Design and synthesis of novel pyrimidine hydroxamic acid inhibitors of histone deacetylases

Article abstract of DOI:10.1016/j.bmcl.2010.09.016

Inhibition of histone deacetylase activity represents a promising new modality in the treatment of a number of cancers. A novel HDAC series demonstrating inhibitory activity in cell proliferation assays is described. Optimisation based on the introduction of basic amine linkers to effect good drug distribution to tumour led to the identification of a compound with oral activity in a human colon cancer xenograft study associated with increased histone H3 acetylation in tumour tissue.

Full text of DOI:10.1016/j.bmcl.2010.09.016

Bioorganic & Medicinal Chemistry Letters 20 (2010) 6657–6660
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Design and synthesis of novel pyrimidine hydroxamic acid inhibitors
of histone deacetylases
Alastair D. G. Donald a, , Vanessa L. Clark a, Sanjay Patel a, Francesca A. Day a, Martin G. Rowlands b,
⇑
Judata Wibata b, Lindsay Stimson b, Anthea Hardcastle b, Sue A. Eccles b, Deborah McNamara a,
Lindsey A. Needham a, Florence I. Raynaud b, Wynne Aherne b, David F. Moffat a
a Chroma Therapeutics, 93 Milton Park, Abingdon, Oxfordshire OX14 4RY, UK
b Cancer Research UK Cancer Therapeutics Unit, The Institute of Cancer Research, 15 Cotswold Road, Sutton, Surrey SM2 5NG, UK
a r t i c l e i n f o
a b s t r a c t
Article history:
Inhibition of histone deacetylase activity represents a promising new modality in the treatment of a num-
ber of cancers. A novel HDAC series demonstrating inhibitory activity in cell proliferation assays is
described. Optimisation based on the introduction of basic amine linkers to effect good drug distribution
to tumour led to the identification of a compound with oral activity in a human colon cancer xenograft
study associated with increased histone H3 acetylation in tumour tissue.
Received 13 August 2010
Revised 2 September 2010
Accepted 2 September 2010
Available online 21 September 2010
Ó 2010 Elsevier Ltd. All rights reserved.
Keywords:
HDAC
Cancer
Optimisation
Histone deacetylases (HDACs) are a family of epigenetic regula-
tor enzymes whose inhibition leads to a reversal of tumour suppres-
sor silencing, cell cycle arrest, differentiation, and/or apoptosis.
Because of these effects, which manifest themselves most strikingly
in transformed cells, HDAC inhibition has emerged as a strategy to
reverse aberrant epigenetic changes associated with cancer.1 Two
such agents have been approved for use in the treatment of cutane-
ous T-cell lymphoma, ZolinzaÒ (Vorinostat/SAHA, Merck),2 and Isto-
daxÒ (Romidepsin, Gloucester Pharmaceuticals/Celgene). A number
of additional HDAC inhibitors have shown anti-tumour activity in
xenograft models and are now in clinical trials.3 A wealth of preclin-
ical data suggests that HDAC inhibitors have potential against a wide
range of additional tumour types either alone or in combination
with other therapies.4
The HDACs comprise three separate classes of enzyme. Both
class I and class II enzymes are metalloenzymes containing a cata-
lytic zinc atom. There are thought to be at least 11 isoforms of the
metalloenzyme HDAC proteins: class I HDACs (HDAC 1, 2, 3 and 8)
are ubiquitously expressed and generally detected in the nucleus
while class II HDACs (HDAC 4, 5, 6, 7, 9 and 10) can shuttle be-
tween the cytoplasm and nucleus.5 HDAC 11 has a related catalytic
domain but has only weak homology with the others. The class III
enzymes are not metalloenzymes and, while relevant to certain
human diseases, are not yet strongly implicated in cancer.6
The major objective of our discovery efforts towards therapeu-
tically useful small molecule HDAC inhibitors was to achieve good
intra-tumoural distribution of drug via oral administration.7,8
Based on an analysis of all previously described HDAC inhibitor
families, addressing their ability to meet our goals of robust phys-
icochemical properties to facilitate improved cell permeability and
oral dosing, and to provide potential for chemical novelty, we
decided to utilise the pyrimidine hydroxamate moiety as the
zinc-binding group (ZBG) in our approach. This seemed likely to
be the most metabolically stable functionality that would be com-
patible with good HDAC inhibitory activity.9,10 Such hydroxamic
acids have been previously described as potent inhibitors of his-
tone deacetylases (e.g., R306465/JNJ-16241199).11
Docking studies to an available structure of the class 1 isoform
HDAC8, where the pyrimidine hydroxamate unit co-ordinated to
the zinc ion at the base of the catalytic site were performed to
identify linker groups that could potentially provide access to the
hydrophobic region at the entrance of the active site.12,13 These
studies suggested fused bicyclic structures such as 2-(hexahydro-
pyrrolo[3,4-c]pyrrol-2-(1H)-yl) substituted pyrimidines, as they
possessed the required length to provide good interaction between
aryl or heteroaryl substituents and the hydrophobic enzyme sur-
face (Fig. 1).
After identification of potential groups for linking to the pyrim-
idine-based ZBG, our studies focused on exploring the hydrophobic
cap group of the 2-(hexahydropyrrolo[3,4-c]pyrrol-2-(1H)-yl)
system. Such compounds were readily accessible from a common
intermediate 3, which was prepared as shown in Scheme 1.
⇑
Corresponding author. Tel.: +44 1235 829165.
E-mail address: adonald@chromatherapeutics.com (A.D.G. Donald).
0960-894X/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2010.09.016

6658
A. D. G. Donald et al. / Bioorg. Med. Chem. Lett. 20 (2010) 6657–6660
Table 1
O
Inhibition of HDAC activity (IC50) for compounds 4–14
HO
NH
N
a
Compd
X
Y
R
IC50 (nM)
H
N
N
SAHA
4
5
6
7
—
—
87 13
64
SO2
SO2
SO2
CO
CO
—
CH2
CH2
CH2
CH2
CH2
N
N
N
N
N
N
N
N
N
N
C
2-Naphthyl
NH
H
3,5-Bis-(trifluoromethyl)phenyl
4-Trifluoromethoxy-phenyl
2-Naphthyl
4-Phenoxypropyl
5-Methoxyindan-1-yl
3,5-Difluorophenyl
4-Chlorophenyl
4-Methoxyphenyl
2-Naphthyl
433
123
Figure 1. Fused bicyclic diamine scaffold identified as a potential core structure for
further elaboration into novel inhibitors of HDACs.
12
2
8
9
122
30 2.9
50
10
11
12
13
14
42
95
H
H
N
N
13 1.9
45
SO2Me
N
H
H
2-Naphthyl
H
H
a, b
N
O
a
Values are mean of three experiments. HDAC activity was measured using Fluor
N
+
N
De Lys reagents (Biomol) and a commercially available HeLa cell nuclear extract
(Cilbiotech s.a.). Values are the mean of two independent determinations or
mean SD for n > 2 determinations.
N
O
N
O
O
OEt
OEt
Table 2
1
2
3
Growth inhibition GI50 for compounds 3–14 in human tumour cell lines
Scheme 1. Reagents and conditions: (a) K2CO3, acetonitrile, room temperature, 1 h
(58%) and (b) 4 M HCl, dioxane, room temperature, 1.5 h (99%).
Compd
HCT116
GI50 (nM)
HeLa
GI50 (nM)
U937
GI50 (nM)
HUT78
GI50 (nM)
a
a
a
a
SAHA
4
5
6
7
750
740
1700
530
590
1800
500
530
490
900
120
700
1460
1132
42% @10
940
210
3950
230
940
690
950
260
390
480
1250
980
390
900
390
300
230
550
72
300
300
700
500
200
800
208
202
190
290
79
Condensation of the commercially available mono-protected
diamine 1 with known pyrimidine 214 followed by acidic deprotec-
tion of the Boc group yielded an intermediate 3 suitable for instal-
lation of a variety of hydrophobic cap groups. Reaction of 3 with a
sulfonyl chloride, acid chloride or aldehyde yielded the required
functionalized pyrimidine ester (Scheme 2).
Conversion of the ester into the hydroxamic acid was achieved
by a three-step procedure—alkaline hydrolysis of the ester, cou-
pling of the resultant acid with an O-protected hydroxylamine
and acidic deprotection to yield the desired hydroxamic acids
4–14.15,16
lM
8
9
10
11
12
13
14
1260
450
340
a
Values are means of three experiments. Growth inhibition was measured using
72 h SRB or MTT assays.
The HDAC inhibitory activities of these compounds are given in
Table 1. Sulfonamide substituted compounds 4–6 showed encour-
aging activity in assays of inhibition of HDAC activity (comparable
to SAHA), but showed modest to poor activity in cell-based assays
of anti-proliferative activity (Table 2).17
Expansion of the SAR to examine the role of other linkers led to
the synthesis of amide substituted compounds (7 and 8) and ter-
tiary amine compounds (9–13). In the naphthalene substituted ser-
ies of compounds (4, 7 and 13), enzyme activity improved
approximately fivefold whilst cellular activity improved by be-
tween 4 and 6-fold (dependent on cell line). A strong preference
for large, lipophilic substituents is evident, with such compounds
demonstrating enhanced potency compared to monocyclic aro-
matic systems such as 11. The two methoxy substituted com-
pounds (9 and 12) also demonstrate this trend; the additional
ring of indane 9 confers a 2–3-fold potency advantage over the ba-
sic 4-methoxybenzyl system of 12. Surprisingly we found that the
pyridine hydroxamic acid 14 was significantly less active in both
enzyme and cellular assays than its direct pyrimidine analogue
13.18
Naphthyl substituted amine 13 emerged as a promising lead
compound due to the desirable combination of excellent potency
and drug-like molecular properties (molecular weight 389.4,
c Log P 1.3, 6 H-bond acceptors, 2 H-bond donors, four rotatable
bonds), and was selected for further in vitro and in vivo studies.
We further hypothesised that the basic nature of the linker unit
of 13 could aid our objective of enhanced drug distribution to
tumour.
R1
X
N
H
N
H
H
H
H
a, b or c
N
N
Incubations of 13 (10 lM) with human and mouse liver micro-
somes as previously described19 indicated acceptable metabolic sta-
bility (72% and 45% of 13 remaining after 30 min, respectively). The
disappointingly low cell permeability (Papp 0.5 ꢀ 10ꢁ6 cm/s), as as-
sessed by PAMPA was identical to that of SAHA. There was no signif-
icant inhibition of cytochrome P450 isoforms 1A2, 2A6, 2C8, 2C19
then d,e, f
N
O
Y
N
O
Y
O
NH
OH
and 3A4 at 50
CYP2D6 was observed.20 Inhibition of the hERG channel by 13 was
modest (4.1 M).
lM but a moderate inhibition (IC50 = 10–50 lM) of
3
4-14
l
Scheme 2. Reagents and conditions: (a) sulfonyl chloride, pyridine (51–66%); (b)
acid chloride, pyridine (80–89%); (c) aldehyde, DCE, sodium triacetoxyborohydride
(71–95%); (d) 6 M NaOH, ethanol, 80 °C (81–100%); (e) O-(1-isobutoxyeth-
yl)hydroxylamine, EDC, HOBt, DIPEA, DCM; (f) TFA, DCM, MeOH (5–90%, two steps).
Cassette PK studies in CD1 mice (3 mg/kg po, 1 mg/kg iv) indi-
cated that 13 was orally available in mice (F = 15%) and distributed
well into tissues (spleen Cmax 218 ng/ml ffi 0.56
lM, plasma Cmax

A. D. G. Donald et al. / Bioorg. Med. Chem. Lett. 20 (2010) 6657–6660
6659
31 ng/ml), and thus a mouse xenograft study was undertaken. Fe-
male CrTac:NCr-Fox1(nu) athymic mice (n = 14) with established
bilateral HCT116 colon carcinoma xenografts were dosed orally,
once-daily for 16 days with 50 mg/kg of 13 (Fig. 2).21
At this dose, 13 was well tolerated as evidenced by an increase
in mean body weight of 7%, although inhibition of tumour growth
was modest with a final tumour weight of 80% of control and a
reduction in mean tumour volume of 28%. Measured peak tumour
concentration of 13 1 h post dosing was 215 75 nmol/kg (SEM)
(1.6 times cellular IC50), whilst the plasma concentration at this
time point was 14 nmol/L (Table 3). At 4 h, tumour concentrations
of 13 were 3.4-fold greater than plasma concentrations, whereas
spleen concentrations were 11-fold higher than plasma. For spleen,
concentrations of 13 were approximately eightfold, sevenfold and
threefold higher than tumour at 30 min, 1 h and 4 h, respectively.
These data are consistent with concentrations of 13 found in the
oral cassette PK study. The modest efficacy shown in this experi-
ment may be a reflection of significant drug efflux as Caco-2 data
was strongly suggestive of this, with permeability A ? B below
Figure 3. Increase in histone H3 acetylation observed in tumour tissue measured
by Western blot using an anti-acetylated H3 antibody (Upstate). Intensity of bands
was measured by densitometry and normalised to GAPDH. Change in acetylation
was calculated as % DMSO controls.
400
300
200
100
0
3
2
1
0
n=2
n=2
n=3
the limit of detection, whereas B ? A was 9.4 ꢀ 10ꢁ6 cm sꢁ1
.
Pharmacodynamic changes consistent with HDAC inhibition
were observed in tumour tissue. A twofold increase in histone
acetylation compared to DMSO controls was observed 1 and 4 h
post final dose (Fig. 3) and paralleled tumour concentrations of
13. Tumour concentrations of 13 were above cellular GI50 for 4 h.
This is consistent with the sustained twofold histone hyperacetyla-
tion observed in tumour tissue at this time point (Fig. 4).
In summary we have identified a novel series of potent HDAC
inhibitors based on a pyrimidine hydroxamic acid ZBG. One of
the compounds identified, 13, demonstrated activity in an
HCT116 human colon carcinoma model when dosed orally once
daily. Details of the further optimisation of pyrimidine hydroxamic
acid HDAC inhibitors will be reported in due course.
n=2
n=2
n=1
30 mins
1h
4h
30min
1h
4h
Tumour concentration
Histone H3 acetylation
Figure 4. Quantitation of tumour concentrations (columns 1–3) and effects on
histone H3 acetylation (columns 4–6) of compound 13. Error bars represent value
for n = 1, mean and range for n = 2, mean and SEM for n = 3. Each value is one
animal. Changes in H3 acetylation are expressed as fold difference compared to
DMSO controls.
1300
Control
1100
Acknowledgment
Compound 13
900
The work of the Cancer Research UK Centre for Cancer Thera-
peutics is funded primarily by Cancer Research UK programme
Grants C307/A2187.
700
500
300
100
Supplementary data
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.bmcl.2010.09.016.
References and notes
0
3
5
8
10
Day
12
15
16
1. Marks, P. A.; Rifkind, R. A.; Richon, V. M.; Breslow, R. Clin. Cancer Res. 2001, 7,
759.
2. Grant, S.; Easley, C.; Kirkpatrick, P. Nat. Rev. Drug Discov. 2007, 6, 21.
3. Paris, M.; Porcelloni, M.; Binaschi, M.; Fattori, D. J. Med. Chem. 2008, 51, 1505.
4. Carew, J.; Giles, F.; Nawrocki, S. Cancer Lett. 2008, 269, 7.
5. Minucci, S.; Pellici, P. Nat. Rev. Cancer 2006, 6, 38.
Figure 2. Relative tumour volumes in HCT116 tumour xenograft study for
compound 13 (50 mg/kg, po, qd).
6. Vequero, A. Int. J. Dev. Biol. 2009, 53, 303.
7. Su, H.; Altucci, L.; You, Q. Mol. Cancer Ther. 2008, 7, 1007.
8. Bieliauskas, A.; Pflum, M. Chem. Soc. Rev. 2008, 37, 1402.
9. Flipo, M.; Charton, J.; Hocine, A.; Dassonneville, S.; Deprez, B.; Deprez-Poulain,
R. J. Med. Chem. 2009, 52, 6790.
10. Van Emelen, K.; Verdonck, M.; Van Brandt, S.; Angibaud, P.; Meerpoel, L.;
Dyatkin, A., International Patent Application WO2003/075929, 2003.
11. Arts, J.; Angibaud, P.; Marien, A.; Floren, W.; Janssens, B.; King, P.; van Dun, J.;
Jannsen, L.; Geerts, T.; Tuman, R. W.; Johnson, D. L.; Andries, L.; Jung, M.;
Janicot, M.; van Emelen, K. Br. J. Cancer 2007, 97, 1344.
12. Details of modelling studies are provided in Supplementary data.
13. Itoh, H.; Suzuki, T.; Miyata, N. Curr. Pharm. Des. 2008, 14, 529.
14. Forbes, I.; Johnson, C.; Thompson, M. Synth. Commun. 1993, 23, 715.
15. Moffat, D.; Patel, S.; Mazzei, F.; Belfield, A.; Van Meurs, S.; International Patent
Application WO2006/123121, 2006.
Table 3
Concentrations of 13 in in vivo study
Timea (h)
[Plasma]b
[Tumour]c
[Spleen]c
0.5
1
4
<LODd
14
25
80
215
154
620
1260
500
a
1 h timepoint values are the means of three samples, 0.5 and 4 h timepoints are
means of two samples.
b
Plasma concentration quoted in nM.
Tumour and spleen concentrations quoted in nmol/kg.
Below limit of detection.
c
d

6660
A. D. G. Donald et al. / Bioorg. Med. Chem. Lett. 20 (2010) 6657–6660
16. Representative Procedure—the preparation of Compound
4
is detailed in
steps were performed in a fashion analogous to that described for compound 4
(see Refs. 15,16).
19. Nutley, B.; Raynaud, F.; Wilson, S.; Fisher, P.; McClue, S.; Goddard, P.; Jarman,
M.; Lane, D.; Workman, P. Mol. Cancer Ther. 2005, 4, 125.
20. Moreno-Farre, J.; Workman, P.; Raynaud, F. Recent Adv. Res. Updates 2006, 7,
207.
Supplementary data (additional characterisation data for compounds 5–14 is
available in Ref. 15).
17. HDAC activity was determined using HeLa cell nuclear extract (Cat. No. CC-01-
20-50) obtained from Cilbiotech s.a., Mons, Belgium and the Fluor De Lys™
System (Biomol International, Exeter, UK). The fluorimetric assay was carried
out as described in the Biomol technical manual using 40
l
g HeLa extract
21. Workman, P.; Aboagye, E. O.; Balkwill, F.; Balmain, A.; Bruder, G.; Chaplin, D. J.;
Double, J. A.; Everitt, J.; Farningham, D. A. H.; Glennie, M. J.; Kelland, L. R.;
Robinson, V.; Stratford, I. J.; Tozer, G. M.; Watson, S.; Wedge, S. R.; Eccles, S. A.
Br. J. Cancer 2010, 102, 1555.
protein for 15 min at room temperature. For IC50 determinations, each
compound was assayed at eight concentrations in duplicate and the data
analysed in GraphPad Prism.
18. Compound 14 was prepared from the product obtained from reaction of ethyl
6-chloronicotinate and amine 1 (72 h, 85 °C, 1,4-dioxane, 76%). Subsequent