Synthesis of substituted pyrazines from N -allyl malonamides Society of Chemistry

The synthesis of pyrazines is described using a sequence that begins with the diazidation of N -allyl malonamides followed by thermal or copper-mediated cyclization. The pyrazine products possess an ester and a hydroxy group at 2- and 3-positions of the heterocyclic core, while alkyl- and aryl groups may be introduced at the other positions. We also show how to modify the pyrazines obtained with our method; examples regarding alkylations, side-chain brominations, hydrogenations and cross-couplings are presented. temperatures to markedly reduced yields. A range of Scheme 1 Formation of heterocycles through

Amongst the countless classes of nitrogen-containing heterocycles, pyrazines 1 are distinguished from their kin by the ease of how to access the symmetrical members. For example, the selfcondensation of a-amino ketones easily provides symmetrical pyrazines in an efficient way. 2 However, this approach provides mixtures of pyrazines when two different a-amino ketones are used. 3 The condensation of 1,2-diamines with 1,2-diketones (followed by dehydrogenation) is another typical method for the preparation of the pyrazine core that has severe limitations with respect to the regiospecity. 4 In addition to these classical methods, a broad range of alternative methods were developed with the goal of addressing as many substituents as possible. 5 Pyrazines were synthesized starting from a-hydroxy ketones, 6 ahalo ketones, 7 nitro epoxides, 8 and 2H-azirines. 9 The pyrazine core was modied through lithiation; 10 halogenated pyrazines were submitted to palladium-catalyzed cross-coupling reactions. 11 A surprisingly small number of reports describe the synthesis of the pyrazine core through cyclization strategies, including electrocyclizations and cyclocondensations. 12 As part of our ongoing studies to unveil the reactivity of geminal diazido compounds, 13 we recently showed that 2,2diazido-3-oxohept-6-enoates gave 3-hydroxypyridines upon thermolysis (Scheme 1). 14 We then questioned whether, starting with diazido compounds 1 derived from N-allyl malonamides, pyrazines 2 may become accessible in a fully analogous way. The challenging and densely substituted pyrazine core of 2 shows potential to produce several variants through further heterocycle modifying endeavors.
Earlier studies demonstrated how nitrogen-containing heterocycles can be synthesized from precursor compounds possessing the geminal diazido moiety. 15 To our knowledge, this is the rst study to successfully convert diazidated N-allyl malonamides into pyrazines through simple thermolysis. We began this study with the synthesis of the diazides 1 as summarized in Scheme 2. To this end, the malonic acid monoethylester 3 was submitted to standard amide-forming conditions using the respective allylic amine 4, HOBT and EDC at room temperature in dichloromethane. The subsequent diazidation of the N-allyl malonamides 5 was achieved with iodine and sodium azide in aqueous DMSO at room temperature, a method we recently developed as a general entry toward diazides derived from 1,3-dicarbonyl compounds. 16 This two-step sequence provided an easy and rapid access to a range of diazidated N-allyl malonamides (1a-1e). Caution! Diazidated compounds are potentially hazardous and should be handled with care. Although we never encountered any problems, we advise the use of protective gear, in particular for scales > 1 mmol.
To our delight, conventional heating of geminal diazide 1a (R ¼ H, R 0 ¼ H) in xylene gave the desired pyrazine core. Moderate yields were obtained when the reaction was carried out in xylene at 140 C with a concentration of 0.05 mol L À1 affording the product 2a in 43%. Unfortunately, our search for better reaction conditions was mainly fruitless: other solvents and temperatures led to markedly reduced yields. A range of Scheme 1 Formation of heterocycles through the thermolysis of geminal diazides. transition metal complex additives had simply no effect on the pyrazine formation; in a few cases (e.g., with iron and palladium salts) pyrazine products were not even detected, despite complete consumption of the diazides. The use of stoichiometric amounts of copper(I)iodide in acetic acid at 120 C, on the other hand, led to a substantial increase in yield, and the pyrazine 2a was obtained in 68% yield in a fully reproducible manner. Although these conditions appeared somewhat unusual in terms of stochiometry and solvent, they typically gave better yields for pyrazine formation, in direct comparison to the simple heating in xylene, while the number of unidenti-ed by-products was reduced.
As summarized in Scheme 3, several diazidated amides 1 were successfully converted into the pyrazines 2 by relying on either the pure heating in xylene (A) or the copper-mediated version in acetic acid (B). Besides the parent structure 2a with a methyl group at 6-position (R ¼ H, R 0 ¼ H), the method also allowed for the construction of pyrazines having other substituents at 6-position (e.g., 2b with R 0 ¼ Ph). It was also possible to introduce additional substituents at 5-position as exemplied through the formation of 2c (R ¼ Ph) and 2d (R ¼ nPr).
We point out that 3-hydroxy-2-pyrazinecarboxylic acid derivatives 2 are, in priciple, capable of prototropic tautomerism. The IR spectra of the neat compounds measured with ATR probes actually suggest the existence of the tautomeric 3oxo-3,4-dihydropyrazine form since the spectra typically show IR signals for two independent carbonyl groups. On the other hand, NMR data in CDCl 3 -solution appear to provide evidence for the aromatic pyrazine system: as exemplied with compound 2a, the chemical shis of À42.0 ppm for N-1 and À85.4 ppm for N-4 (approximated from 1H-15N-HMBC correlations) indicate the pyrazine core. This view is further supported by the downeld shi for H-5 (8.29 ppm, 1 H NMR spectrum). Although NMR data on derivatives of pyrazinecarboxylic acid are rare, 17 a recent study by Holzer, Eller and co-workers 18 on 3-amino-2-pyrazinecarboxylic acid and 3,4dihydro-3-oxo-2-pyrazinecarboxylic acid conrms our structural assignment and suggests that we have an unusual case where the 3-hydroxy-2-pyrazinecarboxylic acid core is existent in solution, rather than its more common 3-oxo-3,4dihydropyrazine tautomer.
Our rationale for the reaction is analogous to what we proposed for the formation of 3-hydroxypyridines through the thermolysis of diazides: 14 the reaction starts with an intramolecular 1,3-dipolar cycloaddition between one azide moiety and the olen to provide a triazoline intermediate. Aer loss of elemental nitrogen and a 1,2-hydride shi, elimination of hydrazoic acid may lead to the heterocyclic product (Scheme 4). As shown in eqn (1), the reaction of diazide 6 having an alkyne moiety led to the bicyclic compound 7 where the triazole substructure remained stable, and the hydrazoic acid was still attached. This structural assignment was supported by NMR, HRMS and IR data, including the characteristic IR band at 2133 cm À1 . (1) Due to the great importance of exibility and diversity in the application of this method, the direct modication of the 3-hydroxypyrazine 2a was then investigated as summarized in Scheme 5. We showed that it is easily possible to modify the parent structure 2a in various ways: for example, the nitrogen was selectively methylated with methyl iodide and cesium carbonate in acetonitrile at room temperature, leading to the Scheme 2 Synthesis of gem-diazido N-allyl malonamides.

Scheme 3 Synthesis of pyrazines 2. [a] ¼ mixture of inseperable compounds.
Scheme 4 Possible mechanism for the formation of pyrazine 2a.
N-methylated dihydro pyrazinone 8 (path a). On the other hand, the alkylation of the hydroxy oxygen was achieved through the use of silver carbonate giving the ethers 9a (with X ¼ Me) and 9b (with X ¼ iPr) (path b). High yields were also obtained for the silylation (e.g., formation of 9d and 9e), and the triate 9c was formed with triic anhydride under standard conditions (path b). The latter compound became an ideal starting point for further diversications with classical palladium-catalyzed cross coupling methodologies. For example, Suzuki coupling of 9c with various boronic acids provided a range of interesting structures 10a-e featuring the densely functionalized pyrazine core (path c). With pyrazines not possessing a free hydroxyl, a further functionalization of the methyl group at 6-position was possible through radical bromination under Wohl-Ziegler conditions; for example, the brominated pyrazines 11a and 11b were formed in moderate yields. The hydrogenation of the aromatic core was achieved using rhodium on alumina in ethanol at 30 bar of hydrogen. The oxo-piperazine 12 was isolated as a mixture of diastereomers in 87% yield. Unfortunately, we were not able to install halogen atoms at the 5-position of the existing pyrazine core 9a; all attempts using metalation or electrophilic substitution protocols failed, at least in our hands.

Conclusions
In summary, we have introduced a new and simple method for the synthesis of densely substituted pyrazines 2. Starting from N-allyl malonamides, the title compounds are accessible through a two-step sequence consisting of diazidation and subsequent thermolysis. We have also shown how the parent pyrazine structure 2 can be further modied at almost every position, in a simple and straightforward manner, thus rendering the pyrazines 2 highly valuable as building blocks for the synthesis of naturally occurring and pharmaceutically active compounds.

Experimental
General procedure for the synthesis of gem-diazido N-allyl malonamides 1 N-Allyl malonamides 5 (1 eq.) were dissolved in DMSO/water (2/ 1) (0.1 M), and NaN 3 (4 eq.) and NaHCO 3 (3 eq.) were added under heavy stirring. The suspension was cooled to 0 C, and iodine (2.05 eq.) was slowly added. The reaction maintained at this temperature for 10 min before warming to room temperature. The mixture was stirred at room temperature until complete consumption of the starting material (monitored via TLC) ($2 h). The reaction mixture was quenched with aqueous saturated solution of Na 2 S 2 O 3 until complete discoloration appeared. The mixture was extracted with EtOAc, and the combined organic phases were washed with brine, dried over anhydrous MgSO 4 and concentrated under reduced pressure. Purication by ash-chromatography on silica gel furnished the corresponding gem-diazido N-allyl malonamides 1. Caution! Diazidated compounds are potentially hazardous. We advise the use of protective gear, in particular for scales > 1 mmol.
General procedure for the synthesis of pyrazines 2 A: The reaction was carried out under argon atmosphere. Gemdiazido N-allyl malonamide 1 (1 eq.) was dissolved in xylenes (0.05 M) and stirred at 140 C for 24 h. The reaction mixture was cooled to room temperature and the solvent was evaporated under reduced pressure. Purication by ash-chromatography on silica gel furnished the corresponding pyrazines 2.
B: The reaction was carried out under argon atmosphere. Gem-diazido N-allyl malonamide 1 (1 eq.) was dissolved in glacial acetic acid (0.05 M). Copper(I)iodide (1.5 eq.) was added, and the reaction was stirred at 120 C for 24 h. The reaction mixture was cooled to room temperature and ltered through a short pad of celite. The ltrate was evaporated under reduced pressure. Purication by ash-chromatography on silica gel furnished the corresponding pyrazines 2.

Conflicts of interest
There are no conicts to declare.