5-Fluorocytosine derivatives as inhibitors of deoxycytidine kinase

Article abstract of DOI:10.1016/j.bmcl.2009.09.082

A series of potent piperidine-linked cytosine derivatives were prepared as inhibitors of deoxycytidine kinase (dCK). Compound 9h was discovered to be a potent inhibitor of dCK and shows a good combination of cellular potency and pharmacokinetic parameters

Full text of DOI:10.1016/j.bmcl.2009.09.082

Bioorganic & Medicinal Chemistry Letters 19 (2009) 6780–6783
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
5-Fluorocytosine derivatives as inhibitors of deoxycytidine kinase
James E. Tarver a, Theodore C. Jessop a, Marianne Carlsen a, David J. Augeri a, Qinghong Fu a, Jason P. Healy a,
Alexander Heim-Riether a, Amy Xu a, Jerry A. Taylor a, Min Shen a, Philip E. Keyes a, S. David Kimball a,
Xuan-Chuan Yu b, Maricar Miranda b, Qingyun Liu b, Jonathan C. Swaffield b, Amr Nouraldeen b,
Alan G. E. Wilson b, Rick Finch b, Kanchan Jhaver b, Ann Marie DiGeorge Foushee b, Steve Anderson b,
Tamas Oravecz b, Kenneth G. Carson a,
*
a Lexicon Pharmaceuticals, 350 Carter Road, Princeton, NJ 08540, United States
b Lexicon Pharmaceuticals, 8800 Technology Forest Place, The Woodlands, TX 77381, United States
a r t i c l e i n f o
a b s t r a c t
Article history:
A series of potent piperidine-linked cytosine derivatives were prepared as inhibitors of deoxycytidine
kinase (dCK). Compound 9h was discovered to be a potent inhibitor of dCK and shows a good combina-
tion of cellular potency and pharmacokinetic parameters. Compound 9h blocks the incorporation of radi-
olabeled cytosine into mouse T-cells in vitro, as well as in vivo in mice following a T-cell challenge.
Ó 2009 Published by Elsevier Ltd.
Received 20 August 2009
Revised 21 September 2009
Accepted 22 September 2009
Available online 25 September 2009
Keywords:
Immunology
Virology
Cancer
Deoxycytidine kinase
Pharmacokinetics
Deoxycytidine kinase (dCK) is an enzyme which is responsible
for 50-phosphorylation of deoxynucleosides deoxycytidine (dC),
deoxyguanosine (dG) and deoxyadenosine (dA).1 This phosphory-
lation is an important part of the nucleoside salvage pathway
and is the first and rate-limiting step in the bio-activation of dA
and dC. Numerous publications have reviewed the role of the sal-
vage pathways in cancer and inflammation.2,3 In addition to its
natural function, dCK is also responsible for in vivo phosphoryla-
tion of several known nucleoside drugs (such as ddC, 3TC, AraC,
Cladribine, and Gemcitabine).4 Knockout mice deficient in dCK
show reduced levels of circulating B- and T-lymphocytes.5 This
suggested that dCK would be a good target for immunomodulation.
Other connections have been made between the nucleoside sal-
vage pathway and modulation of immune function, such as the ge-
netic deficiency in adenosine deaminase, which leads to severe
combined immunodeficiency.6
Neither compound showed significant potency in the cellular
assay (IC50 >4 M). Compound 2 as a dC analog attracted our atten-
l
tion since other dC analogs have been reported to inhibit dCK with
IC50 values in the micromolar or millimolar potency range.8 In gen-
eral, nucleoside derivatives such as 2 demonstrated kinetic behav-
ior consistent with the compounds acting as substrates as well as
inhibitors. It also should be noted that dCK displays feedback inhi-
bition through deoxynucleotide triphosphates.9 An X-ray co-crys-
tal structure of dCK with bound dC has been published.10
Several attempts were made to merge substructures from 1 and
2 to enhance potency. We sought to attach the hydrophobic right-
2
We conducted a high-throughput screen of our library to find
inhibitors of dCK7 and identified hit structures 1 and 2 (Fig. 1). Both
compounds showed submicromolar enzyme potency against dCK
with IC50s of 510 nM for 1 and 120 nM for 2, respectively.
* Corresponding author. Tel.: +1 609 466 6070; fax: +1 609 466 6079.
E-mail address: kcarson@lexpharma.com (K.G. Carson).
Figure 1. Initial hits from HTS screening against dCK.
0960-894X/$ - see front matter Ó 2009 Published by Elsevier Ltd.
doi:10.1016/j.bmcl.2009.09.082

J. E. Tarver et al. / Bioorg. Med. Chem. Lett. 19 (2009) 6780–6783
6781
Table 1
hand side piece of compound 1 to the fluorocytosine core of 2 using
a suitable linker.
In vitro and cell potency of 7a–d
Our synthetic approach is shown in Scheme 1. Treatment of
BOC-protected amino alcohols such as 3 with 5-fluorocytosine un-
der Mitsunobu conditions11 afforded iminophosphorane-protected
cytosine derivatives such as 4, which were isolable via silica gel
chromatography. Subsequent bis-deprotection of 4 under acidic
conditions gave access to coupled fluorocytosine salts such as 5
with >10:1 ratios of O- to N-linkage. In most cases, trituration of
the solid hydrochloride salt afforded O-linked derivative 5 in high
purity and good yield. Substitution regiochemistry at cytosine
could be distinguished by the 13C chemical shifts of the corre-
sponding N- or O-coupled carbon atom of the piperidine.12 In gen-
eral, the major O-linked derivatives were >5X more potent than
their N-linked isomers (see Table 3). Direct-linked scaffold leading
to compound 7a was prepared by direct displacement of 4,6-
dichloropyrimidine with 5-fluorocytosine.
Coupling of linked amines such as 5 to substituted di-halopyr-
imidines was generally fast and highly selective for mono-addition.
Mono-halide intermediates (e.g., 6) could be purified on silica gel,
but were often sufficiently pure after aqueous workup to telescope
the final step. Standard Suzuki coupling14 under microwave condi-
tions afforded biaryl derivatives such as 7a–d, while a second
nucleophilic aromatic substitution with substituted phenols at ele-
vated temperature led to products 8a–p and 9a–j.
2
Example
Linker
Bond
IC50
(
lM)
EC50
13
(lM)
7a
0.72
7b
7c
7d
0.038
0.034
0.003
0.7
N
0.79
N
0.16
N
shown in Table 2, phenoxy-linked piperidine derivatives 8a–8p
showed very promising increases in potency. Unsubstituted phen-
oxy compound 8a compared reasonably well with benzothiophene
7b. Substitution of the phenyl ring allowed significant increases in
potency. As can be observed in 8b–8d and 8h–8j, substitution was
tolerated at all positions, but the strongest potency improvements
(both primary and cellular) were realized by substitution of the 2-
position as in 8d, 8f, 8g, 8j, and 8l. Even large substitutions at the
2-position such as biphenyl derivative 8g maintained good primary
and cellular potency.
In vitro potency (IC50) was obtained using a filter-binding assay
to measure inhibition of human dCK.7 Cell potency (EC50) was
determined by measuring the inhibition of Ara-C phosphorylation
in CCRF-CEM cells by an Ara-C rescue assay.7 IC50 and EC50 values
shown are averages of at least 2 determinations.
As shown in Table 1, piperidine and spaced piperidine scaffolds
served as good linkers between the cytosine mimic and the hydro-
phobic region. Piperidine 7b was chosen for further SAR explora-
tion because it had slightly better solubility than compounds 7a,
7c, or 7d. We sought to further improve both potency and physical
properties by incorporating more hetero atoms and flexibility. As
The steric affects of phenyl substitution seemed to have a larger
impact on activity compared to the electronic factors; for instance,
SAR trends for methoxy-substituted compounds 8b–8d were gen-
3
2
Table 2
In vitro and cell potency of 8a–p
2
Example
R
IC50
(l
M)
EC50 (lM)
8a
8b
8c
8d
8e
8f
8g
8h
8i
8j
8k
8l
8m
8n
8o
8p
H
0.095
0.033
0.023
0.005
0.021
0.009
0.003
0.019
0.022
0.003
0.010
0.003
0.013
0.006
0.009
0.003
0.34
0.89
4-OCH3
3-OCH3
2-OCH3
4-CH3
2-CH3
2- Ph
4-Cl
3-Cl
2-Cl
2-CH2CH3
2-CN
0.036
0.017
0.060
0.019
0.026
0.057
0.035
0.013
0.061
0.014
0.061
0.015
0.027
0.012
Scheme 1. Reagents and conditions: (a) diethylazodicarboxylate, triphenylphos-
phine, THF, rt, overnight; (b) HCl/Dioxane rt, overnight 50% (two steps); (c) 4,6-
dichloropyrimidine, K2CO3, acetonitrile, 4?25 °C, 4–12 h, 90%; (d) PdCl2DPPF,
Na2CO3, boronic acid, acetonitrile, 140 °C, 10 min, 50–90%; (e) ArOH, K2CO3, DMF,
140 °C, 10 min, 60–90%; All compounds were characterized by 1H NMR, LC/MS, and
HPLC.13
3-CN
2-F
2-CH3, 4-Cl
2-CN, 4-OCH3

6782
J. E. Tarver et al. / Bioorg. Med. Chem. Lett. 19 (2009) 6780–6783
Table 3
Table 4
In vitro and cell potency of 9a–j, 10
Mouse pharmacokinetic parameters of 9f, 9h, 9i
Example
1 mg/kg iv
CL (mL/min/kg)
10 mg/kg po
Mol * h)
2
Vdss (L/kg)
AUC (
l
%F
9f
9h
9i
5.7 0.6
5.0 0.3
3.3 0.5
1.6 0.1
0.9 0.1
1.1 0.1
108 10
101 19
236 13
ꢀ100
ꢀ100
ꢀ100
Activity of the compounds was further established in vitro by
testing their ability to inhibit the uptake of 3H-dC by stimulated
primary murine splenic T cells, see Figure 2. Compounds 9f and
9h dramatically inhibited uptake of 3H-dC by splenocytes stimu-
lated with anti-CD3+anti-CD28 antibodies in a standard 48 h pro-
liferation assay.15,16 The same effect was also observed in human
peripheral blood T cells (not shown). Under these conditions,
incorporation of 3H-dT is minimally affected (not shown).
Compound 9h was tested for pharmacologic activity by deter-
mining its ability to inhibit 3H-dC uptake in vivo, see Figure 3. Com-
pound was administered to mice by oral gavage beginning at day
ꢁ1 relative to ip injection of anti-CD3 antibodies to activate T cells,
and continuing through day 2 for a total of 3 doses.17 Compound
treatment resulted in a 75% or greater inhibition of 3H-dC uptake
by spleen T cells in vivo at a dose of 100 mpk. Subsequent experi-
ments revealed that a single dose of compound given on the same
day as anti-CD3 (day 0) was sufficient to block 3H-dC uptake.
Similar results were obtained if 3H-dC incorporation into bone mar-
row was measured without exogenous stimulation (data not
shown).
2
Example
R
X
Y
Z
IC50
(
l
M)
EC50 (lM)
8l
2-CN
2-CN
2-CN
2-CN
2-CN
4-OCH3
3-CN
H
H
CH3
H
H
H
H
H
H
H
CH3
H
Br
F
F
F
F
F
F
F
F
F
F
F
F
F
F
F
H
H
0.003
0.021
0.010
0.005
0.001
0.004
0.004
0.005
0.001
0.002
0.0007
0.063
0.014
0.054
0.14
9a
9b
9c
9d
9e
9f
9g
9h
9i
0.080
0.004
0.014
0.013
0.020
0.002
0.005
0.005
0.35
4-CN
2-OCH3, 5-CN
2-CN, 4-OCH3
2-CN
H
H
H
F
F
F
9j
10
2-CN
120
100
erally very similar to those of chloro-analogs 8h–8j, both in primary
and cellular potency. Within the SAR for mono-substitution, appro-
priate disubstitution of the ring was also tolerated, as in 8o–8p. The
overall scope of Table 2 shows that the phenoxy substituent is a
useful moiety to modulate potency as well as physico-chemical
properties required for cell penetration.
Vehicle Control
80
9h
9i
60
Further analogs covering pyrimidine ring substitutions are
shown in Table 3. Compared to analog 8l, methyl substitutions at
the 2- or 5-position of the central pyrimidine (compounds 9a–
9b) led to somewhat reduced potency, both primary and cellular.
Bromo substitution at the pyrimidine 5-position was tolerated
(9c), while 5-fluoro substitution led to increased potency (for in-
stance: 9d–8l). The potency gain with the 5-fluoropyrimidine
seemed to hold with other aryl substitutions. In fact, the potency
gains from pyrimidine 5-fluorination were even more pronounced
with some of the simple mono-substituted aryl analogs. For exam-
ple, comparison of 9e–8b, and 9f–8m, show dramatic potency
gains, especially in the cellular assay.
Substitution of the cytosine 5-position is also shown in Table 3.
Comparison of compound 9j with 8l shows that the cytosine 5-flu-
oro substituent had little effect on potency. N-Linked piperidine
derivative 10 illustrates the common ꢀ10-fold potency difference
seen between N-linked and O-linked piperidine derivatives (com-
pound 9j).
Compounds such as 9h and 9i were the most potent dCK inhib-
itors discovered in this series, showing excellent potency in both
primary and cellular assays. Shown in Table 4, the pharmacokinetic
behavior of selected compounds was examined in vivo by admin-
istering the compound to mice by intravenous injection (iv) or oral
gavage (po) at doses of 1 mpk and 10 mpk, respectively. Several
compounds of the class showed very favorable pharmacokinetics,
characterized by low clearance, moderate volume of distribution,
robust oral exposure, and high oral bioavailability. Compounds of
this series showed low protein binding. For example, compound
9h showed 73% protein binding in mouse and 45% in human.
40
20
0
0.001
0.01
0.1
1.0
Compound µM
Figure 2. In vitro 3H-dC incorporation into mouse splenocytes blocked by 9h, 9i.
2000
87% inhibition
p=5.0E-08
1500
1000
500
0
VC
9h
Figure 3. In vivo 3H-dC incorporation into mouse spleen, compound 9h.

J. E. Tarver et al. / Bioorg. Med. Chem. Lett. 19 (2009) 6780–6783
6783
10. Sabini, E.; Ort, S.; Monnerjahn, C.; Konrad, M.; Lavie, A. Nat. Struct. Biol. 2003,
10, 513.
11. Mitsunobu, O. Synthesis 1981, 1.
12. Keyes, P.; Hernandez, G.; Ciancetta, G.; Robinson, J.; Lefebvre, B. Magn. Reson.
Chem. 2009, 47, 38. Regiochemistry also confirmed with NOESY and HMBC
experiments.
In summary, we have described a novel class of potent dCK
inhibitors.15 Compounds such as 9h inhibit dCK in enzyme and cel-
lular assays, and inhibit the target in isolated mouse cells and inhi-
bit the incorporation of radiolabeled 3H-dC in vivo.
13. Augeri, David J.; Carlsen, Marianne; Carson, Kenneth G.; Fu, Qinghong; Healy,
Jason P.; Heim-Riether, Alexander; Jessop, Theodore C.; Keyes, Philip E.; Shen,
Min; Tarver, James E.; Taylor, Jerry A.; Xu, Xiaolian. U.S. Pat. Appl. Publ., 2008,
U.S. 2008146571 A1.
Acknowledgements
14. Miyaura, N.; Suzuki, A. Chem. Rev. 1995, 95, 2457.
The authors thank Alan Main, Terry Stouch, and Giovanni
Cianchetta for helpful discussion during manuscript preparation.
15. In vitro dC uptake: Single cell suspensions from murine spleens were
stimulated in vitro in 96-well plates with immobilized anti-CD3 antibodies
and soluble antibodies to CD28 for 48 h to induce DNA synthesis. Compound
was diluted in DMSO and added to each culture at the indicated concentration.
References and notes
DMSO was kept constant in all wells at 0.1%. Before (16 h) harvest 0.5
l
Ci 3H-
dCTP was added to each well. Cultures were harvested onto glass fiber filters
and counted by liquid scintillation.
16. Radu, C. G.; Shu, C. J.; Nair-Gill, E.; Shelly, S. M.; Barrio, J. R.; Satyamurthy, N.;
Phelps, M. E.; Witte, O. N. Nat. Med. 2008, 14, 783.
17. In vivo dC uptake: To measure uptake of dC by T cells in vivo, mice were dosed
with compound by oral gavage beginning on day ꢁ1 relative to ip injection of
anti-CD3 antibodies. On day 0, mice received the second dose of compound
followed by injection of anti-CD3, and a third dose of compound on day 1,
followed by ip injection of 3H-dCTP. Spleens were harvested from the mice on
day 2, dissociated into single cell suspensions, and an aliquot of cell suspension
harvested onto glass fiber filters and counted by liquid scintillation.
1. Arner, E. S. J.; Eriksson, S. Pharmacol. Ther. 1995, 67, 155.
2. Carson, D. A.; Kaye, J.; Seegmiller, J. E. Proc. Natl. Acad. Sci. U.S.A. 1977, 74, 5677.
3. Carson, D. A.; Kaye, J.; Wasson, D. B. J. Immunol. 1981, 126, 348.
4. Shewach, D. S.; Liotta, D. C.; Schinazi, R. F. Biochem. Pharmacol. 1993, 45, 1540.
5. Anderson, S., in preparation.
6. Blackburn, M. R.; Kellems, R. E. Adv. Immunol. 2005, 86, 1.
7. Yu, X. C.; Miranda, M.; Liu, Z.; Patel, S.; Nguyen, N.; Carson, K. ; Liu, Q.;
Swaffield, J. C. J. Biomol. Screen. in press.
8. (a) Krawiec, K.; Kierdaszuk, B.; Shugar, D. Nucleosides Nucleotides 1995, 14, 495;
(b) Ward, A. D.; Baker, B. R. J. Med. Chem. 1977, 20, 88.
9. Shewach, D. S.; Reynolds, K. R.; Hertel, L. Mol. Pharmacol. 1992, 42, 518.