MicroRNA-214 suppresses growth, migration
and invasion through a novel target, high
mobility group AT-hook 1, in human cervical
and colorectal cancer cells
Karthik Subramanian Chandrasekaran
1
, Anusha Sathyanarayanan
1
and Devarajan Karunagaran
*
,1
1
Department of Biotechnology, Bhupat and Jyoti Mehta School of Biosciences, Indian Institute of Technology Madras, Chennai,
Tamil Nadu 600036, India
Background: MicroRNA-214 (miR-214) has been shown to act as a tumour suppressor in human cervical and colorectal cancer
cells. The aim of this study was to experimentally validate high mobility group AT-hook 1 as a novel target for miR-214-mediated
suppression of growth and motility.
Methods: HMGA1 and miR-214 expression levels were estimated in cervical and colorectal clinical specimens using qPCR.
HMGA1 3
0
untranslated region luciferase assays were performed to validate HMGA1 as a target of miR-214. Effect of altering the
expression of miR-214 or HMGA1 on proliferation, migration and invasion of human cervical and colorectal cancer cells was
investigated.
Results: miR-214 expression was poor while that of HMGA1 was high in cervical and colorectal cancer tissues. miR-214-re-
expression or HMGA1 downregulation inhibited proliferation, migration and invasion of cancer cells while miR-214 inhibition had
opposite effects. miR-214 was demonstrated to bind to the wild-type 3
0
untranslated region of HMGA1 but not with its mutant.
Conclusions: Low expression of miR-214 concurrent with elevated levels of HMGA1 may contribute to cervical and colorectal
cancer progression. miR-214-mediated regulation of HMGA1 is a novel mechanism for its tumour-suppressive actions in human
cervical and colorectal cancer cells and opens up avenues for novel therapeutic strategies for these two cancers.
Cervical cancer (CaCx) is common among women predominantly
in the developing countries (Hawes and Kiviat, 2002), and in India,
it is the most common cancer affecting women (Sankaranarayanan
et al, 1996). Human papilloma virus (HPV) is the major risk factor
for cervical cancer (Bouallaga et al, 2000), but independently,
alterations in tumour-suppressor genes and/or oncogenes may also
be necessary for cervical cancer progression (Zur Hausen, 1996).
Although colorectal cancer (CRC) earlier had the lowest rates of
incidence in Asian countries (Haggar and Boushey, 2009), in the
past few years, Asia has witnessed a rapid increase in CRC
incidence particularly in the developing countries, including India
(Mohandas, 2011; Moghimi-Dehkordi, 2012). MicroRNAs are
B22 nucleotide long, small RNA molecules that modulate gene
expression transcriptionally or posttranscriptionally by mainly
binding to the 3
0
untranslated region (UTR) of their target genes
(Calin et al, 2002). miRNAs constitute unique expression
signatures and modulate cellular signalling pathways in many
cancers, including CaCx (Hu et al, 2010; Pereira et al, 2010) and
CRC (Chen et al, 2015; Hur et al, 2015). miR-214 functions as a
tumour suppressor by downregulating oncogenes, such as
GALNT7, Bcl2l2 and TFAM in CaCx (Yang et al, 2009; Peng
et al, 2012; Wang et al, 2013; Wen et al, 2014) and FGF-1 and
ARL2 in CRC (Chen et al, 2014; Long et al, 2015), but the actual
underlying mechanisms are yet to be elucidated. The high mobility
group AT-hook (HMGA) proteins are non-histone chromo-
somal proteins involved in DNA transcription, replication,
*Correspondence: Dr D Karunagaran; E-mail: karuna@iitm.ac.in
Received 25 March 2016; revised 9 June 2016; accepted 12 July 2016; published online 18 August 2016
& 2016 Cancer Research UK. All rights reserved 0007 – 0920/16
FULL PAPER
Keywords: microRNA-214; tumour suppressor; HMGA1; human cervical cancer; human colorectal
cancer; proliferation; migration; invasion
British Journal of Cancer (2016) 115, 741–751 | doi: 10.1038/bjc.2016.234
www.bjcancer.com | DOI:10.1038/bjc.2016.234 741
recombination and repair (Cleynen and Van de Ven, 2008) and, as
recently shown, in the regulation of metabolism (Qiu et al, 2014,
2015). HMGA proteins are easily detected in embryonic, neoplastic
and proliferating undifferentiated cells but not so in non-neoplastic
human adult tissues (Resar, 2010). They were first discovered in
CaCx cells (Lund et al, 1983) and subsequently shown to have an
oncogenic role during CaCx initiation, progression and metastasis
by cooperating with HPV18 E6/E7 oncoproteins and inactivating
p53 (Bandiera et al, 1998; Mellone et al, 2008). Its enhanced
expression correlates with tumorigenesis and metastasis in human
CRC (Fedele et al, 1996; Huang et al, 2009). Importantly, cis-
regulatory elements in the 3
0
UTR mediate posttranscriptional
regulation of HMGA1 (Borrmann et al, 2001). Indeed, in
leukaemia, bladder and prostate cancers, HMGA1 expression is
modulated by different miRNAs targeting its 3
0
UTR (Kaddar et al,
2009; Wei et al, 2011; Lin et al, 2013). In the present study,
miR-214 is demonstrated to directly target wild-type 3
0
UTR of
HMGA1 and reduce its endogenous expression in CaCx and CRC
cells. Downregulation of HMGA1 expression by miR-214 or
siRNA-HMGA1 significantly inhibits proliferation, migration and
invasion of CaCx and CRC cells. Thus novel mechanistic basis for
the tumour-suppressive actions of miR-214 is revealed unraveling
new therapeutic opportunities.
MATERIALS AND METHODS
Human CaCx and CRC tissue samples. Fresh CaCx tissue samples
were collected from consenting patients undergoing treatment for
CaCx at the Institute of Obstetrics and Gynaecology, Chennai, India.
Normal fresh cervical tissues were obtained from patients undergoing
hysterectomy for various non-malignant reasons. Fresh CRC tissue
samples and adjacent normal tissues were collected from consenting
patients undergoing treatment for CRC at the Apollo Hospitals,
Chennai, India. The study was approved by the Institutional Ethics
Committee of Indian Institute of Technology Madras.
Cell lines. Human cervical cancer cell lines, SiHa, CaSki and C33A,
and colorectal cancer cell lines, SW480 and SW620, were maintained
in Dulbecco’s Modified Eagle’s Medium (DMEM) (Life Technologies,
Carlsbad, CA, USA) containing 10% FBS (Life Technologies) and
antibiotics (100 U ml
1
penicillin and 100 mgml
1
streptomycin) in
a humidified atmosphere of 5% CO
2
at 37 1C.
RNA and plasmid transfection. miRNA-control (no. CN-
001000-01-20), miR-214 mimic (no. C-301153-01), siRNA-control
(no. D-001220-01-20), siRNA-HMGA1 (no. M-004597-02), anti-
miR-control (no. IN-002005-01-20 or antimiR-214 (no. IH-
301153-02-0005) were obtained from GE Healthcare Dharmacon
(Lafayette, CO, USA). Transient transfections of the above (with
5n
M of miRNA and antimiR mimics and 50 nM of siRNA mimics)
into CaCx and CRC cells were achieved using Lipofectamine
RNAiMAX (no. 13778150, Life Technologies) while pcDNA 3.1,
pcDNA 3.1-miR-214, pIRES (vector control), pIRES-HMGA1
(kind gift from Edward Whang, Addgene plasmid no. 13466,
Addgene, Cambridge, MA, USA), 3
0
UTR reporter plasmid
constructs were transfected using linear polyethyleneimine (no.
23966-2, MW 25 000, procured from Polysciences, Warrington,
PA, USA) at a ratio of 5 : 1 to DNA. Combinations of miR-
control+pIRES, miR-214+pIRES, miR-control+pIRES-HMGA1,
miR-214+pIRES-HMGA1 were transfected using DharmaFECT
Duo (no. T-2010-01, GE Healthcare Dharmacon).
Western blotting. Total cell lysates were prepared by incubating
cells in RIPA lysis buffer (150 m
M NaCl, 1% NP-40, 0.5%
deoxycholate and 1% SDS) on ice for 1 h, and protein concentra-
tion was quantified by Bradford’s method according to the
manufacturer’s protocol (Bio-Rad, Hercules, CA, USA). Samples
(50 mg protein) were resolved on 10% SDS-PAGE and transferred
to a PVDF membrane (Immunoblot, Bio-Rad) using a Bio-Rad
Mini PROTEAN III apparatus. Anti-HMGA1 (no. 7777S) and
anti-ACTB (no. A5441) antibodies were purchased from Cell
Signaling (Danvers, MA, USA) and Sigma-Aldrich (St Louis, MO,
USA), USA, respectively, while anti-mouse IgG-peroxidase-con-
jugate (no. 115-035-003) and anti-mouse IgG-peroxidase-conju-
gate (no. 111-035-003) secondary antibodies were bought from
Jackson Laboratories (West Grove, PA, USA). Bands detected
using the Enhanced Chemiluminescence Kit (Bio-Rad) were
visualised using ChemiDoc (Bio-Rad) and analysed by densito-
metry (Image Lab, Bio-Rad). ACTB was used as an internal
control. Experiments were repeated at least once to confirm the
results obtained earlier.
RNA isolation and real-time quantitative PCR. Tissue samples
were ground and RNA was extracted using the manufacturer’s
protocol (TRIzol, Life Technologies) and RNA was also isolated
from cells using TRIzol (Life Technologies). Mature miRNA levels
were estimated by performing stem-loop reverse transcription
followed by quantitative PCR; reverse transcription by MMLV
reverse transcriptase (Life Technologies) was performed using
miR-214-specific and RNU6-specific stem-loop primers. PCR
amplification of miR-214 or RNU6 was performed using a forward
primer specific for miR-214 or RNU6 (internal control) and a
universal reverse primer. For estimating HMGA1 mRNA levels,
reverse transcription was carried out by MMLV reverse tran-
scriptase (Life Technologies) using oligo-dT and amplified using
appropriate gene-speci fic PCR primers. Detection and quantitation
of HMGA1 or ACTB (internal control) was carried out using the
DyNAmo ColorFlash SYBR Green qPCR Kit reagent (no. F416L,
Thermo Scientific, Waltham, MA, USA) on Eppendorf realplex4
Mastercycler epgradient S (Eppendorf, Hamburg, Germany).
Relative expression levels of genes analysed were calculated using
2
DCT
(tissue samples) or 2
DDCT
(cancer cells) method.
3
0
UTR luciferase assays. HMGA1 3
0
UTR that contains putative
binding sites for the miR-214 was amplified from human genomic
DNA using Phusion high-fidelity DNA polymerase (New England
Biolabs, Ipswich, MA, USA) and cloned into the 3
0
UTR of Renilla
luciferase gene in the psiCHECK-2 reporter vector (Promega,
Madison, WI, USA). The miR-214-binding site was mutated by
substituting five out of the six bases in the miRNA-binding
sequence (seed sequence) in the 3
0
UTR of HMGA1 using
appropriate primers and the mutant construct thus synthesised
was used as a negative control. CaCx or CRC cells were co-
transfected with pcDNA3.1 or pcDNA3.1-miR-214 and the wild-
type or mutant 3
0
UTR luciferase constructs in a 24-well format,
and 24 h posttransfection, cells were lysed using Passive Lysis
Buffer, and Renilla luciferase activity was measured using the Dual
Luciferase Assay Kit (no. A2492, Promega) and a luminescence
plate reader (Molecular Devices Inc., Sunnyvale, CA, USA),
wherein firefly luciferase acted as the internal control.
Migration and invasion assays. Migration assays were performed
by transfecting CaCx or CRC cells with miR-214 or siRNA-
HMGA1 or antimiR-214 or respective controls and then seeding
5 10
4
cells in DMEM onto the upper part of each Transwell
chamber (BD Biosciences, Franklin Lakes, NJ, USA) and adding
10% FBS containing DMEM to the lower part of the chamber. Cells
adhering to the bottom of the Transwell membrane were stained
with 0.1% crystal violet 48 h later and images were obtained using
an Olympus TL4 inverted light microscope (Shinjuku, Tokyo,
Japan). In addition, stain was collected from stained cells by
washing with 10% acetic acid and quantified by measuring
absorbance at 595 nm (Saito et al, 1997) using a Bio-Rad Model
680 microplate reader (Bio-Rad, Shinagawa-ku, Tokyo, Japan).
Invasion assays were performed in a similar manner but by
BRITISH JOURNAL OF CANCER Mir-214 regulates HMGA1 in cervical and colorectal cancers
742 www.bjcancer.com | DOI:10.1038/bjc.2016.234
allowing the cells to migrate through a GelTrex-coated (no.
A15696-01, Life Technologies) layer in the upper part of a
Transwell chamber.
BrdU incorporation assays. Proliferation assays were performed by
transfecting 0.7 10
4
cells with miR-214 or antimiR-214 or siRNA-
HMGA1 or respective controls and, 48 h posttransfection, were
examined using the BrdU Cell Proliferation Assay Kit (no. 6813S, Cell
Signaling) according to the manufacturer’s protocol.
TCGA data analysis. TCGA open access data directory (http://
cancergenome.nih.gov/) was used to obtain miRNA and mRNA
expression data sets for human CaCx and CRC tumours. Normalised
TCGA level 3 miRNA-seq and RNA-seqV2 data were compiled using
R studio and used for assessing the expression of miRNA and mRNA.
These linear sequencing data expression values were then used to
compute Pearson product–moment correlation coefficient.
Statistical analysis. qPCR was performed in duplicates for all the
clinical tissue specimens and mean expression was calculated within
normal or cancer tissue subsets. As paired normal specimens were
available only in the case of colorectal tumour, miR-214 or HMGA1
expression in a tumour was normalised to its paired normal
counterpart. One-way ANOVA test was performed to evaluate fold
expression relative to mean. In the case of cervical clinical
specimens, mean values for each data set (tumour or normal) were
calculated. In the case of miR-control-, miR-214-, siR-control-, si-
HMGA1-, antimiR-control- or antimiR-214-transfected cells, three
independent experiments were conducted for RNA quantitation
(in vitro), BrdU incorporation, migration and invasion assays, and
after appropriate normalisation, s.e.m. was plotted. After performing
unpaired t-tests, P-values were calculated and represented as
*Pp0.05, **Pp0.01, ***Pp0.001 or ****Pp0.0001.
RESULTS
Expression of miR-214 is lower while that of HMGA1 is higher
in both human cervical and colorectal tumours than in their
corresponding normal tissues. miR-214 is a known tumour-
suppressor miRNA in cervical (CaCx) and colorectal (CRC)
cancers and acts by downregulating a few oncogenes as is the case
with many tumour-suppressor miRNAs. A putative, conserved, 6-
mer, miR-214-binding sequence located at 308–314 bases down-
stream in the HMGA1 3
0
UTR was identified using TargetScan.
HMGA1 was of particular interest among the target genes
identified because of its positive roles in cancer cell proliferation
and invasion in CaCx and CRC cells. Prompted by the inverse
correlation in the expression of this miRNA–target pair deduced
from previous reports that studied miR-214 or HMGA1 expression
separately (Fedele et al, 1996; Bandiera et al, 1998; Yang et al, 2009;
Chen et al, 2014), these two were analysed together by TCGA data
analysis or quantitative PCR in patient samples to strengthen the
possibility of the existence of a regulatory mechanism. TCGA
analysis showed a clear inverse correlation between miR-214 and
HMGA1 both in CaCx and CRC (Figures 1A and B). Pearson
correlation coefficients of 0.239 (P ¼ 0.0019) and 0.274
(P ¼ 0.0005) were obtained for CaCx and CRC tumour samples,
respectively. When a few samples of CaCx and CRC were analysed
to check if this correlation existed in a local population, the same
trend was found. Expression levels of mature miR-214 analysed in
various samples fell into a lower range in CaCx tissues compared
with normal tissues (Figure 1C), with mean values of 20.26 and
66.13, respectively. HMGA1 mRNA levels were found in a
relatively higher, narrow range (Figure 1D) in CaCx relative to
normal tissues with mean values of 22.6 and 8.2, respectively.
Similarly, miR-214 was found to be poorly expressed (mean
0.68
±
0.3), whereas HMGA1 expression was higher (mean
11.52
±
5.4) in CRC tissues than in their paired normal tissues
(Figures 1E and F). Expression levels of miR-214 were higher in 4
out of the 20 CRC samples, whereas HMGA1 expression was lower
in 1 out of the 20 tumours when compared with their paired
normal tissues. These results showing an inverse relationship
between the expression levels of miR-214 and HMGA1 in human
CaCx and CRC together with the identification of a miR-214-
binding site in the HMGA1 3
0
UTR suggested that miR-214 may
directly target and regulate HMGA1.
HMGA1 is directly targeted by miR-214. As the physiological
effects of endogenous miR-214 are difficult to ascertain owing to their
low levels in CaCx and CRC cells, the miRNA was first
re-expressed ectopically. CaCx and CRC cells transfected with miR-
214 mimic showed 3–6 and 2–5 fold more expression, respectively,
than their corresponding miR-control-transfected cells (Figure 2A).
Maintaining this range of expression, the effect of reintroducing miR-
214 on endogenous HMGA1 mRNA and protein levels was
determined, and the results show that miR-214 decreased HMGA1
both at mRNA (Figure 2B) and protein (Figure 2D) levels in human
CaCx and CRC cells. To confirm whether HMGA1 is targeted by
miR-214 by binding to its 3
0
UTR (Figure 2C), HMGA1 3
0
UTR
luciferase assays were performed and it was observed that re-
expression of miR-214 decreased wild-type HMGA1 3
0
UTR-regulated
luciferase activity by B30% in C33A, B46% in CaSki, B30% in
SiHa, B25% in SW480 and B33% in SW620 cells but not in cells
transfected with HMGA1 3
0
UTR-containing mutant miR-214-binding
sites (Figure 2E), confirming that miR-214 binds specifically to the
3
0
UTR of HMGA1 to repress gene expression. Together, these results
suggest that miR-214 negatively regulates endogenous HMGA1
expression by binding to its 3
0
UTR in CaCx and CRC cells.
Re-expression of miR-214 inhibits proliferation, migration and
invasion in CaCx and CRC cells. To evaluate whether miR-214-
mediated targeting and the consequent downregulation/repression
of HMGA1 has a role in tumorigenesis, first the effect of ectopic
expression of miR-214 on cell proliferation was studied using BrdU
incorporation assay in CaCx (C33A and SiHa) and CRC (SW480
and SW620) cells. Re-expression of miR-214 inhibited cell
proliferation significantly in C33A by B28%, SiHa by B13%,
SW480 by B23% and SW620 by B18% (Figure 3A). As the CaCx
and CRC cells used in the current study are known to exhibit
migratory and invasive properties, these cells were next tested by
re-expressing miR-214 and it inhibited migration in C33A by
B31%, SiHa by B23% and SW480 by B25% (Figure 3B) as well
as invasion in C33A by B25%, SiHa by B25% and SW480 by
B33% (Figure 3C). These results confirm that reintroduction of
miR-214 suppresses the aggressive behaviour of CaCx and CRC
cells by inhibiting their proliferation, migration and invasion.
HMGA1 knockdown inhibits proliferation, migration and
invasion in CaCx and CRC cells. Given that miR-214 may have
many potential targets in CaCx and CRC cells, the antiproliferative
effect of miR-214 may not be limited to repression of HMGA1.
To check whether suppression of HMGA1 would simulate miR-214-
mediated effects, siRNA-mediated knockdown of HMGA1 was
performed in CaCx and CRC cells (Figure 4A). Under these conditions,
cell proliferation was inhibited by 10–40% (Figure 4B). Similarly,
diminished levels of HMGA1 reduced migration (Figure 4C) by 30% as
well as invasion (Figure 4D) by 20–50%. From these data, it is inferred
that merely downregulating HMGA1 expression can inhibit prolifera-
tion, migration and invasion in CaCx and CRC cells, producing
phenotypes comparable to miR-214 reintroduction.
Inhibition of miR-214 enhances proliferation, migration and
invasion in CaCx and CRC cells. Although the ectopic expression
of miR-214 effectively inhibited proliferation, migration and invasion in
Mir-214 regulates HMGA1 in cervical and colorectal cancers BRITISH JOURNAL OF CANCER
www.bjcancer.com | DOI:10.1038/bjc.2016.234 743
CaCx and CRC cells, whether the endogenous miR-214 contributes to
these phenomena remained to be elucidated. When the endogenous
miR-214 was inhibited using an antimiR, there was an increase in
HMGA1 levels (Figure 5A) and cell proliferation (Figure 5B), migration
(Figure 5C) and invasion (Figure 5D) were also enhanced by 40–60%,
30–45% and 20–60%, respectively. From these data, it is ascertained
that downregulating miR-214 enhances endogenous HMGA1 levels
concurrently stimulating proliferation, migration and invasion of CaCx
CaCx
miR-214 expression (RPM)
HMGA1 expression (RPKM)
0 20406080100
0
5000
10 000
15 000
20 000
25 000
r =–0.258
P =0.0001
CRC
miR-214 expression (RPM)
HMGA1 expression (RPKM)
020406080
10
0
0
5000
10 000
15 000
20 000
25 000
r = –0.3069
P < 0.0001
Expression
of miR-214 relative to RNU6
(fold, log
10
)
Expression
of miR-214 relative to RNU6
(fold, log
10
)
Expression
of HMGA1 relative to ACTB
(fold, log
10
)
Expression
of miR-214 relative to RNU6
(fold, log
10
)
NT CT
0.1
1
10
100
1000
miR-214
*
*
Mean
NT - 66.13
CT - 20.26
NT
CT
0.0001
0.001
0.01
0.1
1
10
100
Expression
of HMGA1 relative to ACTB
(fold, log
10
)
Expression
ofHMGA1 relative to ACTB
(fold, log
10
)
HMGA1
*
*
Mean
NT - 8.2
CT - 22.6
miR-214
Colorectal tissues
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
1
6
17
18
19
20
0.001
0.01
0.1
1
10
PN
CT
HMGA1
Colorectal tissues
1
2
3
4
5
6
7
8
9
10
11
1
2
13
14
15
16
17
1
8
19
20
0.1
1
10
100
1000
PN
CT
miR-214
PN
CT
0.001
0.01
0.1
1
10
****
HMGA1
PN
CT
0.1
1
10
100
1000
****
B
A
C
D
E
F
Figure 1. Relative expression levels of miR-214 and HMGA1 in normal and tumour cervical and colorectal tissues. TCGA database was used to
plot Pearson product–moment correlation utilising normalised expression data for miR-214 (reads per million (RPM)) and HMGA1 (reads per
kilobase per million mapped reads (RPKM)) from (A) 215 cervical and (B) t 322 colorectal umour samples. Quantitative PCR was performed to
estimate the expression levels of (C) miR-214 from 20 normal cervical tissue specimens (NT) and 20 cervical tumour tissue specimens (CT) or (D)
HMGA1 from 15 normal (NT) and tumour (CT) cervical tissue specimens. Expression levels of miR-214 and HMGA1, respectively, were normalised
to that of RNU6 and ACTB. Similarly, 20 tumour (CT) and their paired normal (PN) colorectal tissue specimens were analysed for the expression
levels of (E) miR-214 and (F) HMGA1. Solid lines in box plots represent mean values. P-values are represented as *Pp0.05, ****Pp0.0001.
BRITISH JOURNAL OF CANCER Mir-214 regulates HMGA1 in cervical and colorectal cancers
744 www.bjcancer.com | DOI:10.1038/bjc.2016.234
and CRC cells, producing cellular phenotypes that are considerably
opposite to the ones produced by reintroduction of miR-214 or
independent downregulation of HMGA1.
miR-214 counters the effects of ectopically expressed HMGA1 on
proliferation, migration and invasion in CaCx and CRC
cells. As miR-214 effectively inhibited proliferation, migration
and invasion in CaCx and CRC cells, it was pertinent to test the
effects of expressing its target HMGA1 on these processes in the
presence of miR-214. Hence, miR-214 and a 3
0
UTR-less HMGA1
(unresponsive to miRNAs) were transfected individually or in
combination in CaCx and CRC cells. As expected, miR-214
decreased HMGA1 expression and ectopic expression of HMGA1
increased it over endogenous levels, and notably, their combined
ectopic expression still led to a decrease in HMGA1 in SiHa and
SW480 cells (Figure 6A). Similar effects on proliferation
(Figure 6B), migration (Figure 6C) and invasion (Figure 6D) were
observed when this combination was used. Although miR-214
expression alone reduced proliferation by 14–29%, migration by
22–30% and invasion by 15–27%, ectopic expression of HMGA1
enhanced these. Combined expression of HMGA1 and miR-214
reduced cell proliferation, migration and invasion by 11–29%,
15–37% and 20–25%, respectively, compared with cells transfected
with miR-control and vector-control. Taken together, these results
ascertain that the tumour-suppressive action of miR-214 prevails
over the protumorigenic effects of HMGA1 even if it is ectopically
expressed over and above the already abundant endogenous levels
in CaCx and CRC cells.
DISCUSSION
In recent years, miRNAs have been increasingly demonstrated to
have crucial roles in gene regulation, cellular signalling, carcino-
genesis and in related processes, including metastasis and
epithelial-to-mesenchymal transition (Suzuki et al, 2014; Kuninty
C33A
SiHa
SW480
S
W620
HMGA1
ACTB
miR-
contr
o
l
miR-214
miR
-
c
ontrol
miR-214
miR
-
control
miR-214
miR-control
m
iR-214
A
B
C
D
E
HMGA1 3′UTR luciferase
Relative luciferase activity (fold)
C33A
CaSki
SiHa
SW480
SW620
0.0
0.5
1.0
1.5
pcDNA-3.1 + WT
miR-214 + WT
pcDNA-3.1 + MUT
miR-214 + MUT
***
*
****
**
***
HMGA1
Expression
of
HMGA1 relative to ACTB (fold)
C33A
SiHa
SW480
SW620
0.0
0.5
1.0
1.5
miR-control
miR-214
**
**
****
*
miR-214
Expression
of miR-214
relative to
RNU6
(fold)
C33A
SiHa
SW480
SW620
0
2
4
6
8
10
miR-control
miR-214
*
*
*
*
1 0.55 1 0.76 1 0.89 1 0.93
HMGA1 3′UTR (WT)
HMGA1 3′UTR (MUT)
miR-214
5′
3′
3′
3′
5′
5′
Figure 2. Changes in HMGA1 expression upon ectopic expression of miR-214 in CaCx and CRC cells. C33A, SiHa, SW480 or SW620 cells were
transfected with control mimic (miR-control) or miR-214 mimic and the expression levels of (A) miR-214 or (B) HMGA1 were estimated by qPCR using
RNU6 or ACTB, respectively, for normalisation, and (D) HMGA1 protein levels were estimated by western blotting and ACTB was used as a loading
control. (E) psiCHECK-2 vector containing HMGA1 3
0
UTR either with wild-type miR-214-binding site (WT) or mutated site (MUT) was co-transfected with
pcDNA 3.1 or pcDNA 3.1-miR-214 (miR-214) in CaCx and CRC cells and luciferase assays were performed. Renilla luciferase activity in miR-214-
transfected cells was normalised to that of vector-transfected cells and Firefly luciferase served as internal control. (C) Wild-type (WT) and mutant (MUT)
miR-214-binding sites in HMGA1 3
0
UTR and miR-214 binding sequence (miR-214) are shown where solid lines and broken lines connect paired and
unpaired bases from WT and MUT with those of miR-214, respectively. P-values are represented as *Pp0.05, **Pp0.01 , ***Pp0.001 or ****Pp0.0001.
Mir-214 regulates HMGA1 in cervical and colorectal cancers BRITISH JOURNAL OF CANCER
www.bjcancer.com | DOI:10.1038/bjc.2016.234 745