Structure-guided design of C2-symmetric HIV-1 protease inhibitors based on a pyrrolidine scaffold

Article abstract of DOI:10.1021/jm701142s

Infections with the human immunodeficiency virus, which inevitably lead to the development of AIDS, are still among the most serious global health problems causing more than 2.5 million deaths per year. In the pathophysiological processes of this pandemic

Full text of DOI:10.1021/jm701142s

2078
J. Med. Chem. 2008, 51, 2078–2087
Structure-Guided Design of C2-Symmetric HIV-1 Protease Inhibitors Based on a Pyrrolidine
Scaffold†
Andreas Blum,‡ Jark Böttcher,‡ Andreas Heine, Gerhard Klebe, and Wibke E. Diederich*
Institut für Pharmazeutische Chemie, Philipps-UniVersität Marburg, Marbacher Weg 6, 35032 Marburg, Germany
ReceiVed September 13, 2007
Infections with the human immunodeficiency virus, which inevitably lead to the development of AIDS, are
still among the most serious global health problems causing more than 2.5 million deaths per year. In the
pathophysiological processes of this pandemic, HIV protease has proven to be an invaluable drug target
because of its essential role in the virus’ replication process. By use of pyrrolidine as core structure, symmetric
3,4-bis-N-alkylsulfonamides were designed and synthesized enantioselectively from D-(-)-tartaric acid as a
new class of HIV protease inhibitors. Structure-guided design using the cocrystal structure of an initial lead
as starting point resulted in a second series of inhibitors with improved affinity. The binding modes of four
representatives were determined by X-ray crystallography to elucidate the underlying factors accounting
for the SAR. With this information for further rational design, the combination of suitable side chains resulted
in a final inhibitor showing a significantly improved affinity of Ki ) 74 nM.
Introduction
virus variants.7 Because a similar mode of action of all approved
inhibitors has resulted in pronounced cross resistance, the need
of a steadfast and continuous search for new inhibitors is evident.
An increased structural diversity of inhibitor scaffolds could
be a possible strategy to at least diminish the accelerated
development of multidrug resistant variants.
Infection with the human immunodeficiency virus (HIVa)
inevitably leads to the development of the acquired immune
deficiency syndrome (AIDS). The WHO estimates that currently
40 million people are infected worldwide and that this number
will increase continuously.1 In the past 20 years, chemotherapy
against HIV infection has been tackled by modern drug
discovery and development. Several stages in the viral replica-
tion have been targeted to reduce the viral load, thus delaying
the progression to AIDS. However, an entire remedy of the
infection or a vaccination is still an unaccomplished goal.2 One
of the most successful approaches in antiviral therapy up to date
is the highly active antiretroviral therapy (HAART) which
combines inhibitors of the viral reverse transcriptase and the
viral protease. This viral protease, the HIV protease, belongs
to the class of aspartic proteases and cleaves precursor polypep-
tides into functional viral proteins, which are essential for the
infectiousness of the virus particles. The HIV protease is a C2-
symmetric homodimer with each monomer consisting of 99
residues. Two glycine-rich flaps are located above the large
active site, which binds six to seven amino acids of the nine
different cleavage sites within the gag and pol polypeptides.
Because inhibition of this enzyme leads to immature virions,
HIV protease has become a target of several marketed drugs.
Up to now, nine inhibitors against the HIV protease have been
approved by the FDA,3–6 but meanwhile the virus has shown a
very high degree of adaptability. The high mutation rate of the
virus is caused by the error-prone viral reverse transcriptase and
its fast replication rate. This leads, especially under the additional
selection pressure of HAART, to the development of resistant
Most of the approved inhibitors are transition state mimetics
addressing the catalytic dyad (D25A/D25B) via a hydroxyl
group. As a different approach, cyclic amines have been
identified as nonpeptidic inhibitors for the aspartic proteases
renin,8,9 ꢀ-secretase,10 and plasmepsins.11,12 Recently, a pyrro-
lidine-based inhibitor for HIV protease has been developed.13,14
The X-ray structure of the protein inhibitor complex revealed
the endocyclic amino function binding to the catalytic dyad.
Electrostatic calculations suggest that the amino functionality
should be protonated and both aspartic acid residues of the
catalytic dyad therefore deprotonated, thus leading to strong
hydrogen bonds as well as electrostatic interactions of the
resulting cyclic ammonium function with the catalytic dyad.15
The sulfonamide side chain of the inhibitor addresses the flap
region by one of its oxygen atoms through hydrogen bonding
to the backbone NH of I50A (Figure 1). A superposition of
this structure with other crystal structures of HIV protease
complexes indicated that the distance between the sulfonyl group
and the endocyclic nitrogen is too large for appropriate flap
interactions. The carboxamide part of inhibitor 1 showed no
directed polar interactions to the enzyme and was therefore not
considered for further inhibitor design.
Guided by the described polar interactions of inhibitor 1 to
the enzyme, a basic pharmacophore model was derived. As core
structure, the 3S,4S-disubstituted pyrrolidine, addressing the
catalytic dyad via its secondary amino group, was retained.
However, as indicated by the X-ray structure of 1, the exocyclic
methylene groups were removed in order to improve the
interaction pattern of the sulfonyl groups with the flap region
of the enzyme. To exploit the C2 symmetry of the enzyme,
exclusively symmetric inhibitors were synthesized. To mimic
the subpocket occupation of 1, arylsulfonamides were introduced
to occupy the S2- and S2′-subpockets. In addition, a hydrophobic
†
PDB codes: 6ad, 2PQZ; 6af, 2QNP; 6cd, 2QNQ; 6fd, 2PWR; 6gd,
2PWC; 6fg, 2QNN.
* To whom correspondence should be addressed. Phone: +49-
6421-28-25810. Fax: +49-6421-28-26254. E-mail: wibke.diederich@
staff.uni-marburg.de.
‡
Equally contributing authors.
a Abbreviations: AIDS, acquired immunodeficiency syndrome; FDA,
Food and Drug Administration; HAART, highly active antiretroviral therapy;
HIV, human immunodeficiency virus; rmsd, root-mean-square deviation;
SAR, structure–activity relationship; vdW, van der Waals; WHO, World
Health Organization.
10.1021/jm701142s CCC: $40.75
2008 American Chemical Society
Published on Web 03/19/2008

Pyrrolidines as HIV Protease Inhibitors
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 7 2079
Scheme 2a
a (a) Tf2O, pyridine, CH2Cl2 -78 to -10 °C; (b) NaN3, DMPU, room
temp; (c) H2, Pd/C, hexane/EtOAc (86% over three steps).
Scheme 3a
a (a) ArSO2Cl, Et3N, CH2Cl2. 4a: Ar ) Ph (86%). 4b: Ar ) o-Me-Ph
(87%). 4c: Ar ) o-Cl-Ph (94%). 4d: Ar ) p-NO2-Ph (69%). 4e: Ar )
p-CN-Ph (71%).
Scheme 4a
Figure 1. Design of C2-symmetric inhibitors starting from the cocrystal
structure of 1.
Scheme 1a
a (a) BnNH2, xylene, Dean–Stark (81%); (b) LAH, THF, reflux (69%);
(c) H2, Pd/C, BOC2O, MeOH (74%).
a (a) RCH2Br, K2CO3, CH3CN, reflux. 4a: Ar ) Ph. (5aa: R ) CHdCH2,
72%. 5ab: R ) CH(CH3)dCH2, 73%. 5ac: R ) CHdC(CH3)2, 68%. 5ad:
R ) Ph, 75%. 5ae: R ) p-Br-Ph, 76%. 5af: R ) p-I-Ph, 76%. 5ag: R )
p-CF3-Ph, 49%.) 4b: Ar ) o-Me-Ph. (5bd: R ) Ph, 78%.) 4c: Ar ) o-Cl-
Ph. (5cd: R ) Ph, 65%.) 4d: Ar ) p-NO2-Ph. (5dd: R ) Ph, 96%.) 4e: Ar
) p-CN-Ph. (5ed: R ) Ph, 67%. 5eg: R ) p-CF3-Ph, 45%.)
moiety was attached to each of the sulfonamide nitrogen atoms
designed to address the S1- and S1′-subsites of the enzyme
(Figure 1).
Scheme 5a
Chemistry
3,4-Difunctionalized pyrrolidines are accessible in enantiopure
form commencing with commercially available tartaric acids.16,17
D-(-)-Tartaric acid was condensed with benzylamine to furnish
the corresponding cyclic imide, which was further reduced with
LiAlH4 to yield the benzyl-protected pyrrolidine-3,4-diol 2a18
(Scheme 1). The synthesis of the corresponding benzyl protected
pyrrolidine-3,4-diamine, accessible via a Mitsunobu reaction of
diol 2a with HN3 and subsequent reduction of the resulting
diazide, has already been described.19 However, following this
synthetic route, the removal of the benzyl-protecting group by
catalytic hydrogenation in the last step of the synthesis to give
rise to the final inhibitors remained unsuccessful, even after
applying higher temperature and hydrogen pressure as well as
different hydrogen sources. Furthermore, we intended to avoid
the usage of the very toxic hydrazoic acid in the Mitsunobu
reaction. Therefore, a change to the easily removable BOC
protecting group was necessary. This exchange was carried out
at the diol stage by hydrogenation of 2a in the presence of
BOC2O,20 yielding the BOC-protected 3,4-diol 2b (Scheme 1).
Although an activation of the 3,4-pyrrolidine diol for the purpose
of nucleophilic displacement reactions by conversion into its
corresponding disulfonates has been described,21,22 in our hands,
the respective mesylate did not undergo substitution with NaN3
in DMSO even at 120 °C. Nevertheless, utilization of the related
but significantly more reactive bis-triflate 2c followed by its
nucleophilic substitution with NaN3 in DMPU at room temper-
a (a) 2 M HCl in Et2O, room temp. Details are given in Table 1.
ature now yielded the 3,4-diazide 2d, which was subsequently
reduced to the corresponding pyrrolidine diamine 3 by catalytic
hydrogenation (Scheme 2) in excellent overall yield.
Diamine 3 was then condensed with suitable sulfonyl
chlorides to the corresponding sulfonamides 4 (Scheme 3),
which were further alkylated with different allyl and benzyl
bromides in the presence of K2CO3 in acetonitrile (Scheme 4),
thus yielding the BOC-protected inhibitor precursors of type 5.
The final deprotection was carried out under nonaqueous acidic
conditions using 2 M HCl in Et2O, furnishing the inhibitors 6
as their hydrochlorides in high yields (Scheme 5; yields are
given in Table 1). In compounds with double lettering codes,
the first letter refers to the sulfonyl substituents (P2/P2′), the
second letter to the N-alkyl moieties (P1/P1′).
The carboxamido-substituted compounds 5fd, and 5fg are
accessible from the corresponding cyano-substituted derivatives
5ed and 5eg by mild hydrolysis, applying 30% H2O2 in DMSO

2080 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 7
Blum et al.
Scheme 6a
Table 1. Ki Values of the Inhibitors 6 toward the HIV Protease and
Yield of the Terminal Deprotection Step
a (a) 30% aqueous H2O2, K2CO3, DMSO, 0 °C to room temp. (5fd: R )
Ph, 92%. 5fg: R ) p-CF3-Ph, 71%.)
Scheme 7a
(Scheme 6).23 The amino-substituted inhibitor 6gd was obtained
from the corresponding nitro-substituted precursor 5dd by
reduction with SnCl2 in EtOAc at elevated temperature.24 Under
these reaction conditions, the BOC protecting group was
removed simultaneously (Scheme 7). The synthesis of inhibitors
of type 6 was achieved in seven or eight steps, respectively,
starting from D-(-)-tartaric acid in an overall yield ranging from
7% to 21% with an average yield of more than 70% for each
step.
a (a) SnCl2 ·2 H2O, EtOAc, 2 h reflux, then 2 M HCl in Et2O 79%.
Results and Discussion
To gain the first insight into the structure–activity-relationship
(SAR), the benzene sulfonamide group in 4a was alkylated with
three differently alkyl-substituted alkenyl moieties and a benzyl
group using allyl bromide (to 5aa), 2-methylallyl bromide (to
5ab), 3,3-dimethylallyl bromide (to 5ac), and benzyl bromide
(to 5ad). After acidic deprotection of the BOC group, inhibitors
6aa-ad were obtained. All four bis-substituted sulfonamides
showed affinity in the micromolar range against HIV protease,
from which those inhibitors with the largest moieties (6ac and
6ad) exhibited the highest affinity (Table 1).
Figure 2. Schematic representation of the binding mode of 6ad
observed in the crystal structure in complex with the HIV protease.
Bond lengths to I50A and I50B of the flap region are given in Å.
structure of pyrrolidine-based inhibitor 1, the protonated cyclic
amino nitrogen is found at its pivotal position forming a
hydrogen bond network to the catalytic aspartic acid residues
D25A (2.6, 3.0 Å) and D25B (2.9, 3.0 Å). Instead of a structural
water molecule usually present in peptidomimetic or substrate-
The X-ray structure of 6ad in complex with the protease was
determined with a resolution of 1.55 Å and surprisingly revealed
an asymmetric binding mode (Figure 2). Similar to the complex

Pyrrolidines as HIV Protease Inhibitors
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 7 2081
Table 2. Conserved Hydrogen Bonds of the Pyrrolidine Nitrogen Atom
to the Catalytic Dyad (D25A/D25B) and the Sulfonyl Oxygen Atoms to
the Flap (I50A/I50B) Observed in the Determined Cocrystal Structuresa
6ad
6af
6cd
6fd
6gd
6fg
D25A Oδ1
D25A Oδ2
D25B Oδ1
D25B Oδ2
I50A N
3.0
2.6
3.0
2.9
2.9
3.3
2.9
2.7
2.8
3.1
3.0
3.1
3.0
2.5
2.8
2.9
3.1
3.6
3.0
2.7
2.8
2.8
3.0
3.2
3.0
2.6
2.8
2.9
3.1
3.7
2.9
2.7
2.8
3.0
2.9
3.1
I50B N
a Values are given in Å.
where for this purpose small hydrophobic substituents were
selected in order to fill remaining space flanked by V32B and
I84B in the S2-pocket as well as V32A and I84A in the S2′-
pocket; (C) para substitution at the P2/P2′-phenyl moieties with
substituents capable of forming hydrogen bonds to D29A and
D30A in the S2′-pocket and D29B and D30B in the S2-pocket.
All three strategies were pursued, and the resulting inhibitors
6ae-ag and 6bd-gd were subsequently analyzed by kinetic
measurements with respect to their affinity against the target
enzyme. The cocrystal structures of selected representatives, at
least one of each modification type, in complex with HIV
protease were determined. This approach allows the detailed
analysis of additional interactions formed by the introduced
substituents and provides an extensive understanding of the
principles accounting for the SAR. Subsequently, this knowledge
should facilitate the selection of the most promising combination
of substituents, thus leading to an even further improved
inhibitor.
Figure 3. Crystal structure of 6ad (green, coded by atom type) in
complex with HIV protease. The protein backbone trace is schematically
illustrated in wheat, the catalytic D25A and D25B as well as I50A and
I50B of the flaps are displayed in gray, coded by atom type. The Fo -
Fc density for the ligand is displayed at a σ level of 3.0 as blue mesh.
(A) Elongation of the P1-benzyl ring in the para position with
hydrophobic residues was easily achieved by alkylation of
benzene sulfonamide 4a with appropriately substituted benzyl
bromides, yielding after deprotection the corresponding bromo-
(6ae), iodo- (6af), and trifluoromethyl- (6ag) substituted inhibi-
tors. These inhibitors indeed showed improved affinity by a
factor of 5 compared to 6ad for the halide substituted ones (6ae,
0.46 µM; 6af, 0.39 µM). In the case of the trifluoromethyl-
substituted inhibitor 6ag, a factor of 3 (0.80 µM) is achieved.
The crystal structure of 6af in complex with HIV protease was
determined with a resolution of 1.41 Å and revealed a binding
mode resembling that of the unsubstituted pyrrolidine 6ad. A
similar hydrogen-bond network of the core structure can be
observed (Table 2). The root-mean-square deviation (rmsd)
between the CR atoms of the complexes of 6ad and 6af is 0.15
Å, resembling the overall similar binding mode. This similarity
is also reflected by an rmsd of 0.45 Å from the respective 6ad
substructure in 6af. The gain in affinity is presumably due to
additional vdW interactions in the S1-subsite with P81A, G48B,
and G49B and in the S1′-subsite with R8B. Obviously, the
position and orientation of the P2/P2′-moiety are not affected
by the additional substituent.
(B) The extension of the P2/P2′-phenyl ring in the ortho
position with small hydrophobic substituents was successfully
accomplished by synthesis of the o-methyl and o-chloro
substituted benzene sulfonamides 4b and 4c, respectively, which
were subsequently alkylated with benzyl bromide yielding after
deprotection the methyl- (6bd) and chloro- (6cd) substituted
inhibitors. Compared to the lead structure 6ad, a 3-fold increase
in affinity was observed (6bd, 0.67 µM; 6cd, 0.77 µM). The
2.30 Å resolved crystal structure of the chloro derivative 6cd
revealed a similar binding mode compared to that of 6ad. The
protein structure remains unaffected (CR rmsd to 6ad is 0.14
Å); however, the binding mode of 6cd slightly deviates from
the expected conformation (rmsd to 6ad substructure is 0.67
Figure 4. Overview of the optimization strategies of 6ad by attachment
of additional substituents. Protein is shown in wheat surface representa-
tion. D29 and D30 surface patches are shown in color, coded by atom
type: (A) hydrophobic P1/P1′ elongation; (B) small hydrophobic P2/P2′
ortho substitution; (C) polar P2/P2′ para substitution to address D29
and D30.
like complexes, one sulfonyl group establishes two hydrogen
bonds to the backbone NHs of I50A (2.9 Å) and I50B (3.3 Å),
each with one sulfonyl oxygen atom. The second sulfonyl group,
however, does not form any polar interactions (Figure 3). The
benzyl substituents occupy the S1/S1′-subpockets, establishing
van der Waals (vdW) contacts with L23A, I50B, V82A, I84A
(S1) and R8B, L23B, I50A, P81B, V82B, I84B (S1′). The phenyl
moieties reside in the S2/S2′-subsites forming vdW contacts with
A28B, V32B, I50A, I84B (S2) and A28A, D30A, V32A, I47A,
I50B (S2′). Additional vdW contacts are found to G27, G48,
and G49 of each subunit. The deviation from a C2-symmetrical
occupation of the subpockets can be analyzed by superposi-
tioning the observed geometry of the inhibitor with its geometry
resulting from a rotation around the protein’s C2-axis. The
average distance of the corresponding ring atoms is 1.7 Å in
the S1- and 2.1 Å in the S2-pocket.
This crystal structure enabled us to rationally design the
second series of most likely more potent inhibitors. Conse-
quently, 6ad was selected as lead structure for further optimiza-
tion. An in-depth analysis of the crystal structure revealed three
very promising symmetric substitution patterns for further lead
optimization (Figure 4): (A) elongation of the P1/P1′-benzyl
moieties with hydrophobic substituents in para position to
address unoccupied space in the S1/S1′-pocket; (B) ortho
substitution at the P2/P2′-phenyl ring systems to improve the
shape match between the binding pocket surface and the ligand,

2082 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 7
Blum et al.
Figure 6. Schematic representation of the hydrogen bond network
observed in the crystal structure of 6fd. Bond lengths are given in Å.
Figure 5. Schematic representation of the hydrogen bond network
observed in the crystal structure of 6gd. Bond lengths are given in Å.
Table 3. Calculated rmsd after Alignment of the Protein CR Atomsa
Å). The observed polar interactions are in agreement with the
structure of 6ad (Table 2); furthermore, the o-chloro substitution
is as intended in additional vdW contacts of the chlorine atom
to V32B and I84B in the S2′-pocket. Moreover, the ortho
substituent induces an unanticipated change in the orientation
of the phenyl moiety in the S2-pocket. There, the chloro
substituent points in the opposite direction toward the S1-pocket
now interacting with G48B and G49B, hence retaining the
orientation of 6ad.
6ad
6af
6cd
6fd
6gd
6fg
Protein CR
6ad
6af
6cd
6fd
6gd
6fg
0.15
0.14
0.17
0.25
0.23
0.28
0.12
0.19
0.15
0.23
0.19
0.18
0.20
0.23
0.20
0.15
0.14
0.25
0.12
0.19
0.17
0.23
0.19
0.18
0.28
0.15
0.20
0.23
0.23
0.20
6ad Substructure
(C) For the decoration of the P2/P2′-phenyl moieties in the
para position with polar groups, the 4-nitrophenylsulfonamide
4d and the 4-cyanophenylsulfonamide 4e were synthesized and
subsequently alkylated with benzyl bromide. The alkylated nitro
compound 5dd was on the one hand directly deprotected to yield
inhibitor 6dd. On the other hand, the nitro group in 5dd was
first reduced to the corresponding amino function, concurrently
removing the protecting group now giving rise to the amino-
substituted inhibitor 6gd. The alkylated cyano compound 5ed
was converted into the carboxamide 5fd and subsequently
deprotected yielding inhibitor 6fd. Compared with the unsub-
stituted pyrrolidine 6ad, the nitro-substituted inhibitor 6dd only
showed a comparable affinity (6dd, 1.72 µM), whereas those
inhibitors with hydrogen bond donor substituents revealed an
8-fold increase in affinity (6fd, 0.26 µM; 6gd, 0.27 µM). To
elucidate the hydrogen bond network established by the polar
substituents, both inhibitors were crystallized in complex with
HIV protease and the resulting crystal structures were deter-
mined (resolution: 6fd, 1.50 Å; 6gd, 1.78 Å). Both crystal
structures show a high similarity to that of 6ad with respect to
the CR atoms of the protein (CR rmsd 6fd, 0.25 Å; 6gd, 0.12
Å) as well as the binding mode of the ligand (rmsd of 6ad
substructure 6fd, 0.54 Å; 6gd, 0.36 Å). The polar interactions
of the central scaffold are similar to those of 6ad (Table 2).
Because of the asymmetric binding mode, the interaction
patterns in the S2- and S2′-subpockets of each ligand differ from
each other (Figures 5 and 6). The p-amino derivative 6gd
establishes two hydrogen bonds in the S2-pocket, one directly
to the backbone carbonyl of D30B (3.2 Å) and one to the side
chain of D30B bridged by a water molecule (2.8, 2.7 Å). In the
S2′-pocket, two hydrogen bonds are also formed, however, now
to the side chains of D30A (2.7 Å) and D29A again mediated
by an interstitial water molecule (3.2, 2.9 Å). The same number
of hydrogen bonds can be observed for the p-carboxamido
substituted inhibitor 6fd. In the S2-pocket the backbone NH of
D30B is addressed via the carboxamide oxygen (2.8 Å), whereas
6ad
6af
6cd
6fd
6gd
6fg
0.45
0.67
0.56
0.54
0.66
0.85
0.36
0.57
0.72
0.28
0.69
0.61
0.81
0.32
0.48
0.45
0.67
0.54
0.36
0.69
0.56
0.66
0.57
0.61
0.85
0.72
0.81
0.28
0.32
0.48
a For the ligands the respective 6ad substructure within each ligand is
compared after CR-alignment of the protein structure. Values are given in
Å.
the amide nitrogen establishes a hydrogen bond to the side chain
of D30B (2.9 Å). In the S2′-pocket a water molecule mediates
the interaction between the ligand amido nitrogen and the D29A
side chain (3.0, 2.7 Å). The carbonyl function of the ligand
accepts a hydrogen bond from the backbone NH of D30A (3.0
Å). The position of the P1/P1′-residues in both ligand structures
is not affected by these additional interactions in the S2- and
S2′-pockets.
All crystal structures analyzed at this stage of the project are
very similar; the rmsd between the CR atoms of any two
complexes is less than 0.28 Å, and all inhibitors also show a
very high degree of similarity in their binding modes (Table
3). The hydrogen-bond network of the lead structure 6ad is
conserved in all structures (Table 2). The rmsd between the
respective substructure and 6ad is less than 0.69 Å. For the
comparison of any two ligands, the rmsd is less than 0.85 Å.
The largest differences are observed between the o-chloro-
substituted ligand (6cd) and the p-amino (6gd) or the p-
carboxamido-substituted ligand (6fd) with 0.72 and 0.85 Å,
respectively. This reflects the different binding mode of 6cd
(Table 3). This different binding mode makes an ortho substitu-
tion at the P2-residue less attractive for further combination with
other modifications (Figure 7), so only the two remaining
substitution strategies were combined. As substituents for the
combined inhibitor, a trifluoromethyl group such as P1/P1′ and
a carboxamido moiety such as the P2/P2′ substituent were
selected.

Pyrrolidines as HIV Protease Inhibitors
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 7 2083
inhibitor 6fg. On the basis of our developed synthetic strategy,
the synthesis of the enantiopure key intermediate 3 commencing
with D-(-)-tartaric acid is straightforward and high-yielding.
Condensation of 3 with appropriately chosen sulfonyl chlorides
renders the corresponding sulfonamides, alkylation of which,
followed by further functional group transformations and
deprotection, gave rise to the desired inhibitors. These are
obtained via a seven- or eight-step synthesis with an overall
yield ranging from 7% to 21%. Following the outlined synthetic
route, the synthesis of a plethora of putative protease inhibitors
is readily feasible. Initial SAR studies as well as the crystal
structure determination resulted in selection of 6ad as the lead
compound for further structural optimization. The analysis of
the cocrystal structure of 6ad revealed three possible strategies
for optimization via symmetric introduction of substituents to
the original P1/P1′- and P2/P2′-phenyl moieties. From each class
of possible modifications, at least one representative was
synthesized and thoroughly analyzed by crystal structure de-
termination of the protein–ligand complex. The observed
interactions of the core structure are highly conserved throughout
this series of inhibitors. These structures provided deeper insights
into the protein–ligand interactions and the underlying principles
of the SAR. Taking this information into account, the most
promising combination of P1/P1′- and P2/P2′-moieties was
selected and the resulting inhibitor 6fg indeed showed the
expected improvement in affinity with Ki ) 74 nM. The
cocrystal structure of this inhibitor confirmed the successful
application of our optimization strategy. The complete structural
characterization of crucial intermediates prevents misdirection
by only taking the affinity of the compounds into account. This
project clearly shows that the rational design of inhibitors based
on the successful cooperation between synthetic medicinal
chemistry and structural biology can lead to a highly efficient
optimization of lead structures.
Figure 7. Superposition of the ligand conformations observed in the
corresponding crystal structures aligned by CR-fit of the proteins.
Ligands are color-coded by atom type: 6af (cyan), 6cd (orange), 6fd
(salmon), 6gd (gray).
Figure 8. Crystal structure of 6fg (cyan, coded by atom type) in
complex with HIV protease. The protein backbone trace is schematically
illustrated in wheat. The catalytic D25A/D25B, D30A/D30B, and I50A/
I50B of the flaps are displayed in gray, coded by atom type. The Fo -
Fc density for the ligand is displayed at a σ level of 3.0 as blue mesh.
The combined inhibitor 6fg was prepared in similar manner
as already described for carboxamide inhibitor 6fd starting from
sulfonamide 4e, followed by alkylation to 5eg and subsequent
hydrolysis of the nitriles to the carboxamide 5fg. After acidic,
nonaqueous deprotection, the inhibitor 6fg was tested for its
affinity against HIV protease. With 74 nM, the trifluoromethyl
and carboxamido-substituted inhibitor 6fg exhibits the highest
affinity within this series. To validate our combination strategy
pursued within this project, the crystal structure of 6fg in
complex with HIV protease was determined with a resolution
of 1.48 Å (Figure 8). No major differences could be observed
comparing the protein structure to the formerly determined
complexes (CR rmsd to 6af, 0.18 Å; to 6fd, 0.23 Å). Our
successful combination strategy is also represented in the
conserved ligand conformation (rmsd of 6ad substructure to 6af,
0.61 Å; 6fd, 0.32 Å). The polar interaction pattern of the
carboxamido-substituted inhibitor 6fd is retained. Compared to
6af, additional vdW contacts of the trifluoromethyl group to
R8A in the S1-subsite are observed. In the structure of 6fg, both
CF3 groups show disorder of a rigid rotor. They were refined
as double conformations.
Experimental Section
Kinetic Assay. Inhibition data for HIV protease were determined
as follows: IC50 values were taken from plots of Vi/V0 versus
inhibitor concentration, in which Vi is the velocity in the presence
of an inhitor and V0 is the velocity in the absence of an inhibitor.
The fluorogenic substrate Abz-Thr-Ile-Nle-(p-NO2-Phe)-Gln-Arg-
NH2 was purchased from Bachem. Recombinant HIV protease was
expressed from Escherichia coli and purified as previously de-
scribed.25 Enzymatic assays were performed in 172 µL of assay
buffer (100 mM MES, 300 mM KCl, 5 mM EDTA, 1 mg/mL BSA,
pH 5.5) by the addition of substrate dissolved in 4 µL of DMSO,
distinct inhibitor concentrations dissolved in 4 µL of DMSO, and
20 µL of HIV-1 protease in assay buffer to a final volume of 200
µL (final DMSO concentration of 4%). The hydrolysis of the
substrate was recorded as the increase in fluorescence intensity
(excitation wavelength 337 nm, emission wavelength 410 nm) over
10 min, during which the signal increased linearly with time.26 The
kinetic constants for HIV protease (Km ) 14.6 µM) were determined
by the method of Lineweaver and Burk with a HIV protease
concentration of 2.8 nM with varied substrate concentrations. The
active-site concentration was quantified by titrating three different
HIV protease concentrations with the strong binding inhibitor
saquinavir (Ki ) 0.3 nM). Ki values were calculated from the
following equation: Ki ) [IC50 - (Et/2)][1 + (S/Km)]-1
Summary and Conclusion
Crystallization of HIV Protease Inhibitor Complexes. HIV
protease inhibitor complexes were crystallized at 18 °C in 0.1 M
BisTris, pH 6.5, 2.0–3.5 M NaCl, and a protein concentration of 7
mg/mL in the space group P21212 (crystal data, Table 4). The
crystals were obtained by cocrystallization of the enzyme with
inhibitor concentrations ranging from 20- to 100-fold the Ki value.
Crystals were further optimized using streak-seeding techniques.
C2-symmetric 3,4-disubstituted pyrrolidines have been de-
veloped as a new class of HIV protease inhibitors. Starting from
the initial lead 6ad, which showed affinity in the low micromolar
range, the activity of this new class of HIV protease inhibitors
could be significantly optimized by means of rational structure-
based design up to the two digit nanomolar range for the final

2084 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 7
Table 4. Crystallographic Data
Blum et al.
6ad, 2PQZ
6af,b 2QNP
6cd, 2QNQ
6fd, 2PWR
6gd, 2PWC
6fg, 2QNN
resolution (Å)
space group
25–1.55
P21212
25–1.41
P21212
30–2.30
P21212
25–1.50
P21212
25–1.78
P21212
25–1.48
P21212
cell dimensions (Å)
a ) 57.5
b ) 86.0
c ) 46.7
1.58–1.55
96.3 [87.3]
13.4 [3.0]
8.6 [33.5]
10–1.55
a ) 57.4
b ) 85.9
c ) 46.7
1.43–1.41
99.5 [94.1]
21.9 [2.83]
5.0 [33.4]
10–1.41
a ) 57.6
b ) 85.9
c ) 46.5
2.35–2.30
97.1 [93.9]
16.1 [5.27]
11.0 [25.2]
25–2.30
a ) 56.8
b ) 84.8
c ) 46.1
1.53–1.50
92.1 [93.7]
34.2 [6.6]
3.9 [23.0]
10–1.50
a ) 57.3
b ) 85.8
c ) 46.5
1.81–1.78
98.2 [99.5]
13.5 [2.9]
9.7 [46.3]
10–1.78
a ) 57.4
b ) 86.2
c ) 46.3
1.51–1.48
99.5 [96.4]
23.5 [3.71]
4.8 [28.7]
10–1.48
highest resoln shell (Å)
completeness (%)a
I/σa
Rsym (%)a
resoln in refinement (Å)
Rcryst (%), F > 4σFo;Fo
Rfree (%), F > 4σFo;Fo
16.4; 18.9
20.4; 23.4
17.6; 19.1
21.3; 22.9
n.d.; 20.5
n.d.; 23.9
16.8; 17.8
20.2; 21.6
16.3; 18.9
21.1; 24.5
17.0; 18.4
19.3; 20.9
a Values in brackets refer to the highest resolution shell. b In the case of 6af, the ligand occupancy was refined to 63% with respect to the protein. For
additional details, refer to Supporting Information.
For cryoprotection the crystals were briefly soaked in mother liquor
containing 25% glycerol.
Tetrahydrofuran was dried over lithium aluminum hydride and
freshly distilled prior to use; other anhydrous solvents were
purchased in Sure/Seal bottles from Sigma-Aldrich or Fluka. All
moisture sensitive reactions were carried out using oven-dried
glassware under a positive stream of argon. Compound 2a has been
prepared according to a literature procedure.18
General Procedure A: Arylsulfonamides. An amount of 1.0
mmol of diamine 3 was dissolved in 6 mL of CH2Cl2, and 2.4 mmol
of triethylamine or diisopropylethylamine and 2.2 mmol of aryl-
sulfonyl chloride were slowly added. The reaction mixture was
stirred for 14–20 h and poured into an aqueous saturated NH4Cl
solution (10 mL). The aqueous phase was separated and extracted
with MTBE (3 × 10 mL). The combined organic phases were
washed with brine (3 × 10 mL), dried over Na2SO4, filtered, and
evaporated. The residual oil was dissolved in 5 mL of CH2Cl2 and
added dropwise to 50 mL of vigorously stirred cold hexane. The
precipitate was filtered off, washed with hexane, and dried. The
resulting product was used in the following steps without further
purification.
General Procedure B: Alkylation of Arylsulfonamides. An
amount of 1.00 mmol of sulfonamide 4 and 3.00 mmol of anhydrous
K2CO3 were suspended in 7.5 mL of dry CH3CN. An amount of
2.40 mmol of benzyl bromide was added, and the mixture was
stirred at the given temperature for 4–12 h. After cooling to room
temperature, the insoluble material was filtered off and washed with
MTBE. The filtrate was evaporated, and the residual oil was purified
by flash chromatography.
General Procedure C: Conversion of Nitriles to Amides. An
amount of 0.30 mmol of nitrile 5ed or 5eg was dissolved in 3 mL
of DMSO and cooled to 0 °C. Then 1.5 mL of 30% H2O2 and 200
mg (0.15 mmol) of K2CO3 were added to the frozen mixture, which
was then allowed to warm to room temperature and stirred for an
additional 30 min. The solution was cooled to 0 °C, and an amount
of 50 mL of H2O was added. The precipitated product was filtered
off and washed with H2O and dissolved in 5 mL of CHCl3. The
solution was washed with H2O (3 × 10 mL) and brine (10 mL),
dried over MgSO4, filtered, and evaporated.
Data Collection, Phasing, and Refinement. The data sets were
collected at the synchrotron BESSY II in Berlin/Germany on PSF
beamline 14.2 (PDB codes 2PQZ for 6ad, 2QNP for 6af, 2PWC
for 6gd, 2QNN for 6fg), on the EMBL/DESY in Hamburg/Germany
on beamline X13 (PDB code 2PWR for 6fd), both equipped with
a MAR-CCD detector, and on a Rigaku R-AXIS IV image plate
detector using Cu KR radiation from an in-house Rigaku rotating
anode (PDB code 2QNQ for 6cd). Data were processed and scaled
with Denzo and Scalepack as implemented in HKL2000.27 The
structures were determined by the molecular replacement method
with Phaser;28 one monomer of the 1.5 Å structure of the HIV-1
protease in complex with a pyrrolidine based inhibitor (PDB code
1XL2), and consecutively a monomer of the determined structure
(PDB code 2PQZ for 6ad), was used as the search model. The
structure (PDB code 2QNP for 6af) was determined by single
isomorphous replacement with anomalous scattering method (SIR-
AS) using the program implemented in the HKL2MAP graphics
user interface;29 the data set of the unsubstituted pyrrolidine was
used as native data. The positions of the iodine atoms were identified
using SHELXD,30 and the phase problem was solved with
SHELXE.31 The resulting phases were applied, and after density
modification with DM,32 ARPwARP33 as implemented in CCP434
was used for automated model building.
Refinement was continued with CNS35 and SHELXL-97.36 For
each refinement step at least 10 cycles of conjugate minimization
were performed, with restraints on bond distances, angles, and
B-values. Intermittent cycles of model building were done with the
program COOT.37 The coordinates have been deposited in the PDB
(http://www.rcsb.org/pdb/) with the following access codes: 6ad,
2PQZ; 6af, 2QNP; 6cd, 2QNQ; 6fd, 2PWR; 6gd, 2PWC; 6fg,
2QNN.
General Synthetic Chemistry. Reported yields refer to the
analytical pure product obtained by column chromatography or
recrystallization. All proton and carbon nuclear magnetic resonance
spectra were recorded on a Jeol Eclipse+ spectrometer (1H NMR,
500 MHz; 13C NMR, 125 MHz) using TMS as internal standard
(0.0 ppm). 13C spectra were referenced to CDCl3 (77.16 ppm),
DMSO-d6 (39.52 ppm), or MeOH-d4 (49.00 ppm). The values of
chemical shifts (δ) are given in ppm, and coupling constants (J )
are given in Hz. Abbreviations are as follows: br ) broad, s )
singulet, d ) doublet, t ) triplet, q ) quartet, sm ) symmetric
multiplet, m )multiplet. Mass spectra were obtained from a double
focusing sectorfield Micromass VG-Autospec spectrometer. Com-
bustion analyses were determined on a Vario Micro Cube by
Elementar Analysen GmbH, and the results indicated by symbols
for the elements were within (0.4% of theoretical values. Melting
points were determined using a Leitz HM-Lux apparatus and are
uncorrected. Flash chromatography was performed using silica gel
60 (0.04–0.063 mm) purchased from Macherey-Nagel, and solvents
are indicated. TLC was carried out using 0.2 mm aluminum plates
coated with silica gel 60 F254 by Merck and visualized by UV
detection. Solvents and reagents that are commercially available
were used without further purification unless otherwise noted.
General Procedure D: BOC Deprotection. An amount of 0.20
mmol of BOC-protected pyrrolidine was dissolved in 5 mL of 2
M HCl in Et2O. After 14 h the solution was decanted from the
precipitated hydrochloride, which was washed thoroughly with dry
ether and dried in vacuum.
(3R,4R)-1-BOC-3,4-dihydroxypyrrolidine (2b). An amount of
6.66 g (34.8 mmol) of N-benzylpyrrolidine18 2a was dissolved in
100 mL of MeOH followed by the addition of 8.36 g (38.3 mmol)
of BOC2O and 666 mg of Pd/C. After the reaction mixture was
stirred under H2 atmosphere for 16 h, the catalyst was removed by
filtration through a pad of Celite and the filtrate was concentrated.
The residual yellow oil was dissolved in 25 mL of hot EtOAc, and
the product crystallized upon cooling. The crystals were collected
by filtration, washed with cold EtOAc, and dried, yielding 4.50 g
of the product 2b. After concentration of the mother liquor, an
additional 0.84 g was obtained resulting in an overall yield of 5.34 g
(74%) of 2b. Beige needles; mp 161 °C; 1H NMR (CDCl3) δ 3.94
(tbr, 2H, J ) 3.3), 3.67 (dd, 1H, J ) 12.0, 4.9), 3.63 (dd, 1H, J )

Pyrrolidines as HIV Protease Inhibitors
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 7 2085
11.9, 5.0), 3.42 (dd, 2H, J ) 12.1, 2.3), 3.37 (dd, 2H, J ) 12.1,
2.3), 1.44 (s, 9H); 13C NMR (CDCl3) δ 153.9, 80.5, 64.2, 63.4,
48.8, 48.5, 28.4; MS (ESI) m/z 204 (25, M + H), 221 (5, M +
NH4), 407 (78, 2M + H), 424 (100, 2M + NH4); HRMS (ESI)
m/z calcd for C18H34N2O8 ·NH4 424.265 891, found 424.264 370.
Anal. (C9H17NO4) C, H, N.
(ESI) m/z 594 (97, M + Na), 1165 (100, 2M + Na), 1736 (72, 3M
+ Na). Anal. (C21H25N5O10S2) H, N. C: calcd, 44.13; found, 43.72.
(3S,4S)-1-BOC-3,4-bis(4-cyanobenzenesulfonylamino)pyrrol-
idine (4e). Following general procedure A, employment of 402 mg
(2.00 mmol) of diamine 3, 887 mg (4.40 mmol) of 4-cyanoben-
zenesulfonyl chloride, and 641 µL (4.40 mmol) of diisoproylethy-
lamine in 12 mL of CH2Cl2 (12 h) furnished sulfonamide 4e (759
mg, 71%). Colorless needles; mp 139 °C; 1H NMR (DMSO-d6) δ
8.39 (sbr, 2H), 8.08 (d, 4H, J ) 8.3), 7.96–7.91 (m, 4H), 3.60 (sm,
2H), 3.28–3.18 (m, 2H), 2.96–2.86 (m, 2H) 1.34 (s, 9H); 13C NMR
(DMSO-d6) δ 152.9, 144.8, 133.3, 127.2, 127.1, 117.6, 115.1, 78.8,
56.6, 55.9, 48.4, 47.9, 27.9; MS (ESI) m/z 554 (100, M + Na),
1085 (13, 2M + Na). Anal. (C23H25N5O6S2) C, H, N.
(3S,4S)-1-BOC-3,4-bis(benzenesulfonylbenzylamino)-
pyrrolidine (5ad). According to general procedure B, employment
of 150 mg (0.31 mmol) of sulfonamide 4a, 89 µL (0.75 mmol) of
benzyl bromide, and 129 mg (0.93 mmol) K2CO3 in 2.5 mL of
CH3CN (7 h reflux) yielded, after purification by flash chromatog-
raphy (hexane/MTBE, 1:1), the N-alkylated sulfonamide 5ad (156
mg, 75%). Colorless crystals; mp 76 °C; 1H NMR (CDCl3) δ 7.92
(sbr, 2H), 7.73 (sbr, 4H), 7.67–7.45 (mbr, 6H), 7.34–7.02 (mbr,
10H), 4.79 (sbr, 1H), 4.45 (d, 2H, J ) 15.4), 4.30 (sbr, 1H),
4.00–3.89 (m, 1H), 3.81–3.72 (m, 1H), 3.29–3.17 (mbr, 1H),
3.11–3.01 (mbr, 1H), 1.37 (s, 9H); 13C NMR (CDCl3) δ 153.6,
140.0, 136.4, 133.1, 129.4, 128.8, 128.2, 128.1, 127.7, 127.3, 80.2,
61.8, 60.2, 50.0, 48.9, 47.6, 46.1, 28.4; MS (ESI) m/z 684 (100, M
+ Na), 1345 (85, 2M + Na). Anal. (C35H39N3O6S2) C, H, N.
(3S,4S)-1-BOC-3,4-bis[benzenesulfonyl-(4-iodobenzyl)ami-
no]pyrrolidine (5af). Following general procedure B, employment
of 482 mg (1.00 mmol) of sulfonamide 4a, 713 mg (2.40 mmol)
of 4-iodobenzyl bromide, and 415 mg (3.00 mmol) of K2CO3 in
7.5 mL of CH3CN (7 h reflux) furnished, after purification by flash
chromatography (hexane/MTBE, 2:1), the N-alkylated sulfonamide
5af (690 mg, 76%). Light-yellow crystals; mp 98 °C; 1H NMR
(CDCl3) δ 7.92–7.43 (mbr, 14H), 6.98–6.74 (mbr, 4H), 4.69 (sbr,
1H), 4.38 (d, 2H, J ) 16.0), 4.20 (sbr, 1H), 4.02–3.90 (mbr, 1H),
3.78–3.66 (mbr, 1H), 3.33–3.09 (mbr, 2H), 3.05–2.95 (mbr, 2H),
1.39 (s, 9H); 13C NMR (CDCl3) δ 153.5, 139.6, 137.9, 136.0, 133.3,
130.0, 129.6, 127.6, 127.2, 93.8, 80.6, 61.9, 60.5, 49.4, 48.4, 47.5,
46.2, 28.5; MS (ESI) m/z 935 (100, M + Na), 1848 (70, 2M +
Na). Anal. (C35H37I2N3O6S2) C, H, N.
(3S,4S)-1-BOC-3,4-diaminopyrrolidine (3). An amount of
2.64 g (13.04 mmol) of diol 2b was suspended in 75 mL of CH2Cl2.
Then a totalof 2.85 mL (35.20 mmol) of dry pyridine was added
and the mixture was cooled to -80 °C. An amount of 5 mL of
Tf2O was slowly added and the reaction mixture was allowed to
warm to -10 °C within 2 h. The CH2Cl2 was carefully removed
by evaporation at room temperature (heating can lead to decom-
position of the labile triflate). Then 23.65 g (364 mmol) of vaccuum-
dried NaN3 and 50 mL of dry DMPU were added to the residual
solid. After the suspension was stirred at room temperature for 3.5 h,
a total of 100 mL of H2O was added. The solution was extracted
with MTBE (5 × 50 mL), and the combined organic extracts were
washed with H2O (3 × 50 mL) and brine (50 mL), dried over
MgSO4, and filtered. To avoid isolation of the explosive diazide, a
total of 6 g silica gel was added and the solvent was carefully
removed at room temperature. The diazide was purified by flash
chromatography (hexane/EtOAc, 5:1). The azide containing frac-
tions were combined, a total of 250 mg Pd/C was added, and the
mixture was stirred under H2 atmosphere for 14 h. The catalyst
was removed by filtration over Celite, and evaporation of the filtrate
yielded the diamine 3 (2.25 g, 86%), which was used without further
purification. Colorless oil; 1H NMR (CDCl3) δ 3.77–3.46 (m, 2H),
3.10–2.96 (m, 4H), 1.57 (sbr, 4H), 1.46 (s, 9H); 13C NMR (CDCl3)
δ 154.4, 79.3, 58.6, 58.1, 52.5, 52.1, 28.4; MS (ESI) m/z 202 (18,
M + H), 224 (68, M + Na), 403 (84, 2M + H), 425 (100, 2M +
Na), 604 (37, 3M + H), 626 (89, 3M + Na); HRMS (ESI) m/z
calcd for C9H20N3O2 202.155 552, found 202.156 485. Anal.
(C9H19N3O2) C, H. N: calcd, 20.88; found, 20.19.
(3S,4S)-1-BOC-3,4-bis-benzenesulfonylaminopyrrolidine
(4a). Following general procedure A, utilization of 183 mg (0.89
mmol) of diamine 3, 256 µL (2.0 mmol) of benzenesulfonyl
chloride, and 340 µL (2.44 mmol) of triethylamine in 5 mL of
CH2Cl2 (22 h) yielded sulfonamide 4a (368 mg, 86%). Colorless
needles; mp 111 °C; 1H NMR (DMSO-d6) δ 8.07 (d, 2H, J ) 6.9),
7.64 (t, 4H, J ) 7.0), 7.66 (tt, 2H, J ) 7.5, 1.5), 7.59 (t, 4H, J )
7.7), 3.52 (mul, 2H), 3.18 (dd, 2H, J ) 11.5, 5.6), 2.91 (dd, 2H, J
) 11.2, 3.9), 1.32 (s, 9H); 13C NMR (DMSO-d6) δ 153.0, 140.7,
132.5, 129.2, 126.4, 78.6, 56.6, 55.8, 48.6, 48.2, 28.0; MS (ESI)
m/z 504 (92, M + Na), 985 (100, 2M + Na). Anal. (C21H27N3O6S2)
C, H, N.
(3S,4S)-1-BOC-3,4-bis[benzyl-(2-chlorobenzenesulfonyl)ami-
no]pyrrolidine (5cd). Following general procedure B, utilization of
550 mg (1.00 mmol) of sulfonamide 4c, 285 µL (2.40 mmol) of benzyl
bromide, and 415 mg (3.00 mmol) K2CO3 in 7.5 mL of CH3CN (7 h
reflux) furnished, after purification by flash chromatography (hexane/
EtOAc, 2:1), the N-alkylated sulfonamide 5cd (477 mg, 65%). Beige
(3S,4S)-1-BOC-3,4-bis-(2-chlorobenzenesulfonylamino)pyr-
rolidine (4c). Following general procedure A, utilization of 418
mg (2.08 mmol) of diamine 3, 623 µL (4.57 mmol) of 2-chlo-
robenzenesulfonyl chloride, and 659 µL (4.98 mmol) of triethy-
lamine in 10 mL of CH2Cl2 (60 h) gave rise to sulfonamide 4c
(1075 mg, 94%). Colorless needles; mp 115 °C; 1H NMR (CDCl3)
δ 8.10 (dbr, 1H, J ) 7.6), 8.04 (dbr, 1H, J ) 7.6), 7.59–7.49 (m,
4H), 7.48–7.41 (m, 2H), 5.92 (sbr, 1H), 5.76 (sbr, 1H), 3.65–3.52
1
crystals; mp 86 °C; H NMR (CDCl3) δ 7.81 (sbr, 1H), 7.65 (sbr,
1H), 7.54–7.29 (m, 4H), 7.25–6.97 (m, 12H), 4.94 (sbr, 1H), 4.70–4.51
(mbr, 3H), 4.48–4.37 (mbr, 1H), 4.26–4.17 (mbr, 1H), 3.71–3.59 (mbr,
1H), 3.46–3.35 (mbr, 1H), 3.28–3.17 (mbr, 1H), 3.13–3.03 (mbr, 1H),
1.40 (s, 9H); 13C NMR (CDCl3) δ 153.7, 138.3, 136.3, 135.6, 133.5,
132.1, 131.9, 131.8, 131.3, 128.5, 128.2, 127.8, 127.0, 80.3, 60.0, 58.9,
49.2, 48.4, 47.7, 47.1, 28.5; MS (ESI): m/z 752, 754, 757 (100, 87, 27
M + Na), 1481, 1483, 1485 (24, 41, 28 2M + Na). Anal.
(C35H37Cl2N3O6S2) C, H, N.
(m, 3H), 3.50–3.41 (m, 1H), 3.14–3.06 (m, 2H), 1.36 (s, 9H); 13
C
NMR (CDCl3) δ 153.9, 136.7, 134.4, 134.3, 131.9, 131.7, 131.5,
127.5, 80.5, 57.3, 56.3, 49.4, 48.1, 28.4; MS (ESI) m/z 572, 574,
578 (56, 46, 11, M + Na), 1121, 1123, 1125 (63, 100, 67, 2 M +
Na). Anal. (C21H25Cl2N3O6S2) C, H, N.
(3S,4S)-1-BOC-3,4-bis[benzyl-(4-nitrobenzenesulfonyl)ami-
no]pyrrolidine (5dd). Following the general procedure B, usage of
750 mg (1.31 mmol) of sulfonamide 4d, 343 µl (2.89 mmol) of benzyl
bromide, and 544 mg (3.94 mmol) of K2CO3, and a catalytic amount
of KI in 10 mL of CH3CN (4 h reflux) gave, after purification by
flash chromatography (hexane/EtOAc, 2:1), the N-alkylated sulfona-
(3S,4S)-1-BOC-3,4-bis(4-nitrobenzenesulfonylamino)pyrrol-
idine (4d). According to general procedure A, utilization of 425
mg (2.10 mmol) of diamine 3, 1.024 g (4.62 mmol) of 4-nitroben-
zenesulfonyl chloride, and 879 µL (6.30 mmol) of triethylamine
in 20 mL of CH2Cl2 (22 h) yielded sulfonamide 4d (833 mg, 69%).
Colorless needles; mp 133 °C; 1H NMR (MeOH-d4) δ 8.40 (d,
2H, J ) 8.5), 8.39 (d, 2H, J ) 8.5), 8.10 (d, 2H, J ) 8.9), 8.06 (d,
2H, J ) 8.9), 3.72 (sm, 2H), 3.44 (dd, 1H, J ) 11.5, 6.4), 3.37
(dd, 1H, J ) 11.5, 6.4), 2.98 (dd, 1H, J ) 11.5, 5.3), 2.91 (dd, 1H,
J ) 11.3, 5.3), 1.32 (s, 9H); 13C NMR (DMSO-d6) δ 152.9, 149.6,
146.4, 128.1, 128.0, 124.5, 78.8, 56.7, 55.9, 48.4, 48.0, 27.9; MS
1
mide 5dd (944 mg, 96%). Yellow crystals; mp 158 °C; H NMR
(CDCl3) δ 8.31 (sbr, 4H), 7.96 (sbr, 2H), 7.83 (sbr, 2H), 7.34–7.19
(mbr, 6H), 7.12 (sbr, 4H), 4.85 (sbr, 1H), 4.45 (sbr, 3H), 4.01 (dbr,
1H, J ) 13.3), 3.89 (dbr, 1H, J ) 14.0), 3.36 (sbr, 1H), 3.19 (sbr,
1H), 3.09 (sbr, 2H), 1.38 (s, 9H); 13C NMR (CDCl3) δ 153.4, 150.1,
145.7, 135.2, 128.9, 128.6, 128.5, 128.4, 124.5, 80.7, 61.5, 60.0, 49.9,
48.0, 47.2, 46.0, 28.3; MS (ESI) m/z 774 (66, M + Na), 1525 (100,
2M + Na). Anal. (C35H37N5O10S2) C, H, N.

2086 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 7
Blum et al.
1
(3S,4S)-1-BOC-3,4-bis[benzyl-(4-cyanobenzenesulfonyl)ami-
no]pyrrolidine (5ed). According to general procedure B, utilization
of 2.53 g (4.75 mmol) of sulfonamide 4e, 1.35 mL (11.40 mmol)
of benzyl bromide, 1.97 g (14.25 mmol) of K2CO3, and a catalytic
amount of KI in 35 mL of CH3CN (4 h reflux) yielded, after
purification by flash chromatography (hexane/EtOAc, 2:1), the
N-alkylated sulfonamide 5ed (2.26 g, 67%). Colorless crystals; mp
208 °C; 1H NMR (CDCl3) δ 7.92 (sbr, 2H), 7.85–7.69 (mbr, 6H),
7.34–7.18 (mbr, 6H), 7.17–7.02 (mbr, 4H), 4.83 (sbr, 1H), 4.41
(sbr, 3H), 4.05–3.92 (mbr, 1H), 3.89–3.77 (mbr, 1H), 3.35 (sbr,
1H), 3.16–2.96 (mbr, 3H), 1.39 (s, 9H); 13C NMR (CDCl3) δ 153.5,
144.2, 135.3, 133.1, 129.0, 128.5, 128.4, 128.2, 127.9, 117.3, 116.8,
80.8, 61.5, 59.9, 50.0, 49.1, 47.2, 46.0, 28.4; MS (ESI) m/z 734
(67, M + Na), 1445 (100, 2M + Na). Anal. (C37H37N5O6S2) C, H,
N.
81%). Beige crystals; mp 211 °C (dec); H NMR (MeOH-d4) δ
7.87 (dd, 4H, J ) 7.1, 1.1), 7.74 (t, 2H, J ) 7.4), 7.63 (t, 4H, J )
7.7), 7.63 (t, 4H, J ) 8.3), 6.89 (d, 4H, J ) 8.3), 4.68 (sm, 2H),
4.34 (d, 2H, J ) 16.0), 3.92 (d, 2H, J ) 16.3), 3.46–3.38 (m, 2H),
3.34–3.24 (m, 2H); 13C NMR (MeOH-d4) δ 140.5, 139.2, 136.9,
135.0, 131.4, 130.9, 128.8, 94.6, 62.2, 51.1, 47.5; MS (ESI) m/z
814 (100, M + H). Anal. (C30H29I2N3O4S2 ·HCl) C, H, N.
(3S,4S)-3,4-Bis[benzyl-(2-chlorobenzenesulfonyl)amino]pyr-
rolidine Hydrochloride (6cd). Following general procedure D,
employment of 138 mg (0.20 mmol) of BOC-protected pyrrolidine
5cd and 5 mL of 2 M HCl in Et2O yielded hydrochloride 6cd (104
1
mg, 78%). Beige crystals; mp 172 °C (dec); H NMR (CDCl3) δ
9.87 (sbr, 2H), 7.73 (d, 2H, J ) 7.8), 7.43 (dd, 2H, J ) 7.3, 0.9),
7.38 (td, 2H, J ) 7.6, 1.1), 7.13 (t, 2H, J ) 7.3), 7.10–7.04 (mbr,
10H), 4.97 (sm, 2H), 4.59 (d, 2H, J ) 15.8), 4.37 (d, 2H, J )
15.8), 3.49 (sbr, 2H), 3.12 (sbr, 2H); 13C NMR (CDCl3) δ 137.9,
135.3, 133.8, 132.2, 131.9, 131.6, 128.8, 128.5, 128.1, 127.2, 58.4,
49.1, 45.6; MS (ESI) m/z 630, 632, 634 (100, 90, 25, M + H).
Anal. (C30H29Cl2N3O4S2 ·HCl·H2O) C, H, N.
(3S,4S)-1-BOC-3,4-bis[(4-cyanobenzenesulfonyl)-(4-trifluoro-
methylbenzyl)amino]pyrrolidine (5eg). According to general
procedure B, usage of 448 mg (0.84 mmol) of sulfonamide 4e,
483 mg (2.02 mmol) of 4-trifluoromethylbenzyl bromide, and 348
mg (2.52 mmol) K2CO3 in 10 mL of CH3CN (7 h, reflux) furnished,
after purification by flash chromatography (hexane/EtOAc, 3:1),
the N-alkylated sulfonamide 5eg (324 mg, 45%). Colorless crystals;
(3S,4S)-3,4-Bis[benzyl-(4-carbamoylbenzenesulfonyl)amino]-
pyrrolidine Hydrochloride (6fd). Following general procedure D,
usage of 135 mg (0.18 mmol) of BOC-protected pyrrolidine 5fd
and 5 mL of 2 M HCl in Et2O gave rise to hydrochloride 6fd (100
mg, 81%). Colorless crystals; mp 176 °C (dec); 1H NMR (DMSO-
d6) δ 8.21 (sbr, 1H), 8.02 (d, 4H, J ) 8.5), 7.75 (d, 4H, J ) 8.5),
7.62 (sbr, 1H), 7.35–7.24 (mbr, 10H), 4.47 (d, 2H, J ) 16.3), 4.33
(sbr, 2H), 4.17 (d, 2H, J ) 16.3), 2.59–2.51 (m, 2H), 2.25 (dd,
2H, J ) 10.6, 4.7); 13C NMR (DMSO-d6) δ 166.3, 141.5, 137.9,
137.5, 128.3, 128.1, 127.8, 127.2, 126.9, 61.7, 48.2, 47.6; MS (ESI)
1
mp 117 °C; H NMR (CDCl3) δ 8.02–7.65 (mbr, 8H), 7.50 (sbr,
4H), 7.23 (sbr, 4H), 4.91–4.60 (mbr, 1H), 4.46 (dbr, 2H, J ) 16.0),
4.32 (sbr, 1H) 4.22–3.89 (mbr, 2H), 3.59–3.21 (mbr, 2H), 3.13 (sbr,
2H), 1.41 (s, 9H); 13C NMR (CDCl3) δ 153.5, 143.7, 139.3, 133.3,
130.9 (q, J ) 33.6), 128.6, 128.0, 125.9, 123.9 (q, J ) 272.3),
117.2, 117.0, 81.3, 62.0, 60.7, 49.7, 48.5, 47.1, 46.3, 28.4; MS (ESI)
m/z 870 (72, M + Na), 1717 (100, 2M + Na). Anal.
(C39H35F6N5O6S2) C, H, N.
m/z 648 (100,
(C32H33N5O6S2 ·HCl·0.5H2O) C, H, N.
M + H), 1295 (10, 2M + H). Anal.
(3S,4S)-1-BOC-3,4-bis[benzyl-(4-carbamoylbenzenesulfonyl)-
amino]pyrrolidine (5fd). Following general procedure C, employ-
ment of 250 mg (0.35 mmol) of nitrile 5ed, 1.6 mL of 30% H2O2,
and 233 mg (1.69 mmol) of K2CO3 in 3.2 mL of DMSO gave rise
to the amide 5fd (241 mg, 92%). Colorless crystals; mp 124 °C;
1H NMR (MeOH-d4) δ 8.08–7.95 (mbr, 4H), 7.87–7.73 (mbr, 4H),
7.34–7.13 (mbr, 10H), 4.64 (sbr, 1H), 4.59–4.49 (mbr, 3H),
3.94–3.86 (mbr, 1H), 3.82–3.74 (mbr, 1H), 3.18–3.06 (mbr, 1H),
3.03–2.81 (mbr, 3H), 1.33 (s, 9H); 13C NMR (MeOH-d4) δ 170.3,
155.0, 144.0, 139.2, 137.9, 129.8, 129.4, 129.0, 128.5, 81.6, 62.1,
61.0, 50.1, 49.9, 47.9, 47.1, 28.5; MS (ESI) m/z 770 (100, M +
Na), 1517 (40, 2M + Na). Anal. (C37H41N5O8S2 ·0.5H2O) C, H,
N.
(3S,4S)-3,4-Bis[benzyl-(4-aminobenzenesulfonyl)amino]pyr-
rolidine Trihydrochloride (6gd). A total of 608 mg (0.81 mmol)
of nitrosulfonamide 5dd was dissolved in 30 mL of EtOAc. After
addition of 1.82 g (8.1 mmol) of SnCl2 ·2H2O, the suspension was
heated to reflux for 2 h and subsequently allowed to reach room
temperature. Following the addition of 10 mL of aqueous saturated
NaHCO3 solution, the aqueous phase was separated and extracted
with EtOAc (3 × 10 mL). The combined organic phases were
washed with brine (3 × 10 mL), dried over Na2SO4, filtered, and
evaporated. The remaining solid was dissolved in 5 mL of dry Et2O,
and 5 mL of 2 M HCl in Et2O was added. After 12 h at room
temperature, the solution was decanted from the precipitate, which
was washed with Et2O and dried in vacuum, yielding trihydrochlo-
(3S,4S)-1-BOC-3,4-bis[(4-carbamoylbenzenesulfonyl)-(4-tri-
fluoromethylbenzyl)amino]pyrrolidine (5fg). Following general
procedure C, employment of 253 mg (0.30 mmol) of nitrile 5eg,
1.5 mL of 30% H2O2, and 200 mg (1.50 mmol) of K2CO3 in 3.0
mL of DMSO gave rise to amide 5fg (190 mg, 71%). Colorless
1
ride 6gd (445 mg, 79%). Yellow crystals; mp 176 °C (dec); H
NMR (MeOH-d4) δ 7.43 (d, 4H, J ) 8.3), 7.37–7.25 (m, 6H), 7.07
(d, 4H, J ) 6.6), 6.71 (d, 4H, J ) 8.5), 4.59 (sbr, 2H), 4.12 (d, 2H,
J ) 15.4), 3.85 (d, 2H, J ) 15.6), 3.48–3.36 (m, 4H); 13C NMR
(MeOH-d4) δ 154.9, 134.5, 131.0, 129.9, 129.6, 129.2, 125.1, 114.7,
62.3, 51.9, 48.2; MS (ESI) m/z 592 (100, M + H), Anal.
(C30H29N5O4S2 ·3HCl·4H2O) C, H, N.
(3S,4S)-3,4-Bis[(4-carbamoylbenzenesulfonyl)-(4-trifluoro-
methylbenzyl)amino]pyrrolidine Hydrochloride (6fg). Following
general procedure D, employment of 148 mg (0.17 mmol) of BOC-
protected pyrrolidine 5fg and 5 mL of 2 M HCl in Et2O furnished
hydrochloride 6fg (110 mg, 80%). Yellow crystals; mp 194 °C
(dec); 1H NMR (DMSO-d6) δ 9.79 (sbr, 2H), 8.23 (sbr, 2H), 8.04
(sbr, 4H), 7.83 (sbr, 4H), 7.64 (sbr, 8H), 7.60 (sbr, 2H), 4.90 (sbr,
2H), 4.58 (sbr, 2H), 3.07 (sbr, 2H), 2.76 (sbr, 2H); 13C NMR
(DMSO-d6) δ 165.9, 141.5, 140.1, 138.1, 128.3, 128.2, 127.5 (q, J
) 32.6), 127.1, 124.5, 123.9 (q, J ) 273.5), 57.6, 47.1, 42.1; MS
(ESI) m/z 784 (100, M + H). Anal. (C34H31F6N5O6S2 ·HCl·4H2O)
C, H, N.
1
crystals; mp 157 °C; H NMR (MeOH-d4) δ 8.04–7.92 (m, 4H),
7.89–7.72 (m, 4H), 7.53 (sbr, 4H), 7.46–7.29 (sbr, 4H), 4.78–4.53
(mbr, 4H), 4.31–4.01 (m, 2H), 3.37–3.12 (mbr, 2H), 2.97 (sbr, 2H),
1.34 (s, 9H); 13C NMR (MeOH-d4) δ 170.2, 155.5, 143.9, 142.6,
139.5, 131.0 (q, J ) 31.7), 129.9, 129.8, 128.6, 126.5, 125.5 (q, J
) 271.6), 81.8, 61.8, 61.0, 49.6, 48.5, 47.6, 47.0, 28.5; MS (ESI)
m/z 906 (99, M + Na). Anal. (C39H39F6N5O8S2) C, H, N.
(3S,4S)-3,4-Bis[benzenesulfonylbenzylamino]pyrrolidine Hy-
drochloride (6ad). According to general procedure D, utilization
of 102 mg (0.15 mmol) of BOC-protected pyrrolidine 5ad and 3
mL of 2 M HCl in Et2O gave rise to hydrochloride 6ad (87 mg,
1
92%). Colorless crystals; mp 221 °C (dec); H NMR (DMSO-d6)
δ 9.65 (sbr, 2H), 7.80 (dd, 4H, J ) 8.4, 1.2), 7.72 (tt, 2H, J ) 7.4,
1.2), 7.60 (tt, 4H, J ) 7.8, 1.6), 7.29–7.20 (mbr, 10H), 4.80 (sm,
2H), 4.39 (d, 2H, J ) 16.5), 3.97 (d, 2H, J ) 16.5), 3.01 (dd, 2H,
J ) 11.3, 7.9), 2.78 (dd, 2H, J ) 11.6, 8.1); 13C NMR (DMSO-d6)
δ 139.1, 136.6, 133.3, 129.4, 128.2, 128.0, 127.4, 127.2, 57.7, 48.1,
43.2; MS (ESI) m/z 562 (100, M + H), 1123 (9, 2M + H). Anal.
(C30H31N3O4S2 ·HCl·0.5H2O) H, N. C: calcd, 59.34; found, 59.75.
(3S,4S)-3,4-Bis[benzenesulfonyl-(4-iodobenzyl)amino]pyrrol-
idine Hydrochloride (6af). Following the general procedure D,
usage of 183 mg (0.20 mmol) of BOC-protected pyrrolidine 5af
and 5 mL of 2 M HCl in Et2O yielded hydrochloride 6af (138 mg,
Acknowledgment. We acknowledge the support of the
beamline staff at BESSY II, Berlin, Germany, as well as the
beamline staff at EMBL/DESY, Hamburg, Germany. The
pET11a plasmid containing the HIV-1 protease gene was kindly
provided by Professor Helena Danielson, University of Uppsala,
Sweden. This work was also supported by a fellowship (A.B.)

Pyrrolidines as HIV Protease Inhibitors
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 7 2087
Bis(diphenylphosphino)-pyrrolidines. The Use of their Rhodium
Complexes for the Asymmetric Hydrogenation of R-(Acylamino)-
acrylic Acid Derivatives. Chem. Ber. 1986, 119 (11), 3326–3343.
(18) Rocha Gonsalves, A. M. d. A.; Serra, M. E. S.; Murtinho, D.; Silva,
V. F.; Matos Beja, A.; Paixão, J. A.; Ramos Silva, M.; Alte da Veiga,
L. Pyrrolidine-based amino alcohols: novel ligands for the enantiose-
lective alkylation of benzaldehyde. J. Mol. Catal. A: Chem. 2003, 195
(1–2), 1–9.
(19) Skarøewski, J.; Gupta, A. Synthesis of C2 symmetric primary vicinal
diamines. Double stereospecific Mitsunobu reaction on the heterocyclic
diols derived from tartaric acid. Tetrahedron: Asymmetry 1997, 8 (11),
1861–1867.
(20) Rodríguez Sarmiento, R. M.; Wirz, B.; Iding, H. Chemoenzymatic
preparation of non-racemic N-Boc-pyrrolidine-3,4-dicarboxylic acid
3-ethyl esters and their 4-hydroxymethyl derivatives. Tetrahedron:
Asymmetry 2003, 14 (11), 1547–1551.
(21) Kuppert, D.; Sander, J.; Roth, C.; Wörle, M.; Weyhermüller, T.; Reiss,
G. J.; Schilde, U.; Müller, I.; Hegetschweiler, K. The coordination
chemistry of cis-3,4-diaminopyrrolidine and related polyamines. Eur.
J. Inorg. Chem. 2001, (10), 2525–2542.
(22) Song, C. E.; Yang, J. W.; Roh, E. J.; Lee, S.-G.; Ahn, J. H.; Han, H.
Heterogeneous Pd-catalyzed asymmetric allylic substitution using
resin-supported trost-type bisphosphane ligands. Angew. Chem., Int.
Ed. 2002, 41 (20), 3852–3854.
(23) Katritzky, A. R.; Pilarski, B.; Urogdi, L. Efficient conversion of nitriles
to amides with basic hydrogen peroxide in dimethyl sulfoxide.
Synthesis 1989, (12), 949–950.
(24) Kettler, K.; Sakowski, J.; Wiesner, J.; Ortmann, R.; Jomaa, H.;
Schlitzer, M. Novel lead structures for antimalarial farnesyltransferase
inhibitors. Pharmazie 2005, 60 (5), 323–327.
(25) Taylor, A.; Brown, D. P.; Kadam, S.; Maus, M.; Kohlbrenner, W. E.;
Weigl, D.; Turon, M. C.; Katz, L. High-level expression and
purification of mature HIV-1 protease in Escherichia coli under control
of the araBAD promoter. Appl. Microbiol. Biotechnol. 1992, 37 (2),
205–210.
(26) Toth, M. V.; Marshall, G. R. A simple, continuous fluorometric assay
for HIV protease. Int. J. Pept. Protein Res. 1990, 36 (6), 544–550.
(27) Otwinowski, Z.; Minor, W. Processing of X-ray Diffraction Data
Collected in Oscillation Mode. In Methods in Enzymology; Carter,
C. W., Jr., Ed.; Academic Press: San Diego, CA, 1997; Vol. 276, pp
307–326.
(28) Storoni, L. C.; McCoy, A. J.; Read, R. J. Likelihood-enhanced fast
rotation functions. Acta Crystallogr. D 2004, 60 (3), 432–438.
(29) Pape, T.; Schneider, T. R. HKL2MAP: a graphical user interface for
macromolecular phasing with SHELX programs. J. Appl. Crystallogr.
2004, 37 (5), 843–844.
(30) Usón, I.; Sheldrick, G. M. Advances in direct methods for protein
crystallography. Curr. Opin. Struct. Biol. 1999, 9 (5), 643–648.
(31) Sheldrick, G. M. Macromolecular phasing with SHELXE. Z. Kristal-
logr. 2002, 217 (12), 644–650.
(32) Cowtan, K. D.; Zhang, K. Y. J. Density modification for macromo-
lecular phase improvement. Prog. Biophys. Mol. Biol. 1999, 72 (3),
245–270.
(33) Lamzin, V. S.; Wilson, K. S. Automated refinement of protein models.
Acta Crystallogr. D 1993, 49 (1), 129–147.
(34) The CCP4 suite: programs for protein crystallography. Acta Crystal-
logr. D 1994, 50 (5), 760–763.
(35) Brünger, A. T.; Adams, P. D.; Clore, G. M.; DeLano, W. L.; Gros,
P.; Grosse-Kunstleve, R. W.; Jiang, J.-S.; Kuszewski, J.; Nilges, M.;
Pannu, N. S.; Read, R. J.; Rice, L. M.; Simonson, T.; Warren, G. L.
Crystallography & NMR system: a new software suite for macromo-
lecular structure determination. Acta Crystallogr. D 1998, 54 (5), 905–
921.
(36) Sheldrick, G. M.; Schneider, T. R., SHELXL: High-Resolution
Refinement. In Methods in Enzymology; Charles, W. C. J., Robert,
M. S., Eds.; Academic Press: San Diego, CA, 1997; Vol. 277, pp
319–343.
from the Fonds der Chemischen Industrie (FCI) and by the
Deutsche Pharmazeutische Gesellschaft (DPhG) “Stiftung zur
Förderung des wissenschaftlichen Nachwuchses” (W.E.D.).
Supporting Information Available: Combustion analysis re-
sults, crystallographic data, and additional details of experimental
procedure. This material is available free of charge via the Internet
at http://pubs.acs.org.
References
(1) UNAIDS: 2006 Report on the Global AIDS Epidemic; UNAIDS:
Geneva, 2006.
(2) Pomerantz, R. J.; Horn, D. L. Twenty years of therapy for HIV-1
infection. Nat. Med. 2003, 9 (7), 867–873.
(3) Wlodawer, A.; Erickson, J. W. Structure-based inhibitors of HIV-1
protease. Annu. ReV. Biochem. 1993, 62 (1), 543–585.
(4) Rodríguez-Barrios, F.; Gago, F. HIV protease inhibition: limited recent
progress and advances in understanding current pitfalls. Curr. Top.
Med. Chem. 2004, 4, 991–1007.
(5) Chrusciel, R. A.; Strohbach, J. W. Non-peptidic HIV protease
inhibitors. Curr. Top. Med. Chem. 2004, 4 (10), 1097–1114.
(6) Randolph, J. T.; DeGoey, D. A. Peptidomimetic inhibitors of HIV
protease. Curr. Top. Med. Chem. 2004, 4 (10), 1079–1095.
(7) D’Aquila, R. T.; Schapiro, J. M.; Brun-Vézinet, F.; Clotet, B.; Conway,
B.; Demeter, L. M.; Grant, R. M.; Johnson, V. A.; Kuritzkes, D. R.;
Loveday, C.; Shafer, R. W.; Richman, D. D. Drug resistance mutations
in HIV-1. Top. HIV Med. 2002, 10 (5), 21–25.
(8) Güller, R.; Binggeli, A.; Breu, V.; Bur, D.; Fischli, W.; Hirth, G.;
Jenny, C.; Kansy, M.; Montavon, F.; Müller, M.; Oefner, C.; Stadler,
H.; Vieira, E.; Wilhelm, M.; Wostl, W.; Märki, H. P. Piperidine-renin
inhibitors compounds with improved physicochemical properties.
Bioorg. Med. Chem. Lett. 1999, 9 (10), 1403–1408.
(9) Vieira, E.; Binggeli, A.; Breu, V.; Bur, D.; Fischli, W.; Güller, R.;
Hirth, G.; Märki, H. P.; Müller, M.; Oefner, C.; Scalone, M.; Stadler,
H.; Wilhelm, M.; Wostl, W. Substituted piperidinesshighly potent
renin inhibitors due to induced fit adaptation of the active site. Bioorg.
Med. Chem. Lett. 1999, 9 (10), 1397–1402.
(10) John, V.; Beck, J. P.; Bienkowski, M. J.; Sinha, S.; Heinrikson, R. L.
Human ꢀ-secretase (BACE) and BACE inhibitors. J. Med. Chem.
2003, 46 (22), 4625–4630.
(11) Prade, L.; Jones, A. F.; Boss, C.; Richard-Bildstein, S.; Meyer, S.;
Binkert, C.; Bur, D. X-ray structure of plasmepsin II complexed with
a potent achiral inhibitor. J. Biol. Chem. 2005, 280 (25), 23837–23843.
(12) Hof, F.; Schütz, A.; Fäh, C.; Meyer, S.; Bur, D.; Liu, J.; Goldberg,
D. E.; Diederich, F. Starving the malaria parasite: inhibitors active
against the aspartic proteases plasmepsins I, II, and IV. Angew. Chem.,
Int. Ed. 2006, 45 (13), 2138–2141.
(13) Specker, E.; Böttcher, J.; Lilie, H.; Heine, A.; Schoop, A.; Müller,
G.; Griebenow, N.; Klebe, G. An old target revisited: two new
privileged skeletons and an unexpected binding mode for HIV-protease
inhibitors. Angew. Chem., Int. Ed. 2005, 44 (20), 3140–3144.
(14) Specker, E.; Böttcher, J.; Brass, S.; Heine, A.; Lilie, H.; Schoop, A.;
Müller, G.; Griebenow, N.; Klebe, G. Unexpected novel binding mode
of pyrrolidine-based aspartyl protease inhibitors: design, synthesis and
crystal structure in complex with HIV protease. ChemMedChem 2006,
1 (1), 106–117.
(15) Czodrowski, P.; Sotriffer, C. A.; Klebe, G. Atypical protonation states
in the active site of HIV-1 protease: a computational study. J. Chem.
Inf. Model. 2007, 47 (4), 1590–1598.
(16) Nagel, U. Asymmetric hydrogenation of R-(acetylamino)cinnamic acid
with a novel rhodium complex; the design of an optimal ligand. Angew.
Chem., Int. Ed. Engl. 1984, 23 (6), 435–436.
(17) Nagel, U.; Kinzel, E.; Andrade, J.; Prescher, G. Enantioselektive
Katalyse, 4. Synthese N-substituierter (R,R)-3,4-Bis(diphenylphos-
phino)-pyrrolidine und Anwendung ihrer Rhodiumkomplexe zur
asymmetrischen Hydrierung von R-(Acylamino)-acrylsäure-Derivaten.
Enantioselective Catalysis, 4. Synthesis N-Substituted (R,R)-3,4-
(37) Emsley, P.; Cowtan, K. Coot: model-building tools for molecular
graphics. Acta Crystallogr. D 2004, 60 (12, Part 1), 2126–2132.
JM701142S