- Email: [email protected]
Effect of antioxidants on the apoptosis of CHO cells and production of tissue plasminogen activator in suspension culture
JOURNAL Or BIOSClEIqCE AND BIOEIqOI/.~I~RIN'O Vol. 91, No. 6, 581-585. 2001
Effect of Antioxidants on the Apoptosis of CHO Cells and Production of Tissue Plasminogen Activator in Suspension Culture ZHANYOU YUN, MUTSUMI TAKAGI,* AND TOSHIOMI YOSHIDA International Center for Biotechnology, Osaka University, 2-I Yamada-oka, Suita, Osaka 565-0871, Japan
Received 12 January 2001/Accepted27 March 2001 The effects of antioxidants on the apoptosis of Chinese hamster ovary (CHO) 1-15,$00 cells and production of tissue plasminogen activator (tPA) in suspension culture were investigated. After cell growth to 2 x 10 s cells/ml in Ham's F12 medium containing 1 0 ~ newborn bovine serum (NBS) in a spinner bottle, CHO cells were maintained for 6 d in Ham's F12 medium containing 0 or 0 . 4 ~ NBS and 10 mM antioxidants, namely, L-ascorbic acid 2-phosphate (VCP) or the reduced form of glutathione (GSH]. The viable cell concentrations at day 6 in the serUm-free culttire with GSH and in the low-serum culture with VCP or GSH were 0.57, 1.04 and 1.69 x 10 s cells/ml, respectively, while those in the serum.free and low-serum cultures without the antioxidants were only 2.33 and 1.17 x 103 cells/ml, respectively. The percentages of apoptutic cells in the serum-free and low-serum cultures with VCP (73.2, 4 4 . 6 ~ ) and GSH (76.9, 3 8 . 6 ~ ) measured using a flowcytometer after annexin V-FITC/propidium iodide staining were markedly lower than those in the cultures without antioxidants (96.3, 9 2 . 5 ~ ) . The percentage of cells having a high mitochondriai membrane potential among the viable cells in the cultures with antioxidants determined using a flowcytometer after 5,5',6,6'-tetrachloro-l,l',3,3'tetraethylbenzimidazulyicarbocyanine iodide staining was clearly higher than those in the cultures without the antioxidants. The production of tPA in the sernm.free and low-serum media with VCP (0.282, 2.02 mg//) or GSH (1.01, 1.61 rag//) was markedly higher than that in the cultures without the antioxidants (0.275, 0.689 mg//). Consequently, the suppression of apoptosis through the maintenance of the membrane potential of mitochondria by VCP or GSH resulted in a marked increase in tPA production by CHO cells in the serum-free and low-serum cultures.
[Key words: apoptosis, CHO cells, suspension culture, antioxidant, tissue plasminogen activator] Mammalian cells can produce secondary metabolite proteins such as monoclonal antibodies and tissue plasminogen activator (tPA) even after the cessation of cell growth during batch culture. Moreover, the production rate after cell growth is markedly higher than that during cell growth due to a high specific production rate during the post-growth period (1). However, it is not easy to maintain a high cell activity level during the production phase due to poor nutritional conditions in a serumfree or low-serum medium, which should be employed in order to reduce the contamination of the product with serum proteins in the downstream production process. Therefore, it is important to develop a strategy to increase productivity by inhibiting cell death during the production phase. Cell death during the production phase may proceed via both the necrotic and apoptotic pathways. Apoptosis in a bioreactor was first reported for the cultivation of murine B-cell hybridoma (2). Solis-Recendez et al. reported that the majority of 32 hybridoma cell lines exhibited apoptosis before reaching maximum cell density even when the cells were cultured in a serum-supplementeA medium (3). Measurement of DNA fragmentation by gel chromatography showed that more than 30% of hybridoma cells were apoptotic at the end of batch culture (4). The percentage of apoptotic cells among the dead cells of both plasmacytoma and hybridoma cell lines was estimated to be 90% based on the nuclear morphology (5, 6). Chinese hamster ovary (CHO) cells, which are one of the most widely used animal cell lines for the production
of pharmaceuticals (7), also died mainly through apoptosis during serum-free cultivation (8). It is expected that the suppression of apoptosis during the production phase in serum-free or low-serum media might increase protein production by mammalian cells such as CHO cells. Apoptosis can be induced by the addition of cytokines such as glucocorticoids, thyroid hormones, TNF a, and TGF 31, a deficiency in growth factors, the presence of cytotoxic factors such as perforins, granzymes, and reactive oxygen species, and application of physical factors such as radiation, hyperthermia, and hydrodynamic tension (9). These factors initiate the signal transduction cascades involving the collapse of mitochondrial membrane potential (A~'m), uncoupling of the respiratory chain, over-production of superoxide anions, local oxidation of inner membrane cardiolipins, release of apoptogenic proteins such as cytochrome c, apoptosis-inducing factor (AIF) and caspases from the mitochondrial intermembrane space into the cytosol, potassium eiflux, cytosol calcium increase, chromatin condensation, and DNA fragmentation (10). Suppression of apoptosis during the cultivation process has been investigated by manipulating nutrient feeding (5, 11) and using antiapoptotic genes such as bcl-2 and crm A (12, 13). Several kinds of apoptosis-suppressing chemicals were also investigated for a wide range of applications in cell culture processes. However, the caspaseinhibiting peptides, such as N-benzyloxycarbonyl-ValAla-Asp-fluoromethyl-kctone (Z-VAD.fmk) and acetylAsp-Glu-Val-Asp-aldehyde (14), as well as the mitochondriai membrane-permeability inhibitors, such as bongkrekic acid and cyclosporin A (14--16), are toxic and
* Corresponding author, e-mail: [email protected] phone: +81-(0)6-6879-7456 fax: +81-(0)6-6879-7454 581
582
YUN ET AL.
expensive for practical application to pharmaceutical manufacturing processes. Although antioxidants such as N-acetylcysteine (14, 17-19), pyrrolidine dithiocarbamate (14, 20) and ascorbic acid (21-23) were reported to suppress apoptosis and are less expensive and not toxic, the effects of antioxidants on the productivity of cells have not been reported except in a study by Mastrangelo et al. (14), in which N-acetylcysteine, pyrrolidine dithiocarbamate, bongkrekic acid and Z-VAD.fmk suppressed the apoptosis caused by viral infection and increased the production of chloramphenicol acetyltransferase by virus-infected AT3-neo cells. However, there are few reports on the effect of antioxidants on apoptosis caused by serum depletion and on protein productivity of common cell lines such as CHO cells. Moreover, there has been no discussion about the mechanism of apoptosis suppression by antioxidants. In this study, the effects of two antioxidants, namely, L-ascorbic acid 2-phosphate (VCP) and the reduced form of glutathione (GSH), on apoptosis and production of tPA by CHO 1-15,500cells during suspension culture in serum-free and low-serum media were investigated. MATERIALS AND METHODS
Cells and media
CHO 1-15,500cells producing tPA (ATCC CRL-9606) (1) were used. Ham's F-12 medium (Gibco, Tokyo) supplemented with 10% NBS (Gibco), streptomycin (0.1 mg/ml), penicillin (100u/ml) and 500 nM of methotrexate (Sigma, MO, USA) was employed for the cell growth phase. The serum concentration in the medium was reduced to 0 or 0.4% and 10mM each of VCP (Sigma) and GSH (Sigma) were added during the production phase. We selected 10 mM as the concentration for antioxidant addition to the spinner bottle culture after preliminary experiments with the still culture of CHO cells in a tissue-culture multiwell plate employing F12 medium supplemented with 0.4% NBS and antioxidants (0-10 mM). Cell cultivation Cells (4.1 × 106 cells) harvested from a dish (55cm 2, Corning Inc., NY, USA) by trypsinization were inoculated into a spinner bottle (10Oml, Shibata Hario Co., Tokyo) containing 30 mi of the cellgrowth medium, and then incubated in a water bath (37°C, 5% CO2, 70rpm). The medium was changed once during the cell-growth phase. The cells at the stationary phase were washed with 40 ml of PBS (pH 7.4), suspended in the production medium (2x 105cells/ml) and incubated further in a spinner bottle for 6 d. Cell cultivation was repeated three times and similar effects of antioxidants were obtained in all three cultivations. Typical data from these cultivations are shown in Results. General analysis Cell concentration and viability were determined by the trypan-blue dye exclusion method. The concentration of tPA was assayed by ELISA (Imulyse tPA, Biopool AB, Sweden). Counting apoptotic and necrotic cells Cells were stained according to the protocol for the Annexin V(AV)-FITC kit (IM2376, Immunotech, Marseille, France). Namely, 1 × 105 cells were washed once with PBS (pH 7.4) and double stained with AV-FITC (5 pM) and propidium iodide (PI, 5 pM) on ice in the dark for 10min. The A V / P I stained cells were kept on ice until measurement using a flowcytometer (EPICS XL, Beck-
J. BIOSCI.BIOENG., 10 e
l
O
S
~
lo ~ 0
2
4
6
Culturetime(d) FIG. 1. Effects of the antioxidants on viable cell concentration during the serum-free and low-serum cultivation of CHO cells. The serum-free (open symbols)and low-serum(closedsymbols)cultivation was performed without (©, e) or with 10 mM antioxidants of VCP (n, • ) or GSH (v, v), respectively. Viable cell concentration was determined by the trypan-blue method. man-Coulter, Miami, USA). The excitation wavelength was 488 nm and the emission intensity was measured at 525 and 675 nm. The settings of the flowcytometer were: FS, 35mV; SS, 242mV; FL1, 499mV; FL4, 588mV; compensation, FL1- FL4 0.3%; and FL4-FL1 0.2%.
Measurement of mitoehondrial membranepotential Cells (1 × 105) were washed once with PBS (pH7.4) and stained with 10pM 5,5',6,6'-tetrachloro-l,l',3,3'-tetra ethylbenzimidazolylcarbocyanine iodide (Jc-1, Molecular Probes, Eugene, USA) in 0.Sml of PBS at 37°C for 10rain in the dark. Stained cells were kept on ice until the analysis using the flowcytometer. The excitation wavelength was 488 nm and emission intensity was measured at 525 and 575 nm. The settings of the flowcytometer were: FS, 35mV; SS, 242mV; FL1, 450mV; FL2, 510mV; compensation, FL1- FL2 3%; and FL2-FL1 22%.
RESULTS Effect of antioxidants on cell viability
Because the objective of this research was to investigate the suppression of apoptosis by antioxidants during the production phase, ceils (2x 105 cells/ml, 97% viability) were transferred from the growth medium (10% serum) to a medium with or without 0.4% serum supplemented with antioxidants, in which cells could not grow, and the viable TABLE 1. Percentagesof apoptotic and necrotic cells during the serum-free and low-serum cultivation Serum Apoptotic cells (%) Necroticcells (%) (%) Antioxidant ld 3d 6d ld 3d 6d -14.4 6 6 . 4 92.5 4.9 8.6 6.3 VCP 22.7 4 8 . 2 73.2 21.9 2 7 . 2 23.5 GSH 27.3 57.3 76.9 13.9 11.3 11.9 0.4 -8.8 53.1 96.3 12.0 2.3 1.4 VCP 9.1 17.6 44.6 11.2 3.3 7.1 GSH 29.7 19.8 38.6 8.4 4.8 7.9 The percentage of apoptotic and necrotic cells among the total number of cells during the serum-free and low-serumcultivation with or without the antioxidants (10 raM) VCP and GSH was determined using a fiowcytometerafter AV/PI staining.
VOL. 91, 2001
EFFECT OF ANTIOXIDANTS ON APOPTOSIS OF CHO CELLS I00
o 0
2
4
g
Culture time (d) FIG. 2. Effect of the antioxidants on mitochondrial membrane potential during the serum-free and low-serum cultivation of CHO cells. Mitocliondrial membrane potential of cells during the serumfree and low-serum cultivation shown in Fig. 1 was analyzed using a flowcytom~er after Jc-1 staining. Symbols are the same as those in Fig. 1. cell concentration and cell viability during the production phase were determined thereafter. While the viable cell concentration decreased rapidly to 1/100 o f the initial concentration in 6 d in the cultures without antioxidants (Fig. 1), a marked decrease in viable cell concentration was not observed in the low-serum culture supplemented with V C P or GSH, resulting in a viability higher than 70% at 6 d, indicating that serum effectively protected cells from death. Moreover, it is notable that the viable cell concentration at 6 d in the serum-free culture supplemented with G S H was maintained at 1/4 o f the initial concentration. Apoptosis during cultivation with the antioxidants In order to clarify which type o f cell death, that is apoptosis or necrosis, was suppressed by the antioxidants, cells in the production phase in the serum-free and lowserum media supplemented with the antioxidants were analyzed after A V / P I staining using a flowcytometer. The proportion o f apoptotic cells a m o n g the total number of cells markedly increased in the serum-free and low-serum media without antioxidants and were higher than 90% at 6 d (Table 1). On the other hand, the proportion o f apoptotic cells was markedly lower at 3 and 6 d in the cultures with the antioxidants than in those without the antioxidants, although the proportion of necrotic cells in the cultures with the antioxidants was slightly higher than in those without the antioxidants. Mitoehonflrial membrane potential of cells during cul-
583
tivation with the antioxidants In order to investigate the mechanism o f apoptosis suppression by the antioxidants, the amount o f cells with high mitochondrial membrane potential (A~m) was estimated during the serumfree and low-serum culture. The percentage o f such cells among the total number o f cells decreased rapidly and was less than 5% at 6 d in the serum-free and low-serum cultures without antioxidants (Fig. 2). The percentage in the low-serum culture with VCP and GSH was maintained at a higher level than that without an antioxidant after a temporary decrease at 1 d. On the other hand, the increments in the percentage by addition o f antioxidants to the serum-free culture were small. In order to better understand the effect o f antioxidants on the A~m, we attempted to estimate the percentage o f cells with high A~m among the viable cells. We observed two clusters of cells in the flowcytometric histogram o f forward scatter (FS) versus side scatter (SS) as shown in Fig. 3. As FS is generally proportional to cell size, the clusters with high and low FS were designated large-ceU cluster and small-cell cluster, respectively. The proportion of cells classified into to the small-cell cluster increased as the culture progressed in the serum-free (Fig. 3) and even in the low-serum media. There was a proportionai relationship between the viability and the percentage o f cells in the large-cell cluster (Fig. 4). This relationship strongly suggests that the cells in the largecell and small-cell clusters were viable and dead cells, respectively. The proportion of cells with high A~m in the large-cell cluster was determined (Table 2). Most cells in the large-cell cluster maintained a high A'~"m up to 3 d in the serum-free and low-serum cultures regardless of the presence of the antioxidants. However, it is notable that the percentages decreased markedly at 6 d in the absence o f the antioxidants in both the serum-free and low-serum cultures, while they remained high in the presence o f the antioxidants. Effect of antioxidants on t P A productivity The concentration of t P A produced at 6 d in the serum-free and low-serum cultures was determined (Table 3). The concentrations of t P A in the serum-free cultures containing VCP and G S H were 1.1 and 3.7 times that in the culture without antioxidants, respectively. VCP and G S H
'A
100
•fl
80
~
40
°1o 20
0
SS
1024
0
SS
1024
0
SS
1024
FIG. 3. Forward scatter and side scatter of CHO cells during serum-free cultivation. Forward scatter and side scatter of CHO cells were analyzed using a flowcytometer at 0, 1 and 3 d during serum-free cultivation. The region surrounded by a solid line indicates the cluster of large cells.
40
60
80
100
Viability (%) FIG. 4. Correlation between the percentage of cells in the largecell cluster and viability during the serum-free and low-serum cultures. The percentage of cells in the cluster of large cells was plotted against the viability measured by the trypan-blue method for all conditions shown in Fig. 1. The sample number was 42 and the square of the correlation coefficient (r2) was 0.91.
584
J. BlOSCl. BIOI~NG.,
Y U N E T AL.
TABLE 2. Percentage of cells with high mitochondrial membrane potential (AYm)among viable cells during the serum-free and low-serum cultivation Serum (%)
Antioxidant VCP GSH
0.4
--
VCP GSH
ld 97.3 88.4 84.6 98.7 72.0 93.7
High A~m cells (o~) 3d 92.8 92.3 88.0 89.0 96.5 96.6
6d 12.7 89.9 70.2 34.2 90.7 95.8
The percentage of cells with high A~m among the cellsin the cluster of large cells during the serum-free and low-serum cultivation with or without the antioxidants (10 mM) VCP and GSH was determined using a flowcytometer after Jc-I staining. in the low-serum culture increased t P A production 3.0 and 2.4 times, respectively. DISCUSSION We employed F12 medium without serum addition as a serum-free medium, and the medium containing 0.4°~ NBS as a low-serum medium. The difference between the effects o f antioxidants on cell apoptosis in the serum-free and low-serum media was clearly shown, though the celis were washed only once with F12 medium after cell growth in 10% NBS medium and there may be some remaining serum even in the serum-free medium. VCP and GSH markedly suppressed the decrease in the viable cell concentration and cell viability during the low-serum culture (Fig. 1). GSH was also effective for the serum-free culture, while the effect of VCP was not apparent. VCP at higher concentration may be effective even for a serum-free culture. It was dearly shown that apoptosis was the major means o f cell death during both the serum-free and lowserum cultivation without antioxidants (Table 1). This result agrees with the reports that approximately 90% o f the death o f both plasmacytoma and hybridoma cells were due to apoptosis (5, 6) and that C H O cells died through apoptosis during serum-free culture (8). The antioxidants VCP and G S H markedly decreased the percentage o f apoptotic cells during both the serumfree and low-serum cultures, while they slightly increased the percentage o f necrotic cells (Table 1). Apoptosis o f other cell lines was reported to be suppressed by antioxidants such as N-acetylcysteine (NAC) (14, 17-19), pyrrolidine dithiocarbamate (PDTC) (14, 20) and ascorbic acid (VC) (21-23). It was strongly suggested that the suppression of apoptosis by VCP and GSH during the serum-free and low-serum cultures resulted in higher viability and viable cell concentration than in cultures without antioxidants. Because reduction o f A~m is a c o m m o n phenomenon TABLE 3. Production of tPA in the serum-free and low-serum cultures Serum (%)
-tPA (nag//) 0.275
0
VCP 0.282
0.4
GSH 1.012
-0.689
VCP 2.017
GSH 1.611
Concentration of tPA produced by CHO cells during the serum-free and low-serum culture for 6 d with or without the antioxidants (10 mM) VCP and GSH was determined by ELISA.
during apoptosis, the influence o f the antioxidants on the A~ m of the total cells during the serum-free and low-serum cultures was compared with their effect on cell viability (Fig. 1) and the percentage o f apoptotic cells among the total cells (Table 1). The percentage o f cells with a high A~m among the total cells remained higher in the low-serum cultures with both VCP and G S H (Fig. 2), so the suppression o f apoptosis by the antioxidants was considered to be correlated with the maintenance o f a high A~ m. We further attempted to estimate the effect o f the antioxidants on the A~m of only viable cells (cells stained by trypan blue). The number o f cells with a high A~m among the viable cells clearly increased in the presence o f the antioxidants in both the serum-free and lowserum cultures (Table 2), which suggests that VCP and G S H prevented the decrease in A~ m before cell death by apoptosis. The sequential mechanism o f the suppression o f apoptosis by the antioxidants, including the change in A~m, should be investigated further. The antioxidants also increased the production o f t P A (Table 3). This indicates that this effect was concomitant to the effect on viable cell maintenance (Fig. 1), based on the evidence that VCP addition to the serum-free culture had only a small effect on viable cell concentration and t P A production. The increase in t P A production by the antioxidants may be a result o f the suppression o f apoptosis. In the future, we would like to examine the effect o f antioxidants on apoptosis and t P A production employing commercially available serum-free media. The antioxidants V C P and G S H appear to suppress the apoptosis o f C H O cells during serum-free and lowserum suspension cultures due to the increase in A~m. Moreover, they markedly increase the t P A production. Consequently, nontoxic and cheap antioxidants such as VCP and GSH should have great potential for application to practical production processes o f pharmaceuticals using mammalian cells. REFERENCES 1. Takagi, M., Hayashi, H., and Yoshida, T.: Effects of osmolarity on metabolism and morphology in adhesion and suspension of Chinese hamster ovary cell producing tissue plasminogen activator. Cytotechnology, 32, 171-179 (2000). 2. AI-Rubeal, M., Mills, D., and Emery, A.N.: Electron microscopy of hybridoma cells with special regard to monoclonal antibody production. Cytotechnology, 4, 13-28 (1990). 3. Solis-Recendez, M.G., Pirani, A., D'Habit, B., Stacey, G.N., and Maugras, M.: Hybridoma cell cultures continuously
undergo apoptosis and reveal a novel 100 bp DNA fragment. J. Biotechnol., 38, 117-127 (1995). 4. Franek, F. and Dolnikova, J.: Nucleosomes occuring in protein-free hybridoma cell cultures. Evidence for programmed cell death. FEBS. Lett., 248, 285-287 (1991). 5. Singh, R.P., AI-Rubeal, M., Gregory, C.D., and Emery,
A.N.: Cell death in bioreactors: a role for apoptosis. Biotechnol. Bioeng., 44, 720-726 (1994). 6. Mercllle, S. and Massie, B.: Induction of apoptosis in nutrient limited cultures of hybridoma cells. Biotechnol. Bioeng., 44, 1140-1154 (1994). 7. Fusseuegger, M. and Bailey, J. E.: Molecular regulation of cellcycle progression and apoptosis in mammalian cells: implications for biotechnology. Biotechnol. Prog., 14, 807-833 0998). 8. Moore, A., Donahue, C.J., Hooley, J.S., Stocks, D.L., Bauer, K.D., and Mather, J.P.: Apoptosis in CHO cell batch
cultures: examination by flow cytometry. Cytotechnology, 17, 1-11 (1995). 9. Motyl, T.: Regulation of apoptosis: involvement of bcl-2- feint-
VOL 91, 2001
EFFECT OF ANTIOXIDANTS ON APOPTOSIS OF CHO CELLS
ed protenins. Reprod. Nutr. Dev., 39, 49-59 (1999). 10. Dallaporta, B., Marehetti, P., de-Pablo, M.A., Maisse, C., Due, H.T., Metivter, D., Zamzami, N., Geuskens, M., and Kroemer, G.: Plasma membrane potential in thymocyte apoptosis. J. Immunol., 162, 6534-6542 (1998). 11. Franek, F. and Sramkova, K.: Protection of B lymphocyte hybridoma against starvation-induced apoptosis: survival signal role of some amino acids. Immunol. Lett., 52, 139-144 (1996). 12. Chnng, J. D., Sinskey, A.J., and Stephanopoulos, G.: Growth factor and bcl-2 mediated survival during abortive proliferation of hybridoma cell line. Biotechnol. Bioeng., 57, 164-171 (1998). 13. Reynolds, J. E., Li, J., Craig, R. W., and Eastman, A.: BCL-2 and MCL-1 expression in Chinese hamster ovary cells inhibits intracellular acidification and apoptosis induced by staurosporine. Exp. Cell Res., 225, 430-436 (1996). 14. Mastrangelo, A.J., Zon, S., Hardwick, J.M., and Betenbangh, M. L: Antiapoptosis chemicals prolong productive lifetimes of mammalian cells upon Sindbis virus vector infection. Biotechnol. Bioeng., 65, 298-305 (1999). 15. Zamzami, N., Mnrehetti, P., Castedo, M., Hirsch, T., Snsin, S.A., Masse, B., and Kroemer, G.: Inhibitors of permeability transition interfere with the disruption of the mitochondrial transmembrane potential during apoptosis. FEBS Lett., 384, 53-57 0996). 16. Halestrap, A.P., Woodfleid, K.Y., and Connem, C.P.: Oxidative stress, thiol reagents, and membrane potential modulate the mitochondrial permeability transition by affecting nucleotide binding to the adenine nucleotide translocase. J. Biol. Chem., 272, 3346-3354 (1997).
585
17. Tindall, B. and Cooper, D.A.: Primary HIV infection: host responses and intervention strategies. AIDS, 5, 1-14 (1991). 18. Cossarizza, A., Mnssini, C., Mongiardo, N., Borghi, V., Sablmfini, A., De-Rienzo, B., and Franeesehi, C.: Mitochondria alterations and dramatic tendency to undergo apoptosis in peripheral blood lymphocytes during acute HIV syndrome. AIDS, 11, 19-26 (1997). 19. Chang, K.H., IOta, K.S., and Kim, J.H.: N-Acetylcysteine increases the biosynthesis of recombinant EPO in apoptotic Chinese hamster ovary cells. Free gadic. Res., 30, 85-91 (1999). 20. Hignchi, M., Proske, R.J., and Yeh, E.T.: Inhibition of mitochondrial respiratory chain complex I by TNF results in cytochrome c release, membrane permeability transition, and apoptosis. Oncogene, 17, 2515-2524 (1998). 21. Ramakrishnan, N. and Catravas, G.N.: N-(2-Mercaptoethyl)1,3-propanediamine (WR-1065) protects thymocytes from programmed cell death. J. Immunol., 148, 1817-1821 (1992). 22. Siow, R.C., Richards, J.P., Pedley, K.C., Leake, D.S., and Mann, G. E.: Vitamin C protects human vascular smooth muscle cells against apoptosis induced by moderately oxidized LDL containing high levels of lipid hydroperoxides. Arterioscler. Thromb. Vase. Biol., 19, 2387-2394 (1999). 23. Blankenship, L.J., Carlisle, D.L., Wise, J.P., Orenstein, J. M., Dye, L. E-3rd, and Patiemo, S. R.: Induction of apoptotic cell death by particulate lead chromate: differential effects of vitamins C and E on genotoxicity and survival. Toxicol. Appl. Pharmacol., 146, 270-280 (1997).
Effect of Antioxidants on the Apoptosis of CHO Cells and Production of Tissue Plasminogen Activator in Suspension Culture ZHANYOU YUN, MUTSUMI TAKAGI,* AND TOSHIOMI YOSHIDA International Center for Biotechnology, Osaka University, 2-I Yamada-oka, Suita, Osaka 565-0871, Japan
Received 12 January 2001/Accepted27 March 2001 The effects of antioxidants on the apoptosis of Chinese hamster ovary (CHO) 1-15,$00 cells and production of tissue plasminogen activator (tPA) in suspension culture were investigated. After cell growth to 2 x 10 s cells/ml in Ham's F12 medium containing 1 0 ~ newborn bovine serum (NBS) in a spinner bottle, CHO cells were maintained for 6 d in Ham's F12 medium containing 0 or 0 . 4 ~ NBS and 10 mM antioxidants, namely, L-ascorbic acid 2-phosphate (VCP) or the reduced form of glutathione (GSH]. The viable cell concentrations at day 6 in the serUm-free culttire with GSH and in the low-serum culture with VCP or GSH were 0.57, 1.04 and 1.69 x 10 s cells/ml, respectively, while those in the serum.free and low-serum cultures without the antioxidants were only 2.33 and 1.17 x 103 cells/ml, respectively. The percentages of apoptutic cells in the serum-free and low-serum cultures with VCP (73.2, 4 4 . 6 ~ ) and GSH (76.9, 3 8 . 6 ~ ) measured using a flowcytometer after annexin V-FITC/propidium iodide staining were markedly lower than those in the cultures without antioxidants (96.3, 9 2 . 5 ~ ) . The percentage of cells having a high mitochondriai membrane potential among the viable cells in the cultures with antioxidants determined using a flowcytometer after 5,5',6,6'-tetrachloro-l,l',3,3'tetraethylbenzimidazulyicarbocyanine iodide staining was clearly higher than those in the cultures without the antioxidants. The production of tPA in the sernm.free and low-serum media with VCP (0.282, 2.02 mg//) or GSH (1.01, 1.61 rag//) was markedly higher than that in the cultures without the antioxidants (0.275, 0.689 mg//). Consequently, the suppression of apoptosis through the maintenance of the membrane potential of mitochondria by VCP or GSH resulted in a marked increase in tPA production by CHO cells in the serum-free and low-serum cultures.
[Key words: apoptosis, CHO cells, suspension culture, antioxidant, tissue plasminogen activator] Mammalian cells can produce secondary metabolite proteins such as monoclonal antibodies and tissue plasminogen activator (tPA) even after the cessation of cell growth during batch culture. Moreover, the production rate after cell growth is markedly higher than that during cell growth due to a high specific production rate during the post-growth period (1). However, it is not easy to maintain a high cell activity level during the production phase due to poor nutritional conditions in a serumfree or low-serum medium, which should be employed in order to reduce the contamination of the product with serum proteins in the downstream production process. Therefore, it is important to develop a strategy to increase productivity by inhibiting cell death during the production phase. Cell death during the production phase may proceed via both the necrotic and apoptotic pathways. Apoptosis in a bioreactor was first reported for the cultivation of murine B-cell hybridoma (2). Solis-Recendez et al. reported that the majority of 32 hybridoma cell lines exhibited apoptosis before reaching maximum cell density even when the cells were cultured in a serum-supplementeA medium (3). Measurement of DNA fragmentation by gel chromatography showed that more than 30% of hybridoma cells were apoptotic at the end of batch culture (4). The percentage of apoptotic cells among the dead cells of both plasmacytoma and hybridoma cell lines was estimated to be 90% based on the nuclear morphology (5, 6). Chinese hamster ovary (CHO) cells, which are one of the most widely used animal cell lines for the production
of pharmaceuticals (7), also died mainly through apoptosis during serum-free cultivation (8). It is expected that the suppression of apoptosis during the production phase in serum-free or low-serum media might increase protein production by mammalian cells such as CHO cells. Apoptosis can be induced by the addition of cytokines such as glucocorticoids, thyroid hormones, TNF a, and TGF 31, a deficiency in growth factors, the presence of cytotoxic factors such as perforins, granzymes, and reactive oxygen species, and application of physical factors such as radiation, hyperthermia, and hydrodynamic tension (9). These factors initiate the signal transduction cascades involving the collapse of mitochondrial membrane potential (A~'m), uncoupling of the respiratory chain, over-production of superoxide anions, local oxidation of inner membrane cardiolipins, release of apoptogenic proteins such as cytochrome c, apoptosis-inducing factor (AIF) and caspases from the mitochondrial intermembrane space into the cytosol, potassium eiflux, cytosol calcium increase, chromatin condensation, and DNA fragmentation (10). Suppression of apoptosis during the cultivation process has been investigated by manipulating nutrient feeding (5, 11) and using antiapoptotic genes such as bcl-2 and crm A (12, 13). Several kinds of apoptosis-suppressing chemicals were also investigated for a wide range of applications in cell culture processes. However, the caspaseinhibiting peptides, such as N-benzyloxycarbonyl-ValAla-Asp-fluoromethyl-kctone (Z-VAD.fmk) and acetylAsp-Glu-Val-Asp-aldehyde (14), as well as the mitochondriai membrane-permeability inhibitors, such as bongkrekic acid and cyclosporin A (14--16), are toxic and
* Corresponding author, e-mail: [email protected] phone: +81-(0)6-6879-7456 fax: +81-(0)6-6879-7454 581
582
YUN ET AL.
expensive for practical application to pharmaceutical manufacturing processes. Although antioxidants such as N-acetylcysteine (14, 17-19), pyrrolidine dithiocarbamate (14, 20) and ascorbic acid (21-23) were reported to suppress apoptosis and are less expensive and not toxic, the effects of antioxidants on the productivity of cells have not been reported except in a study by Mastrangelo et al. (14), in which N-acetylcysteine, pyrrolidine dithiocarbamate, bongkrekic acid and Z-VAD.fmk suppressed the apoptosis caused by viral infection and increased the production of chloramphenicol acetyltransferase by virus-infected AT3-neo cells. However, there are few reports on the effect of antioxidants on apoptosis caused by serum depletion and on protein productivity of common cell lines such as CHO cells. Moreover, there has been no discussion about the mechanism of apoptosis suppression by antioxidants. In this study, the effects of two antioxidants, namely, L-ascorbic acid 2-phosphate (VCP) and the reduced form of glutathione (GSH), on apoptosis and production of tPA by CHO 1-15,500cells during suspension culture in serum-free and low-serum media were investigated. MATERIALS AND METHODS
Cells and media
CHO 1-15,500cells producing tPA (ATCC CRL-9606) (1) were used. Ham's F-12 medium (Gibco, Tokyo) supplemented with 10% NBS (Gibco), streptomycin (0.1 mg/ml), penicillin (100u/ml) and 500 nM of methotrexate (Sigma, MO, USA) was employed for the cell growth phase. The serum concentration in the medium was reduced to 0 or 0.4% and 10mM each of VCP (Sigma) and GSH (Sigma) were added during the production phase. We selected 10 mM as the concentration for antioxidant addition to the spinner bottle culture after preliminary experiments with the still culture of CHO cells in a tissue-culture multiwell plate employing F12 medium supplemented with 0.4% NBS and antioxidants (0-10 mM). Cell cultivation Cells (4.1 × 106 cells) harvested from a dish (55cm 2, Corning Inc., NY, USA) by trypsinization were inoculated into a spinner bottle (10Oml, Shibata Hario Co., Tokyo) containing 30 mi of the cellgrowth medium, and then incubated in a water bath (37°C, 5% CO2, 70rpm). The medium was changed once during the cell-growth phase. The cells at the stationary phase were washed with 40 ml of PBS (pH 7.4), suspended in the production medium (2x 105cells/ml) and incubated further in a spinner bottle for 6 d. Cell cultivation was repeated three times and similar effects of antioxidants were obtained in all three cultivations. Typical data from these cultivations are shown in Results. General analysis Cell concentration and viability were determined by the trypan-blue dye exclusion method. The concentration of tPA was assayed by ELISA (Imulyse tPA, Biopool AB, Sweden). Counting apoptotic and necrotic cells Cells were stained according to the protocol for the Annexin V(AV)-FITC kit (IM2376, Immunotech, Marseille, France). Namely, 1 × 105 cells were washed once with PBS (pH 7.4) and double stained with AV-FITC (5 pM) and propidium iodide (PI, 5 pM) on ice in the dark for 10min. The A V / P I stained cells were kept on ice until measurement using a flowcytometer (EPICS XL, Beck-
J. BIOSCI.BIOENG., 10 e
l
O
S
~
lo ~ 0
2
4
6
Culturetime(d) FIG. 1. Effects of the antioxidants on viable cell concentration during the serum-free and low-serum cultivation of CHO cells. The serum-free (open symbols)and low-serum(closedsymbols)cultivation was performed without (©, e) or with 10 mM antioxidants of VCP (n, • ) or GSH (v, v), respectively. Viable cell concentration was determined by the trypan-blue method. man-Coulter, Miami, USA). The excitation wavelength was 488 nm and the emission intensity was measured at 525 and 675 nm. The settings of the flowcytometer were: FS, 35mV; SS, 242mV; FL1, 499mV; FL4, 588mV; compensation, FL1- FL4 0.3%; and FL4-FL1 0.2%.
Measurement of mitoehondrial membranepotential Cells (1 × 105) were washed once with PBS (pH7.4) and stained with 10pM 5,5',6,6'-tetrachloro-l,l',3,3'-tetra ethylbenzimidazolylcarbocyanine iodide (Jc-1, Molecular Probes, Eugene, USA) in 0.Sml of PBS at 37°C for 10rain in the dark. Stained cells were kept on ice until the analysis using the flowcytometer. The excitation wavelength was 488 nm and emission intensity was measured at 525 and 575 nm. The settings of the flowcytometer were: FS, 35mV; SS, 242mV; FL1, 450mV; FL2, 510mV; compensation, FL1- FL2 3%; and FL2-FL1 22%.
RESULTS Effect of antioxidants on cell viability
Because the objective of this research was to investigate the suppression of apoptosis by antioxidants during the production phase, ceils (2x 105 cells/ml, 97% viability) were transferred from the growth medium (10% serum) to a medium with or without 0.4% serum supplemented with antioxidants, in which cells could not grow, and the viable TABLE 1. Percentagesof apoptotic and necrotic cells during the serum-free and low-serum cultivation Serum Apoptotic cells (%) Necroticcells (%) (%) Antioxidant ld 3d 6d ld 3d 6d -14.4 6 6 . 4 92.5 4.9 8.6 6.3 VCP 22.7 4 8 . 2 73.2 21.9 2 7 . 2 23.5 GSH 27.3 57.3 76.9 13.9 11.3 11.9 0.4 -8.8 53.1 96.3 12.0 2.3 1.4 VCP 9.1 17.6 44.6 11.2 3.3 7.1 GSH 29.7 19.8 38.6 8.4 4.8 7.9 The percentage of apoptotic and necrotic cells among the total number of cells during the serum-free and low-serumcultivation with or without the antioxidants (10 raM) VCP and GSH was determined using a fiowcytometerafter AV/PI staining.
VOL. 91, 2001
EFFECT OF ANTIOXIDANTS ON APOPTOSIS OF CHO CELLS I00
o 0
2
4
g
Culture time (d) FIG. 2. Effect of the antioxidants on mitochondrial membrane potential during the serum-free and low-serum cultivation of CHO cells. Mitocliondrial membrane potential of cells during the serumfree and low-serum cultivation shown in Fig. 1 was analyzed using a flowcytom~er after Jc-1 staining. Symbols are the same as those in Fig. 1. cell concentration and cell viability during the production phase were determined thereafter. While the viable cell concentration decreased rapidly to 1/100 o f the initial concentration in 6 d in the cultures without antioxidants (Fig. 1), a marked decrease in viable cell concentration was not observed in the low-serum culture supplemented with V C P or GSH, resulting in a viability higher than 70% at 6 d, indicating that serum effectively protected cells from death. Moreover, it is notable that the viable cell concentration at 6 d in the serum-free culture supplemented with G S H was maintained at 1/4 o f the initial concentration. Apoptosis during cultivation with the antioxidants In order to clarify which type o f cell death, that is apoptosis or necrosis, was suppressed by the antioxidants, cells in the production phase in the serum-free and lowserum media supplemented with the antioxidants were analyzed after A V / P I staining using a flowcytometer. The proportion o f apoptotic cells a m o n g the total number of cells markedly increased in the serum-free and low-serum media without antioxidants and were higher than 90% at 6 d (Table 1). On the other hand, the proportion o f apoptotic cells was markedly lower at 3 and 6 d in the cultures with the antioxidants than in those without the antioxidants, although the proportion of necrotic cells in the cultures with the antioxidants was slightly higher than in those without the antioxidants. Mitoehonflrial membrane potential of cells during cul-
583
tivation with the antioxidants In order to investigate the mechanism o f apoptosis suppression by the antioxidants, the amount o f cells with high mitochondrial membrane potential (A~m) was estimated during the serumfree and low-serum culture. The percentage o f such cells among the total number o f cells decreased rapidly and was less than 5% at 6 d in the serum-free and low-serum cultures without antioxidants (Fig. 2). The percentage in the low-serum culture with VCP and GSH was maintained at a higher level than that without an antioxidant after a temporary decrease at 1 d. On the other hand, the increments in the percentage by addition o f antioxidants to the serum-free culture were small. In order to better understand the effect o f antioxidants on the A~m, we attempted to estimate the percentage o f cells with high A~m among the viable cells. We observed two clusters of cells in the flowcytometric histogram o f forward scatter (FS) versus side scatter (SS) as shown in Fig. 3. As FS is generally proportional to cell size, the clusters with high and low FS were designated large-ceU cluster and small-cell cluster, respectively. The proportion of cells classified into to the small-cell cluster increased as the culture progressed in the serum-free (Fig. 3) and even in the low-serum media. There was a proportionai relationship between the viability and the percentage o f cells in the large-cell cluster (Fig. 4). This relationship strongly suggests that the cells in the largecell and small-cell clusters were viable and dead cells, respectively. The proportion of cells with high A~m in the large-cell cluster was determined (Table 2). Most cells in the large-cell cluster maintained a high A'~"m up to 3 d in the serum-free and low-serum cultures regardless of the presence of the antioxidants. However, it is notable that the percentages decreased markedly at 6 d in the absence o f the antioxidants in both the serum-free and low-serum cultures, while they remained high in the presence o f the antioxidants. Effect of antioxidants on t P A productivity The concentration of t P A produced at 6 d in the serum-free and low-serum cultures was determined (Table 3). The concentrations of t P A in the serum-free cultures containing VCP and G S H were 1.1 and 3.7 times that in the culture without antioxidants, respectively. VCP and G S H
'A
100
•fl
80
~
40
°1o 20
0
SS
1024
0
SS
1024
0
SS
1024
FIG. 3. Forward scatter and side scatter of CHO cells during serum-free cultivation. Forward scatter and side scatter of CHO cells were analyzed using a flowcytometer at 0, 1 and 3 d during serum-free cultivation. The region surrounded by a solid line indicates the cluster of large cells.
40
60
80
100
Viability (%) FIG. 4. Correlation between the percentage of cells in the largecell cluster and viability during the serum-free and low-serum cultures. The percentage of cells in the cluster of large cells was plotted against the viability measured by the trypan-blue method for all conditions shown in Fig. 1. The sample number was 42 and the square of the correlation coefficient (r2) was 0.91.
584
J. BlOSCl. BIOI~NG.,
Y U N E T AL.
TABLE 2. Percentage of cells with high mitochondrial membrane potential (AYm)among viable cells during the serum-free and low-serum cultivation Serum (%)
Antioxidant VCP GSH
0.4
--
VCP GSH
ld 97.3 88.4 84.6 98.7 72.0 93.7
High A~m cells (o~) 3d 92.8 92.3 88.0 89.0 96.5 96.6
6d 12.7 89.9 70.2 34.2 90.7 95.8
The percentage of cells with high A~m among the cellsin the cluster of large cells during the serum-free and low-serum cultivation with or without the antioxidants (10 mM) VCP and GSH was determined using a flowcytometer after Jc-I staining. in the low-serum culture increased t P A production 3.0 and 2.4 times, respectively. DISCUSSION We employed F12 medium without serum addition as a serum-free medium, and the medium containing 0.4°~ NBS as a low-serum medium. The difference between the effects o f antioxidants on cell apoptosis in the serum-free and low-serum media was clearly shown, though the celis were washed only once with F12 medium after cell growth in 10% NBS medium and there may be some remaining serum even in the serum-free medium. VCP and GSH markedly suppressed the decrease in the viable cell concentration and cell viability during the low-serum culture (Fig. 1). GSH was also effective for the serum-free culture, while the effect of VCP was not apparent. VCP at higher concentration may be effective even for a serum-free culture. It was dearly shown that apoptosis was the major means o f cell death during both the serum-free and lowserum cultivation without antioxidants (Table 1). This result agrees with the reports that approximately 90% o f the death o f both plasmacytoma and hybridoma cells were due to apoptosis (5, 6) and that C H O cells died through apoptosis during serum-free culture (8). The antioxidants VCP and G S H markedly decreased the percentage o f apoptotic cells during both the serumfree and low-serum cultures, while they slightly increased the percentage o f necrotic cells (Table 1). Apoptosis o f other cell lines was reported to be suppressed by antioxidants such as N-acetylcysteine (NAC) (14, 17-19), pyrrolidine dithiocarbamate (PDTC) (14, 20) and ascorbic acid (VC) (21-23). It was strongly suggested that the suppression of apoptosis by VCP and GSH during the serum-free and low-serum cultures resulted in higher viability and viable cell concentration than in cultures without antioxidants. Because reduction o f A~m is a c o m m o n phenomenon TABLE 3. Production of tPA in the serum-free and low-serum cultures Serum (%)
-tPA (nag//) 0.275
0
VCP 0.282
0.4
GSH 1.012
-0.689
VCP 2.017
GSH 1.611
Concentration of tPA produced by CHO cells during the serum-free and low-serum culture for 6 d with or without the antioxidants (10 mM) VCP and GSH was determined by ELISA.
during apoptosis, the influence o f the antioxidants on the A~ m of the total cells during the serum-free and low-serum cultures was compared with their effect on cell viability (Fig. 1) and the percentage o f apoptotic cells among the total cells (Table 1). The percentage o f cells with a high A~m among the total cells remained higher in the low-serum cultures with both VCP and G S H (Fig. 2), so the suppression o f apoptosis by the antioxidants was considered to be correlated with the maintenance o f a high A~ m. We further attempted to estimate the effect o f the antioxidants on the A~m of only viable cells (cells stained by trypan blue). The number o f cells with a high A~m among the viable cells clearly increased in the presence o f the antioxidants in both the serum-free and lowserum cultures (Table 2), which suggests that VCP and G S H prevented the decrease in A~ m before cell death by apoptosis. The sequential mechanism o f the suppression o f apoptosis by the antioxidants, including the change in A~m, should be investigated further. The antioxidants also increased the production o f t P A (Table 3). This indicates that this effect was concomitant to the effect on viable cell maintenance (Fig. 1), based on the evidence that VCP addition to the serum-free culture had only a small effect on viable cell concentration and t P A production. The increase in t P A production by the antioxidants may be a result o f the suppression o f apoptosis. In the future, we would like to examine the effect o f antioxidants on apoptosis and t P A production employing commercially available serum-free media. The antioxidants V C P and G S H appear to suppress the apoptosis o f C H O cells during serum-free and lowserum suspension cultures due to the increase in A~m. Moreover, they markedly increase the t P A production. Consequently, nontoxic and cheap antioxidants such as VCP and GSH should have great potential for application to practical production processes o f pharmaceuticals using mammalian cells. REFERENCES 1. Takagi, M., Hayashi, H., and Yoshida, T.: Effects of osmolarity on metabolism and morphology in adhesion and suspension of Chinese hamster ovary cell producing tissue plasminogen activator. Cytotechnology, 32, 171-179 (2000). 2. AI-Rubeal, M., Mills, D., and Emery, A.N.: Electron microscopy of hybridoma cells with special regard to monoclonal antibody production. Cytotechnology, 4, 13-28 (1990). 3. Solis-Recendez, M.G., Pirani, A., D'Habit, B., Stacey, G.N., and Maugras, M.: Hybridoma cell cultures continuously
undergo apoptosis and reveal a novel 100 bp DNA fragment. J. Biotechnol., 38, 117-127 (1995). 4. Franek, F. and Dolnikova, J.: Nucleosomes occuring in protein-free hybridoma cell cultures. Evidence for programmed cell death. FEBS. Lett., 248, 285-287 (1991). 5. Singh, R.P., AI-Rubeal, M., Gregory, C.D., and Emery,
A.N.: Cell death in bioreactors: a role for apoptosis. Biotechnol. Bioeng., 44, 720-726 (1994). 6. Mercllle, S. and Massie, B.: Induction of apoptosis in nutrient limited cultures of hybridoma cells. Biotechnol. Bioeng., 44, 1140-1154 (1994). 7. Fusseuegger, M. and Bailey, J. E.: Molecular regulation of cellcycle progression and apoptosis in mammalian cells: implications for biotechnology. Biotechnol. Prog., 14, 807-833 0998). 8. Moore, A., Donahue, C.J., Hooley, J.S., Stocks, D.L., Bauer, K.D., and Mather, J.P.: Apoptosis in CHO cell batch
cultures: examination by flow cytometry. Cytotechnology, 17, 1-11 (1995). 9. Motyl, T.: Regulation of apoptosis: involvement of bcl-2- feint-
VOL 91, 2001
EFFECT OF ANTIOXIDANTS ON APOPTOSIS OF CHO CELLS
ed protenins. Reprod. Nutr. Dev., 39, 49-59 (1999). 10. Dallaporta, B., Marehetti, P., de-Pablo, M.A., Maisse, C., Due, H.T., Metivter, D., Zamzami, N., Geuskens, M., and Kroemer, G.: Plasma membrane potential in thymocyte apoptosis. J. Immunol., 162, 6534-6542 (1998). 11. Franek, F. and Sramkova, K.: Protection of B lymphocyte hybridoma against starvation-induced apoptosis: survival signal role of some amino acids. Immunol. Lett., 52, 139-144 (1996). 12. Chnng, J. D., Sinskey, A.J., and Stephanopoulos, G.: Growth factor and bcl-2 mediated survival during abortive proliferation of hybridoma cell line. Biotechnol. Bioeng., 57, 164-171 (1998). 13. Reynolds, J. E., Li, J., Craig, R. W., and Eastman, A.: BCL-2 and MCL-1 expression in Chinese hamster ovary cells inhibits intracellular acidification and apoptosis induced by staurosporine. Exp. Cell Res., 225, 430-436 (1996). 14. Mastrangelo, A.J., Zon, S., Hardwick, J.M., and Betenbangh, M. L: Antiapoptosis chemicals prolong productive lifetimes of mammalian cells upon Sindbis virus vector infection. Biotechnol. Bioeng., 65, 298-305 (1999). 15. Zamzami, N., Mnrehetti, P., Castedo, M., Hirsch, T., Snsin, S.A., Masse, B., and Kroemer, G.: Inhibitors of permeability transition interfere with the disruption of the mitochondrial transmembrane potential during apoptosis. FEBS Lett., 384, 53-57 0996). 16. Halestrap, A.P., Woodfleid, K.Y., and Connem, C.P.: Oxidative stress, thiol reagents, and membrane potential modulate the mitochondrial permeability transition by affecting nucleotide binding to the adenine nucleotide translocase. J. Biol. Chem., 272, 3346-3354 (1997).
585
17. Tindall, B. and Cooper, D.A.: Primary HIV infection: host responses and intervention strategies. AIDS, 5, 1-14 (1991). 18. Cossarizza, A., Mnssini, C., Mongiardo, N., Borghi, V., Sablmfini, A., De-Rienzo, B., and Franeesehi, C.: Mitochondria alterations and dramatic tendency to undergo apoptosis in peripheral blood lymphocytes during acute HIV syndrome. AIDS, 11, 19-26 (1997). 19. Chang, K.H., IOta, K.S., and Kim, J.H.: N-Acetylcysteine increases the biosynthesis of recombinant EPO in apoptotic Chinese hamster ovary cells. Free gadic. Res., 30, 85-91 (1999). 20. Hignchi, M., Proske, R.J., and Yeh, E.T.: Inhibition of mitochondrial respiratory chain complex I by TNF results in cytochrome c release, membrane permeability transition, and apoptosis. Oncogene, 17, 2515-2524 (1998). 21. Ramakrishnan, N. and Catravas, G.N.: N-(2-Mercaptoethyl)1,3-propanediamine (WR-1065) protects thymocytes from programmed cell death. J. Immunol., 148, 1817-1821 (1992). 22. Siow, R.C., Richards, J.P., Pedley, K.C., Leake, D.S., and Mann, G. E.: Vitamin C protects human vascular smooth muscle cells against apoptosis induced by moderately oxidized LDL containing high levels of lipid hydroperoxides. Arterioscler. Thromb. Vase. Biol., 19, 2387-2394 (1999). 23. Blankenship, L.J., Carlisle, D.L., Wise, J.P., Orenstein, J. M., Dye, L. E-3rd, and Patiemo, S. R.: Induction of apoptotic cell death by particulate lead chromate: differential effects of vitamins C and E on genotoxicity and survival. Toxicol. Appl. Pharmacol., 146, 270-280 (1997).