Synthesis and evaluation of C2 functionalized analogs of the α-tubulin-binding natural product pironetin

Article abstract of DOI:10.1016/j.bmcl.2017.10.057

Pironetin is an α-tubulin-binding natural product with potent antiproliferative activity against several cancer cell lines that inhibits cell division by forming a covalent adduct with α-tubulin via a Michael addition into the natural product's α,β-unsatu

Full text of DOI:10.1016/j.bmcl.2017.10.057

Bioorganic & Medicinal Chemistry Letters xxx (2017) xxx–xxx
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Synthesis and evaluation of C2 functionalized analogs of the
binding natural product pironetin
a-tubulin-
⇑
David S. Huang, Henry L. Wong, Gunda I. Georg
Department of Medicinal Chemistry, Institute for Therapeutics Discovery & Development, University of Minnesota, 717 Delaware St. 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:
Pironetin is an
cancer cell lines that inhibits cell division by forming a covalent adduct with
addition into the natural product’s ,b-unsaturated lactone. We designed and prepared analogs carrying
electron-withdrawing groups at the -position (C2) of the ,b-unsaturated lactone with the goal to gen-
erate potent and selective binding analogs. We prepared derivatives containing halogens, a phenyl, and a
methyl group at the C2 position to evaluate the structure-activity relationship at this position. Testing of
the analogs in ovarian cancer cell lines demonstrated 100–1000-fold decreased antiproliferative activity.
Ó 2017 Elsevier Ltd. All rights reserved.
a
-tubulin-binding natural product with potent antiproliferative activity against several
Received 25 August 2017
Revised 21 October 2017
Accepted 24 October 2017
Available online xxxx
a
-tubulin via a Michael
a
a
a
Keywords:
Pironetin
Synthesis
Tubulin
Cytotoxicity
Structure-activity
Main text
Michael acceptors form reversible covalent adducts with thiols
and that ‘‘specific non-covalent interactions in concert with the
Pironetin (1, Fig. 1), a natural product isolated in 19931 and
covalent bond are needed to stabilize the complex” between the
protein and the Michael acceptor for irreversible bond forma-
tion.14,15 Using this principle, they were able to obtain a potent
and selective MSK/RSK-family kinase inhibitor. In related work
they showned that Michael acceptors containing electron-deficient
1994,2 has potent antiproliferative activity and acts via disruption
of tubulin polymerization dynamics by binding to
a
-tubulin.3–7
This mechanism differs from FDA-approved chemotherapeutics,
which disrupt tubulin polymerization by binding to b-tubulin.8–10
X-ray crystallography revealed that the natural product forms a
covalent adduct with cysteine 316 in
tion into the
unique mechanism of action, it has not been developed into a drug
candidate. An in vivo study of pironetin showed poor efficacy and
mice dosed with pironetin exhibited severe weight loss.13 Since the
in vivo toxicity could be due to non-selective formation of covalent
adducts between pironetin and other biomolecules, we hypothe-
aromatic groups and heterocycles at the a-position form reversible
a
-tubulin via Michael addi-
covalent bonds with thiols.16 In a separate study, Yu and coworkers
evaluated the effect of adding a fluorine to the EGFR-TK covalent
inhibitor afatinib.17 They found that a chemically tuned analog,
a
,b-unsaturated lactone.11,12 While pironetin has a
containing a fluorine at the a-position of afatinib’s a,b-unsaturated
amide, was highly potent and had reduced off-target reactivity.
We therefore designed and prepared pironetin analogs with dif-
ferent functional groups at the
lactone to evaluate structure-activity relationships and determine
the feasibility of modifying the -position of the natural product
a-position of the a,b-unsaturated
sized that pironetin analogs that covalently label
a-tubulin but
form reversible bonds with off-target proteins could possess
decreased off-target toxicity. Some support for this hypothesis
was obtained, when we incubated pironetin the with monoethyl
ester of glutathione and observed a glutathione-pironetin covalent
adduct by LC–MS/MS analysis (see SI).
a
to decrease its non-selective covalent adduct formation. While
our group along with other groups had previously evaluated the
structure-activity relationship of different positions of the
unsaturated lactone of pironetin and related analogs,4,5,18–22 the
effect of the addition of a functional group to the -position of
,b-unsaturated lactone has not been reported in the literature.
a,b-
Previous studies reported that the addition of electron-with-
drawing groups at the
a
a
-position of Michael acceptors can
the
a
decrease off-target covalent adduct formation of covalent inhibi-
For the synthesis of the
a-functionalized analogs 2, we envi-
tors. Taunton and coworkers showed that
a-nitrile containing
sioned that the
a,b-unsaturated lactone could be formed via lac-
tonization of intermediate 3 (Scheme 1). The
a,b-unsaturated
lactone of pironetin has been synthesized using this method in
⇑
Corresponding author.
E-mail address: georg@umn.edu (G.I. Georg).
several total syntheses.23–29 The tri-substituted olefin could be
https://doi.org/10.1016/j.bmcl.2017.10.057
0960-894X/Ó 2017 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Huang D.S., et al. Bioorg. Med. Chem. Lett. (2017), https://doi.org/10.1016/j.bmcl.2017.10.057

2
D.S. Huang et al. / Bioorganic & Medicinal Chemistry Letters xxx (2017) xxx–xxx
would be tolerated at the
Suzuki coupling, the secondary alcohol in
a
-position (Scheme 3). Prior to the
O
a-bromopironetin ana-
OMe OH O 1
4
log 2d was protected as the silyl ether. The coupling between vinyl
halide 14 and phenylboronic acid proceeded in moderate yield
under mild conditions to give intermediate 15. We completed
the synthesis of analog 13 following removal of the TBS protecting
group with BF3ÁEt2O.
Me
Me Me
Me
(–)-Pironetin (1)
The antiproliferative activities of the new analogs were evalu-
ated in drug-sensitive OVCAR5 and A2780 ovarian cancer cell lines
Fig. 1. Structure of pironetin.
(Table 1). The addition of any group at the
resulted in decreased antiproliferative activity. The addition of a
methyl group to the -position of pironetin (entry 4) resulted in
an approximate 200-fold decrease in activity. The -chloro and
-fluoro analogs (entries 3 and 5) showed similar GI50 values even
though these halogens along with the methyl group have different
electronic properties. The -phenyl analog 13 was found to be
inactive (entry 7). These results suggest that the decreased activity
of the -functionalized analogs could be due to the steric proper-
ties of the group at the -position instead of the electronic proper-
ties of the various groups.
Although the -functionalized pironetin analogs exhibited
decreased biological activity relative to the natural product, -bro-
mopironetin 2d had unique activity (entry 6). In the OVCAR5 cell
lines, the dose response curve for -bromopironetin 2d showed
biphasic character (Fig. 2). For this biphasic curve, the first inflec-
tion point occurs at a concentration approximately 3-times the
GI50 of pironetin (Table 1 entry 6); while the second point occurs
at double-digit micromolar concentrations. However, a biphasic
dose response curve was not observed when A2780 cell lines with
treated with analog 2d or with pironetin or the other analogs
(Fig. 2). Another interesting aspect for the dose-response curves
a-position of pironetin
synthesized via a selective olefination reaction of aldehyde 4. A
variety of olefination conditions have been reported for the synthe-
sis of tri-substituted olefins carrying an alkoxide, halide, or alkyl
a
a
a
moiety at the a
-position.30–35 An intermediate such as 4 could be
synthesized from aldehyde 5, which has been employed in previ-
a
ous syntheses of pironetin analogs in our group.18
The synthesis of analogs 2 began with the diastereoselective
aldol reaction between aldehyde 6 and thiazolidinethione 7 under
conditions similar to those reported by Crimmins23 and Marco20
for the synthesis of pironetin and related analogs (Scheme 2). Fol-
lowing protection of the secondary alcohol as the TBS ether, the
chiral auxiliary was cleaved with DIBAL-H to provide aldehyde
10. We performed the desired olefination of aldehyde 10 with flu-
orine containing phosphonate ester 11a, because the selective ole-
fination with aryl phosphonate esters containing functional groups
at the 2-position has been reported previously.30,33 The olefination
between aldehyde 10 and phosphonate ester 11a proceeded in 63%
yield to provide intermediate 12a. We subsequently carried out the
olefination reaction with the methyl, chlorine, and bromine con-
taining phosphonate esters 11b–11d to evaluate the SAR at the
a
a
a
a
a
a-position of pironetin. While the olefination with 2-methyl phos-
of a-bromopironetin 2d, was the percentage of cells remaining at
phonate ester 11b proceeded in high yield, we observed incom-
plete conversion for reactions with halogen-containing
phosphonate esters 11c and 11d. We completed the synthesis of
analogs 2 following the deprotection and lactonization of interme-
diates 12 under acidic conditions.
the high drug concentrations. In the dose-response curves of tubu-
lin-binding agents, paclitaxel, pironetin and related analogs in the
A2780 cell line, the dose-response curve plateaus at approximately
10–20% of the control population at the higher tested drug concen-
trations. In the dose response curves of a-bromopironetin 2d in the
We also sought to synthesize analogs containing an aryl group
at this position, since Michael acceptors containing electron-defi-
A2780 cell line, the highest doses of the analog resulted in <10% of
the control population; this was significantly lower than other
evaluated tubulin-binding agents paclitaxel and pironetin (Fig. 3).
cient aromatic groups at the
reversible covalent inhibitors.16 While we proposed a series of ana-
logs containing different aromatic groups at the -position via a
Suzuki coupling with -bromopironetin analog 2d, we initially
synthesized -phenyl analog 13 to determine if an aromatic group
a-position have been reported to be
In summary, we synthesized a-functionalized pironetin analogs
a
to evaluate structure-activity relationships of the C2 position since
substitution at this position could potentially decrease the natural
product’s off-target covalent adduct formation. The analogs
a
a
Scheme 1. Retrosynthesis of a-functionalized pironetin analogs 2 (R = EWG group, PG = protecting group).
Please cite this article in press as: Huang D.S., et al. Bioorg. Med. Chem. Lett. (2017), https://doi.org/10.1016/j.bmcl.2017.10.057

D.S. Huang et al. / Bioorganic & Medicinal Chemistry Letters xxx (2017) xxx–xxx
3
Scheme 2. Synthesis of analogs 2. Reagents and conditions: (a) TiCl4, DIPEA, NMP DCM, À78 °C to À50 °C, 76%; (b) TBSOTf, 2,6-lutidine, DCM, 0 °C to RT, 95%; (c) DIBAL-H,
DCM, 93%; (d) 11, NaH, THF, À78 °C, 19%–87%; (e) HCl, EtOH, RT, 46%–86%.
Scheme 3. Synthesis of
RT, 54%; (c) BF3ÁEt2O, CHCl3, 0 °C, 87%.
a
-phenyl analog 10. Reagents and conditions: (a) TBSOTf, 2,6-lutidine, DCM, À78 °C, 92%; (b) cat. [(tBu)3PPdBr]2, PhB(OH)2, K3PO4, THF:MeCN:H2O,
Please cite this article in press as: Huang D.S., et al. Bioorg. Med. Chem. Lett. (2017), https://doi.org/10.1016/j.bmcl.2017.10.057

4
D.S. Huang et al. / Bioorganic & Medicinal Chemistry Letters xxx (2017) xxx–xxx
Table 1
Antiproliferative activity of a-functionalized pironetin analogs.
Entry
Compound
R
GI50 value (mM, OVCAR5)a
GI50 value (mM, A2780)a
1
2
3
4
5
6
7
Paclitaxel
Pironetin (1)
–
H
F
Me
Cl
Br
Ph
0.0210 0.0020
0.0144 0.0008
3.24 0.07
0.00753 0.00157
0.0243 0.0009
5.11 0.36
10.2 0.4
4.48 0.02
0.901 0.074
>50
2a
2b
2c
2d
13
3.36 0.16
1.08 0.07
0.0449 0.0036, 26.2 1.6b
>50
a
Average of 2 experiments performed in triplicate SEM (n = 6).
0.0684 0.002 calculated GI50 value from fitting a sigmoidal standard dose-response curve to a biphasic graph.
b
Fig. 2. Dose-response curves of pironetin and synthesized analogs in drug-sensitive OVCAR5 ovarian cancer cell line.
Fig. 3. Dose-response curves of pironetin and synthesized analogs in drug-sensitive A2780 ovarian cancer cell line.
Please cite this article in press as: Huang D.S., et al. Bioorg. Med. Chem. Lett. (2017), https://doi.org/10.1016/j.bmcl.2017.10.057

D.S. Huang et al. / Bioorganic & Medicinal Chemistry Letters xxx (2017) xxx–xxx
2. Kobayashi S, Tsuchiya K, Harada T, et al. J Antibiot. 1994;47:697.
5
contained either a halogen or a methyl group at the
phenyl group at the -position was introduced following cross-
coupling reaction with -bromo analog 14. The addition of a func-
tional group at the -position of pironetin resulted in significantly
decreased antiproliferative activity of at least 100-fold for the
majority of the analogs. However, -bromopironetin 2d had only
a 3–40-fold decrease in activity relative to pironetin. The activity
of analog 2d relative to the other -functionalized pironetin ana-
logs could be due to the analog acting through a different mecha-
nism(s) of action in addition to binding to -tubulin. Further
a-position. A
3. Yoshida M, Matsui Y, Ikarashi Y, Usui T, Osada H, Wakasugi H. Anticancer Res.
2007;27:729.
a
a
4. Vogt A, McPherson PA, Shen XQ, et al. Chem Biol Drug Des. 2009;74:358.
5. Marco JA, Garcia-Pla J, Carda M, et al. Eur J Med Chem. 2011;46:1630.
6. Torijano-Gutierrez S, Vilanova C, Diaz-Oltra S, et al. Eur
a
J Org Chem.
2014;2014:2284.
a
7. Vilanova C, Díaz-Oltra S, Murga J, et al. J Med Chem. 2014;57:10391.
8. Dumontet C, Jordan MA. Nat Rev Drug Discovery. 2010;9:790.
9. Jordan MA, Wilson L. Nat Rev Cancer. 2004;4:253.
a
10. Kingston DGI. J Nat Prod. 2009;72:507.
11. Prota AE, Setter J, Waight AB, et al. J Mol Biol. 2016;428:2981.
12. Yang JH, Wang YX, Wang TJ, et al. Nat Commun. 2016;7.
13. Kondoh M, Usui T, Kobayashi S, et al. Cancer Lett. 1998;126:29.
14. Miller RM, Paavilainen VO, Krishnan S, Serafimova IM, Taunton J. J Am Chem Soc.
2013;135:5298.
a
studies will be required to identify potential mechanisms of action
of this analog.
15. Serafimova IM, Pufall MA, Krishnan S, et al. Nat Chem Biol. 2012;8:471.
16. Krishnan S, Miller RM, Tian B, Mullins RD, Jacobson MP, Taunton J. J Am Chem
Soc. 2014;136:12624.
Acknowledgements
17. Xia G, Chen W, Zhang J, et al. J Med Chem. 2014;57:9889.
18. Huang DS, Wong HL, Georg GI. ChemMedChem. 2017;12:520.
19. Lin J, Yue XY, Huang P, Cui DX, Qing FL. Synthesis. 2010;267.
20. Panos J, Diaz-Oltra S, Sanchez-Peris M, et al. Org Biomol Chem. 2013;11:5809.
21. Roldan S, Cardona A, Conesa L, et al. Org Biomol Chem. 2017;15:220.
22. Watanabe H, Watanabe H, Usui T, Kondoh M, Osada H, Kitahara T. J Antibiot.
2000;53:540.
23. Crimmins MT, Dechert AMR. Org Lett. 2009;11:1635.
24. Dias LC, de Oliveira LG, de Sousa MA. Org Lett. 2003;5:265.
25. Gurjar MK, Chakrabarti A, Rao AVR. Heterocycles. 1997;45:7.
26. Gurjar MK, Henri JT, Bose DS, Rao AVR. Tetrahedron Lett. 1996;37:6615.
27. Penning TD, Djuric SW, Haack RA, et al. Synth Commun. 1990;20:307.
28. Watanabe H, Watanabe H, Kitahara T. Tetrahedron Lett. 1998;39:8313.
29. Yadav JS, Ather H, Rao NV, Reddy MS, Prasad AR. Synlett. 2010;1205.
30. Ando K. J Org Chem. 1998;63:8411.
Support from the University of Minnesota through the Vince
and McKnight Endowed Chairs (G.I.G.) and the College of Pharmacy
is greatly appreciated. David Huang thanks the American Founda-
tion for Pharmaceutical Education for a predoctoral fellowship. We
thank Sara K. Coulup for performing the glutathione assay, Dr.
Mathew E. Cuellar of the Institute for Therapeutics Discovery and
Development for analytical support and Dr. Amy P.N. Skubitz in
the Department of Laboratory Medicine and Pathology at the
University of Minnesota for the OVCAR5 and A2780 cells.
A. Supplementary data
31. Karama UJ. Chem Synop Res. 2009;405.
32. Machleidt H, Wessendorf R. Justus Liebigs Ann Chem. 1964;674:1.
33. Olpp T, Bruckner R. Synthesis-Stuttgart. 2004;2135.
34. Seneci P, Leger I, Souchet M, Nadler G. Tetrahedron. 1997;53:17097.
35. Tago K, Kogen H. Tetrahedron. 2000;56:8825.
Supplementary data associated with this article can be found, in
the online version, at https://doi.org/10.1016/j.bmcl.2017.10.057.
References
1. Yoshida T, Koizumi K, Kawamura Y, Matsumoto K, Itazaki H. US Patent
5314911; 1993.
Please cite this article in press as: Huang D.S., et al. Bioorg. Med. Chem. Lett. (2017), https://doi.org/10.1016/j.bmcl.2017.10.057