Polypyrimidine Tract-binding Proteins Are Cleaved by Caspase-3 during Apoptosis*
- From the §National Research Laboratory and the¶National Creative Research Center for Biomolecular Interaction, Department of Life Science, Division of Molecular and Life Sciences, Pohang University of Science and Technology, San31, Hyoja-Dong, Pohang, Kyungbuk 790-784, Korea
Abstract
The polypyrimidine tract-binding protein (PTB), an RNA-binding protein, is required for efficient translation of some mRNAs containing internal ribosomal entry sites (IRESs). Here we provide evidence that the addition of apoptosis-inducing agents to cells results in the cleavage of PTB isoforms 1, 2, and 4 by caspase-3. This cleavage of PTB separated the N-terminal region, containing NLS-RRM1, from the C-terminal region, containing RRM2-3-4. Our data indicate that there are three noncanonical caspase-3 target sites in PTBs, namely Ile-Val-Pro-Asp7↓Ile, Leu-Tyr-Thr-Asp139↓Ser, and Ala-Ala-Val-Asp172↓Ala. The C-terminal PTB fragments localized to the cytoplasm, as opposed to the nucleus where most intact PTBs are found. Moreover, these C-terminal PTB fragments inhibited translation of polioviral mRNA, which contains an IRES element requiring PTB for its activation. This suggests that translation of some IRES-containing mRNAs is regulated by proteolytic cleavage of PTB during apoptosis.
Apoptosis, or programmed cell death, is an essential event in many biological processes, including organ development, development of the immune system, and elimination of damaged or potentially neoplastic cells (1-3). Defective apoptosis may cause or exacerbate a variety of human diseases, including Alzheimer's and Huntington's diseases, autoimmunity, and cancer (4-6). Cells dying by apoptosis display common cytological and molecular features, regardless of the initiating signal. Cytological changes include cytoplasmic and nuclear shrinkage, plasma membrane blebbing, and chromatin condensation. At the molecular level, a family of cysteine proteases, termed caspases, selectively cleaves a series of protein substrates. Initiator caspases, such as caspases 8 and 9, trigger a proteolytic cascade that culminates in the cleavage and activation of downstream caspases such as caspases 3, 6, and 7. These downstream or execution caspases then cleave specific target proteins (7-9). Ultimately, a caspase-activated DNase I cleaves genomic DNA (10), at which point cell death is imminent.
Understanding the mechanisms of cell death requires the identification of caspase targets and the elucidation of the consequences of their proteolytic cleavage. Thus far, more than 70 proteins have been found to be cleaved by caspases (11, 12), and new substrates are continually being identified. For most proteins, the consequences of cleavage are poorly understood. However, in a few cases proteolysis of certain components can be linked to discrete morphological changes associated with cell death. Because apoptosis is an ordered sequence of rather stereotypical alternations in every cell type, one would predict that caspase substrates should be ubiquitously expressed and evolutionary conserved, at least in their aspartate cleavage site. The known substrates of caspases can be loosely categorized into a few functional groups, which include cytoskeletal and structural proteins, cell cycle proteins, signal transduction-regulatory proteins, transcription or translation-regulatory proteins, and cytokine precursors (11, 12). In addition, the recent use of proteomic approaches has found several apoptosis-associated RNA-binding proteins such as splicing factors p54nrb, SRp30c, and ASF-2, as well as heterogeneous nuclear ribonucleoproteins (hnRNPs)1such as hnRNP A1, hnRNP A2/B1, hnRNP C1/C2, hnRNP K, and hnRNP R (13-15).
HnRNPs have been shown to participate in a wide variety of processes in the nucleus, including transcriptional regulation, maintenance of telomere length, immunoglobulin class-switch recombination, alternative pre-mRNA splicing, and pre-mRNA 3′-end processing (16). The view that hnRNPs function only in the nucleus was challenged by findings that some hnRNPs shuttle between the nucleus and the cytoplasm (see Ref. 16 and references therein). Several studies have provided evidence that hnRNPs not only accompany mRNA into cytoplasm but also control the activities of mRNA in the cytoplasm. These studies show that hnRNPs can regulate at least three distinct cytoplasmic events as follows: mRNA localization, mRNA translation, and mRNA turnover.
A number of apoptosis-associated hnRNP proteins have been identified. HnRNPA2 regulates cytoplasmic mRNA localization and translation of myelin basic protein (17-20). HnRNP C increases amyloid precursor protein production by stabilizing amyloid precursor protein mRNA (21), and differentiation-linked binding of hnRNP C on c-sismRNA may regulate IRES-dependent translation (22). HnRNP K and hnRNP E1 silence the translation of 15-lipoxygenase mRNA by inhibiting 80 S ribosome assembly (23). Thus, the control of mRNA localization, translation, and mRNA turnover by hnRNPs is a powerful means to regulate protein expression. The regulation of apoptosis may require a fast response, which could be provided by modifying protein expression rather than by de novosynthesis of mRNA. Therefore, post-translational modifications of hnRNPs, such as phosphorylation and proteolysis, might be used to regulate their functions in apoptotic cells.
Polypyrimidine tract-binding protein (PTB also known as p57 and heterogeneous nuclear ribonucleoprotein I (hnRNP I)) is a member of the hnRNP family and shuttles between the nucleus and the cytoplasm in a transcription-sensitive manner (24). Three isoforms of PTB have been reported (25-27). The prototype of PTB (PTB1) consists of 531 amino acids with a molecular mass of 57 kDa. PTB2 and PTB4 have insertions of 19 and 26 amino acids, respectively, after the amino acid at position 298 of PTB1 (25, 27). Four loosely conserved RNA-recognition motifs (RRMs) are distributed throughout the PTB molecule (see Fig.3 A) (28). PTB exists in oligomeric as well as monomeric forms, and the central region of PTB, including RRM2, plays a key role in its oligomerization (29, 30). PTB contains several RNA-binding domains, among which the C-terminal region of PTB spanning RRM3–4 exerts the strongest RNA binding activity (29-31). The N-terminal region of PTB is responsible for the enhancement of RNA binding activity by HeLa cell cytoplasmic factor(s) (29). PTB was originally identified as a protein binding to the polypyrimidine tracts of adenoviral major late and α-tropomyosin pre-mRNAs and was proposed as a splicing factor (25, 27). Binding of PTB to the polypyrimidine tracts near the branch point of an intron was shown to modulate the alternative splicing of certain pre-mRNAs (32-34). Independently, PTB was shown to interact specifically with the IRESs of several viral and cellular mRNAs (35-45). PTB can either stimulate or inhibit the activity of several viral (46-54) and cellular (43, 44) IRES elements.
Caspase-3 cleaves PTB. A andB, purified recombinant PTB4 (5.0 μg) was incubated with 250 ng of caspase-3 in the presence or absence of 10 μm DEVD-CHO at 37 °C for the indicated times. Reactions were stopped by the addition of 2× Laemmli sample buffer, and samples were analyzed by immunoblotting using anti-PTB monoclonal antibodies (DH17 for A and DH3 for B).C, [35S]methionine-labeled PTB isoforms were incubated with 160 ng of caspase-3 for the indicated times. “Negative control” samples were incubated for 2 h without addition of caspase-3 (lanes 1, 7, and13). Samples were resolved by SDS-PAGE, and the protein bands were detected by autoradiography. The positions of PTB cleavage products are indicated by open arrow, solid arrow, open arrowhead, and solid arrowhead.
The goal of this study was to determine whether PTBs were targets for caspase proteolysis during apoptosis. The proteolytic cleavage of RNA-binding proteins was analyzed by Western blotting, which revealed that PTB was cleaved by caspase-3 in vivo and in vitro, even though only loosely conserved caspase recognition sequences exist in the primary structure of the protein. Caspase-3 cleavage sites in PTB were identified using Edman degradation and confirmed by site-directed mutagenesis. The C-terminal PTB cleavage products yielded by apoptosis were redistributed to the cytoplasm, away from the nucleus where intact PTBs are mainly located. Moreover, the C-terminal region of PTB inhibited translation of polioviral mRNA containing a PTB-dependent IRES element. This indicates that proteolytic cleavage of PTB, an RNA-binding protein that augments IRES-dependent translation, can contribute to drastic changes in gene expression during apoptosis.
EXPERIMENTAL PROCEDURES
Antibodies
Mouse anti-PTB monoclonal antibodies (mAbs) DH3 and DH17 were provided by Dr. E. Wimmer (State University of New York, Stony Brook). The epitopes recognized by these antibodies are between amino acids 1–159 and 292–532, respectively. Anti-caspase-3 polyclonal antibody, anti-hnRNP A1 polyclonal antibody, and anti-hnRNP C1/C2 mAb were provided by Dr. P. Suh (POSTECH, Korea) (55), Dr. M. M. C. Lai (University of Southern California School of Medicine), and Dr. G. Dreyfuss (University of Pennsylvania), respectively. Anti-PARP mAb, anti-green fluorescence protein (anti-GFP) polyclonal antibody, anti-red fluorescence protein (anti-RFP) polyclonal antibody, and rhodamine (TRITC)-conjugated goat anti-mouse IgG were purchased from BD PharMingen, Santa Cruz Biotechnology, CLONTECH, and The Jackson Laboratories, respectively.
Cell Culture, Transfection, and Induction of Apoptosis
Molt-4, Jurkat, and HeLa-S3 cells were routinely cultivated in RPMI 1640 (Invitrogen) supplemented with 10% fetal bovine serum (HyClone). 293T and HeLa-E cells were cultured in Dulbecco's modified Eagle's medium (Invitrogen) supplemented with 10% fetal bovine serum. HeLa-S3 and 293T cells were transfected with DNA by electroporation. HeLa-E cells were seeded on glass coverslips 12 h before transfection with LipofectAMINE (Invitrogen). For induction of apoptosis, cells were incubated in medium for 4–12 h in the presence of the following agents: 20 μm etoposide (Sigma) for Molt-4 and Jurkat, 50 μm etoposide for HeLa-S3, 5 μm camptothecin (Sigma) for Jurkat and HeLa-S3, 2 μm staurosporine for Jurkat and HeLa-S3, 100 ng/ml TRAIL in the presence of 1.5 μg/ml potentiator (Upstate Biotechnology Inc.) for Jurkat and HeLa-S3, and 100 ng/ml TNF-α (Roche Molecular Biochemicals) with 10 μg/ml cycloheximide for HeLa-E. For inhibition of caspases, cells were preincubated for 1 h with various concentrations (0, 0.02, 0.2, 1, 5, 20, or 50 μm) of Z-DEVD-fmk or Z-VAD-fmk (Enzyme Systems Products, Dublin, CA) before treatment with apoptotic inducers. In order to detect changes in RNA-binding proteins during apoptosis, apoptosis inducer-treated cells were washed with ice-cold phosphate-buffered saline and stored at −70 °C. Cell pellets were thawed on ice, suspended in phosphate-buffered saline, sonicated, and the cell lysates were centrifuged at 10,000 × g for 10 min. Thirty micrograms of cell lysate was electrophoresed on SDS-polyacrylamide gels for immunoblotting of hnRNP I (PTB), hnRNP A1, hnRNP C1/C2, PARP, and caspase-3.
Expression and Purification of Caspase-3 and PTB4
The strategies for cloning and purification of the recombinant human caspase-3 and PTB4 are described elsewhere (56, 57).
Plasmid Construction
pTM1H/PTB1 (29), pTM1H/PTB2, and pTM1H/PTB4 (57) are described elsewhere. To construct pTM1H/PTB4(Δ2), with deletion of aa 1 and 2 (Met-Asp), a cDNA fragment corresponding to aa 3–325 of pTM1H/PTB4 was amplified by PCR using primer P11-Δ2 and primer B (see Table Iand Ref. 57). The PCR fragment and pTM1H/PTB4 were digested withEcoRI and SacII and then ligated to produce pTM1H/PTB4(Δ2).
Oligonucleotides used in this study
Site-directed mutagenesis was performed by PCR. All of the inserted PCR fragments were completely sequenced. To construct pSK/PTB4-(1–325) (D139A), DNA fragments were amplified from pTM1H/PTB4. The following four primers were used: primer A (57) and primer P7-SacI for cDNA fragment 1 corresponding to the region aa 1–136, and primer P6-D139A and primer B for cDNA fragment 2 corresponding to the region aa 135–325. The cDNA fragment 1 treated withSacI and EcoRI and the cDNA fragment 2 treated with SacI and PflMI were inserted into pSK/PTB4-(1–325) (Q321A) (57) treated with EcoRI andPflMI. To construct pSK/PTB4-(1–325) (D172A), insert DNA fragments were amplified by PCR from pTM1H/PTB4. The primers used in the PCR are as follows: primer A and primer P9-D172A for cDNA fragment 3 corresponding to the region aa 1–173, and primer P8-D172A and primer B for cDNA fragment 4 corresponding to the region aa 172–325. The cDNA fragment 3 treated with PvuII andEcoRI and the cDNA fragment 4 treated withPvuII and PflMI were inserted into pSK/PTB4-(1–325) (Q321A) treated with EcoRI andPflMI. cDNA fragment 5 corresponding to the region aa 135–325 with D139A/D172A mutation was produced by PCR using primer P6-D139A and primer B from pSK/PTB4-(1–325) (D172A). The cDNA fragment 1 treated with SacI and EcoRI and the cDNA fragment 5 treated with SacI and PflMI were inserted into pSK/PTB4-(1–325) (Q321A) treated withEcoRI and PflMI to construct pSK/PTB4-(1–325) (D139A/D172A). Primers B and P11-Δ2 were used in PCRs generating cDNA fragment 6 (aa 3–325, D139A) from pSK/PTB4-(1–325) (D139A), cDNA fragment 7 (aa 3–325, D172A) from pSK/PTB4-(1–325) (D172A), and cDNA fragment 8 (aa 3–325, D139A/D172A) from pSK/PTB4-(1–325) (D139A/D172A). The cDNA fragments 6–8 treated with SacII and EcoRI were inserted into pTM1H/PTB4 treated with SacII andEcoRI to construct pTM1H/PTB4(Δ2) (D139A), pTM1H/PTB4(Δ2) (D172A), and pTM1H/PTB4(Δ2) (D139A/D172A), respectively. Primer B and P10-D7A(Δ2) were used in generating cDNA fragment 9 (aa 3–325 with D7A mutation) from pTM1H/PTB4 and cDNA fragment 10 (aa 3–325 with D7A/D139A/D172A mutation) from pTM1H/PTB4(Δ2) (D139A/D172A). The cDNA fragments 9 and 10 treated with SacII and EcoRI were inserted into pTM1H/PTB4 treated with SacII and EcoRI to construct pTM1H/PTB4(Δ2) (D7A) and pTM1H/PTB4(Δ2) (D7A/D139A/D172A), respectively.
pEGFP/PTB4 is described elsewhere (57). To construct pEGFP/PTB4-(173–557), the PflMI-PvuII fragment of pSK/PTB4-(1–325) (D172A) was inserted into pEGFP/PTB4 treated withEcoRI-Klenow-PflMI. To construct pEGFP/PTB4-(Δ60–173), pEGFP/PTB4 DNA treated withXbaI-Klenow-NdeI for PTB4-(1–59) was inserted into pEGFP/PTB4-(173–557) treated withHindIII-Klenow-NdeI.
Plasmid pEGFP-RFP was constructed by inserting pEGFP-CI (CLONTECH) DNA treated withEcoRI-Klenow-NdeI into pDsRED-NI (CLONTECH) treated withXhoI-Klenow-NdeI. The PTB4 gene and its derivatives in plasmids pTM1H/PTB4(Δ2), pTM1H/PTB4(Δ2) (D139A), pTM1H/PTB4(Δ2) (D172A), and pTM1H/PTB4(Δ2) (D139A/D172A) were inserted into pEGFP-RFP to construct pEGFP/PTB4(Δ2)/RFP, pEGFP/PTB4(Δ2)/RFP (D139A), pEGFP/PTB4(Δ2)/RFP (D172A), and pEGFP/PTB4(Δ2)/RFP (D139A/D172A), respectively.
Dicistronic plasmid pHR/Polio/F was described by Back et al.(57).
In Vitro Transcription and Translation
Transcription reactions were performed with T7 RNA polymerase (Roche Molecular Biochemicals) at 37 °C for 90 min, as recommended by the manufacturer. The concentration of RNA transcripts was determined using a UV spectrophotometer. Plasmids pTM1H/PTB1, pTM1H/PTB2, pTM1H/PTB4, pTM1H/PTB4(Δ2) (D7A), pTM1H/PTB4(Δ2) (D139A), pTM1H/PTB4(Δ2) (D172A), pTM1H/PTB4(Δ2) (D139A/D172A), and pTM1H/PTB4(Δ2) (D7A/D139A/D172A) digested with SalI were used to generate mRNAs encoding PTB1, PTB2, PTB4, and PTB4 derivatives, respectively. In vitro translations in micrococcal nuclease-treated RRL (Promega) were performed in reaction mixtures containing 30 nm mRNA. Translation reactions were carried out at 30 °C for 1 h in the presence of [35S]methionine (PerkinElmer Life Sciences). Translation reactions were stopped by adding 0.6 mg/ml cycloheximide and 1 mm “cold” methionine.
Determination of Caspase-3 Target Sites on PTB4
Purified PTB4 (5 μg) and 250 ng of caspase-3 were incubated at 37 °C for the indicated time in buffer D (50 mm HEPES, pH 7.5, 10% sucrose, 0.1% CHAPS, and 10 mm dithiothreitol) with or without 10 μm DEVD-CHO (Calbiochem). The reactions were stopped by adding 2× Laemmli sample buffer, and the samples were resolved on 15 or 10% SDS-polyacrylamide gels. PTB-related bands were detected by Western blot analysis using anti-PTB mAbs (DH3 or DH17).
To investigate the caspase-3 activity against PTBs translated in vitro, the radioactive translation mixture (10 μl) was incubated with purified caspase-3 (160 ng) resuspended in buffer D for 2 h (or indicated time) at 37 °C. The caspase reactions were stopped by adding 2× Laemmli sample buffer, and the samples were analyzed by 13% SDS-PAGE followed by autoradiography.
For N-terminal amino acid sequencing of PTB cleavage products, purified PTB4 (40 μg) was incubated with caspase-3 (2 μg) at 37 °C for 30 min or 10 h. The reactions were stopped by adding 2× Laemmli sample buffer. The samples were resolved by 10 or 15% SDS-PAGE and then transferred to polyvinylidene difluoride membranes. The Coomassie Blue-stained protein bands were subjected to microsequencing by an automatic protein sequencer (Applied Biosystems, Inc.) at Korea Basic Science Institute.
Fluorescence Microscopic Analysis
The microscopic observations of endogenous PTB and GFP-fused proteins were accomplished as described previously (57).
Dual-luciferase Reporter Assay
Dual-luciferase assay was carried out as described previously (57).
RESULTS
PTBs Are Cleaved during Apoptosis
Modification of hnRNPs during apoptosis was investigated by Western blot analysis using antibodies against hnRNP I/PTB, hnRNP A1, hnRNP C1/C2, PARP, and caspase-3. Two different monoclonal antibodies were used, DH17 and DH3, that recognize the C- and N-terminal regions of PTB, respectively. Total cell lysates of Molt-4 cells were prepared at various times after induction of apoptosis by etoposide. After 4 h of treatment with etoposide, three ∼41-kDa fragments of PTBs were identified by monoclonal antibody DH17 (Fig.1 A, lane 2), and these fragments were more apparent 6 h post-treatment (Fig.1 A, lanes 2 and 3). Intact PTBs almost disappeared 12 h post-treatment with etoposide (Fig.1 A, lane 4). Monoclonal antibody DH3 detected two or three N-terminal fragments of PTBs (∼15 kDa). Kinetics of the N-terminal fragment generation was similar to that of the C-terminal fragment (compare A with B in Fig. 1). The cleavage of PTBs was inhibited when apoptosis was blocked by pretreatment with Z-VAD-fmk, a broad spectrum caspase inhibitor (Fig.1, A and B, lane 7). Thus, it is very likely that the PTB cleavage is mediated by a caspase. As a control marker of caspase activity, the cleavage of the well characterized caspase-3 and -7 target protein PARP was also measured (Fig.1 E) (7, 58, 59). Similar to PTBs, cleavage of PARP with apparent molecular mass of ∼85 kDa was observed 4 h after etoposide treatment. PARP was completely cleaved 6 h after etoposide treatment, by which time about half the PTBs were cleaved. The amount of full-length caspase-3 decreased gradually as apoptosis proceeded. Etoposide-induced proteolysis of PTBs occurred with approximately the same kinetics as the cleavage of procaspase-3 (compare A and B with F in Fig. 1). In accordance with published data, we also observed the cleavage of heteronuclear ribonucleoprotein (hnRNP) A1 (C in Fig. 1) (13, 14, 60) and hnRNP C1/C2 (D in Fig. 1) (13, 15). However, the efficiency of the cleavages of hnRNP A1 and hnRNP C1/C2 was lower than those of PTBs (compare A and Bwith C and D in Fig. 1).
PTBs are cleaved during apoptosis.Molt-4 cells were incubated with the apoptosis-inducer etoposide (20 μm) for the indicated times, and total cell lysates were analyzed by immunoblotting using antibodies against PTB (Aand B), hnRNPA1 (C), hnRNPC1/C2 (D), PARP (E), and caspase-3 (F). Where indicated, cells were pretreated with the protease inhibitor Z-VAD-fmk (50 μm) for 1 h prior to induction of apoptosis (lane 7). The PTB fragments generated by the induction of apoptosis are indicated by three arrows (open arrow, solid arrow, and broken arrow) in A. Time depicts the duration of etoposide treatment.
PTB Cleavage Is a General Apoptotic Event Dependent on Caspase Activity
In order to investigate whether PTB cleavage is restricted to etoposide-mediated apoptosis in Molt-4 cells, we analyzed PTB cleavage in other cell types using different apoptosis agents. Treatment of Molt-4 (data not shown), Jurkat cells, and HeLa-S3 cells with topoisomerase inhibitors camptothecin and etoposide, a potent general kinase inhibitor staurosporine, tumor necrosis factor-α (TNF-α) (data not shown), or with TNF-related apoptosis-inducing ligand (TRAIL) induced cleavage of PTBs. PTB fragments had the same apparent sizes even though fragmentation efficiencies were different (Fig. 2 A). As shown by the apoptosis-specific cleavage of PARP (Fig. 2 A), the extent of PTB proteolysis correlated well with the potency of agents to cause cell death. Topoisomerase inhibitors were less efficient than TRAIL or staurosporine in inducing apoptosis under the conditions used. These data suggest that PTB cleavage is a general molecular event during apoptosis.
PTB cleavage is a general apoptotic event. A, different cell lines were treated for 12 h with the indicated apoptosis-inducing agents, after which lysates were prepared and analyzed by immunoblotting using antibodies against PARP or PTB (mAb DH17). B, HeLa-S3 cells were pretreated with Z-DEVD-fmk at the indicated concentrations for 1 h and then with staurosporine (2 μm) for 12 h. Cell lysates were immunoblotted with either anti-PTB monoclonal antibody (DH17) or anti-PARP monoclonal antibody. The PTB fragments generated by apoptosis are indicated by arrows.
It is known that caspase-3 is responsible for the cleavage of many proteins (7, 61). In order to evaluate whether caspase-3 is responsible for PTB cleavage during apoptosis, HeLa-S3 cells were pretreated with the group II caspase (caspase-3, -7, and -2) inhibitor Z-DEVD-fmk (Fig.2 B). Cleavage of PTBs and PARP was monitored by Western blot analysis. Preincubation of cells with Z-DEVD-fmk inhibited proteolytic processing of PARP and PTB. Cleavage of PTBs by staurosporine was almost completely inhibited at high concentrations of Z-DEVD-fmk (50 μm) (Fig. 2 B). Similar patterns of inhibition were observed in Molt-4 cells treated with etoposide (data not shown). These results indicate that apoptotic cleavage of PTBs is sensitive to group II caspase (caspase-3, -7, and -2) inhibitors.
PTBs Are Cleaved by Caspase-3
The caspase inhibitor study shown in Fig. 2 B suggests that caspase-3 may be responsible for the cleavage of PTBs during apoptosis, even though the amino acid motif DEXD (or weaker target DXXD) reported as the caspase-3 target sequence (62, 63) is absent in PTBs. To investigate further a possible involvement of caspase-3, HeLa-S3 cell extracts were incubated with purified human recombinant caspase-3, and the cleavage pattern of PTBs was monitored by Western blot analysis using monoclonal antibodies (DH17 and DH3) against PTB. The PTB cleavage pattern was the same as the etoposide cleavage pattern (data not shown). In order to test whether caspase-3 directly cleaves PTB, in vitro cleavage reactions were performed using purified PTB4 and caspase-3 proteins. When purified PTB4 (5 μg) was incubated with purified caspase-3 (250 ng), PTB4 was completely cleaved within 6 min (Fig.3, A and B,lane 4). The size of the C-terminal cleavage fragment, which was detected by monoclonal antibody DH17, was about 41 kDa (indicated by the open arrow in Fig. 3 A). This suggests that caspase-3 cleaves PTB4 near the N terminus of the protein. In addition, the larger fragment (open arrow in Fig. 3 A) was converted to a smaller fragment (solid arrow in Fig.3 A) as the reaction proceeded. This result suggests that PTB4 is initially cleaved to an ∼41-kDa polypeptide, and then the C-terminal region is further processed to an ∼40-kDa polypeptide. This progressive cleavage of PTB would help explain why multiple PTB-related polypeptides were detected by antibodies in apoptotic cell lysates (Fig. 1 A). Another likely reason for the complex pattern is the presence of three different PTB isoforms, 1, 2, and 4 (25-27). The two-step cleavage of PTB was also detected by monoclonal antibody DH3 that recognizes the N-terminal region of PTB. An ∼15-kDa fragment (open arrowhead in Fig. 3 B) was detected early in the incubation (0.5 min) and was gradually converted to a polypeptide of 14 kDa (closed arrowhead in Fig.3 B) as the reaction continued. These PTB cleavage patterns generated by mixing purified PTB and caspase-3 are very similar to those generated during cellular apoptosis (Fig. 1 B).
The cleavage patterns of different PTB isoforms were also investigated using purified caspase-3 and 35S-labeled PTB1, -2, and -4 generated by in vitro translation. All PTB isoforms were almost completely cleaved by caspase-3 within 30 min (Fig.3 C, lanes 5, 6, 11,12, 17, and 18). The two major cleavage products of ∼40 and ∼15 kDa, indicated by the solid arrow and the open arrowhead, and two minor polypeptides of ∼41 and ∼14 kDa, indicated by the open arrow and the solid arrowhead, were generated by caspase-3 in all three isoforms. The minor ∼41-kDa band disappeared as the incubation time extended (Fig. 3 C, open arrow). In contrast, the intensity of an ∼14-kDa minor band (solid arrowhead) increased as the incubation time was extended (Fig. 3 C, lanes 5, 6,11, 12, 17, and 18). These in vitro cleavage data strongly suggest that PTBs are cleaved by caspase-3 in at least three different sites (see below), two sites being highly susceptible to caspase-3 action and one less so.
Determination of Caspase-3 Target Sites in PTB
In order to determine the caspase-3 target sites in PTBs, purified recombinant PTB4 was incubated with purified caspase-3, and the cleavage products were resolved by SDS-PAGE. The cleavage products of ∼41, ∼40, ∼15, and ∼14 kDa from PTB4 shown in Fig. 3, A andB, were subjected to automated Edman degradation to determine their N-terminal sequences. Five residues were unambiguously assigned for each fragment, and their sequences exactly matched sequences within PTB4. The ∼41-kDa fragment of PTB4 matched the amino acid sequence from Ser140 (solid arrow in Fig.4 A), and the ∼40-kDa fragment matched Ala173, which is located between RRM1 and RRM2 (open arrow in Fig. 4 A). The sequence of the ∼15-kDa fragment started from an alanine that is the N-terminal amino acid of PTB4 generated by the pT7-7 expression vector. The sequence of the ∼14-kDa fragment started from Ile8 that is located immediately before the nuclear localization signal (NLS) (solid arrowhead in Fig. 4 A). Note that two cleavage sites reside between RRM1 and RRM2 (Fig. 4 B). Therefore, the C-terminal fragment of PTB contains the protein-protein interaction domain (RRM2) (29, 30) and the major RNA binding domain (RRM3 and -4) (29-31). The PTB polypeptide sequence is evolutionarily conserved, with pig, rat, mouse, and Xenopus PTB showing 97, 95, 89, and 82% identity to human PTB, respectively (Fig. 4 A). Other than the weak cleavage site in Xenopus, where glutamine instead of proline is at the P2 position to the scissile bond, the caspase-3 recognition sites preceding the cleavage sites are well conserved in these PTB homologues. Therefore, it is conceivable that proteolytic cleavage of PTB by caspase-3 during apoptosis may play an important physiological role.
Determination of caspase-3 target sites in PTB. A, evolutionary conservation of caspase cleavage sites in PTB homologues. The N-terminal regions of PTB homologues were aligned. The conserved caspase recognition sites identified by peptide sequencing are indicated by a solid arrowhead, a solid arrow, and an open arrow. hPTB4, pPTB4, rPTB4,mPTB1, and xPTB4 indicate human PTB4, pig PTB4, rat PTB4, mouse PTB1, and Xenopus PTB4, respectively. The NCBI protein accession numbers for these proteins are as follows: hPTB4, CAA46444; pPTB4, CAA63597; rPTB4,Q00438; mPTB1, P17225; xPTB4, AAF00041. B, schematic diagram of caspase-3 cleavage sites in PTBs. The amino acid sequences around caspase-3 cleavage sites in PTBs are shown in single letter symbols of amino acids. The NLS and RRMs in PTB are indicated byopen and solid boxes, and the sizes of the cleaved products are shown in kDa. C, [35S]methionine-labeled PTB4 and its mutants were incubated with (+) or without (−) caspase-3 at 37 °C for 2 h. Protease reactions were stopped by the addition of 2× Laemmli sample buffer, and the samples were analyzed by SDS-PAGE followed by autoradiography. D, HeLa-S3 cells were transfected with plasmids expressing GFP/PTB4/RFP or GFP/PTB4 mutants/RFP, and after 40 h cells were mock-treated or treated with staurosporine (2 μm) for 12 h. Cell lysates were electrophoresed on SDS-polyacrylamide gels before immunoblot analysis for PARP, GFP, and RFP. The positions of cleavage products of transiently expressed GFP/PTB4/RFP and GFP/PTB4 mutants/RFP are indicated by an open arrow, a solid arrow, an open arrowhead, and a solid arrowhead.
Confirmation of Caspase-3 Target Sites by Site-directed Mutagenesis
Caspase-3 requires an aspartate residue at the P1 position to the scissile bond (62, 63). To assess whether aspartates at the 7th, 139th, and 172nd residues in PTB4 are essential for cleavage by caspase-3, we mutated the corresponding aspartate residues to alanine residues. Five different PTB4 mutant proteins were synthesized using micrococcal nuclease-treated RRL in the presence of [35S]methionine. The cleavage patterns of PTB and its derivatives by caspase-3 are shown in Fig. 4 C. Caspase-3 cleaved PTB4 into ∼40-, 15-, and 14-kDa fragments, as indicated by the arrows (c, f, and g) on lane 2 in Fig. 4 C. The ∼41-kDa fragment was not detected in this experiment because of the complete cleavage of PTB at the Asp172/Ala173 site. The cleavage of mutant PTB4 containing Asp7 to Ala (D7A) produced polypeptides of 40- and 15-kDa (arrows, c and f,lane 4 in Fig. 4 C) but not a 14-kDa fragment (arrow g). This finding indicates that the D7A mutation blocks the processing of the 15-kDa fragment to the 14-kDa fragment. PTB4 D139A yielded three cleavage products of 40, 18, and 16 kDa (as indicated by the arrows c–e in Fig. 4 C,lane 6). The cleavage at Asp172/Ala173 site generated ∼40- and ∼18-kDa fragments (arrows c and d), and a subsequent cleavage at Asp7/Ile8 created an ∼16-kDa fragment (arrow e). Mutation of Asp172to Ala (D172A) produced 41-, 15-, and 14-kDa fragments, as indicated bybands at arrows b, f, andg in lane 8 of Fig. 4 C. The 40-kDa polypeptide (arrow c) is absent from the cleavage products. The double mutant PTB4 (D139A/D172A) was cleaved into a fragment that was slightly smaller than the precursor, as indicated by arrow a at lane 10 in Fig. 4 C. No other prominent band was detected. The smaller fragment is likely to be generated by cleavage at Asp7/Ile8. Finally, protein cleavage by caspase-3 was abolished when a triple mutation (D7A/D139A/D172A) was introduced into PTB4 (lane 12 in Fig.4 C).
Similar results to those described above were generated in staurosporine-induced in vivo cleavage assays using PTB4 and mutant PTB4s. DNA encoding PTB4 (which was fused with GFP and RFP at the N terminus and the C terminus, respectively) and its derivatives was transfected into HeLa-S3, and PTB cleavage was monitored by Western blot analysis. Cleavage of endogenous PARP and PTBs by staurosporine-induced apoptosis was monitored by anti-PARP antibody (upper panel in Fig. 4 D) and anti-PTB antibody (DH17) (data not shown), respectively. Following the addition of staurosporine, PARP was almost completely cleaved at a time when about half of the endogenous PTBs were cleaved (upper panel in Fig. 4 D). Wild-type GFP/PTB4/RFP was converted to several fragments, as indicated by a solid arrow in the middle panel and a solid arrowhead in the lower panel of Fig. 4 D. The ∼42-kDa fragment detected by anti-GFP antibody corresponds to the N-terminal region of PTB4 tagged with GFP. Interestingly, the cleavage of PTB4 at the weak caspase-3 target site (Asp7/Ile8) was not apparent in the GFP/PTB4/RFP fusion proteins (lanes 2, 4,6, and 8 of middle panel in Fig.4 D). This is likely to be due to steric hindrance by the bulky GFP protein preventing binding of caspase-3 to the target site in the N-terminal fusion proteins. A protein of about 68 kDa was detected by anti-RFP antibody, which corresponds to the C-terminal domain of PTB4 fused with RFP. The Asp139 to Ala mutant (D139A) produced the slightly larger N-terminal fragment (indicated by theopen arrow) than wild-type PTB4 (indicated by thesolid arrow). In contrast, the D172A mutant yielded the same N-terminal fragment as wild-type PTB4 (compare lane 6 withlane 2 in the middle panel of Fig.4 D). The cleavage efficiency of the GFP/PTB4 (D139A)/RFP mutant was lower than that of the wild-type protein or the D172A mutant PTB4 (note the ratio of substrate versus cleavage product). This indicates that caspase-3 cleaves the Asp139/Ser140 site better than the Asp172/Ala173 site. This is consistent with thein vitro cleavage data shown in Fig. 3 A. The effect of PTB4 mutations on cleavage by caspase-3 was also monitored by Western blot analysis using an antibody against RFP. The D139A mutant produced the same C-terminal fragment as wild-type PTB4 (comparelane 4 with lane 2 in the lower panelof Fig. 4 D). On the other hand, the C-terminal product of D172A was larger (indicated by the arrowhead) than that of wild-type PTB4 (compare lane 6 with lane 2 in thelower panel in Fig. 4 D) The cleavage of PTB4 during apoptosis was abolished when a double mutation (D139A/D172A) was introduced into PTB4 (lane 8 in Fig. 4 D). These data indicate that two caspase-3 target sites, and possibly all three target sites in endogenous PTBs, identified by the in vitroproteolysis of purified proteins are cleaved in vivo during apoptosis.
Redistribution of PTB Fragments in Apoptotic HeLa Cells
During apoptosis, PTBs are mainly divided into two parts as follows: the N-terminal domain, which contains the NLS and RRM1, and the C-terminal domain, which contains RRM2-3-4 (Fig. 4 B). Because the C-terminal fragment lacks the NLS, it is likely to be distributed to the cytoplasm during apoptosis. This was investigated by examining apoptotic HeLa-E cells with a C-terminal specific monoclonal antibody (DH17). In non-apoptotic cells, PTB is mainly localized in the nucleus (Fig. 5 A,V˙apanel 1), which is attributed to the NLS residing in the first 55 amino acids (30, 64, 65). However, PTB-related fluorescence was detected in both cytoplasm and nucleus of apoptotic cells that were treated with TNF-α and cycloheximide (comparepanel 1 with panel 4 in Fig. 5 A).
Subcellular localization of PTB and PTB4/RRM2-3-4. A, the effect of apoptosis induction on PTB localization. HeLa-E cells were mock-treated or treated with TNF-α (100 ng/ml) and cycloheximide (CHX) (10 μg/ml) for 9 h and then stained with antibody DH17. Theleft panels show cells examined using a TRITC filter; themiddle panels show nuclei of cells stained with Hoechst 33258 through a DAPI filter, and right panels are the merged image of the TRITC and DAPI images. B, effects of N-terminal truncation of PTB on its localization. HeLa-E cells were transfected with plasmids expressing GFP-fused with full-length PTB4 (panels 1–3), GFP/PTB4-RRM2-3-4 (panels 4–6), and GFP/PTB4-Δ60–173 (panels 7–9). Fluorescence from GFP was observed using a fluorescein isothiocyanate filter (panels 1, 4, and 7). To visualize nuclei, DNA was stained with Hoechst 33258 and examined using a DAPI filter (panels 2, 5, and 8).Panels 3, 6, and 9 are the merged images.
The distribution pattern of the PTB C-terminal fragment was investigated using GFP fusion techniques. GFP was fused to the N termini of both full-length PTB4 and a truncated PTB containing the C-terminal part of PTB generated by caspase-3. When the DNA constructs were transfected into HeLa-E cells, proteins of the expected sizes were detected by Western blotting (data not shown). Fluorescent microscopic analysis of the cells showed that GFP-tagged full-length PTB4 (GFP/PTB4) localized mainly in the nucleus (panel 1 in Fig.5 B), whereas GFP/PTB4-RRM2-3-4 partially distributed to the cytoplasm (panel 4 in Fig. 5 B). However, GFP/PTB4-(Δ60–173) contains the NLS (aa 1–59) localized in the nucleus, similar to GFP/PTB4 (panel 7 in Fig.5 B). These data suggest that the C-terminal region of PTB is partially redistributed to the cytoplasm from the nucleus after proteolytic cleavage by caspase-3.
Effects of PTB4-RRM2-3-4 Fragment on Viral IRES-dependent Translation
Two unrelated functions have been ascribed to PTB as follows: one is the regulation of pre-mRNA splicing in the nucleus, and the other is the translational modulation of some IRES-dependent mRNAs in the cytoplasm. We investigated the effect of PTB cleavage on PTB-dependent translation using the polioviral IRES as a model system, because the concentration of C-terminal PTB fragment (PTB/RRM2-3-4) is dramatically increased in the cytoplasm during apoptosis (panel 4 in Fig. 5 A). We cotransfected a reporter plasmid yielding a dicistronic mRNA composed sequentially of Renilla luciferase (RLuc), polioviral IRES, firefly luciferase (FLuc), and effector plasmids producing PTB4 and PTB4/RRM2-3-4. The cap-dependent and viral IRES-dependent translations were monitored byRenilla luciferase activity and firefly luciferase activity, respectively. Cotransfection of full-length PTB4 augmented the translation of polioviral IRES-dependent translation in a dose-dependent manner (bars 2–4 in Fig.6) (57). In contrast, truncated PTB4 (GFP/PTB4-RRM2-3-4) strongly inhibited the translation of polioviral mRNA in a dose-dependent manner, without affecting cap-dependent translation (bars 5–7 in Fig. 6). Interestingly, GFP/PTB4-(Δ60–173) containing an NLS did not affect the translation of polioviral mRNA (bars 8–10 in Fig.6). This is probably due to the nuclear localization of the PTB4-(Δ60–173) protein (panel 7 in Fig. 5 B). These data indicate that the C-terminal domain of PTB, which is liberated during apoptosis, has a dominant negative effect on polioviral mRNA translation.
The effects of PTB and its derivatives on polioviral IRES activity. The reporter plasmid expresses a dicistronic mRNA consisting of the Renilla luciferase (RLuc) gene at the first cistron and the poliovirus IRES-firefly luciferase (FLuc) gene at the second cistron. Translation of RLuc and FLuc is accomplished by scanning mechanism and IRES-dependent mechanism, respectively. The effector plasmids produce GFP, GFP-fused full-length PTB4, GFP-fused PTB4-RRM2-3-4, or GFP-fused PTB4-(Δ60–173). 293T cells were cotransfected with 0.75 μg of the reporter plasmid and effector plasmids as indicated in the chart at the bottom. The total amount of effector plasmids was kept the same in each transfection by changing the amount of control plasmid pEGFP-C1. Forty eight hours after transfection, Renilla luciferase and firefly luciferase activities were measured as described under “Experimental Procedures,” and the relative ratio of firefly luciferase to Renilla luciferase activity in each cell lysate was calculated. The columns and barsrepresent the means and S.D. of three independent transfection experiments. The numbers in the chart represent μg of DNA.
DISCUSSION
The cleavage of PTBs during apoptosis is a recent finding (66), and prior to this study little appeared known about this cellular event. Here we show that cleavage of PTBs occurs in cells following the addition of a variety of apoptosis-inducing agents and that caspase-3 is responsible for the cleavage. Three caspase-3 target sites were identified in PTBs by experiments in both in vitro andin vivo systems. The current work also showed the distribution patterns of PTB fragments in the cell and the effect of PTB fragments on IRES-dependent translation.
By using site-directed mutagenesis studies, we showed that three caspase-3 target sites exist in PTBs, and these sites are preceded by Asp7, Asp139, and Asp172 (Fig. 4). The caspase-3 recognition sequences in PTB are highly conserved fromXenopus to human, implying that cleavage of PTB during apoptosis plays an important physiological role in a large group of animals (Fig. 4 A). The caspase-3 target sites in PTB (Ile-Val-Pro-Asp7↓Ile, Leu-Tyr-Thr-Asp139↓Ser, and Ala-Ala-Val-Asp172↓Ala) are not conventional caspase-3 cleavage sites that harbor aspartate residues at both P1 and P4 positions. However, other work has shown the aspartate at the P4 position does not seem to be absolutely required for caspase-3 recognition. For instance, caspase-3 cleaves scaffold attachment factor A at SAL(D/G) (67), human recombinase Rad51 (HsRad51) at AQV(D/G) (68), p21 at SMT(D/F) (69), topoisomerase I at EEE(D/G) (70), and calpastatin at DFT(C/G) (71). These findings may suggest that not only the primary sequence but also the conformation of substrate proteins contribute to caspase-3 recognition.
When we compared the efficiencies of caspase-3-dependent cleavages of three Asp-X pairs in PTB, the Leu-Tyr-Thr-Asp139↓Ser was the most preferred target site, and Ile-Val-Pro-Asp7↓Ile was the least preferred (Fig. 3, A and B, and Fig. 4 D). In addition, cleavage at the latter site was inhibited by the fusion of GFP protein to the N terminus of PTB4 (Fig. 4 D). This result suggests that the sequences surrounding the target sites influence cleavage of PTB. The physiological importance of the differential cleavages remains obscure.
PTB is a shuttling protein that migrates between the nucleus and the cytoplasm (24). In vegetatively growing cells, PTBs reside in the nucleus (panel 1 in Fig. 5 A). Upon induction of apoptosis, PTBs are fairly evenly redistributed between the cytoplasm and the nucleus (panel 4 in Fig. 5 A). By transfecting cells with the GFP/PTB4-RRM2-3-4 construct, which produces a GFP-fused C-terminal fragment of PTB4, we demonstrated that the C-terminal (RRM2-3-4) fragment of PTB4 is partially localized in the cytoplasm (panel 4 in Fig. 5 B). Redistribution of PTB fragments may be attributed to several phenomena occurring during apoptosis. First, the N-terminal region of PTB (up to RRM1) containing an NLS is separated from the C-terminal region (RRM2-3-4). Therefore, the PTB C-terminal domain can stay in the cytoplasm after leaving the nucleus. Second, impairment of the nuclear pore complex occurs during apoptosis. Recently, several reports suggested that the activation of caspases during apoptosis increases permeability of the nuclear pore by cleaving components of the nuclear pore complex (Nup153, Nup214, Tpr, Nup358/RanBP2) (72-74) and of the nuclear transport system (RanGap1) (73). These modifications result in leakage of nuclearly restricted proteins into the cytoplasm. Third, inhibition of transcription during apoptosis may also contribute in part to the relocalization of PTB, because transcriptional inhibition of cellular transcription by actinomycin D induces the partial redistribution of PTB to the cytoplasm (24, 57).
We speculate that the cleavage of PTB by caspase-3 modulates translation of some mRNAs containing IRES elements. Modulation of translation by proteolytic cleavage of a translation initiation factor has been extensively investigated in an attempt to understand the shut down of host cell protein synthesis by poliovirus (75). Upon poliovirus infection, translation of cap-dependent mRNAs in the host cell is inhibited by cleavage of a translational initiation factor eIF4G by virally encoded proteinase 2A. The proteolytic cleavage of eIF4G results in separation of the N-terminal domain that interacts with eIF4E, which in turn binds to the cap structure in an mRNA, from the C-terminal domain that interacts with other translational initiation factors such as eIF4A, an RNA helicase, and eIF3, which in turn interacts with the 40 S ribosomal subunit. Consequently, the cleaved eIF4G cannot direct translation through the cap-dependent scanning mechanism that is utilized by the majority of cellular mRNAs. On the other hand, translation of polioviral mRNAs containing an IRES element is not hampered by the cleavage of eIF4G, because the cap structure is not necessary for IRES-dependent translation. An analogous situation has been observed in apoptotic cells. The eIF4GI is cleaved into three fragments at the early stage of apoptosis (76-80). Caspase-3 is believed to be responsible for the eIF4GI cleavage (79, 81). Similar to the cleavage of eIF4G by viral proteases, the caspase-mediated cleavage of eIF4G coincides with the inhibition of protein synthesis. However, translation of some cellular mRNAs containing IRES elements continues even after apoptosis is induced. For instance, the IRES of the X-linked inhibitor of apoptosis protein is functional when apoptosis is induced by serum starvation and low dose γ-irradiation, conditions that lead to the inhibition of translation of cap-dependent mRNAs (82, 83). In addition, it was reported that the IRESs of c-myc (84) or p97/DAP5/NAT1 (85) are active during apoptosis. Thus, it is generally believed that the translation of some cellular mRNAs through the IRES element is required for maintenance, adaptation, and escape from the apoptotic pathway. It is likely that regulatory mechanisms exist to control translational levels of IRES-containing mRNAs during apoptosis.
Recently, we showed that a poliovirus protein 3Cpro and/or 3CDpro cleaves PTB isoforms and that translation of polioviral mRNA is inhibited by the cleavage of PTB (57). Here we report that PTB is cleaved by caspase-3 during apoptosis. Interestingly, both polioviral 3Cpro and caspase-3 cleave at the interdomain region between RRM1 and RRM2. The resulting PTB fragment (PTB4-RRM2-3-4) generated by caspase-3 is partially localized to the cytoplasm (panel 4 in Fig. 5) and inhibits the function of polioviral IRES (bars 5–7 in Fig. 6) and encephalomyocarditis virus IRES (data not shown). In contrast, PTB4-(Δ60–173), which contains an NLS at the N-terminal end of the molecule, did not affect the translation of polioviral mRNA (bars 8–10 in Fig. 6). It is likely that the C-terminal fragment of PTB inhibits the activity of the polioviral IRES at the translational level. Nevertheless, the mechanism of translational inhibition by PTB fragments remains to be elucidated. We speculate that PTB cleavage during apoptosis may modulate translation of particular species of mRNAs. It is possible that translation of PTB-dependent mRNAs such as apoptotic protease-activating factor-1 and insulin-like growth factor-I receptor is inhibited by the cleavage of PTB (44, 45). On the other hand, translation of mRNAs that are inhibited by intact PTB (43), such as immunoglobulin heavy chain-binding protein, might be enhanced by cleavage of PTB. It is also possible that PTB cleavage impairs nuclear functions of full-length PTB, such as the modulation of alternative splicing of some pre-mRNAs (32-34). The effect of PTB cleavage on the nuclear function of intact PTB remains to be elucidated. Apoptosis is a crucial event in many biological processes. The RNA-binding PTBs are members of the hnRNP family and regulate translation. The present study provides significant novel information regarding the cleavage of PTBs during apoptosis. It is possible that this cleavage influences the progression of apoptosis, particularly through effects on protein translation.
ACKNOWLEDGEMENTS
We thank Dr. E. Wimmer for anti-PTB antibodies; Dr. P. Suh for the anti-caspase-3 antibody; Dr. M. M. C. Lai for the anti-hnRNP A1 antibody; and Dr. G. Dreyfuss for the anti-hnRNP C1/C2 antibody. We thank Dr. S. S. Bae for providing apoptotic cell lysates.
Footnotes
-
↵* This work was supported in part by the G7, National Research Laboratory, and Molecular Medicine Research Group Programs of the Ministry of Science and Technology and by a grant from the Korea Science and Engineering Foundation through the Protein Network Research Center.The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
-
↵‡ To whom correspondence should be addressed: Dept. of Life Science, Division of Molecular and Life Sciences, Pohang University of Science and Technology, San31, Hyoja-Dong, Pohang, Kyungbuk 790-784, Korea. Tel.: 82-54-279-2298; Fax: 82-54-279-8009; E-mail: sungkey@postech.ac.kr.
-
Published, JBC Papers in Press, May 9, 2002, DOI 10.1074/jbc.M203887200
- Abbreviations:
- hnRNP
-
heterogeneous nuclear ribonucleoprotein
- PTB
-
polypyrimidine tract-binding protein
- IRES
-
internal ribosomal entry site
- TNF
-
tumor necrosis factor
- TRAIL
-
tumor necrosis factor-related apoptosis-inducing ligand
- NLS
-
nuclear localization signal
- RRM
-
RNA recognition motif
- GFP
-
green fluorescence protein
- EGFP
-
enhanced GFP
- RFP
-
red fluorescence protein
- aa
-
amino acids
- CHAPS
-
3-[(3-cholamidopropyl)dimethylammonio]-1-propanesulfonic acid
- TRITC
-
tetramethylrhodamine isothiocyanate
- Z
-
benxyloxycarbonyl
- fmk
-
fluoromethyl ketone
- mAb
-
monoclonal antibody
- PARP
-
poly(ADP-ribose) polymerase
- DAPI
-
4,6-diamidino-2-phenylindole
-
- Received April 22, 2002.
- The American Society for Biochemistry and Molecular Biology, Inc.
