One-Pot Synthesis of Indolo[2,3-c]quinolin-6-ones by Sequential Photocyclizations of 3-(2-Azidophenyl)-N-phenylacrylamides

Abstract

A one-pot synthesis of indolo[2,3-c]quinolin-6(7H)-ones was achieved by sequential photocyclizations of 3-(2-azidophenyl)-N-phenylacrylamides in moderate to high yields. The reactions proceeded via photochemical cyclization of aryl azides to form N-phenylindol-2-carbamides and subsequent 6π-electrocyclic reaction and oxidative aromatization to afford the corresponding indolo[2,3-c]quinolin-6(7H)-ones.

Key words

indolo[2,3-c]quinolin-6(7H)-ones - photocyclization - 3-(2-azidophenyl)-N-phenylacrylamides - aryl azides - 6π-electrocyclic reaction

Indoloquinolines are present in many natural alkaloids[ ¹ ] and pharmaceuticals[ ² ] possessing important biological activities, such as alkaloid cryptolepine (A; Figure [1]) (5-methyl-5H-indolo[3,2-b]quinoline), cryptotackieine (B) (5-methyl-5H-indolo[2,3-b]quinoline) and cryptosanguinolentine (C) (5-methyl-5H-indolo[3,2-c]quinoline), all having potent antiplasmodial and antimalarial activities.[1] [2] [3] Isoneocryptolepine (indolo[2,3-c]quinoline) (D), a synthetic analogue of the above natural products, also shows low resistant activity to chloroquine and pyrimethamine, but has much better selectivity index (cytotoxicity/antiplasmodial activity ratio) than cryptolepine, which makes it a better compound for further validation of the indoloquinolines as a potential antiplasmodial drugs. The indoloquinoline alkaloids have been used extensively as lead compounds for the discovery of new antiplasmodial substances.

Figure 1

Many reports have been found for the synthesis of indoloquinoline alkaloids,[1] [2] [3] [4] but for the unnatural isoneocryptolepine (indolo[2,3-c]quinoline) derivatives, especially indolo[2,3-c]quinolin-6(7H)-ones which was found to possess antitumor activities,[ ⁵ ] only a few of synthetic methods were found in literature,[ ⁶ ] for example, Pd(OAc)₂-catalyzed intramolecular Heck reaction of N-(2-iodophenyl)-1H-indole-2-carboxamides[ ^6a ] or of 3-(2-bromophenylamino)quinolines;[ ^6b ] photostimulated intramolecular S_RN1 reactions of N-(2-chlorophenyl)-1H-indole-2-carboxamides;[ ^6c ] and cyclocondensation of 3-formaloxindoles with phenylhydrazines,[ ^6d ] and oxidative photochemical cyclization of N-phenylindole-2-carboxamides.[ ^6e ] However, all these methods generally required either an indole or a quinoline as the precursor and then in order to reach to indolo[2,3-c]quinolin-6(7H)-ones structure, construction of another ring.

We have been long interested in the synthesis of polycyclic heterocycles by one-pot multistep reactions. Recently, we reported a new synthesis of indolo[3,2-c]quinolin-6-ones by the photoreaction of 3-(2-azidobenzylidene)indol-2-ones.[ ⁷ ] In this route, photocyclization of 3-(2-azidobenzylidene)indol-2-ones first lead to formation of indole ring and subsequent ring expansion of indol-2-one afford the quinoline ring. Here, we report a new method for the synthesis of indolo[2,3-c]quinolin-6(7H)-ones (isoneocryptolepinone) by a one-pot two-step photocyclization of 3-(2-azidophenyl)-N-phenylacrylamides (Scheme [1]).

Scheme 1 One-pot two-step photocyclization of 3-(2-azidophenyl)-N-phenylacrylamides

The reactions proceeded via photochemical cyclization of aryl azides 1 to form the corresponding indole ring 2 first and then photochemical 6π-electrocyclic reaction to form the cyclic dihydroquinolin-2-one intermediate which is aromatized by oxidative dehydrogenation to give the product indolo[2,3-c]quinolin-6(7H)-ones 3.

We first optimized the photoreaction conditions using 3-(2-azidophenyl)-N-phenylacrylamide (1a) as a model substrate which was prepared by the Wittig reaction of 2-azidobenzaldehyde with triphenyl phenylcarbamoylmethyl phosphonium salt according the reported procedure (Scheme [2]).[8a] [b]

The reaction of 1a was examined in different oxygen-saturated solvents (Scheme [3]) in a Pyrex flask (λ ≥280 nm) because it was suggested that the irradiation of aromatic amide systems with the 254 nm light largely induced photo-Fries rearrangement,[ ⁹ ] but the irradiation with λ ≥300 nm light rather promoted cyclization.[ ^6e ]

Scheme 2 The procedure for the preparation of 1a

As shown in Table [1], the reaction efficiency was not satisfactory in nonpolar solvent benzene or polar solvent acetonitrile because of the low solubility of substrate 1a in benzene and no 3a was formed in acetonitrile except for the indole-2-carbamide 2a. Thus, the photoreaction was then examined in a mixture of solvents. It was observed that the photoreaction efficiency of 1a increased in solvent mixture and reached its highest value in a 9:1 mixture of benzene and dichloromethane. This result was probably ascribed to the solvent effects on the two photocyclizations because the weak polar solvents such as benzene, cyclohexane and dioxane favored the oxidative photocyclization of amide systems,[6e] [10] whereas polar solvents such as dichloromethane and acetonitrile were favorable to the photochemical cyclization of aryl azides[ ⁷ ] and photo-fries rearrangement.[ ⁹ ] The addition of small quantity of dichloromethane to benzene not only enhanced the reaction of aryl azides, but also increased the solubility of substrates. Therefore, irradiation of substrates 1 in a mixture of solvents benzene and dichloromethane (9:1) in Pyrex flask was selected as the optimal conditions for the synthesis of indolo[2,3-c]quinolin-6(7H)-ones.

Scheme 3 Photoreaction of 1a in different solvents

Table 1 Optimization of Reaction Conditions for the Synthesis of 3a ^a

Entry	Solvent	Time (h)	Conv. (%)b	Yield (%)c
1	MeCN	20	93	0d
2	benzene	20	83	75e
3	benzene–acetone (5:1)	20	70	30
4	benzene–EtOH (5:1)	16	82	45
5	benzene (5:1)	15	85	68
6	benzene–CH2Cl2 (2:1)	14	89	66
7	benzene–CH2Cl2 (5:1)	12	90	70
8	benzene–CH2Cl2 (9:1)	11	96	76
9	benzene–CH2Cl2 (15:1)	13	91	70

^a Reaction conditions: compound 1a (128 mg, 0.5 mmol) was dissolved in anhyd benzene (54 mL) and anhyd CH₂Cl₂ (6 mL). The solution was bubbled with oxygen for 10 min and irradiated at λ ≥280 nm with a medium-pressure mercury lamp (500 W) at ambient temperature.

^b Conversions calculated on the basis of substrate.

^c Yields of isolated product based on consumed substrate.

^d No 3a was separated except for the N-phenyl-2-indolecarbamide (2a).

^e The substrate 1a was not dissolved completely because of its low solubility in benzene.

Scheme 4 A plausible photoreaction mechanism of 1a

We then examined the photocyclization of a group of 3-(2-azidophenyl)-N-phenylacrylamides with different substituents on both azidobenzene ring and aniline ring under the optimized conditions.[ ¹¹ ]The photoreactions of all these substrates (1b–p) afforded the corresponding cyclization products (3b–p) in moderate to high yields after irradiations for 5–25 hours at room temperature (Table 2). All products were fully identified by ¹H NMR, ¹³C NMR, HRMS and IR.[ ¹¹ ] It can be observed from Table 1 that the photoreaction efficiency was different for the substrates with different substituents. The electron-donating groups, such as the methoxy and methyl groups (1f and 1g) on the aniline ring, retarded the photoreaction of 1f and 1g and more time was needed to reach the high conversion (entries 6 and 7; 18 h and 16 h, respectively). In contrast, electron-attracting groups, such as chlorine atom (1h, 1i) and carboxylate group (1j) had no significant effects on the photoreaction rate of 1h–j (entries 8–10) in comparison with 1a, but deceased the yields of 3h–j greatly (entry 10). Moreover, the ortho-ethoxycarbonyl group in 1k led to a significant decrease in both the reaction rate and the yield of product 3k (entry 11). These results were probably derived from the decrease of pericyclic reaction probability of the intermediate product 2k and the increase of photo-Fries rearrangement products of 2k because one ortho-position was blocked in 2k. In fact, small quantities of decomposition product methyl 2-aminobenazoate and photo-Fries rearrangement products could be separated from the reaction mixture besides 2k and 3k after irradiation of 1k for 25 hours. This kind of result was also reported in the photocyclization of benzanilides under the irradiation at λ ≥280 nm.[ ^9b ] On the other hand, the substituents on the azidophenyl ring also had some effects on the photoreactions of 1m–p. For example, meta-chlorine atom in 1l,m (entries 13 and 14) accelerated the photoreaction in comparison to that of 1b,c (entries 2 and 3). Differently, para-ethoxycarbonyl group led to the decrease of both the reaction rate of 1l,m and the yield of products 3l,m (entries 13 and 14) as compared with the reactions of 1d,e (entries 4 and 5); the alkoxy groups in 1l lowered the yield of 3l (entry 12) as compared with that of 3d (entry 4) although the reaction times needed were similar.

Table 2 Photocyclization of (E)-3-(2-Azidophenyl)-N-phenylacrylamides in Benzene–Dichloromethane^a

Entry	Substrate		Time (h)	Conv. (%)b	Product		Yield (%)c
1	1a		11	96	3a		76
2	1b		15	92	3b		68
3	1c		10	94	3c		60
4	1d		6	98	3d		78
5	1e		5	99	3e		85
6	1f		18	92	3f		75
7	1g		16	90	3g		65
8	1h		11	96	3h		80
9	1i		12	96	3i		83
10	1j		9	95	3j		52
11	1k		25	93	3k		40
12	1l		5	96	3l		60
13	1m		8	95	3m		57
14	1n		7	95	3n		56
15	1o		9	95	3o		50
16	1p		9	94	3p		52

^a Reaction conditions: compound 1 (0.5 mmol) was dissolved in anhyd benzene (54 mL) and anhyd CH₂Cl₂ (6 mL). The solution was bubbled with oxygen for 10 min and irradiated at λ ≥280 nm with a medium-pressure mercury lamp (500 W) at ambient temperature.

^b Conversions calculated on the basis of substrate.

^c Yields of isolated product based on consumed substrate.

The intermediate products N-phenylindol-2-carbamides 2, produced in the first photocyclization of 1a–p (Scheme [1]), were monitored in the progress of photoreactions. It was found that no intermediate products 2 were detected in most cases except for the photoreaction of 1c, 1d, 1e and 1k in which a small quantity of intermediate 2c, 2d, 2e and 2k , respectively, could be separated. It meant that the oxidative photocyclization of intermediates 2 (the second photocyclization) proceeded more quickly than the photocyclization of 1 in most cases; only for 1c,1d, 1e and 1k, the photocyclization of intermediates 2c, 2d, 2e and 2k was slower than the photocyclization of 1c, 1d, 1e and 1k.

A reasonable mechanism to account for the formation of indolo[2,3-c]quinolin-6(7H)-ones is presented in Scheme [4]. Photocyclization of azide 1a is believed to occur via attack of nitrene Ia to the β-position of the adjacent double bond to form a zwitterion IIa [ ¹⁰ ] and the subsequent shift of hydrogen atom from β-carbon to nitrogen atom to give the intermediate aryl anilide 2a. Photochemical 6π-electrocyclic reaction of 2a forms the cyclic intermediate (IVa) which then is dehydrogenated by oxygen to give the final product 3a.[10a] [12]

In summary, we have developed an efficient method for the synthesis of indolo[2,3-c]quinolin-6(7H)-ones, indolo[2,3-c]pyrrolo[3,2,1-i,j]quinolin-7-ones and indolo[2,3-c]pyrido[3,2,1-i,j]quinolin-8-ones by a one-pot photocyclization of 3-(2-azidophenyl)-N-phenylacrylamides. The reactions proceed by two sequential cyclization reactions, namely photocyclization of 2-azidocinnamides and 6π-electrocyclic reaction of the formed intermediate aryl anilides.

Acknowledgment

We thank the National Natural Science Foundation of China (Grant Nos. 20872056 and J1103307) for financial support.

• References and Notes



• 1a Cimanga K, De Bruyne T, Pieters L, Claeys M, Vlietinck A. Tetrahedron Lett. 1996; 37: 1703
• 1b Sharaf MH. M, Schiff PL, Tackie AN, Phoebe CH, Martin GE. J. Heterocycl. Chem. 1996; 33: 239
• 1c Pousset J.-L, Martin M.-T, Jossang A, Bodo B. Phytochemistry 1995; 39: 735
• 1d Kirby GC, Paine A, Warhurst DC, Noamese BK, Phillipson JD. Phytother. Res. 1995; 9: 359
• 2a Onyeibor O, Croft SL, Dodson HI, Feiz-Haddad M, Kendrick H, Millington NJ, Parapini S, Phillips RM, Seville S, Shnyder SD, Taramelli D, Wright CW. J. Med. Chem. 2005; 48: 2701
• 2b Lisgarten JN, Coll M, Portugal J, Wright CW, Aymami J. Nat. Struct. Biol. 2002; 9: 57
• 2c Wright CW, Addae-Kyereme J, Breen AG, Brown JE, Cox MF, Croft SL, Gökçek Y, Kendrick H, Phillips RM, Pollet PL. J. Med. Chem. 2001; 44: 3187
• 2d Bierer DE, Dubenko LG, Zhang P, Lu Q, Imbach PA, Garofalo AW, Phuan P.-W, Fort DM, Litvak J, Gerber RE, Sloan B, Luo J, Cooper R, Reaven GM. J. Med. Chem. 1998; 41: 2754
• 3a Van Miert S, Hostyn S, Maes BU. W, Cimanga K, Brun R, Kaiser M, Mátyus P, Dommisse R, Lemière G, Vlietinck A, Pieters L. J. Nat. Prod. 2005; 68: 674
• 3b Cimanga K, De Bruyne T, Lasure A, Van Poel B, Pieters L, Claeys M, Vanden Berghe D, Kambu K, Tona L, Vlietinck AJ. Planta Med. 1996; 62: 22
• 3c Cimanga K, De Bruyne T, Pieters L, Claeys M, Vlietinck A. Tetrahedron Lett. 1996; 37: 1703
• 3d Cimanga K, De Bruyne T, Pieters L, Vlietinck AJ, Turger CA. J. Nat. Prod. 1997; 60: 688
• 3e Grellier P, Ramiaramanana L, Millerioux V, Deharo E, Schrével J, Frappier F, Trigalo F, Bodo B, Pousset J.-L. Phytother. Res. 1996; 10: 317
• 4a Kadam HK, Parvatkar PT, Tilve SG. Synthesis 2012; 44: 1339
• 4b Parvatkar TP, Parameswaran SP, Tilve GS. Curr. Org. Chem. 2011; 15: 1036
• 4c Kraus GA, Guo H. Tetrahedron Lett. 2010; 51: 4137
• 4d Mori T, Ichikawa J. Synlett 2007; 1169
• 4e Amiri-Attou O, Terme T, Vanelle P. Synlett 2005; 3047
• 4f Fresneda PM, Molina P. Synlett 2004; 1
• 5a Fukushima K, Teller MN, Mountain IM, Tarnowski GS, Stock CC. J. Reticuloendoth. Soc. 1979; 26: 187
• 5b Grunberg E, Kramer MJ, Buck M, Trown PW. Chemotherapy 1978; 24: 77
• 6a Putey A, Popowycz F, Do Q.-T, Bernard P, Talapatra SK, Kozielski F, Galmarini CM, Joseph B. J. Med. Chem. 2009; 52: 5916
• 6b Hostyn S, Maes BU. W, Baelen GV, Gulevskaya A, Meyers C, Smits K. Tetrahedron 2006; 62: 4676
• 6c Vaillard VA, Budén ME, Martín SE, Rossi RA. Tetrahedron Lett. 2009; 50: 3829
• 6d Tokmakov GP, Zemlyanova TG, Grandberg II. Chem. Heterocycl. Comp. 1994; 30: 434
• 6e Kanaoka Y, Itoh K. Synthesis 1972; 36
• 7 Shi Z, Ren Y, Li B, Lu S, Zhang W. Chem. Commun. 2010; 46: 3973
• 8a Evans DA, Black WC. J. Am. Chem. Soc. 1993; 115: 4497
• 8b Sun K, Liu S, Bec PM, Driver TG. Angew. Chem. Int. Ed. 2011; 50: 1702
• 9a Ferrini S, Ponticelli F, Taddei M. J. Org. Chem. 2006; 71: 9217
• 9b Mayouf AM, Park Y.-T. J. Photochem. Photobiol. A: Chem. 2002; 150: 115
• 10a Winterfeldt E, Altmann HJ. Angew. Chem. 1986; 80: 486
• 10b Hoboken NJ. Org. React. (N. Y.) 1984; 30: 457
• 11 Typical Experimental Procedure: A solution 1a (128 mg, 0.5 mmol) in a mixture of anhyd benzene (54 mL) and anhyd CH₂Cl₂ (6 mL) was bubbled with oxygen for 10 min and irradiated at λ ≥280 nm with a medium-pressure mercury lamp (500 W) in three 25-mL Pyrex flasks at ambient temperature. The progress of reaction was monitored by TLC at regular intervals. After the solvent was removed under reduced pressure, the residue was separated by silica gel column chromatography eluted with hexane–EtOAc (5:1) to yield the product 3a. Selected spectroscopic data for compounds: 3a: pale yellow solid; mp 294–298 °C. IR (KBr): 3430, 3147, 2921, 1655, 1617, 735 cm^–1. ¹H NMR (400 MHz, DMSO): δ = 7.30–7.36 (m, 2 H), 7.41 (td, J = 8.4, 1.6 Hz, 1 H), 7.45–7.52 (m, 2 H), 7.64 (d, J = 8.0 Hz, 1 H), 8.45 (d, J = 8.0 Hz, 1 H). 8.48 (d, J = 8.4 Hz, 1 H), 11.86 (s, 1 H), 12.35 (s, 1 H). ¹³C NMR (100 MHz, DMSO): δ = 155.67, 138.81, 134.93, 127.59, 125.88, 125.62, 122.97, 122.31, 122.29, 122.22, 120.64, 118.17, 118.01, 116.06, 113.02. ESI–HRMS: m/z [M + H⁺] calcd for C₁₅H₁₁N₂O: 235.0871; found: 235.0870.
• 12a Rowe PA. S. C. D, Hansen DW. Tetrahedron Lett. 1983; 47: 5169
• 12b Kanaoka Y, Nakao S, Hatanaka Y. Heterocycles 1976; 261