Development of antiproliferative phenylmaleimides that activate the unfolded protein response

Article abstract of DOI:10.1016/j.bmc.2010.04.057

The current paper presents the synthesis and evaluation of a series of maleimides that were designed to inhibit the Cdc25 phosphatase by alkylation of catalytically essential cysteine residues. Although in HepB3 cell culture assays the analogues did exhib

Full text of DOI:10.1016/j.bmc.2010.04.057

Bioorganic & Medicinal Chemistry 18 (2010) 4535–4541
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Development of antiproliferative phenylmaleimides that activate
the unfolded protein response
Ulrike Muus a, Curtis Hose b, Wei Yao c, Teresa Kosakowska-Cholody d, David Farnsworth a, Marzena Dyba e,
c,
George T. Lountos f, David S. Waugh f, Anne Monks b, Terrence R. Burke Jr. a, , Christopher J. Michejda
*
a Chemical Biology Laboratory, Molecular Discovery Program, NCI-Frederick, Frederick, MD 21702, USA
b Screening Technologies Branch, SAIC-Frederick, Inc., Frederick, MD 21702, USA
c Structural Biophysics Laboratory, Molecular Discovery Program, NCI-Frederick, Frederick, MD 21702, USA
d Laboratory of Cell and Developmental Signaling, NCI-Frederick, Frederick, MD 21702, USA
e Structural Biophysics Laboratory, SAIC-Frederick, Inc., NCI-Frederick, Frederick, MD 21702, USA
f Macromolecular Crystallography Laboratory, NCI-Frederick, Frederick, MD 21702, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
The current paper presents the synthesis and evaluation of a series of maleimides that were designed to
inhibit the Cdc25 phosphatase by alkylation of catalytically essential cysteine residues. Although in
HepB3 cell culture assays the analogues did exhibit antiproliferative IC50 values ranging from sub-micro-
Received 4 March 2010
Revised 19 April 2010
Accepted 20 April 2010
Available online 24 April 2010
molar to greater than 100 lM, inhibition of Cdc25 through cysteine alkylation could not be demon-
strated. It was also found that analysis using fluorescence activated cell sorting (FACS) following
treatment with the most potent analogue (1t) did not provide data consistent with inhibition at one spe-
cific point in the cell cycle, as would be expected if Cdc25A were inhibited. Further studies with a subset
of analogues resulted in a correlation of antiproliferative potencies with activation of the unfolded pro-
tein response (UPR). The UPR is a regulatory pathway that temporarily suspends protein production
when misfolding of proteins occurs within the endoplastic reticulum (ER). In addition, ER chaperones that
promote proper refolding become up-regulated. If cellular damage cannot be resolved by these mecha-
nisms, then the UPR can initiate apoptosis. The current study indicates that these maleimide analogues
lead to UPR activation, which is predictive of the selective antiproliferative activity of the series.
Published by Elsevier Ltd.
This work is dedicated to the memory of Dr.
Christopher J. Michejda
Keywords:
Cdc25
Antiproliferative
Unfolded protein response
Protein alkylation
1. Introduction
exhibit differential antiproliferative potencies in HEP3B cell
growth assays. When we were unable to demonstrate the origi-
Phenylmaleimides of type 1 (Fig. 1) have recently been reported
to exhibit growth inhibition against liver cancer cell lines in vitro
and antitumor activity in vivo in Fischer 344 rats carrying an ortho-
topic model of hepatocellular carcinoma.1 The design of these com-
pounds was originally based on benzoquinone Cdc25 phosphatase
inhibitors, such as 2-(2-mercaptoethanol)-3-methyl-1,4-naphtho-
quinone (3) and 2,3-bis-[2-hydroxyethylsulfanyl]-1,4-naphthoqui-
none (4) (Fig. 1).2,3 Phosphatase inhibition by naphthoquinones has
been shown to occur by alkylation of the enzyme’s catalytic cys-
teine residue.4 The phenylmaleimides were designed to alkylate
the phosphatase catalytic cysteine via a 1,4-Michael-type mecha-
nism, but to be distinct from the naphthoquinones in not promot-
ing oxidative recycling that can occur with the latter.5 As presented
in our current report, we prepared a series of structurally diverse
N-aryl and N-alkyl maleimides (1 and 2, respectively, Fig. 1) that
nally intended Michael acceptor-based inhibitory effects on
Cdc25 phosphatases, we examined potential causes of antiprolifer-
ation not directly related to inhibition of Cdc25 phosphatases. We
were able to show that the phenylmaleimides differentially acti-
vate unfolded protein response (UPR) pathways and that these lat-
ter actions of the phenylmaleimides could contribute to their
overall antiproliferative profiles.
O
O
S
S
R
OH
OH
OH
R N
S
O
O
1 R = Aryl
3 R = Me
2 R = Aryalkyl
4 R = SCH2CH2OH
* Corresponding author. Tel.: +1 301 846 5906; fax: +1 301 846 6033.
E-mail address: tburke@helix.nih.gov (T.R. Burke).
Deceased.
Figure 1. Structures of N-substituted maleimides (1 and 2) and benzoquinone
Cdc25 phosphatase inhibitors (3 and 4).
0968-0896/$ - see front matter Published by Elsevier Ltd.
doi:10.1016/j.bmc.2010.04.057

4536
U. Muus et al. / Bioorg. Med. Chem. 18 (2010) 4535–4541
O
O
O
O
HS
Br
S
S
Br
Br
R-NH2
OH
OH
OH
HO
HO
R N
R N
HOAc, reflux
imidazole, THF
Br
O
O
5
6a - 6z R = Aryl
1a - 1z R = Aryl
7a - 7d R = Arylalkyl
2a - 2d R = Arylalkyl
Scheme 1. Synthesis of target maleimides with ‘R’ as defined in Tables 1 and 2.
Table 1
2. Results and discussion
2.1. Synthesis
Growth inhibition of phenylmaleimidesa
No.
Rb
IC50
70
(lM)
The synthesis of N-substituted maleimides 1a–1z and 2a–2d
(Scheme 1) began by reacting 3,4-dibromomaleic acid (5) with
the appropriate amines in refluxing acetic acid to give the corre-
sponding 3,4-dibromomaleiides (6a–6z and 7a–7d, respectively)
in yields of 33–93% (Scheme 1). Treatment of the dibromo-inter-
mediates with 2-mercaptoethanol in the presence of imidazole in
tetrahydrofuran followed by silica gel flash chromatographic puri-
fication gave the desired final products (Scheme 1). In an alternate
approach, N-substituted maleimides were brominated to yield a
mixture of 3-bromo and 3,4-dibromo products.6 Reaction with
mercaptoethanol as indicated above, gave the corresponding
mono- and di-substituted adducts, which were subjected to chro-
matographic purification to provide the final products 1a–1z and
2a–2d.
1a
I
1b
9.0
F
1c
1d
1e
9.0
7.0
9.0
O
O
2.2. Evaluation of growth inhibition
1f
9.0
8.0
The antiproliferative effects (IC50 values defined as the concen-
tration of inhibitor causing a 50% reduction of cell number relative
to untreated controls) of 1a–1z and 2a–2d were determined using
a HEP3B cell growth inhibition assay.1 The naphthoquinone-based
Cdc25 phosphatase inhibitors 3 and 4 were also included for com-
parison, since they had served as models in the original design of
the maleimide series of compounds. With few exceptions (i.e.,
1q, 1r and 1w) the N-aryl-substituted maleimides (1a–1z, Table
1) and the N-arylalkyl-substituted maleimides (2a–2d, Table 2)
were found to inhibit cell growth in a dose-dependent manner.
The growth inhibition of 1b–1p, which contained a variety of sub-
stituents at the phenyl 2-, 3- and 4-positions, were all highly uni-
O
1g
Cl
1h
1i
N
O
9.0
6.0
6.0
5.0
9.0
1j
O
1k
1l
O
form (IC50 values of between 4 and 10
order of magnitude greater than the unsubstituted parent (1a,
IC50 = 70 M). This data indicated that most of the aryl substituents
lM) and approximately one
NC
l
F
F
enhanced growth inhibition to similar extents, irrespective of ring
position. In contrast, analogues bearing 4-hydroxy or 4-acetamido
groups (1q and 1r, respectively) showed markedly reduced growth
F
1m
4.0
O
inhibition (IC50 >100 lM). Biphenyl compounds 1s–1z were then
prepared to further investigate substituent effects at the 4-posi-
tion. The most potent analogue was found to be the biphenyl-
4yl-containing compound (1t, IC50 = 0.7 lM), which exhibited
1n
1o
Cl
7.0
9.0
F
100-fold greater antiproliferative potency than the unsubstituted
parent 1a (Table 1). The addition of electron donating or electron
withdrawing groups (1s and 1v, respectively) decreased antiprolif-
erative potencies by from 4 to 6-fold. Introduction of o-methyl
groups onto the biphenyl rings of 1t resulted in greater than a
10-fold loss of potency (1u).
Although there appeared to be wide tolerance for structural
diversity in the biphenyl series (1s–1w), the 4-phenoxyphenyl-
containing analogue (1w) was distinct in showing a significant loss
O
1p
1q
10
HO
>100
>100
3.0
O
1r
1s
HN
O
of toxicity (IC50 >100 lM). This potentially indicated that geomet-
ric spacing of the two phenyl rings could be important for growth
inhibition. That geometry could influence potency was also

U. Muus et al. / Bioorg. Med. Chem. 18 (2010) 4535–4541
4537
techniques were employed in efforts to obtain a structure of
Cdc25A with bound inhibitor. In one protocol a molar excess of
1t was incubated with the catalytic domain of Cdc25A and the
solution was then set up for crystallization. Crystals appeared
within 3–5 days and X-ray diffraction data were collected for
numerous crystals up to a resolution of 2.0 Å. However, the elec-
tron density maps did not show any evidence of bound inhibitor.
Rather, in many crystals electron density maps clearly revealed a
disulfide linkage between the catalytic Cys430 residue and a prox-
imal Cys384 residue.
Table 1 (continued)
No.
Rb
IC50
0.7
(lM)
1t
1u
10
4
1v
O2N
Since co-crystallization studies failed to yield a structure with
bound inhibitor, two strategies employing soaking methodologies
were examined: In the first approach, crystals of the apo-form of
Cdc25A were soaked with 1 mM 1t in the presence of 5% v/v DMSO
to enhance the solubility of the compound in the crystallization
solution. Soaking times varied from 24 h to several days. Although
the resulting crystals diffracted well, the electron densities did not
show signs of bound inhibitor. In most cases, the active sites indi-
cated the presence of the disulfide-linked cysteines. Based on the
hypothesis that these disulfide linkages may prevent the covalent
attachment of 1t to the catalytic cysteine residue, crystals of
Cdc25A were treated with the reducing agent tris-(2-carboxy-
ethyl)phosphine (TCEP) immediately prior to soaking with 1t.
Although crystals from this experiment diffracted at up to 2.0 Å
resolution, no inhibitor could be detected bound to the active site.
MALDI-TOF mass spectral studies were also undertaken to iden-
tify covalent adducts to Cdc25A protein following incubation with
1t. A bromo-containing variant 1t was also examined, since ad-
ducts could be easily identified by their 1:1 isotopic bromine dou-
blets. Inhibitors were reacted with both wild-type Cdc25A and the
C384A mutant. Incubation mixtures were subjected to tryptic di-
gest prior to analysis by MALDI-TOF techniques. These experi-
ments failed to detect peaks corresponding to covalent
attachment of 1t to the Cdc25A protein, indicating that the cata-
lytic cysteine was not alkylated as expected.
1w
O
>100
3.5
1x
1y
4.0
1.8
1z
a
Assays were run using HEP3B cells as described in the experimental procedures.
R group as indicated in Scheme 1.
b
Table 2
Growth inhibition of phenylalkylmaleimidesa
No.
Rb
IC50
5.0
(lM)
2a
Cl
In further work, a Cdc25A assay was set up using a 3-O-methyl
fluoresceine phosphate (OMFP) protocol. Although distinct enzyme
inhibition could be shown for known naphthoquinone-based
inhibitors (including 4), which are hypothesized to function
through a redox-mechanism, inhibition of Cdc25A by 1t could
not be shown. Furthermore, when aliquots of enzyme samples
used in the mass spectrometry studies described above were as-
sayed for inhibition by 1t using a para-nitrophenyl phosphate as-
say, the wild-type enzyme and the two mutants did not show
significant inhibition by 1t. Finally, although Cdc25A is involved
in the control of the G1/S transition,7 fluorescence activated cell
sorting (FACS) analysis following treatment with 1t showed that
inhibition at one specific point in the cell cycle did not appear to
occur as would be expected with inhibition of Cdc25A.8,9
2b
2c
2d
2.5
2.0
1.0
Cl
a
Assays were run using HEP3B cells as described in the experimental procedures.
R group as indicated in Scheme 1.
b
supported by the fact that the growth inhibition of 1t, which had a
conformational minimum at a biphenyl torsional angle of 44°, is
10-fold lower than the bis-methyl-substituted biphenyl compound
1u, where a conformational minimum exists when the phenyl
rings are almost perpendicular to each other. In compounds 2a–
2d alkyl spacers were introduced between the maleimide nitrogen
and the phenyl rings (Table 2). The growth inhibitory potencies of
these analogues were similar to the 1- and 2-naphthyl systems (1y
and 1x) as well as the 2-anthracenyl system (1z) (Table 1). This
indicated that conjugation through the aromatic system was not
a critical determinant of antiproliferative potency.
2.3. Activation of the unfolded protein response (UPR)
The endoplasmic reticulum (ER) is responsible for the synthesis
of secretory, membrane and organelle-targeted proteins.10 Within
the ER nascent polypeptide chains undergo necessary folding and
post-translational modification, which includes the attachment of
oligosaccharides and the formation of disulfide chains.11 Certain
cellular stresses, such as lowered pH values, reduced glucose lev-
els, irradiation, the presence of reactive oxygen species (ROS) or
genotoxic drugs, can adversely affect the ER’s capacity to properly
fold proteins, leading to activation of the UPR.12 The UPR partici-
pates in cytoprotection through the temporary suspension of pro-
tein production, while at the same time upregulating ER
chaperones that promote proper refolding. Under certain circum-
stances the UPR can facilitate induction of apoptosis.13
The compounds in Tables 1 and 2 were designed to be Cdc25
phosphatase inhibitors that could theoretically function by 1,4-Mi-
chael-type alkylation of the catalytic cysteine thiolate anion. With
an IC50 value of 0.7 lM, the biphenyl-4-yl-containing maleimide 1t
was the most potent antiproliferative agent among the analogues
examined. Therefore, crystallographic studies undertaken to obtain
co-crystal structures of 1t bound to Cdc25A protein. Cdc25A crys-
tallizes readily and co-crystallization as well as crystal soaking

4538
U. Muus et al. / Bioorg. Med. Chem. 18 (2010) 4535–4541
Although inhibition of Ccd25 phosphatases did not appear to be
ER stress.14 The capability of phenylmaleimides to react as Michael
acceptors with irreversible expulsion of one or both thioethanol
groups was shown by reacting 1t with N-acetylcysteine. Formation
of the corresponding mono- and bis-cysteine adducts was shown
by HPLC-MS.
In order to investigate whether the extent of UPR activation is
related to growth inhibition by N-substituted maleimides, seven
compounds (1a, 1r, 1t, 1w, 1u, 1v and 2c) from Tables 1 and 2 were
selected for further examination. These analogues were re-evalu-
ated for their antiproliferative potencies in HEP3B cells and in par-
the cause of growth inhibition of the parent phenylmaleimide 1t,
transcriptional profiling of NEP3B cell in response to this com-
pound identified a rapid increase in genes involved in the UPR,
while leaving unchanged other stress inducible genes such as
CDKN1A, DCD25A, CHEK1, BRCA1, MDM2 and GADD45B. Produc-
tion of miss-folded or unfolded proteins causing endoplasmic retic-
ulum (ER) stress and leading to UPR activation could potentially
arise through alkylation of sulfhydryl groups on nascent polypep-
tide chains via 1,4-Michael-type addition onto the maleimide ana-
logues. Such a mechanism is consistent with a report that the
toxicity of arylating quinones is directly coupled to induction of
allel examined at multiple concentrations (1, 10 and 50 lM) for
their effects on the expression of five UPR-related genes (DDIT3,
TRIB3, ATF4, GADD34, HSPA1A), which had been shown to be up-
regulated following 1 h incubation with 10 lM of 1t. These genes
have been most closely allied with the UPR through protein kinase
RNA (PKR)-like ER kinase (PERK) in response to ER stress.15 An-
other critical determinant of the UPR response, is the PERK-medi-
ated phosphorylation at Ser51 of the ‘Eukaryotic Translation
Initiation Factor 2’ on its alpha subunit (termed ‘eIF2
The phosphorylation of eIF2 was monitored concurrently with
a
’).11,16
a
expression of the five UPR-related genes. Figure 2 shows dose–re-
sponse curves that demonstrate the differential sensitivity of
HEP3B cells to the seven selected analogues. These sensitivities
ranged from the most potent analogue (1t) to the analog with no
measurable activity in the assay (1r). Expression data after treat-
ment with 2, 10 and 50
cated that 10 M incubation of each agent for 1 h allowed for a
direct comparison of the magnitude of gene modulation. There
were limited changes at 2 M concentrations, while 50 M con-
lM, of these analogues for 1 and 6 h, indi-
l
l
l
centrations were extremely toxic to the cells, making interpreta-
tion of comparative expression data difficult at these two
concentrations.
Figure 3 identifies gene expression changes ( SD, n = 3) after
10 lM treatment with each analogue (1 h). ATF4, a member of
the b-ZIP family of transcription factors,17 was induced approxi-
mately twofold in response to all the analogues to which the cells
responded and it is one of the few genes that remains actively
translated after the UPR response attenuates protein synthesis in
response to ER stress.18,19 ATF-4 is responsible for transcriptional
activation of a variety of UPR-responsive genes, including GADD34
(PP1R15A), which acts as a negative feedback loop to regulate
Figure 2. Sensitivity of HEP3B cell to indicated maleimide analogues. Cells were
exposed to a range of analogue concentrations for 48 h, then survival was measured
by the vital dye, alamar blue, and dose–response curves were calculated by (OD of
treated/OD of control  100).
Figure 3. Induction of UPR-related gene expression after 1 h treatment with indicated analogs (10
in treated cells compared to control cells (n = 3 SD). Represents situations where drug-induced gene expression is significantly different to control expression (p <0.05).
l
M). Genes were measured by real time PCR and expressed as fold change
*

U. Muus et al. / Bioorg. Med. Chem. 18 (2010) 4535–4541
4539
eIF1
a
phosphorylation,20,21 and provides an adaptive response to
at Ser51 in HEP3B cells treated with the most potent analogs as
compared to treatment with the less potent analogues. The extent
of eIF2a phosphorylation correlated significantly with HEP3B sen-
stress allowing protein synthesis to resume.20 However, GADD34
over-expression has been demonstrated to elicit growth arrest
and apoptosis22,23 in this study, the magnitude of induction of this
gene was significantly correlated (R = À0.824, p = 0.022) with the
potency of growth inhibition.
sitivity to the analogues (Fig. 4B).
The results illustrate an essential paradox of stress response
pathways, including the UPR, where response is designed to facil-
itate both stress adaptation as well as induction of apoptosis,
depending upon the nature of the stress. Here we show that the
ability of the maleimide analogues to induce UPR as measured by
Another UPR-responsive transcriptional target of ATF-4, DDIT3
(CHOP/GADD153) was significantly up-regulated (ꢀ2-fold) by all
the active compounds. This has been identified as the most impor-
tant gene responsible for induction of apoptosis when the UPR fails
to resolve protein-folding defects.24 The Tribbles 3 homolog (TRB3)
has been identified as a target of CHOP/ATF4, and it is involved in
CHOP-dependent cell death during ER stress.25 TRB3 was transcrip-
tionally activated by all the active analogues except 1u. Finally,
HSP1A1 was the most highly induced gene on the arrays. Although
its activation is associated with generalized stress response, it has
been linked to a pro-apoptotic role in ER stress.26
eIF2a activation correlates highly with their antiproliferative
potencies. This finding supports recent data indicating that bort-
ezomib-induced apoptosis is directly related to its ability to main-
tain eIF2a
in a phosphorylated, active state.27
3. Conclusion
Our current paper presents the synthesis and evaluation of
maleimides that were designed to potentially serve as Cdc25 phos-
phatase inhibitors by alkylating catalytically-critical cysteine resi-
dues. In HepB3 cell culture assays, the analogues exhibited
Taken in total, the data convincingly indicates induction of a
series of ER stress induced genes following treatment with the
maleimides. This can be attributed to the UPR response induced
by PERK activation of eIF2
a. Figure 4A demonstrates that there
antiproliferative IC50 values ranging from >100 lM (e.g., 1q, 1r
was a significantly greater activation of eIF2
a
by phosphorylation
Figure 4. Correlation of activation of p-eIF2
ordered by potency. The level of p-eIFf2 (Ser51) induction was calculated and graphed using normalization to actin expression (n = 3 SD). (B) Linear regression curve for
sensitivity and activation (average, n = 3) of eIF2 R = 0.964, p <0.001).
a (Ser51) with HEP3B sensitivity (IC50) to maleimide analogues. (A) A representative western blot of induction by analogues
a
a

4540
U. Muus et al. / Bioorg. Med. Chem. 18 (2010) 4535–4541
and 1w) to sub-micromolar (1t). While the intended phosphatase
inhibition through cysteine alkylation could not be demonstrated,
studies with a subset of analogues correlated antiproliferative
potencies with activation of the UPR response. These results may
indicate the potential utility of UPR activation as means of achiev-
ing selective antiproliferative effects.
layers were washed with brine and dried over magnesium sulfate.
Removal of solvent gave crude product (1a–1z and 2a–2d), which
was purified by silica gel flash chromatography. All compounds
were purified by HPLC chromatography (see Supplementary data
for synthetic details and analytical data of individual compounds).
4.4. Growth inhibition assays
4. Experimental section
4.1. General
HEP3B cells (ATCC No. HB-8064) were purchased from the
American Type Culture Collection (Rockville, MD). Cells were grown
in Eagle’s minimal essential medium (EMEM) supplemented with
10% heat-inactivated fetal bovine serum (FBS), 2 mmol/L
L-gluta-
All chemicals used in syntheses were reagent grade (Aldrich,
Milwaukee, WI) and were used without additional purification.
Hexanes, ethyl acetate, methylene chloride, and methanol em-
ployed in chromatography were HPLC-grade. Flash chromatogra-
phy was performed with Merck silica gel, 70–230 mesh and ISCO
prepacked columns on a Teledyne ISCO CombiFlash Companion
instrument. Analytical thin layer chromatography was performed
on Analtech precoated plates (Uniplate, silica gel GHLF, 250 mi-
crons) containing a fluorescence indicator and compounds were
visualized by UV and Hannesian agent (21 g of ammonium molyb-
date (VI) tetrahydrate, 1 g cerium sulfate tetrahydrate, 31 mL con-
centrated sulfuric acid, 469 mL H2O) by warming with a heat gun.
NMR spectra were recorded on Varian 200 or 400 MHz spectrome-
ters with Me4Si as an internal standard. Coupling constants are re-
ported in Hertz, and chemical shifts are reported in the delta (ppm)
scale; abbreviations s (singlet), d (doublet), t (triplet), q (quartet)
and m (multiplet). Preparative HPLC was performed on a Beckman
Coulter HPLC system equipped with 128P solvent module and 166P
detector using a Phenomenex Jupiter C18 column (250 Â 21.2 mm;
mine, 100 units/mL penicillin and 100 g/mL streptomycin. Cells
l
were grown at 37 °C in humidified atmosphere consisting of 5%
CO2 and 95% air. Cells in exponential growth phase were used
throughout. The sensitivity of HEP3B to seven selected maleimide
analogs was tested following 48 h incubation with 1.6–50 lM of
each analog. Growth inhibition (% treated/control) was calculated
after a further 6 h incubation (37 °C) with the metabolic dye, alamar
blue, (Sigma), and the fluorescence intensity measured on a Tecan
Ultra plate reader (509-nm excitation and 520-nm excitation).
4.5. Transcriptional profiling
Human OperonV2, 20 K arrays, (National Cancer Institute
microarray facility/Advanced Technology Center, Gaithersburg,
Maryland) were utilized according to published protocols (www.
nciarray.nci.nih.gov). Using competitive hybridization of treated
versus untreated samples chemically coupled to a Cy™3 or Cy™5
fluorescently labeled dye (Amersham Biosciences, Little Chalfont
Buckinghamshire, England), fluorescence was read on a GenePix
4100A microarray scanner (Axon Instruments, Union City, Califor-
nia). Data was analyzed using Axon GenePix Pro 4.1 software and
data and image files were uploaded to the National Cancer Insti-
tute/Cancer Center for Research Microarray Center mAdB Gateway
for analysis and comparison of multiple arrays.
particle size 4 lm). Solvent systems were H2O/acetonitrile, 0.1% tri-
fluoroacetic acid, gradient 5–100%. HPLC-MS analyses were per-
formed on an Agilent 1100 Series LC/MSD model G1946A (Agilent
Technologies Inc., Santa Clara, CA) with an API-electrospray source
using a Zorbax 300SB-C18 Poroshell column (2.1 Â 75 mm; particle
size 5 lm) with a solvent system of H2O/acetonitrile 0.5% acetic
acid, gradient 5–100% over 4 min at a flow rate of 1 mL/min. High
resolution mass spectra (HRMS) were acquired on an Agilent
6520 Accurate-Mass Q-TOF LC/MS System, (Agilent Technologies,
Inc., Santa Clara, CA) equipped with a dual electro-spray source,
operated in the positive-ion mode using a Zorbax 300SB-C3 Poro-
4.6. Real time polymerase chain reaction assays
For each sample, five-hundred nanograms of total RNA was re-
verse transcribed using the High Capacity cDNA Reverse Transcrip-
tion kit (ABI, Foster City, CA). Quantitative real time PCR reactions
were conducted and measured using the ABI Prism™ 7500 Se-
quence Detection System and TaqManÒ SYBR chemistries with pri-
mer pairs designed for each gene of interest. Samples were tested
in triplicate wells for the genes of interest and for the endogenous
control, GAPDH. Data was analyzed using the comparative Ct
method as described in the Perkin Elmer User Bulletin #2 (ABI
PrismÒ 7700 Sequence Detection System, 1997) and expressed as
a fold induction of the gene in the drug treated samples compared
to the untreated control samples.
shell column (2.1 Â 150 mm; particle size 5
lm). Data acquisition
and analysis were performed using MassHunter software.
4.2. General procedure A: Preparation of 1-substituted-3,4-
dibro momaleimides (6a–6z)
To a stirred solution of 3,4-dibromomaleic acid (5) in acetic acid
on ice was added phenylamine (R-NH2, Scheme 1; 1.1 equiv) drop-
wise. The mixture was then stirred at reflux under argon (over-
night). After cooling, the mixture was poured into ice water and ex-
tracted (ethyl acetate). The combined organic layers were washed
with saturated sodium bicarbonate and brine and dried over mag-
nesium sulfate. Removal of solvent provided crude product (6a–
6z), which was purified by flash chromatography on silica gel
(see Supplementary data for synthetic details and analytical data
of individual compounds).
4.7. Western blots
Whole cell protein lysates were prepared for protein analysis by
western blot. Protein was quantitated using the Bradford Protein
Assay (Bio-Rad Laboratories, Hercules, California), and approxi-
mately 50
10% Tris glycine gels (Invitrogen, Carlsbad, California) and probed
with phosphor-EIF2 (SER51) (Cell Signaling, Danvers, MA). Pro-
lg of each sample was resolved using SDS–PAGE on
a
4.3. Procedure B: Preparation of 1-substituted-3,4-bis-(2-
hydroxy-ethylsulfanyl)-maleimides (1a–1z and 2a–2d)
teins were visualized using chemiluminescence and imaged using
a Kodak™ X-OMAT 2000A Processor (Rochester, New York).
To a solution of 3,4-dibromomaleimide (6) and imidazole
(2.2 equiv) in THF was added 2-mercaptoethanol and the mixture
was stirred (3 h). The reaction mixture was then partitioned
between ammonium and ethyl acetate and the combined organic
Acknowledgments
We thank the staff of the Macromolecular Crystallography Lab-
oratory Protein Production Core for preparing the Cdc25A protein

U. Muus et al. / Bioorg. Med. Chem. 18 (2010) 4535–4541
4541
4. Nishikawa, Y.; Carr, B. I.; Wang, M.; Kar, S.; Finn, F.; Dowd, P.; Zheng, Z. B.;
Kerns, J.; Naganathan, S. J. Biol. Chem. 1995, 270, 28304.
5. Kar, S.; Wang, M.; Ham, S. W.; Carr, B. I. Biochem. Pharmacol. 2006, 72,
1217.
6. Choi, D.-S.; Huang, S.; Huang, M.; Barnard, T. S.; Adams, R. D.; Seminario, J. M.;
Tour, J. M. J. Org. Chem. 1998, 63, 2646.
7. Boutros, R.; Dozier, C.; Ducommun, B. Curr. Opin. Cell Biol. 2006, 18, 185.
8. Herzenberg, L. A.; Parks, D.; Sahaf, B.; Perez, O. D.; Roederer, M.; Herzenberg, L.
A. Clin. Chem. 2002, 48, 1819.
9. Lyon, M. A.; Ducruet, A. P.; Wipf, P.; Lazo, J. S. Nat. Rev. Drug Disc. 2002, 1,
961.
10. Szegadi, E.; Logue, S. E.; Gorman, A. M.; Samali, A. EMBO Rep. 2006, 7, 880.
11. Malhotra, J. D.; Kaufman, R. J. Antioxid. Redox Signal. 2007, 9, 2277.
12. Schröder, M. Cell. Mol. Life Sci. 2008, 65, 862.
13. Rasheva, V. I.; Domingos, P. M. Apoptosis 2009, 14, 996.
14. Wang, X.; Thomas, B.; Sachdeva, R.; Arterburn, L.; Frye, L.; Hatcher, P. G.;
Cornwell, D. G.; Ma, J. Proc. Nat. Acad. Sci. U.S.A. 2006, 103, 3604.
15. Ron, D.; Walter, P. Nat. Rev. Mol. Cell. Biol. 2007, 8, 519.
16. Harding, H. P.; Zhang, Y.; Ron, D. Nature 1999, 397, 271.
17. Ma, Y.; Hendershot, L. M. J. Biol. Chem. 2003, 278, 34864.
18. Luo, S.; Baumeister, P.; Yang, S.; Abcouwer, S. F.; Lee, A. S. J. Biol. Chem. 2003,
278, 37375.
used in this study. This Work was supported in part by the Intra-
mural Research Program of the NIH, Center for Cancer Research,
NCI-Frederick and the National Cancer Institute, National Institutes
of Health; in part by the Developmental Therapeutics Program in
the Division of Cancer Treatment and Diagnosis of the National
Cancer Institute; and in part with federal funds from the National
Cancer Institute, National Institutes of Health, under contract N01-
CO-12400. The content of this publication does not necessarily re-
flect the views or policies of the Department of Health and Human
Services, nor does mention of trade names, commercial products,
or organizations imply endorsement by the US Government.
Supplementary data
Supplementary data (synthetic details and compound physical
data as well as experimental data showing 1,4 Michael-type addi-
tion of 1t and N-acetyl cysteine presented) associated with this
article can be found, in the online version, at doi:10.1016/
j.bmc.2010.04.057.
19. Zhang, K.; Kaufman, R. J. Neurology 2006, 66, S102.
20. Lee, Y. Y.; Cevallos, R. C.; Jan, E. J. Biol. Chem. 2009, 284, 6661.
21. Novoa, I.; Zeng, H.; Harding, H. P.; Ron, D. J. Cell. Biol. 2001, 153, 1011.
22. Hollander, M. C.; Poola-Kella, S.; Fornace, A. J., Jr. Oncogene 2003, 22, 3827.
23. Yagi, A.; Hasegawa, Y.; Xiao, H.; Haneda, M.; Kojima, E.; Nishikimi, A.;
Hasegawa, T.; Shimokata, K.; Isobe, K. J. Cell. Biochem. 2003, 90, 1242.
24. Malhotra, J. D.; Kaufman, R. J. Semin. Cell Dev. Biol. 2007, 18, 716.
25. Ohoka, N.; Yoshii, S.; Hattori, T.; Onozaki, K.; Hayashi, H. EMBO J. 2005, 24,
1243.
References and notes
1. Kar, S.; Wang, M.; Yao, W.; Michejda, C. J.; Carr, B. I. Mol. Cancer Ther. 2006, 5,
1511.
2. Ham, S. W.; Park, H. J.; Lim, D. H. Bioorg. Chem. 1997, 25, 33.
3. Ham, S. W.; Park, J.; Lee, S. J.; Kim, W.; Kang, K.; Choi, K. H. Bioorg. Med. Chem.
Lett. 1998, 8, 2507.
26. Kadara, H.; Lacroix, L.; Lotan, D.; Lotan, R. Cancer Biol. Ther. 2007, 6, 705.
27. Schewe, D. M.; Aguirre-Ghiso, J. A. Cancer Res. 2009, 69, 1545.