INCORPORATION OF [2-14C, 2',3'-13CC2]TRYPTOPHAN INTO VINDOLINE ESTABLISHED BY 13C S M R EDIYARD LEETE ,\'atural
Products Laboretory,I School of Cheniistry, Cnicersitg of 3finnesota, 1f7 nn eapolzs, M A - 55455
ABSTR ~ C T . - D L - [ ~ - ~ ~ C 2 ',3 , '-13C?]Trypt ophan was prepared from [2-14C]indole, [13C]formaldehyde, and diethyl r2-13C]malonate. T h i s amino acid, labeled with contiguous 13C atoms in its side chain, was fed t o t h e plant Catharanthus roseus, resulting in t h e formation of radioactive vindoline (0.19cc specific incorporation). An examination of t h e 13C nmr spectrum of this alkaloid revealed t h e presence of satellites located about t h e signal of C-G, which &-ere due t o spin-spin coupling of t h e contiguous 13C atoms located a t C-5 and C-6 of vindoline. Since t h e specific incorporation, deduced from the intensity of t h e satellites relative t o t h e central singlet peak, was t h e same as t h a t of t h e 14Clocated in the indole nucleus, t h e intact incorporation of tryptophan into vindoline is unequivocally established
Most of the early work on the biosynthesis of secondary natural products involved the use of putative precursors labeled Ivith I4C, the radioactive isotope of carbon. Often quite lengthy and complicated degradations had t o be carried out to determine the location of the I4C in the ultimate natural product. It is non- possible to avoid these degradations by the use of I3C as a tracer. This isotope is not radioactive, however, its location in a natural product, derived from a precursor n-hich is enriched with I3C above the natural abundance (1.117G) is possible through the use of nmr. The use of precursors labeled with a single I3C atom has been used extensively in the study of the biosynthesis of microbial natural products (1-3)) since one can usually expect high specific incorporations* in such systems. The labeled position in the natural product will have an enhanced signal at that carbon in its 13Cnmr spectrum. Since the intensity of nmr signals is somenhat variable, this method can only be used nhen the enhancement of the signal above the natural abundance is a t least 30-40%. For example. a natural product m s found to have 1.670 I3C a t a certain position, and was derived from a precursor enriched 90% a t the analogous carbon. The specific incorporation would be 0.557,,and the nmr signal for this carbon n-odd be enhanced 4576. I n higher plants the specific incorporation of precursors into alkaloids and other natural products is usually quite low, and only a few examples using this method have been reported (4-9). A much greater dilution of the administered precursor can be tolerated if one labels with contiguous I3C atomc. The natural occurrence of such contiguous carbons is only 1.11% x 1.117,=0.012376. Furthermore, these contiguouq carbon-13 atoms give rise to satellites (due to spin-spin coupling) in the I3C nmr, located about the central singlet peaks, which arise from natural abundance I3C and the incorporation of singly labeled species. This method wis first used in biocynthetic studies by Set0 and com-orkers (10, l l ) , who fed [1,2-13C2]acetic acid to variouc microbial systems. We have used [5,G-13C2]nicotinic acid for investigating the biosynthesis of the tobacco alkaloids (12) and dioscorine (13). 'Contribution S o . 170 from this laboratory. P a r t of this m-ork was presented a t t h e annual meeting of t h e -4merican Society of Pharmacognosy, held a t Purdue L-niversitj-, V e s t Lafayette, Indiana, J u l y 29-August 3, 1979. 2Specific excess in t h e natural product:? excess in t h e administered precursor. F o r radioactive compounds: dpm mXI of t h e natural product/dpm/m?lf in t h e precursor. 130
JAS-FEB
19801
LEETE :
TRI-PTOPHAS ISCORPORATIOS
ISTO VISDOLISE
131
I t has also been shown that [2,3-13C2]ornithineyields nicotine and nornicotine which are labeled symmetrically with contiguous atoms on either side of their pyrrolidine rings (14). I n the present article, 11-e describe the synthesis of tryptophan labeled n-ith contiguous 13C atoms in its side chain, and n-e hare demonstrated its utility for investigating the biosynthesis of the indole alka1oid.s found in Catharmtlrzis roseus. The s>-ntliesisof DL-[2-l4, 2',3'-13C?]tryptoplia~i ( 5 ) is illustrated in figure 1: and e
-2
-5
COOEt
COOE'
6
v
FIG.1. S>-nthesisoi UL-:2--"C, 2',a'-13CS.Tr!ptop1~ar~.( *
I4C,
I3C)
is based on the previously described preparation of unlabeled tryptophan (see Experimental). Tlie amino acid n-as labeled with at C-2 of its indole nucleus o that its specific incorporation into t h e ultimate alkaloids: and the rate of uptake nto the plants could be readily determined by radioactive assay. The 13C nmr spectruni of the IC-enriched trj-ptophan is illustrated in figure 2 . The cheniical shifts were assigned by comparison with model compounds: indole and 3-methylindole (15). Our observed values 1%-erein good agreement n-ith the previously published I3C nmr spectra of natural abundance tryptophan (16-18j. The enriched positions, C-2' and C-3': exhibit the expected satellites n-itli a coiipling constant of 33.2 Hz. The satellites are not symmetrically central peaks. This is due to a n approach to an ;1B spiii unenriched carbons which are adjacent to C-2' and C-3' n triplets: 1J112= 53.5 Hz. 1,J3,31= 44 Hz. The labeled tryptophan Tws fed to Catharauthits roreiis plants by the wick ethod. I n order not t o ovei oad the plants n-ith trJytuphan, the f2eding i n i s pread over 5 da!-s. The plant T1-ei.e harvested after tn-o 11-eeks, and the al1;nloids isolated and separated as prev ousl?- described (2oj. Vindoline (see figure 3 ) is the most abundant alldoid in the plant, and radioactive assay indicated that the Fpecific incorporation of tryptophan into this alkaloid u-as O.IO$. Its nnir spectrum has been previously determined (21)) and n-e obtained essentially 1
JOURNAL OF NATURAL PRODUCTS
132
[VOL.
43, so. 1
-I
I
I I00
50
I
I
50
PP"
I
I
I
FIG.2. Proton nolse-decoupled FT13C XMR spectrum of [2',3'-13C2jtryptophan in D20 a s i t s sodium s a l t . T h e upper spectrum is enhanced ten times in t h e vertical scale, so t h a t t h e natural abundance carbons can be observed.
El I
I
i
I
t - ,I C;.'e
-
d.,
J5,6=33 2 Hz
I 6C
50
DFn
40
FIG.3. Proton noise-decoupled FT x3CXMR spectrum of enriched vindoline (aliphatic region) in CDC1,.
JAX-FEB
19801
LEETE
: TRI PTOPHAS
ISCORPORATIOS
ISTO VISDOLISE
133
the same spectrum except that we found that C-5 (51.9 ppm from 3IeBi) n-as resolved from the eqter O N e gioup (52.1 ppm). The previous nork assigned both these carbons to a signal at 51.9 ppm. The carbons which are expected to be enriched in the vindoline are C-5 and C-6. Figure 3 illustrates that region of the 13C spectrum of the enriched vindoline u-here these carbons resonate. Satellites are readily observable at C-6 (1Jj,6 = 33.2 Hz), again asymmetrically arranged in accord with theory. The specific incorporation of into vindoliiie \\as determined by measuring the intensity of these satellites relative to the central peak, the calculation taking into account the contribution of singly labeled species to the central peak ( 2 2 ) . The satellites at C-5 are not observed m c e they fall under the signals for C - i and C-3. Hou-ever, it is possible to detect qhoulders on these signal. due to the presence of the satellite peaks. The cpecific incorporation of tr) ptophan into vindoline determined from its 13C nmr spectrum na. found to be 0 ?lac, in excellent agreement with the result obtained b?- radioactive abqay Since the tryptophan was labeled n i t h in the indole nucleus. and v i t h 13C in the side chain, the present requltq clearly demonstrate, for the first time. the incorporation of tryptophan as an intact unit into the indole alkaloid vindoline. Previous results (20, 23) only shon ed that the tr) ptophan .ide chain n a i incorporated into vindoline. I t should be pointed out that this method of stud) ing the bios) nthe natural products, using 13Clabeled precur-or*, can J ield information ver? rapidly. I n the present caqe, the n-hole experiment. including feeding, isolation of the vindoline. and determination of it+ 13C nnir spectrum, n a. completed in le-. than a month. Using radioactive tracer, it took over a year to eqtablish that [ 2 ' - ' T ] tryptophan yielded vindoline labeled at C-5 (20). A disadvantage of thiq method of elucidating biosynthetic patlin ays ib that one need, relatively large amount5 of the 13C-labeled natural product for determination of its 13C nmr spectrum, especiall? nhen the specific incorporation is low. Houever, n e can hope that improvement. in instrumentation nil1 make this less of a problem in the future. EXPEK131ESThL3 DL-[2-14C, 2',3'-'3Cz]TRrPTOPH.~s ( 5 ).-[2-14C]Indole (1) (25), dimethylamine, and [13cc] formaldehyde (Merck Sharpe and Dohme of Canada) were condensed in dilute acetic acid according t o the literature method (26) t o afford [2-14C, nzeth~Zene-13C]gramine(2). I t s proton noise-decoupled 13C nmr spectrum (in dimethylsulfoside-de) was essentially t h e same a s t h a t recorded b y Wenkert et nl. ( 2 i ) , except t h a t t h e signal a t 111.6 ppm (C-3) was split into a triplet by t h e adjacent *3C enriched methylene carbon ('J=51 H z ) . Carbon-2 (124.2 ppm) was also split due t o a geminal coupling with the enriched carbon (*5=4.8 Hz). No splitting of t h e S - m e t h y l groups a t 4 4 . i ppm was observed. Mass spectrometry indicated t h a t the gramine was enriched 7 1 5 with I3C. Diethl-1 [2-13Cjmalonate 14) (hferck Sharpe and Dohme of Canada) was converted t o ethyl [2-'3C]acetamidomalonate 13) by reaction with nit roils acid, followed by reduction of t h e resultant nitroso compound, dissolved in acetic anhydride, with h?-drogen in t h e presence of platinum on charcoal 128j. Mass spectrometry indicated a enrichment of sic; in (33. T h e labeled gramine and 13) were condensed in absoliite ethanol in the presence of sodium etlioside and dimethyl sulfate (29) t o yield e t h y l a-carboethosy-aacetamido-$-i3-indolyljpropanoate (6). D L - [ P 4 C , 2',3'-l3C2!Tryptcphan acetate was obtained from this ester as previously described (283. Since tryptophan does not yield a si!bstantial molecular ion in t h e mass spectrometer, the distribution of the 13C labeled species was deduced by mass spectrometery on t h e ester 16) which affords a good molecular ion with no hl- -1 peak. This material was found t o contain 4CC;; l3Cz,4 i 5 and 1 3 5 unenriched. Since the enrichment of the gramine and ethyl acetamidomalonate were i 1 and si5 respec3The l3C nmr spectra were determined by Dr. Robert 11.Riddle on a 1-arian SL-160-15 spectrometer ( 2 5 . 2 N H z ) equipped with a YFT-100 Fonrier-transform accessory. 1Iass spectra were determined by D r . Roger L-pham on an AEI-30 spectrometer. Radioactive materials &-ereassayed in duplicate in a Siiclear Chicago Mark I1 liquid scintillation counter using diorane-ethanol as solvent with t h e usual scintillators (24).
JOURNAL OF SATCRAL PRODUCTS
134
tively, the composition of t h e [3'-I3C] tryptophan.
13C1
[VOL.
43, so. 1
species are calculated t o be: 16.57, [2'-13C]- and 30.5%
ADMISISTRATIOS OF THE DL-[~-'%, 2',3'-'3C2j~p,i,p~op~.is TO Catiiaraniiius roseus ASD ISOLAT I O S OF THE .ix.iLoms.-The plants were growing in soil in a greenhouse and were three months old a t the time of feeding (.J;ily). T h e plants were grown from seeds purchased a t t h e local supermarket labeled: Vinca (Periwinkle), mixed colors. T h e DL-[2-14CI 2t,3t-13C2]tryptophan acetate (0.33 m11, 4.7G x 1C6 dpm(m11) was fed t o 20 plants via cotton wicks inserted into t h e stems of t h e plants near t o ground level. T h e feeding was carried out over 5 days, on each da>- an additional q u a n t i t y of t h e tryptophan acetate dissolved in x-ater x-as added t o t h e beakers in which t h e wicks ivere placed. T h u s each plant received 3.3 p l f of tryptophan per day. Two weeks after t h e initial feeding, t h e plants (fresh weight 2.3 kg) were harvested. The residual activity in t h e beakers was 4.5 x lo3 dpm (0.3yGof t h e t o t a l amount fed). T h e alkaloids were extracted and separated as previously described (20). T-indoline R-as finally purified by tlc on Silica gel PF-254 (Merck) developing with a mixture of chloroform, benzene, methanol, and ammonium hydroxide solution (100:200:30:1). T h e zone (Rf0.6) corresponding t o vindoline was extracted with chloroform, affording after crystallization from e t h e r , colorless needles of vindoline (75 mg), 9.1 x lo3 dpni/mM ( 0 . 1 9 5 specific incorporation). Catharanthine (12 m g ) , 8.3 s lo3 dpni!nihI (0.17yospecific incorporation) was also isolated, but the amount was insufficient for observation of satellites in i t s 13C nmr. DETERhfISATIOS O F THE I3C NMR sPEcTRA.-The DL-[2P4C, 2',3'-W2jtryptophan (64 m g in 0.4 ml of D20, containing an equivalent amount of NaOH, in a 5 mm tube) was run for 13 K transients, 0.73 see acquisition time, 1.47 H z / d a t a point. T h e chemical shifts, ppm from i\lerSi (carbon S o . ) were: 186.3 ( l ' ) ]60.3 (2l), 34.5 @ I ) , 127.8 (2), 114.5 (3), 131.0 (3a), 122.4 (4), 125.3 (5), 122.7 (G), 115.4 ( 7 ) , 139.8 (7a). T h e average distance of t h e inner satellites t o t h e central peaks of C-2' and C-3' was 16.1 H z . T h e theoretical distance, calculated from t h e formula: ~ ( 1 - A B + J A B - ~ 1 - A B 2 + J * ~ 2 ) (30) where l-AB=difference in chemical shift between t h e coupled carbons (651 Hz for C-2' and C-3'), and J A B = t h e coupling constant (33.2 H z for C-2' and C-3'), was 16.2 H z . T h e vindoline (72 m g in 0.4 ml of CDC1, in a 5 m m tube) --as run for 20 K transients, 1.6 see acquisition time, 0.7 H z / d a t a point. T h e inner satellite of C-6 n-as 15.4 Hz from t h e central peak, in excellent agreement with the calculated value (15.3 Hz), where lTj,s=199Hz, and J5,6=33.2 Hz. ACKNOWLEDGhlESTS This investigation was supported b y Research Grant Gh1-13246-22 from t h e Kational Institutes of Health, U. S. Public Health Service. I thank D r . 11.Gorman of the E l i Lilly Co. for generous supplies of vindoline and other alkaloids of Catharanthzis roseus. R e c e i w d 30 July 2979. LITERATURE CITED 1. 11.Tanabe, B i o s y n t h e s i s , 2, 241 (1973); 3, 247 (1975); 4, 204 (1976). 2. E. Sequin and A. I . Scott, S c i m c e , 186, 101 (1974). 3. 4.G . McInnes, J. +. K a l t e r , J . L . C. Wright and L. C. Vining, "Topics in Carbon-13 N N R Spectroscopy , E d i t o r : G . C . Levy, 1-01. 2, p. 123, Wiley-Interscience, New York, 1R76.
4.
5. 6.
7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21.
C:R.
Hutchinson, .i. H. Heckendorf, P. E. Daddona, E. Hagaman and E. Wenkert, J . 4 m . Cheni. Soc., 96, 5609 (1974), 97, 1988 (1975). -4.R. B a t t e r s b r . P . W.Sheldrake and J. 4.Rlilner, Tetrahedroiz Lett.. 3315 (1974). , , E. Leete and G".'B. Bodem, J . Ani. Clien?. SOL.,98, 6321 (1976). E. Leete, K . C. Ranbom and I?. hl. Riddle, P h p t o c h e m i s i r y , 18, 75 (1979). S . Takao, K. Iv-asa, h!. Kamijaiichi and 11.Sugiura, Ciiem. Pharm. Bull., 24, 2859 (1976). A. >-agi, G . Xonaka: 5;. S a k a y a n i a and I. Nishioka, Phytochemistry, 16, 1197 (1977). H. Seto, T. S a t o and H. 1-onehara, J . Ani. Chew. Soc., 95, 8461 (1973). H . Seto, W .Cary and 11.Tanahe, J . C h e w . 5'06. Cheni. Comm., 867 (1973). E. Leete, Bioorg. Cheni., 6, 273 (1977). E. Leete, Piiyiocliewisfrp, 16, 1705 (1977). E. Leete and 1l.-L. 1-11, Phyiochearisfry, in press. J . F. Johnson and R.C . Jankowski, "Carbon-13 S M R Spectra", John Wiley and Sons, S e x - York, PiT (1972), Spectra S o . 283 and 341. J . H. Bredbury and R . S. Korton, Biochiiii. Biophys. Acta, 328, 10 (1973). -1.Allerhand, R . F. Childers and E . Oldfield, Biochemistry, 12, 1335 (1973). K. 1-oelter, S.Fuchs, R . H. Steuffer and K . Zech, M o n a t s h . Cheiir., 105, 1110 (1974). 31. L. Case>-,R . C. Paulick and H. W.Whitlock, J . A m . Ciiem. Soc., 98, 2636 (1976). E . Leete, A. Ahmad and I . Kompis, J . Aiir. Ciienz. Soc., 87, 4168 (1965). E. Wenkert, D . IT. Cochran, F. W .Hagama,n, F. 31. Schell, K . Keuss, 4.S. Katner, P. Potier, C. K a n , hf. P l a t , 11.Koch, H. RIehri, J . Poisson, N . Kunesch and Y. Rolland, -7. .47?2.C i i e i i f . soc., 95, 4990 (1973).
JIS-FEB
22.
1!%o]
LEETE :
T R l P T O P H l S ISCORPORdTIOS ISTO VISDOLISE
135
E. Leete, S . K o ~ v a n k oR, . -1.S e r r m a r k , L . C. 1-ining, -1.C;. IIcInnes and J . L. c'. Wright,
Tetrahedroti L e f t . ,-1103 119i3). D. (;roger, h.Ptolle and h.Mothes, Teirahedroji Let/., 2579 (1964). .1.I:. Friedman and E. Leete, J . .4m.Chem. S o c . , 85, 2141 (1963). E. Leete and L. IIarion, Ciiti. J . Cheiu., 31, 1195 (1953). H. Kuhn and 0 in, Chejji. Ber., 70, 5 6 i (1937). E . Wenkert, J . ndra, C.-J. Chang, D . K. Cochran and F. 11.Scliell, Act. Chriii. Rcz., 7 , -l(i119741. 28. H. I?. Snyder and C. I T . Smith, J . rim. C h e w . .SOL., 66, 350 (1944). 29. S . F. dnderson, S.h r c h e r and C . AI. Suter, J . .Am. Ciieni. Sor., 67, 36 (1945). 31. L. 11.Jackmari and S.Sternhell! "-ipplications of S n c l e a r Magnetic ResoncAnceSpectroscopy in Organic Chemistry", 2nd Edition, Pergarnon Press, 1969! p. 129. 23. 24. 25. 26. 27;.
