Design, synthesis, and structure–activity relationships of pyrimido[4,5-b]indole-4-amines as microtubule depolymerizing agents that are effective against multidrug resistant cells

Article abstract of DOI:10.1016/j.bmcl.2017.05.085

To identify the structural features of 9H-pyrimido[4,5-b]indoles as microtubule depolymerizers, pyrimido[4,5-b]indoles 2–8 with varied substituents at the 2-, 4- and 5-positions were designed and synthesized. Nucleophilic displacement of 2,5-substituted-4

Full text of DOI:10.1016/j.bmcl.2017.05.085

Accepted Manuscript
Design, Synthesis, and Structure-Activity Relationships of Pyrimido[4,5-b]in-
dole-4-amines as Microtubule Depolymerizing Agents that are Effective against
Multidrug Resistant Cells
Ravi Kumar Vyas Devambatla, Wei Li, Nilesh Zaware, Shruti Choudhary,
Ernest Hamel, Susan L. Mooberry, Aleem Gangjee
PII:
S0960-894X(17)30583-8
DOI:
http://dx.doi.org/10.1016/j.bmcl.2017.05.085
BMCL 25030
Reference:
To appear in:
Bioorganic & Medicinal Chemistry Letters
Received Date:
Revised Date:
Accepted Date:
12 April 2017
24 May 2017
27 May 2017
Please cite this article as: Devambatla, R.K.V., Li, W., Zaware, N., Choudhary, S., Hamel, E., Mooberry, S.L.,
Gangjee, A., Design, Synthesis, and Structure-Activity Relationships of Pyrimido[4,5-b]indole-4-amines as
Microtubule Depolymerizing Agents that are Effective against Multidrug Resistant Cells, Bioorganic & Medicinal
Chemistry Letters (2017), doi: http://dx.doi.org/10.1016/j.bmcl.2017.05.085
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Design, Synthesis, and Structure‒Activity Relationships of Pyrimido[4,5-b]indole-4-amines
as Microtubule Depolymerizing Agents that are Effective against Multidrug Resistant Cells
a
a
a
a
b
Ravi Kumar Vyas Devambatla , Wei Li , Nilesh Zaware , Shruti Choudhary , Ernest Hamel ,
c,†,
a,†,
Susan L. Mooberry *, Aleem Gangjee *
a
Division of Medicinal Chemistry, Graduate School of Pharmaceutical Sciences, Duquesne
University, 600 Forbes Avenue, Pittsburgh, PA 15282.
b
Screening Technologies Branch, Developmental Therapeutics Program, Division of Cancer
Treatment and Diagnosis, Frederick National Laboratory for Cancer Research, National
Institutes of Health, Frederick, MD 21702.
c
Department of Pharmacology, Cancer Therapy & Research Center, University of Texas Health
Science Center at San Antonio, 7703 Floyd Curl Drive, San Antonio, TX 78229.
†
*
5
These authors contributed equally to this work.
To whom correspondence should be addressed. For A.G.: phone, 412-396-6070; fax, 412-396-
593; e-mail, gangjee@duq.edu. For S.L.M.: phone, 210-567-4788; fax, 210-567-4300; e-mail,
mooberry@uthscsa.edu.
Keywords: Microtubules; pyrimido[4,5-b]indoles; microtubule depolymerizing agents; multidrug
resistance; conformational restriction
Abbreviations: National Cancer Institute (NCI); microtubule targeting agents (MTAs);
combretastatin A-4 (CA-4); dimethyl sulfoxide (DMSO); Protein Data Bank (PDB); P-
glycoprotein (Pgp); relative resistance (Rr); nuclear magnetic resonance (NMR).
1

Abstract
To identify the structural features of 9H-pyrimido[4,5-b]indoles as microtubule
depolymerizers, pyrimido[4,5-b]indoles 2‒8 with varied substituents at the 2-, 4- and 5-positions
were designed and synthesized. Nucleophilic displacement of 2,5-substituted-4-chloro-
pyrimido[4,5-b]indoles with appropriate arylamines was the final step employed in the synthesis
of target compounds 2‒8. Compounds 2 and 6 had two-digit nanomolar potency (IC ) against
5
0
MDA-MB-435, SK-OV-3 and HeLa cancer cells in vitro. Compounds 2 and 6 also
depolymerized microtubules comparable to the lead compound 1. Compounds 2, 3, 6 and 8 were
effective in cells expressing P-glycoprotein or the βIII isotype of tubulin, mechanisms that are
associated with clinical drug resistance to microtubule targeting drugs. Proton NMR and
molecular modeling studies were employed to identify the structural basis for the microtubule
depolymerizing activity of pyrimido[4,5-b]indoles.
Microtubules are key components of the cytoskeleton for all eukaryotes and are involved in
1
, 2
critical cellular processes including cell division, trafficking, signaling and migration. These
functions depend on the dynamic nature of microtubules, which arises from complex growth and
shortening events related to GTP hydrolysis that is facilitated by microtubule associated
3
proteins. Microtubule targeting agents (MTAs) disrupt microtubule dynamics by altering αβ
tubulin heterodimer addition and loss, which in turn disrupts microtubule-dependent events. In
vitro, microtuble disruption initiates mitotic arrest and cell death, but evidence suggests that
2
, 4
additional mechanisms are involved in the clinical activity of microtubule targeting drugs.
1
These drugs are some of the most successful anticancer agents used clinically. In addition,
MTAs are the only class of cytotoxic anticancer agents effective against p53-mutant cell lines,
2

which constitute 39 of the 58 cell lines in the National Cancer Institute (NCI) 60-cancer cell line
5
, 6
panel.
Figure 1. Structurally diverse microtubule depolymerizing agents.
MTAs are classified as microtubule polymerizing agents, which promote microtubule
polymerization, or as microtubule depolymerizing agents, which inhibit microtubule
polymerization. Microtubule depolymerizers can be further divided into three classes based on
their tubulin binding site: the vinca domain, the maytansine site and the colchicine site. The
vinca alkaloids, including vincristine, vinblastine and vindesine (Figure 1), bind competitively
7
within the vinca site. The vinca alkaloids are indicated for both adult and pediatric cancers.
8
Eribulin (Figure1), a simplified analog of halichondrin B, binds within the vinca domain, and
8
,9
has unique effects on microtubule dynamics. Structurally diverse natural products including
rhizoxin and maytansine (Figure 1) bind to a different site on β-tubulin, referred to as the
1
0
maytansine site.
The occupancy of this site by maytansine causes microtubule
3

1
0
depolymerization by inhibiting longitudinal tubulin interactions. Eribulin and maytansine both
have clinical utility, maytansine is the cytotoxin in the antibody-drug conjugate trastuzumab
1
1
emtansine used for HER2-positive breast cancer and eribulin is approved for the treatment of
1
2
13
metastatic or locally advanced breast cancer and liposarcoma. The colchicine site is a non-
1
4, 15
overlapping binding site located on -tubulin at its interface with -tubulin.
While
colchicine (Figure 1) is too toxic for use in cancer therapy, multiple colchicine-site binding
1
6-19
agents including combretastatin A-4 phosphate (CA-4P, fosbretabulin),
combretastatin A-1
2
0
21, 22
diphosphate (CA-1P, OXi4503), and 2-methoxyestradiol
(Figure 1) were evaluated in early
phase clinical trials. The microtubule polymerizers (stabilizers) that increase the density of
microtubules in cells and disrupt microtubule dynamics include the taxanes, laulimalide,
2
, 23, 24
peloruside A, epothilones, zampanolide and taccalonolides.
However, only paclitaxel,
docetaxel and an epothilone B derivative ixabepilone are used clinically. Paclitaxel and
docetaxel are widely used in the treatment of solid tumors such as breast, prostate, gastric, and
lung cancers amongst others, ixabepilone is used in the US for the treatment of refractory
2
3
metastatic breast cancer.
2
5
Multidrug resistance is a major factor in the failure of cancer chemotherapy. Expression of
the ABC transporter, P-glycoprotein (Pgp) or the expression of III-tubulin are two major
2
6, 27
mechanisms of tumor resistance to taxanes and vinca alkaloids.
Development of MTAs that
2
8
circumvent Pgp and/or βIII-tubulin-mediated resistance could have advantages in patients who
fail to respond to current MTAs. Most of the colchicine site agents circumvent Pgp and III-
2
8, 29
tubulin mediated resistance
and could be beneficial. However, thus far, no colchicine site
agent has been approved as an anticancer agent. Hence, this site provides new opportunities for
drug discovery.
4

2
9
Gangjee et al. reported pyrimido[4,5-b]indole 1 (Figure 2) as a potent microtubule
depolymerizer (EC of 133 nM) with potent in vitro cytotoxic activity in MDA-MB-435 cells
5
0
(
IC of 14.7 nM). Compound 1 is a colchicine site agent and it also circumvents clinically
50
relevant Pgp and III-tubulin mediated resistance. Compounds 2‒8 (Figure 2) were designed to
identify key structural features of 1 responsible for microtubule depolymerizing activities and
binding within the colchicine site.
Figure 2. Parent pyrimido[4,5-b]indole 1 and target compounds 2‒8.
Compounds 2 and 3 (Figure 2) with 2-Me and 2-H substitutions, respectively, were designed
to evaluate the importance of the 2-NH group for microtubule depolymerizing activity. Similar
2
3
0
substitutions were incorporated in the cyclopenta[d]pyrimidine series and resulted in improved
potency in some cases. Compounds 4 and 5, which have different substitutions at the 5-position,
were designed to evaluate the role of the electronics of the C-ring and/or hydrophobicity on
microtubule depolymerization. Compounds 4 and 5, in addition, conformationally restrict the
5

rotation around the N-C4 bond (a) in 1 (Figure 2) as well as the N-C1’ bond (b) in 1 due to steric
hindrance caused by the presence of a larger 5-Me or 5-Cl moiety respectively, instead of the 5-
H in 1. Compounds 6 and 7 were designed as conformationally restricted analogs by
incorporating the bicyclic 6-methoxy-tetrahydroquinoline moiety onto the 4-position of the
pyrimido[4,5-b]indole. Compared to the N-methylanilines, 1‒5, the tetrahydroquinoline moiety
of 6 and 7 eliminates rotation around the b bond, thereby restricting the conformation of the
phenyl ring. The restricted conformation of the phenyl group in 6 and 7 results in a much more
rigid structure than 1 and 4, respectively, but still maintains the phenyl and alkyl substitutions on
4
the N -position. Compound 8 was designed as a bioisostere of 1 by replacement of the 4'-OMe
group with a 4'-SMe moiety to determine the bulk tolerance (O vs S) as well as the importance of
hydrogen bonding.
a
Scheme 1. Synthesis of target compounds 2 and 3
a
Reagents and conditions: (i) POCl , reflux; (ii) 4-methoxy-N-methylaniline, conc. HCl (2
3
drops), BuOH, reflux.
31
32
Synthesis of target compounds 2 and 3 is shown in Scheme 1. Compounds 9 and 10 were
synthesized using reported literature procedures. Treatment of 9 and 10 with POCl afforded the
3
4
-chloro-pyrimido[4,5-b]indoles 11 and 12, respectively. Displacement of the 4-Cl of 11 and 12
6

with 4-methoxy-N-methyl aniline provided target compounds 2 and 3 in 78% and 38% yields,
respectively.
a
Scheme 2. Synthesis of target compounds 4 and 5
a
o
Reagents and conditions: (i) CNCH CO Et, t-BuOK, THF, reflux; (ii) Zn dust, AcOH, 60 C;
2
2
o
o
(
iii) NH (C=NH)Cl·HCl, DMSO , 120 C; (iv) Piv O, 120 C; (v) POCl , reflux; (vi) 4-methoxy-
2 2 2 3
N-methylaniline, conc. HCl (2 drops), i-PrOH, reflux; (vii) 1 N NaOH, i-PrOH, reflux.
The synthesis of target compounds 4 and 5 is shown in Scheme 2. Displacement of the 2-F
of commercially available 13 with ethyl cyanoacetate anion provided 14 in 82% yield. Reduction
of the nitro group of 14 followed by cyclization furnished the indole 15. Cyclocondensation of
1
1
5 with carbamimidic chloride hydrochloride afforded the 2-amino-4-oxo-pyrimido[4,5-b]indole
6. Pivaloyl protection of the 2-NH of 16 gave 17, which, upon chlorination at the 4-position,
2
3
3
provided 18 in 86% yield. Nucleophilic displacement of the 4-Cl of 18 and 19 with 4-methoxy-
N-methyl aniline provided target compounds 4 and 5, respectively.
7

a
Scheme 3. Synthesis of target compounds 6–8
a
Reagents and conditions: (i) compound 21, conc. HCl (2‒3 drops), BuOH, reflux; (ii) compound
2
1, conc. HCl (2 drops), i-PrOH, reflux; (iii) 1 N NaOH, i-PrOH, reflux; (iv) compound 22,
conc. HCl (2 drops), BuOH, reflux.
2
9
Displacement of the 4-Cl group of 20
(Scheme 3) with 6-methoxy-1,2,3,4-
tetrahydroquinoline 21 and concomitant deprotection of the pivaloyl group from the 2-NH2
group afforded target compound 6 in 26% yield. Treatment of 19 (Scheme 2) with 21, followed
by base-mediated deprotection of the 2-NH group, provided target compound 7 in 52% yield.
2
2
9
34
Compound 8 was obtained by treating 20 with 4-thiomethyl-N-methylaniline 22 in BuOH at
reflux.
Effect on cellular microtubules and cancer cell proliferation. Compounds 2–8 were
evaluated for their ability to depolymerize microtubules in A-10 smooth muscle cells with a
phenotypic assay that measures the percent cellular microtubule loss. An EC , the concentration
50
8

that causes 50% cellular microtubule loss, was determined. The antiproliferative effects in
MDA-MB-435 melanoma cells using the sulforhodamine B (SRB) assay and the IC , the
5
0
concentration that cause 50% inhibition of proliferation was determined. The activities of the
compounds were compared with 1, paclitaxel and CA-4 (Table 1).
Table 1. Microtubule depolymerizing activities (EC ) and antiproliferative values (IC ) of
50
50
2
‒8.
EC for microtubule
MDA-MB-435
IC ± SD (nM)
EC /IC
5
0
50
50
Compd
depolymerization (nM)
Ratio
50
a
1
130
14.7 ± 1.5
33.9 ± 3.4
130 ± 7.8
220 ± 25
3,900 ± 200
54.4 ± 4
8.8
3.8
9.2
6.5
2
130
3
1200
4
1400
5
No effects up to 40 µM
6
150
2.8
15
12
7
3700
1100
271 ± 4
8
89.1 ± 10
4.5 ± 0.5
b
paclitaxel
b
CA-4
9.8
4.4 ± 0.5
2.2
a
29 b
35
Results previously published. Results previously published.
The 2-Me analog 2 has a 2fold lower potency for cancer cell cytotoxicity than the 2-NH 1,
2
but was equipotent with 1 for effects on cellular microtubules and had a lower EC /IC ratio,
5
0
50
3
.8 as compared to 8.8 for 1 and this indicates an improved correlation between the microtubule
depolymerizing effects and the cancer cell cytotoxicity. The 2-H analog 3 displayed ~10fold
lower potency in both cell proliferation and microtubule depolymerizing assays, than the 2- NH2
9

1
, indicating the importance of the size of the substitution at the 2-position for microtubule
depolymerizing activity among compounds 1‒3. Replacement of the 5-H of 1 with a 5-Me (4)
resulted in 15 and 11fold lower antiproliferative and microtubule depolymerizing potencies,
respectively. Substitution of a 5-Cl (5) moiety for the 5-H of 1 resulted in a 263fold loss of
potency for the antiproliferative effects and no effects on cellular microtubules at concentration
up to 40 µM. These data indicate that substitution of a Me or Cl at the 5-position results in
decreased potency, perhaps as a consequence of steric hindrance and/or conformational
restriction of the rotation of the bonds (a) and/or (b) (Figure 2). Compound 6, the
tetrahydroquinoline substituted is conformationally restricted around bond (b) and is a 5-
unsubstituted analog, that allows rotation of the (a) bond, is only marginally less active than 1
and 2 for microtubule loss and for inhibiting MDA-MB-435 tumor cells (Table 1). Further,
conformationally restricted 7, the 5-Cl substituted analog, was significantly less potent than 1
and 6 against MDA-MB-435 tumor cells and was less potent at inhibiting microtubule loss as
well. Compound 6 compared to 7 was markedly more potent indicating that conformational
4
restriction of the N -phenyl moiety about (b) bond with a 1,2,3,4-tetrahydroquinoline group is
only slightly detrimental to activity. However, further conformational restriction of (a) bond via
5
-position substitution as in 7 significantly decreases activity. These results together with the
results obtained with the 5-substituted compounds 4 and 5 suggest that conformational restriction
of the (b) bond is not particularly detrimental to MDA-MB-435 antiproliferative activity
(compare 6 with 1 and 2) but further conformational restriction via the (a) bond is detrimental.
The 5-position Me and Cl substitution in 4, 5 and 7 could also cause steric hindrance to binding
at the tubulin site which could also contribute to their lower activities. Replacement of the 4'-
1
0

OMe of 1 with a 4'-SMe group (8) resulted in a 6fold decrease in activity (IC ), clearly
5
0
indicating that this isosteric replacement of the OMe with a SMe is not conducive to activity.
Inhibition of tubulin assembly and colchicine binding. On the basis of their microtubule
depolymerizing activities, compounds 2, 3, 6 and 8 were evaluated for their direct effects on
tubulin assembly and inhibition of colchicine binding (Table 2). Compound 2 inhibited tubulin
assembly with activity comparable to the lead compound 1 and CA-4. Compounds 3, 6 and 8
were marginally (1.6-fold) less potent than 1 as inhibitors of tubulin assembly. Compounds 2, 3,
3
6
and 8 inhibited the binding of [ H]colchicine to tubulin by 62‒71%, whereas the lead
3
compound 1 showed 84% inhibition of [ H]colchicine binding. These data indicate that the
target compounds 2, 3, 6 and 8 have somewhat less affinity for the colchicine site as compared to
1
.
Table 2. Inhibition of tubulin assembly and colchicine binding.
Compd
Inhibition of tubulin assembly Inhibition of colchicine binding
IC ± SD (µM)
(% inhibition ± SD) at 5 µM
5
0
a
1
1.4 ± 0.007
1.2 ± 0.04
2.3 ± 0.4
2.2 ± 0.1
2.3 ± 0.3
1.0 ± 0.09
84 ± 0.5
67 ± 5
62 ± 4
71 ± 5
67 ± 5
99 ± 0.2
2
3
6
8
a
CA-4
aResults previously published.29 bND: not determined.
1
1

Effect on Pgp and βIII-tubulin mediated drug resistance in cancer cells. Pgp and βIII-
3
6
tubulin-mediated drug resistance were observed clinically with the taxanes and vinca alkaloids.
MTAs that are not sensitive to these drug resistance mechanisms would be beneficial in tumors
resistant to other MTAs because of these mechanisms. Hence, target compounds were evaluated
for the ability to overcome Pgp or βIII-tubulin-mediated resistance in isogenic SK-OV-3 and
3
7
HeLa cell line pairs, respectively using the SRB assay (Table 3). The relative resistance value,
designated Rr, was calculated by dividing the IC obtained in the Pgp-expressing SK-OV-3
50
MDR-1-M6/6 cells by the IC obtained in the parental SK-OV-3 ovarian cancer cells. In the SK-
5
0
OV-3 cell line pair, the Pgp-expressing cells are resistant to paclitaxel, with a Rr value of 240,
demonstrating the high susceptibility of paclitaxel to Pgp-mediated transport. Target compounds
2
, 3, 6 and 8 had Rr values of less than 2, similar to that observed with CA-4.
Table 3. Target compounds 2, 3, 6 and 8 circumvent Pgp and βIII-tubulin mediated resistance.
IC ± SD (nM)
IC ± SD (nM)
50
5
0
Compd
SK-OV-3-
SK-OV-3
Rr
HeLa
HeLa WTβIII
Rr
MDR1-M6/6
a
1
27.6 ± 1.8
60.5 ± 2.4
173 ± 8.6
83.2 ± 5.7
156 ± 16
5.0 ± 0.6
34.4 ± 5.9
78.0 ± 8.4
224 ± 21
135 ± 23
160 ± 15
1,200 ± 58
1.2
1.3
1.4
1.6
1
21.3 ± 2.2
50.5 ± 5.3
142 ± 8.1
72.1 ± 7.9
118 ± 13
2.8 ± 0.4
21.4 ± 3.5
31.8 ± 2.5
99.5 ± 12
87.0 ± 4.1
78.4 ± 4
1.0
0.6
0.8
1.2
0.7
8.6
2
3
6
8
b
paclitaxel
240
24.0 ± 3
b
CA-4
5.5 ± 0.5
7.2 ± 1.1
1.3
3.3 ± 0.4
3.3 ± 0.3
1
a
29 b
35
Results previously published. Results previously published.
1
2

A HeLa cell line pair was used to evaluate the effects of βIII-tubulin on the sensitivity of the
cell line for 2, 3, 6 and 8 (Table 3). The IC of each of these compounds were determined and
5
0
the relative resistance (Rr) value was calculated by dividing the IC50 of the βIII-tubulin
expressing cells by the IC value obtained in HeLa parental cells. Paclitaxel is less potent in the
50
βIII-tubulin expressing cells with a Rr value of 8.6. On the other hand, compounds 2, 3, 6 and 8
had Rr values of ≈1.0 implying equal sensitivity to the III-tubulin expressing cell line. These
data suggest that these compounds overcome both Pgp and βIII-tubulin mediated resistance and
in that regard, have advantages over the paclitaxel.
To rationalize the structure-activity relationships summarized above, docking of target
compounds 2‒8 was carried out in the X-ray crystal structure of colchicine in the colchicine site
3
8
39
of tubulin (PDB: 4O2B, 2.30 Å) using Molecular Operating Environment (MOE 2015.10).
Multiple low energy conformations were obtained on docking. The protein was prepared as
3
5
reported previously. Ligands were sketched using the builder function in MOE and minimized
using the Amber10:EHT forcefield. The ligands were then docked in the binding site using the
default settings in the docking protocol. The placement was performed using Triangle Matcher
and scored using London dG. The refinement was carried out using Rigid Receptor and scored
using GBVI/WSA dG. To validate the docking study at the colchicine site, the native ligand
colchicine was re-docked into the binding site using the same set of parameters as described
above. The rmsd of the best docked pose was 0.345 Å, thus validating docking using MOE.
1
3

Figure 3. Superimposition of the docked poses of 1 (black), 2 (green) and 3 (magenta) in the
colchicine (pink) binding site of tubulin (PDB ID: 4O2B).
Figure 3 shows the superimposed docked poses of 1 (black), 2 (green) and 3 (magenta) in
3
8
the colchicine site of tubulin (PDB: 4O2B, 2.30 Å). The pyrimido[4,5-b]indole scaffold of 1, 2
and 3 forms hydrophobic interactions with Alaα180, Valα181, Leuβ248, Asnβ258, Metβ259,
4
Thrβ314 and Lysβ352 and overlaps with the C-ring of colchicine. The N -Me group lies in a
4
pocket lined by hydrophobic residues Leuβ248 and Alaβ354. The N -aryl moiety makes
hydrophobic interactions with Cysβ241, Alaβ250, Leuβ255, and Alaβ316 and overlaps with ring
A of colchicine. The 4'-OMe group points towards residues Cysβ241, Leuβ242 and Leuβ255 and
superposes with the 3-OMe group of colchicine. The 2-NH of 1 undergoes hydrogen bonding
2
with HOH623. The 2-Me group in 2 retains hydrophobic interactions with Leuβ248 (3.90 Å) and
Serα178 (3.99 Å). For compound 3, the 2-H group is incapable of hydrogen bonding or
hydrophobic interactions with the surrounding residues, and this results in a 10fold loss in both
1
4

the IC and EC compared to 1 with its 2-NH moiety. The best docked poses of 1, 2 and 3 had
5
0
50
2
scores of −7.08 kcal/mol, −7.18 kcal/mol and −6.84 kcal/mol, respectively, which suggests lower
affinity of 3 and similar affinity of 1 and 2 in the colchicine site.
Figure 4. Superimposition of the docked poses of 1 (black) and 4 (cyan) in the colchicine (pink)
site of tubulin (PDB ID: 4O2B). Hydrophobic surface is in green and hydrophilic surface is in
pink.
Figure 4 shows the superimposed docked poses of 1 (black) and 4 (cyan) in the colchicine
site of tubulin. Compound 4 retains the key binding interactions at the colchicine site as
described for 1 in Figure 3. The 2-NH of 1 undergoes hydrogen bonding with HOH623,
2
whereas the 2-NH of 4 undergoes hydrogen bonding with HOH606. The lower activity of 4 and
2
5
as compared to 1 can be rationalized in part, on the basis of a small hydrophobic pocket in the
binding site. The 5-Me in 4 lies in a hydrophobic pocket lined by residues Metβ259, Alaβ316
1
5

and Lysβ352. This small hydrophobic pocket results in steric hindrance probably leading to high
energy conformations for 4 and 5, resulting in lower activities of 4 and 5 as compared to 1. In
addition, the best docked pose of 4 had a score of −5.96 kcal/mol which is higher (worse) than
the score (−7.08 kcal/mol) for the best docked pose of 1.
1
Structural explanation of activity with conformational preferences from H NMR data.
1
4
4
The H NMR spectra of the target compounds 2‒8, the lead N -Me compound 1 and the N -H
compound 23 (Figure 5) in DMSO-d provided valuable information regarding the role of
6
conformational restriction in microtubule depolymerization activity of the tricyclic pyrimido[4,5-
4
b]indoles. The N -Me compound 1 showed excellent microtubule depolymerization and tumor
4
29
cell inhibitory activities compared to the N -H analog 23. This difference could be attributed, in
part, to conformational restriction of the bonds connecting the nitrogen to the pyrimidine ring (a)
(N-C4) and to the phenyl ring (b) (N-C1’) (Figure 5). The (a) and (b) bonds in 1 are somewhat
4
restricted compared to those in 23 due to the presence of steric bulk of the N -Me group. For 23,
the “5-H” proton appears at > 6.80, whereas for 1, it is more shielded at = 5.75. This
shielding of the “5-H” proton in 1 is due to the diamagnetic anisotropy in 1 arising from the
orientation of the phenyl ring as depicted by the more favored conformation of 1 in Figure 5. The
4
steric bulk of the N -Me in 1 restricts the conformation and thus positions the phenyl group on
top of the 5-H moiety, resulting in the observed shielding effect on the 5-H group.
1
6

4
Figure 5. Conformational restriction of C4-N (a) and N-C1' (b) bonds in the presence of the N -
1
Me moiety and values ( H NMR) of the 5-H group for 1 and 23.
For target compounds 2 (2-Me) and 3 (2-H), conformational restriction of bond (a) due to
4
1
the N -Me causes shielding of the “5-H” proton (= 5.90, H NMR, Figure 6) similar to that
observed in 1.
1
Figure 6. Shielding (values, H NMR) of the 5-H proton in 2 and 3.
4
Figure 7. Substitution at the 5-position hinders the free rotation of the N -aryl moiety.
1
7

Compounds 4 and 5 were less active than 1 due to increased hydrophobicity and/or steric
hindrance caused by the 5-Me and 5-Cl groups, respectively. The steric hindrance resulting from
the 5-Me (4) or the 5-Cl (5) group conformationally restricts rotation around bond (b) and
decreases the total number of low energy conformations for 4 and 5 (Figure 7). Conformational
4
0
search of 1, 4 and 5 was carried out using Sybyl-X 2.1.1 (default protocol) to assess the number
of low energy conformations within 1 kcal/mol or 5 kcal/mol with 1° rotation about (a) bond
(Table 4). Compounds 4 and 5 both had a lower number of conformations than 1 due to the steric
hindrance, thus making the adoption of the bioactive conformation(s) for 4 and 5 in the
colchicine site more difficult thus contributing to the lower affinity compared to 1. Clearly the 5-
Me and 5-Cl moieties may also cause steric hindrance to tubulin binding which could also
contribute to the lower activities of 4 and 5 compared to 1. The lower potency of the 4’-SMe
analog 8 is attributed to the isosteric replacement of the 4’-OMe in 1 as discussed above, since,
the conformational profile of 8 (calculated using Sybyl-X 2.1.1) was similar to that of 1.
Table 4. Systematic Search (conformations) allowing 1 degree rotation of C -N- C (of pyrimido
4
1'
indole moiety) using Sybyl-X 2.1.1.
Compd Conformations within Conformations within
1
kcal/ mol
5 kcal/ mol
1
4
5
(5-H)
31
22
15
65
37
43
(5-Me)
(5-Cl)
1
8

In conclusion, we reported the design, synthesis and structure‒activity relationship (SAR) of
substituted tricyclic pyrimido[4,5-b]indoles as microtubule depolymerizers. Compound 2,
containing the 2-Me group, was as potent as the 2-NH analog 1 and both 1 and 2 were better
2
than the 2-H analog 3, which indicates that substitution at the 2-position with a 2-Me or 2-NH is
2
important for activity. Substitution at the 5-position of the pyrimido[4,5-b]indole scaffold with a
4
5
1
-Me or 5-Cl is detrimental for activity. Conformational restriction of the N -aryl group using a
,2,3,4-tetrahydroquinoline moiety without a 5-substitution (6) only marginally decreased
potency compared to 1 and 2 in the MDA-MB-435 tumor cell lines. Compound 7 with similar b
bond restriction as 6, but with a sterically bulky 5-Cl substitution substantially decreased
activity. Replacement of the 4'-OMe with a 4'-SMe group (8) also resulted in an increased EC50
for microtubule depolymerization, indicating the probable role of hydrogen bonding of the 4'-
OMe and/or large size interaction with tubulin at this position. Proton NMR and molecular
modeling were utilized to explain how subtle structural changes of the substituents on the
pyrimido[4,5-b]indole scaffold affected the microtubule depolymerization and MDA-MB-435
inhibitory activity. This SAR information will be employed to design the next series of tricyclic
pyrimido[4,5-b]indoles with improved microtubule depolymerizing and tumor cell inhibitory
activities.
Acknowledgements. This work was supported, in part, by a grant from the National Institutes of
Health, National Cancer Institute (CA142868 (AG, SLM)); by the Duquesne University Adrian
Van Kaam Chair in Scholarly Excellence (AG); and by an NSF equipment grant for NMR
instrumentation (NMR: CHE 0614785).
1
9

Disclaimer. The content of this paper is solely the responsibility of the authors and does not
necessarily reflect the official views of the National Institutes of Health.
References
1
2
3
4
.
.
.
.
Dumontet, C.; Jordan, M. A. Nat. Rev. Drug Discov. 2010, 9, 790-803.
Field, J. J.; Kanakkanthara, A.; Miller, J. H. Bioorg. Med. Chem. 2014, 22, 5050-5059.
Bowne-Anderson, H.; Hibbel, A.; Howard, J. Trends Cell Biol. 2015, 25, 769-779.
Komlodi-Pasztor, E.; Sackett, D.; Wilkerson, J.; Fojo, T. Nat. Rev. Clin. Oncol. 2011, 8,
2
44-250.
5
6
.
.
Wang, Z.; Sun, Y. Transl. Oncol. 2010, 3, 1-12.
O'Connor, P. M.; Jackman, J.; Bae, I.; Myers, T. G.; Fan, S.; Mutoh, M.; Scudiero, D. A.;
Monks, A.; Sausville, E. A.; Weinstein, J. N.; Friend, S.; Fornace, A. J.; Kohn, K. W.
Cancer Res. 1997, 57, 4285-4300.
7
.
Lee, J. F. and Harris, L. N. Antimicrotubule agents. In Cancer: Principles & Practice of
Oncology 8th Ed., V. T. DeVita, Jr., T. S. Lawrence and S. A. Rosenberg Eds., Lippincott
Williams & Wilkins, 2008, 447–456.
8
9
.
.
Dybdal-Hargreaves, N. F.; Risinger, A. L.; Mooberry, S. L. Clin. Cancer Res. 2015, 21,
2
445-2452.
Doodhi, H.; Prota, A. E.; Rodriguez-Garcia, R.; Xiao, H.; Custar, D. W.; Bargsten, K.;
Katrukha, E. A.; Hilbert, M.; Hua, S.; Jiang, K.; Grigoriev, I.; Yang, C. P.; Cox, D.;
Horwitz, S. B.; Kapitein, L. C.; Akhmanova, A.; Steinmetz, M. O. Curr. Biol. 2016, 26,
1
713-1721.
2
0

1
1
1
0. Prota, A. E.; Bargsten, K.; Diaz, J. F.; Marsh, M.; Cuevas, C.; Liniger, M.; Neuhaus, C.;
Andreu, J. M.; Altmann, K.-H.; Steinmetz, M. O. Proc. Natl. Acad. Sci. 2014, 111, 13817-
1
3821.
1. Verma, S.; Miles, D.; Gianni, L.; Krop, I. E.; Welslau, M.; Baselga, J.; Pegram, M.; Oh, D.-
Y.; Diéras, V.; Guardino, E.; Fang, L.; Lu, M. W.; Olsen, S.; Blackwell, K. N. Engl. J. Med.
2
012, 367, 1783-1791.
2. Cortes, J.; O'Shaughnessy, J.; Loesch, D.; Blum, J. L.; Vahdat, L. T.; Petrakova, K.; Chollet,
P.; Manikas, A.; Diéras, V.; Delozier, T.; Vladimirov, V.; Cardoso, F.; Koh, H.; Bougnoux,
P.; Dutcus, C. E.; Seegobin, S.; Mir, D.; Meneses, N.; Wanders, J.; Twelves, C. Lancet
2
011, 377, 914-923.
1
1
1
3. Thomas, C.; Movva, S. Onco Targets Ther. 2016, 9, 5619-5627.
4. Lu, Y.; Chen, J.; Xiao, M.; Li, W.; Miller, D. Pharm. Res. 2012, 29, 2943-2971.
5. Massarotti, A.; Coluccia, A.; Silvestri, R.; Sorba, G.; Brancale, A. ChemMedChem 2012, 7,
3
3-42.
1
1
1
6. Liu, P. Q., Y.; Wu, L.; Yang, S.; Li, N.; Wang, H.; Xu, H.; Sun, K.; Zhang, S.; Han, X.;
Sun, Y.; Shi, Y. Anticancer Drugs 2014, 25, 462-471.
7. Ng, Q.-S.; Mandeville, H.; Goh, V.; Alonzi, R.; Milner, J.; Carnell, D.; Meer, K.; Padhani,
A. R.; Saunders, M. I.; Hoskin, P. J. Ann. Oncol. 2012, 23, 231-237.
8. Nathan, P.; Zweifel, M.; Padhani, A. R.; Koh, D.-M.; Ng, M.; Collins, D. J.; Harris, A.;
Carden, C.; Smythe, J.; Fisher, N.; Taylor, N. J.; Stirling, J. J.; Lu, S.-P.; Leach, M. O.;
Rustin, G. J. S.; Judson, I. Clin. Cancer Res. 2012, 18, 3428-3439.
2
1

1
2
9. Zweifel, M.; Jayson, G. C.; Reed, N. S.; Osborne, R.; Hassan, B.; Ledermann, J.; Shreeves,
G.; Poupard, L.; Lu, S.-P.; Balkissoon, J.; Chaplin, D. J.; Rustin, G. J. S. Ann. Oncol. 2011,
2
2, 2036-2041.
0. Patterson, D. M.; Zweifel, M.; Middleton, M. R.; Price, P. M.; Folkes, L. K.; Stratford, M.
R. L.; Ross, P.; Halford, S.; Peters, J.; Balkissoon, J.; Chaplin, D. J.; Padhani, A. R.; Rustin,
G. J. S. Clin. Cancer Res. 2012, 18, 1415-1425.
2
2
1. Bruce, J.; Eickhoff, J.; Pili, R.; Logan, T.; Carducci, M.; Arnott, J.; Treston, A.; Wilding,
G.; Liu, G. Invest. New Drugs 2012, 30, 794-802.
2. Harrison, M.; Hahn, N.; Pili, R.; Oh, W.; Hammers, H.; Sweeney, C.; Kim, K.; Perlman, S.;
Arnott, J.; Sidor, C.; Wilding, G.; Liu, G. Invest. New Drugs 2011, 29, 1465-1474.
3. Rohena, C. C.; Mooberry, S. L. Nat. Prod. Rep. 2014, 31, 335-355.
4. Zhao, Y.; Mu, X.; Du, G. Pharmacol. Ther. 2016, 162, 134-143.
5. Patel, N. H.; Rothenberg, M. L. Invest. New Drugs 1994, 12, 1-13.
6. Kavallaris, M. Nat. Rev. Cancer 2010, 10, 194-204.
2
2
2
2
2
2
7. Fojo, T.; Menefee, M. Ann. Oncol. 2007, 18, 3-8.
8. Stengel, C.; Newman, S. P.; Leese, M. P.; Potter, B. V. L.; Reed, M. J.; Purohit, A. Brit. J.
Cancer 2010, 102, 316-324.
2
3
3
9. Gangjee, A.; Zaware, N.; Devambatla, R. K. V.; Raghavan, S.; Westbrook, C. D.; Dybdal-
Hargreaves, N. F.; Hamel, E.; Mooberry, S. L. Bioorg. Med. Chem. 2013, 21, 891-902.
0. Gangjee, A.; Zhao, Y.; Raghavan, S.; Rohena, C. C.; Mooberry, S. L.; Hamel, E. J. Med.
Chem. 2013, 56, 6829-6844.
1. Showalter, H. D. H.; Bridges, A. J.; Zhou, H.; Sercel, A. D.; McMichael, A.; Fry, D. W. J.
Med. Chem. 1999, 42, 5464-5474.
2
2

3
3
3
3
2. Tichý, M.; Pohl, R.; Xu, H. Y.; Chen, Y.-L.; Yokokawa, F.; Shi, P.-Y.; Hocek, M. Bioorg.
Med. Chem. 2012, 20, 6123-6133.
3. Gangjee, A.; Zaware, N.; Raghavan, S.; Ihnat, M.; Shenoy, S.; Kisliuk, R. L. J. Med. Chem.
2
010, 53, 1563-1578.
4. Abdel-Magid, A. F.; Carson, K. G.; Harris, B. D.; Maryanoff, C. A.; Shah, R. D. J. Org.
Chem. 1996, 61, 3849-3862.
5. Devambatla, R. K. V.; Namjoshi, O. A.; Choudhary, S.; Hamel, E.; Shaffer, C. V.; Rohena,
C. C.; Mooberry, S. L.; Gangjee, A. J. Med. Chem. 2016, 59, 5752-5765.
6. Perez, E. A. Mol. Cancer Ther. 2009, 8, 2086-2095.
3
3
7. Risinger, A. L.; Jackson, E. M.; Polin, L. A.; Helms, G. L.; LeBoeuf, D. A.; Joe, P. A.;
Hopper-Borge, E.; Luduena, R. F.; Kruh, G. D.; Mooberry, S. L. Cancer Res. 2008, 68,
8
881-8888.
3
3
4
8. Prota, A. E.; Danel, F.; Bachmann, F.; Bargsten, K.; Buey, R. M.; Pohlmann, J.; Reinelt, S.;
Lane, H.; Steinmetz, M. O. J. Mol. Biol. 2014, 426, 1848-1860.
9. Molecular Operating Environment (MOE 2013.08); Chemical Computing Group, Inc.:
Montreal, Quebec, Canada, 2013; www.chemcomp.com.
0. SYBYL-X 2.1.1, Tripos International, 1699 South Hanley Rd., St. Louis, Missouri, 63144,
USA.
Supplementary Material. Experimental procedures for the synthesis, compound
characterization, and biological evaluation of the target compounds. This material is available
free of charge via the Internet.
2
3

2
4