Bortezomib, the first selective proteasome inhibitor in this new category of anti-cancer drug, inhibits chymotrypsin-like activity sited at beta 5 subunit of proteasome (PSMB5). To study the mechanism of proteasome inhibitor resistance in tumor cells, we established a series of bortezomib resistant lymphoblastic leukemia cell lines, named JurkatBs, from Jurkat line by repeated drug induction. There were no significant differences observed in the growth curves, colony formation rates, or cell cycle distribution between the JurkatBs and Jurkat cells. However, the effects of bortezomib, namely cytotoxicity, cell cycle arrest at G2 phase and induction of apoptosis, were decreased significantly in JurkatBs cells. There were no significant differences of intracellular mean fluorescence intensities of daunorubicin between JurkatBs and Jurkat cells (P>0.05) after treated with daunorubicin. Accordingly, the JurkatBs showed no cross-resistance to anthracycline, alkaloid and topoisomerase inhibitor. Quantitative PCR analysis showed no significant overexpression of MDR1 gene in JurkatB1, JurkatB2, JurkatB5 cells in comparison with that of Jurkat cells (22.77±5.58, 1.17±0.23, 8.30±2.62 vs 1.00±0.50; P>0.05). P-gp expression analysis was also negative in JurkatBs and Jurkat cells by Western bloting. By using quantitative PCR, we found that the PSMB5 gene was significantly amplified in JurkatB1 (5.82±0.60) and JurkatB5 cells (6.78±1.21) in comparison with that of Jurkat cells (1±0.49)(P<0.001), but not in JurkatB2 cells(0.16±0.03, P=1.000). A specific chromosome abnormality, i(14q), was found in all JurkatB5 cells (highly resistant to bortezomib), 4/16 JurkatB1 cells (moderately resistant), but not in JurkatB2 cells (slightly resistant). This results suggested that i(14q) may be resulted from the amplification of PMSB5 gene, which is located at 14q11. The chymotrypsin-like activities, determined by measuring the release of the fluorescent AMC from the substrate N- Suc-Leu-Leu-Val- Tyr-AMC, increased significantly in JurkatB1 (relative activity 3.27±0.12) and JurkatB5 cells (5.75±0.22) in comparison with that in Jurkat cells (1.00±0.14; P<0.001), but not in JurkatB2 cells (0.92±0.09; P>0.05). These results were coincident with the amplication of PSMB5, which may partly elucidated the bortezomib resistance in JurkatB cells. We then cloned and sequenced the full-length cDNA product of the PSMB5 gene from JurkatBs and Jurkat cells. A mutation at position 322 (G322A) of PSMB5 gene, causing an an amino acid substitution (Ala108Thr), was found in all selected JurkatB clones. The inhibition of chymotrypsin-like activities in JurartB2 (39.66±2.89) and JurartB5 cells (1.71±3.51), incubated with 10nM bortezomib up to 18h, were decreased significantly in comparison with that of Jurkat cells (86.87±0.97; P<.001), suggesting a decreased binding affinity of bortezomib to the chymotrypsin-like active site caused by Ala108Thr in JurkatB cells, which may resulted in conformation change in β5 subunit. In conclusion, we established bortezomib resistant leukemia cell lines with a different mechanism from that of multi-drug resistance (MDR). Both amplification and G322A mutation of PSMB5 gene are the important mechanisms of bortezomib resistance, by increasing chymotrypsin-like activity, and decreasing binding affinity of bortezomib to the chymotrypsin-like site, respectively.

Author notes

Disclosure: No relevant conflicts of interest to declare.