Synthesis, antiviral activity and structure–activity relationship of 1-(1-aryl-4,5-dihydro-1H-imidazoline)-3-chlorosulfonylureas and products of their cyclization

Abstract

Novel 1-(1-aryl-4,5dihydro-1H-imidazoline)-3-chlorosulfonylourea derivatives 3a–3f were synthesized in the reaction of 1-aryl-4,5-dihydro-1H-imidazol-2-amines with chlorosulfonyl isocyanate. The second series of compounds 4a–4f was prepared from the respective 1-(1-aryl-4,5-dihydro-1H-imidazoline)-3-chlorsulfonylureas 3a–3f and 1,1′-carbonyldiimidazole (CDI). The selected compounds were tested for their activity against Herpes simplex virus and coxsackievirus B3 (CVB3). It was determined that three derivatives, i.e 3d, 4a and 4d are active against Herpes simplex virus (HSV-1). Compounds 3d and 4c are active against CVB3. Their favorable activity can be primarily attributed to their low lipophilicity values. Moreover, the lack of substituent in the phenyl moiety or 4-methoxy substitution can be considered as the most beneficial for the antiviral activity.

• Antiviral activity
• 1-(1-Aryl-4,5dihydro-1H-imidazoline)-3-chlorosulfonylourea derivatives
• Coxsackievirus B3,
• Herpes simplex
• structure–activity relationship

Introduction

Viral infections are a permanent health problem of mankind. More and more often used therapeutic methods like organ or bone marrow transplantations, when patients are subject to strong immunosuppression, result in numerous viral infections caused by viruses from the Herpesviridae family. Herpes simplex virus type l (HSV-1) infection can cause several clinical conditions such as keratitis, cutaneous herpes and encephalitis1–3. These pathologies may result from a primary infection or, alternatively, from a reactivation of a latent infection. Human herpesviruses cause mild or asymptomatic infections in patients with properly functioning immunological system4.

The illness is more serious in patients with deteriorated cellular immunity, e.g. human immunodeficiency virus (HIV) infected patients who must receive chronic treatment with antiviral agents favoring the selection of resistant variants5. Infectious complications can include central nervous system inflammation, pneumonia and congenital malformation. A number of nucleoside analogues, especially the guanosine analogue acyclovir have been developed as antiherpetic agents. The therapeutic limitation of these nucleoside analogues is that drug resistant strains develop readily through mutations in viral genes for thymidine kinase and polymerase6. There are many drugs active against herpesviruses now, but despite their high selectivity they have many defects, like numerous side-effects7. Therefore, the continuous search for new compounds as antiviral agents in the HSV therapy is urgently needed.

Myocarditis is defined as inflammation of the heart muscle and is commonly associated with viral infections. Coxsackieviruses, adenoviruses and parvovirus B19 have been implicated as causes of myocarditis. However, the main pathogen associated with this condition appears to be coxsackievirus B3 (CVB3)8. CVB3 is a positive-stranded RNA virus belonging to Picornaviridae. CVB3 infection can lead to both acute and chronic viral myocarditis in infants, children, young adults and immunocompromised individuals. Myocarditis caused by CVB3 often contributes to the development of inflammatory dilatative cardiomyopathy (DCMi), and most patients with DCMi eventually require heart transplantation. Besides heart infections, CVB3 can be responsible for chronic inflammatory diseases of the pancreas and the central nervous system9^,10. CVB3 infections are increasing risk for public health and specific antiviral treatment for CVB3 is not available at the moment. Recently, N-benzenesulfonyl sophocarpinic acid/ester and sophocarpinol derivatives were synthesized and tested for anti CVB3 activity. Among the tested compounds, sophocarpinol exerted the most promising activity against not only CVB3 but also CVB1, CVB2, CVB5 and CVB6 with IC₅₀ ranging from 0.62 to 3.63 μM (SI from 46 to 275), indicating a broad-spectrum anti-CVB activity10. Extract from Selaginella moellendorffii Hieron and its main constituent amentoflavone also exhibit anti-CVB3 activity, both in vitro by preventing cytopathic effect formation in HEp-2 cells and in vivo by reducing mean viral titres in the heart and kidneys as well as mortality of CVB3 infected mice11. It was recently reported that, oxysterol-binding protein (OSBP), a PI4P-binding protein that shuttles cholesterol between membrane compartments, is implicated as another host factor for enterovirus replication12. OSW-1 (3β,16β,17α-trihydroxycholest-5-en-22-one16-O-{O-(2-O-(4-methoxybenzoyl)-β-d-xylopyranosyl)-(1→3)-2-O-acetyl-α-arabinopyranoside, a natural compound extracted from the bulbs of the Ornithogalum saundersiae, and itraconazole, a well-known antifungal agent, are inhibitors of viral RNA replication by targeting oxysterol-binding protein (OSBP) and can be potentially used as broad-spectrum inhibitors of enteroviruses12^,13.

In this study, we present the evaluation of antiviral activity of N-substituted derivatives of 1-arylimidazolidyn-2-ylideneurea against the SV-114^,15 and CVB3. The rationale of this work can be summarized as follows: (1) the design of compounds was based on previously reported imidazo[1,2-a][1,3,5]triazines16 with antiviral activity (Figure 1) and aryl(heteroaryl)sulfonyl ureas17; (2) the set of substituents was selected based on our previous experience enabling to explore structure–activity relationship. In this study, we present synthesis, antiviral activity against the HSV-1 and CVB3 viruses and structure–activity relationship studies for the 1-(1-aryl-4,5dihydro-1H-imidazoline)-3-chlorosulfonyloureas 3a–3f and products their cyclization 4a–4f (Figure 2).

Figure 1. Previously reported antiviral compound16.

Figure 2. The scheme of synthesis of the investigated compounds.

Methods

Chemistry

All commercial reagents and solvents were purchased from Sigma-Aldrich (Sigma Aldrich Corp., St. Louis, MO) and used without purification. Reactions were routinely monitored by thin-layer chromatography (TLC) in silica gel (60 F₂₅₄ plates Merck, Darmstadt, Germany) and the products were visualized with ultraviolet light of 254 nm wavelength. All NMR spectra were acquired on a Bruker AVANCE III 300 MHz spectrometer (Bruker Bioscience, Billerica, MA) equipped with BBO Z-gradient probe. Spectra were recorded at 25 °C using DMSO as a solvent with a non-spinning sample in 5  mm NMR-tubes. MS spectra were recorded on Bruker microTOF-Q II and processed using Compass Data Analysis software. The elementary analysis was performed with the application of Perkin-Elmer analyzer (Perkin Elmer Inc., Waltham, MA). Melting points were determined with a Boetius apparatus (Jena, Germany).

General procedure for the synthesis of 1-(1-aryl-4,5-dihydro-1H-imidazoline)-3-chlorosulfonylureas (3a–3f)

1-Aryl-4,5-dihydro-1H-imidazol-2-amines (1a–1f) (0.01 mol) were dissolved in 25 mL of dichloromethane under the atmosphere of dry nitrogen and added to the solution chlorosulfonyl isocyanate 2 1.41 g (0.01 mol) dissolved in 25 mL of dichloromethne. The mixture was shaken for 24 h at room temperature. Solvent was removed by distillation and the rubber-like residue was treated with warm propan-2-ol. The solid product was filtrated off and purified by crystallization from propan-2-ol.

1-(1-Phenyl-4,5-dihydro-1H-imidazoline)-3-chlorosulfonylurea (3a)

From a general procedure with 1.61 g of 1a and 1.41 g of 2, obtaining 2.0 g of 3a (66% yield), white crystalline solid, mp 200–202 °C; ¹H NMR (DMSO-d₆, 300 MHz) δ = 9.45 (s, 1H, NH); 8.62 (s, 1H, NH); 7.34–6.95 (m, 5H, Ar–H); 4.35 (dd, 2H, CH₂, J = 6.7/J′ = 5.8 Hz); 4.56 (dd, 2H, CH₂, J = 6.8/J′ = 5.6 Hz); ¹³C NMR (DMSO-d₆, 75 MHz): δ = 170.6 (C = O); 159.6 (C = N); 126.6, 126.1, 124.9, 122.1, 114.9, 113.5, (Ar–C); 46.3 C3 (CH₂); 39.7 C2 (CH₂); EIMS m/z 303.2 [M⁺H]⁺. HREIMS (m/z): 302.1049, (calcd for. C₁₀H₁₁ClN₄O₃S 302.75); Anal. Calcd for C₁₀H₁₁ClN₄O₃S C, 39.67; H, 3.66; S, 10.59; Cl, 11.71. Found C, 39.63; H, 3.56; N, 18.39; S, 10.64; Cl, 11.80.

1-[1-(4-Methylphenyl-4,5-dihydro-1H-imidazoline]-3-chlorosulfonylurea (3b)

From a general procedure with 1.75 g of 1b and 1.41 g of 2, obtaining 2.59 g of 3b (82% yield), white crystalline solid, mp 206–208 °C; ¹H NMR (DMSO-d₆, 300 MHz) δ = 9.19 (s, 1H, NH); 8.72 (s, 1H, NH); 7.51–7.05 (m, 5H, Ar–H); 4.24 (dd, 2H, CH₂, J = 6.8/J′ = 5.7 Hz); 4.52 (dd, 2H, CH₂, J = 6.8/J′ = 5.5 Hz); 1.82 (s, 3H, CH₃); ¹³C NMR (DMSO-d₆, 75 MHz): δ = 170.8 (C = O); 165.3 (C = N); 128.6, 128.2, 125.1, 124.1, 116.4, 115.94, (Ar–C); 47.3 C3 (CH₂); 42.6 C2 (CH₂); 18.5 CH₃; EIMS m/z 317.3321 [M⁺H]⁺. HREIMS (m/z): 316.8 [M⁺] (calcd. for C₁₁H₁₃ClN₄O₃S 316.78); Anal. Calcd for C₁₁H₁₃ClN₄O₃S C, 41.70; H, 4.13; N, 17.68; Cl, 11.19; S, 10.12. Found C, 41.61; H, 4.21; N, 17.61, Cl 11.08; S, 10.04.

1-[1-(2-Methoxyphenyl-4,5-dihydro-1H-imidazoline]-3-chlorosulfonylurea (3c)

From a general procedure with 1.91 g of 1c and 1.41 g of 2, obtaining 2.46 g of 3a (74% yield), white crystalline solid, mp 183–184 °C; ¹H NMR (DMSO-d₆, 300 MHz) δ = 9.11 (s, 1H, NH); 8.79 (s, 1H, NH); 7.44–6.91 (m, 4H, Ar–H); 4.31 (dd, 2H, CH₂, J = 6.6/J′ = 5.5 Hz); 4.50 (dd, 2H, CH₂, J = 6.9/J′ = 5.4 Hz); 2.12 (s, 3H, OCH₃); ¹³C NMR (DMSO-d₆, 75 MHz): δ = 171.1 (C = O); 163.6 (C = N); 129.3, 129.1, 128.3, 125.1, 122.8, 120.6, (Ar–C); 45.8 C3 (CH₂); 39.7 C2 (CH₂); 24.5 OCH₃; EIMS m/z 333.1 [M⁺H]⁺. HREIMS (m/z): 332.1540 [M⁺] (calcd. for C₁₁H₁₃ClN₄O₄S 332.78), Anal. Calcd for C₁₁H₁₃ClN₄O₄S: C, 39.70; H 3.93; N, 16.83; Cl, 10.65; S, 9.63. Found C, 39.61; H, 3.98; N, 16.91; Cl, 10.75; S, 9.54.

1-[1-(4-Methoxyphenyl-4,5-dihydro-1H-imidazoline]-3-chlorosulfonylurea (3d)

From a general procedure with 1.91 g of 1d and 1.41 g of 2, obtaining 1.39 g of 3d (42% yield), white crystalline solid, mp 197–199 °C; ¹H NMR (DMSO-d₆, 300 MHz) δ = 9.22 (s, 1H, NH); 8.56 (s, 1H, NH); 7.11–6.83 (m, 4H, Ar–H); 7.31 (s, 1H, NH); 4.30 (dd, 2H, CH₂, J = 7.1/J′ = 5.9 Hz); 4.61 (dd, 2H, CH₂, J = 6.9/J′ = 5.8 Hz); 2.68 (s, 3H, OCH₃); ¹³C NMR (DMSO-d₆, 75 MHz): δ = 171.2 (C = O); 161.5 (C = N); 128.6, 123.1, 122.6, 120.1, 115.4, 113.1, (Ar–C); 45.8 C3 (CH₂); 39.3 C2 (CH₂); 23.2 OCH₃; EIMS m/z 333.1 [M⁺H]⁺. HREIMS (m/z): 332.2349 [M⁺] (calcd for C₁₁H₁₃ClN₄O₄S 332.78), Anal. Calcd for C₁₁H₁₃ClN₄O₄S: C, 39.70; H, 3.93; N, 16.83; Cl, 10.65; S, 9.63. Found C, 39.68; H, 3.85; N, 16.93; Cl, 10.55; S, 9.71.

1-[1-(3-chlorophenyl-4,5-dihydro-1H-imidazoline]-3-chlorosulfonylurea (3e)

From a general procedure with 1.96 g of 1e and 1.41 g of 2, obtaining 1.52 g of 3e (45% yield), white crystalline solid, mp 141–142 °C; ¹H NMR (DMSO-d₆, 300 MHz) δ = 9.41 (s, 1H, NH); 8.93 (s, 1H, NH); 7.60–7.15 (m, 4H, Ar–H); 4.26 (dd, 2H, CH₂, J = 6.8/J′ = 5.5 Hz); 4.54 (dd, 2H, CH₂, J = 6.4/J′ = 5.6 Hz); ¹³C NMR (DMSO-d₆, 75 MHz): δ = 171,3 (C = O); 162.4 (C = N); 129.6, 123.1, 122.5, 121.1, 119.5, 117.1, (Ar–C); 46.9 C3 (CH₂); 41.7 C2 (CH₂); EIMS m/z 338.2 [M⁺H]⁺. HREIMS (m/z): 337.1289 [M⁺] (calcd. for C₁₀H₁₀Cl₂N₄O₃S 337.20) Anal. Calcd for C₁₀H₁₀Cl₂N₄O₃S: C, 35.61; H, 2.98; N, 16.61; Cl, 21.03; S, 9.51. Found C, 35.69; H, 3.05; N, 16.79; Cl, 21.07; S, 9.60.

1-[1-(4-chlorophenyl-4,5-dihydro-1H-imidazoline]-3-chlorosulfonylurea (3f)

From a general procedure with 1.96 g of 1f and 1.41 g of 2, obtaining 2.16 g of 3f (64% yield), white crystalline solid, mp 160–162 °C; ¹H NMR (DMSO-d₆, 300 MHz) δ = 9.17 (s, 1H, NH); 8.82 (s, 1H, NH); 7.36–7.05 (m, 4H, Ar–H); 4.39 (dd, 2H, CH₂, J = 6.5/J′ = 5.6 Hz); 4.76 (dd, 2H, CH₂, J = 6.9/J′ = 5.6 Hz); ¹³C NMR (DMSO-d₆, 75 MHz): δ = 170.9 (C = O); 162.7 (C = N); 129.8, 129.1, 125.9, 122.8, 122.4, 115.3, (Ar–C); 47.8 C3 (CH₂); 40.4 C2 (CH₂); EIMS m/z 338.2 [M⁺H]⁺. HREIMS (m/z): 337.1633 [M⁺] (calcd. for C₁₀H₁₀Cl₂N₄O₃S 337.20) Anal. Calcd. for C₁₀H₁₀Cl₂N₄O₃S: C, 35.61; H, 2.98; N, 16.61; Cl, 21.03; S, 9.51. Found C, 35.51; H, 2.94; N, 16.68, Cl, 21.15; S, 9.59.

General procedure for synthesis 1-aryl-6-chlorosulfonyl-5,7(1H)dioxo-2,3-dihydroimidazo[1,2-a][1,3,5]triazines (4a–4f)

A mixture containing appropriate 1-(1-aryl-2-amine-4,5-dihydro-1H-imidazoline)-3-chlorsulfonylurea (3a–3f) (0.01 mol) and 0.01 mol (1.6 g) of 1,1′-carbonyldiimidazole (CDI) was dissolved in 50 mL of DMF (N,N-dimethlformamide) and heated under reflux for 6 h. The solvent was removed by the low-pressure evaporation and the resulting solid was crystallized from methanol:propan-2-ol (1:1) mixture.

1-Phenyl-6-chlorosulfonyl-5,7(1H)dioxo-2,3-dihydroimidazo[1,2-a][1,3,5]triazine (4a)

From a general procedure with 3.03 g of 3a and 1.6 g of 1,1′-carbonylodiimidazole, obtaining 1.68 g of 4a (51% yield), white crystalline solid, mp 180–182 °C; ¹H NMR (DMSO-d₆, 300 MHz) δ = 7.40–6.85 (m, 5H, Ar–H); 4.26 (dd, 2H, CH₂, J = 6.9/J′ = 5.6 Hz); 4.65 (dd, 2H, CH₂, J = 6.8/J′ = 5.5 Hz); ¹³C NMR (DMSO-d₆, 75 MHz): δ = 171.2 (C = O); 170.5 (C = O); 162.6 (C = N); 129.7, 129.1, 127.1, 121.8, 121.4, 119.7, (Ar–C); 45.8 C3 (CH₂); 39.9 C2 (CH₂); EIMS m/z 329.7 [M⁺H]⁺. HREIMS (m/z): 329.1330 [M⁺] (calcd for C₁₁H₉ClN₄O₄S 328.75) Anal. Calcd for C₁₁H₉ClN₄O₄S: C, 40.19; H, 2.76; N, 17.04; Cl, 10.78; S, 9.75. Found C, 40.11; H, 2.62; N, 17.09; Cl, 10.85; S, 9.64.

1-(4-Methylphenyl)-6-chlorosulfonyl-5,7(1H)dioxo-2,3-dihydroimidazo[1,2-a][1,3,5]-triazine (4b)

From a general procedure with 3.16 g of 3b and 1.6 g of 1,1′-carbonylodiimidazole, obtaining 2.70 g of 4b (79% yield), white crystalline solid, mp 198–200 °C; ¹H NMR (DMSO-d₆, 300 MHz) δ = 7.12–6.67 (m, 4H, Ar–H); 4.41 (dd, 2H, CH₂, J = 7.1/J′ = 5.9 Hz); 4.69 (dd, 2H, CH₂, J = 6.9/J′ = 5.6 Hz); 1.56 (s, 3H, CH₃); ¹³C NMR (75 MHz, DMSO-d₆): δ = 172.1 (C = O); 170.7 (C = O); 161.5 (C = N); 129.6, 128.4, 128.4, 127.2, 122.3, 114.6, (Ar–C); 44.6 C3 (CH₂); 38.2 C2 (CH₂); 14.6 CH₃; EIMS m/z 343.7 [M⁺H]⁺. HREIMS (m/z): 342.3245 [M⁺] (calcd for C₁₂H₁₁ClN₄O₄S 342.78); Anal. Calcd for C₁₂H₁₁ClN₄O₄S: C, 42.04; H, 3.23; N, 16.34; Cl, 10.34; S, 9.35. Found C, 42.11; H, 3.31; N, 16.39; Cl, 10.11; S, 9.31.

1-(2-Methyoxyphenyl)-6-chlorosulfonyl-5,7(1H)dioxo-2,3-dihydroimidazo[1,2-a][1,3,5]-triazine (4c)

From a general procedure with 3.59 g of 3c and 1.6 g of 1,1′-carbonylodiimidazole, obtaining 2.16 g of 4c (64% yield), white crystalline solid, mp 212–214 °C; ¹H NMR (DMSO-d₆, 300 MHz) δ = 7.54–6.95 (m, 4H, Ar–H); 4.33 (dd, 2H, CH₂, J = 6.9/J′ = 5.5 Hz); 4.56 (dd, 2H, CH₂, J = 6.8/J′ = 5.4 Hz); 3.22 (s, 3H, OCH₃); ¹³C NMR (DMSO-d₆, 75 MHz): δ = 171.4 (C = O); 171.4 (C = O); 159.6 (C = N); 127.6, 127.1, 124.9, 121.1, 120.6, 119.1, (Ar–C); 46.8 C3 (CH₂); 41.5 C2 (CH₂); 19.9 (OCH₃); EIMS m/z 359.3 [M⁺H]⁺. HREIMS (m/z): 358.1255 [M⁺] (calcd for C₁₂H₁₁ClN₄O₅S 358.78), Anal. Calcd for C₁₂H₁₁ClN₄O₅S: Calcd C, 40.17; H, 3.09; N, 15.62; Cl, 9.88; S, 8.94. Found C, 40.11; H, 3.18; N, 15.49; Cl, 9.80; S, 8.84.

1-(2-Methyoxyphenyl)-6-chlorosulfonyl-5,7(1H)dioxo-2,3-dihydroimidazo[1,2-a][1,3,5]-triazine (4d)

From a general procedure with 3.59 g of 3d and 1.6 g of 1,1′-carbonylodiimidazole, obtaining 2.16 g of 4d (64% yield), white crystalline solid, mp 219–220 °C; ¹H NMR (DMSO-d₆, 300 MHz) δ = 7.20–6.85 (m, 4H, Ar–H); 4.39 (dd, 2H, CH₂, J = 6.8/J′ = 5.6 Hz); 4.61 (dd, 2H, CH₂, J = 6.8/J′ = 5.7 Hz); 3.23 (s, 3H, OCH₃); ¹³C NMR (DMSO-d₆, 75 MHz): δ = 171.8 (C = O); 170.7 (C = O); 160.2 (C = N); 125.1, 123.9, 122.6, 121.8, 121.56, 118.2, (Ar–C); 45.6 C3 (CH₂); 40.1 C2 (CH₂); 18.6 (OCH₃); EIMS m/z 359.8 [M⁺H]⁺. HREIMS (m/z): 358.1120 [M⁺] (calcd. for C₁₂H₁₁ClN₄O₅S 358.78); Anal. Calcd for C₁₂H₁₁ClN₄O₅S: C, 40.17; H, 3.09; N, 15.62, Cl, 9.88; S, 8.94%. Found C, 40.05; H, 3.19; N, 15.68; Cl, 9.95; S, 8.78.

1-(3-Chlorophenyl)-6-chlorosulfonyl-5,7(1H)dioxo-2,3-dihydroimidazo[1,2-a][1,3,5]-triazine (4e)

From a general procedure with 3.27 g of 3e and 1.6 g of 1,1′-carbonylodiimidazole, obtaining 2.21 g of 4e (61% yield), white crystalline solid, mp 201–202 °C; ¹H NMR (DMSO-d₆, 300 MHz) δ = 7.60–7.15 (m, 4H, Ar–H.); 4.20 (dd, 2H, CH₂, J = 6.8/J′ = 5.8 Hz); 4.56 (dd, 2H, CH₂, J = 6.8/J′ = 5.6 Hz); ¹³C NMR (DMSO-d₆, 75 MHz): δ = 173.4 (C = O); 171.7 (C = O); 160.5 (C = N); 126.6, 123.4, 122.3, 120.9, 119.7, 114.6, (Ar–C); 45.9 C3 (CH₂); 42.1 C2 (CH₂); EIMS m/z 364.1 [M⁺H]⁺. HREIMS (m/z): 363.9861 [M⁺] (calcd. for C₁₁H₈Cl₂N₄OS 363.20); Anal. Calcd for C₁₁H₈Cl₂N₄OS 363.20: C, 36.37; H, 2.22; N, 15.43; Cl, 19.52; S, 8.82. Found C, 36.44; H, 2.35; N, 15.51; Cl 19.40; S, 8.75.

1-(4-Chlorophenyl)-6-chlorosulfonyl-5,7(1H)dioxo-2,3-dihydroimidazo[1,2-a][1,3,5]-triazine (4f)

From a general procedure with 3.27 g of 3f and 1.6 g of 1,1′-carbonylodiimidazole, obtaining 2.16 g of 4f (53% yield), white crystalline solid, mp 231–232 °C; ¹H NMR (DMSO-d₆, 300 MHz) δ = 7.58–7.05 (m, 4H, Ar–H); 4.31 (dd, 2H, CH₂, J = 7.1/J′ = 5.9 Hz); 4.59 (dd, 2H, CH₂, J = 6.8/J′ = 5.8 Hz); ¹³C NMR (DMSO-d₆, 75 MHz): δ = 171.8 (C = O); 170.1 (C = O); 158.4 (C = N); 125.6, 125.1, 123.9, 123.2, 120.5, 117.1, (Ar–C); 46.7 C3 (CH₂); 40.4 C2 (CH₂); EIMS m/z 364.2 [M⁺H]⁺. HREIMS (m/z): 363.1423 [M⁺] (calcd. for C₁₁H₈Cl₂N₄OS 363.20), Anal. Calcd for C₁₁H₈Cl₂N₄OS: Calcd C, 36.37; H, 2.22; N, 15.43; Cl, 19.52; S, 8.82. Found C, 36.50; H, 2.45; N, 15.39; Cl, 19.47; S, 8.89.

Molecular modeling

The investigated compounds were modeled using the LigPrep protocol from the Schrödinger Suite18. In order to sample different protonation states of ligands in physiological pH, Epik module was used19. The compounds were further optimized using Hartree–Fock approach and 6–31 g (d,p) basis set of Spartan 1020. Parameters to evaluate drug-likeness were calculated using VegaZZ version 3.0.121 (number of atoms), Discovery Studio version 3.122 (molar mass, number of rings, lipophilicity, number of rotatable bonds), ACDLabs (molar refractivity, number of hydrogen bond donors and acceptors) and the Schrödinger Suite (a number of rigid bonds) as described previously23–26. ADMET parameters were calculated with Discovery Studio 3.1 (solubility, blood–brain permeation) or Osiris Property Explorer27 (toxicity risks) as previously described23–26. The prediction of toxicity is based on a pre-computed set of structural fragment that give rise to toxicity alerts in case they are encountered in the investigated compound27. These fragment lists were obtained by rigorously shreddering all compounds of the RTECS database known to responsible for a certain toxicity class27. During the shreddering, any molecule was first cut at every rotatable bonds leading to a set of core fragments27. These were applied to reconstruct all possible bigger fragments being a substructure of the investigated compound. Next, a substructure search process determined the occurrence frequency of any fragment (core and constructed fragments) within all compounds of that toxicity class. It also determined these fragment's frequencies within the structures of more than 3000 commercial drugs27. Based on the assumption that drugs present on the market are free of toxic effects, any fragment was considered a risk factor if it occurred often as substructure of harmful compounds but never or rarely in traded drugs27. The predicted toxicity of 1.0 means the lack of a toxic effect and decreasing value means increasing risk of toxicity. For structure–activity relationship studies, HOMO and LUMO energies, lipophilicity and polarizability were calculated with Discovery Studio 3.123 HOMO and LUMO orbitals as well as a map of the electrostatic potential (ESP) onto a surface of the electron density were visualized with ArgusLab28. Molecular surface area, polar surface area, molecular volume and ovality were calculated with VegaZZ21.

Antiviral activity

Antiviral activity assay

Antiviral activity assay were similar for HSV-1 and CBV3. After 24 h incubation, the cell culture was infected with viruses in the dose of 100 TCID₅₀/mL. After 1 h incubation at 37 °C, the suspension of the viruses was removed and the media with 2% of serum together with the tested compounds in the non-toxic concentration were added to the cell cultures. The viruses diluted in the culture media without tested compounds were used as a control. Acyclovir (Polpharma SA, Starogard Gdański, Poland) and ribavirin (Meduna Pharma GmbH, Isernhagen, Germany) were used as reference compounds. After 48 h incubation at 37 °C, the cells were frozen and after thawing, the viruses were titrated in the Vero cell culture. The cytopathic effect (CPE) of the viruses was examined by a light microscope and the titre of virus was estimated according to the Reed–Muench method29. Viral titres were determined by tissue culture infection dose (TCID₅₀) assays.

Cells and viruses

The Vero cell (ECACC No. 84113001-established from the kidney of a normal adult African Green monkey) was used in the experiment. The media in the culture (Dulbecco’s Modified Eagle Medium – DMEM, Sigma-Aldrich, St. Louis, MO), 100 U/mL of penicillin and of streptomycin (Polfa Tarchomin, Warsaw, Poland). The cell culture was incubated at 37 °C in the 5% CO₂ atmosphere. For antiviral activity of examined compounds the HSV-1 (ATCC No. VR-260) and CVB3 (ATCC No. VR-30) from the American Type Culture Collection were used. The viruses were propagated in the Vero cell culture. Virus stocks were stored at −70 °C until used.

Cytotoxicity assay

Compounds were dissolved in dimethyl sulfoxide – DMSO (POCH, Gliwice, Poland) in the concentration of 50 mg/mL and further diluted with a complete test medium. About 100 µl of the Vero cell culture prepared was seeded into 96-well plastic plates (Becton Dickinson and Company, Franklin Lakes, NJ) at a cell density 1.5 × 10⁴ cells per well. After 24 h incubation at 37 °C the media were removed and the cells were treated with a solution of the examined substance diluted in the media with 2% of serum. The cells were submitted to a series of compound concentrations, from 1000 µg/mL to 1.9 µg/mL. Two-fold serial dilutions of compounds were added to the cells in triplicates. The cell cultures were incubated for 72 h at 37 °C in the 5% CO₂ atmosphere.

Cytotoxicity of tested compounds was estimated with the use of the MTT method, described by Takenouchi and Munekata30. The MTT method is a quantitative colorimetric toxicity test, based on the transformation of yellow, soluble tetrazolium salts (3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide) to purple-blue insoluble formasane. This process occurs naturally in mitochondria of living cells. After 72 h incubation with compounds, cell cultures were supplemented with 10 µl per well of 5 mg/mL MTT (Sigma-Aldrich, St. Louis, MO) stock in PBS (BIOMED, Lublin, Poland), and the incubation was continued for 4 h at 37 °C. Then, 100 µl of aqueous solution containing 50% dimethylformamide (POCH, Gliwice, Poland) and 20% SDS (Sigma-Aldrich, St. Louis, MO) to solubilize the insoluble formasane precipitates produced by MTT was added. After the all-night incubation, the absorbance was measured by the Epoch plate reader (BioTek, Winooski, VT) at two wavelengths – 540 and 620 nm. Based on the test results, the cytotoxic concentration (CC₅₀), which is the amount of tested substance that is required to reduce the number of viable cells by 50% compared to the control culture, was determined and was calculated using the Gen 5 2.01.14 software (BioTek, Winooski, VT). Cell viability (%) was calculated as (A_540/620 of the treated/A_540/620 of the control) × 100. When the CC₅₀ value was calculated, non-toxic concentrations of examined compounds were selected to assess their antiviral activity. The investigation was carried out in triplicate.

Results and discussion

Chemistry

New derivatives 3a–3f were obtained in the reaction of 1-aryl-4,5-dihydro-1H-imidazol-2-amines 1a–1f with chlorosulfonyl isocyanate 2 in the dichloromethane solution and under nitrogen atmosphere (Figure 2). The second series of compounds 4a–4f was obtained from the respective 1-(1-aryl-4,5-dihydro-1H-imidazolin)-3-chlorosulfonylureas 3a–3f and 1,1′-carbonyldiimidazole (CDI) in DMF (N,N-dimethlformamide) solution.

Estimation of drug-likeness

The descriptors applied for estimation of drug-likeness are presented in Table 1. Drug-likeness was assessed using Lipinski’s rule as well as the placement of the investigated compounds in the chemical space determined by the databases of the pharmacologically active compounds (CMC, Comprehensive Medicinal Chemistry Database, containing about 7000 compounds and MDDR, MACCS-II Drug Data Report, containing about 100 000 compounds) according to the methodology of PREADMET31 service as described previously23–26. Regarding Lipinski’s rule, all the compounds possess the molar mass below 500, the number of hydrogen bond donors below 5, the number of hydrogen bond acceptors below 10 and the lipophilicity below 532.

Table 1. Parameters for drug-likeness estimation.

Compounds	Molecular mass	logP	HBD	HBA	No. of atoms	Molar refractivity	No. of rings	No. of rigid bonds	No. of rotatable bonds
3a	302.73	0.97	2	7	30	71.85	2	28	3
3b	316.76	1.46	2	7	33	76.28	2	31	3
3c	332.76	0.95	2	8	34	77.66	2	31	4
3d	332.76	0.95	2	8	34	77.66	2	31	4
3e	335.98	1.63	2	7	30	76.45	2	28	3
3f	337.18	1.63	2	7	30	76.45	2	28	3
4a	328.73	1.57	0	8	30	76.57	3	30	2
4b	342.75	2.06	0	8	33	80.99	3	33	2
4c	358.75	1.56	0	9	34	82.38	3	33	3
4d	358.75	1.56	0	9	34	82.38	3	33	3
4e	363.17	2.24	0	8	30	81.17	3	30	2
4f	363.17	2.24	0	8	30	81.17	3	30	2

HBD – a number of hydrogen bond donors; HBA – a number of hydrogen bond acceptors.

Regarding subsequent criteria of drug-likeness, most compounds collected in the CMC database has lipophilicity from −0.4 to 5.6, molar refractivity in the range of 40–130 the number of atoms from 20 to 70 and molar mass from 160 to 48023–26^,33. All the investigated compounds fulfill the first three criteria. Concerning the compounds in MDDR database, the drug-like substances have the number of rings equal or greater than 3, the number of rigid bonds equal or greater than 18, and the number of rotatable bonds equal or greater than 623–26^,34. Compounds 3a–3f have too few rings and all the compounds have too few rotatable bonds. The criterion concerning rigid bonds is fulfilled by all the compounds.

Prediction of ADMET properties

In order to facilitate the selection of compounds for antiviral activity assessment, some ADMET parameters were calculated (Table 2). The plot presented in Figure 3 confirms that most of the tested compounds possess reasonably favorable ADMET properties. Comparing the plot in Figure 3 with lipophilicity values from Table 1 and polar surface areas from Table 5, it can be concluded that compounds from 3c and 3d have less favorable blood–brain permeation properties. All compounds are well absorbed (Figure 3) and well soluble in water as they have of logS over −4 (solubility expressed in mol/dm³)27. All the compounds were predicted to have low mutagenicity risk and possible irritating effects (0.8), however no risk of tumorigenicity and no reproductive effects were predicted (1.0; Table 2).

Figure 3. ADMET properties of the studied compounds.

Table 2. Solubility and toxicity of the investigated compounds.

Compounds	logS	Mutagenicity	Tumorigenicity	Irritating effects	Reproductive effects
3a	−2.470	0.8	1.0	0.8	1.0
3b	−2.969	0.8	1.0	0.8	1.0
3c	−2.650	0.8	1.0	0.8	1.0
3d	−2.625	0.8	1.0	0.8	1.0
3e	−3.300	0.8	1.0	0.8	1.0
3f	−3.294	0.8	1.0	0.8	1.0
4a	−3.269	0.8	1.0	0.8	1.0
4b	−3.760	0.8	1.0	0.8	1.0
4c	−3.266	0.8	1.0	0.8	1.0
4d	−3.237	0.8	1.0	0.8	1.0
4e	−3.923	0.8	1.0	0.8	1.0
4f	−3.915	0.8	1.0	0.8	1.0

S – solubility [mol/dm³].

Antiviral activity determination

The obtained compounds 3d and 4a–4d have been studied as potential antiviral agents against HSV-1 and CVB3 and revealed their high efficacy as inhibitors of virus reproduction. The selection of compounds for antiviral activity determination was justified by availability from synthesis as well as our earlier experience. Acyclovir was used as a reference compound for HSV-1 and ribavirin for CVB3. The influence of 3d and 4a–4d on the Vero cell culture after incubation for 72 h is presented in Table 3.

Table 3. Cytotoxicity of compounds 3d, 4a, 4b, 4c, 4d.

	Cell line/CC50 (µg/mL)*
Compounds	Vero
3d	840.93 ± 147.53
4a	809.50 ± 35.68
4b	819.77 ± 28.89
4c	90.80 ± 11.11
4d	650.07 ± 132.25
Acyclovir	>2000
Ribavirin	>2000

CC₅₀ is the cytotoxic concentration required to reduce the number of viable cells by 50%.

*Mean ± SD values come from three independent experiments.

DMSO used as an eluent for examined compounds in the tested concentration had no toxic effect on cell cultures. All compounds were evaluated for their cytotoxicity on cell line by a standard MTT assay. CC₅₀ values of compounds 3d, 4a, 4b and 4d were contained within the range of 650.07–840.93 µg/mL, except for compound 4c which showed cytotoxicity with CC₅₀ of 90.08 µg/mL. Compounds were tested for in vitro antiviral activity using the cytopathic effect (CPE) inhibitory assay. The results are presented in Table 4.

Table 4. Antiviral activity of the compounds against HSV-1 and CVB3.

Compounds	Concentration [µg/mL]	HSV-1 [TCID50/mL]*	CVB3 [TCID50/mL]*
3d	625	4.69 ± 0.39	7.70 ± 0.24
	800	3.62 ± 0.87	6.25 ± 1.55
4a	500	4.09 ± 0.00	7.17 ± 0.00
	625	4.20 ± 0.24	8.17 ± 0.00
4b	500	4.82 ± 0.35	8.83 ± 0.00
	625	5.08 ± 0.67	8.83 ± 0.00
4c	31	6.38 ± 0.97	8.50 ± 0.00
	62	6.38 ± 0.97	5.34 ± 1.41
4d	250	4.89 ± 0.29	8.33 ± 0.00
	500	3.90 ± 0.16	8.50 ± 0.00
Acyclovir	2000	0	–
Ribavirin	2000	–	0
Control	0	6.38 ± 0.97	8.33 ± 0.32

Data represent the mean ± SD of at least 2–3 independent experiments. TCID₅₀ – 50% Tissue Culture Infectious Dose.

*The virus titres are shown in log.

Table 5. Molecular descriptors for structure–activity determination.

Compounds	Surface (Å^2)	PSA (Å^2)	Volume (Å^3)	Ovality	HOMO (eV)	LUMO (eV)	Polarizability
3d	552.8	101.1	253.5	1.67	−9.33	−2.27	30.79
4a	486.8	90.6	236.5	1.52	−9.63	−2.28	30.35
4b	503.0	90.6	251.6	1.57	−9.46	−2.25	32.10
4c	523.9	99.5	262.4	1.58	−9.62	−2.13	32.65
4d	526.5	99.5	263.9	1.58	−9.52	−2.32	32.65

Non-cytotoxic concentrations of 31, 62, 250, 500, 625 and 800 µg/mL were used for testing the antiviral activity of compounds. The research demonstrated that compound 3d in the concentration of 800 µg/mL influenced the HSV-1 and CVB3 replication by reducing the virus replication level by 2.76 log and 2.08 log, which resulted in reducing the titre by 43.3 and 25%, respectively. The compound 4a in all tested concentrations (500 and 625 µg/mL) demonstrated the antiviral activity. This derivative caused the decrease in the titre of HSV-1 by 2.29 log and 2.18 log, respectively (which corresponds to the following percentage levels of inhibition: 35.9 and 34.2%). In the case of compound 4c, tested in the concentration of 62 µg/mL, the titre of CVB3 was decreased by 2.99 log (35.9%). Whereas compound 4d was active only against HSV-1 it caused the decrease in the virus titre by 2.48 log (38.9%). Compound 4b did not demonstrate any significant antiviral activity. Thus, there is a need of further research to estimate the influence of 3d and 4a–4c on the HSV-1 and CVB3 inactivation, inhibition of viral adsorption and penetration. Prospectively, study of the mechanisms underlying their antiviral activity is necessary for the development of antiviral therapeutics to treat patients infected with HSV-1 and CVB3.

Structure–activity relationship

Molecular descriptors for structure–activity determination are presented in Table 5. The active compound 3d (active against both viruses) compared to inactive compound 4b is characterized by greater surface area, greater polar surface area and also bigger volume. This compound has also the biggest ovality and the smallest lipophilicity value, the biggest HOMO energy and small polarizability in the whole set. In the series of compounds 4a–4d, the active compound 4a (active against Herpes simplex virus) is characterized by the smallest surface area, the smallest volume, molecular weight, polarizability and one of the smallest polar surface area. Simultaneously, it has one of the smallest lipophilicity value. In contrast, another compound active against Herpes simplex virus, 4d has rather big surface area, volume, molecular weight, polarizability and polar surface area but it has still low lipophilicity value. It can be thus generalized that low lipohilicity value favors activity against Herpes simplex in both series. Furthermore, lack of the substituent in phenyl group or 4-methoxy substitution can be considered as the most beneficial for the antiviral activity against this virus. Compound 4c was active against CVB3, which can be also attributed to low lipophilicity value. Figure 4 depicts HOMO and LUMO orbitals for most active compounds 3d and 4a whereas Figure 5 presents maps of electrostatic potential for these derivatives. It can be concluded that for compound 3d HOMO orbital is mainly located on the phenyl ring and imidazole system while LUMO orbital on the chlorosulfonyl group. In the case of compound 4a, HOMO orbital is located on the phenyl group and LUMO orbital in the chlorosulfonyl group. The negative charge is gathered on heteroatoms, mainly on nitrogen atoms, oxygen atoms of the chlorosulfonyl group (for 3d and 4a) and the methoxy group of 3d.

Figure 4. HOMO (A, C) and LUMO (B, D) orbitals for 3d (A, B) and 4a (C, D).

Figure 5. The map of the electrostatic potential (ESP) onto a surface of the electron density for 3d (A) and 4a (B).

Conclusions

In this study, we obtained 12 novel N-substituted derivatives of 1-arylimidazolidyn-2-ylideneurea and tested selected compounds for their activity against Herpes simplex virus and CVB3. It was determined that four derivatives, i.e. 3d, 4a, 4c and 4d are active against these viruses. Their favorable activity in comparison to other tested compounds can be primarily attributed to their low lipophilicity values. Moreover, the lack of substituent in the phenyl moiety or 4-methoxy substitution can be considered as the most beneficial for the antiviral activity. In the subsequent studies, we will work on the molecular mechanism of antiviral activity.

Acknowledgements

The paper was developed using the equipment purchased within the project “The equipment of innovative laboratories doing research on new medicines used in the therapy of civilization and neoplastic diseases” within the Operational Program Development of Eastern Poland 2007–2013, Priority Axis I modern Economy, operations I.3 Innovation promotion.

Declaration of interest

Part of the calculations of this study was performed under a computational grant by Interdisciplinary Center for Mathematical and Computational Modelling (ICM), Warsaw, Poland, grant number G30-18 and under resources and licenses by CSC, Finland.

The research was partially performed during the postdoctoral fellowship of Agnieszka A. Kaczor at University of Eastern Finland, Kuopio, Finland under Marie Curie fellowship.