Design and synthesis of a taxoid library using radiofrequency encoded combinatorial chemistry

Article abstract of DOI:10.1021/jo9708699

Radiofrequency encoded combinatorial (REC) chemistry is a recently developed nonchemical encoding strategy in library synthesis. Encoded chemical libraries of complex molecular structures like Taxol can be constructed employing the noninvasive REC strategy and novel solid phase synthesis techniques, as demonstrated by the synthesis of the first 400- membered taxoid library in a discrete format and in quantities of multimilligrams/member.

Full text of DOI:10.1021/jo9708699

J . Org. Chem. 1997, 62, 6029-6033
6029
Design a n d Syn th esis of a Ta xoid Libr a r y Usin g Ra d iofr equ en cy
En cod ed Com bin a tor ia l Ch em istr y
Xiao-Yi Xiao,* Zahra Parandoosh, and Michael P. Nova
IRORI Quantum Microchemistry, 11025 North Torrey Pines Road, La J olla, California 92037
Received May 15, 1997X
Radiofrequency encoded combinatorial (REC) chemistry is a recently developed nonchemical
encoding strategy in library synthesis. Encoded chemical libraries of complex molecular structures
like Taxol can be constructed employing the noninvasive REC strategy and novel solid phase
synthesis techniques, as demonstrated by the synthesis of the first 400-membered taxoid library
in a discrete format and in quantities of multimilligrams/member.
Taxol1-3 is a newly approved anticancer agent which
is particularly effective against ovarian and breast
cancer. Its unique mechanism of action4 involves promo-
tion of tubulin polymerization and stablization of micro-
tubules, resulting in cell death. However, difficulties
related to formulation and multiple drug resistance
(MDR) limit the application of Taxol and its close analog,
Taxotere, in cancer treatments.5
Modification of natural product scaffolds is an attrac-
tive strategy toward discovering new and better phar-
maceutical agents.6 The advent of combinatorial chem-
istry7 promises to revolutionize traditional drug discovery
process by greatly accelerating lead identification and
optimization. Despite extensive analog studies in the
taxoid field,3 application of this new technique to the
Taxol template has the potential of uncovering new
taxoids with improved pharmaceutical properties. In
search of such molecules, we applied the recently devel-
oped radiofrequency encoded combinatorial (REC) chem-
istry strategy8 to the design and synthesis of a taxoid
library of general formula 1 (Scheme 1). The criteria
used in the molecular design of this library included the
following: (a) potential enhancement of water solubility
(by incorporation of amide, carboxylic acid, and other
polar groups); (b) possible modulation of biological activ-
ity (by variation at C-7, C-2′, and C-3′ with a wide range
of substituents); and, (c) novel solid phase synthesis. The
utilization of microreactors8a as the “split & pool” units,
rather than the resin beads used in conventional “split
& pool” combinatorial synthesis,9 leads to libraries of
discrete compounds in multimilligrams/member quanti-
ties without compromising synthesis efficiency as can be
the case using parallel synthesis.
Resu lts a n d Discu ssion
Atta ch m en t of th e Ta xol Tem p la te. As a starting
point, the first taxoid library in our plan is the combi-
natorialization of the Taxol template through the C-2′
and C-7 hydroxy groups and the C-3′ amino group by
ester and amide formations. This would not only take
advantage of the availability of a vast variety of carboxy-
lic acids as economic and diverse building blocks, but also
produce high quality library compounds through well
established acylation chemistries. The Initial attempts
to link the Taxol template via the C-7 or C-2′ hydroxy
groups using the THP ether linker10 or the related
alkoxyethyl ether linker11 were unsatisfactory due to the
high steric hindrance at these sites. A glutamic acid
“handle” was then incorporated into the molecule (1,
Scheme 1), which would provide a readily accessible
anchoring point (carboxy group), a well-behaving diver-
sity site (R-amino group), and possible solubility enhance-
ment after cleavage (free carboxylic acid). Even though
there has been extensive SAR studies on the modification
of the Taxol core at various positions,3 how the incorpora-
tion of this glutamic acid “handle”, coupled with modi-
fications at C-7 and C-2′, will actually affect Taxol’s
biological activities remains unknown or mere specula-
tion until the library is synthesized and evaluated.
*To whom correspondence should be addressed. Tel: 619-546-
1300; fax: 619-546-3083; e-mail: xyxiao@irori.com.
X Abstract published in Advance ACS Abstracts, August 1, 1997.
(1) Isolation and structure: Wani, M. C.; Taylor, H. L.; Wall, M. E.;
Coggon, P.; McPhail, A. T. J . Am. Chem. Soc. 1971, 93, 2325-2327.
(2) Total Synthesis: (a) Nicolaou, K. C.; Yang, Z.; Liu, J . J .; Ueno,
H.; Nantermet, P. G.; Guy, R. K.; Claiborne, C. F.; Renaud, J .;
Couladouros, E. A.; Paulvannan, K.; Sorensen, E. J . Nature 1994, 367,
630-634. (b) Hotton, R. A.; Somoza, C.; Kim, H.-B.; Liang, F.; Biediger,
R. J .; Boatman, P. D.; Shindo, M.; Smith, C. C.; Kim, S.; Nadizadeh,
H.; Suzuki, Y.; Tao, C.; Vu, P.; Tang, S.; Zhang, P.; Murthi, K. K.;
Gentile, L. N.; Liu, J . H. J . Am. Chem. Soc. 1994, 116, 1597-1600. (c)
Masters, J . J .; Link, J . T.; Snyder, L. B.; Young, W. B.; Danishafsky,
S. J . Angew. Chem., Int. Ed. Engl. 1995, 34, 1723-1726.
(3) Reviews: (a) Nicolaou, K. C.; Guy, R. K.; Potier, P. Sci. Am. 1996,
J une, 94-98. (b) Nicolaou, K. C.; Dai, W.-M; Guy, R. K. Angew. Chem.,
Int. Ed. Engl. 1994, 33, 15-44. (c) Kingston, D. G. I.; Molinero, A. A.;
Rimoldo, J . M. In Progress in Chemistry of Organic Natural Products;
Herz, W., Kirby, G. W., Moore, R. E., Steglich, W., Tamm, C., Eds;
Springer: New York, 1993; Vol. 61, pp 1-206.
(4) (a) Schiff, P. B.; Fant, J .; Horwitz, S. B. Nature 1979, 277, 665-
667. (b) Schiff, P. B.; Horwitz, S. B. Proc. Nat. Acad. Sci., U.S.A. 1980,
77, 1561-1565.
(5) Landino, L. M.; Macdonald T. L. In The Chemistry and Phar-
macology of TaxolTM and its Derivatives; Farin, V., Ed.; Elsevier: New
York, 1995; Chapter 7.
Syn th esis of th e Cor e Str u ctu r e 4. The projected
solid phase synthesis of the targeted library 1 required
the preparation of advanced intermediate 4 (Scheme 2)
from commercial baccatin III.12 Among the reported side-
chain coupling strategies, the oxazolidine approach13 and
(6) (a) Silverman, R. B. The Organic Chemistry of Drug Design and
Drug Action; Academic Press, Inc.: San Diego, 1992. (b) Lavelle, F.;
Gue´ritte-Voegelein, F.; Gue´nard, D. Bull. Cancer 1993, 80, 326-331.
(c) Klein, L. L. Tetrahedron Lett. 1993, 34, 2047-2049.
(7) (a) Balkenhohl, F.; Bussche-Hu¨nnefeld, C.; Lansky, A.; Zechel,
C. Angew. Chem., Int. Ed. Engl. 1996, 35, 2288-2337. (b) Gordon, E.;
Barret, R. W.; Dower, W. J .; Fodor, S. P. A.; Gallop, M. A. J . Med.
Chem. 1994, 37, 1385-1401.
(8) (a) Nicolaou, K. C.; Xiao, X.-Y.; Parandoosh, Z.; Senyei, A.; Nova,
M. P. Angew. Chem., Int. Ed. Engl. 1995, 34, 2289-2291. (b) Moran,
E. J .; Sarshar, S.; Cargill, J . F.; Shahbaz, M. M.; Lio, A.; Mjalli, A. M.
M.; Armstrong, R. W. J . Am. Chem. Soc. 1995, 117, 10787-10788.
(9) Lam, K. S.; Salmon, S. E.; Hersh, E. M.; Hruby, V. J .; Kazmier-
ski, W. M.; Knapp, R. J . Nature 1991, 354, 82-84.
(10) Thompson, L. A.; Ellman, J . A. Tetrahedron Lett. 1994, 35,
9333-9336.
(11) Wang, G. T.; Li, S.; Wideburg, N.; Krafft, G. A.; Kempf, D. J . J .
Med. Chem. 1995, 38, 2995-3002.
S0022-3263(97)00869-4 CCC: $14.00 © 1997 American Chemical Society

6030 J . Org. Chem., Vol. 62, No. 17, 1997
Xiao et al.
Sch em e 1a
Sch em e 2a
aBuilding blocks used in the taxoid library 1: 20 × 20 ) 400
members. Letters beneath building blocks are codes used in
radiofrequency encoding during library synthesis.
the â-lactam approach14 seem to be promising to yield
the desired amino alcohol 2, and the first approach
proved to be fairly successful in our hands. To this end,
baccatin III was protected at C-7 as a Troc carbonate,15
coupled with (4S,5R)-N-Boc-2,2-dimethyl-4-phenyl-5-ox-
azolidinecarboxylic acid14,16 under DCC/4-DMAP condi-
tions, and deprotected with formic acid to yield the amino
alcohol 2 in 72% overall yield. Subsequent coupling
(PyBOP/DIEA) with N-Fmoc-O-(trichloroethyl)-L-glu-
tamic acid, which was conveniently prepared from N-
Fmoc-L-glutamic acid tert-butyl ester using standard
procedures, gave the trichloroethyl ester 3 in 77% isolated
yield. Finally, concomitant removal of the Troc group
and the trichloroethyl ester group with Zn dust furnished
compound 4 in 89% yield.
a(a) Performed according to procedures in references 14-16,
three steps, 72% overall; (b) N-Fmoc-O-(trichloroethyl)glutamic
acid, PyBOP, DIEA, CH2Cl2, rt, 4 h, 77%; (c) AcOH, MeOH, Zn,
60 °C, 2 h, 89%.
4 was reacted with excess 2-chlorotrityl resin (10 mequiv)
via standard trityl ester formation to afford the polymer
bound compound 5 (Scheme 3). The loading reaction was
conveniently monitored by the disappearance of 4 from
the reaction solution using classical solution phase
monitoring methods (TLC, HPLC, etc.), and excess trityl
chloride was quenched by the addition of methanol.
The loaded resin 5 was then distributed into 400
microreactors8a (20 mg each), and the microreactors were
subjected to the following reactions and manipulations.
First, the microreactors were treated with 5% piperidine/
DMF to remove the Fmoc group, yielding amine 6. The
average loading was 2.9 µmol/microreactor as determined
by UV measurement of the released 9-methylidenefluo-
rene. The microreactors were then split into 20 equal
pools. The reactors in each pool were encoded with a
unique rf code (Scheme 1), and each pool was coupled
with the corresponding carboxylic acid (R1, 20 total)
under PyBOP/DIEA/DMF conditions to form the amides
7. The microreactors were re-split into 20 new pools:
each new pool was formed by combining one microreactor
from each previous pool.18 The new pools were radiof-
requency-encoded accordingly. One pool was set aside
(for R2 ) H) and the remaining 19 pools were treated
with the second set of carboxylic acids (R2, 19 total) under
DIC/DMAP/CH2Cl2 conditions, leading to the resin-bound
taxoids 8. The microreactors were then decoded by
Libr a r y Syn th esis. In a standard resin loading
procedure, the reagents in solution are usually in large
excess in order to drive the loading reaction to comple-
tion. Since our Taxol template 4 was derived from the
rather expensive and precious baccatin III, a reversed (or
“resin trapping”17) loading method was devised to im-
mobilize the template onto solid supports, in which the
solid supports are in excess so that complete consumption
of the solution reagents can be achieved. Thus, template
(12) The baccatin III used in this study was from Hauser Labora-
tories, Chemistry Services, Boulder, CO.
(13) Commerc¸on, A.; Be´zard, D.; Bernard, F.; Bourzat, J . D. Tetra-
hedron Lett. 1992, 33, 5185-5188.
(14) (a) Holton, R. A. Eur. Pat. Appl. 1990, EP 400,971. (b) Ojima,
I.; Sun, C. M.; Zucco, M.; Park, Y. H.; Duclos, O.; Kuduk, S. Tetrahedron
Lett. 1993, 34, 4149-4152.
(15) Magri, N. F.; Kingston, D. G. I.; J itrangsri, C.; Piccariello, T.
J . Org. Chem. 1986, 51, 3239-3242.
(16) Denis, J .-N.; Correa, A.; Greene, A. E. J . Org. Chem. 1991, 56,
6939-6942.
(17) During the course of preparing this manuscript, a similar
strategy was applied to remove excess reagents in solution phase
syntheses: Kaldor, S. W. Presented at 25th National Medicinal
Chemistry Symposium of the American Chemical Society, Ann Arbor,
MI, J une, 1996.

Design and Synthesis of a Taxoid Library
J . Org. Chem., Vol. 62, No. 17, 1997 6031
Sch em e 3a
reading the rf codes, distributed into 400 glass vials each
labeled with the corresponding rf code, and treated with
AcOH/CF3CH2OH/CH2Cl2 (2:1:7, v/v/v) to cleave the final
products from the resin in each microreactor, leading to
the desired 400-member taxoid library 1.
The quantity obtained for each compound in the library
ranged between 2 to 4 mg, and the purity between 50%
to 100% as judged by TLC and HPLC analysis. Confir-
mation of the structures of 20 randomly-selected mem-
1
bers by electrospray MS (Table 1) and H NMR spectros-
copy indicated that this synthesis yielded the desired
taxoids in the library. Part of the library was purified
with HPLC for the purpose of comparing the biological
activities of the crude and purified samples.
Con clu sion . We have successfully constructed the
first 400-member taxoid library 1 by employing the
radiofrequency encoded combinatorial (REC) strategy
and novel solid phase synthesis techniques. This has
demonstrated that using these techniques, complex mo-
lecular structures like Taxol can be manipulated on solid
supports, and combinatorial libraries of such molecules
can be built in a discrete format and in multimilligrams/
member quantities. Biological studies of this library and
other Taxol derivative libraries are currently in progress
and will be reported in due time.
Exp er im en ta l Section
Gen er a l P r oced u r es. All moisture sensitive reactions
were performed in sealed glass vessels under nitrogen. Re-
agents and solvents were purchased from either Aldrich or
Fisher. Hygroscopic solvents were dried according to standard
laboratory procedures prior to use. Chlorotrityl resin (100-
200 mesh, 1.05 mmol/g) was purchased from Nova Biochem.
All reagents and nonhygroscopic solvents were used as re-
ceived without further treatment. Thin layer chromatography
(TLC) was performed on EM silica gel 60 F254 glass plates.
Flash column chromatography was performed with EM silica
gel 60 (60-120 mesh). Abbreviations: PyBOP, (benzotriazol-
1-yloxy)-tripyrrolidinophosphonium hexafluorophsophate; DIC,
1,3-diisopropylcarbodiimide; TEA, triethylamine; DIEA, diiso-
propylethylamine; Troc, (2,2,2-trichloroethoxy)carbonyl; rf,
radiofrequency.
7-Tr oc-ba cca tin III. Baccatin III12 (1.70 g, 2.90 mmol) was
dissolved in anhydrous pyridine (34 mL). Troc-Cl (0.80 mL,
5.81 mmol, 3 equiv) was added. The reaction mixture was
stirred at 80 °C for 1 h, cooled to room temperature, diluted
with EtOAc (200 mL), washed with water (3 × 100 mL), 10%
aqueous CuSO4 (50 mL), water (2 × 100 mL), and brine (50
mL), and dried over MgSO4. The MgSO4 was filtered off, and
the solution was concentrated under vacuum. Flash column
chromatography of the residue on silica gel with 20-40%
EtOAc/petroleum ether yielded a white solid (1.71 g, 77%): Rf
a(a) Excess 2-chlorotrityl resin (8.0 g, 10 m equiv), DIEA,
CH2Cl2, rt, 3 h, then MeOH, rt, 0.5 h; (b) resin was distributed
into 400 microreactors; (c) 5% piperidine in DMF, rt, 0.5 h; (d)
microreactors were split into 20 equal pools, and each pool was
treated with carboxylic acid (R1), DIEA, PyBOP, DMF, rt, 4 h; (e)
microreactors were resplit into 20 new pools (see main text), and
each pool was treated with carboxylic acids (R2), DIC, DMAP,
CH2Cl2, rt, 48 h; (f) microreactors were decoded and distributed
into 400 glass vials; (g) AcOH, CH2Cl2, CF3CH2OH, rt, 4.
1
) 0.75 (silica, 80% EtOAc in petroleum ether); H NMR (500
MHz, CDCl3) δ 1.09 (s, 3 H), 1.14 (s, 3 H), 1.82 (s, 3 H), 1.97-
2.10 (m, 2 H), 2.13 (s, 3 H), 2.17 (s, 3H), 2.20-2.29 (m, 2 H),
2.30 (s, 3H), 2.60-2.68 (m, 1 H), 4.02 (d, J ) 6.9 Hz, 1 H),
4.15 (d, J ) 8.5 Hz, 1 H), 4.33 (d, J ) 8.5 Hz, 1 H), 4.65 (d, J
) 12.0 Hz, 1 H), 4.88 (br t, J ) 8.0 Hz, 1 H), 5.00 (d, J ) 9.1
Hz, 1 H), 5.04 (d, J ) 12.0 Hz, 1 H), 5.58-5.65 (m, 2 H), 6.39
(s, 1 H), 7.47-7.52 (m, 2 H), 7.62 (t, J )7.4, 1 H), 8.10 (d, J )
7.8 Hz, 2 H); LRMS (electrospray) m/ z [M + Na]+ 785, calcd
for C34H39Cl3O13Na 785.
7-Tr oc-13-O-(N,O-eth yleid en e-â-p h en ylisoser in yl)ba c-
ca tin III. The preceding compound (1.71 g, 2.24 mmol) was
dissolved in 70 mL toluene. (4S,5R)-N-Boc-2,2-dimethyl-4-
phenyl-5-oxazolidinecarboxylic acid14,16 (1.08 g, 3.36 mmol, 1.5
equiv), 4-DMAP (137 mg, 1.12 mmol, 0.5 eq.), and DCC (0.74
g, 3.59 mmol, 1.6 equiv) were added sequentially. The reaction
mixture was heated at 80 °C for 2 h, cooled to room temper-
ature, diluted with EtOAc/diethyl ether (1:1, 200 mL), and
washed with 10% CuSO4 (2 × 80 mL), water (2 × 80 mL), and
brine (80 mL). The solution was dried over MgSO4, filtered,
and concentrated under vacuum. The crude product was
purified by column chromatography on silica gel with 15-30%
(18) The split of microreactors for the second step is a nonstatistical
redistribution since the ratio between the number of reactors and the
library size is 1:1 (i.e. zero redundancy), and statistical split cannot
be used. Apparently, this method is applicable only in the second step
of a synthesis and only when the reactors are not pooled together after
the first step. A similar split strategy, called directed sorting, which
does not have these limitations, has been recently reported by us: Xiao,
X. Y.; Zhao, C.; Potash, H.; Nova, M. P. Angew. Chem., Int. Ed. Engl.
1997, 36, 780-782.

6032 J . Org. Chem., Vol. 62, No. 17, 1997
Xiao et al.
Ta ble 1. Str u ctu r es a n d An a lytica l Da ta of 20 Mem ber s
fr om th e 400-Mem ber ed Ta xoid Libr a r y 1
concentrated at room temperature under vacuum. The residue
was redissolved in EtOAc (200 mL), washed with saturated
NaHCO3 (2 × 50 mL) and brine (50 mL), dried over MgSO4,
and concentrated under vacuum to yield 1.81 g of the amino
alcohol 2 as a white solid (quantitative): Rf ) 0.42, 5% MeOH
in CH2Cl2.
N-Fmoc-O-(trichloroethyl)glutamic acid (1.25 g, 2.50 mmol,
1.2 equiv), prepared from N-Fmoc-glutamic acid tert-butyl ester
using standard procedures, was dissolved in 50 mL of CH2Cl2.
DIEA (0.55 mL, 3.2 mmol, 1.5 equiv) and PyBOP (1.19 g, 2.29
mmol, 1.1 equiv) were added sequentially. The colorless
solution was stirred at room temperature for 20 min, followed
by addition of amino alcohol 2 (1.93 g, 2.08 mmol, in 100 mL
of CH2Cl2). The reaction mixture was stirred at room tem-
perature for 4 h, diluted with EtOAc (500 mL), and washed
with saturated phosphate buffer (pH ) 4, 3 × 200 mL),
saturated NaHCO3 (3 × 200 mL), and brine (200 mL). The
solution was dried over MgSO4 and concentrated under
vacuum. The residue was flash chromatographed on silica gel
with 0-10% EtOAc/CH2Cl2 to yield the desired ester 3 as a
white solid (2.25 g, 77%): Rf ) 0.55, 5% MeOH in CH2Cl2; 1H
NMR (500 MHz, CDCl3) δ 1.18 (s, 3 H), 1.25 (s, 3 H), 1.84 (s,
3 H), 1.86 (s, 3 H), 2.17 (s, 3 H), 2.30 (d, J ) 8,7 Hz, 2 H), 2.38
(s, 3 H), 1.5-2,8 (m, 9 H), 3.92 (d, J ) 6.8 Hz, 1 H), 4.17-4.30
(m, 3 H), 4.31 (d, J ) 8.6 Hz, 1 H), 4.31-4.45 (m, 2 H), 4.63
(d, J ) 12.0 Hz, 1 H), 4.68 (br d, J ) 19.6 Hz, 2 H), 4.94 (d, J
) 9.0 Hz, 1 H), 5.01 (d, J ) 12.0 Hz, 1 H), 5.45-5.55 (m, 2 H),
5.69 (d, J ) 6.8 Hz, 1 H), 6.20-6.28 (m, 1 H), 6.34 (s, 1 H),
7.05-7.60 (m, 16 H), 7.75 (d, J ) 7.4 Hz, 1 H), 8.11 (d, J ) 7.6
Hz, 1 H); LRMS (electrospray) m/ z [M + H]+ 1405, calcd for
C65H67Cl6N2O20 1405.
13-[N-(N-F m oc-glu t a m yl)-â-p h en ylisoser in yl]b a cca -
tin III (4). Trichloroethyl ester 3 (2.23 g, 1.58 mmol) was
dissolved in MeOH/AcOH (4:3, v/v, 87 mL). Activated zinc
powder (4.2 g, large excess) was added carefully. The reaction
mixture was vigorously stirred at 60 °C for 2 h, cooled to room
temperature, and filtered through a glass wool pad. The clear
solution was concentrated under vacuum, and the residue was
redissolved in a mixture of EtOAc and water (1:1, 500 mL).
The two layers were separated, and the aqueous layer was
extracted with EtOAc (3 × 50 mL). The organic solutions were
combined, dried over MgSO4, and concentrated under vacuum.
Flash column chromatography on silica gel with 5% MeOH/
CH2Cl2 to 0.1% AcOH/5% MeOH/CH2Cl2 yielded the desired
carboxylic acid 4 as a white solid (1.53 g, 89%): Rf ) 0.40, 5%
MeOH and 0.1% AcOH in CH2Cl2 ; 1H NMR (500 MHz, CDCl3)
δ 1.15 (s, 3 H), 1.26 (s, 3 H), 1.67 (s, 3 H), 1.80 (s, 3 H), 2.24
(s, 3 H), 2.32 (s, 3 H), 1.50-2.50 (m, 8 H), 2.51-2.52 (m, 1 H),
3.71-3.80 (m, 1 H), 4.15 (t, J ) 7.0 Hz, 1 H), 4.20 (d, J ) 8.5
Hz, 1 H), 4.28 (d, J ) 8.5 Hz, 1 H), 4.34 (d, J ) 7.0 Hz, 1 H),
4.36-4.50 (m, 2 H), 4.65 (d, J ) 2.5 Hz, 1 H), 4.93 (d, J ) 9.1
Hz, 1 H), 5.50-5.53 (m, 1 H), 5.62 (d, J ) 8.2 Hz, 1 H), 5.67
(d, J ) 7.1 Hz, 1 H), 6.26 (s, 1 H), 7.20-7.60 (m, 16 H), 7.74
(d, J ) 7.4 Hz, 1 H), 8.11 (d, J ) 7.5 Hz, 1 H); LRMS
(electrospray) m/ z [M + Na]+ 1123, calcd for C60H64N2O18Na
1123.
Resin 5. To a solution of carboxylic acid 4 (1.44 g, 1.31
mmol) and DIEA (4.4 mL, 25.2 mmol, 0.5 M final concentra-
tion) in anhydrous CH2Cl2 (50 mL) at room temperature was
added 2-chlorotrityl resin (8.00 g, 1.6 mmol/g, 10 m equiv) in
small portions with constant stirring. The reaction vessel was
sealed and shaken at room temperature for 3 h. The yellow
resin gradually turned into deep purple. TLC analysis of the
reaction solution showed that carboxylic acid 4 was completely
consumed. MeOH (8.4 mL) was then added, and the reaction
was shaken at room temperature for 30 min. The solution
was filtered, and the purple resin was washed using the
following standard washing procedure: MeOH (containing
0.5% TEA) and CH2Cl2 (containing 0.5% TEA) alternately for
a total of four cycles. The resin was dried under vacuum at
room temperature overnight (9.5 g).
a
Codes (in the format of “R1R2”) assigned to building blocks
during library synthesis (See Scheme 1). m/ z refers to [M + H]+
b
(electrospray MS) or [M - H]- (for entries 1, 2, 5, 6, 8, 9, 10, 13,
18, and 19). Calculated values are given in parentheses. c Yields
are based on initial loading of 2.9 µmol per microreactor (see main
text). Product purity was estimated by HPLC analysis and given
in parentheses. HPLC conditions: HP 1050 with an HP ODS
Hypersil C-18 column (5 µm, 100 × 4.6 mm); gradients: 75%
acetonitrile/water to 100% acetonitrile in 10 min; detection at 254
d
nm. Crude samples were purified with preparative HPLC using
similar gradients to that in note c.
EtOAc/petroleum ether to yield 2.23 g of the acetonide as a
white solid (93%): Rf ) 0.85, 50% EtOAc in petroleum ether;
1H NMR (500 MHz, CDCl3) δ 1.10 (br s, 9 H), 1.16 (s, 3 H),
1.24 (s, 3 H), 1.76 (s, 3 H), 1.80 (s, 3 H), 1.92 (s, 3 H), 2.00 (s,
3 H), 2.17 (s, 6 H), 1.50-2.20 (m, 4 H), 2.55-2.70 (m, 1 H),
3.93 (d, J ) 7.0 Hz, 1 H), 4.11 (d, J ) 8.5 Hz, 1 H), 4.27 (d, J
) 8.5 Hz, 1 H), 4.47 (d, J ) 6.6 Hz, 1 H), 6.64 (d, J ) 12.0 Hz,
1 H), 4.91 (d, J ) 9.0 Hz, 1 H), 5.03 (d, J ) 12.0 Hz, 1 H), 5.58
(dd, J ) 7.1 Hz, J ) 10.8 Hz, 1 H), 5.65 (d, J ) 7.0 Hz, 1 H),
6.26 (t, J ) 9.1 Hz, 1 H), 6.36 (s, 1 H), 7.32-7.43 (m, 5 H),
7.48-7.53 (m, 2 H), 7.64 (t, J ) 7.3 Hz, 1 H), 8.04 (d, J ) 7.5
Hz, 1 H); LRMS (electrospray) m/ z [M + Na]+ 1088, calcd for
C51H60Cl3NO17Na 1088.
Resin 6. Resin 5 was distributed into 400 microreactors8a
(20 mg each). The microreactors were then immersed in 5%
piperidine/DMF (400 mL) and shaken at room temperature
for 30 min. Aliquots was removed and measured by UV at
302 nm, and the loading was determined to be 2.9 µmol/
7-Tr oc-13-O-[N-[N-F m oc-O-(tr ich lor oeth yl)glu ta m yl]-
â-p h en ylisoser in yl]ba cca tin III (3). The above acetonide
(2.10 g, 1.97 mmol) was dissolved in cold formic acid (170 mL,
0 °C) and stirred at 0 °C for 3 h. The reaction mixture was

Design and Synthesis of a Taxoid Library
J . Org. Chem., Vol. 62, No. 17, 1997 6033
microreactor. The microreactors were washed thoroughly
according to the standard washing procedure and dried under
vacuum at room temperature for overnight.
1-IR. 1H NMR (500 MHz, CDCl3) δ 8.25-8.16 (m, 1 H),
8.10-8.03 (m, 2 H), 7.90 (d, J ) 8.3 Hz, 1 H), 7.84 (d, J ) 8.3
Hz, 1 H), 7.60-7.30 (m, 15 H), 6.88-6.85 (m, 2 H), 6.44 (s, 1
H), 6.30-6.27 (m, 1 H), 5.84 (dd, J ) 3.2 Hz, J ) 9.5 Hz, H),
5.75-5.65 (m, 2 H), 5.47 (d, J ) 3.6 Hz, 1 H), 5.00 (br d, J )
9.3 Hz, 1 H), 4.90-4.85 (m, 1 H), 4.30 (d, J ) 8.4 Hz, 1 H),
4.21 (d, J ) 8.4 Hz, 1 H), 4.01 (d, J ) 7.1 Hz, 1 H), 2.80-2.72
(m, 1 H), 2.60-1.20 (m, 8 H), 2.53 (s, 3 H), 2.45 (s, 3 H), 2.43
(s, 3 H), 2.08 (s, 3 H), 2.01 (s, 3 H), 1.90 (s, 3 H), 1.46 (s, 3 H),
1.25 (s, 3 H).
1-RF . 1H NMR (500 MHz, CDCl3) δ 8.11 (d, J ) 7.6 Hz, 2
H), 8.05 (d, J ) 7.6 Hz, 2 H), 7.92 (d, J ) 7.6 Hz, 2 H), 7.65-
7.20 (m, 15 H), 6.90 (d, J ) 4.9 Hz, 1 H), 6.55 (br d, J ) 7.3
Hz, 1 H), 6.40 (s, 1 H), 6.32-6.28 (m, 1 H), 5.82 (dd, J ) 2.5
Hz, J ) 9.5 Hz, 1 H), 5.72-5.67 (m, 2 H), 5.00 (br d, J ) 8.5
Hz, 1 H), 4.80-4.76 (m, 1 H), 4.30 (d, J ) 8.3 Hz, 1 H), 4.21
(d, J ) 8.3 Hz, 1 H), 4.00 (d, J ) 7.0 Hz,1 H), 2.85-2.75 (m, 1
H), 2.55-1.00 (m, 8 H), 2.53 (s, 3 H), 2.38 (s, 3 H), 2.09 (s, 3
H), 2.03 (s, 3 H), 1.96 (s, 3 H), 1.48 (s, 3 H), 1.47 (s, 3 H).
1-MO. 1H NMR (500 MHz, CDCl3) δ 8.48 (d, J ) 4.5 Hz, 1
H), 8.07 (d, J ) 7.5 Hz, 2 H), 7.69 (d, J ) 7.5 Hz, 1 H), 7.58-
7.00 (m, 19 H), 6.28 (s, 1 H), 6.17-6.12 (m, 1 H), 5.68-5.60
(m, 2 H), 5.56 (dd, J ) 7.2 Hz, J ) 10.4 Hz, 1 H), 5.35 (d, J )
4.3 Hz,1 H), 4.87 (br d, J ) 9.5 Hz, 1 H), 4.67-4.62 (m, 1 H),
4.26 (d, J ) 8.4 Hz,1 H), 4.14 (d, J ) 8.4 Hz, 1 H), 3.89 (d, J
) 6.9 Hz, 1 H), 3.70-3.60 (m, 5 H), 2.60-1.00 (m, 9 H), 2.51
(s, 3 H), 2.40 (s, 3 H), 2.19 (s, 3 H), 2.08 (s, 3 H), 1.93 (s, 3 H),
1.47 (s, 3 H), 1.46 (s, 3 H).
Resin 7. The above microreactors were split into 20 groups
each containing 20 microreactors. Each group was encoded
with a unique rf code (Scheme 1) and subjected to coupling
with the corresponding carboxylic acid (R1, Scheme 1) under
the following conditions: carboxylic acid (0.1 M), DIEA (0.2
M), PyBOP (0.1 M), DMF (20 mL), room temperature, 2 h;
additional carboxylic acid and PyBOP (half the previous
amounts), room temperature, 2 h. The reactions were quenched
with methanol. The 20 groups of microreactors were kept
separate, washed according to the standard procedure, and
dried under vacuum at room temperature overnight to yield
the resin-bound amides 7.
Resin 8. The above microreactors were resplit into 20 new
groups by combining one microreactor from each of the
previous groups. Each of the 20 new groups was encoded with
a unique rf code. One group, corresponding to R2 ) H, was
set aside. Each of the remaining 19 groups was then subjected
to coupling with one of the second set of carboxylic acids (R2,
Scheme 1) under the following conditions: carboxylic acid (0.4
M), 4-DMAP (0.35 M), DIC (0.4 M), CH2Cl2 (20 mL), room
temperature, 24 h; additional carboxylic acid and DIC (half
the previous amounts), room temperature, 24 h. The reactions
were quenched by the addition of methanol. The microreactors
were pooled together, washed, and dried under vacuum to
afford the desired resin-bound taxoids 8.
Ta xoid Libr a r y 1. The above 400 microreactors were
distributed into 400 glass vials (8 mL capacity). The rf code
in each Microreactor was read, and the vial was labeled with
that code. AcOH/CF3CH2OH/CH2Cl2 (2:1:7, 2 mL) was then
added into each vial. The vials were sealed and shaken at
room temperature for 4 h. The microreactors were taken out
and rinsed with methanol (1 mL/microreactor) and CH2Cl2 1
mL/microreactor). The solutions in the vials were then slowly
evaporated in a SpeedVac under carefully controlled vacuum.
The residues were then redissolved in benzene (2 mL/vial),
frozen, and lyophilized to yield 400 powders with colors
ranging from white to brown. Part of the library (20 com-
1-P K. 1H NMR (500 MHz, CDCl3) δ 9.04 (br d, J ) 4.4 Hz,
1 H), 8.76-8.72 (m, 2 H), 8.64 (d, J ) 4.5 Hz, 1 H), 8.23 (d, J
) 7.7 Hz, 1 H), 8.16 (d, J ) 7.5 Hz, 2 H), 7.92-7.70 (m, 6 H),
7.60-7.20 (m, 10 H), 6.70-6.60 (m, 1 H), 6.44 (br s, 1 H), 6.36
(s, 1 H), 6.32-6.28 (m, 1 H), 6.08 (d, J ) 15.5 Hz, 1 H), 5.91
(dd, J ) 1.7 Hz, J ) 9.3 Hz, 1 H), 5.86 (dd, J ) 7.3 Hz, J )
10.3 Hz, 1 H), 5.72-5.68 (m, 2 H), 5.02 (d, J ) 8.9 Hz, 1 H),
4.73-4.66 (m, 1 H), 4.34 (d, J ) 8.4 Hz, 1 H), 4.26 (d, J ) 8.4
Hz, 1 H), 4.06 (d, J ) 7.0 Hz, 1 H), 2.85-2.74 (m, 1 H), 2.60-
1.10 (m, 8 H), 2.49 (s, 3 H), 2.12 (s, 3 H), 2.03 (s, 3 H), 1.97 (s,
3 H), 1.47 (s, 3 H), 1.46 (s, 3 H).
1
pounds) were analyzed by mass spectrometry and/or H NMR
Ack n ow led gm en t. We thank Professor K. C. Nico-
laou, Mr. W. Ewing, and many other IRORI colleagues
for assistance and valuable discussions. Professor K.
C. Nicolaou is an advisor to IRORI Quantum Micro-
chemistry.
spectroscopy, and the structures were confirmed to be the
desired. Selective 1H NMR data (see Table 1 for purity.
Compounds are named using the rf codes in Scheme 1 in the
format of “1-R1R2”):
1-BU. 1H NMR (500 MHz, CDCl3) δ 8.09 (d, J ) 8.02 Hz,
2 H), 7.87-7.85 (m, 1 H), 7.62-7.50 (m, 2 H), 7.53-7.7.46 (m,
2 H), 7.40-7.25 (m, 4 H), 6.40 (d, J ) 7.5 Hz, 1 H), 6.32 (s, 1
H), 6.19 (m, 1 H), 5.70-5.68 (m, 2 H), 5.70-5.60 (m, 2 H), 5.53
(dd, J ) 7.2 Hz, J ) 10.3 Hz, 1 H), 5.30 (d, J ) 3.8 Hz, 1 H),
4.94 (br d, J ) 9.3 Hz, 1 H), 4.54-4.38 (m, 1 H), 4.29 (d, J )
8.4 Hz, 1 H), 4.18 (d, J ) 8.4 Hz, 1 H), 3.91 (d, J ) 7.0 Hz, 1
H), 2.80-0.80 (m, 29 H), 2.42 (s, 3 H), 2.13 (s, 3 H), 2.08 (s, 3
H), 1.97 (s, 3 H), 1.21 (s, 3 H), 1.15 (s, 3 H), 0.94 (s, 9 H).
Su p p or tin g In for m a tion Ava ila ble: 1H NMR spectra for
intermediates (4 pages). This material is contained in libraries
on microfiche, immediately follows this article in the microfilm
version of the journal, and can be ordered from the ACS; see
any current masthead page for ordering information.
J O9708699