EHS
EHS

Polyphenolic grape stalk and coffee extracts attenuate spinal cord injury-induced neuropathic pain development in ICR-CD1 female mice


Preparation of polyphenolic extracts and quantification of polyphenol content in the extracts

The two polyphenolic extracts used in the present study were obtained from different natural sources. The grape stalk extract (GSE) was obtained from Grape stalks wastes of Cabernet Sauvignon and Merlot varieties, generated in the wine production process in L’Empordà region of Catalonia, genteelly provided by the owner of the winery Celler Roig Parals (Mollet de Peralada, Catalonia). The plant collection was accomplished in accordance with the national guidelines and regulations. The coffee extract (CE) was obtained from commercial decaffeinated ground roasted coffee, blend of Robusta natural (pure Coffea canephora) and Arabica natural (pure Coffea arabica) at a roast point 7 over 10 (dark medium).

To collect the GSE particles, grape stalk residual material was cut with a chopper grinder (WCG75 Pro Prep) and was then sifted with a digital electromagnetic sieve (CISA BA 200N) to achieve a particle size between 0.5 and 1 mm. For CE, processed coffee powder was used. To obtain the extracts, 3 g of either grape stalk or coffee particles were mixed with 50 mL of saline solution (0.9% Vitulia Physiological Serum, Barcelona, Spain) as a solubilizer, which was refluxed and stirred at 100 °C for 2 h. The resulting solutions were first filtered with chromatographic filters (Scharlau Nylon Syringe filter, ø 13 mm, 0.45 μm, NY13045200) and were then filtered and sterilized with 0.22 μm vacuum filter bottles (FPE-204-250, Biofil).

The total amount of polyphenols in both extracts was determined by the Folin-Ciocalteu assay. This method, also called the gallic acid equivalence method (GAE), consists of using the Folin-Ciocaiteu reagent, which is a mixture of phosphomolybdic acid (H3PMo12O40) and phosphotungstic acid (H3PW12O40) that reduces to molybdenum and tungsten oxide in the presence of phenolic and polyphenolic compounds25. The Folin Ciocalteu method was performed using gallic acid as a standard. The calibration curve was obtained by preparing different standard concentrations of gallic acid within the range 100–1000 mg/L. Briefly, 100 µL aliquots of diluted extracts (1:2, 1:3, 1:4), gallic acid standard solutions (100–1000 mg/L) and a blank (saline solution) were placed in different tubes. Then, 3.9 mL of Milli-Q water, 900 µL of 20% sodium carbonate and 600 µL of Folin-Ciocalteu reagent were added. The tubes were shaken and then allowed to incubate for 2 h at room temperature. After incubation, the absorbance against a blank was measured spectrophotometrically at 760 nm (Hitachi U-2000 VIS/UV spectrophotometer). The total polyphenolic content was expressed as milligrams of gallic acid equivalents (GAE) per liter of extracts.

Characterization of phenolic compounds in the extracts

Qualitative determination of polyphenols in GSE was accomplished by high-performance liquid chromatography high resolution mass spectrometry (HPLC-HRMS) carried out by the scientific technical services of the University of Barcelona. Both GSE and CE analysis samples were prepared by filtering through a 0.45 μm syringe filter. GSE was diluted (1/5 and 1/50) with 0.1% formic acid, and the CE sample was diluted (1/10, 1/100 and 1/250) with 0.1% formic acid. Standard solutions of the identified polyphenols were used to confirm their presence and for quantification in both extracts.

HPLC-UV-ESI-TOFMS was performed to identify and quantify the polyphenols present on GSE. The HPLC system consisted of Agilent 1200RR chromatograph equipment. A Luna HST 2.5 μm (10 cm × 2.0 mm) column (Phenomenex) was used. The mobile phase consisted of 0.1% formic acid (A) and acetonitrile (B) with the following gradient (t (min), % B): (0, 0), (0.5, 0), (10, 50), (12, 50), (25, 95), (28, 95), (28.5, 0), and (35, 0). The flow rate was 400 μL/min, with no split before the MS. The oven temperature was 50 °C, and the automatic injection system temperature was 10 °C. The injection volume was 10 μL. This HPLC system was coupled to a QSTAR Elite (ABSciex) mass spectrometer fitted with a TurboIon spray source working in negative ionization mode with the following TOFMS conditions: full scan analysis from m/z 100 to 600 and product ion scan m/z 449.

HPLC-UV-ESI-FTMS was performed to identify and quantify the polyphenols present on CE. The HPLC system consisted of Ulltimate 3000 (Dionex) chromatograph equipment. A Kinetex EVO C18 1.7 μm (10 cm × 2.0 mm) column (Phenomenex) was used. The mobile phase consisted of 0.1% formic acid (A) and acetonitrile 0.1% formic acid (B) with the following gradient (t (min), % B): (0, 0), (0.5, 0), (5, 10), (8, 90), (15, 20), (17, 50), (17.5, 50), (18, 95), (19, 95), (19.5, 0), and (22,0). The flow rate was 500 μL/min, with no split before the MS. The oven temperature was 50 °C, and the automatic injection system temperature was 4 °C. The injection volume was 2 μL. This HPLC system was coupled to an LTQ-Orbitrap Velos (Thermo) mass spectrometer fitted with an electrospray source working in negative ionization mode with the following MS conditions: full scan analysis from m/z 100 to 2000 at 30,000 resolution using FTMS.

Animals

Adult female ICR-CD1 mice (20–30 g) were purchased from Janvier Laboratories (Le-Genest-SaintIsle, France). All animals were housed in groups of 4–5 in standard Marcolon cages (28 × 28 × 15 cm) with wood shaving bedding at 21 ± 1 °C and 40–60% humidity under a 12:12-h light–dark cycle and fed ad libitum with a standard diet of mouse pellets (TEKLAD 2014, Harlan Interfauna Ibérica, Sant Feliu de Codines, Catalonia, Spain). Cages were changed twice weekly. All mice were allowed to acclimatize for at least 1 h to the facility rooms before any functional, behavioral, or surgical procedures, which were all conducted during the light cycle. Sentinel mice were routinely tested for pathogens, and facilities remained pathogen free during the whole experimental period. The number of mice used in this study in all procedures was maintained at a minimum, working with experimental groups consisting of 6 to 12 mice. The animal sample size was calculated using GRANMO (Version 7.12 April 2012) and based on the ethical limits exposed by the Animal Ethics Committee.

All experimental procedures and animal husbandry were conducted following the ARRIVE 2.0 guidelines and performed according to the ethical principles of the IASP for the evaluation of pain in conscious animals26 and the European Parliament and the Council Directive of 22 September 2010 (2010/63/EU), along with the approval Ethical Committee on Animal Experimentation (CEEA) of the University of Barcelona and the Department of Agriculture, Livestock, Fisheries, Food and Natural Environment of the Generalitat de Catalunya, Government of Catalonia (DAAM number 9918-P3). All experiments were carried out at the animal experimentation unit of the Bellvitge campus (University of Barcelona).

Surgical procedure and treatments

Animals were anesthetized with sodium pentobarbital (50 mg/kg, i.p.) and placed prone on a heating pad to maintain constant body temperature levels. After back shaving and disinfection with povidone iodide, T8–T9 thoracic spinal cord segments were exposed via dorsal laminectomy, and using the contusion weight-drop technique, two grams of weight was dropped from 25 mm high onto the metallic stage located over the exposed spinal cord (centered in the midline), to induce mild spinal cord injury17,27,28. Following this procedure, the wound was closed, and the animals were allowed to recover in warmed cages with access to food and water. After the surgical procedure, animals also received 0.5 mL saline solution (i.p.) to restore an eventual volemic deficit. In sham animals, the spinal cord was exposed as described above but not contusioned, and they underwent the same recovery procedures.

Thirty minutes after spinal cord injury and daily during the first week postinjury, different sets of spinal cord injury (SCI) animals received intraperitoneal administration of (i) grape stalk extract (GSE) at doses of 10, 15 and 20 mg/kg or (ii) coffee extract (CE) at doses of 10 and 15 mg/kg. Another set of animals also received saline solution, the dilution vehicle of GSE and CE. The experimental groups of the present study were GSE10 (SCI-mice treated with GSE at 10 mg/kg), GSE15 (SCI-mice treated with GSE at 15 mg/kg), GSE20 (SCI-mice treated with GSE at 20 mg/kg), CE10 (SCI-mice treated with CE at 10 mg/kg), CE15 (SCI-mice treated with CE at 15 mg/kg), saline (SCI-mice treated with saline solution at a volume equivalent to that injected in the animals with the dose of 15 mg/kg) and sham (mice with dorsal laminectomy without spinal cord contusion and without treatment).

Functional evaluation

Locomotor activity

Locomotor activity was evaluated using a circular open field (70 cm diameter × 24 cm wall height), where each animal was allowed to move freely for 5 min. Two independent examiners observed the hindlimb movements of the mouse and scored the locomotor function according to the Basso Mouse Scale for locomotion (BMS)29. The final score of each animal was the mean value of both examiners. The BMS ranges from 0 (no hindlimb movement) to 9 (normal movement-coordinated gait). The objective of evaluating this test is to determine the state of locomotion of the animals after the contusion of the spinal cord and to verify that all of them have scores on the BMS scale higher than 5–6 points, which means that they maintain the ability to move and/or voluntarily remove the hind legs. This test was evaluated before inducing spinal cord injury and at 7, 14, and 21 days postinjury (dpi).

Nociceptive test: mechanical allodynia

Mechanical allodynia was evaluated by assessing 50% withdrawal thresholds using a set of von Frey monofilaments (bending force range 0.04–2 g) following the up-down paradigm27,28. Animals were placed in plastic tube test chambers with a metal mesh floor that allowed full access to the plantar surface of the hind paws. Behavioral accommodation was allowed for approximately 1 h until cage exploration and major grooming activities ceased. Then, von Frey monofilaments were perpendicularly applied to the plantar surface with sufficient force to cause slight buckling against the paw. First, the 0.4 g filament was applied, and then, the strength of the filament was decreased when the mouse responded or increased when it did not respond. Clear paw withdrawal, shaking or licking were considered to be a response. This up-down procedure was limited to four assessments after the first response. Each filament was applied for 2 s with interstimulus intervals of 5–10 s. Both paws were evaluated since the SCI model results in bilateral injury, and it is not possible to use the contralateral paw as a natural intraindividual control. The mechanical threshold that produced 50% of responses was calculated using the Dixon formula30: 50% paw withdrawal threshold (g) = [(10(Xf + κδ)/10,000)], where Xf is the value (in logarithmic units) of the final von Frey filament used, κ is a fixed tabular value for the pattern of positive/negative responses and δ is the mean difference (in logarithmic units) between stimuli. This test was evaluated before inducing SCI and at 7, 14, and 21 dpi and after the evaluation of locomotor activity.

Nociceptive test: thermal hyperalgesia

Thermal hyperalgesia was assessed by measuring the hind paw withdrawal latency in response to a thermal stimulus. Plantar tests were performed according to the Hargreaves method17,27,28,31 using a plantar test analgesimeter (#37370; Ugo Basile, Comerio, Italy). Mice were placed into plastic test enclosures with an elevated glass floor and allowed to acclimate for approximately 1 h until cage exploration and major grooming activities ceased. Then, the light of a projection lamp (100 W) was focused onto the plantar surface of the hind paw with a time limit of 30 s to avoid skin damage. Withdrawal latency was automatically recorded by a time-meter coupled to infrared detectors directed to the plantar surface of the paw. The sum of the mean withdrawal latencies for both hind paws was determined from the average of three separate trials conducted at 5-min intervals. This test was also evaluated before inducing spinal cord injury and at 7, 14 and 21 dpi and after mechanical allodynia evaluation.

Tissue sample collection

At 21 dpi and after functional evaluation, animals were anaesthetized with sodium pentobarbital (90 mg/kg; i.p.) Then, blood was extracted through the insertion of an intracardiac needle. Subsequently, the obtained blood was centrifuged for 15 min at 4000 rpm to obtain the serum, which was frozen immediately in dry ice and stored at − 80 °C until analysis.

On the one hand, half of the animals in each experimental group were perfused intraventricularly with 4% paraformaldehyde solution in phosphate-buffered saline (PBS, 10 mM sodium phosphate buffer, pH 7.4). Afterwards, spinal cord, brainstem and brain tissues were carefully removed. Spinal cord samples were immersed in a Zamboni fixing solution (4% paraformaldehyde and 0.3% picric acid in PBS32). The spinal cord samples were conserved in Zamboni solution at least 14 days after extraction and then were preserved in 30% sucrose solution in PBS. These samples were stored at 4 °C until histological analysis by immunohistochemical techniques. On the other hand, the entire brain, including the brainstem, was removed from each mouse and postfixed in 4% paraformaldehyde at 4 °C until histological analysis. In the other half of the animals in each experimental group, the spinal cord was exposed by a dorsal laminectomy and was then meticulously extracted and stored at − 80 °C. In addition, the entire brain was removed from the cranium of each mouse and stored at − 80 °C. All these samples were used for molecular biology studies.

Histological analysis

Spinal cord samples

For the immunohistochemical analysis, T7-T10 spinal cord segments were embedded in Tissue Freezing Medium (0201-08926, Leica, Barcelona) and cut transversely with a cryostat (CM1520, Leica, Barcelona) into 60 µm thick sections that were collected in six-well porcelain plates. Concretely, the T7-T10 spinal cord block was cut in half at the level of T8, generating a rostral (T7-T8) and a caudal (T9-T10) block. Starting from T8, transverse sections of the spinal cord were made. First, sections were washed 2 times for 10 min with saline phosphate buffer (PBS, 0.1 M, pH = 7.4) and 2 times for 10 min with 0.1 M PBS/0.3% Triton (PBS-Triton). Next, tissue sections were blocked with 1% bovine fetal serum in PBS-Triton (PBS-Triton-FCS) for 1 h and were then incubated with rabbit anti-GFAP (Glial fibrillary acidic protein; 1:200, ab7260, Abcam) or rabbit anti-IBA1 (Ionized calcium-Binding Adaptor molecule 1; 1:200, 019-19741, Fujifilm Wako) for 48 h at 4 °C in a humidified chamber to avoid tissue drying out. As a specificity control, some spinal cord sections were incubated without primary antibody. The sections were washed three times for 10 min with PBS-Triton and then incubated overnight at 4 °C in a humidified chamber with AffiniPure goat anti-rabbit IgG conjugated with cyanine 3 (Cy3) (1:200, Ref # 111-165-144, Jackson ImmunoResearch, USA). Finally, two 10-min washes were again performed with PBS-Triton and one 10-min wash with PBS, and the samples were mounted on previously gelatinized slides. Once mounted, slides were dehydrated through immersion in increasing concentrations of ethanol baths (70%, 96% and 100%) and were covered with cover glass fixed with DPX mounting media (1.01979.0500, Merck, Germany)17.

Histological sections were observed with an epifluorescence microscope (Leica DMR-XA; Leica Microsystems) attached to a digital camera (FMVU-13S2C-CS; Point Gray Research, Canada) used to capture the images (×200). The resulting images were analyzed with the free software ImageJ (Image Processing and Analysis in Java, National Institute of Health, NIH, USA). The ×200 immunolabeled images for GFAP and Iba1 were taken preferentially in the gray matter of the dorsal horn of the spinal cord, although in some cases white matter tracts were also taken very partially. For each animal in the present study, a minimum of eight immunolabeled histological sections were analyzed. Imaging of the sections immunolabeled for IBA1 was performed for reactive and nonreactive microglial cells and was expressed as a percentage of two phenotypes. The percentage of reactive microglial cells was considered an index of the degree of microgliosis. Regarding GFAP immunolabeling, the area immunopositive for GFAP was measured as a percentage of immunoreactivity17. In addition, whole spinal cord coronal sections images were captured to observe the preservation of ventrolateral funiculus in all experimental groups. Whole sections of the spinal cord were taken at ×50.

Brain and brainstem samples: supraspinal structures

Serial PAG coronal sections (12 µm) from the central part of the superior colliculi to the upper edge of the inferior colliculi corresponding to PAG between 6.72 and 8.04 mm from the bregma33 were cut (Leica 1800 cryostat; Leica Microsystems, Wetzlar, Germany). For ACC, serial coronal sections (12 µm) through the prefrontal cortex ACC between 2.2 and 4.2 mm from the bregma33 were also prepared.

The sections were collected on gelatin-coated microscopic slides, air-dried, and processed for immunohistochemical staining. First, the sections were washed with PBS containing 0.05% Tween 20 (PBS-TW20) and 1% bovine serum albumin for 10 min and then treated with 3% normal donkey serum in PBS-TW20 for 30 min. Then, the sections were incubated with rabbit anti-GFAP (1:250; DAKO) or rabbit anti-IBA1 (1:100; Wako) antibodies in a humid chamber at room temperature for 12 h. The binding of primary antibodies was visualized by secondary antibodies (FITC- or TRITC-conjugated, affinity purified goat anti-rabbit; 1:100; Jackson) for 90 min at room temperature. To detect the M2 phenotype subtype of microglial cells, some PAG sections were double immunostained with mouse anti-OX42 (1:100; Santa Cruz) and rabbit anti-CD206 (1:100; Abcam) primary antibodies and developed with purified goat anti-mouse and anti-rabbit FITC- and TRITC-conjugated secondary antibodies, respectively. Immunostained sections were rinsed, stained with Hoechst 33342 to detect the positions of the cell nuclei, and mounted in Vectashield aqueous mounting medium (Vector Laboratories Inc., Burlingame, CA). The control sections were incubated with omission of the primary antibody and displayed no immunostaining.

PAG and ACC sections were analyzed using a Nikon Eclipse NI-E epifluorescence microscope equipped with a Nikon DS-Ri1 camera driven by NIS-elements software (Nikon, Prague, Czech Republic). The areas immunopositive for GFAP or IBA1 were detected by a thresholding technique after subtraction of the background. The areas of immunostaining for GFAP or IBA1 were related to the areas of interest and expressed as the proportion of relative area ± SEM. Quantification of OX42-immunostained and OX42/CD206 double-immunostained microglial cells was carried out using a Leica TCS SP5 confocal microscope at 20× magnification (Leica HC APO L 20×/0.50 W objective). Stacks of confocal images across 6 randomly selected regions of interest were collected from 4 whole sections through PAG. A proportion of OX42/CD206 double immunostained M2 subtype microglial cells to all OX42-immunopositive cells was counted from the results examined by 2 investigators blinded to the tissue source.

Molecular biology analysis: western blot

Spinal cord samples

Spinal cord tissue was homogenized in modified RIPA buffer (50 mM Tris–HCl pH 7.5, 1% Triton X‐100, 0.5% sodium deoxycholate, 0.2% SDS, 100 mM NaCl, 1 mM EDTA, 2 mM PMSF, 1 µg/μL aprotinin, 1 µg/μL leupeptin, and 2 mM sodium orthovanadate) and was then centrifuged at 18,000g at 4 °C for 30 min. The protein concentration from the obtained supernatant was determined by Protein Assay DCTM (Bio‐Rad). The samples were then stored at − 80 °C until use.

Samples containing equal amounts of protein (10–20 µg) were mixed with 2× Laemmli sample buffer (S3401, Sigma-Aldrich) and boiled at 95 °C for 10 min. Samples were fractioned by 10–15% (w/v) SDS-PAGE gels and transferred onto a nitrocellulose membrane and then blocked with either 5% nonfat dry milk or bovine serum albumin (BSA) in Tris-0.1% Tween 20-buffered saline (T-TBS) for 2 h at room temperature.

Membranes were incubated with the following primary antibodies overnight at 4 °C: rabbit anti-IBA1 (1:700, 0019-19741, FUJIFILM Wako Chemicals), rabbit anti-GFAP (1:800, SAB4300647, Sigma-Aldrich), rabbit anti-MCP1/CCL2 (1:1000, ab25124, Abcam), rabbit anti-CCR2 (1:1000, ab203128, Abcam), rabbit anti-CX3CL1 (1:1000, ab25088, Abcam), rabbit anti-CX3CR1 (1:1000, ab8021, Abcam), rabbit anti-extracellular signal-regulated kinases (total ERK ½) (1:1000), and diphosphorylated ERKs (pERK ½) (1:1000). Rabbit anti-GAPDH antibody (1:10,000, G9545, Sigma-Aldrich) was used as a loading control.

The immunoblots were washed three times for 10 min with T-TBS and then incubated for 1:30 h at room temperature with horseradish peroxidase-conjugated goat antirabbit IgG (1/50,000, AP132P, Sigma-Aldrich) and revealed by chemiluminescence Clarity Western ECL Substrate (170-5061, Bio-Rad). Band pixel intensities were quantified by Gel‐Pro Analyzer software (Media Cybernetics, USA) and normalized to the corresponding GAPDH intensity. pERK ½ was normalized to total ERK and subsequently normalized to GAPDH immunoreactivity intensity. In the supplementary material are available the original scanned/revealed full blots. Note that considering that some spinal cord sample blots have been manually developed with a film, without a developer machine and it is not possible to visualize the membrane on the developed films of GFAP, IBA1, CCR2, CX3CR1 and pERK/ERK. The blots were cut prior to hybridization with the antibody of interest and GAPDH, the loading control and consequently not all blot images present exactly the same length.

Supraspinal structures

To obtain the PAG area, a 2 mm thick coronal slice of the mesencephalon was carefully removed at the position from the central part of the superior colliculi to the upper edge of the inferior colliculi corresponding to PAG between 6.72 mm and 8.04 mm from the bregma. For ACC, a 2 mm thick coronal slice of the prefrontal cortex was carefully removed at the position between 2.2 mm and 4.2 mm from the bregma33.

The PAG and ACC tissue samples of individual animals were homogenized in RIPA buffer (Abcam) containing protease inhibitors (LaRoche, Switzerland) and were then centrifuged at 15,000g at 4 °C for 20 min. The protein concentration from the tissue supernatant was measured by a Nanodrop ND-1000 (Thermo Scientific) and normalized to the same levels.

Each sample, containing 50 µg of protein, was separated by SDS–polyacrylamide gel electrophoresis and transferred onto nitrocellulose membranes. The membranes were then blocked with 1% BSA in PBST (3.2 mM Na2HPO4, 0.5 mM KH2PO4, 1.3 mM KCl, 135 mM NaCl, 0.05% Tween 20, pH 7.4) for 1 h at room temperature and incubated with the following primary antibodies overnight at 4 °C: rabbit anti-IBA1 (1:100, FUJIFILM Wako Chemicals), rabbit anti-GFAP (1:10,000, Abcam), rabbit anti-CX3CL1 (1:2000, Abcam), and rabbit anti-CX3CR1 (1:2000, Abcam). Mouse anti-α-tubulin antibody (1:1000, Cell Signaling) was used as a loading control.

Blots were washed in PBST and incubated with peroxidase-conjugated anti-mouse or anti-rabbit secondary antibodies (Sigma, 1:1000) at room temperature for 1 h. Protein bands were visualized using the ECL detection kit (Amersham) on an LAS-3000 chemiluminometer reader (Bouchet Biotech) and analyzed using densitometry image software.

Biochemical analysis of hepato- and nephrotoxicity

Serum samples were analyzed to study the possible hepatotoxicity and nephrotoxicity produced by the different vegetable extract treatments. To this end, serum levels of alanine aminotransferase (ALT/GPT; # 11533, BioSystems, Barcelona, Spain), aspartate aminotransferase (AST/GOT; #11531, BioSystems, Barcelona, Spain) and urea-BUN (# 11536, BioSystems, Barcelona, Spain) were determined by using commercial assay kits according to the manufacturer’s instructions.

Statistical analysis

All functional, histological, and biochemical analyses were performed in a blinded manner using a code for each mouse. The results are shown as the mean ± standard deviation of the mean (SEM). The normal distribution of the data was analyzed by Shapiro–Wilk or Kolmogorov–Smirnov tests before further applying parametric or nonparametric statistical analyses. Data that followed a normal distribution were analyzed using repeated measures MANOVA (Wilks’ criterion) and analysis of variance (ANOVA) followed by Duncan’s test, when applicable. Data that did not follow a normal distribution were analyzed using the Friedman statistic test for nonparametric repeated measures and Kruskal–Wallis followed by the Mann–Whitney U test. In all statistical analyses, the α level was set at 0.05 using the statistical package SPSS 25.0 for Windows.

Ethics approval and consent to participate

All experimental procedures and animal husbandry were performed according to the ethical principles of the I.A.S.P. for the evaluation of pain in conscious animals and the European Parliament and the Council Directive of 22 September 2010 (2010/63/EU), along with the approval Ethical Committee on Animal Experimentation (CEEA) of the University of Barcelona and the Department of Agriculture, Livestock, Fisheries, Food and Natural Environment of the Generalitat de Catalunya, Government of Catalonia (DAAM number 9918-P3). All experiments were carried out at the animal experimentation unit of the Bellvitge campus (University of Barcelona).



Source link

EHS
Back to top button