J . Org. Chem., Vol. 44, No. 8, 1979 1301
4’-Hydroxymethyl Nucleosides and H. Shechter, J. Org. Chem., 40,2446 (1975);A. Diaz, J. Fulcher, R. Cetina, M. Rubio, and R. Reynoso, ibid., 40,2459 (1975). (7) H. E. Simmons, T. L. Cairns, S.A. Vladuchick, and C. M. Hoiness, Org. React., 20, l(1973). (8)D. C. Sanders and H. Shechter, J. Am. Chem. SOC., 95,6858 (1973). (9)D. I. Schuster and C. W. Kim, J. Org. Chem., 40,505 (1975). (10) For previous examples of such directive effects in the cyclopropanation of olefinic ketals, see A. de Meijere, C. Weitemeyer, and 0. Schallner, and I.M. Takakis and W. C. Agosta, J. Org. Chem. Ber., 110, 1504 l,1977), The latter paper also describes two cases in which Chem., 43, 1952 (1978). these effects are not observed. (11) Eu(fod)s is tris(6,6,7,7,8,8,8-heptafluoro-2,2-dimethyl-3,5-octandionato)europium. (12)A similar effect is seen in derivatives of simpler but structurallyrelated ring systems: S. C. Clarke, I(. J. Frayne, and 6. L. Johnson, Tetrahedron, 25, 1265 (1969).A more general discussion with references is given by L. M. Jackman and S. Sternhell, “Applications of Nuclear Magnetic Resonance Spectroscopy in Organic Chemistry”, 2nd ed., Pergamon Press, Oxford, 1969,pp 286 and 287. (13)For earlier examples of this effect in geometrically related systems see M. A. Battiste and M. E. Brennan, Tetrahedron Lett., 5857 (1966). (14)6. Halton, M. A. Battiste, R. Rehberg, C. L. Deyrup, and M. E. Brennan, J. Am. Chem. SOC., 89,5964(1967); R. Bresiow. G. Ryan, and J. T. Groves, S.C. Clarke and 6. L. Johnson, Tetrahedron, 27,3555 ibid., 92,988 (1970);
(1971). (15)W. L. Mock, J. Am. Chsm. SOc.,92,6918 (1970). (16)M. Roberts, H. Hamberger, and S.Winstein, J. Am. Chem. SOC.,92,6346 (1970). (17)H. Tsuruta, K. Kurabayashi, and T. Mukai, Tetrahedron Lett., 3775 (1967); W. Grimme, Chem. Ber, 100, 113 (1967). (18)Conditions tried included ethylene glycol with: (1)pyridinium chloride as catalyst, R. Rausser, A. M. Lyncheski, H. Harris, R. Grocela, N. Murrill, E. Bellamy, D. Ferchinger, W. Gebert, H. L. Herzog, E. 6. Hershberg, and E. P. Oiiveto, J. Org. Chem., 31,26 (1966); (2)ptoluenesulfonic acid, selenium dioxide, and methyl orthoformate in chloroform, E. P. Oliveto. H. Q. Smith, C. Gerold, L. Weber, R. Rausser, and E. 6. Hershberg, J. Am. Chem. Soc., 77,2224(1955); (3)gtoluenesulfonic acid and methyl orthoformate, A. de Meijere, C. Weitemeyer, and 0. Schallner, Chem. Ber., 110,1504
(1977). (19)I. M. Takakis and W. C. Agosta, J. Org. Chem., 43, 1952 (1978). (20)The photochemistry of Z!l and 25 is discussed by I. M. Takakis and W. C. Agosta, J. Am. Chem. Soc., in press.
(21)Conditions used were those given in detail by W. C. .Agosta and S.Wolff, J. Am. Chem. SOC., 98,4182(1976).Methanol was added to trap any ke-
tenes that might be formed.
(22)A preliminary report on the photochemistry of 15 has appeared: I. M. Takakis and W. C. Agosta, Tetrahedron Lett., 2387 1978). (23)This was prepared by thermolysis of cis,cis-bicyclo\6.1 .O]non-2-ene as described by D. S.Glass, R. S.Boikess, and S. Winstein, Tetrahedron Led, 999 (1966),and R. W. Thies, P.-K. Hong, R. Bushwell, and J. L. Boop, J. Org. Chem., 40,585 (1975). (24)D. Devaprabhakara, C. G. Cardenas, and P. D. Gardner, J. Am. Chem. Soc., 85, 1553 (1963). (25)R. Vaidyanathaswamy and D. Devaprabhakara, lnd. J. Chem., 13, 873 11975) \ -,.
(26)R. B. Turner and W. R. Meador, J. Am. Chem. SOC.,79,4133 (1957). (27)R. 6. Woodward and R. Hoffmann, Angew. Chem., 61,797 (1989);Angew. Chem., lnt. Ed. Engl., 8, 781 (1969). (28)S.J. Cristol and R. U. Barbour, J. Am. Chem. SOC., 90,2832 (1968);J. K. Kochi, P. J. Krusic, and D. R. Eaton, ibid., 91, 1877, 1879 (1969);E. C. Friedrich, J. Org. Chem., 34,528(1969);W. G. Dauben, L. Schutte, R. E. Wolf, and E. J. Deviny, ibid., 34, 2512 (1969);A. L. J. Beckwith and G. Phillipou, Chem. Commun., 658 (1971):6. Maillard, D. Forrest, and K. U. Ingold, J. Am. Chem. SOC.,98,7024 (1976). (29)Biradicals 29 and 31 are shown as discrete intermediates for clarity only;
-
decarbonylation concerted with fragmentation of the initial biradical ( i a , 28 + 26, and 30 27) need not be excluded. (See in this regard ref
4.) (30)For 38:see ref 25.For 39: J. G. Traynham, G. R . Franzen, G. A. Knesel, and D. J. Northington, Jr., J. Org. Chem., 32,3285 (1967).For 40: J. G. and H. Gotthardt Traynham and H. H. Hsieh, Tetrahedron Leff., 3905 (1969), and G. S. Hammond, Chem. Ber., 109,3767(1976). (31)The similarly situated chromophore of trans-l,ldimethyl-8,9dihydroindene is reported to absorb at 259.5nm (3300):S.W. Staley and T. J. Henry, J. Am. Chem. Soc., 91,1239 (1969). (32)E. E. van Tamelen and T. L. Burkoth, J. Am. Chem. SOC., 89,151 (1967); S.Masamune and R. T. Seidner, Chem. Commun., 542 (1969). (33)P. Radlick and W. Fenical, Tetrahedron Lett., 4901 (1967). (34)For 44: T. J. Katz and P. J. Garratt, J. Am. Chem. Soc., 86,5194 (1964). For 46: J. Schwartz, Chem. Commun., 833 (1969),and ref 31 for alkylated derivatives. For 47: P. Radlick and G. Alford, J. Am. Chem. Soc., 91,6529 (1969). (35)R. E. Pincock and J. Haywood-Farmer, Tetrahedron Lett., 4759 (1967). (38)E. Vogei, Angew Chem., 73,548 (1961):S.Masamune, P. M. Baker, and K. Hojo, Chem. Commun., 1203 (1969).and references cited therein. (37)I. M. Takakis and W. C. Agosta, Tetrahedron Lett., 531 (1978). (38)A . T. Blomquist, R. E. Burge, Jr.. and A. C. Sucsy, J. Am. Chem. SOC.,74, 3636 (1952).
4‘-Substituted Nucleo’sides. 4. Synthesis of Some 4’-Hydroxymethyl Nucleosides’ Raymond D. Youssefyeh,2 Julien P. H. Verheyden,* and John G. Moffatt Contribution
No. 140 from the I n s t i t u t e of M o l e c u l a r Biology, Syntex Research, Palo Alto, California 94304 Received August 3, 1978
T w o complementary routes have been developed f o r the synthesis o f 4-(acetoxymethyl)-l,2,3,5-tetra-O-acetylD-erythro-pentofuranose (9).T h e f i r s t o f these involves a m i x e d aldol condensation between 1,2-0-isopropylidenea - ~ - x y l opentodialdofuranose a n d formaldehyde w h i c h gives, as i t s major product, 4-(hydroxymethyl)-l,2-O-isopropylidene-P-~-threo-pentofuranose (5a). Inversion o f configuration a t Cs is achieved via a n oxidation-reduction sequence, a n d subsequent acetolysis furnishes 9. A more efficient route t o 9 involves a m i x e d aldol condensation (12a) and formaldehyde followed by debetween 3-0-benzyl-1,2-0-isopropylidene-a-D-r~bo-pentodialdofuranose benzylation, acetylation, and acetolysis. T h e condensation o f 9 w i t h a number o f purine a n d pyrimidine bases and t h e i r analogues l e d t o t h e preparation o f a variety of 4’-hydroxymethyl nucleosides t h a t have been screened for p o tential biological activities.
Recent work from this Laboratory has led to the development of methods for the synthesis of ribonucleosides substituted at the 4‘ position by fluoro,123m e t h ~ x y l ,and ~ , ~azido5 groups. Most of t,hese aspects have been reviewed.6 Nucleosides such as those above bearing electronegative substituents at C4, are frequently rather labile, particularly when the hydroxyl functions are all unsubstituted.l~3-6Hence, it was of interest to undertake the synthesis of nucleosides bearing stable carbon-carbon linked substituents at the C4, position. Such syntheses could, in principle, be carried out via addition reactions to the vinyl ether function of 4’,5’-unsaturated nucleosides, this method being the one used successfully for the 0022-3263/79/1944-1301$01.00/0
preparation of the Q’-flUOrO,-methoxy, and -azido compounds. Preliminary attempts to introduce, e.g., a 4’-cyano function by this route were not, however, overly promising. An alternate approach for the introduction of 4’ substituents is based upon the reactions of nucleoside 5’-aldehydes, a subject that has been of interest to us for some years7 This approach has been carried on in parallel with that reported in the present paper and is described separately.sc This latter work did, indeed, lead to the preparation of several 4‘-hydroxymethyl nucleosides via crossed aldol condensations between suitably protected nucleoside 5’-aldehydes and formaldehyde.8bsc As an alternative to this introduction of a C4,
0 1979 American Chemical Society
1302 J. Org. Chem., Vol. 44, No. 8, I979
Youssefyeh, Verheyden, and Moffatt
carbon linked substituent a t the nucleoside level, it was of interest to develop a synthesis of an appropriate 4’4hy0 droxymethyll-D-ribofuranosederivative9 that could subsequently be condensed with a variety of heterocyclic bases. In this paper we describe the synthesis of such an intermediate and its use in the preparation of a number of 4’-hydroxymethyl nucleosides. Preliminary accounts of this work have already appeared.8aJo The essentials of the key step in the present work find their 4a, R>= OH; R? = H origin in an earlier study by Schaffer,’l who showed that reaction of 1,2-0-isoptopylidene-cr-D-xylo-pentodialdofuranose b. R1= H; R? = OH (4a)12with formaldehyde and aqueous sodium hydroxide led to the isolation of 4- (hydroxymethyl)-1,2-0-isopropylideneP-L-threo-pentofuranose (5a).This product arose via an initial mixed aldol condensation followed by Cannizzaro reduction of the resulting hydroxy aldehyde. Following our own work reported in this paper, Leland and Kotickl3 have also briefly 6 described this reaction and have converted 5a into the corresponding pentaacetate, which has been condensed, via the glycosyl chloride, with NG-benzoyladenine to form 9-[4-
Scheme I
Sa,
R 1 = O H ; R’ = H
b. R ’ = H: R? = OH
7
&,R=H b. R AC 1
9
2
OXO 10
in 60% overall yield from 1,2-0-isopropylidene-a-~-glucofuranose. The melting point of 5a that we observed was several 3a, R ’ = O H , RL = H degrees higher than has been previously reported.llaJ3 b. R1= H; R? = OH Chromatography of the mother liquors from this reaction also (hydroxymethyl)- a - ~ - t h r e opentofuranosyl] adenine (1). led to the isolation, in 12%yield, of a crystalline, isomeric triol In contrast with the work of Leland and Kotick,13 our inwhich we have identified as 4-(hydroxymethyl)-l,2-O-isoterest lay in the preparation of 4-(hydroxymethyl)-P-~- propylidene-a-D-erythro-pentofuranose (5b). This product erythro-pentofuranosyl nucleosides (2), which much more has not previously been reported as arising from the reaction closely resemble the ribonucleosides in that they possess the of 4a with formaldehyde, but has been described as the minor normal cis (D-erythro) orientation of the 2’,3’-diol. Hence, our product from a comparable reaction with 1,2-O-isopropyliprojected key intermediate was considered to be 44acetoxydene-a-D-ribo-pentodialdofuranose (4b).’3 methyl)-1,2,3,5-tetr~-0-acetyl-D-erythro-pentofuranose (9), In order to further investigate steric control of this reaction, the synthesis of which has been approached in two different we have also prepared 4b by periodate oxidation of 1,Z-Oways. In the first of these we undertook an inversion of the isopropylidene-a-D-allof~ranose.~~ The resulting crude alconfiguration a t C3 in the known P-L-threo compound 5a.lla dehyde 4b in this case showed only a single 2,4-dinitrophenylhydrazine positive spot upon TLC examination and The synthetic route followed is outlined in Scheme I. 1,2-0-1sopropylidene-a-D-xylopentodialdofuranose (4a) was converted into the crystalline 1,3-diphenylimidazolidine was prepared by periodate oxidation of 1,2-0-isopropyliderivative, 3b, in an overall yield of 77%. A similar oxidation dene-a-D-glucofuranose as described by Schaffer and Isbe11.12 of 1,2-0-isopropylidene-a-D-allofuranose has been described As directly obtained, the crude aldehyde was found by TLC using lead tetraacetate, but the resulting 4b was not characto give three spots, all giving a positive test with acidic 2,4terized other than by reduction to 1,2-0-isopropylidene-adinitrophenylhydrazine spray.14 It is known that 4a can be D-ribofuranose.18The reaction of crude 4b with formaldehyde isolated as a crystalline dimer in roughly 70% yield. We have was conducted essentially as with 4a and gave the identical found that 4a can also be readily isolated as its crystalline products 5a and 5b in overall yields of 18 and 23%, respec1,3-diphenylimidazolidinederivative (3a) in an overall yield tively. The combined yields of purified 5a and 5b were not as of 74% from 1,2-O-isopropylidene-a-D-glucofuranose by high as those starting with 4a, and some difficulty was expetreatment of the crude oxidation product with N,N’- dirienced in separation of the isomers from some impurities. phenylethylenediamine15 in methanol containing acetic acid. Hence, samples of both 5a and 5b were converted into their We have previously found the 1,3-diphenylimidazolidine tri-0-toluoyl derivatives, which were obtained only as syrups derivative to provide an expeditious method for the isolation that were not more readily separated by chromatography. and subsequent regeneration of a l d e h y d e ~ . ~ c -For ~ J ~ the Clearly, the reaction starting with 4b offers no advantage over present purposes it is not necessary to purify 4a via 3a, and that with the more readily available 4a since almost equal direct treatment of crude 4a with formaldehyde and sodium amounts of the erythro and threo isomers 5b and 5a were hydroxide led to the isolation of crystalline 4-(hydroxyobtained. The mechanism of this isomerization has been admethyl)-1,2-C~-isopropylidene-~-~-threo-pentofuranose (5a) dressed in our concurrent work on the reactions of nucleoside
J . Org. Chem., Vol. 44, No. 8, 1979 1303
4'-Hydroxymet hyl Nucleosides
Table I. 100 MHz 'H NMR Chemical Shifts (ppm) cnmDd
sol- -H-1
ventn
sugar protonsb
H-2
H-3
H-4
3a
D 5.81 id) 4.45 id)
3b
D 5.59 idi 4.37 (ddl :1.66 idd) 4.25 (d)
T,a
D 5.86 (di 4.50 (dd) 4.11 id) D 5.68 id) 4.53 (ddl 4.22 (di D 6 . ! 2 id) 4.93 i d ) 5.65 i s )
5b t ritnluoyl 5a tritoluoyl 5b fi
:3.90 (d)
4.05 (dd)
D 5.96 ( d l i 0 3 iddi ,555 (d) D 5.80 ( d ) .i.:13 id,
base protons
H-5a
H-5b
H-5'a
H-5'b
CS H
2HH)
other
5.67 (d) 5.52 ibr s) 3.39 (d) 3.54 id) :3.47 is! 3.30 id) 3.53 (d) :3.% (d) 3.12 idi 4 5 1 id) 4.68 idi 4.66 ( s i
4.59
(SI
4.82 (si
4.03 (dl
3.13 (di 3.68 idi 3.29 Id) :1.45 td)
Xa
D
D 6 . X id) 4.7; id) 5.*'9id, -1.54 (ddi 4 3 6 (d)
2.99 id) 3.42 id) :XI4 ( d ) 4.16 ( d ) :1.00 id) :X56 idi :1.09 id) : x i ( d )
9
D 6.1): lbr ir.29 ( d ) 5.46 id)
4.0R-4.:3? ( m i
7
CS H
i.$1,
6.29 id. i i )
IO
D ,i.!i9
14a
D .5A3 i d , 4.79tdd) 4.04 idd) 4.30 idd)
12b
D
I3
D 5.M Id) 4.i0 tddl 4.14 it11
is1
5,68 ldl
,S.?il id,
4 7 tdi
4.78(dl
~ . 7 3iddJ 3.71 idd) 4.38 ihr di
4.81 idd! 5.14 ((1)
11
D
15
1 (i 17 1 Ra
D 6.35 id) L 1 6 idd) P 6.64 [d) 5.72 idd) P 6.77 td) 5.61 (dd) D 6.29 id1 6.02 Idd)
18b
D
5.87 (ddi 5.17 id) 5.18 (dl 5.78 i d )
19
5.94 (dl 4.70 (ddi 4.26 id) D 6.(?4 id! 1.8;tddi 4.27 i d )
20a
C
6.10 tdd) 5.58idd, 5.46 id)
20b
D 5.82 ldd) 4.15 ldd) 4.04 ld)
21a
C
21b
P
22
7.17 idi
5.16 Id)
5.62 id)
7.86 (di 4.93 Idd) 5.06 cd) D 5.81 idi 4.59 ( d d ) 4.21 id)
4.07 id1 4.24 i d )
4.11 1s)
9 3 4 id) 5.39 (hi- sJ : 3 x ( d ) :x58 id) :1.49 id) :3.80 id) 4.00 i d ) 4,16 id) 4,2(l ltl)
.LEI i d )
4.23 id) 4,27 Idi 4.42 ( d i 4.44 id, 4.37 ld) 4.52 id1 4.41 i s 1 4.2-4.6 im. 1) .L21 tdl 1.41 i d ) 4.26 idi 4.41
8.56is, 2 ) 8.63.8.94 & ? I , 8.48
.1..5-:1.8 Im. 1) :3.,5-:3.8 ( m . 4)
4.21 id1 4,47 i d i 4,25 !dl
3,:3-:1.7 ( m , 4 ) 4.1-4.6 im. 41 4.19 i s i c 3.50 i s )
