An Orally Bioavailable, Indole-3-glyoxylamide Based Series of Tubulin Polymerization Inhibitors Showing Tumor Growth Inhibition in a Mouse Xenograft Model of Head and Neck Cancer

Article abstract of DOI:10.1021/acs.jmedchem.5b01312

A number of indole-3-glyoxylamides have previously been reported as tubulin polymerization inhibitors, although none has yet been successfully developed clinically. We report here a new series of related compounds, modified according to a strategy of redu

Full text of DOI:10.1021/acs.jmedchem.5b01312

This is an open access article published under a Creative Commons Attribution (CC-BY)
License, which permits unrestricted use, distribution and reproduction in any medium,
provided the author and source are cited.
Article
pubs.acs.org/jmc
An Orally Bioavailable, Indole-3-glyoxylamide Based Series of
Tubulin Polymerization Inhibitors Showing Tumor Growth Inhibition
in a Mouse Xenograft Model of Head and Neck Cancer
Helen E. Colley,*,†,∇ Munitta Muthana,‡,∇ Sarah J. Danson,§ Lucinda V. Jackson,∥ Matthew L. Brett,∥
Joanne Harrison,∥ Sean F. Coole,∥ Daniel P. Mason,∥ Luke R. Jennings,† Melanie Wong,⊥,∇
Vamshi Tulasi,⊥ Dennis Norman,⊥ Peter M. Lockey,⊥ Lynne Williams,‡ Alexander G. Dossetter,#
Edward J. Griffen,#,∇ and Mark J. Thompson*,∥,∇
†School of Clinical Dentistry, University of Sheffield, 19 Claremont Crescent, Sheffield S10 2TA, U.K.
‡Department of Oncology, The University of Sheffield, Medical School, Beech Hill Road, Sheffield S10 2RX, U.K.
§Academic Unit of Clinical Oncology and Sheffield Experimental Medicine Centre, Weston Park Hospital, Whitham Road, Sheffield
S10 2SJ, U.K.
∥Department of Chemistry, University of Sheffield, Brook Hill, Sheffield S3 7HF, U.K.
⊥Charles River, 8−9 Spire Green Centre, Harlow, Harlow, Essex CM19 5TR, U.K.
#MedChemica Limited, Ebenezer House, Ryecroft, Newcastle-Under-Lyme, Staffordshire ST5 2BE, U.K.
S
* Supporting Information
ABSTRACT: A number of indole-3-glyoxylamides have previously been reported as tubulin polymerization inhibitors, although
none has yet been successfully developed clinically. We report here a new series of related compounds, modified according to a
strategy of reducing aromatic ring count and introducing a greater degree of saturation, which retain potent tubulin
polymerization activity but with a distinct SAR from previously documented libraries. A subset of active compounds from the
reported series is shown to interact with tubulin at the colchicine binding site, disrupt the cellular microtubule network, and exert
a cytotoxic effect against multiple cancer cell lines. Two compounds demonstrated significant tumor growth inhibition in a mouse
xenograft model of head and neck cancer, a type of the disease which often proves resistant to chemotherapy, supporting further
development of the current series as potential new therapeutics.
faceted mode of action arising from interaction with micro-
tubules throughout the cell cycle.7,9−11
INTRODUCTION
■
Microtubule-targeting agents (MTAs) are a major class of
cancer chemotherapeutic drugs1,2 that have enjoyed decades of
clinical success in the treatment of a wide range of cancers.
These agents bind to heterodimers of α- and β-tubulin, the
building blocks of cellular microtubules, and in doing so
interfere with normal microtubule dynamics. Depending on
their exact binding site, MTAs can stabilize or destabilize
microtubules, but in either event, disruption of normal function
results and ultimately leads to apoptosis. The precise
mechanism of action of MTAs used in the clinic is still
incompletely understood, and as such continues to be
investigated and debated.3−6 Although it had long been
assumed that MTAs are effective against cancers solely through
interruption of mitosis, on which their effect is certainly
important,7,8 more recent evidence points toward a multi-
Despite the advances in treatment MTAs have undoubtedly
offered, existing agents suffer from some significant drawbacks12
which often compromise clinical efficacy and can lead to
treatment failure. Drug resistance (either innate or acquired) is
a serious problem,13−16 and adverse toxicities such as peripheral
neuropathy17,18 (PN) can prove dose-limiting, indicative of a
narrow therapeutic window.
The search for novel agents in the class has therefore been
intensive19,20 and whereas the more recent approvals, such as
vinflunine,21 eribulin,22−24 ixabepilone,25,26 and the antibody−
drug conjugate trastuzumab emtansine,27 have proven
Received: August 25, 2015
© XXXX American Chemical Society
A
DOI: 10.1021/acs.jmedchem.5b01312
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
beneficial for a limited number of indications, not all clinical
studies have demonstrated improved efficacy, and classical
MTA-associated adverse reactions,28−30 including PN,31−33 are
still encountered with these newer agents. It is worth noting
that eribulin in particular has shown evidence of an improved
safety profile and promise for wider application34 although drug
resistance has also been encountered with this agent.35 Overall,
in spite of some welcome benefits, the newer MTAs have
seemingly failed to displace their older counterparts to a
significant degree, at least to date. Alhough they are expected to
gain broader use in the future, it remains an important goal for
the field to develop novel agents which can demonstrate the
same broad spectrum of efficacy as the more established drugs
while offering significant progress in overcoming resistance and
minimizing unmanageable toxicity.
To this end, many novel small-molecule MTAs have been
reported in recent years,36−39 some with promising preclinical
or early clinical data, although the challenge of demonstrating
superior clinical performance over existing agents is consid-
erable. Nonetheless, such compounds include indibulin 1 and
rosabulin 2 (Figure 1), two experimental drugs with high
structural similarity for which a novel tubulin binding site has
been claimed.40,41
pair analysis (MMPA) is shown to justify such an approach in
the present context.
Tubulin polymerization inhibitors identified during this study
were first applied in models of head and neck cancer (HNC), a
disease in which existing MTAs have shown promise for
improved treatment outcomes if the common problems of
toxicity and drug resistance can be overcome.51−56
HNC is the sixth most common cancer worldwide and is of
particular concern due to rising incidence over the past two
decades. Treatment is often some combination of surgery,
radiotherapy, and chemotherapy, but overall survival rates have
not significantly improved in many years and response rates to
typical first-line chemotherapy regimens, usually a combination
of 5-FU and cisplatin, are problematically low (around
30%).51,57 More recently, targeted therapies have been
introduced58 in an effort to improve treatment of HNC, most
notably the EGFR-targeted monoclonal antibody cetuximab,
which was approved in 2008 for use in locally advanced or
metastatic/recurrent HNC. Other targeted therapies remain
under investigation in HNC, including antiangiogenic agents,
PI3K/Akt/mTor inhibitors, and p53 gene replacement
therapy.58−60 Whereas such targeted approaches have certainly
benefited some patients, their overall impact on HNC
treatment has been mixed,59−62 and personalized or combina-
tion therapy will likely prove necessary to maximize clinical
success of these targeted drugs;60,61 thus, more effective general
first-line chemotherapy treatments are still much needed. Given
the limited but promising data on the potential of MTAs in
HNC, appropriate disease models therefore represent a useful
context for the testing and development of new agents in the
class.
RESULTS AND DISCUSSION
■
Figure 1. Structures of glyoxylamide-based tubulin polymerization
inhibitors that have progressed to clinical trials.
Solubility Prediction Studies. To test our initial
assumptions regarding modifications to 1, directly calculated
solubilities were determined for some of the first-generation
analogues planned initially and detailed in the SAR study below
(5−13, 17, 18, 32, and 33; Tables 1−3). Direct calculation of
aqeuous solubility is known to represent a considerable
challenge, although several methods now exist with accuracy
to well within an order of magnitude for most small
molecules.63 The compounds specified above were evaluated
by the method of Tetko et al.64 using the web-based ALOGPS
2.1 program,65 which estimated at least an order of magnitude
improvement in solubility over 1 in each case (Supporting
Information, Table S11). Given such direct prediction of
solubility is acknowledged as a difficult problem, an alternative
method was applied to further validate our proposed structural
modifications to 1. A matched pairs approach was employed
that has previously shown promise in providing guidance when
seeking to optimize molecular properties.66 The database of
solubility transformations curated by MedChemica67 was used
to give an historical view of how different chemical changes
have influenced measured solubility in order to set reasonable
bounds on what effects should be expected. The chemical
changes were identified using a matched pair finding approach
combining that of Hussein and Rea68 with that of Warner et
al.,69 which captures both the group being changed and
essentially70 the local chemical environment around the change.
The data associated with each chemical change in its local
environment was then recalled from the database and the
recovered values used to construct a solubility network (see
Supporting Information, Figure S16 and accompanying
Little has been published regarding 2, and its development
appears to have been abandoned following phase I clinical
trials.41,42 Compound 1 showed in vivo efficacy and early
evidence of improved therapeutic window,43 suggested to be a
consequence of it targeting a novel binding site on tubulin.40 It
has persisted in phase I clinical development for a number of
years, with the primary issue apparently being low expo-
sure,44,45 in turn likely due to its poor aqueous solubility.
Although both 1 and 2 have encountered problems in the
clinic, the fact remains that they were promising preclinical
agents, and the issues subsequently encountered for 1 do not
seem insurmountable.
In the light of an intriguing report46 suggesting indole-3-
glyoxylamides unsubstituted at N1 can still make effective
tubulin polymerization inhibitors, we supposed that exploring
modifications of 1 and 2 at this locus would be a viable starting
point for the development of new potential MTAs. The initial
focus of the present work was replacement of the p-
chlorobenzyl group of 1, aiming first to improve aqueous
solubility through the dual strategy of reducing aromatic ring
count and increasing saturation. The benefits in terms of
compound developability of both minimizing the number of
aromatic rings47 and increasing the count of sp3 carbon
atoms48−50 are gaining appreciation as a result of several studies
supporting such rationale over recent years. Solubility
prediction by either direct calculation or matched molecular
B
DOI: 10.1021/acs.jmedchem.5b01312
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
legend). Results of the matched pair analysis were in agreement
with those from direct calculation above, predicting at least an
order of magnitude increase in solubility for analogues of 1
replacing the p-chlorobenzyl group with simple aliphatic
substituents. Addition of a methoxy-group to the pyridine
ring, as was ultimately found necessary to recover activity, was
predicted to cause a small compromise in solubility although
these analogues (8−10; Table 1) were still expected to be
much more soluble than 1.
Chemistry. Glyoxylamides derived from N-substituted
indoles 4 were prepared using a straightforward two-step,
one-pot procedure essentially as described previously (Scheme
1a).71 However, for derivatization of intermediates 67
1b) via glyoxylate ester intermediates 67, formed using a
Friedel−Crafts reaction with methyl chlorooxoacetate. This
transformation has been shown effective with the less reactive
azaindole derivatives,72 as also proved the case in our hands,
and was easily adapted for substituted indoles with acceptable
yields. The resulting ester intermediates were then readily
converted into the desired glyoxylamides via direct amidation
catalyzed by 1,5,7-triazabicyclo[4.4.0]dec-5-ene (TBD).73
Some N-substituted indole derivatives not accessible through
direct N-alkylation were also required, and these were
approached by various other methods as appropriate (Scheme
2). 1-Cyclopentylindole 4j was prepared by reaction with
cyclopentylboronic acid using conditions reported for prepara-
Scheme 1. General Synthetic Routes to the Indole-3-
glyoxylamide Screening Libraries, Using Either (a) a One-
Pot Procedure, or (b) a Two-Step Route Suitable for
Scheme 2. Preparation of N-Substituted Indole
Intermediates Accessed via Other Routes
a
a
Modified Indoles
a
Reagents and conditions: (a) NaH, R1-X, DMF or THF, 6−83%; (b)
(i) oxalyl chloride, THF, 3 h, (ii) R2-NH2, DIPEA, DMAP (catalytic),
THF, 18 h, 20−78%; (c) methyl chlorooxoacetate, AlCl3, CH2Cl2, 0
°C, 7−86%; (d) R2-NH2, 30 mol % TBD, toluene, 85 °C, 18 h, 5−
58%. For the sake of clarity, individual structures of 4a−i (Table S8)
and modified indoles 66a−l (Table S9) and their N-alkylated
derivatives 67a−o (Table S10) are provided in the Supporting
Information.
containing modifications at indole positions 4−7, this approach
was surprisingly found to be unsuitable, in contrast to earlier
examples where similar substrates unsubstituted at N1 (i.e., R1
= H) gave the expected products, albeit in reduced yields.71 In
the present study, when compounds of general structure 67
were reacted as outlined in Scheme 1a (with or without heating
to 50 °C in the first step to promote reaction with oxalyl
chloride); no significant formation of the expected indole-3-
glyoxylamide products was observed as evidenced by HPLC or
LC-MS, yet little of the starting indoles remained, suggesting
degradation of these starting materials had occurred. Addition
of DIPEA to the first step of the process, reaction with oxalyl
chloride, to prevent strongly acidic conditions from developing
only appeared to suppress reaction of the indole derivative with
the latter. Thus, an alternative route was employed (Scheme
a
Reagents and conditions: (a) cyclopentylboronic acid, NaHMDS, 10
mol % Cu(OAc)2, DMAP, toluene, 95 °C, 18 h, 34%; (b) isobutene
oxide, NaH, DMF, 24 h, 74%; (c) MeI, NaH, THF, 24 h, 52%; (d)
ethyl 2-bromopropionate, NaH, DMF, 40 °C, 18 h, 74%; (e) LDA,
MeI, THF, −78 °C → rt, 18 h, 86%; (f) LiAlH4, THF, −78 °C → rt,
18 h, 78%; (g) MeI, NaH, THF, 3 h, 79% (4l) or ethyl bromoacetate,
NaH, DMF, 20 h, 29% (4m); (h) methyl acrylate, 50 mol % DBU,
MeCN, 50 °C, 12 h, 43%.
C
DOI: 10.1021/acs.jmedchem.5b01312
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
tion of the cyclopropyl derivative from the requisite boronic
acid.74 Ring-opening of isobutene oxide by indole anion75
followed by O-methylation furnished intermediate 4k. Tertiary
alkyl derivatives 4l and 4m were accessed by O-alkylation of 72,
which in turn was synthesized in three steps from indole 3
utilizing methodology already reported for some substituted
indole starting materials.76 Finally, 4n was prepared by an aza-
Michael reaction between indole and methyl acrylate.77
Glyoxylamide synthesis from these additional intermediates
4j−n was performed according to the general method outlined
above (Scheme 1a).
Scheme 4. Synthesis of Analogues Replacing the Indole
System with an Indolizine Core
a
Variation of the glyoxylamide N-substituent R2 (Scheme 1)
was mostly achieved through the use of commercially available
starting materials, although some such analogues required in-
house synthesis of the amine precursors, R2-NH2 (Scheme 3).
Scheme 3. Preparation of 5-Aminopyridines with Varied 2-
a
Alkoxy Substituents
a
Reagents and conditions: (a) R1CH2CCH, 2 mol % PdCl2(PPh3)2,
4 mol % PPh3, 4 mol % CuI, TEA, 18 h, 63−78%; (b) MeI, NaH,
DMF, 0 °C, 3.5 h, 27%; (c) CuCl, DMA/TEA (7:1), 130 °C, 31−
55%; (d) (i) oxalyl chloride, THF, 0 °C, 3 h, (ii) R2-NH2, DIPEA,
DMAP, THF, 18 h, 20−66%.
a
Reagents and conditions: (a) R5-OH, NaH, iPrOH or DMF or THF,
Primary Screening and Elucidation of SAR. Test
compounds were first assessed for cytotoxicity against FaDu,
an EGFR inhibitor insensitive82 HNC cell line derived from
squamous cell carcinoma of the pharynx. Screens consisted of a
72 h end point MTT assay, and LC50 values were determined
for compounds showing cytotoxic activity at or below 10 μM.
Analogues of 1 replacing the p-chlorobenzyl group at N1 with
various simple alkyl groups showed a loss of cytotoxic activity
(compounds 5−7, Table 1); however, introduction of a
methoxy-group adjacent to the pyridine ring nitrogen rescued
activity (8−10), and switching the positions of these structural
features, as in analogues 11−13, was found to result in a further
order of magnitude improvement in potency.
The foregoing results inspired synthesis of a wider range of
analogues to investigate the SAR with respect to variation of
both R1 and R2. First, systematic changes to the R2 group
revealed some clear trends in activity (Table 2). Removal of the
pyridyl nitrogen (N to CH; 14 and 15) resulted in a modest
compromise to cytotoxic potency of no more than a 3-fold
change, whereas in contrast, the inactivity of 16 indicated the p-
methoxy group is essential for activity. Introduction of a second
ring nitrogen at the meta-position of the aromatic group
(pyrimidines 17 and 18) showed the same modest reduction in
potency as the aniline compounds, whereas introduction of a
second ring nitrogen at the ortho-position (pyridazine 19)
proved highly detrimental to activity. A change of p-methoxy to
ethoxy was tolerated with a 2-fold increase in LC50 (compound
20), whereas introduction of a second methoxy-substituent, as
in 21, was not. Replacement of the p-methoxy group with either
trifluoromethoxy or trifluoromethyl was found severely
detrimental to cytotoxic activity (compare 22 and 23 with 14
and 15, and 24 and 25 with 12 and 13, respectively).
Exchanging the pyridine ring of 12 for a five-membered
heterocycle, in analogous fashion to compound 2, also proved
unproductive. In sum, although the R2 variations explored at
this stage did not lead to improvement in cytotoxic potency in
44−65%; (b) H2, Pd-C, MeOH, 62−95%.
Nucleophilic aromatic substitution of 2-chloro-5-nitropyridine
73 with alkoxides followed by hydrogenolysis of the nitro-
group78 provided easy access to intermediates 77−79 for use in
glyoxylamide syntheses. A small number of analogues
exchanging the indole core for indolizine, as found in model
compound 2, were also evaluated. 3-Alkylindolizines 85−87
were prepared from 2-bromopyridine in two steps by a known
route79,80 comprising Sonogashira cross-coupling followed by a
copper(I)-mediated cyclization. For the most part, this
sequence proceeded smoothly, with the exception of
intermediate 84, whose synthesis proved surprisingly challeng-
ing. Sonogashira reaction between 2-bromopyridine 80 and
pent-4-yn-1-ol readily afforded alcohol 83, but initial attempts
at O-methylation in THF were unsuccessful, giving none of the
desired product and seemingly resulting in intramolecular
reaction between the generated alkoxide and the alkyne (data
not shown). Attempts to convert 83 into 3-(2-hydroxyethyl)-
indolizine resulted in similar mixtures with no formation of the
expected product. Methylation under modified conditions in
DMF at 0 °C did provide a low yield of 84, but its isolation
from the product mixture formed was not straightforward so
these shortcomings inspired an alternative approach whereby
pent-4-yn-1-ol was first methylated to give 5-methoxypent-1-
yne. Because this intermediate is somewhat volatile (lit.81 bp
107−115 °C), the crude material was used without extensive
purification and reacted with 2-bromopyridine under the usual
conditions (Scheme 4). This completed an alternative route to
84, which although still low-yielding, proved operationally
simpler and more amenable to scale than the earlier method.
Once in hand, the 3-alkylindolizine intermediates 84−86 were
successfully elaborated into the desired glyoxylamides by the
same two-step, one-pot procedure used for the majority of the
indole derivatives (Scheme 1a).
D
DOI: 10.1021/acs.jmedchem.5b01312
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Table 1. Cytotoxic Potencies of First-Generation
Compounds in the Present Library
Table 2. Cytotoxicity Data for Second-Generation Analogues
Designed to Interrogate SAR at R2 Position of the Parent
Structure
a
a
a
An expanded form of this Table, including yields, is included in the
b
Supporting Information (Table S1). Reported mean and SEM values
represent the results of at least three independent experiments.
c
Compound 1 was sourced commercially.
a
An expanded form of this Table, including yields, is included in the
b
Supporting Information (Table S2). Reported mean and SEM values
any case, a number of changes were identified that are tolerated
without seriously compromising activity.
represent the results of at least three independent experiments (except
24, n = 2).
In contrast, variation of the R1 group did lead to the sought-
after improvement in cytotoxic potency (Table 3). Changing
the isopropyl group of 12 to sec-butyl slightly improved activity,
although cyclopentyl 29 or isobutyl 30 substituents resulted in
a 4-fold drop-off. Exchanging the oxygen atom in the 2-
methoxyethyl substituent of 13 for a methylene group, as in n-
butyl analogue 31, caused no change in potency, indicating that
the shape, rather than polarity, of this R1 group is the major
determinant of activity. Another change to 13, introducing an
additional methyl group α- to indole N1, resulted in the most
potent compound in the series, 32, with an LC50 against FaDu
of 12 nM. Other analogues of 13, with a ring bridging the side
chain oxygen atom, differed markedly in activity. Tetrahy-
drofuran compound 33 was essentially as potent as its parent,
whereas oxetane analogue 34 was unexpectedly compromised
by more than an order of magnitude. Introducing a tertiary
center β- to N1, as in gem-dimethyl compound 35, was not
tolerated at all. Taken together, cytotoxicity data for this set of
analogues are suggestive of a common binding pocket for the
R1 side chain. Although simple, linear R1 chains give rise to
some of the most potent compounds, a degree of further
substitution is tolerable or even beneficial, whereas too much
substitution severely compromises or abolishes the observed
activity.
The final set of compounds prepared for initial SAR studies
focused on variation at indole positions 4−7 (Table 4). These
analogues 36−51 retained the core structure of compounds 12
or 13 (except for 39, R2 = 4-methoxyphenyl) with systemati-
cally varied modifications at C4 to C7: chloro- or methoxy-
substitution, or a switch to an azaindole (i.e., ring CH to N).
The majority of these changes were not tolerated and abolished
cytotoxic activity, which was particularly unexpected in the case
of azaindoles 47−49 given such a small structural change. For
analogues introducing an additional group, only chloro-
substitution at C4 or C6 gave any retention of activity, albeit
with at least a 20-fold increase in LC50. The single useful
modification identified in this series was a change from indole
to 7-azaindole, with 50 and 51 showing similar potency to their
parent compounds, 12 and 13, respectively.
Evaluation and Early Optimization of in Vitro PK
Properties. A selection of the most promising compounds was
E
DOI: 10.1021/acs.jmedchem.5b01312
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Table 3. Cytotoxicity Data for Second-Generation Analogues
Designed to Interrogate SAR at R1 Position of the Parent
Structure
Table 4. Synthesis and Evaluation of Further Analogues,
a
Modified at Indole Positions 4−7
a
a
An expanded form of this Table, including yields, is included in the
b
Supporting Information (Table S3). Reported mean and SEM values
represent the results of at least three independent experiments.
screened for solubility, microsomal stability, and Caco-2
permeability in an initial effort to identify likely ADME
liabilities. Aqueous solubility thresholds for 12, 13, and 32 were
estimated to be in the region of 10 μM by HPLC (cf. 1, ≤ 1
μM) although reproducibility was difficult, thought to be due to
approaching the detection limit by UV, such that these
measurements were not considered reliable. Results from
dynamic light scattering (DLS) experiments (see Supporting
Information, Figure S6 and accompanying text) were more
supportive of a solubility threshold of 10 μM or higher for
several compounds, including thresholds of at least 20, 80, and
40 μM, respectively, for 12, 13, and 32. Taken together, it
appeared as anticipated that considerable improvement had
already been realized over 1, which in contrast was seen to form
large aggregates at 10 μM by DLS with some evidence for
aggregate formation at 1 μM (markedly raised count rate).
Caco-2 permeability data (Table 5) for selected compounds
were predictive of good intestinal absorption, i.e. Papp (A → B)
> 10 × 10−6 cm s−1, with no evidence of active efflux (Papp ratio
<2). Mouse microsomal half-lives, estimated by extrapolation
from 30 min assay end point values (Table S6), did prove a
concern, however, indicating rapid clearance would likely be an
issue in vivo. Metabolism was previously identified44 as a
possible contributing factor to the suboptimal pharmacokinetics
of 1, making this issue all the more important to address in the
present series of related compounds.
a
An expanded form of this Table, including yields, is included in the
b
Supporting Information (Table S4). Reported mean and SEM values
represent the results of at least three independent experiments.
c
Compounds 37 and 39 were synthesized according to Scheme 1a, in
d
turn meaning intermediate 68b was not synthesized. The R2 group in
39 is 4-methoxyphenyl, not 2-methoxy-5-pyridyl as in all the other
examples.
equally as metabolically unstable as their indole counterparts 12
and 13, if not more so, and also suffered from a small but
notable drop-off in cytotoxic potency. Alternatively, introducing
a tertiary center α- to N1, as in 57, did not offer any
improvement either.
In contrast, adding extra bulk to the methoxy-substituent of
the R2 group to block metabolism at this site greatly improved
stability, conferring an order of magnitude increase in half-life
without adversely compromising cytotoxic potency (compare
59 with 13). However, adding further substitution at this site
did negatively affect activity (compare tert-butyl derivative 60).
A related modification to the methoxy-group in the R1
substituent resulted in modest stabilization (comparing 56
with 13), but in this case the penalty in cytotoxic activity was
too severe to consider it a useful change.
A further selection of analogues was subsequently designed
and synthesized in an attempt to address the shortcomings with
respect to metabolic stability. It had not been possible to extract
conclusive trends from the first round of data, although the R1
group did appear to influence stability (noting t1/2 order 32 >
13 > 12); thus, exchange of the indole for an indolizine core
heterocycle, replacing the C−N bond linking the R1 substituent
with a C−C bond, was attempted but unfortunately did not
prove advantageous. Indolizine analogues 53 and 54 proved
Matched pair analysis for prediction of solubility, as
described above, anticipated a modest but acceptable
compromise for representative compounds introducing more
bulky hydrophobic groups in an attempt to block metabolism
F
DOI: 10.1021/acs.jmedchem.5b01312
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Table 5. Assessment and Preliminary Optimization of Selected in Vitro PK Properties
a
b
An expanded form of this Table, including yields, is included in the Supporting Information (Table S5). Reported mean and SEM values represent
c
the results of at least three independent experiments. Papp values are quoted as ×106 cm −1; values quoted are the mean of two independent
experiments, which were in good agreement in every case. Papp ratio = Papp (B → A)/Papp (A → B). Recovery was <50% in the Caco-2 assay with
d
e
59, indicating potential issues with solubility or binding to cellular proteins/lipids under the assay conditions, which may reduce the accuracy of the
result.
(56, 57, 60; Supporting Information, Figure S17). These
predictions were indeed reflected in the DLS measurements
referred to earlier (Supporting Information Figure S6). Other
analogues prepared specifically with the aim of improving
solubility included 61, 63, and 65, and although all three
compounds appeared fully soluble up to 200 μM by DLS
(Supporting Information, Figure S6a,b), introduction of the
carboxylic acid group abolished cytotoxicity in every case.
Targeting of Tubulin. A subset of cytotoxic compounds
from the initial SAR study was evaluated further to confirm
their presumed mode-of-action, inhibition of tubulin polymer-
ization (Figure 2). GTP-promoted microtubule assembly from
purified porcine brain tubulin was followed by measuring
changes in optical density (OD340) of the solution over a period
of 60 min following initiation of reaction, according to a well-
established protocol.83 Increasing concentrations of 12, 32, 33
(Figure 2a), 51, 57, and 59 were found to suppress the
polymerization reaction in a dose-dependent manner. Assay
end point values, expressed relative to vehicle control, were
considered indicative of extent of reaction, and a secondary plot
of these results against inhibitor concentration permitted
derivation of IC50 values (Figure 2b; full dose−response plots
G
DOI: 10.1021/acs.jmedchem.5b01312
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
mature animal brain, such as would be expected in a
commercially sourced preparation, and an unmodified sample
derived from infant animal brain is required to see inhibition.
This observation, among others, led the authors to conclude
the compound binds to an unprecedented site in the vicinity of
Lys40 of α-tubulin. The apparent inactivity of 1 in the present
tubulin polymerization assays was therefore anticipated, but
similar results for 11 and 13 seemed unusual given the positive
results for closely related analogues.
To investigate further, the above set of compounds, together
with 10, 23, 50, and 57, was tested in an alternative tubulin
polymerization assay which relies on the fluorescence of DAPI
when incorporated into growing microtubules,84 as opposed to
detecting polymer growth through changes in OD340. Under
these conditions, inhibition was now observed for all library
compounds, including 11 and 13 (Supporting Information,
Figures S2,S3). Where results were obtained in both assays,
there was qualitative evidence for agreement between the
relative IC50 values (Supporting Information, Figure S4),
although the very small sample size precludes any firm evidence
of a correlation. Similarly, the trend in potencies obtained in
both polymerization assays (particularly the fluorescence-based
method, with the larger number of compounds screened) is in
good general agreement with that seen earlier in cellular
cytotoxicity activities, suggesting interaction with tubulin was a
major factor leading to the observed toxic phenotype (see
Supporting Information, Figure S5 and accompanying legend).
It is worth noting that although 1 did not show inhibition of
tubulin polymerization in the absorbance-based assay, as stated
above, there was some evidence of activity using the
fluorescence-based method (Supporting Information, Figure
S2), although results between experiments were inconsistent. It
is thus not possible to reach a conclusion either way regarding
inhibition by 1, although it seemed plausible to suggest that
obtaining a clear result was compromised by the poor solubility
of the compound. To investigate this possibility further and
assess whether the initial results for 11 and 13 described above
might likewise be linked to solubility issues, DLS experiments
were carrried out to look for evidence of aggregation over a
concentration range relevant to the inhibition assays (Support-
ing Information, Figure S6a,b). Extensive aggregation was seen
for 1 at 10 μM, forming particles of approximately 1 μm in
diameter, with some evidence for aggregate formation at 1 μM
as already noted above. In contrast, most compounds evaluated
from the present library showed no obvious evidence of
aggregation at 10 μM, which was true of the subset evaluated in
both the tubulin polymerization assays and the DLS screen
(11−13, 32, 33, 53, 57, and 59). It seems apparent that 11 and
13 are not appreciably different in solubility to other
representative library members which did display tubulin
polymerization inhibition in both assays, suggesting the
observed inactivity of the former in the OD340-based assay is
due to an as yet unknown property of these compounds
(because both seem fully soluble over the concentration range
used). In contrast, the DLS results lend support to the idea
that, in the case of 1, poor compound solubility may indeed
interfere with the reliability of these assay readouts.
Figure 2. (a) Tubulin polymerization time-course plots following
changes in OD340 of reaction mixtures in the presence of increasing
concentrations of 33 (blue lines) compared to vehicle control (red
line). (b) Secondary plots of reaction end point data allow
determination of IC50 for inhibition of tubulin polymerization by 33
and other compounds as indicated. Data from multiple experiments
was plotted as individual data points, as shown, and fit globally. (c)
Competition for tubulin binding between [3H]-colchicine (at 0.1 μM)
and vincristine, docetaxel, or podophyllotoxin shows only the latter
competes with binding at the colchicine site, as expected. (d) The
same six compounds listed in (b) above all compete for binding at the
colchicine site with IC50 values as shown, with representative results
graphed for 32, 51, 57, and podophyllotoxin (positive control).
for all six compounds, and colchicine used as a positive control,
are included in the Supporting Information, Figure S1). In
contrast, compounds 11 and 13 unexpectedly showed no
inhibition in the polymerization assay, alongside 1, for which
such a result has already been reported40 (Supporting
Information, Figure S2). Earlier work40 on 1 suggested it
does not inhibit the polymerization of post-translationally
modified tubulin (acetylated on Lys40 of the α-subunit) from
Having gained evidence that compounds from the current
series do interact directly with tubulin, it was considered
important to elucidate their binding site. There are three well-
characterized binding domains on tubulin, the colchicine, Vinca
alkaloid, and taxane binding sites, all located on the β-subunit,
and most (but not all) established MTAs bind at or close to
H
DOI: 10.1021/acs.jmedchem.5b01312
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Figure 3. Representative images of fixed FaDu cells stained for α-tubulin (green) and DNA (blue) after treatment as indicated for 24 h. Control cells
(DMSO-treated) show a well-structured microtubule network and are dividing normally, with mitotic cells in anaphase (center) and prophase (left)
clearly visible in the upper-left panel. In contrast, cells treated with active tubulin polymerization inhibitors showed multiple mitotic defects, most
notably formation of multipolar spindles, and compound 59 appeared especially destructive to the microtubule network at high concentration.
Images in all panels are to the same scale. An enlarged version of these images is provided in the Supporting Information for improved clarity (Figure
S13).
one of them. Taxane domain ligands are known to act as
microtubule stabilizers, whereas drug binding to the Vinca
alkaloid or colchicine sites destabilizes microtubules. Given that
actives from the current series inhibit tubulin polymerization,
binding at either the colchicine or Vinca domains was
anticipated. Evaluation of test compounds at all sites was
probed by competition with tritium-labeled standards (colchi-
cine, vinblastine, or paclitaxel), looking for dose-dependent
displacement of radioactivity from the protein with increasing
test compound concentration. The same set of compounds
evaluated in the tubulin polymerization assays was screened for
binding. Results at the colchicine site proved the most
conclusive, with the assay first validated using competition of
[3H]-colchicine with podophyllotoxin, a known binder at this
site85 (Figure 2c). As expected, no competition with [3H]-
colchicine was observed with either vincristine or docetaxel. All
test compounds gave positive results, with selected IC50 values
presented in Figure 2d together with dose−response plots for
32, 51, and 57 (see Supporting Information for dose−response
plots for all compounds, Figure S7). Good correlation between
FaDu cytotoxicity (LC50 values) and IC50 for competition with
[3H]-colchicine seems apparent (Supporting Information,
Figure S8), although as applies to the polymerization assay
results above, a firm conclusion cannot be drawn from existing
data given the small sample size. Evaluation of 1 again proved
problematic (Figure S9): a partial competition curve was
obtained up to a point where binding leveled off, suggesting a
solubility threshold could have been reached. It is informative
to consider this result in the light of previous data86 reporting
that [3H]-1 was readily displaced from tubulin by excess
colchicine, whereas in the reciprocal experiment, [3H]-
colchicine binding was only partially competed by excess 1.
Considering all aspects together, these results could be
interpreted to mean that 1 binds to the colchicine site with
good affinity but its poor solubility hinders accurate assessment
of this binding, especially where the compound must be present
in excess. It remains to be said that binding of 1 to the
colchicine site is necessarily assessed independently of that at
the other proposed site on the α-subunit, suggesting the
compound could be a dual site binder to tubulin.
ing legends). Briefly, at the Vinca site, docetaxel was found to
complete with [3H]-vinblastine binding just as well as
vincristine, whereas at the taxane site, vincristine competed
with [3H]-paclitaxel binding with an IC50 only 2-fold higher
than that for docetaxel (Supporting Information, Figure S10).
These observations limited the confidence with which clear
results could be drawn from the data; however, none of the
compounds showed any obvious evidence of binding at the
taxane domain in agreement with their activity as inhibitors,
rather than promoters, of tubulin polymerization. At the Vinca
site, some (but not all) of the test compounds displayed IC50
values close to that of vincristine, suggesting the intriguing
possibility that these members of the library are capable of
binding tubulin at both the colchicine and Vinca alkaloid
domains (Supporting Information, Figure S12); however, given
the difficulties encountered with the assay this can only be
considered a preliminary finding, deserving further investigation
in due course.
In sum, a small set of compounds from the cytotoxicity-led
SAR study was found to inhibit tubulin polymerization through
binding at the colchicine site, with good general agreement
between potency in both assays and the earlier LC50 data.
Attention was next focused on whether the expected phenotype
for destabilization of microtubules is seen at the cellular level.
To visualize the effect of 32, 33, and 59 on the microtubule
network, FaDu cells were treated with two concentrations of
each compound for 24 h then fixed and stained for α-tubulin
and DNA (Figure 3). Concentrations were chosen correspond-
ing to approximately 1× and 5× the cytotoxic LC50 (72 h end
point); after exposure for 24 h, the majority of the cell
population will be in mitotic arrest at the higher of these
concentrations (as discussed later; see Figure 4).
All three compounds proved able to induce mitotic spindle
defects at low concentration, and in particular, several cells with
multipolar spindles were seen with 33 and 59. It is also clear
that a high concentration of each compound caused extensive
damage to the microtubule network. Comparison of the images
suggests there may be some differences in the cellular
phenotype produced by each compound, perhaps indicating
subtle variation in their precise mode of action. This
observation could form the basis of future investigations to
understand the effects of compounds from this series in cells in
more detail. In the present context though, it is clear that the
three compounds tested caused obvious defects in the
In our experience, evaluation of the same 12 library
compounds as above at the Vinca alkaloid and taxane binding
sites proved unexpectedly difficult (for a more detailed analysis,
see Supporting Information, Figures S10−S12 and accompany-
I
DOI: 10.1021/acs.jmedchem.5b01312
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
equivalent end points is not possible because all cells are dead
by 72 h (compromising analysis by flow cytometry), whereas
meaningful LC50 values cannot be obtained at 24 h (because
viability is only partially reduced by this stage, typically by only
40−50% even at the top of the dose−response concentration
range). It appears that G2/M accumulation results relatively
soon after treatment, within the first 18−24 h, and a prolonged
period of cell cycle arrest is entered, as is typical with MTAs,87
before ultimate cell death, which is not extensive until around
48 h (data not shown).
Thus, it was established that representative compounds from
the current series cause disruption of the cellular microtubule
network and induce cell cycle arrest in mitosis, consistent with
their tubulin polymerization inhibition properties.
Evaluation in Other Cell Models. It was also considered
important to evaluate some of the most potent compounds
more widely, including cell lines representing other cancer
types. The additional lines used were SCC-4 (squamous cell
carcinoma of the tongue), SK-MEL-28 (malignant melanoma),
SH-SY5Y (neuroblastoma), and MES-SA (uterus sarcoma). A
multidrug-resistant derivative of the latter, MES-SA/Dx5, was
also used because it is known to express very high levels of P-
glycoprotein (P-gp).88,89 When used in combination, these two
lines therefore represent a useful tool to evaluate the propensity
of test compounds for efflux-mediated resistance.
Screening against the additional cell lines was carried out
with a selection of six compounds (13, 32, 33, 51, 57, and 59),
spanning an order of magnitude range of LC50 from the FaDu
screen (12−137 nM), alongside 1 for comparison (Figure 5).
The observed cytotoxic activities, together with the trend in
potency, remained very similar across all lines (Figure 5a; see
also Supporting Information, Figure S14). The multidrug-
resistant cell line MES-SA/Dx5 showed significant cross-
resistance to vinblastine, paclitaxel, and colchicine as expected,
but not to 1 or the present test compounds (Figure 5b),
suggesting the latter are not substrates for efflux by P-gp.
Further support for this assumption is provided by the
bidirectional Caco-2 assay results discussed above (Table 5).
Efflux can also occur by the Caco-2 cell line and may be
mediated by any of multiple transporters including P-gp,
BCRP/ABCG2, MRP1, and MRP2.90−92 Hence, the fact an
efflux ratio of less than 2 was observed for 12, 13, 17, 32, 33,
and 59 suggests these compounds are not good substrates for
any of the transporters characterized in Caco-2.
Figure 4. (a) Exposure of FaDu cells to increasing concentrations of
the compounds indicated, for 24 h, caused dose-dependent cell cycle
arrest in the G2/M phase; for comparisons, ns = P > 0.05, ** = P <
0.01, *** = P < 0.001. (b) Representative histograms for the three
concentrations of 57 (cell count versus DNA staining intensity)
showing the clear shift in cell cycle distribution upon compound
treatment; DMSO control (not shown) was essentially identical to the
0.5 × LC50 (green) plot. (c) A more detailed dose−response analysis
for compounds 33 and 59 provided EC50 values for the observed
mitotic arrest.
The present series of compounds appears to possess general
cytotoxic activity against cancer cell lines and does not seem
susceptible to active efflux based on results from two cell lines
expressing transporter proteins commonly associated with
multidrug resistance.
microtubule network of FaDu cells at concentrations consistent
with the cytotoxic potencies determined earlier. Because the
compounds have also been shown to interact directly with
purified tubulin, these results do appear to confirm engagement
of the same target in cells as the basis of their cytotoxic action.
A small set of the most potent compounds was also assessed
by flow cytometry for their effect on the cell cycle distribution
of proliferating FaDu cells (Figure 4). As can be seen clearly, all
compounds induced accumulation of the majority of the
population at G2/M phase after 24 h exposure. Using multiple
concentrations of 33 and 59 allowed dose−response curves to
be plotted and EC50 values derived for this mitotic arrest, and
these values are approximately 5-fold or 2-fold higher,
respectively, than the LC50 values for cytotoxicity (72 h end
point). This does suggest mitotic arrest as the primary cause of
cell death in the latter assays, although a direct comparison at
A further priority was to evaluate selected compounds from
the series in a suitable 3D cell culture model. Whereas 2D
(monolayer) culture provides a convenient platform for initial
screening experiments, it is not a representative model of a
clinical tumor; thus, 3D cell culture models are increasingly
being developed and applied to better assess the potential for in
vivo activity of preclinical compounds.93 The simplest such 3D
model is the multicellular tumor spheroid (MCTS), a valuable
screening tool which may be formed from either a single cell
type or cocultured cells.94 The FaDu cell line employed for
primary screening in this study readily forms monocultured
MCTS,95 permitting the effect of compounds 32, 33, and 59 on
these spheroids to be monitored in a time- and dose-dependent
fashion (Figure 6). FaDu MCTS are an established and well-
J
DOI: 10.1021/acs.jmedchem.5b01312
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
show efficacy in the 3D MCTS model at concentrations similar
to those observed in the earlier FaDu monolayer screens,
without any prohibitive dropoff in activity. There is certainly
scope for higher-content screening using MCTS which could
be exploited in the future as part of more detailed mode-of-
action studies, but for the purposes of the current work, it was
satisfactory to note successful translation of activity to a 3D
model for representative compounds in the existing series.
Mouse Xenograft Study. Two compounds, 33 and 59,
were selected for a preliminary in vivo efficacy study in a mouse
model of HNC using xenografts of the FaDu cell line. In
addition to the positive results for both in the FaDu MCTS
model described above, these compounds were chosen because
they were the only candidates with both LC50 < 100 nM from
the primary cytotoxicity screen and a microsomal half-life of at
least 30 min. Given concerns over the still suboptimal solubility
of the series, oral dosing was chosen since earlier data suggested
moderate to good absorption by this route (Table 5). Mice
with established xenografts were dosed once daily for a period
of 10 days as shown (Figure 7), and a significant inhibition of
tumor growth was seen initially in all treatment groups.
An unexpected feature of the results is the apparent partial
recovery in tumor volume before the end of the dosing period
(comparing measurements between days 6 and 9), although
numerical analysis revealed this difference is only statistically
significant in the group receiving 59 at 10 mg/kg. There are a
number of possible explanations for this observation, and
particularly in the absence of accompanying pharmacokinetic
data, any interpretation would be speculative at best. The FaDu
cell line does overexpress BCRP/ABCG2,82 so even though
there is no reason to assume acquired resistance is developing
to the current test compounds, the above results do emphasize
the importance that the potential for new drugs to activate
mechanisms of resistance be appropriately investigated at an
early stage of development. On a more encouraging note,
treatment with 33 appeared to result in poor tumor growth
even after administration had ceased.
Figure 5. (a) Distribution of cytotoxicity LC50 values for the
compounds indicated across a range of cell lines representing different
cancer types. (b) Ratio of LC50 values for various compounds between
the multidrug-resistant cell line MES-SA/Dx5 and its parent line,
MES-SA. Varying degrees of resistance are seen to vinblastine,
paclitaxel, and colchicine but not to 1 or a selection of members from
the present series. The cytotoxicity data used in making both these
graphs are tabulated in full (mean, SEM, n) in the Supporting
Information (Table S7a,b).
characterized model of HNC solid tumors that have been
widely employed.94−96
An effect on normal spheroid growth was not evident until
2−3 days after dosing, but from this point onward, an ongoing
dispersion of cells from the spheroid was seen (Figure 6a),
assumed to be due to the cytotoxic effect of the compounds
inducing progressive cell death. As shown for 59 (Figure 6b),
the extent of this effect was dose-dependent at the experiment
end point (7 days post dosing), and size measurements taken at
regular intervals allowed time-course plots of spheroid volume
to be made (also for 59; Figure 6c). This revealed that at
concentrations high enough to ultimately reduce the spheroid
volume (>0.1 μM), it takes several days before the toxic effect
of the compound is sufficient to outweigh new cell growth,
emphasizing the requirement for continued drug exposure in
vivo. Measurements taken at the 7 day end point over a range
of concentrations allowed dose−response plots to be
constructed for all three test compounds and EC50 values for
spheroid disruption to be derived (Figure 6d). Combined
results from two such independent experiments were compared
with the 2D cytotoxic potencies determined earlier (Figure 6e).
Compounds 32 and 59 demonstrated good agreement
between duplicate experiments, with a 3D potency approx-
imately 3- or 2-fold higher, respectively, than the requisite result
in 2D culture. In contrast, more variation was seen in 3D
experiments with 33 for reasons which were not immediately
clear, but 3D results were nonetheless well within an order of
magnitude of those seen in 2D. Hence, all three compounds did
CONCLUSIONS
■
A library of modified indole-3-glyoxylamides has been reported,
capable of inhibiting tubulin polymerization and displaying
cancer cell cytotoxicity with a well-defined SAR. Furthemore,
two compounds from the series displayed evidence for efficacy
in a mouse xenograft model of HNC. This study builds on
previous work by others which has already established a range
of compounds of the indole-3-glyoxylamide class, such as 1 and
2, as promising tubulin polymerization inhibitors. Other
members of the class have been investigated
preclinically,97−99 including 88−91 (Figure 8) arising from an
SAR study100 published shortly after the disclosure of 1, with
both 90 and 91 reportedly demonstrating in vivo activity.
Given the apparent failure of 2 in the clinic, and the
difficulties encountered in attempting to develop 1, we
hypothesized as the basis of the present work that the
physicochemical and PK properties of such compounds might
be improved by removing the benzylic group at N1 common to
existing compounds in the class and replacing it with aliphatic
substituents. This current study details first steps in such a
direction, establishing that compounds modified thus are
capable of retaining potent activity, albeit with a subtly different
but distinct SAR suggestive of changes in binding pose. The
desired improvement in solubility as compared to 1 looks to
have been realized successfully.
K
DOI: 10.1021/acs.jmedchem.5b01312
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Figure 6. Effect of compounds 32, 33, and 59 on multicellular tumor spheroids (MCTS) formed from the FaDu cell line. (a) Images captured at 0,
2, 4, and 7 days postdosing demonstrate ongoing toxicity toward treated MCTS, whereas untreated controls show continued growth over the same
time period; scale bar (lower left panel) 500 μm. (b) Images of MCTS treated with increasing concentrations of 59 captured 7 days postdosing, with
enhanced brightness to emphasize the fragmentation of spheroids which is induced in a dose-dependent manner; scale bar (leftmost panel) 500 μm.
(c) Time-course plots of average volume at selected concentrations of 59 again show dose-dependent toxicity toward FaDu MCTS. (d)
Representative dose−response plots at end point (7 days postdosing) allow EC50 determination for spheroid disruption; values derived from these
data are 39 nM (32), 77 nM (33), and 113 nM (59). (e) Comparison of 3D cytotoxic potencies (mean SEM) with values from 2D (monolayer)
screening reported earlier (Table 5).
Although the reported library has been established as a
workable hit series, further optimization of PK properties is
evidently necessary, although the present results suggest this is
achievable. A better-defined SAR, aspects of which remain
unexplored, is ideally necessary to help inform this undertaking.
A more detailed understanding of the compounds’ mode of
action at the cellular level is also important, especially to
interrogate and direct improvement of their selectivity for the
desired target, tubulin. In the context of the MTA drug class,
selectivity is a critical consideration given that a narrow
therapeutic window remains arguably the most significant
drawback of the existing agents in clinical use. Compounds with
superior therapeutic window are urgently required, and
evidence for such was probably the most appealing aspect of
preclinical data for 1. Efforts to assess and optimize the
therapeutic window must therefore be a priority in the
identification and development of potential new MTAs likely
to be of clinical value, arising either from this series or
elsewhere.
EXPERIMENTAL SECTION
■
Chemistry. General Procedures. Reactions were run under an
inert atmosphere of N2 unless stated otherwise. Anhydrous solvents
were obtained from an in-house “Grubbs” system and stored under N2
until use. All other solvents and reagents were obtained from
commercial sources and used as supplied. NMR spectra were recorded
at the frequencies listed, and 13C NMR spectra were recorded with
complete proton decoupling. J values are reported in Hz. Melting
points were determined using a Gallenkamp melting point apparatus
in capillary tubes and are uncorrected. Analytical HPLC was carried
out using a C18 5 μm column (4.6 mm × 250 mm) and a gradient of
5−100% MeCN/water over 20 or 25 min (holding at 100% for 5
min), with UV detection at 254 nm and a flow rate 1.0 mL min−1.
Purity of all screening compounds was determined to be >95% by
HPLC; a tabulated list of purities is included in the Supporting
Information. Where required, preparative HPLC was performed using
a C18 5 μm column (19 mm × 250 mm) with UV detection at 254 nm
and a flow rate between 15−20 mL min−1, using a gradient specified in
each case. Dynamic light scattering measurements were carried out
using the Malvern Instruments Zetasizer Nano ZS system, equipped
with a 4 mW He−Ne laser at 633 nm, as described.101 To our
knowledge, the compound series presented here is free of obvious
interference motifs (“PAINS”), and the presence of a well-defined SAR
L
DOI: 10.1021/acs.jmedchem.5b01312
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
2.40 (s, 3H), 1.00 (d, 6H, J = 6.0). δC (101 MHz, DMSO-d6) 145.3,
133.8, 130.6, 128.1, 71.4, 70.9, 65.5, 22.3, 21.5. m/z (ES) 281 ([M +
Na]+). HRMS, found 281.0817 (C12H18NaO4S requires 281.0824).
1-Methoxypropan-2-yl 4-Methylbenzenesulfonate (Precursor to
4g). 1-Methoxypropan-2-ol (4.9 mL, 4.5 g, 50 mmol) was dissolved in
anhydrous THF (250 mL) then the solution cooled to 0 °C. Sodium
hydride (60% dispersion in mineral oil, 2.3 g, 58 mmol) was added
portionwise over 10 min, followed 5 min later by the addition of p-
toluenesulfonyl chloride (11.0 g, 58 mmol) and DMAP (50 mg). After
30 min at 0 °C, the mixture was warmed to 40 °C and maintained at
this temperature for 24 h. The reaction was quenched by addition of
satd NH4Cl then the mixture evaporated under vacuum. The residue
was partitioned between Et2O and water, the layers separated, and the
aqueous phase extracted again with Et2O. The combined organic
extracts were washed with water then brine, dried over Na2SO4,
filtered, and evaporated, yielding a thick, yellow oil. Purification by
flash column chromatography on silica gel, eluted with 50 → 75 →
100% CH2Cl2/hexane then 1% MeOH/CH2Cl2, afforded the title
compound as a yellow oil (7.3 g, 60%) which was stored at −20 °C
until further use. δH (400 MHz, DMSO-d6) 7.79 (d, 2H, J = 8.0), 7.47
(d, 2H, J = 8.0), 4.66 (ddd, 1H, J = 3.5, 6.0, 12.5), 3.37−3.27 (m, 2H),
3.13 (s, 3H), 2.42 (s, 3H), 1.14 (d, 3H, J = 6.0). δC (101 MHz,
DMSO-d6) 145.1, 134.2, 130.4, 128.0, 78.8, 74.3, 58.7, 21.5, 17.4. m/z
(ES) 245 ([M + H]+). HRMS, found 245.0845 (C11H17O4S requires
245.0848).
Oxetan-2-ylmethyl 4-Methylbenzenesulfonate (precursor to 4o).
A stirred solution of oxetane-2-methanol (0.95 g, 10.8 mmol) in
CH2Cl2 (30 mL) was cooled to 0 °C then triethylamine (1.65 mL,
1.20 g, 11.8 mmol) and p-toluenesulfonyl chloride (2.26 g, 11.8 mmol)
were added successively, followed by a small amount of DMAP. The
mixture was maintained at 4 °C overnight then poured into water,
extracted, and the layers separated. The aqueous layer was extracted
with additional CH2Cl2 then the combined organic solutions washed
with water, dried over Na2SO4, filtered, and evaporated, giving a pale-
yellow oil. Flash column chromatography, eluted with 15 → 30 →
45% EtOAc/hexane, afforded the title compound as a crystalline white
solid (2.11 g, 81%). δH (400 MHz, CDCl3) 7.84 (d, 2H, J = 8.0), 7.37
(d, 2H, J = 8.0), 4.98−4.91 (m, 1H), 4.66−4.59 (m, 1H), 4.56−4.49
(m, 1H), 4.17 (d, 2H, J = 4.0), 2.78−2.67 (m, 1H), 2.64−2.53 (m,
1H), 2.47 (s, 3H). δC (101 MHz, CDCl3) 144.9, 132.9, 129.9, 128.0,
78.6, 72.0, 69.0, 23.4, 21.7. m/z (ES) 243 ([M + H]+). HRMS, found
243.0694 (C11H15O4S requires 243.0691); mp 55−58 °C (lit.103 58−
59 °C).
Figure 7. Compounds 33 and 59 were administered by oral gavage at
the doses indicated, once daily for 10 days (gray arrowheads), and
produced a significant reduction in tumor growth relative to the
control group. This effect was statistically significant (P < 0.05) in all
treatment groups at days 3, 6, and 9, and significant at the P < 0.0001
level at day 6 in all cases. Only 33 exerted a sustained effect beyond the
dosing period (P < 0.01 at days 13 and 15). Detailed plots of tumor
volume, body weight, and survival for each treatment group, including
results with vinblastine as a comparator/control drug, are presented in
the Supporting Information (Figure S15).
Typical Procedure for Synthesis of N-Alkyl Intermediates 4a−i,
4o, and 67a−o. Indole, or the relevant indole derivative, was
dissolved in anhydrous DMF (or THF if specified) at a concentration
between 0.25−0.5 M, then the solution cooled to 0 °C. Alkylating
agent (1.2 equiv unless stated otherwise) was added, then sodium
hydride (60% dispersion in mineral oil, 1.2 equiv) was added in small
portions over a period of 10−15 min. Once addition was complete, the
reaction mixture was stirred at 0 °C for 30 min then continued at rt
(unless stated otherwise) for a duration specified in each case. The
reaction was quenched by cautious, dropwise addition of water or satd
NH4Cl, then evaporated to dryness. The residue was partitioned
between EtOAc or Et2O and water and extracted, then the aqueous
layer extracted with a second portion of organic solvent. The
combined extracts were washed three times with water, once with
brine, dried over Na2SO4, filtered, and evaporated to afford the crude
product which was purified by flash column chromatography on silica
gel as indicated for each case, then dried and stored at −20 °C until
further use.
1-Ethyl-1H-indole 4a. Prepared from indole (1.17 g, 10 mmol) and
bromoethane (1.35 mL, 1.97 g, 18 mmol), reacting overnight at 30 °C.
Column eluent 5 → 10 → 20% CH2Cl2/hexane. Product obtained as a
colorless oil (1.12 g, 77%). δH (400 MHz, DMSO-d6) 7.55 (d, 1H, J =
8.0), 7.46 (dd, 1H, J = 0.5, 8.0), 7.38 (d, 1H, J = 3.0), 7.14 (dt, 1H, J =
1.0, 7.0), 7.02 (dt, 1H, J = 1.0, 7.0), 6.43 (dd, 1H, J = 0.5, 3.0), 4.19 (q,
2H, J = 7.0), 1.35 (t, 3H, J = 7.0). δC (101 MHz, DMSO-d6) 135.8,
128.6, 128.3, 121.4, 120.9, 119.2, 110.1, 100.9, 40.7, 15.9; m/z (ES)
Figure 8. Some other indole-3-glyoxylamides with reported in vitro or
in vivo antitumor activity. Note 88 is the indole analogue of 2.
(with respect to both tubulin interaction and cytotoxicity) provides
good evidence against nonspecific modes of action.102
2-Isopropoxyethyl 4-Methylbenzenesulfonate (Precursor to 4e).
2-Isopropoxyethanol (1.15 mL, 1.05 g, 10 mmol) and triethylamine
(1.53 mL, 1.11 g, 11 mmol) were dissolved in anhydrous CH2Cl2 and
the solution cooled to 0 °C before addition of p-toluenesulfonyl
chloride (2.10 g, 11 mmol). The mixture was stirred at 0 °C for 30 min
then at rt overnight. The reaction was quenched by addition of satd
NH4Cl then diluted with water and extracted with CH2Cl2 (3 × 20
mL). The combined extracts were dried over Na2SO4, filtered, and
evaporated to give a colorless oil which was purified by flash
chromatography on silica gel, eluted with 1:1 CH2Cl2/hexane →
CH2Cl2 → 1:4 EtOAc/CH2Cl2, affording the title compound as
colorless oil (1.46 g, 57%). δH (400 MHz, DMSO-d6) 7.78 (d, 2H, J =
8.0), 7.48 (d, 2H, J = 8.0), 4.08 (t, 2H, J = 4.5), 3.52−3.45 (m, 3H),
M
DOI: 10.1021/acs.jmedchem.5b01312
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
146 ([M + H]+). HRMS, found 146.0970 (C10H12N requires
146.0970). HPLC purity 98.4%.
6.5). δC (101 MHz, DMSO-d6) 136.4, 129.5, 128.5, 121.3, 120.8,
119.2, 110.4, 100.7, 53.3, 29.6, 20.4. m/z (ES) 174 ([M + H]+).
HRMS, found 174.1275 (C12H16N requires 174.1283). HPLC purity
99.6%.
1-Isopropyl-1H-indole 4b. Prepared from indole (2.93 g, 25 mmol)
and 2-iodopropane (3.0 mL, 5.1 g, 30 mmol), reacting at rt for 90 min.
Column eluent 5 → 10 → 20% CH2Cl2/hexane. Product obtained as a
colorless oil (1.67 g, 42%). δH (400 MHz, DMSO-d6) 7.55 (d, 1H, J =
7.5), 7.51−7.46 (m, 2H), 7.15−7.10 (m, 1H), 7.01 (t, 1H, J = 7.5),
6.45 (d, 1H, J = 3.0), 4.74 (septet, 1H, J = 6.5), 1.45 (d, 6H, J = 6.5).
δC (101 MHz, DMSO-d6) 135.6, 128.6, 125.0, 121.3, 120.9, 119.3,
110.2, 101.2, 46.8, 23.0. m/z (ES) 160 ([M + H]+). HRMS, found
160.1120 (C11H14N requires 160.1126). HPLC purity 99.5%.
1-(2-Methoxyethyl)-1H-indole 4c. Prepared from indole (2.34 g, 20
mmol) and 2-bromoethyl methyl ether (2.26 mL, 3.34 g, 24 mmol),
reacting at rt for 30 min. Column eluent 20 → 30 → 40 → 60%
CH2Cl2/hexane. Title compound is a viscous, pale-yellow oil (2.90 g,
83%). δH (250 MHz, DMSO-d6) 7.58−7.45 (m, 2H), 7.34 (d, 1H, J =
3.0), 7.17−7.09 (m, 1H), 7.06−6.98 (m, 1H), 6.42 (dd, 1H, J = 0.5,
3.0), 4.32 (t, 2H, J = 5.0), 3.64 (t, 2H, J = 5.0), 3.21 (s, 3H). δC (62.8
MHz, DMSO-d6) 136.4, 129.4, 128.6, 121.4, 120.8, 119.3, 110.3, 101.0,
71.5, 58.5, 45.8. m/z (ES) 176 ([M + H]+). HRMS, found 176.1077
(C11H14NO requires 176.1075). HPLC purity 99.4%.
1-Butyl-1H-indole 4d. Prepared from indole (1.17 g, 10 mmol) and
1-bromobutane (1.29 mL, 1.64 g, 12 mmol), reacting at rt for 30 min.
Column eluent 5 → 10 → 20% CH2Cl2/hexane. Product obtained as a
pale-yellow oil (1.43 g, 83%). δH (400 MHz, CDCl3) 7.70 (dt, 1H, J =
1.0, 8.0), 7.42 (dd, 1H, J = 0.5, 8.0), 7.30−7.25 (m, 1H), 7.19−7.14
(m, 2H), 6.55 (dd, 1H, J = 0.5, 3.0), 4.18 (t, 2H, J = 7.0), 1.92−1.84
(m, 2H), 1.41 (sextet, 2H, J = 7.5), 1.03 (t, 3H, J = 7.5). δC (101 MHz,
CDCl3) 136.0, 128.6, 127.8, 121.3, 121.0, 119.2, 109.4, 100.9, 46.2,
32.4, 20.2, 13.8. m/z (ES) 174 ([M + H]+). HRMS, found 174.1276
(C12H16N requires 174.1283). HPLC purity 99.9%.
1-(2-Isopropoxyethyl)-1H-indole 4e. Prepared from indole (469
mg, 4 mmol) and 2-isopropoxyethyl-4-methylbenzenesulfonate (1.14
g, 4.4 mmol) in THF, reacting at rt overnight. Column eluent 50 → 60
→ 100% toluene/hexane. Title compound obtained as a yellow oil
(387 mg, 50%). δH (400 MHz, DMSO-d6) 7.55 (d, 1H, J = 8.0), 7.49
(d, 1H, J = 8.0), 7.36 (d, 1H, J = 3.0), 7.12 (t, 1H, J = 8.0), 6.99 (t, 1H,
J = 8.0), 6.42 (d, 1H, J = 3.0), 4.29 (t, 2H, J = 5.5), 3.68 (t, 2H, J =
5.5), 3.48 (septet, 1H, J = 6.0), 1.00 (d, 6H, J = 6.0). δC (101 MHz,
DMSO-d6) 136.3, 129.5, 128.5, 121.3, 120.8, 119.3, 110.3, 100.8, 71.4,
66.9, 46.4, 22.4. m/z (ES) 204 ([M + H]+). HRMS, found 204.1385
(C13H18NO requires 204.1388).
1-sec-Butyl-1H-indole 4f. Prepared from indole (2.34 g, 20 mmol)
and 2-iodobutane (2.76 mL, 4.42 g, 24 mmol), reacting at rt for 6 h.
Column eluent 5 → 10 → 20% CH2Cl2/hexane. Product obtained as a
thick, colorless oil (0.20 g, 6%). δH (250 MHz, DMSO-d6) 7.56−7.47
(m, 2H), 7.44 (d, 1H, J = 3.0), 7.14−7.06 (m, 1H), 7.03−6.95 (m,
1H), 6.46 (d, 1H, J = 3.0), 4.49 (sextet, 1H, J = 7.0), 1.92−1.72 (m,
2H), 1.43 (d, 3H, J = 7.0), 0.70 (t, 3H, J = 7.0). δC (62.8 MHz,
DMSO-d6) 136.2, 128.4, 125.4, 121.2, 120.8, 119.2, 110.3, 101.3, 52.7,
29.9, 21.2, 11.1. m/z (ES) 174 ([M + H]+). HRMS, found 174.1275
(C12H16N requires 174.1283).
1-((Tetrahydrofuran-2-yl)methyl)-1H-indole 4i. Prepared from
indole (2.34 g, 20 mmol) and tetrahydrofurfuryl bromide (2.70 mL,
3.96 g, 24 mmol), reacting at rt for 2 h. Column eluent 20 → 40 → 60
→ 80% CH2Cl2/hexane. Product obtained as a pale-pink oil (1.33 g,
33%). δH (400 MHz, CDCl3) 7.68 (d, 1H, J = 8.0), 7.42 (d, 1H, J =
8.0), 7.29−7.22 (m, 2H), 7.17−7.12 (m, 1H), 6.55 (d, 1H, J = 3.0),
4.33−4.19 (m, 3H), 3.91−3.85 (m, 1H), 3.82−3.76 (m, 1H), 2.03−
1.94 (m, 1H), 1.92−1.73 (m, 2H), 1.66−1.56 (m, 1H). δC (101 MHz,
CDCl3) 136.5, 128.7, 128.5, 121.5, 120.9, 119.3, 109.5, 101.4, 78.1,
68.3, 50.1, 29.1, 25.7. m/z (ES) 202 ([M + H]+). HRMS, found
202.1225 (C13H16NO requires 202.1232). HPLC purity 99.8%.
1-(Oxetan-2-ylmethyl)-1H-indole 4o. Indole (426 mg, 3.64 mmol)
was dissolved in DMF (10 mL) and the solution cooled to 0 °C.
Sodium hydride (60% dispersion in mineral oil, 160 mg, 4.00 mmol)
was added in portions over 5 min, then after another 5 min stirring,
oxetan-2-ylmethyl 4-methylbenzenesulfonate (0.97 g, 4.00 mmol) was
added. The reaction was allowed to continue for 3 h, after which it was
deemed essentially complete by TLC. It was quenched by dropwise
addition of satd NH4Cl then extracted between EtOAc and water. The
aqueous layer was extracted with further EtOAc then the combined
organic extracts washed three times with water and once with brine.
The solution was dried over Na2SO4, filtered, and evaporated, giving
an oil which was purified by flash column chromatography on silica
gel, eluted with 10 → 20 → 33 → 45% EtOAc/hexane, to provide the
title compound as a colorless oil (530 mg, 78%). δH (400 MHz,
CDCl3) 7.69 (dt, 1H, J = 1.0, 8.0), 7.41 (dd, 1H, J = 1.0, 8.0), 7.31−
7.23 (m, 2H), 7.18−7.13 (m, 1H), 6.59 (dd, 1H, J = 1.0, 3.0), 5.23−
5.15 (m, 1H), 4.67−4.59 (m, 1H), 4.42−4.32 (m, 3H), 2.72−2.62 (m,
1H), 2.42−2.32 (m, 1H). δC (101 MHz, CDCl3) 136.7, 128.8, 128.5,
121.6, 121.0, 119.4, 109.4, 101.7, 81.1, 68.5, 51.3, 24.6. m/z (ES) 188
([M + H]+). HRMS, found 188.1084 (C12H14NO requires 188.1075).
HPLC purity 99.0%.
4-Chloro-1-isopropyl-1H-indole 67a. Prepared from 4-chloroin-
dole (3.03 g, 20 mmol) and 2-iodopropane (2.4 mL, 4.1 g, 24 mmol),
reacting at rt overnight. Column eluent 5 → 10 → 20% CH2Cl2/
hexane. Title compound obtained as a pale-yellow oil (2.04 g, 53%).
δH (400 MHz, CDCl3) 7.34−7.28 (m, 2H), 7.18−7.12 (m, 2H), 6.66
(d, 1H, J = 3.0), 4.69 (septet, 1H, J = 7.0), 1.56 (d, 6H, J = 7.0). δC
(101 MHz, CDCl3) 136.3, 127.3, 126.2, 124.2, 121.8, 119.0, 108.2,
99.9, 47.6, 22.8. m/z (ES) 194 (100%; [M(35Cl) + H]+), 196 (30%;
[M(37Cl) + H]+). HRMS, found 194.0741 (C11H13ClN requires
194.0737). HPLC purity 99.5%.
5-Chloro-1-isopropyl-1H-indole 67b. Prepared from 5-chloroin-
dole (3.78 g, 25 mmol) and 2-iodopropane (3.0 mL, 5.1 g, 30 mmol),
reacting at rt for 2 h. Column eluent 5 → 10 → 20% CH2Cl2/hexane.
Title compound obtained as a viscous, colorless oil (2.46 g, 51%). δH
(250 MHz, DMSO-d6) 7.58 (d, 1H, J = 2.0), 7.56−7.48 (m, 2H), 7.11
(dd, 1H, J = 2.0, 9.0), 6.44 (d, 1H, J = 3.0), 4.73 (septet, 1H, J = 6.5),
1.43 (d, 6H, J = 6.5). δC (62.8 MHz, DMSO-d6) 134.1, 129.7, 126.8,
124.0, 121.2, 120.0, 111.8, 101.1, 47.3, 22.9. m/z (ES) 194 (100%;
[M(35Cl) + H]+), 196 (30%; [M(37Cl) + H]+). HRMS, found
194.0734 (C11H13ClN requires 194.0737). HPLC purity 99.4%.
5-Chloro-1-(2-methoxyethyl)-1H-indole 67c. Prepared from 5-
chloroindole (1.32 g, 8.7 mmol) and 2-bromoethyl methyl ether (0.98
mL, 1.45 g, 10.5 mmol), reacting in THF at rt overnight. Column
eluent 5 → 10 → 20% CH2Cl2/toluene. Product obtained as a
colorless oil (1.07 g, 59%). δH (400 MHz, DMSO-d6) 7.59 (d, 1H, J =
2.0), 7.52 (d, 1H, J = 9.0), 7.43 (d, 1H, J = 3.0), 7.12 (dd, 1H, J = 2.0,
9.0), 6.42 (dd, 1H, J = 0.5, 3.0), 4.33 (t, 2H, J = 5.5), 3.63 (t, 2H, J =
5.5), 3.20 (s, 3H). δC (101 MHz, DMSO-d6) 134.9, 131.2, 129.6,
124.1, 121.3, 119.9, 112.0, 100.8, 71.5, 58.5, 46.0. m/z (ES) 210
(100%; [M(35Cl) + H]+), 212 (25%; [M(37Cl) + H]+). HRMS, found
210.0694 (C11H13ClNO requires 210.0686). HPLC purity 99.4%.
6-Chloro-1-isopropyl-1H-indole 67d. Prepared from 6-chloroin-
dole (3.03 g, 20 mmol) and 2-iodopropane (2.4 mL, 4.1 g, 24 mmol),
reacting at rt for 48 h. Column eluent 5 → 10 → 20% CH2Cl2/hexane.
1-(1-Methoxypropan-2-yl)-1H-indole 4g. Prepared from indole
(2.75 g, 23.5 mmol) and 1-methoxypropan-2-yl 4-methylbenzenesul-
fonate (6.0 mL, 6.9 g, 28 mmol), reacting at 55 °C overnight. Column
eluent 40 → 50 → 67 → 100% toluene/hexane then 10% CH2Cl2/
toluene. Title compound obtained as a pale-yellow oil (3.11 g, 70%).
δH (250 MHz, DMSO-d6) 7.57−7.44 (m, 3H), 7.12 (dt, 1H, J = 1.5,
7.5), 7.05−6.97 (m, 1H), 6.45 (d, 1H, J = 3.0), 4.87−4.72 (m, 1H),
3.71−3.54 (m, 2H), 3.20 (s, 3H), 1.43 (d, 3H, J = 7.0). δC (62.8 MHz,
DMSO-d6) 136.2, 128.5, 125.8, 121.3, 120.8, 119.4, 110.3, 101.4, 75.6,
58.8, 50.6, 18.0. m/z (ES) 190 ([M + H]+). HRMS, found 190.1224
(C12H16NO requires 190.1232). HPLC purity 99.4%.
1-Isobutyl-1H-indole 4h. Prepared from indole (2.34 g, 20 mmol)
and 1-iodo-2-methylpropane (2.76 mL, 4.42 g, 24 mmol), reacting at
rt for 1 h. Column eluent 5 → 10 → 20% CH2Cl2/light petroleum.
Product obtained as a thick, colorless oil (0.54 g, 16%). δH (400 MHz,
DMSO-d6) 7.55 (d, 1H, J = 8.0), 7.46 (dd, 1H, J = 0.5, 8.0), 7.33 (d,
1H, J = 3.0), 7.14−7.09 (m, 1H), 7.03−6.98 (m, 1H), 6.43 (dd, 1H, J
= 0.5, 3.0), 3.96 (d, 2H, J = 7.5), 2.17−2.05 (m, 1H), 0.84 (d, 6H, J =
N
DOI: 10.1021/acs.jmedchem.5b01312
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Product obtained as a colorless oil (1.88 g, 49%). δH (400 MHz,
CDCl3) 7.56 (d, 1H, J = 8.5), 7.42−7.40 (m, 1H), 7.25 (d, 1H, J =
3.5), 7.10 (dd, 1H, J = 2.0, 8.5), 6.53 (dd, 1H, J = 0.5, 3.5), 4.63
(septet, 1H, J = 6.5), 1.56 (d, 6H, J = 6.5). δC (101 MHz, CDCl3)
135.9, 127.3, 127.1, 124.3, 121.8, 120.0, 109.5, 101.4, 47.3, 22.7. m/z
(ES) 194 (100%; [M(35Cl) + H]+), 196 (35%; [M(37Cl) + H]+).
HRMS, found 194.0730 (C11H13ClN requires 194.0737). HPLC
purity 98.7%.
7-Chloro-1-isopropyl-1H-indole 67e. Prepared from 7-chloroin-
dole (1.52 g, 10 mmol) and 2-iodopropane (1.2 mL, 2.0 g, 12 mmol),
reacting at rt overnight. Column eluent 0 → 1→2.5 → 5→10%
CH2Cl2/hexane. Product obtained as a colorless oil (0.50 g, 26%). δH
(400 MHz, CDCl3) 7.55 (dd, 1H, J = 1.0, 8.0), 7.34 (d, 1H, J = 3.5),
7.19 (dd, 1H, J = 1.0, 8.0), 7.02 (t, 1H, J = 8.0), 6.58 (d, 1H, J = 3.5),
5.75 (septet, 1H, J = 6.5), 1.56 (d, 6H, J = 6.5). δC (101 MHz, CDCl3)
131.8, 131.1, 124.9, 123.6, 119.9, 119.7, 116.5, 102.2, 47.9, 23.9. m/z
(ES) 194 (100%; [M(35Cl) + H]+), 196 (35%; [M(37Cl) + H]+).
HRMS, found 194.0731 (C11H13ClN requires 194.0737). HPLC
purity 99.1%.
4-Methoxy-1-isopropyl-1H-indole 67f. Prepared from 4-methox-
yindole (1.25 g, 8.5 mmol) and 2-iodopropane (1.02 mL, 1.73 g, 10.2
mmol), reacting at rt overnight. Column eluent 10 → 20 → 30 → 35%
CH2Cl2/hexane. Title compound obtained as a slightly opaque,
colorless oil (0.61 g, 38%). δH (400 MHz, CDCl3) 7.24−7.16 (m, 2H),
7.06 (d, 1H, J = 8.5), 6.68 (d, 1H, J = 3.0), 6.57 (d, 1H, J = 7.5), 4.69
(septet, 1H, J = 6.5), 4.01 (s, 3H), 1.56 (d, 6H, J = 6.5). δC (101 MHz,
CDCl3) 153.5, 137.0, 122.1, 122.0, 119.1, 103.1, 99.1, 98.4, 55.3, 47.3,
22.8. m/z (ES) 190 ([M + H]+). HRMS, found 190.1235 (C12H16NO
requires 190.1232). HPLC purity 88.7%.
5-Methoxy-1-isopropyl-1H-indole 67g. Prepared from 5-methox-
yindole (1.55 g, 10.5 mmol) and 2-iodopropane (1.26 mL, 2.14 g, 12.6
mmol), reacting at rt overnight. Column eluent 10 → 20 → 25 → 33
→ 40% CH2Cl2/hexane. Product initially obtained as an oil which
crystallized on standing to an off-white solid (0.72 g, 36%). δH (400
MHz, CDCl3) 7.31 (d, 1H, J = 9.0), 7.24 (d, 1H, J = 3.0), 7.14 (d, 1H,
J = 2.5), 6.92 (dd, 1H, J = 2.5, 9.0), 6.48 (d, 1H, J = 3.0), 4.66 (septet,
1H, J = 7.0), 3.89 (s, 3H), 1.55 (d, 6H, J = 7.0). δC (101 MHz, CDCl3)
153.9, 130.9, 128.9, 124.1, 111.6, 110.2, 102.6, 100.7, 55.9, 47.2, 22.8.
m/z (ES) 190 ([M + H]+). HRMS, found 190.1240 (C12H16NO
requires 190.1232). HPLC purity 96.6%; mp 65−66 °C.
5-Methoxy-1-(2-methoxyethyl)-1H-indole 67h. Prepared from 5-
methoxyindole (1.47 g, 10 mmol) and 2-bromoethyl methyl ether
(1.13 mL, 1.67 g, 12 mmol), reacting in THF at 35 °C overnight.
Column eluent 5 → 10 → 17.5 → 20% EtOAc/hexane. Title
compound obtained as a pale-yellow oil (1.36 g, 66%). δH (400 MHz,
DMSO-d6) 7.37 (d, 1H, J = 9.0), 7.29 (d, 1H, J = 3.0), 7.05 (d, 1H, J =
2.5), 6.77 (dd, 1H, J = 2.5, 9.0), 6.33 (dd, 1H, J = 0.5, 3.0), 4.28 (t, 2H,
J = 5.5), 3.75 (s, 3H), 3.62 (t, 2H, J = 5.5), 3.21 (s, 3H). δC (101 MHz,
DMSO-d6) 153.9, 131.6, 129.9, 128.9, 111.5, 111.0, 102.5, 100.6, 71.6,
58.5, 55.8, 45.9. m/z (ES) 206 ([M + H]+). HRMS, found 206.1188
(C12H16NO2 requires 206.1181). HPLC purity 99.3%.
6-Methoxy-1-isopropyl-1H-indole 67i. Prepared from 6-methox-
yindole (1.55 g, 10.5 mmol) and 2-iodopropane (1.26 mL, 2.14 g, 12.6
mmol), reacting at 35 °C overnight. Column eluent 20 → 30 → 40%
CH2Cl2/hexane. Product obtained as a pale-yellow oil (0.72 g, 36%)
which crystallized slowly on storage at −20 °C. δH (400 MHz, CDCl3)
7.54 (d, 1H, J = 8.5), 7.16 (d, 1H, J = 3.0), 6.88 (d, 1H, J = 2.0), 6.82
(dd, 1H, J = 2.0, 8.5), 6.49 (d, 1H, J = 3.0), 4.63 (septet, 1H, J = 6.5),
3.92 (s, 3H), 3.28 (d, 6H, J = 6.5). δC (101 MHz, CDCl3) 156.0,
136.2, 122.9, 122.4, 121.5, 109.1, 101.1, 93.4, 55.8, 46.9, 22.7. m/z
(ES) 190 ([M + H]+). HRMS, found 190.1237 (C12H16NO requires
190.1232). HPLC purity 98.2%. mp 30−31 °C.
130.7, 125.5, 123.7, 119.5, 113.8, 102.4, 101.6, 55.4, 48.9, 24.0. m/z
(ES) 190 ([M + H]+). HRMS, found 190.1236 (C12H16NO requires
190.1232). HPLC purity 94.8%. mp 36−38 °C.
1-Isopropyl-1H-pyrrolo[3,2-b]pyridine 67k. Prepared from 4-
azaindole (1.18 g, 10 mmol) and 2-iodopropane (1.20 mL, 2.04 g,
12 mmol), reacting at rt for 90 min. Column eluent 2.5 → 5→7.5%
MeOH/CH2Cl2. Product obtained as a pale-pink oil (0.56 g, 35%). δH
(400 MHz, CDCl3) 8.47 (dd, 1H, J = 1.0, 4.5), 7.67 (dt, 1H, J = 1.0,
8.5), 7.46 (d, 1H, J = 3.5), 7.11 (dd, 1H, J = 4.5, 8.5), 6.73 (d, 1H, J =
3.5), 4.66 (septet, 1H, J = 6.5), 1.55 (d, 6H, J = 6.5). δC (101 MHz,
CDCl3) 147.0, 143.1, 128.4, 127.0, 116.5, 116.0, 102.3, 47.6, 22.7. m/z
(ES) 161 ([M + H]+). HRMS, found 161.1072 (C10H13N2 requires
161.1079). HPLC purity 98.0%.
1-Isopropyl-1H-pyrrolo[3,2-c]pyridine 67l. Prepared from 5-
azaindole (2.95 g, 25 mmol) and 2-iodopropane (3.0 mL, 5.1 g, 30
mmol), reacting at rt for 2 h. Column eluent 0 → 5→10 → 12.5%
MeOH/CH2Cl2. Product obtained as a pale-orange oil (1.78 g, 44%)
which crystallized on standing to give an off-white solid. δH (250 MHz,
DMSO-d6) 8.81 (d, 1H, J = 1.0), 8.18 (d, 1H, J = 6.0), 7.59 (d, 1H, J =
3.0), 7.51 (d, 1H, J = 6.0), 6.62−6.59 (m, 1H), 4.77 (septet, 1H, J =
6.5), 1.44 (d, 6H, J = 6.5). δC (62.8 MHz, DMSO-d6) 143.7, 140.3,
138.9, 126.6, 125.5, 105.8, 100.9, 47.4, 22.8. m/z (ES) 161 ([M +
H]+). HRMS, found 161.1077 (C10H13N2 requires 161.1079). HPLC
purity 99.7%. mp 71−72 °C.
1-Isopropyl-1H-pyrrolo[2,3-c]pyridine 67m. Prepared from 6-
azaindole (2.95 g, 25 mmol) and 2-iodopropane (3.0 mL, 5.1 g, 30
mmol), reacting at rt for 75 min. Column eluent 0 → 5→10% MeOH/
CH2Cl2. Title compound obtained as a yellow oil (1.88 g, 47%) which
crystallized slowly, forming a waxy, off-white solid on storage at −20
°C. δH (250 MHz, DMSO-d6) 8.92 (s, 1H), 8.11 (d, 1H, J = 5.5), 7.72
(s, 1H, J = 3.0), 7.51 (dd, 1H, J = 1.0, 5.0), 6.50 (dd, 1H, J = 1.0, 3.0),
4.89 (septet, 1H, J = 6.5), 1.48 (d, 6H, J = 6.5). δC (62.8 MHz,
DMSO-d6) 138.3, 133.7, 132.85, 132.78, 129.1, 115.3, 100.6, 47.8,
23.1. m/z (ES) 161 ([M + H]+). HRMS, found 161.1083 (C10H13N2
requires 161.1079). HPLC purity 99.5%. mp 35−36 °C.
1-Isopropyl-1H-pyrrolo[2,3-b]pyridine 67n. Prepared from 7-
azaindole (2.95 g, 25 mmol) and 2-iodopropane (3.0 mL, 5.1 g, 30
mmol), reacting at rt for 2.5 h. Column eluent 33 → 50 → 67 → 100%
CH2Cl2/hexane then 1 → 2% MeOH/CH2Cl2. Title compound
obtained as a slightly cloudy, colorless oil (2.20 g, 55%). δH (400 MHz,
DMSO-d6) 8.24 (dd, 1H, J = 1.5, 4.5), 7.94 (dd, 1H, J = 1.5, 8.0), 7.66
(d, 1H, J = 3.5), 7.07 (dd, 1H, J = 4.5, 8.0), 6.48 (d, 1H, J = 3.5), 5.10
Hz (septet, 1H, J = 7.0), 1.46 (d, 6H, J = 7.0). δC (101 MHz, DMSO-
d6) 146.9, 142.5, 128.8, 126.0, 120.6, 116.0, 99.6, 45.5, 22.9. m/z (ES)
161 ([M + H]+). HRMS, found 161.1077 (C10H13N2 requires
161.1079). HPLC purity 89.6%.
1-(2-Methoxyethyl)-1H-pyrrolo[2,3-b]pyridine 67o. Prepared from
7-azaindole (1.18 g, 10 mmol) and 2-bromoethyl methyl ether (1.13
mL, 1.67 g, 12 mmol), reacting at rt for 5 h. Column eluent 0 → 1→2
→ 4→5% MeOH/CH2Cl2. Product obtained as a pale-yellow oil (1.16
g, 66%). δH (400 MHz, CDCl3) 8.33 (dd, 1H, J = 1.5, 4.5), 7.92 (dd,
1H, J = 1.5, 8.0), 7.35 (d, 1H, J = 3.0), 7.07 (dd, 1H, J = 4.5, 8.0), 6.47
(d, 1H, J = 3.0), 4.50 (t, 2H, J = 5.0), 3.77 (t, 2H, J = 5.0), 3.35 (s,
3H). δC (101 MHz, CDCl3) 147.4, 142.6, 129.0, 128.7, 120.7, 115.7,
99.3, 71.7, 58.9, 44.3. m/z (ES) 177 ([M + H]+). HRMS, found
177.1023 (C10H13N2O requires 177.1028). HPLC purity 98.2%.
1-Cyclopentyl-1H-indole 4k. Indole (2.57 g, 21.9 mmol), DMAP
(7.70 g, 65.7 mmol), cyclopentylboronic acid (5.00 g, 43.9 mmol), and
Cu(OAc)2 (0.40 g, 2.2 mmol) were added to a flame-dried, 3-neck
flask equipped with a reflux condenser. Anhydrous toluene (40 mL)
was introduced via cannula, then sodium hexamethyldisilazane (0.6 M
solution in toluene, 36.5 mL, 21.9 mmol) was added to the resultant
deep-blue suspension. A stream of dry air was introduced, bubbling
through the reaction mixture, which was then heated to 95 °C and
maintained at this temperature overnight. EtOAc (100 mL) and 1 M
HCl (200 mL) were added and the mixture shaken thoroughly then
vacuum filtered. The layers were separated and the aqueous phase
extracted once more with EtOAc. The combined extracts were then
dried over MgSO4, filtered, and evaporated, leaving a thick-brown oil,
which was purified by flash column chromatography on silica gel,
7-Methoxy-1-isopropyl-1H-indole 67j. Prepared from 7-methox-
yindole (1.63 mL, 1.84 g, 12.5 mmol) and 2-iodopropane (1.5 mL, 2.6
g, 15 mmol), reacting at rt overnight. Column eluent 2.5 → 5→10%
EtOAc/hexane. Title compound obtained as a slightly opaque,
colorless oil (0.92 g, 39%) which crystallized slowly on storage at
−20 °C. δH (400 MHz, CDCl3) 7.31−7.25 (m, 2H), 7.05 (t, 1H, J =
8.0), 6.69 (d, 1H, J = 8.0), 6.54 (d, 1H, J = 3.0), 5.51 (septet, 1H, J =
6.5), 4.00 (s, 3H), 1.54 (d, 6H, J = 6.5). δC (101 MHz, CDCl3) 147.7,
O
DOI: 10.1021/acs.jmedchem.5b01312
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
eluted with 5 → 10 → 15% CH2Cl2/light petroleum, yielding the title
compound as a pale-yellow oil (1.37 g, 34%). δH (400 MHz, DMSO-
d6) 7.56−7.48 (m, 2H), 7.44 (d, 1H, J = 3.0), 7.15−7.10 (m, 1H),
7.05−6.99 (m, 1H), 6.44 (d, 1H, J = 3.0), 4.87 (quintet, 1H, J = 7.0),
2.18−2.08 (m, 2H), 1.90−1.78 (m, 4H), 1.77−1.65 (m, 2H). δC (101
MHz, DMSO-d6) 136.3, 128.7, 125.7, 121.3, 120.8, 119.4, 110.5, 101.2,
56.6, 32.6, 24.1. m/z (ES) 186 ([M + H]+). HRMS, found 186.1274
(C11H16N requires 186.1283). HPLC purity 98.8%.
7.43 (d, 1H, J = 8.0), 7.16−7.11 (m, 1H), 7.07−7.02 (m, 1H), 6.50
(dd, 1H, J = 0.5, 3.0), 5.46 (q, 1H, J = 7.5), 4.12 (q, 2H, J = 7.0), 1.73
(d, 3H, J = 7.5), 1.15 (t, 3H, J = 7.0). δC (101 MHz, DMSO-d6) 171.6,
136.3, 128.6, 126.9, 121.7, 120.9, 119.8, 110.2, 102.1, 61.5, 53.5, 17.7,
14.5. m/z (ES) 218 ([M + H]+). HRMS, found 218.1189 (C13H16NO2
requires 218.1181). HPLC purity 98.1%.
Ethyl 2-(1H-Indol-1-yl)-2-methylpropanoate 71. A solution of
intermediate 70 (7.93 g, 36.5 mmol) in dry THF (100 mL) was cooled
to −78 °C, then LDA (2.0 M solution in THF, 22 mL, 44 mmol) was
added over 10 min. The mixture was stirred at −78 °C for 45 min, at
which point methyl iodide (3.77 mL, 8.60 g, 60.6 mmol) was added,
then the reaction continued and allowed to warm slowly to rt
overnight before being quenched by dropwise addition of satd NH4Cl
then EtOH (∼5 mL each). After an additional 30 min stirring, the
mixture was evaporated under vacuum and the residue extracted with
Et2O and water. The aqueous layer was extracted again with Et2O and
the combined organic solutions washed twice with brine, dried over
MgSO4, filtered, and evaporated to give a viscous orange oil.
Additional purification by flash column chromatography on silica gel,
eluted with 20 → 30 → 40 → 50 → 60% CH2Cl2/hexane, gave the
title compound as a pale-yellow oil which crystallized on standing to a
greasy, off-white solid (7.22 g, 86%). δH (400 MHz, DMSO-d6) 7.58
(dt, 1H, J = 1.0, 8.0), 7.52 (d, 1H, J = 3.5), 7.13−7.09 (m, 2H), 7.07−
7.00 (m, 1H), 6.47 (d, 1H, J = 3.5), 4.12 (q, 2H, J = 7.0), 1.81 (s, 6H),
1.06 (t, 3H, J = 7.0). δC (101 MHz, DMSO-d6) 173.7, 135.1, 129.8,
126.0, 121.6, 121.3, 119.7, 111.4, 101.2, 61.8, 60.9, 25.9, 14.3. m/z
(ES) 232 ([M + H]+). HRMS, found 232.1332 (C14H18NO2 requires
232.1338). HPLC purity 97.9%.
2-(1H-Indol-1-yl)-2-methylpropan-1-ol 72. A solution of ester 71
(7.12 g, 30.8 mmol) in anhydrous THF (85 mL) was added over 15
min to a suspension of LiAlH4 (1.28 g, 33.9 mmol) in the same solvent
(40 mL) at −78 °C (some exotherm observed over the addition
period). The reaction temperature was brought back down to −78 °C
then stirring continued overnight, allowing for gradual warming to rt.
The mixture was cooled to −78 °C once more then quenched by very
cautious, dropwise addition of satd NH4Cl, after which it was allowed
to warm slowly to rt once more. The solvent was removed by
evaporation under vacuum, then the residue extracted with Et2O and
water. The mixture was filtered under vacuum to remove suspended
solids, then the layers separated and the aqueous phase extracted with
more Et2O. The combined organic extracts were washed with water
then brine, dried over Na2SO4, filtered, and evaporated. Purification by
flash column chromatography on silica gel, eluted with 0 → 1→2 →
3% MeOH/CH2Cl2, afforded the title compound as a yellow oil which
slowly crystallized on standing to a pale-yellow solid (4.56 g, 78%). δH
(400 MHz, DMSO-d6) 7.69−7.64 (m, 1H), 7.56−7.51 (m, 1H), 7.43
(d, 1H, J = 3.5), 7.10−7.05 (m, 1H), 7.02−6.97 (m, 1H), 6.38 (dd,
1H, J = 1.0, 3.5), 5.08 (t, 1H, J = 4.5), 3.82 (d, 2H, J = 4.5), 1.62 (s,
6H), δC (101 MHz, DMSO-d6) 135.1, 130.3, 127.6, 121.0, 120.7,
118.9, 113.7, 100.1, 67.6, 60.1, 25.2, m/z (ES) 190 ([M + H]+),
HRMS, found 190.1232 (C12H16NO requires 190.1232), HPLC purity
99.3%, mp 68−71 °C.
1-(1H-Indol-1-yl)-2-methylpropan-2-ol 69. A solution of indole
(2.81 g, 24 mmol) in anhydrous DMF (90 mL) was cooled to 0 °C,
then sodium hydride (60% dispersion in mineral oil, 0.96 g, 24 mmol)
added in small portions over 15 min. After a further 10 min stirring at
0 °C, isobutene oxide (533 μL, 433 mg, 6 mmol) was added dropwise
and stirring continued for an additional 50 min, at which point the
cooling was removed. The reaction mixture was stirred at rt for 5.5 h at
rt then 4 h at 30 °C, followed by addition of more isobutene oxide (89
μL, 72 mg, 1 mmol), after which stirring was continued at 30 °C
overnight. The reaction was quenched by dropwise addition of brine,
then evaporated under vacuum. The resultant orange slurry was
partitioned between Et2O and water (∼100 mL each). AcOH (0.2
mL) was added, then the mixture extracted and the layers separated.
The aqueous phase was extracted again with Et2O and the combined
extracts washed three times with brine, dried over MgSO4, filtered, and
evaporated. Further purification was carried out by flash column
chromatography on silica gel, eluted with 0 → 1→2% MeOH/CH2Cl2,
affording the title compound as a pale-pink oil (0.98 g, 74%). δH (400
MHz, DMSO-d6) 7.55−7.50 (m, 2H), 7.33 (d, 1H, J = 3.0), 7.12−7.07
(m, 1H), 6.98 (dt, 1H, J = 1.0, 7.5), 6.42 (dd, 1H, J = 0.5, 3.0), 4.66 (s,
1H), 4.07 (s, 2H), 1.09 (s, 6H). δC (101 MHz, DMSO-d6) 137.3,
130.6, 128.1, 121.1, 120.5, 119.0, 111.1, 100.6, 70.9, 56.6, 27.9. m/z
(ES) 190 ([M + H]+). HRMS, found 190.1237 (C12H16NO requires
190.1232). HPLC purity 99.4%.
1-(2-Methoxy-2-methylpropyl)-1H-indole 4k. Alcohol 69 (874 mg,
4.62 mmol) was dissolved in anhydrous THF (100 mL) at 0 °C, then
sodium hydride (60% dispersion in mineral oil, 231 mg, 5.77 mmol)
added in portions over 5 min. Methyl iodide (359 μL, 819 mg, 5.77
mmol) was introduced 10 min later, then the reaction mixture stirred
at 0 °C for 4.5 h then rt for 4.5 h. TLC analysis (3:1 CH2Cl2/hexane)
indicated reaction was incomplete, so additional methyl iodide (72 μL,
165 mg, 1.16 mmol) was added and stirring continued at 35 °C
overnight. The reaction was then quenched by dropwise addition of
satd NH4Cl, evaporated to dryness and the residue extracted between
EtOAc and water. The aqueous phase was extracted again with EtOAc,
and the combined organic extracts washed with satd NH4Cl, dried
over MgSO4, filtered, and evaporated. The oily residue was purified by
flash column chromatography on silica gel, eluted with 10 → 20 → 33
→ 50% CH2Cl2/hexane, giving the title compound as a viscous, pale-
yellow oil (492 mg, 52%). δH (250 MHz, DMSO-d6) 7.56−7.49 (m,
2H), 7.27 (d, 1H, J = 3.0), 7.11 (dt, 1H, J = 1.0, 7.0), 6.99 (dt, 1H, J =
1.0, 7.0), 6.43 (dd, 1H, J = 1.0, 3.0), 4.15 (s, 2H), 3.16 (s, 3H), 1.07 (s,
6H). δC (62.8 MHz, DMSO-d6) 137.4, 130.4, 128.1, 121.3, 120.6,
119.1, 111.0, 100.9, 75.9, 53.9, 49.5, 22.8. m/z (ES) 204 ([M + H]+).
HRMS, found 204.1384 (C13H18NO requires 204.1388). HPLC purity
99.3%.
1-(1-Methoxy-2-methylpropan-2-yl)-1H-indole 4l. Sodium hy-
dride (60% dispersion in mineral oil, 248 mg, 6.2 mmol) was added
in portions over 5 min to a solution of alcohol 72 (1.01 g, 4.97 mmol)
in anhydrous THF (50 mL) at 0 °C. After 5 min more, methyl iodide
(467 μL, 1.06 g, 7.5 mmol) was added. The mixture was stirred at 0 °C
for 30 min then rt for 3 h and quenched by dropwise addition of satd
NH4Cl. The mixture was evaporated under vacuum, extracted with
Et2O and water, the aqueous phase re-extracted with Et2O, and the
combined extracts dried over Na2SO4, filtered, and evaporated. Flash
column chromatography on silica gel, eluted with 20 → 30 → 40%
CH2Cl2/hexane, provided the title compound as a colorless oil (856
mg, 79%). δH (400 MHz, DMSO-d6) 7.71−7.68 (m, 1H), 7.54 (d, 1H,
J = 8.0), 7.37 (d, 1H, J = 3.5), 7.11−7.06 (m, 1H), 7.03−6.98 (m, 1H),
6.39 (dd, 1H, J = 0.5, 3.5), 3.75 (s, 2H), 3.18 (s, 3H), 1.65 (s, 6H). δC
(101 MHz, DMSO-d6) 135.0, 130.2, 127.5, 121.1, 120.9, 119.1, 113.6,
100.4, 78.2, 59.00, 58.97, 25.5. m/z (ES) 204 ([M + H]+). HRMS,
found 204.1381 (C13H18NO requires 204.1388). HPLC purity 99.6%.
Ethyl 2-(1H-Indol-1-yl)propanoate 70. Indole (5.86 g, 50 mmol)
was dissolved in anhydrous DMF (175 mL) and the solution cooled to
0 °C, after which sodium hydride (60% dispersion in mineral oil, 2.20
g, 55 mmol) was added portionwise over 15 min. Stirring was
continued for a further 15 min, then ethyl 2-bromopropionate (7.14
mL, 9.96 g, 55 mmol) was added. The reaction was then maintained at
40 °C for 5 h and continued at rt overnight before being quenched by
careful, dropwise addition of water. The mixture was evaporated under
vacuum and the residue extracted with Et2O and water. The aqueous
layer was extracted a second time with Et2O then the combined
extracts washed twice with water and twice with brine before drying
over Na2SO4 and evaporating to leave an orange oil. This crude
material was subjected to flash column chromatography on silica gel,
eluted with 20 → 33 → 50 → 65% CH2Cl2/hexane, to provide the
title compound as a somewhat cloudy, pale-yellow oil (8.03 g, 74%).
δH (400 MHz, DMSO-d6) 7.56 (d, 1H, J = 8.0), 7.47 (d, 1H, J = 3.0),
P
DOI: 10.1021/acs.jmedchem.5b01312
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Ethyl 2-(2-(1H-Indol-1-yl)-2-methylpropoxy)acetate 4m. Ethyl
bromoacetate (665 μL, 1.00 g, 6.0 mmol) and alcohol 72 (945 mg,
5.0 mmol) were dissolved in dry DMF, then sodium hydride (60%
dispersion in mineral oil, 0.24 g, 6.0 mmol) was added slowly, in small
portions, over 40 min. After stirring overnight at rt, additional ethyl
bromoacetate (277 μL, 417 mg, 2.5 mmol) was added, followed by
sodium hydride (60% dispersion in mineral oil, 100 mg, 2.5 mmol)
over 30 min. The mixture was stirred at rt for a further 2 h then
quenched by dropwise addition of satd NH4Cl and evaporated under
vacuum. The residue was extracted with EtOAc and satd NH4Cl (100
mL each), then the aqueous layer extracted three more times with
EtOAc. The combined organic extracts were then washed twice with
water, then brine, dried over Na2SO4, filtered, and concentrated under
vacuum. Flash column chromatography on silica gel was carried out,
eluted with 20 → 33 → 50% CH2Cl2/toluene, affording the title
compound as an orange gum (400 mg, 29%). δH (400 MHz, CDCl3)
7.65 (t, 2H, J = 8.5), 7.42 (d, 1H, J = 3.0), 7.21−7.15 (m, 1H), 7.15−
7.09 (m, 1H), 6.51 (d, 1H, J = 3.0), 4.19 (q, 2H, J = 7.0), 3.99 (s, 2H),
3.95 (s, 2H), 1.82 (s, 6H), 1.27 (t, 3H, J = 7.0). δC (101 MHz, CDCl3)
168.6, 133.0, 128.4, 124.4, 119.4, 118.9, 117.1, 111.1, 98.7, 66.6, 59.0,
56.9, 23.4, 12.3. m/z (ES) 276 ([M + H]+). HRMS, found 276.1610
(C16H22NO3 requires 276.1600. HPLC purity 92.0%. Additional
material from mixed column fractions was further purified by
preparative HPLC, using a gradient of 40−70% MeCN/water, to
give an additional portion of the title compound as a thick, orange−
brown oil (40 mg, 3%; combined yield 440 mg, 32%).
Methyl 3-(1H-Indol-1-yl)propanoate 4n. Methyl acrylate (1.35
mL, 1.29 g, 15 mmol) and DBU (772 μL, 5 mmol) were added to a
solution of indole (1.17 g, 10 mmol) in anhydrous MeCN (50 mL).
The mixture was heated at 50 °C for 6 h, rt for 48 h, then 50 °C for 6
h and evaporated to dryness. EtOAc and water (50 mL) each were
added, and after the residue had fully dissolved, the aqueous layer was
adjusted to pH 6 with 1 M HCl and the mixture extracted. The
aqueous layer was extracted with additional EtOAc (3 × 30 mL), then
the combined organic solutions washed twice with water, then brine,
dried over Na2SO4, filtered, and evaporated. Flash column
chromatography on silica gel, eluted with 50 → 67 → 100%
toluene/hexane, yielded the title compound as an oil (865 mg, 43%).
δH (400 MHz, CDCl3) 7.66 (dt, 1H, J = 1.0, 8.0), 7.38 (dd, 1H, J =
0.5, 8.5), 7.32−7.12 (m, 3H), 4.49 (t, 2H, J = 7.0), 3.70 (s, 3H), 2.86
(t, 2H, J = 7.0). δC (101 MHz, CDCl3) 171.7, 135.6, 128.7, 128.0,
121.7, 121.1, 119.5, 109.1, 101.6, 52.0, 41.9, 34.8. m/z (ES) 204 ([M +
H]+). HRMS, found 204.1033 (C12H14NO2 requires 204.1025).
HPLC purity 97.8%.
2-Isopropoxy-5-nitropyridine 74. Sodium hydride (60% dispersion
in mineral oil, 2.40 g, 60 mmol) was added in portions to anhydrous
propan-2-ol (150 mL), then stirring of the suspension continued for
30 min. 2-Chloro-5-nitropyridine (9.51 g, 60 mmol) was then added
and the reaction allowed to continue for 1 h before quenching by
cautious, dropwise addition of water. The mixture was then evaporated
under vacuum and the residue purified by flash column chromatog-
raphy on silica gel, eluted with 1:19 EtOAc/hexane, affording the title
compound as pale-yellow crystals (8.6 g, 64%). δH (400 MHz, DMSO-
d6) 9.25 (d, 1H, J = 3.0), 8.45 (dd, 1H, J = 3.0, 9.0), 6.96 (d, 1H, J =
9.0), 5.37 (septet, 1H, J = 6.0), 1.34 (d, 6H, J = 6.0). δC (101 MHz,
DMSO-d6) 166.7, 145.2, 139.5, 135.1, 112.2, 70.6, 22.0. m/z (ES) 183
([M + H]+). HRMS, found 183.0762 (C8H11N2O3 requires 183.0770).
mp 47−50 °C.
2-tert-Butoxy-5-nitropyridine 75. 2-Chloro-5-nitropyridine (2.38 g,
15 mmol) and tert-butanol (3.6 mL, 45 mmol) were dissolved in dry
DMF (20 mL), then sodium hydride (60% dispersion in mineral oil,
0.72 g, 18 mmol) added slowly in small portions. During the addition,
the reaction mixture was cooled in ice to control the observed
exotherm. After addition was complete, stirring was continued at rt for
4 h, then water added dropwise to quench the reaction. The mixture
was evaporated under vacuum and the residue purified by flash column
chromatography on silica gel, eluted with 2 → 4→6 → 10% EtOAc/
hexane, to give the title compound as a dark-brown oil (1.34 g, 46%).
δH (400 MHz, CDCl3) 9.06 (d, 1H, J = 3.0), 8.30 (dd, 1H, J = 3.0,
9.0), 6.71 (d, 1H, J = 9.0), 1.65 (s, 9H). δC (101 MHz, CDCl3) 167.3,
144.3, 138.6, 133.4, 113.0, 82.8, 28.4. m/z (GC-MS, EI) 196 (M+).
HRMS, found 196.0838 (C9H12N2O3 requires 196.0842). HPLC
purity 91.9%.
Ethyl 2-(5-Nitropyridin-2-yloxy)acetate 76. Sodium hydride (60%
dispersion in mineral oil, 0.80 g, 20 mmol) was added in small
portions, slowly over 1.75 h, to a solution of 2-chloro-5-nitropyridine
(3.18 g, 20 mmol) and ethyl glycolate (1.58 mL, 1.73 g, 16.6 mmol) in
anhydrous THF (100 mL). After stirring at rt for 2 h more, the
reaction was carefully quenched by dropwise addition of water, then
the mixture evaporated under vacuum. The residue was partitioned
between EtOAc and water, the layers separated, and the aqueous phase
extracted with additional EtOAc. The combined extracts were washed
with brine, dried over Na2SO4, filtered, and evaporated, then the
residue purified by flash column chromatography on silica gel, eluted
with 5 → 10 → 15 → 20% EtOAc/hexane, to provide the title
compound as a pale-yellow solid (2.96 g, 65%). δH (400 MHz, CDCl3)
9.05 (d, 1H, J = 3.0), 8.43 (dd, 1H, J = 3.0, 9.0), 7.01 (d, 1H, J = 9.0),
5.01 (s, 2H), 4.27 (q, 2H, J = 7.0), 1.31 (t, 3H, J = 7.0). δC (101 MHz,
CDCl3) 168.1, 165.7, 144.4, 134.4, 111.4, 63.5, 61.5, 14.1. m/z (ES)
227 ([M + H]+). HRMS, found 227.0675 (C9H11N2O5 requires
227.0668). HPLC purity 99.2%. mp 51−52 °C.
Hydrogenation of Nitro-compounds 74−76 to Amines 77−79:
General Procedure. A 0.1 M solution of the nitro-compound in
MeOH was hydrogenated over 10% Pd/C, using a ThalesNano H-
Cube reactor in full H2 mode at a flow rate of 0.5 mL min−1.
Evaporation of the solvent and further drying under vacuum gave the
amine products, which were used without further purification.
6-Isopropoxypyridin-3-amine 77. Hydrogenation of 74 (767 mg,
4.2 mmol) gave the title compound as a peach-colored oil (397 mg,
62%). δH (400 MHz, DMSO-d6) 7.49 (d, 1H, J = 3.0), 6.97 (dd, 1H, J
= 3.0, 9.0), 6.46 (d, 1H, J = 9.0), 4.71 (septet, 1H, J = 6.0), 1.21 (d,
6H, J = 6.0). δC (101 MHz, DMSO-d6) 166.7, 145.2, 139.5, 135.1,
112.2, 70.6, 20.1. m/z (ES) 153 ([M + H]+). HRMS, found 153.1022
(C8H13N2O3 requires 153.1028).
6-tert-Butoxypyridin-3-amine 78. Hydrogenation of 75 (640 mg,
3.86 mmol) gave the title compound as a brown solid (475 mg, 88%).
δH (400 MHz, CDCl3) 7.70 (d, 1H, J = 3.0), 6.99 (dd, 1H, J = 3.0,
9.0), 6.58 (d, 1H, J = 9.0), 3.27 (br s, 2H), 1.49 (s, 9H). δC (101 MHz,
CDCl3) 156.9, 137.2, 133.5, 126.5, 115.0, 78.9, 28.9. m/z (ES) 167
([M + H]+). HRMS, found 167.1188 (C9H15N2O requires 167.1184).
HPLC purity 90.4%. mp 77−78 °C (decomp).
Ethyl 2-(5-Aminopyridin-2-yloxy)acetate 79. Hydrogenation of 76
(1.46 g, 6.46 mmol) provided the title compound as an off-white solid
(1.21 g, 95%). δH (400 MHz, DMSO-d6) 7.42 (dd, 1H, J = 0.5, 3.0),
7.03 (dd, 1H, J = 3.0, 8.5), 6.63 (dd, 1H, J = 0.5, 8.5), 4.80 (s, 2H),
4.73 (s, 2H), 4.11 (q, 2H, J = 7.0), 1.17 (t, 2H, J = 7.0). δC (101 MHz,
DMSO-d6) 169.9, 154.4, 140.5, 131.0, 126.9, 110.6, 62.2, 60.6, 14.5.
m/z (ES) 197 ([M + H]+). HRMS, found 197.0930 (C9H13N2O3
requires 197.0926). HPLC purity 98.9%. mp 101−104 °C.
One-Pot Synthesis of Indole-3-glyoxylamide Derivatives 5−35,
56−60, 62, and 64 (Scheme 1a): General Procedure. The
appropriate starting indole 4a−o (1.5 mmol) was dissolved in
anhydrous THF (15 mL), then oxalyl chloride (144 μL, 209 mg,
1.65 mmol) was added. The solution was stirred at rt for 3 h, then
DIPEA (610 μL, 452 mg, 3.5 mmol) added followed by the requisite
amine (R2-NH2, 1.8 mmol) and DMAP (catalytic amount). The
suspension was stirred at rt overnight then evaporated to dryness. The
residue was extracted with EtOAc and water, then the organic layer
evaporated once more to provide the crude product which was purified
as specified in each case. Unless stated otherwise, purification was
achieved by two successive recrystallizations: first from EtOAc/hexane,
then from propan-2-ol/water. Characterization data, HPLC purity, and
any additional purification details are provided for each example in the
Supporting Information.
Synthesis of Indole-3-glyoxylate Methyl Esters 68a−j from
Indoles 67a−j: General Procedure. The relevant indole compound
67a−j was dissolved in anhydrous CH2Cl2 at a concentration of 0.1 M,
then the solution was cooled to 0 °C. AlCl3 (typically 1.25 equiv) and
methyl chlorooxoacetate (typically 1.25 equiv) were added and the
reaction continued at 0 °C until acceptable conversion was seen as
Q
DOI: 10.1021/acs.jmedchem.5b01312
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
judged by TLC (reaction duration specified below for each case).
Methanol was added to quench the reaction, then the mixture poured
into satd NaHCO3 and extracted thoroughly. If an emulsion resulted,
the entire mixture was filtered under vacuum and the filtrate returned
to the separating funnel. The organic layer was collected and the
aqueous phase extracted with additional CH2Cl2. The combined
organic extracts were dried over Na2SO4 and evaporated. Further
purification was carried out by flash column chromatography on silica
gel, using an eluent indicated in each case, to afford the desired
product 68a−j.
Methyl 2-(4-Chloro-1-isopropyl-1H-indol-3-yl)-2-oxoacetate 68a.
Prepared from 67a (520 μL, 581 mg, 3.0 mmol), AlCl3 (600 mg, 4.5
mmol), and methyl chlorooxoacetate (414 μL, 551 mg, 4.5 mmol).
Reaction duration 1 h. Column eluent 20 → 33 → 50% EtOAc/
hexane. Title compound obtained as a yellow gum (280 mg, 33%). δH
(400 MHz, CDCl3) 8.28 (s, 1H), 7.38−7.21 (m, 3H), 4.71 (septet,
1H, J = 6.5), 3.96 (s, 3H), 1.60 (d, 6H, J = 6.5). δC (101 MHz, CDCl3)
178.4, 164.5, 138.3, 135.7, 127.5, 124.5, 124.2, 113.6, 109.0, 52.7, 48.6,
22.4. m/z (ES) 280 (100%; [M(35Cl) + H]+), 282 (30%; [M(37Cl) +
H]+). HRMS, found 280.0731 (C14H15ClNO3 requires 280.0740).
HPLC purity 98.6%.
Methyl 2-(5-Chloro-1-(2-methoxyethyl)-1H-indol-3-yl)-2-oxoace-
tate 68c. Prepared from 67c (285 mg, 1.36 mmol), AlCl3 (227 mg,
1.70 mmol), and methyl chlorooxoacetate (156 μL, 208 mg, 1.70
mmol). Reaction duration 1 h. Column eluent 20 → 33 → 50 → 67%
EtOAc/hexane. Title compound obtained as a greasy, yellow solid
(111 mg, 28%). δH (400 MHz, CDCl3) 8.45 (s, 1H), 8.44 (dd, 1H, J =
0.5, 2.0), 7.35−7.28 (m, 2H), 4.33 (t, 2H, J = 5.0), 3.97 (s, 3H), 3.76
(t, 2H, J = 5.0), 3.34 (s, 3H). δC (101 MHz, CDCl3) 176.9, 163.0,
141.0, 135.2, 129.6, 128.2, 124.5, 122.4, 112.6, 111.2, 70.6, 59.2, 52.8,
47.5. m/z (ES) 296 (100%; [M(35Cl) + H]+), 298 (35%; [M(37Cl) +
H]+). HRMS, found 296.0694 (C14H15ClNO4 requires 296.0690).
HPLC purity 95.9%. mp 88−90 °C.
Methyl 2-(6-Chloro-1-isopropyl-1H-indol-3-yl)-2-oxoacetate 68d.
Prepared using 67d (347 μL, 387 mg, 2.0 mmol), AlCl3 (667 mg, 5.0
mmol), and methyl chlorooxoacetate (460 μL, 613 mg, 5.0 mmol).
Reaction duration 100 min (at rt). Column eluent 20 → 30 → 40%
EtOAc/hexane. Title compound obtained as a crystalline yellow solid
(142 mg, 25%). δH (400 MHz, CDCl3) 8.49 (s, 1H), 8.38 (d, 1H, J =
8.5), 7.44 (s, 1H), 7.33 (d, 1H, J = 8.5), 4.66 (septet, 1H, J = 6.5), 3.98
(s, 3H), 1.64 (d, 6H, J = 6.5). δC (101 MHz, CDCl3) 176.7, 163.2,
136.9, 136.4, 130.0, 125.8, 124.1, 123.9, 113.1, 110.5, 52.8, 48.7, 22.5.
m/z (ES) 280 (100%; [M(35Cl) + H]+), 282 (35%; [M(37Cl) + H]+).
HRMS, found 280.0742 (C14H15ClNO3 requires 280.0740). HPLC
purity 99.7%. mp 94−95 °C.
Methyl 2-(7-Chloro-1-isopropyl-1H-indol-3-yl)-2-oxoacetate 68e.
Prepared using 67e (441 mg, 2.28 mmol), AlCl3 (380 mg, 2.85 mmol),
and methyl chlorooxoacetate (262 μL, 349 mg, 2.85 mmol). Reaction
duration 1 h. Column eluent 15 → 33 → 50 → 67% CH2Cl2/hexane
then 20 → 33% EtOAc/hexane. Title compound obtained as a waxy
yellow solid (195 mg, 31%). δH (400 MHz, CDCl3) 8.60 (s, 1H), 8.45
(dd, 1H, J = 1.0, 8.0), 7.32 (dd, 1H, J = 1.0, 8.0), 7.25 (t, 1H, J = 8.0),
5.73 (septet, 1H, J = 6.5), 3.98 (s, 3H), 1.64 (d, 6H, J = 6.5). δC (101
MHz, CDCl3) 176.7, 163.2, 137.0, 132.1, 130.2, 126.4, 124.1, 121.6,
117.2, 113.1, 52.8, 49.6, 23.8. m/z (ES) 280 (100%; [M(35Cl) + H]+),
282 (30%; [M(37Cl) + H]+). HRMS, found 280.0734 (C14H15ClNO3
requires 280.0740). HPLC purity 99.2%; mp 62 °C.
Methyl 2-(1-Isopropyl-4-methoxy-1H-indol-3-yl)-2-oxoacetate
68f. Prepared using 67f (405 mg, 2.14 mmol), AlCl3 (428 mg, 3.21
mmol), and methyl chlorooxoacetate (295 μL, 393 mg, 3.21 mmol).
Reaction duration 80 min. Column eluent 10 → 20 → 33 → 50%
EtOAc/pentanes. Title compound obtained as a bright-yellow gum
(120 mg, 20%). δH (400 MHz, CDCl3) 8.13 (s, 1H), 7.26 (t, 1H, J =
8.0), 7.07 (d, 1H, J = 8.0), 6.72 (d, 1H, J = 8.0), 4.68 (septet, 1H, J =
6.5), 3.95 (s, 3H), 3.94 (s, 3H), 1.59 (d, 6H, J = 6.5). δC (101 MHz,
CDCl3) 182.0, 165.6, 154.1, 138.3, 132.6, 124.4, 115.9, 113.5, 103.8,
103.4, 55.6, 52.3, 48.5, 22.6. m/z (ES) 276 ([M + H]+). HRMS, found
276.1232 (C15H18NO4 requires 276.1236). HPLC purity 98.1%.
Methyl 2-(1-Isopropyl-5-methoxy-1H-indol-3-yl)-2-oxoacetate
68g. Prepared using 67g (315 mg, 1.66 mmol), AlCl3 (277 mg,
2.08 mmol), and methyl chlorooxoacetate (191 μL, 255 mg, 2.08
mmol). Reaction duration 50 min. Column eluent 10 → 20 → 30 →
40% EtOAc/hexane. Title compound obtained as a thick, orange oil
(146 mg, 32%). δH (400 MHz, CDCl3) 8.43 (s, 1H), 7.98 (d, 1H, J =
2.5), 7.33 (d, 1H, J = 9.0), 6.98 (dd, 1H, J = 2.5, 9.0), 4.67 (septet, 1H,
J = 6.5), 3.97 (s, 3H), 3.93 (s, 3H), 1.61 (d, 6H, J = 6.5). δC (101
MHz, CDCl3) 176.8, 163.6, 157.1, 135.7, 131.2, 128.3, 114.2, 112.8,
111.1, 104.3, 55.8, 52.7, 48.7, 22.6. m/z (ES) 276 ([M + H]+). HRMS,
found 276.1226 (C15H18NO4 requires 276.1236). HPLC purity 92.1%.
Methyl 2-(5-Methoxy-1-(2-methoxyethyl)-1H-indol-3-yl)-2-oxoa-
cetate 68h. Prepared using 67h (308 mg, 1.50 mmol), AlCl3 (251 mg,
1.88 mmol), and methyl chlorooxoacetate (173 μL, 230 mg, 1.88
mmol). Reaction duration 40 min. Column eluent 20 → 40 → 60 →
80% EtOAc/hexane. Title compound obtained as a viscous, bright-
yellow oil (134 mg, 31%) which crystallized on storage at −20 °C,
giving a yellow solid. δH (400 MHz, CDCl3) 8.38 (s, 1H), 7.97 (d, 1H,
J = 2.5), 7.30 (d, 1H, J = 9.0), 6.98 (dd, 1H, J = 2.5, 9.0), 4.32 (t, 2H, J
= 5.0), 3.97 (s, 3H), 3.93 (s, 3H), 3.76 (t, 2H, J = 5.0), 3.34 (s, 3H). δC
(101 MHz, CDCl3) 177.1, 163.5, 157.1, 140.1, 131.6, 128.2, 114.3,
112.9, 110.9, 104.3, 70.6, 59.1, 55.8, 52.7, 47.5. m/z (ES) 292 ([M +
H]+). HRMS, found 292.1186 (C15H18ClNO5 requires 292.1185).
HPLC purity 97.8%. mp 80−82 °C.
Methyl 2-(1-Isopropyl-6-methoxy-1H-indol-3-yl)-2-oxoacetate
68i. Prepared using 67i (348 mg, 1.84 mmol), AlCl3 (307 mg, 2.30
mmol), and methyl chlorooxoacetate (212 μL, 282 mg, 2.30 mmol).
Reaction duration 70 min. Column eluent 15 → 30 → 40% EtOAc/
hexane. Title compound obtained as a thick, yellow oil (175 mg, 35%)
which crystallized on storage at −20 °C to give a yellow solid. δH (400
MHz, CDCl3) 8.40 (s, 1H), 8.34 (d, 1H, J = 8.5), 7.00 (dd, 1H, J = 2.0,
8.5), 6.89 (d, 1H, J = 2.0), 4.64 (septet, 1H, J = 6.5), 3.96 (s, 3H), 3.90
(s, 3H), 1.62 (d, 6H, J = 6.5). δC (101 MHz, CDCl3) 176.7, 163.5,
157.6, 137.4, 135.3, 123.6, 121.2, 113.2, 112.2, 94.6, 55.8, 52.6, 48.3,
22.4. m/z (ES) 276 ([M + H]+). HRMS, found 276.1243 (C15H18NO4
requires 276.1236). HPLC purity 95.9%. mp 91 °C.
Methyl 2-(1-Isopropyl-7-methoxy-1H-indol-3-yl)-2-oxoacetate
68j. Prepared using 67j (379 mg, 2.0 mmol), AlCl3 (333 mg, 2.5
mmol), and methyl chlorooxoacetate (230 μL, 306 mg, 2.5 mmol).
Reaction duration 1 h. Column eluent 5 → 10 → 20 → 30% EtOAc/
pentanes. Title compound obtained as a viscous orange oil (150 mg,
27%). δH (400 MHz, CDCl3) 8.46 (s, 1H), 8.09 (dd, 1H, J = 1.0, 8.0),
7.26 (t, 1H, J = 8.0), 6.82−6.79 (m, 1H), 5.46 (septet, 1H, J = 6.5),
3.98 (s, 3H), 3.97 (s, 3H), 1.58 (d, 6H, J = 6.5). δC (101 MHz,
CDCl3) 176.8, 163.6, 147.5, 135.8, 129.5, 126.2, 124.2, 115.3, 113.1,
105.4, 55.5, 52.7, 50.7, 23.6. m/z (ES) 276 ([M + H]+). HRMS, found
276.1232 (C15H18NO4 requires 276.1236). HPLC purity 80.2%.
Synthesis of Azaindole-3-glyoxylate Methyl Esters 68k−o from
Azaindoles 67k−o: General Procedure. The relevant azaindole
compound 67k−o was dissolved in anhydrous CH2Cl2 (30 mL/
mmol), then AlCl3 (5 equiv) was added and the suspension stirred for
1 h. Methyl chlorooxoacetate (5 equiv) was added and reaction
continued for the duration specified in each case. Methanol was added
to quench the reaction, then the mixture was evaporated to dryness
under vacuum. Further purification was carried out by flash column
chromatography on silica gel, using an eluent indicated in each case, to
afford the desired product 68k−o.
Methyl 2-(1-Isopropyl-1H-pyrrolo[3,2-b]pyridin-3-yl)-2-oxoace-
tate 68k. Prepared from 67k (365 mg, 2.28 mmol), AlCl3 (1.52 g,
11.4 mmol), and methyl chlorooxoacetate (1.05 mL, 1.40 mg, 11.4
mmol). Reaction duration 4.5 h. In this case, after quenching with
MeOH, extraction between EtOAc and water was carried out and the
organic layer collected and evaporated. Column eluent 2.5 → 5→10%
MeOH/CHCl3. Product obtained as a golden brown gum (40 mg,
7%). Given the poor recovery in this case, it is suspected the majority
of material was either retained by the column or decomposed on silica.
δH (400 MHz, CDCl3) 8.71 (dd, 1H, J = 1.5, 4.5), 8.57 (s, 1H), 7.75
(dd, 1H, J = 1.5, 8.5), 7.24 (dd, 1H, J = 4.5, 8.5), 4.71 (septet, 1H, J =
6.5), 3.97 (s, 3H), 1.62 (d, 6H, J = 6.5). δC (101 MHz, CDCl3) 177.4,
164.1, 146.6, 144.9, 136.7, 129.4, 118.2, 117.9, 112.6, 52.8, 49.1, 22.4.
m/z (ES) 247 ([M + H]+). HRMS, found 247.1082 (C13H15N2O3
requires 247.1083). HPLC purity 95.8%.
R
DOI: 10.1021/acs.jmedchem.5b01312
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Methyl 2-(1-Isopropyl-1H-pyrrolo[3,2-c]pyridin-3-yl)-2-oxoace-
tate 68l. Prepared from 67l (320 mg, 2.0 mmol), AlCl3 (1.33 g, 10
mmol), and methyl chlorooxoacetate (0.92 mL, 1.23 mg, 10 mmol).
Reaction duration 2 h. Column eluent 0 → 2→4 → 6→8% MeOH/
CH2Cl2. Product obtained as a waxy, off-white solid (191 mg, 39%).
δH (400 MHz, CDCl3) 9.66 (d, 1H, J = 1.0), 8.54 (s, 1H), 8.51 (d, 1H,
J = 6.0), 7.36 (dd, 1H, J = 1.0, 6.0), 4.73 (septet, 1H, J = 7.0), 3.99 (s,
3H), 1.65 (d, 6H, J = 7.0). δC (101 MHz, CDCl3) 176.6, 162.8, 145.9,
143.3, 140.4, 136.5, 123.6, 113.2, 105.5, 52.9, 48.9, 22.5. m/z (ES) 247
([M + H]+). HRMS, found 247.1081 (C13H15N2O3 requires
247.1083). HPLC purity 94.1%.
Methyl 2-(1-Isopropyl-1H-pyrrolo[2,3-c]pyridin-3-yl)-2-oxoace-
tate 68m. Prepared from 67m (320 mg, 2.0 mmol), AlCl3 (1.33 g,
10 mmol), and methyl chlorooxoacetate (0.92 mL, 1.23 mg, 10
mmol). Reaction duration 2 h. Column eluent 0 → 1→2 → 4→6%
MeOH/CH2Cl2. Product obtained as a waxy, off-white solid (263 mg,
53%). δH (400 MHz, DMSO-d6) 9.14 (d, 1H, J = 1.0), 8.71 (s, 1H),
8.42 (d, 1H, J = 5.5), 8.07 (dd, 1H, J = 1.0, 5.5), 5.04 (septet, 1H, J =
6.5), 3.92 (s, 3H), 1.58 (d, 6H, J = 6.5). δC (101 MHz, DMSO-d6)
178.7, 163.8, 142.7, 140.0, 135.2, 133.5, 131.8, 116.0, 111.7, 53.2, 49.8,
22.6. m/z (ES) 247 ([M + H]+). HRMS, found 247.1074
(C13H15N2O3 requires 247.1083). HPLC purity 97.4%.
Methyl 2-(1-Isopropyl-1H-pyrrolo[2,3-b]pyridin-3-yl)-2-oxoace-
tate 68n. Prepared from 67n (320 mg, 2.0 mmol), AlCl3 (1.33 g,
10 mmol), and methyl chlorooxoacetate (0.92 mL, 1.23 mg, 10
mmol). Reaction duration 2.5 h. Column eluent 67 → 100% CH2Cl2/
hexane then 0.5 → 1→2% MeOH/CH2Cl2. Product obtained as a
waxy, off-white solid (421 mg, 86%). δH (400 MHz, DMSO-d6) 8.73
(s, 1H), 8.49 (dd, 1H, J = 1.5, 8.0), 8.45 (dd, 1H, J = 1.5, 4.5), 7.39
(dd, 1H, J = 4.5, 8.0), 5.15 (septet, 1H, J = 7.0), 3.92 (s, 3H), 1.56 (d,
6H, J = 7.0). δC (101 MHz, DMSO-d6) 178.8, 163.8, 147.9, 145.3,
138.2, 130.4, 119.9, 119.1, 110.6, 53.2, 47.7, 22.4. m/z (ES) 247 ([M +
H]+). HRMS, found 247.1089 (C13H15N2O3 requires 247.1083).
Methyl 2-(1-(2-Methoxyethyl)-1H-pyrrolo[2,3-b]pyridin-3-yl)-2-
oxoacetate 68o. Prepared from 67o (325 μL, 352 mg, 2.0 mmol),
AlCl3 (1.33 g, 10 mmol), and methyl chlorooxoacetate (0.92 mL, 1.23
mg, 10 mmol). Reaction duration 5 h. Column eluent 0 → 1→2 → 4%
MeOH/CH2Cl2. Product obtained as a pale-yellow oil which
crystallized on standing to a beige solid (324 mg, 62%). δH (400
MHz, CDCl3) 8.68 (dd, 1H, J = 1.5, 8.0), 8.62 (s, 1H), 8.42 (dd, 1H, J
= 1.5, 5.0), 7.30 (dd, 1H, J = 5.0, 8.0), 4.55 (t, 2H, J = 5.0), 3.98 (s,
3H), 3.80 (t, 2H, J = 5.0), 3.36 (s, 3H). δC (101 MHz, CDCl3) 177.1,
162.8, 148.0, 145.0, 140.6, 131.2, 119.6, 119.4, 111.3, 70.6, 59.0, 52.8,
45.3. m/z (ES) 263 ([M + H]+). HRMS, found 263.1038
(C13H15N2O4 requires 263.1032). HPLC purity 98.3%. mp 69−70 °C.
Conversion of Esters 68a and68c−o into Amides 36, 38, and
40−51 (Scheme 1b): General Procedure. The starting ester, the
relevant amine (R2-NH2, 1.2 equiv) and TBD (30 mol %) were
combined in the minimum volume of toluene (typically 1−2 mL),
then the mixture was heated at 85 °C overnight. The solvent was
removed by rotary evaporation then the residue purified further.
Unless stated otherwise, this was achieved as follows. Hexane was
added to a solution of the crude material in boiling EtOAc, with
continuing heating, until cloudiness persisted, at which point the
mixture was left to stand for 1−2 min. A sticky film of brown gum
(mostly unreacted amine component) quickly accumulated at the
bottom of the flask from which the remainder of the suspension could
readily be decanted into a separate flask, at which point, the mixture
was evaporated to dryness. Recrystallization of the remaining residue
from propan-2-ol/water provided the amide product, which was
collected by filtration and dried. Characterization data, HPLC purity,
and any additional purification details are provided for each example in
the Supporting Information.
extracted with EtOAc and water, the aqueous phase was extracted with
more EtOAc, then the combined organic extracts washed with water
then brine, and evaporated under vacuum. Flash column chromatog-
raphy on silica gel, eluted as specified in each individual case, then
afforded the pure alkyne products.
2-(Pentyn-1-yl)pyridine 81. Synthesized using 2-bromopyridine
(954 μL, 1.58 g, 10 mmol), PdCl2(PPh3)2 (140 mg, 0.2 mmol), PPh3
(105 mg, 0.4 mmol), CuI (76 mg, 0.4 mmol), and pent-1-yne (986 μL,
10 mmol). Column eluent 10 → 15 → 20% EtOAc/hexane. Product
obtained as an amber oil (1.13 g, 78%). δH (400 MHz, DMSO-d6) 8.51
(d, 1H, J = 5.0), 7.76 (dt, 1H, J = 2.0, 8.0), 7.45 (d, 1H, J = 8.0), 7.35−
7.31 (m, 1H), 2.43 (t, 2H, J = 7.0), 1.58 (sextet, 2H, J = 7.0), 1.00 (t,
3H, J = 7.0). δC (101 MHz, DMSO-d6) 150.3, 143.4, 137.0, 127.3,
123.3, 90.8, 81.4, 21.9, 20.8, 13.8. m/z (ES) 146 ([M + H]+). HRMS,
found 146.0965 (C10H12N requires 146.0970).
2-(4-Methylpent-1-ynyl)pyridine 82. Synthesized using 2-bromo-
pyridine (3.81 mL, 6.32 g, 40 mmol), PdCl2(PPh3)2 (562 mg, 0.8
mmol), PPh3 (420 mg, 1.6 mmol), CuI (305 mg, 1.6 mmol), and 4-
methylpent-1-yne (4.71 mL, 3.29 g, 40 mmol). Column eluent 5 → 10
→ 15 → 20 → 25% EtOAc/hexane. Product obtained as a brown oil
(4.48 g, 70%). δH (400 MHz, DMSO-d6) 8.53 (d, 1H, J = 4.5), 7.75
(dt, 1H, J = 2.0, 8.0), 7.45 (d, 1H, J = 8.0), 7.33 (ddd, 1H, J = 1.0, 4.5,
6.0), 2.35 (d, 2H, J = 6.5), 1.85 (septet, 1H, J = 6.5), 0.98 (d, 6H, J =
6.5). δC (101 MHz, DMSO-d6) 150.3, 143.5, 137.0, 127.3, 123.3, 89.8,
82.1, 27.94, 27.91, 22.2. m/z (ES) 160 ([M + H]+). HRMS, found
160.1120 (C11H14N requires 160.1126). HPLC purity 99.5%.
5-(Pyridin-2-yl)pent-4-yn-1-ol 83. Synthesized using 2-bromopyr-
idine (2.38 mL, 3.95 g, 25 mmol), PdCl2(PPh3)2 (351 mg, 0.5 mmol),
PPh3 (262 mg, 1.0 mmol), CuI (190 mg, 1.0 mmol), and pent-4-yn-1-
ol (2.33 mL, 2.10 g, 25 mmol). Column eluent 2.5 → 5→10 → 15%
MeOH/CH2Cl2. Product obtained as a yellow oil (2.46 g, 61%). δH
(400 MHz, DMSO-d6) 8.52 (d, 1H, J = 5.0), 7.76 (dt, 1H, J = 1.5, 7.5),
7.44 (d, 1H, J = 7.5), 7.33 (ddd, 1H, J = 1.5, 5.0, 5.5), 4.59 (t, 1H, J =
5.5), 3.53 (q, 2H, J = 5.5), 2.50 (t, 2H, J = 7.0), 1.74−1.67 (m, 2H). δC
(101 MHz, DMSO-d6) 150.3, 143.5, 137.0, 127.3, 123.3, 90.9, 81.1,
59.9, 31.7, 15.6. m/z (ES) 162 ([M + H]+). HRMS, found 162.0912
(C10H12NO requires 162.0919). HPLC purity 99.0%.
2-(5-Methoxypent-1-ynyl)pyridine 84. By Methylation of 83
(Method A). Alcohol 83 (1.14 g, 7.08 mmol) and methyl iodide
(0.53 mL, 1.21 g, 8.5 mmol) were dissolved in anhydrous DMF (25
mL) at 0 °C, then sodium hydride (60% dispersion in mineral oil, 0.34
g, 8.5 mmol) was added in small portions over 5 min. After stirring for
30 min at 0 °C, the cooling was removed and reaction continued at rt
for 3 h. Water was added dropwise to quench any remaining NaH,
then the mixture evaporated under vacuum. The residue was extracted
between EtOAc and water, the aqueous phase extracted with more
EtOAc, then the combined organic extracts washed with brine, water
then brine, dried over Na2SO4, filtered, and evaporated. The resulting
brown oil was purified by preparative HPLC, using a gradient of 5−
15% MeCN in water (containing 0.1% v/v TFA) over 20 min. After
evaporation of pooled fractions, the remaining yellow oil was
partitioned between EtOAc and satd NaHCO3 to regenerate the
free base form of the product. After separation of the layers, the
aqueous phase was extracted again with EtOAc and the combined
organics dried over Na2SO4, filtered, and evaporated to dryness once
more to afford the title compound as a yellow/orange oil (337 mg,
27%). δH (400 MHz, DMSO-d6) 8.56−8.50 (m, 1H), 7.77 (dt, 1H, J =
1.5, 7.5), 7.45 (d, 1H, J = 7.5), 7.34 (dd, 1H, J = 5.0, 7.5), 3.44 (t, 2H, J
= 6.5), 3.25 (s, 3H), 2.49 (t, 2H, J = 6.5), 1.78 (quintet, 2H, J = 6.5).
δC (101 MHz, DMSO-d6) 149.8, 142.9, 136.6, 126.8, 122.9, 89.8, 80.8,
70.4, 57.9, 27.9, 15.3. m/z (ES) 176 ([M + H]+). HRMS, found
176.1075 (C11H14NO requires 176.1075). HPLC purity 99.7%.
By Two-Step Synthesis from Pent-4-yn-1-ol (Method B). A
solution of pent-4-yn-1-ol (3.26 mL, 2.94 g, 35 mmol) and methyl
iodide (4.36 mL, 9.94 g, 70 mmol) in dry THF (60 mL) was cooled to
0 °C, then sodium hydride (60% dispersion in mineral oil, 1.68 g, 42
mmol) was added portionwise over 10 min. After 5 min more, the
cooling was removed and the reaction mixture heated at 50 °C for 2.5
h. The suspension was poured carefully into ice-cold water (250 mL)
then extracted twice with Et2O. The combined extracts were washed
Synthesis of Alkynes 81−83 from 2-Bromopyridine: General
Procedure. PdCl2(PPh3)2 (2 mol %), triphenylphosphine (4 mol %),
and CuI (4 mol %) were added in succession to a 0.5 M solution of 2-
bromopyridine in TEA. The resulting suspension was stirred for 30
min, then the relevant alkyne (1.0 equiv) was added. The reaction
mixture was then stirred overnight in the dark, quenched by addition
of satd NH4Cl, and evaporated under vacuum. The residual slurry was
S
DOI: 10.1021/acs.jmedchem.5b01312
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
with water, dried over Na2SO4, filtered, then evaporated slowly under
vacuum, maintaining the water bath at 4 °C and using an air bleed to
control the evaporation rate. Slow evaporation was necessary to
prevent significant loss of the volatile intermediate, 5-methoxypent-1-
yne. Once the solution of this crude alkyne had been concentrated to
approximately 25 mL, it was used immediately as the alkyne
component of a Sonogashira cross-coupling with 2-bromopyridine
(3.34 mL, 5.53 g, 35 mmol), according to the general procedure
described above, also using PdCl2(PPh3)2 (491 mg, 0.7 mmol), PPh3
(367 mg, 1.4 mmol), and CuI (267 mg, 1.4 mmol). Column eluent 0
→ 1→2.5 → 4→6% MeOH/CH2Cl2. Title compound obtained as an
orange/brown oil (1.11 g, 18% over 2 steps). Analytical data were in
agreement with the sample prepared by method A.
Cyclization of 2-Alkynylpyridines 81, 82, and 84 to 3-
Alkylindolizines 85−87: General Procedure. CuCl (1 equiv) was
added to a 0.45 M solution of starting 2-alkynylpyridine in dry DMA/
TEA (7:1). The mixture was stirred at rt for 10 min then heated to 130
°C for the duration specified in each case. The reaction mixture was
poured into a 1:1 mixture of EtOAc and water, which was then
extracted thoroughly. Following vacuum filtration to remove
suspended solids, the filtrate was poured into a separating funnel
and the layers separated. The aqueous layer was extracted twice more
with EtOAc, repeating the filtration step as necessary. The combined
extracts were washed with water then brine, dried over Na2SO4,
filtered, and evaporated. Purification was carried out by flash column
chromatography on silica gel, using the eluent specified in each case,
and making sure to run the column as quickly as possible to minimize
decomposition of the indolizine product. After evaporation and drying,
the indolizines were stored at −20 °C until further use.
3-Ethylindolizine 85. Prepared from alkyne 81 (500 mg, 3.44
mmol) and CuCl (341 mg, 3.44 mmol). Reaction duration 7.5 h.
Column eluent 0 → 10% EtOAc/hexane. Title compound obtained as
a yellow oil (161 mg, 32%). δH (400 MHz, DMSO-d6) 7.99 (d, 1H, J =
7.0), 7.40 (d, 1H, J = 9.0), 6.66−6.61 (m, 1H), 6.59−6.54 (m, 2H),
6.34 (d, 1H, J = 4.0), 2.80 (q, 2H, J = 7.5), 1.30 (t, 3H, J = 7.5). δC
(101 MHz, DMSO-d6) 132.1, 126.0, 122.7, 119.4, 115.8, 111.1, 110.3,
98.3, 18.9, 12.2. m/z (ES) 146 ([M + H]+). HRMS, found 146.0972
(C10H12N requires 146.0970).
3-Isopropylindolizine 86. Prepared from alkyne 82 (4.01 g, 25.2
mmol) and CuCl (2.49 g, 25.2 mmol). Reaction duration 6.5 h.
Column eluent 0 → 5→10% EtOAc/hexane. Title compound
obtained as a bright-yellow oil (1.24 g, 31%). δH (400 MHz,
DMSO-d6) 8.00 (d, 1H, J = 7.0), 7.40 (d, 1H, J = 9.0), 6.65−6.60 (m,
1H), 6.58−6.52 (m, 2H), 6.35 (d, 1H, J = 4.0), 3.24 (septet, 1H, J =
7.0), 1.28 (d, 6H, J = 7.0). δC (101 MHz, DMSO-d6) 132.2, 130.7,
122.8, 119.6, 115.7, 110.2, 109.4, 98.4, 24.8, 21.6. m/z (ES) 160 ([M +
H]+). HRMS, found 160.1127 (C11H14N requires 160.1126). HPLC
purity 94.2%.
synthesis), was dissolved in 1:1 THF/water to a final concentration of
approximately 0.1 M, then 5 M NaOH (2.5 equiv) was added. The
reaction mixture was stirred at rt for 2 h. Complete reaction was
confirmed by TLC, then the solution was adjusted to pH 7 with 1 M
HCl and evaporated under vacuum. The residue was extracted with
EtOAc and 1 M HCl, the aqueous layer extracted three times further
with EtOAc, then the combined organic solutions dried over Na2SO4,
filtered, and evaporated. Purification was carried out by preparative
HPLC, using the conditions specified in each case. Characterization
data, HPLC purity, and full purification details are provided for each
example in the Supporting Information.
HPLC Solubility Assay. The maximal aqueous solubility of
compounds 12, 13, and 32 was estimated as follows: dilutions of
compound from 10 mM DMSO stock solutions were made into water
at final concentrations ranging from 1−100 μM (typically 1, 2, 4, 6, 10,
15, 20, 30, 50, and 100 μM). These solutions were mixed overnight at
rt then centrifuged in a benchtop microcentrifuge at 13000 rpm
(approximately 17000g) for 5 min to remove any suspended
precipitate. Supernatants were analyzed by HPLC using a C18 5 μm
column, 4.6 mm × 150 mm, and a gradient of 5−55% MeCN/water
over 9 min; ramp to 100% MeCN over 4 min; hold at 100% MeCN
for 5 min, with UV detection at 254 nm. Plots of peak area against
concentration were constructed, and the point at which each plot
deviated from linearity was estimated as the threshold aqueous
solubility of the test compound (above which point the plot leveled
off, indicating saturation).
Biology. Cell Lines and Culture. Cell lines were obtained from the
American Type Culture Collection (FaDu, SCC-4, SK-MEL-28) or
the European Collection of Cell Cultures (MES-SA, MES-SA/Dx5,
SH-SY5Y, Caco-2) and cultured at 37 °C as recommended in each
case, with the exception of FaDu, which was grown in RPMI-1640
medium supplemented with 2 mM L-glutamine. Additionally, all cell
culture medium was supplemented with 100 IU/mL penicillin and 100
μg/mL streptomycin.
Cytotoxicity (MTT) Assay. Cells were distributed into 96-well tissue
culture plates (Greiner Bio-One) at an appropriate density in a volume
of 100 μL/well and incubated for 24 h to allow for cell attachment.
Test compound solutions were prepared by dilution from 10 mM
DMSO stock solutions into full growth medium at 2× the required
final screening concentration and dosed in triplicate at 100 μL/well
(i.e., final assay volume ∼200 μL/well). After incubation for 72 h,
medium was removed from the wells, the cells were washed with PBS,
and 100 μL/well of a solution of MTT in PBS (0.5 mg/mL) was
added. The plate was incubated once more for 45−60 min then the
solution removed and 0.1 M HCl in 2-propanol added (40 μL/well).
After gentle agitation to dissolve the formazan crystals, absorbance
values at 570 nm were read using a microplate reader (reference
wavelength 650 nm). Data was processed using Microsoft Excel,
expressed as viability relative to vehicle control (DMSO only), and
plotted in dose−response form using GraphPad Prism 6 (GraphPad
Software, Inc.), allowing derivation of LC50 values by nonlinear
regression.
3-(2-Methoxyethyl)indolizine 87. Prepared from alkyne 84 (1.09 g,
6.22 mmol) and CuCl (616 mg, 6.22 mmol). Reaction duration 4 h.
Column eluent 5 → 10 → 20% EtOAc/pentanes. Title compound
obtained as a bright-yellow oil (591 mg, 54%). δH (400 MHz, DMSO-
d6) 8.06 (dd, 1H, J = 1.0, 6.5), 7.39 (dt, 1H, J = 1.0, 9.0), 6.64 (ddd,
1H, J = 1.0, 6.5, 9.0), 6.59 (d, 1H, J = 4.0), 6.55 (dt, 1H, J = 1.0, 6.5),
6.35 (d, 1H, J = 4.0), 3.65 (t, 2H, J = 6.5), 3.28 (s, 3H), 3.09 (t, 2H, J =
6.5). δC (101 MHz, DMSO-d6) 132.2, 123.0, 121.9, 119.3, 115.9,
112.7, 110.3, 98.5, 70.8, 58.4, 26.2. m/z (ES) 176 ([M + H]+). HRMS,
found 176.1073 (C11H14NO requires 176.1075). HPLC purity 99.1%.
Synthesis of Indolizine-1-glyoxylamides 52−55 (Scheme 4):
General Procedure. These reactions were carried out using the
same protocol as the one-pot indole-3-glyoxylamide syntheses
described above, except the first step (reaction with oxalyl chloride)
was carried out at 0 °C instead of rt. Characterization data, HPLC
purity, and any additional purification details are provided for each
example in the Supporting Information.
Tubulin Polymerization Assay (OD340 Method). Purified porcine
brain tubulin and other reagents were obtained commercially in kit
form (Cytoskeleton, Inc.). Assays were carried out in half-area 96-well
plates at 37 °C and a final concentration of 3.6 mg/mL tubulin, 9.2%
glycerol, and 900 μM GTP in 100 μL total volume. Test compounds
were preplated at 10× final assay concentration in water (10 μL/well),
then a solution of tubulin (4 mg/mL) in G-PEM buffer (80 mM
PIPES pH 6.9, 2 mM MgCl2, 0.5 mM EGTA) containing 10.2%
glycerol and 1 mM GTP was prepared freshly on ice and promptly
distributed into reaction wells at 90 μL/well. Reactions were followed
at 37 °C over 60 min, monitoring OD340 at 1 min intervals. Data from
each well was normalized relative to initial readings, and plots of
ΔODmax (final−initial values) against compound concentration,
expressed relative to vehicle control (DMSO only), were used to
calculate IC50 values (Supporting Information, Figure S1).
Preparation of 61, 63, and 65 via Ester Deprotection: General
Procedure. These reactions were not run under N2. The starting ester,
either 62, 64, or ethyl 2-(2-(3-(2-(6-methoxypyridin-3-ylamino)-2-
oxoacetyl)-1H-indol-1-yl)-2-methylpropoxy)acetate (prepared from
4m by the general procedure for one-pot indole-3-glyoxylamide
Tubulin Polymerization Assay (DAPI Fluorescence Method).
Purified tubulin and other reagents were sourced in kit form
(Cytoskeleton, Inc.) as above. Assays were run in a total volume of
T
DOI: 10.1021/acs.jmedchem.5b01312
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
55 μL, in half-area 96-well black plates at 37 °C in G-PEM buffer
containing 1.45 mg/mL tubulin, 900 μM GTP, 4.97 μM DAPI, and
13.75% glycerol. Reactions were monitored by measuring DAPI
fluorescence (excitation 360 nm, emission 450 nm) once per minute
over a period of 60 min. Data was normalized and processed as per the
OD340 method (see Supporting Information, Figures S2,S3).
fixative was removed and the cells washed three times with PBS, then
blocked by treatment with 2% FCS and 1% BSA in PBS for 30 min.
The blocking solution was removed and the samples incubated with
FITC-conjugated mouse anti-α-tubulin antibody (clone DM1A,
Sigma, 1:500) in PBS for 40 min. The cells were washed three
times with PBS then incubated with DAPI (5 μg/mL in PBS) for 5
min. After a further three PBS washes, coverslips were removed from
the assay plate and slide mounted using ProLong Diamond Antifade
Mountant (ThermoFisher Scientific). Slides were subsequently imaged
using a spinning disc confocal microscope (Olympus IX81,
PerkinElmer, UK) by capturing confocal z-stack images in 0.5 μm
increments at 100× magnification.
Mouse Microsomal Stability Assay. Stock solutions of test
compounds (2 mM in DMSO) were diluted to 0.1 mM with
MeCN, and these solutions further diluted to 1.25 μM in 50 mM
potassium phosphate buffer (pH 7.4) containing 0.63 mg/mL
micromosal protein from pooled CD-1 (male) mouse liver microsome
preparation (20 mg/mL; BD Biosciences). An aliquot of the resultant
solution (80 μL) was preincubated for 5 min at 37 °C in a shaking
incubator before reaction was initiated by addition of NADPH (5 mM
in potassium phosphate buffer, 20 μL) or buffer alone (for no cofactor
control, 20 μL). Parallel reactions were stopped at 0 and 30 min time
points by the addition of two volumes of ice-cold acetonitrile. The
quenched samples were centrifuged (3000g, 30 min, 4 °C) and the
resultant supernatants analyzed for parent compound by LC-MS/MS,
monitoring ion transitions for parent and daughter ions. Results of
determinations carried out in duplicate are reported as mean percent
parent compound remaining at 30 min. Final incubation conditions
were 1 μM test compound, 0.5 mg/mL microsomal protein, and 1 mM
NADPH.
Bidirectional Caco-2 Permeability Assay. Caco-2 cells were seeded
onto 24-well transwell plates at 20000 cells/well and used as confluent
monolayers after a 21 day culture. Test and control compounds
(propranolol, vinblastine) were prepared at 10 μM in HBSS
supplemented with 25 mM HEPES, pH 7.4, at a final DMSO
concentration of 0.1%. These solutions were added to donor
compartments for both apical to basolateral (A → B) and basolateral
to apical (B → A) measurements. After incubation at 37 °C for 60
min, samples were analyzed by mass spectrometry using an analytical
internal standard. Apparent permeability (Papp) values, expressed as
×10−6 cm s−1, were determined according to eq 2 where Vacceptor and
Vdonor are the volumes of the transwell compartments (apical 125 μL,
basolateral 600 μL), T is the incubation time (in min), A is the area of
cells exposed (in cm2), and the concentrations are relative mass
spectroscopic responses normalized to the internal standard:
Tubulin Competitive Binding Assay. Assays were performed in 96-
well plates in duplicate. Biotinylated porcine tubulin (750 ng/well,
>99% purity, Cytoskeleton Inc.) was incubated with tritiated tubulin
binders (0.1 μM [3H]-colchicine (PerkinElmer), 0.2 μM [3H]-
vinblastine (ARC Inc.), or 0.05 μM [3H]-paclitaxel (BioTrend)) and
test compounds at various concentrations in 200 μL G-PEM buffer
containing 1 mM GTP at a final DMSO concentration of 3% and a
total volume of 200 μL/well. For the [3H]-colchicine binding assay,
the plate was incubated for 2 h at 37 °C, after which 79.2 μg/well
streptavidin yttrium silicate SPA beads (PerkinElmer) was added in 20
μL of G-PEM buffer (final volume 220 μL) then samples incubated for
a further 45 min at 37 °C before bound radioactivity was determined
using a Wallac 1450 Microbeta (PerkinElmer). For the [3H]-
vinblastine binding assay, SPA beads were added together with the
other assay components and the plate incubated for 15 min at rt before
reading. For the [3H]-paclitaxel binding assay, biotinylated tubulin was
precoupled to the SPA beads for 30 min at 4 °C, before the addition of
test compounds and radiolabel, after which the plate was incubated for
a further 25 min at rt before reading. Inhibition at each compound
concentration was calculated according to eq 1, where CPM = counts
per minute, CPM[NSB] is the reading obtained in the presence of a
1000-fold molar excess of unlabeled ligand, and CPM[TB] is the reading
for vehicle control (3% DMSO).
⎡
⎢
⎤
⎥
CPM[sample] − CPM[NSB]
CPM[TB] − CPM[NSB]
%inhibition = 100 −
× 100
⎢
⎣
⎥
⎦
(1)
IC50 values were determined from seven-point semilog concentration−
inhibition curves in XLfit (ID Business Solutions) or GraphPad Prism
6 (GraphPad Software, Inc.) using a four-parameter inhibition model.
Cell Cycle Analysis. FaDu cells were seeded into 12-well plates at
105 cells/well in 1 mL of full growth medium, then incubated
overnight to allow for cell attachment. Test compounds were prepared
at 2× final screening concentration in full growth medium then dosed
at 1 mL/well (final assay volume ∼2 mL), after which plates were
incubated at 37 °C for 24 h. Medium was removed and cells washed
with PBS (500 μL/well) then harvested by detachment with 0.05%
trypsin−EDTA (500 μL/well) followed by centrifugation (2000g, 5
min) in 1.5 mL microcentrifuge tubes. The cells were washed with
flow cytometry buffer (100 mM PBS pH 7.4, 0.1% w/v BSA, 0.1% w/v
NaN3; 500 μL) and centrifuged once more. Buffer was removed and
each cell pellet fixed by dropwise addition of ice-cold 70% ethanol (1.0
mL) with mixing using a vortex mixer. Samples were kept on ice for 30
min then centrifuged (2000g, 5 min) and the fixing solution removed.
After washing with flow cytometry buffer as above, samples were
treated with RNase A (100 μg/mL in PBS, 50 μL) at rt for at least 15
min, then propidium iodide (50 μg/mL in PBS, 200 μL) was added
and incubation continued overnight, in the dark. Samples were
analyzed by flow cytometry using a FACSCalibur instrument (BD
Biosciences), collecting a minimum of 10000 events per sample and
data was then analyzed using Flowing Software 2.5.1 (www.
flowingsoftware.com).
Immunofluorescence Staining. FaDu cells were seeded onto glass
coverslips in 24-well plates at 15000 cells/well in 500 μL of full growth
medium, then incubated for 24 h to allow for cell attachment. Test
compounds were prepared at 2× final assay concentration in full
growth medium then added at 500 μL/well and the plate incubated for
24 h. Medium was removed by aspiration and the cells washed with
PBS, then treated for 3 min with PHEM buffer (60 mM PIPES, 25
mM HEPES, 10 mM EGTA, 2 mM MgCl2, pH 6.9) containing 0.5%
v/v Triton X-100 and 10 μM paclitaxel (this treatment removes
unassembled tubulin dimers while preserving intact microtubules40).
Each culture was fixed by treatment with PHEM containing 8% w/v
paraformaldehyde and 0.3% w/v glutaraldehyde on ice for 10 min. The
⎧
⎩
⎫
⎪
⎛
⎝
⎞
⎠
[acceptor ] × V
⎪
V
acceptor
final
donor
× 106
(2)
⎨⎜
⎟
⎟
⎬
⎪
⎭
P
=
/(T × 60)
×
⎜
app
⎪
[donorinitial] × V
A
donor
Lucifer yellow was added to the apical buffer in each well to assess
cell viability as a control for integrity of the monolayer. Results from
any well with a lucifer yellow Papp > 10 × 10−6 cm s−1 were rejected.
Efflux ratios are expressed as Papp(B → A)/Papp(A → B). Values for
control compounds were as expected: propranolol Papp(A → B) 66.7 ×
10−6 cm s−1, efflux ratio 0.7; vinblastine Papp(A → B) < 0.2 × 10−6 cm
s−1, efflux ratio >288.
Multicellular Tumor Spheroid Cytotoxicity Screen. Multicellular
tumor spheroids were prepared from the FaDu cell line using the
forced-floating method, in 100 μL/well full growth medium in 96-well
tissue culture plates, as described previously.95 After spheroids had
formed, test compounds were dosed essentially as for the 2D
cytotoxicity assay but using n = 6 rather than triplicates,
accommodating a nine-point dose−response experiment with vehicle
control (DMSO) on each plate. Images were captured at regular
intervals (0, 1, 2, 3, 4, and 7 days postdosing) using a Zeiss Axiovert
200 M light microscope with AxioCam MRm camera and AxioVision
4.6 software (Carl Zeiss Inc., Germany) and size measurements made
using AxioVision. These values were converted into spheroid volume
expressed relative to DMSO control and EC50 values determined by
U
DOI: 10.1021/acs.jmedchem.5b01312
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
nonlinear regression using GraphPad Prism 6 (GraphPad Software,
Inc.), as above.
Sheffield) for assistance with mass spectrometry, and Dr. Simon
Tazzyman (University of Sheffield) for valuable help with
confocal microscopy.
Mouse Xenograft Model. Male CD1 athymic mice (aged 7−8
weeks, stock number 000711; Charles River UK) were kept in
ventilated cages with food and water provided ad libitum. FaDu cells,
verified mycoplasma free, were suspended in PBS (pH 7.4) at a density
of 5 × 107 mL−1 and each animal received a subcutaneous injection of
5 × 106 cells (100 μL) from this suspension. Once xenografts were
established, tumors were measured with a digital caliper and the
volume calculated as (length × width2)/2. Mice with palpable tumors
(100−300 mm3) were assigned randomly into groups (n = 9) before
beginning the treatment schedule. The control group received 200 μL
vehicle (95:5 water/glycerol containing 0.25% v/v Tween-20 and 2.3%
v/v DMSO) by oral gavage once daily for 10 days, with the treatment
groups receiving either 33 (10 mg/kg), 59 (10 mg/kg), or 59 (20 mg/
kg) diluted from stock solutions in 100% DMSO in 200 μL of the
same formulation, by the same route and schedule. In a parallel
experiment to compare the effect of vinblastine treatment, two groups
received either vehicle (200 μL PBS) or vinblastine (3 mg/kg in 200
μL PBS) by the intraperitoneal route on days 1, 4, and 7 only. For all
groups, tumor volumes were monitored regularly, and mice were
culled when tumors reached the maximum permitted size (15 mm
diameter). All animal procedures were conducted in accordance with
the UK Home Office Regulations under the Animals (Scientific
Procedures) Act 1986 and the awarded project license number under
which these protocols were performed is PPL:40/3424. In addition,
the University of Sheffield Animal Welfare and Ethical Review Body
approved all the in vivo experiments used in this study.
ABBREVIATIONS USED
■
ABCG2, ATP-binding cassette subfamily G member 2; Akt,
RAC-α serine/threonine-protein kinase; BCRP, breast cancer
resistance protein; DAPI, 4′,6-diamidino-2-phenylindole;
DIPEA, N,N-diisopropylethylamine; DLS, dynamic light
scattering; 5-FU, 5-fluorouracil; GTP, guanosine 5′-triphos-
phate; HBSS, Hanks’ Balanced Salt Solution; HNC, head and
neck cancer; LC50, lethal concentration for 50% of the
population; MCTS, multicellular tumor spheroid; MMPA,
matched molecular pair analysis; MRP1, multidrug resistance-
associated protein 1; MRP2, multidrug resistance-associated
protein 2; MTA, microtubule-targeting agent; mTor, mamma-
lian target of rapamycin; PAINS, pan-assay interference
compounds; P-gp, P-glycoprotein; PN, peripheral neuropathy;
SEM, standard error of the mean; TBD, 1,5,7-triazabicyclo-
[4.4.0]dec-5-ene
REFERENCES
■
(1) Nitika, V.; Kapil, K. Microtubule targeting agents: a benchmark in
cancer therapy. Curr. Drug Ther. 2014, 8, 189−196.
(2) Risinger, A. L.; Giles, F. J.; Mooberry, S. L. Microtubule dynamics
as a target in oncology. Cancer Treat. Rev. 2009, 35, 255−261.
(3) Stanton, R. A.; Gernert, K. M.; Nettles, J. H.; Aneja, R. Drugs that
target dynamic microtubules: a new molecular perspective. Med. Res.
Rev. 2011, 31, 443−481.
(4) Barbier, P.; Tsvetkov, P. O.; Breuzard, G.; Devred, F. Deciphering
the molecular mechanisms of anti-tubulin plant derived drugs.
Phytochem. Rev. 2014, 13, 157−169.
(5) Rohena, C. C.; Mooberry, S. L. Recent progress with microtubule
stabilizers: new compounds, binding modes and cellular activities. Nat.
Prod. Rep. 2014, 31, 335−355.
(6) Jordan, M. A.; Horwitz, S. B.; Lobert, S.; Correia, J. J. Exploring
the mechanisms of action of the novel microtubule inhibitor
vinflunine. Semin. Oncol. 2008, 35, S6−S12.
ASSOCIATED CONTENT
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.jmed-
chem.5b01312.
■
S
Characterization data, purification details, and HPLC
purity for all screening compounds; NMR spectra for all
screening compounds and synthetic intermediates (PDF)
Molecular formula strings (CSV)
AUTHOR INFORMATION
Corresponding Authors
*For M.J.T.: phone +44 (0)114 222 9364; E-mail: m.j.
thompson@sheffield.ac.uk.
■
(7) Field, J. J.; Kanakkanthara, A.; Miller, J. H. Microtubule-targeting
agents are clinically successful due to both mitotic and interphase
impairment of microtubule function. Bioorg. Med. Chem. 2014, 22,
5050−5059.
(8) Zasadil, L. M.; Andersen, K. A.; Yeum, D.; Rocque, G. B.; Wilke,
L. G.; Tevaarwerk, A. J.; Raines, R. T.; Burkard, M. E.; Weaver, B. A.
Cytotoxicity of paclitaxel in breast cancer is due to chromosome
missegregation on multipolar spindles. Sci. Transl. Med. 2014, 6,
229ra43.
*For H.E.C.: phone +44 (0)114 271 7966; E-mail: h.colley@
sheffield.ac.uk.
Author Contributions
∇M.J.T. wrote the manuscript with assistance from the other
authors specified (H.E.C., M.W., E.J.G., and M.M.). All authors
have approved the manuscript prior to submission.
(9) Ogden, A.; Rida, P. C. G.; Reid, M. D.; Aneja, R. Interphase
microtubules: chief casualties in the war on cancer? Drug Discovery
Today 2014, 19, 824−829.
Notes
(10) Furst, R.; Vollmar, A. M. A new perspective on old drugs: non-
̈
The authors declare no competing financial interest.
mitotic actions of tubulin-binding drugs play a major role in cancer
treatment. Pharmazie 2013, 68, 478−483.
ACKNOWLEDGMENTS
■
(11) Mitchison, T. J. The proliferation rate paradox in antimitotic
chemotherapy. Mol. Biol. Cell 2012, 23, 1−6.
We gratefully acknowledge funding from Weston Park Hospital
Cancer Charity (grant refs SG110 to M.J.T., H.E.C., and S.J.D.;
CA135 to M.J.T., M.M., H.E.C., and S.J.D.), Sheffield
Healthcare Gateway Faculty Innovation Fund (University of
Sheffield, to H.E.C. and M.J.T.), and the University of Sheffield
Early Career Researcher Scheme (grants to M.J.T. and H.E.C.).
We thank Michael Briggs, Sharon Wicks, Lisa Strong, Richard
Newman, Roshini Markandu, and Lynsey Crowley (Charles
River) for their contributions to in vitro ADME work, and also
Drs. David Clark, Steve Price, and Mark Albertella (Charles
River) for helpful collaboration; Simon Thorpe (University of
(12) Seligmann, J.; Twelves, C. Tubulin: an example of targeted
chemotherapy. Future Med. Chem. 2013, 5, 339−352.
(13) Kavallaris, M. Microtubules and resistance to tubulin-binding
agents. Nat. Rev. Cancer 2010, 10, 194−204.
(14) Katsetos, C. D.; Draber, P. Tubulins as therapeutic targets in
cancer: from bench to bedside. Curr. Pharm. Des. 2012, 18, 2778−
2792.
(15) Murray, S.; Briasoulis, E.; Linardou, H.; Bafaloukos, D.;
Papadimitriou, C. Taxane resistance in breast cancer: mechanisms,
predictive biomarkers and circumvention strategies. Cancer Treat. Rev.
2012, 38, 890−903.
V
DOI: 10.1021/acs.jmedchem.5b01312
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
(16) Gan, P. P.; Kavallaris, M. Tubulin-targeted drug action:
functional significance of class II and class IVb β-tubulin in Vinca
alkaloid sensitivity. Cancer Res. 2008, 68, 9817−9824.
(36) Lu, Y.; Chen, J.; Xiao, M.; Li, W.; Miller, D. D. An overview of
tubulin inhibitors that interact with the colchicine binding site. Pharm.
Res. 2012, 29, 2943−2971.
(37) Patil, S. A.; Patil, R.; Miller, D. D. Indole molecules as inhibitors
of tubulin polymerization: potential new anticancer agents. Future
Med. Chem. 2012, 4, 2085−2115.
(38) Kaur, R.; Kaur, G.; Gill, R. K.; Soni, R.; Bariwal, J. Recent
developments in tubulin polymerization inhibitors: an overview. Eur. J.
Med. Chem. 2014, 87, 89−124.
(39) Liu, Y.-M.; Chen, H.-L.; Lee, H.-Y.; Liou, J.-P. Tubulin
inhibitors: a patent review. Expert Opin. Ther. Pat. 2014, 24, 69−88.
(40) Wienecke, A.; Bacher, G. Indibulin, a novel microtubule
inhibitor, discriminates between mature neuronal and nonneuronal
tubulin. Cancer Res. 2009, 69, 171−177.
(41) Desai, A.; Ratain, M. J.; Moshier, K.; Tipton, M.; Dooley, W.;
Hocknell, K.; Dahl, T.; Sherman, M.; Limentani, S. A phase I, dose-
escalation trial of STA-5312, a microtubule inhibitor with a novel
binding site, in advanced or metastatic solid malignancies. J. Clin.
Oncol. 2006, 24, 13040, 2006 ASCO Annual Meeting Proceedings
Part I.
(42) (a) STA-5312 administered on alternate weekdays every two
weeks to patients with hematologic malignancies and patients with
solid tumors, NCT00088101. www.clinicaltrials.gov (accessed De-
cember 30, 2014). (b) A phase I study of STA-5312 in subjects with
advanced or metastatic solid tumors, NCT00276913. www.
clinicaltrials.gov (accessed December 30, 2014).
(43) Bacher, G.; Nickel, B.; Emig, P.; Vanhoefer, U.; Seeber, S.;
Shandra, A.; Klenner, T.; Beckers, T. D-24851, a novel synthetic
microtubule inhibitor, exerts curative antitumoral activity in vivo,
shows efficacy toward multidrug-resistant tumor cells, and lacks
neurotoxicity. Cancer Res. 2001, 61, 392−399.
(44) Kuppens, I. E. L. M.; Witteveen, P. O.; Schot, M.; Schuessler, V.
M.; Daehling, A.; Beijnen, J. H.; Voest, E. E.; Schellens, J. H. M. Phase
I dose-finding and pharmacokinetic trial of orally administered
Indibulin (D-24851) to patients with solid tumors. Invest. New Drugs
2007, 25, 227−235.
(17) Carlson, K.; Ocean, A. J. Peripheral neuropathy with
microtubule-targeting agents: occurrence and management approach.
Clin. Breast Cancer 2011, 11, 73−81.
(18) Canta, A.; Chiorazzi, A.; Cavaletti, G. Tubulin: a target for
antineoplastic drugs into the cancer cells but also in the peripheral
nervous system. Curr. Med. Chem. 2009, 16, 1315−1324.
(19) Dumontet, C.; Jordan, M. A. Microtubule-binding agents: a
dynamic field of cancer therapeutics. Nat. Rev. Drug Discovery 2010, 9,
790−803.
(20) Kingston, D. G. I.; Snyder, J. P. The quest for a simple bioactive
analog of paclitaxel as a potential anticancer agent. Acc. Chem. Res.
2014, 47, 2682−2691.
(21) Gerullis, H.; Ecke, T.; Eimer, C.; Wishahi, M.; Otto, T.
Vinflunine as second-line treatment in platin-resistant metastatic
urothelial carcinoma: a review. Anti-Cancer Drugs 2011, 22, 9−17.
(22) Gourmelon, C.; Frenel, J. S.; Campone, M. Eribulin mesylate for
the treatment of late-stage breast cancer. Expert Opin. Pharmacother.
2011, 12, 2883−2890.
(23) Preston, J. N.; Trivedi, M. V. Eribulin: a novel cytotoxic
chemotherapy agent. Ann. Pharmacother. 2012, 46, 802−811.
(24) Cortes, J.; Montero, A. J.; Gluck, S. Eribulin mesylate, a novel
̈
microtubule inhibitor in the treatment of breast cancer. Cancer Treat.
Rev. 2012, 38, 143−151.
(25) Brogdon, C. F.; Lee, F. Y.; Canetta, R. M. Development of other
microtubule-stabilizer families: the epothilones and their derivatives.
Anti-Cancer Drugs 2014, 25, 599−609.
(26) Edelman, M. J.; Shvartsbeyn, M. Epothilones in development for
non-small-cell lung cancer: novel anti-tubulin agents with the potential
to overcome taxane resistance. Clin. Lung Cancer 2012, 13, 171−180.
(27) Sadeghi, S.; Olevsky, O.; Hurvitz, S. A. Profiling and targeting
HER2-positive breast cancer using trastuzumab emtansine. Pharmaco-
genomics Pers. Med. 2014, 7, 329−338.
(28) Coen van Hasselt, J. G.; Gupta, A.; Hussein, Z.; Beijnen, J. H.;
Schellens, J. H. M.; Huitema, A. D. R. Population pharmacokinetic−
pharmacodynamic analysis for eribulin mesylate-associated neutrope-
nia. Br. J. Clin. Pharmacol. 2013, 76, 412−424.
(29) Valero, V. Managing ixabepilone adverse events with dose
reduction. Clin. Breast Cancer 2013, 13, 1−6.
(45) Oostendorp, R. L.; Witteveen, P. O.; Schwartz, B.; Vainchtein,
L. D.; Schot, M.; Nol, A.; Rosing, H.; Beijnen, B. H.; Voest, E. E.;
Schellens, J. H. M. Dose-finding and pharmacokinetic study of orally
administered Indibulin (D-24851) to patients with advanced solid
tumors. Invest. New Drugs 2010, 28, 163−170.
(30) Ebenezer, G. J.; Carlson, K.; Donovan, D.; Cobham, M.;
Chuang, E.; Moore, A.; Cigler, T.; Ward, M.; Lane, M. E.; Ramnarain,
A.; Vahdat, L. T.; Polydefkis, M. Ixabepilone-induced mitochondria
and sensory axon loss in breast cancer patients. Ann. Clin. Transl.
Neurol. 2014, 1, 639−649.
(31) Vahdat, L. T.; Garcia, A. A.; Vogel, C.; Pellegrino, C.; Lindquist,
D. L.; Iannotti, N.; Gopalakrishna, P.; Sparano, J. A. Eribulin mesylate
versus ixabepilone in patients with metastatic breast cancer: a
randomized phase II study comparing the incidence of peripheral
neuropathy. Breast Cancer Res. Treat. 2013, 140, 341−351.
(32) Wozniak, K. M.; Nomoto, K.; Lapidus, R. G.; Wu, Y.; Carozzi,
V.; Cavaletti, G.; Hayakawa, K.; Hosokawa, S.; Towle, M. J.; Littlefield,
B. A.; Slusher, B. S. Comparison of neuropathy-inducing effects of
eribulin mesylate, paclitaxel and ixabepilone in mice. Cancer Res. 2011,
71, 3952−3962.
(33) LaPointe, N. E.; Morfini, G.; Brady, S. T.; Feinstein, S. C.;
Wilson, L.; Jordan, M. A. Effects of eribulin, vincristine, paclitaxel and
ixabepilone on fast axonal transport and kinesin-1 driven microtubule
gliding: implications for chemotherapy-induced peripheral neuropathy.
NeuroToxicology 2013, 37, 231−239.
(34) Polastro, L.; Aftimos, P. G.; Awada, A. Eribulin mesylate in the
management of metastatic breast cancer and other solid cancers: a
drug review. Expert Rev. Anticancer Ther. 2014, 14, 649−665.
(35) Laughney, A. M.; Kim, E.; Sprachman, M. M.; Miller, M. A.;
Kohler, R. H.; Yang, K. S.; Orth, J. D.; Mitchison, T. J.; Weissleder, R.
Single-cell pharmacokinetic imaging reveals a therapeutic strategy to
overcome drug resistance to the microtubule inhibitor eribulin. Sci.
Transl. Med. 2014, 6, 261ra152.
(46) Semenova, M. N.; Kiselyov, A. S.; Titov, I. Y.; Raihstat, M. M.;
Molodtsov, M.; Grishchuk, E.; Spiridonov, I.; Semenov, V. V. In vivo
evaluation of indolyl glyoxylamides in the phenotypic sea urchin
embryo assay. Chem. Biol. Drug Des. 2007, 70, 485−490.
(47) (a) Ritchie, T. J.; Macdonald, S. J. F. The impact of aromatic
ring count on compound developability − are too many aromatic rings
a liability in drug design? Drug Discovery Today 2009, 14, 1011−1020.
(b) Ritchie, T. J.; Macdonald, S. J. F.; Young, R. J.; Pickett, S. D. The
impact of aromatic ring count on compound developability: further
insights by examining carbo- and hetero-aromatic and -aliphatic ring
types. Drug Discovery Today 2011, 16, 164−171.
(48) Lovering, F.; Bikker, J.; Humblet, C. Escape from flatland:
increasing saturation as an approach to improving clinical success. J.
Med. Chem. 2009, 52, 6752−6756.
(49) Ishikawa, M.; Hashimoto, Y. Improvement in aqueous solubility
in small molecule drug discovery programs by disruption of molecular
planarity and symmetry. J. Med. Chem. 2011, 54, 1539−1554.
(50) Mok, N. Y.; Maxe, S.; Brenk, R. Locating sweet spots for
screening hits and evaluating pan-assay interference filters from the
performance analysis of two lead-like libraries. J. Chem. Inf. Model.
2013, 53, 534−544.
(51) Hennemann, B. Palliative chemotherapy of head and neck
cancer: present status and future development. Laryngorhinootologie
2006, 85, 172−178.
(52) Cullen, K. J.; Schumaker, L.; Nikitakis, N.; Goloubeva, O.; Tan,
M.; Sarlis, M. J.; Haddad, R. I.; Posner, M. R. β-Tubulin-II expression
strongly predicts outcome in patients receiving induction chemo-
therapy for locally advanced squamous carcinoma of the head and
W
DOI: 10.1021/acs.jmedchem.5b01312
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
neck: a companion analysis of the TAX 324 trial. J. Clin. Oncol. 2009,
27, 6222−6228.
(53) Schena, M.; Barone, C.; Birocco, N.; Dongiovanni, D.; Numico,
G.; Colantonio, I.; Bertetto, O. Weekly cisplatin paclitaxel and
continuous infusion fluorouracil in patients with recurrent and/or
metastatic head and neck squamous cell carcinoma: a phase II study.
Cancer Chemother. Pharmacol. 2005, 55, 271−276.
(54) Pointreau, Y.; Garaud, P.; Chapet, S.; Sire, C.; Tuchais, C.;
Tortochaux, J.; Faivre, S.; Guerrif, S.; Alfonsi, M.; Calais, G.
Randomized trial of induction chemotherapy with cisplatin and 5-
fluorouracil with or without docetaxel for larynx preservation. J. Natl.
Cancer Inst. 2009, 101, 498−506.
properties: a study of aqueous solubility, plasma protein binding and
oral exposure. J. Med. Chem. 2006, 49, 6672−6682. (b) Gleeson, P.;
Bravi, G.; Modi, S.; Lowe, D. ADMET rules of thumb II: a comparison
of the effects of common substituents on a range of ADMET
parameters. Bioorg. Med. Chem. 2009, 17, 5906−5919. (c) Griffen, E.;
Leach, A. G.; Robb, G. R.; Warner, D. J. Matched molecular pairs as a
medicinal chemistry tool. J. Med. Chem. 2011, 54, 7739−7750.
(67) Fishburn, C. S. Attenuating attrition. SciBX 2013, 6(26),
doi:10.1038/scibx.2013.647.
(68) Hussain, J.; Rea, C. Computationally efficient algorithm to
identify matched molecular pairs (MMPs) in large data sets. J. Chem.
Inf. Model. 2010, 50, 339−348.
(55) Paccagnella, A.; Ghi, M. G.; Loreggian, L.; Buffoli, A.; Koussis,
H.; Mione, C. A.; Bonetti, A.; Campostrini, F.; Gardani, G.; Ardizzoia,
A.; Dondi, D.; Guaraldi, M.; Cavallo, R.; Tomio, L.; Gava, A.
Concomintant chemoradiotherapy versus induction docetaxel, cispla-
tin and 5-fluorouracil (TPF) followed by concomitant chemo-
radiotherapy in locally advanced head and neck cancer: a phase II
randomized study. Ann. Oncol. 2010, 21, 1515−1522.
(69) Warner, D. J.; Griffen, E.; St-Gallay, S. A. WizePairZ: a novel
algorithm to identify, encode, and exploit matched molecular pairs
with unspecified cores in medicinal chemistry. J. Chem. Inf. Model.
2010, 50, 1350−1357.
(70) Papadatos, G.; Alkarouri, M.; Gillet, V. J.; Willett, P.;
Kadirkamanathan, V.; Luscombe, C. N.; Bravi, G.; Richmond, N. J.;
Pickett, S. D.; Hussain, J.; Pritchard, J. M.; Cooper, A. W. J.;
Macdonald, S. J. F. Lead optimization using matched molecular pairs:
inclusion of contextual information for enchanced prediction of hERG
inhibition, solubility and lipophilicity. J. Chem. Inf. Model. 2010, 50,
1872−1886.
(71) Thompson, M. J.; Louth, J. C.; Ferrara, S.; Jackson, M. P.;
Sorrell, F. J.; Cochrane, E. J.; Gever, J.; Baxendale, S.; Silber, M. B.;
Roehl, H. H.; Chen, B. Discovery of 6-substituted indole-3-
glyoxylamides as lead antiprion agents with enhanced cell line activity,
improved microsomal stability and low toxicity. Eur. J. Med. Chem.
2011, 46, 4125−4132.
(72) Zhang, Z.; Yang, Z.; Wong, H.; Zhu, J.; Meanwell, N. A.;
Kadow, J. F.; Wang, T. An effective procedure for the acylation of
azaindoles at C-3. J. Org. Chem. 2002, 67, 6226−6227.
(73) Sabot, C.; Kumar, K. A.; Meunier, S.; Mioskowski, C. A
convenient aminolysis of esters catalyzed by 1,5,7-triazabicyclo[4.4.0]-
dec-5-ene (TBD) under solvent-free conditions. Tetrahedron Lett.
2007, 48, 3863−3866.
(74) Tsuritani, T.; Strotman, N. A.; Yamamoto, Y.; Kawasaki, M.;
Yasuda, N.; Mase, T. N-Cyclopropylation of indoles and cyclic amides
with copper(II) reagent. Org. Lett. 2008, 10, 1653−1655.
(75) Nguyen, T. M.; Duong, H. A.; Richard, J.-A.; Johannes, C. W.;
Pincheng, F.; Ye, D. K. J.; Shuying, E. L. Cascade fluorofunctionali-
satdon of 2,3-unsubstituted indoles by means of electrophilic
fluorination. Chem. Commun. 2013, 49, 10602−10604.
(76) (a) Conner, S. E.; Gossett, L. S.; Green, J. E.; Jones, W. D., Jr.;
Mantlo, N. B.; Matthews, D. P.; Mayhugh, D. R.; Smith, D. L.; Vance,
J. A.; Wang, X.; Warshawsky, A. M.; Winneroski, L. L., Jr.; Xu, Y.; Zhu,
G. Preparation of sulfonamide derivatives, in particular N,N-benzo-
[b]thiophene sulfonamides, as PPAR modulators, especially PPAR
agonists. (Eli Lilly & Co., USA) WO 2004073606 A2, September 2,
2004; SciFinder Scholar AN 2004:718289;. (b) Wilson, D.; Fanning, L.
T. D.; Sheth, U.; Martinborough, E.; Termin, A.; Neubert, T.;
Zimmermann, N.; Knoll, T.; Whitney, T.; Kawatkar, A.; Lehsten, D.;
Stamos, D.; Zhou, J.; Arumugam, V.; Gutierrez, C. Heterocyclic
derivatives as modulators of ion channels and their preparation,
pharmaceutical compositions and use in the treatment of diseases.
(Vertex Pharmaceuticals Inc., USA) WO 2007075892 A2, July 5, 2007;
SciFinder Scholar AN 2007:730896.
(56) Theile, D.; Ketabi-Kiyanvash, N.; Herold-Mende, C.; Dyckhoff,
G.; Efferth, T.; Bertholet, V.; Haefeli, W. E.; Weiss, J. Evaluation of
drug transporters’ significance in for multidrug resistance in head and
neck squamous cell carcinoma. Head Neck 2011, 33, 959−968.
(57) (a) Jacobs, C.; Lyman, G.; Velez-Garcia, E.; Sridhar, K. S.;
Knight, W.; Hochster, H.; Goodnough, L. T.; Mortimer, J. E.; Einhorn,
L. H.; Schacter, L. A phase III randomized study comparing cisplatin
and fluorouracil as single agents and in combination for advanced
squamous cell carcinoma of the head and neck. J. Clin. Oncol. 1992, 10,
257−263. (b) Forastiere, A. A.; Metch, B.; Schuller, D. E.; Ensley, J. F.;
Hutchins, L. F.; Triozzi, P.; Kish, J. A.; McClure, S.; VonFeldt, E.;
Williamson, S. K. Randomized comparison of cisplatin plus
fluorouracil and carboplatin plus fluorouracil versus methotrexate in
advanced squamous-cell carcinoma of the head and neck: a Southwest
Oncology Group study. J. Clin. Oncol. 1992, 10, 1245−1251.
(58) Lee, J.; Moon, C. Current status of experimental therapeutics for
head and neck cancer. Exp. Biol. Med. 2011, 236, 375−389.
(59) Machiels, J. P.; Schmitz, S. New advances in targeted therapies
for squamous cell carcinoma of the head and neck. Anti-Cancer Drugs
2011, 22, 626−633.
(60) Matta, A.; Ralhan, R. Overview of current and future biologically
based targeted therapies in head and neck squamous cell carcinoma.
Head Neck Oncol. 2009, 1, 6.
(61) Raza, S.; Kornblum, N.; Kancharla, V. P.; Baig, M. A.; Singh, A.
B.; Kavalar, M. Emerging therapies in the treatment of locally
advanced squamous cell carcinomas of the head and neck. Recent Pat.
Anti-Cancer Drug Discovery 2011, 6, 246−257.
(62) Denaro, N.; Russi, E. G.; Colantonio, I.; Adamo, V.; Merlano,
M. C. The role of antiangiogenic agents in the treatment of head and
neck cancer. Oncology 2012, 83, 108−116.
(63) (a) Faller, B.; Ertl, P. Computational approaches to determine
drug solubility. Adv. Drug Delivery Rev. 2007, 59, 533−545.
(b) Delaney, J. S. Predicting aqueous solubility from structure. Drug
Discovery Today 2005, 10, 289−295.
(64) (a) Tetko, I. V.; Tanchuk, V. Y.; Kasheva, T. N.; Villa, A. E.
Estimation of aqueous solubility of chemical compounds using E-state
indices. J. Chem. Inf. Model. 2001, 41, 1488−1493. (b) Tetko, I. V.;
Gasteiger, J.; Todeschini, R.; Mauri, A.; Livingstone, D.; Ertl, P.;
Palyulin, V. A.; Radchenko, E. V.; Zefirov, N. S.; Makarenko, A. S.;
Tanchuk, V. Y.; Prokopenko, V. V. Virtual computational chemistry
laboratory − design and description. J. Comput.-Aided Mol. Des. 2005,
19, 453−463. (c) VCCLAB; Virtual Computational Chemistry
Laboratory; http://www.vcclab.org, (accessed October 17, 2015).
(65) Tetko, I. V.; Tanchuk, V. Y.; Kasheva, T. N.; Villa, A. E. P.
Internet software for the calculation of the lipophilicity and aqeuous
solubility of chemical compounds. J. Chem. Inf. Model. 2001, 41, 246−
252.
(77) Yeom, C.-E.; Kim, M. J.; Kim, B. M. 1,8-Diazabycyclo[5.4.0]-
undec-7-ene (DBU)-promoted efficient and versatile aza-Michael
reaction. Tetrahedron 2007, 63, 904−909.
(78) Amato, G.; Roeloffs, R.; Rigdon, G. C.; Antonio, B.; Mersch, T.;
McNaughton-Smith, G.; Wickenden, A. G.; Fritch, P.; Suto, M. J. N-
Pyridyl and pyrimidine benzamides as KCNQ2/Q3 potassium channel
openers for the treatment of epilepsy. ACS Med. Chem. Lett. 2011, 2,
481−484.
(79) Kel'in, A. V.; Sromek, A. W.; Gevorgyan, V. A novel Cu-assisted
cycloisomerization of alkynyl imines: efficient synthesis of pyrroles and
pyrrole-containing heterocycles. J. Am. Chem. Soc. 2001, 123, 2074−
2075.
(66) (a) Leach, A. G.; Jones, H. D.; Cosgrove, D. A.; Kenny, P. W.;
Ruston, L.; MacFaul, P.; Wood, J. M.; Colclough, N.; Law, B. Matched
molecular pairs as a guide in the optimization of pharmaceutical
X
DOI: 10.1021/acs.jmedchem.5b01312
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
(80) Li, H.; Xia, Z.; Chen, S.; Koya, K.; Ono, M.; Sun, L.
Development of a practical synthesis of STA-5312, a novel indolizine
oxalylamide microtubule inhibitor. Org. Process Res. Dev. 2007, 11,
246−250.
(81) Debnar, T.; Dreisigacker, S.; Menche, D. Highly regioselective
opening of zirconacyclopentadienes by remote coordination: concise
synthesis of the furan core of the leupyrrins. Chem. Commun. 2013, 49,
725−727.
cultures. Oncol. Lett. 2015, 10, 1281−1286. (b) Tupper, J.; Greco, O.;
Tozer, G. M.; Dachs, G. U. Analysis of the horseradish peroxidase/
indole-3-acetic acid combination in a three-dimensional tumor model.
Cancer Gene Ther. 2004, 11, 508−513.
(97) Huang, T.-H.; Chiu, S.-J.; Chiang, P.-H.; Chiou, S.-H.; Li, W.-T.;
Chen, C.-T.; Chang, C. A.; Chen, J.-C.; Lee, Y.-J. Antiproliferative
effects of N-heterocyclic indole glyoxylamide derivatives on human
lung cancer cells. Anticancer Res. 2011, 31, 3407−3416.
(98) Li, W.-T.; Yeh, T.-K.; Song, J.-S.; Yang, Y.-N.; Chen, T.-W.; Lin,
C.-H.; Chen, C.-P.; Shen, C.-C.; Hsieh, C.-C.; Lin, H.-L.; Chao, Y.-S.;
Chen, C.-T. BPR0C305, and orally active microtubule-disrupting
anticancer agent. Anti-Cancer Drugs 2013, 24, 1047−1057.
(99) Hu, C.-B.; Chen, C.-P.; Yeh, T.-K.; Song, J.-S.; Chang, C.-Y.;
Chuu, J.-J.; Tung, F.-F.; Ho, P.-Y.; Chen, T.-W.; Lin, C.-H.; Wang, M.-
H.; Chang, K.-Y.; Huang, C.-L.; Lin, H.-L.; Li, W.-T.; Hwang, D.-R.;
Chern, J.-H.; Hwang, L.-L.; Chang, J.-Y.; Chao, Y.-S.; Chen, C.-T.
BPR0C261 is a novel orally active antitumor agent with antimitotic
and anti-angiogenic activities. Cancer Sci. 2011, 102, 182−191.
(100) Li, W.-T.; Hwang, D.-R.; Chen, C.-P.; Shen, C.-W.; Huang, C.-
L.; Chen, T.-W.; Lin, C.-H.; Chang, Y.-L.; Chang, Y.-Y.; Lo, Y.-K.;
Tseng, H.-Y.; Lin, C.-C.; Song, J.-S.; Chen, H.-C.; Chen, S.-J.; Wu, S.-
H.; Chen, C.-T. Synthesis and biological evaluation of N-heterocyclic
indolyl glyoxylamides as orally active anticancer agents. J. Med. Chem.
2003, 46, 1706−1715.
(101) Thompson, M. J.; Louth, J. C.; Ferrara, S.; Sorrell, F. J.; Irving,
B. J.; Cochrane, E. J.; Meijer, A. J. H. M.; Chen, B. Structure−activity
relationship refinement and further assessment of indole-3-glyoxyla-
mides as a lead series against prion disease. ChemMedChem 2011, 6,
115−130.
(102) Baell, J. B. Screening-based translation of public research
encounters painful problems. ACS Med. Chem. Lett. 2015, 6, 229−234.
(103) Fitton, A. O.; Hill, J.; Jane, D. E.; Millar, R. Synthesis of simple
oxetanes carrying reactive 2-sustituents. Synthesis 1987, 1987, 1140−
1142.
(82) Chen, Y.-J.; Huang, W.-C.; Wei, Y.-L.; Hsu, S.-C.; Yuan, P.; Lin,
H. Y.; Wistuba, I. I.; Lee, J. J.; Yen, C.-J.; Su, W.-C.; Chang, K.-Y.;
Chang, W.-C.; Chou, T.-C.; Chou, C.-K.; Tsai, C.-H.; Hung, M.-C.
Elevated BCRP/ABCG2 expression confers acquired resistance to
gefitinib in wild-type EGFR-expressing cells. PLoS One 2011, 6,
e21428.
(83) (a) Shelanski, M. L.; Gaskin, F.; Cantor, C. R. Microtubule
assembly in the absence of added nucleotides. Proc. Natl. Acad. Sci. U.
S. A. 1973, 70, 765−768. (b) Lee, J. C.; Timasheff, S. N. In vitro
reconstitution of calf brain microtubules: effects of solution variables.
Biochemistry 1977, 16, 1754−1764.
̀
(84) Bonne, D.; Heusele, C.; Simon, C.; Pantaloni, D. 4′,6-
Diamidino-2-phenylindole, a fluorescent probe for tubulin and
microtubules. J. Biol. Chem. 1985, 260, 2819−2825.
(85) Cortese, F.; Bhattacharyya, B.; Wolff, J. Podophyllotoxin as a
probe for the colchicine binding site of tubulin. J. Biol. Chem. 1977,
252, 1134−1140.
(86) Wallner, B. P.; Schwartz, B. E.; Komarnitsky, P. B.; Bacher, G.;
Kutscher, B.; Raab, G. Use of indolyl-3-glyoxylic acid derivatives
including indibulin, alone or in combination with further agents for
treating cancer. (Ziopharm Oncology Inc., USA) WO 2008066807 A1,
Jun 5, 2008; SciFinder Scholar AN 2008:673085.
(87) Yamada, H. Y.; Gorbsky, G. J. Spindle checkpoint function and
cellular sensitivity to antimitotic drugs. Mol. Cancer Ther. 2006, 5,
2963−2969.
(88) Harker, W. G.; Sikic, B. I. Multidrug (pleiotropic) resistance in
doxorubicin-selected variants of the human sarcoma cell line MES-SA.
Cancer Res. 1985, 45, 4091−4096.
́ ́
(89) Chen, G.; Jaffrezou, J.-P.; Fleming, W. H.; Duran, G. E.; Sikic, B.
I. Prevalence of multidrug resistance related to activation of the Mdr1
gene in human sarcoma mutants derived by single-step doxorubicin
selection. Cancer Res. 1994, 54, 4980−4987.
(90) Nakamura, T.; Sakaeda, T.; Ohmoto, N.; Tamura, T.; Aoyama,
N.; Shirakawa, T.; Nakamura, T.; Kim, K. I.; Kim, S. R.; Kuroda, Y.;
Matsuo, M.; Kasuga, M.; Okumura, K. Real-time quantitative
polymerase chain reaction for MDR1, MRP1, MRP2 and CYP3A-
mRNA levels in Caco-2 cell lines, human duodenal enterocytes,
normal colorectal tissues, and colorectal adenocarcinomas. Drug
Metab. Dispos. 2002, 30, 4−6.
(91) Gutmann, H.; Fricker, G.; Torok, M.; Michael, S.; Beglinger, C.;
̈
̈
Drewe, J. Evidence for different ABC-transporters in Caco-2 cells
modulating drug uptake. Pharm. Res. 1999, 16, 402−407.
(92) Xia, C. Q.; Liu, N.; Yang, D.; Miwa, G.; Gan, L.-S. Expression,
localization and functional characteristics of breast cancer resistance
protein in Caco-2 cells. Drug Metab. Dispos. 2005, 33, 637−643.
(93) Breslin, S.; O’Driscoll, L. Three-dimensional cell culture: the
missing link in drug discovery. Drug Discovery Today 2013, 18, 240−
249.
(94) Hirschhaeuser, F.; Menne, H.; Dittfeld, C.; West, J.; Mueller-
Klieser, W.; Kunz-Schughart, L. A. Multicellular tumor spheroids: an
underestimated tool is catching up again. J. Biotechnol. 2010, 148, 3−
15.
(95) Colley, H. E.; Hearnden, V.; Avila-Olias, M.; Cecchin, D.;
Canton, I.; Madsen, J.; MacNeil, S.; Warren, N.; Hu, K.; McKeating, J.
A.; Armes, S. P.; Murdoch, C.; Thornhill, M. H.; Battaglia, G.
Polymersome-mediated delivery of combination anticancer therapy to
head and neck cancer cells: 2D and 3D in vitro evaluation. Mol.
Pharmaceutics 2014, 11, 1176−1188.
(96) (a) Kadletz, L.; Heiduschka, G.; Domayer, J.; Schmid, R.;
Enzenhofer, E.; Thurnher, D. Evaluation of spheroid head and neck
squamous cell carcinoma cell models in comparison to monolayer
Y
DOI: 10.1021/acs.jmedchem.5b01312
J. Med. Chem. XXXX, XXX, XXX−XXX