Clinica Chimica Acta 485 (2018) 33–41

Contents lists available at ScienceDirect

Clinica Chimica Acta

journal homepage: www.elsevier.com/locate/cca

Review

MiR-139 in digestive system tumor diagnosis and detection: Bioinformatics T

and meta-analysis
⁎⁎ ⁎
Yu-Hui Wanga,1, Jia Jia,1, Hong Wengb, Bi-Cheng Wangc, , Fu-Bing Wanga,
a
Department of Laboratory Medicine, Zhongnan Hospital of Wuhan University, Wuhan 430071, PR China
b
Center for Evidence-Based and Translational Medicine, Zhongnan Hospital of Wuhan University, Wuhan 430071, PR China
c
Department of Pathology, Zhongnan Hospital of Wuhan University, Wuhan 430071, PR China

A R T I C LE I N FO A B S T R A C T

Keywords: Background: Accumulating evidence has indicated that microRNAs play important roles in the initiation and
miR-139 progression of digestive system tumors. However, previous studies suggest that the accuracy of miRNA detection
Digestive system tumors in digestive system tumors was inconsistent.
Bioinformatics Methods: The candidate miRNAs were obtained from The Cancer Genome Atlas (TCGA). Meta-analysis was
Meta-analysis
performed to evaluate the diagnostic value of these miRNAs in digestive system tumors. Furthermore, the po-
tential target genes of the miRNAs were predicted and assessed with functional analysis.
Results: According to TCGA data, miR-139 was a common biomarker of digestive system tumors. It was mark-
edly reduced in tumor tissues as compared with non-cancerous tissues in digestive system tumors. In the meta-
analysis, the pooled diagnostic odds ratio (DOR) and AUC was 57.51 (95% CI: 14.25–232.04) and 0.96 (95% CI:
0.94–0.97), respectively. Furthermore, the overall sensitivity and specificity was 0.89 (95% CI: 0.73–0.96) and
0.91 (95% CI: 0.75–0.97), respectively. The diagnostic value of tissue miR-139 was higher than the diagnostic
value of blood miR-139. In particular, miR-139 was a superior marker for distinguishing colorectal cancer.
Conclusion: miR-139 could be a potential biomarker for diagnosis of digestive system tumors especially color-
ectal cancer.

1. Introduction binding to the 3′ untranslated regions (UTRs) of mRNA transcripts to

Digestive system tumors, including colorectal cancer, gastric cancer,
liver cancer, esophagus cancer and pancreatic cancer, are common
malignancies causing cancer-related deaths worldwide [1]. According
to the latest statistics, digestive system tumors are the most frequently
diagnosed cancers. Colorectal carcinoma, stomach cancer, liver cancer
and esophageal cancer rank the third, fourth, fifth and seventh in new
cases of male malignancies, respectively, while they rank the second,
fifth and ninth new cases of female cancers, respectively [2]. Despite
the advances in diagnostic and therapeutic approaches, the optimal
treatment for digestive system tumors is still surgical resection. In ad-
dition, the difficulty in early diagnosis and the high recurrence of di-
gestive system tumors are responsible for their poor prognosis [3–5].
Therefore, accurate early diagnosis of digestive system tumors is in
urgent demand.
MicroRNAs (miRNAs) are small single-stranded RNA molecules with
∼22 nucleotides in length. They act in a combinatorial manner via
negatively regulate target gene expression. Thus, they play an im-
portant role in the initiation and progression of cancers [6–8]. miRNAs
are considered to be involved in tumorigenesis through modulating cell
cycle, DNA repair, proliferation, apoptosis, self-renewal and differ-
entiation of tumor cells [9–11]. For example, miR-21 is a classic re-
search hotspot in multiple cancers [12, 13]. miR-221 and miR-222 act
as oncogenic miRNAs by targeting and suppressing PTEN in various
solid tumors [14]. Meanwhile, accumulating evidence demonstrates
that SNP occurring in miR-196a2 and miR-499 potentially increases the
risk of urological cancers [11]. Using miRNAs as novel diagnostic bio-
markers and therapeutic targets will provide new opportunities for the
diagnosis and therapy of digestive system tumors [7, 15].
Microarray and high-throughput sequencing are emerging technol-
ogies that can extensively analyze genomic information. They are
broadly used to identify the relationship between microRNA and can-
cers [16–18]. The information globalization makes it possible for re-
searchers to search for significant differential gene expression in

⁎
Correspondence to: F.B. Wang, Department of Laboratory Medicine, Zhongnan Hospital of Wuhan University, No. 169 Donghu Road, Wuchang District, Wuhan 430071, PR China.
⁎⁎
Correspondence to: B.C. Wang, Department of Pathology, Zhongnan Hospital of Wuhan University, No. 169 Donghu Road, Wuchang District, Wuhan 430071, PR China.
E-mail addresses: wbcheng2006@126.com (B.-C. Wang), wfb20042002@sina.com (F.-B. Wang).
1
These authors contributed equally to this work.

https://doi.org/10.1016/j.cca.2018.06.006
Received 19 February 2018; Received in revised form 2 June 2018; Accepted 4 June 2018
Available online 05 June 2018
0009-8981/ © 2018 Elsevier B.V. All rights reserved.
Y.-H. Wang et al.
diseases through data mining from public databases such as The Cancer
Genome Atlas (TCGA) [19]. Subsequently, it is feasible to use bioin-
formatics analysis to predict dysregulation of miRNAs, potential target
genes and signal pathways in various disorders. Therefore, bioinfor-
matics analysis can be a novel tool to identify crucial miRNAs to direct
early diagnosis, prognosis and therapeutic design for cancer patients. It
is also powerful for understanding the molecular mechanism of miRNAs
in cancers [20].
In our study, a systematic expression profiling analysis was con-
ducted based on the gene expression data from TCGA on digestive
system tumors. miR-139, which was differentially expressed between
the tumor tissues and adjacent normal tissues, stood out to be a
common cancer biomarker. To further substantiate the significance of
miR-139, we performed an integrated meta-analysis to quantitatively
evaluate the effect of miR-139 on the diagnosis of digestive system
tumors. Additionally, bioinformatics analysis was also used to predict
the target genes of miR-139 as well as their potential roles in several
signal pathways. Furthermore, GO enrichment analysis and KEGG
analysis of target genes were conducted. Our results suggest that miR-
139 is a potential tumor biomarker for the screening and diagnosis of
digestive system tumors.
2. Materials and methods
2.1. TCGA data downloading and analysis
High throughput miRNA sequencing data in the project LIHC,
PAAD, ESCA, CHOL, STAD, COAD and READ were downloaded from
The Cancer Genome Atlas (TCGA) database. miRNA expression pro-
filing was conducted on 375 liver cancer tissues and 50 normal liver
tissues, 179 pancreatic cancer tissues and 4 normal pancreatic tissues,
187 esophageal cancer tissues and 13 normal esophageal tissues, 36
cholangiocarcinoma tissues and 9 normal bile ducts tissues, 446 sto-
mach cancer tissues and 45 normal stomach tissues, 619 colorectal
carcinoma tissues and 11 normal colorectal tissues, respectively.
Because the sample size of project CHOL is limited, we pooled data
from project CHOL and LIHC for analysis. The differential miRNA ex-
pression between each cancer and corresponding normal tissue was
determined using R package “edgeR” [21]. Benjamini-Hochberg
method was used to control the false discovery rate (FDR). Differen-
tially expressed miRNAs (DEmiRNAs) were identified using a threshold
of ︱LogFoldChange︱ > 1.5 and FDR < 0.05.
2.2. Literature search strategy
PubMed, Embase, Web of Science, the Cochrane Library, CNKI and
Wanfang database were thoroughly searched for studies (in English or
Chinese) related to the role of miR-139 in human digestive system tu-
mors until May 4, 2018. The keywords used in the search were “cancer”
or “tumor” or “neoplasm” together with “miR-139” or“microRNA-139”
or “miRNA-139” or “hsa-miR-139”. After the search, the published
papers were carefully reviewed and additional relevant references were
also included.
2.3. Screening criteria
Published papers were selected based on the following criteria: (1)
studies are about the diagnostic potential of miR-139 for digestive
system tumors; (2) cancer cases are confirmed by histological ex-
amination; (3) studies provide diagnostic indices such as sensitivity and
specificity, or sufficient information to calculate. Several published
papers were excluded because: (1) they are unrelated to the role of miR-
139 in the diagnosis of digestive system tumors; (2) there are no
available diagnostic test indices; (3) they are not human research; (4)
they are reviews, case reports, meta-analysis and letters; (5) they have
duplicate references.
34
Clinica Chimica Acta 485 (2018) 33–41
2.4. Data extraction and quality assessment
Two independent researchers (Yu-Hui Wang and Hong Weng) ex-
tracted the data from each eligible study for analysis, including the first
author, published papers date, country, cancer type, sample size, con-
trols, specimens, sensitivity and specificity of the detection. The third
reviewer (Jia Ji) joined the discussion to resolve differences.
Methodological quality of selected diagnostic studies was assessed by
the QUADAS-2 (Quality Assessment of Diagnostic Accuracy Studies 2)
[22].
2.5. Prediction and functional analysis of target genes
Target genes of miR-139 were predicted using two prediction da-
tabases including Pictar [23] (Available online: http://pictar.mdc-
berlin.de/) and miRDB [24] (Available online: http://www.mirdb.
org/), and were further validated in DIANA-TarBase v7.0 [25] data-
base with strong evidence including luciferase reporter assay, WB and
RT-PCR (Available online: http://diana.imis.athena-innovation.gr/
DianaTools/index.php?r=tarbase/index). Only genes that were con-
firmed by all three databases were considered target genes. To com-
prehensively study miR-139, GO [26] enrichment analysis and KEGG
[27] analysis were conducted on the target genes using an online tool
DAVID (https://david.ncifcrf.gov/, version 6.8). P-value < 0.05 was set
as the cut-off criterion.
2.6. Statistical analysis
R package “edgeR” of R language (version 3.2.5) was used to ana-
lyze the high-throughput data to determine the differential gene ex-
pression. Statistical analysis was performed using the STATA software
12.0 (Stata Corp, College Station, TX, USA) to calculate two-sided P-
values. In the meta-analysis, P-value < 0.05 was considered significant.
We used the bivariate meta-analysis model to estimate the pooled
sensitivity, specificity, positive likelihood ratio (PLR), negative like-
lihood ratio (NLR) and diagnostic odd ratio (DOR) with the 95% con-
fidence intervals (95% CIs). Meanwhile, we plotted the summary re-
ceiver operator characteristic (SROC) curve and calculated the area
under the curve (AUC) to evaluate the differential diagnosis power. The
heterogeneity of multiple studies was assessed by the Cochran's Q test
and the inconsistency index (I2) test [28]. P-value > 0.10 and I2
value < 50% suggest acceptable heterogeneity, and a fixed effects
model was adopted. In contrast, a random-effects model was adopted to
estimate the pooled results. To evaluate the sources of heterogeneity,
we further performed subgroup analysis and meta-regression. Further-
more, the publication bias analysis was conducted using Deeks' test.
3. Results
3.1. Hsa-miR-139 as a potential biomarker for the diagnosis of digestive
system tumors
Following the screening criteria, 183 DEmiRNAs between liver
cancer tissues and normal liver tissues, 6 DEmiRNAs between pan-
creatic cancer tissues and normal pancreatic tissues, 93 DEmiRNAs
between esophageal cancer tissues and normal esophageal tissues, 190
DEmiRNAs between stomach cancer tissues and normal stomach tis-
sues, 448 DEmiRNAs between colorectal carcinoma tissues and normal
colorectal tissues were obtained. Among these DEmiRNAs, miR-139
was the only common DEmiRNA in all five digestive system tumors.
Meanwhile, further analysis of TCGA data indicated that miR-139 was
markedly reduced in digestive system tumor tissues in comparison with
adjacent non-cancerous tissues (Fig. 1).
Y.-H. Wang et al. Clinica Chimica Acta 485 (2018) 33–41

Fig. 1. Identification of potential biomarkers for the diagnosis of digestive system tumors with Bioinformatics. A. Venn diagram of differentially expressed miRNAs
between tumor tissues and non-cancerous tissues in 5 digestive system tumors. B. Relative expression of miR-139 in 5 digestive system tumors based on TCGA data.

3.2. Characteristics of eligible studies
As shown in Fig. 2, a total of 803 eligible published papers were
found after the search. After a stringent review, 748 papers were ex-
cluded in the step of duplicate removed and title and abstract screening.
55 articles, which are closely related to miR-139 expression for the
diagnosis of digestive system tumors, were selected and further
screened for the eligibility. 51 articles were subsequently excluded due
to insufficient information about diagnostic test indices. Eventually, 4
eligible papers were chosen for our present meta-analysis [29–32].
These papers focus on colorectal cancer (n = 3) and hepatocellular
carcinoma (n = 1) with 223 cases and 182 controls in total. The blood
miR-139 was detected by Quantitative real-time polymerase chain re-
action (qRT-PCR) and TaqMan. Regarding the specimen types, blood
plasma was studied in three papers while blood serum was studied in
one paper. Three studies were done in China [29, 31, 32]. Four pub-
lished studies and TCGA data, with 2065 cases and 314 controls, were
enrolled in our meta-analysis from 2013 to 2017. The general char-
acteristics were presented in Table 1. Consequently, we assessed the
quality of studies along with the QUADAS-2 and showed them in Fig. 3.

Fig. 2. Flow diagram of the paper selection.

35
Y.-H. Wang et al. Clinica Chimica Acta 485 (2018) 33–41

Table 1
Characteristics of diagnostic studies and TCGA sets included in the meta-analysis.
First author Year Country Histology TNM Specimen Methods Case Control Sensitivity Specifcity
stage

Ng 2017 China Colorectal cancer I–IV Serum TaqMan 117 90 0.966 0.978
Yu 2014 China Colorectal cancer I–IV Plasma TaqMan 30 35 0.6 0.857
Li 2014 China Liver cancer I–III Plasma RT-qPCR 31 31 0.806 0.581
Kanaan 2013 America Colorectal cancer I–IV Plasma TaqMan 45 26 0.91 0.57
TCGA-colorectum 2016 America Colorectal cancer I–IV Tissue miRNA-Seq 619 11 0.998 1
TCGA-liver 2016 America Liver cancer I–IV Tissue miRNA-Seq 411 59 0.886 0.983
TCGA-stomach 2016 America Stomach cancer I–IV Tissue miRNA-Seq 446 45 0.866 0.911
TCGA-pancreas 2016 America Pancreatic cancer I–IV Tissue miRNA-Seq 179 4 0.531 0.75
TCGA-esophagus 2016 America Esophageal squamous cell carcinoma I–IV Tissue miRNA-Seq 187 13 0.813 0.923

3.3. Diagnostic accuracy of miR-139 in screening digestive system tumors
We analyzed the pooled sensitivity and specificity of all eligible
published papers. Because of the significant heterogeneity
(I2 = 97.57% for sensitivity and I2 = 89.86% for specificity), the
random-effects model was used to calculate the sensitivity and speci-
ficity. As shown in forest plots (Fig. 4A), the overall sensitivity and
specificity were 0.89 (95% CI: 0.73–0.96) and 0.91 (95% CI:
0.75–0.97), respectively. Additionally, the pooled Positive Likelihood
Ratio (PLR) was 10.19 (95% CI: 3.02–34.45), the Negative Likelihood
Ratio (NLR) was 0.12 (95% CI: 0.04–0.34), and the diagnostic odds
ratio (DOR) was 57.51 (95% CI: 14.25–232.04). The area under the
curve (AUC) was 0.96 (95% CI: 0.94–0.97) according to the corre-
sponding overall summary receiver operator characteristic (SROC)
curve (Fig. 5A). These results indicated that the diagnostic accuracy of
miR-139 in screening digestive system tumors was relatively high.

Fig. 3. Methodological quality of the study on miR-139 for the diagnosis of digestive system tumors.

36
Y.-H. Wang et al.
Fig. 4. Summary estimates of the sensitivity and specificity with forest plots analysis. A. Forest plots for miR-139 in digestive system tumors; B. Forest plots for miR-
139 in colorectal carcinoma; C. Forest plots for miR-139 in tumor tissues; D. Forest plots for miR-139 in blood.
3.4. Diagnostic efficacy of miR-139 in screening colorectal cancer
Colorectal cancer was the most commonly diagnosed cancer in the
papers that were used to evaluate the diagnostic efficacy of miR-139.
Therefore, we further analyzed the significance of miR-139 in distin-
guishing colorectal cancer patients from healthy individuals. The
pooled sensitivity and specificity were 0.96 (95% CI: 0.72–1.00) and
0.94 (95% CI: 0.62–0.99), respectively, with the significant hetero-
geneity (I2 = 97.52% and 96.5%) (Fig. 4B). The PLR, NLR and DOR
were 16.12 (95% CI: 1.77–146.86), 0.04 (95% CI 0.00–0.39) and
149.33 (95% CI: 9.22–2418.70), respectively. The AUC for colorectal
cancer was 0.99 (95% CI: 0.97–0.99) (Fig. 5B).
3.5. Pooled effect of tissue miR-139 or blood miR-139 on the diagnosis of
digestive system tumors
The sensitivity and specificity in the tissue group were 0.91 (95% CI:
0.63–0.98) and 0.97 (95% CI: 0.75–1.00), respectively (Fig. 4C). In
addition, the PLR and NLR were 28.7 (95% CI: 2.60–316.8) and 0.09
(95% CI: 0.02–0. 50), respectively. The heterogeneity was significant in
the sensitivity and specificity data (I2 = 99.02 and I2 = 64.13), we used
the random-effects model. The DOR was 309 (95% CI, 6.00–16,647).
The AUC of the corresponding SROC curve was 0.98 (95% CI:
0.97–0.99), indicating that tissue miR-139 had a relatively high accu-
racy in the diagnosis (Fig. 5C). With regard to the diagnosis using blood
miR-139, the sensitivity, specificity, PLR, NLR and DOR were 0.87
(95% CI: 0.68–0.95), 0.82 (95% CI: 0.53–0.95), 4.9 (95% CI: 1.4–16.8),
0.16 (95% CI: 0.05–0.48) and 31 (95% CI: 4–261), respectively
(Fig. 4D). Moreover, the AUC was 0.92 (95%CI: 0.89–0.94) (Fig. 5D).
37
Clinica Chimica Acta 485 (2018) 33–41
3.6. Meta-regression analysis
We performed meta-regression to further explore the potential
source of heterogeneity (Fig. 6). The meta-regression explored the
characteristics of each study including “Ethnicity (Asian or not)”,
“Control (Healthy control or not)”, “Sample (blood or not)” and
“Method (miRNA-Seq or not)”. Our result suggested that the control
covariate was responsible for the heterogeneity in the specificity.
3.7. Publication bias
To assess potential publication bias of selected papers, the Deeks'
funnel plot asymmetry test was conducted. No significant publication
bias was found in the pooled analysis (p = 0.18) (Fig. 7), indicating no
significant publication bias in our meta-analysis.
3.8. Target genes of miR-139 and the functional and pathway enrichment
To provide a precise and deep insight into the role of miR-139 in the
initiation and progression of digestive system tumors, the target genes
of miR-139 were predicted using miRNA target gene prediction data-
bases. As a result, a total of 30 potential target genes of miR-139 were
found (Table 2).
The functional and pathway enrichment analysis of miR-139 was
performed using the DAVID website (https://david.ncifcrf.gov/, ver-
sion 6.8). The terms of cellular component (CC) and molecular function
(MF) were nucleus (GO: 0005634) and DNA binding (GO: 0003677),
respectively. The enriched biological process (BP) term was positive
regulation of translational initiation (GO: 0045948). However, the
KEGG analysis didn't find an enriched signal pathway (Table 3).
Y.-H. Wang et al.
Fig. 5. SROC analysis of the diagnostic performance of miR-139. A. SROC curves for miR-139 in digestive system tumors; B. SROC curves for miR-139 in colorectal
carcinoma; C. SROC curves for miR-139 in tumor tissues; D. SROC curves for miR-139 in blood.
4. Discussion
Recently, several public databases are increasingly used for studies
on novel biomarkers for the diagnosis and prognosis of cancers. In the
current study, we focused on finding the differential expression of
miRNAs in digestive system tumors for the diagnostic purpose.
According to TCGA data, the expression of miRNA-139 was sig-
nificantly different between the tumor tissues of 5 types of digestive
system tumors and the normal tissues. Therefore, miRNA-139 could be
a biomarker that facilitates the diagnosis of digestive system tumors.
However, a comprehensive and systemic analysis of the clinical sig-
nificance of miR-139 in digestive system tumors is still needed. The
biological function of miR-139 is still unknown.
miR-139, which is a novel miRNA found in various types of diges-
tive system tumors, is also reported to be dysregulated in other tumors
such as cervical cancer [33], bladder cancer [34], and leukemia [35]. A
recent research showed that miR-139 reduced ROCK2 expression to
cause the metastatic growth of hepatocellular carcinoma [36]. In ad-
dition, miR-139 inhibited the IGF-IR/MEK/ERK signaling, and resulted
38
Clinica Chimica Acta 485 (2018) 33–41
in the suppression of MMP2 promoter in colorectal cancer [37]. An-
other study reported that miR-139 inhibited the proliferation and me-
tastasis of colon cancer cells, and promoted the apoptosis of colon
cancer cells via targeting NOTCH1 [38]. Moreover, inhibition of miR-
139 expression by LINC00152 contributed to the tumorigenesis, pro-
gression and chemoresistance of colorectal cancer [39]. Interestingly, in
gastric cancer cells, interaction of HER2 with CD44 induced the epi-
genetic silencing of miR-139 to increase CXCR4 expression. The de-
crease in miR-139 expression is associated with lymph node metastasis
of gastric tumors [40]. What's more, miR-139 might suppress the
growth of colorectal cancer by directly targeting RAP1B [41]. In this
study, we predicted the target genes of miR-139, and found that these
target genes were closely related to the positive regulation of transla-
tional initiation. Hence, miR-139 is a vital oncogene participating in
tumorigenesis. It could be considered a novel biomarker for the diag-
nosis of digestive system tumors. We performed a detailed meta-ana-
lysis to assess the diagnostic value of miR-139 in different patients with
digestive system tumors.
In this study, the pooled sensitivity and specificity were 0.89 (95%
Y.-H. Wang et al.
Fig. 6. Forest plots of multivariable meta-regression and subgroup analysis of
the sensitivity and specificity of miR-139 in digestive system tumors.
CI: 0.73–0.96) and 0.91 (95% CI: 0.75–0.97), respectively. In compar-
ison with the same parameters of the traditional blood-borne bio-
markers, they were both much higher. Additionally, the PLR was 10.19
(95% CI: 3.02–34.45) and the DOR of 57.51 (95% CI: 14.25–232.04),
Fig. 7. Deeks' funnel plot for the assessment of the publication bias.
39
Clinica Chimica Acta 485 (2018) 33–41
indicating that the probability of diagnosis was significantly increased.
The AUC was 0.96, suggesting that the diagnostic performance of miR-
139 was excellent. Due to the high incidence of colorectal cancer, we
analyzed the diagnostic efficacy of miR-139 for colorectal cancer.
Interestingly, the AUC for colorectal cancer was 0.99 (95% CI:
0.97–0.99), and miR-139 showed a superior performance in distin-
guishing colorectal cancer from controls. Importantly, the sensitivity,
specificity, and pooled AUC of tissue miR-139 for the diagnosis of di-
gestive system tumors were much higher than those of blood miR-139.
Our data thus suggest that miR-139, especially miR-139 in the tumor
tissue, is an effective biomarker for the diagnosis of digestive system
tumors. However, it is also possible that samples from different patients
and the evaluation methods caused the distinctive expression patterns
in the tissue and blood. In order to search the sources of the hetero-
geneity, we performed subgroup analysis and meta-regression analysis.
The results showed that the heterogeneity mainly resulted from dif-
ferent sources of controls.
To our knowledge, we are the first to comprehensively analyze the
role of miR-139 in the diagnosis of digestive system tumors using TCGA
data and published papers. Our study also indicated that the target
genes of miR-139 positively regulated translational initiation through
DNA binding in the nucleus. However, it is noteworthy that there are
some limitations in our study. For example, the statistical power was
reduced because of the small sample size of eligible studies.
Additionally, the existence of a severe threshold effect in TCGA data
might impair the accuracy of the result regarding tissue miR-139. What
is more, the publication bias might still exist, especially when most of
the selected papers in our study focus on Chinese population and the
generalizability of our study is limited. Finally, because target genes
were not enough, we can't predict relevant signaling pathways they
enriched in.
In conclusion, our study demonstrates that miR-139 might be a
useful biomarker for the diagnosis of colorectal cancer. A large-scale
and multicenter study will be needed to confirm our results and com-
pletely disclose the functions of miR-139 in colorectal cancer.
Y.-H. Wang et al.
Table 2
Targets of miR-139 in data mining.
miRNA Target genes
hsa-miR-139 AKIRIN2,ASH1L,ATP5G3,AZIN1,C3orf70,DDX3X,DYNLL2,EIF4G2,HIPK1,HNRNPF, JUN,KBTBD2,KLHL28,MGAT4A,NPTX1,PDE3A,PMP22,REV1,SCAPER,TBX1,
TMEM33,TMED10,TMF1,TOP1,TSPAN3, UHMK1,USP6NL,USP24, ZBTB34,ZNF792,
Table 3
GO functional annotation for the most significantly enriched targeted genes of miR-139.
Type Term
GOTERM_BP_DIRECT GO:0045948~positive regulation of translational
initiation
GOTERM_CC_DIRECT GO:0005634~nucleus
GOTERM_MF_DIRECT GO:0003677~DNA binding
Conflict of interest
The authors declare no conflict of interest.
Acknowledgments
This work was supported by Applied Basic Research Program of
Science and Technology Bureau Foundation of Wuhan (No.
2016060101010054), and National Natural Science Foundation of
China (No. 81672114). This work was also funded by Science and
Technology Innovation Fostering Foundation of Zhongnan Hospital of
Wuhan University (cxpy20160025) and Wuhan City health and family
planning medical talented youth development project.
References
[1] R.L. Siegel, K.D. Miller, A. Jemal, Cancer statistics, 2015, CA Cancer J. Clin. 65 (1)
(2015) 5–29.
[2] L.A. Torre, F. Bray, R.L. Siegel, J. Ferlay, J. Lortet-Tieulent, A. Jemal, Global cancer
statistics, 2012, CA Cancer J. Clin. 65 (2) (2015) 87–108.
[3] H.H. Wu, W.C. Lin, K.W. Tsai, Advances in molecular biomarkers for gastric cancer:
miRNAs as emerging novel cancer markers, Expert Rev. Mol. Med. 16 (2014) e1.
[4] H. Brenner, M. Kloor, C.P. Pox, Colorectal cancer, Lancet (London, England) 383
(9927) (2014) 1490–1502.
[5] A. Forner, J.M. Llovet, J. Bruix, Hepatocellular carcinoma, Lancet (London,
England) 379 (9822) (2012) 1245–1255.
[6] A. Salameh, A.K. Lee, M. Cardó-Vila, D.N. Nunes, E. Efstathiou, F.I. Staquicini,
A.S. Dobroff, S. Marchiò, N.M. Navone, H. Hosoya, R.C. Lauer, S. Wen,
C.C. Salmeron, A. Hoang, I. Newsham, L.A. Lima, D.M. Carraro, S. Oliviero,
M.G. Kolonin, R.L. Sidman, K.A. Do, P. Troncoso, C.J. Logothetis, R.R. Brentani,
G.A. Calin, W.K. Cavenee, E. Dias-Neto, R. Pasqualini, W. Arap, PRUNE2 is a human
prostate cancer suppressor regulated by the intronic long noncoding RNA PCA3,
Proc. Natl. Acad. Sci. U. S. A. 112 (27) (2015) 8403–8408.
[7] N. Bushati, S.M. Cohen, microRNA functions, Annu. Rev. Cell Dev. Biol. 23 (2007)
175–205.
[8] B.P. Lewis, I.H. Shih, M.W. Jones-Rhoades, D.P. Bartel, C.B. Burge, Prediction of
mammalian microRNA targets, Cell 115 (7) (2003) 787–798.
[9] M. Carleton, M.A. Cleary, P.S. Linsley, MicroRNAs and cell cycle regulation, Cell
Cycle (Georgetown, Tex.) 6 (17) (2007) 2127–2132.
[10] M. Boehm, F.J. Slack, MicroRNA control of lifespan and metabolism, Cell Cycle
(Georgetown, Tex.) 5 (8) (2006) 837–840.
[11] Y.H. Wang, H.N. Hu, H. Weng, H. Chen, C.L. Luo, J. Ji, C.Q. Yin, C.H. Yuan,
F.B. Wang, Association between polymorphisms in MicroRNAs and risk of ur-
ological Cancer: a meta-analysis based on 17,019 subjects, Front. Physiol. 8 (2017)
325.
[12] F. Meng, R. Henson, H. Wehbe-Janek, K. Ghoshal, S.T. Jacob, T. Patel, MicroRNA-
21 regulates expression of the PTEN tumor suppressor gene in human hepatocel-
lular cancer, Gastroenterology 133 (2) (2007) 647–658.
[13] I.A. Asangani, S.A. Rasheed, D.A. Nikolova, J.H. Leupold, N.H. Colburn, S. Post,
H. Allgayer, MicroRNA-21 (miR-21) post-transcriptionally downregulates tumor
suppressor Pdcd4 and stimulates invasion, intravasation and metastasis in color-
ectal cancer, Oncogene 27 (15) (2008) 2128–2136.
[14] M. Garofalo, C. Quintavalle, G. Romano, C.M. Croce, G. Condorelli, miR221/222 in
cancer: their role in tumor progression and response to therapy, Curr. Mol. Med. 12
(1) (2012) 27–33.
[15] I. Berindan-Neagoe, C. Monroig Pdel, B. Pasculli, G.A. Calin, MicroRNAome
genome: a treasure for cancer diagnosis and therapy, CA Cancer J. Clin. 64 (5)
(2014) 311–336.
Clinica Chimica Acta 485 (2018) 33–41
Gene
DDX3X, UHMK1
AKIRIN2, ASH1L, AZIN1, DDX3X, DYNLL2, HIPK1, HNRNPF, JUN, REV1, TBX1, TMF1,
TOP1, SCAPER, UHMK1, ZNF792,
ASH1L, DDX3X, HIPK1, JUN, TOP1, TBX1, TMF1, ZBTB34,
[16] S.S. Thorgeirsson, J.S. Lee, J.W. Grisham, Functional genomics of hepatocellular
carcinoma, Hepatology (Baltimore, Md.) 43 (2 Suppl 1) (2006) S145–S150.
[17] N. Rosenfeld, R. Aharonov, E. Meiri, S. Rosenwald, Y. Spector, M. Zepeniuk,
H. Benjamin, N. Shabes, S. Tabak, A. Levy, D. Lebanony, Y. Goren, E. Silberschein,
N. Targan, A. Ben-Ari, S. Gilad, N. Sion-Vardy, A. Tobar, M. Feinmesser,
O. Kharenko, O. Nativ, D. Nass, M. Perelman, A. Yosepovich, B. Shalmon, S. Polak-
Charcon, E. Fridman, A. Avniel, I. Bentwich, Z. Bentwich, D. Cohen, A. Chajut,
I. Barshack, MicroRNAs accurately identify cancer tissue origin, Nat. Biotechnol. 26
(4) (2008) 462–469.
[18] A.J. Schetter, S.Y. Leung, J.J. Sohn, K.A. Zanetti, E.D. Bowman, N. Yanaihara,
S.T. Yuen, T.L. Chan, D.L. Kwong, G.K. Au, C.G. Liu, G.A. Calin, C.M. Croce,
C.C. Harris, MicroRNA expression profiles associated with prognosis and ther-
apeutic outcome in colon adenocarcinoma, JAMA 299 (4) (2008) 425–436.
[19] H. Lee, J. Palm, S.M. Grimes, H.P. Ji, The Cancer genome atlas clinical explorer: a
web and mobile interface for identifying clinical-genomic driver associations,
Genome Med. 7 (2015) 112.
[20] L. Cascione, A. Ferro, R. Giugno, A. Lagana, G. Pigola, A. Pulvirenti, D. Veneziano,
Elucidating the role of microRNAs in cancer through data mining techniques, Adv.
Exp. Med. Biol. 774 (2013) 291–315.
[21] M.D. Robinson, D.J. McCarthy, G.K. Smyth, edgeR: a Bioconductor package for
differential expression analysis of digital gene expression data, Bioinformatics 26
(1) (2010) 139–140.
[22] P.F. Whiting, A.W. Rutjes, M.E. Westwood, S. Mallett, J.J. Deeks, J.B. Reitsma,
M.M. Leeflang, J.A. Sterne, P.M. Bossuyt, QUADAS-2: a revised tool for the quality
assessment of diagnostic accuracy studies, Ann. Intern. Med. 155 (8) (2011)
529–536.
[23] A. Krek, D. Grun, M.N. Poy, R. Wolf, L. Rosenberg, E.J. Epstein, P. MacMenamin,
I. da Piedade, K.C. Gunsalus, M. Stoffel, N. Rajewsky, Combinatorial microRNA
target predictions, Nat. Genet. 37 (5) (2005) 495–500.
[24] N. Wong, X. Wang, miRDB: an online resource for microRNA target prediction and
functional annotations, Nucleic Acids Res. 43 (2015) D146–D152 (Database issue).
[25] I.S. Vlachos, M.D. Paraskevopoulou, D. Karagkouni, G. Georgakilas, T. Vergoulis,
I. Kanellos, I.L. Anastasopoulos, S. Maniou, K. Karathanou, D. Kalfakakou,
A. Fevgas, T. Dalamagas, A.G. Hatzigeorgiou, DIANA-TarBase v7.0: indexing more
than half a million experimentally supported miRNA:mRNA interactions, Nucleic
Acids Res. 43 (2015) D153–D159 (Database issue).
[26] M. Ashburner, C.A. Ball, J.A. Blake, D. Botstein, H. Butler, J.M. Cherry, A.P. Davis,
K. Dolinski, S.S. Dwight, J.T. Eppig, M.A. Harris, D.P. Hill, L. Issel-Tarver,
A. Kasarskis, S. Lewis, J.C. Matese, J.E. Richardson, M. Ringwald, G.M. Rubin,
G. Sherlock, Gene ontology: tool for the unification of biology. The Gene ontology
consortium, Nat. Genet. 25 (1) (2000) 25–29.
[27] H. Ogata, S. Goto, K. Sato, W. Fujibuchi, H. Bono, M. Kanehisa, KEGG: Kyoto en-
cyclopedia of genes and genomes, Nucleic Acids Res. 27 (1) (1999) 29–34.
[28] J.P. Higgins, Commentary: heterogeneity in meta-analysis should be expected and
appropriately quantified, Int. J. Epidemiol. 37 (5) (2008) 1158–1160.
[29] W.F. Yu, S.S. Han, Y.J. Yang, B. Liu, C.J. Zhai, L.J. Zhang, Z.R. Zhao, Expression and
clinical significance of microRNA-139-5p in colorectal cancer, Guang Dong Yi Xue
(7) (2014) 1042–1044.
[30] Z. Kanaan, H. Roberts, M.R. Eichenberger, A. Billeter, G. Ocheretner, J. Pan,
S.N. Rai, J. Jorden, A. Williford, S. Galandiuk, A plasma microRNA panel for de-
tection of colorectal adenomas: a step toward more precise screening for colorectal
cancer, Ann. Surg. 258 (3) (2013) 400–408.
[31] T. Li, J. Yin, L. Yuan, S. Wang, L. Yang, X. Du, J. Lu, Downregulation of microRNA-
139 is associated with hepatocellular carcinoma risk and short-term survival, Oncol.
Rep. 31 (4) (2014) 1699–1706.
[32] L. Ng, T.M.-H. Wan, J.H.-W. Man, A.K.-M. Chow, D. Iyer, G. Chen, T.C.-C. Yau, O.S.-
H. Lo, D.C.-C. Foo, J.T.-C. Poon, W.-K. Leung, R.W.-C. Pang, W.-L. Law,
Identification of serum miR-139-3p as a non-invasive biomarker for colorectal
cancer, Oncotarget 8 (16) (2017) 27393–27400.
[33] M.K. Sannigrahi, R. Sharma, V. Singh, N.K. Panda, V. Rattan, M. Khullar, Role of
host miRNA Hsa-miR-139-3p in HPV-16-induced carcinomas, Clin. Cancer Res. 23
(14) (2017) 3884–3895.
[34] M. Yonemori, N. Seki, H. Yoshino, R. Matsushita, K. Miyamoto, M. Nakagawa,
40
Y.-H. Wang et al.
H. Enokida, Dual tumor-suppressors miR-139-5p and miR-139-3p targeting matrix
metalloprotease 11 in bladder cancer, Science (New York, N.Y.) 107 (9) (2016)
1233–1242.
[35] J. Choi, Y.K. Kim, K. Park, J. Nah, S.S. Yoon, D.W. Kim, V.N. Kim, R.H. Seong,
MicroRNA-139-5p regulates proliferation of hematopoietic progenitors and is re-
pressed during BCR-ABL-mediated leukemogenesis, Blood 128 (17) (2016)
2117–2129.
[36] C.C. Wong, C.M. Wong, E.K. Tung, L. Au, J.M. Lee, R.T. Poon, K. Man, I.O. Ng, The
microRNA miR-139 suppresses metastasis and progression of hepatocellular carci-
noma by down-regulating rho-kinase 2, Gastroenterology 140 (1) (2011) 322–331.
[37] K. Shen, Q. Liang, K. Xu, D. Cui, L. Jiang, P. Yin, Y. Lu, Q. Li, J. Liu, MiR-139
inhibits invasion and metastasis of colorectal cancer by targeting the type I insulin-
like growth factor receptor, Biochem. Pharmacol. 84 (3) (2012) 320–330.
[38] L. Zhang, Y. Dong, N. Zhu, H. Tsoi, Z. Zhao, C.W. Wu, K. Wang, S. Zheng, S.S. Ng,
41
Clinica Chimica Acta 485 (2018) 33–41
F.K. Chan, J.J. Sung, J. Yu, microRNA-139-5p exerts tumor suppressor function by
targeting NOTCH1 in colorectal cancer, Mol. Cancer 13 (2014) 124.
[39] Z. Bian, J. Zhang, M. Li, Y. Feng, S. Yao, M. Song, X. Qi, B. Fei, Y. Yin, D. Hua,
Z. Huang, Long non-coding RNA LINC00152 promotes cell proliferation, metastasis,
and confers 5-FU resistance in colorectal cancer by inhibiting miR-139-5p,
Oncogene 6 (11) (2017) 1.
[40] W. Bao, H.J. Fu, Q.S. Xie, L. Wang, R. Zhang, Z.Y. Guo, J. Zhao, Y.L. Meng, X.L. Ren,
T. Wang, Q. Li, B.Q. Jin, L.B. Yao, R.A. Wang, D.M. Fan, S.Y. Chen, L.T. Jia,
A.G. Yang, HER2 interacts with CD44 to up-regulate CXCR4 via epigenetic silencing
of microRNA-139 in gastric cancer cells, Gastroenterology 141 (6) (2011)
2076–2087.e6.
[41] H. Guo, X. Hu, S. Ge, G. Qian, J. Zhang, Regulation of RAP1B by miR-139 sup-
presses human colorectal carcinoma cell proliferation, Int. J. Biochem. Cell Biol. 44
(9) (2012) 1465–1472.