Carbon-13 nuclear magnetic resonance studies of allylic

less) by Cr(acac)s because these carbons are further from the OH site of Cr- ... (1 1) A. Westlake, W. E. Westlake, and F. A. Gunther, J. Agric. F...
0 downloads 0 Views 1MB Size
1282 J . Org. Chem., Vol. 44, No. 8, 1979

Tsuda, Parish, and Schroepfer

was more intense in partially relaxed spectra due to closer proximity to lH nucbei, was less shifted (1.2% less) by Cr(acac)s because these carbons are further from the OH site of Cr(acac)s complexatioin, and was less shifted by the lanthanide shift reagents Eu(fod)s (2.6% less) and Yb(dpm)3 (52% less) than are C-1 and -6. Although the Eu and Yb compounds can complex a t three sites in kelevan, a t carbonyls C-2' and C-6' and at the hydroxyl,-substituted carbon C-2, the preferred conformation, due to intramolecular hydrogen bonding between the OH and C:=O (C-20, places the site of shift reagent binding much further from C-3a and -5b than from C-1 and -6. Shift reagent bound at C-6' has little effect on the relative chemical shifts of C-3a, -5b, -1,and -6. Preferential binding of the shift reagents at the C-2' carbonyl and the hydroxyl group is indicated by the greater effects of the shift reagents on the chemical shifts of C-2 and C-2' than on the chemical shift of C-6'. The 13C spectrum of monohydrokelevan 7 was partially assigned using the same techniques and considerations as for kelevan (3). 'The conversion by CrOa oxidation of 7 to monohydrokepone diol 8 showed that monohydrokelevan had either of the two structures 16 or 17, which differ in the syn (16) or the anti (17) orientation of the ring hydrogen to the hydroxyl group. No doubling of the peaks in the presence of either Eu(fod)?or Yb(dpm)3indicated that only one of these isomers was present in the major product. The Yb shift reagent affected the chjemical shift of the CH carbon somewhat less (8%)than it affected the chemical shifts of the carbons diagonally opposite it (C-1,6 in 17 or C-3a,5b in 16). To distinguish between these two possible structures, we estimated the effects of replacing a chlorine by a hydrogen at either the 5b or the 6 position on the 6c values of 3 by calculating the additive substituent effects15 [6c(8)- 6c(9),ppm] of this replacement in Kepone from the I3C chemical shifts of Kepone diol 9 and rnonohydrokepone diol 8. The predicted 6~ values for both syii and anti forms of 7 are given with the observed chemical shifts in Table 11. Five chemical shifts should differ for the two forms; of these, the average deviation

between predicted and observed shifts is 0.19 ppm for 17 but 1.40 ppm for 16. These data thus show clearly that the anti form 17 is the major product of the photoreaction.

Acknowledgments.We would like to thank G. W. Sovocool and R. L. Harless for mass spectra and C. B. Bryden and S. T. Maher for technical assistance. Registry No.-1, 143-50-0; 2, 2385-85-5; 3, 4234-79-1; 8, 5217196-7; 9,4715-22-4; 10,69069-44-9; 14, 1034-41-9; 17, 69069-45-0.

References and Notes (1)E. G.Alley, B. R. Layton. and J. P. Minyard, Jr., J. Agric. FoodChem., 22, 442 (1974). (2)E. G.Alley and B. R. Layton, in "MassSpectrometry and NMR Spectroscopy in Pesticide Chemistry", Plenum Press, New York, 1974,pp 81-90, (3)E. G. Alley, D. A. Dollar, B. R. Layton, and J. P. Minyard, Jr., J. Agric. Food Chem.. 21, 138 (1973). (4)K. Knoevenagel and R. Himmelreich, 2.Pflanzenkr. Pflanzenschutz, 80, 155 (1973). (5) S. Begum. A. Gab, H. Parlar, and F. Korte. Chemosphere, 2, 235 (1973). (6)H. Parlar, W. Klein, and F. Korte, Chemosphere, 1, 129 (1972). (7)D. A. Carlson. K. D. Konyha, W. B. Wheeler, G. P. Marshall, and R. G. Zaylskie, Science, 194, 939 (1976). (8)E. G. Alley, B. R Layton, and J. P. Minyard, Jr , J. Org. Chem.. 41, 462 (1976). (9)N. K. Wilson and R. D. Zehr, presented at the 29th Southeastern Regional Meeting of the American Chemical Society, Tampa, Fla., Nov 9-11,1977, Abstract 209. (10)R. D. Zehr and N. K. Wilson, J. Chromatogr., submitted for publication. (11) A. Westlake, W. E. Westlake, and F. A. Gunther, J. Agric. Food Chem., 18, 159 (1970). (12)N. K. Wilson and G. W. Sovocool, Org. Magn. Reson., 9,536 (1977). (13)R. L. Harless, D. E. Harris, G. W. Sovocool, R. D. Zehr. N. K. Wilson, and E. 0. Oswaid, Biomed. Mass Spectrom., 5,232 (1978). (14)J. B. Stothers, "Carbon-13 NMR Spectroscopy", Academic Press, New York, 1972. (15)N. K. Wilson and R. D. Zehr, J. Org. Chem., 43, 1768 (1978). (16)Bayer names for some compounds in this paper are as follows: 1,

1.2,3.4.6,7,8,9,10,1O-decachloropentacyclo [ 5.3.0.02~8.03,9.04~8]decan5-one; 9, 5,5-dihydroxy-1.2,3.4.6,7,8,9,10,10decachloropentacyc l 0 [ 5 . 3 . 0 . 0 ~ ~ ~ . O ~ ~ ~ . O ~ ~ ~ 4, ] d e c 1,3,4,6,7,8,9,10,10-nonachloroane; pentacyclo[5.3.0.028.03.9.04.8]decan-5-one; 8, 5,5dihydroxy1,3,4,6,7,8,9,10,1 O-nonachloropentacyclo[5.3.O.O2~8.O3~g.O4~~]decane:

5,

1,3,4,6,7,9,10,l0-octachloropentacyclo[5.3.O~O2~6.O3~g~O4~8]decan-

5-0ne; 10, 5,5dihydroxy-l,3,4,6.7,9,10,1 O-octachloropentacyclo[ 5.3.0.02~6.03~9.04~8]decane.

Carbon-13 Nuclear Magnetic Resonance Studies of Allylic Hydroxysterols. Assignment of Structure to 5a-Cholest-8( 14)-ene-3&7a,15a-triol,an Inhibitor of Sterol Synthesis' Mitsuhiro Tsuda, Edward J. Parish, and George J. Schroepfer, Jr.* Departments of Biochemistry and Chemistry, Rice University, Houston, Texas 77001 Received September 5,1978 5a-Cholest-8(14)-ene-3P,7E,15F-triol, a potent inhibitor of sterol biosynthesis in animal cells in culture, has been with refluxing shown to be formed in 53% yield upon treatment of 3~-benzoyloxy-14a,l5a-epoxy-5a-cholest-7-ene aqueous ethanolic KOH [G. J. Schroepfer, Jr., E. J. Parish, H. W. Chen, and A. A. Kandutsch, J . Biol. Chem , 252, 8975 (1977)l. Detailed analyses of the I3C nuclear magnetic resonance spectra of this compound and of other steroidal allylic alcohols and their derivatives have permitted the establishment of configurations of the 7 and 15 hydroxyl functions as a. The resonances of the individual carbon atoms have been determined for six allylic hydroxysterols as well as a number of carbamate and acetate derivatives. Treatment of 5a-cholest-8(14)-ene-3P,7a,15a-triol with in 87% yield. Also described herein are syntheses of 3P-benzoyloxyacid gave 1~5-oxo-5a-cholest-8(14)-en-3~-ol 8a,l4a-epoxy-5a-cholestan-7a-o1, 3~-benzoyloxy-8a,9a-epoxy-5a-cholestan-7a-01, 7-oxo-5a-cholest-8-en-3~-ol, 5a-cholest-8(14)-ene3@,15a-diacetate,5a-cholest-8(14)-ene3/3,15P-diacetate, 5a-cholest-8(14)-ene 3(3,7a,15a-triacetate, and 7a,l5a-diacetoxy-5a-cholest-8(14)-en-3-one.

Over 20 years ago Barton et al.2,3reported that treatment with perphthalic acid in of 3/3-acetoxyergosta-'7,14,22-triene ether gave, upon washing of the ether solution with dilute 0022-3263/79/1944-1282$01.00/0

aqueous sodium hydroxide, the sodium salt of the half phthalate ester of 3/3-acetoxyergosta-8( 14),22-dien-7[,15[diol. The product was not characterized as such but, upon

0 1979 American Chemical Society

13C NMR Studies of Allylic Hydroxysterols

J.Org. Chem., Vol. 44, No. 8, 1979

1283

treatment with dilute acid, gave the corresponding half ester R in the acid form which was characterized by elemental analysis, optical rotation, and ultraviolet spectral analysis. Heating of the sodium salt of the half ester with hydrochloric acid in 9co-0 'OH methanol gave 3P-acetoxyergosta-8(14),22-dien-15-one. 6 Subsequently, Woodward et aL4l5reported that treatment of I o , R I = H ; R2=R3=OH a mixture enriched with 3P-acetoxy-4,4-dimethyl-5a-choI b , RI=H; R2=R3=OAc lesta-7,14-diene with inonoperphthalic acid gave a product IC, RI=R2:O; R 3 = O H which! on saponification, gave 4,4-dimethyl-5a-cholestId, R , = R , = O ; R,=OAc 8(14)-ene-3@,7[,15(-triol. The latter compound gave 4,4+co-o "OH dimethyl-5a-cholest-8(14)-en-3P-ol-15-oneon treatment with hydrochloric acid in methanol. More recently, Muccino and 7 DjerassiG reported that treatment of 5cu-cholesta-7,14-diene with m-chloroperbenzoic acid in ether followed by saponification of the crude product with ethanolic KOH gave a product which was foirmulated to be 5a-cholest-8(l4)-eneRl 7[,15[-diol. The latter compound, which was not characterized HO 8 upon as such, gave the known 5n-cholest-8(14)-en-15-one 2a, R,=OH treatment with hydrochloric acid in ethanol. Akhtar et al.7,8 Pb,R,=OAc have also recently reported the preparation of 4,4-dimethyl5n-cholest-8(14)-ene-3~,7[,15[-triol by treatment of the corresponding 7,14-diene with perphthalic acid. R We have recently found that a number of 15-oxygenated HO sterols are very potent inhibitors of sterol synthesis in animal cells in c ~ l t u r e . ~In- ' the ~ course of this research we pursued 9 the chemical synthesis; of a 3,7,15-trihydroxysterol for biological testing. A. key p:recursor for the preparation of the desired compound was 3~Lbenzoyloxy-14a,15a-epoxy-5a-cholest-7-ene which. was obtained in 96% yield by treatment of 0 with m-chloropure 3~-benzoyloxy-5tr-cholesta-7,14-diene perbenzoic acid.I4 Una.mbiguous establishment of structure IO was based upon the results of X-ray crystallographic analysis of 3~-p-bromobenzoyloxy-l4~,l5a-epoxy-5a-chole~t-7-ene.~~ Treatment of 3~-benzo:yloxy-14a,15a-epoxy-5cu-cholest-7-ene with refluxing ethanolic KOH gave, in 53% yield, Sa-cholest-8(14)-ene-3~3,7[,15[-triolwhich was characterized by infrared, NMR, and milss spectroscopy and by conversion to 5a-cholest-8(14)-ene-3,7,15-trione.The As"4)-3P,7[,15[-triol l l a , R,=OH 4 0 , R , = OH was found to be a potent inhibitor of sterol synthesis.12Apart IIb, RI=OAc 4 b , R, =OAc from the high activity of this compound in the inhibition of sterol biosynthesis, the triol is of considerable interest since we have now found that 5a-cholest-8(14)-ene-3/3,7[,15[-triol, upon treatment with hydrochloric acid in ethanol, gives 5 a cholest-8(14)-en-3P-ol-15-onein high (87%) yield. The latter compound and several of its derivatives have been found to have significant hypocholesterolemic activity in rats and R'CgH17 m i ~ e . ~ ~ In . - lview g of the potency of the triol in the inhibition of sterol biosynthesis, its utility as an intermediate in the 50, R , = O H chemical synthesis of 5a-cholest-8(14)-en-3@-01-15-one, and Sb, R , = OTAl the unresolved status of the absolute stereochemistry of this and related ;ls(4~)-3/3,7,15-triols, we have pursued determination of the establishment of the structure of the concerned NMR spectra of androstane and cholestane. The observed triol. For the solution of this problem we have primarily used substituent effects were greatly influenced by the position and 13C NMR spectrloscopy. Presented herein are 13C NMR data configuration of the hydroxyl function. The 13C NMR on hs(14)-steroldiols and triols and analyses with respect to chemical shifts have also been reported for 3-cyclohexenol, allylic alcohol substituent effects and their acetylation and a cyclic allylic alcohoL21The assignments reported herein of trichloroacetyl isocyanate (TAI) shifts. The configurations resonance peaks of carbon atoms in close proximity to the of the C-7 and C-15 hydroxyl functions of 5a-cholest-8(14)- hydroxyl substituent were based, to a large extent, on the data ene-3P,7[,15[-triol have been determined by evaluation of the for hydroxyl substituent effects reported by Eggert e t a1.20 hydroxyl substituent effects obtained during the course of this and, in addition, on shifts induced upon acetylation and upon investigation. trichloroacetyl isocyanate formation.22Additional evidence in support of the assignments was obtained not only by sinResults a n d Discussion gle-frequency off-resonance decoupling (SFORD)spectra but The determination of the structure of 5a-cholest-8(14)also from protic solvent induced shifts and spectral comparene-3P,7[,15[-triol by 1.3CNMR required the syntheses (reisons with a selectively deuterated compound. ported herein) of As(14)-.allylic hydroxysterols containing each It is well established that acetylation of hydroxyl functions possible (aand 6) hydroxyl substituent at carbon atoms 7 and in cyclic alcohols causes a downfield shift of the a-carbon 15. Eggert et aLZ0have reported the results of studies of the resonances (-3 ppm), an upfield shift of the &carbon resoeffects of hydroxyl functions a t C-7 and a t C-15 on the 13C nances (-4 ppm), and a smaller upfield shift of the y-carbon

&

&

&

1284 J . Org. (?hem., Vol. 44, No. 8,1979

Tsuda, Parish, and Schroepfer

Table I. I3C Chemical Shifts of AR(14)-Sterols (1-1 (1-2 c:-3 c-4 c:-5 C-6 c-7 C-8 C-9 c-10 r-11 C-12 C-13 (1-14 C-15 C-16 (1-17 C-18 c:-19

C20 (1-21 c.22 C:-23 C-24

c-25 C.26 ('-27

86.5

31.5 il.O 38.2 44.2

28.9 29.6 1:26.1 49.2 :3fi,7 19.9 37.2 42.6 142.4 .)r rI ..