Differential Gene Expression in Nodular Sclerosis Hodgkin Lymphoma: The Role of the Reactive Infiltrate in Disease Pathogenesis.

Abstract

Hodgkin lymphoma is an unusual cancer in that the malignant cells, Hodgkin and Reed Sternberg (H-RS) cells, comprise only 1–2% of the tumor tissue. The ability of the H-RS cell to evade the usual pathways of cell death may be due to complex interactions with reactive cells in the surrounding microenvironment. Few studies have been conducted with whole Hodgkin tumor tissue to better understand gene expression by the non-H-RS cells. In the present experiments, we used high throughput microarray technology to analyze whole Hodgkin tumor tissue from patients with the nodular sclerosis variant of Hodgkin lymphoma with the aim of identifying patterns of genes expressed by the non-H-RS cells. Among the genes significantly changed in their expression in tumor compared to normal samples, we found increased expression of oncogenes, genes known to inhibit apoptosis, and genes involved in regulation of chemotaxis. There was decreased expression of pro-apoptotic genes, tumor suppressor genes, a chemotaxis inhibitor, and transcriptional repressors. RNA was extracted from whole Hodgkin tumor tissue using Trizol (Invitrogen, Carlsbad, CA) and purified using the RNeasy mini-kit (Qiagen, Hilden, Germany). cDNA was amplified using the Biotin Ovation system (Nugen, San Carlos, CA) and biotinylated cDNA was hybridized to Affymetrix (Santa Clara, CA) U133A Plus 2.0 GeneChips. Gene expression data were obtained from 16 nodular sclerosis samples and the results compared to two benign lymph node samples using a Bayesian generalization of the t-test. Samples were analyzed using the Bioconductor package and the results normalized using the Robust Multichip Algorithm (RMA) method. Statistical analysis was performed using the LIMMA (LInear Models for MicroArray) method. Genes with a Bayesian log-odds value (B-value) ≥ 0 were considered to have significant differential expression. Four hundred twenty-nine genes fulfilled this criterion. Differentially expressed genes were sorted according to their molecular function to identify genes with potential relevance in Hodgkin lymphoma pathogenesis. In Hodgkin lymphoma tissue, increased expression of genes with the following functions was found (genes marked with an asterisk were among the ten genes with the most altered expression according to fold change): anti-apoptosis (PPP1CB*, RBPSUH*), oncogene (RUNX1*), and lymphocyte chemoattraction (CKLF). Decreased expression of genes with the following functions was also found: pro-apoptosis (PRKCB1, PAWR, PDCD8, EP400), tumor suppressors (RBBP6, EGR1, FOXP1, PML, PRDM2, SAFB2), transcriptional repressors (PAWR, ZHX2, MORF4L2), chemotaxis inhibition (RGS13*), and initiation of immunoglobulin gene class switching (AICDA). Because of the small percentage of H-RS cells within whole Hodgkin lymphoma tumor tissue, these data largely represent gene expression of the reactive cells surrounding the H-RS cells. Our findings provide insights into the mechanisms by which non-H-RS cells in the microenvironment perpetuate the survival and proliferation of cells in Hodgkin lymphoma.