Novel water-soluble prodrugs of acyclovir cleavable by the dipeptidyl-peptidase IV (DPP IV/CD26) enzyme

Article abstract of DOI:10.1016/j.ejmech.2013.10.001

We herein report for the first time the successful use of the dipeptidyl peptidase IV (DPPIV/CD26) prodrug approach to guanine derivatives such as the antiviral acyclovir (ACV). The solution- and solid-phase synthesis of the tetrapeptide amide prodrug 3 and the tripeptide ester conjugate 4 of acyclovir are reported. The synthesis of the demanding tetrapeptide amide prodrug of ACV 3 was first established in solution and successfully transferred onto solid support by using Ellman's dihydropyran (DHP) resin. In contrast with the valyl ester prodrug (valacyclovir, VACV), the tetrapeptide amide prodrug 3 and the tripeptide ester conjugate 4 of ACV proved fully stable in PBS. Both prodrugs converted to VACV (for 4) or ACV (for 3) upon exposure to purified DPPIV/CD26 or human or bovine serum. Vildagliptin, a potent inhibitor of DPPIV/CD26 efficiently inhibited the DPPIV/CD26-catalysed hydrolysis reaction. Both amide and ester prodrugs of ACV showed pronounced anti-herpetic activity in cell culture and significantly improved the water solubility in comparison with the parent drug.

Full text of DOI:10.1016/j.ejmech.2013.10.001

European Journal of Medicinal Chemistry 70 (2013) 456e468
Contents lists available at ScienceDirect
European Journal of Medicinal Chemistry
journal homepage: http://www.elsevier.com/locate/ejmech
Original article
Novel water-soluble prodrugs of acyclovir cleavable by the dipeptidyl-
peptidase IV (DPP IV/CD26) enzyme
,
Alberto Diez-Torrubia a, Silvia Cabrera a 1, Sonia de Castro a, Carlos García-Aparicio a,
Gwenn Mulder a, Ingrid De Meester b, María-José Camarasa a, Jan Balzarini c,
,
Sonsoles Velázquez a
*
a Instituto de Química Médica (IQM-CSIC), c/ Juan de la Cierva 3, E-28006 Madrid, Spain
b Laboratory of Medical Biochemistry, University of Antwerp, Universiteitsplein 1, B-2610 Antwerp, Belgium
c Rega Institute for Medical Research, KU Leuven, Minderbroedersstraat 10, B-3000 Leuven, Belgium
a r t i c l e i n f o
a b s t r a c t
Article history:
We herein report for the first time the successful use of the dipeptidyl peptidase IV (DPPIV/CD26)
prodrug approach to guanine derivatives such as the antiviral acyclovir (ACV). The solution- and solid-
phase synthesis of the tetrapeptide amide prodrug 3 and the tripeptide ester conjugate 4 of acyclovir
are reported. The synthesis of the demanding tetrapeptide amide prodrug of ACV 3 was first established
in solution and successfully transferred onto solid support by using Ellman’s dihydropyran (DHP) resin.
In contrast with the valyl ester prodrug (valacyclovir, VACV), the tetrapeptide amide prodrug 3 and the
tripeptide ester conjugate 4 of ACV proved fully stable in PBS. Both prodrugs converted to VACV (for 4) or
ACV (for 3) upon exposure to purified DPPIV/CD26 or human or bovine serum. Vildagliptin, a potent
inhibitor of DPPIV/CD26 efficiently inhibited the DPPIV/CD26-catalysed hydrolysis reaction. Both amide
and ester prodrugs of ACV showed pronounced anti-herpetic activity in cell culture and significantly
improved the water solubility in comparison with the parent drug.
Received 2 September 2013
Received in revised form
27 September 2013
Accepted 1 October 2013
Available online 9 October 2013
Keywords:
Acyclovir
Antiviral
Prodrug
Peptide
Dipeptidyl peptidase IV
Solid-phase synthesis
Ó 2013 Elsevier Masson SAS. All rights reserved.
1. Introduction
antiviral drug (glycinamide) through the specific action of DPPIV/
CD26 [7]. Based on these results, we previously reported [8] a novel
Dipeptidyl peptidase IV (DPPIV, also termed CD26) is a member
of the prolyl oligopeptidase family, a group of atypical serine pro-
teases able to hydrolyse the prolyl bond, that mediates the activities
of a number of regulatory peptides [1e4]. It is endowed with an
interesting (dipeptidyl) peptidase catalytic activity and specifically
removes dipeptides from the N-terminus of peptide substrates that
have a proline or, to a lesser extent an alanine at the penultimate
amino acid position. It exists as an extracellular membrane-bound
peptidase, highly expressed on endothelial cells, epithelial cells and
lymphocytes, and as a soluble form that is found in plasma [5,6].
In 2005, it was first demonstrated that a synthetic small mole-
cule [GPG-NH2 (glycylprolylglycinamide)] can be converted to an
type of DPPIV/CD26-directed prodrug technology that provides
conjugates [(Xaa-Pro)n]e[drug] of therapeutic drugs with a di- (or
tetra) peptide moiety as a carrier (Fig. 1). In these prodrugs the
compounds/drugs and the peptide moiety are directly coupled via
an amide bond which is specifically cleaved by DPPIV/CD26 (Fig. 1).
The presence of a proline at the penultimate N-position protects the
amino acid sequence against non-specific proteolytic degradation
since many exopeptidases do not recognize such amino acid
sequence [9]. This approach was successfully applied to a broad
variety of amine-containing drugs having a free amino group on
alkyl chains (the N-3 aminopropyl derivative of the anti-HIV TSAO
compounds) [10,11] and on heteroaromatic (6-aminoquinoline),
carbohydrate (doxorubicine), heterocyclic pyrimidine (ara-C and
TSAO-m5C) and purine (ara-A) rings [12]. Our data revealed that
purified DPPIV/CD26 could recognize these prodrugs as substrates
and efficiently released the parent compound. The DPPIV/CD26-
specific inhibitor Vildagliptin could efficiently block the enzy-
matic conversion in human, bovine and murine serum [12]. Inter-
estingly, it was possible to modulate the enzymatic and serum
Abbreviations: DPP IV, dipeptidyl peptidase IV; ACV, acyclovir; VACV, valacy-
clovir; DKP, diketopiperazine; DHP, dihydropyran.
* Corresponding author. Tel.: þ34 91 258 74 58; fax: þ34 91 5644853.
E-mail addresses: iqmsv29@iqm.csic.es, sonsoles@iqm.csic.es (S. Velázquez).
Present address: Departamento de Química Inorgánica, Facultad de Ciencias,
1
Universidad Autónoma de Madrid, 28049 Madrid, Spain.
0223-5234/$ e see front matter Ó 2013 Elsevier Masson SAS. All rights reserved.
http://dx.doi.org/10.1016/j.ejmech.2013.10.001

A. Diez-Torrubia et al. / European Journal of Medicinal Chemistry 70 (2013) 456e468
457
DPPIV/CD26
Esterases and/or
chemical hydrolysis
DPPIV/CD26
O
C
O
H
N
O
C
(Xaa-Pro)n
Drug
H
H
N
H-Xaa-Pro
C
CH(R)
O Drug
Peptide promoiety
(n = 1, 2)
I
Peptide promoiety
connector
(amino acids)
O
NH2
II
spontaneous
degradation
NH2
(CH2)3
N
O
DPPIV/CD26
NH
N
O
O
C
O
N
self-cleavage
spacer
O
N
SiO
H2N
SiO
H2N
O
Drug
H-Xaa-Pro
Peptide promoiety
III
O
OSi
O
OSi
S
S
O
O
O
O
TSAO-m5C
TSAO-NAP
NH2
NH2
N
N
N
O
O
N
O
N
N
O
N
HO
HO
N
O
HO
N
NH
HO
N
O
HO
N
HO
OH
cytarabine (ara-C)
OH
O
O
vidarabine (ara-A)
OH
O
O
OH
OH
didanosine
HO
Cf1743
H2N
OH
OH
N
H
N
O
O
OH
O
O
6-aminoquinoline
(6-AQ)
O
Doxorubicine (Dox)
NHCOCH3
NH2
OH
acetaminophen
propranolol
O
Fig. 1. Application of the DPP IV/CD26-based prodrug approach to amine-containing
drugs.
N
hydrolysis rate (half-life) of the prodrug conjugates by changing the
nature and the length of the peptide promoiety [13].
N
O
We recently reported [14,15] the application of this prodrug
approach to hydroxy-containing drugs of different nature (primary,
secondary, tertiary or aromatic hydroxyl groups) (Fig. 2). In this
case, the peptide promoiety cannot be linked directly to the hy-
droxyl group through an ester bond because the DPPIV/CD26
enzyme solely recognizes amide bonds. Thus, tripartite conjugates
[Xaa-Pro]e[connector]e[drug] bearing heterobifunctional con-
nectors to link the peptide moiety to the hydroxyl group of the drug
were designed. Tripartite prodrugs bearing an amino acid (i.e.
valine) as a heterobifunctional connector to link the dipeptide to
the hydroxyl group of the drug through an ester bond (released by
chemical or enzymatic hydrolysis) were first studied (conjugates of
general formulae II, Fig. 2). The hydroxy-containing prodrugs
showed different susceptibilities to hydrolysis by DPPIV/CD26 and
serum depending on the nature of the compound. Several prodrugs
showed a remarkable increase in water solubility and a markedly
O
OH
camptothecin
Fig. 2. Two-step activation of DPP IV/CD26-based prodrugs of hydroxy-containing
drugs.
enhanced oral bioavailability in mice compared to the free parent
compound [14]. We also investigated self-cleavage connectors in
the design of novel two-step activation prodrugs (conjugates of
general formulae III, Fig. 2) [16]. Both cyclization and electronic
cascade self-cleavage spacers were explored. Prodrugs based on
dipeptide cyclization spacers efficiently released the parent
nucleoside upon hydrolysis by DPPIV/CD26 followed by sponta-
neous cyclization of the dipeptide conjugate intermediate via

458
A. Diez-Torrubia et al. / European Journal of Medicinal Chemistry 70 (2013) 456e468
acylated at the primary hydroxyl group. Valacyclovir (VACV), the L-
O
N
O
N
valine ester prodrug of ACV [22,23], was found to metabolise easily
upon oral administration (by esterases and/or biphenyl hydrolase-
like protein [24]) and exhibited four-folded higher bioavailability
than ACV [25]. The oral absorption of this prodrug appears to be
mediated by its substrate activity for the PepT1 peptide transporter
abundantly present in the intestinal mucosa. The amino acid ester
prodrugs of ACV have low stability at physiological pH and their
delivery is still not optimal. Dipeptide [26,27] or peptidomimetic
[28] prodrugs of ACV were also synthesized and evaluated for their
chemical stability and pharmacokinetics.
The aim of this study was to assess for the first time the appli-
cability of the DPPIV/CD26 prodrug strategy to guanidine drugs
such as the antiviral acyclovir (1). In this compound both the amino
and the hydroxyl group are amenable for prodrug derivatization.
Depending on the site of attachment of the peptide promoiety, both
peptidyl amide and ester prodrugs of acyclovir were envisioned
(Fig. 3). As the peptide promoiety, we focused on the Val-Pro
dipeptide (efficiently recognized by the DPPIV/CD26 enzyme) and
the Val-Pro-Val-Pro tetrapeptide for the peptidyl amide prodrugs 2
and 3 and on the Val-Pro dipeptide and on the valine as connector
for the peptidyl ester prodrug 4 based on our previous results
(Fig. 3) [8,12,14,15].
N
N
N
N
HN
HN
O
C
H2N
HO
H
(Val Pro)n
HN
HO
O
O
acyclovir (1)
2
(n = 1)
(n = 2)
O
3
N
N
HN
H2N
N
C
O
O
H
Val Pro Val
O
4
Fig. 3. Structure of acyclovir (1) and target amide and ester prodrugs (2e4).
diketopiperazine (DKP). A markedly increased transport through
CaCo-2 cell monolayers and oral bioavailability was observed for
several tetrapeptide prodrugs [16].
9-[2-(Hydroxyethoxy)methyl]guanine (acyclovir 1, Fig. 3) is a
widely used therapeutic agent for the treatment of herpes virus
infections [17]. Acyclovir (ACV) is poorly water-soluble and has an
oral bioavailability of 10e20% [18]. In order to overcome these
limitations, a number of acyclovir prodrugs were previously re-
ported [19e21]. Because the reactivity of the 40-hydroxyl group is
higher than that of the 2-amino group, most of these produgs were
The preparation of target peptidyl amide prodrugs 2 and 3
acylated at the exocyclic amino group of the guanine base repre-
sents a synthetic challenge due to the intrinsically low nucleophi-
licity of this amino function and the low solubility of this
compound. We investigated several procedures for the acylation of
the 2-amino group of acyclovir with peptides using both solution-
phase and solid-phase strategies. It should be noted that although
the amino group of guanine nucleosides (such as ACV) has been
2
O
O
N
H2N
Cbz HN
O
O
N
N
N
HN
HN
O
O
O
O
O
N
N
N
N
N
N
H
H
N
(ii)
O
O
HN
+
5
O
O
O
6
7
DKP (8)
(unstable)
O
N
O
N
(i)
N
N
HN
HN
N
N
O
Val Pro HN
R
Val Pro
(vii)
Val Pro HN
Cbz Val Pro
(iii)
HO
O
N
HN
11 R = Cbz
R = H
O
O
(v)
3
O
N
H2N
N
+
9
O
O
O
(iv)
O
O
N
N
(v), (vi)
HN
5
N
HN
N
H
Cbz Val
Val Pro HN
12
N
Pro
N
Pro HN
Cbz
O
O
O
10
Scheme 1. Synthetic strategies in solution for the synthesis of target acyclovir amide prodrugs 2 and 3. Reagents and conditions: (i) Cbz-Val-Pro-OH, TFFH, collidine, DMF, 80 ꢀC; (ii)
2, Pd(C), MeOH; (iii) Cbz-Val-Pro-Val-Pro-OH, TFFH, collidine, DMF, 80 ꢀC; (iv) Cbz-Pro-OH, TFFH, collidine, DMF, 80 ꢀC; (v) H2, Pd(C), CH3OH; (vi) Cbz-Val-Pro-Val-OH, pyBOP, TEA,
DMF, rt; (vii) NH3/CH3OH sat.
H

A. Diez-Torrubia et al. / European Journal of Medicinal Chemistry 70 (2013) 456e468
459
previously protected with simple acyl groups following different
strategies [29,30], acylation with amino acids or peptides has never
been reported.
We herein describe the synthesis, water solubility and chemical
and enzymatic stability studies of target acyclovir prodrugs (2e4)
in the presence and absence of a dipeptidyl peptidase inhibitor.
O
N
O
N
N
N
N
N
HN
HN
H2N
H2N
Val
Cbz Val Pro Val
(i)
O
H
O
O
O
2. Results and discussion
VACV (13)
14
2.1. Chemistry
(ii)
2.1.1. Solution-phase synthesis of acyclovir amide and ester
prodrugs
We first focused on the synthesis of the Val-Pro dipeptide amide
prodrug of acyclovir 2 using a solution-phase approach. The hy-
droxyl function of acyclovir was first protected with an acetyl group
(compound 5 [31]) before the coupling step under standard acet-
ylation conditions (Ac2O/pyridine) (Scheme 1). Initial attempts of
coupling of 5 with the commercially available dipeptide Cbz-Val-
Pro-OH in the presence of a variety of coupling reagents such as
carbodiimides [1,3-diisopropylcarbodiimide (DIPCDI) with N-
hydroxybenzotriazole (HOBt)] or more potent coupling reagents
such as phosphonium-[(benzotriazol-1-yloxy)-tris[pyrrolidino]
phosphonium hexafluorophosphate (PyBOP)] or uronium salts
[hexafluorophosphate salt of the O-(7-azabenzotriazolyl)tetra-
methyl uronium (HATU)] in the presence of TEA or DIEA as base and
dry DMF as solvent failed to give the desired dipeptide ACV con-
jugate 6 (Scheme 1). Increasing amounts of coupling reagents (3e
5 equiv) and higher reaction temperatures (90e150 ꢀC) in a sealed
pressure tube did not afford the coupled product. Thus, the
exocyclic amino group of the guanine moiety of acyclovir appears
difficult to couple with dipeptides under similar conditions previ-
ously used in our group for the synthesis of dipeptide conjugates of
other pyrimidine and purine (adenine) nucleosides [12]. Further
attempts of coupling by in situ generation of protected dipeptide
chlorides using thionyl chloride or triphosgene [bis(tri-
chloromethyl)carbonate (BTC)] with protected compound 5 in the
presence of pyridine and DMAP or 2,4,6-collidine as base in dry
DMF also failed in all the experimental conditions investigated
(room temperature or 90e150 ꢀC in a sealed pressure tube). We
next explored coupling reactions via acid fluorides. A remarkable
progress in this field was the development of tetramethyl-
fluoroformamidinium hexafluorophosphate (TFFH) by Carpino
et al. [32] which act by in situ generating amino acid fluoride in
difficult peptide coupling reactions. When protected acyclovir 5
was coupled with Cbz-Val-Pro-OH in the presence of 1.5 equiv. of
TFFH as coupling reagent and 2,4,6-collidine as base in dry DMF at
150 ꢀC the dipeptide conjugate 6 was isolated for the first time
although in low yields (5%) after chromatographic purification.
However, when the coupling reaction was performed at 80 ꢀC in the
presence of 5 equivalents of TFFH, the desired protected dipeptide
conjugate 6 was obtained in higher yield (53%). Similarly to what
has been observed with cytidine and adenosine dipeptide prodrugs
[12], the N-deprotected dipeptide acyclovir conjugate 7 was
chemically unstable. Thus, catalytic hydrogenation of protected
conjugate 6 in the presence of 10% Pd/C in methanol showed that
parent ACV derivative 5 was spontaneously released from the N-
deprotected conjugate 7 with concomitant formation of the known
DKP 8 [33] (Scheme 1).
4
Scheme 2. Synthesis of target acyclovir ester prodrug 4. Reagents and conditions: (i)
Cbz-Val-Pro-OH, PyBOP, TEA, DMF, rt; (ii) H2, Pd(C), MeOH.
tetrapeptide coupling, which consisted of coupling of 5 with an
amino acid followed by coupling with the corresponding tripeptide,
was next performed. The required tripeptide Cbz-Val-Pro-Val-OH
was prepared by solid-phase synthesis using a Fmoc strategy on a
2-chlorotrityl polystyrene resin as previously described [12]. Thus,
coupling of compound 5 with Cbz-Pro-OH in the presence of TFFH
and 2,4,6-collidine in dry DMF at 80 ꢀC (under similar conditions
described above for the dipeptide conjugate 6) gave the Cbz pro-
tected prolyl amide derivative 10 in 65% yield after chromato-
graphic purification (Scheme 1). Catalytic hydrogenation of 10 in
the presence of 10% Pd/C in methanol followed by coupling of the
N-deprotected prolyl intermediate derivative with Cbz-Val-Pro-
Val-OH in the presence of 3 equivalents of PyBOP as coupling re-
agent and TEA as base in dry DMF at room temperature afforded the
desired tetrapeptide conjugate 9 in low overall yield (24% from 10).
Use of other coupling reagents usually employed in amide forma-
tion (HATU/HOAt/DIEA or coupling via acyl halides) and higher
reaction temperatures did not lead to improved yields. Treatment
of compound 9 with methanol saturated with ammonia at room
temperature (removal of the O-acetyl group) gave the O-depro-
tected acyclovir conjugate 11 in a 32% yield together with the tet-
rapeptide guanine derivative 12 in a 26% yield due to CeN bond
cleavage. Finally, catalytic hydrogenation of 11 in the presence of
10% Pd/C in methanol afforded the N-deprotected target amide
prodrug 3 in 83% yield (Scheme 1).
The synthesis of the tripeptide ester prodrug of acyclovir 4 was
next carried out by standard peptide coupling of Cbz-Val-Pro-OH
and commercially available -valacyclovir 13 (Scheme 2). Thus,
L
treatment of the hydrochloride salt of free amine of VACV 13 with
Cbz-Val-Pro-OH in the presence of PyBOP as coupling reagent and
TEA as base in dry DMF at room temperature overnight afforded the
N-protected tripeptide conjugate 14 in good yields (81%) after
chromatographic purification. Finally, removal of the Cbz group of
compound 14 by catalytic hydrogenation in the presence of 10% Pd/
C in methanol gave the target tripeptide ester conjugate of ACV 4 in
67% yield.
The target prodrugs 3 and 4 were characterised by 1H NMR,
mass spectrometry and microanalysis, all data confirming the
structure and purity of the novel compounds.
2.1.2. Solid-phase synthesis of acyclovir amide prodrug 3
We next focused on the synthesis of tetrapeptide amide prodrug
3 in order to obtain chemically stable prodrugs. Coupling of 5 with
Cbz-Val-Pro-Val-Pro-OH [12] under the optimized coupling con-
ditions described above for the dipeptide acyclovir conjugate (TFFH
and 2,4,6-collidine at 80 ꢀC) did not afford the desired tetrapeptide
conjugate 9. An alternative strategy to the difficult direct
We next investigated for the first time a solid-phase method for
acylation of the exocyclic amino group of acyclovir by attaching the
hydroxyl group of ACV to a solid-support. In this approach, the use
of a large excess of reagents, which can be easily eliminated
through simple resin washes, may facilitates the completion of the
notoriously difficult coupling reaction of peptides with the guanine

460
A. Diez-Torrubia et al. / European Journal of Medicinal Chemistry 70 (2013) 456e468
moiety observed above in the solution-phase strategy. Our solid-
phase strategies for the synthesis of the tetrapeptide amide pro-
drug of acyclovir 3 involve anchoring acyclovir through the hy-
droxyl group to an appropriate solid-support, followed by a
stepwise coupling of N-protected amino acids (method A), direct
tetrapeptide coupling (method B) or a [1 þ 3] fragment coupling
(method C) and a final N-deprotection/cleavage step.
TFA in DCM and analytical HPLC (254 nm) determination. No sig-
nificant differences in loadings were observed between PS- and
PEG-based 2-chlorotrityl resins. Supported acyclovir compound 15
(Scheme 3) was then reacted with Fmoc-Pro-OH in dry DMF in the
presence of a variety of coupling reagents (TFFH, HATU, PyAOP,
DIPCDI, HCTU), additives (HOAt) and bases (2,4,6-collidine, DIEA or
DMAP). The coupling reactions were heated in a pressure tube or
using microwave irradiation at 80 ꢀC. In all cases, only traces of the
desired prolyl acyclovir derivative 16 were detected by HPLC-MS
after cleavage (1% TFA) of a small amount of resin being the ma-
jor compound acyclovir.
We first focused on the selection of a convenient solid-support
to anchor acyclovir via the hydroxyl group followed by an explor-
atory model coupling reaction with a conveniently N-protected
amino acid (Scheme 3). We initially investigated the widely used
acid labile 2-chlorotrityl chloride resin in solid phase peptide
synthesis using Fmoc strategy. The steric bulk of the 2-chlorotrityl
resin prevents diketopiperazine formation during the attachment
of the first two amino acids. This would allow us to explore the
stepwise coupling of amino acids (method A) that could not be used
in the solution-phase approach. Both, the classical polystyrene
(PS)-based solid support and the more hydrophilic ChemMatrixÒ, a
totally poly(ethylene glycol)(PEG)-based resin developed by
Albericio et al. [34] were used. This resin swells better than PS-
based resins in polar solvents such as acetonitrile, DMF or DMSO
that were required to dissolve the highly polar acyclovir compound.
A variety of bases (DIEA or py), solvents (DMF or DMSO) and re-
action conditions (rt or 60 ꢀC) were explored for the loading of the
primary alcohol of acyclovir onto the 2-chlorotrityl resin (PS or
PEG). The best loading conditions (DIEA, DMSO, rt) were deter-
mined by cleavage of a small amount of resin in the presence of 1%
In order to reduce steric hindrance of the resin, that may explain
the failure of the amino acid coupling (even in the PEG resin which
minimizes the steric effects caused by the support), in a second
attempt, we chose base-labile aminomethyl-resins (both PS and
ChemMatrixÒ PEG-resin) using
a BOC strategy (Scheme 3).
Acyclovir was first esterified with succinic anhydride according to
the method of Colla et al. [22a,35], followed by coupling of com-
pound 17 through its carboxylic group to the aminomethyl-resins.
The efficiency of the coupling reaction was easily monitored by
the Kaiser ninhydrin test. Coupling of 17 to the aminomethyl-PS
resin in the presence of DIPCDI/HOAt in DMSO at room tempera-
ture for 2 h did not result in a negative Kaiser test even after three
coupling cycles (incomplete coupling). However, reaction of 17 to
the more soluble ChemMatrixÒ PEG-support under the same
coupling conditions provided a negative Kaiser test after two
coupling cycles. The resulting supported-bound acyclovir 18
O
O
N
HN
N
HN
N
R Pro HN
N
N
H2N
N
Cl
O
Fmoc-Pro-OH
(ii)
cleavage
(iii)
HO
O
O
16 R = Fmoc
19 R = Boc
15
cleavage
(vii)
2-chlorotritylchloride resin
(PS or PEG)
(i)
O
O
N
Boc-Pro-OH
(vi)
N
N
N
HN
HN
H2N
NH2
H2N
N
N
(PS or PEG)
(v)
(iv)
HOOC
(viii)
HNOC
ACV
(1)
O
O
O
O
O
O
17
18
DHP resin
O
N
HN
N
H2N
N
O
O
O
Fmoc-Pro-OH
(ix)
cleavage
(x)
O
16
20
Scheme 3. Resin selection for the solid-phase synthesis of acyclovir prodrugs. Reagents and conditions: (i) DIEA, DMSO, rt; (ii) Fmoc-Pro-OH, HATU, HOAt, or PyAOP, HOAt, or
DIPCDI, DMAP, TFFH, DIEA or HCTU, DIPEA, DMF, rt or 90 ꢀC; (iii) TFA, 1% CH2Cl2; (iv) succinic anhydride, pyridine, DMF, rt; (v) DIPCDI, HOAt, DMSO, rt; (vi) Boc-Pro-OH, HATU, HOAt,
or PyAOP, HOAt, DMF, rt; (vii) DBU or 2 N NH3, EtOH; (viii) PPTS, DMF, 80 ꢀC; (ix) Fmoc-Pro-OH, HCTU, DIPEA, DMF, MW, 80 ꢀC; (x) TFA:H2O, 95:5, rt.

A. Diez-Torrubia et al. / European Journal of Medicinal Chemistry 70 (2013) 456e468
461
(Scheme 3) was then coupled to Boc-Pro-OH in the presence of
HATU/HOAt and PyAOP/HOAt as coupling reagents in dry DMF.
Monitoring of this reaction was carried out by cleaving a small
amount of resin from the reaction and then analysing the crude by
HPLC-MS. Unfortunately, the standard basic cleavage conditions
(DBU or 2 N NH3 in ethanol) were found to be completely inef-
fective and no acyclovir or Boc-prolyl acyclovir 19 were detected by
HPLC-MS. When stronger cleavage conditions were used (ammonia
gas overnight) CeN bond cleavage occurred and a signal corre-
sponding to the guanine moiety was detected in the HPLC-MS
chromatogram as the major product.
At this point, we examined Ellman’s dihydropyran (DHP) PS-
resin as an alternative solid phase support for the inmobilization
of primary or secondary alcohols [36]. The DHP linker was chosen
because it is less sterically hindered than the trityl linkers and the
bound alcohols can be cleaved from the resin under acidic condi-
tions. Acyclovir (1) was coupled to the DHP resin (0.90 mmol/g of
resin) by employing pyridinium p-toluenesulfonate (PPTS) in DMF
at 80 ꢀC [36] to achieve maximum dissolution. The reaction was
carried out using microwave irradiation on a Biotage Initiator
reactor for 8 h maintaining the vessel temperature at 80 ꢀC. The
loading level of the resin (0.843 mmol/g) was based on the mass
balance recovered acyclovir, which was obtained by subjecting a
portion of the resin to cleavage by 95:5 trifluoroacetic acid (TFA)/
water for 20 min. The resulting acyclovir on DHP-resin (20, Scheme
4) was coupled with Fmoc-Pro-OH in the presence of 4 equivalents
of HCTU as coupling reagent, DIEA as base in dry DMF at room
temperature overnight. After cleavage of a small portion of resin
(95:5 TFA/water), the desired Fmoc-Pro-acyclovir compound 16
was detected by HPLC-MS analysis (254 nm) albeit with recovering
of acyclovir in a 1:5 proportion (20% conversion). The use of other
coupling reagents such as TFFH/2,4,6-collidine or HATU/HOAt did
not improve conversion. These results showed for the first time that
the coupling of Fmoc-Pro-OH with the exocyclic amino of acyclovir
was viable in a solid-phase approach using a DHP-resin and HCTU
as coupling reagent. A variety of coupling reaction conditions were
next examined. The best results were obtained by repeating
coupling cycles under microwave heating at 80 ꢀC
(5 cycles ꢁ 10 min) using 1.2 equivalents of HCTU and DIEA in dry
DMF. Under these conditions, an improved conversion (50%) to the
prolyl-acyclovir compound 16 was detected by HPLC-MS after
cleavage of a small portion of resin.
Next, we focused on the synthesis of target tetrapeptide amide
prodrugs of acyclovir 3 from acyclovir attached to the DHP resin 20
(Scheme 4). We first investigated the stepwise introduction of
amino acids (method A). Thus, coupling of Fmoc-Val-OH onto the
prolyl-acyclovir resin 21 under optimized conditions (HCTU/DIEA/
80 ꢀC/MW) gave the expected dipeptide derivative 22. However,
Fmoc deprotection of resin 22 under standard conditions (piperi-
dine 20% in DMF) and coupling of the third amino acid failed to give
the tripeptide-acyclovir derivative 23 and only acyclovir was
detected by HPLC-MS after cleavage of a small portion of the resin.
This could be explained by DKP formation before the coupling of
the third amino acid similarly to what had occurred in the above
solution-phase approach.
We next examined the direct tetrapeptide coupling (method B)
on the acyclovir DHP-resin (20) under the optimized coupling
conditions (HCTU/DIEA/80 ꢀC/MW). Unfortunately, although the
desired tetrapeptide prodrug 25 was detected by HPLC-MS after
cleaving a small portion of the resin, the acyclovir starting material
(1) was recovered as major compound.
These results prompted us to investigate the amino acid and
tripeptide coupling strategy (method C) as an alternative for the
synthesis of the target tetrapeptide prodrug using a Boc protecting
strategy. Thus, supported prolyl derivative 21 was reacted with
tripeptide Boc-Val-Pro-Val-OH in the presence of HCTU as coupling
reagent under the optimized coupling conditions to give the
desired supported tetrapeptide-acyclovir derivative 26 (Scheme 4).
O
N
O
N
O
N
N
N
N
N
HN
Fmoc Pro Val Pro HN
(ii), (i)
HN
Fmoc Val Pro HN
(ii), (i)
HN
Fmoc Pro HN
N
N
O
O
O
O
O
O
A
Stepwise incorporation
of amino acids
22
23
21
O
N
O
(i)
B
N
N
N
N
HN
HN
Direct
tetrapeptide
coupling
Cbz Val Pro Val Pro HN
Cbz Val Pro Val Pro HN
N
(iii)
O
O
ACV
O
HO
+
25 (low conversions)
20
cleavage
(i)
24
C
O
[1 + 3] fragment coupling
(i)
N
N
HN
Boc Val Pro Val Pro HN
(ii), (i)
N
(iii)
O
ACV
21
3
+
O
deprotection/
cleavage
26
Scheme 4. Synthetic strategies in solid phase for the synthesis of target tetrapeptide acyclovir prodrug 3 from acyclovir attached to the DHP resin (20). Reagents and conditions: (i)
R-XXa-OH or peptide, HCTU, DIPEA, DMF, MW, 80 ꢀC, 5 ꢁ 10 min; (ii) 20% piperidine, DMF, rt; (iii)TFA/H2O 95:5, rt.

462
A. Diez-Torrubia et al. / European Journal of Medicinal Chemistry 70 (2013) 456e468
The final product was cleaved from the resin using the standard
acidic cleavage protocol (95:5, TFA/water) which result in the
simultaneous deprotection of the Boc group. Purification of the
cleavage crude by reverse-phase chromatography on a Biotage
Isolera provided the target prodrug 3 in a 20% overall yield together
with unreacted acyclovir. The overall yield for the four-step process
(loading, amino acid and tripeptide couplings and cleavage) was
moderate but the target prodrug could be isolated at high purity
after chromatographic purification without isolation of
intermediates.
fully stable in PBS. Even after 24 h, no traces of free ACV were
observed (Fig. 5). Exposure to CD26 resulted in rapid and efficient
conversion to Val-ACV. The appearance of ACV after 24 h to the
amount of w40% is apparently due to spontaneous (chemical)
conversion as earlier demonstrated in Fig. 4. Administration of
HS and BS to [ValPro-Val]-ACV also efficiently converted the
prodrug to Val-ACV. Within 15 min, 80% (HS) and 40% (BS) of the
ACV tripeptide derivative had been converted to Val-ACV. Thus,
in contrast with Val-ACV, [ValPro-Val]-ACV proved chemically
fully stable, but was efficiently converted by CD26 and human/
bovine sera to Val-ACV.
Finally, the conversion of the tetrapeptide amide prodrug 3 was
investigated. This ACV prodrug was virtually completely stable in
PBS. Only a few percent of the compound converted to ACV after
24 h (Fig. 6). CD26 efficiently hydrolysed the tetrapeptide to ACV.
Interestingly, no traces of dipeptide-ACV that were expected to
appear during the 2-step conversion of the tetrapeptide-ACV to
ACV could be demonstrated. This observation strongly suggests
that the ACV dipeptide derivative is highly unstable and sponta-
neously converts to ACV as soon as formed in the CD26-catalysed
hydrolysis reaction. Also, in HS and BS, solely ACV could be found
in the efficient hydrolysis reaction during the eventual conversion
of the ACV-tetrapeptide to ACV (Fig. 6).
2.2. Water solubility of prodrug conjugates
The water-solubility of the acyclovir target prodrugs 3 and 4
were determined and compared to that of the poorly soluble parent
compound. The amide prodrug 3 and the ester prodrug 4 improved
the water-solubility (22.87 mg/mL and 12.19 mg/mL) more than 17-
fold and 9-fold, respectively, in comparison with the parent com-
pound 1 (1.29 mg/mL). This result suggests that the CD26 prodrug
approach could be useful for increasing the water-solubility of polar
drugs.
2.3. Biological studies
To demonstrate the involvement of CD26 in the conversion of
[ValPro-ValPro] amide prodrug 3 and [ValPro-Val] ester prodrug 4
in HS, the enzyme reaction was investigated in the presence of
increasing concentrations of Vildagliptin, a potent inhibitor of
dipeptidyl peptidase activity (Fig. 7). The CD26 inhibitor dose-
dependently prevented the release of the ValPro moiety of both
tetrapeptide- and tripeptide-ACV prodrugs and proved particularly
effective in case of the tetrapeptide-ACV derivative (Fig. 7) that was
2.3.1. Susceptibility of the prodrugs to the action of purified CD26
and CD26 present in human and bovine serum
The stability of valyl ester prodrug 13 (Val-ACV) in PBS and sera
was investigated. It was found that during the first two hours, only
a few percent of the compound was converted to the parental ACV
derivative (Fig. 4) in both PBS and human/bovine serum (HS and BS)
(Fig. 4). However, after 24 h, almost 50% of the prodrug proved to be
converted to ACV in PBS and HS and this conversion had even
proceeded to w60% in BS (Fig. 4). CD26 had no additional influence
on this chemical conversion process as expected.
still intact by >80% after 24 h in the presence of 25 mM Vildagliptin.
Without this inhibitor, the tetrapeptide-ACV was virtually
completely converted to ACV within 1 h.
To investigate whether these prodrugs can survive the low pH
environment of the stomach in case they would be administered
Next, the tripeptide ester prodrug of ACV 4 was investigated.
In contrast with Val-ACV (13), the tripeptide ester conjugate was
Fig. 4. Stability in PBS and conversion of valacyclovir 13 to the parent compound in the presence of DPPIV/CD26 and 10% human (HS) or bovine (BS) serum.

A. Diez-Torrubia et al. / European Journal of Medicinal Chemistry 70 (2013) 456e468
463
Fig. 5. Stability in PBS and conversion of [ValPro]-[Val] ester prodrug 4 to the parent compound in the presence of DPPIV/CD26 and 10% human (HS) or bovine serum (HS).
orally, their stability was examined in sodium phosphate buffer pH
3.2. It was found that both the [ValPro-ValPro] amide ACV prodrug
3 and the [ValPro-Val] ester ACV prodrug 4 were fully stable at such
low pH for at least 6 h.
were 32 and 73 mM versus 18 mM for ACV. Thus, the peptide prodrugs
displayed marked anti-herpetic activity in cell culture within a
comparable order of magnitude as parent ACV.
3. Conclusions
2.3.2. Antiviral evaluation
The antiviral activity of ACV and its peptide amide and ester
prodrugs were also evaluated in confluent HEL cell cultures. Whereas
ACV proved substantially active against herpes simplex virus type 1
In conclusion, various synthetic studies toward peptide amide
and ester prodrugs of acyclovir cleavable by the DPPIV/CD26
enzyme are reported. Solid-phase synthesis for the challenging
acylation of the exocyclic amino group of acyclovir with amino
acids or peptides is a convenient procedure that does not require
protection of the hydroxyl group of acyclovir. In our opinion the
method could be extended to other nucleosides containing amino
functions and/or other acylating reagents. Purified CD26 as well as
(KOS) and type 2 (G) at an EC50 of 0.32 and 0.50
[ValPro-Val] and [ValPro-ValPro]-ACV prodrugs showed EC50’s of
0.60 and 1.8 M against HSV-1 and 1.6 and 4.9 M against HSV-2. The
mM, respectively, the
m
m
antiviral activities of the [ValPro-Val] and [ValPro-ValPro]-ACV pro-
drugs against a thymidine kinase-deficient HSV-1 (KOS ACVr) strain
PBS
CD26
120
120
100
80
100
80
14.3' (ValPro-ValPro-ACV)
14.3' (ValPro-ValPro-ACV)
60
60
5.0' (ACV)
5.0' (ACV)
40
20
0
40
20
0
0
0,25
1
2
24
0
0,25
1
2
24
Time (hours)
Time (hours)
HS 20%
BS 20%
120
100
80
60
40
20
0
120
100
80
60
40
20
0
14.3' (ValPro-ValPro-ACV)
5.0' (ACV)
5.0' (ACV)
14.3' (ValPro-ValPro-ACV)
0
0,25
1
2
24
0
0,25
1
2
24
Time (hours)
Time (hours)
Fig. 6. Stability in PBS and conversion of [ValPro]-[ValPro] amide prodrug 3 to the parent compound in the presence of DPPIV/CD26 and 10% human (HS) or bovine serum (BS).

464
A. Diez-Torrubia et al. / European Journal of Medicinal Chemistry 70 (2013) 456e468
Fig. 7. Hydrolysis of [ValPro]-[ValPro] amide prodrug 3 and [ValPro]-[Val] ester prodrug 4 in the presence of a specific inhibitor of DPPIV/CD26.
(CD26-containing) serum efficiently converted the tetrapeptide
amide prodrug 3 to ACV, and the tripeptide ester prodrug 4 to Val-
ACV. In the former case, no traces of intermediate ValPro-ACV were
observed in the hydrolysis process. The anchoring of the peptide
promoiety to either the amino or the hydroxyl group of ACV led to
successful amide or ester prodrugs. The stability results indicate
that the tetrapeptide amide prodrug of ACV exhibit faster release of
the parent drug compared to peptide ester prodrugs of acyclovir.
Both types of prodrugs afforded an efficient anti-herpetic activity in
cell culture.

A. Diez-Torrubia et al. / European Journal of Medicinal Chemistry 70 (2013) 456e468
465
4. Experimental section
598.3 [M þ H]þ. Anal. Calcd. for C28H35N7O8: C, 56.27; H, 5.90; N,
16.41. Found: C, 56.42; H, 6.10; N, 14.33.
4.1. Chemistry
4.1.1.2. 9-[(2-Acetoxyethoxy)methyl]-2-N-[N-(bencyloxycarbonyl)
prolyl]guanine (10). According to the coupling procedure described
for compound 6, a solution of dry 9-[(acetoxyethoxy)methyl]gua-
nine 5 [31] (0.20 g, 0.74 mmol) in dry DMF (4 mL) was successively
reacted with Cbz-Pro-OH (0.94 g, 3.74 mmol), TFFH (0.98 g,
3.74 mmol) and 2,4,6-collidine (0.50 mL, 3.74 mmol) at 80 ꢀC
overnight. The final residue was purified by CCTLC on the Chro-
matotron (CH2Cl2/MeOH, 30:1) to afford 10 (244 mg, 65%) as a
Microanalyses were obtained with a Heraeus CHN-O-RAPID
instrument. Electrospray mass spectra were measured on
a
quadropole mass spectrometer equipped with an electrospray
source (HewlettePackard, LC/MS HP 1100). Spectra were recorded
with Varian Inova-300 or Varian Inova-400 spectrometers oper-
ating at 300 or 400 MHz for 1H NMR with Me4Si as internal
standard. Analytical thin-layer chromatography (TLC) was per-
formed on silica gel 60 F254 (Merck). Separations on silica gel
were performed by flash column chromatography with silica gel
60 (230e400 mesh) (Merck) or preparative centrifugal circular
thin layer chromatography (CCTLC) on a ChromatotronR (Kieselgel
60 PF254 gipshaltig (silica gel containing gypsum) (Merck)), layer
thickness of 1 mm, flow rate of 5 mL/min. Liquid chromatography
was performed using a force flow (flash chromatography) Hori-
zon HPFG System (Biotage) with Flash 25 or 40 silica gel
cartridges.
The coupling reactions on solid phase were carried out using
microwave radiation in a Biotage Initiator reactor in a 5 mL vial.
Excluding the coupling reaction on the microwave reactor, the rest
of the SPPS reactions were stirred using an IKA-100 orbital shaker.
After cleavage, the acidic crudes were sedimented in Et2O on a
Hettlich Universal 320R centrifuge at 5000 rpm. All the crude and
samples were lyophilized using mixtures water/acetonitrile on a
Telstar 8-80 instrument. The monitoring of the reactions was per-
formed by HPLC/MS through a HPLC-waters 12695 connected to a
Waters Micromass ZQ spectrometer.
white foam. 1H NMR (400 MHz, DMSO-d6):
OCOCH3, -CH2, -CH2, Pro), 3.42e3.66 (m, 4H,
4.04 (m, 2H, CH2OCO), 4.46 (m, 1H, -CH, Pro), 4.87e5.08 (m, 2H,
d
1.58e2.21 (m, 7H,
-CH2, Pro, OCH2),
b
g
d
a
CH2, CBz), 5.44 (s, 2H, NCH2O), 7.05e7.34 (m, 5H, Ar, Cbz), 8.13 (s,
1H, H-8), 11.87 (bs, 1H, NH-1), 11.93 (bs, 1H, 2-NH). MS (ESIþ) 499.2
[M þ H]þ. Anal. Calcd. for C23H26N6O7: C, 55.42; H, 5.26; N, 16.86.
Found: C, 55.71; H, 5.21; N, 16.68.
4.1.1.3. 9-[(2-Acetoxyethoxy)methyl]-2-N-[N-(bencyloxycarbonyl)-
valyl-prolyl-valyl-prolyl]-guanine (9). Method A. A solution of N-
protected prolyl derivative 10 (0.065 g, 0.144 mmol) in methanol
(5 mL) containing Pd/C (10%) (40% wt/wt) was hydrogenated at
25 psi at room temperature for 2 h. The reaction mixture was
filtered and the filtrate was evaporated to dryness under reduced
pressure. The residue (N-deprotected prolyl intermediate) was
dissolved in dry DMF (2 mL) and was successively treated with
PyBOP (0.112 g, 0.216 mmol) and triethylamine (50
mL,
0.36 mmol). The reaction mixture was stirred at room tempera-
ture for 2 h and an extra amount of PyBOP (0.112 g, 0.216 mmol)
Chemicals and reagents were obtained from commercial sour-
ces and were used without further purification. 2-chlorotrityl resin
(PS) and HCTU were purchased from GL Biochem (Shanghai) Ltd.
Aminomethyl resin (PS) and dihydropyrane resin were purchased
from Novabiochem and aminomethyl resin (PEG) and 2-chlorotrityl
resin (PEG) from Chematrix. Acyclovir was purchased from Ark
Pharm Inc. (USA) and valacyclovir from Sigma. The dipeptide de-
rivative Cbz-Val-Pro-OH and the tripeptide Boc-Val-Pro-Val-OH
were purchased from Bachem Feinchemikalien and Peptide Pro-
tein Research Ltd, respectively. The 50-O-acetyl derivative of
acyclovir (5) [31] and the tetrapeptide Cbz-Val-Pro-Val-Pro-OH and
the tripeptide Cbz-Val-Pro-Val-OH were synthesized as previously
described [12].
and triethylamine (50 mL, 0.36 mmol) were added. The reaction
mixture was stirred at room temperature overnight and the sol-
vent was evaporated to dryness. The residue was dissolved in
ethyl acetate 26 (50 mL), washed with 10% aqueous citric acid
(3 ꢁ 20 mL), 10% aqueous NaHCO3 (3 ꢁ 20 mL), water (3 ꢁ 20 mL)
and brine (3 ꢁ 20 mL). The organic layer was dried (Na2SO4),
filtered and evaporated to dryness. The final residue was purified
by CCTLC on the Chromatotron (CH2Cl2/MeOH, 40:1) to afford 9
(27 mg, 24% overall yield from 10) as a white foam. 1H NMR
(400 MHz, DMSO-d6):
(m, 13H, OCOCH3, -CH, Val,
-CH2, Pro, OCH2), 3.99e4.20 (m, 4H,
d
0.83e0.92 (d, 12H,
-CH2, -CH2, Pro), 3.44e3.73 (m, 5H,
-CH, Val, CH2OCO), 4.42e
a-CH, Pro), 4.92e5.08 (m, 2H, CH2, Cbz), 5.48 (s, 2H,
g-CH3, Val), 1.60e1.99
b
b
g
d
a
4.52 (m, 2H,
NCH2O), 7.31e7.47 (m, 6H, Ar, Cbz, NH), 7.88 (d, 1H, NH, Val,
J ¼ 9.1 Hz), 8.10 (d, 1H, NH, Val, J ¼ 8.3 Hz), 8.15 (s, 1H, H-8), 11.89
(bs, 1H, NH-1), 11.99 (bs, 1H, 2-NH). MS (ESIþ) 794.5 [M þ H]þ.
Anal. Calcd. for C38H51N9O10: C, 57.49; H, 6.48; N, 15.88. Found: C,
57.70; H, 6.21; N, 15.63.
Method B. A solution of N-protected prolyl derivative 10 (0.065 g,
0.144 mmol) in methanol (5 mL) containing Pd/C (10%) (40% wt/wt)
was hydrogenated at 25 psi at room temperature for 2 h. The re-
action mixture was filtered and the filtrate was evaporated to
dryness under reduced pressure. The residue (N-deprotected prolyl
intermediate) was dissolved in dry DMF (2 mL) and was succes-
sively reacted with Cbz-Val-Pro-Val-OH (0.098 g, 0.22 mmol), HATU
4.1.1. Solution-phase synthesis of acyclovir prodrugs
4.1.1.1. 9-[(2-Acetoxyethoxy)methyl]-2-N-[N-(bencyloxycarbonyl)-
valyl-prolyl]guanine (6). To a solution of dry 9-[(acetoxyethoxy)
methyl]guanine 5 [31] (0.10 g, 0.37 mmol) in dry DMF (2 mL) was
successively added Cbz-Val-Pro-OH (0.47 g, 1.87 mmol), TFFH
(0.49 g, 1.87 mmol) and 2,4,6-collidine (0.25 mL, 1.87 mmol). The
reaction mixture was stirred at 80 ꢀC overnight and the solvent was
evaporated to dryness. The residue was dissolved in ethyl acetate
(100 mL) and washed with 10% aqueous citric acid (3 ꢁ 20 mL), 10%
NaHCO3 (3 ꢁ 20 mL), H2O (3 ꢁ 20 mL) and brine (3 ꢁ 20 mL). The
organic layer was dried (Na2SO4), filtered and evaporated to dry-
ness. The crude was purified by flash column chromatography and
CCTLC on the Chromatotron (CH2Cl2/MeOH, 30:1) to afford 6
(125 mg, 53%) as a white foam. 1H NMR (400 MHz, acetone-d6):
(0.084 g, 0.22 mmol), HOAt (0.03 g, 0.22 mmol) and DIEA (30.1 mL,
0.22 mmol). The reaction mixture was stirred at room temperature
for 2 h and an extra amount of HATU (0.084 g, 0.22 mmol), HOAt
d
0.90 (d, 3H,
J ¼ 6.4 Hz),1.91e1.95 (m, 7H, OCOCH3,
2.20 (m, 1H, -CH2, Pro), 3.49e3.53 (m, 2H,
OCH2), 3.83 (m, 1H, -CH, Val), 4.07 (m, 2H, CH2OCO), 4.55 (m, 1H,
g
-CH3, Val, J ¼ 6.4 Hz), 0.95 (d, 3H,
-CH, Val, -CH2,
-CH2, Pro), 3.68 (m, 2H,
g
-CH3, Val,
(0.03 g, 0.22 mmol) and DIEA (30.1 mL, 0.22 mmol) were added. The
b
b
g-CH2, Pro),
reaction mixture was stirred at room temperature overnight and
the solvent was evaporated to dryness. The residue was dissolved in
ethyl acetate (50 mL), washed with 10% aqueous citric acid
(3 ꢁ 20 mL), 10% aqueous NaHCO3 (3 ꢁ 20 mL), water (3 ꢁ 20 mL)
and brine (3 ꢁ 20 mL). The organic layer was dried (Na2SO4),
filtered and evaporated to dryness. The final residue was purified by
b
d
a
a
-CH, Pro), 5.02 (dd, 2H, CH2, Z, J ¼ 12.4 Hz, J ¼ 14.9 Hz), 5.49 (bs,
2H, NCH2O), 7.35 (m, 5H, Ar, Z), 7.53 (d, 1H, NH, Val, J ¼ 7.9 Hz), 8.15
(s, 1H, H-8), 11.89 (bs, 1H, NH-1), 11.99 (bs, 1H, 2-NH). MS (ESIþ)

466
A. Diez-Torrubia et al. / European Journal of Medicinal Chemistry 70 (2013) 456e468
CCTLC on the Chromatotron (CH2Cl2/MeOH, 40:1) to afford 9
(10 mg, 9% overall yield from 10) as a white foam.
10.61 (s, 1H, NH-1). MS (ESIþ) 655.3 [M þ H]þ. Anal. Calcd. for
C
31H42N8O8: C, 56.87; H, 6.47; N, 17.11. Found: C, 56.96; H, 6.61; N,
17.23.
4.1.1.4. 9-[(2-Hydroxyethoxy)methyl]-2-N-[N-(bencyloxycarbonyl)-
valyl-prolyl-valyl-prolyl]-guanine (11). A solution of protected pro-
drug 9 (0.100 g, 0.126 mmol) was reacted with methanol saturated
with ammonia (10 mL) at room temperature overnight and the
solvent was evaporated to dryness. The residue was purified by
CCTLC (CH2Cl2/MeOH, 20:1). The fastest moving fractions gave 11
(30 mg, 32%) as a white foam. 1H NMR (400 MHz, DMSO-d6):
4.1.1.7. 9-[2-(Valyl-prolyl-valyloxyethoxy)methyl]guanine (4). A so-
lution of 14 (0.08 g, 0.12 mmol) in MeOH:H2O:THF (5:2:2 mL)
containing Pd/C (10%) (0.04 g) was hydrogenated at 1 atm at room
temperature for 3 h. The reaction mixture was filtered and the
filtrate was washed with MeOH (20 mL) and was evaporated to
dryness under reduced pressure to give 4 (40 mg, 67%) as a white
d
0.80e0.93 (d, 12H,
CH2, -CH2, Pro), 3.24e3.72 (m, 8H,
4.00e4.22 (m, 2H, -CH, Val), 4.49, 4.68 (2m, 2H,
g
-CH3, Val), 1.60e2.26 (m, 10H,
-CH2, Pro, OCH2, CH2OCO),
-CH, Pro), 4.93e
b
-CH, Val,
b
-
foam. 1H NMR (300 MHz, DMSO-d6):
Val), 1.73e2.04 (m, 6H, -CH, Val, -CH2,
3.68 (m, 4H, OCH2, -CH2, Pro), 4.09e4.54 (m, 5H, CH2OCO, 2
Val, -CH, Pro), 5.34 (bs, 2H, NCH2O), 6.57 (bs, 2H, 2-NH2), 7.80 (s,
d
0.82e0.97 (d, 12H,
-CH2, Pro), 3.19e3.37e
-CH,
g-CH3,
g
d
b
b
g
a
a
d
a
5.17 (m, 2H, CH2, Cbz), 5.48 (s, 2H, NCH2O), 7.09e7.46 (m, 6H, Ar,
Cbz, NH), 7.39 (d, 1H, NH, Val, J ¼ 8.3 Hz), 7.90 (d, 1H, NH, Val,
J ¼ 9.0 Hz), 8.14 (s, 1H, H-8), 11.92 (bs, 2H, NH-1, 2-NH). MS (ESIþ)
752.5 [M þ H]þ. Anal. Calcd. for C36H49N9O10: C, 57.79; H, 6.50; N,
16.89. Found: C, 57.51; H, 6.57; N, 16.77.
a
1H, H-8), 8.07 (d, 1H, NH, Val, J ¼ 8.4 Hz), 8.44 (d, 1H, NH, Val,
J ¼ 8.4 Hz), 10.61 (s, 1H, NH-1). MS (ESIþ) 521.3 [M þ H]þ. Anal.
Calcd. for C23H36N8O6: C, 53.06; H, 6.97; N, 21.52. Found: C, 52.89;
H, 7.13; N, 21.64.
From the slowest moving fractions of 2-N-[N-(bencylox-
ycarbonyl)-valyl-prolyl-valyl-prolyl]guanine (30 mg, 26%) were
isolated due to partial rupture of the CeN bond of the acyclovir
4.1.2. Solid-phase synthesis of acyclovir prodrugs
4.1.2.1. DHP resin attachment of acyclovir (20). A suspension of
acyclovir (2.5 g, 11.25 mmol) in dry DMF (3 mL) was stirred at 80 ꢀC
to achieve maximum dissolution. Then, this suspension was
transferred into a 5-ml glass tube containing DHP resin (2.5 g,
2.25 mmol) previously swollen in DCM/DMF/DCM/DMF
(4 ꢁ 0.5 min) and pyridinium p-toluenesulfonate (PPTS) (1.13 g,
4.50 mmol) was added. The mixture was irradiated in a Biotage
Initiator MW reactor for 8 h at 80 ꢀC. Then, the reaction mixture was
transferred to a polypropylene cartridge and the resin was washed
with DMSO (3ꢁ), DMF (3ꢁ) and DCM (3ꢁ) and dried under reduced
pressure for 24 h. The loading level of acyclovir onto the resin was
determined according to the following procedure. Resin 20
(216 mg) was treated with 95:5 TFA/H2O (2 mL) for 20 min and
washed thoroughly with DMF/DCM (3ꢁ) and the combined filtrates
concentrated under reduced pressure. In this way, 41 mg of
acyclovir were recovered which corresponds to a resin loading of
0.843 mmol/g.
prodrug. 1H NMR (400 MHz, DMSO-d6):
d
0.80e0.95 (d, 12H,
-CH2, -CH2, Pro), 3.20e3.78
-CH, Val), 4.51, 4.75 (2m, 2H,
-CH, Pro), 4.90e5.19 (m, 2H, CH2, Cbz), 7.00e7.48 (m, 6H, Ar, Cbz,
g-CH3,
Val), 1.62e2.29 (m, 10H,
b
-CH, Val,
b
g
(m, 4H, -CH2, Pro), 4.12e4.26 (m, 2H,
d
a
a
NH), 7.45 (d, 1H, NH, Val, J ¼ 8.3 Hz), 7.95 (d, 1H, NH, Val, J ¼ 9.0 Hz),
8.19 (s, 1H, H-8), 12.00 (bs, 2H, NH-1, 2-NH), 13.00 (bs, 1H, NH7/9).
MS (ESIþ) 679.5 [M þ H]þ. Anal. Calcd. for C33H43N9O7: C, 58.48; H,
6.40; N, 18.60. Found: C, 58.61; H, 6.57; N, 18.87.
4.1.1.5. 9-[(2-Hydroxyethoxy)methyl]-2-N-[valyl-prolyl-valyl-prolyl]
guanine (3). A solution of the N-protected tetrapeptide prodrug
derivative 11 (0.025 g, 0.033 mmol) in methanol (1 mL) containing
Pd/C (10%) (40% wt/wt) was hydrogenated at 25 psi at room tem-
perature for 2 h. The reaction mixture was filtered and the filtrate
was evaporated to dryness under reduced pressure. The residue
was dissolved in water and lyophilized to give the final deprotected
conjugate 3 (17 mg, 83%) as a white foam. 1H NMR (300 MHz,
DMSO-d6):
10H, 2 -CH, Val1 y Val2, 2
3.31e3.72 (m, 6H, 2 -CH2, Pro1 y Pro2, OCH2), 3.93e4.26 (m, 4H,
CH, Pro, -CH, Val2, CH2OH), 4.36e4.52 (m, 2H, 2 -CH, Val1, -CH,
Pro), 5.38 (s, 2H, NCH2O), 6.67 (bs, 1H, NH-1), 7.99e8.05 (m, 4H, H-
8, NHþ3 ), 8.52 (m, 1H, NH, Val2), 10.86 (bs, 1H, NH-2). EM (ESIþ):
619.2 [M þ H]þ. Anal. Calcd. for C28H43N9O7: C, 54.44; H, 7.02; N,
20.41. Found: C, 54.77; H, 6.89; N, 20.24.
d
0.78e1.01 (m, 12H, 2
g
-CH3, Val1 y Val2), 1.70e2.15 (m,
4.1.2.2. Support bound 9-[(2-hydroxyethoxy)methyl]-2-N-[Na-(9-
fluorenylmethoxycarbonyl)-prolyl]guanine (21). To a solid phase
syringe, acyclovir on DHP-resin (20) (0.071 g, 0.056 mmol) was
added. The resin was swollen with DCM, DMF, DCM and DMF
(3 ꢁ 30 s per solvent). Then, Fmoc-Pro-OH (23 mg, 0.067 mmol),
HCTU (28 mg, 0.067 mmol) and DIPEA (0.023 mL, 0.134 mmol) were
added to the resin. The mixture was transferred to a microwave vial
and irradiated in a Biotage Initiator MW reactor for 10 min at 40 ꢀC.
Then, the vial was opened, the supernatant removed and new
coupling mixture added. This process was repeated 5 times in total
(5 ꢁ 10 min). Finally, the Fmoc-Pro-acyclovir on DHP-resin 21 was
transferred to a fritted syringe, drained and washed extensively
(DMF/DCM/DMF/DCM, 5 ꢁ 0.5 min). To follow the progress of the
reaction, a cleavage of a small portion of the resin 21 with TFA/H2O
(95:5 (v/v), 1 mL) for 20 min was carried out. The resin was thor-
oughly washed with DMF and DCM and the filtrates were evapo-
rated to dryness. The product was lyophilized from CH3CN/H2O and
analysed by HPLC-MS. In the chromatogram a 1.5:1 mixture of 16
(tR ¼ 13.03 min, 545.15 [M þ H]þ, 1089.37 [2M þ H]þ) and acyclovir
(tR ¼ 1.32 min, 226.14 [M þ H]þ) was detected.
b
b-CH2, Pro1 y Pro2, 2
g-CH2, Pro1 y Pro2),
d
a-
a
a
a
4.1.1.6. 9-[2-((N-Bencyloxycarbonyl)valyl-prolyl-valyloxyethoxy)
methyl]guanine (14). To a solution of VACV (13) (0.10 g, 0.27 mmol)
in dry DMF (2 mL), was successively added pyBOP (0.198 g,
0.38 mmol), Cbz-Val-Pro-OH (0.13 g, 0.38 mmol) and TEA (90.3 mL,
0.65 mmol). The reaction mixture was stirred at room temperature
for 24 h and the solvent was evaporated to dryness. The residue was
dissolved in ethyl acetate (50 mL) and washed with 10% aqueous
citric acid (3 ꢁ 20 mL), 10% NaHCO3 (3 ꢁ 20 mL), H2O (3 ꢁ 20 mL)
and brine (3 ꢁ 20 mL). The organic layer was dried (Na2SO4),
filtered and evaporated to dryness. The final residue was purified by
flash column chromatography (ethyl acetate/MeOH, 4:1) to afford
14 (150 mg, 81%) as a white foam. 1H NMR (300 MHz, DMSO-d6):
d
0.86 (d, 12H,
CH2, Pro), 3.39e3.70 (m, 4H, OCH2,
Val), 4.10 (m, 1H, -CH, Val), 4.20 (m, 2H, CH2OCO), 4.44 (m, 1H, a-
g
-CH3, Val), 1.73e2.01 (m, 6H,
b
-CH, Val,
b
-CH2,
g
-
4.1.2.3. 9-[(2-Hydroxyethoxy)methyl]-2-N-[valyl-prolyl-valyl-prolyl]
guanine (3). To a solid phase syringe, Fmoc-Pro-acyclovir on DHP-
resin 21 (0.21 mmol) was added. The resin was swollen with DCM,
DMF, DCM and DMF (3 ꢁ 30 s per solvent). Then, Boc-Val-Pro-Val-
OH (0.104 g, 0.252 mmol), HCTU (0.104 g, 0.252 mmol) and DIPEA
(0.086 mL, 0.504 mmol) were added to the resin. The mixture was
d-CH2, Pro), 4.02 (m, 1H,
a
-CH,
a
CH, Pro), 5.01 (dd, 2H, CH2, Cbz, J ¼ 12.4 Hz, J ¼ 14.9 Hz), 5.34 (bs,
2H, NCH2O), 6.49 (bs, 2H, 2-NH2), 7.34 (m, 5H, Ar, Cbz), 7.42 (d, 1H,
NH, Val, J ¼ 8.3 Hz), 7.66 (s,1H, H-8), 8.09 (d,1H, NH, Val, J ¼ 8.3 Hz),

A. Diez-Torrubia et al. / European Journal of Medicinal Chemistry 70 (2013) 456e468
467
transferred to a microwave vial and irradiated in a Biotage Initiator
MW reactor for 10 min at 40ꢀ C. Then, the vial was opened, the
supernatant was removed and new coupling mixture was added.
This process was repeated 5 times in total (5 ꢁ 10 min). Finally, the
compound was cleaved from the resin 26 with TFA/H2O (95:5 (v/v),
3 mL) for 20 min. The resin was thoroughly washed with DMF and
DCM and the filtrates were evaporated to dryness. The product was
lyophilized from CH3CN/H2O and purified by reverse-phase chro-
matography on a Biotage Isolera instrument using a reverse phase
column (KP-C18-HS 12 g) to give target prodrug 3 (25 mg, 20%
overall yield). As mobile phase, mixtures of A:B were used, where
A ¼ 0.05% TFA water and B ¼ acetonitrile with a flow rate of 7 mL/
min. The prodrug 3 was purified using a gradient from 0% of B to
100% of B in 30e45 min and was detected at 254 nm. All the
characterization data were identical to those of prodrug 3 obtained
from the solution-phase strategy.
gradient to 65% A þ 35% B; 8 min linear gradient to 60% A þ 40% B;
7 min linear gradient to 50% A þ 50% B; 5 min 90% A þ 10% B; 5 min
equilibration at 90% A þ 10% B.
Test prodrugs (25 mM) were exposed to CD26, 20% HS or 20% BS
(as described above) in the presence or absence of different con-
centrations of the DPPIV/CD26 inhibitor Vildagliptin (ranging be-
tween 50 and 0.05 mM). Conversion of the prodrugs to the parent
compounds was evaluated after 15, 30, 45 and 60 min, and analysed
as described above.
4.3.2. Anti-HSV-1 and -HSV-2 activity assay
The ACV prodrugs were evaluated against herpes simplex virus
type 1 (HSV-1) strain KOS, and herpes simplex virus type 2 (HSV-2)
strain G. The antiviral assay was based on inhibition of virus-
induced cytopathicity in human embryonic lung (HEL) fibroblasts.
Confluent cell cultures in microtiter 96-well plates were inoculated
with 100 CCID50 of virus (1 CCID50 being the virus dose to infect 50%
of the cell cultures) in the presence of varying concentrations of the
test compounds. Viral cytopathicity was recorded as soon as it
reached completion in the control virus-infected cell cultures that
were not treated with the test compounds. Antiviral activity was
expressed as the EC50 or compound concentration required to
reduce virus-induced cytopathicity by 50%.
4.2. Water-solubility studies
Water-solubility of the prodrugs 3 and 4 and the parent com-
pound was determined by HPLC analysis. The HPLC runs were
carried out on a Waters 484 System using Novapack C18 reverse
phase column. Flow rate: 1 mL/min. Detection: UV 254 nm.
Gradient solvent system A/B (acetonitrile/water): initial 15%
A þ 85% B; 5 min linear gradient to 25% A þ 75% B; 5 min linear
gradient to 35% A þ 65% B; 10 min linear gradient to 45% A þ 55% B;
5 min linear gradient to 60% A þ 40% B and 5 min linear gradient to
100% A. Excess amount of the prodrug or the parent drug was
suspended in deionized water (pH ¼ 5.5), sonicated for 10 min at
room temperature, and then equilibrated overnight at room tem-
perature. The samples were centrifuged at 14,000 rpm in an
eppendorf microcentrifuge for 1.5 min at room temperature. An
aliquot of the clear supernatant was removed and diluted to a
concentration within the range of a five-point standard curve.
Water-solubility was calculated from each peak area of the solution
by HPLC analysis compared with the sample dissolved in dime-
thylsulfoxide as the standard, the exact concentration of which is
known.
Acknowledgements
We thank Ria Van Berwaer and Leentje Persoons for excellent
technical assistance. We also thank the Spanish MEC/MICINN
(Project SAF2009-13914-C02 and SAF2012-39760-C02), the
Comunidad de Madrid (BIPEDD-2-CM ref S-2010/BMD-2457), and
the KU Leuven (GOA No. 10/014) for financial support. A Juan de la
Cierva contract to S.d.C. (JDC-MICINN) from the Spanish Ministry of
Science and Innovation is also gratefully acknowledged. We very
much appreciated the helpful discussions with Dr. Judit Tulla-
Puche and Prof. Fernando Albericio (IRB Barcelona, Spain).
Appendix A. Supplementary data
Supplementary data related to this article can be found at http://
dx.doi.org/10.1016/j.ejmech.2013.10.001.
4.3. Biological methods. Compounds and enzymes
Human soluble CD26 was purified as described [27] or obtained
from Sigma (St. Louis, MO). Fetal bovine serum (FBS) was obtained
from Integro (Dieren, The Netherlands) and human serum was
provided by the Blood Bank, Leuven, Belgium. Vildagliptin was
custom synthesized by GLSynthesis Inc. (Worcester, MA, USA).
References
[1] D.A. Fox, R.E. Hussey, K.A. Fitzgerald, O. Acuto, C. Poole, L. Palley, J.F. Daley,
S.F. Schossman, E.L. Reinhert, Ta1, a novel 105 kDa human T cell activation
antigen defined by a monoclonal antibody, J. Immunol. 133 (1984) 1250e
1256.
[2] I. De Meester, G. Vanhoof, A.-M. Lambeir, S. Scharpe, CD26, let it cut or cut it
down, Immunol. Today 20 (1999) 367e375.
[3] A.-M. Lambeir, C. Durinx, S. Scharpe, I. De Meester, Dipeptidyl-peptidase IV
from bench to bedside: an update on structural properties, functions and
clinical aspects of the enzyme DPPIV, Crit. Rev. Clin. Lab. Sci. 40 (2003) 209e
294.
4.3.1. Conversion of the prodrugs of ACV to the corresponding
parent compound
The test compounds have been evaluated for their substrate
activity against purified CD26, human serum (HS), and bovine
serum (BS) in Eppendorf tubes. The 400
mL reaction mixtures
[4] A. Yaron, F. Naider, Proline-dependent structural and biological properties of
peptides and proteins, Crit. Rev. Biochem. Mol. 28 (1993) 31e81.
[5] R. Mentlein, Dipeptidyl-peptidase IV (CD26)-role in the inactivation of regu-
latory peptides, Regul. Pept. 85 (1999) 9e24.
[6] R. Mentlein, Cell-surface peptidases, Int. Rev. Cytol. 235 (2004) 165e202.
[7] J. Balzarini, E. Andersson, D. Schols, P. Proost, J. Van Damme, B. Svennerholm,
P. Horal, A. Vahlne, Obligatory involvement of CD26/dipeptidyl peptidase IV in
the activation of the antiretroviral tripeptide glycylprolylglycinamide (GPG-
NH2), Int. J. Biochem. Cell Biol. 36 (2004) 1848e1859.
contained 50 M test compound (amide and ester prodrugs of ACV
m
3 and 4) in PBS (containing 0.1% DMSO). The reaction was started by
the addition of purified CD26 (1.5 mU) or 20% (final concentration)
of HS (in PBS) or 20% (final concentration) of BS (in PBS) at 37 ꢀC. At
different time points (as indicated in the figures) 100
withdrawn from the reaction mixture, added to 200 L of cold
methanol, and put on ice for 10 min. Then the mixtures were
centrifuged at 13,000 rpm for 5 min at 4 ꢀC and 250
L supernatant
was analysed by HPLC on a reverse phase RP-8 column (excitation
wavelength: 340 nm; envision wavelength: 415 nm), using
following buffers and gradients. Buffer A: 50 mM NaH2PO4 þ 5 mM
heptanesulfonic acid, pH 3.2. Buffer B: acetonitrile. Gradient: 90%
A þ 10% B; 12 min linear gradient to 75% A þ 25% B; 13 min linear
mL was
m
[8] C. García-Aparicio, M.-C. Bonache, I. De Meester, A. San-Félix, J. Balzarini, M.-
J. Camarasa, S. Velázquez, Design and discovery of
a novel dipeptidyl-
m
peptidase IV (CD26)-based prodrug approach, J. Med. Chem. 49 (2006)
5339e5351.
[9] G. Vanhoof, F. Goossens, I. De Meester, D. Hendriks, S. Scharpé, Proline motifs
in peptides and their biological processing, FASEB J. 9 (1995) 736e744.
[10] (a) J. Balzarini, M.J. Pérez-Pérez, A. San-Félix, D. Schols, C.F. Perno,
A.M. Vandamme, M.J. Camarasa, E. De Clercq, 20,50-Bis-O-(tert-butyldime-
thylsilyl)-30-spiro-500-(400-amino-100,200-oxathiole-200,200-dioxide)pyrimidine

468
A. Diez-Torrubia et al. / European Journal of Medicinal Chemistry 70 (2013) 456e468
(TSAO) nucleoside analogues: highly selective inhibitors of human immuno-
[23] K.J. Vigil, R.F. Chemaly, Valacyclovir: approved and off-label uses for the
treatment of herpes virus infections in immunocompetent and immuno-
compromised adults, Expert Opin. Pharmacother. 11 (2010) 1901e1913 and
references therein.
[24] I. Kim, X. Chu, S. Kim, C.J. Provoda, K. Lee, G.L. Amidon, Identification of a
human valacyclovirase: biphenyl hydrolase-like protein as valacyclovir hy-
drolase, J. Biol. Chem. 278 (2003) 25348e25356.
deficiency virus type 1 that are targeted at the viral reverse transcriptase,
Proc. Natl. Acad. Sci. U. S. A. 89 (1992) 4392e4396;
(b) M.J. Camarasa, M.J. Pérez-Pérez, A. San-Félix, J. Balzarini, E. De Clercq, 30-
Spironucleosides (TSAO derivatives), a new class of specific human immu-
nodeficiency virus type 1 inhibitors: synthesis and antiviral activity of 3’-
spiro-500-[400-amino-100,200-oxathiole-200,200-dioxide]pyrimidine
nucleosides,
J. Med. Chem. 35 (1992) 2721e2727.
[25] S. Weller, M.R. Blum, M. Doucette, T. Burnette, D.M. Cederberg, P. de Miranda,
M.L. Smiley, Pharmacokinetics of the acyclovir pro-drug valaciclovir after
escalating single-and multiple-dose administration to normal volunteers, Clin.
Pharmacol. Ther. 54 (1993) 595e605.
[26] (a) Y.E. Nashed, A.K. Mitra, Synthesis and characterization of novel
dipeptide ester prodrugs of acyclovir, Spectrochim. Acta Part A 59 (2003)
2033e2039;
[11] M.J. Camarasa, A. San-Félix, S. Velázquez, M.J. Pérez-Pérez, F. Gago, J. Balzarini,
TSAO compounds: the comprehensive story of a unique family of HIV-1 specific
inhibitors of reverse transcriptase, Curr. Top. Med. Chem. 4 (2004) 945e963.
[12] A. Diez-Torrubia, C. García-Aparicio, S. Cabrera, I. De Meester, J. Balzarini,
M.J. Camarasa, S. Velázquez, Application of the dipeptidyl-peptidase IV
(DPPIV/CD26)-based prodrug approach to different amine-containing drugs,
J. Med. Chem. 53 (2010) 559e572.
[13] C. García-Aparicio, A. Diez-Torrubia, J. Balzarini, A.-M. Lambeir, S. Velázquez,
M.J. Camarasa, Efficient conversion of tetrapeptide-based TSAO prodrugs to
the parent drug by dipeptidyl-peptidase IV (DPPIV/CD26), Antivir. Res. 76
(2007) 130e139.
(b) R.S. Talluri, S.K. Samanta, R. Gaudana, A.K. Mitra, Synthesis, meta-
bolism and cellular permeability of enzymatically stable dipeptide pro-
drugs of acyclovir, Int. J. Pharm. 361 (2008) 118e124.
[27] C.R. Santos, R. Capela, C.S.G.P. Pereira, E. Valente, L. Gouveia, C. Pannecouque,
E. De Clercq, R. Moreira, P. Gomes, Structureeactivity relationships for
dipeptide prodrugs of acyclovir: Implications for prodrug design, Eur. J. Med.
Chem. 44 (2009) 2339e2346 and references therein.
[28] G. Hristov, I. Stankova, Chemical stability of new acyclovir analogues with
peptidomimetics, Sci. Pharm. 79 (2011) 259e264.
[14] A. Diez-Torrubia, J. Balzarini, G. Andrei, R. Snoeck, I. De Meester,
M.J. Camarasa, S. Velázquez, Dipeptidyl-peptidase IV-dependent water-solu-
ble prodrugs of highly lipophilic bicyclic nucleoside analogues, J. Med. Chem.
54 (2011) 1927e1942.
[15] A. Diez-Torrubia, S. Cabrera, A.-M. Lambeir, J. Balzarini, M.J. Camarasa,
S. Velázquez, Dipeptidyl peptidase IV (DPPIV/CD26)-based prodrugs of
hydroxy-containing drugs, ChemMedChem 7 (2012) 618e628.
[16] A. Diez-Torrubia, S. Cabrera, I. De Meester, M.J. Camarasa, J. Balzarini,
S. Velázquez, Dipeptidyl-peptidase IV-activated prodrugs of anti-varicella
zoster virus bicyclic nucleoside analogues containing different self-cleavage
spacers systems, ChemMedChem 7 (2012) 1612e1622.
[17] H.J. Schaeffer, L. Beauchamp, P. De Miranda, G.B. Elion, D.J. Bauer, P. Collins, 9-
(2-Hydroxyethoxymethyl)guanine activity against viruses of the herpes
group, Nature 272 (1978) 583e585.
[18] P. de Miranda, M.R. Blum, Pharmacokinetics of acyclovir after intravenous and
oral administration, J. Antimicrob. Chemother. 12 (Suppl. B) (1983) 29e37.
[19] E. De Clercq, H.J. Field, Antiviral prodrugs-the development of successful pro-
drug strategies for antiviral chemotherapy, Br. J. Pharmacol. 147 (2006) 1e11.
[20] H. Gao, A.K. Mitra, Synthesis of acyclovir, ganciclovir and their prodrugs: a
review, Synthesis-Stuttgart 3 (2000) 329e351.
[21] R. Karaman, K.K. Dajani, A. Qtait, M. Khamis, Prodrugs of acyclovir-a
computational approach, Chem. Biol. Drug Des. 79 (2012) 819e834 and ref-
erences therein.
[29] H. Gao, A.K. Mitra, Regioselective synthesis of acyclovir and its various pro-
drugs, Synth. Commun. 31 (2001) 1399e1419.
[30] X. Li, Q. Wu, D. Lv, X. Lin, Controllable synthesis of polymerizable ester and
amide prodrugs of acyclovir by enzyme in organic solvent, Bioorg. Med. Chem.
14 (2006) 3377e3382.
[31] P. Herdewijn, R. Charubala, E. De Clercq, W. Pfleiderer, Nucleotides. Part.
XXXII. Synthesis of 20e50 Connected oligonucleotides. Prodrugs for antiviral
and antitumoral nucleosides, Helv. Chim. Acta 72 (1989) 1739e1748.
[32] L.A. Carpino, A. El-Faham, Tetramethylfluoroformamidinium hexa-
fluorophosphate: a rapid-Acting peptide coupling reagent for solution and
solid phase, J. Am. Chem. Soc. 117 (1995) 5401e5402.
[33] (a) E. Giralt, R. Eritja, J. Sosa, C. Kuklinski, E. Pedroso, The use of Nbb resin in
cyclic dipeptide (“diketopiperazine”) synthesis, Synthesis (1985) 181e184;
(b) E. Pedroso, A. Grandas, X. de las Heras, R. Eritja, E. Giralt, Diketopiperazine
formation in solid phase peptide synthesis using p-alkoxybenzyl ester resins
and Fmoc-amino acids, Tetrahedron Lett. 27 (1986) 743e746.
[34] F. García-Martín, M. Quintanar-Audelo, Y. García-Ramos, L.J. Cruz, C. Gravel,
R. Furic, S. Côté, J. Tulla-Puche, F. Albericio, ChemMatrix, a poly(ethylene
glycol)-based support for the solid-phase synthesis of Complex peptides,
J. Comb. Chem. 8 (2006) 213e220.
[22] (a) L. Colla, E. De Clercq, R. Busson, H. Vanderhaeghe, Synthesis and antiviral
activity of water-soluble esters of acyclovir [9-[(2-hydroxyethoxy)methyl]
guanine], J. Med. Chem. 26 (1983) 602e604;
[35] H. Rosemeyer, F. Seela, Polymer-linked acycloguanosine, Makromol. Chem.
185 (1984) 687e695.
(b) L.M. Beauchamp, T.A. Krenitsky, Acyclovir prodrugs: the road to valacy-
clovir, Drugs Future 18 (1993) 619e628.
[36] L.A. Thompson, J.A. Ellman, Straightforward and general method for coupling
alcohols to solid supports, Tetrahedron Lett. 35 (1994) 9333e9336.