| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
|
FJ
EXPRESS SUMMARY ARTICLE The Full-length version of this article is also available, published online April 27, 2001 as doi:10.1096/fj.00-0531fje. |
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Department of Experimental Oncology, Istituto Nazionale Tumori, 20133 Milano, Italy;
* Department of Medical Chemistry and Biochemistry, Medical School, University of Milan, Italy;
Kekule Institut fur Organishe, Chemie und Biochemie, 53121 Bonn, Germany; and
Children’s Hospital Research Foundation, Division and Program in Human Genetics, TCHRF 1042, Cincinnati, Ohio 45229-3039, USA
2Correspondence: Department of Experimental Oncology, Istituto Nazionale Tumori, Via G. Venezian 1, 20133 Milano, Italy. E-mail: delia{at}istitutotumori.mi
SPECIFIC AIMS
N-(4-hydroxyphenyl)retinamide (HPR) is a synthetic retinoid of clinical use in breast cancer chemoprevention. To understand its mechanism of action, we undertook cDNA differential display analysis to identify genes specifically regulated by HPR.
PRINCIPAL FINDINGS
1. HPR up-regulates prosaposin gene expression
The c-DNA differential display screening was performed on mRNA
extracted from T47D and MCF7 breast cancer cell lines treated or not
for 24 h with 3 µM HPR. One RT-PCR band whose intensity was
particularly elevated after retinoid exposure was excised from the gel
and tested on Northern blots, where it hybridized to a transcript of
1.6 kb. Most important, the intensity of the signal was 2.4- to
3.0-fold higher after HPR treatment. The DNA sequence analysis of this
RT-PCR band revealed homology with prosaposin, a gene coding for a
multifunctional glycoprotein, precursor of four saposins that are
localized partly within lysosomes, acting as a cofactors for
sphingolipids hydrolysis, and partly secreted.
2. Prosaposin protein up-regulation by HPR but not by ATRA
Treatment of MCF7 and T47D cells with HPR induced a significant
rise in intracellular prosaposin protein at 48 h (but not at
24 h) of 8.3- and 4-fold, respectively (Fig. 1
). By contrast, neither saposin A, C, or D were affected by the
retinoid. Together with the Northern results, these data demonstrate
that the transcriptional up-regulation of prosaposin by HPR occurs much
earlier than protein up-regulation.
|
The activity of all-trans retinoic acid (ATRA) and other retinoic acid derivatives is dependent on the binding to RAR and RXR family of nuclear transcription factors whereas that of HPR is mostly receptor independent. These differences between ATRA and HPR also included the regulation of prosaposin since, unlike HPR, ATRA did not affect its expression.
3. Prosaposin up-regulation by oxidative stress
We and others have previously reported that HPR behaves as a
pro-oxidant as it induces intracellular reactive oxygen species (ROS).
These free radicals have a key role in the apoptotic response to HPR.
To determine the role of ROS on prosaposin, we examined the effect of
noncytotoxic doses of the water-soluble oxidants
H2O2 and diethylmaleate
(DEM) (the latter functioning as a glutathione-depleting agent).
Treatment of T47D and MCF7 with these agents for 48 h induced a
significant rise in prosaposin levels (Fig. 1)
, more substantial with
H2O2 (3.5- and 4.2-fold
increase, respectively) than with DEM (1.9- and 1.7-fold increase,
respectively). On the other hand, the up-regulation of prosaposin by
HPR was markedly suppressed when cell cultures also contained the
antioxidant Vit-C (Fig. 1)
.
4. Ectopic expression of prosaposin antagonizes cell growth and
adhesion
The effect of deregulated prosaposin was analyzed on cells stably
transfected with an expression vector carrying the prosaposin cDNA
(T47D-PSAP) or with an empty plasmid (T47D-M). Many T47D-PSAP clones
(all confirmed for prosaposin overexpression) exhibited decreased
proliferation, impaired spreading and adhesion to culture plates.
Compared with T47D-M, the growth of T47D-PSAP3 was suppressed by >50%
at 96 h, as revealed by time- and serum-dependent thymidine uptake
experiments. Furthermore, the growth of T47D-PSAP3, but not that of
T47D-M or T47D, rose significantly in the presence of 15% serum,
indicating that the antiproliferative effect of ectopic prosaposin can
be partly overcome by increased serum concentrations.
The defective spreading and adhesion of T47D-PSAP clones, evident
even 72 h after seeding, was striking considering that parental
and mock-transfected T47D attached to culture plates within 4–6 h.
However, when T47D-PSAP cells were seeded in a medium containing 15%
of serum plus 50% spent medium from T47D cells, the adhesion occurred
much earlier (24–36 h after seeding), suggesting the possibility that
deregulated prosaposin causes an inhibition of integrin-mediated cell
anchorage. We thus performed immunofluorescence flow cytofluorometric
analyzed to determine the integrin receptors status in T47D-M and
T47D-PSAP3 cells. Whereas no expression of the 
1 and 3 subunits
was detected in these cells (Fig. 2
), the 6 and ß4 subunits, which form a laminin-specific
heterodimer, were by contrast well expressed in T47D-M [84% of
labeled cells: mean fluorescence intensity (MFI) 3.7÷3.9], but poorly
in T47D-PSAP3 (15%; MFI 1.7÷1.89). Likewise, the ß1 subunit, which
is involved in the establishment of skeletal metastases in advanced
breast cancer, was strongly expressed in T47D-M (99.8%, MFI: 12.7) but
weakly in T47D-PSAP3 (56%; MFI: 3.9) (Fig. 2)
. These findings thus
indicate an inverse relationship between the levels of prosaposin and
some integrin receptors involved in adhesion.
|
5. Prosaposin overexpression and HPR both increase ceramide levels
Evidence linking ceramide production to prosaposin or to HPR
responses prompted us to quantitate the intracellular ceramide in our
cells. The cells were pulsed with [1–3H]sphingosine and cultured for
48 h with or without HPR, after which lipids were extracted,
separated by HPTLC, and quantitated by digital autoradiography. The
amount of radioactivity incorporated into ceramide was 1.12 ±
0.19 (nCi/mg of protein±SD) in T47D and T47D mock cells,
2.00 ± 0.48 in T47D-PSAP3, 1.2 ± 0.06 in T47D-PSAP5,
1.98 ± 0.27, and 3.44 ± 0.01 in T47D treated with 3 µM
and 10 µM HPR, respectively. T47D-PSAP5 express lower levels of
ectopic prosaposin than T47D-PSAP3. Cells treated with 3 µM HPR
showed levels of ceramide comparable to T47D-PSAP3.
These results establish a direct relationship between prosaposin levels and ceramide biosynthesis. Ceramide levels also increase in response to HPR.
6. Down-regulation of integrin receptors and up-regulation of
prosaposin in ceramide-treated cells
The increased intracellular ceramide and decreased integrin
receptors in T47D-PSAP suggested a relationship between these events.
To verify this possibility, we analyzed the effects of exogenous
ceramide on integrin receptors. T47D cells were cultured for up to 5
days with noncytotoxic doses of synthetic C2-ceramide and then analyzed
by immunofluorescence flow cytometry. Compared with controls,
ceramide-treated cells evidenced a down-regulation of 6, ß4 and,
to a lesser extent, ß1 integrin receptors. Moreover, C2-ceramide
caused a five to sevenfold increase in prosaposin levels at both the
transcriptional and protein level. Noteworthy: the down-regulation of
integrin receptors was also observed after 5 days treatment of cells
with the long chain natural ceramide as well as with 1 µM HPR.
These findings suggest a feedback mechanism between ceramide and prosaposin that somehow affects integrin receptors regulation.
CONCLUSIONS AND SIGNIFICANCE
HPR is a synthetic retinoid with pharmacological
antiproliferative and apoptotic effects in vitro against several
malignant cell lines. We have performed a cDNA differential display
analysis to identify genes regulated by HPR, whose activity may help
elucidate the mechanism of action of this retinoid. This investigation
has led to identify prosaposin as a gene that is markedly up-regulated
in breast cancer cells by HPR. Human prosaposin is a glycoprotein of
70 kDa precursor of four saposins designated A, B, C, and D, each
composed of 80 amino acids that, in lysosomes, activate sphingolipid
hydrolases to degrade sphingomyelin into phosphorylcholine and
ceramide. A secreted form of prosaposin found in several body fluids
such as milk, cerebrospinal fluid, and seminal plasma is active in the
stimulation of neuronal outgrowth.
We and others have previously shown that HPR induces intracellular free radicals whose scavenging by antioxidants antagonizes apoptosis. Prompted by this evidence, we analyzed the effect of ROS on prosaposin. We have shown that Vit-C prevents the up-regulation of prosaposin induced by HPR and that treatment with H2O2 or DEM enhances the levels of prosaposin expression. These findings thus imply that prosaposin up-regulation represents a signaling response to oxidative stress, which in the case of HPR reflects its intrinsic pro-oxidant activity.
To better understand the role of prosaposin, we established stable transfectants of T47D cells ectopically expressing prosaposin. The proliferation, spreading, and adhesiveness of these clones were consistently impaired. Compared with parental or mock cells, the growth of T47D-PSAP was in fact reduced by >50% at 72 h after seeding, although it could be increased by higher serum concentration. These findings hence indicate that deregulated prosaposin has an antiproliferative effect (partly overcome by increasing the serum concentration), but whether this arises from prosaposin altering the threshold sensitivity of the growth factor responsive signal transduction pathway is unknown.
The impaired attachment of T47D-PSAP clones, a defect partly
rescued by growth in a conditioned medium, strongly suggested that
prosaposin overexpression interferes with the integrin receptor pathway
that mediates cell anchorage. In mammalian cells, the 24 known integrin
receptors are heterodimers formed from the combination of 17 and 8
ß subunits that, in addition to playing a role in adhesive
interactions, activate signaling pathways involved in cytoskeletal
organization, cell proliferation, and cell survival. We have found
greatly reduced levels of 6, ß1, and ß4 subunits in T47D-PSAP
relative to T47D-M or T47D parental cells, thus implicating prosaposin
in the regulation of integrins. To gain insight into the mechanism
underlying this event, we assessed ceramide in view of the fact that
prosaposin can activate ceramide biosynthesis through degradation of
sphingomyelin and that increased levels of ceramide are associated with
inhibition of integrin-mediated cell anchorage. We have shown increased
amounts of ceramide in prosaposin transfectants and a relationship
between prosaposin and ceramide levels. In addition, HPR at a 3 µM
dose increases ceramide levels similar to those observed in T47D-PSAP3.
It unclear how ceramide can affect the regulation of integrin receptors. We point out, however, that integrin receptors can undergo ‘inside-out’ regulation by the mitogen-activated protein kinase (MAPK) cascade and that ceramide can either modulate this pathway or interact with it by transmodulation of a proapoptotic signaling through JNK to MAPK. On this basis, it can be speculated that integrin receptors down-regulation may arise from an interaction of prosaposin-generated ceramide with the MAPK kinase pathway.
In conclusion, we have shown that HPR up-regulates the sphingolipid
activator protein prosaposin through oxidative stress. This response is
associated with increased ceramide production and defective attachment
paralleled by integrin
receptorsdown-regulation and growth inhibition, thus suggesting a direct link
between these events, as represented in the schematic diagram. We
cannot exclude that the cancer chemopreventive activity of HPR partly
arises from its effect, via prosaposin, on integrin receptors whose
down-regulation is associated with reduced metastatic capacity of tumor
cells.FIGURE 3
|
FOOTNOTES
1 To read the full text of this article, go to http://www.fasebj.org/cgi/doi/10.1096/fj.00-0531fje ; to cite this
article, use FASEB J. (April 27, 2001) 10.1096/fj.00-0531fje 
This article has been cited by other articles:
|
|
 |
|
  Y. H. Zeidan, R. W. Jenkins, and Y. A. Hannun Remodeling of cellular cytoskeleton by the acid sphingomyelinase/ceramide pathway J. Cell Biol., April 21, 2008; 181(2): 335 - 350. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 W. Hu, R. Xu, G. Zhang, J. Jin, Z. M. Szulc, J. Bielawski, Y. A. Hannun, L. M. Obeid, and C. Mao Golgi Fragmentation Is Associated with Ceramide-induced Cellular Effects Mol. Biol. Cell, March 1, 2005; 16(3): 1555 - 1567. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Pirkmaier, K. Yuen, J. Hendley, M. J. O'Connell, and D. Germain Cyclin D1 Overexpression Sensitizes Breast Cancer Cells to Fenretinide Clin. Cancer Res., May 1, 2003; 9(5): 1877 - 1884. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Prinetti, L. Basso, V. Appierto, M. G. Villani, M. Valsecchi, N. Loberto, S. Prioni, V. Chigorno, E. Cavadini, F. Formelli, et al. Altered Sphingolipid Metabolism in N-(4-Hydroxyphenyl)- retinamide-resistant A2780 Human Ovarian Carcinoma Cells J. Biol. Chem., February 14, 2003; 278(8): 5574 - 5583. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
