| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Regular Articles |
1 Inhibitory Receptors by Epidermal Growth Factor Receptor Signaling
Departments of 1 Neurosurgery and 2 Pathology and Laboratory Medicine, University of Pennsylvania School of Medicine, Philadelphia, Pennsylvania
 |
ABSTRACT |
|---|
 Top ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
1 is a membrane glycoprotein and a member of the SIRP receptor family. These transmembrane receptors have been shown to exert negative effects on signal transduction by receptor tyrosine kinases via immunoreceptor tyrosine-based inhibitory motifs in the carboxyl domain. Previous work has demonstrated that SIRPs negatively regulate many signaling pathways leading to reduction in tumor migration, survival, and cell transformation. Thus, modulation of SIRP expression levels or activity could be of great significance in the field of cancer therapy. The aim of the present study was to determine the factors that regulate levels of SIRP1 in human glioblastoma cells that frequently overexpress the epidermal growth factor receptor (EGFR) because SIRPs have been shown to negatively regulate EGFR signaling. Northern blot analysis and immunoprecipitation assays showed variable expression levels of endogenous SIRP transcripts in nine well-characterized glioblastoma cell lines. We examined SIRP1 regulation in U87MG and U373MG cells in comparison with clonal derivatives that express a truncated form of erbB2, which negatively regulates EGFR signaling by inducing the formation of nonfunctional heterodimeric complexes. Mutant erbB2-expressing cells contained more SIRP1 mRNA when compared with the parental cells in presence or absence of serum. Similarly, immunoprecipitation assays showed increased SIRP1 protein levels in erbB-inhibited cells when compared with parental cells. Messenger RNA stability assays revealed that the increased mRNA levels in EGFR-inhibited cells were due to an induction of transcription. Consistent with this finding, expression of the erbB2 mutant receptor up-regulated SIRP1 promoter activity in all cell lines tested. Interestingly, pharmacological inhibition of the kinase activities of EGFR, erbB2, and src and activation of mitogen-activated protein kinase, but not phosphatidylinositol 3'-kinase, significantly up-regulated SIRP1 promoter activity. Based on these observations, we hypothesize that down-modulation of EGFR signaling leads to transcriptional up-regulation of the inhibitory SIRP1 gene. These data may be important in the application of erbB-inhibitory strategies and for design of therapies for the treatment of glial tumors and other epithelial malignancies. |
INTRODUCTION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
and ß, which differ by the presence (SIRP) or absence (SIRPß) of an intracytoplasmic domain (2)
. The SIRP subtype has an apparent molecular weight of between 85,000 and 95,000 in humans and rats and 100,000 to 120,000 in the mouse and is heavily glycosylated.
Several groups have discovered various SIRP-like molecules, which are designated as SHPS-1, p84, BIT, MFR, MyD-1, and SIRP1 (1
, 2
, 4, 5, 6, 7)
. The intracytoplasmic domain of SIRP proteins contains a four tyrosine-based regulatory motif called the immunoreceptor tyrosine-based inhibitory motif, which, on phosphorylation, recruits the SH2 domain-containing protein tyrosine phosphatases SHP-1 and SHP-2. This physical association has been shown to induce negative regulatory effects on cell proliferation induced by growth factor stimulation of receptor tyrosine kinases and by viral oncogene products (1, 2, 3
, 8)
. On the other hand, SIRPß contains no intracytoplasmic domain and associates with other adapter molecules to modulate signaling (9)
. Two such adapter molecules, DAP-12 (also known as KARAP; refs. 10, 11, 12
) and DAP-10 (also called KAP-10; refs. 13
and 14
), have been shown to interact with SIRPß. In contrast to SIRP, SIRPß molecules possess immunoreceptor tyrosine-based activating motifs, which confer a positive regulatory effect (10
, 15
, 16)
.
SIRP1 proteins are differentially expressed in a variety of tissues but are abundant in myeloid cells, neurons, and brain (2
, 17, 18, 19, 20)
. Various studies have linked SIRP1 receptors to different biological processes, most notably cell growth and motility. SIRP1 proteins were reported to interact with CD47 on cerebral epithelium to facilitate monocyte migration across the brain, which is considered to be a critical event in various neuroinflammatory disorders, such as multiple sclerosis or human immunodeficiency virus-associated dementia (21)
. A study using mice expressing a mutant SHPS-1 [SHPS-1-cyto(–/–)] lacking most of the cytoplasmic region indicated that SHPS-1 (SIRP1) contributes to the survival of circulating platelets by down-regulating the macrophage phagocytic response (22)
. Overexpression studies of SHPS-1 in fibroblasts have shown that SHPS-1 plays a crucial role in integrin-mediated cytoskeletal reorganization, cell motility, and adhesion-induced activation of Rho and in the negative regulation of growth factor-induced activation of mitogen-activated protein kinases [MAPKs (23)
]. SIRP1/SHPS-1 receptors have also been shown to positively or negatively regulate MAPK pathways after stimulation of many different receptors, including epidermal growth factor receptor (EGFR) and the insulin receptor. A previous study has shown that overexpression of SHPS-1 resulted in increased SHPS-1/SHP-2 complex formation, which potentiated the Ras/Raf/MAPK pathway in response to insulin (24)
. On the contrary, SIRP1 overexpression in NIH3T3 cells inhibited DNA synthesis and MAPK phosphorylation after epidermal growth factor (EGF) or insulin stimulation (2)
. Previously, we reported that SIRP1 associates with SHP-2 to negatively regulate EGFR-mediated phosphatidylinositol 3'-kinase (PI3K) signaling and resulted in reduced transformation, reduced cell migration, and cell spreading and enhanced apoptosis after DNA damage in human glioblastoma cells (25)
. Recently, we demonstrated that an association between the SHP-2 phosphatase and Gab1 adapter protein is critical for EGFR-mediated positive signaling (26)
and that SHP-2 plays a positive role in regulating the response of astrocytes and fibroblasts to growth factors (27)
. Therefore, modulation of SIRP1 activity and expression levels may play an important role in regulating EGFR-mediated Gab1/SHP-2 association and cell signaling. However, mechanisms that regulate SIRP1 expression and function have not been well characterized.
In this study, we have evaluated the transcriptional regulation of SIRP1 in human cancer cells to understand SIRP expression and function. We have chosen human glioblastoma cells due to the central role played by EGFR in glial tumorigenesis (28, 29, 30, 31)
, the coupling between EGFR and SIRP, and the high levels of endogenous SIRP expression in the mammalian brain. We used nine well-characterized human glioblastoma cell lines to determine factors responsible for transcriptional regulation of SIRP1 levels. Northern blot and immunoprecipitation analyses in U87MG and U373MG cells in comparison with erbB-inhibited clonal derivative U87MG/T691 and U373MG/T691 cells showed that phenotypic inhibition and serum starvation resulted in increased SIRP1 expression. These results suggest that down-regulated EGFR signaling cooperated with serum starvation to up-regulate SIRP1 expression. To further evaluate mechanisms regulating inhibitory SIRP1 receptors in glioblastoma cells, we used a luciferase reporter construct carrying an isolated 
2.0-kb fragment of the 5'-untranslated region (UTR) from the SIRP1 gene for promoter studies. A luciferase reporter plasmid containing this SIRP1 promoter region showed increased luciferase activity when cotransfected with a plasmid carrying a truncated erbB2 receptor, which interferes with EGFR signaling in different cell lines. Furthermore, pharmacological inhibition of tyrosine kinase activity of EGFR and erbB2 increased SIRP1 promoter activity. Interestingly, pharmacological inhibition of p42/44 MAPK and src kinase showed statistically significant up-regulation of promoter activity as compared with control cells. Transcriptional up-regulation of SIRP1 expression is therefore coupled to down-modulation of EGFR signaling. Collectively, our data suggest that mechanisms up-regulating inhibitory SIRP1 receptor expression and/or function could be used as an adjunct to existing therapies for erbB-driven tumors including glioblastomas and other epithelial malignancies.
|
MATERIALS AND METHODS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
1 antibody was kindly provided by Dr. Gibbes R. Johnson (Food and Drug Administration, Bethesda, MD). Anti-ß-actin monoclonal antibody (clone AC-15) was purchased from Sigma (St. Louis, MO). TRIzol and all tissue culture supplies were obtained from Invitrogen (Carlsbad, CA). Diethyl pyrocarbonate and protein A-Sepharose CL-4B were purchased from Sigma. Magnacharge nylon membrane was purchased from Osmonics Inc. (Westborough, MA). [-32P]dCTP (300 Ci/mmol) was obtained from NEN Life Science Products, Inc. (Boston, MA), and the enhanced chemiluminescence detection kit was obtained from Amersham Pharmacia Biotech (Buckinghamshire, United Kingdom). A light chemiluminescence reporter gene assay system for the detection of luciferase activity was purchased from Promega (Madison, WI).
Cell Lines and Culture Conditions.
Human glioblastoma cell lines U87MG, U373MG, U343, U251, U118, LN18, LN229, SF767, and T98G were maintained in Dulbecco’s modified Eagle’s medium (Cellgro, Herndon, VA), with 10% fetal bovine serum (Hyclone, Ogdon, UT) at 37°C in 95% air/5% CO2. U87MG/T691 and U373MG/T691, clonal derivatives of U87MG and U373MG, respectively, were supplemented with 0.4 mg/mL G418 (Invitrogen) to maintain stable expression of truncated ErbB2/Neu receptor (T691stop) with large cytoplasmic deletion, which include the tyrosine kinase domain of p185/neu (31)
. Another derivative, U87MG/SIRP1, which stably expresses the SIRP1 protein, was supplemented with 40 ng/mL hygromycin (Roche Molecular Biochemicals, Indianapolis, IN; ref. 25
).
Messenger RNA Stability Assay.
To measure the half-life of the SIRP1 message, we incubated U373MG and U373MG/T691 cells with 10 µg/mL actinomycin D, and samples were harvested for total RNA at the designated intervals thereafter.
Northern Blot Analysis.
Total RNA was isolated using TRIzol, according to the manufacturer’s directions. Briefly, 10 to 20 µg of total cellular RNA were fractionated on 1% formaldehyde agarose gel, transferred to Magnacharge nylon membrane by capillary blotting, and fixed by baking at 80°C under vacuum. Labeling of radioactive probes was performed using [-32P]dCTP and a Prime-It kit (Stratagene, La Jolla, CA) according to the manufacturer’s instructions. Hybridization was carried out at 65°C, after which membranes were washed to a stringency of 1x SSC, 0.1% SDS at 65°C. Autoradiography was carried out at –80°C. A 2.2-kb SIRP1 cDNA fragment excised with NheI and PstI from pIRES hygro2 plasmid was used to make radioactive probe for hybridization. To verify equal loading, all gels were stained with ethidium bromide. Furthermore, SIRP1 mRNA levels were normalized with glyceraldehyde 3-phosphate dehydrogenase (GAPDH) mRNA levels, which remained relatively constant under different treatment conditions.
Cell Lysis, Immunoprecipitation, and Immunoblotting.
Cells were scraped off in lysis buffer containing 50 mmol/L Tris (pH 7.5), 150 mmol/L NaCl, 2 mmol/L EGTA, 0.1% Triton X-100, 1 mmol/L phenylmethylsulfonyl fluoride, 100 µg/mL aprotinin and leupeptin, and 1 mmol/L sodium orthovanadate. After a 30-minute incubation on ice, soluble fractions were collected, and protein concentrations were determined using a Dc Protein Assay Kit (Bio-Rad, Hercules, CA). Approximately 400–500 µg of soluble fraction were incubated with anti-SIRP1-antibody (2 µg) for 2 hours at 4°C. Immune complexes were collected with protein A-Sepharose CL-4B for 1 hour; washed three times with wash buffer containing 50 mmol/L Tris (pH 7.5), 133 mmol/L NaCl, 2 mmol/L EGTA, and 0.1% Triton X-100; and boiled for 5 minutes in 1x SDS-PAGE sample buffer [250 mmol/L Tris (pH 6.8), 10% SDS, 10% ß-mercaptoethanol, and 40% glycerol]. Protein samples were resolved by SDS-PAGE, transferred to nitrocellulose, and incubated with anti-SIRP1 antibody (1:3,000) for 1 hour, followed by incubation with horseradish peroxidase-linked anti-IgG secondary antibody (1:3,000; Amersham Pharmacia Biotech). Immunoreactive proteins were detected by enhanced chemiluminescence as described by the supplier.
Isolation of the 5'-Flanking Region Fragment of SIRP1.
Based on the published sequence of SIRP1/PTPNS1/SHPS-1 (GenBank accession number NM_080792, NM_004648), the 5'-UTR of SIRP1 was used to BLAST search the expressed sequence tag database for the transcripts containing the longest 5'-UTR. A clone (BG421441) that contained an additional 2,000 nucleotides at the 5'-end was used to search the human genome for the sequence upstream of the SIRP1 5'-UTR. A Homo sapiens PAC clone RP4-539M6 from chromosome 22 was retrieved (locus AC004832) and analyzed for putative transcription start sites. Using the Promoter Prediction tool of the Berkeley Drosophila Genome Project, a putative transcription start site was predicted to be located 50 bp upstream of the BG421441 5'-UTR with a score of 0.99. Translation is predicted to originate at position +261.
Plasmid Construct.
Based on the sequence of the RP4-539M9 PAC clone from chromosome 22 (AC004832 locus), two oligonucleotide primers, a plus primer (5'-CTTACGCGTAACTCATGGGCATTAAGATCAATTACTTGGCCAGGTGAGG-3', –1908/–1868) and a minus primer (5'-CAAAGATCTTTGCGCAAACTTGTTTTTCTGAGGTCAGCGCTGCGAGC-3', +176/+136), were designed to amplify the UTR of the SIRP1 gene using normal human genomic DNA (Promega) as a template. The 2084-bp fragment was confirmed by both plus and minus primers by DNA sequencing. Plasmid pGL3-2084 SIRP1 was generated by fusing this fragment with pGL3-Basic vector at the MluI and BglII sites.
Transient Transfections.
Transfections were performed using the FuGene reagent (Roche Molecular Biochemicals) according to the manufacturer’s instructions. Briefly, cells were seeded at a density of 1 x 105 cells/well in a 12-well plate 1 day in advance. Transfections were performed in duplicate using 6 µL of FuGene and 1 µg of the reporter plasmid carrying the SIRP1 promoter. The pSV2-ß-galactosidase vector (0.4 µg; Promega) was used to control transfection efficiency. Forty-eight hours after transfection, cells were harvested by removing the media, washing twice with PBS, and directly adding 100 µL of lysis buffer per well. Of this lysate, 10 µL were used for luciferase determination, and 50 µL were used for ß-galactosidase determination. These determinations were performed using the luciferase kit and the ß-galactosidase enzyme assay system (Promega).
|
RESULTS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
1.1 molecules are expressed at different levels in a variety of tissues (2)
. Two SIRP transcripts of 3.9 and 2.5 kb have been detected by Northern hybridization in heart, brain, placenta, lung, liver, skeletal muscle, kidney, and pancreas, with higher expression in brain, liver, and heart (2)
. Similarly, other workers have reported transcripts of 4.4 and 2.4 kb in different tissues, suggesting the occurrence of alternative splice forms (19
, 32)
. In the present study, we investigated the presence of SIRP1 transcripts in RNAs from nine well-characterized human glioblastoma cell lines by Northern blotting using a 32P-labeled, 2.2-kb SIRP1 cDNA fragment. We also used U87MG glioblastoma cells ectopically expressing SIRP1 protein (U87MG/SIRP1) as a positive control. We observed one prominent mRNA transcript of 3.9 kb and three additional transcripts of 6.0, 2.4, and 2.3 kb in SIRP1-expressing U87MG cells (Fig. 1A)
. There was variable expression of SIRP1 transcripts in glioblastoma cells lines with greatest SIRP1 abundance in U87MG, U373MG, U343, and LN229 cells (Fig. 1A)
.
|
1 protein by immunoprecipitating SIRP polypeptides with an anti-SIRP1 antibody in glioblastoma cell lysates, followed by immunoblotting with a polyclonal anti-SIRP1 antibody. The immunoblot results showed a prominent single Mr 90,000 band in U87MG/SIRP1 cells (Fig. 1B)
. The expression of SIRP1 in different cell lines was variable, and the highest levels of SIRP1 protein were found in U373 and LN229 cells. Collectively, our results indicate that the majority of glioblastoma cell lines express SIRP1 with multiple transcripts that may represent various alternative splice forms of the SIRP1 gene.
ErbB Inhibition and Serum Starvation Cooperate to Up-Regulate SIRP1 Expression in Human Glioblastoma Cells.
EGF treatment induces phosphorylation of the SIRP1/SHPS-1 receptor (25
, 33, 34, 35)
, suggesting that EGFR signaling may modulate SIRP1 expression and/or activation. In our previous work, we showed that overexpression of a truncated erbB2 mutant receptor (T691stop) reduced cell transformation and conferred increased susceptibility to apoptosis in U87MG glioblastoma cells by inhibiting EGFR/erbB signaling (30
, 31)
. Similarly, we also showed that overexpression of SIRP1 in U87MG glioblastoma cells reduced cell transformation and motility, and enhanced DNA damage-induced apoptosis (25)
. These studies suggested a possible link between EGFR signaling and SIRP expression and/or function. However, there are no reports demonstrating that SIRP expression is regulated by EGFR function or kinase activation. To evaluate the relationship between EGFR signaling and SIRP expression, we performed Northern blot analysis using a 32P-labeled, 2.2-kb SIRP1 cDNA on total RNAs from U87MG and U373MG glioblastoma cells in the presence or absence of serum. Parental cells were compared with their clonal derivatives expressing either the constitutively active EGFRvIII oncoprotein (also termed 
EGFR) or the truncated erbB2 mutant (T691stop), all of which have well-characterized phenotypes with distinct transforming efficiencies (30
, 31)
. EGFR (or EGFRvIII) lacks a portion of the extracellular domain, is constitutively phosphorylated, and confers a more malignant phenotype (30
, 36) . T691stop is a truncated form of erbB2 with a large cytoplasmic deletion, which includes the tyrosine kinase domain (p185neu) and the entire COOH terminus. This mutant receptor inhibits cell transformation and confers increased susceptibility to apoptosis by inducing the formation of nonfunctional heterodimeric complexes (30
, 31)
. Interestingly, U87MG/T691 and U373MG/T691 cell clones showed increased SIRP1 mRNA expression when compared with their parental controls (Figs. 2A
and 3A
). However, U87MG.EGFR cells showed no change in SIRP1 mRNA expression as compared with parental U87MG cells (Fig. 3A)
. These studies suggest that down-modulation of EGFR signaling leads to up-regulation of SIRP1 expression. Moreover, on serum starvation, all cells showed up-regulation of SIRP1 expression, suggesting that low serum might be an additional stimulus that increases SIRP1 mRNA synthesis. We then evaluated SIRP1 protein expression under similar conditions by immunoprecipitation assay using an anti-SIRP1 antibody and Western blot analysis. Consistent with the mRNA expression results, we observed increased SIRP1 protein levels in EGFR-inhibited clones and with serum starvation of cells (Figs. 2B
and 3B
).
|
|
1 receptor expression was determined by Northern blotting. EGF treatment reduced SIRP1 receptor expression in a dose-dependent manner without substantially affecting expression of the housekeeping GAPDH gene (Fig. 4A and B)
. The most effective concentration of EGF for down-regulating SIRP1 expression was 50 ng/mL, and saturation occurred after 100 ng/mL (Fig. 4A)
. A significant decrease in SIRP1 receptor expression was apparent at 6 hours, and the plateau was reached after 12 hours (Fig. 4B)
. These observations suggest that ligand-induced activation of EGFR/erbB2 complexes regulates expression of SIRP1 inhibitory receptor.
|
1 Messenger RNA Involves Induction of Gene Transcription.1 mRNA expression. To evaluate whether increase in SIRP1 expression is regulated at the level of transcription or on RNA stability, we inhibited the transcription of newly synthesized mRNA with actinomycin D (10 µg/mL) at different time intervals in both U373MG parental and EGFR-inhibited U373MG/T691 cells (Fig. 5A)
. We then plotted the relative SIRP1 mRNA levels (Fig. 5B)
and calculated the half-lives of the 3.9- and 6.0-kb mRNA species in both cell lines. The calculated half-lives for both mRNA species were nearly identical in both cell lines (3.9-kb species, 5 hours; 6.0-kb species, 2 hours), indicating that SIRP1 mRNA stability does not account for the differences in SIRP1 levels in the two isogenic cell lines. This observation suggested that there might be a difference in transcriptional regulation of the SIRP1 gene between the two cell lines distinguished by the level of endogenous EGFR activation.
|
1, we identified a 2084-bp region upstream of the SIRP1 5'-UTR on chromosome 22 and cloned this fragment into the pGL3 basic reporter plasmid carrying a luciferase gene (pGL3-SIRPprom-Luc) as described in Materials and Methods (Fig. 6A)
. The pGL3-SIRP-Luc reporter was cotransfected with either empty vector or the T691 inhibitory receptor into different glioblastoma cells to determine the effect of EGFR inhibition on SIRP1 promoter activity. We observed a 2- to 3.5-fold increase in SIRP1 promoter activity in all cells transfected with the T691 EGFR inhibitory receptor when compared with empty vector controls (Fig. 6B)
. To further support this observation, we treated U87MG parental cells with the EGFR kinase inhibitor AG1478 and the erbB2 kinase inhibitor AG825 to evaluate the effects on SIRP1 promoter activity. As expected, we observed enhanced promoter activity with both the inhibitors, suggesting that differences in the SIRP1 mRNA levels between two isogenic glioblastoma cell lines distinguished only by levels of functional activation of the erbB kinases could only be attributed to differences in activation of the SIRP1 promoter (Fig. 6C)
.
|
1 Promoter Activity.1 promoter activity, we examined whether pathways downstream of the EGFR activation loop played any role in SIRP1 promoter activation. U87MG cells were transiently transfected with the pGL3-SIRPprom-Luc reporter, serum-starved, and then treated with various pharmacological inhibitors. We observed that treatment of U87MG cells with the p42/44MAPK inhibitors PD98059 and U0126 up-regulated SIRP1 promoter activity (Fig. 7)
. Interestingly, treatment with PP1, a src kinase inhibitor, also increased SIRP1 promoter activity. Our results support a previous study showing that overexpression of v-src suppressed SHPS1 (rodent homologue of SIRP) mRNA expression via the Ras-MAPK pathway to promote transformation in fibroblast cells (37)
and suggest that src kinase and MAPK pathways may play an important role in regulating SIRP expression at the level of transcription. However, LY294002 and wortmannin, pharmacological inhibitors of the PI3K/Akt pathway, did not affect SIRP1 promoter activity (Fig. 7)
. This observation suggests that transcriptional activation of the SIRP1 gene may be regulated by EGFR/erbB2 and src kinases via extracellular signal-regulated kinase (ERK) 1/2.
|
Expression Is Dependent on New Protein Synthesis. expression required new protein synthesis, phenotypically inhibited U87MG cells expressing the ErbB2 (T691stop) mutant were incubated with different concentrations of cycloheximide and puromycin in the presence or absence of EGF (100 ng/mL). Inhibition of translation reduced mRNA levels of the 3.9-kb SIRP1 transcript, indicating that additional protein synthesis is required to stimulate SIRP1 expression on inhibition of the EGFR signaling module (Fig. 8A and B)
. However, there was an increase in the levels of 6.0-kb transcript on cycloheximide and puromycin treatment in the presence or absence of EGF, suggesting that the 6.0-kb transcript could be a precursor mRNA and that ligand-mediated EGFR activation may cooperate with inhibition of translation to block generation of mature SIRP1 mRNAs, to down-regulate SIRP1 expression (Fig. 4A and B)
.
|
1 gene (Fig. 8C)
. |
DISCUSSION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
1 is tyrosine-phosphorylated and activated by a variety of ligands, including EGF, platelet-derived growth factor, insulin, neurotrophins, lysophosphatidic acid, adhesion to fibronectin, growth hormone, colony-stimulating factor, and serum (1, 2, 3
, 8
, 17
, 24
, 25
, 33
, 35
, 38, 39, 40)
. Tyrosine phosphorylation of SIRP1 leads to recruitment of either of the two SH2 domain-containing protein tyrosine phosphatases, SHP-1 and SHP-2 (2
, 17
, 19)
. Binding of these protein tyrosine phosphatases to SIRP stimulates the catalytic activity of these phosphatases, which in turn exerts a negative regulatory effect on cell signaling. Previously, we reported (25)
that in glioblastoma cells, EGF stimulation led to an association between tyrosine-phosphorylated SIRP1 and SHP-2, resulting in reduced transformation, cell migration, and cell spreading and enhanced apoptosis after DNA damage. These observations suggest that modulation of SIRP1 levels or function may be important for both biology and therapy. However, there are no reports regarding the specific factors regulating the expression of SIRP1 inhibitory proteins.
In this study, we provide the first evidence, to our knowledge, of transcriptional regulation of SIRP1. Variable amounts of SIRP1 mRNA and protein in different glioblastoma cell lines suggest that expression of SIRP1 may be under some regulatory control mediated by EGFR kinase function. It has been established both in vitro and in vivo systems that EGFR interacts and forms active heterodimeric complexes with neu/c-erbB2 (p185neu) in response to EGF and neu differentiation factor (heregulin) to potentiate cell signaling and transformation (41, 42, 43, 44, 45, 46)
. In accordance with previous reports, we demonstrated that erbB/EGFR inhibition due to nonfunctional EGFR/erbB2 complexes resulted in inhibition of glioblastoma transformation and increased susceptibility to apoptosis (30
, 31)
. Furthermore, we have also shown that overexpression of SIRP1 in glioblastoma cells led to reduced EGFR-mediated transformation, motility, and spreading and resulted in increased susceptibility to apoptosis (25)
. These observations strongly indicate a link between EGFR/erbB2 signaling and SIRP1 expression and function in glioblastoma cells. The present observations using Northern blot analysis and immunoprecipitation assays showed elevated SIRP1 levels in cells expressing an erbB2 mutant receptor in comparison with parental cells. Based on these results, we report that modulation of EGFR receptor assembly and subsequent kinase activation at the cell surface modulate SIRP1 gene expression. SIRP1 may then participate in exerting negative regulatory effects on cell signaling cascades in erbB-driven human cancer cells, including glioblastoma cells.
We further extended our work to examine whether up-regulation of SIRP1 by EGFR blockade is controlled at the level of transcription. We observed that blockade of mRNA synthesis by actinomycin D in parental and mutant ErbB2-expressing cells did not affect the half-life of SIRP1 mRNAs in both the cell phenotypes, indicating that up-regulation of SIRP1 levels was not due to alteration in mRNA transcript stability. This suggested that down-modulation of EGFR signaling might up-regulate SIRP1 expression at the level of transcription. To more clearly define this observation, we cloned a 2084-bp fragment representing the 5'-UTR of SIRP1 transcripts into a luciferase reporter vector. A significant increase in promoter activity was observed when different cell lines were cotransfected with the T691 erbB2 inhibitory receptor and pGL3-SIRP-Luc reporter vector. This is in agreement with our Northern blot results that showed the ability of the erbB2 mutant to induce SIRP1 mRNA expression. Furthermore, pharmacological inhibition of EGFR and erbB2 tyrosine kinase activity increased SIRP1 promoter activity, supporting our hypothesis that down-modulation of erbB kinase activation and signaling leads to enhanced SIRP1 gene expression, which results in reduced transformation and increased apoptosis (25)
. Interestingly, pharmacological inhibition of src kinases and ERK1/2, but not PI3K/Akt, significantly up-regulated SIRP1 promoter activity, suggesting that expression of SIRP1 is regulated by EGFR and src kinases via the MAPK pathway.
We also investigated the effects of cycloheximide and puromycin, two well-known protein synthesis inhibitors, on SIRP1 mRNA expression. Surprisingly, inhibition of protein translation reduced the 3.9-kb mRNA but increased 6.0-kb mRNA levels in the presence or absence of EGF. One possible explanation is that the 6.0-kb mRNA may be an unspliced pre-mRNA. Thus, blocking of translation would block the synthesis of components of spliceosome required to splice 6.0-kb mRNA. Furthermore, there are two hypotheses that may explain our observations. One study showed that mRNA degradation is coupled to translation. Thus, blocking of protein synthesis prolongs mRNA half-life (47)
. Another hypothesis is that degradation is dependent on a labile protein whose synthesis is blocked by the translational inhibitor (48)
. Additionally, some translational inhibitors, such as cycloheximide, cause ribosomes to "freeze" on the mRNA, potentially shielding it from degradation by cytoplasmic RNases (49, 50, 51, 52)
. Our results demonstrating EGF-mediated down-regulation of SIRP1 expression (Fig. 4)
and a substantial increase in the 6.0-kb mRNA by cycloheximide or puromycin in the presence of EGF (Fig. 8)
suggest that EGFR activation might regulate the protein synthesis machinery to inhibit splicing and/or degradation of 6.0-kb SIRP1 mRNA. On the other hand, cycloheximide treatment down-modulated SIRP1 promoter activity, which correlates with the cyclohexaimide-induced decrease in the 3.9-kb mRNA transcript. This observation clearly suggest that SIRP1 up-regulation in EGFR-inhibited cells requires de novo protein synthesis and the synthesis of new intermediatory molecules that are essential for the activation of transcriptional machinery.
Serum starvation is a routinely used technique for pushing cells into a quiescent state to study the mechanism for a particular pathway or the mode of action of a specific enzyme under optimal conditions. It is generally believed that serum-starved cells are arrested in G1 or G0 phase, with most of the genes functioning at their basal expression level (53, 54, 55)
. However, there are some reports showing that serum starvation leads to up-regulation of certain genes. A study in breast carcinoma, bladder carcinoma, and osteosarcoma cell lines and in SV40-immortalized keratinocytes showed that serum starvation led to an increase in human alternative reading frame (ARF) mRNA levels, suggesting that ARF responds not only to oncogenic hyperproliferative signals but also to suboptimal growth conditions (56)
. The ARF protein is encoded by the INK4a locus as an alternative INK4a transcript (57, 58, 59)
. ARF has been shown to have tumor suppressor activity (60)
and to induce G1 cell cycle arrest in a p53-dependent manner (60, 61, 62)
. Similarly, a study in human colon carcinoma cells demonstrated that serum starvation up-regulated vascular endothelial growth factor mRNA expression and promoter activity via ERK1/2 activation (63)
. Interestingly, it was reported that in NIH3T3 and MCF-7 breast carcinoma cells, serum starvation up-regulated protein kinase C
, which associates with the cyclin E/cyclin-dependent kinase 2 complex and exerts a negative regulatory effect on cell cycle progression (64)
. A study using human T-cell lymphotrophic virus-infected T-cell lines N1186 [interleukin (IL)-2 dependent] and N1186-94 (IL-2-independent) showed that N1186 cells became arrested in G1 after serum and IL-2 deprivation and this resulted in elevated levels of p27Kip1 bound to the cyclin E/cyclin-dependent kinase 2 complex. However, N1186-94 cells failed to arrest in G1 on serum deprivation, suggesting that serum starvation cooperates with down-modulation of IL-2 receptor signaling to produce antiproliferative effects (65)
. In the present study, we observed that serum starvation led to up-regulation of SIRP1 levels in both parental and EGFR-inhibited T691-expressing cells, which is in agreement with the previously discussed studies. Collectively, the present observations and our previous reports on the effect of EGFR and SIRP1 on transformation, motility, and susceptibility to apoptosis (25
, 30
, 31)
suggest that an erbB-inhibited phenotype and serum starvation cooperate to induce SIRP1 gene transcription, which might result in activation of certain cell cycle-regulatory proteins that exert the negative effects on cell signaling reported previously (25)
. Present studies are under way to define mechanisms of SIRP1 inhibitory functions.
Future studies will aim to identify various transcriptional factors required for SIRP1 gene expression and to delineate putative cis-regulatory element-binding sites on the SIRP1 promoter to gain understanding of the regulatory mechanisms required for the up-regulation of SIRP1 gene expression. Defining the coupling between EGFR down-modulation and SIRP1 expression and functional activation may be important for the development of future therapies for human cancers, in particular central nervous system and lymphoid-related malignancies.
|
FOOTNOTES |
|---|
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Requests for reprints: Donald M. O’Rourke, 502 Stemmler Hall, Department of Neurosurgery, University of Pennsylvania School of Medicine, 36th and Hamilton Walk, Philadelphia, PA 19104. Phone: 215-898-2871; Fax: 215-898-9217; E-mail: orourked{at}mail.med.upenn.edu
Received 1/26/04. Revised 6/23/04. Accepted 7/14/04.
|
REFERENCES |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
and 
, U373MG cells; 
and 
, U373MG/T691 cells. 
