Apoptosis 2002; 7: 195–207
C 2002 Kluwer Academic Publishers

Induction of cytochrome c release and apoptosis by Hck-SH3 domain-mediated signalling requires caspase-3 V. Radha, Ch. Sudhakar, P. Ray and G. Swarup Centre for Cellular and Molecular Biology, Hyderabad 500 007, India

The function of key components of signal transduction, the Src family tyrosine kinases is dependent on catalytic activity as well as on intermolecular interaction achieved by their SH2 and SH3 modular domains. We have analyzed the effect of overexpression of the hematopoietic cell kinase (Hck) and its N-terminal unique and SH3 domains on cell survival. Overexpression of the N-terminal unique and SH3 domains (Hck-USH3) induced about 25% of expressing Cos-1 cells to undergo apoptosis 30 hrs after transfection. The full length p59 and p56 forms and the unique domain alone induced low levels of cell death. The unique and SH3 domain of a closely related kinase, Lyn did not induce apoptosis. Overexpression of a mutant USH3 domain (Gly → Ala), that disrupts membrane localization, did not induce high level of apoptosis. Cells overexpressing Hck-USH3 showed activation of caspase-3 and release of cytochrome c from mitochondria into cytosol. Caspase-3 defective MCF-7 cells were resistant to apoptosis and cytochrome c release induced by Hck-USH3, which were restored by introducing the caspase-3 gene. These results suggest that Hck SH3 domain mediated signalling at the plasma membrane triggers a pathway leading to caspase-3 dependent cytochrome c release and apoptosis. Keywords: apoptosis; caspase-3; cytochrome c; Hck; SH3 domain.

Introduction Programmed cell death or apoptosis that eukaryotic cells undergo is essential for maintenance of homeostasis in the various developmental stages and physiological processes in multicellular organisms.1 In vivo, apoptosis takes place to remove unwanted and altered cells without causing inflammation and therefore, deregulation of apoptotic controls could either lead to hyperproliferative or degenerative disorders.2,3 Apoptosis can be induced by a variety of extrinsic factors such as nutrient or growth Correspondence to: G. Swarup, Centre for Cellular and Molecular Biology, Hyderabad 500 007, India. Tel: +91-40-7160222-41 (ext: 2616); Fax: +91-40-7160311, +91-40-7160591; e-mail: [email protected]

factor deprivation, presence of cytokines or abnormal cell microenvironment. It can also be induced as a result of an imbalance in intrinsic factors which are intracellular molecules that convey death or survival signals or effector molecules that cause the morphological changes in the cell.4 The basic machinery for apoptosis is present in all cells and some of the signalling pathways appear to be common to those involved in other cellular processes.5 Nevertheless, multiple pathways that have feed back loops and are not necessarily linear, appear to be activated in a stimulus and cell type dependent manner, bringing in a complexity that is presently not very well understood. A family of proteases known as caspases (which are aspartate specific cysteine proteases) play important roles in execution as well as commitment phase of apoptosis.6,7 Different caspases appear to have distinct roles in this process as suggested by the phenotype of knockout mice.8 In one of the pathways of apoptosis induced by a variety of agents, mitochondrial cytochrome c is released into cytosol where it forms a complex with APAF-1 and activates caspase-9 which then cleaves procaspase-3 leading to the generation of active caspase-3.9 Active caspase-3 is one of the main executioners of apoptosis which cleaves many proteins including other caspases and antiapoptotic protein Bcl 2.10 In eukaryotic cells, tyrosine phosphorylation regulates transmission of signals across the plasma membrane to the nucleus and, therefore, maintaining a balance between tyrosine kinases and phosphatases may play a crucial role in regulating apoptotic pathways.11−13 The Src family tyrosine kinases are key components of signal transduction and the involvement of some members of this family, has been shown in Fas mediated and HIV-induced apoptosis.14−18 The hematopoietic cell kinase (Hck) shows restricted expression, primarily in mature myelomonocytic cells and B lymphocytes.19,20 It has been shown to transduce signals from cytokines and FcY receptors.21−23 Hck has the basic structure possessed by all Src family members, with a catalytic domain in the C-terminal half of the molecule. Preceding the catalytic domain are two domains called

Apoptosis · Vol 7 · No 3 · 2002

195

V. Radha et al.

SH2 and SH3 which are protein interaction modules.16 SH2 domains associate with peptide sequences containing a phosphotyrosine residue whereas SH3 domains interact with proline rich sequences in proteins.24 Among the various Src family members, the extent of homology is greatest in the catalytic domain, decreasing towards the N-terminal region, where, preceeding the SH3 domain are sequences with no homology and are therefore called the unique domain. The first few residues at the Nterminal are modified by acylation helping anchorage of the molecule at the plasma membrane.25,26 Hck protein exists as two isoforms in murine cells, p59 and p56 (p61 and p59 in human) which arise due to the usage of two different translation start sites.27 A large number of substrates and ligands that interact with the various domains have been identified for the Src family members, but in many instances, the contribution of these to downstream effector functions has not been delineated.16,28 Several physiological and pharmacological agents have been used for in vitro and in vivo studies as well as in cell free systems in an attempt to find molecules regulating apoptosis and to delineate the pathways leading to apoptosis. Here we report that ectopic expression of Hck results in low level of apoptosis which increased severalfold when Hck SH3 domain was expressed alongwith the unique domain (USH3). This apoptosis was dependent on membrane localization of these domains and did not require functional p53. Caspase-3 was required for USH3 induced apoptosis which was activated during this process. Mitochondrial cytochrome c was released into cytosol by overexpression of USH3 and unexpectedly caspase3 was required for this release of cytochrome c. Thus our observations suggest an additional role for caspase-3 upstream of cytochrome c release in the apoptotic pathway induced by expression of SH3 and unique domains of Hck.

Materials and methods Materials Antibody against cleaved caspase-3 (activated form) was a rabbit polyclonal obtained from Cell Signalling Technology. Rabbit polyclonal Hck antibody (Santa Cruz) was against a peptide from unique domain of Hck. A mouse monoclonal antibody 3E9 which recognised an epitope in SH3 domain of Hck has been described earlier.29 Cytochrome c antibody was a rabbit polyclonal from Santa Cruz. Polyclonal Lyn antibody raised in rabbit against a peptide from the unique domain was also obtained from Santa Cruz. DEVD-cmk, YVAD-cmk and DEVD-AMC were from Bachem. Cell lines and transfections. All cell lines used were maintained in DMEM with 10% FCS and antibiotics in a 196 Apoptosis · Vol 7 · No 3 · 2002

humidified CO2 incubator at 37◦ C. Transfections for transient or stable expression were performed using Qiagen column purified plasmids and Lipofectamine reagent (Life Technologies) according to the manufacturer’s specifications and as described earlier.13,30 For obtaining stable clones of MCF-7 cells expressing caspase-3 (or carrying control vector), transfected MCF-7 cells were maintained in medium containing 100 µg/ml G418 and stable pools of colonies were obtained after 15 days of selection. Apoptosis assays. Cells plated on coverslips were transfected with the expression vectors and at the indicated time, fixed for indirect immunofluorescence as described earlier.13,30 To visualize the cells expressing the Hck constructs, they were stained with either a polyclonal antiHck antibody (Santa Cruz) or a monoclonal antibody 3E9 that recognises the Hck-SH3 domain. Fluorescein conjugated anti rabbit or anti-mouse second antibodies were used. Cells were mounted in 90% glycerol in PBS containing 0.1% p-phenylenediamine and 0.5 µg/ml of DAPI. Cells were visualised using a fluorescence microscope for FITC staining (positive for expression) and DAPI staining to examine nuclear morphology. Apoptotic cells were identified by the appearance of condensed chromatin, apoptotic bodies, cell shrinkage or membrane blebbing.13 In MCF-7 cells which do not show apoptotic body formation and membrane blebbing due to lack of functional caspase-3, apoptotic cells were identified by the presence of chromatin condensation and cell shrinkage along with a rounded morphology and loss of refractility.13 Apoptotic index was determined as the percentage of apoptotic cells among the total number of expressing cells. For each experiment a minimum 200 cells were counted per coverslip from microscopic fields. The number of apoptotic cells among the non-expressing cells was also counted from the same coverslips as background control. Caspase peptide inhibitors were added to transfected cells at 250 µM concentration, 6 hours after transfection wherever indicated. To quantitate the total number of cells undergoing cell death, Cos-1 cells (5 × 104 ) were transfected in duplicate 35 mm dishes with either control plasmid pCl, or plasmids expressing the various Hck constructs. After 48 hours of transfection cell viability was determined by counting trypan blue excluding cells using a hemocytometer. Terminal deoxynucleotidyl transferase-mediated dUTP nick end labelling (TUNEL) assay was performed according to the manufacturer’s specification (Boehringer). Cos-1 cells on coverslips were transfected with control plasmid or Hck-USH3 and 30 hrs later fixed in 4% formaldehyde in PBS for 30 min. Cells were permeabilized in cold PBS with 0.1% Triton X-100 and 0.1% sodium citrate for 2 min. Coverslips were washed in PBS and 50 µl of TUNEL reaction mix added and left for 60 min at 37◦ C in a humidified dark chamber. After washing thrice

Hck SH3 mediated caspase-3 dependent apoptosis

in PBS, coverslips were mounted and observed under a fluorescence microscope. After TUNEL labeling, HckUSH3 transfected cells were stained with Hck antibody and AMCA conjugated secondary antibodies to visualize the transfected cells and confirm TUNEL positivity of cells that show apoptotic morphology. Caspase activity assay. Control vector or Hck-USH3 transfected Cos-1 cells were lysed in cold lysis buffer (20 mM Tris 8.0, 137 mM NaCl, 10% glycerol, 2 mM DTT, 1 mM EDTA, 1% NP40, 1 mM PMSF and 2 µg/ml Soyabean trypsin inhibitor, aprotinin and leupeptin). The supernatant after centrifugation was used at the required protein concentration in a 50 µl reaction volume containing assay buffer (similar to lysis buffer except Tris pH 7.5 was used) and 50 µM DEVD-AMC and incubated at 37◦ C for 1 hr. The reaction was diluted to 500 µl in assay buffer and fluorescence measured at 360 nm excitation and 460 nm emission. Peptide inhibitors DEVD-cmk or YVAD-cmk were added to the reactions at 0.1 and 1 µM concentration. Cos-1 cells on coverslips were transfected with either control vector or Hck-USH3 and dual stained with antibodies for Hck and cleaved caspase-3 for detection of caspase-3 activation. Western blotting. Whole cell lysates were prepared from cells transfected with the indicated vectors and subjected to SDS-PAGE and western blotting as described earlier.31 Cytochrome c release. Release of cytochrome c from mitochondria was visualized by immunostaining of cells. Dual immunostaining for cytochrome c and Hck was performed by sequential staining of transfected cells on coverslips using anti cytochrome c antibody (Santa Cruz); FITC conjugated anti rabbit IgG; anti-Hck monoclonal antibody and AMCA conjugated anti-mouse IgG. After each step cells were washed thrice in PBS and finally mounted in the mountant. Cytochrome c was visualised as fluorescein staining and Hck as AMCA staining using a flourescence microscope. Cytochrome c release was identified in cells that showed diffuse cytosolic staining instead of the particulate mitochondrial staining.32 Proportion of cells that showed cytochrome c release was scored from microscopic fields. Expression vectors All the expression vectors derived from Hck cDNA were cloned in pCl plasmid (Promega) which drives the expression under the control of CMV promoter. The full length rat Hck cDNA was cloned in Eco RI site of pCl to obtain the construct p59 which expressed p59 and p56 isoforms. A shorter rat Hck cDNA lacking sequences upstream of ATG start site was cloned in Eco RI site of pCl vector to obtain p56 construct which expressed only the p56

isoform. The sequence of these two rat Hck cDNAs has been described previously.20,29 In order to express unique and unique plus SH3 domains the desired regions were amplified by PCR and the PCR products were cloned in Eco RI site of pCl vector after digesting with Eco RI. For unique region the cDNA coding for amino acids 1-59 (where amino acid 1 is Met) was amplified. For unique plus SH3 domains, (USH3 construct) the cDNA corresponding to amino acids 1-114 was amplified. In order to obtain mUSH3 construct, which has Ala in place of Gly at position 2, the PCR primer had GCA instead of GGA codon. The cDNA coding for unique and SH3 domains (N-terminal 122 amino acids) of Lyn tyrosine kinase p56lyn , was obtained by RT-PCR using mouse spleen RNA as template. The resulting cDNA was cloned in Eco RI site of pCl plasmid. Sequence of all these constructs was determined after cloning in pCl vector (using T7 primer and a primer designed for sequencing) using an automated sequencer ABI Prism model 377 from Applied Biosystems. The plasmids for murine Bcl2 gene and poxvirus gene CrmA have been described earlier.33,34 Human caspase-3 cDNA cloned in pcDNA3 has been described.35 In this plasmid the expression of caspase-3 is driven by CMV promoter.

Results Overexpression of Hck-SH3 domain induces apoptosis To study the possible function of Hck in effecting cell survival, we examined the phenotype of Cos-1 cells after transient transfection using expression vectors carrying either full length Hck isoforms (p59 and p56) or truncated vectors with only the unique domain or unique and SH3 domains (USH3) (Figure 1). Expression of the protein as well as cell morphology were monitored after immunostaining of cells with anti Hck antibody and nuclei with DAPI. Full length Hck as well as the truncated forms showed a similar pattern of localization with prominent staining of the plasma membrane and some staining of cytoplasmic organelles (Figure 2). Transfection with full length constructs had no effect on morphology of most of the cells but a large number of cells expressing Hck-USH3 showed features of apoptosis (Figure 2). Observation of the DNA staining in these cells showed irregular nuclear morphology characterized by intense staining of condensed and fragmented chromatin. Cells showing these changes, which are indicative of apoptosis, were counted and it was found that Hck-USH3 expressing cells showed a high apoptotic index of 25.4%, while the full length Hck expressing cells showed much lower number of apoptotic cells (8.8%). Nonexpressing cells showed apoptosis in the range of 2–3% (Figure 1A). Overexpression of green fluorescence protein in Cos-1 cells Apoptosis · Vol 7 · No 3 · 2002

197

V. Radha et al. Figure 1. Overexpression of Hck-USH3 domain induces apoptosis. (A) Cos-1 cells were transiently transfected with the indicated constructs and apoptotic index determined at 30 hrs after transfection. The percentage of apoptotic cells was calculated as the number of expressing cells with apoptotic morphology divided by the total number of expressing cells. A minimum of 200 expressing cells and 1000 non-expressing cells were counted from microscopic fields in each coverslip. Data represent mean ± s.d. of at least 3 independent experiments. (B) Western blot of whole cell lysates after transfection with the indicated constructs showing size and expression. (C) Line diagram of the various constructs used in the apoptosis assays. Numbers indicate position of amino acids.

induced an average apoptosis of 3.75% in expressing cells showing that high levels of expression of any exogenous protein does not induce cell death. To delineate whether this property was specific to the unique or the SH3 domain, we expressed the unique domain alone, and found that it did not induce apoptotic changes in most of the expressing cells (Figures 1 and 2). TUNEL assay was used to confirm that the morphological changes visualised microscopically were indeed due to apoptosis (Figure 3). This assay is based on the ability of terminal transferase to extend the 3 -hydroxyl end of chromosome breaks and it stains those cells that have fragmented DNA which is one of the hallmarks of apoptosis. In order to quantitate the total number of cells undergoing cell death, we assayed 198 Apoptosis · Vol 7 · No 3 · 2002

Table 1. Effect of Hck constructs on survival of Cos-1 cells Expression plasmid Control plasmid, pCl

Percent survival 100

p59

90.2 ± 8.7 (n = 6)

p56

96.2 ± 5.4 (n = 4)

USH3

50.6 ± 7.7 (n = 5)

mUSH3

78.2 ± 5.7 (n = 6)

Unique

102 ± 3.3 (n = 4)

Cell survival is expressed as the percentage of live cells remaining 48 hrs after transfection with various constructs compared to transfection with control plasmid pCl.

the number of surviving cells remaining after 48 hours of transfection with various Hck constructs. These results showed that USH3 protein induced much more cell death than any other Hck construct (Table 1). We analysed the extent of expression and size of expressed proteins from the various constructs by Western blotting. The size of expressed proteins from all the constructs was as expected. The level of expression of unique and USH3 proteins was equal. However, the level of full length Hck proteins was lower, which was due to lower number of transfected cells generally obtained with these constructs as determined by counting the expressing cells in immunofluorescence staining experiments. Therefore, lower level of apoptosis observed with full length Hck proteins may not be due to lower level of expression than USH3 protein, but we cannot entirely rule out this possibility. In any case it is clear from these data (Figures 1 and 2) that kinase and SH2 domains are not required for Hck induced apoptosis. We also quantitated the extent of apoptosis at different times after transfection and found that percentage of apoptotic cells were greater at 30 hrs compared to 24 hrs, with no significant increase at 48 hrs. (data not shown). Therefore, in all further experiments extent of apoptosis was quantitated at 30 hrs after transfection. In some instances, it has been observed that overexpression of a protein sensitizes cells and apoptosis occurs under conditions of stress such as growth factor withdrawal. In our experiments, cells were fed with serum 6 hrs after transfection and maintained in serum containing medium until fixation. Since Hck-USH3 overexpression induced apoptosis under these conditions, it was not dependent on withdrawal of growth factors. To investigate whether the property of Hck-USH3 to induce apoptosis is cell type specific, we performed similar experiments using other cell types. Since p53 is a central modulator of many apoptotic stimuli, we used cell lines that had either normal p53 (A549 lung carcinoma) or were nonfunctional for p53 (HeLa cervical carcinoma and SW620 colon cancer). It was seen that HckUSH3 overexpression resulted in apoptosis (in 25–30% of

Hck SH3 mediated caspase-3 dependent apoptosis Figure 2. Effect of expression of Hck and its variant constructs on cellular morphology. Cos-1 cells were transfected with the indicated constructs and immunostained to visualize exogenously expressed protein (FITC). Nuclear morphology observed by DAPI staining and phase constrast pictures of the same cells are shown. Shown are representative fields 30 hrs after transfection. Cells overexpressing Hck-USH3 show apoptotic morphology (condensed/fragmented nuclei) indicated by arrows in a fraction of the cells. Arrow heads indicate apoptotic cells that show membrane blebbing. Cells overexpressing all other Hck forms generally show normal morphology with intact nuclei.

expressing cells) in HeLa cells (Figure 4), A549 and SW 620 cells (data not shown) suggesting the independence of the pathway for functional p53. Hck-USH3 also induced apoptosis in a normal mouse fibroblast cell line C3H

indicating that nontransformed cells are also susceptible to apoptosis. The SH3 domain of Hck shows homology with that of other Src family members, particularly that of Lyn (71% Apoptosis · Vol 7 · No 3 · 2002

199

V. Radha et al. Figure 3. TUNEL positivity in USH-3 expressing cells. (A) Fluorescence and phase contrast micrographs of control vector and USH3 transfected Cos-1 cells after TUNEL assay. (B) Hck-USH3 transfected cells stained with Hck antibody and AMCA conjugated secondary antibodies after TUNEL labeling. Arrows indicate USH3 expressing cells that are TUNEL positive and arrow heads indicate USH3 expressing cells with normal morphology and no TUNEL positivity.

amino acids identical). Therefore, we examined the ability of overexpressed Lyn-USH3 (derived from p56 isoform of Lyn) to induce apoptosis in Cos-1 cells. Lyn-USH3 also localized primarily to the plasma membrane, but did not induce apoptosis to the extent seen in cells expressing Hck-USH3 (Figure 5). Therefore the apoptosis inducing property is not a general feature of Src family kinases. The antiapoptotic protein Bcl2 has been shown to protect cells against apoptosis induced by various agents.36,37 We therefore cotransfected Bcl2 expression plasmid along with Hck-USH3 and found that Bcl2 can rescue cells from Hck-USH3 induced apoptosis (Table 2).

Table 2. Effect of cotransfection of Bcl2 and Crm A on USH3 induced apoptosis

Membrane localization is necessary for apoptotic activity of Hck-USH3

apoptosis, a point mutation (Gly → Ala) that disrupts the site of myristylation and thereby prevents membrane attachment was introduced (mUSH3). When Cos-1 or HeLa cells were transfected with this vector (mUSH3) predominant nuclear and cytoplasmic staining was observed, with very minimal amount remaining at the plasma membrane (Figure 2). After 30 hours of transfection, only 8.1% of over-expressing Cos-1 cells (Figure 1) and 6.4% of

Hck has been shown to transmit signals originating from transmembrane receptors and its localization at the plasma membrane is concurrent with such an activity.22,23,25 To determine whether membrane localization of the overexpressed Hck-USH3 protein is necessary to induce 200 Apoptosis · Vol 7 · No 3 · 2002

Expression plasmid

% Apoptosis in % Apoptosis in non-expressing expressing cells cells

USH3: pCl 1 : 2 (n = 4)

27.53 ± 2.7

2.4 ± 2.1

USH3: Bcl2 1 : 2 (n = 4)

9.62 ± 1.3

2.6 ± 1.1

USH3: CrmA 1 : 2 (n = 4)

27.2 ± 0.67

3.0 ± 1.0

Cos-1 cells were cotransfected with the indicated plasmids and the proportion of USH3 expressing cells showing apoptotic phenotype indicated as percentages. pCl plasmid was used as control.

Hck SH3 mediated caspase-3 dependent apoptosis Figure 4. Induction of apoptosis by USH3 in HeLa cells. The indicated constructs were transiently transfected into HeLa cells and stained with Hck antibody. Apoptotic index was determined as the percentage of apoptotic cells among the total number of expressing cells. The data represent mean ± s.d. of at least 3 experiments.

since it was expressed at least at the same level as USH3 (Figure 1B). Activation of caspases during Hck-USH3 induced apoptosis Apoptotic cell death has primarily been shown to involve caspases that not only cause the morphological changes by cleaving a variety of substrates, but also act to convey apoptotic signals.6,38 The caspase-3 family of enzymes which are known executioners of apoptosis have been shown to be activated during apoptosis induced by various agents.39 We therefore performed DEVDase activity assays using lysates from control vector and Hck-USH3 transfected Cos-1 cells and found about five fold increase in DEVDase activity in Hck-USH3 transfected cells (Figure 6A). This activity was inhibited by the presence of the peptide DEVD-cmk but not by YVAD-cmk in the assay reaction (data not shown). Since caspase-3 or other closely related caspases may contribute to the DEVDase activity, we monitored caspase-3 cleavage leading to its activation in Hck-USH3 expressing cells using an antibody that recognises only cleaved caspase-3. As shown in Figure 6C, cleaved caspase-3 is seen only in Hck-USH3 transfected cells.

Figure 5. Lyn-USH3 overexpression does not induce apoptosis. Cos-1 cells were transiently transfected with either Hck-USH3 or Lyn-USH3 constructs and the extent of apoptosis induced was determined in expressing and non-expressing cells after staining with Hck or Lyn antibodies respectively.

mUSH3 expressing HeLa cells (Figure 4) showed the morphological changes of apoptosis and cell survival at 48 hrs was higher (Table 2) suggesting that membrane anchoring of Hck-USH3 is required for induction of apoptosis by USH3. Lower level of induction of apoptosis by mUSH3 protein was not due to lower level of expression

Caspase-3 is required for Hck-USH3 induced apoptosis To examine the role of caspase-3 in Hck-USH3 induced apoptosis, we tested the effect of various caspase inhibitors. Hck-USH3 induced apoptosis could be reduced by DEVD-cmk, a caspase-3 family peptide inhibitor, but not by YVAD-cmk, a caspase-1 family inhibitor (Figure 6B). A general peptide inhibitor of caspases ZVAD fmk also partially inhibited apoptosis induced by Hck-USH3 (data not shown). Cotransfection of Hck-USH3 with the cowpox viral protein Crm A that inhibits most of the members of caspase-1 and caspase-8 family,40 but is not effective against caspase-3, did not inhibit apoptosis (Table 1). Crm A plasmid was functional since it could inhibit apoptosis induced by a protein tyrosine phosphatase.13 These results indicated that caspase-3 or a related enzyme is necessary for Hck-USH3 induced apoptosis. To provide direct evidence for the requirement of caspase-3, we used the human breast carcinoma cell line, MCF-7 which lacks functional caspase-3 due to a deletion in the gene.41 Expression of Hck-USH3 into MCF-7 cells did not induce high level of apoptosis as determined by morphological criteria of chromatin condensation, cell shrinkage along with rounded morphology and loss of refractility (Figure 7). In order to confirm that the inability of HckUSH3 to induce significant level of apoptosis in MCF-7 cells is due to lack of caspase-3 and not due to some other Apoptosis · Vol 7 · No 3 · 2002

201

V. Radha et al. Figure 6. Caspase activation upon Hck-USH3 overexpression and inhibition of USH3 induced apoptosis by caspase inhibitors. (A) Extracts of Cos-1 cells transfected with control plasmid or USH3 were assayed for caspase activity using the substrate Ac-DEVD-AMC. (B) Effect of caspase inhibitors YVAD-cmk and DEVD-cmk on apoptosis induced by USH3 in Cos-1 cells. (C) Cleaved caspase-3 expression in USH3 expressing Cos-1 cells. Fluorescence and phase contrast micrographs of control vector and Hck-USH3 transfected Cos-1 cells dual stained for Hck (FITC) and cleaved caspase-3 (AMCA). Arrows indicate USH3 expressing cells that show cleaved caspase-3. Arrowheads show non-apoptotic USH3 expressing cells that are not positive for cleaved caspase-3.

defect, we made stable clones of MCF-7 cells expressing the vector alone with neomycin resistance gene (nr MCF2) or a caspase-3 construct (P2 and P3). Transient transfection of Hck-USH3 into the control nr MCF clone showed apoptosis similar to MCF-7 but a larger number of apoptotic cells (about 20%) were observed in the caspase-3 expressing clones P2 and P3 (Figure 7B). Expressing cells also showed apoptotic bodies (Figure 7A) consistent with the presence of caspase-3 in these cells.41,42 Hck-USH3 triggers cytochrome c release from mitochondria One of the features frequently observed in apoptotic cells is the release of mitochondrial cytochrome c into the cytoplasm.43 This leads to the formation of the apoptosome whereby caspase-9 is activated leading to activation of caspase-3.44 We attempted to find out if Hck-USH3 induced release of cytochrome c. We analysed cytochrome c subcellular localization by double labelled immunofluorescence in Cos-1 cells over-expressing Hck-USH3. While 202 Apoptosis · Vol 7 · No 3 · 2002

the cells not expressing USH3 showed a particulate staining in the cytoplasm typical of mitochondrial localization, many of the Hck-USH3 expressing cells exhibited a diffuse fluorescence throughout the entire cell consistent with release from mitochondria (Figure 8). This pattern of staining was similar to that observed for cytochrome c release upon sodium butyrate induced apoptosis. Several expressing cells that did not show an apoptotic morphology also showed cytochrome c release indicating that this event occurs prior to appearance of apoptotic changes. Since cytochrome c release is an event which occurs upstream of caspase-3 activation,44 we reasoned that Hck-USH3 overexpression should show cytochrome c release in MCF-7 cells which lack functional caspase-3 protein. We therefore stained control vector and Hck-USH3 transfected MCF-7 cells for cytochrome c localization. Unexpectedly we found that there was no difference in the number of expressing cells showing cytochrome c release compared to the vector transfected or non-expressing cells indicating that Hck-USH3 does not trigger cytochrome c release in these cells (Figure 9). When this experiment was

Hck SH3 mediated caspase-3 dependent apoptosis Figure 7. Requirement of caspase 3 for induction of apoptosis by Hck-USH3. (A) USH3 transfected MCF-7 and P2 cells showing expression (FITC), nuclear morphology (DAPI) and cellular morphology (Phase). Arrows indicate apoptotic cells. (B) Quantitative representation of the extent of apoptosis induced by USH3 in MCF-7, neomycin resistant control, nrMCF2 and caspase-3 expressing stable clones P2 and P3. (C) Western blot showing expression of caspase-3 in stable cell lines, P2 and P3.

Figure 8. Hck-USH3 expression causes cytochrome c release. Cos-1 cells transfected with Hck-USH3 plasmid for 30 hrs were dual labelled for Hck-staining (AMCA) and cytochrome c (FITC). Hck-USH3 expressing cells exhibit diffusely staining cytochrome c that obscures the nucleus (arrow heads) in contrast to neighbouring non-expressing cells that show punctate cytochrome c staining around a well defined nucleus.

cytochrome c. These results suggest that the release of cytochrome c induced by Hck-USH3 expression requires caspase-3. In response to other apoptotic stimuli such as treatment with sodium butyrate, cytochrome c release was seen in MCF-7 cells (data not shown) indicating that this event is not always dependent on caspase-3, but in the pathway of Hck-USH3 induced apoptosis, cytochrome c release is dependent on the presence of caspase-3.

Discussion

done using caspase-3 expressing clones of MCF-7 cells (P2 and P3) we observed a number of USH3 expressing non apoptotic cells showing cytochrome c release (Figure 9). In addition to these, about 20% of the USH3 expressing cells which showed apoptosis, also showed release of

In the present study we found that overexpression of the tyrosine kinase Hck affects cell survival to some extent, but overexpression of its SH3 domain along with unique domain induces high level of programmed cell death. Expression of unique and SH3 domains of a closely related tyrosine kinase, Lyn, did not induce apoptosis showing that the apoptosis inducing activity is not a general property of Src family kinases. Since the full length Hck protein also has an SH3 domain, it could be inferred that the SH3 domain independent of the catalytic and SH2 domains, may interact differently or more effectively with cellular Apoptosis · Vol 7 · No 3 · 2002

203

V. Radha et al. Figure 9. Cytochrome c release is dependent on caspase-3 in Hck-USH3 induced apoptosis. Upper panel shows images of MCF-7 cells and caspase-3 expressing clone P2 transfected with Hck-USH3 and double labelled for Hck expression (AMCA) and cytochrome c (FITC). Diffuse cytosolic staining as opposed to punctate mitochondrial staining was indicative of cytochrome c release. Arrows indicate cells showing cytochrome c release. The number of USH3 expressing non-apoptotic cells showing release of mitochondrial cytochrome c was counted from microscopic fields and represented as a percentage of the total expressing non-apoptotic cells (lower panel). This is in addition to USH3 expressing apoptotic cells (about 20%) all of which showed cytochrome c release. The percentage of nonexpressing cells showing release is also shown.

proteins as compared to the full length molecule. Alternatively, the kinase or SH2 domain may be suppressing the apoptotic activity of SH3 domain by activating signals for cell survival. Activation of Src family kinases appears to be required for apoptosis induced by certain stimuli and it is also possible that the low level of apoptosis induced by full length Hck is due to suppression of uncontrolled kinase activity by the negative regulatory SH3 and SH2 domains. Tight regulation of the process of apoptosis is necessary and it is possible that the apoptosis inducing activity of Hck is tightly regulated, which is disregulated by deletion of the C-terminal regions. It has earlier been shown that p56lck interaction with CD4 aids HIV induced 204 Apoptosis · Vol 7 · No 3 · 2002

apoptosis and this function is not dependent on the kinase activity of the Lck molecule.45 The c-Src kinase has a role in cellular adhesion which requires both SH3 and SH2 domains but does not require catalytic activity.46 Thus it appears that apart from their catalytic activity, the Src family kinases may have kinase independent roles dependent on their protein interaction domains. Caspase-3 makes an essential contribution to cell death in many but not all pathways of apoptosis.39 Caspase3 knockout mice have shown that cells exhibit a tissue, cell type and stimulus specific dependence on caspase3 for apoptosis.47 We have shown that caspase-3 defective MCF-7 cells are resistant to apoptosis induced by

Hck SH3 mediated caspase-3 dependent apoptosis

Hck-USH3, which is restored by introducing caspase-3 gene. These results suggest that overexpression of the Hck-USH3 activates a pathway leading to induction of apoptosis that is dependent on caspase-3. This pathway is likely to be independent of caspase-1 and caspase-8 since viral protein Crm A did not inhibit USH3 induced apoptosis. Inhibition by Bcl2 also suggests that receptor mediated pathway of apoptosis which utilizes caspase-8 is not involved. In this study we have found that cytochrome c is released from mitochondria in cells undergoing apoptosis upon Hck-USH3 overexpression. This release is not seen in MCF-7 cells which lack caspase-3 suggesting that caspase-3 is required for cytochrome c release from mitochondria. This was supported by the observation that introduction of caspase-3 cDNA in MCF-7 cells restored cytochrome c release from mitochondria. Though the earlier known conventional pathway describes cytochrome c release as an upstream event in caspase-3 activation, it has recently been shown that hepatocytes from wild type mice but not caspase-3 deficient mice release cytochrome c upon Fas stimulation.48 It has been suggested that this cytochrome c release by caspase-3 is brought about by cleavage of Bcl2 and other upstream regulatory proteins.48 Caspases have been found to induce release of cytochrome c (perhaps by their action on cytosolic factors) but this has been interpreted as an amplification event (functioning in a positive feedback loop) initiated by the release of cytochrome c.49−51 However most of these data (but not all) can also be interpreted in terms of requirement of a caspase for the release of mitochondrial cytochrome c. Which caspase is required for release of cytochrome c? This is likely to be dependent on the cell type and the stimulus. Microinjection of cytochrome c in various cells induces apoptosis but not in MCF-7 cells which lack caspase-3.52 However cytochrome c induced apoptosis in MCF-7 cells was restored by transfection with caspase-3 but not by other caspases suggesting that caspase-3 is required for cytochrome c mediated apoptosis.52 It has also been shown that caspase-3 is essential for the processing of caspase-9 in MCF-7 cells.53 Our results presented here suggest that caspase-3 is required for release of cytochrome c upon over expression of USH3. These observations are not contradictory to previous reports showing a requirement of caspase-3 for cytochrome c mediated apoptosis.44,52 Thus it appears that caspase-3 may be required at two distinct steps in some apoptotic pathways, upstream as well as downstream of cytochrome c release. In those systems of apoptosis where cytochrome c is released, its function is suggested to be activation of caspase9 which then activates caspase-3 and other executioner caspases. If caspase-3 is required for the release of cytochrome c then what is the role of released cytochrome c? One possibility is that cytochrome c is released as a

consequence of apoptosis. This is unlikely since the role of cytochrome c in inducing apoptosis is well documented by microinjection experiments.52 The other possibility is that the released cytochrome c functions in a positive amplification loop to activate caspase-3. Molecules with only the N-terminal regulatory sequences of Src family kinases may be generated in vivo by proteolytic cleavage (of the kinase domain). The N-terminal SH3 and SH2 domains which would be membrane bound may serve as adaptor molecules to activate independent signaling pathways as happens in our study of overexpression of Hck USH3. Similarly overexpression of a truncated Src that lacks the kinase domain was shown to downregulate the Akt survival pathway and induce apoptosis in osteoclasts.54 Caspase-3 cleavage of Fyn and Lyn in the unique domain has been shown during Fas induced apoptosis.55 While Fyn and Lyn are cleaved by caspase-3 in vitro, Hck is not. Though the full length Hck lacks potential caspase-3 cleavage sites, sites for possible cleavage by other caspases are present raising the possibility that the truncated Hck molecules may have physiologically relevant function.

Conclusion Our results show that an apoptotic signalling pathway exists in the cell which requires caspase-3 for release of cytochrome c from mitochondria. This apoptotic pathway has the following features (a) it is induced by overexpression of SH3 and unique domains of Hck tyrosine kinase, (b) it is initiated at the inner side of plasma membrane, (c) it is p53 independent and inhibited by Bcl2, (d) it shows caspase 3 dependent cytochrome c release, and (e) it shows caspase 3 dependent apoptosis. References 1. Jacobson MD, Weil M, Raff MC. Programmed cell death in animal development. Cell 1997; 88: 347–354. 2. Barr PJ, Tonei LD. Apoptosis and its role in human disease. Biotechnology 1994; 12: 487–493. 3. Thompson CB. Apoptosis in the pathogenesis and treatment of disease. Science 1995; 267: 1456–1462. 4. Stellar H. Mechanism and genes of cellular suicide. Science 1995; 267: 1445–1449. 5. Jarpe MB, Widmann C, Knall C, et al. Antiapoptotic versus proapoptotic signal transduction checkpoints and stop signs along road to death. Oncogene 1998; 17: 1475–1482. 6. Earnshaw WC, Martins LM, Kaufmann SH. Mammalian caspases: Structure, activation, substrates and functions during apoptosis. Ann Rev Biochem 1999; 68: 383–424. 7. Slee E, Adrain C, Martin SJ. Serial killers: Ordering caspase activation events in apoptosis. Cell Death and Differen 1999; 6: 1067–1074. 8. Zheng TS, Hunot S, Kuida K, Flawell RA. Caspase knockouts: Matters of life and death. Cell Death and Differen 1999; 6: 1043– 1053. Apoptosis · Vol 7 · No 3 · 2002

205

V. Radha et al. 9. Li P, Nijhawan D, Dudihardjo I, et al. Cytochrome C and dATPdependent formation of Apaf-1/caspase-9 complex initiates an apoptotic protease cascade. Cell 1997b; 91: 479–389. 10. Nicholson DW. Caspase structure, proteolytic substrates, and function during apoptotic cell death. Cell Death and Differen 1999; 6: 1028–1042. 11. Yang C, Chang J, Gorospe M, Passaniti A. Protein tyrosine phosphatase regulation of endothelial cell apoptosis and differentiation. Cell Growth & Differen 1996; 7: 161–171. 12. Neel BG, Tonks NK. Protein tyrosine phosphatases in signal transduction. Curr Op Cell Biol 1997; 9: 193–204. 13. Radha V, Sudhakar Ch, Swarup G. Induction of p53 dependent apoptosis upon overexpression of a nuclear protein tyrosine phosphatase. FEBS Letts 1999; 453: 308–312. 14. Wei S, Liu JH, Epling-Burnette PK, et al. Critical role of Lyn kinase in inhibition of neutrophil apoptosis by GM-CSF. J Immunol 1996; 157: 5155–5162. 15. Yousefi S, Hoessli DC, Blaser K, Mills GB, Simon HU. Requirement of Lyn and Syk tyrosine kinases for the prevention of apoptosis by cytokines in human eosinophils. J Exp Med 1996; 183: 1407–1414. 16. Thomas SM, Brugge JS. Cellular functions regulated by Src family kinases. Ann Rev Cell Dev Biol 1997; 13: 513–609. 17. Simon HU, Yousefi S, Dibbert B, et al. Role of tyrosine phosphorylation and Lyn tyrosine kinase in Fas receptor mediated apoptosis in eosinophils. Blood 1998; 92: 547–557. 18. AL-Ramadi BK, Zhang H, Bothwell ALM. Cell cycle arrest and apoptosis hyper susceptibility as a consequence of Lck deficiency in nontrans formed T lymphocytes. Proc Natl Acad Sci 1998; 95: 12498–12503. 19. Ziegler SF, Marth JD, Lewis DB, Perlmutter RM. Novel protein-tyrosine kinase gene (hck) preferentially expressed in cells of hematopoietic origin. Mol Cell Biol 1987; 7: 2276– 2285. 20. Rema V, Swarup G. Molecular cloning of a membrane bound tyrosine specific protein kinase from rat spleen. Ind J Biochem Biophys 1990; 27: 433–437. 21. English BK, Ihle JN, Myracle A, Yi T. Hck tyrosine kinase activity modulates TNF production by murine macrophages. J Exp Med 1993; 178: 1017–1022. 22. Ernst M, Gearing DP, Dunn AR. Functional and biochemical association of Hck and the LIF/IL-6 receptor signal transducing subunit gp130 in embryonic stem cells. EMBO J 1994; 13: 1574–1584. 23. Ghazizadeh S, Bolen JB, Fleit HB. Physical and functional association of Src related PTKs and FCr Rll in THP-1 cells. J Biol Chem 1994; 269: 8878–8884. 24. Pawson T. Protein modules and signalling networks. Nature 1995; 373: 573–580. 25. Resh MD. Myristylation and palmitylation of Src family members. Cell 1994; 76: 411–413. 26. Robbins SM, Quintrell NA, Bishop MJ. Myristoylation and differential palmitoylation of the Hck PTKs govern their attachment to membranes and association with coveolae. Mol Cell Biol 1995; 15: 3507–3515. 27. Lock P, Ralph S, Stanley E, Bonlet I, Ramsay R, Dunn AR. Two isoforms of murine hck, generated by utilization of alternative translational initiation codons, exhibit different patterns of subcellular localization. Mol Cell Biol 1991; 11: 4363–4370. 28. Erpel T, Courtneidge SA. Src family protein tyrosine kinases and cellular signal transduction pathways. Curr Op Cell Biol 1995; 7: 176–182. 29. Gouri BSV, Rema V, Kamatkar S, Swarup G. Nucleotide sequence of a cDNA coding for rat hck tyrosine kinase and characterization of its gene product. J Biosci 1994; 19: 117–129.

206 Apoptosis · Vol 7 · No 3 · 2002

30. Kamatkar S, Radha V, Nambirajan S, Reddy RS, Swarup G. Two splice variants of a tyrosine phosphatase differ in substrate specificity, DNA binding and subcellular location. J Biol Chem 1996; 271: 26755–26761. 31. Radha V, Nambirajan S, Swarup G. Subcellular localization of a PTPase: Evidence for association with chromatin. Biochem J 1994; 299: 41–47. 32. Juin P, Hueber A, Littlewood T, Evan G. c-Myc-induced sensitization to apoptosis is mediated through cytochrome c release. Genes & Develop 1999; 13: 1367–1381. 33. Tewari M, Dixit VM. Fas and TNF-induced apoptosis is inhibited by the poxvirus CrmA gene product. J Biol Chem 1995; 270: 3255–3260. 34. Zhang J, Alter N, Reed JC, Borner C, Obeid LM, Hannun YA. Bcl-2 interrupts the ceramide-mediated pathway of cell death. Proc Natl Acad Sci 1996; 93: 5325–5328. 35. Tewari M, Quan LT, O’Rourke K, et al. Yama/CPP-32β, a mammalian homolog of CED-3, is a Crm A-inhibitable protease that cleaves the death substrate poly (ADP-ribose) polymerase. Cell 1995; 81: 801–809. 36. Adams JM, Cory S. The Bcl2 protein family: Arbiters of cell survival. Science 1998; 281: 1322–1326. 37. Newton K, Strasser A. The Bcl-2 family and cell death regulation. Curr Op Gen & Dev 1998; 8: 68–75. 38. Salvesen GS, Dixit VM. Caspases: Intracellular signaling by proteolysis. Cell 1997; 91: 443–446. 39. Porter AG, Janicke RU. Emerging roles of caspase-3 in apoptosis. Cell Death Differ 1999; 6: 99–104. 40. Garcia-Calvo M, Peterson EP, Leiting B, Ruel R, Nicholson DW, Thornberry NA. Inhibition of human caspases by peptide based and macromolecular inhibitors. J Biol Chem 1998; 273: 32608–32613. 41. Janicke RU, Sprengart ML, Wati MR, Porter AG. Caspase 3 is required for DNA fragmentation and morphological changes associated with apoptosis. J Biol Chem 1998; 273: 9357–9360. 42. Woo M, Hakem R, Soengas MS, et al. Essential contribution of caspase-3 to apoptosis and its associated nuclear changes. Genes Dev 1998; 12: 806–819. 43. Green DR, Reed JC. Mitochondria and apoptosis. Science 1998; 281: 1309–1312. 44. Zou H, Henzel WJ, Liu X, Lutschg A, Womy X. Apaf-1, a human protein homologous to C-elegans CED-4 participates in Cyt C. Dependent activation of caspase-3. Cell 1997; 90: 405–413. 45. Corbeil J, Tremblay M, Rickman D. HIV-induced apoptosis requires the CD4 receptor cytoplasmic tail and is accelerated by interaction of CD4 with p56lck . J Exp Med 1996; 183: 39–48. 46. Kaplan KB, Swedlow JR, Morgan DO, Varmus HE. C-Src enhances spreading of Src− /− fibroblasts on fibronectin by a kinase independent mechanism. Genes Dev 1995; 9: 1505–1517. 47. Kuida K, Zheng TS, Na SQ, et al. Decreased apoptosis in the brain and premature lethality in CPP 32-deficient mice. Nature 1996; 384: 368–372. 48. Woo M, Haken A, Elia AJ, et al. In vivo evidence that caspase-3 is required for fas-mediated apoptosis of hepatocytes. J Immunol 1999; 163: 4909–4916. 49. Cosulich SC, Savory PJ, Clarke PR. Bcl-2 regulates amplification of caspase activation by cytochrome C. Curr Biol 1999; 9: 147–150. 50. Bossy-Wetzel E, Green DR. Caspases induce cytochrome release from mitochondria by activating cytosolic factors. J Biol Chem 1999; 274: 17484–17490. 51. Kirsch DG, Dosef A, Chau BN, et al. Caspase-3-dependent cleavage of Bcl-2 promotes release of cytochrome C. J Biol Chem 1999; 274: 21155–21161.

Hck SH3 mediated caspase-3 dependent apoptosis 52. Li F, Anu S, Wang Y, Armstrong RC, Tomasolli KJ, Fritz LC. Cell-specific induction of apoptosis by microinjection of cytochrome c. J Biol Chem 1997a; 272: 30299–30305. 53. Blanc C, Deveraux QL, Krajewski S, et al. Caspase-3 is essential for procaspase-9 processing and cisplatin-induced apoptosis of MCF-7 breast cancer cells. Cancer Res 2000; 60: 4386– 4390.

54. Xing L, Venegas AM, Chen A, et al. Genetic evidence for a role for Src family kinases in TNF family receptor signaling and cell survival. Genes & Devp 2001; 15: 241–253. 55. Luciano F, Ricci J-E, Auberger P. Cleavage of Fyn and Lyn in their N-terminal unique regions during induction of apoptosis: A new mechanism for Src kinase regulation. Oncogene 2001; 20: 4935–4941.

Apoptosis · Vol 7 · No 3 · 2002

207