Design and synthesis of 2-amino-isoxazolopyridines as Polo-like kinase inhibitors

Article abstract of DOI:10.1016/j.bmcl.2008.08.091

A series of 2-amino-isoxazolopyridines was designed and synthesized as Polo-like kinase (Plk) inhibitors. Key SAR and crystallographic data are discussed. More advanced analogues inhibit Plk1 with good enzymatic activity and modest cell-based activity. Differential selectivity among the three Plk isoforms is observed.

Full text of DOI:10.1016/j.bmcl.2008.08.091

Bioorganic & Medicinal Chemistry Letters 18 (2008) 5186–5189
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Design and synthesis of 2-amino-isoxazolopyridines as Polo-like
kinase inhibitors
,
*
Emily J. Hanan , Raymond V. Fucini , Michael J. Romanowski, Robert A. Elling, Willard Lew,
Hans E. Purkey, Erica C. VanderPorten, Wenjin Yang
Sunesis Pharmaceuticals, Inc., 395 Oyster Point Boulevard Suite 400, South San Francisco, CA 94080, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
A series of 2-amino-isoxazolopyridines was designed and synthesized as Polo-like kinase (Plk) inhibitors.
Key SAR and crystallographic data are discussed. More advanced analogues inhibit Plk1 with good enzy-
matic activity and modest cell-based activity. Differential selectivity among the three Plk isoforms is
observed.
Received 12 June 2008
Revised 24 August 2008
Accepted 26 August 2008
Available online 29 August 2008
Ó 2008 Elsevier Ltd. All rights reserved.
Keywords:
2-Amino-isoxazolopyridines
Polo-like kinase (Plk)
SAR
Kinase inhibitor
Polo-like kinases (Plks) are a family of serine-threonine kinases
that play fundamental roles in regulating cell division.1–3 Exposure
to Plk inhibitors leads to multiple defects in mitosis, causing cell
death.1,4–6 These enzymes present promising targets for cancer
therapy due to frequent overexpression in various human tu-
mors.1,4,7 As part of our drug discovery efforts, we identified and
developed pyrazolopyridine-based compounds 1 and 2 as a new
class of Plk inhibitors (Fig. 1).8 Additional work on developing
the core structure of this series led to the identification of several
alternate series; in this communication we report the synthesis
and SAR of novel 2-amino-isoxazolopyridine-based compounds. A
co-crystal structure of an isoxazolopyridine inhibitor was obtained
in complex with zebrafish Plk1 (zPlk1), a homolog possessing a sin-
gle conserved substitution for a residue lining the active site com-
pared to human Plk1. The phenotype of isoxazolopyridine
inhibitor-treated cells is consistent with Plk1 inhibition.
After the discovery of 1, compound 3 was designed as a varia-
tion on the core structure, with the goal of identifying a distinct
chemical series that could provide improvements in activity and
physicochemical properties. With the observation that 3 exhibits
low micromolar biochemical activity, the compounds in Table 1
were synthesized to develop initial structure–activity relationships
(SAR) around the right hand phenyl ring. Compounds 4–9 were
prepared from 2,5-dichloroisonicotinic acid as shown in Scheme
1. Following esterification of the carboxylic acid, ketones were gen-
erated using aryl Grignard reagents at low temperature. The isox-
azole ring was formed by heating the ketones in the presence of
hydroxylamine and potassium hydroxide.9 Amination of the pyri-
dine ring took place under standard Buchwald–Hartwig
conditions.10
Compounds 13 and 15 were prepared as shown in Scheme 2. A
Heck reaction of 1,3-dibromo-5-chlorobenzene and allyloxy-tert-
butyl-dimethylsilane followed by rhodium-catalyzed reduction
provided compound 10.11 Halogen–lithium exchange and reaction
with 2,5-dichloro-isonicotinic acid methyl ester gave ketone 11.12
Cyclization and amination proceeded as described above, and
deprotection of the alcohol with tetrabutylammonium fluoride
provided compound 13. Oxidation to the corresponding acid 14,
followed by acyl chloride formation and reaction with ammonia
gave the desired amide 15.
The data in Table 1 indicate that simple substitution around the
right hand phenyl ring is tolerated, with substitution favored at the
3-position (compounds 4 and 7). Bis-methoxy substitution at the
* Corresponding author. Tel.: +1 650 266 3500; fax: +1 650 266 3505.
E-mail address: info@sunesis.com (E.J. Hanan).
Figure 1. Plk1 inhibitors that served as a starting point for 3. Biochemical Plk1
These authors contributed equally to this work.
activity: 1 = 1.301 lM; 2 = 0.121 lM.
0960-894X/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2008.08.091

E. J. Hanan et al. / Bioorg. Med. Chem. Lett. 18 (2008) 5186–5189
5187
Table 1
Based on a co-crystal structure of a pyrazolopyridine analog de-
rived from 1 with zPlk1, we theorized that this strategy would al-
low us to engage Asp180 (hPlk1 Asp194) of the DFG-loop.8 As
shown by 13 and 15, this strategy delivered good improvements
in biochemical activity (15–30-fold improvement over parent 3).
A co-crystal structure of 13 with zPlk1 confirms that the propanol
chain is situated in a hydrophobic pocket contributes additional
affinity through lipophilic interactions, and the terminal hydroxyl
group is clearly within distance to engage Asp180 through
water-mediated hydrogen-bonding (Fig. 2).13
Plk1 inhibition leads to a mitotic arrest and cell death, thus we
tested the effects of inhibitors on cell cycle progression in the HCT
116 colorectal cancer cell line. Mitotic arrest was determined as a
doubling of DNA content (4N) as measured by a high-content cell
screening (HCS) assay. As indicated in Table 1, compounds (e.g., 4
and 7) with comparable enzymatic activity show variable ability
to arrest the cell cycle as measured in the HCS assay. Physicochem-
ical or pharmacokinetic properties such as solubility, cell-perme-
ability or efflux are not entirely sufficient to explain these
observations. All compounds tested were moderately to highly cell
permeable in an MDCK assay and moderately to highly stable in
human liver microsomes. It is possible, at least in part, that bio-
SAR for phenyl substitution of 3a
Compound
R
Plk1 IC50
(lM)
HCS IC50 (l
M)b
3
4
5
6
7
8
9
13
15
H
3-CH3
4-CH3
3,5-CH3
3-Cl
3,4-OCH3
3,4-Methylenedioxy
3-Cl-5-(3-propanol)
3-Cl-5-(3-propionamide)
1.616
0.549
1.170
0.330
0.625
4.927
0.214
0.099
0.051
>20
6.69
>20
>20
>20
>20
7.32
10.16
8.35
a
See Supplementary Data for assay details.
Mitotic arrest as measured by doubling of DNA content in high-content
b
screening (HCS) of HCT 116 cells.
chemical potency <1 lM prerequisites cell activity for this series.
Compounds 18–23 were synthesized to determine if any ana-
logs without a direct phenyl attachment could be identified with
Scheme 1. Reagents and conditions: (a) TMSCHN2, MeOH/CH2Cl2, 25 °C, 0.5 h,
100%; (b) ArMgBr or ArMgCl, THF, À78 °C, 25–90%; (c) NH2OHÁHCl, KOH, iPrOH/H2O,
80 °C, 40–60%; (d) (S)-a-methyl-benzylamine, Pd2dba3, 1,3-bis(2,6-diisopropyl-
phenyl)imidazolium chloride, NaOtBu, toluene, 110 °C, 15–24 h, 20–50%.
3- and 4-positions (compound 8) results in a slight decrease in
activity compared to the parent compound, but tying the methy-
lene groups together (as in compound 9) results in a 7.5-fold
improvement in Plk1 inhibition over the parent compound 3. Sim-
ple bis-substitution at the 3- and 5-positions is also well tolerated
(compound 6), which led us to maintain chloro-substitution at the
3-position and extend the substituent at the 5-position.
Figure 2. Structure of isoxazolopyridine 13 in the zPlk1 active site.
Scheme 2. Reagents and conditions: (a) allyloxy-tert-butyl-dimethylsilane, Pd(OAc)2, PPh3, TEA, DMF, 80 °C; (b) H2, Rh/Al2O3, EtOAc/iPrOH, 25 °C, 24 h, 85% (2 steps);
(c) tBuLi, 2,5-dichloro-isonicotinic acid methyl ester, À78 °C, 78%; (d) NH2OHÁHCl, KOH, iPrOH/H2O, 80 °C; (e) TBSCl, Et3N, DCE, 60 °C, 78% (2 steps); (f) (S)-
a-methyl-
benzylamine, Pd2dba3, 1,3-bis(2,6-diisopropylphenyl)imidazolium chloride, NaOtBu, toluene, 110 °C, 7 h, 70%; (g) TBAF, THF, 25 °C, 2 h, 70%; (h) Jones Reagent, acetone, 0 °C,
1 h, 56%; (i) oxalyl chloride, DMF, DCM, 25 °C, 0.5 h; (j) ammonia, Et3N, dioxane, 25 °C, 1 h, 51% (2 steps).

5188
E. J. Hanan et al. / Bioorg. Med. Chem. Lett. 18 (2008) 5186–5189
Table 2
SAR for phenyl replacementsa
Compound
R
Plk1 IC50 (lM)
18
19
20
21
23
Benzyl
2-Thiophene
3-(4-Methyl)thiophene
2-Methyl-cyclohexane
3-(N-3-Propionamide)piperidinyl
>20
3.007
0.779
5.689
7.479
Scheme 3. Reagents and conditions: (a) i—LDA, THF, 0.5 h, À78 °C; ii—methyl-
phenyl acetate, 0.5 h, À78 °C, 45%; (b) NH2OHÁHCl, pyridine, 2 h, 100 °C; (c) NaOH,
a
H2O, 3 h, 100 °C, 10% (2 steps); (d) (S)-
NaOtBu, toluene, 80 °C, 15 h, 5%.
a
-methyl-benzylamine, Pd2dba3, Peppsi-iPr,
See Supplementary Data for assay details.
activity (19 is comparable in activity to 3, 20 is comparable to 4).
Saturated rings were not as well tolerated, neither compound 21
nor 23 providing any improvement over initial hit 3. None of the
compounds described in Table 2 exhibited measurable cell-based
improved activity. The synthesis of compound 18 is described in
Scheme 3. 2,5-Dichloropyridine was lithiated and quenched with
methylphenylacetate to provide 16.12 Treatment with hydroxyl-
amine in pyridine generated the oxime, which cyclized to the isox-
azole 17 when treated with aqueous sodium hydroxide.9
Amination under standard conditions provided the desired com-
pound 18 (Scheme 3). Thiophenes 19 and 20 were generated as de-
scribed in Scheme 4; reaction of lithiated thiophenes with pyridine
methyl esters gave the desired ketones, which were cyclized and
aminated to provide the final compounds. Compound 21 was syn-
thesized in an analogous manner to 18, starting with 2-methyl-1-
cyclohexane-carboxylic acid. Intermediate 22 was also synthesized
in the same fashion. Deprotection of 22 followed by alkylation with
3-bromopropionamide provided compound 23 (Scheme 5).
activity (>20 lM in the cell-cycle assay).
An examination of alternate amino groups appended to the pyr-
idine ring revealed that limited substitution on the benzyl ring was
tolerated, with only fluoro substituents exhibiting comparable
activity and all other substitution resulting in inactive compounds.
A variety of substituted anilines provided activity within a few fold
of that seen with the benzyl amine, including typical solubilizing
group appendages. None of these variations provided a significant
advantage over the (S)-a-methyl-benzyl amine (data not shown).
Next, we assessed Plk-isoform selectivity of this series by profil-
ing biochemical activities for several compounds against human
Plk1, Plk2, and Plk3. As shown in Table 3, three structurally similar
compounds all show distinct selectivity profiles. Compound 4
Table 2 summarizes the structure–activity relationships for the
inhibition of Plk1 for these non-phenyl analogs. Appending a ben-
zyl group to the isoxazole ring ablated all activity (compound 18).
The heterocyclic compounds 19 and 20 demonstrate that the phe-
nyl ring may be exchanged for a thiophene with minimal loss of
inhibits Plk1 and Plk2 in the 1
against Plk3 up to 10 M. The 3,4-methylenedioxy version (com-
pound 9) is highly active against Plk2 (0.004 M), with 6-fold
selectivity against Plk3 and 50-fold selectivity against Plk1. The
most active compound against Plk1 (15, 0.051 M), is 3-fold less
lM range, but shows no inhibition
l
l
l
active against Plk2 and 27-fold less active against Plk3. Interest-
ingly, all three of these compounds result in comparable inhibition
in the cell-cycle assay, consistent with the role of these isoforms in
mediating cell cycle progression. (Comparable activity is also seen
in MTT proliferation assays; data not shown.) Further confirmation
for inhibition of Plk1 is seen in HCT 116 cells treated with com-
pound 15, where the expected monopolar spindle phenotype is ob-
served (Supplementary Data, Fig. S1).
In conclusion, we have developed a novel 2-amino-isoxazolo-
pyridine series as Plk inhibitors. Starting from compound 3 with
modest biochemical potency we identified a series of Plk inhibitors
with good biochemical potency and modest cell-based activity. A
co-crystal structure confirmed that these compounds bind to
Plk1 in an ATP-competitive manner. The phenotype observed in
cells treated with these inhibitors is consistent with inhibition of
Plk1. Lastly, we identified compounds differentially selective for
Plk1, 2, and 3, which could serve as useful tools for elucidating
the specific cellular functions of these Plk isoforms.
Scheme 4. Reagents and conditions: (a) TMSCHN2, MeOH/CH2Cl2, 25 °C, 0.5 h,
100%; (b) bromo-thiophene, tBuLi, THF, À78 °C, 70–85%; (c) NH2OHÁHCl, KOH,
iPrOH/H2O, 80 °C, 10–40%; (d) (S)-
a-methyl-benzylamine, Pd2dba3, 1,3-bis(2,6-
diisopropylphenyl)imidazolium chloride, NaOtBu, toluene, 110 °C, 15–24 h, 10–30%.
Table 3
Plk isoform profiles of selected compoundsa
Compound
Plk1 IC50 (lM)
Plk2 IC50 (
lM)
Plk3 IC50 (lM)
4
9
15
0.549
0.214
0.051
1.449
0.004
0.172
>10
0.024
1.382
Scheme 5. Reagents and conditions: (a) HCl, dioxane, 25 °C, 3 h, 100%; (b) 3-
bromopropionamide, Et3N, THF, 40 °C, 24 h, 14%.
a
See Supplementary Data for assay details.

E. J. Hanan et al. / Bioorg. Med. Chem. Lett. 18 (2008) 5186–5189
5189
8. This work reported in a separate communication: ‘Design and synthesis of 2-
amino-pyrazolopyridines as Polo-like kinase inhibitors’, Fucini, R. V.; Hanan, E.
J.; Romnowski, M. J.; Elling, R. A.; Lew, W.; Barr, K. J.; Zhu, J.; Yoburn, J. C.; Liu,
Y.; Fahr, B. T.; Fan, J.; Lu, Y.; Pham, P.; Choong, I. C.; VanderPorten, E. C.; Bui, M.;
Purkey, H. E.; Evanchik, M. J.; Yang, W. Bioorg. Med. Chem. Lett. 2008,
doi:10.1016/j.bmcl.2008.08.095.
Supplementary data
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.bmcl.2008.08.091.
9. Strupczewski, J. T.; Allen, R. C.; Gardner, B. A.; Schmid, B. L.; Stache, U.;
Glamkowski, E. J.; Jones, M. C.; Ellis, D. B.; Huger, F. P.; Dunn, R. W. J. Med. Chem.
1985, 28, 761.
References and notes
10. (a) Lee, S.; Hartwig, J. F. J. Org. Chem. 2001, 66, 3402; (b) Grasa, G. A.; Miciu, M.
S.; Huang, J.; Nolan, S. P. J. Org. Chem. 2001, 66, 7729; (c) Organ, M. G.; Abdel-
Hadi, M.; Avola, S.; Dubovyk, I.; Hadei, N.; Kantchev, E. A. B.; O’Brien, C. J.;
Sayah, M.; Valente, C. Chem. Eur. J. 2008, 14, 2443.
1. Eckerdt, F.; Yuan, J.; Strebhardt, K. Oncogene 2005, 24, 267.
2. Lowery, D. M.; Lim, D.; Yaffe, M. B. Oncogene 2005, 24, 248.
3. Xie, S.; Xie, B.; Lee, M. Y.; Dai, W. Oncogene 2005, 24, 277.
4. Strebhardt, K.; Ullrich, A. Nat. Rev. Cancer 2006, 6, 321.
5. Plyte, S.; Musacchio, A. Curr. Biol. 2007, 17, R280.
6. Peters, U.; Cherian, J.; Kim, J. H.; Kwok, B. H.; Kapoor, T. M. Nat. Chem. Biol. 2006,
2, 618.
7. Takai, N.; Hamanaka, R.; Yoshimatsu, J.; Miyakawa, I. Oncogene 2005, 24, 287.
11. Ziegler, C. B.; Heck, R. F. J. Org. Chem. 1978, 43, 2941.
12. Voisin, A. S.; Bouillon, A.; Lancelot, J. C.; Rault, S. Tetrahedron 2005, 61, 1417.
13. Expression, purification, and crystallization of Plk1: Elling, R. A.; Fucini, R. V.;
Romanowski, M. J. Acta Cryst. 2008, 64, 909. Details for this X-ray crystal
structure are provided as Supplementary Data.