A convenient protection for 4-oxopyrimidine moieties in nucleosides by the pivaloyl group

Article abstract of DOI:10.1135/cccc2009084

Application of the pivaloyl group as a protection for the N3 position of thymidine and uridine was investigated. Pivaloylation of thymidine is a very rapid reaction proceeding under mild conditions with excellent regioselectivity for sugar or thymine moiety, depending on the amines used. Several pivaloylated thymidine derivatives were obtained by treatment of unprotected thymidine with pivaloyl chloride under various experimental conditions. Stability of the N3-pivaloyl protecting group under basic and acidic conditions was evaluated and the conditions for its selective removal were found.

Full text of DOI:10.1135/cccc2009084

Protection for 4-Oxopyrimidine Moieties
33
A CONVENIENT PROTECTION FOR 4-OXOPYRIMIDINE MOIETIES
IN NUCLEOSIDES BY THE PIVALOYL GROUP+
Michal SOBKOWSKI
Institute of Bioorganic Chemistry, Polish Academy of Sciences, Noskowskiego 12/14,
61-704 Poznań, Poland; e-mail: msob@ibch.poznan.pl
Received Jun e 29, 2009
Accepted August 31, 2009
Publish ed on lin e February 8, 2010
Application of th e pivaloyl group as a protection for th e N3 position of th ym idin e an d
uridin e was in vestigated. Pivaloylation of th ym idin e is a very rapid reaction proceedin g un -
der m ild con dition s with excellen t regioselectivity for sugar or th ym in e m oiety, depen din g
on th e am in es used. Several pivaloylated th ym idin e derivatives were obtain ed by treatm en t
of un protected th ym idin e with pivaloyl ch loride un der various experim en tal con dition s. Sta-
bility of th e N3-pivaloyl protectin g group un der basic an d acidic con dition s was evaluated
an d th e con dition s for its selective rem oval were foun d.
Keyw ord s: Th ym idin e; Uridin e; Pivaloyl; Acylation s; Protectin g groups; Regioselectivity.
In preparation of n ucleoside derivatives, th e im ido fun ction ality in
4-oxopyrim idin e m oieties is often left un protected due to its relatively low
reactivity. However, for a n um ber of reaction s aim ed at sugar2 or ph osph o-
rus3,4 m oieties, such protection is h igh ly desirable or n ecessary to preven t
nucleobase m odifications. The N3H function contains the m ost acidic hydro-
gen in th ym idin e or uridin e m olecules (pKa 9.2–10.1, depen din g on sub-
stituen ts5), an d h en ce is pron e to react with soft electroph iles, e.g. in
alkylation with m ild alkylation agen ts6,7 or in th e reaction s un der th e
Mitsun obu con dition s8,9. Addition ally, in th e deproton ated th ym in e m oi-
ety th e n egative ch arge is delocalized, m akin g th e O4 fun ction also reactive.
Derivatization of th is position of th ym idin e (or uridin e) was observed in
con den sation s in volvin g azole derivatives (e.g. MSNT++)10–12 or triazole13,14
+ A prelim in ary com m un ication of th is work was presen ted at th e XIVth Symposium on the
Chemistry of Nucleic Acid Components, Český Krumlov, Czech Republic, June 2008; see ref.1
++Abbreviation s: DMAP, N,N-dim eth ylam in opyridin e; MSNT, 1-(m esityl-2-sulfonyl)-3-nitro-
1,2,4-triazole; Pv, pivaloyl (trimethylacetyl); Py, pyridine; TEA, trieth ylam in e.
Collect. Czech. Chem. Commun. 2010, Vol. 75, No. 1, pp. 33–57
© 2010 Institute of Organic Chemistry and Biochemistry
doi:10.1135/cccc2009084

34
Sobkowski:
as well as wh en aren esulfon yl ch lorides15,16a or ch loroph osph ates17,18 were
used as con den sin g agen ts. Th e resultin g O4-substituted com poun ds react
readily with n ucleoph iles, leadin g, for exam ple, in th e presen ce of am m on ia
or am in es to cytidin e derivatives13,16a,19
.
Th e protection of 4-oxopyrim idin e is realized by blockin g eith er th e O4
fun ction (with pyridyl2, aryl12,20 or p-n itroph en yleth yl21 protectin g groups)
or th e N3 position (aren esulfen yl22, acyl16,23–27 or Boc-type28 protectin g
groups). Curren tly, N3-ben zoyl protection is used routin ely, often with a
substituen t in th e ben zen e rin g in order to adjust th e stability of th e protec-
tion group7,27,29
.
Sin ce th e pion eerin g works of Todd30,31 an d Kh oran a32–34, acyl groups
(usually ben zoyl an d its derivatives, acetyl, pivaloyl, an d adam an tan e-
1-carbon yl) h ave been also com m on ly used for th e protection of carbo-
h ydrate residues in n ucleosides. Acylation of th ym idin e in pyridin e is
ch em oselective for sugar h ydroxy groups35,36; h owever, on ly bulky acyl
groups (e.g. pivaloyl37–40 or adam an tan e-1-carbon yl41) can be easily in tro-
duced in a regioselective m an n er to th e 5′ position , wh ile for less sterically
dem an din g groups special procedures are required, e.g. in volvin g activation
of th e h ydroxy group42,43 or en gagin g en zym atic reaction s44–46
.
Th e kn own m eth ods for acylation of n ucleosides, both in sugar an d
aglycon e m oieties, often require len gth y procedures an d in volve th e tran -
sien t protection –deprotection strategy47. In th is paper, application of th e
pivaloyl (trim eth ylacetyl) group as a con ven ien t protection of N3 position s
an d h ydroxy groups of th ym idin e an d uridin e is presen ted. (Apart from
ch em ical application s, pivaloylated n ucleosides are of in terest as th erapeu-
tics. For exam ple, 5′-O-pivaloylth ym idin e was foun d to act as a prodrug de-
rivative of th ym idin e for possible use in th e preven tion of toxic effects in
som e th erapies48.)
RESULTS AND DISCUSSION
Recen tly we h ave foun d th at in th e con den sation s of protected n ucleoside
3′-H-ph osph on ates with alcoh ols, th e yield of H-ph osph on ate diesters sig-
n ifican tly decreased wh en th e reaction was carried out in th e presen ce of
trialkylam in es or stron g n ucleoph ilic catalysts (e.g. DMAP)49. Wh ile such
results could be expected due to th e kn own pron en ess of H-ph osph on ates
to side reaction s in th e presen ce of stron g organ ic bases50–54, an in triguin g
observation was m ade th at n o by-products with m odified ph osph orus
groups could be detected in reaction m ixtures by 31P NMR spectroscopy. It
appeared subsequen tly th at th e m ain reason for th e observed deterioration
Collect. Czech. Chem. Commun. 2010, Vol. 75, No. 1, pp. 33–57

Protection for 4-Oxopyrimidine Moieties
35
of th e yield was partial con sum ption of pivaloyl ch loride (PvCl) used as a
con den sin g agen t49. However, in som e cases also an oth er reaction , rapid
pivaloylation of sugar an d/or n ucleobase m oiety, was detected. Th is was a
rath er surprisin g observation , as th e procedures for acylation of n ucleosides
usually require lon g (often overn igh t) reaction tim es, wh ile un der very m ild
con dition s used for th e con den sation of n ucleoside H-ph osph on ates (dilute
solution of reactan ts in DCM, 3 equiv. of an am in e, 1.2 equiv. of PvCl), n o
com petin g acylation was an ticipated.
Pivaloylation of Thymidine
In follow-up experim en ts, th e reactivity of PvCl in th e reaction s with pro-
tected th ym idin e derivatives (3′,N3-diprotected for an alysis of acylation of
5′-OH, an d 3′,5′-diprotected for N3-acylation ) was in vestigated in th e pres-
en ce of th ree am in es of differen t basicity55 an d n ucleoph ilicity: DMAP (a
stron g n ucleoph ilic catalyst; pKa 9.7), pyridin e (a m oderate n ucleoph ilic
catalyst; pKa 5.2), TEA (poor n ucleoph ilic catalyst; pKa 11.0), or TEA/Py
m ixture (Table I).
In th e presen ce of DMAP pivaloylation of th e 5′-OH group of
N3,3′-O-diben zoylth ym idin e was com plete with in a few secon ds, wh ile
N3-acylation of 3′-O-(dim eth oxytrityl)-5′-O-pivaloylth ym idin e (7) required
several m in utes for com pletion (N vs O ch em oselectivity >30:1). For
TABLE I
Th e h alf-tim es of pivaloylation an d ben zoylation at N3H an d 5′-OH of 3′,5′-O- or 3′-O,N3-
protected th ym idin e derivatives
Pivaloylation , t1/2
Ben zoylation , t1/2
Am in e
N3H
5′-OH
N3H
5′-OH
DMAP (5 equiv.)
~30 s
~10 h
~4 h
<1 s
~2 h
~0.5 h
~24 h
~1.5 h
<1 s
~10 s
~60 s
~10 s
~60 s
~60 sa
~2 s
Py (5 equiv.)
Py (62 equiv. = 50% v/v)
TEA (5 equiv.)
~5 m in
~60 s
<1 s
>4 h
~1 h a
~15 m in a
a,b
TEA (2.5 equiv.) + Py (2.5 equiv.)
TEA (5 equiv.) + Py (5 equiv.)
–
<1 s
<1 sa,c
a
b
c
Un der n itrogen . Not com plete due to side reaction s. Com plete after ca. 4 h .
Collect. Czech. Chem. Commun. 2010, Vol. 75, No. 1, pp. 33–57

36
Sobkowski:
pyridin e (5 equiv.), both reaction s were m uch slower, an d th e selectivity of
O- vs N-acylation was sign ifican tly lower (ca. 5:1). However, wh en basicity
of th e reaction m edium was in creased by raisin g th e con ten ts of pyridin e to
50% (v/v), th e selectivity for 5′-OH reach ed a value of ca. 50:1, com parable
to th at in th e presen ce of DMAP.
In th e presen ce of trieth ylam in e (TEA), com plete reversal of ch em o-
selectivity (N3 vs 5′-O) of pivaloylation was foun d an d th e acylation in th e
th ym in e m oiety was over 240 tim es faster th an in th e deoxyribose. (A sim i-
lar pattern of regioselectivity in ben zoylation of th e 5-(trifluorom eth yl) an -
alogue of th ym idin e prom oted by various am in es was n oted by Yam ash ita
et al.25) Th e rate of N3-pivaloylation (t1/2 ~ 1 m in ) could be n otably in -
creased with out affectin g h igh ch em oselectivity wh en th e reaction was
con ducted in th e presen ce of both TEA an d pyridin e. Un der such con di-
tion s, N3-Pv derivative was form ed th ree orders of m agn itude faster th an
th e 5′-O-Pv on e (t1/2 < 1 s vs 1 h ), an d required on ly a few secon ds for com -
pletion . (In th e presen ce of TEA an d pyridin e, pivaloyl ch loride (or oth er
acyl ch lorides) was con sum ed with in several m in utes in a side reaction
givin g un iden tified dark-brown products if th e reaction was n ot protected
from atm osph eric oxygen .) Th is h igh in crease in th e rate of pivaloylation
m igh t be explain ed in term s of syn ergistic effects of trieth ylam in e an d
pyridin e, with base catalysis by TEA operatin g on th e pyrim idin e rin g an d a
n ucleoph ilic catalysis by pyridin e operatin g on PvCl. Th e observed diverse
reactivity of PvCl in th e reaction with th ym idin e due to th ese two types of
FIG. 1
Estim ated kin etics of N3- an d 5′-O-acylation of th ym idin e correlated with possible base an d
n ucleoph ilic catalysis
Collect. Czech. Chem. Commun. 2010, Vol. 75, No. 1, pp. 33–57

Protection for 4-Oxopyrimidine Moieties
37
catalysis is drafted in Fig. 1. (Acylation of th ym idin e with adam an tan e-
1-carbon yl ch loride proceeded with sim ilar selectivity as pivaloylation ,
wh ile acetylation (Ac2O or AcCl) sh owed little regioselectivity. For
benzoylation, the influence of am ines on the directing the reaction to N3H or
5′-OH position s was sim ilar as for pivaloylation , but th e reaction s differed
sign ifican tly in term s of kin etics an d product distribution (Table I)1.)
Th e above fin din gs were exploited for rapid an d efficien t preparation of
N3-pivaloyl- (TPv, 2), 5′-O-pivaloyl- (PvT, 3), 3′-O,N3-dipivaloyl- (PvTPv, 4),
an d 3′,5′-O-dipivaloylth ym idin e (Pv2T, 5) by reaction of un protected
th ym idin e (1) with PvCl in th e presen ce of appropriate am in es: pyridin e
or DMAP for O-pivaloylation , or TEA/ Py for N-pivaloylation (Fig. 2 an d
Table II). Th e procedures reported h ere are eith er im provem en ts of th e
37,40
kn own m eth ods (for 3
an d 5 56,57) or n ew approach es (for 2 an d 4).
Mon opivaloylated th ym idin e derivatives are fairly well soluble in water an d
dieth yl eth er, wh ile rath er poorly in ben zen e or toluen e (n oteworth y,
an alogous ben zoylated or dim eth oxytritylated th ym idin e derivatives sh ow
an opposite solubility profile). Takin g advan tage of th is feature it was possi-
FIG. 2
Regioselective pivaloylation of th ym idin e (1a)
Collect. Czech. Chem. Commun. 2010, Vol. 75, No. 1, pp. 33–57

38
Sobkowski:
ble to replace th e laborious colum n ch rom atograph y with sim ple solven t
extraction s to obtain th e desired com poun ds in good yields an d purity
>98%. Due to a rath er poor solubility of th ym idin e in MeCN, in th e pre-
parative run s th e 1:1 Py–MeCN solution s were used.
Th us, N3-pivaloylth ym idin e (2) was prepared by treatin g th ym idin e (1a;
0.1 M solution in 1:1 MeCN–pyridin e m ixture con tain in g 15 equiv. of TEA)
with 1.2 equiv. of PvCl un der in ert gas atm osph ere. After 10 m in th e reac-
tion was com plete an d th e reaction m ixture con sisted of ca. 90% of th e de-
sired TPv (2) an d ca. 5% each of un reacted 1a an d PvTPv (4). Un reacted
th ym idin e an d by-products were rem oved by successive extraction s with
Et2O–TEAB 9:1 to rem ove h ydroph ilic com poun ds (DCM could be used as
well; h owever, dark by-products form ed acciden tally if oxygen was n ot
com pletely rem oved from th e reaction vessel could be efficien tly elim i-
n ated usin g eth er on ly) an d with water–toluen e 9:1 to rem ove PvTPv an d
lipoph ilic im purities. Attem pted crystallization of th e product was un suc-
cessful an d 2 was obtain ed as a wh ite foam (yield 70%).
5′-O-Pivaloylth ym idin e 3 was syn th esized by reactin g th ym idin e (0.1 M
solution in 1:1 MeCN–pyridin e) with 2 equiv. of PvCl. Th e reaction was
quen ch ed with water after 15 m in as th is reaction tim e was foun d to yield
th e best ratio of PvT to un reacted an d bis-pivaloylated th ym idin es.
Altern atively, wh en th e reaction was perform ed in th e presen ce of 5 equiv.
of DMAP, sim ilar rapid pivaloylation of 5′-OH was ach ieved with alm ost a
TABLE II
Con dition s an d results of pivaloylation of th ym idin e (1a) an d uridin e (1b)
Reaction con dition s
Nucleoside Product
Isolated yield, %
Am in e
PvCl, equiv. Reaction tim e, m in
1a
1a
1a
1a
1b
1b
10
TPv (2)
Py/TEA
1.1
2.0
2.2
2.2
1.1
5
15
15
70
87
PvT (3)
PvTPv (4)
Pv2T (5)
UPv (9)
Pv3UPv (10) Py/DMAP
Pv3U (11)a
Py
1.Py; 2.+TEA
Py/DMAP
Py/TEA
10 + 10
15
74
85
15
61
15
100
98
–
–
15
a
Selective deprotection of 10 (HCl/i-PrOH, 80 °C).
Collect. Czech. Chem. Commun. 2010, Vol. 75, No. 1, pp. 33–57

Protection for 4-Oxopyrimidine Moieties
39
stoich iom etric am oun t of PvCl (1.1 equiv.). Th us, th e risk of m ultiacylation
was avoided. In both cases, th e reaction s afforded PvT (3) (ca. 90%) accom -
pan ied by un reacted 1a an d bis-acylated Pv2T (5) (ca. 5% each ). After aque-
ous work-up, pure com poun d 3 was obtain ed by crystallization from
toluen e37 (yield 85–87%).
Efficien t on e-pot preparation of 5′-O,N3-dipivaloylth ym idin e (4) required
a com bin ation of th e m eth ods for O- an d N-acylation . To th is en d,
th ym idin e in 1:1 MeCN–pyridin e solution was treated with 2.2 equiv. of
PvCl in th e first step. After 10 m in (tim e required for ca. 90% form ation of
5′-O-pivaloylth ym idin e with out bis-acylation ), TEA (5 equiv.) was added,
wh ich prom oted a rapid N3-pivaloylation to give PvTPv (4). Aqueous
work-up (TEAB buffer pH 7.0), followed by lyoph ilization afforded 4 in 74%
yield.
For th e preparation of 3′,5′-O-dipivaloylth ym idin e (5), in order to secure
rapid pivaloylation of th e 3′-OH group (ca. 10 m in ), 2.5 equiv. of PvCl was
used in th e presen ce of DMAP (5 equiv.) as a n ucleoph ilic an d base catalyst.
After aqueous work-up (5% citric acid to rem ove DMAP followed by TEAB
buffer) Pv2T (5) was lyoph ilized from ben zen e (yield 85%).
A special atten tion was paid to recogn ize wh eth er pivaloylation at th e
th ym in e m oiety took place in th e N3 or O4 site. Wh ile N3 is th e typical
acylation position in 4-oxopyrim idin e group of n ucleosides, ben zoylation
of th ym idin e derivatives in th e presen ce of TEA led to a m ixture of N3- an d
O4-acylated products1, wh ich were used as referen ce com poun ds for TLC
an alysis on th eir pivaloylated an alogues. In every case th e pivaloylated
derivatives sh owed TLC m obilities com parable to th eir N3-ben zoylated
con gen ers. For in stan ce, th e O4- an d N3-ben zoylated derivatives of
3′,5′-O-bis(tert-butyldim eth ylsilyl)th ym idin e h ad th e RF values of 0.11 an d
0.59, respectively (DCM–Et2O 95:5), while the product of pivaloylation of the
sam e com pound had the RF value of 0.62. Also TPv (2) and N3-benzoylthymi-
dine had similar RF values, 0.31 and 0.29, respectively (DCM–MeOH 9:1).
In th e 13C NMR spectra th e ch em ical sh ifts for th e carbon s C5 h ave been
establish ed to be of an an alytical value for determ in ation of N3 vs O4 sub-
stitution in uridin e derivatives16. Th us, acylation or alkylation in th e N3
position was foun d to in fluen ce th e ch em ical sh ift for C5 on ly m argin ally,
wh ile in O4-substituted derivatives, ca. 4 ppm upfield sh ift was observed.
Th e reson an ces for C6 carbon s in O4-esters were sh ifted ca. 6 ppm
down field, wh ile th ose in N3-am ides rem ain ed un affected16a. Sin ce
pivaloylation of uracil an d th ym in e m oieties did n ot in fluen ce sign ifican tly
the chem ical shifts for C5 and C6 (1b, 9, 10: δC5 ~103 ppm , δC6 138–140 ppm ;
Collect. Czech. Chem. Commun. 2010, Vol. 75, No. 1, pp. 33–57

40
Sobkowski:
1a, 2, 4: δC5 ~111 ppm , δC6 ~136 ppm ), th e site of derivatization was in -
ferred to be th e n itrogen N3.
Regioisom ers 2, 3 an d 4, 5 could be clearly distin guish ed by TLC (N3-Pv
derivatives sh ow h igh er m obility th an th eir O-Pv con gen ers). Th eir struc-
1
tures were also con firm ed by H an d 13C NMR spectroscopy+++. A clear-cut
differen ce between 13C ch em ical sh ifts of th e pivaloyl carbon yl carbon s
in N-Pv (2 , 4 , 9 , an d 1 0 : δC ~178 p p m ) an d O-Pv (3 , 4 , 5 , 1 0 , an d 1 1 :
δC ~184 ppm ) groups was foun d in all an alyzed com poun ds.
Th e above described con dition s for selective pivaloylation could be
adopted for acylation of partly protected th ym idin e derivatives. For exam ple,
5′-O,N3-dipivaloylth ym idin e (4) could be prepared by 5′-O-pivaloylation of
TPv (2) or by N3-pivaloylation of PvT (3) accordin g to th e procedures for
pivaloylation of 5′-OH an d N3H sites of th ym idin e. Pivaloylation of
5′-O-(dim eth oxytrityl)th ym idin e in th e presen ce of TEA yielded 5′-O-(di-
m ethoxytrityl)-N3-pivaloylth ym idin e (7) in 87% yield after TEAB–DCM
work-up an d precipitation in to cycloh exan e. Com poun d 3 is a con ven ien t
substrate for th e preparation of 3′-O-(dim eth oxytrityl)th ym id in e (8) in
a procedure an alogous to th at described by De Nin o et al. wh o used
5′-O-(4-n itroben zoyl)th ym idin e as startin g com poun d58 (see Experim en tal).
Stability of N3-Pivaloyl Group
Pivalate esters of th ym idin e are stable un der m oderately acidic (pH > 1)40
an d m oderately basic con dition s. Th ey m ay be cleaved by treatm en t with
solution s of am m on ia, prim ary am in es or tetrabutylam m on ium h ydroxide,
as well as alkali m etal h ydroxides37,59. N3-Pivaloylth ym idin e (2) sh owed a
sign ifican tly differen t profile of stability. In con trast to pivalic esters, it un -
derwen t slow h ydrolysis in n eutral aqueous solution s; h owever, in stan dard
aqueous work-up of th e reaction m ixture n o detectable am oun ts of th e
depivaloylated products could be foun d. Hydrolysis of TPv in water at room
tem perature was com plete with in 7 days. Th e reaction rate was sign ifican tly
en h an ced at elevated tem perature an d a com plete rem oval of th e pivaloyl
group from TPv (2) was accom plish ed with in 10 m in at 90 °C in water. Th e
+++N3-Pivaloylated derivatives of 4-oxopyrim idin e n ucleosides sh owed un usual NMR pattern s
of 1H an d 13C NMR sign als, e.g. broaden in g of som e sign als an d/or th eir splittin g in to pairs
of sign als, depen din g on tem perature an d th e solven t used (Table IV). Th ese suggested
in volvem en t of th ose TPv derivatives in som e kin d of equilibria, th e n ature of wh ich is
un kn own at presen t.
Collect. Czech. Chem. Commun. 2010, Vol. 75, No. 1, pp. 33–57

Protection for 4-Oxopyrimidine Moieties
41
depen den ce of th e h ydrolysis rate at 80 an d 90 °C on pH of th e solution
was in vestigated in ph osph ate buffers in th e pH ran ge 2–12 (Figs 3 an d 5b).
In con trast to acidic an d basic con dition s, in n eutral m edia th e rate of h ydro-
lysis was very low (several h ours). Th e h igh rate of h ydrolysis of TPv in
water (in com parison with th at in pH 6 buffer) m ay be plausibly attributed
to acid autocatalysis sin ce pH of an un buffered solution decreased sign ifi-
can tly in th e course of th e reaction due to released pivalic acid.
In terestin gly, h ydrolysis of TPv in 1:1 MeCN–water m ixture required ca.
5 h for th e com pletion (at 80 °C), in com parison with 0.5 h in water at th e
sam e tem perature (Fig. 4), wh ile in 95:5 MeCN–water n o traces of h ydroly-
sis could be observed even after several h ours of reflux at 78 °C. Th e retar-
dation effect was even stron ger in 1:1 i-PrOH–water m ixture (com pletion
tim e ca. 10 h ). Th e rate of h ydrolysis of TPv was also sign ifican tly reduced
in th e presen ce of DMF an d DMSO. It m ust be n oted, h owever, th at upon
addition of a large volum e of organ ic solven t to water, pH of th e m ixture
in creased by ca. 0.3 (for MeCN) an d 1.2 (for i-PrOH), presum ably as a result
of lower solubility of CO2. Th is ph en om en on could accoun t partly for th e
observed retardation of h ydrolysis; n everth eless, th e reaction rate was still
low in spite of sign ifican t am oun ts of pivalic acid released in to th e m edium
durin g deacylation . Th us, h ydrolysis of TPv was repeated in buffered m ix-
tures; th is revealed th e pseudo-first order of th e reaction (Fig. 5). Th e in flu-
en ce of aceton itrile on h ydrolysis rates was h igh er in less acidic solution s
FIG. 3
Kin etics of deacylation of 0.01 M N3-pivaloylth ym idin e (2) in water (pH 6.0) an d in pH 2–7
buffers at 90 °C
Collect. Czech. Chem. Commun. 2010, Vol. 75, No. 1, pp. 33–57

42
Sobkowski:
(Fig. 5b), an d th is was in good agreem en t with th e sign ifican t rate decrease
observed in n early n eutral (in itially) MeCN–water m ixtures (Fig. 4).
Durin g h ydrolysis of Pv2T (5) un der sim ilar con dition s, in water an d in
buffers (pH 2–5), th e N3-Pv group was cleaved with full ch em oselectivity,
wh ile un der n early n eutral con dition s an accom pan yin g m in or cleavage of
FIG. 4
In fluen ce of MeCN an d i-PrOH on th e kin etics of deacylation of 0.01 M N3-pivaloylth ym idin e
(2) at 80 °C
a
b
time, s
pH
FIG. 5
Hydrolysis of 0.01 M TPv (2) in aqueous or aqueous–MeCN phosphate buffers (50 mmol l–1) at 80 an d
90 °C (±1 °C). a Plots of th e h ydrolysis progress at 80 °C. b Correlation of log kobs vs pH for h y-
drolysis of TPv (2)
Collect. Czech. Chem. Commun. 2010, Vol. 75, No. 1, pp. 33–57

Protection for 4-Oxopyrimidine Moieties
43
th e 5′-O-Pv group was observed (th e 3/2 ratio ca. 85:15). Un der basic con di-
tion s (pH ≥ 10), h ydrolysis of th e 5′-O-Pv group was th e dom in an t reaction .
Apart from th erm al h ydrolysis, th e stability of N3-pivaloyl group un der
various acidic an d basic con dition s was estim ated. Th e approxim ate
h alf-life tim es of deacylation are given in Table III. In con trast to h ydrolysis
in buffers at elevated tem perature, n o sign ifican t differen ces between acid
h ydrolysis in aqueous an d MeCN solution s were observed. However, in
weakly polar solven ts (DCM, ch loroform ), th e acid-catalyzed depivaloyl-
ation of TPv was ca. twen ty tim es faster th an th at in polar solven ts (MeCN,
H2O). Un der basic con dition s th e h ydrolysis was ca. fifteen tim es faster in
water th an in MeOH, wh ich in dicates a sign ifican t solven t effect on
cleavage of th e N3–Pv bon d. In th e presen ce of a lim ited am oun t of water
(a few per cen t), th e accom pan yin g tran sacylation , m ain ly to PvT (3) was
observed.
Dilute aqueous am m on ia (0.5 m ol l–1) in MeCN cleaved th e N3-pivaloyl
group sim ilarly to pure water (t1/2 ≈ 24 h ), wh ile con cen trated aqueous NH3
required several h ours to com plete deacylation . Th e deprotection tim e in
con cen trated am m on ia could be reduced to ca. 2 h by in creasin g th e tem -
perature to 55 °C.
In th e case of N3-ben zoyl protected th ym idin e an d its an alogues, th e
m ost frequen tly used deprotection reagen t was sodium h ydroxide in organ ic
solven ts60–63. Un der such con dition s rapid an d efficien t deben zoylation
could be ach ieved. (N3-Ben zoyl group could be also cleaved with in 1 h in
con cen trated am m on ia at room tem perature24 or by prolon ged h eatin g in
ben zyl alcoh ol (30 h /90 °C)64. However, it is stable in 80% AcOH an d 2%
TFA/CHCl3 (ref.24).) For in stan ce, Ram asam y an d Stoisavljevic63 reported
99% yield of deprotection using 1 M NaOH/MeOH/THF for 30 min. In contrast,
TPv required ca. 5 h for deacylation usin g aqueous 0.5 M NaOH, wh ile in
m eth an olic solution th e reaction was com plete in ca. 24 h .
TABLE III
Half-tim es of acid- an d base-catalyzed deacylation of 0.01 M N3-pivaloylth ym idin e (2) at
room tem perature
0.5 M TFA or HCl 0.5 M TFA
80% AcOH 0.5 M NaOH 0.5 M NaOH Aqueous
in MeCN or H2O in DCM or CHCl3
in H2O
in MeOH
25% NH3
10 h
0.5 h
10 h
(5 h )a
6 h
1 h
a
Com pletion tim e.
Collect. Czech. Chem. Commun. 2010, Vol. 75, No. 1, pp. 33–57

44
Sobkowski:
Th e aforem en tion ed resistan ce of TPv to m eth an olic NaOH solution
open ed a ch an ce for usin g th e N3-pivaloyl group as a protection orth ogon al
to base-labile groups presen t in oth er position s of th e n ucleoside, such as
ester groups. To verify th is possibility, PvTPv (5) was treated with 0.5 M NaOH
(10 equiv.) in MeOH (room tem perature) to cleave 5′-O-pivaloyl, or with
boilin g water or in a buffer solution (pH 3) at 90 °C to cleave N3-pivaloyl
group (Fig. 6). Un der basic con dition s, depivaloylation was h igh ly regio-
selective favourin g th e 5′-O deprotection , an d required ca. 1 h for comple-
tion. (Un der sim ilar con dition s, N3-ben zoyl group is cleaved rapidly60–63
.
Regioselective cleavage of ester ben zoyl groups from Bz2TBz proceeds by
treatm en t with 2.2 equiv. of NaOH overn igh t64.) Th e m ain produ ct of
th e reaction was th e expected TPv (2), accom pan ied by ca. 3% of fully
deprotected th ym idin e (1a), wh ich could be easily wash ed out durin g
work-up. Un der acidic con dition s (water or buffer), th e N3 position was
deprotected exclusively to form PvT (3) with in 15–30 m in . On th e oth er
h an d, dm tTPv could be selectively deprotected to TPv sin ce un der m ild
acidic con dition s required for com plete detritylation (80% acetic acid,
room tem perature, 30 m in ), th e N3-pivaloyl group was cleaved on ly m ar-
gin ally (ca. 3%). Th us, th e pivaloyl group can be con sidered a m oderately
acid-labile an d base-labile protectin g group for th e N3 position of
th ym idin e, wh ich resists detritylation an d de-esterification con dition s, but
can be cleaved rapidly in hot water (30 m in) or in aqueous am m onia at 55 °C
(2 h ) or with 0.5 M TFA at room tem perature (4 h ).
Pivaloylated Derivatives of Uridine
Th e in creased reactivity of th e ribose h ydroxy groups with electroph iles,
th e lack of th e 5-m eth yl group, an d differen t con form ation al preferen ces
are th e m ain features of uridin e (1b) com pared with th ym idin e (1a) th at
Fig. 6
Selective deprotection of Pv2T (5) un der basic or m ildly acid con dition s (80 °C/water)
Collect. Czech. Chem. Commun. 2010, Vol. 75, No. 1, pp. 33–57

Protection for 4-Oxopyrimidine Moieties
45
m ay in fluen ce its pron en ess to acylation an d stability of th e protectin g
groups (Fig. 7, Table II).
Syn th esis of N3-pivaloyluridin e (UPv, 9) proceeded sm ooth ly un der th e
sam e con dition s as for TPv; h owever, due to h igh er solubility of UPv in wa-
ter, m ultiple DCM extraction s of th e TEAB buffer ph ase were required in
th e work-up step. Th e product was obtain ed as wh ite foam in 61% yield.
Attem pted selective O-pivaloylation of uridin e un der con dition s devised
for preparation of PvT led, in con trast, to a m ixture of several partly
acylated products, sim ilarly as it was observed by Kam aike et al.65 Notewor-
th y, peracylation of uridin e with an excess of PvCl an d DMAP was foun d to
be a rapid reaction (com pletion tim e 15 m in ). Th e in term ediate product
(Pv3UPv, 10) un derwen t selective N-depivaloylation on brief h eatin g in
0.5 M HCl/i-PrOH (15 m in ). Th e produced 2′,3′,5′-O-tripivaloyluridin e
(Pv3U, 11) was isolated in 98% yield after lyoph ilization . Th is process is
th us a cost- an d tim e-effective altern ative to th e literature m eth od for th e
FIG. 7
Regioselective pivaloylation of uridin e (1b)
Collect. Czech. Chem. Commun. 2010, Vol. 75, No. 1, pp. 33–57

46
Sobkowski:
preparation of 11 th at requires 20 + 7 h reaction tim e for pivaloylation
usin g th e Pv2O/Bi(OTf)3 reagen t system 57
.
N3-Pivaloyl group in UPv (9) appeared to be sign ifican tly m ore labile th an
th at of TPv; it was cleaved com pletely with in several m in utes in 80% AcOH
at room tem perature or in water at 90 °C. Th e h alf-tim e of deacylation of
UPv at pH 2 an d room tem perature was ca. 6 h , wh ile in 25% aqueous NH3
at room tem perature, 30 m in (form ation of som e side-products was n oted
in th e latter case, ca. 20%). In 0.5 M NaOH, th e TLC spot of UPv disappeared
with in less th an 15 m in (H2O) or 30 m in (MeOH). However, n o form ation
of uridin e was observed. In terestin gly, in perpivaloylated derivative, Pv3UPv
(10), th e N-pivaloyl group was by orders of m agn itude m ore resistan t to acids
th an in UPv an d required several h ours in 1 M HCl/alcoh ol solution or m ore
th an 48 h in 80% AcOH for com plete cleavage. Th e reaction could be sig-
n ifican tly accelerated by h eatin g, with out affectin g th e regioselectivity of
deacylation (vide supra). Con cludin g, th e pivaloyl group in th e N3 position
of uridin e seem s to h ave lim ited syn th etic application s as a tran sien t pro-
tectin g group.
Mechanism of Alkaline Hydrolysis of TPv (2)
Th e attem pts to evaluate th e kin etics of NaOH-prom oted deacylation of TPv
in aqueous solution by TLC m on itorin g its progress failed sin ce th e gradual
disappearan ce of th e spot of substrate 2 was n ot accom pan ied by sim ulta-
n eous form ation of an y oth er UV-absorbin g (254 n m ) product in th e first
20–30 m in . After th at period, th e spot of th ym idin e (1a) started to appear
in th e TLC of th e reaction m ixture an d, after ca. 5 h , th e reaction was com -
plete. Apart from depivaloylation , form ation of som e oth er products took
place (ca. 10%). Th ese side reaction s were dom in an t wh en a sam ple of th e
reaction m ixture from th e early stages (5–60 m in ) was n eutralized or acidi-
fied.
UV spectroph otom etric quan tification of th e reaction progress (Fig. 8) re-
vealed a gradual decrease in in ten sity of th e peak at 272.5 n m th at reach ed
th e lowest value after th e 20–25 m in reaction . After th at tim e a n ew m axi-
m um started to rise at 269 n m un til th e en d of reaction after ca. 5 h .
Tem porary disappearan ce of a C6=C5–C4=O ch rom oph ore at ca. 270 n m
was presum ably caused by th e addition of h ydroxide ion s to th is system ,
presum ably at C6. Th us, due to a low rate of depivaloylation un der alkalin e
con dition s (Fig. 9, path A), th e form ation of 6-h ydroxy-N3-pivaloyl-
th ym idin e (9, path B) could take place. Subsequen t cleavage of th e pivaloyl
group was apparen tly associated with deh ydration to th ym idin e (1a), as
Collect. Czech. Chem. Commun. 2010, Vol. 75, No. 1, pp. 33–57

Protection for 4-Oxopyrimidine Moieties
47
FIG. 8
UV spectra of th e reaction m ixture durin g h ydrolysis of TPv (2) in aqueous 0.5 M NaOH. Reac-
tion tim es: 1–20 m in (th ick lin es); 20–300 m in (th in lin es). Th e ch rom oph ores in th e th ym in e
m oiety are m arked in grey
FIG. 9
Putative m ech an ism of depivaloylation an d degradation of TPv (2). Stage 1 (upper row): TPv in
0.5 M NaOH. Stage 2 (lower row): degradation of in term ediate 9 un der n eutral or acid con di-
tion s
Collect. Czech. Chem. Commun. 2010, Vol. 75, No. 1, pp. 33–57

48
Sobkowski:
eviden ced by re-appearan ce of th e peak at 270 n m (Fig. 8). Upon acidifica-
tion (pH ≤ 7), com poun d 9 was un stable. TLC sh owed form ation of several
products, of wh ich deoxyribose was readily iden tified, wh ile two oth er iso-
lated com poun ds, 12 an d 13 (Fig. 9), were assign ed ten tatively on th e basis
1
of th eir TLC m obilities, an d UV, H/13C NMR an d MS spectra. It is worth
n otin g th at an open -ch ain form of h ydrated th ym in e was postulated as an
in term ediate in cis–trans isom erization of 6-hydroxy-5,6-dihydrothym idine66.
Furth er in vestigation s on elucidation of th e reaction s of N3-acylated
4-oxopyrim idin es un der alkalin e con dition s are in progress.
CONCLUSIONS
Acylation of n ucleosides is usually con sidered a tim e-dem an din g process. In
con trast, in th e presen t work it was foun d th at un der properly ch osen reac-
tion con dition s th e pivaloylation of 4-oxopyrim idin es could be com pleted
with in several m in utes. Moreover, th e reaction s with th ym idin e were
h igh ly regioselective an d th e site of acylation could be con trolled by a
proper ch oice of am in e actin g as a base an d/or n ucleoph ilic catalyst of th e
reaction . Th u s, stron gly basic bu t p oorly n u cleop h ilic trieth ylam in e
prom oted acylation of th e n ucleobase m oiety, wh ile in th e presen ce of
n ucleoph ilic catalysts (pyridin e or DMAP) th e reaction proceeded
selectively in th e sugar part. Th e pivaloylated products could be purified
by organ ic-aqueous extraction s; th ey were isolated in good yields.
Th e pivaloyl group could be cleaved from N3-pivaloylth ym idin e deriva-
tives un der acidic or basic con dition s; h owever, it was sufficien tly stable to
with stan d acid-prom oted detritylation or treatm en t with alkali in organ ic
solven ts used for th e rem oval of ester protectin g groups. Th us, th e pivaloyl
group em erged as a con ven ien t protectin g group for th e N3 position of
th ym idin e.
N3-Pivaloylation of uridin e could be ach ieved with a sim ilar regio-
selectivity as th ym idin e, wh ile th e an alogous selective O-pivaloylation was
n ot successful. Neverth eless, a con ven ien t procedure for th e preparation
of 2′,3′,5′-O-tripivaloyluridin e was devised. N3-Pivaloyluridin e was less
stable th an N3-pivaloylth ym idin e both un der basic an d acid con dition s.
EXPERIMENTAL
NMR spectra (δ, ppm ; J, Hz) were collected on Bruker Avan ce II 400 MHz an d Varian Un ity
300 MHz spectrom eters. 1H an d 13C NMR sign al assign m en t was based on 2D correlation
spectra. UV spectra were recorded on a Jasco V-650 spectroph otom eter. ESI HR-MS spectra
were recorded on
a
Bruker m icrOTOF-Q in strum en t operated in positive ion m ode.
Collect. Czech. Chem. Commun. 2010, Vol. 75, No. 1, pp. 33–57

Protection for 4-Oxopyrimidine Moieties
49
Th ym idin e was purch ased from Rasayan In c. an d uridin e from Ph arm a–Waldh of. Th ese an d
oth er com m ercial reagen ts an d solven ts (Sigm a–Aldrich , Merck, an d POCh –Polan d) were
used as purch ased un less oth erwise n oted. An h ydrous solven ts con tain in g below 20 ppm of
water (by Karl Fisch er coulom etric titration , Metroh m 684 KF) were stored over m olecular
sieves 4 Å. PvCl was distilled an d used with in on e m on th . A 1 M trieth ylam m on ium h ydrogen -
carbon ate (TEAB) buffer was prepared by bubblin g CO2 th rough aqueous TEA at 0 °C un til
pH 7.0 was reach ed.
N3-(Trim eth ylacetyl)th ym idin e (TPv; 2)
Th ym idin e (2.42 g, 10 m m ol) was dissolved in 1:1 MeCN–pyridin e m ixture (100 m l) con -
tain in g trieth ylam in e (7 m l, 5 equiv.), an d PvCl (1.4 m l, 1.1 equiv.) was added wh ile stirrin g
vigorously un der N2 atm osph ere. After 15 m in th e reaction was quen ch ed by addition of
water an d th e solven ts were evaporated un der reduced pressure. Dieth yl eth er (200 m l) was
added an d th e precipitate was filtered off, wash ed with 50 m l of eth er an d discarded. Th e
eth er solution was extracted with 1 M TEAB buffer (25 m l) an d water (5 m l), an d th e com -
bin ed aqueous ph ases were wash ed with 100 m l of eth er an d discarded. Th e organ ic fraction
con tain in g com poun d 2 was evaporated an d th e residue was partition ed between water
(200 m l) an d toluen e (20 m l). Th e aqueous ph ase was evaporated to dryn ess un der reduced
pressure in a 30 °C bath , an d th e residue was dissolved in DCM. If TLC revealed som e h ydro-
lysis of TPv durin g th e work-up, addition al extraction with 200 m l DCM/10 m l water was
applied. Th e DCM solution was evaporated (repeated twice) givin g a wh ite foam . Yield 2.3 g
(70%). RF 0.31 (DCM–MeOH 9:1). For 1H NMR, see Table IV. 13C NMR (75 MHz, DMSO-d6):
184.03 (C=O, Pv); 162.19 (C4); 148.71 (C2); 136.77 (C6); 109.25 (C5); 87.58 (C1′); 84.43
(C4′); 70.21 (C3′); 61.14 (C5′); 43.29 (C(CH3)3); 39.60 (C2′); 26.86 (C(CH3)3); 12.07 (C5).
13C NMR (101 MHz, CDCl3): 184.16 (C=O, Pv); 162.91 (C4); 149.21 (C2); 136.38 (C6, br);
110.82 (C5, br); 87.06 (C4′); 86.26 an d 86.01 (C1′); 71.19 (C3′); 62.11 (C5′); 43.89 (C(CH3)3);
40.35 an d 39.89 (C2′); 27.22 (C(CH3)3); 12.42 (C5).
5′-O-(Trim eth ylacetyl)th ym idin e (PvT; 3)
Th ym idin e (2.42 g, 10 m m ol) was dissolved in 1:1 MeCN–pyridin e m ixture (100 m l) an d
PvCl (2.52 m l, 2.0 equiv.) was added wh ile stirrin g vigorously. After 15 m in th e reaction was
quen ch ed with water an d th e solven ts were evaporated un der reduced pressure. Th e oily resi-
due was dissolved in 200 m l of DCM an d extracted successively with a 5% aqueous citric
acid (50 m l)–1 M TEAB buffer (50 m l). DCM was evaporated an d th e resultin g wh ite foam
was dissolved in toluen e (250 m l, 90 °C). Th e solution was cooled to –20 °C yieldin g a wh ite
p recipitate37. Yield 2.83 g (87%). RF 0.39 (DCM–MeOH 9:1). For 1H NMR, see Table IV.
13C NMR (101 MHz, CDCl3): 178.44 (C=O, Pv); 163.88 (C4); 150.52 (C2); 134.93 (C6);
111.29 (C5); 85.06 (C1′); 84.58 (C4′); 71.55 (C3′); 63.90 (C5′); 40.49 an d 38.85 (C2′ an d
C(CH3)3); 27.19 (C(CH3)3); 12.49 (C5).
N3,5′-O-Bis(trim eth ylacetyl)th ym idin e (PvTPv; 4)
Th ym idin e (2.42 g, 10 m m ol) was dissolved in a 1:1 MeCN–pyridin e m ixture (100 m l) an d
PvCl (2.77 m l, 2.2 equiv.) was added wh ile stirrin g vigorously un der N2 atm osph ere. After
10 m in TEA (7.0 m l, 5.0 equiv.) was added. After an oth er 10 m in , th e reaction was
quen ch ed with water an d evaporated un der reduced pressure. Dieth yl eth er (100 m l) was
Collect. Czech. Chem. Commun. 2010, Vol. 75, No. 1, pp. 33–57

50
Sobkowski:
Collect. Czech. Chem. Commun. 2010, Vol. 75, No. 1, pp. 33–57

Protection for 4-Oxopyrimidine Moieties
51
Collect. Czech. Chem. Commun. 2010, Vol. 75, No. 1, pp. 33–57

52
Sobkowski:
added an d th e precipitate was filtered off, wash ed with 50 m l of eth er an d discarded. Th e
eth er solution was extracted successively with 1 M TEAB buffer (150 m l) an d water (150 m l),
an d evaporated givin g a wh ite foam . Occasion al by-products causin g yellow coloration of
th e product could be rem oved from a DCM solution usin g ch arcoal. Th e product was dis-
solved in warm 9:1 n-h exan e–Et2O m ixture an d kept at –20 °C overn igh t yieldin g wh ite fi-
brous precipitate. Yield 3.1 g (74%). RF 0.35 (DCM–MeOH 95:5). For 1H NMR, see Table IV.
13C NMR (101 MHz, CDCl3): 183.82 an d 183.54 (N-C=O, Pv); 178.33 (O-C=O, Pv); 162.70
(C4); 148.88 (C2); 134.88 an d 134.52 (C6); 110.82 (C5); 85.43 an d 85.12 (C1′); 84.44 (C4′);
71.21 (C3′); 63.75 (C5′); 43.80 (N-C(O)-C(CH3)3); 40.62 an d 40.46 (C2′); 38.77
(O-C(O)-C(CH3)3); 27.17 (C(CH3)3); 27.11 (C(CH3)3); 12.39 (C5).
3′,5′-O-Bis(trim eth ylacetyl)th ym idin e (Pv2T; 5)
Th ym idin e (2.42 g, 10 m m ol) was dissolved in a 1:1 MeCN–pyridin e m ixture (100 m l) con -
tain in g DMAP (6.1 g, 5 equiv.), an d PvCl (2.77 m l, 2.2 equiv.) was added wh ile stirrin g vig-
orously. After 15 m in th e reaction was quen ch ed by addition of water an d th e solven ts were
evaporated un der reduced pressure. Toluen e (200 m l) was added, th e precipitate was filtered
off, wash ed with 50 m l of toluen e an d discarded. Th e solution was extracted successively
with 5% citric acid (150 m l) an d 1 M TEAB buffer (150 m l), an d evaporated givin g a wh ite
foam . Th e product was dissolved in h ot n-h exan e an d kept at –20 °C overn igh t yieldin g a gel
wh ich was filtered off, wash ed with cold h exan e an d dried to wh ite powder. Yield 3.50 g
(85%). RF 0.37 (DCM–MeOH 95:5). For 1H NMR, see Table IV. 13C NMR (101 MHz, CDCl3):
178.05 (C=O, Pv); 177.88 (C=O, Pv); 163.36 (C4); 150.11 (C2); 134.34 (C6); 111.47 (C5);
84.75 (C1′); 82.58 (C4′); 73.92 (C3′); 63.85 (C5′); 38.84 (C(CH3)3); 38.64 (C(CH3)3); 37.74
(C2′); 27.23 (C(CH3)3); 26.97 (C(CH3)3); 12.48 (C5).
3′-O-(Dim eth oxytrityl)-5′-O-(trim eth ylacetyl)th ym idin e (7)
5′-O-Pivaloylth ym idin e (3; 1.63 g, 5 m m ol) was dried by co-evaporation with pyridin e (2 ×),
and dissolved in the sam e solvent (25 m l). DMAP (61 m g, 0.5 m m ol) and TEA (2.8 m l, 10 m m ol)
were added followed by dim eth oxytrityl ch loride (3.39 g, 10 m m ol), an d th e reaction
m ixture was kept at room tem perature for 48 h un til th e tritylation was com plete (TLC).
Meth an ol was added, an d after typical work-up in a DCM–1 M TEAB buffer m ixture, a yellow
foam was obtain ed, con tain in g 3′-O-(dim eth oxytrityl)-5′-O-pivaloylth ym idin e (7) an d fast
m igratin g (TLC) products of decom position of excess DMTrCl. Th is m ixture was used for
depivaloylation (see below) or was purified by silica gel colum n ch rom atograph y, 0–3% gra-
dien t of MeOH in DCM, an d was lyoph ilized from ben zen e yieldin g a wh ite am orph ous
solid. Yield 2.7 g (87%). RF 0.45 (DCM–MeOH 95:5). 1H NMR (300 MHz, CDCl3): 9.54 s, 1 H
3
3
(H6); 7.14–7.50 m , 9 H (Ar); 6.88 d, 2J = 8.7, 4 H (Ar); 6.37 dd, J1′-2′a = 5.6, J1′-2′b = 8.4, 1 H
(H1′); 4.40 dm , 3J = 6.3, 1 H (H3′); 4.10 m , 1 H (H4′); 3.97 dd, J5′a-5′b = 12.3, J4′-5′a = 2.4, 1 H
(H5′a); 3.90 dd, J5′a-5′b = 12.3, J4′-5′b = 4.5, 1 H (H5′b); 3.82 s, 6 H (OMe); 2.06 br dd, J2′a-2′b
2
3
2
3
2
=
=
3
2
3
3
13.7, J1′-2′a = 5.6, 1 H (H2′a); 1.90 s, 3 H (5-Me); 1.54 ddd, J2′a-2′b = 13.7, J1′-2′b = 8.5, J2′b-3′
6.6, 1 H (H2′b); 1.15 s, 9 H (Pv). 13C NMR (101 MHz, CDCl3): 177.67 (C=O, Pv); 163.83
(C4); 150.27 (C2); 134.67 (C6); 110.95 (C5); 84.80 (C1′); 83.47 (C4′); 73.56 (C3′); 63.75
(C5′); 39.47 an d 38.58 (C2′ an d C(CH3)3); 26.99 (C(CH3)3); 12.36 (C5); 158.69, 144.76,
135.87, 130.05, 128.07, 127.95, 127.07, 113.28, 87.29, 55.13 (DMTr).
Collect. Czech. Chem. Commun. 2010, Vol. 75, No. 1, pp. 33–57

Protection for 4-Oxopyrimidine Moieties
53
3′-O-(Dim eth oxytrityl)th ym idin e (8)
Crude 3′-O-dim eth oxytrityl-5′-O-pivaloylth ym idin e (7; 5 m m ol) from th e previous step was
dissolved in m eth an ol (5 m l). A 1 M m eth an olic NaOH (20 m l) was added an d th e reaction
m ixture was left stan din g at room tem perature overn igh t. Th e work-up was don e accordin g
to th e procedure of De Nin o et al.58 After com plete depivaloylation (TLC), water (100 m l)
was added, an d th e solution extracted with dieth yl eth er (2 × 40 m l) an d DCM (3 × 25 m l).
DCM fraction s were com bin ed, evaporated, an d th e resultin g oil was triturated in dieth yl
eth er affordin g a fin e wh ite powder. Yield 2.3 g (85%). RF 0.52 (1.08 relatively to 5′-O-DMTrT;
DCM–MeOH 9:1). Th e spectral data were con sisten t with literature58
.
N3-(Trim eth ylacetyl)uridin e (UPv; 9)
Sim ilar procedure as for TPv (2) was used with an exception th at in th e first stage of work-up,
th e TEAB buffer was extracted five tim es with a ten -fold volum e of DCM (UPv is well soluble
in water an d poorly in Et2O). Yield 2.0 g (61%). RF 0.23 (DCM–MeOH 9:1). 1H NMR (400 MHz,
3
3
3
DMSO-d6): 8.07 d, J5-6 = 8.1, 1 H (H6); 5.87 d, J5-6 = 8.1, 1 H (H5); 5.77 d, J1′-2′ = 4.5, 1 H
(H1′); 5.49 br, 1 H (2′-OH); 5.15 br, 2 H (3′-OH, 5′-OH); 4.07 m , 1 H (H2′); 3.99 m , 1 H
2
3
2
(H3′); 3.88 m , 1 H (H4′); 3.65 dd, J5′a-5′b = 12.3, J4′-5′a = 2.4, 1 H (H5′a); 3.57 dd, J5′a-5′b
=
12.3, J4′-5′b = 2.5, 1 H (H5′b); 1.23 s, 9 H (Pv). 13C NMR (101 MHz, DMSO-d6): 184.01 br
(C=O, Pv); 161.57 (C4); 149.08 (C2); 141.31 (C6); 101.36 (C5); 88.34 (C1′); 85.18 (C4′);
73.78 (C3′); 69.77 (C2′); 60.69 (C5′); 43.40 (C(CH3)3); 26.86 (C(CH3)3).
3
2′,3′,5′-O-Tris(trim eth ylacetyl)uridin e (Pv3U; 11)
Uridin e (1b; 2.44 g, 10 m m ol) was dissolved in a MeCN–pyridin e m ixture (1:1, 100 m l) con -
tain in g DMAP (8.5 g, 7 equiv.), an d PvCl (6.3 m l, 5 equiv.) was added wh ile stirrin g vigor-
ously. After 15 m in th e reaction was quen ch ed by addition of water an d th e solven ts were
evaporated un der reduced pressure. Toluen e (200 m l) was added an d th e precipitate was fil-
tered off, wash ed with 50 m l of toluen e an d discarded. Th e solution was extracted succes-
sively with 5% citric acid (150 m l) an d 1 M TEAB buffer (150 m l), an d evaporated to obtain a
white foam of Pv3UPv (10; quantitative yield). RF 0.76 (DCM–MeOH 95:5). 1H NMR (300 MHz,
3
CDCl3): 7.44 br, 1 H (H6); 6.07 br, 1 H (H1′); 5.77 d, J5-6 = 8.1, 1 H (H5); 5.26 br, 2 H (H2′
2
2
an d H3′); 4.41 br d, J5′a-5′b = 11.4, 1 H (H5′a); 4.31 br, 1 H (H4′); 4.25 br d, J5′a-5′b = 11.4,
1 H (H5′b); 1.16, 1.22, 1.23, 1.29, 4 × s, 4 × 9 H (4 × Pv). 13C NMR (101 MHz, CDCl3): 182.7 br
(N-C=O, Pv); 177.07 (O-C=O, Pv); 161.51 (C4); 148.92 (C2); 138.3 br (C6); 103.02 (C5);
86.4 br (C1′); 80.97 (C4′); 72.8 br (C3′); 70.5 br (C2′); 63.41 (C5′); 38.76 an d 38.79 (C(CH3)3);
26.91, 27.00, 27.17 (C(CH3)3).
Com poun d 10 obtain ed in th e previous step was dissolved in i-PrOH (110 m l), 11 M HCl
was added (2 m l) an d th e m ixture was kept in an 80 °C bath for 15 m in . After th at tim e TLC
sh owed com plete rem oval of N-pivaloyl group. After evaporation of th e solven t, work-up
in th e DCM–1 M TEAB buffer m ixture an d lyoph ilization from ben zen e, pure product 10
was obtain ed as a wh ite am orph ous solid. Yield 2.44 g (98%). RF 0.34 (DCM–MeOH 95:5).
3
3
1H NMR (300 MHz, CDCl3): 7.42 d, J5-6 = 8.1, 1 H (H6); 6.09 d, J1′-2′ = 5.7, 1 H (H1′); 5.75 d,
3J5-6 = 8.1, 1 H (H5); 5.26 br, 2 H (H2′ an d H3′); 4.41 dd, J5′a-5′b = 11.2, J4′-5′a = 2.4, 1 H
2
3
2
3
(H5′a); 4.28 m , 1 H (H4′); 4.27 dd, J5′a-5′b = 11.2, J4′-5′b = 1.7, 1 H (H5′b); 1.16, 1.21, 1.25,
3 × s, 3 × 9 H (3 × Pv). 13C NMR (101 MHz, CDCl3): 177.03, 177.09, 177.79 (C=O, Pv);
Collect. Czech. Chem. Commun. 2010, Vol. 75, No. 1, pp. 33–57

54
Sobkowski:
163.02 (C4); 150.10 (C2); 138.86 (C6); 103.21 (C5); 86.48 (C1′); 80.67 (C4′); 72.76 (C3′);
70.20 (C2′); 63.29 (C5′); 38.75, 38.76, 38.79 (C(CH3)3); 26.92, 27.00 an d 27.15 (C(CH3)3).
Typical Procedure for 5′-O-Depivaloylation
N3,5′-O-Dipivaloylth ym idin e (4; 2.05 g, 5 m m ol) was dissolved in 0.5 M m eth an olic NaOH
(100 m l) an d th e reaction m ixture was kept at room tem perature un til TLC sh owed com -
plete deprotection (ca. 1 h ). Th e reaction m ixture was cooled to 0 °C an d n eutralized with
0.5 M aqueous HCl (90 m l), con cen trated un der reduced pressure to ca. 20 m l an d extracted
with DCM (3 × 200 m l). Th e organ ic ph ase was dried with an h ydrous Na2SO4 an d evapo-
rated un der reduced pressure yieldin g com poun d 2 as a wh ite foam . Yield 1.40 g (86%).
Typical Procedure for N3-Depivaloylation
N3,5′-O-Dipivaloylth ym idin e (4; 2.05 g, 5 m m ol) was suspen ded in water (50 m l) an d
refluxed un til TLC sh owed com plete deprotection of th e N3 position (ca. 15 m in ). Th e reac-
tion m ixture was cooled to 0 °C, n eutralized with 1 M NaOH (5 m l) an d worked-up as above.
Yield of PvT (3) 1.47 g (90%).
Kin etics of Acylation of Th ym idin e at 5′-OH an d Th ym in e Moiety
3′-O-(Dim eth oxytrityl)-5′-O-pivaloylth ym idin e (7) an d N3,3′-O-diben zoylth ym idin e were
used for determ in ation of th e rates of acylation of N3H an d 5′-OH groups of th ym idin e, re-
spectively. Nucleoside derivative (0.1 m m ol) was dried by co-evaporation with toluen e, dis-
solved in MeCN or 1:1 MeCN–pyridin e (1 m l) con tain in g an appropriate am oun t of am in e
(TEA, pyridin e, DMAP or th eir m ixtures, usually 5 equiv.) an d acyl ch loride (3 equiv.) was
added. Durin g reaction s 50-µl sam ples were tran sferred to 10% aqueous MeCN (500 µl) an d
an alyzed by TLC after com pletion of th e reaction . In selected experim en ts th e sam ples were
an alyzed also im m ediately after sam plin g. Th e t1/2 was assign ed wh en th e spots due to th e
substrate an d th e product were of th e sam e in ten sity.
Kin etics of Hydrolysis
A 0.02 M aqueous solution of acylated th ym idin e derivative was m ixed with th e sam e volum e
of a 0.1 M ph osph ate buffer or an oth er in vestigated solution an d th e m ixture was kept at
room tem perature or in water bath at 80 or 90 °C (±1 °C). Durin g th e reaction 10-µl sam ples
were taken , diluted with 100 µl of aceton itrile an d kept at 0 or at –20 °C if th e reaction was
con tin ued for a lon ger tim e th an 4 h . After com pletion of h ydrolysis, 10-µl sam ples were
applied to a TLC plate (silica gel F254), developed an d an alyzed visually by com parin g rela-
tive in ten sities of substrate/product spots at various stages of th e reaction un der a UV lam p.
Each experim en t was perform ed th ree tim es an d a m ean value was used for calculation s. For
buffers at pH > 4 a sigm oid deviation from lin earity was observed in th e latter stages of th e
reaction (> ca. 80% of con version ). In th ose cases th e in itial lin ear data poin ts were taken
for calculation s of kobs (Fig. 5a).
Base Hydrolysis of N3-(Trim eth ylacetyl)th ym idin e (TPv; 2)
N3-(Trim eth ylacetyl)th ym idin e (2; 33 m g, 0.1 m m ol) was dissolved in 0.5 M NaOH an d th e
reaction progress was followed by TLC an d UV spectroscopy (200–400 n m ). To th is en d,
Collect. Czech. Chem. Commun. 2010, Vol. 75, No. 1, pp. 33–57

Protection for 4-Oxopyrimidine Moieties
55
aliquots of th e reaction m ixture were an alyzed eith er in 0.5 M NaOH solution or in 0.1
M
ph osph ate buffers of pH 2, 4, 6, 7, an d 8 (for TLC, 50 µl aliquot in 1 m l of a buffer; for UV,
5 µl aliquot in 2.5 m l of a buffer).
4,6-Dih ydroxy-5-m eth yl-1-trim eth ylacetyl-3,4-dih ydropyrim idin -2(1H)-on e (12)
N3-(Trim eth ylacetyl)th ym idin e (2; 660 m g, 2.0 m m ol) was dissolved in 0.5 M NaOH (40 m l).
After 20 m in 5 M HCl was added to th e reaction m ixture un til pH ~ 2 was reach ed. With in
several m in utes wh ite fin e crystals started to precipitate. After 15 m in th e crystals were
filtered off, wash ed twice with water an d dried un der vacuum . Yield 30%. RF 0.51
(DCM–MeOH 9:1). 1H NMR (400 MHz, DMSO-d6): 12.53 br s, 1 H (NH, exch an ge with wa-
ter); 11.96 d, 3J = 11.2, 1 H (O6H, exch an ge with water); 10.32 s, 1H (O4H, exch an ge with
water); 7.34 d, 3J = 11.2, 1 H (H6); 1.78 s, 3 H (5-Me); 1.18 s, 9 H (Pv). 13C NMR (101 MHz,
DMSO-d6): 179.80 (C=O, Pv); 168.76 (C4); 151.37 (C2); 132.83 (C6); 105.41 (C5); ~39
(C(CH3)3, overlapped with DMSO); 26.22 (C(CH3)3); 16.61 (C5). ESI MS(+), m/z: 251.0987
[M + Na]+.
N-[(β-D-2-Deoxyribosyl)carbam oyl]-3-h ydroxy-2-m eth yl-N-(trim eth ylacetyl)acrylam ide (13)
N3-(Trim eth ylacetyl)th ym idin e (2; 660 m g, 2.0 m m ol) was dissolved in 0.5 M NaOH (40 m l).
After 20 m in Dowex 50W (H+ form ) was added to th e reaction m ixture un til pH ~ 6 was
reach ed. Th e resin was filtered off, wash ed twice with water an d th e solven t was evaporated
to dryn ess. Com poun d 13 was isolated by colum n ch rom atograph y (silica gel, 0–30% MeOH
3
in DCM). Yield 44%. RF 0.23 (DCM–MeOH 7:3). 1H NMR (300 MHz, D2O): 6.14 t, J1′-2′
=
4
3
3
7.1, 1 H (H1′); 5.79 d, J6-5Me = 1.7, 1 H (H6); 4.26 dt, J3′-4′ = 6.0, J2′-3′ = 3.8, 1 H (H3′); 3.84 m ,
1 H (H4′); 3.67 dd, J5′a-5′b = 12.6, J4′-5′a = 3.9, 1 H (H5′a); 3.54 dd, J5′a-5′b = 12.6, J4′-5′b
5.7, 1 H (H5′b); 1.96 d, J6-5Me = 1.7, 3 H (5-Me); 1.19 s, 9 H (Pv). 13C NMR (101 MHz, D2O):
181.25 (C=O, Pv); 175.72 (C4); 154.25 (C4); 144.72 (C2); 116.95 (C5); 84.92 an d 85.23 (C1′
an d C4′); 70.47 (C3′); 61.44 (C5′); 40.00 (C2′); 35.75 (C(CH3)3); 25.94 (C(CH3)3); 16.88 (C5).
ESI MS(+), m/z: 367.1414 [M + Na]+.
2
3
2
3
=
4
The author would like to thank Prof. A. Kraszewski and Prof. J. Stawinski for helpful discussions,
Mrs E. Pasternak and Mrs A. Teubert for running NMR experiments, and Dr. Ł. Marczak for MS
spectra. Financial support from the Polish Ministry of Science and Higher Education is gratefully
acknowledged.
REFERENCES
1. Sobkowski M. in: Collection Symposium Series (M. Hocek, Ed.), Vol. 10, p. 277. Institute of
Organic Chemistry and Biochemistry, Academy of Sciences of the Czech Republic,
Prague 2008.
2. Welch C. J., Zhou X. X., Nyilas A., Remaud G., Chattopadhyaya J. in: Biophosphates and
Their Analogues – Synthesis, Structure, Metabolism and Activity (K. S. Bruzik and W. J. Stec,
Eds), p. 107. Elsevier Science, Amsterdam 1987.
3. Wozniak L. A., Bukowiecka-Matusiak M., Burzynska-Pedziwatr I., Stec W. J.: Tetrahedron
Lett. 2009, 50, 2620.
4. Higashida R., Oka N., Kawanaka T., Wada T.: Chem. Commun. 2009, 2466.
Collect. Czech. Chem. Commun. 2010, Vol. 75, No. 1, pp. 33–57

56
Sobkowski:
5. Acharya P., Cheruku P., Chatterjee S., Acharya S., Chattopadhyaya J.: J. Am. Chem. Soc.
2004, 126, 2862.
6. Furukawa Y., Kobayashi K., Kanai Y., Honjo M.: Chem. Pharm. Bull. 1965, 13, 1273.
7. Borsting P., Freitag M., Nielsen P.: Tetrahedron 2004, 60, 10955.
8. Mitsunobu O.: Synthesis 1981, 1.
9. He G. X., Bischofberger N.: Nucleosides Nucleotides 1997, 16, 257.
10. Reese C. B., Ubasawa A.: Tetrahedron Lett. 1980, 21, 2265.
11. Reese C. B., Ubasawa A.: Nucleic Acids Symp. Ser. 1980, 7, 5.
12. Jones S. S., Reese C. B., Sibanda S., Ubasawa A.: Tetrahedron Lett. 1981, 22, 4755.
13. Sung W. L.: Nucleic Acids Res. 1981, 9, 6139.
14. Wallis M. P., Mahmood N., Fraser W.: Farmaco 1999, 54, 83.
15. Zhou X. X., Welch C. J., Chattopadhyaya J.: Acta Chem. Scand., Ser. B 1986, 40, 806.
16. a) Sekine M.: J. Org. Chem. 1989, 54, 2321; b) Kamimura T., Masegi T., Sekine M., Hata T.:
Tetrahedron Lett. 1984, 25, 4241
17. Sung W. L.: J. Org. Chem. 1982, 47, 3623.
18. Adamiak R. W., Biala E., Skalski B.: Nucleic Acids Res. 1985, 13, 2989.
19. Zhou X. X., Chattopadhyaya J.: Tetrahedron 1986, 42, 5149.
20. Reese C. B., Skone P. A.: J. Chem. Soc., Perkin Trans. 1 1984, 1263.
21. Himmelsbach F., Schulz B. S., Trichtinger T., Charubala R., Pfleiderer W.: Tetrahedron
1984, 40, 59.
22. Welch C. J., Bazin H., Heikkila J., Chattopadhyaya J.: Acta Chem. Scand., Ser. B 1985, 39,
203.
23. Welch C. J., Chattopadhyaya J.: Acta Chem. Scand., Ser. B 1983, 37, 147.
24. Matsuzaki J., Hotoda H., Sekine M., Hata T.: Tetrahedron Lett. 1984, 25, 4019.
25. Yamashita J., Takeda S., Matsumoto H., Terada T., Unemi N., Yasumoto M.: Chem.
Pharm. Bull. 1987, 35, 2090.
26. Sekine M., Fujii M., Nagai H., Hata T.: Synthesis 1987, 1119.
27. Zhu X. F., Williams H. J., Scott A. I.: Synth. Commun. 2003, 33, 1233.
28. Kamimura T., Masegi T., Hata T.: Chem. Lett. 1982, 965.
29. Pedersen S. L., Nielsen P.: Org. Biomol. Chem. 2005, 3, 3570.
30. Brown D. M., Haynes L. J., Todd A. R.: J. Chem. Soc. 1950, 3299.
31. Michelson A. M., Todd A. R.: J. Chem. Soc. 1955, 2632.
32. Rammler D. H., Khorana H. G., Lapidot Y.: J. Am. Chem. Soc. 1963, 85, 1989.
33. Smith M., Khorana H. G., Rammler D. H., Goldberg I. H.: J. Am. Chem. Soc. 1962, 84, 430.
34. Gilham P. T., Khorana H. G.: J. Am. Chem. Soc. 1958, 80, 6212.
35. Gait M. J. (Ed.): Oligonucleotide Synthesis: A Practical Approach. Oxford University Press,
Oxford 1984.
36. Yamashita J., Matsumoto H., Kobayashi K., Noguchi K., Yasumoto M., Ueda T.: Chem.
Pharm. Bull. 1989, 37, 2287.
37. Weimann G., Khorana H. G.: J. Am. Chem. Soc. 1962, 84, 4329.
38. Vial J.-M., Agback P., Chattopadhyaya J.: Nucleosides Nucleotides Nucleic Acids 1990, 9, 245.
39. Seio K., Wada T., Sekine M.: Helv. Chim. Acta 2000, 83, 162.
40. Ora M., Maki E., Poijarvi P., Neuvonen K., Oivanan M., Lonnberg H.: J. Chem. Soc., Perkin
Trans. 2 2001, 881.
41. Gerzon K., Kau D.: J. Med. Chem. 1967, 10, 189.
42. Mitsunobu O., Kimura J., Fujisawa Y.: Bull. Chem. Soc. Jpn. 1972, 45, 245.
43. Liguori A., Perri E., Sindona G., Uccela N.: Tetrahedron 1988, 44, 229.
Collect. Czech. Chem. Commun. 2010, Vol. 75, No. 1, pp. 33–57

Protection for 4-Oxopyrimidine Moieties
57
44. Wong C. H., Chen S. T., Hennen W. J., Bibbs J. A., Wang Y. F., Liu J. L. C., Pantoliano M. W.,
Whitlow M., Bryan P. N.: J. Am. Chem. Soc. 1990, 112, 945.
45. Ferrero M., Gotor V.: Monatsh. Chem. 2000, 131, 585.
46. Li N., Zong M. H., Ma D.: Eur. J. Org. Chem. 2008, 5375.
47. Breiner R. G., Rose W. C., Dunn J. A., MacDiarmid J. E., Bardos T. J.: J. Med. Chem. 1990,
33, 2596.
48. Ensminger W. D., Rosowsky A.: Biochem. Pharmacol. 1979, 28, 1541.
49. Sobkowski M., Stawinski J., Kraszewski A.: New. J. Chem. 2009, 33, 164.
50. Sobkowski M., Stawinski J., Sobkowska A., Kraszewski A.: J. Chem. Soc., Perkin Trans. 1
1994, 1803.
51. Kers A., Kers I., Stawinski J., Sobkowski M., Kraszewski A.: Tetrahedron 1996, 52, 9931.
52. Dubey I. Y., Efimov V. A., Lyapina T. V., Fedoryak D. M.: Bioorg. Khim. 1992, 18, 911.
53. Garegg P. J., Regberg T., Stawinski J., Stromberg R.: Nucleosides Nucleotides 1987, 6, 655.
54. Sigurdsson S., Stromberg R.: Nucleosides Nucleotides Nucleic Acids 2003, 22, 1.
55. Perrin D. D.: Dissociation Constants of Organic Bases in Aqueous Solution. Butterworth,
London 1965.
56. Pragnacharyulu P. V. P., Abushanab E.: Tetrahedron Lett. 1997, 38, 7025.
57. Orita A., Tanahashi C., Kakuda A., Otera J.: J. Org. Chem. 2001, 66, 8926.
58. De Nino A., Liguori A., Procopio A., Roberti E., Sindona C.: Carbohydr. Res. 1996, 286, 77.
59. Ogilvie K. K.: Can. J. Chem. 1973, 51, 3799.
60. Sekine M., Nakanishi T.: J. Org. Chem. 1990, 55, 924.
61. Hossain N., Rozenski J., De Clercq E., Herdewijn P.: Tetrahedron 1996, 52, 13655.
62. Altmann K. H., Chiesi C. S., Garcia-Echeverria C.: Bioorg. Med. Chem. Lett. 1997, 7, 1119.
63. Ramasamy K. S., Stoisavljevic V.: Nucleosides Nucleotides Nucleic Acids 1999, 18, 1845.
64. Maguire A. R., Hladezuk I., Ford A.: Carbohydr. Res. 2002, 337, 369.
65. Kamaike K., Uemura F., Yamakage S. I., Nishino S., Ishido Y.: Nucleosides Nucleotides
Nucleic Acids 1987, 6, 699.
66. Cadet J., Ducolomb R., Teoule R.: Tetrahedron 1977, 33, 1603.
Collect. Czech. Chem. Commun. 2010, Vol. 75, No. 1, pp. 33–57