Enhancer retargeting of CDX2 and UBTF::ATXN7L3 define a subtype of high-risk B-progenitor acute lymphoblastic leukemia

Abstract

Transcriptome sequencing has identified multiple subtypes of B-progenitor acute lymphoblastic leukemia (B-ALL) of prognostic significance, but a minority of cases lack a known genetic driver. Here, we used integrated whole-genome (WGS) and -transcriptome sequencing (RNA-seq), enhancer mapping, and chromatin topology analysis to identify previously unrecognized genomic drivers in B-ALL. Newly diagnosed (n = 3221) and relapsed (n = 177) B-ALL cases with tumor RNA-seq were studied. WGS was performed to detect mutations, structural variants, and copy number alterations. Integrated analysis of histone 3 lysine 27 acetylation and chromatin looping was performed using HiChIP. We identified a subset of 17 newly diagnosed and 5 relapsed B-ALL cases with a distinct gene expression profile and 2 universal and unique genomic alterations resulting from aberrant recombination-activating gene activation: a focal deletion downstream of PAN3 at 13q12.2 resulting in CDX2 deregulation by the PAN3 enhancer and a focal deletion of exons 18-21 of UBTF at 17q21.31 resulting in a chimeric fusion, UBTF::ATXN7L3. A subset of cases also had rearrangement and increased expression of the PAX5 gene, which is otherwise uncommon in B-ALL. Patients were more commonly female and young adult with median age 35 (range,12-70 years). The immunophenotype was characterized by CD10 negativity and immunoglobulin M positivity. Among 16 patients with known clinical response, 9 (56.3%) had high-risk features including relapse (n = 4) or minimal residual disease >1% at the end of remission induction (n = 5). CDX2-deregulated, UBTF::ATXN7L3 rearranged (CDX2/UBTF) B-ALL is a high-risk subtype of leukemia in young adults for which novel therapeutic approaches are required.

Graphical abstract

Figure 1.

Identification of a distinct subtype in B-ALL. (A) Gene expression profiling of 2004 B-ALL cases including 22 cases of CDX2/UBTF subtype (red) shown in a 2-dimensional tSNE plot. Each dot represents a sample. The top 1000 most variable genes (on the basis of median absolute deviation) were selected and processed by the tSNE algorithm with a perplexity score of 30. Major B-ALL subtypes are highlighted in different colors. (B) Differentially expressed genes in CDX2/UBTF cases compared with other B-ALL cases are shown in the volcano plot. Genes exclusively expressed in CDX2/UBTF B-ALL are colored in red and annotated. Significantly low expression of MME (CD10) and FLT3 in CDX2/UBTF B-ALL is also shown. Histone cluster genes are colored in orange showing upregulation in CDX2/UBTF B-ALL. (C) Immunophenotype of CD10 and cytoplasmic IgM in CDX2/UBTF B-ALL at diagnosis or relapse. Most cases are negative for CD10 and positive for IgM. (D) Expression of NTRK3 gene by normalized read counts comparing CDX2/UBTF and other B-ALL cases is shown. (E) Pathway analysis using top 216 differentially expressed genes (fold change >2, adjusted P < 1 × 10⁻³⁰) in CDX2/UBTF B-ALL revealed enrichment of histone and ribosomal RNA related pathways.

Figure 2.

Detection of novel UBTF::ATXN7L3 fusion gene. (A) Protein structures of wild-type UBTF, ATXN7L3, and UBTF::ATXN7L3 fusion (exon 17 type) are shown. UBTF exon 17 is rearranged to the 5′ untranslated region (UTR) of ATXN7L3 and encodes an in-frame chimeric fusion protein. Direct sequencing of a representative case (SJALL060150) is shown. (B) Immunoblotting showing wild-type UBTF (right second lane), ATXN7L3 (left third lane), and UBTF::ATXN7L3 fusion (exon 17 type, 4th lane) proteins. Protein lysate was collected from human embryonic kidney 293T cells with transient transfection of wild-type UBTF and ATXN7L3, and UBTF::ATXN7L3 fusion containing plasmids. All samples were run in the same gel. After transferring, the membrane was cut, probed with 2 different antibodies, and then the 2 halves scanned together. (C) The deleted UBTF regions detected by WGS are shown in black bars. (D) The schema of UBTF deletion and direct sequencing of a representative case (SJBALL020169) is shown. Nontemplated nucleotides are inserted at the breakpoint.

Figure 3.

Enhancer retargeting of FLT3-PAN3 regions drives deregulation of CDX2 in CDX2/UBTF subtype. (A) The deleted FLT3/PAN3 regions detected by WGS (type II deletion; top) and previously reported deleted 13q12.2 regions (type I deletion; bottom) are shown in black bars. Amplification of 13q12.2 found in Nalm-16 (hypodiploid B-ALL) is shown in a red bar. Type II deletion includes the promoter of FLT3 but not in type I deletion. Type II deletion does not affect the enhancer in PAN3. (B) Expression of FLT3 and CDX2 genes by normalized read counts comparing CDX2/UBTF and other B-ALL cases are shown. CDX2/UBTF B-ALL cases exhibited lower FLT3 expression and higher CDX2 expression. (C) Allele-specific expression of CDX2 is confirmed by comparison of allele frequencies at bases of heterozygous single-nucleotide polymorphism among RNA-seq (red), WGS (blue), and HiChIP (green). (D) H3K27ac HiChIP data are shown for representative patient samples (n = 3, SJALL060150 and SJBALL020169 [CDX2/UBTF B-ALL with type II deletion] and SJBALL113 [B-other with type I deletion]) and cell lines (n = 2, Nalm-16 [hypodiploid with PAN3 genic enhancer amplification] and Reh [ETV6::RUNX1 with no type I or type II deletion]). The heatmap shows raw read counts for all pairwise 5 kb genomic bins in the viewing window (chr13:28,457,037-28,911,626) for sample SJALL060150. The black box outlines read-pairs mapping to the CDX2 gene (highlighted in gold) and PAN3 genic enhancers (black dotted outline). Below the heatmap are RNA-seq and 1-dimensional HiChIP coverage tracks for all 5 samples. The corresponding heatmap for each sample can be found in supplemental Figure 8. Black bars indicate sample-specific deletions, whereas the Nalm-16 amplification is shown in red. Significant (false discovery rate < 0.01) loops called with MAPS are shown as blue arcs. Significant loops linking the PAN3 genic enhancers to CDX2 are shown in red.

Figure 4.

PAX5 deregulation in a subset of CDX2/UBTF subtype induced by the structural variants at PAX5-ZCCHC7 region. (A) The SVs at PAX5/ZCCHC7 region detected by WGS were shown in green (inversion) and blue (deletion) bars. Below the coverage tracks of 1-dimensional HiChIP and RNA-seq for representative CDX2/UBTF inversion case (SJALL060150_D) are shown. SVs include the promoter and exon 1 of PAX5 (exon 1a) but do not affect the alternative PAX5 exon 1b. (B) Expression of PAX5 by normalized read counts in each B-ALL subtype is shown. CDX2/UBTF cases with PAX5 rearrangements are shown in green (inversion) and blue (deletion). PAX5 is expressed highly in CDX2/UBTF subtype, especially for the case with PAX5/ZCCHC7 inversion. (C) Sashimi plot showing the junctions with the number of reads split across the junction (junction depth) in the PAX5 gene for representative CDX2/UBTF samples: inversion (n = 1, SJALL060150_D), deletion (n = 2, SJBALL016287_D1 and SJBALL020169_D1), and without PAX5 rearrangements (n = 2, SJBALL020165_D1 and SJALL0554518_D1). PAX5 exon 1b is shown in red bar. The ratio of exon 1b usage to exon 1a is shown in the right of plot. The cases with PAX5 rearrangements exhibited higher exon 1b usage ratio than cases without rearrangements.