Sonic Hedgehog Agonists Induce Repair Schwann Cells

Abstract

Peripheral nerve regeneration relies on repair Schwann cells (SCs) to support axonal regrowth and functional recovery. This study aimed to identify drugs that promote this repair phenotype, which is regulated by the expression of the transcription factor c-Jun. Purmorphamine (PUR) and Smoothened agonist (SAG) are both Sonic Hedgehog (SHH) agonists that have been implicated in promoting regeneration after neurological injury in animal models. Here, we have demonstrated that SHH agonists significantly increased c-Jun expression in rat primary SCs and promoted morphological and functional changes consistent with the repair SC phenotype, including an elongated bipolar morphology and increased secretion of neurotrophic factors. Notably, PUR consistently demonstrated a greater potency in driving these effects compared to SAG at the same concentrations. We also identified 2.5 μM PUR as an effective dosage producing these measurable effects in vitro. Co-culturing dorsal root ganglion (DRG) neurons with PUR-treated SCs resulted in a marked increase in neurite elongation, suggesting cell-based or contact dependent features of repair SCs contribute to axon growth. These findings demonstrate that SHH agonists effectively reprogram SCs into a repair phenotype, which constitutes a potential therapeutic strategy for enhancing nerve regeneration and functional recovery in peripheral nerve injury treatment.

Introduction

Peripheral nerve injuries are a common source of lasting disability, frequently leading to severe functional impairments¹. This limitation is affected by the extent of axonal regeneration, which relies on the function of Schwann cells (SCs) in the peripheral nerve environment². Following injury, SCs de-differentiate into a well-characterized repair phenotype that is responsible for supporting nerve repair processes³. Repair SCs maintain elongated and branched structures to support uninterrupted and robust regeneration within repair-associated structures known as Bands of Büngner³. Additionally, these repair SCs provide a milieu of trophic support to promote axon regeneration. Principal among this trophic support include growth factors, where neurons and their axons express receptors for different growth factors, which permit an environment in the distal nerve conducive to motor and sensory axon growth⁴.

Whereas the acute injury microenvironment downstream from injury stimulates repair SCs, there are features specific to the type of peripheral nerve injury that determine how effectively and for how long this activation is sustained. An increase in the gap length resulting from a nerve transection significantly impairs regeneration. Given that the axonal growth rate is estimated at 1-3 mm/day, longer gaps require a longer time for regenerating axons to reinnervate targets, sometimes in the range of several months⁵. Additionally, prolonged distances reduce the efficacy of SC-secreted growth factors and extracellular matrix (ECM) proteins to migrate across the lesion site to sustain regeneration, often leading to diminished SC activity³. Long nerve gaps can thus result in chronic denervation of the distal nerve and end-target atrophy^{6, 7}.

Additionally, the repair SC phenotype is also transient and based on the microenvironmental signaling cues that result from nerve injury^{8, 9}. In the peripheral nerve injury environment, the number of SCs, repair SCs, and the expression of neurotrophic factors increases, but returns to uninjured levels or lower within 2-3 months⁵. Additionally, from the loss of SCs, an accumulation of endoneurial fibroblasts, collagen, and other ECM modifications ensue, potentially further hindering axon regeneration^{10, 11}. This overall condition, characterized by decreased SC activity with concurrent increased fibroblasts and ECM deposition within the distal nerve, has been described as chronic denervation and results from a sustained loss of axonal support in the nerve environment. Given this, a delay in therapeutic intervention can lead to this chronic denervation where SCs become quiescent and can also undergo potential senescence onset, contributing to atrophy of the tissue. This fact underscores an urgent need for treatments that can reprogram or enhance the repair capacity of SCs^{10, 11}.

A key regulator of this repair phenotype is the transcription factor c-Jun, which is essential for the downstream expression of growth-promoting genes^12-14. Loss of the repair SC phenotype is caused by a decrease in the activation and maintenance of high levels of c-Jun¹². Further, it has been demonstrated that genetically restoring c-Jun levels in SCs can restore their capacity to support axon regeneration^{9, 15}. Thus, c-Jun has been termed a “global amplifier” of the repair phenotype of SCs^{12, 13}. Additionally, c-Jun also regulates genes that are involved in SC proliferation and migration, which may contribute to the formation of the Bands of Büngner, a necessary configuration of SCs to guide regenerating axons to their targets^{9, 15}. Therefore, c-Jun represents a key therapeutic target to drive the repair phenotype of SCs.

The expression of c-Jun has been linked to Sonic Hedgehog (SHH) pathway signaling¹⁶. SHH is a key developmental pathway that regulates many forms of cell growth and differentiation, particularly in the nervous system. c-Jun expression is upregulated by SHH signaling, which also itself leads to consistent modulation of downstream genes that lead to c-Jun transcription^{9, 15}. Additionally, c-Jun itself can also interact with components of the SHH pathway to amplify its effects, potentially promoting cell survival and proliferation¹⁶. Mechanistically, c-Jun binds to injury-induced enhancer sites for Shh, which is only upregulated in repair SCs relative to their other phenotypes.^{17, 18} Expression of Shh then enhances further activation of c-Jun through a well characterized autocrine signaling loop, which helps sustain the repair SC phenotype.^{9, 14} This link between c-Jun and SHH signaling is especially relevant in the context of nerve repair, where both SHH pathway activation and elevated c-Jun levels are observed, thus highlighting the central role of this signaling pathway in promoting the SC repair phenotype after injury.^{9, 15, 19} Axonal injury has been demonstrated to activate the phosphorylation and nuclear translocation of c-Jun, which remains elevated during the nerve regenerative process.^{20, 21} Conversely, c-Jun mutations in rodents have demonstrated limited axonal regeneration following nerve transection.^{12, 21} Therefore, sufficient SHH pathway signaling -via Shh protein or its agonists-may induce repair SCs when the injury signal is absent to activate c-Jun, such as in the setting of chronic denervation.

This study investigates potential pharmacologic approaches to reprogram SCs toward the repair phenotype. Specifically, we evaluated the effects of two SHH pathway agonists, purmorphamine (PUR) and Smoothened agonist (SAG), known to play roles in tissue repair and neural regeneration in animal models after various types of injury. We hypothesize that SHH agonists, such as PUR and SAG, can enhance the expression of repair-associated genes in SCs, providing a promising strategy to reprogram SCs into their repair phenotype and support regeneration. By analyzing key features of the repair SC phenotype, such as morphological changes and c-Jun expression, as well as the impact of SHH agonist treatment on neurite outgrowth from neurons, this study aims to establish a new therapeutic pathway for addressing peripheral nerve injuries.

Materials and Methods

Cell culture

Primary SCs (Catalog #R1700; ScienCell Research Laboratories, Carlsbad, CA), isolated from rat sciatic nerve, were cultured in Schwann cell media (SCM, Catalog #1701; ScienCell Research Laboratories). Culture flasks were pre-coated with 0.01% poly-L-lysine (Catalog #P4707, Sigma-Aldrich, St. Louis, MO) to enhance cell adhesion. The cells were maintained in a humidified atmosphere at 37 °C with 5% CO₂. The media was changed every three days. Cells used in this experiment were up to passage six and plated at a density of .2 x 10⁴ cells/cm² Dorsal root ganglia (DRG) neurons (Catalog #R8820N-10; ScienCell Research Laboratories) were plated at density of 1000 neurons/cm² on 0.01% poly-L-lysine coated tissue culture plates. Rat Ganglion Neuron Culture Media (Catalog #R817K-100, ScienCell Research Laboratories) was used with the provided growth supplements. Media was changed daily. All cultures were maintained in water jacketed incubators at 37 °C with 5% CO₂.

SHH agonist treatment

Purmorphamine (PUR, Catalog #SML0868, Sigma-Aldrich, St. Louis, MO) and Smoothened agonist (SAG, Catalog #566661, Sigma-Aldrich) were prepared as stock solutions at 50 mM in dimethyl sulfoxide (DMSO) and 10 mM in sterile phosphate-buffered saline (PBS, Catalog #P2272, Sigma-Aldrich), respectively. Rat SCs were cultured on poly-L-lysine-coated cover glass and treated with PUR or SAG at concentrations of 0, 1, 2.5, 5, and 10 μM. The cells were incubated with the agonists for 48 hours to assess dose-dependent effects on the repair phenotype.

Immunocytochemistry for SC morphologies and phenotypes

SC morphologies and phenotypes were examined by immunocytochemistry. Cells were fixed with 4% (w/v) paraformaldehyde (PFA) for 30 min, followed by blocking with 5% goat serum in PBS to prevent nonspecific binding. Primary antibodies, such as those against S-100 (Catalog #GA504, Dako Omnis, Singapore) and c-Jun (Catalog #UM800005, OriGene, Rockville, MD) were applied, and the cells were incubated overnight at 4 °C. All antibodies used are summarized in Supplemental Table 1. After three washes with PBS, the samples were incubated with the secondary antibodies for 2 hours at room temperature. Nuclei were counterstained with 4′,6-diamidino-2-phenylindole (DAPI) for 30 min. The samples were imaged using an Olympus FluoView 1000 spectral confocal microscope. For each well, five images were taken from different regions to ensure comprehensive analysis.

Morphological analysis

To evaluate SC morphology, cells were fixed with 4% PFA. The samples were then stained with Alexa Fluor 647 Phalloidin (Catalog #A22287, Invitrogen, Waltham, MA) and DAPI. The cells were then imaged for analyses at 1000× overall magnification on a Leitz Laborlux S microscope. For analysis, a semi-automated digital image-analysis system linked to morphometry macros developed for peripheral nerve analysis (Clemex Vision Professional, Clemex Technologies, Longueuil, Quebec, Canada) was used. Six-eight random fields per sample were imaged, resulting in between 15-25 cells included per sample and averaged to represent that sample. Binary histomorphometry analysis of the digitized information based on gray and white scales allowed for measurements of the total area and shape factor.

Quantitative reverse transcription polymerase chain reaction (qRT-PCR)

To evaluate gene expression, qRT-PCR was performed with a focus on a broad panel of select genes indicative of repair SCs (c-Jun, Sox10, Bdnf, Mbp, and Ki67). For each sample, the RNA was extracted using chloroform and an Rneasy Kit (Qiagen, Valencia, CA) according to the manufacturer’s instructions. The extracted RNA concentration was measured using a NanoDrop 1000 Spectrophotometer (Thermofisher Scientific, Waltham, MA) and adjusted for cDNA preparation. The cDNA was generated with SuperScript II reverse transcriptase (Invitrogen, Carlsbad, CA). To quantify gene expression, real-time PCR was per-formed using a Step One Plus thermocycler (Applied Biosystems, Foster City, CA) and TaqMan Master Mix (Thermofisher Scientific) reagents with specific oligonucleotide prime pairs (Supplementary Table 2). The PCR conditions were a hot start at 95 °C for 10 min, followed by 50 °C for 2 min, 95 °C for 15 s and 60 °C for 1 min intervals, repeated for 40 cycles. The gene expression of the selected genes was normalized to an internal control gene (Actb). The data were analyzed using StepOne Software v3.0.1 (Applied Biosystems, Foster City, CA). Analysis of each sample for a group was considered a replicate.

Protein detection assay

The RayBio^® C-Series Rat Growth Factor Array 1 (Catalog #AAR-CGF-1-4, RayBiotech, Norcross, GA) was used to compare levels of multiple growth factors pertinent to nerve growth, including BDNF, GDNF, NGF, and FGF. For analysis, 1 mL of cell supernatant was collected from cultured SCs treated with 2.5 μM PUR or SAG for 48 hours, as well as from untreated SCs. Antibody arrays were equilibrated to room temperature, blocked with provided buffer for 30 minutes, and incubated with 1 mL of each sample. A total of 3 replicates were obtained per group (PUR, SAG, and untreated), with each representing a distinct culture of SCs plated at a density of 2 × 10⁴ SCs. Antibody array membranes were equilibrated to room temperature prior to use with provided reagents and instructions. Incubation time was 2 hours at room temperature for samples, biotinylated antibody, and labeled streptavidin. Chemiluminescence signals were imaged with a CCD camera. Images were then analyzed using the built-in Gel Analysis function. Background subtraction and positive control normalization were both applied to ensure data accuracy.

Cell migration assay

The effect of PUR or SAG on SC migration was investigated in vitro using ThinCert cell-culture inserts with a polyethylene terephthalate filter (8.4 mm diameter, 8 μm pore size) to form dual compartments in a 24-well tissue culture plate. Then 2 × 10⁴ SCs in 100 μL of the media were added to the upper compartments. The lower compartments were filled with media with different amounts of PUR or SAG (0, 1, 2.5, 5, and 10 μM). The cultures were incubated at 37 °C and 5% CO₂. After 24 h, SCs were fixed with 4% PFA. The SCs that migrated across the filter membrane were stained with Alexa Fluor 546-phalloidin (red) and DAPI (blue). Images were captured using an Olympus FluoView 1000 spectral confocal microscope. At least six random fields per membrane were analyzed to quantify cell migration for each group.

Neuron-based assays

DRG-derived neurons were used to determine how SCs affected neurite outgrowth as a measure of their functional properties. Rat SCs were cultured on 0.01% poly-L-lysine coated glass coverslips in 24 well plates at a density of 2 x 10⁴ cells/cm² as previously described. Then, SCs were treated with either PUR or SAG at 2.5 μM for 48hrs. Untreated groups were also plated in a similar fashion receiving no drug supplementation. After 48hrs, the media from groups was aspirated and collected for SC-conditioned media experiments. This collected conditioned media was aliquoted for immediate use, with the remainder stored in −80°C. For experiments involving SC-conditioned media, DRG neurons were plated at density of 1000 neurons/cm²24 well plate, followed by the immediate addition of conditioned media from either PUR-treated, SAG-treated, or untreated SCs. Separately, DRG neurons were plated and co-cultured at the same density upon the SCs used for SC-conditioned media experiments. As media was previously collected, SC-DRG neuron co-culture groups were incubated with rat ganglion neuron media as previously described. All groups were incubated at 37 °C with 5% CO₂until study end.

Quantification of neurite extension

Neurite extension was quantified following 48 hours of culture. Cells were fixed with 4% PFA and mounted on glass microscope slides. Slides were then stained with primary antibodies against βIII-tubulin (Catalog #T2200, Sigma-Aldrich), followed by secondary antibody with Alexa Fluor 555 conjugation. Stained cells were imaged using an Olympus FluoView 1000 spectral confocal microscope. Images were quantified using ImageJ (NIH). To obtain average neurite length, the length from each neurite was measured and averaged per neuron, with each neuron having one or two neurite projections. The longest neurite per neuron was identified and subsequently analyzed. n = 12 neurons total from 3 distinct replicate wells.

Statistical Analysis

All data are presented as mean ± standard error of the mean (SEM). For statistical analyses, we employed GraphPad Prism version 9. Data sets were tested for normality using the Shapiro–Wilk test. A Student’s t-test was performed for comparison between 2 groups with a Gaussian distribution. We utilized a one-way analysis of variance (ANOVA) to compare the means across multiple groups. Following ANOVA, we conducted Tukey's post hoc test to identify significant differences between groups, with a predetermined alpha level of P < 0.05 indicating statistical significance.

Results

SHH agonists induced dose-dependent c-Jun expression in rat primary SCs

Treatment with SHH agonists PUR and SAG for 48 hours resulted in an increase in c-Jun expression in rat SCs. Representative images reveal that higher concentrations of both SHH agonists correspond to a greater density of c-Jun-expressing cells, thus demonstrating this pattern (Figure 1a). As the concentration of SAG increased from 1 to 10 μM, there was a corresponding dose-dependent increase in c-Jun+ cells (Figure 1b; 0 μM: 11.6%, 1 μM: 22.0%, 2.5 μM: 24.4%, 5 μM: 32.8%, and 10 μM: 44.8%).

Figure 1. SHH agonists increased expression of c-Jun in rat primary SCs.

(a) Representative images of rat SCs with SHH agonists PUR or SAG at concentrations of 0, 1, 2.5, 5, and 10 μM. c-Jun expression was visualized via immunostaining (red), and nuclei were counterstained with DAPI (blue). Scale bar = 200 μm. (b, c) Quantification of the percentage of c-Jun+ cells based on the average of 5 representative images for n = 3 replicate wells per condition. Data represented as mean ± SEM.

A similar dose-dependent increase was observed with PUR treatment. Compared to SAG treatment, PUR treatment resulted in a higher percentage of c-Jun+ cells per equivalent dose (Figure 1c; 0 μM: 8.2%, 1 μM: 19.6%, 2.5 μM: 47.0%, 5 μM: 52.8%, and 10 μM: 74.8%). At 10 μM, PUR treatment resulted in a 67% higher percentage in c-Jun+ cells compared to SAG treatment. These findings demonstrate the potent effect of both PUR and SAG in activating c-Jun expression in SCs, suggesting a robust link between SHH signaling and SC response. Furthermore, the consistently higher percentage of c-Jun+ cells with PUR suggests that it may be a more effective SHH agonist than SAG.

SHH agonists promoted morphological changes consistent with the repair phenotype of SCs

We then investigated the downstream changes of SCs with increased SHH signaling as induced by PUR and SAG. Notable alterations in SC morphology, such as cell shape and size, were associated with PUR and SAG treatment (Figure 2a). These changes were visually characteristic of the repair SC phenotype, especially at high doses of PUR. Increasing doses of PUR resulted in the SC cytoskeleton becoming more elongated and smaller, which is consistent with repair phenotype morphology. SAG treatment also induced qualitative shape changes, though a more modest effect could be observed at 10 μM SAG (Figure 2a).

Figure 2. SHH agonists induce morphological changes consistent with the repair SC phenotype.

(a) Representative images of SC morphology following treatment with SHH agonists. Cells were stained for F-actin using phalloidin (green) to visualize the cytoskeleton and DAPI (blue) for nuclei. Scale bar = 100 μm. (b) Quantification of total cell volume, cell body surface area, and length of SC demonstrates a significant reduction in with SHH agonist treatment at 2.5 μM for PUR and 10 μM for SAG. # P< 0.05. Data represented as mean ± SEM. Data based on average of 5 representative images for n = 3 replicate wells per condition.

The volume, area, and length of SCs calculated using digital image analysis confirmed these morphological changes. PUR treatment resulted in a significant decrease in SC volume starting at 2.5 μM, after which the values plateau with increased concentrations (Figure 2b; 0 μM: 9213 μm³, 1 μM: 9907 μm³, 2.5 μM: 5183 μm³, 5 μM: 5185 μm³, and 10 μM: 5419 μm³). Similarly, the area of SCs significantly at 2.5 μM, with no further decreases at higher doses (Figure 2b; 0 μM: 462.3 μm², 1 μM: 492.1 μm², 2.5 μM: 318.5 μm², 5 μM: 325.6 μm², and 10 μM: 321.3 μm²). The same trend was observed when comparing length of PUR-treated SCs. The length decreased significantly at the 2.5 μM, but not further decreases were observed at higher ones (Figure 2b; 0 μM: 11.2 μm, 1 μM: 11.7 μm, 2.5 μM: 9.4 μm, 5 μM: 9.6 μm, and 10 μM: 9.2 μm).

While SC area and volume were significantly decreased after treatment with PUR at 2.5 μM concentration, more modest effects were observed with SAG treatment. The volume of SCs treated with SAG decreased to significant levels only at 10 μM (Figure 2c; 0 μM: 9213 μm³, 1 μM: 7908 μm³, 2.5 μM: 8398 μm³, 5 μM: 7512 μm³, and 10 μM: 3845 μm³). The area of SCs demonstrated this same pattern, with significant decreases only noted at 10 μM (Figure 2c; 0 μM: 462.3 μm², 1 μM: 436.6 μm², 2.5 μM: 421.3 μm², 5 μM: 411.5 μm², and 10 μM: 266.0 μm²). Similarly, the length was only significantly reduced at 10 μM (Figure 2c; 0 μM: 11.2 μm, 1 μM: 11.1 μm, 2.5 μM: 10.5 μm, 5 μM: 10.7 μm, and 10 μM: 8.7 μm)

Comparing the effects between PUR and SAG on the morphology of SCs demonstrates that both agonists induce significant decreases in cell volume, area, and length, but at different threshold concentrations. This threshold concentration is higher for SAG at 10 μM compared to 2.5 μM for PUR, representing the significantly higher potency of PUR in driving these morphological changes. Collectively, these changes suggest further a shift towards the repair phenotype, underscoring the role of SHH signaling in promoting this cellular transition.

SHH agonists upregulated repair-associated gene expression in rat primary SCs

To further characterize the SC phenotype induced by SHH agonists, we investigated the expression of key repair-associated genes using qPCR. Figure 3 shows that both SAG and PUR effectively upregulate the expression of c-Jun, Sox10, Bdnf, and Mbp in a dose-dependent manner.

Figure 3. SAG and PUR upregulate repair-associated gene expression in SCs.

The relative expression levels of repair-associated genes were significantly upregulated in primary SCs as the dose of (a) PUR and (b) SAG increased. # P< 0.05. Data represented as mean ± SEM. n = 3 replicates.

SAG significantly enhanced the expression of these repair-associated genes (Figure 3a; c-Jun: 3.0-fold; Sox10: 7.6-fold; Bdnf: 3.3-fold; Mbp: 2.2-fold). PUR also upregulated expression of these genes with increasing dosages. At a dose of 10 μM, these genes were significantly highly expressed in SCs compared to those with the vehicle control (Figure 3b; c-Jun: 5.0-fold; Sox10: 6.9-fold; Bdnf: 5.0-fold; Mbp: 3.5-fold).

Expression levels of c-Jun, an essential regulator of the repair SC phenotype, were confirmed to be elevated, aligning with the immunocytochemistry results in Figure 1 that showed similar upregulation with SHH signaling. Sox10 is important for SC lineage maintenance, thus correlating while Bdnf (brain-derived neurotrophic factor) enhances neuronal survival and supports axon regrowth. Finally, Mbp (myelin basic protein) plays a crucial role in remyelination. These results demonstrate significant upregulation of a repair-oriented gene expression profile in SCs, underscoring the potency of SAG and PUR in driving SC plasticity and regenerative function.

SHH agonist treatment differentially modulated growth factor secretion from SCs

To further investigate the functional effects of SHH agonists on SC secretory activity, we characterized the release of neurotrophic factors following PUR and SAG treatment. Having identified the minimum threshold concentration that produces measurable effects in SCs, we selected 2.5 μM as the concentration to base further comparison between agonists. Using a protein-based detection array, we quantified multiple neurotrophic factors that are critical markers for neuronal growth and survival (Figure 4a). Here, we normalized levels of released growth factors to those released by untreated cells.

Figure 4. SHH agonists regulate the expression of neurotrophic growth factors.

(a) Representative image of protein array membrane showing the detection of neurotrophic factors in media collected from Schwann cells treated with SHH agonists. (b) Quantitative analysis reveals relative changes in neurotrophic factor release. Significantly higher levels of secreted BDNF, NGF, and IGFBP-5 and VEGF-A were demonstrated in response to PUR treatment. On the other hand, SCs treated with either PUR or SAG released significantly lower levels of GDNF and GMCSF. Data represented as mean ± SEM, normalized relative to levels from untreated SCs. # P< 0.05. n = 3 replicate wells.

Treatment with PUR and SAG resulted in higher BDNF levels relative to control (Figure 4b; SAG: 1.3-fold; PUR: 1.7-fold). Levels of nerve growth factor (NGF) were also significantly elevated after PUR treatment compared with SAG treatment (Figure 4b; NGF: SAG: 0.8-fold; PUR: 2.8-fold). On the other hand, levels of GDNF and GMCSF were decreased in SCs treated with PUR or SAG (Figure 4b; GDNF: SAG: 0.4-fold; PUR: 0.3-fold; GMCSF: SAG: 0.7-fold; PUR: 0.2-fold). Some of the highest levels of increased secreted growth factors were insulin growth factor binding protein 5 (IGFBP-5) and vascular endothelial growth factor (VEGF-A), with PUR resulting in significantly higher levels relative to those from untreated control and SAG treated cells (Figure 4b; IGFBP-5: SAG: 0.5-fold; PUR: 4.7-fold; VEGF-A: SAG: 0.5-fold; PUR: 7.0-fold). These results demonstrate that SHH agonists differentially regulate the secretion of key neurotrophic and angiogenic factors, with PUR exerting a stronger effect on the release of NGF, IGFBP-5 and VEGF-A. The enhanced secretion of these factors suggests that PUR may play a more potent role in modulating SC-mediated neuronal growth and vascular support, further highlighting its potential in regeneration applications.

SHH agonists increased active SC migration across a permeable membrane

We used a similar experimental paradigm to investigate the effects of SHH agonists on the migratory abilities of SCs using a transwell system as demonstrated by our schematic (Figure 5a). This system allowed us to quantify the number of cells migrating through 8 μm pores following 24 hours of incubation. Figure 5b shows the representative images of DAPI-stained cells on the lower membrane surface.

Figure 5. SHH agonists increase SC migration across a permeable membrane.

SCs were incubated for 24 hrs on a polyethylene terephthalate membrane with 8μm pores and 6.5mm in diameter. (a) Schematic illustration of SC migration assay using a transwell system. (b) Representative image showing DAPI-stained nuclei of migrated SCs on the underside of the membrane. (c and d) Quantification of migrated SC density after 24 hours of incubation with various concentrations of SHH agonists PUR and SAG. (e) Comparison of migrated SC density at 2.5 μM concentrations of PUR and SAG. Data represented as mean ± SEM. Data based on average of 5 representative images for n = 3 replicate wells per condition.

Treatment with PUR did not result in statistically higher average density of migrating cells on a dose-dependent manner, although it is worth noting that significant variance of cell counts existed between analyzed fields (Figure 5c; 0 μM: 435 cells/cm²; 1 μM: 983 cells/cm²; 2.5 μM: 1309 cells/cm²; 5 μM: 1672 cells/cm²; 10 μM: 1278 cells/cm²). Similar trends were noted with SAG treatment, with no statistical difference between samples given high variance (Figure 5d; 0 μM: 435 cells/cm²; 1 μM: 903 cells/cm²; 2.5 μM: 1081 cells/cm²; 5 μM: 1475 cells/cm²; 10 μM: 943 cells/cm²). We demonstrate the results of these experiments with the effective dose of 2.5 μM (Figure 5e; untreated: 435 cells/cm²; PUR: 1309 cells/cm²; SAG: 1081 cells/cm²).

SC-conditioned media from SHH agonist-treated SCs did not change neurite elongation

To evaluate the functional effects of SHH agonist-induced SCs on neurons, we first investigated whether SC-conditioned media could enhance induce neurite elongation. Neurons were treated with media collected from SCs that had been exposed to PUR or SAG at 2.5 μM for 48 hours, as well as with standard culture media as a control. Additionally, a direct treatment with BDNF at 10 ng/mL served as a positive comparison for neurite outgrowth (Figure 6a).

Figure 6. SC-conditioned media alone does not enhance neurite elongation in rat sensory DRG neurons.

(a) Representative images of DRG neurons treated with SC-conditioned media collected after 48 hours of treatment with PUR (2.5 μM), SAG (2.5 μM), or standard culture media. Cells stained with Alexa-Fluor Phalloidin 647 (red) and DAPI (blue). Direct treatment with brain-derived neurotrophic factor (BDNF, 10 ng/mL) was included as a positive control. Scale bar = 100 μm. (b) Quantification of neurite elongation shows no significant differences among neurons treated with conditioned media from the different groups compared to the untreated control. # P< 0.05. n = 12 neurons from 3 distinct replicate culture wells. Data represented as mean ± SEM.

Quantitative analysis reveals that there was no significant increase in neurite elongation among the groups treated with SC-conditioned media, whether this was from SCs treated with PUR, SAG, or remained untreated (Figure 6b; untreated: 80 μm, PUR: 61 μm, SAG: 57 μm). In contrast, neurons treated directly with BDNF exhibited enhanced neurite outgrowth compared to standard neuron media controls, underscoring the efficacy of BDNF in promoting neuronal elongation (Figure 6b; no BDNF: 63 μm, BDNF, 139 μm). These results indicate that, under the conditions tested, conditioned media from SHH agonist-treated SCs alone is not sufficient to produce significant neurite elongation, suggesting that additional non-secretory factors, such as direct SC-neuron interactions, may be required to produce robust neurite outgrowth.

Co-culture with SHH agonist-treated SCs increased neurite elongation

Given the direct contact through which SCs interact with axons in peripheral nerve, we explored whether co-culturing neurons with SHH-agonist-treated SCs could neurite elongation. Specifically, neurons were co-cultured with SCs that had been pre-treated for 48 hours with 2.5 μM of PUR or SAG, as well as with SCs maintained in standard culture media as a control. Co-culture with SHH agonist-treated SCs demonstrated increased neurite extension with these agents compared to the control (Figure 7a). Quantitative analysis shows that average neurite elongation significantly improved in the PUR-treated co-culture group relative to control (Figure 7b; untreated: 279 μm, PUR-treated SC co-culture: 462 μm, SAG-treated SC co-culture: 427 μm). However, the longest neurite length was not changed between groups (Figure 7c; untreated: 339 μm, PUR-treated SC co-culture: 508 μm, SAG-treated SC co-culture: 488 μm). This finding emphasizes the functional implications of SHH agonist treatment in SCs. While SC-conditioned media alone (Figure 6) did not yield significant neurite growth, the direct presence of SHH-treated SCs within the neuronal environment promotes substantial neurite elongation. This suggests that SHH-activated SCs may provide essential contact-dependent cues or secrete additional factors in close proximity that facilitate axonal elongation. These results underscore the potential of SHH-activated SCs to foster a more supportive regenerative microenvironment for neuronal repair.

Figure 7. Neuronal co-culture with SCs treated with SAG and PUR significantly enhances neurite elongation.

(a) Representative images of rat sensory DRG neurons co-cultured with SCs first treated for 48 hours with SAG or PUR. Scale bar = 100 μm. Neurons were labeled with Tuj1 (red) and DAPI (blue). Background SCs labeled with S-100 (green) and DAPI (blue). Neurite traces show below each representative images. (b) Quantification of neurite elongation demonstrates a significant increase in neurons co-cultured with PUR-treated SCs compared to the control. (c) Length of longest neurite per neuron is not significantly affected by PUR or SAG treatment. # P< 0.05. n = 12 neurons total from 3 distinct replicate wells. Data represented as mean ± SEM.

Discussion

This study demonstrates the significance of SHH pathway activation in SCs as a key driver to promote their functional repair phenotype. Our findings underscore that activating SHH signaling in SCs through agonists, such as PUR and SAG, promotes a robust shift towards their repair phenotype, marked by enhanced expression of c-Jun, characteristic morphological adaptation, increased neurotrophic factor secretion, and facilitated neurite growth of primary neurons. These changes suggest that SHH signaling may prime SCs to respond to neuronal injury and contribute to the regenerative process.

Our findings suggest that SHH agonists are effective in stimulating c-Jun expression in a dose-dependent manner. Given that c-Jun is a critical transcription factor that determines the repair SC phenotype^{12, 14, 22, 23}, these results suggest that the SHH signaling pathway plays a pivotal role in both initiating and sustaining this phenotype. These results are observable through immunocytochemical analysis and qPCR. Gene expression analyses further support the broad regulatory capacity of SHH signaling on the SC repair phenotype, underscoring its ability to activate genes involved in SC lineage maintenance, axonal support, and remyelination. By driving expression of factors involved in the regulation of neuronal axons, such as BDNF^24-26, as well as markers of new axonal assembly and myelination, including myelin basic protein (MBP)^{27, 28}, SHH signaling may enhance the potential of SCs to sustain axons and contribute to their structural repair. There is some evidence that constitutive expression of c-Jun may delay or inhibit myelination, suggesting that c-Jun negatively regulates myelinating processes, including downregulating MBP, in a dose-dependent manner.²³ Given that PUR treatment resulted in increases in both c-Jun and MBP, we thus hypothesize that SHH pathway activation with PUR may represent a deeper alignment with the repair phenotype, wherein SCs are primed to create an environment that aids neuronal survival, re-establishment of connections, eventual myelination, and functional recovery^{4, 13}.

Beyond c-Jun expression, morphological adaptations induced by SHH agonists, such as elongation and reduced cell body size, highlight structural shifts that may be vital for SC-neuron interaction during regeneration, particularly in the formation of the bands of Büngner¹³. The reorganization of SC morphology in response to SHH signaling not only aligns with cellular characteristics typically observed in repair SCs, but also suggests that these changes might facilitate improved movement, axonal contact, and guidance of regenerating axons¹³.

A unique highlight of this study was identifying the concentration of these drugs at which an effect was observable. Specifically, 2.5 μM of PUR was determined to be the concentration that reliably produced an effect on SCs and their functional behavior. This concentration was selected as a benchmark for both agonists, allowing for direct comparisons on their efficacy. Notably, while SAG also produced a dose-dependent increase in the expression of pro-repair genes, its effects were consistently less pronounced across multiple assays compared to PUR and exhibited diminished activity at this concentration. These results suggest a more potent effect of PUR on SHH pathway activation, potentially enhancing the SC capacity to sustain the repair phenotype. Understanding the differential activation of neurotrophic factors by SHH agonists could guide the development of more refined, agonist-specific treatments that optimize SC-mediated regeneration in peripheral nerve injuries.

Another key aspect of this study was demonstrating the differential effects of PUR and SAG on neurotrophic factor secretion. The significant increase in BDNF and NGF levels in SCs treated with PUR suggests enhanced expression of factors critical to neuronal growth and survival via SC-mediated mechanisms. BDNF is known to be a critical factor that enhances peripheral nerve regeneration by promoting neuron survival²⁵, axonal myelination^{24, 26} and is known to be upregulated after injury^{24-26, 29}. NGF facilitates nerve regeneration by enhancing neuron survival²⁹ and axonal elongation^{30, 31} as well as enhancing SC survival¹³ and repair functions such as myelin debris clearance³².Despite both acting as agonists by targeting Smoothened on the cell membrane to activate SHH signaling, they produced distinct biological effects. This suggests that the two agonists might engage different downstream signaling pathways or interact with cellular components in unique ways, leading to different outcomes in neurotrophic factor secretion. Alternatively, the differential effects may be related to other factors such as the strength, duration, or specificity of SHH pathway activation by PUR compared to SAG. Other factors that were specifically and highly increased with PUR treatment alone include IGFBP-5, known to more specifically mediate axon-SC interactions, particularly in degenerating axons³³ in addition to homologs with functions implicated in autocrine growth and survival³⁴. Similarly, VEGF-AA was noted to be specifically elevated with PUR treatment only and is known to play a role in stimulating SC infiltration and migration^{35, 36}, neovascularization of injured tissues³⁵, and crucially enhancing axonal outgrowth^{37, 38}.On the other hand, our experiments demonstrated a decrease in the levels of GDNF and GMCSF released from SCs after treatment with either PUR or SAG. GDNF has been well-characterized as a factor involved in the survival of peripheral nerve neurons and functional recovery after nerve injury^{29, 39-41}. Rat sciatic nerves with constitutively overexpressed levels of c-Jun resulted have also been shown to have massively higher levels of GDNF²³. However, very high levels of c-Jun and GDNF resulted in hypomyelination of injured nerves, suggesting that levels of GDNF have a mixed role in regeneration. Furthermore, it has been demonstrated that GDNF in injured sciatic nerve negatively affects SC maturation and myelination at very high levels whereas it allows for sprouting of neurons at lower levels⁴². This suggests that there is a fine balance in GDNF levels for optimal therapeutic outcomes. GMCSF has also been implicated in peripheral nerve regeneration following injury by stimulating macrophage infiltration and early axonal; however, it did not by itself improve myelin clearance or functional outcomes.⁴³ Although it is possible that GMCSF may also function in a similar dose-dependent response in regenerating nerve, further studies are warranted to better characterize its function in this context.

Functional changes in SC migration may also suggest another dimension to the role of SHH in neural repair. Enhanced migration following SHH agonist treatment suggests that SHH signaling might also support the innate ability of SCs to reach injury sites and establish proximity with damaged axons—a crucial feature for effective repair. In addition, the migration of SCs is also necessary to facilitate processes such as clearing of debris, guidance of axons, and the formation of the bands of Büngner¹³. This characteristic of SHH-activated SCs is indispensable to create the environment conductive to axonal growth after injury. By enabling SCs to move more effectively to areas of damage, SHH agonists may help overcome one of the key barriers to nerve repair: the timely recruitment of repair SCs. This effect would be particularly relevant in longer gaps or chronic denervation injuries, where distance and senescence of SCs limit regeneration. Overall, these combined findings suggest that SHH signaling activates both the repair SC phenotype as well as the dynamic capabilities of SCs to respond to nerve injury.

Importantly, while conditioned media from SHH-treated SCs was enriched with neurotrophic factors such as BDNF and NGF, it was insufficient to stimulate significant neurite outgrowth in isolation. Whereas SC conditioned media has been shown to stimulate the regeneration of central nervous system neurons^{44, 45}, we did not observe this in our spinal neuron model. This finding points toward a layered mechanism of SHH-driven SC support in peripheral nerve, one that likely combines both soluble, secreted factors and the need for direct SC-neuron contact to drive axonal elongation within a peripheral nerve repair context. The dual contributions of these mechanisms reflect the complexity of the microenvironment in the context of injury and regeneration. In exploring the co-culture of SCs with neurons, it becomes clear that physical interactions between SHH-activated SCs and neurons are essential to promote neurite elongation. These results suggest that contact-mediated signals may be indispensable in creating the microenvironment conducive to regeneration. There are well-described axon-SC interactions that are mediated by contact-dependent mechanisms^{46, 47}. The presence of SHH-treated SCs in the neuronal environment enhanced neurite growth, suggesting a potential role for this type of SC-neuron interactions in repair mediated by mechanisms beyond secreted factors alone. These interactions may involve contact-mediated signals such as cell surface receptors, adhesion molecules, and extracellular matrix components that cannot be replicated by soluble factors alone.

In comparison to other approaches, our study underscores the advantage of treating SCs with SHH agonists to stimulate a comprehensive repair phenotype. Unlike methods that rely solely on neurotrophic factor supplementation, SHH agonists simultaneously regulated gene expression, morphology, migration, and secretion of growth factors, thus comprising a more integrated repair process. A few limitations of this study should be considered: Firstly, the in vitro environment does not fully capture the nerve regenerative environment, which involves multiple cell types (e.g. macrophages), as well as ECM interactions and other systemic factors. Additionally, the long-term effects of SHH agonist treatment were not investigated, and whether prolonged treatment may impact remyelination remains unknown. Finally, SHH pathway activation was not assessed using canonical downstream targets following administration of PUR or SAG, which limits their validation within this study at the concentrations tested.

Future studies could build upon these findings to characterize the precise molecular pathways that determine these SHH-mediated effects, particularly those involved in neurotrophic factor secretion and the paracrine-mediated neurite elongation from PUR-treated SCs. Pursuing the specific mechanisms that govern these observed effects would lead to greater precision to develop strategies to produce repair-specific outcomes. Further work should also prioritize identifying and characterizing the molecular pathways activated by SHH agonists to avoid unintended effects while optimizing their therapeutic outcomes. Understanding the chronic effects of SHH agonist treatment on cell behavior will be necessary to develop nerve regenerative therapies that they are scalable for clinical application. Ultimately, comprehensive research studies will be critical to the development of in vivo therapies for peripheral nerve healing, ensuring more targeted and effective approaches to promote tissue regeneration.

Conclusion

In summary, SHH signaling emerges as a powerful modulator of SC repair capabilities, with effects that span gene expression, morphological changes, migratory enhancement, neurotrophic factor secretion, and pro-regenerative direct neuron interactions. This study reveals the complex ways in which SHH signaling promotes SCs to foster a regenerative microenvironment, pointing to the therapeutic potential of targeting SHH signaling in neural repair strategies. By deepening our understanding of SHH-driven SC responses, we open pathways to additional research focused on translating these findings to therapeutic contexts to treat peripheral nerve injuries, where enhancing SC-driven repair could hold significant promise. Future studies should focus on elucidating the specific molecular and structural interactions underlying the contact-dependent mechanisms between SHH-activated SCs and neurons, further refining therapeutic strategies to optimize SC-driven nerve repair.