(C) PLOS One
This story was originally published by PLOS One and is unaltered.
. . . . . . . . . .



The RSK2-RPS6 axis promotes axonal regeneration in the peripheral and central nervous systems [1]

['Charlotte Decourt', 'Univ. Grenoble Alpes', 'Inserm', 'Grenoble Institut Neurosciences', 'Grenoble', 'Julia Schaeffer', 'Beatrice Blot', 'Antoine Paccard', 'Blandine Excoffier', 'Mario Pende']

Date: 2023-04

Unlike immature neurons and the ones from the peripheral nervous system (PNS), mature neurons from the central nervous system (CNS) cannot regenerate after injury. In the past 15 years, tremendous progress has been made to identify molecules and pathways necessary for neuroprotection and/or axon regeneration after CNS injury. In most regenerative models, phosphorylated ribosomal protein S6 (p-RPS6) is up-regulated in neurons, which is often associated with an activation of the mTOR (mammalian target of rapamycin) pathway. However, the exact contribution of posttranslational modifications of this ribosomal protein in CNS regeneration remains elusive. In this study, we demonstrate that RPS6 phosphorylation is essential for PNS and CNS regeneration in mice. We show that this phosphorylation is induced during the preconditioning effect in dorsal root ganglion (DRG) neurons and that it is controlled by the p90S6 kinase RSK2. Our results reveal that RSK2 controls the preconditioning effect and that the RSK2-RPS6 axis is key for this process, as well as for PNS regeneration. Finally, we demonstrate that RSK2 promotes CNS regeneration in the dorsal column, spinal cord synaptic plasticity, and target innervation leading to functional recovery. Our data establish the critical role of RPS6 phosphorylation controlled by RSK2 in CNS regeneration and give new insights into the mechanisms related to axon growth and circuit formation after traumatic lesion.

Funding: This work was supported by a grant from ANR to SB (ANR-18-CE16-0007). HN is supported by the NRJ Foundation and the European Research Council (ERC-St17-759089). This work was supported by the French National Research Agency under the “Investissements d’avenir” programme (ANR-17-EURE-0003 to CD). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Copyright: © 2023 Decourt et al. This is an open access article distributed under the terms of the Creative Commons Attribution License , which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

In this study, we focus on the mouse lumbar DRG as a model of central and peripheral nervous system regeneration. We analyze a mouse line with unphosphorylable RPS6 to decipher its impact on regeneration. We show that RPS6 phosphorylation on Ser 235–236 residues is essential not only for PNS regeneration but also for the preconditioning effect. Among the 4 RSK, RSK2 is strongly expressed by DRG and its expression is regulated by axon injury. We further show that RSK2 modulates RPS6 phosphorylation to promote spinal cord regeneration, spinal synaptic plasticity, target innervation, and functional recovery in mice. Together, our results provide evidence that the RSK2/RPS6 axis is critical in nervous system regeneration.

Proteins from the p90 S6 kinase (RSK) family are also known regulators of RPS6 phosphorylation [ 25 ]. The RSK protein family is composed of 4 isoforms (RSK1-4), with high homology (from 80% to 87%) [ 26 ]. RSK are mostly activated by extracellular signal-regulated kinase (ERK) and regulate important processes in cells, such as growth, survival, proliferation, and cell cycle progression [ 26 ]. Interestingly, Mao and colleagues found that RSK1 contributes to axon regeneration through activation of pro-regenerative proteins [ 27 ]. However, the contribution of the RSK-RPS6 axis in CNS regeneration has not been addressed yet.

Altogether, the phosphorylation status of RPS6 stands as critical to promote axon regeneration. Yet, the exact role of RPS6 phosphorylation and the mechanisms regulating this posttranslational modification in the process of CNS regeneration remain elusive. Surprisingly, mTOR inhibition in DRG cultures does not impact RPS6 phosphorylation [ 17 ]. This result suggests that other signaling pathways might be controlling RPS6 phosphorylation, beside the mTOR pathway.

One major readout of mTOR activation is the phosphorylation of the ribosomal protein S6 (RPS6) [ 18 ], which belongs to the small 40S subunit of the ribosome, the functional unit of protein synthesis. RPS6 is an RNA-binding protein that stabilizes the ribosome by interacting with the ribosomal RNA [ 19 ]. Among all ribosomal proteins, RPS6 has attracted most attention as it was the first one shown to have inducible posttranslational modifications [ 20 ]. For almost 40 years, RPS6 phosphorylation has been studied, yet, many unknowns remain about its physiological functions [ 21 ]. Interestingly, in the retina, RGC subpopulations that are the most resilient to injury have high endogenous levels of RPS6 phosphorylation, which is maintained after injury [ 10 ]. Nevertheless, whether this phosphorylation is directly associated with mTOR activation remains elusive. Moreover, in some cases, injury signals may trigger specific events to prime neurons towards a pro-regenerative response. This feature has been elegantly described in the model of the dorsal column lesion in the spinal cord [ 22 ]. As part of the CNS, the dorsal column, formed by the central branch of DRG neurons, is not able to regenerate after spinal cord injury. However, a prior lesion of the DRG peripheral branch, which forms, for example, the sciatic nerve at the lumbar level, primes DRG neurons to regenerate their axon in the central branch: this is called the preconditioning effect [ 23 , 24 ]. Interestingly, the level of RPS6 phosphorylation increases in DRG neurons after sciatic nerve injury [ 10 , 18 ].

CNS regenerative failure has both extrinsic and neuronal intrinsic components [ 1 , 2 ]. The mTOR (mammalian target of rapamycin) pathway is one of the key neuronal signaling pathway controlling axon regeneration. Indeed, it has been shown that the activation of mTOR pathway via PTEN ( P hosphatase and TEN sin homolog) deletion in neurons, triggers robust axon regeneration in the visual system and in the corticospinal tract [ 3 – 7 ]. Subsequently, combinatorial/synergistic approaches, which often include mTOR pathway activation, have led to long-distance regeneration [ 8 , 9 ]. Additionally, the analysis of specific retinal ganglion cells (RGC) subpopulations regenerative capacity revealed osteopontin and IGF as regulators of axon regeneration through mTOR activation [ 10 ]. In the PNS, mTOR has also been shown to regulate axon regeneration. However, its exact contribution to this process remains unclear. Indeed, one target of mTOR, S6 kinase 1 (S6K1), inhibits axon regeneration through negative feedback on mTOR [ 11 , 12 ]. In contrast, TSC2 genetic deletion or PTEN inhibition (negative regulators of mTOR pathway) leads to a modest increase of axon regeneration after sciatic nerve lesion [ 13 – 15 ]. Moreover, pharmacological inhibition of mTOR, in cultured DRG (dorsal root ganglia) neurons induces only a mild effect [ 16 , 17 ].

In contrast to developing neurons or the ones from the peripheral nervous system (PNS), mature neurons from the central nervous system (CNS) fail to regenerate their axons after an insult (neurodegenerative diseases or traumatic lesions). Patients must bear irreversible and permanent motor, cognitive and/or sensory disabilities. The continuous increase of such nervous system disorders worldwide, along with the lack of efficient therapies, makes axon regeneration and functional recovery major challenges of public health.

Results

RPS6 phosphorylation and preconditioning effect are not controlled by the mTOR pathway RPS6 phosphorylation is commonly used as a readout of mTOR pathway activation, particularly during nervous system regeneration [3,8]. As RPS6 phosphorylation is key for the preconditioning effect and sciatic nerve regeneration, we asked which signaling pathway controls its phosphorylation in DRG. To do so, we used a pharmacological approach. We performed sciatic nerve crush unilaterally on wild-type mice and 3 days later, we isolated DRG neurons to put them in culture. We used Cycloheximide as a global inhibitor of translation, Rapamycin and Torin1 as inhibitors of mTOR, PF-4708671 as a S6 kinase inhibitor, BRD7389 as an inhibitor of the p90 RSK (S4A Fig), and DMSO as control [29]. One hour after plating, we treated cultures with the drug of interest, then we assessed neurite growth after 16 h. We found that Cycloheximide-mediated inhibition of global translation totally blocks neurite outgrowth, both in naive and in preconditioned DRG cultures (Figs 3A–3C and S4B–S4D). This result shows that protein translation is key for neurite outgrowth in naive and preconditioned cultures. PPT PowerPoint slide

PNG larger image

TIFF original image Download: Fig 3. RSK controls the preconditioning effect in mature DRG neurons. (A) Representative microphotographs of preconditioned cultures of mature DRG neurons treated with DMSO (control), translation inhibitor (Cycloheximide, 2 nM), mTOR inhibitors (Torin1, 5 nM or Rapamycin, 0.1 nM), S6K1 inhibitor (PF-4708671-8uM). Scale bar: 250 μm. (B) Quantification of longest neurite length per neuron 16 h after plating (mean ± SEM, two-way ANOVA, 3–4 independent DRG cultures, approximately 50–100 cells counted per condition per culture (except for Cycloheximide)). (C) Percentage of neurons growing a neurite 16 h after plating (mean ± SEM, two-way ANOVA, 10 random microscopy fields quantified per condition). (D) Representative microphotographs of naive and preconditioned cultures of mature DRG neurons treated with DMSO (control) or RSK inhibitor (BRD-7389 (3 μm)). Scale bar: 250 μm. (E) Quantification of longest neurite length per neuron 16 h after plating in naive condition (mean ± SEM, two-way ANOVA, 3 independent DRG cultures, approximately 50–100 cells counted per condition per culture for DMSO condition; all neurons found with a neurite were quantified in BRD7389 condition). (F) Percentage of neurons growing a neurite 16 h after plating in naive DRG treated with DMSO or BRD7389 (mean ± SEM, paired t test, 10 random microscopy fields quantified per condition). (G) Quantification of the longest neurite length 16 h after plating in PC DRG treated with DMSO or BRD7389 (mean ± SEM, two-way ANOVA, approximately 50–100 cells per condition per culture for DMSO condition; all neurons growing a neurite were quantified in BRD7389 condition). (H). Percentage of neurons growing a neurite 16 h after plating in naive DRG treated with DMSO or BRD7389 (mean ± SEM, paired t test, 5 independent DRG cultures, 10 random microscopy fields were quantified per condition). ⁎⁎⁎p < 0.001, ⁎⁎p < 0.01, ⁎p < 0.05. Raw data can be found in Supporting information (S1 Data). DRG, dorsal root ganglion; mTOR, mammalian target of rapamycin; PC, precondtionned. https://doi.org/10.1371/journal.pbio.3002044.g003 Interestingly, in naive conditions, inhibition of mTOR or S1 kinase did not prevent neurite outgrowth (S4B–S4D Fig). We found no difference in the length of the longest neurite nor in the total number of neurons that grow a neurite between control and mTOR inhibition (Torin1, Rapamycin) treatments (S4C and S4D Fig). Inhibition of S6K with PF-4708671 caused a slight increase of the number of neurons that grow a neurite (6.7% ± 1.6% for DMSO versus 9.0% ± 1.2% for PF-4708671) (S4C and S4D Fig). In preconditioned DRG cultures, mTOR inhibition via Torin1 has a mild effect on the extent of growth (691 ± 58 μm for DMSO versus 594 ± 4 μm for Torin1) (Fig 3A–3C). Unlike Torin1, Rapamycin-treated DRG have fewer growing neurites (35.6% ± 2.6% for DMSO versus 25.8% ± 1.9% for Rapamycin) (Fig 3A–3C). When preconditioned neurons are treated with S6K inhibitor, a slight increase of the longest neurite length is observed (610 μm ± 15 for DMSO versus 694 μm ± 13 for PF-4708671) but the total number of neurons growing a neurite is unchanged (Fig 3A–3C). Altogether, our results show that mTOR nor its downstream effector S6K1 are the main actors of the preconditioning effect. Strikingly, the inhibition of the RSK family with BRD7389 completely blocked neurite outgrowth, both in naive DRG cultures and in preconditioned DRG cultures (Fig 3D–3H). We verified that this effect was not due to drug toxicity as the number of Tuj1-positive cells is similar between DMSO and BRD7389 treatments. This result suggests that RSK is a family of kinases involved in the preconditioning effect.

RSK2 controls the preconditioning effect and sciatic nerve regeneration Next, we asked whether RSK2 was involved in the preconditioning effect. To this end, we modulated RSK2 expression in vivo by intrathecal injection of AAV8 vectors and analyzed the neurite growth of both naive and preconditioned DRG in culture (S7A Fig). Strikingly, overexpression of RSK2 in vivo caused naive DRG to grow significantly longer neurites with fewer ramifications, a phenotype that is identical to the preconditioning effect (Fig 5A–5D). We found that this effect is specific of RSK2, as overexpression of RSK3 in naive DRG does not mimic the preconditioning effect (S7B–S7E Fig). Conversely, inhibition of RSK2 expression in vivo resulted in the loss of the preconditioning effect in DRG of the sciatic nerve injured side. Indeed, in absence of RSK2, preconditioned DRG neurons resemble the naive ones, with shorter, highly ramified neurites (Fig 5H–5K). PPT PowerPoint slide

PNG larger image

TIFF original image Download: Fig 5. RSK2 controls the preconditioning effect and axon regeneration in the PNS. (A) Representative microphotographs of DRG dissociated cultures showing that RSK2 overexpression in naive cultures phenocopies the preconditioning effect. Scale bar: 250 μm. (B–D) Quantification of A. (B) Longest neurite length per neuron 16 h after plating (mean ± SEM, Ordinary one-way ANOVA, 3 independent DRG cultures, approximately 50–100 cells counted per condition per culture). (C) Mean distance between 2 ramifications (mean ± SEM, Ordinary one-way ANOVA, 3 independent DRG cultures, approximately 50 cells counted per condition per culture) and (D) percentage of neurons growing a neurite 16 h after plating (mean ± SEM, two-way ANOVA Tukey’s multiple comparisons test, 10 random microscopy fields quantified per condition per culture). (E) Representative images of the sciatic nerve sections 3 days post-injury from mice intrathecally injected with AAV8-PLAP (control) or AAV8-RSK2. Regenerating axons are labeled with anti-SCG10 antibody (white). The red dashed line indicates the injury site. Scale bar: 500 μm. (F) Quantification of regenerative axons from E (mean ± SEM, multiple t test, at least 6 animals per condition). (G) Regeneration index at 3 dpi (mean ± SEM, unpaired t test, at least 5 animals per condition). (H) Representative microphotographs of DRG dissociated cultures showing that RSK2 inhibition in preconditioned cultures phenocopies the naive condition. Scale bar: 250 μm. (I–K) Quantification of G. (I) Longest neurite length per neuron 16 h after plating (mean ± SEM, Ordinary one-way ANOVA, 3 independent DRG cultures, approximately 50–100 cells counted per condition per culture). (J) Mean distance between 2 ramifications (mean ± SEM, Ordinary one-way ANOVA, 3 independent DRG cultures, approximately 50 cells counted per condition per culture) and (K) percentage of neurons growing a neurite 16 h after plating (mean ± SEM, two-way ANOVA Tukey’s multiple comparisons test, 10 random microscopy fields were quantified per condition). (L) Representative images of the sciatic nerve sections 3 days post-injury from mice injected intrathecally with AAV-Sh-Scrambled or AAV-Sh-RSK2. Regenerating axons are labeled with anti-SCG10 antibody (white). The red dashed line indicates the injury site. (M) Quantification of regenerative axons from L (mean ± SEM, multiple t test, at least 6 animals per condition). (N) Regeneration index at 3 dpi (mean ± SEM, unpaired t test, at least 6 animals per condition). (O) Representative microphotographs of DRG dissociated cultures showing that RSK2 inhibition in PTEN deleted preconditioned cultures phenocopies the naive condition. Scale bar: 250 μm. (P–R) Quantification of O. (P) Longest neurite length per neuron 16 h after plating (mean ± SEM, Ordinary one-way ANOVA, 3 independent DRG cultures, approximately 50–100 cells counted per condition per culture). (Q) Mean distance between 2 ramifications (mean ± SEM, Ordinary one-way ANOVA, 3 independent DRG cultures, approximately 50 counted cells per condition per culture) and (R) percentage of neurons growing a neurite 16 h after plating (mean ± SEM, two-way ANOVA Tukey’s multiple comparisons test, 10 random microscopy fields quantified per condition). (S) Representative images of the sciatic nerve sections 3 days post-injury from mice injected intrathecally with AAV8-Sh-Scrambled or AAV8-Sh-RSK2 and AAV8-CRE in PTENfl/fl mice. Regenerating axons are labeled with anti-SCG10 antibody (white). The red dashed line indicates the injury site. (T) Quantification of regenerative axons from S (mean ± SEM, multiple t test, at least 6 animals per condition). (U) Regeneration index at 3 dpi (mean ± SEM, unpaired t test, at least 5 animals per condition). ⁎⁎⁎p < 0.001, ⁎⁎p < 0.01, ⁎p < 0.05, ns: not significant. Raw data can be found in Supporting information (S1 Data). DRG, dorsal root ganglion; PNS, peripheral nervous system. https://doi.org/10.1371/journal.pbio.3002044.g005 In parallel, we analyzed the regeneration of the sciatic nerve in vivo. We injected intrathecally AAV-RSK2, AAV-shRNA-RSK2, or corresponding controls in 4-week-old animals and performed unilateral sciatic nerve crush 3 weeks later (S7A Fig). The extent of axon regeneration was analyzed by SCG10 immunostaining at 3 dpi. Similarly to the effect on DRG cultures, we found that RSK2 overexpression enhances sciatic nerve regeneration (Fig 5E–5G), with axons extending up to 5 mm from the lesion site. This effect is specific to RSK2, as overexpression of RSK3 did not affect sciatic nerve regeneration (S7F–S7H Fig). In contrast, RSK2 inhibition blocks axon regeneration in the sciatic nerve (Fig 5L–5N). Together, our results demonstrate that RSK2 is critical for peripheral nerve regeneration. To show that mTOR and RSK2 act independently on the preconditioning effect, we activated the mTOR pathway through intrathecal injection of AAV-Cre in PTENf/f mice to delete PTEN [3]. We verified the efficiency of multiple AAV infections and the AAV8-Cre induced recombination by injecting a mix of AAV8-Cre and AAV-GFP intrathecally in the reporter mouse line STOP-tdTomatof/f (S7I and S7J Fig). We found that almost 100% of neurons were expressing tdTomato, 2 weeks after injection and that 94% of these DRG were co-infected with AAV-Cre and AAV-GFP (S7I and S7J Fig). As expected, mTOR activation in naive DRG neurons does not induce the preconditioning effect (Fig 5O–5R). We then analyzed the axon growth outcome of RSK2 inhibition together with mTOR activation in preconditioned DRG neurons. We observed that mTOR activation does not modify the preconditioned effect. In contrast, inhibition of RSK2 in PTEN-deleted neurons blocks the preconditioning effect: neurons grow shorter and highly ramified neurites, as in control condition without preconditioning (Fig 5O–5R). In parallel, we analyzed axon regeneration of sciatic nerve in these mice. We found that PTEN deletion does not significantly improve sciatic nerve regeneration, even if a trend is observed with the regeneration index (S7K–S7M Fig). When AAV8-Sh-RSK2 and AAV-Cre were injected in PTENf/f mice, we found that mTOR pathway activation does not counteract the inhibition of axon regeneration induced by RSK2 knockdown (Fig 5S–5U). Altogether, our results show that RSK2 controls the preconditioning effect and PNS regeneration independently of mTOR.

[END]
---
[1] Url: https://journals.plos.org/plosbiology/article?id=10.1371/journal.pbio.3002044

Published and (C) by PLOS One
Content appears here under this condition or license: Creative Commons - Attribution BY 4.0.

via Magical.Fish Gopher News Feeds:
gopher://magical.fish/1/feeds/news/plosone/

You are viewing proxied material from magical.fish. The copyright of proxied material belongs to its original authors. Any comments or complaints in relation to proxied material should be directed to the original authors of the content concerned. Please see the disclaimer for more details.