Ureido-based peptidomimetic inhibitors of herpes simplex virus ribonucleotide reductase: An investigation of inhibitor bioactive conformation

Article abstract of DOI:10.1021/jm950825x

We have been investigating peptidomimetic inhibitors of herpes simplex virus (HSV) ribonucleotide reductase (RR). These inhibitors bind to the HSV RR large subunit and consequently prevent subunit association and subsequent enzymatic activity. This report

Full text of DOI:10.1021/jm950825x

2178
J . Med. Chem. 1996, 39, 2178-2187
Ur eid o-Ba sed P ep tid om im etic In h ibitor s of Her p es Sim p lex Vir u s
Ribon u cleotid e Red u cta se: An In vestiga tion of In h ibitor Bioa ctive
Con for m a tion
Neil Moss,* Pierre Beaulieu, J ean-Simon Duceppe, J ean-Marie Ferland, J ean Gauthier, Elise Ghiro,
Sylvie Goulet, Ingrid Guse, Montse Llina`s-Brunet, Raymond Plante, Louis Plamondon, Dominik Wernic, and
Robert De´ziel
Bio-Me´ga/ Boehringer Ingelheim Research Inc., 2100 Cunard, Laval, Que´bec, Canada H7S 2G5
Received November 8, 1995X
We have been investigating peptidomimetic inhibitors of herpes simplex virus (HSV) ribo-
nucleotide reductase (RR). These inhibitors bind to the HSV RR large subunit and consequently
prevent subunit association and subsequent enzymatic activity. This report introduces a new
series of compounds that contain an extra nitrogen (a ureido function) at the inhibitor
N-terminus. This nitrogen improves inhibitor binding potency 50-fold over our first published
inhibitor series. Evidence supports that this improvement in potency results from a new
hydrogen-bonding contact between the inhibitor and the RR large subunit. This report also
provides evidence for the bioactive conformation around two important amino acid residues
contained in our inhibitors. A tert-butyl group, which contributes 100-fold to inhibitor potency
but does not directly bind to the large subunit, favors an extended â-strand conformation that
is prevalent in solution and in the bound state. More significantly, the bioactive conformation
around a pyrrolidine-modified asparagine residue, which contributes over 30 000-fold to inhibitor
potency, is elucidated through a series of conformationally restricted analogues.
In tr od u ction
ing potency over our first published inhibitor series6
without significantly increasing inhibitor size.
We are currently exploring the use of peptidomimetic
inhibitors of herpes simplex virus (HSV) ribonucleotide
reductase (RR) as a potential treatment for herpes
simplex infections. HSV RR plays a pivotal role in viral
DNA biosynthesis by catalyzing the conversion of ribo-
nucleoside diphosphates into the corresponding deoxy-
ribonucleotides.1 Our inhibitors act as mimics of the
C-terminus of the HSV RR small subunit (R2). This
C-terminal region binds to the RR large subunit (R1)
thus enabling subunit association and subsequent cata-
lytic activity.2 Inhibition results from competitive bind-
ing of the inhibitors to R1 which prevents formation of
the catalytically active holoenzyme. An attractive
characteristic of this type of inhibition is that com-
pounds based on the HSV R2 C-terminal sequence are
selective for HSV RR over human RR.3 Herpes and
human RRs operate by the same mechanism and have
similar structural features. However, the critical R2
C-terminal regions of these enzymes have little se-
quence homology; consequently, mimics of either en-
zyme’s R2 C-terminus do not cross-inhibit the other
ribonucleotide reductase.4
We have also been trying to obtain information on the
bioactive conformation of our peptidomimetic inhibitors.
This information may eventually prove useful in the
design of better inhibitors. At present no X-ray struc-
tural information exists for either subunit of HSV RR.
An X-ray structure has been published for the Escheri-
chia coli R1 bound to a peptide corresponding to the 20
C-terminal amino acids of the E. coli R2.7 Although this
fascinating structure reveals the binding mode of the
E. coli C-terminal peptide, the information is not
directly translatable to HSV RR. The R2 C-termini for
E. coli and HSV have little sequence homology, and thus
the binding modes are expected to be quite different.4
Transferred NOE spectroscopy has been used to study
the interaction between a peptide and E. coli R1.8
Unfortunately, our attempts to apply this technique to
HSV RR have proven unsuccessful. Thus, in the
absence of structural information, we have instead
deduced local inhibitor bioactive conformations by cor-
relating inhibitory potency with the conformation of
unbound inhibitors elucidated by NMR and molecular
modeling. More importantly, we have also correlated
potency with the structure of conformationally restricted
inhibitors. We previously used this approach to provide
evidence for the bioactive conformation of the aspartic
acid region of our inhibitors.9 Herein we describe how
conformational restrictions have provided insight into
the bioactive conformation of other important regions
of our inhibitors.
We recently reported that inhibitors of HSV RR
subunit association could also prevent HSV replication
in cell culture and reduce the severity of HSV-induced
keratitis in a murine ocular model.5 This work provided
the first illustration that peptidomimetic compounds can
prevent enzyme subunit association in vivo. In our
continuing search for better HSV RR inhibitors, we have
maintained an interest in improving inhibitor-binding
potency. In this paper, we report a new structural
modification that substantially improves inhibitor-bind-
Resu lts a n d Discu ssion
Ur eid o-Ba sed In h ibitor s. Early enzyme inhibition
tests of peptides that corresponded to the C-terminus
of the HSV R2 showed that the five C-terminal amino
acids, Val-Val-Asn-Asp-Leu, contained the most impor-
tant functionality for binding to R1.10 We initiated
* Current address: Boehringer Ingelheim Pharmaceuticals, Inc.,
R&D Center, 175 Briar Ridge Rd., Ridgefield, CT 06877.
X Abstract published in Advance ACS Abstracts, April 15, 1996.
S0022-2623(95)00825-9 CCC: $12.00 © 1996 American Chemical Society

Ureido-Based Peptidomimetic Inhibitors of HSV RR
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 11 2179
Ta ble 1
Ta ble 2
tor size. In this regard, we were delighted when we
found that substituting the diethylacetyl N-terminus in
compound 1 with an [(ethylpropyl)amino]carbonyl group
(see compound 4) increased inhibitor potency over 50-
fold. This new modification provided a more potent
inhibitor than the analogous compound containing the
(phenylpropionyl)valine moiety (compound 3) but with
a molecular weight closer to that of compound 1.
The fact that insertion of a NH group could have such
a pronounced effect on binding potency intrigued us.
Our previous structure-activity studies demonstrated
that truncation of the N-terminal diethylacetyl group
to an acetyl group in an inhibitor similar to compound
1 lowered binding potency over 1000-fold (compound
inactive at 1000 µM).6 We also knew that replacement
of the N-terminal valine in compound 3 with glycine
lowered potency ca. 5000-fold.11 As it was evident that
a lipophilic group at this position was essential for good
inhibitor potency, we speculated that the N-terminal
NH in compound 4 might serve to more favorably
position the lipophilic group for binding to R1. How-
ever, sequential truncation of the (ethylpropyl)amino
group in 4 to a methyl group (compounds 5 and 6) cul-
minated in a 2800-fold loss of potency. This potency
loss corresponded to the potency loss observed with the
other two inhibitor series. This argued against the NH
group in compound 4 having a significant influence on
the lipophilic binding interaction between the inhibi-
tor N-terminus and R1. The strength of this lipo-
philic binding interaction could be augmented, how-
ever, by the addition of two methyl groups (cf. com-
pounds 4 and 7).
structure-activity studies based on this sequence.
Compound 1 (Table 1) serves as a representative of our
first published inhibitor series based on this initial
investigation.6 Compound 1 contains two significant
modifications over the R2 C-terminal sequence. Firstly,
the diethylacetyl N-terminus in compound 1 was found
to effectively replace the N-terminal valine, and sec-
ondly, the replacement of the asparagine side-chain NH2
with a pyrrolidine was shown to improve inhibitor-
binding potency over 100-fold. We were initially at-
tracted to the diethylacetyl N-terminus of compound 1
because it provided inhibitors equipotent to analogues
containing the more peptidic N-acetylvaline N-terminus
(e.g., compound 2). Unfortunately, no inhibitor contain-
ing either the diethylacetyl or the N-acetylvaline N-
terminus prevented viral replication in HSV cell culture.
However, replacement of the acetyl group in compound
2 with a 3-phenylpropionyl group (compound 3) in-
creased binding potency over 30 times.11 Compound 3
provided the starting point for structure-activity stud-
ies that led to the discovery of ribonucleotide reductase
inhibitors effective at preventing HSV replication in
vitro and in vivo.5 The binding potency increase as-
sociated with the 3-phenylpropionyl group (illustrated
by compound 3) and related derivatives proved to be
very important for inhibitor antiviral activity.
The compounds shown in Table 2 provide evidence
that the N-terminal NH of this new ureido-based
inhibitor series participates in a hydrogen-bonding
interaction with R1. Replacement of this NH with a
CH2 lowers binding potency more than 40 times (cf.
compounds 8 and 9). More significantly, the corre-
sponding carbamate analogue (compound 10), which
should impart similar geometrical constraints to the
inhibitor as the ureido group in compound 8, is 200
It is generally believed that smaller molecules stand
a better chance of diffusing through biological mem-
branes. Consequently, we have been conscious of
identifying structural modifications that can improve
binding potency without significantly increasing inhibi-

2180 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 11
Moss et al.
Ta ble 3
F igu r e 1.
times less potent. The N-terminal oxygen in compound
10 obviously cannot act as a hydrogen bond donor. The
oxygen would not only be unable to partake in the same
hydrogen-bonding interaction as the NH but may also
face a repulsive interaction with the hydrogen bond
acceptor on R1. This could explain the substantial
potency difference observed. The hydrogen bond be-
tween the N-terminal NH of the urea-based inhibitors
and R1 likely constitutes a novel binding interaction not
utilized by the R2 C-terminus.
Unlike the N-terminal NH of the ureido group, the
tert-leucine NH does not contribute significantly to
binding potency. The similar IC50 values of compounds
8 and 11 indicate that the tert-leucine NH does not
interact with R1 nor is it necessary for defining local
inhibitor conformation.
It is also noteworthy that substituting the valine
moiety in compound 4 (Table 1) with tert-leucine im-
proves inhibitor-binding potency 5-fold (compound 8,
Table 2). This result is consistent with our previous
findings. Replacement of the C-terminal valine with
tert-leucine in compound 3 has also been shown to
improve binding potency 5-fold.11 This result demon-
strates that the new ureido-based N-terminus is not
influenced by the adjacent inhibitor functionality any
differently than other N-terminal groups.
The significant potency increases obtained with this
new urea-based N-terminus have of course encouraged
us to investigate the effect of these inhibitors against
HSV in vitro and in vivo. Incorporating previously
published modifications at the inhibitor C-terminus into
the ureido-based inhibitors does provide compounds
with antiviral efficacy. Details regarding the antiviral
properties of these urea-based inhibitors will be pub-
lished in a separate account.
In vestiga tion of In h ibitor Bioa ctive Con for m a -
tion . Understanding an inhibitor’s bioactive conforma-
tion and how the molecule binds to its target requires
knowledge of what inhibitor functionalities are impor-
tant for binding potency and why. Figure 1 uses
compound 8 to summarize information learned from our
first published structure-activity studies6 and the
previous section of this paper. The isopropyl group of
the C-terminal leucine and the N-terminal pentyl group
likely bind to lipophilic binding pockets on R1, while
the aspartic acid carboxyl group probably participates
in a strong hydrogen-bonding interaction. The absence
of any one of these three groups causes over a 1000-
fold loss in binding potency. Unlike the three previous
groups, the tert-butyl group in compound 8 does not
appear to interact directly with R1. Instead, it probably
reduces conformational mobility in this region of the
inhibitor and helps favors a conformation which re-
sembles the bioactive conformation. At this position,
inhibitor potency improves with increased substitution
of the amino acid â-carbon and is not strongly influenced
by the type of functionality present.6 On the basis of
this information, we looked for the presence of predomi-
nant conformations around the inhibitor tert-butyl group
in solution and investigated whether we could correlate
this to the local bioactive conformation.
(a ) Bioa ctive Con for m a tion of ter t-Leu cin e Resi-
d u e. â-Branched amino acids such as valine and
isoleucine are known to favor an extended peptide
conformation (i.e., the peptide backbone conformation
found in â-sheets).12 It seems reasonable that tert-
leucine would also favor this typical extended conforma-
tion, perhaps to a greater extent than valine or isoleu-
cine. NMR coupling constants and NOE data for the
tert-leucine portion of compound 8 and previously
published inhibitors9 (DMSO solution) show a relatively
large Ha-Hb coupling constant (8.5-10 Hz) and a strong
NOE between Hb and Hc (see compound 8 in Table 3).13
These data are typical for an extended peptide confor-
mation. Given that the conformation depicted in com-
pound 8 (Table 3) appears to predominate in DMSO
solution, we prepared conformationally restricted cyclic
inhibitors to investigate if this conformation also re-
sembles the bioactive conformation.
The strong NOE between Hb and Hc observed for
compound 8 suggested mimicking the proximity of these
two hydrogens by connecting the carbon and nitrogen
atoms attached to Hb and Hc with a 5- or 6-membered
ring. It was already known from a related inhibitor
series that replacing Hc with a methyl group had little

Ureido-Based Peptidomimetic Inhibitors of HSV RR
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 11 2181
effect on binding potency.14 This implied that the Hc
did not directly interact with R1 and thus would be
amenable to structural modification. For synthetic
simplicity, we prepared two cyclic analogues which
lacked the tert-butyl group. These cyclic compounds
were based on connecting the carbon and nitrogen atoms
attached to Hb and Hc in glycine analogue 12 (see
compounds 13 and 14, Table 3). This simplification was
justifiable even though the glycine analogue 12 lost over
200 times potency compared with the corresponding tert-
leucine derivative 8. As previously mentioned, this loss
of potency is likely a consequence of increased inhibitor
conformational mobility and not a consequence of a lost
binding interaction. Compound 12 can still achieve the
same peptide backbone conformation as compound 8;
however, compound 12 will populate many additional
conformational states.
Ta ble 4
The 6-membered ring compound 13 is ca. 3 times
more potent than the acyclic reference analogue 12,
while the 5-membered ring compound 14 is essentially
equipotent. These conformational restrictions produce
reasonably potent inhibitors and thus provide evidence
that the close proximity of Hb and Hc observed for
compound 8 in DMSO solution also exists in the bound
state. Since, by design, the cyclic amino acids in
compounds 13 and 14 have the unnatural R-configura-
tion, we prepared the (R)-alanine derivative 15 as a
further point of reference. The 50-170-fold difference
between the (R)-alanine inhibitor 15 and compounds 14
and 13 more impressively exemplifies the suitability of
these restrictions to approximate the optimal ψ angle
for inhibitor binding (Hb-CO torsion angle). The mod-
est potency of compound 13 compared to compound 8 is
probably not a consequence of a nonoptimal ψ angle in
13. More likely, the presence of the â-ring methylene
in compounds 13 and 14 forces an unfavorable steric
interaction with the adjacent urea carbonyl in the bound
state. This would be predicted if the optimal φ angle
for inhibitor binding is close to that shown in compounds
8 and 12 (Ha-Hb torsion angle around 180°). If one
considers the 170-fold difference in potency between
compounds 13 and 15 as a partial measure of how poorly
the (R)-methyl group in 15 orients the adjacent C-
terminal amide (ψ angle), it may be reasonable to
assume that the (R)-alkyl groups in compounds 13-15
have a similar negative influence on the orientation of
the adjacent urea (φ angle). Interestingly, this magni-
tude of potency difference is similar to the potency
difference between compounds 8 and 13.
(b) Bioa ctive Con for m a tion of P yr r olid in e-Mod -
ified Asp a r a gin e Resid u e. The central amino acid
side chain (asparagine modified with a pyrrolidine)
contributes enormously to binding potency in all our
inhibitor series. Replacement of this side chain with a
methyl group can reduce binding potency over 30 000
times (cf. compounds 8 and 23, Table 4). Our first
structure-activity paper indicates that both the carbo-
nyl group and the pyrrolidine (more specifically dialky-
lation of the side-chain nitrogen) contribute almost
equally and independently to inhibitor potency.6 How-
ever, it is not clear whether or to what extent these two
side-chain functionalities engage in direct binding with
R1 or contribute to favoring inhibitor bioactive confor-
mation. As this central region of our inhibitors con-
tributes so much to binding potency, we undertook an
investigation to gain information about this important
region through studying its bioactive conformation.
NMR data from previously published inhibitors9
provided information on the solution conformation
(DMSO) of the pyrrolidine-modified asparagine moiety.
The most distinctive piece of information was the
relatively large and small coupling constants observed
between Hd and He (9-11 Hz) and Hd and Hf (4-5 Hz),
respectively (see Figure 2). These data are typical for a

2182 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 11
Moss et al.
a methylamine or ethylamine substituent usually binds
10 times less effectively than the corresponding pyrro-
lidine analogue (cf. compounds 8 and 19, Table 4, and
compounds from ref 6). This result provides strong
indirect evidence that an ø2 angle of 0° for the pyrroli-
dine-modified asparagine side chain approximates the
actual angle favored for binding to R1. Significantly,
compound 18 also proved 3 times more potent than the
corresponding acyclic analogue, compound 19. This
observation provides more convincing evidence that the
pyrrolidine-modified asparagine side chain prefers to be
oriented toward the inhibitor C-terminus (ø1 ) 180°).
We also investigated the 5-membered ring derivative
20. The 5-membered ring cannot readily accommodate
the ø1 angle predicted by NMR. Thus, this derivative
was prepared to verify whether the ø1 angle predicted
by NMR was in fact necessary for good inhibitor
potency. As it turned out, compound 20 proved to be
ca. 3 times less potent than compound 16.
F igu r e 2.
relatively fixed ø1 torsion angle. However, the data are
consistent with two possibilities (ø1 ) -60° or 180°,
conformers A and B, Figure 2). The Hc-Hd coupling
constants (7-8 Hz), in contrast, do not provide conclu-
sive evidence of a fixed φ torsion angle. These coupling
values are also typical for a conformationally mobile
torsion angle. Information about the ø2 angle is more
difficult to obtain by NMR. Molecular mechanics cal-
culations of a pyrrolidine-modified asparagine (N-acetyl
N′-methylamide derivative) do not find a predominant
ø2 angle. Conformers B and C shown in Figure 2 (ø2 )
0° and 90°, respectively) have essentially the same
energy and together with conformer A constitute the
three lowest energy structures found.
It was not immediately obvious how we could deter-
mine whether any of the conformations suggested by
NMR and molecular modeling actually corresponded to
the inhibitor bioactive conformation. However, as pre-
viously mentioned, we knew that Hc could be replaced
by a methyl group with little effect on inhibitor potency.
This suggested the possibility of connecting the amide
nitrogen and the â-carbon by replacing Hc and He with
a two- or three-carbon bridge. Thus, we first prepared
conformationally restricted inhibitors 16 and 17 (Table
4). These two derivatives essentially lock the ø1 angle
of the pyrrolidine-modified asparagine side chain in
either of the two directions suggested by the NMR
data.14 Compound 16, which orients the side-chain
carbonyl toward the inhibitor C-terminus, lost only 10
times potency compared to compound 8, while compound
17, which orients the side-chain carbonyl toward the
inhibitor N-terminus, lost a much more substantial
1000-fold potency. These results provided the first
evidence that the pyrrolidine-modified asparagine side
chain prefers to be oriented toward the inhibitor C-
terminus when bound to HSV R1.
The fact that compound 16 binds 10 times less
effectively than compound 8 suggests that if the con-
formational restriction in 16 resembles the bioactive
conformation, it must not be optimally orienting the
local inhibitor functionality for binding. One evident
disfavored orientation in compound 16 would be the ø2
angle shown (i.e., ø2 ) 0°, see conformer B, Figure 2). A
steric interaction between the pyrrolidine ring and the
methylenes of the conformational restriction (see high-
lighted bonds, compound 16) would destabilize this
conformation. The presence of the conformational
restriction should not adversely affect a 90° ø2 angle (see
conformer C, Figure 2). To test whether this potential
steric interaction had any relation to inhibitor bioactive
conformation, we prepared compound 18, a conforma-
tionally restricted inhibitor that minimizes the steric
interaction when ø2 ) 0°. Compound 18 turned out to
be 3 times more potent than compound 16, even though
The previous results show how the conformational
restrictions in compounds 16, 18, and 20 provide
information on the side-chain bioactive conformation of
the pyrrolidine-modified asparagine. We were curious
whether these conformational restrictions could also
provide insight into the bioactive conformation of the
peptide backbone (i.e., the φ angle, Figure 2). This
question is interesting considering that the conforma-
tional restrictions in compounds 16 and 18 lock φ at
close to 180°. Previous NMR data for this torsion angle
indicate either conformational mobility or an angle
closer to 140-150° (J H -H ) 7-8 Hz, Figure 2).16 To
d
c
examine the effect of the 5- and 6-membered rings on
the preferred φ angle for binding to R1, we had to
remove the side-chain carbonyl and pyrrolidine groups.
Although the potency of the resulting inhibitors, com-
pounds 21-23, is substantially lower, the relative
potency differences between compounds 21-23 are
informative. The 6-membered pipecolic acid moiety in
compound 21 produces an inhibitor 7 times less potent
than the acyclic alanine compound 23, while the 5-mem-
bered proline in compound 22 provides an inhibitor
equipotent with 23. These results indicate that, in
contrast to the side-chain conformation, the 5-membered
ring does a better job than the 6-membered ring of
mimicking the preferred backbone conformation (φ
angle) for binding to R1. Interestingly, the relevant φ
angle in compound 20 was predicted by molecular
mechanics calculations to be around 135°. This value
is close to the φ angle predicted by NMR coupling
constants for acyclic analogues.
By combining the new information presented in this
paper regarding inhibitor bioactive conformation with
that previously published for the aspartic acid portion
of our inhibitors, a partial picture of the overall inhibitor
bioactive conformation starts to emerge. Figure 3 uses
compound 8 to provide a prediction based on our
conformational restrictions (dashed bonds, lower struc-
ture). At present, we do not have good direct evidence
for the bioactive conformation of the important C-
terminal leucine moiety. We also do not have direct
evidence (via conformational restriction) for the confor-
mational relationship between the aspartic acid moiety
and the pyrrolidine-modified asparagine. However,
ROESY data collected for previously published inhibi-
tors9 show a very strong NOE between the aspartic acid

Ureido-Based Peptidomimetic Inhibitors of HSV RR
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 11 2183
Sch em e 1
F igu r e 3.
NH and the asparagine HR.13 Although this informa-
tion cannot be translated into the bioactive conforma-
tion, the strength of the NOE combined with the low-
nanomolar binding potency of these inhibitors prompts
us to hypothesize that these two hydrogens may also
be close in the inhibitor-bound state.
From the conformation represented in Figure 3, it
appears that our inhibitors bind to R1 in a predomi-
nantly extended conformation. It is also noteworthy
that three bulky conformational restrictions can be
incorporated into our inhibitors without dramatically
decreasing binding potency. This suggests that one face
of the inhibitor is primarily responsible for the interac-
tion with R1.
Ma ter ia ls. Common N-Boc-L-amino acids, N-Boc-L-tert-
leucine, and L-leucine O-benzyl ester pTsOH salt were obtained
from commercial sources. Boc-2(S)-amino-4-pyrrolidino-4-
oxobutanoic acid (pyrrolidine-modified asparagine moiety) and
Boc-2(S)-amino-4-(methylamino)-4-oxobutanoic acid (used in
the preparation of inhibitor 19) were prepared as outlined in
ref 6. 3-Ethylpentanoic acid (used in the preparation of
inhibitor 9) could be prepared by the procedure in ref 18.
Syn th esis of Con for m a tion a lly Restr icted Am in o Acid
Der iva tives F ou n d in In h ibitor s 16-18 (Com p ou n d s 31-
33). The synthesis of inhibitors 16-18 required the prepara-
tion of N-Boc amino acids 31-33, respectively (Scheme 1). The
synthetic sequence outlined in Scheme 1 facilitated prepara-
tion of both cis and trans isomers 31 and 32 and allowed
unambiguous assignment of the structures of the products that
resulted from monoester hydrolysis of the N-Boc derivative of
intermediates 27 and 28.19 An alternate synthesis of com-
pound 27 has been reported.20 Chiral approaches to dicar-
boxypiperidines have been published.19 However at the con-
ception of this work, the achiral approach outlined in Scheme
1 was deemed expeditious.
A solution of compound 24 (45 g, 0.23 mol) in methanol (250
mL) containing 20% Pd(OH)2/C (2 g) was shaken under 50 psi
of H2 for 16 h. Filtration and concentration provided com-
pound 25 as a pale yellow liquid of sufficient purity for the
subsequent reaction (∼46 g): 1H NMR (400 MHz, CDCl3) δ
3.72 (s, 3 H), 3.68 (s, 3 H), 3.63 (d, J ) 3 Hz, 1 H), 3.05-2.97
(m, 2 H), 2.71-2.63 (m, 1 H), 2.18-2.08 (m, 1 H), 1.81-1.73
(m, 1 H), 1.51-1.46 (m, 2 H).
To a solution of diester 25 (3.0 g, 10 mmol) in ethanol (30
mL) were added water (120 mL) and CuCO3‚Cu(OH)2 (10 g,
45 mmol). The resultant slurry was stirred at 70 °C for 22 h
and then at room temperature for 16 h. H2S was bubbled
through the dark blue suspension (∼15 min), and the resultant
black suspension was filtered through Celite (rinsing with 4:1
water-ethanol). The clear colorless filtrate was concentrated
in vacuo to provide a white foam (∼1.9 g). This material was
dissolved in water (10 mL), and 1 N aqueous NaOH (10 mL,
10 mmol), dioxane (15 mL), and a solution of Boc2O (2.6 g, 12
mmol) in dioxane (5 mL) were added consecutively. The
reaction mixture was stirred at room temperature for 3-5 h.
Additional 1 N NaOH was added if solution pH dropped below
11. The reaction mixture was diluted with water (100 mL)
and extracted with ether (2 × 50 mL). The aqueous phase
Con clu sion
This report reveals that introduction of an appropri-
ately positioned NH near the inhibitor N-terminus can
improve inhibitor-binding potency >50-fold over our
first published inhibitor series. This NH probably takes
advantage of a hydrogen-bonding interaction not used
by the R2 C-terminus. We also provide evidence for the
bioactive conformation around the tert-leucine and
pyrrolidine-modified asparagine residues of our inhibi-
tors. We obtained this evidence by correlating the
structure of conformationally restricted inhibitors with
inhibitor-binding potency. It appears that peptide-
based inhibitors as well as the R2 C-terminus likely bind
to R1 in a predominantly extended conformation.
Exp er im en ta l Section
Ribon u cleotid e Red u cta se Bin d in g Assa y. The inhibi-
tory effect of our compounds in an HSV RR radioligand binding
assay was measured according to a published protocol.17 The
reported IC50 values are the mean of at least four separate
determinations, and the standard deviation from the mean is
also reported.
Molecu la r Mech a n ics Ca lcu la tion s. The pyrrolidine-
modified asparagine (N-acetyl N′-methylamide derivative) was
subjected to a Monte Carlo conformational search. A total of
1200 generated structures were minimized (Polak-Ribier
conjugate gradient method, 2000 iterations) with the MM3
force field as implemented in MacroModel v4.0. A dielectric
constant of 80 was used to lessen the contribution of intramo-
lecular hydrogen bonds.
NMR. The NMR data that provided a basis for preparing
our conformationally restricted inhibitors were obtained fol-
lowing the protocol in ref 9.

2184 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 11
Moss et al.
was acidified with solid citric acid and extracted with EtOAc
(2 × 50 mL). The combined organic extracts were washed with
brine, dried (MgSO4), filtered, and concentrated to afford N-Boc
amino acid derivative 26 as a white foam (2.5 g, 85%).
Interpretation of the NMR spectrum of this material was
difficult due to the mixture of diastereoisomers and rotamers
present.
Sch em e 2
To a solution of 26 (2.5 g, 8.4 mmol) in dry acetonitrile (30
mL) at 0 °C were added DBU (1.26 mL, 9.2 mmol) and BnBr
(1.1 mL, 9.2 mmol). The cooling bath was removed, and the
reaction mixture was stirred at room temperature for 4-6 h.
The solvent was remove in vacuo, and the residue was
partitioned between EtOAc (70 mL) and saturated aqueous
NaHCO3 (70 mL). The organic phase was washed with water
and brine, dried (MgSO4), filtered, and concentrated. The
resultant crude product was purified by silica gel flash
chromatography (eluent, hexane-EtOAc, 4:1) to provide a
clear colorless oil (2.8 g, 88%). The two diastereoisomers could
not be readily separated at this stage, so the mixture was
treated with 4 N HCl in dioxane (20 mL) for 30 min. The
volatiles were removed in vacuo, and the residue was parti-
tioned between EtOAc and saturated aqueous NaHCO3. The
organic phase was washed with brine, dried (MgSO4), filtered,
and concentrated. The two diastereoisomers could now be
readily separated by silica gel flash chromatography (EtOAc
eluent) to provide compounds 27 (1.2 g) and 28 (0.47 g) as
colorless viscous liquids. Compound 27: 1H NMR (400 MHz,
CDCl3) δ 7.36-7.30 (m, 5 H), 5.21-5.15 (m, 2 H), 3.71 (d, J )
3.5 Hz, 1 H), 3.56 (s, 3 H), 3.07-2.97 (m, 2 H), 2.72-2.66 (m,
1 H), 2.20-2.10 (m, 2 H), 1.82-1.75 (m, 1 H), 1.58-1.46 (m, 2
H); FAB-MS 278 (M+ + H).
H), 3.67-3.30 (br m, 4 H), 2.83-2.30 (br m, 3 H), 2.08-1.40
(br m, 7 H), 1.46 (s, 9 H); FAB-MS 327 (M+ + H).
Compound 32: 1H NMR (400 MHz, CDCl3) δ 4.97-4.91 (br,
1 H), 3.84-3.12 (br m, 7 H), 2.05-1.64 (m, 8 H), 1.47 (s, 9 H);
FAB-MS 327 (M+ + H).
Compound 33: 1H NMR (400 MHz, DMSO-d6), spectrum
shows a 65:35 mixture of rotamers, δ 8.43-8.38 (br, 0.35 H),
7.87-7.80 (br, 0.65), 5.01-4.97 (br, 0.65 H), 4.89-4.85 (br, 0.35
H), 4.05-4.00 (br, 0.35 H), 3.83-3.74 (br, 0.65 H), 3.3-2.38
(br m, 3 H, overlaps with DMSO and H2O), 2.60 (br d, J ) 3.5
Hz, 1.05 H), 2.57 (br d, J ) 4 Hz, 1.95 H), 2.01-1.30 (m, 3 H),
1.39 (br s, 9 H); FAB-MS 287 (M+ + H).
Syn th esis of th e Con for m a tion a lly Restr icted Am in o
Acid Der iva tive F ou n d in In h ibitor 20 (Com p ou n d 38).
The synthesis of N-Boc amino acid 38 started from diethyl cis-
2,3-pyrrolidinedicarboxylate (33)21 (Scheme 2). A solution of
33 (1.67 g, 7.76 mmol) in 6 N aqueous HCl (13 mL) was
refluxed for 22 h. The reaction mixture which contained a fine
black precipitate was filtered through a 4 µm Milex filter unit.
Concentration of the filtrate and drying in vacuo provided a
yellow solid (1.48 g, 97%) which was used in the next reaction
without further purification:22 1H NMR (400 MHz, DMSO-d6)
δ 4.43 (d, J ) 7 Hz, 1 H), 3.45 (ddd, J ) 8, 3.5, 3 Hz, 1 H),
3.28-3.23 (m, 2 H), 2.35-2.25 (m, 1 H), 2.19-2.11 (m, 1 H);
FAB-MS 244 (M+ + H).
Amino acid 34 was converted to N-Boc dibenzyl ester
derivative 35 by using procedures similar to those described
above (two of the reactions used in conversion of 25 to 27). An
extra 1 equiv of NaOH was required for the N-Boc reaction,
and an extra 1 equiv each of DBU and BnBr was used for the
benzylation reaction. Compound 35 was isolated as a clear
colorless liquid: 1H NMR (400 MHz, CDCl3), spectrum shows
a 65:35 mixture of rotamers, δ 7.35-7.20 (m, 10 H), 5.11-
4.89 (m, 3.65 H), 4.77 (d, J ) 12 Hz, 0.35 H), 4.69 (d, J ) 8.5
Hz, 0.35 H), 4.55 (d, J ) 8.5 Hz, 0.65 H), 3.78-3.65 (m, 1 H),
3.44-3.25 (m, 2 H), 2.48-2.36 (m, 1 H), 2.19-2.08 (m, 1 H),
1.45 (s, 3.15 H), 1.33 (s, 5.85 H); FAB-MS 440 (M+ + H).
To a solution of 35 (2.83 g, 6.45 mmol) in 3:1 THF-H2O (65
mL) was added 1 N aqueous LiOH (7.8 mL, 7.8 mmol) at a
rate to maintain reaction homogeneity. After 2 h, the reaction
mixture was diluted with water (150 mL) and extracted with
ether (2 × 100 mL). The aqueous phase was acidified to pH
2-3 with 1 N aqueous HCl and extracted with EtOAc (2 × 70
mL). The combined organic extracts were dried (MgSO4),
filtered, and concentrated to afford a white solid (1.6 g, NMR
shows a mixture of isomers). This material could be crystal-
lized from 5:1 heptane-EtOAc to provide 37 as fine needles
(0.8 g, 35%): mp 131-132 °C; 1H NMR (400 MHz, CDCl3),
spectrum shows a 65:35 mixture of rotamers, δ 7.34-7.29 (m,
5 H), 5.18-5.04 (m, 2 H), 4.69 (d, J ) 8.5 Hz, 1 H), 4,57 (d, J
) 8.5 Hz, 1 H), 3.78-3.64 (m, 1 H), 3.46-3.24 (m, 2 H), 2.44-
2.33 (m, 1 H), 2.19-2.08 (m, 1 H), 1.46 (s, 3.15 H), 1.33 (s,
5.85 H); FAB-MS 350 (M+ + H). Anal. Calcd for C18H23NO6:
C, 61.88; H, 6.64; N, 4.01. Found: C, 61.80; H, 6.64; N, 3.94.
Proof of structure was obtained by X-ray crystallography
(structure available as Supporting Information).
Compound 28: 1H NMR (400 MHz, CDCl3) δ 7.39-7.31 (m,
5 H), 5.19-5.11 (m, 2 H), 3.70 (d, J ) 9 Hz, 1 H), 3.55 (s, 3 H),
3.10-3.04 (m, 1 H), 2.72-2.63 (m, 2 H), 2.06-2.01 (m, 1 H),
1.52-1.61 (m, 2 H), 1.52-1.40 (m, 1 H); FAB-MS 278 (M+
+
H). The two coupling constants reported for compounds 27
and 28 (NCHCO2) were consistent with the relative stereo-
chemistries shown.
To a solution of compound 27 or 28 (1.7 mmol) in THF (5
mL) was added Boc2O (2 mmol), and the resultant mixture
was stirred at room temperature for 16 h. The reaction
mixture was concentrated and the residue purified by flash
chromatography (eluent, hexane-EtOAc, 4:1) to provide clear
colorless oils in essentially quantitative yield. To a solution
of this oil (1.7 mmol) in 3:1 THF-water (13 mL) was added a
solution of aqueous LiOH (1 N, 2 mL, 2 mmol) dropwise at a
rate to maintain a clear homogeneous reaction mixture. After
complete addition, the reaction mixture was stirred at room
temperature for 1-2 h, diluted with water (60 mL), and
extracted with ether to remove any unreacted starting mate-
rial. The aqueous phase was acidified to pH ∼2 with 0.5 N
aqueous HCl and extracted with EtOAc (2 × 50 mL). The
combined organic extracts were washed with brine, dried
(MgSO4), filtered, and concentrated to provide compound 29
or 30 as a sticky solid (∼1 mmol, each product was contami-
nated with material resulting from benzyl ester hydrolysis).
NMR of this material was complex (rotamers); however, it
indicated that methyl ester hydrolysis was favored over benzyl
ester hydrolysis by ca. 2:1. Crude 29 or 30 was subjected to
amide bond formation with either pyrrolidine or methylamine
hydrochloride by using the typical coupling procedure outlined
in Inhibitor Synthesis. In each case, the desired isomer could
be isolated pure by flash chromatography (∼20-30% yield
from compound 27 or 28).
The pure benzyl ester derivative (0.5 mmol) was mixed with
methanol (5 mL) and 10% Pd/C (50 mg) and stirred under 1
atm of H2 for 1-2 h. The reaction mixture was filtered and
concentrated to provide N-Boc amino acid derivatives 31 and
32 as white foams. These compounds were of sufficient purity
for coupling to â-benzyl-L-aspartic acid L-leucine O-benzyl
ester. Analytical samples could be obtained by reverse phase
HPLC (Whatman Partisil 10 ODS-3 column, eluent 0-60%
acetonitrile in water, both solvents containing 0.06% TFA)
followed by lyophilization. Compound 31: 1H NMR (400 MHz,
CDCl3), spectrum shows a 65:35 mixture of rotamers, δ 5.48-
5.44 (br, 0.65 H), 5.28-5.22 (br, 0.35 H), 4.02-3.84 (br m, 1
Compound 37 was converted to the N-Boc amino acid
derivative 38 as described for compounds 31-33. Compound
38 was obtained as a white foam (93% yield): 1H NMR (400
MHz, CDCl3), spectrum shows a 60:40 mixture of rotamers, δ
4.62-4.58 (m, 0.4 H), 4.43 (br d, J ) 7.5 Hz, 0.6 H), 3.83-

Ureido-Based Peptidomimetic Inhibitors of HSV RR
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 11 2185
Sch em e 3
Sch em e 4
3.29 (br m, 7 H), 2.62-2.50 (br m, 1 H), 2.05-1.81 (br m, 5
H), 1.48 (s, 3.6 H), 1.43 (s, 5.4 H); FAB-MS 312 (M+ + H).
P r ep a r a tion of th e ter t-Bu tylsu ccin oyl Moiety F ou n d
in In h ibitor 11 (Com p ou n d 40). Scheme 3 outlines the
preparation of compound 40. To a solution of (R)-tert-butyl-
succinic anhydride23 (39) (50 g, 0.32 mol) in dry pyridine (1.2
L) at -40 °C was added a solution of 3-aminopentane (33.5 g,
0.38 mol) in pyridine (0.1 L). The reaction mixture was
allowed to warm to room temperature and stirred for 16 h.
Volatiles were removed under aspirator vacuum (50 °C), and
the residue was dissolved in EtOAc (500 mL). This solution
was washed with 20% citric acid (4 × 100 mL) and brine (2 ×
100 mL), dried (MgSO4), filtered, and concentrated to provide
a colorless liquid. This material was dissolved in 3:1 hexane-
ether (400 mL) (warming required) and allowed to crystallize
for 4 h at room temperature and then for 12 h at 5 °C.
Filtration and washing with 10% ether in hexane provided
compound 40 as a white solid (58.5 g, 75%): mp 124-127 °C;
[R]23D -7.5° (c 1.00, CHCl3); 1H NMR (400 MHz, CDCl3) δ 5.40
(d, J ) 9 Hz, 1 H), 3.79-3.70 (m, 1 H), 2.78 (dd, J ) 11.5, 3.5
Hz, 1 H), 2.53 (dd, J ) 14.5, 11.5 Hz, 1 H), 2.41 (dd, J ) 14.5,
3.5 Hz, 1 H), 1.58-1.45 (m, 2 H), 1.40-1.28 (m, 2 H), 1.02 (s,
9 H), 0.87 (t, J ) 7.5 Hz, 3 H), 0.85 (t, J ) 7.5 Hz, 3 H); FAB-
MS 244 (M+ + H). Anal. Calcd for C13H25NO3: C, 64.17; H,
10.36; N, 5.76. Found: C, 63.99; H, 10.54; N, 5.70.
P r ep a r a t ion of Ca r b a m a t e N-Ter m in u s F ou n d in
In h ibitor 10 (Com p ou n d 42). Scheme 3 outlines the prepa-
ration of compound 42. L-tert-Leucine O-benzyl ester hydro-
chloride salt (41) (92 mg, 0.42 mmol) was treated with a
commercially available solution of phosgene in toluene (1.93
M, 4 mL). The suspension was refluxed (dry ice condenser)
for 1 h, after which time excess phosgene was removed by
bubbling nitrogen through the reaction solution. 3-Pentanol
(180 µL, 1.65 mmol) was added to the isocyanate solution, and
the resultant mixture was stirred at 95 °C for 3 h. The
reaction mixture was diluted with EtOAc (20 mL) and washed
with saturated aqueous sodium bicarbonate and brine. Drying
(MgSO4), filtering, and concentrating afforded a thick colorless
liquid (90 mg). This material was mixed with ethanol (6 mL)
and 10% Pd/C (5 mg) and stirred under an atmosphere of H2
for 2 h. Filtration and concentration provided compound 42
as a thick oil in sufficient purity for the subsequent coupling
reaction (65 mg): 1H NMR (400 MHz, DMSO-d6) δ 7.07 (d, J
) 9 Hz, 1 H), 4.55-4.44 (m, 1 H), 3.83 (d, J ) 9 Hz, 1 H),
1.56-1.42 (m, 4 H), 0.76 (s, 9 H), 0.86-0.80 (m, 3 H); FAB-
MS 246 (M+ + H).
washed with brine, dried (MgSO4), filtered, and concentrated.
The crude product was purified by silica gel flash chromatog-
raphy (hexane-EtOAc, 10:1) to provide 43 as a clear viscous
liquid (1.78 g, 84%): 1H NMR (400 MHz, CDCl3) δ 7.40-7.28
(m, 5 H), 5.95-5.82 (m, 1 H), 5.33-5.16 (m, 4 H), 4.87 (dd, J
) 9, 4 Hz, 1 H), 4.57-4.53 (m, 2 H), 2.49-2.32 (m, 3 H), 2.23-
2.13 (m, 1 H), 1.45 (s, 9 H), 1.39 (s, 9 H).
Compound 43 was converted to compound 44 by established
procedures: deallylation,27 reduction of the acid to the alco-
hol,26 and oxidation.28 Compound 44: 1H NMR (400 MHz,
CDCl3) δ 9.71 (s, 1 H), 7.41-7.30 (m, 5 H), 5.28 (d, J ) 13 Hz,
1 H), 5.20 (d, J ) 13 Hz, 1 H), 4.82 (dd, J ) 8.5, 4 Hz, 1 H),
2.60-2.41 (m, 3 H), 2.22-2.12 (m, 1 H), 1.45 (s, 9 H), 1.39 (s,
9 H).
To a solution of tripeptide salt 46 (0.5 mmol) in EtOH (7
mL) were added a solution of aldehyde 44 or 45 (0.5 mmol) in
MeOH (3 mL) and solid NaOAc to adjust pH to ∼5. The
resultant mixture was cooled to 0 °C, and NaCNBH3 (0.6
mmol) was added in three portions over 15 min. The reaction
mixture was stirred at room temperature for 3 h, the solvent
was evaporated, and the residue was partitioned between
EtOAc and brine. The organic phase was washed with brine,
dried (MgSO4), filtered, and concentrated. The crude reductive
amination product was purified by silica gel flash chromatog-
raphy to provide white foams (50-80%). This material was
treated with a 1:1 mixture of TFA and dichloromethane (10
mL) for 1 h. The volatiles were removed in vacuo, and the
residue was dissolved in acetonitrile (10 mL). NMM (3 equiv)
and BOP reagent (1 equiv) were added, and the reaction
mixture was stirred at room temperature for 1 h. The solvent
was evaporated, and the residue was partitioned between
EtOAc and saturated aqueous NaHCO3. The organic phase
was washed with 1 N HCl and brine, dried (MgSO4), filtered,
and concentrated. The crude products were purified by silica
gel flash chromatography to provide compounds 47 as white
foams (80-85%). Compounds 47 (0.2 mmol) were mixed with
1:1 THF-MeOH (10 mL) and 10% Pd/C (50 mg) and stirred
under an atmosphere of H2 for 2 h. The reaction mixture was
filtered and concentrated, and the residue was dissolved/
suspended in dry dichloromethane (10 mL). Triethylamine
(0.7 mmol) and 1-ethylpropyl isocyanate (0.7 mmol) were
added, and the resultant mixture was stirred at room tem-
perature for 16 h, after which time the volatiles were removed
in vacuo. The residue was dissolved in water (∼5 mL), and
the resultant solution was acidified to pH 4 with 1 N HCl.
This solution was passed through a preparative HPLC C18
reverse phase column (Vydac, 15 µm particle size) eluting with
a 0.06% aqueous TFA-0.06% TFA in acetonitrile gradient (0-
43% acetonitrile) to provide, after concentration and lyo-
philization of the appropriate fractions, compounds 13 and 14
as fluffy white solids (50-80%). Full characterization of these
compounds is provided in the Supporting Information.
In h ibitor Syn th esis. All inhibitors were prepared by
solution phase peptide synthesis, in which N-Boc-amino acid
derivatives were coupled sequentially from C- to N-terminus
by using benzotriazol-1-yl-1,1,3,3-tetramethyluronium tet-
rafluoroborate (TBTU) as the coupling agent. Removal of the
P r ep a r a tion of In h ibitor s 13 a n d 14. The synthesis of
compounds 13 and 14 relied on the preparation of aldehyde
derivatives 44 and 45,25 respectively (Scheme 4). The suc-
cessful reductive amination of compound 44 required the
R-amino group to be diprotected.26 N-Cbz-â-allyl-D-aspartic
acid tert-butyl ester, available by standard amino acid protec-
tion reactions, was converted to compound 43 by the following
procedure. To a solution of the former (1.67 g, 4.43 mmol) in
CH3CN (10 mL) were added Boc2O (1.06 g, 4.87 mmol) and
DMAP (54 mg, 0.44 mmol). The reaction mixture was stirred
at room temperature for 16 h, after which time the solvent
was evaporated and the residue partitioned between EtOAc
and saturated aqueous NaHCO3. The organic phase was

2186 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 11
Moss et al.
N-Boc protective group was effected with 4 N HCl in dioxane.
The following procedure is representative. To a solution of
N-Boc-amino acid (1 mmol) in dry acetonitrile (2.5 mL) were
added TBTU (1 mmol) and N-methylmorpholine (1 mmol).
After ca. 5 min, this solution was added to a solution of
L-leucine O-benzyl ester pTsOH salt or peptide hydrochloride
salt (1 mmol) in dry acetonitrile (2.5 mL) containing N-
methylmorpholine (2 mmol). The reaction mixture was stirred
at ambient temperature for 2-6 h (reaction monitored by TLC)
and then poured into a mixture of EtOAc (50 mL) and
saturated aqueous sodium bicarbonate (50 mL). The organic
phase was washed with another portion of sodium bicarbonate,
1 N aqueous HCl (2 × 50 mL), and brine (50 mL). Drying
(MgSO4), filtration, and concentration provided the peptide,
usually of sufficient purity to continue to the next step without
further purification. N-Boc peptide derivatives could be puri-
fied if necessary by conventional flash chromatography. The
N-Boc peptide product (1 mmol) was then treated with 4 N
HCl in dioxane (5 mL) for 30 min. The solvent was removed
under vacuum, and the resulting hydrochloride salt was
subjected to high vacuum before its use in the next coupling
reaction.
The conformationally restricted amino acid derivatives 31-
33 and 38 used in the preparation of inhibitors 16-18 and
20, respectively, were incorporated as racemates. The dias-
tereoisomers obtained on coupling these racemic amino acid
derivatives with â-benzyl-L-aspartic acid L-leucine O-benzyl
ester were readily separated by flash chromatography. Both
diastereoisomers were elaborated to the corresponding final
peptide, and in each case, one of the diastereoisomers proved
to have inhibitory activity at least 100-1000 times greater
than the other (potency differences affected by traces of
contaminating active isomer). For compounds 31-33, the first
diastereoisomer to elute provided the most potent inhibitor,
while for compound 38, the bottom isomer provided the most
potent inhibitor. The configuration of the active diastereo-
isomer was assigned as corresponding to L based on precedent
with other active inhibitors containing L-asparagine deriva-
tives (e.g., compounds 8 and 19) and the fact that a previously
made peptide inhibitor containing D-asparagine had signifi-
cantly reduced inhibitory activity.12
The various N-terminal alkylureido functionalities were
introduced as follows. To a solution of the appropriate peptide
hydrochloride salt (0.5 mmol) in dry dichloromethane (5 mL)
were added N-methylmorpholine (1.5 mmol) and the relevant
isocyanate (1 mmol, usually neat). The reaction mixture was
stirred at room temperature for 16 h, after which time the
volatiles were removed and the residue was partitioned
between EtOAc and 0.1 M aqueous HCl. The organic phase
was washed with 5% aqueous sodium bicarbonate and brine,
dried (MgSO4), filtered, and concentrated. The resultant crude
product was purified by silica gel flash chromatography to
provide the desired ureido derivatives usually as white foams
(64-76% yield). Isopropyl and methyl isocyanates were
obtained from commercial sources. 1-Propylbutyl and 1-eth-
ylpropyl isocyanates were prepared from commercially avail-
able 4-aminoheptane and 3-aminopentane, respectively, by
following the procedure in ref 24.
Pitner, P. J . J ones, and S. F. Leonard for relevant NMR
observations from our previously published inhibitors
and to L. Tong for the X-ray structure of compound 37.
Su p p or tin g In for m a tion Ava ila ble: Full tabulation of
1H NMR, FAB mass spectra, amino acid analysis, elemental
analysis, and HPLC purity data for new inhibitors and the
X-ray structure of compound 37 (16 pages). Ordering informa-
tion is given on any current masthead page.
Refer en ces
(1) Cory, J . G. Role of ribonucleotide reductase in cell division. In
Inhibitors of Ribonucleoside Diphosphate Reductase Activity;
Cory, J . G., Cory, A. H., Eds.; Pergamon Press Inc.: New York,
1989; pp 1-17.
(2) Filatov, D.; Ingemarson, R.; Graslund, A.; Thelander, L. The role
of herpes simplex virus ribonucleotide reductase small subunit
carboxyl terminus in subunit interaction and formation of iron-
tyrosyl center structure. J . Biol. Chem. 1992, 267, 15816-15822.
(3) (a) Dutia, B. M.; Frame, M. C.; Subak-Sharpe, J . H.; Clarke, W.
N.; Marsden, H. S. Specific inhibition of herpes virus ribonucle-
otide reductase by synthetic peptides. Nature 1986, 321, 439-
441. (b) Cohen, E. A.; Gaudreau, P.; Brazeau, P.; Langelier, Y.
Specific inhibition of herpes simplex virus ribonucleotide reduc-
tase activity by
a nonapeptide derived from the carboxyl
terminus of subunit 2. Nature 1986, 321, 441-443.
(4) Cosentino, G.; Lavalle´e, P.; Rakhit, S.; Plante, R.; Gaudette, Y.;
Lawetz, C.; Whitehead, P. W.; Duceppe, J .-S.; Le´pine-Frenette,
C.; Dansereau, N.; Guilbault, C.; Langelier, Y.; Gaudreau, P.;
Thelander, L.; Guindon, Y. Specific inhibition of ribonucleotide
reductases by peptides corresponding to the C-terminal of their
second subunit. Biochem. Cell Biol. 1991, 69, 79-83.
(5) Liuzzi, M.; De´ziel, R.; Moss, N.; Beaulieu, P.; Bonneau, A.-M.;
Bousquet, C.; Chafouleas, J . G.; Garneau, M.; J aramillo, J .;
Krogsrud, R. L.; Lagace´, L.; McCollum, R. S.; Nawoot, S.;
Guindon, Y. A potent peptidomimetic inhibitor of HSV ribo-
nucleotide reductase with antiviral activity in vivo. Nature 1994,
372, 695-698.
(6) Moss, N.; De´ziel, R.; Adams, J .; Aubry, N.; Bailey, M.; Baillet,
M.; Beaulieu, P.; DiMaio, J .; Duceppe, J .-S.; Ferland, J .-M.;
Gauthier, J .; Ghiro, E.; Goulet, S.; Grenier, L.; Lavalle´e, P.;
Le´pine-Frenette, C.; Plante, R.; Rakhit, S.; Soucy, F.; Wernic,
D.; Guindon, Y. Inhibition of herpes simplex virus type
1
ribonucleotide reductase by substituted tetrapeptide derivatives.
J . Med. Chem. 1993, 36, 3005-3009.
(7) Ulin, U.; Eklund, H. Structure of ribonucleotide reductase
protein R1. Nature 1994, 370, 533-539.
(8) Bushweller, J . H.; Bartlett, P. A. Investigation of an octapeptide
inhibitor of escherichia coli ribonucleotide reductase by trans-
ferred nuclear Overhauser effect spectroscopy. Biochemistry
1991, 30, 8144-8150.
(9) Moss, N.; De´ziel, R.; Ferland, J .-M.; Goulet, S.; J ones, P.-J .;
Leonard, S. F.; Pitner, T. P.; Plante, R. Herpes simplex virus
ribonucleotide reductase subunit association inhibitors: the
effect and conformation of â-alkylated aspartic acid derivatives.
Bioorg. Med. Chem. 1994, 2, 959-970.
(10) Gaudreau, P.; Paradis, H.; Langelier, Y.; Brazeau, P. Synthesis
and inhibitory potency of peptides corresponding to the subunit
2 C-terminal region of herpes virus ribonucleotide reductase. J .
Med. Chem. 1990, 33, 723-730.
(11) Moss, N.; Beaulieu, P.; Duceppe, J .-S.; Ferland, J .-M.; Gauthier,
J .; Ghiro, E.; Goulet, S.; Grenier, L.; Llinas-Brunet, M.; Plante,
R.; Wernic, D.; De´ziel, R. Peptidomimetic inhibitors of HSV
ribonucleotide reductase: a new class of antiviral agents. J . Med.
Chem. 1995, 38, 2723-2730.
(12) Chou, P. Y.; Fasman, D. G. Prediction of the secondary structure
of proteins from their amino acid sequence. Adv. Enzymol. 1978,
47, 45-148.
The last reaction involved in the preparation of our inhibi-
tors involved removal of the two benzyl ester protective groups
by catalytic hydrogenolysis (10 mol % 10% Pd/C in methanol
under 1 atm of H2 for 3 h). The resultant inhibitor was often
obtained in >95% purity (HPLC and NMR), but when neces-
sary it was purified by preparative HPLC on a C18 reverse
phase column (Vydac, 15 µm particle size) eluting with 0.06%
aqueous TFA-0.06% TFA in acetonitrile gradients.
In h ibitor Ch a r a cter iza tion a n d P u r ity. All peptide
derivatives showed satisfactory 400 MHz 1H NMR spectra,
FAB mass spectra (M+ + H) and/or (M+ + Na), amino acid
analysis including peptide recovery, and HPLC purity in two
solvent systems (>95%).
(13) A full account of the NMR-derived solution structure of these
compounds and others will be published separately.
(14) Unpublished observation.
(15) Even though the restricted amino acid derivative in compound
17 contains two substituents in the axial position on the
piperidine ring, the conformation shown predominates in solu-
tion (4.5-5 Hz). Substantial A1,3 strain between the two back-
bone amides likely destabilizes the alternate chair form. A slight
twisting of the chair conformations shown for compounds 16-
18 may exist, but this should not invalidate the general
conclusions made.
(16) Pardi, J .; Billeter, M.; Wu¨thrich, K. Calibration of the angular
dependence of the amide proton-CR proton coupling constants
in a globular protein. J . Mol. Biol. 1984, 180, 741-751.
(17) Krogsrud, R. L.; Welchner, E.; Scouten, E.; Liuzzi, M. A solid-
phase assay for the binding of peptidic subunit association
inhibitors to the herpes simplex virus ribonucleotide reductase
large subunit. Anal. Biochem. 1993, 213, 386-390.
Ack n ow led gm en t. We are grateful to M. Liuzzi and
E. Scouten for determining the IC50s of our inhibitors
in the binding assay. We are also grateful to T. P.

Ureido-Based Peptidomimetic Inhibitors of HSV RR
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 11 2187
(18) Kepner, R. E.; Winstein, S.; Young, W. G. Allylic rearrangements.
XXIV. Abnormal bimolecular substitution. The condensation of
butenyl and pentenyl chlorides with sodium malonic ester. J .
Am. Chem. Soc. 1949, 71, 114-119.
(19) Monoester hydrolysis of related derivatives is known to be
nonselective. Gmeiner, P.; Feldman, P. L.; Chu-Moyer, M. Y.;
Rapoport, H. An efficient and practical total synthesis of (+)-
vicamine from L-aspartic acid. J . Org. Chem. 1990, 55, 3068-
3074.
(20) J ohansen, J . E.; Christie, B. D.; Rapoport, H. Iminium salts from
R-aminoacid decarbonylation. Application to the synthesis of
octahydroindolo[2,3-a]quinolizines. J . Org. Chem. 1981, 46,
4914-4920.
(21) Ro¨der, E.; Wiedenfeld, H.; Bourauel, T. Synthese von 2,3-Bis-
(ethoxycarbonyl)-1H-pyrrol-1-propionsa¨ure-ethylester. Liebigs
Ann. Chem. 1987, 1117-1119.
(23) Polonski, T. Chiroptical properties and molecular geometry of
substituted succinic anhydrides and imides. J . Chem. Soc. Perkin
Trans. 1 1988, 629-637.
(24) Goldesmidt, V. S.; Wick, M. U¨ ber peptid-synthesen I. Liebigs
Ann. Chem. 1952, 575, 217-231.
(25) Valerio, R. M.; Alewood, P. F.; J ohns, R. B. Synthesis of optically
active 2-(tert-butoxycarbonylamino)-4-dialkyloxyphosphorylbu-
tanoate protected isosteres of O-phosphonoserine for peptide
synthesis. Synthesis 1988, 786-789.
(26) Zydowsky, T. M.; Dellaria, J . F.; Nellans, H. N. Efficient and
versatile synthesis of dipeptide isosteres containing γ- or δ-lac-
tams. J . Org. Chem. 1988, 53, 5607-5616.
(27) De´ziel, R. Mild palladium(0)-catalyzed deprotection of allyl
esters. A useful application in the synthesis of carbapenems and
other â-lactam derivatives. Tetrahedron Lett. 1987, 28, 4371-
4372.
(28) Dess, D. B.; Martin, J . C. Readily accessible 12-I-5 oxidant for
the conversion of primary and secondary alcohols to aldehydes
and ketones. J . Org. Chem. 1983, 48, 4155-4156.
(22) A chiral synthesis of the neutral amino acid of compound 34
has been reported. Humphrey, J . M.; Bridges, R. J .; Hart, J . A.;
Chamberlin, A. R. 2,3-Pyrrolidinedicarboxylates as neurotrans-
mitter conformer mimics: enantioselective synthesis via chela-
tion-controlled enolate alkylation. J . Org. Chem. 1994, 59, 2469-
2472.
J M950825X