Caspase-3 activation during apoptosis caused by glutathione–doxorubicin conjugate

Glutathione–doxorubicin (GSH–DXR) effectively induced apoptosis in rat hepatoma cells (AH66) at a lower concentration than DXR. After 24 h of drug treatment, DNA fragmentation of the cells was observed at the concentration of 1.0 μM DXR or 0.01 μM GSH–DXR. Increase in caspase-3 activity and DNA fragmentation were observed within 12 h and 15 h after treatment with either drug. Intracellular caspase-3 activity was increased in a dose-dependent manner after treatment with DXR or GSH–DXR, and caspase-3 activity correlated well with the ability to induce DNA fragmentation. When the cells were treated with either DXR or GSH–DXR for only 6 h, apoptotic DNA degradation and caspase-3 activation occurred 24 h after treatment. DNA fragmentation caused by these drugs was prevented completely by simultaneous treatment with the caspase-3 inhibitor, acetyl–Asp–Glu–Val–Asp-aldehyde (DEVD-CHO), at 10 μM. By contrast, DNA fragmentation was not prevented by the caspase-1 inhibitor, acetyl–Tyr–Val–Ala–Asp–aldehyde (YVAD-CHO), at the same concentration as DEVD-CHO, and caspase-1 was not activated at all by the treatment of AH66 cells with both DXR and GSH–DXR. These results demonstrate that DXR and GSH–DXR induce apoptotic DNA fragmentation via caspase-3 activation, but not via caspase-1 activation, and that GSH–DXR enhances the activation of caspase-3 approximately 100-fold more than DXR. Moreover, the findings suggested that an upstream apoptotic signal that can activate caspase-3 is induced within 6 h by treating AH66 cells with the drug. © 1999 Cancer Research Campaign


Conjugation of DXR with GSH
GSH-DXR was prepared as described previously (Asakura et al, 1997a). In brief, 1 mg of GSH and 0.5 mg of DXR in 0.5 ml of 0.15 M sodium chloride (NaCl) containing 0.1% glutaraldehyde were incubated at room temperature for 30 min. After incubation, GSH-DXR was separated from GSH and DXR using Dowex 50Wx8 (H ϩ form, 5 ϫ 15 mm). The concentration of DXR was measured by absorbance at 495 nm.

Preparation of cell extract
After treatment of AH66 cells with DXR or GSH-DXR, harvested cells were washed with ice-cold 0.15 M NaCl and lysed with ice-cold 0.5% Triton X-100 containing 10 mM Tris-HCl (pH 8.0) and 10 mM EDTA. The cell lysate was spun down at 10 000 g for 10 min and the supernatant was used for the assays of DNA fragmentation and caspase activity.

DNA fragmentation assay
After treatment of the cells (2 ϫ 10 6 ) with DXR or GSH-DXR in the presence or absence of a caspase family inhibitor, the cell extract containing fragmented DNA was incubated with 0.5 mg ml Ϫ1 RNase A at 37ЊC for 60 min, then with 0.5 mg ml Ϫ1 proteinase K at 37ЊC for 60 min. After incubation, fragmented DNA precipitated by isopropanol was dissolved with 10 mM  Tris-HCl (pH 8.0), 1 mM EDTA, 5% glycerol and 0.05% bromophenol blue. The DNA fragments, separated by 2% agarose gel electrophoresis, were stained with ethidium bromide, and photographed on a UV transilluminator. The 123 base pair DNA ladder (Gibco, BRL, NY, USA) was used as the standard DNA fragments.

Assay of caspase-3 activity
Reaction mixtures, which contained 100 µM of DEVD-MCA, the appropriate protein concentration of cell extract, 50 mM HEPES-NaOH (pH 7.5), 10% glycerol and 2 mM dithiothreitol with or without 0.1 µM DEVD-CHO, were monitored for AMC liberation at 37ЊC for 15 min in a spectrofluorometer at an excitation wavelength of 380 nm and an emission wavelength of 460 nm . The caspase-3 proteolytic activity was expressed as the difference between nmol AMC liberations in the presence and absence of the inhibitor per min per mg protein. When the activity of caspase-1 was assayed, 20 µM YVAD-MCA and 0.1 µM YVAD-CHO were substituted for 100 µM DEVD-MCA and 0.1 µM DEVD-CHO in the reaction mixture respectively.

Protein determination
Protein concentration was assayed by a Bio-Rad protein assay kit (Bio-Rad Lab., Tokyo, Japan) using BSA as the standard.

Induction of apoptosis by GSH-DXR
When AH66 cells were continuously exposed to 3 µM DXR or 0.1 µM GSH-DXR for 24 h, the cell viability determined with a colourimetric assay (Ohkawa et al 1993;Asakura et al, 1997aAsakura et al, , 1997b, was decreased to approximately 50% compared with the non-treated cells (data not shown). Inter-nucleosomal DNA fragmentation, a biochemical feature of the apoptotic process, was observed in the cells treated with the drugs. DNA fragmentation occurred within 15 h after continuous treatment with 3 µM DXR or 0.1 µM GSH-DXR ( Figure 1A), and concentrations of DXR and GSH-DXR as low as 1.0 and 0.01 µM, respectively, were found to induce DNA fragmentation at 24 h of incubation ( Figure 1C). This result indicates that GSH-DXR is a potent inducer of apoptosis as compared with DXR. In our recent report (Asakura et al, 1997b), the cytotoxicity of GSH-DXR in AH66 cells was 170-fold higher than that of DXR. Therefore, the extent of cytotoxicity for DXR and GSH-DXR corresponded to the magnitude of apoptosis induced by treatment with these drugs. Moreover, GSH-DXR showed approximately tenfold more cytotoxic activity than other large molecular weight derivatives of DXR, such as DXR conjugated with BSA or with oxidized glutathione against AH66 cells (Asakura et al, 1997a). On the other hand, cytotoxicities of DXR coupled to several small peptides, such as glycylglycine and glycylglycylglycine, demonstrated almost the same cytotoxic activity as DXR (Asakura et al, 1997a). The magnitude of apoptosis induced by treatment with these conjugates also corresponded to the extent of cytotoxicity for these drugs (data not shown).

Inhibition of apoptosis by caspase inhibitor
In order to determine the kind of proteases involved in the apoptotic process, the effects of two cysteine protease inhibitors, YVAD-CHO (Thornberry et al, 1992) and DEVD-CHO , on the apoptosis of AH66 cells were determined. As shown in Figure 2A, DEVD-CHO strongly inhibited drug-induced DNA fragmentation in a dose-dependent manner, and 10 µM of the inhibitor completely blocked DNA fragmentation. By contrast, YVAD-CHO (10 µM) did not exhibit any 3.2 ± 1.7 58.7 ± 4.3 41.3 ± 5.5 10.8 ± 3.1 5.3 ± 1.9 60.2 ± 7.2 127.4 ± 10.8 59.6 ± 6.9 33.3 ± 2.7 7.4 ± 2.0 120.9 ± 20.4 B A Figure 2 Prevention of drug-induced apoptosis by caspase inhibitor. AH66 cells were treated with 3 µM DXR or 0.1 µM GSH-DXR, and DEVD-CHO (0.1, 1 or protective effect on drug-induced apoptosis. This result suggests that DXR-and GSH-DXR-induced DNA fragmentation occur via activation of caspase-3, but not of caspase-1. Several reports have described that anthracycline was able to induce inter-nucleosomal DNA fragmentation in treated cells (Kaufmann et al, 1993;Bose et al, 1995;Yamashita et al, 1995;Chen et al, 1996;Mizushima et al, 1996), but the magnitude of the process induced by GSH-DXR treatment was markedly more potent than that induced by DXR.

Activation of caspase-3 by treatment with GSH-DXR
When AH66 cells were treated continuously with 3 µM DXR or 0.1 µM GSH-DXR, caspase-3 proteolytic activity in the cells did not increase until 9 h, and increased linearly thereafter to a level approximately 20-or 50-fold higher than that of the non-treated control by 24 h respectively ( Figure 1B). By treating with a higher concentration of the drugs (100 µM DXR or 10 µM GSH-DXR), enhancement of caspase-3 activity was not observed until 9 h (data not shown). Therefore, caspase-3 was not activated until 9 h after treatment with the drugs independent of the drug concentration. These drugs increased caspase-3 activity in a dose-dependent manner ( Figure 1D). However, caspase-1 proteolytic activity was not increased in cells treated with DXR or GSH-DXR in any time period (data not shown). This result suggests that GSH-DXR enhances the activity of caspase-3 about 100-fold more than DXR-induced activation and the magnitude of the activation induced by treatment with the drugs correlates with the extent of DNA fragmentation.
When AH66 cells were treated simultaneously with DEVD-CHO and the drug (3 µM DXR or 0.1 µM GSH-DXR) for 24 h, cellular caspase-3 activity failed to increase ( Figure 2B). However, YVAD-CHO (10 µM) did not affect drug-induced activation of caspase-3. It has been reported that active caspase-3 is generated from its inactive precursor form by other active caspases via Fas (Enari et al, 1996). Therefore, this result suggests that caspase-1 does not participate in proteolytic activation of caspase-3 in druginduced apoptosis.

DEVD-CHO prevents drug-induced apoptosis pathway
To examine whether or not DEVD-CHO could inhibit the initial DXR-or GSH-DXR-induced DNA damage or the apoptotic signal pathway itself, the cells were co-treated with the drug and DEVD-CHO in various time schedules (Figure 3). When the cells were treated with DXR or GSH-DXR for 6 h, DNA fragmentation was induced 24 h after the treatment with the drug. It was demonstrated that treatment of AH66 cells with the drug for 6 h was enough to commit the cells to apoptosis (Figure 3, lanes 2 and 6). Although the apoptotic signal was induced by 6-h treatment with DXR and GSH-DXR, caspase-3 activation and DNA fragmentation occurred 12 h and 15 h after treatment respectively (Figure 1  A,B). On the other hand, after treating AH66 cells for 6 h with DXR or GSH-DXR, the addition of 10 µM DEVD-CHO blocked both DNA fragmentation and caspase-3 activation for as long as 24 h (Figure 3, lanes 3 and 8). However, when DEVD-CHO was washed out 12 h after the treatment with the drug, apoptotic DNA degradation occurred 12 h after the wash-out (Figure 3, lanes 5 and 9). This result suggests that DEVD-CHO can inhibit the following apoptotic signal pathway, but does not affect the initial drug-induced DNA damage in AH66 cells. By washing out the inhibitor, caspase-3 activity in the cells was increased as compared with that in the non-treated cells (5.1 to 72.5 pmol mg Ϫ1 min Ϫ1 ). These results indicate that DEVD-CHO does not cause the GSH-DXR-induced apoptotic signal to disappear, but the signal for caspase-3 activation is temporarily suppressed. Moreover, the findings suggested that an upstream apoptotic signal able to activate caspase-3 was already induced by treatment of AH66 cells with DXR or GSH-DXR for 6 h.
GSH-DXR was synthesized by the conjugation between both amino groups of DXR and of GSH via glutaraldehyde. Since the SH group of GSH-DXR determined by fluorescent method using o-phthalaldehyde showed the same concentration as DXR, it was demonstrated that the SH group of GSH was present (data not shown). In our recent reports (Asakura et al., 1997a(Asakura et al., , 1997b, the SH group on the cysteine of GSH-DXR was important for enhancement of the cytotoxicity, and GSH-DXR inhibited potent