Total Synthesis of Amaryllidaceae Pyrrolophenanthridinium Alkaloids via the Ziegler-Ullmann Reaction: Tortuosine, Criasbetaine, and Ungeremine

Full text of DOI:10.1021/jo991902p

J . Org. Chem. 2000, 65, 3227-3230
3227
conferring antitumor properties within the pyrrolophenan-
thridinium class of alkaloids.6
Tota l Syn th esis of Am a r yllid a cea e
P yr r olop h en a n th r id in iu m Alk a loid s via th e
Ziegler -Ullm a n n Rea ction : Tor tu osin e,
Cr ia sbeta in e, a n d Un ger em in e
Lucy M. Stark,1 Xiao-Fa Lin, and Lee A. Flippin*
Department of Medicinal Chemistry, Neurobiology Unit,
Roche Bioscience, 3401 Hillview Avenue,
Palo Alto, California 94304-1397
Received December 10, 1999
In tr od u ction
In connection with a program in our laboratory directed
toward the total synthesis of pyrrolophenanthridinium
alkaloids for antitumor screening,2 we required a short
and general approach to the Narcissus tortuosus con-
stituent tortuosine, 1,3 its structural relative criasbetaine,
2,4 from Crinum asiaticum, and ungeremine, 3.5 Struc-
ture-activity relationship (SAR) data for antitumor
efficacy within the small class of known Amaryllidaceae
pyrrolophenanthridinium alkaloids is very limited in
scope; however, several encouraging preclinical studies
of ungeremine have appeared over the past two decades.6
Recently, significant efficacy against human ovarian and
stomach cancers has been claimed for semisynthetic
ungeremine acetate in human clinical trials conducted
in the People’s Republic of China.7 The results of limited
SAR studies related to the preclinical development of
ungeremine have led to the suggestions that C-ring
betaine functionality, A-ring oxygenation, and an optimal
OA-NB-OC angle, may be important determinants in
Tortuosine and criasbetaine were initially selected as
synthetic targets for the current study for the following
reasons: (1) No directly comparable biological studies of
1 and 2 have been reported, although the two compounds
may provide a relatively subtle SAR probe of the effects
of pyrrolophenanthridinium alkaloid C-ring functionality
on antitumor efficacy, (2) compounds 1 and 2 have not
been previously reported by total synthesis, and (3) the
poor availability of these compounds is currently a
serious limitation to further study. Tortuosine and cri-
asbetaine have so far been isolated in low yield from their
respective natural sources. Compounds 1 and 2 have also
been prepared in very limited quantities, for the purpose
of structure confirmation, by SeO2-mediated degradation
of stereochemically complex Amaryllidaceae alkaloids;
however, the required precursor alkaloids for these
semisynthetic preparations (galanthine f 1 and meth-
ylpseudolycorine f 2)3c,4e are themselves essentially
inaccessible. Two previous total syntheses of 3 have been
reported; however, neither of the extant approaches is
reported to give practical quantities of ungeremine.8
Interestingly, all of the ungeremine acetate employed in
preclinical and human clinical studies in the PRC was
apparently obtained via the SeO2-mediated degradation
of lycorine.6,7
(1) Roche Bioscience Student Intern, 1998.
(2) A preliminary account of the synthesis of tortuosine and crias-
betaine has been presented: Stark, L. M.; Lin, X.-F.; Flippin, L. A.
Book of Abstracts, 217th ACS National Meeting, ORGN-398, 1999,
Anaheim, CA.
(3) (a) Amanza, G.; Fernandez, J . M.; Wakori, E. W.; Viladomat, F.;
Codina, C.; Bastida, J . Phytochemistry 1996, 43, 1375. (b) Bastida, J .;
Fernandez, J . M.; Viladomat, F.; Codina, C.; de La Fuente, G.
Phytochemistry 1995, 38, 549. (c) Fales, H. M.; Wildman, W. C. J . Am.
Chem. Soc. 1956, 78, 4151.
(4) (a) Tran, C. K. Tap Chi Duoc Hoc, 1998, 7, 11 [Chem. Abstr.
1999, 130, 158320]. (b) Ghosal, S.; Singh, S. K.; Unnikrishnan, S. G.
Phytochemistry 1990, 29, 805. (c) Ghosal, S.; Singh, S. K.; Srivastava,
R. S. Phytochemistry 1986, 25, 1975. (d) Ghosal, S.; Kumar, Y.; Singh,
S. K.; Kumar, A. J . Chem. Res., Synop. 1986, 3, 112. (e) Fales, H. M.;
Giuffrida, L.; Wildman, W. C. J . Am. Chem. Soc. 1956, 78, 4145.
(5) (a) Bastida, J .; Codina, C.; Porras, C. L.; Paiz, L. Planta Med.
1996, 62, 74 [Chem. Abstr. 1996, 124, 226600]. (b) Abou-Donia, A. H.;
Abib, A. A.; El Din, A. S.; Evidente, A.; Gaber, M.; Scopa, A.
Phytochemistry 1992, 31, 2139. (c) Allayarov, Kh.; Abdusamatov, A.;
Yunusov, S. Yu. Khim. Prir. Soedin. 1970, 6, 143 [Chem. Abstr. 1970,
73, 69771]. (d) Normatov, M.; Abduazimov, Kh. A.; Yunusov, S. Yu.
Uzbeksk. Khim. Zh. 1965, 9, 25 [Chem. Abstr. 1965, 63, 7061f]. (e) See
also ref 4e.
Resu lts a n d Discu ssion
Using a modification of conditions first described by
Iwao and Kuraishi, 1-(tert-butoxycarbonyl)-5-methoxy-
indoline, 4, was lithiated at the C(7) position.9 The 7-lithio
indoline intermediate was subsequently transmetalated
with CuI-P(OEt)3 complex. The resulting organocopper
(6) (a) Chen, J .; Chen, K.; J iang, H.; Lin, M.; J i, R Prog. Nat. Sci.
1997, 7, 329. (b) He, H. M.; Weng, Z. Y. Yaoxue Xuebo 1989, 24, 302
[Chem. Abstr. 1989, 111, 108492]. (c) Wu, Y.; Wu, Y.; Yu, C.; Zhang,
S.; Su, Z.; J iang, S. Shanghai Yixue 1988, 11, 683 [Chem. Abstr. 1989,
111, 17263]. (d) Wang, X.; Yu, W.; Shen, Z.; Yang, J .; Xu, B. Zhongguo
Yaoli Xuebue 1987, 8, 86 [Chem. Abstr. 1987, 106, 95753]. (e) Huang,
C. C.; Han, C. S.; Yue, X. F.; Shen, C. M.; Wang, S. W.; Wu, F. G.; Xu,
B. J . Natl. Cancer Inst. 1983, 71, 841. Xu, B. Trends Pharmacol. Sci.
1981, 2, 271. (f) Furusawa, E.; Irie, H.; Combs, D.; Wildman, W. C.
Chemotherapy (Basel) 1980, 26, 28. (g) Zee-Cheng, R. K. Y.; Yan, S.-
J .; Cheng, C. C. J . Med. Chem. 1978, 21, 199.
(8) (a) Siddiqui, M. A.; Snieckus, V. Tetrahedron Lett. 1990, 31, 1523.
(b) Lauk, U.; Durst, D.; Fischer, W. Tetrahedron Lett. 1991, 32, 65.
(9) (a) Iwao, M.; Kuraishi, T. Org. Synth. 1996, 73, 85. (b) Parnes,
J . S.; Carter, D. S.; Kurz, L. J .; Flippin, L. A. J . Org. Chem. 1994, 59,
3497. (c) Iwao, M.; Takehara, H.; Obta, S.; Watanabe, M. Heterocycles
1994, 38, 1717. (d) Iwao, M.; Kuraishi, T. Heterocycles 1992, 34, 1031.
(7) (a) Xu, B.; Wang, X.-W. Drugs Future 1997, 22, 123. (b) Weng,
Z.-Y.; Wang, Z.-Y.; Yan, X.-M. Acta Pharm. Sin. 1982, 17, 744 [Chem.
Abstr. 1983, 98, 89710].
10.1021/jo991902p CCC: $19.00 © 2000 American Chemical Society
Published on Web 04/26/2000

3228 J . Org. Chem., Vol. 65, No. 10, 2000
Notes
Sch em e 1
Sch em e 2a
a
(i) NaCNBH3-AcOH; then BOC2 f 7 (60%); (ii) H2 (1 atm)/10% Pd-C f 8 (99%); (iii) TIPSCl-imidazole, DMF: 8 f 9 (73%) or
TBDMSCl-imidazole, DMF: 8 f 10 (94%).
slurry underwent the Ziegler-Ullmann coupling reac-
tion10 with 2-iodoaryl imine 5 to give, after mild hydro-
lytic workup, biarylaldehyde 6 in 60% yield. Removal of
the N-t-BOC group from 6 with a saturated ether solution
of anhydrous HCl also promoted concomitant cyclization
and dehydration to afford tortuosine, 1, in 81% yield
(Scheme 1).
and cyclization-dehydration proceeded rapidly from both
of these intermediates; however, removal of the silyl
group from the intermediate pyrrolophenanthridinium
salts, 14, was very slow, and prolonged exposure to the
reaction conditions eventually resulted in the formation
of uncharacterized side-products. Using this protocol,
yields of isolated criasbetaine hydrochloride, 2b, never
exceeded ca. 10%. In another approach, the TIPS group
was first removed from 11 (2 equiv of TBAF, THF, 23
°C, 30 min) to give a phenol followed by treatment with
saturated HCl in CHCl3; however, this sequence gave a
We anticipated that a similar approach using a suitable
5-hydroxyindoline synthon would lead to short syntheses
of criasbetaine, 2, and ungeremine, 3. In the event,
commercial 5-benzyloxyindole was reduced with Na-
CNBH3 to give the corresponding indoline which, without
isolation, was protected as a t-BOC carbamate, 7. The
O-benzyl group was removed from compound 7 using H2/
Pd-C in ethanol to give 1-(tert-butoxycarbonyl)-5-hy-
droxyindoline, 8, in 99% yield and the hydroxyl group of
compound 8 was efficiently protected either as a triiso-
propylsilyl (TIPS) ether, 9, or a tert-butyldimethylsilyl
(TBDMS) ether, 10. Initially we investigated the utility
of 9 and 10 for the synthesis of criasbetaine. For reasons
that are still unclear, the lithiation-transmetalation-
Ziegler coupling sequence using TIPS ether 9 afforded
biarylaldehyde 11 in very poor yield (8-16%) despite
several attempts to optimize it; however, TBDMS ether
10 behaved satisfactorily in the identical metalation-
coupling sequence, reliably giving biarylaldehyde 12 in
40-48% yield over several independent trials.
1
crude product mixture that appeared by H NMR analy-
sis to suffer extensive tert-butyl substitution on the C-ring
oxygen atom. Alternately, reaction of crude intermediate
14 (R1 ) TBDMS; R2 ) R3 ) CH3; X ) Cl: See Scheme
2) with 2 equiv of TBAF (THF, 23 °C, 16 h) gave <5%
yield of criasbetaine along with polar side-products that
no longer appeared to contain a C(6)-N(7) double bond
(1H NMR analysis). Finally, it was found that treatment
of TBDMS-protected intermediate 12 with refluxing
trifluoroacetic acid (neat, 16 h) promoted the desired
t-BOC removal, cyclization-dehydration, desilylation
sequence very cleanly to afford criasbetaine trifluoro-
acetate, 2c, in near-quantitative yield. Incorporation of
the appropriate A-ring functionality into our optimized
approach also provided a new and highly concise syn-
thesis11 of ungeremine trifluoroacetate, 3c (Scheme 2).
Synthetic tortuosine chloride, 1, and criasbetaine
hydrochloride, 2b, were submitted for quantitative as-
Several attempts to prepare criasbetaine hydrochloride
by allowing either 11 or 12 to react in CHCl3 saturated
with HCl proved inefficient. Removal of the t-BOC group
(11) Snieckus synthesis (ref 8a): five linear steps from 5-hydroxy-
indole, 17% overall yield. Lauk synthesis (ref 8b): eight linear steps
from 7-bromo-5-nitroindoline, 2% overall yield. Current method: five
linear steps from 5-benzyloxyindole, 35% overall yield.
(10) Ziegler, F. E.; Fowler, K. W.; Rodgers, W. B.; Wester, R. T. Org.
Synth. 1987, 65, 108 and references therein.

Notes
J . Org. Chem., Vol. 65, No. 10, 2000 3229
3.00 (m, 2H), 1.16 (s, 9H); 13C NMR δ 191.7, 157.1, 154.1, 153.2,
148.8, 140.7, 137.1, 135.5, 127.6, 126.2, 116.1, 112.0, 111.2, 108.8,
80.9, 56.6, 56.5, 56.2, 51.1, 30.3, 28.3; LSIMS m/z 414 ((M +
H)+, 22%), 314 (42%), 296 (100%). Anal. Calcd for C23H27NO6:
C, 66.81; H, 6.58; N, 3.39. Found: C, 66.61; H, 6.60; N, 3.60.
Tor tu osin e (1). Biarylaldehyde 6 (0.807 g; 1.95 mmol) was
dissolved in 5 mL of CHCl3. To this solution was added 16 mL
of a satd solution of anhydrous HCl in ether to give a copious
yellow precipitate. The precipitate was collected by suction
filtration and washed with ether to give 0.675 g of crude solid.
The crude material was separated by column chromatography
(9:1 CH2Cl2-MeOH) to give 0.516 g (80%) of tortuosine chloride,
1: mp 245-248 °C (dec) (lit. mp 242-243 °C); 1H NMR (DMSO-
d6) δ 9.69 (s, 1H), 8.27 (s, 1H), 8.10 (s, 1H), 7.92 (s, 1H), 7.49 (s,
1H), 5.25 (t, J ) 6.5 Hz, 2H), 4.19 (s, 3H), 4.07 (s, 3H), 4.00 (s,
3H), 3.68 (t, J ) 6.5 Hz, 2H); 13C NMR (DMSO-d6) δ 162.4, 157.0,
151.4, 142.3, 138.8, 131.6, 129.5, 124.4, 121.3, 116.7, 110.5, 104.4,
101.9, 57.7, 57.3, 56.6, 27.6; LSIMS m/z 296 (M+, 100%); HR-
sessment of growth inhibitory efficacy against 60 human
tumor cell lines at the National Cancer Institute.12 The
results of this in vitro study are summarized and
contrasted with comparable data provided by the NCI
for ungeremine hydrochloride, 3b, in the Supporting
Information. In consonance with their close structural
similarity, criasbetaine and ungeremine exhibit some-
what similar (although not identical) potencies and
selectivity profiles across the tumor cell-lines in which
they were tested in common. Although clear evidence of
selectivity against the CNS tumor-cell panel was found
for both 2b and 3b, neither betaine proved highly potent
against any of the cell-lines tested. Tortuosine chloride,
1, also exhibited general selectivity for the CNS tumor-
cell panel; however, this compound displayed relatively
high specific potency against the SF-268 glioblastoma
CNS cell-line: Com p ou n d n u m ber (pGI50): 2b (5.4),
3b (6.0), and 1 (7.8). The unusual selectivity and efficacy
of tortuosine chloride against SF-268 cells also compares
favorably with the maximum in vitro growth inhibition
results reported for ellipticinium-type13 and protober-
berine-type14 CNS-selective antitumor agents against this
cell line.
LSIMS m/z Calcd for
296.1291.
C
18H18NO3 (M+): 296.1287. Found:
1-(ter t-Bu toxyca r bon yl)-5-(ben zyloxy)in d olin e (7). To a
solution of 5-(benzyloxy)indole (15.0 g; 67 mmol) in 200 mL of
acetic acid was added sodium cyanoborohydride (12.7 g; 202
mmol), and the reaction mixture was stirred for 30 min at 23
°C. The reaction mixture was cooled with an ice bath, made basic
with aqueous NaOH, and extracted with ether to give 14.3 g of
a yellow oil. The crude oil was combined with di-tert-butyl
dicarbonate (18.3 g; 84 mmol) in 200 mL of THF, and the
reaction mixture was allowed to stir overnight at room temper-
ature. The reaction mixture was concentrated with a rotary
evaporator, and the crude product was recrystallized from
ether-pentane to afford 13.1 g (60%) of 7: mp 89.3-90.2 °C;
1H NMR (CDCl3) δ 7.8-7.65 (br m, 1H), 7.45-7.25 (m, 5H), 6.8-
6.7 (m, 2H), 5.02 (s, 2H), 3.96 (br t, J ) 8.6 Hz, 2H), 3.05 (t, J
) 8.6 Hz, 2H), 1.55 (br s, 9H); 13C NMR (CDCl3) δ 154.8, 152.6,
137.6, 136.9 (br), 132.7 (br), 128.6, 127.9, 127.5, 80.6 (br), 71.0,
47.9, 28.6, 27.6. Anal. Calcd for C20H23NO3: C, 73.82; H, 7.12,
N, 4.30. Found: C, 73.93; H, 7.09; N, 4.56.
A convincing mechanistic rationale for tortuosine’s
remarkable in vitro profile remains to be experimentally
determined; however, on the basis of its performance
against SF-268 cancer cells, tortuosine chloride has
recently been selected for preclinical in vivo antitumor
assay at the National Cancer Institute. The results of
these investigations will be reported in due course.
Exp er im en ta l Section
Gen er a l. Metalation reactions were performed under an
argon atmosphere using oven-dried glassware and freshly puri-
fied solvents. Melting points are uncorrected. 1H NMR (300
MHz) and 13C NMR (75 MHz) were recorded at room tempera-
ture. Column chromatography was performed using 230-400
mesh silica gel.
1-(ter t-Bu toxyca r bon yl)-5-(ter t-bu tyld im eth ylsilyloxy)-
in d olin e (10). BOC-protected indoline 7 (5.77 g; 17.7 mmol) and
10% Pd-C were mixed in 100 mL of absolute ethanol, and the
mixture was stirred under 1 atm of hydrogen overnight. The
reaction mixture was filtered through Celite and concentrated
under vacuum to give 3.75 g (90%) of highly pure 1-(tert-
butoxycarbonyl)-5-hydroxyindoline, 8.15 The BOC-protected hy-
droxyindoline (3.75 g; 15.9 mmol), tert-butyldimethylsilyl chlo-
ride (5.28 g; 35.0 mmol), and imidazole (2.71 g; 39.8 mmol) were
mixed in dry DMF, and the reaction mixture was stirred at 23
°C overnight. After an extractive (ether) workup, column chro-
matography (99:1 hexanes-EtOAc) afforded 5.22 g (94%) of 10:
mp 58.4-59.5 °C; 1H NMR (CDCl3) δ 7.75-7.6 (br m, 1H), 6.64
(br d, J ) 0.8 Hz, 1H), 6.61 (br d, J ) 0.8 Hz, 1H), 3.95 (br t, J
) 8.7 Hz, 2H), 3.02 (t, J ) 8.7 Hz, 2H), 1.55 (br s, 9H), 0.97 (s,
9H), 0.16 (s, 6H); 13C NMR (CDCl3) δ 152.6; 150.9; 118.4; 116.8;
115.0, 47.7, 28.5, 27.5, 25.7, 18.2, -4.5; EIMS m/z 349 (M+, 13%),
ter t-Bu tyl 5-Meth oxy-7-(2-for m yl-4,5-d im eth oxyp h en yl)-
2,3-d ih yd r oin d ole-1-ca r boxyla te (6). 5-Methoxy-N-BOC-in-
doline, 4,9d (1.62 g; 6.5 mmol) and 2 mL of TMEDA were
dissolved in 40 mL of ether at -45 °C (internal thermometer).
sec-BuLi (10 mL of a 1.3 M solution in cyclohexane) was added
dropwise. The reaction mixture was stirred at -45 to -50 °C
for 2 h, and CuI-P(EtO)3 complex (4.64 g; 13.0 mmol) was added
in one portion. The resulting orange slurry was stirred at -45
°C for 0.5 h, and a solution of imine 5a (2.43 g; 6.5 mmol) in 5
mL of THF was added in one portion. The reaction mixture was
allowed to warm to room temperature (orange slurry f greenish
black solution) and stirred overnight. The solution was acidified
with dilute aqueous HCl to pH ) 4, and the mixture was refluxed
for 10 min. The mixture was extracted with ether, the combined
ether layers were dried (MgSO4) and concentrated with a rotary
evaporator to give a yellow oil. Chromatography of the crude
product (3:1 hexanes-EtOAc) gave 1.61 g (60%) of 6: mp 142.4-
294 (22%), 293 (100%); HR-EIMS Calcd for
349.2073. Found: 349.2073.
C19H31NO3Si:
1-(ter t-Bu t oxyca r b on yl)-5-(t r iisop r op ylsilyloxy)in d o-
lin e (9). Protection of compound 8 using the general procedure
with TIPSCl gave 73% of 9 as a colorless oil: 1H NMR (CDCl3)
δ 7.75-7.6 (br m, 1H), 6.7-6.6 (br m, 2H), 3.95 (br t, J ) 8.3
Hz, 2H), 3.05 (t, J ) 8.3 Hz, 2H), 1.55 (br s, 9H), 1.3-1.15 (m,
3H), 1.09 (d, J ) 6.8 Hz, 18 H); 13C NMR (CDCl3) δ 152.5, 151.4,
136.5 (br s), 132.2 (br s), 118.0, 116.5, 114.9, 80.4 (br s), 47.7,
28.5, 27.3, 17.9, 12.6. HR-EIMS Calcd for C22H37NO3Si: 391.2543.
Found: 391.2542.
ter t-Bu tyl 5-(ter t-Bu tyld im eth ylsilyloxy)-7-(2-for m yl-4,5-
d im et h oxyp h en yl)-2,3-d ih yd r oin d ole-1-ca r b oxyla t e (12).
Indoline 10 (1.00 g; 2.9 mmol) and 0.86 mL of TMEDA were
dissolved in 15 mL of dry THF. At -40 °C (internal thermom-
eter) sec-Buli (4.4 mL of a 1.3 M solution in cyclohexane) was
added dropwise to the reaction mixture. The mixture was
allowed to stir for 2 h at -40 to -50 °C, and CuI-P(OEt)3
1
143.3 °C; H NMR (CDCl3) δ 9.78 (s, 1H), 7.47 (s, 1H), 6.91 (s,
1H), 6.84 (d, J ) 2.6 Hz, 1H), 6.64 (d, J ) 2.6 Hz, 1H), 4.13-
4.04 (m, 2H), 3.958 (s, 3H), 3.955 (s, 3H), 3.80 (s, 3H), 3.07-
(12) Monks, A.; Scudiero, D.; Skehan, P.; Shoemaker, R.; Paull, K.;
Vistica, D.; Hose, C.; Langley, J .; Cronise, P.; Vaigro-Wolff, A.; Gray-
Goodrich, M.; Campbell, H.; Mayo, J .; Boyd, M. R. J . Natl. Cancer Inst.
1991, 83, 757.
(13) (a) J urayj, J .; Haugwitz, R. D.; Varma, R. K.; Paull, K. D.;
Barrett, J . F.; Cushman, M. J . Med. Chem. 1994, 37, 2190. (b) Acton,
E.; Narayanan, V. L.; Risbood, P. A.; Shoemaker, R. H.; Vistica, D. T.;
Boyd, M. R. J . Med. Chem. 1994, 37, 2185. (c) Anderson, W.;
Gopalsamy, A.; Reddy, P. S. J . Med. Chem. 1994, 37, 1955.
(14) (a) Sanders, M. M.; Liu, A. A.; Li, T.-K.; Wu, H.-Y.; Desai, S.
D.; Mao, Y.; Rubin, E. H.; LaVoie, E. J .; Makhey, D.; Liu, L. F. Biochem.
Pharmacol. 1998, 56, 1157. (b) For a recent review see Gatto, B.;
Capranico, G.; Palumbo, M. Curr. Pharm. Des. 1999, 5, 195.
(15) Padwa, A.; Dimitroff, M.; Waterson, A. G.; Wu, T. J . Org. Chem.
1998, 63, 3986.

3230 J . Org. Chem., Vol. 65, No. 10, 2000
Notes
complex (2.04 g; 5.72 mmol) was added in one portion to give a
clear orange solution. The solution was stirred for 0.5 h at -40
°C, and a solution of imine 5a 10 (1.07 g; 2.9 mmol) in 5 mL of
THF was added in one portion. The reaction mixture was allowed
to warm to room temperature over about 1 h, and the resulting
green-black solution was allowed to stir at room-temperature
overnight. The reaction mixture was quenched with aqueous
ammonium chloride and extracted with ether. The ether layer
was acidified to pH ) 4 with aqueous HCl, refluxed for 5 min,
washed with satd NaCl, dried (MgSO4), and concentrated under
vacuum to give a yellow, viscous oil. Chromatographic purifica-
tion of the crude product (85:15 hexane EtOAc) gave 0.72 g (48%)
of 12: mp 119.8-121.5 °C; 1H NMR (CDCl3) δ 9.78 (s, 1H), 7.47
(s, 1H), 6.88 (s, 1H), 6.75 (d, J ) 2.5 Hz, 1H), 6.58 (d, J ) 2.5
Hz, 1H), 4.11-4.04 (m, 2H), 3.953 (s, 3H), 3.951 (s, 3H), 3.00 (t,
J ) 7.7 Hz, 2H), 1.16 (s, 9H), 0.98 (s, 9H), 0.20 (s, 6H); 13C NMR
(CDCl3) δ 191.5, 153.8, 153.0, 152.6, 148.4, 140.6, 136.7, 135.6,
127.2, 126.0, 122.1, 116.7, 111.6, 108.5, 80.7, 60.5, 56.34, 56.29,
50.9, 29.9, 29.1, 25.8, 18.4, -4.2; EIMS m/z 513 (M+, 26%), 413
(57%), 396 (100%); HR-LSIMS Calcd for C28H39NO6Si: 513.2547.
Found: 513.2545.
ter t-Bu tyl 5-(ter t-Bu tyld im eth ylsilyloxy)-7-(2-for m yl-4,5-
(m et h ylen ed ioxy)p h en yl)-2,3-d ih yd r oin d ole-1-ca r b oxy-
la te (13). Indoline 10 (1.05 g; 3.0 mmol) and 0.9 mL of TMEDA
were dissolved in 15 mL of THF at -40 °C. sec-BuLi (4.4 mL of
a 1.3 M solution in cyclohexane) was added to the reaction
mixture over 5 min. The reaction mixture was maintained
between -40 to -50 °C for 2 h, and CuI-P(OEt)3 complex (2.14
g; 6.0 mmol) was added in one portion to give a clear orange
solution. The reaction mixture was stirred for 0.5 h at -40 °C,
and a solution of imine 5b10 (1.07 g; 3.0 mmol) in 5 mL of THF
was added in one portion. The reaction mixture was allowed to
warm to room temperature, stirred overnight, and worked up
as above to give a crude oil. Flash chromatography of the crude
material (9:1 hexanes-EtOAc) afforded 1.20 g (80%) of 13.
Recrystallization of the chromatographed product from hexane
gave 1.04 g (69%) of analytically pure 13: mp 152.3-153.3 °C;
1H NMR (DMSO-d6) δ 9.62 (s, 1H), 7.26 (s, 1H), 6.92 (s, 1H),
6.86 (d, J ) 2.3 Hz, 1H), 6.51 (d, J ) 2.3 Hz, 1H), 6.16 (d, J )
9.7 Hz, 2H), 4.05-3.90 (m, 2H), 3.05-2.95 (m, 2H), 1.13 (s, 9H),
0.95 (s, 9H), 0.18 (s, 6H); 13C NMR (DMSO-d6) δ 190.2, 152.3,
152.0, 147.3, 142.2, 137.4, 135.5, 127.2, 126.9, 121.4, 116.9, 109.6,
105.3, 102.5, 80.1, 50.8, 29.5, 27.9, 25.9, 18.3, -4.2. Anal. Calcd
for C27H35NO6Si: C, 65.16; H, 7.09; N, 2.81. Found: C, 65.21;
H, 7.15; N, 2.92.
Cr ia sbeta in e Tr iflu or oa ceta te (2c). Biarylaldehyde 12
(0.335 g; 0.65 mmol) was refluxed in 5 mL of trifluoroacetic acid
for 16 h. The reaction mixture was concentrated with a rotary
evaporator, and the residue was triturated with ether. The crude
residue was purified by column chromatography (85:15 CH2Cl2-
MeOH) to afford 0.265 g (99%) of criasbetaine trifluoroacetate
monohydrate 2c: mp 238-240 °C; 1H NMR (DMSO-d6) δ 10.95
(s, 1H), 9.56 (s, 1H), 8.14 (s, 1H), 7.98 (s, 1H), 7.90 (s, 1H), 7.40
(s, 1H), 5.21 (t, J ) 6.8 Hz, 2H), 4.16 (s, 3H), 4.01 (s, 3H), 3.68
(t, J ) 6.8 Hz, 2H); 13C NMR (DMSO-d6) δ 161.1, 156.8, 151.4,
141.5, 139.0, 130.8, 129.1, 124.8, 121.3, 116.6, 110.5, 104.3, 104.2,
57.3, 56.5, 56.1, 27.7; 19F NMR (DMSO-d6 solution; external
reference: CCl3F) δ -73.91; LSIMS m/z 282 (M + H)+, 100%);
HR-LSIMS Calcd for C17H16NO3 (M + H)+: 282.1130. Found:
282.1130. Anal. Calcd for C17H15NO3 + CF3CO2H + H2O: C,
55.21; H, 4.39; N, 3.39. Found: C, 54.94; H, 3.86; N, 3.52.
Un ger em in e Tr iflu or oa ceta te (3c). Biarylaldehyde 13
(0.200 g; 0.4 mmol) was added to 5 mL of trifluoroacetic acid,
and the solution was refluxed for 14 h. The reaction mixture
was concentrated with a rotary evaporator, and the residue was
triturated with ether to give 137 mg of a green powder.
Recrystallization of the crude solid from methanol gave pale
yellow crystalline 3c (100 mg; 66% yield): mp >250 °C; 1H NMR
(DMSO-d6) δ 11.08 (br s, 1H), 9.54 (s, 1H), 8.25 (s, 1H), 7.86 (s,
1H), 7.79 (d, J ) 1 Hz, 1H), 7.38 (d, J ) 1 Hz, 1H), 6.44 (s, 2H),
5.19 (br t, J ) 6.5 Hz, 2H), 3.67 (br t, J ) 6.5 Hz, 2H); 13C NMR
(DMSO-d6) δ 161.3, 155.7, 150.3, 141.6, 139.0, 131.5, 130.9,
125.1, 122.7, 117.1, 107.5, 104.3, 104.0, 101.8, 56.1, 27.7; 19F
NMR (DMSO-d6 solution; external reference: CCl3F) δ -73.93.
Anal. Calcd for
C18H12F3NO5: C, 57.00; H, 3.19; N, 3.69.
Found: C, 56.77; H, 3.11; N, 3.83.
Ack n ow led gm en t. We thank the National Cancer
Institute for the in vitro antitumor screening of com-
pounds 1 and 2b, and for generously supplying compa-
rable data for 3b.
Su p p or t in g In for m a t ion Ava ila b le: 1H and 13C NMR
spectra of all new compounds and comprehensive screening
results for 1, 2b, and 3b in the NCI human tumor cell-line
panel. This material is available free of charge via the Internet
at http://pubs.acs.org.
J O991902P