Apparent Involvement Of The Interferon, RNA Processing, and Wnt Signaling Pathways In Monoclonal B Lymphocytosis

Chronic lymphocytic leukemia (CLL) develops in a stepwise fashion, starting in a restricted set of normal B lymphocytes that clonally expand, presumably due to antigen stimulation, reaching levels exceeding normal homeostasis but less than required for a diagnosis of CLL. Immunologic, genetic, and epidemiologic studies suggest these clonal expansions, termed monoclonal B lymphocytosis (MBL), are requisite precursors of CLL. Although ∼5% of normal subjects over age of 60 years old exhibit MBL, only ∼1% evolve to overt CLL each year. Hence, it is likely the genetic factors leading to the development of MBL from normal B cells are not sufficient to automatically lead to CLL and additional genetic lesions are needed for the final conversion to leukemia. To understand the development of MBL and its evolution to CLL, we investigated gene expression profiles of normal blood (N) B cells, MBL cells, and CLL cells using microarray technology.

RNA was purified from 31 N CD19⁺ B cells, 21 CD19⁺CD5⁺CD20^dimIgL-restricted MBL cells, and 65 CD5⁺CD19⁺ CLL cells. Microarray assays were performed using Illumina Human HT12 BeadChips. Genes differentially expressed between the 3 populations were identified (MBL vs N and MBL vs CLL) and sets of significant genes (≥1.5 fold change and P<0.01) were analyzed using Ingenuity Pathway Analysis (IPA).

Focusing on comparisons between MBL and N B cells, 1040 genes were higher and 868 lower in MBL than N B cells. Genes higher in MBL fall into different IPA categories including “PI3K/AKT Signaling” and “Inflammatory Disease”, thereby further underscoring a potential role of antigenic/inflammatory stimuli at the origin of MBL. In the “Inflammatory Disease” category, genes belonging to IFNα pathway were over-expressed in MBL. Eighteen of these IFN-associated genes overlapped with 74 genes in a type I IFN signature characteristic of systemic lupus erythematosus (SLE). Interestingly, 4 genes belonging to the “Post-Translational Modification” category that were more expressed in MBL - SPOP, SPK1, SRRM1, and ADAR (all P<0.0001) - have been directly linked to SLE. Finally, 21 genes in the Wnt/b-catenin pathway were also overexpressed in MBL compared with N B cells. Among these genes are WNT ligands (WNT3, 10A, and 5b), receptors (ROR1 and FZD3), effectors (TCF3, 4, and 25 and LEF1), targets (CCND1 and 3), inhibitors (GSKB, TLE4, PIN1, and AES) (all P<0.0001).

In contrast, the 683 genes more expressed in CLL than MBL B cells fall into the “Cell Death and Survival” and “Cancer” categories. Of note, several of the genes in the Wnt pathway over-expressed between MBL and N were lower in CLL B cells; however, since these comparisons were not paired (i.e., MBL clones that evolved to CLL were not studied), some of these differences may be spurious since we expect that not all MBL cases would evolve into CLL.

Our studies implicate several pathways in the development and evolution of MBL. First, an interferon pathway, similar to that activated in SLE, may be operational in MBL, hinting such signaling is involved in the amplification of MBL clones and associating CLL precursors with autoimmunity, a well documented fact for overt CLL cells. Second, over-representation of genes involved in RNA processing suggests this function is also important in MBL. This again associates MBL with CLL and autoimmunity as splicing factor mutations have been identified in CLL and the small nuclear ribonucleoproteins involved in splicing are often autoantibody targets in SLE. Of note, mutations that affect SF3B1 and its splicing function have been recently described in a subset of CLL patients and implicated in its pathogenesis. Finally, our data strongly incriminate the Wnt/b-catenin pathway in MBL. Since both Wnt ligand and receptor genes are upregulated, autocrine as well as paracrine loops may be operative. It is interesting that the WNT effector, LEF1, which has been reported as upregulated in MBL previously, is an interferon-responsive gene, possibly linking those two pathways. Finally, if not an artifact of sample availability, it is curious that most of the Wnt pathway genes diminish in expression at the level of manifest CLL. Additional studies are needed to determine the functional relevance of our observations to MBL and CLL B-cell biology.

No relevant conflicts of interest to declare.