The potential therapeutic role of selenium in diffuse large B-cell lymphoma

Abstract

The clinical background to this work confirms the very poor outcome of patients with recurrent diffuse large B-cell lymphoma (DLBCL) and highlights the need for better therapies. One such potential option would be to add selenium (Se) to conventional chemotherapy. Previous work has demonstrated the ability of non-toxic concentrations of Se to sensitise DLBCL cell lines to chemotherapy and to protect normal cells from chemotherapy-induced toxicity. The aims of this study were therefore to identify mechanisms of Se action and potential biomarkers of Se activity. The form of Se used was methylseleninic acid (MSA), a precursor of methylselenol, which is the metabolite thought to be responsible for the anti-tumour effects of organic Se compounds. DLBCL cell lines differed in their sensitivity to MSA, which may relate to differences in intracellular glutathione depletion by MSA. MSA sensitivity, however, was not related to the induction of DNA damage or to the p53 status of lymphoma cell lines. Although cytotoxic concentrations of MSA induced apoptosis, chemo-sensitising concentrations did not enhance apoptosis or alter pro-apoptotic pathways. MSA induced endoplasmic reticulum (ER) stress in a concentration-dependent manner, however, in an MSA-resistant cell line, this led to autophagy and cell survival. Thus, ER stress induction is not a mechanism of chemo-sensitisation. MSA inhibited HDAC activity in DLBCL cell lines but only in a cell-based assay, suggesting that a metabolite of MSA is responsible for this effect. In addition, MSA inhibited the hypoxia-induced induction of HIF-1α in DLBCL cell lines. Peripheral blood mononuclear cells (PBMCs) were relatively resistant to MSA and this was associated with increased expression of two pro-survival proteins, GRP78 and NF-κB. In addition, the metabolism of MSA differed between PBMCs and DLBCL cell lines, suggesting that methylselenol is formed more efficiently in the latter. In contrast, keratinocytes and fibroblasts were relatively sensitive to MSA, but MSA was unable to protect keratinocytes from the toxicity of chemotherapeutic agents. These results differ 3 from those obtained in DLBCL cell lines in which MSA enhances the activity of chemotherapeutic agents. Combining MSA and bortezomib in mantle cell lymphoma cell lines unexpectedly resulted in an antagonistic interaction. This was associated with the induction of ER stress and autophagy and increased expression of two pro-survival proteins, Bcl-2 and Mcl-1. A proteomics approach identified novel protein changes induced by chemo-sensitising concentrations of MSA in two DLBCL cell lines. Several potential biomarkers of Se activity were identified; GRP78, NF-κB, vascular endothelial growth factor and acetylated histone H3. In conclusion, Se in the form of MSA affects many intracellular pathways in DLBCL cell lines, such that it has not been possible to identify a single unifying mechanism of Se action. However, differences have been observed between PBMCs and DLBCL cell lines and this work has identified novel protein changes and mechanisms of Se action