Structure-based design of novel, urea-containing FKBP12 inhibitors

Article abstract of DOI:10.1021/jm950798a

The structure-based design and subsequent chemical synthesis of novel, urea-containing FKBP12 inhibitors are described. These compounds are shown to disrupt the cis-trans peptidylprolyl isomerase activity of FKBP12 with inhibition constants (K(i,app)) approaching 0.10 μM. Analyses of several X- ray crystal structures of FKBP12-urea complexes demonstrate that the urea- containing inhibitors associate with FKBP12 in a manner that is similar to, but significantly different from, that observed for the natural product FK506.

Full text of DOI:10.1021/jm950798a

1872
J . Med. Chem. 1996, 39, 1872-1884
Str u ctu r e-Ba sed Design of Novel, Ur ea -Con ta in in g F KBP 12 In h ibitor s
Peter S. Dragovich,* J ohn E. Barker, J udy French, Michael Imbacuan, Vincent J . Kalish, Charles R. Kissinger,
Daniel R. Knighton, Cristina T. Lewis, Ellen W. Moomaw, Hans E. Parge, Laura A. K. Pelletier,
Thomas J . Prins, Richard E. Showalter, J ohn H. Tatlock, Kathleen D. Tucker, and J . Ernest Villafranca
Agouron Pharmaceuticals, Inc., 3565 General Atomics Court, San Diego, California 92121
Received October 30, 1995X
The structure-based design and subsequent chemical synthesis of novel, urea-containing
FKBP12 inhibitors are described. These compounds are shown to disrupt the cis-trans
peptidylprolyl isomerase activity of FKBP12 with inhibition constants (Ki,app) approaching 0.10
µM. Analyses of several X-ray crystal structures of FKBP12-urea complexes demonstrate
that the urea-containing inhibitors associate with FKBP12 in a manner that is similar to, but
significantly different from, that observed for the natural product FK506.
In tr od u ction
FK506 (tacrolimus, Figure 1) is a potent immunosup-
pressive natural product whose mechanism of action has
recently been elucidated at the cellular level.1,2 FK506
strongly associates with the immunophilin FKBP12
(FK506-binding protein), a member of a ubiquitous
family of proteins which catalyze the cis-trans isomer-
ization of peptidyl proline amide bonds.3,4 FK506
potently inhibits FKBP12 isomerase activity, presum-
ably by binding as a transition state analog, and it was
initially speculated that such isomerase inhibition was
responsible for the immunosuppressive properties which
the molecule displays.5 This theory was supported by
the earlier finding that the structurally unrelated
immunosuppressant cyclosporin A (CsA) inhibits the
function of a different cellular target, cyclophilin (cy-
closporin-binding protein, CyP), which also possesses
isomerase activity.6,7 However, rapamycin, another
highly immunosuppressive natural product, also po-
tently inhibits FKBP12 isomerase activity but antago-
nizes FK506-related immunosuppressive effects.8,9 Ad-
ditional experiments have established that the complex
formed by FK506 and FKBP12 inhibits the function of
a second cellular target, the calcium-dependent, calm-
odulin-activated protein phosphatase calcineurin (CN,
PP2B).10,11 This inhibition, in turn, is believed to
prevent nuclear translocation of the NF-AT (nuclear
factor of activated T-cells) cytosolic component thereby
blocking Ca2+-dependent IL-2 gene transcription neces-
sary for T-cell activation.12
It has been implied from the above experimental data
that FK506 can be divided into two distinct regions: one
which associates with FKBP12 and one which contacts
calcineurin upon formation of the FKBP12-FK506
complex.13 The former region, the binding domain,
contains the pipecolinic acid R-ketoamide functional
group and extends from the pyranose fragment on one
side of the molecule to the cyclohexyl group on the other.
The latter domain, the effector region, includes portions
F igu r e 1. Natural product immunosuppressants.
of the remainder of the molecule. Such assignments are
supported by NMR solution structures14 and X-ray
crystal structures15 of FK506 complexed with FKBP12
and were recently confirmed by the determination of the
X-ray crystal structures of the calcineurin-FKBP12-
FK506 ternary complex.16 An important conclusion of
the dual-domain hypothesis is that only molecules which
incorporate both functional domains would be expected
to exhibit immunosuppressive effects similar to those
induced by FK506. Indeed, the synthetic molecule
506BD, which contains the FKBP12-binding domain of
FK506 but lacks an appropriate effector region, potently
inhibits FKBP12 isomerase activity but displays no
immunosuppressive properties.13,17 In addition, selected
X Abstract published in Advance ACS Abstracts, March 1, 1996.
0022-2623/96/1839-1872$12.00/0 © 1996 American Chemical Society

Structure-Based Design of Novel FKBP12 Inhibitors
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 9 1873
Ta ble 1
alterations to either the effector region of FK506 or
certain FKBP12 residues which surround the FK506-
binding pocket can diminish the immunosuppressive
properties of the corresponding FKBP12-FK506 com-
plexes (presumably by disrupting their association with
calcineurin) without significantly affecting FKBP12-
FK506 interactions.18,19 Such results, coupled with the
X-ray analysis of the calcineurin-FKBP12-FK506
ternary complex,16 indicate that elements of both FK506
and FKBP12 contribute to a composite surface which
is recognized by calcineurin and underscore the need
for FK506-like immunosuppressants to associate with
FKBP12 in a manner that does not preclude the
formation of such a surface.
In h ibitor Design a n d Str u ctu r e-Activity
Stu d ies
As part of a program to develop novel immunosup-
pressants that function in a manner similar to that of
FK506, we sought to design simple FKBP12-binding
moieties to which potential effector regions could be
attached. The apparent inhibition constant of FKBP12
isomerase activity (Ki,app) exhibited by a molecule bind-
ing at the FKBP12 active site is known to correlate well
with measured dissociation constants (Kd).20a,b This
inhibition constant can be accurately determined by
known assay techniques and was therefore chosen for
use in this study as an indicator of FKBP12-ligand
association strength (full details of the biochemical
assay methods are presented in the Experimental
Section).
Small synthetic molecules which incorporate a (S)-
pipecolyl-derived R-ketoamide moiety are known to be
potent inhibitors of FKBP12 isomerase activity (Ki,app
) 1-100 nM).20 However, in an effort to obtain immu-
nosuppressive compounds with pharmacological profiles
that differed from FK506, the development of FKBP12
inhibitors which did not contain the R-ketoamide func-
tional group was considered. Molecules belonging to
this category are typically not potent FKBP12 inhibi-
tors,21 although several groups have reported sub-
micromolar FKBP12 isomerase inhibition by pipecolyl-
derived sulfonamides.22 In addition, it was reasoned
that an ideal binding moiety would depart the FKBP12
active site in a manner similar to that observed for
FK506 in order to facilitate the future design of effector
regions which spatially resemble the natural product.
The above considerations led to the general design of
an inhibitor in which the R-ketoamide functional group
typically found in potent FKBP12-binding molecules
was replaced by a urea moiety (Table 1). Such a
molecule would forfeit the beneficial interactions made
by the R-ketoamide ketone carbonyl with the ꢀ-hydro-
gens of three aromatic FKBP12 residues on the floor of
the binding cavity (Tyr-26, Phe-36, and Phe-99) but was
anticipated to form a hydrogen bond between the urea
NH and the carboxylate of Asp-37 analogous to that
made by the hemiketal hydroxyl group of FK506.14,15
Analysis of computer-generated models suggested that,
due to the planar geometry of the urea functional group,
inclusion of a methylene unit adjacent to the urea NH
would allow appended moieties (R1) to better contact the
FKBP12 residues which interact with the FK506 pyra-
nose ring (Phe-36, Ile-90, Ile-91, and His-87). The
development of two series of urea-containing compounds
which function as inhibitors of FKBP12 isomerase
activity is described below.
Initially, the identification of an appropriate FK506
pyranose mimetic for incorporation into an acyclic urea
derived from (S)-pipecolinic acid benzyl ester was
undertaken. Since previous studies of R-ketoamide-
containing FKBP12 inhibitors had demonstrated that
large hydrophobic groups could function as effective
pyranose mimics, the inclusion of similar moieties in
the urea series seemed appropriate.20 Accordingly, a
number of ureas were prepared by combining (S)-
pipecolinic acid benzyl ester with various primary
amines which incorporated hydrophobic moieties of
differing sizes (Table 1). Ureas containing small (1) or
flat (2) hydrophobic pyranose replacement groups dis-
played moderate FKBP12 inhibition levels in the low
micromolar range. Incorporation of the larger tert-butyl
moiety into the urea design substantially increased the
FKBP12 affinity of the resulting inhibitor (compare 2
and 3). Inclusion of an unprotected hydroxyl group in
this portion of the inhibitor (4) was detrimental to
FKBP12 affinity, presumably due to poor interactions
with the hydrophobic FKBP12 residues which contact
the pyranose mimics. Crystallographic analysis of the
FKBP12-3 complex indicated that additional protein-
ligand hydrophobic interactions could be realized by
incorporating a portion of the tert-butyl moiety into a
hydrocarbon ring. Accordingly, compounds 5 and 6
were prepared with such a modification, and both
inhibited FKBP12 more potently than compound 3. The
preparation of compounds 5 and 6 concluded the devel-
opment of pyranose mimics for use with the acyclic urea
series. However, since molecules which contained the
cyclic pyranose mimics proved more difficult to crystal-
lize with FKBP12 than their tert-butyl counterparts, the
latter fragment was utilized in many subsequent inhibi-
tor designs.

1874 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 9
Dragovich et al.
Ta ble 2
Ta ble 3
compd no.a
R1
CH2Ph
R2
Ki,app (µM)
14
15
16
17
CH3
CH2Ph
CH3
67
14
83
24
b
CH2CH2CH3
b
CH2Ph
a
b
Prepared in racemic form. Same as above.
inhibitors was also undertaken (Table 3). Although
such compounds would forfeit the presumed hydrogen
bonding interactions of the acyclic ureas with Asp-37
of the protein, their cyclic nature would prevent cis-
trans isomer equilibration and thus potentially increase
FKBP12 affinity by rigidifying the inhibitor in a ben-
eficial conformation. Computational analysis utilizing
the FKBP12-FK506 crystal structure15 suggested that
the proposed bicyclic framework could be accommodated
in the FKBP12 active site with small movements of Asp-
37 and Tyr-82 outward from the center of the binding
pocket. Since the binding geometry of the acyclic and
cyclic ureas was expected to be similar, it was antici-
pated that structure-activity information gained from
development of the former inhibitor series could be
utilized for optimization of the latter. In the event,
several racemic cyclic ureas were prepared and dis-
played mid to low micromolar levels of FKBP12 inhibi-
tion (Table 3). However, crystallographic analysis of the
FKBP12-17 complex revealed that the binding geom-
etry of the cyclic urea was significantly different from
that observed for the related acyclic molecules described
above and suggested that independent structure-activ-
ity studies would be required for optimization of the new
compounds. Therefore, subsequent synthetic efforts
were focused on the development of the more easily
prepared acyclic urea series.
a
Same as above.
Having identified several suitable pyranose replace-
ments for use in the development of the urea-containing
inhibitors, variation of the ester functionality was then
examined with the dual purpose of providing increased
FKBP12 affinity and a departure point for the subse-
quent attachment of appropriate effector regions (Table
2). Incorporation of simplified versions of the FK506
cyclohexyl moiety (C29-C34) into the inhibitor design
increased the FKBP12 affinity of the resulting com-
pounds 2-5-fold as compared to their benzyl ester
counterparts. Initially, it was believed that the in-
creased FKBP12 affinity displayed by all of these
compounds was derived from additional hydrophobic
interactions between the larger ester fragments and the
region of FKBP12 which contacts the FK506 cyclohexyl
moiety. However, crystallographic analysis of the
FKBP12-8 complex revealed this assumption to be
incorrect and also suggested that the corresponding 3-(3-
methoxyphenyl)propyl ester should display similar
FKBP12 affinity. This prediction was verified experi-
mentally with the preparation and evaluation of com-
pound 9 (Table 2). Replacement of the tert-butyl
pyranose mimetics of the most active molecules with the
larger cyclohexyl moiety resulted in slightly more potent
FKBP12 inhibitors relative to the original compounds
(compare 10 with 12 and 11 with 13). These ureas rank
among the most potent non-R-ketoamide-containing
FKBP12 inhibitors reported to date and should function
as suitable FKBP12-binding moieties for the attachment
of appropriate effector regions.23
X-r a y Str u ctu r e An a lyses
The iterative analysis of protein-ligand interactions
by X-ray crystallography is an essential element of a
structure-based inhibitor design program.24 Accord-
ingly, several crystal structures of FKBP12-inhibitor
complexes were obtained during the development of the
urea-containing compounds described above. These
structures established the binding geometries that the
urea inhibitors adopted when complexed with FKBP12,
suggested improvements for ligand design, allowed
comparison of FKBP12-urea interactions with those
reported for related R-ketoamide inhibitors (including
FK506), and identified departure points from the
FKBP12-binding pocket for subsequent effector region
attachment. An overview of the general characteristics
of the FKBP12-urea crystal structures obtained is
presented below followed by discussion of specific details
that were observed in individual FKBP12-inhibitor
complexes.
In addition to the acyclic ureas described above, the
development of a series of cyclic ureas as FKBP12
The overall FKBP12 structure observed for all new
FKBP12-inhibitor complexes was not significantly dif-

Structure-Based Design of Novel FKBP12 Inhibitors
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 9 1875
Ta ble 4. Root-Mean-Square Differences between Atom
Positions of FKBP12-FK506 and FKBP12-Ligand Complexes
compd no.
R-carbons
backbone atoms
3
4
7
8
9
0.39
0.38
0.40
0.50
0.53
0.38
0.40
0.38
0.40
0.48
0.53
0.39
17
F igu r e 3. Crystal structure of the FKBP12-3 complex
(alternate view). A portion of the water-accessible surface of
the protein is shown as pink dots. FK506 (hydrogen atoms
omitted) from the FKBP12-FK506 crystal structure is super-
imposed and shown in yellow.
methyl group contacts the side chain of Phe-46 in the
FKBP12-FK506 complex and most likely influences its
crystallographically observed position. Importantly, the
locations at which the acyclic urea inhibitors exited the
FKBP12-binding cavity were spatially similar to those
observed for FK506 in the FKBP12-FK506 complex,
presumably simplifying the task of subsequent effector
region design.
F igu r e 2. Crystal structure of the FKBP12-3 complex.
Hydrogen bonds are indicated as white dashed lines. A portion
of the water-accessible surface of the protein is shown as pink
dots. Selected side chains of FKBP12 residues from the
FKBP12-FK506 crystal structure are superimposed and
shown in yellow.
One of the most surprising aspects of the FKBP12-
inhibitor crystal structures was the binding geometry
adopted by the pipecolyl fragment of the urea ligands
(Figure 3). This geometry was significantly different
from that observed for related pipecolyl-derived R-ke-
toamides (such as FK506) and resulted in a departure
of the urea functional group from the plane defined by
the FK506 amide moiety. Due to the planar nature of
the urea, the locations of attached pyranose mimetics
were also altered slightly from the position of the FK506
pyranose moiety. In addition, the three aromatic
FKBP12 residues on the floor of the active site whose
ꢀ-hydrogens interact with the FK506 ketone carbonyl
moiety (Tyr-26, Phe-36, and Phe-99) did not move
appreciably from their locations in the FKBP12-FK506
complex. Thus, a significant gap between protein and
ligand was observed directly below the urea NH in all
the FKBP12-inhibitor complexes examined (Figure 3).
The presence of this, presumably unfavorable, gap in
the FKBP12-urea complexes may be primarily respon-
sible for the somewhat weaker FKBP12 affinity exhib-
ited by a typical acyclic urea inhibitor relative to similar
R-ketoamide-containing compounds.20
In spite of the altered binding geometry described
above, good hydrophobic interactions were observed
between the tert-butyl group of compound 3 and the
FKBP12 residues which interact with the FK506 pyra-
nose moiety (Phe-36, Ile-90, Ile-91, and His-87; Figures
2 and 3). Importantly, the positions of these residues
in the FKBP12-3 complex were virtually unchanged
from their locations in the FKBP12-FK506 crystal
structure (Figure 2). Since crystallographic analysis of
the calcineurin-FKBP12-FK506 ternary complex has
established that this region of FKBP12 forms part of
the composite surface which interacts with calcineurin,
the ability of the urea inhibitors to associate with
FKBP12 without perturbing these residues was note-
ferent from that found in the uncomplexed protein or
the FKBP12-FK506 or FKBP12-rapamycin crystal
structures.14,15,25 Comparison of the FKBP12-inhibitor
complexes with that of FKBP12-FK506 revealed small
root-mean-square (rms) deviations between R-carbon
atom positions as well as backbone atom locations
(Table 4). Similar gross protein structure was reported
for the complexes formed between FKBP12 and several
non-macrocyclic R-ketoamide inhibitors.15d,20b
The acyclic urea inhibitors, typified by compound 3,
bound to FKBP12 in a manner that was similar to, but
significantly different from, that observed for related
R-ketoamide-containing inhibitors such as FK506. The
pipecolyl ring of each compound was positioned above
the indole side chain of Trp-59, with hydrogen bonds
observed between the pipecolyl ester carbonyl and Ile-
56-NH and between the urea carbonyl and Tyr-82-OH
(Figures 2 and 3). Comparable inhibitor locations and
hydrogen-bonding interactions have been observed in
crystal structures of related R-ketoamides complexed
with FKBP12.20b,i As anticipated, an additional (albeit
long) hydrogen bond was observed between the urea NH
and the carboxylate of Asp-37, accompanied by move-
ment of the Asp-37 side chain toward the inhibitor
(typically 0.87 Å relative to its position observed in the
FKBP12-FK506 complex). Other FKBP12 residues
which form hydrogen bonds with Asp-37 (Arg-42 and
Tyr-26) also moved slightly from their observed locations
in the FKBP12-FK506 crystal structure. In addition,
movement (∼1.1 Å) of the Phe-46 aromatic side chain
toward the inhibitor (relative to its position in the
FKBP12-FK506 structure) was also observed in many
FKBP12-urea complexes (Figure 2). This movement
probably results from the absence of inhibitor function-
ality which mimics the C19 methyl group of FK506. This

1876 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 9
Dragovich et al.
F igu r e 4. Crystal structure of the FKBP12-8 complex. The
ligand 7 is superimposed and shown in gray. Hydrogen atoms
have been omitted from both compounds 7 and 8 for clarity. A
portion of the water-accessible FKBP12 surface is shown as
pink dots. The side chain of Phe-46 from the FKBP12-FK506
crystal structure is superimposed and shown in yellow.
F igu r e 5. Crystal structure of the FKBP12-17 complex.
Hydrogen bonds are indicated as white dashed lines. A portion
of the water-accessible surface of the protein is shown as pink
dots. Selected side chains of FKBP12 residues from the
FKBP12-FK506 crystal structure are superimposed and
shown in yellow.
worthy.16,26 In addition, the benzyl ester fragment of 3
made favorable inter- and intramolecular hydrophobic
contacts with Phe-46 and the tert-butyl portion of the
molecule, respectively. Analysis of the FKBP12-3
complex also identified possible locations for additional
FKBP12-ligand hydrophobic interactions in the pyra-
nose-binding region as described in the structure-
activity section above.
Crystallographic comparison of the FKBP12-4 and
FKBP12-3 complexes indicated that 4 bound to FKBP12
in a manner almost identical with that of 3 with the
hydroxyl group oriented away from the pyranose-
binding pocket toward solvent (data not shown). The
reduced FKBP12 affinity of compound 4 relative to 3
presumably arises from unfavorable interactions be-
tween the polar hydroxyl group and the hydrophobic
FKBP12 residues which contact both the FK506 pyra-
nose group and the tert-butyl moiety of 3. Interestingly,
the electron density observed for the pyranose mimetic
region of compound 4 was well-defined, in contrast to
that noted for the corresponding portions of all other
urea inhibitors when complexed with FKBP12. This
increased definition presumably results from reduced
conformational mobility imparted by the hydroxyl group
of 4 and suggests that the pyranose mimics of other
urea-containing FKBP12 ligands may adopt multiple
binding geometries.
As was previously observed for a related R-ketoamide-
containing FKBP12 inhibitor, the phenethyl fragment
of compound 7 interacted with FKBP12 in a region near
where the cyclohexyl moiety of FK506 (C29-C34) was
located in the FKBP12-FK506 crystal structure (cy-
clohexyl-binding region; Figure 4).20b The phenyl ring
of 7 was rotated 90° relative to the plane defined by the
cyclohexyl ring of FK506, presumably to maximize
beneficial hydrophobic interactions with FKBP12. As
was noted for compound 3 above, the benzylic ester
fragment of 7 made beneficial inter- and intramolecular
hydrophobic interactions with Phe-46 of FKBP12 and
the tert-butyl pyranose mimic, respectively.
the structure of 7 (Figure 4). Thus, a single (meta) aryl
methoxy group interacted with FKBP12 at a site near
to the location of the FK506 C19 methyl group in the
FKBP12-FK506 crystal structure. Good hydrophobic
interactions were observed between this methoxy group
(and the aryl ring itself) and Phe-46, the side chain of
which was positioned as observed in the FKBP12-
FK506 crystal structure. The remaining aryl methoxy
groups of 8 did not appear to significantly contact
FKBP12, and their removal from the inhibitor design
was therefore predicted to have a negligible effect on
the resulting FKBP12-ligand interactions. This pre-
diction was verified by the preparation and analysis of
compound 9 which was crystallographically observed to
associate with FKBP12 in a manner similar to that
described for compound 8 (data not shown). The
observed binding of 8 and 9 is believed to arise from a
prearrangement of the compounds in aqueous solution
in which intramolecular hydrophobic interactions be-
tween the ester appendage and the pyranose mimetics
are maximized.27 Compound 7 may be similarly prear-
ranged, but in this case the benzylic ester portion of the
molecule contacts the tert-butyl pyranose mimic, allow-
ing the phenethyl fragment to occupy the cyclohexyl-
binding region of FKBP12. Similar factors may also
influence the FKBP12 association of other inhibitors
which contain branched pipecolyl esters derived from
secondary alcohols.20
In addition to the FKBP12-acyclic urea complexes
examined above, the crystal structure of the cyclic urea
inhibitor 17 complexed with FKBP12 was also deter-
mined. The pipecolyl-derived portion of the 6,6-bicyclic
structure was located above the indole ring of Trp-59,
analogous to the positioning of the acyclic urea inhibi-
tors examined above (Figure 5). The anticipated move-
ment of Asp-37 away from the ligand was also observed
(0.37 Å relative to its position in the FKBP12-FK506
crystal structure), presumably to accommodate the
larger bicyclic structure in the FKBP12 active site. The
FKBP12 residues which form hydrogen bonds to Asp-
37 (Arg-42 and Tyr-26) were also slightly displaced away
from the binding cavity but without significant pertur-
bation of the general protein structure. As was noted
for the acyclic ureas above, the binding geometry of the
bicyclic inhibitor was significantly different from that
In contrast, the crystal structure of the FKBP12-8
complex revealed that the trimethoxyphenyl fragment
of 8 contacted the protein in a region occupied crystal-
lographically by a portion of the FK506 effector domain
rather than the cyclohexyl moiety as anticipated from

Structure-Based Design of Novel FKBP12 Inhibitors
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 9 1877
Sch em e 1a
Sch em e 2. Possible Cyclization Outcomes of Proposed
N-Acyliminium Ion Intermediates in the Preparation of
14-17
a
Reagents and conditions (R1, R2 ) see Tables 1 and 2; Bn )
benzyl): (a) 1.0 equiv of OdCdNCH2R1, 1.0 equiv of Et3N, CH2Cl2,
23 °C, 1 h, 75-90%; (b) 1.0 equiv of R2OH, 2.0 equiv of DCC, 0.3
equiv of CSA, 0.3 equiv of DMAP, CH2Cl2, 23 °C, 4-12 h, 80-
90%; (c) 4.0 M HCl in 1,4-dioxane, 23 °C, 1 h.
tion, basic hydrolysis) invariably resulted in the forma-
tion of the corresponding hydantoins (eq 1). Thus,
observed for related pipecolyl-derived R-ketoamides.
This geometry difference was even more pronounced in
the structure of 17 and caused a portion of the bicyclic
structure to protrude above the FKBP12-binding pocket
(Figure 5). Not unexpectedly, such binding geometry
resulted in a gap between FKBP12 and the inhibitor
which was larger than that detected previously for the
related acyclic ureas. However, favorable hydrophobic
interactions were maintained between the propyl side
chain of 17 and the FKBP12 pyranose-binding residues
(Phe-36, Ile-90, Ile-91, and His-87) with minimal per-
turbation of these residues from their observed positions
in the FKBP12-FK506 crystal structure. As was noted
for other benzyl pipecolyl-containing inhibitors, favor-
able inter- and intramolecular hydrophobic interactions
were made by the benzyl ester of 17 with Phe-46 of
FKBP12 and the propyl chain of the ligand, respectively.
inhibitors containing ester groups other than benzyl
were prepared by carbodiimide-mediated esterification
of (S)-N-Boc-pipecolinic acid30 followed by removal of the
nitrogen-protecting group and use of the resulting amine
in the unsymmetrical urea synthesis described above
(Scheme 1). The alcohol required for the preparation
of compound 7 was synthesized by the reported litera-
ture method,20b while those utilized in the syntheses of
8 and 9 were prepared by borane-mediated reduction
of the corresponding carboxylic acids.31 Interestingly,
all the acyclic ureas examined in the course of this work
(compounds 1-13) displayed only one significant isomer
1
in their H NMR spectra (CDCl3, 23 °C). The origin of
this apparent conformational bias is unknown as is the
distribution of acyclic urea isomers in solvents other
than chloroform (e.g., water).32
Syn th esis
The 6,6-bicyclic molecules utilized in this study were
somewhat more difficult to prepare than the acyclic
ureas described above. It was envisioned that cycliza-
tion of an appropriately positioned olefin onto a tran-
sient N-acyliminium ion intermediate with subsequent
trapping of the resulting carbocation would allow for
construction of the desired 6,6-bicyclic framework
(Scheme 2).33,34 Examination of molecular models sug-
gested that the cyclization which produced the undes-
ired bicyclic structure (pathway B) would be severely
disfavored by steric clashes between the ester and urea
carbonyl groups relative to that which afforded the
desired bicyclic framework (pathway A).33a Subsequent
functional group manipulation of the cyclization prod-
uct(s) would then provide the target molecules. The
preparation of ureas 14 and 15 in racemic form is
described below.
Treatment of 3-bromopropionaldehyde dimethyl ac-
etal with equimolar amounts of sodium hydride and
benzylamine in N,N-dimethylformamide at 80 °C for 5
h afforded the monoalkylation product 23 in moderate
yield following workup and flash column chromatogra-
phy (Scheme 3). This compound was efficiently coupled
with racemic allylglycine methyl ester (24) utilizing the
previously described method for unsymmetrical urea
preparation to afford urea 25 after purification on
neutral silica gel. Exposure of 25 to a 1:1 mixture of
The simple benzyl pipecolyl-derived FKBP12 inhibi-
tors were conveniently synthesized by modification of
a literature procedure for the preparation of unsym-
metrical ureas (Scheme 1).28 Thus, appropriate primary
amines were first transformed into the corresponding
isocyanates by addition, with an equimolar amount of
triethylamine, to a solution of triphosgene in dichlo-
romethane at 23 °C followed by a short (∼1 h) reflux
period. These intermediates were not isolated but
treated directly with (S)-pipecolinic acid benzyl ester20a
and an additional 1 equiv of triethylamine to provide
the desired ureas in high yield (70-95%) following
workup and flash column chromatography. In contrast
to the related literature procedure, slow (e.g., syringe
pump) addition of the primary amine substrates was
not required in order to obtain good yields of the ureas.
For the preparation of compound 4, the hydroxyl group
of 3-amino-2,2-dimethyl-1-propanol was converted to the
corresponding tert-butyldimethylsilyl ether prior to urea
formation. The silyl group was subsequently removed
in good yield by exposure of the protected urea to
triethylamine trihydrofluoride. The primary amines
required for the syntheses of compounds 5 and 6 were
prepared by literature methods.29
Attempted deprotection of the benzyl pipecolyl-
derived ureas by a variety of methods (e.g., hydrogena-

1878 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 9
Dragovich et al.
Sch em e 3a
desired 6,6-bicyclic ring system was confirmed by analy-
sis of the FKBP12-17 crystal structure (see above).
Con clu sion s
The ability of the urea functional group to replace the
R-ketoamide moiety present in potent FKBP12-binding
molecules was demonstrated by the synthesis and
biological evaluation of several acyclic and cyclic urea-
containing compounds. Such molecules disrupted
FKBP12 isomerase activity with apparent inhibition
constants (Ki,app) approaching 0.10 µM. X-ray crystal-
lographic analysis of several ureas complexed with
FKBP12 established that the compounds associated
with the protein in a manner similar to, but significantly
different from, that observed for related R-ketoamide-
containing FKBP12 inhibitors (including the natural
product FK506). The crystal structures also provided
insight for the possible extension of the simple FKBP12-
binding domains to include additional functionality
necessary to impart immunosuppressive properties to
this class of compounds.
Exp er im en ta l Section
a
Reagents and conditions (Bn ) benzyl): (a) 1.0 equiv of NaH,
1.0 equiv of BnNH2, DMF, 23 f 80 °C, 5 h, 36%; (b) 0.33 equiv of
(Cl3CO)2CO, 1.0 equiv of Et3N, CH2Cl2, 23 f 40 °C, 1 h, then 1.0
equiv of 23, 1.0 equiv of Et3N, 15 min, 88%; (c) 1:1 TFA:CH2Cl2,
23 °C, 2 h, 71%; (d) 6.0 equiv of ClC(S)O-p-tolyl, 6.5 equiv of
DMAP, CH3CN, 82 °C, 10 h, 50%; (e) 1.5 equiv of (TMS)3SiH, 0.2
equiv of AIBN, toluene, 80 °C, 1 h, 91%; (f) 10 equiv of NaOH,
CH3OH, 23 °C, 4 h, then 3.0 equiv of EDC, 3.0 equiv of BnOH,
0.15 equiv of DMAP, CH2Cl2, 23 °C, 1 h, 72%.
Gen er a l. All reactions were performed in septum-sealed
flasks under a slight positive pressure of argon unless other-
wise noted. All commercial reagents and solvents were used
as received from their respective suppliers with the following
exceptions. Tetrahydrofuran (THF) was distilled from sodium
benzophenone ketyl prior to use. Dichloromethane (CH2Cl2)
was distilled from calcium hydride prior to use. 2,2,2-Trif-
luoroethanol was distilled from sodium prior to use. The
succinyl-Ala-Leu-Pro-Phe-p-nitroanilide peptide used in en-
zyme assays was purchased from Bachem. Flash column
chromatography38 was performed using silica gel 60 (Merck
art. 9385). 1H NMR spectra were recorded at 300 MHz
utilizing either a Varian UNITYplus 300 or a General Electric
QE-300 spectrometer. Chemical shifts are reported in ppm
(δ) downfield relative to internal tetramethylsilane, and
coupling constants are given in Hertz. Infrared absorption
spectra were recorded using a Perkin-Elmer 1600 series FTIR
spectrometer. Continuous spectrophotometric enzyme assays
were performed using either a Beckman DU-12 or a Perkin-
Elmer Lambda 2 spectrophotometer. Elemental analyses were
performed by Atlantic Microlab, Inc., Norcross, GA. Melting
points are uncorrected.
P r otein -Liga n d Cr ysta l Str u ctu r e Deter m in a tion .
Wild-type human FKBP12 and an FKBP12 mutant (R13Q,
K17Q, K92Q) designed to facilitate crystallization were ex-
pressed in Escherichia coli and purified according to standard
procedures39 with the following modifications: Cells were
disrupted by microfluidization (Microfluidics Corp. Model
M110Y), a fast flow DEAE Sepharose (Pharmacia) column was
used instead of a DE-52 (Whatman) column, and two am-
monium sulfate cuts (46% with retention of supernatant, 86%
with retention of pellet) replaced the heat treatment step.
Samples of the ligands were dissolved in DMSO at a concen-
tration of 0.5 M and mixed in a 5:1 molar ratio with wild-type
(compounds 3, 4, 7, and 17) or mutant (compounds 8 and 9)
FKBP12. The FKBP12-ligand complexes were spun on a
desktop centrifuge at 15 000 rpm for 5 min to remove particu-
lates prior to use.
trifluoroacetic acid and dichloromethane at ambient
temperature for 2 h provided the 6,6-bicyclic compound
26 in 71% yield following a basic quench and flash
column chromatography. Prolonged reaction times
significantly reduced the yield of the cyclic urea, and
small amounts of additional uncharacterized products
were formed during the cyclization reaction.35 The
relative stereochemistry of 26 was established by an
X-ray crystal structure determination which confirmed
that the desired 6,6-bicyclic ring system had been
formed. In addition, the crystal structure indicated that
trapping of the intermediate carbocation occurred from
the same face as the ester moiety (cf. Scheme 2). It is
uncertain whether the observed alcohol stereochemistry
resulted from a directed intermolecular trapping event
or by an intramolecular cyclization involving the ester
group with subsequent methanolysis of the intermediate
lactone.
Attempts to deoxygenate 26 proved troublesome,
presumably due to steric effects resulting from the cis-
1,3-relationship of the alcohol and ester groups. The
desired transformation was eventually accomplished by
reducing the corresponding thionoformate 27 with tris-
(trimethylsilyl)silane/AIBN to afford methyl ester 14 in
moderate yield.36,37 Basic hydrolysis of 14 and esteri-
fication of the resulting carboxylic acid with benzyl
alcohol afforded benzyl ester 15 in good yield. The
preparation of the 6,6-bicyclic-N-propyl compounds 16
and 17 was analogous to that described for the N-benzyl
ureas above and proceeded through intermediates 28-
31 (see the Experimental Section). The stereochemistry
of the corresponding N-acyliminium ion cyclization
product 30 was assigned by analogy to the N-benzyl
series described above, and successful preparation of the
Crystals of the complexes were obtained by vapor diffusion
at room temperature. A given solution of FKBP12-ligand
complex (3 µL) was mixed with reservoir solution (3 µL) to
form the required hanging drops. Crystals of the complexes
were obtained using the following reservoir solutions: 2% poly-
(ethylene glycol) 400, 1.4 M Na/K phosphate, pH 7.5 (FKBP12-
3); 2.0 M sodium formate, 0.1 M bis-Tris, pH 6.0 (FKBP12-
4); 1.0 M sodium formate, 0.1 M HEPES, pH 7.0 (FKBP12-
7); 1.4 M Na/K phosphate, 0.1 M cacodylate, pH 6.5 (FKBP12-
8); 1.4 M ammonium phosphate, 0.1 M bis-Tris, pH 5.5

Structure-Based Design of Novel FKBP12 Inhibitors
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 9 1879
Ta ble 5. X-ray Data Collection and Refinement Statistics for FKBP12-Ligand Complexes
3
4
7
8
9
17
cell constants (Å):
a
b
c
31.47
43.10
77.76
31.36
43.61
77.79
31.52
43.66
77.77
33.46
36.59
92.51
33.25
36.44
93.37
31.44
43.39
77.66
data collection:
no. observations
no. unique reflections
resolution range (Å)
25 848
9029
20-1.8
0.051
0.88
30 819
12 282
20-1.65
0.031
28 315
9526
20-1.8
0.058
0.93
34 598
10 231
20-1.8
0.034
17 242
6639
20-2.1
0.064
0.94
44 047
9920
20-1.8
0.064
0.96
a
Rsym
completeness
refinement:
0.92
0.92
resolution range (Å)
10-1.8
0.173
0.014
2.62
9.33
10.18
115
10-1.65
0.185
0.013
2.73
12.73
12.88
72
10-1.8
0.190
0.014
2.65
10.75
11.98
120
10-1.8
0.192
0.015
2.96
26.17
20.69
63
10-2.1
0.184
0.016
2.95
30.49
21.97
37
10-1.8
0.179
0.014
2.67
13.20
15.02
85
final R factorb
rms bond dev (Å)
rms angle dev (deg)
av B, protein (Å2)
av B, ligand (Å2)
no. solvent molecules
a
b
∑h∑i|Ih,i - Å Ih |/h∑iIh,i, where Ih is the mean intensity for the i observations of reflection h. ∑|Fo - Fc|/∑Fo, where Fo and Fc are the
observed and calculated structure factor amplitudes, respectively.
(FKBP12-9); and 1.5 M ammonium sulfate, 0.1 M cacodylate,
pH 6.5 (FKBP12-17).
starting model. After initial refinement of the protein coor-
dinates, a model of 9 was fit to a difference electron density
map.
All X-ray data sets (Table 5) were collected at room tem-
perature using a San Diego Multiwire Systems area detector
with a Rigaku RU-200 copper rotating anode operating at 50
kV and 180 mA. Crystals of complexes formed with wild-type
FKBP12 grew in space group P212121 with one complex in the
asymmetric unit. Crystals of complexes formed with the
mutant FKBP12 grew in space group P212121 with altered cell
dimensions and also contained one complex per asymmetric
unit.
Refinement of all FKBP12-ligand complexes was performed
using a combination of simulated annealing and conventional
positional refinement along with refinement of individual
atomic temperature factors. The structures were checked and
corrected using a series of simulated-annealing omit maps.
Display of electron density and manual adjustment of the
model were performed using the program X-FIT.42 Water
molecules were assigned to 4σ peaks in difference electron
density maps when there was good hydrogen-bond geometry
to protein atoms. Refinement results are summarized in Table
5.
The structure of the FKBP12-3 complex was determined
by molecular replacement using the protein coordinates from
the FKBP12-FK506 crystal structure as a search model.15
Molecular replacement and refinement calculations were
carried out using the program X-PLOR.40 A cross-rotation
search was performed in 2.5° intervals using 15-4 Å resolution
data and a maximum cross-vector length of 40 Å. The rotation
corresponding to the highest peak in the search (6.4σ) was
optimized by Patterson correlation refinement and used to
En zym e Assa ys. Wild-type human recombinant FKBP12
was purified as described above. Peptidylprolyl isomerase
activity was determined spectrophotometrically as described
using 100 µM succinyl-Ala-Leu-Pro-Phe-p-nitroanilide as a
substrate.43 The substrate was dried in vacuo over P2O5 at
room temperature, dissolved in 2,2,2-trifluoroethanol contain-
ing 240 mM LiCl, and subsequently stored under argon at -20
°C. The percentage of cis-succinyl-Ala-Leu-Pro-Phe-p-nitro-
anilide present in the equilibrium mixture was usually about
45%, allowing significant portions of the reaction progress
curves to be obtained. Continuous spectrophotometric assays
were performed at 10 or 15 °C in the presence of 50 mM
HEPES (pH 8.0), 100 mM NaCl, 100 µM 2-mercaptoethanol,
1% 2,2,2-trifluoroethanol, 2.5 mM LiCl, 1% DMSO, 90 nM
FKBP12, 100 µM cis-peptide substrate, and 6 mg/mL R-chy-
motrypsin coupling enzyme. Data for reaction progress curves
were collected for 5 min, and initial velocities were determined
by nonlinear least-squares fits of each progress curve to a
single exponential. Apparent Ki’s were determined by analyz-
ing the concentration dependence of each inhibitor on prolyl
isomerase activity compared to a control that contained all
components of the reaction mixture except inhibitor. These
data were fit to an equation for competitive tight-binding
inhibition with a correction for uncatalyzed isomerization.
Standard errors for the fit data were <1%. All data were
analyzed using the software program KineTic (BioKin, Ltd.,
Madison, WI). The kcat and Km values determined for the
succinyl-Ala-Leu-Pro-Phe-p-nitroanilide substrate, purified
recombinant FKBP12, and FKBP12 mutant were similar to
those reported previously.43 Using these methods, Ki,app’s of
0.28 nM for FK506 and 0.20 nM for rapamycin were measured
in agreement with previous reports.43,44
generate starting coordinates for the translation search.
A
translation search over a 1.0 Å grid using 15-3 Å resolution
data produced a clear solution (peak height of 13.8σ). The
resulting model produced an R-factor of 0.379 for data (F>2σF)
in the resolution range 10-3.0 Å after rigid-body refinement.
The model was refined with the simulated annealing and
conjugate gradient minimization protocols in X-PLOR using
all data in the 10-1.8 Å resolution range. A model of 3,
constructed using the program Quanta,41 was then fit to a
difference electron density map, and the combined model was
refined.
Structures of FKBP12 complexed with compounds 4, 7, and
17 were determined using the protein coordinates from the
FKBP12-3 complex crystal structure as a starting model. In
each case, the ligand was fit to a difference electron density
map after initial refinement of the protein coordinates. The
structure of the FKBP12-8 complex was determined by
molecular replacement using the coordinates of FKBP12 from
the FKBP12-4 complex structure as a search model. The
rotation corresponding to the highest peak (6.8σ) in a cross-
rotation search using 15-4 Å data was optimized by Patterson
correlation refinement and used to generate starting coordi-
nates for the translation search. The translation function
solution (peak height of 11.9σ) produced a model with an
R-factor of 0.341 for all data in the resolution range 10-3.0 Å
after rigid-body refinement. This model was refined with the
conjugate gradient minimization protocol in X-PLOR using all
data in the 10-1.8 Å resolution range, and a model of 8 was
fit to a difference electron density map. The structure of the
FKBP12-9 complex was determined using the protein coor-
dinates from the FKBP12-8 complex crystal structure as the
Rep r esen ta tive P r oced u r e for Syn th esis of Acyclic
Ur ea s 1-6. (2S)-1-[(2-Meth yla llyl)ca r ba m oyl]p ip er id in e-
2-ca r boxylic Acid Ben zyl Ester (1). A solution of 2-me-
thylallylamine (0.099 g, 1.39 mmol, 1 equiv) and Et3N (0.194
mL, 1.39 mmol, 1.0 equiv) in CH2Cl2 (3 mL) was added over 2
min to a solution of triphosgene (0.137 g, 0.462 mmol, 0.33

1880 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 9
Dragovich et al.
equiv) in CH2Cl2 (5 mL) at 23 °C. The resulting mixture was
refluxed for 1 h and then cooled to 23 °C. A solution of (2S)-
piperidine-2-carboxylic acid benzyl ester20a (0.306 g, 1.39 mmol,
1.0 equiv) and Et3N (0.194 mL, 1.39 mmol, 1.0 equiv) in CH2-
Cl2 (5 mL) was then added over 5 min. The mixture was
stirred at 23 °C for 1 h and then partitioned between water
(100 mL) and a 1:1 mixture of EtOAc and hexanes (2 × 100
mL). The combined organic layers were dried over Na2SO4
and concentrated. Purification of the residue by flash column
chromatography (40% EtOAc in hexanes) afforded urea 1
(0.320 g, 73%) as a colorless oil: Rf ) 0.45 (50% EtOAc in
hexanes); IR (cm-1) 3354, 1738, 1630; 1H NMR (CDCl3, major
isomer) δ 1.19-1.28 (m, 1 H), 1.42-1.56 (m, 1 H), 1.65-1.71
(m, 2 H), 1.73 (s, 3 H), 2.26 (d, 1 H, J ) 14.0), 3.17 (td, 1 H, J
) 12.2, 3.2), 3.50-3.55 (m, 1 H), 3.80 (d, 2 H, J ) 5.6), 4.71-
4.74 (m, 1 H), 4.80 (s, 1 H), 4.83 (s, 1 H), 5.11-5.12 (m, 1 H),
5.12 (d, 1 H, J ) 12.4), 5.19 (d, 1 H, J ) 12.4), 7.28-7.38 (m,
5 H). Anal. (C18H24N2O3) C, H, N.
(2S )-1-[[(1′-Me t h ylcyclop e n t yl)m e t h yl]ca r b a m oyl]-
p ip er id in e-2-ca r boxylic a cid ben zyl ester (5): Rf ) 0.35
(40% EtOAc in hexanes); IR (cm-1) 3364, 1738, 1631; 1H NMR
(CDCl3, major isomer) δ 0.95 (s, 3 H), 1.25-1.56 (m, 10 H),
2.24 (s, br, 2 H), 3.18 (m, 3 H), 3.50 (m, 1 H), 3.75 (m, 1 H),
4.28 (m, 1 H), 4.62 (t, 1 H, J ) 5.9), 5.05 (m, 1 H), 5.12 (d, 1
H, J ) 12.5), 5.18 (d, 1 H, J ) 12.4), 7.36 (s, 5 H). Anal.
(C21H30N2O3) C, H, N.
(2S )-1-[[(1′-Me t h ylc yc loh e xyl)m e t h yl]c a r b a m oyl]-
p ip er id in e-2-ca r boxylic a cid ben zyl ester (6): Rf ) 0.66
(5% CH3OH in CH2Cl2); IR (cm-1) 3362, 1739, 1631; 1H NMR
(CDCl3, major isomer) δ 0.84 (s, 3 H), 1.17-1.30 (m, 6 H),
1.34-1.54 (m, 6 H), 1.60-1.73 (m, 3 H), 2.18-2.26 (m, 1 H),
3.06 (dd, 1 H, J ) 13.5, 5.9), 3.10-3.20 (m, 2 H), 3.43-3.51
(m, 1 H), 4.56-4.62 (m, 1 H), 5.04-5.08 (m, 1 H), 5.12 (d, 1 H,
J ) 12.4), 5.18 (d, 1 H, J ) 12.4), 7.29-7.34 (m, 5 H). Anal.
(C22H32N2O3) C, H, N.
Rep r esen ta tive P r oced u r e for Ester ifica tion of (S)-N-
Boc-p ip ecolin ic Acid . (2S)-P ip er id in e-1,2-d ica r boxylic
Acid 1-ter t-Bu tyl 2-[(1′R)-1′,3′-Dip h en ylp r op yl] Diester
(18). 1-[3-(Dimethylamino)propyl]-3-ethylcarbodiimide hy-
drochloride (0.418 g, 2.18 mmol, 1.5 equiv) and DMAP (0.025
g, 0.21 mmol, 0.14 equiv) were added sequentially to a solution
of (1R)-1,3-diphenylpropanol20b (0.309 g, 1.46 mmol, 1 equiv)
and (S)-N-Boc-pipecolinic acid
(2S)-1-(Ben zylca r ba m oyl)p ip er id in e-2-ca r boxylic a cid
ben zyl ester (2): Rf ) 0.21 (30% EtOAc in hexanes); IR (cm-1
)
1
3355, 1737, 1711, 1631; H NMR (CDCl3) δ 0.82-0.90 (m, 1
H), 1.19-1.28 (m, 1 H), 1.41-1.50 (m, 1 H), 1.65-1.76 (m, 2
H), 2.24-2.29 (m, 1 H), 3.15 (td, 1 H, J ) 12.3, 3.0), 3.47-
3.52 (m, 1 H), 4.42-4.45 (m, 2 H), 4.89 (t, 1 H, J ) 4.9), 5.12-
5.13 (m, 1 H), 5.13 (d, 1 H, J ) 12.4), 5.20 (d, 1 H, J ) 12.4),
7.24-7.40 (m, 10 H). Anal. (C21H24N2O3) C, H, N.
30 (0.500 g, 2.18 mmol, 1.5 equiv)
in CH2Cl2 (10 mL) at 23 °C. The resulting solution was stirred
for 3 h at 23 °C and then partitioned between water (100 mL)
and a 1:1 mixture of EtOAc and hexanes (2 × 100 mL). The
combined organic layers were dried over Na2SO4 and concen-
trated. The residue was purified by flash column chromatog-
raphy (5% EtOAc in hexanes) to provide ester 18 (0.574 g, 93%)
(2S)-1-[(2′,2′-Dim eth ylp r op yl)ca r ba m oyl]p ip er id in e-2-
ca r boxylic a cid ben zyl ester (3): Rf ) 0.31 (30% EtOAc in
hexanes); IR (cm-1) 3365, 1739, 1632; 1H NMR (CDCl3, major
isomer) δ 0.88 (s, 9 H), 1.42-1.57 (m, 1 H), 1.63-1.77 (m, 3
H), 2.20-2.28 (m, 1 H), 3.00-3.21 (m, 3 H), 3.47-3.52 (m, 1
H), 4.60-4.64 (m, 1 H), 5.08-5.10 (m, 1 H), 5.12 (d, 1 H, J )
12.6), 5.18 (d, 1 H, J ) 12.6), 7.28-7.37 (m, 5 H). Anal.
(C19H28N2O3) C, H, N.
(2S)-1-[(3′-H yd r oxy-2′,2′-d im et h ylp r op yl)ca r b a m oyl]-
p ip er id in e-2-ca r boxylic Acid Ben zyl Ester (4). tert-Bu-
tyldimethylsilyl chloride (12.50 g, 83.1 mmol, 1.0 equiv), Et3N
(12.71 mL, 91.0 mmol, 1.1 equiv), and DMAP (0.10 g, 0.83
mmol, 0.01 equiv) were added sequentially to a solution of 2,2-
dimethyl-3-aminopropanol (8.56 g, 83.1 mmol, 1.0 equiv) in
CH2Cl2 (100 mL) at 0 °C. The resulting mixture was main-
tained at 0 °C for 2 h and then partitioned between water (150
mL) and CH2Cl2 (2 × 50 mL). The combined organic layers
were washed sequentially with water (100 mL) and brine (100
mL) and then dried over Na2SO4 and concentrated. The
residue was purified by flash column chromatography (gradi-
ent elution 10% f 40% EtOAc in hexanes) to afford 3-[(tert-
butyldimethylsilanyl)oxy]-2,2-dimethylpropylamine (15.20 g,
84.1%) as a clear oil: Rf ) 0.30 (50% EtOAc in hexanes); IR
(cm-1) 3319; 1H NMR (CDCl3) δ 0.03 (s, 6 H), 0.79 (s, 6 H),
0.86 (s, 9 H), 1.10 (s, br, 2 H), 2.49 (s, 2 H), 3.28 (s, 2 H). Anal.
Calcd for C11H27NOSi: C, 60.77; H, 12.52; N, 6.44. Found:
C, 61.02; H, 12.37; N, 6.32.
The urea derived from (2S)-piperidine-2-carboxylic acid
benzyl ester20a and 3-[(tert-butyldimethylsilanyl)oxy]-2,2-dim-
ethylpropylamine was prepared utilizing the general procedure
described above: Rf ) 0.68 (40% EtOAc in hexanes); IR (cm-1)
3371, 1740, 1632; 1H NMR (CDCl3, major isomer) δ 0.53 (s, 6
H), 0.82 (s, 3 H), 0.84 (s, 3 H), 0.85 (s, 9 H), 1.23-1.62 (m, 8
H), 2.25 (m, 1 H), 3.19 (m, 2 H), 3.46 (s, 2 H), 5.02 (s, br, 1 H),
5.11 (d, 1 H, J ) 12.3), 5.19 (d, 1 H, J ) 12.3), 5.51 (t, 1 H, J
) 6.1), 7.29 (s, 5 H). Anal. Calcd for C25H42N2O4Si: C, 64.90;
H, 9.15; N, 6.05. Found: C, 64.78; H, 9.18; N, 5.98.
Triethylamine trihydrofluoride (0.60 mL) was added to a
solution of this urea (0.041 g, 0.09 mmol) in THF (3 mL) at 23
°C. The reaction mixture was stirred at 23 °C for 4 h and then
partitioned between water (50 mL) and EtOAc (2 × 50 mL).
The combined organic layers were washed with brine (50 mL)
and then dried over Na2SO4 and concentrated. The residue
was purified by flash column chromatography (30% EtOAc in
hexanes) to afford urea 4 (0.024 g, 76%) as a colorless oil: Rf
) 0.31 (40% EtOAc in hexanes); IR (cm-1) 3354, 1738, 1621;
1H NMR (CDCl3, major isomer) δ 0.82 (s, 3 H), 0.85 (s, 3 H),
1.19-1.75 (m, 8 H), 2.25 (m, 1 H), 2.92-3.25 (m, 4 H), 4.41 (t,
1 H, J ) 7.3), 4.99 (m, 1 H), 5.21 (s, 2 H), 7.33 (s, 5 H). Anal.
(C19H28N2O4) C, H, N.
as a colorless oil: Rf ) 0.74 (20% EtOAc in hexanes); IR (cm-1
)
1739, 1695; 1H NMR (CDCl3, 1:1 mixture of isomers) δ 1.05-
1.29 (m), 1.34 (s), 1.37-1.43 (m), 1.48 (s), 1.59-1.72 (m), 2.03-
2.34 (m), 2.53-2.69 (m), 2.81-2.98 (m), 3.91-4.05 (m), 4.74-
4.75 (m), 4.96-4.97 (m), 5.80 (dd, J ) 7.7, 5.8), 7.08-7.39 (m).
Anal. (C26H33NO4) C, H, N.
(2S)-P ip er id in e-1,2-d ica r boxylic a cid 1-ter t-bu tyl 2-[3′-
(3,4,5-tr im eth oxyp h en yl)p r op yl] d iester (19): Rf ) 0.65
(50% EtOAc in hexanes); IR (cm-1) 1738, 1694; 1H NMR
(CDCl3, 1:1 mixture of isomers) δ 0.86-0.89 (m), 1.20-1.38
(m), 1.62-1.74 (m), 2.20-2.24 (m), 2.63 (t, J ) 7.6), 2.84-3.03
(m), 3.83 (s), 3.85 (s), 4.11-4.17 (m), 4.75 (s, br), 4.91 (s, br),
6.40 (s). Anal. (C23H35NO7) C, H, N.
(2S)-P ip er id in e-1,2-d ica r boxylic a cid 1-ter t-bu tyl 2-[3′-
(3-m eth oxyp h en yl)p r op yl] d iester (20): Rf ) 0.60 (30%
EtOAc in hexanes); IR (cm-1) 1739, 1696; 1H NMR (CDCl3,
1:1 mixture of isomers) δ 1.21-1.40 (m), 1.45 (s), 1.47 (s), 1.53-
1.72 (m), 1.92-2.01 (m), 2.17-2.24 (m), 2.67 (t, J ) 7.8), 2.88-
2.98 (m), 3.80 (s), 3.92-4.08 (m), 4.16 (t, J ) 6.4), 4.74-4.76
(m), 4.90-4.91 (m), 6.73-7.18 (m), 7.19-7.23 (m). Anal.
(C21H31NO5) C, H, N.
(2S)-P ip er id in e-1,2-d ica r boxylic a cid 1-ter t-bu tyl 2-(3′-
p h en oxyben zyl) d iester (21): Rf ) 0.65 (30% EtOAc in
1
hexanes); IR (cm-1) 1742, 1697; H NMR (CDCl3, mixture of
isomers) δ 0.86-0.91 (m), 0.96-1.27 (m), 1.38 (s), 1.45 (s),
1.56-1.73 (m), 2.17-2.20 (m), 2.86-2.98 (m), 3.88-4.03 (m),
4.75 (s, br), 4.94 (s, br), 5.07-5.20 (m), 6.91-7.14 (m), 7.23-
7.38 (m). Anal. (C24
H29NO5) C, H, N.
(2S)-P ip er id in e-1,2-d ica r boxylic a cid 1-ter t-bu tyl 2-(4′-
p h en ylbu tyl) d iester (22): Rf ) 0.68 (30% EtOAc in hex-
1
anes); IR (cm-1) 1739, 1698; H NMR (CDCl3, 1:1 mixture of
isomers) δ 1.13-1.35 (m), 1.42 (s), 1.45 (s), 1.52-1.75 (m),
2.17-2.19 (m), 2.61-2.66 (m), 2.81-2.99 (m), 3.89-4.08 (m),
4.15 (s, br), 4.70 (s, br), 4.87 (s, br), 7.15-7.20 (m), 7.23-7.30
(m). Anal. (C21H31NO4) C, H, N.
Rep r esen ta tive P r oced u r e for th e Syn th esis of Acyclic
Ur ea s 7-13. (2S)-1-[(2′,2′-Dim eth ylp r op yl)ca r ba m oyl]-
p ip er id in e-2-ca r boxylic Acid (1′′R)-1′′,3′′-Dip h en ylp r op yl
Ester (7). A solution of HCl in 1,4-dioxane (4.0 M, 2.0 mL)
was added to a solution of 18 (0.570 g, 1.35 mmol) in the same
solvent (2 mL) at 23 °C. The reaction mixture was stirred for
2 h at 23 °C and then carefully partitioned between saturated
NaHCO3 (100 mL) and EtOAc (3 × 100 mL). The combined
organic layers were dried over Na2SO4 and concentrated to
afford the crude amine product (0.383 g, 88%). The free amine

Structure-Based Design of Novel FKBP12 Inhibitors
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 9 1881
thus obtained was immediately coupled with 2,2-dimethylpro-
pylamine utilizing the procedure described above for the
preparation of compounds 1-6 to afford urea 7 (77%) as a
hexanes (2 × 150 mL). The combined organic layers were
dried over Na2SO4 and concentrated. The residue was purified
by flash column chromatography (SiO2 pretreated with 5%
Et3N in hexanes, gradient elution 40% f 30% hexanes in
EtOAc) to afford acetal 23 (1.85 g, 36%) as a pale yellow oil:
Rf ) 0.34 (50% EtOAc in hexanes, SiO2 pretreated with 5%
Et3N in hexanes); IR (cm-1) 3331; 1H NMR (C6D6) δ 1.71-1.77
(m, 2 H), 2.57 (t, 2 H, J ) 6.8), 3.11 (s, 6 H), 3.57 (s, 2 H), 4.43
(t, 1 H, J ) 5.7), 7.09-7.21 (m, 3 H), 7.27-7.30 (m, 2 H). Anal.
(C12H19NO2) C, H, N.
colorless oil: Rf ) 0.30 (30% EtOAc in hexanes); IR (cm-1
)
3356, 1734, 1627; 1H NMR (CDCl3, major isomer) δ 0.88 (s, 9
H), 1.05-1.32 (m, 1 H), 1.43-1.57 (m, 1 H), 1.68-1.79 (m, 3
H), 2.02-2.14 (m, 1 H), 2.19-2.32 (m, 2 H), 2.51-2.69 (m, 2
H), 3.00 (dd, 1 H, J ) 13.3, 5.8), 3.07-3.18 (m, 2 H), 3.54 (dd,
1 H, J ) 12.1, 3.2), 4.59 (t, 1 H, J ) 5.9), 5.06-5.08 (m, 1 H),
5.79 (dd, 1 H, J ) 7.6, 6.0), 7.13-7.20 (m, 4 H), 7.24-7.34 (m,
6 H). Anal. (C27H36N2O3) C, H, N.
2-Am in op en t-4-en oic Acid Meth yl Ester (24). p-Tolu-
enesulfonic acid monohydrate (14.4 g, 75.7 mmol, 1.2 equiv)
was added to a solution of 2-aminopent-4-enoic acid (allylgly-
cine; 7.25 g, 62.9 mmol, 1 equiv) in CH3OH (125 mL) at 23 °C.
The reaction mixture was refluxed for 18 h and then cooled to
23 °C and concentrated. The colorless oil thus obtained was
partitioned between saturated aqueous NaHCO3 (150 mL) and
a 9:1 mixture of CH2Cl2 and CH3OH (3 × 150 mL). The
combined organic layers were dried over Na2SO4 and concen-
trated. The residue was distilled under reduced pressure to
afford 2-aminopent-4-enoic acid methyl ester (24) (5.00 g, 62%)
(2S)-1-[(2′,2′-Dim eth ylp r op yl)ca r ba m oyl]p ip er id in e-2-
ca r boxylic a cid 3′′-(3,4,5-tr im eth oxyp h en yl)p r op yl ester
(8): Rf ) 0.31 (50% EtOAc in hexanes); IR (cm-1) 3371, 1736,
1
1633; H NMR (CDCl3, major isomer) δ 0.91 (s, 9 H), 1.24-
1.31 (m, 1 H), 1.49-1.57 (m, 1 H), 1.67-1.78 (m, 3 H), 1.91-
2.00 (m, 2 H), 2.18-2.25 (m, 1 H), 2.62 (t, 2 H, J ) 7.7), 3.04
(dd, 1 H, J ) 13.4, 6.0), 3.12 (dd, 1 H, J ) 13.4, 6.3), 3.21 (td,
1 H, J ) 12.1, 3.3), 3.48-3.53 (m, 1 H), 3.83 (s, 3 H), 3.85 (s,
6 H), 4.15 (t, 2 H, J ) 6.5), 4.65 (t, 1 H, J ) 5.9), 5.04-5.05
(m, 1 H), 6.39 (s, 2 H). Anal. (C24H38N2O6) C, H, N.
(2S)-1-[(2′,2′-Dim eth ylp r op yl)ca r ba m oyl]p ip er id in e-2-
ca r boxylic a cid 3′′-(3-m eth oxyp h en yl)p r op yl ester (9): Rf
) 0.23 (30% EtOAc in hexanes); IR (cm-1) 3365, 1737, 1626;
1H NMR (CDCl3, major isomer) δ 0.90 (s, 9 H), 1.49-1.55 (m,
1 H), 1.67-1.77 (m, 3 H), 1.91-2.00 (m, 2 H), 2.21-2.25 (m, 1
H), 2.66 (t, 2 H, J ) 7.7), 2.98 (d, 1 H, J ) 6.3), 3.04 (dd, 1 H,
J ) 13.3, 6.1), 3.12 (dd, 1 H, J ) 13.3, 6.0), 3.19 (td, 1 H, J )
12.1, 3.1), 3.49-3.54 (m, 1 H), 3.80 (s, 3 H), 4.11-4.16 (m, 2
H), 4.64 (t, 1 H, J ) 5.8), 5.03-5.04 (m, 1 H), 6.73-6.78 (m, 3
H), 7.18-7.23 (m, 1 H). Anal. (C22H34N2O4) C, H, N.
(2S)-1-[(2′,2′-Dim eth ylp r op yl)ca r ba m oyl]p ip er id in e-2-
ca r boxylic a cid 3-p h en oxyben zyl ester (10): Rf ) 0.30
(30% EtOAc in hexanes); IR (cm-1) 3365, 1740, 1632; 1H NMR
(CDCl3, major isomer) δ 0.88 (s, 9 H), 1.09-1.27 (m, 1 H),
1.43-1.48 (m, 1 H), 1.50-1.57 (m, 1 H), 1.61-1.76 (m, 2 H),
2.19-2.25 (m, 1 H), 2.99-3.19 (m, 3 H), 3.32-3.48 (m, 1 H),
4.62 (t, 1 H, J ) 5.9), 5.06-5.10 (m, 2 H), 5.16 (d, 1 H, J )
12.8), 6.93-7.15 (m, 5 H), 7.26-7.38 (m, 4 H). Anal.
(C25H32N2O4) C, H, N.
(2S)-1-[(2′,2′-Dim eth ylp r op yl)ca r ba m oyl]p ip er id in e-2-
ca r boxylic a cid 4′′-p h en ylbu tyl ester (11): Rf ) 0.25 (30%
EtOAc in hexanes); IR (cm-1) 3364, 1736, 1631; 1H NMR
(CDCl3, major isomer) δ 0.90 (s, 9 H), 1.22-1.31 (m, 2 H),
1.47-1.54 (m, 1 H), 1.64-1.74 (m, 5 H), 2.19-2.24 (m, 1 H),
2.62-2.66 (m, 2 H), 3.05-3.11 (m, 3 H), 3.17 (td, 1 H, J )
12.1, 2.9), 3.48-3.52 (m, 1 H), 4.13-4.15 (m, 2 H), 4.62 (t, 1
H, J ) 5.4), 5.02 (d, 1 H, J ) 3.3), 7.16-7.21 (m, 3 H), 7.26-
7.31 (m, 2 H). Anal. (C22H34N2O3) C, H, N.
as a colorless liquid: bp ) 69-70 °C, 1.1 mmHg; IR (cm-1
)
3379, 1740; 1H NMR (CDCl3) δ 2.33-2.43 (m, 1 H), 2.46-2.55
(m, 1 H), 3.557 (dd, 1 H, J ) 7.2, 5.1), 3.73 (s, 3 H), 5.12 (s, 1
H), 5.16-5.19 (m, 1 H), 5.68-5.80 (m, 1 H). Anal. (C6H11
NO2) C, H, N.
-
2-[3′-Ben zyl-3′-(3′′,3′′-d im eth oxyp r op yl)u r eid o]p en t-4-
en oic Acid Meth yl Ester (25). A solution of 2-aminopent-
4-enoic acid methyl ester (24) (1.05 g, 8.13 mmol, 1.0 equiv)
and Et3N (1.13 mL, 8.11 mmol, 1.0 equiv) in CH2Cl2 (10 mL)
was added via cannula to a solution of triphosgene (0.803 g,
2.71 mmol, 0.33 equiv) in CH2Cl2 (50 mL) at 23 °C. The
reaction mixture was stirred for 15 min at 23 °C and then
refluxed for 1.5 h. After cooling to 23 °C, a solution of 23 (1.70
g, 8.12 mmol, 1 equiv) and Et3N (1.13 mL, 8.11 mmol, 1.0
equiv) in CH2Cl2 (10 mL) was added via cannula. The
resulting solution was stirred at 23 °C for 15 min and then
partitioned between water (150 mL) and a 1:1 mixture of
EtOAc and hexanes (2 × 150 mL). The combined organic
layers were dried over Na2SO4 and concentrated. Purification
of the residue by flash column chromatography (SiO2 pre-
treated with 5% Et3N in hexanes, gradient elution 30% f 50%
EtOAc in hexanes) provided urea 25 (2.61 g, 88%) as a colorless
oil: Rf ) 0.24 (30% EtOAc in hexanes, SiO2 pretreated with
5% Et3N in hexanes); IR (cm-1) 3357, 1745, 1646; 1H NMR
(C6D6) δ 1.64-1.77 (m, 2 H), 2.38-2.58 (m, 2 H), 3.07 (s, 3 H),
3.08 (s, 3 H), 3.11-3.25 (m, 2 H), 3.28 (s, 3 H), 4.30-4.48 (m,
3 H), 4.83-4.98 (m, 3 H), 5.58-5.72 (m, 2 H), 7.01-7.15 (m, 3
H), 7.21-7.24 (m, 2 H). Anal. (C19H28N2O5) C, H, N.
(2S )-1-[[(1′-Me t h ylc yc loh e xyl)m e t h yl]c a r b a m oyl]-
p ip er id in e-2-ca r boxylic a cid 3-p h en oxyben zyl ester (12):
Rf ) 0.13 (20% EtOAc in hexanes); IR (cm-1) 3357, 1741, 1632;
1H NMR (CDCl3, major isomer) δ 0.85 (s, 3 H), 1.14-1.34 (m,
6 H), 1.35-1.58 (m, 6 H), 1.60-1.77 (m, 3 H), 2.16-2.25 (m, 1
H), 3.04 (dd, 1 H, J ) 13.4, 5.9), 3.08-3.21 (m, 2 H), 3.42-
3.50 (m, 1 H), 4.55-4.62 (m, 1 H), 5.05-5.12 (m, 2 H), 5.15 (d,
1 H, J ) 12.4), 6.92-7.15 (m, 6 H), 7.25-7.37 (m, 3 H). Anal.
(C28H36N2O4) C, H, N.
(2S )-1-[[(1′-Me t h ylc yc loh e xyl)m e t h yl]c a r b a m oyl]-
p ip er id in e-2-ca r boxylic a cid 4′′-p h en ylbu tyl ester (13):
Rf ) 0.66 (5% CH3OH in CH2Cl2); IR (cm-1) 3372, 1738, 1631;
1H NMR (CDCl3, major isomer) δ 0.87 (s, 3 H), 1.21-1.33 (m,
6 H), 1.35-1.56 (m, 6 H), 1.62-1.75 (m, 7 H), 2.16-2.24 (m, 1
H), 2.60-2.66 (m, 2 H), 3.05 (dd, 1 H, J ) 13.4, 5.9), 3.10-
3.21 (m, 2 H), 3.45-3.53 (m, 1 H), 4.10-4.16 (m, 2 H), 4.56-
4.61 (m, 1 H), 4.98-5.02 (m, 1 H), 7.14-7.21 (m, 3 H), 7.25-
7.31 (m, 2 H). Anal. (C25H38N2O3) C, H, N.
tr a n s-2-Ben zyl-6-h yd r oxy-1-oxoocta h yd r op yr id o[1,2-
c]p yr im id in e-8-ca r boxylic Acid Meth yl Ester (26). Tri-
fluoroacetic acid (5 mL) was added to a solution of urea 25
(1.10 g, 3.02 mmol) in CH2Cl2 (5 mL) at 23 °C. The reaction
mixture was stirred for 2 h at 23 °C and then concentrated.
The resulting yellow oil was carefully partitioned between
saturated aqueous NaHCO3 (100 mL) and EtOAc (4 × 100
mL). The combined organic layers were dried over Na2SO4
and concentrated. The residue was purified by flash column
chromatography (5% CH3OH in CH2Cl2) to afford alcohol 26
(0.682 g, 71%) as a white solid: mp ) 144 °C; Rf ) 0.14 (30%
1
hexanes in EtOAc); IR (cm-1) 3394 (br), 1740, 1616; H NMR
(CDCl3) δ 1.40-1.49 (m, 1 H), 1.60 (d, 1 H, J ) 2.2), 1.75-
1.99 (m, 4 H), 2.51 (dq, 1 H, J ) 14.6, 2.4), 3.09 (dq, 1 H, J )
11.7, 2.5), 3.36 (td, 1 H, J ) 11.8, 3.8), 3.75 (s, 3 H), 3.86-3.95
(m, 1 H), 4.16-4.19 (m, 1 H), 4.55 (d, 1 H, J ) 15.3), 4.66 (d,
1 H, J ) 15.3), 5.27 (dd, 1 H, J ) 6.8, 1.8), 7.22-7.36 (m, 5 H).
Anal. (C17H22N2O4) C, H, N.
Ben zyl(3,3-d im eth oxyp r op yl)a m in e (23). Sodium hy-
dride (0.983 g of a 60% dispersion in mineral oil, 24.6 mmol,
1.0 equiv) and 3-bromopropionaldehyde dimethyl acetal (5.0
g, 24.6 mmol, 1.0 equiv) were added sequentially to a solution
of benzylamine (2.69 mL, 24.6 mmol, 1 equiv) in DMF (60 mL)
at 23 °C. The resulting suspension was stirred for 3 h at 23
°C and then maintained at 80 °C for an additional 4 h. The
clear reaction mixture was cooled to 23 °C and partitioned
between water (150 mL) and a 1:1 mixture of EtOAc and
tr a n s-2-Ben zyl-1-oxo-6-[[(p-tolyloxy)th iocar bon yl]oxy]-
oct a h yd r op yr id o[1,2-c]p yr im id in e-8-ca r b oxylic
Acid
Meth yl Ester (27). O-p-Tolyl chlorothionoformate (1.74 mL,
11.3 mmol, 6.0 equiv) and DMAP (1.50 g, 12.3 mmol, 6.5 equiv)
were added sequentially to a solution of alcohol 26 (0.600 g,
1.88 mmol, 1 equiv) in CH3CN (25 mL) at 23 °C. The resulting
cloudy suspension was refluxed for 10 h and then cooled to 23
°C and partitioned between water (100 mL) and EtOAc (4 ×
100 mL). The combined organic layers were dried over Na2-

1882 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 9
Dragovich et al.
SO4 and concentrated. Flash chromatographic purification of
the residue (gradient elution 30% f 50% EtOAc in hexanes)
afforded thionoformate 27 (0.444 g, 50%) as a white solid: mp
1.46 (m, 1 H), 1.50-1.62 (m, 2 H), 1.72-1.99 (m, 4 H), 2.47
(dq, 1 H, J ) 14.3, 2.5), 3.12-3.27 (m, 2 H), 3.34-3.49 (m, 2
H), 3.71 (s, 3 H), 3.83-3.91 (m, 1 H), 4.14-4.16 (m, 1 H), 5.20
(dd, 1 H, J ) 6.8, 1.7). Anal. (C13H22N2O4) C, H, N.
) 146-149 °C; Rf ) 0.53 (50% EtOAc in hexanes); IR (cm-1
)
1
1740, 1637; H NMR (CDCl3) δ 1.55-1.65 (m, 1 H), 1.86 (qd,
1 H, J ) 11.9, 4.7), 1.95-2.06 (m, 2 H), 2.16-2.21 (m, 1 H),
2.37 (s, 3 H), 2.96-3.03 (m, 1 H), 3.10-3.16 (m, 1 H), 3.41 (td,
1 H, J ) 11.9, 3.7), 3.80 (s, 3 H), 3.89-3.99 (m, 1 H), 4.54 (d,
1 H, J ) 15.3), 4.72 (d, 1 H, J ) 15.3), 5.40 (dd, 1 H, J ) 6.8,
1.5), 5.56-5.58 (m, 1 H), 6.98 (dd, 2 H, J ) 6.7, 1.7), 7.20-
7.36 (m, 7 H). Anal. (C25H28N2O5S) C, H, N.
tr a n s-1-Oxo-2-pr opyl-6-[[(p-tolyloxy)th iocar bon yl]oxy]-
oct a h yd r op yr id o[1,2-c]p yr im id in e-8-ca r b oxylic
a cid
m eth yl ester (31): mp ) 170 °C dec; Rf ) 0.32 (50% EtOAc
in hexanes); IR (cm-1) 1744, 1639; 1H NMR (CDCl3) δ 0.91 (t,
3 H, J ) 7.5), 1.52-1.66 (m, 3 H), 1.76-1.90 (m, 1 H), 1.94-
2.07 (m, 2 H), 2.18 (dq, 1 H, J ) 14.3, 2.8), 2.37 (s, 3 H), 2.96
(dq, 1 H, J ) 15.0, 2.5), 3.15-3.28 (m, 2 H), 3.39-3.54 (m, 2
H), 3.76 (s, 3 H), 3.90 (tt, 1 H, J ) 11.4, 3.2), 5.35 (dd, 1 H, J
) 6.9, 1.9), 5.53-5.57 (m, 1 H), 6.98 (d, 2 H, J ) 8.4), 7.21 (d,
2 H, J ) 8.4). Anal. (C21H28N2O5S) C, H, N.
tr a n s-1-Oxo-2-pr opyloctah ydr opyr ido[1,2-c]pyr im idin e-
8-ca r boxylic a cid m eth yl ester (16): Rf ) 0.16 (30% EtOAc
in hexanes); IR (cm-1) 1741, 1635; 1H NMR (CDCl3) δ 0.89 (t,
1 H, J ) 7.3), 1.12-1.32 (m, 2 H), 1.49-1.84 (m, 6 H), 1.95
(dq, 1 H, J ) 13.1, 3.4), 2.21-2.27 (m, 1 H), 3.13 (ddd, 1 H, J
) 11.5, 5.0, 2.8), 3.19-3.52 (m, 4 H), 3.71 (s, 3 H), 5.24 (dd, 1
H, J ) 6.2, 1.9). Anal. (C13H22N2O3) C, H, N.
tr a n s-1-Oxo-2-pr opyloctah ydr opyr ido[1,2-c]pyr im idin e-
8-ca r boxylic a cid ben zyl ester (17): Rf ) 0.24 (30% EtOAc
in hexanes); IR (cm-1) 1737, 1636; 1H NMR (CDCl3) δ 0.87 (t,
3 H, J ) 7.5), 1.15-1.25 (m, 2 H), 1.45-1.81 (m, 6 H), 1.90
(dq, 1 H, J ) 13.1, 3.2), 2.25-2.31 (m, 1 H), 3.10 (ddd, 1 H, J
) 11.7, 4.9, 2.5), 3.17-3.41 (m, 4 H), 5.11 (d, 1 H, J ) 12.5),
5.19 (d, 1 H, J ) 12.5), 5.32 (dd, 1 H, J ) 6.4, 2.0), 7.30-7.36
(m, 5 H). Anal. (C19H26N2O3) C, H, N.
tr a n s-2-Ben zyl-1-oxooctah ydr opyr ido[1,2-c]pyr im idin e-
8-ca r boxylic Acid Meth yl Ester (14). Tris(trimethylsilyl)-
silane (0.439 mL, 1.42 mmol, 1.5 equiv) and 2,2′-azobis(2-
methylpropionitrile) (AIBN; 0.031 g, 0.189 mmol, 0.20 equiv)
were added sequentially to a solution of thionoformate 27
(0.444 g, 0.948 mmol, 1 equiv) in toluene (50 mL) at 23 °C.
The resulting solution was deoxygenated by alternately evacu-
ating the reaction flask and filling with argon (10 cycles) and
then heated to 80 °C for 1 h. After cooling to 23 °C, the
reaction mixture was partitioned between water (100 mL) and
CH2Cl2 (3 × 120 mL). The combined organic layers were dried
over Na2SO4 and concentrated. Purification of the residue by
flash column chromatography (gradient elution 20% f 40%
EtOAc in hexanes) provided 14 (0.260 g, 91%) as a colorless
oil: Rf ) 0.21 (30% EtOAc in hexanes); IR (cm-1) 1739, 1635;
1H NMR (CDCl3) δ 1.19-1.30 (m, 2 H), 1.64-1.84 (m, 4 H),
1.87-1.94 (m, 1 H), 2.26-2.30 (m, 1 H), 3.07 (dq, 1 H, J )
11.6, 2.5), 3.31 (td, 1 H, J ) 11.9, 4.1), 3.48-3.55 (m, 1 H),
3.75 (s, 3 H), 4.59 (s, 2 H), 5.31 (dd, 1 H, J ) 6.2, 1.8), 7.22-
7.35 (m, 5 H). Anal. (C17H22N2O3) C, H, N.
Ack n ow led gm en t. We are grateful for many helpful
discussions throughout the course of this work with
Prof. Henry Rappoport, Drs. Robert Babine, Wesley
Chong, Thomas Hendrickson, Susumu Katoh, Hiroshi
Kawakami, Nobuyuki Okajima, Anna Tempczyk, and
Leora Zalman, Mr. Hiroki Tada, Ms. Angelica Linton,
and Mr. Ted Bleckman. We also thank Dr. Raj K.
Chadha of The Scripps Research Institute for obtaining
the X-ray crystal structure of compound 26.
tr a n s-2-Ben zyl-1-oxooctah ydr opyr ido[1,2-c]pyr im idin e-
8-ca r boxylic Acid Ben zyl Ester (15). Sodium hydroxide
(0.185 g, 4.63 mmol, 10 equiv) was added to a solution of 14
(0.140 g, 0.463 mmol, 1 equiv) in a 4:1 mixture of CH3OH and
water (5 mL) at 23 °C. The resulting suspension was stirred
for 4 h at 23 °C and then partitioned between aqueous HCl (1
M, 100 mL) and EtOAc (4 × 100 mL). The combined organic
layers were dried over Na2SO4 and concentrated to provide
the corresponding carboxylic acid (0.105 g, 78% crude yield).
1-[3-(Dimethylamino)propyl]-3-ethylcarbodiimide hydrochlo-
ride (0.209 g, 1.09 mmol, 3 equiv), benzyl alcohol (0.113 mL,
1.09 mmol, 3 equiv), and DMAP (0.020 g, 0.163 mmol, 0.15
equiv) were added sequentially to a solution of the crude
carboxylic acid obtained above (0.105 g, 0.364 mmol, 1 equiv)
in CH2Cl2 at 23 °C. The resulting suspension was stirred for
1 h at 23 °C and then partitioned between water (100 mL)
and a 1:1 mixture of EtOAc and hexanes (2 × 120 mL). The
combined organic layers were dried over Na2SO4 and concen-
trated. The residue was purified by flash column chromatog-
raphy (gradient elution 15% f 30% EtOAc in hexanes) to
afford 15 (0.100 g, 72%) as a colorless oil: Rf ) 0.30 (30%
Su p p or tin g In for m a tion Ava ila ble: ORTEP diagram of
compound 26 with relevant crystal structure data (7 pages).
Ordering information is given on any current masthead page.
Refer en ces
(1) Tanaka, H.; Kuroda, A.; Marusawa, H.; Hatanaka, H.; Kino, T.;
Goto, T.; Hashimoto, M.; Taga, T. Structure of FK506: A Novel
Immunosuppressant Isolated from Streptomyces. J . Am. Chem.
Soc. 1987, 109, 5031-5033.
(2) For recent reviews on this class of immunosuppressants, see:
(a) Schreiber, S. L.; Albers, M. W.; Brown, E. J . The Cell Cycle,
Signal Transduction, and Immunophilin-Ligand Complexes. Acc.
Chem. Res. 1993, 26, 412-420. (b) Armistead, D. M.; Harding,
M. W. Immunophilins and Immunouppressive Drug Action.
Annu. Rep. Med. Chem. 1993, 28, 207-215. (c) Belshaw, P. J .;
Meyer, S. D.; J ohnson, D. D.; Romo, D.; Ikeda, Y.; Andrus, M.;
Alberg, D. G.; Schultz, L. W.; Clardy, J .; Schreiber, S. L.
Synthesis, Structure and Mechanism in Immunophilin Research.
Synlett 1994, 381-392. (d) Wiederrecht, G.; Etzkorn, F. The
Immunophilins. Perspect. Drug Discovery Des. 1994, 2, 57-84.
(e) O’Keefe, S. J .; O’Neill, E. A. Cyclosporin A and FK506:
Immunosuppression, Inhibition of Transcription and the Role
of Calcineurin. Perspect. Drug Discovery Des. 1994, 2, 85-102.
(3) (a) Siekierka, J . J .; Hung, S. H. Y.; Poe, M.; Lin, C. S.; Sigal, N.
H. A Cytosolic-Binding Protein for the Immunosuppressant
FK506 has Isomerase Activity but is Distinct from Cyclophilin.
Nature 1989, 341, 755-757. (b) Harding, M. W.; Galat, A.;
Uehling, D. E.; Schreiber, S. L. A Receptor for the Immunosup-
pressant FK506 is a Cis-Trans Peptidyl-Prolyl Isomerase. Nature
1989, 341, 758-760. (c) Siekierka, J . J .; Wiederrecht, G.;
Greulich, H.; Boulton, D.; Hung, S. H. Y.; Cryan, J .; Hodges, P.
J .; Sigal, N. H. The Cytosolic-Binding Protein for the Immuno-
suppressant FK506 is Both a Ubiquitous and Highly Conserved
Peptidyl-Prolyl Cis-Trans Isomerase. J . Biol. Chem. 1990, 265,
21011-21015.
1
EtOAc in hexanes); IR (cm-1) 1736, 1634; H NMR (CDCl3) δ
1.13-1.30 (m, 2 H), 1.65-1.91 (m, 5 H), 2.29-2.34 (m, 1 H),
3.06 (dq, 1 H, J ) 11.6, 2.5), 3.26 (td, 1 H, J ) 11.8, 4.0), 3.45-
3.52 (m, 1 H), 4.52 (d, 1 H, J ) 15.2), 4.65 (d, 1 H, J ) 15.2),
5.16 (d, 1 H, J ) 12.4), 5.22 (d, 1 H, J ) 12.4), 5.36-5.38 (m,
1 H), 7.21-7.38 (m, 10 H). Anal. (C23H26N2O3) C, H, N.
(3,3-Dim eth oxypr opyl)pr opylam in e (28): IR (cm-1) 3324;
1H NMR (C6D6) δ 0.85 (t, 3 H, J ) 7.4), 1.33-1.41 (m, 2 H),
1.74-1.81 (m, 2 H), 2.40 (t, 2 H, J ) 7.0), 2.60 (t, 2 H, J )
6.8), 3.14 (s, 6 H), 4.49 (t, 1 H, J ) 5.7). Anal. (C8H19NO2) C,
H, N.
2-[3′-(3′′,3′′-Dim eth oxyp r op yl)-3′-p r op ylu r eid o]p en t-4-
en oic a cid m eth yl ester (29): Rf ) 0.21 (30% EtOAc in
hexanes, SiO2 pretreated with 5% Et3N in hexanes); IR (cm-1
)
3356, 1745, 1642; 1H NMR (C6D6) δ 0.73 (t, 3 H, J ) 7.4), 1.42-
1.49 (m, 2 H), 1.76-1.85 (m, 2 H), 2.38-2.61 (m, 2 H), 2.96-
3.22 (m, 4 H), 3.12 (s, 3 H), 3.13 (s, 3 H), 3.26 (s, 3 H), 4.39 (t,
1 H, J ) 5.7), 4.84-4.90 (m, 1 H), 4.94-5.03 (m, 2 H), 5.45 (d,
br, 1 H, J ) 7.7), 5.66-5.77 (m, 1 H). Anal. (C15H28N2O5) C,
H, N.
(4) Additional members of a group of proteins which bind FK506
have been identified and collectively named FKBPs. The 12 kDa
protein originally termed FKBP is currently referred to as
FKBP12. For the characterization of other (higher molecular
weight) members of this family of proteins, see: Baughman, G.;
tr a n s-6-Hyd r oxy-1-oxo-2-p r op ylocta h yd r op yr id o[1,2-
c]p yr im id in e-8-ca r boxylic a cid m eth yl ester (30): mp )
122-123 °C; Rf ) 0.09 (30% hexanes in EtOAc); IR (cm-1) 3388
(br), 1741, 1617; 1H NMR (CDCl3) δ 0.90 (t, 3 H, J ) 7.4), 1.37-

Structure-Based Design of Novel FKBP12 Inhibitors
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 9 1883
(16) (a) Griffith, J . P.; Kim, J . L.; Kim, E. E.; Sintchak, M. D.;
Thomson, J . A.; Fitzgibbon, M. J .; Fleming, M. A.; Caron, P. R.;
Hsiao, K.; Navia, M. A. X-ray Structure of Calcineurin Inhibited
by the Immunophilin-Immunosuppressant FKBP12-FK506 Com-
plex. Cell 1995, 82, 507-522. (b) Kissinger, C. R.; Parge, H. E.;
Knighton, D. R.; Lewis, C. T.; Pelletier, L. A.; Tempczyk, A.;
Kalish, V. J .; Tucker, K. D.; Showalter, R. E.; Moomaw, E. W.;
Gastinel, L. N.; Habuka, N.; Chen, X.; Maldonado, F.; Barker,
J . E.; Bacquet, R.; Villafranca, J . E. Crystal Structures of Human
Calcineurin and the Human FKBP12-FK506-Calcineurin Com-
plex. Nature 1995, 378, 641-644.
(17) For a discussion of the possible intricacies of calcineurin inhibi-
tion by the FKBP12-FK506 complex, see refs 15d and 16a.
(18) For an excellent review of FK506 structure-activity studies,
see: Goulet, M. T.; Rupprecht, K. M.; Sinclair, P. J .; Wyvratt,
M. J .; Parsons, W. H. The Medicinal Chemistry of FK506.
Perspect. Drug Discovery Des. 1994, 2, 145-162.
(19) (a) Aldape, R. A.; Futer, O.; DeCenzo, M. T.; J arrett, B. P.;
Murcko, M. A.; Livingston, D. J . Charged Surface Residues of
FKBP12 Participate in Formation of the FKBP12-FK506-Cal-
cineurin Complex. J . Biol. Chem. 1992, 267, 16029-16032. (b)
Yang, D.; Rosen, M. K.; Schreiber, S. L. A Composite FKBP12-
FK506 Surface That Contacts Calcineurin. J . Am. Chem. Soc.
1993, 115, 819-820. (c) Futer, O.; DeCenzo, M. T.; Aldape, R.
A.; Livingston, D. J . FK506 Binding Protein Mutational Analy-
sis. J . Biol. Chem. 1995, 270, 18935-18940.
Wiederrecht, G. J .; Campbell, N. F.; Martin, M. M.; Bourgeois,
S. FKBP51, a Novel T-Cell Specific Immunophilin Capable of
Calcineurin Inhibition. Mol. Cell. Biol. 1995, 15, 4395-4402 and
references therein. See also: Fischer, G. Peptidyl-Prolyl cis/
trans Isomerases and Their Effectors. Angew. Chem., Int. Ed.
Engl. 1994, 33, 1415-1436.
(5) (a) Rosen, M. K.; Standaert, R. F.; Galat, A.; Nakatsuka, M.;
Schreiber, S. L. Inhibition of FKBP Rotamase Activity by
Immunosuppressant FK506: Twisted Amide Surrogate. Science
1990, 248, 863-866. (b) Albers, M. W.; Walsh, C. T.; Schreiber,
S. L. Substrate Specificity for the Human Rotamase FKBP: A
View of FK506 and Rapamycin as Leucine-(Twisted Amide)-
Proline Mimics. J . Org. Chem. 1990, 55, 4984-4986.
(6) Ru¨egger, A.; Kuhn, M.; Lichti, H.; Loosli, H.-R.; Huguenin, R.;
Quiquerez, C.; von Wartburg, A. Cyclosporin A, a Peptide
Metabolite from Trichoderma polysporum (LINK ex PERS.)
Rifai, with a Remarkable Immunosuppressive Activity. Helv.
Chim. Acta 1976, 59, 1075-1092.
(7) (a) Takahashi, N.; Hayano, T.; Suzuki, M. Peptidyl-Prolyl cis-
trans Isomerase is the Cyclosporin A-Binding Protein Cyclophi-
lin. Nature 1989, 337, 473-475. (b) Fischer, G.; Wittman-
Liebold, B.; Lang, K.; Kiefhaber, T.; Schmid, F. X. Cyclophilin
and Peptidyl-Prolyl cis-trans Isomerase are Probably Identical
Proteins. Nature 1989, 337, 476-478.
(8) Findlay, J . A.; Radics, L. On the Chemistry and High Field
Nuclear Magnetic Resonance Spectroscopy of Rapamycin. Can.
J . Chem. 1980, 58, 579-590.
(20) (a) Armistead, D. A.; Boger, J . S.; Meyers, H. V.; Saunders, J .
O.; Tung, R. D. Immunosuppressive Compounds. U.S. Patent 5
192 773, 1993. (b) Holt, D. A.; Luengo, J . I.; Yamashita, D. S.;
Oh, H.-J .; Konialian, A. L.; Yen, H.-K.; Rozamus, L. W.; Brandt,
M.; Bossard, M. J .; Levy, M. A.; Eggleston, D. S.; Liang, J .;
Schultz, L. W.; Stout, T. J .; Clardy, J . Design, Synthesis, and
Kinetic Evaluation of High-Affinity FKBP Ligands and the X-ray
Crystal Structures of Their Complexes with FKBP12. J . Am.
Chem. Soc. 1993, 115, 9925-9938. (c) Holt, D. A.; Konialian-
Beck, A. L.; Oh, H.-J .; Yen, H.-K.; Rozamus, L. W.; Krog, A. J .;
Erhard, K. F.; Ortiz, E.; Levy, M. A.; Brandt, M.; Bossard, M.
J .; Luengo, J . I. Structure-Activity Studies of Synthetic FKBP
Ligands as Peptidyl-Prolyl Isomerase Inhibitors. Bioorg. Med.
Chem. Lett. 1994, 4, 315-320. (d) Luengo, J . I.; Konialian-Beck,
A.; Levy, M. A.; Brandt, M.; Eggleston, D. S.; Holt, D. A.
Synthesis and Structure-Activity Relationships of Macrocyclic
FKBP Ligands. Bioorg. Med. Chem. Lett. 1994, 4, 321-324. (e)
Yamashita, D. S.; Oh, H.-J .; Yen, H.-K.; Bossard, M. J .; Brandt,
M.; Levy, M. A.; Newman-Tarr, T.; Badger, A.; Luengo, J . I.;
Holt, D. A. Design, Synthesis and Evaluation of Dual Domain
FKBP Ligands. Bioorg. Med. Chem. Lett. 1994, 4, 325-328. (f)
Wang, G. T.; Lane, B.; Fesik, S. W.; Petros, A.; Luly, J .; Krafft,
G. A. Synthesis and FKBP Binding of Small Molecule Mimics
of the Tricarbonyl Region of FK506. Bioorg. Med. Chem. Lett.
1994, 4, 1161-1166. (g) Birkenshaw, T. N.; Caffrey, M. V.;
Cladingboel, D. E.; Cooper, M. E.; Donald, D. K.; Furber, M.;
Hardern, D. N.; Harrison, R. P.; Marriott, D. P.; Perry, M. W.
D.; Stocks, M. J .; Teague, S. J .; Withnall, W. J . Synthetic
FKBP12 Ligands. Design and Synthesis of Pyranose Replace-
ments. Bioorg. Med. Chem. Lett. 1994, 4, 2501-2506. (h) Caffrey,
M. V.; Cladingboel, D. E.; Cooper, M. E.; Donald, D. K.; Furber,
M.; Hardern, D. N.; Harrison, R. P.; Stocks, M. J .; Teague, S.
J . Synthesis and Evaluation of Dual Domain Macrocyclic
FKBP12 Ligands. Bioorg. Med. Chem. Lett. 1994, 4, 2507-2510.
(i) Armistead, D. M.; Badia, M. C.; Deininger, D. D; Duffy, J . P.;
Saunders, J . O.; Tung, R. D.; Thomson, J . A.; DeCenzo, M. T.;
Futer, O.; Livingston, D. J .; Murcko, M. A.; Yamashita, M. M.;
Navia, M. A. Design, Synthesis and Structure of Non-macrocyclic
Inhibitors of FKBP12, the Major Binding Protein for the
Immunosuppressant FK506. Acta Crystallogr. 1995, D51, 522-
528. (j) Tatlock, J . H.; Kalish, V. J .; Parge, H. E.; Knighton, D.
R.; Showalter, R. E.; Lewis, C. T.; French, J . V.; Villafranca, J .
E. High-Affinity FKBP-12 Ligands Derived from (R)-(-)-Car-
vone. Synthesis and Evaluation of FK506 Pyranose Ring Re-
placements. Bioorg. Med. Chem. Lett. 1995, 5, 2489-2494.
(21) (a) Hauske, J . R.; Dorff, P.; J ulin, S.; DiBrino, J .; Spencer, R.;
Williams, R. Design and Synthesis of Novel FKBP Inhibitors.
J . Med. Chem. 1992, 35, 4284-4296. (b) Andres, C. J .; Mac-
donald, T. L.; Ocain, T. D.; Longhi, D. Conformationally Defined
Analogs of Prolylamides. trans-Prolyl Peptidomimetics. J . Org.
Chem. 1993, 58, 6609-6613. (c) Babine, R. E.; Bleckman, T. M.;
Kissinger, C. R.; Showalter, R.; Pelletier, L. A.; Lewis, C.; Tucker,
K.; Moomaw, E.; Parge, H. E.; Villafranca, J . E. Design,
Synthesis and X-ray Crystallographic Studies of Novel FKBP-
12 Ligands. Bioorg. Med. Chem. Lett. 1995, 5, 1719-1724. (d)
Reference 20c.
(9) (a) Bierer, B. E.; Mattila, P. S.; Standaert, R. F.; Herzenberg, L.
A.; Burakoff, S. J .; Crabtree, G.; Schreiber, S. L. Two Distinct
Signal Transmission Pathways in T Lymphocytes are Inhibited
by Complexes Formed Between an Immunophilin and Either
FK506 or Rapamycin. Proc. Natl. Acad. Sci. U.S.A. 1990, 87,
9231-9235. (b) Dumont, F. J .; Melino, M. R.; Staruch, M. J .;
Koprak, S. L.; Fischer, P. A.; Sigal, N. H. The Immunosuppres-
sive Macrolides FK506 and Rapamycin act as Reciprocal An-
tagonists in Murine T-cells. J . Immunol. 1990, 144, 1418-1424.
(10) (a) Klee, C. B.; Draetta, G. F.; Hubbard, M. J . Calcineurin. Adv.
Enzymol. Relat. Areas Mol. Biol. 1988, 61, 149-200. (b) Pallen,
C.; Sharma, R. K.; Wang, J . H. Regulation of Calcineurin:
A
Multifunctional Calmodulin-Stimulated Phosphatase. In Cal-
cium Binding Proteins, Characterization and Properties, Vol. 1;
Thompson, M. P., Ed.; CRC Press: Boca Raton, FL, 1988; pp
51-82.
(11) (a) Liu, J .; Farmer, J . D., J r.; Lane, W. S.; Friedman, J .;
Weissman, I.; Schreiber, S. L. Calcineurin is a Common Target
of Cyclophilin-Cyclosporin A and FKBP-FK506 Complexes. Cell
1991, 66, 807-815. (b) Fruman, D. A.; Klee, C. B.; Bierer, B. E.;
Burakoff, S. J . Calcineurin Phosphatase Activity in T Lympho-
cytes is Inhibited by FK506 and Cyclosporin A. Proc. Natl. Acad.
Sci. U.S.A. 1992, 89, 3686-3690.
(12) (a) Flanagan, W. M.; Corthe´sy, B.; Bram, R. J .; Crabtree, G. R.
Nuclear Association of a T-Cell Transcription Factor Blocked by
FK506 and Cyclosporin A. Nature 1991, 352, 803-807. (b)
Dumont, F. J .; Staruch, M. J .; Koprak, S. L.; Melino, M. R.; Sigal,
N. H. Distinct Mechanisms of Suppression of Murine T-Cell
Activation by the Related Macrolides FK506 and Rapamycin.
J . Immunol. 1990, 144, 251-258. (c) Tocci, M. J .; Matkovich,
D. A.; Collier, K. A.; Kwok, P.; Dumont, F.; Lin, S.; Degudicibus,
S.; Siekierka, J . J .; Chin, J .; Hutchinson, N. I. The Immuno-
suppressant FK506 Selectively Inhibits Expression of Early
T-Cell Activation Genes. J . Immunol. 1989, 143, 718-726.
(13) Bierer, B. E.; Somers, P. K.; Wandless, T. J .; Burakoff, S. J .;
Schreiber, S. L. Probing Immunosuppressant Action with a
Nonnatural Immunophilin Ligand. Science 1990, 250, 556-559.
(14) Lepre, C. A.; Pearlman, D. A.; Cheng, J .-W.; DeCenzo, M. T.;
Livingston, D. J .; Moore, J . M. Solution Structure of FK506
Bound to the R42K, H87V Double Mutant of FKBP-12. Bio-
chemistry 1994, 33, 13571-13580. See also: Meadows, R. P.;
Nettesheim, D. G.; Xu, R. X.; Olejniczak, E. T.; Petros, A. M.;
Holzman, T. F.; Severin, J .; Gubbins, E.; Smith, H.; Fesik, S.
W. Three-Dimensional Structure of the FK506 Binding Protein/
Ascomycin Complex in Solution by Heteronuclear Three- and
Four-Dimensional NMR. Biochemistry 1993, 32, 754-765.
(15) (a) Van Duyne, G. D.; Standaert, R. F.; Karplus, P. A.; Schreiber,
S. L.; Clardy, J . Atomic Structure of FKBP-FK506, an Immu-
nophilin-Immunosuppressant Complex. Science 1991, 252, 839-
842. (b) Van Duyne, G. D.; Standaert, R. F.; Karplus, P. A.;
Schreiber, S. L.; Clardy, J . Atomic Structures of the Human
Immunophilin FKBP-12 Complexes with FK506 and Rapamycin.
J . Mol. Biol. 1993, 229, 105-124. (c) Clardy, J . Structural
Studies on FK506, Cyclosporin A and Their Immunophilin
Complexes. Perspect. Drug Discovery Des. 1994, 2, 127-144. (d)
Wilson, K. P.; Yamashita, M. M.; Sintchak, M. D.; Rotstein, S.
H.; Murcko, M. A.; Boger, J .; Thomson, J . A.; Fitzgibbon, M. J .;
Black, J . R.; Navia, M. A. Comparative X-ray Structures of the
Major Binding Protein for the Immunosuppressant FK506
(Tacrolimus) in Unligated Form and in Complex with FK506
and Rapamycin. Acta Crystallogr. 1995, D51, 511-521.
(22) (a) Duffy, J . P. Novel Immunosuppressive Compounds. Inter-
national Patent Application WO 92/21313, 1992. (b) Reference
20c.
(23) A non-macrocyclic molecule which associates with FKBP12 at
the 0.20 µM level and displays calcineurin inhibitory properties
has been reported; see: Andrus, M. B.; Schreiber, S. L. Structure-

1884 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 9
Dragovich et al.
Based Design of an Acyclic Ligand That Bridges FKBP12 and
Calcineurin. J . Am. Chem. Soc. 1993, 115, 10420-10421. See
also: Furber, M. FKBP12-Ligand-Calcineurin Interactions: Ana-
logues of SBL506. J . Am. Chem. Soc. 1995, 117, 7267-7268.
(24) Greer, J .; Erickson, J . W.; Baldwin, J . J .; Varney, M. D.
Application of the Three-Dimensional Structures of Protein
Target Molecules in Structure-Based Drug Design. J . Med.
Chem. 1994, 37, 1035-1054 and references therein.
(25) Michnick, S. W.; Rosen, M. K.; Wandless, T. J .; Karplus, M.;
Schreiber, S. L. Solution Structure of FKBP, a Rotamase Enzyme
and Receptor for FK506 and Rapamycin. Science 1991, 252,
836-839.
(26) The observed locations in the FKBP12-FK506 complex of the
FKBP12 residues determined to interact with calcineurin are
assumed to be desirable. However, slight movement of these
residues from the above positions has been noted in the crystal
structure of the calcineurin-FKBP12-FK506 ternary complex.16a
(27) (a) Wiley, R. A.; Rich, D. H. Peptidomimetics Derived From
Natural Products. Med. Res. Rev. 1993, 13, 368-378. (b) Petros,
A. M.; Luly, J . R.; Liang, H.; Fesik, S. W. Conformation of an
FK506 Analog in Aqueous Solution is Similar to the FKBP-
Bound Conformation of FK506. J . Am. Chem. Soc. 1993, 115,
9920-9924.
(28) Majer, P.; Randad, R. S. A Safe and Efficient Method for
Preparation of N,N′-Unsymmetrically Disubstituted Ureas Uti-
lizing Triphosgene. J . Org. Chem. 1994, 59, 1937-1938.
(29) Kotani, R. Demjanov Rearrangement of 1-Methylcyclohexyl-
methylamine. J . Org. Chem. 1965, 30, 350-354.
(30) J ohnson, R. L.; Rajakumar, G.; Yu, K.-L.; Mishra, R. K. Synthesis
of Pro-Leu-Gly-NH2 Analogues Modified at the Prolyl Residue
and Evaluation of Their Effects on the Receptor Binding Activity
of the Central Dopamine Receptor Agonist, ADTN. J . Med.
Chem. 1986, 29, 2104-2107.
(31) Brown, H. C.; Krishnamurthy, S.; Stocky, T. P. Selective
Reductions. XIX. The Rapid Reaction of Carboxylic Acids with
Borane-Tetrahydrofuran. A Remarkably Convenient Procedure
for the Selective Conversion of Carboxylic Acids to the Corre-
sponding Alcohols in the Presence of Other Functional Groups.
J . Org. Chem. 1973, 38, 2786-2792.
(32) The identity of the major acyclic urea isomer (cis or trans) in
chloroform solution is currently unknown.
(33) For use of related N-acyliminium ion cyclization reactions in
alkaloid synthesis, see: (a) Hart, D. J . Effect of A(1,3) Strain on
the Stereochemical Course of N-Acyliminium Ion Cyclizations.
J . Am. Chem. Soc. 1980, 102, 397-398. (b) Hart, D. J .; Kanai,
K.-i. New Approach to Lythraceae Alkaloids: Total Synthesis
of (()-Vertaline. J . Org. Chem. 1982, 47, 1555-1560. (c) Hart,
D. J .; Yang, T.-K. N-Acyliminium Ion Rearrangements: Gen-
eralities and Application to the Synthesis of Pyrrolizidine
Alkaloids. J . Org. Chem. 1985, 50, 235-242.
(34) Review: Speckamp, W. N.; Hiemstra, H. Intramolecular Reac-
tions of N-Acyliminium Intermediates. Tetrahedron 1985, 41,
4367-4416.
(35) The formation of diastereomeric alcohol and olefinic products
in related N-acyliminium ion cyclization reactions has been
reported. See refs 33b,c.
(36) Robins, M. J .; Wilson, J . S.; Hansske, F. Nucleic Acid Related
Compounds. 42. A General Procedure for the Efficient Deoxy-
genation of Secondary Alcohols. Regiospecific and Stereoselective
Conversion of Ribonucleosides to 2′-Deoxynucleosides. J . Am.
Chem. Soc. 1983, 105, 4059-4065.
(37) Schummer, D.; Ho¨fle, G. Tris(trimethylsilyl)silane as a Reagent
for the Radical Deoxygenation of Alcohols. Synlett 1990, 705-
706.
(38) Still, W. C.; Kahn, M.; Mitra, A. Rapid Chromatographic
Technique for Preparative Separations with Moderate Resolu-
tion. J . Org. Chem. 1978, 43, 2923-2925.
(39) Standaert, R. F.; Galat, A.; Verdine, G. L.; Schreiber, S. L.
Molecular Cloning and Overexpression of the Human FK506-
Binding Protein FKBP. Nature 1990, 346, 671-674.
(40) X-PLOR, version 3.1. Bru¨nger, A. T. X-PLOR v3.1 Manual; Yale
University Press: New Haven, CT, 1992.
(41) Quanta, version 4.1, is produced by BIOSYM/Molecular Simula-
tions, Inc., 16 New England Executive Park, Burlington, MA
01803-5297.
(42) McRee, D. E. A Visual Protein Crystallographic Software System
for X11/Xview. J . Mol. Graph. 1992, 10, 44-47.
(43) Kofron, J . L.; Kuzmic, P.; Kishore, V.; Colo´n-Bonilla, E.; Rich,
D. H. Determination of Kinetic Constants for Peptidyl Prolyl cis-
trans Isomerases by an Improved Spectrophotometric Assay.
Biochemistry 1991, 30, 6127-6134.
(44) (a) Harrison, R. K.; Stein, R. L. Substrate Specificities of the
Peptidyl Prolyl cis-trans Isomerase Activities of Cyclophilin and
FK-506 Binding Protein: Evidence for the Existence of a Family
of Distinct Enzymes. Biochemistry 1990, 29, 3813-3816. (b)
Reference 9a.
J M950798A