HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

J. Biol. Chem., Vol. 277, Issue 49, 47826-47833, December 6, 2002

This Article Abstract Full Text (PDF) All Versions of this Article: 277/49/47826    most recent M204901200v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Jarad, G. Articles by Schelling, J. R. Search for Related Content PubMed PubMed Citation Articles by Jarad, G. Articles by Schelling, J. R. Social Bookmarking                      What's this?

Fas Activation Induces Renal Tubular Epithelial Cell ₈ Integrin Expression and Function in the Absence of Apoptosis^*

George Jarad, Bingcheng Wang^§, Shenaz Khan, Jay DeVore, Hui Miao, Karen Wu, Stephen L. Nishimura^¶, Barbara A. Wible, Martha Konieczkowski, John R. Sedor^**, and Jeffrey R. Schelling

From the Departments of  Medicine, ^§ Pharmacology,  Biochemistry, and ^** Physiology and Biophysics, Rammelkamp Center for Education and Research, MetroHealth Medical Center Campus, Case Western Reserve University School of Medicine, Cleveland, Ohio 44109-1998 and the ^¶ Departments of Pathology and Lung Biology, Pulmonary Division, University of California at San Francisco School of Medicine, San Francisco, California 94143

Received for publication, May 17, 2002, and in revised form, September 3, 2002

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Cell fate following Fas (CD95) ligand or agonistic anti-Fas antibody stimulation is determined by multiple factors, including Fas expression level, microdomain localization, and modulating cytokines. Highly expressed Fas clusters and activates a canonical apoptosis signaling pathway. In less susceptible cells, Fas transduces apoptosis-independent signals, which are not well defined, but have been linked to inflammation, angiogenesis, and fibrosis. To identify apoptosis-independent Fas pathways, cultured renal tubular epithelial cells were stimulated with agonistic anti-Fas antibodies under conditions that did not cause cell death. Analysis of filter cDNA microarrays revealed ₈ integrin subunit mRNA induction in Fas-stimulated cells. ₈ integrin mRNA expression increased within 3-6 h of Fas ligation due to enhanced mRNA stabilization, and mRNA increases were sustained for 48-72 h. Expression of plasma membrane ₈ integrin, as well as its heterodimer partner _v, was increased by Fas activation with a similar kinetic pattern. Fas-induced _v8 expression correlated with increased migration to vitronectin, the ligand for _v8. Results from studies with function-blocking antibodies against other _v integrins or suppression of ₈ integrin expression by RNA interference demonstrated that induced ₈ integrin expression mediated Fas-stimulated migration. We conclude that _v8 integrin induction defines an unexpected role for Fas in cell migration, rather than as a cell death receptor.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Fas (CD95, APO-1) is a ubiquitously expressed member of the tumor necrosis factor receptor superfamily, which mediates diverse cellular responses, including proliferation, inflammation, angiogenesis, and apoptosis. We have previously demonstrated that Fas-dependent renal tubular epithelial cell (RTC)¹ apoptosis mediates tubular atrophy (1, 2), a hallmark of progressive renal disease. Plasma membrane Fas is expressed as a pre-assembled glycoprotein homotrimer (3, 4). In highly susceptible (type I) cells, Fas binding by Fas ligand in vivo or agonistic anti-Fas antibodies in vitro causes clustering of Fas multimers within ceramide-rich lipid rafts and ezrin-containing cytoskeletal compartments (5-7), which leads to apoptosis following rapid aggregation of adaptor molecules and caspase complexes at the cytoplasmic Fas death domain. Fas-overexpressing cells have even been associated with ligand-independent apoptosis (8), suggesting that Fas surface density and caspase proximity promote apoptosis signaling within microdomains (9).

There is a spectrum of Fas responses, however. In contrast to type I cells, which rapidly activate caspases through signals generated at the plasma membrane, type II cells are relatively resistant to Fas-induced apoptosis, with a more prolonged signal transduction cascade that ultimately involves release of cytochrome c, Apaf-1, and apoptosis-inducing factor from mitochondria, leading to apoptosome formation, activation of cytosolic caspases, and DNA degradation. A third group, characterized by the RTC, is even more resistant to Fas-dependent apoptosis, despite constitutive Fas surface expression (2, 10, 11), although these cells can be converted to a type I or II phenotype upon induction of Fas expression (1, 2, 10, 12, 13).

Multiple explanations for diminished basal RTC apoptosis sensitivity have been proposed, such as inadequate clustering of Fas under conditions of low surface density and altered intracellular pro- and anti-apoptotic molecule activities (13); but the physiologic role of constitutively expressed RTC Fas is not understood. One theory is that apoptosis is a default process, whereby the continuous presence of survival factors is required for evasion of apoptosis (14), and when RTC undergo pathologic cell death, in the context of acute or chronic renal failure, apoptosis programs must be efficiently executed, without abrupt need for synthesis of cell death machinery. According to this paradigm, the sole function for constitutively expressed RTC Fas would be as a death receptor that is perpetually poised for activation. Alternatively, RTC Fas could regulate cell death-independent processes under homeostatic or pathophysiologic circumstances and only rarely function as a death receptor after survival factors have been depleted, such as in RTC deletion associated with tubular atrophy. Indeed, Fas activation has been associated with multiple apoptosis-independent processes in other tissues, including proliferation, fibrosis, inflammation, and cytokine secretion (reviewed in Ref. 15); angiogenesis (16); and in RTC, c-Jun NH₂-terminal kinase (JNK) activation (13). These observations suggest that RTC Fas may transduce dual apoptosis and non-apoptosis pathways, similar to other family members such as the tumor necrosis factor receptor and CD40.

To identify RTC Fas-regulated, apoptosis-independent pathways, we utilized a high throughput, cDNA hybridization array strategy. Unexpectedly, _v8 integrin expression and function were found to be up-regulated by RTC Fas stimulation. Although there is precedence for cross-talk between apoptosis and integrin pathways inasmuch as apoptosis has been associated with integrin detachment from extracellular matrix ligand (17) or unligated integrins in adherent cells (18), our results represent the first description of adhesion molecule up-regulation by a death receptor.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Antibodies-- Rabbit antiserum was generated against the human ₈ integrin cytoplasmic domain according to previously described methods (19). Anti-_v3 (clone LM609), anti-_v5 (clone P1F6), anti-_v6 (clone 10D5), and anti-_v (AV1) antibodies were purchased Chemicon International, Inc. (Temecula, CA). Agonistic anti-human Fas IgM (clone CH11) was from Kamiya (Seattle, WA). Agonistic anti-human Fas IgG (clone DX2), agonistic anti-mouse Fas IgG (clone Jo2), and anti-poly(ADP-ribose) polymerase IgG were obtained from Pharmingen. Anti-lamin A/C IgG (clone sc-7292) was from Santa Cruz Biotechnology (Santa Cruz, CA). Anti--tubulin IgG (clone B-5-1-2) was a product of Sigma.

Cell Lines-- HRPT cells (a gift from Dr. L. C. Racusen) were derived from human proximal tubules and have been extensively characterized, including demonstration of constitutive Fas expression (1, 2, 20). HRPT cells were maintained in DMEM/nutrient mixture F-12 (Invitrogen) supplemented with 10% fetal bovine serum (Hyclone Laboratories, Logan, UT), penicillin G (100 units/ml), and streptomycin sulfate (100 µg/ml) (both from Sigma). MCF-7 breast carcinoma cells (American Type Culture Collection, Manassas, VA) were maintained in minimal essential medium supplemented with 10% fetal bovine serum and 1% insulin/transferrin/selenium. Stable ₈ integrin transfectants were generated from HRPT cells, which were cultured to 50% confluence in 10-cm dishes and then transfected with 2 µg of human ₈ integrin cDNA subcloned into the pcDNAIneo vector (21) plus cationic liposomes (40 µl/dish; Superfect, QIAGEN Inc.) for 3 h in serum-free DMEM. Cells were cultured in complete medium containing G418 (Sigma). Individual G418-resistant clones were isolated, subcultured, and assayed for ₈ integrin expression by biotin surface labeling. Stable transfectants with persistent ₈ integrin overexpression were utilized after three to five passages.

cDNA Microarray-- Total RNA was extracted from HRPT cells by established methods (22); poly(A) RNA was isolated using Oligotex beads (QIAGEN Inc.); and 0.5 µg of poly(A) RNA was labeled with [-³³P]dCTP and incubated with chemokine and cytokine microarray filters (R&D Systems, Minneapolis, MN) according to the manufacturer's instructions. Individual hybridization bands were digitized by a PhosphorImager (Amersham Biosciences), quantified with ImageQuant Version 5 software (Amersham Biosciences), and normalized to -actin mRNA intensity from the same filter.

Northern Blot Analysis-- Methods have previously been described in detail (23). Poly(A) RNA was isolated as described above; and 2.0 µg of poly(A) RNA/lane was fractionated on a denaturing 1.0% agarose gel containing 0.67% formaldehyde, transferred to nylon membranes, and cross-linked by UV light exposure. To assess ₈ integrin mRNA levels, full-length human ₈ cDNA probes (24) were labeled with [-³²P]dCTP to a specific activity of 1.0 × 10⁸ cpm/µg of DNA (RTS Random Prime DNA labeling system, Invitrogen). Hybridization and high stringency washes were conducted according to previously described methods (23). Blots were stripped and rehybridized with a 300-nucleotide PCR product amplified from human -actin cDNA as a control for housekeeping gene expression.

Detection of Proteins Surface-labeled with ¹²⁵I-- Cells cultured in 10-cm dishes were washed with phosphate-buffered saline, lifted with trypsin/EDTA, washed again, and resuspended in 1 ml of phosphate-buffered saline. Cell surfaces were labeled by incubation with lactoperoxidase (200 µl, 1 mg/ml), Na¹²⁵I (3000 mCi/mmol, 0.5 mCi/10⁶ cells), and hydrogen peroxide (20 µl, 0.12%) for 5 min on ice with mild agitation every 30-60 s. Additional hydrogen peroxide (20 µl, 0.12%) was added, and the cell suspension was incubated on ice for 5 min. Cells were lysed with Triton X-100 buffer (25 mM Tris (pH 7.4), 50 mM NaCl, 25 mM NaF, 10% glycerol, and 1% Triton X-100). Lysates were precleared with rabbit serum, and _v and ₈ integrins surface-labeled with ¹²⁵I were immunoprecipitated from aliquots with equal radioactivity using 1 µg of monoclonal anti-_v IgG and 2 µl of anti-₈ antiserum, respectively, and resolved by 8% SDS-PAGE under nonreducing conditions according to previously described methods (25). Gels were dried and exposed to film overnight.

Detection of Proteins Surface-labeled with Biotin-- Cells were washed with ice-cold phosphate-buffered saline and surface-labeled with 1 mg/ml EZ-Link sulfosuccinimidyl 6-(biotinamido)hexanoate (Pierce). The labeling reaction was quenched with 0.1 M glycine, and cells were lysed with Triton X-100 buffer. Surface-biotinylated integrins were immunoprecipitated with specific anti-integrin antibodies, resolved by 6% SDS-PAGE under nonreducing conditions, transferred to PVDF membranes according to previously described methods (25), and probed with peroxidase-conjugated streptavidin (Pierce).

Apoptosis Assays-- RTC were rendered susceptible to Fas-dependent apoptosis by transfection and overexpression of mouse fas cDNA according to previously described methods (13). Transfected and untransfected control cells were then incubated with agonistic anti-mouse Fas IgG (clone Jo2; 5 µg/ml, 48 h, 37 °C). Whole cell lysates (20 µg of protein/lane) were resolved by SDS-PAGE and immunoblotted with mouse anti-human poly(ADP-ribose) polymerase IgG and peroxidase-conjugated goat anti-mouse IgG as described above. Apoptosis was defined by cleavage of poly(ADP-ribose) polymerase, which is a caspase-3 substrate.

Transcription and Translation Assays-- In initial experiments, RTC were pretreated with actinomycin D (0.5 µg/ml, 1 h, 37 °C) or cycloheximide (5 µg/ml, 1 h, 37 °C), followed by stimulation with agonistic anti-Fas antibodies (clone CH11; 150 ng/ml, 6 h, 37 °C) in the continued presence of actinomycin D or cycloheximide. Plasma membrane ₈ integrin protein expression was then determined by biotin surface labeling. For mRNA stability assays, RTC were stimulated with or without agonistic anti-Fas antibodies (clone CH11; 150 ng/ml, 6 h, 37 °C), and then actinomycin D (0.5 µg/ml) was added for up to 12 h. Steady-state ₈ integrin and -actin mRNA levels were determined by Northern blotting, digitized by a PhosphorImager, and quantified with ImageQuant Version 5 software. Normalized values were plotted on a logarithmic scale against time of actinomycin D incubation.

Migration Assay-- Haptotaxis migration assays were performed as previously described (26). Briefly, RTC were stimulated with agonistic anti-Fas antibodies (clone CH11; 150 ng/ml, 18 h, 37 °C) and then plated at a density of 1.2 × 10⁵ cells/well in the upper chamber of permeable supports (8.0-µm pore, Corning Costar, Corning, NY) precoated on the underside with vitronectin (100 ng/10 µl), which was purified from human plasma according to previously described methods (27). Nonspecific binding was blocked with bovine serum albumin (2%, 3 h, room temperature). In some experiments, function-blocking antibodies against _v3 or _v5 were added to the upper and lower chambers according to previously described methods (28, 29). Cells were fixed in paraformaldehyde (4%, 30 min, room temperature) and stained with crystal violet (0.5% in 20% methanol, 30 min, room temperature). Cells that did not migrate were gently removed from the upper chamber with a Q-tip. While blinded to the experimental condition, we viewed migrating cells at magnification ×40 and counted them. Mean values from six randomly selected fields per insert are reported. For filters containing cells, which were too dense to count, crystal violet was eluted with 1 ml of sodium acetate (100 mM in 50% ethanol at pH 5.2), and the absorbance was quantitated by spectrophotometry ( = 550 nm). Standard curves generated with serially diluted, crystal violet-stained cells yielded linear values up to A = 0.5.

Small Interfering RNA (siRNA) Transfection-- siRNA design and transfection protocol were conducted according to the methods of Elbashir et al. (30). The siRNA oligonucleotide sequence targeting ₈ integrin (AAACCAGGTACAAGGCATCTA) corresponded to nucleotides 1859-1879 in the coding region of the human cDNA sequence (24). An NCBI Protein Database BLASTn search revealed only the ₈ integrin cDNA as an exact match with the selected sense or antisense sequences. Lamin double-stranded RNA transfection, according to published methods (30), was used as a control for nonspecific effects of siRNA incorporation. Synthetic, annealed, double-stranded RNA constructs containing a 3'-dTdT overhang were purchased from Dharmacon Research, Inc. (Lafayette, CO). Oligonucleotide transfection was conducted according to the manufacturer's recommendations (Invitrogen). Briefly, HRPT cells were plated in 24-well dishes in antibiotic-free DMEM supplemented with 10% fetal bovine serum. At 24 h, cells at ~50% confluence were incubated with 150 µl of Opti-MEM (Invitrogen)/well. Oligonucleotides (3 µl, 20 µM stock) suspended in 50 µl of Opti-MEM were combined with OligofectAMINE reagent (Invitrogen) and maintained at room temperature for 25 min. The siRNA and OligofectAMINE mixture was supplemented with Opti-MEM and incubated with each well (4 h, 37 °C), followed by addition of DMEM plus 30% fetal bovine serum for 24 h. Wells were washed with phosphate-buffered saline and incubated for an additional 24 h in DMEM plus 10% fetal bovine serum. Transfected cells were analyzed for expression of ₈ integrin and lamin A/C and migration as described above.

Statistics-- Data are representative of two to four experiments for each condition. Histogram results are expressed as means ± S.E. Comparisons between groups were made by one-way analysis of variance with Bonferroni's test for multiple comparisons. Statistical significance is defined as p < 0.05.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Fas Activation Induces ₈ Integrin mRNA Expression-- We have previously demonstrated in experimental models that activation of up-regulated RTC Fas contributes to apoptosis and renal disease progression (1, 2). On the other hand, stimulation of constitutively expressed, low level RTC Fas transduces intracellular signals (13), but does not lead to apoptosis (2, 11), suggesting that Fas may regulate pathophysiologically relevant RTC functions in the absence of apoptosis.

To identify apoptosis-independent Fas pathways in RTC, cultured HRPT cells, which constitutively express Fas (1), were stimulated with either agonistic anti-Fas IgM or isotype control IgM (150 ng/ml, 6 h, 37 °C) under conditions that did not cause apoptosis, as determined by previously described poly(ADP-ribose) polymerase cleavage methods (13) (data not shown). mRNA expression patterns from both groups were screened by hybridization array using nylon filters spotted with 375 chemokine and cytokine cDNAs. This strategy was chosen because previous reports indicated that Fas stimulation induces cytokine and chemokine synthesis and secretion (15). The arrays revealed induction of four gene products in Fas-stimulated cells, and no genes were down-regulated by Fas activation (data not shown). Of the Fas-induced transcripts, ₈ integrin subunit mRNA expression was most markedly increased, with a mean 3-fold elevation.

The kinetics of Fas-dependent ₈ integrin gene expression were examined in RTC stimulated with agonistic anti-Fas IgM or isotype control IgM for durations ranging from 3 to 48 h, and mRNA expression was determined by Northern analysis. As shown in Fig. 1, steady-state ₈ integrin mRNA was detected in unstimulated cells at all time points, consistent with previous data demonstrating constitutive expression in kidney (24). Fas-stimulated ₈ integrin mRNA levels were slightly increased at 3 h, peaked between 6 and 18 h, and remained elevated at 48 h.

View larger version (27K): [in this window] [in a new window]   Fig. 1.   Fas activation induces 8 integrin mRNA expression. HRPT cells were stimulated with agonistic anti-Fas IgM (150 ng/ml, 37 °C) or isotype control IgM (150 ng/ml, 37 °C) for the indicated time periods. Poly(A) RNA (2 µg/lane) was fractionated by formaldehyde-containing 1% agarose gel electrophoresis and transferred to nylon membranes as described under "Materials and Methods." Blots were probed with 32P-labeled full-length human 8 integrin cDNA (upper panels) or a 300-nucleotide PCR product amplified from human -actin cDNA (lower panels). Results are representative of four different experiments.

Fas Activation Stimulates Surface _v8 Integrin Protein Expression-- Integrins are heterodimeric, single transmembrane-spanning receptors for extracellular matrix ligands, and the ₈ integrin subunit dimerizes exclusively with the _v subunit (24). To determine whether RTC Fas activation induces plasma membrane _v8 protein expression, cells were stimulated with agonistic anti-Fas antibodies or isotype control IgM, surface-labeled with ¹²⁵I, and immunoprecipitated with anti-_v integrin IgG or anti-₈ integrin antiserum. As shown in Fig. 2A, anti-_v immunoprecipitates contained 150- and 90-kDa bands, suggesting that _v8 is induced by Fas activation. Although the smaller band could represent other similarly sized _v partners such as ₃ and ₅ integrin subunits, immunoprecipitation with anti-₈ integrin antiserum yielded identical results (Fig. 2A), indicating that Fas stimulation up-regulates surface _v8 expression. Concordant increases in _v and ₈ integrins may result from increased association of endogenous _v with newly synthesized ₈, in agreement with previous observations for _v3 (31). However, the anti-_v immunoprecipitation data from Fig. 2A suggest that _v expression may actually be induced by Fas stimulation.

View larger version (36K): [in this window] [in a new window]   Fig. 2.   Fas activation stimulates surface v8 integrin protein expression. A, subconfluent HRPT cell monolayers were stimulated with agonistic anti-Fas IgM (150 ng/ml, 6 h, 37 °C) or isotype control IgM (150 ng/ml, 6 h, 37 °C) and then surfaced-labeled with 125I as described under "Materials and Methods." Cell lysates were precleared with preimmune serum, and aliquots with equal radioactivity were immunoprecipitated (IP) with anti-v IgG (first and second lanes) or anti-8 integrin antiserum (third and fourth lanes) and resolved by SDS-PAGE. Gels were dried and exposed to film as described under "Materials and Methods." Results are representative of three different experiments. B, subconfluent HRPT cell monolayers were stimulated with agonistic anti-Fas IgM (150 ng/ml, 37 °C) or isotype control IgM (150 ng/ml, 37 °C) for the indicated time periods and then surfaced-labeled with biotin as described under "Materials and Methods." Cell lysates were immunoprecipitated with anti-v integrin IgG (upper panels) or anti-8 integrin antiserum (lower panels), resolved by SDS-PAGE, transferred to PVDF membranes, and probed with peroxidase-conjugated streptavidin. Results are representative of four different experiments. C, subconfluent HRPT cell monolayers were stimulated with isotype control IgG (1 µg/ml, 18 h, 37 °C), agonistic anti-Fas IgG (clone DX2; 1 µg/ml, 18 h, 37 °C), isotype control IgM (150 ng/ml, 18 h, 37 °C), or agonistic anti-Fas IgM (clone CH11; 150 ng/ml, 18 h, 37 °C) and surfaced-labeled with biotin. Cell lysates were immunoprecipitated with anti-8 integrin antiserum, resolved by 6% SDS-PAGE, transferred to PVDF membranes, and probed with peroxidase-conjugated streptavidin. Results are representative of three different experiments.

To determine whether Fas-induced plasma membrane _v8 integrin expression is sustained, RTC were stimulated with agonistic anti-Fas IgM or isotype control IgM for 6-48 h, surface-labeled with biotin, and immunoprecipitated with anti-_v IgG and anti-₈ integrin antiserum. Fig. 2B demonstrates that Fas activation stimulated plasma membrane _v and ₈ integrin expression at 6 h, and Fas-mediated ₈ integrin expression was observed for up to 48 h, which parallels the Fas-induced ₈ integrin mRNA expression kinetics.

The specificity of ₈ integrin induction by Fas activation with the agonistic anti-Fas IgM stimulus was assessed by determination of cell-surface ₈ integrin expression in RTC incubated with a different agonistic antibody (clone DX2). Fas activation under these conditions did not stimulate apoptosis (data not shown). More importantly, DX2 exposure induced ₈ integrin expression (Fig. 2C), consistent with results generated with the anti-Fas IgM clone CH11. We conclude that ₈ integrin induction is not restricted to specific agonistic antibody stimuli, but is a general phenomenon of Fas activation.

Fas Induction of ₈ Integrin Expression Is Unique to Apoptosis-independent Conditions and RTC-- To determine whether Fas-regulated ₈ integrin induction is specific to apoptosis-independent conditions, ₈ integrin expression was determined in HRPT cells stimulated to undergo apoptosis following transient transfection with mouse fas cDNA and incubation with agonistic anti-mouse Fas IgG, as previously described (13). Fas stimulation under these conditions was associated with an increase in apoptosis as defined by poly(ADP-ribose) polymerase cleavage, but ₈ integrin expression was unchanged following apoptosis stimulation with agonistic anti-Fas antibodies (Fig. 3A). These data indicate that Fas induction of ₈ integrin expression is restricted to RTC with low basal Fas expression levels, which are relatively resistant to Fas-dependent apoptosis (2, 10, 11, 13).

View larger version (42K): [in this window] [in a new window]   Fig. 3.   Fas induction of 8 integrin expression is unique to apoptosis-independent conditions and RTC. A, HRPT cells were transiently transfected with a murine fas cDNA expression vector (mFas transfect.; 2 µg/well), followed by incubation with or without agonistic anti-mouse Fas IgG (clone Jo2; 5 µg/ml, 18 h, 37 °C). Untransfected, unstimulated cells served as a control. Cell lysates were immunoprecipitated with anti-8 integrin antiserum, resolved by 6% SDS-PAGE, transferred to PVDF membranes, and probed with peroxidase-conjugated streptavidin. Apoptosis was determined by immunoblot analysis of the same lysates for poly(ADP-ribose) polymerase (PARP) cleavage (depicted by the arrow). Results are representative of three separate experiments. B, MCF-7 cells were exposed to agonistic anti-Fas antibodies (clone CH11; 150 ng/ml, 6 h, 37 °C) or isotype control IgM (150 ng/ml, 6 h, 37 °C) and surfaced-labeled with biotin. Unstimulated, biotin-labeled HRPT cells were included as a control for relative 8 integrin expression levels. Cell lysates were immunoprecipitated with anti-8 integrin antiserum, resolved by 6% SDS-PAGE, transferred to PVDF membranes, and probed with peroxidase-conjugated streptavidin. Results are representative of two separate experiments.

To investigate whether Fas-dependent ₈ integrin induction is a generalized epithelial cell phenomenon, MCF-7 breast carcinoma cells were exposed to agonistic anti-Fas antibodies, and ₈ integrin expression was determined by immunoprecipitation of surface-biotinylated protein with anti-₈ integrin antiserum. MCF-7 ₈ integrin expression was detectable, but base-line levels were diminished compared with HRPT cells (Fig. 3B). Moreover, ₈ integrin expression was not regulated by Fas stimulation in MCF-7 cells, suggesting that ₈ integrin induction by Fas activation is specific to RTC.

Fas Induces ₈ Integrin Expression by an Enhanced mRNA Stabilization Mechanism-- To determine the mechanism of Fas-dependent ₈ integrin induction, surface ₈ integrin protein expression was measured in RTC co-incubated with agonistic anti-Fas antibodies and/or the mRNA transcription inhibitor actinomycin D or the protein translation inhibitor cycloheximide. Exposure of either actinomycin D or cycloheximide to anti-Fas antibodies for >12 h resulted in significant cell death (data not shown), in agreement with previous reports of enhanced Fas-dependent apoptosis in selected cell lines due to inhibition of endogenous anti-apoptotic proteins (32, 33). Fig. 4A confirms that ₈ integrin expression was induced in response to agonistic anti-Fas IgG or anti-Fas IgM. Actinomycin D co-incubation had no effect on ₈ integrin induction (Fig. 4B), suggesting that Fas-induced ₈ integrin expression is post-transcriptionally regulated. Cycloheximide completely inhibited Fas-dependent increases in surface ₈ subunit expression (Fig. 4C), indicating that Fas modulation of ₈ integrin expression requires new protein synthesis.

View larger version (28K): [in this window] [in a new window]   Fig. 4.   Fas induces 8 integrin expression by an enhanced mRNA stabilization mechanism. A, subconfluent HRPT cell monolayers were stimulated with agonistic anti-Fas IgG (clone DX2; 1 µg/ml, 6 h, 37 °C) or agonistic anti-Fas IgM (clone CH11; 150 ng/ml, 6 h, 37 °C) and then surfaced-labeled with biotin. Cell lysates were immunoprecipitated with anti-8 integrin antiserum, resolved by 6% SDS-PAGE, transferred to PVDF membranes, and probed for surface 8 integrin expression with peroxidase-conjugated streptavidin. Results are representative of two different experiments. B, subconfluent HRPT cells were pretreated with actinomycin D (0.5 µg/ml, 1 h, 37 °C), stimulated with agonistic anti-Fas IgG (clone DX2; 1 µg/ml, 6 h, 37 °C) or agonistic anti-Fas IgM (clone CH11; 150 ng/ml, 6 h, 37 °C) in the continued presence of actinomycin D, and then surfaced-labeled with biotin. Cell lysates were probed for surface 8 integrin expression as described for A. Results are representative of three different experiments. C, subconfluent HRPT cells were pretreated with cycloheximide (5 µg/ml, 1 h, 37 °C), stimulated with agonistic anti-Fas IgG (clone DX2; 1 µg/ml, 6 h, 37 °C) or agonistic anti-Fas IgM (clone CH11; 150 ng/ml, 6 h, 37 °C) in the continued presence of cycloheximide, and then surfaced-labeled with biotin. Cell lysates were probed for surface 8 integrin expression as described for A. Results are representative of three different experiments. D, HRPT cells were stimulated with or without agonistic anti-Fas IgM (150 ng/ml, 6 h, 37 °C). Actinomycin D (0.5 µg/ml, 37 °C) was then added for the indicated time periods. Steady-state 8 integrin and -actin mRNA levels were determined by Northern blotting as described under "Materials and Methods." Band intensities were quantitated by PhosphorImager analysis. Results represent mean of 8 integrin mRNA/-actin mRNA ratios (expressed on a logarithmic scale along the y axis) from two experiments. Data are normalized to mRNA ratios at the 0-h time point. *, p < 0.05 compared with anti-Fas IgM-treated cells at the same time point.

To determine the mechanism of post-transcriptional regulation, ₈ integrin mRNA stability was determined in Fas-stimulated versus unstimulated RTC. Fig. 4D shows steady ₈ integrin mRNA decay in the control group from 0 to 12 h, whereas ₈ integrin mRNA levels were relatively unchanged in Fas-stimulated cells over the same time period. These data demonstrate that Fas activation induces ₈ integrin expression by an enhanced mRNA stabilization mechanism.

Fas Activation Does Not Regulate Expression or Function of Other _v Integrin Heterodimers-- Before determining the functional consequences of _v8 induction by Fas (29), we first characterized the expression of other RTC _v integrins that could potentially mimic _v8 effects such as cell migration (21, 34). Basal and Fas-induced surface expression of _v1, _v3, _v5, and _v6 were determined by biotin surface labeling and immunoprecipitation. _v3 and _v5 were abundant under basal (no additions) and control (isotype control antibody exposure) conditions, consistent with previous reports in cortical RTC (35). In contrast to _v8, Fas activation did not induce expression of either _v3 or _v5 (Fig. 5A). _v1 and _v6 were undetectable in basal or Fas-stimulated cells (data not shown).

View larger version (49K): [in this window] [in a new window]   Fig. 5.   Fas activation does not regulate expression or function of other v integrin heterodimers. A, subconfluent HRPT cells were stimulated with agonistic anti-Fas IgM (150 ng/ml, 18 h, 37 °C) or isotype control IgM (150 ng/ml, 18 h, 37 °C) and surfaced-labeled with biotin. Lysates were immunoprecipitated with anti-v3 IgG (upper panel), anti-v5 IgG (middle panel), or anti-8 integrin antiserum (lower panel). Results are representative of four different experiments. B, HRPT cells were plated in the upper chamber of permeable supports at 1.2 × 105 cells/well. The underside of the support was precoated with 10 µg/ml vitronectin. Function-blocking anti-v3 IgG was added to the media in the upper and lower chambers at the indicated concentrations at time point 0. Cells migrating to the lower chamber were fixed after 12 h, stained with crystal violet, and counted. Results represent means ± S.E. from six fields viewed at magnification ×40 in two different experiments. Results are representative of three different experiments. C, the experiment was identical to that described for B, except that function-blocking anti-v5 IgG was added to the upper and lower chambers. *, p < 0.05 compared with control and isotype IgG groups. Results are representative of three different experiments. HPF, high power (×40) field.

To assess the functional significance of _v3 and _v5, RTC haptotaxis on vitronectin-coated permeable supports was quantitatively assayed in the presence of function-blocking anti-_v3 and anti-_v5 antibodies. Anti-_v3 IgG incubation had no effect on migration (Fig. 5B), even after exceeding concentrations that have been shown to prevent motility in cells that abundantly express _v3 (28). In contrast, haptotaxis was inhibited by ~50% with anti-_v5 IgG (Fig. 5C). The results demonstrate that basal RTC migration to vitronectin is partly mediated by _v5, with no contribution from _v3. However, these findings do not preclude other functions for RTC _v3 such as stable adhesion formation.

Fas Activation Stimulates RTC Migration by a ₈ Integrin-dependent Mechanism-- To test whether Fas-regulated ₈ integrin is functional, RTC were exposed to isotype control IgM or agonistic anti-Fas IgM, and cell migration to vitronectin matrix was assayed. Fas stimulation for 4-6 h was associated with 15-20% increases in the number of migrating cells (data not shown). Greater increases in migration were observed after 18 h (Fig. 6A). Migration was increased in proportion to surface ₈ integrin expression levels in Fas-stimulated and ₈ integrin-transfected RTC (Fig. 6, A and B), suggesting that Fas transduces a migratory phenotype through induction of _v8 integrin function.

View larger version (30K): [in this window] [in a new window]   Fig. 6.   Fas activation stimulates RTC migration by a 8 integrin-dependent mechanism. A, HRPT cells were pretreated with no additions (Control), isotype control IgM (150 ng/ml, 18 h, 37 °C), or agonistic anti-Fas IgM (150 ng/ml, 18 h, 37 °C) and then plated in the upper chamber of permeable supports at 1.2 × 105 cells/well. The underside of the support was precoated with vitronectin (10 µg/ml). HRPT cells stably transfected with human 8 integrin cDNA (8 transf.) were included as a positive control group. Cells migrating to the lower chamber were fixed after 18 h and stained with crystal violet, which was eluted and quantitated by spectrophotometry as described under "Materials and Methods." Results represent means ± S.E. of absorbance readings from four experiments. *, p < 0.05 compared with control (no additions) and isotype IgM groups. B, HRPT cells were treated as described for A with no additions, isotype control IgM (150 ng/ml, 18 h, 37 °C), or agonistic anti-Fas IgM (150 ng/ml, 18 h, 37 °C) or stably transfected with human 8 integrin cDNA and then surfaced-labeled with biotin. Cell lysates were immunoprecipitated with anti-8 integrin antiserum, resolved by 6% SDS-PAGE, transferred to PVDF membranes, and probed with peroxidase-conjugated streptavidin. Results are representative of four different experiments. C, HRPT cells treated with agonistic anti-Fas IgM (150 ng/ml, 18 h, 37 °C) or isotype control IgM (150 ng/ml, 18 h, 37 °C) were plated in the upper chamber of permeable supports at 1.2 × 105 cells/well. The support was precoated on the underside with vitronectin (10 µg/ml). The indicated groups were co-incubated with function-blocking anti-v5 IgG (1 µg/ml) in both the upper and lower chambers at the time of plating. Cells migrating to the lower chamber were fixed after 18 h, stained with crystal violet, and counted. Results represent means ± S.E. from six fields viewed at magnification ×40 in three different experiments. *, p < 0.05 compared with control and isotype IgM groups.

_v8 ligand specificity was examined by determining migration on an alternative extracellular matrix protein, fibronectin, which has been implicated as a ligand for _v8 in other systems (36). RTC migration to fibronectin was not observed in basal, Fas-stimulated, or ₈ integrin-overexpressing RTC (data not shown), indicating that fibronectin is not a ligand for RTC _v8. The data are consistent with previous affinity chromatography studies showing that epithelial cell _v8 does not bind fibronectin (21), as well as with the recent report by Mu et al. (37), who demonstrated that epithelial cell _v8 ligands are restricted to vitronectin and TGF-1.

Because _v5 was the only other _v integrin to affect RTC migration (Fig. 5), the specificity of Fas-induced _v8 for RTC migration to vitronectin was addressed by haptotaxis assays in the presence of inhibitory anti-_v5 antibodies. This strategy has previously been employed to assess ₈ integrin function in astrocytes, wherein residual migration after _v5 inhibition was attributed to _v8 (29). As shown in Fig. 6C, RTC Fas stimulation induced a significant increase in haptotaxis. Co-incubation with function-blocking anti-_v5 IgG resulted in diminution of base-line migration by 50%, in agreement with data from Fig. 5. However, the Fas-induced, 1-fold increase in haptotaxis was maintained in anti-_v5 IgG-treated cells, indicating that _v5 contributes to basal (but not Fas-dependent) migration and that Fas-dependent migration is due to functional up-regulation of _v8.

To more directly determine the functional effect of Fas-induced ₈ integrin expression, migration to vitronectin matrix was assessed in RTC with ₈ integrin expression down-regulation by RNA interference. Fig. 7A demonstrates the inhibition of basal and Fas-stimulated surface ₈ integrin protein expression in RTC transfected with a ₈ integrin-targeted siRNA construct. Fas-stimulated increases in RTC migration were completely blocked by ₈ integrin siRNA preincubation (Fig. 7B), whereas suppressed expression of an irrelevant gene (lamin A/C) by siRNA had no effect on basal or RTC Fas-dependent ₈ integrin expression or migration (data not shown). The results demonstrate that Fas activation induces ₈ integrin expression and function.

View larger version (40K): [in this window] [in a new window]   Fig. 7.   Fas-activated RTC migration is inhibited by 8 integrin RNA interference. A, HRPT cells were transfected in duplicate wells with OligofectAMINE in the absence of double-stranded RNA (Control) or with 8 integrin siRNA according to the protocol described under "Materials and Methods." After 24 h, the medium was changed, and agonistic anti-Fas IgM (150 ng/ml, 18 h, 37 °C) was added to one well for each condition. Cells were surface-labeled with biotin, lysed, and probed for 8 integrin expression by immunoprecipitation with anti-8 integrin antiserum (upper panel) or immunoblotted for -tubulin expression as a loading control (lower panel). Results are representative of four different experiments. B, HRPT cells were transfected in duplicate wells with OligofectAMINE in the absence of double-stranded RNA or with 8 integrin siRNA and incubated with anti-Fas IgM as described for A. Cells were then plated in the upper chamber of permeable supports at 1.2 × 105 cells/well. The support was precoated on the underside with vitronectin (10 µg/ml). Cells migrating to the lower chamber were fixed after 18 h and stained with crystal violet, which was eluted and quantitated by spectrophotometry as described under "Materials and Methods." Results are representative of four different experiments. *, p < 0.05 compared with all other groups.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Among receptor-mediated apoptosis pathways, signals transduced by Fas stimulation have been so extensively characterized that Fas is viewed as the prototypical death receptor in susceptible cells (3). However, in some cell types, Fas may mediate apoptosis-independent processes such as proliferation, angiogenesis, fibrosis, and inflammation (15, 16). Perhaps the most striking example of apoptosis-independent Fas function is derived from transgenic mouse and pancreatic islet cell transplantation studies in which -cells were genetically engineered to express Fas ligand in an effort to confer immune privilege through apoptotic deletion of invading, Fas-expressing T-cells (38-40). These studies surprisingly revealed extensive neutrophil infiltration and islet cell destruction, rather than preserved pancreatic morphology. In RTC, stimulation of overexpressed Fas transmits typical apoptosis signals (2, 13), whereas apoptosis-independent signals such as JNK activation are generated through constitutively expressed, low abundance Fas (13). Our study extends these observations by demonstrating that ₈ integrin induction is an additional pathway that is regulated by constitutively expressed RTC Fas.

Although Fas stimulation of ₈ integrin expression and function was unexpected, apoptosis and integrin pathways have previously been linked, albeit in an antagonistic fashion, whereby cells undergo apoptosis when integrins are no longer ligated to appropriate extracellular matrix or matrix-embedded growth factors. Frisch and Francis (17) described the process of anoikis, a specific type of apoptosis that is initiated upon integrin detachment from the extracellular matrix, which normally serves as a survival factor. More recently, Stupack et al. (18) demonstrated another form of integrin-related cell death in _v3-expressing cells maintained within a three-dimensional collagen gel, which is an inappropriate ligand for _v3. Under these conditions, caspase-8 docked at the cytoplasmic domain of the unligated ₃ integrin, resulting in cell death receptor-independent apoptosis. In contrast to these observations of detached or unligated integrins causing apoptosis, we now demonstrate, for the first time, that an integrin is up-regulated by death receptor activation. Furthermore, results from studies with apoptosis-sensitive HRPT and MCF-7 cells revealed that Fas-dependent ₈ integrin induction is restricted to apoptosis-independent conditions, thereby defining a new role for Fas in a context other than cell death.

Of the genes regulated by agonistic anti-Fas antibodies in our hybridization array experiments, ₈ integrin was selected for further study because it was induced by the greatest magnitude and because of potential pathophysiologic relevance since integrin-extracellular matrix interactions are pivotal in renal disease pathogenesis. In chronic renal diseases, increased RTC _v expression is associated with histologic damage and disease progression (41). Furthermore, mRNA expression of the _v ligand vitronectin is not present in normal kidney, but becomes detectable within the tubular basement membrane and interstitium in renal diseases (reviewed in Ref. 42). The tubular basement membrane usually acts as a barrier for RTC migration to the interstitium. However, regions of tubular basement membrane disruption have recently been identified in chronic renal disease biopsies (43, 44), suggesting that tubular atrophy, a strong predictor of renal disease progression, can be initiated by a mechanism involving RTC Fas-regulated induction of _v8 expression, which leads to directional RTC motility from a luminal location to the fibrotic, vitronectin-rich renal interstitium. Alternatively, up-regulated ₈ integrin-dependent RTC motility could represent an adaptive process by facilitating migration of regenerating, _v8-expressing RTC along denuded regions of intact, vitronectin-containing tubular basement membrane.

Recent studies have demonstrated that, in addition to serving as a stimulus for cell motility, _v8 also modulates cell growth and differentiation (37, 45). In the studies of lung epithelial cells by Cambier et al. (45), _v8 expression was associated with cell cycle withdrawal and inhibition of proliferation, which was not due to increased apoptosis. This same group subsequently showed that the negative cell growth regulation was mediated, at least in part, by _v8 binding of latent TGF-1 and metalloproteinase-dependent TGF-1 activation (37). These studies raise the intriguing possibility that RTC _v8 could act as a TGF-1 receptor and potentially play a role in the pathogenesis of interstitial fibrosis since renal fibrogenesis is regulated by TGF-1 (46).

After establishing that ₈ integrin was induced by Fas activation, we determined the mechanism to be enhanced mRNA stabilization, which is consistent with regulatory mechanisms for ₃ and ₅ integrin synthesis (35, 47). Although post-transcriptional mRNA regulation has not been as extensively investigated as transcriptional regulation mechanisms, Chen et al. (48) have described cis-elements within the 5'- and 3'-untranslated regions of the interleukin-2 gene that mediate mRNA stability, although specific binding sequences were not identified. These investigators subsequently identified two RNA-binding proteins that are targeted by the JNK signaling pathway to specifically bind to the 5'-untranslated region and the initial portion of the coding region and that confer a prolonged interleukin-2 mRNA half-life in activated T-cells (49). A similar mechanism for ₈ integrin mRNA regulation is plausible inasmuch as previous reports from our laboratory demonstrate Fas-dependent JNK activation in the absence of apoptosis (13). Alternatively, there are several domains within the 3'-untranslated region of the ₈ integrin mRNA that contain consensus AU-rich response elements, which have been associated with cytokine mRNA stabilization by JNK-independent signals (50).

Our data were generated from cells stimulated with agonistic anti-Fas antibodies; and although this has been a widely accepted strategy to achieve Fas-activated apoptosis in vitro and in vivo (51), agonistic anti-Fas IgG does not cluster Fas as effectively as transmembrane Fas ligand in some systems (4, 6, 52, 53), suggesting that apoptosis-independent pathways can be triggered by a less potent stimulus. This is unlikely in our system because ₈ integrin expression was induced by both agonistic anti-Fas IgG and anti-Fas IgM, and pentameric anti-Fas IgM should be sufficient to cluster Fas and to induce apoptosis in susceptible cells (6, 54). We speculate that in apoptosis-resistant cells, Fas signals transduced by agonistic antibodies may more closely mimic in vivo soluble Fas ligand-stimulated pathways because soluble Fas ligand has similarly been shown to bind Fas without stimulating apoptosis (52, 55). In support of this possibility, soluble Fas ligand and agonistic anti-Fas antibodies both induce apoptosis-independent JNK activation in RTC (13).² Another potentially confounding issue related to agonistic anti-Fas antibodies is whether ₈ integrin induction could be due to immunoglobulin receptor stimulation rather than to specific Fas activation. Because Fc receptors have not been identified in RTC, this is an unlikely explanation for our findings. Nevertheless, to address nonspecific effects of immunoglobulins on ₈ integrin induction, most studies included isotype IgG or IgM control groups. Because isotype immunoglobulin controls had no effect on ₈ integrin expression or function compared with unstimulated control cells, we conclude that ₈ integrin induction is specifically due to Fas activation.

In conclusion, we have defined a new role for Fas in the facilitation of _v8 integrin-dependent cell migration. These findings are unique inasmuch as synergism between cell death receptor and integrin pathways has not previously been described and suggest that additional apoptosis-independent cell phenotypes may be mediated by receptors traditionally relegated to cell death functions.

	FOOTNOTES

^* This work was supported by National Institutes of Health Grants DK54178, DK38558, DK57933, and CA96533 and United States Department of Defense Grant DAMD17-99-1-9019.The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Established Investigator of the American Heart Association. To whom correspondence should be addressed: Case Western Reserve University, Rammelkamp Center for Education and Research, MetroHealth Medical Center Campus, 2500 MetroHealth Dr., G531, Cleveland, OH 44109-1998. Tel.: 216-778-4993; Fax: 216-778-8248; E-mail: jrs15@po.cwru.edu.

Published, JBC Papers in Press, September 24, 2002, DOI 10.1074/jbc.M204901200

² G. Jarad, S. Khan, and J. R. Schelling, unpublished data.

	ABBREVIATIONS

The abbreviations used are: RTC, renal tubular epithelial cell(s); HRPT, human renal proximal tubule; JNK, c-Jun NH₂-terminal kinase; DMEM, Dulbecco's modified Eagle's medium; PVDF, polyvinylidene difluoride; siRNA, small interfering RNA; TGF-1, transforming growth factor-1.

	REFERENCES

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES