Truncated Actin-Targeting Macrolide Derivative Blocks Cancer Cell Motility and Invasion of Extracellular Matrix

Article abstract of DOI:10.1021/jacs.0c12404

Cancer metastasis is a complex process involving highly motile tumor cells that breach tissue barriers, enter the bloodstream and lymphatic system, and disseminate throughout the body as circulating tumor cells. The primary cellular mechanism contributing to these critical events is the reorganization of the actin cytoskeleton. Mycalolide B (MycB) is an actin-targeting marine macrolide that can suppress proliferation, migration, and invasion of breast and ovarian cancer cells at low nanomolar doses. Through structure-activity relationship studies focused on the actin-binding tail region (C24-C35) of MycB, we identified a potent truncated derivative that inhibits polymerization of G-actin and severs F-actin by binding to actin's barbed end cleft. Biological analyses of this miniature MycB derivative demonstrate that it causes a rapid collapse of the actin cytoskeleton in ovarian cancer cells and impairs cancer cell motility and invasion of the extracellular matrix (ECM) by inhibiting invadopodia-mediated ECM degradation. These studies provide essential proof-of-principle for developing actin-targeting therapeutic agents to block cancer metastasis and establish a synthetically tractable barbed end-binding pharmacophore that can be further improved by adding targeting groups for precision drug design.

Full text of DOI:10.1021/jacs.0c12404

pubs.acs.org/JACS
Article
Truncated Actin-Targeting Macrolide Derivative Blocks Cancer Cell
Motility and Invasion of Extracellular Matrix
Bhavin V. Pipaliya,# Daria N. Trofimova,# Rebecca L. Grange,# Madhu Aeluri,# Xu Deng, Kavan Shah,
Andrew W. Craig, John S. Allingham,* and P. Andrew Evans*
Cite This: J. Am. Chem. Soc. 2021, 143, 6847−6854
Read Online
sı
*
Metrics & More
Article Recommendations
Supporting Information
ACCESS
ABSTRACT: Cancer metastasis is a complex process involving
highly motile tumor cells that breach tissue barriers, enter the
bloodstream and lymphatic system, and disseminate throughout
the body as circulating tumor cells. The primary cellular
mechanism contributing to these critical events is the reorganiza-
tion of the actin cytoskeleton. Mycalolide B (MycB) is an actin-
targeting marine macrolide that can suppress proliferation,
migration, and invasion of breast and ovarian cancer cells at low
nanomolar doses. Through structure−activity relationship studies
focused on the actin-binding tail region (C24−C35) of MycB, we
identified a potent truncated derivative that inhibits polymerization
of G-actin and severs F-actin by binding to actin’s barbed end cleft.
Biological analyses of this miniature MycB derivative demonstrate that it causes a rapid collapse of the actin cytoskeleton in ovarian
cancer cells and impairs cancer cell motility and invasion of the extracellular matrix (ECM) by inhibiting invadopodia-mediated
ECM degradation. These studies provide essential proof-of-principle for developing actin-targeting therapeutic agents to block
cancer metastasis and establish a synthetically tractable barbed end-binding pharmacophore that can be further improved by adding
targeting groups for precision drug design.
which was isolated from the marine sponge Mycale sp.7 MycB
(1) is a macrolide toxin that suppresses motility and invasion of
HER2+ breast (HCC1954) and ovarian (SKOV3) cancer cell
INTRODUCTION
■
Cancer metastasis, in which malignant cells spread from the
primary site to colonize distant organs, accounts for over 90
percent of cancer-related deaths. In many patients, metastasis has
already occurred by the time they are diagnosed with primary
lines at low nanomolar doses and halts cell growth completely at
higher doses (200 nM).8 Like several other natural actin-
targeting macrolides,9,10 MycB (1) consists of a structurally
cancer. Consequently, this mechanism represents one of the
complex macrolactone (the “ring”) and a long aliphatic side
chain (the “tail”) that terminates with an (E)-N-methyl-N-
vinylformamide moiety (MVF) (Figure 1A). X-ray crystallo-
graphic analysis of actin−macrolide complexes, in combination
with structure−activity relationship studies, indicate that the
ring appears to be important for making the initial interaction
with actin on a large and shallow hydrophobic surface that is
exposed in both its globular (G-actin) and filamentous (F-actin)
state (Figure 2).3,11 The tail component intercalates into the
narrow hydrophobic cleft splitting subdomains 1 and 3 at the
barbed end of actin, which forms a critical interface during actin
most significant challenges for successful cancer treatment.
Cancers that overexpress human epidermal growth factor
receptor 2 (HER2) are invasive, with high rates of metastasis
and poor prognosis.1 HER2-driven metastasis is particularly
prevalent in subsets of breast and ovarian cancers.2 Although
many signaling pathways are implicated in metastasis, the engine
driving metastasis is fuelled by rapid and dynamic actin
polymerization.
In an analogous manner to microtubules, which are targeted
by some of the best classes of natural product-derived cancer
chemotherapeutics, the actin cytoskeleton is targeted by many
natural products that elicit extremely potent cytotoxic effects in
cancer cells.3,4 Unfortunately, clinical use of these agents is
impeded by their low natural abundance, structural complexity,
and their inability to discriminate between tumor cells and
normal tissues.5,6 In a program directed toward identifying novel
small molecules that target actin, we have focused on the design,
synthesis, and biological evaluation of truncated analogues of the
actin-depolymerizing natural product, mycalolide B (1, MycB),
Received: November 27, 2020
Published: May 3, 2021
https://doi.org/10.1021/jacs.0c12404
J. Am. Chem. Soc. 2021, 143, 6847−6854
© 2021 American Chemical Society
6847

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
against the “actin addiction” of cancer metastasis. To accomplish
this, we directed our efforts toward addressing the following
questions: What design features are required to offset the
binding free energy loss derived from the removal of the
macrocycle and retain the ability to block actin polymerization
and disrupt existing actin filaments? Does the actin-disrupting
activity of the analogue suppress migration and invasion of
metastatic cancer cells ex vivo? Is the analogue conducive to
large-scale synthesis and functionalization with linkers for
bioconjugation to ligand-targeted drug delivery systems?19,20
With these questions in mind, we prepared a panel of >30
MycB (1) “tail” analogues that explored variations of the C24
substituents, the C24 and C30 esters, the N-vinylformamide
substituent, and the stereochemistry at C30 (see Table S2 in the
Supporting Information for pertinent details). All analogues
were evaluated biochemically for inhibition of G-actin polymer-
ization and F-actin depolymerization. The synthetic strategy for
the construction of the most potent truncated analogue (2) is
highlighted in Figure 1B. This particular analogue was tested in
HER2+ cancer cells for its effects on the actin cytoskeleton
structure and functions related to cancer cell metastasis. By
determining its cocrystal structure with G-actin, we are able to
provide a structural explanation for its activity and a framework
for designing even more potent analogues.
Figure 1. (A) Structure of mycalolide B (1). (B) Synthetic strategy for
the preparation of the truncated analogue (2).
The synthesis of 2 was initiated with the preparation of the
C29−C34 fragment 7 in five steps from the lactate derived
enantiomerically enriched ketone 321 (Scheme 1A). Paterson
asymmetric anti-aldol reaction between the (E)-boron enolate
derived from benzoyloxy ketone 3 with (R)-3-(benzyloxy)-2-
methylpropanal (4)22 in the presence of chlorodicyclohexylbor-
ane and dimethylethylamine afforded the anti-aldol adduct 5
with excellent yield and good diastereoselectivity (dr = 9:1
23,24
1
determined by 500 MHz H NMR).
Methylation25 of the
secondary alcohol 5 with (CH3)3OBF4 (i.e., Meerwein’s salt) in
the presence of proton sponge (MilliporeSigma) followed by a
one-pot benzoyl deprotection and ketone reduction using
lithium borohydride afforded the diol 6 in 76% yield over two
steps. Oxidative cleavage of the diol 6 with sodium periodate
gave the corresponding aldehyde, which was treated with the
vinyl organocerium reagent to afford the allylic alcohols 7 and 8
in 55% and 27% yield, respectively, over two steps. The
undesired diastereomer 8 could be separated and converted to
the desired diastereomer 7 via the following two-step sequence:
Oxidation of the secondary alcohol 8 using Dess-Martin
periodinane (DMP) followed by a stereoselective reduction
with sodium borohydride in the presence of cerium(III)
chloride furnished the desired secondary alcohol 7 in 80%
yield and with significantly improved diastereoselectivity (dr =
9:1 determined by 500 MHz 1H NMR).
Figure 2. X-ray crystal structure of the actin−MycB complex. MycB (1)
(shown as sticks) binds to the barbed end cleft of actin, between
subdomains 1 and 3. Selected carbon atoms of the MycB (1) core are
labeled. All images of structural models were generated with PyMol.
subunit polymerization, and is the binding site for a myriad of
actin-binding proteins.12,13
While a recent total synthesis of MycB (1) was achieved in
more than 50 total steps,14 the structural complexity and limited
availability from natural sources (89 mg of MycB (1) was
isolated from 1.8 kg of sponge)7 impede its development as an
anticancer agent. We and others have recognized that an
appropriately designed and optimized synthetic analogue of the
macrolide tail should inhibit G-actin polymerization, destabilize
F-actin, and display similar levels of cytotoxicity to that exhibited
by the natural product.4,15 Unfortunately, the simplified tail
analogues that have been developed to date are substantially less
potent, and their ability to inhibit cancer cell migration and
invasion has not been evaluated.4,16−18
The synthesis of the C24−C28 fragment 12 commenced from
the commercially available enantiomerically enriched homo-
allylic alcohol 9 (Scheme 1B). Benzylation of secondary alcohol
9 using sodium hydride and benzyl bromide gave the benzyl
ether 10 in 87% yield. Ozonolysis of olefin 10 provided the
aldehyde, which was subjected to the Leighton asymmetric
crotylation using the chiral cis-crotylsilane generated in situ from
(Z)-but-2-en-1-yltrichlorosilane and (1S,2S)-2-tert-butyl-6-
({[2-(methylamino)cyclohexyl]amino}methyl)phenol (L*) in
the presence of 1,8-diazabicyclo[5.4.0]undec-7-ene (DBU), to
furnish the homoallylic alcohol 11 in excellent yield and
RESULTS AND DISCUSSION
■
Synthesis of Mycalolide B Analogues. Guided by the
crystal structure of the actin−MycB complex (Figure 2, PDB ID:
6MGO), we sought to identify the MycB (1) pharmacophore
that would become the focus for future precision drug design
1
diastereoselectivity (dr ≥19:1 determined by 500 MHz H
NMR).26 The secondary alcohol 11 was methylated and the
benzyl ether oxidatively cleaved to enable the protection of the
6848
https://doi.org/10.1021/jacs.0c12404
J. Am. Chem. Soc. 2021, 143, 6847−6854

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
Scheme 1. Synthesis of C29−C34 and C24−C28 Fragments
Scheme 2. Synthesis of Truncated MycB Analogue 2
resulting secondary alcohol as the TBS ether 12 for selective
deprotection later in the synthesis.
15. The formamide moiety was then installed using a copper(I)-
catalyzed Goldberg coupling of vinyl iodide 15 with N-
ethylformamide in the presence of 1,10-phenanthroline and
cesium carbonate in dimethylacetamide at elevated temper-
ature.35 Yamaguchi esterification of the C24 secondary alcohol
with the mixed anhydride obtained from 4-azidobutyric acid,
followed by azido reduction using Lindlar’s catalyst36 under an
atmosphere of hydrogen furnished the primary amine 2 in 64%
yield over three steps. Hence, the synthesis of the truncated
analogue 2 was accomplished from the enantiomerically
enriched ketone 3 in 11.7% overall yield via this convergent
and modular sequence.
Biochemical and Biological Activity of MycB Ana-
logues. A quantitative analysis of the inhibition of G-actin
polymerization by analogue 2 was carried out using the pyrene−
actin fluorescence assay, with MycB (1) as a positive control.
The polymerization of 9 μM skeletal muscle G-actin was
monitored by the increase in fluorescence over the duration of
600 s after the addition of actin polymerization buffer at 50 s.
Representative traces of actin polymerization in the presence of
increasing concentrations of MycB (1) or analogue 2 are shown
in Figure 3A,B. These data show that analogue 2 caused a
The construction of the two fragments set the stage to
examine the cross-coupling and the completion of the MycB
truncated analogue 2 (Scheme 2). The fragments 7 and 12 were
subjected to a Type II alkene cross-metathesis using 10 mol %
Hoveyda-Grubbs 2nd generation catalyst to furnish the cross-
coupled product,27−29 which was reduced with diimide,
generated in situ from 2,4,6-triisopropylbenzenesulfonyl hydra-
zide30,31 and triethylamine, to afford the secondary alcohol 13 in
65% yield over two steps. Yamaguchi esterification32 of the
secondary alcohol 13 using the mixed anhydride derived from
(R)-tetrahydrofuran-2-carboxylic acid and 2,4,6-trichloroben-
zoyl chloride in the presence of 4-dimethylaminopyridine
(DMAP), followed by hydrogenolysis of the terminal benzyl
ether, furnished the primary alcohol 14 in excellent yield (87%
over two steps). The key vinyl iodide 15 was then prepared via a
three-step sequence in 70% overall yield. Dess-Martin period-
inane mediated oxidation of the primary alcohol 14 to the
aldehyde, followed by Takai olefination33,34 with anhydrous
CrCl2 and iodoform, and subsequent cleavage of the TBS ether
with 3 N hydrochloric acid, afforded the required vinyl iodide
6849
https://doi.org/10.1021/jacs.0c12404
J. Am. Chem. Soc. 2021, 143, 6847−6854

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
concentration-dependent decrease in both the rate and the final
yield of the polymerized actin. The inhibitory concentration of
analogue 2 required to decrease the rate of G-actin polymer-
ization by half (IC50) was 200 15 nM, while the IC50 for MycB
(1) was 20 15 nM (Figure S1). Thus, although the molecular
weight and number of stereocenters in analogue 2 are
approximately half that of MycB (1), it retains substantial
inhibitory effects on actin polymerization.
To assess the ability of analogue 2 to disrupt preassembled
actin polymers, we monitored the decrease in the fluorescence of
0.2 μM pyrene-F-actin immediately after the addition of
increasing concentrations of MycB (1) or analogue 2. A
comparison of the concentration−response curves in Figure
3C,D indicates that MycB (1) is a more potent F-actin
depolymerization agent than analogue 2. This outcome reaffirms
previous findings that the macrocyclic ring is a major contributor
to the F-actin-severing capability of barbed-end binding natural
products like MycB (1).37 However, all concentrations of
analogue 2 clearly stimulate rapid F-actin fluorescence decay
within seconds of addition, indicating that some of the F-actin
depolymerization activity of analogue 2 involves filament
severing. Moreover, the relative amount of analogue 2 required
to decrease the pyrene fluorescence to the same baseline level as
MycB (1) was similar to the associated G-actin polymerization
inhibition. These data show that, although less potent, a
synthetic barbed-end binding analogue can elicit the same
bioactivities as the natural product.
On the basis of our recent finding that MycB (1) treatment led
to rapid disruption of F-actin cytoskeleton in SKOV3 cells,8 we
also examined the ability of analogue 2 to disrupt F-actin in
SKOV3 cells. Upon treatment of SKOV3 cells with analogue 2 at
1 μM or above for 1 h, cortical F-actin and membrane
projections began to collapse and large F-actin aggregates began
to form, which is consistent with the outcome with 0.1 μM
MycB (1) (Figure 4). Smaller effects on F-actin were observed
upon treatment with analogue 2 at a 0.1 μM dose, although some
aggregates were detectable near the cell cortex. This indicates
that analogue 2 enters the cell quickly and can disrupt intact F-
Figure 3. Effect of MycB (1) and analogue 2 on polymerization and
depolymerization of purified actin. (A and B) The change in
fluorescence signal during polymerization of 9 μM pyrene−G-actin
was measured in the presence of 0−250 nM MycB (1) or 0−1000 nM
analogue 2 and plotted as a function of time. The polymerization rate of
actin in the absence of toxin was normalized to 100%. (C and D)
Depolymerization of pyrene-F-actin (0.2 μM) was monitored in the
presence of 0−5 μM MycB (1) or 0−40 μM analogue 2 (final
concentration).
Figure 4. Rapid F-actin disruption in SKOV3 cancer cells following
treatment with analogue 2 or MycB (1). Representative confocal
micrographs are shown for SKOV3 cells in DMEM, DMEM in the
presence of a vehicle (1% DMSO), MycB (1) at 0.1 μM, or analogue 2
at specified concentrations for 1 h. F-actin was stained with Phalloidin
(red) and nuclei with DAPI (blue). Images from red and blue channels
are merged (scale bars indicate 15 μm).
6850
https://doi.org/10.1021/jacs.0c12404
J. Am. Chem. Soc. 2021, 143, 6847−6854

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
actin polymers in cancer cells, leading to loss of their typical
motile phenotype, such as a prominent leading-edge projection.
To test the longer-term effects of analogue 2, we performed a
dose response assay of SKOV3 cell invasion using a modified
wound closure assay with an overlay of extracellular matrix
(ECM) proteins (Matrigel). Treatment with analogue 2 at 5 or
10 μM caused significant inhibition of SKOV3 cell invasion over
48 h (Figure 5A). At the higher dose, treatment with analogue 2
resulted in comparable suppression of cell invasion to that of
MycB (1) at 1 μM (Figure 5B).
Since cancer cells require F-actin-based protrusions with
extracellular matrix degradation activity (invadopodia) to close
the wound area in these invasion assays, we performed assays to
measure the effect of analogue 2 and MycB (1) on invadopodia
activity in SKOV3 cells. Plating of SKOV3 cells on coverslips
coated with DQ-collagen, a quenched form of fluorescent
collagen, the invadopodia-derived matrix metalloproteinases
(MMP) can digest DQ-collagen, leading to increased
fluorescence (ECM degradation). Confocal images from these
assays revealed ECM degradation in vehicle-control-treated
cells, but not in those treated with an MMP inhibitor (Figure
S2). Treatments with MycB (1) or analogue 2 led to F-actin
disruption and loss of ECM degradation (Figure S2).
Quantification of these results revealed that the treatment
with analogue 2 at doses as low as 0.1 μM significantly reduced
ECM degradation by SKOV3 cells (Figure 5C). Together, these
data provide the first demonstration of potent antimetastatic activity
by a truncated actin-targeting natural product analogue.
Crystal Structure of Analogue 2 Bound to Actin. To
provide a structural explanation for these activities of analogue 2,
and to provide a framework for designing more potent
analogues, we crystallized analogue 2 bound to G-actin and
determined its X-ray crystal structure to a resolution of 1.7 Å
(Figure 6A). The data clearly demonstrate that the analogue 2
forms an extended conformation within the cleft separating actin
subdomains 1 and 3, which mimics most barbed-end binding
actin toxins. This site is where the DNase I loop of the lower
longitudinal actin subunit would bind during F-actin polymer-
ization.
The actin−2 binding interface is formed by 12 residues that
make hydrophobic contacts along the central aliphatic region of
analogue 2, and by two water-bridged hydrogen bonds between
the terminal N-ethylformamide group and amino acid residues
Ile136 and Ala170 deep within the cleft (Figure 6B and Figure
S3). The (R)-tetrahydrofuran ester side chain at C30 also
interacts with a network of water molecules that bridge to the
side chain hydroxyl group of Tyr169. At the other end of
analogue 2, the terminal amino group of the C24 ester side chain
sits above a large hydrophobic patch on actin, where the ring of
most barbed end-targeting macrolide toxins binds. However, the
polar amino terminus of analogue 2 projects away from actin to
hydrogen bond with nearby surface waters. By extending this
positively charged terminal amino group, it may be possible to
establish a new interaction with a negatively charged region
(Glu334) at the entrance of the barbed end cleft, mimicking
either the salt bridge or H-bond formed by the F-actin-severing
proteins twinfilin and cofilin, respectively.38,39 Functionalization
of the terminal amino group with a linker for bioconjugation to
cancer-cell-specific monoclonal antibodies could also be ex-
plored to focus the antimetastatic activity of this compound on
tumor cells.
Figure 5. Inhibition of SKOV3 cell invasion by MycB (1) and analogue
2. (A) The change in wound confluence of SKOV3 cells was measured
in the presence of analogue 2 (2.5−10 μM), MycB (1) (1 μM), or 2%
DMSO (control) and plotted as a function of time, or (B) was
quantified after 48 h using an IncuCyte Zoom system. (C) Graph
depicts quantification of ECM degradation activity of SKOV3 cells
plated on DQ-collagen. Cells were grown for 8 h prior to treatment with
a vehicle, MMP inhibitor (MMPi; 10 μM), MycB (1) (25 nM), or
analogue 2 (0.1 or 1 μM). After 4 h of treatment, cells were fixed,
stained with AF568-Phalloidin, and imaged by confocal microscopy.
The amount of green fluorescence per field was analyzed to determine
the extent of DQ-collagen degradation (relative ECM degradation; n =
3 experiments; mean SEM, ***P < 0.001).
formamide and with Thr148 via the (R)-tetrahydrofuran ester
moiety in place of the 2,3-di-O-methylglycerate found in MycB
(1). These novel features could contribute to the potent
biochemical and cell biological effects of this compound by
enhancing the disruption of protein−protein contacts rather
than by improving the affinity for actin. Specifically, the
protrusive (R)-tetrahydrofuran ester moiety could provide
Analogue 2 also forms unique hydrophobic interactions with
Leu346 through the N-ethyl substituent on the N-vinyl-
6851
https://doi.org/10.1021/jacs.0c12404
J. Am. Chem. Soc. 2021, 143, 6847−6854

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
F-actin stability. Gratifyingly, this analogue is active in SKOV3
cells and rapidly disrupts the actin cytoskeleton, causing cells to
round up at low concentrations of analogue 2. The crystal
structure of analogue 2 bound to actin suggests that some of the
features of analogue 2 that differ from MycB (1) are located at
positions that do not contact actin directly, but instead face
toward the solvent. Design, synthesis, and functional analysis of
new analogues that explore modifications at these positions are
currently underway.
EXPERIMENTAL SECTION
■
Proteins and Reagents. Actin and pyrene actin were purchased
from Cytoskeleton Inc. (Denver, CO). Mycalolide B was purchased
from Wako Pure Chemical Industries Ltd., Japan, and stored at −20 °C.
Pyrene−G-actin Polymerization Assay. A mixture of rabbit
skeletal muscle G-actin and pyrene−G-actin (2:1 ratio)40 was diluted in
G-buffer (2 mM Tris-HCl, 0.2 mM CaCl2, and 0.8 mM NaN3) to a final
concentration of 9 μM, incubated on ice for 30 min, and further
centrifuged for 1 h at 130000g at +4 °C. Then 63 μL of this mixture was
mixed with 1.4 μL of DMSO, MycB (1), or analogue 2 (0−1000 nM) in
a fluorimeter cuvette and incubated at 25 °C for 50 s. Actin
polymerization reactions were initiated by the addition of 7 μL of
actin polymerization buffer (100 mM Tris-HCl (pH 7.5), 500 mM KCl,
20 mM MgCl2, and 10 mM ATP) to a pyrene actin−toxin mixture.
Fluorescence emission at 406 nm was monitored by using a Fluorolog
fluorimeter (Jobin Yvon Horiba) with an excitation wavelength of 365
nm for a period of 600 s. The rate of polymerization was calculated as an
increase in the fluorescent intensity between 300 and 350 s and divided
at 50 s. The polymerization rate of actin in the absence of the toxin was
normalized to 100%. The rate of polymerization in the presence of the
toxin was expressed as a percentage relative to the no-toxin value and
plotted as a function of concentration. The IC50 value was deduced as
the concentration of the toxin at which a 50% rate of polymerization
was observed.41 All experiments were performed in triplicate and
presented as mean standard error (SEM).
Pyrene−F-actin Depolymerization Assay. A mixture of rabbit
skeletal muscle G-actin and pyrene−G-actin (2:1) was diluted to 5 μM
in A-buffer (100 mM Tris-HCl (pH 7.5), 500 mM KCl, 20 mM MgCl2,
and 10 mM ATP) and incubated for 1 h at room temperature to form F-
actin. Toxin solutions (96 μL) containing a range of concentrations of
MycB (1) or analogue 2 in buffer F (10 mM Tris-HCl (pH 7.5), 50 mM
KCl, 2 mM MgCl2, 1 mM ATP, 0.2 mM DTT, and 0.2 mM CaCl2) were
added to 4 μL of the F-actin solution. Fluorescence emission at 406 nm
was monitored with an excitation wavelength of 365 nm for a period of
900 s. All experiments were performed in triplicate.
Cell Culture. The SKOV3 cell line was obtained from American
Type Culture Collection and authenticated by STR profiling. The cell
line was maintained in Dulbecco’s Modified Eagle’s Medium (DMEM,
Multicell) supplemented with 10% fetal bovine serum (FBS, Multicell)
and 1% antibiotic-antimycotic solution (Multicell) and cultured in a
humidified incubator at 37 °C with 5% CO2.
Figure 6. X-ray crystal structure of actin bound to analogue 2. (A)
Close-up view of analogue 2 bound within the cleft between
subdomains 1 and 3 (shown in sticks and surface representation,
respectively). (B) Close-up view of the binding sites of MycB (left) and
analogue 2 (right). Residues and waters that form the binding pockets
for each compound are shown in stick and ball representation,
respectively. H-bonds are shown as dashed lines. (C) The steric clash
between the DNase I loop of actin subunits that results from analogue 2
was modeled by superimposing the actin−2 structure onto a cryo-EM
structure of F-actin (PDB ID: 3J8I, pink cartoon and sticks
representation).
Spinning Disc Confocal Microscopy. SKOV3 cells were seeded
on glass coverslips, coated with human FN, and allowed to attach and
spread for 24 h prior to treatments with media, media + 1% DMSO
(vehicle), 100 nM MycB (1), or 100−5000 nM analogue 2 for 1 h. Cells
were then fixed with 4% PFA, permeabilized in 0.2% Triton X-100 and
stained with TRITC-Phalloidin (1:400) to visualize the actin
cytoskeleton and DAPI (1:400) to identify the nuclei. Images were
acquired in red and blue channels using a Quorum WaveFX-X1
spinning disc confocal system (Quorum Technologies Inc., Guelph,
Canada) in the Queen’s University Biomedical Imaging Centre.
Confocal micrographs were merged, and scale bars were added using
Metamorph software.
Cell Invasion Assay. Directional cell invasion measurements were
performed on SKOV3 cells. Cells (2.5 × 104) were seeded in triplicate
in a 96-well ImageLock plate (Essen BioScience). At approximately
90% confluence a scratch was induced using the IncuCyte Wound-
maker (a 96-well wound-making tool, Essen BioScience). After
additional steric hindrance that prevents the addition of actin
protomers at the barbed end during filament assembly (Figure
6C). This is an important consideration that other studies have
not taken into account during analogue design. Hence, we
speculate that improving Kd alone may be insufficient to improve
the activity.
CONCLUSIONS
■
In summary, we have developed a concise, divergent, and
scalable route to a bioactive and synthetically tractable analogue
of MycB (1). Notably, this route permitted the assessment of the
effect of structural variations at C24, C30, and the MVF, which
identified a combination of tail modifications that lead to the
potent inhibition of actin polymer formation and disruption of
6852
https://doi.org/10.1021/jacs.0c12404
J. Am. Chem. Soc. 2021, 143, 6847−6854

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
wounding and the removal of nonadherent cells, wells were overlaid
with 5% Matrigel and incubated at 37 °C for 30 min to allow the
Matrigel to solidify. DMEM containing the indicated doses of MycB
(1) or analogue 2 was added to each well. The plate was then inserted
into the IncuCyte Zoom Live Cell Analysis System at 37 °C and imaged
every 2 h for 48 h. The percentage of wound closure was determined
through IncuCyte Zoom Scratch Wound Analysis software package.
ECM Degradation Assay. To test the effects of MycB (1) and
analogue 2 on ECM degradation by invadopodia, SKOV3 cells (2 ×
104) were seeded on #1.5 acid-washed glass coverslips that were coated
with a mixture of Matrigel and DQ-Collagen (8:1). Eight hours later,
cells were treated with DMSO control, MycB (1) (25 nM), or analogue
2 (0.1 or 1 μM) for 4 h. MMP inhibitor-Ilomastat (GM6001,
MedChem Express, 10 μM) was added as a positive control for
inhibition of MMP-mediated cleavage of DQ-collagen by invadopo-
dia.42 The formation of green dots represents the number of
invadopodia. The amount of green fluorescence per field in confocal
micrographs was analyzed and quantified using Image J software.
Crystallization of the Actin−MycB and Actin−2 Complex.
MycB (1) was mixed with 10 mg/mL G-actin from rabbit muscle in G-
buffer (5 mM Tris (pH 8), 0.2 mM CaCl2, 0.2 mM ATP, and 0.5 mM
DTT) at a 1:1 molar ratio. Analogue 2 and Latrunculin B (LatB) were
mixed with G-actin in G-buffer at a 2:1:1 molar ratio. Mixtures were
incubated on ice for 30 min. Crystals of the actin−MycB complex that
were suitable for X-ray diffraction data collection grew in 3 days from 2
μL of sitting drops containing the actin−MycB complex in a 1:1 ratio
with a precipitant solution containing 0.1 M sodium cacodylate/HCl
(pH 6.5), 20% polyethylene glycol (PEG) 8000, and 0.2 M magnesium
acetate at 293 K. Prior to diffraction data collection, crystals were
transferred into a cryoprotectant composed of 0.1 M sodium
cacodylate/HCl (pH 6.5), 20% PEG 8000, 0.2 M magnesium acetate,
and 22% ethylene glycol (EG) and were then flash-cooled in liquid N2.
Crystals of the actin−2−LatB complex grew in 25 days from 2 μL of
hanging drop containing a 1:2:1 ratio of a complex with a precipitant
solution of 0.1 M Tris-HCl (pH 8.5) and 25% PEG 3350.
Cryoprotectant was composed of 0.1 M Tris-HCl (pH 8.5), 27%
PEG 3350, and 20% EG.
P. Andrew Evans − Department of Chemistry, Queen’s
University, Kingston ON K7L 3N6, Canada; Xiangya School
of Pharmaceutical Sciences, Central South University,
Changsha, Hunan 410013, P.R. China; orcid.org/0000-
0001-6609-5282; Email: andrew.evans@chem.queensu.ca
Authors
Bhavin V. Pipaliya − Department of Chemistry, Queen’s
University, Kingston ON K7L 3N6, Canada; orcid.org/
0000-0003-2680-2178
Daria N. Trofimova − Department of Biomedical & Molecular
Sciences, Queen’s University, Kingston ON K7L 3N6,
Canada; orcid.org/0000-0002-5451-0765
Rebecca L. Grange − Department of Chemistry, Queen’s
University, Kingston ON K7L 3N6, Canada
Madhu Aeluri − Department of Chemistry, Queen’s University,
Kingston ON K7L 3N6, Canada
Xu Deng − Cancer Biology & Genetics Division, Queen’s Cancer
Research Institute, Kingston ON K7L 3N6, Canada; Xiangya
School of Pharmaceutical Sciences, Central South University,
Changsha, Hunan 410013, P.R. China; orcid.org/0000-
0001-7683-1626
Kavan Shah − Cancer Biology & Genetics Division, Queen’s
Cancer Research Institute, Kingston ON K7L 3N6, Canada
Andrew W. Craig − Cancer Biology & Genetics Division,
Queen’s Cancer Research Institute, Kingston ON K7L 3N6,
Canada
Complete contact information is available at:
https://pubs.acs.org/10.1021/jacs.0c12404
Author Contributions
#B.V.P., D.N.T., R.L.G., and M.A. contributed equally to this
work.
X-ray Diffraction Data Collection and Structure Determi-
nation. Diffraction data were collected from a single crystal at beamline
08ID-1 of the Canadian Light Source (Saskatoon, Canada) at 100 K
and was indexed, integrated, and scaled with HKL2000.43 The actin−
MycB complex structure was solved by molecular replacement from
chain A of the actin−ulapualide A structure (PDB accession code 1S22)
with PHASER.44 The actin−2−LatB complex structure was solved by
molecular replacement with the actin−MycB structure (PDB accession
code 6MGO) with PHASER. The structures were refined with Refmac5
and manually optimized using Coot.45,46 During model building and
refinement of the actin−MycB structure, it was discovered that the C5−
C6 double bond was reduced and that C5 was linked to a sulfur. Data
processing and refinement statistics are summarized in Table S1.
Coordinates and structure factors of actin−MycB and actin−2−LatB
have been deposited in the Protein Data Bank with Accession Codes
6MGO and 6W7V, respectively.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We acknowledge the generous financial support from the
Collaborative Health Research Projects (CIHR/NSERC) granting
program (Award Nos. 134763 and 151976 to J.S.A., P.A.E., and
A.W.C.). NSERC is acknowledged for supporting a Tier 1
Canada Research Chair (P.A.E.). We are also indebted to Tyler
Vance, Sean Phippen, and Shuaiqi Guo for technical assistance
with X-ray data collection and processing at the Canadian Light
Source, Saskatoon, Canada (beamline 08ID-1), and we thank
the beamline group for making these experiments possible. We
also appreciate the technical assistance of Kim Munro, David
McLeod, Sarah Nersesian, Stephanie Young, and Matthew
Gordon for contributions to the biological evaluation. We thank
Sarah Nersesian for her assistance with the artwork in the
Abstract Graphic.
ASSOCIATED CONTENT
* Supporting Information
■
sı
The Supporting Information is available free of charge at
https://pubs.acs.org/doi/10.1021/jacs.0c12404.
Experimental procedures, spectral data, supplementary
REFERENCES
■
1
figures, and copies of H and 13C NMR spectra for all
(1) Rubin, I.; Yarden, Y. The basic biology of HER2. Ann. Oncol. 2001,
12 (Suppl 1), S3−S8.
compounds (PDF)
(2) Ross, J. S.; Slodkowska, E. A.; Symmans, W. F.; Pusztai, L.; Ravdin,
P. M.; Hortobagyi, G. N. The HER-2 receptor and breast cancer: ten
years of targeted anti-HER-2 therapy and personalized medicine.
Oncologist 2009, 14, 320−368.
(3) Allingham, J. S.; Klenchin, V. A.; Rayment, I. Actin-targeting
natural products: structures, properties and mechanisms of action. Cell.
Mol. Life Sci. 2006, 63, 2119−2134.
AUTHOR INFORMATION
Corresponding Authors
John S. Allingham − Department of Biomedical & Molecular
Sciences, Queen’s University, Kingston ON K7L 3N6, Canada;
Email: allinghj@queensu.ca
■
6853
https://doi.org/10.1021/jacs.0c12404
J. Am. Chem. Soc. 2021, 143, 6847−6854

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
(4) Ohno, O.; Morita, M.; Kitamura, K.; Teruya, T.; Yoneda, K.; Kita,
M.; Kigoshi, H.; Suenaga, K. Apoptosis-inducing activity of the actin-
depolymerizing agent aplyronine A and its side-chain derivatives.
Bioorg. Med. Chem. Lett. 2013, 23, 1467−1471.
(24) Lister, T.; Perkins, M. V. A retro-Claisen approach to
dolabriferol. Org. Lett. 2006, 8, 1827−1830.
(25) Evans, D. A.; Ratz, A. M.; Huff, B. E.; Sheppard, G. S. Total
synthesis of the polyether antibiotic lonomycin A (Emericid). J. Am.
Chem. Soc. 1995, 117, 3448−3467.
(5) Newman, D. J.; Cragg, G. M. Marine natural products and related
compounds in clinical and advanced preclinical trials. J. Nat. Prod. 2004,
67, 1216−1238.
(26) Suen, L. M.; Steigerwald, M. L.; Leighton, J. L. A new and more
powerfully activating diamine for practical and scalable enantioselective
aldehyde crotylsilylation reactions. Chem. Sci. 2013, 4, 2413−2417.
(27) Kingsbury, J. S.; Harrity, J. P. A.; Bonitatebus, P. J., Jr.; Hoveyda,
A. H. A recyclable Ru-based metathesis catalyst. J. Am. Chem. Soc. 1999,
121, 791−799.
(6) Paterson, I.; Fink, S. J.; Lee, L. Y. W.; Atkinson, S. J.; Blakey, S. B.
Total synthesis of aplyronine C. Org. Lett. 2013, 15, 3118−3121.
(7) Fusetani, N.; Yasumuro, K.; Matsunaga, S.; Hashimoto, K.
Mycalolides A−C, hybrid macrolides of ulapualides and halichon-
dramide, from a sponge of the genus Mycale. Tetrahedron Lett. 1989, 30,
2809−2812.
(8) Nersesian, S.; Williams, R.; Newsted, D.; Shah, K.; Young, S.;
Evans, P. A.; Allingham, J. S.; Craig, A. W. Effects of modulating actin
dynamics on HER2 cancer cell motility and metastasis. Sci. Rep. 2018, 8,
17243.
(9) Spector, I.; Braet, F.; Shochet, N. R.; Bubb, M. R. New anti-actin
drugs in the study of the organization and function of the actin
cytoskeleton. Microsc. Res. Tech. 1999, 47, 18−37.
(10) Fenteany, G.; Zhu, S. Small-molecule inhibitors of actin
dynamics and cell motility. Curr. Top. Med. Chem. 2003, 3, 593−616.
(11) Kita, M.; Kigoshi, H. Marine natural products that interfere with
multiple cytoskeletal protein interactions. Nat. Prod. Rep. 2015, 32,
534−542.
(12) Klenchin, V. A.; Allingham, J. S.; King, R.; Tanaka, J.; Marriott,
G.; Rayment, I. Trisoxazole macrolide toxins mimic the binding of
actin-capping proteins to actin. Nat. Struct. Mol. Biol. 2003, 10, 1058−
1063.
(13) Yeung, K.-S.; Paterson, I. Actin-binding marine macrolides: total
synthesis and biological importance. Angew. Chem., Int. Ed. 2002, 41,
4632−4653.
(14) Kita, M.; Oka, H.; Usui, A.; Ishitsuka, T.; Mogi, Y.; Watanabe, H.;
Tsunoda, M.; Kigoshi, H. Total synthesis of mycalolides A and B
through olefin metathesis. Angew. Chem., Int. Ed. 2015, 54, 14174−
14178.
(15) Pereira, J. H.; Petchprayoon, C.; Hoepker, A. C.; Moriarty, N. W.;
Fink, S. J.; Cecere, G.; Paterson, I.; Adams, P. D.; Marriott, G. Structural
and biochemical studies of actin in complex with synthetic macrolide
tail analogs. ChemMedChem 2014, 9, 2286−2293.
(28) Garber, S. B.; Kingsbury, J. S.; Gray, B. L.; Hoveyda, A. H.
Efficient and recyclable monomeric and dendritic Ru-based metathesis
catalysts. J. Am. Chem. Soc. 2000, 122, 8168−8179.
(29) Chatterjee, A. K.; Choi, T.-L.; Sanders, D. P.; Grubbs, R. H. A
general model for selectivity in olefin cross metathesis. J. Am. Chem. Soc.
2003, 125, 11360−11370.
(30) Cusack, N. J.; Reese, C. B.; Risius, A. C.; Roozepeikar, B. 2,4,6-
Tri-isopropylbenzenesulphonyl hydrazide: A convenient source of di-
imide. Tetrahedron 1976, 32, 2157−2162.
(31) Pasto, D. J.; Taylor, R. T. Reduction with diimide. Org. React.
1991, 40, 91.
(32) Inanaga, J.; Hirata, K.; Saeki, H.; Katsuki, T.; Yamaguchi, M. A
rapid esterification by means of mixed anhydride and its application to
large-ring lactonization. Bull. Chem. Soc. Jpn. 1979, 52, 1989−1993.
(33) Takai, K.; Nitta, K.; Utimoto, K. Simple and selective method for
RCHO→(E)-RCHCHX conversion by means of a CHX3-CrCl2
system. J. Am. Chem. Soc. 1986, 108, 7408−7410.
(34) Takai, K. Addition of organochromium reagents to carbonyl
compounds. Org. React. 2004, 64, 253.
(35) Shen, R.; Lin, C. T.; Bowman, E. J.; Bowman, B. J.; Porco, J. A., Jr.
Lobatamide C: total synthesis, stereochemical assignment, preparation
of simplified analogues, and V-ATPase inhibition studies. J. Am. Chem.
Soc. 2003, 125, 7889−7891.
(36) Corey, E. J.; Nicolaou, K. C.; Balanson, R. D.; Machida, Y. A
useful method for the conversion of azides to amines. Synthesis 1975,
590−591.
(37) Tanaka, J.; Yan, Y.; Choi, J.; Bai, J.; Klenchin, V. A.; Rayment, I.;
Marriott, G. Biomolecular mimicry in the actin cytoskeleton:
mechanisms underlying the cytotoxicity of kabiramide C and related
macrolides. Proc. Natl. Acad. Sci. U. S. A. 2003, 100, 13851−13856.
(38) Paavilainen, V. O.; Oksanen, E.; Goldman, A.; Lappalainen, P.
Structure of the actin-depolymerizing factor homology domain in
complex with actin. J. Cell Biol. 2008, 182, 51−59.
(39) Tanaka, K.; Takeda, S.; Mitsuoka, K.; Oda, T.; Kimura-Sakiyama,
C.; Maéda, Y.; Narita, A. Structural basis for cofilin binding and actin
filament disassembly. Nat. Commun. 2018, 9, 1860.
(40) Doolittle, L. K.; Rosen, M. K.; Padrick, S. B. Measurement and
analysis of in vitro actin polymerization. Methods Mol. Biol. 2013, 1046,
273−293.
(41) Butler, S. C.; Miles, C. O.; Karim, A.; Twiner, M. J. Inhibitory
effects of pectenotoxins from marine algae on the polymerization of
various actin isoforms. Toxicol. In Vitro 2012, 26, 493−499.
(42) Sharma, V. P.; Eddy, R.; Entenberg, D.; Kai, M.; Gertler, F. B.;
Condeelis, J. Tks5 and SHIP2 regulate invadopodium maturation, but
not initiation, in breast carcinoma cells. Curr. Biol. 2013, 23, 2079−
2089.
(43) Otwinowski, Z.; Minor, W. Processing of X-ray diffraction data
collected in oscillation mode. Methods Enzymol. 1997, 276, 307−326.
(44) McCoy, A. J.; Grosse-Kunstleve, R. W.; Adams, P. D.; Winn, M.
D.; Storoni, L. C.; Read, R. J. Phaser crystallographic software. J. Appl.
Cryst. 2007, 40, 658−674.
(16) Suenaga, K.; Miya, S.; Kuroda, T.; Handa, T.; Kanematsu, K.;
Sakakura, A.; Kigoshi, H. Synthesis and actin-depolymerizing activity of
mycalolide analogs. Tetrahedron Lett. 2004, 45, 5383−5386.
(17) Kitamura, K.; Teruya, T.; Kuroda, T.; Kigoshi, H.; Suenaga, K.
Synthesis of actin-depolymerizing compounds. Bioorg. Med. Chem. Lett.
2009, 19, 1896−1898.
(18) Tetlow, D. J.; Winder, S. J.; Aïssa, C. The synthesis and biological
evaluation of a kabiramide C fragment modified with a WH2 consensus
actin-binding motif as a potential disruptor of the actin cytoskeleton.
Chem. Commun. 2016, 52, 807−810.
(19) Zhao, Z.; Ukidve, A.; Kim, J.; Mitragotri, S. Targeting strategies
for tissue-specific drug delivery. Cell 2020, 181, 151−167.
̌
̌
(20) Anzicek, N.; Williams, S.; Housden, M. P.; Paterson, I. Toward
aplyronine payloads for antibody-drug conjugates: total synthesis of
aplyronines A and D. Org. Biomol. Chem. 2018, 16, 1343−1350.
(21) Paterson, I.; Wallace, D. J.; Cowden, C. J. Polyketide synthesis
using the boron-mediated, anti-aldol reactions of lactate-derived
ketones: total synthesis of (−)-ACRL toxin IIIB. Synthesis 1998,
639−652.
(22) Paquette, L. A.; Guevel, R.; Sakamoto, S.; Kim, I. H.; Crawford, J.
Convergent enantioselective synthesis of vinigrol, an architecturally
novel diterpenoid with potent platelet aggregation inhibitory and
antihypertensive properties. 1. Application of anionic sigmatropy to
construction of the octalin substructure. J. Org. Chem. 2003, 68, 6096−
6107.
(23) Paterson, I.; Blakey, S. B.; Cowden, C. J. Studies in marine
macrolide synthesis: stereocontrolled synthesis of a C21−C34 subunit
of the aplyronines. Tetrahedron Lett. 2002, 43, 6005−6008.
(45) Murshudov, G. N.; Skubák, P.; Lebedev, A. A.; Pannu, N. S.;
Steiner, R. A.; Nicholls, R. A.; Winn, M. D.; Long, F.; Vagin, A. A.
REFMAC5 for the refinement of macromolecular crystal structures.
Acta Cryst. 2011, D67, 355−367.
(46) Emsley, P.; Cowtan, K. Coot: model-building tools for molecular
graphics. Acta Cryst. 2004, D60, 2126−2132.
6854
https://doi.org/10.1021/jacs.0c12404
J. Am. Chem. Soc. 2021, 143, 6847−6854