Effects of human breast stromal cells on conjugated linoleic acid (CLA) modulated vascular endothelial growth factor-A (VEGF-A) expression in MCF-7 cells.

Abstract

Conjugated linoleic acid (CLA), a naturally occurring compound found in ruminant dairy and beef products, has been shown to possess anti-cancer ability in vivo and in vitro. There are several CLA isomers in ruminant-produced food products, among which t10,c12-CLA and c9,t11-CLA are most potent. Vascular endothelial growth factor-A (VEGF-A) has been implicated as an angiogenesis-activating cytokine. Our previous results indicated that CLA induced suppression of VEGF-A in MCF-7 cells, which may be one of CLA's anticancer mechanisms.
The effects of 10,c12-CLA and c9,t11-CLA on VEGF-A mRNA and protein expression in MCF-7 cells, which were co-cultured with human breast stromal cells isolated from breast tissues of surgical specimens of mammoplasty and breast cancer patients, were detected by RT-PCR and Western blot analysis.
VEGF-A mRNA and protein expressions were significantly (p < 0.05) elevated in co-cultured MCF-7 cells in comparison with cultured MCF-7 cells alone. Normal human breast stromal cells contribute greater effects in increasing VEGF-A protein expression in MCF-7 cells. Both t10,c12-CLA and c9,t11-CLA significantly (p < 0.05) decreased VEGF-A mRNA and protein levels in co-cultured MCF-7 cells. t10,c12-CLA appeared to be the more active isomer than c9, t11-CLA.
The results indicate that dietary CLA might serve as a chemo-therapeutic agent in human breast cancers by down-regulating VEGF-A expression.

Full text

Page 1

Abstract. Background: Conjugated linoleic acid (CLA), a
naturally occurring compound found in ruminant dairy and
beef products, has been shown to possess anti-cancer ability in
vivo and in vitro. There are several CLA isomers in ruminant-
produced food products, among which t10,c12-CLA and
c9,t11-CLA are most potent. Vascular endothelial growth
factor-A (VEGF-A) has been implicated as an angiogenesis-
activating cytokine. Our previous results indicated that CLA
induced suppression of VEGF-A in MCF-7 cells, which may
be one of CLA’s anticancer mechanisms. Materials and
Methods: The effects of t10,c12-CLA and c9,t11-CLA on
VEGF-A mRNA and protein expression in MCF-7 cells, which
were co-cultured with human breast stromal cells isolated from
breast tissues of surgical specimens of mammoplasty and
breast cancer patients, were detected by RT-PCR and Western
blot analysis. Results: VEGF-A mRNA and protein expressions
were significantly (p<0.05) elevated in co-cultured MCF-7
cells in comparison with cultured MCF-7 cells alone. Normal
human breast stromal cells contribute greater effects in
increasing VEGF-A protein expression in MCF-7 cells. Both
t10,c12-CLA and c9,t11-CLA significantly (p<0.05) decreased
VEGF-A mRNA and protein levels in co-cultured MCF-7
cells. t10,c12-CLA appeared to be the more active isomer than
c9,t11-CLA. Conclusion: The results indicate that dietary CLA
might serve as a chemo-therapeutic agent in human breast
cancers by down-regulating VEGF-A expression.
CLA is produced by rumen fermentation of linoleic acid
and is deposited in the subcutaneous fat (sub-Q) and
intramuscular fat (IM or marbling) layer in beef cattle, and
is also present in dairy milk fat (1). There are several CLA
isomers in ruminant-produced foods, among which c9,t11-
CLA and t10,c12-CLA are most potent (2). CLA has been
found to possess anti-carcinogenic, anti-diabetic, anti-
atherogenic and anti-adipogenic activities in mouse and
human cell lines, and in in vivo animal studies using mice
(2). Human studies of CLA and breast cancer revealed that
a diet composed of CLA-rich foods, particularly cheese, may
protect against breast cancer in postmenopausal women (3).
It was recently reported, in a large epidemiological study,
that CLA dietary intake was associated with the regulation
of estrogen receptor (ER) expression in breast cancer
patients (4). CLA has been found to possess the ability to
reduce the risk of having an ER-negative tumor in
premenopausal breast cancer, which may lead to a better
therapeutic outcome for breast cancer patients as their
cancer will probably respond to anti-estrogen therapy (4).
CLA studies in our laboratory (5, 6) also demonstrated the
anti-mammary tumor effects of CLA on (I) anti-
angiogenesis by: suppressing the predominant vascular
endothelial growth factor (VEGF)-A isomers, VEGF-A121
and VEGF-A165, mRNA expression in MCF-7 cells; (II)
decreasing estrogenic agent-induced breast cancer cell
proliferation; and (III) regulating ER· and ER‚ in
epithelial-stromal cell interactions in human breast cancer.
Angiogenesis is the process of forming new blood vessels
from the existing vascular network. Cancer cells growing
into tiny tumors need to link up to the organ’s blood vessels
(7). Tumor-associated angiogenesis is considered to go
through two phases and is believed to be separated by the
"angiogenic switch" (8). The first is the so-called avascular
phase and tumor diameters were not more than 1-2 mm.
4061
Correspondence to: Young C. Lin, D.V.M., Ph.D., Laboratory of
Reproductive and Molecular Endocrinology, College of Veterinary
Medicine, The Ohio State University, Columbus, OH, U.S.A. Tel:
+1 614 292-9706, Fax: +1 614 292-7599, e-mail: lin.15@osu.edu
Key Words: Breast cancer, CLA, VEGF-A, linoleic acid.
ANTICANCER RESEARCH 25: 4061-4068 (2005)
Effects of Human Breast Stromal Cells on Conjugated
Linoleic Acid (CLA) Modulated Vascular Endothelial
Growth Factor-A (VEGF-A) Expression in MCF-7 Cells
LI-SHU WANG1, YI-WEN HUANG1, YASURO SUGIMOTO1,2, SULING LIU3,
HSIANG-LIN CHANG1, WEIPING YE1, SHERRY SHU1and YOUNG C. LIN1,2
1Laboratory of Reproductive and Molecular Endocrinology,
College of Veterinary Medicine, The Ohio State University, Columbus, OH 43210;
2OSU Comprehensive Cancer Center and
3University of Michigan Comprehensive Cancer Research Center, Ann Arbor, MI 48105, U.S.A.
0250-7005/2005 $2.00+.40

Page 2

This state of tumors is balanced between proliferation and
apoptosis. The second is the so-called vascular phase
referring to the potential tumor growth stage (8). The
angiogenic switch is believed to correlate with the balance
change between activating angiogenesis molecules, such as
VEGF (9-10), matrix metalloproteinase (MMP)-9 (9),
cyclooxygenase-2 (11) and hypoxia inducible factor 1 alpha
(HIF-1·) (12), and inhibiting angiogenesis molecules, such
as thrombospondin-1 (TSP1) (13-14), MMP2 (15) and Eph
receptor A2 (EphA2) (16-17). If angiogenesis was blocked,
tumor growth stopped at a diameter of about 1-2 mm,
implicating the therapeutic values of angiogenesis inhibitors
in the treatment of human cancers (7, 18). Molecules
involved in the angionenesis process are mainly secreted by
cancer epithelial cells and act on surrounding normal host
tissue (19). This signaling activates certain genes in the host
tissue that, in turn, produce proteins to encourage growth
of the tumor microvasculature, which is a critical early step
in tumor stroma generation (19). One of the most potent
angiogenesis-activating molecules is the VEGF family
discovered as a tumor-secreted protein. Unbalanced
secretion of VEGF family members induces abnormal
vessels in tumors (8, 19). The VEGF family, including
VEGF-A, -B, -C, -D and -E, is associated with angiogenesis
and lymph-angiogenesis, and acts by binding to VEGF
receptors, the tyrosine kinase receptor, mainly expressed on
the surface of vascular endothelial cells (18). Current
interest is focused on VEGF-A, which is a well-studied
VEGF family member, because of its important and unique
properties (8, 19). VEGF-A is a member of the dimeric
glycoprotein family and belongs to the platelet-derived
growth factor (PDGF) superfamily (19). The VEGF-A gene
is located on chromosome 6p21.3. Based on differential
exon splicing, four VEGF-A isomers, containing 121, 165,
189 or 206 amino acids, represented as VEGF-A121,
VEGF-A165, VEGF-A189and VEGF-A206, respectively, are
formed which contain different primary structures. In
human cells, isomers VEGF-A121, VEGF-A165
VEGF-A189
are the three major VEGF-A isomers
(corresponding murine proteins are one amino acid shorter)
(8, 19, 21). In a pancreatic-islet cancer model, the
angiogenesis switch was proved to be VEGF-A-related (10).
It has been demonstrated that ~90% of primary breast
carcinoma with high microvessel density overexpressed
VEGF-A (20). VEGF-A is overexpressed in most malignant
tumors, inducingendothelial
proliferation and protecting against endothelial cell
apoptosis and senescence (19). Another important feature
of VEGF-A is that the hypoxic state of the growing tumor
triggers expression of hypoxia-inducible factor, which binds
to the VEGF-A promoter to stimulate VEGF-A production
(21). Regulation of VEGF-A expression by hormones in
rodent and human uterus models has been shown that
and
cell migration and
estrogen- and progestins-induced VEGF-A expression can
be blocked by anti-estrogen and anti-progestin, respectively.
In addition, progestin also up-regulated VEGF-A in a
hormone-responsive breast cancer cell line (22). Studies
from breast cancer patients showed that the VEGF-A level
in plasma and serum collected from breast cancer patients is
linked with estrogen receptor status (23).
The local microenvironment of cancer is thought to be
crucial for cancer progression, because cancer epithelial
cells are surrounded by variable types of stromal cells (24).
From experimental cancer models, the extracellular
microenvironment has been demonstrated to influence
tumor formation, the rate of cellular proliferation, the
ability of the cancer cells to metastasize and the extent of
invasiveness. In cancers,
microenvironment are mediated, in part, by paracrine
signaling between epithelial cancer cells and the
surrounding stromal cells (25, 26). It has been suggested
that stromal cells may serve as a local reservoir for CLA
and, thus, may inhibit breast cancer cell progression (27).
Although VEGF-A is mainly expressed in malignant
epithelial cells, lesser amounts of VEGF-A can also be
detected in stromal cells and vascular endothelial cells,
which may imply both autocrine and paracrine signals of
this cytokine in the tumor microenvironment (19). Our
current study investigated whether CLA-modulated breast
stromal cells involved the microenvironment and whether
these effects are capable of suppressing the VEGF-A
angiogenesis biomarker in the hormone-responsive breast
cancer cell line, MCF-7, which may lead to suppression of
human breast neoplasms.
the influences of the
Materials and Methods
+98% purity t10,c12-CLA and c9,t11-CLA were purchased from
Matreya, Inc. (PA, USA) and the CLA stock solution was
prepared based on the paper published by Dr. Ip’s group (28). The
main difference in our CLA stock is that dextran-coated charcoal
(DCC, Dextran T-70; Pharmacia; activated charcoal; Sigma)-
treated fetal bovine serum (FBS, GibcoBRL, Bethesda, MD,
USA) was added to pure t10,c12-CLA and c9,t11-CLA to form a
CLA-serum protein complex.
Immortalized cell line. MCF-7 cells were purchased from the
American Type Culture Collection (ATCC, Manassas, VA, USA).
MCF-7 cells were planted in 75-cm2culture flasks in a humidified
incubator (5% CO2, 95% air, 37ÆC) and cultured in phenolred-free
high-calcium Dulbecco’s Modified Eagle’s Medium and Ham’s F12
Medium (DMEM/F12, 1.05 mM CaCl2) supplemented with 5%
FBS. The medium was renewed every two days.
Isolation of stromal cells from human breast tissues. Normal and
cancerous human breast tissues were obtained through the Tissue
Procurement Program at The Ohio State University Hospital in
Columbus, Ohio, USA. Tissues were placed in DMEM/F12 and
stored at 4ÆC. The isolation of stromal cells from human breast
ANTICANCER RESEARCH 25: 4061-4068 (2005)
4062

Page 3

tissues and culture condition have been described in detail
previously (29). Briefly, the tissues were minced and digested in
0.1% collagenase I (GibcoBRL) supplemented with 5% FBS and
antibiotic-antimycotic (100 unit/ml penicillin G sodium, 100 mg/ml
amphotericin B) (GibcoBRL) in a 37ÆC humidified incubator (5%
CO2: 95% air) overnight. The digested mixture was centrifuged at
200 xg for 5 min at 25ÆC. The cell pellet was re-suspended and
allowed to settle by gravity 3 times. Stromal cells in the supernatant
were then centrifuged at 200 xg for 5 min at 25ÆC and the pelleted
stromal cells were re-suspended in phenol red-free high-calcium
DMEM/F12 (1.05 mM CaCl2) supplemented with 5% FBS. Using
this method to separate primary cultured stromal cells and primary
cultured epithelial cells from human breast tissues has been shown
to produce a high purity of stromal cells, as described previously in
our laboratory (29). Although the primary cultured epithelial cell
type could also grow in the high-calcium medium, stromal cells
grew much faster than epithelial cells during the first week of
culture, thus leading to stromal cell dominance in our culture
conditions. In the current study, stromal cells were isolated and co-
cultured with MCF-7 cells.
Co-culture system and treatments. Treatments and total RNA and
protein extractions were performed on the primary cultured human
breast stromal cells not propagated beyond the third passage, and the
viabilities of the MCF-7 cells and breast stromal cells were greater
than 95%, as determined by the trypan blue dye exclusion method
(30). Co-cultures of MCF-7 cells and breast stromal cells were
performed by using flat-bottomed cell culture plates. Stromal cells
(0.5x106cells/well) were seeded on the nucleopore polycarbonate
membrane (0.4 Ìm pore size) of the cell culture inserts (upper
chamber). MCF-7 cells (1.0x106cells/well) were seeded on the
bottom plates (lower chamber). Because of the difference in the cell
sizes of MCF-7 (~5 Ìm) and stromal cells (~10 Ìm), the MCF-7 cell
number seeded on the bottom chamber was twice that of the stromal
cells. The seeded MCF-7 and stromal cells were not over 90%
confluence on the day of harvest. Since MCF-7 and stromal cells were
originally both cultured in phenolred-free high-calcium DMEM/F12
(1.05 mM CaCl2), in the co-culture of these two cell types, phenolred-
free high-calcium DMEM/F12 (1.05 mM CaCl2) supplemented with
5% DCC-treated FBS was used. After 2 days, the cells were treated
with 40 ÌM t10,c12-CLA, 40 ÌM c9,t11-CLA or vehicle as control in
the same medium for 3 days.
Reverse transcription-polymerase chain reaction (RT-PCR). At the
end of the treatment, breast stromal cells on cell culture inserts
were removed and discarded. Total RNA from MCF-7 cells on the
bottom plates was isolated in 1 ml TRIZOL® Reagent
(GibcoBRL), according to the manufacturer’s instructions. RT-
PCR was used to produce cDNA and performed in a gradient
mastercycler (Eppendorf®) as described in detail previously (29).
The PCR conditions for VEGF-A121and VEGF-A165were
optimized for MgCl2concentration, annealing temperature and
cycle number. The newly synthesized cDNAs were used as
templates for PCR after adjusting the reagent concentrations to
1.25 mM MgCl2, 0.24 ÌM primers, 2.5 Ìl 10X PCR Buffer
(GibcoBRL) and 1 U Platinum®
(GibcoBRL). The reactant was incubated at 95ÆC for 5 min. Then,
32 cycles of amplification were performed with each cycle
consisting of denaturation at 95ÆC for 1 min, annealing at 70ÆC for
30 sec and extension at 72ÆC for 1 min. For 36B4 (loading control),
Taq DNA polymerase
the PCR conditions were as described before (30, 31) and only
modified in terms of MgCl2concentration: 1.25 mM MgCl2.
The primer sequences for VEGF-A121and VEGF-A165were 5’-
ATC TTC AAG CCG TCC TGT GTG-3’ (sense) and 5’-TCA
CCG CCT CGG CTT GTC ACA-3’ (antisense). The primer
sequences for 36B4 were 5’-AAA CTG CTG CCT CAT ATC CG-
3’ (sense) and 5’-TTG ATG ATA GAA TGG GGT ACT GAT G-
3’ (antisense). The final PCR products (10 Ìl), mixed with 1 Ìl 10x
loading buffer, were separated on a 1.5% agarose gel containing
ethidium bromide. The lengths of the PCR products were 231 bp
for VEGF-A121, 363 bp for VEGF-A165,and 563 bp for 36B4. The
specific bands were quantified by ImageQuaNT software
(Molecular Dynamics, Sunnyvale, CA, USA). The results are
presented as the ratio of VEGF-A121or VEGF-A165to 36B4.
Western blot analysis. At the end of the treatment, breast stromal
cells on the cell culture inserts were removed and discarded. MCF-7
cells on the bottom plates were washed with ice-cold PBS and then
lysed with extraction reagent (Pierce, Rockford, IL, USA) and
protease inhibitor (Pierce) in ice. Cell lysates were separated by
centrifugation at 15,000 rpm in a cool room (4ÆC) for 30 min. An
equivalent amount of protein (50 Ìg) from each supernatant with
sample buffer (62.5 mM Tris-HCl, pH 6.8, 25% glycerol, 2% SDS,
0.004% bromophenol blue and 5% ‚-mercaptoethanol) was boiled
for 5 min and resolved in a 4-15% SDS-polyacrylamide ready gel
(Bio-Rad, Hercules, CA, USA). After electrophoreses, the proteins
were transferred to a PVDF (polyvinylidene difluoride) membrane
(Millipore, Billerica, MA, USA) by semi-dry transfer system (Bio-
Rad) at room temperature. The transblotted membrane was washed
twice with phosphate-buffered saline containing 0.1% Tween 20
(PBST). After blocking with PBST containing 10 % non-fat milk for
1 h at room temperature, the membrane for VEGF-A was
incubated with the diluted (1:500) VEGF-A mouse monoclonal
antibody (Santa Cruz, CA, USA, SC-7269) and the membrane for
‚-actin was incubated with the diluted (1:1000) ‚-actin goat
polyclonal antibody (Santa Cruz) in PBST, 5% non-fat milk at 4ÆC
overnight. After that, the membrane for VEGF-A was incubated
with diluted (1:3000) anti-mouse secondary antibody (Amersham,
Piscataway, NJ, USA) and the membrane for ‚-actin was incubated
with diluted (1:5000) donkey anti-goat secondary antibody
(Amersham) in PBST, 5% non-fat milk for 1 h at room
temperature. The immunoblots were enhanced by ECL Plusì
Western Blotting Detection reagent (Amersham, Buckinghamshire,
UK) and visualized by the Fuji image system (FUJIFILM Medical
Systems U.S.A., Inc., Stamford, CT, USA).
Statistics. The data, from 3 replicated wells as one cell group, was
presented as the mean±standard deviation (SD) and was analyzed
using StatView®(SAS Institute Inc. Cary, NC, USA) ANOVA
unpaired t-test. A p-value less than 0.05 was considered to be
statistically significant.
Results
c9,t11-CLA and t10,c12-CLA inhibited cell proliferation of
co-cultured MCF-7 cells. Based on the CLA concentration in
normal physiological human serum (10-70 ÌM) and in humans
who take CLA long-term supplementation (50-350 ÌM), the
CLA concentration applied in in vitro studies has ranged
Wang et al: Down-regulation of Angiogenesis Biomarker VEGF-A by CLA
4063

Page 4

from 12.5-250 ÌM, the mid-normal to supraphysiological-
pharmacological levels (31-32). CLA’s biological activities
on breast cancer cells regarding the ER status have
demonstrated that CLA (25-200 ÌM) appeared to inhibit
ER·-positive breast cancer cell growth but not ER·-
negative breast cancer cells (31). Our preliminary studies
showed that: i) the effective dose range in inhibiting human
breast cancer epithelial and stromal cell proliferation is
t10,c12-CLA at 10-80 ÌM and c9,t11-CLA at 40-160 ÌM for
3-day treatment; and ii) these experimental data also
indicated different biological effects of t10,c12-CLA and
c9,t11-CLA; finally iii) CLA was more effective on the
estrogen-responsive breast cancer cell line, MCF-7,
(unpublished data) than the estrogen-non-responsive breast
cancer cell line, MDA-MB-231. Therefore, we chose the
relatively lower but effective dose, 40 ÌM, of these two CLA
isomers to investigate the effects of breast stromal cells in
CLA-modulated VEGF-A expression in estrogen-responsive
MCF-7 cells.
The purity of the human breast stromal cells used for the
current study were examined by morphology and confirmed
by immunohistochemical staining previously reported from
our laboratory (29). We have been able to show that breast
stromal cells exhibited typical spindle-shaped morphology
and that the majority of breast stromal cells (>95%) were
immunopositive for the presence of vimentin, while no
expression of cytokeratin was detected, which confirmed the
fibroblastic nature of the breast stromal cells.
The effects of the CLA isomers, c9,t11-CLA and t10,
c12-CLA, on the proliferation of MCF-7 cells, which were
either cultured alone, or co-cultured with breast cancer
stromal cells (referred to hereafter as CASC), or co-
cultured with normal breast stromal cells (referred to
hereafter as NSC), were detected.
The proliferation rate of co-cultured MCF-7 cells was
significantly (p<0.05) higher in comparison with MCF-7
cells cultured alone (Figure 1). c9,t11-CLA and t10,c12-CLA
both decreased cell proliferation in cultured alone MCF-7
cells in a similar way. However, only t10,c12-CLA inhibited
cell proliferation in co-cultured MCF-7 cells. Our results
suggested that: i) NSC and CASC both contribute similar
effects in increasing MCF-7 cell proliferation; ii) t10,
c12-CLA is more potent in inhibiting MCF-7 growth.
c9,t11-CLA and t10,c12-CLA decreased VEGF-A121and
VEGF-A165mRNA expressions in co-cultured MCF-7 cells. In
human cells, differential exon splicing of VEGF-A formed
VEGF-A isomers which contain 121, 165, 189 and 206
amino acids, represented as VEGF-A121, VEGF-A165,
VEGF-A189and VEGF-A206, respectively. The vascular
permeability properties of VEGF-A121and VEGF-A165
angiogenesis cytokines have been suggested to be the initial
and crucial step in tumor angiogenesis (19, 33-34). To
compare the effects of t10,c12-CLA and c9,t11-CLA on
either VEGF-A121or VEGF-A165mRNA expressions in
MCF-7 cells, MCF-7 cells were cultured alone, or co-
cultured with CASC, or co-cultured with NSC. VEGF-A121
and VEGF-A165mRNA expression levels in MCF-7 cells
were determined by RT-PCR after the cells had been
treated by both CLA isomers, c9,t11-CLA and t10,c12-CLA,
for 3 days and the results are shown in Figures 2 and 3,
respectively.
The experimental data demonstrated that: i) VEGF-A121
mRNA expression of co-cultured MCF-7 cells was
significantly (p<0.05) higher in comparison to MCF-7 cells
cultured alone. Similarly, CASC and NSC contributed to up-
regulate VEGF-A165mRNA expression in MCF-7 cells; ii)
c9,t11-CLA and
t10,c12-CLA
VEGF-A121and VEGF-A165mRNA expressions in cultured
alone MCF-7 cells; iii) t10,c12-CLA decreased both VEGF-
A121and VEGF-A165mRNA expressions in MCF-7 cells
which were either co-cultured with CASC or with NSC; iv) in
both down-regulated
ANTICANCER RESEARCH 25: 4061-4068 (2005)
4064
Figure 1. CLA inhibited proliferation of co-cultured MCF-7 cells. MCF-
7, MCF-7 (+CASC), and MCF-7 (+SC) represent MCF-7 cells were
cultured alone, MCF-7 cells were co-cultured with breast cancer stromal
cells (CASC), and MCF-7 cells were co-cultured with normal breast
stromal cells (SC), respectively. Cells were treated by two CLA isoforms
and proliferation rate of MCF-7 cells was quantified by using CellTiterì
96 AQueous assay and optical density was read at 490nm (OD 490 nm)
in 96-well plates by an ELISA plate reader for each group separately. Bars
represent mean±SD, n=3. *p<0.05 stands for control group in MCF-7
(+CASC) and MCF-7 (+SC) versus control group in MCF-7 cells
cultured alone; +p<0.05 for control versus 40 ÌM c9,t11-CLA and 40
ÌM t10,c12-CLA.

Page 5

contrast, c9,t11-CLA-regulated VEGF-A121and VEGF-A165
mRNA expressions was more complicated because c9,
t11-CLA only decreased VEGF-A121mRNA expression of
MCF-7 cells in co-culture with NSC but not with CASC; v)
moreover, t10,c12-CLA appeared to be more active than
c9,t11-CLA in down-regulating VEGF-A121and VEGF-A165
mRNA expressions. Our results suggested that surrounded
stromal cells provide paracrine signals that play important
roles in increasing VEGF-A121and VEGF-A165 mRNA
expressions in malignant epithelial cells, which may lead to
tumor progression. CLA isomers, especially t10,c12-CLA, can
interrupt breast microenvironment paracrine signals which
may suppress characteristic of human breast neoplasms.
c9,t11-CLA and t10,c12-CLA decreased VEGF-A protein
expression in co-cultured MCF-7 cells. To compare the effects
of t10,c12-CLA and c9,t11-CLA on VEGF-A protein
expression in co-cultured MCF-7 cells, MCF-7 cells were
cultured alone, or co-cultured with CASC, or co-cultured
with NSC. The VEGF-A protein levels in MCF-7 cells were
determined by Western blot analysis after the cells had been
treated by both CLA isomers for 3 days and the results are
shown in Figure 4.
VEGF-A121, VEGF-A165and VEGF-A189are three
major VEGF-A isomers in human cells and, among them,
VEGF-A121and VEGF-A165are the predominant isomers
(19). Therefore, detected VEGF-A protein might represent
Wang et al: Down-regulation of Angiogenesis Biomarker VEGF-A by CLA
4065
Figure 2. CLA down-regulated VEGF-A121mRNA expression in co-
cultured MCF-7 cells. (A) Ethidium bromide-stained PCR products
separated in a 1.5% agarose gel. MCF-7, MCF-7 (+CASC) and MCF-7
(+SC) represent MCF-7 cells cultured alone, MCF-7 cells co-cultured
with breast cancer stromal cells (CASC), and MCF-7 cells co-cultured
with normal breast stromal cells (SC), respectively. The cells were treated
by two CLA isoforms and total RNA from MCF-7 cells was isolated from
each group separately. 36B4 was used as loading control. (B) The mRNA
ratio of VEGF-A121to 36B4 was measured by densitometry. Bars
represent mean±SD, n=3. *p<0.05 stands for control group in MCF-7
(+CASC) and MCF-7 (+SC) versus control group in MCF-7 cells
cultured alone; +p<0.05 for control versus 40 ÌM c9,t11-CLA and 40
ÌM t10,c12-CLA.
Figure 3. CLA down-regulated VEGF-A165mRNA expression in co-
cultured MCF-7 cells. (A) Ethidium bromide-stained PCR products
separated in a 1.5% agarose gel. MCF-7, MCF-7 (+CASC) and MCF-7
(+SC) represent MCF-7 cells cultured alone, MCF-7 cells co-cultured
with breast cancer stromal cells (CASC) and MCF-7 cells co-cultured with
normal breast stromal cells (SC), respectively. The cells were treated by
two CLA isoforms and total RNA from MCF-7 cells was isolated from
each group separately. 36B4 was used as loading control. (B) The mRNA
ratio of VEGF-A165to 36B4 was measured by densitometry. Bars
represent mean±SD, n=3. *p<0.05 stands for control group in MCF-7
(+CASC) and MCF-7 (+SC) versus control group in MCF-7 cells
cultured alone; +p<0.05 for control versus 40 ÌM c9,t11-CLA and 40
ÌM t10,c12-CLA.