13 c and 1 h nuclear magnetic resonance spectroscopy of c-19 and 6β-methyl substituted steroids:...

13C and 'N nuclear magnetic resonance spectroscopy of C-19 and 6P-methyl substituted steroids: long-range

shift effects in conformational analysis

KATHERINE NASFAY SCOTT' A N D THOMAS HAROLD MARECI I/i.tc~rcir~s Atlriii/iistrcitiori No.\pitol, cir~ti Drpcirtt?ii~tzt qfRciciiologI', Ut!i~,et..,ity c!f'Floriclri, G'iiiries~~illr, F L 32610, U.5 .A

Received April 13. 1978

KATHERINE N A S F A ~ SCOTT and THOMAS HAROLD MARECI. Can. J. Chem. 57, 27 (1979). I3C and ' H nmr spectra were obtained and assigned for nine C-19 substituted cholest-5-enes,

three 6b-substituted 19-norcholest-5(10)-enes, and several related steroids. ' T cheniical shift cffects have previously not been studied in either C-19 substituted steroids or in cholest-5(10)- enes. I n the present study, substituent effects o n the l3C chenlical shifts of the T , P, 7 , and 6 carbons were evaluated in detail. Although the substituent in C-19 substituted and 6(i-methyl substituted steroids is less rigidly oriented with respect to the rest of the molecule than in ring- substituted steroids, similar shift erects mere observed. In cholest-5-enes thc obserLed 13C and 'H shift effects and the temperature dependence of the 13C shifts indicate that the preferred orientation of the C-19 substituent is rillti to C-1. The relative stabilities of the rotaniers can be attributed to the orientation of the C-19 substituent ~ v i t h respect to the double bond. This interpretation is supported by the fact that the preferred orientation of the iodine in 6p-iodomethyl-19-norchoIest-5(10)-en-3~-ol bas the same spacial relationship with respect to the double bond, i.e., gnirclre to C-5 and C-7.

KATHERINE NASFAY SCOTT et THOMAS HAROLD MARECI. Can. J. Chem. 57. 27 (1979). On a enregistrk les spectres rriln de ' 3C et de '11 et fait les attributions pour neuf cholcstenes-5

substitues en C-19, trois nor-19 choiesti'nes-5(10) substitues en 6P et plusicurs steroides apparentes. Les effets de diplacement chiniique de I3C n'ont encore Fait l'objet d'une etude ni pour les stiroides substitues en C-19, ni pour les cliolestenes-j(l0). Dans le present t r a ~ a i l , on Cvalue en detail les ei'iets des substituants sur lcs deplacenients chirniques de I3C des carbones G!, (i, et 6. Bien que le substituant dans les steroides substitu(.s en C - i 9 o u sur le groupe methyle en positioil 6p, soit oriente nioins rigidenlent par rapport au reste tle la inolecule que dans les steroides substitues sur le cycle, o n a observe des effets de deplacement semblables. Dans Les cholestenes-5, les effets de deplacernent de 13C et de ' H et l'influence de la temperature sur les deplacernents de 'T iindiquent que le substituant en C-19 adopte une orientation preferentielle otiri par rapport i C-I . On peut attribuer les stabilitks relati\es des rotanieres B I'orientation d u substituant en C-19 par rapport a la double liaison. Ccttc intcrprdtation est renforcee du fait que l'orientation preferentielle de I'iode dans I'iodomethyl-60 nor-19 choiesten-5(10) 01-3b possede la memc relation spatiale par rapport i la double liaison, soit goricl~e par rapport a C-5 et C-7.

[Traduit par le journal]

Introduction

Since steroids are rigid structures, long-range effects of substituents on l3C chernical shifts have previously been correlated with the spacial relationship between the carbons and the ring substituent. Substituents y-galtc,/?e or -eclipsed to a given carbon produce shielding: ~vhen anti to the carbon the effect is generally deshielding (ref. 1 and references cited therein). The direction and the iiiagnitude of the shifts depends on the degree of substitution of the intervening carbons. This conclusion has also been reached from theoretical co~lsiderations (2). 111 ring- substituted hydroxy steroids, methyl-trans-decalols, and n~ethylnorbornanols, substituents in the sj-17-

axial configuration substantially deshield 6 carbons (3-5). Deshielding of carbons four bonds from the substituent is not peculiar to hydroxy substituents or cyclic systems, but has also been observed for methyl substi t~~ents and acyclic compounds (6, 7). However, the crowded sjx-axial 6 co~lfiguration has ca. 3 kca1,'mol interaction energy (8); therefore, more stable orientations of the substituent arc fakored and deshielding d effects are seldom observed in acyclic systems.

The y and 6 efects are of very obvious utility in stereochemical assignment of rigid systems. I n addi- tion, since these and 6 efTects are frequently quite substantial (2 ppm or larger): it secmed to us that they might be utilized for ronfor~iiational analysis of

'Author to whom correspondence may be addressed. less rigid syste~ns which have rotation of the sub- 0008-4042/79/0l0027- 1 1x01 .00/0

:#I979 National Research Council of Canada/Conieil national d e I-echerches du Canada

Can

. J. C

hem

. Dow

nloa

ded

from

ww

w.n

rcre

sear

chpr

ess.

com

by

UN

IVE

RSI

TY

OF

SOU

TH

CA

RO

LIN

A o

n 11

/15/

14Fo

r pe

rson

al u

se o

nly.

2 8 C A N . J . C H E M . VOL. 7 . 1979

stituent about one bond. Accordingly we examined sr~bstituelit effects in some C-19 and 6P-methyl substituted steroids.

The effect of introducing substituents at the C-19 methyl of steroids has previously not been investigated (9). Rotation of the methyl group about the bond bet~veen C-10 and C-19 allo~vs for various orientations of the C-19 substituent relative to tlie rest of the molecule. An ntnr investigation of C-13 substituted steroids could show whether substituent effects in this less rigid system folloiv the trends established for substit~ltion on the steroid ring and ivhetlier these effects could be used to establisli any orientational preference of the substituent. For this reason ive report I3Cmr and 'Hlnr data on the several C-19 substituted cholest-5-enes and related steroids slioivn in Fig. 1 . We examined several 19-hy-

CHzI 7 8

(i R = O H 0

1 1 (, R = OCH, h R = OCCH, (i R = 0CH2CH,

FIG. 1 . Compounds studied: In, cholest-5-en-3P-01: 2h. choIest-S-en-3p,l9-diol 3-acetate: 3ii, cholest-5-ene-3P,19- diol 19-p-toluenesulfonate; 40, 19-iodocholest-5-en-3p-01: 5, estr-5(10)-ene-3B,17/3-diol 17-propionate; 6, estr-S(10)-ene- 3B,171?-dione; 7, 6P-niethyl-19-norcholest-S(iO)-en-3~-ol: 80, 6~-iodoniethyl-l9-norcholest-S(10)-en-3~-ol.

droxy substituted cholet-5-enes; since the effect of hpdroxy siabstitution has been most widely studied in steroids. Because we wanted to study conformational effects, bulky substituents, such as iodo and 11-toluenesulfonql, were also examined. Our results sho~v that the preferred orientation for the C-19 substituent is allti to C-1. We attribute the relative stabilities of the rotarners in the cholest-5-enes to tlie orientation of the C-19 substituent with respect to the double bond. To test this interpretation, we also investigated the orielitational preference of a different double-bonded steroid, 6b-iodomethyl-19- norcholest-5(lO)-en-3 P-01.

'"I labelled 4a and 8a are two adrenal radio- imaging agents approved for l iu~uan use and are thus of considerable biological and radiopharmaceutical interest.

Results and Discussion

Spec.ti.~I A.v.cigtztl~crzts The 13C che~nical shifts obtained in this study are

sumnlarized i l l Table 1. Chemical shifts for l a , Ib. and I c have been reported previously (10-13). The chemical shifts in Table I for these compounds agree with tlie literature values to 0.6 ppm or better, which is in the range expected for solvent and concentration effects. Our assignments for Brr, Bb, and I c agree L\ ith those originally reported (10-1 3) except for the reversal of the assignments of C-12 and C-16; as noted for cholesterol (14) and sorne cholestanes (15) and later confirmed for several related steroids (16-19).

111 assigning the remaining compounds. it Ivas assumed that the D .ring and the C-17 side chain were least affected by substiturio~i. Indeed, these resonances changed little in the cholesterol series of compounds ant1 Mere assigned on this basis. The expected shifts of the remaining carbons for each of tlie compounds were predicted from the observed shifts of appropriate liiodel cornpounds and from knoiqn substituenl emests. Agreement between observed and predicted shifts was sufficient to permit preliminary assignment. Then the specific assign- ~ ~ i e n t s were confirmed by e~tablishing the type of carbon (methyl, methylene. methine, quaternary) by a combination of low power broadband decoupling (LPBBD) or single frequency OK-resonance decoupling (SFORD) of the protons. U~lequivocal assignment of all but a few resonances could be achieved by the above procedure. Arllbiguities in assignment re- mained only for those few resonances for which the spectrum contained close-lying resonances of carbons with the same number of attached hydrogens or where the SFORD spectra did not establish the num-

Can

. J. C

hem

. Dow

nloa

ded

from

ww

w.n

rcre

sear

chpr

ess.

com

by

UN

IVE

RSI

TY

OF

SOU

TH

CA

RO

LIN

A o

n 11

/15/

14Fo

r pe

rson

al u

se o

nly.

Can

. J. C

hem

. Dow

nloa

ded

from

ww

w.n

rcre

sear

chpr

ess.

com

by

UN

IVE

RSI

TY

OF

SOU

TH

CA

RO

LIN

A o

n 11

/15/

14Fo

r pe

rson

al u

se o

nly.

30 CAN .I. CHEM

her of attached I~ydrogens. Possible alternative assignments are indicated in Table 1.

The expected cliernical shifts of the individual compounds rvcre predicted as follows: The shifts of 2h and of 2c n,ere predicted from the observed shifts of Ih and Ic and the -OH substitue~it effect ob- scrbcd in 2.2-dimcthyl-I-propanol (1 lb, 20). The o b s e r ~ e d steroid shifts of 3 were very similar to those of 2b and 2c and \\ere assigned on this basis. We cietermined the shifts for 11-toluenesulfonyl chloride (C-1' 141.8; C-2'. C-6' 127.1 ; C-3', C-5' 130.3: C-4' 146.9; CH, 21.8) and this permitted the assignment of the tosyl resonances of 3. The shifts of 4 uere predicted fro111 the o b s e r ~ e d shifts of 1 and the iodine substituent effect obserbed for I-iodopentane ( l l c . 21).

T o facilitate our assignrxerit of 8 ~ , which is the first 13Cmr assignment of a j(l0)-ene steroid, we obtained and assigned spectra 5, 6. 7, and 8c. Coni- pound 7 was obtained by LiAIH, reduction of 8u and 8c was obtained by methylating 8cr. Since there arc major discrepancies among the previously reported 'HIIIS data for $0 (22-25), we report the assignment of the 13Cmr spectrum of this cornpound in somekvhat more detail to show that '"~nr unequivocally confirms the structure as 6P-iodornethyl- 19-norchoIest-5(10)-en-3P-ol. Each of 7 , 80, and 8c shows the same eight resonances as those observed for the C-17 side chain in all the other cholesterol compounds of this series and these resonances vere assigned to C-20 through C-27. The cheinical shifts of C-18 and the C- and D-ring carbons of 7 .8~1 , and 8c were predicted from the observed shifts of Irr and the effect of the removal of the C-19 ruethy]. This effect xvas obtained by cornparing the shifts of testosterone and 19-nortestosterone (1 0, 1 10, 19). The agreement bet~veen observed and predicted shifts was 0.1-1.0 ppm. The larger deviations, \vliich oc- curred for the C-ring carbons. undoubtedly reflect the additional structural differences in the A and B rings betiveen I n and compounds 7 , 80, and 8c.

The principal differences between the 1 3 C ~ i ~ r spectrum of $0 and its isomer, 40, occur in the A- and B-ring resonances, which means that the structural differences betiveen 4rr and 80 n ~ u s t be in the A and B rings. By LPBBD ute showed that both olefinic carbons of 8u had no directly attached hydrogens: therefore, the double bond had to be either between C-5 and C-10 o r C-9 and C-10. In the 'Hmr spectrum of 80, the CH,I protons forni the AB part of as1 ABX pattern, confirming that the CH,I is attached to 3 C H carbon. Therefore, the C--C(CH,I) possibility for quaternary olefinic carbons could be eliminated. In the I3Cmr spectrum of 5a-pregn-9(10)-ene-I6r- methyl-17a,2 I-diol-3,20-dione 21-acetate, we ob-

tained olefinic resonances a t 145.6 and 117.0 ppln for the C-9 to C-10 double bond. According to the CH,I substituent effects observed in 4. the attachment of CH,I anywhere on the A or B rings couid not have changed these olefinic shifts to those observed in Ma (125.5 and 133.9 ppm). thus the C-9 to C-10 double bond possibility for 8u is eliminated.

We then assigned 13Cmr spectra of 5 (an authentic sample (26) \\as kindly provided by Prof. S. G. Levine for this purpose) and of 6. The C- and D-ring carbon shifts of 6 nrere predicted from the obserhed shifts of 19-norandrost-4-ene-3.17-dione (10. 1 1, 19). The propionate side-chain carbons of 5 could easily be identified froill the corresponding shifts (27.9 and 9.2 ppm) which we obtained for isopropyl propionate. The shifts of C-18 and the C- and D-ring carbons of 5 uere predicted from the shifts of testosterone 17-acetate (10, 11u) and the cfrect of the removal of the 19-methyl. The A- and B-ring carbons of 5 and 6 could then be matched carbon by carbon and assigned if the effect of carboxylation a t C-3 was added to the observed shifts of 5 . This effect \+,as obtained by comparing the observed shifts of cliolestan- 3P-ol with those of cholestan-3-one (10, 1 lrr,b, 19).

Once we had assigned the spectra of 5 and 6 , the A- and B-ring resonances of 7 could be predicted by adding the 6P-methyl effect (4) to the observed shifts of 5. The shifts of 8a were then obtained by adding the iodine substituent effect to the observed shifts of 7. Carbons 2: 3, and 4 of $0 were further identified by con~parison u i th 8c, which gave the same resollances as 8rr, except C-3 was deshielded by 9.1 ppln and C-2 and C-4 were shielded by 3.6 and 3.1 ppm. respecti\ely. Thus. by the above detailed analysis of the l " ~ r spectra of the five conipounds 5 . 6, 7 , 8cr, and 8c, we could unequivocally co~ifirm that the structure for 8n is as indicated.

E f e c t o f 0xj.gcr1 01, Joditie S~rbstirictiotl r uric/ p Eff~~ct.c The 13C chemical shift effects of oxygen or iodine

substitution are summarized in Table 2. Hydroxyla- tion a t C-19 deshields the a carbon by 43.4 ppm and the p carbon by 4.9 ppm. These shift effects are similar to the 41.4 and 5.0 ppm deshieldings observed for the R and 0 carbons of 2.2-dirnethyl-1-propanol ( I lb , 20). Tosylation a t C-19 gives 50.6 and 3.6 ppm deshielding of the a and carbons, respectively. Thus thep-toluenesulfonyl substituent deshields the a carbon 7.2 ppm inore and the P carbon 1.3 ppln less than the OH. Although these effects are smaller, they lire analogous to the changes observed between the shifts of ethyl sulfate (69.6 and 14.5 ppm, ref. 1 % ) and ethanol (57.3 and 17.9 ppm, refs. l 1 b and 20).

The a carbon in each of thc iodine substituted

Can

. J. C

hem

. Dow

nloa

ded

from

ww

w.n

rcre

sear

chpr

ess.

com

by

UN

IVE

RSI

TY

OF

SOU

TH

CA

RO

LIN

A o

n 11

/15/

14Fo

r pe

rson

al u

se o

nly.

SCOTT A N D MARECI

Q M y y - 66 ,cia 0 0 -

1 1 I l l I l l

d-"i m c m m n 3 66 366 60-1

"" b, -7 YP!? G& ~ w o w d - a

I I 1 ' 1

Can

. J. C

hem

. Dow

nloa

ded

from

ww

w.n

rcre

sear

chpr

ess.

com

by

UN

IVE

RSI

TY

OF

SOU

TH

CA

RO

LIN

A o

n 11

/15/

14Fo

r pe

rson

al u

se o

nly.

3 1 C 4 N . 3 . CHEhl VOL. 57. 1979

steroids is shielded by about the same extent (8.1 to 8.7 ppm)? nhich is about 2 pprn larger than the 6.5 shielding observed in I-iodopentane. The reason for the larger shielding in the steroids is not apparent. The fi carbon in the cholest-5-enes is deshielded by 1.6 to 2.0 ppm and in the 5(10)-ene by 8.2 ppm. These values al-e srnaller than the 11.3 ppm deshielding of the P carbon in I-iodopentane (21: 1 I c ) . The lesser shielding of the carbons in the steroids can be attributed to the greater degree of substitution for C-6 in 8c1 and particularly C-10 in 4. Decreased P substituent effects ~vith increased branching were fi rst observed for norbornyl derivatives (28) and the11 confirmed for the decalols (3h) and for various steroids (3b. 5: 29).

y Effects 111 C-19 substituted steroids, rotation about the

C-10 to C-19 bond gives the staggered rotarners. 9 , IT, and I11 (Fig. 2). The rotamers are nonequivalent and may have unequal population. For the 6P-iodo- ~nethyl conlpounds rotation abvut the iodomethyl to C-6 bond yields the staggered rotamess: A, B, and C (Fig. 2). At roo111 temperature. none of the rotamess are 'frozen out' and the observed chemical shift is a rotationally averaged shift. However. from the observed 13C and IH chemical shifts we were able to establish the relative populations of the rotamers.

A recent reinterpretation of the y effect (30) and the data on hydroxy steroids in ref. 5 show that large (2.3 to 8.3 pprn) upfield -{-gauche effects are possible if the substituent removes a 1,3-diaxial proton- proton interaction. If no such interaction is removed by the substituent, the y-gazrclie effect will be much smaller (0.1 to 0.9 ppm). In the present compounds, 10-H gives a 1.3-diaxial type interaction with the C-19 protons and 7P-H with the 6P-methyl protons. These interactions are removed by the substituent only in rotamer I1 and rotalner A, respectively. The

FIG 2 Uppir lo\\ The thiee staggered rotamers (I, 11, and 111) of the C-19 CH2X gioup in C-19 subst~tuted cholest-5- enes The C-19 to C-10 bond is into the plane of the papel Loher rob+ The three staggeled rotarners ( A , B, and C) of the h/l-lodomethyl group of 6/l-1odonieth) 1- 19-norcholest- 5(10)-en-3p-01 The 6P-iodornethyl to C-6 bond 15 into the plane of the papel

substantial shielding 1 - 2.23 to -4..7 ppm) of C-1 in the 19-0 compounds shcv - that rotainer I1 is appreciably populated in these ci?,npounds, whereas the lack of shielding effect 01: 6 -1 ( - 0.1 to 1-0.2 ppm) indicates that rotamer B H is iiot well populated in the 19-1 compounds. (Tiie dir" ,rence in relative populations of the 19-1 and 19-0 compounds will be dis- cussed in subseq~ient sections.) The 3.0 ppm shielding of C-7 in 8 shows that rotamer A has appieciable population.

Since C-5 has no hydrogen, the large slzieiding (-3.9 to -6.4 pprn) of this carbon cannot be ac- counted for by the removal of the 1,3-diaxial intes- action but ]nust be due to linear electric field shift effects. (See subsequent sections.)

6 Effcjct~ In 10-methyl-trci~.s-decalols, ruethylnorbornanols,

and some hydroxy steroids, larger positive ( + 2 ppm or greater) 6 substituent effects were observed for carbons i i i the .s!.~~-axial confornlation relative to the OH (3, 4). In 31 monohydroxylated cholestanes and androstanes, about 80 exan~ples of interaction between the O H and the 8 carbon were examined (5). Of the five possible orientations, 6 , - 5, (3h, 5), only the 6 , sjn-axial configuration showed large (2.0 to 3.8 ppm) deshielding effects (5). Although n ~ o s t of the observed interactions were between O H and CH,. several interactions rvere between O H and ring nlethylene carbons (e.g., IP-OH and C-11).

In the present series of 19-0 compounds, C-6 is consistently strongly desliielded (5.3 to 5.6 ppm). C-6 is 6 , relative to the 19-0 only in rotarner I. (See Table 3.) Thus the strong deshielding of C-6 in the cornpounds indicates that rotamer 1 has appreciable population. In this configuration, C-8 too is 6 , to the 19-0 . Indeed, C-8 is the only other 6 carbon which shows a substantial deshielding (0.8 to 1.4 ppm).

In the present 19-0 compounds, the 6 , effect on C-6 is larger and on C-8 someuhat srnallei than the 1.4 to 3.8 ppm deshielding effects previously observed in ring-substituted steroids, methy1-tran.c- decalols, and methylnorbornanols (3-5). Some of the deshielding of C-6 may be due to linear electric field shift (LEFS) effects ( 3 1 , 32), since these effects can be substantial on .rp2 carbons (33, 34). The oxygen- carbon bond acts as a dipole, polarizing the charge distribution of the C=C in such a manner as to shield the carbon closer to the polar group and deshield the carbon farther away. Using eqs. [2]-[4] of ref. 33a, n e estimated the LEF'S on C-5 and C-6 in the 19-OH compounds. The appropriate distances and angle\ were obtained from molecular models, dipole moments from ref. 35, and LEF coeficients from ref. 33b. The LEFS thus estimated for C-5 and C-6 in the rotamers are : 1 - 4.8 and 1 .0; 11 - 1.7 and 1.4;

Can

. J. C

hem

. Dow

nloa

ded

from

ww

w.n

rcre

sear

chpr

ess.

com

by

UN

IVE

RSI

TY

OF

SOU

TH

CA

RO

LIN

A o

n 11

/15/

14Fo

r pe

rson

al u

se o

nly.

SCOTT A h D M 4 R E C I 3 3

IIT - 0.1 and 0.6 p p ~ n , respectively. Since the value of the EEF coeiiicient may involve error's up to a factor of 2 (33b) , clearly these LEFS are oiily esti- mates. Nonetheless, they account for a substaiitial part of the observed shift effect on C-5 but only for a fraction of the observed effect on C-6. The LEFS suggest that 1 is most populated.

There are additional reasons why the 6 , effect on C-6 is large. First. the .c./~' hybridized C-6 does not necessarily experience tlie same 6, effect as the .c.p3 hybridized carbons. Whether the 6, effect on sp2 carbons is generally lasger than on .ip\arbons cannot be stated at this time, since the present series of compounds and those in ref. 9 represent the only report of 6 , effects on s / 7 2 carbons. Second, in monohydroxylated cholestanes and androstanes, srnaller 6 , effects mere observed for the more rigid molecules and also for larger 013 to 6 , carbon distances (5). Indeed, the very small (-0.1 and +0.4 ppm) 6 , effect on C-7 in the 15P-hydrox!l coinpou~ids was attributed to the considerably larger (ca. 307;) than i~sual internuclear distance between the oxygen and 6 , carbon. In rotamer B of the 19-0 compounds, the C-8 to 0 intern~~clear distance has the usual (2.5 F 0.2 A ) value. u,hereas the C-6 to 0 interiiuc1ea.r distance is considerably larger (3.1 k 0.2 A). However. there is reason to believe (see the following two paragraphs) that the preferred orientation of the substituent is not precisely atlfi C-1 but slightly rotated toward the C-5 eclipsed position. This \vould increase the substituent to C-8 distance and account for the louer 6 , effect on this carbon.

Recently. a shorter T , (hence slower rotation rate) was observed for the C-19 niethyl in androst-Sene than in androstan- 17-one (36) and was attributed to a louer energy of the preferred conformer in the 5-ene. I n the 17-ketone. the C-19 inethyl group has .rj3ti-

axial interactions \kith the axial protons on carbons 2, 4: 6, 8, and 1 1 . I n the Sene, the interaction with the axial proton on C-G has been removed; this is believed to account for tlie lower energy of the preferred rotamer in this compound.

In the present work, the stability of one staggered rotamer over another is niost probably due to the presence of the C=C. Since C-6 has no H; the bulky substituent ( O H or I ) in 1 experiences fewer nonbonded proton interactions, which accounts for the lower energy. Furthermore, the nonbonded interaction betueen the substituent and 8P-H can be decreased if the substituent is slightly rotated toward the C-5 eclipsed position and since there is no 6b-H. this \vill not result in an increased interaction between the substituent and the 6p-H.

Thus we postulate that because of the double bond, the nonbonded interaction between the substituent and the 8P-H in I is less severe than the interaction

between the substituent and 1 I P-H in IT. And f i ~ r - thernlore that it is much less than the combined noilbonded interactions between the substituent and the axial protons 011 C-2 and C-4 in 111. If this is so, then the relative populations of the rotamers should be 11 > TI >> 111. Furthermore. the Inore bulky the siibstit~~ellt on C-19, the more rotamer I mould be stabilized, relative to rotamer II. This may \!#ell ex- plain ivhy rotainer 11 is populated for the i 9 - 0 compounds but not for the 19-1 co~npounds. (The cova- lent radius of iodine is alniost twice that of oxygeii.)

For the 6p-iodomethyl compound similar con- siderations would predict that the nonbonded interaction b e t ~ e e n the substituent and 8P-H in B is less severe than tlie interaction between the substituent and 4P-H in C. Hence B would be of lower energy than C. But because there are no 6 , interactions i r i rotainer A, this rotanler may be of lomest energy. We will give experimental evidence in subsequent paragraphs to shorn that the relative populations of rotarners I. 11, IHI and of A. B, C are indeed as predicted.

The strong deshielding of C-6 in the 19-1 compounds and of C-10 in 8a is consistent \vith the 6 , interaction bet\+,een these carbons and iodine, if the most populated rotamers are % and B, respectively. Since C-8 is also 6 , to the iodine in these rotamers, the much smaller substituent effect at C-8 inost likely arises from the slight rotation of the substituent tobvard the C-5 eclipsed position. as previously dis- cussed for the 19-0 compounds. The difference in iodine substitueilt effect between C-8 and C-6 is identical to the difference in oxygen substituent effect bet\\een these carbons.

The lack of deshielding effects at C-2 and C-4 for the 19-substitutecl compounds and at C-4 for 8 sl~ows the absence of 6 , interaction for these carbons, hence negligible population of rotamers IIB and C. (See Table 3.) Thus by the analysis of all the ;/ and 6 effects, we established that the rotamers wit11 appreciable populations are H for the 19-1 compounds, 1 and HI for the 19-0 compounds, and A and B for 8.

TABLE 3. Spacial relationshipa between the substituent and the 6 carbon in the staggered rotaniersb

C-19 6p-Methyl substitiitcd substituted

-

6 carbon I" 116 HIPh 6 carbon Ah Bb C1'

C-2 62 53 61 C-4. 65 63 61 C-4 63 62 51 C-8 63 6, 62 C-6 6, 6, 6, C-10 6, 61 63 C-8 61 63 6, C-I I 63 51 6.2

"The spacial reiationship betueen the substituent and the 6 carbon is designated by 6 , t o 6 5 (refs. 36 and 5).

bThe rotamers 1, 11, 111 and A, B, C are ahown in Fig. 2 .

Can

. J. C

hem

. Dow

nloa

ded

from

ww

w.n

rcre

sear

chpr

ess.

com

by

UN

IVE

RSI

TY

OF

SOU

TH

CA

RO

LIN

A o

n 11

/15/

14Fo

r pe

rson

al u

se o

nly.

34 CAN J . CHEM. VOL. 57. 1979

TABLE 4. Ten~perature dependence of the C-19 hydroxy and of the 6D-~odoniethyl lodine substituent effect o n l3C s h ~ f t ~ "

u D Y 6

Substitue~lt Carbon A ppm Carbon A ppm Carbon A ppm Carbon h ppni

OAli compounds 0.5 in CDC13. 5~1ubstitue1it effect a t -34-C (shifts of 26 - s h ~ f t s of 16) siibtiacted from subst~tiient effect at -20 C (shllts of 26 -- shifts of l b ) . 'Subsiituent effect at '34'C (shifts of 100 - shifts of 9) s~ibtracted fron? subst~tuent effect at 2 0 ' C (shifts of l O i r - shrfts oi 9) . Posttite \ a l~ ics

indicate increased deshieid~ng at -20'C. Spectra \\ere l u n at 3500 Hz s\\eep ,\idth, thercrorc, the shifts are kiio\rn ~ i t h :- 0.04 ppni precision.

\lie then e x a ~ n ~ ~ l e d the temperatule dependence o f the 19-hydroxy substltuent effect to establ~sh 15 hethel rotarner I or 11 11, of lower energy Slm~larly, the teinperature dependence of the 6P-iodomethyl iodlne substltuent effect was used to establish the relat~ve population of retainers A and B.

Tei~~per.uturc~ De/1endc,17ce of the C-19 HV~/I.O.YJ crnd of tlze 6~-I0~/01~1ethj~l Iodi~le Substit~teizt EfSc~cts

spectra were obtained for cornpounds Ih , 2b, 7, and 80 oker the temperature range of +34 to -20'C. The solubilities in chlorofor~n decreased to such an extent that it was not practical to obtain spectra below -201C. Thus the conformational equilibrium could be studied only in the fast ex- change limit. This generally provides only a first approximation to the equilibrium i l lc); therefore, the relative rotamer populations obtained in this lnanner should be considered indicative rather than conclusive

By cornparlng the chernlcal sli~fts at +34 and - 20 C. the temperature dependences shobtn In Table 4 \$ere obta~ned As expected, the ternperatule dependence of the substltuent effects o \er the 54 C temperature range mas small. houe\er, many of the observed efyects were 4 to 14 t ~ m e s as large as the exper~rnental error. and therefore can be cons~dered significant. The r and 0 effects shou substantla1 changes. whlch are not read114 analyzable, slnce these effects are a composite of ~ n d u c t ~ \ e . resonance. and sterlc effects (1 lb. 20).

The temperature dependence of the l~qdroxy y effects is on!y on the order of the experimental error. The hydroxy 6 effects are larger (2 to 14 times as large as the experimental error). If rotamer I is of lower energy, then decreasing the temperature should further increase the population of rotamer li relative to rotarner 11. inspection of Table 3 and the expected shifts of the 6 carbons (6 , strong deshielding, 62 moderate deshielding, 6, and 6, very sinall effect,

6, inoderate shielding (5) ) allo\\s us to predict deshielding of C-2, C-6. C-8 and shielding of C-4 and C-11 at lower temperature. This is indeed the experi- inentally observed trend.

For 8u the temperature dependence of the 6 effects is snialler and of the y effect slightly larger than for 26. Half of these efTects (at C-4 and C-10) indicate that rotanier B is of lower energy, whereas the other two effects (at C-7 and C-8) indicate that rotarner A is of lower energy. This can only be interpreted to mean that neither A nor B predominates. Thus from the careful consideration of the y and 6 substituent effects and their temperature dependence. the fol- lo~ving relative rotamer populations are indicated: I > 11 >> III a,nd A - B >> C . These relative populations are consistent with the hypothesis that it is the presence of the double bond in these compounds which accounts for the relative energies of the rotamers. Moreover. since the energy of B is comparable to that of A, the stabilizing effect of the double bond is substantial.

Recent data on three 19-hydroxy-4-en-3-one steroids (9) show strong shielding of C-5 and strong deshieldi~ig of C-4. More importantly. C-2 is now moderately shielded ( +O.6 to + 1.2 ppm). These shifts indicate that rotamer III has an appreciable population; that is, the 19-OH has the same orientation with respect to the C=C as it did in the 5-ene steroids.

X-ray crystallography showed the 19-OH conformation to be 111 for 19-hydroxyandrost-4-ene- 3,17-dione but to be out-of-ring (rather than I) for androst-5-ene-3P, 17P,19- trio1 17-p- bromobenzoate (37). The discrepancy betwee11 the X-ray and our data for the 5-ene may indicate co~lformational differences bet~veen the crystalline form and the solution, or interaction between the 19-OH and the bulky 17P substituent. or both. However, when there are two substituents on C-19, as in [19-R]- and [19-S]- 19-rnethylandrost-5-e!le- JB,17P,19-triol, the methyl was in the over-B-ring position. The fact that both

Can

. J. C

hem

. Dow

nloa

ded

from

ww

w.n

rcre

sear

chpr

ess.

com

by

UN

IVE

RSI

TY

OF

SOU

TH

CA

RO

LIN

A o

n 11

/15/

14Fo

r pe

rson

al u

se o

nly.

SCOTT .4ND MAKECI 3 5

TABLE 5. Proton chemical shifts"

Proton In l b l c 20 2c 3a 3h 3c

TABLF 6 S~ ibs t~ t i~en t effects on protoll chernlcal sh~fts"

Subst~tuent Compoundsb 3B-H 6P-H 18-CH, 19-CH, 21-CH, OCH,

19-OH 2b - l h 0.00 t 0 . 3 8 +0.06 +2.83 +2.60 0.00 2~ - 10 0 . 0 +0.40 i 0 . 0 5 +2.82 ~ 2 . 5 7 -0.01 0.00

19-OTs 3n - lii -0.01 f 0 . 2 4 -0.14 i 3 . 1 4 -2.95 -0 .03 36 - Ib -0 .04 t 0.22 -0.13 t 3 . 1 3 t 2 . 9 7 -0.02 3c - l c 0 . 0 +0.21 -0.07 +3.11 -2.97 -0.04 - 0.05

19-1 4u - In 0 . 0 t 0 . 2 9 r 0 . 0 9 +2.59 1 2 . 2 9 -0.01 40 - l h -0.01 f 0 . 2 8 + O . 10 i 2 . 5 7 i 2 . 2 9 t 0 . 0 1 4c - l c 0 . 0 t 0 . 2 8 +0.11 +2.58 -L2.30 0.00 +0.01

3D-OCOCH," 10 - In + 1.06 i 0.03 -0 .01 -0.01 3b - 30 + 1.03 t 0 . 0 1 0.00 0.00 0.00 0.00 46 - 40 t l . 1 t 0 . 0 2 0.00 -0.01 tO.O1 +0.01

3D-0CH3 l c - In -0.48 1 0 . 0 1 0.00 0.00 3c - 30 - 0 . 5 -0.02 t 0 . 0 7 -0.03 -0.01 -0.01 4c -- 40 -0 .5 0.00 +0.02 0.00 0.00 +0.01

"In ppm. Positive values ~ndicare downfield shift. Substitution had no effect ( + O . O l ppm) o n the chemical shifts of the 26-CH,, thc 27-CH, and the 38-OCOCH1 protons.

hChemlcal shifts of the t n o compounds that \%ere compared to obtain the substitucnt effects. CAcylation o f the 3P-OH. "Methylation o f the 3s-OH.

diastereorners have tile bulkier methyl group in the over-B-ring position suggests that this is the least energy conformation (37).

B.' H N~rcleur. Magt~etic Resonance Spectral Assignmer~ts I n the 'Hmr spectra only the 3-H, 6-H, and methyl

protons u-ere assigned. The results are summarized in Table 5. The spectral assignments relied heavily on previously published steroidal assignme~lts (38). We have previously published the 'Hmr data for compounds 4cr and 8cr elsewhere (25), but have in- cluded them in Table 5 for the sake of completeness. Our observed 'Hmr data for 8a agree with those reported by Kojima and co-workers (22, 24) but not with those of Basmadjian et al. (23). The peak posi- tions and intensities reported by Basmadjian et a/ . are not possible for steroids containing the C-20 cholestane side chain (25, 38a, 39, 40).

Substituent E;fSrcts otz Pi.ototz Clzemical Slzifts The more inlportant effects.of substitution on the

proton chemical shifts are summarized in Table 6. The largest substituent effect is the nonequivalence of the 19-CH, protons in all the C-19 substituted compounds. This nonequi~alence is the result of the rotationally averaged chemical shifts of the protons in the three rotamers shown in Fig. 2. Since the rotamers are different, the rotationally averaged chemical shifts of the protons will be different, even if the rotamers were of equal probability. Therefore, the observed nonequivaience of the 19-CH, protons does not yield any information about the relative population of the rotamers. However, the nonequivalence of the protons confirms that the rotamers are sufficiently different to cause appreciable (0.2 ppm) chenlical shift difference.

The 19-OH deshields the 18-CH, by 0.06 ppm. Since the 18-CH, is many bonds removed from the 19-OH substituent, the effect must be long range in nature, such as magnetic anisotropy and dipole- dipole interactions. Both of these effects are inversely proportional to the cube of the distance between the affected proton and the substituent (38b). Table 7

Can

. J. C

hem

. Dow

nloa

ded

from

ww

w.n

rcre

sear

chpr

ess.

com

by

UN

IVE

RSI

TY

OF

SOU

TH

CA

RO

LIN

A o

n 11

/15/

14Fo

r pe

rson

al u

se o

nly.

T A B L E 7. Effect of P-hydroxy substitution on C-18 1iiethq1 proton chemical shift5

--

18-Ha ippm) ----- ~ ----- li

Substirue~it Ref. 38c Ref. 41 ( A )

6P-OH 1 0 . 0 4 2 '0.033 4 . 8 7rj-OM + 0.033 + 0 . 0 2 2 4 . 2 8P-OH -k0. Id3 2 . 0

I I P-OH t 0 . 2 4 2 + 0 . -142 2 , 0 12B-OH 1 0 . 0 6 7 2 . 6

,'Relari\e to i~nsubatitutcd '~nd i i l~ rane Positc\c \ ' ~ i i ~ e ~ ~ncilcate donr:iield shiit.

'Appl-okimare 1 ~ 0 . 2 A ) dist'incc fioi11 the oylgen ni~cleiic to the ne;lset.t proton u f the C-I8 n le t l~> i , as rnc.lsi~r-ed from rnoleciiiai ~ n o d c l s . Til t \*me d~srance un tiis sotamers of 2 1s 1. 3.1 h ; l I . 3 . l A : J t i . 5 . l 4.

sho~vs some of the previously observed (38c, 41) eft'ecrs of B- and C-ring P-OH substitution on the 18-CH,. This table also contai~ls the approximate distance fi-om the oxjgen nucleus to the nearest proton in the 18-CH, and this distance in the three rotamers of Fig. 7. Only the 1ZP-OH substituel~t effect of 0.067 ppni is comparable to the 0.06 p p ~ n ei'fcct obser\,ed for the 19-OH substituent. The 8P- and I 1 p-OH eft'ects are much larger because of their close proximity to the 18-CH, and the 6P- and 7P- OH effects are smaller because of the larger distance. I t is noteworthy that only in rotamers I and Hid is the distance comparable to the distance in the 12P-OH compound. The distance in rotalner III is considerably larger and ~ t o u l d yield sinaller substituent effects than observed, Therefore, the 19-OH substituent effects on the 18-CH, proton chemical shifts further support the conclusion reached from chemical shift effects that rotamers H and 8% predominate at room temperature. Furtherlnore. the 0.38 to 0.40 ppm deshielding of the 6P-H by the 19-OH suggests close proxin~ity of the oxygen and the 6P-H and, hence. the predominance of rotamer 1 .

The 19-1 si~bstituent effect 011 the proton chemical shifts are comparable to those of the 19-OH substituent. This suggests that for the 19-1 substituted compo~~nds . also, rotamer 1 predominates. The 19-OTs substituent deshields the nlethyl protons at C-18 and C-21, and also the 3P-0-methyl of 3c,, undoubtedly because of the ring current effect of the aromatic ring. Ho\\ever, because the deshielding effect of the aromatic ring can extend to distances of several ring radii (42). and because the p-toluenesulfonyl substituent group is internally flexible: no statement can be made from the 'Hmr data about the relative population o f the rotamers in the 19-OTs compounds.

Conclusions

Since this is the first investigation of long-range shift effects in the conformational analysis of substituted methyls in steroids, i t remains to be seen

\vhether this method is generally applicable. How- ever. the present study sho~vs the feasibility of uti- lizing y and 8 eiTects for conformational analysis of solne steroid derivatikes even \vhen the rotamers are rzpidly intercon\erting. The method should be applicable to short side chain conformers in other rigid ring systems and, to a Inore limited extent, to highly branched acyclic compounds. A large shielding :J effect is expected for rotaniers in \vIiich the substituent remo\ es a 1.3-diaxial proton-proton interaction. The 8 , interaction will act as an indicator by deshielding cirbons in I-otamers ~ ~ i t h appreciable ~.~opulation. Moreoher, if the carbons 6 , to the substituent have proton5 to give nonbonded interaction nit11 the substituent, the rotamer is expected to be destabilized in proportion to the number of these interactions. Thus the observed ;/ and 6 effects may establish uhich ~.otamers are appreciably populated. In the case of more than one highly populated rota- mel-, the temperature dependence of these shift effects Iilay establish the relatike stabilities of the rotamers.

Experimental r\'r~clc~cri ~24crgi1ctic Kcso,iat~cc

Nuclear magnetic 1-esonance spectra were obtained \\it11 n Rruker model HX-90 spectrometer equipped \vith a fast Fourier transform system using a Nicolet modcl 1083 corn- puter. Proton spectra \vere obtained at 90.00 MHz and ' " C spectra at 22.63 MHz spectrometer operating frequei~cies. The samples were dissolved in a stock solution of deuteriochioro- form which contained 5% by volume hesaii~iorohenzc~ie (the flaorinc signal serbed as the field-freque~~cy stabilization signal) and tetramethylsilane in by volurnc concentration for ' H m r spectra aild 10% by ~ o l u m e concentration for I3Cnir spectra. Solution concentrations ranged from 0.05 lo 2.0 \I and were dictated hy the amount of sample a\ailable. The "Cnir shifts could he detcrniincd with 0.05 ppm precision and 'Hmr shifts \\ith 0.01 ppin precision, cvcept for 0.1 ppm precision for the very broad H-3 resonance.

,Vfcirc~iicr/c Cotnpounds I(/, If], 6, and 5r-pregn-9(lO)-e11e-l6x-methyl-

17.21-diol-3,20-dione 21-acetate \\ere commercially available. An authentic sample of 5 (26) wai kindly s ~ ~ p p l i e d by Dr. S. C;. Le\ine.

The remaining compounds Mere rynthesized b> kno\rn methoclr in the following sequence: l b -, 5%-bron~o-cholestm- 3P:G[J-diol 3-acetate (43-46) 5x-bromo-60.19-oxidocholes- tan-3P-ol 3-acetate (43, 44, 46, 47) 1 2 1 1 (43, 44, 48): 20 -+ 38 (33-45); 30 -t 46 (44, 49): 30 and 40 Jvere prepared from 30 and 40, respectively (43, 44); 40 -t 80 (25): 80 i 7 (22).

C o n ~ p o u n d Ic, was synthesized by the action of perchloric acid on 1 0 in trimethylorthoformate (43, 50). The same method was also used to prepare 8c from 8u. Compound 3c was similarly prepared from 4tr by the use of tr-icthyl ortho- formate (43). The reaction sequence for the remaining 311- n~ethoxy derivatives, 2c-4c, started \vith I c and \\as 3 5 for the .%acetates clescribed in the prekious paragraph (43).

All colnpoulids tested gave satisfactory elenier~tal analyses and had melting points as reported in the literature. All solid co~lipounds were found to be spectroscopically pure by I3Cnir and 'Hnir. Compounds 7, $a, and We \\ere oils. High pressure liquid chromatographic separation of 80, as preiiously dc-

Can

. J. C

hem

. Dow

nloa

ded

from

ww

w.n

rcre

sear

chpr

ess.

com

by

UN

IVE

RSI

TY

OF

SOU

TH

CA

RO

LIN

A o

n 11

/15/

14Fo

r pe

rson

al u

se o

nly.

SCOTT AYD hlARLCI 3 7

scribed, yieldcd a pale yellow glass, mhich was shown by 21. T. D. BROWN. P1i.D. Thesis. tini\ersityof Utah. Salt Lake ' T m r to be > 98 n~ol", pure (25). Compounds 7 and 8c \yere Cit! , UT. 1965. not purified b j high pressure liquid chromatogr-aphy but the 22. 11. KOJIM.A. R1. M \ E D A . H . O G ~ M A . K. N I I I A . and 'T. 110. i rnp~~ri t ies detected by '3C~i i r and ' H m r did not interfere ~ i t h J . C h e ~ n . Soc. Chem. Cornmun. 47 ( 1975). the spectroscopic analysis, except possibly for the assignment 23. 6. P. B A S \ I . \ D J I . ~ ~ . K . R. H L I Z ~ L . R. D. Icr. and W. H. of C-5 in Nc (see Table 1). BFIF.RM.ALTES. J . Labelled Cornpound\. 11. 427 (1975).

24. RI. h I - \ r n ~ . M . KOJIRIA. H . OGAM \ . K . NI I T A . and T . 1.10. Acknowledgmeaits Stei-oida. 26. 241 (1975).

Financial support \\{as provided by the Medical Research Service of the Veterans ad mini st ratio^^. We arc indcbted to Drs. H. L. Holland. P. R. P. Diakow, and 6. J. Taylor for letting us use sosne of their data before publication. This led to a reassign- ~ncll t of C-l and C-7 in our 19-0 compounds and aided the discussio~i of results for the 5-ene compounds. We thank Dr . S. 6. Levine for providing a saniple of 5 and Dr . J . A . Owoyale for the syntheses of the 3P-methosy and 3P-ethoxy derikatives. We are greatly indebted to Dr. M . W. Couch for the syntheses of a11 other co~npounds and to Dr. C. M . Willia~ns for initiating the research project.

1 . W . H. ALER. L . rZ1. B R O M N ~ . S . F L \ G . and J . B. S-101 H ~ R S . Org. hlngn. re so!^. 1 1 . 73 (1978).

2. K. S E I D X I A U and G. E. M A C I E L . J . Am. Chem. Soc. 99.659 (1977): D. J . G o R ~ ~ ~ ~ E I u . J . Am. Chem. Soc. 99. 2254 (1977).

3. ( 0 ) S. H. GROVER. J . k'. G U I H R I E . J . B. S ~ O T H F K ~ . and C. 'r. T , \u . J . Lfagn. Reaon. 10. 127 (1973): ( h ) S . H. GROLEK a ~ l d J . B . ST-OTHERS. Can. J . Chern. 52. 870 (1974): ( c ) J . R . S I 0-1 HERS. C . T. T A N . illid K . C TEO. 3. Rlagn. Reson. 20. 570 (1975): (I/) J . B. STOTHERS, C. T . T A N , and K . C . TEO. Can. J . Chem. 54. 1211 (1976).

4. J . U'. BI.UNI- . Au\t. J . Chem. 28. 1017(1975). 5. H . E C ~ G L K I . C . L. VAN .Ax1 \\'FRP. N . S. BHI\CCA, ilnd C .

DJLRASSI. J. Org. C h e ~ n . 11. 71 (1976). 6. L . P . Llxnt\1.4\ and T . Q. '\u.\hrs. Anal. Chem. 43. 1245

- 8. P. J . FLORI . Statistical mechanics of chain molecules. John

M'ileyand Son5. Neas Yol-k. NY. 1969. Chapt. 3. 9. H . LA HOLI-AND. P. R. P. D I - \ K O \ ~ . and G. J . TA\ . I O R . Can.

.I. Cheni. f h i \ i5sue. 10. H . J . REICH. Rl. J.\UTEL . \ I . M . T . MESSL. F. J . WEIGERT.

and J . D. K o n r ~ r s . J . Am. Chem. Soc. 98.7445 (1969). 1 1 . J . B. S I O T H ~ K S . Carbon-13 nnil- spectroscopy. Academic

Pre\ \ . Nem Yol-k, YY. 1972. (if) pp. 439-451; ( h ) pp. 140-141: ( c ) p. 133: ((1) p. 56: ( e ) pp. 420-432.

12. L . F. J o ~ u s o \ and W. C . J A ~ K O ~ S K I . Carbon-13 nmr spectra. John Wiley and Soni . New York. NU. 1972. (11) S p e c t r ~ ~ m No. 494: (bl Spectrum No. 13.

13. H . H . X ' 1 ~ u - r s c ~ and I . C. P. SXIITH. Can. J . Chem. 51. 1384 (1973).

14. W. B. S V I T H . D. L. DAVEYPOR.~. J . A. S W A ~ Z ' ~ . and G. A. PATE. 3. Magn. Reson. 12. 15 ( 1973).

15. H . E G G F R ~ andC. DJER-ASSI. J . 01-g. Chem.28.3788(1973). 16. D. LEIBFRIT~. and J . D. ROBERTS. J . Am. Chem. Soc. 95.

4996 (1973). 17. R . J . .ABRAHA\? and J . R. MOYAS~ERIOS. J . Chem. Sot.

Perkin Trans. 2.662 (1974). 18. R. J . CUSHLEV and J . D. FILIPENKO. Org. Magn. Reson. 8,

308 (1976). 19. J . W. BLUXT and J. B. S r o r ~ t ~ s . Or&. Mayn. R e w n . 8.439

(1977). 20. 5. D. ROBERTS, F. J . WEICERT,J. I . KROSCHWITL, and H . j

REICH. J . Am. Chem. Soc. 92. l'338(1970).

25. K . N. Scori- . hl. I+'. C o c c ~ . T . H. ~ I . \ R E c I . and C . 14. WILLI IS. Steroids. 28. 295 (1976).

26. S. 6. L L V I Y F . N. H. EcD>. and C . F. LLFFI L R . J . 01-y. Chem. 31, 3995 (1966).

27. H . S P I E S ~ . ( K E and W . G. S C H \ ~ I U ~ R . .I. Chem. Phqs. 35. 722 (1961).

28. J . B. G I ~ L I Z ~ F R . M. JXL-TELIJ . .~ . B. D L Y C F . R . A. SMITH. and J . D. K O B ~ R I S . J . Am. Chent. Soc. 92.7107 (1970).

29. N. S. B H A C C ~ . D. I). G I . \ V N I ~ I . U'. S. J A \ K O W S K I . and $1. E. WOLFF. J . ,Am. Chem. Soc. 95. 8421 ( 1973).

30. H . B E I ~ R B E C K andJ . K . S A L U D L R S . Can. J . Chem. 54,2985 (1976).

31. .A. D. B L C ~ I Y G H . A R I . C i ~ n . J . Chem. 38, 300 (1960). 32. J . F t r n r ~ . L. H . SL I C L I F F L . nnil S . hf. W \ I - K E R . k101.

Ph!s. 11. 117 (1966). 33. ( ( I ) J . G. B \ I ( H ~ L O R . J . H . P R L S T F G ~ R D . R. J . CUSHLEI .

and S . R. I J ~ ~ s r c \ . J . .Am. Chem. Soc. 95.6358 (1973): ( h ) J . G . B.\TCHFI.OR. J . A m . Chem. Soc. 97. 3410(1975).

34. H.-S. S c r r \ ~ . i n t ~ . 6. Gsc H\\ , inu I V L R . iinii L . B U ( ' H H ~ I I

J . hlagn. Reion. 26. 175 ( 1977). 35. V . I . R l ~ v k r \ . 0. .A. Os~l>ov. and Y. .A. Z~11>-\nov. Dipole

mornenis in organic chernistr). Plen~im Press. New York. VY. 1970. Chapt. l i l .

36. J . W . A P S I \ I ~ N . H . BEIFRBL(.K. and J . K . S ~ U N D ~ R S . Can. J . Chem. 53. 338(1975).

37. Y. Os.%\\-.\. K . SHIRATA. D. ROHRER. C . WEEKS. and W. L . DL \x . 1. .Am. Chem. Soc. 97.4400 (1975).

38. N. S. HHACC 4 and D. H . WILLI.\XIS. Applications of nmr spectroacopq in organic chemi\tr) . Holden-Da!. San FI-an- cisco. C,A. 1964. (11) pp. 7 and 35: ( h ) p. 24 and reference\ cited: (c , ) pp. 20-22.

39. N . S . BH..\cc A. L . F. J o ~ u s o u . and J . S . S H O O L L R ' ~ . High resolution nmr spectl-a catalog. Vol. 1. V~I- inn :\\sociate\, Palo Alto. C.4. 1962. Spectr~irn No. 363.

40. N. S. B H \ c ~ , A . D. P. HOLLIS. L . F. JOHTLSON. ;111d E . .A. Pl t~c . High I-esol~~tion nml- spectl-a c;~talog. Vol. 2. Varian Xs\ociates. Pnlo Alto. CA. 1963. Spectrum No. 698.

41. W . . A R ~ O L D . W. ~ I E I S T L R , and 6. E ~ G L L R T . Helv. Chim. i-\cta. 57. 1559 (1974).

42. C. E . JOHNSON. JR. and F. ,A. B O \ , E \ ~ . J . Chem. Phyi. 29. 1012 (1958): J . U'. ERISLE>. J . FFEUEY. and L. H. SU ICLIFFE. High resolution nucleai- magnetic I-e\on:ince spectroscopq. Vol. 1. Per-gamon Press. London. England. 1965. pp. 595-604.

43. J . 4. OMJOYALL. Ph.D. Dissertation. Uni~ersi tq of Florida. Guinesville, FL. 1975.

44. hl. W. C O U C H , K . N. SCOTT. and C . h4. WII I - IA\ IS .

Steroids. 27, 451 (1976). 45. M . AKHTAR and D. H. R. BARTOI. J . ,4111. Chem. Soc. 86.

1528 (1964). 46. D. R. J ~ h f t s and C . W. S H O P P F ~ ~ . J . Chem. Soc. 4224

(1954). 47. R. 11. M O R I A R ~ Y and T . D. J . D'SILVA. Tetl-nhedl-on. 21.

547 (1965). 48. J . K-\LVOD.A. K . HELSLER. J . UERER\V.\SSFR. G . A V ~ E R ,

and A. WETTSTEIN. Heiv. Chim. Acta. 46. 1361 (1963). 49. R . E. COLZSELL. V. V. R ~ N A U ~ , R. J. BLAIR. W. H.

B ~ I E R W A L I ES. and P. A . WEINHOLD. Steroids, 16. 317 (1970).

50. J . P. D u s z ~ . J . P. JOSEPH. and S . B E R N S T E I ~ . Steroids, 8 , 495 (1966).

Can

. J. C

hem

. Dow

nloa

ded

from

ww

w.n

rcre

sear

chpr

ess.

com

by

UN

IVE

RSI

TY

OF

SOU

TH

CA

RO

LIN

A o

n 11

/15/

14Fo

r pe

rson

al u

se o

nly.

13 c and 1 h nuclear magnetic resonance spectroscopy of c-19 and 6β-methyl substituted steroids:...

Documents