Muscle-targeted Klotho Gene Therapy Ameliorates ALS Hallmarks by Addressing Multiple Disease Mechanisms in SOD1G93A Mice

doi:10.21203/rs.3.rs-4510655/v1

Download PDF

Article

Muscle-targeted Klotho Gene Therapy Ameliorates ALS Hallmarks by Addressing Multiple Disease Mechanisms in SOD1^G93A Mice

https://doi.org/10.21203/rs.3.rs-4510655/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

The antiaging protein a-Klotho (KL) exhibits neuroprotective and myoregenerative properties, mitigating age-related neurodegeneration and promoting muscle regeneration. This study harnesses its pleiotropic properties in the context of Amyotrophic Lateral Sclerosis (ALS), a motoneuron disease lacking effective treatments due to its diverse pathophysiological mechanisms. By overexpressing secreted KL in skeletal muscles of SOD1^G93A mice with myotropic viral vectors we aimed to directly protect muscles and exert a paracrine effect on motoneuron (MN) terminals. Secreted KL preserved MNs and neuromuscular junctions, and mitigated glial reactivity, resulting in maintained muscle mass, improved neuromuscular function, delayed disease onset, and extended survival. Even when administered during symptomatic stages, KL slowed down ALS progression. Transcriptomic and proteomic studies in muscles revealed significant correction of pathophysiological mechanisms involved in ALS disease, unveiling novel roles for KL. These findings highlight the potential application of muscle-secreted KL in ALS regardless of its origin and suggest broader therapeutic implications.

Biological sciences/Neuroscience/Diseases of the nervous system/Amyotrophic lateral sclerosis

Biological sciences/Neuroscience/Molecular neuroscience

Biological sciences/Biotechnology/Gene therapy

Biological sciences/Neuroscience/Motor control/Neuromuscular junction

Biological sciences/Molecular biology/Transcriptomics

Amyotrophic lateral sclerosis (ALS) is a devastating neurodegenerative disorder of the motor system characterized by the degeneration of motoneurons (MNs) and muscle denervation. This results in progressive muscle weakness ultimately leading to paralysis and patient death within 2–5 years of diagnosis, often due to respiratory failure¹. Sporadic ALS (sALS) accounts for the majority of the cases, with only 5–10% being familiar forms (fALS) caused by inherited mutations. Over 30 genes have been linked to ALS and several animal models carrying mutations in these genes have been developed ^2,3. The transgenic mouse featuring a high copy number of the human SOD1 gene with a glycine-to-alanine transition at the 93rd codon being the most extensively used⁴.

Multiple pathological mechanisms underlying the disease process have been identified using ALS models. These mechanisms include oxidative stress, inflammation, excitotoxicity, mitochondrial dysfunction, impaired proteostasis, endoplasmic reticulum stress, and RNA disturbances. The interplay of these factors likely results in the selective death of MNs⁵. Promising preclinical and clinical trials targeting a single or a limited number of disease mechanisms have largely failed³. A growing viewpoint supports a more integrated approach by targeting multiple mechanisms simultaneously. Here we propose the therapeutic exploration of the pleiotropic protein a-Klotho (KL), which counteracts several mechanisms involved in neurodegeneration, such as oxidative stress^6,7, demyelination^8,9, senescence^10,11, inflammation^10,12, and synaptic dysfunction^13,14. KL is also critical for promoting muscle regeneration^15–17 and avoiding fibrosis¹⁸. KL-deficient transgenic mice exhibit altered gait and reduced large neurons in the ventral horn of the spinal cord, where astrogliosis and neurofilament phosphorylation are also evident^19,20. These changes resemble the MN degeneration in ALS, with a significant reduction of cytoplasmic RNA and apparent chromatolysis as described in patients’ samples¹⁹. SOD1^G93A mice crossed with KL-overexpressing mice showed slightly delayed disease onset and slowed disease progression, with only mild strength improvements²¹. This limitation in efficacy was likely attributed to the absence of KL upregulation in the muscles²².

Changes in skeletal muscle innervation manifest early in ALS and precede MN death^23,24. Electrophysiological data correlate with structural alterations in distal nerves^25–28, muscles^29–31, and terminal Schwann cells at early disease stages^32,33, suggesting that axonal retraction leads, in a “dying-back” pattern, to the death of MNs^26,34. This emphasizes that therapeutic approaches should target both MNs and the neuromuscular junctions (NMJ) to maximize their positive impact. To exploit the neuroprotective and myoregenerative properties of KL, we have systemically delivered myotropic MyoAAV vectors³⁵ to overexpress the KL secreted form (sKL) in skeletal muscles of SOD1^G93A mice, thus enhancing paracrine effects on MN terminals.

AAV8-mediated delivery of sKL to muscles preserves neuromuscular and motor function in SOD1^G93A mice

To overexpress sKL in muscles we intravenously administered AAV8 vectors with the human desmin promoter. With this formerly validated systemic administration^36–38, we achieve widespread and restricted gene expression in skeletal muscles. In a first study, we treated 6-week-old female SOD1^G93A mice with 3 x 10¹⁴ vg/kg of AAV8-Des-sKL or null vectors via tail vein injection. We followed them with motor and nerve conduction tests until 16 weeks of age along with sex and age-matched wild type (WT) mice (Fig. 1A). The electrophysiological evaluation revealed that the amplitude of the compound muscle action potentials (CMAP) was notably preserved along time in treated compared to mock SOD1^G93A mice. This preservation was significant at the three time points examined for the plantar interosseus (PL) muscle (Fig. 1B, C) and at 8 and 16 weeks for the tibialis anterior (TA) muscle (Fig. 1D, E). Interestingly, muscle-secreted KL was able to prevent the decline of central motor pathways at the early stages, as demonstrated by the significant maintenance of motor evoked potential (MEP) amplitudes at 8 and 12 weeks (Fig. 1G, H). Muscle overexpression of sKL failed to preserve body weight (data not shown) but maintained the gastrocnemius muscle (GM) mass at the terminal stage (95.1 ± 2.6 mg) compared to SOD1^G93A mock-treated mice (76.2 ± 6.6 mg). These results support the role of sKL in preventing denervation and muscle atrophy (Fig. 1F).

Motor performance and coordination in the rotarod test were significantly improved in sKL-treated SOD1^G93A animals throughout the disease progression (Fig. 1I). The force of treated SOD1^G93A female mice was maintained in the grip strength test at the late stage of 16 weeks, compared to mock-injected controls (Fig. 1J). These functional benefits were associated with MNs protection (Fig. 1K, P) in the lumbar spinal cord (13.7 ± 1.5, mean MN number in ventral horns ± SEM) compared to SOD1^G93A mock mice (8.6 ± 0.7), in line with electrophysiological findings. Preserved CMAP amplitudes in muscles were corroborated by the significant maintenance of the number of occupied motor endplates (75.2 ± 5.6%) in mice receiving gene therapy compared to mock group (40.6 ± 6.4%) (Fig. 1L, Q).

Neuroglial reactivity in the spinal cord at 16 weeks of age was reduced for both microglial (Fig. 1M, R) and astroglial cells (Fig. 1O, S). Thus, although sKL does not cross the blood-brain barrier, secretion of sKL from skeletal muscles mitigates the characteristic neuroinflammatory response in the spinal cords of SOD1^G93A mice.

Targeting sKL to muscles with low doses of MyoAAV vectors promotes functional improvements in SOD1 ^G93A mice

The previous analyses established that KL overexpression preserves neuromuscular and motor function in SOD1^G93A female mice. However, extremely high systemic AAV doses in patients have been linked to adverse effects and pose a considerable manufacturing challenge^39,40. Thus, we explored MyoAAV, a capsid 9 variant with higher muscle tropism³⁵, to achieve comparable outcomes with lower viral doses. Following the same protocol (Fig. 1A), we treated female and male SOD1^G93A mice with three doses of MyoAAV vectors: 2 x 10¹², 8 x 10¹², and 1.6 x 10¹³ vg/kg. A clear dose-dependent effect was observed in the maintenance of CMAP (Fig. S1A, B) and MEP amplitudes (Fig. S1C).

Then, MyoAAV-Des-sKL vectors were given at 1.6 x 10¹³ vg/kg, representing a ~ 19-fold dose reduction compared to AAV8 vectors. Significant preservation of CMAPs amplitudes for PL and TA muscles was achieved in treated female mice by 12 weeks, almost doubling the mean values of SOD1^G93A null-treated mice at 16 weeks of age (Fig. 2A, B). SOD1^G93A male mice exhibit a more accelerated decline in neuromuscular function than females, with a significant reduction in CMAP amplitude of TA muscle at 8 weeks compared to WT mice (Fig. S2B). sKL treatment preserved CMAP amplitude for both muscles also in male mice throughout disease progression (Fig. S2A, B). Wild-type animals administered with MyoAAV-Des- sKL showed no differences in CMAP amplitudes compared to the WT null-injected group.

Treatment in SOD1^G93A mice significantly preserved the number of functional motor units (MU) in TA muscle at 16 weeks of age, confirming improved neuromuscular function (Fig. 2D and S2D). Additionally, treatment fostered a slight increase in single motor unit amplitude in females (Fig. 2E), reaching statistical significance in SOD1^G93A males (Fig. S2E). These findings indicate that sKL treatment sustains functional innervation of hindlimb muscles. MEPs amplitudes tended to be higher in treated mice compared to SOD1^G93A mock-treated controls (Fig. 2C and S2C) during follow-up, suggesting sKL secretion from muscles helped to preserve the descending corticospinal tract. However, significant differences were observed only at 12 weeks for females (Fig. 2C) and at 8 weeks for males (Fig. S2C).

sKL-treated SOD1^G93A mice showed improved rotarod performance throughout the disease course, with significant differences at end stage (Fig. 2G and S2G). At 16 weeks of age, most mock-treated SOD1^G93A females (56%) and males (70%) did not reach the cutoff time, contrasting with 22% and 32%, respectively, of sKL-treated SOD1^G93A mice. Force in the grip strength test (Fig. 2F and S2F) and running velocity in the treadmill test (Fig. 2H and S2H) were higher in treated SOD1^G93A mice compared to matched SOD1^G93A mock-treated animals, and closer to values observed in WT animals. Hindlimb tremors and clasping, detected earlier and preceding dysfunctions in the locomotor and strength tests⁴¹, were delayed in mice overexpressing sKL compared to mock-injected controls (Fig. 2J and S2J). As observed in AAV8-based gene therapy, overexpression of sKL directed by MyoAAV did not improve the decline in body weight (data not shown) but significantly preserved gastrocnemius mass in both sexes (Fig. 2I and S2I). Importantly, sKL treatment led to a significant extension in median lifespan of SOD1^G93A animals, with a 6.9% (9 days) increase in females and a 13.7% (16 days) increase in males (Fig. 2K and S2K). While the median survival extension was modest in treated females, some had prolonged lifespans, with the longest-lived mouse reaching 172 days, in contrast to null-treated mice that survived up to 149 days.

Analysis of AAV viral genomes confirmed the transduction of GM muscles, spinal cord, and liver (data not shown); however, desmin promoter largely confined mRNA expression to muscles (Fig. S1D, E). Examination of TA muscles corroborated KL protein overexpression in treated mice, with a 3.3-fold increase in expression for WT sKL and a 4.6-fold increase for SOD1^G93A sKL, compared to null-treated matched groups (Fig. S1I, J). ELISA further validated KL protein overexpression in GM muscles of transgenic mice, showing an average 2.7-fold increase after treatment (Fig. S1F). Levels of KL in serum demonstrated a significant 18-fold increase in treated SOD1^G93A mice, indicating its release to the bloodstream from transduced muscles (Fig. S1G). In the spinal cord, since KL does not cross the blood-brain barrier, no significant changes were observed in KL protein (Fig. S1H), despite the leakage of the desmin promoter detected at mRNA level (Fig. S1E).

Importantly, MyoAAV-mediated muscle overexpression of sKL led to protection of MNs (Fig. 3A, B) in the ventral horn of the spinal cord (16.1 ± 0.7, average ± SEM of MNs per section), approximately doubling the counts in SOD1^G93A mock-treated mice (8.3 ± 0.7). This treatment also preserved the number of innervated muscle endplates (78.2 ± 3.5%) at the late stage of the disease (Fig. 3C, D), in comparison to the mock-treated group (47.0 ± 3.9%). No differences in NMJ innervation or MN numbers were observed between mock and sKL-treated WT animals. Neuroglial reactivity in the spinal cord at the final stage was notably reduced in SOD1^G93A mice treated with MyoAAV-Des-sKL vectors. Conversely, overexpression of sKL in WT mice did not lead to changes in glial markers expression (Fig. 3E-H).

Given that sKL preserved the amplitude of MEPs, we quantified the number of pyramidal neurons in layer V of the primary motor cortex in female mice. Cortical layer V neurons were less numerous in SOD1^G93A null animals (1,211 ± 50.7 per mm²) compared to WT mice (1,441 ± 54.5) (Fig. 3I-K), particularly in posterior regions of the cortex. We found a significant rescue of layer V neurons in sKL-treated SOD1^G93A mice (1,451 ± 56.6), indicating upper MN protection. These histological findings confirm the improvements in CMAP amplitude, motor unit number, MEP amplitude, and locomotion detected in electrophysiological and functional tests in SOD1^G93A sKL mice. In essence, a lower dose of MyoAAV vectors demonstrated more favorable outcomes in preserving the neuromuscular system compared to those observed with AAV8-Des-sKL gene therapy.

KL muscle secretion slows down functional decline in symptomatic SOD1^G93A mice

Treating at the symptomatic stage presents a challenge in the rapidly progressing SOD1^G93A mouse model, especially for AAV gene therapies, considering the time required for transgene expression. Nevertheless, electrophysiological follow-up of SOD1^G93A mice administered at disease onset, i.e. 11 weeks of age, evidenced that the progressive fall in CMAP amplitude in PL and TA muscles was attenuated shortly after treatment. In treated female SOD1^G93A mice, CMAP amplitude reached significant differences at 16 weeks for both PL and TA muscles, compared to mock-injected mice (Fig. 4A, B). In males, the effect on CMAPs was also evident after administration (Fig. S3A, B), yet at 16 weeks only remained significant for PL muscles. Calculating the CMAP preservation for each animal in percentage relative to the week before administration reveals an effective mitigation of CMAP amplitude decline (Fig. 4C, D and Fig. S3C, D). At this symptomatic stage, treatment with sKL did not improve the decline of MEPs amplitude in either males or females, despite a trend toward preservation (Fig. 4E and S3E). Enhanced performance in the rotarod test was also evident (Fig. 4F and S3F), along with significant maintenance of grip strength at the disease final stage (Fig. 4G and S3G).

Results of nerve conduction studies were related to an increased number of remaining spinal MNs in SOD1^G93A mice (12.71 ± 0.7 MNs per slice), compared to mock-treated mice (8.33 ± 0.7), at 16 weeks of age (Fig. 4H, I). Additionally, there was significant preservation of innervated NMJ in the treated group (77 ± 2.5%) as opposed to the null treated controls (51.2 ± 6.2%) (Fig. 4J, K). Taken together, we demonstrate that even if given during the overt progression of the disease, overexpression of sKL in muscles reduces MN degeneration and promotes functional improvements in the SOD1^G93A mouse model.

Widespread overexpression of sKL fails to preserve neuromuscular function in SOD1^G93A mice

Considering the observed rise in sKL serum concentration in treated SOD1^G93A mice (Fig. S1G), we questioned whether the beneficial effects shown above originated from elevated circulating sKL rather than robust expression in skeletal muscles. To elucidate if overexpression of sKL in muscles was specifically required, we conducted a proof-of-concept study in which a small batch of SOD1^G93A female mice were intravenously administered with 1.6x10¹³ vg/kg of AAV9-CAG-sKL vectors (Fig. S4A). To achieve stronger secretion of sKL to the bloodstream, we used AAV9 vectors and the ubiquitous CAG promoter for whole-body overexpression of sKL. This resulted in a 2.5-fold increase of KL serum levels compared to the MyoAAV-based approach (Fig. S4B). However, similar protein levels of sKL in the GM muscle compared to those of control animals evidenced the inefficiency of overexpressing sKL in muscles using this dose and route of administration (Fig. S4C). The approach failed to show beneficial effects on CMAP (Fig. S4D, E) and MEPs amplitudes (Fig. S4F) in AAV9-treated SOD1^G93A females when compared to the mock group. These data reinforce the need to precisely direct sKL overexpression to skeletal muscles.

Overexpression of sKL normalized altered muscle gene expression patterns.

To determine the effect of sKL on gene expression signature of GM muscles, we conducted bulk RNA sequencing (RNAseq) and proteomic studies. We explored sKL molecular mechanisms responsible for the improvement of neuromuscular functions in animals treated presymptomatically with MyoAAV vectors. Principal component analysis (PCA) identified three distinct clusters in both data sets, including SOD1^G93A null group, the WT group regardless of administration, and an intermediate group represented by SOD1^G93A sKL-treated animals, positioned between the two other clusters (Fig. 5A, B). This is also illustrated in the heatmap depicting all significant genes across the four conditions, with SOD1^G93A sKL animals showing color hues more like those of WT mice, both for RNA and proteins (Fig. 5D, E). Differential expression analysis identified 721 differentially expressed genes (DEGs) in SOD1^G93A compared to WT mice (log₂fold change >|1.5|, FDR < 0.1). In contrast, only 119 DEGs were detected in transgenic sKL versus WT null mice, consisting of 37 upregulated and 82 downregulated genes (Table S1). SOD1^G93A mice showed 157 underabundant and 360 overabundant proteins in GM muscle compared to WT null mice (fold change >|1.5|, FDR < 0.05), which was corrected by treatment to no underabundant proteins and only 9 overabundant proteins (Fig. 5C and Table S2). Importantly, sKL overexpression in WT mice only induced changes in 44 gens, with no differentially represented proteins.

Gene set enrichment analysis (GSEA) of RNAseq data between sKL and null-treated SOD1^G93A mice showed tendencies for enriched pathways related to ubiquitin-mediated proteolysis, regulation of protein exit from the endoplasmic reticulum, protein maturation, export, and targeting. Mitochondrion organization, SNARE interactions in vesicular transport, and negative regulation of oxidative-stress-induced neuron cell death also showed a tendency towards enrichment. On the contrary, sKL appeared to downregulate aging-related mechanisms, apoptotic pathways, myoblast migration, and deoxyribonuclease activity (Fig. 6A).

We then analyzed expression of the top dysregulated proteins corrected by the treatment. Most of the downregulated and normalized proteins belonged to functions related with microtubule and cytoskeletal dynamics, such as myosin chains, myosin kinases, and myosin- or actin-binding proteins (Fig. 6B). Reduction of neurofilament polypeptides in the muscle of SOD1^G93A mice, which play a crucial role in motor axons, partially resolved after sKL overexpression. We also found normalization of various RNA trafficking and processing proteins, such as exportin-5, TIAL1, ribosomal proteins, and RNA-binding protein EWSR1 (Fig. 6C). While not being among the top upregulated, numerous proteins belonging to the ubiquitin-ligase family (Parkin7, E3A, E3B, ARIH1, ARIH2, UBR4, HERC2, RNF123, WWP1, and RNF14), ubiquitin conjugation factors and ubiquitin-conjugating enzymes (E2 D1, E2 H, E2 K, E2 L3, E2 R, E4A, and UBC12) demonstrated a significant positive fold-change, thereby affecting protein turnover (data not graphed). Elevated levels of H-2 class I histocompatibility antigen, allograft inflammatory factor 1-like (IBA1), tapasin, or galectin-3 in SOD1^G93A muscles indicate an inflammatory state, which was mitigated by sKL overexpression.

To identify correlations between mRNA and protein levels, and validate both technologies, we integrated our RNAseq and proteomic data. To combine data sets, we compared p-values (p < 0.1) for the top and bottom differentially expressed genes and proteins when evaluating the sKL group against the null group of SOD1^G93A animals. Major downregulated pathways were related to DNA topological changes and nuclear speck formation. This association could be related to the observed decrease in genes and proteins within the Gene Ontology (GO) annotations of mRNA binding, single-stranded RNA binding, and ribonucleoprotein granule (Fig. 6D). Consistent downregulations in both RNA and protein levels were also observed for those increasing inflammatory response, downregulating macrophage migration, and glial cell activation. sKL induced antioxidant effects by inhibiting signaling pathways related to upregulation of ROS biosynthesis and oxidases, or by increasing thioredoxin peroxidase activities. Other upregulated pathways in GM muscles were centered around functions associated with skeletal muscle fiber contraction, muscle development, and differentiation (Fig. 6E), while regenerative myogenesis was normalized. This correlates with increased glycogen, ribonucleoside, energy reserves, and cysteine and methionine metabolic processes. Expression of genes within the Ca²⁺/calmodulin-dependent protein kinase signaling pathway, crucial for synaptic plasticity and muscle contraction⁴², was also activated.

Overall, transcriptomic and proteomic data confirmed multiple beneficial effects of secreted KL in skeletal muscles on pathological mechanisms in ALS, revealing novel roles for KL in proteolysis and RNA homeostasis.

A growing body of studies supports the concept that ALS is characterized by a dying-back process, starting with distal axonopathy^{23,24,26–28,30,33}. Neuromuscular denervation precedes axonal degeneration, and both occur before clinical motor signs and spinal MN death. Thus, merely preserving MN soma is insufficient to halt ALS progression. Axons cannot reestablish connections at the NMJ after significant retraction^23,24,43. Early abnormalities are detected in muscles of ALS animal models, contributing to axonal detachment of NMJ^29,31. To target the neuromuscular axis, we overexpressed sKL in skeletal muscles throughout the body. This approach aimed to harness sKL myoregenerative effects within muscles and exert a paracrine neuroprotective effect on MN terminals, preventing NMJ detachment and promoting MN survival.

Traditionally, ALS clinical trials have focused on a single or a few of the disease mechanisms. ALS is believed to arise from various interacting pathophysiological mechanisms, leading to extensive neuromuscular network disruption⁵. ALS heterogeneity suggests that different mechanisms exert varying influence across patients, rendering limited target strategies ineffective³. Consequently, an emerging perspective advocates for a synergistic approach targeting multiple pathways. KL is distinguished for its exceptional capacity to influence various signaling pathways linked to neurodegeneration. A recent study suggested that KL holds potential to mitigate the pathological progression of ALS²¹. In that study, SOD1^G93A mice were crossed with full-length KL-overexpressing mice, where KL was driven by the EF1a promoter and showed no detectable overexpression in muscles^21,22. We speculated that this lack of overexpression in muscles might account for the comparatively poor results observed in their neurological score and the hanging wire grip test. In our approach, while very low mRNA expression was measured in the spinal cord of treated mice compared to muscle expression, no differences in sKL protein levels were detectable by ELISA. This discards an effect of sKL protein in the central nervous system as a contributor to ALS phenotype improvement.

To target sKL specifically to muscles in SOD1^G93A mice, we systemically administered AAV8 vectors with the human desmin promoter, mitigating off-target effects in other tissues^36–38. To further ensure the biosafety of our approach, we chose the secreted KL isoform, which lacks a domain found in the other KL forms and associated with toxicity in mineral metabolism and bone structure⁴⁴. AAV8-Des-sKL improved functional outcomes in treated SOD1^G93A mice, but the high dose (3x10¹⁴ vg/kg) limited clinical translation. Consequently, we transitioned to the recently described MyoAAV serotype, which has enhanced muscle tropism compared to naturally occurring AAVs³⁵. MyoAAV vectors encoding sKL preserved SOD1^G93A mice neuromuscular function in a dose-dependent manner, achieving comparable results to AAV8-Des-sKL therapy with an almost 20-fold lower dosage. Electrophysiological data showed preserved CMAPs amplitudes for hindlimb muscles, nearly doubling values at the end stage relative to SOD1^G93A control mice. This was attributed to a partial yet significant maintenance of motor units, despite a concurrent reduction in their size. Higher MEP amplitudes also evidenced preservation of the corticospinal tract at early stages and a tendency at end stage.

On the contrary, sKL failed to mitigate the characteristic body weight loss of SOD1^G93A mice during advanced stages. Thus, the impact of sKL overexpression on maintaining muscle mass in treated mice is even more meaningful. Given that KL inhibits insulin and IGF-1 signaling pathways, possibly resembling a caloric restriction model, improvements in body weight were not anticipated in sKL-overexpressing SOD1^G93A mice^20,22.

Overexpression of sKL in muscles showed a remarkable effect on NMJ innervation and spinal MNs protection, explaining delayed disease onset, attenuation of its progression, neuromuscular benefits, and improved motor function. MN degeneration in SOD1^G93A mice is a non-cell autonomous process, with glia playing an important role. Ablation of mutant SOD1 from microglia⁴⁵ or astrocytes⁴⁶ can extend survival by slowing disease progression. In our study, sKL attenuated neuroinflammatory glial reactivity in the spinal cord. These findings suggest that KL secretion by muscles influences not only MNs directly connected to muscles through their terminals but also glial cells within the spinal cord. Given the absence of KL protein overexpression in the spinal cord and its inability to cross the blood-brain barrier^47,48, this effect may stem from reduced MN death or from secondary mediators that indirectly mitigate neuroinflammation. Importantly, these protective effects extended beyond the spinal cord and muscles, significantly preserving the numbers of pyramidal neurons in layer V of the primary motor cortex. Although corticospinal tract integrity remained compromised, as indicated MEPs decline, the rescue of upper MNs might explain the higher amplitude of these potentials, confirming preserved connectivity between upper and lower MNs in treated animals.

Survival analysis revealed a significantly prolonged lifespan in females and a more pronounced effect in SOD1^G93A males following treatment. This is intriguing given that SOD1^G93A male mice typically exhibit worse disease outcomes and shorter lifespans than females. However, it is consistent with previous reports of KL overexpression, where KL-overexpressing males displayed a longer life expectancy compared to females²².

Despite muscles being modest contributors to circulating protein levels, sKL successfully entered the bloodstream. We investigated whether the observed improvements were influenced by high sKL concentrations in the blood or within the muscles. Using AAV9 vectors with a ubiquitous promoter, we increased circulating sKL concentrations while maintaining muscle levels similar to controls. In this case, there were no improvements in CMAPs and MEPs amplitudes, indicating that muscle-targeted sKL overexpression, rather than high blood levels, is crucial for preserving neuromuscular function in SOD1^G93A animals. This highlights the importance of using AAV serotypes with enhanced muscle tropism for this therapeutic strategy, dismissing the systemic administration of recombinant sKL protein or mRNA.

Given the multiple therapeutic effects of sKL overexpression in SOD1^G93A mice, we treated animals at the symptomatic stage of 11 weeks. This is crucial for assessing therapeutic efficacy and ensuring the findings are more representative of the stage at which human patients are typically diagnosed. Nerve conduction tests evidenced a slowing of the progressive decline in neuromuscular function, with clear deceleration shortly after treatment for both muscles tested and in both sexes. Remarkably, sKL overexpression in muscles safeguarded remaining MNs and NMJs and provided substantial improvements in late stage rotarod performance and grip strength in SOD1^G93A animals.

We also investigated the impact of sKL overexpression on RNA and protein expression patterns in both WT and SOD1^G93A mice, with a focus on ALS pathology. In WT mice, sKL muscle overexpression led to subtle RNA expression alterations without affecting protein levels, suggesting minimal effects on young muscles and highlighting its safety and specificity in pathological conditions. In contrast, sKL overexpression had a significant impact on SOD1^G93A mouse muscles, correcting dysregulated RNAs and proteins, particularly upregulating proteins endogenously downregulated by the disease. Functional comparison between RNA and protein revealed most enriched pathways in SOD1^G93A sKL compared to SOD1^G93A null animals were related to muscle contraction and development. Among the most prominently upregulated proteins in SOD1^G93A null muscles were ANKRD1, COL19A1, and RRAD, with sKL treatment correcting their expression at both mRNA and protein levels. Previous studies reported increased mRNA of these genes as reliable markers of disease progression^49,50.

The integration of RNAseq and proteomic data uncovered significant corrections in inflammatory response pathways and RNA disturbances, crucial ALS hallmarks overlooked when analyzing RNAseq data alone. The role of KL in modulating inflammation has been previously documented^10,12. Consistent with this, our histological observations demonstrated corrections in glial reactivity within the spinal cord, suggesting a potential beneficial effect of KL on neuroinflammatory prodegenerating mechanisms in ALS. Furthermore, the antioxidant properties of KL, previously observed in other disease models^6,7, through mechanisms involving the thioredoxin-peroxiredoxin system, were confirmed in our results.

Mutations in TARDBP, FUS, C9ORF72, and other RNA-binding proteins associated with ALS underscore the central role of RNA metabolism abnormalities in ALS and frontotemporal dementia^1,2. Our study reveals that sKL overexpression decreased the expression of genes and proteins associated with RNA-binding, nuclear speck, and ribonucleoprotein granule functions. Additionally, secreted KL induced ubiquitin-mediated proteolysis pathways, especially evident at mRNA level but also significant in proteomic studies. Approximately 97% of ALS cases exhibit TAR DNA-binding protein 43 (TDP-43) proteinopathy, often accompanied by abnormal inclusions of p62, FUS, C9ORF72 dipeptide repeats, SOD1, or other abnormal proteins³. Promoting proteostasis and RNA homeostasis are new roles for KL, previously unreported. These new functions, coupled with KL’s ability to counteract oxidative stress, enhance mitochondrial structure, and mitigate inflammation, position it as a promising therapeutic candidate for ALS regardless of disease etiology.

In summary, our data provides consistent evidence that targeting sKL to skeletal muscles allows significant protection against muscle denervation and MN death. Early intervention is crucial for optimizing therapeutic efficacy in ALS; however, even treatment initiation at a symptomatic stage showed promising outcomes in hindering disease progression. This proof-of-concept study offers a safe and effective approach for ALS therapy, correcting multiple shared pathways of the disease and maximizing its translational potential.

Study design

This proof-of-concept study explored the potential of AAV8 and MyoAAV to deliver and overexpress the secreted Klotho gene to muscles in WT and SOD1^G93A mice. At 6 weeks of age, mice received a tail vein injection of 3x10¹⁴ vg/kg of AAV8-Des-sKL vectors or one of three doses (2x10¹², 8x10¹², and 1.6x10¹³ vg/kg) of MyoAAV-Des-sKL vectors. Control mock mice received equivalent doses of AAV8-Des-Null or MyoAAV-Des-Null vectors, carrying a non-coding stuffer sequence. Another batch of animals was administered during the 11th week of age with MyoAAV-Des-sKL or null vectors at 1.6x10¹³ vg/kg. All experiments were performed by researchers blinded to treatment, and to genotype when possible. Mice were assigned randomly to groups to ensure that littermates or mice from the same cage were not grouped. Sample sizes were determined based on prior observations in our laboratory, together with G*Power software in some of the studies. For RNAseq and Mass spectrometry (MS)/proteomics four female mice were employed per group.

Animals

Transgenic mice expressing the G93A mutation in the human SOD1 gene and non-transgenic wild-type (WT) littermates with the B6SJLF1/J background were used for the characterization and gene therapy studies. High-copy number SOD1^G93A mice (Tg[SOD1-G93A]1Gur) were obtained from The Jackson Laboratory in Bar Harbor, ME. These mice were bred by mating transgenic males with F1 hybrid females from Janvier Laboratories and maintained as hemizygotes. Animals were housed under standard conditions with unrestricted access to food (Teklad 2018) and water at the animal services of the Universitat Autònoma de Barcelona. Proper care, handling and euthanasia of the animals was carried out following European Union Council Directives (Directive 2010/63/EU) and Spanish regulations for the use of laboratory animals. The experimental procedures were approved by the Ethics and the Biosafety Committees of the Universitat Autònoma de Barcelona (CEEAH 4812 and HR-128-18).

Virus production

The cDNA of the mouse secreted a-Klotho isoform was cloned between AAV2 ITRs under the regulation of the human desmin promoter (provided by B. Byrne, University of Florida, FL, USA). The woodchuck hepatitis virus responsive element (WPRE) was added at 3′ to stabilize mRNA expression, followed by the SV40 poly(A) signal. For the null control, the non-coding RNA transcript variant IV of the TMEM185A gene was cloned into the same plasmid. AAV2/8 and Myo stocks were produced at the facilities of the Viral Production Unit of UAB-VHIR (http://viralvector.eu). In brief, vectors were created through the triple transfection system in HEK293-AAV cells, using expression plasmids including the Rep2Cap8 (provided by J.M. Wilson, University of Pennsylvania, Philadelphia, USA) or Rep2CapMyo plasmid containing AAV genes ³⁵ and the pXX6 plasmid containing adenoviral genes, serving as a helper virus ⁵¹. Following a 48-hour incubation period, AAV vectors were collected, subjected to benzonase treatment, purified using an iodixanol gradient, and quantified using the Picogreen (Invitrogen) system for viral genome quantification ⁵².

Electrophysiological tests

Motor nerve conduction tests were performed in anesthetized (ketamine/xylazine 90/10 mg/kg i.p.) animals by stimulating the sciatic nerve using single pulses of 20µs duration (Grass S88) delivered by two needle electrodes positioned at the sciatic notch, with increasing intensity until a maximal response was recorded. The CMAP (M wave) was recorded using microneedle electrodes carefully positioned under a dissecting microscope at the TA and PL muscles ²⁷. To assess central descending motor pathways, MEPs were recorded from the TA muscles following transcranial electrical stimulation to the motor cortex using 0.1 ms single pulses at maximal intensity. Stimulation electrodes were inserted subcutaneously above the skull, with the cathode situated over the sensorimotor cortex and the anode at the nose. All potentials were amplified (x100 to x1000), band-pass filtered between 10Hz and 2KHz, displayed on a digital oscilloscope (Tektronix 450S), digitized, and stored in LabChart (ADInstruments). CMAP maximal amplitude and minimal onset latency were measured, whereas for MEPs the maximal amplitude was measured for the two components of the response. For the MUNE, the incremental technique protocol was followed by stimulating the sciatic nerve starting with subthreshold intensity and applying pulses of gradually increasing intensity, recording quantal increases in CMAP. Increments exceeding 50 µV in amplitude were considered as recruitment of a new MU. The mean amplitude of single motor units (SMUA) was computed as the average of consistent increases. The estimated number of MU was then determined using the equation: MUNE = CMAP maximal amplitude/SMUA.

Behavioral tests and survival analysis

The Rotarod test was conducted weekly to assess the locomotor coordination, strength, and balance of the animals²⁷. Mice were placed on a rotating rod (Ugo Basile®) at a constant speed of 14 rpm, and the duration each animal remained on the rod was measured. Five trials were given to each mouse, and the longest time until falling was recorded, with a cut-off time of 180 seconds. To evaluate grip strength, mice were positioned on a metallic grid of the Grip Strength Meter (Ugo Basile®) permitting all four paws to grasp the bars while gently pulling them by the tail. The peak pull force just before the mouse released the grid was recorded. Each animal underwent a minimum of three trials with rest intervals in between. For the treadmill running test, mice at 16 weeks of age were placed on a motor-driven 5-lane treadmill (Panlab, Harvard Apparatus). The protocol consisted of a single running session, starting at 5 cm/s and gradually increasing to 30 cm/s over 5 minutes. The maximum speed at which the animals ran before consistently encountering the shock grid was recorded. The onset of clasping behavior was also monitored weekly by gently lifting the mouse by the tail and holding it in a vertical position. To evaluate survival, groups of animals were observed until they reached a humane endpoint. This endpoint was defined by loss of the righting reflex, indicating the moment when a mouse could not turn and sustain an upright standing position for 30 seconds when placed on its side.

Histological analysis

Animals were transcardially perfused with PFA 4% in PBS at 16 weeks of age. Spinal cords underwent post-fixation in 4% PFA in PBS for 2 hours and brains for 24 hours, followed by cryoprotection in a 30% sucrose solution. Muscle samples were directly cryopreserved.

The lumbar spinal cord was serially cut in 20-µm-thick transverse sections. For spinal MN counting, L4-L5 spinal cord sections separated by 100 µm were stained in cresyl violet. Sections were analyzed under a microscope (Nikon Eclipse Ni) and MNs were identified by their localization in the lateral protrusion of the ventral horn, and following strict size and morphological criteria ²⁷. For assessing glial reactivity, spinal cord sections were permeabilized with TBS-0.3% TritonX-100 (Sigma) and blocked with 10% normal donkey serum. Slices were incubated overnight in a blocking solution with primary antibodies anti-Iba-1 (1:500; 19-19741, Rafer, Fujifilm) or anti-GFAP (1:500, 130300; Invitrogen) with anti-Vimentin (1:600, ab92547, Abcam). Following TBS-0.1%Tween20 washes (Thermofisher), sections were then incubated overnight with secondary antibodies Alexa Fluor 488-conjugated (1:600; A21206, Invitrogen) or Cy3-conjugated (1:200; 712-165-150, Jackson IR). Slices were mounted in DAPI Fluoromount-G (ThermoFisher). Labeling was examined under confocal microscopy (LSM 700 Axio Observer, Carl Zeiss). Photographs of the ventral horn were captured and after setting a threshold for background, the integrated density of GFAP (astrocytes) or Iba-1 (microglia) labeling was measured using ImageJ software (NIH).

To detect KL in TA muscles, transversal sections separated by 100 µm were cut at 10-µm-thickness. An in-house made rabbit antibody anti-sKL (K113, 1:1000) ⁵³, was used and the signal was amplified with the Alexa Fluor™ 555 Tyramide SuperBoost™ Kit (B40932, Fisher Scientific). Samples underwent antigen retrieval with citrate buffer (10mM citrate, pH 6,0) and the primary antibody anti-collagen type IV (1:200, 134001, Setareh Biotech) was added simultaneously with blocking buffer and incubated overnight. The next day, slides were incubated with the secondary biotinylated horse anti-rabbit IgG (1:200, BA-1100, Vector) and Alexa Fluor 488 Donkey anti-Goat IgG (1:200, A11055, Invitrogen). Slices were incubated with Streptavidin-HRP (Str-HRP), after which the tyramide signal was developed and stopped within 7 min. Slides were mounted with DAPI Fluoromount-G. Images of the muscles were taken under confocal microscopy and the sKL signal was quantified using ImageJ software (NIH) as the integrated density.

For NMJ analysis, GM muscles were longitudinally sectioned at a thickness of 50 µm. Free-floating sections were boiled in PBS for 5 min and permeabilized in PBS-0.3%Triton-X100 (Sigma). Subsequently, sections were blocked with normal donkey serum 5% and incubated for 48 hours with primary antibodies anti-NF200 (1:1000, AB5539, Merk) and anti-synaptophysin (1:500, ab32127, Abcam). Secondary antibodies Alexa Fluor 594-conjugated goat x chicken (1:200, A11042, Invitrogen), Alexa Fluor 594-conjugated donkey x chicken (1:200, A21207, Invitrogen), and Alexa Fluor 488-conjugated to α-bungarotoxin (1:650, B13422, Invitrogen) were applied overnight. Muscle sections were mounted with DAPI Fluoromount G medium. Images were obtained using a scanning confocal microscope (LSM 700 Axio Observer, Carl Zeiss). The NMJs were classified as either innervated endplates (when presynaptic terminals overlaid or touched the endplate), or vacant (when presynaptic terminals were not within the endplate). A minimum of 4 fields were examined per muscle.

Coronal brain sections were cut at 40-µm thickness and serially collected for the analysis of cortical MNs. The region of interest, corresponding to the primary motor cortex, was defined using a Mouse Brain Atlas from Bregma 1.10 to -0.22 mm. Free-floating sections spaced 200 µm were permeabilized with PBS-0.2%Tx-100 (Sigma), blocked with 10% normal goat serum, and incubated for 48h with mouse antibody against NeuN (1:1000, MAB377, Merck) and subsequently secondary antibody goat anti-mouse Alexa 488 (1:1000, A11001, ThermoFisher) was added. Hoechst (1:10000, 33342, ThermoFisher) in PBS was applied for nuclear contrast, and slices were mounted using Fluoromount G medium (ThermoFisher). M1 cortex tile images were captured by confocal microscopy (LSM 980 Airyscan 2, Carl Zeiss) and analyzed by Imaris Software (version 9.2). The region corresponding to layer V was manually selected, and a 3D surface of neurons was generated for quantifying neuron volume, surface area, and number. For each mouse, three to four sections were quantified, and somas with an area greater 100 µm² were counted (53).

Nucleic acid extraction and RT-PCR

Lumbar spinal cords and pulverized gastrocnemius muscles were homogenized separately using a Tissue Lyser LT at 50 Hz with QIAzol and purified with the RNeasy Plus Mini Kit (Qiagen) to extract total RNA. RNA concentration was measured using a NanoDrop ND-1000 (Thermo Scientific). RNA was treated with DNase I (FisherScientific), and mRNA was reverse transcribed with the iScript cDNA Synthesis Kit (Bio-Rad). Expression analysis was performed using the CFX384 Touch Real-Time PCR Detection System (Bio-Rad) and the SensiFAST™ Probe No-ROX Kit (Bioline Meridian). The PCR protocol included an initial step at 50°C for 15 minutes, heat activation at 95°C for 1 minute, followed by 39 cycles of denaturation at 95°C for 10 seconds and extension at 61°C for 1 minute. Relative quantification to the UXT housekeeping gene was determined for secreted Klotho using the Pfaffl method. The following primers and probes were used: sKL (5’-CAATGGCTTTCCTCCTTTACC-3’, 5’-GAGGCCGACACTGGGTTTTG-3’, 5’-FAM/AGGGACATTTCCCTGTGACTTTGCTTGG/BHQ1-3’); UXT (5’-TCATGGCGACGCCCCCTAAAC-3’, 5’-AAAGCCTCGTAGCGCAGCACT-3’, 5’-HEX/CGGGCCTTGGATACGGTGGGG/BHQ1-3’). For viral genome copy number determination, DNA from spinal cord and livers was extracted with the DNA QIAamp Mini Kit (Qiagen), while DNA from muscles was extracted using 0.1 mg/ml proteinase K (Roche Diagnostics) followed by phenol/chloroform extraction. Primers and probes were designed for the peptidylprolyl isomerase B (PPIB) sequence (5’-AGGAGTGTGCCTCTTGACC-3’, 5’-CACCTTCCGTACCACATCCT-3’, 5’-HEX/CCGCTCAAC/Zen/CTCTCCTCTCCTGCC/31ABKFQ-3’) and for amplifying the SV40 poly(A) within the viral construct (5’-AGCAATAGCATCACAAATTTCACAA-3’, 5’-CAGACATGATAAGATACATTGATGAGTT-3’, 5’-FAM/AGCATTTTTTTCAC TGCATTCTAGTTGTGGTTTGTC/BHQ1-3’). Viral genome copies per cell were determined using a standard curve generated from known amounts of plasmid DNA containing the construct and a PPIB PCR product purified with the Wizard® SV Gel and PCR Clean-Up System (Promega) in 10 ng/ml salmon sperm DNA (Sigma). PPIB was used as a genomic reference, assuming 1 mg of mouse genomic DNA contains 3x10⁵ haploid genomes. The RT-PCR protocol was run with an extension temperature of 58°C.

Protein extraction and detection

For measuring KL concentration, GM muscle and spinal cord samples were kept on ice and homogenized in lysis buffer (25mM HEPES, 0.1% Igepal, 5mM MgCl_2, 1.3mM EDTA, 1mM EGTA, 0.1 mM PMSF) with Protease inhibitor cocktail set (Calbiochem), using a dispersing instrument (IKA™ Disperser ULTRA-TURRAX™ T10 Basic). Samples were centrifuged for 5 min at 14,000g at 4°C. The supernatant was quantified using Pierce™ BCA Protein Assay Kit (ThermoFisher) and the absorbance read (Spark, TECAN). KL concentrations in these samples and serum were detected using a specific ELISA kit (27601, IBL) according to the manufacturer’s instructions.

For proteomic studies, samples were sonicated in lysis buffer (2% SDS, 0.1M dithiothreitol, 100 mM Tris/HCl, pH = 7.6) at 40Hz using a Vibra-cellTM ultrasonic processor (Sonics & Materials Inc) for five 30-second cycles, alternating with 30-second cooling periods on ice. Afterward, they were incubated at 95ºC for 5 minutes with optional soft shaking. A second round of five sonication-ice cycles followed. Finally, the samples were centrifuged at 16,000g and 13ºC for 20 minutes, and the supernatant was quantified with Pierce™ 660 nm Protein Assay Reagent (ThermoFisher). FASP digestion of samples was performed with trypsin (1 µg/µL, Promega), followed by cleanup using C18 tips (PolyLC)⁵⁴. Tryptic samples were quantified at the peptide level using Pierce™ Quantitative Peptide Assays (ThermoFisher) and subjected to nanoLC-MS/MS analysis. Data were acquired using a Data Independent Acquisition (DIA), library-free, label-free strategy.

Differential abundance analysis was carried out using limma, employing a label-free quantification approach with six distinct contrasts: SOD1 Null vs SOD1 sKL, SOD1 Null vs WT sKL, SOD1 Null vs WT Null, SOD1 sKL vs WT sKL, SOD1 sKL vs WT Null, and WT Null vs WT sKL. For protein quantification, MaxLFQ was employed⁵⁵, computed from normalized precursor quantities provided by DIANN using the SwissProt Mouse database⁵⁶ and the ImFit function for model fitting from the limma package⁵⁷. Standard cutoffs for fold change (|FC| > 1.5) and adjusted p-value (p_adj < 0.05) were applied to identify significant proteins using the Benjamini & Hochberg FDR method⁵⁸. Data preprocessing involved excluding protein groups identified solely by a single unique peptide. Proteins with no missing values in at least one condition were retained, and "multiple-member" protein groups redundantly represented by others in the DIA-NN output were eliminated.

RNA library preparation, sequencing, and analysis

Libraries for RNA-seq were prepared at the Functional Genomics Core Facility of IRB-Barcelona. mRNA was isolated from 1.2 µg and 0.12 µg of total RNA extracted from muscle and spinal cord samples, respectively. The extracted mRNA was used to create dual-indexed cDNA libraries employing the Illumina stranded mRNA ligation kit (Illumina) and UD Indexes Set A (Illumina). Muscle and spinal cord batches, each consisting of 16 samples, underwent ten and thirteen cycles of PCR amplification, respectively. The prepared sequencing-ready libraries underwent quantification using the Qubit dsDNA HS assay (Invitrogen) and quality control using the Tapestation HS D5000 assay (Agilent). A unified equimolar pool of the thirty-two libraries was generated and subsequently submitted for sequencing on a NovaSeq6000 S4 (Illumina). The sequencing process yielded more than 362 Gbp, with a minimum of 23.4 million paired-end reads per sample.

For bioinformatics analyses, paired-end reads were aligned with STAR to the Mus musculus genome⁵⁹, gene expression levels were quantified using the R package Rsubread⁶⁰ and data was transformed with variance-stabilizing transformation (VST) for exploration purposes. DESeq2 was used for differential expression analysis with independent filtering of features applied⁶¹, and multiple contrasts were adjusted with Benjamini-Hoechberg FDR⁵⁸. For enrichment analysis, ROAST-GSA hybrid of methods that combines the statistical inference based on limma was employed^57,62,63. To annotate genes according to GO terms labels and KEGG pathways the Bioconductors' R package “org.Mm.eg.db” was used (Carlson M.: Genome wide annotation for Mouse), and GO slim terms downloaded from the GO project webpage (geneontology.org). All these analyses were performed on DESeq2's rlog transformed counts⁶¹. To compare RNAseq and proteomic data, hypergeometric tests for functional enrichment were performed with the top/bottom differentially expressed genes/proteins, filtering them by a p-value lower than 0.1.

Statistical analyses

Data are generally presented as mean ± standard error of the mean (SEM), and statistical significance is established at p-values ≤ 0.05 unless stated otherwise. Statistical analysis and graphical representations were performed using GraphPad Prism (GraphPad Software) and R software. Shapiro-Wilk normality test was applied to verify the normal distribution of the data. Comparisons between the two groups were performed with two-tailed unpaired Student's t-tests. One-way analysis of variance (ANOVA) was used for comparisons of one variable between all groups. Two-way ANOVA was performed when the effect of the dependent variable was tested on the genotype of the animals and the treatment. Following ANOVA, Tukey’s multiple comparisons post-hoc tests were performed to identify significant differences between specific pairs of groups. Mantel-Cox (log-rank) test was used for assessment of the clasping onset and survival analysis. Data from cortical neurons were analyzed following the approach outlined by Becker et al.⁶⁴, using linear mixed models (LMM) implemented with the lme4 R package. The statistics used in RNA-seq and proteomic studies were conducted by the MS/proteomics and the Biostatistics/Bioinformatics facilities at IRB-Barcelona, and are described in their respective sections.

Competing interests:

The authors declare the following competing interests: AB, SV, MHG, XN, and MC are listed as inventors on a patent related to the technology described in this article. This patent is “Treatment of neuromuscular diseases via gene therapy that expresses Klotho protein”, with the filing number PCT/EP2023/059677.

Author contributions:

MC, XN, SV, and AB conceived the project. SV, JD, MC, MHG, XN, and AB were involved in the development and design of the methodology. SV, RGY, NGC, JS, JD, JRS, GC, MLJ, LRE, MHG, and XN participated in the project investigation and experimental procedures. SV performed data visualization, analysis, and curation. MHG, XN, and AB administered and supervised the project. SV wrote the original draft of the manuscript. SV, XN, and AB contributed to writing, reviewing, and editing.

Acknowledgments:

The authors are grateful to J.I. Pons and F. DO. Monteiro from the IRB-Barcelona Functional Genomics for their technical assistance on RNA sequencing. M. Díaz, M. Gay, and M. Vilaseca performed Mass spectrometry/Proteomics at the IRB-Barcelona Mass Spectrometry and Proteomics Core Facility, which is granted in the framework of the 2014–2020 ERDF Operational Programme in Catalonia, co-financed by the European Regional Development Fund (ERDF). Reference: IU16-015983. We acknowledge C. SO. Attolini from the Biostatistics/Bioinformatics service from IRB-Barcelona for performing RNAseq analyses and their integration to proteomic data. We would like to thank the staff from Servei de Microscòpia of Universitat Autònoma de Barcelona for the advice and assistance in confocal microscopy and Imaris software image processing. We also appreciate the help of Neus Hernández, Susana Miravet, Jorge Lunar, Jessica Jaramillo, and Monica Espejo for their technical support. This work was supported by Agència de Gestió d’Ajuts Universitaris i de Recerca (AGAUR), Generalitat de Catalunya (grants 2021 PROD 00030 and 2021-SGR 00529 to AB, grant 2021SGR 00488 to XN, and the predoctoral grant 2019FI_B 00120 to SV), by Ministerio de Ciencia, Innovación y Universidades (PID2020-116735RB-I00 /AEI/10.13039/ 501100011033 to AB), and by Instituto de Salud Carlos III of Spain (CIBERNED CB06/05/1105, and TERAV RD21/0017/0008), co-funded by European Union (NextGenerationEU, Recovery, Transformation and Resilience Plan).

Taylor, J. P., Brown, R. H. & Cleveland, D. W. Decoding ALS: From genes to mechanism. Nature vol. 539 197–206 Preprint at https://doi.org/10.1038/nature20413 (2016).
Shaw, P. J. Molecular and cellular pathways of neurodegeneration in motor neurone disease. Journal of Neurology, Neurosurgery and Psychiatry vol. 76 1046–1057 Preprint at https://doi.org/10.1136/jnnp.2004.048652 (2005).
Mead, R. J., Shan, N., Reiser, H. J., Marshall, F. & Shaw, P. J. Amyotrophic lateral sclerosis: a neurodegenerative disorder poised for successful therapeutic translation. Nature Reviews Drug Discovery vol. 22 185–212 Preprint at https://doi.org/10.1038/s41573-022-00612-2 (2023).
Ripps, M. E., Huntley, G. W., Hof, P. R., Morrison, J. H. & Gordon, J. W. Transgenic mice expressing an altered murine superoxide dismutase gene provide an animal model of amyotrophic lateral sclerosis. Proceedings of the National Academy of Sciences 92, 689–693 (1995).
Mancuso, R. & Navarro, X. Amyotrophic lateral sclerosis: Current perspectives from basic research to the clinic. Progress in Neurobiology vol. 133 1–26 Preprint at https://doi.org/10.1016/j.pneurobio.2015.07.004 (2015).
Zeldich, E. et al. The neuroprotective effect of Klotho is mediated via regulation of members of the redox system. Journal of Biological Chemistry 289, 24700–24715 (2014).
Yamamoto, M. et al. Regulation of oxidative stress by the anti-aging hormone klotho. Journal of Biological Chemistry 280, 38029–38034 (2005).
Chen, C. Di et al. The antiaging protein klotho enhances oligodendrocyte maturation and myelination of the CNS. Journal of Neuroscience 33, 1927–1939 (2013).
Zeldich, E., Chen, C. Di, Avila, R., Medicetty, S. & Abraham, C. R. The Anti-Aging Protein Klotho Enhances Remyelination Following Cuprizone-Induced Demyelination. Journal of Molecular Neuroscience 57, 185–196 (2015).
Roig-Soriano, J. et al. AAV‐mediated expression of secreted and transmembrane αKlotho isoforms rescues relevant aging hallmarks in senescent SAMP8 mice. Aging Cell 21, (2022).
Shaker, M. R., Aguado, J., Chaggar, H. K. & Wolvetang, E. J. Klotho inhibits neuronal senescence in human brain organoids. NPJ Aging Mech Dis 7, 18 (2021).
Zhu, L. et al. Klotho controls the brain–immune system interface in the choroid plexus. Proceedings of the National Academy of Sciences 115, (2018).
Castner, S. A. et al. Longevity factor klotho enhances cognition in aged nonhuman primates. Nat Aging 3, 931–937 (2023).
Li, Q. et al. Klotho regulates CA1 hippocampal synaptic plasticity. Neuroscience 347, 123–133 (2017).
Sahu, A. et al. Age-related declines in α-Klotho drive progenitor cell mitochondrial dysfunction and impaired muscle regeneration. Nat Commun 9, (2018).
Sahu, A. et al. Regulation of aged skeletal muscle regeneration by circulating extracellular vesicles. Nat Aging 1, 1148–1161 (2021).
Ahrens, H. E., Huettemeister, J., Schmidt, M., Kaether, C. & von Maltzahn, J. Klotho expression is a prerequisite for proper muscle stem cell function and regeneration of skeletal muscle. Skelet Muscle 8, (2018).
Yuan, Q. et al. A Klotho-derived peptide protects against kidney fibrosis by targeting TGF-β signaling. Nat Commun 13, 438 (2022).
Anamizu, Y. et al. Klotho insufficiency causes decrease of ribosomal RNA gene transcription activity, cytoplasmic RNA and rough ER in the spinal anterior horn cells. Acta Neuropathol 109, 457–466 (2005).
Kuro-O, M. et al. Mutation of the Mouse Klotho Gene Leads to a Syndrome Resembling Ageing. NATURE vol. 390 (1997).
Zeldich, E. et al. Klotho Is Neuroprotective in the Superoxide Dismutase (SOD1G93A) Mouse Model of ALS. Journal of Molecular Neuroscience 69, 264–285 (2019).
Kurosu, H. et al. Suppression of Aging in Mice by the Hormone Klotho. Science (1979) 309, 1829–1833 (2005).
Dewil, M., dela Cruz, V. F., Van Den Bosch, L. & Robberecht, W. Inhibition of p38 mitogen activated protein kinase activation and mutant SOD1G93A-induced motor neuron death. Neurobiol Dis 26, 332–341 (2007).
Gould, T. W. et al. Complete Dissociation of Motor Neuron Death from Motor Dysfunction by Bax Deletion in a Mouse Model of ALS. The Journal of Neuroscience 26, 8774–8786 (2006).
Dengler, R. et al. Amyotrophic lateral sclerosis: macro-EMG and twitch forces of single motor units. Muscle Nerve 13, 545–50 (1990).
Fischer, L. R. et al. Amyotrophic lateral sclerosis is a distal axonopathy: evidence in mice and man. Exp Neurol 185, 232–240 (2004).
Mancuso, R., Santos-Nogueira, E., Osta, R. & Navarro, X. Electrophysiological analysis of a murine model of motoneuron disease. Clinical Neurophysiology 122, 1660–1670 (2011).
Mancuso, R., Osta, R. & Navarro, X. Presymptomatic electrophysiological tests predict clinical onset and survival in SOD1 ^G93A ALS mice. Muscle Nerve 50, 943–949 (2014).
Marcuzzo, S. et al. Hind limb muscle atrophy precedes cerebral neuronal degeneration in G93A-SOD1 mouse model of amyotrophic lateral sclerosis: A longitudinal MRI study. Exp Neurol 231, 30–37 (2011).
Wong, M. & Martin, L. J. Skeletal muscle-restricted expression of human SOD1 causes motor neuron degeneration in transgenic mice. Hum Mol Genet 19, 2284–2302 (2010).
Jokic, N. et al. The neurite outgrowth inhibitor Nogo-A promotes denervation in an amyotrophic lateral sclerosis model. EMBO Rep 7, 1162–1167 (2006).
Martineau, É., Arbour, D., Vallée, J. & Robitaille, R. Properties of Glial Cell at the Neuromuscular Junction Are Incompatible with Synaptic Repair in the SOD1 ^G37R ALS Mouse Model. The Journal of Neuroscience 40, 7759–7777 (2020).
Frey, D. et al. Early and Selective Loss of Neuromuscular Synapse Subtypes with Low Sprouting Competence in Motoneuron Diseases. (2000).
Azzouz, M. et al. Progressive motor neuron impairment in an animal model of familial amyotrophic lateral sclerosis. Muscle Nerve 20, 45–51 (1997).
Tabebordbar, M. et al. Directed evolution of a family of AAV capsid variants enabling potent muscle-directed gene delivery across species. Cell 184, 4919–4938.e22 (2021).
Mòdol-Caballero, G. et al. Specific Expression of Glial-Derived Neurotrophic Factor in Muscles as Gene Therapy Strategy for Amyotrophic Lateral Sclerosis. Neurotherapeutics 18, 1113–1126 (2021).
Mòdol-Caballero, G. et al. Gene therapy for overexpressing Neuregulin 1 type I in skeletal muscles promotes functional improvement in the SOD1 < sup > G939 < sup > ALS mice. Neurobiol Dis 137, (2020).
Mòdol-Caballero, G. et al. Gene Therapy Overexpressing Neuregulin 1 Type I in Combination With Neuregulin 1 Type III Promotes Functional Improvement in the SOD1G93A ALS Mice. Front Neurol 12, (2021).
Duan, D. Lethal immunotoxicity in high-dose systemic AAV therapy. Molecular Therapy 31, 3123–3126 (2023).
Arjomandnejad, M., Dasgupta, I., Flotte, T. R. & Keeler, A. M. Immunogenicity of Recombinant Adeno-Associated Virus (AAV) Vectors for Gene Transfer. BioDrugs vol. 37 311–329 Preprint at https://doi.org/10.1007/s40259-023-00585-7 (2023).
Mead, R. J. et al. Optimised and Rapid Pre-clinical Screening in the SOD1G93A Transgenic Mouse Model of Amyotrophic Lateral Sclerosis (ALS). PLoS One 6, e23244 (2011).
Wayman, G. A., Lee, Y.-S., Tokumitsu, H., Silva, A. & Soderling, T. R. Calmodulin-Kinases: Modulators of Neuronal Development and Plasticity. Neuron 59, 914–931 (2008).
Ruegsegger, C. et al. Aberrant association of misfolded SOD1 with Na+/K + ATPase-α3 impairs its activity and contributes to motor neuron vulnerability in ALS. Acta Neuropathol 131, 427–451 (2016).
Roig-Soriano, J. et al. Differential toxicity profile of secreted and processed α-Klotho expression over mineral metabolism and bone microstructure. Sci Rep 13, (2023).
Boillée, S. et al. Onset and Progression in Inherited ALS Determined by Motor Neurons and Microglia. Science (1979) 312, 1389–1392 (2006).
Yamanaka, K. et al. Astrocytes as determinants of disease progression in inherited amyotrophic lateral sclerosis. Nat Neurosci 11, 251–253 (2008).
Leon, J. et al. Peripheral Elevation of a Klotho Fragment Enhances Brain Function and Resilience in Young, Aging, and α-Synuclein Transgenic Mice. Cell Rep 20, 1360–1371 (2017).
Hu, M. C. et al. Renal Production, Uptake, and Handling of Circulating αKlotho. Journal of the American Society of Nephrology 27, 79–90 (2016).
Calvo, A. C. et al. Genetic biomarkers for ALS disease in transgenic SOD1 G93A mice. PLoS One 7, (2012).
Calvo, A. C. et al. Collagen XIX Alpha 1 Improves Prognosis in Amyotrophic Lateral Sclerosis. Aging Dis 10, 278 (2019).
Zolotukhin, S. et al. Recombinant adeno-associated virus purification using novel methods improves infectious titer and yield. Gene Ther 6, 973–985 (1999).
Piedra, J. et al. Development of a Rapid, Robust, and Universal PicoGreen-Based Method to Titer Adeno-Associated Vectors. Hum Gene Ther Methods 26, 35–42 (2015).
Massó, A. et al. Secreted and transmembrane αklotho isoforms have different spatio-temporal profiles in the brain during aging and Alzheimer’s disease progression. PLoS One 10, (2015).
Wiśniewski, J. R., Zougman, A., Nagaraj, N. & Mann, M. Universal sample preparation method for proteome analysis. Nat Methods 6, 359–362 (2009).
Cox, J. et al. Accurate Proteome-wide Label-free Quantification by Delayed Normalization and Maximal Peptide Ratio Extraction, Termed MaxLFQ. Molecular & Cellular Proteomics 13, 2513–2526 (2014).
Demichev, V., Messner, C. B., Vernardis, S. I., Lilley, K. S. & Ralser, M. DIA-NN: neural networks and interference correction enable deep proteome coverage in high throughput. Nat Methods 17, 41–44 (2020).
Ritchie, M. E. et al. limma powers differential expression analyses for RNA-sequencing and microarray studies. Nucleic Acids Res 43, e47–e47 (2015).
Benjamini, Y. & Hochberg, Y. Controlling the False Discovery Rate: A Practical and Powerful Approach to Multiple Testing. Journal of the Royal Statistical Society: Series B (Methodological) 57, 289–300 (1995).
Dobin, A. et al. STAR: ultrafast universal RNA-seq aligner. Bioinformatics 29, 15–21 (2013).
Liao, Y., Smyth, G. K. & Shi, W. The R package Rsubread is easier, faster, cheaper and better for alignment and quantification of RNA sequencing reads. Nucleic Acids Res 47, e47–e47 (2019).
Love, M. I., Huber, W. & Anders, S. Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biol 15, 550 (2014).
Wu, D. et al. ROAST: rotation gene set tests for complex microarray experiments. Bioinformatics 26, 2176–2182 (2010).
Efron, B. & Tibshirani, R. On testing the significance of sets of genes. Ann Appl Stat 1, (2007).
Becker, L. A. et al. Therapeutic reduction of ataxin-2 extends lifespan and reduces pathology in TDP-43 mice. Nature 544, 367–371 (2017).

Yes there is potential Competing Interest. The authors declare the following competing interests: AB, SV, MHG, XN, and MC are listed as inventors on a patent related to the technology described in this article. This patent is “Treatment of neuromuscular diseases via gene therapy that expresses Klotho protein”, with the filing number PCT/EP2023/059677.

SupplementaryInformation.docx

Download PDF

Version 1

posted

You are reading this latest preprint version

Muscle-targeted Klotho Gene Therapy Ameliorates ALS Hallmarks by Addressing Multiple Disease Mechanisms in SOD1^G93A Mice

Status:

Version 1

Abstract

Figures

INTRODUCTION

RESULTS

AAV8-mediated delivery of sKL to muscles preserves neuromuscular and motor function in SOD1^G93A mice

KL muscle secretion slows down functional decline in symptomatic SOD1^G93A mice

Widespread overexpression of sKL fails to preserve neuromuscular function in SOD1^G93A mice

DISCUSSION

METHODS

Study design

Animals

Virus production

Electrophysiological tests

Behavioral tests and survival analysis

Histological analysis

Nucleic acid extraction and RT-PCR

Protein extraction and detection

RNA library preparation, sequencing, and analysis

Statistical analyses

Declarations

Competing interests:

Author contributions:

Acknowledgments:

References

Additional Declarations

Supplementary Files

Status:

Version 1

Muscle-targeted Klotho Gene Therapy Ameliorates ALS Hallmarks by Addressing Multiple Disease Mechanisms in SOD1G93A Mice

Status:

Version 1

Abstract

Figures

INTRODUCTION

RESULTS

AAV8-mediated delivery of sKL to muscles preserves neuromuscular and motor function in SOD1G93A mice

KL muscle secretion slows down functional decline in symptomatic SOD1G93A mice

Widespread overexpression of sKL fails to preserve neuromuscular function in SOD1G93A mice

DISCUSSION

METHODS

Study design

Animals

Virus production

Electrophysiological tests

Behavioral tests and survival analysis

Histological analysis

Nucleic acid extraction and RT-PCR

Protein extraction and detection

RNA library preparation, sequencing, and analysis

Statistical analyses

Declarations

Competing interests:

Author contributions:

Acknowledgments:

References

Additional Declarations

Supplementary Files

Status:

Version 1

Muscle-targeted Klotho Gene Therapy Ameliorates ALS Hallmarks by Addressing Multiple Disease Mechanisms in SOD1^G93A Mice

AAV8-mediated delivery of sKL to muscles preserves neuromuscular and motor function in SOD1^G93A mice

KL muscle secretion slows down functional decline in symptomatic SOD1^G93A mice

Widespread overexpression of sKL fails to preserve neuromuscular function in SOD1^G93A mice