Expedient Synthesis of N-Methyl Tubulysin Analogues
SCHEME 3. Stereoselective Synthesis of
N-tert-Butanesulfinyl Tubuvaline Ethyl Ester (28)
the ratio of tetrahydrooxazines 29 and 30 to 88:12. An
examination of solvents using 0.01 equiv of TsOH established
that more polar solvents such as THF and dioxane gave
decomposition products (entries 11 and 12), while toluene gave
a slightly better ratio than benzene with similar conversions after
5 h (entries 8 and 13). Because even low concentrations of acid
catalyzed not only tetrahydrooxazine formation but also undes-
ired sulfinyl group cleavage, the condensation reaction with
paraformaldehyde was performed in the absence of any acid
(Table 1, entry 14). Gratifyingly, heating 28 with paraformal-
dehyde in toluene resulted in complete conversion to the desired
N-sulfinyl-protected tetrahydrooxazine 29.
When the reaction was performed on a preparative scale, 29
proved to be stable to silica gel chromatography and was
obtained in 87% yield (Scheme 5). Reduction of 2927 with
support-bound cyanoborohydride (MP-BH3CN) resin under
acidic conditions then readily provided 31 in 97% yield as the
free amine after silica gel chromatography in the presence of
NH4OH. Accordingly, N-methyl tubuvaline ethyl ester (31) was
prepared in 49% yield over seven steps and serves as a versatile
and scalable intermediate for the construction of N-methyl
tubulysin analogues with variation at the Mep, Ile, or Tup/Tut
positions.
Coupling of N-Boc-isoleucine (Boc-Ile-OH) with N-methyl
tubuvaline 31 to form the hindered tertiary amide linkage was
next attempted. However, using a variety of coupling reagents
and reaction parameters, ester 32 was surprisingly the predomi-
nant product (Table 2). Bisacylated product 33 was also
observed in some examples, but no more than 8% of the desired
amide product 34 was observed in any case. The predilection
for O-acylation with a secondary alcohol was quite unexpected,
although it is not without precedent.28,29 For amino alcohol 31,
selective esterification is possibly aided by intramolecular
general base catalysis provided by the proximal secondary
amine. EDC, which exists as an HCl salt, HOAt, and excess
base gave primarily ester 32 contaminated with significant
amounts of bisacylated product 33, but no amide product 34
was observed (entry 1). Decreasing the amount of base used in
the reaction decreased the amount of 33 (entry 2), as did using
the less active HOBt in place of HOAt (entry 3). EDC with
HOBt and 1.2 equiv of base provided only monoacylated
products with a 97:3 ratio of ester 32 to desired amide 34. Using
these conditions in DMF, however, failed to give a significantly
improved product ratio (entry 4). Similar base and HOBt/HOAt
dependence was observed when support-bound cyclohexylcar-
The direct alkylation of N-sulfinyl-1,3-amino alcohol 28 under
basic conditions was first attempted using dihalogenated me-
thylene reagents such as CH2Br2. This type of reactivity is
precedented for N-acyl- and N-alkyl-1,2-amino alcohol scaf-
folds,23 as well as for N-alkyl-1,3-amino alcohols,24 but to our
knowledge, not for N-acyl-1,3-amino alcohols. Exploration of
the alkylation of 28 with CH2Br2 using a variety of solvents
and bases as precedented for N-acyl-1,2-amino alcohol systems
(Table 1, entries 1-4) proved unsuccessful. We therefore
explored condensation of 28 with paraformaldehyde under acidic
conditions to yield tetrahydrooxazine 30 (Scheme 4). Formation
of tetrahydrooxazines and oxazolidines by reaction of formal-
dehyde with 1,3- and 1,2-amino alcohols with N-aryl- and
N-alkyl substituents is common. Reactions with N-acyl- and
N-sulfonyl-1,2-amino alcohols have also been reported in a few
cases,25 but there are no reports using N-acyl-1,3-amino alcohols.
In our initial attempt to condense 28 with paraformaldehyde, 1
equiv of TsOH in benzene with heating at 70 °C resulted in
complete conversion to sulfinyl-deprotected tetrahydrooxazine
30 in less than 2 h (Table 1, entry 5). Unfortunately, 30 was
prohibitively unstable and hydrolyzed upon purification via silica
gel or reverse-phase chromatography. Attempts to convert the
crude product mixture directly to the desired N-methyl tubu-
26
valine 31 via reduction with NaCNBH3 gave instead the
undesired dimethylated product due to contaminating amounts
of paraformaldehyde that remained after workup. Therefore, our
focus turned to the preparation of tetrahydrooxazine 29, which
should be more stable and isolable.
Incrementally decreasing the amount of TsOH from stoichio-
metric amounts to 0.001 equiv (Table 1, entries 6-9) increased
(26) For examples of borohydride reduction of 1,3-N-alkyl-tetrahydroox-
azines, see: (a) Kirst, H. A.; Wind, J. A.; Leeds, J. P.; Willard, K. E.; Debono,
M.; Bonjouklian, R.; Greene, J. M.; Sullivan, K. A.; Paschal, J. W.; Deeter,
J. B.; Jones, N. D.; Ott, J. L.; Felty-Duckworth, A. M.; Counter, F. T. J. Med.
Chem. 1990, 33, 3086–3094. (b) Mandville, G.; Bloch, R. Eur. J. Org. Chem.
1999, 9, 2303–2307. (c) Bandarage, U. K.; Chen, L.; Fang, X.; Garvey, D. S.;
Glavin, A.; Janero, D. R.; Letts, L. G.; Mercer, G. J.; Saha, J. K.; Schroeder,
J. D.; Shumway, M. J.; Tam, S. W. J. Med. Chem. 2000, 43, 4005–4016. (d)
Charmantray, F.; Demeunynck, M.; Carrez, D.; Croisy, A.; Lansiaux, A.; Bailly,
C.; Colson, P. J. Med. Chem. 2003, 46, 967–977. (e) Mueller, A.; Polborn, K.;
Wanner, K. T. J. Hetercycl. Chem. 2007, 44, 575–590.
(23) (a) Vega-Perez, J. M.; Espartero, J. L.; Alcudia, F. J. Carbohydr. Chem.
1993, 12, 477–486. (b) Vega-Perez, J. M.; Espartero, J. L.; Alcudia, F. Carbohydr.
Res. 1992, 235, C5-C7. (c) Vega-Perez, J. M.; Candela, J. I.; Blanco, E.; Iglesias-
Guerra, F. Tetrahedron: Asymmetry 2002, 13, 2471–2483. (d) Vega-Perez, J. M.;
Vega, M.; Blanco, E.; Iglesias-Guerra, F. Tetrahedron: Asymmetry 2004, 15,
3617–3633. (e) Hajji, C.; Testa, M. L.; Salud-Bea, R. D. L.; Zaballos-Garcia,
E.; Server-Carrio, J.; Sepulveda-Arques, J. Tetrahedron 2000, 56, 8173–8177.
(24) Comins, D. L.; Sahn, J. J. Org. Lett. 2005, 7, 5227–5228.
(25) (a) Bertolini, G.; Aquino, M.; Ferrario, F.; Pavich, G.; Zaliani, A.; Sala,
A. J. Org. Chem. 1996, 61, 3358–3361. (b) Wagner, B.; Beugelmans, R.; Zhu,
J. Tetrahedron Lett. 1996, 37, 6557–6560. (c) Baeckvall, J. E.; Oshima, K.;
Palermo, R. E.; Sharpless, K. B J. Org. Chem. 1979, 44, 1953–1957. (d) Schrey,
A.; Osterkamp, F.; Straudi, A.; Rickert, C.; Wagner, H.; Koert, U.; Herrschaft,
B.; Harms, K. Eur. J. Org. Chem. 1999, 11, 2977–2990. (e) Aurelio, L.;
Brownlee, R. T. C.; Hughes, A. B.; Sleebs, B. E. Aust. J. Chem. 2000, 53, 425–
433. (f) Reddy, G. V.; Rao, G. V.; Sreevani, V.; Iyengar, D. S. Tetrahedron
Lett. 2000, 41, 949–951. (g) Aurelio, L.; Brownlee, R. T. C.; Hughes, A. B.
Org. Lett. 2002, 4, 3767–3769. (h) Ashley, E. R.; Cruz, E. G.; Stoltz, B. M.
J. Am. Chem. Soc. 2003, 125, 15000–15001. (i) Bennett, N. J.; Prodger, J. C.;
Pattenden, G. Tetrahedron 2007, 63, 6216–6231. (j) Dai, C.-F.; Cheng, F.; Xu,
H.-C.; Ruan, Y.-P.; Huang, P.-Q. J. Comb. Chem. 2007, 9, 386–394.
(27) To our knowledge no examples exist of reducing 1,3-N-acyl-, 1,3-N-
sulfonyl-, or 1,3-N-sulfinyl-tetrahydrooxazines to the corresponding 1,3-amino
alcohol. For examples with 1,2-N-acyl-oxazolidines, see refs 25f, i, and 25j, as
well as: (a) Reddy, G. V.; Rao, G. V.; Iyengar, D. S. Chem. Commun. 1999, 4,
317–318. (b) Reddy, G. V.; Rao, G. V.; Sreevani, V.; Iyengar, D. S. Tetrahedron
Lett. 2000, 41, 953–954.
(28) (a) Ksander, G. M.; Erion, M.; Yuan, A. M.; Diefenbacher, C. G.; El-
Chehabi, L.; Cote, D.; Levens, N. J. Med. Chem. 1994, 37, 1823–1832. (b)
Kilonda, A.; Compernolle, F.; Peeters, K.; Joly, G. J.; Toppet, S.; Hoornaert,
G. J. Tetrahedron 2000, 56, 1005–1012. (c) Cheng, Q.; Kiyota, H.; Yamaguchi,
M.; Horiguchi, T.; Oritani, T. Bioorg. Med. Chem. Lett. 2003, 13, 1075–1077.
(29) Welsh, L. H. J. Am. Chem. Soc. 1947, 69, 128–134.
J. Org. Chem. Vol. 73, No. 12, 2008 4365