Synthesis of multi ring-fused imidazo [1,2- a ]isoquinoline-based fluorescent scaffold as anti-Herpetic agent

Abstract

Background

Natural product-inspired synthesis is a key incorporation in modern diversity-oriented synthesis to yield biologically novel scaffold. Inspired by β-carboline fused system, we have designed molecules with multi ring fused scaffold by modifying the tricyclic pyrido[3,4-b]indole ring with imidazo[1,2-a]isoquinoline.

Methods

A highly convergent approach with new C–N and C–C bond formation to synthesize multiring fused complex scaffold imidazo[1,2-a]isoquinolinies as fluorophores. N-nucleophile-induced ring transformation of 2H-pyran-2-one followed by in situ cis-stilbene-type oxidative photocyclization yielded new C–C bond formation without additional oxidant. The cytotoxicity, effective concentrations, and the mode of action of the synthesized analogs were determined by 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT),, plaque reduction, time of addition, and reverse transcriptase Polymerase Chain Reaction (PCR).

Results

Novel imidazo[1,2-a]isoquinoline analogs were prepared, and the results revealed that trans isomer of cyclopropyl analog (EC₅₀ 35 and 37.5 µg/ml) and trans isomer of citric acid salt of phenyl analog (EC₅₀ 38.2 and 39.8 µg/ml) possess significant anti-Herpes Simplex Virus (HSV) activity with selectivity index of >10. The kinetic study demonstrated that both the analogs inhibited HSV-1F and HSV-2G at 2–4 h postinfection. Finally, western blot and reverse transcriptase PCR assays revealed that both the analogs suppressed viral immediate early transcription.

Conclusion

Novel imidazo[1,2-a]isoquinoline analogs were synthesized from pyranone with appropriate amines. Two compounds showed better antiviral profile on HSV-infected Vero cells, compared to the standard drug acyclovir (ACV). Overall, we discovered a promising scaffold to develop a nonnucleoside lead targeting the viral immediate early transcription for the management of HSV infections.

Introduction

Diversity-oriented natural product based synthesis has drawn a special attention due to its potential to discover significant biologically active molecules. ¹ More than 70% of new chemical moieties discovered in the recent past were natural products or natural product derived or mimetics. ² The traditional total synthesis of natural products has been the key to medicinal chemistry since ages, although this type of synthesis is cumbersome and time consuming. Modern approach diversifies the synthesis and explores a wider chemical space.^3,4

β-Carboline alkaloids, saturated or unsaturated, both of them have been reported to be active against either HSV-1 or HSV-2 or both (Figure in supporting information).^5–8 Previously, we had described that βC alkaloids exhibit anti-HSV activity, although with high toxicity (safe up to 50 mg/kg body weight).^6,7 Moreover, in continuation to our pyranone scaffold exploration toward antiviral molecule synthesis,^9,10 chiral amines were inserted into pyranone moiety to yield a natural product like “N-containing multi ring heterocycles” with imidazo[1,2-a]isoquinoline core. This has several biological implications,^11,12 including anti-HSV property. ¹³ Furthermore, the organic molecules with fluorescent properties have added advantages to its biological applications. For example, it can act as DNA staining dyes, work as intracellular probe or they can also be covalently linked to specific protein to study its mode of action. ¹⁴

Diverse synthetic approaches are utilized to synthesize a fused isoquinoline system. The traditional routes such as Bischler−Napieralski, Pictet−Gams, and Pomeranz−Fritsch reactions require harsh conditions and highly active substrates. The transition metal-catalyzed reactions provide an efficient route ¹⁵ and are widely used ¹⁶ for coupling of C(aryl)–C, C(aryl)–N, C(aryl)–O, and C(aryl)–S bonds but is generally expensive. Metal-catalyzed ¹⁷ or non-metal-catalyzed ¹⁸ cycloaddition reactions are yet another efficient way to synthesize N-containing multiple fused ring system. Herein we present a one-pot synthesis of fluorophoric [1,2-a]isoquinoline analogs using key intermediate pyranone and N-nucleophiles to demonstrate anti-HSV potential of novel scaffold.

Materials and methods

General chemistry

The reactions were carried out in oven-dried glassware. The chemicals and solvents were purchased from Spectrochem, Across, Rankem, or Sigma Aldrich. Melting points were recorded on Veego melting point apparatus. Analytical thin layer chromatography (TLC) was performed on precoated plates (silica gel 60 F-254). Purification by gravity column chromatography was carried out on silica gel (100–200 mesh). Elico UV/Vis spectrophotometer was used for recording the UV spectra; while ¹H, ¹³ C NMRs were determined in a Varian/Bruker (300/400 MHz) spectrometer using CDCl₃ or DMSO-d₆ as solvent. Peaks are recorded with: s (singlet), bs (broad singlet), d (doublet), t (triplet), q (quartet), m (multiplet), and J (coupling constant, Hertz). Fluorescence emission spectrum was measured in Shimadzu RF-5301 spectrofluorometer. All the intermediates and final compounds were adequately characterized by spectroscopic analysis, and one of the compounds [3b] was further confirmed by a single crystal X-ray analysis (Figure 1).

Figure 1.

ORTEP diagram of [3b] (crystallized in ethyl acetate: n-hexane 30:70) showing the X-ray molecular structure at 30% probability level.

Procedure for the synthesis of [3a–f]

The suspension of NaH (3.3 mmol) and [2] (1.23 mmol) was placed in dry THF at room temperature; [1a–c] (0.825 mmol) was added and stirred for 24 h. Upon completion, the reaction mixture was poured in ice water. The pH was maintained between 3 and 4 by adding 1.5 N HCl. The precipitate formed was filtered, dried, dissolved in ethyl acetate, and added with silica gel in 1:5 (calculated on 100% yield basis), stirred for 12 h under 40 W yellow light bulb. The product formed was charged on silica gel column and eluted with EtOAc/n-hexane mixture to obtain [3a–f].

Procedure for synthesis [3g–h]

Similar procedure was followed using [1d] and [2] up to the addition of 1.5 N HCl. The precipitate formed was filtered, dried, and the diastereomers were separated using column chromatography using EtOAc/n-hexane mixture as eluent to obtain [3g–h] as a mixture of stereoisomers.

Procedure for synthesis [4a–b]

[3a–b] was dissolved in ethyl acetate and 1 eq citric was added to it and stirred for 4 h. Solid was precipitated, which was filtered and washed with diethyl ether to obtain [4a] and [4b] from [3a] and [3b], respectively.

Characterization of all the new compounds are given in supporting information.

Measurement of fluorescence spectra

Fluorescence emission spectrum was measured on Shimadzu RF-5301 spectrofluorometer. The molecules were excited at λ max (380 nm) using methanol as the solvent. Fluorescence spectrum was measured at different pH using HCl and NaOH solutions in the range of 0.5–12. Spectrum was further obtained using aprotic solvent DMF instead of MeOH dimethylformamide.

Antiviral activity study

Cells and viruses

Vero cells (ATCC, USA) were cultured in Dulbecco’s Modified Eagle’s medium (DMEM; nvitrogen, USA) with 5% fetal bovine serum (FBS; Invitrogen, USA), 100 U/ml penicillin, and 100 U/ml streptomycin at 37°C in 5% CO₂. The viral strains used were HSV-1F (ATCC 733) and HSV-2G (ATCC 734), purchased from the ATCC, USA.

In vitro cytotoxicity and anti-HSV activity study

The in vitro cytotoxicity and antiviral activity were determined by MTT and plaque reduction assays (PRAs).^7,19 In brief, the Vero cells were exposed to various concentrations of the analogs and incubated at 37°C in 5% CO₂, using ACV and DMSO (0.1%) as controls. After 72 h, MTT assay was carried out following manufacturer’s protocol (MTT; Sigma) and the OD was read at 570 nm. The 50% cytotoxic concentration (CC₅₀) was calculated by linear regression of the dose–response curves. For antiviral activity, the PRA was used. Briefly, Vero cells infected with HSV-1F and HSV-2G (100 PFU) were exposed to different concentrations of test drugs for 1 h and then overlaid with 1% methylcellulose containing test drug. The plaques developed after 72 h were counted and virus titers were calculated by scoring the plaque-forming units. The effective concentration of test drugs that reduced plaques’ number by 50% (EC₅₀) was interpolated from the dose–response curves. ⁷

Mode of action

Time of addition assay

Following three different approaches, for understanding the preinfection, cells were treated with test drugs either for 3 or 1 h and then infected with HSV-1F in DMEM containing 2% FBS at 37°C. For studying the coinfection, cells were subsequently infected and treated with test drugs. These were removed after 1 h and added with fresh media. Whereas for postinfection (pi) the cells were first infected with HSV and then treated with the test drugs at 0–24 h time intervals, with an incubation period of 1 h in each case. The mixture was removed and added with fresh media. The PRA was carried out as described,^7,19 after incubation at 37°C in 5% CO₂ for 48 h. DMSO (0.1 %) and ACV (5.0 microg/ml) were used as control for all these experiment.

Attachment and penetration assay

To investigate the role of test drugs on viral attachment the Vero cells were prechilled at 4°C for 1 h and challenged with HSV-1 (100 PFU/well) in the presence of [3g] (35.0 µg/ml) and [4b] (38.2 µg/ml) or DMSO (0.1%) for 3 h at 4°C. Then, the wells were washed with prechilled PBS to remove unabsorbed virus and overlaid with 1% methylcellulose to form plaques. While for penetration assay, the infected-untreated or infected-treated cell monolayers, after 3 h of incubation at 4°C, was reincubated for 20 min at 37°C to facilitate viral penetration. At the end of the incubation period extracellular nonpenetrated virus was inactivated by citrate buffer (pH 3.0) for 1 min, then washed with PBS, and overlaid with fresh medium. The plaques developed after 48 h of incubation at 37°C were stained and counted. ²⁰

Quantitative reverse transcriptase PCR (RT-PCR)

The effects of analog addition on HSV-1 RNA expression within the cell was performed by RT-PCR analysis. Briefly, the HSV-1 (5 Multiplicity of infection (MOI)) infected Vero cells were treated with [3g] (35.0 µg/ml) and [4b] (32.5 µg/ml) or DMSO (0.1%) for 2, 4, and 8 h p.i., and then RNA was isolated using RNeasy Mini kit (Qiagen, Germany) following the manufacturer’s protocol. Aliquots of 1 µg of RNA were used to generate cDNA with a high-capacity cDNA reverse transcription kit (ABI, Foster City, CA). The cDNA (10%) was then subjected to standard PCR amplification ²⁰ using the required primers against HSV-1. The sequences of primers used were as follows: ICP4 (5′-GACGTTGTGGACTGGGAAG-3′ and 5′-ACTTAACAGGTCGTTGCCG-3′), ICP27 (5′-CCTTTCTCCAGTGCTACCTG-3′ and 5′-GCCAGAATGACAAACACGAAG-3′), and GAPDH (5′-CGAGATCCCTCCAAAATCAA-3′ and 5′-ACAGTCTTCTGGGTGGCAGT-3′).

Results and discussions

A convergent one-pot route was followed to synthesize multi ring-fused imidazo[1,2-a]isoquinolines [3a–h] from simple precursors 4-(methylthio)-2-oxo-6-phenyl-2H-pyran-3-carbonitrile [2] and N-nucleophiles based on 2-(imidazolidin-2-ylidene)-1-aryl ethanones [1a–d]. The reaction undergoes condensation and Mallory type photocyclization in absence of I₂ at room temperature by the use of simple 40 W yellow light bulb. Initially, we tried benzene, one of the most preferred solvent for Mallory oxidative photocyclization,^21,22 which resulted a low yield (<35% crude). After optimization, ethyl acetate was found better to achieve 65% crude yield. Further, the addition of silica gel (1:5) into the reaction medium, the crude yield improved to 85%. The starting materials [1a–d] were synthesized by refluxing appropriate ketene dithioacetals with cis/trans-1,2-diamino cyclohexane in toluene (Scheme 1). ²³ The condensation of [1a–d] with [2] ²⁴ was carried out in presence of NaH in THF to yield uncyclized intermediates (Scheme 1) which underwent further oxidative photocyclization to form [3a–f] while that from [1d] did not undergo cyclization and yielded [3g–h] as final compounds.

Scheme 1.

Synthesis of key intermediates [1a–d]. Reagents and conditions: (i) Toluene, 120°C, 24 h. Synthesis of [3a–h]. Reagents and conditions: (ii) NaH, THF, 24°C rt, 4 h; (iii) hυ, O₂ 48 h.

The reaction of [2] was carried out with [1d] in which the aromatic ring is replaced with cyclopropyl group. The recyclized product [3g–h] did not undergo any further cyclization and was collected as a mixture of stereoisomers due to the induced chirality at the C–CN bond (Scheme 1). Each compound contained a mixture of E and Z isomers that was not separated further. One of the isomers was crystallized and characterized in a pure form (NMR data attached in supporting information) confirmed the formation of [3g].

The products [3a–f] was purified and characterized by spectroscopic methods. The structures for the regioisomers [3e–f] were assigned with the help of H-H NOESY and H-H COSY spectrum. It is interesting to mention that an ambiguity arose due to the presence of multiplet at 4.29–4.43 which is directly attached to the CH_3, as revealed by COSY spectra. However, this multiplet corresponds to two protons. HSQC spectra suggest that this multiplet of two protons is attached to two different carbons. This implies that the multiplet of two protons at 4.29–4.43 is not due to two protons of CH₂ but a merged peak for CH–NH and one of the protons of CH_2. The formation of [3a–f] from their respective intermediates can be substantiated by the mass spectrum of the intermediate. The mass spectrum of intermediate [I] shows the presence of base peak at (M⁺+ 1) at 394.3; however, the mass spectrum of [3a] (Figure in supporting information) shows the presence of molecular ion (M⁺) at 391.6. Thus, the changes in TLC pattern and the mass spectrum clearly indicate the conversion of intermediate [I] to [3a] (Scheme 1; Spectral data in supporting information). Moreover, the more soluble compounds [4a] and [4b] were prepared as a citric acid salt of [3a] and [3b] to understand the effect of solubility to the activity against HSV.

Fluorescent properties

Significant fluorescent properties were shown by [3a–f] as shown in Figure 2. However, this property was not for the compounds [3g–h] which have cyclopropyl group in place of the aromatic group. This indicates the presence of aromatic group adjunct to the system, which extends the conjugation, is an absolute necessity for the fluorophoric property. Moreover, inclusion of a sulfur heteroatom in [3c–d] increased the fluorescence intensity to a large extent. The aza-fused systems are known to respond to change in pH and solvent due to the formation of different protonation states. However, no evident change was observed in fluorescence spectra with the change in pH or solvent. This clearly suggests that lone pairs of N are not involved in the conjugation and hence does not contribute to fluorescence and the planar conjugated imidazo[1,2-a]isoquinolines are the sole participants. Similar pH independent xanthamide dyes have been reported earlier. ²⁵

Figure 2.

(a) Absorbance and emission spectra of [3a] and (b) Fluorescence spectra of [3a], [3c], and [3e] in 1 μM solution at pH 7.

Cytotoxicity and antiviral activity study

Cytotoxicity assays, presented in Table 1, revealed that two of the molecules exhibited cytotoxicity (CC₅₀) against Vero cells between 225 and 455 µg/ml, which is much higher than their EC₅₀ (35.00 and 37.5 µg/ml for [3g], 38.2 and 39.8 µg/ml for [4b], respectively) against HSV-1F and HSV-2G, indicating significant antiviral activity, compared to ACV (1.5 and 22 µg/ml). It was interesting to note that upon the addition of the polar citrate group [4b] or by the replacement of the hydrophobic aromatic phenyl/thiophenyl group [3a–d] with cyclopropyl [3g] there was a marked increase in the biological activity. Furthermore, the selectivity index (>10) of [3g] and [4b] against the tested strains indicated its significant activity against HSV. The results further showed that [3g] and [4b] inhibited in a dose-dependent manner, and maximum (>99%) inhibition of plaque formation was observed at 58.4 and 71.8 µg/ml against HSV-1F, and 63.5 and 76.2 µg/ml against HSV-2G, respectively.

Table 1.

Anti-HSV activity spectrum on HSV-infected Vero cell.

Compound	CC50 a	EC50 b		Selectivity index c
	(µg/ml)	HSV-1	HSV-2	HSV-1	HSV-2
[3a] cis	225	25	26.5	9	8.49
[3b] trans	250	26.5	27.2	9.43	9.19
[3c] cis	325	>100	–	>3.25	–
[3d] trans	260	>100	–	>2.6	–
[3e]	220	>100	–	>2.2	–
[3f]	340	>100	–	>3.4	–
[3g] cis	410	35.0	37.5	11.71	10.93
[3h] trans	455	48.1	48.7	9.46	9.36
[4a] cis	440	47.8	51.5	9.21	8.54
[4b] trans	400	38.2	39.8	10.47	10.05
ACV	130	1.5	2.2	86.66	59.09

ACV: Acyclovir; CC: cytotoxic concentration; EC: Effective Concentration; HSV: Herpes Simplex Virus.

a

The 50% cytotoxic concentration for Vero cells in µg/ml.

b

Concentration of compound in µg/ml producing 50% inhibition of virus-induced cytopathic effect. All experiments were repeated for three times.

c

Selectivity index = CC₅₀/EC₅₀.

To investigate the possible stage of viral life cycle interfered by the test analogs, we performed the time-of-addition assay. The results demonstrated that [3g] and [4b] exert their inhibitory effect during 1–4 h p.i., with maximum inhibition at 2–4 h p.i. (Figure 3), i.e. the early event of viral life cycle including attachment and penetration, whereas ACV was effective at 6–12 h p.i. Again, we investigated these steps separately. Result showed that [3g] and [4b] had no effect on HSV-1F attachment or penetration into the Vero cells (Figure in supporting information) like ACV, indicating that these compounds were unable to inhibit viral entry to the host cell, but perhaps interferes with the viral early replication process.

Figure 3.

Time of addition assay of [3g] and [4b] against HSV-1F (a) and HSV-2G (b). Vero cells were infected with HSV-1F (a) and 2G (b), untreated or treated with [3g] or [4b] or ACV at their EC100 concentrations at various time points: preinfection (3 h, 1 h), coinfection (0 h), and postinfection (1–24 h) were subjected to the plaque reduction assay. Each bar represents the mean ± SEM of three independent experiments. (c) Effect of [3a] and [3b] on IE gene expression: HSV-1 infected (5 m.o.i.) Vero cells were treated with [3g] and [4b] for 2–8 h postinfection. Then RNA was isolated and subjected to cDNA synthesis, followed by the quantitative reverse transcriptase PCR of ICP4 and ICP27, using GAPDH as control. GAPDH: Glyceraldehyde 3-phosphate dehydrogenase; HSV: Herpes Simplex Virus; ICP: Infected Cell Polypeptide; PCR: Polymerase Chain Reaction.

The expressions of two IE gene products ICP4 and ICP27 were determined to analyze the effect of [3g] and [4b] on HSV-1 replication cycle.^26,27 The RT-PCR data reveal that ICP4 and ICP27 were well expressed in untreated controls but failed to express in virus-infected drug-treated cells (Figure 3(c)). This indicated that both these analogs may interfere with the IE transcription of HSV-1 at 2–4 h p.i. Importantly, two major products of IE gene, ICP4 and ICP27, expressed by HSV, have intense effects on viral gene expression and are essential for virus replication.

During lytic infection, the gene expression of HSV occurs in a regulated fashion commencing with the synthesis of IE gene products, followed by the synthesis of early and late gene products.^28,29 The HSV synthesizes five IE proteins as infected cell polypeptides (ICP) 0, 4, 22, 27, and 47. Except for ICP47, all these proteins are reported to affect the expression of viral genes, and the genes encoding ICP22 and ICP47 can be deleted from the viral genome with no effect on the growth and viability of the virus in most cell types.^30,31Mutant HSV-1 deleted for the ICP0 gene grows poorly in cell culture, particularly at low MOIs but remains viable. ³² On the other hand, viral mutants carrying deletions or temperature-sensitive lesions in ICP4 and ICP27 demonstrated that these proteins provide essential functions during productive infection.^26,27 We previously reported that Harmaline, an isolated βC alkaloid, recruits lysine-specific demethylase-1 and it binds on IE complex of ICP0 promoter which leads to the inhibition of viral IE gene synthesis and suppressed the expression of ICP4 and ICP27. ⁶ It could hence be suggested that both the test analogs interfere with the viral IE-transcriptional events, which are a critical component of herpes virus reactivation. Our study demonstrated that the test analogs may help to prevent the multiplication of HSV and provide an interesting molecular target for the development of lead for better therapeutic and efficient management of HSV infection with minimal animal toxicity (safe up to 200 mg/kg, shown in supporting information).