Overexpression of the transmembrane tyrosine phosphatase LAR activates the caspase pathway and induces apoptosis

Research Paper

247

Overexpression of the transmembrane tyrosine phosphatase LAR activates the caspase pathway and induces apoptosis Liang-Ping Weng, Junying Yuan* and Qiang Yu Background: The protein tyrosine phosphatase family comprises transmembrane receptor-like and cytosolic forms. Although the exact biological functions of these enzymes are largely unknown, they are believed to counter-balance the effects of protein tyrosine kinases. We have previously identified and characterized a mammalian transmembrane protein tyrosine phosphatase, called LAR (leukocyte common antigen related gene), whose expression is often associated with proliferating epithelial cells or epithelial progenitor cells. This study investigates the potential role of LAR in the regulation of cell growth and death in mammals. Results: We overexpressed in mammalian cells in culture either the full-length wild-type LAR or a truncation mutant containing only the extracellular domain of the molecule, and found that whereas the truncated LAR could be readily overexpressed in various cell lines, cells overexpressing the wild-type LAR were negatively selected. Using an inducible expression system, we demonstrated that overexpression of the wild-type LAR, but not the truncated LAR, activated the caspase pathway directly and induced p53-independent apoptosis.

Addresses: Pulmonary Center, Department of Medicine, and Department of Biochemistry, Boston University Medical Center, Boston, Massachusetts 02118, USA. *Department of Cell Biology, Harvard Medical School, Boston, Massachusetts 02115, USA. Correspondence: Qiang Yu E-mail: [email protected] Received: 7 November 1997 Revised: 29 December 1997 Accepted: 19 January 1998 Published: 10 February 1998 Current Biology 1998, 8:247–256 http://biomednet.com/elecref/0960982200800247 © Current Biology Ltd ISSN 0960-9822

Conclusion: Our data suggest that LAR might regulate cellular signals essential for cell survival. Overproduction of LAR may tilt the balance between the tyrosine phosphorylation and dephosphorylation of proteins whose activities are critical for cell survival, and therefore lead to cell death. In addition, our observations that overexpression of LAR induces cell death without affecting cell adhesion suggest that LAR may activate the caspase pathway and induce cell death directly. This work is the first example of the involvement of a receptor-like protein tyrosine phosphatase in cell-death control and provides the basis for searching for molecules and mechanisms linking signal transduction by protein tyrosine phosphorylation to the caspase-mediated cell-death pathway.

Background Signal transduction involving protein tyrosine phosphorylation cascades is a fundamental cellular mechanism regulating cell survival, growth, differentiation and migration. The tyrosine phosphorylation status of a protein is controlled by two antagonistic families of enzymes, protein tyrosine kinases, which add phosphate, and protein tyrosine phosphatases, which remove phosphate. The protein tyrosine kinases, which were originally discovered as biologically important molecules such as retroviral oncoproteins and growth factor receptors, have been studied extensively during the past two decades and their significance in cellular physiology is widely appreciated. The protein tyrosine phosphatases, on the other hand, were mostly identified based on their sequence homology to pre-existing tyrosine phosphatases that were defined by their enzymatic activities in vitro. In comparison with the tyrosine kinases therefore, the tyrosine phosphatases are much less well studied and their biological functions are poorly understood.

LAR is a member of a family of transmembrane, receptorlike protein tyrosine phosphatases or PTPs [1–4], originally identified by sequence homology to the leukocyte common antigen CD45 [1]. LAR expression is detected in a wide variety of tissues and is regulated during development [5]. In cultured cells, the levels of LAR mRNA and protein are influenced by growth factors and cell density [6,7]. Structural and biochemical characterization of LAR suggest that it might participate in transmembrane signaling involving protein tyrosine phosphorylation by counterbalancing the activity of the protein tyrosine kinases [1,2]. Functional studies of LAR in cultured cells suggest that it might have a role in regulating signal transduction mediated by receptor tyrosine kinases [8–12]. Attempts to find physiological substrates of LAR using interaction-trap strategies identified two proteins that physically interact with LAR: one colocalized with LAR at focal adhesions of human breast adenocarcinoma cells [13] and the other contained guanine nucleotide exchange factor domains of the Rac/Rho family of GTPases [14], suggesting that LAR

248

Current Biology, Vol 8 No 5

may also function in regulating cell adhesion, migration and/or signaling mediated by the focal adhesion complex. There is also evidence that the Drosophila homolog of LAR, DLAR, is selectively expressed in central nervous system and is required for motor axon guidance during embryonic development, presumably by regulating cell adhesion [15,16]. Taken together, these observations suggest that LAR might be a multifunctional protein and have roles in diverse aspects of cell physiology involving regulation of protein tyrosine phosphorylation. Programed cell death or apoptosis is a fundamental part of tissue homeostasis during normal development and in pathological conditions [17]. The cell death machinery is regulated by both internal genetic programs and external physiological stimuli. Studies of the molecular pathways, mechanisms and regulation of the cell death process have revealed three gene families that are the central components of apoptotic pathways: the tumor necrosis factor (TNF) family and their receptors [18], the interleukin-1 βconverting enzyme (ICE) or caspase family of cysteine proteases [19] and the proto-oncogenes of the bcl-2 gene family [20]. At the same time, a large number of molecules, which normally function in directing cell proliferation and differentiation, have been found to be involved in regulating apoptotic pathways [21]. Protein tyrosine phosphorylation and tyrosine kinases have been implicated as playing critical roles in regulating cell survival. For example, in addition to its ability to promote cell proliferation and differentiation, the basic fibroblast growth factor (bFGF), by activating tyrosine kinase activity of its receptor, is a potent suppressor of programed cell death during development of the limb bud [22], lens fibre cells in the mouse [23], and during TNFαinduced apoptosis [24]. Insulin and insulin-like growth factor (IGF) are effective inhibitors of apoptosis induced by trophic factor removal [25] and overexpression of c-Myc [26] or ICE [27]. Activation of survival pathways that involve insulin and IGF has been hypothesized to be mediated by tyrosine phosphorylation activity of the insulin and IGF receptors and by phosphatidylinositol 3kinase [28,29]. These results suggest that tyrosine

phosphorylation of certain proteins is critical for cell survival. The involvement of PTPs in cell-death control has been suggested by a number of experiments using tyrosine phosphatase inhibitors [30,31], in which apoptosis was found to be inhibited following inactivation of tyrosine phosphatases. The exact roles of tyrosine phosphatases in these experiments and in cell-death control in general, are however not clear. To define the biological functions of LAR and to investigate the potential role of PTPs in regulating cell growth, death and differentiation, we overexpressed LAR in various mammalian cell lines. We found that overexpression of LAR in many types of cells, especially nontransformed cells, was cytotoxic. Using an inducible expression system, we demonstrated that overexpression of LAR activated the caspase pathway and induced apoptosis. Our results suggest that, in mammalian cells, LAR and LARlike PTPs might play a role in regulating tyrosine phosphorylation of a set of proteins which are critical for cell growth and survival. These proteins are important because they might function to link signal transduction pathways involving protein phosphorylation cascades with the caspase-mediated cell-death pathway. Our study provides an assay system to search for these proteins. This is the first demonstration of the involvement of a receptor-like PTPs in the regulaton of cell death.

Results Cells overexpressing wild-type LAR are negatively selected

To investigate possible roles of LAR in signal transduction and regulation of cell growth, we attempted to alter LAR activity by overexpressing wild-type and mutant LAR in cultured cells. Two expression constructs were generated, one encoding the full-length wild-type LAR, and the other encoding only the extracellular domain of LAR (Figure 1). The two constructs were tested in COS-7 cells by transient transfection followed by immunoprecipitation assays. Both proteins were expressed at comparable levels in COS-7 cells (Figure 2). Cells transfected with the wild-type LAR cDNA expressed three protein species, a 190 kDa precursor, a 145 kDa proteolytically cleaved extracellular domain subunit and a 85 kDa intracellular

Figure 1 Immunoglobulinlike domains

Fibronectin type III repeats

LAR

ss

ss

ss

LAR-∆TP

ss

ss

ss

Transmembrane domain

Phosphatase domains

Current Biology

The full-length wild-type LAR and the truncated LAR mutant, LAR-∆TP, which contains only the extracellular domain of LAR. The full-length LAR has an extracellular portion composed of immunoglobulin-like domains and fibronectin type III repeats and an intracellular portion containing two tandemly repeated phosphatase domains. Notice that the LAR protein is cleaved into two subunits which remain associated by a non-covalent linkage [2].

Research Paper Tyrosine phosphatase LAR and apoptosis Weng et al.

1

LAR-∆TP

LAR

Figure 3

Control

Figure 2

249

2

3

4

5

6

7

8

9 10 11

LAR

145

LAR-∆TP

145

kDa

200–

Current Biology

190

145

97–

185

68– Current Biology

Transient expression of LAR in COS-7 cells. COS-7 cells were transiently transfected either with a construct that encodes the wild-type LAR or one that encodes the truncated LAR-∆TP. Expression of LAR protein was analyzed by immunoprecipitation of [35S]methionine-labeled proteins from the transfected COS-7 cells using anti-LAR antiserum α294. The control is an immunoprecipitate from untransfected COS-7 cells. The 190 kDa precursor of LAR is indicated, as are the 145 kDa extracellular and 85 kDa phosphatase-containing subunits, which become dissociated under the denaturing conditions of SDS–PAGE. The band below the 85 kDa subunit is a degradation product of LAR.

phosphatase-domain-containing subunit [2]. Cells transfected with the truncated LAR cDNA, on the other hand, expressed only the 145 kDa subunit (Figure 2). We next tried to establish stable cell clones that overproduce LAR. We introduced the two cDNA expression constructs by transfection into a rat embryonic fibroblast cell line 208F, from which the rat LAR cDNA was previously isolated and in which the LAR protein was characterized [2]. Although we could obtain drug-resistant colonies from

Western blot analysis of LAR expression in stable transfectants. Rat fibroblast 208F cells were transfected with the wild-type LAR or the truncated LAR-∆TP. After selection, stably transfected clones were isolated. Equal amounts of total protein from the isolated clones were analyzed by western blotting using anti-LAR antiserum αN-131, which recognizes only the 145 kDa extracellular subunit of LAR (arrowheads). Protein samples are numbered according to the isolated clones. Notice that, in lanes 10 and 11 for the wild-type LAR transfection, two aberrant LAR proteins were expressed.

both transfections, the number of colonies obtained from the cells transfected with wild-type LAR was consistently less than the number obtained with the truncated LAR (data not shown). Furthermore, most of the clones that were transfected with wild-type LAR showed either no increase or less than onefold increase in LAR protein level when analyzed by western blotting or by immunoprecipitation (Figure 3 and data not shown). The few LAR-expressing clones, obtained from transfections with the wild-type LAR, expressed aberrant forms of the protein (Figure 3 and data not shown). In contrast, more than 50% of the clones transfected with the truncated LAR cDNA expressed 5–10-fold higher levels of the mutant LAR protein over those of the endogenous LAR protein (Figure 3 and data not shown). It appeared that overexpression of wild-type LAR in rat fibroblasts had negative effects on cell growth, resulting in negative selection against cells overexpressing the LAR protein. The only cells which could grow after transfection were the cells which had no increase or very small increases in LAR expression or expressed aberrant forms of the LAR protein. Overexpression of truncated LAR did not affect cell morphology, growth or protein tyrosine phosphorylation (data not shown). Overexpression of LAR induces cell death

To understand the reason for the failure to obtain stable transfectants that overproduce wild-type LAR, we transiently co-transfected the 208F cells with an indicator plasmid encoding the Escherichia coli β-galactosidase protein, together with either of the two LAR expression plasmids. As controls, we co-transfected the cells with the vector plasmid or with a plasmid expressing ICE, which is known to induce apoptosis when overexpressed in rat fibroblast cells [32]. The number and the morphology of the transfected cells were analyzed 24 hours after transfection. As shown in

250

Current Biology, Vol 8 No 5

Figure 4

tetracycline-suppressible promoter [33]. Our initial efforts to use the fibroblast cell line 208F for the tetracycline expression system failed because of the leaky activity of the tetracycline-suppressible promoter in this cell line. We then selected an osteosarcoma cell line U2OS, with which the Tet-suppressible system has been successfully used (L Zhu, personal communication).

(a)

Vector

Proportion of cells that are rounded (%)

(b)

Constructs expressing either wild-type or truncated LAR were transfected into U2OS cells and stable cell clones, which could be induced to express the two proteins, were established. In general, comparable amounts of LAR protein were induced in the two sets of transfectants. Nevertheless, there were 5–20-fold variations in LAR protein levels between clones compared with the level of LAR in uninduced cells (Figure 5a). The LAR protein was detectable after six hours of induction and reached its maximum level 12–24 hours following induction, depending on the clone (data not shown).

LAR

30

20

10

0 Vector

LAR-∆TP

LAR

ICE Current Biology

Cell rounding induced by transient transfection of LAR. Rat fibroblast 208F cells were transiently co-transfected with a plasmid encoding the β-galactosidase reporter and constructs containing no cDNA insert (Vector), or cDNA inserts encoding the wild-type LAR, the truncated LAR-∆TP, or ICE. The ratio of the reporter plasmid to the cotransfected second plasmid was 1:5. Cells were fixed and stained with X-gal 24 h after transfection. Transfected cells stain blue with X-gal. (a) Examples of round cells (blue) resulting from LAR transfection. Cells transfected with vector or wild-type LAR are as indicated. (b) Proportion of cells that are round resulting from each transfection. About 700 blue cells from each transfection were counted and the number of rounded blue cells was expressed as a percentage of the total number of blue cells.

When cultured in the presence of tetracycline, which suppresses LAR expression, the parental U2OS cells and the LAR tansfectants appeared as a flat monolayer of cells (Figure 5b and data not shown). Rounded, floating cells began to appear in the cultures transfected with wild-type LAR, 12–24 hours after the induction of LAR expresion. The floating cells stained positive with propidium iodide, indicating that they were dead (Figure 5d,e). In the majority of clones expressing wild-type LAR protein, more than 50% of the cells in each culture died after LAR overexpression for 48 hours (Figure 5c,f). In contrast, there were no apparent morphological changes nor did cell death occur in the cultures transfected with truncated LAR under the same conditions (data not shown). The percentage of cell death varied between clones but did not correlate entirely with the level of LAR protein (Figure 5a,c), indicating that cell death was not solely caused by the high level of LAR protein that was expressed in the cells. LAR-induced cell death is p53-independent but requires caspase activity

Figure 4, about 30% of the β-galactosidase-positive cells (blue) transfected with the wild-type LAR appeared as rounded dying cells (Figure 4a,b). In contrast, less than 10% of the blue cells co-transfected with the truncated LAR, or with the vector plasmid, appeared as rounded cells. As was seen with the LAR-transfected cells, about 30% of the blue cells transfected with ICE appeared as rounded cells (Figure 4b). The rounded cells obtained following transfection with LAR or ICE were indistinguishable on the basis of morphology, suggesting that both were undergoing apoptosis (Figure 4a and data not shown). To verify that overexpression of LAR indeed induced cell death, we expressed LAR under the control of the

Cells that were induced to die by LAR overexpression had the characteristic morphology of cells undergoing apoptosis, including membrane blebbing, cell volume loss, nuclear condensation and cell fragmentation (Figure 6a and data not shown). To confirm the occurrence of DNA fragmentation, chromosomal DNA was isolated from induced and uninduced cells and visualized by agarose gel electrophoresis. The characteristic internucleosomal DNA fragmentation was apparent only in LAR-induced cells (Figure 6d). DNA fragmentation was further confirmed using the terminal-deoxynucleotidylmediated dUTP nick end-labeling (TUNEL) assay. About half of the rounded cells in the LAR-induced culture were TUNEL-positive (Figure 6b). In contrast,

Research Paper Tyrosine phosphatase LAR and apoptosis Weng et al.

no TUNEL-positive cells were found in the uninduced control culture (Figure 6c).

Tet

– + – + – + –

+ – +

U2OS/LAR-∆TP3

U2OS/LAR3

U2OS/LAR2

U2OS/LAR1

U2OS

(a)

U2OS/LAR-∆TP2

U2OS/LAR-∆TP1

To understand the molecular basis of the LAR-induced cell death, we investigated the involvement of several key mol-

–+

– 190 145

(b)

(c)

(d)

(e)

251

ecules that have been implicated as playing key roles in apoptotic processes. The requirement for functional p53 in LAR-induced cell death was investigated by introducing the LAR expression construct into another osteosarcoma cell line, Saos2, in which the p53 gene was deleted. Although the LAR protein level induced in the Saos2 cells was not as high as in the U2OS cells (data not shown), LAR induced similar apoptotic cell death in these p53-deficient cells (Figure 7), indicating that p53 is not required for LARinduced apoptosis. To investigate the involvement of caspases in LAR-induced death, we induced LAR expression in the presence of caspase inhibitors. LAR-induced cell death was inhibited by ZVAD, a modified peptide which inhibits several caspases [34], in a dose-dependent manner (Figure 8 and data not shown). At a concentration of 50 µM, ZVAD completely inhibited cell death as well as cell detachment (Figure 8a,b), suggesting that cell detachment may be a consequence of caspase activation. ZVAD had no apparent effect on uninduced cells (Figure 8a). We next determined which member of the caspase family was required for the LARinduced cell death using specific caspase inhibitors. Cell death was partially inhibited by the caspase-3 inhibitor ZDEVD but not by the ICE inhibitor ZYVAD (Figure 8b). Because ZVAD is more effective than ZDEVD, our data suggest that LAR may induce apoptosis partly through activation of caspase-3, but other caspases may also be involved. Caspase-3, but not ICE, is activated during LAR-induced cell death

To confirm the above suggestion, we examined the proteolytic activation of ICE and caspase-3 [35,36] upon (f)

150

100

50

U2 O S/ LAR3

U2 O S/ LAR2

U2 O S/ LAR1

U2 OS /L AR-∆ T P3

U2 OS /L AR-∆ T P2

U2 OS /L AR-∆ T P1

0 U2 OS

Vi ab il it y ( %)

Figure 5

Current Biology

Cell death induced by overexpression of LAR under the control of the tetracycline-suppressible promoter. (a) Induction of LAR overexpression. Clones of U2OS cells containing the tetracyclinesuppressible LAR cDNA expression constructs were induced to express LAR by removing tetracycline from the medium. Cells, which were stably transfected with the wild-type LAR construct (U2OS/LAR) or the truncated LAR construct (U2OS/LAR-∆TP), were cultured in the presence (Tet+) or absence (Tet–) of tetracycline for 24 h. Three independent clones of cells were used for each LAR construct. Equal amounts of total protein from these cells were analyzed by western blotting using anti-LAR antiserum αN-130, which recognizes the 190 kDa precursor and the 145 kDa extracellular subunit of LAR. Parental U2OS cells was used as a control. (b) Morphology of U2OS/LAR cells in the absence of induction. (c–e) Massive cell death occurred 24–48 h after induction of LAR overexpression. The dead cells are shown under higher magnification in (d,e) and stained with propidium iodide in (e). (f) Cell survival after LAR induction. Parental U2OS cells and U2OS cells stably transfected with truncated LAR (U2OS/LAR-∆TP) or wild-type LAR (U2OS/LAR) were seeded at equal densities and cultured with or without tetracycline. Viable cells after culture for 48 h were counted and the ratio of the number of viable cells in Tet– cultures to the number in Tet+ cultures was calculated and is presented in the graph as the percentage of cells surviving. The data are mean values from three separate experiments.

252

Current Biology, Vol 8 No 5

Figure 6 Characterization of LAR-induced cell death. (a) Morphology of LAR-overexpressing cells is typical of apoptotic cells with membrane blebbing and fragmented cell bodies. (b) Result of TUNEL assay on induced, LARoverexpressing cells. Apoptotic cells appear yellow because of double staining by fluorescein (green) and propidium iodide (red). (c) Result of TUNEL assay on control uninduced cells, which show only propidium iodide staining. (d) Agarose gel electrophoresis of DNA from control uninduced cells (lane 2) or from the induced, LARoverexpressing dead cells (lane 3). Evidence for DNA fragmentation characteristic of apoptosis is the laddering of the DNA sample from LAR-overexpressing cells (lane 3). The 1 kb DNA molecular weight ladder is in lane 1.

overexpression with LAR. We also investigated whether poly(ADP-ribose) polymerase (PARP), a known substrate of caspases was cleaved [36]. Cells were collected at different time points after induction of LAR overexpression. Cell lysates were analyzed by western blot analysis using specific antibodies against ICE, caspase-3 and PARP. Both caspase-3 and PARP were cleaved in the LAR-overexpressing cells but not in the control cells (Figure 9). The level of ICE protein, on the other hand, was not affected by LAR overexpression (Figure 9), indicating that caspase-3, but not ICE, was activated upon LAR overexpression and might be the protease responsible for LAR-induced apoptosis.

also tested whether the parental U2OS cells could undergo anoikis by plating the cells onto poly(2-hydroxyethyl methacrylate), which prevents cell attachment [38]. The cells were stained with propidium iodide after incubation for 24 hours. No significant cell death was detected during this period (data not shown). In contrast, LAR-induced cell death occurred less than 12 hours after Figure 7

100

Overexpression of LAR does not affect cell adhesion

75

% Viability

Activation of caspases has been shown to occur during anoikis, a form of apoptosis caused by detachment of cells from their substratum [37]. The structure of the LAR extracellular domain suggests that LAR might interact with extracellular matrix proteins and it is therefore possible that overexpression of LAR might interfere with cell adhesion. LAR-induced cell death could have been caused by disrupted cell adhesion and the cells could have died through anoikis. Nevertheless, the observation that both cell death and cell detachment were inhibited by caspase inhibitors (Figure 8) suggested that LARinduced cell death might not be a consequence of disrupted cell adhesion. We therefore investigated effects of LAR overexpression on cell adhesion. Uninduced cells and cells induced to express LAR for 8–24 hours were plated onto microtiter plates coated with various concentrations of fibronectin. Adherent cells were examined after incubation for 30 minutes. No obvious changes in cell adhesion to fibronectin were observed following induction of LAR overexpression (Figure 10 and data not shown). Similar results were obtained when laminin or collagen were used as substrates (data not shown). We

50

25

0 Saos2

Saos2/LAR Current Biology

LAR-induced cell death in cells that are defective for the p53 gene. Parental Saos2 osteosacoma cells and cells stably transfected with the wild-type LAR cDNA under the control of the tetracycline-regulated promoter (Saos2/LAR) were each seeded equally into pairs of cultures with or without tetracycline. After 48 h, viable cells from all cultures were counted and the ratio of viable cells from the Tet– and the Tet+ cultures was calculated for each cell line and is presented as the percentage of cells surviving. The data are presented as mean values from three independently isolated transfectants.

Research Paper Tyrosine phosphatase LAR and apoptosis Weng et al.

Figure 8

(a)

253

Figure 9

Tet

+

+

–

–

ZVAD

–

+

–

+

Time Tet +

6h

9h –

+

12 h –

+

D –

–

Caspase-3

ICE

(b)

125

PARP

100 Viability (% )

Tubulin 75

Current Biology

50

25

0 Control

ZVAD

ZDEVD

ZYVAD

Activation of caspase-3 by LAR induction. The U2OS/LAR stable transfectants were cultured in the presence or absence of tetracycline for the indicated times. Both adherent and floating cells were harvested and equal amounts of total cellular protein were resolved by SDS–PAGE followed by western blot analysis. The blot was sequentially probed with different antibodies as indicated. Notice the time-dependent reduction of caspase-3 and the appearance of the PARP cleavage product (arrowhead) in the Tet– (LAR+) culture. D, protein isolated from floating dead cells. Anti-tubulin antibody was used as a control to show equal loading of protein between lanes.

Current Biology

Inhibition of LAR-induced cell death by caspase inhibitors. The U2OS/LAR stable transfectant cells were cultured in the presence (Tet+) or the absence (Tet–) of tetracycline and with or without the general caspase inhibitor ZVAD (50 µM), caspase-3 inhibitor ZDEVD (250 µM), or ICE inhibitor ZYVAD (250 µM) respectively for 12 h. (a) Morphology of LAR+ (Tet–) and LAR– (Tet+) cells in the presence or absence of ZVAD. (b) Proportion of cells that are viable after induction of LAR overexpression and treatment with the caspase inhibitors. Viability was calculated as the ratio (%) of the number of viable cells in Tet– cultures to the number in Tet+ cultures. The data are mean values from three separate experiments.

induction of LAR overexpression, suggesting that LAR may not induce cell death through anoikis.

Discussion We have demonstrated here that overexpression of LAR activates caspase-3 and induces apoptosis, raising the intriguing possibility that LAR might play a role in regulating growth/survival signals and controlling cell death/survival in vivo. The ability of LAR to induce apoptosis requires its phosphatase domain — a truncated version of LAR lacking the catalytic domain does not induce apoptosis and the overexpression of this mutant is well-tolerated by the cells, suggesting that overproduction of the LAR protein per se is not sufficient to cause cell death. We believe that dephosphorylation of certain tyrosine phosphorylated proteins by increased LAR

phosphatase activity is the key to caspase activation and induction of cell death. Although the direct substrates of LAR have yet to be identified, we present evidence to suggest that LAR may activate the caspases directly to induce apoptosis. How might overexpression or activation of LAR induce apoptosis? There are at least three possible ways. The first possible mechanism comes from the finding that tyrosine phosphorylation is critical for signal transduction of many important growth factors, such as the platelet-derived growth factor (PDGF), the epidermal growth factor (EGF), the fibroblast growth factor (FGF) and IGF. It has been reported that LAR might directly dephosphorylate the insulin receptor [11] and that LAR modulates signal transduction from a number of growth factor receptor tyrosine kinases [9]. We observed that LAR-induced cell death can be partially inhibited by basic FGF (our unpublished observations), suggesting at least that activation of some receptor tyrosine kinases can overcome the cell death caused by overexpression or activation of the LAR tyrosine phosphatase. Therefore, LAR might dephosphorylate one or several of the growth factor receptors or their downstream signaling components and thus block growth and/or survival signals, which might lead to activation of caspase and induction of cell death [39]. Second, cell adhesion and signal transduction mediated by cell-adhesion molecules are also critical for cell survival

254

Current Biology, Vol 8 No 5

Figure 10

Absorbance at 562 nm

1.0 0.8 0.6 0.4

LAR+

Orthovanadate has also been found to block DNA fragmentation, a hallmark of apoptosis [48], and is as effective as basic FGF in preventing the apoptosis of endothelial cells that results from treatment with TNFα [49]. Thus, phosphorylation of certain proteins on tyrosine residues is likely to be important for cell survival. We hypothesize that LAR overexpression leads to dephosphorylation of protein(s) that are critical for cell survival resulting in cell death and cell detachment.

LAR– 0.2 0 0 5 10 15 20 Concentration of fibronectin coated onto plates (µg/ml) Current Biology

Effects of LAR overexpression on cell adhesion. Microtiter plates were coated with fibronectin at the indicated concentrations. Cells were precultured in normal tissue culture plates for 8 h with or without tetracycline to induce LAR expression. Both induced (LAR+) and uninduced (LAR–) cells were then trypsinized, and each sample was plated in triplicate into three wells of the microtiter plate per concentration of fibronectin, and then incubated for 30 min at 37oC as described in the Materials and methods. Nonadherent cells were removed. The remaining adherent cells were fixed, stained with crystal violet and quantified by measuring the optical density at 562 nm. Representative mean data are shown for three wells from one of three separate experiments.

Our data also suggest that LAR-induced apoptosis does not require functional p53. In apoptosis that is induced by DNA damage, p53 plays a critical role [50] but in other cases of apoptosis induced, for example, by glucocorticoids or phorbol ester, p53 is not required, indicating that both p53-dependent and p53-independent apoptotic pathways exist [51,52]. Our results provide another example of apoptosis that is p53-independent and not induced by DNA damage. We have also analyzed possible effects of LAR on cell-cycle progression prior to cell death and found no evidence indicating that LAR overexpression causes cells to arrest at a particular phase of the cell cycle (our unpublished observations). Thus, LAR-induced cell death might not be a consequence of deregulated cellcycle progression.

Conclusions [40–43]. Mutation of DLAR results in defects in axonal guidance, suggesting that DLAR is involved in cell–cell or cell–matrix signaling and/or adhesion [16]. Serra-Pages et al. [13] reported that the LAR protein is localized at focal adhesions. It is thus possible that the LAR phosphatase has a role in regulating cell adhesion and/or signal transduction mediated by cell-adhesion molecules. Because disruption of cell adhesion during anoikis causes activation of caspases [37,44,45], cell death could occur as a result. Our data, however, show that LAR overexpression has no effect on cell adhesion and that LAR-induced cell death and cell detachment are inhibited by caspase inhibitors. It is therefore unlikely that LAR induces cell death by interfering with cell adhesion. Nevertheless, our data does not exclude the possibility that LAR affects signal transduction mediated by cell-adhesion molecules. This might limit growth signals or create conflicting signals through the focal adhesion complexes, with apoptosis as the consequence. Last but not least, LAR might activate caspases and cause cell to die through a novel mechanism. Yang et al. [46] have reported that loss of substrate attachment and apoptosis of endothelial cells following removal of growth factors are associated with dephosphorylation of tyrosine residues at the cell periphery. The dephosphorylation of total cellular proteins that accompanies apoptosis can be blocked by orthovanadate, an inhibitor of PTPs [46,47].

The present study provides the first evidence of the involvement of a receptor-like PTP in regulating cell death. Our results show that, when overexpressed, LAR can activate the caspase pathway directly, without affecting cell adhesion, and can induce p53-independent apoptosis. We believe that overexpression of LAR increases the total activity of LAR, resulting in dephosphorylation of a set of proteins which are crucial for cell survival. This study provides a foundation for identifying these important molecules and to link protein tyrosine phosphorylation with the caspase protease pathway.

Materials and methods Plasmid construction and transfection Two mammalian cell expression vectors, pBabe [53], which expresses inserted genes from the constitutive moloney murine leukemia virus long terminal repeat (LTR) promoter, and pUHD10-3 [33], which contains a tetracycline-suppressible promoter, were used to generate the two sets of LAR expression constructs. The wild-type LAR cDNA expression constructs were generated by ligating a 6.3 kb SalI–BamHI fragment, which contains the 240 bp 5′ untranslated region (UTR), the entire 5661 bp protein-coding region, and the 450 bp 3′ UTR of the rat LAR cDNA [2], into the two expression vectors. LAR-∆TP constructs were generated by ligating a 3.8 kb SalI–SacI LAR cDNA fragment, which contains the same 5′ UTR followed by 3524 bp of protein coding sequences, into the same expression vectors. Cell transfections were done using the standard calcium phosphate precipitation method [54]. For transient transfections of COS-7 cells, 5 µg of LAR expression plasmid DNAs were used in each transfection assay. Cell lysates were harvested 24 h after transfection. For transient transfection assays of the 208F cells, exponentially growing cells were co-transfected with 1 µg β-galactosidase expression plasmid and 5 µg LAR or

Research Paper Tyrosine phosphatase LAR and apoptosis Weng et al.

255

ICE cDNA expression plasmids. The cells were fixed for β-galactosidase staining 24 h after transfection. To generate stable 208F, U2OS, and Saos2 transfectants, the cells were transfected with 2.5 µg plasmid DNA per 5 × 104 cells on a 35 mm plate. The cells were diluted into 100 mm plates 12 h after transfection and were selected in the presence of puromycin (Sigma) at 1 µg/ml for 1 week. The individual clones were then isolated and propagated. The U2OS and Saos2 transfectant cell clones were maintained in 10% FBS/DMEM with 250 µg/ml Geneticin (GIBCO), 0.2 µg/ml puromycin (Sigma), and 1 µg/ml tetracycline.

Rabbit polyclonal antisera αC-292 and αN-130 are anti-LAR antisera raised against synthetic peptides of the C-terminus and the N-terminus of the LAR protein respectively [2]. and were used at 1:500 dilution. Rabbit polyclonal antisera α−294 was raised against a E. coli-produced full length LAR protein in a similar way as described in [2] and were used at 1:500 dilution. Mouse monoclonal anti-ICE, anti-PARP, and anti-caspase-3 were used at 1:2000 dilution. Mouse monoclonal antibody α-tubulin was purchased from Sigma and used at 1:10000 dilution.

Cells cultures

Adhesion assays

The rat embryonic fibroblast cell line 208F [55] and the COS-7 cells were grown in 10% CF (GIBCO)/DMEM (GIBCO). The two human osteosarcoma cell lines U2OS and Saos2, which were stably transfected with the tetracycline-controlled transactivator expression plasmid pUHD15-1 [33], were gifts from Liang Zhu and were maintained in 10% FBS/DMEM (GIBCO) with 250 µg/ml Geneticin (GIBCO).

Adhesion assays were carried out according to Cary et al. [56]. Briefly, 96-well Immunol-4 microtiter plates (Dynatech Laboratories) were precoated with human plasma fibronectin (Gibco BRL) at various concentrations in DMEM for 60 min at room temperature, washed twice with DMEM, and blocked with 1% BSA for 60 min at 37°C. Cells were initially cultured in regular tissue culture plates with or without tetracycline for 8 h to induce LAR expression. After induction, both LAR+ and LAR– cells were trypsinized, washed once with DMEM containing 0.5 mg/ml soybean trypsin inhibitor (Sigma), twice with DMEM containing 1% BSA, resuspended in DMEM, and then added to the fibronectin coated plates at 5 × 104 cells per well. An aliquot of cells at this point was saved for confirming LAR expression. After 30 min incubation at 37°C in a humidified atmosphere containing 5% CO2, nonadherent cells were washed twice with PBS, and adherent cells were fixed with 70% ethanol for 20 min. Cells were stained with 5 mg/ml crystal violet in 20% methanol for 10 min and washed 4 times with PBS. Bound stain was extracted with 0.1 M sodium citrate pH 4.2 for 30 min. The optical density (OD) was read at 562 nm. The background OD value from a well without cells was subtracted from the OD value for each well. LAR expression was confirmed by western blot analysis.

β-Galactosidase staining Cells were fixed with 0.5% glutaraldehyde/PBS for 10 min at room temperature, washed 3 times with PBS, and then incubated in X-gal staining solution (5 mM potassium ferricyanide, 5 mM potassium ferrocyanide, 1 mM MgCl2, 1 mg/ml X-gal in PBS) for 2–4 h at 37°C.

Induction and analysis of apoptosis To induce LAR expression, cells were trypsinized and washed twice with medium, and plated into tetracycline-free medium at 5 × 104 cells per 35 mm diameter dish. Cell viability were measured by plating equal numbers of cells into Tet+ and Tet– media and then cultured. At the end of the induction, viable cells, which remain attached to the dish, were trypsinized and counted with a Coulter counter. The ratio of the number of viable cells in Tet– cultures to the number in Tet+ cultures were calculated and presented as the percentage of cells surviving. Propidium iodide staining was performed by adding 1% volume of 69 µM propidium iodide in 38 mM sodium citrate pH 7.3 to the dishes followed by incubation for 10 min at 37°C. TUNEL analysis of DNA fragmentation was performed using an in situ apoptosis detection kit (ApopTag) following the instructions recommended by the manufacturer (Oncor). Gel electrophoretic analysis of DNA fragmentation was performed on a 1.2% agarose gel using genomic DNA extracted from detached cells with the Easy DNA kit (Invitrogen). The caspase inhibitors were purchased from Enzyme Systems. To test the effects of the caspase inhibitors on LAR-induced cell death, the U2OS/LAR transfectants were cultured in pairs of Tet+ and Tet– cultures and in the presence of the caspase inhibitors, 50 µM Z-Val-Ala-Asp-CH2F, 250 µM Z-Asp-Glu-Val-Asp-CH2F or 250 µM Z-Tyr-Val-Ala-Asp-CH2F for 12 h.

Protein extraction, immunoprecipitation, and immunoblotting Metabolic labeling of cells with [35S]methionine and immunoprecipitation was performed as described [2]. Briefly, [35S]methioninelabeled cells were washed with PBS and lysed in cold RIPA buffer (150 mM NaCl, 1% NP-40, 0.5% deoxycholate, 0.1% SDS, 50 mM Tris pH 8.0), and immunoprecipitated with anti-LAR antibody. The proteins were resolved by 6% gels by SDS–PAGE. For immunoblotting, Laemmli sample buffer was added directly to the cells to a final concentration of 106 cells/ml. The cell lysates were boiled for 5 min and aliquots equal to approximately 5 × 104 cells were subjected to SDS–PAGE (6% or 12% gels) and transferred to nitrocellulose membrane. The membranes were blocked with 5% nonfat dry milk in 10 mM Tris pH 7.5, 100 mM NaCl, 0.1% Tween 20 for 1 h at room temperature, incubated with appropriate primary antibodies for 2 h at room temperature or overnight at 4°C, followed by either secondary antibodies conjugated to alkaline phosphatase or horseradish peroxidase (Promega) for 1 h at room temperature. Proteins were detected with the ProtoBlot Western Blot AP System (Promega) or the ECL kit (Amersham).

Antibodies

Acknowledgements We thank Jerome S. Brody and Mary C. Williams for helpful discussion and critical reading of the manuscript, Liang Zhu for providing reagents for the tetracycline-inducible system, and Xin Wang for technical assistance. This work was supported by the National Heart, Lung, and Blood Institute (RO1 HL54044) and by an American Heart Association Established Investigatorship to J.Y.

References 1. Streuli M, Krueger NX, Hall LR, Schlossman SF, Saito H: A new member of the immunoglobulin superfamily that has a cytoplasmic region homologous to the leukocyte common antigen. J Exp Med 1988, 168:1523-1530. 2. Yu Q, Lenardo T, Weinberg RA: The N-terminal and C-terminal domains of a receptor tyrosine phosphatase are associated by non-covalent linkage. Oncogene 1992, 7:1051-1057. 3. Krueger NX, Streuli M, Saito H: Structural diversity and evolution of human receptor-like protein tyrosine phosphatases. EMBO J 1990, 9:3241-3252. 4. Tonks NK: Protein tyrosine phosphatases and the control of cellular signaling responses. Adv Pharmacol 1996, 36:91-119. 5. Katsura H, Williams MC, Brody JS, Yu Q: Two closely related receptor-type tyrosine phosphatases are differentially expressed during rat lung development. Dev Dyn 1995, 204:89-97. 6. Celler JW, Luo X, Gonez LJ, Bohmer FD: mRNA expression of two transmembrane protein tyrosine phosphatases is modulated by growth factors and growth arrest in 3T3 fibroblasts. Biochem Biophys Res Commun 1995, 209:614-621. 7. Longo FM, Martignetti JA, Le Beau JM, Zhang JS, Barnes JP, Brosius J: Leukocyte common antigen-related receptor-linked tyrosine phosphatase. Regulation of mRNA expression. J Biol Chem 1993, 268:26503-26511. 8. Kulas DT, Zhang WR, Goldstein BJ, Furlanetto RW, Mooney RA: Insulin receptor signaling is augmented by antisense inhibition of the protein tyrosine phosphatase LAR. J Biol Chem 1995, 270:2435-2438. 9. Kulas DT, Goldstein BJ, Mooney RA: The transmembrane proteintyrosine phosphatase LAR modulates signaling by multiple receptor tyrosine kinases. J Biol Chem 1996, 271:748-754.

256

Current Biology, Vol 8 No 5

10. Li PM, Zhang WR, Goldstein BJ: Suppression of insulin receptor activation by overexpression of the protein-tyrosine phosphatase LAR in hepatoma cells. Cell Signal 1996, 8:467-473. 11. Ahmad F, Goldstein BJ: Functional association between the insulin receptor and the transmembrane protein-tyrosine phosphatase LAR in intact cells. J Biol Chem 1997, 272:448-457. 12. Zhang WR, Li PM, Oswald MA, Goldstein BJ: Modulation of insulin signal transduction by eutopic overexpression of the receptortype protein-tyrosine phosphatase LAR. Mol Endocrinol 1996, 10:575-584. 13. Serra-Pages C, Kedersha NL, Fazikas L, Medley Q, Debant A, Streuli M: The LAR transmembrane protein tyrosine phosphatase and a coiled-coil LAR-interacting protein co-localize at focal adhesions. EMBO J 1995, 14:2827-3288. 14. Debant A, Serra-Pages C, Seipel K, O’Brien S, Tang M, Park SH: The multidomain protein Trio binds the LAR transmembrane tyrosine phosphatase, contains a protein kinase domain, and has separate rac-specific and rho-specific guanine nucleotide exchange factor domains. Proc Natl Acad Sci USA 1996, 93:5466-5471. 15. Tian SS, Tsoulfas P, Zinn K: Three receptor-linked protein-tyrosine phosphatases are selectively expressed on central nervous system axons in the Drosophila embryo. Cell 1991, 67:675-680. 16. Krueger NX, Van Vactor D, Wan HI, Gelbart WM, Goodman CS, Saito H: The transmembrane tyrosine phosphatase DLAR controls motor axon guidance in Drosophila. Cell 1996, 84:611-622. 17. Ellis RE, Yuan JY, Horvitz HR: Mechanisms and functions of cell death. Annu Rev Cell Biol 1991, 7:663-698. 18. Nagata S, Golstein P: The Fas death factor. Science 1995, 267:1449-1456. 19. Yuan J: Molecular control of life and death. Curr Opin Cell Biol 1995, 7:211-4. 20. Boise LH, Gottschalk AR, Quintans J, Thompson CB: Bcl-2 and Bcl2-related proteins in apoptosis regulation. Curr Top Microbiol Immunol 1995, 200:107-121. 21. Harrington EA, Fanidi A, Evan GI: Oncogenes and cell death. Curr Opin Genet Dev 1994, 4:120-129. 22. MacCabe JA, Noveroske JK: The control of cell death in the early chick embryo wing bud. Poult Sci 1997, 76:105-110. 23. Chow RL, Roux GD, Roghani M, Palmer MA, Rifkin DB, Moscatelli DA, Lang RA: FGF suppresses apoptosis and induces differentiation of fibre cells in the mouse lens. Development 1995, 121:43834893. 24. Gardner AM, Johnson GL: Fibroblast growth factor-2 suppression of tumor necrosis factor alpha-mediated apoptosis requires Ras and the activation of mitogen-activated protein kinase. J Biol Chem 1996, 271:14560-14566. 25. D’Mello SR, Borodezt K, Soltoff SP: Insulin-like growth factor and potassium depolarization maintain neuronal survival by distinct pathways: possible involvement of PI 3-kinase in IGF-1 signaling. J Neurosci 1997, 17:1548-1560. 26. Harrington EA, Bennett MR, Fanidi A, Evan GI: c-Myc-induced apoptosis in fibroblasts is inhibited by specific cytokines. EMBO J 1994, 13:3286-3295. 27. Jung Y, Miura M, Yuan J: Suppression of interleukin-1 betaconverting enzyme-mediated cell death by insulin-like growth factor. J Biol Chem 1996, 271:5112-5117. 28. Miller TM, Tansey MG, Johnson EM Jr, Creedon DJ: Inhibition of phosphatidylinositol 3-kinase activity blocks depolarization- and insulin-like growth factor I-mediated survival of cerebellar granule cells. J Biol Chem 1997, 272:9847-9853. 29. Minshall C, Arkins S, Freund GG, Kelley KW: Requirement for phosphatidylinositol 3’-kinase to protect hemopoietic progenitors against apoptosis depends upon the extracellular survival factor. J Immunol 1996, 156:939-947. 30. Bergamaschi G, Rosti V, Danova M, Ponchio L, Lucotti C, Cazzola M: Inhibitors of tyrosine phosphorylation induce apoptosis in human leukemic cell lines. Leukemia 1993, 7:2012-2018. 31. Yousefi S, Green DR, Blaser K, Simon HU: Protein-tyrosine phosphorylation regulates apoptosis in human eosinophils and neutrophils. Proc Natl Acad Sci USA 1994, 91:10868-10872. 32. Miura M, Zhu H, Rotello R, Hartwieg EA, Yuan J: Induction of apoptosis in fibroblasts by IL-1b-converting enzyme, a mammalian homolog of the C. elegans cell death gene ced-3. Cell 1993, 75:653-660. 33. Gossen M, Bujard H: Tight control of gene expression in mammalian cells by tetracycline-responsive promoters. Proc Natl Acad Sci USA 1992, 89:5547-5551.

34. Dolle RE, Hoyer D, Prasad CV, Schmidt SJ, Helaszek CT, Miller RE, Ator MA: P1 aspartate-based peptide alpha-((2,6dichlorobenzoyl)oxy)methyl ketones as potent time-dependent inhibitors of interleukin-1 beta-converting enzyme. J Med Chem 1994, 37:563-564. 35. Thornberry NA, Bull HG, Calaycay JR, Chapman KT, Howard AD, Kostura MJ, et al.: A novel heterodimeric cysteine protease is required for interleukin-1b processing in monocytes. Nature 1992, 356:768-774. 36. Nicholson DW, Ali A, Thornberry NA, Vaillancourt JP, Ding CK, Gallant M, et al.: Identification and inhibition of the ICE/CED-3 protease necessary for mammalian apoptosis. Nature 1995, 376:37-43. 37. Reed JC, Ruoslahti E: Anchorage dependence, integrins, and apoptosis. Cell 1994, 77:477-478. 38. Frisch SM, Francis H: Disruption of epithelial cell-matrix interactions induces apoptosis. J Cell Biol 1994, 124:619-626. 39. Hale AJ, Smith CA, Sutherland LC, Stoneman VEA, Longthorne VL, Culhane AC, Williams GT: Apoptosis: molecular regulation of cell death. Eur J Biochem 1996, 236:1-26. 40. Burridge K, Chrzanowska-Wodnicka M: Focal adhesions, contractility, and signaling. Annu Rev Cell Dev Biol 1996, 12:463518. 41. Clark EA, Brugge JS: Integrins and signal transduction pathways: the road taken. Science 1995, 268:233-239. 42. Yamada KM, Miyamoto S: Integrin transmembrane signaling and cytoskeletal control. Curr Opin Cell Biol 1995, 7:681-689. 43. Richardson A, Parsons T: A mechanism for regulation of the adhesion-associated proteintyrosine kinase pp125FAK. Nature 1996, 380:538-540. 44. Boudreau N, Sympson CJ, Werb Z, Bissel MJ: Suppression of ICE and apoptosis in mammary epithelial cells by extracellular matrix. Science 1995, 267:891-893. 45. Boudreau N, Werb Z, Bissell MJ: Suppression of apoptosis by basement membrane requires three-dimensional tissue organization and withdrawal from the cell cycle. Proc Natl Acad Sci USA 1996, 93:3509-3513. 46. Yang C, Chang J, Gorospe M, Passaniti A: Protein tyrosine phosphatase regulation of endothelial cell apoptosis and differentiation. Cell Growth Differentiation 1996, 7:161-171. 47. Lund-Johansen F, Frey T, Ledbetter JA, Thompson PA: Apoptosis in hematopoietic cells is associated with an extensive decrease in cellular phosphotyrosine content that can be inhibited by the tyrosine phosphatase antagonist pervanadate. Cytometry 1996, 25:182-190. 48. Huang TS, Shu CH, Shih YL, Huang HC, Su YC, Chao Y, et al.: Protein tyrosine phosphatase activities are involved in apoptotic cancer cell death induced by GL331, a new homolog of etoposide. Cancer Lett 1996, 110:77-85. 49. Ahmad F, Goldstein BJ: Effect of tumor necrosis factor-alpha on the phosphorylation of tyrosine kinase receptors is associated with dynamic alterations in specific protein-tyrosine phosphatases. J Cell Biochem 1997, 64:117-127. 50. Fisher DE: Apoptosis in cancer therapy: crossing the threshold. Cell 1994, 78:539-542. 51. Clarke AR, Purdie CA, Harrison DJ, Morris RG, Bird CC, Hooper ML, Wyllie AH: Thymocyte apoptosis induced by p53-dependent and independent pathways. Nature 1993, 362:849-852. 52. Lowe SW, Schmitt EM, Smith SW, Osborne BA, Jacks T: p53 is required for radiation-induced apoptosis in mouse thymocytes. Nature 1993, 362:847-849. 53. Morgenstern JP, Land H: A series of mammalian expression vectors and characterisation of their expression of a reporter gene in stably and transiently transfected cells. Nucl Acids Res 1990, 18:1068. 54. Gorman C, Padmanabhan R, Howard BH: High efficiency DNAmediated transformation of primate cells. Science 1983, 221:551553. 55. Pragnell IB, Spandidos DA, Wilkie NM: Consequences of altered oncogene expression in rodent cells. Proc R Soc Lond [Biol] 1985, 226:107-119. 56. Cary LA, Chang JF, Guan JL: Stimulation of cell migration by overexpression of focal adhesion kinase and its association with Src and Fyn. J Cell Sci 1996, 109:1787-1794.