Gene Expression Profiling Reveals Aberrant T-cell Marker Expression on Tumor Cells of Waldenström's Macroglobulinemia

Abstract

Purpose: That the malignant clone of Waldenström's macroglobulinemia (WM) demonstrates significant intraclonal heterogeneity with respect to plasmacytoid differentiation indicates the mechanistic complexity of tumorigenesis and progression. Identification of WM genes by comparing different stages of B cells may provide novel druggable targets.

Experimental design: The gene expression signatures of CD19⁺ B cells (BC) and CD138⁺ plasma cells (PC) from 19 patients with WM were compared with those of BCs from peripheral blood and tonsil and to those of PCs from the marrow of healthy (N-PC) and multiple myeloma donors (MM-PC), as well as tonsil (T-PC). Flow cytometry and immunofluorescence were used to examine T-cell marker expression on WM tumor cells.

Results: Consistent with defective differentiation, both BCs and PCs from WM cases expressed abnormal differentiation markers. Sets of 55 and 46 genes were differentially expressed in WM-BC and WM-PC, respectively; and 40 genes uniquely dysregulated in WM samples were identified. Dysregulated genes included cytokines, growth factor receptors, and oncogenes not previously implicated in WM or other plasma cell dyscrasias. Interestingly, strong upregulation of both IL6 and IL6R was confirmed. Supervised cluster analysis of PC revealed that marrow-derived WM-PC was either MM-PC-like or T-PC-like, but not N-PC-like. The aberrant expression of T-cell markers was confirmed at the protein level in WM-BC.

Conclusions: We showed that comparative microarray profiles allowed gaining more comprehensive insights into the biology of WM. The data presented here have implications for the development of novel therapies, such as targeting aberrant T-cell markers in WM.

Fig. 1.. Distinct expression patterns of B-cell markers and transcription factors in WM.

Samples included 7 PB-BC, 7 T-BC, 12 WM-BC, 9 WM-PC, 9 T-PCs, 10 N-PC, and 11 MM-PC and are distributed along the x axis; the normalized signal is plotted on the y axis. A, B-cell markers for CD45, CD19, CD20, CD138, CD38, and CD27; the expression of normalized signal appears in blue, red, yellow, cyan, purple, and green, respectively. B, Transcription factors for XBP1, PRDM1, IRF4, and PAX5; the expression of normalized signal appears in blue, red, yellow, and cyan, respectively.

Fig. 2.. Unsupervised hierarchical cluster analysis of WM cases and their controls.

A, Two-way cluster analysis of all 12 WM-BC, 7 PB-BC, and 7 T-BC cases sorted with CD19/CD20 based on the expression of 6,504 genes. Mean-centered gene expression is depicted by a normalized-signal pseudocolor scale. Red and green indicate the genes overexpressed and underexpressed, respectively. Each row represents a gene, and each column a tissue sample. B, Cluster analysis of all 9 WM-PC, 9 T-PC, 10 N-PC, and 11 MM-PC cases sorted with CD138 based on the expression of 7,870 genes.

Fig. 3.. Delineation of the top significantly expressed genes distinguishing WM cases from the controls.

A, Expression of the 55 top significantly expressed genes in WM-BC samples. The WM-BC-defined clusters for tumor B-cells (horizontal red bar), PB-BC (horizontal green bar), and T-BC (horizontal blue bar) are indicated. B, Expression of the 46 significantly expressed genes in WM-PC. The WM-PC-defined clusters for tumor plasma cells (horizontal red bar), N-PC (horizontal green bar), and T-PC (horizontal blue bar) are indicated. The clustering analysis illustrates that the expression pattern of WM-PC is more similar to that of T-PC than N-PC.

Fig. 4.. Aberrant T cell marker expression in WM-B cells.

A, A two dimensional hierarchical cluster of 12 WM-BC and 9 WM-PC samples clustered based on the correlation of experimental expression signatures of 13 probe sets representing T-cell markers or related genes. The clustering was presented graphically as a colored image. Along the vertical axis, the analyzed genes were arranged as ordered by the similar patterns of expression. Experimental samples were arranged the same way along the horizontal axis; those with the similar patterns of expression across all genes were adjacent to each other. The samples labelled the Red-bar were derived from 7 WB-BC samples. B, Cell analyses of T-cell marker expression in WM-BC samples. Two-color dot plots of flow cytometry analysis showed the population of CD19 and CD45 double stain cells derived from bone marrow aspirates of WM patients, the strong CD19 and CD45 positive B cells were labelled with green color (Top panel); Two-color dot plots presented dual antibody analyses of CD3 and CD8 in CD19⁺CD45⁺ enriched WM-B cells (green-labelled in the top panel), about 4.49% (0.1 ~ 13.4%) CD3⁺CD8⁺ cells were found in the CD19⁺CD45⁺ WM-B cells (Bottom panel). C, Immunofluorescence confirms CD3 expression on CD19-positive WM-B cells. The CD19⁺CD45⁺ WM-B cells sorted out from above 5 WM patients were also performed CD19 (red) and CD3 (green) double stains by immunofluorescence; the yellow color indicated the merged signals. DAPI (blue color) was used to stain cell nuclei.