Investigating the Cytoprotective Mechanisms of the Tardigrade Damage Suppressor (Dsup) Protein in Human Cells Under Hypoxic Stress

To test whether the Dsup transfection confers protection against hypoxia, we exposed Dsup+ and Dsup− cells to 100 μM cobalt chloride (CoCl₂) (22) for 24, 48, and 72 h. As shown in Figure 1A, Dsup+ cells exposed to hypoxia showed an increased survival rate at each time point. After 72 h of hypoxia, cells tended to die, but death extension was lower when compared to Dsup− cells (p < 0.01). The mechanisms of this protection may be linked to the increased transcription of genes involved in hypoxia-responsive mechanisms. In fact, when analyzed by real-time PCR after 72 h of hypoxia, in Dsup+ cells, we observed an increased fold change of approximately 1 of more than 20 hypoxia response-related genes belonging to these classes: transcription factors, immune and stress response, signal transduction, apoptosis, oxidoreductases, proteases, and protein metabolism as well as induction of HIF1α (Table 1). Networking among these proteins evaluated with STRING is shown in Figure 1B.

We also performed a proteomic analysis on these cells, and an average of 3200 spots per gel was detected by image analysis software. Only spots with a p-value ≤ 0.05 and a fold change ≥ 2 were considered. This differential analysis highlighted 54 differentially abundant spots, of which 19 were down and 35 were upregulated (Figure S1 and Table S1). Of 54 spots, 45 proteins were identified by MALDI ToF mass spectrometry. We performed a principal component analysis (PCA) of the %V data of the differential spots to easily identify the largest variation in the data. This revealed a differential protein pattern between Dsup+ and Dsup− cells after 72 h of hypoxia, since samples have a different distribution into the 2D plane (Figure 1C, green and red dots, respectively). PROF1, KVRU, MCA3, THIO, PPIA, ALDOA, EFTA, and NH2L1 protein abundance seem to have the strongest role in this sample separation. Enrichment analysis by MetaCore software, by Gene Ontology (GO) Biological Processes (BPs), reveals that up-regulated proteins in Dsup+ after 72 h of CoCl₂ are involved in the negative regulation of stress-activated MAPK cascade, negative regulation of neutrophil aggregation, negative regulation of peroxidase activity, nitric oxide storage, methylglyoxal biosyntetic process, aldheyde biosyntetic process, piruvate metabolic process, and mono carboxylic acid metabolic process, while down-regulated proteins act on protein insertion and organization into outer mitochondrial membrane, chaperone-mediated protein complex assembly, positive regulation of protein polymerization, chaperone-mediated autophagy, organelle organization, positive regulation of ATP-dependent activity and chromosome organization (Figure 1D).

After an ischemic injury, tissue death can be linked to sudden reperfusion of the ischemic area. We simulated this event by replacing the medium containing 100 μM CoCl₂ with complete medium (10% FCS) after 72 h of hypoxia and allowing cells to grow for 24 h (5% CO₂, 37 °C). We then performed an MTT assay, and we observed a significantly increased survival rate in Dsup+ cells compared to Dsup− cells (Figure 2A). In Dsup−, during reperfusion, cell proliferation is balanced by cell death, and cell number returned to that measured after 48 h of hypoxia. On the contrary, in Dsup+ cells, we observed a huge increase in cell proliferation, indicating an acquired capacity of the cells to overcome reperfusion injury.

To understand the modifications at the protein level induced by reperfusion, we performed a proteomic analysis, also comparing Dsup+ and Dsup− cells after the reperfusion event. We found 38 differential spots, 25 down- and 13 up-regulated in Dsup+ cells, of which 31 were identified by mass spectrometry (Figure S1 and Table S1b). Complete proteomic data from Dsup+/− 72 h CoCl₂ and reperfusion have been used to perform PCA and heatmap analysis to have a whole panorama of the proteins and samples' behavior. In Figure 2B, it is reported that the PCA, PC1 vs. PC2, and PC1 vs. PC3 visualizations highlight the differential protein pattern, well distinguishing the four groups of analysis, suggesting a differential response of Dsup+ cells not only to hypoxia but also to reperfusion with respect to Dsup− cells. Moreover, heatmap analysis reported in Figure 2C shows that Dsup has an immediate impact on cell response to hypoxia, upregulating proteins in cluster one of Dsup+ cells with respect to Dsup− cells. On the other hand, during reperfusion, there is a reversal of the situation: cluster one is strongly downregulated in both Dsup+/− conditions. Characteristic protein profiles of Dsup+ cells, during reperfusion, are observed in cluster three, where proteins showed an upregulation, and cluster four, where proteins showed a downregulation with respect to Dsup− cells. Upregulated proteins in Dsup+ cells after reperfusion are involved in the insertion and organization of the mitochondria and mitochondrial membrane proteins, in chaperon-mediated protein complex assembly and autophagy, in regulation of the apoptotic process, and in response to stress (Figure 2D). Instead, the downregulated proteins in Dsup+ cells after reperfusion are involved in organophosphate metabolic processes, regulation of RNA splicing, purine-containing compound metabolic processes, the creatine cycle, ribose phosphate metabolic processes, and phosphagen biosynthetic processes (Figure 2D).

From the moment that reperfusion injury involves the generation of reactive oxygen species (ROS) that is fueled by the reintroduction of oxygen when the blood flow is reestablished, we deeply investigated molecular pathways involved in Dsup protection against H₂O₂ exposure (20). Firstly, we demonstrated that after 72 h of CoCl₂ and 24 h of reperfusion, ROS production is significantly increased (p < 0.01) in the cell media of DSUP− cells, and the GSH/GSSG ratio is strongly decreased (p < 0.001), indicating a rise in oxidative stress (Supplementary Figure S2A,B). Importantly, in Dsup+ cells treated with hypoxia/reperfusion, we observed a reduction in ROS levels similar to that measured in untreated cells (Figure 3A), and the GSH/GSSG ratio was maintained at basal levels (Figure 3B).

As Dsup transfection triggers an increased overall survival of human cells in response to oxidative stress (20), we applied a differential proteomic approach in Dsup− and Dsup+ cells after H₂O₂ treatment to investigate the molecular mechanisms by which Dsup responds to this stress and can counteract ROS-related damage. Proteomic analysis evidenced 60 differentially abundant spots, of which 58 were identified by mass spectrometry, 24 up- and 36 down-regulated, as reported in Figure S3 and Table S2. PCA in Figure 4A shows the distribution of Dsup− and Dsup+ samples along the two principal components, which together represent 97.18% of the variance of the %V of differential spots. According to the PCA, the Dsup+ proteomic profile well differentiates from that of Dsup− cells, evidencing the Dsup impact in modulating the response of human cells to oxidative stress. Particularly, the increased abundance of ubiquitin carboxyl-terminal hydrolase isozyme L3 (UCHL3), thioredoxin (THIO), and 60 S acidic ribosomal protein P0 (RLA0), and the decreased abundance of two protein species of polyubiquitin-B (UBB) have a greater relevance in discriminating Dsup−/+ cells. Based on heatmap analysis (Figure 4B), differential spots similarly behave in terms of abundance within the same condition, leading Dsup− and Dsup+ samples to cluster well into two distinct groups. On the other hand, the heatmap clearly highlights two main clusters of spots (rows) that oppositely behave in Dsup+ cells with respect to the Dsup− cells: one with 36 protein spots that are less abundant in Dsup+ cells and one with 24 protein spots that are highly abundant in Dsup+ cells. Considering these two main clusters of protein spots, an enrichment analysis by GO biological processes (BPs) was performed. The most represented BPs associated with the up-regulated proteins in Dsup+ cells after oxidative stress are protein folding, glycolytic processes, mRNA splicing via the spliceosome, cell-redox homeostasis, regulation of macro-autophagy, and post-translational protein modifications. On the other hand, the most represented BPs associated with the down-regulated proteins in Dsup+ cells after oxidative stress are anaphase-promoting complex dependent catabolic processes, pre-replicative complex assembly, regulation of cellular amino acid metabolic processes, protein refolding, antigen processing and presentation via MHC class I, TAP-dependent regulation of transcription from RNA polymerase II promoter in response to hypoxia, negative regulation of G2/M transition of mitotic cell cycle, ATP metabolic process, regulation of mRNA stability, post-translational protein modification, and proteasome-mediated ubiquitin-dependent protein catabolic processes (Figure 4C).

Mitochondrial activity has a central role in response to hypoxia as well as oxidative stress. Treatment with CoCl₂ highlighted proteins involved in insertion into the mitochondrial membrane, such as voltage-dependent anion channel (VDAC) proteins VDAC1 and VDAC2, which we found dysregulated, by proteomic analysis, after hypoxic stress and after reperfusion (Table S1). VDAC1 and VDAC2 have roles in pore-forming proteins and are known to play an essential role in cellular metabolism and in the early stages of apoptosis. For example, VDAC constitutes a major pathway by which metabolites such as ADP/ATP, succinate, and citrate are exchanged between the cytosol and mitochondria. VDAC1 in the plasma membrane establishes a novel level of apoptosis regulation, putatively via its redox activity. Western blot analysis of VDAC after CoCl₂ and H₂O_2, and reperfusion showed VDAC increased after CoCl₂ and H₂O₂ treatment and after reperfusion in both Dsup+/−, with respect to the basal condition (Figure 5A). Of note, after reperfusion, in Dsup+ cells, there was not a significant increase in VDAC, indicating that in these cells, a lower stress condition is maintained.

Our proteomic data showed dysregulation of PRDX1, PRDX3, and PRDX6, reflecting a regulation of peroxidase activity after hypoxic insult and cell redox homeostasis regulation after oxidative stress induced by H₂O₂. By Western blot analysis, we analyzed the levels of PRDX3, a thiol-specific peroxidase that catalyzes the reduction of hydrogen peroxide and organic hydroperoxides to water and alcohols, playing a role in cell protection against oxidative stress. PRDX3 levels did not change in abundance in Dsup+/− cells in the presence of hypoxia (CoCl₂) with respect to the basal condition. Its different behavior (downregulation) is observed in Dsup+ cells after H₂O₂ treatment. Similarly, PRDX3 tends to be downregulated also in Dsup+ cells after reperfusion (Figure 5A).

Another interesting protein, highlighted by differential proteomic analysis among Dsup+/− cells after 72 h of hypoxic stress, subsequent reperfusion, is profilin 1. This is an actin-binding protein involved in the dynamic transformation and reorganization of actin. Its validation by Western blot showed PROF-1 upregulation in Dsup+ cells at basal conditions and after CoCl₂ treatment for 72 h. On the other hand, PROF-1 levels decrease in Dsup+ cells after H₂O₂ treatment and overall, after reperfusion, suggesting an alternative role in the presence of oxidative stress with respect to hypoxic stress.

Among up-regulated proteins in Dsup+ cells following reperfusion and H₂O₂ treatment, we found those involved in autophagy mechanisms, whereas the same proteins are down-regulated in Dsup+ cells after CoCl₂ treatment. Immunohistochemical analysis clearly shows (Figure 5B) that Dsup presence induces an increase in autophagy both at baseline and following H₂O₂ treatment and reperfusion compared to untransfected cells.

The HEK293TT cell line (mycoplasma-free, verified by N-GARDE Mycoplasma PCR reagent set, Euroclone, Milan, Italy) was kindly donated by Prof. Sandra Donnini (University of Siena). pCXN2KS-Dsup was a gift from Prof. Kunieda Takekazu (Addgene plasmid #90019; https://www.addgene.org/90019/↗ RRID: Addgene_90019 accessed on 20 February 2021). Cells were cultured in Dulbecco's modified Eagle's Medium (DMEM) supplemented with 10% fetal bovine serum (FBS). An empty vector (plasmid without Dsup) was kindly donated by Dr. Jlenia Brunetti (University of Siena). Both constructs were transfected into HEK293TT cells using Lipofectamine^® 2000 Reagent (Life Technologies, Milna, Italy) and selected by 700 µg/mL G418 (Euroclone) for three weeks. Transfection was verified by endpoint PCR as already reported [20].

The MTT metabolic assay (Vybrant^® MTT Cell Proliferation Assay Kit, Molecular Probes, Eugene, OR, USA) was used to quantify cell viability. Both Dsup+ and empty vector cells (Dsup−) were seeded at a 100,000 cells/mL density in a 96-well plate. After 24 h of incubation, to induce hypoxia, cells were treated for 24, 48, and 72 h with 100 μM CoCl₂ [35]. To simulate reperfusion, we allowed cells to grow in normal medium for 24 h after 72 h of treatment with CoCl₂. To induce oxidative stress, cells were treated with 250 µM hydrogen peroxide (H₂O₂) O/N. After each stimulation, cells were incubated at 37 °C for 4 h with the tetrazolium dye MTT (3-(4,5-Dimethylthiazol-2-yl)-2,5-Diphenyltetrazolium Bromide). Solubilization was carried out with DMSO (50 µL) at 37 °C for 10 min. Then, the number of viable cells was determined by measuring absorbance at 540 nm in a microplate reader (Tecan). Each experiment was run in triplicate and was repeated at least three times.

Total RNA was extracted from cell pellets using the SV Total RNA Isolation System (Promega, Milan, Italy) following the manufacturer's instructions. The TaqMan™ Array Human Hypoxia (ThermoFisher Scientific, Milan, Italy) was run on 100 ng of cDNA obtained by retrotranscription of RNA with the iScript cDNA Synthesis kit (Biorad, Milan, Italy). The hypoxia assay contains 44 hypoxia-related assays and 4 assays to candidate endogenous control genes. The assay was plated in duplicate. Quantification was determined by using the 2−ΔCT method.

Three independent cell cultures per condition, as biological replicates, in both cells transfected with Dsup (Dsup+) and cells transfected with an empty vector (Dsup−) were used for proteomic analyses. Three main conditions were analyzed, such as oxidative stress by H₂O₂ stimulation, 72 h-hypoxia by CoCl₂ stimulation, and reperfusion by restoring normal O₂ conditions.

Cell cultures were collected by trypsinization and washed with phosphate-buffered saline (PBS) three times, followed by a centrifugation of 5 min at 1600× g after each wash. Cell pellets were kept at −80 °C until use.

Samples were dissolved in a lysis buffer composed of 7 M urea, 2 M thiourea, 4% w/v 3-[(3-cholamidopropyl) dimethylammonia]-1-propanesulfonate hydrate (CHAPS), and 1% w/v dithioerythritol (DTE). After having determined the concentration of the proteins contained in the samples by Bradford assay, each sample was resolved following the two-dimensional electrophoresis (2DE) technique [36]. Image analysis was then performed by Melanie™ Classic 9 software (SIB Swiss Institute of Bioinformatics, Geneva, Switzerland), which allowed the determination of the percentage of relative volume (%V) of each spot. Based on these values, a statistical analysis was performed to identify statistically significant differential spots by ANOVA test via XLStat (Addinsoft, Paris, France), according to p-value ≤ 0.05 and fold change ≥ 1.8 for H₂O₂ analysis, and a fold change ≥ 2 for hypoxia/reperfusion analysis.

The identification of the differential spots was performed by MALDI-ToF mass spectrometry through peptide mass fingerprinting (PMF). Differential spots were detected in the MS-compatible silver-stained gels and manually excised from them. To avoid possible interferences due to the staining, spots were destained first in a solution of 30 mM potassium ferricyanide and 100 mM anhydrous sodium sulfate, and then in 200 mM ammonium bicarbonate. Then, they went through a dehydration process in 100% acetonitrile (ACN). After the following rehydration, protein spots were digested in a trypsin solution overnight at 37 °C, then placed on a MALDI target and covered with a matrix solution of 5 mg/mL α-cyano-4-hydroxycinnamic acid (CHCA) in 50% v/v ACN and 0.5% v/v trifluoroacetic acid (TFA). The UltrafleXtreme™ MALDI-ToF/ToF mass spectrometer (Bruker Daltoniks, Bremen, Germany), equipped with a 200 Hz smartbeam™ I laser in the positive reflector mode, was used to carry out the mass spectrometry analysis with the following parameters: 170 ns of delay; ion source 1:20 kV; ion source 2:17.70 kV; lens voltage: 7.80 kV; reflector voltage: 20.80 kV; and reflector 2 voltage: 10.95 kV. The applied laser wavelength and frequency were 353 nm and 100 Hz, respectively, and the percentage was set to 50%. MS spectra were obtained considering the averages of 1500 laser shots targeted at five different positions within the spot. The acquisition of mass spectra was carried out by FleXcontrol software v 3.0 (Bruker), and the processing was carried out via the FleXanalysis software version 3.0 (Bruker), considering the peptides obtained from the trypsin autoproteolysis as internal standards for calibration. In the resulting mass lists, common contaminants (such as matrix-related ions, trypsin autolysis, and keratin peaks) were pinpointed and excluded. The identification of the protein spots was performed using the PMF search with MASCOT (Matrix Science Ltd., London, UK, http://www.matrixscience.com↗; accessed on 15 March 2024). The following parameters were set up: Homo sapiens as taxonomy, SwissProt as database, 20 ppm as mass tolerance, one admissible missed cleavage site, carbamidomethylation (iodoacetamide alkylation) of cysteine as a fixed modification, and oxidation of methionine as a variable modification. The following inclusion criteria were considered for the protein identifications: a p-value < 0.02 (referred to as "expect", as reported in Tables S1 and S2), a minimum of three matched peptides, and a minimum MASCOT score of 55.

The mass spectrometry proteomics data have been deposited to the ProteomeXchange Consortium the PRIDE [37] partner repository with the dataset identifier PXD067125.

Twenty-five micrograms of three different biological replicates per condition were suspended in LEAMMLI buffer (Tris HCl pH 6.8, 62.5 mM, 20% v/v glycerol, 2% w/v SDS, 5% v/v β-mercaptoethanol, and 0.032% w/v bromophenol blue), and samples were heated at 95 °C for 7 min. After cooling, samples were loaded in 12% polyacrylamide gels. The Mini-PROTEAN electrophoresis system (Bio-Rad, Heracles, CA, USA) was used to carry out mono-dimensional electrophoresis. After SDS-PAGE, the 1 h equilibration of gels took place in the transfer buffer (192 mM glycine, 25 mM Tris, 20% v/v methanol), and then, gels were transferred to nitrocellulose membranes (Cytiva, Washington, DC, USA). To evaluate if the protein transfer had correctly taken place, membranes were stained with Red Ponceau (0.2% w/v Ponceau S in 3% acetic acid).

Membranes were tested for three different primary antibodies, such as anti-VDAC (rabbit polyclonal, PA1-954, Invitrogen, Waltham, MA, USA, WB dilution 2 μg/mL), anti-PROF1 (rabbit polyclonal, P7624, Sigma-Aldrich, St. Louis, MO, USA, WB dilution 2 μg/mL), and anti-PRDX3 (mouse monoclonal, sc-59663, Santa Cruz Biotechnology, Dallas, TX, USA, dilution 1:200). Secondary antibody incubation varies according to primary antibody composition; in detail, secondary peroxidase-conjugated goat anti-rabbit antibody was used at 1:7000 dilution and secondary peroxidase-conjugated goat anti-mouse antibody at 1:3000 dilution. Western blot normalization was performed on total β-actin (anti-β-actin mouse monoclonal, A5441, Sigma, NY, USA, dilution: 1:40,000). Immunoreaction was detected by a chemiluminescence detection system (Cytiva), and images were analyzed with ImageJ v 1.x (1.51p, NIH, Chicago, IL, USA) to obtain band intensity. Statistical analysis was performed by XLSTAT (Addinsoft).

To assess autophagy, 5 × 104 cells were plated in 24-well plates and allowed to grow for 24 h. Cells were then stimulated O/N with 250 µM H₂O₂ or with the hypoxia/reperfusion protocol (72 h of 100 μM CoCl₂ and 24 h of normal medium). Autophagy was assessed using the Autophagy Assay Kit (Sigma-Aldrich, NY, USA) following the kit instructions. Briefly, the medium was removed, and 100 µL of the Autophagosome Detection Reagent was added to each well. Cells were incubated at 37 °C with 5% CO₂ for 1 h. After incubation, cells were washed 4 times with the wash buffer by gently adding 100 μL of wash buffer to each well. Fluorescence intensity (lex = 360/lem = 520 nm) was measured using a fluorescence microscope (Nikon Eclipse Ts2). Autophagy was indicated by bright blue dot staining of autophagic vacuoles.

The ROS-Glo™ H₂O₂ Assay (Promega) was used to measure the level of H₂O₂ directly in the cell culture. Briefly, 1.5 × 103 cells were plated in 96-well plates and treated with the hypoxia/reperfusion protocol (72 h of 100 μM CoCl₂ and 24 h of normal medium). According to the kit instructions, 20 µL of H₂O₂ Substrate Solution was incubated at 37 °C for the last 6 h. This reacts directly with H₂O₂ to generate a luciferin precursor. Then, 100 µL of the ROS-Glo™ Detection Solution was added for 20 min to convert the precursor to luciferin. The light signal is proportional to the level of H₂O₂ present in each sample.

The GSH/GSSG-Glo™ Assay (Promega) was used to measure the ratio of GSH/GSSG. Briefly, 5 × 103 cells were plated in 96-well plates and treated with the hypoxia/reperfusion protocol. At the end of reperfusion, we removed the medium and performed a wash with PBS to remove any residue of cell culture media that may interfere with the assay (as reported in the protocol). Then, according to the kit instruction, 50 µL of Total Glutathione Lysis Reagent (for GSH + GSSG wells) or Oxidized Glutathione Lysis Reagent (for GSSG wells) was added. Total Glutathione Lysis Reagent lyses cells and reduces GSSG to GSH to measure total glutathione. Oxidized Glutathione Lysis Reagent lyses cells and blocks the GSH present in the sample to produce a form that cannot contribute to the luminescent signal, then it converts GSSG into new GSH. In both formulations, the luciferin-NT is converted to luciferin in a GSH-dependent reaction, which is coupled to a firefly luciferase reaction. The amount of light emitted by luciferase depends on the amount of luciferin formed, which in turn depends on the amount of GSH present. The plate was placed on a plate shaker for 5 min, and 50 µL of Luciferin Generation Reagent was added to each well for 30 min (room temperature). Finally, 100 µL of Luciferin Detection Reagent was added to each well to block the luciferin generation reaction and initiate a luminescent signal.

GraphPad Prism software version 5 was used for statistical analysis. For qPCR analysis, at least three separate replicates were performed. Sample differences were assessed by a one-way ANOVA test. Survival data were analyzed using a paired test for non-parametric data (Wilcoxon signed-rank test). Principal Component Analysis (PCA) and heatmap analysis were performed by XLSTAT for both H₂O₂ and hypoxia/reperfusion proteomic analyses, comparing Dsup− and Dsup+ cells, considering the %V of statistically significant differential protein spots. PCA was carried out Pearson's correlation and heatmap analysis using the Euclidean distance.

Enrichment analyses were carried out using the DAVID Bioinformatics platform and MetaCore software (https://portal.genego.com/↗, accessed in 7 April 2024) (Clarivate Analytics, Boston, MA, USA). Enrichment by Gene Ontology (GO) was performed for cellular compartments and biological processes, as well as enrichment by pathway maps and process networks.

Results from Western blot were analyzed by the Kruskal–Wallis test with the multiple comparison Conover-Iman. For all comparisons, a p value of <0.05 was considered significant (* < 0.05; ** < 0.01; *** < 0.001).