Journal of Cancer Therapy, 2012, 3, 452-459
http://dx.doi.org/10.4236/jct.2012.324058 Published Online September 2012 (http://www.SciRP.org/journal/jct)
Expression Profile of Epithelial Protein Lost in
Neoplasm-Alpha (EPLIN-α) in Human Pulmonary Cancer
and Its Impact on SKMES-1 Cells in Vitro
Yinan Liu1,2,3, Andrew J. Sanders1,2, Lijian Zhang1,3*, Wen G. Jiang1,2*
1Cardiff University-Peking University School of Oncology Joint Institute, Cardiff, UK; 2Metastasis and Angiogenesis Research
Group, Cardiff University School of Medicine, Cardiff, UK; 3Key Laboratory of Carcinogenesis and Translational Research (Minis-
try of Education), Department of Thoracic Surgery II, Peking University Cancer Hospital and Institute, Beijing, China.
Received July 27th, 2012; revised August 31st, 2012; accepted September 14th, 2012
ABSTRACT
Epithelial Protein Lost in Neop lasm (EPLIN) is a cytoskeletal associated protein implicated in regulatin g actin dynamo-
ics and cellular motility and whose expression is frequently downregu lated in a number of human cancers. The current
study examined the expression levels of EPLIN-α in a pulmonary cancer cohort and its association with clinical patho-
logical factors using quantitative polymerase chain reaction. Additionally, EPLIN-α was over-expressed in the SKMES-
1 pulmonary cancer cell line through transfection with a plasmid containing the expression sequence for EPLIN-α. The
role of EPLIN-α was subsequently examined using a variety of in vitro functional assays. Decreased levels of EPLIN-α
were seen in cancerous tissues compared to normal background tissue. Lower levels of EPLIN-α were also associated
with higher TNM stage and nodal involvement. In vitro over-expression of EPLIN-α inhibited SKMES-1 growth rates
(p = 0.05 vs plasmid control) and motility (p = 0.002 vs plasmid co ntrol), thou gh did not hav e any sign ificant effects on
cell-matrix adhesion or cell invasion. Taken together, the current study indicates that lower levels of EPLIN-α may be
associated with poorer prognosis and more advanced pulmonary cancer, where this molecule appears to play a suppres-
sive role on cell growth and migration.
Keywords: EPLIN-α; Pulmonary Cancer; Cell Migration; ECIS
1. Introduction
Pulmonary cancer is one of the most fatal cancers, with
distant metastasis being a key factor in the poor progno-
sis associated with this cancer. Epithelial protein lost in
neoplasm (EPLIN) has previously been found to localise
with actin stress fibres and play important roles in regu-
lating actin dynamics and linking the catenin-cadherin
complex to F-actin [1-3], suggesting a key role for this
molecule in regulating cellular motility.
Since the discovery of EPLIN, as a molecule different-
tially expressed between normal oral epithelial cells and
HPV-immortalised oral epithelial cell lines, scientific in-
terest has focused on the role of this molecule in cancer,
with a number of studies reporting aberrant expression of
EPLIN in cancerous cells and tissues of various cancer
types including, breast, prostate and oesophageal can cers
[4-8]. EPLIN, also known as Lima-1 (LIM domain and
actin binding-1), contains a centrally located LIM do-
main and exists as two isoforms, known as EPLIN-α and
EPLIN-β, with the EPLIN-β isoform containing an addi-
tional 160 amino acids at the N-terminus. The human
gene for EPLIN consists of 11 exons, spanning more
than 100 kb, with different transcriptional starting points
across this gene accounting for the EPLIN-α (90 kDa) and
EPLIN-β (110 kDa) isoforms [8,9 ]. Previous studies have
suggested roles for EPLIN-α in cytokinesis, where loss
of EPLIN-α in cancerous cells results in the mis-locali-
sation of cytokinesis proteins resulting in enhanced ge-
nomic instability [10]. Numerous studies have also im-
plicated the loss of EPLIN-α as contributing to the en-
hanced motility and invasiveness of cancer cells [1-3,5,6,
11,12]. EPLIN-α can bind actin monomers through the
NH2 and COOH protein ends and can also bind F-actin,
demonstrating a role in actin stabilisation [2]. Extracel-
lular signal regulated kinase (ERK) phosphorylation of
EPLIN has also been shown to reduce the C-terminal
protein affinity for F-actin, suggesting that ERK phos-
phorylation may be key to EPLINs role in regulating
actin dynamics [12]. EPLIN-α has also demonstrated the
capacity to link the cadherin-catenin complex to F-actin
*Corresponding authors.

Expression Profile of Epithelial Protein Lost in Neoplasm-Alpha (EPLIN-α) in Human Pulmonary Cancer and
Its Impact on SKMES-1 Cells in Vitro 453
through its ability to bind α-catenin [3]. Thus, EPLIN-α
appear to play key roles in regulating actin dynamics and
motility in normal cells. The loss of this molecule in can -
cerous cells likely impacts on these key functions, lead-
ing to a more invasive phenotype. Clinical studies have
further highlighted the importance of EPLIN-α in cancer
progression with lower levels of this molecule being as-
sociated with poor prognosis predictive factors such as
TNM stage, grade and patient survival rate in numerous
cancers including breast, prostate and oesophageal can-
cers [5,6,11]. In support of clinical observations, addi-
tional work has demonstrated that the forced expression
of EPLIN-α in prostate, breast and oesophageal cancer
cells can impact on aggressive traits in vitro and in vivo
[5-7]. Taken together all these studies indicate that
EPLIN-α may act as a potential prognostic indicator and
that the molecule may act as a protective factor in cancer
progression.
Whilst the importance and role of EPLIN-α in a num-
ber of cancers is beginning to become apparent, its role
in pulmonary cancer is currently unstudied. In the present
study, the expression of EPLIN-α was examined in a
cohort of pulmonary cancer samples and cell lines. Pul-
monary cancer cells were forced to express this molecule,
though transfection with a mammalian expression plas-
mid containing th e full sequence of EPLIN-α, in order to
establish the functional role of EPLIN-α in pulmonary
cancer cells in vitro.
2. Materials and Methods
2.1. Cell Lines and Maintenance
Human pulmonary cancer cell lines, SKMES-1 and A549
were purchased from ECACC (European Collection of
Animal Cell Culture, Salisbury, England, UK) and main-
tained in Dubecco’s Modified Eagle Medium (DMEM)
(Sigma, Dorset, UK) supplemented with penicillin, stre-
ptomycin and 10% foetal calf serum (Sigma, Dorset,
UK). The cells were incubated at 37˚C, 5% CO2 and 95%
humidity.
2.2. Pulmonary Tissue Sample Collection
Paired normal and tumour pulmonary tissues (n = 83
pairs) were collected immediately after surgery and
stored at –80˚C until use. Ethical approval and informed
consents were obtained from the local ethics committee
and patients respectively. The clinical follow-up was
routinely performed following surgery and the median
follow-up period was 120 months. Details of histology
were obtained from pathology reports and confirmed by
a consultant pathologist.
2.3. Generation of SKMES-1 Cells
Over-Expressing EPLIN-α
A plasmid containing the expression sequence for
EPLIN-α has previously been generated and used within
our laboratories [5-7,13]. In brief, the full length se-
quence for EPLIN-α was generated from mammary tis-
sue cDNA using a combination of the following amplify-
cation primers, 5’ATGGAAAATTGTCTAGGAGA’3
(EPLINaExF1), 5’ATGGAAAATTGTCTAGGAGAA’3
(EPLINaExF2) and 5’TCACTCTTCATCCT CATCCTC’3
(EPLINaExR1), together with a high fidelity PCR mas-
termix (AbGene, Epsom, UK). The product size was
verified before being T-A cloned into the pEF6/V5-His
mammalian expression vector (Invitrogen, Paisley, UK).
Plasmids were then inserted into chemically competent
TOP10 bacteria, amplified and plated under ampicillin
(100 µg/ml) selection. Colonies which grew were sub-
jected to orientation analysis and bacteria containing
plasmids with correctly orientated inserts were selected,
amplified and harvested for plasmid extraction (GenElute,
Sigma, Dorset, UK). This resource was used to transfect
the human pulmonary SKMES-1 cancer cell line, which
demonstrated minimal expression levels of EPLIN-α.
SKMES-1 cells were transfected with either the EPLIN-α
expression plasmid or a closed pEF6 plasmid and sub-
jected to a period of selection (5 µg/ml blasticidin) be-
fore subsequent maintenance of the cells at 0.5 µg/ml
blasticidin. Following the selective pe riod, the expression
of EPLIN-α was examined in the wild type and trans-
fected cells, at the transcriptional level, using RT-PCR.
Those cells containing the expression plasmid and dis-
playing enhanced EPLIN-α expression were designated
SKMES-1EPLIN-α exp, those containing the pEF6 control
plasmid were designated SKMES-1pEF6 and unaltered
wild type cells were designated SKMES-1WT.
2.4. RNA Extraction, Reverse
Transcription-PCR, and Quantitative PCR
RNA isolation was performed on either a 25 cm2 tissue
culture flask or homogenised tissue samples using TRI
Reagent (Sigma, Dorset, UK). RNA was subsequently
quantified using a spectrophotometer (WPA UV 1101,
Biotech Photometer, Cambridge, UK) before standardis-
ing all samples to a concentration of 500 ng for use in
reverse transcription (iScript™ Reverse Transcription
Supermix kit, Bio-Rad Laboratories, Hemel Hempstead,
UK). Routine RT-PCR was carried out using specific
primers for EPLIN-α (Table 1). Amplification conditions
were as follows: 94˚C for 5 minutes, 32 cycles of 94˚C
for 40 seconds, 55˚C for 40 seconds and 72˚C for 60
seconds, followed by a final extension of 72˚C for 10
minutes and 4˚C hold. PCR products were separa ted on a

Expression Profile of Epithelial Protein Lost in Neoplasm-Alpha (EPLIN-α) in Human Pulmonary Cancer and
Its Impact on SKMES-1 Cells in Vitro
454
Table. 1. Primer sequences used in PCR and Q-PCR.
Primer set Sense Anti-sense
GAPDH probe ATGATATCGCCGCGCTCA CGCTCGGTGAGGATCTTCA
EPLIN probe/Q-PCR AAGCAAAAATGAAAACGAAG ACTGAACCTGACCGTACAGACACCCACCTTAGCAA
GAPDH Q-PCR CTGAGTACGTCGTGGAGTC ACTGAACCTGACCGTACACAGAGATGATGACCCTTTTG
Z-sequence on Q-PCR primers is 5’ ACTGAACCTGACCGTACA 3’.
2% agarose gel and documented, following ethidium bro-
mide staining, using a digital camera mounted over a UV
transluminator.
The level of EPLIN-α transcript present in the clinical
samples was determined using real-time quantitative
PCR, based on the Amplifluor technology and modified
from a method reported previously [5]. Briefly, primers
were designed in a similar way as to those used in con-
ventional PCR. To one of the primer pairs, an additional
sequence was added, known as the Z sequence, which is
complementary to the universal Z probe (Intergen Inc.,
Oxford, England, UK). The reaction was carried out us-
ing the following: Hot-start Q-master mix (Abgene, Ep-
som, UK), 10 pmol of specific forward primer, 1 pmol
reverse primer (containing the Z sequence), 10 pmol of
FAM-tagged probe (Intergen), and cDNA from approxi-
mately 50 ng RNA. The reaction was carried out using an
IcyclerIQ™ (Bio-Rad, surrey, UK). The following con-
ditions were used in the reaction: 94˚C for 12 minutes, 60
cycles of 94˚C for 15 seconds, 55˚C for 40 seconds and
72˚C for 20 seconds. The levels of the EPLIN-α trans-
cripts are presented as transcript copies per 50 ng RNA
and values are calculated based on an internal standard
that was simultaneously amplified during the same quan-
titative real-time PCR.
2.5. In Vitro Cell Growth Assay
Cells were seeded into triplicate 96-well plates at a seed-
ing density of 3000 cells per well. The plates were incu-
bated for differing periods (overnight, 3 and 5 days).
Following incubation for the appropriate time, plates
were fixed in 4% formalin (v/v) and stained with 0.5%
(w/v) crystal violet. Crystal violet stain was extracted
using 10% (v/v) acetic acid and the absorbance was de-
tected at a wavelength of 540 nm using a spectropho-
tometer.
2.6. In Vitro Invasion Assay
This was undertaken following a previously described
method [5,6]. Transwell inserts with 8.0 µm pores were
coated with 50 µg Matrigel (BD Bioscience, Oxford, UK)
and air dried. The Matrigel was then rehydrated before
seeding 30,000 cells to each insert. After 72 hours, any
cells that had invaded to the underside of the insert were
fixed in 4% formalin and stained in crystal violet. Cell
invasion was quantified by assessing the number of
stained cells that had invaded to the underside of the in-
sert in random microscopic fields.
2.7. In Vitro Matrix Adhesion Assay
Wells of a 96 well plate were pre-coated with 5 µg of
Matrigel and air dried. The Matrigel layer was then re-
hydrated before seeding 45,000 cells onto the matrix
layer and incubating for 40 minutes. Following this in-
cubation, non-adherent cells were removed through nu-
merous washes with BSS before fixing adherent cells in
formalin and staining with crystal violet. Cell-matrix
adhesion was then quantified under the microscope by
assessing the number of adherent cells in random micro-
scopic fields.
2.8. ECIS Based Cellular Motility Assays
Cell motility was assessed using the Electric Cell-sub-
strate Impedance Sensing (ECIS) system (Applied Bio-
physics Inc, NJ, US) and 96W1E arrays. One hundred
and twenty thousand cells (SKMES-1pEF6 or SKMES-
1EPLIN-α exp) were seeded into each well. These cells were
allowed to settle and attach to the array wells to form a
confluent monolayer. This monolayer was subsequently
wounded, through the application of an electrical charge
to the electrode. The change in resistance of the cell layer
was recorded for a number of hours as the surrounding
cells recovered and migrated onto the electrode to close
the wound. Additional analysis was performed to assess
the importance of PLCγ signalling in EPLIN-α regulated
migration using a PLCγ inhibitor (CalBiochem, Notting-
ham, England, UK). To accomplish this, similar experi-
ments were set up to examine the motility rates of
SKMES-1pEF6 and SKMES-1EPLIN-α exp cells, both in the
presence and absence of 100nM PLCγ inhibitor.
2.9. Statistical Analysis
The Minitab 14 statistical software package was used to
carry out statistical analysis. Normally distributed data
was analysed using two-sample, two-tailed t-tests and
non-normally distributed data was analysed using the
Mann-Whitney test. Two Way ANOVA analysis of mi-

Expression Profile of Epithelial Protein Lost in Neoplasm-Alpha (EPLIN-α) in Human Pulmonary Cancer and
Its Impact on SKMES-1 Cells in Vitro 455
gration data was undertaken using the SIGMAPLOT 11
statistical package. Statistical comparisons were made
between the SKMES-1pEF6 and SKMES-1EPLIN-α exp cells.
3. Results
3.1. The Expression of EPLIN-α in Pulmonary
Cancer
The expression of EPLIN-α was examined in pulmonary
cancer cell lines and in a clinical cohort of pulmonary
cancer and normal background samples. The transcript
expression of EPLIN-α in the SKMES-1 and A549 pul-
monary cancer cell lines was determined using RT-PCR
(Figure 1(a)). Levels of EPLIN-α were found to be
minimal in both of the cell lines tested . Quantitative PCR
was also used to detect EPLIN-α transcript levels in the
clinical pulmonary cohort. EPLIN-α expression was found
to be significantly reduced in pu lmonary tumour samples
(139 ± 49 copies/50ng RNA), compared to the normal
background tissues (3373 ± 1266 copies/50ng RNA, p =
0.013) (Figure 1(b)).
Figure 1. Expression of epithelial protein lost in neoplasm-α
(EPLIN-α) in pulmonary cancer cell lines and tissues. (a)
EPLIN-α expression was found to be very weak in both
SKMES-1 and A549 pulmonary cancer cell lines using RT-
PCR; (b) EPLIN-α transcript levels were also significantly
decreased in tumour tissue compared to normal pulmonary
tissues (p = 0.013).
3.2. Association of EPLIN-α with Clinical
Pathological Details
To assess the importance of EPLIN-α expression in dis-
ease progression, EPLIN-α transcript levels in the pul-
monary cancer samples were analysed against important
clinical statuses, such as histological type, grade, node
status, and TNM staging. Similar levels of EPLIN-α
transcript was seen in squamous cancers (175 ± 70) and
adenocarcinomas (159 ± 100). Reduced transcript levels
of EPLIN-α were observed in small cell lung cancer (35
± 17), compared to squamous cancers, though this was
not quite significant (p = 0.06). Significant reductions in
EPLIN-α transcript levels were seen in other pulmonary
cancers (pulmonary cancers other than squamous, ade-
nocarcinoma or small cell) compared to squamous cancer
(p = 0.019). As sample number was very small (n = 3),
the statistical comparisons of the other types such as car-
cinoid was not possible (Figure 2(a)). Expression of
EPLIN-α was discovered to be elevated in tumour sam-
ples of a lower TNM1 stage (391 ± 169) compared to the
more advanced TNM2 (74 ± 41, p = 0.08 vs TNM1) and
TNM3 (45 ± 15, p = 0.05, vs TNM1) tumours (Figure
2(b)). Higher levels of EPLIN-α transcript were detected
in lymph node negative tumours (N0, 271 ± 113) than in
those with local lymph node involvement (N1) (75 ± 34,
p = 0.1, vs N0) and those with advanced lymph node
metastasis (N2, 42 ± 17, p = 0.05, vs N0) (Figure 2(c)).
It was also noted that samples from patients who had
vessel cancerous embolus had a significantly lower ex-
pression level of EPLIN-α compared to embolus negative
patients (21 ± 18 vs 190 ± 77 respectively, p = 0.037)
(Figure 2(d)).
3.3. Impact of EPLIN-α on in Vitro Growth,
Invasion, and Cell-Matrix Adhesion
To further evaluate the biological function of EPLIN-α in
pulmonary cancer, SKMES-1 cells were transfected with
a mammalian EPLIN-α expression plasmid. Enhanced
EPLIN-α expression was confirmed in cells transfected
with the expression plasmid (SKMES-1EPLIN-α exp) com-
pared to wild type (SKMES-1WT) and empty plasmid
transfected (SKMES-1pEF6) cells using RT-PCR (Figure
3(a)). Over-expression of EPLIN-α resulted in a reduced
rate of growth in vitro (Figure 3(b)), with significant re-
ductions in growth rates seen between SKMES-1EPLIN-α exp
and SKMES-1pEF6 over a 5 day incubation period (p =
0.05). In contrast to this, forced expression of EPLIN-α
appeared to have little impact on cell-matrix adhesion
and cellular invasion, with no significant differences in
either of these traits being observed between SKMES-1pEF6
control cells and SKMES-1EPLIN-α exp cells (p > 0.05 in
both cases) (Figures 3(c) and (d)).

Expression Profile of Epithelial Protein Lost in Neoplasm-Alpha (EPLIN-α) in Human Pulmonary Cancer and
Its Impact on SKMES-1 Cells in Vitro
456
Figure 2. Association of EPLN-α with histological type, nodal status, and TNM stage. (a) EPLIN-α transcript levels were
found to be highest in the squamous and adenocarcinoma cancers, reduced levels of EPLIN-α were observed in the small cell
cancers (p = 0.06 vs squamous) and combined other types of pulmonary cancer (p = 0.019); (b) Decreased levels of EPLIN-α
were also associated with a more advanced TNM stage. Substantial decreases in EPLIN-α expression were observed between
TNM1 and TNM2 (p = 0.08) and TNM1 and TNM3 (p = 0.05); (c) Similarly, decreased expression of EPLIN-α was also asso-
ciated with lymph node involvement, with highest levels of EPLIN-α being observed in patients with no lymph node involve-
ment (N0). Decreased levels were seen in those patients with local lymph node involvement (N1; p = 0.1 vs N0), with lowest
levels observed in patients who had advanced lymph node involvement (N2; p = 0.05 vs N0); (d) Significantly reduced
EPLIN-α expression was also apparent in patients with local advanced cancers with vessel embolus (p = 0.037).
3.4. EPLIN-α Over-Expression Inhibits
Pulmonary Cell Migration Rates
An Electric Cell Impedance Sensing (ECIS) method was
used to investigate the impact of EPLIN-α over-expres-
sion on cell motility (Figure 4). Forced expression of
EPLIN-α in SKMES-1 pulmonary cancer cells dramati-
cally inhibited monolayer recovery following electrical
wounding. Significant differences in recovery (assessed
through change in resistance over the electrode as cells
migrated to re-cover the electrode) over the 6 hour period
were seen between SKMES-1pEF6 control and SK-
MES-1EPLIN-α exp cells (p = 0.002) (Figure 4(a)). Treat-
ment with the PLCγ inhibitor seemed to h ave little effect
on the recov ery rate s of either SK MES-1 pEF6 control cells
or SKMES-1EPLIN-α exp cells (Figure 4(b)). In both cases,
treatment with PLCγ inhibitor did not significantly im-
pact on the migratory rate of either cell line compared to
the untreated equivalent (p > 0.05).
4. Discussion
Epithelial protein lost in neoplasm is a cytoskeletal asso-
ciated protein involved in the regulation of actin dynam-
ics and subsequently in cell motility [2,3]. The expres-
sion of EPLIN-α has been found to be down-regulated in
a number of oral, breast, oesophageal and prostate cancer
cell lines compared to their normal counterparts [5-8].
Previous studies from our laboratories have provided
data supporting a tumour/metastasis suppressive role for
EPLIN-α, where enhanced levels of EPLIN-α can nega-
tively impact on key metastatic and angiogenic traits in
vitro and in vivo [5,6,13]. This is supported by data from
our clinical cohorts indicating that reduced EPLIN-α
levels are associated with poor NPI prognosis and lower
patient surviv al rates in breast cancer patients [5].
The clinical data obtained in the cu rrent study, appears
to be in line with previous observations. Transcript levels
of EPLIN-α where found to be significantly reduced in
tumour tissues compared to normal tissues. We further
analyzed the quantity of EPLIN-α transcript in pulmo-
nary cancer samples against the corresponding clinical
data. Relatively high levels of EPLIN-α transcript were
detected in squamous pulmonary cancer and adenocarci-
nomas compared to small cell cancers and other types of

Expression Profile of Epithelial Protein Lost in Neoplasm-Alpha (EPLIN-α) in Human Pulmonary Cancer and
Its Impact on SKMES-1 Cells in Vitro 457
Figure 3. Impact of EPLIN-α over-expression on SKMES-1 pulmonary cancer cells. (a) Successful over-expression of EPLIN-
α in SKMES-1 cells was seen in cells containing the expression plasmid (SKMES-1EPLIN-α exp) compared to wild type
(SKMES-1WT) and plasmid control (SKMES-1pEF6) cells using RT-PCR; (b) Over-expression of EPLIN-α resulted in a de-
creased rate of growth, in comparison to the plasmid control line, over a 5 day incubation period (p = 0.05). No significant
effect on SKMES-1 invasiveness; (c) Or cell matrix-adhesion; (d) was observed following over-expression of EPLIN-α (p >
0.05). Mean values +/− SEM are shown.
Figure 4. Impact of EPLIN-α over-expression on SKMES-1 cell motility. (a) Over-expression of EPLIN-α significantly inhi-
bited cell motility in SKMES-1 cells, with significant differences being seen between SKMES-1pEF6 and SKMES-1EPLIN-α exp
cells over time (p = 0.002); (b) Inhibition of PLCγ signalling by treatment of cells with a PLCγ inhibitor did not have any sig-
nificant effects on the motility rate of either cell line (vs untreated control, p > 0.05). Representative data is shown.

Expression Profile of Epithelial Protein Lost in Neoplasm-Alpha (EPLIN-α) in Human Pulmonary Cancer and
Its Impact on SKMES-1 Cells in Vitro
458
pulmonary cancers, such as carcinoid. Expression levels
of EPLIN-α appeared to associate with the TNM stage of
the cancer, with highest levels being seen in the lower
TNM1 stage and levels reducing at TNM2 and TNM3
stages with close to significant and significant diffe-
rences observed between TNM1 vs TNM2 and TNM1 vs
TNM3 respectively. This observation is in line with ob-
servations from the clinical breast cohort [5], where
lower level of EPLIN-α were seen in the higher TNM
stages and similarly indicates that a loss of EPLIN-α ex-
pression occurs during the advancement of pulmonary
cancer. Decreased expression levels of EPLIN-α were
also found to associate with lymphatic involvement.
Highest levels of EPLIN-α transcripts were observed in
cancers with no lymphatic involvement (N0) and large
reductions in EPLIN-α expression, in comparison to no
lymphatic involvement, were seen in tissues where local
lymph node involvement was apparent (N1), with further
reductions in EPLIN-α expression being apparent in the
tissues from patients with advanced lymph node in-
volvement (N2). A recent study has also highlighted re-
duced expression of EPLIN in lymph node metastasis in
a number of epithelial cancers, including prostate, colo-
rectal, breast and squamous cell carcinoma of the head
and neck [11]. The data presented here further suggests
that EPLIN-α may be a key contributing factor in deter-
mining the likelihood of cancer cells metastasising
through the lymphatic route in a variety of human can-
cers. Finally, our clinical data indicated that reduced
EPLIN-α levels were associated with patients who had
local advanced cancers with vessel cancerous embolus.
Thus, the clinical results indicate that, similar to other
studies on different cancer types, lower levels of EPLIN-
α are associated with a more aggressive pulmonary can-
cer and an increased likelihood of metastatic spread.
To support our clinical findings, we generated a
SKMES-1 human lung cancer line over-expressing EP-
LIN-α and examined the impact of this over-expression
on the cellular functions of this cell line. Forced expres-
sion of EPLIN-α brought about a reduction in SKMES-1
cell growth and motility co mpared to control cells. How-
ever, over-expression of EPLIN-α in this cell line did not
seem to influence SKMES-1 invasiveness. This data is
somewhat in line with the established role of EPLIN in
motility. Previous studies from our laboratory have also
found the over-expression of EPLIN-α to negatively im-
pact on the cellular invasiveness and motility of MDA-
MB-231 breast cancer cells [5] and negatively affect the
motility of HECV endothelial cells [13], though in oeso-
phageal cancer cells over-expression of EPLIN-α had a
greater influence on cellular invasiveness than motility
[7]. Additionally, treatment with a PLCγ inhibitor seemed
to have similar effects on both control and EPLIN-α
over-expressing cells, suggesting that the role that
EPLIN-α plays in cellular motility in this cell line may
not be linked to the PLCγ signalling pathway. Thus, our
data suggests that, in this lung cancer cell line, EPLIN-α
may not play as great a role in cell invasion. The exact
reason behind this observation, which is in contrast to
other studies, is currently unknown. Further work to cla-
rify this observation is required, examining the role of
EPLIN-α in other pulmonary cancer cell lines using both
over-expression and knockdown studies where possib le.
Collectively, the present study suggests that EPLIN-α
is inversely associated with the aggressiveness and clini-
cal outcome of human pulmonary cancers and can influ-
ence in vitro cell growth and migration of lung cancer
cells. Further data examining EPLIN-α expression in
larger cohorts and also using in vivo metastasis models
are required to fully explore the potential importance of
EPLIN-α in pulmonary cancer.
5. Acknowledgements
Dr Yinan Liu is a recipient of the Albert Hung China
Medical Scholarship of Cardiff University. The authors
wish to thank Cancer Research Wales for their kind sup-
port of this wo rk .
REFERENCES
[1] Y. Song, R. E. Maul, C. S. Gerbin and D. D. Chang, “In-
hibition of Anchorage-Independent Growth of Trans-
formed NIH3T3 Cells by Epithelial Protein Lost in Neo-
plasm (EPLIN) Requireslocalization of EPLIN to Actin
Cytoskeleton,” Molecular Biology of the Cell, Vol. 13,
No. 4, 2002, pp. 1408-1416. doi:10.1091/mbc.01-08-0414
[2] R. S. Maul, Y. Song, K. J. Amann, S. C. Gerbin, T. D.
Pollard and D. D. Chang, “EPLIN Regulates Actin Dy-
namics by Cross-Linking and Stabilizing Filaments,” The
Journal of Cell Biology, Vol. 160, No. 3, 2003, pp. 399-
407. doi:10.1083/jcb.200212057
[3] K. Abe and M. Takeichi, “EPLIN Mediates Linkage of
the Cadherin-Catenin Complex to F-Actin and Stabilizes
the Circumferential Actin Belt,” Proceedings of the Na-
tional Academy of Sciences, Vol. 105, No. 1, 2008, pp.13-
19. doi:10.1073/pnas.0710504105
[4] D. D. Chang, N.-H. Park, C. T. Denny, S. F. Nelson and
M. Pe, “Characterization of Transformation Related
Genes in Oral Cancer Cells,” Oncogene, Vol. 16, No. 15,
1998, pp. 1921-1930. doi:10.1038/sj.onc.1201715
[5] W. G. Jiang, T. A Martin, J. M. Lewis-Russell, A. Doug-
las-Jones, L. Ye and R. E. Mansel, “Eplin-Alpha Expres-
sion in Human Breast Cancer, the Impact on Cellular Mi-
gration and Clinical Outcome,” Molecular Cancer, Vol. 7,
2008, pp. 71-80. doi:10.1186/1476-4598-7-71
[6] A. J. Sanders, T. A. Martin, L. Ye, M. D. Mason and W.
G. Jiang, “EPLIN Is a Negative Regulator of Prostate
Cancer Growth and Invasion,” Journal of Urology, Vol.

Expression Profile of Epithelial Protein Lost in Neoplasm-Alpha (EPLIN-α) in Human Pulmonary Cancer and
Its Impact on SKMES-1 Cells in Vitro 459
186, No. 1, 2011, pp. 295-301.
doi:10.1016/j.juro.2011.03.038
[7] Y. Liu, A. J. Sanders, L. Zhang and W. G. Jiang,
“EPLIN-Alpha Expression in Human Oesophageal Can-
cer and Its Impact on Cellular Aggressiveness and Clini-
cal Outcome,” Anticancer Research, Vol. 32, No. 4, 2012,
pp. 1283-1289.
[8] R. S. Maul and D. D. Chang, “EPLIN, Epithelial Protein
Lost in Neoplasm,” Oncogene, Vol. 18, No. 54, 1999, pp.
7838-7841. doi:10.1038/sj.onc.1203206
[9] S. Chen, R. S. Maul, H. R. Kim and D. D. Chang, “Char-
acterization of the Human EPLIN (Epithelial Protein Lost
in Neoplasm) Gene Reveals Distinct Promoters for the
Two EPLIN Isoforms,” Gene, Vol. 248, No. 1-2, 2000,
pp. 69-76. doi:10.1016/S0378-1119(00)00144-X
[10] M. Chircop, V. Oakes, M. E. Graham, M. P. Ma, C. M.
Smith, P. J. Robinson and K. K Khanna, “The Actin-
Binding and Bundling Protein, EPLIN, Is Required for
Cytokinesis,” Cell Cycle, Vol. 8, No. 5, 2009, pp. 757-
764. doi:10.4161/cc.8.5.7878
[11] S. Zhang, X. Wang, A. O. Osunkoya, S. Iqbal, Y. Wang,
Z. Chen, S. Muller, S. Josson, I. M. Coleman, P. S. Nel-
son, Y. A. Wang, R. Wang, D. M. Shin, F. F. Marshall, O.
Kucuk, W. L.Chung, H. E. Zhau and D. Wu, “EPLIN
Downregulation Promotes Epithelial-Mesenchymal Tran-
sition in Prostate Cancer Cells and Correlates with Clini-
cal Lymph Node Metastasis,” Oncogene, Vol. 30, No. 50,
2011, pp. 4941-4952. doi:10.1038/onc.2011.199
[12] M. Y. Han, H. Kosako, T. Watanabe and S. Hattori, “Ex-
tracellular Signal-Regulated Kinase/Mitogen-Activated
Protein Kinase Regulates Actin Organization and Cell
Motility by Phosphorylating the Actin Cross-Linking
Protein EPLIN,” Molecular and Cellular Biology, Vol. 27,
No. 23, 2007, pp. 8190-8204.
doi:10.1128/MCB.00661-07
[13] A. J. Sanders, L. Ye, M. D. Mason and W. G. Jiang, “The
Impact of EPLINα (Epithelial Protein Lost in Neoplasm)
on Endothelial Cells, Angiogenesis and Tumorigenesis,”
Angiogenesis, Vol. 13, No. 4, 2010, pp. 317-326.
doi:10.1007/s10456-010-9188-7