Aza-Flavanones as potent cross-species microRNA inhibitors that arrest cell cycle

Article abstract of DOI:10.1016/j.bmcl.2011.10.061

aza-Flavanones have been identified as a new class of selective microRNA inhibitors. These compounds were found to arrest cell cycle via a novel cross species microRNA-dependent regulatory pathway interpreting an unexpected link between cell cycle arrest and microRNA mediated control in cancer.

Full text of DOI:10.1016/j.bmcl.2011.10.061

Bioorganic & Medicinal Chemistry Letters 22 (2012) 645–648
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
aza-Flavanones as potent cross-species microRNA inhibitors that arrest
cell cycle
Srivari Chandrasekhar a, , Sreerangam N.C.V.L. Pushpavalli b, Srinivas Chatla b, Debasmita Mukhopadhyay b,
⇑
Bogonda Ganganna a, Kandi Vijeender a, Pabbaraja Srihari a, Chada Raji Reddy a, M. Janaki Ramaiah b,
Utpal Bhadra b,
⇑
a Natural Products Chemistry Division, CSIR-Indian Institute of Chemical Technology, Hyderabad 500 607, India
b Functional Genomics and Gene Silencing Group, Centre for Cellular and Molecular Biology, Hyderabad 500 607, India
a r t i c l e i n f o
a b s t r a c t
Article history:
aza-Flavanones have been identified as a new class of selective microRNA inhibitors. These compounds
were found to arrest cell cycle via a novel cross species microRNA-dependent regulatory pathway inter-
preting an unexpected link between cell cycle arrest and microRNA mediated control in cancer.
Ó 2011 Elsevier Ltd. All rights reserved.
Received 18 August 2011
Revised 18 October 2011
Accepted 19 October 2011
Available online 24 October 2011
Keywords:
aza-Flavones
MicroRNA inhibitors
Drosophila
Cell cycle arrest
Apoptosis
Small tiny endogenous microRNA (21–23-nt) control multiple
genes during development, evolution and human diseases.1,2 Re-
cently, miRNAs have been implicated in various processes such as
apoptosis and oncogenesis.3 Misexpression of microRNAs by differ-
ent small molecules has important therapeutic relevance because
miRNA controls more than 30–40% of human genes at post-tran-
scriptional level.4 Nearly thousand miRNAs have been identified in
human and some of them were conserved in Drosophila suggesting
their evolutionary role. Functionally, one microRNA can control sev-
eral genes or conversely many microRNAs have only one target.3 Pri-
mary miRNAs (Pri-miRNA) contain stem-loop structure from which
precursor pre-miRNA (Pre-miRNA) is formed by the action of RNase
III enzyme Drosha.5–7 Animal pre-miRNAs exported into cytoplasm
are further cleaved by the Dicer enzyme to form mature miRNA.5–7
Misregulation of miRNA expression has been observed in wide vari-
ety of cancers. Cancer-associated miRNAs are used to develop effec-
tive biomarkers for individualized medicine and potential
therapeutic targets.8–10 Numerous miRNAs having proliferative
and anti-apoptotic effects were assessed for treating various can-
cers, some of them exhibited promising results in cultured cells.11
MicroRNAs have been reported to be involved in tumorigenesis
and can act as oncogenes. A small number of structurally diverse
microRNA inhibitors have been identified to control cell prolifera-
tion in tumor cells.12,13 The functional significance of microRNA
inhibitors arresting tumor cell growth are currently the focus of
intensive cancer research. In continuation of our ongoing pro-
gramme on total synthesis of natural products and investigating
their biological properties, we have demonstrated that a cyclopep-
tide azumamide with unnatural amino acid displayed better cellu-
lar histone deacetylase inhibitory activity compared to parent
natural product.14 Even though, the cyclic peptide analogue was
exhibiting interesting biological property as a non-selective cell cy-
cle inhibitor, further studies were hampered by issues in tedious
scale-up procedures. We conceived a thought that a classical aza-
trick being demonstrated by Danishefsky15 and Chandrasekhar
et al.16 might be a fundamentally challenging exercise to be inves-
tigated. As the aza-flavanone scaffold is found in large number of
natural products with broad spectrum of biological activities,17–20
they might therefore be biologically pre-validated starting materi-
als for further compound development.
Also conversion of flavanones to aza-flavanones might provide a
new insight into understanding the SAR. Towards this, a series of
aza-flavanones 1–9 (Fig. 1) were synthesized following our own
protocol involving an organo-catalytic approach21 and investigated
as microRNA inhibitors. It was found that this class of compounds
act as microRNA inhibitors and also control cell proliferation by
altering the miRNA levels which is unprecedented.
For cytotoxicity of the compounds, MTT (MTT = 3-(4,5-dimeth-
⇑
Corresponding authors.
E-mail address: srivaric@iict.res.in (S. Chandrasekhar).
ylthiazol-2-yl)-2,5-diphenyltetrazolium-bromide)
assay
was
0960-894X/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2011.10.061

646
S. Chandrasekhar et al. / Bioorg. Med. Chem. Lett. 22 (2012) 645–648
O
O
O
OH
N
H
N
H
N
H
OMe
OH
1
2
5
3
O
O
O
OAc
N
H
N
H
N
H
OAc
Br
4
6
O
O
O
O
OTs
N
H
N
H
N
H
NH
OTs
8
9
7
Figure 1. Structures of diverse aza-flavanones.
Figure 2. (a) Scheme for luciferase expression assay under the control of a miRNA
binding sequence in the 30 UTR provides an efficient miRNA screening in Drosophila
and human cells. (b) Change in luciferase signals of cells treated with compound 3
carried out which is based on mitochondrial dehydrogenase activ-
ity. These inhibitors have shown about 50% death of MCF-7 (Hu-
man breast carcinoma) cells at 4 lM concentration (IC50 value)
(Figs. S1–2), and S2 cells (Drosophila melanogaster) (Fig. S3) sug-
gesting their therapeutic potentialities. Further flow cytometry as-
say to examine cell cycle arrest caused by the compounds 3 and 4
and 4 (4 lM) relative to DMSO (0.1%) control. (c) Level of mature miR-4644 and Pri-
miR-4644 in MCF-7 and HeLa cells treated with compounds 3 and 4 relative to
DMSO (0.1%) control as determined by quantitative Real time PCR. Bars represent
S.D. determined by triplicate measurements.
at the same concentration (4 lM, IC50) showed a G1 phase arrest in
MCF-7 cells (Fig. S4). The cytotoxic effects of these conjugates in
MCF-7 cells (Breast cancer cells) have been compared with non
cancerous HEK (Human embryonic kidney cells) as well as a
MCF-10A (breast epithelial cells). We have found that these com-
pounds exhibited a profound cytotoxic effect in MCF-7 cells com-
pared to the normal non cancerous cells. There was two-fold
increase in cytotoxicity in MCF-7 cells compared to HEK cells
whereas in comparison to MCF-10A the increase in cytotoxicity
was three-fold.
expressed.7 The intensity of luciferase signal in the cells carrying
miR-14 constructs was reduced intensively relative to that of cells
with mismatched miR-14 sequence reporters. Thus, Luc-miR-14
and control reporters (Luc-control seq) are specific and react
against miR-14 only (Fig. S6). A homology search of miR-14 in hu-
man identified human miR-4644, a recently validated microRNA
(Fig. S7) which expresses abundantly in breast tumor tissues.17
The level of endogenous miR-4644 in human MCF-7 and HeLa cells
was also detected by introducing same Luc-miR-4644 reporter,
that led to a 90% decrease in the intensity of the luciferase signals
relative to that of the control luciferase linker construct. This result
indicates a high level of endogenous miR-4644 expression in MCF-
7 and HeLa cells (Fig. S8).
Subsequently, nine aza-flavanones (as initial hits) were tested
against lentiviral miR-4644-luciferase vector containing MCF-7
and HeLa cells. An average 450–500% increase of luciferase signals
was found relative to DMSO treated cells (0.1% dimethylsulfoxide
(DMSO)) (Fig. S9) that had no effect on the luciferase signal
(Fig. S10). Through several rounds of testing in two cancer cell lines
and structural modification of aza-flavanones, a preliminary struc-
ture–activity relationship was developed. The aza-flavanone 1 and
methylated hydroxyl group substituted aza-flavanone 2 reduced
the relative activities. However, structural modification by adding
two hydroxyl group or bromide group onto the aza-flavanone core
uncovered a higher activity (compounds 3 and 4), for which three-
fold increase in the luciferase signal was found at a concentration
Compounds containing a hydroxyl group or a halogen atom in
para-position of the 2-aryl substituent (compounds 3 and 4) were
preferred over analogues bearing other substituents. Replacement
of the hydroxyl groups on compound 3 with acetoxy or tosyloxy
group (compounds 6 and 9) led to the loss of miR inhibitory activity.
The analogues with other substituents (1, 2, 5, 7, 8 and 9) were found
tobe less active (Fig. S1). Thus, amongst the compounds synthesized,
the compounds 3 and 4 were found to be the best inhibitors that ar-
rest cells and block cell growth. Thus, these two compounds have
been selected for further studies as lead optimized analogues.
To test, whether aza-flavanones control the expression of
microRNAs (miRNAs) involved in cell death and cell proliferation,
a cell culture based assay was developed to screen against miRNA
targets. The miR-14 in model organisms Drosophila was selected as
a target for its anti-apoptotic activity. The miR-14 also acts in the
insulin producing neurosecretory cells in the adult Drosophila brain
to control metabolism.22–24 A direct comparison between in vitro
and in vivo screening for small molecules is important for thera-
peutic potentialities, which is easy and affordable in Drosophila.
Two luciferases reporter constructs for miR-14 activity were
constructed introducing either complementary sequence of miR-
14 or mismatched sequences in control vector assembly (Fig. S5).
The mismatched sequences carry no recognizable targets for any
natural miRNAs. A luciferase reporter gene was cloned upstream
to the assembly. These plasmids serve as sensors to detect the
presence of mature miR-14 molecules (Figs. 3a and S5). The repor-
ter constructs (Luc-miR-14 and Luc-control seq) were introduced
separately into Drosophila S2 cells, in which miR-14 is abundantly
of IC50 values (4 lM) (Figs. S9 and 2a), while, a wide range of mod-
ification of compound 5–9 reduces the luciferase signals to a 21%
and 56%, respectively (Fig. S9). Thus compounds 3 and 4 are the
most effective small-molecule inhibitors for miR-4644 function.
They induce close to 475% increase in the intensity of the luciferase
reporter signal at 4 lM (Fig. S9). However compounds 3 and 4 do
not affect the luciferase signals in MCF-7 cells that express the
Luci-linker control sequence (Fig. 2b). This result indicates that
compounds 3 and 4 only increases the intensity of the luciferase
signal through inhibition of the miRNA-4644 pathway in the hu-
man tumor cell lines. To quantitate Pri-miR-4644 and mature

S. Chandrasekhar et al. / Bioorg. Med. Chem. Lett. 22 (2012) 645–648
647
The expression of EGFP is very much reduced in larval wing
discs, which indicate that miR-14 is expressed in the wing discs.
The EGFP transgene carrying fly that does not have the miRNA
binding sites is termed as control sensor and expresses EGFP in
all tissues. The schematic diagram of the entire gene construct is
shown in Figure 3a.
In fly, absence of miR-14 binding sequence (deleted) increased
expression of apoptotic effector caspase (drice).22–24 The miR-14
construct carrying miR-14 binding sites is expressed largely in
the larval wing discs (Fig. 3a). Endogenous miR-14 in wing discs
binds to the miR complementary sites and reduces the EGFP activ-
ity. As a result, level of EGFP expression is reduced strongly in the
wing discs. If oral feeding of aza-flavanones leads to subtle changes
in microRNA expression it is then reflected by the changes in EFGP
expression. The miR-14 and control sensor flies were fed with
50 lM of compounds 3 and 4 and emerging larval wing discs were
dissected. The EGFP expression was viewed after scanning the
discs under confocal microscope. The parameter of microscope
during scanning was always kept constant in control and miR-14
discs for reliable comparison. In miR-14 transgenic flies EGFP
was expressed at a high level in the wing discs when fed with
two compounds (3 and 4), while buffer fed (DMSO used as solvent
of the compound) larval discs have not shown any change in the
levels of EGFP expression (Fig. 3b). We isolated the total protein
from the wing discs and carried out western blotting using anti-
GFP antibody to quantify the level of EGFP expression. An upregu-
lation of EGFP protein was observed (Fig. 3c). Conversely a strong
reduction in miR-14 levels was found measuring miR-14 levels
by quantitative real time PCR studies. The miR-14 was down-reg-
ulated into five folds relative to buffer fed control Figure 4. The
expression of EGFP was high in the control sensor (which lacks
the miRNA binding sites in the 30 UTR) and remained the same in
both buffer and compound fed larval wing discs indicating that
the compounds effect the expression of the EGFP by altering the
levels of miR-14. Hence our data showed that aza-flavanones (3
and 4) down regulate miR-14 in vivo which is known to affect
the effector caspase, drice in Drosophila.22–24
Figure 3. (a) Schematic diagram of control sensor and miR-14 construct. EGFP
reporter gene is driven by tubulin promoter and fused with two miR-14 binding
sites at the 30end. (b) The EGFP expression shows the effect of compounds 3 and 4
The amount of Pri-miR-14 and mature miR-14 was estimated
by same Real time PCR assay using specific primers derived from
Pri-miR-14 and miR-14 sequences (Supplementary data). The Pri-
miR-14 and miR-14 levels were decreased dramatically after feed-
ing with compounds (3 and 4) relative to DMSO fed control flies
(Fig. 4). However the level of miR-3 remains same after feeding
on larval wing discs carrying control sensor and miR-14 constructs. (c)
A
quantitative Western blot of EGFP protein in the miR-14 Sensor flies fed with
control buffer and compounds (3 and 4) at 50 lM concentration were shown. b-
Actin acts as control.
miR-4644 after administrating compounds 3 and 4 in culture MCF-
7 and HeLa cells, a series of quantitative real time PCR was con-
ducted. The Pri-miR and mature miR levels were reduced close to
20–25% of the control DMSO level in MCF-7 and HeLa cells
(Fig. 2c). Later, functionally different miR-30A luciferase reporter
and miR-30A was introduced in HeLa cells. HeLa cells that express
both luciferase construct and miR-30A were also treated with com-
pounds 3 and 4 at a same concentration (4 lM). In this case, no
change in the intensity of the luciferase signal was detected
(Figs. 2c and S9). The results suggest that aza-flavanones are poten-
tially specific towards miR-4644 in human and their fly homologue
miR-14 and do not have a general effect on the common miRNA
pathway.
To understand whether aza-flavanones are directly involved in
Drosophila miR-14 regulation in vivo, two transgenic fly strains
were used that carry Enhanced Green Fluorescent Proteins (EGFP)
gene as reporter. The reporter gene is fused downstream of tubulin
promoter and GFP expressed in all tissues. Two copies of the target
sequences of miRNA-14 sites were placed in the 30 UTR. When miR-
14 is expressed, it will bind to the miRNA binding sites at the 30
UTR and the expression of EGFP reporter is reduced.
Figure 4. Level of mature miR-14 and Pri-miR-14 in compounds 3 and 4 Drosophila
larvae relative to buffer (including DMSO 0.1%) fed larvae control as determined by
quantitative Real time PCR. Bars represent S.D. determined by triplicate
measurements.

648
S. Chandrasekhar et al. / Bioorg. Med. Chem. Lett. 22 (2012) 645–648
the compounds. Therefore, two compounds (3 and 4) reduce only
miR-14 expression that control cell proliferation in fly.
mental procedures and Figs. S1–S11) associated with this article
can be found, in the online version, at doi:10.1016/
j.bmcl.2011.10.061.
We identified novel function of two aza-flavanone compounds;
in one path they act as potential cell cycle regulator, while in other
way they function as regulators of miR-4644 in human that control
cell proliferation and cell death in breast tumors.25 Using miR-14
homologues in human we found that compounds 3 and 4 inhibit
fly miR-14 and human counterpart miR-4644 suggesting their
common regulation during evolutionary changes. Since miRNAs
are multi-specific in nature and function, single miRNA can control
subsets of gene networks. The miRNA based therapy delivers an
appeal for targeting miR-14 dependent gene circuits in two differ-
ent species, therefore offers a broad range therapy potential
against cancer regulatory pathways in two animals. Moreover,
two compounds of aza-flavanones employ a common regulatory
strategy to control tumor progression, either by inhibiting tumor
progressing microRNA in two different animals or by cell cycle
control. It suggests that aza-flavanones deliver cell cycle arrest by
controlling tumor progressing microRNA paths.
In addition to in vitro human cell based screening, we have car-
ried out a novel cross species in vivo high throughput screening
method using transgenic model organism Drosophila that not only
identify the target of small molecules, but classify their in-depth
and elaborated biological functions in different physiological envi-
ronment. These studies would provide a powerful mute for under-
standing the exact role of aza-flavanones in biogenesis and
processing of microRNAs and their regulatory pathways during
animal evolution. In addition, these molecules potentially offer
the development of new and powerful miRNA based tools in multi-
ple organisms and can be highly applicable for the novel drug dis-
covery against cancer.
References and notes
1. He, L.; Hannon, G. J. Nat. Rev. Genet. 2004, 5, 522.
2. Kerscher, A. E.; Slack, F. J. Nat. Rev. Cancer 2006, 6, 259.
3. Lai, E. C. Genome Biol. 2004, 5, 115.1.
4. Ambros, V. Nature 2004, 431, 350.
5. Bushati, N.; Cohen, S. M. Annu. Rev. Cell Dev. Biol. 2007, 23, 175.
6. Forstermann, K.; Tomari, Y.; Du, T.; Vagin, V. V.; Denli, A. M.; Bratu, D. P.;
Klattenhoff, C.; Theurkauf, W. E.; Zamore, P. D. Plos Biol. 2005, 3, e236.
7. Stark, A.; Brennecke, J.; Russel, R. B.; Cohen, S. M. Plos Biol. 2003, 1, 397.
8. Brennecke, J.; Hipfner, D. R.; Stark, A.; Russel, R. B.; Cohen, S. M. Cell 2003, 113,
25.
9. Ambros, V. Cell 2001, 107, 823.
10. Bernstein, E.; Caudy, A. A.; Hammond, S. M.; Hannon, G. J. Nature 2001, 409,
363.
11. Gumireddy, K.; Young, D. D.; Xiong, X.; Hogenesch, J. B.; Huang, Q.; Deiters, A.
Angew. Chem., Int. Ed. 2008, 47, 7482.
12. Yan, L. X.; Huang, X. F.; Shao, Q.; Huang, M. Y.; Deng, L.; Wu, Q. L.; Zeng, Y. X.;
Shao, J. Y. RNA 2008, 14, 2348.
13. Young, D. D.; Connelly, C. M.; Grohmann, C.; Deiters, A. J. Am. Chem. Soc. 2010,
132, 7976.
14. Chandrasekhar, S.; Rao, Ch. L.; Seenaiah, M.; Naresh, P.; Jagadeesh, B.;
Manjeera, D.; Sarkar, A.; Bhadra, M. P. J. Org. Chem. 2009, 74, 401.
15. Stachel, S. J.; Lee, C. B.; Spassova, M.; Chappell, M. D.; Bornmann, W. G.;
Danishefsky, S. J.; Chou, T.-C.; Guan, Y. J. Org. Chem. 2004, 64, 4369.
16. Chandrasekhar, S.; Babu, G. S. K.; Reddy, Ch. R. Tetrahedron: Asymmetry 2009,
20, 2216.
17. Prakash, O.; Kumar, D.; Saini, R. K.; Singh, S. P. Synth. Commun. 1994, 24, 2167.
18. Singh, O. V.; Kapil, R. S. Synth. Commun. 1993, 23, 277.
19. Kalinin, V. N.; Shostakovsky, M. V.; Ponomaryov, A. B. Tetrahedron Lett. 1992,
33, 373.
20. Xia, Y.; Yang, Z. Y.; Xia, P.; Bastow, K. F.; Tachobana, Y.; Kuo, S. C.; Hamel, E.;
Hackl, T.; Lee, K. H. J. Med. Chem. 1998, 41, 1155.
21. Chandrasekhar, S.; Vijeender, K.; Sridhar, Ch. Tetrahedron Lett. 2007, 48, 4935.
22. Varghese, J.; Lim, S. F.; Cohen, S. M. Gene Dev. 2010, 24, 2748.
23. Brennecke, J.; Hipfner, D. R.; Stark, A.; Russell, R. B.; Cohen, S. M. Cell 2003, 113,
25.
Acknowledgments
24. Xu, P.; Vernooy, S. Y.; Guo, M.; Hay, B. A. Curr. Biol. 2003, 13, 790.
25. Persson, H.; Kvist, A.; Rego, N.; Staaf, J.; Vallon-Christersson, J.; Luts, L.; Loman,
N.; Jonsson, G.; Naya, H.; Hoglund, M.; Borg, A.; Rovira, C. Cancer Res. 2011, 71,
78.
S.C. thanks RSC, UK for the award of journals grant for interna-
tional authors, which helped to conceive this work and U.B. thank
CSIR, New Delhi for their generous funding (NWP-34). Thanks to
Dr. Stephen Cohen for transgenic fly stocks.
Supplementary data
Supplementary data (experimental procedure for the prepara-
tion of aza-flavones, their characterization data, biological experi-