A mice.

We also monitored the levels of ER stress markers in the spinal cord (Figures 5A,B). The protein levels of Sig-1R did not change at 8 and 16 weeks in SOD1^G93A mice, as previously described (Mancuso et al., 2012), and administration of Sig-1R ligands PRE-084 and BD1063 did not modify the levels of this receptor (Figures 5C,D). During disease progression, WB analyses revealed a marked increase in the chaperone BiP at 16 weeks of age in SOD^G93A mice (Figure 5D). There were no differences in the ratio p-IRE1/IRE between experimental groups at both time points evaluated, whereas there was a significant reduction of the ratio XBP1s/XBP1unspliced at 8 weeks, that was reverted to increased levels at 16 weeks in SOD1^G93A mice, with a significant difference for the BD1063 treated compared to WT mice. We also observed a significant increase of CHOP levels in SOD1^G93A mice at 8 weeks of age, though no differences were found at 16 weeks of age.

DISCUSSION

The results of this study demonstrate the beneficial effects of the Sig-1R agonist PRE-084 as well as the Sig-1R antagonist BD1063, on preserving neuromuscular function and attenuating MNs loss in the SOD1^G93A mouse model of ALS. The compound SA4503 showed lower effects in ameliorating the progression of the disease. Whereas previous studies have reported that PRE-084 (Mancuso et al., 2012) and SA4503 (Ono et al., 2014) treatment ameliorated the progression and survival of spinal MNs, we report for the first time that a Sig-1R antagonist, such as BD1063, has also neuroprotective effects in preclinical models of ALS. We have recently reported that all PRE-084, SA4503 and BD1063 also prevent MN death in a model of spinal root injury in mice (Gaja-Capdevila et al., 2021), emphasizing the interesting effects of Sig-1R ligands for improving motor neurodegenerative conditions.

Although similar outcomes were obtained in terms of MN preservation in vitro in the SCOC with the three Sig-1R ligands tested, there were noticeable differences in the in vivo model of ALS. Glutamate excitotoxicity is one of the main pathophysiological mechanisms implicated in ALS. Astrocytes are the main regulators of extracellular glutamate levels and excitatory amino acid transporter 2 (EAAT2) is the main astroglial synaptic glutamate reuptake transporter protein. The loss of EAAT2 has been reported in both ALS patients (Bristol and Rothstein, 1996) and rodent models (Howland et al., 2002). Our in vitro results showed that two considered Sig-1R agonists, PRE-084 and SA4503, and the antagonist BD1063 induced protection against chronic excitotoxicity, at the doses of 3–30 μM. Indeed, some reports reveal that binding to Sig-1R prevented neuronal death in vitro studies. Our group described that PRE-084 (at 10 μM) protected the SCOC against acute glutamate toxicity (Guzmán-Lenis et al., 2009). Ono and collaborators found that SA4503 (at 10 μM) protected the NSC34 cell line against SOD1^G93A and serum free neurotoxicity (Ono et al., 2014). However, those protective effects were inhibited by the co-addition of a Sig-1R antagonist, BD1063 or BD1047, respectively. In our study we assessed each compound alone, without combining them, demonstrating that BD1063 had also similar neuroprotective effects.

Several reports have focused on Sig-1R ligands to modulate ALS progression. PRE-084 (0.25 mg/kg) daily administrated in SOD1^G93A from 8 to 16 weeks of age showed neuroprotection, improving MN function and survival and extending the lifespan of SOD1^G93A mice (Mancuso et al., 2012). SA4503 (1 mg/kg) treatment from 5 weeks of age extended survival time, but did not affect the onset time in the SOD1^G93A mice (Ono et al., 2014). However, we found a delay of 1 week of disease onset compared to untreated mice. In addition, in our study we evaluate the number of MN survival to compare with the other two Sig-1R ligands. Even though SA4503 did not significantly preserve spinal MNs as PRE-084 and BD1063 treatment did, all the pharmacological treatments significantly preserved NMJ innervation in the hindlimb muscles at 16 weeks of age. This is a remarkable result, since NMJ disruption is an early event in ALS pathogenesis (Fischer et al., 2004; Mancuso et al., 2011a). Since SA4503 and PRE-084 had been shown to prolong the lifespan of SOD1^G93A mice, a further study is needed to elucidate whether BD1063 has similar effect. Recently, pridopidine, a small molecule that modulates axonal transport deficit and causes a reduction in mutant SOD1 aggregates in the spinal cord of SOD1^G93A mice though Sig-1R, was reported to attenuate also NMJ disruption (Ionescu et al., 2019). Taken together, it seems that Sig-1R ligands improve several cellular and histological hallmark pathologies related to ALS.

Despite the three Sig-1R compounds assessed, PRE-084, SA4503 and BD1063, have demonstrated potential to bind the Sig-1R, the neuroprotective effects observed in the SOD1^G93A mouse model significantly differed between the three ligands. In the same line, Wang and others compared the Sig-1R ligands SA4503, PRE-084 and pentazocine (PTZ) in a model of severe retinopathy (Wang et al., 2020). In vitro results yielded similar outcomes, whereas neither PRE-084 or SA4503 afforded in vivo protection comparable to PTZ. A wide range of evidence is now available to support the role of Sig-1R in the treatment of several CNS disorders, such Parkinson disease or ischemia (Nguyen et al., 2017; Haga et al., 2019; Zhao et al., 2019). It is important to note that in this study Sig-1R ligands were classified as agonists or antagonist according to the BiP/Sig-1R association assay, in which the interaction between Sig-1R and the chaperone BiP is used to identify the functional nature (agonistic or antagonistic) of compounds. However, classification of Sig-1R ligands causes a lot of controversy because some studies revealed different cellular function between compounds considered as agonists of Sig-1R. Indeed, Sig-1R ligands of the same type may act differently in each pathology/degenerative disease, even in opposite way (Tadić et al., 2017). For example, it has been shown that the agonist SA4503 normalized cytosolic Ca²⁺ levels following activation by kainate and by bradykinin in embryonic MNs, whereas PRE-084 (also Sig-1R agonist) did not exert any significant effect (Tadić et al., 2017). Furthermore, we emphasize that the traditional concept of agonist or antagonist is controversial for Sig-1R ligands, that may act as modulators of this receptor promoting activity in different pathways with a delicate balance of effects (for review see Herrando-Grabulosa et al., 2021). Moreover, the total absence of Sig-1R in SOD1^G93A ALS mice model accelerates the ALS pathology (Mavlyutov et al., 2013). Therefore, considering the positive results obtained, BD1063 might act pharmacologically as a partial agonist in our animal model. Further comparative studies with other ligands classified as antagonists such as BD1047 and studies related with ligand classification in vivo are needed.

A body of evidence has demonstrated the contribution of neuroinflammation, the role of non-neuronal cells including microglia and astrocytes in ALS pathogenesis. Genetic deletion of mutant SOD1 selectively in microglia increased the lifespan of ALS mice, despite the mutant protein was expressed in MNs and all other cell types (Boillée et al., 2006). We observed a reduction in microglia activation following treatment with PRE-084 and with SA4503 at a dose of 0.25 mg/kg (see Figure 3D), but not with BD1063. The effect of PRE-084 on reducing the microglial reactivity and improving the MN environment was already found in the SOD1^G93A ALS model (Mancuso et al., 2012), as well as in a mouse model of spinal muscular atrophy (SMA^2B/-), in which PRE-084 treatment mitigated reactive gliosis restoring the altered M1/M2 balance (Cerveró et al., 2018). On the other hand, we did not observe any effect with any of the three Sig-1R ligands assessed on reactive astrogliosis. Contrarily, after spinal root injury, where MN death also occurs, PRE-084 reduced astroglia immunoreactivity (Penas et al., 2011; Gaja-Capdevila et al., 2021). All these studies highlight the role of Sig-1R ligands modulating the neuroinflammatory response.

The etiology underlying the development of ALS remains poorly understood, but abnormal protein aggregation and altered proteostasis are common features of sporadic and familial ALS forms (Medinas et al., 2017). To elucidate the mechanisms through which BD1063 and PRE-084 promote MN preservation we analyzed autophagic flux and ER stress. Autophagy is an intracellular lysosome degradation system responsible for the clearance of cytoplasmic components and organelles. Enhancement of autophagy has been reported in ALS with alterations in several steps. Thus, the level of LC3-II, which is correlated with the extent of autophagosome formation, was found increased in SOD1^G93A transgenic mice at symptomatic stage (Morimoto et al., 2007; Tian et al., 2011), as we also found in this study. The marker of the late stage autophagosome, the autophagy adaptor p62, interacts with polyubiquitinated misfolded mutant SOD1 (mSOD1), sequestering mSOD1 into protein inclusions, so fusion of the autophagosome to the lysosome becomes insufficient at the end stage (Tian et al., 2011). Furthermore, autophagy is activated in the ventral spinal cord MNs in sporadic ALS patients, observed by immunostaining for LC3 and p62 (Mizuno et al., 2006; Sasaki, 2011). In our results, we also observed that there was an accumulation of SQSTM1/p62 in the MNs of spinal cord and in the glia cells in SOD1^G93A mice. However, there was no difference between untreated group and animals treated with the Sig-1R ligands. It is worth to mention that in ALS the role of autophagy is confusing and it is still unknown whether activation or inhibition of autophagy may influence in the disease progression (Nguyen et al., 2019), and depending on the model used (Medinas et al., 2017).

On the other hand, increased expression of ER stress markers was observed in post-mortem tissues from ALS patients (Hetz et al., 2009), and correspondingly, in this work we observed up-regulation of BiP and XBP1s in the spinal cord of symptomatic SOD1^G93A mice. The unfolded protein response (UPR) and the autophagy pathway have been linked (Hetz et al., 2009); thus, XBP1 conditional deletion in the nervous system ameliorated SOD1 mice pathogenesis, through up-regulation of the autophagy pathway boosting the degradation of mSOD1 aggregates. In the context of ALS, the functional significance of ER stress is still unclear, because activating UPR is a protective response to increase protein folding and quality control mechanisms, whereas chronic stress may represent a deleterious signaling due to irreversible disturbance of ER homeostasis. Although we found some differences in the markers analyzed by WB between WT and SOD1^G93A mice, we did not observe changes induced by treatment with the Sig-1R ligands, indicating that autophagy and ER stress pathways are not significantly modified through Sig-1R modulation. However, specific immunolabeling analyses of the spinal cord may reveal changes in specific populations, either MNs or glial cells, that may be obscured in protein analyses of the whole spinal cord tissue. Thus, further experiments are needed to elucidate the mechanisms of action by which Sig-1R ligands yield neuroprotective effects.