CROSSTALK BETWEEN LIVER-RELATED MICRORNAS AND WNT Β- [602815]

1 
 CROSSTALK BETWEEN LIVER-RELATED MICRORNAS AND WNT/ Β-
CATENIN PATHWAY IN HEPATOCELLULAR CARCINOMA PATIENTS
Abeer M. Ashmawy1*, Khaled M. Elgeshy2, El-Said T. Abdel Salam2, Mohamed
Ghareeb3, Mohamed H. kobaisi4, Sabry K. Sharawy1, Abdel Hady A. Abdel Wahab1
1Department of Cancer Biology, Biochemistry Unit, and 3Department of Medical
Oncology, National Cancer Institute, Cairo University, Egypt.2Department of Botany
and microbiology, Faculty of Science, Cairo University, Egypt.4Department of
Pathology, National Institute of Urology and Nephrology, Cairo, Egypt.
*Correspondence author: (E-mail: [anonimizat] .)
ABSTRACT
Hepatocellular carcinoma (HCC) is one of the most common cancers
worldwide. MicroRNAs not only post transcriptionally regulate gene expression but also respond to signaling molecules to affect cell functions and have been implicated
in regulating Wnt pathway specifically in HCC. Herein, we verified a crosstalk
between Wnt/ β-catenin signaling pathway and microRNAs expression in tissues of 27
primary HCC patients and 6 control subjects. Expression level of 13 different miRNAs was examined using real-time PCR. Five proteins involved in the Wnt/ β-
catenin pathway were estimated by im munohistochemistry. MiRNA-10a, miR-30e,
miR-215, miR-125b and miR-148a were significantly positive correlated with β-
catenin, APC and c-myc and this association with Wnt/ β-catenin cascade proteins
provides a better understanding of miRNAs ro le in HCC, which could be exploited to
develop new therapeutic target strategies.
Key words : HCC,Wnt/ β-catenin, MicroRNAs

2 
 INTRODUCTION
Hepatocellular carcinoma (HCC) is the fifth most commonly diagnosed cancer
worldwide but the third leading cause of cancer-related death ar ound the world(Block,
Mehta et al. 2003). The involvement microRNAs (miRNAs) as regulators for HCC
development has been become a focus of research in molecular biology. MiRNAs are small, evolutionarily conserved, single stranded RNA molecules of approximately
21–24 nucleotides(Papaconstantinou, Karakatsan is et al. 2012). By targeting different
genes in tumor development, miRNAs function as oncogenes or tumor suppressor
genes. Several reports revealed a deregulat ion of different miRNAs in different types
of cancer including HCCs (Sun, Lu et al. 2013).The Wnt pathway is a highly regulated signaling pathway that controls numerous stages of tissue homeostasis. This pathway is closely regulated at both transcriptional-level regulations to post-
translational modification; thus aberrant Wnt activity often results in developmental disorders and diseases including but not limited to cancer The canonical Wnt-
signaling cascade refers to the transduction of series of signals mediated via the interaction of specific Wnt ligands with their target receptor resulting in the accumulation of β-catenin in the nucleus (MacDonald, Tamai et al. 2009). The
cytoplasmic stability of β-catenin is usually maintained at a minimal level by the
destruction complex composed of a sca ffold combination of tumour suppressor
protein adenomatous polyposis coli (APC), Axin2, casein kinase1 (CK1) and glycogen synthase kinase 3 β (GSK-3β) ,in the absence of Wnt ligand interaction
(OFF-state), the membrane receptor complex is inactivated, eventually results in β-
catenin ubiquitination and degradation (Anastas and Moon 2013). Deregulation of this
pathway has previously been described in he patoblastoma as well as in HCC. In liver
carcinogenesis, early deregulation of theWnt pathway occurs, leading to increased

3 
 proliferation, migration, andinvasiveness of cancer cells (Allegretta and Filmus
2011).Recent advances in biomedical research have allowed experimental and bioinformatics approaches to identify mi RNAs as regulators of components of the
Wnt-signaling pathways and vice versa. Thus, both miRNAs and Wnt-signaling
pathways form a network involved in the regulation of key biological processes. In this study we aimed to find the associati on between liver-related miRNAs with Wnt/
β-catenin proteins in liver tissue cells of HCC patients.
MATERIALS AND METHODS
Patients and Samples
Twenty-seven newly diagnosed primary HCC patients as well as 6 healthy
volunteers from liver donor prior transplantation, used as controls. Written informed
consents were obtained from patients and co ntrols and the study was approved by the
Institutional Review Board (IRB) of the National Cancer Institute, Cairo University. Liver biopsy was taken from each patient and control volunteer using fine needle
aspiration technique. Tissues were cut immediately into two parts; one was preserved
in 10% neutral buffered formalin for histopathological and immunohistochemcial
studies. The other part was collected in RNAlater solution (Qiagen, Germany) and stored at -80°C for miRNA analysis. The rele vant clinico-pathological characteristics
of the studied subjects are shown in table 1.
RNA isolation/cDNA synthesis/quantitative real-time PCR
Isolation of small RNA molecules including miRNAs from tissue biopsies
collected was performed using miRNeasy Mini kit (Qiagen). The concentration and
purity of total RNA was assessed us ing a Nano-Drop 2000 spectrophotometer

4 
 (Thermo Fisher Scientific technologies, Waltham, MA, USA). 1µg of total RNA was
reverse transcribed into c DNA in a final volume of 20µl using miScript Reverse
Transcription kit (Qiagen) according to the manufacturer's protocol. The relative
expression of 13 miRNAs was quantified by miScript SYBER Green PCR kit
(Qiagen) using miScript universal primer and quantiTech SYBER green PCR master
mix. Thirteen different miRNAs used in the current study are: miR-10a, miR-24, miR-30e, miR-99a, miR-106b, miR-122, miR-125b, miR-148a, miR-155, miR-183, miR-199a, miR-199a-3p and miR-215. A ll primers used are obtained from
Qiagen .The real-time PCR program was standardized to 95°C for 15 min. followed by
40 cycles at 94°C for 15 sec., 55°C for 30 sec. and 70°C for 30 sec. Each sample was examined in triplicate, and the relative expression values for miRNAs were normalized first using mean expression value of all miRNAs examined and calculated
using the 2 –delta Ct method (Mestdagh, Van Vlierberghe et al. 2009). Average fold change was calculated using 2 fold changes as cut off value. Hierarchical clustering
analysis and heatmap have been performed using GENE-E software (Broad Institute,
Inc.).
Immunohistochemistry (IHC) assay
IHC analysis for β–catenin, APC, c-myc, cyclin D1 and survivin proteinswas
performed on paraffin embedded tissue section from the 27 HCC cases and 6
controls.The Sections were countered stained and mounted then reviewed and scored under light microscope as described previ ously(Detre, Saclani Jotti et al. 1995). The
following primary antibodies were used: APC (Abcam, dilution 1:100); B-catenin
(Invitrogen, dilution 1:100); c-myc (Thermo Scientific, dilution 1:100); cyclin D1 (R&D system, dilution. 1:50); and survivin (Enzo life, dilution 1:40).

5 
 Statistical analysis
Univariate analysis was the analysis tool used consistently throughout the study. Chi
Square Trend statistical analysis was used to determine whether distributions were identical or not. For each parameter to be investigated hypothesis testing was used
where H0 "null hypothesis" = no difference between two distributions, and H1 "alternative" = a significant difference between two distributions. SPSS for windows
software was used to tabulate and graph results as well as calculate the P value
"probability"of the Chi Square Trend analysis . Trends were considered significant at
alpha <0.05.
RESULTS
Expression profile of miRNAs in HCC patients
Most of miRNAs (11 out of 13) showed a downregulation with different fold
change ranging from 3.35 for miR-24 up to 16.95 for miR-148a compared to normal subjects. Interestingly, miR-155 and miR- 183 showed significant upregulation in
HCC patients with 6.9 and 26.9 fold increases, respectively (p<0.001) compared to
normal subjects. The expression levels for all miRNAs are summarized in table 2 and
heatmap analysis was done using GENE-E software as shown in Figure 1.
Analysis of Wnt/ β-catenin signaling pathway in HCC patients
Tissue sections from human colon adenocarcinoma used as positive control for
β-catenin, APC and survivin whereas human breast carcinoma we employed as
positive control for c-myc and cyclin D1. Our results revealed a strong membranous
and pale cytoplasmic β-catenin staining with absence of nuclear expression in normal
hepatocytes. Interestingly, tumor cells in HCC sections, β-catenin showed subcellular

6 
 relocalization from cytoplasmic and membranous to nuclear and cytoplasmic
localization (Figure2a,b). Nuclear stai ning was observed in 22 cases of the 27
(81.5%) cases analyzed. In the 22 cases, 5 cases showed weak nuclear staining, 5
cases exhibited moderate staining, and strong signals were observed in 12 cases .For
APC protein expression, we found that all normal liver subjects exhibited cytoplasmic staining. The intensity of APC immunostaining and the number of immunoreactive cells were absent in tumorous epithelial cells of 14/27 (52%) patients. While, the
expression of APC was similar to normal cells in 2 patients (7.5%), decreased APC
expression was observed in 3 (11%) patients (moderate expression) and 8 (30%)
patients (weak expression) (Figure2c,d) . Strong nuclear c-myc expression was
detected in 16/27 (59%) patients , moderate nuclear c-myc expression was observed in 10/27 (37%) cases, while only one patie nt exhibited weak expression of c-myc
compared to normal liver tissues and positive control breast carcinoma tissues (Figure2e, f). Furthermore, we observed over expression of cyclin D1 in tumor
tissues (30% with strong and 37% with moderate overexpression) compared to normal tissue.Seven patients showed weak cyclin D1 protein expression,and only 2cases didn't show any expression of cyclin D1 (F igure 2g, h).The expression of survivin
protein was absent in all normal tissue liver sections analyzed. Except for one patient
who showed similarly no suvivin expression, as illustrated in (Figure 2i,j). HCC patients showed differential expression patte rns. Survivin was detected with weak (2
cases, 7.5%), moderate (10 cases, 37%) and marked intensity (11 cases, 41%). Noteworthy, 3 cases were not available for analysis for survivin due to insufficient
paraffin section samples.
Correlation between miRNA l evels and protein expressions

7 
 Our data revealed significant positive correlations between the level of APC
protein with each of miR-106b (p= 0.02), miR-125b (p=0.05), and miR-148a (p=0.04). β-catenin protein expression was positively correlated with the level of
miR-30e (p= 0.03) and miR-125b ( p=0.03). C-myc protein level showed also
significant positive correlation with miR-10a (p=0.04), miR-30e (p=0.004) and miR-215 (p=0.01). Unfortunately we could not find any correlation regarding the other miRNAs studied.
Correlation between miRNA expression levels /protein expression levels with several
clinico-pathological parameters
The present study indicated a signific ant correlation between miR-125b with
presence of HCV infection, being higher ex pressed in HCC/HCV+ patients compared
to HCC/HCV free (foldchanges were 0.17 vs 0.03, respectively; p=0.049). Furthermore, proteinexpression of surviv in detected in HCC/HCV+ patients was
higher compared to HCC/HCV free patients (p=0.05). High expression of miR-155
(12.8 fold change) in HCC associated with cirrhosis was observed compared to non
cirrhotic patients (3.43 fold change, p=0. 027). Finally, highly significant correlation
was observed in miR-215 level with tumor grades where the mean fold changes recorded in early grade (grade I) was 0.01 compared to its level in grade II & III(0.017 fold change p=0.006). No significant correlation was recorded with other
clinico-pathological factors analyzed. Figur e 3 (a,b,c) shows representative miRNAs
expression level with some clinico-pathological features.
DISCUSSION
The deregulation of the Wnt pathway is an early event in hepatocarcinogenesis
(Waisberg and Saba 2015). In this study we evaluate the crosstalk between Wnt/ β-

8 
 catenin signaling proteins and microRNAs in hepatocellular carcinoma patients. The
results from this study showed up regulation of miR-155and miR-183 compared to normal healthy tissue. Furthermore, miR-155 upregulation was found to be significantly
more frequent in HCC associated with cirrhos is, indicating its oncogeneic activity. This
finding is in consistent with previously pub lished studies with respect to describing
miR-155 and miR-183 as oncogenic miRNas were upregulated in several types of cancers ( Peng, Dong et al. 2015, Yuan, Li et al. 2015). However our data didn't exhibit
such significant correlation this might be attributed to the etiological, environmental and
genetic variations in addition to the small sample size used in the present study. Eleven
miRNAs were found to be significantly down regulated in HCC compared to normal liver tissues indicting a tumor suppression activity. MiR-10a and miR-30e and miR-215
were positively correlating with c-myc expression.MiR-30e and miR-125b were
positively correlating with β-catenin. These data indicate a possible mechanism for the
tumor suppression mediated by the indicated miRNAs, being transcription regulators of
different proteins of Wnt pathway. miR-10a interacts with the 5
,untranslated region of
mRNAs encoding ribosomal proteins to enhance their translation and may affect the ability of cells to undergo transformation(Orom , Nielsen et al. 2008). Still the role of
miR-30e played in HCC initiation or progression is not yet fully understood. Our
results demonstrated a significant down-r egulation of miR-30e in HCC patients
compared to healthy liver cases in Consistent with (Bhattacharya, Steele et al. 2016) .
Interestingly, we reported a significant correlation between MiR-30e and survivin in HCC patients. In-silico analyses revealed the involvement of miR-30e in regulating
autophagy pathway, explaining the suggested correlation between miR-30e and
Survivin which in turn was proven to be tightly related with autophagy (Shi, An et al. 2014). MiR-125b was found to be down-regulated in our results and positively

9 
 correlated with APC, supporting many literatures where it was found to be involved in
blocking Hepatocarcinogenesis (Li, Fang et al. 2015). Also, miR-125b was found to suppress cell growth, induce cell cycle arrest at G1 phase, and inhibit migration and
invasion of HCC cells (Liang, Wong et al. 2010).Furthermore, it was shown that miR-
125b suppresses EMT and EMT-associated traits of HCC including chemoresistance, migration, and stemness in HCC cells (Zhou, Lu et al. 2015). The dual correlation of
miR125b with both β-catenin and APC may because APC is not the sole factor to
determine the activity of β-catenin.In the current study, miR-148a expression was
reported to be significantly lower in HCC pa tients, reassuring what previously stated:
down expression of miR-148a associated with hepatocarcinogenesis and deterioration of
HCC (Pan, Huang et al. 2014). It was shown that down regulation of miR-148a in HCC yielding over-expression ofUbiquitin specific protease 4(USP4) in mesenchymal type
liver-tumor cells facilitating migration an d proliferation (Heo, Kim et al. 2014). The
significant correlation between miR-148a and APC protein in the present study might be
explained through the phosphatase activity of PTEN which is dispensable for the
regulation of APC and thereby contributed to hepatocarcinogenesis (Chu, Mo et al. 2014). Finally, down regulation of miR-215 reported in HCC of the present work was
consistent with several studies( Shimizu, Ta kehara et al. 2010, Gong, Zhang et al. 2013)
and its association with c-myc may be due to activating Wnt/ β-catenin through
increasing β-catenin phosphorylation(Tong, Liu et al. 2015). In conclusion, our data
suggest a crosstalk between the above indicated miRNAs and Wnt/ β-catenin signaling
proteins and their role in regulation of HCC progression. Better understand of this
crosstalk could be exploited for development new therapeutic agents through targeting
the canonical Wnt signaling pathway in HCC.

10 
 REFERENCES
Allegretta, M. and J. Filmus (2011). "Thera peutic potential of targeting glypican-3 in
hepatocellular carcinoma." Anticancer Agents Med Chem 11(6): 543-548.
Anastas, J. N. and R. T. Moon (2013). "WNT signalling pathways as therapeutic
targets in cancer." Nat Rev Cancer 13(1): 11-26.
Bhattacharya, S., R. Steele, S. Shrivastava, S. Chakraborty, A. M. Di Bisceglie and R.
B. Ray (2016). "Serum miR-30e and miR-223 as Novel Noninvasive Biomarkers for Hepatocellular Carcinoma." Am J Pathol
186(2): 242-247.
Block, T. M., A. S. Mehta, C. J. Fimmel and R. Jordan (2003). "Molecular viral
oncology of hepatocellular carcinoma." Oncogene 22(33): 5093-5107.
Chu, R., G. Mo, Z. Duan, M. Huang, J. Chang, X. Li and P. Liu (2014). "miRNAs
affect the development of hepatocellular carcinoma via dysregulation of their
biogenesis and expression." Cell Commun Signal 12: 45.
Detre, S., G. Saclani Jotti and M. Dowsett (1995). "A "quickscore" method for
immunohistochemical semiquantitation: va lidation for oestrogen receptor in
breast carcinomas." J Clin Pathol 48(9): 876-878.
Heo, M. J., Y. M. Kim, J. H. Koo, Y. M. Yang, J. An, S. K. Lee, S. J. Lee, K. M.
Kim, J. W. Park and S. G. Kim (2014). "microRNA-148a dysregulation discriminates poor prognosis of hepatocel lular carcinoma in association with
USP4 overexpression." Oncotarget
5(9): 2792-2806.
Li, J., L. Fang, W. Yu and Y. Wang (2015). "MicroRNA-125b suppresses the
migration and invasion of hepatocellular carcinoma cells by targeting transcriptional coactivator with PDZ-binding motif." Oncol Lett
9(4): 1971-1975.

11 
 Liang, L., C. M. Wong, Q. Ying, D. N. Fan, S. Huang, J. Ding, J. Yao, M. Yan, J. Li,
M. Yao, I. O. Ng and X. He (2010). "MicroRNA-125b suppressesed human liver cancer cell proliferation and metastasis by directly targeting oncogene LIN28B2."
Hepatology
52(5): 1731-1740.
MacDonald, B. T., K. Tamai and X. He (2009). "Wnt/beta-catenin signaling:
components, mechanisms, and diseases." Dev Cell 17(1): 9-26.
Mestdagh, P., P. Van Vlierberghe, A. De Weer, D. Muth, F. Westermann, F.
Speleman and J. Vandesompele (2009). "A novel and universal method for microRNA RT-qPCR data normalization." Genome Biol
10(6): R64.
Orom, U. A., F. C. Nielsen and A. H. Lund (2008). "MicroRNA-10a binds the 5'UTR
of ribosomal protein mRNAs and enhances their translation." Mol Cell 30(4):
460-471.
Pan, L., S. Huang, R. He, M. Rong, Y. Dang and G. Chen (2014). "Decreased
expression and clinical significance of miR-148a in hepatocellular carcinoma
tissues." Eur J Med Res 19: 68.
Peng, Y., W. Dong, T. X. Lin, G. Z. Zhong, B. Liao, B. Wang, P. Gu, L. Huang, Y.
Xie, F. D. Lu, X. Chen, W. B. Xie, W. He, S. X. Wu and J. Huang (2015). "MicroRNA-155 promotes bladder cancer growth by repressing the tumor suppressor DMTF1." Oncotarget
6(18): 16043-16058.
Shi, K., J. An, L. Shan, Q. Jiang, F. Li, Y. Ci, P. Wu, J. Duan, K. Hui, Y. Yang and C.
Xu (2014). "Survivin-2B promotes autophagy by accumulating IKK alpha in the
nucleus of selenite-treated NB4 cells." Cell Death Dis5: e1071.
Jie Sun, Haiqi Lu, Xian Wang, and Hongchuan Jin. MicroRNAs in Hepatocellular
Carcinoma: Regulation, Function, and Clin ical Implications. The ScientificWorld

12 
 Journal Volume 2013, Article ID 924206, 14 pages.
ttp://dx.doi.org/10.1155/2013/924206.
Tong, Y. Q., B. Liu, H. Y. Zheng, J. Gu, H. Liu, F. Li, B. H. Tan, M. Hartman, C.
Song and Y. Li (2015). "MiR-215, an activator of the CTNNBIP1/beta-catenin pathway, is a marker of poor prognosis in human glioma." Oncotarget
6(28):
25024-25033.
Waisberg, J. and G. T. Saba (2015). "Wnt-/-beta-catenin pathway signaling in human
hepatocellular carcinoma." World J Hepatol 7(26): 2631-2635.
Yuan, D., K. Li, K. Zhu, R. Yan and C. Dang (2015). "Plasma miR-183 predicts
recurrence and prognosis in patients with colorectal cancer." Cancer Biol
Ther 16(2): 268-275.
Zhou, G., M. Lu, J. Chen, C. Li, J. Zhang, J. Chen, X. Shi and S. Wu (2015).
"Identification of miR-199a-5p in serum as noninvasive biomarkers for detecting
and monitoring osteosarcoma." Tumour Biol 36(11): 8845-8852.

13 
 Legends for Figures
Figure1. A heatmap of 13 differenti ally expressed miRNAs in human
hepatocellular carcinoma (HCC). The expression of the indicated miRNAs was
analyzed using real-time qPCR. A heatmap was generated using the fold change of
each miRNA for HCC in comparison to the average of normal samples. Red:
upregulated miRNAs, green: downregula ted miRNAs. Relatedness in miRNA
expression across samples is shown by hierarchical tree on the Y axis.
Figure 2. Immunohistochemistry (IHC) staining of different Wnt/ β-catenin
proteins in paraffin-embedded HCC tumors. (A-B) Weak and strong cytoplasmic
and nuclear staining of β-catenin. (C-D) Moderate and strong cytoplasmic staining of
APC. (E-F) Moderate and strong nuclear staining of c-myc. (G-H) Weak and strong nuclear staining of cyclin D1. (I-J) Absent and strong cytoplasmic staining of surviving (DAB X400).
Figure 3 Correlations between expression levels of miRNAs and some clinico-
pathological factors in HCC patients. (A) miR-125b expression was significantly
higher in HCC patients infected with HCV (HCC/HCV
+) compared with HCV-
negative HCC patients (HCC/HCV-). (B) Expression levels of miR-155 was higher in
HCC patients showed cirrhosis than those with no signs for cirrhosis. (C) Expression levels of miR-215 was correlated with tumor grade, being upregulated expressed in
higher grades (grade II & III) compared to lo wer grade (grade I). Box plots describe
the relative expression of the indicated miRNAs. Shown are the means ±SD of three independent experiments. Statistical analysis was performed using the paired Student's t-test. *P<0.05

14 
 Table 1:Clinico-pathological features for primary hepatocellular carcinomapatients
Clinico-pathological features Primary HCC patients (N = 27)
Age (mean ± SEM)
Gender (M:F ratio)
AFP level (ng/ml)
ALT (IU/L) AST (IU/L)
HCV
Negative
Positive
Cirrhosis Negative
Positive
Grade
I
II III 60.70 ± 7.95 (45-75)
2.86 (20:7)
138.98 ± 25.72 (1-351)
88.48 ± 22.21 (10 – 609) 56.11 ± 16.21 (17 – 185)

6/27 (22%)
21/27 (78%)
17/27 (63%)
10/27 (37%)

4/27 (15%)
21/27 (78%) 2/27 (7%)
SEM: standard error of the means, M: male, F: female, AFP: alpha-
fetoprotein, ALT:Alanine-amin otransferase, AST: aspartate-
aminotransferase, IU: international unite, L: liter, N: number of patients
Table 2: List of tissue miRNA expres sion (average fold change) that are
significantly deregulated in primary HCC patients compared to control
subjects.
miRNA ID -log
2(fold changes) Up/down expression p-value
miR-10a 0.244 Down 0.043
miR-24 0.298 Down 0.027
miR-30e 0.183 Down 0.030
miR-99a 0.147 Down 0.012
miR-106b 0.268 Down 0.038
miR-122 0.101 Down 0.0003
miR-125b 0.184 Down 0.004
miR-148a 0.059 Down 0.001
miR-155 6.908 Up 0.017
miR-183 26.983 Up 0.042
miR-199a 0.363 Down 0.036
miR-199a-3p 0.117 Down 0.039
miR-215 0.157 Down 0.013

15 
 Figure 1:

F

Figure 2

16

17 
 Figure 3