No abstract

Methyl 5-bromolevulinate (30%) is obtained directly by brominating levulinic acid in methanol. The synthesis of 5-aminolevulinic acid over the 5-phthalimido ester thus becomes a three-step one.

Ethyl esters of 2-alkyl-and 2,4-dialkylpyrrole-3-carboxylic acids are obtained generally by extensions of the Hantzsch synthesis, benzyl and t-butyl esters when the 2-alkyl group is methyl. Hemopyrrole is obtained from butanal and ethyl acetoacetate in three steps. Pyrroles bearing higher alkyl groups or carbobenzoxy groups are reductively alkylated like the corresponding methylpyrroles and carbethoxy derivatives; t-butyl esters d o not survive.

UROPORPHYRIN 4389min., 0.5 g. of sodium acetate in 10 ml. of acetic acid was added and air passed in for 24 hr. to precipitate the crystalline product (29 mg., 667,). Its solution in 250 ml. of chloroform was filtered through alumina (Grade V) and the filtrate concentrated to 30 ml. The crystals which separated were dissolved in 60 ml. of hot chloroform. After concentrating to 30 ml. and cooling, flat needles (23 mg.), m.p. >350", separated from the solution. The visible spectrum of a chloroform solution recorded spectrophotometrically was of the "etio" type in contrast to that noted visually": maxima (and e) a t 521 mp (13 X lo3), 558 mp (7.1 X lo3), 598 mp(5.1 X lo3) and 655 mp (4.5 X lo3).Anal. Calcd. for C36H38N400: C , 66.04; H, 5.85; I T , 8.56; OEt, 27.53. Found: C, 66.24; H , 5.90; N, 8.63; OEt, 27.23. (b).-When this synthesis was carried out in the same way except that an equivalent amount of 707, perchloric acid was used instead of the hydriodic acid, aeration precipitated about 1 mg. (27,) of a product shown spectroscopically to contain 0.27 mg. of porphyrin.(c).-The aldehyde \'Ib (102 mg.) was heated with concentrated hydrochloric acid on the steam-bath. The precipitate was dissolved in chloroform, the solution filtered through a column of deactivated alumina and concentrated. The product which crystallized (7 mg.) had the same visible spectrum in chloroform as did that of method ( a ) above. In acetic acid-hydrochloric acid both had strong bands a t 626 and 575 mp, a very weak one a t 530 mp. Evidently Fischer observed the first two bands and a third band due to an impurity.12 1,4-Dimethyl-2 ,J-dicarbethoxyporphin-5,8-dipropionicAcid-6,7-diacetic Acid Tetramethyl Ester.-The pyrromethanes VIa (16 mg.) and Ie (26 mg.) were dissolved in 25 ml. of acetic acid, and acetic acid containing 0.16 ml. of 56% hydriodic acid was added. After 20 min., 0.5 g. of anhydrous sodium acetate was added. Aeration precipitated 16 mg. (407,) of the crystalline product, m.p. 250.5-251.5'.Anal. Calcd. for C42H48N4o12: C, 63.15; H , 5.80; S , 7.01; alkoxy (as OMe), 23.31. Found: C, 62.90; H, 5.7i; K, 6.89; alkoxy (as OMe), 22.38.2,3-Dimethylporphin-6,7-diacetic 4,5,pionic Acid Hexamethyl Ester.--ii solution of 175 mg. of the pyrromethene hydrobromide \;,a1 122 mg. of the pyrromethane IC and 300 mg. of anhydrous sodium acetate in 200 ml. of acetic acid was heated on the steam-bath for 1 hour. The cooled solution was then aerated for 1 hour. The crystalline but impure product (100 mg.) was separated and washed with acetic acid, then with ether. I t was esterified with methanolic hydrogen chloride, brought into chloroform, the solution filtered through deactivated alumina and the chloroform replaced by methanol. Crystallized again from chloroform-methanol it formed tiny bent hairs (82 mg., 3370), m.p. 233-238' after sintering and a solid phase change from 230". Paper c h r o m a t~g r a p h y~~ of the free acid showed only the hexacarboxylic acid spot. I t was degraded to coproporphyrin 2 methyl ester, m.p. 285-287" (uncor.).

5-Amino-4-carbethoxythiazoles (111) are readily converted into the corresponding thioureidoderivatives (V) : the latter lose the elements of ethanol in cold dilute sodium hydroxide, passing into thiazolopyrimidines (VI) which are analogues of purines. Attempts t o synthesise such compounds by attaching the thiazole to the pyrimidine ring were unsuccessful, one approach leading to the related ring system present in the tbiazolopyrimidine (IX).ONE of the prime objectives of the work described in this series has been, almost from its inception, the provision of simple and flexible methods of synthesising purines. The present memoir describes some experiments culminating in the preparation of thiazolopyrimidines which are skeletally similar to the purines. The following paper (Part XIII) records some further compounds of the thiazolopyrimidine group obtained by a somewhat different route, and a proof of the structures assigned by the conversion of one of the new compounds into a true purine of known structure. Finally, Part XIV generalises the method as one of wide applicability for the facile production of purines.In recent years several syntheses of purines have been described in which the pyrimidine ring is affixed to an already existing iminazole group. Thus Sarasin and Wegmann (Helv. Chim. Acta, 1924, 7, 713) used 4-amino-5-carbamyl-1-methylglyoxaline (I ; R = H), completing the 6-membered ring by the use of ethyl carbonate, and related compounds (I; R = Me or Et) with ethyl CO*NHR NH, \ / 7==? NMe N ' c' Mann and Porter (J., 1945, 751) used the chloroformate. The related synthesis of \ C < (111.) purines by Allsebrook, Gulland, and Story ( J . , 1942, 232) may be said to consist in effecting the bond a in the skeletal structure (11) ; those of Montequi ( A n d . Soc. Fis. Quirn., 1927, 25, 182)and of Baxter and Spring (J., 1945, 232) in making the link b, and that of Mitter and Chatterjee ( J . Indian Chem. SOC., 1934, 11, 867) in closing the bond c. None of these reactions appeared, however, to afford the desired flexibility and certain of them at least used intermediates which were the result of a series of tedious and sometimes drastic operations.It was observed (Cook, Heilbron, and Levy, J., 1947, 1598) that interaction of carbon disulphide and ethyl aminocyanoacetate yielded the thiazole (111; R = SH). In the present