Synthesis, Characterization and Biological Activity of Some Dithiourea Derivatives

Novel dithiourea derivatives have been designed as HIV-1 protease inhibitors using Autodock 4.2, synthesized and characterized by spectroscopic methods and microanalysis. 1-(3-Bromobenzoyl)-3-[2-({[(3-bromophenyl)formamido]methanethioyl}amino)phenyl]thiourea (10) and 3-benzoyl-1{[(phenylformamido)methanethioyl]amino}thiourea (12) gave a percentage viability of 17.9 ± 5.6% and 11.2 ± 0.9% against Trypanosoma brucei. Single crystal X-ray diffraction analysis of 1-benzoyl-3-(5-methyl-2-{[(phenylformamido)methanethioyl]amino}phenyl)thiourea (1), 3-benzoyl-1-(2-{[(phenylformamido)methanethioyl]amino}ethyl)thiourea (11), 3-benzoyl-1-{[(phenylformamido)methanethioyl]amino}thiourea (12) and 3-benzoyl-1-(4-{[(phenylformamido)methanethioyl]amino}butyl)thiourea (14) have been presented. 1-(3-Bromobenzoyl)-3-[2-({[(3-bromophenyl)formamido]methanethioyl}amino)phenyl]thiourea (10) gave a percentage inhibition of 97.03 ± 0.37% against HIV-1 protease enzyme at a concentration of 100 μM.


1. Synthesis and Spectroscopic Characterization
The phenyl thiourea derivatives are formed by the attack of the thione carbon of the starting benzoyl isothiocyanate by the two amino groups of the other starting molecule. Scheme 1 gives the synthetic pathway for the synthesis of the diamine derivatives.
Spectroscopic characterization. The dithiourea derivatives were obtained by the reaction of ammonium thiocyanate with the respective benzoyl chloride in acetone and heating under reflux for 2 h to yield the benzoyl isothiocyanate derivatives. The addition of the diamines and further heating under reflux for 3 h gave the final products 1-14. The 1 H NMR gave signals between δ 14.24 and 11.24 ppm for the NH proton of the amide. Table 1 gives the structures of all the synthesized compounds and their yields.
Aromatic protons gave signals between δ 11.72 and 7.01 ppm. In the 13 C NMR the thione signal was observed between δ 180.8 and 171.5 ppm whilst the carbonyl occurred between δ 168.3 and 161.0 ppm. Signals for aromatic carbons were observed between δ 159.0 and 113.3 ppm. The IR gave signals for the N-H stretching between 3440 and 3071 cm -1 , whilst the aliphatic C-H stretching occurred between 2993 and 2727 cm -1 . The C=S stretching was observed between 1683 and 1670 cm -1 , with the carbonyl stretching occurring between 1687 and 1640 cm -1 and the C=C stretching observed between 1597 and 1506 cm -1 .
The crystal structure of compound 11 was reported at 293 K, 12 but this work gives the crystal structure at 200 K. Both measurements gave a monoclinic space group P2 1 /c with two molecules in the unit cell. The cell parameters obtained at 273 K were slightly higher than the measurement at 200 K.
The crystal structure of compound 12 has been reported at 273 K, 13 whilst this work presents the crystal structure at 200 K. Both measurements gave a monoclinic space group P2 1 /n with two molecules in the unit cell and each molecule bonded to two molecules of dimethylsulfox-   (14) showing 50% probability displacement ellipsoids and atom labelling.
ide. The cell parameters for the determination at 273 K gave consistently higher values than the measurement at 200 K. The measurement at 273 K gave a lower density (1.347 g cm -3 ) than the measurement at 200 K (1.380 g cm -3 ) The crystal structure of compound 14 has been reported at 298 K, 2 whilst this work reports the crystal structure at 200 K. Both measurements gave a monoclinic space group P2 1 /c with two molecules in the unit cell. The cell parameters for the determination at 298 K gave consistently higher values than the measurement at 200 K. The measurement at 298 K gave a lower density (1.380 g cm -3 ) than the measurement at 200 K (1.416 g cm -3 )

Biological Studies
The compounds were tested for their cytotoxicity using Hela cells, and tested against HIV-1 protease with ritonavir as a positive control and Plasmodium falciparum strain 3D7 (20 µM) with chloroquine as a positive control. The compounds were also tested for their activity against Trypanosoma brucei (20 µM) with pentamidine as a positive control.
HIV-1 protease activity. Table 5 and Figure 6 give the HIV-1 screening results for the diamine derivatives of benzoyl isothiocyanate and their in silico results. The screening of the compounds 1-14 was completed at 100 µM and 10 µM of inhibitor and ritonavir, respectively. The predicted inhibition constant for compound 1 (4-methyldithiourea) was 0.19 µM whilst the HIV-1 assay gave a % inhibition of 17.69 ± 9.61%, compound 2 (unsubstituted) gave a predicted inhibition constant of 0.13 µM and a percentage inhibition of 10.30 ± 6.12%. For compound 3 (4-nitro derivative) a predicted inhibition constant of 0.47 µM and a percentage inhibition of 31.03 ± 0.42% were obtained whilst compound 5 (3-nitro derivative) gave predicted inhibition constant of 0.11 µM and a percentage inhibition of 32.68 ± 11.03%. Though the predicted inhibition constant of the 3-nitro derivative seems to be better than that of the 4-nitro, the percentage inhibition for both compounds are not too different, due to their interaction with solvent molecules which does not greatly change  their orientations in the active site. Compounds 4 (4-chloro derivative), 6 (3-methoxy derivative) and 8 (4-methoxy derivative) showed no activity in the HIV-1 protease assay with the predicted inhibition constants of 0.21, 1.90 and 0.81 µM, suggesting that in solution the methoxy substituent on the dithiourea makes the whole molecule inactive against HIV-1 protease at a concentration of 100 µM whilst a chloro substitution at position 4 makes the molecule ineffective at inhibiting HIV-1 protease because the chloro group interacts with the surrounding groups that interfere with its ability to fit well into the active site for effective inhibition of the protease. When the chloro group is attached at position 3, such as in compound 9 (3-chloro derivative), it gave a predicted inhibition constant of 0.06 µM which was the best predicted inhibition constant from the set and a percentage inhibition of 1.78 ± 11% confirming that the chloro group undergoes too much interaction with polar groups in solution hence the substantial departure from the predicted inhibition. Compound 7 (4-bromo derivative) gave predicted inhibition constant of 0.12 µM and a percentage inhibition of 29.62 ± 4.10% whilst compound 10 (3-bromo derivative) gave predicted inhibition constant of 0.095 µM and a percentage inhibition of 97.03 ± 0.37%. Compound 10 gave the best percentage inhibition of all the compounds. In these class of compounds, a bromo substituent at position 3 gives the best percentage inhibition among this class of compounds. The size of the bromo group allows the substituent to fit the active site for effective binding to the aspartate groups and the bridging water molecules in the active site. The other diamine derivatives gave lower predicted inhibition constants than those with the phenyl backbone. In the computation, the lack of rigidity in these molecules accounts for their lower predicted inhibition constants even though their protease % inhibition is comparable to the ortho-phenylenediamine derivatives.  Figure 7 gives the 2D representation of compound 10 in the protease active site. The presence of a polar group on this class of compounds improves the extent of interaction at the active site both in the docking studies and the bioassays making this class of compounds active against the protease.  Anti-malaria test. The bar graph ( Figure 8) and table (Table 6) below show the percentage parasite (Plasmodium falciparum strain 3D7) viability ± SD obtained after a 48 h incubation with 20 µM of the individual compounds 1-12. The anti-malaria test showed varying degrees of activity, with compound 6 giving the best percentage viability of 57.2 ± 1.3, whilst compound 11 was the least active with a percentage viability of 98.0 ± 13.4. Chloroquine was used as the standard in the antimalarial test.