Identification of KD5170: A novel mercaptoketone-based histone deacetylase inhibitor

Article abstract of DOI:10.1016/j.bmcl.2008.10.029

We report the identification of KD5170, a potent mercaptoketone-based Class I and II-histone deacetylase inhibitor that demonstrates broad spectrum cytotoxic activity against a range of human tumor-derived cell lines. KD5170 exhibits robust and sustained histone H3 hyperacetylation in HCT-116 xenograft tumors following single oral or iv dose and inhibition of tumor growth following chronic dosing.

Full text of DOI:10.1016/j.bmcl.2008.10.029

Bioorganic & Medicinal Chemistry Letters 18 (2008) 6093–6096
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Identification of KD5170: A novel mercaptoketone-based histone
deacetylase inhibitor
Joseph E. Payne a, Céline Bonnefous a, Christian A. Hassig b, Kent T. Symons b, Xin Guo c,
Phan-Manh Nguyen b, Tami Annable c, Paul L. Wash a, Timothy Z. Hoffman a, Tadimeti S. Rao c,
Andrew K. Shiau b, James W. Malecha a, Stewart A. Noble a, Jeffrey H. Hager c, Nicholas D. Smith a,
*
a Department of Chemistry, Kalypsys, Inc., 10420 Wateridge Circle, San Diego, CA 92121, USA
b Department of Biology, Kalypsys, Inc., 10420 Wateridge Circle, San Diego, CA 92121, USA
c Department of Pharmacology, Kalypsys, Inc., 10420 Wateridge Circle, San Diego, CA 92121, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
We report the identification of KD5170, a potent mercaptoketone-based Class I and II-histone deacetylase
inhibitor that demonstrates broad spectrum cytotoxic activity against a range of human tumor-derived
cell lines. KD5170 exhibits robust and sustained histone H3 hyperacetylation in HCT-116 xenograft
tumors following single oral or iv dose and inhibition of tumor growth following chronic dosing.
Ó 2008 Elsevier Ltd. All rights reserved.
Received 26 August 2008
Revised 6 October 2008
Accepted 7 October 2008
Available online 11 October 2008
Keywords:
Histone deacetylase inhibitor
HDAC
Mercaptoketone
Prodrug
HCT-116
KD5170
Sulfonamide
The acetylation state of the terminal NH2 of lysine residues in
the histone chromatin core is a key regulator of gene transcrip-
tion.1 Histone acetyltransferases (HATs) and histone deacytalases
(HDACs) maintain this balance and have been shown to act as tran-
scriptional coactivators and transcriptional repressors, respec-
tively.2 Recently HDACs have emerged as a promising new target
in cancer therapy as tumor cells up-regulate or recruit HDACs to
regions in the genome that harbor tumor suppressor genes, result-
ing in their silencing.3,4
The majority of HDAC inhibitors (HDACi), including the recently
approved SAHA (ZolinzaTM),5 use a hydroxamic acid motif to che-
late zinc in the active site of the enzyme (Fig. 1).6 However, this
motif typically confers undesired properties such as poor pharma-
cokinetics, off-target cross-reactivity and poor solubility.7 Addi-
tional HDACi in clinical trials include the benzamide class8 such
as SNDX-275 and the macrocyclic peptide FK-228,9,10 which
undergoes intracellular disulfide reduction to liberate an alkylthiol,
a functionality well established as a potent inhibitor of zinc-depen-
dent enzymatic processes (Fig. 1).7
KD5150 (Fig. 1) is a novel HDACi that was identified from an ul-
tra high-throughput screen aimed at identifying non-hydroxamate,
non-benzamide small-molecule HDACis.11 KD5150 is a thioacetate
prodrug that undergoes facile hydrolysis in vivo to release a mer-
captoketone that binds to the zinc in the active site of the HDAC
enzyme.12
Although KD5150 was effective at reducing tumor burden in a
mouse xenograft model following oral administration,11 it suffered
from poor aqueous solubility (0.2 mg/mL), precluding intra-venous
(iv) administration and perhaps sub-optimal oral bioavailability.
To improve the pharmaceutical potential of the series and allow
for iv dosing an effort was undertaken to increase the aqueous sol-
ubility of KD5150. In addition, it was felt that improved physico-
chemical properties would potentially increase the biological
activity of series, thereby increasing therapeutic potential. Herein,
we describe our attempts to improve the physicochemical proper-
ties of KD5150 leading to the identification of the clinical candidate
KD5170.
KD5150 contains a catechol group distal to zinc-binding mer-
captoketone group and it was proposed that this site could allow
for the incorporation groups to improve the water solubility via
an ether linkage. Compounds of this type were prepared as shown
in Scheme 1.
* Corresponding author.
E-mail address: Nicholas_Smith77@yahoo.com (N.D. Smith).
0960-894X/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2008.10.029

6094
J. E. Payne et al. / Bioorg. Med. Chem. Lett. 18 (2008) 6093–6096
Additionally, the compounds were tested for cytotoxicity against
the colorectal cancer line HCT-116 and the myeloma line U266.
KD5150 is a potent HDACi in the cell based assay with an
EC50 = 0.038
and 2.0 M against the U266 tumor lines (Table 1). Previous SAR
had shown that a methoxy substituent at the meta-position (entry
2) lost activity relative to KD5150 (EC50 = 0.19 M), whereas a
para-methoxy substation (entry 3) maintained activity
(EC50 = 0.029 M). Hence the para-position was chosen as the
lM and cytotoxicity of 1.2 lM against the HCT-116
l
l
l
primary focus for the attachment of functionality that would be
expected to improve aqueous solubility. Thus, incorporation of a
pyridine ring via a 1, 2, or 3 carbon chain (entries 4, 5, and 6) led
to a loss in potency in the cell-based assay (EC50 ꢀ 0.20
lM) com-
pared to the para-methoxy ether (entry 3). Similarly, use of a teth-
ered alcohol (entry 7) or ether (entry 8) gave a loss in potency to
(EC50 ꢀ 0.10
lM) and a significant decrease in cytotoxicity against
the HCT-116 and U266 cell lines relative to KD-5150.
Figure 1. HDAC Inhibitors.
However, the use of tethered aliphatic amines gave more
encouraging results. Thus although a 2-carbon linked dimethyl-
amino group (entry 9) eroded potency (EC50 = 0.093 lM) com-
pared to KD5150, extending the chain to 3- (entry 10) or 4-
carbons (entry 11) gave compounds with equivalent cell based
Aniline 4 was prepared from the commercially available acid
chloride 1 in 3 steps. Thus addition of the anion of dimethylmalo-
nate to acid chloride 1 gave keto-malonate 2 which was doubly
decarboxylated to give ketone 3. Finally, high pressure amination
with ammonia gave the desired aniline 4.
potencies (EC50 ꢀ 0.040
tethered dimethylamino group (entry 10), had an improved cyto-
toxicity against HCT-116 (EC50 = 0.14 M) and U266
(EC50 = 0.18 M) cell lines compared to KD5150. The SAR of 10
lM) to KD5150. Further, the 3-carbon
l
Aniline 4 was reacted with pipsyl iodide (5) in pyridine to give
sulfonamide 6. The iodide of 6 was readily derivatized with com-
mercial alcohols using copper catalyzed chemistry13 to give ethers
7 wherein groups aimed at improving aqueous solubility were ap-
pended at the para-position of the phenyl ring. Subsequently 7 was
converted to the mercaptoketone prodrug 8 using pyrrolidone
hydrotribromide in HBr/AcOH to affect ketone bromination (HBr/
AcOH was required to minimize ring bromination). Displacement
of the bromine with potassium thioacetate then gave the final thi-
oesters 8. Using this sequence, the structure–activity relationship
(SAR) of this region of the molecule was rapidly investigated with
a particular focus on the incorporation of groups that would im-
prove aqueous solubility. Compounds prepared in this manor were
tested for their pan-HDAC inhibitor activity by their ability to in-
duce histone H3 hyperacetylation in a HeLA cell-based assay.14
l
was examined more closely; thus cyclization of the dimethyl-
amino group of 10 by conversion to a pyrrolidinyl (entry 12),
piperidinyl (entry 13) or morpholinyl (entry 14) group led to a
slight loss in potency in the H3 cell-based assay (EC50 = 0.05–
0.07
HCT-116 and U266 lines. Restricting the confirmation of 10 as
in piperidines 15 (EC50 = 0.11 M) and 16 (EC50 = 0.044 M) led
lM) and a corresponding reduction of cytotoxicity against
l
l
to a general loss in potency, although conformation 16 was clearly
preferred to 15.
With robust potency in the histone H3 hyper-acetylation cell-
based assay and sub-micromolar activity against the HCT-116
and U-266 tumor lines, the dimethylamino-tethered analog 10
(forthwith KD5170) was selected for further profiling. As shown
Scheme 1. Synthesis of
a-thioketones incorporating a water solubilizing group. Reagents and Conditions: (a) Dimethylmalonate, MgCl2, Et3N, toluene, 5 h, 92%; (b) H2O,
DMSO, 130 °C, 2 h, 32%. (c) NH3, 120 °C, 10 h, 89%; (d) Pyridine, 60 °C, 1 h, 69%; (e) CuI (5 mol %), 1,10-phenanthralene (10 mol %), Cs2CO3 (2.5 equiv), ROH (20 equiv), 120 °C,
6 h, 40–80%; (f) i—Pyrrolidone hydrotribromide, HBr/AcOH, DMF, 50 °C, 2 h; ii—Potassium thioacetate, DMF, rt, 1 h, 45–70%.

J. E. Payne et al. / Bioorg. Med. Chem. Lett. 18 (2008) 6093–6096
6095
Table 1
ally, KD5170 showed minimal inhibition of the five major
cytochrome P450 enzymes (IC50 > 50 M) and minimal inhibition
of the hERG channel in cell-based patch clamp assays
(EC50 > 30 M).
Cell-based HDAC inhibition and cytotoxicity of selected compounds
l
l
To determine the cytotoxicity of KD5170 against cell lines de-
rived from a variety of human tumor types, KD5170 was screened
against the NCI panel of 60 human tumor-derived cell lines (NCI-
60). As can be seen in Table 3 from a selection of the results,14
KD5170 displayed micromolar to submicromolar activity against
a broad spectrum of tumor cell lines.15
Entry
R
Cell H3a EC50
M)
HCT-116 EC50
M)
U226 EC50
(lM)
(
l
(
l
A key assay utilized in the program was the pharmacodynamic
assessment of histone H3 hyperacetylation following single dose
administration of HCT-116 tumor-bearing nude mice. Briefly, mice
were dosed with compound po or iv, and tumors excised at 4 h and
fold increase in H3 hyperacetylation over control determined by
quantitative immunoblotting of tumor lysates.14 As shown in Fig-
ure 2, KD5170 (42 mg/kg po) induced robust histone H3 acetyla-
tion in HCT-116 xenograft tumors. Importantly, the magnitude of
this effect was greater that than observed with KD5150 (50 mg/
kg po), (9-fold versus 5-fold over control). Furthermore, as the
aqueous solubility of KD5170 allowed for iv administration, this
1
2
3
KD5150
m-oMe
p-oMe
0.038
0.190
0.029
1.2
4.3
0.54
2.0
NT
NT
4
5
6
0.22
0.42
0.75
NT
0.19
9.6
0.22
NT
NT
Table 2
0.11
0.12
0.093
8.6
NA
29
Aqueous solubility, CYP P450 and hERG channel inhibition for KD5150 and KD5170
7
8
9
Aq solubility (mg/mL)
EC50
M)
CYP inhibitionb
(
l
hERGd
(lM)
5.3
KD5170a
KD5150
10
0.2
>50
>30
>30
>50c
0.66
NT
a
HBr salt.
b
c
Against 3A4, 2D6, 2C9, 2C19, and 1A2.
2C9 = 10
Patch clamp.
lM.
d
0.045
0.029
0.061
0.14
0.28
0.82
0.18
0.12
0.4
10
11
Table 3
Selected cytotoxic activities of KD5170 against cell lines from the NCI-60 panel
Line
Origin
EC50
(lM)
Line
Origin
EC50 (lM)
HT29
MB435
H460
Colon
Breast
Lung
1.0
PC3
Prostate
Kidney
Leukemia
CTCL
0.74
0.59
0.39
0.17
0.18
0.23
0.76
768-0
HL60
MJ
12
13
SKOV3
Ovary
0.053
0.067
1.0
0.8
19
0.33
14
15
16
0.110
0.044
3.7
1.2
0.18
0.30
NT, not tested.
a
HeLa cell-based histone H3 hyperacetylation assay.
in Table 2, the HBr salt of KD5170 is highly soluble in aqueous
media (10 mg/mL) compared to KD5150 (<0.2 mg/mL). Addition-
Figure 2. Histone H3 hyperacetylation pharmacodynamic biomarker: comparison
of KD5150 (po) and KD5170 (po and iv).

6096
J. E. Payne et al. / Bioorg. Med. Chem. Lett. 18 (2008) 6093–6096
orally QOD in a 14-day xenograft, KD5170 demonstrated signifi-
cant tumor growth inhibition at 42 and 84 mg/kg with T/C values16
of 44% and 25%, respectively.
In summary, we have identified KD5170, a novel mercaptoke-
tone-based broad spectrum Class I and II-HDAC inhibitor with
potent activity in cell based assays that monitor histone acetyla-
tion, and broad spectrum activity against a variety of human tu-
mor cell lines. KD5170 exhibits high aqueous solubility and
when dosed orally or intravenously, gave a robust and sustained
pharmacodynamic response as monitored by histone H3 hyper-
acetylation in xenograft tumors. This robust pharmacodynamnic
response translated into tumor growth inhibition with chronic
oral administration of nude mice bearing HCT-116 xenograft tu-
mors. These data and others support the continued pre-clinical and
clinical development of KD5170 as an oncology therapeutic with po-
tential across a wide range of cancer indications.
Figure 3. KD5170 induces a robust and sustained pharmacodynamic response in
xenograft tumor tissues.
References and notes
1. Wang, G. G.; Allis, C. D.; Chi, P. Trends Mol. Med. 2007, 13, 363.
2. Xu, W. S.; Parmigiani, R. B.; Marks, P. A. Oncogene 2007, 26(37), 5541.
3. Gallinari, P.; Di, M. S.; Jones, P.; Pallaoro, M.; Steinkuhler, C. Cell Res. 2007, 17,
195.
4. Glozak, M. A.; Seto, E. Oncogene 2007, 26, 5420.
5. (a) Marks, P. A. Oncogene 2007, 26, 1351; (b) Marks, P. A.; Breslow, R. Nat.
Biotechnol. 2007, 25, 84.
6. Paris, M.; Porcelloni, M.; Binaschi, M.; Fattori, D. J. Med. Chem. 2008, 51(6), 1505.
7. Suzuki, T.; Miyata, N. Curr. Med. Chem. 2005, 12, 2867.
8. (a) Bolden, J. E.; Peart, M. J.; Johnstone, R. W. Nat. Rev. Drug Discov. 2006, 5, 769;
(b) Zhou, N.; Moradei, O.; Raeppel, S.; Leit, S.; Frechette, S.; Gaudetee, F.;
Paquin, I.; Bernstein, N.; Bouchain, G.; Vaisburg, A.; Jin, Z.; Gillespie, J.; Wang, J.;
Fournel, M.; Yan, P. T.; Trachy-Bourget, M. C.; Kalita, A.; Lu, A.; Rahil, J.;
MacLeod, R.; Li, Z.; Besterman, J. M.; Delormw, D. J. Med. Chem. 2008, 51, 4072.
9. Konstantinopoulos, P. A.; Vanoros, G. P.; Papavassiliou, A. G. Cancer Chemother.
Pharmacol. 2006, 58, 711.
10. Furumai, R.; Matsuyama, A.; Kobashi, N.; Lee, K. H.; Nishiyama, M.; Nakajima,
H.; Tanaka, A.; Komatsu, Y.; Nishino, N.; Yoshida, M.; Horinouchi, S. Cancer Res.
2002, 62, 4916.
11. Wash, P. L.; Hoffman, T. Z.; Wiley, B.M.; Bonnefous, C.; Smith, N. D.; Sertic, M.
S.; Lawrence, C. M.; Hassig, C. A.; Symons, K. T.; Guo, X.; Nguyen, P.; Annable,
T.; Noble, S. A.; Hager, J. H.; Malecha, J. W. Bioorg. Med. Chem Lett., 2008
(preceding paper).
Figure 4. KD-5170 demonstrates tumor growth regression in HCT-116 tumor
bearing mice (po dosing).
dosing route was examined and gave a similar robust pharmacody-
namic response (12-fold over control).
12. The binding mode of the mercaptoketone maybe mono- or bi-dentate:
Vanommeslaeghe, K.; Loverix, S.; Geerlings, P.; Tourwe, D. Bioorg. Med. Chem.
Lett. 2005, 13, 6070.
The time course of histone H3 hyperacetylation in HCT-116 tu-
mors was also examined as shown in Figure 3. Following a 30 mg/
kg oral dose of KD5170, tumor H3 hyperacetylation rapidly in-
creased to reach a maximum at 4 h. Significant hyperacetylation
was also observed at 8 hours with levels decreasing by 24 h
(although still above T0 levels).
Finally, having demonstrated a sustained pharmacodynamic re-
sponse, KD5170 was examined for its anti-tumor potential in HCT-
116 tumor bearing nude mice. As shown in Figure 4, when dosed
13. Wolter, M.; Nordmann, G.; Job, G. E.; Buchwald, S. L. Org. Lett. 2002, 4, 973.
14. Hassig, C. A.; Symons, K. T.; Guo, X.; Nguyen, P.; Annable, T.; Wash, P. L.; Payne,
J. E.; Jenkins, D. A.; Bonnefous, C.; Trotter, C.; Wang, Y.; Anzola, J. V.; Milkova, E.
L.; Hoffman, T. Z.; Dozier, S. J.; Wiley, B. M.; Saven, A.; Malecha, J. W.; Davis, R.
L.; Shiau, A. K.; Noble, S. A.; Rao, T. S.; Smith, N. D.; Hager, J. H. Mol. Cancer Ther.
2008, 7, 1054.
15. The cytotoxic activities (EC50) ranged from 0.1 to 7.7
median of 1.6 and 0.9 M, respectively.
lM with a mean and
l
16. T/C: (Treated final volume À Treated initial volume)/(Control final
volume À Control initial volume) Â 100.
A T/C value of 0 equals tumor
stasis.