Intramolecular aminolysis of trichloroethyl esters: A mild macrocyclization protocol for the preparation of cryptophycin derivatives

Article abstract of DOI:10.1016/S0957-4166(98)00287-0

The bifunctional catalyst 2-hydroxypyridine (2-pyridone) is shown to promote the intramolecular aminolysis of the polyfunctionalized long-chain amino trichloroethyl ester 8 to afford cryptophycin-51 (4). This process for the construction of the macrocycli

Full text of DOI:10.1016/S0957-4166(98)00287-0

TETRAHEDRON:
ASYMMETRY
Pergamon
Tetrahedron: Asymmetry 9 (1998) 2777–2781
Intramolecular aminolysis of trichloroethyl esters: a mild
macrocyclization protocol for the preparation of cryptophycin
derivatives
Andrew H. Fray ∗
Chemical Process R & D, Lilly Research Laboratories, A Division of Eli Lilly and Company, Indianapolis,
Indiana 46285-4813, USA
Received 6 July 1998; accepted 8 July 1998
Abstract
The bifunctional catalyst 2-hydroxypyridine (2-pyridone) is shown to promote the intramolecular aminolysis of
the polyfunctionalized long-chain amino trichloroethyl ester 8 to afford cryptophycin-51 (4). This process for the
construction of the macrocyclic core of cryptophycin derivatives is noteworthy for its convenience, avoidance of
expensive coupling reagents, and use of mild reaction conditions. © 1998 Elsevier Science Ltd. All rights reserved.
The cryptophycins are novel macrolides isolated from blue–green algae (Nostoc sp. strain GSV224)
which have been shown by Moore and coworkers1 to be potent tumor selective cytotoxins in vivo.
Following the isolation and first total synthesis2a of cryptophycin A (1), there has been considerable
interest in the development of novel synthetic routes to this class of compounds. Generally these efforts
have been focused on the synthesis of the penultimate olefin intermediate 2, followed by epoxidation of
the olefin. In a similar fashion, Moore and Tius2b synthesized cryptophycin-52 (3), a synthetic analog of 1
bearing a 3-amino-2,2-dimethyl propionic acid subunit, through the intermediacy of olefin cryptophycin-
51 (4). This sequence has been improved to support the multigram preparation of 3 and other analogs
for preclinical and clinical evaluation. In the course of addressing the issues involved with forming the
macrocyclic ring of 4, I found that 2-hydroxypyridine significantly accelerated the macrolactamization
of functionalized long-chain amino trichloroethyl esters like 8. I therefore report here a mild, convenient
process for the macrocyclization of key seco-derivatives for cryptophycin construction.
∗
E-mail: afray@lilly.com
0957-4166/98/$ - see front matter © 1998 Elsevier Science Ltd. All rights reserved.
PII: S0957-4166(98)00287-0

2778
A. H. Fray / Tetrahedron: Asymmetry 9 (1998) 2777–2781
1. Results and discussion
Several synthetic strategies for the construction of the macrocyclic core of the cryptophycins3 (and the
structurally related arenastatins4) have predominantly involved ring-closure via formation of an amide
linkage.5 The literature describes simultaneous or sequential removal of C and N-terminus protecting
groups to reveal an amino carboxylic acid, which is converted to the macrolactam via conventional
peptide coupling reagents. A problematic issue associated with this sequence included the poor impurity
profile of the crude product, which was due to the nature of the reagents and reaction conditions needed to
remove the protecting groups and induce ring closure. These conditions usually require chromatographic
purification of the macrocycle, which is expensive and difficult to implement on a large scale. These
issues were encountered during the development of the original2a macrocyclization sequence shown in
Scheme 1.
Scheme 1. Moore–Tius macrocyclization sequence
The three-step sequence commenced with the conversion of trichloroethyl ester 5 to the corresponding
carboxylic acid 6 using zinc and acetic acid (HOAc). Conversion of crude 6 to the amino carboxylate
7 was accomplished with trifluoroacetic acid (TFA) followed by a basic workup. In the final step,
ring closure was achieved in the presence of pentafluorophenyl diphenylphosphinate (FDPP) in DMF
to provide 4 in 61% yield (overall from 5) after chromatography. To reduce reagent expense, as well
as minimize the effects of waste-stream issues associated with the use of zinc, DMF, and FDPP, we
hypothesized that the trichloroethyl ester moiety might be sufficiently activated to allow a more direct
route involving intramolecular aminolysis resulting in macrocyclic ring closure (Scheme 2).
Scheme 2. Macrolactamization via intramolecular aminolysis
The following reaction protocol gave encouraging results. Removal of the N-Boc group of 5 was
accomplished within 15 min with neat TFA, followed by concentration and basic extractive work-up (0.5
N NaOH/EtOAc) to give nearly pure amino ester 8. As summarized in Table 1, HPLC analyses of stirred
room temperature solutions (0.02 M) of crude amino trichloroethyl ester 8 in DMF, CH3CN and toluene
indicated percent conversion of 8 to macrocycle 4 ranging from 13–22% over a period of 30 h (entries
1, 3, and 4). A heated solution of ester 8 in toluene exhibited an expected reaction rate increase and
yield (entry 2). However, since we were anxious to maintain the lowest reaction temperature compatible
with the chemical stability of the substrate (to minimize the production of impurities), the prospect of
chemically promoting the macrolactamization reaction was explored.
The catalysis of ester aminolysis reactions has been extensively studied, although the intramolec-
ular variants6 are rare and have generally involved the formation of five- and six-membered rings.
Very few aminolyses involving the formation of macrocycles6e have been reported. Examples of cat-

A. H. Fray / Tetrahedron: Asymmetry 9 (1998) 2777–2781
2779
Table 1
Initial screening: percent Conversions (8 to 4), Rxn concentration: 0.02 M
alysts for aminolysis reactions include amides,7a carboxylate anions,7b 2-hydroxypyridine,7c–g DMAP,7h
imidazole6d and nucleosides.7g
Accordingly, potential candidates were tested for their ability to accelerate ring closure. Toluene was
chosen as the reaction solvent based upon literature precedent of successful aminolyses accelerated by
DMAP7h and benzoate7b in this non-polar medium. As summarized in Table 2, 2-hydroxypyridine (entry
7) and benzoate anion (entry 8) have been shown to significantly accelerate the macrolactamization
process. In contrast, ring closure was very slow when the reaction was allowed to stir at room temperature
without a promoter (entry 1) or in the presence of amine bases such as imidazole, i-Pr2NEt, pyridine or
DMAP (entries 3–6). Hydroxybenzotriazole (HOBt) exhibited a modest rate increase (entry 2), perhaps
due to its structural similarity to 2-hydroxypyridine. TFA (1 equiv.) completely prevented ring closure
(entry 9) as an expected consequence of increased protonation of the amino terminus of 8.
Table 2
Catalyst screening in toluene, 0.02 M, 20°C. Percent conversions (8 to 4)
1.1. Origin of rate acceleration
Menger’s8a seminal study of the bimolecular aminolysis of activated esters showed that, in a non-
polar medium, the rate-limiting step is the breakdown of the tetrahedral intermediate to provide the
amide product and an alcohol. Furthermore, Menger provided evidence8b that the aminolysis of aryl
esters in aprotic solvents proceeds by a general base-catalyzed mechanism. Thus, benzoate anion, which
is a 10 times3 better proton acceptor than piperidine in toluene, functions to remove the ammonium
proton of the tetrahedral intermediate and thereby enables its collapse to give the neutral amide (TS-A,
Fig. 1). In contrast, 2-hydroxypyridine (TS-B) is thought to manifest its rate enhancing effect through
a bifunctional7d mechanism. Thus, stabilization of the ionic aminolysis transition state occurs through
simultaneous donation and acceptance of protons.
Having identified tetrabutylammonium benzoate and 2-hydroxypyridine as effective promoters for ring
closure, the effect of each on yield and product quality of macrocycle 4 was evaluated. Freshly prepared
0.02 M solutions of amino ester 8 were allowed to stir in the presence of each reagent until complete
consumption of 8 was observed. Each reagent was completely removed from the organic product solution

2780
A. H. Fray / Tetrahedron: Asymmetry 9 (1998) 2777–2781
Fig. 1. Transition state structures depicting various modes of stabilization by catalysts
via extraction with aqueous NaHCO3. Upon concentration, the crude oils were induced to crystallize
1
(1:1 EtOAc:hexanes), to afford macrocycle 4 in 93–98% purity as a crystalline powder. H NMR and
mass spectral analysis of the filtrate revealed that the symmetrical dimer 9 (Scheme 2) was the principle
impurity.
However, a limitation was noted during the evaluation of n-Bu4N benzoate as a promoter. This reagent
was prepared by the addition of 0.1 equiv. of n-Bu4NHSO4 to a toluene suspension of sodium benzoate (1
equiv.). When amino ester 8 was allowed to stir in this suspension, the reaction times for macrocyclization
on a gram-scale became widely variable; reactions took 25–100 h to proceed to completion (Table 3, entry
1). The variability of the reaction times are not clearly understood, but appear to be related to the amount
of adventitious water present in the reaction solution. The addition of 3 Å powdered molecular sieves
appeared to inhibit macrocyclization.
Table 3
Isolated yields of macrocycle 4. All reactions run at 0.02 M
Thus, 2-hydroxypyridine was found to be optimal for ring closure and facile workup. This reagent is
inexpensive and is easily removed by aqueous extraction. Furthermore, on a large-scale, macrocycle 4
was afforded in 98.4% purity and in 80% yield (Table 3, entry 2).
In summary, a novel process for the construction of the macrocyclic core of cryptophycin derivatives is
exemplified by the efficient, large-scale preparation of cryptophycin intermediate 4. This process show-
cases a 2-hydroxypyridine-promoted, intramolecular aminolysis (macrolactamization) of trichloroethyl
ester 8 and is noteworthy for its convenience, avoidance of expensive coupling reagents, and use of mild
conditions. Details, generalization of the method, and studies directed towards dimer (9) reduction will
be published in due course.
Acknowledgements
Helpful discussions with Professor Marvin Miller (University of Notre Dame), Dr. Michael Martinelli
and Dr. Eric Moher are gratefully acknowledged. Mass spectral evidence generated by Mr. David Robbins
proved invaluable in the characterization of dimer 9. The technical contributions of Mr. John Brennan,
who successfully demonstrated the macrolactamization process on a large-scale, are also gratefully
acknowledged.
References
1. Golakoti, T.; Ogino, J.; Heltzel, C. E.; Husrebo, T. L.; Jensen, C. M.; Larsen, L. K.; Patterson, G. M. L.; Moore, R. E.;
Moobery, S. L.; Corbett, T. H.; Valeriote, F. A. J. Am. Chem. Soc. 1995, 117, 12030–12049.

A. H. Fray / Tetrahedron: Asymmetry 9 (1998) 2777–2781
2781
2. (a) Barrow, R. A.; Hemscheidt, T.; Liang, J.; Paik, S.; Moore, R. E.; Tius, M. A. J. Am. Chem. Soc. 1995, 117, 2479–2490;
(b) Moore, R. E.; Tius, M. A.; Barrow, R. A.; Liang, J.; Corbett, T. H.; Valeriote, F. A.; Hemscheidt, T. K., US Pat. Appl.
US 95-400057.
3. Salamonczyk, G. M.; Han, K.; Guo, Z.-W.; Sih, C. J. J. Org. Chem. 1996, 61, 6893–6900; DeMuys, J.-M.; Rej, R.; Nguyen,
D.; Go, B.; Fortin, S.; Lavellee, J.-F. Bioorg. Med. Chem. Lett. 1996, 6, 1111–1116; Rej, R.; Nguyen, D.; Go, B.; Fortin,
S.; Lavellee, J.-F. J. Org. Chem. 1996, 61(18), 6289–6285.
4. Kobayashi, M.; Wang, W.; Ohyabu, N.; Kurosu, M.; Kitagawa, I. Chem. Pharm. Bull. 1995, 43(9), 1598–1600.
5. Exceptions are cyclization by ester bond formation: Gardinier, K. M.; Leahy, J. W. J. Org. Chem. 1997, 62(21), 7098–7099,
and by olefin bond formation: Kobayashi, M.; Kurosu, M.; Wang, W.; Kitagawa, I. Chem. Pharm. Bull. 1994, 42,
2394–2396.
6. Intramolecular aminolyses: (a) Gisin, B. F.; Merrifield, R. B. J. Am. Chem. Soc. 1972, 94, 3102; (b) Kirk, K. L.; Cohen,
L. A. J. Am. Chem. Soc. 1972, 94, 8142; (c) Fife, T. H.; DeMark, B. R. J. Am. Chem. Soc. 1976, 98, 6978; (d) Fife, T. H.;
DeMark, B. R. J. Am. Chem. Soc. 1979, 101, 7379 and references therein; (e) Wasserman, H. H.; Robinson, R. P.; Carter,
C. G. J. Am. Chem. Soc. 1983, 105(6), 1697–1698; Mutoh, R.; Shirai, R.; Koiso, Y.; Iwasaki, S. Heterocycles 1995, 41(1),
9–12; Ishihara, K.; Kuroki, Y.; Hanaki, N.; Ohara, S.; Yamamoto, H. J. Am. Chem. Soc. 1996, 118, 1569–1570.
7. Bimolecular aminolyses: (a) amides: Su, C.-W.; Watson, J. J. Am. Chem. Soc. 1974, 96(6), 1854–1857; (b) carboxylate:
Menger, F. M.; Vitale, A. C. J. Am. Chem. Soc. 1973, 95(15), 4931–4934; (c) 2-hydroxypyridine: Rony, P. R. J. Am. Chem.
Soc. 1969, 91(22), 6090–6096; (d) Nakamizo, N. Bull. Chem. Soc. Jpn 1969, 42(4), 1071–1077; (e) Robins, M. J.; Sarker,
S.; Xie, M.; Zhang, W.; Peterson, M. A. Tetrahedron Lett. 1996, 37(23), 3921–3924; (f) Openshaw, H. T.; Whittaker, N. J.
Chem. Soc. C. 1969, 1, 89–91; (g) nucleosides: Melander, C.; Horne, D. A. J. Org. Chem. 1997, 62(26), 9295–9297; (h)
DMAP: Cossy, J.; Thellend, A. Synthesis 1989, 753–755.
8. (a) Menger, F. M.; Smith, J. H. Tetrahedron Lett. 1970, 48, 4163; (b) Menger, F. M.; Smith, J. H. J. Am. Chem. Soc. 1972,
94(11), 3824–3829.
9. The use of excess 2-hydroxypyridine was found to give complete reactions within 14–16 h (overnight).