Optimization of antitumor modulators of pre-mRNA splicing

Article abstract of DOI:10.1021/jm401370h

The spliceosome regulates pre-mRNA splicing, which is a critical process in normal mammalian cells. Recently, recurrent mutations in numerous spliceosomal proteins have been associated with a number of cancers. Previously, natural product antitumor agents have been shown to interact with one of the proteins that is subject to recurrent mutations (SF3B1). We report the optimization of a class of tumor-selective spliceosome modulators that demonstrate significant in vivo antitumor activity. This optimization culminated in the discovery of sudemycin D6, which shows potent cytotoxic activity in the melanoma line SK-MEL-2 (IC₅₀ = 39 nM) and other tumor cell lines, including JeKo-1 (IC₅₀ = 22 nM), HeLa (IC₅₀ = 50 nM), and SK-N-AS (IC ₅₀ = 81 nM). We also report improved processes for the synthesis of these compounds. Our work supports the idea that sudemycin D6 is worthy of further investigation as a novel preclinical anticancer agent with application in the treatment of numerous human cancers.

Full text of DOI:10.1021/jm401370h

Article
pubs.acs.org/jmc
Optimization of Antitumor Modulators of Pre-mRNA Splicing
Chandraiah Lagisetti,†,§ Gustavo Palacios,†,§ Tinopiwa Goronga,† Burgess Freeman,‡ William Caufield,‡
and Thomas R. Webb*,†,∥
‡
†Department of Chemical Biology and Therapeutics, Preclinical PK Shared Resource, St. Jude Children’s Research Hospital, 262
Danny Thomas Place, Memphis, Tennessee 38105-2794, United States
S
* Supporting Information
ABSTRACT: The spliceosome regulates pre-mRNA splicing, which
is a critical process in normal mammalian cells. Recently, recurrent
mutations in numerous spliceosomal proteins have been associated
with a number of cancers. Previously, natural product antitumor
agents have been shown to interact with one of the proteins that is
subject to recurrent mutations (SF3B1). We report the optimization
of a class of tumor-selective spliceosome modulators that demonstrate
significant in vivo antitumor activity. This optimization culminated in
the discovery of sudemycin D6, which shows potent cytotoxic activity
in the melanoma line SK-MEL-2 (IC50 = 39 nM) and other tumor cell
lines, including JeKo-1 (IC50 = 22 nM), HeLa (IC50 = 50 nM), and
SK-N-AS (IC50 = 81 nM). We also report improved processes for the synthesis of these compounds. Our work supports the idea
that sudemycin D6 is worthy of further investigation as a novel preclinical anticancer agent with application in the treatment of
numerous human cancers.
There is increasing evidence that aberrant splicing of pre-
mRNA is a driver of tumorigenesis4 and that modulation of this
process may be a valid target for cancer therapy.17−20
Subsequent to the characterization of FR and PD as
spliceosome modulators, recurrent mutations in SF3B1 were
independently identified in myelodysplastic syndromes
(MDS),21 chronic lymphocytic leukemia (CLL),22 acute
myeloid leukemia (AML),23 and uveal melanoma.24 The
selective sensitivity of tumors to agents such as PD and
meayamycin has been the topic of significant research efforts
and has very recently been linked to overexpression of MYC25
or the perturbation of the pre-mRNA splicing of Mcl-1,26
respectively. Thus, the collective progress in natural-product
screening, target identification, and high-throughput tran-
scriptome sequencing has led to a remarkable convergence of
independent research areas, which have simultaneously
identified new oncology drug targets and new small-molecule
therapeutic agents.17,20
Numerous total synthetic approaches to FR have been
reported, which is one of these active natural products. Most of
these syntheses directly target the highly intricate natural
product itself or corresponding synthetically complex ana-
logues.28−32 In spite of much progress, it remains challenging to
synthetically access large quantities of the natural-product
spliceosome modulators or their close analogues because of the
structural complexity and the presence of eight or more
stereocenters.17 As part of our effort to develop a class of
druglike and elegant natural-product analogues, we previously
INTRODUCTION
■
In most eukaryotic cells, the splicing of pre-mRNA is a critical
process that is catalyzed and regulated by the spliceosome
machinery.1 The spliceosome is a complex of many proteins
and RNA that includes five small-nuclear ribonucleoproteins
(snRNPs) (U1, U2, U4/U6, and U5) and more than 150 other
associated proteins.2,3 The spliceosome catalyzes removal of
noncoded introns from pre-mRNA, joins coded exons together
to generate mature RNA, and also controls the process of
alternative splicing, which is an important mechanism for
regulating cellular function.4 Two bacterial natural products,
FR901464 (FR)5−7 and pladienolide (PD)8,9 (Figure 1), have
been reported to target the SF3B subunit of the spliceo-
some.10,11 Subsequent to those initial discoveries, the list of
bacterial natural products targeting the SF3B subunit has grown
(Figure 1) to include other bacterial natural products including
herboxidiene (GEX1A)12 (isolated from Streptomyces sp.
A7847) and the thailanstatins (isolated from Burkholderia
thailandensis).13 A similar interaction with the SF3B subunit is
also likely for the macrolide natural product FD-895 given its
potent biological activity and its high structural and
pharmacophore similarity to PD.14,15 These bacterial fermenta-
tion products show cytotoxic IC50’s in the low nanomolar range
in several tumor cell lines and have been reported to have a
similar distinctive effect on the cell cycle in mammalian cell
lines, including cell cycle arrest in the G1 and G2/M phases.5,16
Several of these natural products have also been reported to
show potent antitumor activity in vitro and in vivo.17 Work in
this area led to the development of the semisynthetic PD
analogue E7107 (Figure 1) for clinical studies.11
Received: September 6, 2013
Published: December 11, 2013
© 2013 American Chemical Society
10033
dx.doi.org/10.1021/jm401370h | J. Med. Chem. 2013, 56, 10033−10044

Journal of Medicinal Chemistry
Article
continue to be a major therapeutic challenge in the field of
pediatric and adult cancer chemotherapy.37 We have previously
reported the initial SAR with the early generations of these
compounds in addition to the initial results of the in vitro
evaluation of these analogues in a panel of tumor cell lines.33,34
This work led to the identification of several types of drug-
sensitive tumor lines, which included several melanoma lines.
Because of this observed drug sensitivity combined with the
critical need for new therapeutic agents for melanoma, we have
also emphasized in vivo experiments with melanoma. As such,
melanoma as a model disease serves to exemplify a major
potential clinical application for a sudemycin lead candidate.
We now report progress in lead-optimization efforts that have
led to novel sudemycin analogues possessing in vivo activity in
a murine melanoma xenograft model. Additionally, we also
report the results from in vitro dual-agent synergy/antagonism
experiments with these newly optimized sudemycin derivatives
when combined with a set of approved and investigational
antitumor drugs.
In the following, we describe additional SAR results and new
compounds that possess important structural modifications
required for enhanced in vivo antitumor activity. In addition to
substantial new data on the biological activity of these
compounds, we also report new and substantially improved
synthetic processes including the incorporation of a novel
sulfone reagent for the key Julia−Kocienski coupling step. The
synthetic refinements allow for a much more efficient and
scalable process that makes these sudemycin analogues
available for clinical development.
Figure 1. Natural product spliceosome modulators (FR901464,10
pladienolides,11 herboxidiene,12 thailanstatins,13 and FD-895)15 along
with semisynthetic analogues (E710711 and spliceostatin10) and a
structurally similar totally synthetic analogue class (meayamycins).27
RESULTS AND DISCUSSION
■
reported the design and synthesis of FR analogues (the
sudemycins)33−35 and pladienolide analogues,36 which are
active simplified compounds that effectivelly modulate alternate
splicing.35 Most of our related lead-optimization work has
focused on the sudemycins, which are chemically stable
synthetic molecules that were designed on the basis of the
known FR structure−activity relationships (SAR) and the
application of a hypothetical consensus pharmacophore model
that is derived from molecular overlays of FR and PD.33−35
Although the application of small-molecule splicing modu-
lators may have broad anticancer potential, we chose to focus
significant attention on melanoma for the purposes of the
present study because melanoma is one of several cancers that
Synthesis. Previously we have reported the practical
enantioselective and diastereospecific synthesis of key inter-
mediates for the sudemycin series that includes a key Julia−
Kocienski olefination (Scheme 1, top panel).33,34 Our original
synthetic scheme, although quite useful for the synthesis of new
analogues, did not represent an efficient process for large
amounts of sudemycin analogues for in vivo studies. We now
report a much improved procedure (Scheme 1, bottom panel,
and Scheme 2) using an alternate retrosynthetic dissection,38
which makes use of the novel and highly crystalline sulfone 5.
As can be seen in Scheme 1, this process gives a much higher
overall yield of the key coupling product. As can also be seen in
a
Scheme 1. Comparison of the Original Synthetic Approach to the Sudemycins to the Improved Route
a
Bt = 2-benzo[d]thiazole.
10034
dx.doi.org/10.1021/jm401370h | J. Med. Chem. 2013, 56, 10033−10044

Journal of Medicinal Chemistry
Article
of our spliceosome-modulator lead-optimization work, we
identified potency (as measured initially by cytotoxicity and
confirmed by potent effects on the modulation of MDM2
splicing), druglike solubility, and rapid onset of cytotoxicity (as
measured by drug washout experiments) as the primary initial
qualities needed in a candidate drug molecule. Compounds
were evaluated by initial screening in a 72 h cytotoxicity against
a small panel of tumor cell lines (generally, JeKo1 and PC-3)
and were selected for further evaluation on the basis of potency
(IC50 < 250 nM) in one of the cell lines of interest. Finally, the
best leads were ranked in order of the most promising solubility
and metabolic stability characteristics. We have found the
structure−activity of these spliceosome-modulatory com-
pounds to be very sensitive to even small changes in the
general scaffolds (derivatives of 1 or 2, shown in Figure 2). Our
work33,34 and previously reported SAR work on analogues that
have close structural similarity to FR have shown pronounced
limits on the activity tolerance for many structural changes.27 In
these cell-based assays, biological activity can be affected by
numerous factors, includeing transport into the cell and into
the nucleus (in order to reach the spliceosome) and cellular
metabolic stability in addition to the affinity for the
spliceosome. Therefore, many of the modifications that we
examined resulted in a significant reduction in cytotoxic
activity, presumably by negatively impacting one of these
important factors. These analogues were prepared using the
previously published chemistry or that shown in Scheme 3.
Because FR and our first series of active compounds all
contain a metabolically labile ester group, we have previously
investigated and reported the activity of the alcohol metabolites
of our compounds, which were found to have cytotoxicity IC50
values of >1 μM against all cell lines examined,35 consistent
with the pharmacophore model that we have previously
proposed.33 Therefore, we have now expanded on our previous
reports via the synthesis and evaluation of new ester and
optimized carbamate derivatives of the general structure shown
in Figure 2 to identify potent compounds with improved
solubility and improved metabolic stability.34 As can be seen by
inspection of Table 1, a new ester (19a) and a diverse
collection of carbamate derivatives (19b−o) were prepared and
screened for cytotoxicity. With regard to the following SAR
analysis, our focus was on the identification of potent
compounds with IC50’s significantly below 1 μM. Because of
the inherent error in cytotoxicity measurements, it is possible
that some IC50’s may actually overlap. However, several general
trends can be seen. For example, although some level of
substitution was possible, the compounds that contained longer
chain groups (or polar solubilizing groups) were less potent
than 1. As can be seen by inspection of Table 1 together with
our previously published work on carbamate derivatives of 18,34
there was limited tolerance for the length of the alkyl chains in
Scheme 2. Optimized Synthetic Approach to Key
Intermediate 5
a
a
(i) KOtBu, Me3SOI, DMSO; (ii) polymethylhydrosiloxane, Ti-
(OiPr)4; (iii) 2-benzthiazoleSH, Ph3P, diisopropyl diazodicarboxylate,
THF; (iv) ammonium molybdate, 30% H2O2.
Scheme 2, Julia−Kocienski reagent 5 can be readily derived
from the previously reported spiroepoxy alcohol 8.33 Additional
refinements to the synthesis of 5 include the replacement of
hazardous NaH with the much more practical potassium t-
butoxide for the generation of the sulfur ylide in the
diastereospecific epoxidation step (to prepare 7, see the
Experimental Section). We were also able to replace the
hazardous pyrophoric reagent KH with potassium t-butoxide as
a refinement in the preparation of the intermediate (S,Z)-4-
((tert-butyldimethylsilyl)oxy)pent-2-enoic acid used in this
scheme (see the Supporting Information).33 A further enhance-
ment in the synthetic process is the replacement of the di-
isobutyl aluminum hydride with the safer and more cost-
effective polymethylhydrosiloxane (PMHS)/Ti(OiPr)4 combi-
nation39 for the selective reduction of the ester to alcohol in the
presence of a spiroepoxide group (Scheme 2). This new route
also replaces the t-butoxycarbonyl protecting group with azide
protection.40,41 (All synthetic protocols for these improvements
are included in the Experimental Section and Supporting
Information.) These refinements allow for the practical
synthesis of >10 g amounts of sudemycin derivatives in a
standard research laboratory setting and will facilitate the
development of synthetic processes suitable for larger scale
synthesis that will be required for clinical studies.
Structure−Activity Relationships (SAR). Several factors
must be optimized simultaneously to develop a lead compound
that displays adequate tumor efficacy in vivo. These factors
include (1) potent on-target cell-based activity against the
tumor, (2) reasonable metabolic stability (so that the drug will
reach the tumor), (3) solubility (so that the drug can be
formulated for in vivo delivery), and, importantly, (4) rapid
onset of action (so that the drug acts before it is metabolized).
The development of a compound that optimizes all of these
parameters simultaneously is a significant challenge. In the case
Figure 2. Structures of the previously reported compounds sudemycin C133 and sudemycin F135 along with the new preclinical lead compounds
sudemycin D1 and D6, which have not previously been disclosed.
10035
dx.doi.org/10.1021/jm401370h | J. Med. Chem. 2013, 56, 10033−10044

Journal of Medicinal Chemistry
Article
a
Scheme 3. Improved Synthesis of Sudemycin Analogues
a
(i) Borane/THF, 0 °C to rt; (ii) (1) TFA, (2) K2CO3, (3) imidazolesulfonyl azide, K2CO3; (iii) (1) Swern oxidation, (2) ethyl 2-
(triphenylphosphoranylidene)propanoate; (iv) DIBAL-H, CH2Cl2, −78 °C; (v) pyridine, Dess−Martin periodinane; (vi) 5/NaHMDS, THF, −78
°C; (vii) (1) Ph3P, H2O, 55 °C, (2) 2.DIPEA, HBTU, (S,Z)-4-((TBS)oxy)pent-2-enoic acid; (viii) for carbamates (1) TBAF/THF, (2) 4-
nitrophenylchloroformate/THF; (ix) R1R2NH/THF or R′′COCl/pyr for esters.
a
Table 1. SAR of the New Ester and Carbamate Derivatives Prepared Using Previously Reported Procedures
R′
ID
JeKo1
PC3
R′
ID
JeKo1
PC3
-C(CH3)3
19a
19b
19c
19d
19e
19f
0.34
>1
1.2
N-methyl-N-t-butylamine
morpholine
piperidine
19i
19j
19k
19l
19m
1
0.78
1.32
1.47
0.97
0.94
0.3
2.7
7.8
7.2
4.6
5.2
4.4
>10
>10
n-butylamine
>10
>10
7.3
diethylamine
>1
t-butylamine
>1
pyrollidine
azetidine
N-methyl-N-propylamine
2-methoxy ethylamine
N-methyl-N-butylamine
N-methyl-N-isopropylamine
0.61
>1
3.2
>10
3.2
-N(CH3)2
19g
19h
0.69
0.7
-NHCH3
19n
19o
0.9
3.4
-NH2
1.9
a
All activities are based on the mean of three experiments after a 72 h exposure to compound. The cytotoxicity IC50 values are all reported in
micromolar units, with the standard error of the mean (SEM) being less than 30% in all cases (determined as previously reported).34
the groups used in the substitution of carbamate derivatives of
these compounds. For example, although pivaloyl ester 19a
showed good activity, the n-butyl, diethyl, and t-butyl
carbamates (19b, 19c, and 19d, respectively) failed to show
submicromolar activity in the two cell lines used in this screen,
which represents a significant reduction in activity in the
cytotoxicity screen (Table 1; the SEM was less than 30% in all
cases and was determined as previously reported for the
compounds in this table).34 In the case of the SAR of N,N-
dialkyl carbamates, it was noted that the presence of a N-methyl
group was better tolerated than other groups, as can be seen
with the N-methyl-N-propyl (19e), N-methyl-N-butyl (19g),
and other N-methyl-N-alkyl derivatives (19h and 19i) when
compared to 19b or 19c. In addition, it is noted that cyclic
carbamates (19j−m) showed activity near 1 μM, so no
significant improvement was seen with these cyclic compounds.
The most active early esterase stable compound in this series
was the dimethylamino carbamate derivative 1, which we
named sudemycin F1.35 On the basis of this SAR data, it could
be hypothesized that a compact and relatively nonpolar pocket
near (or on) the SF3B subunit interaction site exists for the
ester and carbamate functionalities because activity goes down
with both the length of hydrophobic groups and with increasing
polarity (Table 2), although other effects, such as cellular
uptake, undoubtably influence activity as well.
We also explored the replacement of the carbonyloxy group
that is present in the active carbamates and esters with other
dual hydrogen-bond-acceptor groups. If the key pharmaco-
phore feature of the interaction of the carbonyloxy group is the
presence of the dual 1,3 dihydrogen-bond acceptors, then other
functional groups might also fulfill this interaction and at the
same time be fully resistant to esterase metabolism. To explore
this hypothesis, we prepared a set of compounds shown in
Scheme 4. These heterocyclic derivatives could all be
reasonable fits to the pharmacophore model if simple
hydrogen-bond dual acceptors are required; however, none of
these compounds showed cytotoxic IC50’s below 10 μM. This
suggests that the steric requirements for this fit may be very
constrained.
In addition, we explored the importance of the NH group in
these analogues; the pharmacophore model predicted the
importance of the NH hydrogen-bond donor of the amide
group, and as predicted the N-methyl derivative of 19p (19q,
see the Supporting Information) showed a complete abrogation
10036
dx.doi.org/10.1021/jm401370h | J. Med. Chem. 2013, 56, 10033−10044

Journal of Medicinal Chemistry
Article
a
Table 2. Comparison of the Most Active Lead Compounds
a
All activities are based on the mean of three experiments after a 72 h exposure to compound. The cytotoxicity IC50 values are all reported in
micromolar units. The standard error of the mean (SEM) for cytotoxicity values was less than 30% in all cases and was determined as previously
reported34 or as described in the Experimental Section. Previously reported compounds are included for comparison and are indicated by the
b
appropriate reference. (Protocols for all assays are provided in the Supporting Information.) ND = not determined. Major splicing changes
c
d
observed with 1 μM drug treatment of Rh18 cells. In mouse plasma at 37 °C, as determined by HPLC. Solubility (as determined by HPLC) in
e
f
PBS. The time to achieve the majority of the observed 72 h cytotoxicity on the basis of drug-washout experiments. 72 h cytotoxicity measured using
an MTT assay, as previously reported.34 72 h cytotoxicity using a Cell Titer Glow assay, as described in the Experimental Section.
g
ranked lead compounds, as shown in Table 2. Compound 3 has
the best profile and stands out because it possesses potent cell-
based activity combined with druglike measured solubility (45
μM in PBS). Compound 3 also showed potent cytotoxic
activity in several other tumor lines investigated (see the
Supporting Information), including the melanoma line SK-
MEL-2 (IC50 = 39 nM) as well as Hela (IC50 = 50 nM), SK-N-
AS (IC50 = 81 nM), and PC-3 (IC50 = 142 nM).
Scheme 4. Synthesis of Heteroaryl Derivatives
Drug Synergy Studies. There is increasing evidence that
certain synergistic anticancer drug combinations can be both
more effective and less toxic that single-agent drug protocols.42
In consideration of this fact, we next turned our attention to the
in vitro anticancer drug combination properties of our splicing
modulators in tumor cell lines by exploration of the combined
dual-agent cytotoxic effects of compounds 2 or 3 with
successive members of a panel of 23 diverse types of clinically
useful anticancer drugs and bioactive agents (Table 3).
Although many of the combinations investigated (10 of the
23 agents) did not demonstrate either antagonism or synergy,
we were able to uncover a number of drugs that show a
significant combination effect in the first round of this screening
efforts using the first drug panel shown in Table 3. Although
the identification of antagonistic or synergistic drug combina-
tions remains an empirical process, it is generally believed that
agents that are antagonistic may both be targeting related
signaling pathways and that agents that are synergistic are
targeting distinct and complementary pathways.42 In principle,
it is also possible that an agent can also show synergy in other
ways, for example, by interfering with an important drug
metabolic process and therefore stabilizing an agent that is a
component of the combination therapy.
of activity. These results taken together demonstrate the
exquisite sensitivity of the spliceosome modulator pharmaco-
phore, which is consistent with other published work.27,28
However, despite these constraints, we were able to prepare a
set of highly active compounds, as summarized in Table 2.
To explore further the SAR, we also prepared a set of
exploratory compounds based on the less soluble series
(containing the cyclohexyl ring rather than the 1,3-dioxane
ring) to explore a set of conservative changes that could give
active low-molecular-weight analogues with better solubility
properties (Table 2). We were very pleasantly surprised to find
that we had, for the first time, prepared compounds with
improved cytotoxic activity, which also showed improved
measured solubility properties. Most strikingly, carbamate 3
showed low double-digit potency in contrast to the reduced
activity seen in the monomethyl carbamate derivative 19n when
compared to 1. We then performed careful analysis of the top-
Subsequent to the first screen, we focused our attention on
agents showing synergy; the most notable of these was
BEZ235, a dual inhibitor of phosphatidylinositol-3-kinase
(PI3K) and the mammalian target of rapamycin (mTOR),43
which is currently under investigation in numerous clinical
trials. Given the strong synergy that we observed with BEZ235,
10037
dx.doi.org/10.1021/jm401370h | J. Med. Chem. 2013, 56, 10033−10044

Journal of Medicinal Chemistry
Article
a
supporting the possibility that this epoxide metabolic pathway
is significant for these drugs in vitro and potentially in vivo. It
should be noted that because agents such as AUDA do not
show cytotoxic effects the combination index (CI) can not be
calculated for them. Of course, these drug-synergy studies are
relevant for the design of potential future clinical studies that
would include the use of relevant drug combinations of
sudemycin D6.
Table 3. Drug Combinations Showing Antagonism or
b
Synergy of Cytotoxicity of SK-MEL-2 with Compound 2
first drug panel
interaction
drug
mechanism of action
type
CI in SK-MEL-2
0.21
BEZ235
dual inhibitor of PI3K
and mTOR
strong
synergism
c
bortezomib
proteasome inhibitor
antagonism
n.d.
In Vivo Tumor Xenograft Studies. Previously, our group
reported the in vivo efficacy of the early lead sudemycin C1,35
which inhibited tumor growth in a lymphoma xenograft model
when delivered intravenously.34 Although the sudemycin C1-
treated mice showed significant inhibition of tumor growth in
this model (when compared to that of the vehicle-treated
group), tumor proliferation was not completely inhibited. It is
likely that the incomplete inhibition seen in our earlier study
was a consequence of the short in vivo half-life of sudemycin
C1. We propose that the observed short half-life is due to the
presence of a labile ester group that is present in sudemycin C1,
which we have previously shown.34 It is well-known that esters
are substrates for plasma and microsomal esterases that can
rapidly metabolize esters to the corresponding alcohols, and
this topic has recently been reviewed.48 We have previously
reported that hydrolysis of sudemycin C1 leads to an inactive
alcohol (see the Supporting Information for the assay
protocol).34 The other metabolically labile group that is
required for activity in the sudemycins and the natural
products27 is the epoxide group, which can be a target for
epoxide hydrolases, which would produce the inactivated diol
metabolites.46 This led us to conclude that it was necessary to
deliver the drug via slow continuous jugular catheter (JVC)
infusion in our in vivo efficacy studies.
Thus, following maximum tolerated dose (MTD) (data not
shown) and pharmacokinetic studies (see the Supporting
Information, in vivo pharmacokinetic studies), we sought to
determine the efficacy of lead compound 3 in the inhibition of
tumor growth in mice (Figure 4). Thus, SK-MEL-2 tumors
were transplanted into male C.B-17 scid mice. The animals
were subsequently infused over 4 h with either vehicle (eight
mice) or compound 3 (nine mice) for 5 consecutive days. No
differences in fatalities or behavioral changes were observed
between the control infusion and the drug-treatment groups. In
the drug-treated group (50 mg/kg of compound 3 over 4 h
once a day for 5 days) (Figure 4), we observed 100% inhibition
of tumor growth during the infusion period and significant
inhibition subsequent to this dosing period. When compound 3
was infused over a period of 4 h, it also induced moderate
toxicity (as measured by weight loss); we observed an average
maximum loss of weight of 12.7 2.3% at 5 days of treatment.
Nevertheless, the drug-treated group regained weight immedi-
ately after the cessation of the drug infusion, and the animals
returned to a normal weight range compared to the vehicle-
infused mice by day 22 of the study. This demonstrates the
single-agent antitumor effects of splicing modulator 3 in this
xenograft model of an aggressive metastatic human melanoma.
camptothecin DNA topoisomerase I
inhibitor
moderate
synergism
0.74
harringtonine inhibitor of protein
synthesis
antagonism
n.d.
imatinib
tyrosine kinase inhibitor antagonism
broad-spectrum HDAC moderate
inhibitor
n.d.
panobinostat
0.76
synergism
silibinin
sorafenib
SU 9516
taxol
multiple Targets
slight
synergism
synergism
0.89
0.43
0.83
n.d.
multikinase inhibitor
(inhibits sEH)
cyclin-dependent
kinase-2 inhibitor
moderate
synergism
microtubule inhibitor
antagonism
follow-up drug panel
interaction
type
drug
mechanism of action
CI in SK-MEL-2
0.57
PX-866
irreversible inhibitor of
PI3K
synergism
temsirolimus inhibitor of mTORC1
AUDA inhibitor of sEH
synergism
synergism
0.68
n.d.
a
For the complete table including the full list of drugs that were tested,
b
see Supporting Information Tables 2 and 3. See the Supporting
Information for combination indexes for each drug combination
c
investigated. The notation n.d. indicates that the CI value was not
determined. For some compounds such as AUDA, it is not possible to
determine a CI because AUDA does not show cytotoxicity as a single
agent. Drug interaction was established by combination of various
concentrations of compound 2 combined with the indicated drug at a
constant concentration.
we decided to investigate the combined effects of selective
inhibitors of the two individual targets (i.e., PI3K and
mTOR).44 For this we chose PX-866 and temsirolimus,
which are selective inhibitors of PI3K and mTOR, respectively.
Interestingly, both of these agents showed synergy with lead
compound 2, although not as strong as that observed with the
dual-acting agent BEZ235 (Table 3, follow-up drug panel). As
evident in Figure 3, this synergy is also seen with our best lead
for both 2 and 3 in all cell lines investigated. The second
strongest synergistic agent in the first drug panel was the
multikinase inhibitor sorafenib, which is also reported to have
off-target inhibition of soluble epoxide hydrolase (sEH).45
Some mammalian epoxide hydrolases are known to be
important in the metabolism of epoxide-containing xeno-
biotics.46 Because the sudemycins (and related drugs) have an
epoxide group as a critical feature of their pharmacophore,33 it
is possible that this group is metabolized and inactivated by
epoxide hydrolases present in tumor cells during the course of
the cytotoxicity screening. We prepared the diol product of
epoxide ring-opening of compound 3 and found that this
compound showed no detectable activity at a concentration of
10 μM (see the Supporting Information, Scheme S4 and
associated experimental procedures). Therefore, we hypothe-
sized that EH inhibitors may stabilize this class of drugs, so we
explored the use the EH inhibitor AUDA. As shown in Table 3,
AUDA also demonstrates synergy with compound 2,
CONCLUSIONS
■
The regulation of the pre-mRNA splicing process is abnormal
in most cancers, and numerous spliceosomal proteins are
known to be recurrently mutated in a variety of tumors. We
have optimized a class of totally synthetic natural-product
analogues targeting the spliceosome, leading to sudemycin D6
(compound 3), and further explored the SAR of this series.
10038
dx.doi.org/10.1021/jm401370h | J. Med. Chem. 2013, 56, 10033−10044

Journal of Medicinal Chemistry
Article
Figure 3. Interactions between (A) compound 2 and BEZ 235 and (B) compound 3 and BEZ 235 result in strong synergism. The IC50 curves of
drug combinations in the neuroblastoma cell line SK-NA-S are plotted. The combination indexes (CI) shown to the right were calculated at the
concentration of drugs required to inhibit 50% of cell proliferation (ED50) in various cell lines. Each experimental condition was carried out in
triplicate; error bars represent the SEM. Drug interactions were assessed with CalcuSyn 2.1 (Biosoft, Cambridge, UK). The program is based on
Chou−Talalay’s combination index theorem, which renders a quantitative definition of the drugs’ interaction defined as the CI.47
(malignant melanoma), and WiDr (colorectal adenocarcinoma) cells
were grown in EMEM medium containing 10% FBS; and neuro-
blastoma SK-N-AS cells were grown in RPMI 1640 medium
containing 10% FBS. All cell culture media was supplemented with
L-glutamine and penicillin/streptomycin. All cells were grown under
standard conditions in CO2 incubators at 37 °C with 5% CO2 and
100% relative humidity.
Sudemycin D6 shows potent cytotoxic activity in the melanoma
line SK-MEL-2 (IC50 = 39 nM) as well as several other tumor
lines that were investigated, including JeKo-1 (IC50 = 26 nM),
HeLa (IC50 = 50 nM), SK-N-AS (IC50 = 81 nM), and PC-3
(IC50 = 142 nM). Sudemycin D6 also shows rapid and potent
modulation of MDM2 alternate splicing in the Rh18 cell line as
well as acceptable plasma stability, which led to its selection for
in vivo studies. These in vivo studies demonstrate the activity of
sudemycin D6 in a mouse melanoma xenograft model. In
addition to this work, we also report improved synthetic
processes for this class of compounds (including 2 and 3) as
well as data from in vitro drug-synergy studies using a panel of
anticancer drugs. This work supports the idea that sudemycin
D6 is worthy of further investigation as a novel preclinical
anticancer agent, which may have application in the treatment
of several types of human cancers.
In Vitro Cytotoxicity Assays. The concentration of drug required to
inhibit cell proliferation by 50% (IC50) was determined as the
percentage of viable cells remaining after a 72 h exposure to the drug.
Adherent cells were seeded in 100 μL of cell culture medium in 96-
well plates. Depending on the cell proliferation rate, cells were seeded
at 4000−6000 cells/well. Suspension cell cultures were initiated with
20 000 cells/well. The cells were incubated overnight at 37 °C, and
then the drug treatments were started by adding 50 μL of medium
containing the corresponding drug dilution and DMSO. The final
concentration of DMSO onto the cells was 0.5% v/v in all wells. Cell
viability was determined using the CellTiter-Glo Luminescent Cell
Viability Assay (Promega, Madison, WI). The CellTiter-Glo reagent
was prepared according to the manufacturer’s instructions, and then 50
μL of reagent was added to each well for 1 h at rt. The luminescence
was measured on an EnVision Multilabel Plate Reader (PerkinElmer,
Waltham, MA). The IC50 values were calculated from the dose−
response curve analysis with GraphPad Prism 6.01 (GraphPad
Software, La Jolla, CA) using the nonlinear regression model 4
parameter logistic. Nine concentrations were generated by 1:2 serial
dilutions for each IC50 assay. Each drug concentration was evaluated in
triplicate or quadruplicate.
EXPERIMENTAL SECTION
■
Biology General Aspects. Cell Culture. Cancer cell lines were
purchased from the American Type Culture Collection (Manassas,
VA) with the exception of the pediatric rhabdomyosarcoma Rh18,
which was kindly provided by Dr. P. Houghton (Ohio State
University, Nationwide Children’s Hospital, Columbus, OH). The
following cell media was used: JeKo-1 (mantle cell lymphoma) cells
were grown in RPMI 1640 medium containing 20% fetal bovine serum
(FBS); PC-3 (prostate adenocarcinoma) cells were grown in Ham’s F-
12K with 10% FBS; HeLa (cervical adenocarcinoma), SK-MEL-2
10039
dx.doi.org/10.1021/jm401370h | J. Med. Chem. 2013, 56, 10033−10044

Journal of Medicinal Chemistry
Article
NaH2PO4, pH 7.4) or 50 mg/kg of lead compound 3 for 5 consecutive
days. Infusions were carried out by placing the animals inside a
continuous infusion system (Instech Laboratories, Plymouth Meeting,
PA) and delivering the vehicle or drug solution via the jugular catheter
at a rate of 4 μL/min with a SP230iw syringe pump (WPI, Sarasota,
FL). The total volume infused into the animals did not exceed 850 μL
per day. Tumor size and animal weight were both continuously
measured during and after infusion periods, five times per week
(weekdays). Tumor volumes were determined on the basis of the
following equation: Tuvol = (δ1/2 + δ2/2)1.5708, where δ1 and δ2 are
diameters measured with a caliper at right angles to each other. The
size of the tumors ranged from 0.2 to 0.5 cm3 at the times the infusions
were initiated.
Chemistry General Aspects. All materials and reagents were used
as is unless otherwise indicated. Air- or moisture-sensitive reactions
were carried out under a nitrogen atmosphere. THF, toluene,
acetonitrile, N,N-dimethyl formamide, and CH2Cl2 were distilled
before use. All compounds were purified on Biotage prepacked silica
gel columns. TLC analysis was performed using glass TLC plates (0.25
mm, 60 F-254 silica gel). Visualization of the developed plates was
accomplished by staining with ethanolic phosphomolybdic acid, ceric
ammonium molybdate, or ethanolic ninhydrin followed by heating on
a hot plate (120 °C). All compounds for biological assays possessed a
purity of >95% as determined by ultra-high-pressure liquid
chromatography on a Waters Acquity UPLC/PDA/ELSD/MS system
carried out with a BEH C18 2.1 × 50 mm column using gradient
elution with stationary phase: BEH C18, 1.7 mm, solvents: A: 0.1%
formic acid in water, B: 0.1% formic acid in acetonitrile. NMR spectra
were obtained on Bruker Avance II 400 MHz. The values dH 7.26 and
dC 77.0 ppm were used as references for NMR spectroscopy in CDCl3.
The coupling constants deduced from 1H NMR data were obtained by
first-order coupling analysis. Analytical and/or preparative SFC
(supercritical fluid chromatography) systems (AD-H) were used for
analysis and purification of final chiral compounds. IR spectra were
collected using a Nicolet IR 100 (FT IR). FT IR analyses were
prepared neat or as neat films on KBr plates, and the data are reported
in wavenumbers (cm−1) unless specified otherwise. Melting points
(mp) were obtained on a Buchi apparatus and are uncorrected. The
University of Illinois Mass Spectroscopy Laboratories and High
Throughput Analytical Center at St. Jude Children’s Research Hospital
collected the high-resolution mass spectral data.
Figure 4. Compound 3 inhibits SK-MEL-2 tumor growth in a mouse
xenograft model. The arrows indicate a 4 h infusion of 50 mg/kg of
either vehicle or compound 3. Tumor growth was monitored daily five
times per week. n = 8 for the vehicle group and 9 for the drug-treated
group. Mice were euthanized when the tumor burden reached the size
limit (>4 cm3). Error bars represent the SEM. Statistical evaluations
were assessed using one-tailed Student’s t-test, and the symbols above
the lines indicate significant differences at * = p < 0.05 and ¶ = p <
0.01.
Drug Combination Assays. Preliminary cytotoxic screenings for
each drug used in drug combination assays were performed to
determine their cytotoxic dose range. The drug combination assays
were initiated at a dose that was 4 or more times higher than the
concentration of each drug required for the IC50 value. Eight
concentrations were established by 1:3 serial dilutions of the single
drugs and their combinations with the splicing modulators. All drug
concentrations were evaluated in triplicate in 96-well cell culture
plates. Assays of individual drugs and drug combinations were run on
the same cell culture plate. Cell viability was determined using the
CellTiter-Glo reagent. Drug interactions, in terms of synergism,
additive effect, or antagonism, were based on the drugs’ CI.47
Chemistry Procedures. (S,Z)-5-(((2R,5R)-2-((2E,4E)-5-((3R,5S)-
7,7-Dimethyl-1,6-dioxaspiro[2.5]octan-5-yl)-3-methylpenta-2,4-
dien-1-yl)-1,3-dioxan-5-yl)amino)-5-oxopent-3-en-2-yl Dimethyl-
carbamate (1). A stirred solution of (S,Z)-5-(((2R,5R)-2-((2E,4E)-
5-((3R,5S)-7,7-dimethyl-1,6-dioxaspiro[2.5]octan-5-yl)-3-methylpen-
ta-2,4-dien-1-yl)-1,3-dioxan-5-yl)amino)-5-oxopent-3-en-2-yl (4-nitro-
phenyl) carbonate (2.4 g, 4.09 mmol) in dichloroethane (35 mL) in an
ice bath was treated with dimethylamine (6.14 mL, 12.27 mmol, 2 M
in THF) dropwise. The resulting yellow solution was stirred at the
same temperature for 10 min, by which time LC/MS indicated that all
starting material had been consumed. The solvent was then removed
under reduced pressure. The yellow viscous crude residue was
dissolved in a minimum volume of methylene chloride and treated
with hexane. The resulting yellow solids were filtered and washed with
5% methylene chloride in hexane (25 mL). The filtrate was
concentrated and purified on silica gel (100 g, 8−70% acetone in
Combination Index for Constant-Ratio Drug Combinations.
Luminescence data obtained from the cell viability determinations
were formatted per the requirements of the analytical software
(viability data range >0 and <1). The formatted data were analyzed
with CalcuSyn 2.1 (Biosoft, Cambridge, UK), which provides
quantitative CI values for the drug-interaction analyses (see ranges
in Supporting Information Table S2). All CI values were calculated at
the combination dose that produced a cytotoxic effect in 50% of the
cells (ED50). AUDA does not show any cytotoxicity at any achievable
concentration, so it was not possible to determine a CI for those drugs.
Therefore, in these two cases, the drug interaction was evaluated by
comparing the IC50 value of sudemycin standard (D1 or D6) with that
of the IC50 values resulting from the serial dilutions of the splicing
modulator combined with a constant dose of the nontoxic drug.
Infusions in Tumor-Bearing Mice. Xenograft tumors were first
established from an SK-MEL-2 cell culture at the St. Jude Children’s
Research Hospital Xenograft Core. After these tumors were
established as a mouse xenograft model, they were routinely
maintained in the C.B-17 scid strain (Taconic Farms, Germantown,
NY). Experimental animals were prepared by transplanting small
pieces of dissected tumors from donor animals into recipient mice. All
animal studies were performed in accordance with the St. Jude
Children’s Research Hospital Animal Care and Use Committee. SK-
MEL-2 xenograft tumors were transplanted into male C.B-17 scid mice
on day −21. On days −7 to −5, a jugular vein catheter was surgically
implanted into each mouse. Beginning on day 1, the animals (eight for
vehicle and nine for drug treatment) received daily infusions of vehicle
(10% 2-hydroxypropyl-β-cyclodextrin dissolved in 50 mM Na2HPO4/
1
hexane) to give 1.72 g (86%) of carbamate as a fluffy solid. H NMR
(400 MHz, CDCl3) δ 7.93 (d, J = 8.2 Hz, 1H), 6.27 (d, J = 15.8 Hz,
1H), 5.91−5.84 (m, 1H), 5.84−5.78 (m, 2H), 5.58 (dd, J = 15.7, 6.6
Hz, 1H), 5.50 (t, J = 7.2 Hz, 1H), 4.58 (t, J = 5.3 Hz, 1H), 4.45 (ddd, J
= 11.0, 6.6, 1.8 Hz, 1H), 4.06−3.87 (m, 5H), 2.89 (s, 6H), 2.58−2.54
(m, 2H), 2.54−2.47 (m, 2H), 2.03−1.86 (m, 2H), 1.73 (d, J = 0.9 Hz,
3H), 1.41−1.35 (m, 6H), 1.27 (s, 3H), 1.24−1.11 (m, 2H). 13C NMR
(101 MHz, CDCl3) δ 165.23, 156.30, 141.45, 135.70, 135.48, 127.86,
125.92, 123.67, 102.04, 73.04, 70.15, 70.12, 69.50, 69.39, 55.66, 51.05,
43.80, 42.45, 38.59, 36.31, 35.90, 34.20, 31.54, 23.77, 20.44, 12.59.
HRMS (ESI) m/z calcd for C26H40N2O7Na (M + Na)+, 515.2733;
found, 515.2734.
10040
dx.doi.org/10.1021/jm401370h | J. Med. Chem. 2013, 56, 10033−10044

Journal of Medicinal Chemistry
Article
(S,Z)-5-(((1R,4R)-4-((2E,4E)-5-((3R,5S)-7,7-Dimethyl-1,6-
dioxaspiro[2.5]octan-5-yl)-3-methylpenta-2,4-dien-1-yl)cyclohexyl)-
amino)-5-oxopent-3-en-2-yl Dimethylcarbamate (2). A solution of
activated carbonate 17 (1g, 1.7 mmol) in dichloroethane (15 mL) at 0
°C was treated with dimethylamine (2.57 mL, 5.15 mmol, 2 M in
THF) dropwise. The resulting yellow solution was allowed to stir at 0
°C for 10 min, by which time the TLC indicated that all of the starting
material had been consumed. The solids formed (p-nitro phenol) in
the reaction were filtered and washed with 5% methylene chloride in
hexane (2 × 15 mL). The filtrate was concentrated, and the resulting
residue was purified on silica column (10−80% EtOAc in hexane) to
give 755 mg (90%) of dimethyl carbamate derivative 2 as a fluffy solid.
1H NMR (400 MHz, CDCl3) δ 8.23 (d, J = 8.0 Hz, 1H), 6.27 (d, J =
15.6 Hz, 1H), 5.82 (d, J = 11.4 Hz, 1H), 5.64−5.46 (m, 4H), 4.51−
4.39 (m, 1H), 4.18−4.06 (m, 1H), 2.91 (s, 6H), 2.57 (s, 2H), 2.14−
2.04 (td, J = 7.0, 4.6 Hz, 2H), 2.04−1.85 (m, 3H), 1.75−1.65 (m, 2H),
1.71 (s, 3H), 1.63−1.51 (m, 4H), 1.42−1.32 (s, 2H), 1.39 (s, 3H),
1.34 (d, J = 6.1 Hz, 3H), 1.28 (s, 3H), 1.24−1.10 (m, 2H). 13C NMR
(101 MHz, CDCl3) δ 165.15, 156.45, 136.36, 136.27, 133.60, 132.30,
126.85, 126.27, 73.02, 69.61, 69.53, 55.67, 51.03, 45.24, 42.45, 38.63,
36.66, 36.26, 35.89, 34.55, 31.53, 29.64, 29.50, 27.76, 27.43, 23.76,
20.66, 12.44. IR (neat film) 3267, 2874, 2804, 1656, 1635, 1511, 1370,
1172, 1038 cm−1. HRMS (ESI) m/z calcd for C28H45N2O5 (M + H)+,
489.3323; found, 489.3334.
HRMS (ESI) m/z calcd for C16H20NO4S2 (M + H)+, 354.0834; found,
354.0843.
2-((((3R,5S)-7,7-Dimethyl-1,6-dioxaspiro[2.5]octan-5-yl)methyl)-
thio)benzo[d]thiazole (9). A solution of alcohol 8 (19.5 g, 113 mmol)
in THF (750 mL) was treated with Ph3P (32.7 g, 6.31 mmol) and 2-
mecapto benzothiazole (20.5 g, 119 mmol) and allowed to stir at rt.
After 15 min, the reaction solution was cooled to 0 °C as a solution of
diisopropyl azodicaboxylate (26 mL, 125 mmol) in toluene (70 mL)
was added dropwise. The resulting yellow suspension was allowed to
stir for 1 h at 0 °C. The reaction mixture was then concentrated, and
the residue was triturated with 10% EtOAc in hexane (500 mL). The
solids were filtered and washed with 7.5% EtOAc in hexane (3 × 100
mL). The filtrate was then evaporated, and the resulting residue was
purified by flash chromatography (750 g, 5−30% EtOAc in hexane) to
give 27.6 g (76%) of sulfide derivative 9 as a viscous oil. [α]2D4 +49.9°
1
(c 1.3, CHCl3). H NMR (400 MHz, CDCl3) δ 7.87−7.82 (m, 1H),
7.76−7.72 (m, 1H), 7.44−7.37 (m, 1H), 7.32−7.27 (m, 1H), 4.35−
4.24 (m, 1H), 3.54 (ddd, J = 19.8, 13.2, 5.8 Hz, 2H), 2.57 (q, J = 4.6
Hz, 2H), 2.03−1.92 (m, 2H), 1.42−1.34 (m, 4H), 1.25 (s, 3H), 1.16
(dd, J = 13.9, 1.9 Hz, 1H). 13C NMR (101 MHz, CDCl3) δ 167.14,
153.17, 135.33, 125.98, 124.16, 121.46, 120.95, 73.58, 67.46, 55.56,
50.94, 42.42, 38.76, 37.04, 31.26, 23.72. IR (neat film) 2973, 2910,
1459, 1428, 1252, 1090, 1055, 1020 cm−1. HRMS (ESI) m/z calcd for
C16H20NO2S2 (M + H)+, 322.0935; found, 322.0930.
2-((1S,4S)-4-Azidocyclohexyl)ethanol (11). A stirred solution of
alcohol 10 (10 g, 41.1 mmol) in methylene chloride (160 mL) was
cooled in an ice bath as trifluoroacetic acid (35.1 mL, 431 mmol) was
slowly added. The resulting solution was allowed to warm to rt and
stirred for an additional 4 h. The solvent was then evaporated under
reduced pressure, and the resulting residue was dissolved in MeOH
(200 mL), cooled in an ice bath, and then treated with solid K2CO3
(17.0 g, 123 mmol). This mixture was allowed to stir for 1 h, filtered
through a sintered funnel, and washed with methylene chloride (2 ×
50 mL). The combined filtrates were concentrated to give the free
amine (5.8 g) as an oil. The crude amine was used in the next step
without further purification.
(S,Z)-5-(((1R,4R)-4-((2E,4E)-5-((3R,5S)-7,7-Dimethyl-1,6-
dioxaspiro[2.5]octan-5-yl)-3-methylpenta-2,4-dien-1-yl)cyclohexyl)-
amino)-5-oxopent-3-en-2-yl Methylcarbamate (3). A stirred sol-
ution of activated carbonate 17 (100 mg, 0.17 mmol) in 1,2-
dichloroethane (0.3 mL) was cooled in an ice bath and slowly treated
with methylamine (77 μL, 0.51 mmol, 2 M in THF). The resulting
yellow solution was stirred at 0 °C for 10 min. The solid precipitate (p-
nitro phenol) was removed by filtration, and the residue was washed
with a mixture of methylene chloride and hexane. The combined
filtrates were concentrated, and the residue was purified on a silica
column (25g, 15−100% EtOAc in hexane) to give 75 mg (92%) of
1
methyl carbamate 3 as a white viscous solid. H NMR (400 MHz,
CDCl3) δ 7.94 (d, J = 7.7 Hz, 1H), 6.26 (d, J = 15.7 Hz, 1H), 5.82 (d,
J = 10.8 Hz, 1H), 5.66−5.44 (m, 4H), 4.77 (s, 1H), 4.52−4.39 (m,
1H), 4.18−4.06 (m, 1H), 2.78 (d, J = 5.0 Hz, 3H), 2.57 (s, 2H), 2.09
(t, J = 7.1 Hz, 2H), 2.05−1.87 (m, 2H), 1.80−1.67 (m, 2H), 1.72 (s,
3H), 1.65−1.49 (m, 4H), 1.48−1.41 (m, 1H), 1.40 (s, 3H), 1.38−1.34
(m, 2H), 1.31 (d, J = 5.8 Hz, 3H), 1.28 (s, 3H), 1.24−1.12 (m, 2H).
13C NMR (101 MHz, CDCl3) δ 165.17, 157.03, 136.66, 136.16,
133.70, 132.10, 126.80, 126.06, 73.06, 69.52, 69.15, 55.68, 51.04,
45.36, 42.46, 38.57, 36.62, 34.42, 31.52, 29.57, 29.44, 27.68, 27.39,
27.33, 23.76, 20.60, 12.44. HRMS (ESI) m/z calcd for C27H43N2O5
(M + H)+, 475.3172; found, 475.3188.
A stirred solution of amine derivative (5.8 g, 40.5 mmol) in MeOH
(225 mL) was sequentially treated with K2CO3 (12.5 g, 91 mmol) and
CuSO4·5H2O (0.1 g, 0.4 mmol). The mixture was cooled in an ice
bath and 1H-imidazole-1-sulfonylazide HCl [CAUTION: Subsequent to
our development of this procedure, this reagent (as the HCl salt) was
shown to be an explosion hazard, and safe alternatives have been
discovered, which include the hydrogen sulfate salt. Because of this, we do
not recommend that the HCl salt be used in the future]41,49 (9.4 g, 54.7
mmol) was added portionwise to the resulting suspension. The
reaction was warmed to rt and stirred overnight. The reaction mixture
was then concentrated under vacuum and diluted with water (100
mL). The product was extracted with EtOAc (2 × 250 mL), and the
combined organic layers were dried over Na2SO4. This was
concentrated, and the resulting residue was purified on a silica column
(160 g, 6−40% acetone in hexane) to give 5 g (73%) of the azide
2-((((3R,5S)-7,7-Dimethyl-1,6-dioxaspiro[2.5]octan-5-yl)methyl)-
sulfonyl)benzo[d]thiazole (5). A stirred solution of sulfide 9 (29 g, 90
mmol) in EtOH (700 mL) was cooled in an ice bath and treated with
a mixture of ammonium molybdate·4H2O (33.4 g, 27 mmol) and 30%
H2O2 (123 mL, 1.08 mol) that had been adjusted to pH 4 to 5 with
7.0 phosphate buffer (49 mL, 0.26 M) at 0 °C before the addition. The
resulting suspension was allowed to stir for 14 h at rt. The reaction
solution was divided into two portions for further workup. Each
portion was diluted with water (400 mL), and CH2Cl2 (400 mL) and
the aqueous layer were extracted with CH2Cl2 (2 × 300 mL) and
washed with brine (1 × 300 mL). The two organic portions were
combined and dried over Na2SO4. The solvent was evaporated under
reduced pressure to give the crude product as a white viscous solid.
This was crystallized using methylene chloride and hexane to give 29 g
(91%) of pure sulfone 5 as a white solid. mp 124−126 °C. [α]2D4 +4.5°
1
derivative 11 as a colorless oil. H NMR (400 MHz, CDCl3) δ 3.84−
3.74 (m, 1H), 3.69 (t, J = 6.1 Hz, 2H), 1.88−1.72 (m, 2H), 1.66−1.43
(m, 8H), 1.40−1.18 (m, 2H). 13C NMR (101 MHz, CDCl3) δ 60.57,
57.92, 39.01, 32.67, 29.14, 27.45. IR (neat film) 3274, 2872, 2802,
2061, 1419, 1236, 1032 cm−1. HRMS (ESI) m/z calcd for C8H16N3O
(M + H)+, 170.1288; found, 170.1281.
(E)-Ethyl 4-((1S,4S)-4-Azidocyclohexyl)-2-methylbut-2-enoate
(12). A cold (−78 °C) stirred solution of oxalyl chloride (20.6 mL,
2 M in methylene chloride, 41.4 mmol) in anhydrous methylene
chloride (80 mL) was treated with DMSO (4.2 mL, 59.1 mmol)
dropwise. The resulting mixture was stirred for 15 min at −78 °C and
treated with a solution of alcohol 11 (5 g, 29.5 mmol) in methylene
chloride (60 mL). The resulting slurry was allowed to stir at −78 °C
for 60 min and then treated with diisopropylethylamine (26.3 mL, 148
mmol). The cooling bath was removed, and the reaction mixture was
allowed to stir at rt for 1 h to afford a yellow solution. This solution
was then cooled in an ice bath and then treated with ethyl 2-
(triphenylphosphoranylidene)propanoate (16.0 g, 44.3 mmol) with
stirring for 10 min. The resulting stirred suspension was then allowed
1
(c 1.1, CHCl3). H NMR (400 MHz, CDCl3) δ 8.25−8.17 (m, 1H),
8.04 −7.97 (m, 1H), 7.67−7.52 (m, 2H), 4.62−4.54 (m, 1H), 3.91
(dd, J = 14.8, 9.2 Hz, 1H), 3.43 (dd, J = 14.7, 2.8 Hz, 1H), 2.53 (dd, J
= 17.9, 4.5 Hz, 2H), 1.96−1.74 (m, 2H), 1.25 (s, 3H), 1.23−1.20 (m,
1H), 1.01 (dd, J = 14.0, 1.8 Hz, 1H), 0.59 (s, 3H). 13C NMR (101
MHz, CDCl3) δ 166.96, 152.69, 136.85, 127.80, 127.43, 125.40,
122.16, 73.53, 64.03, 60.14, 54.92, 50.80, 42.02, 37.02, 30.27, 22.90. IR
(neat film) 2973, 2916, 1475, 1423, 1325, 1147, 1088, 1027 cm−1.
10041
dx.doi.org/10.1021/jm401370h | J. Med. Chem. 2013, 56, 10033−10044

Journal of Medicinal Chemistry
Article
to warm to rt and allowed to stir overnight. Water (100 mL) and ether
(300 mL) were then added to the reaction mixture. The organic layer
was separated and washed sequentially with water (100 mL) and brine
(50 mL) and dried over Na2SO4. This was concentrated, and the
resulting residue was purified on a silica column (160 g, 2−20% EtOAc
in hexane) to give 7 g (94%) of conjugated ester 12 as an oil. 1H NMR
(400 MHz, CDCl3) δ 6.75 (tq, J = 7.7, 1.5 Hz, 1H), 4.29−4.14 (q, J =
7.1 Hz, 2H), 3.86−3.72 (m, 1H), 2.11 (t, J = 7.5 Hz, 2H), 1.87−1.76
(m, 5H), 1.62−1.48 (m, 5H), 1.35 −1.30 (m, 2H), 1.31 (t, J = 7.1 Hz
3H). 13C NMR (101 MHz, CDCl3) δ 168.19, 140.43, 128.69, 60.45,
57.70, 36.43, 35.39, 29.22, 27.40, 14.30, 12.54. IR (neat film) 2874,
2804, 2061, 1677, 1236 cm−1. HRMS (ESI) m/z calcd for C13H20N3O2
(M − H)+, 250.1707; found, 250.1716.
(E)-4-((1S,4S)-4-Azidocyclohexyl)-2-methylbut-2-en-1-ol (13). A
solution of diisobutylaluminum hydride (66 mL, 1.0 M in hexanes)
was added dropwise to a cold (−78 °C) stirred solution of ester 12 in
methylene chloride (150 mL). The resulting mixture was allowed to
stir at −78 °C for 2.5 h. The reaction was quenched with MeOH (9
mL) and diluted with saturated sodium potassium tartrate tetrahydrate
(150 mL) and EtOAc (160 mL). The resulting mixture was then
allowed to stir vigorously for 2 h at rt to obtain a clear solution. The
aqueous layer was extracted with EtOAc (2 × 200 mL), and the
combined organic layers were dried over Na2SO4. This was
concentrated, and the residue was purified on a silica column (160
g, 5−40% EtOAc in hexane) to give 5.3 g (91%) of pure alcohol 13 as
an oil. 1H NMR (400 MHz, CDCl3) δ 5.41 (t, J = 7.2 Hz, 1H), 4.01 (s,
2H), 3.83−3.73 (m, 1H), 1.97 (t, J = 6.9 Hz, 2H), 1.85−1.75 (m, 4.3
Hz, 2H), 1.66 (s, 3H), 1.60−1.48 (m, 2H), 1.41−1.23 (m, 4H). 13C
NMR (101 MHz, CDCl3) δ 135.69, 124.36, 68.96, 57.95, 36.85, 29.25,
27.32, 13.85. IR (neat film) 3275, 2870, 2801, 2060, 1418, 1236, 993
cm−1. HRMS (ESI) m/z calcd for C11H19N3ONa (M + Na)+,
232.1426; found, 232.1423.
3H), 1.30−1.24 (m, 2H), 1.27 (s, 3H), 1.25−1.12 (m, 2H). 13C NMR
(101 MHz, CDCl3) δ 136.19, 133.93, 131.74, 127.02, 73.01, 69.59,
57.94, 55.67, 51.03, 42.45, 38.62, 37.03, 34.88, 31.53, 29.24, 27.37,
27.32, 23.76, 12.50. IR (neat film) 2916, 2872, 2802, 2060, 1418, 1253,
956 cm−1. HRMS (ESI) m/z calcd for C20H32N3O2 (M + H)+,
346.2489; found, 346.2492.
ASSOCIATED CONTENT
* Supporting Information
■
S
Full experimental details for the in vitro and in vivo biological
methods, all details for the synthesis of all new compounds, and
details for the single-crystal X-ray structure of compound 5.
This material is available free of charge via the Internet at
http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
*E-mail: thomasr.webb@gmail.com.
■
Present Address
∥SRI International, 333 Ravenswood Avenue, Menlo Park,
California 94025-3493, United States. E-mail: thomas.webb@
sri.com. (Address as of January 2014.)
Author Contributions
§These authors contributed equally to this work
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
(E)-4-((1S,4S)-4-Azidocyclohexyl)-2-methylbut-2-enal (14). A sol-
ution of alcohol 13 (3.3 g, 15.7 mmol) and pyridine (12 mL) in
anhydrous methylene chloride (80 mL) at 0 °C was treated with
Dess−Martin periodinate (12.4 g, 23.6 mmol) portionwise. The
resulting solution was stirred for 2 h at rt and added to a 1:1 mixture
(300 mL) of a saturated aqueous NaHCO3 solution and saturated
aqueous sodium thiosulfate solution. This mixture was allowed to stir
for 30 min at rt. The aqueous layer was separated and extracted with
EtOAc (2 × 400 mL), and the combined organic fractions were dried
over Na2SO4. This was concentrated, and the resulting residue purified
on silica column (2−20% EtOAc in hexane) to give 3.0 g (92%) of
We thank Peter Vogel, DVM, Ph.D. for the histopathologic
evaluation of all tissues of animals subjected to infusions with
sudemycin D6 during the MTD study and Christopher L.
Morton for the generation of the xenograft model and the
tumor-bearing animals used in this study. We also thank John
Bollinger, Ph.D. for his determination of the high-resolution
single-crystal X-ray structure of compound 5 and Christopher
Calabrese, Ph.D. for his help in the establishment of the
infusion model in mice. We thank Lei Yang of the High
Throughput Analytical Chemistry Center for the compound
plasma stability and compound solubility results. We also thank
Alan Pourpak, Xiaoli Cui, and Dr. Steve Morris for the
cytotoxicity data for the derivatives of compound 19. This work
was supported in part by NIH grant CA140474 and by the
American Lebanese Syrian Associated Charities and St. Jude
Children’s Research Hospital.
1
aldehyde 14 as an oil. H NMR (400 MHz, CDCl3) δ 9.41 (s, 1H),
6.50 (td, J = 7.6, 1.0 Hz, 1H), 3.88−3.78 (m, 1H), 2.30 (t, J = 6.9 Hz,
2H), 1.89−1.79 (m, 2H), 1.75 (d, J = 0.4 Hz, 3H), 1.64−1.51 (m,
5H), 1.44−1.29 (dt, J = 14.5, 11.2 Hz, 2H). 13C NMR (101 MHz,
CDCl3) δ 195.16, 152.83, 140.31, 57.50, 36.46, 35.72, 29.22, 27.33,
9.41. IR (neat film) 2874, 2803, 2061, 1659, 1418, 1236 cm−1. HRMS
(ESI) m/z calcd for C11H18N3O (M + H)+, 208.1445; found,
208.1455.
ABBREVIATIONS USED
(3R,7S)-7-((1E,3E)-5-((1S,4R)-4-Azidocyclohexyl)-3-methylpenta-
1,3-dien-1-yl)-5,5-dimethyl-1,6-dioxaspiro[2.5]octane (15). A mix-
ture of sulfone 5 (4.7 g, 13.4 mmol) and aldehyde 14 (2.4 g, 11.8
mmol) was concentrated with benzene (2 × 25 mL) under reduced
pressure, dissolved in anhydrous THF (80 mL), and cooled to −78
°C. This stirred solution was treated dropwise with NaHMDS (14.7
mL, 1 M in THF, 14.7 mmol) over 10 min. The resulting yellow
suspension was stirred at −78 °C for 1 h, allowed to warm in an ice
bath for 15 min, and then allowed to warm to rt and stirred for 0.5 h.
The reaction mixture was quenched with pH 7 buffer solution (120
mL) and extracted with Et2O (3 × 150 mL). The organic layer was
dried (Na2SO4), filtered, and concentrated in vacuo. The solids were
removed by filtration after the addition of 10% CH2Cl2 in hexane. The
resulting residue was purified on silica column (100 g, 2−20% EtOAc
in hexane) to give 3.3 g (81%) of diene 15 (E/Z ratio 95:5) as an oil.
1H NMR (400 MHz, CDCl3) δ 6.26 (d, J = 15.5 Hz, 1H), 5.54 (dd, J =
■
FR, FR901464; PD, pladienolide; HBTU, N,N,N′,N′-tetra-
methyl-O-(1H-benzotriazol-1-yl)uronium hexafluorophos-
phate; NaHMDS, sodium bis(trimethylsilyl)amide; MeOH,
methanol; EtOAc, ethyl acetate; AcOH, acetic acid; CI,
combination index; CL, plasma clearance value; Cmax, maximum
plasma concentration; EH, epoxide hydrolase; FBS, fetal bovine
serum; JVC, jugular vein catheter; LLOQ, lower limit of
quantification; MLEM, maximum likehood via the expectation
maximization algorithm; RSE, relative standard error; SEM,
standard error of the mean; snRNPs, small nuclear ribonucleic
particles; Tmax, time to reach Cmax; Tuvol, tumor volume
REFERENCES
■
15.7, 6.7 Hz, 1H), 5.46 (t, J = 7.7 Hz, 1H), 4.51−4.39 (m, 1H), 3.82−
3.72 (m, 1H), 2.57 (s, 2H), 2.10−2.02 (m, 2H), 2.02−1.86 (m, 2H),
1.84−1.73 (m, 3H), 1.71 (s, 3H), 1.59−1.46 (m, 4H), 1.42−1.38 (s,
(1) Kramer, A. The structure and function of proteins involved in
mammalian pre-mRNA splicing. Annu. Rev. Biochem. 1996, 65, 367−
409.
10042
dx.doi.org/10.1021/jm401370h | J. Med. Chem. 2013, 56, 10033−10044

Journal of Medicinal Chemistry
Article
(2) Golas, M. M.; Sander, B.; Will, C. L.; Luhrmann, R.; Stark, H.
Molecular architecture of the multiprotein splicing factor SF3b. Science
2003, 300, 980−984.
(3) Staley, J. P.; Guthrie, C. Mechanical devices of the spliceosome:
Motors, clocks, springs, and things. Cell 1998, 92, 315−326.
(4) Germann, S.; Gratadou, L.; Dutertre, M.; Auboeuf, D. Splicing
programs and cancer. J. Nucleic Acids 2012, 2012, 269570-1−269570-
9.
(22) Damm, F.; Nguyen-Khac, F.; Fontenay, M.; Bernard, O. A.
Spliceosome and other novel mutations in chronic lymphocytic
leukemia and myeloid malignancies. Leukemia 2012, 26, 2027−2031.
(23) Murati, A.; Brecqueville, M.; Devillier, R.; Mozziconacci, M. J.;
Gelsi-Boyer, V.; Birnbaum, D. Myeloid malignancies: Mutations,
models and management. BMC Cancer 2012, 12, 304.
(24) Harbour, J. W.; Roberson, E. D.; Anbunathan, H.; Onken, M.
D.; Worley, L. A.; Bowcock, A. M. Recurrent mutations at codon 625
of the splicing factor SF3B1 in uveal melanoma. Nat. Genet. 2013, 45,
133−135.
(25) Hubert, C. G.; Bradley, R. K.; Ding, Y.; Toledo, C. M.; Herman,
J.; Skutt-Kakaria, K.; Girard, E. J.; Davison, J.; Berndt, J.; Corrin, P.;
Hardcastle, J.; Basom, R.; Delrow, J. J.; Webb, T.; Pollard, S. M.; Lee,
J.; Olson, J. M.; Paddison, P. J. Genome-wide RNAi screens in human
brain tumor isolates reveal a novel viability requirement for PHF5A.
Genes Dev. 2013, 27, 1032−1045.
(26) Gao, Y.; Koide, K. Chemical perturbation of Mcl-1 pre-mRNA
splicing to induce apoptosis in cancer cells. ACS Chem. Biol. 2013, 8,
895−900.
(27) Osman, S.; Albert, B. J.; Wang, Y.; Li, M.; Czaicki, N. L.; Koide,
K. Structural requirements for the antiproliferative activity of pre-
mRNA splicing inhibitor FR901464. Chem.Eur. J. 2011, 17, 895−
904.
(28) Thompson, C. F.; Jamison, T. F.; Jacobsen, E. N. Convergent
assembly of chiral components prepared by asymmetric catalysis. J.
Am. Chem. Soc. 2000, 122, 10482−10483.
(29) Horigome, M.; Motoyoshi, H.; Watanabe, H.; Kitahara, T. A
synthesis of FR901464. Tetrahedron Lett. 2001, 42, 8207−8210.
(30) Thompson, C. F.; Jamison, T. F.; Jacobsen, E. N. FR901464:
Total synthesis, proof of structure, and evaluation of synthetic
analogues. J. Am. Chem. Soc. 2001, 123, 9974−9983.
(31) Albert, B. J.; Sivaramakrishnan, A.; Naka, T.; Koide, K. Total
synthesis of FR901464, an antitumor agent that regulates the
transcription of oncogenes and tumor suppressor genes. J. Am.
Chem. Soc. 2006, 128, 2792−2793.
(5) Nakajima, H.; Hori, Y.; Terano, H.; Okuhara, M.; Manda, T.;
Matsumoto, S.; Shimomura, K. New antitumor substances, FR901463,
FR901464 and FR901465. II. Activities against experimental tumors in
mice and mechanism of action. J. Antibiot. 1996, 49, 1204−1211.
(6) Nakajima, H.; Sato, B.; Fujita, T.; Takase, S.; Terano, H.;
Okuhara, M. New antitumor substances, FR901463, FR901464 and
FR901465. I. Taxonomy, fermentation, isolation, physico-chemical
properties and biological activities. J. Antibiot. 1996, 49, 1196−1203.
(7) Nakajima, H.; Takase, S.; Terano, H.; Tanaka, H. New antitumor
substances, FR901463, FR901464 and FR901465. III. Structures of
FR901463, FR901464 and FR901465. J. Antibiot. 1997, 50, 96−99.
(8) Mizui, Y.; Sakai, T.; Iwata, M.; Uenaka, T.; Okamoto, K.;
Shimizu, H.; Yamori, T.; Yoshimatsu, K.; Asada, M. Pladienolides, new
substances from culture of Streptomyces platensis Mer-11107. III. In
vitro and in vivo antitumor activities. J. Antibiot. 2004, 57, 188−196.
(9) Asai, N.; Kotake, Y.; Niijima, J.; Fukuda, Y.; Uehara, T.; Sakai, T.
Stereochemistry of pladienolide B. J. Antibiot. 2007, 60, 364−369.
(10) Kaida, D.; Motoyoshi, H.; Tashiro, E.; Nojima, T.; Hagiwara,
M.; Ishigami, K.; Watanabe, H.; Kitahara, T.; Yoshida, T.; Nakajima,
H.; Tani, T.; Horinouchi, S.; Yoshida, M. Spliceostatin A targets SF3b
and inhibits both splicing and nuclear retention of pre-mRNA. Nat.
Chem. Biol. 2007, 3, 576−583.
(11) Kotake, Y.; Sagane, K.; Owa, T.; Mimori-Kiyosue, Y.; Shimizu,
H.; Uesugi, M.; Ishihama, Y.; Iwata, M.; Mizui, Y. Splicing factor SF3b
as a target of the antitumor natural product pladienolide. Nat. Chem.
Biol. 2007, 3, 570−575.
(12) Hasegawa, M.; Miura, T.; Kuzuya, K.; Inoue, A.; Won, K. S.;
Horinouchi, S.; Yoshida, T.; Kunoh, T.; Koseki, K.; Mino, K.; Sasaki,
R.; Yoshida, M.; Mizukami, T. Identification of SAP155 as the target of
GEX1A (Herboxidiene), an antitumor natural product. ACS Chem.
Biol. 2011, 6, 229−233.
(32) Albert, B. J.; Sivaramakrishnan, A.; Naka, T.; Czaicki, N. L.;
Koide, K. Total syntheses, fragmentation studies, and antitumor/
antiproliferative activities of FR901464 and its low picomolar
analogue. J. Am. Chem. Soc. 2007, 129, 2648−2659.
(13) Liu, X.; Biswas, S.; Berg, M. G.; Antapli, C. M.; Xie, F.; Wang,
Q.; Tang, M. C.; Tang, G. L.; Zhang, L.; Dreyfuss, G.; Cheng, Y. Q.
Genomics-guided discovery of thailanstatins A, B, and C as pre-mRNA
splicing inhibitors and antiproliferative agents from Burkholderia
thailandensis MSMB43. J. Nat. Prod. 2013, 76, 685−693.
(14) Mandel, A. L.; Jones, B. D.; La Clair, J. J.; Burkart, M. D. A
synthetic entry to pladienolide B and FD-895. Bioorg. Med. Chem. Lett.
2007, 17, 5159−5164.
(15) Villa, R.; Mandel, A. L.; Jones, B. D.; La Clair, J. J.; Burkart, M.
D. Structure of FD-895 revealed through total synthesis. Org. Lett.
2012, 14, 5396−5399.
(16) Sakai, Y.; Tsujita, T.; Akiyama, T.; Yoshida, T.; Mizukami, T.;
Akinaga, S.; Horinouchi, S.; Yoshida, M. GEX1 compounds, novel
antitumor antibiotics related to herboxidiene, produced by Strepto-
myces sp. II. The effects on cell cycle progression and gene expression.
J. Antibiot. 2002, 55, 863−872.
(17) Webb, T. R.; Joyner, A. S.; Potter, P. M. The development and
application of small molecule modulators of SF3b as therapeutic
agents for cancer. Drug Discovery Today 2013, 18, 43−49.
(18) Venables, J. P. Aberrant and alternative splicing in cancer.
Cancer Res. 2004, 64, 7647−7654.
(19) Kalnina, Z.; Zayakin, P.; Silina, K.; Line, A. Alterations of pre-
mRNA splicing in cancer. Genes, Chromosomes Cancer 2005, 42, 342−
357.
(33) Lagisetti, C.; Pourpak, A.; Jiang, Q.; Cui, X.; Goronga, T.;
Morris, S. W.; Webb, T. R. Antitumor compounds based on a natural
product consensus pharmacophore. J. Med. Chem. 2008, 51, 6220−
6224.
(34) Lagisetti, C.; Pourpak, A.; Goronga, T.; Jiang, Q.; Cui, X.; Hyle,
J.; Lahti, J. M.; Morris, S. W.; Webb, T. R. Synthetic mRNA splicing
modulator compounds with in vivo antitumor activity. J. Med. Chem.
2009, 52, 6979−6990.
(35) Fan, L.; Lagisetti, C.; Edwards, C. C.; Webb, T. R.; Potter, P. M.
Sudemycins, novel small molecule analogues of FR901464, induce
alternative gene splicing. ACS Chem. Biol. 2011, 6, 582−589.
(36) Gundluru, M. K.; Pourpak, A.; Cui, X.; Morris, S. W.; Webb, T.
R. Design, synthesis and initial biological evaluation of a novel
pladienolide analog scaffold. MedChemCommun 2011, 2, 904−908.
(37) Jennings, L.; Murphy, G. M. Predicting outcome in melanoma:
where are we now? Br. J. Dermatol. 2009, 161, 496−503.
(38) Smith, A. B., 3rd.; Dong, S.; Brenneman, J. B.; Fox, R. J. Total
synthesis of (+)-sorangicin A. J. Am. Chem. Soc. 2009, 131, 12109−
12111.
(39) Lawrence, N. J.; Drew, M. D.; Bushell, S. M. Polymethylhy-
drosiloxane: A versatile reducing agent for organic synthesis. J. Chem.
Soc., Perkin Trans. 1 1999, 0, 3381−3391.
(40) Goddard-Borger, E. D.; Stick, R. V. An efficient, inexpensive,
and shelf-stable diazotransfer reagent: Imidazole-1-sulfonyl azide
hydrochloride. Org. Lett. 2007, 9, 3797−3800.
(41) Ye, H.; Liu, R.; Li, D.; Liu, Y.; Yuan, H.; Guo, W.; Zhou, L.;
Cao, X.; Tian, H.; Shen, J.; Wang, P. G. A safe and facile route to
imidazole-1-sulfonyl azide as a diazotransfer reagent. Org. Lett. 2012,
15, 18−21.
(20) Bonnal, S.; Vigevani, L.; Valcarcel, J. The spliceosome as a target
of novel antitumour drugs. Nat. Rev. Drug Discovery 2012, 11, 847−
859.
(21) Ogawa, S. Splicing factor mutations in myelodysplasia. Int. J.
Hematol. 2012, 96, 438−442.
10043
dx.doi.org/10.1021/jm401370h | J. Med. Chem. 2013, 56, 10033−10044

Journal of Medicinal Chemistry
Article
(42) Lehar, J.; Krueger, A. S.; Avery, W.; Heilbut, A. M.; Johansen, L.
M.; Price, E. R.; Rickles, R. J.; Short, G. F., 3rd.; Staunton, J. E.; Jin, X.;
Lee, M. S.; Zimmermann, G. R.; Borisy, A. A. Synergistic drug
combinations tend to improve therapeutically relevant selectivity. Nat.
Biotechnol. 2009, 27, 659−666.
(43) Maira, S. M.; Stauffer, F.; Brueggen, J.; Furet, P.; Schnell, C.;
Fritsch, C.; Brachmann, S.; Chene, P.; De Pover, A.; Schoemaker, K.;
Fabbro, D.; Gabriel, D.; Simonen, M.; Murphy, L.; Finan, P.; Sellers,
W.; Garcia-Echeverria, C. Identification and characterization of NVP-
BEZ235, a new orally available dual phosphatidylinositol 3-kinase/
mammalian target of rapamycin inhibitor with potent in vivo
antitumor activity. Mol. Cancer Ther. 2008, 7, 1851−1863.
(44) LoPiccolo, J.; Blumenthal, G. M.; Bernstein, W. B.; Dennis, P. A.
Targeting the PI3K/Akt/mTOR pathway: Effective combinations and
clinical considerations. Drug Resist. Updates 2008, 11, 32−50.
(45) Liu, J. Y.; Park, S. H.; Morisseau, C.; Hwang, S. H.; Hammock,
B. D.; Weiss, R. H. Sorafenib has soluble epoxide hydrolase inhibitory
activity, which contributes to its effect profile in vivo. Mol. Cancer Ther.
2009, 8, 2193−2203.
(46) Decker, M.; Arand, M.; Cronin, A. Mammalian epoxide
hydrolases in xenobiotic metabolism and signalling. Arch. Toxicol.
2009, 83, 297−318.
(47) Chou, T. C.; Talalay, P. Quantitative analysis of dose-effect
relationships: The combined effects of multiple drugs or enzyme
inhibitors. Adv. Enzyme Regul. 1984, 22, 27−55.
(48) Laizure, S. C.; Herring, V.; Hu, Z.; Witbrodt, K.; Parker, R. B.
The role of human carboxylesterases in drug metabolism: Have we
overlooked their importance? Pharmacotherapy 2013, 33, 210−222.
(49) Fischer, N.; Goddard-Borger, E. D.; Greiner, R.; Klapotke, T.
̈
M.; Skelton, B. W.; Stierstorfer, J. Sensitivities of some imidazole-1-
sulfonyl azide salts. J. Org. Chem. 2012, 77, 1760−1764.
10044
dx.doi.org/10.1021/jm401370h | J. Med. Chem. 2013, 56, 10033−10044