High cell density induces expression from the carbonic anhydrase 9 promoter.

Full text

Page 1

228 BioTechniques Vol. 36, No. 2 (2004)
BENCHMARKS
1279-1288.
9.Ellis, T.H., S.J. Poyser, M.R. Knox, A.V.
Vershinin, and M.J. Ambrose. 1998. Poly-
morphism of insertion sites of Ty1-copia class
retrotransposons and its use for linkage and
diversity analysis in pea. Mol. Gen. Genet.
260:9-19.
10.Melayah, D., E. Bonnivard, B. Chalhoub,
C. Audeon, and M.-A. Grandbastien. 2001.
The mobility of the tobacco Tnt1 retrotrans-
poson correlates with its transcriptional acti-
vation by fungal factors. Plant J. 28:159-168.
11.Arnold, C. and I.J. Hodgson. 1991. Vector-
ette PCR: a novel approach to genomic walk-
ing. PCR Methods Appl. 1:39-42.
12.Kidwell, M.G. 2002. Transposable elements
and the evolution of genome size in eukary-
otes. Genetica 115:49-63.
13.Schuler, G.D. 1997. Sequence mapping by
electronic PCR. Genome Res. 7:541-550.
14.Rombauts, S.,Y. Van De Peer, and P. Rouzé.
2003. AFLPinSilico, simulating AFLP finger-
prints. Bioinformatics 19:776-777.
15.Miura A., S. Yonebayashi, K. Watanabe,
T. Toyama, H. Shimada, and T. Kakutani.
2001. Mobilization of transposons by a mu-
tation abolishing full DNA methylation in
Arabidopsis. Nature 411:212-214.
16.Altschul, S.F., T.L. Madden, A.A. Schaffer,
J. Zhang, Z. Zhang, W. Miller, and D.J.
Lipman. 1997. Gapped BLAST and PSI-
BLAST: a new generation of protein database
search programs. Nucleic Acids Res. 25:
3389-3402.
17.Breslauer K.J., R. Frank, H. Blocker, and
L.A. Marky. 1986. Predicting DNA duplex
stability from the base sequence. Proc. Natl.
Acad. Sci. USA 83:3746-3750.
Received 22 August 2003; accepted 19
November 2003.
Address correspondence to Quang Hien Le,
Laboratoire de Biologie Cellulaire, INRA
Centre de Versailles, 78026 Versailles,
France. e-mail: hien.le@versailles.inra.fr
Efficient ways to control the level
and timing of the expression of spe-
cific genes in cultured cells and tis-
sues, including tumors, have received
considerable attention (1,2). Inducible
expression systems should have a
minimal activity in the basal state but
should allow rapid accumulation of the
heterologous protein upon stimulation.
The most advanced inducible systems
employ combinations of functional
domains from prokaryotic, eukary-
otic, or viral proteins to create chimeric
transactivators capable of modulating
gene expression in a drug-dependent
manner (1). The second component in
these systems is an inducible promoter,
which consists of a multimerized
transactivator binding sequence linked
upstream of a minimal promoter. In
the presence of inducer, the chimeric
activator binds specifically to its DNA
recognition sequence and activates the
transcription of the target gene (1).
Although very specific and effective,
these chimeric systems usually require
specialized cell lines or have to be
prepared in several relatively time-
consuming steps. Therefore, it may be
advantageous to use systems that are
easier to generate and yet retain signifi-
cant inducibility. Here we describe the
cell density-dependent activity of the
carbonic anhydrase 9 (CA9) promoter
and propose its utility as an inducible
expression system.
The expression of carbonic anhy-
drase IX (CAIX, previously known
as MN) has been detected in a large
number of carcinomas and carci-
noma-derived cell lines but not in the
corresponding normal tissues (Refer-
ences 3, 4, and references therein). The
mechanism of CAIX induction in dense
cultures was the subject of our previous
study (5). Earlier, oxygen levels in
sparse (106 cells in 100-mm plates)
and dense (106 cells in 34.8-mm plates)
LNCaP human prostate cells were es-
tablished as 13% (96 mmHg) and 9%
(70 mmHg), respectively (6). Reoxy-
genation by stirring abrogated CAIX
expression, suggesting that CAIX
expression in cultured cells is indeed
triggered by an intermediate decrease
of O2 tension due to increased O2
consumption and not by cell contacts
per se. This decreased O2 tension, also
termed pericellular hypoxia (6), is too
high for an appreciable stabilization of
hypoxia-inducible factor 1α (HIF-1α),
but it is sufficient for the activation of a
phosphatidylinositol 3′-kinase (PI3-K)-
dependent pathway (5). Earlier studies
defined the CA9 promoter in the (-173;
+31) region (the numbers in parenthe-
ses indicate each position relative to the
transcription start), which appears to
contain the critical regulatory elements
for CA9 transcriptional activation (7).
Among these, the hypoxia-response el-
ement (HRE) (8) and SP1/SP3 binding
protected region 1 (PR1) (9) are crucial
for CA9 transcriptional activity.
The striking effect of cell density on
CAIX expression prompted us to inves-
tigate the utility of the CA9 promoter
as a cell density-inducible expression
system. The (-173; +31) CA9 promoter
fragment was cloned in the pGL2-Ba-
sic (Promega, Madison, WI, USA) and
pEGFP-1 (BD Biosciences Clontech,
Palo Alto, CA, USA) vectors. The
(-2361; +298) vascular endothelial
growth factor (VEGF) gene fragment
was also cloned in the pGL2-Basic
vector. The simian virus 40 (SV40)
early promoter-driven pGl2 control
vector was obtained from Promega.
To prevent the possible modulation of
CA9 promoter activity by the SV40
promoter/enhancer sequence present
in pEGFP-1 (GenBank® accession
no. U55761; positions 1694–1925),
High cell density induces expression from the
carbonic anhydrase 9 promoter
Milota Kaluzová1,2, Stefan Kaluz1,2, and Eric J. Stanbridge1
1University of California at Irvine, Irvine, CA, USA and 2Slovak Academy of Sciences,
Bratislava, Slovak Republic
BioTechniques 36:228-234 (February 2004)

Page 2

230 BioTechniques Vol. 36, No. 2 (2004)
BENCHMARKS
this sequence was deleted. The human
cervical carcinoma HeLa and osteosar-
coma Saos-2 cell lines were grown in
Dulbecco’s modified Eagle’s medium
(DMEM) (Cambrex, Baltimore, MD,
USA), supplemented with 10% fetal
calf serum (FCS) (Invitrogen, Carls-
bad, CA, USA), 1.102 U/mL penicillin
(Sigma, St. Louis, MO, USA), 1.102
µg/mL streptomycin (ICN Biomedi-
cals, Costa Mesa, CA, USA), and 125
ng/mL amphotericin B (Sigma). The
cells were transiently transfected with
a pGL2 construct (expressing firefly
luciferase) and pRL-CMV (expressing
Renilla luciferase) (Promega) using the
Effectene® kit (Qiagen, Valencia, CA,
USA), according to the manufacturer’s
instructions. Following an 8-h exposure
to the transfection mixture, the cells
were rinsed with phosphate-buffered
saline (PBS), trypsinized, replated at
varying densities, and harvested at the
indicated times. The firefly and Renilla
luciferase activities were assayed with
the Dual-Luciferase® Reporter Assay
System (Promega) in the Monolight™
2010 luminometer (BD Biosciences,
San Jose, CA, USA). Promoter activity
was expressed as the average of ratios
of firefly to Renilla luciferase activities
from three independent experiments.
For the green fluorescent protein
(GFP) experiments, HeLa cells were
cotransfected with the pEGFP-1 con-
struct containing the CA9 fragment and
pCEP4 (conferring resistance to hygro-
mycin) (Invitrogen) at a 10:1 ratio. The
stably transfected cells were selected in
the presence of 400 U hygromycin/mL
media for 3 weeks. The mass popula-
tion of transfected cells was plated at
20,000 and 160,000/cm2 and tested for
fluorescence 24 h later.
Initially, we investigated the time
course of the cell density-dependent in-
duction of the CA9 promoter activity in
transiently transfected HeLa cells. The
transfected cells were replated at vary-
ing densities and harvested after 24,
48, and 72 h incubation. After 24 h, a
basal activity of the CA9 promoter was
observed in all tested densities, except
160,000/cm2, where a 3-fold induction
was observed (Figure 1A). Forty-eight
hours after replating, a moderate induc-
tion was observed in cells plated at
40,000 and 80,000/cm2 (2.5- and 3.3-
fold, respectively), whereas a 9-fold in-
duction was observed in cells plated at
160,000/cm2 (Figure 1A). The most ef-
ficient induction was observed after 72
h. The activity of the CA9 promoter in
cells plated at 40,000 and 80,000/cm2
increased 4- and 6-fold, respectively,
and a more than 15-fold induction was
observed in the cells plated at 160,000/
cm2 (Figure 1A). These observations
confirm the strong dependence of CA9
promoter activity on the density of
replated cells (transfection efficiency
Figure 1. Cell density-dependent activity of the carbonic anhydrase 9 (CA9) promoter. (A) Lucif-
erase activity in transiently transfected HeLa cells. Hela cells were cotransfected with the (-173; +31)
CA9 promoter fragment in pGL2-Basic and pRL-CMV. Following an 8-h exposure to the transfection
mixture, the cells were trypsinized and plated. The cells were then harvested, and the reporter assay was
performed. The activity of the CA9 promoter is expressed as the ratio of firefly to Renilla activity for
each density and time point. Each of the points represents the mean value (×– ± ) from at least three
individual experiments. (B) Luciferase activity generated from the CA9, vascular endothelial growth fac-
tor (VEGF), and simian virus 40 (SV40) promoters in transiently transfected Saos-2 cells. Saos-2 cells
were transfected as in panel A. The cells were harvested 24 h after replating at 20,000 and 160,000/cm2.
The activity of each promoter is expressed as the fold increase of the ratio of firefly to Renilla activity
in dense (160,000/cm2) over sparse (20,000/cm2) cultures. (C) Green fluorescent protein (GFP) activ-
ity in stably transfected HeLa cells. HeLa cells were cotransfected with the (-173; +31) CA9 promoter
fragment in pEGFP-1 and pCEP4 at a 10:1 ratio. The mass population of the stably transfected cells was
plated at 20,000 and 160,000/cm2 and tested for fluorescence 24 h later. A representative photograph is
shown for each cell density (magnification: 100×). AU, arbitrary unit.

Page 3

BENCHMARKS
was the same for all cells because the
same transfection mixture was used).
Basal activity (slightly above the
background) in sparsely plated cells
(20,000/cm2) indicates the tight control
of the CA9 promoter activity. Similar
results were obtained in a number of
other cell lines, such as Saos-2, HT
1080 (fibrosarcoma), MCF-7 (breast
carcinoma), M006 (glioblastoma), and
SiHa and CaSki (cervical carcinoma)
(data not shown).
VEGF expression is also regulated
by cell density (10,11), although the
regulatory elements involved have not
been identified. Therefore, we wanted
to relate the cell density-dependent in-
duction of the CA9 and VEGF promot-
ers. Transiently transfected Saos-2 cells
were harvested 24 h after replating
at 20,000 and 160,000/cm2. Reporter
activities indicated about a 12-fold in-
crease of the CA9 promoter activity in
dense cultures, compared to about a 3-
fold increase of VEGF promoter activ-
ity (Figure 1B). Increasing cell density
had no effect on the SV40 promoter
activity (Figure 1B). Thus, a direct
comparison of the CA9 and VEGF pro-
moters revealed a substantially higher
induction of the CA9 promoter under
conditions of increasing cell density.
To further probe the efficiency of
the CA9 promoter in controlling cell
density-dependent expression, HeLa
cells were stably transfected with the
GFP vector. In agreement with the
results of transient transfections, the
activity of the GFP was also dependent
on the initial density of the plated cells.
No fluorescence was observed in cells
plated at 20,000/cm2, whereas signifi-
cant fluorescence was observed in cells
plated at 160,000/cm2 (Figure 1C). The
fluorescence observed was extensive,
but focal areas of low expression were
noticed. This heterogeneous fluorescent
pattern is not the consequence of using
a mass population of stably transfected
cells because similar results were ob-
served with several individual clones
(data not shown). In our opinion, this
irregular expression pattern reflects
the heterogeneity of conditions (with
respect to decreased O2 tension) prevail-
ing in dense cultures. Although it would
seem that the cells in culture should be
exposed to the same conditions, in fact,
the cells throughout the high cell density
culture experience varying microenvi-
ronmental conditions, and we believe
that the CA9 promoter functions as a
sensor of the heterogeneity of O2 levels
in this environment.
Transformed cells, unlike their
normal counterparts, do not display
density-dependent inhibition of cell
division. As a result, they do not stop
dividing after a confluent monolayer
is formed and instead produce increas-
ingly denser cultures by multilayering.
This increasing density is associated
with alterations in the expression pat-
terns of multiple genes. Although sev-
eral mechanisms responsible for the

Page 4

234 BioTechniques Vol. 36, No. 2 (2004)
BENCHMARKS
cell density-dependent modulation of
gene expression have been proposed
[e.g., involvement of soluble factor(s),
changes in cell shape, and direct cell-
cell contacts or their combination] (10),
in the case of the CA9 promoter, the
primary cause is the decreased O2 ten-
sion in dense cultures due to increased
O2 consumption (5).
There has recently been consider-
able progress in dissecting the CA9
promoter function (5,7–9). The coop-
eration between transcription factors
binding HRE and PR1 is critical for
the inducibility of the CA9 promoter
by cell density (5), while the remain-
ing positive regulatory elements am-
plify the signal. In the absence of an
inducing signal, the activity of the CA9
promoter is repressed by the silencer
element PR4, which leads to the tight
control of basal expression (7).
Using two different reporters, we
were able to demonstrate the tight
regulation of the CA9 promoter by cell
density and associated lower O2 ten-
sion in several cell lines. Apparently,
the regulation of the CA9 promoter by
cell density in transformed cell lines is
a general phenomenon, and we there-
fore hypothesize that the CA9 promoter
will display cell density-dependent
activity in most cells that are capable
of reaching a certain density that will
generate mild hypoxia. The only re-
quirement is that the cells must be HIF-
1α, SP1/SP3, and PI3-K competent (5).
Indeed, we have shown that many dif-
ferent types of malignant cells behave
in this fashion.
Induction by cell density does not
involve treatment with an exogenous
agent and no special conditions are re-
quired, except for replating cells trans-
fected with the expression construct
under the control of the CA9 promoter
at high density. The proposed cell den-
sity-inducible system utilizing the CA9
promoter thus represents an easy and
readily applicable alternative to other
inducible expression systems.
The unique organization of the CA9
promoter (HRE immediately upstream
of transcription start) results in tight
control under normoxic conditions
(21% O2) and efficient induction under
conditions of mild to strong hypoxia
(5%–0.5% O2). The CA9 promoter is
therefore an excellent candidate for var-
ious O2-sensing applications, including
the monitoring of hypoxia in vivo or in
vitro or the expression of suicide genes
in hypoxic tumors. It is possible that
combinations of multiple HRE and
PR1 regulatory elements in front of the
CA9 or a heterologous promoter will be
more efficient for directing cell densi-
ty-dependent expression and will yield
more homogenous expression than the
wild-type CA9 promoter.
ACKNOWLEDGMENTS
This study was supported by a grant
from the California Cancer Research
Program (00-00789V-20240) and The
Avon Foundation. The (-2361; +298)
VEGF gene fragment was kindly pro-
vided by Dr. Chris Hughes (Depart-
ment of Biochemistry and Molecular
Biology, University of California,
Irvine, CA).
REFERENCES
1.Harvey, D.M. and C.T. Caskey. 1998. Induc-
ible control of gene expression: prospects
for gene therapy. Curr. Opin. Chem. Biol. 2:
512-518.
2.Rossi, F.M. and H.M. Blau. 1998. Recent ad-
vances in inducible gene expression systems.
Curr. Opin. Biotechnol. 9:451-456.
3.Závada, J., Z. Závadová, S. Pastoreková, F.
Ciampor, J. Pastorek, and V. Zelník. 1993.
Expression of MaTu-MN protein in human
tumor cultures and in clinical specimens. Int.
J. Cancer 54:268-274.
4.Ivanov, S., S.Y. Liao, A. Ivanova, A.
Danilkovitch-Miagkova, N. Tarasova, G.
Weirich, M.J. Merril, M.A. Proescholdt,
et al. 2001. Expression of hypoxia-inducible
cell-surface transmembrane carbonic anhy-
drases in human cancer. Am. J. Pathol. 158:
905-919.
5.Kaluz, S., M. Kaluzová, A. Chrastina, P.L.
Olive, S. Pastoreková, J. Pastorek, M.I. Le-
rman, and E.J. Stanbridge. 2002. Lowered
oxygen tension induces expression of the
hypoxia marker MN/carbonic anhydrase IX
in the absence of hypoxia-inducible factor 1α
stabilization: a role for phosphatidylinositol
3′-kinase. Cancer Res. 62:4469-4477.
6.Sheta, E.A., H. Trout, J.J. Gildea, M.A.
Harding, and D. Theodorescu. 2001. Cell
density mediated pericellular hypoxia leads
to induction of HIF-1α via nitric oxide and
Ras/Map kinase mediated signaling pathways.
Oncogene 20:7624-7634.
7.Kaluz, S., M. Kaluzová, R. Opavský, S. Pas-
toreková, A. Gibadulinová, F. Dequiedt, R.
Kettmann, and J. Pastorek. 1999. Transcrip-
tional regulation of the MN/CA 9 gene coding
for the tumor-associated carbonic anhydrase
IX. Identification and characterization of a
proximal silencer element. J. Biol. Chem.
274:32588-32595.
8.Wykoff, C.C., N.J. Beasley, P.H. Watson,
K.J. Turner, J. Pastorek, A. Sibtain, G.D.
Wilson, H. Turley, et al. 2000. Hypoxia-in-
ducible expression of tumor-associated car-
bonic anhydrases. Cancer Res. 60:7075-7083.
9.Kaluzová, M., S. Pastoreková, E. Svastová,
J. Pastorek, E.J. Stanbridge, and S. Kaluz.
2001. Characterization of the MN/CA 9 pro-
moter proximal region-a role for SP and AP1
factors. Biochem. J. 359:669-677.
10.Mukhopadhyay, D., L. Tsiokas, and V.P.
Sukhatme. 1998. High cell density induces
vascular endothelial growth factor expression
via protein tyrosine phosphorylation. Gene
Expr. 7:53-60.
11.Koura, A.N., W. Liu, Y. Kitadai, R.K.
Singh, R. Radinsky, and L.M. Ellis. 1996.
Regulation of vascular endothelial growth fac-
tor expression in human colon carcinoma cells
by cell density. Cancer Res. 56:3891-3894.
Received 28 October 2003; accepted
18 November 2003.
Address correspondence to Stefan Kaluz,
Department of Microbiology and Mo-
lecular Genetics, Medical Science I B210,
University of California at Irvine, College
of Medicine, Irvine, CA 92697-4025, USA.
e-mail: skaluz@uci.edu