Differences in mutant p53 protein stability and functional activity in teniposide-sensitive and -resistant human leukemic CEM cells
We examined p53 protein stability and DNA damage-induced p53-dependent responses in a human leukemic CEM cell line and two teniposide-resistant sublines, CEM/VM-1 and CEM/VM-1-5 (∼40 and 400-fold resistant to teniposide, respectively). Although all cell lines contain the same p53 mutations at codons 175 (Arg→His) and 248 (Arg→Gln), the constitutive levels of p53 were progressively increased with the resistance of the cells to teniposide. By pulse-chase experiments, we found that the half-lives of mutant p53 protein were approximately 12, 17, and >30 h in CEM, CEM/VM-1, and CEM/VM-1-5 cells, respectively. The prolonged half-lives of p53 in these cells is consistent with the fact that the protein harbors the indicated mutations. Of note, however, is the fact that the increased p53 protein half-lives in the two drug-resistant cell lines corresponds to a proportional decrease in MDM2 protein levels but an increase in p53-MDM2 binding interactions. This suggests that MDM2-mediated p53 degradation may be altered in our leukemic cell lines. The DNA damage-induced p53 response is fully functional in the drug-sensitive CEM cells containing a mutant p53, but this pathway is attenuated in the drug-resistant cells. Specifically, while the mutant p53 was phosphorylated at serine-15 in response to ionizing radiation in all these cell lines, mutant p53 induction in response to teniposide or ionizing radiation and induction of the p53-target genes, p21 and GADD45 only occurred in the drug-sensitive CEM cells. As assessed by MTT cytotoxicity assay, CEM cells were also significantly more sensitive to ionizing radiation, compared to the drug-resistant cell lines, and this correlated with p53 induction. Collectively, these results suggest that changes in constitutive mutant p53 protein levels, p53-MDM2 binding interactions, and altered regulation of the DNA damage-inducible p53-dependent pathway may play a role in drug- and radiation-responsiveness in these cells.
Alterations of the p53 gene, including point mutations and allelic loss, have been linked to a wide variety of human cancers (Hollstein et al., 1991), suggesting that the loss of wild type (wt) p53 function is associated with oncogenic transformation (Eliyahu et al., 1989; Finlay et al., 1989). p53 is a nuclear phosphoprotein that, when induced, can mediate either an arrest in the cell cycle prior to or near the restriction point in G1 or programmed cell death following exposure to certain DNA-damaging agents (Ullrich et al., 1992; Donehower and Bradley, 1993), thus maintaining genomic integrity and aiding in the suppression of malignant transformation (Hartwell and Kastan, 1994). The induction of wt p53 protein levels in response to DNA damage is regulated by post-transcriptional mechanisms and correlates with a subsequent increase in p53 activity (Maltzman and Czyzyk, 1984; Kastan et al., 1991; Tishler et al., 1993). p53 protein activity is believed to be regulated in part by phosphorylation. In response to both ionizing radiation and ultraviolet irradiation, p53 is phosphorylated at highly conserved serine residues at the amino-terminus (Giaccia and Kastan, 1998; Prives, 1998). This post-translational modification of p53 in response to DNA damaging agents correlates with both increased levels of p53 as well as p53-mediated transactivation of several genes, including the growth arrest-related gene products p21, GADD45, and MDM2, and the cell death-promoting gene product, bax (Barak et al., 1993; El-Deiry et al., 1993; Harper et al., 1993; Zambetti and Levine, 1993; Smith et al., 1994; Miyashita and Reed, 1995). Unlike p21 or GADD45, DNA damage-induced p53 transactivation of MDM2 blocks p53-mediated transcriptional activity (Momand et al., 1992; Zauberman et al., 1993). Inhibition of p53 activity is mediated by binding of MDM2 to the amino-terminus of p53 that then targets p53 for rapid degradation (Haupt et al., 1997; Kubbutat et al., 1997). However, DNA damage-induced phosphorylation of p53 can block binding of the p53 inhibitor, MDM2, thereby preventing MDM2-mediated degradation of p53 (Prives, 1998).
Several studies support the view that loss of wt p53 function can contribute to enhanced tumorigenesis either by loss of p53-mediated checkpoint control, resulting in enhanced genetic instability, or by failure to induce cell death in inappropriate physiological situations (for review see Morgan and Kastan, 1997). However, in contrast to wt p53, there is little understanding of the role mutant (m) p53 plays in tumor cell responsiveness to therapies that include DNA damaging agents. In the present study, we attempted to assess the role of mp53 in the p53-dependent DNA damage response pathway of drug-sensitive and teniposide (VM-26)-resistant human leukemic CEM cell lines. VM-26 resistance in these cells is associated with both decreased activity of DNA topoisomerase IIα (Danks et al., 1988), and with increased expression of MRP (Wang and Beck, 1998). Topoisomerase IIα in these drug-resistant cells harbors point mutations in the cDNA (Bugg et al., 1991; Danks et al., 1993). CEM cells also contain two missense p53 mutations at codons 175/248 located on separate alleles (Cheng and Haas, 1990). This p53 mutant has the ability to transactivate a reporter construct containing a p53 high affinity DNA binding site (Park et al., 1994), suggesting that some mp53 gene products may have different biological properties.
In the work reported herein, we found that (1) increased p53 protein half-lives in the drug-resistant cell lines corresponds to a proportional decrease in MDM2 protein levels but an increase in p53-MDM2 binding interactions, suggesting that MDM2-mediated p53 degradation may be altered in our drug-resistant leukemic cell lines; and (2) the DNA damage-induced p53-dependent responses were progressively attenuated in the drug-resistant cells compared to the drug-sensitive cells, suggesting that only the drug-sensitive CEM cells express a functionally active form of a mutant p53. Our data suggest that changes in mp53 protein stability, mp53-MDM2 binding interactions, and altered regulation of the DNA damage-induced p53-dependent signal transduction pathway may all play a role in drug and radiation responsiveness in these cell lines.
Levels of expression of mp53 protein in VM-26-sensitive and -resistant CEM cells
It has been established that the parental leukemic CEM cell line contains heterozygous mutations located in exon 5 (codon 175, Arg→His) and exon 7 (codon 248, Arg→Gln), with each mutation residing on a different allele, resulting in the cells having no wild-type p53 protein (Cheng and Haas, 1990). In order to compare the basal expression levels of mp53 protein between VM-26-sensitive (CEM) and -resistant (CEM/VM-1 and CEM/VM-1-5) cells, p53 was immunoprecipitated from 35S-labeled cell extracts using three different anti-p53 antibodies (PAbs 421, 1620, and 240). As shown in Figure 1, the basal levels of the p53 protein recognized by monoclonal antibody PAb 421 are increased in the drug-resistant cells compared to the drug-sensitive parental cell line, CEM. Furthermore, this increase in p53 protein appears to coincide with the resistance of these cells to VM-26 (CEM/VM-1 is approximately 40-fold resistant and CEM/VM-1-5 is approximately 400-fold resistant to VM-26; Danks et al., 1988). The monoclonal antibody PAb 1620, which only recognizes the conformational epitope of wt p53 in immunoprecipitates, also detected increased p53 protein levels in the drug-resistant cells (Figure 1). In contrast, the monoclonal antibody PAb 240, which only has selective reactivity for mp53 in immunoprecipitates, precipitated low levels of the p53 protein; these were seen to be unchanged among the drug-sensitive and-resistant cell lines (Figure 1).
p53 protein expression in CEM, CEM/VM-1 and CEM/VM-1-5 cells. p53 was immunoprecipitated from 35S-labeled CEM, CEM/VM-1, and CEM/VM-1-5 cellular extracts using p53 antibodies that react with wild-type and mutant p53 (PAb 421), with a conformational epitope of wild-type p53 (PAb 1620), or with mutant p53 (PAb 240). Proteins were resolved by 10% SDS–PAGE, and the gels were dried and then directly autoradiographed
Full size image
Results from Figure 1 support the idea that a compound heterozygous p53 mutation at codons 175 and 248 (Cheng and Haas, 1990) can exhibit a wild-type conformational change, known as pseudo wild-type p53. This is supported by the observation that the mp53 protein was preferentially precipitated by an antibody specific for wt p53 conformation (PAb 1620) rather than by an antibody specific for mutant conformation (PAb 240). Therefore, increased levels of conformationally wt 53 protein may contribute to the enhanced levels of mp53 in the resistant cell lines.
Nucleotide sequencing of p53 in VM-26-sensitive and -resistant CEM cells
To determine whether there were any additional p53 mutations in the VM-26-resistant cell lines, exons 2-11 of the p53 cDNA in CEM, CEM/VM-1, and CEM/VM-1-5 cells were amplified by PCR and examined by direct sequence analysis. Nucleotide sequencing of the entire coding region of the p53 gene in CEM, CEM/VM-1, and CEM/VM-1-5 cells confirmed the same heterozygous mutations at codons 175 (exon 5) and 248 (exon 7) (data not shown) reported by others (Cheng and Haas, 1990). No other mutations in the coding region of the p53 gene were observed in the drug-sensitive or -resistant CEM cells (data not shown), indicating that the overexpression of p53 in drug-resistant cells, compared to the drug-sensitive parental cell line, is not due to additional mutations in the p53 gene.
p53 half-lives in VM-26-sensitive and -resistant CEM cells
To determine whether the increases in mp53 protein levels in our VM-26-resistant cells compared to the-sensitive cells is due to increased stability, the p53 protein half-life was determined by pulse-chase experiments. As shown in Figure 2, the long half-life (∼12 h) of p53 in the drug-sensitive CEM cells reflects its mutant status. Moreover, the mp53 half-lives in CEM/VM-1 and CEM/VM-1-5 cells were progressively increased compared to CEM, in association with the resistance of the cells to VM-26. These results demonstrate that the progressively longer half-lives of the mp53 protein in the VM-26-resistant cells are due to impairment of their degradation.
The half-lives of the p53 protein in CEM, CEM/VM-1, and CEM/VM-1-5 cells were examined by pulse-chase experiments as described in ‘Materials and methods’. Cellular protein was extracted from 35S-labeled cells at the times indicated and p53 was immunoprecipated using the anti-p53 antibody, PAb 421. Proteins were resolved by 10% SDS–PAGE, and the gels were dried and then directly autoradiographed. This figure is representative of at least three independent experiments. t1/2, half-life
Full size image
Expression of MDM2 protein in VM-26-sensitive and -resistant cells treated with DNA damaging agents
Since MDM2 can negatively regulate p53 levels by binding to the amino-terminus of p53, thereby targeting p53 for rapid degradation (Haupt et al., 1997; Kubbutat et al., 1997), we asked whether the levels of MDM2 were altered in the drug-resistant cell lines, CEM/VM-1 and CEM/VM-1-5, compared to the parental CEM; if so, this may help explain the altered p53 half-lives in these cells. As revealed by Western blot analysis, the constitutive levels of MDM2 protein were progressively decreased with the resistance of the cells to VM-26 (Figure 3a,b). In contrast, the constitutive levels of p53 were progressively increased with the resistance of the cells to VM-26 (Figure 3a). Interestingly, the levels of MDM2 decreased with time in response to 2 Gy irradiation (Figure 3a), or in response to equitoxic doses of VM-26 (Figure 3b) in both the drug-sensitive and -resistant cells. As controls, it is seen in Figure 3c that MDM2 protein can be induced by 6 h in response to 2 Gy irradiation in an ovarian tumor cell line containing wild-type p53 (PA-1) but not in a cell line containing mutant p53 (SW626). This p53-dependent induction of MDM2 in response to DNA damage has been reported previously (Barak et al., 1993). Thus, in our CEM cell lines, the apparent inverse correlation between p53 and MDM2 protein levels suggests that MDM2 may play an important role in mp53 stability in our cell lines.
Levels of MDM2 protein expression in VM-26-sensitive and -resistant cells treated with DNA damaging agents. (a) Immunoblot analysis of p53 and MDM2 was done on CEM, CEM/VM-1, and CEM/VM-1-5 cells untreated or treated with 2 Gy ionizing radiation (IR). Cells were harvested at the indicated times after treatment and lysates were resolved by 10% SDS–PAGE. After electrophoretic transfer, the nitrocellulose was cut and immunoblotted with either anti-p53 (Ab-6, top panel), or anti-MDM2 (Ab-1, bottom panel) as described in ‘Materials and methods’. (b) Immunoblot analysis of MDM2 was done on drug-sensitive and -resistant cells untreated or treated with the indicated equitoxic doses of VM-26. Cells were harvested 6 h after treatment and lysates were resolved and immunoblotted as described above. (c) Immunoblot analysis of MDM2 was performed on the human ovarian cell lines PA-1 and SW626 untreated or treated with 2 Gy ionizing radiation (IR) and cells were harvested at the indicated times and immunoblotted using anti-MDM2 (Ab-1) antibody as described in `Materials and methods'. These Western blots are representative of at least three separate experiments
Full size image
p53-MDM2 binding interactions in VM-26-sensitive and -resistant CEM cells
The possibility exists that enhanced p53 stabilization in the drug-resistant cells may be due in part to modifications of the p53 protein that prevent p53-MDM2 interactions, thus resulting in abrogation of MDM2-mediated p53 degradation. To address this possibility, a co-immunoprecipitation experiment was performed to analyse p53-MDM2 binding interactions in CEM, CEM/VM-1, and CEM/VM-1-5 cells. Immunoprecipitation and immunoblotting of p53 revealed that p53 protein levels were progressively increased in association with the resistance of the cells to VM-26, as expected (Figure 4, bottom panel). Of importance, immunoprecipitation of p53 and immunoblotting for MDM2 revealed that p53-MDM2 interactions occur in both drug-sensitive and -resistant cells (Figure 4, top panel). Interestingly, the amount of MDM2 co-immunoprecipitated appeared to be increased in association with the resistance of the cells to VM-26. As a negative control, an immunoprecipitation reaction was also performed using a mouse immunoglobulin of the same isotype as the anti-p53 antibody (Figure 4). Thus, the observed increase in p53-MDM2 binding interaction in the drug-resistant cells suggests that enhanced p53 stabilization in the drug-resistant cells is not due to an abrogation of p53-MDM2 binding. Furthermore, the increase in p53 protein levels and in p53-MDM2 binding in the drug-resistant cells, suggests that the pathway leading to MDM2-mediated p53 degradation may be altered.
p53-MDM2 binding interactions in VM-26-sensitive and -resistant CEM cells. Immunoprecipitation of p53 was done on CEM, CEM/VM-1 and CEM/VM-1-5 cells as described in ‘Materials and methods’. Immunoprecipitated p53 was electrophoresed on a 10% SDS polyacrylamide gel. Protein was electrophoretically transferred and the membrane was cut in half and immunoblotted with either anti-p53 (Ab-7, bottom panel), or anti-MDM2 (Ab-1, top panel) as described in ‘Materials and methods’. These Western blots are representative of at least three separate experiments
Full size image
Effect of VM-26 on mp53 protein induction in VM26-sensitive and -resistant CEM cells
To determine whether mp53 in the drug-sensitive and-resistant cells is functional, we performed p53 immunoblot experiments to ask whether mp53 can be induced in response to VM-26 in CEM, CEM/VM-1 and CEM/VM-1-5 cells. As shown in Figure 5, the level of mp53 protein in CEM cells treated with 0.2 μM VM-26 (a 48 h IC50 value; Chen and Beck, 1993) or 1 μM VM-26 increased in a time-dependent manner up to 24 h after treatment. However, in the drug-resistant cells, treatment with equitoxic doses of VM-26 (8 μM for CEM/VM-1 and 100 μM for CEM/VM-1-5; Kim and Beck, 1994) produced no alterations in the level of mp53 protein (Figure 5b). The lack of responsiveness of mp53 in CEM/VM-1-5 cells might be due to the high (possibly maximum) constitutive overexpression of mp53, therefore making the results difficult to interpret. Densitometric scanning of the p53 protein levels clearly revealed the differences in responsiveness of the mp53 protein after VM-26 treatment in drug-sensitive and -resistant cells (Figure 5c). These findings suggest that the accumulation of mp53 protein in response to VM-26 in the drug-sensitive CEM cells might play a role in their response to DNA damage, since induction of mp53 protein by VM-26 appears to be attenuated or absent in the drug-resistant cells.
Levels of p53 protein expression after VM-26 treatment of CEM, CEM/VM-1 and CEM/VM-1-5 cells. (a, b) Immunoblot analysis of p53 was done on drug-sensitive and-resistant cells that were harvested at the indicated times after treatment with various doses of VM-26. Whole cell extracts were electrophoresed by 10% SDS–PAGE and p53 was detected using PAb 421. (c) Quantitation of p53 protein levels after VM-26 treatment by densitometric scanning. ^, CEM; •, CEM/VM-1; □, CEM/VM-1-5
Full size image
Ionizing radiation induces phosphorylation of p53 in both drug-sensitive and -resistant cells
To better understand whether the DNA damage-inducible p53-dependent pathway is functional in the drug-sensitive compared to the drug-resistant cell lines, p53 phosphorylation in response to ionizing radiation was examined in CEM, CEM/VM-1, and CEM/VM-1-5 cells. p53 is phosphorylated at amino-terminal serine residues in response to DNA damage (Giaccia and Kastan, 1998; Prives, 1998). This de novo phosphorylation is believed to be involved in the subsequent induction and activation of p53 (Siliciano et al., 1997). By Western blot analysis we examined the levels of total p53 protein and p53 protein phosphorylated on serine-15 derived from untreated, irradiated, or ALLN-treated drug-sensitive and -resistant cells using antibodies specific for p53 (Ab-6) and phosphoserine-15 p53. Figure 6 shows a representative immunoblot of mp53 (top panel) and serine-15 phosphorylated mp53 (bottom panel) in drug-sensitive and -resistant CEM cells. In response to 2 Gy irradiation, the levels of mp53 protein increased in CEM cells, whereas these levels did not appear to change in the drug-resistant cells, although their constitutive levels were higher than those of the CEM cells (Figure 6, top panel). Interestingly, ionizing radiation-induced phosphorylation of p53 on serine-15 was observed for both drug-sensitive and -resistant cells (Figure 6, bottom panel). In contrast very little or no p53 serine-15 phosphorylation was observed in either unirradiated (control) or in cell lines treated with the proteosome inhibitor, ALLN, despite the fact that there were equivalent amounts of total p53 protein in irradiated cells compared to ALLN-treated cells (Figure 6). These results suggest that the IR-induced signal transduction pathway involving p53 phosphorylation remains intact in both the VM-26-sensitive and -resistant cell lines alike.
Effect of ionizing radiation on p53 phosphorylation in VM-26-sensitive and-resistant cells. Western blot analysis of p53 was done on CEM, CEM/VM-1 and CEM/VM-1-5 cells that were either untreated (control) or treated with 2 Gy irradiation or 20 μM ALLN. Cells were harvested at 24 h after treatment and lysates were resolved by 10% SDS–PAGE. After electrophoretic transfers, duplicate membranes were immunoblotted with either anti-p53 (Ab-6, top panel), or anti-phosphoserine-15 p53 (bottom panel) as described in `Materials and methods'. These Western blots are representative of at least three separate experiments
Full size image
Effect of DNA damaging agents on p21 and GADD45 protein induction and cell death in VM-26-sensitive and -resistant cells
Phosphorylation and induction of p53 in response to DNA damaging agents result in p53-mediated transactivation of several genes including p21, GADD45, and MDM2 (Kastan et al., 1992; El-Deiry et al., 1993). To determine whether these p53-target genes can be induced in response to DNA damage in our drug-sensitive and -resistant cells containing a mutant p53, levels of p21 and GADD45 were examined by Western blot in VM-26-treated or ionizing radiation-treated CEM, CEM/VM-1, and CEM/VM-1-5 cells. Figure 7a reveals that at 6 h after treatment of the drug-sensitive CEM cells with VM-26, the level of p21 protein increased in a dose-dependent manner (up to fourfold increase). In contrast, p21 protein did not appear to be significantly induced in either of the drug-resistant cell lines in response to equitoxic doses of VM-26 (Figure 7a). Moreover, as with p53, p21 appeared as a doublet and was constitutively up-regulated in the drug-resistant cells. Similarily, GADD45 was induced in the drug-sensitive CEM cells within 6 h in response to DNA damage, whereas in the drug-resistant cells there was little (CEM/VM-1) or no (CEM/VM-1-5) detectable GADD45 induction (Figure 7b). Equivalent amounts of total cellular protein were loaded for each blot as determined by probing for β-actin (Figure 7a,b).
Levels of p21 and GADD45 protein expression after treatment with DNA damaging agents in VM-26-sensitive and-resistant cells. (a) Immunoblot analysis of p21 was done on CEM cells treated with either 0.2 or 1.0 μM VM-26 and on CEM/VM-1 and CEM/VM-1-5 cells treated with 8.0 and 100 μM, respectively. (b) Immunoblot analysis of GADD45 was performed on drug-sensitive and -resistant cells treated with 2 Gy ionizing radiation (IR). All cells were harvested at 6 h following treatment and lysates were resolved by 12% SDS–PAGE. p21 was detected using an anti-p21 antibody (Ab-5, panel a) and GADD45 was detected using an anti-GADD45 antibody (panel b), as described in ‘Materials and methods’. Equal loading of total cellular protein was determined by blotting for β-actin. These Western blots are each representative of at least three separate experiments
Full size image
In response to DNA damage, p53 can mediate either a cell cycle arrest or programmed cell death (Ullrich et al., 1992; Donehower and Bradley, 1993). Cell death in response to ionizing radiation was therefore examined in VM-26-sensitive and -resistant cell lines as assessed by a MTT cytotoxicity assay. The CEM cells were statistically significantly more radiosensitive to 4 and 6 Gy of ionizing radiation than the CEM/VM-1 or CEM/VM-1-5 cell lines as determined by the Student's t-test (Figure 8). The CEM/VM-1 cells exhibited an intermediate sensitivity to irradiation whereas the CEM/VM-1-5 cells were the most radioresistant. Enhanced radiation-induced cell death corresponded to an ionizing radiation-induced increase in p53 protein levels only in the parental CEM cells (data not shown). These changes in sensitivity between the drug-sensitive and resistant cells do not appear to be due to altered cell cycle regulation, as treatment of these cells with 4 Gy of ionizing radiation induced a G2/M cell cycle checkpoint by 24–48 h in all cell lines examined (data not shown).
Radiation sensitivity in VM-26-sensitive and -resistant cells. CEM, CEM/VM-1, and CEM/VM-1-5 cells were exposed to 0, 2, 4, 6 and 10 Gy of ionizing radiation and incubated at 37°C in a humidified chamber containing 5% CO2, 95% air. Cell survival was determined 96 h later using the MTT cytotoxicity assay as described in ‘Materials and methods’. s.e. error bars
Full size image
Comparison of results in Figures 6,7,8 indicates that while mp53 was phosphorylated at serine-15 in response to DNA damage in the drug-sensitive and -resistant cells alike, induction of mp53 as well as the p53-target genes, p21 and GADD45, and enhanced cell death in response to ionizing radiation or VM-26 only occurred in the drug-sensitive CEM cells. Therefore, the DNA damage-inducible p53-dependent pathway that leads to cell death appears to be functional in drug-sensitive CEM cells containing mutant p53, but this pathway, post-DNA damage-induced p53 phosphorylation, is attenuated or missing in the drug-resistant cells.
Effect of DNA damaging agents on bax and bcl-2 protein expression in VM-26-sensitive and -resistant cells
To determine whether resistance to DNA damage-induced cell death in the drug-resistant cells could be associated in part with differential expression of the p53-regulated apoptosis-related genes, bax and bcl-2, we measured levels of these proteins in VM-26-treated CEM, CEM/VM-1, and CEM/VM-1-5 cells. Bcl-2 protects cells from drug-induced cell death (Reed, 1994; Boise et al., 1995), while bax, a p53 target gene, promotes apoptosis (Oltvai et al., 1993; Miyashita et al., 1994; Selvakumaran et al., 1994; Sedlak et al., 1995). Altered levels of these apoptosis-related proteins are believed to be a determinant of cellular resistance to various death-inducing cytotoxic drugs (Craig, 1995). Figure 9a,b shows a representative immunoblot of bcl-2 and bax in our drug-sensitive and -resistant cells. In response to 10 μM VM-26, the levels of bcl-2 (26 kDa) and bax (21 kDa) proteins did not appear to change in both drug-sensitive and -resistant cells (Figure 9a,b). One exception, however, was that by 24–48 h post treatment, bax levels were reduced in the CEM cells, which is a result of extensive cell death. At 8 h-post treatment, CEM cells also exhibited a bcl-2 cleavage product of ∼23 kDa that increased in intensity with longer treatment, while in the resistant cells, the appearance of this cleaved product in response to VM-26 was delayed or not seen (Figure 9a). It is known that bcl-2 is cleaved in its loop domain by caspase-3 in response to apoptotic stimuli, creating a 23 kDa cleavage product that has pro-apoptotic activity (Cheng et al., 1997).
Effect of DNA damaging agents on bax and bcl-2 protein expression in VM-26-sensitive and -resistant cells. (a) Immunoblot analysis of bcl-2 was done on CEM, CEM/VM-1, and CEM/VM-1-5 cells treated with 10 μM VM-26. (b) Immunoblot analysis of bax was done on CEM, CEM/VM-1, and CEM/VM-1-5 cells treated with 10 μM VM-26. All cells were harvested at 1, 3, 8, and 24 h following treatment and lysates were separated by 12% SDS–PAGE, as described in ‘Materials and methods’. These Western blots are each representative of at least three separate experiments
Full size image
Interestingly, while we detected no substantial changes in bcl-2 or bax protein expression in response to drug, we did observe that the constitutive levels of these proteins varied between the drug-sensitive and -resistant cells. On average, bcl-2 protein levels were increased up to twofold in the drug-resistant cells and bax protein levels were reduced by approximately 50% in the drug-resistant compared to -sensitive cells (Table 1). The high ratio of bcl-2 to bax in the VM-26-resistant cells is consistent with the idea that increased levels of anti-apoptotic genes and decreased levels of pro-apoptotic genes can play a role in drug resistance (Craig, 1995). That there is a higher ratio in CEM/VM-1 compared to CEM/VM-1-5 cells suggests that other factors are likely to play a role in the resistance of these cells to anticancer drugs and ionizing radiation.
Table 1: Relative protein levels and ratios of bcl-2 and bax inVM-26-sensitive and -resistant cell linesa
Full size table
We examined p53 protein stability and DNA damage-induced p53-dependent responses in a human leukemic CEM cell line and in the VM-26-resistant sublines, CEM/VM-1 and CEM/VM-1-5. From this study we arrived at two novel observations. First, although both the drug-sensitive and -resistant cells contain the same p53 mutations, we observed an increase in p53 protein half-life that corresponded to a decrease in MDM2 protein levels and an increase in p53-MDM2 binding interactions only in the drug-resistant cells, suggesting that MDM2-mediated p53 degradation may be altered in these leukemic cell lines. Second, because treatment of the drug-sensitive CEM cells with DNA damaging agents resulted in the phosphorylation of mp53 at serine-15, mp53 protein induction, induction of the p53 target genes, p21 and GADD45, and enhanced radiosensitivity, this suggests that only the drug-sensitive CEM cells express a functionally active form of a mutant p53, compared to the drug-resistant cells. Collectively, these observations suggest that changes in mp53 protein stability, p53-MDM2 binding interactions, and alterations in a functionally active mp53 response pathway may play a role in the drug and radiation responsiveness of these cells. Furthermore, since CEM/VM-1 and CEM/VM-1-5 cells also exhibit radioresistance and have increased bcl-2/bax ratios compared to CEM cells, this suggests not only that the overall DNA damage response pathway leading to cell death is impaired in these resistant cells, but also that selection for resistance to VM-26 can lead to partial cross-resistance to ionizing radiation.
Mutant p53 protein appears to play an active role in the DNA damage response in the CEM cells. This is not surprising since several studies have reported that mp53 can exhibit functional properties similar to those of wt p53. For example, doxorubicin treatment led to an increase of mp53 protein in association with apoptotic cell death in human squamous carcinoma cells expressing only mp53 (Kwok et al., 1994). Moreover, transfection of a mp53 gene into cells expressing wt p53 in cisplatin-resistant ovarian carcinoma cells produced a significant increase in sensitivity to cisplatin (Brown et al., 1993). Further, it has been reported that different p53 mutants exhibit both DNA binding and transcriptional activation functions (Zhang et al., 1993; Park et al., 1994). Surprisingly, the p53 mutant expressed in the sensitive CEM cell line was found to bind to DNA as well as transactivate a reporter construct (Park et al., 1994). This novel observation was explained by the possibility that oligomerization of these two different mp53 proteins (codon 175 mutant p53 and codon 248 mutant p53) may result in trans-complementation such that these p53 oligomers may form normal functioning proteins (Park et al., 1994). In addition, one recent study has identified small synthetic compounds that stabilize the active conformation of the DNA binding domain of mp53, enabling mp53 to activate transcription (Foster et al., 1999). The possibility exists that in our drug-sensitive CEM cells mp53 may also be ‘stabilized’ in such a way that allows functional activity.
While mp53 plays a functionally active role in DNA damage-induced cell death in the drug-sensitive cells, this response appears to be down-regulated and/or attenuated in the VM-26-resistant cell lines. Thus, one contributing factor to the drug resistance of these cells (i.e. attenuation of the p53-dependent pathway leading to cell death) may involve the striking changes observed in the half-lives of the mp53 protein in the drug-sensitive and -resistant cell lines. Pulse-chase experiments revealed that the half-lives of mp53 protein were ∼12, 17 and >30 h in CEM, CEM/VM-1, and CEM/VM-1-5 cells, respectively. These data suggest that the prolonged half-life of mp53 protein may play a critical role in the sensitivity of tumor cells to DNA damaging agents. Although the molecular mechanism associated with increased p53 protein and its extended half-life in our drug-resistant cell lines compared to the drug-sensitive cells is unclear, the impaired degradation of mp53 protein in drug resistant cells may account in part for its prolonged half-life. p53 protein degradation is believed to be controlled through a tightly regulated ubiquitination pathway (Maki et al., 1996). Experiments are in progress in our laboratory to determine whether this pathway is altered in these cell lines. The extended half-life and loss of p53 function in the drug-resistant cells compared to the drug-sensitive cells is not due to nuclear exclusion, as equivalent p53 protein levels were apparent by Western blot in both nuclear and cytoplasmic fractions derived from drug-resistant cells (Morgan and Beck, unpublished observation).
Our studies do reveal, however, that constitutive levels of MDM2, a negative regulator of p53 (Barak et al., 1993), are lower in the VM-26-resistant cells, compared to the drug-sensitive cells. The MDM2 protein functions as a ubiquitin ligase for p53, thereby targeting p53 for degradation (Honda et al., 1997). Interestingly, our data reveal that there is both an increase in p53 protein and in p53-MDM2 binding interactions in the VM-26-resistant cells, suggesting that p53-targeted degradation mediated by MDM2 may be altered in such a way that it stabilizes p53 but does not target p53 for degradation. One study has in fact revealed dominant-negative forms of MDM2 that stabilize p53 but lack the degradative function (Kubbutat et al., 1999). Furthermore, stabilization of p53 by these mutant MDM2 proteins also abrogated p53 functional activity (Kubbutat et al., 1999). Whether such alterations in MDM2 also exist in our drug-resistant cells expressing constitutively higher levels of non-functional mp53 compared to the drug-sensitive cells remains to be determined. We are presently examining MDM2 in our cells for its regulation, for mutations, and to further define the biochemical interactions between MDM2 and mp53 in the VM-26-sensitive and -resistant cells.
In addition to the decreased constitutive levels of MDM2 in our drug-resistant cells, our studies also revealed that MDM2 protein levels were further decreased in response to DNA damage in both the drug-sensitive and -resistant cell lines. This result was unexpected since MDM2 expression is normally activated in response to DNA damage in a p53-dependent manner (Barak et al., 1993). However, another study demonstrated that treatment of a tumor cell line with the topoisomerase II inhibitor, etoposide, led to down-regulation of MDM2 at both the RNA and protein levels, whereas the other p53-responsive genes, GADD45 and p21, were induced in a dose-dependent manner (Arriola et al., 1999). Hypoxia treatment of a human colorectal carcinoma cell line also resulted in the down-regulation of MDM2 protein levels (Alarcon et al., 1999). These findings, with ours, suggest that inhibition of MDM2 expression may be regulated through a p53-independent mechanism that may also be cell-type specific.
It remains to be determined how mp53 may be involved in tumor cell susceptibility to cell death by anticancer agents and, furthermore, how changes in mp53 protein stability in the VM-26-resistant cells play a role in anticancer drug responsiveness. One contributing factor to the resistance phenotype of the CEM/VM-1-5 cells appears to be the increased level of MRP (∼ 30 times more MRP mRNA and protein than in CEM cells; Wang and Beck, 1998). This overexpression of MRP is believed to be regulated in part by a loss of wt p53 function. MRP promoter activity is suppressed by wt p53 protein, while mutant p53 is permissive for its expression in these cells (Wang and Beck, 1998). Whether an increased half-life of the mp53 protein in the CEM/VM-1-5 cells plays a direct role in this enhanced expression of MRP protein levels remains to be determined.
Overall, our results suggest that the unresponsiveness of mp53 protein in the drug-resistant cells may play a role in the resistance of these cells to drug- and radiation-induced cell death. Studies are presently underway to clarify the relationships between mp53 responsiveness, p53 stability, p53-MDM2 biochemical interactions, and cell cycle and cell death responses in drug-resistant and -sensitive cell lines. Further insight into the differences in signaling pathways between drug-resistant and -sensitive cells should ultimately enhance our knowledge of drug responsiveness.
Materials and methods
Cell lines, and drug and irradiation treatments
The human leukemic CEM cells and the drug-resistant sublines, CEM/VM-1 and CEM/VM-1-5, developed in our laboratory (Danks et al., 1987, 1988), were cultered in SMEM (BioWhittaker) supplemented with 10% fetal bovine serum (Sigma Chemical Co.), and 2 mM L-glutamine (BioWhittaker) as described previously (Danks et al., 1987; Chen and Beck, 1993). The two human ovarian cell lines, PA-1 (harboring a wild-type p53), and SW626 (harboring a mutant p53) (American Type Culture Collection), were grown in RPMI 1640 supplemented with 10% fetal bovine serum and 2 mM L-glutamine. All cell lines were incubated at 37°C in a humidified chamber containing 5% CO2, 95% air. VM-26 was generously provided by Bristol Myers Squibb, through the courtesy of Dr Byron H Long (Princeton, NJ, USA). Exponentially growing cells were irradiated at the indicated doses with a 137Cs source at a dose rate of approximately 3.55 Gy/min. The proteosome inhibitor ALLN (Sigma) was added to the cells at a final concentration of 20 μM. Cells were harvested at indicated times after DNA damage or ALLN treatments.
For amplification of p53 exons, single-stranded p53 cDNAs were synthesized from total RNA (5 μg) derived from CEM, CEM/VM-1 and CEM/VM-1-5 cells using Superscript II RT (Life Technologies) in the presence of 3′ gene-specific primers. Following the RT reaction, cDNAs were PCR-amplified according to standard conditions: denaturing at 94°C, 30 s; annealing at 50°C, 30 s; extension at 72°C, 1.5 min, using a commercial kit (Ampli-Taq, Perkin-Elmer Corp.). PCR clones were sequenced by the dideoxy chain termination method with the Sequenase kit version 2 (Amersham) according to the manufacturer's instructions. The following primers were used for the amplification of p53 exons: 2-3, AGCCAGATCGCCTTCCGGGTC (sense) and TTCATCTGGACCTGGGTCTT (antisense); exon 4, ATTTGATGCTGTCCCCGGAC (sense) and ACAGACTTGGCTGTCCCAGA (antisense); exon 5, TACGGTTTCCGTCTGGGCTT (sense) and GTGCTGTGACTGCTTGTAGA (antisense); exon 6, GCCATCTACAAGCAGTCACAG (sense) and CACCACACTATGTC GAAAAGT (antisense); exon 7, GGAAATTTGCGTGTGGAGTA (sense) and CTTCCAGT GTGATGATGGTG (antisense); exon 8–9, ATGAACCGGAGGCCCATCCT (sense) and CATTCAGCTCTCGGAACATC (antisense); and exon 10–11, CCAAAGAAGAAACCACTGGAT (sense) and TCTCAGTGGGGAACAAGAAG (antisense).
Immunoprecipitation of p53 from 35S-labeled cellular extracts
Cells (6×106 total) were washed and resuspended in methionine-free medium and incubated for 30 min at 37°C. This medium was then replaced with methionine-free medium containing 100 μCi 35S-methionine/ml (Trans 35S-label, 1000 Ci/mmol, ICN), and the cells were incubated for 1 h. After washing with PBS, the cells were incubated on ice for 15 min in 1×lysis buffer (0.58 M Na2HPO4, 0.17 M NaH2PO4, 0.68 M NaCl, 1% Triton X-100, 0.1% sodium deoxycholate, 0.1% SDS, 0.5 mM PMSF, 10 μg aprotinin/ml). The lysates were centrifuged at 10 000 g for 30 min to remove insoluble material. For immunoprecipitation, the cellular extract was pre-cleared with protein-A agarose (Oncogene) for 1 h. After removing the protein-A agarose, 1 μg of p53 antibody (either PAbs 421, 1620, or 240; Oncogene) was added to the supernatants, which were incubated with rotation at 4°C for 1 h, after which protein-A agarose was added for an additional 1 h. The immunoprecipitates were collected by centrifugation, and the pellets were washed four times with 1×lysis buffer. The pellets were resuspended in SDS sample buffer (65 mM, Tris-HCl pH 6.8, 5% 2-mercaptoethanol, 3% SDS, 10% glycerol), boiled for 5 min, loaded onto 10% SDS polyacrylamide gels, and the proteins were separated by electrophoresis (Laemmli, 1970). The gels were dried and directly autoradiographed.
For pulse-chase experiments, cells (6×106 total) were washed and resuspended in methionine-free medium and incubated for 30 min at 37°C. This medium was then replaced with methionine-free medium containing 100 μCi 35S-methionine/ml, and the cells were incubated for 1 h, after which the medium was replaced with 50 mM methionine-containing medium and cells were harvested at various time points as indicated in the figures. Immunoprecipitations were performed as described above.
Co-immunoprecipitation of p53 and MDM2
Exponentially growing cells were washed in 1×ice-cold PBS and lysed at 4°C in 1×lysis buffer (250 mM NaCl, 1% NP-40, 20 mM Tris-HCl at pH 7.5, 1 mM EDTA, 2 μg aprotinin/ml, and 0.5 mM PMSF). Lysates were clarified at 10 000 g for 15 min and pre-incubated with protein-G/protein-A agarose at 4°C for 1 h, while rocking. These lysates were then centrifuged (10 000 g) and the supernatant was incubated with 5 μg of mouse monoclonal anti-p53 antibodies (Ab-2 and Ab-6; Oncogene) on ice for 2 h. As a control for nonspecific binding, an immunoprecipitation reaction was also done using a primary mouse monoclonal antibody of the same isotype as the anti-p53 antibodies (mouse IgG2a, Kappa immunoglobulin, Sigma). Protein-G /protein-A agarose (Oncogene) was added and the lysates were further incubated at 4°C for 1 h, with rocking. The immunoprecipitate beads were collected by centrifugation, and the pellets were washed three times with ice-cold 1×wash buffer (150 mM NaCl, 1% NP-40, 20 mM Tris-HCl at pH 7.5, 1 mM EDTA, 2 μg aprotinin/ml, and 0.5 mM PMSF). The beads were boiled in SDS sample buffer, and immunoprecipitated p53 was electrophoresed on a 10% SDS polyacrylamide gel. Protein was electrophoretically transferred to Immobilon-P membrane (Millipore) and the levels of p53 and MDM2 protein were detected by immunoblot analysis using either a polyclonal sheep anti-p53 antibody (Ab-7; Oncogene), or a monoclonal anti-MDM2 antibody (Ab-1; Neomarkers). Bound antibody was detected using the enhanced chemiluminescence. (ECL) detection method (Amersham) according to the manufacturer's instructions.
Untreated or treated cells (∼5×105 cells/ml) were harvested and solubilized in Laemmli sample buffer (Laemmli, 1970). For detection of p53, p53 serine-15 phosphorylation, or MDM2, cellular extracts (100 μg/well) were separated in 10% SDS-polyacrylamide gels. For analysis of p21, GADD45, bcl-2, or bax protein, cellular extracts (100 μg/well) were separated in 12% SDS-polyacrylamide gels. All gels were electrophoretically transferred onto nitrocellulose and filters were blocked in 5% BSA in TBST buffer (40 mM Tris-HCl, pH 7.2, 150 mM NaCl, 0.05% Triton X-100) for 1 h at room temperature. After incubation with either a monoclonal anti-p53 antibody (Ab-6; Oncogene), a polyclonal anti-phospho-53-serine15 antibody (New England Biolabs), a monoclonal anti-MDM2 antibody (Ab-1; Neomarkers), a monoclonal anti-p21 antibody (Ab-5; Neomarkers), a polyclonal anti-GADD45 antibody (Zymed), or polyclonal anti-bcl-2 or -bax antibodies (Santacruz Biotechnology) for 2 h, filters were washed three times in TBS buffer containing 0.05% Triton X-100. Bound antibody was detected using the enhanced chemiluminescence (ECL) detection method (Amersham) according to the manufacturer's instructions. Autoradiographic signals were quantified by densitometric scanning using a GS-700 Imaging Densitometer and Molecular Analyst Software (Bio-Rad). All blots were stripped and re-probed using anti-β-actin antibody (Oncogene Research Products) to control for protein loading.
MTT cytotoxicity assay
Exponentially growing CEM, CEM/VM-1, and CEM/VM-1-5 cells were irradiated at the indicated doses and then immediately plated at 500 cells/well in 96-well microtiter plates (200 μl/well), and the cells were incubated at 37°C for 96 h. After drug exposure, 25 μl of MTT compound (Sigma; 4 mg/ml in SMEM without FBS) were added to each well and the cells were incubated at 37°C for 4 h. The plates were centrifuged in a swinging bucket rotor (1000 g, 5 min), and the cells were incubated with 200 μl of DMSO for 15 min at 25°C. The metabolic activity of the cells was measured by quantifying the conversion of the yellow MTT to a purple metabolite, MTT-formazan. Optical density was read at 540 nm using a microplate reader. Four replicates were measured for each drug concentration.
Alarcon R, Koumenis C, Geyer RK, Maki C and Giaccia AJ.. 1999 Cancer Res. 59: 6046–6051.
Arriola EL, Lopez AR and Chresta CM.. 1999 Oncogene 18: 1081–1091.
Barak Y, Juven T, Haffner R and Oren M.. 1993 EMBO J. 12: 461–468.
Boise LH, Gottschalk AR, Quintans J and Thompson CB.. 1995 Curr. Topics. Microbiol. Immunol. 200: 107–121.
Brown R, Clugston C, Burns P, Edlin A, Vasey P, Vojtesek B and Kaye SB.. 1993 Int. J. Cancer 55: 678–684.
Bugg BY, Danks MK, Beck WT and Suttle DP.. 1991 Proc. Natl. Acad. Sci. USA 88: 7654–7658.
Chen M and Beck WT.. 1993 Cancer Res. 53: 5946–5953.
Cheng EHJ, Kirsch DG, Clem RJ, Ravi R, Kastan MB, Bedi A, Ueno K and Hardwick JM.. 1997 Science 278: 1966–1968.
Cheng J and Haas M.. 1990 Mol. Cell. Biol. 10: 5502–5509.
Craig RW.. 1995 Semin. Cancer Biol. 6: 35–43.
Danks MK, Schmidt CA, Cirtain MC, Suttle DP and Beck WT.. 1988 Biochemistry 27: 8861–8869.
Danks MK, Warmoth MR, Friche E, Granzen B, Bugg BY, Harker WG, Zwelling LA, Futshcer BW, Suttle DP and Beck WT.. 1993 Cancer Res. 53: 1373–1379.
Danks MK, Yalowich J and Beck WT.. 1987 Cancer Res. 47: 1297–1301.
Donehower LA and Bradley A.. 1993 Biochem. Biophys. Acta 1155: 181–205.
El-Deiry WS, Tokino T, Velculescu VE, Levy DB, Parsons R, Trent JM, Lin D, Mercer WE, Kinzler KW and Vogelstein B.. 1993 Cell 75: 817–825.
Eliyahu D, Michalovitz D, Eliyahu S, Pinhasi-Kimhi O and Oren M.. 1989 Proc. Natl. Acad. Sci. USA 86: 8763–8767.
Finlay CA, Hinds PW and Levine AJ.. 1989 Cell 57: 1083–1093.
Foster BA, Coffey HA, Morin MJ and Rastinejad F.. 1999 Science 286: 2507–2510.
Giaccia AJ and Kastan MB.. 1998 Genes. Dev. 12: 2973–2983.
Harper JW, Adami GR, Wei N, Keyomarsi K and Elledge SJ.. 1993 Cell 75: 805–816.
Hartwell LH and Kastan MB.. 1994 Science 266: 1821–1828.
Haupt Y, Maya R, Kazaz A and Oren M.. 1997 Nature 387: 296–299.
Hollstein M, Sidransky D, Vogelstein B and Harris CC.. 1991 Science 253: 49–53.
Honda R, Tanaka H and Yasuda H.. 1997 FEBS Lett. 420: 25–27.
Kastan MB, Onyekwere O, Sidransky D, Vogelstein B and Craig RW.. 1991 Cancer Res. 51: 6304–6311.
Kastan MB, Zhan Q, El-Diery WS, Carrier F, Jacks T, Walsh WV, Plunkett BS, Vogelstein B and Fornace AJ.. 1992 Cell 71: 587–597.
Kim R and Beck WT.. 1994 Cancer Res. 54: 4958–4966.
Kubbutat MHG, Jones SN and Vousden KH.. 1997 Nature 387: 299–303.
Kubbutat MHG, Ludwig RL, Levine AJ and Vousden KH.. 1999 Cell Growth Differ. 10: 87–92.
Kwok TT, Mok CH and Menton-Brennan L.. 1994 Cancer Res. 54: 2834–2836.
Laemmli UK.. 1970 Nature (Lond.) 227: 680–685.
Maki CG, Huibregtse JM and Howley PM.. 1996 Cancer Res. 56: 2649–2654.
Maltzman W and Czyzyk L.. 1984 Mol. Cell. Biol. 4: 1689–1694.
Miyashita T, Krajewski S, Krajewska M, Wang HG, Lin HK, Liebermann DA, Hoffman B and Reed JC.. 1994 Oncogene 9: 1799–1805.
Miyashita T and Reed JC.. 1995 Cell 80: 293–299.
Momand J, Zambetti GP, Olson DC, George DL and Levine AJ.. 1992 Cell 69: 1237–1245.
Morgan SE and Kastan MB.. 1997 Adv. Cancer Res. 71: 1–25.
Oltvai ZN, Milliman CL and Korsmeyer SJ.. 1993 Cell 74: 609–619.
Park DJ, Nakamura H, Chumakov AM, Said JW, Miller CW, Chen DL and Koeffler HP.. 1994 Oncogene 9: 1899–1906.
Prives C.. 1998 Cell 95: 5–8.
Reed JC.. 1994 J. Cell. Biol. 124: 1–6.
Sedlak TW, Oltvai ZN, Yang E, Wang K, Boise LH, Thompson CB and Korsmeyer SJ.. 1995 Proc. Natl. Acad. Sci. USA 92: 7834–7838.
Selvakumaran M, Lin H, Miyashita T, Wang HG, Krajewski S, Reed JC, Hoffman B and Liebermann D.. 1994 Oncogene 9: 1791–1798.
Siliciano JD, Canman CE, Taya Y, Sakaguchi K, Appella E and Kastan MB.. 1997 Genes. Dev. 11: 3471–3481.
Smith ML, Chen I-T, Zhan Q, Bae I, Chen C-Y, Gilmer TM, Kastan MB, O'Connor PM and Fornace Jr AJ.. 1994 Science 266: 1376–1380.
Tishler RB, Calderwood SK, Coleman CN and Price BD.. 1993 Cancer Res. 53: 2212–2216.
Ullrich SJ, Anderson CW, Mercer WE and Appella E.. 1992 J. Biol. Chem. 267: 15259–15262.
Wang Q and Beck WT.. 1998 Cancer Res. 58: 5762–5769.
Zambetti GP and Levine AJ.. 1993 FASEB J. 7: 855–865.
Zauberman A, Barak Y, Ragimov N, Levy N and Oren M.. 1993 EMBO J. 12: 2799–2808.
Zhang W, Funk WD, Wright WE, Shay JW and Deisseroth AB.. 1993 Oncogene 8: 2555–2559.
We are grateful to Linda Rawlinson and the Biomedical Communications Department at St. Jude Children's Research Hospital for preparation of some of the artwork. This work was supported in part by Research Grants CA40570 and CA30103 (to WT Beck) from the National Cancer Institute, DHHS, Bethesda, MD, and in part by the University of Illinois at Chicago.
AffiliationsDivision of Molecular Pharmacology, Department of Molecular Genetics, University of Illinois at Chicago, Chicago, Illinois, IL 60607, USASusan E Morgan, Pu Chen Wang, Uppoor G Bhat & William T BeckDepartment of Surgical Oncology, Hiroshima University, Kasumi, Minami-ku, Hiroshima 734-8553, JapanRyungsa KimDepartment of General Surgery, Fukuoka Dental College Hospital, Fukuoka 814-01, JapanHiroki KusumotoDepartment of Cell and Gene Therapy, St. Jude Children's Research Hospital, Memphis, Tennessee, TN 38101, USATaihe LuDepartment of Pharmaceutics and Pharmacodynamics, University of Illinois at Chicago, Chicago, Illinois, IL 60607, USAWilliam T BeckCurrent address: HPD Biologics Development, Abbott Laboratories, Abbott Park, Illinois, IL 60064, USASusan E Morgan
AuthorsSearch for Susan E Morgan in:Nature Research journals • PubMed • Google ScholarSearch for Ryungsa Kim in:Nature Research journals • PubMed • Google ScholarSearch for Pu Chen Wang in:Nature Research journals • PubMed • Google ScholarSearch for Uppoor G Bhat in:Nature Research journals • PubMed • Google ScholarSearch for Hiroki Kusumoto in:Nature Research journals • PubMed • Google ScholarSearch for Taihe Lu in:Nature Research journals • PubMed • Google ScholarSearch for William T Beck in:Nature Research journals • PubMed • Google Scholar
Correspondence to William T Beck.
Rights and permissions
To obtain permission to re-use content from this article visit RightsLink.
About this article
31 January 2000
14 August 2000
14 August 2000
12 October 2000