Subscriber access provided by KEAN UNIV
Article
Truncated G-quadruplex Isomers Cross-talk with the Transcription Factors to maintain Homeostatic Equilibria in c-MYC Transcription Pallabi Sengupta, Apoorva Bhattacharya, Gaurisankar Sa, Tanya Das, and Subhrangsu Chatterjee Biochemistry, Just Accepted Manuscript • DOI: 10.1021/acs.biochem.9b00030 • Publication Date (Web): 28 Mar 2019 Downloaded from http://pubs.acs.org on March 29, 2019
Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.
Page 1 of 33 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Biochemistry
1
Truncated G-quadruplex Isomers Cross-talk with the Transcription Factors to maintain Homeostatic
2
Equilibria in c-MYC Transcription
3
Pallabi Sengupta1, Apoorva Bhattacharya2, Gaurisankar Sa2, Tanya Das2, Subhrangsu Chatterjee1*
4
1
5
India
6
2
7
West Bengal, India
8
* To whom correspondence should be addressed. Tel: 033-25693340; Email:
[email protected] 9
Abstract
Department of Biophysics, Bose Institute, P 1/12, C. I. T. Road, Scheme–VIIM, Kolkata–700054, West Bengal, Division of Molecular Medicine, Bose Institute, P 1/12, C. I. T. Road, Scheme – VIIM, Kolkata – 700054,
10
The Nuclease Hypersensitive Element III1 (NHE III1) upstream c-MYC promoter harbours a transcription-
11
silencing G-quadruplex (Pu27) element. Dynamic turnover of various transcription factors (TF) across Pu27 to
12
control c-MYC transcription homeostasis is enigmatic. Here, we revealed that native Pu27 evolves truncated G-
13
quadruplex isomers (Pu19, Pu22, Pu24, Pu25) in cells that are optimal intracellular targets of specific TFs in
14
sequence- and structure-dependent manner. NMR and Isothermal titration calorimetry envisaged that NM23-
15
H2 (Nucleoside diphosphate kinase) and Nucleolin induce conformational fluctuations in Pu27 to sample
16
specific conformationally restricted conformer(s). Structural investigations appended that the flanking guanines
17
at 5′-Pu27 control solvent exposure at G-quartets upon NM23-H2/Nucleolin binding driving Pu27 unfolding
18
and folding respectively. Transient chromatin immunoprecipitations confirmed that NM23-H2 drives the
19
conformation switch to Pu24 that outcompetes Nucleolin recruitment. Similarly, Nucleolin arrests Pu27 in Pu22
20
conformer minimizing NM23-H2 binding at Pu27. hnRNPK (Heterogeneous nuclear ribonucleoprotein K)
21
positively regulates NM23-H2 and Nucleolin association at Pu27 despite their antagonism. Based on these
22
results, we simulated the transcription kinetics in a feed-forward loop where the transcription output responds
23
to hnRNPK-induced early activation via NM23-H2 association, which favours Pu24 formation at NHE III1
24
reducing Nucleolin occupancy and driving quadruplex unfolding to initiate transcription. NM23-H2 further
25
promotes hnRNPK deposition across NHE III1 altering Pu27 plasticity that lately enriches Nucleolin abundance
26
to drive Pu22 formation and reduce NM23-H2 binding to extinguish transcription. This mechanism involves
27
three positive (NM23-H2-hnRNPK, NM23-H2-CNBP, hnRNPK-Nucleolin) and one negative (NM23-H2-
28
Nucleolin) feedback loops controlling optimal turnover and residence time of TFs at Pu27 to homeostatically
29
regulate c-MYC transcription.
30
Introduction
31
G-quadruplex nucleic acids sparked a new wave of interest as cancer-associated targets. They are
32
formed over tandem repeats of guanines with a stable core of π–π stacked G-quartets in a co-planar Hoogsteen
33
hydrogen-bonding arrangement1. These non-canonical structures of versatile topologies are ubiquitously
1 ACS Paragon Plus Environment
Biochemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 2 of 33
1
disseminated in the telomeres, proximal promoters, and untranslated regions of the proto-oncogenes2. In vivo
2
mapping of these conserved motifs by quadruplex-specific antibodies witnessed their abundance into
3
translocation hot-spots and primary tumors than normal tissues3, 4. G-quadruplexes play key functions in cellular
4
processes (e.g., transcription5, translation6, 7, mRNA stability8) and neoplastic transformation9, 10, allowing them
5
as the promising anti-cancer targets for clinical applications.
6
c-MYC is a major oncogenic driver in cancers11. Its overexpression through constitutive transcription
7
is strongly associated with autonomous proliferation, chromosomal translocation, relentless replication, and
8
impaired apoptosis in cancer cells,12, 13. MYC knockout experiments in in vivo studies revealed that partial
9
silencing of c-MYC results into acute or sustained tumor regression driving the cancer cells to undergo
10
proliferation arrest, apoptosis, differentiation, and cellular senescence12,
14
11
therapeutics holds a promising outcome for anti-neoplastic treatment. c-MYC oncogene harbours a Nuclease
12
Hypersensitive Element (NHE III1) located –142 to –115 base pairs upstream P1 promoter, which conserves G-
13
quadruplex (Pu27) and i-motif structures in offset orientation in the non-coding and coding strands
14
respectively15-17. These structures are evolved by negative supercoiling and torsional stresses due to localized
15
unfolding of double-helical DNA during transcription18. In normal proliferating cells, these tertiary DNA
16
scaffolds maintain an equilibrium between double-helical and tetra-stranded conformations to impede the
17
transcription and hold a threshold level of c-MYC transcripts19. In cancer cells, reciprocal translocation shifts c-
18
MYC-NHE III1 under the control of an ectopic promoter or leads to the removal of Pu27 resulting constitutive
19
transcription20-23. Introduction of synthetic 22-mer quadruplex significantly depleted c-MYC transcription in
20
Pu27-deleted cells with overexpressed c-MYC, which underscores the clinical importance of G-quadruplex to
21
restore its basal level in cancer cells 17, 23.
. Therefore, c-MYC-targeted
22
The major caveat of the study stems from the dynamic and polymorphic conformation of Pu27 limiting
23
its structure determination. The 27-mer sequence consists of six guanine tracts and exhibits disparity in the
24
length of intervening loops17, 24-26. This allows dynamic shuffling of multiple quadruplex loop isomers at c-MYC
25
promoter under cellular milieu rendering c-MYC quadruplex-targeting therapeutics by small compounds highly
26
challenging. They fail to conform to the polymorphic skeleton of Pu27 due to conformational rigidity and suffer
27
from promiscuity to other biomolecular quadruplex targets
28
issue, Pu27 is truncated into shorter stretches 16, 30-3233
29
27, 2810
resulting off-target effects. To resolve the
or both truncated and modified by base substitutions and/or
29
insertions
30
structure calculations. However, these truncated conformers are topologically divergent from wild-type element
31
and restrict the pharmacophore design for specific interactions with ambiguous wild-type Pu27, fulfilling
32
geometric and chemical restraints.
that reduce structural polymorphism and favour one particular topology necessary for
33
Many transcription factors are recruited at these secondary motifs to control the kinetic inertia of its
34
folding and unfolding in response to appropriate stimuli. While NM23-H2 (Nucleoside diphosphate kinase)
35
unfolds Pu27 quadruplex into transcriptionally active single-stranded form34, Nucleolin stabilizes Pu27 to
36
extinguish transcription35. Sp1-induced enhanced negative supercoiling at NHE III1 recruits hnRNPK 2 ACS Paragon Plus Environment
Page 3 of 33 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Biochemistry
1
(Heterogeneous nuclear ribonucleoprotein K) at i-motif (C-rich element), which eventually depletes Nucleolin
2
enrichment at Pu27 and upregulates c-MYC transcription36. CNBP (Cellular nucleic-acid binding protein) first
3
induces the transcriptionally inactive quadruplex at NHE III1, but later forms a transient complex with NM23-
4
H2 to activate transcription37. However, little is known about the dynamic assembly and disassembly of these
5
transcription factors at Pu27 motif that control the temporal expression of c-MYC transcripts. Whether these
6
transcription factors, each having its partially overlapping temporal windows of expression, exhibit mutually
7
exclusive binding to wild-type Pu27 is an enigma. How these transcription factors sharing overlapping DNA
8
binding specificities at NHE III1 stimulate or sequester either’s activity upon quadruplex binding and how the
9
oscillation of truncated quadruplex isomers at Pu27 regulates the binding turnover of these proteins at NHE III1
10
are less understood. It is likely that the cooperative binding of transcription factors depends upon the dynamic
11
character and flexible positioning of wild-type quadruplex, which functions as a buffer to absorb the negative
12
supercoiling stress into NHE III1 and coordinates the binding turnover and residence time of the transcription
13
factors to impart a homeostatic equilibria in c-MYC transcription.
14
Here, we identified that wild-type Pu27 presents an ensemble of four truncated quadruplex variants of
15
different topologies, rather than a single subset of conformer in the cells. These truncated conformers could be
16
targeted by quadruplex-specific small molecule probe and differentially regulate c-MYC transcription in cancer
17
cells. NM23-H2 and Nucleolin are the key transcription factors that induce considerable conformational changes
18
to target Pu27 upon binding and allow the polymorphic wild-type structure to conform to specific restricted
19
isomers at NHE III1 driving quadruplex unfolding and folding respectively. These transcription factors
20
synchronize a dynamic equilibrium between specific short-lived G-quadruplex conformers that are transiently
21
formed at the overlapping stretches of Pu27 as the native G-quadruplex unfolds into transcriptionally active
22
single-stranded form. This forms a recurring circuit at polymorphic Pu27 providing optimal kinetic advantage
23
for the binding turnover and residence time of transcription factors at NHE III1 involving both positive and
24
negative feedback loops to poise a homeostatic regulation in c-MYC transcription. Prevalence of the discrete
25
isomers perturbs the homeostasis resulting in aberrant c-MYC expression in cancer cells. This study provides a
26
detailed insight of protein-protein and protein-quadruplex crosstalk across c-MYC –NHE III1 that encourages
27
better pharmocophore design against the dynamic skeleton of Pu27 and development of specific probes to arrest
28
Pu27 in particular conformers to selectively modulate c-MYC transcription.
29 30
Materials and Methods
31
Oligonucleotide sequences: The native (Pu27) and truncated quadruplex–forming oligonucleotide sequences
32
(Pu22, Pu24, Pu25, and Pu19) (Table 1), residing at the Nuclease Hypersensitive Element III1 (NHE III1)
33
upstream c-MYC-P1 promoter are procured from Eurofins Genomics India Pvt. Ltd. The lyophilized pellets of
34
oligonucleotides are reconstituted into 10 mM Potassium phosphate buffer (10 mM K2HPO4 + 10 mM KH2PO4)
35
supplemented with 0.1 M Potassium Chloride (KCl) and 1 mM Ethylenediaminetetraacetic acid (EDTA) at pH
3 ACS Paragon Plus Environment
Biochemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 4 of 33
1
7.0. The sequences are annealed by heating at 950C for 5 minutes followed by gradual cooling to room
2
temperature to allow the formation of G-quadruplex structures in vitro.
3
Table 1: Oligonucleotide sequences used in biophysical experiments (Isothermal titration Calorimetry and
4
Nuclear Magnetic Resonance Spectroscopy): Oligonucleonucleotide sequences Pu27 (wild-type)
5′- T GGGG A GGG T GGGG A GGG T GGGG AA GG -3′
Pu19
5′ – T GGGG A GGG T GGGG A GGG T – 3′
Pu22
5′ – GGA GGG T GGGG A GGG T GGGG AA – 3′
Pu24
5′ – GGA GGG T GGGG A GGG T GGGG AA GG – 3′
Pu25
5′ – T GGGG A GGG T GGGG A GGG T GGGG AA – 3′
5 6
Cell Culture and Treatment: Human breast ductal carcinoma cell line (T47D) (NCCS, Pune), breast
7
adenocarcinoma cell lines (MCF-7 and MDAMB 231) (ATCC), and human cervical adenocarcinoma cell line
8
(HeLa) (ATCC) are cultured separately in complete Dulbecco’s modified Eagle’s medium (DMEM) (Himedia;
9
AL007G) respectively, supplemented with 10% (v/v) fetal bovine serum (FBS), 2 mM L-glutamine, 50 μg/mL
10
gentamycin, 1% Pen-Strep, and 2.5 μg/mL Amphotericin B in a fully humidified CO2 incubator (ESCO cell
11
culture CO2 Incubator, Model no. CCL-1708-8-UV) at 37°C and 5% CO2. Human gastric adenocarcinoma cell
12
line (AGS) is a generous gift from Dr. Debaprasad Mandal. These cells are grown in F–12K media having 10%
13
FBS, 2 mM L-glutamine, 50 μg/mL gentamycin, 1% Pen-Strep, and 2.5 μg/mL Amphotericin B in CO2
14
incubator at 37°C and 5% CO2.
15 16
Dual-luciferase assays: Human cancer cell lines, showing elevated c-MYC levels (T47D, MCF-7, HeLa) and
17
moderate c-MYC expression levels (MDAMB 231 and AGS) are taken up for dual-luciferase assay38, 39 to examine
18
the effect of native and truncated quadruplex structures in c-MYC promoter activation. Cells are subcultured into 24
19
well plates at a density of 2.5x104 cells/well. The reporter plasmids having the promoter constructs (with or without
20
native and truncated quadruplexes) are transformed into One Shot® Mach1™ T1 Competent E. coli cells (a kind
21
gift from Prof. Gautam Basu) to amplify the plasmids. Transformed cells are spread over LB (Luria broth) agar
22
plates having 150 μg/ml Ampicillin and incubated at 37 °C for 8 – 10 hours. Singular colony is picked from each
23
plate and further grown into LB media containing Ampicillin for 12–16 hours at 37 °C incubator and shaker (200
24
rpm). Plasmids are isolated using QIAGEN Plasmid Mini Kit (Catalog no. 27104) and 500 ng of pGL4.72[hRlucCP]
25
reporter plasmids are co-transfected with 50 ng pGL3-control vector (used as internal control) by Lipofectamine
26
2000 (Invitrogen) reagent as per manufacturer’s protocol. Cells are then incubated at 37 °C temperature and 5%
27
CO2 for 48 hours. After 24 hours of transfection, cells are treated with Chelerythrhine at an increasing concentration
28
gradient (50, 75, 100, 150, 250, 500 nM). Cell are taken out after 24 hours of treatment, washed with 1xPBS 4 ACS Paragon Plus Environment
Page 5 of 33 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Biochemistry
1
(Phosphate buffered saline) and scraped in ice using 1xPLB (Passive lysis buffer). Luciferase activities are
2
monitored by the dual-luciferase assay system (Promega; Catalog no. 0000219665) as per the manufacturer’s
3
protocol. Luminescence is detected in Thermo Scientific Varioskan Flash Spectral Scanning Multimode Reader.
4
Chromatin Immunoprecipitation (ChIP) assays: The chromatin immunoprecipitation studies are conducted
5
to monitor the promoter occupancy of the transcription factors (NM23-H2, Nucleolin, CNBP, and hnRNP K)
6
across the c-MYC-NHE III1(67). The transcription factors are transiently knocked down using siRNA-mediated
7
approach and occupancy of individual protein is examined under knocked down conditions. We have grown
8
T47D cells in 10-cm2 culture flask at a density of 1×106 cells per well followed by siRNA treatment for 48
9
hours. Cells are cross-linked with 1% formaldehyde at room temperature for 10 min and the reaction is stopped
10
by 0.125 M glycine at room temperature for 10 min. Fixed cells are lysed and sonicated to yield DNA fragments
11
of 200–500 bp. ChIP-grade antibodies (NM23-H2 (L-15) Antibody (sc-14790, Santa Cruz), and Anti-Nucleolin
12
antibody [4E2]-ChIP Grade (Abcam)) are added to the sonicated chromatin and incubated overnight at 4 0C.
13
Then, 10 μl protein A magnetic beads (Dynal, Invitrogen), previously washed in RIPA buffer are added to the
14
samples and bead–protein complexes are washed three times with RIPA buffer and twice with TE buffer. The
15
genomic DNA is eluted for 2 h at 680C in complete elution Buffer (20 mM Tris, pH 7.5, 5 mM EDTA, 50 mM
16
NaCl, 1% SDS and 50 μg/ml proteinase K) and further combined with the DNA eluted from second elution for
17
10 min at 680C in 100 μl of elution buffer (20 mM Tris, pH 7.5, 5 mM EDTA and 50 mM NaCl). ChIP-isolated
18
DNA is purified using MinElute Purification kit (Qiagen) and is amplified by PCR (Polymerase Chain Reaction)
19
reactions using forward and reverse primers(68) (Supplementary Table S10A) specific to the quadruplex-
20
enriched regions at c-MYC promoter and Phusion® High-Fidelity PCR Kit (NEB). Anti-rabbit IgG is employed
21
for mock immunoprecipitation. Quantification of the binding is performed by Real time PCR (Supplementary
22
section ESI) using fold enrichment method.
23
Fold enrichment: ∆∆Ct=Ct (target)-Ct (IgG),
24
Fold enrichment: 2^(-∆∆Ct )
25
Transient Chromatin Immunoprecipitation (t-ChIP) assays: T47D cells are grown overnight in 100-mm
26
dishes up to ∼60–70% confluency; cells are then transfected with 1 μg of the promoter construct using
27
Lipofectamine 2000 reagent. After 48 hours of incubation, cells are cross-linked with formaldehyde, harvested,
28
and chromatin immunoprecipitations are performed. The remainder of the procedure follows standard protocols
29
for ChIP analysis, as has been published previously40 and described on the University of California at Davis
30
Genome Center web site (genomecenter.ucdavis.edu/farnham/). The resulting DNA amplicons are analyzed by
31
qPCR reactions with a forward primer to a downstream portion of the luciferase construct that is common to all
32
constructs and reverse primers specific to the quadruplex-backbone (Supplementary Table S10B). Antibodies
33
used in the ChIP procedure include Anti-hnRNP K antibody - ChIP Grade (Abcam), Anti-Nucleolin antibody -
34
ChIP Grade (Abcam), Anti-CNBP polyclonal antibody (Abcam) and NM23-H2 (L-15) Antibody (from Santa
35
Cruz Biotechnology).
36
Isothermal titration calorimetry (ITC):
5 ACS Paragon Plus Environment
Biochemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 6 of 33
1
CHE and trimmed quadruplex-forming sequences: Thermodynamic attributes of the interaction profiles between
2
Chelerythrine, NM23-H2, Nucleolin and putative quadruplexes (Table 1) are monitored in iTC200
3
Microcalorimeter at 25°C
4
degassed under vacuum for 10 minutes ahead of the titrations to ensure the removal of bubbles, if any. 20 µM
5
Chelerythrine, prepared into the annealing buffer is contained into the calorimeter cell while the syringe is filled
6
with 500 µM quadruplex sequences. The sequences are injected into the cell containing Chelerythrine at an interval
7
of 150 s. A control experiment is performed in parallel by injecting the same concentration of oligonucleotides
8
(Pu27, Pu22, Pu24, Pu25, and Pu19) into identical buffer without Chelerythrine to subtract the heat of dilution from
9
Chelerythrine-quadruplex binding experiments before curve-fitting. The number of injections are set at 20 to
10
achieve the binding saturation. The heat of reaction per injection (µcal/s) is determined by integration of the peak
11
areas using in-built Origin 7.0 software, which provides the best-fit values of the enthalpy of binding (∆𝐻), the
12
stoichiometry of binding (n), and the dissociation constant (Kd). The data points are further simulated with “one –
13
site” and/or “sequential” binding models. The quality of fitting curve is inspected by the reduced Chi-squared values
14
(𝜒𝑅2 ). Given the total concentration of Chelerythrine, L inside the calorimetric cell is known, after each consecutive
15
injection i:
16
41
. Oligonucleotide sequences and Chelerythrine solution are freshly prepared and
𝑣
[𝑀] 𝑇,𝑖 = [𝑀]0 (1 − 𝑉)𝑖 ……….. Eq. (1) 𝑣
17
[𝐿] 𝑇,𝑖 =
𝑖
[𝐿]0 {1 − (1 − )} 𝑉
...... Eq. (2)
18
Where [M]0 is the initial concentration of quadruplex in the syringe and [L]0 is the concentration of Chelerythrine,
19
L in the cell. V signifies the cell volume and v is the injection volume. (1 − v/V) is the factor that accounts for the
20
change in the concentration of reactants due to dilution upon sequential titrations. Therefore, using the mass action
21
law and the conservation of mass for each species:
22
[𝑀] 𝑇 = [𝑀] + [𝑀𝐿] = [𝑀] + 𝐾𝑎,𝐿 [𝑀][𝐿] ……….. Eq. (3)
23
[𝐿] 𝑇 = [𝐿] + [𝑀𝐿] = [𝐿] + 𝐾𝑎,𝐿 [𝑀][𝐿] …………..Eq. (4)
24
Where 𝐾𝑎,𝐿 is the binding constant for Chelerythrine for the quadruplexes. Solving the series of equations yields the
25
concentration of complexes, [ML] in the calorimetric cell after each injection i. The heat released or absorbed due
26
to each injection, 𝒒𝒊 , is the heat associated with the formation/dissociation of each complex in the injection i: 𝑣 𝑉
27
𝒒𝒊 = 𝑉[∆𝐻𝑎,𝐿 {[𝑀𝐿]𝑖 − [𝑀𝐿]𝑖−1 (1 − )}]…… Eq. (5)
28
∆𝐺 = 𝑅𝑇𝑙𝑛(𝐾𝑎 )……… Eq. (6)
29
∆𝐺 = ∆𝐻 − 𝑇∆𝑆………Eq. (7) 6 ACS Paragon Plus Environment
Page 7 of 33 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Biochemistry
1
Here, ∆𝐻𝑎,𝐿 is the binding or association enthalpy for each ligand. T is temperature at which the experiment is
2
performed. ∆𝐺 is the binding free energy and ∆𝑆 is the entropic contribution in each binding event.
3
1D and 2D NMR (Nuclear Magnetic Resonance) studies:
4
1D 1H NMR spectroscopy: NMR experiments are performed in Bruker AVANCE III 500 MHz NMR spectrometer,
5
equipped with a 5 mm SMART probe. The wild-type (Pu27) and truncated quadruplex-forming oligonucleotide
6
sequences (Pu25, Pu24, Pu22, and Pu19) are prepared in 350 µM buffer containing 90% water and 10% D2O. 1-D
7
experiments are carried out in 5 mm NMR tubes having an active sample volume of 600 µL. The spectra are
8
referenced to an internal standard, TSP (3-(trimethylsilyl)-2, 2′, 3, 3′-tetradeuteropropionic acid) at 0.0 ppm. The
9
imino protons of putative quadruplex-forming sequences are observed in the one-dimensional proton spectra (1014
10
12 ppm) using Bruker Pulprog ‘zgesgp’ with a spectral width
of 20 ppm, number of scans (ns) of 512, and
11
calibrated pulse length (p1) of 12.48 µs.
12
NMR titrations: NMR titrations are carried out by adding the aliquots of purified proteins (NM23-H2 and Nucleolin)
13
into 350 μM of Pu27 dissolved into 1x PBS (Phosphate buffered saline) containing 90% water and 10% D2O. NMR
14
samples are then mixed to homogeneity and allowed to reach thermal equilibria. Proton spectra are acquired at each
15
point of titration at 25°C.
16
Mathematical modelling of the transcription kinetics: To elucidate the impact of the parameter values on the
17
homeostatic mechanisms in quadruplex-driven c-MYC transcription, we proposed an intuitive mathematical
18
model based on the mass-action kinetics42 (Supplementary Table S11). We undertook prior assumptions for
19
further simplification of the model – (i) c-MYC transcription output is mainly controlled by the quadruplex
20
scaffold(s) disregarding other factors involved in the transcription. (ii) Transcription factors, recruited at the
21
quadruplex scaffold are synthesized at a constant rate and are transiently degraded by siRNA treatment. Inputs
22
for the model denote the transcription factors’ concentrations at NHE III1 while c-MYC mRNA expression
23
denotes the output and addresses steady-state kinetics between quadruplex and single-stranded forms, regulated
24
by the competitive interaction between NM23-H2 and Nucleolin and the positive feedback loops (NM23-H2–
25
hnRNP K, NM23-H2–CNBP, and hnRNP K–Nucleolin). This hierarchical clustering of c-MYC
26
quadruplex/protein kinetic network is modelled by nonlinear ordinary differential equations (ODEs).
27
To study the kinetic behavior of different components of the network, the set of nonlinear ODEs are solved by
28
XPP (http://www.math.pitt.edu/~bard/xpp/xpp.html) using the parameter sets given in the downstream
29
equations. The parameters are manually tuned to generate the temporal experimental profile. The parameter sets
30
for numerical integration of nonlinear ODEs are estimated using Parameter Estimation Toolkit
31
(http://mpf.biol.vt.edu/pet/).
32
First, we performed time-dependent siRNA knockdown of NM23-H2, Nucleolin, and hnRNP K and determined
33
their mRNA expression level at 0, 12, 24, 48, 72, and 96 hours of siRNA treatment by quantitative PCR. We
34
considered these data points to mathematically simulate the fluctuations of their mRNA expression level by 7 ACS Paragon Plus Environment
43
Biochemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 8 of 33
1
differential equation. siRNA mediated transfection leads to a transient knockdown of these transcription factors
2
after which the expression level of the proteins rise and approach to that of the control:
3 4
I.
siRNA mediated degradation of NM23-H2: 𝑘1
5
siRNA degradation: 𝑠𝑖𝑅𝑁𝐴𝑁𝑀 → 𝜙
6
Rate of siRNA degradation over time:
7
siRNA mediated degradation of NM23-H2: [𝑁𝑀] + 𝑠𝑖𝑅𝑁𝐴𝑁𝑀 → 𝜙
8
Rate of change in NM23-H2 concentration over time:
9
𝑑[𝑠𝑖𝑅𝑁𝐴]𝑁𝑀 𝑑𝑡
= −𝑘1 [𝑠𝑖𝑅𝑁𝐴] 𝑘5′
II.
𝑑[𝑁𝑀] 𝑑𝑡
= 𝑘5 − 𝑘5′ [𝑁𝑀] − 𝑘1 [𝑠𝑖𝑅𝑁𝐴]𝑁𝑀
siRNA mediated degradation of Nucleolin: 𝑘2
10
siRNA degradation: 𝑠𝑖𝑅𝑁𝐴 𝑁𝑢 → 𝜙
11
Rate of siRNA degradation over time:
12
siRNA mediated degradation of Nucleolin: [𝑁𝑢] + 𝑠𝑖𝑅𝑁𝐴 𝑁𝑢 → 𝜙
13
Rate of change in Nucleolin concentration over time:
𝑑[𝑠𝑖𝑅𝑁𝐴]𝑁𝑢 𝑑𝑡
= −𝑘2 [𝑠𝑖𝑅𝑁𝐴] 𝑘4′
𝑑[𝑁𝑢] 𝑑𝑡
= 𝑘4 − 𝑘4′ [𝑁𝑢] − 𝑘2 [𝑠𝑖𝑅𝑁𝐴]𝑁𝑢
14 15
III.
siRNA mediated degradation of hnRNP K: 𝑘3
16
siRNA degradation: 𝑠𝑖𝑅𝑁𝐴ℎ𝑛 → 𝜙
17
Rate of siRNA degradation over time:
18
siRNA mediated degradation of Nucleolin: [ℎ𝑛] + 𝑠𝑖𝑅𝑁𝐴ℎ𝑛 → 𝜙
19
Rate of change in Nucleolin concentration over time:
𝑑[𝑠𝑖𝑅𝑁𝐴]ℎ𝑛 𝑑𝑡
= −𝑘3 [𝑠𝑖𝑅𝑁𝐴] 𝑘6′
𝑑[ℎ𝑛] 𝑑𝑡
= 𝑘6 − 𝑘6′ [ℎ𝑛] − 𝑘3 [𝑠𝑖𝑅𝑁𝐴]ℎ𝑛
20
Then, hRluc expression level was determined using the same qPCR that demonstrates c-MYC promoter activity
21
at similar time windows. We generated mathematically simulated curves for the following based on the
22
experimental data points and coupling of the following differential equations:
23
IV.
Quadruplex-mediated Transcription regulation at c-MYC promoter
24
[𝑁𝑀. 𝑄] + [𝑄] 𝑇 + [𝑁𝑢] ⇌ [𝑁𝑢. 𝑄] + [𝑄](𝑇−𝑁𝑢.𝑄) + [𝑁𝑀]
25
[𝑁𝑢. 𝑄] + [𝑄](𝑇−𝑁𝑢.𝑄) + [𝑁𝑀] ⇌ [𝑁𝑀. 𝑄] + [𝑄](𝑇−𝑁𝑀.𝑄) + [𝑁𝑢]
26
[ℎ𝑛] + [𝑄] 𝑇 + [𝑁𝑀] ⇋ [ℎ𝑛. 𝑄] + [𝑁𝑀. 𝑄] + [𝑄](𝑇−𝑁𝑀.𝑄)
27
[𝑁𝑀. 𝑄] + [ℎ𝑛. 𝑄] ⇌ [𝑁𝑀. 𝑠𝑠] + [𝑠𝑠] 𝑇 + [𝑁𝑀] + [ℎ𝑛]
28
[𝑠𝑠] 𝑇 → 𝑅
29 30
𝑑[𝑁𝑀. 𝑄] = 𝑘8 [𝑁𝑀][𝑄](𝑇−𝑁𝑢.𝑄) + 𝑘9 [ℎ𝑛][𝑄] 𝑇 − 𝑘8′ [𝑁𝑀. 𝑄] − 𝑘8′ [𝑁𝑢] − 𝑘8′ [𝑄](𝑇−𝑁𝑀.𝑄) 𝑑𝑡 𝑑[𝑁𝑢. 𝑄] = 𝑘7 [𝑁𝑢][𝑄] 𝑇 − 𝑘7′ [𝑁𝑢. 𝑄] − 𝑘9′ [𝑁𝑀] − 𝑘7′ [𝑄](𝑇−𝑁𝑢.𝑄) 𝑑𝑡
8 ACS Paragon Plus Environment
Page 9 of 33 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
𝑑[ℎ𝑛. 𝑄] = 𝑘9 [ℎ𝑛][𝑄] 𝑇 + 𝑘8 [𝑁𝑀][𝑄] 𝑇 − 𝑘9′ [𝑁𝑀. 𝑄] − 𝑘9′ [ℎ𝑛. 𝑄] 𝑑𝑡 𝑑[𝑠𝑠] = 𝑘8 [𝑁𝑀. 𝑄] − 𝑘10′ [𝑁𝑀. 𝑠𝑠] 𝑑𝑡 𝑑[𝑚𝑅𝑁𝐴] = 𝑘10 [𝑠𝑠] 𝑇 𝑑𝑡
1 2 3 4
Biochemistry
Results
5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
Figure 1. Identification of the biologically relevant intracellular conformers of native c-MYC quadruplex (Pu27). (A) Diagram of the pGL4.72[hRlucCP] vector (Linear) having the insert containing c-MYC promoter sequences (P1 and P2) and upstream Nuclease Hypersensitive Elements (NHE III1) ahead of the hRluc coding region. hRluc, Renilla luciferase gene; hCL1 and hPEST, protein destabilizing sequences; oriC, origin of replication; Amp R, ampicillin resistance gene; SV40 (Simian virus 40 polyadenylation signal cassette). c-MYC promoter sequences are cloned into KpnI and HindIII restriction sites with or without the wild-type (Pu27) and truncated quadruplex elements (Pu25, Pu24, Pu22, and Pu19) into NHE III 1. Pu27 contains six guanine tracts (I-VI) while in Pu24 and Pu25, G-tract-I and VI are removed respectively. Pu22 lacks G-tract I and VI while G-tracts V and VI are absent in Pu19. (B) c-MYC promoter activity using reporter plasmids with or without wild-type (Pu27) and truncated quadruplex-forming sequences (Pu25, Pu24, Pu22, and Pu19) in different cancer cell lines (MCF-7, T47D, MDAMB 231, HeLa, and AGS). Relative promoter activities of GQ-null and G-quadruplex-harbouring constructs are determined by the Rluc/Fluc values. Error bars represent mean ± SE (N = 3). Statistical differences compared to that of wild-type Pu27C construct in the luciferase activities used one-way ANOVA followed by Dunnett’s Test (*P < 0.05, **P < 0.01, ***P < 0.001). (C) Mapping of S1 nuclease-sensitive sites in pGL4.72[hRlucCP] plasmids containing c-MYC promoter inserts (Pu27C, Pu25C, Pu24C, Pu22C, and Pu19C). Lane 1, DNA size marker (1 kb); lane 2, cleavage with S1 nuclease; lane 3, digestion with S1 nuclease followed by BamHI; lane 4 digestion with BamHI; lane 5, digestion with BamH1 followed by S1 nuclease; lane 6, digestion with HindIII; lane 7, digestion with HindIII and BamHI.
9 ACS Paragon Plus Environment
Biochemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 10 of 33
1
Multiple loop isomers of Pu27 trigger differential c-MYC promoter regulation in cancer cells. The wild-
2
type Pu27 structure located in c-MYC-NHE III1 shows conformational heterogeneity26. Pruning and
3
modifications in Pu27 sequence evolve selective quadruplex species in vitro16, 30, 31. Despite having structural
4
evidences of truncated isomers in Pu27, their formation and role in c-MYC transcription under cellular milieu
5
remained questions. To examine the intracellular formation of quadruplex isomers in c-MYC-NHE III1, we
6
truncated the parent sequence from 5ꞌ- and/or 3ꞌ-termini to generate the trG-Q sequences (e.g., Pu19, Pu22,
7
Pu24, and Pu25). Then, we investigated their role in c-MYC-P1 promoter regulation by dual-luciferase assays in
8
multiple cancer cell lines (MCF-7, T47D, MDAMB 231, HeLa, and AGS cells) expressing moderate to high c-
9
MYC transcripts. We used c-MYC promoter constructs with or without the native and trG-Q sequences in the
10
upstream of hRluc gene and monitored the magnitude of c-MYC promoter activities (Figure 1(A)) in contrast
11
to that observed in Pu27C. We found a constitutive activation of c-MYC promoter in the absence of quadruplex
12
scaffold at NHE III1 (GQ-null construct) while the wild-type and putative trG-Q stretches significantly inhibited
13
c-MYC promoter activation in cancer cells (Figure 1(B)) compared to that of GQ-null. The significant find of
14
this study was the differential regulation of c-MYC promoter by truncated quadruplex conformers in different
15
cancer cells as compared to native Pu27. Pu25 insert, which lacks the sixth G-tract
16
alteration in the promoter activity compared to Pu27C in MCF-7, T47D, AGS, and HeLa cells (Figure 1(B),
17
and Supplementary Table S1). However, in MDAMB 231, a sharp rise of promoter activation (P value < 0.01)
18
was observed for Pu25C (Figure 1(B)). Pu24 insert having a truncated 5ꞌ-G-tract (I) elevated c-MYC promoter
19
activation (~1.5-3 folds) in all the cell lines compared to that of Pu27C (P value
