Palladium-Catalyzed Intramolecular Carbopalladation/Cyclization Cascade: Access to Polycyclic N-Fused Heterocycles

Abstract

An efficient palladium-catalyzed intramolecular carbopalladation/cyclization cascade toward tetra- and pentacyclic N-fused heterocycles has been developed. This transformation proceeds via the palladium-catalyzed coupling of aryl halides with internal propargylic esters or ethers followed by the 5-endo-dig cyclization leading to polycyclic pyrroloheterocycles in moderate to excellent yields.

Nitrogen-containing heteroaromatic molecules and their analogues are pharmaceutically significant scaffolds, widely present in naturally occurring and synthetic biologically active molecules.¹ For example, molecules containing an indolizine motif, such as Lamellarin D² and other closely related cores,³ exhibit a wide array of biological activities, including human DNA topoisomerase I inhibition,⁴ ability to reverse multidrug resistance,⁵ and to induce apoptosis through a mitochondria mediated pathway toward broad range of cancer cell lines.⁶ These significant biological activities brought substantial attention to the synthesis of Lamellarins and their analogues.^7,8 Although several routes toward their core exist, new methods allowing for the efficient construction of heterocycles, particularly those with modified core and different substitution patterns, are in high demand. We have recently reported a two-component coupling methodology toward fully substituted N-fused heterocycles.^9,10 Although quite general with respect to the heterocyclic core, this approach is limited to the synthesis of bicyclic and tricyclic heteroaromatic molecules only (eq 1). Herein, we report a novel Pd-catalyzed arylation cyclization cascade that allows for easy assembly of various tetra-

(1)

and penta-cyclic isochromanone-annelated and other heterocycles.

We envisioned, that internal Ar-Pd-X species would undergo carbopalladation of the propargylic moiety of 1 with subsequent 5-endo-dig cyclization to produce 2 (Table 1). This idea was tested on cyclization of easily available propargyl-containing pyridine 1¹¹ in the presence of PdCl₂(PPh₃)₂ catalyst. However, only trace amounts of desired product 2 were observed (Table 1, entries 1-3). Switching to PdCl₂(MeCN)₂ catalyst did not provide any improvement of reaction outcome (entries 4-6). However, employment of the ligand-free catalyst Pd(OAc)₂ led to considerable enhancement of the reaction yields (entries 7-9). Solvent and concentration screening revealed the optimal conditions (entry 12).

Table 1.

Optimization of Reaction Conditions^a

entry	X	Pd	base	solvent	yield, %b
1	Br	PdCl2(PPh3)2	Cs2CO3	NMP	-
2	Br	PdCl2(PPh3)2	K3PO4	NMP	-
3	Br	PdCl2(PPh3)2	K2CO3	NMP	traces
4	Br	PdCl2(MeCN)2	Cs2CO3	NMP	-
5	Br	PdCl2(MeCN)2	K3PO4	NMP	-
6	Br	PdCl2(MeCN)2	K2CO3	NMP	traces
7	Br	Pd(OAc)2	Cs2CO3	NMP	25
8	Br	Pd(OAc)2	K3PO4	NMP	32
9	Br	Pd(OAc)2	K2CO3	NMP	37
10	Br	Pd(OAc)2	K2CO3	DMF	72c,d
11	Br	Pd(OAc)2	K2CO3	NMP	68c,d
12	Br	Pd(OAc)2	K2CO3	DMA	80c,d
13	I	Pd(OAc)2	K2CO3	DMA	21c,d

^aReactions were run in the presence of 5 mol % of Pd-catalyst in appropriate solvent (0.5M) at 130 °C for 8 hours unless otherwise noted.

^bGC/MS yields.

^cReactions was performed in appropriate solvent (0.33M) at 130 °C for 8 hours.

^dIsolated yield.

With these optimized conditions in hand, the scope of this cascade cyclization was examined (Table 2). Hence, benzyloxy-propargylic esters 1, possessing different alkyl (entries 1-3), alkenyl (entry 4), aryl (entries 5-7) substituents at the triple bond underwent smooth conversion to give the corresponding polycyclic heterocycles 2a-g in good yields. The generality of this transformation was further extended by employment of a number of functionalized benzyloxy-propargylic esters that smoothly were converted into the corresponding polycyclic N-fused heterocycles 2h-k (entries 8-11). Importantly, this transformation performed with comparable efficiency with other heterocyclic cores; isoquinoline and quinoline propargylic ester were effectively transformed to the pentacyclic N-fused heterocycles 2l and 2m, respectively (entries 12, 13).

Table 2.

Carbopalladation/Cyclization Reaction of Propargylic Esters 1^a

entry	1		2		yield, % b
1		1a		2a	80
2		1b		2b	65
3		1c		2c	71
4		1d		2d	70
5		1e		2e	88
6		1f		2f	76
7		1g		2g	84
8		1h		2h	91
9		1i		2i	64
10		1j		2j	90
11		1k		2k	73
12		1l		2l	58
13		1m		2m	72

^aAll reactions were performed on 0.15 mmol scale in DMA (0.33 M) at 100 to 130 °C for time specified in the Supporting Information.

^bYield of the isolated product after flash chromatography on silica gel.

Furthermore, we found that propargylic ethers 3¹² could also be utilized in this transformation providing access to another polycyclic scaffold, employing slightly modified reaction conditions¹³ (Table 3). Interestingly, this reaction proved even more general with respect to the functional group compatibility (Table 3, entries 1-6). Pentacyclic heterocycle 4g was also efficiently obtained using this methodology.

Table 3.

Carbopalladation/Cyclization Reaction of Propargylic Ethers 3^a

entry	3		4		yield, %b
1		3a		4a	63c
2		3b		4b	26c
3		3c		4c	38c
4		3d		4d	35c
5		3e		4e	64c
6		3f		4f	47c
7		3g		4g	70c
8		3h		4h	83c
9		3i		4i	71d
10		3j		4j	64d
11		3k		4k	68d
12		3l		4l	48d
13		3m		4m	83d

^aAll reactions were performed on 0.15 mmol scale in appropriate solvent (0.33 M) at 120 to 135 °C for time specified in the Supporting Information.

^bIsolated yield.

^cReaction conditions A: Pd(OAc)₂ (5 mol %), K₂CO₃ (2.0 equiv), n-Bu₄NCl (1.0 equiv), DMF, 120 °C.

^dReaction conditions B: Pd(OAc)₂ (5 mol %), K₂CO₃ (2.0 equiv), LiCl (1.0 equiv), p-xylene, 135 °C.

Notably, this transformation could also be successfully performed on the carbocyclic analog, giving access to the fused indenofuran derivative 4h (entry 8). Propargylic silyl ethers 3i-m also proved efficient to undergo the carbopalladation/cyclization sequence, thus affording access toward novel silacyclic scaffolds (entries 8-13).⁹

In summary, we have developed a new synthetic protocol for rapid and efficient assembly of three distinct tetra- and pentacyclic cores from easily accessible starting materials. This cascade carbopalladation/cyclization approach is complementary to the previously developed methods^10,11,14 toward multi-substituted N-fused heterocycles.