Discovery of antitubulin agents with antiangiogenic activity as single entities with multitarget chemotherapy potential

Article abstract of DOI:10.1021/ml4004793

Antiangiogenic agents (AA) are cytostatic, and their utility in cancer chemotherapy lies in their combination with cytotoxic chemotherapeutic agents. Clinical combinations of vascular endothelial growth factor receptor-2 (VEGFR2) inhibitors with antitubulin agents have been particularly successful. We have discovered a novel, potentially important analogue, that combines potent VEGFR2 inhibitory activity (comparable to that of sunitinib) with potent antitubulin activity (comparable to that of combretastatin A-4 (CA)) in a single molecule, with GI₅₀ values of 10^-7 M across the entire NCI 60 tumor cell panel. It potently inhibited tubulin assembly and circumvented the most clinically relevant tumor resistance mechanisms (P-glycoprotein and β-III tubulin expression) to antimicrotubule agents. The compound is freely water-soluble as its HCl salt and afforded excellent antitumor activity in vivo, superior to docetaxel, sunitinib, or Temozolomide, without any toxicity.

Full text of DOI:10.1021/ml4004793

Letter
pubs.acs.org/acsmedchemlett
Discovery of Antitubulin Agents with Antiangiogenic Activity as
Single Entities with Multitarget Chemotherapy Potential
Aleem Gangjee,†,* Roheeth Kumar Pavana,† Michael A. Ihnat,‡ Jessica E. Thorpe,‡ Bryan C. Disch,
Anja Bastian,‡ Lora C. Bailey-Downs,‡ Ernest Hamel,§ and Rouli Bai§
†Division of Medicinal Chemistry, Graduate School of Pharmaceutical Sciences, Duquesne University, Pittsburgh, Pennsylvania
15282, United States
‡Department of Pharmaceutical Sciences, University of Oklahoma College of Pharmacy, Oklahoma City, Oklahoma 73117, United
States
§Screening Technologies Branch, Developmental Therapeutics Program, Division of Cancer Treatment and Diagnosis, Frederick
National Laboratory for Cancer Research, National Cancer Institute, Frederick, Maryland 21702, United States
S
* Supporting Information
ABSTRACT: Antiangiogenic agents (AA) are cytostatic, and
their utility in cancer chemotherapy lies in their combination
with cytotoxic chemotherapeutic agents. Clinical combinations
of vascular endothelial growth factor receptor-2 (VEGFR2)
inhibitors with antitubulin agents have been particularly
successful. We have discovered a novel, potentially important
analogue, that combines potent VEGFR2 inhibitory activity
(comparable to that of sunitinib) with potent antitubulin
activity (comparable to that of combretastatin A-4 (CA)) in a
single molecule, with GI50 values of 10−7 M across the entire NCI 60 tumor cell panel. It potently inhibited tubulin assembly and
circumvented the most clinically relevant tumor resistance mechanisms (P-glycoprotein and β-III tubulin expression) to
antimicrotubule agents. The compound is freely water-soluble as its HCl salt and afforded excellent antitumor activity in vivo,
superior to docetaxel, sunitinib, or Temozolomide, without any toxicity.
KEYWORDS: Antitubulin, antimitotic, antiangiogenic, VEGFR2 inhibition, combination chemotherapy
of multiple agents, avoid drug−drug interactions, and be used
ntiangiogenic agents (AA) have defined a new paradigm
1
Afor cancer treatment. The principal mediator of angio- at lower doses to minimize toxicities and tumor cell resistance.
It has been our long-standing interest to design single entities
with multiple targets.5−8 We chose inhibition of tubulin as the
mechanism for the cytotoxic effect because antitubulin agents
are some of the most effective drugs in cancer chemotherapy.9
Additionally, evidence for the remarkable clinical success of
VEGFR2 inhibitors in combination with antitubulin agents is
abundant in the literature.10 Structural similarities of anti-
tubulin11 agents and RTK inhibitors12 previously reported from
our laboratory provided the rationale to incorporate structural
features of both antitubulin activity and RTK inhibitory activity
in single molecules to explore the possibility of providing dual
inhibitory activities in single compounds. We11 previously
reported the design, synthesis, and antitumor activity of
pyrrolo[3,2-d]pyrimidine (1; Figure 1), which inhibits tubulin
assembly. Substituted 7-benzyl pyrrolo[2,3-d]pyrimidines with
general structure 2 (Figure 1) were reported as antiangiogenic,
antimetastatic, and antitumor agents.9 Substituents on the
benzyl group and the location of the benzyl group dictate RTK
genesis is the vascular endothelial growth factor (VEGF) and its
receptor tyrosine kinase (RTK) VEGFR2.2 Rapid regrowth of
vasculature in tumors after the removal of VEGFR2 inhibitors
from the treatment regimen attests to their cytostatic
mechanism.3 VEGFR2 inhibitors are known to cause regression
of impaired tumor blood vessels but transiently normalize the
surviving vessels, which have less leakiness and are more like
normal blood vessels. Such treatment therefore can transiently
increase blood flow in the surviving vasculature.4 Administering
a cytotoxic agent during this transient tumor vasculature
normalization provides improved drug delivery to the tumor,
dependent on the surviving vasculature. However, administer-
ing separate antiangiogenic and cytotoxic agents may miss the
timing window. Single entities with both antiangiogenic and
cytotoxic components should allow the cytotoxicity to be
manifested even at low concentrations in the tumor as soon as
the antiangiogenic effect and vasculature normalization occur.
Hence, these agents might not need the cytotoxic component
to be as potent as conventional cytotoxic drugs and as such
could perhaps avoid or reduce dose-limiting toxicities
associated with conventional chemotherapy. Additionally,
such single agents could circumvent pharmacokinetic problems
Received: November 21, 2013
Accepted: February 27, 2014
Published: February 27, 2014
© 2014 American Chemical Society
480
dx.doi.org/10.1021/ml4004793 | ACS Med. Chem. Lett. 2014, 5, 480−484

ACS Medicinal Chemistry Letters
Letter
Figure 1. Hybrid design from lead compounds.
Figure 2. Synthesis of target compounds.
Table 1. Inhibition of Tubulin Assembly and Cellular VEGFR2 Receptor Tyrosine Kinase and the Growth of β-III and P-gp
Overexpressing Cells
inhibition of tubulin
assembly IC50 (μM)
VEGFR2 inhibition
IC50 (nM)
WT HeLa
IC50 (nM)
β-III overexpressing
HeLa IC50 (nM)
parental OVCAR-8
IC50 (nM)
P-gp overexpressing NCI/
ADR-RES IC50 (nM)
compd
1·HCl
3·HCl
4·HCl
5·HCl
6·HCl
CA
10 0.6
182.3 20.6
21.3 3.2
50.1 8.8
40.6 7.8
30.1 5.1
nd
nd
nd
nd
21
1
280 40
>10,000
>10,000
280 40
>10,000
>10,000
1000 300
700 200
>20 (no act)
>20 (no act)
>40 (no act)
1.2 0.01
>5,000
>5,000
1.8 0.4
2.5 0.7
paclitaxel
semaxinib
sunitinib
5.3
2
16
1
10
0
5,000
0
12.9
18.9 2.7
inhibitory activity in pyrrolo[2,3-d]pyrimidines. Hence, having
determined the antitubulin effects of 1, it was of interest to
engineer RTK inhibitory activity without loss of antitubulin
activity by incorporating the 7-benzyl group onto the
pyrrolo[3,2-d]pyrimidine scaffold. The results of this hybrid
design afforded compound 3 (Figure 1). Compounds 4 and 5
were designed to evaluate the importance of the 4′-OCH3 and
the 4-NCH3 moieties, respectively. Compound 6 was designed
by incorporating a 2′-OCH3 group to explore the effect of
substituents at this position. The antimitotic activity of 3 also
prompted its evaluation against cell lines which overexpress the
multidrug resistance protein P-glycoprotein (Pgp) and the β-
tubulin isoform βIII. Tumors with either of these proteins are
associated with significant resistance to antitubulin agents,
especially with resistance to paclitaxel.13
Inhibition of VEGFR2. We chose to evaluate the VEGFR2
inhibitory activity of our compounds using human tumor cells
known to express high levels of the VEGFR2. Compound 3·
HCl has inhibitory potency comparable with the clinically used
sunitinib and clinically evaluated semaxinib against VEGFR2
(Table 1).
In order to rule out off target toxicity, we evaluated 3·HCl
with the assistance of a kinase profiling service (Luceome
Biotechnologies)15 against 48 other kinases (Supporting
Information) and found no significant activity at 10 μM (data
not shown).
Structure−Activity Relationship. The 7-benzyl moiety
improved cellular VEGFR2 inhibition by about 9-fold and
decreased inhibition of tubulin assembly by about 2-fold
(comparing activities of 3·HCl and lead compound 1·HCl).
The methyl group attached to the nitrogen bridge and the 4′-
methoxy group were crucial for inhibition of tubulin
(comparing activities of 5·HCl and 4·HCl with those of 3·
HCl). Removal of either of these groups resulted in complete
loss of the ability to inhibit tubulin assembly and also decreased
inhibition in VEGFR2 cellular activities.
CAM Assay. Compound 3·HCl was tested for its effects on
blood vessel formation in the chicken chorioallantoic
membrane (CAM) antiangiogenic activity assay and was
found to have an IC50 value of 2.9 0.32 μM comparable to
sunitinib, IC50 = 1.3 0.07 μM.
Chemistry. Acid catalyzed condensation of pyrrole 714
(Figure 2) with acetonitrile afforded bicyclic 8 in 72% yield.
Next, 8 was chlorinated with POCl3 to generate 9 in 87% yield.
Compound 9 was subjected to nucleophilic displacement
reactions using appropriately substituted aromatic amines to
afford 3−6 in 69−82% yields.
Inhibition of Tubulin Assembly. Quantitative studies were
conducted to determine the effects of these compounds on the
polymerization of purified bovine brain tubulin. Compounds
were compared with combretastatin A-4 (CA) as inhibitors of
tubulin assembly (Table 1). Compound 3·HCl was an effective
inhibitor of bovine tubulin assembly, albeit almost 18-fold less
active than CA.
NCI 60 Tumor Panel. (Table 2, Supporting Information).
Compound 3·HCl, showed a 3-digit nanomolar GI50 in all the
NCI 60 tumor cell lines, showing moderate cytotoxic activity
481
dx.doi.org/10.1021/ml4004793 | ACS Med. Chem. Lett. 2014, 5, 480−484

ACS Medicinal Chemistry Letters
Letter
Figure 3. Treatment with 3·HCl decreased primary tumor growth and tumor vascular density in the MDA-MB-435 flank xenograft model. *, P <
0.05; **, P < 0.01; ***, P < 0.001 by one way ANOVA with a Neuman−Keuls post-test. Human BLBCs, MDA-MB-435, were implanted into the
lateral flank of NCr athymic nu/nu nude mice at 500,000 cells, and the mice were treated with carrier, docetaxel, sunitinib, or 3·HCl twice weekly at
their MTDs until the end of the experiment. Data is representative of 6−8 animals. (A) Tumor size was assessed by measuring tumor length, width,
and depth twice weekly using Vernier calipers. Tumor volume was calculated with the ellipsoid formula: volume = 0.52 (length × width × depth)
and graphically represented as days after implantation. Statistics on these graphs were two way ANOVA with repeated measures post-test. (B)
Tumor vascular density was examined at the end of the experiment by staining tumor sections with CD31/PECAM-1 antibodies and a Vectastain
ABC kit. Vascular density was determined by counting CD31-positive vessels and graphed as percent vessel density as compared with carrier treated
animals. (C) Graphical representation of percent change in animal weight as determined by measuring animal weight at the beginning and end of the
experiment.
Figure 4. Treatment with 3·HCl decreased primary tumor growth and tumor vascular density in a U251 flank xenograft mouse model. *, P < 0.05;
**, P < 0.01; ***, P < 0.001 by one way ANOVA with a Neuman−Keuls post-test. U251 human glioma cells were implanted into the lateral flank of
NCr athymic nu/nu nude mice at 500,000 cells, and the mice were treated with carrier Temozolomide or 3·HCl twice weekly at their MTDs until
the end of the experiment. Data is representative of 6−8 animals. (A) Tumor size was determined as described in the legend of Figure 3. Statistics on
this graph were two way ANOVA with repeated measures post-test. (B) Tumor vascular density was examined at the end of the experiment as
described in the legend of Figure 3. (C) Graphical representation of percent change in animal weight as determined by measuring animal weight at
the beginning and end of the experiment.
(as desired) against tumor cells. The antiangiogenic component
of 3·HCl is not active in these cell culture assays.
Effects on Microtubules, Cell Cycle, and Apoptosis.
Tubulin staining in MDA-MB-435 cells and immunofluor-
escence imaging revealed that combretastatin A-4 phosphate
(CA4P), a prodrug of CA converted to the active molecule
probably by a phosphatase in the medium, and 3·HCl led to
less prominent and less aggregated microtubules as compared
with control cells (Supporting Information), indicating that
both compounds have similar effects on cellular microtubules.
Human MDA-MB-435 basal-like breast cancer cells
(BLBCs)16 were treated with an IC50 dose of 3·HCl, docetaxel,
or CA4P and subjected to flow cytometric and cell cycle
analysis. Analysis of DNA content and cell cycle determination
showed an increase in the G2/M phase of cells treated with 3·
HCl, as was the case with the antitubulin agents docetaxel and
CA4P (Supporting Information), indicating that all three
compounds arrest cells in mitosis, as is known for docetaxel
and CA4P. Western blot analysis showed that 3·HCl induced
levels of the apoptosis markers, cleaved PARP, and cleaved
caspase 3 (Supporting Information) after 48 h of treatment.
Effect on Pgp Overexpression and βIII-Tubulin Mediated
Resistance to Paclitaxel. The ability of 3·HCl to overcome
βIII-tubulin drug resistance was evaluated by using an isogenic
HeLa cell line pair. While paclitaxel was 3-fold less potent in the
βIII-tubulin overexpressing cell line than in the wild type HeLa
cells, compound 3·HCl inhibited both cell lines with equal
potency without regard to their expression of βIII-tubulin.
The ability of 3·HCl to circumvent Pgp-mediated drug
resistance was evaluated by using an ovarian cancer cell line pair
(Table 1). In this cell line pair, paclitaxel, which is a well-known
Pgp substrate, was 500-fold less potent in the Pgp over-
expressing cell line than in the wild type OVCAR-8 cells.
Surprisingly, 3·HCl is about 1.4-fold more potent in the Pgp
overexpressing cell line. Hence, 3·HCl is a poor substrate for
transport by Pgp and thus has advantages over some clinically
useful tubulin-targeting drugs such as paclitaxel, docetaxel,
vinblastine, and vinorelbine.13
482
dx.doi.org/10.1021/ml4004793 | ACS Med. Chem. Lett. 2014, 5, 480−484

ACS Medicinal Chemistry Letters
Letter
Figure 5. Treatment with 3·HCl decreased primary tumor growth and lung metastases in the 4T1 orthotopic breast model. *, P < 0.05 by one way
ANOVA with a Neuman−Keuls post-test. 4T1-Lucerase/GFP tagged cells were implanted orthotopically into BALB/c mice at 7500 cells, and the
mice were treated with carrier, docetaxel, or 3·HCl at their MTDs twice weekly until the end of the experiment. (A) Tumor volume was as described
in the legend of Figure 3. (B) Animal weights were recorded before each treatment and graphed as percent weight change. (C) After 32 days, animals
were euthanized and lungs were excised and immediately imaged at 25× using the LumaScope fluorescent imaging system. The number of
metastases in both lungs was counted visually and represented graphically.
Taken together, these data show that 3·HCl interferes with
cellular microtubules, arrests cells in mitosis, and induces
apoptosis.
Identification of Binding Site on Tubulin. Compound 3·
HCl failed to inhibit [3H]vinblastine binding to tubulin and
nucleotide exchange, as measured by [8-14C]GTP binding to
tubulin (Supporting Information). However, it inhibited
[3H]colchicine binding by 55% ( 2) at a concentration of 50
μM, indicating that 3·HCl is probably a colchicine site-binding
antitubulin agent.
human tumor), we chose the mouse 4T1 triple negative
orthotopic allograft model. It was found that 3·HCl was as
active as docetaxel at reducing primary tumor growth in the
4T1 model (Figure 5A). Significantly, 3·HCl reduced lung
metastases in this model better than did docetaxel, a key
concept because most cancer patients die of metastatic disease
(Figure 5C). Finally, treatment with 3·HCl did not result in as
much weight loss as with docetaxel, indicating less systemic
toxicity to the animals (Figure 5B).
In summary, we have reported the design, synthesis, and
biological evaluation for an agent that was designed to have
both cytotoxic and antiangiogenic effects. The antiangiogenic
effect is mediated via VEGFR2 inhibition. The cytotoxic effect
is mediated by tubulin inhibition and is independent of
overexpression of Pgp and βIII-tubulin. The compound caused
cellular microtubule depolymerization, arrested cells in the G2/
M phase, and triggered apoptotic cell death. In vivo, this
compound reduced tumor size and vascularity in two flank
xenograft models (the BLBC MDA-MB-435 and U251 glioma
models) and in a 4T1 triple negative breast orthotopic allograft
model. In these in vivo models, the activity of 3·HCl was
superior to those of Temozolomide (U251), docetaxel, and
sunitinib (MDA-MB-435 and 4T1) without overt toxicity to
the animals.
In Vivo Efficacy and Toxicity of 3·HCl in Two Flank Tumor
Models. First, the maximal tolerated dose (MTD) of 3·HCl,
docetaxel, sunitinib, and Temozolomide was determined in
mice (Supporting Information) and was used to examine the
efficacy of 3·HCl as compared with the antimicrotubule agent
docetaxel and the VEGFR-2 inhibitor sunitinib on tumor
growth in a flank xenograft mouse model in vivo. When MDA-
MB-435 cancer cells were implanted into the flank of athymic
mice, 3·HCl significantly decreased the growth of primary
tumors, both as compared with carrier treated mice and as
compared with mice treated with docetaxel or with sunitinib
(Figure 3A). Further, 3·HCl decreased tumor vascular density
(as measured by CD31/PECAM-1 immunohistochemical
staining) as compared with carrier-treated and docetaxel-treated
mice. However, sunitinib (Figure 3B) was more effective than
3·HCl in reducing tumor vascular density. Finally, all of the
treated animals gained weight during the study, indicating a lack
of gross systemic toxicity (Figure 3C) in this experiment.
We also examined the effect of treatment with 3·HCl on a
human U251 glioma model implanted in immunodeficient mice
in comparison with the standard clinically used agent
Temozolomide (Figure 4). Compound 3·HCl decreased
primary tumor growth and tumor vascular density as compared
with carrier and with Temozolomide without overt toxicity to
the animals.
ASSOCIATED CONTENT
■
S
* Supporting Information
Tumor cell inhibitory activity (NCI) GI50 (nM) of 3·HCl, and
the Experimental Section. This material is available free of
charge via the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: gangjee@duq.edu. Fax: 412-396-5593.
Effect of 3·HCl on Primary Tumor Growth and Lung
Metastasis in the 4T1 Orthotopic Breast Cancer Model in
Vivo. We next sought to determine the in vivo efficacy of 3·HCl
in a relevant triple negative breast cancer model where
metastasis can be assessed. Because the human MDA-MB-231
triple negative model requires 10 million cells, Matrigel, and
immune deficient animals to grow in vivo and because the
MDA-MB-468 line only metastasizes to the liver (unlike the
Funding
This work was supported, in part, by NIH Grants CA136944
(to A.G.), CA114021 (to A.G.), the Duquesne University
Adrian Van Kaam Chair in Scholarly Excellence (to A.G.), and
NSF equipment grant CHE 0614785 (for the NMR).
Notes
The authors declare no competing financial interest.
483
dx.doi.org/10.1021/ml4004793 | ACS Med. Chem. Lett. 2014, 5, 480−484

ACS Medicinal Chemistry Letters
Letter
ACKNOWLEDGMENTS
We thank the NCI for performing the in vitro antitumor
evaluation in their 60 tumor preclinical screening program.
■
ABBREVIATIONS
■
AA, antiangiogenic agents; VEGFR2, vascular endothelial
growth factor receptor-2; RTK, receptor tyrosine kinase; Pgp,
P-glycoprotein; CAM, chicken chorioallantoic membrane;
CA4P, combretastatin A-4 phosphate; BLBCs, basal-like breast
cancer cells
REFERENCES
■
(1) Young, R. J.; Reed, M. W. Anti-angiogenic therapy: Concept to
clinic. Microcirculation 2012, 19, 115−25.
(2) Shibuya, M. Vascular endothelial growth factor and its receptor
system: Physiological functions in angiogenesis and pathological roles
in various diseases. J. Biochem. 2013, 153, 13−19.
(3) Bergers, G.; Hanahan, D. Modes of resistance to anti-angiogenic
therapy. Nat. Rev. Cancer 2008, 8, 592−603.
(4) Jain, R. K. Normalizing tumor microenvironment to treat cancer:
Bench to bedside to biomarkers. J. Clin. Oncol. 2013, 31, 2205−2218.
(5) Gangjee, A.; Zaware, N.; Raghavan, S.; Ihnat, M.; Shenoy, S.;
Kisliuk, R. L. Single agents with designed combination chemotherapy
potential: Synthesis and evaluation of substituted pyrimido[4,5-
b]indoles as receptor tyrosine kinase and thymidylate synthase
inhibitors and as antitumor agents. J. Med. Chem. 2010, 53, 1563−
1578.
(6) Gangjee, A.; Zhao, Y.; Ihnat, M. A.; Thorpe, J. E.; Bailey-Downs,
L. C.; Kisliuk, R. L. Novel tricyclic indeno[2,1-d]pyrimidines with dual
antiangiogenic and cytotoxic activities as potent antitumor agents.
Bioorg. Med. Chem. 2012, 20, 4217−4225.
(7) Gangjee, A.; Li, W.; Lin, L.; Zeng, Y.; Ihnat, M.; Warnke, L. A.;
Green, D. W.; Cody, V.; Pace, J.; Queener, S. F. Design, synthesis, and
X-ray crystal structures of 2,4-diaminofuro[2,3-d]pyrimidines as
multireceptor tyrosine kinase and dihydrofolate reductase inhibitors.
Bioorg. Med. Chem. 2009, 17, 7324−7336.
(8) Gangjee, A.; Zeng, Y.; Ihnat, M.; Warnke, L. A.; Green, D. W.;
Kisliuk, R. L.; Lin, F.-T. Novel 5-substituted, 2,4-diaminofuro[2,3-
d]pyrimidines as multireceptor tyrosine kinase and dihydrofolate
reductase inhibitors with antiangiogenic and antitumor activity. Bioorg.
Med. Chem. 2005, 13, 5475−5491.
(9) Dumontet, C.; Jordan, M. A. Microtubule-binding Agents: A
dynamic field of cancer therapeutics. Nat. Rev. Drug Discovery 2010, 9,
790−803.
(10) www.clinicaltrials.gov.
(11) Gangjee, A.; Pavana, R. K.; Li, W.; Hamel, E.; Westbrook, C.;
Mooberry, S. L. Novel water-soluble substituted pyrrolo[3,2-d]-
pyrimidines: design, synthesis, and biological evaluation as antitubulin
antitumor agents. Pharm. Res. 2012, 29, 3033−3039.
(12) Gangjee, A.; Zaware, N.; Raghavan, S.; Yang, J.; Thorpe, J. E.;
Ihnat, M. A. N(4)-(3-Bromophenyl)-7-(substituted benzyl) pyrrolo-
[2,3-d]pyrimidines as potent multiple receptor tyrosine kinase
inhibitors: design, synthesis, and in vivo evaluation. Bioorg. Med.
Chem. 2012, 20, 2444−2454.
(13) Ganguly, A.; Cabral, F. New insights into mechanisms of
resistance to microtubule inhibitors. Biochim. Biophys. Acta 2011, 1816,
164−171.
(14) Elliott, A. J.; Morris, P. E.; Petty, S. L.; Williams, C. H. An
improved synthesis of 7-substituted pyrrolo[3,2-d]pyrimidines. J. Org.
Chem. 1997, 62, 8071−8075.
(15) Luceome Biotechnologies, 1775 S. Pantano Rd, Suite 100,
Tucson, AZ 85710.
(16) Chambers, A. F. MDA-MB-435 and M14 cell lines: Identical but
not M14 melanoma? Cancer Res. 2009, 69, 5292−5293.
484
dx.doi.org/10.1021/ml4004793 | ACS Med. Chem. Lett. 2014, 5, 480−484