Molecular Mechanisms of the Cytotoxic Effect of Recombinant Selenoprotein SELENOM on Human Glioblastoma Cells

Currently, selenobiology is an actively developing area, primarily due to the study of the role of the trace element selenium and its organic and inorganic compounds in the regulation of vital processes occurring in the cell. In particular, the study of the functions of selenium nanoparticles has gained great popularity in recent years. However, a weak point in this area of biology is the study of the functions of selenoproteins, of which 25 have been identified in mammals to date. First of all, this is due to the difficulties in obtaining native forms of selenoproteins in preparative quantities, due to the fact that the amino acid selenocysteine is encoded by one of the three stop codons of the TGA universal genetic code. A complex system for recognizing a given codon as a selenocysteine codon has a number of features in pro- and eukaryotes. The selenoprotein SELENOM is one of the least studied mammalian selenoproteins. In this work, for the first time, studies of the molecular mechanisms of regulation of the cytotoxic effect of this protein on human glioblastoma cells were carried out. The cytotoxicity of cancer cells in our experiments was already observed when cells were exposed to 50 μg of SELENOM and increased in proportion to the increase in protein concentration. Apoptosis of human glioblastoma cells was accompanied by an increase in mRNA expression of a number of pro-apoptotic genes, an increase in endoplasmic reticulum stress, and activation of the UPR IRE1α signaling pathway. The results obtained also demonstrate a dose-dependent depletion of the Ca2+ pool under the action of SELENOM, which proves the important role of this protein in the regulation of calcium homeostasis in the cell.


Introduction
Selenium (Se) is a biologically active trace element that is part of a number of important enzymes in the form of the amino acid selenocysteine, and is thus associated with all organs and systems. To date, 25 human selenium proteins have been identified, while murine and rat selenoproteomes are represented by 24 proteins. Most of them can be classified into two groups based on the localization of the selenocysteine residues [1]. The first group includes selenoproteins characterized by the presence of selenocysteine in the C-terminal sequence of these proteins. Thus, in mammalian thioredoxin reductases containing the GCUG motif, selenocysteine is localized in the mobile C-terminal position [2], which allows the catalytic center to transfer reduced equivalents from the hard-to-reach disulfide active site to the active site of the substrate protein.
The second selenoprotein group includes proteins containing selenocysteine at the N-terminal position of the short domains. This is a large group of α/β proteins in which, according to their secondary structure, the selenocysteine-containing catalytic center is located between the α-helix and the β-sheet, similar to the arrangement of the CXXC motif (sequence in which two cysteines are separated by two other amino acids) of oxidoreductases, involved in thiol-disulfide exchange. Selenocysteine in the redox center can replace both N-and C-terminal cysteine with the formation of CXXU/UXXC motifs [3]. This group When studying the functions of any protein, it is necessary to obtain it in preparative quantities, which is often a difficult task. This is especially true in cases where it is necessary to express the gene of a eukaryotic protein in a bacterial system. In the case of selenoproteins, this task is complicated by the fact that the amino acid selenocysteine (Sec), which is part of them, is encoded by one of the three stop codons of translation. In order for it to be recognized as selenocysteine and not as a premature termination of translation, as a rule, many authors work with cysteine homologues of selenoproteins. In rare cases, it is possible to choose conditions for the synthesis of native forms of selenoproteins in preparative amounts, which is explained by the peculiarities of recognition of the selenocysteine codon and differences in cis-and trans-active factors in bacterial and eukaryotic systems [14,15].
Despite the fact that earlier we were able to select the conditions for the production of the cysteine homologue hSELENOM, in the framework of this work we further improved them, which allowed us to obtain an even higher yield of purified protein (up to 10 mg/mL), taking into account the fact that this protein tends to form inclusion bodies. Figure 1A shows the results of the selection of optimal conditions for achieving high expression of the open reading frame (ORF) of the gene encoding the cysteine homologue of hSELENOM. Based on the above data, it can be concluded that the maximum level of expression of the ORF gene encoding the human cysteine homologue SELENOM (TGA-> TGT) in the pET 23b (+) plasmid was observed when bacterial cells were treated with 1 mM IPTG for 3 h at 25 • C and 1 mM IPTG for 1 h at 37 • C. However, when cells were incubated with 0.5 mM IPTG for 3 h at 25 • C, only approximately 20-30% of the total protein precipitated, with most remaining in the soluble phase. Therefore, in our case, the optimal conditions for the synthesis of recombinant hSELENOM in bacterial cells of the Rosetta (DE3)pLysS strain (Novagen, USA) were achieved by inducing the lactose operon with 0.5 mM IPTG for 3 h at 25 • C. Figure 1B shows the results of PAGE electrophoresis visualizing the various stages of protein purification under study.

Optimal Conditions for the Synthesis of Human SELENOM in a Bacterial System Were
Observed When Bacterial Cells Were Incubated with 0. 5

MM IPTG for 3 h at 25 °C
When studying the functions of any protein, it is necessary to obtain it in preparat quantities, which is often a difficult task. This is especially true in cases where it is n essary to express the gene of a eukaryotic protein in a bacterial system. In the case of lenoproteins, this task is complicated by the fact that the amino acid selenocysteine (Se which is part of them, is encoded by one of the three stop codons of translation. In ord for it to be recognized as selenocysteine and not as a premature termination of trans tion, as a rule, many authors work with cysteine homologues of selenoproteins. In ra cases, it is possible to choose conditions for the synthesis of native forms of selenop teins in preparative amounts, which is explained by the peculiarities of recognition of t selenocysteine codon and differences in cis-and trans-active factors in bacterial and e karyotic systems [14,15]. Despite the fact that earlier we were able to select the conditions for the producti of the cysteine homologue hSELENOM, in the framework of this work we further i proved them, which allowed us to obtain an even higher yield of purified protein (up 10 mg/mL), taking into account the fact that this protein tends to form inclusion bodies Figure 1A shows the results of the selection of optimal conditions for achieving hi expression of the open reading frame (ORF) of the gene encoding the cysteine homolog of hSELENOM. Based on the above data, it can be concluded that the maximum level expression of the ORF gene encoding the human cysteine homologue SELENOM (TG > TGT) in the pET 23b (+) plasmid was observed when bacterial cells were treated wit mM IPTG for 3 h at 25 °C and 1 mM IPTG for 1 h at 37°C. However, when cells were cubated with 0.5 mM IPTG for 3 h at 25 °C, only approximately 20-30% of the total p tein precipitated, with most remaining in the soluble phase. Therefore, in our case, t optimal conditions for the synthesis of recombinant hSELENOM in bacterial cells of t Rosetta (DE3)pLysS strain (Novagen, USA) were achieved by inducing the lactose o eron with 0.5 mM IPTG for 3 h at 25 °C. Figure 1B shows the results of PAGE elect phoresis visualizing the various stages of protein purification under study.

Recombinant hSELENOM Reduced Proliferative Activity and Induced Apoptosis in Human Glioblastoma Cells
On Figure 2A shows the results of MTT analysis of the proliferative activity of human glioblastoma cells (A-172 lines) after a 24 h incubation in a nutrient medium supplemented with various amounts (50, 100, and 200 µg) of hSELENOM. The seeding density of cells in the experimental groups was the same, and hSELENOM and buffer application were made 2 days after cell seeding. At this time, the cells are in the active stage of proliferation. According to the presented results, it can be concluded that recombinant hSELENOM was able to reduce the proliferative activity of A-172 cancer cells in direct proportion to the increase in the concentrations of this protein in the nutrient medium. As a control, cells were incubated under the same conditions, but with the addition of protein elution buffer. According to Figure 2A, it can be concluded that a significant decrease (by approximately 40%) in the proliferative activity of glioblastoma cells was observed after exposure of A-172 cells to 100 µg of protein, while a doubling of the protein concentration contributed to a decrease in the proliferative activity of cells by approximately 70-75%.

Recombinant hSELENOM Reduced Proliferative Activity and Induced Apoptosis in Human Glioblastoma Cells
On Figure 2A shows the results of MTT analysis of the proliferative activity of human glioblastoma cells (A-172 lines) after a 24 h incubation in a nutrient medium supplemented with various amounts (50, 100, and 200 μg) of hSELENOM. The seeding density of cells in the experimental groups was the same, and hSELENOM and buffer application were made 2 days after cell seeding. At this time, the cells are in the active stage of proliferation. According to the presented results, it can be concluded that recombinant hSELENOM was able to reduce the proliferative activity of A-172 cancer cells in direct proportion to the increase in the concentrations of this protein in the nutrient medium. As a control, cells were incubated under the same conditions, but with the addition of protein elution buffer. According to Figure 2A, it can be concluded that a significant decrease (by approximately 40%) in the proliferative activity of glioblastoma cells was observed after exposure of A-172 cells to 100 μg of protein, while a doubling of the protein concentration contributed to a decrease in the proliferative activity of cells by approximately 70-75%.   Figure 3A,B show the results of the analysis of the presence of signs of apoptotic death of glioblastoma cells. According to the data obtained, it can be concluded that hSELENOM was able to induce apoptosis in glioblastoma cells. The number of apoptotic cells increased with increasing protein concentration in the nutrient medium. According to the results presented in Figure 3B, we can confidently say that almost 70% of the cells were in a state of apoptotic death when A-172 cells are treated with 100 µg hSELENOM. Doubling the amount of protein enhanced the cytotoxic effect and the number of apoptotic cells increased up to almost 90% compared with the control. to the results presented in Figure 3B, we can confidently say that almost 70% of the cells were in a state of apoptotic death when A-172 cells are treated with 100 μg hSELENOM. Doubling the amount of protein enhanced the cytotoxic effect and the number of apoptotic cells increased up to almost 90% compared with the control.

Recombinant hSELENOM Induced Increased mRNA Expression of Pro-apoptotic Genes and the IRE1α Signaling Pathway UPR
The above results were confirmed by the analysis of mRNA expression patterns for a number of pro-apoptotic genes. Thus, 24 h exposure of human glioblastoma cells to various amounts of hSELENOM led to the increased expression of a number of pro-apoptotic genes by two or more times ( Figure 4A). Interestingly, lower amounts of this protein (50 and 100 μg) contributed to the greatest increase in the expression of these genes, while 200 μg of the protein did not significantly affect their expression.
Increased mRNA expression of genes such as CHOP, GADD34, PUMA, and CAS-4 may also indicate the activation of endoplasmic reticulum stress (ER stress) in these cells. In addition, when analyzing the mRNA expression patterns shown in Figure 4A, it is

Recombinant hSELENOM Induced Increased mRNA Expression of Pro-Apoptotic Genes and the IRE1α Signaling Pathway UPR
The above results were confirmed by the analysis of mRNA expression patterns for a number of pro-apoptotic genes. Thus, 24 h exposure of human glioblastoma cells to various amounts of hSELENOM led to the increased expression of a number of pro-apoptotic genes by two or more times ( Figure 4A). Interestingly, lower amounts of this protein (50 and 100 µg) contributed to the greatest increase in the expression of these genes, while 200 µg of the protein did not significantly affect their expression.
Increased mRNA expression of genes such as CHOP, GADD34, PUMA, and CAS-4 may also indicate the activation of endoplasmic reticulum stress (ER stress) in these cells. In addition, when analyzing the mRNA expression patterns shown in Figure 4A, it is easy to see that 50 and 100 µg hSELENOM caused an increase in the expression of the anti-apoptotic BCL-2 gene, which was not observed when cells were exposed to 200 µg hSELENOM.
Direct evidence of increased ER stress was also provided by the fact that this protein promoted an increase in the expression of a marker of one of the three signaling pathways of the cell's UPR response to the growth of misfolded proteins. Therefore, according to the results shown in Figure 4B, there was an increase in the expression of the spliced form of the XBP1s transcription factor. This may indicate the activation of the IRE1α signaling pathway UPR. These data were confirmed by Western blotting, the results of which are shown in Figure 5A,B. easy to see that 50 and 100 μg hSELENOM caused an increase in the expression of the anti-apoptotic BCL-2 gene, which was not observed when cells were exposed to 200 μg hSELENOM. Direct evidence of increased ER stress was also provided by the fact that this protein promoted an increase in the expression of a marker of one of the three signaling pathways of the cell's UPR response to the growth of misfolded proteins. Therefore, according to the results shown in Figure 4B, there was an increase in the expression of the spliced form of the XBP1s transcription factor. This may indicate the activation of the IRE1α signaling pathway UPR. These data were confirmed by Western blotting, the results of which are shown in Figure 5A,B. In addition, we analyzed how the expression of seven ER-resident selenoproteins changed, since their role in the regulation of apoptosis, ER stress, post-translational modification of proteins in the ER, and calcium homeostasis has been repeatedly proven [16][17][18][19]. According to the real-time PCR results shown in Figure 4C, we can confidently speak about a significant increase in the mRNA of the endogenous selenoprotein SE-LENOM by more than 3-5 times compared with the control when treating human glioblastoma cells with 50, 100, and 200 μg of recombinant hSELENOM. At the same time, we did not observe an increase in mRNA expression of the genes of the other six ER proteins. In addition, we analyzed how the expression of seven ER-resident selenoproteins changed, since their role in the regulation of apoptosis, ER stress, post-translational modification of proteins in the ER, and calcium homeostasis has been repeatedly proven [16][17][18][19]. According to the real-time PCR results shown in Figure 4C, we can confidently speak about a significant increase in the mRNA of the endogenous selenoprotein SELENOM by more than 3-5 times compared with the control when treating human glioblastoma cells with 50, 100, and 200 µg of recombinant hSELENOM. At the same time, we did not observe an increase in mRNA expression of the genes of the other six ER proteins.

Recombinant hSELENOM Depleted Endoplasmic Reticulum Ca 2+ Pools
To study the effects of exogenous hSELENOM on the Ca 2+ signaling system of cells, the cultures were loaded with the calcium-sensitive Fura-2AM probe and, using fluorescence microscopy, changes in the baseline and thapsigargin-induced changes in the levels of cytosolic calcium ( Figure 7B) also did not cause Ca 2+ signaling in the cells. Ca 2+ responses to ATP application were, on average, similar in all three exper-  Figure 7B) also did not cause Ca 2+ signaling in the cells. Ca 2+ responses to ATP application were, on average, similar in all three experimental groups and close to control cells ( Figure 7C), which were washed out with the working medium (HBSS). level of [Ca 2+ ]i of A-172 human glioblastoma cells after application of 10 μM thapsigargin (TG) in a calcium-free medium supplemented with 0.5 mM EGTA, incubated with various amounts of hSELENOM or an equivalent volume of diluent buffer. Cellular Ca 2+ signals averaged over 100 cells for each curve in one experiment are shown. (C) Averaged values of baseline [Ca 2+ ]i and [Ca 2+ ]i levels after addition of TG (10 μM) in calcium-free medium and incubated for 24 h with different concentrations of hSELENOM. Data shown are the mean value ± SE of the [Ca 2+ ]i levels obtained from several hundred cells. Number of passages = 3; n/s-data not significant (p > 0.05), ** p < 0.01, ***p < 0.001.
The results presented above showed the dose-dependent effect of hSELENOM on the Ca 2+ signaling system of cancer cells, which was observed during a long 24 h incubation of the cells with the protein. Application of SELENOM 200 μg to A-172 cells in an acute experiment did not cause cellular Ca 2+ responses ( Figure 7A), including no increase in the baseline level of [Ca 2+ ]i during the 30 min registration of Fura-2 fluorescence. In the control sample, the application of buffer ( Figure 7B) also did not cause Ca 2+ signaling in the cells. Ca 2+ responses to ATP application were, on average, similar in all three experimental groups and close to control cells ( Figure 7С), which were washed out with the working medium (HBSS).

Activation of the Mechanism of Actin-Dependent Endocytosis under the Action of Recombinant hSELENOM on Human Glioblastoma Cells
Since the effects of recombinant hSELENOM on A-172 cancer cells were realized within a day, it can be assumed that the protein needs to penetrate into the cells, which takes time and probably activates the mechanisms of endocytosis. To test this hypothesis, cells were exposed to actin-dependent endocytosis inhibitor Cytochalasin D (Cyt D, 10 µM) for 2 h, after which 200 µg hSELENOM was added for 24 h (with Cytochalasin D still in the culture medium). It is known that Cytochalasin D blocks or largely suppresses all actin-dependent pathways of endocytosis, depending on the duration of incubation (from 2 to 24 h) [20].
After the addition of hSELENOM to the medium, a decrease in the baseline level of Fura-2 fluorescence by 50% was observed, and the TG-induced signal was reduced by 70%, relative to the control ( Figure 8B). Pre-incubation of A-172 cells with Cytochalasin D for 24 h did not significantly affect the baseline level of [Ca 2+ ] i ( Figure 8A,B), but caused a 30% decrease in the Ca 2+ signal of cells after the addition of TG, relative to the control ( Figure 8B). At the same time, the use of hSELENOM in the background of Cytochalasin D led to the similar levels Fura-2 fluorescence as cells incubated with Cytochalasin D without hSELENOM. The TG-induced increase in [Ca 2+ ] i, on average, also did not change significantly, which indicates the activation of the mechanisms of actin-dependent endocytosis under the action of recombinant hSELENOM in human glioblastoma cells.
for 24 h did not significantly affect the baseline level of [Ca 2+ ]i ( Figure 8A,B), but caused a 30% decrease in the Ca 2+ signal of cells after the addition of TG, relative to the control ( Figure 8B). At the same time, the use of hSELENOM in the background of Cytochalasin D led to the similar levels Fura-2 fluorescence as cells incubated with Cytochalasin D without hSELENOM. The TG-induced increase in [Ca 2+ ]i, on average, also did not change significantly, which indicates the activation of the mechanisms of actin-dependent endocytosis under the action of recombinant hSELENOM in human glioblastoma cells.
It is known that SELENOM is expressed in many tissues, but to a greater extent in the brain [4], is sensitive to selenium deficiency in the brain and can serve as a molecular biomarker of selenium status in this organ, and is involved in the regulation of human neurogenerative diseases [6]. Similar results were shown in HT22 hippocampal and C8-D1A cerebellar cells: stable overexpression of SELENOM prevented hydrogen peroxideinduced oxidative damage to cells [7]. According to the available data, it becomes clear that SELENOM is important for normal brain development, and its expression in various brain cancer cells is quite high. Previously, we conducted studies on this cell line under hSELENOM-KD conditions; therefore, the purpose of this work was to investigate the role of exogenous hSELENOM on the same line of human cancer cells and to conduct a comparative analysis of the regulation mechanisms of key processes occurring in cancer cells and controlling their cytotoxicity.
In this work, it was shown that the incubation of A-172 cells in a nutrient medium containing recombinant hSELENOM for 24 h led to a decrease in the proliferative properties of these cells and a decrease in their viability (Figures 2 and 3). These indicators tended to decrease after exposure to 50 µg of protein, while a significant cytotoxic effect was observed when cells were exposed to 100 and 200 µg of the studied protein, with a decrease in the proliferative properties of cells by 40 and 70-75%, respectively. We performed a similar series of experiments using human embryonic kidney (HEK 293) cells as an example ( Figure 2B), but no significant changes in the proliferative activity of these cells after exposure to hSELENOM at concentrations of 50 and 100 µg were observed. A marked decrease (approximately 20-25%) in the viability of HEK 293 cells was observed only after they were treated with a high concentration (200 µg) of the protein under study.
This selectivity of recombinant hSELENOM towards cancer cells is a very important property and can be explained by the fact that the microtumor environment (in the nutrient medium in this case) most likely contains a large amount of reduced thiols. On the other hand, many cancer cells are known to have elevated levels of reduced glutathione [41]. hSELENOM penetrates into cells, most likely via actin-dependent endocytosis as shown in this work (Figure 8), and interacts with reduced glutathione due to the presence of the CXCS active center (where C is cysteine and X is any amino acid). Thus, the ratio of oxidized/reduced glutathione is either 1:1 or shifts towards the oxidized form and is 3:1. These changes in the redox status can significantly affect the correct folding of proteins in the lumen of the endoplasmic reticulum, which leads to the accumulation of misfolded proteins and is the cause of ER stress activation. In the case of prolonged ER stress, the cell cannot cope and goes into apoptosis, which was demonstrated in this work. Interestingly, with a decrease in the activity of this selenoprotein in the same A-172 cell line, no significant changes were observed in the proliferative properties and viability of these cancer cells [13].
Furthermore, here we have shown (Figures 4 and 5) that treatment of cancer cells with hSELENOM resulted in the increased expression of a key marker of one of the three known UPR (unfolding protein response) signaling pathways-IRE1α. It is known that during the growth of proteins with incorrect folding in the lumen of the ER, the cell responds to this situation by activating UPR signaling pathways. In this case, we observed that hSELENOM caused an increase in the expression of the spliced form of the XBP1s transcription factor, which may be due to the activation of both IRE1α and ATF-6 UPR signaling pathways, since the active form of the transcription factor ATF-6 can bind to the stress-sensitive element CCAAT (N)9CCACG in the XBP1 sequence and enhance its expression [42,43]. This transcription factor controls the proliferation of ER structures, protein maturation, and its own release from the ER. XBP1 also promotes the removal of proteins with irreversibly disrupted conformations from the ER [44]. However, real-time PCR did not reveal any significant change in ATF-6 mRNA expression; therefore, the version of activation of the IRE1α-signaling UPR pathway after 24 h treatment of A-172 cells with recombinant hSELENOM seems to be the most convincing mechanism. The activation of ER stress can be supported by the data on increased expression of genes such as CHOP, GADD34, PUMA, and CAS-4 (Figures 3 and 4). CHOP (CCAAT/enhancer-binding protein-homologous protein) is usually activated during ER stress [45][46][47][48]. Activated CHOP induces a number of genes encoding proteins involved in apoptosis, such as GADD34 (growth retardation protein induced by DNA damage) and PUMA (p53-regulated modulator of apoptosis), which is activated by p53 during ER stress.
Enhancement of mRNA expression of the gene encoding caspase-4 is known to be possible in two ways. First, it was found that TRAF2 (a factor associated with the TNF receptor) interacts with pro-caspase-12 under ER stress [49], which activates caspase-9, which forms apoptosomes that are necessary for the activation of effector caspase-3 [50][51][52]. Caspase-12 has been identified in rodents, while caspase-4 presumably plays its role in humans [53]. Secondly, the activation of caspase-12 under conditions of ER stress seems to be mediated by the Ca 2+ signaling pathway. When ER stress occurs, conformational changes and/or oligomerization of pro-apoptotic Bax and Bak proteins (proteins belonging to the Bcl-2 family) occur on the ER membrane, which leads to damage to calcium depots in the ER and the release of Ca 2+ into the cytosol [54,55]. An increased flow of Ca 2+ activates m-calpain, which is a member of the family of Ca 2+ -dependent cysteine proteases [51]. Calpain, in turn, cleaves pro-caspase-12 to caspase-12, which leads to the activation of apoptosis [56][57][58].
The second scenario is supported by the data obtained by us in the course of this work showing that recombinant hSELENOM can dose-dependently deplete the Ca 2+ pool of the ER. Our experiments showed that the application of hSELENOM in an acute experiment did not cause an increase in Ca 2+ ions in the cytosol of A-172 cells (Figure 7). We observed the effect of Ca 2+ depletion of the endoplasmic reticulum pool 24 h after incubation of A-172 cancer cells with hSELENOM. It can be assumed that endocytosis of hSELENOM into cancer cells led to the leakage of Ca 2+ ions from the ER and a gradual increase in [Ca 2+ ] i . This increase in Ca 2+ ions in the cytosol leads to the activation of m-calpain which in turn leads to the activation of CAS-4 and the induction of apoptosis. It is known that leakage of Ca 2+ ions from the ER pool can activate apoptosis [59,60]. This cascade of events can lead to disruption of Ca 2+ homeostasis of the cells, and a reduced baseline level of [Ca 2+ ] i after treatment of glioblastoma A-172 cells with hSELENOM.
Interestingly, the treatment of human glioblastoma cells with recombinant hSELENOM led to an increase in SELENOM mRNA expression. It is unlikely that the recombinant protein may be involved in the regulation of the mRNA expression of the native protein.
Most likely, the increasing oxidative and ER stress in cancer cells caused by the action of recombinant hSELENOM activates the expression of the native protein gene, which was repeatedly shown in our early works [16][17][18].
On the other hand, when the same cancer cell line was treated with hSELENOM-KD, an increase in the expression of the CHOP, GADD34, and PUMA genes was also observed, but no increase in the expression of the CAS-4 mRNA and key UPR markers was observed [13]. This may indicate that the activation of ER stress is caused directly by overexpression of either the native hSELENOM protein or the recombinant hSELENOM.
Thus, in this work, it was reliably shown that incubation of human glioblastoma cells with the recombinant cysteine homologue of the human selenoprotein hSELENOM leads to a decrease in the proliferative activity of these cancer cells and activation of the IRE1α signaling pathway UPR. Activation of apoptosis was also observed, most likely through disturbance of calcium homeostasis caused by depletion of the ER depot, activation of m-calpain, and subsequent activation of CAS-4.

Cell Culture and Bacterial Strains
The human glioblastoma (A-172) and human embryonic kidney (HEK 293) cell lines used in this study were purchased from ATCC, Manassas, VA, USA. The cells were grown in DMEM nutrient medium supplemented with 10% serum and antibiotics.

Isolation of RNA and Reverse Transcription
Isolation of total RNA from cells was performed using the RNA extract reagent (Evrogen, Moscow, Russia) according to the manufacturer's protocol. To prevent contamination of genomic DNA, RNA samples were treated with DNase I at 37 • C for 1 h, after which the enzyme was inactivated by adding 50 mM EDTA and heating to 60 • C for 10 min. The amount of total RNA used in the reverse transcription reaction averaged 1 µg. The reverse transcription reaction was carried out according to the protocol and using a kit for the synthesis of the first strand cDNA (Evrogen, Moscow, Russia). The reaction was carried out in the presence of oligo(dT) primers.

Cloning and Site-Directed Mutagenesis
To insert the open reading frame (ORF) of the gene encoding human SELENOM (hSELENOM) (NM_080430.3) into the pET23b(+) plasmid (Novagene, Sacramento, CA, USA) to express hSELENOM protein in a bacterial system, a routine cloning method was performed using restriction and ligation enzymes (Thermo Fisher Scientific, USA). Preliminarily, total RNA was isolated from human embryonic kidney (HEK 293) cell line and used in a reverse transcription reaction. RNA isolation and reverse transcription were performed as described above. Restriction was carried out using EcoR I and SalI restriction enzymes for 3 h at 37 • C, after which the cut PCR products were purified from restriction enzymes according to the Evrogen protocol. Next, the ligation reaction was performed using T4 DNA ligase (Thermo Fisher Scientific, Waltham, MA, USA) for 2 h at room temperature. Ligation was carried out at 4 • C for 16 h or 3 h at room temperature, and then the ligation mixture was used to transform the bacterial strain Top 10. Site-directed mutagenesis was performed by PCR using overlapping primers, the sequences of which are shown in Table 1.

Isolation and Purification of Recombinant hSELENOM
Purification of the recombinant hSELENOM was carried out using affinity chromatography on nickel agarose (Thermo Fisher Scientific, USA).The procedure for isolating and purifying the recombinant protein was carried out under native conditions, in which the cells were resuspended in buffer (50 mM NaH 2 PO 4 , 300 mM NaCl, 10 mM imidazole, and 1 mM PMSF, pH 8.0). The cells were lysed using an ultrasonic disintegrator at a frequency of 22 kHz for 10 min at 4 • C, and then centrifuged at 10,000 rpm for 20 min. After that, the amount of protein in the soluble fraction and sediment was estimated by PAGE electrophoresis. The soluble protein fraction was filtered through a 0.45 µM filter to remove cell debris and applied to Poly-Prep Chromatography Columns (Bio-Rad, Hercules, CA, USA) containing nickel agarose. Protein binding to nickel agarose was carried out at 4 • C for 3 h, after which it was thoroughly washed with a buffer (50 mM NaH 2 PO 4 , 300 mM NaCl, 20 mM imidazole, pH 8.0); then the protein was eluted from the column by incubation with the same buffer, but the content of imidazole in it was 150-250 mM. The recombinant proteins were analyzed by PAGE electrophoresis using 10 or 12% polyacrylamide resolving gels. For further work with the purified recombinant protein, stepwise dialysis was performed to reduce the concentration of imidazole in the elution buffer. If the protein was eluted with the addition of large volumes of buffer, then it was further concentrated using special membranes for an Amicon Ultra 4-50 kDa centrifuge filter (Merk Millipore, Darmstadt, Germany).

Real-Time PCR
The real-time PCR reaction was performed using the qPCRmix-HS SYBR mixture (Evrogen, Moscow, Russia) according to the manufacturer's protocol. The removal of fluorescence occurred at the stage of elongation; the melting curve was measured from 60 to 95 • C with a step of 0.3 • C. The results were expressed as 2 −∆(∆Ct) , where ∆(∆Ct) is the difference between the ∆Ct values between the gene under study and a housekeeping gene. The sequences of all primers used in the real-time PCRs are presented in Table 2. The specificity of the synthesized PCR products was checked by fragment electrophoresis in 2-3% agarose gels.

Registration of Changes in Cytosolic Ca 2+
Registration of the dynamics of the cytosolic Ca 2+ levels in cells was carried out using the fluorescent calcium probe Fura-2AM. The cells were loaded with a probe that was diluted with Hanks' balanced salt solution (HBSS) (156 mM NaCl, 3 mM KCl, 2 mM MgSO 4 , 1.25 mM KH 2 PO 4 , 2 mM CaCl 2 , 10 mM glucose and 10 mM HEPES, pH 7.4) at 37 • C for 40 min. Further washing was carried out for 15 min. The changes in [Ca 2+ ] i were recorded using an Axiovert 200M inverted motorized fluorescence microscope (Carl Zeiss Microscopy GmbH, Jena, Germany). The reagents were added and washed with a perfusion system providing a perfusion rate of 15 mL/min. The Fura-2 was excited and recorded using an HCX PL APO 20.0 × 0.70 IMM UV lens with a refractive index of 1.52. The resulting time-lapse series of images were processed using ImageJ software (Java 1.6.0_12, RRID: SCR_003070, LOCI, University of Wisconsin, Madison, WI, USA). Background fluorescence was subtracted frame by frame using ImageJ's Math Subtract plugin. All experiments were carried out at a temperature of 28-30 • C.

MTT Analysis and Apoptosis Detection Assay
MTT analysis was performed using the MTT kit (Abcam, UK) according to the manufacturer's protocol. The optical density of the solutions in each test well was carried out at OD = 570 nm using a tablet reader. This procedure was repeated three times An apoptosis/necrosis reagent kit (#ab176749, Abcam) was used to detect healthy cells and cells with signs of apoptosis. To detect cells with signs of apoptotic death, the Apopxin Green Indicator was used and they were visualized using a fluorescent microscope using the FITC channel (Ex/Em = 490/525 nm). CytoCalcein 450 was used to detect healthy cells and the cells were visualized with the violet channel (Ex/Em = 405/450 nm).

Conclusions
In the work, a recombinant hSELENOM, in which the TGA codon was replaced by TGT, in preparative amounts was obtained. It has been shown that incubation of human glioblastoma cells (A-172 line) with 50, 100, and 200 µg of hSELENOM for 24 h contributed to a dose-dependent decrease in the proliferative properties of these cells, while the cytotoxic effect of this protein was practically absent when non-cancer human embryonic kidney cells (HEK 293) were treated. Recombinant hSELENOM penetrated into A-172 cells by actin-dependent endocytosis, inducing apoptosis of the cells, causing an increase in the expression of mRNA of pro-apoptotic genes and the UPR IRE1α signaling pathway, and also depleting the ER pool of Ca 2+ .