Figure 1

MYC was overexpressed and related to aberrant choline metabolism in DLBCL. (a) MYC protein was High (⩾40%) in 47 of 108 (44%) patients with DLBCL. (b) The score plots of the principal component analysis (PCA) models (left panel) were set up using data from 15 MYC-High DLBCL cases (green plots), 15 MYC-Low DLBCL cases (blue plots) of the training set and 10 repeated quality controls (QCs, red plots). The score plots of the orthogonal partial least-squared discrimination analysis (OPLS-DA) models (middle panel) discriminated the MYC-High group (green plots) from the MYC-Low group (blue plots), which was confirmed by response permutation test (RPT, right panel). (c) Enrichment of KEGG pathways was performed on all the dysregulated metabolites using ConsensusPathDB. X-axis displayed P-values that were generated by –Log(P) function. The P<0.05 and—Log(P)>1.30 were considered significant. (d) Genesets of choline metabolism in cancer and glycerophospholipid metabolism pathway were analyzed by gene network and pathway analysis on microarray data of nine non-GCB DLBCL cases according to MYC expression (left panel). Significantly dysregulated genes involved in the metabolic pathways were shown in right panel. Shades of red and green represented high or low expression. (e) The OPLS-DA score plots confirmed a global metabolic difference between the MYC-High group (N=25, green plots) and the MYC-Low group (N=25, blue plots) in the validation set. (f) B-lymphoma cell line DB was either treated with 10058-F4 (40 μM) for 48 h or transfected with MYC siRNA. Expression of Choline, PC(16:0/22:6), LPC(16:0) and LPC(18:0) were detected by UPLC-triple quadrupole mass spectrometry. Data in f were represented as mean±s.e.m. Assay in f was set up in triplicate.