Synthesis of novel N-branched acyclic nucleoside phosphonates as potent and selective inhibitors of human, plasmodium falciparum and plasmodium vivax 6-oxopurine phosphoribosyltransferases

Article abstract of DOI:10.1021/jm300662d

Hypoxanthine-guanine-(xanthine) phosphoribosyltransferase (HG(X)PRT) is crucial for the survival of malarial parasites Plasmodium falciparum (Pf) and Plasmodium vivax (Pv). Acyclic nucleoside phosphonates (ANPs) are inhibitors of HG(X)PRT and arrest the growth of Pf in cell culture. Here, a novel class of ANPs containing trisubstituted nitrogen (aza-ANPs) has been synthesized. These compounds have a wide range of K_i values and selectivity for human HGPRT, PfHGXPRT, and PvHGPRT. The most selective and potent inhibitor of PfHGXPRT is 9-[N-(3-methoxy-3-oxopropyl)-N-(2-phosphonoethyl)-2-aminoethyl] hypoxanthine (K_i = 100 nM): no inhibition could be detected against the human enzyme. This compound exhibits the highest ever reported selectivity for PfHGXPRT compared to human HGPRT. For PvHGPRT, 9-[N-(2-carboxyethyl)-N-(2- phosphonoethyl)-2-aminoethyl]guanine has a K_i of 50 nM, the best inhibitor discovered for this enzyme to date. Docking of these compounds into the known structures of human HGPRT in complex with ANP-based inhibitors suggests reasons for the variations in affinity, providing insights for the design of antimalarial drug candidates.

Full text of DOI:10.1021/jm300662d

Article
pubs.acs.org/jmc
Synthesis of Novel N-Branched Acyclic Nucleoside Phosphonates As
Potent and Selective Inhibitors of Human, Plasmodium falciparum
and Plasmodium vivax 6-Oxopurine Phosphoribosyltransferases
Dana Hockova,*,† Dianne T. Keough,‡ Zlatko Janeba,† Tzu-Hsuan Wang,‡ John de Jersey,‡
́
and Luke W. Guddat‡
†Institute of Organic Chemistry and Biochemistry, Academy of Sciences of the Czech Republic, vvi Flemingovo nam. 2, CZ-166 10
Prague 6, Czech Republic
‡The School of Chemistry and Molecular Biosciences, The University of Queensland, Brisbane, 4072 QLD, Australia
ABSTRACT: Hypoxanthine−guanine−(xanthine) phosphoribosyl-
transferase (HG(X)PRT) is crucial for the survival of malarial
parasites Plasmodium falciparum (Pf) and Plasmodium vivax (Pv).
Acyclic nucleoside phosphonates (ANPs) are inhibitors of HG(X)-
PRT and arrest the growth of Pf in cell culture. Here, a novel class of
ANPs containing trisubstituted nitrogen (aza-ANPs) has been
synthesized. These compounds have a wide range of Ki values and
selectivity for human HGPRT, PfHGXPRT, and PvHGPRT. The
most selective and potent inhibitor of Pf HGXPRT is 9-[N-(3-
methoxy-3-oxopropyl)-N-(2-phosphonoethyl)-2-aminoethyl]-
hypoxanthine (Ki = 100 nM): no inhibition could be detected against the human enzyme. This compound exhibits the highest
ever reported selectivity for Pf HGXPRT compared to human HGPRT. For PvHGPRT, 9-[N-(2-carboxyethyl)-N-(2-
phosphonoethyl)-2-aminoethyl]guanine has a Ki of 50 nM, the best inhibitor discovered for this enzyme to date. Docking of
these compounds into the known structures of human HGPRT in complex with ANP-based inhibitors suggests reasons for the
variations in affinity, providing insights for the design of antimalarial drug candidates.
other class of HG(X)PRT inhibitors known so far are the
transition state analogues, (1S)-1-(9-deazahypoxanthin-9-yl)-
1,4-dideoxy-1,4-imino-D-ribitol 5-phosphate (immucillinHP)
and (1S)-1-(9-deazaguanin-9-yl)-1,4-dideoxy-1,4-imino-D-ribi-
tol 5-phosphate (immucillinGP).8 The activity of HG(X)PRT
is necessary for both survival and reproduction of many
protozoan parasites.9 This is because, unlike mammalian cells,
protozoan parasites such as Plasmodium, the causative agent of
malaria, are auxotrophic for the purine ring. Thus, they depend
on HG(X)PRT for the synthesis of the 6-oxopurine nucleoside
monophosphates required for RNA/DNA production.10,11 The
reaction catalyzed by HG(X)PRT is given in Figure 1.
There are four major species of Plasmodium that infect
humans and result in symptoms of malaria: falciparum, vivax,
malariae, and ovale. Pf and Plasmodium vivax (Pv) are the most
lethal and widespread, with both species infecting around 500
million people per year, resulting in at least one million deaths
per annum, mainly among children.12 Drugs such as artemisinin
and combination therapies, quinine, chloroquine, and prima-
quine, are the only known treatments for malaria but, because
of increasing resistance13 as well as cost-effectiveness, there is
an urgent need for the discovery of new targets and therapeutic
leads for the development of potent antimalarials.
INTRODUCTION
■
Acyclic nucleoside phosphonates (ANPs) are reverse tran-
scriptase inhibitors, and several ANP-based drugs are in current
clinical use for the treatment of serious viral infections (e.g.,
Cidofovir, Adefovir, Tenofovir).1,2 These compounds consist of
a nucleobase, either 6-aminopurine or pyrimidine, linked to a
phosphonate group by an acyclic linker. These nucleotide
analogues are excellent templates for drug design because of the
absence of the labile glycosidic bond of the nucleotides and the
stability of the phosphonate moiety compared with the
phosphate ester bond of the nucleotides. Such bonds can be
enzymatically hydrolyzed within the cell to yield inactive
compounds. The chemical structure of the ANPs also allows
the synthesis of prodrugs to increase membrane permeability
and bioavailability. Another structural advantage of this class of
molecules is the flexibility of the acyclic chain, which enables
the compounds to adopt a range of conformations to facilitate
binding to not only reverse transcriptases and DNA
polymerases but potentially to other enzymes using nucleotides
as substrates or products.
Some ANPs with 6-oxopurine bases have been found to
arrest parasitemia in Plasmodium falciparum (Pf) grown in
culture with IC50 values as low as 1 μM.3,4 These compounds
are potent inhibitors of hypoxanthine−guanine−(xanthine)
phosphoribosyltransferase (HG(X)PRT), and it is proposed
that this is how they exert their antimalarial effect.4−7 The only
Received: May 11, 2012
Published: June 22, 2012
© 2012 American Chemical Society
6209
dx.doi.org/10.1021/jm300662d | J. Med. Chem. 2012, 55, 6209−6223

Journal of Medicinal Chemistry
Article
Pf HGXPRT.4 They have Ki values for human HGPRT of 1
and 3.6 μM, respectively, and 0.1 and 0.3 μM for Pf HGXPRT.
In all of the new compounds (aza-ANPs), there are five atoms
between the N-9 atom in the purine ring and the phosphonate
phosphorus atom. Different side chains were attached to the
linker nitrogen atom in attempts to improve the affinity and/or
selectivity for the Plasmodium enzymes. Herein is reported the
synthesis of 24 new ANPs and a comparison of their inhibition
constants for human HGPRT, PfHGXPRT, and PvHGPRT.
Docking of the aza-ANPs into the known crystal structures4 of
human HGPRT in complex with 2-(phosphonoethoxy)ethyl
guanine or hypoxanthine (PEEG or PEEHx) and 3-hydroxy-2-
(phosphonomethoxy)propyl guanine (HPMPG) provides in-
sights into how these compounds may bind in the active site.
Figure 1. The reaction catalyzed by HG(X)PRT. R = −H
(hypoxanthine), −NH2 (guanine), −OH (xanthine).
The subject of this paper is the design and synthesis of a new
series of ANPs with novel linkers designed to increase affinity
and/or selectivity for Pf HGXPRT and PvHGPRT over the
corresponding human HGPRT. These compounds are based
on the 2-(phosphonoethoxy)ethyl (PEE) ANPs but with a
nitrogen atom replacing oxygen in the phosphonate tail (Figure
2). This scaffold was chosen as both 2-(phosphonoethoxy)ethyl
RESULTS AND DISCUSSION
■
Chemical Synthesis. The design of the aza-ANPs is based
on previous results from SAR studies.4−7 ANPs containing a
nitrogen atom in the acyclic chain (aza-ANPs, Figure 2B) have
three main advantages for inhibitor development: (1)
derivatives with different linker lengths and nitrogen position
(determining the location of the side chain) can be easily
synthesized, (2) a variety of synthetic methods can be applied
to attach a side chain with various functional groups to the NH-
moiety, and (3) difficulties with the synthesis and character-
ization of stereoisomers, typical for C-branched ANPs, are
avoided.
Synthesis of several compounds containing a nucleobase as
well as an acyclic chain with substituted nitrogen has recently
been published.14−16 However, only one type of acyclic
nucleoside phosphonate (with an unsubstituted NH-moiety)
has previously been reported.17
A number of different synthetic procedures can be used for
carbon−nitrogen bond formation. One of these is the aza-
Michael addition, and thus reaction of vinylphosphonate with
an amino group was chosen for the synthesis of the β-
aminophosphonate moiety. The use of various solvents and
catalysts for the reaction has been described in the literature,
but we applied the recently published methodology using only
Figure 2. Comparison of the structure of PEE compounds (A) with
the aza-ANPs (B). Y and n denote the variety of functional groups in
the side chain.
guanine (PEEG) and 2-(phosphonoethoxy)ethyl hypoxanthine
(PEEHx) are good inhibitors of human HGPRT and
a
Scheme 1
a
(i) H2O, rt; (ii) TrCl, Et3N, DMAP, CH2Cl2; (iii) K2CO3, MeCN.
6210
dx.doi.org/10.1021/jm300662d | J. Med. Chem. 2012, 55, 6209−6223

Journal of Medicinal Chemistry
Article
a
Scheme 2
a
(i) PPh3, DIAD, THF; (ii) addition of H2O, heating; (iii) 75% aq trifluoroacetic acid; (iv) Me3SiBr, MeCN, DMF, 2,6-lutidine; (v) (1) aq NaOH,
THF, MeOH, (2) Me3SiBr, MeCN, DMF, 2,6-lutidine; (vi) (1) aq NaOH, MeOH, (2) Me3SiBr, MeCN, DMF, 2,6-lutidine; (vii) DMF, heating.
water without any catalyst or organic cosolvent.18 Thus, aza-
Michael addition of diethyl vinylphosphonate to 1 equiv of 2-
aminoethanol produced desired phosphonate 1 (77% yield)
together with bisphosphonate 2a (8% yield).
triphenylphosphoranylidene intermediate was very stable when
the trityl group was present. After deprotection under acidic
conditions (the 6-chloro group was unexpectedly completely
preserved), the triphenylphosphoranylidene 4i was decom-
posed using standard procedures.
The chlorine atom of phosphonates 3a−3h and 4a−4g,4i
was next displaced with hydroxyl by nucleophilic aromatic
substitution in acidic conditions (75% aqueous trifluoroacetic
acid) to form hypoxanthine and guanine phosphonates 5a−5g
and 6a−6g,6i. The tritylated derivative 3h was simultaneously
deprotected to give the hydroxyderivative 5i. Instead of direct
Mitsunobu reaction of 6-oxopurines with poor regioselectivity,
this two-step approach was used for the preparation of 9-
substituted hypoxanthine and guanine derivatives to avoid the
separation of N7 and N9 regioisomers.
To form free phosphonic acids 7b−7g,7i and 8b−8g, 8i,
ester groups of diethyl phosphonates 5b−5g,5i and 6b−6g,6i
were cleaved under standard conditions using Me3SiBr/
acetonitrile followed by hydrolysis; 2,6-lutidine was added to
neutralize contamination of Me3SiBr by HBr. In the tetraethyl
bisphosphonates 5a and 6a, both phosphonate moieties were
simultaneously deprotected and free bisphosphonic acids 7m
and 8m were obtained. Alkaline hydrolysis of carboxylic acid
esters 5b−5d, 6b, and 6d with aqueous sodium hydroxide
followed by cleavage of phosphonate esters under the above-
mentioned conditions afforded the corresponding phosphonic
acids 7j−7l, 8j, and 8l with free carboxylic acid moiety in the
side chain. The phosphonic acid 8k was prepared by the
alkaline hydrolysis of the cyanoderivative 6f again followed by
cleavage of the phosphonate ester (Scheme 2).
Diethyl 2-(2-hydroxyethylamino)ethylphosphonate 1 was
used as the starting material for the synthesis of a series of
branched phosphonates 2b−2g (Scheme 1). Two types of
reactions were used to introduce suitable substituents to the
NH-group of the phosphonate 1: (a) alkylation with the
corresponding halogenoderivative was used to attach a one-
carbon or three-carbon chain bearing a cyano or carboxylic
ester group (compounds 2b, 2d, 2e, and 2g, isolated yields 64−
77%), (b) Another aza-Michael addition of methyl acrylate or
acrylonitrile to the phosphonate 1 was employed to prepare
compounds 2c and 2f with a two-carbon chain (isolated yields
50% and 78%, respectively). To complement the series, a
protected hydroxyethyl derivative 2h was synthesized by aza-
Michael addition of diethyl vinylphosphonate to diethanol-
amine followed by tritylation of one of the hydroxy groups
(Scheme 1).
The series of hydroxyderivatives 2a−2h prepared by the
above-mentioned straightforward approach was used to
introduce branched acyclic moieties to the N9-position of 6-
chloropurine or 2-amino-6-chloropurine (Scheme 2) via
Mitsunobu reaction.19,20 6-Chloropurine phosphonates 3a−3h
were produced in good yields (about 80%). In the case of 2-
amino-6-chloropurine phosphonates 4a−4g, the Mitsunobu
reaction had to be followed by heating in water/tetrahydrofur-
an to decompose the triphenylphosphoranylidene intermediate
rising from the presence of the free amino group.17 Anomalous
behavior was observed for the tritylated derivative 4h: the
6211
dx.doi.org/10.1021/jm300662d | J. Med. Chem. 2012, 55, 6209−6223

Journal of Medicinal Chemistry
Article
Table 1. Comparison of the Ki Values of the PEE Compounds with Aza-ANP
a
ND = not determined.
During the purification of the cyanomethyl phosphonic acids
7e and 8e, partial cleavage of the side-chain at higher
temperature was observed. To complement the series of target
ANPs, we refluxed these unstable cyanomethyl derivatives in
dimethylformamide for several days to obtain compounds 7n
and 8n with an unsubstituted NH-moiety (Scheme2). All target
ANPs described above, 7b−7g, 7i−7n, 8b−8g, and 8i−8n,
were purified by preparative HPLC or ion-exchange chroma-
tography.
The cytotoxicity of the prepared aza-ANPs was tested on
several cell lines (HEL, Vero, CRFK), and none of the
compounds exhibited any significant toxicity (CC50 > 100 μg/
mL) except for the cyano derivative 8e (CC50 = 53 μg/mL,
HEL cells). The title compounds were also tested for their in
vitro inhibition of the cell growth in mouse leukemia L1210
cells, human T-lymphoblastoid CCRF-CEM cell line, human
promyelocytic leukemia HL-60 cells, and human cervix
GMP are similar to that found for this new aza-ANP 8n: 5.8,
10, and 26.2 μM for human HGPRT, PfHGXPRT, and
PvHGPRT, respectively. The Ki values for IMP for these three
enzymes are 5.4, 3.6, and 62 μM for human HGPRT,
Pf HGXPRT and PvHGPRT, respectively, not significantly
different from those found for GMP. When PEEG or PEEHx
bind to human HGPRT, the loop does become visible and
partially closes.4 Thus, it is likely that one of the contributing
factors leading to the weaker binding of 7n and 8n compared
with PEEG and PEEHx is the failure of the loop to become
ordered. Therefore, when 7n or 8n bind to the enzymes, the
structure is likely to be similar to that when GMP or IMP are
present and commencing to leave the active site.
Addition of a Side Chain with a Second Phosphonate
Group to the Basic Linker. In an attempt to improve the
affinity of these new ANPs for the Plasmodium enzymes, a
second phosphonate group was attached via a two-carbon chain
to the linker nitrogen atom (7m and 8m). The hypothesis was
that this second phosphonate group may be able to occupy part
of the pyrophosphate binding site, increasing the affinity. The
Ki values for these new compounds are given in Table 3. For
human HGPRT, Pf HGXPRT, and PvHGPRT, these values are
3, 0.4, and 13 μM, respectively (7m, hypoxanthine as the purine
base). When hypoxanthine is replaced by guanine, Ki values are
lower for all three enzymes. For human HGPRT, Pf HGXPRT,
and PvHGPRT, they are 0.1. 0.3 and 0.9 μM, respectively
(8m). Thus, the addition of this second phosphonate group
does affect the affinity of these ANPs for all three enzymes,
decreasing the Ki for all these enzymes significantly. The
decrease in the Ki values when the hydrogen on the nitrogen
atom in the linker is replaced by a phosphonate group is
reflected by comparison of their Ki ratios (Table 2).
́
́ ́
carcinoma HeLa S3 cells (Dr. H. Kaiserova-Mertlıkova,
IOCB, Prague), and no cytostatic activity was observed.
Inhibition of Human HGPRT, Pf HGXPRT, and
PvHGPRT by the Aza-ANPs. Effect of the Substitution of
the Oxygen Atom with Nitrogen. The substitution of the
oxygen atom in the acyclic linker (as in the PEE compounds;4
Figure 2A) by −NH− (compounds 7n and 8n) resulted in an
increase in the Ki values for human HGPRT and PfHGPRT
(Table 1).
These new ANPs are also weak inhibitors for PvHGPRT.
The Ki for 8n is 11 μM, similar to that found for Pf HGXPRT,
while for the hypoxanthine derivative, 7n, no inhibition was
detected (Ki is ≥200 μM). One possible explanation for these
increased Ki values is that the substitution of the partially
negatively charged oxygen atom by the positively charged
nitrogen changes the conformation of the linker, resulting in
different interactions between the ligand and the active site
amino acid residues. These aza-ANPs are competitive inhibitors
with PRib-PP. Thus, they bind in the same location in the active
site as the naturally occurring substrates or products of the
reaction. However, they may not be able to induce the
conformational changes which occur in the enzyme on the
binding of the transition state analogue inhibitors.8 When
immucillinGP 5′-phosphate, PPi and, Mg2+ bind to human
HGPRT, a large flexible loop closes over the active site, firmly
anchoring them in position. ANPs are analogues of the
nucleoside monophosphate products of the reaction, IMP and
GMP. In the crystal structure of GMP bound to human
HGPRT, this loop is only partially visible. The Ki values4 for
However, as the Ki values for all three enzymes are similar
(8m and 7m), selectivity is lost (Table 3). A series of
compounds with different substituents at the nitrogen atom
Table 2. Decrease in the Ki Value When the Hydrogen Atom
Attached to the Nitrogen Atom in the Linker Is Replaced by
a Phosphonate Group
Ki ratio (8n/8m) guanine
Ki ratio (7n/7m)
hypoxanthine as the base
enzyme
as the base
humanHGPRT
PfHGXPRT
PvHGPRT
15-fold
53-fold
12-fold
13-fold
>500-fold
>20-fold
6212
dx.doi.org/10.1021/jm300662d | J. Med. Chem. 2012, 55, 6209−6223

Journal of Medicinal Chemistry
Article
Table 3. Ki Values of Aza-ANPs for Human HGPRT, Pf HGXPRT, and PvHGPRT
a
b
ND, not determined. No inhibition could be observed under conditions where measurable inhibition would have been detected if the Ki value had
been less than or equal to the specified value.
were thus synthesized to try to increase both potency and
selectivity.
preferred the hypoxanthine derivative by 17-fold. Thus, it can
be deduced that for this class of ANPs, human HGPRT and
PvHGPRT usually prefer guanine as the purine base while
Pf HGXPRT is not as discriminatory.
Aza-ANPs with Various Substituents to the Linker
Nitrogen Atom. The basic structure of aza-ANPs was modified
by attaching various substituents to the nitrogen atom of the
linker. These substituents differed in the chemical nature of the
functional groups and the number of carbon atoms between the
trisubstituted nitrogen and the functional group. The structures
and Ki values for the three enzymes are given in Table 3.
Effect of the Purine Base on the Ki Values for All Three
Enzymes. For human HGPRT, guanine is always the preferred
base, with selectivity ratios in favor of guanine over
hypoxanthine of 23−220 (compounds 8b−8m, 7b−7n). As a
substrate, guanine binds slightly more tightly than hypoxan-
thine (cf. Km for guanine is 1.9 μM and for hypoxanthine is 3.4
μM),21 perhaps because guanine has the ability to form an extra
hydrogen bond between the 2-amino group and the active site
amino acid residues. However, this Km difference is only 2-fold,
suggesting that it is the structure or chemical nature of the
linker and the groups attached to the linker that contribute to
the increase of the Ki ratio in favor of guanine-containing aza-
ANPs. PfHGXPRT exhibited similar Ki values for the majority
of these compounds irrespective of whether guanine or
hypoxanthine was the base. There were two exceptions: 8i/7i
(13-fold in favor of guanine) and 8d/7d (1.3 μM for guanine
and ≥220 μM for hypoxanthine). PvHGPRT favored the ANPs
with guanine as the purine base, with ratios between 2 and 820.
The exception was compounds 8l and 7l, where PvHGPRT
Effect of the Nature of the Linker and the Different
Attachments. For the parasite enzymes, the optimal number of
carbon atoms between the trisubstituted nitrogen and the
substituent functional group was usually two (Table 3).
Compounds where the functional group was −COOMe or
−CN or −COOH or −OH were all effective inhibitors of
Pf HGXPRT. PvHGPRT is different in that it has a general
preference for cyano derivatives although its lowest Ki value was
found when the base was guanine and the functional group was
−COOH (8f; 50 nM). The most selective compounds tested
for the malarial enzymes were 7c and 7k as no inhibitory
activity could be observed against human HGPRT (Ki ≥200
μM), but they were good inhibitors of Pf HGXPRT (0.1 and
0.4 μM, respectively).
Docking of the ANPs into the Active Site of Human
HGPRT. Attempts were made to dock all of the ANPs in Tables
1 and 3 into the active site of the known crystal structures of
human HGPRT. Docking into the free structure of human
HGPRT or into the enzyme in complex with immucillinGP and
Mg2+-pyrophosphate was unsuccessful. These structures are not
good models for the binding of the ANPs because a number of
conformational changes occur when substrates enter the active
site, products of the reaction are released or when the transition
state analogues bind. Thus, these two structures differ too
6213
dx.doi.org/10.1021/jm300662d | J. Med. Chem. 2012, 55, 6209−6223

Journal of Medicinal Chemistry
Article
dramatically from the conformation expected when the aza-
ANPs bind.
site residues. The docked structure of 8m (Figure 4) suggests
that this second phosphonate group does not, however, make
The structures which did provide good solutions were
human HGPRT in complex with 2-(phosphonoethoxy)ethyl
(PEE) guanine or PEE hypoxanthine (Figure 2A) and 3-
hydroxy-2-(phosphonomethoxy)propyl (HPMP) guanine.4
These structures are expected to most closely resemble that
of the enzyme when in complex with the aza-ANPs. In general,
the docking poses with the highest scores locate the purine ring
in the purine binding pocket and the phosphonate group in the
5′-phosphate binding site.
Comparison of the Docked Structure of 8n (Guanine As
the Base) with the Crystal Structure of PEEG in Complex with
Human HGPRT. Superimposition of the docked structure of 8n
in complex with human HGPRT onto the known structure of
PEEG in complex with this enzyme showed that the
phosphonate groups were virtually superimposable. Differences
in the superimposition of the two ANPs in the active site lay in
the position of the linker. The replacement of a partially
negatively charged oxygen atom with a positively charged
nitrogen atom may explain the apparent difference between
these two structures (Figure 3). The disubstituted nitrogen of
Figure 4. Comparison of the docked structures of 8i with 8m. In these
two structures, the trisubstituted nitrogen is differently oriented,
resulting in the phosphonate group of 8m being located near the NZ
atom of K68 while the hydroxyl group of 8i is located in the 5′-
phosphate binding pocket, forming a hydrogen bond with the
phosphorus oxygen atoms.
the same interactions with the amino acid residues as does
pyrophosphate when it is bound in the active site in complex
with immucillinGP and magnesium ions. The reason for this
may be that, when PPi binds, the structure of the enzyme is very
different from that in its absence. The phosphorus oxygen
atoms of the second phosphonate group in 8m are further away
from the NE1 of R199 (11 Å compared with 2.8 and 2.7 Å
when PPi is bound) and 6.6 Å from the main chain nitrogen
atom of S103 compared with 2.7 Å. Probably the biggest
structural change between the immucillinGP complex and the
ANP complexes with human HGPRT is the position of the
large flexible loop which, in the former, is firmly closed over the
active site while, in the latter, it is only partially closed. The
other difference between these two structures is the location of
the K68 side chain. When PPi binds, this side chain has to
rotate by 180° to allow room for this compound. In the crystal
structures where ANPs are bound (PDB codes 3GGJ, 3GEP),
this rotation does not occur as there are no groups attached to
the ANPs to occupy this site. Because 8m is unable to dock into
the transition state analogue structure, it is deduced that this
movement does not occur and that the second phosphonate
group does not occupy this site. Mg2+ has not been modeled
into the 8m structure. A hypothesis to explain the lower Ki for
8m, compared with 8n, is that the oxygen atoms of the
phosphonate group interact with the enzyme via magnesium
ion(s).
Comparison of the Docked Structure of 8m with 8i
(Guanine As the Base). In the pose which gave the best
docking score for all of these compounds, the hydroxyl group is
located close to the phosphonate group (1.8 Å: Figure 4). This
compound is a good inhibitor of human HGPRT (Ki = 0.07
μM) and also of the parasite enzymes (0.2 μM for Pf HGXPRT
and 1.4 μM for PvHGPRT). The docking studies suggest that,
for 8i, the hydroxyl group of the side chain can make internal
hydrogen bonds with the phosphonate group. As a result of
this, the PPi binding site is largely unoccupied. The docking
studies show that when 8i is bound there still remains
unoccupied space in the 5′-phosphate binding pocket. Thus,
there is a possibility that this region could be exploited to
decrease the Ki value.
Figure 3. Comparison of the crystal structure4 of PEEG in complex
with human HGPRT with that of the docked structure of 8n. This
figure shows the change in the location of the disubstituted nitrogen
and the movement of the purine ring.
8n is located 1.4 Å from the corresponding oxygen atom in the
PEEG structure (Figure 3). This moves the nitrogen atom in
the linker closer to the OD2 atom of D107 (2.91 Å compared
with 4.69 Å between the oxygen atom in the linker in the PEEG
structure and this atom) and away from the back surface of the
active site. D107 is located in the large flexible loop (residues
100−120), which closes over the active site during catalysis. In
the PEEG structure, this mobile loop might be less hindered
and better able to open and close over the active site explaining,
in part, the lower Ki values of PEEG compared with 8n. The
changes in the linker structure also results in a subtle change in
the predicted orientation of the purine ring with the 6-oxo
group being located further away (2.8 Å compared to 2.4 Å)
from the NZ atom of K165. This difference may also contribute
to the observed decrease in affinity of the aza-ANPs. For the
parasite enzymes, the change in the structure of the linker has a
much more drastic effect on the affinity with these aza-ANPs
binding extremely weakly. Thus, for Pf HGXPRT and
PvHGPRT, the orientation of the purine ring in the active
site may be more compromised.
The Docked Structure of 8m (Guanine As the Base). The
addition of a second phosphonate group to the nitrogen atom
in the linker (as in 8m) resulted in a decrease in Ki for human
HGPRT by 15-fold (8m cf. 8n). This could be attributed to the
extra interactions between this group and the enzyme’s active
Comparison of the Docked Structures of 8f and 7f (Same
Linker, Different Base). In these two docked structures, no
6214
dx.doi.org/10.1021/jm300662d | J. Med. Chem. 2012, 55, 6209−6223

Journal of Medicinal Chemistry
Article
difference was observed in the location of the purine base and
the phosphonate group (Figure 5). There is, however, a large
Figure 6. The docking of 8d containing a −COOMe attachment to
the active site of HGPRT. The location of PPi when it binds in the
enzyme has been superimposed from the structure of human HGPRT
in complex with immucillinHP, PPi, and Mg2+.
Figure 5. Comparison of the docked structures of 8f with 7f. In these
two structures, the purine ring and the phosphonate group are located
in the same place and the difference in the positioning of the cyano
group is shown.
positive charge in the linker due to the substitution of −NH−
for −O−. However, the trisubstituted nitrogen compounds
exhibit low Ki values and, in some instances, excellent
selectivity. The low Ki values of some of these compounds
for PfHGXPRT and PvHGPRT and the high selectivity in favor
of these enzymes compared with human HGPRT suggest that
they are lead compounds for development as antimalarial drugs.
Docking of these compounds suggests how these compounds
bind in the active site of human HGPRT and provides
preliminary explanations for their Ki values. Further chemical
modifications can now be made to improve both potency and
selectivity. Work is ongoing to obtain crystal structures of the
three enzymes in complex with the aza-ANPs to verify the
docking studies and to determine more precisely how these
molecules bind in the active site.
difference in the Ki values for the human enzyme depending on
the attached base [cf. 0.07 μM with guanine (8f) and 2.4 μM
with hypoxanthine (7f)]. In contrast, the nature of the base did
not have any effect on the Ki values for either Pf HGXPRT or
PvHGPRT. Examination of the two docked structures in the
human enzyme suggests that the 34-fold difference in the Ki
values may be due to the cyano group being located in different
regions in the active site (Figure 5). Comparison of the docked
structures of 7f and 8f show that the cyano substituents in these
two poses are 4.6 Å apart. This structural change is due to the
slightly different conformations adopted by the linker. The
docked structure of 8f into the human HGPRT−3-hydroxy-2-
(phosphonomethoxy)propyl guanine complex shows that the
nitrogen in the cyano group is further away from the NZ atom
of K68 (4.6 Å) but is only 2.8 Å, if the NZ adopts the same
structure as it does in human HGPRT−PEEG complex. This
situation is reversed for 7f, i.e., the cyano nitrogen is closer to
the NZ atom in the human HGPRT−3-hydroxy-2-
(phosphonomethoxy)propyl guanine complex (3.63 Å) but
5.5 Å distant in the human HGPRT−PEEG structure. As the Ki
values do not differ for the parasite enzymes irrespective of the
base, one deduction that could be made is that, in these two
ANPs, the linker adopts the structure found when attached to
guanine. Thus, it would be expected that the cyano group
would be located closer to the corresponding lysine residue
(K68 in human).
Docked Structure of 8d. This compound is a good inhibitor
of human HGPRT (0.1 μM) and is a reasonable inhibitor of
PfHGXPRT (1.3 μM) but binds very weakly, if at all, (Ki ≥200
μM) to PvHGPRT. Docking suggests that its side chain
functional group (−COOMe) does not bind in the same
location as the phosphonate group in 8m (cf. Figures 4 and 6).
This new orientation could be attributed to the longer side
chain in 8d. The functional group is predicted to bind in a
hydrophobic pocket which is not located in the same area as
PPi. Therefore, attachment of other functional groups able to
occupy the PPi site may lead to increased potency.
EXPERIMENTAL SECTION
■
Synthesis and Analytical Chemistry. Unless otherwise stated,
solvents were evaporated at 40 °C/2 kPa and the compounds were
dried over P2O5 at 2 kPa. NMR spectra were recorded on Bruker
Avance 500 (1H at 500 MHz, 13C at 125.8 MHz) and Bruker Avance
400 (1H at 400 MHz, 13C at 100.6 MHz) spectrometers with TMS as
internal standard or referenced to the residual solvent signal. Mass
spectra were measured on a ZAB-EQ (VG Analytical) spectrometer.
The purity of the tested compounds was determined by the
combustion analysis (C, H, N) and was higher than 95%. The
chemicals were obtained from commercial sources (Sigma-Aldrich) or
prepared according to the published procedures. Dimethylformamide
and acetonitrile were distilled from P2O5 and stored over molecular
sieves (4 Å). THF was distilled from sodium/benzophenone under
argon. Preparative HPLC purifications were performed on columns
packed with 7 μm C18 reversed phase resin (Waters Delta 600
chromatograph column), 17 mm × 250 mm, in ca. 200 mg batches of
mixtures using gradient MeOH/H2O as eluent. Chromatography on
Dowex 1 × 2 (acetate form) was as follows: after application of the
aqueous solution of the crude product onto the column, it was washed
with water until the UV absorption was low. The column was then
washed with a gradient of acetic (0−1 M) followed by a gradient of
formic acid (0.125−2 M).
Synthesis of Diethyl 2-(2-Hydroxyethylamino)-
ethylphosphonate (1) and Tetraethyl 2, 2′-(2-
Hydroxyethylazanediyl)bis(ethane-2,1-diyl)diphosphonate
(2a). The mixture of 2-aminoethanol (10.7 g, 175 mmol), H2O (250
mL), and diethyl vinylphosphonate (16.4 g, 100 mmol) was stirred for
2 days at roomtemperature. Solvent was evaporated and the residue
codistilled with ethanol. The resulting compound was purified by
chromatography on silica gel (CHCl3−MeOH). Monophosphonate 1
CONCLUSION
■
An efficient method has been developed for the synthesis of a
new type of ANPs containing a di- or trisubstituted nitrogen in
the acyclic moiety. In the disubstituted compounds, the affinity
for all the enzymes is decreased, possibly because of the new
6215
dx.doi.org/10.1021/jm300662d | J. Med. Chem. 2012, 55, 6209−6223

Journal of Medicinal Chemistry
Article
was obtained as a major product, bisphosphonate 2a as a minor
product.
J(P,C) = 135.0 (C−P); 16.11 d, 2 C, J(P,C) = 5.8 (Et); 16.64 (C−
CN). MS (ESI): m/z = 293 [M + H]+.
1: oil, 17.22 g, 77%. 1H NMR (DMSO-d6): 3.97 m, 4 H (Et); 3.42 t,
2 H, J = 5.7 (CH2O); 2.70 m, 2 H (CH2N); 2.55 t, 2 H, J = 5.7
(CH2N); 1.89 m, 2 H (CH2P); 1.22 t, 6 H, J = 7.1 (Et). 13C NMR
(DMSO-d6): 60.48 d, 2 C, J(P,C) = 6.2 (Et); 59.95 (C−O); 50.85
(C−N); 42.54 d, J(P,C) = 2.5 (C-NP); 25.48 d, J(P,C) = 136.4 (C−
P); 15.91 d, 2 C, J(P,C) = 5.8 (Et). MS (ESI): m/z = 226 [M + H]+.
2: oil, 3.05 g, 8%. 1H NMR (DMSO-d6): 4.42 t, 1 H, J = 5.6 (OH);
3.97 m, 8 H (Et); 3.42 q, 2 H, J = 5.9 (CH2O); 2.67 m, 4 H (CH2N);
2.47 t, 2 H, J = 6.0 (CH2N); 1.87 m, 4 H (CH2P); 1.22 t, 12 H, J = 7.1
(Et). 13C NMR (DMSO-d6): 60.72 d, 4 C, J(P,C) = 6.3 (Et); 58.90
(C−O); 54.53 (C−N); 46. 27, 2 C (C−N); 22.34 d, 2 C, J(P,C) =
135.2 (C−P); 16.11 d, 4 C, J(P,C) = 5.8 (Et). MS (ESI): m/z = 390
[M + H]+.
Diethyl 2-(2-Cyanoethylamino)ethylphosphonate (2f). Method B,
starting from acrylonitrile, yield 78%. 1H NMR (DMSO-d6): 4.44 t, J =
5.4 (OH); 3.98 m, 4 H (Et); 3.44 q, 2 H, J = 6.0 (CH2O); 2.75 t, 2 H,
J = 6.8(CH2N); 2.72 m, 2 H (CH2N); 2.57 t, 2 H, J = 6.8 (CH2CN);
2.53 t, 2 H, J = 6.1 (CH2N); 1.91 m, 2 H (CH2P); 1.22 t, 6 H, J = 7.1
(Et). 13C NMR (DMSO-d6): 119.94 (CN); 60.76 d, 2 C, J(P,C) = 6.2
(Et); 59.14 (C−O); 54.66 (C−N); 48.83 (C−N); 46.77 (C−N);
22.60 d, J(P,C) = 133.9 (C−P); 16.10 d, 2 C, J(P,C) = 5.3 (Et); 15.51
(C−CN). MS (ESI): m/z = 279 [M + H]+.
Diethyl 2-(2-Cyanopropylamino)ethylphosphonate (2g). Method
A, KI (0.4 g, 2.6 mmol) added, additional heating 70 h, 80 °C, starting
from 4-chlorobutyronitrile, yield 69%. 1H NMR (DMSO-d6): 4.40 t, J
= 5.4 (OH); 3.98 m, 4 H (Et); 3.42 dd, 2 H, J = 11.3 and 5.7 (CH2O);
2.66 m, 2 H (CH2N); 2.50 m, 6 H, J = 6.8 (CH2N and CH2CN); 1.89
m, 2 H (CH2P); 1.64 m, 2 H, (CH2); 1.23 t, 6 H, J = 7.0 (Et). 13C
NMR (DMSO-d6): 120.62 (CN); 60.71 d, 2 C, J(P,C) = 6.3 (Et);
58.92 (C−O); 55.46 (C−N); 47.47 (C−N); 41.65 (C−N); 22.26 d,
J(P,C) = 136.6 (C−P); 16.11 d, 2 C, J(P,C) = 5.8 (Et). MS (ESI): m/
z = 287 [M + Na]+.
Diethyl 2-((2-Hydroxyethyl)(2-(trityloxy)ethyl)amino)-
ethylphosphonate (2h). The mixture of diethanolamine (2.1 g, 20
mmol), H2O (50 mL), and diethyl vinylphosphonate (3.3 g, 20 mmol)
was stirred at room temperature overnight. The solvent was
evaporated and the residue codistilled with ethanol. The residue was
purified by chromatography on silica gel (CHCl3−MeOH) to obtain
diethyl 2-(bis(2-hydroxyethyl)amino)ethylphosphonate, yield 5.03 g
(93%). 1H NMR (DMSO-d6): 4.35 t, 2 H, J = 5.5 (OH); 3.98 m, 4 H
(Et); 3.41 q, 4 H, J = 6.0 (CH2O); 2.71 m, 2 H (CH2N); 2.49 m, 4 H
(CH2N); 1.89 m, 2 H (CH2P); 1.22 t, 6 H, J = 7.1 (Et). 13C NMR
(DMSO-d6): 60.71 d, 2 C, J(P,C) = 6.3 (Et); 58.97, 2 C (C−O);
55.68, 2 C (C−N); 47.43 d, J(P,C) = 1.5 (C−N); 22.42 d, J(P,C) =
134.9 (C−P); 16.12 d, 2 C, J(P,C) = 5.8 (Et). MS (ESI): m/z = 270
[M + H]+. This intermediate (3.4 g, 12.6 mmol) was dissolved in
CH2Cl2 (50 mL), a catalytical amount of dimethylaminopyridine was
added followed by Et3N (1.2 mL), and then tritylchloride (3.5 g, 14.5
mmol) dissolved in CH2Cl2 (30 mL) was added dropwise. The
reaction mixture was stirred overnight at room temperature and
solvent evaporated. The residue was purified by column chromatog-
raphy on silica gel (CHCl3−MeOH) to afford the product as yellowish
oil (6.01 g, 93%). 1H NMR (DMSO-d6): 7.38 m, 6 H, 7.33 m, 6 H and
7.25 m, 3 H (Ar); 4.35 t, J = 5.5 (OH); 3.92 m, 4 H (Et); 3.38 m, 2 H
(CH2O); 3.01 t, 2 H, J = 6.0 (CH2O); 2.70 m, 2 H (CH2N); 2.66 t, 2
H, J = 6.0 (CH2N); 2.47 t, 2 H, J = 6.3 (CH2N); 1.87 m, 2 H (CH2P);
1.17 t, 6 H, J = 7.1 (Et). 13C NMR (DMSO-d6): 143.83, 3 C (Ar);
128.07, 6 C (Ar); 127.68, 6 C (Ar); 126.78, 3 C (Ar); 85.98 (Tr);
62.11 (C−O-Tr); 60.65 d, 2 C, J(P,C) = 6.2 (Et); 59.1 (C−O); 55.88,
2 C (C−N); 53.06 (C−N); 47.73 (C−N); 22.62 d, J(P,C) = 135.2
(C−P); 16.09 d, 2 C, J(P,C) = 5.4 (Et). MS (ESI): m/z = 512 [M +
H]+.
Synthesis of 2b−2g from Diethyl 2-(2-Hydroxyethylamino)-
ethylphosphonate (1). Method A. Alkylation: The mixture of 1 (6
g, 26.7 mmol) and K2CO3 (3.7 g, 26.7 mmol) in dry acetonitrile (80
mL) was cooled to −10 °C and a corresponding halogenoderivative
(40 mmol) was added. The reaction mixture was stirred at −10 °C for
1 h and then at room temperature for 2 days. The solvent was then
removed by evaporatation. Water and CHCl3 were added, and the
organic layer was separated, washed with brine, and dried over
anhydrous MgSO4. After filtration, solvent was evaporated and the
residue was purified by column chromatography on silica gel (CHCl3−
MeOH), the product was obtained as yellow oil.
Method B. Michael addition: A corresponding vinylderivative (45
mmol) was added to the solution of 1 (6.75 g, 30 mmol) in water (70
mL). The reaction mixture was stirred overnight at room temperature,
evaporated, and the residue codistilled with toluene. The resulting
compound was purified by column chromatography on silica gel
(CHCl3−MeOH), and the product was obtained as yellow oil.
Ethyl 2-((2-(Diethoxyphosphoryl)ethyl)(2-hydroxyethyl)amino)-
acetate (2b). Method A, starting from ethyl bromoacetate, yield
64%. 1H NMR (DMSO-d6): 4.43 t, J = 5.5 (OH); 4.07 q, 2 H, J = 7.1
(Et); 3.97 m, 4 H (Et); 3.43 m, 2 H (CH2O); 3.41 s, 2 H (CH2CO);
2.83 m, 2 H (CH2N); 2.63 t, 2 H, J = 6.0 (CH2N); 1.90 m, 2 H
(CH2P); 1.21 t, 6 H, J = 7.1 (Et); 1.18 t, 3 H, J = 7.1 (Et). 13C NMR
(DMSO-d6): 60.73 d, 2 C, J(P,C) = 6.3 (Et); 59.59 (Et); 59.22 (C−
O); 55.36, 54.19, and 47.34 (C−N); 23.53 d, J(P,C) = 134.1 (C−P);
16.12 d, 2 C, J(P,C) = 5.8 (Et); 14.01 (Et). MS (ESI): m/z = 312 [M
+ H]+.
Methyl 3-((2-(Diethoxyphosphoryl)ethyl)(2-hydroxyethyl)amino)-
propanoate (2c). Method B, starting from methyl acrylate, yield 50%.
1H NMR (DMSO-d6): 4.35 t, J = 5.4 (OH); 3.97 m, 4 H (Et); 3.58 s,
3 H (Me); 3.40 q, 2 H, J = 5.8 (CH2O); 2.71 t, 2 H, J = 7.0 (CH2CO);
2.66 m, 2 H (CH2N); 2.46 t, 2 H, J = 6.2 (CH2N); 2.40 t, 2 H, J = 7.0
(CH2N); 1.86 m, 2 H (CH2P); 1.22 t, 6 H, J = 7.1 (Et). 13C NMR
(DMSO-d6): 172.36 (CO); 60.70 d, 2 C, J(P,C) = 6.2 (Et); 59.01 (C−
O); 54.98, 48.64, and 46.75 (C−N); 51.06 (Me); 31.86 (C−CO);
22.34 d, J(P,C) = 134.9 (C−P); 16.12 d, 2 C, J(P,C) = 5.8 (Et). MS
(ESI): m/z = 312[M + H]+.
Synthesis of N9-Substituted 6-Chloropurines 3a−3h via
Mitsunobu Reaction, General Procedure. To a solution of
triphenylphosphine (6.3 g, 24 mmol) in dry THF (100 mL) cooled
to −30 °C under argon atmosphere, diisopropylazadicarboxylate
(DIAD, 4.4 mL, 23 mmol) was added slowly. The mixture was stirred
for 30 min, and this preformed complex was added to chloropurine
(3.5 g, 22.6 mmol), dry THF (70 mL), and corresponding diethyl
phosphonate 2a−2h (11.2 mmol) at −30 °C under argon. The
resulting mixture was slowly warmed to room temperature and stirred
for 48 h. Solvent was evaporated, and the crude mixture was purified
by chromatography on silica gel (MeOH−CHCl3). The pure product
was obtained as yellowish foam.
Methyl 3-((2-(Diethoxyphosphoryl)ethyl)(2-hydroxyethyl)amino)-
butanoate (2d). Method A, additional heating 70 h, 80 °C, starting
1
from methyl 4-chlorobutyrate, yield 77%. H NMR (DMSO-d6): 4.36
t, J = 5.4 (OH); 3.97 m, 4 H (Et); 3.57 s, 3 H (Me); 3.40 q, 2 H, J =
6.0 (CH2O); 2.65 m, 2 H (CH2N); 2.43 t, 2 H, J = 6.2 (CH2N); 2.39
t, 2 H, J = 7.0 (CH2N); 2.32 t, 2 H, J = 7.3 (CH2N); 1.85 m, 2 H
(CH2P); 1.60 p, 2 H, J = 7.2 (CH2); 1.22 t, 6 H, J = 7.1 (Et). 13C
NMR (DMSO-d6): 60.76 d, 2 C, J(P,C) = 6.2 (Et); 59.03 (C−O);
55.16, 52.21, and 46.83 (C−N); 51.06 (Me); 30.87 (C−CO); 22.04 d,
J(P,C) = 134.8 (C−P); 22.07; 16.19 d, 2 C, J(P,C) = 2.8 (Et). MS
(ESI): m/z = 326 [M + H]+.
Diethyl 2-(2-Cyanomethylamino)ethylphosphonate (2e). Method
A, starting from chloroacetonitrile, yield 65%. H NMR (DMSO-d6):
Tetraethyl 9-[(N,N-(Bis-2-phosphonoethyl))-2-aminoethyl]-6-
chloropurine (3a). Starting from diethyl phosphonate 2a, yield 80%.
1H NMR (DMSO-d6): 8.78 s, 1 H and 8.77 s, 1 H (H-2 and H-8);
4.33 t, 2 H, J(1′,2′) = 5.8 (H-1′); 3.92 m, 8 H (Et); 2.85 t, 2 H, J(2′,1′)
= 5.8 (H-2′); 2.64 dd, 4 H, J = 9.4, Jg = 15.6 (H-3′ and H-5′); 1.70 m, 4
H (H-4′ and H-6′); 1.19 t, 12 H, J = 7.0 (Et). 13C NMR (DMSO-d6):
151.86 (C-4); 151.14 (C-2); 148.70 (C-6); 147.93 (C-8); 130.58 (C-
1
4.61 t, J = 5.4 (OH); 3.98 m, 4 H (Et); 3.82 s, 3 H (CH2CN); 3.48 q,
2 H, J = 5.5 (CH2O); 2.66 m, 2 H (CH2N); 2.53 t, 2 H, J = 5.8
(CH2N); 1.97, 2 H (CH2P); 1.23 t, 6 H, J = 7.1 (Et). 13C NMR
(DMSO-d6): 116.13 (CN); 60.86 d, 2 C, J(P,C) = 6.2 (Et); 58.92 (C−
O); 55.03 (C−N); 51.51 (C−N); 46.89 (C−N); 22.87 (C); 22.20 d,
6216
dx.doi.org/10.1021/jm300662d | J. Med. Chem. 2012, 55, 6209−6223

Journal of Medicinal Chemistry
Article
5); 60.73 d, 4 C, J(P,C) = 6.3 (Et); 51.06 (C-2′); 46.04, 2 C (C-3′ and
C-5′); 42.00 (C-1′); 22.26 d, 2 C, J(P,C) = 135.0 (C-4′and C-6′); 16.05
d, 4 C, J(P,C) = 5.8 (Et). MS (ESI): m/z = 526 [M + H]+.
Diethyl 9-[(N-(2-(Trityloxy)ethyl)-N-(2-phosphonoethyl))-2-ami-
noethyl]-6-chloropurine (3h). Starting from diethyl phosphonate
2h, yield 76%. 1H NMR (DMSO-d6): 8.65 s, 1 H and 8.62 s, 1 H (H-2
and H-8); 7.25 m, 15 H (Tr); 4.29 t, 2 H, J(1′,2′) = 5.7(H-1′); 3.87 m,
4 H (Et); 2.87 t, 2 H, J(2′,1′) = 5.7 (H-2′); 2.77 t, 2 H, J(6′,5′) = 5.8
(H-6′); 2.67 m, 4 H (H-3′ and H-5′); 1.73 m, 2 H (H-4′); 1.13 t, 6 H, J
= 7.0 (Et). 13C NMR (DMSO-d6): 151.73 (C-4); 151.01 (C-2);
148.65 (C-6); 147.70 (C-8); 143.59, 3 C (Tr); 130.50 (C-5); 127.92,
6 C (Tr); 127.61, 6 C (Tr); 126.75, 3 C (Tr); 86.02 (Tr); 61.96 (C-
6′); 60.66 d, 2 C, J(P,C) = 6.2 (Et); 52.42, 52.18, and 47.23 (C-2′, C-3′
and C-5′); 41.98 (C-1′); 22.28 d, J(P,C) = 134.0 (C-4′); 16.04 d, 2 C,
J(P,C) = 5.7 (Et). MS (ESI): m/z = 648 [M + H]+.
Diethyl 9-[(N-(2-Ethoxy-2-oxoethyl)-N-(2-phosphonoethyl))-2-
aminoethyl]-6-chloropurine (3b). Starting from diethyl phosphonate
2b, yield 79%. 1H NMR (DMSO-d6): 8.77 s, 1 H and 8.72 s, 1 H (H-2
and H-8); 4.34 t, 2 H, J(1′,2′) = 5.8 (H-1′); 3.90 m, 4 H (Et); 3.41 s, 2
H (H-5′); 3.01 t, 2 H, J(2′,1′) = 5.8 (H-2′); 2.22 m, 2 H (H-3′); 1.67 m,
2 H (H-4′); 1.17 t, 6 H, J = 7.0 (Et). 13C NMR (DMSO-d6): 170.64
(CO); 151.99 (C-4); 151.19 (C-2); 149.11 (C-6); 147.99 (C-8);
130.60 (C-5); 60.79 d, 2 C, J(P,C) = 6.1 (Et); 59.78 (Et); 53.45,
51.83, and 47.04 (C-2′, C-5′ and C-3′); 41.94 (C-1′); 22.97 d, J(P,C) =
134.2 (C-4′); 16.15 d, 2 C, J(P,C) = 5.7 (Et); 14.11 (Et). MS (ESI):
m/z = 448 [M + H]+.
Synthesis of N9-Substituted 2-Amino-6-Chloropurines 4a−4i
via Mitsunobu Reaction, General Procedure. Starting from 2-
amino-6-chloropurine and diethyl phosphonates 2a−2h the procedure
was identical as described above for 6-chloropurine. Then after the
stirring of reaction mixture overnight, water (30 mL) was added and
the mixture was heated at 80 °C for 30 h. Solvent was evaporated, the
residue was codistilled with toluene or ethanol, and the crude mixture
purified by chromatography on silica gel (MeOH−CHCl3). The pure
product was isolated as yellow foam.
Diethyl 9-[(N-(3-Methoxy-3-oxopropyl)-N-(2-phosphonoethyl))-
2-aminoethyl]-6-chloropurine (3c). Starting from diethyl phospho-
nate 2c, yield 56%. 1H NMR (DMSO-d6): 8.77 s, 1 H and 8.64 s, 1 H
(H-2 and H-8); 4.32 t, 2 H, J(1′,2′) = 5.8 (H-1′); 3.94 m, 4 H (Et);
3.45 s, 3 H (Me); 2.86 t, 2 H, J(2′,1′) = 5.8 (H-2′); 2.66 m, 4 H (H-3′
and H-5′); 2.19 t, 2 H, J(6′,5′) = 6.7 (H-6′); 1.74 m, 2 H (H-4′); 1.20 t,
6 H, J = 7.0 (Et). 13C NMR (DMSO-d6): 171.98 (CO); 151.85 (C-4);
151.12 (C-2); 148.66 (C-6); 147.78 (C-8); 130.62 (C-5); 60.74 d, 2
C, J(P,C) = 6.3 (Et); 51.33, 48.06, and 46.15 (C-2′, C-3′ and C-5′);
50.95 (Me); 41.90 (C-1′); 31.67 (C-6′); 21.93 d, J(P,C) = 134.7 (C-
4′); 16.09 d, 2 C, J(P,C) = 5.7 (Et). MS (ESI): m/z = 448 [M + H]+.
Diethyl 9-[(N-(4-Methoxy-4-oxobutyl)-N-(2-phosphonoethyl))-2-
aminoethyl]-6-chloropurine (3d). Starting from diethyl phosphonate
2d, yield 83%. 1H NMR (DMSO-d6): 8.76 s, 1 H and 8.71 s, 1 H (H-2
and H-8); 4.32 t, 2 H, J(1′,2′) = 5.7 (H-1′); 3.95 m, 4 H (Et); 3.50 s, 3
H (Me); 2.83 t, 2 H, J(2′,1′) = 5.8 (H-2′); 2.65 dd, 2 H, J = 15.8 and
8.5 (H-3′); 2.35 t, 2 H, J(5′,6′) = 6.8 (H-5′); 1.96 t, 2 H, J(7′,6′) = 7.4
(H-7′); 1.76 m, 2 H (H-4′); 1.33 m, 2 H (H-6′); 1.21 t, 6 H, J = 7.1
(Et). 13C NMR (DMSO-d6): 172.94 (CO); 151.83 (C-4); 151.13 (C-
2); 148.75 (C-6); 147.79 (C-8); 130.61 (C-5); 60.73 d, 2 C, J(P,C) =
6.3 (Et); 51.42, 51.40, and 46.22 (C-2′, C-3′ and C-5′); 50.93 (Me);
42.01 (C-1′); 30.29 (C-6′); 21.80 d, J(P,C) = 134.6 (C-4′); 21.77 (C-
7′); 16.08 d, 2 C, J(P,C) = 5.8 (Et). MS (ESI): m/z = 462 [M + H]+.
Diethyl 9-[(N-(2-Cyanomethyl)-N-(2-phosphonoethyl))-2-amino-
ethyl]-6-chloropurine (3e). Starting from diethyl phosphonate 2e,
yield 87%. 1H NMR (DMSO-d6): 8.78 s, 1 H and 8.70 s, 1 H (H-2 and
H-8); 4.41 t, 2 H, J(1′,2′) = 5.8 (H-1′); 3.92 m, 4 H (Et); 3.85 s, 2
H(H-5′); 2.95 t, 2 H, J(2′,1′) = 5.8 (H-2′); 2.67 dd, 2 H, J = 9.8 and
16.00 (H-3′); 1.78 m, 2 H (H-4′); 1.18 t, 6 H, J = 7.1 (Et). 13C NMR
(DMSO-d6): 151.93 (C-4); 151.20 (C-2); 148.72 (C-6); 147.65 (C-
8); 130.57 (C-5); 115.95 (CN); 60.84 d, 2 C, J(P,C) = 6.2 (Et); 52.03
and 47.07 (C-2′ and C-3′); 41.27 and 40.99 (C-1′ and C-5′); 23.02 d,
J(P,C) = 136.4 (C-4′); 16.04 d, 2 C, J(P,C) = 5.8 (Et). MS (ESI): m/z
= 401 [M + H]+.
Diethyl 9-[(N-(2-Cyanoethyl)-N-(2-phosphonoethyl))-2-amino-
ethyl]-6-chloropurine (3f). Starting from diethyl phosphonate 2f,
yield 85%. 1H NMR (DMSO-d6): 8.77 s, 1 H and 8.71 s, 1 H (H-2 and
H-8); 4.35 t, 2 H, J(1′,2′) = 5.9 (H-1′); 3.93 m, 4 H (Et); 2.91 t, 2 H,
J(2′,1′) = 5.9 (H-2′); 2.74 t, 2 H, J(5′,6′) = 6.7 (H-5′); 2.67 dd, 2 H, J =
8.5 and 16.0 (H-3′); 2.47 t, 2 H, J(6′,5′) = 7.0 (H-6′); 1.73 m, 2 H (H-
4′); 1.19 t, 6 H, J = 7.1 (Et). 13C NMR (DMSO-d6): 151.93 (C-4);
151.17 (C-2); 148.74 (C-6); 147.65 (C-8); 130.65 (C-5); 119.52
(CN); 60.79 d, 2 C, J(P,C) = 6.2 (Et); 51.26, 48.02, and 46.14 (C-2′,
C-5′ and C-3′); 41.85 (C-1′); 22.22 d, J(P,C) = 134.6 (C-4′); 16.09 d,
2 C, J(P,C) = 5.7 (Et); 15.42 (C-6′). MS (ESI): m/z = 415 [M + H]+.
Diethyl 9-[(N-(2-Cyanopropyl)-N-(2-phosphonoethyl))-2-amino-
ethyl]-6-chloropurine (3g). Starting from diethyl phosphonate 2g,
yield 81%. 1H NMR (DMSO-d6): 8.89 s, 1 H and 8.85 s, 1 H (H-2 and
H-8); 4.45 t, 2 H, J(1′,2′) = 6.0 (H-1′); 4.06 m, 4 H (Et); 2.97 t, 2 H,
J(2′,1′) = 6.0 (H-2′); 2.77 dd, 2 H, J = 9.0 and 15.7 (H-3′); 2.55 t, 2 H,
J(5′,6′) = 6.7 (H-5′); 2.33 t, 2 H, J(7′,6′) = 7.2 (H-7′); 1.89 m, 2 H (H-
4′); 1.56 m, 2 H (H-6′); 1.32 t, 6 H, J = 7.1 (Et). 13C NMR (DMSO-
d6): 152.36 (C-4); 151.20 (C-2); 148.82 (C-6); 147.77 (C-8); 129.17
(C-5); 120.23 (CN); 60.75 d, 2 C, J(P,C) = 6.3 (Et); 51.51, 51.05, and
46.20 (C-2′, C-5′ and C-3′); 41.97 (C-1′); 22.57 (C-6′); 16.09 d, 2 C,
J(P,C) = 5.7 (Et); 13.43 (C-7′). MS (ESI): m/z = 429 [M + H]+.
Tetraethyl 9-[(N,N-(Bis-2-phosphonoethyl))-2-aminoethyl]-2-
amino-6-chloropurine (4a). Starting from diethyl phosphonate 2a,
1
yield 79%. H NMR (DMSO-d6): 8.17 s, 1 H (H-8); 6.89 s, 2 H
(NH2); 4.07 t, 2 H, J(1′,2′) = 5.7 (H-1′); 3.93 m, 8 H (Et); 2.76 t, 2 H,
J(2′,1′) = 5.7 (H-2′); 2.63 dd, 4 H, J = 15.6 and 9.2 (H-6′ and H-5′);
1.72 m, 4 H (H-4′ and H-6′); 1.19 t, 12 H, J = 7.1 (Et). 13C NMR
(DMSO-d6): 159.54 (C-2); 153.95 (C-4); 149.05 (C-6); 143.78 (C-
8); 123.15 (C-5); 60.77 d, 4 C, J(P,C) = 6.3 (Et); 51.03 (C-2′); 46.12,
2 C (C-3′ and C-5′); 41.43 (C-1′); 22.33 d, 2 C, J(P,C) = 134.8 (C-
4′and C-6′); 16.09 d, 4 C, J(P,C) = 5.8 (Et). MS (ESI): m/z = 541 [M
+ H]+.
Diethyl 9-[(N-(2-Ethoxy-2-oxoethyl)-N-(2-phosphonoethyl))-2-
aminoethyl]-2-amino-6-chloropurine (4b). Starting from diethyl
1
phosphonate 2b, yield 47%. H NMR (DMSO-d6): 8.14 s, 1 H (H-
8); 6.87 s, 2 H (NH2); 4.08 t, 2 H, J(1′,2′) = 5.4 (H-1′); 4.03 dd, 2 H, J
= 14.2 and 7.2 (Et); 3.91 m, 4 H (Et); 3.41 s, 1 H (H-5′); 2.94 t, 2 H,
J(2′,1′) = 5.4 (H-2′); 2.76 dd, 2 H, J = 15.92 and 8.4 (H-3′); 1.68 m, 2
H (H-4′); 1.18 t, 6 H, J = 7.0 (Et); 1.15 t, 3 H, J = 7.1 (Et). 13C NMR
(DMSO-d6): 170.70 (CO); 159.59 (C-2); 154.05 (C-4); 149.05 (C-
6); 143.78 (C-8); 123.17 (C-5); 60.84 d, 2 C, J(P,C) = 6.1 (Et); 59. 79
(Et); 53.52, 51.76, and 47.10 (C-2′, C-3′ and C-5′); 41.32 (C-1′); 23.34
d, J(P,C) = 133.9 (C-4′); 16.13 d, 2 C, J(P,C) = 5.7 (Et); 13.98 (Et).
MS (ESI): m/z = 463 [M + H]+
Diethyl 9-[(N-(3-Methoxy-3-oxopropyl)-N-(2-phosphonoethyl))-
2-aminoethyl]-2-amino-6-chloropurine (4c). Starting from diethyl
phosphonate 2c, yield 59%. 1H NMR (DMSO-d6): 8.05 s, 1 H (H-8);
6.87 s, 2 H (NH2); 4.05 t, 2 H, J(1′,2′) = 5.8 (H-1′); 3.94 m, 4 H (Et);
3.51 s, 3 H (Me); 2.75 t, 2 H, J(2′,1′) = 5.8 (H-2′); 2.67 t, 2 H, J(5′,6′)
= 6.8 (H-5′); 2.63 m, 2 H (H-3′); 2.26 t, 2 H, J(6′,5′) = 6.8 (H-6′); 1.73
m, 2 H (H-4′); 1.20 t, 6 H, J = 7.1 (Et). 13C NMR (DMSO-d6): 172.13
(CO); 159.52 (C-2); 153.92 (C-4); 149.00 (C-6); 143.61 (C-8);
123.18 (C-5); 60.77 d, 2 C, J(P,C) = 6.1 (Et); 51.03 (Me); 51.35,
48.08, and 46.24 (C-2′, C-3′ and C-5′); 41.26 (C-1′); 31.71 (C-6′);
22.12 d, J(P,C) = 135.2 (C-4′); 16.11 d, 2 C, J(P,C) = 5.6 (Et). MS
(ESI): m/z = 463 [M + H]+.
Diethyl 9-[(N-(4-Methoxy-4-oxobutyl)-N-(2-phosphonoethyl))-2-
aminoethyl]-2-amino-6-chloropurine (4d). Starting from diethyl
1
phosphonate 2d, yield 80%. H NMR (DMSO-d6): 8.11 s, 1 H (H-
8); 6.86 s, 2 H (NH2); 4.06 t, 2 H, J(1′,2′) = 5.7 (H-1′); 3.95 m, 4 H
(Et); 3.53 s, 3 H (Me); 2.74 t, 2 H, J(2′,1′) = 5.7 (H-2′); 2.74 dd, 2 H,
J = 15.8 and 8.4 (H-3′); 2.36 t, 2 H, J(5′,6′) = 6.8 (H-5′); 2.07 t, 2 H,
J(7′,6′) = 7.4 (H-7′); 1.75 m, 2 H (H-4′); 1.42 m, 2 H, (H-6′); 1.21 t, 6
H, J = 7.0 (Et). 13C NMR (DMSO-d6): 173.08 (CO); 159.51 (C-2);
153.92 (C-4); 149.05 (C-6); 143.60 (C-8); 123.19 (C-5); 60.76 d, 2
C, J(P,C) = 6.3 (Et); 50.97 (Me); 51.43, 51.40, and 46.28 (C-2′, C-3′
and C-5′); 41.32 (C-1′); 30.42 (C-6′); 21.94 d, J(P,C) = 134.5 (C-4′);
21.84 (C-7′); 16.10 d, 2 C, J(P,C) = 5.7 (Et). MS (ESI): m/z = 477
[M + H]+.
6217
dx.doi.org/10.1021/jm300662d | J. Med. Chem. 2012, 55, 6209−6223

Journal of Medicinal Chemistry
Article
Diethyl 9-[(N-(2-Cyanomethyl)-N-(2-phosphonoethyl))-2-amino-
ethyl]-2-amino-6-chloropurine (4e). Starting from diethyl phospho-
nate 2e, yield 73%. 1H NMR (DMSO-d6): 8.12 s, 1 H (H-8); 6.90 s, 2
H (NH2); 4.15 t, 2 H, J(1′,2′) = 5.7 (H-1′); 3.92 m, 4 H (Et); 3.86 s, 2
H (H-5′); 2.86 t, 2 H, J(2′,1′) = 5.7 (H-2′); 2.67 dd, 2 H, J = 15.9 and
9.6 (H-3′); 1.79 m, 2 H (H-4′); 1.19 t, 6 H, J = 7.1 (Et). 13C NMR
(DMSO-d6): 159.56 (C-2); 154.03 (C-4); 149.08 (C-6); 143.40 (C-
8); 124.07 (C-5); 115.91 (CN); 60.90 d, 2 C, J(P,C) = 6.2 (Et); 52.02,
47.03, 40.98, and 40.48 (C-2′, C-3′, C-5′ and C-1′); 23.13 d, J(P,C) =
136.1 (C-4′); 16.07 d, 2 C, J(P,C) = 5.7 (Et). MS (ESI): m/z = 416
[M + H]+.
Diethyl 9-[(N-(2-Cyanoethyl)-N-(2-phosphonoethyl))-2-amino-
ethyl]-2-amino-6-chloropurine (4f). Starting from diethyl phospho-
nate 2f, yield 78%. 1H NMR (DMSO-d6): 8.14 s, 1 H (H-8); 6.89 s, 2
H (NH2); 4.08 t, 2 H, J(1′,2′) = 6.0 (H-1′); 3.93 m, 4 H (Et); 2.83 t, 2
H, J(2′,1′) = 6.0 (H-2′); 2.76 t, 2 H, J(5′,6′) = 6.8 (H-5′); 2.66 dd, 2 H,
J = 16.0 and 8.3 (H-3′); 2.53 t, 2 H, J(6′,5′) = 6.6 (H-6′); 1.73 m, 2 H
(H-4′); 1.19 t, 6 H, J = 7.1 (Et). 13C NMR (DMSO-d6): 159.55 (C-2);
153.96 (C-4); 149.08 (C-6); 143.49 (C-8); 123.17 (C-5); 119.73
(CN); 60.84 d, 2 C, J(P,C) = 6.3 (Et); 51.23, 48.15, and 46.31 (C-2′,
C-3′ and C-5′); 41.30 (C-1′); 22.41 d, J(P,C) = 134.5 (C-4′); 16.09 d,
2 C, J(P,C) = 5.7 (Et); 15.45 (C-6′). MS (ESI): m/z = 430 [M + H]+.
Diethyl 9-[(N-(2-Cyanopropyl)-N-(2-phosphonoethyl))-2-amino-
ethyl]-2-amino-6-chloropurine (4g). Starting from diethyl phospho-
nate 2g, yield 66%. 1H NMR (DMSO-d6): 8.14 s, 1 H (H-8); 6.88 s, 2
H (NH2); 4.08 t, 2 H, J(1′,2′) = 5.8 (H-1′); 3.95 m, 4 H (Et); 2.76 t, 2
H, J(2′,1′) = 5.8 (H-2′); 2.64 dd, 2 H, J = 15.7 and 8.8 (H-3′); 2.44 t, 2
H, J(5′,6′) = 6.7 (H-5′); 2.27 t, 2 H, J(7′,6′) = 7.2 (H-7′); 1.77 m, 2 H
(H-4′); 1.49 m, 2 H (H-6′); 1.21 t, 6 H, J = 7.1 (Et). 13C NMR
(DMSO-d6): 159.53 (C-2); 153.94 (C-4); 149.14 (C-6); 143.55 (C-
8); 123.17 (C-5); 120.35 (CN); 60.78 d, 2 C, J(P,C) = 6.3 (Et); 51.48,
51.02, and 46.33 (C-2′, C-3′ and C-5′); 41.28 (C-1′); 22.63 (C-6′);
21.95 d, J(P,C) = 136.2 (C-4′); 16.11 d, 2 C, J(P,C) = 5.7 (Et); 13.49
(C-7′). MS (ESI): m/z = 444 [M + H]+.
s, 1 H (H-2 and H-8); 4.15 t, 2 H, J(1′,2′) = 6.0 (H-1′); 3.94 m, 8 H
(Et); 2.79 t, 2 H, J(2′,1′) = 6.0 (H-2′); 2.63 dd, 4 H, J = 15.6 and 9.16
(H-3′ and H-5′); 1.73 m, 4 H (H-4′ and H-6′); 1.20 t, 12 H, J = 7.1
(Et). 13C NMR (DMSO-d6): 156.43 (C-6); 148.20 (C-4); 145.31 (C-
2); 140.54 (C-8); 123.66 (C-5); 60.97 d, 4 C, J(P,C) = 6.1 (Et); 51.14
(C-2′); 46.28, 2 C (C-3′ and C-5′); 41.02 (C-1′); 21.75 d, 2 C, J(P,C)
= 135.9 (C-4′and C-6′); 16.08 d, 4 C, J(P,C) = 5.7 (Et). MS (ESI): m/
z = 508 [M + H]+.
Diethyl 9-[(N-(2-Ethoxy-2-oxoethyl)-N-(2-phosphonoethyl))-2-
aminoethyl]hypoxanthine (5b). Starting from diethyl phosphonate
3b, yield 75%. 1H NMR (DMSO-d6): 12.26 s, 1 H (NH); 8.09 s, 1 H
and 8.02 s, 1 H (H-2 and H-8); 4.17 t, 2 H, J(1′,2′) = 6.0 (H-1′); 4.04
q, 2 H, J(1′,2′) = 7.1 (Et); 3.91 m, 4 H (Et); 3.42 s, 2 H (H-5′); 2.96 t,
2 H, J(2′,1′) = 6.0 (H-2′); 2.76 dd, 2 H, J = 8.4 and 16.2 (H-3′); 1.69
m, 2 H (H-4′); 1.18 t, 6 H, J = 7.1 (Et);1.16 t, 3 H, J = 7.2 (Et). 13C
NMR (DMSO-d6): 170.72 (CO); 156.59 (C-6); 148.31 (C-4); 145.20
(C-2); 140.72 (C-8); 123.66 (C-5); 60.82 d, 2 C, J(P,C) = 6.2 (Et);
59.78 (Et); 53.51, 52.22, and 47.19 (C-2′, C-5′ and C-3′); 41.59 (C-1′);
23.40 d, J(P,C) = 134.3 (C-4′); 16.14 d, 2 C, J(P,C) = 5.8 (Et); 14.01
(Et). MS (ESI): m/z = 430 [M + H]+.
Diethyl 9-[(N-(3-Methoxy-3-oxopropyl)-N-(2-phosphonoethyl))-
2-aminoethyl]hypoxanthine (5c). Starting from diethyl phosphonate
3c, yield 66%. 1H NMR (DMSO-d6): 12.25 s, 1 H (NH); 8.02 s, 2 H
(H-2 and H-8); 4.14 t, 2 H, J(1′,2′) = 5.9 (H-1′); 3.94 m, 4 H (Et);
3.51 s, 3 H (Me); 2.78 t, 2 H, J(2′,1′) = 5.9 (H-2′); 2.68 t, 2 H, J(5′,6′)
= 6.7 (H-5′); 2.63 m, 2 H (H-3′); 2.27 t, 2 H, J(6′,5′) = 6.7 (H-6′); 1.74
m, 2 H (H-4′); 1.20 t, 6 H, J = 7.0 (Et). 13C NMR (DMSO-d6): 172.14
(CO); 156.50 (C-6); 148.17 (C-4); 145.09 (C-2); 140.53 (C-8);
123.69 (C-5); 60.75 d, 2 C, J(P,C) = 6.2 (Et); 51.02 (Me); 51.89,
48.15, and 46.31 (C-2′, C-5′ and C-3′); 41.51 (C-1′); 31.77 (C-6′);
22.11 d, J(P,C) = 134.7 (C-4′); 16.08 d, 2 C, J(P,C) = 5.7 (Et). MS
(ESI): m/z = 430 [M + H]+.
Diethyl 9-[(N-(4-Methoxy-4-oxobutyl)-N-(2-phosphonoethyl))-2-
aminoethyl]hypoxanthine (5d). Starting from diethyl phosphonate
3d, yield 42%. 1H NMR (DMSO-d6): 12.26 s, 1 H (NH); 8.07 s, 1 H
and 8.02 s, 1 H (H-2 and H-8); 4.15 t, 2 H, J(1′,2′) = 6.0 (H-1′); 3.95
m, 4 H (Et); 3.53 s, 3 H (Me); 2.77 t, 2 H, J(2′,1′) = 6.0 (H-2′); 2.63
dd, 2 H, J = 8.4 and 15.9 (H-3′); 2.36 t, 2 H, J(5′,6′) = 6.8 (H-5′); 2.06
t, 2 H, J(7′,6′) = 7.4 (H-7′); 1.76 m, 2 H (H-4′); 1.42 m, 2 H (H-6′);
1.21 t, 6 H, J = 7.0 (Et). 13C NMR (DMSO-d6): 173.11 (CO); 156.52
(C-6); 148.21 (C-4); 145.11 (C-2); 140.57 (C-8); 123.74 (C-5);
60.75 d, 2 C, J(P,C) = 6.3 (Et); 51.91, 51.50, and 46.32 (C-2′, C-5′ and
C-3′); 50.97 (Me); 41.60 (C-1′); 30.40 (C-7′); 21.92 d, J(P,C) = 134.7
(C-4′); 21.90 (C-6′); 16.11 d, 2 C, J(P,C) = 5.8 (Et). MS (ESI): m/z =
444 [M + H]+.
Diethyl 9-[(N-(2-Trityloxyoxyethyl)-N-(2-phosphonoethyl))-2-
aminoethyl]2-amino-6-chloropurine (4h). 4h was isolated as the
triphenylphosphoranylidene intermediate MS (ESI) m/z = 923 [M +
H]+, starting from diethyl phosphonate 2h, yield 73%. This compound
was detritylated to enable cleavage of the triphenylphosphoranylidene
moiety and hydrolysis of the 6-chlorogroupduring the next steps (see
synthesis of 4i and 6i).
Diethyl 9-[(N-(2-Hydoxyoxyethyl)-N-(2-phosphonoethyl))-2-
aminoethyl]2-amino-6-chloropurine (4i). Intermediate 4h (5 g, 5.4
mmol) in 75% aqueous trifluoroacetic acid was stirred overnight at
room temperature, solvents were evaporated and the residue was
codistilled with water and ethanol. The crude mixture was purified by
chromatography on silica gel (MeOH−CHCl3) to obtain 2.2 g (60%)
of detritylated compound 4i in the form of triphenylphosphoranyli-
Diethyl 9-[(N-(2-Cyanomethyl)-N-(2-phosphonoethyl))-2-
aminoethyl]hypoxanthine (5e). Starting from diethyl phosphonate
3e, yield 95%. 1H NMR (DMSO-d6): 12.30 s, 1 H (NH); 8.08 s, 1 H
and 8.04 s, 1 H (H-2 and H-8); 4.24 t, 2 H, J(1′,2′) = 5.8 (H-1′); 3.93
m, 4 H (Et); 3.86 s, 2 H (H-5′); 2.88 t, 2 H, J(2′,1′) = 5.8 (H-2′); 2.65
dd, 2 H, J = 9.4 and 15.9 (H-3′); 1.78 m, 2 H (H-4′); 1.19 t, 6 H, J =
7.0 (Et). 13C NMR (DMSO-d6): 156.51 (C-6); 148.30 (C-4); 145.27
(C-2); 140.45 (C-8); 123.66 (C-5); 115.99 (CN); 60.91 d, 2 C, J(P,C)
= 6.2 (Et); 52.38 and 46.34 (C-2′ and C-3′); 41.05 and 40.85 (C-5′ and
C-1′); 23.12 d, J(P,C) = 136.3 (C-4′); 16.08 d, 2 C, J(P,C) = 5.7 (Et).
MS (ESI): m/z = 383 [M + H]+.
1
dene derivative. H NMR (DMSO-d6): 8.37 s, 1 H (H-8); 7.95 m, 6
H, 7.86 m, 3 H and 7.22 m, 6 H (Ar); 4.41 t, 2 H, J(1′,2′) = 6.2 (H-1′);
4.06 m, 4 H (Et); 23.70 t, 2 H, J(6′,5′) = 5.0 (H-6′); 3.28 m, 4 H, (H-2′
and H-3′); 3.09 m, 2 H (H-5′); 2.16 m, 2 H (H-4′); 1.29 t, 6 H, J = 7.1
(Et). 13C NMR (DMSO-d6): 162.70 d, J(P,C) = 35.3 (C-2); 154.71 d,
J(P,C) = 2.8(C-4); 151.34 (C-6); 147.52 (C-8); 136.13 d, J(P,C) =
2.2, 134.95 d, J(P,C) = 11.2 and 131.00 d, J(P,C) = 13.6 (PPh3);
128.33 (C-5); 122.27 d, J(P,C) = 104.9 (PPh3); 63.84 d, 2 C, J(P,C) =
6.6 (Et); 58.26 (C-6′); 56.01 (C-5′); 53.82 (C-2′); 49.31 (C-3′); 41.73
(C-1′); 22.43 d, J(P,C) = 139.4 (C-4′); 16.68 d, 2 C, J(P,C) = 6.0 (Et).
MS (ESI): m/z = 681 [M + H]+.
Diethyl 9-[(N-(2-Cyanoethyl)-N-(2-phosphonoethyl))-2-
aminoethyl]hypoxanthine (5f). Starting from diethyl phosphonate
1
3f, yield 95%. H NMR (DMSO-d6): 12.27 s, 1 H (NH); 8.10 s, 1 H
and 8.03 s, 1 H (H-2 and H-8); 4.17 t, 2 H, J(1′,2′) = 6.1 (H-1′); 3.94
m, 4 H (Et); 2.85 t, 2 H, J(2′,1′) = 6.1 (H-2′); 2.76 t, 2 H, J(5′,6′) = 6.8
(H-5′); 2.66 dd, 2 H, J = 8.3 and 15.9 (H-3′); 2.49 m, 2 H (H-6′); 1.74
m, 2 H (H-4′); 1.20 t, 6 H, J = 7.1 (Et). 13C NMR (DMSO-d6): 156.57
(C-6); 148.24 (C-4); 145.21 (C-2); 140.49 (C-8); 123.71 (C-5);
119.70 (CN); 60.81 d, 2 C, J(P,C) = 6.2 (Et); 51.80, 48.22, and 46.38
(C-2′, C-5′ and C-3′); 41.54 (C-1′); 22.41 d, J(P,C) = 134.4 (C-4′);
16.11 d, 2 C, J(P,C) = 5.8 (Et); 15.49 (C-6′). MS (ESI): m/z = 397
[M + H]+.
Synthesis of Hypoxanthine Derivatives 5a−5g and 5i and
Guanine Derivatives 6a−6g and 6i, General Procedure. The
corresponding 6-chloropurine derivative 3a−3h (2 mmol) or the
corresponding 2-amino-6-chloropurine derivative 4a−4g,4i was
dissolved in trifluoroacetic acid (aqueous, 75%, 20 mL) and stirred
overnight. The solvent was evaporated and the residue codistilled with
water (3×) and ethanol. After chromatography on silica gel (MeOH−
CHCl3), the pure products were colorless foams.
Tetraethyl 9-[(N,N-(Bis-2-phosphonoethyl))-2-aminoethyl]-
hypoxanthine (5a). Starting from diethyl phosphonate 3a, yield
67%. 1H NMR (DMSO-d6): 12.29 s, 1 H (NH); 8.14 s, 1 H and 8.03
Diethyl 9-[(N-(2-Cyanopropyl)-N-(2-phosphonoethyl))-2-
aminoethyl]hypoxanthine (5g). Starting from diethyl phosphonate
6218
dx.doi.org/10.1021/jm300662d | J. Med. Chem. 2012, 55, 6209−6223

Journal of Medicinal Chemistry
Article
3g, yield 76%. 1H NMR (DMSO-d6): 12.28 s, 1 H (NH); 8.10 s, 1 H
and 8.03 s, 1 H (H-2 and H-8); 4.17 t, 2 H, J(1′,2′) = 6.1 (H-1′); 3.96
m, 4 H (Et); 2.79 t, 2 H, J(2′,1′) = 6.1 (H-2′); 2.65 dd, 2 H, J = 8.7 and
15.8 (H-3′); 2.44 t, 2 H, J(5′,6′) = 6.7 (H-5′); 2.24 t, 2 H, J(7′,6′) = 7.2
(H-7′); 1.79 m, 2 H (H-4′); 1.50 m, 2 H (H-6′); 1.21 t, 6 H, J = 7.1
(Et). 13C NMR (DMSO-d6): 156.48 (C-6); 148.23 (C-4); 145.19 (C-
2); 140.55 (C-8); 123.73 (C-5); 120.34 (CN); 60.77 d, 2 C, J(P,C) =
6.2 (Et); 51.90, 51.06, and 46.34 (C-2′, C-5′ and C-3′); 41.53 (C-1′);
22.66 (C-6′); 21.91 d, J(P,C) = 134.7 (C-4′); 16.12 d, 2 C, J(P,C) =
5.7 (Et); 13.41 (C-7′). MS (ESI): m/z = 411 [M + H]+.
Diethyl 9-[(N-(2-Hydroxyethyl)-N-(2-phosphonoethyl))-2-
aminoethyl]hypoxanthine (5i). Starting from diethyl phosphonate
3h, during the hydrolysis of 6-chloro group, the trityl moiety was
simultaneously cleaved to form hydroxydetivative 5i. The crude
product was without chromatography used directly in the next step.
MS (ESI): m/z = 388 [M + H]+.
1′); 23.17 d, J(P,C) = 136.2 (C-4′); 16.08 d, 2 C, J(P,C) = 5.7 (Et).
MS (ESI): m/z = 398 [M + H]+.
Diethyl 9-[(N-(2-Cyanoethyl)-N-(2-phosphonoethyl))-2-
aminoethyl]guanine (6f). Starting from diethyl phosphonate 4f,
yield 95%. 1H NMR (DMSO-d6): 10.63 s, 1 H (NH); 7.71 s, 1 H (H-
8); 6.50 s, 2 H (NH2); 3.94 m, 6 H (H-1′ and Et); 2.77 m, 4 H(H-2′
and H-5′); 2.66 dd, 2 H, J = 8.3 and 15.9 (H-3′); 2.54 t, 2 H, J(6′,5′) =
6.8 (H-3′); 1.75 m, 2 H (H-4′); 1.20 t, 6 H, J = 7.1 (Et). 13C NMR
(DMSO-d6): 156.66 (C-6); 153.41 (C-2); 150.96 (C-4); 137.69 (C-
8); 119.82 (CN); 116.27 (C-5); 60.84 d, 2 C, J(P,C) = 6.3 (Et); 51.68,
48.30, and 46.48 (C-2′, C-3′ and C-5′); 22.46 d, J(P,C) = 134.5 (C-4′);
16.11 d, 2 C, J(P,C) = 5.8 (Et); 15.53 (C-6′). MS (ESI): m/z = 412
[M + H]+.
Diethyl 9-[(N-(2-Cyanopropyl)-N-(2-phosphonoethyl))-2-
aminoethyl]guanine (6g). Starting from diethyl phosphonate 4g,
yield 95%. 1H NMR (DMSO-d6): 10.66 s, 1 H (NH); 7.70 s, 1 H (H-
8); 6.51 s, 2 H (NH2); 3.96 m, 6 H (H-1′ and Et); 2.74 m, 2 H(H-2′);
2.66 m, 2 H (H-3′); 2.48 m, 2 H(H-5′); 2.29 t, 2 H, J(7′,6′) = 7.1 (H-
7′); 1.82 m, 2 H (H-4′); 1.54 m, 2 H (H-6′); 1.21 t, 6 H, J = 7.0 (Et).
13C NMR (DMSO-d6): 156.65 (C-6); 153.42 (C-2); 150.98 (C-4);
Tetraethyl 9-[(N,N-(Bis-2-phosphonoethyl))-2-aminoethyl]-
1
guanine (6a). Starting from diethyl phosphonate 4a, yield 85%. H
NMR (DMSO-d6): 10.76 s, 1 H (NH); 7.86 s, 1 H (H-8); 6.58 s, 2 H
(NH2); 4.07 t, 2 H, J(1′,2′) = 5.6 (H-1′); 3.97 m, 8 H (Et); 2.95 t, 2 H,
J(2′,1′) = 5.6 (H-2′); 2.83 m, 4 H (H-6′ and H-5′); 1.91 m, 4 H (H-4′
and H-6′); 1.22 t, 12 H, J = 7.1 (Et). 13C NMR (DMSO-d6):
156.30(C-6); 153.60 (C-2); 150.74 (C-4); 137.76 (C-8); 116.25 (C-
5); 61.01 d, 4 C, J(P,C) = 6.2 (Et); 51.07 (C-2′); 46.31, 2 C (C-3′ and
C-5′); 41.02 (C-1′); 21.26 d, 2 C, J(P,C) = 134.2 (C-4′and C-6′); 16.10
d, 4 C, J(P,C) = 5.7 (Et). MS (ESI): m/z = 523 [M + H]+.
Diethyl 9-[(N-(2-Ethoxy-2-oxoethyl)-N-(2-phosphonoethyl))-2-
aminoethyl]guanine (6b). Starting from diethyl phosphonate 4b,
yield 95%. 1H NMR (DMSO-d6): 10.60 s, 1 H (NH); 7.71 s, 1 H (H-
8); 6.46 s, 2 H (NH2); 4.05 q, 2 H, J = 7.1 (Et); 3.97 t, 2 H, J(1′,2′) =
6.0 (H-1′); 3.92 m, 4 H (Et); 3.41 s, 1 H (H-5′); 2.90 t, 2 H, J(2′,1′) =
6.0 (H-2′); 2.76 dd, 2 H, J = 16.2 and 8.3 (H-3′); 1.718 m, 2 H (H-4′);
1.19 t, 6 H, J = 7.0 (Et); 1.16 t, 3 H, J = 7.1 (Et). 13C NMR (DMSO-
d6): 170.76 (CO); 156.69 (C-8); 153.42 (C-2); 151.03 (C-4); 137.94
(C-8); 116.11 (C-5); 60.87 d, 2 C, J(P,C) = 6.2 (Et); 59. 81 (Et);
53.58, 52.10, and 47.20 (C-2′, C-3′ and C-5′); 41.14 (C-1′); 23.40 d,
J(P,C) = 136.4 (C-4′); 16.15 d, 2 C, J(P,C) = 6.1 (Et); 14.03 (Et). MS
(ESI): m/z = 445 [M + H]+.
137.79 (C-8); 120.41 (CN); 116.28 (C-5); 60.84 d, 2 C, J(P,C) = 6.3
(Et); 54.22, 51.75, and 46.43 (C-2′, C-3′ and C-5′); 40.90 (C-1′); 22.04
d, J(P,C) = 155.6 (C-4′); 22.58 (C-6′); 16.12 d, 2 C, J(P,C) = 5.7 (Et);
13.48 (C-6′). MS (ESI): m/z = 426 [M + H]+.
Diethyl 9-[(N-(2-Hydroxyethyl)-N-(2-phosphonoethyl))-2-
aminoethyl]guanine (6i). Prior the above-mentioned general
procedure, diethyl phosphonate 4i had to be heated for 24 h at 80
°C in the mixture of THF (60 mL) and water (30 mL) to cleave the
triphenylphosphoranylidene moiety. Solvents were evaporated and the
residue treated with trifluoroacetic acid, yield 85%. 1H NMR (DMSO-
d6): 10.84 s, 1 H (NH); 7.72 s, 1 H (H-8); 6.65 s, 2 H (NH2); 4.01 t, 2
H, J(1′,2′) = 5.9 (H-1′); 3.95 m, 4 H (Et); 2.85 m, 2 H(H-2′); 2.79 m,
2 H(H-5′); 2.69 m, 2 H (H-3′); 1.83 m, 2 H (H-4′); 1.20 t, 6 H, J = 7.0
(Et). 13C NMR (DMSO-d6): 156.85 (C-6); 153.58 (C-2); 151.01 (C-
4); 137.90 (C-8); 116.26 (C-5); 61.00d, 2 C, J(P,C) = 6.2 (Et); 58.92
(C-6′); 55.12 (C-5′); 52.62 (C-2′); 48.50 (C-3′); 22.08 d, J(P,C) =
134.4 (C-4′); 16.12 d, 2 C, J(P,C) = 5.7 (Et). MS (ESI): m/z = 403
[M + H]+.
Diethyl 9-[(N-(3-Methoxy-3-oxopropyl)-N-(2-phosphonoethyl))-
2-aminoethyl]guanine (6c). Starting from diethyl phosphonate 4c,
yield 73%. 1H NMR (DMSO-d6): 10.74 s, 1 H (NH); 7.79 s, 1 H (H-
8); 6.56 s, 2 H (NH2); 4.07 t, 2 H, J(1′,2′) = 5.7 (H-1′); 3.57 s, 3 H
(Me); 3.99 m, 4 H (Et); 3.51 m, 4 H, (H-2′ and H-5′); 2.90 m, 4 H
(H-3′ and H-6′); 1.91 m, 2 H (H-4′); 1.22 t, 6 H, J = 7.0 (Et). 13C
NMR (DMSO-d6): 171.89 (CO); 156.29 (C-6); 153.58 (C-2); 150.72
(C-4); 137.61 (C-8); 117.13 (C-5); 61.01 d, 2 C, J(P,C) = 6.1 (Et);
51.25 (Me); 51.52, 48.05, and 46.55 (C-2′, C-3′ and C-5′); 39.22 (C-
1′); 30.99 (C-6′); 22.94 d, J(P,C) = 134.4 (C-4′); 16.10 d, 2 C, J(P,C)
= 5.8 (Et). MS (ESI): m/z = 445 [M + H]+.
Diethyl 9-[(N-(4-Methoxy-4-oxobutyl)-N-(2-phosphonoethyl))-2-
aminoethyl]guanine (6d). Starting from diethyl phosphonate 4d,
yield 57%. 1H NMR (DMSO-d6): 10.60 s, 1 H (NH); 7.65 s, 1 H (H-
8); 6.47 s, 2 H (NH2); 3.95 m, 6 H (H-1′ and Et); 3.55 s, 3H (Me);
2.70 m, 2 H, (H-2′); 2.62 m, 2 H (H-5′); 2.37 t, 2 H, J(5′,6′) = 6.7 (H-
5′); 2.14 t, 2 H, J(7′,6′) = 7.3 (H-7′); 1.76 m, 2 H (H-4′); 1.48 m, 2 H
(H-6′); 1.21 t, 6 H, J = 7.0 (Et). 13C NMR (DMSO-d6): 173.20 (CO);
156.67 (C-6); 153.33 (C-2); 150.96 (C-4); 137.75 (C-8); 116.38 (C-
5); 60.78 d, 2 C, J(P,C) = 6.2 (Et); 51.00 (Me); 51.84, 51.52, and
46.36 (C-2′, C-3′ and C-5′); 41.05 (C-1′); 30.51 (C-7′); 16.11 d, 2 C,
J(P,C) = 5.7 (Et). MS (ESI): m/z = 459 [M + H]+.
Synthesis of the Free Phosphonic Acids 7b−7g, 7i, 8b−8g,
and 8i: General Procedure. A mixture of the corresponding diethyl
ester 5a−5g,5i or 6a−6g,6i (1 mmol), acetonitrile (20 mL), 2,6-
lutidine (0.1 mL), and BrSiMe3 (1 mL) was stirred for 2 days at room
temperature. After evaporation and codistillation with acetonitrile, the
residue was treated with aqueous methanol (2: 1, 30 mL) for 1 h and
evaporated. The residue was purified by preparative HPLC (water−
methanol) or by chromatography on Dowex 1 × 2 (acetate form).
9-[(N-(2-Ethoxy-2-oxoethyl)-N-(2-phosphonoethyl))-2-
aminoethyl]hypoxanthine (7b). Starting from diethyl phosphonate
5b, yield 39%, white foam. 1H NMR (DMSO-d6): 12.26 s, 1 H (NH);
8.08 s, 1 H and 8.02 s, 1 H (H-2 and H-8); 4.17 t, 2 H, J(1′,2′) = 6.0
(H-1′); 4.03 q, 2 H, J = 7.1 (Et); 3.34 s, 2 H (H-5′); 2.94 t, 2 H, J(2′,1′)
= 6.0 (H-2′); 2.76 m, 2 H (H-3′); 1.48 m, 2 H (H-4′); 1.16 t, 3 H, J =
7.1 (Et). 13C NMR (DMSO-d6): 170.78 (CO); 156.61 (C-6); 148.30
(C-4); 145.21 (C-2); 140.71 (C-8); 123.60 (C-5); 59.77 (Et); 53.79,
52.16, and 48.34 (C-2′, C-5′ and C-3′); 41.50 (C-1′); 14.02 (Et). Anal.
Calcd for C13H20N5O6P·1/2MeOH: C, 41.65; H, 5.70; N, 17.99.
Found: C, 41.53; H, 5.45; N, 17.78. MS (ESI−): m/z = 372 [M −
H]−.
9-[(N-(3-Methoxy-3-oxopropyl)-N-(2-phosphonoethyl))-2-
aminoethyl]hypoxanthine (7c). Chromatography on Dowex 1 × 2
(acetate form, elution by 0.25 M acetic acid). Starting from diethyl
Diethyl 9-[(N-(2-Cyanomethyl)-N-(2-phosphonoethyl))-2-
aminoethyl]guanine (6e). Starting from diethyl phosphonate 4e,
yield 95%. 1H NMR (DMSO-d6): 10.66 s, 1 H (NH); 7.63 s, 1 H (H-
8); 6.52 s, 2 H (NH2); 4.04 t, 2 H, J(1′,2′) = 5.8 (H-1′); 3.94 m, 4 H
(Et); 3.87 s, 2 H(H-5′); 2.82 t, 2 H, J(2′,1′) = 5.8 (H-2′); 2.65 dd, 2 H,
J = 9.3 and 16.0 (H-3′); 1.81 m, 2 H (H-4′); 1.20 t, 6 H, J = 7.1 (Et).
13C NMR (DMSO-d6): 156.67 (C-6); 153.42 (C-2); 151.05 (C-4);
1
phosphonate 5c, yield 60%, white foam. H NMR (DMSO-d6): 12.29
s, 1 H (NH); 8.05 s, 1 H and 8.04 s, 1 H (H-2 and H-8); 4.26 t, 2 H,
J(1′,2′) = 6.1 (H-1′); 3.54 s, 3 H (Me); 2.97 t, 2 H, J(2′,1′) = 6.1 (H-
2′); 2.86 t, 2 H, J(5′,6′) = 6.8 (H-5′); 2.81 m, 2 H (H-3′); 2.42 t, 2 H,
J(6′,5′) = 6.8 (H-6′); 1.67 m, 2 H (H-4′). 13C NMR (DMSO-d6):
171.68 (CO); 156.51 (C-6); 148.21 (C-4); 145.29 (C-2); 140.47 (C-
8); 123.72 (C-5); 51.24 (Me); 51.61, 48.02, and 47.48 (C-2′, C-5′ and
C-3′); 40.64 (C-1′); 30.89 (C-6′); 24.16 d, J(P,C) = 132.2 (C-4′). Anal.
137.55 (C-8); 116.28 (C-5); 115.94 (CN); 60.93 d, 2 C, J(P,C) = 6.2
(Et); 51.75 and 46.43 (C-2′ and C-3′); 41.01 and 40.20 (C-5′ and C-
6219
dx.doi.org/10.1021/jm300662d | J. Med. Chem. 2012, 55, 6209−6223

Journal of Medicinal Chemistry
Article
Calcd for C13H20N5O6P·3/2H2O: C, 39.00; H, 5.79; N, 17.49. Found:
C, 39.00; H, 5.50; N, 17.57. MS (ESI−): m/z = 372 [M − H]−.
9-[(N-(4-Methoxy-4-oxobutyl)-N-(2-phosphonoethyl))-2-
aminoethyl]hypoxanthine (7d). Starting from diethyl phosphonate
5d, yield 41%, white solid. 1H NMR (DMSO-d6): 12.30 s, 1 H (NH);
8.08 s, 1 H and 8.03 s, 1 H (H-2 and H-8); 4.26 m, 2 H, (H-1′); 3.55 s,
3 H (Me); 2.96 m, 2 H (H-2′); 2.82 m, 2 H (H-5′); 2.55 m, 2 H (H-
3′); 2.14 t, 2 H, J(7′,6′) = 6.8 (H-7′); 1.65 m, 2 H (H-4′); 1.53 m, 2 H
(H-6′). 13C NMR (DMSO-d6): 172.93 (CO); 156.50 (C-6); 148.23
(C-4); 145.26 (C-2); 140.50 (C-8); 123.76 (C-5); 51.07 (Me); 51.58,
51.31, and 47.57 (C-2′, C-5′ and C-3′); 40.82 (C-1′); 30.32 (C-7′);
24.19 d, J(P,C) = 129.4 (C-4′); 21.16 (C-6′). Anal. Calcd for
C14H22N5O6P·H2O: C, 41.48; H, 5.97; N, 17.28. Found: C, 41.79; H,
5.81; N, 17.04. MS (ESI−): m/z = 386 [M − H]−.
2 C (C-3′ and C-5′); 38.17 (C-1′); 22.30 d, 2 C, J(P,C) = 132.4 (C-
4′and C-6′). Anal. Calcd for C11H19N5O7P: C, 33.43; H, 4.85; N,
17.72. Found: C, 33.96; H, 4.71; N, 17.46. MS (ESI−): m/z = 394 [M
− H]−.
9-[(N-(2-Ethoxy-2-oxoethyl)-N-(2-phosphonoethyl))-2-
aminoethyl]guanine (8b). Starting from diethyl phosphonate 6b,
yield 35%, white solid. 1H NMR (DMSO-d6): 10.54 s, 1 H (NH); 7.70
s, 1 H (H-8); 6.43 s, 2 H (NH2); 4.05 q, 2 H, J = 7.1 (Et); 3.97 t, 2 H,
J(1′,2′) = 6.0 (H-1′); 3.39 s, 2 H (H-5′); 2.88 t, 2 H, J(2′,1′) = 6.0 (H-
2′); 2.79 dd, 2 H, J = 7.4 and 16.0 (H-3′); 1.56 m, 2 H (H-4′); 1.17 t, 3
H, J = 7.1 (Et). 13C NMR (DMSO-d6): 170.78 (CO); 156.74 (C-6);
153.38 (C-2); 151.06 (C-4); 137.89 (C-8); 116.11 (C-5); 59.82 (Et);
53.81, 52.28, and 48.05 (C-2′, C-5′ and C-3′); 40.94 (C-1′); 26.23,
J(P,C) = 131.7 (C-4′); 14.05 (Et). Anal. Calcd for C13H21N6O6P·2/
3H2O: C, 39.00; H, 5.62; N, 20.99. Found: C, 38.85; H, 5.55; N,
20.92. MS (ESI−): m/z = 387 [M − H]−.
9-[(N-(2-Cyanomethyl)-N-(2-phosphonoethyl))-2-aminoethyl]-
hypoxanthine (7e). To improve solubility, dimethylformamide (2
mL) was added to the reaction mixture. Starting from diethyl
9-[(N-(3-Methoxy-3-oxopropyl)-N-(2-phosphonoethyl))-2-
aminoethyl]guanine (8c). To improve solubility, dimethylformamide
(6 mL) was added to the reaction mixture. Starting from diethyl
1
phosphonate 5e, yield 59%, white solid. H NMR (DMSO-d6): 12.29
s, 1 H (NH); 8.06 s, 1 H and 8.03 s, 1 H (H-2 and H-8); 4.25 t, 2 H,
J(1′,2′) = 5.7 (H-1′); 3.83 s, 2 H (H-5′); 2.85 t, 2 H, J(2′,1′) = 5.7 (H-
2′); 2.66 dd, 2 H, J = 7.5 and 15.8 (H-3′); 1.52 m, 2 H (H-4′). 13C
NMR (DMSO-d6): 156.51 (C-6); 148.28 (C-4); 145.27 (C-2); 140.44
(C-8); 123.59 (C-5); 116.11 (CN); 52.07 and 48.33 (C-2′ and C-3′);
41.22 and 40.82 (C-5′ and C-1′); 24.83 d, J(P,C) = 134.1 (C-4′). Anal.
Calcd for C11H15N6O4P·2/3H2O: C, 39.06; H, 4.87; N, 24.84. Found:
C, 38.92; H, 4.68; N, 24.56. MS (ESI−): m/z = 298 [M − H]−.
9-[(N-(2-Cyanoethyl)-N-(2-phosphonoethyl))-2-aminoethyl]-
hypoxanthine (7f). Starting from diethyl phosphonate 5f, yield 67%,
white foam. 1H NMR (DMSO-d6): 12.26 s, 1 H (NH); 8.09 s, 1 H and
8.03 s, 1 H (H-2 and H-8); 4.18 t, 2 H, J(1′,2′) = 6.2 (H-1′); 2.82 t, 2
H, J(2′,1′) = 6.2 (H-2′); 2.74 t, 2 H, J(5′,6′) = 6.7 (H-5′); 2.68 m, 2 H
(H-3′); 2.60 m, 2 H (H-6′); 1.55 m, 2 H (H-4′). 13C NMR (DMSO-
d6): 156.53 (C-6); 148.24 (C-4); 145.20 (C-2); 140.47 (C-8); 123.68
(C-5); 119.66 (CN); 51.87, 48.23, and 47.13 (C-2′, C-5′ and C-3′);
41.47 (C-1′); 24.98 d, J(P,C) = 131.9 (C-4′); 15.53 (C-6′). Anal. Calcd
for C12H17N6O4P·4/3H2O: C, 39.56; H, 5.44; N, 23.07. Found: C,
39.70; H, 5.27; N, 23.07. MS (ESI−): m/z = 339 [M − H]−.
9-[(N-(2-Cyanopropyl)-N-(2-phosphonoethyl))-2-aminoethyl]-
hypoxanthine (7g). Starting from diethyl phosphonate 5g, yield 51%,
white solid. 1H NMR (DMSO-d6): 12.34 s, 1 H (NH); 8.12 s, 1 H and
8.05 s, 1 H (H-2 and H-8); 4.32 t, 2 H, J(1′,2′) = 6.0 (H-1′); 3.10 t, 2
H, J(2′,1′) = 6.0 (H-2′); 2.92 dd, 2 H, J = 7.5 and 15.2 (H-3′); 2.75 t, 2
H, J(5′,6′) = 6.8 (H-5′); 2.38 t, 2 H, J(7′,6′) = 7.2 (H-7′); 1.74 m, 2 H
(H-4′); 1.68 m, 2 H(H-6′). 13C NMR (DMSO-d6): 156.45 (C-6);
148.26 (C-4); 145.40 (C-2); 140.40 (C-8); 123.76 (C-5); 120.04
(CN); 51.48, 50.75, and 47.57 (C-2′, C-5′ and C-3′); 40.23 (C-1′);
23.75 d, J(P,C) = 131.5 (C-4′); 21.46 (C-6′); 13.52 (C-6′). Anal. Calcd
for C13H19N6O4P·2H2O: C, 40.00; H, 5.94; N, 21.54. Found: C, 40.44;
H, 5.85; N, 21.31. MS (ESI−): m/z = 353 [M − H]−.
1
phosphonate 6c, yield 31%, white solid. H NMR (DMSO-d6): 10.55
s, 1 H (NH); 7.62 s, 1 H (H-8); 6.46 s, 2 H (NH2); 3.98 t, 2 H, J(1′,2′)
= 6.1 (H-1′); 3.54 s, 3 H (Me); 2.75 m, 6 H (H-2′, H-3′, H-5′); 2.36 t,
2 H, J(6′,5′) = 6.9 (H-6′); 1.67 m, 2 H (H-4′). 13C NMR (DMSO-d6):
172.04 (CO); 156.66 (C-6); 153.32 (C-2); 150.93 (C-4); 137.66 (C-
8); 116.24 (C-5); 51.14 (Me); 51.94, 48.13, 47.31 (C-2′, C-5′ and C-
3′); 40.59 (C-1′); 31.45 (C-6′); 24.58 d, J(P,C) = 131.5 (C-4′). Anal.
Calcd for C13H21N6O6P·H2O: C, 38.43; H, 5.71; N, 20.68. Found: C,
38.13; H, 5.86; N, 20.48. MS (ESI−): m/z = 387 [M − H]−.
9-[(N-(4-Methoxy-4-oxobutyl)-N-(2-phosphonoethyl))-2-
aminoethyl]guanine (8d). Starting from diethyl phosphonate 6d,
1
yield 58%, white foam. H NMR (DMSO-d6): 10.66 s, 1 H (NH);
7.69 s, 1 H (H-8); 6.56 s, 2 H (NH2); 4.11 t, 2 H, J(1′,2′) = 5.4 (H-1′);
3.56 s, 3 H (Me); 3.02 t, 2 H, J(2′,1′) = 5.4 (H-2′); 2.96 dd, 2 H, J =
15.0 and 8.3 (H-3′); 2.68 t, 2 H, J(5′,6′) = 6.6 (H-5′); 2.23 t, 2 H,
J(7′,6′) = 7.2 (H-7′); 1.71 m, 2 H (H-4′), 1.63 m, 2 H (H-6′). 13C
NMR (DMSO-d6): 172.88 (CO); 156.65 (C-6); 153.46 (C-2); 150.96
(C-4); 137.54 (C-8); 116.28 (C-5); 51.13 (Me); 51.50, 51.26, and
47.83 (C-2′, C-3′ and C-5′); 30.30 (C-7′); 24.10 d, J(P,C) = 129.3 (C-
4′), 20.61 (C-6′). Anal. Calcd for C14H23N6O6P·H2O: C, 40.00; H,
5.99; N, 19.99. Found: C, 39.99; H, 5.91; N, 19.77. MS (ESI): m/z =
401 [M − H]−.
9-[(N-(2-Cyanomethyl)-N-(2-phosphonoethyl))-2-aminoethyl]-
guanine (8e). To improve solubility, dimethylformamide (10 mL) was
added to the reaction mixture. Starting from diethyl phosphonate 6e,
yield 89%, white solid. 1H NMR (DMSO-d6): 11.39 s, 1 H (NH); 9.05
s, 1 H (H-8); 7.07 s, 2 H (NH2); 4.19 t, 2 H, J(1′,2′) = 5.2 (H-1′); 3.82
s, 2 H (H-5′); 4.19 t, 2 H, J(2′,1′) = 5.2 (H-2′); 2.71dd, 2 H, J = 8.5 and
16.8 (H-3′); 1.65 m, 2 H (H-4′). 13C NMR (DMSO-d6): 155.02 (C-
6); 153.73 (C-2); 149.81 (C-4); 137.49 (C-8); 116.08 (CN and C-5);
51.68 and 47.78 (C-2′ and C-3′); 41.80 and 41.27 (C-1′ and C-5′);
25.74 d, J(P,C) = 134.2 (C-4′). Anal. Calcd for C11H16N7O4P·MeOH:
C, 38.61; H, 5.40; N, 26.26. Found: C, 38.12; H, 5.15; N, 26.39. MS
(ESI−): m/z = 340 [M − H]−.
9-[(N-(2-Cyanoethyl)-N-(2- phosphonoethyl))-2-aminoethyl]-
guanine (8f). To improve solubility, dimethylformamide (4 mL)
was added to the reaction mixture. Starting from diethyl phosphonate
6f, yield 36%, white foam. 1H NMR (DMSO-d6): 10.53 s, 1 H (NH);
7.70 s, 1 H (H-8); 6.44 s, 2 H (NH2); 3.97 t, 2 H, J(1′,2′) = 6.3 (H-1′);
2.73 m, 6 H (H-2′, H-3′ and H-5′); 2.53 t, 2 H, J(6′,5′) = 6.5 (H-6′);
1.59 m, 2 H (H-4′). 13C NMR (DMSO-d6): 156.69 (C-6); 153.34 (C-
2); 150.99 (C-4); 137.65 (C-8); 119.94 (CN); 116.26 (C-5); 51.90,
48.31, and 47.24 (C-2′, C-5′ and C-3′); 40.89 (C-1′); 25.14 d, J(P,C) =
131.4 (C-4′); 15.56 (C-6′). Anal. Calcd for C12H18N7O4P·MeOH: C,
40.31; H, 5.72; N, 25.31. Found: C, 40.22; H, 5.78; N, 25.09. MS
(ESI−): m/z = 354 [M − H]−.
9-[(N-(2-Cyanopropyl)-N-(2-phosphonoethyl))-2-aminoethyl]-
guanine (8g). To improve solubility, dimethylformamide (14 mL)
was added to the reaction mixture. Starting from diethyl phosphonate
6g, yield 58%, white solid. 1H NMR (DMSO-d6): 10.61 s, 1 H (NH);
7.71 s, 1 H (H-8); 6.50 s, 2 H (NH2); 4.06 t, 2 H, J(1′,2′) = 5.8 (H-1′);
9-[(N-(2-Hydroxyethyl)-N-(2-phosphonoethyl))-2-aminoethyl]-
hypoxanthine (7i). To improve solubility, dimethylformamide (2.5
mL) was added to the reaction mixture. Chromatography on Dowex 1
× 2 (acetate form, elution by 0.25 M acetic acid). Starting from diethyl
1
phosphonate 5i, yield 36%, white foam. H NMR (DMSO-d6): 12.39
s, 1 H (NH); 8.16 s, 1 H and 8.07 s, 1 H (H-2 and H-8); 4.45 t, 2 H,
J(1′,2′) = 5.8 (H-1′); 3.60 m, 2 H (H-6′); 3.37 t, 2 H, J(2′,1′) = 5.8 (H-
2′); 3.28 m, 2 H (H-3′); 3.04 t, 2 H(H-5′); 1.86 m, 2 H (H-4′). 13C
NMR (DMSO-d6): 156.48 (C-6); 148.25 (C-4); 145.52 (C-2); 140.41
(C-8); 123.82 (C-5); 56.48, 54.51, 51.63, and 48.47 (C-6′, C-2′, C-5′
and C-3′); 23.53 d, J(P,C) = 131.2 (C-4′). Anal. Calcd for
C11H18N5O5P: C, 39.88; H, 5.48; N, 21.14. Found: C, 40.23; H,
5.524; N, 20.92. MS (ESI−): m/z = 330 [M − H]−.
9-[(N,N-(Bis-2-phosphonoethyl))-2-aminoethyl]hypoxanthine
(7m). Chromatography on Dowex 1 × 2 (acetate form, elution by
0.125 M formic acid). Starting from diethyl phosphonate 5a, yield
85%, white foam. 1H NMR (DMSO-d6): 12.64 s, 1 H (NH); 8.51 s, 1
H and 8.15 s, 1 H (H-2 and H-8); 4.64 t, 2 H, J(1′,2′) = 6.8 (H-1′);
3.67 t, 2 H, J(2′,1′) = 6.8 (H-2′); 3.35 m, 4 H (H-3′ and H-5′); 2.10 m,
4 H (H-4′ and H-6′). 13C NMR (DMSO-d6): 155.78 (C-6); 148.08
(C-4); 146.40 (C-2); 140.15 (C-8); 122.37 (C-5); 49.84 (C-2′); 47.51,
6220
dx.doi.org/10.1021/jm300662d | J. Med. Chem. 2012, 55, 6209−6223

Journal of Medicinal Chemistry
Article
2.89 m, 2 H(H-2′); 2.86 m, 2 H (H-3′); 2.62 t, 2 H, J(5′,6′) = 6.8 (H-
5′); 2.34 t, 2 H, J(7′,6′) = 7.2 (H-7′); 1.70 m, 2 H (H-4′); 1.67 m, 2 H
(H-6′). 13C NMR (DMSO-d6): 156.60 (C-6); 153.41 (C-2); 150.96
(C-4); 137.60 (C-8); 120.27 (CN); 116.21 (C-5); 51.74, 50.86, and
47.47 (C-2′, C-3′ and C-5′); 40.23 (C-1′); 24.23 d, J(P,C) = 130.73 (C-
4′); 21.96 (C-6′); 13.51 (C-6′). Anal. Calcd for C13H20N7O4P·3/
2H2O: C, 39.40; H, 5.85; N, 24.74. Found: C, 39.12; H, 5.46; N,
24.33. MS (ESI): m/z = 368 [M + H]+.
9-[(N-(2-Hydroxyethyl)-N-(2-phosphonoethyl))-2-aminoethyl]-
guanine (8i). To improve solubility, dimethylformamide (14 mL) was
added to the reaction mixture. Starting from diethyl phosphonate 6i,
yield 62%, white solid. 1H NMR (DMSO-d6): 10.74 s, 1 H (NH); 7.75
s, 1 H (H-8); 6.62 s, 2 H (NH2); 4.27 t, 2 H, J(1′,2′) = 6.2 (H-1′); 3.64
t, 2 H, J(6′,5′) = 5,4 (H-6′); 3.38 t, 2 H, J(2′,1′) = 6.2 (H-2′); 3.27 m, 2
H (H-3′); 3.11 t, 2 H, J = 5.4 (H-5′); 1.90 m, 2 H (H-4′). 13C NMR
(DMSO-d6): 157.01 (C-6); 153.95 (C-2); 151.26 (C-4); 137.87 (C-
8); 116.58 (C-5); 56.29 (C-6′), 54.62, 51.98, and 59.11 (C-5′, C-2′ and
C-3′); 38.89 (C-1′); 23.64 d, J(P,C) = 130.8 (C-4′). Anal. Calcd for
C11H19N6O5P·H2O: C, 36.27; H, 5.81; N, 23.07. Found: C, 35.99; H,
5.65; N, 22.72. MS (ESI−): m/z = 345 [M − H]−.
H (H-6′). 13C NMR (DMSO-d6): 174.03 (CO); 156.49 (C-6); 148.25
(C-4); 145.29 (C-2); 140.49 (C-8); 123.74 (C-5); 51.68, 51.48, and
47.72 (C-2′, C-5′ and C-3′); 40.68 (C-1′); 30.65 (C-7′); 24.20 d,
J(P,C) = 131.5 (C-4′); 21.12 (C-6′). Anal. Calcd for C13H20N5O6P·3/
2H2O: C, 40.59; H, 6.39; N, 17.75. Found: C, 40.64; H, 6.21; N,
17.62. MS (ESI−): m/z = 372 [M − H]−.
9-[(N-(Carboxymethyl)-N-(2-phosphonoethyl))-2-aminoethyl]-
guanine (8j). Starting from diethyl phosphonate 6b, yield 54%, white
1
solid. H NMR (DMSO-d6): 10.58 s, 1 H (NH); 7.73 s, 1 H (H-8);
6.46 s, 2 H (NH2); 4.00 t, 2 H, J(1′,2′) = 6.1 (H-1′); 3.34 s, 2 H (H-5′);
2.93 t, 2 H, J(2′,1′) = 6.1 (H-2′); 2.84 dd, 2 H, J = 8.1 and 16.1 (H-3′);
1.61 m, 2 H (H-4′); 1.17 t, 3 H, J = 7.1 (Et). 13C NMR (DMSO-d6):
172.13 (CO); 156.70 (C-6); 153.43 (C-2); 151.00 (C-4); 137.88 (C-
8); 116.06 (C-5); 54.01, 52.51, and 48.18 (C-2′, C-5′ and C-3′); 40.78
(C-1′); 25.97, J(P,C) = 133.6 (C-4′). Anal. Calcd for C11H17N6O6P·3/
2H2O: C, 34.11; H, 5.21; N, 21.70. Found: C, 33.82; H, 5.19; N,
21.29. MS (ESI−): m/z = 359 [M − H]−.
9-[(N-(Carboxypropyl)-N-(2-phosphonoethyl))-2-aminoethyl]-
guanine (8l). Starting from diethyl phosphonate 6d, yield 58%, white
1
solid. H NMR (DMSO-d6): 10.61 s, 1 H (NH); 7.68 s, 1 H (H-8);
6.54 s, 2 H (NH2); 4.08 t, 2 H, J(1′,2′) = 5.8 (H-1′); 2.95 t, 2 H, J(2′,1′)
= 5.8 (H-2′); 2.89 m, 2 H (H-3′); 2.63 t, 2 H, J(5′,6′) = 7.0 (H-5′); 2.17
t, 2 H, J(7′,6′) = 7.2 (H-7′); 1.68 m, 2 H (H-4′), 1.60 m, 2 H (H-6′).
13C NMR (DMSO-d6): 174.08 (CO); 156.64 (C-6); 153.42 (C-2);
151.00 (C-4); 137.53 (C-8); 116.27 (C-5); 51.78, 51.50, and 47.85
(C-2′, C-3′ and C-5′); 39.99 (C-1′); 30.87 (C-7′); 24.30 d, J(P,C) =
132.91 (C-4′), 20.96 (C-6′). Anal. Calcd for C13H21N6O6P·H2O: C,
38.43; H, 5.71; N, 20.68. Found: C, 38.51; H, 5.64; N, 20.53. MS
(ESI−): m/z = 387 [M − H]−.
9-[(N,N-(Bis-2-phosphonoethyl))-2-aminoethyl]guanine (8m).
Starting from diethyl phosphonate 6a, yield 51%, white solid. 1H
NMR (DMSO-d6): 8.36 s, 1 H (H-8); 7.98 s, 4.62 t, 2 H, J(1′,2′) = 6.0
(H-1′); 3.73 t, 2 H, J(2′,1′) = 6.3 (H-2′); 3.52 dd, 4 H, J = 8.2 and 16.4
(H-3′ and H-5′); 2.02 m, 4 H (H-4′ and H-6′). 13C NMR (DMSO-d6):
156.74 (C-6); 155.08 (C-2); 151.22 (C-4); 138.5 (C-8); 116.32 (C-
5); 51.21 (C-2′); 49.44, 2 C (C-3′ and C-5′); 39.56 (C-1′); 22.78 d, 2
C, J(P,C)
= 129.41 (C-4′and C-6′). Anal. Calcd for
C15H23N6O4P·2H2O: C, 29.60; H, 5.42; N, 18.83. Found: C, 29.36;
H, 5.37; N, 19.14. MS (ESI−): m/z = 409 [M − H]−.
9-[(N-(2-Carboxyethyl)-N-(2-phosphonoethyl))-2-aminoethyl]-
guanine (8k). A mixture of cyanoderivative 6f (0.41 g, 1 mmol),
methanol (15 mL), and aqueous NaOH (25%, 3 mL) was refluxed for
4 h and then stirred at room temperature overnight. After evaporation
and codistillation with toluene/ethanol and acetonitrile, the residue
was dissolved in acetonitrile (20 mL). 2,6-Lutidine (0.1 mL) and
BrSiMe3 (1 mL) were added and the mixture stirred for 2 days at room
temperature. After evaporation and codistillation with acetonitrile, the
residue was treated with aqueous methanol (2: 1, 30 mL) for 1 h and
evaporated. The residue was purified by preparative HPLC (water−
methanol), yield 45%, white solid. 1H NMR (D2O): 7.64 s, 1 H (H-8);
4.73 t, 2 H, J(1′,2′) = 6.0 (H-1′); 3.84 t, 2 H, J(2′,1′) = 6.0 (H-2′); 3.62
t, J(5′,6′) = 6.5,2 H (H-5′); 3.58 m, 2 H (H-3′); 2.92 t, 2 H, J(6′,5′) =
6.5 (H-6′); 2.08 m, 2 H (H-4′). 13C NMR (D2O): 174.21 (CO);
156.44 (C-6); 155.32 (C-2); 151.01 (C-4); 138.57 (C-8); 116.21 (C-
5); 51.80, 50.29, and 49.72 (C-2′, C-3′ and C-5′); 40.29 (C-1′); 28.54
(C-6′); 22.62 d, J(P,C) = 129.3 (C-4′). Anal. Calcd for
C12H19N6O6P·4/3H2O: C, 36.18; H, 5.48; N, 21.10. Found: C,
36.41; H, 5.23; N, 20.86. MS (ESI−): m/z = 373 [M − H]−.
Synthesis of Phosphonic Acids 7n and 8n, General
Procedure. The corresponding cyanomethyl derivative 7e or 8e
(0.25 mmol) in dimethylformamide (10 mL) was heated at 120 °C for
3 days. The solvent was then evaporated and the residue codistilled
with toluene. The residue was purified by preparative HPLC (water−
methanol).
9-[(N-(2-Phosphonoethyl))-2-aminoethyl]hypoxanthine (7n).
Starting from diethyl phosphonate 7e, yield 49%, white solid. 1H
NMR (D2O + NaOD): 8.04 s, 1 H and 7.88 s, 1 H (H-2 and H-8);
4.21 t, 2 H, J(1′,2′) = 6.2(H-1′); 2.96 t, 2 H, J(2′,1′) = 6.2 (H-2′); 2.69
dd, 2 H, J = 7.3 and 16.3 (H-3′); 1.50 m, 2 H (H-4′). 13C NMR (D2O
+ NaOD): 168.34 (C-6); 154.10 (C-2); 150.33 (C-4); 123.76 (C-5);
47.80, 45.21, and 43.89 (C-2′, C-3′ and C-1′); 29.99 d, J(P,C) = 127.2
(C-4′). Anal. Calcd for C9H14N5O4P·H2O: C, 35.41; H, 5.28; N, 22.94.
Found: C, 35.06; H, 5.29; N, 23.39. MS (ESI−): m/z = 286 [M −
H]−.
Synthesis of Carboxylic Acids 7j−7l, 8j: and 8l, General
Procedure. A mixture of corresponding ester 5b−5d, 6b, or 6d (1
mmol), tetrahydrofurane (10 mL), methanol (10 mL), and aqueous
NaOH (10 M, 0.2 mL) was refluxed for 2 h and then stirred at room
temperature overnight. After evaporation and codistillation with
toluene/ethanol and acetonitrile, the residue was dissolved in
acetonitrile (20 mL) and dimethylformamide (8 mL). 2,6-Lutidine
(0.1 mL) and BrSiMe3 (1 mL) were added, and the mixture was stirred
for 2 days at room temperature. After evaporation and codistillation
with acetonitrile, the residue was treated with aqueous methanol (2: 1,
30 mL) for 1 h, evaporated, and codistilled with water. The residue
was purified by preparative HPLC (water−methanol).
9-[(N-(Carboxymethyl)-N-(2-phosphonoethyl))-2-aminoethyl]-
hypoxanthine (7j). Starting from diethyl phosphonate 5b, yield 35%,
white solid. 1H NMR (DMSO-d6): 12.29 s, 1 H (NH); 8.11 s, 1 H and
8.03 s, 1 H (H-2 and H-8); 4.20 t, 2 H, J(1′,2′) = 6.0 (H-1′); 3.35 s, 2
H (H-5′); 2.97 t, 2 H, J(2′,1′) = 6.0 (H-2′); 2.80 m, 2 H (H-3′); 1.54 m,
2 H (H-4′). 13C NMR (DMSO-d6): 172.22 (CO); 156.61 (C-6);
148.30 (C-4); 145.32 (C-2); 140.72 (C-8); 123.63 (C-5); 53.86,
52.34, and 48.25 (C-2′, C-5′ and C-3′); 41.39 (C-1′); 26.05 d, J(P,C) =
131.4 (C-4′). Anal. Calcd for C11H16N5O6P·1/2H2O: C, 37.29; H,
4.84; N, 19.77. Found: C, 37.48; H, 4.65; N, 19.49. MS (ESI−): m/z =
344 [M − H]−.
9-[(N-(2-Carboxyethyl)-N-(2-phosphonoethyl))-2-aminoethyl]-
hypoxanthine (7k). Starting from diethyl phosphonate 5c, yield 60%,
white foam. 1H NMR (DMSO-d6): 12.26 s, 1 H (NH); 8.05 s, 1 H and
8.03 s, 1 H (H-2 and H-8); 4.21 t, 2 H, J(1′,2′) = 6.1 (H-1′); 2.86 t, 2
H, J(2′,1′) = 6.1 (H-2′); 2.77 t, 2 H, J(5′,6′) = 6.9 (H-5′); 2.74 m, 2 H
(H-3′); 2.30 t, 2 H, J(6′,5′) = 6.9 (H-6′); 1.56 m, 2 H (H-4′). 13C NMR
(DMSO-d6): 173.12 (CO); 156.52 (C-6); 148.21 (C-4); 145.24 (C-
2); 140.54 (C-8); 123.68 (C-5); 51.99, 48.29 and 47.54 (C-2′, C-5′ and
C-3′); 41.16 (C-1′); 31.58 (C-6′); 24.59 d, J(P,C) = 131.6 (C-4′). Anal.
Calcd for C12H18N5O6P·5/3H2O: C, 37.02; H, 5.52; N, 17.99. Found:
C, 37.18; H, 5.55; N, 17.75. MS (ESI−): m/z = 358 [M − H]−.
9-[(N-(Carboxypropyl)-N-(2-phosphonoethyl))-2-aminoethyl]-
hypoxanthine (7l). Starting from diethyl phosphonate 5d, yield 56%,
white solid. 1H NMR (DMSO-d6): 12.31 s, 1 H (NH); 8.10 s, 1 H and
8.03 s, 1 H (H-2 and H-8); 4.28 t, 2 H, J(1′,2′) = 6.0 (H-1′); 3.00 t, 2
H, J(2′,1′) = 6.0 (H-2′); 2.84 m, 2 H (H-3′); 2.60 t, 2 H, J(5′,6′) = 6.9
(H-5′); 2.12 t, 2 H, J(7′,6′) = 7.2 (H-7′); 1.65 m, 2 H (H-4′); 1.56 m, 2
9-[(N-(2-Phosphonoethyl))-2-aminoethyl]guanine (8n). Starting
from diethyl phosphonate 8e, yield 30%, white solid. 1H NMR (D2O +
NaOD): 7.59 s, 1 H (H-8); 4.05 t, 2 H, J(1′,2′) = 6.3 (H-1′); 2.88 t, 2
H, J(2′,1′) = 6.3 (H-2′); 2.66 dd, 2 H, J = 7.2 and 16.3 (H-3′); 1.470 m,
2 H (H-4′). 13C NMR (D2O + NaOD): 168.68 (C-6); 161.54 (C-2);
151.81 (C-4); 138.94 (C-8); 117.99 (C-5); 47.63, 45.02, and 43.25
6221
dx.doi.org/10.1021/jm300662d | J. Med. Chem. 2012, 55, 6209−6223

Journal of Medicinal Chemistry
Article
(C-2′, C-3′ and C-1′); 29.82 d, J(P,C) = 127.6 (C-4′). Anal. Calcd for
C13H21N6O6P·H2O: C, 38.43; H, 5.71; N, 20.68. Found: C, 38.51; H,
5.64; N, 20.53. HRMS calcd for C9H14N6O4P, 301.08196; found,
301.08196. MS (ESI−): m/z = 301 [M − H]−.
Determination of Ki Values. The Ki values were determined
using a spectrophotometric assay at 25 °C, 0.1 M Tris-HCl, 10 mM
MgCl2, and pH 7.4 with PRib-PP as the variable substrate and guanine
as the fixed substrate.22 The Ki values are Ki(app) as they were measured
at a single concentration of the second substrate. The concentration of
the second substrate (guanine) was saturating: 60 μM. Ki(app) was
calculated using the equation Km(app) = Km (1 + [I]/Ki(app)).
Docking of the AzaANPs into the Active Site of Human
HGPRT. Models for the ANPs were generated using the Dundee
PRODRUG server.23 Coordinates for the human HGPRT structure
were obtained from the Protein Data Bank. Only Chain A from the
coordinate files were used for docking. All of the water molecules and
ligands were removed prior to docking. Compounds 7f and 8f had the
highest docking scores when docked into 3GEP (human HGPRT in
complex with 3-hydroxy-2-(phosphonomethoxy)propyl guanine).4 All
of the other compounds had the highest docking scores when docked
into 3GGJ (human HGPRT in complex with 2-(phosphonoethoxy)-
ethyl guanine or hypoxanthine).4 The program GOLD24 was used for
all the docking calculations. All hydrogen atoms were added to the
protein within GOLD. The active site center was defined by the
coordinates of a carboxylate oxygen atom on the side chain of D13.
The search radius around this point was 15 Å. For each ligand and
protein, 20 independent docking searches were undertaken. Scoring
was made with the default setting, ChemPLP.25
́
́
(4) Keough, D. T.; Hockova, D.; Holy, A.; Naesens, L.; Skinner-
Adams, T.; de Jersey, J.; Guddat, L. W. Inhibition of Hypoxanthine−
Guanine Phosphoribosyltransferase by Acyclic Nucleoside Phospho-
nates: A New Class of Antimalarial Therapeutics. J. Med. Chem. 2009,
52, 4391−4399.
́
́ ́
(5) (a) Hockova, D.; Holy, A.; Masojídkova, M.; Keough, D. T.; de
Jersey, J.; Guddat, L. W. Synthesis of branched 9-[2-(2-
phosphonoethoxy)ethyl]purines as a new class of acyclic nucleoside
phosphonates which inhibit Plasmodium falciparum hypoxanthine−
guanine−xanthine phosphoribosyltransferase. Bioorg. Med. Chem.
2009, 17, 6218−6232. (b) Krec
Hockova,
̌ ́ ̌
́
merova, M.; Dracínsky, M.;
́
D.; Holy, A.; Keough, D. T.; Guddat, L. W. Synthesis of
́
purine N9-[2-hydroxy-3-O-(phosphonomethoxy)propyl] derivatives
and their side chain modified analogues as potential antimalarial
agents. Bioorg. Med. Chem. 2012, 20, 1222−1230.
̌
́ ̌ ́
(6) Keough, D. T.; Hockova, D.; Krecmerova, M.; Cesnek, M.; Holy,
́
A.; Naesens, L.; Brereton, I.; Winzor, D. J.; de Jersey, J.; Guddat, L. W.
Plasmodium vivax hypoxanthine−guanine phosphoribosyltransferase: a
target for antimalarial chemotherapy. Mol. Biochem. Parasitol. 2010,
173, 165−169.
̌
(7) Cesnek, M.; Hockova,
́
D.; Holy, A.; Drací
̌
nsky, M.; Baszczyn
̌
ski,
́
́
O.; de Jersey, J.; Keough, D. T.; Guddat, L. W. Synthesis of 9-
phosphonoalkyl and 9-phosphonoalkoxyalkyl purines: evaluation of
their ability to act as inhibitors of Plasmodium falciparum, Plasmodium
vivax and human hypoxanthine−guanine−(xanthine) phosphoribosyl-
transferases. Bioorg. Med. Chem. 2012, 20, 1076−1089.
(8) (a) Li, C. M.; Tyler, P. C.; Furneaux, R. H.; Kicska, G.; Xu, Y. M.;
Grubmeyer, C.; Girvin, M. E.; Schramm, V. L. Transition-state analogs
as inhibitors of human and malarial hypoxanthine−guanine phosphor-
ibosyltransferases. Nature Struct. Biol. 1999, 6, 582−587. (b) Shi, W.;
Li, C. M.; Tyler, P. C.; Furneaux, R. H.; Grubmeyer, C.; Schramm, V.
L.; Almo, S. C. The 2.0 Å structure of human hypoxanthine−guanine
phosphoribosyltransferase in complex with a transition-state analog
inhibitor. Nature Struct. Biol. 1999, 6, 588−593.
AUTHOR INFORMATION
Corresponding Author
*Phone: +420 220183262. E-mail: lasice@uochb.cas.cz.
■
Notes
The authors declare no competing financial interest.
(9) Berg, M.; Van der Veken, P.; Goeminne, A.; Haemers, A.;
Augustyns, K. Inhibitors of the Purine Salvage Pathway: A Valuable
Approach for Antiprotozoal Chemotherapy. Curr. Med. Chem. 2010,
17, 2456−2481.
ACKNOWLEDGMENTS
■
This work was supported by the subvention for development of
research organization (Institute of Organic Chemistry and
Biochemistry) RVO 61388963, by the Grant Agency of the
Czech Republic (grant no. P207/11/0108), Centre for New
Antivirals and Antineoplastics (1M0508), funds from the
National Health and Medical Research Council, Australia
(grant nos. 569703 and 1030353), and by Gilead Sciences
(Foster City, CA, USA).
́
́
(10) de Jersey, J.; Holy, A.; Hockova, D.; Naesens, L.; Keough, D. T.;
Guddat, L. W. 6-Oxopurine Phosphoribosyltransferase: A Target for
the Development of Antimalarial Drugs. Curr. Top. Med. Chem. 2011,
11, 2085−2102.
(11) Cassera, M. B.; Zhang, Y.; Hazleton, K. Z.; Schramm, V. L.
Pyrimidine Pathways as Targets in Plasmodium falciparum. Curr. Top.
Med. Chem. 2011, 11, 2103−2115.
(12) Snow, R. W.; Guerra, C. A.; Noor, A. M.; Myint, H. Y.; Hay, S. I.
The global distribution of clinical episodes of Plasmodium falciparum
malaria. Nature 2005, 434, 214−217.
ABBREVIATIONS USED
■
ANPs, acyclic nucleoside phosphonates; aza-ANPs, acyclic
nucleoside phosphonates with build-in nitrogen atom in the
acyclic moiety; CRFK, feline renal cells; HEL, human
embryonic lung cells; HG(X)PRT, hypoxanthine-guanine-
(xanthine) phosphoribosyltransferase; PEEG, 2-
(phosphonoethoxy)ethyl guanine; PEEHx, 2-
(phosphonoethoxy)ethyl hypoxanthine; Pf , Plasmodium falci-
parum; PPi, pyrophosphate; PRib-PP, 5-phosphoribosyl-1-
pyrophosphate; Pv, Plasmodium vivax; Vero, kidney epithelial
cells
(13) Hyde, J. E. Drug-resistant malariaan insight. FEBS J. 2007,
274, 4688−4698.
(14) Hickman, D. T.; Sreenivasachary, N.; Lehn, J.-M. Synthesis of
components for the generation of constitutional dynamic analogues of
nucleic acids. Helv. Chim. Acta 2008, 91, 1−20.
(15) Sreenivasachary, N.; Hickman, D. T.; Sarazin, D.; Lehn, J.-M.
DyNAs: Constitutional dynamic nucleic acid analogues. Chem.Eur. J.
2006, 12, 8581−8588.
(16) Levy, D. E.; Marlowe, C.; Kane-Maguire, K.; Bao, M.;
Cherbavaz, D. B.; Tomlinson, J. E.; Sedlock, D. M.; Scarborough, R.
M. Hydroxamate based inhibitors of adenylyl cyclase. Part 1: The
effect of acyclic linkers on P-site binding. Bioorg. Med. Chem. Lett.
2002, 12, 3085−3088.
(17) Zhou, D.; Lagoja, I. M.; Van Aerschot, A.; Herdewijn, P.
Synthesis of aminopropyl phosphonate nucleosides with purine and
pyrimidine bases. Collect. Czech. Chem. Commun. 2006, 71, 15−34.
(18) Matveeva, E. V.; Petrovskii, P. V.; Odinets, I. L. Efficient
synthesis of racemic β-aminophosphonates via aza-Michael reaction in
water. Tetrahedron Lett. 2008, 49, 6129−6133.
REFERENCES
■
(1) Holy, A. Phosphonomethoxyalkyl analogs of nucleotides. Curr.
́
Pharm. Des. 2003, 9, 2567−2592.
(2) De Clercq, E.; Holy, A. Acyclic nucleoside phosphonates: a key
́
class of antiviral drugs. Nature Rev. Drug Discovery 2005, 4, 928−940.
(3) Smeijsters, L. J.; Franssen, F. F.; Naesens, L.; de Vries, E.; Holy,
́
A.; Balzarini, J.; de Clercq, E.; Overdulve, J. P. Inhibition of the in vitro
growth of Plasmodium falciparum by acyclic nucleoside phosphonates.
Int. J. Antimicrob. Agents 1999, 12, 53−61.
6222
dx.doi.org/10.1021/jm300662d | J. Med. Chem. 2012, 55, 6209−6223

Journal of Medicinal Chemistry
Article
(19) Mitsunobu, O. The use of diethyl azodicarboxylate and
triphenylphosphine in synthesis and transformation of natural
products. Synthesis 1981, 1−28.
(20) Ludek, O. R.; Meier, C. Influence of the N3-protection group on
N1- vs. O2-Alkylation in the Mitsunobu reaction. Eur. J. Org. Chem.
2006, 941−946.
(21) Keough, D. T.; Skinner-Adams, T.; Jones, M. K.; Ng, A. L.;
Brereton, I. M.; Guddat, L. W.; de Jersey, J. Lead compounds for
antimalarial chemotherapy: purine base analogs discriminate between
human and P. falciparum 6-oxopurine phosphoribosyltransferases. J.
Med. Chem. 2006, 49, 7479−7486.
(22) Keough, D. T.; Ng, A. L.; Winzor, D. J.; Emmerson, B. T.; de
Jersey, J. Purification and characterization of Plasmodium falciparum
hypoxanthine−guanine−xanthine phosphoribosyltransferase and com-
parison with the human enzyme. Mol. Biochem. Parasitol. 1999, 98,
29−41.
(23) Schuttelkopf, A. W.; van Aalten, D. M. F. PRODRG: a tool for
̈
high-throughput crystallography of protein−ligand complexes. Acta
Crystallogr., Sect. D: Biol. Crystallogr. 2004, D60, 355−1363.
(24) Jones, G; Willett, P.; Glen, R. C.; Leach, A. R.; Taylor, R.
Development and validation of a genetic algorithm flexible docking. J.
Biol. Chem. 1997, 267, 727−748.
(25) Korb, O.; Stutzle, T.; Exner, T. E. Empirical scoring functions
̈
for advanced protein−ligand docking with PLANTS. J. Chem. Inf.
Model. 2009, 49, 84−96.
6223
dx.doi.org/10.1021/jm300662d | J. Med. Chem. 2012, 55, 6209−6223