Gene expression differences within an in vitro germinal center

Eight families in Iceland have been identified with multiple cases of monoclonal gammopathies (MG), including monoclonal gammopathy of undetermined significance (MGUS), multiple myeloma (MM) and Waldenström’s macroglobulinemia (WM). It has also been shown that there is a multi-step transformation fr...

Full description

Bibliographic Details
Main Author: Schiffhauer, Heather Rene, 1987-
Other Authors: Háskóli Íslands
Format: Thesis
Language:English
Published: 2012
Subjects:
Gen
Online Access:http://hdl.handle.net/1946/11831
Description
Summary:Eight families in Iceland have been identified with multiple cases of monoclonal gammopathies (MG), including monoclonal gammopathy of undetermined significance (MGUS), multiple myeloma (MM) and Waldenström’s macroglobulinemia (WM). It has also been shown that there is a multi-step transformation from MGUS to MM. An in vitro poke-weed mitogen (PWM) stimulation assay identified 13 disease-free relatives of patients within four of the eight families as having hyper-responding (HR) B-lymphocytes (B cells) producing Ig at > 2 Standard Deviations (SD) above controls. We established an in vitro model of the germinal center (GC) reaction and performed gene expression (GE) analysis on unstimulated samples and those collected after 14 days in the GC culture to investigate if the GC model faithfully mimics in vivo GC. We also screened for differences between controls (related and unrelated) and the HR group. B cells were isolated from peripheral blood (PB) samples from each HR (n=11), unrelated control (UC; n=11), and related (RC; n=9). mRNA was extracted immediately from B cells at day 0 and then from B cells after 14 days of stimulation in the in vitro GC culture and used to synthesize cDNA. Cy-3 labeled samples were loaded onto oligonucleotide microarrays. GE was normalized and appropriate linear-mixed effects models were applied to prepared data for analysis between study groups and stimulation days. Maximum likelihood and likelihood ratio tests and Benjamini and Hochberg correction were applied. Analysis methods included unsupervised hierarchical cluster analysis and ontology group testing using software programs recommended by the GO database. Day 0 and day 14 GE results clustered separately on a heatmap. Approximately one-third (14,277) of gene transcripts were significantly different after 14 days in the GC culture (padjusted < 0.001). Significant genes were ranked by t-statistic to identify the most up- and down- regulated genes. These genes grouped into multiple ontology classes as expected. The most ...