WDR26 and MTF2 are therapeutic targets in multiple myeloma

Abstract

Unbiased genetic forward screening using retroviral insertional mutagenesis in a genetically engineered mouse model of human multiple myeloma may further our understanding of the genetic pathways that govern neoplastic plasma cell development. To evaluate this hypothesis, we performed a tumor induction study in MYC-transgenic mice infected as neonates with the Moloney-derived murine leukemia virus, MOL4070LTR. Next-generation DNA sequencing of proviral genomic integration sites yielded rank-ordered candidate tumor progression genes that accelerated plasma cell neoplasia in mice. Rigorous clinical and biological validation of these genes led to the discovery of two novel myeloma genes: WDR26 (WD repeat-containing protein 26) and MTF2 (metal response element binding transcription factor 2). WDR26, a core component of the carboxy-terminal to LisH (CTLH) complex, is overexpressed or mutated in solid cancers. MTF2, an ancillary subunit of the polycomb repressive complex 2 (PRC2), is a close functional relative of PHD finger protein 19 (PHF19) which is currently emerging as an important driver of myeloma. These findings underline the utility of genetic forward screens in mice for uncovering novel blood cancer genes and suggest that WDR26-CTLH and MTF2-PRC2 are promising molecular targets for new approaches to myeloma treatment and prevention.

Keywords: Carboxy-terminal to LisH (CTLH) complex; Forward genetic screen; Moloney murine leukemia virus; Plasma cell neoplasia; Polycomb repressive complex 2 (PRC2).

Fig. 1

Discovery of WDR26 and MTF2 in unbiased genetic forward screen using Myc-transgenic mice. a Schema of workflow that led to the nomination of WDR26 and MTF2 as candidate myeloma genes. Filters used to pare down the list of 100 input genes to 2 candidate genes are indicated on the right. b Accelerated tumor development in iMyc^ΔEµ gene-insertion mice treated with MOL4070LTR (mean tumor onset 178 ± 94 days; range 46–348 days) compared to mice not infected with virus (mean tumor onset 384 ± 86 days; range 245–505 days). Virus was injected IP (5 × 10⁴ colony forming units/10 µL) using a 30-gauge needle. c Tumor pattern in virus-infected mice from b. d Tissue section of plasmacytoma (left) and B lymphoma exhibiting plasmablastic differentiation (right) stained according to hematoxylin and eosin (H&E) and immunostained using antibody to CD138, respectively. e Ideogrammatic representation of mouse autosomes plus chromosome X indicating the genomic location of the top 100 candidate B cell and plasma cell tumor genes detected. Genes that passed Filters 1, 2 and 3 in a are labeled using orange, green and red dots, respectively. Thin red or blue lines denote whether cRIS mapping predicts increased or decreased gene expression due to proviral insertion. f Network of MYC-interacting proteins visualized by STRING (Search Tool for the Retrival of Interacting Genes). Proteins (n = 27) that interact with MYC directly or indirectly are depicted in red or blue, respectively. The minimum required interaction score was 0.5. Red and black lines within the network circle denote direct and indirect interactions with MYC, respectively. The symbols for WDR26 and MTF2 are enlarged for enhanced visibility. Network visualization relied on Cytoscape 3.8.2. g KEGG (Kyoto Encyclopedia of Genes and Genomes) pathway analysis of the top 100 candidate genes. Enrichment scores are denoted by ovals which indicate both the number of pathway genes involved (count) and level of statistical significance (blue saturation). h GO (Gene Ontology) term enrichment analysis of biological processes using the top 100 candidate genes from a as input. i Magnitude of RNAi-dependent knockdown (KD) of WDR26 and MTF2 expression in three different HMCLs relative to HMCLs transfected with scrambled message (control). j Programmed cell death in HMCLs exhibiting low WDR26 or MTF2 expression (KD) compared to HMCLs cells containing normal message levels (control). Gene knockdown relied on Sigma MISSION® Endoribonuclease-prepared siRNAs (esiRNAs, 50 ng), a heterogeneous mix of siRNAs that target the same message. k Growth inhibition (PrestoBlue™) of HMCLs harboring low levels of WDR26 or MTF2 message. Genes were knocked down using Mission EHU esiRNAs 150,671 (WDR26) and 042,951 (MTF2)

Fig. 2

Clinical and biological validation of WDR26 and MTF2 in myeloma. a WDR26 (top) and MTF2 (bottom) expression levels (GSE 2658 and 5900 datasets) in bone marrow plasma cells from healthy individuals (BMPC, n = 22) or patients with monoclonal gammopathy of undetermined significance (MGUS, n = 44), smoldering myeloma (SMM, n = 12) or frank myeloma (MM, n = 559). MM data are from GSE2658, all others from GSE5900. b Comparison of mean mRNA levels of WDR26 (top) or MTF2 (bottom) in patients with standard-risk myeloma (SR, n = 690) or high-risk myeloma (HR, n = 287), using data from the Multiple Myeloma DREAM Challenge study. c Overall survival (OS) of patients with myeloma in the MMRF CoMMpass study stratified according to WDR26 (top) or MTF2 (bottom) message levels in malignant plasma cells. The top quartile (n = 194, red) and bottom quartile (n = 194, blue) are compared. HR, hazard ratio. d Western analysis of WDR26 (left) or MTF2 (right) in normal (N) or gene-targeted (KO) myeloma cell lines, OPM2, H929 and MM1.S. KO protocols including gRNA sequences are available upon request. e Growth of HMCLs in bulk suspension culture. Cells deficient in WDR26 (blue) or MTF2 (red) are compared to parental cells (black) used as control. f Clonogenic growth of OPM2 (top), H929 (center) and MM1.S cells (bottom) lacking WDR26 (blue) or MTF2 (red) or containing the proteins (black). Representative images of soft-agar plates are shown to the left. The bar diagram to the right displays mean colony numbers ± SD based on three independent experiments. g Representative flow cytometric scatter plots of apoptotic death (red, labeled rectangles) of WDR26 or MTF2 deficient HMCLs compared to normal cells. h Mean values of apoptosis based on three independent measurements. Standard deviations are indicated by short vertical lines (***p < 0.001). i Bioluminescence images of NSG mice on days 10, 20, 30 and 40 following challenge with OPM2 cells (upper panel) or H929 cells (lower panel) deficient of WDR26 (center column) or MTF2 (right column). Parental cells proficient of these proteins served as control (left column). j Quantitative analysis of bioluminescence signal strength in mice from h. k Kaplan–Meier survival curves of mice depicted in h. Statistical comparison relied on log rank analysis. l Flow cytometry histogram distinguishing GFP-expressing tumor cells in the bone marrow of xenotransplanted mice from h (smaller peaks, right) from bone marrow cells not expressing GFP (larger peaks, left). m Abundance of GFP-expressing tumor cells in the bone marrow of mice from h