Efficient syntheses of 1,3-unsubstituted 1 H -pyrazolo[3,4-b ]quinolines

The synthesis of pyrazolo[3,4-b ]quinolines from 2-chloro-3-formylquinolines is described. Direct reaction of the chloro-aldehydes with hydrazine was unsuccessful, but a very efficient sequence was developed in which protection of the aldehyde group was followed by displacement of the halogen by hydrazine, release of the aldehyde then leading to spontaneous ring closure giving the tricycles. The products were obtained in excellent yields


Introduction
1H-Pyrazolo [3,4-b]quinolines are of interest in a number of diverse contexts: for example as antiviral agents 1 in particular as inhibitors of Herpex simplex virus type 1 replication, 2 as activators of caspases and inducers of apoptosis, 3 as antimicrobial agents, [4][5][6][7] as parasiticidic agents, 8 as antimalarials, 9 in photoluminescence 10 and electroluminescence 11 studies, and in electroluminescent devices. 12The use of 1H-pyrazolo [3,4-b]quinoline derivatives as optical brighteners has been long known 13 and interest in them as dyestuffs 14,15 and as colourants within polymers, 16 continues to the present.
Two main strategies have been employed for the synthesis of 1H-pyrazolo [3,4-b]quinolines: (1) some routes begin with a preformed pyrazole (or pyrazolone) the pyridine ring being constructed during the sequence; (2) other routes start from a preformed quinoline with the pyrazole ring being constructed during the sequence.The new work in this paper falls into the latter category.8][19] An alternative pyridine ring construction results from the condensation of a 5-halo-4-acylpyrazole with an aniline (Scheme 1 route b). 20,21ess frequently employed routes which also start from preformed pyrazoles include the condensation of 5-aminopyrazoles with ortho-halo aromatic aldehydes, 22 the Friedel-Crafts type closure of an ortho-pyrazol-5-ylamino benzoic acid 9 and the condensation of 4arylidenepyrazolin-5-ones with anilines. 23 Scheme 1 (2) The reaction of 2-chloro-3-cyanoquinolines with hydrazine producing 3-amino-1Hpyrazolo [3,4-b]quinolines (Scheme 2) has been used widely.

Scheme 2
There has been a similar wide interest in the combination of 2-chloro-3-acylquinolines with hydrazine or arylhydrazines to form 1H-pyrazolo [3,4-b]quinolines.In particular, and partly because of the ready availability of 2-chloro-3-formylquinolines from Meth-Cohn's convenient and efficient synthesis of such molecules from acetanilides and the Vilsmeier reagent, 29 most work has involved these chloro-aldehydes, though these are easily converted into corresponding ketones by Grignard addition then oxidation of the resulting alcohol. 24,30The reaction of such ketones with hydrazine or arylhydrazines produces 3-substituted 1H-pyrazolo [3,4-b]quinolines simply by refluxing in methanol (Scheme 3); 24,30 it was suggested 30 that the size of the ketone substituent favours formation of the Z-stereoisomer of a first-formed hydrazone, thus having the geometry required for a subsequent ring-closing displacement of the chloride.

Scheme 3
At the commencement of our work, there were conflicting reports as to the facility with which hydrazine and arylhydrazines would react, to give comparable products, with 2-chloro-3formylquinolines i.e. to produce 3-unsubstituted 1H-pyrazolo [3,4-b]quinolines.Thus, although some hetarylhydrazines were described as reacting with 6-bromo-31 or 7-methoxy-32 -2-chloro-3formylquinolines in methanol at reflux or with microwave heating, other reports of attempts to bring about such cyclocondensations implied difficulties.It was shown 28 for example, that an Ehydrazone is formed from such chloro-aldehydes which would not ring-close in refluxing ethanol, having the inappropriate imine geometry, however other workers apparently did not encounter such difficulties using hydrazine hydrochlorides in refluxing ethanol for extended reaction times, 33 or reflux in acetonitrile. 34Much more vigorous conditions can be applied to the interaction between 2-chloro-3-formylquinolines and hydrazine and arylhydrazines which overcome these difficulties: microwave heating with p-toluenesulfonic acid is recommended. 35n a similar study, a 2-methylthio-3-benzoylquinoline formed the undesirable hydrazone isomer which would not ring close; here again, the difficulty was overcome by reacting the ketone with hydrazine using microwave heating in the presence of p-toluenesulfonic acid. 36 device which was utilised first by Meth-Cohn 39,37 and later by Singh 28 to avoid the difficulties associated with formation of the undesired geometrical isomer of the hydrazone, was to initially protect the aldehyde as an acetal thus forcing a first step to be nucleophilic displacement of the chloride by the hydrazine, hydrolytic release of the aldehyde then leading to the desired pyrazolo-quinoline.

Results and Discussion
Our aim in this work was develop a simple and mild procedure for the synthesis of 1,3unsubstituted 1H-pyrazolo [3,4-b]quinolines (Scheme 4).We view such compounds as ideal starting materials for the flexible synthesis of a large range of substituted 1H-pyrazolo [3,4b]quinolines through manipulations at N-1 and C-3.Benzene ring substituents will be provided by the 2-chloro-3-formylquinoline ring syntheses, which, it will be recalled, are absolutely straightforward using Meth-Cohn's procedure and start from simple and readily available benzenamines.Alkylations at N-1 will involve pyrazole ring anions which can be easily formed using a strong base and which react with alkylating agents. 38N-Arylations will rely on highly efficient and well established copper- 39 or palladium-40 -catalysed processes.
The introduction of substituents at C-3 will involve the intriguing prospect of developing further the relatively little explored electrophilic substitution of indazoles, in this case 7azaindazoles.There is considerable interest in indazole analogues of the many biologically significant indoles, natural and synthetic, in the medicinal and biological fields.Halogenation of the target 1H-pyrazolo [3,4-b]quinolines, would be predicted to produce 3-halo-derivatives, with all the potential therein for formation of nucleophilic organometallic reagents or for use directly in cross-coupling processes.
2-Chloro-3-formylquinolines 1a-f were produced in good yields from precursor acetanilides (acetanilide itself, 2-methyl-, 3-methyl-, 4-methyl-, 3-methoxy-and 4-methoxyacetanilide). 29eaction of 1a with hydrazine produced a hydrazone which could not be cyclised by refluxing in ethanol.We concluded that the hydrazone had E geometry and thus was sterically prevented from attacking the quinoline C-2.In order to avoid this difficulty, each of the aldehydes was converted into an ethylene acetal 2, and aldehyde 1a was also converted into the corresponding thioacetal 5a.
With the aldehyde masked, each of the chloro-quinolines 2a-f and 5a was reacted with hydrazine hydrate in refluxing ethanol, giving the quinolin-2-ylhydrazines 4a-f and 5b in good yields.Finally, mild aqueous acidic removal of the acetal protection from 3a-f led directly, in one pot, to the cyclised 1H-pyrazolo [3,4-b]quinolines 4a-f.This protocol is far superior to the bismuth trichloride method employed 28 for the comparable ring closure of two examples, 3a and 3e, and requires nothing more than simple basification at the end of the process to liberate the

Experimental Section
General Experimental Procedures.All the 2-chloro-3-formylquinolines (1a-f) were prepared by the Meth-Cohn method. 29Melting points were recorded on a Philip Harris C4954718 apparatus and are uncorrected. 1H and 13 C NMR spectra were recorded on a Bruker Avance AQS 300 MHz spectrometer, at 300 MHz and 75 MHz respectively.Chemical shifts d are in parts per million (ppm) measured in CDCl 3 or DMSO-d 6 as solvent and relative to TMS as the internal standard.Infrared spectra were recorded on a Thermonicolet-Nexus 670 FT-IR instrument and elemental analyses were carried out on an Exeter analytical model CE440 (CHN) and a Leco elemental analyzer Truspec (CHNSO), and Cl was determined by Chemical Analyses Labs Ltd.

General method for the formation of ethylene acetals (2)
A solution of 2-chloro-3-formylquinoline (20 mmol) in benzene (100 mL) containing ethylene glycol (3.57g, 3.2 mL, 57 mmol) and a crystal of toluene-p-sulfonic acid was heated under reflux using a Dean-Stark water separator until no more water collected in the side arm (5-12 h).The cooled solution was treated with saturated aqueous sodium carbonate (50 mL), dried and evaporated giving 2 which was recrystallised from petroleum (b.p. 100-120 ºC)-toluene.

General method for the hydrazinolysis reactions giving 3
A solution of 2a-f (20 mmol) and hydrazine hydrate 80% (12.5 g, 12.1 ml, 200 mmol) was heated under reflux for 24 h.Then cooled solution was evaporated and residue recrystallised from ethanol to give 3a-f in high yields.In 50-mL flask fitted with a condenser and a dropping funnel was added cupric oxide (0.48 g, 6 mmol), anhydrous cupric chloride (1.61 g, 12 mmol) and acetone (40 mL).The resulting suspension was brought to reflux with vigorous stirring and a solution of 5b (1.31 g, 5 mmol) in acetone (9 ml) and DMF (1 mLl) was added over 5 min.Reflux was maintained for 30 min then the mixture was cooled and filtered.The insoluble material was washed with dichloromethane (3×20 mL) and the combined organic layers were washed with aqueous sodium carbonate (50 ml), dried over sodium sulfate and filtered.Evaporation of the solvent gave 4a as yellow powder in 85% yield. 7,[24][25][26][27][28]