![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
|
|
|
|
|
||
Vol. 18, Issue 7, 2481-2490, July 2007
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
*Laboratory of Integrin Signaling, Van Andel Research Institute, Grand Rapids, MI 49503; and Division of Public Health Sciences, Program in Cancer Biology, Fred Hutchinson Cancer Research Center, Seattle, WA 98109
Submitted April 3, 2006;
Revised April 20, 2007;
Accepted April 24, 2007
Monitoring Editor: Richard Assoian
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
3
1 and 64 integrins. When placed in culture primary prostate basal epithelial cells secrete and adhere to their own LM5-rich matrix. Adhesion to LM5 is required for cell survival that is dependent on integrin-mediated, ligand-independent activation of the epidermal growth factor receptor (EGFR) and the cytoplasmic tyrosine kinase Src, but not PI-3K. Integrin-mediated adhesion via 31, but not 64 integrin, supports cell survival through EGFR by signaling downstream to Erk. PC3 cells, which do not activate EGFR or Erk on LM5-rich matrices, are not dependent on this pathway for survival. PC3 cells are dependent on PI-3K for survival and undergo caspase-dependent death when PI-3K is inhibited. The death induced by inhibition of EGFR or Src in normal primary prostate cells is not mediated through or dependent on caspase activation, but depends on the induction of reactive oxygen species. In addition the presence of an autophagic pathway, maintained by adhesion to matrix through 31 and 64, prevents the induction of caspases when EGFR or Src is inhibited. Suppression of autophagy is sufficient to induce caspase activation and apoptosis in LM5-adherent primary prostate epithelial cells. |
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
31 and 64, are critically involved in mediating survival. Genetic loss of LM5, or its receptors 3, 6, or 4 integrins, in vivo results in cell detachment and induction of caspase-mediated apoptosis, even in the presence of soluble factors (Ryan et al., 1999
; DiPersio et al., 2000). This detachment-induced form of apoptosis has been termed anoikis (Frisch and Screaton, 2001). In vitro anoikis can be rescued by expression of an activated form of FAK, Rac, or Akt (Frisch et al., 1996; Rytomaa et al., 2000; Coniglio et al., 2001), suggesting that integrin-mediated signaling through these molecules is required to maintain cell survival. However, studies in which specific signaling pathways are inhibited while integrins are still engaged suggest alternative pathways, such as Ras/Erk or Jnk, are required for integrin-mediated survival (Almeida et al., 2000; Manohar et al., 2004). Whether signaling from multiple pathways is involved in mediating integrin-dependent survival and whether different pathways are unique to specific cell types have not been extensively investigated.
In addition to classical caspase-mediated apoptosis, such as that observed during anoikis, several other mechanisms of cell death have been described (Melino et al., 2005). Other forms of cell death include caspase-independent cell death, autophagy, or cornification. The role of integrins in regulating cell survival through suppression of these other death pathways is unknown. However, some of the same integrin-induced signal transduction pathways that have been linked to survival are also important for regulating these alternative cell death pathways. For example the Ras/Erk and PI-3K pathways act as positive and negative regulators, respectively, of autophagy in several cell types (Kondo et al., 2005). Additionally, epithelial cells have been shown to undergo death by cornification in response to inhibition of Erk and Jnk, but not PI-3K (Uzgare and Isaacs, 2004). Finally, death induced by over expression of Ras, or suppression of Raf in melanoma cells leads to caspase-independent cell death (Chi et al., 1999; Panka et al., 2006). Whether integrin-induced activation of specific signaling pathways plays a role in regulating any of these cell death mechanisms has not been determined.
Although studies with various established cell lines have been extremely useful for elucidating potential signaling pathways involved in integrin-mediated survival, it is important to place the findings in the context of a defined organ system where the specific cell type, the integrins expressed, and the matrix being studied are better defined. Basal epithelial cells in the prostate gland express 64 and 31 integrins and adhere to a basement membrane rich in LM5 (Knox et al., 1994). When these cells are placed in culture they retain in vitro a majority of the properties seen in vivo, including the ability to secrete and organize their own LM5-rich matrix (Gmyrek et al., 2001; Yu et al., 2004).
Our work and that of others have demonstrated that integrin engagement is sufficient to activate receptor tyrosine kinases (Plopper et al., 1995; Miyamoto et al., 1996; Wang et al., 1996; Moro et al., 1998; Danilkovitch-Miagkova et al., 2000; Kuwada and Li, 2000; Marcoux and Vuori, 2003; Bill et al., 2004). We demonstrated that adhesion of normal epithelial cells to matrix is sufficient to induce activation of the epidermal growth factor receptor (EGFR), independently of ligand (Bill et al., 2004). In addition, we demonstrated that integrin-mediated activation of a subset of signaling pathways, namely the Ras/Erk and PI-3K/Akt pathways, are dependent on integrin-induced EGFR activation. Because both of these pathways have been implicated in regulating integrin-mediated survival, we hypothesized that integrin-mediated survival of epithelial cells via Ras/Erk or PI-3K/Akt pathways could be mediated through integrin-dependent activation of EGFR. To test this hypothesis, we assessed the ability of primary prostate epithelial cells (PECs) adherent to their endogenous LM5-rich matrix to survive in the context of EGFR and downstream signaling inhibitors.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
). Blocking antibodies for 4 integrin (ASC-8) and 3 integrin (P1B5) were purchased from Chemicon (Temecula, CA) and GoH3 6 integrin antibody was obtained from Becton-Dickinson.
Cell Culture
Primary cultures of human PECs were derived from normal human prostatic tissue and cultured as described previously (Gmyrek et al., 2001). Human samples were obtained after institutional IRB approval. PECs were maintained in Keratinocyte-SFM medium (Invitrogen, Carlsbad, CA) supplemented with bovine pituitary extract and epidermal growth factor (EGF). All experiments were conducted on cells between passages 3 and 5. PC3 cells were obtained from ATCC. PC3 cells were maintained in F12K medium (Invitrogen) supplemented with 10% fetal bovine serum, 2 mM glutamine, 50 U of penicillin, and 50 mg of streptomycin/ml.
Integrin Signaling
Preparation of cells for adhesion to extracellular matrices was carried out as described in Miranti (2002). Briefly, cells were growth factor-starved for 48 h, trypsinized, treated with soybean trypsin inhibitor (Invitrogen), washed in PBS, and placed in suspension in growth factor-free medium for 30–60 min. Cells were then either plated on tissue culture plates blocked with 1% BSA (Sigma, St. Louis, MO) to allow deposition of endogenous LM5-rich matrix or directly replated on LM5-coated plates obtained from culturing PECs as described previously for LM5-secreting cells (Xia et al., 1996). In some cases, PC3 cells were also plated on laminin 1 (LM1; Invitrogen). Similar results were obtained in PC3 cells on LM1 as on LM5. Occasionally cells were also treated with 2–10 ng/ml EGF (Upstate Biotechnology) or 50 ng/ml HGF (Calbiochem, La Jolla, CA) for 10 min. A suspension control was maintained at 37°C. Two hours after plating on the matrix cells were lysed either in Triton X-100 (50 mM Tris, pH 7.5, 100 mM NaCl, 0.5 mM EDTA, 1% Triton X-100, 50 mM NaF, 50 mM -glycerophosphate, 5 mM sodium pyrophosphate, 1 mM Na3VO4, 1 mM PMSF, 100 U/ml aprotinin, 10 µg/ml pepstatin, and 10 µg/ml leupeptin) or RIPA (10 mM Tris, pH 7.2, 158 mM NaCl, 1 mM EDTA, 0.1% SDS, 1% NaDOC, 1% Triton X-100, 1 mMNa3VO4, 1 mM PMSF, 100 U ml aprotinin, 10 µg/ml pepstatin, and 10 µg/ml leupeptin) buffers. Pharmacological inhibitors, PD168393, AG1478, LY294002, SU6656, or PP2, purchased from Calbiochem, were added to suspension cells 20 min before plating on matrix; except for SU6656, which was added 16 h before placing cells in suspension. All working concentrations of the pharmacological inhibitors were determined by titrating to the minimum inhibitor concentration that effectively blocked the target of the pharmacological inhibitor for the duration of our experiments. Inhibitor effectiveness was monitored by Western blotting. Specifically, PD168393 and AG1478 were tested for their ability to inhibit EGFR tyrosine phosphorylation, p130Cas tyrosine phosphorylation (a Src substrate) was used to test SU6656 and PP2, phosphorylation of Akt was used for LY294002, and U0126 was tested against phosphorylated Erk. Titrations were performed for each drug in each cell type.
Antibody Blocking Assays
Blocking Integrins.
For integrin blocking studies, PECs were starved and placed in suspension and then plated on 1% bovine serum albumin (BSA)-blocked eight-chamber slides in the presence of 10 µg/ml blocking anti-4 integrin antibody (ASC-8), anti-3 integrin antibody (P1B5), anti-6 integrin antibody (GoH3), or IgG. Cells were allowed to adhere to endogenous LM5-rich matrix for 48–72 h in the presence of the indicated antibodies. Cells were monitored for viability by Annexin V staining, caspase activation, autophagy induction with LC3-GFP, or for Erk activation by immunoblotting.
Blocking EGF Binding. Thirty minutes before plating, growth-factor–starved PEC suspension cells were pretreated with 0–10 µg/ml mAb to EGFR (AB225 [HB-8508, ATCC]) or 10 µg/ml nonspecific mouse IgG for 30 min with occasional mixing. Cells were allowed to adhere to LM5-rich matrix in the presence or absence of 2 ng/ml EGF for 2 h. Cells were lysed, and EGFR tyrosine phosphorylation and Erk activation were monitored in immunoprecipitates or cell lysates, respectively.
Cell Survival Assays. PECs were starved and placed in suspension as described above and then plated on 1% BSA-blocked tissue culture plates to allow deposition and adhesion to endogenous LM5 matrix. Serum-starved PC3 cells were plated on 1% BSA-blocked tissue culture plates precoated with 10 µg/ml laminin. Pharmacological inhibitors, 1 µM staurosporine (Promega, Madison, WI), 0.5 µM PD168393, 1 µM AG1478, 10 µM U0126, 10 µM LY294002, 0.5–2 µM SU6656, 10 µM PP2, 50 µM butylated hydroxyanisole (BHA), 1.25 mM N-acetylcysteine (NAC), or 10 mM 3-methyladenine (3MA), were then added. Cells were allowed to adhere for 4 h and then nonadherent cells were removed and drugs were replaced. Cells were incubated for an additional 72 h. LY294002 was replenished 48 h after plating.
To assess cell death, cells were stained with Annexin V using a kit obtained from Molecular Probes (Invitrogen). Staining was carried out according the supplied protocols. For all staining procedures both attached and floating cells were collected. Attached cells were removed by trypsinization and pooled with floating cells, and all cells were washed one time. For Annexin V staining, cells were resuspended in Annexin binding buffer (10 mM HEPES, 140 mM NaCl, 2.5 mM CaCl, pH 7.4) containing Alexa-fluor–conjugated Annexin V and incubated in the dark for 15 min. Samples were put on ice and immediately analyzed. Extent of staining was monitored by fluorescence-activated cell sorting (FACS) using a FACS Caliber (Becton-Dickinson) and CellQuest version 3.1.3 acquisition and analysis software (Becton-Dickinson) immediately after staining. On several occasions Annexin V staining was also monitored in adherent cells (without trypsinization) by microscopy using a Nikon Eclipse TE300 fluorescence microscope (Melville, NY) and OpenLab version 3.1.7 image analysis software (Improvision, Lexington, MA).
Caspase Activity Assays. Caspase 3 and 7 activity in PEC and PC3 cells was directly measured using a CaspaseGlo 3/7 kit (Promega) following the manufacturer's suggested protocol. For PECs 10,000 cells/well were plated on endogenous LM5 in BSA-coated 96-well plates in the presence of DMSO, 1 µM staurosporine (Promega), 0.5 µM PD168393, 10 µM PP2, or 10 mM 3MA, with or without 20 µM zVAD (Promega). For PC3 cells, 10,000 cells/well were plated on 1% BSA blocked 96-well plates precoated with 10 µg/ml laminin, respectively. Cells were plated in the presence of DMSO, 1 µM staurosporine, 0.5 µM PD168393, 10 µM LY294002, 2 µM SU6656, and 10 µM PP2. CaspaseGlo reagent was added at various times after inhibitor treatment and incubated for 1 h at room temperature in the dark. Relative light intensity was measured in each well using a Fluoroskan Assent FL fluorometer and software (Labsystems, Franklin, MA).
Immunoprecipitation and Immunoblotting. Immunoprecipitation mixtures containing 500–1000 µg protein were incubated with the appropriate antibodies for 3 h at 4°C with either protein A– or protein G–conjugated agarose beads (Pierce, Rockford, IL) to capture the complexes. All immunoprecipitated complexes were washed three times with their respective lysis buffer. Immunoprecipitated samples from adhesion assays were resuspended in 2x SDS sample buffer. In some cases 50–75 µg of total cell lysates were placed directly in 2x SDS sample buffer. All resuspended samples were boiled and subjected to SDS-PAGE, transferred to a polyvinylidene difluoride (PVDF) membrane. The PVDF membranes were blocked with 5% BSA in Tris-buffered saline containing 0.1% Tween 20 (TBST) for 2 h, followed by a 2-h incubation with the appropriate primary antibodies in 5% BSA/TBST. After several washes, blots were incubated with a horseradish peroxidase–conjugated secondary antibody (Bio-Rad, Richmond, CA) for 1 h in 5% BSA/TBST and visualized with a chemiluminescence reagent and captured by a CCD camera in a Bio-Rad Chemi-Doc Imaging System. Levels of activation, relative to total levels of protein, from blots captured by CCD camera were quantified using Quantity One software (Bio-Rad). Blots were stripped in low-pH 2% SDS at 65°C for 60 min, rinsed, and reprobed for total levels of protein in the immunoprecipitates or cell lysates.
Autophagy Assay. LC3-GFP in the pBABE expression vector was kindly provided by Dr. Jay Debnath (University of California, San Francisco, CA). The LC3-GFP cDNA BamHI/SalI restriction fragment was subcloned into the BglII and SalI restriction sites of pShuttle-CMV (Strategene, La Jolla, CA). pAd-Easy (Strategene) adenoviral recombinants containing LC3-green fluorescent protein (GFP) were generated in BJ5183-AD1 bacteria. HEK293 cells were transfected with adenoviral recombinant DNA and adenoviruses purified using a kit from Clontech and titrated by GFP expression in PECs.
PECs were infected at an moi of 2 with adenoviruses expressing LC3-GFP fusion protein. Twenty-four hours later, cells were growth factor-starved or left in complete medium and allowed to adhere to their own LM5. For antibody blocking experiments, antibodies were added at the time plating on matrix. Localization of LC3-GFP was monitored by standard fluorescence microscopy at 24 and 48 h after plating using a Nikon Eclipse TE300 fluorescence microscope and OpenLab version 3.1.7 image analysis software (Improvision).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
). Adhesion of PECs to this LM5-rich matrix induces activation of the ErbB family receptor tyrosine kinase EGFR (Figure 1A) and its downstream target Erk (Figure 1B). Adhesion to the LM5-rich matrix failed to stimulate detectable levels of Akt activation (Figure 1C). However, Akt was activated by treatment with EGF or human growth factor (HGF), demonstrating that the PI-3K/Akt pathway is intact in these cells. Inhibition of EGFR activity with the EGFR-specific inhibitors AG1478 (not shown) or PD168393 blocked integrin-induced EGFR tyrosine phosphorylation (Figure 1A) and downstream signaling to Erk (Figure 1B). Thus integrin-mediated activation of Erk on LM5-rich matrix is dependent on EGFR.
|
; Moro et al., 2002); however, in PECs, inhibition of Src activation by the Src kinase-specific inhibitor SU6656 (Blake et al., 2000) had no effect on EGFR activation or its downstream target Erk (Figure 1, A and B). Inhibition of Src did block integrin-mediated activation of its downstream substrate p130Cas (Figure 1D). Reciprocally inhibition of EGFR did not block integrin-mediated tyrosine phosphorylation of the Src substrate p130Cas. These data demonstrate that adhesion of PECs to LM5-rich matrix regulates two independent signaling pathways: one that activates EGFR, independent of ligand, and signals to Erk and one that activates Src and its downstream target p130Cas, independently of EGFR. To rule out the possibility that residual EGF was responsible for activation of EGFR and Erk, we used a blocking antibody that prevents EGF binding to EGFR, and thus EGFR activation by EGF. Stimulation of PECs with EGF effectively activates EGFR and Erk in PECs. Pretreatment of PECs with the EGFR blocking antibody did not significantly inhibit integrin-induced activation of EGFR or Erk (Figure 1E). In contrast, the same EGFR-blocking antibody blocked EGF-induced activation of EGFR and Erk, thereby reducing its activation to similar levels seen on matrix alone (Figure 1E). These data indicate that the ability of integrins to activate EGFR occurs independently of EGFR ligand.
The ability of integrins to activate Erk in epithelial cells is dependent on EGFR activation and is regulated in part by the level of EGFR expression (Moro et al., 1998; Bill et al., 2004). For instance, over expression of EGFR in fibroblasts (where Erk activation is not dependent on EGFR) leads to EGFR-dependent Erk activation in fibroblasts (Moro et al., 1998). Adhesion of the epithelial cell line PC3, which expresses threefold less EGFR than PECs (Figure 1F), to the LM5-rich matrix produced by PECs fails to activate EGFR or Erk (Figure 1, G and H). EGFR is active in PC3 cells because stimulation with EGF activates both EGFR and Erk. In contrast, adhesion to matrix does not increase Akt activation, which is partially constitutively activated (Figure 1I). This is likely due to the loss of Pten expression in these cells (Vlietstra et al., 1998). Consequently, constitutively active Akt is not blocked by EGFR inhibition. Thus, we predict that survival of these two cell lines on LM5-rich matrix is likely to be mediated by different signaling pathways.
EGFR and Src Independently Regulate Integrin-mediated Cell Survival in PECs
Treatment of LM5-rich adherent PECs with two EGFR-specific inhibitors, AG1478 or PD168393, results in the induction of cell death as measured by Annexin V staining (Figure 2, A and B). Maximal Annexin V staining is observed 72 h after drug treatment and occurs in over 85% of the cells (Figure 2A). Inhibition of Erk activation, the downstream target of EGFR, with U0126, but not inhibition of PI-3K with LY294002, induced cell death to the same extent as loss of EGFR signaling (Figure 2C). Together with the signaling data shown in Figure 1 these findings indicate that Erk signaling downstream of EGFR is required for PEC survival on LM5-rich matrix. Inhibition of Src by SU6656 or PP2 also induced cell death (Figure 2, B and C), with a time course and effectiveness that is similar to that seen with EGFR or Erk inhibition. Simultaneous inhibition of EGFR and Src did not increase the amount of cell death observed (Figure 2C), suggesting that although these molecules lie on separate signaling pathways (see Figure 1), they may regulate cell survival through a similar downstream mechanism. All drugs were effective at inhibiting their respective signaling pathways at the concentrations used (see Methods and Materials).
|
PC3 cells do not express the PI-3K/Akt inhibitor Pten, and consequently Akt is activated independent of matrix in PC3 cells and is not blocked when EGFR signaling is inhibited (Figure 1H). However, inhibition of the PI-3K pathway with LY294002 induced cell death in LM-adherent PC3 cells (Figure 2D). The extent of cell death induced by inhibition of PI-3K in PC3 cells was only twofold compared with the fourfold increase observed with inhibition of EGFR in PECs, suggesting that other mechanisms might be involved in regulating PC3 cell survival on matrix. We blocked Src signaling in PC3 cells with either SU6656 or PP2 and found that integrin-mediated survival in PC3 cells, like PECs, is also dependent on Src (Figure 2D). Inhibition of Src resulted in a three to fourfold increase in cell death. Thus cells that are able to activate EGFR/Erk signaling on LM5 are dependent on this pathway for survival. Cells unable to activate the EGFR/Erk pathway use other signaling pathways, such as PI-3K, for survival.
LM5-dependent Survival through EGFR Is Mediated by Engagement of 3 Integrin in PECs
PECs express 31 and 64, both of which mediate adhesion to LM5 (Delwel et al., 1994; Niessen et al., 1994). As expected, placing PECs in suspension rapidly induces cell death, with 80–90% of cells displaying Annexin V positivity within 6 h (Figure 3A). To determine which of these integrin receptors, 31 or 64, is mediating survival, we used specific blocking antibodies raised against 3 or 4 integrin. PECs treated with either 3 or 4 blocking antibodies were still able to adhere to LM5-rich matrix; however, treatment with anti-3 antibody blocked cell spreading (not shown). Blocking antibody to 3 integrin induced cell death to a similar extent as that seen with inhibition of EGFR or Src (Figure 3, B and C), i.e., 70–80%. Blocking 4 integrin on the other hand did not compromise survival. Furthermore, cells treated with blocking antibodies to 3 integrin, but not 4, were also defective in activating Erk (Figure 3D) and EGFR (not shown). These data indicate that signaling through 3 integrin regulates LM5-mediated activation of EGFR and its subsequent activation of Erk. Thus signaling from LM5 through 31 to EGFR and downstream to Erk is critical for regulating survival.
|
|
Cell Death in PECs Is Not Due To Autophagy
Nutrient and serum deprivation can induce cells to enter a state of survival termed autophagy (Lum et al., 2005). Knowing that our experiments were conducted under growth factor starvation conditions, we suspected that the cell death we were observing could be autophagic in nature (Baehrecke, 2005). To address this possibility, we first determined whether placing PECs under starvation conditions was sufficient to induce autophagy. LC3 protein is generally present throughout the cell, and upon induction of autophagy it is processed and incorporated into autophagic vacuoles. PECs were infected with an adenovirus that expresses an LC3-GFP fusion protein. Induction of autophagy is indicated by a shift from very diffuse LC3-GFP fluorescence throughout the cell to punctate fluorescence within the cytoplasm (Boya et al., 2005). As early as 24 h after plating growth factor-starved LC3-GFP expressing PECs on LM5, punctate fluorescence was evident. By 48 h, multiple punctate fluorescent areas were observed in over 90% of cells under growth factor starvation conditions (Figure 5, A and B). Punctate fluorescence was rarely observed (in <10% in cells) in normal growth media at 24 or 48 h after plating on LM5 (Figure 5, A and B). Thus adhesion to LM5-rich matrix in growth factor-deprived cells leads to induction of autophagy. Furthermore, adhesion of growth factor-deprived PECs to their LM5-rich matrix is sufficient to mediate cell survival for at least 8 d. However, removal from matrix and placement in suspension results in maximum Annexin V positivity within 6 h (Figure 3A). Pretreatment of GFP-LC3 expressing cells with integrin-blocking antibodies to 3 integrin resulted in a threefold reduction in LC3 punctate staining versus a 1.5-fold reduction with 4 or 6 blocking antibodies (Figure 5C). Thus adhesion of growth-factor deprived PECs to LM5-rich matrix primarily via 31, but also to some extent through 64, is required to maintain autophagy.
|
).
|
31 integrin regulates cell survival by maintenance of autophagy and signaling through EGFR/Erk and if blocking autophagy and EGFR induces caspase-mediated death, then blocking 31 integrin should also induce caspase activation. Cells pretreated with 3 blocking antibodies, but not 6 or 4 blocking antibodies, induced a fivefold increase in caspase 3/7 activity (Figure 6C). Caspase-independent death has been linked to the generation of reactive oxygen species (ROS) in several cell systems. To determine if the mechanism by which inhibition of EGFR signaling induces cell death is due to the generation of ROS, PECs were pretreated with two different ROS inhibitors, 50 µM butylated hydroxyanisole (BHA) or 1.25 mM N-acetylcysteine (NAC). Treatment of PECs with NAC or BHA alone did not significantly increase or decrease the basal level of Annexin V staining. However, pretreatment with either NAC or BHA prevented the induction of Annexin V staining in cells treated with the EGFR inhibitor PD168393 (Figure 7). Thus loss of integrin-mediated signaling through EGFR results in an increase in ROS, which is required for the subsequent induction of caspase-independent death.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
31, and to a lesser extent 64, is required for autophagy. Under starvation conditions cell survival is also dependent on at least two additional independent integrin signaling pathways: 1) integrin-mediated activation of EGFR and subsequent signaling to Erk and 2) integrin-mediated activation of Src, the former being dependent on 31 integrin. Interestingly, there was no activation of the PI-3K/Akt signaling pathway in PECs on LM5; consequently there was no dependence on this pathway for survival in normal PECs. In the presence of an intact autophagy pathway, inhibition of EGFR/Erk or Src is sufficient to induce cell death, but this death is mediated through a caspase-independent mechanism that is dependent on the generation of reactive oxygen species. On the other hand, disruption of autophagy, pharmacologically or by blocking 31, leads to caspase activation and death.
|
). These differences may reflect the cell type, fibroblasts versus epithelial cells, or the matrix that was used, fibronectin versus laminin. Although LM5 is the predominant matrix secreted by PECs, we cannot rule out the possibility that additional matrix materials may also be present that could be contributing to PEC survival. Nonetheless, we have demonstrated that integrin-mediated signaling through 31 is required for survival mediated by both autophagy and EGFR/Erk signaling on this PEC-generated matrix. Because PECs rapidly secrete their own LM5-rich matrix, it has not been possible to determine the role of different matrices in regulating long-term survival of PECs. However, in short-term 1-h adhesion assays we have observed that adhesion of PECs to LM1, but not the LM5-rich matrix, is sufficient to activate Akt, suggesting signaling pathways on other matrices may be important in survival. We are currently developing siRNA-based methods for eliminating LM5 from the PEC matrix, which will allow us to investigate the role of EGFR and other signaling pathways in mediating survival on other matrices.
Integrin-mediated survival of primary keratinocytes on LM5 has been shown to involve signaling to Erk (Manohar et al., 2004). As in our studies, keratinocyte survival on LM5 was dependent on 31 integrin. Whether Erk activation in keratinocytes is dependent on integrin signaling through EGFR has not been reported. However, autocrine ligand-mediated signaling through EGFR to Erk was shown to contribute to cell survival of keratinocytes in suspension (Jost et al., 2001). We have ruled out a role for autocrine ligand involvement in PECs, because ligand binding blocking antibodies do not block integrin-mediated EGFR activation or downstream signaling to Erk. Given the similar findings in primary prostate epithelial cells and keratinocytes, we predict that keratinocyte survival on LM5 should also involve integrin-mediated activation of EGFR and subsequent downstream signaling to Erk.
Whether integrin-mediated activation of other receptor tyrosine kinases is involved in regulating survival on matrix is not known. However, it was recently demonstrated that ligand-independent activation of c-Met in PC3 cells was required for cell survival (Shinomiya et al., 2004). The specific integrins, matrix, and signaling pathways involved in c-Met–mediated survival are currently unknown. Furthermore, whether c-Met regulates integrin-mediated survival in normal primary PECs is also unknown. Our data indicate that PC3 cells, which express low levels of EGFR relative to PECs, do not activate EGFR or Erk upon integrin engagement and do not depend on this pathway for integrin-mediated survival. Instead survival of PC3 cells requires PI-3K and Src. c-Met is known to activate these signaling pathways in response to HGF, but whether c-Met participates in integrin-mediated signaling to PI-3K or Src in PC3 cells has not been determined.
Surprisingly, interference with integrin signaling through EGFR/Erk or Src leads to caspase-independent death. This was unexpected, because cell death due to anoikis has been reported to be caspase-dependent. Our results suggest that the mechanism of cell death induced during complete loss of cell adhesion through integrins is different from the mechanism of cell death induced by interfering with specific signaling downstream of integrin engagement. One possible explanation is that loss of attachment is likely to interfere with several different signaling pathways simultaneously, whereas our studies individually dissected distinct pathways. In fact, blocking 31 alone was sufficient to induce caspase activity in PECs, similar to what has previously been observed in keratinocytes derived from integrin 3 null mice (Manohar et al., 2004). We have demonstrated that at least one of the signaling pathways activated by 31 is EGFR/Erk. Src activation is also important for integrin-mediated survival, but inhibition of both EGFR and Src was also not sufficient to induce caspases, indicating additional survival pathways are involved.
Previous studies in mammary epithelial cells demonstrated that cells in the center of acinar structures undergo autophagy and die during morphogenesis (Melino et al., 2005). However, the current prevailing theory on the primary role of autophagy is to promote temporary survival under growth factor and nutrient deprivation conditions, rather than to be a direct mechanism for programmed cell death (Lum et al., 2005). Interestingly, inhibiting autophagy in LM5-adherent PECs induced caspase activation and cell death and did not rescue cells induced to die by inhibition of EGFR/Erk, suggesting that cell death was not dependent on autophagy. Thus adhesion of starved PECs to a LM5-rich matrix induces an autophagic state that permits survival, but further assault by inhibiting EGFR/Erk activation leads to caspase-independent death.
Blocking 31 integrin significantly reduced the extent of autophagy induced under starvation conditions. Thus, in addition to regulating EGFR/Erk, 31 integrin is also required to maintain autophagy. Although blocking 64 integrin also reduced autophagy, the effect was not as dramatic as blocking 31. Furthermore, 64 was not required for EGFR/Erk activation and disruption of this integrin alone failed to induce caspase activation or cell death. Therefore, the small reduction in autophagy seen by blocking 64 may not be sufficient enough to overcome other survival pathways that are active in the cell, such as EGFR/Erk and/or Src. Interestingly, inhibition of both autophagy and EGFR signaling lead to a small (although not statistically significant in the four assays examined), but consistent increase in caspase activation, suggesting that in the absence of autophagy signaling through EGFR/Erk may still contribute to cell survival. One possibility is that signaling through EGFR/Erk helps to maintain autophagy. However, if this is true, then there must be other pathways involved, because inhibition of EGFR/Erk alone is not sufficient to induce caspase-mediated death.
It is interesting to note that signaling through PI-3K is actually inhibitory to the development of autophagy (Rusten et al., 2004). Adhesion of PECs to LM5 does not activate the PI-3K pathway. This suggests that the absence of strong PI-3K signaling permits the survival of PECs on LM5 through autophagy. It is striking that cell death induced by loss of EGFR/Erk or Src signaling is not sufficient to activate caspases. This suggests the existence of a strong anticaspase mechanism present in PECs. Recent studies have suggested that the presence of an autophagic state can be inhibitory to the activation of caspases (Degenhardt et al., 2006; Abedin et al., 2007). One model proposes that the autophagy pathway selectively targets damaged mitochondria for destruction by walling them off from the cytoplasm and thus preventing the release of enzymes required for the induction of caspases. Therefore, it is possible that the absence of a PI-3K pathway may allow this shift to a caspase inhibitory state. Thus in the EGFR/Erk inhibited cells, caspase-mediated death is dominantly inhibited, forcing other death mechanisms to be activated when this level of stress is induced.
Many human prostate cancers have reduced levels of the negative PI-3K regulator, Pten, and the PI-3K/Akt pathway is constitutively activated in those tumors (McMenamin et al., 1999). Furthermore, the development of prostate cancer is accompanied by the loss of LM5 in the basement membrane (Davis et al., 2001). Therefore, given that autophagy is driven by LM5-mediated adhesion and suppression of PI-3K signaling, we would predict that loss of LM5 and increased PI-3K signaling would prevent the induction of an autophagy survival pathway in tumor cells and make them more sensitive to caspase-mediated death.
Several caspase-independent mechanisms of cell death have been described, including activation of cathepsins, calpeptins, ROS, and release of numerous destructive enzymes from the mitochondria (Kroemer and Martin, 2005). To date we have been unable to detect release of cytochrome C from mitochondria in PECs treated with EGFR inhibitors, and inhibition of calpeptin did not rescue the death induced by inhibiting EGFR/Erk signaling (not shown). However, by blocking the generation of ROS with two different inhibitors, we were able to prevent the cell death induced by inhibition of EGFR/Erk. Thus adhesion to matrix and signaling through EGFR/Erk may act to limit ROS production. Integrin 1 null mesangial cells have been reported to have enhanced ligand-independent EGFR activation and excessive ROS production, suggesting that integrin signaling can help to modulate EGFR activation and limit ROS production (Chen et al., 2007). Another report suggests that 1 negatively regulates EGFR activation by stimulating the activity of TC-PTP (Mattila et al., 2005). However, if this same mechanism is acting in PECs, then loss of EGFR/Erk signaling would lead to reduced ROS production rather than its increase. Thus the mechanism by which loss of EGFR/Erk signaling leads to enhanced ROS production is not clear. Interestingly, a recent report indicates that high levels of ROS are required for starvation-induced autophagy (Scherz-Shouval et al., 2007). Therefore it is possible that one side effect of EGFR/Erk inhibition may be to further enhance autophagy through increased generation of ROS.
|
ACKNOWLEDGMENTS |
|---|
|
Footnotes |
|---|
Present address: Department of Physiology, Michigan State University, Lansing, MI 48824. 
Address correspondence to: Cindy K. Miranti (cindy.miranti{at}vai.org).
Abbreviations used: EGFR, epidermal growth factor receptor; PEC, prostate epithelial cells; LM, laminin; PI-3K, phosphatidyl inositol 3-kinase.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Almeida, E. A., Ilic, D., Han, Q., Hauck, C. R., Jin, F., Kawakatsu, H., Schlaepfer, D. D., and Damsky, C. H. (2000). Matrix survival signaling: from fibronectin via focal adhesion kinase to c-Jun NH(2)-terminal kinase. J. Cell Biol 149, 741–754.
Baehrecke, E. H. (2005). Autophagy: dual roles in life and death? Nat. Rev. Mol. Cell Biol 6, 505–510.[CrossRef][Medline]
Bill, H. M., Knudsen, B., Moores, S. L., Muthuswamy, S. K., Rao, V. R., Brugge, J. S., and Miranti, C. K. (2004). Epidermal growth factor receptor-dependent regulation of integrin-mediated signaling and cell cycle entry in epithelial cells. Mol. Cell. Biol 24, 8586–8599.
Blake, R. A., Broome, M. A., Liu, X., Wu, J., Gishizky, M., Sun, L., and Courtneidge, S. A. (2000). SU6656, a selective src family kinase inhibitor, used to probe growth factor signaling. Mol. Cell. Biol 20, 9018–9027.
Boya, P. et al. (2005). Inhibition of macroautophagy triggers apoptosis. Mol. Cell. Biol 25, 1025–1040.
Chen, X. et al. (2007). Integrin 11 controls reactive oxygen species synthesis by negatively regulating epidermal growth factor receptor-mediated Rac activation. Mol. Cell. Biol 27, 3313–3326.
Chi, S. et al. (1999). Oncogenic Ras triggers cell suicide through the activation of a caspase-independent cell death program in human cancer cells. Oncogene 18, 2281–2290.[CrossRef][Medline]
Coniglio, S. J., Jou, T. S., and Symons, M. (2001). Rac1 protects epithelial cells against anoikis. J. Biol. Chem 276, 28113–28120.
Danilkovitch-Miagkova, A., Angeloni, D., Skeel, A., Donley, S., Lerman, M., and Leonard, E. J. (2000). Integrin-mediated RON growth factor receptor phosphorylation requires tyrosine kinase activity of both the receptor and c-Src. J. Biol. Cell 275, 14783–14786.
Davis, T. L., Cress, A. E., Dalkin, B. L., and Nagle, R. B. (2001). Unique expression pattern of the 64 integrin and laminin-5 in human prostate carcinoma. Prostate 46, 240–248.[CrossRef][Medline]
Degenhardt, K. et al. (2006). Autophagy promotes tumor cell survival and restricts necrosis, inflammation, and tumorigenesis. Cancer Cell 10, 51–64.[CrossRef][Medline]
Delwel, G. O., de Melker, A. A., Hogervorst, F., Jaspars, L. H., Fles, D. L., Kuikman, I., Lindblom, A., Paulsson, M., Timpl, R., and Sonnenberg, A. (1994). Distinct and overlapping ligand specificities of the 3A1 and 6A1 integrins: recognition of laminin isoforms. Mol. Biol. Cell 5, 203–215.[Abstract]
DiPersio, C. M., van Der Neut, R., Georges-Labouesse, E., Kreidberg, J. A., Sonnenberg, A., and Hynes, R. O. (2000). 31 and 64 integrin receptors for laminin-5 are not essential for epidermal morphogenesis and homeostasis during skin development. J. Cell Sci 113, 3051–3062.[Abstract]
Frisch, S. M., and Screaton, R. A. (2001). Anoikis mechanisms. Curr. Opin. Cell Biol 13, 555–562.[CrossRef][Medline]
Frisch, S. M., Vuori, K., Kelaita, D., and Sicks, S. (1996). A role for Jun-N-terminal kinase in anoikis; suppression by bcl-2 and crmA. J. Cell Biol 135, 1377–1382.
Gmyrek, G. A., Walburg, M., Webb, C. P., Yu, H. M., You, X., Vaughan, E. D., Vande Woude, G. F., and Knudsen, B. S. (2001). Normal and malignant prostate epithelial cells differ in their response to hepatocyte growth factor/scatter factor. Am. J. Pathol 159, 579–590.
Jost, M., Huggett, T. M., Kari, C., and Rodeck, U. (2001). Matrix-independent survival of human keratinocytes through an EGF receptor/MAPK-kinase-dependent pathway. Mol. Biol. Cell 12, 1519–1527.
Knox, J. D., Cress, A. E., Clark, V., Manriquez, L., Affinito, K. S., Dalkin, B. L., and Nagle, R. B. (1994). Differential expression of extracellular matrix molecules and the 6-integrins in the normal and neoplastic prostate. Am. J. Pathol 145, 167–174.[Abstract]
Kondo, Y., Kanzawa, T., Sawaya, R., and Kondo, S. (2005). The role of autophagy in cancer development and response to therapy. Nat. Rev. Cancer 5, 726–734.[CrossRef][Medline]
Kroemer, G., and Martin, S. J. (2005). Caspase-independent cell death. Nat. Med 11, 725–730.[CrossRef][Medline]
Kuwada, S. K., and Li, X. (2000). Integrin 5/1 mediates fibronectin-dependent epithelial cell proliferation through epidermal growth factor receptor activation. Mol. Biol. Cell 11, 2485–2496.
Lum, J. J., DeBerardinis, R. J., and Thompson, C. B. (2005). Autophagy in metazoans: cell survival in the land of plenty. Nat. Rev. Mol. Cell Biol 6, 439–448.[CrossRef][Medline]
Manohar, A., Shome, S. G., Lamar, J., Stirling, L., Iyer, V., Pumiglia, K., and DiPersio, C. M. (2004). 31 integrin promotes keratinocyte cell survival through activation of a MEK/ERK signaling pathway. J. Cell Sci 117, 4043–4054.
Marcoux, N., and Vuori, K. (2003). EGF receptor mediates adhesion-dependent activation of the Rac GTPase: a role for phosphatidylinositol 3-kinase and Vav2. Oncogene 22, 6100–6106.[CrossRef][Medline]
Mattila, E., Pellinen, T., Nevo, J., Vuoriluoto, K., Arjonen, A., and Ivaska, J. (2005). Negative regulation of EGFR signalling through integrin-11-mediated activation of protein tyrosine phosphatase TCPTP. Nat. Cell Biol 7, 78–85.[CrossRef][Medline]
McMenamin, M. E., Soung, P., Perera, S., Kaplan, I., Loda, M., and Sellers, W. R. (1999). Loss of PTEN expression in paraffin-embedded primary prostate cancer correlates with high Gleason score and advanced stage. Cancer Res 59, 4291–4296.
Melino, G., Knight, R. A., and Nicotera, P. (2005). How many ways to die? How many different models of cell death? Cell Death Differ 12, (Suppl 2), 1457–1462.[CrossRef][Medline]
Miranti, C. K. (2002). Application of cell adhesion to study signaling networks. Methods Cell Biol 69, 359–383.[Medline]
Miyamoto, S., Teramoto, H., Gutkind, J. S., and Yamada, K. M. (1996). Integrins can collaborate with growth factors for phosphorylation of receptor tyrosine kinases and MAP kinase activation: roles of integrin aggregation and occupancy of receptors. J. Cell Biol 135, 1633–1642.
Moro, L. et al. (2002). Integrin-induced epidermal growth factor (EGF) receptor activation requires c-Src and p130Cas and leads to phosphorylation of specific EGF receptor tyrosines. J. Biol. Chem 277, 9405–9414.
Moro, L., Venturino, M., Bozzo, C., Silengo, L., Altruda, F., Beguinot, L., Tarone, G., and Defilippi, P. (1998). Integrins induce activation of EGF receptor: role in MAP kinase induction and adhesion-dependent cell survival. EMBO J 17, 6622–6632.[CrossRef][Medline]
Niessen, C. M., Hogervorst, F., Jaspars, L. H., de Melker, A. A., Delwel, G. O., Hulsman, E. H., Kuikman, I., and Sonnenberg, A. (1994). The 64 integrin is a receptor for both laminin and kalinin. Exp. Cell Res 211, 360–367.[CrossRef][Medline]
Panka, D. J., Wang, W., Atkins, M. B., and Mier, J. W. (2006). The Raf inhibitor BAY 43–9006 (Sorafenib) induces caspase-independent apoptosis in melanoma cells. Cancer Res 66, 1611–1619.
Plopper, G. E., McNamee, H. P., Dike, L. E., Bojanowski, K., and Ingber, D. E. (1995). Convergence of integrin and growth factor receptor signaling pathways within the focal adhesion complex. Mol. Biol. Cell 6, 1349–1365.[Abstract]
Rusten, T. E., Lindmo, K., Juhasz, G., Sass, M., Seglen, P. O., Brech, A., and Stenmark, H. (2004). Programmed autophagy in the Drosophila fat body is induced by ecdysone through regulation of the PI3K pathway. Dev. Cell 7, 179–192.[CrossRef][Medline]
Ryan, M. C., Lee, K., Miyashita, Y., and Carter, W. G. (1999). Targeted disruption of the LAMA3 gene in mice reveals abnormalities in survival and late stage differentiation of epithelial cells. J. Cell Biol 145, 1309–1323.
Rytomaa, M., Lehmann, K., and Downward, J. (2000). Matrix detachment induces caspase-dependent cytochrome c release from mitochondria: inhibition by PKB/Akt but not Raf signalling. Oncogene 19, 4461–4468.[CrossRef][Medline]
Scherz-Shouval, R., Shvets, E., Fass, E., Shorer, H., Gil, L., and Elazar, Z. (2007). Reactive oxygen species are essential for autophagy and specifically regulate the activity of Atg4. EMBO J 26, 1749–1760.[CrossRef][Medline]
Shinomiya, N., Gao, C. F., Xie, Q., Gustafson, M., Waters, D. J., Zhang, Y. W., and Vande Woude, G. F. (2004). RNA interference reveals that ligand-independent met activity is required for tumor cell signaling and survival. Cancer Res 64, 7962–7970.
Uzgare, A. R., and Isaacs, J. T. (2004). Enhanced redundancy in Akt and mitogen-activated protein kinase-induced survival of malignant versus normal prostate epithelial cells. Cancer Res 64, 6190–6199.
