Abstract
There is significant morphologic, immunophenotypic and molecular overlap between some subtypes of diffuse large B-cell lymphoma and classical Hodgkin lymphoma (cHL). In addition, recent studies reveal similar gene expression profiles in PMBCL and cHL suggesting a common cell of origin. We utilized a differential isotopic strategy to determine the global proteomic differences between cell lines derived from cHL (L428), DLBCL (SUDHL-9) and PMBCL (Karpas 1106P). Protein was collected from cell lysates and subjected to labeling by isobaric tags (iTRAQ) for relative quantification and analyzed by reverse-phase liquid chromatography and electrospray ionization tandem mass spectrometry. 200 mg of total cell lysates from each cell line was used and liquid chromatography and tandem mass spectrometry (LC-MS/MS) analyses were performed in duplicate. The proteins were identified using X!Tandem. After normalization, quantitative data were subjected to false discovery rate (FDR) calculation to identify differentially expressed proteins through Mixture Modeling. Significantly differentially expressed proteins were scored at a false discovery rate (FDR) cut-off of ≤ 0.13. This approach yielded 66 proteins with differential expression patterns that were discriminative for the 3 different cell lines. Several proteins that have been previously reported to be differentially expressed between cHL and PMBCL were identified including Fascin, Galectin-1, Galectin-3, STAT1 and SWAP70. In addition, several previously unreported proteins involved in numerous cellular functions were differentially expressed (cell adhesion, signaling, immunity). We have validated a subset of these by Western blot (WB) analysis and immunohistochemistry. For example, the π class of glutathione S-transferases (GSTP), was over-expressed six-fold in K1106P cells vs. L428 cells, and virtually absent in the SUDHL-9 cell line, as confirmed by WB. This enzyme class has been shown to be overexpressed in many human cancers and to be involved in therapy refractoriness. More recently, GSTP has been implicated as a negative regulator of cellular death/apoptosis through the MAP kinase pathway. These factors, and its unusual absence in the SUDHL-9 cell line made GSTP a rational candidate for further functional analysis. Briefly, SUDHL-9 cells and the cHL cell lines L428 and KMH2 were exposed to increasing concentrations (20–80 μM) of the GSTP inhibitor and thiol modifier molecule ethacrynic acid (EA), and subjected to viability assays (WST-1) and to WB for apoptotic markers. Both cHL cell lines were equally resistant to EA concentrations up to 60μM by WST-1 assays, while SUDHL-9 cells were exquisitely sensitive, undergoing total cell necrosis at the lowest (20 μM) concentration. WB analysis for markers of apoptosis showed a higher level of apoptosis in SUDHL-9 cells compared to the HL-derived KMH2 and L428 cells, as evidenced by poly (ADP-ribose) polymerase (PARP) cleavage into its 85 kDa apoptosis-related fragment. L428 cells, expressing 40% the GSTP levels of KMH2 cells by WB, were more susceptible to apoptosis at 60μM EA than KMH2 cells, as evidenced by decrease in S-phase Kinase-associated Protein 2 (p45SKP2), with concomitant increase in p27Kip1, a negative regulator of G1-S progression. These results demonstrate an inverse relationship between GSTP levels and susceptibility to EA-induced cell death in lymphoma. This study demonstrates the utility of large-scale mass spectrometry-based proteomics for the discovery of proteins that may serve as potential diagnostic marker panels for the distinction of PMBCL, DLBCL and cHL, as well as candidate therapeutic targets.
Disclosures: No relevant conflicts of interest to declare.
Author notes
Corresponding author