The tumor suppressor protein p53 is intricately regulated by various signaling molecules, including non-coding small RNAs, called microRNAs (miRNAs). The in silico analysis and the inverse expression status in various cell lines raised the possibility of miR-27a being a new regulator of p53. Using luciferase reporter assay and various mutational and functional analysis, we identified two putative binding sites of miR-27a on the 3′-UTR of p53. The overexpression of miR-27a in the human colorectal cancer cell line HCT-116+/+ resulted in the decreased expression of the endogenous p53 protein levels. During hypoxia of the HCT-116+/+ cells, p53 showed increased accumulation after 3 h, and the levels were significantly up-regulated until 24 h of hypoxia. The p53 expression dynamics during hypoxia of the HCT-116+/+ cells were found to be inversely regulated by miR-27a expression. Moreover, using a cell viability assay, we established that after 3 h of hypoxia, the accumulation of p53 results in a decreased number of the viable HCT-116+/+ cells and the overexpression of miR-27a resulted in an increased number of viable HCT-116+/+ cells with a concomitant decrease in p53 expression. Additionally, our data indicated that miR-27a and p53 depict inverse expression dynamics in 50% of the human colorectal cancer samples studied, when compared with that in the adjacent normal samples. Our data established that miR-27a and the tumor suppressor protein p53 are part of the same signaling network that has important implications during hypoxia and tumorigenesis.
p53 is regarded as critical for the genomic stability, cell growth and proliferation [1,2]. It is mutated in more than 50% of human tumors, and in many cases, disruption of p53's function(s) is a prerequisite for the initiation and progression of various tumors . Functionally, p53 is a transcription factor, whose activity is regulated by many cellular stresses, such as DNA damage , hypoxia  and viral entry . Activation of p53 selectively transcribes a set of its target genes that ultimately determines the cellular response . Depending on the cell type and the degree of stress, the p53 protein induces cell cycle arrest, senescence or apoptosis so as to prevent the proliferation of the damaged cells, which have the potential of becoming cancerous . Moreover, numerous signaling molecules, in turn, regulate the p53 protein. One of the key cellular components that regulate the p53 protein level is the E3 ubiquitin protein ligase murine double minute 2 (Mdm2) . Mdm2 binds to the N-terminus of the p53 protein and induces its degradation by promoting the ubiquitylation and subsequent proteasomal degradation .
Among the various stress conditions, hypoxia is significantly known to influence p53 expression, which, in turn, mediates various physiological and pathophysiological cellular responses to such conditions . It has been reported that the exposure to severe hypoxia leads to the accumulation of p53 inside cells, which ultimately leads to cell apoptosis . Most of the studies suggest that hypoxia leads to stabilization of hypoxia-inducible factor-alpha-1α (HIF-1α) inside cells , which in turn suppresses Mdm2-mediated degradation of p53 and thus stabilizes p53 protein levels . However, the suppression of Mdm2 does not account fully for the accumulation of p53 during hypoxia. Interestingly, previous studies have unveiled the role of microRNAs (miRNAs) in modulating the expression of p53 . miRNAs are endogenous class of small non-coding RNAs that regulate gene expression post-transcriptionally. miRNAs bind to the complementary sites on the 3′-UTR of the target mRNAs and regulate the expression by either repressing the translation or degrading the target mRNA . miRNAs are known to regulate the vast number of mRNA targets, which are involved in cell proliferation , senescence , metabolism , differentiation and apoptosis [18,19]. The deregulation of the miRNA expression has a deleterious effect on the normal functioning of the cell. Aberrant expression of miRNAs has been observed in various human malignancies, including breast [20,21], prostate , gastric  and colorectal  cancer. miR-125b and miR-504 have been shown to negatively regulate the human p53 gene at the translational level [25,26]. Conversely, p53 also regulates various miRNA genes by binding to the p53-responsive elements present in the promoter regions . For example, the miR-34 gene family contains p53 consensus sequences in the promoter that are activated by p53, which in turn regulates apoptosis, senescence and cell cycle arrest . Moreover, increasing evidence is emerging that support the role of miRNAs in tumorigenesis and angiogenesis that are driven by hypoxia [19,29].
Thus, there exists a complex interplay between miRNAs, p53 and various stress conditions, including hypoxia. In the present study, we further tried to explore the role of other miRNAs that may regulate p53 expression with implications during hypoxia and cellular transformations.
Materials and methods
Thirty human colorectal cancer tissues and paired normal adjacent tissues were obtained from patients who underwent radical resection at the Department of General Surgery, Sheri-Kashmir Institute of Medical Sciences (SKIMS), Srinagar, J&K, India, from 2013 to 2015. In all cases, the clinical diagnosis was confirmed through histological examination. All samples were collected as per the protocol approved by the Research Ethical Committee, SKIMS, and the written informed consent of each patient was also given. For each case, samples from the primary tumor and the corresponding normal colorectal tissue were collected for comparison and immediately preserved in RNA later® (Qiagen, Germany) and then stored at −80°C until for further use.
The pmirGLO Dual-Luciferase miRNA Target expression vector was obtained on request from Yusuke Takahashi, University of Oklahoma, USA. The pmirGLO vector is a dual-luciferase vector having the firefly luciferase gene under the control of phosphoglycerate kinase (PGK) promoter and the Renilla luciferase gene driven by the SV40 promoter. The human p53-3′-UTR (1443 to 2564 bp; NM_000546) was PCR-amplified from the cDNA [obtained from HCT-116+/+ cells, known to have wild-type (WT) p53], using 3′-UTR-FP and 3′-UTR-RP primer pairs (see Supplementary Table S1 for primer description) and Q5 high-fidelity DNA polymerase (New England Biolabs, UK). The desired PCR product was gel-purified, using the QIAquick gel extraction kit (Qiagen), restriction-digested (NheI/SalI) and ligated downstream of the firefly gene of the pmirGLO vector at NheI and SalI (New England Biolabs) restriction sites to obtain the pmirGLO-p53-3′-UTR construct. In addition, three mutant constructs (Mut-1, Mut-2 and Mut-1+2) of the p53 3′-UTR in pmirGLO were made using the QuikChange Site-Directed Mutagenesis Kit in accordance with the manufacturer's protocol (Agilent Technologies, CA, USA). The Mut-1 construct, having a 4-base substitution at site-1 (436–442), was prepared using the partially overlapping primers 3′-UTR M1-FP and 3′-UTR M1-RP. The Mut-2 construct, having a 4-base substitution at site-2 (588–594), was prepared using the partially overlapping primers 3′-UTR M2-FP and 3′-UTR M2-RP. Similarly, the double-mutant construct Mut-1+2, having 4-base substitutions at site-1 and site-2, was prepared using the above primers. The description of all the mutant constructs is provided in Supplementary Table S2.
To construct a plasmid expressing mature miR-27a-3p, a 300 bp DNA fragment, encompassing miR-27a precursor gene, was PCR-amplified from the human genomic DNA using 27a-FP and 27a-RP primer pairs. The desired amplicon was gel-purified, restriction-digested using EcoRI/BamHI restriction enzymes (New England Biolabs) and ligated at EcoRI/BamHI sites of the pcDNA3.1(−) vector (Invitrogen, CA, USA). Likewise, pcDNA 3.1(−) constructs, expressing mature miR-125b and miR-28, were made using primer pairs 125b-FP, 125b-RP and 28-FP, 28-RP, respectively. To prepare a vector construct having p53-coding region upstream of WT p53 3′-UTR or mutant (site-1+2) p53 3′-UTR, the firefly luciferase-coding region of the pmirGLO-p53-3′-UTR construct was replaced with WT p53-coding region. For this purpose, the firefly luciferase-coding region of the pmirGLO-p53-3′-UTR construct was removed using the ApaI (upstream of firefly luciferase) and Nhe-I (downstream of firefly luciferase) restriction enzymes and gel-purified after ApaI/NheI digestion. The full-length p53-coding sequence was PCR-amplified from the cDNA (obtained from HCT-116+/+ cells) using primer pairs p53-FP and p53-RP. The desired PCR amplicon was gel-purified, digested with ApaI/NheI restriction enzymes and ligated at ApaI/NheI sites of the digested pmirGLO-p53-3′-UTR vector. Similarly, the p53 cDNA construct, having Mut-1+2 3′-UTR, was prepared from the pmirGLO-Mut-1+2 p53-3′-UTR construct, using the similar strategy as described above. All primers used in the present study were synthesized using the services of Imperial Life Sciences (Gurgaon, India). All constructs were sequenced at SciGenom Labs (Kerala, India).
Cell culture and transfection
Cell lines U2OS, HCT-116+/+ (having WT p53), NIH3T3, C6, human embryonic kidney (HEK)-293 and IMR32 cells used in the present study were obtained from National Centre for Cell Science (Pune, India). All cell lines were grown in Dulbecco's modified Eagle's medium (DMEM) (Sigma–Aldrich, MA, USA), supplemented with 10% fetal bovine serum (Sigma–Aldrich, MO, USA) and 100 units of penicillin/ml and 100 µg of streptomycin/ml (Hyclone, South Logan, UT). Cells were incubated at 37°C in a humidified CO2 incubator supplemented with 5% CO2 (Eppendorf Brunswick, USA). All transient transfections including that of miR-27a constructs, anti-miR-27a, scrambled miRNA and pmiRGLO constructs were carried out using polyethyleneimine (PEI) reagent (Polysciences, Inc., Warrington, PA) and in accordance with the manufacturer's protocol. For the hypoxia studies, 70% confluent HCT-116+/+ cells were exposed to 1% O2 in the hypoxia chamber (New Brunswick Galaxy 170R-230-1200), using the hypoxic environment by triple-gas mixture (1% O2, 5% CO2 and 94% N2). The cells were exposed to the hypoxic environment for the stipulated time points of 3, 6, 12 and 24 h. In the case of miR-27a transfections, HCT-116+/+ cells were seeded at 50% confluency on day 0. Next day, the cells were transiently transfected with 4 µg of miR-27a construct using PEI reagent (Polysciences). Twenty-four hours post-transfection, the cells were exposed to hypoxia (95% N2, 1% O2 and 5% Co2) for 24 h and then further processed for either RNA or protein isolation.
Dual-luciferase reporter assay
For luciferase assay, cells were seeded in 12-well plates at 50% confluency 1 day before transfection. The next day, pmiR-GLO luciferase constructs (Cont., WT, Mut-1, Mut-2 and Mut-1+2) were co-transfected along with pcDNA 3.1(−) expressing miR-27a into the HCT-116+/+ cells, using PEI transfection reagent and in accordance with the manufacturer's instructions. Luciferase activities [firefly (FLuc) and Renilla (RLuc)] were measured 48 h post-transfection using the Dual-Luciferase assay kit (Promega, Madison, WI, USA) and the microplate luminometer (Tecan Technologies, Switzerland) as per the manufacturer's instructions. The firefly luciferase (FLuc) activity and the Renilla luciferase (RLuc) activity were measured, and the data were represented as the ratio of the firefly-to-Renilla luciferase activity (FLuc/RLuc). Each construct was studied in triplicate, and the each experiment was repeated three times. Data are expressed as means ± SEM.
RNA isolation and real-time PCR
Total RNA from different cell lines and human colorectal tissues was isolated using TRIzol® reagent (Invitrogen, USA) in accordance with the manufacturer's instructions. After DNase-I (New England Biolabs) treatment, total RNA was subjected to reverse transcription using a standard protocol. The first cDNA strand of the miRNAs was synthesized using Superscript III (Invitrogen, USA) and two-step quantitative real-time PCR (qRT-PCR) method with hairpin-looped reverse transcription (RT) primers specific for miR-27a-3p, miR-22a-3p, Let-7d-5p, miR-125b, miR-28 and miR-16. miR-16 was used as an endogenous control. The RT primers were designed as per Chen et al. . The miRNA stem-loop RT primer sequences used for first-strand cDNA synthesis are shown in Supplementary Table S1. The first-strand cDNA was subjected to qRT-PCR in real-time PCR 7500 (Applied Systems) using maxima SYBR Green® mix (Thermo Scientific, USA) and miRNA-specific forward primer and a universal reverse primer and in accordance with the manufacturer's protocol. The sequences of forward primers, used for the amplification of various miRNAs along with a universal reverse primer, are given in Supplementary Table S1. All reactions were run in triplicate.
Protein extraction and Western blot assay
From the human colorectal tissue, the protein extract was obtained by disintegrating the tissue in 0.5% trypsin–EDTA solution (Sigma, St. Louis, MO, USA), followed by incubation at 37°C for 5 min. The resultant cell suspension and the cells harvested from the culture were washed with phosphate-buffered saline (Sigma–Aldrich, USA) followed by lysis, using NP-40 lysis buffer [20 mM Tris/HCl (pH 8), 137 mM NaCl, 10% glycerol, 1% NP-40 and 2 mM EDTA]. Protease inhibitor cocktail, PMSF and phosphatase inhibitors (1 mM PMSF, 5–10 mM NaF; Sigma–Aldrich, USA) were freshly added to the lysis buffer. The supernatant was collected and the protein concentration was estimated using the Bradford assay using the standard protocol. Protein extract, preheated at 100°C for 5 min in reducing SDS sample buffer containing 50 mM Tris/HCl (pH 6.8), 2% SDS, 10% glycerol, 0.1% bromophenol blue, 100 mM β-mercaptoethanol, was run on 12% SDS–polyacrylamide gel. After gel electrophoresis, separated proteins were transferred onto the PVDF membrane (Millipore, USA) by the semidry transfer method in accordance with the manufacturer's instructions (Hoefer, USA). For immune-detection, the PVDF membrane was processed using the standardized protocol. Immune-detection was performed using rabbit anti-p53 (1:1000; Santa Cruz Biotechnology, Dallas, TX, USA; Cat.# SC-6243) or anti-HIF-1α (1:100; Abcam, Clone ID EP1215Y, Cat.# 2015-1). Mouse anti-β-actin (1:1000; Santa Cruz Biotechnology, Cat.# SC-47778) was used as the endogenous control. The secondary detection was performed using ﬂuorescent anti-mouse IRDye 680 (1:20 000) and anti-rabbit IRDye 800 (1:10 000) secondary antibodies (LI-COR, USA). The ﬂuorescence was detected using the Odyssey infrared detection system (LI-COR).
Cell viability assay
For cell viability assay, ∼104 cells were seeded in a 12-well plate with all necessary controls and incubated overnight at 37°C in a CO2 incubator. The next day, the cells were transfected with the desired plasmids. All transfections were tested at least in triplicate, and the assay was repeated independently three times. Forty-eight hours post-transfection, the culture medium was removed completely, followed by the addition of 300 μl of MTT solution (Invitrogen, Eugene, OR, USA; 5 mg/ml in PBS). After 4 h of incubation at 37°C, MTT was carefully removed, and 300 µl of MTT solvent was added to dissolve formazan crystals at room temperature for 30 min in the dark. The optical density of plate was read at 570 nm on ELISA Plate Reader (BioTek Instruments, Inc.) within 10 min.
Quantification of the Western blots
Protein bands of the Western blots were quantified using the Licor infrared scanner and in accordance with the protocol described by the LI-COR Odyssey system. Briefly, for the protein quantification, the standard was prepared by adding different concentrations of fluorescently labeled LI-COR secondary antibodies on the PVDF membrane. The fluorescent spots on the membrane were identiﬁed using a shape tool, and the fluorescence emitted was measured. Similarly, the fluorescence emitted by the Western blot bands was measured, and the amount of protein was estimated using the standard. Each protein band was normalized with the control protein band of β-actin.
In the present study, representative experiments from three independent experiments are shown. Results for each experiment are given as means of triplicates ± SEM. Statistically significant differences between the sample groups were determined using Student's t-test. A value of P < 0.05 was considered significant.
Two evolutionarily conserved miR-27a-binding sites are present in the 3′-UTR of p53
To investigate the role of the miRNAs in regulating the expression of p53, in silico analysis was performed for the putative miRNA-binding sites in the 3′-UTR of p53. For this purpose, the online bioinformatics software TargetScan 6.2 (http://www.targetscan.org/) was used. On the basis of the total number of binding sites of a particular miRNA, their context score and the conservation among different species, three miRNAs — Let-7d-5p, miR-27a-3p and miR-22a-3p — were selected for further study (Supplementary Figure S1). Next, we determined the endogenous expression levels of the three miRNAs (Let-7d, miR-27a and miR-22a) and the p53 protein in six different cell lines. The mature miRNA expression was quantified by qRT-PCR, and p53 expression was detected by the Western blotting as described in the Materials and methods section. Interestingly, out of the three miRNAs, only the expression levels of miR-27a (Figure 1A) showed significant inverse correlation with the expression levels of p53 (Figure 1B). The cell lines with high endogenous miR-27a levels, such as human osteocarcinoma U2OS, rat glioma C6, human HCT116+/+ and the human neuronal IMR-32, depicted decreased expression of the endogenous p53 (compare Figure 1A and B). Conversely, the cell lines, such as human HEK-293T and mouse NIH 3T3 cells, showed low endogenous miR-27a levels, and the expression of p53 was found to be highest. Moreover, the in silico analysis indicated that the 3′-UTR of the p53 mRNA possesses two miR-27a-binding sites (sites 1 and 2; Figure 1C). Both the seed regions of the site-1 and the site-2 (shown as upper-case letters) showed 7 bp complementary with miR-27a (Figure 1D). The significance of the miR-27a-binding sites on the p53 3′-UTR was further determined by their evolutionary conservation. As shown in Figure 1E, the seed region of site-1 showed complete conservation among different organisms, whereas the seed region of site-2 showed ∼90% conservation.
Two evolutionarily conserved miR-27a-binding sites are present in the 3′-UTR of p53.
Overexpression of miR-27a decreased the endogenous p53 protein levels in the human colon cancer cell line HCT-116+/+
To determine whether miR-27a regulates the expression of the endogenous p53 protein levels, miR-27a, cloned in the pcDNA3.1(–) plasmid, was transiently transfected in the HCT-116+/+ cells. Forty-eight hours post-transfection, the overexpression of miR-27a was confirmed by qRT-PCR. As shown in Figure 2A, the expression of miR-27a was ∼6-fold more than that of those cells in which empty vector was transfected (plasmid only). As a positive control for p53 regulation, the miR-125b-pcDNA3.1 construct was also transiently transfected (Figure 2A). Additionally, as a negative control, miR-28 (known to have no binding site on the 3′-UTR of p53) was transiently transfected in the HCT-116+/+ cells (Figure 2A). Moreover, total protein extract was prepared from the miR-27a-, miR-125b- and miR-28-transfected cells, and Western blotting was performed using rabbit anti-p53 antibody. As shown in Figure 2B, the miR-27a-transfected cells (lane II) showed decreased expression of p53 when compared with that of the control cells (lane I). Similarly, the miR-125b-transfected cells showed decreased expression of p53 (lane III) when compared with that of control cells (lane I). The expression of miR-28 did not show any effect on the expression of p53 (lane IV). The detection of β-actin by the Western blotting, using mouse anti-β-actin antibody, was used as a protein loading control. The densitometry analysis of the protein bands shown in Figure 2B (n = 3) indicated that miR-27a transfection resulted in ∼2-fold decrease in the endogenous expression of p53 (Figure 2C).
Overexpression of miR-27a decreased the endogenous p53 protein levels in HCT-116+/+ cells.
To further validate the regulation of p53 mRNA by miR-27a, we synthesized anti-miR-27a and a scrambled miRNA (as a negative control) as described in the Materials and methods section. The anti-miR-27a was co-transfected along with pcDNA-miR-27a vector in the HCT116+/+ cells. Forty-eight hours post-transfection, protein extract was prepared, and Western blotting was performed using the rabbit anti-p53 antibody. As shown in Figure 2D, the transfection of miR-27a (M) resulted in the decreased expression of p53 (lane II) when compared with that of the control (C) vector (lane I). The co-transfection of anti-miR-27a (A) along with miR-27a (M + A; lane III) significantly abrogated the inhibitory effect of miR-27a on the expression of p53 (compare lane III with lane II). As the specificity control, the co-transfection of the scrambled miRNA (Sc) along with miR-27a did not abrogate the effect of miR-27a on the expression of p53 (M + Sc; lane IV). Moreover, the transfection of Sc had no effect on the expression of p53 (lane V). The densitometry analysis of the protein bands shown in Figure 2D (n = 3) indicated that the transfection of miR-27a (M) resulted in significant ∼2.2-fold decrease in the expression of p53, whereas the co-transfection of anti-miR-27a along with miR-27a (M + A) decreased the inhibitory effect of miR-27a on the expression of p53 (Figure 2E). These results confirm the specific role of miR-27a in regulating p53 expression.
miR-27a specifically binds the 3′-UTR of p53 mRNA
The in silico analysis indicated that the 3′-UTR of p53 mRNA possesses two putative binding sites for miR-27a. To prove that these two sites are important for the regulation of p53 expression by miR-27a, we subcloned the 3′-UTR of the WT p53 downstream of the firefly luciferase gene using pmirGLO reporter vector (WT). Moreover, we prepared the mutant constructs by mutating either site-1 (Mut-1) or site-2 (Mut-2) and both site-1 and site-2 (Mut-1+2). The schematic representation of all these constructs is shown in Figure 3A. All of the above constructs were transiently transfected in HCT-116+/+ cells, and in some cases miR-27a was co-transfected along with the WT or mutant constructs. Forty-eight hours post-transfection, cells were harvested and the cell extract was prepared followed by the measurement of the Renilla (RLuc) and firefly luciferase (FLuc) activity as described in the Materials and methods section. The firefly to Renilla luciferase ratio (FLuc/RLuc) of the control vector (Cont.) was set at 1. As shown in Figure 3B, the FLuc/RLuc ratio of the WT construct was found to be ∼0.8, indicating down-regulation of the firefly luciferase (under the control of WT p53-3′-UTR) by the endogenous miR-27a. Furthermore, the FLuc/RLuc ratio was significantly reduced (∼70%) in the cells that were co-transfected with miR-27a and the WT construct, when compared with those transfected with the WT construct alone (Figure 3B). The expression of miR-27a in the above samples was validated using qRT-PCR (results not shown). The above data indicated that miR-27a binds the p53 WT 3′-UTR and hence reduces the expression of the firefly luciferase as depicted by the decreased FLuc/Rluc ratio. This was further confirmed by co-transfecting the miR-27a with either Mut-1 or Mut-2 or with Mut-1+2 constructs. As shown in Figure 3B, the inhibitory effect of miR-27a on the FLuc/RLuc ratio was appreciably reduced when the Mut-1 construct was used, whereas the Mut-2 construct showed significantly more reduced effect of miR-27a (compare Mut-2 + miR-27a with WT). Moreover, the Mut-1+2 construct showed a slightly higher FLuc/RLuc ratio than that of the WT. The above results corroborate the in silico data that site-1 and site-2 on the 3′-UTR of p53 are important for the inhibitory effect of miR-27a on the expression of p53.
miR-27a specifically binds the 3′-UTR of human p53 mRNA.
miR-27a and p53 protein depict inverse expression dynamics during hypoxia stress
p53 shows differential expression status during different cellular stress conditions, like hypoxia. To understand the expression dynamics of p53 and miR-27a during hypoxia, HCT-116+/+ cells were exposed to hypoxic conditions after incubation in a hypoxia chamber with the triple gas mixture (95% N2, 1% O2 and 5% CO2) for different time points (Figure 4A). The cells were harvested, protein extract was prepared and Western blotting was performed using the rabbit anti-p53 antibody. As shown in Figure 4A, p53 showed increased expression after 3 h of hypoxia (lane II) when compared with that of control cells (lane I). Moreover, the expression of p53 showed almost a linear increase in the expression until 24 h of hypoxia (compare lane V with lane I). The densitometry analysis of the protein bands shown in Figure 4A (n = 3) indicated that after 3 h of hypoxia, the expression of p53 increased ∼1.5-fold when compared with that of the control (0 h) cells, whereas the expression was ∼2.5-fold higher till 24 h of hypoxia (Figure 4B). As a positive control for the hypoxic conditions, the expression of HIF-1α was detected by Western blotting using anti-HIF-1α antibody (Supplementary Figure S3). Next, we studied the expression of miR-27a under similar hypoxic conditions using qRT-PCR. As shown in Figure 4C, miR-27a showed a ∼1.3-fold decrease in expression after 3 h of hypoxia, whereas there was a ∼5.5-fold decrease in the expression after 24 h of hypoxia when compared with that of the control cells (0 h). To further demonstrate the interplay between miR-27a and p53 during hypoxia, we transiently transfected HCT-116+/+ cells with the miR-27a expression vector, and after 24 h the cells were exposed to 24 h of hypoxia using the triple gas mixture. After hypoxia, protein extract was prepared, and Western blotting was performed using the rabbit anti-p53 antibody. As shown in Figure 4D, 24 h of hypoxia to the untransfected cells resulted in an increase in p53 expression (lane II) when compared with that of the control cells (lane I). However, the cells that were transfected with miR-27a did not show any increase in the expression of p53 (lane III) after 24 h of hypoxia. The overexpression of miR-27a was confirmed by qRT-PCR (results not shown). The above data indicated that the decreased expression of miR-27a after 24 h of hypoxia results in the accumulation of p53 in the hypoxic cells. miR-27a and p53 have opposite effect on the cell viability
miR-27a and p53 protein depict inverse expression dynamics during hypoxia stress.
To ascertain the physiological significance of the increased p53 expression during 24 h of hypoxia, we performed a cell viability assay. For this purpose, we transiently transfected HCT-116+/+ cells with miR-27a, WT-p53, miR-27a plus WT-p53, or miR-27a plus p53-Mut-1/2 constructs. Forty-eight hours post-transfection, the cell extract was prepared, and the cell viability assay was performed as described in the Materials and methods section. As shown in Figure 5A, 24 h of hypoxia (hypoxia only) resulted in a significant decrease in cell viability when compared with that of the control cells (vehicle control). The miR-27a-transfected cells depicted increased cell viability (miR-27a only) when compared with hypoxia-only cells. Moreover, the WT-p53-transfected cells showed a significant decrease in the cell viability (compare WT-p53 cells with vehicle control cells). Furthermore, the miR-27a and WT-p53-co-transfected (miR-27a + WT-p53) cells showed increased cell viability when compared with WT-p53 cells. Interestingly, the cells co-transfected with miR-27a and Mut-1/2 p53 (miR-27a + p53-Mut-1/2) showed decreased cell viability when compared with the miR-27a + WT-p53 cells. To confirm the differential expression of p53 during the cell viability assay, we performed the Western blotting using anti-p53 antibody from the same samples used for the cell viability assay. As shown in Figure 5B, the hypoxia-only cells showed increased p53 expression (lane II) when compared with that of the vehicle control cells (lane I). The expression of miR-27a resulted in the decreased expression of p53 (lane III), whereas the transfection of the WT-p53 showed increased expression of p53 (lane IV). The co-transfection of WT-p53 and miR-27a (lane V) did not depict increased expression of p53 (compare lane V with lane VI). Interestingly, the co-transfection of miR-27a and Mut-1/2-p53 showed increased expression of p53 (lane VI). The data in Figure 5B are completely in accordance with the cell viability assay (Figure 5A). These results clearly demonstrate that miR-27a and p53 play opposite roles during cell viability.
miR-27a and p53 have opposite effects on cell viability.
miR-27a and p53 show differential expression in the human colorectal cancer samples
In addition to in vitro studies, we sought to corroborate the negative regulation of the p53 protein by endogenous miR-27a under in vivo conditions. For this purpose, resected tumor and normal tissue samples of 30 patients with colorectal cancer were analyzed for p53 protein and miR-27a expression. p53 expression in the normal (N) and the tumor (T) samples was detected by Western blotting using the rabbit anti-p53 antibody. As shown in Figure 6A (representative blot), the expression of p53 was reduced in the tumor samples when compared with that of their adjacent normal (N). The detection of β-actin was used for the protein loading control. The densitometry analysis of protein bands shown in Figure 6A (n = 30) indicated that the p53 expression was significantly decreased in the tumor samples when compared with that in the normal samples (Figure 6B). For the expression analysis of mature miR-27a, the same tissue samples were used to detect the miR-27a levels using qRT-PCR. As shown in Figure 6C, miR-27a showed increased expression in 80% of the tumor (T) samples when compared with that in the normal (N) samples. The above data indicate that miR-27a and p53 show differential expression in the human colorectal cancer samples.
miR-27a and p53 show differential expression in human colorectal cancer samples.
The tumor suppressor p53 protein is a multifunctional protein and is regarded as the main guardian of genomic stability . Being so crucial for cells, p53 protein expression and activity are tightly regulated. The expression and activity of p53 are controlled by numerous extracellular and intracellular molecules . While most of the molecules and mechanisms that regulate the expression of p53 have been elucidated, the discovery of non-coding miRNAs has diversified the list of signaling molecules that regulate p53 expression. Working on the same lines, we investigated the role of miRNAs in regulating the expression of p53. Accordingly, in silico analysis indicated that the p53 3′-UTR possesses many putative miRNA-binding sites. Our initial data, from the various cell lines, indicated that there exists a strong inverse correlation between the expression of p53 and miR-27a. The evolutionary conservation of the miR-27a-binding site on the p53 3′-UTR mRNA between different organisms signifies its importance in regulating p53 expression. This was further demonstrated by creating mutant constructs and by performing the luciferase assays. Similar methods have been employed to establish the role of the novel miRNAs in regulating p53 expression . During the compilation of the results of the present study, our identification of the miR-27a as a novel p53-regulating non-coding RNA got a strong endorsement from the very recent paper by Towers et al. . In line with our studies, they have demonstrated that the Six1 oncoprotein up-regulates miR-27a in various tumor cells and that in turn results in the down-regulation of the p53 protein.
Recently, various miRNAs have been identified, and most of them have been shown to negatively regulate p53 expression . One such example is miR-125b that negatively regulates p53 expression during embryonic development, and this leads to the inhibition of cell apoptosis, which is necessary for normal tissue growth . Similarly, miR-34a is known to restore p53 functions and thus inhibits cell growth and induces apoptosis . Besides the post-translational regulation, the p53 expression levels are also highly dynamic particularly during various pathophysiological and stress conditions . Accordingly, we demonstrated that during hypoxia, p53 expression is regulated by miR-27a. Our data indicated that the p53 protein is up-regulated after 3 h of hypoxia, and the levels show an increasing trend until 24 h of hypoxia, while miR-27a showed decreased expression levels after 3 h of hypoxia. We further confirmed the inverse correlation between p53 and miR-27a by transfecting HCT-116+/+ cells with miR-27a, and demonstrated that miR-27a prevents the accumulation of p53 after 24 h of hypoxia. The accumulation of the p53 protein following stress conditions has been described previously . Various mechanisms have been proposed for the accumulation of p53 after hypoxia, such as the stabilization of the p53 protein  either by Mdm2-dependent [34,36] or Mdm2-independent pathways . However, none of these mechanisms have been shown to fully account for the accumulation of p53 after hypoxia. Moreover, other pathways have been proposed that include the increased translation of p53 mRNA . In this context, we propose that the down-regulation of miR-27a during hypoxia may be one such molecular mechanism responsible for the accumulation of p53. The physiological significance of p53 accumulation during hypoxia may be to prevent the propagation of hypoxic insult to the daughter cells. Our argument is further strengthened by the observation that the accumulation of p53 after hypoxia leads to replication arrest and activation of various genes leads to cell cycle arrest . With reference to this, our data indicate that the hypoxia-induced accumulation of p53 results in the decreased cell viability. Moreover, the overexpression of miR-27a leads to decreased expression of p53 with the concomitant increase in the number of viable cells. The role of miR-27a in increasing the number of viable cells by decreasing the expression of p53 demonstrates that miR-27a has a positive effect on the cell viability. miR-27a has been mostly studied in the context of human cell transformations . The studies have shown that miR-27a, along with other miRNAs, acts as the repressor of various tumor suppressor proteins, and their altered activities have been associated with the higher proliferation rates of pancreatic ductal adenocarcinoma cells . On the basis of various other studies, miR-27a has been described to act as an oncomiR  and it is well established that it is up-regulated in various human cancers . Moreover, our data also indicated that miR-27a is significantly overexpressed in human colorectal cancer samples when compared with the adjacent normal samples. Furthermore, our data established that there exists an inverse correlation between p53 and miR-27a expression in the colorectal cancer samples studied. Our data have been endorsed by other studies in which it was shown that miR-27a is up-regulated in colon cancer cells, and targeting miR-27a has a negative effect on the proliferation of the colon cancer cells . Similar to our study, miR-504 has been shown to negatively regulate p53 at the translational level, and this has been linked to the inhibition of cell apoptosis as well as cell cycle arrest in response to stress. miR-504 also enhances tumorigenesis of cells in vivo and is known to be unregulated in some types of human tumors, ultimately leading to down-regulation of p53 .
Overall, the present study describes the role of a new miRNA (miR-27a) in regulating the expression of p53 and the interplay between these two during the cell hypoxia and cell transformation. It is well established that the inner cells of the highly proliferative solid tumors survive the hypoxic conditions . These transformed cells have developed strategies to survive and proliferate under the extreme hypoxic conditions that are known to exist in the interiors of these tumors. On the basis of our data, we put forward an argument that under hypoxic conditions, the cells present in the interior of the solid tumors up-regulate miR-27a and that, in turn, may lead to the repression of the tumor suppressor p53 protein, thus favoring cell proliferation over cell apoptosis. In accordance with our data, various studies have established that p53-positive tumors show significant hypoxia-induced apoptosis, while p53-null tumors were more aggressive even under hypoxic conditions .
FLuc, firefly luciferase; HEK, human embryonic kidney; HIF-1α, hypoxia-inducible factor-1α; Mdm2, murine double minute 2; miRNAs, microRNAs; PEI, polyethyleneimine; PGK, phosphoglycerate kinase; qRT-PCR, quantitative real-time PCR; RLuc, Renilla luciferase; RT, reverse transcription; Sc, scrambled miRNA; WT, wild-type.
R.M. performed the subcloning procedures, mutational analysis, Western blotting experiments, luciferase assays and helped in drafting the manuscript. S.N.L. performed qRT-PCR, colorectal tissue sampling and Western blotting. M.U.H. conceived the study, designed, co-ordinated the experiments and drafted the manuscript. All authors read and approved the final manuscript.
The present work was supported by a project grant [SB/SO/BB-014/2014] from the Science and Engineering Research Board, Department of Science and Technology (DST) New Delhi to M.U.H. A UGC Junior research fellowship to R.M. [19-06/2011Ci EU-IV] and ICMR Junior research fellowship to S.N.L. [JRF-2012/HRD-29] are highly acknowledged.
We acknowledge the kind help of the Yusuke Takahashi, University of Oklahoma, USA for providing the pmirGLO vector. We are also grateful to Dr Mohd Jamal Dar (Senior Scientist, IIIM Jammu) for helping with the luciferase data.
The Authors declare that there are no competing interests associated with the manuscript.