Abstract
B-cell precursor acute lymphoblastic leukemia (BCP-ALL) is accompanied by genomic mutations and rearrangements that commonly affect cytokines, transcription factors or signalling molecules that drive B-cell development or contribute to the pre-B cell receptor (pre-BCR) checkpoint. Deletions of the long arm of chromosome 6 [del(6q)] occur in ~10% of BCP-ALL and are also frequent in mature B and T-cell malignancies. Loss of function of the 6q genes EPHA7 and PRDM1, have been implicated in the genesis of lymphoma and BACH2, as a mediator of pre-BCR negative selection, is functionally a candidate tumour suppressor gene. However loss of these or other 6q genes have not been demonstrated, for example through biallelic inactivation, to contribute to BCP-ALL.
Analysis of our own and published SNP6.0 data from ALL patients defined 5 focal recurrent regions of deletion on 6q, 4 mapping to 6q15-6q21, coincident with previously published common regions of deletion in ALL. These 4 regions contain 22 candidate genes, including EPHA7 but not BACH2 or PRDM1, which nevertheless mapped close to focal deletions and were also classed as candidate tumour suppressors.
To develop the clone tracking assay, we adapted the SIN-SIEW lentiviral construct that expresses EGFP under the control of a spleen focus forming virus (SFFV) promoter. Candidate gene consensus coding sequence (CCDS) or a control luciferase cDNA were cloned between the promoter and an internal ribosomal entry site immediately upstream of EGFP. Transduction of the control (pSLIEW) or candidate gene SIN-SIEW-CCDS constructs consistently expressed EGFP in 697, a BCP-ALL cell line with del(6)(q14.1-22.3). For clone tracking, SIN-SIEW-CCDS constructs were assigned to 4 pools that also included pSLIEW. Pools were transduced into 697 cells that were both cultured in vitro and transplanted by intra-femoral injection into NOD/LtSz-scid IL2Rƴ null (NSG) mice. DNA was isolated from transduced cells immediately before transplant and then at 3 to 5 day intervals from cultured cells or from cells recovered from mouse bone marrow, spleen or liver at end stage disease. The pSLIEW construct facilitated monitoring of disease progression by in vivo imaging and also served as a control to measure CCDS construct copy number changes against. To quantify changes in integrated SIN-SIEW-CCDS, we developed a multiplex targeted Illumina sequencing approach.
In vitro, highly significant (p<0.01) reductions in copy number relative to pSLIEW over time, occurred for constructs expressing FOXO3, POU3F2, SIM1, PRDM13, C6orf168 and both α and β isoforms of PRDM1 (Fig 1a). With the exception of C6orf168, these genes also strongly suppressed leukemia development in vivo in all tissues analysed (Fig 1b). The known tumour suppressor genes, BACH2 and EPHA7, had no effect on cell growth in vitro. In vivo a moderate reduction for one of two EPHA7 CCDS was observed though curiously cells expressing BACH2 increased in relative copy number by approximately 3 fold. RNA sequencing data from 697 and published array data for normal pre-B cells and cases of BCP-ALL showed no, or extremely low, levels of expression for POU3F2, SIM1, PRDM13 and C6orf168 making it unlikely that they function as tumour suppressor genes in BCP-ALL. However significant expression of the transcription factors FOXO3 and PRDM1 were seen across data sets. Western blot confirmed expression of FOXO3 and PRDM1 in 697 and other BCP-ALL cell lines and demonstrated substantial increases in the corresponding proteins after transduction of 697 with FOXO3 and PRDM1 SIN-SIEW CCDS constructs. Over-expression of FOXO3 and both isoforms of PRDM1 decreased the proportion of cells in S and G2 phases of the cell cycle, but failed to induce apoptosis as measured by Annexin-5 staining. Comparison of total mRNA sequencing profiles of 697 cells, FACS sorted for ectopic expression of FOXO3, PRDM1 or control construct, showed distinctive patterns of up or down regulated mRNA. The roles of FOXO3 and PRDM1 in early B-cell development are currently undefined but notably our data suggests they influence expression of components of the pre-BCR and related signalling pathways and therefore may contribute to the pre-BCR checkpoint.
Disclosures
No relevant conflicts of interest to declare.
DNA hypermethylation contributes to colorectal cancer metastasis by regulating the binding of CEBPB and TFCP2 to the CPEB1 promoter
