Substituted Piperazines as Novel Potential Radioprotective Agents

The increasing risk of radiation exposure underlines the need for novel radioprotective agents. Hence, a series of novel 1-(2-hydroxyethyl)piperazine derivatives were designed and synthesized. Some of the compounds protected human cells against radiation-induced apoptosis and exhibited low cytotoxicity. Compared to the previous series of piperazine derivatives, compound 8 exhibited a radioprotective effect on cell survival in vitro and low toxicity in vivo. It also enhanced the survival of mice 30 days after whole-body irradiation (although this increase was not statistically significant). Taken together, our in vitro and in vivo data indicate that some of our compounds are valuable for further research as potential radioprotectors.


Introduction
Ionizing radiation (IR) was discovered in the 19th century; since then, its use has had a dramatic impact on medicine, space travel, research, and the energy industry. Unfortunately, the catastrophic effects of IR have also been exploited by the development of nuclear weapons; moreover, a number of accidents have occurred at nuclear power reactors resulting in ecological disasters and loss of lives [1][2][3].
Exposure to IR is associated with cell death, genetic mutations, and carcinogenesis [4]. It induces DNA damage in the form of double-strand DNA breaks, which is considered the underlying mechanism of the resulting cell death: apoptosis.
Apoptosis is a very complex process, which is tightly controlled in mammalian cells [5]. Its multi-level regulation has been the subject of medical research for a long time because some pathologies (such as cancer and autoimmune and neurodegenerative disease) are closely associated of intermediate 10b (Scheme S6 in Supplementary Materials). This intermediate was isolated during the procedure of preparation of 9 as the second expected product in a yield of 45%. The next step involved the alkylation of 1-(2-hydroxyethyl)piperazine (1) by the intermediate (10b), resulting in compound 10 in 90% yield (Scheme S6 in Supplementary Materials).
Molecules 2019, 24, x; doi: www.mdpi.com/journal/molecules Supplementary Materials). This intermediate was isolated during the procedure of preparation of 9 as the second expected product in a yield of 45%. The next step involved the alkylation of 1-(2hydroxyethyl)piperazine (1) by the intermediate (10b), resulting in compound 10 in 90% yield (Scheme S6 in Supplementary Materials). Finally, the compounds (2-10) were converted to their salts by treatment with hydrochloric acid in methanol. All the final compounds were characterized by their 1 H-and 13 C-NMR spectra and high-resolution mass spectrometry (HRMS). All analytical measurements confirmed the structure and purity of over 95% of the final compounds as determined by LC-MS analysis. All the spectral data of the compounds are provided in the Experimental section.
Scheme 1. Synthesized compounds as bases with potential radioprotective properties.

Molecular Docking with Anti-Apoptotic Protein Bcl-2
The docking calculations with ligand anti-apoptotic protein Bcl-2 (B-cell lymphoma 2) were performed for all of the investigated compounds. All of them showed hydrophobic interaction of the aromatic part (tetrahydroacridine, naphtyl, or phenyl moiety) with the hydrophilic pocket of the Bcl-2 protein (Leu134, Phe101, Tyr105, Phe109, Met112, Phe150 and Val130), except for compound 8.
The quarternary hydrogen of all compounds interacted with Asp108, except for compound 9.
Scheme 1. Synthesized compounds as bases with potential radioprotective properties.
Finally, the compounds (2-10) (Table 1) were converted to their salts by treatment with hydrochloric acid in methanol. All the final compounds were characterized by their 1 H-and 13 C-NMR spectra and high-resolution mass spectrometry (HRMS). All analytical measurements confirmed the structure and purity of over 95% of the final compounds as determined by LC-MS analysis. All the spectral data of the compounds are provided in the Experimental section.

Molecular Docking with Anti-Apoptotic Protein Bcl-2
The docking calculations with ligand anti-apoptotic protein Bcl-2 (B-cell lymphoma 2) were performed for all of the investigated compounds. All of them showed hydrophobic interaction of the aromatic part (tetrahydroacridine, naphtyl, or phenyl moiety) with the hydrophilic pocket of the Bcl-2 protein (Leu134, Phe101, Tyr105, Phe109, Met112, Phe150 and Val130), except for compound 8. The quarternary hydrogen of all compounds interacted with Asp108, except for compound 9.
Molecules 2020, 25, 532 4 of 16 A hydrogen bond of uncharged piperazine nitrogen with Arg143 was observed frequently. Some of the compounds showed a hydrogen bond of aniline nitrogen with Asp108. Other interactions could not be judged as group characteristics but rather individual features.
Interaction of compounds 4, 8, and 10 with Bcl-2 are shown in Figure 1. A hydrogen bond of uncharged piperazine nitrogen with Arg143 was observed frequently. Some of the compounds showed a hydrogen bond of aniline nitrogen with Asp108. Other interactions could not be judged as group characteristics but rather individual features.
Interaction of compounds 4, 8, and 10 with Bcl-2 are shown in Figure 1. In this study, the cytotoxic effects of nine new compounds (2, 3, 4, 5, 6, 7, 8, 9, 10) were tested on 10 human cell lines (nine cancer and one non-cancer cell line). The cell lines were treated with each compound at concentrations of 10 µM and 100 µM for 48 h. The proliferation capacity of the treated cells was determined by WST-1 test and compared against the proliferation rate of the untreated control cells (100%).
These experiments revealed that compound 2 (10 µM) was cytotoxic across all the 10 cell lines tested, as indicated by the percentage of viable cells which was less than 44%, especially in MOLT-4 cells. Accordingly, compound 2 was not tested at 100 µM. The remaining compounds had no evident cytotoxic effect on any of the tested cell lines.
Further experiments at 100 µM revealed a significantly lower number of living cells when compound 3 was used. The cytotoxic effect of this compound was detected in all the tested cell lines, decreasing cell viability by 70% and more, especially in MOLT-4 and MRC-5 cells. However, the remaining compounds (4, 5, 6, 7, 8, 9, and 10) had no significant cytotoxic effects on the studied cell lines (Table 2). Table 2. Cytotoxic effect of novel compounds in vitro. Values represent cell viability after treatment with 10 µM (A) and 100 µM (B) of the tested compounds. The results are expressed as relative viability (%) when compared to an untreated control (100%). Compound 2 was not tested at 100 µM. DOXdoxorubicin (1 µM). Each value represents the mean of three independent experiments ± SD. Values from In this study, the cytotoxic effects of nine new compounds (2, 3, 4, 5, 6, 7, 8, 9, 10) were tested on 10 human cell lines (nine cancer and one non-cancer cell line). The cell lines were treated with each compound at concentrations of 10 µM and 100 µM for 48 h. The proliferation capacity of the treated cells was determined by WST-1 test and compared against the proliferation rate of the untreated control cells (100%).
These experiments revealed that compound 2 (10 µM) was cytotoxic across all the 10 cell lines tested, as indicated by the percentage of viable cells which was less than 44%, especially in MOLT-4 cells. Accordingly, compound 2 was not tested at 100 µM. The remaining compounds had no evident cytotoxic effect on any of the tested cell lines.
Further experiments at 100 µM revealed a significantly lower number of living cells when compound 3 was used. The cytotoxic effect of this compound was detected in all the tested cell lines, decreasing cell viability by 70% and more, especially in MOLT-4 and MRC-5 cells. However, the remaining compounds (4, 5, 6, 7, 8, 9, and 10) had no significant cytotoxic effects on the studied cell lines (Table 2). Table 2. Cytotoxic effect of novel compounds in vitro. Values represent cell viability after treatment with 10 µM (A) and 100 µM (B) of the tested compounds. The results are expressed as relative viability (%) when compared to an untreated control (100%). Compound 2 was not tested at 100 µM. DOX-doxorubicin (1 µM). Each value represents the mean of three independent experiments ± SD. Values from the intervals 0-25% (red color), 26-50% (orange color) and 50-75% (yellow color) are highlighted with different colors.

In Vitro Toxicity Determination of New Inhibitors
In further experiments, MTT (3-(4,5-di-methylthiazol-2-yl)-2,5-diphenyltetrazolium bromide) assay was employed to evaluate the toxicological indexes of the newly synthesized compounds, including IC 50 (half-maximal inhibitory concentration) and MTC (maximum tolerated concentration). Since the assay is applicable only for adherent cells, the A-549 cell line was selected for these evaluations.
In A549 cells, the lowest IC 50 was found for compound 2 (0.04 ± 0.002 mM). The tolerance of other compounds increased in the following order from 2 to 3, 9, 6, 4, 5, 7, 10, and 8, with corresponding IC 50 values being 3.5-, 6.0-, 11.3-, 16.0-, 33.8-, 63.8-, 126.5-, and 625-fold higher than that for compound 2 (Table 3). Phosphatidylserine, a major component of the inner leaflet of the cytoplasmic membrane bilayers, is frequently used as an indicator of cell death. The externalization of this phospholipid can be detected through Annexin V, which binds to phosphatidylserine in the presence of calcium. Cell viability was quantified by Annexin V-Alexa Fluor ® 488/propidium iodide (PI) staining and flow cytometry analysis to determine whether the tested compounds could prevent radiation-induced cell death in MOLT-4 cells. The percentage of viable cells was 92% for non-irradiated control (0.1% DMSO vehicle) and 40% for untreated irradiated cells (1 Gy) (Figure 2A). A 60-min pre-treatment with the compounds at 100 µM significantly increased cell viability after irradiation (1 Gy), with the following values: 61% (compound 4), 69% (compound 6), 64% (compound 7), 57% (compound 8), and 51% (compound 10) ( Figure 2B). Compound 9 exhibited no significant radioprotective effect in vitro. Although compound 5 did not show statistically significant radioprotective effect (59%), it was further evaluated in vivo due to its low cytotoxicity.

Pre-Treatment with Compounds 4, 5, 6, 7, 8, and 10 Reduced Radiation-Induced Apoptosis in Vitro
Phosphatidylserine, a major component of the inner leaflet of the cytoplasmic membrane bilayers, is frequently used as an indicator of cell death. The externalization of this phospholipid can be detected through Annexin V, which binds to phosphatidylserine in the presence of calcium. Cell viability was quantified by Annexin V-Alexa Fluor ® 488/propidium iodide (PI) staining and flow cytometry analysis to determine whether the tested compounds could prevent radiationinduced cell death in MOLT-4 cells. The percentage of viable cells was 92% for non-irradiated control (0.1% DMSO vehicle) and 40% for untreated irradiated cells (1 Gy) (Figure 2A). A 60-min pre-treatment with the compounds at 100 µM significantly increased cell viability after irradiation (1 Gy), with the following values: 61% (compound 4), 69% (compound 6), 64% (compound 7), 57% (compound 8), and 51% (compound 10) ( Figure 2B). Compound 9 exhibited no significant radioprotective effect in vitro. Although compound 5 did not show statistically significant radioprotective effect (59%), it was further evaluated in vivo due to its low cytotoxicity.

The Compounds Administered to Mice at MTD Caused No Pathologies
Mild to moderate signs of intoxication could be observed once the MTD (maximum tolerated dose) was reached. In vivo, MTD is described as the highest dose of a given compound that can be administered before showing signs of toxicity (based on the same dose regimen or series) and whose excess would be expected to be fatal, yielding unreliable results [18]. Therefore, a staged approach (including core and intermediate dose levels) has become a suitable strategy in this regard.
The symptoms described in Tables S1 (in Supplementary Materials) and S2 (in Supplementary Materials) diminished spontaneously within 2 h. After observing these symptoms, MTD thresholds were set at 100, 200, 100, 200, 2000, and 650 mg/kg for compounds 4, 5, 6, 7, 8, and 10, respectively. Animals receiving MTD were further examined by necropsy 48 h after administration. Plasma (reflecting biochemical changes in blood), intestine with mesentery (site of administration), and liver and kidney (elimination organs) were collected to evaluate the toxic potential of the tested compounds. For evaluation of the impact of the compounds on organ function, we performed blood biochemistry test 48 h after administration. The test focused on seven important indicators: glucose and amylase for normal function of pancreas, alanine aminotransferase (ALT), aspartate aminotransferase (AST), and alkaline phosphatase (ALP) for the normal function of liver, and urea and creatinine for the normal function of kidneys. According to necropsy, blood biochemistry and histopathology analyses, the tested compounds caused no pathologies, were found harmless, and were suitable for in vivo survival tests (Table S3 in Supplementary Materials). The only histopathological change observed was limited focal necrosis in one male administered with compound 4 (one small focus), one female administered with compound 10 (five small foci), and one control male (two small foci). According to Thoolen et al. [19], focal necrosis can be occasionally seen in untreated rodents and therefore is not pathognomonic of hepatotoxicity, which is consistent with our findings.

Pre-Treatment with the Compounds Increased Survival of Whole-Body Irradiated Mice
To estimate the radioprotective effect on whole-body irradiated mice, compounds 4 (50 mg/kg), 5 (100 mg/kg), 6 (50 mg/kg), 7 (100 mg/kg), 8 (1000 mg/kg), and 10 (325 mg/kg) were injected (intraperitoneally, i.p.) 5 min before receiving 7.15 Gy of gamma radiation. These compounds were selected based on the cytotoxicity screening in vitro, and applied at a dose corresponding to 50% of the MTD in vivo. Notably, the tested compounds did not affect the survival of non-irradiated mice. The follow-up period was 30 days.
The lethal effect of IR could be observed after 10 days in the non-treated group, and only four mice remained alive after the 30-day period of the study. Mice pre-treated with the compounds retained longer (or equal) survival than the non-treated control with the exception of the group pre-treated with compound 7, which showed an earlier lethal effect at 9 days post-irradiation. In addition, compounds 5 and 6 caused a decline in survival at the same time as the non-treated control (10 days post-irradiation). In contrast, the groups pre-treated with 4, 8, and 10 had a higher survival rate, with a drop after day 12 post-irradiation.

Discussion
In recent years, medical radiation uses have increased significantly together with the risk of radiation/nuclear events. Current pre-clinically tested radioprotectors for radiotherapy involve immune modulators, nutraceuticals (isoflavonoids), free-radical scavengers, corticosteroids, or recombinant cytokines. Several promising radioprotective agents have been synthesized de novo.

Discussion
In recent years, medical radiation uses have increased significantly together with the risk of radiation/nuclear events. Current pre-clinically tested radioprotectors for radiotherapy involve immune modulators, nutraceuticals (isoflavonoids), free-radical scavengers, corticosteroids, or recombinant cytokines. Several promising radioprotective agents have been synthesized de novo.
Ex-RAD( ® ), which is also known as the sodium salt of 4-carboxystyryl-4-chlorobenzylsulfone, was reported as a small molecule with radioprotective properties based on targeting p53 and its downstream regulators [20]. Ex-RAD prevented radiation-induced apoptosis and increased the survival rate of whole-body irradiated mice [21]. Tang et al. prepared a group of novel benzyl naphthyl sulfoxide derivatives from Ex-RAD. Interestingly, one of the new derivatives exhibited a remarkable increase (100%) in mice survival after irradiation [22].
Another attempt to develop radioprotective agents was made by Hosseinimehr, who synthesized a group of 2-imino-3-[(chromone-2-yl)carbonyl] thiazolidines. Some of these compounds protected mice from whole-body gamma irradiation with high statistical significance [23].
Obviously, numerous papers were published regarding novel agents for enhanced radiation protection in the last decade; notwithstanding, the only radioprotector approved by the FDA is amifostine [24].
Amifostine was reported to possess a high dose-reducing factor and to prevent radiation-induced injury repeatedly [25,26]. However, FDA has approved it for limited clinical indications only and not for non-clinical applications. Despite the efforts that have been put into increasing amifostine effectiveness, the issues with its toxicity and side effects have not been resolved. Thus, further studies are needed to improve the drug design and safety of future radioprotectors.
When searching for appropriate drug targets in radioprotection, many researchers turned their attention toward the regulation of apoptotic machinery, as both acute and chronic radiotoxicity has been associated with cell death. IR induces apoptosis through a complex DNA damage pathway. While this effect is desirable during the radiotherapy of cancer cells, it leads to radiotoxicity in normal healthy tissues.
There has been growing evidence that a complex of pro-apoptotic and anti-apoptotic proteins (e.g., PUMA-Bcl-2, PUMA-Mcl-1) play the crucial role in apoptosis regulation, and their selective inhibition represents an exciting therapeutic opportunity [27].
Consistently, according to our previous observations, we identified PUMA together with its interaction partners as the key molecule of radiation-induced cell death signaling, and thus a promising target in medical radioprotection, but also in the therapy of neurodegenerative and cardiovascular diseases [28].
Previously, we reported a group of 1-(4-(2-hydroxyethyl)piperazin-1-yl)-3-phenoxypropan-2-ol derivatives with a possible mechanism of radioprotective effect based on interference with Bcl-2 family protein-protein interaction [14]. The goal of this work was to describe the second generation of piperazine derivates, as we anticipate a similar mechanism of action with an improved pharmacological profile. Although the molecular explanation of the potential radioprotective effect is a subject of parallel study (currently undergoing), the molecular docking data indicate possible interaction with the hydrophilic pocket of anti-apoptotic protein Bcl-2.
In the presented study, 10 new compounds were synthesized based on 1-(2-hydroxyethyl) piperazine derivatives. Compound 2 has already been published as a cholinesterase inhibitor with intended application for the treatment of Alzheimer's disease [15]. Since it contained a piperazine structural moiety linked through an aminoethyl group to an aromatic moiety, we presumed its inhibition of other protein-protein interactions. Unfortunately, the compound was discarded in baseline screening due to high cytotoxicity. Therefore, the substance was modified by substituting the piperazine core with a hydroxyethyl group. By substituting the secondary amino group of the piperazine of compound 2, we presumed that there would be a reduction in the cytotoxic effect. This phenomenon was confirmed during the baseline screening but at a higher concentration (100 µM). Compound 3 was eventually found to be excessively toxic to the cell lines and excluded from further testing.
Compound 9 showed no radioprotective effect in vitro. A possible explanation could be that the compound does not contain a basic piperidine moiety, which seems to be essential for this property. Such phenomenon has already been published by our group [14]. Compound 9 could not be tested in vivo due to the low solubility. Compared to the other compounds, the low solubility was due to the presence of two aromatic moieties that also significantly affect the ClogP value (Table 1). All other substances (4, 5, 6, 7, 8, and 10) have been tested in vitro and in vivo based on cytotoxic screening and physicochemical properties.
For structurally similar substances 5, 6, 7, and 10, there is an evident correlation between ClogP and cytotoxic effect on various cell lines. Compound 6, substituted with a naphthalene moiety (with the highest ClogP and hence MTC), appears to be the most promising radioprotector of all the compounds evaluated in vitro. On the other hand, no significant effect was observed under in vivo conditions. The effect is comparable to substances 5 and 7. According to the results for compounds 5 and 10, the presence or absence of a hydroxyl group on the linker plays no significant role concerning the potential radioprotective effect.
Substance 4 is structurally comparable to those prepared recently by our group [14]. In addition, a halogen atom (Cl) was added to the structure, which could influence other binding interactions. In vitro and in vivo tests indicate that the compound (despite its relatively higher toxicity) possesses a radioprotective effect in vivo.
The structurally different compound 8 contains no aromatic moiety. This compound was synthesized by linking two 1-(2-hydroxyethyl)piperazine moieties with an alkyl linker. The compound is minimally non-toxic and excellently soluble in media. After in vivo evaluation, it appeared to be the most effective compound in this group in terms of radioprotection. In addition, according to the safety profile, it might be considered an appropriate compound for further studies.
Regarding the structure-activity relationship, we observed that acridine moiety increased the cytotoxicity of the compounds 2 and 3 and therefore were omitted. Interestingly, lower cytotoxicity was achieved by substitution of the secondary amine at the 4-position of piperazine nitrogen, whereby a tertiary amine was formed (compound 3). The radioprotective properties are reduced in compounds with the methoxy group. Furthermore, the 1-(2-hydroxyethyl)piperazine moiety seems to be crucial, as the compounds lacking this structure possess increased cytotoxicity and insolubility (compound 9). The longest survival was found after the treatment with compound 8. Moreover, it is lacking aromatic moieties and is entirely non-toxic and well soluble. Thus, it might be the most promising radioprotective compound that is suitable for further development.

Molecular Docking
The structure of receptor was gained from the Protein Data Bank, PDB ID 4LXD (Bcl_2-Navitoclax Analog complex) [29]. The structure has a resolution of 1.9 Å and it does not have any Ramachandran's outliers. Therefore, it was found to be suitable for molecular docking. The structure was prepared by the DockPrep function of UCSF Chimera (version 1.4) and converted to pdbqt-files by AutodockTools (v. 1.5.6) [30,31]. Three-dimensional structures of ligands were minimized by Avogadro (v. 1.1.0) and converted to a pdbqt-file format by Open Babel (v. 2.3.1) [32,33]. The docking calculations were done by Autodock Vina (v. 1.1.2) with the exhaustiveness of 8 [34]. Calculation was repeated 10 times for each ligand, and the best-scored result was selected for manual inspection. The visualization of enzyme-ligand interactions was prepared using Pymol (v. 1.7.4.5) [The PyMOL Molecular Graphics System, Version 1.7.4.5, Schrödinger, LLC, Mannheim, Germany].

Cell Culture and Treatment with Novel Compounds
The cell lines Jurkat, MOLT-4, A2780, A549, HT-29, PANC-1, HeLa, MCF-7, SAOS-2, and MRC-5 (all purchased from Sigma Aldrich, Czech Republic) were included in this study. All cell lines were maintained according to the provider's guidelines, and plated and treated in 96-well plates (500 to 30 × 10 3 cells per well). The tested compounds 2, 3, 4, 5, 6, 7, 8, 9, and 10 were dissolved in 0.1% DMSO and kept in 10 mM stock solutions. Before use, the stock solution was diluted in culture media at ratios of 1:100 and 1:1000. The studied cell lines were treated with the compounds 2, 3, 4, 5, 6, 7, 8, 9, and 10 at concentrations of 10 and 100 µM for 48 h. Control cells were tested in culture media with and without DMSO (0.1%). The cells were also treated with 1 µM doxorubicin as a cytotoxicity control.
IC 50 and MTC were determined through MTT assays. The MTT assay was performed according to [35]. In brief, the A549 cell line was plated in 96-well plates (100 µL, 9 × 10 3 cells per well) and mounted in DPX (Dibutylphthalate Polystyrene Xylene) media (Merck). The histopathological analysis was performed on a BX-51 microscope (Olympus, Tokyo, Japan).

Animals and Gamma Radiation
Adult female BALB/c mice (21 ± 0.8 g) were used for all experiments (Velaz a.s., Prague, Czech Rep.). All the performed methods were in accordance with NIRS Guidelines for the Care and Use of Laboratory Animals and approved by the Committee for animal rights of the Ministry of Defence of the Czech Republic no.ČJ MO 255172/2019-684800. The mice, arranged in groups of 10 animals each, received a dose of 7.15 Gy from a 60 Co gamma source (Chisotron, Chirana, Czech Republic) at a dose rate of 1.04 Gy/min. Non-irradiated and non-treated control groups were included. Their survival and health status were checked on a daily basis for 30 days.

Statistical Analysis
The charts were made with the GraphPad Prism 6 biostatistics software (GraphPad Software, La Jolla, CA, USA). The evaluated groups were analyzed by Student's t-test or one-way ANOVA, followed by a post hoc Tukey test (p < 0.05).

Conclusions
In summary, all the compounds were tested in vitro, revealing that compounds 2 (10 µM) and 3 (100 µM) were highly cytotoxic, while compound 9 was not cytotoxic, but was insufficiently soluble for necessary concentrations, making it inappropriate for further tests in vivo. After determining the MTD in vivo of the remaining compounds (4, 5, 6, 7, 8, and 10), we used only 50% of the calculated MTD and assessed the radioprotective effect through the long-term survival of the irradiated mice.
Our results show that compounds 4 and 10 were able to prolong survival but failed to prevent fatal radiation-induced injury in the long term. On the other hand, pre-treatment with compound 8 led to an increased survival rate in the irradiated animals. Although this increase was not statistically significant, we propose compound 8 as a valuable candidate for further research due to its low toxicity and appropriate yield during synthesis.