The Multi-targeted Effect of Fascaplysin on the Proliferation and Dedifferentiation of Schwann Cells Inhibits Peripheral Nerve Degeneration by Blocking CDK4/6 and Androgen Receptor Authors: Hyung-Joo Chung, Ja-Eun Kim, Youngbuhm Huh, Jin San Lee, So-Woon Kim, Kiyong Na, Jiwon Kim, Seung Hyeun Lee, Hiroyuki Konishi, Seung Geun Yeo, Dong Keon Yon, Dokyoung Kim, Junyang Jung, Na Young Jeong Published: 2024 ABSTRACT AbstractPeripheral neurodegenerative diseases induced by irreversible peripheral nerve degeneration (PND), such as diabetic peripheral neuropathy, have a high prevalence worldwide and reduce the quality of life. However, there is no agent effective against the irreversible PND. After peripheral nerve injury, Schwann cells play an important role in regulating PND. However, because PND involves multiple biochemical events in Schwann cells, a one-drug-single-target therapeutic strategy is not feasible for PND. Here, we suggested that fascaplysin (Fas), a compound with multiple targets (CDK4/6), could overcome these problems. Fas exerted a significant inhibitory effect on axonal degradation, demyelination, and Schwann cell proliferation and dedifferentiation during in vitro and ex vivo PND. To discover the most likely novel target for PND, a chemo-bioinformatics analysis predicted the other on-targets of Fas and identified androgen receptor (AR) which were involved in Schwann cell differentiation and proliferation. AR interacted with Fas, and nuclear import of the AR/Fas complex was inhibited in Schwann cells, altering the expression patterns of transcription factors during PND. Therefore, Fas may have therapeutic potential for irreversible peripheral neurodegenerative diseases.Keywords: Fascaplysin, Schwann cells, Peripheral nerve degeneration, CDK4/6, Androgen receptor, Multi-target drug INTRODUCTION The aging of the population has led to an increase in the incidence of various incurable diseases, reducing quality of life. These include peripheral neurodegenerative diseases, such as diabetic peripheral neuropathy. Because the pathogenesis of peripheral neurodegenerative diseases depends on a set of biochemical events and bioreceptors, the one-drug-single-target approach to drug discovery for multifactorial diseases, such as neurodegenerative diseases and tumors, is of little utility Thus, a trial for single multi-targeted drug discovery is especially timely and meaningful. In peripheral neurodegenerative diseases, damage to peripheral nerves, such as gene mutation or hyperglycemia, leads to irreversible peripheral nerve degeneration (PND) and disturbed nerve regeneration [1]. During PND, Schwann cells, peripheral neuroglial cells, morphologically and biochemically mediate demyelination, axonal degradation, and remyelination of peripheral nerves, curing which they exhibit dedifferentiation and proliferation [2]. These changes are not caused by malfunction or non-function of a single protein or a signaling pathway, but multiple biochemical events. Hydrogen sulfide (H2S), a gaseous neurotransmitter, pathway is one of the molecular events in Schwann cells during PND [3-5]. Cystathionine γ-lyase (CSE) mediates H2S production in Schwann cells [5]. Controlling H2S pathways in Schwann cells could have therapeutic potential for peripheral neurodegenerative diseases [3-5]. However, because PND involves multiple pathways, including mitogen-activated protein kinase (MAPK) [6], notch [7], phosphoinositide 3-kinases (PI3K) [8], and H2S pathways, multiple targeted drugs are likely to be necessary to delay or stop irreversible PND compared with one target drug for H2S pathways. Fascaplysin (Fas), a fused benzoyl-linked β-carboline alkaloid derived from the marine sponge Fascaplysinopsis berguist, is a pentacyclic quaternary ammonium salt [9]. Fas inhibits cyclin-dependent kinases (CDK) 4/cyclin D (CCND) 1 complex and the CDK6/CCND2 complex due to CDKs’ different regioselectivity; therefore, Fas is a multi-targeted compound [10]. In addition, Fas exhibits antitumor, antibacterial, and antimalarial activities and exerts an antineurodegenerative effect in Alzheimer diseases [11-14]. Therefore, the pharmacological activities of Fas could be considered multi-targeted effects on PNDs. Here, we evaluated the effect of Fas on PND using in vitro and ex vivo models and on the degenerative characteristics of Schwann cells. We also predicted a novel target of Fas in dedifferentiated Schwann cells via chemo-bioinformatics analysis and a genetically modified zebrafish model. We aimed to clarify the polypharmacological mechanism of Fas in Schwann cells against PND using integrated in silico, in vitro, and ex vivo approaches. The results indicate that the potent anti-neurodegenerative activity of Fas in peripheral nerves has therapeutic potential for peripheral neurodegenerative diseases. MATERIALS AND METHODS Compounds Arcyiaflavin A (Arc A, #2457) and SU9516 (SU, [3Z]-3-[1H-imidazol-5-ylmethylidene]-5-methoxy-1H-indol-2-one, #2907) were purchased from Tpcris (Bristol, UK). Brusatol (Bru, #FB30016) was purchased from Carbosynth (Compton, UK). Trans-4-[6-(Ethylamino)-2-[1-(phenylmethyl)-1H-indol-5-yl]amino]-4-pyrimidinyl]amino]-cyclohexanol (CINK4, #SML1352), purvalanol B (Pur B, #P5234), PD0332991 (PD, #PZ0199) and trigonelline (Tri, #T5509) were purchased from Sigma-Aldrich (St. Louise, MO, USA). Fascaplysin (Fas, #341251), NSC625987 (NSC, 1,4-dimethoxy-10H-acridine-9-thione, #219477), BIBN4096 (BIBN, #4561) and purvalanol A (Pur A, #540500) were purchased from Calbiochem (San Diego, CA, USA). BMS927711 (BMS, #HY-15498) and MK0974 (MK, #HY-32709) were purchased from MedChemExpress (Monmouth Junction, NJ, USA). The stock solution was reconstituted according to manufacturer’s instruction. Animals All the procedures in the mice experiment were tried to minimize the suffering and the number of animals according to the guideline of Kyung Hee University Committee on Animal Research with approval (Protocol number #KHSASP-21-463). 5-week-old C57BL/5 male mice were purchased from OrientbioTM (Seongnam, South Korea). All experimental mice were in the environment with controlled humidity (50%) and temperature (23°C±1°C) on a 12-hour light/dark cycle and provided with free access to feeds and water. For in vivo experiments, the nerve axotomy was performed as previously described [5]. Briefly, under sterile condition, sciatic nerves were axotomized after induction of anesthesia by intraperitoneal injection of pentobarbital (35 mg/kg) and then, the distal stumps of the sciatic nerves were treated with a 10-mm blind PVC tube packed with gel foam presoaked in Fas for 3 days. At 3 days after axotomy, mice were sacrificed by CO2 asphyxiation and then, the distal part of the nerves was harvested. Control was intact sciatic nerves. Sciatic nerve explant culture The sciatic nerve explant culture was performed using mice according to a previously described method [5]. Briefly, sciatic nerves were isolated from the 5-week-old C57BL/6 male mice with a fine iris scissor (Fine Science Tools, Foster City, CA, USA) under sterile surgical conditions. The connective tissues surrounding the nerve fibers were removed under a stereomicroscope in the phosphate-buffered saline (PBS). Each nerve was then cut into 3 or 4 pieces in 2 to 3 mm length with fine iris scissors. The nerve explants were transferred to a culture medium (Dulbecco’s modified eagle’s medium [DMEM], #SH30243.01, HyClone, Logan, UT, USA) added with 10% fetal bovine serum (FBS) and 1% penicillin/streptomycin] and incubated in a humidified incubator containing 5% CO2 at 37°C with or without compounds for three days. For immunohistochemistry, the sciatic nerves were fixed with 4% ice-cold paraformaldehyde (PFA) in PBS for 16 hours at 4°C. For cryo-section, the sciatic nerves were cut into16-μm thickness sections using a cryostat (CM1520, LEICA, Wetzlar, Germany) transversely and longitudinally. For teased-nerve preparation, we removed connective sheathes from the nerve and were carefully teased apart from nerve fascicles upon gelatin-coated slides. Immunohistochemistry The teased nerve fibers on slides were post-fixed in 4% PFA and washed with PBS. The slides were permeabilized in PBS-containing 0.3% Triton X-100 (PBS-T, Biosesang, Seongnam, South Korea). Sample slides were blocked in 5% bovine serum albumin (BSA) containing PBS-T. Primary antibodies were anti-neurofilament (NF; #841001, Biolegend, San Diego, CA, USA), anti-myelin basic protein (MBP; #ab980, Millipore, Burlington, MA, USA), anti-S100-β (#S2532, Sigma-Aldrich), anti-lysosomal-associated membrane protein 1 (LAMP1; #sc-19992, Santa Cruz Biotechnology, Dallas, TX, USA), anti-nerve growth factor receptor p75 (p75; #07-476, Millipore), anti-androgen receptor (AR; #ab133273, Abcam, Cambridge, UK) and anti-Ki67 (#ab16667, Abcam) were incubated at 4°C for overnight. Secondary antibodies were obtained from Invitrogen (Alexa fluor-488 donkey anti-mouse [#A32766], alexa fluor-488 donkey anti-rabbit [#A32790], alexa fluor-594 donkey anti-mouse [#A10037], and alexa fluor-594 donkey anti-rabbit [#A10042], Waltham, Massachusetts, US) and applied under light-protected environment for 2 hours at room temperature (RT). A laser scanning confocal microscope (LSM 700, Carl Zeiss, Jena, Germany) was used to detect immunofluorescence on the slides. The quantification of the relative intensity was carried out by counting the relative number of the pixels showing each intensity unit. Morphometric indices To verify the degree of peripheral nerve degeneration morphologically, stripe index was selected as a broad indicator by counting the number of transverse stripes in a nerve of 2 mm length. The number of ovoid-shaped structures as an ovoid index, which represents the myelin fragmentation, was measured from a 200 µm length of a nerve fiber. Cell culture and transfection SW10 cells, a mouse neuronal Schwann cell line, were obtained from the American Type Culture Collection (ATCC, #CRL-2766TM, Manassas, VA, USA) and maintained at 33°C in DMEM supplemented with 10% FBS. LNCaP clone FGC cells, a human prostate cancer cell line, were obtained from the Korean Cell Line Bank (KCLB, #21740, Seoul, South Korea) and maintained at 37°C in RPMI-1640 (#SH30027.01, HyClone) supplemented with 10% FBS. All cell lines were maintained with 5% CO2 at designated temperatures. Cells were transfected (Lipofectamine 2000, Invitrogen, Carlsbad, CA, USA) with siRNAs (20 nM) (Control [Sense: GAUAGCAAUGACGAAUGCGUA, Antisense: UACGCAUUCGUCAUUGCUAUC], or Ar [Sense: GACUCAGCUGCCCCAUCCA, Antisense: UGGAUGGGGCAGCUGAGUC] siRNA) according to the manufacturer’s protocol. Transfected cells were generally analyzed 48 h later. Cell viability assay Cell viability of the SW10 cell line was determined using a tetrazolium compound, MTS (3-[4,5-Dimethylthiazol-2-yl]-5-[3-carboxymethoxyphenyl]-2-[4-sulfophenyl]-2H-tetrazolium) purchased from Promega (CellTiter 96® AQueous One Solution Cell Proliferation Assay, #G3582, Madison, WI, USA) described previously [15]. Briefly, cells were plated on 96-well, flat-bottom microplates (#3628, Corning Life Sciences, Corning, NY, US), at a density of 5.0 × 103 cells per well in 100 µL DMEM supplemented with 5% FBS mixed with Fas (10 nM to 100 µM) and cultured at humidified incubator (5% CO2, 33°C). Cell viability was assessed after Fas treatment for 48 hours using 2 mg/ml MTS solution dissolved in dimethyl sulfoxide (DMSO). The cells were incubated with 20 µL MTS solution for two hours at 33°C. The absorbance was measured at 490 nm using MULTISKAN FC microplate reader (Thermo Fisher Scientific, Waltham, MA, USA). Percentage cell viability was calculated as Cell viability (%)=Atreated-AbackgroundAuntreated-Abackground×100. Normalized growth rate inhibition (GI) was assessed using A(untreated) at the time of treatment. The half-maximal inhibitory concentration (IC50), half-maximal effective concentration (EC50), the concentration for 50% of maximal inhibition of cell proliferation (GI50), and 50% cytotoxic concentration (CC50) were calculated by linear regression analysis as previously described [16]. Three independent replicates of the experiment were performed. The data were analyzed and plotted with Graph Pad Prism software version 9.0 (GraphPad Software, Inc., San Diego, CA, USA). Western blot analysis Western blot was performed as previously described [5]. Primary antibodies, anti-cyclin dependent kinase 4 (CDK4, #sc-56277, Santa Cruz Biotechnology), anti-cyclin D1 (CCND1, #ma5-1451, Invitrogen), anti-NF (#n2787, Sigma-Aldrich), anti- p75 (#07-476, Millipore), anti-phospho-c-Jun (#sc-822, Santa Cruz Biotechnology), anti-phospho-c-Fos (Cat#5348S, Cell signaling Technology, Danvers, MA, USA) and anti-AR (#ab133273, Abcam) were applied to the blot for 1 hour at RT. Horseradish peroxidase-conjugated secondary antibody was applied for 1 hour at RT. Specific signals were visualized using enhanced chemiluminescence (ECL SelectTM Western Blotting Detection Reagent, Cat#RPN2235, AmershamTM, Amersham, UK) as per the manufacturer’s instruction, and β-actin (#a5441, Sigma-Aldrich) was used as a loading control. Real-time quantitative polymerase chain reaction (RT-qPCR) Total RNA was isolated from SW10 cell lysates or sciatic nerve explants with or without Fas treatment using TRizol® as per the manufacturer’s instructions (#15596-026, Invitrogen), and complementary DNA (cDNA) was reverse-transcribed using a SuperScriptTM III reverse transcriptase (#18080044, Invitrogen). RT-qPCR was performed with a thermal cycler (TP700, Takara, Kusatsu, Japan) with 45 cycles, using 50 ng cDNA, 200 pmole/µL gene-specific primers (Supporting Table S1), and TB green premix Ex Taq (#RR420A, Takara). Glyceraldehyde 3-phosphate dehydrogenase (Gapdh) was used as a reference, and relative quantification was calculated using the double-delta Ct (∆∆Ct) method [17]. RT-qPCR was run at least triplicate for each assay. A detailed primer list was shown in Supplemental Table 1. RNA-seq analysis SW10 cells were grown and cultured in humidified 5% CO2 incubator at 33°C. Fas (10 µM) was treated in the cultures for 18 hours prior to harvesting. Three independent biological replicates were prepared for RNA isolation and RNA sequencing. Next generation RNA-seq analysis was conducted by Macrogen, Inc., (Seoul, South Korea). In brief, total RNAs were isolated separately using the TRIzol® reagent. The RNA concentration was calculated by Quant-IT RiboGreen (#R11490, Invitrogen). To assess the integrity of the total RNA, samples are run on the RNA ScreenTape System (2200 TapeStation system, Agilent, Santa Clara, CA, USA). Only high-quality RNA preparations with RNA integrity number (RIN) greater than 7.0, were used for RNA library construction. The libraries were independently constructed with 1 µg of total RNA with Illumina TruSeq Stranded mRNA Sample Prep Kit (#RS-122-2101, Illumina, San Diego, CA, USA,). NovaSeq 6000 platform (Illumina) was used for RNA-seq. Transcript-level quantification of RNA-seq data was carried out by Robinson and Oshlack [18] using Ensemble release GRCm38 (mm10) as the reference. Transcript-level quantification was aggregated to gene expression level using the exact test R package. False discovery rate (FDR) was controlled by adjusting p-value using Benjamini-Hochberg algorithm. Protein and ligand structure preparation Prior to molecular docking, the 3-dimensional (3D) structure of the protein was collected from Protein Data Bank [19] and AlphaFold Protein Structure Database [20]. The ligand was separated from the protein. The polar hydrogen atoms and Kollman charges were added to the protein, and the water molecules were eliminated. Discovery Studio Visualizer (version 21.1.0, BIOVIA, San Diego, CA, USA) minimized the final prepared file and saved it in PDB format for further analysis [21]. The 3D structure of Fas was downloaded from PubChem database (PubChem CID #73293) and changed into chemical file formats for virtual screening analysis using Open Babel software (http://openbabel.org). Molecular docking Molecular docking analysis was performed with AutoDock vina 4.2.1 connected with a PyRx working interface using both prepared target and ligand libraries as the previous method [21]. The target protein structure and Fas were used as a receptor and a ligand, respectively. Targets and Fas were represented in a grid space of a universal force field. The result of each docked molecule was determined according to the docking energy of the receptor-ligand complex. The top conformations of the structures with target and Fas complexes were obtained in PDBQT format. Discovery Studio Visualizer was used to determine and represent the target-ligand interaction sites. Drug-target interaction networks The search tool for retrieval of interacting genes (STRING) database [22], which integrates the known and predicted protein interactions or the drug and target protein interactions, can be applied to predict the functional interactions of molecules. The network was defined as a graph, where nodes and edges represent molecules and their interactions, respectively. STRING version 11.5 server for the network between the partners and the interactions were analyzed. Gene ontology (GO) analysis Database for Annotation, Visualization and Integrated Discovery (DAVID) bioinformatics resources (2021) were utilized to categorize functional pathways [23]. Enrichment score of each biological process term was tested with statistical significance (p-value <0.05). Inter-relationship of the output GO terms were visualized in semantic space as a scatter plot using Reduce and visualize gene ontology (REVIGO). Drug affinity responsive target stability (DARTS) assay DARTS was performed as a previous report [24]. Briefly, LNCaP cells were lysed with cold M-PER lysis buffer (Cat#78501, Thermo Fisher Scientific, Waltham, MA, USA) and centrifuge at 18,000×g for 10 min at 4°C. Lysates were diluted to the same final volume in Tris-NaCl-CaCl2 (TNC) buffer [50 mM NaCl, 10 mM CaCl2 and 50 mM Tris-HCl (pH 8.0)]. Fas (100 µM) or the same volume of DMSO were added and incubated for 30 min at RT. Pronase (Cat#10165921001, Roche, Basel, Switzerland) solution was diluted serially by mixing with 1× TNC buffer to create 1:500, 1:1,000, and 1:5,000 (ratio of pronase vs protein amount) pronase stock aliquots. Pronase was added into both DMSO and Fas-treated cell lysates and incubated for 30 minutes at RT. Digestion was stopped by adding a protease inhibitor cocktail (HaltTM Protease Inhibitor Cocktail, #78440, Thermo Fisher Scientific) and the samples were incubated on ice for 10 minutes. 6× SDS-PAGE loading buffer was added to the samples and then the samples were boiled for 10 minutes immediately prior to the western blot analysis. Statistical analysis Statistical significance was determined by the two-tailed Student t-test, using GraphPad Prism 9 software (GraphPad Software, Inc.), and a p-value <0.05 was considered statistically significant. All data were expressed as mean±standard deviation (SD). The sample size (n=3) is recognized as a minimum number to detect a pre-specified effect size. RESULTS Inhibitory effect of Fas on PND in vitro and ex vivo CSE produces H2S in Schwann cells, and its downstream factors are involved in several cellular events [3, 5]. First, after screening inhibitors associated with H2S signaling pathway, we investigated their effects on other pathways, predicted their molecular targets and evaluated the interaction between the inhibitors and new targets. Nuclear factor erythroid 2-related factor 2 (NRF2) is a basic leucine zipper protein that regulates the synthesis of antioxidant proteins, and its expression is regulated by H2S [25]. To investigate whether the H2S/NRF2 pathway in Schwann cells is related to prevention of peripheral nerve damage, we evaluated the effects of brusatol (Bru) [26] and trigonelline (Tri) [27] on PND (Supplemental Table S2). Calcitonin gene-related peptide (CGRP) is a member of the calcitonin family of peptides, and H2S and nitric oxide (NO) induce CGRP expression [28]. In particular, α-CGRP functions as a neuropeptide in the nervous system [29]. We used BIBN4096 (BIBN), BMS927711 (BMS), and MK0974 (MK) [30, 31] to assess the relationship between H2S/CGRP and PND (Supplemental Table S2). CDKs regulate cell cycle progression by binding to cyclins [32, 33]. To investigate the effect of H2S/CDKs on PND, we used arcyiaflavin A (Arc A), Fas, NSC625987 (NSC), SU9516 (SU), PD0332991 (PD), and purvalanol (Pur) A/B [34, 35] (Supplemental Table S2). Before starting the drug screening, we evaluated H2S-pathway-related gene expression using an ex vivo sciatic nerve culture system to check its availability in Schwann cells (Fig. 1A) by semi-PCR analysis. Ccnd1, Calcrl, Hbegf, Hmox1, Cdkn1a, Il1a, Il6, Ccl2 and Fbp1 were all upregulated on day 3 of in vitro culture (3DIV) (Fig. 1B). Drug targets of H2S pathway inhibitors. (A) Illustration of mouse ex vivo sciatic nerve explant experiment. Degenerating sciatic nerves showed morphological changes in transverse stripes and myelin sheath. (B) Upregulation of H2S pathway-related gene expression in ex vivo sciatic nerves confirmed by semi-PCR analysis (left) and quantification of band intensity (right). Control (Con), intact nerves; 3DIV, 3 days in vitro. *p<0.05, **p<0.01. (C) Targets (pink circle) of each H2S pathway inhibitor were collected from Pubchem (biological tested target) and the network was prepared by Cytoscape (www.cytoscape.org). During PND, the transverse stripes (bands of Fontana) in sciatic nerves disappeared at 3DIV [5]. We performed a pharmacological activity assay using an ex vivo culture [5]. Bru, Tri, BIBN, BMS, and MK, which are associated with the H2S/NRF2 or H2S/CGRP signaling pathway, did not inhibit disappearance of the transverse stripes in the sciatic nerve explants at 3DIV (Fig. 2A). Among the compounds, only Fas inhibited disappearance of the transverse stripes in sciatic nerve explants at 3DIV (Fig. 1A). Next, we calculated the ovoid index to morphometrically identify myelin fragmentation, a major phenotype of PND. Among the compounds associated with the H2S/NRF2, H2S/CGRP, or H2S/CDK pathways, Fas inhibited ovoid formation caused by myelin fragmentation (Fig. 2B). Because Fas showed significant pharmacological activity at ≥10 μM (Fig. 2B), we used 10 μM Fas in subsequent experiments. Taken together, our results indicate that Fas modulates the H2S/CDK pathway during PND. Attenuation of peripheral nerve degeneration by Fas. (A) Sciatic nerve explants were cultured for 3DIV with or without H2S pathway-related inhibitors (detail information as shown in Methods, Compound section), and then the explants were visualized under a stereoscope (left panel). Size bar=600 μm. The quantitative graph (right panel) showed the number of transverse stripes within 2 mm of a sciatic nerve fiber under a stereoscope (stripe index). n=3 mice. (B) The effect of tested drugs on ovoid formation in ex vivo sciatic nerves for 3DIV (left panel). Ovoid structures were observed under differential interference contrast (DIC)-filtered microscope. The fibers at 3DIV showed the formation of ovoid-like myelin fragmentation. Size bar=50 μm. Ovoid index (right panel) means the number of ovoid-shaped structures (n=3 of mice using each group). (C) Target information of H2S-related inhibitors (n=3) using ChEMBL database. Heatmap displayed drug-target interactions. (D) In vitro anti-proliferation activity of Fas against SW10 Schwann cells after 48 h incubation. Experiments were done in triplicates. The chemical structure of Fas and drug-quantitative indices are shown in the right panel. *p<0.05, **p<0.01, and ***p<0.001. We visualized Fas-target interactions as a heatmap using the ChEMBL database [36]. Fas strongly interacted with CDK4/cyclin D1 and moderately with CDK5 (Fig. 2C, Supplemental Table 2). Targets of Fas collected from the Pubchem database also hint at the possibility as a multi-target drug (Fig. 1C). A MTS assay in SW10 Schwann cells yielded IC50 (0.71 μM), EC50 (0.47 μM), GI50 (0.26 μM), and CC50 (3.02 μM) values indicating that 0.26 μM Fas inhibited Schwann cell proliferation in vitro (Fig. 2D). Therefore, Fas prevented effectively PND in vitro and ex vivo. Pharmacological effects of Fas on peripheral neurodegenerative phenotypes In peripheral nerves, main structures are peripheral nerve axons and Schwann cells, the latter being core regulators of PND and the regeneration [37]. Therefore, we investigated the effect of Fas on the phenotypes of Schwann cells during PND. To investigate whether Fas inhibits axonal degradation and demyelination, we performed immunostaining with antibodies for NF (axonal marker) and MBP (myelin sheath marker) using ex vivo mouse sciatic nerve explant culture. NF immunostaining showed consecutive lines (green) in control nerve fibers whereas nerve fibers at 3DIV revealed an increased number of broken consecutive lines (Fig. 3A). The lines of Fas-treated nerve fibers were intact (Fig. 3A, D). MBP staining was observed as consecutive double lines whereas nerve fibers at 3DIV displayed cleaved lines and irregular margins of MBP staining (Fig. 3B), indicating Schwann cells demyelination. However, nerve fibers treated with Fas at 3DIV showed the same pattern as the control, consecutive lines and regular margins, during ex vivo culture (Fig. 3B, E, F). Fas regulating the degenerative phenotypes of Schwann cells. Sciatic nerve fibers were immunostained with antibodies for (A) neurofilament (NF; an axonal marker, green), (B) myelin basic protein (MBP; a myelin sheath marker, red), S100 (a marker for Schwann cells, green) and DAPI (blue), (C) transcriptional regulator, p-c-Jun (red, arrows indicate p-c-Jun/DAPI double-positive cells), (H) dedifferentiated Schwann cell marker including lysosomal-associated membrane protein 1 (LAMP1, green) and (I) nerve growth factor receptor p75 (p75, red), and (J) a marker for cell proliferation, Ki67 (green, arrows indicate Ki67/DAPI double-positive cells). All the experiments used ex vivo sciatic nerves at 3DIV in the absence or presence of Fas and control (no injury). (D~G, K~M) Quantitative indices and intensities of control, 3DIV, and 3DIV/Fas treatment were shown on the lower panel of each image. NF (D); MBP (E, F); p-c-Jun (G); LAMP1 (K); p75 (I); Ki67 (M). Size bar=50 µm. n=3 of mice using each group. ***p<0.001. (N) Western blot analysis showed protein expression of CDK4, CCND1, NF, p75, c-Jun and p-c-Jun in ex vivo sciatic nerves (n=3) in the presence or absence of Fas. Transcriptional regulation controls the Schwann-cell phenotype under normal and degenerating conditions [38]. To assess whether Fas affects transcriptional regulation during PND, we immunostained samples for p-c-Jun as a negative transcriptional regulator of myelin maintenance. P-c-Jun was not expressed under normal conditions but was highly expressed in the nucleus in untreated Schwann cells at 3DIV (Fig. 3C). Fas-treated nerve fibers showed no change in p-c-Jun nuclear expression in Schwann cells at 3DIV compared to untreated fibers (Fig. 3C, G). Schwann cells during PND have similar requirements to immature Schwann cells during peripheral nerve development, known as Schwann cell dedifferentiation. To determine whether Fas regulates Schwann cell dedifferentiation during PND, we investigated the expression of the dedifferentiation markers LAMP1 and p75 NTR in Schwann cells [5, 39]. Immunostaining showed that LAMP1 and p75 NTR were inhibited by Fas at 3DIV (Fig. 3H, I, K, L). To identify the effect of Fas on Schwann cell proliferation during PND, we immunostained ex vivo sciatic nerve fibers with an anti-Ki67. Ki67 was not immunostained in the control fibers. Degenerated nerve fibers were double-positive for DAPI/Ki67 at 3DIV, whereas Fas-treated nerve fibers did not express Ki67 at 3DIV (Fig. 3J, M). Western blot for NF, p75 NTR, and p-c-Jun showed that Fas inhibited axonal degradation, dedifferentiation, and negative regulation and the CDK4/cyclin D1 complex as a main target of Fas (Fig. 3N). Therefore, Fas prevents development of the morphological and biochemical phenotypes of degenerating Schwann cells during PND. Pharmacological effect of Fas on the proliferation and differentiation of Schwann cells To investigate the mechanism of the anti-neurodegenerative effect of Fas, RNA sequencing (RNA-seq) of SW10 Schwann cells [40] was performed (Supplemental Table 3). Differentially expressed genes (DEGs) analysis revealed 1,178 upregulated and 2,018 downregulated genes between the control and Fas-treated groups (Supplemental Table 3). Our GO analyses showed 5 categorized groups, regulation of biological process (1 and 2), cell communication (3), cellular component organization (4), and developmental process (5) (Fig. 4A, Supplemental Table 3). Among the top 4 biological functions regulated by Fas in the group 1 and 2, we investigated the gene expression of cell differentiation and cell proliferation which should be positively and negatively regulated, respectively, during PND (Fig. 4B, Supplemental Table 3). Transcriptional alteration of Fas-treated dedifferentiated Schwann cells. (A) Gene ontology (GO) analysis was performed on DEGs in Fas-treated SW10 cells compared with control, using a three-fold (log2 value) upregulated cut-off. GO scatter plot was shown using reduce and visualize gene ontology (REVIGO). GO terms are represented by the bubbles and the color are indicated as the p-values. Horizontal and vertical axes of the plot indicated semantic space. Similar GO terms are shown closer in the plot. Bubbles are clustered into 5 groups regarding to semantic similarities. (B) Representative GO terms of group 1 and 2 were shown and they were selected among the phenotypic GO terms in Schwann cells. (C) Volcano plot of the genes associated with cell differentiation and cell proliferation. Blue and pink colors indicated cell differentiation and cell proliferation, respectively. The genes related to both phenotypes are shown in black color. (D) Venn diagram was drawn with transcription factors involved in both cell dedifferentiation and proliferation from our RNA-seq data. (E) Validation of downregulated transcription factors including Ar, Arid2, and Pbrm1 are performed by RT-qPCR. ***p<0.001. Transcription factors mediate the myelination, dedifferentiation, proliferation, and demyelination of Schwann cells [37]. To identify key transcription factors associated with both cell differentiation and cell proliferation regulated by Fas, we sorted out 320 transcription factor genes from our RNA-seq data and then found 3 genes, Ar, Arid2, and Pbrm1, which are related to both cell dedifferentiation and cell proliferation in Schwann cells (Fig. 4C, D). We verified the expression levels of several transcription factors by RT-qPCR. Six genes (Ar, Arid2, Gli3, Prbm1, Runx1, and Zeb1) were downregulated in Schwann cells (Fig. 4E). The RNA-seq results support the notion that Fas regulates the expression of transcription factors to modulate the differentiation and dedifferentiation of Schwann cells. Identification of novel targets of Fas by using chemo-bioinformatics Various biochemical events and several bioreceptors in Schwann cells initiate and progress PND [2]. To identify novel targets of Fas, we hypothesized that the targets could be involved in the proliferation and differentiation of Schwann cells (Fig. 4). We first calculated a free energy of binding between Fas and proteins affecting the proliferation and differentiation of Schwann cells (Supplemental Table S3) [22]. Androgen receptor (AR), Polybromo 1 (PBRM1), Adenomatous polyposis coli protein (APC), and Wnt family member 9A (WNT9A) had free energies of binding of <-8 kcal/mol, and AR was the lowest binding affinity (-10.5 kcal/mol) with Fas (Fig. 5A). In the binding pocket of AR, Fas interacted with Met745, Met787, Leu873, Leu707, and Leu704 (Fig. 5B). We also collected two diseases, non-insulin-dependent diabetes mellitus (DM) and neuropathy (NP) to screen genes involved in diabetic peripheral neuropathies as a common irreversible PND. Disease-related genes (DM, n=3,134; NP, n=523) were obtained from the DisGeNET [41]; the 53 genes including Ar are presented in a Venn diagram (Fig. 5C). These results indicated that AR could be a potential on-target of Fas in Schwann cells during PND. Selecting novel targets of Fas using chemo-bioinformatic analysis. (A) Binding free energy between Fas and 109 proteins affecting cell proliferation and cell differentiation was calculated using AutoDock VINA tools. (B) 3-dimensional view of the best target-ligand conformation of interaction between Fas and androgen receptor (AR, PBD ID: 2PIV) were analyzed using Discovery Studio Visualizer 4.5. Dot lines represented H-bonds interactions. (C) Venn diagram showing the comparison of 3,134 genes associated with non-insulin-dependent diabetes mellitus (DM), 523 genes associated with neuropathy (NP), and 3,196 DEGs from our RNA-seq analysis. 53 Shared proteins are represented in the black square boxes. Proteins for DM (UMSL CUI#C0011860) and NP (UMLS CUI#C0442874) were collected from DisGeNET (v7.0) database. (D) Venn diagram showed the comparison between 536 genes from ChIP-seq analysis and 342 genes from RNA-seq results (Fas-treated SW10 cells). 137 genes were overlapped between two datasets. All genes are encoding transcription factors. (E, F) Validation of downregulated (E) and upregulated (F) transcription factor genes using RT-qPCR. ***p<0.001. Data represent mean±SD. Next, we investigated the expression of transcription factors regulated by AR in Schwann cells. A total of 137 transcription factors were overlapped between our RNA-seq data and 536 genes in a previous ChIP-seq analysis of transcription factors modulated by AR (Fig. 5D) [42, 43]. Verification with the several upregulated or downregulated genes in RNA-seq showed similar expression patterns in SW10 cells by RT-qPCR (Fig. 5E, F). Inhibitory effect of Fas on nuclear transport of AR in Schwann cells AR as a type of nuclear receptor is activated by binding androgenic hormones in the cytoplasm and then translocated into the nucleus [44]. We first investigated AR expression pattern as an optimal target in Schwann cells by immunostaining using ex vivo mouse sciatic nerves. Ex vivo sciatic nerves showed that, in the control, AR was not found Schwann cells. By contrast, during PND, AR was detected in the nuclei of Schwann cells at 12 hours in vitro (12HIV or 0.5DIV) (Fig. 6A). Interestingly, AR was located in the cytoplasm of Schwann cells in ex vivo Fas-treated sciatic nerves (Fig. 6B). Thus, these results indicated pharmaco-anatomically that AR is an on-target of Fas and makes up a significant portion of the transcriptional regulation controlled by Fas in Schwann cells. Androgen receptor (AR) as a new on-target of Fas in Schwann cells. (A) Ex vivo sciatic nerve longitudinal sections were immunostained with antibodies for androgen receptor (AR; red), S100 (green), and DAPI (blue). Percentage of AR/DAPI double-positive profiles at control (Con), 12 hours in vitro (12HIV), 1 day in vitro (1DIV), and 3DIV with or without Fas treatment were shown on the right panel. Size bar=50 µm. In each group, 3 mice were used. (B) Ex vivo sciatic nerve cross sections at 3DIV with Fas treatment were immunostained with AR (red) and S100 (green). Higher magnification images (lower panel) showed the location of AR in the cytoplasm (a white arrow). Schematic illustration described the cytoplasmic retention of Fas/AR complex during PND (left panel). (C) DARTS assay using Fas 10 μM showed the interaction between Fas and AR. LNCaP cell lysates were incubated with Fas or vehicle (DMSO) for 30 minutes, followed by digestion with different dilutions of Pronase (1: 500, 1:1,000, and 1:5,000) for 30 minutes. Western blot analysis used antibodies against AR or β-actin (upper panel). AR level was normalized to β-actin and fold change relative to control was calculated by using the AR/β-actin ratio (lower panel). (D). Western analysis showed that AR-siRNA significantly decreased only AR expression, but not CDK4 or CCND1 in SW10 cells. (E). Validation of AR- or Fas-regulated transcription factor genes was processed by using RT-qPCR. E2f7 and E2f8 are AR-independent genes. *p<0.05, **p<0.01, and *** or ### p<0.001. (F) Protein lysates (30 μg) from sciatic nerves at 3 days after axotomy were analyzed by Western blot (n=4 mice). LAMP1, lysosomal-associated membrane protein 1; p75, neurotrophin receptor. Next, we performed DARTS analysis using LNCaP clone FGC cells because of their characteristics highly expressing AR [45], which is based on the notion that drug-target binding reduces its susceptibility to protease digestion due to stabilization of the complex [46]. Fas prevented the digestion of AR, meaning that Fas directly interacts with AR (Fig. 6C). Therefore, Fas bound to AR and the Fas/AR complex was retained in the cytoplasm of Schwann cells during PND. We also investigated the individual effect of AR on its transcriptional regulation in Schwann cells. The expression of CDK4/CCND1 as an original target of Fas was not changed by AR gene silencing (Fig. 6D). In the verification by using several transcription factors regulated by Fas (Fig. 5D), the Tead1 and Zeb1 genes regulated by AR were transcriptionally inhibited by both AR gene silencing and Fas, whereas E2f7 and E2f8 genes irrelevant to AR were unchanged by the silencing but Fas changed their expressions (Fig. 6E). Verification by using representatively dedifferentiation markers, LAMP1 and p75, also showed that Fas-dependent inhibition of PND also occurred in in vivo mice model as well as ex vivo model (Fig. 6F). Thus, these results indicated that AR is an on-target of Fas and its transcriptional regulation is also controlled by Fas in Schwann cells meaning that Fas has multi-targeted effects on both AR and CDK4/CCND1. DISCUSSION Fas is a kinase inhibitor with high selectivity for CDK4 and relatively less selectivity for CDK2, CDK5, and CDK6 [47, 48]. Because it is compatible with the CDK4/CCND1 complex, which is related to cell proliferation, Fas has antitumor and antibacterial activities [11,14]. However, in Schwann cells, targeting CDK4/CCND1 did not inhibit PND significantly. Because Schwann cell proliferation is a characteristic of peripheral nerve development, degeneration, and regeneration, much effort has focused on regulating CDK4/CCND1 expression in Schwann cells. However, although CDK4-deficient mice influenced Schwann cell proliferation during peripheral nerve development, Schwann cell differentiation was unaffected [49]. The CDK4/CCND1 complex is also not associated with Schwann cell mitotic responses in CCND1-deficient mice after nerve injury [50]. Therefore, targeting only CDK4/CCND1 in degenerating Schwann cells would not ameliorate PND. Thus, our finding that Fas inhibits PND leads to the existence of other on-targets of Fas in Schwann cells, in other words, Fas has more complicated multi-targeted properties in Schwann cells during PND. First, we selected H2S signaling pathway to cluster compounds with similar mechanisms and performed drug screening using ex vivo culture system as an effective drug screening platform for PND [5] (Fig. 2). Among them, Fas showed good in vitro and ex vivo efficacy (Fig. 2, 3) and reasonable pharmacological activity indices (Fig. 2). Drug screening for inhibiting PND is important to consider unusually GI50 value compared with other diseases [16]. During PND, Schwann cells should not die under all circumstances because Schwann cell proliferation is a physiological event, but not a pathological property such as cancer and a Schwann cell needs to be just protected against its irreversible dysfunction. Therefore, the GI50 value was used to evaluate the inhibitory activity of PND. In our ex vivo models, Fas effectively penetrated the blood-nerve barrier (BNB) in peripheral nerves (Fig. 2A). Fas may be structurally close to a lead-like compound for in vivo brain drug efficacy because Fas has physicochemical properties that affect drug penetration of the blood-brain barrier (BBB): lipophilicity (cLogP and cLogD at pH 7.4), hydrogen bonding donor (HBD) and hydrogen bonding acceptor (HBA), molecular weight (MW), and polar surface area (PSA) [51]; the MW, cLogP, cLogD (pH 7.4), HBD, HBA, and PSA of Fas were 271.3 g/mol, 3.8, 0.38, 3, 1, and 36.74 Å2, respectively. Thus, we can infer ex vivo ability to penetrate peripheral nerves from its physicochemical properties in brain although BNB is structurally different from BBB, and it could be the reason why Fas is the most effective drug in the ex vivo studies (Fig. 2A). Next, we confirmed the pharmacological action of Fas via bioinformatics analysis. GO enrichment analysis by using RNA-seq exhibited that Fas regulated positively cell differentiation and negatively cell proliferation (Fig. 4A, B). Schwann cell dedifferentiation and proliferation are major characteristics during PND. To delay or prevent PND, myelin structures in the cytoplasmic membrane of Schwann cells should be maintained. In other words, a differentiated form of Schwann cells should be maintained (Fig. 3B). The positive regulation of Schwann cell differentiation by Fas means to inhibit its induction for dedifferentiation during PND (Fig. 4A, B). Negative regulation of Schwann cell proliferation in GO terms also matches up to the inhibitory effect of Fas on the ex vivo Ki67 expression (Fig. 3J). In addition, molecular docking is computational modeling of the interaction between a small molecule and a protein target, and can predict the best fit of the ligand-protein complex [52]. Among the 109 proteins associated with the proliferation and differentiation of Schwann cells, AR, PBRM1, APC, and WNT9A showed the highest binding stability (-10.5 kcal/mol) with Fas, but only AR was shared by DM genes and NP genes (Fig. 5A~C). The effect of AR on PND has been little known in Schwann cells. Therefore, studying its pathophysiology and pharmacological mechanism as an expected target of Fas may be very attractive for drug discovery and development against peripheral neurodegenerative diseases. During PND, AR was located in the nucleus of Schwann cells at the early stage of PND, 12HIV but not 3DIV (Fig. 6A, B). The DARTS analysis also showed that AR was a potential target of Fas (Fig. 6C). Therefore, during PND, the formation of a complex with Fas inhibits nuclear transport or promotes dysfunction of AR because the complex is maintained in the cytoplasm but not degraded at 3DIV (Fig. 6B). It is possible that androgens from paracrine of Schwann cells at the early stage of response to nerve injury accelerate the nuclear transport of AR in Schwann cells [53]. Taken together, our chemo-bioinformatic findings revealed that Fas is involved in cell proliferation and differentiation, and interacts with multiple targets, including AR and CDK4/6. In conclusion, Fas potently inhibited PND. Because irreversible PND is a determinant of chronic peripheral neurodegenerative diseases, inhibition of PND may delay or prevent the development of the diseases. However, because PND is multifactorial, targeting a single factor will not achieve a therapeutic effect. Our findings showed that Fas is a ligand for multiple factors, including AR, CDK4, and CDK6, in Schwann cells during PND. Fas promotes cell cycle arrest at G0/G1 phase and transcriptionally regulates AR to prevent PND. Therefore, based on our encouraging results, Fas may provide a possibility as a promising lead-like compound in drug discovery for PNDs. Supplemental Materials ACKNOWLEDGEMENTS We thank Dr. Yoo Lim Chun, Mr. Jung Sig Cho and Dr. Ki-Hoon Park (Department of Anatomy and Neurobiology, Kyung Hee University, Seoul, Korea) for their valuable discussion. This work was supported by Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Science, ICT and Future Planning, No. 2021R1A2C1004184 (to H-.J.C), No. 2021R1A2C1004133 (to N.Y.J.) and No. 2023R1A2C1003763 (to J.J.). References Associated Data Supplementary Materials