BH4-domain peptide from Bcl-xL exerts anti-apoptotic activity ... - Nature

Oncogene (2003) 22, 8432–8440

& 2003 Nature Publishing Group All rights reserved 0950-9232/03 $25.00 www.nature.com/onc

BH4-domain peptide from Bcl-xL exerts anti-apoptotic activity in vivo Rie Sugioka1,2, Shigeomi Shimizu1,2, Toshihiro Funatsu3, Hiroshi Tamagawa3, Yoshiki Sawa3, Toru Kawakami4 and Yoshihide Tsujimoto1,2,* 1

Department of Post-Genomics & Diseases, Osaka University Medical School, 2-2 Yamadaoka, Suita, Osaka 565-0871, Japan; CREST and SORST of Japan Science and Technology Corporation (JST), 2-2 Yamadaoka, Suita, Osaka 565-0871, Japan; 3 Department of Surgery, Osaka University Medical School, 2-2 Yamadaoka, Suita, Osaka 565-0871, Japan; 4Laboratory of Protein Architectonics, Research Center for Structural Biology, Institute for Protein Research, Osaka University, 3-2 Yamadaoka, Suita, Osaka 565-0871, Japan 2

The Bcl-2 family of proteins regulates apoptosis chiefly by controlling mitochondrial membrane permeability. It has previously been shown that the BH4 domain of Bcl-2/BclxL is essential for the prevention of apoptotic mitochondrial changes, including the release of cytochrome c and apoptotic cell death. We have previously reported that BH4 peptide fused to the protein transduction domain of HIV-1 TAT protein (TAT-BH4) significantly inhibits etoposide-induced apoptosis in a cell line. This time, we investigated whether TAT-BH4 peptide was cytoprotective in ex vivo and in vivo rodent models. Intraperitoneal injection of TAT-BH4 peptide greatly inhibited X-rayinduced apoptosis in the small intestine of mice and partially suppressed Fas-induced fulminant hepatitis. In addition, this peptide markedly suppressed heart failure after ischemia–reperfusion injury in isolated rat heart, probably by preventing mitochondrial dysfunction. These findings demonstrate that TAT-BH4 peptide exerts antiapoptotic activity both in vivo and ex vivo, and imply that it may be a useful therapeutic agent for diseases involving mitochondrial dysfunction and apoptosis. Oncogene (2003) 22, 8432–8440. doi:10.1038/sj.onc.1207180 Keywords: Bcl-xL; BH4 domain; apoptosis; mitochondria; TAT

Introduction Apoptosis is an essential physiological process that leads to the selective elimination of cells in metazoans, and it is involved in a variety of biological events. The mitochondria often play a crucial role in apoptosis by releasing apoptogenic factors, such as cytochrome c and Smac/Diablo, into the cytoplasm (reviewed by Wang, 2001). After it reaches the cytoplasm, cytochrome c forms a large protein complex with Apaf-1 and *Correspondence: Y Tsujimoto, Osaka University Medical School, Laboratory of Molecular Genetics (Room B8), 2-2 Yamadaoka, Suita, Osaka 565-0871, Japan; E-mail: [email protected] Received 20 March 2003; revised 9 September 2003; accepted 9 September 2003

procaspase-9, leading to autoactivation of caspase-9, and activated caspase-9 subsequently activates the effector caspases such as caspase-3 (reviewed by Wang, 2001). Then the activated effector caspases cleave various cellular proteins to execute apoptosis. Smac/ Diablo facilitates caspase activation by interacting with and inhibiting the IAPs, a family of endogenous caspase inhibitors (Du et al., 2000; Verhagen et al., 2000). The Bcl-2 family of proteins is one of the bestcharacterized regulators of apoptosis, consisting of antiapoptotic members, such as Bcl-2 and Bcl-xL, as well as pro-apoptotic members that include multidomain Bax and Bak plus various BH3-only proteins including Bid, Bik, and Bim (reviewed by Adams and Cory, 1998; Green and Reed, 1998; Tsujimoto and Shimizu, 2000a). The proteins of this family mainly regulate mitochondrial membrane permeability (reviewed by Martinou and Green, 2001; Zamzami and Kroemer, 2001). It has been proposed that the release of mitochondrial apoptogenic factors occurs through one of the following: (1) a pore formed by oligomerized multidomain pro-apoptotic Bcl-2 family members such as Bax and Bak (Eskes et al., 2000; Saito et al., 2000; Wei et al., 2000), (2) the voltage-dependent anion channel (VDAC), which is directly regulated by Bcl-2 family members (Shimizu et al., 1999, 2000a, 2001; Tsujimoto and Shimizu, 2000b), and (3) a large transient lipidic pore triggered by Bax (Kuwana et al., 2002). Anti-apoptotic Bcl-2 family members share sequence homology at multiple regions of the four BH (Bcl-2 homology) domains (reviewed by Adams and Cory, 1998; Green and Reed, 1998; Tsujimoto and Shimizu, 2000a). It has been shown that the BH1 and BH2 domains, and probably the BH3 domain as well, are required for dimerization with pro-apoptotic Bcl-2 family proteins when inhibiting their pro-apoptotic activity. Unlike other BH domains, the BH4 domain is conserved among many anti-apoptotic proteins from the Bcl-2 family and this domain seems to be essential for anti-apoptotic activity because its deletion from Bcl-2/ Bcl-xL abrogates their anti-apoptotic effect (Borner et al., 1994; Hanada et al., 1995; Hunter et al., 1996; Lee et al., 1996; Huang et al., 1998; Shimizu et al., 2000b). It has been reported that the BH4 domain is required for binding to other molecules, including

Bcl-xL-BH4 domain inhibits apoptosis in vivo R Sugioka et al

8433

calcineurin (Shibasaki et al., 1997), Raf-1 (Wang et al., 1996), and Bax (Hirotani et al., 1999), although it was not determined whether this domain was sufficient for the binding. Recently, we showed that the BH4 domain is essential and sufficient for blocking the apoptotic increase of mitochondrial membrane permeability through direct interaction with the VDAC (Shimizu et al., 2000a, b). This finding led us to speculate that BH4 peptide exerts anti-apoptotic activity. In a previous study, we used a protein-transduction domain (PTD), which is found in several membrane-permeable proteins (reviewed by Schwarze et al., 2000) (such as Drosophila Antenapedia, HSV VP-22, and HIV-1 TAT) and has proven useful for delivering polypeptides into cells after fusion to peptides of interest (Schwarze et al., 1999). We showed that fusion of the PTD of HIV-1 TAT (nine amino-acid residues long) to BH4 peptide (TAT-BH4)

caused significant inhibition of etoposide-induced apoptosis in a cell line (Shimizu et al., 2000b). Here, we extended this finding and showed that TATBH4 peptide could be used both in vivo and ex vivo to protect cells and mitochondria. Our results suggest that TAT-BH4 peptide may be a useful therapeutic agent for diseases involving apoptosis and mitochondrial dysfunction. Results TAT-BH4 peptide is incorporated into cells and exerts anti-apoptotic activity Using TAT-BH4 peptide (Figure 1a) with an eosinlabeled amino terminal, we have previously shown that this peptide is efficiently incorporated into HeLa cells

Figure 1 Efficient incorporation of TAT-BH4 peptide into HeLa cells and mouse liver. (a) Amino-acid sequences of the peptides used in this study. Ac and b-A stand for acetylated and b-alanine, respectively. The sequences of TAT (RKKRRQRRR) and BH4 (4SNRELVVDFLSYKLSQKGYS23) were derived from nine amino-acid residues within the PTD of HIV-1 TAT protein and from the BH4 domain of human Bcl-xL, respectively. Deleted and substituted residues are highlighted. (b, c) Eosin-labeled TAT-BH4 peptide (330 nM) was added to the culture medium of HeLa cells. This concentration of TAT-BH4 peptide was used to detect definitive fluorescence. After 30 min, the cells were washed and examined by phase-contrast (left) and fluorescence (right) microscopes (b). Efficiency of incorporation was also measured by flow cytometry (c). Profiles of cells incubated with (red) or without (blue) eosinlabeled TAT-BH4 peptide are shown. Under this condition, cell membrane remained intact, as confirmed by the absence of PI staining. (d) Incorporation of TAT-BH4 peptide in vivo. Mice were intraperitoneally injected with BH4 peptide (3.3 mg/g BW) or TAT-BH4 peptide (5 mg/g BW). After 30 min, the livers were perfused and excised. Crude lysate was prepared and an equivalent amount of proteins was subjected to dot blot analysis using the indicated antibodies (upper two panels). In the lower two panels, BH4 and TATBH4 peptide (0.8 nmol, upper spots; 0.4 nmol, lower spots) were also spotted on the membrane Oncogene


8434

(a human cervical carcinoma cell line) when added to the culture medium (Shimizu et al., 2000b). As shown in Figure 1b, TAT-BH4 peptide, but not BH4 itself (data not shown), was incorporated into almost all the cells within 30 min, confirming our previous observation (Shimizu et al., 2000b). This efficient incorporation of TAT-BH4 peptide was confirmed by flow cytometric analysis (Figure 1c). Moreover, for the use of TAT-BH4 peptide in animals, we examined whether TAT-BH4 peptide injected intraperitoneally in mice was delivered to organs. As shown in Figure 1d, TAT-BH4 peptide but not BH4 peptide injected intraperitoneally accumulated in mouse liver. We have previously shown that TAT-BH4 peptide inhibits etoposide-induced apoptosis of HeLa cells (Shimizu et al., 2000b). To confirm the anti-apoptotic activity of TAT-BH4 peptide, we used different cell lines and apoptotic stimuli in the present study. TAT-BH4 peptide inhibited X-ray-induced apoptosis of PC12 cells (a rat pheochromocytoma cell line) (Figure 2a). The anti-cell death activity of TAT-BH4 peptide was also confirmed by a clonogenicity assay: X-ray-irradiated cells in the presence of TAT-BH4 peptide formed more colonies than X-ray-irradiated cells in the absence of TAT-BH4 peptide (data not shown). TAT-BH4 peptide exhibited increasing anti-apoptotic activity up to 33 nM, while its activity decreased at 330 nM (Figure 2b), probably due to toxicity. TAT-only and two TATBH4 mutant peptides (mt1 and mt2, see Figure 1a) (the mutations abrogated the anti-apoptotic activity of BclxL (Lee et al., 1996)) did not show anti-apoptotic activity (Figure 2c and d), indicating that BH4 itself had the anti-apoptotic effect. The maximum anti-apoptotic activity of TAT-BH4 peptide against X-ray-induced apoptosis was almost half of that shown by overexpression of Bcl-xL (Figure 2a). TAT-BH4 peptide was also tested against other apoptotic stimuli. TAT-BH4 peptide partially inhibited etoposide-induced apoptosis, but not tunicamycin-induced apoptosis (Figure 2e and f). Both of these forms of apoptosis were almost completely inhibited by overexpression of Bcl-2 or BclxL (Figure 2e and data not shown). These results indicated that, unlike Bcl-2 and Bcl-xL, TAT-BH4 peptide only inhibits some (but not all) types of apoptosis. TAT-BH4 peptide inhibits X-ray-induced apoptosis in the mouse small intestine TAT-BH4 peptide efficiently inhibited X-ray-induced apoptosis of cultured cells (Figures 2a–d). To examine whether TAT-BH4 peptide could also inhibit apoptosis in vivo, we assessed its effect on X-ray-induced apoptosis in the mouse small intestine, which is known to be highly sensitive to X-ray damage. When control mice were irradiated with X-rays (20 Gy), caspase-3(-like) activity was markedly increased after 2 h and then decreased rapidly (Figure 3a, closed squares), probably due to degradation and/or leakage of the enzymes from the cells. In contrast, when TAT-BH4 peptide was preinjected intraperitoneally into the mice, an increase of Oncogene

Figure 2 Inhibition of some forms of apoptosis by TAT-BH4 peptide. (a–d) Anti-apoptotic effect of TAT-BH4 peptide on X-rayinduced apoptosis. (a) PC12 cells were treated with X-rays (20 Gy) in the presence (open circles) or absence (closed squares) of TATBH4 peptide (33 nM). PC12 cells overexpressing human Bcl-xL were also exposed to X-rays (20 Gy) in the absence of TAT-BH4 peptide (open squares). At the indicated time, cells were stained with Hoechst 33342, and apoptotic cells were counted under a fluorescent microscope, (b–d) PC12 cells were exposed to X-rays (20 Gy) in the presence of TAT-BH4 peptide at the indicated concentrations (b) or with the indicated peptides at 33 nM (c, d). After 24 h, apoptotic cells were counted by examining the nuclear morphology (b, c), and were measured by LDH release (d). (e) Anti-apoptotic effect of TAT-BH4 peptide on etoposide-induced apoptosis. HeLa cells were treated with etoposide (200 mM) for 30 h in the absence or presence of TAT-BH4 peptide (33 nM). HeLa cells overexpressing human Bcl-xL were also treated with etoposide in the absence of TAT-BH4 peptide (Bcl-xL). Apoptotic cells were detected from the nuclear morphology, (f) No effect of TAT-BH4 peptide on tunicamycin-induced apoptosis. HeLa cells were treated with 2 mg/ml of tunicamycin for 24 h in the absence or presence of TAT-BH4 peptide (33 nM). Apoptotic cells were detected from the nuclear morphology

caspase-3(-like) activity was suppressed (Figure 3a, open circles). By injecting various amounts of TAT-BH4 peptide, the minimal amount of the peptide, which exhibited the maximum cytoprotection, was determined (data not shown). The X-ray-induced activation of caspase-3 and inhibition by TAT-BH4 peptide were confirmed by Western blot analysis using anti-active caspase-3 antibody (Figure 3b, lower panel) and CaspACE FITC-VAD-FMK staining, which specifically


8435

Figure 3 Inhibition of X-ray-induced apoptosis in the mouse small intestine by TAT-BH4 peptide. (a–c) Inhibition of X-ray-induced caspase-3 activation by TAT-BH4 peptide. (a) Mice were pretreated with (open circles) or without (closed squares) TAT-BH4 peptide at 20 mg/g BW for 2 h and then exposed to X-rays (20 Gy). Cytosolic fractions were prepared from equivalent portions of the small intestine at the indicated times after irradiation, and caspase-3(-like) activity was measured, (b) Mice pretreated with the indicated peptides at 20 mg/g BW were exposed to X-rays (20 Gy). Cytosolic fractions were prepared from equivalent portions of the small intestine at 2 h after irradiation, and an equivalent amount of proteins was subjected to Western blot analysis using anti-active caspase3 antibody (lower) or were assayed for caspase-3(-like) activity (upper). Values are means þ s.d. (n>6 in each group). #Po0.01 vs group of nonirradiated mice, *Po0.01 vs group of X-ray-irradiated TAT-BH4 () mice, (c) Mice pretreated with or without TAT-BH4 peptide were exposed to X-rays (20 Gy). At 3 h after irradiation, FITC-VAD-FMK (13.2 mg/g BW) was intraperitoneally injected and incubated for 40 min. Activated caspases (stained green) were detected by fluorescence microscopy. (d) Inhibition of X-ray-induced apoptosis in the mouse small intestine by TAT-BH4 peptide. Mice were pretreated with TAT-BH4 or TAT-BH4 mt2 peptide (20 mg/g BW) or DMSO and were exposed to X-rays (20 Gy). Tissues were frozen after 8 h and apoptotic cells were visualized by the TUNEL method. All samples were simultaneously subjected to the TUNEL assay. Representative photographs are shown of a low magnification ( 100). Apoptotic cells are stained brown, and non-apoptotic cells are stained green

binds to active caspases (Figure 3c). Inhibition of caspase-3 activation was not observed with TAT-BH4 mutant peptides, as assessed by measurement of enzyme activity and by Western blot analysis (Figure 3b), indicating that BH4 was vital for the inhibition of caspase-3 activation. Furthermore, to examine whether X-ray-induced apoptosis in small intestine is rescued by TAT-BH4 peptide, TUNEL analysis was performed. At 8 h after X-ray irradiation of mice, cells of small intestine were TUNEL-positive in the absence of TAT-BH4 peptide, whereas little TUNEL-positive cells were detected in the small intestine of non-irradiated mice (Figure 3d). In contrast, treatment with TAT-BH4 peptide but not TAT-BH4 mutant peptide dramatically inhibited X-rayinduced apoptosis of the small intestine (Figure 3d).

These results demonstrated that, when administered intraperitoneally to mice, TAT-BH4 peptide was able to inhibit X-ray-induced apoptosis in small intestine. TAT-BH4 peptide partially suppresses Fas-induced fulminant hepatitis in vivo To explore the possibility that TAT-BH4 peptide could be used to protect cells from death in a setting that was more relevant to human disease, we used two models, which were fulminant hepatitis and cardiac ischemia– reperfusion injury. Mice. were intraperitoneally injected with TAT-BH4 peptide (1 mg/g body weight (BW)) and then were injected with an agonistic anti-Fas antibody (Jo-2) at 0.5 mg/g BW. TAT-BH4 peptide markedly inhibited the Oncogene


8436

Figure 4 Partial suppression of anti-Fas antibody-induced fulminant hepatitis by TAT-BH4 peptide. (a, b) Serum AST level and liver histology. (a, left panel, b) TAT-BH4 peptide (1 mg/g BW) or DMSO was injected intraperitoneally into C57BL6 mice. After 30 min, 0.5 mg/g BW of an agonistic anti-Fas antibody, Jo-2 was injected intraperitoneally. At the indicated time, serum AST activity was measured (a). Values are means þ s.d. (n ¼ 4). #Po0.01 vs group of Jo-2-injected TAT-BH4 () mice at 7 h after Jo-2 injection. Representative histological appearance of livers at 5.5 h after Jo-2 injection is shown of low (b, left panels) and high (b, right panels) magnifications. Arrows indicate apoptotic nuclei. (a, right panel) Mice (n>8) were intraperitoneally injected with 5 mg/g BW of TATBH4 or TAT-BH4 mt2 and then 1.5 mg/g BW of Jo-2. At 4 h after Jo-2 injection, serum AST activity was measured. *Po0.01 vs group of Jo-2-injected TAT-BH4 () mice at 4 h after Jo-2 injection. (c) Effect of TAT-BH4 peptide on the survival after injection of anti-Fas antibody. Mice (n ¼ 20) were intraperitoneally injected with (open circles) or without (closed squares) 5 mg/g BW of TAT-BH4 peptide and then 1.5 mg/g BW of Jo-2 (P ¼ 0.0015)

release of aspartate aminotransferase (AST) into the serum by 7 h after Jo-2 injection (Figure 4a, left panel). However, this effect of TAT-BH4 peptide was abrogated at 9 h after Jo-2 injection (data not shown), indicating that TAT-BH4 peptide has the ability to delay Jo-2-induced hepatic injury. Histological examination of resected livers indicated that hepatic apoptosis and hemorrhagic necrosis were suppressed in the presence of TAT-BH4 peptide (Figure 4b), supporting the protective effect of TAT-BH4 peptide against Jo-2induced hepatic injury. A larger amount of TAT-BH4 peptide (5 mg/g BW), but not mutant peptide, suppressed hepatic injury induced by a higher dose of Jo-2 (1.5 mg/g BW) (Figure 4a, right panel). Moreover, mouse survival after Jo-2-induced hepatic injury was also improved, although partially, by TAT-BH4 peptide (Figure 4c). Note that TAT-BH4 peptide exhibited increasing antiapoptotic activity up to 5 mg/g BW in this system, while Oncogene

its activity significantly decreased at more than 5 mg/g BW. These results indicated that TAT-BH4 peptide partially inhibits the occurrence of Fas-induced fulminant hepatitis. TAT-BH4 peptide improves ex vivo ischemia–reperfusioninduced cardiac dysfunction Since we and others have shown that ischemia–reperfusion triggers apoptosis and necrosis, both of which are inhibited by Bcl-2/Bcl-xL (Shimizu et al., 1996a; Saikumar et al., 1998; Yamabe et al., 1998), we examined the protective effect of TAT-BH4 peptide on ischemia–reperfusion-induced injury of the heart. Hearts harvested from rats were used instead of an in vivo study for the sake of experimental simplicity. Cardiac function was assessed by measuring the left ventricular maximum dP/dt, as described in Materials


8437

and methods. After 40 min of warm ischemia, control hearts treated with TAT-BH4 mutant peptides (mt1 and mt2) showed little cardiac function, probably because of the low level of intracellular ATP and only slight recovery (12 and 15.4%, respectively) after reperfusion (Figure 5a). When TAT-BH4 peptide was administered before ischemia, although cardiac function decreased during ischemia to a similar level as that of the control hearts, it recovered to a much higher level (65.3%) after reperfusion (Figure 5a). Isolated mitochondria were prepared after reperfusion, and mitochondrial respiration was measured as described in Materials and methods. The respiratory control ratio (RCR) of mitochondria from TAT-BH4 mutant peptide-pre-

treated hearts only recovered slightly after reperfusion (24.1%), whereas the RCR of mitochondria from TATBH4 peptide-pretreated hearts showed marked recovery (75.8%) (Figure 5b). The recovery rate of the RCR was similar to that of cardiac function, indicating that the effect of TAT-BH4 peptide was probably ascribable to mitochondrial protection. Moreover, we examined whether TAT-BH4 peptide directly inhibited hypoxia– reoxygenation-induced mitochondrial damage, which mimics ischemia–reperfusion injury, using isolated liver mitochondria. Mitochondrial damage was assessed by measuring the release of mitochondrial AST (mAST), a matrix enzyme, and cytochrome c into the media. As shown in Figure 5c and d, TAT-BH4 peptide but not

Figure 5 Inhibition of cardiac ischemia–reperfusion injury by TAT-BH4 peptide through preserving mitochondria. (a) Explanted rat hearts (n ¼ 5) were preperfused with modified Krebs–Henselite solution for 20 min, and TAT-BH4 peptide (open circles) or TAT-BH4 mutant peptides (mt1, closed squares; mt2, open squares) (0.4 mg) were given. Then the hearts were subjected to warm ischemia for 40 min, followed by reperfusion for 60 min, and cardiac function was measured every 15 min, as described in Materials and methods. An asterisk indicates maximum dP/dt before ischemia treatment. (b) Effect of TAT-BH4 peptide on mitochondrial respiration during ischemia–reperfusion. The same experiment as in (a) was performed, and, at the indicated time, mitochondria were isolated from the heart, and then, mitochondrial respiration at state III–IV was measured by O2 electrode, as described in Materials and methods. Data are expressed as RCR (state III/IV). Open circles: TAT-BH4 peptide, open squares: TAT-BH4 mt2 peptide. An asterisk indicates the RCR before ischemia treatment. Values are means7s.d. (n ¼ 5 in each group). (c, d) Effect of TAT-BH4 peptide on release of mAST and cytochrome c from isolated mitochondria during hypoxia and reoxygenation. Mitochondria were incubated in N2 gas and then aerated in the absence (closed circles) or presence of TAT-BH4 peptide (open circles) or TAT-BH4 mt2 peptide (open squares). At the indicated times, the samples were centrifuged and the supernatants were subjected to mAST assay (c) and to Western blot analysis for the detection of cytochrome c (d). Data are representative of three independent experiments Oncogene


8438

TAT-BH4 mutant peptide prevented hypoxia–reoxygenation-induced mAST and cytochrome c release from the mitochondria. These results indicated that TATBH4 peptide could preserve cardiac function by preventing ischemia–reperfusion-induced mitochondrial damage. Discussion We and others have previously reported that the BH4 domain is essential for the anti-apoptotic activity of Bcl2/Bcl-xL (Borner et al., 1994; Hanada et al., 1995; Hunter et al., 1996; Lee et al., 1996; Huang et al., 1998; Shimizu et al., 2000b). We have also reported that BH4 peptide can inhibit VDAC activity like Bcl-xL and that it inhibits apoptotic changes of isolated mitochondria (Shimizu et al., 2000b). In addition, TAT-BH4 peptide suppresses apoptosis in HeLa cells induced by etoposide (Shimizu et al., 2000b). Here, we confirmed that TATBH4 peptide was able to inhibit some types of apoptosis in cultured cells, and also demonstrated that it had a significant protective effect in two in vivo mouse models, which were X-ray-induced apoptosis of the small intestine and agonistic anti-Fas-antibody-induced fulminant hepatitis. We also showed that TAT-BH4 peptide could effectively suppress ischemia–reperfusion injury of the explanted rat heart, most likely by preventing ischemia–reperfusion-induced mitochondrial dysfunction. Since TAT-only as well as TAT-BH4 mutant peptides showed no such effects, our results clearly indicate that the BH4 domain itself possesses anti-apoptotic activity. Although we assume that TATBH4 peptide exerts its cytoprotective effect by inhibiting the VDAC, we did not exclude the possibility that this peptide has some other activity. These findings suggest that TAT-BH4 peptide might be a useful agent for the treatment of diseases involving apoptotic death or mitochondrial dysfunction. In contrast to Bcl-2/Bcl-xL, which prevents various types of apoptosis, TAT-BH4 peptide was unable to inhibit some forms of apoptosis such as that induced by tunicamycin. It seems that Bcl-2/Bcl-xL has two independent anti-apoptotic mechanisms: one is represented by the BH4 domain itself and the other also requires the BH4 domain, but the domain alone is not sufficient (reviewed by Tsujimoto, 2003). The former mechanism might be related to the anti-apoptotic activity of Bcl-2/Bcl-xL that is mediated by closing the VDAC (Shimizu et al., 2000b), and the latter might be related to a mechanism involving heterodimerization with pro-apoptotic Bcl-2 family members (Oltvai et al., 1993; Yang et al., 1995). It is conceivable that, depending on the apoptotic stimulus or which proapoptotic members are activated, death signals are suppressed more efficiently by one mechanism than the other, because Bcl-2/Bcl-xL seems to show significant differences of its affinity for different pro-apoptotic Bcl2 family members. Therefore, since TAT-BH4 peptide does not show heterodimerization, it may not suppress well certain forms of apoptosis. Oncogene

Why is the anti-apoptotic activity of TAT-BH4 peptide incomplete compared with that of Bcl-2/BclxL? This may also be partly explained by the existence of two independent anti-apoptotic mechanisms involving Bcl-2/Bcl-xL, as described above. Another explanation is a lower affinity of TAT-BH4 peptide for the VDAC than Bcl-2/Bcl-xL. We have previously shown that BH4 peptide can prevent apoptotic changes of isolated mitochondria when it is added at a molar concentration about 25-fold higher than that of recombinant Bcl-xL (Shimizu et al., 2000b). This is probably because other regions, including the BH1 domain (which we have shown is required for interaction of Bcl-xL with the VDAC (Shimizu et al., 1999)), facilitate binding of the BH4 domain to the VDAC or assist this domain to form its proper structure. Although a higher concentration of TAT-BH4 peptide is probably able to exert antiapoptotic activity comparable to that of Bcl-2/Bcl-xL, this is hampered by the toxicity of the peptide (Figure 2b). Such toxicity is probably ascribable to the TAT region, since HIV-1 TAT protein is known to induce apoptosis (Macho et al., 1999; Park et al., 2001). Thus, it might be possible to obtain a form of TAT-BH4 peptide with stronger anti-apoptotic activity by modifying the TAT region. Although we have shown that the BH4 domain of Bcl-xL exerts anti-apoptotic activity, one might argue that not all anti-apoptotic Bcl-2 family members possess the BH4 domain. Thus, members lacking the BH4 domain may exert anti-apoptotic activity simply by heterodimerization with pro-apoptotic members. Alternatively, since many Bcl-2 family members that lack the BH4 domain still possess an a helix at the N-terminal region, where the BH4 domain of Bcl-2/Bcl-xL resides, this a helix may have a role similar to that of the BH4 domain. TAT-BH4 peptide seems to have advantages as a therapeutic agent when compared with gene therapy using Bcl-2/Bcl-xL. First of all, the peptide can easily be injected intravenously or intraperitoneally, and it is unnecessary to introduce it specifically at affected sites, whereas the gene must be given locally into the target areas. Introduction of the Bcl-2/Bcl-xL gene might also increase the risk of tumorigenesis (Hockenbery, 1992), whereas the TAT-BH4 peptide does not pose this risk because of its relatively short life. The present TAT-BH4 peptide could be improved with respect to activity and toxicity by modifying the BH4 and TAT regions, respectively. Therefore, this peptide is a promising agent for diseases involving accelerated apoptotic death or mitochondrial dysfunction, including ischemia–reperfusion injury of the heart and brain. Materials and methods Chemicals and antibodies Tunicamycin was obtained from Sigma. An anti-active caspase-3 polyclonal antibody and CaspACE FITC-VADFMK were purchased from Promega. An anti-Fas monoclonal antibody (Jo-2) and an anti-cytochrome c monoclonal antibody were obtained from BD Pharmingen. An anti-Bcl-xL


8439 (S-18) antibody recognizing the BH4 domain was purchased from Santa Cruz. DEVD-MCA was purchased from The Peptide Institute. Other chemicals were obtained from Wako Biochemicals. Synthesis of peptides Synthesis of the peptides used in this study was performed as described previously (Shimizu et al., 2000b). Eosin-labeled peptides were synthesized by conjugating eosin to the cysteine residue at the N-terminus. All peptides were dissolved in dimethyl sulfoxide (DMSO) before use. Cell lines HeLa cells (a human cervical carcinoma cell line) and PC12 cells (a rat pheochromocytoma cell line) were maintained in RPMI 1640 medium supplemented with 10% fetal bovine serum. Stable transfectants of PC12 and HeLa cells expressing human Bcl-xL were obtained by transfection of the pUCCAGGS vector (Niwa et al., 1991) bearing human bcl-xL cDNA using electroporation (Shimizu et al., 1996a). Analysis with eosin-labeled TAT-BH4 peptide HeLa cells were preincubated with 330 nM of eosin-labeled TAT-BH4 peptide for 30 min at 371C and then were washed, after which the incorporated TAT-BH4 peptides were detected by fluorescent microscopy (Olympus) and by flow cytometric analysis (FACS Calibur, Becton Dickinson). Induction of apoptosis PC12 cells and HeLa cells were pretreated with 33 nM of TATBH4 peptide for 30 min at 371C. Then the cells were exposed to 20 Gy of X-ray irradiation, 200 mM etoposide, or 2 mg/ml tunicamycin. After staining with 1 mM Hoechst 33342, the extent of apoptosis was calculated as the percentage of cells with nuclear condensation or chromatin condensation. LDH release was measured by LDH-cytotoxic test (WAKO). All values are expressed as means and s.d. In vivo and ex vivo experiments To investigate X-ray-induced apoptosis of the small intestine, C57BL6J mice were intraperitoneally injected with TAT-BH4 (20 mg/g BW) or TAT-BH4 mutant peptides and were exposed to 20 Gy of X-ray irradiation after 2 h. At the indicated times, an equivalent segment of small intestine was excised and homogenized in isotonic buffer (20 mM potassium HEPES (pH 7.4), 10 mM KCl, 1.5 mM MgCl2, 250 mM sucrose, 1 mM EDTA, 0.1 mM PMSF, and 1 mM DTT), and the cytosolic fraction was obtained by centrifugation (45 000 g, for 1 h). Then, an equivalent amount of proteins was subjected to assay of caspase-3(-like) activity and Western blotting with antiactive caspase-3 antibody. For CaspACE FITC-VAD-FMK staining, mice were injected intraperitoneally with 13.2 mg/g BW of CaspACE FITC-VAD-FMK at 3 h after X-ray exposure and small intestine was excised 40 min later. At 8 h after X-ray exposure, another segment of small intestine was also excised and frozen at 801C in Tissue-Tek for TUNEL staining. To investigate fulminant hepatitis, 1 mg/g BW of TAT-BH4 peptide was preinjected intraperitoneally into C57BL6J mice, and 0.5 mg/g BW of anti-Fas monoclonal antibody (Jo-2) was injected intraperitoneally after 30 min. At the indicated times, serum aspartate aminotransferase (AST) activity was measured and the livers were dissected out to allow histological

examination. For survival assay, a similar experiment was performed except for 5 mg/g BW of TAT-BH4 peptide and 1.5 mg/g BW of Jo-2 antibody. The cardiac ischemia–reperfusion experiments were carried out as described (Suzuki et al., 1997). Briefly, Sprague–Dawley rats (B250 g) were anesthetized with pentobarbital and anticoagulated with intravenous heparin. Their hearts were quickly excised and perfused with modified Krebs–Henselite buffer (mKH; 120 mM NaCl, 4.5 mM KCl, 20 mM NaHCO3, 1.2 mM KH2PO4, 1.2 mM MgCl2, 2.5 mM CaCl2, and 10 mM glucose; gassed with 95% O2 þ 5% CO2 to obtain pH 7.4 at 371C) at a pressure of 1 m H2O using a Langendorff apparatus. A thinwalled latex balloon was inserted into the left ventricle via the left atrium to monitor the left ventricular pressure. After stabilization, maximum dP/dt (an indicator of cardiac systolic function) was measured with the LV diastolic pressure stabilized at 10 mmHg. Next, TAT-BH4 peptide (0.4 mg) or its mutant peptides were given through the side-arm of the perfusion apparatus at a rate of 20 mg/min for 20 min. Subsequently, the heart was subjected to warm ischemia at 371C for 40 min, followed by 60 min of reperfusion. The balloon was deflated during ischemia. Maximum dP/dt was measured before ischemia and after 0, 15, 30, 45, and 60 min of reperfusion. Finally, mitochondria were isolated from the perfused hearts, as described elsewhere (Solem and Wallace, 1993). Briefly, each heart was homogenized with a glass Teflon Potter homogenizer in medium containing 0.3 M mannitol, 10 mM HEPES/K þ (pH 7.4), 0.2 mM EDTA/K þ , 0.17 mg/ml Nagase, and 0.1% BSA (fatty acid free). After centrifugation at 2500 g for 5 min, the mitochondria were washed twice in the same medium without EDTA and Nagase. Hypoxia–reoxygenation experiments using mitochondria isolated from rat livers were performed as described (Shimizu et al., 1994). Briefly, isolated mitochondria were subjected to hypoxia using N2 (99%) gas for 30 min, and then aerated (reoxygenation) for 60 min. TAT-BH4 and TAT-BH4 mutant2 peptides were added just before hypoxic treatment. Histological examination From tissues frozen in Tissue-Tek, 7 mm sections were cut and stained with hematoxylin and eosin using a standard method. To visualize apoptotic cells, all samples were simultaneously subjected to TUNEL staining using an ApopTag Detection Kit (Intergen). Biochemical analysis Mitochondrial respiration was measured using an O2 electrode in the presence of succinate (state IV) or succinate plus ADP (state III), and the RCR was calculated as the ratio of state III to state IV. Caspase-3(-like) activity was measured as described elsewhere (Shimizu et al., 1996b). Briefly, the cytosol was suspended in 50 mM Tris-Cl (pH 7.4), 1 mM EDTA, and 10 mM EGTA, and was then incubated at 251C with 10 mM of an enzyme substrate (Ac-DEVD-MCA). The release of 7-amino-4-methylcoumarin (AMC) was measured using a fluorescence microplate reader (CytoFluor, Applied Biosystems) with excitation at 360 nm and emission at 460 nm. One unit was defined as the amount of enzyme that liberated 1 mmol of AMC in 1 min. AST (EC 2.6.1.1) activity was measured by the standard procedure recommended by the German Society for Clinical Chemistry. All values are expressed as means and s.d. Statistical evaluation Statistical evaluation was performed by the paired t-test, or analysis of variance. Scheffe’s test was used for Oncogene


8440 individual comparisons of groups when a significant change was observed by analysis of variance. Statistical evaluation for survival assay was performed by the log-rank test. A P-value of less than 0.01 was considered statistically significant.

Abbreviations BH, Bcl-2 homology; PTD, protein transduction domain; VDAC, voltage-dependent anion channel; RCR, respiratory

control ratio; mAST, mitochondrial aspartate aminotransferase. Acknowledgements This study was supported in part by a grant for Scientific Research on Priority Areas, a grant for Center of Excellence Research, a grant for Scientific Research from the Ministry of Education, Science, Sports, and Culture of Japan, and by Special Coordination Funds of Promoting Science and Technology from the Science and Technology Agency of Japan.

References Adams JM and Cory S. (1998). Science, 281, 1322–1326. Borner C, Martinou I, Mattmann C, Irmler M, Schaerer E, Martinou JC and Tschopp J. (1994). J. Cell Biol., 126, 1059–1068. Du C, Fang M, Li Y, Li L and Wang X. (2000). Cell, 102, 33–42. Eskes R, Desagher S, Antonsson B and Martinou JC. (2000). Mol. Cell. Biol., 20, 929–935. Green DR and Reed JC. (1998). Science, 281, 1309–1312. Hanada M, Aime-Sempe C, Sato T and Reed JC. (1995). J. Biol. Chem., 270, 11962–11969. Hirotani M, Zhang Y, Fujita N, Naito M and Tsuruo T. (1999). J. Biol. Chem., 274, 20415–20420. Hockenbery DM. (1992). Semin. Immunol., 4, 413–420. Huang DC, Adams JM and Cory S. (1998). EMBO J., 17, 1029–1039. Hunter JJ, Bond BL and Parslow TG. (1996). Mol. Cell. Biol., 16, 877–883. Kuwana T, Mackey MR, Perkins G, Ellisman MH, Latterich M, Schneiter R, Green DR and Newmeyer DD. (2002). Cell, 111, 331–342. Lee LC, Hunter JJ, Mujeeb A, Turck C and Parslow TG. (1996). J. Biol. Chem., 271, 23284–23288. Macho A, Calzado MA, Jimenez-Reina L, Ceballos E, Leon J and Munoz E. (1999). Oncogene, 18, 7543–7551. Martinou JC and Green DR. (2001). Nat. Rev. Mol. Cell Biol., 2, 63–67. Niwa H, Yamamura K and Miyazaki J. (1991). Gene, 108, 193–199. Oltvai ZN, Milliman CL and Korsmeyer SJ. (1993). Cell, 74, 609–619. Park IW, Ullrich CK, Schoenberger E, Ganju RK and Groopman JE. (2001). J. Immunol., 167, 2766– 2771. Saikumar P, Dong Z, Weinberg JM and Venkatachalam MA. (1998). Oncogene, 17, 3341–3349. Saito M, Korsmeyer SJ and Schlesinger PH. (2000). Nat. Cell Biol., 2, 553–555. Schwarze SR, Hruska KA and Dowdy SF. (2000). Trends Cell Biol., 10, 290–295. Schwarze SR, Ho A, Vocero-Akbani A and Dowdy SF. (1999). Science, 285, 1569–1572.

Oncogene

Shibasaki F, Kondo E, Akagi T and McKeon F. (1997). Nature, 386, 728–731. Shimizu S, Kamiike W, Hatanaka N, Nishimura M, Miyata M, Inoue T, Yoshida Y, Tagawa K and Matsuda H. (1994). Transplantation, 57, 144–148. Shimizu S, Eguchi Y, Kamiike W, Itoh Y, Hasegawa J, Yamabe K, Otsuki Y, Matsuda H and Tsujimoto Y. (1996a). Cancer Res., 56, 2161–2166. Shimizu S, Eguchi Y, Kamiike W, Matsuda H and Tsujimoto Y. (1996b). Oncogene, 12, 2251–2257. Shimizu S, Narita M and Tsujimoto Y. (1999). Nature, 399, 483–487. Shimizu S, Ide T, Yanagida T and Tsujimoto Y. (2000a). J. Biol. Chem., 275, 12321–12325. Shimizu S, Konishi A, Kodama T and Tsujimoto Y. (2000b). Proc. Natl. Acad. Sci. USA, 97, 3100–3105. Shimizu S, Matsuoka Y, Shinohara Y, Yoneda Y and Tsujimoto Y. (2001). J. Cell Biol., 152, 237–250. Solem LE and Wallace KB. (1993). Toxicol. Appl. Pharmacol., 121, 50–57. Suzuki K, Sawa Y, Kaneda Y, Ichikawa H, Shirakura R and Matsuda H. (1997). J. Clin. Invest., 99, 1645–1650. Tsujimoto Y and Shimizu S. (2000a). FEBS Lett., 466, 6–10. Tsujimoto Y and Shimizu S. (2000b). Cell Death Differ., 7, 1174–1181. Tsujimoto Y. (2003). J. Cell. Physiol., 195, 158–167. Verhagen AM, Ekert PG, Pakusch M, Silke J, Connolly LM, Reid GE, Moritz RL, Simpson RJ and Vaux DL. (2000). Cell, 102, 43–53. Wang HG, Rapp UR and Reed JC. (1996). Cell, 87, 629–638. Wang X. (2001). Genes Dev., 15, 2922–2933. Wei MC, Lindsten T, Mootha VK, Weiler S, Gross A, Ashiya M, Thompson CB and Korsmeyer SJ. (2000). Genes Dev., 14, 2060–2071. Yamabe K, Shimizu S, Kamiike W, Waguri S, Eguchi Y, Hasegawa J, Okuno S, Yoshioka Y, Ito T, Sawa Y, Uchiyama Y, Tsujimoto Y and Matsuda H. (1998). Biochem. Biophys. Res. Commun., 243, 217–223. Yang E, Zha J, Jockel J, Boise LH, Thompson CB and Korsmeyer SJ. (1995). Cell, 80, 285–291. Zamzami N and Kroemer G. (2001). Nat. Rev. Mol. Cell Biol., 2, 67–71.