Toxicity of Glycyl-l-Prolyl-l-Glutamate Pseudotripeptides: Cytotoxic, Oxidative, Genotoxic, and Embryotoxic Perspectives

The tripeptide H-Gly-Pro-Glu-OH (GPE) and its analogs began to take much interest from scientists for developing effective novel molecules in the treatment of several disorders including Alzheimer's disease, Parkinson's disease, and stroke. The peptidomimetics of GPEs exerted significant biological properties involving anti-inflammatory, antiapoptotic, and anticancer properties. The assessments of their hematological toxicity potentials are critically required for their possible usage in further preclinical and clinical trials against a wide range of pathological conditions. However, there is so limited information on the safety profiling of GPE and its analogs on human blood tissue from cytotoxic, oxidative, and genotoxic perspectives. And, their embryotoxicity potentials were not investigated yet. Therefore, in this study, measurements of mitochondrial viability (using MTT assay) and lactate dehydrogenase (LDH) release as well as total antioxidant capacity (TAC) assays were performed on cultured human whole blood cells after treatment with GPE and its three novel peptidomimetics for 72 h. Sister chromatid exchange (SCE), micronucleus (MN), and 8-oxo-2-deoxyguanosine (8-OH-dG) assays were performed for determining the genotoxic damage potentials. In addition, the nuclear division index (NDI) was figured out for revealing their cytostatic potentials. Embryotoxicity assessments were performed on cultured human pluripotent NT2 embryonal carcinoma cells by MTT and LDH assays. The present results from cytotoxicity, oxidative, genotoxicity, and embryotoxicity testing clearly propounded that GPEs had good biosafety profiles and were trouble-free from the toxicological point of view. Noncytotoxic, antioxidative, nongenotoxic, noncytostatic, and nonembryotoxic features of GPE analogs are worthwhile exploring further and may exert high potentials for improving the development of novel disease-modifying agents.


Introduction
Te small peptide glycine-L-proline-L-glutamate (GPE) (Figure 1) is naturally cleaved from the N-terminal sequence of insulin-like growth factor (IGF-1) by proteases. Tere is a commercially available form of this tripeptide known as Glypromate [1,2]. GPE and its analogs exert key biological functions. Principally GPE and its modifed peptidomimetics exhibit neuro-modulatory and neuroprotective properties by providing protection against toxic insults by N-methyl-D-aspartate (NMDA) and amyloid beta, giving countenance to the release of neurotransmitters as well as reducing infammatory environment and activating diferent pathways involved in prosurvival [3][4][5]. In addition to these actions antioxidative, proliferative, and antiapoptotic features by GPE analogs were also previously documented [6][7][8][9][10].
Te usage of GPE as a therapeutic tripeptide is limited due to its certain biochemical and pharmacokinetic natures. In that, the presence of degradation process by peptidases and metabolic stability issues are considered as the main limitedness. Correspondingly, GPE exhibited poor delivery to the central nervous system (CNS) due to a very short plasma half-life time (t1/2 < 1 h). Hence, it seems to be very critical to develop efective modifying strategies towards neuropeptides involving GPE for dissipating available limitations. In fact, diferent strategies are being introduced for modifying GPE tripeptide [11][12][13]. In a recent efort, the amino acids modifcation of the GPE sequence was practiced for the purpose of enhancing its resistance against proteases and still maintaining its cytoprotective action. Ultimately, three diferent novels of GPE peptidomimetics shortened as GPE1, GPE2, and GPE3 were synthesized via the reducement of the peptidic bond to an aminomethylenic group at the Pro-Glu (GPE1), Gly-Pro (GPE3), or both the junctions (GPE2) (Figure 1). Tese modifed GPEs exhibited longer plasma half-life time (t1/2 > 4.5 h) [5].
Recently small peptides such as GPEs have attracted extensive attention as multipotent therapeutics for the treatment of several diseases including Alzheimer's and Parkinson's diseases, stroke, diabetes as well as some types of cancer [10,14,15]. Although the peptidomimetics feld has emerged as a fruitful strategy for executing drug candidates, so little attention has been given to their potential toxic side efects and the interaction potentials of these bioactive tripeptides with DNA, RNA, other proteins, or other subcellular structures were neglected. In this context, the investigation of cytologic and genetic damage potentials of novel GPEs such as GPE1, GPE2, and GPE3 is literally worthwhile for developing further and may contribute to the generation of novel disease-modifying agents [6]. Based on these earlier fndings, the main objective of the current study was to investigate the cytotoxic, oxidative, and genotoxic damage potentials of novel GPE analogs in comparison to natural GPE in peripheral human whole blood (PHWB) cells for the frst time. To further evaluate the toxicity potentials of GPE analogs, these three novel GPEs were also assessed for their embryotoxic efects on cultured human pluripotent NT2 embryonal carcinoma cells. We assessed the in vitro efects of these peptidomimetics on cell viability using MTT and LDH release assays, antioxidant capacity using TAC assays, and DNA damage response using SCE, MN, and 8-OH-dG assays. In addition, NDI analysis was performed on cytostatic action potentials by novel GPEs.

Materials and Methods
Synthesis of GPE and GPE1-3 was performed as reported by Marinelli et al. [5]. All reagents for the synthesis of compounds were purchased by Sigma-Aldrich, MO, USA).

Experimental Design.
For assessing the biosafety of novel tripeptides, human peripheral blood samples were used. Te cell cultures were designed due to minor modifcations of the previously recommended protocol [16]. Heparinized blood samples were obtained from fve healthy male volunteers, nonsmoking and nonalcoholic and with no recent history of exposure to mutagens; aged 24-28 years (26.4 ± 1.8). Te study was approved by the local Ethics Committee (2019/ 03-30) and was in accordance with the Declaration of Helsinki and the International Conference on Harmonization for Good Clinical Practice. Written informed consent was obtained from all patients. Te 0.6 ml of blood was cultured in 6.4 ml of culture medium (Karyotyping Medium, Gibco, MA, USA) with 5.0 mg/ml of phytohemagglutinin (Sigma--Aldrich, USA). Te GPE and novel GPEs were added into the culture tubes at six diferent concentrations (0.1, 1, 10, 25, 50, and 100 μM) just before the incubation for 72 h.

Cytotoxicity
Testing. Cell viability was determined using MTT and LDH release assays. Te cell proliferation rate was determined by using commercially available MTT kits (Cayman Chemical Company, USA). Triton-X (%1, Sigma-Aldrich) was used as a positive control in cytotoxicity testing. In brief, cells were incubated with GPEs at 37°C for 72 h. MTT agent was added into the cultures for 3.5 h and generated formazan crystals were dissolved in dimethyl sulfoxide (Sigma-Aldrich). Finally, absorbances of samples in the well plates were read at 570 nm by an ELISA plate reader (Bio-Tek, Winooski, VT, USA) [17].
LDH release assay was carried out by using the commercially available LDH-cytotoxicity assay kit (Cayman Chemical, USA) due to the provider's guide. In brief, after the 72 h incubation period, the obtained 100 μL supernatant was transferred to a fresh well of 96-well plate containing 100 μL of the reaction mixture from the kit and incubated for 30 min at room temperature. After this extra incubation, the absorbances of the sample were read at 490 nm using a microplate reader (Bio-Tek). Finally, the total amount of released LDH was calculated using the equation % LDH released � (LDH amount in medium/total LDH amount) × 100 [13,18].

TAC Analysis.
Plasma samples, obtained by centrifugation of the whole blood cultures at 2000 g for 10 min 72 h after incubation with GPEs, were analyzed using commercially available TAC assay kits (Rel Assay Diagnostics, Gaziantep, Turkey). Ascorbic acid (C 6 H 8 O 6 , Sigma-Aldrich, 20 μM) was used as the positive control group [19].

Genotoxicity Testing.
For scoring SCE formations, 5bromo-2-deoxyuridine (BrdU, Sigma-Aldrich) was added at the initiation step. Exactly 70 h and 45 min after the beginning of the incubations, demecolcine (N-Diacetyl-Nmethylcolchicine, Sigma-Aldrich) was added into the culture tubes. After hypotonic treatment (0.075 M·KCl), three repetitive cycles of fxation with methanol/acetic acid solution (3 : 1, v/v), centrifugation, and resuspension steps, the suspension was dropped onto microscopic slides, following diferentially stained for examination of the SCE rates according to the fuorescence plus Giemsa (FPG) procedure. For each application, thirty metaphases (containing 42-46 chromosomes per cell) were scored and the values obtained were presented as SCEs per cell [20,21]. A negative control and a positive control (mitomycin-C (MMC), Sigma, 10 −7 M) were used to ensure the validity of the genotoxicity testing.
Te MN test was implemented by adding cytochalasin B (Sigma) after 44 h of culture initiation. At the end of the 72 h incubation period, the lymphocytes were fxed with ice-cold methanol/acetic acid (1 : 1, v/v). Te fxed cells were transferred to microscopic slides and then stained with Giemsa solution (Sigma). Te criteria for scoring MN were applied according to the previous protocol [22]. At least 2.000 binucleated cells were examined per treatment for the presence of one, two, or more MN. Cytostatic potentials of GPEs were also assessed by the nuclear division index (NDI). To determine NDI rates, 500 cells per application were examined for the presence of one, two, or more than two nuclei, and the nuclear division index (NDI) was determined as where 1N represents the number of cells with one nucleus, 2N with two nuclei, and >2N with more than two nuclei, and C represents the number of scored cells [23]. 8-hydroxy-2′-deoxyguanosine assay kits were provided from Cayman Chemical ® for measuring 8-OH-dG levels in vitro due to the provider's guide [24,25].

Statistical Analysis.
Te obtained data are presented by mean ± SD. from fve independent repetitions. For statistical evaluation, one-way analysis of variance (ANOVA) and Duncan's test were performed using the statistical program SPSS version 20.0. Statistical decisions were made with a signifcance level of 0.05.

Results
Te MTT and LDH release assays were applied to determine the cytotoxic response to diferent concentrations of GPEs (0.1 to 100 μM). Te positive control (as %1 Triton-X) caused signifcant (p < 0.05) decreases of cell viability in comparison to the untreated (negative control) group at rates of 75.14% and 72.89, in MTT and LDH release assays, respectively. By contrast with this status, when the cultured PHWB cells were exposed to all GPEs, the observed cell viability rates were not statistically diferent from the untreated cultures (Figures 2  and 3). Te cytotoxicity testing determined the noncytotoxic feature of GPE and its novel analogs.
We determined the TAC levels after application with GPE and GPE analogs. Te results of the TAC assay revealed that all examined tripeptides led to statistically signifcant (p < 0.05) elevations of antioxidant capacity levels as compared to untreated PHWB cells. Te established decreasing order of efectiveness of all tested tripeptides in elevating antioxidant capacity was GPE3 > GPE > GPE2 > GPE1 (Table 1).
Te results of the SCE testing were shown in Table 2 and  Table 2. No evidence of marked clastogenicity and aneugenic efect were observed following 72 h treatments with GPEs ( Figure 5). Tus, it is concluded that treatments with GPEs did not induce SCE or MN formations in cultured human peripheral blood lymphocytes.
As shown in Table 2, MMC led to a statistically signifcant reduction in the NDI as compared to the control groups. But GPEs at all tested concentrations did not decrease the NDI to the same extent as the negative control. Terefore, GPEs were concluded as noncytostatic even at the highest concentration (100 μM) for the 72 h treatment period.
Te level of 8-OH-dG in cultured human blood cells of the control and GPE, GPE1, GPE2, and GPE3 treated groups is shown in Figure 6. Te level of 8-OH-dG, which is a sensitive marker of oxidative DNA damage, was elevated after exposure to MMC. In fact, it was determined that MMC (at 10 −7 M) signifcantly elevated 8-OH-dG levels (about 4.5 folds) in human blood cultures after 72 h. Contrariwise, 8-OH-dG levels were not increased in human blood cell cultures that were applied with all GPEs at diferent concentrations ranging from 0.1 to 100 μM.

Journal of Toxicology
After treatment with CPA, this agent induced signifcant decreases in cell viability rates in NT2 cells as compared to untreated cells. In fact, 10 μg/ml of CPA induced cell death at rates of 82.55% and 86.80%, in MTT and LDH release assays, respectively. On the contrary, all tested concentrations of GPEs did not alter the cell viability rates in comparison to untreated cell cultures. Hence, GPEs were concluded to exert nonembryotoxic action in in vitro conditions.

Discussion
Te present results from cytotoxicity and genotoxicity testing clearly revealed that GPEs had good biosafety profles and were trouble-free from the toxicological point of view. As a matter of fact, in this study, all treatments with GPE and GPE analogs did not change the viability (by MTT and LDH release assays) of cultured human blood cells as compared to untreated cultures. In accordance with our fndings, the GPE did not alter the cell viability rates detected using ELISA and autoradiography methods in cultured rat hippocampal neurons [4,26]. Likewise, the performed MTT and LDH release assays propounded noncytotoxic properties of both lipoic acid conjugated GPE (LA-GPE) and natural GPE on diferentiated human SH-SY5Y neuroblastoma cell cultures [6]. Moreover, treatment with GPE (100 μM) elevated the proliferation rates of mouse embryonic neural stem cells [27].
In addition to the results of these in vitro studies, a previous in vivo investigation on the cortex and hippocampus regions of experimental rats revealed the neuroprotective action of natural GPE. Te cytoprotective action by GPE was attributed to an associated mechanism for central and systemic antiapoptotic efects of IGF-1 [28]. Te exact molecular mechanisms underlying the noncytotoxic and cytoprotective actions of GPE on human blood cells are still unclear. However, GPE was shown to inhibit apoptosis and promote cell survival in human corneal fbroblast cultures via stimulating a signaling pathway through phosphatidylinositol (PI) 3-kinase and then protein kinase B also known as Akt [29]. It was reported that activated Akt modulates fundamental cellular events involving cell proliferation and survival via phosphorylating variegated substrates [30]. Our results also revealed that not only GPE but also GPE1, GPE2, and GPE3 were noncytotoxic in human blood cells. Te observed feature of the novel GPE analogs might be associated with the activation of the PI3K/Akt pathway. In fact, PI3K/Akt pathway activation inhibited apoptosis of chicken splenic lymphocytes in vitro and stimulated their proliferation [31].
Systemic administration of GPE was found to activate nuclear erythroid 2-related factor 2 (Nrf2), a transcription factor-induced antioxidant response in Parkinson's modelgenerated animals [32]. Likewise, the neuroprotection by LA-GPE and GPE was reported to be related to the expression of antioxidant proteins in diferentiated human SH-SY5Y cells [12]. In supporting the presence of antioxidant properties by GPE, the present results proved that GPE along with its three analogs exhibited a high level of antioxidative action on human blood cells. Furthermore, the present fndings also indicated that GPE3 exhibited higher potency in supporting antioxidant capacity than GPE in human blood cells. Te diferences in antioxidant potentials of GPEs were highly contingent upon the sequence and composition of amino acids. Indeed, Gly residue makes a major contribution to antioxidant capacity because of that Gly is a proton-donating source and may combat free radicals. In addition, the single hydrogen atom of Gly may positively infuence the antioxidant capacity. Similar to Gly residue, the benefcial role of Pro residue in advocating antioxidant activity was previously put forward on peptide purifed from protein hydrolyzate of Saccharomyces cerevisiae [7,[33][34][35].
In a previous report, it was reported that cytoprotective action by Nrf2 was glutathione-dependent. And the increased cell death rates were directly associated with elevations in the amounts of reactive oxygen and nitrogen radicals that entailed DNA damage [36]. GPE, GPE1, GPE2, and GPE3 were found to be nongenotoxic (by SCE, MN, and 8-OH-dG testing) and noncytostatic (by NDI analysis) in human lymphocytes in the present study. At this point, the determination of cyto-and geno-toxicological profles is of critical importance for the use of GPE and its novel analogs, most particularly GPE3 as drug candidates. Nominately it was previously identifed that eptastigmine readily entered into the CNS and suppressed acetylcholinesterase for a prolonged period. But its observed haemato-toxic efects have intervened in eptastigmine usage in further clinical trials against Alzheimer' Disease [37,38].
A few investigations were undertaken to reveal the teratogenic and embryotoxic action potentials by diferent neuropeptides like L-prolyl-L-leucyl-glycinamide, cyclo(glycyl-Lleucyl)[Cyclo(Gly-Leu)] and cyclo(1-aminocyclo-pentanecarbonyl-L-alanyl) (cyclo(Acp-Ala)) and found positive results on in vivo models [39]. Interestingly, some antioxidant-featured amino acid derivatives L-methionine-or N-acetylcysteine-led to developmental toxicity in sea urchin embryos and gametes [40]. On the contrary, our in vitro results revealed that tested GPE and their GPE analogs exerted nonembryotoxic features. In this regard, the embryotoxicity or teratogenicity evaluation of peptide-based drug and drug candidates is critically important in the drug development process. So, it was reported that about 5% of the pregnancies used potential embryotoxic or teratogenic drugs during pregnancy and 0.66% of them were in the risk   [41]. Our fndings suggest that GPEs do not exert potentially embryotoxic action for impairing the pluripotent lineage of NT2 cells, and these data will contribute to gain new embryotoxicity insights into their potential for safe use during pregnancy.
In a conclusion, proven noncytotoxic, nongenotoxic, noncytostatic and nonembryotoxic features of GPE analogs are worthwhile exploring further and have high potentials for improving the development of novel disease-modifying agents in the treatment of variegated disorders. Table 3.    Journal of Toxicology 7