Design, synthesis and biological evaluation of sulfamethazine derivatives as potent neuraminidase inhibitors

Abstract

Aim: The purpose of this study is to design and synthesize a new series of sulfamethazine derivatives as potent neuraminidase inhibitors. Materials & methods: A sulfamethazine lead compound, ZINC670537, was first identified by structure-based virtual screening technique, then some novel inhibitors X1–X10 based on ZINC670537 were designed and synthesized. Results: Compound X3 exerts the most good potency in inhibiting the wild-type H5N1 NA (IC₅₀ = 6.74 μM) and the H274Y mutant NA (IC₅₀ = 21.09 μM). 150-cavity occupation is very important in determining activities of these inhibitors. The sulfamethazine moiety also plays an important role. Conclusion: Compound X3 maybe regard as a good anti-influenza candidate to preform further study.

GRAPHICAL ABSTRACT

A series of novel sulfamethazine neuraminidase inhibitors were designed and synthesized.

Plain language summary

Summary points.

• A novel lead compound ZINC670537 was screened out by means of structure-based virtual screening, molecular dynamic simulation and bioactivity test.

• A series of novel sulfamethazine neuraminidase inhibitors were designed and synthesized.

• Compound X3 has the best inhibitory activity, and its activity inhibiting the H274Y mutant NA is comparable to that of OSC.

• Molecular docking studies have shown that the 150-cavity plays an important role in discovery of novel sulfamethazine NA inhibitors.

• The sulfamethazine moiety interacts tightly with three key residues (Arg118, Arg292, Arg371) by forming hydrogen bondings, which is beneficial to improve activity of compound.

• The results of this work provide new insights into the design of more potent NA inhibitors.

1. Introduction

Influenza is one of the acute contagious respiratory diseases caused by influenza virus in the family Orthomyxoviridae (a group of RNA viruses) [1]. Influenza viruses are categorized as types A, B and C. These major types generally produce similar symptoms but are completely unrelated antigenically, so that infection with one type confers no immunity against the others. Types A and B viruses may cause an annual influenza pandemic, while C viruses have milder symptoms. Influenza A viruses are prone to antigenic mutations and are the main cause of seasonal or pandemic influenza worldwide. Influenza B virus, however, is more genetically stable than type A and the epidemics it causes are generally much less severe [2,3]. The World Health Organization reported that 290,000 to 650,000 people die from seasonal influenza every year [4]. The illness remains a great threat to global health, and the treatment or prevention of influenza is still a tough task.

Regarding the treatment and prevention of influenza, individual protection may be bolstered by injection of a vaccine containing two or more circulating influenza viruses. However, the prevention effect is not very ideal due to the difference of individual immunity and the strong ability of antigenic variation of influenza virus. Especially the efficacy of vaccines mainly depend on effective prediction of the infectious strains for each year, and the inaccurate predictions can lead to vaccination 25% less effective [5,6]. Therefore, the development of novel antiviral drugs against influenza infections is reasonably attractive in the event that a full-blown pandemic outbreak.

So far, three classes of anti-flu drugs had been approved by the US FDA, including M2 ion-channel inhibitors like amantadine and rimantadine, polymerase acidic protein (PA) inhibitors like baloxavir marboxil, and neuraminidase (NA) inhibitors like zanamivir, oseltamivir, peramivir and laninamivir octanoate. However, the clinical use for some approved antivirals had been hampered due to undesirable side effects and emergence of transmissible resistant variants [7,8]. Nevertheless, NA inhibitors have turned into the preferable or even the best choice in treating or preventing influenza in clinical use in many cases. Despite the success of NA inhibitors, drug-resistance issue is a growing concern in recent years due to their extensive clinical application. For example, the increased reports of oseltamivir-resistant strains of the influenza virus, such as the most frequent NA substitutions reported in H5N1 (H274Y) and H3N2 (E119V) subtypes have created some cause for concern [9]. Thus, the development of new NA inhibitors with highly efficiency, enhanced anti-drug resistance and good pharmacokinetic profiles is becoming increasingly urgent.

Neuraminidase is a crucial surface antigenic glycoprotein of the influenza virus, which exerts essential role in the life cycle of influenza viruses and had been confirmed to be an attractive target for anti-influenza drugs [10–13]. It can facilitate viral shedding via cleaving terminal sialic acid residues from glycoconjugates, which is crucial for virus replication and infectivity [10,11]. It also plays essential roles in the release of viral progeny from infected cells and promote the reproduction of the virus [12,13]. Anti-flu drugs with NA as a drug target can be categorized as the following types based on their structures: sialic acid derivatives [11,14], cyclohexene and cyclopentane derivatives [15,16], thiophene derivatives [17], triazole derivatives [18], acylhydrazone derivatives [19] and some natural products, etc. [20]. Recently, great interest had been focused on an open large cavity in Group-1 NAs, termed as ‘150-cavity’ [21], which is near the active site of NA and consisting of residues Gly147-Arg152. The 150-cavity is large enough to accommodate ligands with enhanced efficiency and affinity, making it a desired target for further drug modifications [22–24]. Many medicinal chemistry research teams had designed various NA inhibitors based on the 150-cavity. For instance, Zhang et al. [22] performed structure-based optimization of some N-substituted oseltamivir derivatives targeting 150-cavity to fight against oseltamivir-resistant N1-H274Y variant. Jia et al. [23] reported the design, synthesis and biological evaluation of a series of novel oseltamivir derivatives via the structural modifications at C₅-NH2 of oseltamivir targeting 150-cavity. Ju et al. [24] had designed and synthesized three series of novel oseltamivir amino derivatives to explore the druggable chemical space inside the 150-cavity of NAs. Some compounds in series C were demonstrated to be more potent than oseltamivir carboxylate (OSC) in inhibiting both wild-type and oseltamivir-resistant group-1 NAs. Therefore, as a promising binding site, the 150-cavity is of great significance for the development of novel and efficient NA inhibitors.

Sulfanilamide and pyrimidine structures exhibit obvious antimicrobial and biological activity and had been confirmed to have important applications in many drugs [25,26]. Zhang et al. [27] reported a design, synthesis and structure–activity relationship of new arylpyrazole pyrimidine ether derivatives as fungicides. Abd El-All et al. [28] found that a series of new thiophene[2,3-d]pyrimidine derivatives have significant inhibitory activity on NA of influenza A H3N2 virus. Wang et al. [29] discovered a sulfathiazole–amantadine hydrochloride cocrystal, which is the first codrug simultaneously comprising antiviral and antibacterial components.

Virtual screening analyses databases with computational technique to identify potential new lead compounds, which can be divided into structure-based virtual screening and ligand-based virtual screening. If the structures of the therapeutic targets are known, then structure-based virtual screening method such as protein ligand molecular docking may be chosen [30]. In the present work, a series of novel sulfamethazine derivatives were discovered as NA inhibitors targeting 150-cavity by using of structure-based virtual screening, molecular dynamic (MD) simulation, organic synthesis and biological activity determination. First, a new lead compound was screened out by the structure-based virtual screening method, and then the lead compound was performed modification to obtain a series of target compounds. Finally, ten new sulfamethazine compounds were synthesized and evaluated for their NA inhibition activities.

2. Chemistry materials & methods

All the chemical regents were commercially available regents without further purification. Reactions were monitored by thin layer chromatography (TLC) with precoated silica gel 60 F254. ¹H NMR and ¹³C NMR spectra were recorded on a Bruker AVANCE III (MA, USA) at 400 MHz, 100 MHz in DMSO-d₆ using tetramethylsilane (TMS; δ = 2.50 ppm) as internal standard. Chemical shifts are reported in δ (parts per million [ppm]) with TMS and coupling constants (J) in Hz. HRMS were recorded on a solariX 70 FTMS spectrometer (Bruker) using methanol and acetonitrile as the solvent. Data were acquired in the positive ion mode at resolving power of 100,000. Melting points were measured using a WRS-2A digital melting point apparatus (Shanghai Shenguang Instrument Co., Ltd., Shanghai China). Analytic HPLC was performed on Agilent technologies 1260 series (CA, USA) with water contains 0.1% CH₃COOH (Solvent B, 40%)/CH₃CN (Solvent A, 60%) as eluent and the targeted products were detected by DAD in the detection rang of 254–320 nm. Product purities were confirmed to be >95% by this method.

2.1. General procedure for synthesis of compound 2

A mixture of sulfamethazine (5.57 g, 20 mmol) and potassium carbonate (4.15 g, 30 mmol) in DMF (80.00 ml) was stirred at 0°C. Chloroacetyl chloride (2.39 ml, 30 mmol) was added dropwise to a cooled stirred mixture for 1.5 h. After completion of the reaction, the precipitate was filtered, washed by distilled water. The crude product was recrystallized by 95% aqueous ethanol to obtain the compound 2.

2.1.1. 2-chloro-N-(4-(N-(4,6-dimethylpyrimidin-2-yl)sulfamoyl)phenyl)acetamide (2)

White solid, yield 73.3%, ¹H NMR (400 MHz, DMSO-d₆) δ 8.96 (s, 1H), 7.92–7.85 (m, 2H), 7.76–7.70 (m, 2H), 7.01 (d, J = 0.9 Hz, 1H), 4.20 (s, 2H), 2.33 (s, 6H). ¹³C NMR (100 MHz, DMSO-d₆) δ 167.13, 162.56, 155.88, 142.28, 132.44, 128.77, 119.25, 116.22, 42.89, 23.35.

2.2. General procedure for synthesis of compound X10

A mixture of compound 2 (10.64 g, 30 mmol), potassium carbonate (4.15 g, 30 mmol) and 4,6-dihydroxy-2-mercaptopyrimidine (2.89 g, 20 mmol) in DMF (80.00 ml) was stirred at 60°C for 10 h. After completion of the reaction, the precipitate was filtered, washed by distilled water. The crude product was recrystallized by 95% aqueous ethanol to obtain the pure target compound X10.

2.2.1. 2-((4,6-dihydroxypyrimidin-2-yl)thio)-N-(4-(N-(4,6-dimethylpyrimidin-2-l)sulfamoyl)phenyl)acetamide (X10)

3-White solid, yield 25.6%, m.p. 250.9–252.3°C, purity 99.80%; ¹H NMR (400 MHz, DMSO-d₆) δ 8.86 (s, 1H), 7.99–7.89 (m, 2H), 7.85–7.71 (m, 2H), 6.73 (d, J = 7.8 Hz, 1H), 5.32 (s, 1H), 4.09 (s, 2H), 2.25 (s, 6H). ¹³C NMR (100 MHz, DMSO-d₆) δ 168.29, 167.87, 162.90, 156.62, 152.18, 142.97, 135.32, 129.69, 118.71, 113.98, 86.20, 36.34, 34.59. HRMS (ESI) calcd for C₁₈H₁₈N₆O₅S₂ [M+Na]⁺: 485.0672; Found: 485.0673.

2.3. General procedure for synthesis of compound 3

A mixture of sulfamethazine (5.57 g, 20 mmol) and potassium carbonate (4.15 g, 30 mmol) in DMF (80.00 ml) was stirred at 0°C. Oxalate monoethyl ester chloride (3.20 ml, 30 mmol) or malonate monoethyl ester chloride (3.86 ml, 30mmol) or succinate monoethyl ester chloride (4.18 ml, 30mmol) was added dropwise to a cooled stirred mixture for 0.5 h. This solution was stirred at room temperature for 6 h. After completion of the reaction, the precipitate was filtered, washed by distilled water. The crude product was recrystallized by 95% aqueous ethanol to obtain the compound 3.

2.3.1. Ethyl-2-((4-(N-(4,6-dimethylpyrimidin-2-yl)sulfamoyl)phenyl)amino)-2-oxoacetate (3a)

White solid, yield 71.9%, ¹H NMR (400 MHz, DMSO-d₆) δ 7.82–7.75 (m, 4H), 7.01 (s, 1H), 5.99 (s, 1H), 4.29 (q, J = 7.1 Hz, 2H), 2.32 (s, 6H), 1.40 (t, J = 7.1 Hz, 3H). ¹³C NMR (100 MHz, DMSO-d₆) δ 162.58, 160.27, 156.33, 155.88, 141.56, 132.47, 128.70, 119.39, 116.22, 63.01, 23.35, 13.95.

2.3.2. Ethyl-3-((4-(N-(4,6-dimethylpyrimidin-2-yl)sulfamoyl)phenyl)amino)-3-oxopropanoate (3b)

White solid, yield 69.1%, ¹H NMR (400 MHz, DMSO-d₆) δ 9.85 (s, 1H), 7.87–7.81 (m, 2H), 7.77–7.71 (m, 2H), 7.01 (d, J = 0.8 Hz, 1H), 4.14 (q, J = 7.0 Hz, 2H), 3.53 (s, 2H), 2.33 (s, 6H), 1.25 (t, J = 7.0 Hz, 3H). ¹³C NMR (100 MHz, DMSO-d₆) δ 168.43, 165.35, 161.62, 155.88, 143.26, 132.45, 128.75, 119.44, 116.21, 61.44, 41.34, 23.35, 14.05.

2.3.3. Ethyl-4-((4-(N-(4,6-dimethylpyrimidin-2-yl)sulfamoyl)phenyl)amino)-4-oxobutanoate (3c)

White solid, yield 65.5%, ¹H NMR (400 MHz, DMSO-d₆) δ 9.84 (s, 1H), 9.17 (s, 1H), 7.85–7.79 (m, 2H), 7.78–7.71 (m, 2H), 7.01 (d, J = 0.8 Hz, 1H), 4.11 (q, J = 7.0 Hz, 2H), 2.76–2.69 (m, 2H), 2.69–2.61 (m, 2H), 2.32 (s, 6H), 1.21 (t, J = 6.9 Hz, 3H). ¹³C NMR (100 MHz, DMSO-d₆) δ 171.43, 170.84, 161.61, 155.88, 142.53, 132.45, 128.79, 119.36, 116.17, 60.44, 32.45, 30.32, 23.35, 14.19.

2.4. General procedure for synthesis of compound 4

Potassium hydroxide (1.68 g, 30 mmol) dissolved in 5 ml of distilled water was added to a solution of compound 3 (10 mmol) in 150 ml ethanol at 0°C. The reaction mixture was allowed to warm to room temperature and stirred for additional 4 h. After the reaction was completed, 100 ml of distilled water was added to the reaction solution and the solution was acidified with concentrated hydrochloric acid to pH 1. After removing the solvent in vacuo, compound 4 was recrystallized from water.

2.4.1. 2-((4-(N-(4,6-dimethylpyrimidin-2-yl)sulfamoyl)phenyl)amino)-2-oxoacetic acid (4a)

White solid, yield 60.7%, ¹H NMR (400 MHz, DMSO-d₆) δ 7.86–7.75 (m, 4H), 7.00 (s, 1H), 3.41 (s, 1H), 2.32 (s, 6H). ¹³C NMR (100 MHz, DMSO-d₆) δ 162.99, 162.56, 160.57, 155.88, 142.04, 132.47, 128.74, 119.28, 116.22, 23.35.

2.4.2. 3-((4-(N-(4,6-dimethylpyrimidin-2-yl)sulfamoyl)phenyl)amino)-3-oxopropanoic acid (4b)

White solid, yield 55.5%, ¹H NMR (400 MHz, DMSO-d₆) δ 9.97 (s, 1H), 9.06 (s, 1H), 7.85–7.79 (m, 2H), 7.78–7.72 (m, 2H), 7.03–6.99 (m, 1H), 3.39 (s, 2H), 2.33 (s, 6H). ¹³C NMR (100 MHz, DMSO-d₆) δ 170.79, 166.39, 162.56, 155.88, 143.07, 132.44, 128.77, 119.24, 116.22, 43.23, 23.35.

2.4.3. 4-((4-(N-(4,6-dimethylpyrimidin-2-yl)sulfamoyl)phenyl)amino)-4-oxobutanoic acid (4c)

White solid, yield 51.8%, ¹H NMR (400 MHz, DMSO-d₆) δ 9.84 (s, 1H), 7.86–7.80 (m, 2H), 7.77–7.70 (m, 2H), 7.03–6.99 (m, 1H), 2.66–2.52 (m, 4H), 2.36 (s, 6H). ¹³C NMR (100 MHz, DMSO-d₆) δ 175.39, 170.96, 162.58, 155.88, 142.17, 132.45, 128.71, 119.34, 116.22, 32.55, 30.74, 23.35.

2.5. General procedure for synthesis of target compounds X1–X9

Compound 4 (5 mmol) and substituted aniline (7.5 mmol) were dissolved in DMF (50 ml), and then HOBT (1.35 g, 10 mmol) and EDCI (3.35 g, 17.5 mmol) were added into the solution. The reaction mixture was stirred at 25°C for 6 h under nitrogen flow condition. After completion of the reaction, the precipitate was filtered, washed by distilled water. The crude product was recrystallized by 95% aqueous ethanol to give the appropriate target compounds X1–X9.

2.5.1. N¹-(4-(N-(4,6-dimethylpyrimidin-2-yl)sulfamoyl)phenyl)-N⁴-(4-fluorophenyl)succinamide (X1)

White solid, yield 47.8%, m.p. 187.3–190.0°C, purity 99.32%; ¹H NMR (400 MHz, DMSO-d₆) δ 10.93 (s, 1H), 9.10 (s, 1H), 8.02–7.98 (m, 4H), 7.32 (d, J = 7.2 Hz, 1H), 7.14–6.95 (m, 4H), 2.88 (d, J = 7.0 Hz, 4H), 2.25 (s, 6H). ¹³C NMR (100 MHz, DMSO-d₆) δ 161.48, 160.02, 159.47, 156.68, 142.58, 142.51, 141.60, 136.63, 130.69, 130.61, 129.47, 125.32, 120.12, 115.92, 115.71, 113.58, 113.37, 34.59, 23.31. HRMS (ESI) calcd for C₂₂H₂₂FN₅O₄S [M+Na]⁺: 494.1269; Found: 494.1260.

2.5.2. N¹-(4-(N-(4,6-dimethylpyrimidin-2-yl)sulfamoyl)phenyl)-N⁴-(4-methoxyphenyl)succinamide (X2)

White solid, yield 41.6%, m.p. 190.9–193.3°C, purity 97.48%; ¹H NMR (400 MHz, DMSO-d₆) δ 10.94 (s, 1H), 10.54 (s, 1H), 9.07 (s, 1H), 8.01–7.88 (m, 4H), 7.22 (d, J = 7.9 Hz, 1H), 6.88–6.62 (m, 5H), 3.73 (s, 3H), 2.96–2.80 (m, 4H), 2.25 (s, 6H). ¹³C NMR (100 MHz, DMSO-d₆) δ 159.98, 159.78, 159.54, 156.69, 141.62, 141.12, 136.62, 129.87, 129.47, 126.37, 124.28, 121.35, 120.12, 119.25, 114.72, 112.17, 110.69, 55.39, 35.02, 23.32. HRMS (ESI) calcd for C₂₃H₂₅N₅O₅S [M+Na]⁺: 506.1469; Found: 506.1463.

2.5.3. N¹-(3,4-dimethoxyphenyl)-N⁴-(4-(N-(4,6-dimethylpyrimidin-2-yl)sulfamoyl)phenyl)succinamide (X3)

White solid, yield 43.7%, m.p. 203.7–205.2°C, purity 99.90%; ¹H NMR (400 MHz, DMSO-d₆) δ 11.20 (s, 1H), 10.93 (s, 1H), 9.03 (s, 1H), 8.07–7.93 (m, 4H), 6.89–6.81 (m, 1H), 6.78–6.71 (m, 3H), 3.73 (d, J = 9.0 Hz, 6H), 2.78 (m,4H), 2.26 (s, 6H). ¹³C NMR (100 MHz, DMSO) δ 159.95, 159.59, 156.64, 149.09, 147.78, 141.61, 131.96, 129.53, 129.46, 127.08, 120.97, 120.84, 120.29, 120.12, 112.98, 112.35, 55.99, 55.87, 34.57, 23.32. HRMS (ESI) calcd for C₂₄H₂₇N₅O₆S [M+Na]⁺: 536.1574; Found: 536.1569.

2.5.4. N¹-(4-(N-(4,6-dimethylpyrimidin-2-yl)sulfamoyl)phenyl)-N⁴-phenylsuccinamide (X4)

White solid, yield 46.3%, m.p. 188.2–190.5°C, purity 96.38%; ¹H NMR (400 MHz, DMSO-d₆) δ 10.93 (s, 1H), 10.70 (s, 1H), 8.03–7.95 (m, 4H), 7.32 (d, J = 7.8 Hz, 1H), 7.08–6.98 (m, 5H), 2.89 (d, J = 7.2 Hz, 4H), 2.25 (s, 6H). ¹³C NMR (100 MHz, DMSO) δ 171.17, 171.12, 165.86, 156.70, 142.30, 140.06, 133.33, 128.91, 128.83, 124.15, 121.23, 120.07, 114.18, 31.87, 31.82, 23.43. HRMS (ESI) calcd for C₂₂H₂₃N₅O₄S [M+Na]⁺: 476.1363; Found: 476.1363.

2.5.5. N¹-(4-(N-(4,6-dimethylpyrimidin-2-yl)sulfamoyl)phenyl)-N³-(4-fluorophenyl)malonamide (X5)

White solid, yield 39.9%, m.p. 213.4–215.5°C, purity 98.06%; ¹H NMR (400 MHz, DMSO-d₆) δ 11.20 (s, 1H), 10.97 (s, 1H), 9.62 (s, 1H), 8.04–7.95 (m, 4H), 7.37 (m, 2H), 7.20–7.09 (m, 2H), 6.76 (d, J = 3.1 Hz, 1H), 4.40 (s, 2H), 2.25 (s, 6H). ¹³C NMR (100 MHz, DMSO-d₆) δ 162.99, 160.57, 160.28, 159.53, 159.26, 156.65, 141.61, 135.26, 130.03, 129.95, 129.53, 129.46, 127.08, 120.84, 120.29, 120.15, 115.64, 115.43, 115.31, 42.46, 23.31. HRMS (ESI) calcd for C₂₁H₂₀FN₅O₄S [M+Na]⁺: 480.1112; Found: 480.1109.

2.5.6. N¹-(4-chlorophenyl)-N³-(4-(N-(4,6-dimethylpyrimidin-2-yl)sulfamoyl)phenyl)malonamide (X6)

Yellow solid, yield 44.4%, m.p. 214.8–216.9°C, purity 95.23%; ¹H NMR (400 MHz, DMSO-d₆) δ 11.61 (s, 1H), 11.19 (s, 1H), 8.06–7.93 (m, 4H), 7.88–7.80 (m, 2H), 7.40–7.30 (m, 2H), 7.31 (d, J = 8.5 Hz, 1H), 4.04 (s, 2H), 2.27 (s, 6H), 2.00 (s, 1H). ¹³C NMR (100 MHz, DMSO) δ 162.99, 162.90, 162.77, 160.57, 159.26, 156.65, 141.61, 135.26, 130.03, 129.95, 129.53, 129.46, 128.06, 127.66, 127.08, 120.84, 120.15, 115.64, 115.43, 45.14, 23.31. HRMS (ESI) calcd for C₂₁H₂₀ClN₅O₄S [M+Na]⁺: 496.0816; Found: 496.0865.

2.5.7. N¹-(4-(N-(4,6-dimethylpyrimidin-2-yl)sulfamoyl)phenyl)-N²-(4-fluorophenyl)oxalamide (X7)

White solid, yield 43.8%, m.p. 240.4–241.6°C, purity 99.72%; ¹H NMR (400 MHz, DMSO-d₆) δ 11.16 (s, 1H), 10.97 (s, 1H), 8.08–7.97 (m, 4H), 7.91–7.85 (m, 2H), 7.47–7.15 (m, 2H), 6.74 (d, J = 0.8 Hz, 1H), 2.25 (s, 6H). ¹³C NMR (100 MHz, DMSO-d₆) δ 160.57, 159.54, 159.24, 158.65, 158.17, 156.61, 141.62, 136.79, 134.46, 129.54, 127.09, 122.98, 122.90, 120.85, 120.28, 116.03, 115.81, 113.85, 23.28. HRMS (ESI) calcd for C₂₀H₁₈FN₅O₄S [M+Na]⁺: 466.0956; Found: 466.0957.

2.5.8. N¹-(4-(N-(4,6-dimethylpyrimidin-2-yl)sulfamoyl)phenyl)-N²-(4-nitrophenyl)oxalamide (X8)

Yellow solid, yield 38.8%, m.p. 199.5–201.2°C, purity 98.78%; ¹H NMR (400 MHz, DMSO-d₆) δ 11.45 (s, 1H), 11.23 (s, 1H), 8.33–8.25 (m, 2H), 8.19–8.12 (m, 2H), 8.05–7.90 (m, 4H), 6.77 (d, J = 0.8 Hz, 1H), 2.26 (s, 6H), 2.00 (s, 1H). ¹³C NMR (100 MHz, DMSO-d₆) δ 161.60, 159.08, 159.07, 155.87, 143.30, 141.90, 141.85, 132.49, 128.77, 125.51, 119.30, 118.01, 116.17, 23.35. HRMS (ESI) calcd for C₂₀H₁₈N₆O₆S [M+Na]⁺: 493.0901; Found: 493.0896.

2.5.9. N¹-(4-acetylphenyl)-N²-(4-(N-(4,6-dimethylpyrimidin-2-yl)sulfamoyl)phenyl)oxalamide (X9)

White solid, yield 39.0%, m.p. 193.5–194.3°C, purity 97.94%; ¹H NMR (400 MHz, DMSO-d₆) δ 11.20 (d, J = 8.5 Hz, 2H), 8.25–7.83 (m, 8H), 6.76 (s, 1H), 2.57 (s, 3H), 2.26 (s, 6H). ¹³C NMR (100 MHz, DMSO-d₆) δ 197.25, 159.31, 159.23, 156.63, 142.34, 141.59, 133.42, 129.79, 129.54, 127.03, 124.59, 120.45, 120.29, 119.39, 110.54, 27.07, 23.31. HRMS (ESI) calcd for C₂₂H₂₁N₅O₅S [M+Na]⁺: 490.1156; Found: 490.1155.

2.6. Biological assays

The neuraminidase (H5N1 and H5N1-H274Y) was purchased from Sino Biological Inc. (Beijing, China). 2-N-mopholino-ethanesulfonic acid (MES) and 4-methylumbelliferyl-α-D-N-acetylneuraminic acid sodium salt hydrate (4-MUNANA) were purchased from Sigma (Kawasaki, Japan). 4-methylumbelliferone (4-MU) was purchased from Shanghai Standard Technology Co., Ltd (Shanghai, China). The enzyme assay was performed by using the previously reported method with slight modifications [31]. The tested compounds were dissolved in DMSO first and then diluted into six concentration gradients. For each tested compound 5 and X1–X10, a set of solutions with concentrations span the range 0.064–200 μM. In general, 10 μl of NA, 70 μl buffer solution (33 mM MES, 4 mM CaCl₂) and 10 μl of different concentrations of samples were added to each well of the 96-well plate. Then the 96-well plate was placed and shocked for 1 min in the multifunctional fluorescent enzyme-labeled instrument and the temperature was set at 37°C, so that the NA enzyme and the sample to be tested could be fully mixed. The mixture was incubated at 37°C for 15 min. 10 μl of 100 μM fluorescent substrate (4-MUNANA) solution was also added to each well. Next, the plate was placed in the multifunctional fluorescent enzyme-labeled instrument again. It was shaken for 1 min and then incubated at 37°C for 60 min. The reaction was terminated by adding 150 μl of stop solution (14 mmol·l^-1 NaOH containing 83% ethanol). Finally, the resulting fluorescence was measured at an excitation wavelength of 355 nm and an emission wavelength of 460 nm, respectively. Parallel experiments were performed three times. The oseltamivir acid was used as a positive control in the enzyme inhibition assay. The inhibition curves were drawn by GraphPad Prism 5.0 software and the IC₅₀ values were calculated by using enzyme inhibition rate and concentration data. The IC₅₀ (μM) is presented as mean ± SD from at least three independent tests.

3. Results & discussion

3.1. Virtual screening

In this study, the H5N1 NA (Protein Data Bank [PDB] ID: 2HU0) acts as target protein. The crystal structure of NA was downloaded from the RCSB Protein Data Bank. The x-ray coordinates of 2HU0 were shown in Supplementary Table S1. Before docking operation some preparations for target protein, such as dehydration, hydrogenation and modification of protein structure is indispensable. Figure 1 displays the lead compound discovery process by way of virtual screening. First, one database, Data set 1, was downloaded from the ZINC 12 database. It is consisting of about 350,000 small bioactive molecules and had been filtered by the Lipinski's Rule of Five (RO5) to guarantee druggability. Then, initial screening was preformed by using of Sybyl-X 2.1 (Tripos Inc, MO, USA) with its surflex-dock module. It is also necessary to consider the molecular ring flexibility in the docking process. Data set 2, consisting of the top higher total scores 100 molecules, was obtained by comparison of the docking scores of the extracted ligand with that of molecules in Data set 1. Further binding mode analysis revealed that some molecules can interact with both the 150-cavity and the active site. For the selected molecules, structural diversity should be considered as one of the key focuses. Ultimately, the top 20 molecules were selected and composed of Data set 3.

Figure 1.

The lead compound discovery process by way of structure-based virtual screening.

MD: Molecular dynamics.

Binding free energy calculations based on MD simulations is an important technique in studying the binding affinities between targets and inhibitors. To hit new lead compounds, MD simulations were further performed for the 20 molecules of Data set 3 with the package of Amber 12.0 [32]. The globle simulation convergence and system equilibrium were monitored with root-mean-square deviation (RMSD) of backbone atoms (C, C_α, N and O). Supplementary Figure S1 shows the RMSD values (Å) relative to the simulation time (ns). Two methods, Molecular Mechanics/Generalized Born (MM/GBSA) and Molecular Mechanics/Poisson Surface Area (MM/PBSA), were used to compute binding free energies of NA enzyme and ligands [33–35]. In contrast, MM/PBSA is usually considered to be better than MM/GBSA in calculating binding free energy. Furthermore, it was reported that binding free energy has a good correlate with compound's activity. The smaller the ΔG_bind of a molecule is, the stronger the activity of the molecule is. Accordingly, three compounds ZINC670537, ZINC33315666 and ZINC16112011 (Figure 2) with smaller ΔG_bind values were obtained to constitute Data set 4. The corresponding ΔG_bind values of these three compounds are tabulated in Supplementary Table S2. The binding free energy of oseltamivir carboxylate (OSC) was also calculated for comparison. The ΔG_bind values of ZINC670537, ZINC33315666 and ZINC16112011 predicted by MM/GBSA method are -12.61, -13.51 and -13.02 kcal·mol^-1. The ΔG_bind values predicted by MM/PBSA method are -13.84, -11.83 and -11.25 kcal·mol^-1. For the reference OSC, the corresponding values calculated by MM/GBSA and MM/PBSA methods are -13.40 and -10.08 kcal·mol^-1, respectively. Therefore, from the calculation results of MM/PBSA, ZINC670537 (compound 5) has the smallest binding free energy and probably has satisfying NA inhibitory activity. ZINC670537 was further examined for biological activity to confirm the reliability of theoretical prediction. Results indicate that it has satisfying activity (IC₅₀ = 16.50 μM). Finally, compound 5 had been chosen as a lead compound for further modification.

Figure 2.

The chemical structures of Data set 4 and the lead compound 5.

A series of new NA inhibitors were further designed with lead 5 as a template. Figure 3 shows the bonding model of 5 with the NA enzyme (PDB: 2HU0). It can be seen that the pyrimidine group could well extend into the 430-cavity (composed of key residues Pro431, Thr439 and Arg430, etc.). However, except for the oxygen atoms of the sulfonamide group, which can form four hydrogen bonds with the key residues Arg152 and Arg118, and the amide oxygen atom, which can form a hydrogen bond with Arg371, the pyrimidine and benzene rings do not form any hydrogen bonds with the binding sites, such as 150-cavity and active site. Consequently, great emphasis should be put on introduction of some hydrogen bond receptors, such as halogen atoms, acetyl, methoxy and nitro groups into pyrimidine-4,6-diol or benzene rings to form more hydrogen bonds with the binding sites, and thus enhance the inhibitory activity of the inhibitors. Thus, ten new compounds X1–X10 had been designed (Figure 4). MD simulation were subsequently carried out for these ten new compounds, and the RMSD values (Å) versus simulation time (ns) are shown in Supplementary Figure S2. It is shows that all curves tend to be equilibrium after 7.5 ns. The calculated ΔG_bind values for X1–X10 are listed in Table 1. It is shown that all ΔG_bind values are much smaller than that of the reference OSC calculated by the two methods of MM/GBSA and MM/PBSA, suggesting that their activity are theoretically good and have the potential value for further synthesis and study. Especially, as the above described, calculation results from MM/PBSA is generally regarded to be more reliable than that from MM/GBSA. From the MM/PBSA results one can see that X3 has the lowest binding free energy among the ten newly designed compounds and it maybe have the strongest activity.

Figure 3.

Binding mode of lead compound 5 (ZINC670537) with neuraminidase (Protein Data Bank ID: 2HU0).

Figure 4.

Synthesis of target compounds X1–X10. Reagents and conditions: (A) Chloroacetyl chloride, K₂CO₃, DMF, 0°C, 1.5 h; (B) K₂CO₃, DMF, 60°C, 10 h; (C) K₂CO₃, DMF, rt, 6 h; (D) KOH, ethanol, rt, 4 h; (E) DMF, 1-hydroxybenzotriazole,1-(3-dimethylaminopropyl)-3-ethylcarbodiimide hydrochloride, rt, 6 h.

rt: Room temperature.

Table 1. The calculated binding free values (kcal·mol^-1) of all designed compounds.

No.	VDW	EEL	ΔGgas	ΔGGB	ΔGSA	ΔGsolv (GB)	ΔGbind (GB)	ΔGPB	ΔGSA	ΔGsolv (PB)	ΔGbind (PB)
OSC	-30.36	-24.18	-54.54	45.44	-4.30	41.14	-13.40	48.15	-3.69	44.46	-10.08
X1	-24.79	-34.07	-58.86	41.66	-3.41	38.25	-20.61	45.36	-3.18	42.18	-16.68
X2	-26.46	-11.22	-37.68	28.92	-3.83	25.09	-12.59	27.06	-3.49	23.57	-14.11
X3	-38.76	-7.70	-46.46	22.37	-4.70	17.68	-28.78	29.72	-3.81	25.91	-20.55
X4	-34.80	-13.56	-48.36	34.58	-4.46	30.11	-18.25	40.24	-3.56	36.68	-11.68
X5	-37.97	-15.02	-52.99	31.98	-5.29	26.68	-26.30	42.83	-4.23	38.60	-14.39
X6	-28.11	-13.09	-41.20	31.19	-3.24	27.95	-13.24	33.04	-2.98	30.06	-11.14
X7	-26.55	-14.44	-40.99	28.99	-3.22	25.77	-15.22	26.73	-3.05	23.68	-17.31
X8	-26.78	-29.34	-56.12	41.13	-3.67	37.46	-18.66	42.43	-3.04	39.40	-16.73
X9	-20.34	-17.67	-38.01	24.01	-2.46	21.55	-16.47	26.13	-2.23	26.12	-14.11
X10	-43.13	-30.11	-73.24	50.80	-5.75	45.05	-28.19	60.38	-4.63	55.74	-17.50

EEL: Electrostatic energy; GB: Generalized born; OSC: Oseltamivir carboxylate; PB: Poisson Boltzmann; VDW: Van de Waals.

3.2. Chemistry

Compounds (X1–X10) were synthesized by a feasible synthetic route shown in Figure 4 With sulfamethazine and chloroacetyl chloride or substituted monoethyl acyl chloride as raw materials, target compounds were derived via nucleophilic substitution, ester hydrolysis and condensation steps. Structure conformation was performed by the means of ¹³C NMR, ¹H NMR and HRMS and high performance liquid chromatography (HPLC) was used to detect the purity data. The corresponding spectra from structural characterization are shown in Supporting Information.

3.3. In vitro inhibitory activity

The inhibition of wild-type H5N1 NA for X1–X10 was tested. The results show that most of the compounds have fairly good inhibitory activity. Table 2 lists the corresponding IC₅₀ values. It can be seen that two compounds X3 and X10 exhibit the strongest activity among the ten new synthesized compounds, with IC₅₀ values of 6.74 and 10.71 μM. Especially, the inhibitory activity of X3 is more close to that of the positive control OSC. The IC₅₀ values of X1(15.33 μM), X5(17.56 μM), X7(13.56 μM) and X8(16.23 μM) are slightly higher than that of X10. With an IC₅₀ value of 32.94 μM, the activity of X6 is the weakest. From structure viewpoint, compounds X1–X9 are different in substituted groups of phenyl, the methoxy in X2 and X3 is electron-donating group, whereas the corresponding substituted groups in other compounds, except X4 are all electron-withdrawing groups (e.g. -F, -Cl, -NO₂, -COCH₃), indicating that the presence of electron-donating groups in benzene ring may improve compounds' activity. X3 has two electron-donating substituted groups and it has the strongest activity. For X10, The two hydroxyl groups in pyrimidine ring are also electron-donating groups. In combination with the binding free energy calculation results of Tables 1 and 2, among all synthesized compounds, X3 and X10 have the most negative ΔG_bind values and their activities against NA are also the most strongest. Two compounds X2 and X9 have equal ΔG_bind values (-14.11 kcal·mol^-1) predicted by MM/PBSA, and their IC₅₀ values are also very close (20.74 μM and 20.33 μM). X4 (IC₅₀ = 31.80 μM) and X6 (IC₅₀ = 32.94 μM) exert the weakest activity, and their binding free energies are also the highest among the ten synthesized compounds (-11.68 and -11.14 kcal·mol^-1 predicted by MM/PBSA). Therefore, the theoretical and experimental results are consistent quite well.

Table 2. Inhibition activity of target compounds (X1–X10) against the wild-type H5N1 NA.

Compound	Structure	IC50 (μM)
X1		15.33 ± 1.05
X2		20.74 ± 2.63
X3		6.74 ± 0.13
X4		31.80 ± 4.37
X5		17.56 ± 4.24
X6		32.94 ± 5.66
X7		13.56 ± 3.77
X8		16.23 ± 0.98
X9		20.33 ± 4.72
X10		10.71 ± 0.51
OSC	–	0.12 ± 0.02

To better explore the reasons for the activity discrepancy of compounds and perform further study for structure–activity relationships, molecular docking analysis was carried out with the surflex-dock module of SYBYL-X 2.1. The binding modes of inhibitors X3, X4, X6 and X10 with NA are exhibited in Figure 5. From the 2D diagram, one can see that the pyrimidine groups of X3 and X10 can well extend into a large hydrophobic 150-cavity (consisting of some key residues Gly147-Arg156). The sulfonamide group establishes steady hydrogen bonds with Arg118, Arg371, Arg152 and Arg156, playing a role as a bridge connecting the active site and 150-cavity, and anchoring molecular skeleton in the binding sites. For X3, five hydrogen bonds are formed by sulfonamide oxygen atoms and residues Arg156 and Arg152, two hydrogen bonds are formed by the amide oxygen atoms of the succinamide chain and residues Arg371 and Arg118, and one hydrogen bond is formed by one methoxy group in the phenyl and Ser246. For X10, four hydrogen bonds are formed by sulfonamide oxygen atoms and residues Arg371, Arg118, the sulfur and hydroxyl oxygen atoms on the pyrimidine group, and amide oxygen atom form four hydrogen bonds with Asp151, Arg152, respectively. Amide nitrogen atom generates one hydrogen bond with Tyr347. Through the comparison of X3 and X10, although these two compounds can well implanted into two binding sites of 150-cavity and active site, and both form strong hydrogen bonds with two essential residues of Arg371 and Arg118, X3 forms more hydrogen bonds with residues near the 150-cavity, and can combine more stably with the 150-cavity. For X4, only three hydrogen bonds are formed by amide oxygen atom and Arg118, Arg371, Tyr406. Regarding X6, although the pyrimidine group can well occupy the 150-cavity, the chlorobenzene group has not generated any hydrogen bonds. Therefore, theoretically, X6 has the weakest activity towards NA in all compounds, and its calculated binding free energy is also the highest.

Figure 5.

Binding modes of some representative compounds X3, X4, X6 and X10 with neuraminidase (Protein Data Bank: 2HU0).

It is known that the H274Y mutant NA is resistant to oseltamivir. For the target compounds (X1–X10), further bioactivity assay was managed to examine the inhibitory activities against the mutant NA. The corresponding results are shown in Table 3. Oseltamivir carboxylate (OSC), with an IC₅₀ value of 18.23 μM, approximately 152-fold more potent than the IC₅₀ value against the wild-type NA. For mutant NA it is obviously insensitive. As shown in Table 3, X3 (IC₅₀ = 21.09 μM) still exerts the most potency against the H274Y mutant, the activity is similar to that of OSC. The inhibitory activity of X10 (IC₅₀ = 32.24 μM) is a little weaker than that of X3. The activities of X4 and X6 are still the weakest among all NA inhibitors.

Table 3. Inhibition activity of target compounds (X1–X10) against the H274Y mutant NA.

Compound	IC50 (μM), H5N1(H274Y)
X1	49.58 ± 5.07
X2	151.07 ± 14.11
X3	21.09 ± 0.78
X4	177.20 ± 2.51
X5	52.13 ± 1.89
X6	193.70 ± 14.67
X7	38.61 ± 4.99
X8	50.20 ± 6.18
X9	131.30 ± 11.76
X10	32.24 ± 3.56
OSC	18.23 ± 0.32

4. Conclusion

In this study, a novel lead compound ZINC670537 (IC₅₀ = 16.50 μM) was screened out by means of virtual screening, MD simulation and bioactivity test. Further structural modification of ZINC670537 resulted in ten new sulfamethazine NA inhibitors (X1–X10). Most new synthesized inhibitors exert good potency comparable to oseltamivir carboxylate approved by US FDA. Especially, compound X3 has the best activity, with IC₅₀ values of 6.74 μM (wild-type H5N1 NA) and 21.09 μM (H274Y mutant NA), respectively, and the activity inhibiting the H274Y mutant NA is comparable to that of OSC. Study results show that the excellent activity of X3 could be ascribed to the 3,4-dimethoxy substituted phenyl, which is connected to 150-cavity via the succinamide moiety. For these sulfamethazine NA inhibitors, 150-cavity is very important in determining their activities. In addition, the sulfamethazine moiety also plays important roles, as it can interact tightly with three key residues (Arg118, Arg292, Arg371) of the active site by forming strong hydrogen bondings. These results of this study broaden and advance the path toward new high-efficient NA inhibitors.

Funding Statement

This work was supported by the Collaborative Innovation Fund (XTCX2023) and the Shanghai Municipal Education Commission.

Supplemental material

Supplemental data for this article can be accessed at https://doi.org/10.1080/17568919.2024.2342688

Financial disclosure

This work was supported by the Collaborative Innovation Fund (XTCX2023) and the Shanghai Municipal Education Commission. The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.

Competing interests disclosure

The authors have no competing interests or relevant affiliations with any organization or entity with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

Writing disclosure

No writing assistance was utilized in the production of this manuscript.