Journal of Crystallographic and Spectroscopic Research, Vol. 23, No. 6, 1993

Molecular structure of Lantadene-B&C, triterpenoids of Lantana camara, red variety: Lantadene-B, 22p-angeloyloxy3-oxoolean-12-en-28-oic acid; Lantadene-C, 22 p-(S)-2'-methyl butanoyloxy-3-oxoolean-12-en-28-oic acid I M. Nethaji, 2 C. Rufes, 3 C. Sadasivan, 3 Vasantha Pattabhi, *'3 and O. P. Sharma 4

Received June 23, 1992," accepted August 28, 1992 (I) Lantadene-B: C35H5205, Mr = 5 5 2 . 8 0 , M o n o c l i n i c C 2 , a = 2 5 . 6 5 ( 1 ) , b = 6 . 8 1 9 ( 9 ) , c = 1 8 . 7 5 ( 1 ) ] 1 , / 3 = 1 0 0 . 6 1 ( 9 ) , V = 3 2 2 3 ( 5 ) ] 1 3 , Z = 4, Dx = 1.14 g c m -3 C u K s ()x = 1 . 5 4 1 8 A ) , /z = 5.5 c m - l , F ( 0 0 0 ) = 1208, R = 0 . 1 1 8 , wR = 0 . 1 3 2 f o r 1527 o b s e r v e d reflections w i t h Fo

>- 2a(Fo). (II) Lantadene-C: C35H5405"CH3OH, Mr = 5 8 6 . 8 5 , M o n o c l i n i c , P 2 1 , a = 9 . 8 2 2 ( 3 ) , b = 1 0 . 9 0 9 ( 3 ) , c = 1 6 . 1 2 0 ( 8 ) ] t , /3 = 9 9 . 8 2 ( 4 ) , V = 1702(1)]13, Z = 2, Dx = 1.145 g c m -3, M o K s (X = 0 . 7 1 0 7 1 1 ) , /x = 0 . 7 0 8 c m -1 F ( 0 0 0 ) = 6 4 4 , R = 0 . 0 9 8 , w R = 0 . 0 9 4 f o r 1073 o b s e r v e d reflections. The rings A , B, C, D , and E are trans, trans, trans, cis fused and are in chair, chair, sofa, half-chair, chair conformations, respectively, in both the s t r u c t u r e s . I n the unit cell the molecules are stabilized b y O - - H 9 9 9 O hydrogen bonds in both the structures, however

an additional

C--H-

in the case of Lantadene-C.

9 9 O i n t e r a c t i o n is o b s e r v e d

Introduction

2g

Ingestion of Lantana foliage causes cholestasis and hepatotoxicity in animals (Sharma et al., 1988). Lantadene-A, B, and C (I) are the pentacyclic triterpenoids isolated from the leaves o f L a n t a n a camara, Red variety (Sharma et al., 1989). Lantadene-A (form II) and Lantadene-C (form I and form II) are hepatotoxic to guinea pigs (Sharma et al., 1989, 1991); while Lantadene-B an isomer of Lantadene-A has been found to be nontoxic (Sharma, 1991). In recent years, it has been realized that it is of importance to investigate the three dimensional structure of xenobiotics (Williams, 1989). Such knowledge would be helpful in designing rational therapeutic strategies against the adverse biological effects of xenobiotics and toxins. The crystal structure of Lan-

3O

H

P, II ~-C-R 31

28

H

25H.. I

lt%L

163 H3

Ja

~1

23,,~2Z. H

.I

R

:

~ R :

/N

-- C = C

LANTA

H

34

i

~Ha

--C = C " / 32

33 ~.fiH 3 3,~

LANT B

35

~ H3

R :

~C--C

32

H

/ ~H

LANT C

~,3CH3

Scheme 1 ~DCB Contribution No 809. 2 Department of Inorganic and Physical Chemistry, Indian Institute of

tadene-A (form I) has been determined by our group (Pattabhi et al., 1990). Crystal structure determinations of Lantadene-B and C were undertaken to elucidate their molecular structure and conformation and to compare them with Lantadene-A in order to correlate their con-

Science, Bangalore-560 012, India. 3Department of Crystallography and Biophysics, University of Madras, Guindy Campus, Madras-600 025, India. 4Biochemistry Laboratory, Indian Veterinary Research Institute, Regional Station, Palampur, H.P.-176061, India.

469 0277-8068/93/0600-0469507.00/0 9 1993PlenumPublishingCorporation

470 formation with biological activity. The differences in molecular geometry of Lantadene A, B, and C have been discussed in terms of the manifestation of their biological action.

Experimental (I) Lantadene-B Thin colorless needles, 0.1 x 0.075 x 0.225 ram, were obtained from a cyclohexane-methanol mixture. The three dimensional intensity data were collected on an E N R A F - N O N I U S CAD-4 automated diffractometer with monochromated Cu Kc~ ( k = 1.5418 A) co/20 scan mode with 2 < 20 < 120 ~ absorption correction with an average correction factor 0.90 and maximum and minimum values of 0.998 and 0.841. - 2 8 _< h _< 27, 0 <__k _< 8, - 2 0 _< l <_ 20, 5245 reflections measured, using Cu Kc~, 2195 unique and 1527 observed with Fo >_ 2a (Fo), cell constants from 15 reflections 14 < 20 < 40 ~ . Three standard reflections monitored every hour showed no significant variation in intensity. The intensity data were corrected for Lorentz and polarization effects but not for absorption. Structure solution was by direct methods using the program SnELX-86 (Sheldrick, 1986). Hydrogen atoms were fixed by geometrical considerations and were refined as riding on the atoms to which they are bonded and associated with isotropic temperature factors. Full matrix least-squares refinement o n Fob s using SHELX-76 (Sheldrick, 1976) with non H-atoms anisotropic gave a final R = 0.118 and wR = 0.132 with individual weighting scheme based on counting statistics where (w = 0.3178/aZFo + 0.034105F 2) and (A/a)m~x = 0.8. R = 0.14 for the opposite enantomer. The non-H-atoms were divided into three blocks and refined in order to compensate for the low parameter to observation ratio. The final Ap map has no peaks > 0.45 e A -3. Bond length and angles calculations were done using PARST (Nardelli, 1983). (II) Lantadene-C As for (I) but with special values described below. Thin transparent colorless needles 0.3 • 0.3 x 0.2 m m were obtained from chloroform-methanol mixture. Monochromated Cu Kc~ (X = 1.5418A) radiation, co/20 scan mode with w -- (1.30 + 3.90 tan 0), aperture width = (0.35 + 1.05 tan 0) mm, 2 < 0 < 50, 0 < h <_ 1 1 , 0 <_ k_< 12, - 1 9 <_ l _< 19, 3436 measured, 2896 unique and 1073 observed with IFol > 5.a (Fo) ,

Nethaji, Rufes, Sadasivan, Pattabhi, and Sharma cell constants with 20 reflections 9.8 < 0 < 11.2 ~ two standard reflections 1 3 6 and 1 - 5 - 2 monitored every 100 reflections showed no significant variation in intensity. Structure solution was by direct methods using SHELXS-86 (Sheldrick, t986), nine H atoms from Ap synthesis, one solvent molecule (methanol) was located from a difference Fourier, full matrix least-squares refinement was carried out (in two blocks) on Fo using SHELX-76 (Sheldrick, 1976) with non-H-atoms, but for C33 and C34, anistropic after C33, C34 and H atoms isotropic, converged to a final R factor o f 0.098 and wR = 0.094 with an individual weighting scheme based on counting statistics, where w = 1/a 2 (Fo) + 0.0001 Fo2 (A/a)max = 0.15, (A/a)min = 0.00, final AO map had no peaks > 0 . 2 5 e A -3 S = 2.95 for 408 parameters. There was no change in the residual index for the opposite enantiomer. Scattering factors are as given in SHELX-76.

Discussion A stereo view of the molecules is shown in Figs. 1(a) and (b). All the figures were drawn using the programs PLUTO (Motherwell and Clegg, 1978) and CInTRA (Gautham, 1989). Average standard deviations in bond lengths is 0.02 A, and 0.04 A and in bond angles is 1.1 and 2 ~ in Lantadene-B and C, respectively. The accuracy of the structure determination is affected by the poor quality of the crystals. In both the structures some of the bond lengths deviate from standard values but these latter are within the limits of experimental errors. Atoms C33, C34, and C35 in Lantadene-C and C30 and C34 in Lantadene-B have high thermal parameters which is reflected in the bond lengths and bond angles involving these atoms. The packing of the molecule in Lantadene-C is stabilized by C1 9 - - O 4 ( 3 . 2 9 ) A , C2 9 9 9 0 4 (3.14)A and a O - - H 9 9 9 O hydrogen bond between carboxyl group oxygen (03) and solvent oxygen (OW). 0 3 9 9 9 OW is 2.65(3)A. As the protons attached to the solvent oxygen and 03 have not been located the hydrogen bond geometry is not discussed. In Lantadene-B, the packing is stabilized by a 0 3 9 9 - 0 2 hydrogen bond only ( 0 2 " 9 9 0 3 is 2.70(2)A). The tings A, B, C, D, and E are in chair, chair, sofa, half-chair and chair conformations, respectively, and fused in the trans-trans-trans-cis fashion forming an extended structure in both the molecules. The presence of a double bond in the C ring between C13 and C12 accounts for its twist conformation. Lantadene-A, B, and C are similar except for the

Structure of Lantadene-B and C

471

0

(a)

0

0

0 0 (b) Fig. 1. Stereoviewsof (a) Lantadene-B and (b) Lantadene-C.

sidechain (at C22) atoms, C32 and C33 which are connected by a single bond in Lantadene-C and by double bond in Lantadene-A. The conformational differences are also observed only at this end of the molecule. Table

3 gives the relevant torsion angles for comparison. From Table 3 it is evident that the side chain conformations in Lantadene-A and B are identical while that of Lantadene-C varies. Figure 2 shows a superposition of the

472

Nethaji, Rufes, Sadasivan, Pattabhi, and Sharma

0"0 0

...

.~ = . ~

:~'--Lonta Lantb ....... Lantc

0

Fig. 2. Superposition of Lantadene A, B, and C.

three molecules. The RMS deviation o f atoms when Lantadene A and C, C and B and A and B are superimposed are, 2.413, 2.416 and 0.87 A respectively. The sidechain atoms and 0 2 have an average RMS deviation of 2.76 A ; while the other atoms have an average deviation o f 0.21 A. in the case o f the Lantadene-A and B superposition. The differences observed in sidechain conformation o f Lantadene-A, B and C are mainly due to the presence or absence o f the double bond at C32. Lantadene-C has an asymmetric carbon at this end (viz C32) while this has been lost in Lantadene-A and B due to the presence o f the double bond. These differences indicate that Lantadene-C may be binding to a receptor

for its activity where the asymmetric carbon and carboxyl oxygen (viz 0 5 ) have an important role to play. Further the presence of two methyl groups at C33 in Lantadene-B may be posing steric hinderance to the active site of the receptor. One is tempted to speculate that the C34 atom in Lantadene-A (form I), which is in cis conformation with respect to C31, may rotate to the trans position in form II as in the case of Lantadene-C, which makes it active. However, the potency of Lantadene-A (form II) and Lantadene-C may differ due to the absence o f an asymmetric carbon at C32 in Lantadene-A.

References Gautham, N. (1989) CHITRA program for plotting molecular and crystal structures. (Univ. of Madras, India). Motherwell, W. D. S., and Clegg, W., (1978) PLUTO program for plotting molecular and crystal structures. (Univ. of Cambridge, England). Nardelli, M. (1983) Comp. Chem., I, 95. Sharma, O. P., Makkar, H. P. S., and Dawara, R. K. (1988) Toxicon 26, 975. Sharma, O. P., and Sharma P. D. (1989) J. Sci. Indust. Res. 48, 471. Sharma, O. P., Dawara, R. K., and Pattabhi, V. (1991)J. Biochem. Toxicol. 6, 57. Sheldrick, G. M. (1976) SHELXS76. Program for crystal structure determination. (Univ. of Cambridge, England). Sheldrick G. M. (1986) SHELXS86, Program for crystal structure solution. (Univ. of Gottingen, FRG). Pattabhi, V., Sukumar, N., and Sharma, O. P. (1990) Acta Cryst. C46, 810. Williams, K. M. (1989) Clin. Exp. Physiol. 16, 465.

Structure factor data have been deposited with the British Library, Boston Spa, Wetherby, West Yorkshire, UK, as supplementary publication No. 63286 (35 pages).