Gene expression Studies in WM. An important accomplishment in our investigation of WM has been our studies into the global analysis of gene expression in WM tumor cells, and their respective microenvironment. We analyzed the gene expression profiles (GEP) of CD19+ (which constitute the B-cell compartment) and CD138+ cells (which constitute the plasma cell compartment) obtained from bone marrow aspirates of untreated WM patients and compared their profiles to their normal counterparts from healthy donors (HD) using Affymetrix microarrays. By using analogous bone marrow tissues, we succeeded in performing the first reported direct gene expression comparison of WM and healthy donor CD19+, CD138+ and microenvironmental (CD19- and CD138- depleted) cells. Gene expression analysis was performed using dChip software. Supervised hierarchical cluster analysis for genes with > 2 fold change in expression and a False Discovery Rate (FDR) < 2% was performed and demonstrated a wide spectrum of differentially expressed genes between WM patient and HD (Figure 1).

Figure 1. A heat map generated from gene expression studies demonstrates genes which are up- (red) or down- regulated between healthy donor (HD) and WM patient (WM) bone marrow isolated CD19+ cells. Similar heat maps were generated for CD138+ isolated bone marrow cells.

Among the significantly over expressed genes in both CD19+ and CD138+ cell populations were: BCL-2, TACI, CD40, FLIP, cIAP-2, IGLL1, CCR2, CLLU1. Functional analyses undertaken with Ingenuity software showed that the most relevant biological pathways associated with these genes were found in the CD19+ compartment and included cell cycle and G1/S checkpoint regulators, as well as death receptors, ERK/MARK, SAPK/JNK and P53 signaling pathways, whereas in the CD138+ compartment, the most relevant pathways were those of the B cell receptor (BCR), glucocorticoid receptor and death receptor signaling. Interestingly, microenvironmental cells in WM patients demonstrated a unique transcriptional profile (Figure 2) including the Toll like receptors (TLR 1,5,7,8), interferon and cytokines (IFI16, IFNAR1, IL-10R, IL-8R), as well as genes encoding extracellular matrix components (Fibronectin and Hepatocyte Growth Factor).

Figure 2. A heat map generated from gene expression studies demonstrates genes which are up- (red) or down- regulated between healthy donor (HD) and WM patient (WM) bone marrow microenvironmental cells, showing that the bone marrow environment itself differs between HD and WM patients, and provides signals which could support WM cell growth.

These results may have important implications for the treatment of WM suggesting that inhibitors of BCL-2, of caspase activation, as well as inhibitors of proliferative pathways mediated by AKT and SAPK/JNK2 and MEK4 may have a therapeutic role for WM. The results of these studies were presented at the 5th International Workshop on WM (Hatjiharissi et al, www.wmworkshop.org).

microRNA studies in WM. In 2008, we began a large collaborative effort with the French IFM Group, and the Munshi Lab at Harvard Medical School to evaluate and compare microRNA expression in WM, multiple myeloma and healthy cells. MicroRNAs are small noncoding RNAs which regulate the expression of protein-coding genes by inducing cleavage of targeted transcripts or inhibiting translation. Given their emerging significance in the pathogenesis of other B-cell disorders, we evaluated expression of 384 microRNAs in tumor cells from WM patients and compared these to healthy donors (HD), as well as CD138+ cells from myeloma (MM) patients. MicroRNA profiling was performed in 13 WM and 79 MM patients, and 13 HDs. Data obtained from the microRNA arrays were analyzed using SDS, RQ manager, R and dChip softwares. MicroRNA profiling demonstrated significant upregulation of miRs -192, -125b, -21, -155, and downregulation of miRs-181c, -572, and -650 in WM patient derived bone marrow CD19+ cells in comparison to their healthy donor counterparts. Analysis of bone marrow derived CD138+ cells of WM patients was also compared to CD138+ cells from healthy donors and demonstrated differential expression of 40 microRNAs that included upregulation of miR-192, -193b, -17-3p, -585, -148b, and downregulation of miR-29c, -155, -126, -148a, -125a, -181d, -30a-3p, let-7b, let-7c, and others. Comparative analysis of WM CD138+ cells to MM CD138+ cells identified 17 microRNAs that were upregulated and 4 which were downregulated. Unsupervised hierarchical clustering of filtered microRNAs, based on their DCt values, identified two major groups within the WM and MM population (groups A and group B). Samples of Group A WM patients aligned with the microRNA profile of MM patients and appeared to have be associated with more aggressive disease, whereas Group B WM patients clustered with normal plasma cell microRNA profile, and were aligned with patients having a more indolent course. These results await further validation. Additionally, microRNA analysis from WM patients revealed alterations in critical signaling pathways which effect including apoptosis, hematopoietic cell differentiation and proliferation and survival through modulation of HOX, c-myc, and Bcl-2. The results of these studies were presented at the 5th International Workshop on WM (Adamia et al, www.wmworkshop.org).

Epigenetic modification in WM. In 2007, we initiated studies aimed at delineating epigenetic changes in WM with a potential role in effecting growth and survival pathways. Our initial studies focused on genes susceptible to hypermethylation and implicated previously in B-cell tumorigenesis. These studies identified that DLC-1, a tumor suppressor gene with a negative growth regulatory role which was previously shown to be hypermethylated and under-expressed in multiple myeloma patients was also a target of hypermethylation in WM. As shown in Figure 3, by MSP-PCR, DLC-1 was hypermethylated (partially and fully) in bone marrow derived CD19+ cells from 12/17 (71%) WM patients, but in none of 4 healthy donors, as well as in BCWM.1 (partially) and WM-WSU (fully) WM cells (Figure 4).

Figure 3. DNA methylation status of DLC-1 gene in bone marrow CD19+ cells from healthy donors (N1-4) and WM patients (P1-17) as determined by MSP-PCR.

Figure 4. DNA methylation status of DLC-1 gene in WM and MM cell lines as determined by MSP-PCR.

As predicted from methylation studies, mRNA expression of DLC-1 was significantly lower in WM patients compared to normal controls in two different experiments (Figure 5), as well as BCWM.1 cells, whereas in fully methylated WM-WSU cells, DLC-1 expression was absent (Figure 6).

Figure 5. mRNA expression levels of DLC-1 in healthy donor and WM patient CD19+ bone marrow cells.

Figure 6. mRNA expression levels of DLC-1 in WM and MM cell lines. In view of the above results, we evaluated the impact of hypomethylating agents on the growth and survival of BCWM.1 and WM-WSU WM cells, and examined the impact of these agents on expression levels of DLC-1. As show in Figure 7, the FDA approved hypomethylating agent 5 azacytidine (5-AzaC) induced expression of DLC-1 in a dose dependent manner in both BCWM.1 and WM-WSU WM cells.

Figure 7. mRNA expression levels of DLC-1 in BCWM.1 and WM-WSU WM cells co-cultured with 5-AzaC.

5-AzaC also exhibited significant dose-dependent cytotoxicity against the BCWM.1 WM cells. Treatment of BCWM1 cells with 2 uM of 5-AzaC rapidly induced cell cycle arrest at G1. More than 40% of BCWM1 cells underwent to apoptosis in 48 hours. 5-AzaC also induced significant apoptosis in primary samples of WM but no significant cytotoxic effect was seen in peripheral blood mononuclear cells from healthy donors. Cleavage of caspase 3, 7, 8 and 9 and PARP-1 were associated with 5-AzaC induced apoptosis, suggesting that both the mitochondrial and the death receptor pathways were involved in the apoptotic process. 5-AzaC also rapidly induced down-regulation of raptor which is a key component of the mammalian target of rapamycin complex 1 (mTORC1), and consequently inhibited phosphorylation of S6 kinase 1 (S6K1) which is a down-stream target of mTORC1 and an important regulator of protein translation. Rapamycin, a potent mTORC1 inhibitor, demonstrated additive inhibition of BCWM.1 cell growth when combined with 5-AzaC.

Taken together, the above data support a role for the investigation of hypomethylating agents in WM, and suggest multiple pathways of activity including reversal of hypermethylating effects on tumor suppressor genes as well as by inhibition of mTORC1 activity by down-regulating raptor in WM cells. 5-AzaC alone or combined with mTOR inhibitors (such as RAD001) represent a novel and potentially important platform for advancing the treatment of WM. The results of these studies were presented at the 5th International Workshop on WM (Xu et al, www.wmworkshop.org).