Selective Synthesis of the para-Quinone Region of Geldanamycin

Article abstract of DOI:10.1021/ol035400g

(Matrix presented) The quinone portion of the ansamycin geldanamycin was made with complete selectivity from the 1,4-dihydroquinone generated from a 1,4-bis-methoxymethyl (MOM) ether intermediate. Palladium catalysis with air gave the desired product in 9

Full text of DOI:10.1021/ol035400g

ORGANIC
LETTERS
2003
Vol. 5, No. 21
3859-3862
Selective Synthesis of the para-Quinone
Region of Geldanamycin
Merritt B. Andrus,* Erik J. Hicken, Erik L. Meredith, Bryon L. Simmons, and
John F. Cannon
Department of Chemistry and Biochemistry, Brigham Young UniVersity, C100 BNSN,
ProVo, Utah 84602-5700
mbandrus@chem.byu.edu
Received July 25, 2003
ABSTRACT
The quinone portion of the ansamycin geldanamycin was made with complete selectivity from the 1,4-dihydroquinone generated from a 1,4-
bis-methoxymethyl (MOM) ether intermediate. Palladium catalysis with air gave the desired product in 98% isolated yield. The structure was
established using NMR, UV, and X-ray analysis with comparisons to geldanamycin, ortho-quino-geldanamycin and a model compound.
The total synthesis of the antitumor antibiotic geldanamycin
(GA) was recently reported.1 Key steps included the devel-
opment of two asymmetric glycolate aldol reactions and a
dealkylative quinone-forming step with nitric acid. This
compound has generated considerable interest due to its
ability to bind heat shock protein 90 (Hsp90) and lower
cellular levels of various oncogenic kinases.2 Geldanamycin
is the most potent member of the ansamycin family, which
includes both the herbimycins and macbecins (Scheme 1).
17-Allylamino-GA, a semisynthetic analogue, is currently
in clinical trials.3
Geldanamycin presents a unique challenge in that it
possesses a trisubstituted quinone with a methoxyl group at
C17. The trimethoxylactam precursor employed in the
synthesis of GA provided some unexpected results. Unlike
the dimethoxy precursors employed for the synthesis of the
other ansamycins,4 the trimethoxylactam precursor for GA
1 gave either an unusual azaquinone product or decomposi-
tion upon treatment with the standard oxidants.5 Only with
nitric acid was GA generated as a minor component along
with the nonnatural ortho-quinone product 2 in a 1:10 ratio
(Scheme 2). The low yield and poor selectivity of the
dealkylative approach prompted the investigation of a new
route to the desired methoxy-p-quinone portion of GA using
Scheme 1
(1) Andrus, M. B.; Meredith, E. L.; Simmons, B. L.; Soma Sekhar, B.
B. V.; Hicken, E. J. Org. Lett. 2002, 4, 3549.
(2) (a) Whitesell, L.; Mimnaugh, E. G.; De Costa, B.; Myers, C. E.;
Neckers, L. M. Proc. Natl. Acad. Sci. U.S.A. 1994, 91, 8324. (b) Lawrence,
D. S.; Niu, J. Pharmacol. Ther. 1998, 77, 81. (c) Stebbins, C. E.; Russo,
A. A.; Schneider, C.; Rosen, N.; Hartl, F. U.; Pavletich, N. P. Cell, 1997,
89, 239. (d) Roe, S. M.; Prodromou, C.; O’Brien, R.; Ladbury, J. E.; Piper,
P. W.; Pearl, L. H. J. Med. Chem. 1999, 42, 260.
10.1021/ol035400g CCC: $25.00 © 2003 American Chemical Society
Published on Web 09/18/2003

Scheme 2
Scheme 3
phenol and dihydroquinone intermediates. The new approach
establishes, for the first time, a selective route to the GA
quinone, opening up a new route to GA and analogues of
this important template. In addition, the structure of the
methoxyquinone product is unambiguously established using
NMR, UV, and X-ray analysis with comparison to GA,
ortho-quino GA 2, and a simple model compound.
As a model for the GA synthesis, trimethoxy benzamide
3 was produced and explored for dealkylative quinone
formation (Scheme 3). An early GA-model study reported
by Schill and co-workers employed a trimethoxy substrate
that was dealkylated to the hydroxyquinone.6 The R,â-
unsaturated precursor to 3 was made from the corresponding
benzaldehyde, which was described in a previous report.7
Treatment with the standard oxidants, ceric ammonium
nitrate (CAN)8 or silver oxide, rapidly produced quinone
product in essentially quanitative isolated yields. CoF2 and
MnO2 gave lower yields of product, 50 and 25%, respec-
tively.9 Hypervalent iodine-based oxidants failed to give
product in this case.10 Unlike lactam 1 where conformational
effects appear to govern the reaction outcome, no trace of
aza-quinone was detected with 3. In this case, one-electron
removal gives radical cation 4 where the p-methoxyls
stabilize the charge through lone-pair donation. Addition of
water, loss of methanol, a proton, and an additional electron
gives intermediate 5.11 Water can then attack either ortho or
para to the carbonyl prior to quinone formation. In contrast
to previous reports of 1,2,4-trimethoxybenzenes where para-
quinone products were obtained,5 quinone 6 was also shown
to be the unexpected ortho-quinone. Single-crystal X-ray
analysis unambiguously identified the structure as shown.12
UV data of 6 was also obtained and compared to GA and
ortho-quino-GA 2. The λmax at 300 nm for the π-π* (CHCl3,
K-band) of 6 is very close to the energy of this transition
for 2 (λmax 303 nm).13 The corresponding value for GA is
much higher at 311 nm. Attack at the ortho position may be
due to steric factors or destabilization of the allylic cation at
the para position of 5. The amide may force the methyl ether
at this position to adopt a conformation that does not promote
lone-pair donation in this case.
(3) GA derivatives: (a) Rinehart, K. L.; McMillan, M. W.; Witty, T.
R.; Tipton, C. D.; Shield, L. S. Li, L. H.; Reusser, F. Bioorg. Chem. 1977,
6, 353. (b) Schnur, R. C.; Corman, M. L J. Org. Chem. 1994, 59, 2581. (c)
Schnur, R. C.; Corman, M. L.; Gallaschun, R. J.; Cooper, B. A.; Dee, M.
F.; Doty, J. L.; Muzzi, M. L.; Moyer, J. D.; DiOrio, C. I.; Barbacci, E. G.;
Miller, P. E.; O’Brien, A. T.; Morin, M. J.; Foster, B. A.; Pollack, V. A.;
Savage, D. M.; Sloan, D. E.; Pustilnik, L. R.; Moyer, M. P. J. Med. Chem.
1995, 38, 3806.
(4) Herbimycin A: (a) Nakata, M.; Osumi, T.; Ueno, A.; Kimura, T.
Tamai, T.; Tatsuta, K. Tetrahedron Lett. 1991, 32, 6015. Macbecin I: (b)
Baker, R.; Castro, J. L. J. Chem. Soc., Perkin Trans. 1 1990, 47. (c) Evans,
D. A.; Miller, S. J.; Ennis, M. D. J. Org. Chem. 1993, 58, 471. (d) Panek,
J. S.; Xu, F.; Rondon, A. C. J. Am. Chem. Soc. 1998, 120, 4113.
(5) Trimethoxy substrates normally give p-quinone products: (a) Cam-
eron, D. W.; Feutrill, G. I.; Patti, A. F.; Perlmutter, P.; Sefton, M. A. Aust.
J. Chem. 1982, 35, 1501. (b) Witiak, D. T.; Loper, J. T.; Ananthan, S.;
Almerico, A. M.; Verhoef, V. L.; Filppi, J. A. J. Med. Chem. 1989, 32,
1636. (c) Kozuka, T.; Bull. Chem. Soc. Jpn. 1982, 55, 2415. (d) Cheng, A.
C.; Castagnoli, N. J. Med. Chem. 1984, 27, 513. (e) Michael, J. P.; Cirillo,
P. F.; Denner, L.; Hosken, G. D.; Howard, A. S.; Tinkler, O. S. Tetrahedron
1990, 46, 7923. (f) Luly, J. R.; Rapoport, H. J. Org. Chem. 1981, 46, 2745.
(g) Kitahara, Y.; Nakahara, S.; Shimizu, M.; Yonezawa, T.; Kubo, A.
Heterocycles 1993, 36, 1909.
The shortcomings of the dealkylation route prompted the
investigation of the phenol and dihydroquinone strategies.
An unsaturated amide with additional functionality was
considered to be an improved model for GA. The route to
the p-di-MOM-protected hydroquinone began with meth-
oxyhydroquinone 8 (Scheme 4). Protection, formylation,14
and nitration gave 10a (R ) MOM) in high overall yield.
(6) Schill, G.; Merkel, C.; Zu¨rcher, C. Liebigs Ann. Chem. 1977, 288.
Trimethoxyamide was converted to the hydroxyquinone with boron tribro-
mide and alkylated with diazomethane. These conditions led only to
decomposition with 1, and other precursors led to GA.
(7) Andrus, M. B.; Meredith, E. L.; Soma Sekhar, B. B. V. Org. Lett.
2001, 3, 259.
(11) For a recent mechanistic investigation of dealkylative quinone
formation, see: Rathore, R.; Bosch, E.; Kochi, J. K. J. Chem. Soc., Perkin
Trans. 2 1994, 1157.
(12) X-ray data: orthorhombic space group P212121, a ) 7.359, b )
8.645, c ) 21.12 Å, independent data R1 ) 0.038. The Supporting
Information contains full details.
(8) Krohn, K.; Frese, P.; Flo¨rke, U. Chem. Eur. J. 2000, 6, 3887.
(9) Tomatsu, A.; Takemura, S.; Hashimoto, K.; Nakata, M. Synlett 1999,
1474.
(10) Tohma, H.; Morioka, H.; Harayama, Y.; Hashizume, M.; Kita, Y.
Tetrahedron Lett. 2001, 42, 6899.
(13) For a discussion of UV-structure correlation, see: Spectrometric
Identification of Organic Compounds, 5th ed.; Silverstein, R. M., Bassler,
G. C., Morrill, T. C., Eds.; John Wiley & Sons: New York, 1991; Chapter
7.
(14) Gilman, H.; Thirtle, J. R. J. Am. Chem. Soc. 1944, 66, 858.
3860
Org. Lett., Vol. 5, No. 21, 2003

Scheme 4
Under these conditions, the extra protection step was needed
group to the aniline, and treatment with tigloyl chloride22
gave the unsaturated benzamides 17 (R ) MOM, Me).
Dimethoxy 17 (R ) Me) was treated with TMSCl and
sodium iodide to give the phenol 18 (Table 1).23 The
because the C3-MOM group was removed during the
nitration step. For the phenol approach, 2,4-dimethoxyben-
zaldehyde 11 underwent Baeyer-Villiger oxidation,15 pro-
tection, formylation, and nitration with ammonium nitrate
in trifluoroacetic anhydride16 to give 10b (R ) Me).
Surprisingly, use of ammonium nitrate with 9 as with nitric
acid, also led to MOM removal. Alternative conditions for
nitration at this point gave multiple products or lower yields.17
These intermediates were separately converted to the benzyl
bromides 13 using sodium borohydride and MsCl in the
presence of lithium bromide and triethylamine according to
Kajiwara’s conditions.18 Use of standard conditions, PBr3
and pyridine or Ph3P with carbon tetrabromide, led to MOM
removal or low yield. Evans asymmetric alkylation with
S-oxazolidinone 14 proceeded with high yield and selectiv-
ity.19 Reduction with lithium borohydride provided the
alcohols 15 (R ) MOM, Me). Mitsunobu homologation, with
acetone cyanohydrin under the conditions of Ito,20 gave the
intermediate cyano compounds. Reduction to the aldehydes,
with DIBAL followed by addition of water, gave the cor-
responding aldehydes. Asymmetric allylborane addition fol-
lowing the conditions of Brown21 gave homoallylic alcohols
16. Protection as the TBS ether, reduction of the arylnitro
Table 1. Oxidation Conditions for Quinone Formation
entry
reagent
CAN
% yielda
1
2
3
4
5
6
7
8
9
19
20
b
b
15
DDQ
AgO/HNO3
HNO3/AcOH
PhIO
CuCl2/O2
Salacomine/O2
Co(salen)/O2
K3Fe(CN)6
27
26
14
a Isolated yield following silica gel chromatography. b Multiple products
were obtained without formation of 19.
(15) Matsumoto, M.; Kobayashi, H.; Hotta, Y. J. Org. Chem. 1984, 49,
4740.
(16) Crivello, J. V. J. Org. Chem. 1981, 46, 3056.
(17) (a) NO2+BF4-: Dwyer, C. L.; Holzapfel, C. W. Tetrahedron 1998,
54, 7843. (b) Cu(NO3)2‚(H2O)2.5: Boger, D. L.; Garbaccio, R. M. J. Org.
Chem. 1999, 64, 8350. (c) Nitropyridinium: Olah, G. A.; Narang, S. C.;
Olah, J. A.; Pearson, R. L.; Cupas, C. A. J. Am. Chem. Soc. 1980, 102,
3507. (d) La(NO3)3, Ouertani, M.; Girard, P.; Kagan, H. B. Tetrahedron
Lett. 1982, 23, 4315.
(18) Takatori, K.; Nishihara, M.; Nishiyama, Y.; Kajiwara, M. Tetra-
hedron 1998, 54, 15861.
(19) Evans, D. A.; Ennis, M. D.; Mathre, D. J. J. Am. Chem. Soc. 1982,
104, 1737.
deprotection was accompanied by TBS ether removal. The
standard oxidants in this case,24 CAN and DDQ (2,3-
(22) O’Hare, D.; Green, J. C.; Marder, T.; Collins, S.; Stringer, G.;
Kakkar, A. K.; Kaltsoyannis, N.; Kuhn, A.; Lewis, R.; Mehnert, C.; Scott,
P.; Kurmoo, M.; Pugh, S. Organometallics 1992, 11, 48.
(23) Cuevas, C.; Perez, M.; Martin, M. J.; Chicharro, J. L.; Fernandez-
Rivas, C.; Flores, M.; Francesch, A.; Gallego, P.; Zarzuelo, M.; de la Calle,
F.; Garcia, J.; Polanco, C.; Rodriguez, I.; Manzanares, I. Org. Lett. 2000,
2, 2545.
(20) (a) Wilk, B. K. Synth. Commun. 1993, 23, 2481. (b) Tsunoda, T.;
Uemoto, K.; Nagino, C.; Kawamura, M.; Kaku, H.; Ito, S. Tetrahedron
Lett. 1999, 40, 7355.
(21) Jadhav, P. K.; Bhat, K. S.; Perumal, P. T.; Brown, H. C. J. Org.
Chem. 1986, 51, 432.
(24) (a) Majetich, G.; Zhang, Y. J. Am. Chem. Soc. 1994, 116, 4979.
(b) Fukuyama, T.; Yang, L. J. Am. Chem. Soc. 1987, 109, 7881.
Org. Lett., Vol. 5, No. 21, 2003
3861

that treatment of 20 with catalytic palladium on carbon
(10%), following the conditions of Rapoport,27 with the flask
open to air gave quinone 19 in near quantitative yield. The
UV spectra showed a λmax at 310 nm consistent with the
same π-π* band observed for geldanamycin (311 nm).
These p-quinone UV bands are clearly distinct from the
π-π* bands seen with the o-quinones 2 and 6 (303 and 300
nm, respectively). Additional NMR experiments were also
performed to confirm the structure of 19. DEPT and
Scheme 5
1
HETCOR were used to assign the 13C NMR shifts and H
correlations for 19. HMBC was then used to establish the
connectivity and rule out alternatives. A key observation in
this regard was the through-bond coupling observed for the
amide hydrogen to the adjacent carbonyl of the quinone. In
the alternative o-quinone in this case, the amide hydrogen
would be found four and five bonds removed from the ring
carbonyls and a correlation would not be observed.
A new route to the p-quinone portion of geldanamycin
has been developed involving dihydroquinone oxidation
using 1,4-di-MOM protection. For the first time, selective
access to this unique trisubstituted quinone has been achieved.
New routes to this important target can now be undertaken
with confidence.
dichloro-5,6-dicyanoquinone), gave only a very low yield
of the desired product 19 (entries 1 and 2). The product 19
was shown to be the p-quinone. Spectral comparisons and
NMR data are discussed below. Silver oxide and nitric acid
gave multiple products without formation of 19 (entries 3
and 4). Iodosobenzene, as used recently for the synthesis of
25
saframycin and longithorone,
was also low yielding,
producing 19 in only 15% isolated yield (entry 5). At this
point, various catalysts with oxygen as oxidant that have been
used with more complex substrates were explored (entries
6-8). Copper(II) chloride failed to give product, while
salcomine,17b used for CC-1065, and Co(salen)26 again gave
only low yields of 19. Potassium ferricyanide (entry 9) also
gave a low yield.
Treatment of di-MOM 17 (R ) MOM) with TMSCl and
sodium iodide gave dihydroquinone 20 in 79% yield (Scheme
5). Again, the TBS ether was removed. Looking ahead, silyl
ether removal at C11 is needed prior to quinone formation
in the route to GA. Use of HF‚pyridine could be used to
selectively remove the TBS ether and maintain the MOM
ethers in this case (82%). TBAF and HF‚Et3N were not
effective for this transformation. Gratifyingly, it was found
Acknowledgment. We are grateful to The NIH
(GM57275) and the BYU Cancer Center for funding, Dr.
Bruce Jackson for MS assistance, and Dr. Li Du for NMR
assistance. We also thank Ryan J. Spencer for synthetic
assistance.
Supporting Information Available: Experimental pro-
cedures and characterization, including two-dimensional
NMR and X-ray data for 19. This material is available free
of charge via the Internet at http://pubs.acs.org.
OL035400G
(25) Myers, A. G.; Kung, D. W. J. Am. Chem. Soc. 1999, 121, 10828.
(b) Layton, M. E.; Morales, C. A.; Shair, M. D. J. Am. Chem. Soc. 2002,
124, 773.
(27) (a) Luly, J. R.; Rapoport, H. J. Org. Chem. 1984, 49, 1671. For
other routes and oxidants involving 1,4-di-MOM substrates: (b) Roush,
W. R.; Coffey, D. S.; Madar, D. J. J. Am. Chem. Soc. 1997, 119, 11331.
(c) Noland, W. E.; Kedrowski, B. L. Org. Lett. 2000, 2, 2109.
(26) (a) Lipshutz, B. H.; Mollard, P.; Pfeiffer, S. S.; Chrisman, W. J.
Am. Chem. Soc. 2002, 124, 14282. (b) With Cu, see: Matsumoto, M.;
Kobayashi, H.; Hotta, Y. Synth. Commun. 1985, 15, 515.
3862
Org. Lett., Vol. 5, No. 21, 2003