Community Research and Development Information Service - CORDIS

Periodic Report Summary 1 - MODELING MLL-AF4 ALL (Cell of origin/Leukemia Initiating Cell in infant Pro-B MLL-AF4+ ALL)

The mixed lineage leukemia gene (MLL) is frequently rearranged in acute leukemias. So far, more than 80 MLL fusions have been described. Infant lymphoblastic leukemia (ALL) harbouring the t(4;11) translocation is associated with very brief latency (<1 year) and poor prognosis (20% 5-year OS). It has a low prevalence (1 newborn/1x106 cases) and presents a mixed pro-B ALL phenotype (CD19+CD10-CD15+CD56+). Several evidences (retrospective analyses of neonatal samples and twin concordance studies) demonstrate its prenatal origin during embryo/fetal development. However, there are still many open questions about this disease and the nature of the target cell for transformation. For example, it is unknown if MLL-AF4 is a one-hit driver leukemia or requires additional cooperating (epi)genetic mutations (Sanjuan-Pla, et al. Blood, 2015). The cytogenetic abnormality of this pro-B ALL is the balanced chromosomal translocation t(4;11)(q21;q23). This leads to MLL-AF4 and AF4-MLL fusion genes, with the first being considered as the leukemia initiating event. Recently, AF4-MLL was assumed to be causative for the disease phenotype. Retroviral expression of AF4-MLL into murine hematopoietic stem/progenitor cells (HSPC) followed by transplantation led to pro-B ALL or MLL when co-transduced HSPC expressing both fusion proteins were used (Bursen et al., Blood, 2010). However, AF4-MLL is only expressed in ca. 70% patients and cell lines showed addiction in vitro to MLL-AF4 but not AF4-MLL. Therefore, so far, we lack mouse/human models that faithfully recapitulate the disease latency and phenotype. This is in part because the cell in which MLL-AF4/AF4-MLL exerts its oncogenic effect is unknown. Our hypothesis is that to improve our understanding of this leukemia, we should define the nature/developmental context of the cell in which MLL-AF4 occurs. Therefore, we aimed to express MLL fusions in different ontogenic target cells. We proposed to:

1. To analyze if three potential blood developmental stages of HSCs, aorta-gonada-mesonephros (AGM), fetal liver (FL) and neonatal bone marrow (BM), influence the transforming capacity of MLL-AF4 or AF4-MLL.
2. To determine the Leukemia Initiating Cell (LIC) activity in vitro and in vivo.

Description of the work performed (1-2 years)

For aim 1, the following deliverables were planned. Tasks performed since the beginning of the project are indicated.

D1.1. Husbandry and breeding to expand mouse colonies. Wild-type C57Bl/6 (C45.2), B6SJL (CD45.1) and transgenic Vwf-EGFP mice were housed and expanded.

D.1.2. Dissection of embryonic tissues.

- E11.5 AGM region. Total unfractionated cells were prepared. AGM transplantation was optimized by comparing intravenous over intra-femoral administration. Given the low HSC content of non-explanted AGM, we tested that reasonable engraftment level of un-manipulated AGM cells could be achieved before lentiviral infection.

- E14.5 fetal liver. Fetal livers (FL) of littermates born from plugged females were dissected and pooled before isolating c-kit+ HSPC using MACS.

D.1.3.A. Optimization of FACS protocols to efficiently isolate HSC populations and lentiviral transduction efficiency. I started working with c-kit+ FL cells. I encountered several pitfalls:

a) production of low titer viral supernatants; b) low efficiency of lentiviral transduction; c) insufficient number of infected cells for functional studies. This caveat was anticipated in the proposal contingency plan because the fusion genes resulting from t(4;11) translocation are rather huge. They encode an open reading frame of 6.912 bp (MLL-AF4) and 8.739 bp (AF4-MLL). These cassettes were cloned into retro- (MSCV) or lentiviral vector systems (pRRL and pLVX). With pRRL constructs, <2% c-kit+ FL cells were generally infected. Troubleshooting consisted of testing: a) alternative infection protocols; b) different expression systems; c) chemicals to improve transduction efficiency (retronectin, polybrene, rapamycin, etc.); d) PEI vs calcium chloride method; e) homemade vs commercially prepared viruses. To bypass these pitfalls to express MLL-AF4 and AF4-MLL, I have undertaken a new approach: in vivo recreation of MLL rearrangement by gene targeting and by etoposide exposure. Rearrangaments involving the MLL gene have been reported to occur only in mice with Atm null background. Epidemiological and genetics studies have suggested that MLL-r many result from transplacental exposure to DNA topoisomerase-II (TOPO-II) inhibitors during gestation, such as chemotherapeutic agents o pesticides among others. However, exposure of TOPOII inhibitors is not sufficient per se for rearrangement of MLL and the genetic background, such as mutations in the DNA damage response may influence the likelihood of MLL-r.

D.1.3.B. Functional effects in vitro of MLL-AF4 or AF4-MLL in FL-HSCs. Since in our first attempts, the infection of AF4-MLL was very hard to achieve and yielded insufficient number of infected cells, we focused first on sorted MLL-AF4 infected cells for the in vitro studies. We performed colony assays, proliferation kinetics, apoptosis and cell cycle analysis.

D1.4. Functional effects in vivo (transplantation) of MLL-AF4 or AF4-MLL in FL-HSCs. The limitations described above impeded to perform the in vivo studies planned as part of Deliverable 1.4.

Potential impact and use of the final results

The in vitro studies of MLL-AF4 transduced c-kit+ FL cells are novel since this target cell was never tested in this pro-B ALL. Taking into account the in vitro data, no immortalization or proliferative advantage was observed. This suggests that: a) c-kit+ FL cells do not constitute the cell type of transformation; b) expression levels of the fusion gene are supra-physiologic; c) transforming ability is unique of the human context as recently proposed by J. Mullloy et al.; d) (epi) genetic oncogenic events are missing.

Due to the impossibility to perform in vivo experiments, to reliably answer if this system is capable of inducing leukemogenesis, we gave up the viral transductions. We focus on different murine models of infant MLL-r leukemia in order to: a) identify the cell of origin of infant MLL-ENL leukemia; b) compare the transformation vulnerability of hematopoietic cells at different ontogenic stages (AGM, FL and BM); c) develop a faithful model of infant MLL-ENL leukemia. First treatment of infant MLL-r leukemia is based on glucocorticoids (Interfant protocol) but drug resistance causes unfavorable evolution. Our model will be useful to explore the leukemogenic mechanisms in order to develop a rational basis for novel therapies. New therapies are in high demand since current treatment (chemotherapy, radiotherapy and BM transplantation) are extremely inefficient in these infants and associated with large toxicity and devastating side effects. Such a model could help to develop diagnostic approaches/drugs for early intervention. The project will impact on pediatric hemato/oncology: leukemia in infants is rare but generates tremendous interest due to its aggressive clinical presentation.

Reported by

Fundació Institut de Recerca Contra la Leucemia Josep Carreras


Life Sciences
Follow us on: RSS Facebook Twitter YouTube Managed by the EU Publications Office Top