Nuclear magnetic resonance fragment-based identification of novel FKBP12 inhibitors

Article abstract of DOI:10.1021/jm0707424

Peptidyl-prolyl cis-trans isomerases are a group of cytosolic enzymes initially characterized by their ability to catalyze the cis-trans isomerization of peptidyl-prolyl bonds. This represents a significant event for protein folding because cis-proline in

Full text of DOI:10.1021/jm0707424

J. Med. Chem. 2007, 50, 6607–6617
6607
Nuclear Magnetic Resonance Fragment-Based Identification of Novel FKBP12 Inhibitors
John L. Stebbins,†,‡ Ziming Zhang,†,‡ Jinhua Chen,†,‡ Bainan Wu,† Aras Emdadi,§ Megan E. Williams,§ John Cashman,4 and
Maurizio Pellecchia*,†
Infectious and Inflammatory Disease Center, Cancer Center, Burnham Institute for Medical Research, 10901 North Torrey Pines Road,
La Jolla, California 92037, UniVersity of California at San Diego, DiVision of Biological Sciences, 9500 Gilman DriVe,
La Jolla, California 92093, and Human BioMolecular Research Institute, 5310 Eastgate Mall, San Diego, California 92121
ReceiVed June 25, 2007
Peptidyl-prolyl cis–trans isomerases are a group of cytosolic enzymes initially characterized by their ability
to catalyze the cis–trans isomerization of peptidyl-prolyl bonds. This represents a significant event for
protein folding because cis-proline introduces critical bends within the protein conformation. FK506-binding
proteins (FKBPs) represent one of the three families of enzymes sharing peptidyl-prolyl cis-trans isomerase
activity. Inhibitors of FKBP12, in particular, have potent neurotrophic properties both in vivo and in vitro.
Here, we describe a fragment-based unbiased nuclear magnetic resonance drug discovery approach for the
identification of novel classes of chemical inhibitors against FKBP12. Compared to FK506, the fragment-
based FKBP12 inhibitors developed herein possess significant advantages as drug candidates.
Introduction
approach 22,23 for the identification of novel classes of chemical
inhibitors against FKBP12.
Peptidyl-prolyl cis–trans isomerases (PPIsa) are a group of
cytosolic enzymes initially characterized by their ability to
catalyze the cis–trans isomerization of peptidyl-prolyl bonds.1
This represents a significant event for protein folding because
cis-prolines introduce essential bends within protein conforma-
tions. FK506-binding proteins (FKBPs) are one of three families
of enzymes sharing PPI activity. FKBPs are found in all classes
of organisms, some highly conserved while others are unique.2
These proteins have multiple cellular roles, the best known being
as receptors for medically important immuno-suppressors such
as FK506. However, most cellular FKBP functions are still
unknown. FKBP12 represents the minimal peptide sequence that
harbors the two hallmark properties of FKBPs: PPI activity and
immunophilin FK506 binding.
Results and Discussion
The relatively small size of FKBP12 (10 kDa) makes it
amenable to protein-NMR spectroscopy techniques, such as two-
dimensional (2D) [15N, 1H] or 2D [13C, 1H] experiments,
designed for the detection of ligand binding. These approaches
are preferential to transfer ligand-based techniques as they
provide intrinsic structural binding site information and are
largely void of false positives or false negatives.23,24 However,
the use of 2D [15N, 1H]25 or 2D [13C, 1H] correlation spectra 26,27
as a screening method requires relatively large amounts of
protein per sample (>100 µM) and relatively lengthy measure-
ment times. To ameliorate these problems we propose a primary
screen employing simple one-dimensional (1D) 1H NMR
experiments, acquired in the presence and absence of a mixture
of potential ligands. Eventual resonance overlap with resonance
lines of test compounds is resolved by using 13C-labeled samples
and 1D 13C-edited 1H NMR experiments. This protocol has the
intrinsic advantage of requiring relatively small protein con-
centrations (10–50 µM) and relatively short measurement times
(typically from 15 min to 1 h depending on the spectrometer
used and the protein’s molecular weight). Because of the 13C
filter, signals arising from the test ligands or from solvent and
buffers are effectively suppressed, thereby enabling the screening
of compounds in mixtures without protein spectrum interference.
Another advantage of this technique is that observables can also
be detected in the aliphatic region of the protein spectrum (1
ppm and below), a region rarely populated by signals from
organic molecules. In such cases, the use of 13C labeling would
not be necessary. This is illustrated in Figure 1 where the
Inhibitors of the prolyl isomerase activity of FKBP12 possess
potent neurotrophic properties in vivo and in vitro,3–6 a function
that may be distinct from its immunosuppressive activity. This
finding makes them obvious therapeutic candidates for a variety
of neurodegenerative diseases such as Parkinson’s disease,
Alzheimer’s disease, amyotrophic lateral sclerosis, diabetic
neuropathy, nerve injury, cerebral ischemia, and traumatic brain
injury. As a result, research in both industry and academia aimed
at the development of novel FKBP12 inhibitors has been
initiated. However, the small molecule inhibitors of FKBP12
developed thus far are invariably based on the natural immu-
nophilin FK506.3,7–21 Here we describe a fragment-based
unbiased nuclear magnetic resonance (NMR) drug discovery
* To whom corresondence should be addressed. Tel.: 858-646-3159. Fax:
858-713-9925. E-mail: mpellecchia@burnham.org.
†
Burnham Institute for Medical Research.
These authors contributed equally to this work.
University of California at San Diego.
Human BioMolecular Research Institute.
1
‡
aliphatic region of the 1D H NMR spectrum of FKBP12 is
§
shown in the absence and presence of compound 1 (BI-12B10).
The protein spectrum of the aliphatic region is generally not
affected by DMSO (up to 5%) or small changes in buffer
conditions, making identification of potential ligands straight-
forward and unambiguous. In addition, initial potency ranking
can be obtained by direct 1D 1H NMR chemical shift titrations.
An approximate KD value of 3 µM was obtained for 1 (Figure
1), which compares well with the reported KD value of 2 µM.25
4
aAbbreviations: PPI, peptidyl-prolyl cis-trans isomerase; FKBP,
FK506-bonding protein; NMR, nuclear magnetic resonance; HSQC, het-
eronuclear single quantum correlation; DMAP, dimethylaminopyridine;
DMSO, dimethylsulfoxide; DMF, dimethylformamide; EDC, N-ethyl-N′-
(3-dimethylaminopropyl)-carbodiimide hydrochloride; HOBt, 1-hydroxy-
1H-benzotriazole hydrate; MeOH, methanol; LiOH, lithium hydroxide;
NaOH, sodium hydroxide; NEt3, triethylamine; THF, tetrahydrofuran; rt,
room temperature.
10.1021/jm0707424 CCC: $37.00
2007 American Chemical Society
Published on Web 11/27/2007

6608 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 26
Stebbins et al.
Figure 2. Chemical structures and estimated binding affinities of hit
compounds from the NMR screen.
shown in Figure 2. As an example, we report in Figure 3 data
relevant to compound 2 (PubChem ID number 14717014). This
compound induces chemical shift changes upon binding in the
aliphatic region of the spectrum similar to those observed with
1.25 Compounds 3 (14717006) and 4 (14717025) resemble
previously identified FKBP12 inhibitors, which provides a
further validation of this approach. Compound 2, while also
displaying similarities to 1 and other FK506-derived inhibitors,28,29
presents novel structural features that could not have been easily
anticipated by simply modifying these compounds. For example,
the previously identified FK506-derived inhibitors require the
presence of an R-keto amido group (Figure 1), a moiety that is
critical for the ability of these compounds to bind FKBP12.
Attempts to replace the R-keto amido group with more stable
functional groups have been recently reported.17,29 As 2 lacks
this undesired functionality, it represents a more promising
starting point for the development of novel FKBP12 inhibitors.
Structure–Activity Relationships. Initial structure–activity
relationship data were obtained by selecting and testing 51
commercially available derivatives of 2 (Table 2), 12 derivatives
of 3 (Supporting Information, Table 1) and 13 derivatives of
compound 4 (Supporting Information, Table 2). Compound
dissociation constants for FKBP12 were determined by 1D-
Figure 1. Chemical structure of (A) FK506 and (B) 1. (C) 1D 1H
aliphatic NMR-based ligand screening. 1H NMR spectra of the FKBP12
aliphatic region measured at 27 °C in the presence (red) and in the
absence (black) of 1.
Table 1. Statistics of the Assembled Library for the Proposed
Fragment-Based Screeninga
database
number of compounds
diversity (%)
MayBridge
491
485
1010
976
2962
602
79
95
82
89
82
49
Life Chemicals
ChemBridge (I)
ChemBridge (II)
Overall
Natural Product
a The structures have been deposited in PubChem (http://pubchem.ncbi.
nlm.nih.gov). The diversity was computed using UNITY, as implemented
in SYBYL, version 7.0 (TRIPOS Inc., St. Louis, MO).
Fragment-Based Drug Discovery. As is typical of fragment-
based drug discovery, we assembled a scaffold library composed
of ∼3000 fragments (Table 1). Given the ability of the described
NMR-based screening technique to detect even weak binding
events, with affinities in the micromolar to millimolar range,
such a library of representative compounds should suffice to
identify preferential structures. Hits would subsequently be used
as starting points for iterative optimizations, using several
different strategies, the simplest being selecting compound
analogues. The selection of a sublibrary is also necessary to
overcome the intrinsic low throughput of NMR-based strategies
versus traditional biochemical assays. The scaffold library used
in this study was assembled from three different sources (Table
1), and the chemical structures of the library components have
been deposited into PubChem (http://pubchem.ncbi.nlm.nih.
gov). In addition, the library also includes a collection of 602
natural products (MicroSource) amenable to NMR screening.
The ability of library components to interact with FKBP12
was determined by 1D aliphatic-1H NMR measurements using
a 32 µM protein sample and mixtures of 10 compounds per
sample (125 µM each). Compound mixtures producing signifi-
cant chemical shift variations (e.g., larger than 0.08 ppm) were
subsequently deconvoluted. Ultimately, by employing this
strategy, we identified the classes of novel FKBP12 ligands
1
aliphatic H NMR and isothermal titration calorimetry (ITC;
Figure 3).
Based on the SAR data obtained with 2 (Table 2) and its
similarity with 1, the following chemical modifications were
made: the synthesis of bidentate compounds derivatizing the
morpholino group with secondary site binders and the modifica-
tion of the linker region between the phenyl ring and the
morpholino groups. In addition, SAR data relative to the phenyl
ring of 2 suggests that a bulkier, hydrophobic group such as an
isopropyl in the para position may be preferred (Schemes 1–4).
These strategies lead to the identification of submicromolar
binders for FKBP12 (Table 3).
Mapping Studies. To further characterize the binding af-
finities of the novel compounds for FKBP12, we collected and
overlaid 2D [15N, 1H] spectra of FKBP12 in presence of various
compounds. Among the five compounds tested, compound 6
(BI-69A8) and compound 10 (BI-69A5) resulted in significant
cross-peak movements, such as G51 and Y26, under the same

Identification of NoVel FKBP12 Inhibitors
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 26 6609
binding affinities (Table 3). Mapping chemical shift perturba-
tions (Figure 4) on the three-dimensional (3D) structure of
FKBP12 upon binding identifies the binding pocket for 6.
Similar chemical shift mapping data for compounds reported
in Table 3 were also obtained (not shown). The nature of the
chemical shifts of the compounds classes listed in Figure 2
clearly indicates that compounds occupy different subpockets
within the binding domain of the enzyme. In agreement with
this observation, a displacement assay based on NMR chemical
shift mapping shows that while 2 and 3 occupy different
subpockets their simultaneous presence is not tolerated (data
not shown).
Furthermore, we performed molecular docking studies by
using the X-ray coordinates of FKBP12 in complex with a
previously derived FK506 mimic compound.17 From these
models and in agreement with NMR chemical shift mapping
data, the morpholino group adopts a binding mode that is similar
to the piperidine of the previously reported compounds (such
as 1), while the sulfur atom and the p-isopropyl group extend
to occupy a smaller hydrophobic pocket at the edge of the
binding site (Figure 5). However, because of the insertion of
the oxygen atom, the morpholino moiety is shifted with respect
to the corresponding piperidines in the previously reported
compounds. As a result, the carbonyl of the amide group of 6
is involved in the crucial hydrogen bonding interaction with
the hydroxyl group of Tyr 26 (Figure 5).17 In FK506 and related
compounds, this hydrogen bonding was previously recognized
as being essential for the binding affinity and was attributed to
the R-keto group.3,16–21 Only recently, attempts to replace this
R-keto group were made with a fluorinated compound derivative
(Figure 5D). In agreement with this model, corresponding
substitutions of the morpholino group with previously reported
second site binders such as a pyridine ring (Table 3) resulted
in less active compounds, presumably due to the different
conformational arrangement of the aliphatic ring in the molecule.
Similarly, as one would predict by the model (Figure 5),
replacing the methylene group in 6 with a -CF2- group does
not result in increased activity (Table 3).
Therefore, we have obtained a molecular model that can
explain both experimental NMR chemical mapping studies and
observed SAR related to this series of compounds. We anticipate
that this model could be very useful in formulating hypotheses
for further derivatizations of 6 and related compounds.
Cell-Based Evaluation. To test the hypothesis that the
purported FKBP12 binders would stimulate neurite outgrowth
similar to that seen with FK506, we treated primary cortical
neurons with either 0.1 µM or 1 µM of test compound for 72 h
and then calculated average neurite length. As expected, FK506
treatment resulted in an increase in average neurite length
(Figure 6). Treatment with several of the compounds identified
in this study resulted in stimulation of neurite outgrowth. In
comparison to control treatment, most compounds resulted
in robust neurite outgrowth, an example of which can be seen
in Figure 6. In most cases, neurite outgrowth was stimulated to
a level comparable to that seen with FK506 (Figure 6). In
agreement with binding affinity determinations (Figure 2), 3
demonstrated the weakest effect, while 6 (Table 3) was the most
potent. Thus, we feel that these compounds have biological
activity similar to FK506 in this assay.
Figure 3. Binding data relative to 2. (A) 1D 1H aliphatic NMR spectra
of 32 µM FKBP12 in the presence of different concentrations of ligand.
(B) Fractional changes (R) of chemical shifts in FKBP12 as a function
of ligand concentration. The experimental data were fitted to the
nonlinear equation as described in the methods. KD ) 15.0 µM. (C)
ITC data: The top panel shows the heat signals for ligand injections
into a sample cell containing FKBP12 (100 µM). The bottom panel
shows the integrated heat of each injection after correcting for the heat
of dilution of the ligand. The curve represents the best fit to a model
involving a single set of independent sites. KD ) 9.4 µM, which is in
close agreement with the NMR titration value reported above.
While these data strongly suggest FKBP12 is the relevant
target of the compounds, we cannot rule out the possibility that
the molecules could interact with other closely related targets
(e.g., FKBP12.6, FKBP13, FKBP25, FKBP38, and FKBP38).28
However, when tested against Pin1, another proline cis–trans
molar ratio (Figure 4). In agreement with 1D chemical shift
titration, 2D mapping suggests that 6 and 10 have the strongest

6610 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 26
Stebbins et al.
Table 2. Structure–Activity Relationships Data Relative to Derivatives of 2

Identification of NoVel FKBP12 Inhibitors
Table 2. Continued
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 26 6611
Scheme 1a
analysis, there was no detected conversion of esters 8, 1, 10,
and 12 into their corresponding carboxylic acids, suggesting
the major route of metabolism was not ester hydrolysis. To
investigate the apparent metabolic stability of this series of
compounds, their ability to inhibit the conversion of testosterone
to 6-hydroxy testosterone by human liver microsomes at
concentrations of 10 µM was examined. At 10 µM, 12 and
FK506 completely inhibited the formation of 6-hydroxy test-
osterone. The other compounds examined showed estimated IC50
values of approximately 10 µM. Detailed determination of the
IC50 of 12 and FK506 for inhibition of testosterone hydroxylase
showed they possessed IC50 values of 0.81 µM and 2.26 µM,
respectively. It is likely that the pyridine moiety of 12 and the
vinyl group of FK506, respectively, are responsible for the
potent testosterone 6-hydroxylase inhibition activity. Regardless,
the fragment-based approach has successfully resulted in the
elaboration of smaller molecular weight compounds devoid of
the metabolic inhibitory activity of FK506. Development of
drug-like materials without the potential for adverse drug-drug
interactions may hold promise for a new class of neuroprotective
agents.
a Reagents and conditions: (a) NaOH, H2O, 80 °C, 2 h, then HCl; (b)
EDC, HOBt, NEt3, methyl morpholine-3-carboxylate, DMF, rt, 12 h; (c)
LiOH, THF/H2O, rt, 5 h, then HCl; (d) EDC, DMAP, 3-(pyridin-3-
yl)propan-1-ol, DMF, rt, 12 h.
In conclusion, our fragment-based data clearly demonstrate
the effectiveness and usefulness of our approaches in the rapid
identification and optimization of novel molecular tools. We
feel that this method will be useful for further lead development
and optimization of therapeutic candidates for a variety of
neurodegenerative diseases.
Scheme 2a
Material and Methods
Protein Expression and Purification. The gene coding for the
human FKBP12 was amplified with PCR and subcloned into
pET21a using the NdeI and XhoI cloning sites. The resul-
ting proteins contain eight extra C-terminal amino acid residues
(LEHHHHHH). The protein was expressed in the Escherichia coli
strain BL21(DE3) and purified using Ni2+ affinity chromatography.
The uniformly 15N-labeled FKBP12 was produced by growing the
bacteria in M9 minimal media containing 15NH4Cl as the sole
nitrogen source. The NMR samples were dissolved in 20 µM
sodium phosphate buffer (pH 7.5) containing 90%/10% (H2O/2H2O)
a Reagents and conditions: (a) Na2CO3, DMF, 60 °C; (b) LiOH, THF/
H2O; (c) EDC, HOBt, NEt3, methyl morpholine-3-carboxylate, DMF, rt,
12 h.
isomerase,30,31 the compounds did not show any appreciable
binding (data not shown). Therefore, our compounds may prove
to be molecular probes useful in the elucidation of the roles of
FKBP12 related proteins in cell.
Metabolic Stability Studies. Compounds 8 (BI-69B7), 9 (BI-
12B8), 1, 10, 11 (BI-69A6), 12 (BI-69A7), and 6 (Table 3), as
well as FK506, were assayed for their metabolic stability in
the presence of pooled human liver microsomes and an NADPH
generating system. With the exception of 11 and FK506, which
showed half-lives of greater than 60 min, the compounds
exhibited half-lives of about 40 min. On the basis of RP-HPLC
2
or 99.5% H2O.
NMR Spectroscopy. Spectra were acquired on a 600 MHz
Bruker Avance spectrometer equipped with TXI probe and z-
shielded gradient coils or on a 600 MHz Bruker Avance equipped
with TCI cryoprobe. Ligand binding was detected at 27 °C by
comparing the aliphatic region of 1D 1H spectra of 32 µM FKPB12
in the presence and in the absence of 125 µM compounds.
Compounds were initially tested at mixtures of 10, and then
individual compound mixtures that caused significant perturbations
in the spectrum (>0.08 ppm) were characterized further.

6612 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 26
Stebbins et al.
Scheme 3a
a Reagents and conditions: (a) mCPBA, CH2Cl2, rt, 72 h.
Scheme 4a
to attach for 48 h, after which various doses of experimental
compounds were applied. After a 24 h incubation, cells were
transfected with GFP expressing plasmid (pBOS-EGFP) using
Lipofectamine 2000, after which various doses of experimental
compounds were reapplied. Neurite length of at least 10 neurite
containing cells was quantified using Image-Pro Plus (Media
Cybernetics, Bethesda, MD).
Immunostaining of Dissociated Cells on Coverslips. Cells were
fixed for 15 min at room temperature with 4% paraformaldehyde/
0.1% Triton X-100 followed by blocking nonspecific binding for
2 h with 3% BSA and 0.1% Triton X-100 in PBS. Cells were then
incubated with primary antibody for 2 h at room temperature in
3% BSA and 0.3% Triton X-100 in PBS and then with secondary
antibody at room temperature for 1 h in 3% BSA and 0.3% Triton
X-100 in PBS. Primary antibody used was rabbit polyclonal anti-
GFP (1:1000, Invitrogen catalog number A11122). Secondary
antibody used was goat antirabbit Alexa Flour 488 (1:1000,
Invitrogen, catalog number A11034).
Synthesis of Phenylthioacetylmorpholine and Analogs. All
commercially available reagents were used without further purifica-
tion. Column chromatography was performed with silica gel 60
(35–75 Å). 1H NMR and 13C NMR for QC analysis were acquired
on a Varian Inova 300 MHz spectrometer. Chemical shifts are
reported in ppm from residual solvent peaks (2.50 and 7.26 for
DMSO-d6 and CDCl3 for 1H NMR, and 49.0 and 77.16 for DMSO-
d6 and CDCl3 for 13C NMR, respectively). High resolution ESI-
TOF mass spectra were acquired at the Center for Mass Spectrom-
etry, the Scripps Research Institute, La Jolla, CA. Compounds 6,
7, 8, 10, 12, 16, 17, and 18 were all found to be in excess of 95%
pure as established by LC-MS.
a Reagents and conditions: (a) EDC, HOBt, DMF; (b) (COCl)2, DMF
(cat), CH2Cl2.
Dissociation equilibrium constants (KD) of ligands were deter-
mined by monitoring the protein chemical shift changes as a
function of ligand concentration. Data were collected for a set of
resolved FKBP12 1H NMR resonances and fitted to a single binding
site model according to the following equation:25 R ) (δobs - δf)/
∆
b-f, where R is the mole fraction of the protein–ligand complex
used for the KD calculation, δobs is the observed chemical shift for
NMR titrations, δf is the chemical shift of the free protein, and
∆
is the difference in chemical shift between the free and the
b-f
2-(4-Isopropylphenylthio) Acetic Acid; Compound 13 (BI-
69A4). 4-Isopropylbenzenethiol (762 mg, 5.0 mmol) was dissolved
in a NaOH solution (500 mg, 12.5 mmol) in H2O (1.2 mL).
2-Chloroacetic acid (472.5 mg, 5.0 mmol) was added and the
reaction mixture was heated at 80 °C for 2 h under N2. The reaction
mixture was dissolved in H2O (100 mL) and was adjusted to pH
2.0 with concentrated HCl. It was extracted with EtOAc. The
combined organic phase was extracted with Na2CO3 solution (10%).
The Na2CO3 solution was adjusted to pH 2.0 with concentrated
HCl and extracted with EtOAc. It was dried over Na2SO4 and
solvent was removed to give 13 as a white solid without further
purification (450 mg, 43%). 1H NMR (300 MHz, DMSO-d6) δ
7.28–7.18 (m, 4 H), 3.73 (s, 2 H), 2.84 (sept, J ) 6.9 Hz, 1 H),
1.17 (d, J ) 6.9 Hz, 6 H). 13C NMR (75 MHz, CDCl3) δ 176.1,
148.6, 131.0, 127.5, 37.3, 33.9, 24.0. HRMS m/z calcd for
C11H13O2S [M - H]-, 209.0642; found, 209.0635.
Methyl 4-(2-(4-Isopropylphenylthio)acetyl)morpholine-3-car-
boxylate (10). Triethylamine (150 mg, 1.5 mmol), EDC (114.6 mg,
0.60 mmol), and HOBt (91.8 mg, 0.60 mmol) were added to a
solution of 13 (105 mg, 0.50 mmol) in DMF (5.0 mL) under N2.
After the reaction mixture was stirred at rt for 30 min, methyl
morpholine-3-carboxylate (90.5 mg, 0.50 mmol) was added. After
the reaction mixture was stirred at rt for 12 h, DMF was removed
under reduced pressure. It was purified by silica gel flash chroma-
tography (EtOAc-hexanes ) 1:4) to give 10 as a colorless oil (70
mg, 42%). The product exits in two amide rotamers in about 10:1
fully complexed protein.
All NMR data were processed and analyzed using TOPSPIN2.0
(Bruker BioSpin Corp., Billerica, MA) and SPARKY.32 For
chemical shift mapping, the NMR samples contained 0.1 mM 15N-
labeled FKBP12, 100 mM potassium phosphate buffer (pH 6.5),
1
5% DMSO, and 5% D2O. 2D [15N, H]-HSQC experiments were
acquired using 32 scans with 2048 and 128 complex data points in
1
the H and 15N dimensions at 300 K.
Binding Constant Determination by Isothermal Titration
Calorimetry. Isothermal titration calorimetry (ITC) was performed
on a VP-ITC calorimeter from Microcal (Northampton, MA). A
total of 8 µL of ligand solution (1.0 mM) were injected into the
cell containing 100 µM FKBP12 per injection. In each experiment,
37 injections were made. All titrations were performed at 23 °C in
PBS buffer supplemented with 10% DMSO. Experimental data were
analyzed using Microcal Origin software provided by the ITC
manufacturer (Microcal, Northampton, MA).
Neurite Outgrowth Assay. E18/19 cortical cells from Long-
Evans rats were cultured as previously described,33 with the
following minor modifications. The cortex was dissected in ice-
cold HBSS (6.5 g/L glucose), digested in 10 U/mL Papain in
dissociation media (2 × 20 min), and dissociated in culture media.
The dissociated neurons were plated on polylysine laminin-coated
24-well plates at 3 × 105 cells per well in glutamine-free Basal
Media Eagle (Sigma) supplemented with glutamine (to 1 mM), N2
(to 1%; Gibco), and fetal bovine serum (5%). Cells were allowed

Identification of NoVel FKBP12 Inhibitors
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 26 6613
Table 3. Synthesized Compounds and Relative Binding Affinities for FBKP12a
a *Indicates an estimated value, as compounds bind in the intermediate exchange rate, typical of those with submicromolar affinity.
ratio. For the major rotamer: 1H NMR (300 MHz, CDCl3) δ
7.44–7.35 (m, 2 H), 7.21–7.14 (m, 2 H), 5.02 (s, 1 H), 4.41 (d, J
) 12 Hz, 1 H), 3.94–3.79 (m, 2 H), 3.76 (s, 3 H), 3.73 (s, 1 H),
3.64–3.45 (m, 3 H), 3.48–3.36 (m, 1 H), 2.94–2.82 (m, 1 H),
1.26–1.18 (m, 6 H). 13C NMR (75 MHz, CDCl3) δ 169.7, 169.1,
148.5, 131.4, 127.3, 67.7, 66.3, 52.6, 52.5, 43.8, 36.9, 33.8, 23.9.
HRMS m/z calcd for C17H24NO4S [M + H]+, 338.1426; found,
338.1423.
4-(2-(4-Isopropylphenylthio)acetyl)morpholine-3-carboxylic
Acid (11). A solution of 10 (337 mg, 1.0 mmol) in THF (10.0
mL) was mixed with a solution of LiOH (120 mg, 5.0 mmol) in
H2O (10.0 mL). After the mixture was stirred at rt for 5 h, THF
was removed under reduced pressure and the solution was washed
with ether (3 × 20 mL). The solution was adjusted to pH 1.0 with
concentrated HCl and was extracted with ether. The combined ether
solution was dried over Na2SO4. The solvent was removed to give
11 without further purification (200 mg, 62%). 1H NMR (300 MHz,
CDCl3) δ 7.62 (b, 1 H), 7.42–7.36 (m, 2 H), 7.20–7.14 (m, 2 H),
5.04 (d, J ) 3.3 Hz, 1 H), 4.43 (d, J ) 11.7 Hz, 1 H), 3.94–3.70
(m, 3 H), 3.70–3.50 (m, 3 H), 3.50–3.34 (m, 1 H), 2.87 (sept, J )
6.9 Hz), 1.23 (s, 3 H), 1.21 (s, 3 H). 13C NMR (75 MHz, CDCl3)
δ 173.5, 169.7, 148.7, 131.7, 127.3, 67.5, 66.1, 52.4, 43.8, 36.8,
33.7, 23.8. HRMS m/z calcd for C16H22NO4S [M + H]+, 324.1264;
found, 324.1263.
at rt for 12 h, DMF was removed under reduced pressure and the
residuewaspurifiedbysilicagelflashchromatography(EtOAc-hexanes
1
) 1:2–1:1) to give 12 as a white solid (80 mg, 53%). H NMR
(300 MHz, CDCl3) δ 8.46–8.42 (m, 2 H), 7.50–7.45 (m, 1 H), 7.42
-7.34 (m, 2 H), 7.23–7.19 (m, 1 H), 7.19–7.13 (m, 2 H), 5.00 (d,
J ) 3.6 Hz, 1 H), 4.44–4.34 (m, 1 H), 4.25–4.16 (m, 2 H), 3.94–3.82
(m, 1 H), 3.78–3.42 (m, 1 H), 3.70–3.54 (m, 3 H), 3.48–3.38 (m,
1 H), 2.86 (sept, J ) 6.9 Hz, 1 H), 2.67 (t, J ) 7.2 Hz, 2 H),
2.04–1.92 (m, 2 H), 2.21 (d, J ) 6.9 Hz, 6 H). 13C NMR (75 MHz,
CDCl3) δ 169.3, 150.0, 147.7, 136.0, 131.4, 127.4, 123.5, 67.9,
66.4, 64.7, 52.7, 43.9, 37.0, 33.8, 30.0, 29.3, 24.0. HRMS m/z calcd
for C24H31N2O4S [M + H]+, 443.1999; found, 443.2007.
Ethyl 2,2-Difluoro-2-(4-isopropylphenylthio)acetate; Com-
pound 14 (JCIII150). Ethyl 2-bromo-2,2-difluoroacetate (1.0 g,
5.0 mmol), 4-isopropylbenzethiol (0.75 g, 5.0 mmol), and sodium
carbonate (0.53 g, 5.0 mmol) were mixed in DMF (10.0 mL) and
heated to 60 °C under N2 for 12 h. DMF was removed and the
residue was purified by silica gel flash chromatography (EtOAc-
hexanes ) 1:9) to give 14 as a colorless oil (0.70 g, 51%). 1H
NMR (300 MHz, DMSO-d6) δ 7.53 (d, J ) 8.4 Hz, 2 H), 7.37 (d,
J ) 8.1 Hz, 2 H), 4.32 (q, 2 H), 2.94 (sept, J ) 6.9 Hz, 1 H), 1.20
(d, J ) 6.9 Hz, 6 H), 1.14 (t, J ) 7.2 Hz, 3 H). 13C NMR (75
MHz, DMSO-d6) δ 161.2, 160.8, 160.4, 151.9, 136.6, 127.8, 123.8,
120.4, 120.0, 116.2, 63.9, 33.2, 23.5, 13.5. HRMS m/z calcd for
C13H16F2O2SNa [M + Na]+, 297.0731; found, 297.0735.
3-(Pyridine-3-yl) propyl 4-(2-(4-isopropylphenylthio)acetyl)
morpholine-3-carboxylate (12). 3-(Pyridin-3-yl)propan-1-ol (56
mg, 0.41 mmol) was added to a solution of 11 (110 mg, 0.34 mmol)
in DMF (5.0 mL). EDC (95 mg, 0.51 mmol) and DMAP (20 mg,
0.17 mmol) were then added. After the reaction mixture was stirred
2,2-Difluoro-2-(4-isopropylphenylthio)acetic Acid; Compound
15 (JCIII151). A solution of ethyl 2,2-difluoro-2-(4-isopropylphe-
nylthio)acetate (440 mg, 1.6 mmol) in THF (5.0 mL) was mixed
with a solution of LiOH (115 mg, 4.8 mmol) in H2O (5.0 mL) and

6614 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 26
Stebbins et al.
Figure 4. Chemical shift mapping studies. (A) Changes in chemical shifts of FKBP12 15N-HSQC spectrum upon addition of 6. The molar ratio
of FKBP12 and the compounds are 1:1. (B) Sequential variation of chemical shift changes due to the binding of 6. Chemical shift changes differences
(∆δ) between 15N-HSQC spectra of free FKBP12 and 6-bound FKBP12 were calculated as ∆δ ) [(∆HN)2 + (0.17∆15N)2]1/2. (C) Mapping on a
3D surface representation of the calculated chemical shift changes between free and 6-bound FKBP12. Colors are white (∆δ < 0.1), yellow (0.1
< ∆δ < 0.2), orange (0.2 < ∆δ < 0.4), and red (0.4 < ∆δ < 0.6).
stirred at rt for 2 h. THF was removed under reduced pressure
and the solution was washed with ether (3 × 10 mL). It was
acidified to pH 3.0 and was extracted with ether (3 × 20 mL). It
was washed with brine and dried over Na2SO4 to give a colorless
oil without further purification (390 mg, 99%). 1H NMR (300 MHz,
DMSO-d6) δ 7.52 (d, J ) 8.1 Hz, 2 H), 7.35 (d, J ) 8.1 Hz, 2 H),
2.93 (sept, J ) 6.9 Hz, 1 H), 1.20 (d, J ) 6.9 Hz, 6 H). 13C NMR
(75 MHz, DMSO-d6) δ 162.9, 162.5, 162.1, 151.4, 136.4, 127.7,

Identification of NoVel FKBP12 Inhibitors
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 26 6615
Figure 5. Molecular docking studies. (A) Surface representation of the X-ray structure of FKBP12 in complex with a previously derived inhibitor
(PDB ID 1J4R).17 (B) Docked structure of a previously reported derivative of 1.25 (C) Docked structure of 6. (D) Overlay of the docked structures
shown in B and C. In all figures, hydrogen-bonding interactions between the critical Tyr26 and each compound are highlighted by a dashed line.
124.3, 121.2, 120.5, 116.7, 33.2, 23.6. HRMS m/z calcd for
C11H11F2O2S [M - H]+, 245.0453; found, 245.0457.
3-Chloroperoxybenzoic acid (mCPBA, 245 mg, 70%, 1.0 mmol)
was added in small portions to a solution of 3-(pyridin-3-yl)propyl
4-(2-(4-isopropylphenylthio)acetyl)morpholine-3-carboxylate (110
mg, 0.25 mmol) in CH2Cl2 (25 mL). After the solution had been
stirred at rt for 72 h, it was washed with excess of cold Na2SO3
and Na2CO3 solution to remove remaining mCPBA. It was washed
with brine and dried over Na2SO4. Solvent was removed under
reduced pressure and the residue was purified by silica gel flash
chromatography (EtOAc-hexanes ) 1:1) to give a colorless oil
(40 mg, 32%). 1H NMR (300 MHz, CDCl3; a mixture of rotamers)
δ 8.18–8.04 (m, 2 H), 7.88–7.78 (m, 2 H), 7.50–7.36 (m, 2 H),
7.24–7.16 (m, 1 H), 7.15–7.08 (m, 1 H), 4.50–4.12 (m, 5 H),
4.00–3.78 (m, 2 H), 3.74–3.44 (m, 3 H), 3.16–2.92 (m, 1 H),
2.72–2.60 (m, 2 H), 2.10–1.92 (m, 2 H), 1.27 (d, J ) 6.9 Hz, 6 H).
13C NMR (75 MHz, CDCl3) δ 168.7, 161.8, 156.3, 140.2, 139.2,
137.4, 136.2, 128.6, 127.6, 126.7, 125.8, 67.7, 66.5, 64.4, 59.7,
53.1, 44.8, 34.4, 29.3, 29.1, 23.7. HRMS m/z calcd for C24H31N2O7S
[M + H]+, 491.1846; found, 491.1842.
2-(4-Isopropylphenylthio)-1-morpholinoethanone (6). To a
solution of 2-(4-isopropylphenylthio)acetic acid (13; 210 mg, 1.0
mmol) in DMF (20.0 mL) was added EDC (228 mg, 1.2 mmol),
HOBt (184 mg, 1.2 mmol), NEt3 (300 mg, 3.0 mmol), and
morpholine (174 mg, 2.0 mmol). After the reaction mixture was
stirred at rt for 72 h, DMF was removed under reduced pressure.
It was purified by silica gel flash chromatography (EtOAc-hexanes
) 1:7) to give a colorless oil (100 mg, 36%). 1H NMR (300 MHz,
CDCl3) δ 7.39 (d, J ) 8.4 Hz, 2 H), 7.17 (d, J ) 8.1 Hz, 2 H),
3.69 (s, 2 H), 3.66–3.56 (m, 6 H), 1.48–3.42 (m, 2 H), 2.87 (sept,
J ) 6.9 Hz, 1 H), 1.22 (d, J ) 6.9 Hz, 6 H). 13C NMR (75 MHz,
CDCl3) δ 167.5, 148.6, 131.4, 127.4, 66.9, 66.6, 46.9, 42.4, 37.0,
33.8, 24.0. HRMS m/z calcd for C15H22NO2S [M + H]+, 280.1366;
found, 280.1361.
Methyl 4-(2,2-Difluoro-2-(4-isopropylphenylthio)acetyl)mor-
pholine-3-carboxylate (8). Triethylamine (500 mg, 5.0 mmol),
EDC (384 mg, 2.0 mmol), and HOBt (306 mg, 2.0 mmol) was
added to a solution of 2,2-difluoro-2-(4-isopropylphenylthio)acetic
acid (390 mg, 1.60 mmol) in DMF (5.0 mL) under N2. After the
reaction mixture was stirred at rt for 30 min, methyl morpholine-
3-carboxylate (295 mg, 1.63 mmol) was added. After the reaction
mixture was stirred at rt for 12 h, DMF was removed under reduced
pressure. It was purified by silica gel flash chromatography
(EtOAc-hexanes ) 1:7) to give a colorless oil (120 mg, 20%).
1H NMR (300 MHz, DMSO-d6) δ 7.62–7.52 (m, 2 H), 7.40–7.34
(m, 2 H), 4.93 (s, 1 H), 4.32–4.22 (m, 1 H), 4.06–3.96 (m, 1 H),
3.92–3.80 (m, 1 H), 3.74 (s, 1 H), 3.72–3.64 (m, 1 H), 3.52–3.02
(m, 2 H), 3.02–2.88 (m, 2 H), 3.00–2.89 (m, 1 H), 1.24–1.08 (m,
6 H). 13C NMR (75 MHz, DMSO-d6) δ 168.9, 160.4, 151.6, 136.8,
127.7, 120.7, 66.9, 65.3, 53.2, 52.8, 43.9, 33.2, 23.6. HRMS m/z
calcd for C17H22F2NO4S [M + H]+, 374.1232; found, 374.1237.
Methyl 4-(2-(4-Isopropylphenylsulfonyl)acetyl)morpholine-
3-carboxylate; Compound 16 (BI-69A9). 3-Chloroperoxybenzoic
acid (mCPBA, 245 mg, 70%, 1.0 mmol) was added in small portions
to a solution of methyl 4-(2-(4-isopropylphenylthio)acetyl)mor-
pholine-3-carboxylate (10, 100 mg, 0.30 mmol) in CH2Cl2 (10 mL).
After the solution had been stirred at rt for 72 h, it was washed
with an excess of cold Na2SO3 and Na2CO3 solution to remove
remaining mCPBA. It was washed with brine and dried over
Na2SO4. Solvent was removed under reduced pressure and the
residuewaspurifiedbysilicagelflashchromatography(EtOAc-hexanes
1
) 1:4) to give a colorless oil (80 mg, 72%). H NMR (300 MHz,
CDCl3; a mixture of rotamers) δ 7.88–7.78 (m, 2 H), 4.98 (m, 1
H), 4.50–4.14 (m, 3 H), 3.96–3.82 (m, 2 H), 3.75 (s, 3 H), 3.70–3.40
(m, 3 H), 2.98 (sept, J ) 6.6 Hz, 1 H), 1.253 (d, J ) 6.6 Hz, 6 H).
13C NMR (75 MHz, CDCl3) δ 169.2, 161.7, 156.1, 136.0, 128.6,
127.5, 67.6, 66.5, 59.6, 52.9, 52.8, 44.7, 34.3, 23.6. HRMS m/z
calcd for C17H24NO6S [M + H]+, 370.1319; found, 370.1328.
3-(3-(4-(2-(4-Isopropylphenylsulfonyl)acetyl)morpholine-3-car-
bonyloxy)propyl)pyridine 1-Oxide; Compound 17 (BI-69A10).
2,2-Difluoro-2-(4-isopropylphenylthio)-1-morpholinoetha-
none (7). To a solution of 2,2-difluoro-2-(4-isopropylphenylthio)
acetic acid (15; 120 mg, 0.50 mmol) in CH2Cl2 (10.0 mL) was
added EDC (144 mg, 0.75 mmol), HOBt (114.8 mg, 0.75 mmol),
NEt3 (150 mg, 1.5 mmol), and morpholine (87.2 mg, 1.0 mmol).
After the solution was stirred at rt for 24 h, the solution was washed

6616 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 26
Stebbins et al.
Figure 6. Compound-mediated augmentation of neurite outgrowth in cultured primary cortical rat embryonic cells. Freshly explanted cortical cells
expressing transfected green fluorescent protein (GFP) were treated with (A) vehicle or (B) 1 µM 6 for 72 h. (C) The average neurite length (L)
in microns was calculated from at least 10 neurons per indicated treatment (0.1 and 1 µM, for left and right bars, respectively).
with saturated NH4Cl solution, water, saturated NaHCO3 solution,
and brine. It was dried over Na2SO4, and the solvent was removed.
It was purified by silica gel flash chromatography (CH2Cl2, 100%)
to give as a colorless oil (80 mg, 51%). 1H NMR (300 MHz, CDCl3)
δ 7.54 (d, J ) 8.1 Hz, 2 H), 7.26 (d, J ) 8.1 Hz, 2 H), 3.80–3.64
(m, 8 H), 2.93 (sept, J ) 6.9 Hz, 1 H), 1.26 (d, J ) 6.9 Hz, 6 H).
13C NMR (75 MHz, CDCl3) δ 160.1, 151.7, 137.0, 127.5, 123.8,
121.8, 120.0, 66.8, 47.0, 43.8, 34.1, 23.0. HRMS m/z calcd for
C15H20F2NO2S [M + H]+, 316.1177; found, 316.1177.
CORD37 and energy-minimized with Sybyl. For each compound,
10 solutions were generated and subsequently ranked according to
Chemscore (Cambridge Crystallographic Data Centre, Cambridge,
U.K.).34,35 Top solutions were used to represent the docked
geometry of the compounds and compared with the X-ray coordi-
nates of compounds previously reported.
Human Liver Microsomal Stability. Solvents, reagents, and
chemicals were purchased from VWR (San Diego, CA) in the
highest grade commercially available. Human liver microsomes
were purchased from BD Gentest (Boston, MA). Testosterone and
6-hydroxy testosterone and the reagents of the NADPH-generating
system were purchased from Sigma (Milwaukee, WI). A mixture
of the NADPH-generating system consisting of 50 µL of 2.5 mM
DETAPAC, 40 µL of 5 mM NADP+, 40 µL of 5 mM glucose-6-
phosphate, and 60 µL of glucose-6-phosphate dehydrogenase (50
IU/mL) per sample was prepared. Each incubation received 280
µL of 50 mM potassium phosphate buffer (pH 7.4) and 40 µL of
a 20 mg/mL solution of pooled human liver microsomes (BD
Gentest, Boston, MA). Each assay was initiated by the addition of
20 µL of a 5 mM stock of the test compound. The FK506 stock
was at 0.5 mM. Aliquots were removed at 0, 10, 25, 40, and 60
min The assay aliquots were extracted with dichloromethane/
isopropanol (3:1). The tubes were centrifuged for 5 min at 4000
rpm to separate the organic and aqueous layers. The organic layer
was decanted and concentrated under a stream of argon, and the
residue was reconstituted with 200 µL of methanol and analyzed
by RP-HPLC using a gradient of water with 0.02% perchloric acid
going to 100% acetonitrile over 12 min at a flow rate of 1.5 mL/
min at 254 nm.
3-(4-Isopropylphenyl)-1-morpholinopropan-1-one; Compound
18 (BI-69B10). To a solution of 3-(4-isopropylphenyl)propanoic
acid (192 mg, 1.0 mmol) in CH2Cl2 (10.0 mL) was added oxalyl
chloride (381 mg, 3.0 mmol) at 0 °C. DMF (0.5 drop) was added
and the solution was stirred at rt for 3 h. The solvent was removed
and was coevaporated with CH2Cl2 (5 mL) three times. It was
dissolved in CH2Cl2 (10.0 mL) and cooled to 0 °C. Morpholine
(100 mg, 1.2 mmol) and NEt3 (200 mg, 2.0 mmol) was added
dropwise. After it was stirred at rt for 3 h, it was washed with
water (10.0 mL) and dried over Na2SO4, and the solvent was
removed. It was purified by silica gel flash chromatography
1
(EtOAc-CH2Cl2 ) 1:3) to give a white solid (200 mg, 77%). H
NMR (300 MHz, CDCl3) δ 7.14 (d, 4 H), 3.61 (s, 4 H), 3.48 (t, J
) 4.2 Hz, 2 H), 3.34 (t, J ) 4.5 Hz, 2 H), 2.94 (t, J ) 7.5 Hz, 2
H), 2.87 (sept, J ) 6.6 Hz, 1 H), 2.60 (t, J ) 7.5 Hz, 2 H), 1.23 (d,
J ) 6.9 Hz, 6 H). 13C NMR (75 MHz, CDCl3) δ 171.1, 147.0,
138.4, 128.5, 126.6, 66.9, 66.5, 46.1, 42.0, 35.0, 33.8, 31.2, 24.1.
HRMS m/z calcd for C16H24NO2 [M + H]+, 262.1801; found,
262.1805.
Molecular Modeling. Docking studies were performed with
GOLD version 3.0 (Cambridge Crystallographic Data Centre,
Cambridge, U.K.)34,35 and analyzed with Sybyl (Tripos, St. Louis,
MO). Molecular surfaces were generated with MOLCAD.36 The
X-ray coordinates of FKBP12 PDB-ID 1J4R were used to dock
the compounds. Molecular models were generated with CON-
Testosterone 6-Hydroxylase Inhibition. A mixture of the
NADPH-generating system described above was prepared and 90
µL was placed into each incubation. Each incubation then received
140 µL of 50 mM potassium phosphate buffer (pH 7.4). Each
incubation received 20 or 30 µL of the microsome preparation

Identification of NoVel FKBP12 Inhibitors
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 26 6617
(14) Wang, G. T.; Lane, B.; Fesik, S. W.; Petros, W.; Luly, J.; Krafft,
G. A. Synthesis and FKBP binding of small molecule mimics of
tricarbonyl region of FK506. Bioorg. Med. Chem. Lett. 1994, 4 (9),
1161–1166.
(15) Yamashita, D. S.; Hye-Ja-Oh; Yen, S.-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 (2), 325–328.
followed by addition of the test compound. The assay was initiated
by the addition of 4 µL of a 1.45 mg/mL testosterone substrate
stock to incubations, giving a final volume of 250 µL. The
incubations were shaken for 10 min at 37 °C. Each incubation was
terminated at 10 min by the addition of 1 mL of cold ethyl acetate.
Sodium carbonate (20 mg) was added to each incubation, followed
by mixing, and the mixture was centrifuged for 5 min at 4000 rpm
to separate the organic and aqueous layers. The organic layer was
decanted and concentrated under a stream of argon and reconstituted
with 200 µL of methanol. The mixture was analyzed by RP-HPLC
with a HS Supelco column at 254 nm, employing an isocratic
mobile phase of water/MeOH/MeCN/HClO4 (30:60:10:0.02, v:v)
at 1 mL/min. Under these conditions, testosterone had a retention
time of 6.8 min and 6-hydroxy testosterone eluted at 4.0 min.
(16) Snyder, S. H.; Sabatini, D. M. Immunophilins and the nervous system.
Nat. Med. 1995, 1 (1), 32–37.
(17) Dubowchik, G. M.; Vrudhula, V. M.; Dasgupta, B.; Ditta, J.; Chen,
T.; Sheriff, S.; Sipman, K.; Witmer, M.; Tredup, J.; Vyas, D. M.;
Verdoorn, T. A.; Bollini, S.; Vinitsky, A. 2-Aryl-2,2-difluoroacetamide
FKBP12 ligands: Synthesis and X-ray structural studies. Org. Lett.
2001, 3 (25), 3987–3990.
(18) Schreiber, S. L. Chemistry and biology of the immunophilins and their
immunosuppressive ligands. Science 1991, 251 (4991), 283.
(19) Wu, Y. Q.; Wilkinson, D. E.; Limburg, D.; Li, J. H.; Sauer, H.; Ross,
D.; Liang, S.; Spicer, D.; Valentine, H.; Fuller, M.; Guo, H.; Howorth,
P.; Soni, R.; Chen, Y.; Steiner, J. P.; Hamilton, G. S. Synthesis of
ketone analogues of prolyl and pipecolyl ester FKBP12 ligands. J. Med.
Chem. 2002, 45 (16), 3558–3568.
(20) 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., Sect. D 1995, 51 (Pt 4), 522–528.
Acknowledgment. This work was supported in part by the
NIH Grants U54H9003916, XO1MH078942, and R01HL082574
(to M.P.). We thank Dr. Andrey Bobkov (BIMR) for technical
assistance. We thank Dr. Anirvan Ghosh (UCSD) for his generosity
with primary cortical cells and use of his laboratory facilities.
Supporting Information Available: Tables S1 and S2 report
the SAR data relative to two other hits reported in Figure 2. Figure
1
S3 consists of H and 13C NMR spectra for synthesized FKBP12
ligands. This material is available free of charge via the Internet at
http://pubs.acs.org.
(21) Xiao, H.; Wang, L. L.; Shu, C. L.; Yu, M.; Li, S.; Shen, B. F.; Li, Y.
Establishment of a cell model based on FKBP12 dimerization for
screening of FK506-like neurotrophic small molecular compounds.
J. Biomol. Screening 2006, 11 (3), 225–235.
(22) Hajduk, P. J.; Greer, J. A decade of fragment-based drug design:
Strategic advances and lessons learned. Nat. ReV. Drug DiscoVery
2007, 6 (3), 211–219.
(23) Pellecchia, M. Solution nuclear magnetic resonance spectroscopy
techniques for probing intermolecular interactions. Chem. Biol. 2005,
12 (9), 961–971.
(24) Pellecchia, M.; Becattini, B.; Crowell, K. J.; Fattorusso, R.; Forino,
M.; Fragai, M.; Jung, D.; Mustelin, T.; Tautz, L. NMR-based
techniques in the hit identification and optimisation processes. Expert
Opin. Ther. Targets 2004, 8 (6), 597–611.
(25) Shuker, S. B.; Hajduk, P. J.; Meadows, R. P.; Fesik, S. W. Discovering
high-affinity ligands for proteins: SAR by NMR. Science 1996, 274
(5292), 1531–1534.
(26) Hajduk, P. J.; Augeri, D. j.; Mack, J.; Mendoza, R.; Yang, J.; Betz,
S. F.; Fesik, S. W. NMR-based screening of proteins containing 13C-
labeled methyl groups. J. Am. Chem. Soc. 2000, 122, 7898–7904.
(27) Pellecchia, M.; Meininger, D.; Dong, Q.; Chang, E.; Jack, R.; Sem,
D. S. NMR-based structural characterization of large protein-ligand
interactions. J. Biomol. NMR 2002, 22 (2), 165–173.
References
(1) Kay, J. E. Structure-function relationships in the FK506-binding
protein (FKBP) family of peptidylproplyl cis-trans isomerases.
Biochem. J. 1996, 314 (2), 361.
(2) Harrar, Y.; Bellini, C.; Faure, J.-D. FKBPs: At the crossroads of folding
and transduction. Trends Plant Sci. 2001, 6 (9), 426.
(3) Steiner, J. P.; Hamilton, G. S.; Ross, D. T.; Valentine, H. L.; Guo,
H.; Connolly, M. A.; Liang, S.; Ramsey, C.; Li, J. H.; Huang, W.;
Howorth, P.; Soni, R.; Fuller, M.; Sauer, H.; Nowotnik, A. C.; Suzdak,
P. D. Neurotrophic immunophilin ligands stimulate structural and
functional recovery in neurodegenerative animal models. Proc. Natl.
Acad. Sci. U.S.A. 1997, 94 (5), 2019–2024.
(4) Gold, B. G.; Katoh, K.; Storm-Dickerson, T. The immunosuppressant
FK506 increases the rate of axonal regeneration in rat sciatic nerve.
J. Neurosci. 1995, 15 (11), 7509–16.
(5) Lyons, W. E.; George, E. B.; Dawson, T. M.; Steiner, J. P.; Snyder,
S. H. Immunosuppressant FK506 promotes neurite outgrowth in
cultures of PC12 cells and sensory ganglia. Proc. Natl. Acad. Sci.
U.S.A. 1994, 91 (8), 3191–3195.
(6) Sharkey, J.; Butcher, S. P. Immunophilins mediate the neuroprotective
effects of FK506 in focal cerebral ischaemia. Nature 1994, 371 (6495),
336–339.
(7) Caffrey, M. V.; Cladingboel, D. E.; Cooper, M. E.; Donald, D. K.;
Furber, M.; Harden, 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 (21), 2507–2510.
(8) Hauske, J. R.; Dorft, P.; Julin, S.; DiBrino, J.; Spencer, R.; Williams,
R. Design and synthesis of novel FKBP inhibitors. Med. Chem. 1993,
35, 4248–4296.
(9) Holt, D. A.; Konialian-Beck, A. L.; Hye-Ja-Oh; Yen, K.-K.; Rozamus,
L. W.; Kroga, 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-propyl isomerase inhibitors. Bioorg. Med.
Chem. Lett. 1994, 4 (2), 315–320.
(28) Hamilton, G. S.; Steiner, J. P. Immunophilins: Beyond immunosup-
pression. J. Med. Chem. 1998, 41 (26), 5119–5143.
(29) Wang, X. J.; Etzkorn, F. A. Peptidyl-prolyl isomerase inhibitors.
Biopolymers 2006, 84 (2), 125–146.
(30) Yaffe, M. B.; Schutkowski, M.; Shen, M.; Zhou, X. Z.; Stukenberg,
P. T.; Rahfeld, J. U.; Xu, J.; Kuang, J.; Kirschner, M. W.; Fischer,
G.; Cantley, L. C.; Lu, K. P. Sequence-specific and phosphorylation-
dependent proline isomerization: A potential mitotic regulatory mech-
anism. Science 1997, 278 (5345), 1957–1960.
(31) Zhou, X. Z.; Lu, P. J.; Wulf, G.; Lu, K. P. Phosphorylation-dependent
prolyl isomerization: A novel signaling regulatory mechanism. Cell.
Mol. Life Sci. 1999, 56 (9–10), 788–806.
(32) Goddard, T. D.; Kneller, D. G. SPARKY 3; University of CA, San
Francisco, 2007.
(33) Threadgill, R.; Bobb, K.; Ghosh, A. Regulation of dendritic growth
and remodeling by Rho, Rac, and Cdc42. Neuron 1997, 19 (3), 625–
634.
(34) Jones, G.; Willett, P.; Glen, R. C. Molecular recognition of receptor
sites using a genetic algorithm with a description of desolvation. J.
Mol. Biol. 1995, 245 (1), 43–53.
(10) Holt, D. A.; Luengo, J. I.; Yamashita, D. S.; Hye-Ja-Oh.; Konialian-
Beck, A. L.; Yen, H.-K.; Rozamus, L. W.; Brandt, M.; Bossard, M. J.;
Levy, M. A.; Eggleston, D. S.; Liang, J.; Shultz, 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
FKPB12. J. Am. Chem. Soc. 1993, 115, 9925–9938.
(11) Stocks, M. J.; Birkinshaw, T. N.; Teague, S. J. The contribution to
binding of the pyranoside substituents in the excised bindind domain
of FK-506. Bioorg. Med. Chem. Lett. 1994, 4 (12), 1457–1460.
(12) Teague, S. J.; Cooper, M. E.; Donald, D. K.; Furber, M. Synthesis
and study on nonmacrocyclic FK506 derivative. Bioorg. Med. Chem.
Lett. 1994, 4 (13), 1581–1584.
(13) Teague, S. J.; Stocks, M. J., The affinity of the excised binding domain
of FK-506 for the Immunophilin FKBP12. Bioorg. Med. Chem. Lett.
1993, 3 (10), 1947–1950.
(35) Jones, G.; Willett, P.; Glen, R. C.; Leach, A. R.; Taylor, R.
Development and validation of a genetic algorithm for flexible docking.
J. Mol. Biol. 1997, 267 (3), 727–748.
(36) Teschner, M.; Henn, C.; Vollhardt, H.; Reiling, S.; Brickmann, J.
Texture mapping: A new tool for molecular graphics. J. Mol. Graph.
1994, 12 (2), 98–105.
(37) Pearlman, R. S. CONCORD; Tripos: St. Louis, MO, 1998.
JM0707424