Synthesis of S-Calpeptin. A mixture of Calpeptin (420 mg, 1.16 mmol, 1.0 eq.) and sodium bisulfite (121 mg, 1.16 mmol, 1.0 eq.) was dissolved in EtOAc/EtOH/H2O (4:2:1, 15 mL) (Figure S8A). The resulting reaction mixture was heated to 40 °C and stirred for 2 h. After cooling to room temperature, the solution was filtered, concentrated, and dried in vacuo to give the desired bisulfite adduct S-Calpeptin (524 mg, 1.12 mmol, 97 % yield) as a white crystalline solid. The product was further characterized by UV (Figure S8B), mass spectrometry (Figure S8C+D) and nuclear magnetic resonance spectroscopy (Figure S9).
Protein purification. Recombinant production of cathepsins was previously described for CatL and CatV 41, CatB42, and CatK43. Mpro was purified as previously described44, with the only difference that 1 mM dithiothreitol in the storage buffer was replaced by 1 mM tris(2-carboxyethyl)phosphane-hydrochloride (TCEP).
Cathepsin and Mpro inhibition test. Calpeptin, S-Calpeptin and GC-376 were tested for their inhibitory properties on human CatB, CatK, CatL, and CatV. All experiments were performed in solution of 50 mM sodium acetate, pH 5.5, 50 mM NaCl and 5 mM DTT. Measurements were taken at 37 °C in 96-well black flat bottom microplates (Greiner, Germany) using Tecan INFINITE M1000 pro plate reader (Tecan, Switzerland) with excitation and emission wavelengths of 370 and 460 nm, respectively. Initial velocities were calculated from the initial linear portions of their curves, assuming steady-state kinetics. IC50 and Ki values were calculated using GraphPad Prism software. Z-FR-AMC (CatK and CatV), Z-RR-AMC (CatB and CatL), and QS145 or Acetyl-VKLQ-AMC (Mpro) substrates were used to monitor reactions.
For initial screening, cathepsins (1 or 10 nM) were mixed with different concentrations of inhibitors (50 µM - 1 nM) to determine the range of their inhibition. For cathepsin – inhibitor pairs that exhibited inhibition in high nanomolar or micromolar range, Ki was determined using 10 nM cathepsin solutions with several inhibitor and substrate concentrations. For those showing inhibition in pM range, cathepsins (0.33 nM) were incubated with 15 inhibitor concentrations of 1.8-fold dilution series (0.01 – 37.5 nM) and Ki was calculated using Morrison equation. These measurements were done in four independent duplicates or triplicates.
To check whether the covalent bond between cathepsins and inhibitors is reversible, we incubated 1 nM CatL with S-Calpeptin and Calpeptin (concentration range 1 – 80 nM) for 60 min at 37 °C. We observed no increase in relative inhibition over time, thus concluding that the covalent bond between cathepsins and inhibitors is reversible.
Inhibition assays of Mpro by Calpeptin, S-Calpeptin and GC-376 were performed in 20 mM HEPES, pH 7.3, 1 mM EDTA, 100 mM NaCl and 1 mM DTT at 37 °C. Ki for S-Calpeptin and Calpeptin were determined using 100 nM Mpro and varying inhibitor/prodrug concentrations (0.4 µM - 97µM) and substrate concentrations. Inhibition by GC-376 was determined using 100 nM Mpro and varying inhibitor concentrations (12.5 nM - 800 nM).
Cell culture and virus production with SARS-CoV-2-GFP
VERO E6 cells and their respective culturing conditions were described previously46. All cell lines were tested to be mycoplasma-free. For SARS-CoV-2-GFP strain47 production VERO E6 cells (in DMEM, 5% FCS, 100 µg/mL Streptomycin, 100 IU/mL Penicillin) were inoculated with virus stock at an MOI of 0.05. After an incubation period of 60 h (37° C, 5% CO2), virus-containing supernatant was collected, spun twice (1000g, 10 min) and stored at -80 °C. For determination of viral titers, a plaque assay was conducted. Confluent monolayers of VERO E6 cells were inoculated with 5-fold serial dilutions of virus supernatants for 1 h at 37 °C. Hereafter, virus inoculum was discarded, serum-free MEM (Gibco, Life Technologies) containing 0.5% carboxymethylcellulose (Sigma-Aldrich) was added and incubated for 48 h (37 °C, 5% CO2). After fixation with 4% formaldehyde (20 min at room temperature), cells were washed extensively with 1X PBS and stained with 1% crystal violet and 10% ethanol in H2O for 20 min. After another extensive washing step with 1X PBS, plaques were counted, and the virus titer was calculated.
Cell lines and cell culture
VERO-CCL81 cell line was obtained from ATCC (ATCC® CCL-81™). LC-HK2 cell line was derived from a tumor induced in nude rat by the inoculation of LC-HK1 cells which was spontaneously established from an explant of a cervical human non-small cell lung cancer metastasis and exhibits similarity to the original lung tumor characteristics, both morphologically and biochemically48–50. The LC-HK2 cells have pleiomorphic morphology, maintaining a small population of mono- or multinucleated giant cells and are aneuploid presenting hyperdiploid DNA content, as established for a human tumor lineage. In addition, LC-HK2 co-express both cytokeratin and vimentin intermediate filaments and are rich in actin microfilaments organized in stress fibers and network pattern exhibiting clusters48–50.
This enables us to identify phenotypes, cellular and molecular changes that accompany the infection process, similar to those that become infected in vivo. Since LC-HK2 holds the biochemical and physiological properties of human lung tissue, it was used in our studies as a human cellular model for SARS-CoV-2 infection.
Both VERO-CCL81 and LC-HK2 cell lines were cultivated using Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% fetal bovine serum (FBS), in an incubator at 37 °C and 5% CO2 atmosphere.
Cell viability assay
VERO-CCL81 was seeded in 96-well plates at a density of 1.5×104 cells/well, while LC-HK2 was seeded at a density of 3×104 cells/well, following 24 hours incubation at 37 °C and 5% CO2 atmosphere. The cell culture media was replaced by serial dilutions of the compounds and the cell viability was determined 72 h post-treatment via CellTiter-Glo® Luminescent Cell Viability Assay (Promega), following manufacturer’s instructions. Luminescent signal was recorded using a the CLARIOstar multi-mode microplate reader (BMG Labtech, Germany). Graphs were generated using GraphPad Prism software version 8.0 (La Jolla California USA, www.graphpad.com). Samples deemed to be technical failures and extreme outlier were removed.
Viral infection
VERO-CCL81 was seeded in 96-well plates at a density of 1.5×104 cells/well, while LC-HK2 was seeded at a density of 3×104 cells/well following 24 hours incubation at 37 °C and 5% CO2 atmosphere. The cells were pre-treated for 2 h with serial dilutions of compounds in fresh DMEM supplemented with 2.5 % FBS. The compounds were removed, and SARS-CoV-2 strain 27 diluted in DMEM supplemented with 2.5 % FBS, was added to the VERO-CCL81 cells at a MOI of 0.01 and to LC-HK2 cells at M.O.I of 0.05, allowing absorption for 1 h. The viral inoculum was removed, and cells were gently washed with phosphate-buffered saline (PBS) without calcium and magnesium. Fresh DMEM supplemented with 2.5 % FBS containing serial dilutions of compounds was added back onto the cells. VERO-CCL81 was incubated for further 48 h, while LC-HK2 was incubated for further 72 h post-infection to assess viral loading or cytopathic effect. All SARS-CoV-2 infections were performed in a biosafety level 3 laboratory at the Institute of Biomedical Sciences, University of São Paulo, Brazil.
Viral loading determination via RT-qPCR
For viral loading evaluation, viral RNA was purified from cellular supernatant using MagMAX™ Viral/Pathogen Nucleic Acid Isolation Kit (Thermo Fisher Scientific) and the samples were processed using the semi-automated NucliSENS® easyMag® platform (bioMérieux, Lyon, France), following the manufacturer’s instructions. The detection of viral RNA was carried out on a QuantStudio™ 3 Real-Time PCR System (Thermo Fisher Scientific) using the AgPath-ID™ One-Step RT-PCR Kit (Thermo Fisher Scientific) and a sequence of primers and probe for E gene (Corman et al., 2020). The viral titers were calculated using a standard curve generated with serial dilutions of a template of known concentration and expressed in tissue culture infectious dose (TCID50)/mL. Infected cells with 0.5% DMSO were used as control. EC50 were calculated by fitting the data using GraphPad Prism version 8.00 (La Jolla California USA, www.graphpad.com). Samples deemed to be technical failures and extreme outlier were removed.
Cytopathic effect inhibition assay
When cytopathic effect occurs due to viral infection, ATP depletion can be measured and correlated with the viral burden51. The cytopathic effect following 48 h post-infection of VERO-CCL81 and 72 h post-infection of LC-HK2 was measured via CellTiter-Glo® (CTG) Luminescent Cell Viability Assay (Promega), following manufacturer’s instructions. Luminescent signal was recorded using a CLARIOstar multi-mode microplate reader (BMG Labtech, Germany). Percent CPE inhibition was defined as [(test compound - virus control)/(cell control - virus control)] *10051. EC50 values were fitted by sigmoidal function using GraphPad Prism version 8.00 (La Jolla California USA, www.graphpad.com). Samples deemed to be technical failures and extreme outlier were removed.
Fluorescence microscopy and image analysis
LC-HK2 was cultivated on coverslips (13 mm diameter) to approximately 80% confluence. The cells were pre-treated for 2 h with 1.5 µM of S-Calpeptin diluted in DMEM supplemented with 2.5 % FBS. The compounds were removed, and SARS-CoV-2 strain diluted in DMEM supplemented with 2.5 % FBS was added to VERO-CCL81 cells at M.O.I of 0.01 and to LC-HK2 cells at M.O.I of 0.05, allowing absorption for 1 h. The viral inoculum was removed, and cells were gently washed with PBS without calcium and magnesium. The cells were incubated in the continued presence of 1.5 µM of S-Calpeptin diluted in DMEM supplemented with 2.5 % FBS and were examined by fluorescence microscopy at the time points 32, 40, 48, 56, 64 and 72 h post-infection.
To this end, the cells were fixed and permeabilized with 3.7% formaldehyde solution (Sigma-Aldrich, F1635) containing 0.2% Triton X-100 in PBS 1x. The fixed specimens were washed with PBS-T (PBS 1x, 0.05% Tween 20) and blocked in PBS-AT (3% BSA, 0.5% Triton X-100 in PBS 1x) at room temperature for 30 min. The cells were incubated for 1 h at room temperature with primaries antibodies diluted in PBS-AT: rabbit anti-cathepsin L (Sigma-Aldrich - ZRB1636, 1:100 dilution) and mouse anti-SARS-CoV-2 spike (Thermo Fisher Scientific - MA536245, 1:200). Hoechst 3342 (Thermo Fisher Scientific - H1399, 10 µg/mL final concentration) was used to counterstain the DNA. The cells were gently washed with PBS-T and incubated with respective secondary antibodies diluted in PBS-AT: chicken anti-mouse AlexaFluor-488 (Thermo Fisher Scientific - A-21200, 1:500 dilution) and donkey anti-rabbit AlexaFluor-568 (Thermo Fisher Scientific - A10042, 1:500 dilution). The coverslips were mounted on glass slides using ProLong Gold mounting media (Thermo Fisher Scientific). From the coverslips of LC-HK2 a series of Z stack images were captured in 0.5 μm thick sections using a ZEISS AxioObserver Z1 equipped with ApoTome2 and bi-gas incubation chamber with a 100X oil-immersion objective. The images were captured using fluorescence range intensity adjusted identically within each experimental series. The entire fixed cell volume was displayed as 2D maximum projections using Image J FIJI (National Institutes of Health) or processed for 3D rendering using Zen 2.6 blue software (ZEISS).
Vesicles size quantifications were obtained from 2D maximum projections of Z stacks exported as greyscale TIFs (16-bit depth) using Automated Image Analysis in Zen 2.6 blue software (ZEISS). Compartments in diameter were generated from thresholding adjustments during overall segmentation. Any objects not meeting the threshold criteria in the segmentation were identified and manually excluded from analyzes. To avoid artefact formation, only identified compartments with diameter range from 0.2 μm ~ 1 μm which correspond to late endosomes or lysosomes (Huotari & Helenius, 2011), were accepted in the output.
P-values were generated by unpaired t-test using GraphPad Prism version 8.00 (La Jolla California USA, www.graphpad.com). Samples deemed to be technical failures and extreme outlier were removed. P<0.05 was considered as statistically significant.
Infection assays for inhibitor screening. VERO E6 were seeded at a density of 10,000 cells per well in 96-well plates. After an incubation time of 24 h, cells were inoculated with SARS-CoV-2-GFP (MOI 0.05) for 24 h. Using an EssenBioscience IncuCyte with IncuCyte 2020C Rev1 software, live-cell imaging was performed by taking pictures every 3 h (scan type: standard; image channels: Phase, Green to detect GFP, Red to detect RFP; objective: 10X). Integrated intensity of detected signal in the green channel was calculated by the IncuCyte 2020C Rev1 software. Data were fitted to a four-parameter logistic function to derive EC50 values using software Origin 2021b.
X-ray crystallography. Mpro was crystallized with compounds (Calpeptin, S-Calpeptin and GC-376) by adding 0.24 µL protein solution (6.25 mg/mL) in 20 mM HEPES buffer (pH 7.8) supplemented with 1 mM TCEP, 1 mM EDTA and 150 mM NaCl, with 0.21 µL reservoir solution containing 100 mM MIB buffer (malonate, imidazole, boric acid) pH 7.5, 25% PEG 1500 (w/w) and 5% (v/v) DMSO, and 0.05 µL reservoir solution containing microseeds obtained by mechanical fragmentation with glass beads (Jena Bioscience). In the case of co-crystallization with Calpeptin, the DMSO contained 5 mM of the respective compound. GC-376 and S-Calpeptin was soaked into crystals, using 5 mM compound concentration and 1 h incubation. CatL was crystallized after removing the protection group S-methyl methanethiosulfonate from the active site Cys25 in the presence of 5 mM DTT. The glycosylation at the position Thr110 was changed by mutation to an alanine to avoid glycosylation. CatL at concentration of 7 mg/mL was equilibrated against 27% (w/v) PEG 8000, 1 mM TCEP and 0.1 M sodium acetate at pH 4.0 in MRC maxi plates by sitting drop vapor diffusion. Crystals grew to maximum size after approximately three days at 20 °C and were transferred to compound soaking solution, which contained 22% (w/v) PEG 8000, 1 mM TCEP and 0.1 M sodium acetate at pH 4.0 as well as 5 % (v/v) DMSO and addidional 10% (v/v) PEG 400 for cryoprotection. The final concentration of the compounds was adjusted to approximately 1 mM. After 12-16 ho the crystals were flash frozen in liquid nitrogen and measured at the PETRAIII synchrotron. The rod-like crystals were routinely measured at three positions and the datasets were merged to obtain complete datasets. All crystallographic data were processed with XDS and refinement was in general performed with phenix interspersed with manual model building in COOT. CatK with S-Calpeptin was crystallized in 30 % of PEG-3350 and 0.2 M CaCL2 at 293 K with sitting drop vapor diffusion experiments. CatV with Calpeptin was crystallized in 77 % MPD and 23 % of 60 mM TRIS, pH 8.0 at 278 K with sitting drop vapor diffusion experiments. The cathepsin structures were refined using MAIN52. Further structure analysis and visualization was done by using PyMOL and Chimera.
In vivo studies, virus stock and titration. The SARS-CoV-2 (B.1.1.28, SARS-CoV-2/SP02/2020HIAE, GenBank MT126808.1) used in animal inoculations was first isolated in VERO E6 cells from a nasopharyngeal swab of one of the first patients reported with SARS-CoV-2 in Brazil27. The sample was confirmed to be free of 15 other viral agents (Endemic Coronavirus - CoV-NL63, -229E, -HKU1 and -OC43, Enterovirus, Influenza A and B, Parainfluenzavirus 1, 2, 3 and 4, Rhinovirus, Human Metapneumovirus, Respiratory Syncytial Virus, and Adenovirus) by qRT-PCR and SARS-CoV-2 presence was confirmed using a specific qRT-PCR assay27,54.
For virus stock production, VERO CCL-81 cells were infected with a M.O.I. (multiplicity of infection) of 0.1 during 48 h, at 37 °C and 5 % CO2, and the virus was subsequently stored at -80°C. For titration, virus samples were ten-fold serially diluted (10-1– 10-12) in Dulbecco’s Modified Eagle Medium (DMEM), with 2.5% of fetal bovine serum (FBS), and inoculated in sextuplicate on 96-well plates containing 2 x 104 VERO CCL-81 cells/well. After 72 h at 37 °C with 5% CO2, the plates were microscopically inspected for CPE due to SARS-CoV-2 and the monolayer fixed and stained with Naphthol Blue Black (Sigma-Aldrich) solution (0.1% naphthol blue w/w with 5.4% of acetic acid and 0.7% of sodium acetate) for visual confirmation. Viral titer was calculated using the Spearman & Kärber algorithm55 and expressed in TCID50 (Tissue Culture Infectious Dose)/mL. The isolate used in this study was on its 3rd passage in VERO CCL-81 cells.
S-Calpeptin toxicity experiment in Golden Syrian hamsters. Twelve conventional male, Golden Syrian hamsters (Mesocricetus auratus; 6-8 weeks-old) were acquired from the Instituto Gonçalo Muniz, Fiocruz, Salvador, Brazil and housed in the Animal Facility of the Department of Preventive Veterinary Medicine and Animal Health of the College of Veterinary Medicine, University of São Paulo, Brazil. Animals were acclimatized for seven days, received food and water ad libitum and were divided into four groups (n=3 each), homogenized based on weight. Animals of each group received a daily subcutaneous dose of S-Calpeptin suspended in 1:100 dimethyl sulfoxide (DMSO) (1 mg/kg, 2 mg/kg, or 3 mg/kg of body weight) or a solution of 1:100 DMSO for seven consecutive days, starting at day zero, alternating the inoculation site from the right to the left flank in a maximum volume of 30 μL (Figure S10A). Animals were checked and weighted daily. After seven days of inoculation, animals were euthanized with morphine (2.5 mg/kg) subcutaneously followed by an intraperitoneal overdose of ketamine (600 mg/kg) and xylazine (30 mg/kg). Immediately following cardiac arrest, total blood was collected via cardiac puncture and placed in microtubes containing EDTA (BD Biosciences, USA). Animals were necropsied and samples of the brain, liver, pancreas, gastro-intestinal tract, kidney, lungs, spleen, and trachea were collected and fixed in 10% buffered formalin. Total blood was then subjected to a complete blood cell count at the private veterinary clinical pathology laboratory Lab&Vet, São Paulo, Brazil. All procedures were approved by the Committee on Animal Use and Experimentation from the College of Veterinary Medicine, University of São Paulo, Brazil (protocol # 8711260321).
Anti-viral experiment with Golden Syrian hamsters. A total of 36 conventional male, Golden Syrian hamsters (Mesocricetus auratus; 6-8 weeks-old) were acquired from the Instituto Gonçalo Muniz, Fiocruz, Salvador, Brazil and housed at the Animal Biosafety Level 3 Laboratory of the Department of Parasitology, Institute of Biomedical Sciences, University of São Paulo, Brazil. Animals were acclimatized for seven days and kept individually in microisolators with food and water ad libitum. Animals were checked and weighted daily.
Hamsters were separated into four groups (G1-G4), homogenized based on weight. On day zero (infection day), all animals were initially anesthetized with 100 mg/kg of ketamine and 10 mg/kg of xylazine intraperitoneally. Hamsters from G1 (n=15) and G2 (n=15) were then inoculated intranasally with 105 TCID50 of SARS-CoV-2 (in 50 μL of DMEM, 2.5% FBS), while hamsters from G3 (n=3) and G4 (n=3) were inoculated intranasally with 50 μL of DMEM, 2.5% FBS. From day 1 to day 7 post infection (p.i.), animals from G1 and G3 received 1 mg/kg of S-Calpeptin diluted in 1:100 DMSO subcutaneously once a day (volume ranged from 16-26 μL/animal), while animals from G2 and G4 received 16-26 μL of 1:100 DMSO daily, also subcutaneously (Figure S5A and B).
On days 3, 5, and 7 p.i., subgroups of 5 animals each from G1 and G2 were euthanized using 5 mg/kg of morphine subcutaneously followed by 600 mg/kg of ketamine and 30 mg/kg of xylazine intraperitoneally. Animals from G3 (n=3) and G4 (n=3) were euthanized with the same protocol on day 7 p.i. (Figure S5A). Hamsters were necropsied and samples from trachea and lungs were collected for viral load determination (in DMEM, 2% FBS with 10.000 U/mL of penicillin, 10 mg/mL of streptomycin, 25 μg of amphotericin B/mL, and 2 mm glass beads) and histopathology (in 10% buffered formalin). Organ samples were individually weighted and rapidly frozen in liquid nitrogen and transferred to -80 °C for posterior analysis. Clean, sterile instruments were used between organ collection to avoid viral cross-contamination. All five lung lobes (right cranial lobe, right middle lobe, right caudal lobe, accessory lobe, and left lung) were equally represented in all samples and individually identified to be evaluated by histopathology analysis. All procedures were approved by the Committee on Animal Use and Experimentation from the Institute of Biomedical Sciences, University of São Paulo, Brazil (protocol # 9498230321).
Viral load quantification from tissue samples. The viral load was quantified by determining the TCID50/g of tissue and the number of viral RNA copies/β-actin RNA copies/g of tissue. Briefly, all samples of trachea and lungs were thawed and subjected to disruption using 2 mm glass beads in a TissueLyser II equipment (Qiagen, Germany) at 30 Hz for 2 min twice, followed by centrifugation at 13,000 rpm (Eppendorf 5804R centrifuge) for 1 min. The supernatants were used for viral load quantification by TCID50 determination and for viral RNA copies determination via quantitative reverse transcription PCR (qRT-PCR). For TCID50 determination, the samples were serially diluted ten-fold (10-1 to 10-12) with DMEM containing 2.5% FBS and inoculated in six replicates in 96-well plates containing VERO CCL-81 at a density of 5x104 cells/well. After incubation at 37°C with 5% CO2 for 72 h, the plates were microscopically inspected for CPE caused by SARS-CoV-2. The monolayers were fixed and stained with a Naphthol Blue Black (Sigma–Aldrich) solution (0.1% naphthol blue w/w with 5.4% acetic acid and 0.7% sodium acetate) and analyzed to confirm the results. The viral titer was calculated using the Spearman & Kärber algorithm and reported in TCID50/mL. For RT-qPCR, total nucleic acids were extracted using a Magmax Core Kit with a MagMAX Express Magnetic Particle Processor (Thermo Fisher Scientific) according to the manufacturer’s instructions. The detection of SARS-CoV-2 RNA was performed based on a previously described protocol using a one-step qRT–PCR assay kit (AgPath-ID™ One-Step RT–PCR Reagents, Applied Biosystems Inc.) and an ABI 7500 SDS real-time PCR machine (Applied Biosystems). The duplex reaction was performed using specific primers for the E gene of SARS-CoV-2 and primers for β-actin gene (ActB Rv 5’ CAC CAT CAC CAG AGT CCA TCA C 3’, ActB F CTG AAC CCC AAA GCC AAC; ActB_P- HEX TGT CCC TGT ATG CCT CTG GTC GTA ZEN/IOWA BLACK) that was used as a housekeeping gene. The number of RNA copies/mL was quantified based on a standard curve obtained by serially diluting a synthetic dsDNA sequence (Gene Blocks, IDT) corresponding to the amplification fragment of the target gene. Data were obtained from three replicates in one biological experiments.
Histopathology. Organ samples from the toxicity and the anti-viral experiments were collected on necropsy and fixed in 10% buffered formalin for 24 hours. Three to five sections from each organ were obtained and samples were routinely processed for histological examination. Examination of hematoxylin and eosin-stained slides was done blindly by a board-certified anatomic pathologist on a Nikon® E200 optical microscope. For the anti-viral experiment, lung lesions were semi-quantitatively evaluated based on parameters described in the literature56–58 and according to an in-house developed scoring system (Table S3). A score was attributed to each parameter for each lung lobe of every animal. The final score of each lung parameter was constituted by the sum of scores attributed for each lung lobe. Therefore, the maximum value each lung parameter was able to reach was 15. For the trachea, a score of 0 was given to normal trachea, whereas scores of 1, 2 or 3 were given to mild, moderate, and severe tracheitis, respectively.
Statistical analyses. Data from weight loss during the infection period was subjected to a normality test according to D’Agostino & Pearson and then analyzed using one-way ANOVA followed by Tukey HSD. Viral load and histopathology scores were compared among groups using Mann-Whitney U test or Kruskall-Wallis with Dunn test for pairwise comparisons. Results were considered statistically significant when p≤0.05 and analyses were performed in GraphPad Prism (version 9.1.1).