Signatures of Response to the Proteasome Inhibitor Bortezomib in Diffuse Large B-Cell Lymphoma.

The proteasome inhibitor, bortezomib (VELCADE®, formerly PS341), has significant anti-tumor activity in several lymphoid malignancies. Reported targets of this broad-based inhibitor include the NF K B pathway (I K B A). Recently defined subtypes of large B-cell lymphoma (LBCL) exhibit constitutive activation of NF K B, prompting us to analyze the efficacy of bortezomib in a panel of 10 DLBCL cell lines. Six of the diffuse LBCL cell lines were sensitive to bortezomib treatment at doses below 10 nM (range IC50 = 2.9 to 6.9 nM) whereas 4 cell lines were resistant at 10 nM (IC50 = 14.8 to 70.2 nM). Baseline proteasomal function, as defined by cleavage of the 20S proteasome-specific fluorogenic peptide LLVY-AMC, was similar in sensitive and resistant DLBCLs; however, the IC50 for bortezomib proteasomal inhibition was somewhat lower in sensitive vs. resistant lines (sens. vs res., p = .04, one-sided t test). Baseline NF K B activity varied widely in the DLBCL cell lines and did not differ in cell lines that were sensitive vs. resistant to bortezomib. Ten nM bortezomib did not inhibit NF K B activity in resistant DLBCL cell lines whereas the same dose reduced NF K B activity in sensitive DLBCL cell lines (sens. vs. res., p < .005, rank test [Mann-Whitney]). However, 5 of 6 sensitive DLBCL cell lines had very low baseline NF K B levels (< 0.5 relative absorbance units) suggesting that NF K B inhibition was not a major factor in bortezomib response and prompting further analysis of additional bortezomib targets. Three sensitive and 1 resistant DLBCL cell line were selected for detailed analyses of transcriptional profiles following bortezomib treatment. We developed an algorithm for identifying genes that were significantly up- or down-regulated in the bortezomib-sensitive cell lines but unchanged in the resistant line. In addition, we utilized gene set enrichment analysis (GSEA) and gene ontogeny (GO) termed enrichment to interpret the molecular signatures of response. Genes down-regulated in response to bortezomib included critical B-cell transcription factors, components of the B-cell receptor signaling cascade and genes regulating mitosis and cell cycle control; up-regulated genes included heat shock proteins (HSP) and multiple proteasomal components. Consistent with the functional data, down-regulation of NF K B target genes was not a common feature in all bortezomib-sensitive cell lines. In contrast, target genes of the c-MYC transcription factor were significantly down-regulated and c-MYC activity was decreased in sensitive (but not resistant) DLBCL cell lines following bortezomib treatment (sens. vs. res., p < .005, rank test). Taken together, the results provide insights into likely mechanisms of action of bortezomib in DLBCL, highlighting c-MYC as a potentially important target and identifying HSP as a complementary target to overcome bortezomib resistance.