The S100B/RAGE Pathway in Amyotrophic Lateral Sclerosis-induced Astrocyte Activation and Motor Neuron Death: a Potential Target for Therapy

Project location: ITALY, Rome
Project start date: September 2016 - Project end date: February 2018
Project number: 2016-033
Beneficiary: Università Cattolica del Sacro Cuore

Very short description of the project goals 
Amyotrophic Lateral Sclerosis (ALS) is a fatal  neurodegenerative disease affecting brain and spinal motor neurons; loss of these neurons induces death, with a median survival time from 20 to 48 months from symptom onset, essentially due to respiratory paralysis. Although pathophysiologic mechanisms contributing to the disease onset and development are essentially unknown, a crucial role for the astrocytic glial cell type in  the inflammatory processes sustaining the disease appears to be increasingly evident. The S100B protein is a molecule concentrated in astrocytes and, when released at high concentration, is regarded to play a pivotal  role in neuroinflammatory processes, displaying toxicity through its binding to the Receptor for Advanced Glycation Endproducts (RAGE). A series of preliminary published and unpublished evidences from our and other groups suggest that S100B may in fact be involved  in ALS onset and progression.This project is aimed at investigating a role for the S100B protein as mediator of astrocyte pro-inflammatory activation and consequent motor neuron toxicity in ALS, in order to establish the basis to individuate the protein as a therapeutic target for the disease. 
 
Detailed description of the activities implemented and of the methodologies applied  
Immunofluorescence microscopy. The localization of selected antigens (S100B, GFAP, RAGE, TP3, CNPase, ChAT)) was studied by double immunofluorescence analysis in spinal cord of diseased and control transcardiacally perfused (4% paraformaldehyde) animals in free floating. Immunofluorescence was analyzed by means of the confocal laser scanning microscope LSM 510 META, Zeiss equipped with three lasers: Argon/2, HeNe543, and HeNe633. The brightness and contrast of the digital images were adjusted using the LSM Image Browser software. 
Nissl staining and motor neuron count. Spinal cord sections were randomly selected and stained with 1% cresyl violet. The whole ventral horn of the spinal cord was photographed at ×10 magnification with Zeiss Axiophot microscope. Large neurons, with cell bodies ≥200 μm and a definable cytoplasm with a nucleus and nucleolus were then counted. 
Cell cultures, transfection and silencing. C6 rat astrocytoma cells were grown in Dulbecco's modified Eagle's medium supplemented with 10% fetal bovine serum at 37°C in an atmosphere of 5% CO2 in air. To perform transfection, C6 cells were plated at 70% confluency and, 24 h later, transfected with pCMV, pCMV-SOD1wt, or pCMV-SOD1G93A using Metafectene Pro reagent. Cells were harvested 24 h after transfection. For the S100B silencing, primary astrocytes were plated and transfected with 100 ng/ml of scramble siRNA or with 50 ng/ml each of a pair of S100B siRNAs using Metafectene Pro reagent. To obtain an improved silencing efficiency, both negative controls and si-s100B-treated cells were transfected twice, at 24 and 48 hours, and harvested at 72 hours post the first transfection. 
The human astrocytoma cell line U87 was stably transfected  with three lentiviral PLKO.1 vectors, two of which containing short hairpin RNA targeting S100B (VA and VB clone). Control cells were transduced with non-targeting PLKO.1 vector (scrambled). In order to shed light on the molecular mechanisms responsible for the proinfiammatory activity of S100B in astrocyties, the expression of genes specifically altered by S100B-directed shRNA in presence or absence of inflammatory stimuli, such as  after 24h treatment with 10µM of amyloid beta protein (Aβp), was evaluated by qPCR. 
Protein extraction, SDS-PAGE and Western blotting. Protein lysates from cervical and lumbar spinal cord segments were obtained according to the following procedure: tissues were crushed with a potter in RIPA buffer (PBS, 1% Nonidet P-40, 0.5% sodium deoxycholate, 0.1% SDS) containing a protease inhibitor cocktail . After an incubation of 30 minutes in ice, the lysates were centrifuged for 20 min at 14.000g at 4°C. Supernatants were collected and assayed for protein quantification with the Bradford detection kit .To isolate total-protein extracts from cellular cultures, cells were harvested in ice-cold RIPA buffer added with protease inhibitor cocktail. Lysates were kept on ice for 30 min and then centrifuged for 10 min at 14.000g at 4°C. Supernatants were collected and assayed for protein quantification with the BCA protein assay. Protein samples were separated by SDS-PAGE and transferred onto PVDF membranes. Membranes were blocked for 1 h in Tris-buffered saline solution with 0.1% Tween-20 (TBS-T) containing 5% nonfat dry milk and then incubated overnight at 4°C with indicated primary antibodies diluted in TBS-T containing 2% nonfat dry milk. After rinsing with TBS-T solution, membranes were incubated for 1 h with the appropriated peroxidase-conjugated secondary antibody diluted in TBS-T containing 1% nonfat dry milk, then washed and developed using the ECL chemiluminescence detection system . Densitometric analyses were performed using ImageJ software program. 
S100B analysis by ELISA kit. Culture medium from C6 cultures transfected with pCMV, pCMV-SOD1wt, or pCMV-SOD1G93Awas collected and measured for S100B content using S100B human ELISA kit . The HRP-generated signal was detected by a microplate reader at 450 nm, and the OD values were converted in S100B concentration using a standard curve. 
Semiquantitative PCR and Real Time qPCR. For quantitative Real-Time PCR and semiquantitative PCR, RNAs were isolated using TRIzol for tissue extraction and RNeasy kit for the cell cultures. RNAs were quantified and reverse-transcribed with random primers by GoScript Reverse Transcription System. qRT-PCR was performed with GoTaq qPCR Green Master Mix using adequate primers.. Semiquantitative PCR was performed with BioMix Red  
 
Results obtained (final)  
Using the recognized experimental model for ALS offered by rodents (rats or mice) genetically modified to carry the human-mutated SOD1G93A,   , both S100B and its receptor  RAGE where shown to be  progressively  upregulated in spinal cord astrocytes of diseased animals with a  timing pattern correlated to the level of neurodegeneration, also accompanying the typical features of activated astrocytes. Also the distribution of the receptor RAGE was dysregulated: during the progression of the disease: its localization shifted from neurons to astrocytes and its different molecular forms (isoforms) were differently expressed in different phases of the disease. Altogether, these results are consistent with the possibility that an autocrine loop of S100B occurs in astrocytes, being increasingly expressed/secreted by astrocytes, and affecting again astrocytes through the receptor RAGE, possibly depending on signals released by motor neurons progressively dying during the progression of the disease. Additional data were obtained, showing that transfection of mutated SOD1G93A  in an astrocyte cell line (C6 cells) resulted in overexpression and increased release of S100B, which fit this hypothesis. Finally, the inhibition of S100B expression, obtained by transfecting silencing mRNA targeting S100B  in astrocytes derived from diseased mutated SOD1G93A, reduced the expression of reactivity-linked/ proinflammatory genes known to be increased in ALS astrocytes (GFAP, TNF, CXCL10, CCL6). 
The role played by S100B in the induction of  proinflammatory features in astrocytes was further confirmed producing an astrocyte  line (astrocytoma U87 cell line) where the S100B synthesis is permanently inhibited by genetic ablation (S100B KO). The knockdown of S100B expression in this  cell line was associated with a downregulation of NF-Kb and iNOS. In addition, a different behavior  after proinflammatory stimulation (amyloid beta protein)in the expression of of inflammatory molecules ( NF-KB, iNOS, IL-6) between wild type and the S100B KO  cell line was observed, the expression of these molecules being significantly reduced in S100B KO cells. 
The above data pose S100B as a key player in tuning the inflammatory response of astrocytes and in managing the interplay between astrocytes and surrounding cells. Whether the aberrant dysregulation of S100B and its receptor RAGE are among the causes or the hallmarks of ALS has yet to be elucidated. Nevertheless, these  findings propose S100B-RAGE axis as a relevant contributor to the pathogenesis of the disease, and its blockade may be suggested as a rational target for therapeutic intervention in ALS. It is reasonable therefore to speculate that the inhibition of S100B in vivo may reduce the proinflammatory features of astrocytes and influence in this way the progression of the disease. 
In addition,  since converging evidence points out that S100B plays a pivotal role in different neural injuries, which appear to share some common features reasonably attributable to neuroinflammation regardless their origin -multiple sclerosis, Parkinson’s disease, Alzheimer’s disease, acute brain injury-(for review, Michetti et al, 2019) the protein might be candidated as a common therapeutic target for these diseases. An extension of the present studies to other neuroinflammatory/neurodegenerative diseases (e.g. multiple sclerosis) will be needed in order to verify this intriguing and promising possibility.     
The results obtained have been published in international scientific journals, where the support by the Nando and Elsa Peretti Foundation has been acknowledged: 

Serrano A, Donno C, Giannetti S, Perić M, Andjus P, D'Ambrosi N, Michetti FThe astrocytic S100B protein with Its receptor RAGE Is aberrantly expressed in SOD1G93A models, and its inhibition decreases the expression of proinflammatory genes. Mediators Inflamm. 2017;. doi: 10.1155/2017/1626204 
 
Geloso MC, Corvino V, Marchese E, Serrano A, Michetti F, D'Ambrosi N. The dual role of microglia in ALS: mechanisms and therapeutic approaches. 
Front Aging Neurosci. 2017. doi: 10.3389/fnagi.2017.00242 
 
Michetti F, D'Ambrosi N, Toesca A, Puglisi MA, Serrano A, Marchese E, Corvino V, Geloso MC. The S100B story: from biomarker to active factor in neural injury. J Neurochem. 2019, 148: 168-87 

• download images (PowerPoint file)
• Figure 1. S100B increase in GFAP-positive cells in the lumbar spinal cord from SOD1G93A rats. Double immunofluorescence labelling was performed with anti-S100B (green) and anti-GFAP (red) in the lumbar spinal cord from wild-type (WT), presymptomatic (ps-SOD1G93A), and end-stage (es-SOD1G93A) rats. The images were acquired from the grey matter (GM) and from the white matter (WM) at two magnifications (scale bars: 50 μm left columns and 20 μm right column). Merged panels also show To-Pro-3 nuclear staining (TP3, blue). In WM from WT rats, insets show a double immunolabelling with anti-S100B (green) and anti-CNPase (red) and merged inset also with To-Pro-3 staining (scale bar: 10 μm).  
• Figure 2. RAGE increase in GFAP-positive cells in the lumbar spinal cord from SOD1G93A rats. Double immunofluorescence was performed with anti-RAGE (red) and anti-GFAP (green) on lumbar spinal cord from wild-type (WT), presymptomatic (ps-SOD1G93A), and end-stage (es-SOD1G93A) rats. The images were acquired from grey (GM) and white matter (WM) at two magnifications (scale bars: 50 μm left column and 20 μm right column). Merged panels also show To-Pro-3 nuclear staining (TP3, blue). In GM from WT rats, insets show a double immunofluorescence with anti-RAGE (red) and anti-ChAT (green) (scale bar: 10 μm). 
• Figure 3.Intracellular and extracellular increases of S100B by transfection of SOD1G93A in C6 cells. C6 cells were transiently transfected with pCMV (mock), pCMV-SOD1wild-type (SOD1wt), or pCMV-SOD1G93A(SOD1G93A). (a) Cells were lysed and analyzed by Western blotting with anti-S100B, anti-RAGE, and anti-SOD1. The lower panel shows signal quantifications expressed in arbitrary units (AU), relative to mock, and reported as mean ± s.d. (n = 3 independent experiments). ∗P < 0.05. (b) ELISA assay for S100B contained in the supernatant of C6 cells transfected with mock, SOD1wt, or SOD1G93A plasmids. Mean ± s.d. (n = 3 independent experiments). ∗P < 0.05.  
• Figure 4. Reduction of proinflammatory genes expression by S100B silencing in SOD1G93A primary astrocytes. Primary astrocytes were transfected with scramble (si-scr) or anti-S100B siRNAs (si-S100B) at 24 and 48 h after plating. 72 hours after plating, cells were lysed, and RNA (a, c) or protein was extracted (b). (a, c) cDNA from si-scr and si-S100B-treated cells was analyzed by real-time qPCR for S100B, GFAP, TNFα, CXCL10, and CCL6 expression. The panels show the quantification of each mRNA expressed in arbitrary units (AU) and reported as mean ± s.d., relative to corresponding si-scr (n = 3 independent experiments). ∗P < 0.05. (b) Western blot analysis with anti-S100B, anti-GFAP, and anti-GAPDH. In the right panels, the quantification of S100B (blue) and GFAP (red) bands normalized to GAPDH and relative to si-scr. Data are expressed as mean ± s.d. (n = 3 independent experiments). ∗P < 0.05.  
• Figure 5. The knockdown of S100B expression in astrocytoma cell line U87 is associated with a downregulation of NF-Kb and iNOS. These effects are more evident in the clone expressing the lowest level of S100B (VA clone).  
• Figure 6. The knockdown of S100B expression shows a protective effect against Aβp mediated cytotoxicity reducing activation of genes involved in astrogliosis.