Structural-based design and synthesis of novel 9-deazaguanine derivatives having a phosphate mimic as multi-substrate analogue inhibitors for mammalian PNPs

Article abstract of DOI:10.1016/j.bmc.2010.01.062

9-(5′,5′-Difluoro-5′-phosphonopentyl)-9-deazaguanine (DFPP-DG) was designed as a multi-substrate analogue inhibitor against purine nucleoside phosphorylase (PNP) on the basis of X-ray crystallographic data obtained for a binary complex of 9-(5′,5′-difluoro-5′-phosphonopentyl)guanine (DFPP-G) with calf-spleen PNP. DFPP-DG and its analogous compounds were synthesized by the Sonogashira coupling reaction between a 9-deaza-9-iodoguanine derivative and ω-alkynyldifluoromethylene phosphonates as a key reaction. The experimental details focused on the synthetic chemistry along with some insights into the physical and biological properties of newly synthesized DFPP-DG derivatives are disclosed.

Full text of DOI:10.1016/j.bmc.2010.01.062

Bioorganic & Medicinal Chemistry 18 (2010) 2275–2284
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Structural-based design and synthesis of novel 9-deazaguanine derivatives
having a phosphate mimic as multi-substrate analogue inhibitors
for mammalian PNPs
Sadao Hikishima a, Mariko Hashimoto a, Lucyna Magnowska b, Agnieszka Bzowska b, Tsutomu Yokomatsu a,
*
a School of Pharmacy, Tokyo University of Pharmacy and Life Sciences, 1432-1 Horinouchi, Hachioji, Tokyo 192-0392, Japan
b Department of Biophysics, Institute of Experimental Physics, Warsaw University, Zwirki i Wigury 93, 02 089 Warsaw, Poland
a r t i c l e i n f o
a b s t r a c t
9-(50,50-Difluoro-50-phosphonopentyl)-9-deazaguanine (DFPP-DG) was designed as a multi-substrate
analogue inhibitor against purine nucleoside phosphorylase (PNP) on the basis of X-ray crystallographic
data obtained for a binary complex of 9-(50,50-difluoro-50-phosphonopentyl)guanine (DFPP-G) with calf-
spleen PNP. DFPP-DG and its analogous compounds were synthesized by the Sonogashira coupling reac-
Article history:
Received 10 December 2009
Revised 27 January 2010
Accepted 28 January 2010
Available online 4 February 2010
tion between a 9-deaza-9-iodoguanine derivative and
x-alkynyldifluoromethylene phosphonates as a
key reaction. The experimental details focused on the synthetic chemistry along with some insights into
the physical and biological properties of newly synthesized DFPP-DG derivatives are disclosed.
Ó 2010 Elsevier Ltd. All rights reserved.
Keywords:
Purine nucleoside phosphorylase
9-Deazaguanines
Phosphate mimic
Multi-substrate analogues inhibitors
1. Introduction
structure are expected to act as a ‘multi-substrate analogue’ inhib-
itor of PNP. Therefore, a number of metabolically stable acyclic
Phosphorylation and dephosphorylation of biological molecules
including proteins, sugars, and nucleosides is one of the most funda-
mental processes in the mammalian organism. Molecular design to
regulate the activity of enzymes which catalyze the phosphoryla-
tion/dephosphorylationprocesseshas a brightfutureinthedevelop-
ment of functional molecules as therapeutic reagents and biological
tools.1 Purine nucleoside phosphorylase (PNP, E.C. 2.4.2.1), a nucle-
oside processing enzyme, is ubiquitous and essential in providing
precursors for RNA and DNA synthesis and energy metabolism.2
PNP catalyzes the reversible phosphorolytic cleavage of the glyco-
sidicbondofribo-anddeoxyribonucleosides, inthepresenceofinor-
ganic orthophosphate (Pi) as a second substrate, to generate purine
bases and ribose(deoxyribose)-1-phosphate (Scheme 1).
In mammals, PNP is crucial for the proper functioning of cellular
immune systems, since PNP deficiency in humans leads to the
impairment of T-cell function, usually with no apparent effects
on B-cell function.2 Consequently, PNP inhibition has become a tar-
get for drug design on selective immunosuppressive agents.
PNP accomplishes the reversible phosphorylation of purine
nucleosides via a ternary complex of enzyme, nucleoside, and
orthophosphate. Compounds that contain covalently linked ele-
ments of both substrates (nucleoside and orthophosphate) in their
nucleotides containing a purine and phosphate-like moiety con-
nected by a linker have been synthesized.3 Of the PNP inhibitors re-
ported, 9-(50,50-difluoro-50-phosphonopentyl)guanine (DFPP-G)
developed by Halazy et al., is one of the most potent and structur-
ally simple multi-substrate analogue inhibitors of PNP.4
During our previous studies directed toward the design and
synthesis of a multi-substrate analogue inhibitor of trimeric PNP
based on the use of difluoromethylenephosphonic acid as a phos-
phate mimic,5 we have recently succeeded in crystallizing a binary
complex with DFPP-G and calf-spleen PNP.6 High-resolution X-ray
differentiation data confirmed for the first time that DFPP-G acts as
a multi-substrate analogue inhibitor as it binds to both nucleoside-
and phosphate-binding sites of PNP. In addition, the putative
hydrogen bonds identified in the base-binding site indicate that
the contact of guanine O6 with Asn243 Od1 is not a direct contact
but is mediated by a water molecule (Fig. 1 and 2). The bridging
water molecule is entropically disfavored, and exclusion of the
water molecule from the complex may induce a tighter bounding
to PNP. On the basis of these findings and hypothesis, we designed
a new candidate, 9-(50,50-difluoro-50-phosphonopentyl)-9-deaza-
β-purine nucleosides
Pi
+
purine base
α-D-pentose-1-phosphate
+
* Corresponding author. Tel./fax: +81 42 676 3239.
E-mail address: yokomatu@toyaku.ac.jp (T. Yokomatsu).
Scheme 1. PNP-catalyzed reversible phosphorylation of purine nucleosides.
0968-0896/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2010.01.062

2276
S. Hikishima et al. / Bioorg. Med. Chem. 18 (2010) 2275–2284
N
2. Results and discussion
2.1. Synthesis
O
N
H2O3PF2C
NH
N
NH2
2.1.1. Synthetic strategies
DFPP-G
To synthesize DFPP-DG derivatives, a strategy which has a high
diversity for adjusting the length of the linker is desirable. In this
context, we consider Pd-catalyzed cross-coupling reactions be-
tween base unit 1 and CF2-phosphonate units 2–4 under Heck, Still
and Sonogashira conditions. This strategy would lead to a variety
of analogues of DFPP-DG upon adjusting the number of carbon
atoms (n) in compounds 2–4 (Scheme 2).
Preliminary studies of the coupling reaction of 1 with simple
linker units such as 3-buten-1-ol, tri-n-butylvinylstanane, and 3-
butyn-1-ol under Heck, Still, and Sonogashira reaction condi-
tions,10 respectively, revealed that only the Sonogashira reaction
with 3-butyn-1-ol gave the corresponding coupling-product in
good yield, while the Still and Heck reactions gave no desired
products. Therefore we decided to examine the cross-coupling
reaction of 1 with
x-alkynyldifluoromethylenephosphonate units
4 under Sonogashira conditions for the synthesis of DFPP-DG
derivatives.
Figure 1. Structure of DFPP-G and its interactions with calf-spleen PNP in the
binary complex (figure from Ref. 6).
2.1.2. Synthesis of base unit 1
A literature review revealed that suitably protected 9-deaza-
guanine derivative 6 has been previously prepared from readily
accessible pyrimidinone 5,11 and compound 6 was transformed
to 9-iodo-9-deazaguanine derivative 7 by NIS-mediated selective
iodination at the 9-position.12 Then, according to the method in lit-
erature, we were able to obtain multi-grams of 7 without any dif-
ficulty. To protect the acidic proton at the 7-position, 7 was
nosylated by a conventional method to give 1 in a quantitative
yield (Scheme 3).
Asn 243
Asn 243
H
H
O
O
N
H
H
O
N
H
N
O
N
H
O
H
N
NH
NH
N
NH2
NH2
N
2.1.3. Preparation of CF2-phosphonate units 4
Treatment of TMS-protected bromopropyne 8 with BrZnCF2
PO3Et2 in the presence of CuBr in DMF according to our previously
reported protocols13 gave 9a in a 70% yield. Tetrabutylammonium
floride (TBAF)-induced desilylation of 9a in THF in the presence of
CF2P(O)(OH)2
DFPP-G
CF2P(O)(OH)2
DFPP-DG
AcOH gave
x-alkynyldifluoromethylene phosphonate unit 4a (C3-
phosphonate unit) in a 97% yield. To synthesize C4–C7-phospho-
nate units 4b–e, alkylation of LiCF2PO3Et2 with readily available
triflates 10–13 was applied according to the method described
by Berkowitz.14 While the reaction with 11–13 under the condi-
tions proceeded smoothly to give 9c–e in high yields, the reaction
with 10 was problematic due to the competitive formation of large
amounts of enyne products by b-elimination of HF under the con-
ditions. However, 9b was isolated in a very low yield (8%). Com-
pounds 9b–e were treated with TBAF in THF to give the requisite
C4–C7-phosphonate units 4b–e in modest and good yields
(Scheme 4).
Figure 2. Binding patterns of DFPP-G and expected binding patterns of DFPP-DG
with the base-binding site of PNP.
guanine (DFPP-DG) as an inhibitor against PNP, in which the guan-
ine base of DFPP-G is replaced by 9-deazaguanine (Fig 2).
Newly designed DFPP-DG is expected to bind to the nucleoside-
binding site through direct hydrogen-bonds to Asn243 as shown in
Figure 2. Such hydrogen-bond patterns in which N(7) of the base is
protonated and donates a hydrogen to Asn243Od1, while the
Asn243Nd1 donates a hydrogen to the exocyclic O6 of the base,
are generally observed with binary complexes with PNP and
ground-state PNP inhibitors such as 9-arylmethy-9-deazaguanine
derivatives.2,7 In addition, similar hydrogen-bond patterns were
detected in a ternary complex with the enzyme, phosphate, and
immucillin G, a transition-state analogue inhibitor of PNP.8 Syn-
thetic and biological studies of DFPP-DG and the related analogues
have been reported in a preliminary communication.9 In this paper,
we now disclose the experimental details focused on the synthetic
chemistry along with additional insights into the physical and bio-
logical properties of newly synthesized DFPP-DG derivatives as an
extension of our previous works. We also disclose an improved
CF2PO3Et2
CF2PO3Et2
n
2
O
N
Ns
Bu3Sn
POM
N
n
n
N
3
N
CHNMe2
I
1
CF2PO3Et2
base unit
4
CF2-phosphonate units
Scheme 2. Cross-coupling strategy for synthesis of DFPP-DG analogues.
synthesis of
a x-alkynyldifluoromethylenephosphonate unit,
which are previously synthesized in a very low yield.

S. Hikishima et al. / Bioorg. Med. Chem. 18 (2010) 2275–2284
2277
O
2.1.5. Sonogashira reaction with phosphonate units 4b-e
O
H
N
O2N
CH3
The Sonogashira coupling reaction between base unit 1 and C5-
phosphonate unit 4c was first examined to verify proper reaction
conditions, since 4c was readily available in a large scale. These re-
sults are summarized in Table 1.
POM
NIS/DMF
ref.12
4-steps
ref. 11
NH
N
NH2
N
N=CHNMe 2
N
6
5
When the Sonogashira reaction of 1 was carried out with a
slight excess of 4c in Et3N in the presence of PdCl2(PPh3)2
(2 mol %) and CuI (1 mol %) at 60 °C for 15 h according to the pro-
cedure reported by Larock,17 desired coupling product 18c was ob-
tained in a 58% yield along with a large amount of recovered 1
(36%) (entry 1). The yield of 18c was not increased upon using
the Pd(0) catalysis under the same conditions (entry 2). In an effort
to increase the yield, modified conditions using representative sol-
vents composed by 2 equiv of Et3N were examined (entries 4–9).
This survey identified MeCN was a good solvent to induce a good
yield (69%) of 18c and minimize the recovered yield of 1 (entry 6).
The findings were applicable to the Sonogashira reactions with
phosphonate units 4b, 4d, and 4e. The results are also included in
Table 1. It should be noted that the prolonged reaction time (15 h)
of the reactions with phosphonate units 4d and 4e resulted in low
yields and modest yields were obtained upon diminishing the reac-
tion time to 3–4 h (entries 8 and 9).
O
Ns
O
H
POM
N
POM
NsCl/NaH
CH2Cl2
N=CHNMe2
N
N
N
N=CHNMe2
N
N
I
I
1
7
POM:pivaloyloxymethyl
Scheme 3. Preparation of 9-iodo-9-deazaguanine derivatives.
Br
(a)
TMS
8
CF2PO3Et2
n
R
(b)
or
9a-c
9d,e
4a-e
R=TMS
R=TIPS
R=H
OTf
(c)
n
(d)
2.1.6. The Sonogashira reaction of 1 with phosphonate unit 4a
The Sonogashira reaction of 1 with phoshonate unit 4a was
examined according to the optimized conditions described in the
previous section (Table 1, entry 6). However, this reaction yielded
the desired coupling product 18a in a low yield along with the de-
nosyl derivative 19a and de-nosyl enyne product 20 in a compara-
tive yield (Scheme 5 and Table 2, entry 1).
To gain an insight into a mechanism of the formation of 19a and
20, isolated 18a was treated with Et3N in MeCN at 60 °C for 22 h.
This reaction gave 19a and 20 in 8% and 38% yields, respectively.
However, phosphonate unit 4a was inert to the same conditions.
On the basis of these findings, the mechanism shown in Figure 3
is proposed for the formation of the by-products. In this
mechanism, the initial-formed cross-coupling product 18a is
transformed to enyne derivative 21 through Et3N-induced b-elim-
ination of HF, since the acidity of propalgyl protons increase upon
incorporation of the heteroaromatic. Enyne 21 is rapidly de-nosy-
lated to give 20 by action of the floride ion derived from triethyl-
ammonium floride (Et3N+HFꢀ),18 that is produced in the first-step
reaction. Alternatively, the route where compound 19a is first
de-nosylated, followed by b-elimination of HF to give 20, is also
predictable.
R
n=1 R=TMS
n=2 R=TMS
n=3 R=TIPS
n=4 R=TIPS
10
11
12
13
a: n=0; b: n=1; c: n=2; d: n=3; e: n=4
Scheme 4. Synthesis of
x-alkynyldifluoromethylenephosphonate units. Reagents
and conditions: (a) BrZnCF2PO3Et2, CuBr, DMF, rt; (b) TBAF, THF, AcOH; (c)
LiCF2PO3Et2, THF, ꢀ78 °C; (d) TBAF, THF.
2.1.4. Improved synthesis of CF2-phosphonate 4b
As described in the previous section, we encountered a signifi-
cant difficulty in preparation of CF2-phosphonate unit 4b via a
conventional alkylation reaction of LiCF2PO3Et2 with the
corresponding triflate 10. The difficulty attributes to extremely
unstable of triflate 10 and the alkylation product 9b under the ba-
sic conditions. Therefore, to obtain 4b in a reasonable yield and
quantities, a new reaction, which proceeds under a neutral condi-
tion, should be developed as a key reaction. Fuchs has reported
that organic iodides undergo chemospecific alkynylation with
acetylenic triflones through photochemical irradiation.15,16 The
Fuch method possess an attractive feature and would be applicable
to the synthesis of 4b, since the reaction medium is almost neutral.
Then, we examined alkynylation reaction of 3-iodo-1,1-dif-
luorophosphonate 15, readily prepared from the corresponding
alcohol 14, with triisopropylsilylacetylenic triflone 16 (Scheme 5).16
According to the reported method, a 0.15 M solution of 15
(1.0 equiv) and 16 (1.2 equiv) in benzene was irradiated by
500 W Hg-lamp (Pyrex filter) in the presence of a catalytic amount
of hexabutyldistannane for 16 h at ambient temperature. As ex-
pected, this reaction provided the desired adduct 17 in 42% yield.
Compound 17 was desilylated with TBAF in THF to give 4b in
67% yield. Using this sequence, required 4b was obtained in mod-
est overall yield (26 %).
Keeping the above mechanistic prediction in mind, the cross-
coupling reaction between 1 and 4a was carried out in MeCN in
the presence of a slight excess of Et3N for diminishing reaction
time (1 h) to suppress the formation of the by-products (Table 2,
entry 2). This reaction gave the desired coupling product 18a in a
59% yield, along with recovering 1 in a 41% yield. As expected,
the formation of the by-products was significantly suppressed un-
der the conditions.
2.1.7. Preparation of DFPP-DG and its derivatives
Compounds 18a–e, thus obtained, were readily transformed to
DFPP-DG and its analogues (Scheme 6). Thiophenol-mediated re-
moval of the nosyl protecting group for 18b, followed by hydroge-
nation, gave 22b in a 90% yield. Compound 22b was treated with
NaOMe in MeOH to give 23b in an 82% yield. Removal of the ethyl
protecting group and the dimethylaminomethylene group for 23b
was achieved in concd HCl at reflux (20 h) to give DFPP-DG (25) as
a white solid in a 93% yield. Compounds 18a and 18c–e were,
respectively, transformed to nor-DFPP-DG (24), homo-DFPP-DG
(26), 6C-DFPP-DG (27), and 7C-DFPP-DG (28) by the same proce-
dure as a series of DFPP-DG in good overall yield.
CF2PO3Et2
TIPS
SO2CF3
+
X
14 X=OH
15 X=I
16
(a)
CF2PO3Et2
(b)
R
17 R=TIPS
4b R=H
(c)
Scheme 5. Improved synthesis of phosphonate unit 4b. Reagents and conditions:
(a) I2, PPh3, imidazole, CH2Cl2; (b) (n-Bu3Sn)2, h , benzene; (c) TBAF, THF.
m

2278
S. Hikishima et al. / Bioorg. Med. Chem. 18 (2010) 2275–2284
Table 1
Sonogashira coupling between 1 and 4b–ea
O
POM
N
Ns
O
N
Ns
N
POM
Pd-cat
CuI (1mol%)
N
CHNMe2
N
N
n
Et3N (eq)
CF2PO3Et2
N
N
CHNMe
2
solvent
60 ºC
I
4b-e
1
CF2PO3Et2
18b-e
Solvent
n
Entry
Phosphonate Unit
Pd-cat (mol %)
Et3N (equiv)
Time (h)
Yield (%)
18
1
1
2
3
4
5
6
7
8
9
4c (n = 2)
4c (n = 2)
4c (n = 2)
4c (n = 2)
4c (n = 2)
4c (n = 2)
4b (n = 1)
4d (n = 3)
4e (n = 4)
PdCl2(PPh3)2 (2)
Pd(PPh3)4 (2)
>60
>60
>60
2
2
2
2
2
2
15
15
15
15
15
15
15
3
None
None
None
THF
58
22
40
15
41
64
66
75
68
36
78
49
69
59
6
8
21
19
PdCl2(PPh3)2 (20)
PdCl2(PPh3)2 (2)
PdCl2(PPh3)2 (2)
PdCl2(PPh3)2 (2)
PdCl2(PPh3)2 (2)
PdCl2(PPh3)2 (2)
PdCl2(PPh3)2 (2)
DMF
MeCN
MeCN
MeCN
MeCN
4
a
All reactions were carried out in the presence of 1 mol % of CuI.
O
Ns
N
Table 2
CF2PO3Et2
4a
Sonogashira reaction of 1 with phosphonate unit 4aa
NPOM
O
Ns
N
N
R
CuI, Et3N
I
N
PdCl2(PPh3)2
NPOM
CF2PO3Et2
+
1
MeCN, 60 ºC
4a
N
N
CHNMe2
R = CHNMe2
I
Ns
N
O
H
N
1
O
H
N
R
NPOM
O
O
NPOM
N
N
N
N
R
NPOM
CHNMe2
NPOM
N
R
+
N
N
N
N
CHNMe2
CF2PO3Et2
19a
CF2PO3Et2
18a
CF2PO3Et2
F
20
Et3N
Et3N
Et2O3P
Et3NH F
18a
19a
R=Ns
R=H
Ns
N
O
O
H
N
Entry
Et3N (equiv)
Time (h)
Yield (%)
18a
19a
20
1
NPOM
NPOM
1
2
2
1.2
14
1
15
59
17
11
NDb
41
N
N
NDb
NDb
N
R
N
R
a
All reactions were carried out in the presence of 1 mol % of CuI and 2 mol % of
PdCl2(PPh3)2.
F
F
b
21
20
Not detected.
PO3Et2
PO3Et2
Figure 3. The predictive mechanism of the formation of by-products in Sonogashira
reaction with phosphate unit 4a.
2.2. Biophysical/biochemical results and discussion
2.2.1. Spectral data for the new analogues
not shown), to cation formation, most probably as a result of pro-
tonation at nitrogen N(3) and anion formation resulting in depro-
tonation of N(1)-H, as shown in Scheme 7.
On the basis of the pKa1 and pKa2 values, the active form of
DFPP-DG derivative is most probably the neutral form under the
physiological conditions. Some time ago it was hypothesized, on
the basis of unusual enhancement of the fluorescence in the PNP/
guanine complex that trimeric PNPs bind preferentially to the
Spectrophotometric titrations of the DFPP-DG analogues and
9-deazaguanine were carried out in the pH range 1–14. These titra-
tions indicated that all DFPP-DG analogues show two ionizable
sites at the heterocyclic base: one in the pH range 4–5 and the
second one about pH 11 (Fig. 4). These were attributed, by compar-
ison with the spectra of three model compounds, 9-deazaguanine,
1-methyl-9-deazaguanine and 1,7-dimethyl-9-deazaguanine (data

S. Hikishima et al. / Bioorg. Med. Chem. 18 (2010) 2275–2284
2279
O
O
N
R
0.014
0.012
0.010
0.008
0.006
0.004
0.002
H
N
POM
N
NR
N
(b)
CHNMe2
N
N
n
N=CHNMe2
CF2PO3Et2
22a-e
R=POM
R=H
n
CF2PO3Et2
(c)
23a-e
O
18a-e
19a-e
R=Ns
(a)
H
N
R=H
NH
a: n=0; b: n=1;
c: n=2; d: n=3;
e: n=4
(d)
NH2
N
n
CF2PO3H2
24 (nor-DFPP-DG) n=0
27 (6C-DFPP-DG) n=3
28 (7C-DFPP-DG) n=4
25 (DFPP-DG) n=1
26 (homo-DFPP-DG) n=2
-12 -9 -6 -3
0
3
6
9
12 15 18
I [nM]
Scheme 6. Synthesis of DFPP-DG analogues. Reagents and conditions: (a) thio-
phenol, K2CO3, DMF; (b) H2, Pd–C, MeOH–CHCl3; (c) NaOMe, MeOH; (d) concd HCl.
Figure 5. Inhibition of human erythrocyte PNP by DFPP-DG, (s) 0
lM (d) 4.99 lM,
(h) 10.98 M, (j) 18.71 M, shown in Dixon plot form. Reactions were carried out
l
l
in 50 mM hepes buffer pH 7.0, at 25 °C, with 7-methylguanosine as a variable
substrate, and in the presence of 1 mM phosphate. Kinetic data are presented in the
form of the Dixon plot only to visualize the type of inhibition. Data (vo, co and [I])
0.10
0.08
0.06
0.04
0.02
0.00
were analyzed and kinetic constants were obtained with the use of the weighted
least-squares non-linear regression using the program Leonora.21
2.2.2. Inhibition data
The inhibitory properties of new compounds with calf-spleen
and human erythrocyte PNPs were determined with 7-methylgu-
anosine (m7Guo) as a variable substrate using methods previously
described for other inhibitors of trimeric PNPs.22 The substrate em-
ployed was m7Guo, because in the case of trimeric PNPs, its kinet-
ics, in striking contrast to those observed in the natural substrates
inosine and guanosine, is fairly well described by the Michaelis–
Menten model.22a With fixed concentrations of one substrate—
inorganic phosphate—apparent inhibition constants (Kai pp) were
determined from initial velocity data with variable concentrations
of the inhibitor and the second substrate—m7Guo. Dixon plots dis-
played a competitive mode of inhibition as shown in Figure 5 for
DFPP-DG and human erythrocyte PNP. Kinetic data are presented
in the form of the Dixon plot only to visualize the type of inhibi-
tion. Data sets were analyzed and the apparent inhibition con-
stants were calculated with the use of the weighted least-squares
non-linear regression using the program Leonora23 as described
in the Experimental section. The results obtained for DFPP-DG
and analogues are summarized in Table 3.
Not only the apparent inhibition constants but also IC50s were
calculated form the experimental data (see Experimental section).
This allowed for a more reliable comparison with literature data
available for previously synthesized PNP inhibitors. For a compar-
ison of the inhibitory activity of DFPP-G22a and a transition state-
analogue inhibitor—immucillin H24—are also included in Table 3.
All compounds studied were found to be very potent inhibitors
of 7-methylguanosine (m7Guo) phosphorolysis with apparent inhi-
bition constants, Kai pp, in the nM range. Inhibition is competitive
versus nucleoside (m7Guo), and apparent Ki values decrease with
decreasing phosphate (the fixed substrate) concentration (see
Table 3). This indicates that the inhibitors studied bind to both
nucleoside- and phosphate-binding sites, hence they act as mul-
ti-substrate analogue inhibitors.
1
2
3
4
5
6
7
pH
8
9 10 11 12 13 14
0.35
0.30
0.25
0.20
0.15
0.10
0
1
2
3
4
5
6
7
8
9 10 11 12 13 14
pH
Figure 4. Spectrophotometric titrations of DFPP-DG (upper panel) and 9-deaza-
guanine (bottom panel) showing two ionization sites in the pH range 1–14. The
solid curves are the least-squared fits to the points of titration, conducted
separately in the pH range 1–9 and 7–14 assuming a single proton ionization site
in the particular pH range examined.
O
O
N
O
N
H
N
H
N
H
N
NH
NH
N
p
Ka1=5.1
pKa2=10.6
NH2
NH2
NH2
N
H
CF2P(O)(OH)2
CF2P(O)(OH)2
CF2P(O)(OH)2
As predicted by previous structural studies,6 DFPP-DG allows
for more favorable interactions with the base-binding site of calf
spleen and human erythrocyte PNPs when compared with
DFPP-G, and therefore, yields Kai pp and IC50s lower than those ob-
served for DFPP-G (Table 3). Other derivatives with shorter and
longer linkers exhibited weaker inhibitory effects, except for
Scheme 7. Chemical structures of neutral and ionic forms of DFPP-DG and pKa
assignments that were done by comparison with UV spectra and pKa of model
compounds, 1-methyl-9-deazaguanine and 1,7-dimethyl-9-deazaguanine.
anion of guanine.19 However, subsequent crystallographic20 and
solution studies21 have shown unequivocally that this is not the case.

2280
S. Hikishima et al. / Bioorg. Med. Chem. 18 (2010) 2275–2284
Table 3
Inhibitory properties of DFPP-G, DFPP-DG and their analogues vs calf-spleen and human erythrocyte PNPsa
b
b
Kai pp (nM) for
Kai pp (nM) for
Kei q (nM) for
Compound
Pi concd (mM)
IC50 (nM) for
IC50 (nM) for
calf PNP
human PNP
calf PNP
human PNP
calf PNP
DFPP-G
DFPP-G
DFPP-DG
DFPP-DG
nor-DFPP-DG
homo-DFPP-DG
6C-DFPP-DG
7C-DFPP-DG
Immucillin H
1
18.7
—
10.2
—
20.2
—
20.4
—
170
9.5
22
6.9 0.7c
2.7 0.2
4.4 0.6
1.0 0.2
—
10.8 0.7
—
8.1 0.6
1.0 0.2
—
0.72 0.13
0.025
1
0.025
1
1
1
1
50
0.085 0.013
—
10.2
42
—
5.7 0.6
5.3 0.4
0.36 0.04
0.23 0.03
0.94 0.10
0.023 0.05d
21
2
13
1
—
—
6.1 0.9
5.6 0.8
—
—
41 8d
a
All reactions were carried out in 50 mM hepes buffer pH 7.0, at 25 °C, with 7-methylguanosine as a variable substrate, in the presence of a fixed concentration of
phosphate.
b
With 25 lM of 7-methylguanosine.
Data from Iwanow et al. (Ref. 22b).
Data from Miles et al. (Ref. 24).
c
d
homo-DFPP-DG vs human erythrocyte PNP, which exhibited even
The preparation of these nucleotide analogues was achieved using
the Sonogashira reaction between a 9-iodo-9-deazaguanine deriva-
better binding properties than those observed for DGPP-DG (Kiapp
=
5.3 nM as compared with 8.1 nM, see Table 3).
tive and x-alkynyldifluoromethylenephosphonate units as a key
These inhibition constants, Kai pp, determined by the classical ap-
proach and shown in Table 3 should be treated as apparent values
since the reaction rates observed in the presence of DFPP-DG and
analogues exhibit some inhibition in the initial velocity experi-
ments, with increasing inhibition as a function of time (Fig. 6).
Therefore, the inhibition constant for binding of DFPP-DG and ana-
logues to calf-spleen PNP was also determined at equilibrium, as
described in Experimental section. Approach to equilibrium was
followed by measuring the velocity observed after various times
of incubation (0.5–120 min), and for various inhibitor concentra-
reaction. The apparent inhibitionconstants of DFPP-DG, determined
by the initial velocity experiments, proved to be more potent than
those of DFPP-G and immucillin H. The equilibrium inhibition con-
stant for DFPP-DG is Kei q = 85 13 pM and those of DFPP-DG ana-
logues are also in the pM range. Hence these compounds are
almost as potent inhibitors as the transition state inhibitors,
immucillins.
DFPP-G4 and some of its analogues with cyclic and acyclic linkers
between the purine base and difluoromethylene phosphonic acid
groups,5,22b were synthesized following a suggestion, derived from
tions (in the range 1–50 nM),
were determined as shown in Figure 6, upper panels, by fitting
Eq. 1 to the (t) dependence, separately for each inhibitor concen-
tration [I]. From the set of s obtained for various inhibitor concen-
trations, the inhibition constant at equilibrium, Kieq
was
determined by fitting Eq. 2 to the s[I]/kc dependence, and for
v
(t, [I]). Steady-state velocities, vs
electronic and steric considerations,27 that
a-fluoroand a,a-difluoro
alkanephosphonates should better mimic phosphate esters than the
corresponding phosphonates, hence they are expected to better
interact with the phosphate-binding site of PNPs. As expected, on
the basis of these findings, such derivatives proved to be competitive
inhibitors of trimeric PNPs exhibiting inhibitory activity in pM
concentrations. Some of them also showed interesting biological
properties, for example, they significantly slow down proliferation
of T-lymphocytes isolated from patients with autoimmune thyroid
disease—Hashimoto’s thyroiditis, compared to the inhibitory effects
on the growth of human blood T-lymphocytes isolated from healthy
donors.22b It isthereforeverylikelythatthenewlysynthesizedseries
of DFPP-DG and its analogues are even more promising candidates
as in vivo PNP inhibitors. Detailed studies of their effect on the
proliferation of various cell lines are now in progress.
v
v
,
v
DFPP-DG was found to be Kieq = 85 13 pM (Table 3), hence two
orders of magnitude lower than the apparent inhibition constant
determined in the standard initial velocity experiment,
Kai pp = 4.4 nM (Table 3). For other compounds Kei q was also at least
10-fold lower that the apparent inhibition constant, Kai pp deter-
mined by the standard initial velocity procedure (se Table 3).
The time-dependence of inhibition may be due to low enzyme
and inhibitor concentrations leading to slow-binding inhibition
caused by the fact that equilibrium may not be reached in the time
scale of the initial velocity studies.25 However, it is also possible
that the time-dependence of inhibition of DFPP-DG is due to the
slow-binding of the inhibitor with the enzyme (one step mecha-
nism) or slow-onset (i.e., two step) mechanism. In the two-step
mechanism binding involves the rapid formation of the enzyme/
inhibitor collision complex followed by a slow conformational
change leading to a more tightly bound enzyme/inhibitor complex,
as observed for immucillin H and analogues.24 Studies that aim to
answer the question, which of the two mechanisms is responsible
for the time-dependence of inhibition of DFPP-DG, were under-
taken and the results obtained are more consistent with the two-
step mechanism.26
4. Experimental section
4.1. Chemicals
4.1.1. 2N-(N,N-Dimethylaminomethylidene)-9-iodo-1N-
(pivaloyloxy)methyl-9-deazaguanine (7)
To a stirred solution of 610 (1.88 g, 5.89 mmol) in DMF (59 mL)
was added NIS (1.32 g, 5.89 mmol) at room temperature. After
being stirred for 30 min, the solvent was evaporated in vacuo.
The residue was treated with MeOH and the resulting precipitates
were filtered to give 7 (2.56 g, 98%) as a white solid: mp >228 °C
(colored); 1H NMR (400 MHz, DMSO-d6) d 12.25 (br s, 1H, NH),
8.50 (s, 1H, CHNMe2), 7.45 (d, 1H, 8-H, J = 2.2 Hz), 6.17 (s, 2H,
CH2OC(O)tBu), 3.16, 2.97 (each s, 6H, CHN(CH3)2), 1.08 (s, 9H,
tBu); 13C NMR (100 MHz, DMSO-d6) d 176.64, 156.68, 153.65,
153.44, 144.56, 132.08, 113.73, 65.49, 57.57, 40.56, 38.22, 34.50,
26.69 (3 carbons); ESI-MS (LR) m/z 446 (MH+); ESI-MS (HR) calcd
for C15H21N5O3I (MH+): 446.0689, found: 446.0677.
3. Conclusion
In summary, we have designed and synthesized 9-(50,50-difluoro-
50-phosphonopentyl)-9-deazaguanine (DFPP-DG) and its analogues
asnewmulti-substrate analogueinhibitorsagainstPNPsonthebasis
of the structure for the binary complex of DFPP-G/calf-spleen PNP.

S. Hikishima et al. / Bioorg. Med. Chem. 18 (2010) 2275–2284
2281
homo-DFPP-DG
6C-DFPP-DG
4
3
2
1
0
4
3
2
1
0
0
5
10 15 20 25 30 35 40 45 50
0
5
10 15 20 25 30 35 40 45 50
time of incubation [min]
time of incubation [min]
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0.0
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0.0
0
5
10
15
20
25
30
35
40
0
5
10
15
20
25
30
35
40
homo-DFPP-DG [nM]
6C-DFPP-DG [nM]
Figure 6. Time-dependence of inhibition of (upper panels) and inhibition in equilibrium (lower panels) of calf spleen PNP by homo-DFPP-DG (left panels) and 6C-DFPP-DG
(right panels). Time-dependence of inhibition: 4 73 nM of homo-DFPP-DG (h) and 2.09 nM of 6C-DFPP-DG (j) were incubated with 3.95 nM and 2.45 nM of calf PNP (in
terms of subunits), respectively, and initial velocity was measured as described in Experimental section. Similar experiments were repeated for various inhibitor
concentrations (not shown here) and velocity in equilibrium, vs, was determined by fitting Eq. 1 to the time-dependence observed. The equilibrium velocities, vs were plotted
as a function of the inhibitor concentration (lower panels). The inhibition constants at equilibrium, Kei q, were obtained by fitting Eq. 2 to the dependence of equilibrium
velocities
v
s versus inhibitor concentration, [I]. The Keq value obtained from these data is 0.36 0.04 nM in the case of homo-DFPP-DG, and 0.23 0.03 nM in the case of 6C-
i
DFPP-DG.
4.1.2. 2N-(N,N-Dimethylaminomethylidene)-9-iodo-7N-(2-
nitrobenzene)sulfonyl-1N-(pivaloyloxy)methyl-9-deazaguanine
(1)
4.1.3. Diethyl 1,1-difluoro-5-(triisopropylsilyl)pent-4-
ynylphosphonate (17)
To a solution of 15 (100 mg, 0.29 mmol) and 16 (110 mg,
0.35 mmol) in benzene (1.95 mL) was added dropwise a solution
To a stirred suspension of 7 (2.33 g, 5.23 mmol) in CH2Cl2
(106 mL) was added NaH (50% purity, 276 mg, 5.75 mmol) at room
temperature. After being stirred for 10 min, the mixture was trea-
ted with o-nitrobenzenesulfonyl chloride (1.39 g, 6.28 mmol). The
mixture was stirred for 6 h at room temperature. The reaction
was quenched with MeOH, poured into water and extracted with
CHCl3. The organic extracts were washed in brine, dried over
MgSO4 and evaporated. The resulting residue was chromato-
graphed on silica gel (hexane/CHCl3 = 1:1) to give 1 (3.51 g, ca.
100%) as a pale yellow solid: mp 238–240 °C; 1H NMR (400 MHz,
DMSO-d6) d 8.57 (s, 1H, CHNMe2), 8.30 (dd, 1H, CH of Ns,
J = 1.3 Hz, 7.9 Hz), 8.10 (dd, 1H, CH of Ns, J = 1.2 Hz, 7.8 Hz), 8.00–
7.92 (m, 3H, CH of Ns ꢁ 2, 8-H), 6.01 (s, 2H, CH2OC(O)tBu), 3.20,
2.99 (each s, 6H, CHN(CH3)2), 1.03 (s, 9H, tBu); 13C NMR
(100 MHz, DMSO-d6) d 176.42, 157.70, 156.55, 151.80, 150.37,
147.29, 136.28, 135.08, 132.99, 132.59, 130.08, 125.22, 112.19,
67.34, 65.56, 40.93, 38.18, 34.79, 26.61 (3 carbons); ESI-MS (LR)
m/z 631 (MH+); ESI-MS (HR) calcd for C21H24N6O7SI (MH+):
631.0472, found: 631.0470.
of hexabutyldistannane (74 lL, 0.15 mmol). After the solution
was irradiated by 500 W Hg-lamp (Pyrex filter) for 16 h at the
room temperature, the reaction mixture was poured into the ice-
water and extracted with AcOEt. The extracts were washed with
brine, dried over MgSO4, and evaporated. The resulting residue
was chromatographed on silica gel (hexane/AcOEt = 8:1) to give
17 (48 mg, 42%) as a pale yellow oil: 1H NMR (400 MHz, CDCl3) d
4.28 (dq, 4H, OCH2CH3 ꢁ 2, JH,H = JH,P = 7.2 Hz), 2.55 (m, 2H, 3-
H ꢁ 2), 2.35 (m, 2H, 2-H ꢁ 2), 1.39 (t, 6H, OCH2CH3 ꢁ 2,
J = 7.2 Hz), 1.03 (m, 21H, Si(CH(CH3)2)3); 13C NMR (100 MHz,
CDCl3) d 119.89 (dt, JC,P = 215.4 Hz, JC,F = 260.3 Hz), 106.16, 81.41,
64.73 (d, JC,P = 6.6 Hz,
2 carbons), 33.91 (dt, JC,P = 14.8 Hz,
JC,F = 20.6 Hz), 18.72 (3 carbons), 16.54 (d, JC,P = 5.5 Hz, 2 carbons),
11.89 (m), 11.33 (6 carbons); 19F NMR (282 MHz, CDCl3) d ꢀ50.43
(dt, 2F, JF,H = 18.8 Hz, JF,P = 107.1 Hz); 31P NMR (122 MHz, CDCl3) d
7.29 (t, 1P, JP,F = 107.1 Hz); ESI-MS (LR) m/z 397 (MH+);
ESI-MS (HR) calcd for C18H36F2O3PSi (MH+): 397.2139, found:
313.2114.

2282
S. Hikishima et al. / Bioorg. Med. Chem. 18 (2010) 2275–2284
4.1.4. Diethyl 1,1-difluoropent-4-ynylphosphonate (4b)
To a solution of 17 (40.5 mg, 0.10 mmol) in THF (3.5 mL) was
added dropwise a 1.0 M THF solution of TBAF (204 lL, 0.20 mmol)
73.82, 65.39, 64.35 (d, JC,P = 6.7 Hz, 2 carbons), 40.46, 38.24,
34.45, 33.09 (dt, JC,P = 14.4 Hz, JC,F = 20.4 Hz), 26.70 (3 carbons),
16.23 (d, JC,P = 5.0 Hz, 2 carbons), 11.59 (m); 19F NMR (376 MHz,
DMSO-d6) d –50.82 (dt, 2F, JF,H = 20.0 Hz, JF,P = 105.0 Hz); 31P NMR
(162 MHz, DMSO-d6) d 7.50 (t, 1P, JP,F = 105.0 Hz); ESI-MS (LR)
m/z 558 (MH+); ESI-MS (HR) calcd for C24H34F2N5O6P (MH+):
558.2292, found: 558.2271.
at 0 °C. After the stirring was continued for 5 min at the same tem-
perature, the reaction mixture was poured into water and ex-
tracted with AcOEt. The organic extracts were washed with
brine, dried over MgSO4, and evaporated. The resulting residue
was chromatographed on silica gel (hexane/AcOEt = 6:1) to give
4b (16.4 mg, 67%) as a pale yellow oil: 1H NMR (400 MHz, CDCl3)
d 4.28 (dq, 4H, OCH2CH3 ꢁ 2, JH,H = JH,P = 7.3 Hz), 2.50 (dt, 2H, 3-
H ꢁ 2, J3,5 = 2.6 Hz, J3,2 = 7.5 Hz), 2.35 (m, 2H, 2-H ꢁ 2), 1.98 (t,
1H, 5-H, J5,3 = 2.6 Hz), 1.39 (t, 6H, OCH2CH3 ꢁ 2, J = 7.3 Hz); 13C
NMR (100 MHz, CDCl3) d 119.67 (dt, JC,P = 215.7 Hz, JC,F = 260.5 Hz),
4.1.7. 9-[50,50-Difluoro-50-(diethylphosphono)pentyl]-2N-(N,N-
dimethylaminomethylidene)-1N-(pivaloyloxy)methyl-9-
deazaguanine (22b)
A solution of 19b (1.00 g, 1.79 mmol) in MeOH/CHCl3 (1:1)
(180 mL) was hydrogenated over 10% Pd–C (1.0 g) for 20 min at
room temperature under an atmospheric pressure. The catalyst
was removed through Celite, and the filtrate was concentrated in
vacuo. The resulting residue was chromatographed on silica gel
(3% MeOH in CHCl3) to give 22b (1.01 g, ca. 100%) as a white solid:
UV kmax (MeOH) 303 nm, 265 nm; mp 111–118 °C; 1H NMR
(400 MHz, DMSO-d6) d 12.24 (s, 1H, NH), 8.56 (s, 1H, CHNMe2),
7.29 (s, 1H, 8-H), 6.11 (s, 2H, CH2OC(O)tBu), 4.17 (m, 4H,
OCH2CH3 ꢁ 2), 3.23, 3.04 (each s, 6H, CHN(CH3)2), 2.65 (m, 2H,
10-H ꢁ 2), 2.05 (m, 2H, 40-H ꢁ 2), 1.65 (m, 2H, 20-H ꢁ 2), 1.53 (m,
2H, 30-H ꢁ 2), 1.26 (t, 6H, OCH2CH3 ꢁ 2, J = 7.1 Hz), 1.10 (s, 9H,
tBu); 13C NMR (100 MHz, DMSO-d6) d 176.46, 158.81, 153.70,
152.17, 128.00, 120.99 (dt, JC,P = 213.6 Hz, JC,F = 259.6 Hz), 112.68,
112.15, 65.39, 64.06 (d, JC,P = 7.0 Hz, 2 carbons), 41.16, 38.30,
34.96, 33.12 (dt, JC,P = 14.6 Hz, JC,F = 20.3 Hz), 29.54, 26.64 (3 car-
bons), 22.94, 19.87 (m), 16.23 (d, JC,P = 5.2 Hz, 2 carbons); 19F
81.94, 69.10, 64.59 (d, JC,P = 7.0 Hz,
2 carbons), 33.37 (dt,
JC,P = 15.2 Hz, JC,F = 20.9 Hz), 16.38 (d, JC,P = 5.4 Hz, 2 carbons),
10.88 (m); 19F NMR (282 MHz, CDCl3)
d -50.61 (dt, 2F,
JF,H = 18.6 Hz, JF,P = 106.9 Hz); 31P NMR (122 MHz, CDCl3) d 6.97
(t, 1P, JP,F = 106.9 Hz); ESI-MS (LR) m/z 241 (MH+); ESI-MS (HR)
calcd for C9H16F2O3P (MH+): 241.0805, found: 241.0810.
4.1.5. 9-[50,50-Difluoro-50-(diethylphosphono)pent-10-ynyl]-2N-
(N,N-dimethylaminomethylidene)-7N-(2-nitrobenzene)
sulfonyl-1N-(pivaloyloxy)methyl-9-deazaguanine (18b)
To a solution of 4b (144 mg, 600
added Et3N (139 L, 1.00 mmol), PdCl2(PPh3)2 (7.0 mg, 10.0
and CuI (1.0 mg, 5.00 mol) at room temperature. After being stirred
for1 hat thesametemperature, tothemixturewasadded1(315 mg,
500 mol). After being stirred for 15 h at 60 °C, the mixture was
l
mol) in MeCN (2.5 mL) was
l
lmol)
l
l
NMR (376 MHz, DMSO-d6)
d
ꢀ49.41 (dt, 2F, JF,H = 20.4 Hz,
evaporated. The resulting residue was chromatographed on silica
gel (hexane/AcOEt = 1:1) to give 18b (244 mg, 66%) as a pale yellow
solid: UV kmax (MeOH) 320 nm, 216 nm; mp 192–194 °C; 1H NMR
(400 MHz, DMSO-d6) d 8.59 (s, 1H, CHNMe2), 8.34 (dd, 1H, CH of
Ns, J = 1.4 Hz, 7.9 Hz), 8.06 (dd, 1H, CH of Ns, J = 1.3 Hz, 7.9 Hz),
8.01–7.92 (m, 3H, CH of Ns ꢁ 2, 8-H), 6.02 (s, 2H, CH2OC(O)tBu),
4.23 (dq, 4H, OCH2CH3 ꢁ 2, JH,H = JH,P = 7.1 Hz), 3.18, 2.99 (each s,
JF,P = 107.1 Hz); 31P NMR (162 MHz, DMSO-d6) d 8.10 (t, 1P,
JP,F = 107.1 Hz); ESI-MS (LR) m/z 562 (MH+); ESI-MS (HR) calcd for
C24H39F2N5O6P (MH+): 562.2605, found: 562.2585.
4.1.8. 9-[50,50-Difluoro-50-(diethylphosphono)pentyl]-2N-(N,N-
dimethylaminomethylidene)-9-deazaguanine (23b)
To a solution of 22b (924 mg, 1.65 mmol) in MeOH (83 mL) was
added NaOMe (446 mg, 8.25 mmol). After being stirred for 22 h at
room temperature, the reaction mixture was quenched with 1 M
HCl solution at 0 °C and evaporated. The resulting residue was
chromatographed on silica gel (3% MeOH in CHCl3) to give 23b
6H, CHN(CH3)2), 2.74 (t, 2H, 30-H ꢁ 2, J3 ,4 = 7.6 Hz), 2.40 (m, 2H, 40-
0
0
t
H ꢁ 2), 1.30 (t, 6H, OCH2CH3 ꢁ 2, J = 7.1 Hz), 1.02 (s, 9H, Bu); 13C
NMR (100 MHz, DMSO-d6) d 176.46, 157.86, 156.65, 152.11,
149.73, 147.32, 136.44, 134.26, 133.02, 132.82, 129.88, 125.28,
119.96 (dt, JC,P = 214.0 Hz, JC,F = 260.3 Hz), 112.25, 103.42, 92.42,
70.74, 65.48, 64.37 (d, JC,P = 6.4 Hz, 2 carbons), 40.84, 38.22, 34.75,
32.59 (dt, JC,P = 14.9 Hz, JC,F = 20.3 Hz), 26.64 (3 carbons), 16.22 (d,
JC,P = 4.8 Hz, 2 carbons), 11.58 (m); 19F NMR (376 MHz, DMSO-d6) d
–50.74 (dt, 2F, JF,H = 19.5 Hz, JF,P = 104.0 Hz); 31P NMR (162 MHz,
DMSO-d6) d 7.44 (t, 1P, JP,F = 104.0 Hz); ESI-MS (LR) m/z 743 (MH+);
ESI-MS (HR) calcd for C30H38F2N6O10PS (MH+): 743.2075, found:
743.2028.
(604 mg, 82%) as
a
white solid: mp 110–113 °C; 1H NMR
(400 MHz, DMSO-d6) d 11.28 (s, 1H, 1-NH), 10.88 (s, 1H, 7-NH),
8.46 (s, 1H, CHNMe2), 6.99 (d, 1H, 8-H, J = 2.7 Hz), 4.16 (m, 4H,
OCH2CH3 ꢁ 2), 3.10, 2.98 (each s, 6H, CHN(CH3)2), 2.54 (t, 2H, 10-
H ꢁ 2, J1 ,2 = 7.4 Hz), 2.03 (m, 2H, 40-H ꢁ 2), 1.65 (tt, 2H, 20-H ꢁ 2,
0
0
J2 ,1 = J2 ,3 = 7.4 Hz), 1.51 (m, 2H, 30-H ꢁ 2), 1.25 (t, 6H,
OCH2CH3 ꢁ 2, J = 7.1 Hz); 13C NMR (100 MHz, DMSO-d6) d 156.66,
155.19, 153.54, 144.00, 124.73, 121.08 (dt, JC,P = 213.6 Hz,
JC,F = 258.5 Hz), 114.94, 64.05 (d, JC,P = 6.8 Hz, 2 carbons), 40.29,
34.41, 33.16 (dt, JC,P = 14.2 Hz, JC,F = 20.5 Hz), 29.34, 23.07, 20.04
(m), 16.22 (d, JC,P = 5.0 Hz, 2 carbons); 19F NMR (376 MHz, DMSO-
d6) d ꢀ49.37 (dt, 2F, JF,H = 20.5 Hz, JF,P = 106.8 Hz); 31P NMR
(162 MHz, DMSO-d6) d 8.16 (t, 1P, JP,F = 106.8 Hz); ESI-MS (LR) m/
0
0
0
0
4.1.6. 9-[50,50-Difluoro-50-(diethylphosphono)pent-10-ynyl]-2N-
(N,N-dimethylaminomethylidene)-1N-(pivaloyloxy)methyl-9-
deazaguanine (19b)
To a solution of 18b (152 mg, 172
l
mol) in DMF (5.8 mL) was
mol.) at
z
448 (MH+); ESI-MS (HR) calcd for C18H29F2N5O4P (MH+):
added PhSH (21.0 L, 206 mol), K2CO3 (71.0 mg, 516 l
l
l
room temperature. After being stirred for 1.5 h at the same tem-
perature, the reaction mixture was diluted with CHCl3 and washed
with brine, dried over MgSO4, and evaporated. The resulting resi-
due was chromatographed on silica gel (CHCl3) to give 19b
(86.5 mg, 90%) as a yellow solids: mp 161–166 °C; 1H NMR
(400 MHz, DMSO-d6) d 12.09 (s, 1H, NH), 8.51 (s, 1H, CHNMe2),
7.46 (d, 1H, 8-H, J = 3.2 Hz), 6.18 (s, 2H, CH2OC(O)tBu), 4.22 (m,
4H, OCH2CH3 ꢁ 2), 3.15, 2.97 (each s, 6H, CHN(CH3)2), 2.67 (t, 2H,
448.1925, found: 448.1927.
4.1.9. 9-[50,50-Difluoro-50-(diethylphosphono)pentyl]-9-
deazaguanine (DFPP-DG; 25)
A solution of 23b (342 mg, 764 lmol) in concd HCl solution
(12 mL) was refluxed for 20 h. The reaction mixture was diluted
with water and washed with AcOEt. The water layer was concen-
trated and co-evaporated with EtOH. The resulting precipitate
was suspended with water/MeOH (1:1) and collected by filtration
to give a solid. The solid was heated for 24 h at 40 °C in vacuo to
give 25 (238 mg, 93%) as a pale brown solid: mp >287 °C (colored);
30-H ꢁ 2, J3 ,4 = 7.7 Hz), 2.34 (m, 2H, 40-H ꢁ 2), 1.29 (t, 6H,
0
0
OCH2CH3 ꢁ 2, J = 7.1 Hz), 1.08 (s, 9H, tBu); 13C NMR (100 MHz,
DMSO-d6)
d 176.65, 156.79, 153.98, 153.49, 144.40, 130.30,
119.93 (dt, JC,P = 216.2 Hz, JC,F = 261.8 Hz), 113.29, 97.80, 88.35,
UV kmax (0.1 N HCl) 230 nm (e = 17,000), 274 nm (e = 12,700), UV

S. Hikishima et al. / Bioorg. Med. Chem. 18 (2010) 2275–2284
2283
kmax (pH 7.0 hepes buffer) 231 nm
(
e
= 17,500), 273 nm
= 18,400), 267 nm
= 6100); 1H NMR (400 MHz, D2O, NaOD) d
use of the weighted least-squares non-linear regression using the
program Leonora developed by Cornish-Bowden.23 Models of com-
petitive, uncompetitive, and mixed-type inhibition29 were com-
pared by analysis of variance. Inhibition constants were
determined by fitting the appropriate equation (for competitive,
uncompetitive and mixed-type inhibition models) to the whole
data set obtained for the inhibitor, that is, for all triplets: vo, co,
[I], where co is initial substrate concentration, vo is initial velocity
and [I] is inhibitor concentration.
(
(
e
e
= 10,100), UV kmax (0.1 N NaOH) 231 nm (
e
= 6800), 285 nm (
e
7.15 (s, 1H, 8-H), 2.62 (t, 2H, 10-H ꢁ 2, J1 ,2 = 7.0 Hz), 2.01 (m, 2H,
40-H ꢁ 2), 1.69–1.62 (m, 4H, 20-H ꢁ 2, 30-H ꢁ 2); 13C NMR
(100 MHz, D2O, NaOD) d 167.25, 162.22, 148.56, 128.03 (dt,
JC,P = 184.9 Hz, JC,F = 257.3 Hz), 127.71, 117.39, 116.50, 36.48 (dt,
JC,P = 12.3 Hz, JC,F = 21.5 Hz), 32.44, 25.92, 23.11 (m); 19F NMR
0
0
(376 MHz, D2O, NaOD) d ꢀ47.96 (dt, 2F, JF,H = 21.4 Hz, JF,P
=
87.7 Hz); 31P NMR (162 MHz, D2O, NaOD)
d
7.33 (t, 1P,
IC50s were also calculated from the experimental data by fitting
the exponential decay equation to the initial velocities, vo versus
inhibitor concentration [I] obtained for fixed concentration of sub-
JP,F = 87.7 Hz); ESI-MS (LR) m/z 337 (MH+); ESI-MS (HR) calcd for
C11H16F2N4O4P (MH+): 337.0877, found: 337.0856.
strates: 25 lM of 7-methylguanosine and 1 mM of phosphate.
4.2. Biological studies
The time dependence of inhibition was studied by incubation of
calf-spleen PNP (1–4 nM subunits), DFPP-DG and its analogues
(1–50 nM), and one substrate—1 mM phosphate—in hepes buffer
pH 7.0 at 25 °C in a total volume of 1.2 mL. After a given time inter-
val, t (0.5–20 min), 0.03 mL of the second substrate, m7Guo
4.2.1. Enzymatic procedures, inhibition types and inhibition
constants
Kinetic studies, if not otherwise indicated, were conducted at
25 °C in 50 mM hepes/NaOH buffer pH 7.0 in the presence of
1 mM phosphate buffer for the determination of the inhibition
constants and in the presence of 50 mM phosphate buffer for
determination of the enzyme specific activity.
(2000
l
M), was mixed with 0.97 mL of the incubated solution.
The final concentration of m7Guo was therefore 60
lM, with all
other concentration changed by only 3%, so may be treated as
equal to the initial values. The initial velocities observed after var-
One unit of PNP is defined as the amount of enzyme that causes
ious incubation times for each inhibitor concentration, vo (t, [I]),
phosphorolysis of 1
l
mol of Inosine (Ino) to hypoxanthine and ri-
were measured, that is, the rate of phosphorolysis (7-methylgua-
nine formation) using the spectrophotometric assay as described
above.
bose-1-phosphate per min under standard conditions i.e. at 25 °C
with 0.5 mM Ino and 50 mM sodium phosphate buffer pH 7.0. The
phosphorolysis of Ino was therefore measured to determine enzyme
specific activity. A standard coupled xanthine oxidase procedure28
was used with observation wavelength kobs = 300 nm and molar
For each inhibitor concentration, the velocity at equilibrium,
that is, at infinite time,
vo (1, [I]) (later referred to as vs[I]—stea-
dy-state velocity observed in the presence of inhibitor at [I]
concentration, was determined. This was done by fitting the one-
phase exponential decay to each set of velocities observed with
extinction coefficient difference De .
300 nm = 9600 Mꢀ1 cmꢀ1 22a
PNP is known for its non-hyperbolic kinetics and deviations form
the classical Michaelis–Menten kinetics depend on the nucleoside
substrate and concentration of the co-substrate, phosphate.22a
Therefore, inhibition type and inhibition constants were deter-
mined, if not otherwise indicated, using 7-methylguanosine as the
variable substrate since it was shown that for this substrate the clas-
sical Michaelis–Menten29 equation is sufficient for data analysis.22a
The phosphorolysis of 7-methylguanosine was examined spec-
trophotometrically by a direct method.30 Observation wavelength,
kobs = 260 nm corresponds to the maximal difference between the
extinction coefficients of the nucleoside substrate, m7Guo, and
various [I],
vo (t, [I]):
voðt; ½IꢂÞ ¼ A expðꢀktÞ þ vs½Iꢂ
ð1Þ
In separate experiments kc was determined as the initial veloc-
ity obtained at time t = 0 (no incubation) with a saturating concen-
tration of m7Guo (120
v
l
M) and in the absence of inhibitor, that is,
o (0; [0]) = kc. It was also found that 120 min incubation has no
influence on enzyme activity, hence one may assume that o (0;
v
[0]) = vo (120 min; [0]). The Michaelis constant was determined
as previously described,22a and the value obtained Km = 17
lM,
the respective purine base, 7-methylguanine:
De
= 4 600 Mꢀ1 cm
ꢀ1 at 260 nm at pH 7.0 for the mixture of cationic and zwitterionic
forms of m7Guo.21a,30
was used in the following calculations.
The inhibition constant at equilibrium, Kei q, was finally deter-
mined as reported previously for immucillins24, from the equation:
The reaction mixture for the direct method and for the coupled
method had 1 mL volume in a 10-mm path-length cuvette and was
kept at 25 °C. It contained 50 mM Hepes pH 7.0, both substrates of
the phosphorolytic reaction (phosphate buffer of the same pH as
the main buffer, and a nucleoside, m7Guo or Ino). In the case of
Ino phosphorolysis, xanthine oxidase was also present (about
0.1 U/mL), while in the case of inhibition studies an inhibitor was
also included in the reaction mixture. If not otherwise indicated,
the reaction was started by the addition of PNP (calf or human).
Standard initial rate procedures were employed in all kinetic
studies In the case of inhibition studies for each combination of
the initial substrate concentration co and the inhibitor concentra-
tion [I], the rates were determined at least two times. The initial
Vs½Iꢂ=kc ¼ ½Sꢂ=ðKmð1 þ ½Iꢂ=KeqÞ þ ½SꢂÞ
ð2Þ
i
where [S] is the concentration of m7Guo (60
constants.
lM), and Km and kc are
4.2.2. Extinction coefficients and pKa determination
Extinction coefficients and pKa were determined spectrophoto-
metrically. Britton–Robisnon universal buffers31 were used to get
buffers covering the broad pH range 1.8–12. In addition, spectra
at pH 1.0 and 13.0 (in 0.1 N HCl and in 0.1 N NaOH) were recorded.
Two milligrams of an inhibitor (DFPP-DG or one of its ana-
logues) were dissolved in 1–10 mL DMSO (depending on solubility
of the compound examined) to get stock solutions of several mM
concentrations. From the stock solution samples, desired pH was
prepared in a two-step dilution. In the first, step 0.1–0.3 mL of
the stock was added to 50 mL of water, and in the next step,
0.9 mL of the solution in water was mixed with 0.1 mL of the de-
sired buffer. The spectra of these solutions were recorded in a
range of 220–360 nm and used to determine extinction coefficients
and pKa. pKa was determined by fitting the sigmoidal dose–
velocities, vo, were measured directly from the computer control-
ling the spectrophotometer. A linear regression program was used
for the determination of the slopes, with their standard errors, of
absorbance versus time.
Data were initially presented in a form of Michaelis–Menten,
Lineveawer–Burk, and Dixon plots to visualize kinetic data ob-
tained in the particular experiment and to visually inspect the type
of inhibition observed, but not to calculate kinetic constants. Initial
rates were analyzed and kinetic constants were obtained with the

2284
S. Hikishima et al. / Bioorg. Med. Chem. 18 (2010) 2275–2284
M.; Koyanagi, S.; Soeda, S.; Shimeno, H.; Shibuya, S.; Yokomatsu, T. Bioorg. Med.
Chem. 2006, 14, 1660.
6. Luic, M.; Koellner, G.; Yokomatsu, T.; Shibuya, S.; Bzowska, A. Acta Crystallogr.,
Sect. D 2004, 60, 1417. PDB-reference number for the PNP-DFPP-G complex:
1V48.
7. (a) Montgomery, J. A.; Niwas, S.; Rose, J. D.; Secrist, J. A., III; Babu, Y. S.; Bugg, C.
E.; Erion, M. D.; Guida, W. C.; Ealick, S. E. J. Med. Chem. 1993, 36, 55; (b) Erion,
M. D.; Niwas, S.; Rose, J. D.; Ananthan, S.; Allen, M.; Secrist, J. A., III; Babu, Y. S.;
Bugg, C. E.; Guida, W. C.; Ealick, S. E.; Montgomery, J. A. J. Med. Chem. 1993, 36,
3771.
8. Kicska, G. A.; Tyler, P. C.; Evans, G. B.; Furneaux, R. H.; Schramm, V. L.; Kim, K. J.
Biol. Chem. 2002, 277, 3226.
9. Hikishima, S.; Hashimoto, M.; Magnowska, L.; Bzowska, A.; Yokomatsu, T.
Bioorg. Med. Chem. Lett. 2007, 17, 4173.
10. For a review of Pd-mediated cross-coupling reactions to nuclosides: Agrofoglio,
L. A.; Gillaizeau, I.; Saito, Y. Chem. Rev. 2003, 103, 1875.
11. Taylor, E. C.; Young, W. B. J. Org. Chem. 1995, 60, 7947.
12. Lwonard, P.; Wiglenda, T.; Seela, F. Nucleoside, Nucleotides, Nucleic Acids 2001,
20, 1279.
13. (a) Yokomatsu, T.; Suemune, K.; Murano, T.; Shibuya, S. J. Org. Chem. 1996, 61,
7207; (b) Yokomatsu, T.; Murano, T.; Suemune, K.; Shibuya, S. Tetrahedron
1997, 53, 815; (c) Yokomatsu, T.; Ichimura, A.; Kato, J.; Shibuya, S. Synlett 2001,
287.
14. (a) Berkowitzt, D. B.; Eggen, M.-J.; Shen, Q.; Sloss, D. G. J. Org. Chem. 1993, 58,
6174; (b) Shen, Q.; Sloss, D. G.; Berkowitz, D. B. Synth. Commun. 1994, 24, 1519.
15. Xiang, J.; Fuchs, P. L. Tetrahedron Lett. 1998, 39, 8597.
16. Bian, J.; Wingerden, M. V.; Ready, J. M. J. Am. Chem. Soc. 2006, 128, 7428.
17. Zhang, H.; Larock, R. C. J. Org. Chem. 2002, 67, 7048.
18. Recently, TBAF-mediated de-nosylation of nosyl esters have been reported:
Koyama, Y.; Yamaguchi, R.; Suzuki, K. Angew. Chem., Int. Ed. 2008, 47, 1084.
19. Porter, D. J. J. Biol. Chem. 1992, 267, 7342.
20. (a) Koellner, G.; Luic´, M.; Shugar, D.; Saenger, W.; Bzowska, A. J. Mol. Biol. 1997,
265, 202; (b) Mao, C.; Cook, W. J.; Zhou, M.; Fedorov, A. A.; Almo, S. C.; Ealick, S.
E. Biochemistry 1998, 37, 7135.
response curve to the absorbance versus pH, that is, assuming
titration of one ionizable proton:
aꢀpH
A ðpHÞ ¼ Ao þ
D
A ð1 þ 10pK
Þ
where Ao is absorbance of one ionic form and
DA is the difference of
absorbance of two ionic forms ( A could be positive or negative,
D
depending of the observation wavelength). Such a fitting was done
separately for several (3–5) observation wavelengths. The mean
values were then calculated for a particular compound and those
are given in Section 2.
DMSO concentrations in solutions used for spectral and inhibi-
tory experiments were typically up to 0.2% and never exceeded
0.6%.
Acknowledgements
This work was supported in part by a Grant-in-Aid for Scientific
Research (C) from the Ministry of Education, Culture, Sports, Sci-
ence and Technology of Japan and by the Polish Ministry of Science
and Higher Education grant N301 003 31/0042. The Promotion and
Mutual Aid Corporation for Private School of Japan is thanked for
supporting this work.
Supplementary data
Supplementary data (additional experimental details: synthetic
procedures and physical data for all new compounds not described
in the experimental section, and materials and instrumentation for
the biological studies) associated with this article can be found, in
the online version, at doi:10.1016/j.bmc.2010.01.062.
21. (a) Wiezchowski, J.; Bzowska, A.; Stepniak, K.; Shugar, D. Z. Naturforsch. 2004,
ˇ
´
59c, 713; (b) Stepniak, K.; Zinic, B.; Wiezchowski, J.; Bzowska, A. Nucleosides,
Nucleotides, Nucleic Acids 2007, 26, 841; (c) Ghanem, M.; Saen-oon, S.; Zhadin,
N.; Wing, C.; Cahill, S. M.; Schwartz, S. D.; Callender, R.; Schramm, V. L.
Biochemistry 2008, 47, 3202.
22. (a) Bzowska, A. Biochim. Biophys. Acta 2002, 1596, 293; (b) Iwanow, M.;
Magnowska, L.; Yokomatsu, T.; Shibuya, S.; Bzowska, A. Nucleosides,
Nucleotides, Nucleic Acids 2003, 22, 1567; (c) Wierzchowski, J.; Bzowska, A.;
Ste˛pniak, K.; Shugar, D. Z. Naturforsch 2004, 59c, 713; (d) Wielgus-Kutrowska,
B.; Bzowska, A. Biochim. Biophys. Acta 2006, 1764, 887.
References and notes
1. (a) Zhang, Z.-Y. Account Chem. Res. 2003, 36, 385; (b) Thaimattam, R.; Banerjee,
R.; Miglani, R.; Iqbal, J. Curr. Pharm. Design 2007, 13, 2751.
23. Cornish-Bowden, A. Analysis of the enzyme kinetic data; Oxford University
Press: New York, 1995.
2. Bzowska, A.; Kulikowska, E.; Shugar, D. Pharmacol. Ther. 2000, 88, 349.
3. (a) Nakamura, C. E.; Chu, S.-H.; Stoeckler, J. D.; Parks, R. E. P., Jr. Biochem.
Pharmacol. 1986, 35, 133; (b) Ealick, S. E.; Babu, Y. S.; Bugg, C. E.; Erion, M. D.;
Guida, W. C.; Montgomery, J. A.; Secrist, J. A., III Proc. Natl. Acad. Sci. U.S.A. 1991,
88, 11540; (c) Elliott, R. D.; Niwes, S.; Riordan, J. M.; Montgomery, J. A.; Sercrist,
J. A., III Nucleoside Nucleotides 1992, 11, 97; (d) Kelly, J. L.; McLean, E. W.;
Crouch, R. C.; Averett, D. R.; Tuttle, J. V. J. Med. Chem. 1995, 38, 1005.
4. Halazy, S.; Ehrhard, A.; Danzin, C. J. Am. Chem. Soc. 1991, 113, 315.
5. (a) Yokomatsu, T.; Abe, H.; Sato, M.; Suemune, K.; Kihara, T.; Soeda, S.;
Shimeno, H.; Shibuya, S. Bioorg. Med. Chem. 1998, 6, 2495; (b) Yokomatsu, T.;
Hayakawa, Y.; Suemune, K.; Kihara, T.; Soeda, S.; Shimeno, H.; Shibuya, S.
Bioorg. Med. Chem. Lett. 1999, 9, 2833; (c) Yokomatsu, T.; Hayakawa, Y.; Kihara,
T.; Koyanagi, S.; Soeda, S.; Shimeno, H.; Shibuya, S. Bioorg. Med. Chem. 2000, 8,
2571; (d) Yokomatsu, T.; Yamagishi, T.; Suemune, K.; Abe, H.; Kihara, T.; Soeda,
S.; Shimeno, H.; Shibuya, S. Tetrahedron 2000, 56, 7099; (e) Hikishima, S.; Isobe,
24. Miles, R. W.; Tyler, P. C.; Furneaux, R. H.; Bagdassarian, C. K.; Schramm, V. L.
Biochemistry 1998, 37, 8615.
25. Morrison, J. F.; Walsh, C. T. Adv. Enzymol. Relat. Areas Molec. Biol. 1988, 61, 201.
26. Breer, K.; Glavaš-Obrovac, L.; Suver, M.; Hikishima, S.; Hashimoto, M.;
Yokomatsu, T.; Wielgus-Kutrowska, B.; Magnowska, L.; Bzowska, A. FEBS J., in
press.
27. Blackburn, G. M.; Kent, D. E. J. Chem., Perkin Trans. 1 1986, 913.
28. Stoeckler, J. D.; Agarwal, R. P.; Agarwal, K. C.; Parks, R. E., Jr. Methods Enzymol.
1978, 51, 530.
29. Segel, I. H. Enzyme kinetics; John Wiley and Sons: New York, 1975.
30. Kulikowska, E.; Bzowska, A.; Wierzchowski, J.; Shugar, D. Biochim. Biophys. Acta
1986, 874, 355.
31. Data for Biochemical Research; Dawson, R. M. C., Elliott, D., Elliott, W. H., Jones,
K. M., Eds.; Oxford University Press: New York, 1969; pp 145–179.