Microwave assisted synthesis of 2-amino-4-chloro-pyrimidine derivatives: Anticancer and computational study on potential inhibitory action against COVID-19

2.1 Materials

Pure grade reagents were utilized, procured from Sigma Aldrich. SMP10 (Stuart) melting point apparatus was used for the determination of melting points. IR spectra were acquired on a Perkin Elmer instrument employing ATR accessory. NMR (¹H and ¹³C) was performed via Bruker 400 MHz NMR spectrometers. “δ“, ppm (parts per million) was used to report proton chemical shifts while “J”, coupling constant was denoted by Hz (Hertz) and the terms singlet “s“, doublet “d”, triplet “t“, multiplet “m” and broad “br“ were used. TLC on silica gel Al foil was visualized under UV lamp.

2.2 General procedure for the synthesis of 2-amino-4-chloro-pyrimidine derivatives (1–7)

2 mmol of 2-amino-4-chloro-pyrimidine was weighed and transferred into microwave (CEM, Discovery, USA) reaction vial; 1 mL of anhydrous propanol was added to it and keeps stirring at room temperature. 2 mmol of different substituted amine was added to the reaction vial. After stirring 200 µL triethylamine was also introduced. Reaction was performed at 120–140 °C for 15–30 min and monitored by TLC. After cooling the obtained precipitate was dispersed in saturated sodium bicarbonate in water for the product to be extracted using ethyl acetate. Anhydrous sodium sulfate (Na₂SO₄) was used for drying of extract, following filtration, and being concentrated under reduced pressure to yield 1–7 derivatives (Scheme 1).

Close

Scheme 1

Synthesis of 2-amino-4-chloro-pyrimidine derivatives (1–7).

Export to PPT

2.3 Anticancer activity

2.4 Docking studies

3.1 Anticancer activity

The anticancer activities of synthesized derivatives (1–7) were tested against human colon colorectal (HCT116) and breast cancer (MCF7) cell lines by employing MTT assay. EC₅₀ of derivatives are presented in Table 1 and cell viability (%) is shown in the Figs. 1–2. Among the synthesized derivatives, derivative 6 having bromophenyl piperazine moiety at 4 position of the pyrimidine ring exhibited highest anticancer activity on both human colon colorectal (HCT116) with EC₅₀ 89.24 ± 1.36 µM and breast cancer (MCF7) cell lines with EC₅₀ 89.37 ± 1.17 µM. Second most active derivative was 1, with 4-methyl piperazine moiety and EC₅₀ values were 209.17 ± 1.23 µM and 221.91 ± 1.37 µM against human colon colorectal (HCT116) and on breast cancer cell lines (MCF7) respectively. While EC50 of doxorubicin as a positive control was 2 µM and 0.98 µM against human colon colorectal (HCT116) and on breast cancer cell lines (MCF7) respectively.

Table 1 EC₅₀ values for synthesized derivatives against HCT116 and MCF7 cell lines.

S.No.	R	EC50 (µM ± SD)
		HCT116	MCF7
1		209.17 ± 1.23	221.91 ± 1.37
2		272.48 ± 1.59	271.20 ± 1.27
3		368.20 ± 2.02	335.23 ± 1.47
4		334.54 ± 1.06	191.35 ± 1.22
5		349.41 ± 1.14	252.07 ± 1.78
6		89.24 ± 1.36	89.37 ± 1.17
7		458.89 ± 1.41	275.79 ± 1.51

Close

Fig. 1

% Cell viability against colon colorectal (HCT116) using pyrimidine derivatives (1–7).

Export to PPT

Close

Fig. 2

% Cell viability against breast cancer cell lines (MCF7) using pyrimidine derivatives (1–7).

Export to PPT

3.2 Molecular docking

Because of the tremendous therapeutic and biomedical importance, a series of seven pyrimidine substituted derivatives were synthesized and characterized in order to determine their binding efficiency versus the main protease of SARS-CoV-2 using molecular docking and ADMET studies. The molecular interaction results of synthesized pyrimidine derivatives (1–7) with 3CLpro (PDB ID: 6LU7) analyzed by docking experimentation are shown in Table 2 and Figs. 3 and 4. The results revealed that most of the pyrimidine derivatives had superior or equivalent affinity for the 3CLpro, as determined by binding energy scores. Derivative 6 was the top candidate with a binding energy docking score of −8.12 kcal/mol along with inhibition constant of 1.11 µM. 6 interacted with the 3CLpro via formation of three hydrogen bonds (bond lengths 1.98–3.74 Å). Amino acids Phe140 and Asn142 were crucially involved in the hydrogen bonds formation. His41, Met49, Tyr54, Leu141, His163, His164, Met165, Glu166, His172, Asp187, Arg188 amino acid residues participated in hydrophobic interaction (Table 2) It was also observed that Met165 formed pi-sulfur bond, His42 formed Pi-Pi stacking while Met49 was involved in Pi-Alkyl interaction (Fig. 4f1). The derivative 6 was closely followed by 7 with binding energy score of −7.83 kcal/mol and inhibition constant 1.83 µM having two hydrogen bond formations after the involvement of Phe140 and Asn142 and bond length of 2.06–2.93 Å were noted. Leu141, His172, His163, Glu166, Gln189, Arg188, Asp187, Tyr54, His41 amino acid residues participated in hydrophobic interaction (Table 2). The amino acid residues like Cys145, Met49 and Pro52 were involved in Alkyl/pi-alkyl interaction while Met165 was form Pi-Sulfur bond (Fig. 4g1). The binding energy scores of derivative 3, 4 and 5 were −6.61, −7.03 and −7.06 and kcal/mol, respectively, with the formation of five, two and five hydrogen bonds together with hydrophobic interactions. It is interesting to note that derivative 3, 4, 5, 6 and 7 exhibited better binding affinity than the control standard Remdesivir drug, which exhibited binding energy of −6.41 kcal/mol, inhibition constant of 19.91 µM and His41, Asn142, Gly143, Cys145, Met165, Glu166, Leu167, Pro168, His172, Gln189, Thr190, Gln192 amino acid residues participated in hydrophobic interaction (Table 2).

Close

Fig. 3

a, b, c and d displaying the 3D interaction of SARS-CoV2 main protease (PDB ID: 6LU7) with compound 1, 2, 3 and 4. Compounds were shown by ball stick pattern in the center and main protease were shown in light grey color. Close interacting residues are shown by stick pattern with dark blue color. a1, b1, c1 and d1 displaying 2D representation of hydrophobic interaction with amino acid residues. Hydrogen bonds are shown by green color dotted lines. 3D and 2D graphics were produced by Discovery Studio Visualizer 2020.

Export to PPT

Close

Fig. 4

e, f, g and h displaying the 3D interaction of SARS-CoV2 main protease (PDB ID: 6LU7) with compound 5, 6, 7 and standard Remdesivir. Compounds were shown by ball stick pattern in the center and main protease were shown in light grey color. Close interacting residues are shown by stick pattern with dark blue color. e1, f1, g1 and h1 displaying 2D representation of hydrophobic interaction with amino acid residues. Hydrogen bonds are shown by green color dotted lines. The 3D and 2D graphics were produced by Discovery Studio Visualizer 2020.

Export to PPT

3.3 ADME, drug-likeliness and toxicity results

ADME prediction data obtained from Swiss ADME are represented in Table 3. All the derivatives shown high GI absorption. It was found that except derivatives 1 and 3, all derivatives shown BBB permeability. Most of the derivatives shown negative log Kp values. Drug-likeness analysis revealed that all derivatives have zero violation for the required parameters of Lipinski’s rule of five (Lipinski et al., 2001) (Table 4). All the derivatives shown molecular weight in the ranged of 192.36–334.21 g/mol. No violation of rotatable bonds was observed, and the values were found between 1 and 2 bonds in comparison with standard cutoff value < 9 bonds. Observed TPSA values were between found in the range of 55.04–84.06 Å². The n-octanol–water partition coefficient, log P_o/w, is the standard used for Lipophilicity (Pliska et al., 1996). Consensus Log P was ranged between 0.14 and 1.89 for all the derivatives (Table 4). All derivatives showed bioavailability score of 0.55 and had a synthetic accessibility score of 2.21 to 2.44, which was within the typical standard range (Table 4).

The BOILED-Egg model is used for the simultaneous assessment of two simple physicochemical characteristics i.e., brain access and gastrointestinal absorption of drugs/chemicals and presented in Fig. 5. Further the toxicity analysis performed using pkCSM server (Pires, et al., 2015) demonstrated no AMES toxicity and the Max. tolerated dose (Human) values was −0.03 and −0.018, respectively for derivate 6 and 7. The range of values was −0.018–0.354 for derivatives 3,4,5,6 and 7. Minnow toxicity analysis values for all derivatives were found ranged between 1.666 and 3.241 (Table 5).

Close

Fig. 5

showing BOILED-Egg graph of compounds 1–7 Where yellow part showing the points situated in BOILED-Egg’s yolk are molecules predicted to passively permeate through the blood–brain barrier (BBB), White part showing the points situated in BOILED-Egg’s white are molecules predicted to passively absorbed by the gastrointestinal tract, Blue dots showing molecules predicted to be effluated from the central nervous system by the P-glycoprotein. Red dots showing molecules predicted not to be effluated from the central nervous system by the P-glycoprotein.

Export to PPT

4.1 Chemistry

The microwave synthesis of pyrimidine derivatives (1–7) was done in a single step. 2-amino-4-chloro-pyrimidine and different substituted amine was transferred into microwave reaction vial and reacted at 120–140 °C for 15–30 min in the presence of anhydrous propanol triethylamine (monitored by TLC). It further involved the products dispersion in aqueous saturated sodium bicarbonate solution and products extraction with ethyl acetate. After drying over anhydrous sodium sulfate (Na₂SO₄), filtered products were subjected to reduced pressure for concentrated yield of 1–7 derivatives (Scheme 1).

4.2 Anticancer activity

The replacement of 4-methyl piperazine by 4-methyl piperidine (derivative 2) resulted in decrease of anticancer activity and EC₅₀ values were 272.48 ± 0.02 µM, 271.20 ± 0.06 µM for both cancer cell lines (HCT116 and MCF7) respectively. Substitution on pyrimidine ring with 4-Pyrimidin-2-yl-piperazine (Derivative 3), 4-flurophenyl piperazine (Derivatives 4), 2-flurophenyl piperazine (Derivative 5), and 4-chlorophenyl piperazine (Derivative 7) didn’t improve anticancer activity against HCT116 and derivative 7 exhibited least activity against human colon colorectal (HCT116). However, against MCF7, derivatives 4, 5, and 7 demonstrated improved activity with EC₅₀ 191.35 ± 1.22, 252.07 ± 1.78 and 275.79 ± 1.51 µM respectively. % cell viability as shown in Figs. 1–2 indicates the concentration dependent cell viability in all derivatives, at higher concentration (300 µM) % cell viability was observed lower (23–58 %) while at low concentration (50 µM) % cell viability was ranged 52–86 % against human colon colorectal (HCT116). While against breast cancer cell line (MCF7), % cell availability was 84–100 % at lower concentration, when concentration was increased to 300 µM, % cell viability was ranged 27–48 %.

4.3 Molecular docking

Pyrimidine derivatives have received a great deal of attention in the treatment of a broad spectrum of pharmacological conditions, including HIV, cardiovascular disease, diabetes, fungal, bacterial, viral, and cancer (Patil, 2018; Rane et al., 2021). It has been reported that pyrimidine derivatives are used as herbicides, antioxidants, bronchodilators, antipyretics, anti-leishmanials agents, as well as analgesics and anti-inflammatory medications (Rane et al., 2021). Because pyrimidine base is found in both DNA and RNA, therefore, it can be explored in the treatment of viruses that have both DNA and RNA as genetic materials (Sharma et al., 2014). The first step in drug development is to design the targeted therapeutic derivatives utilizing cheminformatics techniques. Over the last decade, simulated drug discovery approaches have surfaced as important tools in the drug development process, and they have been used to uncover protein inhibitors as well as analyze protein-drug and protein–protein interactions (Keretsu et al., 2020). The 3C-Like proteinase (3CLpro) of the SARS-CoV-2 has been characterized as a promising therapeutic target for COVID-19 treatment due to it plays an essential role in virus replication and post-translational modifications of replicase polyproteins.

The results of binding efficiency of 1–7 versus the main protease of SARS-CoV-2 revealed that most of the pyrimidine derivatives had superior or equivalent affinity for the 3CLpro, as determined by binding energy scores. Derivative 6 was the top candidate and interacted with the 3CLpro via formation of three hydrogen bonds (bond lengths 1.98–3.74 Å). Amino acids Phe140 and Asn142 were crucially involved in the hydrogen bonds formation. His41, Met49, Tyr54, Leu141, His163, His164, Met165, Glu166, His172, Asp187, Arg188 amino acid residues participated in hydrophobic interaction. It was also observed that Met165 formed pi-sulfur bond, His42 formed Pi-Pi stacking while Met49 was involved in Pi-Alkyl interaction (Fig. 4f1). The derivative 6 was closely followed by 7 having two hydrogen bond formations after the involvement of Phe140 and Asn142 and bond length of 2.06–2.93 Å were noted. Leu141, His172, His163, Glu166, Gln189, Arg188, Asp187, Tyr54, His41 amino acid residues participated in hydrophobic interaction. The amino acid residues like Cys145, Met49 and Pro52 were involved in Alkyl/pi-alkyl interaction while Met165 was form Pi-Sulfur bond (Fig. 4g1). It is interesting to note that derivative 3, 4, 5, 6 and 7 exhibited better binding affinity than the control standard Remdesivir drug, which exhibited binding energy of −6.41 kcal/mol, inhibition constant of 19.91 µM and His41, Asn142, Gly143, Cys145, Met165, Glu166, Leu167, Pro168, His172, Gln189, Thr190, Gln192 amino acid residues participated in hydrophobic interaction (Table 2).

GLU-166 has been shown to be required for maintaining the right shape of the S1 pocket and the active conformation of the enzyme. In general, N-terminal residues in the protein formed a H-bond with GLU-166 of the promotor, forming the S1 pocket of the main protease, which serves as a catalytic pocket (Zhang et al., 2020). In the present study, it has been found that derivatives 1, 2 formed hydrogen bond with Glu166, while derivatives 3, 4, 5, 6, 7 formed van der waals interactions (Fig. 3a1-d1 and Fig. 4e1-g1). It has been also reported that the main protease formation is quite analogous to that of serine proteases. However, instead of a triad of catalytic residues, main protease has a catalytic dyad (involving His41 and Cys145) (Griffin, 2020). In this present work, it was found that His41 was involved in Alkyl/pi-alkyl interaction with derivatives 2, Pi cation with derivatives 4, Pi-Pi stacking with derivatives 6 and Van der waals interaction with derivatives 3, 5 and 7 (Fig. 3a1-d1 and Fig. 4e1-g1). Derivatives 1, 3 and 5 formed Pi-Sulfur bond with Cys145 during interaction. Cys145 also involved in van der waals interaction with derivatives 6 while it formed alkyl/pi alkyl bonds with derivatives 4 and 7 (Fig. 3a1-d1 and Fig. 4e1-g1). In the current study, the key amino acids residues of SARS-CoV2 main protease i.e., His41, Cys145, and Glu166 were found to be actively involved in the interaction with our designed derivative.

4.4 ADME, drug-likeliness and toxicity results

The Lipinski rule is commonly used to assess a derivatives potential as a pharmacological candidate. This rule filters simulation assessments and aids in screening potential therapeutic agents during the early drug designing stages, thus reducing time, exercise and costs in clinical drug development (Gombar et al., 2003; Hughes et al., 2011). The rule assesses some of the fundamental molecular features that a derivative contains, such as absorption, distribution, metabolism, and excretion, for a specific spectrum of derivatives and if a derivative contains at least two of the characteristics, it is considered to be safe (Bojarska, et al., 2020; Lipinski, 2004). ADME prediction data obtained from Swiss ADME demonstrated that all derivatives had high GI absorption. It was found that except derivatives 1 and 3, all derivatives shown BBB permeability and most of the derivatives are less skin permeable.

Drug-likeness analysis revealed that all derivatives have zero violation for the required parameters of Lipinski’s rule of five (Lipinski, et al., 2001) (Table 4). No violation of rotatable bonds was observed, and the values were found between 1 and 2 bonds in comparison with standard cutoff value < 9 bonds. Observed TPSA values were optimum as the recommended values should be between 20 and 130 Å² (Ertl, et al., 2000). The n-octanol–water partition coefficient, log P_o/w, is the standard used for Lipophilicity (Pliska, et al., 1996). Consensus Log P should not be higher than 6 (Mannhold, et al., 2009), while, in the present work, it was not higher than 1.89 for all the derivatives (Table 4).

The determination of a bioavailability score is required in order to calculate the likelihood that a candidate will have at least 10 % oral bioavailability in rats or significant Caco-2 permeability, and it should not be<0.25 (Martin, 2005) and our derivatives showed bioavailability score of 0.55. The predicted synthetic accessibility score should be ranged from 1 (very easy) to 10 (very difficult) (Ertl and Schuffenhauer, 2009; Ansari, et al., 2020) and all the derivatives had scores of 2.21 to 2.44, which was within the typical standard range (Table 4).

The BOILED-Egg model is used for the simultaneous assessment of two simple physicochemical characteristics i.e., brain access and gastrointestinal absorption of drugs/chemicals and these two pharmacokinetic behaviors are crucial at various stages of the drug discovery and development processes (Daina and Zoete, 2016) In the present study, it was observed that derivatives 2, 4, 5, 6 and 7 were expected to pass across the blood–brain barrier (BBB), whereas derivatives 1 and 3 were predicted to be absorbed passively by the gastrointestinal tract (Fig. 5). Further investigation found that the P-glycoprotein was unable to effluate the derivatives 1 and 2 from the central neurological system, however the derivatives 3, 4, 5, 6 and 7 were able to be effluated from the central neurological system (Fig. 5).

Further the toxicity analysis performed using pkCSM server (Pires, et al., 2015) demonstrated that all derivatives have shown no skin sensitization properties. The best identified derivatives were 6 and 7 which shows no AMES toxicity and the Max. tolerated dose (Human) values was −0.03 and −0.018, respectively for derivate 6 and 7 which is within the standard range i.e., the value smaller than/equal to 0.477 log(mg/kg/day) is considered low, whereas more than 0.477 log(mg/kg/day) is considered to be high. In this present analysis the values found for derivatives 3,4,5,6 and 7 were within the standard range. Minnow toxicity analysis values for all derivatives were found satisfactory with the values ranged between 1.666 and 3.241 because LC50 values below 0.5 mM (Log LC50 < -0.3) will be considered as high acute toxicity (Table 5).