Aryl H -Phosphonates 18. Synthesis, properties, and biological activity of 2′,3′-dideoxynucleoside (N -heteroaryl)phosphoramidates of increased lipophilicity

Article abstract of DOI:10.1016/j.ejmech.2015.06.004

Recently, AZT (N-pyridyl)phosphoramidates were reported as a new type of potential anti-HIV therapeutics. In continuation of that work, here we present new (N-heteroaryl)phosphoramidate derivatives of antiviral 2′,3′-dideoxynucleosides containing other types of N-heteroaryl moieties, particularly those with higher lipophilicity. The present studies comprise mechanistic investigations using ³¹P NMR correlation analysis, which permitted improvements in the synthetic procedures. The obtained compounds were tested in biological systems to establish their cytotoxicity and anti-HIV activity. The results were analyzed with respect to possible correlations between biological and physico-chemical properties of the phosphoramidates studied, to get some insight into their antiviral mode of action.

Full text of DOI:10.1016/j.ejmech.2015.06.004

Accepted Manuscript
Aryl H-Phosphonates 18. Synthesis, properties, and biological activity of 2',3'-
dideoxynucleoside (N-heteroaryl)phosphoramidates of increased lipophilicity
Krystian Kolodziej, Joanna Romanowska, Jacek Stawinski, Jerzy Boryski, Aleksandra
Dabrowska, Andrzej Lipniacki, Andrzej Piasek, Adam Kraszewski, Michal Sobkowski
PII:
S0223-5234(15)30079-9
10.1016/j.ejmech.2015.06.004
EJMECH 7931
DOI:
Reference:
To appear in: European Journal of Medicinal Chemistry
Received Date: 6 March 2015
Revised Date: 22 April 2015
Accepted Date: 1 June 2015
Please cite this article as: K. Kolodziej, J. Romanowska, J. Stawinski, J. Boryski, A. Dabrowska, A.
Lipniacki, A. Piasek, A. Kraszewski, M. Sobkowski, Aryl H-Phosphonates 18. Synthesis, properties, and
biological activity of 2',3'-dideoxynucleoside (N-heteroaryl)phosphoramidates of increased lipophilicity,
European Journal of Medicinal Chemistry (2015), doi: 10.1016/j.ejmech.2015.06.004.
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT
Aryl H-Phosphonates 18. Synthesis, properties, and biological activity of 2',3'-
dideoxynucleoside (N-heteroaryl)phosphoramidates of increased lipophilicity
a
a
a
a
Krystian Kolodziej, Joanna Romanowska, Jacek Stawinski, Jerzy Boryski, Aleksandra
a
b
b
a*
Dabrowska, Andrzej Lipniacki, Andrzej Piasek, Adam Kraszewski, and Michal
a*
Sobkowski
a
Institute of Bioorganic Chemistry, Polish Academy of Sciences, Noskowskiego 12/14, 61-704 Poznań, Poland
Corresponding authors: Adam.Kraszewski@ibch.poznan.pl (AK) and msob@ibch.poznan.pl (MS)
*
b
National Institute of Medicines, Chełmska 30/34, 00-725 Warsaw, Poland
Keywords:
pro-nucleotides;
AZT;
2',3'-dideoxynucleosides;
H-phosphonates;
phosphoramidates; heteroaryl amines; anti-HIV; lipophilicity
Abstract
Recently, AZT (N-pyridyl)phosphoramidates were reported as a new type of potential
anti-HIV therapeutics. In continuation of that work, here we present new (N-
heteroaryl)phosphoramidate derivatives of antiviral 2',3'-dideoxynucleosides containing other
types of N-heteroaryl moieties, particularly those with higher lipophilicity. The present studies
3
1
comprise mechanistic investigations using P NMR correlation analysis, which permitted
improvements in the synthetic procedures. The obtained compounds were tested in biological
systems to establish their cytotoxicity and anti-HIV activity. The results were analyzed with
respect to possible correlations between biological and physico-chemical properties of the
phosphoramidates studied, to get some insight into their antiviral mode of action.
Introduction
2',3'-Dideoxy derivatives of nucleosides (e.g. AZT, d4T, 3TC) are known to inhibit viral
reverse transciriptases and some of them are used extensively in treatment of retroviral
infections [1]. Nevertheless, these compounds are far from being optimal drugs and are still
under thorough investigations aimed to improve their pharmacological features. One of the
challenges is to overcome the AZT toxicity associated with accumulation of AZT
monophosphate (AZTMP), which formed in cells rapidly from the parent nucleoside in a
reaction catalyzed by thymidine kinase (TK). In contrast, phosphorylation of AZTMP to
AZTDP by thymidylate kinase is slow, leading to accumulation of AZTMP and decreasing
total AZT antiviral efficacy. The third phosphorylation, AZTDP → AZTTP, catalyzed by
unspecific 5'-nucleoside diphosphate kinase, is rapid again [2]. Moreover, the large amounts
of the accumulated AZTMP inhibit thymidylate kinase by binding to its ATP site [3], and in
1

ACCEPTED MANUSCRIPT
consequence, affect also phosphorylation of the natural TMP, and finally, DNA replication.
The simplest way to by-pass this problem would be administration of proper doses of already
phosphorylated AZT; unfortunately, due to the negative charge of the phosphate residue,
AZTMP it is not able to pass the cell membrane. Thus, a so-called 'pro-nucleotide approach'
was developed, [4] in which the phosphate group of a therapeutic nucleotide analogue was
appropriately protected and in such uncharged form, was able to surmount the lipid barrier of
a cell. Subsequently, inside the cell, the phosphate protecting groups were cleaved off
yielding the desired AZTMP.
Recently, we developed AZT (N-pyridyl)phosphoramidates (cf. Fig. 1) as a new type of
pro-nucleotides, which exhibited high anti-HIV activity and low cytotoxicity on one hand,
and very good chemical stability, resistance to enzymatic degradation, and excellent solubility
in water, on the other one [5]. Additionally, preliminary tests showed low acute toxicity of
these species in animal models: 7ab (Ar = 2-pyridyl): GHS class 5 (LD = 2500 mg/kg, rats,
5
0
oral); 7ab (Ar = 4-pyridyl): GHS class 2 (LD = 50 mg/kg, rats, oral) or class 3 (LD = 200
5
0
50
mg/kg, rats, iv); 7ac (Ar = 3-pyridyl): GHS class 5 (beyond classification, no lethality at 2000
mg/kg, rats, oral). In the same study, LD for AZT was found to be 2500 mg/kg (oral) and
5
0
5
00 mg/kg (iv) [6].
High biological activity indicated that despite their anionic character, AZT (N-
pyridyl)phosphoramidate monoesters were able to enter the cell effectively. We rationalized
this by assuming that in aqueous solutions, apart form typical ionic forms these compounds
exist – to small but still significant extent – as protonated, neutral species (Fig. 1), which
apparently can penetrate the cellular lipid membranes [5]. Compatibility of (N-
pyridyl)phosphoramidates with aqueous and hydrophobic media may be also a result of the
dual nature of the pyridine ring, which is both lipophilic due to its hydrocarbon fragment, and
hydrophilic due to the presence of the nitrogen atom.
FIGURE 1
These speculations prompted us to study the impact of the structure, lipophilicity and
basicity of the parent heterocyclic amines on anti-HIV activity of the derived nucleoside
phosphoramidates.
Results and discussion
In order to elucidate the topics delineated above, three series of heteroaryl 5'-
phosphoramidates were prepared: derivatives of aminoquinolines, aminocyanopyridines, and
aminothiazolines. Among 14 possible isomers of aminoquinolines and aminoisoquinolines, 8
2

ACCEPTED MANUSCRIPT
were readily commercially available and these were included in this study. Various positions
of the nitrogen atoms in these isomers lead not only to different spatial arrangement of the
(iso)quinoline groups, but also have a significant impact on lipophilicity and basicity of the
designed nucleoside phosphoramidates (Table 1, entries 5–11). Permeability through cell
membranes, which is a key feature of a compound considered as a prospective drug, usually
correlates linearly with solubility of a substance in nonpolar solvents (Overton's rule), and
may be conveniently expressed as an octanol-water partition coefficient, log P [7,8]. Most of
amino(iso)quinolines have log P in a range of 1.2–1.9, which is roughly an order of
magnitude higher than those of the unsubstituted aminopyridines (log P = 0.2, 0.3, and 0.5 for
*
m, p, and o isomer, respectively ) used in the previous study [5]. Thus, phosphoramidates
containing a quinoline residue may be expected to penetrate the cell membranes more
effectively than their pyridyl congeners. Next, pK 's of amines 2d–j are spread over the range
a
of 4.0–9.6. This should influence the acid-base equilibria of the derived phosphoramidates,
and consequently, the fraction of their neutral, protonated forms under physiological pH,
which are expected to surmount the cell membrane easier than the ionic forms. Finally, the
aforementioned different spatial arrangement of the isomeric (iso)quinoline groups in
nucleoside phosphoramidates may have additional impact on their biological activity in view
of the recent findings by Herdewijn et al., who observed that some nucleoside 5'-
phosphoramidates can be used as substrates by HIV reverse transcriptase (without conversion
into the respective triphosphates), and that structural variations in the amide part had
fundamental impact on their substrate properties [9-11].
The two members in the second set of heterocyclic amines included in this study contain
a nitrile group in the pyridine ring (Table 1, entries 12–13), which decreased significantly
their basicity (estimated pK 0.5–3 vs. 6.0–6.7 for the respective aminopyridines). Thus,
a
protonation of the amide residue in the derived phosphoramidates (cf. Fig. 1) should be
attenuated, and this may in turn lead to an increase in population of neutral protonated forms,
which are expected to be particularly effective in crossing the lipid membranes. The third
group, 2-aminothiazoles (Table 1, entries 14–16), was chosen to verify whether
phosphoramidates containing significantly different heteroaromatic group can be synthesized
under the same conditions and if they retain the antiviral properties of (N-
pyridyl)phosphoramidates. It is worth to note that the thiazole motif is present in many
*
Experimental values, included in reports generated by ChemBioDraw v. 13.
3

ACCEPTED MANUSCRIPT
molecules showing biological activity and successful syntheses of the proposed here model
compounds may be a starting point for development of new bioactive derivatives.
Chemistry
The phosphoramidates of type 7 were prepared according to modified published
protocols, i.e. by condensation of nucleoside 5'-H-phosphonate with an amine followed by
oxidation (method A, Fig. 2) or by phosphorylation of nucleoside with bis(triazoyl)-(N-
3
1
heteroaryl)phosphoramidates (method B, Fig. 4) [5]. All reactions were monitored by
P
NMR spectroscopy.
FIGURE 2
In the method A, nucleoside H-phosphonates 1 [12] were condensed with heteroaryl
amines of type 2 (1.1 equiv.) using diphenyl chlorophosphate (DPCP; 1.7 equiv.) as a
condensing agent to yield H-phosphonamidates 3, which were oxidized in situ to the final
phosphoramidates 7 (Fig. 2). Since in same cases lower yields of the products were obtained,
3
1
we inspected more carefully the course of the reactions using P NMR spectroscopy. In the
standard procedure, the addition of aqueous iodine in pyridine to the solution the in situ
3
1
formed mixture of 3 + 4, resulted in relatively rapid (ca. 5 min) disappearance of P NMR
signal of H-phosphonamidate 3 at ca 7–9 ppm with a simultaneous built-up of the signal of
the final product 7 (δ ca. -3 ppm). Additionally, regeneration of some amounts of H-
P
phosphonate 1 was observed for several amines 2, which were more basic than pyridine. Most
likely, this was a result of hydrolysis of 3, which in these cases was rapid enough to compete
with relatively sluggish iodination of H-phosphonamidates. Under the above conditions no
†
intermediates could be observed in the reaction mixture. However, when iodination of the
mixture of 3 + 4 was carried out under anhydrous conditions, formation of two groups of
31
signals in a 1:1 ratio was observed in the P NMR spectra: a multiplet at ca. -7 ppm and two
doublets at ca. -24 ppm. The reaction was complete within ca. 5 min, as judged from
disappearance of the intermediate 3. Also, a decrease in intensity of the signal at ca. -10 ppm,
assigned to diphenyl phosphate 4, was observed. Upon addition of an excess of water, both
3
1
the new signals disappeared simultaneously within ca. 5 min, and in the final P NMR
spectrum only signals of 7 and 4 were present.
†
Phosphoroiodidate diesters are extremely reactive, particularly in the presence of pyridine (an efficient
nucleophilic catalyst for this reaction) [13,14], and the failure to detect signals of 5 (or the derived
+
31
phosphopyridinium P-Py species [14], not shown in Fig. 2) in P NMR spectra was not surprising.
4

ACCEPTED MANUSCRIPT
The most plausible explanation of these results is formation of a mixed pyrophosphate 6
two diastereomers), for which the multiplet at ca. -7 ppm was assigned to the amidate ester
(
phosphorus atom [-P(O)(OR)(NHAr)], while the two doublets at ca. -24 ppm, to the diester
one [-P(O)(OPh) ]. This was confirmed by spiking the reaction mixture containing the
2
assumed mixed pyrophosphate 6 with a sample of this compound obtained in the reaction of
‡
(N-heteroaryl)phosphoramidate with DPCP.
These results prompted us to carry out the oxidation process in two steps: (i) oxidation
of H-phosphonamidate 3 using anhydrous iodine in pyridine to generate pyrophosphate 6,
followed by (ii) its hydrolysis with an excess of water. Both steps proceeded with similar rate
3
1
and were completed within < 5 min yielding the desired product 7 quantitatively ( P NMR).
The above procedure was used successfully for the synthesis of most AZT (N-
heteroaryl)phosphoramidates 7ax (70–80% isolated yields; Table 1). A wide scope of this
approach was demonstrated by preparation of several representative derivatives of ddU, d4T,
ABC, and 3TC nucleosides (7b–e, Table 2). Synthesis of H-phosphonates 1b-e [12] and their
subsequent coupling with aromatic amines 2, followed by oxidation (vide supra) proceeded
3
1
nearly quantitatively ( P NMR); however, due to similar R 's of the products and other
f
compounds present in the reaction mixtures, in several cases a repeated chromatography was
required
heteroaryl)phosphoramidates. This led to 10–15% decrease in the yields (Table 2, entries 18,
0, 21, 25, 26, 28, & 29). Conversely, phosphoramidate 7cg (d4T – 6AQ; entry 24) was
to
obtain
satisfactory purity
of
the
desired
nucleoside
(N-
2
prepared with an excellent isolated yield of 97%, confirming the effectiveness and high
synthetic potential of the chemistry used.
FIGURE 3
Similarly as it was observed previously for 4APy (2a) [5], condensations of AZT H-
phosphonate 1a with more basic amines led to poor results (Table 1, entries 6 & 10),
presumably due to coincidence of several factors. The first was an undesired superfluous
activation of the initially formed H-phosphonamidate esters 3 (Fig. 3) with an excess of a
condensing agent (diphenyl chlorophosphate, DPCP) that resulted in formation of nucleoside
bis(N-heteroaryl)phosphordiamidites 8 (δ ca. 95 ppm). Since such by-products were not
P
‡
Attempted isolation of pyrophosphates of type 6 failed due to their rather high reactivity. In separate
experiments we found that such pyrophosphates reacted with amines and alcohols with preferential nucleophilic
attack at the phosphoramidate phosphorus atom. Unfortunately, this regioselectivity was not complete (ca. 75%).
Thus, potential applications of pyrophosphates 6 for synthesis of phosphoramidate diesters or phosphordiamidate
monoesters require further studies.
5

ACCEPTED MANUSCRIPT
observed for amines with pK < 7, it is plausible that in the presence of more basic amines the
a
abundance of tervalent tautomers of type 3' became significant, and their reaction with a
second molecule of DPCP and the amine yielded the putative diamides 8. Upon their
oxidation, the respective bis(N-heteroaryl)phosphordiamidates 9 (δ ca. -5 ppm) were formed
P
in amounts up to 20%. Apart from decreasing the yield of conversion of 1 to 7, their presence
§
hampered the chromatographic purification of the desired monoamide products.
FIGURE 4
An additional problem associated with the more basic heteroaryl amines was their rapid
reaction with the condensing agents, e.g., DPCP or others, while such side-reaction was
negligible for amines with pK < 7. In consequence, both the consumed amines and DPCP had
a
to be replenished in several portions during the reaction. This in turn generated large amounts
of HCl, which reduced nucleophilicity of the amines by protonation and led to their massive
precipitation from the reaction mixtures in a form of hydrochlorides.
TABLE 1
TABLE 2
Thus, for incorporation of amines with pK > 7 into phosphoramidates 7, an approach
a
based on P(V) derivatives was developed (method B, Fig. 4). Thus,
tris(triazoyl)phosphoramidate 10 [15] was reacted with an amine (2) to yield monosubstituted
amides of type 11. Similarly as the previously described di(triazoyl)[N-(pyridin-4-
yl)]phosphoramidate 11a [5], also derivatives of 2APy (11b), 4A2MQ (11e), 1AIQ (11i), and
2
AQ (11k) precipitated from the reaction mixtures and could be readily isolated as pure
products (the last were unstable during storage and should be prepared directly before use). In
contrast, derivatives of other, less basic heteroarylamines remained in solution and attempts to
use them directly for the next reaction steps gave poor results. Thus, the methods A and B
complement each other with a borderline of pK ≈ 7 of the amine used.
a
To produce phosphoramidates 7 containing the more basic amines, the corresponding
bistriazolides 11b, 11i, and 11k were dissolved in pyridine (at elevated temperature, ~75 ºC, if
necessary) and reacted with the added AZT to yield, after hydrolysis, products 7ax (Fig. 4) in
§
AZT bis[N-(pyrid-4-yl)]phosphordiamidate 9aa appeared to be very stable under neutral and basic conditions.
The half-life time in 0.1 M NaOH was ca. 10 days, and in 10% aqueous pyridine + 10% triethylamine, ca 5 days.
In contrast, in 0.1 M HCl the hydrolysis was very rapid (< 3 min). In all instances, monoamidate 7aa was formed
as the final product. Initial experiments revealed high anti-HIV activity of diamidate 9aa (EC 30 nM; EC 400
50
90
nM) and lack of cytotoxicity at concentrations up to 200 µM. Further studies on this type of compounds are in
progress in this laboratory.
6

ACCEPTED MANUSCRIPT
ca. 50–60% isolated yields. For 11e, concomitant formation of bis(AZT) phosphate diester
was observed and the yield of desired product 7ae was low, ca. 20% after chromatographic
purification.
Similarly as we noted previously for AZT (N-pyridyl)phosphoramidates [5], during
purification of all synthesized phosphoramidates of type 7 in a form of triethylammonium
salts, a significant loss of the cation took place and the products were obtained partly or
completely in a form of acids. This is a serious issue since non-stoichiometric and variable
amounts of the cation compromise the accuracy of sample preparation, while under-
stoichiometric counter-ion contents causes solubility problems and affects the chemical
stability of phosphoramidates (acid-catalyzed degradation). Thus, to obtain the products with
a 1:1 anion:cation ratio, a new purification procedure was developed (see Experimental Part).
The issue of the cation loss during purification is a subject of a separate publication [21].
Stability of phosphoramidates 7
The phosphoramidates of type 7 were examined with respect to their chemical and
enzymatic stability. All of them appeared to be intact during incubation for 5 days at 37 ºC in
3
1
RPMI buffer (HPLC & P NMR analysis). A pilot experiment revealed outstandingly high
stability of phosphoramidate 7aa (AZT – 4APy) in 0.1 M HCl and 0.1 M NaOH (no
3
1
hydrolysis of the phosphoramidate group within 7 days, P NMR). For such high chemical
resistance one can expect a very long shelf-life of this type of compounds. In RPMI + 10%
fetal bovine serum (FBS) a rather slow decomposition was observed (Fig. 5, Tables 1 & 2),
the half-life time was usually several days; however, derivatives of ddU (7b) were in general
less stable. Most of phosphoramidates of d4T (7c), ABC (7d), and 3TC (7e) appeared to be
completely resistant to degradation by the FBS solution. In order to compare stability of
compounds with t1/2 > 7 days, the amount of phosphoramidates 7 which remained after 3 days
of incubation in 10% FBS/RPMI was determined (Tables 1 & 2).
FIGURE 5
The final products of hydrolysis of the compounds 7a–c in FBS solution were identified
by RP HPLC as the respective nucleosides and amines of type 2. Unfortunately, in some cases
a separation of all compounds present in reaction mixtures could not be achieved, e.g., for
phosphoramidate 7am the formed AZTMP and 2-aminothiazole 2m co-migrated and the
analysis of the products was hampered (Fig. 5A). For several phosphoramidates (e.g., 7bo,
Fig. 5B), however, the peaks for all compounds were well resolved and their assignment was
confirmed by spiking with original samples. This provided a clear-cut evidence for formation
7

ACCEPTED MANUSCRIPT
of the anticipated nucleoside monophosphates, which, rather surprisingly, remained in a
steady concentration for many days (ca. 15% of A.U. of all species absorbing at 254 nm),
presumably as a result of similar rate of formation and degradation. Provided that an
analogous steady-state type of kinetics is kept in vivo, such continuous delivery of NMTP's by
gradual decomposition of phosphoramidates 7 would comply with a pro-nucleotide scenario,
in which phosphoramidates in the cells are first hydrolyzed – at least partly – to
monophosphate esters that serve subsequently as substrates for nucleotidyl kinases.
Peaks which could be assigned to phosphoramidates of heteroarylamines (ArNHPO H )
3
2
that hypothetically may be formed by cleavage of the P-O bond in compounds 7, were not
detected in neither case, indicating that such pathway of decomposition is probably of minor,
if any, significance.
It must be stressed that the formula of FBS-containing media is significantly different
from the composition of fluids present inside cells and the above experiments should be
treated just as tentative ones to get some insight into processes occurring in living cells.
Biological evaluation
The biological goal for the preparation of (N-heteroaryl)phosphoramidates of type 7
was their evaluation as anti-HIV agents. As a first step, we examined the parent
heteroaromatic amines of type 2, which likely are the catabolic products of the studied
phosphoramidates, and found that these were of low or very low cytotoxicity (Table 1), and
did not show any anti-HIV activity up to concentration of 10 µM. Thus, their biological
activity was assumed to be negligible. In contrast, all AZT (N-heteroaryl)phosphoramidates of
type 7a were found to be potent anti-HIV agents even in nanomolar concentrations. EC50
values were in the range of 1–20 nM and EC90 values, ca. 10–200 nM (Table 1). Among
them, the most promising compounds (EC90 < 25 nM) contained isoquinoline (7ai & 7aj),
cyanopyridine (7al & 7am), and thiazoline amide groups (7an, 7ao & 7ap), while those with
quinoline group (7ad–7ah) were in general slightly less active (EC = 30–220 nM).
9
0
All the tested phosphoramidates were practically non-toxic for CEM-T4 cells, even at
the highest concentrations, which were limited only by solubility of the compounds. Thus, CC
indices could not be determined and the values of selectivity indices may be only presumed.
Apparently, these are very high, in many cases likely at the level of hundreds of thousands or
more, making AZT (N-heteroaryl)phosphoramidates of type 7a valuable candidates for further
development as anti-HIV drugs.
8

ACCEPTED MANUSCRIPT
As it was mentioned in the introduction, the working hypothesis, which governed the
choice of the heteroaryl groups for phosphoramidates of type 7, was that increased
lipophilicity should facilitate cell membrane permeability and in consequence, should lead to
superior pharmacological properties. To evaluate correctness of these assumptions, water-
octanol partition coefficients of the compounds studied were determined (micro shake-flask
partitioning approach [22]). Initial experiments revealed that octanol-water partition of
phosphoramidates of type 7 was dependent on pH to a small extent only, e.g., log P7ad = -1.45
at pH 4; -1.72 at pH 7.4; -1.40 at pH 11. Thus, it was assumed that log P ≈ log D and further
measurements were done at the physiological pH 7.4 only. The values of log P determined are
included in Tables 1 and 2, and deserve some comments. (i) The influence of lipophilicity of
the nucleoside residues on the total lipophilicity of phosphoramidates 7 appeared to be rather
limited. Some trend of lower log P's for derivatives of hydrophilic ddU [log PddU = -0.96 (lit. -
1
.0 [23]), log P7bx = -1.7 – -2.4], d4T [log Pd4T = -0.79 (lit. -0.6 [24,25]), log P7cx = -2.0 – -
.4], and 3TC [log P3TC = -0.94 (lit. -0.93 [26]) , log P7eo = -2.3] than those for AZT [which is
3
equally lipo- and hydrophilic; log PAZT = 0.06 (lit. -0.01 – 0.1 [27-29]), log P7ax = -0.7 – -2.1]
could be noted. However, similarly low log P's were found for phosphoramidates of much
more lipophilic ABC [log PABC = 1.16 (lit. 0.85 [26]), log P7dx = -1.6 – -1.9]. (ii) There is no
apparent relationship between lipophilicity of the free heteroarylamines and AZT
phosphoramidates 7ax containing them (Table 1). (iii) There are large discrepancies between
experimental and calculated log P values for compounds 7a (Table 1). These three
observations indicate that the common additive methods for predicting log P values is not
applicable for phosphoramidates of type 7, possibly due to acid-base equilibria, electronic
effects, and functional group interactions (intra- or intermolecular) etc., which apparently
control the overall lipophilicity for this class of compounds.
Thus, it was not surprising to see lack of correlation between lipophilicities and
antiviral potency of compounds 7 (Table 1 and 2, Fig. 6). For example, in the AZT series
(7ax), similarly very low EC values were found for the least hydrophilic 7ai (log P = -0.7,
EC = 2.0 nM) and the most hydrophilic 7an and 7ao (log P = -2.1, EC = 2.1 nM), while
50
50
the least antiviral effective 7ah (EC = 20 nM) has an intermediate log P value of -1.5.
5
0
We also did not observe any correlation between anti-HIV activity of amides 7 and
their stability in RPMI-FBS or basicity of the parent amines 2 (which should govern the
isoelectric point of amides 7, and consequently, their cell membrane permeability at given
pH).
9

ACCEPTED MANUSCRIPT
Most likely, these results reflect a complex nature of anti-HIV mode of action of the
compounds studied, including involvement in acid-base equilibria, the aforementioned
different cell membrane permeability, stability inside the cells, and enzyme substrate
properties.
FIGURE 6
In a next set of experiments, phosphoramidates 7b-e bearing other dideoxynucleoside
residues (ddU, d4T, ABC, and 3TC) were tested as anti-HIV agents (Table 2). d4T and ABC
nucleosides are significantly less active than AZT, and this was true for their
phosphoramidates 7c and 7d in comparison to 7a. Also, similarly to the AZT series, the best
phosphoramidates in 7c and 7d series were at a similar level of anti-HIV activity as their
parent nucleosides, while 3TC phosphoramidate (7eo) was ca. 2 orders of magnitude less
active than 3TC. Nevertheless, all the studied phosphoramidates of type 7c–e showed very
low toxicity for cells, and this makes them promising compounds for further evaluation as
antiviral agents.
Phosphoramidates in the 7b series are a special case, since their parent nucleoside,
dideoxyuridine (ddU), is biologically inert since is not phosphorylated by kinases to ddUMP.
Moreover, even when ddU was delivered to the cells in a form of masked mono- and
diphosphates, these were poorly phosphorylated to ddUTP (which is a potent HIV RT
inhibitor) [30]. Previously, we reported on inhibition of HIV by ddU [N-(pyridin-4-
yl)]phosphoramidate (EC 0.2 µM) but not by analogous 2- and 3-aminopyridine derivatives
5
0
[
5]. In this work, among the studied compounds of type 7b, a moderate activity (EC 4.2 µM)
50
**
was found for [N-(quinolin-3-yl)]phosphoramidate 7bd only. Degradation of all derivatives
of type 7b in the FBS medium proceeded in similar rates (Table 2) toward ddU via ddUMP (a
clear-cut formation of ddUMP was observed by HPLC, cf. Fig. 5B). It was rather unlikely that
in an anti-HIV assay the formation of ddUMP was significantly more effective for 7bd than
for 7bg, 7bl, or 7bo. To explain the observed activity of some ddU phosphoramidates and
differences within this group, one may invoke recent results by Herdevijn et al., who observed
that some nucleoside 5'-(N-amino acid)phosphoramidates could mimic dNTP's in nucleic acid
biosynthesis catalyzed by HIV RT and other polymerases, and that propensity of a
phosphoramidate to be a substrate in such reaction depended on the structure of AA residue
[
9-11]. Provided that similar mode of action applies for (N-heteroaryl)phosphoramidates of
*
*
EC50 values of 0.2 [5] and 4.2 µM are still much better than these of EC50 = 15 and 26 µM, found as the best
results for classical pro-nucleotides of ddUMP [31] and ddUDP [30], respectively.
1
0

ACCEPTED MANUSCRIPT
type 7, 3-aminoquinoline derivative 7bd itself could serve as substrate for incorporation into
DNA by HIV RT compromising further chain growth, while the other amides of type 7b,
would be worse substrates.
This hypothesis may explain as well why some exceptionally stable phosphoramidates
of type 7 (e.g. 7cd, 7dl, 7eo) are able to inhibit proliferation of HIV despite their complete
resistance to degradation by FBS, which may suggest a very low level of the respective
nucleosides or NMP's produced inside the cells. These assumptions still require experimental
verification.
Conclusions
In the chemical part of this contribution a number of new nucleoside (N-
heteroaryl)phosphoramidates were prepared in good yields. Mechanistic investigations
revealed that in situ oxidation of nucleoside H-phosphonamidates involved formation of
mixed pyrophosphoramidates as intermediates. This finding led to development of an
improved synthetic protocol, in which oxidation and hydrolysis were carried out as one-pot
two-stage process. Additionally, side products formed in reactions in which highly basic
amines were used, were identified as so far unknown nucleoside di(N-
heteroaryl)phosphorodiamidates, which are potentially valuable as a new type of antiviral
agents.
The obtained compounds were not harmful for CEM-T4 cells, while were able to
suppress HIV-1 proliferation in nanomolar (AZT derivatives) or micromolar (ddU, d4T,
ABC, and 3TC derivatives) concentrations.
There are no clear-cut evidences how 2',3'-dideoxynucleoside (N-
heteroaryl)phosphoramidates of type 7 are incorporated into DNA. In a pro-nucleoside
scenario, they are dephosphorylated in the cells, and can be considered as a kind of storage for
gradual delivery of antiviral nucleoside analogues. However, some results (anti-HIV activity
of ddU derivatives) suggest more desirable pro-nucleotide mode of action, i.e. cleavage of the
P-N bond with formation of phosphorylated nucleosides, and by-passing the first
phosphorylation stage. Alternatively, phosphoramidates 7 may be recruited as direct
substrates for HIV reverse transcriptase. Anti-HIV activity of several amides that are fully
resistant to degradation by FBS may point to the last mechanism.
It is intriguing that irrespective of significantly different type of the heteroaromatic
residues, which resulted in variations in enzymatic stability, lipophilicity, and (presumably)
1
1

ACCEPTED MANUSCRIPT
ionic equilibria, all AZT derivatives studied showed rather similar level of antiviral activity
(
EC ≈ 1–20 nM; EC ≈ 10–200 nM).
5
0
90
Concluding, efficient synthesis, stability, solubility, high anti-HIV activity, and very
low toxicity found for the phosphoramidates studied is calling for further studies on this class
of compounds as potential antiviral drugs.
Acknowledgments Financial support form the National Science Centre of Poland, Projects
No. 2011/01/B/ST5/06414 & 2011/03/B/ST5/03102, is greatly acknowledged.
Experimental
Material and methods
1
13
31
H, C, and P NMR spectra were recorded on Varian Unity BB VT 300 MHz or Bruker
Avance II 400 MHz machines. Mass spectra were recorded with the ESI technique with
negative ionization with accuracy below 5 ppm. Amount of water in anhydrous solvents was
controlled using Karl Fischer coulometric titration (Metrohm 684 KF coulometer). The
3
1
31
progress of reactions was monitored by P NMR spectroscopy. P NMR data are collected in
Table 3. HPLC analyses were performed on a Nucleosil 100-5C18 column (5.0 ꢀm, 4.6 mm x
2
5
50 mm) using Shimadzu Prominence UFLC and Waters Breeze HPLC systems and 0 →
0%/ 30 min gradient of A+B solvent systems (A, 0.01 M aqueous triethylammonium acetate;
B, A + acetonitrile, 1 : 4, v/v) at 36 ºC, flow rate 1.0 mL/min. Biological experiments were
carried out as described previously [5].
2
’,3’-Dideoxynucleoside H-phosphonates 1a-e
AZT H-phosphonate 1a was obtained as described previously [32]. H-Phosphonates of ddU
1b), d4T (1c), ABC (1d), and 3TC (1e) were prepared by a slightly modified diphenyl H-
(
phosphonate procedure [12]. A nucleoside (1 mmol) was rendered anhydrous by evaporation
with dry pyridine and dissolved in 5 mL of the same solvent. To this solution diphenyl H-
phosphonate (3 equiv.) was added while stirring and left for 0.5 hour. The reaction was
quenched by addition of water (1 mL) and triethylamine (TEA, 1 mL) and the solution
concentrated to a viscous oil. The residue was dissolved in minimal amount of water and
without work-up applied to silica-gel chromatographic column, eluted subsequently with 0 →
3
0% (v/v) MeOH gradient in 99:1 ethyl acetate – TEA. Fractions containing pure product
were combined, evaporated and lyophilized from benzene.
1
2

ACCEPTED MANUSCRIPT
+
3
'-Azido-3'-dideoxythymidin-5'-yl H-phosphonate, TEAH salt (AZT H-phosphonate,
+
1
a) and 2',3'-dideoxyuridin-5'-yl H-phosphonate, TEAH salt (ddU H-phosphonate 1b).
Analytical data were in agreement with the literature [32].
+
2
',3'-Didehydro-3'-deoxythymidin-5'-yl H-phosphonate, TEAH
salt (d4T H-
1
phosphonate; 1c). Yield 72%. RP HPLC R 12.78 min. H NMR (400 MHz) δ (D O) 1.21
f
H
2
(9H, t, J = 7.3 Hz), 1.89 (3H, d, J = 1.0 Hz), 3.21 (6H, q, J = 7.3 Hz), 4.07 (2H, m), 5.09 (1H,
s), 5.98 (1H, m), 6.49 (1H, m), 6.69 (1H, d, J = 637.4 Hz), 6.95 (1H, m), 7.60 (1H, d, J = 1.2
1
3
Hz). C NMR (100 MHz) δ (D O) 7.7, 11, 46.2, 63.4 (d, J = 4.5 Hz), 85.4 (d, JPC = 7.5
c
2
PC
Hz), 89.5, 110.7, 124.9, 133.8, 137.7, 151.7, 166.2. HRMS m/z 287.1864, calcd. for
-
C H N O P 287.0438.
10
12
2
6
(1S,cis)-4-[2-Amino-6-(cyclopropylamino)-9H-purin-9-yl]-2-cyclopentene-1-methyl H-
+
phosphonate, TEAH salt (ABC H-phosphonate; 1d). Yield 47%. RP HPLC R 17.67 min.
f
1
H NMR (400 MHz) δ (D O) 0.49 (2H, m), 0.76 (2H, m), 1.17 (9H, t, J = 7.3 Hz), 1.42 (1H,
H
2
m), 2.66 (2H, m), 3.06 (6H, q, J =7.3 Hz), 3.76 (2H, m), 5.22 (1H, t, J = 6.2 Hz), 5.77 (1H,
1
3
m), 6.10 (1H, m), 6.61 (1H, d, J = 630.8 Hz), 7.63 (1H, s). C NMR (100 MHz) δ (D O) 6.2,
c
2
7
.6, 22.8, 33.8, 45.1 (d, JPC = 7.0 Hz), 46, 58.6, 65.4 (d, JPC = 4.6 Hz), 112.6, 128.9, 136.4,
-
1
37.4, 155.1, 159. HRMS m/z 349.1162, calcd. for C H N O P 349.1183.
1
4
18
6
3
+
α-L
-2',3'-Dideoxy-3'-thiacytadin-5'-yl H-phosphonate, TEAH
salt (3TC H-
1
phosphonate; 1e). Yield 93%. RP HPLC R 12.03 min. H NMR (400 MHz) δ (D O) 1.30
f
H
2
(
9H, t, J = 7.3 Hz), 3.21 (6H, q, J = 7.3 Hz), 3.26 (1H, m), 3.58 (1H, dd, J = 12.3 Hz), 4.18
1H, m), 4.31 (1H, m), 5.47 (1H, t, J = 7.2 Hz), 6.09 (1H, d, J = 7.6 Hz), 6.36 (1H, t, J = 9.9
(
1
3
Hz), 8.06 (1H, d, J = 7.6 Hz), 6.82 (1H, d, J = 642.4 Hz). C NMR (100 MHz) δ (D O) 7.7,
c
2
3
6.2, 46.2, 63.6 (d, JPC = 4.2 Hz), 83.9 (d, JPC = 7.2 Hz), 86.9, 95.4, 141.5, 156, 165.2. HRMS
-
m/z 292.0085, calcd. for C H N O PS 292.0162.
8
11
3
5
2
’,3’-Dideoxynucleoside (N-heteroaryl)phosphoramidates 7
Method A
All phosphoramidates containing amines of pK < 7 were prepared from H-phosphonates of
a
type 1 by a slightly modified literature methods [5,21]. Nucleoside H-phosphonate 1 (1 mmol)
and heteroaryl amine 2d, 2f–i, or 2j–p (1.1 mmol) were rendered anhydrous by the
evaporation of the added pyridine (3 x 10 mL) and then dissolved in 10 mL of DCM
containing 10% (v/v) of pyridine. To this, DPCP (1.7 mmol) was added, and the reaction
3
1
mixture was left for 5 min yielding H-phosphonamidates 3 (for P NMR data, see Table 1).
Without isolation of these intermediates, a solution of iodine (2 mmol) in pyridine (1 mL) was
1
3

ACCEPTED MANUSCRIPT
added, the reaction mixture was left for 5 min and then, a solution of water (50 mmol) in
pyridine (1 mL) was added. After another 5 min, the excess of iodine was decomposed with
the added ethanethiol, and the mixture was concentrated to an oil under reduced pressure.
Crude phosphoramidate 7 was dissolved in a minimum amount of MeOH, applied to a silica-
gel column, and eluted with 0–20% gradient of MeOH in ethyl acetate containing 5% TEA.
Fractions containing the pure product were pooled and evaporated to dryness under reduced
pressure. The residue was dissolved in a minimal amount of water containing ca. 2 equiv. (in
respect to phosphoramidate) of TEA and lyophilized. Such final procedure secured a
+
1
stoichiometric amount of the TEAH cation (1 equiv.) [21], which was confirmed by H NMR
spectroscopy.
+
3
'-Azido-3'-deoxythymidin-5'-yl [N-(quinolin-3-yl)]phosphoramidate, TEAH salt (7ad).
1
Yield 78%. RP HPLC R 20.32 min. H NMR (400 MHz) δ (D O) 1.21 (9H, t, J = 7.3 Hz),
f
H
2
1
6
7
3
.45 (3H, s), 2.27 (2H, m), 3.10 (6H, q, J = 7.3 Hz), 3.99 (1H, m), 4.08 (1H, m), 4.18 (2H, m),
.01 (1H, t, J = 5.6 Hz), 6.89 (1H, m), 7.45 (2H, m), 7.56 (1H, m), 7.69 (1H, d, J = 2.6 Hz),
1
3
.75 (1H, d, J = 8.3 Hz), 8.50 (1H, d, J = 2.6 Hz). C NMR (100 MHz) δ (D O) 7.7, 11.1,
c
2
5.3, 46, 59.2, 64.5 (d, JPC = 4.7 Hz), 81.9 (d, JPC = 9.6 Hz), 84, 110.5, 119.2 (d, JPC = 3.5
Hz), 125.9, 126.4, 126.6, 126.8, 127.8, 135.3, 135.9, 140.9, 143.1 (d, JPC = 10.5 Hz), 151.5,
-
1
66.2. HRMS m/z 472.1258, calcd. for C H N O P 472.1140.
1
9
19
7
6
+
3
'-Azido-3'-deoxythymidin-5'-yl [N-(quinolin-5-yl)]phosphoramidate, TEAH salt (7af).
1
Yield 77%. RP HPLC R 19.26 min. H NMR (400 MHz) δ (D O) 1.15 (9H, t, J = 7.3 Hz),
f
H
2
1
5
=
.51 (3H, s), 2.18 (2H, m), 3.01 (6H, q, J = 7.3 Hz), 3.96 (2H, m), 4.03 (1H, m), 4.15 (1H, m),
.97 (1H, t, J = 6.6 Hz), 7.22 (1H, s), 7.31 (1H, m), 7.38 (1H, m), 7.43 (2H, m), 8.44 (1H, d J
1
3
8.4 Hz), 8.63 (1H, m). C NMR (100 MHz) δ (D O) 7.7, 10.8, 35.3, 46.1, 59.8, 64.2 (d,
c
2
JPC = 4.9 Hz), 82.4 (d, JPC = 9 Hz), 84.6, 110.4, 113.6 (d, JPC = 2 Hz), 119.7, 119.9, 120,
1
29.5, 130.8, 136.6, 137.2, 146.7, 149.3, 150.9, 165.6. HRMS m/z 472.1172, calcd. for
-
C H N O P 472.1140.
19
19
7
6
+
3
'-Azido-3'-deoxythymidin-5'-yl [N-(quinolin-6-yl)]phosphoramidate, TEAH salt (7ag).
1
Yield 73%. RP HPLC R 19.14 min. H NMR (400 MHz) δ (D O) 1.14 (9H, t, J = 7.3 Hz),
f
H
2
1.51 (3H, s), 2.18 (2H, m), 3.01 (6H, q, J = 7.32), 3.94 (2H, m), 4.06 (1H, m), 4.13 (1H, m),
5.98 (1H, t, J = 6.5 Hz), 7.07 (1H, m), 7.23 (1H, dd, J = 4.4, 8.4 Hz), 7.26 (1H, d, J = 2.3 Hz),
7.40 (1H, dd, J = 2.4, 9.1 Hz), 7.70 (1H, d, J = 9.1 Hz), 7.87 (1H, d, J = 8.2 Hz), 8.43 (1H,
1
3
m). C NMR (100 MHz) δ (D O) 7.7, 10.9, 35.2, 46.2, 59.5, 64 (d, J = 5 Hz), 82.3 (d, JPC
c
2
PC
=
9.9 Hz), 84.6, 110.3, 110.5, 121.2, 123.7 (d, JPC = 9.9 Hz), 126.4, 128, 128.6, 136.5, 136.7,
-
1
40.3, 145.7, 150.5, 164.9. HRMS m/z 472.1131, calcd. for C H N O P 472.1140.
1
9
19
7
6
1
4

ACCEPTED MANUSCRIPT
+
3
'-Azido-3'-deoxythymidin-5'-yl [N-(quinolin-8-yl)]phosphoramidate, TEAH salt (7ah).
1
Yield 69%. RP HPLC R 22.35 min. H NMR (400 MHz) δ (D O) 1.28 (6H, t, J = 7.3 Hz),
f
H
2
1
.46 (3H, s), 2.37 (2H, m), 3.20 (6H, q, J = 7.3 Hz), 4.06 (1H, m), 4.16 (2H, m), 4.31 (1H, m),
.03 (1H, t, J = 6.5 Hz), 7.24 (1H, s), 7.37 (1H, m), 7.41 (2H, m), 7.53 (1H, dd, J = 4.3, 8.2
6
1
3
Hz), 8.26 (1H, m), 8.76 (1H, m). C NMR (100 MHz) δ (D O) 7.7, 10.7, 35.2, 46.2, 59.7,
c
2
6
9
4.5 (d, JPC = 5 Hz), 82.5 (d, JPC = 8.9 Hz), 84.6, 110.3, 110.5, 110.6, 121.2, 123.7 (d, JPC =
.8 Hz), 126.4, 128, 128.6, 136.5, 136.7, 140.3, 145.7, 150.5. HRMS m/z 472.1138, calcd. for
-
C H N O P 472.1140.
19
19
7
6
+
3
'-Azido-3'-deoxythymidin-5'-yl [N-(isoquinolin-5-yl)]phosphoramidate, TEAH salt
1
(
7aj). Yield 69%. RP HPLC R 20.07 min. H NMR (400 MHz) δ (D O) 1.26 (6H, t, J = 7.3
f
H
2
Hz), 1.45 (3H, s), 2.34 (2H, m), 3.17 (6H, q, J = 7.3 Hz), 4.03 (1H, m), 4.12 (2H, m), 4.27
(1H, m), 6.00 (1H, t, J = 6.5 Hz), 7.23 (1H, s), 7.44 (2H, m), 7.73 (1H, m), 7.92 (1H, d, J =
1
3
6
4
1
.2 Hz), 8.35 (1H, d, J = 6.2 Hz), 9.04 (1H, s). C NMR (100 MHz) δ (D O) 7.7, 10.7, 35.1,
c 2
6.1, 59.8, 64.4 (d, JPC = 4.7 Hz), 82.4 (d, JPC = 9 Hz), 84.6, 110.4, 114.7, 117.4, 120, 127.3,
28. 128.3, 136, 136.6, 139.7, 150.6, 151.3, 165.1. HRMS m/z 472.1040, calcd. for
-
C H N O P 472.1140.
19
19
7
6
+
3
'-Azido-3'-deoxythymidin-5'-yl [N-(5-cyanopyridin-2-yl)]phosphoramidate, TEAH salt
1
(
7al). Yield 74%. RP HPLC R 18.84 min. H NMR (400 MHz) δ (D O) 1.29 (9H, t, J = 7.3
f
H
2
Hz), 1.85 (3H, d, J = 1 Hz), 2.51 (2H, m), 3.21 (6H, q, J = 7.3 Hz), 4.07 (1H, m), 4.12 (1H,
m), 4.18 (1H, m), 4.40 (1H, m), 6.13 (1H, t, J = 6.5 Hz), 7.06 (1H, d J = 8.8 Hz), 7.50 (1H, d,
1
3
J = 1.2 Hz), 7.85 (1H, dd, J = 2.3, 8.9 Hz), 8.41 (1H, d, J=2.1 Hz). C NMR (100 MHz) δ
c
(
D O) 7.7, 11.1, 35.4, 46.2, 59.6 (d, JPC = 5 Hz), 64.4 (d, JPC = 9.3 Hz), 82.3, 84.8, 99.2,
2
1
4
3
10.7, 117.8, 128, 137.2, 140.4, 150.9, 151.7, 157.1 (d, JPC = 5.Hz), 165.7. HRMS m/z
-
47.0983, calcd. for C H N O P 447.0935.
1
6
16
8
6
+
'-Azido-3'-deoxythymidin-5'-yl [N-(6-cyanopyridin-3-yl)]phosphoramidate, TEAH salt
1
(
7am). Yield 76%. RP HPLC R 18.59 min. H NMR (400 MHz) δ (D O) 1.30 (9H, t, J = 7.3
f
H
2
Hz), 1.84 (3H, s), 2.51 (2H, m), 3.22 (6H, q, J = 7.3 Hz), 4.03 (1H, m), 4.14 (2H, m), 4.36
(1H, m), 6.12 (1H, t, J = 6.5 Hz), 7.39 (1H, s), 7.48 (1H, dd, J = 2.6, 8.6 Hz), 7.67 (1H, d, J =
1
3
6
.5 Hz), 8.27 (1H, d, J = 2.5 Hz). C NMR (100 MHz) δ (D O) 8.2, 11.8, 35.8, 46.6, 64.8 (d,
c 2
JPC = 4.8 Hz), 82.4 (d, JPC = 9.3 Hz), 85.31, 111.2, 117.8, 121.3, 123.4 (d, JPC = 5.6 Hz),
1
29.7, 137.3, 139.9 (d, JPC = 8.6 Hz), 143.1, 152.9, 168.2. HRMS m/z 447.0845, calcd. for
-
C H N O P 447.0935.
16
16
8
6
+
3
'-Azido-3'-deoxythymidin-5'-yl [N-(thiazol-2-yl)]phosphoramidate, TEAH salt (7an).
1
Yield 72%. RP HPLC R 17.46 min. H NMR (400 MHz) δ (D O) 1.28 (9H, t, J = 7.3 Hz),
f
H
2
1
5

ACCEPTED MANUSCRIPT
1
.88 (3H, d, J = 0.8 Hz), 2.49 (2H, m), 3.20 (6H, q, J = 7.3 Hz), 4.06 (1H, m), 4.13 (2H, m),
.40 (1H, q, J = 4.8 Hz), 6.18 (1H, t, J = 6.5 Hz), 6.84 (1H, d, J = 3.8 Hz), 7.16 (1H, d, J = 3.8
4
1
3
Hz), 7.63 (1H, d, J = 1.1 Hz). C NMR (100 MHz) 8.2, 10.8, 11.7, 36.1, 46.6, 60.4, 64.5 (d,
JPC = 4.8 Hz), 82.9 (d, JPC = 9.1 Hz), 84.9, 111.2, 125.4, 133.5, 137.3, 151.3, 163.9 (d, JPC
=
-
4
.8 Hz), 166.1. HRMS m/z 428.0424, calcd. for C H N O PS 428.0547.
1
3
15
7
6
+
3
'-Azido-3'-deoxythymidin-5'-yl [N-(4-methylthiazol-2-yl)]phosphoramidate, TEAH
1
salt (7ao). Yield 75%. RP HPLC R 18.96 min. H NMR (400 MHz) δ (D O) 1.27 (9H, t, J =
f
H
2
7
.3 Hz), 1.86 (3H, s), 2.23 (3H, s), 2.47 (H, m), 3.18 (6H, q, J = 7.3 Hz), 4.04 (1H, m), 4.11
(H, m), 4.40 (1H, m), 6.14 (1H, t, J = 6.6 Hz), 6.77 (1H, d, J = 1.1 Hz), 7.60 (1H, d, J = 1.1
13
Hz). C NMR (100 MHz) 8.2, 11.6, 36.1, 46.6, 60.3, 64.5 (d, J = 4.9 Hz), 82.8 (d, JPC = 9
PC
Hz), 85, 111.2, 111.4, 136.9, 137.3, 151.4, 165.6 (d, J = 5.1 Hz), 166.3. HRMS m/z 442.0618,
-
calcd. for C H N O PS 442.0704.
1
4
17
7
6
+
3
'-Azido-3'-deoxythymidin-5'-yl [N-(benzo[d]thiazol-2-yl)]phosphoramidate, TEAH salt
1
(
7ap). Yield 69%. RP HPLC R 21.78 min. H NMR (400 MHz) δ (D O) 1.27 (9H, t, J = 7.3
f
H
2
Hz), 1.75 (3H, s), 2.40 (2H, m), 3.17 (6H, q, J = 7.3 Hz), 4.07 (2H, m), 4.17 (1H, m), 4.36
1H, m), 6.02 (1H, t, J = 6.5 Hz), 7.17 (1H, t, J = 7.6 Hz), 7.32 (1H, t, J = 7.7 Hz), 7.38 (1H,
(
13
s), 7.49 (1H, d, J = 8.1 Hz), 7.64 (1H, d, J = 7.9 Hz). C NMR (100 MHz) δ (D O) 7.7, 11.1,
c
2
3
5.5, 46.2, 59.3, 64.4 (d, JPC = 4.9 Hz), 82.1 (d, JPC = 8.4 Hz), 84.4, 110.6, 118.1, 120.7,
1
22.7, 125.8, 130.3, 136.7, 148.5, 150.7, 164.7, 165.4. HRMS m/z 478.0542, calcd. for
-
C H N O PS 478.0704.
17
17
7
6
+
2
',3'-Dideoxyuridin-5'-yl [N-(quinolin-3-yl)]phosphoramidate, TEAH salt (7bd). Two
1
chromatographic purifications were required. Yield 55%. RP HPLC R 17.81 min. H NMR
f
(
400 MHz) δ (D O) 1.19 (9H, t, J = 7.3 Hz), 1.72 (2H, m), 1.98 (1H, m), 2.24 (1H, m), 3.09
H 2
(6H, q, J = 7.3 Hz), 3.84 (1H, m), 4.09 (1H, m), 4.26 (1H, m), 5.06 (1H, d, J = 8.0 Hz), 5.83
(1H, dd, J = 3.3, 7.3 Hz), 7.07 (1H, d, J = 8.1 Hz), 7.42 (2H, m), 7.56 (1H, m), 7.71 (2H, m),
1
3
8
5
1
.50 (1H, d, J = 2.6 Hz). C NMR (100 MHz) δ (D O) 7.7, 22.7, 24.4, 30.3, 46, 66 (d, J
=
c
2
PC
.1 Hz), 79.4 (JPC = 9.5 Hz), 85.2, 100.9, 119.2 (JPC = 4.1 Hz).126, 126.2, 126.7, 126.9, 128,
35.6, 140.4 (JPC = 10.4 Hz), 142.7 (JPC = 10.8 Hz), 150.2, 164.6. HRMS m/z 417.0958,
-
calcd. for C H N O P 417.0969.
1
8
18
4
6
+
2
',3'-Dideoxyuridin-5'-yl [N-(quinolin-6-yl)]phosphoramidate, TEAH salt (7bg). Two
1
chromatographic purifications were required. Yield 52%. RP HPLC R 16.70 min. H NMR
f
(400 MHz) δ (D O) 1.27 (9H, t, J = 7.3 Hz), 1.80 (1H, m), 1.95 (1H, m), 2.09 (1H, m), 2.36
H 2
(1H, m), 3.19 (6H, q, J = 7.3 Hz), 3.91 (1H, m), 4.14 (1H, m), 4.37 (1H, m), 5.11 (1H, m),
1
6

ACCEPTED MANUSCRIPT
5
.12 (1H, d, J = 8.0 Hz), 5.96 (1H, dd, J = 3.3, 7.4 Hz), 7.23 (1H, d, J = 8.0 Hz), 7.38 (1H, d,
J = 2.5 Hz), 7.48 (2H, m), 7.82 (1H, d, J = 9.1 Hz), 8.18 (1H, d, J = 8.2 Hz), 8.57 (1H, d, J =
1
3
3
.0 Hz). C NMR (100 MHz) δ (D O) 7.7, 24.5, 30.1, 46.1, 66.1 (d, J = 5.2 Hz), 79.6 (d,
c 2 PC
JPC = 9.8 Hz), 85.6, 100.8, 110.4 (d, JPC = 4.9 Hz), 121.1, 123.7 (d, JPC = 9.7 Hz), 126.2,
1
28.8, 136.7, 139.9, 140.5, 140.9, 145.4, 150.5, 164.82. HRMS m/z 417.0636, calcd. for
-
C H N O P 417.0969.
18
18
4
6
+
2
',3'-Dideoxyuridin-5'-yl [N-(5-cyanopyridin-2-yl)]phosphoramidate, TEAH salt (7bl).
1
Yield 67%. RP HPLC R 15.78 min. H NMR (400 MHz) δ (D O) 1.30 (9H, t, J = 7.3 Hz),
f
H
2
1
.92 (1H, m), 2.12 (2H, m), 2.45 (1H, m), 3.22 (6H, q, J = 7.33Hz), 3.96 (1H, m), 4.16 (1H,
m), 4.35 (1H, m), 5.68 (1H, d, J = 8.1 Hz), 6.04 (1H, dd, J = 3.2, 7.3 Hz), 7.09 (1H, d, J = 8.8
1
3
Hz), 7.70 (1H, d, J = 8.1 Hz), 7.87 (1H, dd, J = 2.3, 8.9 Hz), 8.43 (1H, d, J = 2.0 Hz). C
NMR (100 MHz) δ (D O) 7.7, 24.4, 30.2, 46.2, 65.9 (d, J = 5.2 Hz), 79.7 (d, JPC = 9.2 Hz),
c
2
PC
8
3
2
5.9, 99.2, 101.1, 110.5, 117.8, 140.6, 141.7, 150.9, 151.8, 157.1, 165.5. HRMS m/z
-
92.0699, calcd. for C H N O P 392.0765.
1
5
15
5
6
+
',3'-Dideoxyuridin-5'-yl [N-(4-methylthiazol-2-yl)]phosphoramidate, TEAH salt (7bo).
1
Two chromatographic purifications were required. Yield 53%. RP HPLC R 15.90 min. H
f
NMR (400 MHz) δ (D O) 1.27 (9H, t, J = 7.3 Hz), 1.94 (1H, m), 2.08 (2H, m), 2.25 (3H, d,
H
2
J = 1 Hz), 2.42 (1H, m), 3.18 (6H, q, J = 7.3 Hz), 3.95 (1H, m), 4.13 (1H, m), 4.33 (1H, m),
5
.71 (1H, d, J = 8.1 Hz), 6.04 (1H, dd, J = 3.3, 7.1 Hz), 6.80 (1H, d, J = 1 Hz), 7.80 (1H, d, J
13
=
8.1 Hz). C NMR (100 MHz) δ (D O) 7.7, 10.3, 24.3, 30.5, 46.1, 65.5 (d, J = 5.3 Hz),
c 2 PC
7
9.8 (d, JPC = 9.1 Hz), 85.9 (d, JPC = 12.9 Hz), 101.1, 125, 132, 141.6, 151, 163.9 (d, JPC = 4.9
-
Hz), 165.7. HRMS m/z 387.0485, calcd. for C H N O PS 387.0533.
1
3
16
4
6
+
2
',3'-Didehydro-3'-deoxythymidin-5'-yl [N-(quinolin-3-yl)]phosphoramidate, TEAH
1
salt (7cd). Yield 67%. RP HPLC R 16.46 min. H NMR (400 MHz) δ (D O) 1.25 (9H, t, J =
f
H
2
7
.3 Hz), 1.39 (3H, s), 3.16 (6H, q, J = 7.3 Hz), 3.77 (1H, m), 4.19 (1H, d, J = 11.1 Hz), 5.16
(1H, d, J = 7.7 Hz), 5.71 (1H, d, J = 5.8 Hz), 6.37 (1H, d, J = 6.1 Hz), 6.77 (1H, m), 7.55 (2H,
13
m), 7.66 (1H, m), 7.68 (1H, m), 7.79 (1H, d, J = 8.3 Hz), 8.46 (1H, d, J = 2.6 Hz). C NMR
(100 MHz) δ (D O) 7.7, 10.5, 46.1, 66.2 (d, J = 5 Hz), 85 (d, JPC = 11.1 Hz), 89.3, 110, 119
c 2 PC
(JPC = 3.3 Hz), 125.2, 126, 126.2, 126.9, 127, 127.7, 132.6, 135.2, 136, 140.4, 142.5 (d, JPC
=
-
1
1.2 Hz), 150.7, 164.3. HRMS m/z 429.0933, calcd. for C H N O P 429.0969.
1
9
18
4
6
+
2
',3'-Didehydro-3'-deoxythymidin-5'-yl [N-(quinolin-6-yl)]phosphoramidate, TEAH
1
salt (7cg). RP HPLC R 15.78 min. Yield 97%. H NMR (400 MHz) δ (D O) 1.17 (9H, t, J =
f
H
2
7
.3 Hz), 1.36 (3H, s), 3.06 (6H, t, J = 7.3 Hz), 3.74 (1H, m), 4.10 (1H, m), 5.05 (1H, m), 5.65
(1H, m), 6.63 (1H, m), 6.84 (1H, d, J = 1 Hz), 7.15 (1H, d, J = 2.4 Hz), 7.32 (2H, m), 7.61
1
7

ACCEPTED MANUSCRIPT
1
3
(
1H, d, J = 9.8 Hz), 7.98 (1H, d, J=8.1 Hz), 8.47 (1H, m). C NMR (100 MHz) δ (D O) 7.7,
c 2
1
1
1
2
0.7, 46, 65.8 (d, JPC = 4.9 Hz), 85.1 (d, JPC = 10.8 Hz), 89.3, 109.7, 110 (d, JPC = 4.9 Hz),
21.1, 123.2 (d, JPC = 9.4 Hz), 124.8, 126.2, 128.4, 133.1, 136.1, 136.3, 139.8, 140.3, 145.5,
-
50.6, 164.3. HRMS m/z 429.0838, calcd. for C H N O P 429.0969.
1
9
18
4
6
',3'-Didehydro-3'-deoxythymidin-5'-yl
[N-(5-cyanopyridin-2-yl)]phosphoramidate,
+
TEAH salt (7cl). Two chromatographic purifications were required. Yield 63%. RP HPLC
1
R 15.47 min. H NMR (400 MHz) δ (D O) 1.28 (9H, t, J = 7.3 Hz), 1.73 (3H, d, J = 0.8 Hz),
f
H
2
3
.20 (6H, t, J = 7.3 Hz), 3.87 (1H, m), 4.15 (1H, m), 5.09 (1H, m), 5.89 (1H, m), 6.44 (1H,
m), 6.81 (1H, m), 6.98 (1H, d, J = 8.9 Hz), 7.30 (1H, d, J = 1.1 Hz), 7.79 (1H, d, J = 8.8 Hz),
1
3
8
8
1
2
.35 (1H, d, J = 2.1 Hz). C NMR (100 MHz) δ (D O) 7.8, 10.9, 46.1, 65.9 (d, J = 5.2 Hz),
c 2 PC
5 (d, JPC = 10.2 Hz), 89.6, 99.2, 110.2, 110.4, 117.7, 124.7, 133.5, 137.2, 140.1, 151.3,
-
51.6, 157 (d, JPC = 5.7 Hz), 165.3. HRMS m/z 404.0794, calcd. for C H N O P 404.0765.
16
15
5
6
',3'-Didehydro-3'-deoxythymidin-5'-yl
[N-(4-methylthiazol-2-yl)]phosphoramidate,
+
TEAH salt (7co). Two chromatographic purifications were required. Yield 61%. RP HPLC
1
R 16.01 min. H NMR (400 MHz) δ (D O) 1.29 (9H, t, J = 7.3 Hz), 1.79 (3H, d, J = 0.8 Hz),
f
H
2
2
5
.25 (3H, d, J = 1.1 Hz), 3.21 (6H, q, J = 7.3 Hz), 3.88 (1H, m), 4.10 (1H, m), 5.11 (1H, s),
1
3
.93 (1H, m), 6.47 (1H, m), 6.73 (1H, s), 6.88 (1H, m), 7.39 (1H, d, J = 1.1 Hz). C NMR
(
100 MHz) δ (D O) 7.7, 10.3, 10.9, 46.2, 65 (d, J = 4.9 Hz), 85 (d, JPC = 10.2 Hz), 89.7,
c 2 PC
1
10.3, 124.5, 124.8, 132, 133.7, 137.5, 151.7, 163.3, 165.9. HRMS m/z 399.0532, calcd. for
-
C H N O PS 399.0533.
14
16
4
6
(1S,cis)-4-[2-Amino-6-(cyclopropylamino)-9H-purin-9-yl]-2-cyclopentene-1-methyl [N-
+
(
5-cyanopyridin-2-yl)]phosphoramidate, TEAH salt (7dl). Two chromatographic
1
purifications were required. Yield 35%. RP HPLC R 19.39 min. H NMR (400 MHz) δ
f
H
(
D O) 0.69 (2H, m), 0.99 (2H, m), 1.27 (9H, t, J = 7.3 Hz), 1.93 (2H, m), 2.50 (1H, m), 2.88
2
(1H, m), 3.10 (1H, m), 3.19 (6H, q, J = 7.3 Hz), 3.57 (1H, m), 3.97 (1H, m), 5.10 (1H, m),
5
.88 (1H, m), 6.19 (1H, m), 6.71 (1H, d, J = 8.8 Hz), 7.47 (1H, dd, J = 2.2, 8.8 Hz), 7.50 (1H,
13
s), 8.12 (1H, d, J = 1.98Hz). C NMR (100 MHz) δ (D O) 6.3, 7.7, 22.8, 33.4, 44.8 (d, J =
PC
c
2
4
1
4
.0 Hz), 46.1, 58.4, 67.1 (d, JPC = 21.6 Hz), 98.5, 109.9, 112.6, 117.6, 128.3, 129.2, 136.5,
-
38, 139.9, 151.3, 154.8, 156.6, 158.7. HRMS m/z 466.1461, calcd. for C H N O P
2
0
21
3
3
66.1510.
(1S,cis)-4-[2-Amino-6-(cyclopropylamino)-9H-purin-9-yl]-2-cyclopentene-1-methyl [N-
+
(
4-methylthiazol-2-yl)]phosphoramidate, TEAH salt (7do). Two chromatographic
1
purifications were required. Yield 51%. RP HPLC R 19.94 min. H NMR (400 MHz) δ
f
H
(
D O) 0.66 (2H, m), 0.89 (2H, m), 1.26 (9H, t, J = 7.3 Hz), 1.42 (1H, m), 1.97 (3H, d, J = 1.0
2
1
8

ACCEPTED MANUSCRIPT
Hz), 2.65 (1H, m), 2.81 (1H, s), 3.09 (1H, s), 3.18 (6H, q, J = 7.3 Hz), 3.66 (1H, m), 3.90 (1H,
1
3
m), 5.17 (1H, m), 5.88 (1H, m), 6.17 (1H, m), 6.44 (1H, s), 7.57 (1H, s). C NMR (100 MHz)
δ (D O) 6.4, 7.7, 10.1, 22.6, 33.4, 45. (d, J = 9.6 Hz), 46.1, 58.7, 66.8 (d, JPC = 5.3 Hz),
c
2
PC
1
22.6, 124.5, 128.3, 132.1, 136.6, 138, 154.8, 158.5, 163. HRMS m/z 461.1461, calcd. for
-
C H N O PS 461.1278.
18
22
8
3
α-L
-2',3'-Dideoxy-3'-thiacytadin-5'-yl
[N-(4-methylthiazol-2-yl)]phosphoramidate,
+
1
TEAH salt (7eo). Yield 69%. RP HPLC R 12.03 min. H NMR (400 MHz) δ (D O) 1.27
f
H
2
(9H, t, J = 7.3 Hz), 2.22 (3H, d, J = 0.8 Hz), 3.12 (1H, m ), 3.17 (6H, q, J = 7.3 Hz), 3.49 (1H,
dd, J = 5.6, 12.2 Hz), 4.11 (1H, m), 4.24 (1H, m), 5.38 (1H, q, J = 7.6 Hz), 5.92 (1H, d, J =
1
3
7
.6 Hz), 6.25 (1H, t, J = 5.0 Hz), 6.77 (1H, s), 7.84 (1H, d, J = 7.6 Hz). C NMR (100 MHz)
δ (D O) 7.7, 10.4, 36.2, 46.1, 65.3 (d, J = 5.1 Hz), 83.4 (d, JPC = 8.9 Hz), 86.7, 95.4, 125,
c
2
PC
1
32, 141.1, 156.3, 163.9 (d, JPC = 4.9 Hz), 165.3. HRMS m/z 404.0546, calcd. for
-
C H N O PS2 404.0257.
12
15
5
5
Method B
Phosphoramidates containing amines of pK > 7 were prepared by the tristriazolide approach
a
as it was described previously [5]. The only difference was solubility of di(triazoyl)(N-
heteroaryl)phosphoramidates of type 11 in pyridine: derivative 11k was soluble at r.t., 11i
required heating at 75 °C for ca. 10, while 11e dissolved only partly and was used as a
3
1
suspension.
P
NMR data of intermediates 11 and AZT (triazoyl)(N-
heteroaryl)phosphoramidates 12 are listed in Table 3. Purification of the final products 7 was
done as in Method A.
+
3
'-Azido-3'-deoxythymidin-5'-yl [N-(quinolin-2-yl)]phosphoramidate, TEAH salt (7ak).
1
Yield 43%. RP HPLC R 22.40 min. H NMR (400 MHz) δ (D O) 1.26 (9H, t, J = 7.3 Hz),
f
H
2
1.60 (3H, s), 2.33 (2H, t, J = 6.7 Hz ), 3.17 (6H, k, J = 7.3 Hz), 4.03 (1H, m), 4.09 (1H, m),
4.20 (1H, m), 4.27 (1H, m), 5.92 (1H, t, J = 6.5 Hz), 7.14 (1H, s), 7.21 (1H, d, J = 9.1 Hz),
7.36 (1H, t, J = 7.4 Hz), 7.54 (1H, t, J = 7.4 Hz), 7.60 (1H, t, J = 7.7 Hz), 7.66 (1H, d, J = 8.0
13
Hz), 8.03 (1H, d, J = 9.1 Hz). C NMR (100 MHz) δ (D O) 7.7, 11, 35.3, 46.2, 59.2, 64,8 (d,
c
2
JPC = 5.1 Hz), 82 (d, JPC = 8.0 Hz), 84.2, 110.6, 112.7, 122.3, 122.8, 124.5, 127.5, 130.7,
-
1
4
3
36.4, 139.7, 150.3, 153,1, 164.9, 180.8. HRMS m/z 472.1062, calcd. for C H N O P
19 19 7 6
72.1140.
+
'-Azido-3'-deoxythymidin-5'-yl [N-(isoquinolin-1-yl)]phosphoramidate, TEAH salt
1
(
7ai). Yield 61%. RP HPLC R 21.94 min. H NMR (400 MHz) δ (D O) 1.27 (9H, t, J = 7.3
f
H
2
1
9

ACCEPTED MANUSCRIPT
Hz), 1.50 (3H, s), 2.39 (2H, m), 3.18 (6H, q, J = 7.3 Hz), 4.10 (1H, m), 4.14 (1H, m), 4.24
(1H, m), 4.33 (1H, m), 5.94 (1H, t, J = 6.5 Hz), 7.17 (1H, d, J = 6 Hz), 7.25 (1H, s), 7.41 (1H,
1
3
s), 7.56 (1H, m), 7.70 (2H, m), 7.84 (1H, d, J = 6.6 Hz), 8.03 (1H, d, J = 8.4 Hz). C NMR
(
100 MHz) δ (D O) 8.1, 11.2, 35.8, 46.5, 60.1, 65 (d, J = 4.5 Hz), 82.8 (d, J = 7.5 Hz), 84.9,
c 2
1
4
10.8, 113.7, 118.6, 122.5, 126.9, 130.7, 136.8, 137, 138.6, 151.2, 152.4, 166. HRMS m/z
-
72.1144, calcd. for C H N O P 472.1140.
1
9
19
7
6
+
3
'-Azido-3'-deoxythymidin-5'-yl bis[N-(pyrid-4-yl)]phosphordiamidate, TEAH salt
(9aa). The compound was formed as a side product during attempted preparation of 7aa by
1
Method A and was isolated chromatographically from the reaction mixture. Yield ca. 10%. H
NMR (400 MHz) δ (D O) 1.71 (3H, s), 2.50 (2H, m), 4.16 (1H, m), 4.30 (1H, m), 4.39 (1H,
H
2
m), 4.44 (1H, m), 6.18 (1H, t, J = 6.8 Hz), 6.99 (2H, dd, J = 1.6, 5.2 Hz), 7.01 (2H, dd, J =
1
3
1
.6, 5.0 Hz), 7.28 (1H, d, J = 0.8 Hz), 8.13 (2H, d, J = 6.4 Hz), 8.15 (2H, d, J = 5.6 Hz). C
NMR (100 MHz) δ (D O) 11.6, 35.2, 59.1, 64.5, 81.3, 84.8, 110.9, 112.9, 119.7, 124.0,
c
2
+
1
5
29.3, 136.3, 147.1, 147.4. HRMS (positive mode) m/z 500.1514, calcd. for C H N O P
20 23 9 5
00.1554.
Determination of octanol-water partition coefficient
Partition coefficients P were determined using micro shake-flask partitioning approach [22].
.01 M triethylammonium acetate buffer pH 7.4 and n-octanol stock solutions were initially
0
saturated one with another, separated, and left for 48 h. 20 µL of 40 mM aqueous solution of a
sample (ca. 2 mg/100 µL) was diluted with 480 µL of the buffer and shaken vigorously with
0
.5 mL of the added octanol at 22 °C for 24 h. The mixture was centrifuged and a half of each
layer was carefully pipetted out. From these, 15 µL aliquots were injected for HPLC and the
ratio of concentrations in each phase was determined according to the area of appropriate
peaks. For compounds of the lowest lipophilicity, a measured volume of the octanol layer was
evaporated to dryness, the residue dissolved in 1/10 volume of water, and used for HPLC
analysis.
TABLE 3
References
[
1] G. Gumina, G.Y. Song, C.K. Chu, L-Nucleosides as chemotherapeutic agents, FEMS
Microbiol. Lett. 202, (2001) 9-15.
2
0

ACCEPTED MANUSCRIPT
[
2] H. Lee, J. Hanes, K.A. Johnson, Toxicity of nucleoside analogues used to treat AIDS
and the selectivity of the mitochondrial DNA polymerase, Biochemistry 42, (2003)
14711-14719.
[
[
3] A. Lavie, I. Schlichting, I.R. Vetter, M. Konrad, J. Reinstein, R.S. Goody, The
bottleneck in AZT activation, Nature Medicine 3, (1997) 922-924.
4] For recent reviews, see: (a) S. Peyrottes, C. Perigaud, Mononucleotide Prodrug
Synthetic Strategies, in: P. Merino (Ed.), Chemical Synthesis of Nucleoside Analogues,
Wiley, 2013; pp. 229-261; (b) U. Pradere, E.C. Garnier-Amblard, S.J. Coats, F.
Amblard, R.F. Schinazi, Synthesis of Nucleoside Phosphate and Phosphonate Prodrugs,
Chem. Rev. 114, (2014) 9154-9218; (c) A.J. Wiemer, D.F. Wiemer, Prodrugs of
phosphonates and phosphates: Crossing the membrane barrier, Topics in Current
Chemistry 360, (2014) 115-160.
[
5] J. Romanowska, M. Sobkowski, A. Szymanska-Michalak, K. Kolodziej, A. Dabrowska,
A. Lipniacki, A. Piasek, Z.M. Pietrusiewicz, M. Figlerowicz, A. Guranowski, J.
Boryski, J. Stawinski, A. Kraszewski, Aryl H-phosphonates 17: (N-
Aryl)phosphoramidates of pyrimidine nucleoside analogues and their synthesis, selected
properties, and anti-HIV activity, J. Med. Chem. 54, (2011) 6482-6491.
[
[
6] J. Romanowska, M. Sobkowski, A. Szymanska-Michalak, K. Kolodziej, J. Boryski, J.
Stawinski, A. Kraszewski, Unpublihed results. Animal studies were done by the
Department of Toxicological Studies at the Institute of Industrial Organic Chemistry,
Branch Pszczyna, Poland.
7] H.A.L. Filipe, A. Salvador, J.M. Silvestre, W.L.C. Vaz, M.J. Moreno, Beyond Overton's
Rule: Quantitative Modeling of Passive Permeation through Tight Cell Monolayers,
Molecular Pharmaceutics 11, (2014) 3696-3706.
[
[
8] C.Y. Wei, A. Pohorille, Permeation of membranes by ribose and its diastereomers, J.
Am. Chem. Soc. 131, (2009) 10237-10245.
9] O. Adelfinskaya, P. Herdewijn, Amino acid phosphoramidate nucleotides as alternative
substrates for HIV-1 reverse transcriptase, Angew. Chem. Int. Ed. 46, (2007) 4356-
4358.
[
[
10] O. Adelfinskaya, M. Terrazas, M. Froeyen, P. Marliere, K. Nauwelaerts, P. Herdewijn,
Polymerase-catalyzed synthesis of DNA from phosphoramidate conjugates of
deoxynucleotides and amino acids, Nucleic Acids Res. 35, (2007) 5060-5072.
11] I. Zlatev, A. Giraut, F. Morvan, P. Herdewijn, J.J. Vasseur, d-Di-carboxybutyl
phosphoramidate of 2'-deoxycytidine-5'-monophosphate as substrate for DNA
polymerization by HIV-1 reverse transcriptase, Bioorg. Med. Chem. 17, (2009) 7008-
7014.
[
12] J. Jankowska, M. Sobkowski, J. Stawinski, A. Kraszewski, Studies on aryl H-
phosphonates. 1. An efficient method for the preparation of deoxyribo- and
ribonucleoside 3'-H-phosphonate monoesters by transesterification of diphenyl H-
phosphonate, Tetrahedron Lett. 35, (1994) 3355-3358.
2
1

ACCEPTED MANUSCRIPT
[
[
13] P.J. Garegg, T. Regberg, J. Stawinski, R. Stromberg, Nucleoside phosphonates: Part 7.
Studies on the oxidation of nucleoside phosphonate esters, J. Chem. Soc. , Perkin Trans.
1
(1987) 1269-1273.
3
1
14] J. Nilsson, A. Kraszewski, J. Stawinski, Reinvestigation of the P NMR evidence for
the formation of diorganyl phosphoropyridinium intermediates, J. Chem. Soc. , Perkin
Trans. 2 (2001) 2263-2266.
[
[
[
[
15] A. Kraszewski, J. Stawinski, Phosphoryl tris-triazole - a new phosphorylating reagent,
Tetrahedron Lett. 21, (1980) 2935-2936.
16] D.D. Perrin, Dissociation Constants of Organic Bases in Aqueous Solution, Butterworth,
London, 1965.
17] A. Albert, R. Goldacre, J. Phillips, The strength of heterocyclic bases, J. Chem. Soc.
(1948) 2240-2249.
18] A.R. Osborn, K. Schofield, L.N. Short, Studies of the amino-isoquinolines, amino-
cinnolines, and amino-quinazolines. (A) The basic strengths and ultraviolet absorption
spectra. (B) The infrared spectra, J. Chem. Soc. (1956) 4191-4206.
[
19] A.S. Al Attas, M.M. Habeeb, D.S. Al Raimi, Synthesis and spectroscopic studies of
charge transfer complexes between chloranilic acid and some heterocyclic amines in
ethanol, J. Mol. Struct. 928, (2009) 158-170.
[
[
20] S.B. Goncharenko, M.M. Kaganskii, Y. Portnov, V.G. Granik, Ionization constants of
the drugs amiridin and tacrine and their analogs, Pharm. Chem. J. 26, (1992) 769-772.
21] K. Kolodziej, J. Romanowska, J. Stawinski, A. Kraszewski, M. Sobkowski, The case of
triethylammonium cation loss during purification of certain nucleotide analogues. A
cautionary note, Anal. Bioanal. Chem. 407, (2015) 1775-1780.
[
[
[
22] H. Ford, C.L. Merski, J.A. Kelley, A rapid microscale method for the determination of
partition coefficients by HPLC, Journal of Liquid Chromatography 14, (1991) 3365-
3386.
23] M.A. Siddiqui, J.S. Driscoll, V.E. Marquez, J.S. Roth, T. Shirasaka, H. Mitsuya, J.J.
Barchi, J.A. Kelley, Chemistry and anti-HIV properties of 2'-fluoro-2',3'-
dideoxyarabinofuranosylpyrimidines, J. Med. Chem. 35, (1992) 2195-2201.
24] A.Q. Siddiqui, C. McGuigan, C. Ballatore, F. Zuccotto, I.H. Gilbert, E. De Clercq, J.
Balzarini, Design and synthesis of lipophilic phospharamidate d4T-MP prodrugs
expressing high potency against HIV in cell culture: Structural determinants for in vitro
activity and QSAR, J. Med. Chem. 42, (1999) 4122-4128.
[
[
25] F.M. Uckun, S. Pendergrass, S. Qazi, P. Samuel, T.K. Venkatachalam, Phenyl
phosphoramidate derivatives of stavuldine as anti-HIV agents with potent and selective
in-vitro antiviral activity against adenovirus, Eur. J. Med. Chem. 39, (2004) 225-234.
26] O. Janneh, S.H. Khoo, Interactions of tenofovir, lamivudine, abacavir and didanosine in
primary human cells, Pharmaceutics 3, (2011) 326-337.
2
2

ACCEPTED MANUSCRIPT
[
[
27] C.A.S. Bergstrom, M. Strafford, L. Lazorova, A. Avdeef, K. Luthman, P. Artursson,
Absorption classification of oral drugs based on molecular surface properties, J. Med.
Chem. 46, (2003) 558-570.
28] M. Shafiee, S. Deferme, A.L. Villard, D. Egron, G. Gosselin, J.L. Imbach, T. Lioux, A.
Pompon, S. Varray, A.M. Aubertin, G. Van den Mooter, R. Kinget, C. Perigaud, P.
Augustijns, New bis(SATE) prodrug of AZT 5'-monophosphate: In vitro anti-HIV
activity, stability, and potential oral absorption, J. Pharm. Sci. 90, (2001) 448-463.
[
[
[
[
29] H. Schott, P.S. Ludwig, A. Immelmann, R.A. Schwendener, Synthesis and in vitro anti-
HIV activities of amphiphilic heterodinucleoside phosphate derivatives containing the
2',3'-dideoxynucleosides ddC, AZT and ddI, Eur. J. Med. Chem. 34, (1999) 343-352.
30] F. Pertenbreiter, J. Balzarini, C. Meier, Nucleoside mono- and diphosphate prodrugs of
',3'-dideoxyuridine and 2',3'-dideoxy-2',3'-didehydrouridine, ChemMedChem 10,
2015) 94-106, and the references cited therein.
2
(
31] Y. Mehellou, J. Balzarini, C. McGuigan, An investigation into the anti-HIV activity of
',3'-didehydro-2',3'-dideoxyuridine (d4U) and 2',3'-dideoxyuridine (ddU)
2
phosphoramidate 'ProTide' derivatives, Org. Biomol. Chem. 7, (2009) 2548-2553.
32] J. Romanowska, A. Szymanska-Michalak, M. Pietkiewicz, M. Sobkowski, J. Boryski, J.
Stawinski, A. Kraszewski, A new, efficient entry to non-lipophilic H-phosphonate
monoesters - Preparation of anti-HIV nucleotide analogues, Lett. Org. Chem. 6, (2009)
496-499.
2
3

ACCEPTED MANUSCRIPT
Figure 1. Possible protonation sites in AZT N-(pyrid-4-yl)phosphoramidic acid.
2
4

ACCEPTED MANUSCRIPT
H O
2
(
if I /H O/Py is used for oxidation)
~-10 ppm
P
2
2
NuO
O
H
O
PhO
O
NuO
O
H
PhO
O
O
Py
P
+ ArNH2 +
P
P
+
P
+ Cl
PhO Cl
NHAr PhO
1
2
DPCP
3
4
~7-9 ppm
~
3 ppm
~630 Hz
~-6 ppm
P
P
P
1J ~650-670 Hz
1J
HP
HP
two diastereomers
I2
two diastereomers
~-7 ppm ~-24 ppm
~-3 ppm
P
P
P
NuO
O
P
NHAr
H O NuO NHAr O OPh
NuO
I
O
2
P
P
P
not observed
O
O
O
OPh
NHAr in 31P NMR
7
6
5
DPCP
(
verification of the structure of 6)
Ar: heteroaromatic group (see Table 1)
Py: pyridine
HN
N
N
Nu =
H2N
Thy
N
O
Ura
Thy
Cyt
N
O
O
S
N3
a, AZT
O
e, 3TC
b, ddU
c, d4T
d, ABC
Figure 2. Synthesis of (N-heteroaryl)phosphoramidates 7. Method A
2
5

ACCEPTED MANUSCRIPT
Figure 3. A putative mechanism for the formation of nucleoside di(N-heteroaryl)phosphordiamidates 9 for
amines of pK > 7 (see Table 1)
a
2
6

ACCEPTED MANUSCRIPT
Figure 4. Synthesis of AZT (N-heteroaryl)phosphoramidates 7ax. Method B
2
7

ACCEPTED MANUSCRIPT
0
5
10
15
20
25
30
0
5
10
15
20
25
30
RT [min]
RT [min]
Figure 5. Exemplary HPLC traces of phosphoramidates of type 7 incubated with FBS in RPMI buffer, 37 °C. A,
phosphoramidate 7an (AZT AT); B, phosphoramidate 7bo (ddU AMT); n.i. = not identified
–
–
2
8