TEL-AML1 transgenic zebrafish model of precursor B cell acute lymphoblastic leukemia

Abstract

Acute lymphoblastic leukemia (ALL) is a clonal disease that evolves through the accrual of genetic rearrangements and/or mutations within the dominant clone. The TEL-AML1 (ETV6-RUNX1) fusion in precursor-B (pre-B) ALL is the most common genetic rearrangement in childhood cancer; however, the cellular origin and the molecular pathogenesis of TEL-AML1-induced leukemia have not been identified. To study the origin of TEL-AML1-induced ALL, we generated transgenic zebrafish expressing TEL-AML1 either ubiquitously or in lymphoid progenitors. TEL-AML1 expression in all lineages, but not lymphoid-restricted expression, led to progenitor cell expansion that evolved into oligoclonal B-lineage ALL in 3% of the transgenic zebrafish. This leukemia was transplantable to conditioned wild-type recipients. We demonstrate that TEL-AML1 induces a B cell differentiation arrest, and that leukemia development is associated with loss of TEL expression and elevated Bcl2/Bax ratio. The TEL-AML1 transgenic zebrafish models human pre-B ALL, identifies the molecular pathways associated with leukemia development, and serves as the foundation for subsequent genetic screens to identify modifiers and leukemia therapeutic targets.

Fig. 1.

TEL-AML1 transgenic zebrafish with ubiquitous and lymphoid-restricted expression. (A) Diagrams of the human TEL-AML1c (TA) cDNA, alone or fused in-frame to EGFP, expressed from the ubiquitous Xenopus elongation factor-1 (XEF) 0.7 kb, the zebrafish β-actin (ZBA) 4.5-kb promoters, or from the lymphoid zebrafish Recombination Activation Gene-2 (ZRAG2) 6.5-kb promoter. (B) Transgenic embryos (T) from the ZBA-EGFP-TA line expressing EGFP-TEL-AML1 at 3 dpf compared with nontransgenic (NT) sibling. Fish were oriented with anterior to the left and dorsal to the top (T) or dorsal to the bottom (NT). (C) Ventral view of a 7-dpf RAG2-EGFP-TA zebrafish with EGFP-TEL-AML1-labeled cells in the bilateral thymus (Th) (arrowheads). (Scale, 1 mm in B and 2 mm in C.)

Fig. 2.

Evidence of B-lymphoid differentiation arrest in TEL-AML1 transgenic zebrafish. (A) Lymphoid hyperplasia in TEL-AML1 transgenic zebrafish. Hematoxylin and eosin (H&E) staining of kidney marrow between the tubules (T) from wild-type (WT), and transgenic ZBA-EGFP-TA fish (TA) with lymphoid hyperplasia (×10 and ×63). Touch preps from transgenic fish stained with Giemsa (×63) show increased basophilic immature cells with nucleoli consistent with lymphoid progenitors (arrowheads), compared with wild-type. (B) In vitro, mitogen-induced B cell clonogenic assays indicate reduced colony numbers from TEL-AML1 (TA)-expressing XEF-TA transgenic fish, compared with wild-type (WT). Sorted EGFP-positive kidney progenitors from the ZBA-EGFP-TA transgenic fish developed significantly less EGFP-positive colonies (EGFP-TA) than from control fish (EGFP). May–Grunwald/Giemsa-stained B cell colonies from wild-type and transgenic cells, shown in low (×10) and high (×100) power. Colony cells from wild-type and TEL-AML1 transgenic cells (from two different colonies) expressed the constant region of IgM by RT-PCR.

Fig. 3.

Leukemic features of TEL-AML1 transgenic zebrafish. (A and B) H&E stain of sagittal sections from wild-type (A) and an F1 XEF-EGFP-TA transgenic zebrafish (B), with diffuse infiltrates of basophilic blast-like cells, most dense in the kidney region (arrows indicate head and tail kidney). Leukemic cells infiltrated distal organs including the brain (D), ovary (F), liver (H), and muscle (J), compared with wild-type sections from these organs (C, E, G, and I). Peripheral blood smear from wild-type fish (K) showing normal nucleated RBCs, lymphocytes, and a monocyte, whereas a leukemic blood smear (L) shows clusters of lymphoblasts. The kidney section revealed complete infiltration of the marrow between the tubules shown in low (N) and high (P) power, compared with wild-type (M and O). (Q and R) Leukemic cells express EGFP-TEL-AML1. (Q) No staining without the primary anti-EGFP antibody. (R) TEL-AML1 lymphoblasts showed strong nuclear and cytoplasmic staining with anti-EGFP antibody (arrows). (Scale bars, 3 mm in A and B; 100 μm in C–M and O; and 50 μm in N, P and Q and R.)

Fig. 4.

Molecular analysis of TEL-AML1 induced leukemia in transgenic zebrafish. (A) Southern blotting of RT-PCR products from five leukemia samples and positive control. (B) Semiquantitative RT-PCR showing the relative expression of each regulatory gene expressed as a ratio to β-actin to normalize the number of leukemic blasts. First, the expression range (dotted lines) in kidney marrow cells from wild-type and nonleukemic TEL-AML1 transgenic fish was established (C, control; n = 15). The expression levels from each of the five TEL-AML1 leukemic zebrafish are depicted (L, leukemias n = 5). (C) Down-regulation of zebrafish TEL transcripts in TEL-AML1 induced leukemia measured by quantitative one-step real-time RT-PCR. The value is presented as a ratio normalized to β-actin when a ratio of 1 represents the normalized expression ratio in wild-type. (D) Apoptotic signal changes in leukemic fish displayed as the mean (± SD) transcript copy number relative to β-actin. A survival determinant high Bcl2/Bax ratio was 7- to 8-fold higher than control in L2 (∗, P = 0.004), L3 (∗, P = 0.0043), and L5 (∗, P = 0.0063). Data represent three independent experiments done in triplicates.