DNA damage and antioxidant properties of CORM-2 in normal and cancer cells

In this study, we compared the effect of tricarbonyldichlororuthenium (II) dimer (CORM-2) and its CO-depleted molecule (iCORM-2) on human peripheral blood mononuclear cells (PBMCs) and human promyelocytic leukemia HL-60 cells. We determined cell viability, DNA damage and DNA repair kinetics. We also studied the effect of both compounds on DNA oxidative damage, free radical level and HO-1 gene expression. We showed that at low concentrations both CORM-2 and iCORM-2 stimulate PBMCs viability. After 24-h incubation, CORM-2 and iCORM-2, at the concentration of 100 µM, reduce the viability of both PBMCs and HL-60 cells. We also demonstrated that CORM-2 and iCORM-2, in the 0.01–100 µM concentration range, cause DNA damage such as strand breaks and alkaline labile sites. DNA damage was repaired efficiently only in HL-60 cells. CORM-2 significantly reduces oxidative stress induced by 1 mM H2O2 in normal and cancer cells. On the contrary, iCORM-2 in HL-60 cells increases the level of free radicals in the presence of 1 and 5 mM H2O2. We also revealed that both CORM-2 and iCORM-2 induce HO-1 gene expression. However, CORM-2 induces this gene to a greater extent than iCORM-2, especially in HL-60 cells at 100 µM. Finally, we showed that CORM-2 and iCORM-2 reduce H2O2-induced DNA oxidative damage. Furthermore, CORM-2 proved to be a compound with stronger antioxidant properties than iCORM-2. Our results suggest that both active CORM-2 and inactive iCORM-2 exert biological effects such as cyto- and genotoxicity, antioxidant properties and the ability to induce the HO-1 gene. The released CO as well as iCORM-2 can be responsible for these effects.


Methods
Cells culture. Peripheral blood mononuclear cells (PBMCs) were isolated from a leucocyte-buffy coat collected from the blood of healthy non-smoking donors from the Blood Bank in Lodz, Poland, as described previously 21 . The study protocol was approved by the Committee for Research on Human Subjects of the University of Lodz (17/KBBN-UŁ/III/2019).
DNA damage. CORM-2 and iCORM-2 were added to the suspension of the cells to give final concentrations of 0.01, 0.025, 0.05, 0.1, 1, 10, 50 and 100 μM. Both PBMCs and HL-60 cells were incubated for 2 h at 37 °C in 5% CO 2 . The experiment included a positive control, i.e. a cells sample incubated with hydrogen peroxide (H 2 O 2 ) at 25 μM for 15 min on ice. The cells after treatment with CORM-2 and iCORM-2 were washed and suspended in the IMDM medium. A freshly prepared suspension of the cells in LMP agarose dissolved in PBS was spread onto microscope slides. The slides were processed as described previously 21,22 .
DNA repair. PBMCs and HL-60 cells were incubated with CORM-2 and iCORM-2 for 2 h at 37 °C in 5% CO 2 at 100 µM and then were washed and suspended in fresh IMDM medium preheated to 37 °C. DNA repair was assessed by the extent of residual DNA damage detected at each time-point using the comet assay as described previously 22 . www.nature.com/scientificreports/ Effect of CORM-2 on DNA oxidative damage. Based on the results obtained by Babu et al. 10 , we investigated CORM-2 as a compound showing potential to reduce DNA oxidative damage. We prepared two experimental schemes of CORM-2 and iCORM-2 incubation with H 2 O 2 : pre-incubation and pre-incubation + coincubation . In the first scheme, the cells were initially incubated with 40 µM CORM-2 or iCORM-2 for 1 h at  37 °C in 5% CO 2 ; then the cells were washed and incubated with 25 or 50 µM H 2 O 2 for 15 min on ice. In the  second scheme, the cells were initially incubated with 40 µM CORM-2 or iCORM-2 for 1 h at 37 °C in 5% CO 2 ;  then the cells were incubated simultaneously with 40 µM CORM-2 or iCORM-2 and 25 or 50 µM H 2 O 2 for  15 min on ice. After incubation in all the schemes, the cells were washed, suspended in LMP agarose and spread onto microscope slide. The slides were processed as described previously 21,22 .
Comet assay. The comet assay was performed under alkaline conditions according to a procedure described previously 21,22 .
Evaluation of oxidative stress. In order to measure the production of reactive oxygen species (ROS), the

Statistical analysis.
The values of the cell viability experiment were presented as mean ± SD from six repeats. The values of the comet assay were expressed as mean + standard error of the mean from three experiments; data from three experiments were pooled, and the statistical parameters were calculated. The statistical analysis was conducted using the Mann-Whitney test (samples with distributions departing from normality) and the Student's t-test (samples with the normal distribution). HO-1 gene expression was calculated by double delta Ct. Statistics were performed using Student's two-tailed t test. Data were presented as a mean ± SD, relative to control. HO-1 expression was normalized to GAPDH (as a reference gene).
The differences were considered to be statistically significant when the P value was < 0.05.

Results
Cell viability. We used the resazurin reduction assay to determine cell viability after incubation with CORM-2 and iCORM-2. This assay is based on the application of an indicator dye to measure oxidation-reduction reactions, which principally occur in the mitochondria of live cells. The non-fluorescent dark blue dye (resazurin) becomes fluorescently pink at 570 nm and fluorescently red at neutral pH (resorufin), when reduced by metabolically active cells.
We observed an increase in the RFU value after 2 h incubation with CORM-2 at 100 µM in both PBMCs (P < 0.001) ( Fig. 2A) and HL-60 cells (P < 0.01) (Fig. 2D). Under these conditions, we observed a decrease in RFU in PBMCs in the case of iCORM-2 at 0.1 µM (P < 0.05) and 10 µM (P < 0.01). We did not observe any changes in the RFU values after 6 h incubation of the cells with CORM-2 and iCORM-2 ( Fig. 2B,E). After 24 h incubation, we noticed an increase in the RFU values for CORM-2 (P < 0.001) and iCORM-2 (P < 0.05) at the concentration of 0.1 µM, but only in PBMCs (Fig. 2C). In both cell types, the RFU values decreased after incubation with CORM-2 and iCORM-2 at 100 µM (Fig. 2C,F). Our results indicate that CO released from CORM-2 at a low concentration can increase cell viability. However, both CORM-2 and iCORM-2 were eventually cytotoxic at 100 µM for normal and cancer cells.
DNA damage and repair. Figure 3 shows the level of DNA damage analyzed by the comet assay under alkaline conditions. The comet assay in the alkaline version is a sensitive and simple method of determining the level of DNA damage, including single-and double-strand breaks and alkali-labile sites in living cells 25 . We observed a significant increase in the level of DNA damage in PBMCs incubated with CORM-2 and iCORM-2 compared to DMSO (Fig. 3A). We also observed a significant increase of DNA damage in the case of HL-60 cells after incubation with CORM-2 and iCORM-2 compared to negative control and DMSO (P < 0.001) (Fig. 3B). We did not detect any significant differences in the level of DNA damage between CORM-2 and iCORM-2 in PBMCs and HL-60 cells, except HL-60 cells incubated with 0.1 µM (P < 0.001) (Fig. 3B). In this case, we observed a significant decrease of DNA damage in cells incubated with iCORM-2 compared to the level of DNA damage in cells incubated with CORM-2. Figure 4 shows the example images of comets from this experiment. www.nature.com/scientificreports/  www.nature.com/scientificreports/ Figure 5 shows DNA damage in PBMCs (A) and HL-60 cells (B) incubated with CORM-2 and iCORM-2 at 100 µM immediately after 2 h incubation as well as 30, 60 and 120 min later. We observed changes over time in the level of DNA damage after washout of the test compounds. We have determined that the damage was repaired, when the cells which were incubated with CORM-2 or iCORM-2 reached the level of DNA damage in control cells after 120 min repair incubation. We detected that DNA damage induced by CORM-2 in PBMCs was not repaired within 120 min post-incubation (Fig. 5A). We observed a significant difference (P < 0.001) in the level of DNA damage between cells incubated with CORM-2 and negative control. We also observed a significant difference in the level of DNA damage between PBMCs incubated with CORM-2 and iCORM-2 after 60 min (P < 0.05) and 120 min (P < 0.01) (Fig. 5A).
HL-60 cells incubated with CORM-2 and iCORM-2 were able to ensure total repair of DNA damage within the repair incubation time of 120 min (Fig. 5B).
PBMCs and HL-60 cells exposed to 25 µM H 2 O 2 for 15 min on ice (positive control) were able to effectively repair DNA damage within 120 min (Fig. 5A,B). We did not observe any changes in the level of DNA damage during repair incubation of cells exposed to DMSO (data not shown).
The effect of CORM-2 on oxidative stress was very strong in HL-60 cells (Fig. 6B). The effect of CORM-2 was significant even in the cells which were not incubated with H 2 O 2 . We also observed that CORM-2 reduces oxidative stress in cells incubated with 1 mM H 2 O 2 (P < 0.001) . In the case of HL-60 cells pre-incubated with iCORM-2, we noticed a significant increase in oxidative stress, especially in the cells incubated with 5 mM H 2 O 2 (P < 0.001).  Figure 7 shows the effects on HO-1 gene expression. Both in PBMCs and HL-60 cells we observed significant HO-1 gene upregulation after incubation with CORM-2 and iCORM-2 (P < 0.001). The observed increase was dependent on the concentration of both CORM-2 and iCORM-2. In PBMCs incubated with iCORM-2 at the concentrations of 40 µM and 100 µM, we detected a slightly lower increase of HO-1 gene expression (P < 0.05 and P < 0.001, respectively) compared to PBMCs incubated with CORM-2 (Fig. 7A).
In HL-60 cells incubated with CORM-2 at 100 µM we noticed a 100-fold increase of HO-1 gene expression (Fig. 7B). Similarly like in PBMCs, we observed a slightly lower increase of HO-1 gene expression after incubation with iCORM-2 at the concentration of 40 µM compared to HL-60 cells incubated with 40 µM CORM-2 (P < 0.001). In the case of HL-60 cells incubated with iCORM-2 at the concentration of 100 µM, we detected approximately 50% lower expression of the HO-1 gene compared to the cells incubated with 100 µM CORM-2 (P < 0.001).

Effect of CORM-2 and iCORM-2 on DNA oxidative damage.
In this experiment we induced DNA oxidative damage in PBMCs and HL-60 cells by using H 2 O 2 and we investigated the effect of CORM-2 and iCORM-2 on the level of DNA oxidative damage. We performed these studies in two different experimental systems described in Materials and Methods. The results obtained in the experiment with pre-incubation of the cells with CORM-2 or iCORM-2 clearly showed a significant decrease of DNA oxidative damage induced www.nature.com/scientificreports/ by H 2 O 2 (P < 0.001) (Fig. 8A,B). This effect was more pronounced in the case of CORM-2 and DNA damage induced by 25 µM H 2 O 2 . Similarly, in the experiment with pre-incubation and co-incubation with CORM-2 or iCORM-2 we observed a significant decrease of DNA oxidative damage in both PBMCs and HL-60 cells (P < 0.001) (Fig. 8C,D).

Discussion
In this study, we examined the effect of CORM-2 and iCORM-2 on human peripheral blood mononuclear cells (PBMCs) and human promyelocytic leukemia HL-60 cells. We determined cell viability, DNA damage and their repair kinetics. We also studied the effect of both compounds on DNA oxidative damage, free radical level and HO-1 gene expression.
The mitochondria are the most recognized cellular targets for carbon monoxide. CO prevents cell death by limiting mitochondrial membrane permeabilization, which inhibits the release of pro-apoptotic factors into the cytosol 26 . It was found that CORM-2 significantly attenuated 6-hydroxydopamine (6-OHDA)-induced apoptotic cell death in a dose-dependent manner in C6 glioma cells 16 . CORM-2 decreased the Bax/Bcl2 ratio and caspase-3 activity, which had been increased by 6-OHDA. Winburn et al. showed that both CORM-2 and iCORM-2 decreased cisplatin-induced caspase-3 activity in MDCK (Madin-Darby canine kidney Cells) and HeK (human embryonic kidney) cells suggesting an anti-apoptotic effect 27 . On the other hand, it was shown in the same study that both CORM-2 and iCORM-2 induced cellular toxicity by decreased cell viability, abnormal cell cytology, increased apoptosis and necrosis, cell cycle arrest and reduced mitochondrial enzyme activity 27 .
In our study we showed that both CORM-2 and iCORM-2 induce DNA damage, including DNA single and double strand breaks and alkali labile sites in PBMCs and HL-60 cells (Fig. 3A,B). We did not observe any significant differences between CORM-2 and iCORM-2 in the level of induced DNA damage. HL-60 cells are much more sensitive to both CORM-2 and iCORM-2 compared to normal PBMCs. We observed a significant increase of DNA damage in HL-60 cells after incubation with CORM-2 and iCORM-2 compared to negative control (P < 0.001) and DMSO (P < 0.001). The difference in the level of DNA damage between PBMCs and HL-60 cells can be associated with the overexpression of transferrin receptors present on the surface of cancer cells. As www.nature.com/scientificreports/ previously described, leukemic cells are known to overexpress transferrin receptors 28 . CORM-2 contains ruthenium, which can mimic iron. It was shown that ruthenium can be up taken into cells by transferrin receptors 29 . DNA damage induced by CORM-2 and iCORM-2 in HL-60 cells was effectively repaired within 120 min postincubation (Fig. 5B). PBMCs did not repair DNA damage induced by CORM-2. On the contrary, we observed an increase of DNA damage in these cells during post-incubation repair (Fig. 5A). DNA damage induced by iCORM-2 in PBMCs was effectively repaired. Possible reasons for this may include differences in the activity of DNA repair systems in normal and cancer cells. The presence of c-MYC oncogene in HL-60 may be responsible for more effective DNA repair compared to PBMCs. It was found that increased expression of c-MYC results in a decrease of BIN1 (Bridging Integrator 1) protein expression. BIN1 binds to PARP1 and inhibits its activity. Decreased expression of BIN1 causes induction of PARP and can stimulate DNA repair 30 . On the other hand, it was shown that HO-1 induction or exposure to CO induces homologous recombination-mediated DNA repair through ataxia-telangiectasia mutated/ataxia telangiectasia and Rad3-related (ATM/ATR) protein 31 . Effective repair of CORM-2-induced DNA damage may be also associated with high mobility group box 1 protein (HMGB-1). It was found that CORM-2 treatment prevents nuclear-cytoplasmic translocation of this protein in primary mouse renal proximal tubular epithelial cells (RPTECs) 32 . HMGB1 protein accumulated in cellular nuclei as a result of CORM-2 action might stimulate DNA repair 33 . www.nature.com/scientificreports/ We suggest that DNA damage induced by CORM-2 and iCORM-2 may result from the presence of ruthenium in these molecules. Many types of interactions of ruthenium-containing compounds with DNA have been described in the literature such as coordinative, intercalative, minor groove binding, sequence specificity of DNA binding, the ability of ruthenium compounds to condense and cleave DNA, binding to A-and Z-DNA, DNA quadruplexes and other unusual DNA structures 34 . Ruthenium, similarly to other transition metals such as Fe, Cr and Cu, can damage cells by producing free radicals. It was shown that CORM-2 caused DNA damage in bacteria cells 35 . Bacteria cells treated with CORM-2 contained higher levels of free iron arising from the destruction of iron-sulfur proteins. Moreover, Tavares et al. showed that CORM-2 generated hydroxyl radicals in a cell-free solution, a process that was abolished by scavenging CO 35 . It was demonstrated that the radical formation from CORM-2 is closely associated with the presence of CO ligands as no radical species were observed in the inactive compound devoid of CO, iCORM-2 35 . Our study of the cell system revealed that CORM-2 at 40 µM does not induce free radicals (Fig. 6A,B). On the contrary, this compound reduces free radicals even in cells not exposed to oxidative stress. We observed free radical generation in HL-60 cells pre-incubated with 40 µM iCORM-2 (Fig. 6B). In the presence of 5 mM H 2 O 2 , especially in HL-60 cells pre-incubated with 40 µM iCORM-2, we demonstrated a significant increase in the level of free radicals compared to cells pre-incubated with 40 µM CORM-2 (P < 0.001) (Fig. 6B). Our results clearly indicate that the ability of CORM-2 to reduce oxidative stress depends on CO.
The antioxidant properties of CORM-2 are very well documented. Studies of serum plasma showed that CORM-2 reduces lipid peroxidation induced by H 2 O 2 and also by H 2 O 2 /Fe 6 . Moreover, in human umbilical vein endothelial cells, CORM-2 administration diminishes oxidative stress induced by hypoglycemia 36 . In vivo, in hypothalamic paraventricular nucleus of Male Dahl Salt-Sensitive rats, inflammation and oxidative stress were induced through high-salt-induced hypertension. As it was shown, a microinjection of CORM-2 decreased the level of ROS, while the levels of CU/Zn-SOD and HO-1 were elevated 13 . CORM-2 also reduces the ROSdependent doxorubicin cardiotoxicity in mice. In this study, mice were treated with doxorubicin and cardiotoxicity was evaluated by markers such as creatine kinase, lactate dehydrogenase, malondialdehyde and total antioxidant status in serum. Co-treatment of CORM-2 led to a significant reduction of those markers, while the level of HO-1 was markedly elevated 8 . Another study showed that a very low concentration of CORM-2 (50 nM) significantly reduced trimethyltin-induced superoxide production in SH-SY5Y neuroblastoma cells 37 . Moreover, pretreatment with CORM-2 significantly inhibited airborne particulate matter-induced mitochondria-derived ROS production in human pulmonary alveolar epithelial cells (HPAEpiCs) 12 . An experiment carried out on the human gastric cancer AGS cell line confirmed the antioxidant properties of CORM-2 involving significant inhibition of IL-1β-induced ROS production 15 . Pretreatment with CORM-2 inhibited angiotensin-II-induced ROS generation in human aortic smooth muscle cells (HASMCs). Therefore, CORM-2 can play the role of a protective antioxidant in heart and blood vessels 9 . An in vivo study performed on rats exhibited a decrease in oxidative damage of DNA after exposure to CORM-2. The concentration of 8-OHG measured in gastric mucosa cells after exposure to ischemia/reperfusion was significantly lower for CORM-2 compared to the vehicle (DMSO and saline in ratio 1:10) 7 .
The cytoprotective properties of CORM-2 observed by us and by other researchers may result from the ability of this compound to induce HO-1 7,16,17,38,39 . It was shown that CORM-2-induced HO-1 expression was mediated through a Pyk2/PDGFR/PI3K/Akt/FoxO1/Sp1-dependent manner and exerted a cytoprotective effect in human cardiomyocytes 38 . Moreover, it was observed that CORM-2 activates the c-SRC/EGFR/PI3K/Akt/JNK1/2 and p38 MAPK pathways, which cause Nrf2 activation and HO-1 expression in human tracheal smooth muscle cells (HTSMCs) 39 . In human hepatocellular carcinoma cell lines (HCC), the HO-1/CO axis conferred resistance to the TGF-β growth inhibitory signal by increasing Smad3 phosphorylation at Thr-179 via the ERK1/2 pathway 17 . HO-1 is considered to be a potential target in cancer therapy, including leukemia 40 . Our results indicate that both CORM-2 and iCORM-2 induce expression of HO-1 in PBMCs and HL-60 cells (Fig. 7A,B). However, after incubation of HL-60 cells with 100 µM CORM-2, we observed a twofold increase in HO-1 expression compared to HL-60 cells incubated with iCORM-2 (Fig. 7B). This result confirms previous conclusions, mentioned above, that CO released from CORM-2 increases HO-1 expression. The increase in HO-1 expression in PBMCs and HL-60 cells observed after incubation with iCORM-2 is probably due to the presence of ruthenium. Therefore, it should be assumed that an increase in HO-1 expression noticeable after incubation of the cells with CORM-2 is due to both the release of CO and the presence of iCORM-2.
The results of our studies regarding the reduction of the level of free radicals and induction of the HO-1 gene by CORM-2 and iCORM-2 prompted us to investigate whether these compounds can protect DNA against oxidative damage. Using the comet assay we measured the level of DNA oxidative damage in PBMCs and HL-60 cells pre-incubated only as well as pre-and co-incubated with CORM-2 and iCORM-2 at the concentration of 40 µM (Fig. 8). We observed a significant reduction in DNA oxidative damage in the two experimental systems for both CORM-2 and iCORM-2. However, the protective effect was significantly greater in the case of CORM-2. The results recorded confirm our assumptions that the ability of CORM-2 to reduce DNA oxidative damage is caused by both the released CO and also by the metal core of CORM-2. Some ruthenium complexes have been shown to have antioxidant properties [41][42][43] . Two main oxidation states, i.e. Ru(II) and Ru(III), are accessible for ruthenium species in physiological solution. In both oxidation states the Ru ion is a six-coordinate complex with octahedral geometry and has good affinity to nitrogen and sulfur ligands 44 . CORM-2 after releasing CO can form Ru(CO) 2 adducts, preferentially with histidine residues, as demonstrated with synthetic peptides using mass-spectrometry analysis 45 . Moreover, it was shown that functional consequences of these adducts can be diverse. While KCa1.1 channels were activated, channels Kv11.1, Kv10.1, and Kv1.5 were inhibited by CORM-2 in a CO-independent manner. Thus, CORM-2 seems can serious side effects as a drug.
Our results indicate that not only CORM-2 but also iCORM-2 has a biological effect on normal and cancer cells. Depending on the incubation time and concentration, they can be cytotoxic or stimulate cell viability.