Identification of Differentially Expressed Genes in the Rituximab- Resistant Clone (Ramos RR1) Compared to Wildtype Sensitive Ramos: Therapeutic Implications in Rituximab-Resistance

Abstract

Treatment of patients with relapsed or refractory B-NHL with rituximab alone or in combination with CHOP has resulted in significant clinical response. However, a subset of patients does not initially respond or develops resistance to such therapies. The mechanism underlying rituximab resistance is not fully understood. We have explored a possible mechanism by generating in the laboratory rituximab resistant clones (Ramos RR, Daudi RR, and 2F7 RR) and reported that the clones, unlike the parental wild type (wt), no longer respond to rituximab-induced cell signaling and chemo- immuno-sensitization. For instance, we have demonstrated that the resistant clones exhibit hyperactivation of cell survival signaling pathways, such as the NF-kB and Raf-1/MEK/ERK pathways, and overexpress several anti-apoptotic gene products that regulate apoptosis (Jazirehi et al., Cancer Research 67:1270–1281, 2007). To further characterize the molecular basis of rituximab-resistance, we analyzed the gene expression profile of Ramos and Ramos RR1 using oligonucleotide microarrays. There were only a few genes that were significantly modified and we have focused on such genes for analysis. First, we analyzed genes that were present in Ramos RR1 and silenced in wtRamos, namely, KIAA0738, DHTKDU1, PTPRO, REG3A, and ATXN10. Analysis of some of these gene products revealed their possible role in the regulation of resistance. For instance, REG3A, regenerating islet-derived 3α, a growth promoting lectin, has been reported to be involved in various biological functions including proliferation and resistance to apoptosis and silenced by hypermethylation. This suggests that REG3A expression in Ramos RR1 may be due to inhibition of hypermethylation. Another gene product silenced in wtRamos but expressed in Ramos RR1 is the protein tyrosine phosphatase receptor type O (PTPRO), which is known to be hypermethylated in response to rituximab therapy. Overexpression of PTPRO inhibits BCR-triggered syk tyrosyl phosphorylation and cell signaling. Hence, rituximab has been reported to trigger cells through the association of BCR with CD20 on the membrane. Second, we analyzed differentially expressed genes between Ramos RR1 and wtRamos. Two gene products, MAP3K14, and ACN9 were most overexpressed, whereas PAM, SDCCAG33, and NPAS4 were most downregulated in Ramos RR1. Overexpression of MAP3K14 (NIK) is in agreement with the hyperactivated state of Ramos RR1, and in addition to our findings of hyperactivation of the canonical NF-κB pathway, would suggest that the non-canonical pathway is also activated. Hyperactivation of NIK in Ramos RR1 may be due to its stabilization by NF-κB-dependent degradation of TRAF-3 and our findings are consistent with NIK overexpression in other blood malignancies. PAM (phosphatydyl-glycine α amidating monooxygenase) is an enzyme that has been reported to play a critical role in an autocrine pathway promoting proliferation. Its downregulation in Ramos RR1 suggests PAM-independence. Findings will be presented on the validation and roles of the above gene products in the regulation of rituximab resistance and approaches to target such gene products to reverse resistance.