Seminal plasma induces angiogenic chemokine expression in cervical cancer cells and
regulates vascular function
Kurt J. Salesa,⁎,1, Jason R. Sutherlanda,1, Henry N. Jabbourb, Arieh A. Katza,⁎
aMRC/UCTResearch Group for ReceptorBiology,Institute of Infectious Disease and Molecular Medicine and Division ofMedicalBiochemistry, Faculty ofHealthSciences,University of Cape Town,
7925, South Africa
bMRC Human Reproductive Sciences Unit, The Queen's Medical Research Institute, The University of Edinburgh, Edinburgh, EH16 4TJ, UK
a b s t r a c ta r t i c l ei n f o
Received 24 April 2012
Received in revised form 14 June 2012
Accepted 15 June 2012
Available online 23 June 2012
Cervical cancer is one of the leading gynecological malignancies in women. We have recently shown that
seminal plasma (SP) can regulate the inflammatory cyclooxygenase-prostaglandin pathway and enhance
the growth of cervical epithelial tumours in vivo by promoting cellular proliferation and alteration of vascular
function. This study investigated the molecular mechanism whereby SP regulates vascular function using an
in vitro model system of HeLa cervical adenocarcinoma cells and human umbilical vein endothelial cells
(HUVECs). We found that SP rapidly enhanced the expression of the angiogenic chemokines, interleukin
(IL)-8 and growth regulated oncogene alpha (GRO) in HeLa cells in a time-dependent manner. We investigat-
ed the molecular mechanism of SP-mediated regulation of IL-8 and GRO using a panel of chemical inhibitors
of cell signalling. We found that treatment of HeLa cells with SP elevated expression of IL-8 and GRO by trans-
activation of the epidermal growth factor receptor, activation of extracellular signal-regulated kinase and in-
duction of cyclooxygenase enzymes and nuclear factor kappa B. We investigated the impact of IL-8 and GRO,
released from HeLa cells after treatment with SP, on vascular function using a co-culture model system of
conditioned medium (CM) from HeLa cells, treated with or without SP, and HUVECs. We found that CM
from HeLa cells induced the arrangement of endothelial cells into a network of tube-like structures via the
CXCR2 receptor on HUVECs. Taken together our data outline a molecular mechanism whereby SP can alter
vascular function in cervical cancers via the pro-angiogenic chemokines, IL-8 and GRO.
© 2012 Elsevier B.V. All rights reserved.
Chronic inflammation has been linked to increased cancer risk in
numerous organs including the liver and colon [1,2]. Non-steroidal
anti-inflammatory drug use and suppression of the inflammatory cy-
clooxygenase (COX)-prostaglandin (PG) axis have been shown over
the last two decades to have beneficial effects in reducing inflamma-
tion and have been associated with an inverse risk of developing can-
cer. These observations have been confirmed by recent clinical trials
that have shown that long term aspirin treatment can be beneficial
for reducing the burden of colorectal cancer .
In sub-Saharan African countries, cancer of the uterine–cervix is
the leading gynecological malignancy [4,5]. The main etiological fac-
tor associated with cervical cancer is infection of the cervix by
human papillomavirus (HPV) . Following infection, HPV oncogenes
have been shown to regulate the COX-PG axis in cervical cancer cells
to promote inflammation, persistent infection and tumorigenesis
[7,8]. We and others have shown that the inflammatory COX-PG
axis is elevated in cervical cancers . COX enzymes, of which there
are two isoforms in humans, namely COX-1 and COX-2, catalyse the
rate limiting conversion of arachidonic acid to PG . Following
their biosynthesis, PG are actively transported from the cell, where
they act locally in an autocrine/paracrine manner via PG receptors
. We have shown that PG, biosynthesised following induction of
COX-1 and COX-2 in cervical adenocarcinoma cells, elevates the ex-
pression of potent pro-angiogenic factors such as basic fibroblast
growth factor 2 (FGF2), vascular endothelial growth factor (VEGF)
and angiopoietins [9,10]. Following their biosynthesis and release
from neoplastic cervical epithelial cells, angiogenic factors can then
exert a paracrine activity on endothelial cells to enhance blood supply
via angiogenesis to facilitate tumour growth . These observations
have led us to propose that suppression of the inflammatory COX-PG
axis with affordable non-steroidal anti-inflammatory drugs like aspi-
rin could be beneficial for preventing cervical cancer progression in
resource-poor countries in Africa, by suppressing potent inflammato-
ry and angiogenic pathways that can promote disease progression.
Biochimica et Biophysica Acta 1823 (2012) 1789–1795
⁎ Corresponding authors at: MRC/UCT Research Group for Receptor Biology, Room
N.2.02 Wernher and Beit North, Institute of Infectious Disease and Molecular Medicine
and Division of Medical Biochemistry, UCT Faculty of Health Sciences, Observatory
7925, South Africa.
E-mail addresses: email@example.com (K.J. Sales), firstname.lastname@example.org (A.A. Katz).
1These authors contributed equally to this work.
0167-4889/$ – see front matter © 2012 Elsevier B.V. All rights reserved.
Contents lists available at SciVerse ScienceDirect
Biochimica et Biophysica Acta
journal homepage: www.elsevier.com/locate/bbamcr
In addition to HPV regulation of the COX-PG axis, we have recently
shown that seminal plasma (SP) can elevate the COX-PG axis and en-
hance the growth of cervical cancer cells in vitro and in nude mouse
xenografts in vivo . In our recent study, we found that the
SP-induced tumour growth in nude mice was associated with an al-
teration in the vasculature, with xenograft tumours arising from ani-
mals injected with SP yielding significantly larger blood vessels
compared with controls . These observations led us to propose
that the enhanced tumour growth we observed in xenografts arising
from mice injected with SP was due to an increase in the capacity of
blood vessels to supply the tumour with nutrients and oxygen .
Our observations imply that in sexually active women, repeated ex-
posure of neoplastic epithelial cells to seminal fluid could enhance tu-
morigenesis by inducing inflammatory and angiogenic pathways in
the cervical epithelium.
Several angiogenic factors have been shown to regulate blood ves-
sel branching, sprouting and maturation to facilitate tumour angio-
genesis, including VEGF and FGF2 [12,13]. In the present study, we
investigated the regulation of two potent pro-angiogenic chemokines,
interleukin (IL)-8 (also called CXCL8) and growth regulated oncogene
thelial cell function in vitro. Both IL-8 and GRO are related proteins,
which belong to the C-X-C motif superfamily of chemokines and func-
tion via the same G protein-coupled receptor, CXCR2, present on en-
dothelial cells [14–16]. Here we show that SP induces the expression
of IL-8 and GRO in a time-dependent manner via the epidermal
growth factor receptor (EGFR) and extracellular signal-regulated ki-
nase (ERK)-mediated induction of COX enzymes in cervical cancer
(HeLa) cells. Using a co-culture model system of conditioned medium
from HeLa cells, treated with or without SP and endothelial cells, we
found that the IL-8 and GRO can regulate vascular function in vitro
via the CXCR2 receptor on endothelial cells.
2. Materials and methods
2.1. Human ethics
Ethics approval for the study was obtained from the University of
Cape Town Research Ethics Committee (REC/REF: 152/99). Written
informed consent was obtained from all subjects before sample
Culture medium, penicillin–streptomycin and fetal-calf serum
(FCS) were purchased from Highveld Biological (PTY) Limited (Cape
Town, South Africa). PBS, BSA, Percoll and Tri-reagent® were pur-
chased from Sigma Chemical Company (Cape Town, South Africa).
from Cell Signalling Technologies/New England Biolabs (Hertfordshire,
UK). SB225002 was purchased from Calbiochem (Nottingham, UK).
2.3. Cell culture
Wild type HeLa-S3 cells, authenticated and verified as cervical ad-
enocarcinoma cells, were purchased from BioWhittaker (Berkshire,
UK) and were maintained as described . Conditioned medium
(CM) was prepared by seeding HeLa cells at a density of 2×106cells
in a 75cm flask. Cells were treated with 20ml of serum free DMEM in
the presence of 1:100 dilution of SP or vehicle (PBS) for 24h to create
HeLa SP conditioned medium (SP CM) or vehicle conditioned medium
(V CM). Conditioned medium from four independent experiments
was pooled, aliquoted and stored at −20°C until required for network
assays. For HeLa cell experiments, cells were incubated in the presence
of vehicle (PBS) or 1:100 dilution of SP or SP and inhibitors of EGFR
kinase (AG1478; 200nM), ERK1/2 kinase (PD98059; 50μM), COX-1
(SC560; 10μM), COX-2 (NS398; 10μM) or nuclear factor kappa B
(NFκB, SN50; 100μg/ml) for the time indicated. Human umbilical
vein endothelial cells (HUVECs) (Lonza, Walkersville, USA) were cul-
tured in Endothelial Basal Medium (EBM-2) with 2% FCS and growth
supplements (VEGF, FGF, PGDF, IGF, EGF, ascorbic acid, heparin and
gentamicin) subsequently referred to as Endothelial Growth Medium
(EGM-2) (Lonza, Walkersville, USA). Under experimental conditions
HUVECs were incubated with EBM-2 plus 1% FCS with the addition of
ascorbic acid and gentamicin (EBM1%) (Lonza, USA).
2.4. Semen donors and preparation
Semen was obtained from 10 healthy males attending the Andrology
Cape Town. Ejaculates were collected in sterile specimen jars following
voluntary self-masturbation. Seminal plasma (SP) was isolated from
the pooled ejaculate by density gradient centrifugation as described pre-
viously . The seminal plasma was aliquoted and stored at −80°C
until required and used at a concentration of 1:100 for in vitro studies
2.5. Taqman quantitative RT-PCR
HeLa cell RNA was extracted and reverse transcribed as described
previously . Quantitative RT-PCR was performed under standard
operating conditions using an ABI Prism 7500 Real-time PCR machine
(Applied Biosystems, Warrington, UK), using sequence-specific
primers and 6-carboxyfluoresceine-labelled probes [11,15,16]. RNA
loading of all target genes was normalised using the 18S ribosomal
RNA as an internal standard. Data were analysed by the comparative
Ct method and processed using Sequence Detector v1.6.3 (Applied
Biosystems). All results were calculated relative to an endogenous
control of HeLa cell cDNA included in each experiment. Fold increase
was determined by dividing the relative expression of the treatment
group by the relative expression of the control group. Data are pres-
ented as mean±SEM.
2.6. Western blot analysis
Immunoblot analysis was carried out using a LI-COR Odyssey™ in-
frared imager (LI-COR Biosciences, Cambridge, UK) . Briefly 20μg
of total protein was isolated from HeLa cells and resolved on a sodium
dodecyl sulphate polyacrylamide gel and transferred to PVDF mem-
brane. Immunoblots were blocked in Odyssey Blocking buffer™
(LI-COR Biosciences) before overnight incubation with primary rab-
bit phospho-p42/44 and mouse p42/44 antibodies (diluted 1:1000
in Odyssey blocking buffer) at 4°C. The following day, blots were
washed and incubated with the goat anti-mouse IRDye™ 800
(1:10,000) (Rockland Immunochemicals Inc., Gillbertsville, PA,
USA) and goat anti-rabbit Alexafluor 680 (1:5000) (Invitrogen, Pais-
ley, UK) for 60min at room temperature. Immunoreactive proteins
were detected and quantified using the Odyssey infrared imaging
system (LI-COR Biosciences). ERK1/2 phosphorylation was calculated
by dividing the value obtained from the phosphorylated ERK1/2
channel (700nm) by the value obtained from total ERK1/2 channel
(800nm) and expressed as fold above vehicle controls. Results are
expressed as mean±SEM.
Secreted IL-8 and GRO were measured in the HeLa cell CM using a
commercially available enzyme linked immunosorbent assay (R&D
Systems, Oxford, UK). CM was prepared as described in Section 2.3.
A control flask containing 1:100 dilution of SP in serum free medium
alone (without HeLa cells) was prepared in parallel to the HeLa cell
CM described in Section 2.3 to investigate the levels of IL-8 and GRO
K.J. Sales et al. / Biochimica et Biophysica Acta 1823 (2012) 1789–1795
in the SP. The ELISA was carried out as per the manufacturer's protocol
on individual batches of CM prior to pooling for network assays. Data
are presented as mean±SEM from four independent experiments.
2.8. Tube-like structure (endothelial network) assays
Endothelial network (tube-like structure; TLS) formation was
measured using 12-well Transwell plates (Corning Costar, Cambridge,
UK). The upper chambers were coated with 80μl of growth factor
(GF)-reduced Matrigel (BD Biosciences, MA, USA) in the absence/
presence of the CXCR2 antagonist SB225002 (100nM) and incubated
at 37C for 30min to allow thin gel formation. HUVECs were plated
onto the gel (2.5×104cells/well) in EBM 1%. In the lower chamber
CM or medium containing 1:100 dilution of SP was added. Transwell
plates were incubated at 37°C in a 5% CO2atmosphere for 16h to
allowtheHUVECs to form polygonalnetworksof three dimensional en-
dothelial TLS. Subsequently, wells were fixed with 100% ice cold meth-
anol and stained with haematoxylin. To assess the ability of HUVECs to
form networks of TLS, five photomicrographs (×10 magnification) of
each well were captured using an inverted microscope and camera
(Axiovert 200, Carl Zeiss, Germany). TLS formation was quantified by
counting the number of network branch points (as indicated by the
arrow in Fig. 5) per photomicrograph by an investigator blinded to
treatment. Experiments were repeated four times in duplicate. Fold dif-
ference was determined by dividing the value obtained from SP or SP
CM treated cells by the value obtained from V CM treated cells. Data
are represented as percentage increase in network formation with V
CM=100% and are presented as mean±SEM.
2.9. Statistical analysis
The data in this study was analysed by t-test or one-way ANOVA
using Graph Pad Prism 4.0c (Graph Pad, San Diego, CA). Paired T-tests
wereconducted onthe untransformedmeans of thereplicates between
SP and control or inhibitor and inhibitor and SP for each experiment.
Unpaired T-tests were conducted on SP versus SP and inhibitor after
conversion to fold/percentage increase. ANOVA was used to determine
significant difference between the various time points for IL-8 and GRO
by real-time PCR and phosphorylation of ERK1/2 in response to SP
3.1. Seminal plasma induces expression of IL-8 and GRO in HeLa cells
We investigated whether exposure of neoplastic cervical epithelial
cells to SP would induce expression of the angiogenic chemokines
IL-8 and GRO. HeLa cervical adenocarcinoma cells were used as a
model system and were treated with 1:100 dilution of SP or vehicle
(PBS) for 4, 8, 12 and 24h and the mRNA expression of IL-8 and
GRO was determined by quantitative RT-PCR analysis. We found
that SP significantly elevated expression of IL-8 and GRO at all time
points investigated with maximal expression observed for IL-8 after
12h (Fig. 1A; 250.3±20.7 fold increase compared to vehicle treat-
ment; Pb0.01) and for GRO after 8h of SP treatment (Fig. 1B; 18.4±
2.5-fold increase compared to vehicle treatment; Pb0.01).
3.2. IL-8 and GRO expression in HeLa cells is regulated by SP via the
We investigated the pathway whereby SP mediates the induction
of IL-8 and GRO in HeLa cells using a panel of small molecule chemical
inhibitors of cell signalling. HeLa cells were treated for 8h with vehi-
cle (Fig. 2A and B; white bar) or 1:100 dilution of SP (Fig. 2A and B;
black bar) in the presence of chemical inhibitors (Fig. 2A and B;
dark grey bar) of EGFR kinase (AG1478; 200nM), ERK1/2 kinase
(PD98059; 50μM), COX-1 (SC560; 10μM), COX-2 (NS398; 10μM) or
nuclear factor kappa B (NFκB, SN50; 100μg/ml). We found that all
chemical inhibitors significantly reduced the SP-mediated induction
of IL-8 in HeLa cells with an average percentage inhibition of 65%,
70%, 55%, 98% and 70% for AG1478, PD98059, SC560, NS398 and
SN50 respectively (Fig. 2A; dark grey bar; Pb0.01). Similarly we
found that AG1478, PD98059, SC560, NS398 and SN50 inhibited
SP-mediated induction of GRO in HeLa cells by 65%, 70%, 50%, 98%
and 60% respectively (Fig. 2B; dark grey bar; Pb0.01). Treatment of
cells with inhibitor alone (Fig. 2A and B; light grey bar) had no signif-
icant effect on expression of IL-8 and GRO in HeLa cells compared
with vehicle treated cells (Fig. 2A and B; white bar).
We have previously shown that SP can mediate signalling to
ERK1/2 via transactivation of the EGFR . We used immunoblot
analysis to investigate whether the EGFR, COX and NFκB pathways
lay upstream or downstream of ERK1/2 in the present study. HeLa
cells were treated with vehicle or 1:100 dilution of SP for 0, 5, 10,
20, 30 and 60min and ERK1/2 phosphorylation was measured by im-
munoblot analysis. We found a rapid time-dependent increase in
ERK1/2 phosphorylation, reaching a maximum after 10min of stimu-
lation (Fig. 3A; 7.5±1.2 fold increase above control; Pb0.05). We
next treated HeLa cells with vehicle or 1:100 dilution of SP alone or
in the presence of chemical inhibitors of EGFR kinase (AG1478),
ERK1/2 kinase (PD98059), COX-1 (SC560), COX-2 (NS398) or NFκB
(SN50) for 10min and measured ERK1/2 phosphorylation. We
found that SP-mediated ERK1/2 phosphorylation was significantly re-
duced by an average of 80% and 98% respectively in cells co-treated
with SP and inhibitors of EGFR kinase (Fig. 3B; Pb0.05) and ERK1/2
kinase (Fig. 3B; Pb0.01). We found that co-treatment of HeLa cells
with SP and inhibitors of COX-1, COX-2 or NFκB had no inhibitory ef-
fect on ERK phosphorylation in the presence of SP (Fig. 3B). These
findings suggested that the EGFR pathway lay upstream of ERK1/2,
whereas the COX-NFκB pathways lay downstream of ERK1/2.
Since our data suggest that SP sequentially transactivates the EGFR
and then activates ERK1/2 to up-regulate IL-8 and GRO via the
COX-NFκB pathway, we investigated the regulation of COX-1 and
Fig. 1. Upregulation of IL-8 and GRO expression in HeLa cells in response to treatment
with seminal plasma. (A) IL-8 and (B) GRO mRNA expression was measured by quan-
titative RT-PCR analysis and is significantly increased by SP (1:100) after 4, 8, 12 and
24h (H) compared to vehicle (PBS) control, (*Pb0.05; **Pb0.01). Data are represented
as mean±SEM from 5 independent experiments.
K.J. Sales et al. / Biochimica et Biophysica Acta 1823 (2012) 1789–1795
COX-2 mRNA expression by SP. HeLa cells were treated for 8h with
vehicle (Fig. 4A and B; white bar) or 1:100 dilution of SP (Fig. 4A
and B; black bar) in the presence of our panel of chemical inhibitors
(Fig. 4A and B; dark grey bar). We found that SP significantly induced
COX-1 (Fig. 4A; black bar; Pb0.05) and COX-2 (Fig. 4B; black bar;
Pb0.05) mRNA expression. Moreover we found that COX-1 mRNA ex-
pression was inhibited by an average of 65% and 70% respectively by
co-treatment of cells with EGFR kinase (AG1478) and ERK1/2 kinase
(PD98059; Fig. 4A; Pb0.01) inhibitors. Co-treatment of HeLa cells
with inhibitors of COX-1 (SC560), COX-2 (NS398) or NFκB (SN50;
Fig. 4A) had no significant effect on reducing the SP-mediated induc-
tion of COX-1. In contrast, SP-mediated induction of COX-2 mRNA ex-
pression was significantly reduced by an average of 60%, 98% and 97%
respectively by co-treatment of HeLa cells with SP and the EGFR ki-
nase (AG1478), ERK1/2 kinase (PD98059), and COX-2 (NS398;
Fig. 4B; Pb0.01) inhibitors. Co-treatment of cells with inhibitors of
COX-1 (SC560) or NFκB (SN50; Fig. 4B) had no effect on reducing
the SP-mediated induction of COX-2. These data suggest that SP se-
quentially activates EGFR and ERK1/2 signalling to induce inflamma-
tory COX enzyme expression and regulate IL-8 and GRO expression
via the NFκB pathway.
3.3. IL-8 and GRO, released from HeLa cells in response to SP treatment,
regulates vascular function
We measured IL-8 and GRO in the SP (1:100 dilution) and in the
CM by ELISA. We found IL-8 to be present at 8.5±2.3pg/ml in the di-
luted SP (1:100), consistent with other studies . In contrast, we
did not detect GRO, at the dilution of SP used in this study, by
ELISA. Relative to diluted SP in culture medium alone, we found a
3.1±0.5 fold increase in IL-8 levels in the SP CM. In addition, we
found that SP treatment of HeLa cells for 24h increased the secretion
of IL-8 and GRO proteins by 6.0±4.5 and 9.8±2.3 fold, respectively in
HeLa cell CM compared to vehicle treatment (Pb0.05). To assess the
effects of the CM on vascular function, we used an in vitro endothelial
cell tube-formation (network formation) assay . For this we used
a co-culture system of conditioned medium (CM) from HeLa cells
treated with vehicle (V CM) or 1:100 dilution of SP for 24h (SP CM)
or serum‐free culture medium containing 1:100 dilution of SP alone
(SP only) and HUVECs. Treatment of HUVECs with SP CM significantly
increased endothelial cell network formation, as indicated by the abil-
ity of HUVECs to form stabilised polygonal structures comprised of
three dimensional capillary-like tubes (as indicated by the arrow in
Fig. 5), compared to V CM-treated cells (Fig. 5; Pb0.001). Treatment
of HUVECs with SP CM in the presence of the CXCR2 antagonist
SB225002 inhibited endothelial network formation to the levels ob-
served for V CM treated HUVECS, confirming that these alterations
in endothelial cell function were mediated by IL-8 and GRO via their
common receptor CXCR2 (Fig. 5). Interestingly, co-culture of HUVECs
with medium containing 1:100 dilution of SP alone also significantly
increased network formation, but to a lesser extent compared with
CM from HeLa cells treated with SP (Fig. 5; Pb0.05).
Inflammation involves tissue remodelling events which act across
multiple cellular compartments in the face of infection or injury in
order to regulate homeostasis. If left unchecked, unabated inflamma-
tion results in pathology. Indeed, chronic inflammation is estimated
to contribute to more than 25% of all new cancer cases globally
Fig. 2. Regulation of IL-8 and GRO expression in HeLa cells by seminal plasma via the
EGFR, ERK, COX and NFκB pathways. (A) IL-8 and (B) GRO mRNA expression as mea-
sured by quantitative RT-PCR analysis. HeLa cells were treated for 8h with SP
(1:100) or control in the presence/absence of chemical inhibitors of EGFR kinase
(AG1478; 200nM), ERK1/2 kinase (PD98059; 50μM), COX-1 (SC560; 10μM), COX-2
(NS398; 10μM) or NFκB (SN50; 100μg/ml). b is significantly different from a, c is sig-
nificantly different from a and b; Pb0.05. Data are represented as mean±SEM from 4
Fig. 3. ERK phosphorylation in HeLa cells by seminal fluid. (A) ERK1/2 phosphorylation
in HeLa cells treated with SP (1:100) or control for 0, 2, 5, 10, 20, 30, 60min. (B) ERK1/2
presence of inhibitors of EGFR kinase (AG1478; 200nM), ERK1/2 kinase (PD98059;
50μM), COX-1 (SC560; 10μM), COX-2 (NS398; 10μM) or NFκB (SN50; 100μg/ml).
Cell lysates were subjected to immunoblot analysis and quantified as described in
the Materials and methods (*Pb0.05; **Pb0.01). Data are represented as mean±SEM
from 5 independent experiments.
K.J. Sales et al. / Biochimica et Biophysica Acta 1823 (2012) 1789–1795
[2,19]. Our data over the past 10years have shown that the COX-PG
pathway, a critical pathway regulating inflammation and cancer is
significantly up-regulated in cervical carcinomas . Moreover, we
have shown that SP can regulate this pathway in vitro and can en-
hance the growth of neoplastic cervical cells xenografted in nude
mice in vivo [11,17].
Here we demonstrate that SP can induce the expression of the po-
tent chemokines, IL-8 and GRO in cervical cancer (HeLa) cells.
Chemokines have emerged as important regulators of cell prolifera-
tion, angiogenesis and inflammation [20–24]. Seminal plasma has
been shown to regulate the release of inflammatory cytokines, in-
cluding IL-8 and GRO from normal cervical and vaginal cells .
These modulate inflammation within the cervical compartment
after coitus by regulating the influx of leukocytes into the epithelial
and stromal compartments [25,26]. Although the precise role of the
post‐coital inflammatory response is unclear, it has been proposed
to facilitate immune tolerance to male transplantation antigens con-
tained within the ejaculate and prime the uterus for conception .
However in sexually active women with neoplastic cervical lesions,
activation of these inflammatory and pro-tumorigenic pathways by
SP could put women at risk of disease progression by exacerbating
the release of inflammatory mediators and inducing the polarisation
and recruitment of tumour associated macrophages and neutrophils
into the cervix. These have been shown to promote tissue remodeling,
angiogenesis and tumorigenesis [27,28].
We investigated the pathways whereby SP regulates IL-8 and GRO
expression in HeLa cells, using a panel of small molecule chemical in-
hibitors targeted to specific signal transduction pathways with
known roles in regulating cell growth, differentiation and angiogene-
sis. We found that SP induced IL-8 and GRO mRNA expression via the
EGFR-mediated activation of ERK1/2 signalling since inhibitors of
EGFR kinase and ERK1/2 kinase inhibited SP-mediated induction of
IL-8 and GRO. Although we have not isolated the precise ligand in
the SP responsible for activation/transactivation of the EGFR in our
study, SP is rich in epidermal growth factor (EGF), transforming
growth factor beta (TGFβ), inflammatory cytokines and prostaglan-
dins [18,29–31]. It is feasible that the EGFR could be phosphorylated
and activated directly by the EGF in the SP to induce IL-8 and GRO ex-
pression. Alternatively, PGE2which is present at high concentrations
in SP can transactivate the EGFR via the E-series PG receptors (EP2
and EP4 receptors) as we have previously shown [17,32]. In our pre-
sent study, the net effect of SP signalling to IL-8 and GRO in cervical
cancer cells is most likely mediated by a combination of ligands,
such as EGF activating its receptor directly and PGE2, transactivating
the EGFR via the phosphorylation of intracellular tyrosine kinases
Fig. 4. Regulation of COX-1 and COX-2 expression in HeLa cells by seminal plasma.
(A) COX-1 and (B) COX-2 mRNA expression as measured by quantitative RT-PCR
analysis. HeLa cells were treated for 8h with SP (1:100) or control in the presence/
absence of chemical inhibitors of EGFR kinase (AG1478), ERK1/2 kinase (PD98059),
COX-1 (SC560), COX-2 (NS398) or NFκB (SN50). b is significantly different from a,
c is significantly different from a and b; Pb0.05. ns; not significantly different for SP
versus SP and inhibitor. Data are represented as mean±SEM from 4 independent
Fig. 5. The effect of SP conditioned medium on endothelial network (TLS) formation. The ability of endothelial cells to differentiate into polygonal networks of three-dimensional
tube-like structures (endothelial network formation) was investigated using HUVECs stimulated with vehicle conditioned medium (V CM), SP conditioned medium (SP CM), 1:100
dilution of SP in culture medium alone or SP CM in the presence of CXCR2 antagonist SB225002. (*Pb0.05, **Pb0.01, ***Pb0.001). The arrowhead indicates a network branch point
within the polygonal capillary network. Data are represented as percentage increase compared to V CM and presented as mean±SEM from 4 independent experiments.
K.J. Sales et al. / Biochimica et Biophysica Acta 1823 (2012) 1789–1795
Furthermore, in our present study we show that activation of
EGFR and ERK1/2 signalling leads to the elevation of COX-1 and
COX-2 expression, as an intermediate step in the SP-mediated induc-
tion of IL-8 and GRO. We have recently shown that SP treatment of
HeLa cells elevates COX-1 and COX-2 expression in vitro and in
HeLa cells xenografted in nude mice in vivo . Moreover, we
have shown that the PG released as a consequence of elevated COX
enzyme expression in HeLa cells can regulate tumorigenic and angio-
genic gene expression in vitro [10,33]. However, the present study is
the first to demonstrate that SP regulates chemokine expression in
cervical cancer cells by first inducing the inflammatory COX-PG path-
way. Furthermore our study also shows that COX-2 expression, but not
of HeLacells with SP and the selective COX-2 inhibitor NS398 abolished
the SP-mediated induction of COX-2. The auto-regulation of COX-2 ex-
pression via a positive feedback loop has been demonstrated in several
model systems and is proposed to be the mechanism behind the role of
COX-2 in regulating chronic inflammatory events and tumorigenesis
[33,34]. Interestingly we found that the NFκB inhibitor SN50 abolished
IL-8 and GRO expression, but not COX-1 or COX-2, in response to SP.
These data suggest that IL-8 and GRO are regulated by NFκB via the
COX-PG pathway. Indeed PGE2has recently been shown to up-regulate
IL-8 expression in the endometrium via NFκB . Furthermore, consti-
tutive activation of NFκB leading to activation of chemokines has been
associated with cellular transformation, tumorigenesis and angiogenesis
in several solid tumours in humans . It is therefore plausible that
repeated exposure of neoplastic cervical epithelial cells to SP could
promote tumorigenesis by elevating NFκB signalling and inducing
In our recent study, we found that tumours arising from HeLa cell
xenografts in nude mice grew larger and had larger blood vessels in
animals injected with seminal plasma, compared with control ani-
mals . In order to investigate a potential mechanism behind the
regulation of vasculature in HeLa cell xenografts by SP, we investigat-
ed whether the IL-8 and GRO, released from HeLa cells in response to
SP, could modulate vascular function using an in vitro co-culture sys-
tem. A critical process required for angiogenesis is endothelial cell
stabilisation and differentiation into three dimensional capillary net-
works [35–37]. Recently, Maybin and colleagues showed that IL-8 can
increase angiogenesis in menstrual endometrial explants ex vivo by
lial cell tube-formation (network) assay to determine the role of IL-8
and GRO on vascular function in vitro. This assay measures the ability
of endothelial cells to form polygonal networks comprising three di-
mensional capillary-like tubular structures (TLS) when cultured on a
dium from HeLa cells treated with SP contains elevated IL-8 and GRO
and promotes the formation of HUVECs into polygonal endothelial net-
works, which were more stable and dense compared with HUVECs in-
cubated with V CM. Interestingly, incubation of HUVECs with SP alone
could alsoincrease TLS formation but toa lesser extent thanthe SP con-
ditioned medium. This is possibly due to the lower levels of IL-8 found
in SP compared with SP conditioned medium. IL-8 and GRO act via a
common G protein-coupled receptor CXCR2, present on endothelial
cells . Using a specific non-peptide antagonist of CXCR2, which pre-
vents the binding of IL-8 and GRO to their receptor, we show that con-
ditioned medium from SP treated HeLa cells, which contains both
IL-8 present in SP as well as IL-8 and GRO biosynthesised in HeLa
cells in response to SP signalling, promotes endothelial cell net-
work formation via CXCR2 on endothelial cells, since incubation
of HUVECs with SP CM and the CXCR2 antagonist disrupted endo-
thelial TLS formation. Stabilised networks of endothelial tube-like
structures would then allow other pro-angiogenic factors such as
VEGF and FGF2 to interact with endothelial cells to promote prolif-
eration and angiogenesis. These findings suggest a potential mech-
anism for the alteration of the vasculature we observed in HeLa cell
xenografts in nude mice injected with SP, in our previous study
In conclusion, our study as summarised schematically in Fig. 6,
outlines the regulation of pro-angiogenic chemokines IL-8 and
GRO by SP in neoplastic cervical epithelial cells and their role in
vascular function. SP induces ERK1/2 via transactivation of the
EGFR. This in turn induces biosynthesis and signalling of PG to
NFκB via upregulation of COX enzyme expression. IL-8 and GRO,
once released from neoplastic epithelial cells can induce the ar-
rangement of endothelial cells into networks of tube-like structures
to alter vascular function. This alteration of vascular function can be
further enhanced by the endogenous IL-8 present in the SP. These
findings highlight that suppression of the inflammatory COX-PG
axis can attenuate the release of pro-inflammatory cytokines from
cervical epithelial cells exposed to SP. Our findings further suggest
that investigation of non-steroidal anti-inflammatory drugs as inter-
vention strategies for preventing cervical inflammation and tumor-
igenesis is warranted.
Conflict of interest
There are no conflicts of interest to declare.
This study was supported by UK MRC core funding to HNJ
(U.1276.00.004.00002.01), by the South African MRC to AAK and KJS
and by Poliomyelitis Research Foundation and Cancer Association of
South Africa (CANSA) to KJS. The funding bodies played no role in
Fig. 6. Schematic summary highlighting the role of SP in regulating vascular function.
Our findings demonstrate that SP activates ERK1/2 signalling via the transactivation
of the EGFR. This induces expression of inflammatory COX-1 and COX-2 genes resulting
in production of PG and activation of PG receptors [10,33]. This pathway then regulates
COX-2 expression in an autoregulatory manner and regulates IL-8 and GRO via NFκB
. In turn, IL-8 and GRO act via a common receptor, CXCR2 on endothelial cells to
alter vascular function by promoting the formation of endothelial cell tube‐like struc-
tures and network formation. In addition endogenous IL-8 present in SP can enhance
the effect of the IL-8 and GRO, released from HeLa cells in response to SP signalling,
on endothelial network formation.
K.J. Sales et al. / Biochimica et Biophysica Acta 1823 (2012) 1789–1795
the study design; nor in the collection, analysis and interpretation of
data; or in the writing of the report; or the decision to submit the
article for publication.
 J.D. Lewis, J.J. Deren, G.R. Lichtenstein, Cancer risk in patients with inflammatory
bowel disease, Gastroenterol. Clin. North Am. 28 (1999) 459–477.
 L.M. Coussens, Z. Werb, Inflammation and cancer, Nature 420 (2002) 860–867.
 J. Burn, A.M. Gerdes, F. Macrae, J.P. Mecklin, G. Moeslein, S. Olschwang, D. Eccles,
D.G. Evans, E.R. Maher, L. Bertario, M.L. Bisgaard, M.G. Dunlop, J.W. Ho, S.V.
Hodgson, A. Lindblom, J. Lubinski, P.J. Morrison, V. Murday, R. Ramesar, L. Side,
R.J. Scott, H.J. Thomas, H.F. Vasen, G. Barker, G. Crawford, F. Elliott, M.
Movahedi, K. Pylvanainen, J.T. Wijnen, R. Fodde, H.T. Lynch, J.C. Mathers, D.T.
Bishop, Long-term effect of aspirin on cancer risk in carriers of hereditary colorec-
tal cancer: an analysis from the CAPP2 randomised controlled trial, Lancet (2011)
 F. Sitas, J. Madhoo, J. Wessie, Incidence of histologically diagnosed cancer in South
Africa, 1993–1995, National Cancer Registry of South Africa, South African Insti-
tute for Medical Research, Johannesburg, 1998.
 M. Arbyn, X. Castellsague, S. de Sanjose, L. Bruni, M. Saraiya, F. Bray, J. Ferlay,
Worldwide burden of cervical cancer in 2008, Ann. Oncol. 22 (2011) 2675–2686.
 H. zur Hausen, Papillomaviruses in the causation of human cancers—a brief his-
torical account, Virology 384 (2009) 260–265.
 K. Subbaramaiah, A.J. Dannenberg, Cyclooxygenase-2 transcription is regulated
by human papillomavirus 16 E6 and E7 oncoproteins: evidence of a
corepressor/coactivator exchange, Cancer Res. 67 (2007) 3976–3985.
 J.M. Oh, S.H. Kim, Y.I. Lee, M. Seo, S.Y. Kim, Y.S. Song, W.H. Kim, Y.S. Juhnn, Human
papillomavirus E5 protein induces expression of the EP4 subtype of prostaglandin
E2 receptor in cyclic AMP response element-dependent pathways in cervical can-
cer cells, Carcinogenesis 30 (2009) 141–149.
 K.J. Sales, H.N. Jabbour, Cyclooxygenase enzymes and prostaglandins in reproduc-
tive tract physiology and pathology, Prostaglandins Other Lipid Mediat. 71 (2003)
 K.J. Sales, A.A. Katz, B. Howard, R.P. Soeters, R.P. Millar, H.N. Jabbour, Cyclooxy-
genase-1 is up-regulated in cervical carcinomas: autocrine/paracrine regulation
of cyclooxygenase-2, prostaglandin e receptors, and angiogenic factors by
cyclooxygenase-1, Cancer Res. 62 (2002) 424–432.
 J.R. Sutherland, K.J. Sales, H.N. Jabbour, A.A. Katz, Seminal plasma enhances cervi-
cal adenocarcinoma cell proliferation and tumour growth in vivo, PLoS One 7
 M.C. Keightley, P. Brown, H.N. Jabbour, K.J. Sales, F-Prostaglandin receptor regu-
lates endothelial cell function via fibroblast growth factor-2, BMC Cell Biol. 11
 J.A. Maybin, N. Hirani, H.N. Jabbour, H.O. Critchley, Novel roles for hypoxia and
prostaglandin E2 in the regulation of IL-8 during endometrial repair, Am. J. Pathol.
178 (2011) 1245–1256.
 D. Wang, H. Wang, J. Brown, T. Daikoku, W. Ning, Q. Shi, A. Richmond, R. Strieter,
S.K. Dey, R.N. DuBois, CXCL1 induced by prostaglandin E2 promotes angiogenesis
in colorectal cancer, J. Exp. Med. 203 (2006) 941–951.
 K.J. Sales, D. Maldonado-Perez, V. Grant, R.D. Catalano, M.R. Wilson, P. Brown,
A.R. Williams, R.A. Anderson, E.A. Thompson, H.N. Jabbour, Prostaglandin
F(2alpha)-F-prostanoid receptor regulates CXCL8 expression in endometrial
adenocarcinoma cells via the calcium-calcineurin-NFAT pathway, Biochim.
Biophys. Acta 1793 (2009) 1917–1928.
 A.E. Wallace, K.J. Sales, R.D. Catalano, R.A. Anderson, A.R. Williams, M.R. Wilson, J.
Schwarze, H. Wang, A.G. Rossi, H.N. Jabbour, Prostaglandin F2alpha-F-prostanoid
receptor signalling promotes neutrophil chemotaxis via chemokine (C-X-C motif)
ligand 1 in endometrial adenocarcinoma, Cancer Res. 69 (2009) 5726–5733.
 M. Muller, K.J. Sales, A.A. Katz, H.N. Jabbour, Seminal plasma promotes the ex-
pression of tumorigenic and angiogenic genes in cervical adenocarcinoma cells
via the E-series prostanoid 4 receptor, Endocrinology 147 (2006) 3356–3365.
 J.A. Politch, L. Tucker, F.P. Bowman, D.J. Anderson, Concentrations and significance
of cytokines and other immunologic factors in semen of healthy fertile men, Hum.
Reprod. 22 (2007) 2928–2935.
 S. Perwez Hussain, C.C. Harris, Inflammation and cancer: an ancient link with
novel potentials, Int. J. Cancer 121 (2007) 2373–2380.
 A. Arici, E. Seli, L.M. Senturk, L.S. Gutierrez, E. Oral, H.S. Taylor, Interleukin-8 in the
human endometrium, J. Clin. Endocrinol. Metab. 83 (1998) 1783–1787.
 A. Arici, E. Seli, H.B. Zeyneloglu, L.M. Senturk, E. Oral, D.L. Olive, Interleukin-8 in-
duces proliferation of endometrial stromal cells: a potential autocrine growth fac-
tor, J. Clin. Endocrinol. Metab. 83 (1998) 1201–1205.
 U.A. Kayisli, N.G. Mahutte, A. Arici, Uterine chemokines in reproductive physiology
and pathology, Am. J. Reprod. Immunol. 47 (2002) 213–221.
tors in endometriosis, Hum. Reprod. 20 (2005) 794–801.
 A. Richmond, G.H. Fan, P. Dhawan, J. Yang, How do chemokine/chemokine recep-
tor activations affect tumorigenesis? Novartis Found. Symp. 256 (2004) 74–89
(discussion 89–91, 106–111, 266–109).
 D.J. Sharkey, A.M. Macpherson, K.P. Tremellen, S.A. Robertson, Seminal plasma
differentially regulates inflammatory cytokine gene expression in human cervical
and vaginal epithelial cells, Mol. Hum. Reprod. 13 (2007) 491–501.
 D.J. Sharkey, K.P. Tremellen, M.J. Jasper, Robertson S.A. Gemzell-Danielsson, Seminal
fluid induces leukocyte recruitment and cytokine and chemokine mRNA expression
in the human cervix after coitus, J. Immunol. 188 (5) (2012) 2445–2454.
 F. Balkwill, A. Mantovani, Inflammation and cancer: back to Virchow? Lancet 357
 S.K. Biswas, A. Mantovani, Macrophage plasticity and interaction with lympho-
cyte subsets: cancer as a paradigm, Nat. Immunol. 11 (2010) 889–896.
 R.W. Kelly, Immunosuppressive mechanisms in semen: implications for contra-
ception, Hum. Reprod. 10 (1995) 1686–1693.
 S.A. Robertson, Seminal plasma and male factor signalling in the female reproduc-
tive tract, Cell Tissue Res. 322 (2005) 43–52.
 Y. Hirata, M. Uchihashi, M. Hazama, T. Fujita, Epidermal growth factor in human
seminal plasma, Horm. Metab. Res. 19 (1987) 35–37.
plasma and prostaglandin E2 up-regulate fibroblast growth factor 2 expression in
endometrial adenocarcinoma cells via E-series prostanoid-2 receptor-mediated
transactivation of the epidermal growth factor receptor and extracellular
signal-regulated kinase pathway, Hum. Reprod. 22 (2007) 36–44.
 H.N. Jabbour, K.J. Sales, S.C. Boddy, R.A. Anderson, A.R. Williams, A positive feed-
back loop that regulates cyclooxygenase-2 expression and prostaglandin F2alpha
synthesis via the F-series-prostanoid receptor and extracellular signal-regulated
kinase 1/2 signalling pathway, Endocrinology 146 (2005) 4657–4664.
 H. Fujino, J.W. Regan, Prostaglandin F(2alpha) stimulation of cyclooxygenase-2 pro-
moter activity by the FP(B) prostanoid receptor, Eur. J. Pharmacol. 465 (2003) 39–41.
 D.H. Ausprunk, J. Folkman, Migration and proliferation of endothelial cells in pref-
ormed and newly formed blood vessels during tumor angiogenesis, Microvasc.
Res. 14 (1977) 53–65.
 J. Folkman, C. Haudenschild, Angiogenesis in vitro, Nature 288 (1980) 551–556.
 D. Hanahan, J. Folkman, Patterns and emerging mechanisms of the angiogenic
switch during tumorigenesis, Cell 86 (1996) 353–364.
K.J. Sales et al. / Biochimica et Biophysica Acta 1823 (2012) 1789–1795