Identifying Targets for Therapy in High Risk t(4;14) Myeloma Using Multi-Level Molecular and Phenotypic Analysis of Isogenic MMSET and MMSET Knock out Cell Lines

Introduction

In a subset of multiple myeloma (MM) patients the t(4;14) deregulates the histone methyltransferase, MMSET and growth factor receptor, FGFR3 and has been associated with poor prognosis. In some patients this has been ameliorated by the introduction of proteasome inhibitors but 50% of t(4;14) cases have a high-risk gene expression profile (GEP70) at presentation and derive less benefit from their use. The t(4;14), therefore, constitutes a very significant target for therapy. Despite international efforts at drug design it has proven difficult to target MMSET directly. Targeting FGFR3 has been attempted and may be effective when the gene is mutated, but this only occurs in a small proportion. We have used multi-level molecular (proteomic, gene expression, metabolomic) and phenotypic analysis to identify downstream molecules upregulated in t(4;14) MM as an alternative approach to defining novel targets for therapy.

Materials/Methods

The MM cell line KMS11 has t(4;14) which re-locates MMSET downstream of the immunoglobulin heavy chain gene enhancer. From KMS11 the isogenic paired cell lines non-translocated allele knock out (NTKO, retains high MMSET expression) and translocated allele knock out (TKO, low MMSET expression) have been derived using homologous recombination (Lauring et al, Blood 2008).

We performed Affymetrix GEP (HG_U133_plus_2) on mRNA extracted from KMS11, NTKO and TKO cells in triplicate and analysed using Partek®. Probes were included in the analysis if they had fold change (FC) >2 or <-2 (FDR <0.05). Mass spectrometry was performed on lysates from cells grown in media supplemented with heavy/light labeled amino acids. Proteins meeting quality metrics and FC >2 or <-2 were included. Pathway analysis was performed in WebGestalt.

A high throughput CellTitre-Blue® viability screen using 484 targeted anti-cancer compounds (80, 200, 800nM) was performed in KMS11vTKO cells (n=3), and the relative reduction in viability compared.

Results

GEP results from 1349 patients revealed increased expression of DSG2, BACE2, SULF2 and PTP4A3 along with MMSET and FGFR3 in the t(4;14) subgroup, suggesting novel targets for therapy.

GEP experiments in KMS11, NTKO and TKO cell lines showed increased expression of DSG2 and BACE2 genes in MMSET high vs low cells, indicating this occurs downstream of MMSET. In total 283 genes had altered expression common between MMSET high and low cell line comparisons (246 increased in MMSET high cells, 37 increased in MMSET low). Kegg pathways including cell adhesion molecules (7 genes, P =0.001), tight junctions (6, P =0.005) and apoptosis (5, P =0.007) were enriched.

SILAC mass spectrometry identified 58 proteins with altered expression common between MMSET high and low cell line comparisons (41 increased in MMSET high cells, 17 increased in MMSET low). Kegg pathways including steroid biosynthesis (2 proteins, P =0.0014) and metabolic pathways (8, P =0.0125) were enriched suggesting metabolomic analysis (in progress) may yield interesting results.

Combining mRNA and proteomic analysis, 13 genes/proteins were consistently altered in high vs low MMSET cell lines (12 increased in MMSET high). The most readily targetable of those upregulated is carbonic anhydrase 2, for which there are therapeutic agents available. Another, CD55, accelerates the decay of complement components. This hinders the formation of the membrane attack complex, required for the activity of antibody therapies, suggesting their efficacy may be impaired in this subgroup. 5/12 of those upregulated were associated with the actin cytoskeleton highlighting another novel target for therapy.

Compounds that resulted in preferential reduction in cell viability in MMSET high cells included histone deacetylase inhibitors. MMSET has been shown to interact with HDAC1 suggesting a mechanism by which this selectivity might be mediated. Other classes associated with selectivity included PI3K/mTOR and topoisomerase inhibitors.

Conclusions

Using a multi-modality approach we have identified a number of potential targets in t(4;14) MM. We have also identified a potential biomarker of antibody-targeted therapy resistance in this subgroup, suggesting molecular subgroup analysis of clinical trials using these agents will be important.

Our targets will be further validated as potential avenues for therapeutic intervention with the aim of improving outcomes in this poor prognostic group.