The transcription factor Hepatic leukemia factor (HLF), initially identified as part of a leukemic fusion protein (E2A-HLF) causing B-lymphoid leukemia in children, is expressed in a wide range of tissues. In the mouse hematopoietic system, HLF is highly expressed in hematopoietic stem cells while being hardly detectable in more restricted progenitor populations. We have previously shown that in adult mice, HLF is essential for preserving the hematopoietic stem cell (HSC) pool during regeneration by maintaining their quiescence (Komorowska et al, Cell reports, 2017). Notably, despite this highly specific expression profile and a critical role in hematopoietic regeneration, mice with double knock-out of HLF alleles (KO) appeared to age normally (18-24 months survival > 90% for both WT and KO). In this study, we investigated the role of HLF in HSC during aging and leukemogenesis.

We first confirmed that HLF is expressed all through ontogeny by performing quantitative PCR on HSC and progenitor populations from mouse fetal liver (E14.5 FL) and bone marrow (BM) of aged mice (18-month-old mice). We found that not only is HLF expressed in hematopoietic cells from embryonic to aged mice, but also that the expression profile was highly conserved. In fact, we found that HLF expression decreased during hematopoietic differentiation independently of ontogenic age (HLF/HPRT relative expression: 6 in HSC (LSK-CD150+CD48-), 3 in HSPC (LSK), and <0,2 in CMP (Lin-kit+sca1-CD34+CD16/32-), GMP (Lin-kit+sca1-CD34+CD16/32+), and MEP (Lin-kit+sca1-CD34-CD16/32-). Additionally, hematopoietic parameters as well as lineage distribution in peripheral blood and bone marrow of aged KO mice were within normal range (at the exception of a 50% reduction in platelet blood count, also reported in adult mice). While we found no significant difference in the frequency of immunophenotypic HSC in total BM (LSK-CD34-Flt3-; WT: 0,0006 %, KO: 0,0004 %), aged KO grafts (200 000 total BM cells in 1:1 ratio with competitor) demonstrated a defective capacity to reconstitute primary recipients (donor contribution, WT: 60%, KO: 45 %). In secondary recipients, the defect was further aggravated (donor contribution; WT: 60 %, KO: 20%). Because neither SLAM nor CD34 markers can be used in KO FL, we quantified the percentage of EPCR+ cells within the LSK compartment as a surrogate marker for HSC and found a 2-fold decrease compared to WT controls. In competitive transplantations, KO E14.5 fetal liver cells demonstrated an impaired regenerative capacity in primary (donor contribution; WT: 88.8% (SEM 6.2), KO 63% SEM (16.2); n>5) and secondary recipients (donor contribution: WT: 67% (SEM 2.7), KO: 8.4% (SEM 3.6), n=4).

In addition to its function in healthy hematopoiesis, studies have suggested that HLF is a downstream mediator of the MLL/AF9 (MA9) leukemic fusion protein. To address the contribution of HLF to MA9 leukemogenesis, mice were transplanted with either KO- or WT-BM-derived c-Kit+ cells expressing the MA9 fusion protein. Preliminary results showed that HLF did not affect the latency of the leukemia in primary recipients with mice succumbing to a lethal disease at days 40-41 after transplant independently of the leukemic graft genetic background. However, KO leukemic cells had a significantly delayed leukemia development in secondary recipients (time to death; WT: 32-39 days, KO: 39-68 days). Moreover, semi-quantitative PCR showed that not only was HLF expressed in leukemia cells harvested from primary recipients but also that the expression level displayed a tendency to an increase in leukemia cells from secondary recipients. Altogether, our data suggest that while HLF may be dispensable for leukemic transformation, it promotes the development of MLL-AF9 leukemia in mice. We are currently in the process of confirming the role of HLF in MA9 leukemia development.

Taken together, we here show that HLF is essential for HSCs from embryonic to aged mice. We further show that HLF is critical for serial propagation of AML cells, suggesting that HLF is needed to maintain leukemia stem cells.


No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.