MicroRNA expression profiling of nasopharyngeal carcinoma

Article (PDF Available)inOncology Reports 25(5):1353-63 · March 2011with14 Reads
DOI: 10.3892/or.2011.1204 · Source: PubMed
Nasopharyngeal carcinoma (NPC) is posing a serious health problem worldwide. The association between its pathogenesis and microRNAs (miRNA) has not been elucidated. In this study, miRNA expression profiling was performed to screen the miRNA expression changes in 8 NPC tissues and 4 normal nasopharyngeal tissues. Thirty-four miRNAs were identified to be differentially expressed; of these, one miRNA (miR-18a) was overexpressed and 33 miRNAs (miR-34b, miR-34c, let-7 family, etc.) were underexpressed in NPC tissues compared to the normal samples. Validation was performed by real-time quantitative PCR for two altered miRNAs (miR-34b and let-7g) and one non-differentially expressed miRNA (miR-30c). Unsupervised hierarchical clustering analysis showed that the aberrant miRNAs were correlated with the clinical stage of NPC patients. In addition to several biological pathways that are well characterised in NPC and which were significantly targeted by the underexpressed miRNAs, two novel pathways, nervous system development and sensory perception of sound, were identified to be strongly associated with NPC development. Furthermore, a c-Myc centered miRNA regulatory network was inferred in NPC. Our study reveals that aberrantly expressed miRNAs play important roles in NPC tumorigenesis and may serve as potential targets for novel therapeutic strategies in the future.
ONCOLOGY REPORTS 25: 1353-1363, 2011
Abstract. Nasopharyngeal carcinoma (NPC) is posing a
serious health problem worldwide. The association between
its pathogenesis and microRNAs (miRNA) has not been
elucidated. In this study, miRNA expression proling was
performed to screen the miRNA expression changes in 8
NPC tissues and 4 normal nasopharyngeal tissues. Thirty-
four miRNAs were identied to be differentially expressed;
of these, one miRNA (miR-18a) was overexpressed and 33
miRNAs (miR-34b, miR-34c, let-7 family, etc.) were under-
expressed in NPC tissues compared to the normal samples.
Validation was performed by real-time quantitative PCR
for two altered miRNAs (miR-34b and let-7g) and one non-
differentially expressed miRNA (miR-30c). Unsupervised
hierarchical clustering analysis showed that the aberrant
miRNAs were correlated with the clinical stage of NPC
patients. In addition to several biological pathways that are
well characterised in NPC and which were significantly
targeted by the underexpressed miRNAs, two novel path-
ways, nervous system development and sensory perception
of sound, were identied to be strongly associated with NPC
development. Furthermore, a c-Myc centered miRNA regula-
tory network was inferred in NPC. Our study reveals that
aberrantly expressed miRNAs play important roles in NPC
tumorigenesis and may serve as potential targets for novel
therapeutic strategies in the future.
MicroRNAs (miRNAs) are a newly identied class of short
(about 19-22 nt), endogenous, non-coding, single-stranded
RNA molecules, which are conserved in sequence between
distantly related organisms. More and more evidence has
proved miRNAs play important roles in diverse biological
functions, including development, differentiation, proliferation
and apoptosis (1,2). MiRNAs regulate protein expression by
binding to the 3' untranslated regions (3' UTR) of messenger
RNAs (mRNAs) and suppressing translation or inducing
degradation or deadenylation with varying degrees of sequence
complementarity (3). In humans, strong links between cancer
and miRNA deregulation have been suggested by recent
studies. The relevance of miRNAs to cancer was rst reported
in 2002 when Calin et al observed that miR-15a and miR-16-1
were down-regulated or even deleted in the majority of
chronic lymphocytic leukemia patients (4). They also found
that many known miRNAs are located in the regions of
human chromosomes with high frequencies of copy number
alterations in cancers (4). So far, distinct miRNA profiles
have been reported and the potential role of miRNAs in
various human haematological and solid cancers have been
indicated by many studies. Lu et al used miRNA expressions
to classify human cancers associated with the developmental
lineage and differentiation state (5). Johnson et al reported
that the down-regulation of let-7 is a signicant cause of lung
cancer tumorigenesis (6). This body of evidence indicates that
the aberrations of miRNA expression in cancer are strongly
correlated with tumor carcinogenesis and progression.
Nasopharyngeal carcinoma (NPC) is a non-lymphomatous,
squamous cell malignancy arising from the epithelial cells
lining of the nasopharynx (7). NPC was vastly common
in Southeast Asia, particularly amongst the Cantonese
population of southern China, such as Guangdong Province
and Guangxi Province. The incidence is ~30-80/100,000
people per year and has remained high for decades. Because
of the high risk and incidence region specicity, it is important
to elucidate the mechanism of NPC carcinogenesis. Thus far,
little research has been conducted to investigate the role
of miRNAs in NPC carcinogenesis. Zhang et al proposed
that the interactions between miR-141 and tumor-related
genes c-Myc, Splunc1, Brd3, Ubap1, and Pten contribute to
the progression of NPC (8). Xia et al found that miR-200a
targeting ZEB2 and CTNNB1 inhibits NPC cell growth,
microRNA expression proling of nasopharyngeal carcinoma
and YAO LI
State Key Laboratory of Genetic Engineering, Institute of Genetics, School of Life Science,
Fudan University, Shanghai;
Institute of Nasopharyngeal Carcinoma, The People's Hospital
of Guangxi Zhuang Nationality Autonomous Region, Nanning, P.R. China
Received November 2, 2010; Accepted January 17, 2011
DOI: 10.3892/or.2011.1204
Correspondence to: Professor Yao Li, State Key Laboratory of
Genetic Engineering, Institute of Genetics, School of Life Science,
Fudan University, Shanghai, P.R. China
E-mail: yaoli@fudan.edu.cn
Professor Jia-Xin Chen, Institute of Nasopharyngeal Carcinoma,
The People's Hospital of Guangxi Zhuang Nationality Autonomous
Region, Nanning, P.R. China
E-mail: cjx166@yahoo.com.cn
Contributed equally
Key words: microRNA, nasopharyngeal carcinoma, microarray,
differentially expressed gene, pathway enrichment
migration and invasion (9). Wong et al demonstrated that
miRNAs of let-7 family suppress nasopharyngeal carcinoma
cell proliferation through down-regulating c-Myc expression
(10). Shi et al validated that underexpressed miR-100 leads
to Plk1 overexpression, which in turn contributes to NPC
progression (11). In addition to the investigation of individual
miRNA, high-throughput techniques have been applied to
identify deregulated miRNAs in NPC. Chen et al used a stem-
loop real-time-PCR assay to examine the expression levels
of 270 human miRNAs in 13 NPC samples and 9 adjacent
normal tissues. They identied 35 miRNAs whose expression
levels were signicantly altered in NPC samples and inferred
some cancer-related pathways enriched with targets of down-
regulated miRNAs (12). Sengupta et al performed a miRNA
microarray experiment to explore the expressions of 207
miRNAs between 31 NPCs and 10 normal tissues. They
demonstrated the involvement of miR-29c in NPC metastasis
by regulating mRNAs encoding extracellular matrix proteins
(13). However, the relationship between miRNAs and NPC
tumorigenesis is still unclear.
In this study, we recruited the Illumina miRNA microarray
platform, which covers a total of 735 human miRNAs, to prole
miRNA expression and used it to analyze human miRNAs in
8 NPC samples and 4 normal nasopharyngeal tissues. We
identied 34 signicantly differentially expressed miRNAs
and explored the relationship between their expression and
NPC progression. To evaluate the biological consequences of
the miRNAs dysregulation, a relative stringent target predic-
tion followed by pathway enrichment analysis was conducted
to identify the functional pathways specically regulated by
the aberrantly expressed miRNAs. We expect our analysis
to provide some clues to estimate the mechanism of miRNA
effects on NPC carcinogenesis.
Materials and methods
Tissue samples. Nasopharyngeal carcinoma tissue samples
were obtained from 8 poorly differentiated squamous NPC
patients before treatment at the Institute of Nasopharyngeal
Carcinoma, The People's Hospital of Guangxi Zhuang
Nationality Autonomous Region, Nanning, China. In addition,
normal nasopharyngeal tissues were obtained from 4 different
donors in the same hospital. All samples obtained were with
consent in accordance with approval granted by the Ethics
Committee of the Institute of Nasopharyngeal Carcinoma, The
People's Hospital of Guangxi Zhuang Nationality Autonomous
Region. Fresh NPC and nasopharyngeal tissues were imme-
diately snap-frozen in liquid nitrogen and stored at -80˚C
until use. All of the samples were used to do microarray and
quantitative real-time PCR (qRT-PCR). The NPC samples
we used are listed in Table I. The diagnosed stages I, II, III
and IV were classied by otorhinolaryngologic pathologist
according to the 2008 GuangZhou Staging of NPC.
RNA extraction and miRNA array hybridization. Total
RNA was extracted from 8 NPC tissues and 4 nasopharyn-
geal tissues using TRIzol (Invitrogen, Carlsbad, CA, USA)
reagent according to the manufacturer's instructions. All RNA
samples were examined for concentration and purity based
on the agarose gel electrophoresis and absorbance ratio at
260-280 nm to make sure the RNA without any protein and
We used the Illumina human v1 miRNA panel (based on
miRBase release 9.0) for miRNA analysis. RNA (100 ng) was
amplied using the Illumina Total Prep RNA amplication
kit (Ambion Cat. No. IL1791, Austin, TX, USA) to generate
biotinylated cRNA. An aliquot (1.5 mg/30 ml) of the labeled
cRNA for each sample, prepared in a probe cocktail that
included GEX-HYB hybridization buffer, was hybridized
to an Illumina human v1 miRNA panel at 58˚C for 16 h.
After hybridization, the chips were washed, coupled with
streptavadin-Cy3 and scanned in the Illumina BeadArray
Reader. The expression profiles have been deposited in
NCBI's Gene Expression Omnibus (GEO) with accession
number GSE22587.
Microarray data analysis. Data analysis and visualization
were performed using Illumina BeadStudio Gene Expression
Software (Illumina, Inc., San Diego, CA, USA). With Illumina
gene expression array, each probe is measured at least 30
times independently on random distributed beads. This
large number of technical replicates allows robust estimation
of the hybridization intensity and the measurement error
for each probe. We first transformed raw data generated
from BeadStudio using a variance stabilization transforma-
tion algorithm (14) and then normalized them using cubic
spline algorithm and background subtraction provided by
BeadStudio Software.
In general, a large fraction of miRNAs were either not
expressed or non-detectable, such that these miRNAs are
considered as weak signals. In this research, we adopted a
two-step strategy to deal with the weak signals. First, we
only chose the miRNA probes with detection p-value 0.01
in at least half of the chips for the further analysis, which
means when the detection p-value was 0.01 in at least
4 NPC tissues and 2 normal nasopharyngeal tissues, the
miRNA probe would be included in subsequent analysis. The
detection p-value, calculated by comparing the distribution
of the transcript signal to that of the negative control signal,
was set at 0.01 to identify transcript that were expressed
above background. Note that the detection p-values were
automatically reported in BeadStudio. This procedure reduced
the number of miRNA probes from 735 to 276 for nal data
analysis. Second, for the remaining 276 miRNA probes, when
Table Ι. The information of NPC samples used in the study.
Sample Gender Age TNM stage Cancer stage
Tumor 1 Female 35 T2NIM0
Tumor 2 Male 46 T2NIM0
Tumor 3 Female 24 T3NIM0
Tumor 4 Male 46 T4N2M0
Tumor 5 Female 39 T3N1M0
Tumor 6 Male 69 T3N1M0
Tumor 7 Male 65 T2N0M0
Tumor 8 Male 56 T4N3M0
ONCOLOGY REPORTS 25: 1353-1363, 2011
the detection p-value >0.01 in one chip, this corresponding
signal intensity was considered to be missed. To avoid
losing useful information and to facilitate data analysis, we
proposed an approach to impute the missing data. It can be
noted that for each chip, the minimum intensity of miRNAs
with detection p0.01 is always higher than the maximum
intensity of miRNAs with detection p-value >0.01, thereby the
minimum signal intensity of miRNAs with detection p≤0.01
can be considered as the threshold in this chip to separate the
valid intensity from background noise. Then, for each chip,
we ranked the signal intensities of all probes with detection
p≤0.01. If the signal intensity of a probe was lower than the
threshold, we provided the threshold as the intensity for this
miRNA probe. Otherwise, we would preserve its original
int ensit y.
To detect signicantly differentially expressed miRNAs,
we employed the software named SAM (signicant analysis of
microarray, version 3.02) (15). The cut-off was set at q-value
≤0.05 (a false discovery rate below 0.05 that the given miRNA
is expressed at different levels between NPC samples and
normal samples). The software Cluster and TreeView was used
to perform unsupervised hierarchical clustering analysis (16).
Real-time quantitative PCR. The quantitative real-time
RT-PCR was used to validate our microarray results. The
quantification of miRNA was performed by using Bulge-
Loop miRNA qPCR (RiBo Co., Cat. No. MQP-0101,
Guangzhou, China) following the manufacturer's instruction
in an ABI7900 Thermocycler (Applied Biosystems, Foster
City, CA, USA). Reverse transcriptase reactions included
purified total RNA (2 µg), 5 nmol/l stem-loop RT primer,
5X RT buffer, 0.2 mmol/l each of dNTPs, 100 U RT reverse
transcriptase and 20 U RNase inhibitor. The RNA template
and stem-loop RT primer were mixed to 11 µl and incubated
for 10 min at 70˚C, 2 min in ice. The other components
were added to 25 µl. The 25l reactions were incubated
in the ABI7900 thermocycler plate for 60 min at 42˚C,
10 min at 70˚C, and held at C. The 20 µl PCR included
2 µl RT product, 1X SYBR-Green Mix, 500 nmol/l miRNA
forward primer and 500 nmol/l miRNA reverse primer,
as recommended by the manufacturer. The reactions were
incubated at 95˚C for 20 sec, then followed by 40-45 cycles
of 95˚C for 10 sec, 60˚C for 20 sec and 70˚C for 10 sec. The
2-Ct method was used as relative quantication measure of
differential expression. All reactions were run in triplicate.
U6 small nuclear RNA was used as the internal control for
determining the relative miRNA expression level. In relation
to the expression of small nuclear U6 RNA, the expres-
sion level of specific miRNA for each RNA sample was
calculated, reecting by the value of ΔCt (Ct of miRNA – Ct
of U6). Two sided Wilcoxon rank sum test was performed
to analyze the expression differences of the three selected
miRNAs between NPC samples and normal samples with
the SPSS software (version 11.5).
Pathway analysis of the target genes of differentially expressed
miRNAs. Distinct miRNA target prediction methods may
result in considerably different target gene sets, herein, to rule
out the possibility of bias introduced by only one miRNA
target prediction method, we adopted the method introduced
by Ozen et al (17). If a given target was identied using at
least three of four different algorithms, including TargetScan
(release 4.1, containing 669 human miRNAs) (18), miRanda
(January 2008 release, total 475 human miRNAs) (19), PITA
(version 6, total 418 human miRNAs) (20), and PicTar (May
2007 release, total 178 human miRNAs) (21), it was consid-
ered likely to be a genuine miRNA target.
In addition, we queried a published mRNA expression
dataset to determine if any of the potential targets were not
only signicantly differentially expressed in NPC compared
to normal tissues, but also expressed in an inverse manner
relative to corresponding miRNAs, as would be predicted if
there was deregulation due to action of the miRNA targeting
them. Sengupta et al has reported that they conducted a
mRNA microarray experiment containing 31 nasopharyngeal
carcinoma (NPC) tissue samples and 10 normal nasopharyn-
geal tissues to identify the molecular mechanism of NPC (22).
For this mRNA expression dataset, we mapped the probe or
probeset IDs to NCBI Refseq IDs. When multiple probe sets
are mapped to the same Refseq ID, their values are averaged
to represent the expression level of this Refseq gene. We then
performed two-sided t-test for each gene to compare the
expressions between NPC and normal tissues. To address the
multiple test issues, we employed BH correction method to
adjust p-values (23). The mRNA was considered as deregu-
lated if it satised the condition of adjusted p-value 0.05.
A total of 634 genes possibly targeted by underexpressed
miRNA genes were signicantly increased in NPC, whereas a
total of 50 target genes of the only one overexpressed miRNA,
miR-18a, were signicantly decreased in NPC.
We then enriched these target genes in curated pathways
and Gene Ontology (GO) categories by the aid of GenMAPP
2.0 software, which is a widely used pathway visualization
and analysis tool for biological data (24). For convenience,
we uniformly denoted the identified pathway or GO term
as pathway. We used the GenMAPP software to map the
deregulated mRNA genes in the dataset of Sengupta et al in
all pathways. To evaluate whether miRNA targets are more
likely than expected by chance to be contained within one
pathway, we test if the targets of the aberrantly expressed
miRNA are enriched in this pathway. From all differentially
expressed genes in the dataset of Sengupta et al, a rate of
differential expression was computed for the entire array as
(number of total miRNA targeted deregulated genes/number
of total deregulated genes). For each of a pre-specified
pathway, a rate of differential expression was computed
as (number of miRNA targeted deregulated genes in this
pathway/number of total deregulated genes contained in
the pathway). When a pathway is unrelated with miRNA
modulation, differential expression occurs irrespective of
this pathway, and the differential expression rates for the
complete array and the pathway are the same. We tested this
null hypothesis against the alternative that the rates were
higher in the list of a pathway by comparing the differential
expression rates for gene targeted by aberrantly expressed
miRNAs and entire array. P-values were estimated using
one-sided hypergeometric test and the BH method was used
to address the problem of multiple test. The pathways statisti-
cally enriched with by miRNA targets were identied using a
threshold of adjusted p≤0.10.
Differential expression of miRNAs between NPC and normal
tissue. A total of 735 human miRNAs were detected with
Illumina microarray platform in each sample, which include
the sequences information from the Sanger Institute miRBase
Database (Release 9.0) and the public, accepted miRNAs
reported before. After the process of dealing with weak
signals, 276 miRNAs were included in subsequent differential
expression analysis. We applied the two class unpaired method
provided by SAM software and identified 33 significantly
underexpressed miRNAs and 1 signicantly overexpressed
miRNA (Table II).
MiR-18a, which belongs to miR-17-92 cluster, was the only
one overexpressed gene in our results. It was reported that
miR-18a prevents translation of ERα, potentially blocking the
protective effects of estrogen and promoting the development
of hepatocellular carcinoma in women (25). MiR-17-92 cluster
(encoding miR-17-5p, miR-17-3p, miR-18a, miR-19a, miR-20a,
miR-19b-1, and miR-92-1) located in intron 3 of C13orf 25
gene. miR-17-92 promotes tumor angiogenesis by regulating
the expressions of Tsp1 and CTGF (26). He et al reported that
the levels of the primary or mature miRNAs derived from the
miR-17-92 locus are often substantially increased in B-cell
lymphoma samples and cell lines. Enforced expression of the
miR-17-92 cluster acted with c-Myc expression to accelerate
tumor development in a mouse B-cell lymphoma model.
Thirty-three underexpressed miRNAs included the
miR-34b/miR-34c cluster, the miR-195/miR-497 cluster and a
majority of members of let-7 family. Among these miRNAs,
miR-34b and miR-34c were the top two underexpressed
miRNAs in NPC. Genes encoding miRNAs in the miR-34
family are direct transcriptional targets of p53, whose induc-
tion by DNA damage and oncogenic stress depends on p53
both in vitro and in vivo. Ectopic expression of miR-34
induces cell cycle arrest in both primary and tumor-derived
cell lines, which is consistent with the observed ability of
miR-34 to down-regulate the program of genes promoting
cell cycle progression. The p53 network suppresses tumor
formation through the coordinated activation of multiple
transcriptional targets, and miR-34 may act in concert with
other effectors to inhibit inappropriate cell proliferation
(27). Moreover, induction of miR-34b and miR-34c leads to
apoptosis or cellular senescence, whereas reduced miR-34b/c
expression attenuates p53-mediated cell death (28,29). On
the other hand, miR-34 enhances the activities of p53 by
inhibiting silence information regulator I (SIRTI). The posi-
tive feedback regulatory network based on p53 and miR-34
families play an important role in suppression of oncogenesis
and deterioration (30).
It should be mentioned that most members of let-7 family
in our microarray data were down-regulated. The let-7
miRNA is a founding member of the miRNA family and is
conserved in invertebrates and vertebrates, including humans,
where the let-7 family consists 11 very closely related genes
(31). Our microarray platform detected eight members of let-7
family (let-7a, let-7b, let-7c, let-7d, let-7e, let-7f, let-7g, let-7i)
and seven in them showed signicantly underexpression in
NPC. Johnson et al reported that the let-7 family negatively
regulates let-60/RAS. Loss of let-60/RAS suppresses let-7, and
3'UTR (untranslated region) of let-60/RAS contains multiple
let-7 complementary sites (LCSs), allowing let-7 to regulate
RAS expression (6). The expression level of let-7 is lower in
lung tumors than in normal lung tissues, while RAS protein
is signicantly higher in lung tumors. The inverse relationship
between let-7 and RAS suggested a possible regulation mecha-
nism for cancer cell proliferation. In addition, Johnson et al
showed that let-7 overexpression caused human cancer cells
to decrease cell cycle progression and let-7 directly regulated
a few key cell cycle proto-oncogenes, e.g., RAS, CDC25a,
CDK6, and cyclin D, thus controlling cell proliferation by
reducing ux through the pathways promoting the G1 to S
transition (32).
Validation of microarray results using quantitative real-time
RT-PCR. Quantitative RT-PCR method with a stem-loop
primer was employed to validate the reliability of miRNA
Table II. The differentially expressed miRNAs in NPCs com-
pared to normal tissues.
Fold change (log ratio
miRNA ID q-value (%) NPC/Normal)
hsa-miR-18a 0 2.23
hsa-miR-34c 0 -2.17
hsa-miR-642 0 -1.89
hsa-miR-34b 0 -1.85
HS_38.1 0 -1.83
hsa-miR-449 0 -1.56
hsa-miR-10b 0 -1.43
hsa-miR-92b 0 -1.37
hsa-miR-625 0 -1.22
hsa-miR-497 0 -1.16
hsa-miR-576 0 -1.15
hsa-miR-100 0 -0.82
hsa-let-7b 0 -0.75
hsa-miR-150 0 -0.65
hsa-miR-152 0 -0.63
hsa-let-7d 0 -0.58
hsa-miR-342 0 -0.56
hsa-miR-26b 0 -0.49
hsa-let-7g 0 -0.49
hsa-miR-29a 0 -0.48
HS_204.1 0 -0.47
hsa-let-7f 0 -0.46
hsa-miR-29b 0 -0.38
hsa-let-7e 0 -0.35
hsa-let-7a 0 -0.32
hsa-miR-30d 0 -0.28
hsa-miR-375 2.47 -1.74
hsa-miR-155 2.47 -0.50
hsa-miR-195 2.47 -0.33
hsa-miR-768-3p 2.47 -0.29
HS_210 4.3 -1.15
hsa-let-7c 4.3 -0.62
hsa-miR-425-5p 4.3 -0.46
hsa-miR-221 4.3 -0.24
ONCOLOGY REPORTS 25: 1353-1363, 2011
array result. Since mature miRNAs is very short (only
19-22 nt) without the polyA tail, the amplication could not
be conducted using the Oligo(dT) methods. Here we used the
stem-loop primer to extend the mature miRNA on the reverse
step designed by Chen et al (33). Stem-loop RT primers are
better than conventional ones in terms of RT efciency and
To validate our miRNA microarray result, we carried
out quantitative RT-PCR to investigate the expressions of 3
miRNAs in 8 NPC tissues and 4 normal tissues. MiR-34c is
signicantly underexpressed in NPC with a great fold change;
let-7g is also signicantly underexpressed in NPC, however,
with a moderate fold change; whereas miR-30c shows no
signicantly differential expression between NPC and normal
tissue. As expected in RT-PCR results, both miR-34c and
let-7g reveal signicant lower expression in NPC tissues with
respect to normal tissues, whereas the difference of miR-30c
expression between NPC and normal tissue is of no statistical
signicance (Fig. 1A). In addition, for each miRNA, we corre-
lated the fold changes from RT-PCR and microarray (Fig. 1B).
For miR-34c, the fold change is great in both microarray and
RT-PCR. For let-7g, which fold change is consistently rela-
tive low in microarray experiments, the fold change of this
miRNA in RT-PCR is about -1. Nevertheless, the expression
of miR-30c is the opposite between microarray and RT-PCR.
It can be explained by the fact that miR-30c has no differential
expression between NPC and normal tissues. The logarithm of
its fold change in microarray experiment is very close to zero,
therefore, so long as the difference of expressions in RT-PCR
experiment is not statistically signicant, it is possible that the
logarithm of fold change is either positive or negative. Taken
together, it is indicated that RT-PCR results are consistent
with our microarray data.
Cluster analysis on dif ferentially expressed miRNAs.
Unsupervised hierarchical clustering analysis was done by
using total 34 differentially expressed miRNAs, and this
analysis resulted in clearly segregated NPC samples from
the normal samples (Fig. 2). Moreover, as shown in Fig. 2,
the tumor tissues can be classified into two groups. We
associated the groups with tumor progression stage listed in
Table I. Surprisingly, three II stage tumors (T1, T2, T7) and
Figure 1. (A) Conrmatory studies of selected miRNAs by real-time RT-PCR. The graph shows the two aberrant miRNAs (miR-34c and let-7g) the
microarray results also showed the signicant alteration in real-time RT-PCR. (B) Fold change comparison between the microarray and real-time RT-PCR
in three miRNAs. The miR-34c and let-7g showed the same trends in microarray and real-time RT-PCR. However, the miR-30c did not show the same trend,
which can be explained by the expression difference of miR-30c between NPC and normal tissue and has no statistical signicance, so it is possible that the
logarithm of fold change is either positive or negative.
one III stage tumor (T3) are in one group; and the other group
includes two III stage tumor (T5,T6) and two IV stage tumor
(T4, T8). This demonstrates that hierarchical clustering of
miRNA expression data by using multiple miRNAs can group
NPCs into classes with clinical relevance, also suggesting
that these differentially expressed miRNAs may be of special
interest in future NPC research because they are involved in
the progression of cancer.
Identication of pathways enriched with targets of differ-
entially expressed miRNAs. To evaluate the biological
consequences of the miRNA abnormal expressions, we exam-
ined the pathway enrichment of targets of down-regulated
and up-regulated miRNAs, respectively. For example, focal-
adhesion is an important pathway involve in tumor formation
and progression, where the kinases is an important mediator
of growth-factor signaling, cell proliferation, cell survival
and cell migration. The genes in this pathway generally
increase their expression in human tumors (34). We observed
that the targets of underexpressed miRNAs are enriched
in focal-adhesion pathway. The rate of total up-regulated
miRNA targets to total up-regulated mRNA genes are 0.261
(584/2231). Thirty-four up-regulated mRNA genes can be
mapped into Focal-adhesion pathway, among which 18
genes are regulated by underexpressed miRNAs. The rate
of 0.529 (18/34) is signicantly higher than the background
rate of 0.261 (adjusted p0.05), indicating that the activation
of focal-adhesion pathway has strong association with the
underexpression of some miRNAs in NPC.
Six GO categories and no pathways were found to be statis-
tically enriched with the targets of the overexpressed miRNA
(Table IIIA). The possible explanation for the lack of relevant
pathways is that only one miRNA was identied as overex-
pressed in NPC, whose targets are not enough to be enriched
in some pathways. In our study, microarray expression
analysis indicated that almost all miRNAs are underexpressed
in NPC tissues (33,34). As miRNAs are negative regulators
of protein-coding genes, underexpression of these miRNAs
are expected to cause an up-regulation of their target genes
and alterations of the associated cellular pathways in NPC
tissues. In total 22 GO categories and 6 pathways were found
to be statistically enriched with targets of the underexpressed
miRNAs (Table IIIB), among which some well known cell
processes linked with tumor pathogenesis, such as cell-cell
signaling, cell adhesion, cell differentiation, cell mobility,
and cell senescence, are recognized as strongly associated
with miRNA modulation. Several published pathway analyses
have reached a conclusion that WNT signaling pathway is
abnormally regulated in NPC (12,35,36). In our study, 7 of 11
up-regulated genes in this pathway of ‘Wnt receptor signaling
pathway, are targeted by underexpressed miRNAs, consistent
with their observations. Moreover, the whole genome-wide
mRNA expression analysis shows that the pathway extracel-
lular matrix structural constituent’ is signicantly enriched
with up-regulated mRNA genes. Sengupta et al identified
that most of the genes, encoding extracellular matrix proteins,
can be regulated by miR-29c, which is underexpressed in
NPC (13). Although miR-29c expression is not signicantly
decreased in our study, miR-29a and miR-29b are found to
be underexpressed in NPC. MiR-29a and miR-29b share a
majority of targets in the category of extracellular matrix
structural constituent’ with miR-29c. In fact, 9 of 13 up-regu-
lated collagens (Table IIIB) can be mediated by miR-29a and
miR-29b, as reported by Sengupta et al (13), indicating that
the pathways of extracellular matrix structural constituent’
and ‘collagen’ are associated with tumor cell invasiveness and
metastatic potential of NPC.
Interestingly, two novel pathways are identified to be
subject to miRNA regulation. One is sensory perception
of sound, where 9 of 15 up-regulated genes are targeted by
underexpressed miRNAs. Hearing impairment is frequently
associated with nasopharyngeal carcinoma, nonetheless,
whether the high frequency hearing loss is the cause or the
effect of NPC is still unclear (37-39). It is conjectured that
the up-regulation of genes involved in sound perception is a
mechanism of compensation for the impairment of hearing
ability. The other pathway is ‘nervous system develop-
Figure 2. Unsupervised hierarchical clustering of 34 differentially expressed
miRNAs in 4 normal (yellow) and 8 NPC (red) samples. Samples were
clustered using Pearson correlation (uncentered) and average linkage.
Figure 3. c-Myc centred miRNA regulatory network in NPC. c-Myc and
miR-18a are marked with ascending arrow because their expression are
increased in NPC. Reduced miRNAs are marked with descending arrow.
The up-regulating and down-regulating regulatory effect are indicated by an
arrow and a bar, respectively.
ONCOLOGY REPORTS 25: 1353-1363, 2011
Table III. Statistically enriched pathways by targets of differentially expressed miRNAs.
A, Specically targeted pathways by overexpressed miRNA.
Enriched GO/Pathway Target
Metal ion binding (F) 13 192 1.20E-02
Zinc ion binding (F) 11 185 4.25E-02
Regulation of transcription\DNA-dependent (P) 8 119 4.19E-02
Nucleus (C) 14 307 7.07E-02
Transcription factor activity (F) 5 72 8.17E-02
Transcription (P) 6 102 8.71E-02
B, Specically targeted pathways by underexpressed miRNAs.
Enriched GO/Pathway Target
Neurogenesis 17 23 3.65E-04
Hs_Kit_Receptor_Signaling 9 11 1.58E-02
Hs_B_Cell_Receptor_Signaling 12 18 2.06E-02
Hs_IL-2_Signaling 10 14 2.18E-02
Hs_Focal_Adhesion 18 34 2.59E-02
Hs_Senescence_and_Autophagy 12 21 6.27E-02
Cell differentiation (P) 25 47 7.02E-03
Synapse (C) 8 9 9.14E-03
Actin binding (F) 15 24 9.17E-03
Nervous system development (P) 18 29 9.66E-03
Regulation of translation (P) 7 8 1.54E-02
Multicellular organismal development (P) 40 94 1.58E-02
Phosphate transport (P) 14 25 3.08E-02
Membrane (C) 140 429 3.19E-02
Collagen (C) 9 13 3.20E-02
Transcription factor activity (F) 44 116 4.63E-02
Protein binding (F) 260 871 4.72E-02
Sequence-specic DNA binding (F) 21 46 4.96E-02
Basement membrane (C) 9 14 4.98E-02
Extracellular matrix structural constituent (F) 13 24 5.00E-02
Receptor activity (F) 40 102 5.19E-02
Sensory perception of sound (P) 9 15 6.55E-02
Endocytosis (P) 8 13 7.98E-02
Cell-cell signaling (P) 19 43 8.35E-02
Wnt receptor signaling pathway (P) 7 11 8.72E-02
Heart development (P) 9 16 8.75E-02
Cell motility (P) 12 24 8.76E-02
Transcriptional activator activity (F) 11 21 8.85E-02
The number of genes targeted by differentially expressed miRNAs in a GO/Pathway.
Total overexpressed/underexpressed genes in a GO/
The p-value adjusted by the BH method to display the enrichment of miRNA targets in this GO/Pathway, representing the associa-
tion of this pathway with miRNA modulation. P, F and C in parentheses are denoted as the corresponding term from GO biological process,
molecular function and cellular component.
ment’, including neurogenesisand synapse. Especially for
the pathway of synapse’, almost all up-regulated synapse
genes (8/9) are regulated by underexpressed miRNAs, such
as SNAP25, SV2B, and SYT1. There are three possible
explanations for the relationship between nervous system
development and nasopharyngeal carcinoma. First, not only
the skull base, but the adjacent central nervous system, is
commonly invaded by NPC, and the occurrence of central
nervous system metastasis from nasopharyngeal carcinoma
(NPC) have been published (40,41). Second, nasopharyngeal
carcinoma patients after radiotherapy treatment are prone
to central nervous system (CNS) infection (42). Third, the
activation of Wnt pathway also induces neuronal circuit
development, such as synaptic differentiation, mature synapse
modulation and synaptic plasticity (43). These two novel
pathways enriched with miRNA targets provide an intriguing
clue to the biological consequences of the miRNAs abnormal
expression in NPC.
Reconstruction of c-Myc centered miRNA regulatory network.
c-Myc is an oncogene involved in cell proliferation, differen-
tiation, or cell death (44). It also exhibits signicantly elevated
expression in NPCs as compared to normal tissues (22). A few
of the miRNAs are directly regulated by c-Myc or regulate
c-Myc, leading to the altered expressions of various mRNAs
encoding tumor suppressors that block cell proliferation
(45). Here, based on our identied differentially expressed
miRNAs, we inferred a c-Myc centred miRNA regulatory
network in NPC (Fig. 3), including two levels of regulation.
First, the members of let-7 family and miR-34 family can
directly down-regulate c-Myc expression. Ectopic expression
of let-7 family in nasopharyngeal carcinoma cells resulted
in inhibition of cell proliferation through down-regulation of
c-Myc expression (10). MiR-34 family, including miR-34b and
miR-34c, are critical regulators of the c-Myc expression and
serve to remove c-Myc to prevent inappropriate replication
which may otherwise lead to genomic instability (46). Second,
c-Myc, as a transcription factor, can directly modulate an
array of miRNAs. The link between miR-18a and c-Myc
is well known. miR-18a, one of six members of miR-17-92
cluster, is directly activated by c-Myc as a way to ne tune
the activity of another c-Myc target E2F1 (47). Recently, Mott
et al experimentally validated that miR-29 expression can be
suppressed by c-Myc, promoting the malignant phenotype
(48). Kim et al proposed that c-Myc repressing several
miRNAs, including miR-100 and miR-221, plays a role in cell
cycle progression (45). It is of great interest to elucidate the
regulatory mechanisms of miRNAs in the c-Myc pathway
in NPC carcinogenesis. Further studies are warranted that
more and more miRNAs would be added in this regulatory
network. Thus, the inferred c-Myc centred network proposes
a relevant role of c-Myc in NPCs, and opens an important
new facet to our understanding of epigenetic alterations in
Two set of high-throughput data have been published to
screen the abnormally expressed miRNAs between NPC and
its normal counterpart. Chen et al employed a quantitative
Figure 4. The comparison of three high-throughput NPC dataset. (A) Venn diagram of differentially expressed miRNAs identied in our dataset (dataset
Li), dataset Chen et al (12) and dataset Sengupta et al (13) (B) the data correlation between two datasets. The x-axis and y-axis show the expressions of
miRNAs (log2 transformed fold change) in the corresponding dataset. Regress linear line and the Pearson correlation coefcient (R) are displayed in each
ONCOLOGY REPORTS 25: 1353-1363, 2011
RT-PCR assay to compare the expressions of 270 miRNAs
and identied 35 dys-regulated miRNAs (12). Sengupta et al
recruited a microarray spotted with 207 miRNAs to examine
their expressions. Among these miRNAs, 37 miRNAs exhibit
signicantly differential expressions between 31 NPC samples
and 10 normal samples (Wilcoxon rank sum test, at 5% false
discovery rate) (13). For convenience, we call our dataset
and these two published dataset as dataset Li, dataset Chen
and dataset Sengupta. Eight miRNAs are shared in the list
of abnormal expressions by dataset Li and dataset Chen;
whereas only 3 miRNAs are differentially expressed in both
dataset Li and dataset Sengupta (Fig. 4A). The inconsistency
may be explained by heterogeneity of samples and use of
different expression measuring platforms. Chen et al chose
corresponding adjacent normal nasopharynx tissue from
patients as normal samples. For dataset Sengupta, the normal
references include nasopharyngeal tissues and tissues from
other part of NPC patients, whereas our normal samples
are normal nasopharyngeal tissues collected from different
donors. In addition, the three datasets fail to show good inter-
platform concordance. Because Chen et al enumerated only
the expressions of their signicantly differentially expressed
miRNAs, there are only 30 miRNAs whose expressions are
available in all three datasets. To assess the data consistency,
we calculated the Pearson correlation coefcients of these 30
miRNAs between any two platforms. As illustrated in Fig. 4 ,
the highest correlation is 0.50, indicating that good agreement
was not observed between the platforms. It is noticed that the
correlation between dataset Li and dataset Sengupta is much
higher, suggesting the better data reproducibility in similar
techniques because Sengputa et al and our study adopted
the microarray platform, while Chen et al used a stem-loop
RT-PCR assay to quantify the expression levels of miRNAs
in NPC tissues. Nevertheless, several miRNAs are commonly
detected to be abnormally expressed. For example, miR-18a is
overexpressed in the datasets Li and Chen, which is reported
as potential oncogenes in various tumors (49). Reduced
miR-29c, up-regulating the mRNAs encoding extracellular
matrix proteins, contributes the invasion of NPC (13). In
both dataset Chen and Sengupta, miR-29c is consistently
down-regulated. In our dataset, the expression of miR-29c in
NPCs is moderately lower than those in normal tissues, yet
not meeting the statistical signicance level. However, the
other two members of miR-29 family, miR-29a and miR-29b
are signicantly down-regulated in our dataset. In fact, the
signicant underexpression of miR-29b is also observed in
dataset Sengupta. The seed sequence of miR-29c is identical to
that of miR-29a and miR-29b, leading to heavily overlapping
targets of the three members of miR-29 family. Therefore, it
is suggesting that the mechanism of decreased miR-29c to
elevate the expression of genes related to extracellular matrix
may be compensated by underexpression of miR-29a and
miR-29b in our study. Two underexpressed miRNAs, miR-34b
and miR-34c are overlapped in all three datasets. The reduced
expression of miR-34b and miR-34c frequently occur in
various cancers, which is believed to be strongly associated
with p53 network (27-29,50). Interestingly, the down-modu-
lated let-7 family members are observed only by our study.
Wong et al have experimentally confirmed the reduced
expression level of let-7 family in nasopharyngeal carcinoma
cells (10). The main functionality of down-regulated tumor-
suppressive let-7 family is interpreted to directly induce a few
key proto-oncogenes, such as RAS and c-Myc (51). The let-7
miRNAs may be of special interest to be tumor suppressors
in future nasopharyngeal carcinoma research.
In our study, a group of miRNAs were identified as
significantly differentially expressed between NPC and
normal tissues. The number of overexpressed miRNAs is
much lower than that of down-regulated miRNAs. Only
1 miRNAs showed overexpression in NPC, whereas 33
miRNAs were underexpressed in NPC. The demonstration
of the widespread down-regulation of miRNAs in NPC is
consistent with the findings by Chen et al and Sengupta
that a large number of miRNAs are down-modulated in
NPC tissues (12,13). Accumulated research has reported the
widespread down-regulation of miRNA expression in human
cancers (5,17,52,53). Lu et al used a bead-based detection
system to investigate the miRNA expression in multiple
human cancers. Their research displayed that a majority of
miRNAs are underexpressed in cancer tissues as compared to
normal tissues (5). Mattie et al also found widespread down-
regulation of miRNA on two biopsies of breast cancer versus
pooled normal tissue using a microarray-based approach
(53). Ozen et al observed that almost all of the miRNAs
detected were down-regulated in the majority of the prostate
cancer samples (17). Zhang et al speculated that DNA copy
number loss may contribute to the widely down-regulation of
miRNAs because miRNAs are frequently located in cancer-
associated regions of the human genome (54). In addition,
Thomson indicated that the widespread down-regulation of
miRNAs observed in human cancers might be due to a failure
in miRNA processing, especially the activity of the enzyme
Drosha, which digests the primary miRNA (pri-miRNA)
in the nuclease to release hairpin, precursor miRNA (pre-
miRNA) (55). The decreased activity of Drosha leads to the
deciency of biogenesis of pre-miRNA and mature miRNA,
and then the widespread down-regulation of miRNA can be
observed. This provides a hint that the deregulation of genes
taking part in the biogenesis of miRNAs may play critical
roles in tumorigenesis (56).
In conclusion, our study identied 34 aberrant miRNAs
occurred in human NPC tissues with respect to normal tissues
using the high-sensitive and high-throughput microarray
technology. Several well characterised biological pathways
were identied to be signicantly enriched with targets of
the under-expressed miRNAs. Two novel pathways, nervous
system development and sensory perception of sound,
were identied strongly associated with NPC development.
Furthermore, our study revealed that a c-Myc centred miRNA
regulatory network may play roles in NPC tumorigenesis.
Our work indicates that miRNAs are potential diagnosis
biomarkers and probable factors involved in the pathogenesis
of NPC. This study of differentially expressed miRNAs
may lead to nding their potential for improving diagnosis,
prognosis and their impact on further therapeutic strategies.
This research was supported by grant 30860081 from the
National Natural Science Funds. This research was also
supported by the Guangxi Science and Technology Key
Project (No. 0719006-2-4).
1. Bartel DP: MicroRNAs: genomics, biogenesis, mechanism, and
function. Cell 116: 281-297, 2004.
2. Ambros V: The functions of animal microRNAs. Nature 431:
350-355, 2004.
3. Wu L, Fan J and Belasco JG: MicroRNAs direct rapid deadeny-
lation of mRNA. Proc Natl Acad Sci 103: 4034-4039, 2006.
4. Calin GA, Dumitru CD, Shimizu M, et al: Frequent deletions
and down-regulation of micro-RNA genes miR15 and miR16
at 13q14 in chronic lymphocytic leukemia. Proc Natl Acad Sci
USA 99: 15524-15529, 2002.
5. Lu J, Getz G, Miska EA, et al: MicroRNA expression proles
classify human cancers. Nature 435: 834-838, 2005.
6. Johnson SM, Grosshans H, Shingara J, et al: RAS is regulated
by the let-7 microRNA family. Cell 120: 635-647, 2005.
7. Wei WI and Sham JST: Nasopharyngeal carcinoma. Lancet 365:
2041-2054, 2005.
8. Zhang L, Deng T, Li X, et al: microRNA-141 is involved in
a naso p ha r ynge al ca rci no m a- relat ed genes net work.
Carcinogenesis 31: 559-566, 2010.
9. Xia H, Ng SS, Jiang S, et al: miR-200a-mediated downregulation
of ZEB2 and CTNNB1 differentially inhibits nasopharyngeal
carcinoma cell growth, migration and invasion. Biochem Biophys
Res Commun 391: 535-541, 2010.
10. Wong TS, Man OY, Tsang CM, et al: MicroRNA let-7 suppresses
nasopharyngeal carcinoma cells proliferation through down-
regulating c-Myc expression. J Cancer Res Clin Oncol: May 4,
2010 (Epub ahead of print).
11. Shi W, Alajez NM, Bastianutto C, et al: Signicance of Plk1
regulation by miR-100 in human nasopharyngeal cancer. Int J
Cancer 126: 2036-2048, 2010.
12. Chen HC, Chen GH, Chen YH, et al: MicroRNA deregulation
and pathway alterations in nasopharyngeal carcinoma. Br J
Cancer 100: 1002-1011, 2009.
13. Sengupta S, Den Boon JA and Chen I: MicroRNA 29c is down-
regulated in nasopharyngeal carcinomas, up-regulating mRNAs
encoding extracellular matrix proteins. Proc Natl Acad Sci USA
105: 5874-5878, 2008.
14. Lin SM, Du P, Huber W and Kibbe WA: Model-based variance-
stabilizing transformation for Illumina microarray data. Nucleic
Acids Res 36: e11, 2008.
15. Tusher VG, Tibshirani R and Chu G: Signicance analysis of
microarrays applied to the ionizing radiation response. Proc Natl
Acad Sci USA 98: 5116-5121, 2001.
16. Eisen MB, Spellman PT, Brown PO and Botstein D: Cluster
analysis and display of genome-wide expression patterns. Proc
Natl Acad Sci USA 95: 14863-14868, 1998.
17. Ozen M, Creighton CJ, Ozdemir M and Ittmann M: Widespread
deregulation of microRNA expression in human prostate cancer.
Oncogene 27: 1788-1793, 2008.
18. Lewis BP, Shih IH, Jones-Rhoades MW, Bartel DP and
Burge CB: Prediction of mammalian microRNA targets. Cell
115: 787-798, 2003.
19. John B, Enright AJ, Aravin A, Tuschl T, Sander C and Marks DS:
Human MicroRNA targets. PLoS Biol 2: e363, 2004.
20. Kertesz M, Iovino N, Unnerstall U, Gaul U and Segal E: The
role of site accessibility in microRNA target recognition. Nat
Genet 39: 1278-1284, 2007.
21. Krek A, Grun D, Poy MN, et al: Combinatorial microRNA target
predictions. Nat Genet 37: 495-500, 2005.
22. Sengupta S, Den Boon JA, Chen IH, et al: Genome-wide
expression proling reveals EBV-associated inhibition of MHC
class I expression in nasopharyngeal carcinoma. Cancer Res 66:
7999-8006, 2006.
23. Reiner A, Yekutieli D and Benjamini Y: Identifying differentially
expressed genes using false discovery rate controlling proce-
dures. Bioinformatics 19: 368-375, 2003.
24. Doniger SW, Salomonis N, Dahlquist KD, Vranizan K,
Lawlor SC and Conklin BR: MAPPFinder: using Gene Ontology
and GenMAPP to create a global gene-expression prole from
microarray data. Genome Biol 4: R7, 2003.
25. Liu WH, Yeh SH, Lu CC, et al: MicroRNA-18a prevents estrogen
receptor-alpha expression, promoting proliferation of hepatocel-
lular carcinoma cells. Gastroenterology 136: 683-693, 2009.
26. Dews M, Homayouni A, Yu D, et al: Augmentation of tumor
angiogenesis by a Myc-activated microRNA cluster. Nat Genet
38: 1060-1065, 2006.
27. He L, He X, Lim LP, et al: A microRNA component of the p53
tumour suppressor network. Nature 447: 1130-1134, 2007.
28. Tarasov V, Jung P, Verdoodt B, et al: Differential regulation of
microRNAs by p53 revealed by massively parallel sequencing.
Cell Cycle 6: 1586-1593, 2007.
29. Welch C, Chen Y and Stallings RL: MicroRNA-34a functions as
a potential tumor suppressor by inducing apoptosis in neuroblas-
toma cells. Oncogene 26: 5017-5022, 2007.
30. Yamakuchi M and Lowenstein CJ: MiR-34, SIRT1 and p53: the
feedback loop. Cell Cycle 8: 712-715, 2009.
31. Reinhart BJ, Slack FJ, Basson M, et al: The 21-nucleotide
let-7 RNA regulates developmental timing in Caenorhabditis
elegans. Nature 403: 901-906, 2000.
32. Johnson CD, Esquela-Kerscher A, Stefani G, et al: The let-7
microRNA represses cell proliferation pathways in human cells.
Cancer Res 67: 7713-7722, 2007.
33. Chen C, Ridzon DA, Broomer AJ, et al: Real-time quantication
of microRNAs by stem-loop RT-PCR. Nucleic Acids Res 33:
e179, 2005.
34. McLean GW, Carragher NO, Avizienyte E, Evans J, Brunton VG
and Frame MC: The role of focal-adhesion kinase in cancer-a
new therapeutic opportunity. Nat Rev Cancer 5: 505-515,
35. Zeng ZY, Zhou YH, Zhang WL, et al: Gene expression
proling of nasopharyngeal carcinoma reveals the abnormally
regulated Wnt signaling pathway. Hum Pathol 38: 120-133,
20 07.
36. Shi W, Bastianutto C, Li A, et al: Multiple dysregulated
pathways in nasopharyngeal carcinoma revealed by gene expres-
sion proling. Int J Cancer 119: 2467-2475, 2006.
37. S u m i t s a wa n Y, Va s e e n o n V, H a n p r a s e r t p o n g C ,
Roongrotwattanasiri K, Chitapanarux I and Isaradisaikul S:
High frequency hearing loss following treatment for nasopha-
ryngeal carcinoma. J Med Assoc Thai 93: 324-329, 2010.
38. Chan SH, Ng WT, Kam KL, et al: Sensorineural hearing loss
after treatment of nasopharyngeal carcinoma: a longitudinal
analysis. Int J Radiat Oncol Biol Phys 73: 1335-1342, 2009.
39. Chen WC, Jackson A, Budnick AS, et al: Sensorineural hearing
loss in combined modality treatment of nasopharyngeal carci-
noma. Cancer 106: 820-829, 2006.
40. Gunn GB, Villa RD, Sedler RR, Hardwicke F, Fornari GA and
Mark RJ: Nasopharyngeal carcinoma metastasis to the pituitary
gland: a case report and literature review. J Neurooncol 68:
87-90, 2004.
41. Ngan RK, Yiu HH, Cheng HK, Chan JK, Sin VC and Lau WH:
Central nervous system metastasis from nasopharyngeal
carcinoma: a report of two patients and a review of the literature.
Cancer 94: 398-405, 2002.
42. Liang KL, Jiang RS, Lin JC, et al: Central nervous system
infection in patients with postirradiated nasopharyngeal
carcinoma: a case-controlled study. Am J Rhinol Allergy 23:
417-421, 2009.
43. Cerpa W, Toledo EM, Varela-Nallar L and Inestrosa NC: The
role of Wnt signaling in neuroprotection. Drug News Perspect
22: 579-591, 2009.
44. Pelengaris S, Khan M and Evan G: c-MYC: more than just a
matter of life and death. Nat Rev Cancer 2: 764-776, 2002.
45. Kim JW, Mori S and Nevins JR: Myc-induced microRNAs
integrate Myc-mediated cell proliferation and cell fate. Cancer
Res 70: 4820-4828, 2010.
46. Cannell IG, Kong YW, Johnston SJ, et al: p38 MAPK/
MK2-mediated induction of miR-34c following DNA damage
prevents Myc-dependent DNA replication. Proc Natl Acad Sci
USA 107: 5375-5380, 2010.
47. O'Donnell KA, Wentzel EA, Zeller KI, Dang CV and Mendell JT:
c-Myc-regulated microRNAs modulate E2F1 expression. Nature
435: 839-843, 2005.
48. Mott JL, Kurita S, Cazanave SC, Bronk SF, Werneburg NW and
Fernandez-Zapico ME: Transcriptional suppression of mir-29b-1/
mir-29a promoter by c-Myc, hedgehog, and NF-kappaB. J Cell
Biochem 110: 1155-1164, 2010.
49. He L, Thomson JM, Hemann MT, et al: A microRNA poly-
cistron as a potential human oncogene. Nature 435: 828-833,
50. Yamakuchi M, Ferlito M and Lowenstein CJ: miR-34a repres-
sion of SIRT1 regulates apoptosis. Proc Natl Acad Sci USA 105:
13421-13426, 2008.
ONCOLOGY REPORTS 25: 1353-1363, 2011
51. Boyerinas B, Park SM, Hau A, Murmann AE and Peter ME:
The role of let-7 in cell differentiation and cancer. Endocr Relat
Cancer 17: F19-F36, 2010.
52. Sun R, Fu X, Li Y, Xie Y and Mao Y: Global gene expression
analysis reveals reduced abundance of putative microRNA
targets in human prostate tumours. BMC Genomics 10: 93,
20 09.
53. Mattie MD, Benz CC, Bowers J, et al: Optimized high-
throughput microRNA expression profiling provides novel
biomarker assessment of clinical prostate and breast cancer
biopsies. Mol Cancer 5: 24, 2006.
54. Zhang L, Volinia S, Bonome T, et al: Genomic and epigenetic
alterations deregulate microRNA expression in human epithelial
ovarian cancer. Proc Natl Acad Sci USA 105: 7004-7009, 2008.
55. Thomson JM, Newman M, Parker JS, Morin-Kensicki EM,
Wright T and Hammond SM: Extensive post-transcriptional
regulation of microRNAs and its implications for cancer. Genes
Dev 20: 2202-2207, 2006.
56. Cheng C, Fu X, Alves P and Gerstein M: mRNA expression
profiles show differential regulatory effects of microRNAs
between estrogen receptor-positive and estrogen receptor-
negative breast cancer. Genome Biol 10: R90, 2009.
    • "Thus far, several high-throughput techniques have been applied to identify deregulated miRNAs in NPC. Most studies to date have applied microarray technique on nasopharyngeal carcinoma and non-cancer nasopharyngitis tissues, and some of them could identify the miRNAs as a prognostic factor or recurrent marker from the distinctive miRNA expression profile8910. But the results failed to show good interplatform concordance. "
    [Show abstract] [Hide abstract] ABSTRACT: Background: MicroRNAs (miRNAs) have been shown to play a critical role in the development and progression of nasopharyngeal carcinoma (NPC). Although accumulating studies have been performed on the molecular mechanisms of NPC, the miRNA regulatory networks in cancer progression remain largely unknown. Laser capture microdissection (LCM) and deep sequencing are powerful tools that can help us to detect the integrated view of miRNA-target network. Methods: Illumina Hiseq2000 deep sequencing was used to screen differentially expressed miRNAs in laser-microdessected biopsies between 12 NPC and 8 chronic nasopharyngitis patients. The result was validated by real-time PCR on 201 NPC and 25 chronic nasopharyngitis patients. The potential candidate target genes of the miRNAs were predicted using published target prediction softwares (RNAhybrid, TargetScan, Miranda, PITA), and the overlay part was analyzed in Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) biological process. The miRNA regulatory network analysis was performed using the Ingenuity Pathway Analysis (IPA) software. Results: Eight differentially expressed miRNAs were identified between NPC and chronic nasopharyngitis patients by deep sequencing. Further qRT-PCR assays confirmed 3 down-regulated miRNAs (miR-34c-5p, miR-375 and miR-449c-5p), 4 up-regulated miRNAs (miR-205-5p, miR-92a-3p, miR-193b-3p and miR-27a-5p). Additionally, the low level of miR-34c-5p (miR-34c) was significantly correlated with advanced TNM stage. GO and KEGG enrichment analyses showed that 914 target genes were involved in cell cycle, cytokine secretion and tumor immunology, and so on. IPA revealed that cancer was the top disease associated with those dysregulated miRNAs, and the genes regulated by miR-34c were in the center of miRNA-mRNA regulatory network, including TP53, CCND1, CDK6, MET and BCL2, and the PI3K/AKT/ mTOR signaling was regarded as a significant function pathway in this network. Conclusion: Our study presents the current knowledge of miRNA regulatory network in NPC with combination of bioinformatics analysis and literature research. The hypothesis of miR-34c regulatory pathway may be beneficial in guiding further studies on the molecular mechanism of NPC tumorigenesis.
    Full-text · Article · Dec 2016
    • "MiR-223 up-regulation in Hela cells inhibits cell proliferation by targeting IGF-1R [9]. However, in the gastric cancer development, miR-223 acts as an oncogene to promote cell invasion and migration via affecting expressions of EPB41L3 and FBXW7/hCdc4 genes [10, 26, 27]. In addition, miR-223 has been report to regulate the proliferation and invasion of human breast cancer cells for targeting Caprin-1 [28]. "
    [Show abstract] [Hide abstract] ABSTRACT: Mounting evidence suggests that miRNAs have major functions in tumor pathogenesis, and this study aimed to identify the candidate miRNA and investigate its role in nasopharyngeal carcinoma (NPC). MiRNA and mRNA expressions were screened by microarray assays. The cell proliferation, colony formation and migration ability were measured by MTT, soft agar and wound healing assays, respectively. The tumor growth suppression was evaluated by xenografting in nude mice. The plasma miR-223 levels in NPC patients were detected by TaqMan analysis. Real-time quantitative PCR and Western blotting were used to confirm miR-223 and MAFB expression levels. The targeting relationship between miR-223 and MAFB was verified using dual luciferase reporter assay. The miR-223 expression was decreased in CNE-1, CNE-2 cells as compared with NP69 cells, an immortalized human nasopharyngeal epithelial cell line, and its level also reduced in NPC patients' plasma as compared with healthy controls. Exogenous expression of miR-223 in CNE-2 cells could inhibit cell proliferation both in vitro and in vivo. Extrogenous miR-223 in CNE-2 cells would decrease the ability of colony formation and migration. MAFB, a transcription factor of Maf family members, was identified as a target gene of miR-223. We found that migration and invasion abilities were inhibited by MAFB silencing. MiR-223 negatively regulates the growth and migration of NPC cells via reducing MAFB expression, and this finding provides a novel insight into understanding miR-223 regulation mechanism in nasopharyngeal carcinoma tumorigenesis.
    Full-text · Article · Jun 2015
    • "Previous evidences have implicated the importance of miRNAs dysregulation in NPC tumorigenesis. The miRNA aberrant expression could promote an aggressive tumor phenotype by changing the expression of mRNA targets (Hu et al., 2009; Li et al., 2011). For instance, Sengupta et.al reported that mir-29c had a lower expression level in NPC tumors than in normal epithelium. "
    [Show abstract] [Hide abstract] ABSTRACT: Accumulative evidences indicated that microRNAs (miRNAs) can function as tumor suppressors and oncogenes, in which genetic variations are implicated in various cancer susceptibility. However, it remains unclear whether single nucleotide polymorphisms (SNPs) in mature miRNA sequence alter nasopharyngeal carcinoma (NPC) susceptibility. In this study, we analyzed associations between eight SNPs in miRNA mature sequences (i.e., rs3746444T>C in hsa-mir-499, rs4919510C>G in hsa-mir-608, rs13299349G>A in hsa-mir-3152, rs12220909G>C in hsa-mir-4293, rs2168518G>A in hsa-mir-4513, rs8078913T>C in hsa-mir-4520a, rs11237828T>C in hsa-mir-5579, and rs9295535T>C in hsa-mir-5689) and NPC susceptibility in southern Chinese with 906 NPC cases and 1,072 cancer-free controls, and validated the significant findings in eastern Chinese with 684 cases and 907 healthy controls. Functional assays were further performed to identify the biological effects of these polymorphisms. We found that rs4919510C>G polymorphism shown a consistent association with NPC risk in the southern Chinese (GC+GG versus CC genotype, odds ratio [OR] =1.36, 95% confidence interval [CI] =1.10-1.70) and the eastern Chinese (GC+GG versus CC: OR=1.37, 95%CI=1.08-1.74). After merged the two population, the ORs and 95%CI were 1.38 and 1.18 to1.62. Moreover, the rs4919510C>G adverse genotypes significantly interacted with Epstein-Barr virus (EBV) infection on increasing NPC risk (P=0.001). The functional assay further showed that the CNE-2 cell lines that transfected with miR-608-rs4919510G allele expression vector exerted more colony number formations than cell lines that transfected with miR-608-rs4919510C allele expression vector (P=0.001). These data suggested that rs4919510C>G of miR-608 may be a susceptible biomarker of NPC in Chinese. Copyright © 2015. Published by Elsevier B.V.
    Full-text · Article · Apr 2015
Show more