Dimerization of a heat shock protein 90 inhibitor enhances inhibitory activity

Article abstract of DOI:10.1039/c3ob41722k

Heat shock protein 90 (hsp90) accounts for 1-2% of the total proteins in normal cells and it functions as a dimer. Hsp90 behaves as a molecular chaperone that folds, assembles, and stabilizes client proteins. We have developed a novel hsp90 inhibitor, and

Full text of DOI:10.1039/c3ob41722k

Organic &
Biomolecular Chemistry
View Article Online
View Journal | View Issue
PAPER
Dimerization of a heat shock protein 90 inhibitor
enhances inhibitory activity†
Cite this: Org. Biomol. Chem., 2014,
12, 765
Hendra Wahyudi, Yao Wang and Shelli R. McAlpine*
Heat shock protein 90 (hsp90) accounts for 1–2% of the total proteins in normal cells and it functions as
a dimer. Hsp90 behaves as a molecular chaperone that folds, assembles, and stabilizes client proteins. We
have developed a novel hsp90 inhibitor, and herein we describe the synthesis and biological activity of the
dimerized variant of this inhibitor. Tethering a monomer inhibitor together produced a dimerized com-
pound that more effectively inhibits hsp90 over the monomer.
Received 22nd August 2013,
Accepted 5th November 2013
DOI: 10.1039/c3ob41722k
www.rsc.org/obc
to hsp90.18–34 Of these inhibitors, SM122 (Fig. 2) displayed
reasonable cytotoxicity (IC50 = 3.5 µM against pancreatic
Introduction
Heat shock protein 90 (hsp90) is a ubiquitous molecular cha- cancer cell line PL45; IC50 = 3.9 µM against colon cancer cell
perone that facilitates the assembly and regulates the folding line HCT-116) and it had a unique phenotype of allosterically
of proteins that are important in molecular signalling events.1 inhibiting access to hsp90’s C-terminal domain. SM122 has
In normal cells, expression of hsp90 accounts for 1–2% of cyto- already been reported to preferentially bind to hsp90’s closed
solic protein, whereas in cancer cells it is 3–6%.1–7 This over- twisted conformation (Fig. 1); based on FP anisotropy of
expression of hsp90 is primarily driven by high levels of cell SM122 against Hsc82 (full-length yeast homolog of hsp90),
signalling events that occur during oncogenic cell growth, Kd of SM122 to open state Hsc82 is 103 46 μM versus 55
which causes stress and leads to the heat shock response 21 μM in closed-stated conformation.31 Thus, we investigated
being induced (including hsp90). To date, there are over 100 the hypothesis of whether the combination of two
client proteins associated with hsp90 that are involved in the SM122 molecules using a polyethylene glycol (PEG) chain
molecular signalling pathways of cancer cells.1,8–12 Hsp90 would enhance the inhibitory activity against hsp90 dimer.29,34
functions as a dimer and it contains three domains: N-, Regulating signal transduction events of two proteins was
middle- and C-domains.13,14 The ATP binding site (located in successfully accomplished by Scheiber et al. utilizing dimer-
the N-domain) is the target for all hsp90 inhibitors in clinical ized molecules.35,36 This strategy was exemplified in the syn-
trials.15 The major drawback to these clinical candidates, thesis of FK1012, a dimer variant of immunosuppressant
including 17AAG (our positive control) is the induction of a FK506, where the dimer diminished the toxicity of FK506
heat shock response that occurs when the inhibitors bind to associated with calcineurin inhibition, but retained immuno-
hsp90’s ATP-binding domain.16 In order to avoid the heat suppressant activity at considerably lower doses than FK506.35
shock rescue response, compounds that do not target the ATP-
binding site are utilized, and several examples of allosteric
modulators and C-terminal inhibitors have shown that the
heat shock response is not induced when modulating the
C-terminus, but only when blocking the ATP binding site of
the N-terminus.17–20
We have developed several hsp90 inhibitors, all of which
bind to hsp90 between its N-middle domain, and allosterically
modulate C-terminal clients and co-chaperones from binding
Department of Chemistry, University of New South Wales, Sydney, NSW 2052,
Australia. E-mail: s.mcalpine@UNSW.edu.au; Fax: +61-2-9385-6111;
Tel: +61-4-1672-8896
†Electronic supplementary information (ESI) available: 1H-, 13C-NMR spectra,
LCMS, HRMS, solubility, luciferase assay and protein binding assay data for
SM122-PEG conjugate and DIMERs are provided. See DOI: 10.1039/c3ob41722k
Fig. 1 Design element behind dimerizing SM122. Our goal is to trap the
closed-twisted conformation of hsp90.
This journal is © The Royal Society of Chemistry 2014
Org. Biomol. Chem., 2014, 12, 765–773 | 765

View Article Online
Paper
Organic & Biomolecular Chemistry
Fig. 2 Development of SM122 variants: structures of SM122, and
SM122–PEG conjugate or SM122–PEG DIMER.
In 2011, dimerisation of novobiocin, an hsp90 C-terminal
inhibitor, utilizing variety of linker lengths and types produced
coumermycin A1 analogues that were more potent and selec-
tive than monomeric novobiocin.37 Considering that hsp90
functions as a dimer, one strategy for enhancing SM122 as a
clinical candidate is to halt hsp90 function by stabilizing the
protein dimer (Fig. 1) using a dimerized molecule that pre-
ferred the closed-twisted conformation.
Herein we describe our dimerization approach for advan-
cing our novel hsp90 inhibitor, SM122, into a more effective
inhibitor over the parent compound. Dimerizing SM122 using
PEG linkers (Fig. 2), we tested our compounds in hsp90 activity
assays (luciferase refolding assays) in order to evaluate their
activity compared to the parent SM122. We also prepared
control compounds SM122–PEG-4 and SM122–PEG-8 that were
monomers with PEG (Fig. 2). We observed that the inhibitory
effects of DIMERs on hsp90 chaperone function are signifi-
cantly greater than either the monomer or the PEG-monomer
controls (∼2-fold, Fig. 3a). Examining the compounds’ ability
to inhibit binding between hsp90 and its essential co-chaper-
one proteins HOP and FKBP52 showed that the inhibitory
effects of SM122–PEG DIMER is 18–39% greater than SM122–
PEG or SM122.
Results and discussion
Design
Previous work on SM122 had shown that replacing a leucine
with a lysine and attaching the compound to a PEG-biotin
moiety did not inhibit binding of SM122 to hsp90.29,34 Thus,
we chose this position to attach the PEG linker. In the syn-
thesis of SM122–PEG conjugate, two different PEG chains
(4-PEG and 8-PEG units) were chosen to examine the impact of
the PEG length on the hsp90 binding and as controls for the
Fig. 3 (a–c) Luciferase refolding assay: comparison of compounds’
inhibitory activity using an hsp90-mediated protein folding assay
(*P-value ≤ 0.05).
766 | Org. Biomol. Chem., 2014, 12, 765–773
This journal is © The Royal Society of Chemistry 2014

View Article Online
Organic & Biomolecular Chemistry
Paper
dimerized SM122 compounds.38 In the SM122–PEG DIMER preloaded 2-chlorotrityl-phenylalanine (Phe) resin. Subsequent
synthesis, we used two different linker lengths, 5-PEG and coupling of Fmoc-protected amino acids and amine deprotec-
9-PEG units, in order (a) to correlate to the monomer-PEG tions led to a solid phase bound linear analog. Cleavage of the
analogs (4-PEG and 8-PEG) and (b) to probe the optimal linker compound from the resin and cyclization yielded 1. Deprotec-
length that would facilitate binding to hsp90.
tion of the lysine residue generated 2.
We chose the PEG length based upon the crystal structure
Conversion of 2 into SM122–PEG was accomplished by
of full yeast hsp90-p23/Sba1 complex as reported by coupling a methyl-capped PEGylated (4 or 8 PEG units)
Pearl.18,39–41 Lee et al. also showed that in an open state of N-hydroxysuccinimidyl ester to 2 (1.0 equivalents of the linker,
hsp90, the inner distance between the Middle C-domains are Scheme 2). Generating the dimer was accomplished in a
30 Å apart, and the outer distance is 70 Å apart.42 Based on the similar manner, but utilized a PEGylated (5 or 9 PEG units)
work of Agard et al.40 the open conformation of hsp90’s homobifunctional bis-N-hydroxysuccinimidyl ester to
N-middle domain is estimated to be 80 Å apart. Since the (0.5 equivalents of the linker, Scheme 2).
crystal structure of human hsp90 in the closed twisted confor-
2
Hsp90-dependent luciferase refolding assay
mation is currently unknown, we selected PEG lengths
∼30–40 Å (5-PEG and 9-PEG units), which approximately equal
to the inner distance or half of the outer distance of the
N-middle domain of the two monomers of one hsp90
protein.34,40
Assessing the direct inhibition of the hsp90 protein folding
machinery with our compounds provides information on
which analogue binds most effectively to hsp90 and inhibits
its protein folding function. Utilizing an hsp90-dependent
luciferase-refolding assay with rabbit reticulocyte lysate (RRL)
refolding system produces a luminescent signal when hsp90 is
active and functioning. Addition of hsp90 inhibitors to the
RRL system shows a decrease in luminescence that is directly
correlated to the compound’s inhibition activity of hsp90.
Our data show that DMSO (negative control, black line,
Fig. 3a) behaved similar to the PEG conjugates (SM122–PEG-4
and SM122–PEG-8, red and orange lines respectively, Fig. 3a)
and the PEG motifs alone (i.e. no compound, see MS (PEG)s
and BS (PEG)s, ESI†). Both 17-AAG and SM122 inhibited hsp90
activity (light blue and dark blue lines respectively, Fig. 3a).
However, the inhibitory activity was significantly improved
when using the dimers SM122–PEG-5 DIMER and SM122–
PEG-9 DIMER (dark and light green respectively, Fig. 3a). Sig-
nificantly, SM122–PEG-5 DIMER and SM122–PEG-9 DIMER are
∼15- to 18-fold more effective than their SM122–PEG conju-
gates and PEG motifs respectively (Table, Fig. 3a). Our data
indicate that the inhibitory activity against the hsp90-mediated
Synthesis
Synthesis of 2, a precursor for both SM122–PEG and SM122–
PEG DIMER, was afforded via solid phase peptide synthesis
(Scheme 1). Using a synthetic approach previously reported for
making biotinylated tagged derivatives29 we started with
Scheme 1 Solid phase peptide synthesis of compound 2, a precursor
to SM122–PEG conjugate and SM122–PEG DIMER.
Scheme 2 Synthesis of SM122–PEG conjugates and SM122–PEG
DIMERs.
This journal is © The Royal Society of Chemistry 2014
Org. Biomol. Chem., 2014, 12, 765–773 | 767

View Article Online
Paper
Organic & Biomolecular Chemistry
refolding of denatured luciferase is due to the dimerization of
SM122 rather than an increase the solubility effect of PEG.
Indeed a deleterious effect is seen when adding PEG to the
compound. We believe that this is due to the inherent “sticky”
quality of PEG, which blocks access to the compound if not
utilized for compound binding.
Running additional luciferase assays (Fig. 3b), where
multiple compound concentrations were run, clearly show the
same trends where even at twice the concentration of the
dimer, SM122 never reaches the inhibitory ability of SM122–
PEG-9. Specifically, comparing SM122 (1 µM) and SM122–
PEG-9 DIMER (0.5 µM) shows that the DIMER is still signifi-
cantly more active than the monomer despite the same
number of SM122 molecules being present (Fig. 3c). This
trend is also observed at 5 µM of SM122 and 2.5 µM of DIMER.
Taken together, these results provide strong evidence that
dimerization is a successful way to improve the efficiency of
SM122 molecule in suppressing the chaperone function of
hsp90 protein acting as a dimer.
Protein binding assay
In order to evaluate the direct impact of these compounds on
hsp90, we examined their efficiency in blocking the binding
interaction between hsp90 and HOP (hsp organizing protein)
as well as hsp90 and FKBP52 (an immunophilin that regulates
hormone receptors). It has been demonstrated that both pro-
teins, HOP and FKBP52, bind to the C-terminus of hsp90, and
SM122 blocks the interaction between hsp90 and both of these
proteins.18,43 Thus, this is an ideal assay to compare the effec-
tiveness of SM122–PEG DIMER and SM122. Protein binding
assays showed that 10 µM of SM122–PEG-9 DIMER was 1.3-
fold more effective in inhibiting binding between hsp90 and
HOP than 20 μM of SM122 (P-value = 0.002, Fig. 4a), indicating
that the DIMER is more efficient at interfering with the hsp90–
HOP interaction than SM122 monomer alone. Additional evi-
dence supports these conclusions, where 10 µM of SM122–
PEG-9 DIMER blocked the binding between hsp90-FKBP52 by
1.4-fold greater than 20 μM of SM122 alone (Fig. 4b).
Interestingly, both the 20 and 10 µM SM122 have the same
effect at blocking the binding between hsp90 and HOP, which
suggests that the monomer compound has maximized its
binding. This could be related to the solubility, however, as
discussed in the next section, because PEG dramatically
increases their solubility, addition of PEG to the monomer
should dramatically improve SM122’s ability to block binding.
As is observed in both Fig. 4a and 4b, addition of PEG does
not enhance the activity of SM122 for either protein with
hsp90 (compare SM122 activity versus SM122–PEG-8).
Fig. 4 The inhibitory effects of SM122, SM122–PEG-8 and SM122–
PEG-9 DIMER on the binding between hsp90 and its co-chaperone (a)
HOP and (b) FKBP52 (*P-value ≤ 0.05).
Table 1 Solubility of SM122, SM122–PEG conjugates, and SM122–PEG
DIMERs
Compound
(μg mL−1
)
(μM)
SM122
SM122–PEG-4
SM122–PEG-5 DIMER
SM122–PEG-8
SM122–PEG-9 DIMER
0.66 0.28
6.45 0.36
7.00 0.71
198.64 3.89
10.18 1.72
0.84 0.35
6.36 0.35
3.69 0.37
166.69 3.27
4.91 0.83
Table 1 below) show that SM122–PEG-8, the least active hsp90
inhibitor, was the most soluble compound, identified as ∼200-
fold more soluble than SM122. In contrast, the DIMERs
(SM122–PEG-5 DIMER and SM122–PEG-9 DIMER) were only
∼4- and ∼6-fold respectively more soluble than SM122. The
PEG monomer compounds (SM122–PEG-4 and PEG-8 respect-
ively) were completely ineffective in the luciferase assay. These
indicate that solubilizing the compound is not enough to
produce activity. Given that the RRL luciferase assay is run
with large quantities of cytosol, it is likely that the sticky PEG
Solubility determination
Since the PEG linker is known to increase solubility of drug
molecules, one hypothesis is that the increase in solubility of
the DIMER is responsible for the observed improvement of
activity. Thus, the solubility of SM122–PEG conjugates and
DIMERs were determined. The method for solubility evalu-
ation is annotated in the ESI.† The results (summarized in
768 | Org. Biomol. Chem., 2014, 12, 765–773
This journal is © The Royal Society of Chemistry 2014

View Article Online
Organic & Biomolecular Chemistry
Paper
linkers block the ability of SM122 to find hsp90 and inhibit its (Fig. 5, Model A). That is, the two SM122 molecules in the PEG
function. dimer are not acting as two separate SM122 molecules. If they
In contrast, biochemical assays, which did not have sticky had been behaving independently, we would see that the
cytosolic proteins in the assays, showed that the SM122–PEG-8 dimer would be twice as effective at the same concentration
analog was as effective at inhibiting the binding between HOP and at half the concentration the dimer would be as effective
and hsp90 (Fig. 4a). These data support the conclusion that as the monomer. As discussed above this is not the case, in all
solubility plays little role in the effectiveness of the com- assays, the dimer is more effective than the monomer regard-
pounds. Interestingly, the effect of PEG does appear to be dele- less of concentration, which is indicative of a synergistic
terious when comparing the binding of 20 µM SM122 to binding event of both SM122 molecules tethered together
SM122–PEG-8 but not when comparing the activity of 10 µM of (Model B, Fig. 5).
each molecule. In both cases, however, the data indicate that
In summary, we have shown that dimerizing SM122, a novel
solubility plays little- to no-role in the biological activity of the hsp90 inhibitor, will effectively modulate hsp90’s activity. Fur-
molecules. Finally, the DIMER, although not as soluble as the thermore, we have shown that dimerization of SM122 produces
PEG monomer, is significantly more active in both binding a synergistic inhibitor, that more effectively inhibits hsp90
assays.
activity from refolding proteins, and blocks binding between
client and co-chaperone proteins and hsp90 than monomer
alone. We also show that the DIMER effectiveness is not
related to its improved solubility, and that the DIMER is most
effective when bound to the same hsp90 dimer versus two
separate hsp90 dimers. These studies demonstrate that dimer-
izing an hsp90 inhibitor is more effective at blocking hsp90
activity than using monomer alone.
Dimer binding mode
The data observed from luciferase refolding assay and protein
binding assay also revealed the possible interactions that
occurred when SM122–PEG DIMER bound to hsp90. Based on
the cooperative binding theory, when one SM122 unit binds to
the hsp90 dimer it may induce an allosteric modulation across
one side of the dimer protein, resulting in a conformational
shift that allows a second SM122 molecule to bind to the other
side of the hsp90 dimer with lower binding energy than
monomer alone. Alternatively, the binding event of SM122 can
compromise the secondary binding event. In our case, indu-
cing a positive binding event and lowering the energy for a
second SM122 molecule to bind explains the data observed in
Fig. 3b. Specifically, SM122 (1 µM) and SM122–PEG-9 DIMER
(0.5 µM) and SM122 (5 µM) and SM122–PEG-9 DIMER
(2.5 µM) both show that the DIMER is still significantly more
active than the monomer despite the same number of SM122
molecules being present (Fig. 3c). Thus, the dimer is more
effective than two equivalents of the monomer (Model B,
Fig. 5).
Experimental
General information
All reactions were carried out under N2 atmosphere with dry
solvent under anhydrous conditions, unless indicated other-
wise. Reagents were commercially obtained without further
purification, unless otherwise stated. Reaction was monitored
via thin-layer chromatography (TLC) carried out on silica gel
plates using UV light at λ = 254 nm for visualization, and
potassium permanganate in water with heat and developing
agents. Silica gel was used for flash chromatography. NMR
spectra were obtained at 298 K. LC/MS was recorded on an
LCMS system connected to a trap running in positive electro-
spray ionization (ESI+) mode. The mobile phase was composed
of DDI water with 0.1% (v/v) formic acid (solvent A), and HPLC
grade acetonitrile with 0.1% (v/v) formic acid (solvent B). The
gradient elution was conducted as follows: flow rate 0.5 mL
min−1; initial 70% solvent A, 30% solvent B; at 4 minutes
100% solvent B; at 12 minutes 70% solvent A, 30% solvent
B. Semipreparative reversed-phase HPLC was carried out on an
LCMS system. The mobile phase was composed of DDI water
with 0.1% (v/v) formic acid (solvent A), and HPLC grade aceto-
nitrile with 0.1% (v/v) formic acid (solvent B). The gradient
elution was conducted as follows: flow rate 2 mL min−1; initial
70% solvent A, 30% solvent B hold for 35 minutes; at
35 minutes 100% solvent B hold for 13 minutes; at 48 minutes
70% solvent A, 30% solvent B hold for 2 minutes.
An alternative theory is that Model A (Fig. 5) may exist.
However, since the DIMER showed higher inhibitory efficacy
than the monomer at the same concentration levels (Fig. 3c), a
single SM122–PEG DIMER must bind to more than one hsp90
dimer protein. However our data shows that the two “mono-
mers” of the SM122–PEG dimer are not acting independently
General solid phase synthesis remarks
Stepwise solid phase peptide synthesis was performed in a
polypropylene solid-phase extraction cartridge fitted with a
20 µm polyethylene frit purchased from Applied Separations
Fig. 5 Dimer binding mode: Model A versus Model B.
This journal is © The Royal Society of Chemistry 2014
Org. Biomol. Chem., 2014, 12, 765–773 | 769

View Article Online
Paper
Organic & Biomolecular Chemistry
(Allentown, PA). 2-Chlorotrityl resins were purchased in Macrocyclization procedure (syringe pump)
pre-loaded form with L-Phe. Resins were swelled in dimethyl-
Three coupling agents (DMTMM, HATU, and TBTU) were used
formamide (DMF) for 30 minutes prior to assembly of the
linear five-residue peptide sequence. Solid-phase syntheses
were performed on a 0.5 mmol scale based on resin-loading.
All operations were performed at room temperature under
open atmosphere unless stated otherwise.
at 0.8 equiv. each. These coupling agents were dissolved in 4/5
of a calculated volume of dry CH2Cl2 that would give a 0.001 M
overall concentration when included in the volume used for
the deprotected peptide. The crude, dry, double deprotected
peptide (free acid and free amine) was dissolved in the other
1/5 solvent volume of CH2Cl2. DIPEA (8.0 equiv.) was then
added to the solution containing coupling reagents dissolved
in CH2Cl2. The double deprotected peptide was then added to
the bulk solution dropwise using a syringe pump at a rate of
30 mL h−1. The reaction was monitored via LCMS and gene-
rally complete in 1–2 hours. Upon completion, the reaction
was worked up by washing with aqueous HCl (pH 1) and satu-
rated sodium bicarbonate. After back extraction of aqueous
layers with large quantities of CH2Cl2, the organic layers were
combined, dried, filtered and concentrated. All macrocycles
were first purified by flash column chromatography using an
ethyl acetate–hexane gradient on silica gel. Finally, when
necessary, reversed-phase HPLC was used for additional purifi-
cation using a gradient of acetonitrile and deionized water
with 0.1% TFA.
General solid phase peptide synthesis
Fmoc-protected amino acids were coupled using 3.0 equiv. of
amino acid, 3.0 equiv. of 1-hydroxybenzotriazole (HOBt), and
6.0 equiv. of diisopropylcarbodiimide (DIC). Couplings were
performed in DMF at 0.2 M with respect to the incoming
Fmoc-protected amino acid. Couplings were allowed to
proceed for a minimum of 2 hours, and were assayed via nin-
hydrin test to verify completion. Once complete, the coupling
reaction solution was drained, and the resin subjected to
Fmoc deprotection. (Note: Fmoc and N-methyl amino acids are
coupled according to the cycle above, however for subsequent
coupling onto the secondary amino terminus, 1-hydroxybenzo-
triazole was substituted with 1-hydroxy-7-azabenzotriazole
(HOAt) and the coupling was allowed to proceed overnight.)
Synthesis
General solid phase amine deprotection
Macrocycle Phe-D-N-Me-Phe-Val-Leu-Lys (Cbz) (SM122). Syn-
thesis of Macrocycle Phe-D-N-Me-Phe-Val-Leu-Lys (Cbz) was
referenced from the published data (Sellers, R. P., Alexander,
L. D., Johnson, V. A., Lin, C.-C., Savage, J., Corral, R., Moss, J.,
Slugocki, T. S., Singh, E. K., Davis, M. R., et al. A third gene-
ration of Sansalvamide A derivatives: Design and synthesis of
Hsp90 Inhibitors. Bioorg. Med. Chem., 2010, 18, 6822–6856;
compound 28).
Following coupling completion, the peptide-resin was treated
as follows for removal of the Fmoc protecting group: DMF
wash (3 × 1 minute), 20% piperidine–DMF (1 × 5 minutes),
20% piperidine–DMF (1 × 10 minutes), DMF wash (2 ×
1 minute), IPA wash (1 × 1 minute), DMF wash (1 × 1 minute),
IPA (1 × 1 minute), DMF (3 × 1 minute). A ninhydrin test was
performed to verify completion.
Macrocycle Phe-N-Me-D-Phe-Val-Lys (Boc)-Lys (Cbz) (com-
pound 1). Synthesis of Macrocycle Phe-D-N-Me-Phe-Val-Lys
(Boc)-Lys (Cbz) was referenced from the published data
(Kunicki, J. B., Petersen, M. N., Alexander, L. D., Ardi, V. C.,
McConnell, J. R., and McAlpine, S. R. Synthesis and Evaluation
of Biotinylated Sansalvamide A Analogs and their Modulation
of Hsp90. Bioorg. Med. Chem. Lett., 2011, 21, 4716–4719; com-
pound 2-T-IV).
General N-terminal solid phase amine deprotection
Once the final N-terminal amino acid residue had been
coupled, the peptide-resin was treated as follows for removal
of the Fmoc protecting group: DMF wash (3 × 1 minute), 20%
piperidine–DMF (1 × 5 minutes), 20% piperidine–DMF (1 ×
10 minutes), DMF wash (3 × 1 minute), IPA wash (3 ×
1 minute), MeOH (3 × 1 minute). The fully-assembled peptide-
resin was then drained and dried in vacuo overnight.
Macrocycle Phe-N-Me-D-Phe-Val-Lys-Lys (Cbz) (compound 2)
Cleavage of linear peptide
Macrocycle Phe-N-Me-D-Phe-Val-Lys-Lys (Cbz) was synthesized
by removing the Boc from macrocycle Phe-N-Me-D-Phe-Val-Lys
(Boc)-Lys (Cbz). Utilizing 206 mg (0.23 mmol, 1.0 equiv.) of
Phe-N-Me-D-Phe-Val-Lys (Boc)-Lys (Cbz), 0.46 mL of trifluoro-
acetic acid, 1.84 mL of CH2Cl2 (183.5 mg, Quant. yield).
The full-length, linear peptide was cleaved from the resin by
swelling and shaking the peptide-resin for 24 hours in a 1 : 1
(v/v) 2,2,2-trifluoroethanol (TFE)–CH2Cl2 (10 volumes per gram
of dried resin). The cleavage solution was filtered through a
Buchner filter, and the drained resin was washed with
additional CH2Cl2 (5 volumes per gram of initial dried
Compound SM122–PEG-5 DIMER
peptide-resin) to fully extract the cleaved peptide from the Utilizing 102 mg (0.13 mmol, 1.0 equiv.) of Phe-N-Me-D-Phe-
resin. Solvents in the combined filtrates were evaporated by Val-Lys-Lys (Cbz), 68.0 μL (64 μmol, 0.5 equiv.) BS(PEG)5, and
rotary evaporation and the solids dried in vacuo overnight. The 0.18 mL (8.0 equiv.) of DIPEA dissolved in 13.0 mL CH2Cl2,
solids were then reconstituted in CH2Cl2, evaporated by rotary under N2. The reaction was monitored via LCMS and generally
evaporation and dried in vacuo overnight again to remove complete in 1–2 hours. Upon completion, the reaction was
residual entrapped TFE.
worked up by washing with aqueous HCl (pH 1) and saturated
770 | Org. Biomol. Chem., 2014, 12, 765–773
This journal is © The Royal Society of Chemistry 2014

View Article Online
Organic & Biomolecular Chemistry
Paper
sodium bicarbonate. After back extraction of aqueous layers 2H, NH); 7.01–7.37 (m, 30H, Ph); 7.65–7.72 (m, 2H, NH).
with large quantities of CH2Cl2, the organic layers were com- δC NMR (75 MHz, CDCl3): δ 18.0, 19.6, 22.7, 23.4, 28.9, 29.2,
bined, dried, filtered and concentrated. The crude material 29.6, 30.3, 30.5, 30.7, 32.9, 36.8, 37.8, 38.2, 40.8, 54.0, 54.9,
was purified using column chromatography (silica gel, EtOAc– 55.3, 57.1, 57.3, 59.0, 66.5, 67.3, 70.5, 71.8, 126.9, 127.9, 128.0,
MeOH) to yield 70 mg (58% yield) of the compound. Rf: 0.39 128.5, 128.6, 128.7, 128.8, 129.4, 136.1, 136.5, 136.8, 156.7,
(EtOAc–MeOH 9 : 1); δH NMR (300 MHz, CDCl3): δ 0.80–0.85 169.7, 171.7, 172.1, 172.7; LC/MS (ESI): m/z called for
(d, J = 6.71 Hz, 6H, CHCH(CH3)2); 0.88–0.94 (d, J = 6.71 Hz,
6H, CHCH(CH3)2); 0.95–1.08 (m, 4H, CH2γ Lys (PEG)); HRMS (ESI-TOF):
1.27–1.30 (m, 4H, CH2δ Lys (Cbz)); 1.34–1.46 (m, 4H, CH2γ Lys
(Cbz)); 1.48–1.58 (m, 4H, CH2δ Lys (PEG)); 1.69–1.82 (m, 4H,
C
110H156N14O25 (M + H+) = 2073.14, found 1038.00 (half-mass);
Na+ H+ found 2097.1290
110H157N14O25Na requires 2097.1342.
Compound SM122–PEG-4. Utilizing 124 mg (0.16 mmol,
M
+
+
C
CH2β Lys (PEG)); 1.92–2.03 (m, 4H, CH2β Lys (Cbz)); 2.14–2.23 1.0 equiv.) of Phe-N-Me-D-Phe-Val-Lys-Lys (Cbz), 155.0 μL
(m, 4H, CH2β Val); 2.47–2.50 (t, 4H, COCH2 (PEG)); 2.83–2.88 & (155 μmol, 1.0 equiv.) MS (PEG)4, and 0.22 mL (8.0 equiv.) of
3.05–3.14 (m, 4H, CH2β Phe); 2.85–2.92 (s, 6H, NMe); DIPEA dissolved in 20.0 mL CH2Cl2, under N2. The reaction
2.96–3.02 & 3.14–3.24 (m, 4H, CH2β D-Phe); 3.04–3.20 (m, 4H, was monitored via LCMS and generally complete in 1–2 hours.
CH2ε Lys (PEG)); 3.15–3.30 (m, 4H, CH2ε Lys (Cbz)); 3.54–3.60 Upon completion, the reaction was worked up by washing with
(m, 4H, CH2 (PEG)); 3.61–3.78 (m, 12H, CH2 (PEG)); 3.75–3.83 aqueous HCl (pH 1) and saturated sodium bicarbonate. After
(m, 2H, CHα Lys (PEG)); 3.93–4.01 (m, 2H, CHα Lys (Cbz)); back extraction of aqueous layers with large quantities of
4.44–4.52 (m, 2H, CHα Val); 4.72–4.82 (m, 2H, CHα Phe); CH2Cl2, the organic layers were combined, dried, filtered and
5.07–5.13 (s, 4H, CH2 Cbz); 5.16–5.23 (m, 2H, CHα D-Phe); concentrated. The crude material was purified using column
5.49–5.57 (m, 2H, NH); 6.80–6.87 (m, 2H, NH); 6.87–6.95 (m, chromatography (silica gel, EtOAc–MeOH) to yield 116 mg
2H, NH); 7.01–7.37 (m, 30H, Ph); 7.65–7.72 (m, 2H, NH). (73.5% yield) of the compound. Rf: 0.35 (EtOAc–MeOH 19 : 1).
δC NMR (75 MHz, CDCl3): δ 18.0, 19.6, 22.7, 23.4, 28.9, 29.2, δH NMR (300 MHz, CDCl3): δ 0.80–0.85 (d, J = 6.71 Hz, 3H,
29.6, 30.3, 30.5, 30.7, 32.9, 36.8, 37.8, 38.2, 40.8, 54.0, 54.9, CHCH(CH3)2); 0.88–0.94 (d, J = 6.71 Hz, 3H, CHCH(CH3)2);
55.3, 57.1, 57.3, 59.0, 66.5, 67.3, 70.5, 71.8, 126.9, 127.9, 128.0, 0.95–1.08 (m, 2H, CH2γ Lys (PEG)); 1.27–1.30 (m, 2H, CH2δ Lys
128.5, 128.6, 128.7, 128.8, 129.4, 136.1, 136.5, 136.8, 156.7, (Cbz)); 1.34–1.46 (m, 2H, CH2γ Lys (Cbz)); 1.48–1.58 (m, 2H,
169.7, 171.7, 172.1, 172.7; LC/MS (ESI): m/z called for CH2δ Lys (PEG)); 1.69–1.82 (m, 2H, CH2β Lys (PEG)); 1.92–2.03
C
102H140N14O21 (M + H+) = 1897.03, found 950.00 (half-mass); (m, 2H, CH2β Lys (Cbz)); 2.14–2.23 (m, 2H, CH2β Val);
HRMS (ESI-TOF):
M
+
Na+
+
H+ found 1921.0233 2.47–2.50 (t, 2H, COCH2 (PEG)); 2.83–2.88 & 3.05–3.14 (m, 2H,
CH2β Phe); 2.85–2.92 (s, 3H, NMe); 2.96–3.02 & 3.14–3.24 (m,
C102H141N14O21Na requires 1921.0293.
Compound SM122–PEG-9 DIMER. Utilizing 180 mg 2H, CH2β D-Phe); 3.04–3.20 (m, 2H, CH2ε Lys (PEG)); 3.15–3.30
(0.23 mmol, 1.0 equiv.) of Phe-N-Me-D-Phe-Val-Lys-Lys (Cbz), (m, 2H, CH2ε Lys (Cbz)); 3.36–3.42 (s, 3H, OCH3 (PEG));
161.0 μL (113 μmol, 0.5 equiv.) BS (PEG)9, and 0.31 mL 3.54–3.60 (m, 2H, CH2 (PEG)); 3.61–3.78 (m, 10H, CH2 (PEG));
(8.0 equiv.) of DIPEA dissolved in 20.0 mL CH2Cl2, under N2. 3.67–3.75 (m, 2H, CH2 (PEG)); 3.75–3.83 (m, 1H, CHα Lys
The reaction was monitored via LCMS and generally complete (PEG)); 3.93–4.01 (m, 1H, CHα Lys (Cbz)); 4.44–4.52 (m, 1H,
in 1–2 hours. Upon completion, the reaction was worked up by CHα Val); 4.72–4.82 (m, 1H, CHα Phe); 5.07–5.13 (s, 2H, CH2
washing with aqueous HCl (pH 1) and saturated sodium bicar- Cbz); 5.16–5.23 (m, 1H, CHα D-Phe); 5.49–5.57 (m, 1H, NH);
bonate. After back extraction of aqueous layers with large 6.80–6.87 (m, 1H, NH); 6.87–6.95 (m, 1H, NH); 7.01–7.37 (m,
quantities of CH2Cl2, the organic layers were combined, dried, 15H, Ph); 7.65–7.72 (m, 1H, NH). δC NMR (75 MHz, CDCl3):
filtered and concentrated. The crude material was purified δ 18.0, 19.6, 22.7, 23.4, 28.9, 29.2, 29.6, 30.3, 30.5, 30.7, 32.9,
using column chromatography (silica gel, EtOAc–MeOH) to 36.8, 37.8, 38.2, 40.8, 54.0, 54.9, 55.3, 57.1, 57.3, 59.0, 66.5,
yield 123 mg (52% yield) of the compound. Rf: 0.56 (EtOAc– 67.3, 70.5, 71.8, 126.9, 127.9, 128.0, 128.5, 128.6, 128.7, 128.8,
MeOH 6 : 4); δH NMR (300 MHz, CDCl3): δ 0.80–0.85 (d, J = 129.4, 136.1, 136.5, 136.8, 156.7, 169.7, 171.7, 172.1, 172.7;
6.71 Hz, 6H, CHCH(CH3)2); 0.88–0.94 (d, J = 6.71 Hz, 6H, LC/MS (ESI): m/z called for C54H77N7O12 (M + H+) = 1015.56,
CHCH(CH3)2); 0.95–1.08 (m, 4H, CH2γ Lys (PEG)); 1.27–1.30 found 1016.00.
(m, 4H, CH2δ Lys (Cbz)); 1.34–1.46 (m, 4H, CH2γ Lys (Cbz));
1.48–1.58 (m, 4H, CH2δ Lys (PEG)); 1.69–1.82 (m, 4H, CH2β Lys C54H77N7O12Na requires 1038.5528.
(PEG)); 1.92–2.03 (m, 4H, CH2β Lys (Cbz)); 2.14–2.23 (m, 4H, Compound SM122–PEG-8. Utilizing 124 mg (0.16 mmol, 1.0
HRMS (ESI-TOF):
M
+
Na+, found 1038.5504
CH2β Val); 2.47–2.50 (t, 4H, COCH2 (PEG)); 2.83–2.88 & equiv.) of Phe-N-Me-D-Phe-Val-Lys-Lys (Cbz), 68.0 μL (155 μmol,
3.05–3.14 (m, 4H, CH2β Phe); 2.85–2.92 (s, 6H, NMe); 1.0 equiv.) MS (PEG)8, and 0.22 mL (8.0 equiv.) of DIPEA dis-
2.96–3.02 & 3.14–3.24 (m, 4H, CH2β D-Phe); 3.04–3.20 (m, 4H, solved in 20.0 mL CH2Cl2, under N2. The reaction was moni-
CH2ε Lys (PEG)); 3.15–3.30 (m, 4H, CH2ε Lys (Cbz)); 3.54–3.60 tored via LCMS and generally complete in 1–2 hours. Upon
(m, 4H, CH2 (PEG)); 3.61–3.78 (m, 28H, CH2 (PEG)); 3.75–3.83 completion, the reaction was worked up by washing with
(m, 2H, CHα Lys (PEG)); 3.93–4.01 (m, 2H, CHα Lys (Cbz)); aqueous HCl (pH 1) and saturated sodium bicarbonate. After
4.44–4.52 (m, 2H, CHα Val); 4.72–4.82 (m, 2H, CHα Phe); back extraction of aqueous layers with large quantities of
5.07–5.13 (s, 4H, CH2 Cbz); 5.16–5.23 (m, 2H, CHα D-Phe); CH2Cl2, the organic layers were combined, dried, filtered and
5.49–5.57 (m, 2H, NH); 6.80–6.87 (m, 2H, NH); 6.87–6.95 (m, concentrated. The crude material was purified using column
This journal is © The Royal Society of Chemistry 2014
Org. Biomol. Chem., 2014, 12, 765–773 | 771

View Article Online
Paper
Organic & Biomolecular Chemistry
chromatography (silica gel, EtOAc–MeOH) to yield 97 mg The luciferase activity in the refolding reaction with DMSO
(52.5% yield) of the compound. Rf: 0.26 (EtOAc–MeOH 19 : 1); (control) at 120 minutes was considered as 100% refolding.
δH NMR (300 MHz, CDCl3): δ 0.80–0.85 (d, J = 6.71 Hz, 3H, Reported values were averaged from three independent
CHCH(CH3)2); 0.88–0.94 (d, J = 6.71 Hz, 3H, CHCH(CH3)2); experiments. Error bars indicate the standard deviation.
0.95–1.08 (m, 2H, CH2γ Lys (PEG)); 1.27–1.30 (m, 2H, CH2δ Lys Statistical analysis was performed using GraphPad Prism
(Cbz)); 1.34–1.46 (m, 2H, CH2γ Lys (Cbz)); 1.48–1.58 (m, 2H, software.
CH2δ Lys (PEG)); 1.69–1.82 (m, 2H, CH2β Lys (PEG)); 1.92–2.03
Protein binding assay. The inhibitory effects of SM122,
(m, 2H, CH2β Lys (Cbz)); 2.14–2.23 (m, 2H, CH2β Val); SM122–PEG-8 and SM122–PEG-9 DIMER on the binding
2.47–2.50 (t, 2H, COCH2 (PEG)); 2.83–2.88 & 3.05–3.14 (m, 2H, between hsp90 and its co-chaperone (HOP, StressMarq Bio-
CH2β Phe); 2.85–2.92 (s, 3H, NMe); 2.96–3.02 & 3.14–3.24 (m, sciences; FKBP52, Abcam) were determined using protein
2H, CH2β D-Phe); 3.04–3.20 (m, 2H, CH2ε Lys (PEG)); 3.15–3.30 binding assay. 200 nM (final concentration) of human native
(m, 2H, CH2ε Lys (Cbz)); 3.36–3.42 (s, 3H, OCH3 (PEG)); protein hsp90 (Life Technologies) was incubated with 10 μM of
3.54–3.60 (m, 2H, CH2 (PEG)); 3.61–3.78 (m, 26H, CH2 (PEG)); compound or DMSO (Sigma-Aldrich, final concentration of
3.67–3.75 (m, 2H, CH2 (PEG)); 3.75–3.83 (m, 1H, CHα Lys DMSO was 1%) in binding buffer (20 mM tris-HCl, 150 mM
(PEG)); 3.93–4.01 (m, 1H, CHα Lys (Cbz)); 4.44–4.52 (m, 1H, NaCl, 1% Triton-X-100, pH 7.4) for 1 hour at room temperature
CHα Val); 4.72–4.82 (m, 1H, CHα Phe); 5.07–5.13 (s, 2H, CH2 (RT). After incubation 100 nM (final concentration) of Histi-
Cbz); 5.16–5.23 (m, 1H, CHα D-Phe); 5.49–5.57 (m, 1H, NH); dine tagged recombinant co-chaperone was added and incu-
6.80–6.87 (m, 1H, NH); 6.87–6.95 (m, 1H, NH); 7.01–7.37 (m, bated for another 1 hour at RT. The recombinant protein was
15H, Ph); 7.65–7.72 (m, 1H, NH). δC NMR (75 MHz, CDCl3): fished out with Talon-Metal Affinity Resin (Clontech Labora-
δ 18.0, 19.6, 22.7, 23.4, 28.9, 29.2, 29.6, 30.3, 30.5, 30.7, 32.9, tories) followed by three washes of the beads in binding buffer
36.8, 37.8, 38.2, 40.8, 54.0, 54.9, 55.3, 57.1, 57.3, 59.0, 66.5, and finally boiling the beads with 5 × Laemmli sample buffer.
67.3, 70.5, 71.8, 126.9, 127.9, 128.0, 128.5, 128.6, 128.7, 128.8, Samples were analyzed using 4–20% Tri-Glycine gels (Life
129.4, 136.5, 136.8, 156.7, 169.7, 171.7, 172.1, 172.7; LC/MS Technologies), followed by standard Western blot to detect
(ESI): m/z called for C62H93N7O16 (M + H+) = 1191.67, found hsp90 and its co-chaperone. The respective ratio of hsp90 to
1192.00; HRMS (ESI-TOF):
C62H93N7O16Na requires 1214.6577.
M
+
Na+, found 1214.6552 co-chaperone was analyzed via Image J and transformed to a
percent of hsp90 bound to co-chaperone. Reported values were
averaged from three independent experiments. Error bars indi-
cate the standard deviation. Statistical analysis was performed
using GraphPad Prism software.
Biological methods
Luciferase refolding assay. Firefly luciferase (12.5 mg mL−1
;
Novus Biologicals) was diluted to a concentration of 2 mg
mL−1 in stability buffer (25 mM Tricine, pH 7.8, 10 mM MgCl2,
1 mM dithiothreitol, 0.1 mM EDTA, 10% (v/v) glycerol, and
10 mg mL−1 bovine serum albumin), and was heat denatured
at 41 °C for 30 minutes. The denatured protein was further
diluted (1 : 20, v/v) in stability buffer to form 0.1 mg mL−1
stock solution and placed on ice before refolding. 0.5 μL of
compound or DMSO (Sigma-Aldrich) as a control was incu-
bated with 48.5 μL of 50% diluted rabbit reticulocyte lysate
(RRL; Promega) in Mili-Q water at 30 °C for 5 hours. Refolding
was initiated by adding 1.0 μL of the denatured luciferase
stock into the RRL refolding system, treated either with com-
pounds or DMSO in advance. Reactions were performed at
30 °C. At the indicated time points (15, 30, 45, 60, 75, 90, 105
and 120 minutes), 5 μL of each reaction mixture was removed
and added to 45 μL of Bright-Glo™ luciferase assay buffer
(Promega) mixed with Bright-Glo™ luciferase assay substrate
(Promega), which was preloaded in a white, flat-bottomed, 96
well plate (Greiner Bio-One). After incubating for 2 minutes at
room temperature, the luminescence was measured using a
luminometer (Berthold Orion Microplate Luminometer). Luci-
ferase activity in refolding reactions at each time point was cal-
culated by the formula:
Acknowledgements
We thank the University of New South Wales for a scholarship
to H.W. and a tuition fee waiver to Y.W. We thank NHMRC
(1043561) for support of Y.W.
Notes and references
1 L. Neckers, Trends Mol. Med., 2002, 8, S55–S61.
2 G. Chiosis, H. Huezo, N. Rosen, E. Mimnaugh, L. Whitesell
and L. Neckers, Mol. Cancer Ther., 2003, 2, 123–129.
3 M. G. Hollingshead, M. Alley, A. M. Burger, S. Borgel,
C. Pacula-Cox, H.-H. Fiebig and E. A. Sausville, Cancer
Chemother. Pharmacol., 2005, 56, 115–125.
4 M. Senju, N. Sueoka, A. Sato, K. Iwanaga, Y. Sakao,
S. Tomimitsu, M. Tominaga, K. Irie, S. Hayashi and
E. Sueoka, J. Cancer Res. Clin. Oncol., 2006, 132, 150–158.
5 Y.-S. Chang, L.-C. Lee, F.-C. Sun, C.-C. Chao, H.-W. Fu and
Y.-K. Lai, J. Cell. Biochem., 2006, 97, 156–165.
6 D. Matei, M. Satpathy, L. Cao, Y.-C. Lai, H. Nakshatri and
D. B. Donner, J. Biol. Chem., 2007, 282, 445–453.
7 G. Crevel, H. Bates, H. Huikeshoven and S. Cotterill, J. Cell
Sci., 2001, 114, 2015–2025.
Luciferase activity ð%Þ ¼ ðLIsample=LIDMSO at 120 minÞ ꢀ 100
where LI indicates the luminescence intensity in each reaction.
772 | Org. Biomol. Chem., 2014, 12, 765–773
This journal is © The Royal Society of Chemistry 2014

View Article Online
Organic & Biomolecular Chemistry
Paper
8 J. E. Bohonowych, U. Gopal and J. S. Isaacs, J. Oncol., 2010, 27 P. S. Pan, R. C. Vasko, S. A. Lapera, V. A. Johnson,
2010, 412985.
R. P. Sellers, C.-C. Lin, C.-M. Pan, M. R. Davis, V. C. Ardi and
S. R. McAlpine, Bioorg. Med. Chem., 2009, 17, 5806–5825.
28 R. P. Sellers, L. D. Alexander, V. A. Johnson, C.-C. Lin,
J. Savage, R. Corral, J. Moss, T. S. Slugocki, E. K. Singh,
M. R. Davis, S. Ravula, J. E. Spicer, J. L. Oelrich,
A. Thornquist, C.-M. Pan and S. R. McAlpine, Bioorg. Med.
Chem., 2010, 18, 6822–6856.
9 S. Takayama, J. C. Reed and S. Homma, Oncogene, 2003, 22,
9041–9047.
10 L. Neckers, J. Biosci., 2007, 32, 517–530.
11 C. Jolly and R. I. Morimoto, J. Natl. Cancer Inst., 2000, 92,
1564–1572.
12 T. U. Mayer, T. M. Kapoor, S. J. Haggarty, R. W. King,
S. L. Schreiber and T. J. Mitchison, Science, 1999, 286, 971– 29 J. B. Kunicki, M. N. Petersen, L. D. Alexander, V. C. Ardi,
974.
J. R. McConnell and S. R. McAlpine, Bioorg. Med. Chem.
Lett., 2011, 21, 4716–4719.
30 M. R. Davis, E. K. Singh, H. Wahyudi, L. D. Alexander,
J. Kunicki, L. A. Nazarova, K. A. Fairweather, A. M. Giltrap,
K. A. Jolliffe and S. R. McAlpine, Tetrahedron, 2012, 68,
1029–1051.
31 L. D. Alexander, J. R. Partridge, D. A. Agard and
S. R. McAlpine, Bioorg. Med. Chem. Lett., 2011, 21, 7068–
7071.
32 D. M. Ramsey, J. R. McConnell, L. D. Alexander,
K. W. Tanaka, C. M. Vera and S. R. Mcalpine, Bioorg. Med.
Chem. Lett., 2012, 22, 3287–3290.
13 M. Maruya, M. Sameshima, T. Nemoto and I. Yahara,
J. Mol. Biol., 1999, 285, 903–907.
14 T. Nemoto, Y. Ohara-Nemoto, M. Ota, T. Takagi and
K. Yokoyama, Eur. J. Biochem., 1995, 233, 1–8.
15 F. Koga, K. Kihara and L. Neckers, Anticancer Res., 2009, 29,
797–807.
16 M. V. Powers, K. Jones, C. Barillari, I. Westwood, R. L. M.
v. Montfort and P. Workman, Cell Cycle, 2010, 9, 1542–
1550.
17 J. D. Eskew, T. Sadikot, P. Morales, A. Duren, I. Dunwiddie,
M. Swink, X. Zhang, S. Hembruff, A. Donnelly,
R. A. Rajewski, B. S. Blagg, J. R. Manjarrez, R. L. Matts, 33 V. C. Ardi, L. D. Alexander, V. A. Johnson and
J. M. Holzbeierlein and G. A. Vielhauer, BMC Cancer, 2011,
11, 468.
18 R. C. Vasko, R. A. Rodriguez, C. N. Cunningham, V. C. Ardi,
D. A. Agard and S. R. McAlpine, ACS Med. Chem. Lett., 2010,
1, 4–8.
19 V. C. Ardi, L. D. Alexander, V. A. Johnson and
S. R. McAlpine, ACS Chem. Biol., 2011, 6, 1357–1366.
20 J. M. McConnell, L. D. Alexander and S. R. McAlpine, Bioorg.
Med. Chem. Lett., DOI: 10.1016/j.bmcl.2013.11.059.
21 C. L. Carroll, J. V. C. Johnston, A. Kekec, J. D. Brown,
E. Parry, J. Cajica, I. Medina, K. M. Cook, R. Corral,
S. R. McAlpine, ACS Chem. Biol., 2011, 6, 1357–1366.
34 L. D. Alexander, J. R. Partridge, D. A. Agard and
S. R. McAlpine, Bioorg. Med. Chem. Lett., 2011, 21, 7068–
7071.
35 S. T. Diver and S. L. Schreiber, J. Am. Chem. Soc., 1997, 119,
5106–5109.
36 J. D. Klemm, S. L. Schreiber and G. R. Crabtree, Annu. Rev.
Immunol., 1998, 16, 569–592.
37 B. R. Kusuma, L. B. Peterson, H. Zhao, G. Vielhauer,
J. Holzbeierlein and B. S. Blagg, J. Med. Chem., 2011, 54,
6234–6253.
P.-S. Pan and S. R. McAlpine, Org. Lett., 2005, 7, 3481–3484. 38 M. L. Miller, E. E. Roller, R. Y. Zhao, B. A. Leece, O. Ab,
22 T. J. Styers, R. A. Rodriguez, P.-S. Pan and S. R. McAlpine,
E. Baloglu, V. S. Goldmacher and R. V. J. Chari, J. Med.
Chem., 2004, 47, 4802–4805.
Tetrahedron Lett., 2006, 47, 515–517.
23 K. Otrubova, K. L. McGuire and S. R. McAlpine, J. Med. 39 M. M. U. Ali, S. M. Roe, C. K. Vaughan, P. Meyer,
Chem., 2007, 50, 1999–2002.
24 P. S. Pan, K. McGuire and S. R. McAlpine, Bioorg. Med.
Chem. Lett., 2007, 17, 5072–5077.
B. Panaretou, P. W. Piper, C. Prodromou and L. H. Pearl,
Nature, 2006, 440, 1013–1017.
40 D. R. Southworth and D. A. Agard, Mol. Cell, 2011, 42, 771–
781.
25 R. A. Rodriguez, P.-S. Pan, C.-M. Pan, S. Ravula,
S. A. Lapera, E. K. Singh, T. J. Styers, J. D. Brown, J. Cajica, 41 S. K. Wandinger, K. Richter and J. Buchner, J. Biol. Chem.,
E. Parry, K. Otrubova and S. R. McAlpine, J. Org. Chem.,
2007, 72, 1980–2002.
26 K. Otrubova, G. H. Lushington, D. Vander Velde,
2008, 283, 18473–18477.
42 C.-C. Lee, T.-W. Lin, T.-P. Ko and A. H.-J. Wang, PLoS One,
2011, 6, e19961.
K. L. McGuire and S. R. McAlpine, J. Med. Chem., 2008, 51, 43 Z. L. Hildenbrand, S. K. Molugu, N. Herrera, C. Ramirez,
530–544.
C. Xiao and R. A. Bernal, Oncotarget, 2011, 2, 43–58.
This journal is © The Royal Society of Chemistry 2014
Org. Biomol. Chem., 2014, 12, 765–773 | 773