Chiral Pyridine-3,5-bis- (L-phenylalaninyl-L-leucinyl) Schiff Base Peptides as Potential Anticancer Agents: Design, Synthesis, and Molecular Docking Studies Targeting Lactate Dehydrogenase-A

A series of branched tetrapeptide Schiff bases 3–6 were designed and synthesized from corresponding tetrapeptide hydrazide 2 as a starting material.In vitroevaluation of the synthesized compounds 4–6 against breast MCF-7 carcinoma cells identified their excellent anticancer potency, with IC50 ranging from 8.12 ± 0.14 to 17.55 ± 0.27 μM in comparison with the references, cisplatin and milaplatin (IC50= 13.34 ± 0.11and 18.43 ± 0.13 μM, respectively). Furthermore, all derivatives demonstrated promising activity upon evaluation of theirin vitroandin vivosuppression of p53 ubiquitination and inhibition assessment for LDHA kinase. Finally, molecular docking studies were performed to predict the possible binding features of the potent derivatives within the ATP pocket of LDHA in an attempt to get a lead for developing a more potent LDHA inhibitor with anti-proliferative potency.


Introduction
Peptides are short linear chains of amino acids (usually less than 50 AA in length) and are often stabilized by disulfide bonds [1]. Recently, there has been a noticeable increase in the number of publications covering the potential effects of peptides on lipid metabolism and blood pressure (BP), in addition to their antimicrobial, anti-inflammatory, analgesic, antioxidant, immunomodulatory, and anticancer activities [2][3][4].
Still, cancer is one of the most prominent causes of death in developed countries and there is a great demand for the discovery of new chemotherapies with more selectivity for cancer cells and less side effects [4,5]. The anticancer effects of peptides have been extensively explored with a raised number of approved peptide-based drugs [6][7][8]. The great potential of their antitumor activity has been exhibited through the action on multiple molecular pathways like enzymes involved in carcinogenesis with no genotoxicity [9].
Moreover, tumor-targeting peptides are effective alternative entities for the selective delivery of high doses of chemotherapeutic drugs or diagnostic agents to tumor sites while sparing normal tissues. Several peptide hormones have already been applied for tumor targeting, for example, the cyclic octapeptide analogue of somatostatin, octreotide, has been utilized for radiotargeting in neuroendocrine tumor [10]. Additionally, the linear peptide analogue of Luteinizing hormone-releasing hormone (LHRH), AN-152, has been used to target the LHRH receptor of breast cancer, ovarian cancer, and prostate cancer [11]. Recently, most peptidopyridine derivatives that have been synthesized consist of a wide range of pharmaceutical activities, including anticancer, analgesic, anti-convulsant, antiparkinsonian, analgesic, anti-inflammatory, and antimicrobial activities [12][13][14][15].
Schiff bases have been important candidates owing to their coordination chemistry, and they can be synthesized and used as ligands in building complex with different metal ions. On the other hand, Schiff bases have been interested with a broad spectrum of biological and pharmacological activities, such as antibacterial, DNA binding, anti HIV, Largicidal, antifertility, anticancer, and anti-inflammatory activities [16][17][18][19][20][21][22][23]. The broad-spectrum biological activities of Schiff bases have been attributed to the -C = N-imine bond, where the electrophilic carbon and nucleophilic nitrogen provide excellent binding characters with different nucleophiles and electrophiles, resulting in the inhibition of targeted diseases, enzymes, or DNA replication [24].
The lactate dehydrogenase (LDH) is one of the most prominent proteins and is responsible for the interconversion of lactate and pyruvate, parallel with the interconversion of NAD + and NADH. LDH is present in humans as three subunits-LDHA, LDHB, and LDHC [25]. It is clear that cancer cells mainly produce energy through glycolysis and LDHA is the final enzyme involved in this pathway [26]. As LDHA is a vital supporter of glucose metabolism in cancer cells and its increased level is a marker for many tumors, LDHA can be defined as an anticancer drug target [27]. Previous literatures reported that V-shape structures illustrate good LDHA inhibition, so molecular modification of the potentially identified LDHA inhibitors A-C ( Figure 1) might provide more effective hits [28][29][30].
In the present investigation and as a continuation of our previous work [31][32][33][34][35][36][37][38] in peptide and heterocyclic chemistry, chiral N3,N5-bis[(l-phenylalaninyl-l-leucinyl) substituted aryl or heterocyclic Schiff base]pyridine candidates were successfully designed, synthesized, and plausibly characterized by analytical techniques including IR, nuclear magnetic resonance (NMR), and mass spectroscopy evidences. In addition, their cytotoxicity effects on MCF-7 cancer cell lines, in vivo and in vitro inhibition of p53 ubiquitination, and lactate dehydrogenase-A were demonstrated in detail. Furthermore, molecular docking illustrates the binding affinity to LDHA kinase, which could facilitate the discovery of novel anticancer and LDHA inhibitory agents.

Evaluation of In vitro Anticancer Potentials Against Breast Cancer
Following the synthesis of the new bis-dipeptide derivatives, we evaluated the anticancer potentials of these derivatives in vitro against breast cancer cells (MCF-7) as well as normal non tumorigenic MCF-10A cells. The obtained results were expressed as IC 50 values in µM and were also compared for both cell lines with positive reference controls; that is cisplatin and milaplatin. The obtained results are presented in Figure 2. Generally, it can be seen that our newly synthesized bis-dipeptides have potential anticancer activities against MCF-7 cells. Furthermore, the obtained IC 50 values ranged from 8.12 ± 0.14 to 17.55 ± 0.27 µM for compounds 4b and 5a, respectively. Additionally, all compounds were much active than milaplatin (IC 50 = 18.43 ± 0.13 µM). Similarly, except for 5a, 5b, and 6a, all other compounds were more active than cisplatin (IC 50 = 13.34 ± 0.11 µM). The most active compound (4b) was more active by about 39.1% and 55.9% than cisplatin and milaplatin, respectively. Generally, the anticancer activities of the prepared compounds can be arranged in descending order as follows: 4b > 4a> 4c > 5e > 5d > 5c > 6b > 5b>6a > 5a. This arrangement can be related with the structure function relationship. Pyridine containing derivatives 4a-c were more active than benzene ring-containing ones (5a-d), which were in turn less active than the five-membered ring compounds (6a,b). Furthermore, the nitrogen-position in pyridine nucleus affected the activities where the 3-pyridyl >2-pyridyl >4-pyridyl ones. This may be attributed to the aromaticity and the location of the nitrogen atom in the pyridine ring. On the other hand, substitution on the benzene ring affected the anticancer activity (nitro 5e > bromo 5d > chloro 5c > methoxyl 5b > methyl 5a), owing to the difference in the electron withdrawing or donation capacity of the substituting group. Finally, the substitution with five-membered rings afforded notably increased activity in the thiophene derivative 6b than the furan one 6a, owing to the difference in electronegativity between the oxygen and sulfur atoms. Moreover, the larger size of the sulfur atom makes the thiophene ring more similar in size to a phenyl ring and gives it a more pronounced aromatic character, compared with the furan ring, thus more closely resembling the benzene ring of the active compounds. Bis-dipeptide Schiff bases have been previously reported to exhibit anticancer activities owing to their ionophoric properties, which may contribute to apoptosis induction [40,41].
It is worth mentioning that comparing IC 50 results between both MCF-7 and non-tumorigenic MCF-10A cells revealed that our newly synthesized compounds are far less toxic against normal cells. The IC 50 values obtained for the most and least potent derivatives (4b and 5a, respectively) were higher by about 142.7-and 145.9-fold, respectively, against MCF-10A. It is generally preferable that pharmaceutical candidates with potential anticancer properties should exhibit less toxicity against normal cells. The fact that normal cells are usually less affected by possible anticancer candidates may be attributed to the fact that, upon cancer initiation and development, cancer cells overcome normal growth regulation pathways present in normal cells, and thusthey tend to be affected by other different regulatory growth pathways, which are not evident in the case of normal cells.

In vivo Anti-Breast Cancer
In this section, we developed a breast cancer animal xenograft model for the evaluation of thein vivo anticancer potential of our newly prepared compounds. We chose the derivatives that showed the lowest IC 50 values in the previous in vitro investigation; namely, 4a, 4b, 4c, and 5e. The results obtained ( Figure 3) showed that all four tested compounds have a great potential to decrease tumor growth in animal model, where the decrease in tumor growth effect increased gradually with the treatment period up to eight days of treatment. At that point, the highest decrease in tumor growth obtained ranged from 81.3% to 74.2% for compounds 4a and 5e, respectively. After wards, the noticeable effects were more or less constant up to 14 days of treatment. However, by the end of the treatment period (20 days), the percentage of tumor growth inhibition decreased to about 75.4%, 73.1%, 70.5%, and 60.8% for compounds 4a, 4a, 4c, and 5e, respectively. Our results are in good agreement with those previously published on the in vivo anticancer potential of Schiff bases owing to apoptosis induction [42].

In vitro and In vivo Inhibition of p53
Several cancer prevention drugs depend mainly on the ubiquitination activities against p53 as a suppressor protein. When E3 ubiquitin protein ligase HDM2 binds to p53, the resulting mechanism is the inhibition of its ability as a transcription activator. Accordingly, the regulative mechanism is acting negatively. Therefore, blocking the p53 binding site on HDM2 provides for a better possible antitumorigenic molecule. Currently, murine double minute 2 (MDM2) is considered as the regulator of choice when it comes to investigatingp53 inhibition [43].
The results obtained ( Figure 4) showed that all prepared derivatives exhibited potential in vitro as well as in vivo suppression of p53 ubiquitination when compared with the positive control (diphenyl imidazole). The obtained in vitro IC 50 values ranged from 21.34 ± 0.13 to 44.33 ± 0.34 µM for compounds 4a and 5a, respectively. For thein vivo results, the IC 50 values ranged from 0.12 ± 0.0011 to 0.31 ± 0.0010 µM for compounds 4a and 6a, respectively. Furthermore, it can be seen that inhibitory effects of 4a were higher by about 12.2-and 15.7-fold compared with the positive control for in vitro and in vivo p53 ubiquitination, respectively.

Kinase Inhibition Studies
The newly synthesized compounds 4-6 were evaluated for their ability to inhibit LDHA in comparison with galloflavin as a known isoform nonselective LDHA inhibitor [44,45]. The choice of galloflavin as a reference was based on the fact that it is potent anticancer agent and commercially available [46][47][48][49][50]. The results presented in Figure 5 show that all investigated derivatives exhibited better LDHA inhibition compared with galloflalvin, where the IC 50 values ranged from 60.54 ± 0.56 to 141.56 ± 0.98 µM for compounds 5c and 5e, respectively. These values correspond to about 2.66-and 1.14-fold in increase in the inhibitory activity compared with that recorded for galloflavin IC 50 = 161.05 ± 0.32 µM).

Figure 5.
In vitro kinase inhibition activity of the tested compounds 4-6 against human lactate dehydrogenase-A.

Molecular Modeling Studies
In order to further understand and illustrate the interaction of compounds 4-6 with the LDHA active site, the docking modeling studies were done in an open loop conformation of the enzyme. The docking studies were developed using Molecular Operating Environment software, MOE 10.2008 software [51,52] with X-ray crystallographic structure of LDHA (PDB ID: 4ZVV) [53]. The root mean square deviation (RMSD) value of 8.93 was calculated via redocking of the co-crystallized ligand, GNE-140, into the pocket site of LDH-A with docking score energy of −9.73 kcal/mol. The docking scores of the screened compounds 4-6 were all in the range of −15.60 to −11.29 kcal/mol. Representations of the docking results of these compounds with LDHA are depicted in Table 1and Figures 6-8.   As the reported binding mode of GNE-140 with LDHA [53], the oxygen of the hydroxyl group binds to the active site via two hydrogen bond acceptors with the sidechains of Asn137 and His192 (distance: 2.75 and 2.74 Å), respectively. Additionally, the carbonyl oxygen forms H-bonding with the sidechain of Arg168 (distance: 2.99 Å) ( Figure 6).
The increased inhibitory activity of the newly synthesized derivatives 4-6 against LDHA enzyme could be partially explained by the evidence obtained from the docking studies (Table 1). By focusing on the highly potent inhibitors 4a and 5c (Figures 7 and 8), the essential residue Arg168 connected to the newly formed 2-pyridyl (in 4a) and 4-chlorophenyl moieties (in 5c) by arene-cation interactions. Moreover, the compound 4a was fitted nicely in the active pocket through two hydrogen bond interactions between nitrogen of 2-pyridyl moiety and oxygen of the amide group with the key amino acids His192 and Asn137 (distance: 2.66 and 2.54 Å), respectively.
The enhanced inhibitory potency of 5c may be attributed to its stabilization within the ATP binding pocket via further arene-cation interaction between the other 4-chlorophenyl and Arg98, and two H-bond acceptors between the oxygens of the two amide groups located at opposite sites around the parent pyridine ring with the sidechain of Tyr238 and the backbone of Thr247 (distance: 3.06 and 2.70 Å), respectively. Additionally, the essential amino acids Ala97, Asn137, and Pro138 were located between the two wings of the symmetrical side chains facing the central core (pyridine moiety) ( Figure 8).

Chemistry
Melting points were determined in open glass capillary tubes with an electro thermal digital melting point apparatus (model: IA9100), and are uncorrected. Elemental microanalysis for carbon, hydrogen, and nitrogen (Microanalytical Unit, NRC) was found within the acceptable limits of the calculated values. Infrared spectra (KBr) were recorded on a Nexus 670 FTIR Nicolet, Fourier Transform infrared spectrometer. Proton and carbon nuclear magnetic resonance ( 1 H-and 13 C NMR) spectra were run in (DMSO-d 6 ) on Jeol 500 MHz instruments. Mass spectra were run on a MAT Finnigan SSQ 7000 spectrometer, using the electron impact technique (EI). Analytical thin layer chromatography (TLC) was performed on silica gel aluminum sheets, 60 F 254 (E. Merck). Compound 2 was prepared according to our previous reported procedure [39].

In Vitro Cytotoxic Activity against MCF-7 Cancer Cells
Human breast cancer (MCF-7) cells, obtained from Sigma-Aldrich Chemie GmbH, Taufkirchen, Germany, were propagated in RPMI-1640 supplemented with 10% heat inactivated FBS, 2 mM L-glutamine, and 1% standard antibiotic solution. Cells were incubated in a 5% CO 2 humidified incubator at 37 • C and passaged bi-weekly. The in vitro anti-proliferative activity of the newly synthesized derivatives was assayed using the standard MTT technique [54][55][56].
The results were expressed as IC 50 . Experiments were repeated at least in triplicate, to obtain good reproducibility between replicate wells with standard errors below 10%. Furthermore, the cytotoxic effects of the prepared derivatives were evaluated against normal nonmalignant cells, non-tumorigenic MCF-10A, in order to find out if the synthesized derivatives have toxicity against normal cells. Additionally, the results were compared with reference compounds (cisplatin and milaplatin) as positive controls.

In Vivo Human Breast Cancer Xenograft
The breast cancer xenograft model protocol was approved by the Institutional Animal Use and Care Committee of the University of Alabama at Birmingham (50-01-05-08B). Female athymic pathogen-free nude mice (nu/nu, 4-6 weeks) were used. Firstly, MCF-7 xenografts were initiated by implanting pellets slowly releasing estrogen for two months (1.7mg 17β-estradiol/pellet) subcutaneously in the female nude mice. After 24 h, confluent MCF-7 cells were harvested, washed two times with serum-free medium, re-suspended, and injected subcutaneously (s.c.) (5 × 10 6 cells, total volume 0.2 mL) into the left inguinal area of the mice. Caliper measurement was used to measure tumor growth in two perpendicular diameters of the implant after 48 h, and its volume was determined. Mice grafted with MCF-7 were divided into different groups (7-10 mice/group). Untreated mice received the solvent only. Treated groups received different prepared derivatives as previously described [57].

In Vitro and In Vivo p53 Ubiquitination
Different prepared derivatives 4-6 were evaluated for their inhibitory potential against p53 ubiquitination according to the previously reported procedure [57].

LDHA Inhibition Assay
Lactate dehydrogenase assay kit (Abcam, ab102526) and recombinant human LDHA protein (Abcam, ab93699) were used to assess the inhibitory effect of the tested compounds. First, we generated an NADH standard curve for colorimetric detection by measuring the OD (450 nm) at different molar amounts of NADH. The linear regression equation of the curve was derivatized in Graphpad Prism. Tested compounds were dissolved in DMSO and their inhibitory activity was assessed at 140 and 300 µM strength. Galloflavin (Sigma, SML0776) at the same molar concentration was used as a known LDH inhibitor. The master reaction mix composed of LDH assay buffer and 10 ng human LDHA substrate per reaction was prepared. Tested compounds were added and the final reaction volume was adjusted to 50 µL. Absorbance at 450 nm was taken after 2-3 min (T-initial) and continued every minute for 118 min (T-final). The change in measurement over time (delta A450) was calculated as T-final minus T-initial. The amount of NADH generated by the assay between T-initial and T-final was deduced by comparing the A450 of each sample to the standard NADH standard curve. LDHA activity was calculated in milliunits per mL as the amount of NADH generated by the assay / (reaction time in minutes (118) x reaction volume in mL (0.005) [45].

Molecular Docking Studies
The three-dimensional X-ray structure of LDHA (PDB code: 4ZVV) [53] was chosen as the template for the modeling study of the screened compounds 4-6. The crystal structure was derived from the RCSB Protein Data Bank and the molecular docking procedure was performed using MOE, 10.2008 software following the reported procedure [51,52].

Conclusions
In conclusion, a series of new tetrapeptide Schiff bases 3-6were designed, synthesized, and evaluated for their antitumor activities against breast MCF-7 cell lines. These compounds exhibited potent antiproliferative activities in comparison with the standards, cisplatin and milaplatin. All of these derivatives were subjected to evaluation ofin vitroandin vivosuppression of p53 ubiquitination and inhibition assay for LDHA enzyme. Molecular docking study was further performed to illustrate the possible binding interactions with LDHA that might be a promising base for the further development of novel anticancer agents with excellent LDH inhibitory activity.
Author Contributions: The listed authors contributed to this work as described in the following: A.E.-G.E.A. and R.E.A.M. gave the concepts of the work, interpreted the results, did the experimental part, and prepared the manuscript; M.E.-N. analyzed the data; E.S.N. and E.A.E. contributed to the anticancer activity and the modeling studies. All authors read and approved the final manuscript.