Total synthesis and evaluation of C25-benzyloxyepothilone C for tubulin assembly and cytotoxicity against MCF-7 breast cancer cells

Article abstract of DOI:10.1016/j.bmcl.2008.07.024

The total synthesis of C25-benzyloxy epothilone C is described. A sequential Suzuki-Aldol-Yamaguchi macrolactonization strategy was utilized employing a novel derivatized C8-C12 fragment. The C25-benzyloxy analog exhibited significantly reduced biological activity in microtubule assembly and cytotoxicity assays. Molecular modeling simulations indicated that excessive steric bulk in the C25 position may reduce activity by disrupting key hydrogen bonds that are crucial for epothilone binding to β-tubulin.

Full text of DOI:10.1016/j.bmcl.2008.07.024

Bioorganic & Medicinal Chemistry Letters 18 (2008) 4904–4906
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Total synthesis and evaluation of C25-benzyloxyepothilone C for tubulin
assembly and cytotoxicity against MCF-7 breast cancer cells
Oliver E. Hutt a, Bollu S. Reddy a, Sajiv K. Nair a, Emily A. Reiff a, John T. Henri a, Jack F. Greiner a,
Ting-Lan Chiu c, David G. VanderVelde a, Elizabeth A. Amin c, Richard H. Himes b, Gunda I. Georg a,c,
*
a Department of Medicinal Chemistry, University of Kansas, 1251 Wescoe Hall Drive, Lawrence, KS 66045, USA
b Department of Molecular Bioscience, University of Kansas, 1200 Sunnyside Avenue, Lawrence, KS 66045, USA
c Department of Medicinal Chemistry, and Institute for Therapeutics Discovery and Development, University of Minnesota, 717 Delaware Street SE, Minneapolis, MN 55414, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
The total synthesis of C25-benzyloxy epothilone C is described. A sequential Suzuki–Aldol–Yamaguchi
macrolactonization strategy was utilized employing a novel derivatized C8–C12 fragment. The C25-ben-
zyloxy analog exhibited significantly reduced biological activity in microtubule assembly and cytotoxic-
ity assays. Molecular modeling simulations indicated that excessive steric bulk in the C25 position may
reduce activity by disrupting key hydrogen bonds that are crucial for epothilone binding to b-tubulin.
Ó 2008 Published by Elsevier Ltd.
Received 6 May 2008
Accepted 1 July 2008
Available online 11 July 2008
Keywords:
Epothilones
C-25 epothilone analogue
Total synthesis
Tubulin assembly
Cytotoxicity
S
N
S
N
The epothilone class of natural products is a family of potent
cytotoxic polyketide marcolides first isolated in 1992 by Höfle
et al. from the myxobacterium Sorangium cellulosum.1,2 Like paclit-
axel, they disrupt microtubule dynamics, resulting in cell cycle ar-
rest and apoptosis.3,4 Most importantly, the epothilones maintain
impressive efficacy against several paclitaxel-resistant cancer cell
lines and they are now generally recognized as the next generation
of clinically relevant anti-mitotic agents. Some members of the
epothilone family contain an epoxide while others, typified by epo-
thilone C (1) and epothilone D (2), contain a double bond in place
of the epoxide moiety (Fig. 1).
R
8
25
15
O
25
OH
BnO
OH
1
O
O
OH
O
O
OH O
1 R = H epothilone C
2 R = Me epothilone D
3
Figure 1. Structures of epothilones.
Resistance to paclitaxel has arisen through three main mecha-
nisms: overexpression of P-glycoprotein (Pgp), which lowers the
intracellular concentration of the drug, overexpression of the b-
tubulin isotype b-III, and tubulin point-mutations in key amino
acid residues important to taxane binding.5 The epothilones appear
to be able to evade Pgp efflux and retain activity in cell lines that
have become resistant to paclitaxel because of mutations in b-
tubulin. This implies that, despite sharing a common binding site
on b-tubulin, the mode of binding of paclitaxel and the epothilones
is significantly different.6 As a result, early attempts to find a
shared pharmacophore between paclitaxel and epothilone B were
unsuccessful.7–10 The clinical importance of this class of com-
pounds has led to an explosion of synthetic activity7 and structural
studies have led to two b-tubulin-epothilone binding models. The
first is based on electron crystallography (EC) of Zn2+ induced tubu-
lin sheets,11 while the second is derived from NMR studies.12
One approach to gaining supporting evidence for either of these
binding hypotheses is through photoaffinity labeling of microtu-
bules with an appropriate photoreactive epothilone analog. To-
ward this end we have synthesized the C25-benzyloxy derivative
3 (Fig. 1) as a potential precursor to a photoaffinity analog. We
now report the synthesis of 3 and the resulting microtubule assem-
bly and cytotoxicity studies.
The synthetic strategy for the synthesis of 3 is outlined in Figure
2 and involved the preparation and successive coupling of building
blocks 4, 5, and 6.13,14 We envisioned that the C11–C12 bond
would be formed by a Suzuki reaction, the C6–C7 bond by an Aldol
reaction, and the use of the Yamaguchi procedure to form the
macrocycle.
* Corresponding author. Tel.: +1 785 864 4498; fax: +1 785 864 5836.
E-mail address: georg239@umn.edu (G.I. Georg).
0960-894X/$ - see front matter Ó 2008 Published by Elsevier Ltd.
doi:10.1016/j.bmcl.2008.07.024

O. E. Hutt et al. / Bioorg. Med. Chem. Lett. 18 (2008) 4904–4906
4905
The primary TBS ether functionality of 11 was then selectively
cleaved in 82% yield on treatment with fluorosilicic acid
(H2SiF6),14,17 and the resulting alcohol was subsequently oxidized
with 2-iodoxybenzoic acid (IBX) in EtOAc to give the aldehyde 12
in 95% yield. The stage was now set to install the final fragment.
Accordingly, ketone 6 was treated with LDA to generate the Z-eno-
late, which was subsequently treated with aldehyde 12 to deliver
the syn-Aldol product 13 in 54% yield.
The secondary alcohol function of 13 was protected as the TBS
ether. The primary TBS ether was then selectively deprotected
(84% yield) with fluorosilicic acid, and the resulting alcohol was
sequentially oxidized with Dess–Martin periodane (DMP) and per-
chlorate to the C1 acid in preparation for the Yamaguchi
macrolactonization.
1. Suzuki Coupling
12
S
N
11
B
I
15
7
O
BnO
OH
5
4
2. Aldol
Reaction
1
TBSO
6
3. Yamaguchi
Macrolactonization
TBSO
O
6
Figure 2. Synthetic strategy for the construction of 3.
O
O
O
O
TiCl4, DIEA,
The allylic alcohol group at C15 was deprotected and the subse-
quent Yamaguchi macrolactonization delivered the core epothi-
lone skeleton 14. Deprotection of the TBS groups furnished the
target compound 3 (Scheme 2).
CH2Cl2, -78 °C
N
O
N
O
BOM-Cl
76%
BnO
Me
Ph
Me
Ph
With the desired probe in hand we could begin to test its suit-
ability for biological studies. As the two most commonly used
photoaffinity labels are aryl azide and aryl diazirines, the steric de-
mand of the benzyl function of 3 would serve as an acceptable sur-
rogate to probe the suitability of this approach.18 Unfortunately, at
8
7
LiBH4, Et2O
0 °C, 68%
a concentration of 100
a microtubule assembly assay (EC50 for epothilone B = 2.5
and was essentially devoid of cytotoxicity against the cancer cell
line MCF-7 (2.1
M, ED50/ED50Epo B = 1050).19 The available evi-
l
M, compound 3 was completely inactive in
TBSOTf, CH2Cl2
OH
OTBS
l
M)19
2,6-lutidine, 0 °C
BnO
BnO
98%
l
9
10
dence suggests that the loss of activity is not due to a change in
the solution conformation of the free compound, but to an unfavor-
able interaction in the bound form. Comparison of the solution
NMR data of 3 with epothilone A,20 the C12–C13 b-epoxide of epo-
thilone C (1) indicates there are only local and relatively small con-
formational adjustments to the steric demand of the OBn group:
the H6–H7 coupling constant declined from 9 Hz to 7 Hz. This rep-
resents a change in the torsion angle of approximately 15 degrees.
Thus, both the C8 methyl group of the parent compound and the
OBn group in 3 essentially point away from the macrocycle and to-
ward a hydrophobic surface of the protein in the models of the
bound state. These data suggested that the OBn group was too
large to fit into this sterically constrained hydrophobic area.
To probe this hypothesis a molecular modeling study was con-
ducted. The pre-docked conformation of C25-benzyloxyepothilone
C was first prepared by modifying the experimental EpoA structure
and optimizing the resulting geometry using the MMFF94s force
field in MOE (Chemical Computing Group, Inc.). The optimized
Scheme 1. Synthesis of fragment 10.
Fragments 4 and 6 were synthesized as previously reported.15
The synthesis of the precursor to fragment 5 is outlined in Scheme
1.16 The oxazolidinone 7 was readily obtained through the acyla-
tion of the oxazolidinone core with pent-4-enoyl chloride. The C8
stereochemistry was then established through alkylation of 7 with
BOM–Cl to furnish benzyloxy intermediate 8 in 76% yield. The aux-
iliary was then reductively cleaved on treatment of 8 with LiBH4 to
provide alcohol 9 in 68% yield. The alcohol moiety of 9 was subse-
quently protected as the TBS ether 10 in preparation for the
planned Suzuki coupling (Scheme 2).
With fragment 10 in hand the synthesis of 3 was accomplished
as shown in Scheme 2. The alkene function of 10 was subjected to
hydroboration with 9-BBN, and the ensuing Sukuzi coupling with
vinyl iodide 4 delivered the desired cis alkene 11 in 92% yield.
1. 9-BBN, THF
2. 4, PdCl2(dppf)⋅CH2Cl2
S
S
N
1. 40% HF, MeCN, Et2O
glass splinters, 82%
N
BnO
OTBS
CsCO3, Ph3As, H2O
92%
O
BnO
OTBS
12
BnO
OTBS
OTBS
2. IBX, EtOAc, 70°C, 95%
11
10
6, LDA, THF
54%
1. TBSOTf, 67%
S
N
S
N
S
N
2. 40% HF, MeCN, Et2O
glass splinters, 84%
3. DMP, 95%
20% TFA
OH
BnO
BnO
OTBS
BnO
OH
83%
4. NaClO2, Na2HPO4, 59%
5. TBAF, 48%
6. 2,4,6-trichlorobenzoyl
O
O
TBSO
TBSO
O
OH O
O
OR1
O
OR1
O
chloride, DMAP, Ph-Me, 58%
14 R1 = TBS
3
13 R1 = TBS
Scheme 2. Synthesis of target compound 3.

4906
O. E. Hutt et al. / Bioorg. Med. Chem. Lett. 18 (2008) 4904–4906
Figure 3. Docked configurations (Surflex-Dock, Tripos, Inc.) of epothilone A (a, left panel) and C25-benzyloxyepothilone C (b, right panel) are shown with MOLCAD (Tripos,
Inc.) electron density surfaces of the tubulin binding site (1TVK11), onto which hydrogen bond donor and acceptor regions have been mapped. Red areas represent hydrogen
bond donors; blue areas represent hydrogen bond acceptors; and gray indicates regions in which no hydrogen bonding takes place. Key ligand–receptor interactions are
shown in both pictures. In (a), hydrogen bonding occurs between the EpoA thiazole nitrogen and the imidazole NH of His227; between the EpoA C1 carbonyl and two amino
groups in Arg276; between the EpoA C3 hydroxyl and the Thr274 backbone carbonyl; between the EpoA C5 carbonyl and the Thr274 backbone NH; and between the EpoA C7
OH and Arg282 and Pro272. In the docked configuration of C25-benzyloxyepothilone C (b), all of these interactions disappear due to sterically driven ligand rearrangement,
except for a single hydrogen bond between the epothilone C1 carbonyl and Arg276. Validation was performed by comparing the docked configuration of EpoA to the
experiment (RMSD = 1.575 Å).
7. Wang, M.; Xia, X.; Kim, Y.; Hwang, D.; Jansen, J. M.; Botta, M.; Liotta, D. C.;
Snyder, J. P. Org. Lett. 1999, 1, 43.
8. Ojima, I.; Chakravarty, S.; Inoue, T.; Lin, S.; He, L.; Horwitz, S. B.; Kuduk, S. D.;
C25-benzyloxyepothiloneC usedfor docking was in good agreement
with the predicted solution conformation of EpoA and with the NMR
experimental results. When the C25-benzyloxy analog was docked
into the tubulin binding site (Fig. 3), as predicted, the OBn group
proved too large. The subsequent reorientation of the molecule in
the binding site disrupted two crucial hydrogen-bonding interac-
tions: between the thiazole nitrogen and the imidazole NH of
His227 and between the C7–OH and Arg282 and Pro272.
In conclusion, we have synthesized the first C25 functionalized
epothilone derivative21 as a model to test the suitability of this po-
sition for the placement of a photoreactive function. Unfortunately,
this analog was inactive due to the steric demand at the C25 posi-
tion, which disabled key hydrogen bonds resulting in significantly
weaker ligand binding.
Danishefsky, S. J. Proc. Natl. Acad. Sci. U.S.A. 1999, 96, 4256.
9. He, L.; Jagtap, P. G.; Kingston, D. G. I.; Shen, H.-J.; Orr, G. A.; Horwitz, S. B.
Biochemistry 2000, 39, 3972.
10. Giannakakou, P.; Gussio, R.; Nogales, E.; Downing, K. H.; Zaharevitz, D.;
Bollbuck, B.; Poy, G.; Sackett, D.; Nicolaou, K. C.; Fojo, T. Proc. Natl. Acad. Sci.
U.S.A. 2000, 97, 2904.
11. Nettles, J. H.; Li, H.; Cornett, B.; Krahn Joseph, M.; Snyder James, P.; Downing
Kenneth, H. Science 2004, 305, 866.
12. Reese, M.; Sanchez-Pedregal, V. M.; Kubicek, K.; Meiler, J.; Blommers, M. J. J.;
Griesinger, C.; Carlomagno, T. Angew. Chem., Int. Ed. 2007, 46, 1864.
13. Balog, A.; Meng, D.; Kamenecka, T.; Bertinato, P.; Su, D.-S.; Sorensen, E. J.;
Danishefsky, S. J. Angew. Chem., Int. Ed. 1997, 35, 2801.
14. Schinzer, D.; Bauer, A.; Schieber, J. Chem. Eur. J. 1999, 5, 2492.
15. Reiff, E. A.; Nair, S. K.; Reddy, B. S. N.; Inagaki, J.; Henri, J. T.; Greiner, J. F.; Georg,
G. I. Tetrahedron Lett. 2004, 45, 5845.
16. Evans, D. A.; Urpi, F.; Somers, T. C.; Clark, J. S.; Bilodeau, M. T. J. Am. Chem. Soc.
1990, 112, 8215.
17. Pilcher, A. S.; DeShong, P. J. Org. Chem. 1993, 58, 5130.
Acknowledgment
18. Dorman, G.; Prestwich, G. D. Trends Biotechnol. 2000, 18, 64.
19. Liu, Y.; Ali, S. M.; Boge, T. C.; Georg, G. I.; Victory, S.; Zygmunt, J.; Marquez, R. T.;
Himes, R. H. Comb. Chem. High Throughput Screening 2002, 5, 39.
20. Victory, S. F.; Vander Velde, D. G.; Jalluri, R. K.; Grunewald, G. L.; Georg, G. I.
Bioorg. Med. Chem. Lett. 1996, 6, 893.
The authors gratefully acknowledge financial support from the
National Institutes of Health: National Cancer Institute CA79641.
21. The spectroscopic data of all intermediates were in agreement with their
structures. Spectroscopic data for compound 3: 1H NMR (400 MHz, CDCl3)
d 7.37–7.29 (m, 5H), 6.97 (s, 1H), 6.56 (s, 1H), 5.47 (m 1H), 5.37 (m, 1H),
5.24 (d, J = 7.8 Hz, 1H), 4.55 (d, J = 11.9 Hz, 1H), 4.47 (d, J = 11.0 Hz, 1H),
4.04 (obscured br d, J = 7.0 Hz, 1H), 4.01 (m, 1H), 3.63 (dd, J = 2.9 Hz,
J = 9.2 Hz, 1H), 3.57 (dd, J = 3.6 Hz, J = 9.2 Hz, 1H), 3.23 (s, 1H), 3.18 (t,
J = 6.9 Hz, 1H), 2.70 (s and obscured m, 4H), 2.51 (d, J = 2.8 Hz, 2H), 2.49
(s, 1H), 2.22–2.15 (m, 2H), 2.10 (s, 3H), 1.98–1.90 (m, 1H), 1.70–1.50 (m,
4H), 1.35 (m, 1H), 1.31 (s, 3H), 1.20 (d, J = 6.8 Hz, 3H), 1.11 (s, 3H); 13C
References and notes
1. Höfle, G.; Bedorf, N.; Gerth, K.; Reichenbach, H. Offenlegungsschrift der
Bundesrepublik Deutschland 1993, DE 4138042, A1.
2. Gerth, K.; Bedorf, N.; Hoefle, G.; Irschik, H.; Reichenbach, H. J. Antibiot. 1996, 49,
560.
3. Bollag, D. M.; McQueney, P. A.; Zhu, J.; Hensens, O.; Koupal, L.; Liesch, J.; Goetz,
M.; Lazarides, E.; Woods, C. M. Cancer Res. 1995, 55, 2325.
4. Altmann, K.-H. Curr. Pharm. Des. 2005, 11, 1595.
5. Fojo, T.; Menefee, M. Ann. Oncol. 2007, 18, v3.
6. Gupta, M. L., Jr.; Bode, C. J.; Georg, G. I.; Himes, R. H. Proc. Natl. Acad. Sci. U.S.A.
2003, 100, 6394.
NMR (100 MHz, CDCl3)
d 218.8, 170.7, 164.9, 152.1, 138.0, 137.9, 133.9,
128.4, 127.7, 127.6, 123.9, 119.8, 116.2, 78.9, 75.73, 73.4, 73.4, 72.0, 52.4,
45.1, 41.8, 38.7, 31.4, 28.3, 27.8, 25.2, 21.9, 21.8, 19.2, 15.4, 15.4; MS
(FAB) m/e 584.4 (M+H); ½a D20
ꢁ 23 (c 0.29, CHCl3).
ꢀ