Studies on the synthesis of N′-acetyl AZA-analogues of Ganciclovir - Unexpected liability of N′-(2-hydroxyethyl)-azanucleosides under basic conditions

Article abstract of DOI:10.1080/15257770.2010.519367

The O′-pivaloyl diesters of N′-acetyl-azanucleosides were obtained from N-[1,3-di(pivaloyloxy)prop-2-yl]-N-(pivaloyloxymethyl)acetamide and a silylated nucleobase under Vorbruggen′s conditions. Unexpectedly, de-pivaloylation of the diesters under basic conditions afforded the corresponding nucleobase and N-acetylserinol. Mechanistic investigations showed that these products result from the following cascade of spontaneous transformations initiated by the mono de-pivaloylation of the starting diesters. N′-Deacetylation of the resultant mono-esters via the intramolecular N-O acetyl migration is the key step of the cascade; the corresponding NH-azanucleosides in the form of O-acetyl-O′-pivaloyl diesters are formed. Fragmentation of these diester intermediates gives the nucleobase and O-acetyl-O'-pivaloylserinol. Conversion of the latter to N-acetylserinol involves the selective O-N acetyl migration followed by de-pivaloylation of the resulting N-acetyl-O-pivaloylserinol. Copyright Taylor and Francis Group, LLC.

Full text of DOI:10.1080/15257770.2010.519367

This article was downloaded by: [University of Colorado at Boulder Libraries]
On: 30 December 2014, At: 22:13
Publisher: Taylor & Francis
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered
office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK
Nucleosides, Nucleotides and Nucleic
Acids
Publication details, including instructions for authors and
subscription information:
http://www.tandfonline.com/loi/lncn20
Studies on the Synthesis of
N′-Acetyl AZA-Analogues of
Ganciclovir—Unexpected Liability of N′-
(2-Hydroxyethyl)-Azanucleosides Under
Basic Conditions
Mariola Koszytkowska-Stawińska a
a Faculty of Chemistry , Warsaw University of Technology ,
Warszawa, Poland
Published online: 04 Oct 2010.
To cite this article: Mariola Koszytkowska-Stawińska (2010) Studies on the Synthesis of N′-
Acetyl AZA-Analogues of Ganciclovir—Unexpected Liability of N′-(2-Hydroxyethyl)-Azanucleosides
Under Basic Conditions, Nucleosides, Nucleotides and Nucleic Acids, 29:10, 768-785, DOI:
10.1080/15257770.2010.519367
To link to this article: http://dx.doi.org/10.1080/15257770.2010.519367
PLEASE SCROLL DOWN FOR ARTICLE
Taylor & Francis makes every effort to ensure the accuracy of all the information (the
“Content”) contained in the publications on our platform. However, Taylor & Francis,
our agents, and our licensors make no representations or warranties whatsoever as to
the accuracy, completeness, or suitability for any purpose of the Content. Any opinions
and views expressed in this publication are the opinions and views of the authors,
and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content
should not be relied upon and should be independently verified with primary sources
of information. Taylor and Francis shall not be liable for any losses, actions, claims,
proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or
howsoever caused arising directly or indirectly in connection with, in relation to or arising
out of the use of the Content.
This article may be used for research, teaching, and private study purposes. Any
substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing,
systematic supply, or distribution in any form to anyone is expressly forbidden. Terms &

Conditions of access and use can be found at http://www.tandfonline.com/page/terms-
and-conditions

Nucleosides, Nucleotides and Nucleic Acids, 29:768–785, 2010
Copyright ꢀC Taylor and Francis Group, LLC
ISSN: 1525-7770 print / 1532-2335 online
DOI: 10.1080/15257770.2010.519367
STUDIES ON THE SYNTHESIS OF Nꢀ-ACETYL AZA-ANALOGUES
OF GANCICLOVIR—UNEXPECTED LIABILITY OF
Nꢀ-(2-HYDROXYETHYL)-AZANUCLEOSIDES UNDER
BASIC CONDITIONS
Mariola Koszytkowska-Stawin´ ska
Faculty of Chemistry, Warsaw University of Technology, Warszawa, Poland
The Oꢁ-pivaloyl diesters of Nꢁ-acetyl-azanucleosides were obtained from N-[1,3-
2
di(pivaloyloxy)prop-2-yl]-N-(pivaloyloxymethyl)acetamide and
a silylated nucleobase under
Vorbru¨ggen’s conditions. Unexpectedly, de-pivaloylation of the diesters under basic conditions
afforded the corresponding nucleobase and N-acetylserinol. Mechanistic investigations showed
that these products result from the following cascade of spontaneous transformations initiated by
the mono de-pivaloylation of the starting diesters. Nꢁ-Deacetylation of the resultant mono-esters
via the intramolecular N-O acetyl migration is the key step of the cascade; the corresponding
NH-azanucleosides in the form of O-acetyl-Oꢁ-pivaloyl diesters are formed. Fragmentation of these
diester intermediates gives the nucleobase and O-acetyl-Oꢁ-pivaloylserinol. Conversion of the latter
to N-acetylserinol involves the selective O-N acetyl migration followed by de-pivaloylation of the
resulting N-acetyl-O-pivaloylserinol.
Keywords Nucleoside analogues; aza-nucleosides; N-(2-hydroxyethyl)amides; N-O acyl
migration; O-N acyl migration
INTRODUCTION
Acyclovir and ganciclovir (Figure 1) are drugs used for treatment of
virus infections.[1] In search of their novel analogues, azanucleosides shown
in Figure1 have recently been developed at this laboratory. The com-
pounds were derived from N -(2-hydroxyethyl)arenosulfonamides (deriva-
tives 1[2]), N -(1,3-dihydroxyprop-2-yl)metanosulfonamide (derivatives 2[3]),
or 5,5-bis(hydroxymethyl)pyrrolidin-2-one (derivatives 3[4]). Among them
the thymine, guanine, or cytosine derivatives 3 showed a modest antiviral
activity.
Received 23 June 2010; accepted 24 August 2010.
This work was financially supported by Warsaw University of Technology. The author thanks Prof.
Wojciech Sas and Dr. Włodzimierz Buchowicz, Warsaw University of Technology, for their support and
fruitful discussions.
Address correspondence to Mariola Koszytkowska-Stawin´ska, Faculty of Chemistry, Warsaw Univer-
sity of Technology, ul. Noakowskiego 3, 00-664, Warszawa, Poland. E-mail: mkoszyt@ch.pw.edu.pl
768

Nꢁ-Acetyl Aza-Analogues of Ganciclovir
769
FIGURE 1 Selected antiviral drugs and their aza analogues 1–3.
Herein, the studies on the synthesis of azanucleosides 6 derived from N -
(1,3-dihydroxyprop-2-yl)acetamide are reported (Scheme 1). Structurally,
compounds 6 are a combination of derivatives 2 and 3. In relation to 3,
a greater conformational freedom of 6 resulting from the replacement of
the pyrrolidin-2-one ring with the flexible acyclic backbone, such as that
of 2, would be expected to have an effect on their biochemical prop-
erties. According to the methodology developed for the aforementioned
azanucleosides 1–3, the synthetic approach to 6 involved coupling of N -
[1,3-di(pivaloyloxy)prop-2-yl]-N -(pivaloyloxymethyl)acetamide 4 with a nu-
cleobase under Vorbru¨ggen’s conditions followed by ammonolysis of the re-
sulting Oꢁ-pivaloylated azanucleosides 5 (Scheme 1). Azanucleosides 6 were
labile under basic conditions. The nature of the final products obtained in
a majority of the examined reactions, that is, those of the corresponding
nucleobase (BH) and N -acetylserinol 7, suggested that the liability of 6 was
associated with the presence of the β-hydroxyamide system in their structure;
the N O acyl migration process seemed to be involved in releasing of the
nucleobase from the examined compounds. In the course of investigating
these observations, it has been found that the N -O acyl migration in N -
acyl-(β-hydroxyalkyl)amines under basic[5] conditions has been postulated
as an initial step of a base-induced hydrolysis of pseudoephedrine amides.
Nevertheless, this hypothesis was not supported by experimental evidences.
On the other hand, the interconversion of N -acyl-(β-hydroxyalkyl)amines
into O-acyl-(β-hydroxyalkyl)ammonium salts under acidic conditions via N -
O acyl migration is well documented in the literature.[6] The mechanistic
SCHEME 1 Synthetic strategy for Oꢁ-pivaloylated azanucleosides 5 and their further transformations.

770
M. Koszytkowska-Stawin´ska
considerations[6] and experimental evidences[7,8] reveal that the intercon-
version is reversible and, under basic conditions, the resulting O-acyl-(β-
hydroxyalkyl)ammonium salt smoothly isomerises to the parent N -acyl-(β-
hydroxyalkyl)amine via O-N acyl migration. The latter direction of this
pH-dependent equilibrium was successfully employed for creation of pro-
drugs of HIV protease inhibitors,[9] or anticancer agents.[10] To the best of au-
thor’s knowledge, the chemical behavior of N -acyl-(β-hydroxyalkyl)amines
has not been considered with reference to the stability and potential ap-
plications of nucleoside analogues. Therefore, the reported studies on the
liability of 6 were focused on two aspects: (a) the nature of the β-hydroxy
group participation in the liberation of the nucleobase, and (b) the differ-
ence in the behavior of 6 and the 5-(hydroxymethyl)-containing lactams 3
under the examined conditions.
It is worth noting that nucleobase derivatives capable of releasing a
free nucleobase have been studied as potential carriers for nucleobases,
mainly for fluorouracil. The derivatives with the following functionality
were examined in this respect: a carbamoyl [RNHC(O) ], carbamoyl-
methyl [RNHC(O)CH2 ], aminomethylcarbamate [ROC(O)NHCH2 ], or
1-amido-1-carboxyalkyl [RC(O)NHC(CO2H)(alkyl)].[11,12]
RESULTS AND DISCUSSION
The synthesis of the Oꢁ-pivaloylated azanucleosides 5a-d is shown
in Scheme 2. The starting N -[1,3-bis(pivaloyloxy)prop-2-yl]acetamide 4
SCHEME 2 Reagents and conditions: (i) PivCl, 90◦C, 4 hours; (ii) Ac2O, pyridine, room temerpature,
1 day; (iii) ClCH2OPiv, NaH, DMF, 4 days; (iv) 1). nucleobase (N 4-benzoylcytosine, thymine, or 2-
thiouracil), BSA, acetonitrile, room temperature 1 hour; 2). 4, TMSOTf, acetonitrile, room temperature,
2 days; (v) 1). N 2-acetyl-O6-(diphenylcarbamoyl)guanine, BSA, CH2Cl2, 80◦C, 15 minutes; 2). 4, TMSOTf,
toluene, 80◦C, 1 hour.

Nꢁ-Acetyl Aza-Analogues of Ganciclovir
771
was prepared by: (i) O-pivaloylation of serinol hydrochloride 8 with pi-
valoyl chloride (PivCl); (ii) N -acetylation of 9 with acetic anhydride
(Ac2O); and (iii) N -pivaloyloxymethylation of 10 with chloromethyl pi-
valate (ClCH2OPiv). The reaction (iii) was performed under the condi-
tions optimized during N -pivaloyloxymethylation of N -cyclohexylacetamide
(50% yield[13]), N -cyclohexylmethanesulfonamide (75% yield[13]), and
N -(1,3-dipivaloyloxyprop-2-yl)methanesulfonamide (72% yield[3]). The N -
pivaloyloxymethylation of 10 occurred with 48% yield, that is, a lower yield
than N -(1,3-dipivaloyloxyprop-2-yl)methanesulfonamide, but comparable to
that of N -cyclohexylacetamide. Presumably, the observed tendency in the N -
pivaloyloxymethylation results from a lower N-H acidity of amides than the
N-H acidity of sulfonamides.[14]
Coupling of 4 with a silylated nucleobase in the presence of trimethylsi-
lyl trifluoromethanesulfonate (TMSOTf) afforded 5a-d in good to satis-
factory yield (88–43%). Derivatives 5a, 5b, or 5c were obtained from N 4-
benzoylcytosine, thymine or 2-thiouracil, respectively, (conditions iv). Com-
pound 5d was prepared from N 2-acetyl-O6-(diphenylcarbamoyl)guanine
(conditions v). Effectiveness of the couplings with N 4-benzoylcytosine
(88%), thymine (81%), or N 2-acetyl-O6-(diphenylcarbamoyl)guanine (47%)
was higher than (or at least comparable to) those previously per-
formed.[3,4,13] The coupling with 2-thiouracil gave 5c in 43% yield. In the
previous studies on the synthesis of 2ꢁ-azanucleosides, 2-thiouracil was not
examined. However, the literature data reveal that coupling of 2-thiouracil
is often less effective than the coupling of uracil or thymine with the same
sugar mimic under Vorbru¨ggen’s conditions.[15]
According to the literature data, ammonolysis is the most common pro-
cedure used for the removal of the O-acyl protecting groups or the N -acyl
protecting groups from nucleoside analogues.[15] In the previous studies, the
ammonolysis of the protected azanucleosides was routinely accomplished by
the use of aqueous ammonia in methanol (Scheme 3). The deprotection of
the nucleobase moiety of the cyclohexyl derivatives 11 was effective at room
SCHEME 3 The routine ammonolysis conditions used for the preparation of azanucleosides 1a, 3, and
12.

772
M. Koszytkowska-Stawin´ska
SCHEME 4 Reagents and conditions: 5, NH4OHconc/MeOH, sealed tube, 1 hour: (i) 70◦C; (ii) room
temperature.
temperature.[13] In contrast to 11, the removal of the Oꢁ-pivaloyl function
required heating of the reaction mixture at 70◦C at least for 1 day. Under
those conditions, the ammonolysis of derivatives with both the Oꢁ-pivaloyl
function and the N ꢁ-acyl one in the azasugar moiety (such as 13[2a] or 14[4]
)
occurred chemoselectively. The desired products 1a or 3, resulting from the
selective removal of the Oꢁ-pivaloyl function, were obtained in high yields.
Preliminary trials on the ammonolysis of 5a-d with aqueous ammonia in
methanol at 70◦C (Scheme 4, conditions i) afforded N -acetylserinol 7 and
the corresponding nucleobase (BH = cytosine, thymine, 2-thiouracil, or gua-
nine) as the only products. The reactions performed at room temperature
gave the same results for 5b, 5c, or 5d (conditions ii), whereas the reaction
of the N 4-benzoylcytosine derivative 5a yielded diester 5e (12%) resulting
from deprotection of the cytosine moiety, monoester 6a (4%), 7 (58%),
and cytosine (60%). The corresponding nucleobase was also liberated when
5a or 5b was treated, in a methanolic solution, with tetrabutylammonium
hydroxide or a solid potassium carbonate at room temperature.
The results from the ammonolysis of 5a-d, particularly the formation of
the monoester 6a from the N 4-benzoylcytosine derivative 5a, suggested that
the liberation of the nucleobase from 6 was initiated by the deprotection of
the hydroxy group. The cleavage of azanucleosides leading to the formation
of an azasugar and a free nucleobase was reported when the removal of
the N ꢁ-protecting group was attempted.[16] Therefore, it is assumed that the
liberation of the free nucleobase from 6 has resulted from their initial N ꢁ-
deacetylation. Scheme 5 illustrates two possibilities of the N ꢁ-deacetylation
of 6, as well as further transformations leading to N -acetylserinol 7.

Nꢁ-Acetyl Aza-Analogues of Ganciclovir
773
SCHEME 5 Plausible pathways for liberation of the nucleobase from 6.
Path (a) involves the intramolecular N-O acetyl migration leading to
the O-acetyl-Oꢁ-pivaloyl amine 15, followed by the liberation of the free
nucleobase (BH) and formation of O-acetyl-Oꢁ-pivaloylserinol 16. The in-
tramolecular O-N acetyl migration in 16 gives N -acetyl-O-pivaloylserinol
17 which, after ammonolysis of the O-pivaloyl protection, transforms into
N -acetylserinol 7. O-Acetyl-Oꢁ-pivaloylserinol 16 may also isomerise to N -
pivaloyl-O-acetylserinol 18 via the intramolecular O-N pivaloyl migration.
This interconversion leads to N -pivaloylserinol 19 as an accompanying prod-
uct. The fact that 19 was not detected among the products of the ammonoly-
sis may suggest that, if path (a) is operative, 16 does not undergo the O-N
pivaloyl migration or the migration occurs much slower than the O-N acetyl
migration leading to 17. Path (b) involves ammonolysis of the acetamide link-
age in 6 leading to the O-pivaloyl amine 20 and acetamide (AcNH2), followed
by releasing of the free nucleobase (BH) and formation of O-pivaloylserinol
21. The intramolecular O-N pivaloyl migration in 21 gives N -pivaloylserinol
19, whereas its competing reaction with the previously formed acetamide
leads to O-acetyl-Oꢁ-pivaloylamine 16 and/or N -acetyl-O-pivaloylserinol 17.
Consequently, formation of 21 in the tested ammonolysis reactions may give
the free nucleobase accompanied by: (i) N -pivaloylserinol 19, as the only
product formed from 21; (ii) N -acetylserinol 7 resulted from reaction(s) of
21 with acetamide, when 21 does not undergo the O-N pivaloyl migration;
or (iii) a mixture of 7 and 19. The fact that the ammonolysis of 5 has af-
forded the free nucleobase and 7 as the only products may suggest that, when

774
M. Koszytkowska-Stawin´ska
SCHEME 6 Reagents and conditions: NH4OHconc/MeOH, sealed tube, 1 day: (ia) or (ib) room tem-
perature; (iia) or (iib) 70◦C.
path (b) is operative, the reaction of 21 with acetamide (i.e., transamidation
to 17, or alcoholysis to 16) predominates over the O-N pivaloyl migration.
However, the literature data on transamidation reveal that such exchange
reaction occurs between the salt of amine and amide (usually a substituted
urea[17]) at an elevated temperature, or between neutral amine and a specific
derivative of N ,N -diacetylaniline.[18] Analogously, the alcoholysis of amides
proceeds under catalysis by a mineral acid,[19] alkoxide,[20] or a transition-
metal salt,[21] for instance. To gain an information on a behavior of the
potential intermediate 21 under the ammonolysis conditions, the reactions
shown in Scheme 6 were performed.
Treatment of O,Oꢁ-dipivaloylserinol hydrochloride 9 with a methanolic
solution of ammonium hydroxide at room temperature for 1 day gave 19 and
O,N -dipivaloylserinol 22 in 51% and 28% yield, respectively. The same prod-
ucts were obtained from the reaction performed at 70◦C; the yield of 19 and
22 were 65% and 11%, respectively. As expected, 19 and 22 were also the only
products obtained, when an equimolar mixture of 9 and acetamide (AcNH2)
was treated with a methanolic solution of ammonium hydroxide under the
previously used conditions; the yield of the products were comparable to
those obtained from the reactions curried out in the absence of acetamide;
acetamide was recovered in the yields exciding 90%. Compounds corre-
sponding to potential acetylation products of O,Oꢁ-dipivaloylserinol (i.e., the
N -acetyl derivatives 7, 10 or 17) were not obtained (for preparation of the
authentic reference 17 see, Experimental section). Furthermore, treatment
of an equimolar mixture of ethanolamine 23, an unbranched counterpart
of 21, and acetamide with a methanolic solution of ammonium hydroxide
under the previously used conditions did not afford N -acetylethanolamine
24, as the potential product resulting from O-acylation of 23 by acetamide
and the subsequent O-N acyl migration. Instead, the starting material (by
NMR) was recovered. These results indicate that the N ꢁ-deacetylation of 6 by
the ammonolysis of the acetamide function leading to intermediate 21, that
is, in accordance with path (b), should give 19 as the only product. In view
of these results, it is suggested that the liberation of the nucleobase from 6
under the tested conditions is a consequence of their initial N ꢁ-deacetylation

Nꢁ-Acetyl Aza-Analogues of Ganciclovir
775
via the intramolecular N-O acetyl migration, that is, in accordance with path
(a). Although an equilibrium of the N -acyl-(β-hydroxyalkyl)amine-O-acyl-
(β-hydroxyalkyl)amine interconversion is shifted toward the amide under
basic[6] or hydrolytic[22] conditions, the liberation of the free nucleobase
from 15, typical for N ꢁ-unprotected azanucleosides,[16] seems to be the driv-
ing force of the process. Taking into account the ability of 9 to undergo
rearrangement to O,N -dipivaloylserinol 22 under the ammonolysis condi-
tions (Scheme 6), it is suggested that the transformation of the postulated
16 into 17 via the O-N acetyl migration proceeded faster than its transforma-
tion into 18 via the O-N pivaloyl migration under the examined conditions;
probably due to a less steric effect of the acetyl group in relation to that of the
pivaloyl one.[9a] Therefore, N -acetylserinol 7 is the only product resulting
from the serinol moiety of 6.
As mentioned earlier, the behavior of 6 under the ammonolysis con-
ditions was in contrast to that previously observed for the 5,5-bis(hydro
xymethyl)pyrrolidin-2-on-1-yl nucleosides 3[4] as well as to those reported
in the literature for N -acetyl-5-(hydroxymethyl)pyrrolidin-2-yl nucleosides
25 (Figure 2).[23] To rationalize this difference in behavior of the com-
pounds, energetic requirements for the N -acyl-(β-hydroxyalkyl)amine-O-
acyl-(β-hydroxyalkyl)amine interconversion were considered. The literature-
based pathway for the interconversion (shown in Figure 2) involves:
(i) transformation of N -(β-hydroxyalkyl)amide A into alkoxide B by the
FIGURE 2 Total strain energy (ES) evaluated for intermediates involved into the N -(2-
hydroxyalkyl)amide-O-(2-aminoalkyl)ester interconversion.

776
M. Koszytkowska-Stawin´ska
action of ammonium hydroxide;[24] (ii) the intramolecular attack of the
alkoxide anion on the carbonyl carbon to form 1,3-oxazolidin-2-olate C;[6]
and (iii) conversion of C to the final O-(β-aminoalkyl)ester D. The sec-
ond step of the sequence, that is, the formation of 1,3-oxazolidin-2-olate
C, is the rate-determining step of the overall process.[25] Therefore, in line
with the Hammond postulate,[26] the reaction is believed to take place via
the transition state, which should closely resemble 1,3-oxazolidin-2-olate C.
Consequently, the ease of the conversion of N -(β-hydroxyalkyl)amides A to
O-(β-aminoalkyl)esters D might be discussed in relation to the stability of
1,3-oxazolidin-2-olates C. Since the Ruzicka hypothesis,[27] the stability of the
ring being formed is considered as a major factor affecting the ease of ring
closure and it is expressed as strain energy of the ring. Therefore, the total
strain energy (ES) in both 1,3-oxazolidin-2-olates C and the corresponding
starting alkoxides B related to compounds 3, 6, or 25 was evaluated by the
MM2 molecular mechanics method.[28] Taking into account the fact that
complex structures of the considered systems might result in many local
minima of the evaluated strain energy, calculations were performed on the
simplified systems corresponding to alkoxides B (i.e., Model B; Figure 2)
and 1,3-oxazolidin-2-olates C (i.e., Model C; Figure 2) where a nucleobase
ring (Figure 2, entries 1–3) as well as the pivaloyloxymethyl function (en-
try 1) (or the hydroxymethyl one, entry 2) were replaced with hydrogen
atoms.
The calculations showed that the monocyclic intermediates C formed
from 6 were much less strained than both the fused bicyclic intermediates
C formed from 25 and the bridged ones formed from 3 (Figure 2).[29] The
lowest strain energy in 1,3-oxazolidin-2-olates C formed from 6 is believed
to reflect in the lowest energy in their transition states. For that reason,
in contrast to 3 or 25, derivatives 6 proved to be capable of rearranging
to the corresponding esters 15 (shown in Scheme 5) under relatively mild
conditions.
CONCLUSION
In summary, the pivaloyl esters of N ꢁ-acetyl-azanucleosides 5a-d were
obtained from N -[1,3-bis(pivaloyloxy)prop-2-yl)-N -(pivaloyloxymethyl)]
acetamide 4 and the corresponding silylated nucleobase under Vorbru¨ggen’s
conditions. The ammonolysis of 5a-d gave the corresponding nucleobase
(BH) and N -acetylserinol 7 as the only products in a majority of the reac-
tions performed (Scheme 4); only 5a afforded the monoester 6a at room
temperature. It was proved that the free nucleobase and 7 result from
the following cascade of spontaneous transformations initiated by the de-
pivaloylation of 5 (Scheme 5, path a): (i) the intramolecular N -O acetyl

Nꢁ-Acetyl Aza-Analogues of Ganciclovir
777
migration in monoesters 6; (ii) fragmentation of the O-acetyl-Oꢁ-pivaloyl NH-
intermediates 15 to the free nucleobase (BH) and O-acetyl-Oꢁ-pivaloylserinol
16; (iii) the O-N acetyl migration in 16; and (iv) de-pivaloylation of N -
acetyl-O-pivaloylserinol 17 leading to N -acetylserinol 7. The selective trans-
formation of 16 into 17 (the precursor of 7) represents an example of
a structural competition[7b,30] in the N -acyl-(β-hydroxyalkyl)amine-O-acyl-
(β-hydroxyalkyl)amine interconversion. The transformation results from
predomination of the O-N acetyl migration in 16 over the alternative O-N
pivaloyl one under the examined conditions. Occurrence of the O-N pi-
valoyl migration under the ammonolysis conditions has been confirmed
by the formation of N -pivaloylserinol 19 from O,Oꢁ-dipivaloylserinol hy-
drochloride 9 (Scheme 6). The observation that the O-N acyl migration
can occur selectively when O-acyl groups of a starting amino-diester, for in-
stance of a derivative relative to 16, are significantly different in steric bulk
can be a clue to effective design a synthesis. The literature data on the se-
lective O-N acyl migration in unsymmetrical diesters of serinol is rather
limited. Among the known compounds of this type,[31] only 1-acetoxy-
2-amino-3-benzoyloxy-1-phenylpropane hydrochloride and 1-benzoyloxy-2-
amino-3-acetoxy-1-phenylpropane hydrochloride were examined in this re-
spect.[31a] The compounds underwent the O-N acetyl migration to give the
corresponding benzyloxyacetamides under alkaline conditions.
The molecular modeling supports the postulate that the proclivity of
6 to undergo N ꢁ-deacetylation via the intramolecular N-O acyl migration
under basic conditions may be attributed to the formation of the monocyclic
1,3-oxazolidin-2-olate intermediate (Figure 2, entry 1). It is clear that this
intermediate is much less strained, that is, it is much easier to be formed,
than the fused or bridged bicyclic one required for the N ꢁ-deacylation of 3
or 25, respectively, by the same mode.
In light of the reports on N -N acyl migrations in nucleoside analogues
with an N -acyl-(β-aminoalkyl)amine backbone,[32] the findings on the li-
ability of 6 are expected to be of interest for investigators involved in
the synthesis of nucleoside analogues with an N -acyl-(β-hydroxyalkyl)amine
backbone, for example, as potential antimicrobials,[32a] intermediates for
preparing of novel anticancer agents,[33] or precursors of peptide nucleic
acid monomers.[34] The results show that possible proclivity of compounds
from this class to undergo N -O acyl migration should be considered before
their treatment with ammonium hydroxide or potassium carbonate, that is,
with agents typically used for removal of protecting groups from nucleoside
derivatives.
On the other hand, the transformations leading to liberation of the
nucleobase from 6 may be of interest for whose who are involved in cre-
ation of novel prodrug strategies. From a structural point of view, N -acyl-
(β-hydroxyalkyl)aminomethyl functionality, when linked to the NH acidic

778
M. Koszytkowska-Stawin´ska
group of a drug, can be considered as a useful promoiety. The intramolecu-
lar N -O acyl migration would be expected to initiate the interconversion
cascade leading to release of the parent drug from the resultant N -(β-
acyloxyalkyl)aminomethyl-containing intermediate. This prodrug concept
is expected to be worthy of consideration as a supplement to the current
prodrug strategies for drugs containing the NH acidic group.[11,35]
EXPERIMENTAL
General
High resolution mass spectra (Electrospray Ionisation, ESI) were per-
formed on a Mariner spectrometer in a positive ionization mode. The in-
frared (IR) spectra were recorded on a Specord M80 (Carl-Zeiss Jena) spec-
1
trometer; absorption maxima (νmax) are given in cm−1. H- and 13C-NMR
spectra were recorded on a Varian Gemini 200 spectrometer (1H at 200 MHz,
13C at 50 MHz). 1H and 13C chemical shifts are reported in ppm relative to
the solvent signals: CDCl3, δH(residual CHCl3) 7.26 ppm, δC 77.16 ppm) or
DMSO-d6, δH(residual DMSO) 2.50 ppm, δC 39.52 ppm; signals are quoted
as s (singlet), d (doublet), dd (doublet of doublets), dtt (doublet of triplets
of triplets), m (multiplet), br s (broad singlet), and q (quartet). Coupling
constants J are reported in Hz. Precoated Merck silica gel 60 F254 (0.2 mm)
plates were used for thin-layer chromatography (TLC), and the spots were
detected under UV light (254 nm). Column chromatography was performed
using silica gel (200–400 mesh, Merck). Anhydrous MgSO4 was employed
as a drying agent. Solvents were distilled off under reduced pressure on a
rotary evaporator.
1. N-[1,3-bis(pivaloyloxy)prop-2-yl]acetamide (10)
A mixture of serinol hydrochloride 8 (8.72 g, 68 mmol) and pivaloyl
chloride (16.44 g, 136 mmol, 16.8 cm3) was heated on an oil bath at 90◦C
for 4 hours. When the evolution of hydrogen chloride ceased, the mixture
became solid. The crude 9 was kept at a vacuum desiccator over KOH for
1 day. Yield 18.83 g (93%, crude). Analytical sample was crystallized from
dry acetone. δH (DMSO-d6) 1.15 (18H, s) 3.55–3.78 (1H, m), 4.06–4.37 (4H,
m), 8.77 (3H, br s). δC (DMSO-d6) 26.88, 48.06, 58.66, 61.05, 177.26. νmax
(KBr) 3648, 2928, 2856, 1728, 1460, 1156. HRMS m/z 260.1856 calcd for
C13H26NO4 (M+H)+, found 260.1861.
A mixture of the crude 9 (17.93 g, 61 mmol), dry pyridine (120 cm3)
and acetic anhydride (12.25 g, 120 mmol, 11.3 cm3) was kept at room tem-
perature for 1 day. The mixture was poured into cold water (30 cm3). The
organic phase was separated and the aqueous phase was extracted with chlo-
roform (3 × 30 cm3). The extracts and the organic phase were combined,
washed with water and dried. The solvent was distilled off. Product 10 was

Nꢁ-Acetyl Aza-Analogues of Ganciclovir
779
isolated from the residue by column chromatography (chloroform). Yield
48% (8.83 g, m.p. 56–58◦C). δH (CDCl3) 1.18 (18H, s), 1.97 (3H, s), 4.04
(2H, dd, 2J 11.4, 3J 5.5), 4.21 (2H, dd, 2J 11.4, 3J 5.9), 4.51 (1H, dtt, 2J 8.6, 3J
5.9, 3J 5.5), 5.81 (1H, d, 2J 8.6). δC (CDCl3) 23.42, 27.26, 39.04, 47.94, 62.72,
169.89, 178.53. νmax (KBr) 2928, 1732, 1644, 1460, 1376, 1152. HRMS m/z
324.1781 calcd for C15H27NO5Na (M+Na)+, found 324.1790.
2. N-(1,3-dipivaloyloxyprop-2-yl)-N-(pivaloyloxymethyl)acetamide (4)
A mixture of 10 (3.0 g, 10 mmol), sodium hydride (60% suspension
in mineral oil, 1.0 g, 25.0 mmol) and dry dimethylformamide (DMF) (40
cm3) was stirred at room temperature for 1 hour and then cooled on an ice
bath. Chloromethyl pivalate (6.89 g, 46 mmol, 4.9 cm3) was added to the
resulting suspension in one portion. After 4 days of stirring at room temper-
ature the mixture was poured into ice water (120 cm3) and the phases were
separated. The aqueous phase was extracted with ethyl acetate (4 × 30 cm3).
Extracts and the organic phase were combined, washed with brine and dried.
The low boiling solvents were distilled off and the residual DMF was removed
under vacuum (80◦C, 0.05 mmHg). Product 4 was isolated from the residue
by column chromatography (dichloromethane). Yield 40% (1.69 g, m.p.
43–45◦C). δH (CDCl3) 1.18 (18H, s), 1.21 (9H, s), 2.19 (3H, s), 4.16–4.31 (4H,
m), 4.88–5.09 (1H, m), 5.44 (2H, s). δC (CDCl3) 21.63, 27.10, 27.23, 38.91,
51.28, 62.13, 71.66, 172.12, 177.98. νmax (nujol) 2925, 1732, 1672, 1460, 1376,
1280, 1156. HRMS m/z 438.2462 calcd for C21H37NO7Na (M+Na)+, found
438.2483.
3. General Procedure for the Preparation of Pyrimidine Azanucleosides 5a-c
A mixture of the pyrimidine nucleobase (thymine, N 4-benzoylcytosine,
or 2-thiouracil) (2.0 mmol), N ,O-bis(trimethtlsilyl)acetamide (BSA, 815 mg,
4.0 mmol, 1.0 cm3) and dry acetonitrile (10 cm3) was stirred at room tem-
perature under an argon atmosphere for 1 hour, and then a solution of 4
(416 mg, 1.0 mmol) in dry acetonitrile (1 cm3) and neat trimethylsilyl trifluo-
romethanesulfonate (TMSOTf, 370 mg, 1.7 mmol, 0.3 cm3) were added. The
mixture was left for 2 days at room temperature, and then ethyl acetate (50
cm3) and a saturated solution of sodium bicarbonate (1 cm3) were added.
The mixture was stirred for 1 hour, and then filtered through a Celite pad.
The organic phase was separated, washed with brine and dried. The solvent
was distilled off. Product 5a (5b or 5c) was isolated from the residue by
column chromatography; the eluting solvents are given bellow.
3.1. 4-(Benzoylamino)-1-[N-(1,3-dipivaloyloxyprop-2-yl)acetylaminomethyl]-1H-
pyrimidin-2-one (5a). According to the general procedure, 5a was obtained
from 4 and N 4-benzoylcytosine. Eluting solvent: CHCl3/acetone, 97/3, v/v.
Yield 88% (466 mg, oil). δH (CDCl3) 1.12 (18H, s), 2.21 (3H, s), 4.21–4.48
(5H, m), 5.60 (2H, s), 7.44–7.62 (4H, m), 7.87–7.92 (2H, m), 8.31 (1H, d,
3J 7.6), 9.33 (1H, br s). δC (CDCl3) 22.45, 27.17, 38.89, 52.37, 55.52, 61.07,

780
M. Koszytkowska-Stawin´ska
97.62, 127.80, 129.11, 133.29, 133.36, 150.07, 155.93, 162.70, 174.42, 177.97.
νmax (nujol) 2906, 1724, 1660, 1624, 1460, 1376, 1292, 1164. HRMS m/z
551.2476 calcd for C27H36N4O7Na (M+Na)+, found 551.2501.
3.2. 1-[N-(1,3-dipivaloyloxyprop-2-yl)acetylaminomethyl]-5-methyl-1H,3H-pyri
midin-2,4-dione (5b). According to the general procedure, 5b was obtained
from 4 and thymine. Eluting solvent: CHCl3/acetone, 95/5, v/v. Yield 81%
4
(356 mg, m.p. 137–140◦C). δH (CDCl3) 1.14 (18H, s), 1.88 (3H, d, J 1.2),
2.21 (3H, s), 4.16–4.58 (5H, m), 5.40 (2H, s), 7.71 (1H, q, 4J 1.2), 8.66 (1H, br
s). δC (CDCl3) 12.44, 22.48, 27.10, 38.86, 50.53, 55.28, 60.89, 111.702, 140.65,
151.65, 164.04, 174.31, 177.90. νmax (nujol) 3648, 2924, 1736, 1688, 1632,
1460, 1376, 1178. HRMS m/z 462.2211 calcd for C21H33N3O7Na (M+Na)+,
found 462.2229.
3.3. 1-[N-(1,3-dipivaloyloxyprop-2-yl)acetylaminomethyl]-5-methyl-1H,3H-pyri
midin-4-on-2-thione (5c). According to the general procedure, 5c was ob-
tained from 4 and 2-thiouracil. Eluting solvent: CHCl3/acetone, 95/5, v/v.
Yield 43% (188 mg, oil). δH (CDCl3) 1.12 (18H, s), 2.21 (3H, s), 4.17–4.91
(5H, m), 5.96 (1H, d, 3J 8.0), 6.03 (2H, s), 7.88 (1H, d, 3J 8.0) 10.93 (1H, br
s). δC (CDCl3) 22.49, 27.19, 38.88, 55.97, 60.85, 62.69, 108.02, 144.55, 159.82,
174.45, 176.63, 177.91. νmax (nujol) 2908, 1732, 1680, 1460, 1376, 1284, 1148.
HRMS m/z 464.1826 calcd for C20H31N3O6NaS (M+Na)+, found 464.1845.
4. 2-(Acetylamino)-6-(diphenylcarbamoyl)-9-[N-(1,3-dipivaloyloxyprop-
2-yl)acetylaminomethyl]-9H-purine (5d)
A mixture of N 2-acetyl-O6-(diphenylcarbamoyl)guanine (780 mg,
2 mmol), N ,O-bis(trimethtlsilyl)acetamide (BSA, 814 mg, 4 mmol, 1 cm3)
and dry ethylene chloride (10 cm3) was heated at 80◦C in a sealed tube
under an argon atmosphere for 15 min. The solvent was removed under
vacuum at 40◦C (0.05 mmHg), and the residue was dissolved in dry toluene
(20 cm3). A solution of 4 (416 mg, 1.0 mmol) in dry toluene (1 cm3) and
neat trimethylsilyl trifluoromethanesulfonate (TMSOTf, 370 mg, 1.7 mmol,
0.3 cm3) were added. The mixture was heated at 80◦C for 1 hour, and then
cooled to room temperature. Ethyl acetate (50 cm3) and saturated solu-
tion of sodium bicarbonate (1 cm3) were added. The resulting mixture was
stirred for 1 hour at room temperature and filtered through a Celite pad.
The organic phase was separated from the filtrate and washed with water,
brine and dried. The solvent was distilled off. Product 5d was isolated from
the residue by column chromatography (chloroform/acetone, 95/5, v/v).
Yield 47% (331 mg, foam). δH (CDCl3) 1.04 (18H, s), 2.18 (3H, s), 2.41 (3H,
s), 4.22–4.31 (2H, m), 4.39–4.50 (3H, m), 5.64 (2H, s), 7.23–7.44 (10H, m),
8.28 (1H, br s), 8.47 (1H, s). δC (CDCl3) 22.13, 25.12, 27.06, 38.78, 48. 18,
55.42, 61.04, 127.01, 129.24, 141.85, 146.40, 150.30, 151.94, 156.28, 169.49,
173.60, 177.95. νmax (nujol) 2900, 1723, 1656, 1632, 1448, 1370. HRMS m/z
724.3065 calcd for C36H43N7O8Na (M+Na)+, found 724.3073.

Nꢀ-Acetyl Aza-Analogues of Ganciclovir
781
5. Reaction of 5a-d with Ammonium Hydroxide at 70◦C (Scheme 4, Path i)
A mixture of 5a-d (1 mmol), concentrated ammonium hydroxide
(20 cm3) and methanol (20 cm3) was kept at 70◦C (in a sealed tube) for
1 day. Then the reaction mixture was evaporated to dryness and the residue
was treated with methanol (15 cm3). The resulting suspension was filtered;
the solid was washed with cold methanol (3 × 10 cm3) and dried in the
air to give the corresponding nucleobase in the yields given in Scheme 4.
The filtrate and the washings were combined and evaporated to dryness.
N -Acetylserinol 7 was isolated from the residue by column chromatogra-
phy (chloroform-methanol, 95/5, v/v); the yields are given in Scheme 4;
1H MNR spectrum was consistent with that reported in the literature.[36] δC
(DMSO-d6) 22.79, 52.89, 60.20, 169.15.
6. Reaction of 5a-d with Ammonium Hydroxide at Room Temperature
(Scheme 4, Path ii)
A mixture of 5a-d (1 mmol), concentrated ammonium hydroxide
(20 cm3) and methanol (20 cm3) was kept at room temperature for 1 day.
Then the reaction mixture was evaporated to dryness and the residue was
treated with methanol (15 cm3). The resulting suspension was filtered; the
solid was washed with cold methanol (3 × 10 cm3) and dried in the air to
give the corresponding nucleobase in the yields given in Scheme 4. The
filtrate and the washings were combined and evaporated to dryness. The
residue was passed through a chromatographic column.
6.1. Chromatography of the residue from the reaction of 5a (chloroform-
Methanol, 99/1, v/v) afforded 5e (12%, 51 mg, oil), 6a (4%, 14 mg, oil),
and N-acetylserinol 7 (58%)
6.1.1.
4-(Amino)-1-[N-(1,3-dipivaloyloxyprop-2-yl)acetylaminomethyl]-1H-
pyrimidin-2-one (5e). δH (CDCl3) 1.12 (18H, s), 2.17 (3H, s), 4.19–4.39 (5H,
3
m), 5.44 (2H, s), 5.73 (1H, d, J 7.2), 6.56 (1H, br s), 6.60 (1H, br s), 7.90
3
(1H, d, J 7.2). δC (CDCl3) 22.47, 27.18, 38.88, 55.41, 61.29, 61.97, 95.43,
146.56, 156.76, 165.93, 174.27. νmax (nujol) 2908, 1721, 1627, 1426, 1355,
1141. HRMS m/z 447.2233 calcd for C20H32N4O6Na (M+Na)+, found
447.2214.
6.2.2. 4-Amino-1-{N -[1-hydroxy-3-(pivaloyloxymethyl)prop-2-yl]acetylam
inomethyl}-1H-pyrimidin-2-one (6a). δH (CDCl3) 1.15 (9H, s), 2.20 (3H, s),
3.64–4.31 (6H, m), 5.21 (1H, d, 2J 13.6), 5.36 (1H, d, 2J 13.6), 5.74 (1H, d, 3J
7.2), 6.52 (1H, br s), 6.56 (1H, br s), 7.86 (1H, d, 3J 7.2). δC (CDCl3) 22.77,
27.19, 38.85, 53.53, 59.14, 59.76, 61.93, 95.54, 146.59, 157.30, 166.15, 175.17.
νmax (nujol) 2915, 1721, 1618, 1434, 1366, 1158. HRMS m/z 363.1639 calcd
for C15H24N4O5Na (M+Na)+, found 363.1639.
6.2. Chromatography of the residue from the reaction of 5b-d
(chloroform-Methanol, 95/5, v/v) gave N-acetylserinol 7 in the yields given
in Scheme 4.

782
M. Koszytkowska-Stawin´ska
7. Reaction of 9 with Ammonium Hydroxide
(Scheme 6, Reaction i(a) or ii(a))
A mixture of 9 (2 mmol), concentrated ammonium hydroxide (40 cm3)
and methanol (40 cm3) was kept for 1 day at room temperature (Scheme 6,
reaction ia) or at 70◦C (Scheme 6, reaction iia). The reaction mixture was
evaporated to dryness. Column chromatography of the residue (chloroform-
methanol, 99/1 → 8/2, v/v)afforded 19 (184 mg, 51%, reaction ia; 233 mg,
65% reaction iia) and 22 (151 mg, 28%, reaction ia; 58 mg, 11% reaction
iia).
7.1. N-(1,3-Dihydroxypropan-2-yl)pivalamide (19). A white solid, m.p.
2
59–61◦C. δH (CDCl3) 1.17 (9H, s), 3.63 (2H, ABX, X = H(2), J AB 11.0,
3J AX 5.3), 3.75 (2H, ABX, X = H(2), 2J AB 11.0, 3J BX 4.1), 3.82–3.96 (1H, m),
3
4.35 (2H, br s, OH), 6.51 (1H, 6.30 (d, J HH 7.6, NH). δC (CDCl3) 27.57,
38.94, 52.46, 61.85, 180.05. νmax (KBr) 3342, 3237, 1632, 1530, 1397, 1364.
HRMS m/z 198.1102 calcd for C8H17NO3Na (M+Na)+, found 198.1097.
7.2. 3-Hydroxy-2-pivalamidopropyl pivalate (22). A colorless oil. δH
(CDCl3) 1.10 (9H, s), 1.11 (9H, s), 3.50 (1H, ABX, X = H(2), 2J AB 11.4, 3J AX
4.5), 3.61 (1H, ABX, X = H(2), 2J AB 11.4, 3J BX 3.5), 4.41–4.22 (3H, m), 6.30
3
(1H, d, J HH 7.0, NH). δC (CDCl3) 27.11, 27.43, 38.73, 38.83, 50.55, 61.54,
62.86, 179.05, 179.24. νmax (KBr) 3363, 1728, 1640, 1524, 1397, 1367, 1155.
HRMS m/z 282.1645 calcd for C13H25NO4Na (M+Na)+, found 282.1653.
8. Reaction of 9 with Ammonium Hydroxide in the Presence of AcNH2
(Scheme 6, Reaction i(b) or ii(b))
A mixture of 9 (2 mmol) and acetamide (2 mmol), concentrated am-
monium hydroxide (40 cm3) and methanol (40 cm3) was kept for 1 day at
room temperature (Scheme 6, reaction ib) or at 70◦C (Scheme 6, reaction
iib). The reaction mixture was evaporated to dryness. The solid residue was
shaken with chloroform (25 cm3) and filtered. The white solid (AcNH2) was
washed with chloroform (10 cm3) and dried. The filtrate was evaporated
to dryness. Column chromatography of the residue (chloroform-methanol,
99/1 → 8/2, v/v) gave 19 (176 mg, 49%, reaction ib; 223 mg, 62% reac-
tion iib) and 22 (156 mg, 29%, reaction ib; 59 mg, 11% reaction iib). The
NMR spectra of the products 19 and 22 were consistent with those described
previously. The NMR spectra (DMSO-d6) of the recovered AcNH2 showed:
δH 1.76 (3H, s), 6.74 (1H, br s, NH) and 7.32 (1H, br s, NH); δC 22.55 and
172.06.
9. 2-Acetamido-3-hydroxypropyl pivalate (17)
A mixture of 10 (550 mg, 1.8 mmol), concentrated ammonium hydrox-
ide (40 cm3) and methanol (40 cm3) was kept for 1 day at room tempera-
ture. The reaction mixture was evaporated to dryness and the oil residue was
passed through a short chromatographic column. The column was eluted
with a chloroform-methanol mixture (95/5, v/v, 500 cm3) to yield 17 (85 mg,

Nꢁ-Acetyl Aza-Analogues of Ganciclovir
783
23%, oil) and then with methanol (200 cm3) to give 7 (111 mg, 46%). δH
(CDCl3) 1.16 (9H, s), 1.96 (3H, s), 3.47–3.85 (3H, m), 4.12–4.24 (3H, m),
3
6.46 (d, J HH 6.2, 1H, NH). δC (CDCl3) 23.23, 27.19, 38.96, 50.44, 61.49,
62.51, 170.98, 179.22. νmax (KBr) 3287, 1715, 1650, 1543, 1399, 1370, 1161.
HRMS m/z 240.1207 calcd for C10H19NO4Na (M+Na)+, found 240.1204.
REFERENCES AND NOTES
1. De Clercq, E. Antiviral drugs in current clinical use. J. Clin. Virol. 2004, 30, 115–133.
2. a) Gawin, R.; De Clercq, E.; Naesens, L.; Koszytkowska-Stawin´ska, M. Synthesis and antiviral evaluation
of acyclic azanucleosides developed from sulfanilamide as a lead structure. Bioorg. Med. Chem. 2008,
16, 8379–8389; b) Koszytkowska-Stawin´ska, M.; Sas, W.; Kaleta, K.; De Clercq, E. Synthesis and antiviral
properties of aza–analogues of acyclovir. Nucleos. Nucleot. Nucl. Acids 2007, 26, 51–64.
3. Koszytkowska-Stawin´ska, M.; Sas, W.; De Clercq, E. Synthesis of aza-analogues of Ganciclovir. Tetrahe-
dron 2006, 62, 10325–10331.
4. Koszytkowska-Stawin´ska, M.; Kołaczkowska, E.; Adamkiewicz, E.; De Clercq, E. Synthesis and antivi-
ral properties of aza-analogues of ganciclovir derived from 5,5-bis(hydroxymethyl)pyrrolidin-2-one.
Tetrahedron 2007, 63, 10587–10595.
5. a) Myers, A.G.; Yang, B.H.; Chen, H.; McKinstry, L.; Kopecky, D.J.; Gleason, J.L. Pseudoephedrine
as a practical chiral auxiliary for the synthesis of highly enantomerically enriched carboxylic acids,
alcohols, aldehydes, and ketones. J. Am. Chem. Soc. 1997, 119, 6496–6511; b) Myers, A.G.; Gleason, J.L.
Asymetric synthesis of α-amino acids by the alkylation of pseudoephedrine glycinamide: L-allylglycine
and N-Boc-L-allylglycine. Org. Synth. 2004, Coll. Voll. 10, 12.
6. Phillips, A.; Baltzly, R. Rearrangements between primary ethanolamides of carboxylic acids and the
corresponding aminoethylesters. J. Am. Chem. Soc. 1947, 69, 200–204.
7. a) Markey, S.P.; Dudding, T.; Wang, T-C.L. Base- and acid-catalyzed interconversions of O-acyl- and
N-acyletanolamines: a cautionary note for lipid analyses. J. Lipid Res. 2000, 41, 657–662; b) Van
Overloop, H.; Van der Hoeven, G.; Van Veldhoven, P.P. N-Acyl migration in ceramides. J. Lipid Res.
2005, 46, 812–816.
8. Eberhard, H.; Seitz, O. N→O-Acyl migration shift in Fmoc-based synthesis of phosphopeptides. Org.
Biomol. Chem. 2008, 6, 1349–1355.
9. a) Hamada, Y.; Ohtake, J.; Sohma, Y.; Kimura, T.; Hayashi, Y.; Kiso, Y. New water-soluble prodrugs
of HIV protease inhibitors based on O→N intramolecular acyl migration. Bioorg. Med. Chem. 2002,
10, 4155–4167; b) Hamada, Y.; Matsumoto, H.; Kimura, T.; Hayashi, Y.; Kiso, Y. Effect of the acyl
groups on O→N acyl migration in the water-soluble prodrugs of HIV-1 protease inhibitor. Bioorg.
Med. Chem. Lett. 2003, 13, 2727–2730; c) Hamada, Y.; Matsumoto, H.; Yamaguchi, S.; Kimura, T.;
Hayashi, Y.; Kiso, Water-soluble prodrugs of dipeptide HIV protease inhibitors based on O→N acyl
migration: design, synthesis and kinetic study. Bioorg. Med. Chem. 2004, 12, 159–170.
10. Skwarczynski, M.; Sohma, Y.; Kimura, M.; Hayashi, Y.; Kimura, T.; Kiso, Y. O,N intramolecular acyl
migration strategy in water-soluble prodrugs of Taxoids. Bioorg. Med. Chem. Lett. 2003, 13, 4441–
4444.
11. Guarino, V.R.; Stella, V.J. Prodrugs of amides, imides and other NH acid compounds. In: Stella,
V.J.; Borchardt, R.T.; Hageman, M.; Oliyai, R.; Maag, H.; Tilley, J., eds., Prodrugs: Challenges and
Rewards-Part 2, AAPS Press/Springer, New York, 2007.
12. a) Xiong, J.; Zhu, H.-F.; Zhao, Ya-J.; Lan, Y.-J.; Jiang, Ji-W.; Yang, J.-J.; Zhang, S.-F. Synthesis and antitu-
mor activity of amino acid ester derivatives containing 5-fluorouracil. Molecules 2009, 14, 3142–3152;
b) Madec-Lougerstay, R.; Florent, J.-C.; Monneretm, C. Synthesis of self-immolative glucuronide spac-
ers based on aminomethylcarbamate. Application to 5-fluorouracil prodrugs for antibody-directed
enzyme prodrug therapy. J. Chem. Soc., Perkin Trans. 1, 1999, 1369–1376; c) Nishitani, T.; Iwasaki,
T.; Mushika, Y.; Miyoshi, M. Synthesis of 2-(Pyrimidin-l-yl)- and 2-(Purin-9-yl)-2-amino Acids. J. Org.
Chem. 1979, 44, 2019–2023.
13. Koszytkowska-Stawin´ska, M.; Sas, W. Facile synthesis of acyclic azanucleosides from N-
pivaloyloxymethyl amides and sulfonamides: synthesis of aza-analogues of Ganciclovir. Tetrahedron
Lett. 2004, 45, 5437–5440.

784
M. Koszytkowska-Stawin´ska
14. Bordwell, F.R. Equilibrium acidities in dimethyl sulfoxide solution. Acc. Chem. Res. 1988, 21, 456–463.
15. a) Vorbru¨ggen, H.; Ruh-Pohlenz, C. Handbook of Nucleoside Synthesis, John Wiley: New York, 2001.
16. a) Altmann, K-H. Pyrrolidine based analogs of 3ꢁ-deoxythymidine. Tetrahedron Lett. 1993, 34,
7721–7724; b) Rassu, G.; Pinna, L.; Ulgheri, F.; Casiraghi, G. Total syntheses of N-boc-protected
3ꢁ-deoxy-4ꢁ-azathymidine and 4ꢁ-azauridine. Tetrahedron Lett. 1994, 35, 4019–4022; c) Pickering,
L.; Malhi, B.S.; Coe, P.L.; Walker, R.T. The synthesis of 4ꢁ-t-butylcarbamyl- and 4ꢁ-p-
toluenesulphonamidyl-2ꢁ,3ꢁ-dideoxy pyrimidine nucleoside analogues. Nucleosides Nucleotides 1994,
13, 1493–1506.
17. Chimishkyan, A.L.; Snagovskii, Yu.S.; Gulyaev, N.D.; Leonova, T.V.; Kusakin, M.S. Transamination
characteristics of substituted ureas. III. Confirmation of the mechanism via kinetic studies. Zh. Org.
Khim. 1985, 21, 2136–2143.
18. Murakami, Y.; Kondo, K.; Miki, K.; Akiyama, Y.; Watanabe, T.; Yokoyama, Y. The ortho-
Substituted N,N-Diacetylaniline as a Selective Acetylating Reagent. Tetrahedron Lett. 1997, 38, 3751–
3754.
19. Marvell, E.; Nelson, H.; Joncich, M.; Geiszler, A.; Williams, M. Kinetics of the Alcoholysis of o- and
p-Nitroacetanilides. J. Am. Chem. Soc. 1957, 79, 1076–1078.
20. Verkade, P.E.; Witjens, P.H. Deacetylation of acetamino compounds containing a nitro group in the
orto or para position. Recl. Trav. Chim. Pay B 1943, 62, 201–204.
21. a) Kawaguchi, S.; Araki, K. Mild, rapid and selective alcoholysis of terpyridine-appended amide
substrates by Cu2+-catalysis: protonation state and reactivity of the complex. Inorg. Chem. Acta, 2005,
358, 947–956; b) Brown, R.S.; Neverov, A.A. Acyl and phosphoryl transfer to methanol promoted by
metal ions. J. Chem. Soc., Perkin Trans. 2 2002, 1039–1049.
22. Anelli, P.L.; Brocchetta, M.; Canipari, S.; Losi, P.; Manfredi, G.; Tomba, C. Synthesis and hydrolytic
lability of α-phenoxyacetamides containing hydroxy groups in the N-alkyl residue. Gazz. Chim. Ital.
1997, 127, 135–142.
23. Varaprasad, C.V.; Averett, D.; Ramasamy, K.S. Synthesis and structural studies of monocyclic 4ꢁ-aza-
L-nucleosides. Tetrahedron 1999, 55, 13345–13368.
24. Fersht, A.R. Acyl-transfer reactions of amides and esters with alcohols and thiols. A reference system
for the serine and cysteine proteinases. Concerning the N promotion of amides and amide-imidate
equilibria. J. Am. Chem. Soc. 1971, 93, 3504–3515.
25. Carey, F.A.; Sundberg, R.J. Advanced Organic Chemistry, 3rd ed. 1990, Plenum Press, New York, Part
A, pp. 439.
26. Hammond, G.S. A correlation of reaction rates. J. Am. Chem. Soc. 1955, 77, 334–338.
27. Ruzicka, L. The many-membered carbon rings. Chem. Ind. (London) 1935, 13, 2–8.
28. The ChemBio3D (Ultra 11.0) program was used for the MM2 calculations.
29. In the considered heterocyclic systems C, the total strain order is consisted with that established
for the reference hydrocarbons: monocyclic < fused bicyclic < bridged bicyclic; Liebman, J.F.;
Greenberg, A. A survey of strained organic molecules. Chem. Rev. 1976, 76, 311–365.
30. McCasland, G.E. N→O Acyl migration in epimeric acetyl inosamines. J. Am. Chem. Soc. 1955, 73,
2295–2297.
31. a) Fodor, G.; Kiss, J.; Sallay, I. Configuration Correlation of Chloramphenicol with nor-ψ-
ephedrine. J. Chem. Soc. 1951, 1858–1863; b) Bushnev, A.S.; Struchkova, M.I.; Zvonkova, E.N.;
Evstigneeva, R.P.; Preobrazhenskii, N.A. Infrared absorption spectra and spatial structure of ce-
ramide and its derivatives. Zh. Org. Khim. 1971, 7, 2132–2136; c) Bushnev, A.S.; Zvonkova, E.N.;
Evstigneeva, R.P. Synthesis of 2-heptadecyl-4-(1-benzoyloxy)hexadecyl-2-oxazoline. Zh. Org. Khim.
1971, 7, 2313–2316; d) Zvonkova, E.N.; Bushnev, A.S.; Mitsner, B.I.; Evstigneeva, R.P. Phosphoryla-
tion of 3-O-benzoylceramide. Zh. Org. Khim. 1973, 9, 31–35.
32. a) McCarthy, O.; Musso-Buendia, A.; Kaiser, M.; Brun, R.; Ruiz-Perez, L.M.; Johanson, N.G.;
Pacanowska, D.G.; Gilbert, I.H. Design, synthesis and evaluation of novel uracil acetamide deriva-
tives as potential inhibitors of Plasmodium falciparum dUTP nucleotidohydrolase. Eur. J. Med. Chem.
2009, 44, 678–688; b) Eriksson, M.; Christensen, L.; Schmidt, J.; Haaima, G.; Orgel, L.; Nielsen, P.E.
Sequence dependent N-terminal rearrangement and degradation of peptide nucleic acid (PNA) in
aqueous solution. New J. Chem. 1998, 22, 1055–1059; c) Christensen, L.; Fitzpatrick, R.; Gildea, B.;
Petersen, K.H.; Hansen, H.F.; Koch, T.; Egholm, M.; Buchardt, O.; Nielsen, P.E.; Coull, J.; Berg, R.H.
Solid-Phase synthesis of peptide nucleic acids. J. Peptide Sci. 1995, 1, 175–183.

Nꢁ-Acetyl Aza-Analogues of Ganciclovir
785
33. Hadj-Bouazza, A.; Teste, K.; Colombeau, L.; Chaleix, V.; Zerrouki, R.; Kraemer, M.; Catherine, O.S.
Synthesis and biological activity of mustard derivatives of thymine. Nucleos. Nucleot. Nucl. Acids 2008,
27, 439–448.
34. a) Katritzky, A.R.; Narindoshvili, T. Chiral peptide nucleic acid monomers (PNAM) with mod-
ified backbones. Org. Biomol. Chem. 2008, 6, 3171–3176; b) Murata, A.; Wada, T. Synthesis of a
novel ester analog of nucleic acids bearind a serine backbone. Bioorg. Med. Chem. Lett. 2006, 16,
2933–2936; c) Hamzavi, R.; Dolle, F.; Tavitian, B.; Dahl, O.; Nielsen, P.E. Modulation of the phar-
macokinetic properties of PNA: preparation of galactosyl, mannosyl, fucosyl, N-acetylgalactosaminyl,
and N-acetylglucosaminyl derivatives of aminoglycine peptide nucleic acid monomers and their in-
corporation into PNA oligomers. Bioconiugate Chem. 2003, 14, 941–954; d) Bajor, Z.; Sa´gi, G.; Tegyey,
Z.; Kraicsovits, F. PNA-DNA chimeras containing 5-alkynyl-pyrimidine PNA units. Synthesis, Binding
properties, and enzymatic stability. Nucleos. Nucleot. Nucl. Acids 2003, 22, 1963–1983; e) Rana, V.S.;
Kumar, V.A.; Ganesh, K.N. Oligonucleotides with (N-thymin-1-ylacetyl)-1-arylserinol backbone: chi-
ral acyclic analogs with restricted conformational flexibility. Tetrahedron 2001, 57, 1311–1321.
35. Krise, J.P.; Oliyai, R. Prodrugs of amines. In: Stella, V.J.; Borchardt, R.T.; Hageman, M.; Oliyai, R.;
Maag, H.; Tilley, J. Eds., Prodrugs: Challenges and Rewards-Part 2, AAPS Press/Springer, 2007.
36. Harada, M.; Moire, T.; Hirokawa, Y.; Kato, S. An efficient synthesis of 6-substituted aminohexahydro-
1H-1,4-diazepines from 2-substituted aminopropenals. Chem. Pharm. Bull. 1996, 44, 2205–2212.