Our studies have focused on characterizing a role for the lipid phosphatase inositol polyphosphate 4-phosphate Type II (INPP4B) in Acute Myeloid Leukemia (AML). Although INPP4B has been reported to antagonize the oncogenic PI3K pathway, and is frequently disrupted in several epithelial cancers, our recent studies provide new evidence for a paradoxical oncogenic role for INPP4B in AML. Our analysis of several AML patient databases led to the discovery that high levels of INPP4B expression were associated with poor clinical outcome, including significantly decreased survival metrics and poor response to chemotherapy. Frequency distribution analysis utilizing a fit of mixture model identified a population of high INPP4B expressers comprising 25% of patients across all the datasets. INPP4Bhigh patients had lower rates of CR compared with their INPP4Blow counterparts (57% vs 74%, respectively; P = 0.01). Accordingly, the mean expression of INPP4B was significantly lower in CR vs NR samples (61.8 ± 2.0 vs 84.9 ± 10.7, P = 0.003). Thus, we have characterized INPP4B as a useful clinical prognostic factor associating with unfavourable outcome in AML.

We have since established that INPP4B expression is significantly enriched in human and mouse hematopoietic stem cells (HSC), with an approximate 3-fold higher relative-expression in HSC and MPP compared to common myeloid precursors (CMP), granulocyte-macrophage progenitors (GMP) and multilymphoid progenitor (MLP). We demonstrate compelling evidence that INPP4B protein plays a critical functional role in both human and murine HSC maintenance, using in vivo models of leukemia and Inpp4b-knockout. Inpp4b-knockout mice show a disrupted bone-marrow stem-niche with reduced long-term HSC, HSC and GMP (49.1%, 32.4% and 19.7% lower than wild-type, respectively).

To further test the role of INPP4B in HSC we conducted functional assays in vivo. Competitive transplant of bone marrow into lethally-irradiated host mice shows that Inpp4b-knockout cells preferentially reconstitute myeloid blood lineages in the short term, with 78.5% of myeloid cells being Inpp4b-knockout at 16 weeks post-transplant of 1:1 ratio of Inpp4b-knockout to wild-type cells. Targeting proliferating cells with chemotherapeutic agents induces HSC cycling and differentiation to replenish the blood system, and has been used as an indirect measure of HSC functionality in vivo. We then injected mice weekly with 5'-fluorouracil, a non-mutagenic myelosuppressive agent that kills cycling hematopoietic cells, (150mg/kg every 7 days, i.p.) to model repeated hematopoietic injury. We observed a reduced overall survival in Inpp4b-knockout mice compared to wild-type controls, indicating an impaired ability of Inpp4b-knockout HSCs to repopulate the hematopoietic system post-insult. We generated MLL-AF9 leukemic mouse models in Inpp4b-wild type and Inpp4b -knockout bone marrow and determined that Inpp4b is not an absolute requirement for formation of leukemia.

In order to better understand the mechanism underlying the role of Inpp4b in hematopoiesis and leukemia we have an undertaken an unbiased -omics approach to identify key changes associated with Inpp4b-loss in stem and progenitor cells. We have performed quantitative-proteomics and next-generation sequencing on HSC, CMP, GMP and MEP populations. We have also investigated altered signalling dynamics associated with Inpp4b-loss- quantifying phospho-signalling changes in stem-enriched cell populations by means of high-dimensional single-cell mas-cytometry.

Based on our current data, we hypothesize that INPP4B plays a critical role in HSC maintenance and contributes to worse leukemia. We further hypothesise it is this stem-related function of INPP4B that is responsible for the poor therapy response observed in patients with elevated INPP4B. Given the aging population, and the association of AML and age on prognosis, our success has the potential to have an impact on an ever-growing number of patients.

Disclosures

No relevant conflicts of interest to declare.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution