Severe congenital neutropenia (CN) is a pre-leukemic bone marrow failure syndrome. Recently we reported a high frequency of cooperating RUNX1 and CSF3R mutations in CN patients that developed AML or MDS. Only a combination of these two mutations induced elevated proliferation and diminished myeloid differentiation of CD34+ cells in vitro.

To confirm these clinical data in an in vitro model, we generated human induced pluripotent stem cells (hiPSCs) from PBMNCs of a CN patient harbouring p.C151Y mutation in ELANE after acquisition of AML. During GCSF treatment, this patient acquired G-CSFRmutation p.Q741*, which leads to a truncated G-CSF receptor and was detected six years prior to overt AML. Three years later, he acquired an additional RUNX1 (p.R139G) mutation, which is located in the RUNT-homology domain (RHD). Subsequently, he developed AML (FAB M1) with trisomy 21. Reprogramming of PBMNCs isolated from the time-point of AML (ca. 80 % of AML blasts) resulted in the generation of hiPSCs clones harbouring either only ELANE p.C151Y mutation (CN-iPSC clone, derived from non-leukemia PBMNCs) or additional CSF3R and RUNX1 mutations and trisomy 21 (CN/AML-iPSC clone, derived from AML blasts), which was subsequently validated by Sanger sequencing and by digital PCR.

These iPSCs clones have been tested for their pluripotency and self-renewal capacity. Both iPSC clones expressed the pluripotent stem cell surface markers SSEA-4 and TRA-1-60 and displayed alkaline phosphatase activity. Further they highly expressed mRNA of the pluripotent stem cell markers SOX2, ABCG2, DNMT and NANOG and were able to differentiate into all three germ layers (meso-, endo- and ectoderm).

Embryoid body (EB)-based hematopoietic / neutrophilic differentiation of CN-iPS clones using serum-free APEL stem cell differentiation medium showed comparable amounts of CD34+ and CD33+ cells, but ~ 2-fold reduction of CD16+ cells, compared to healthy donor (HD) iPSCs. CN/AML-iPSCs were not able to differentiate into mature granulocytes at all and revealed 10-fold reduced counts of CD34+ and CD33+hematopoietic cells. Morphological examinations of Giemsa-stained cytospin slides confirmed these results. Additionally, CN/AML-iPSCs showed a highly reduced number of CFU-G and CFU-GM colonies in CFU-Assay.

To investigate the intracellular mechanisms of leukemogenic transformation in CN, we analyzed gene expression profiles of hematopoietic cells generated from CN-iPSCs vs CN/AML-iPSCs and HD-iPSCs for various time points of differentiation in our EB based-system. Our previous microarray-based analysis of bone marrow CD33+ cells of this CN/AML patient revealed that genes overexpressed in early hematopoietic stem/progenitor cells (HSPCs) as compared to more mature progenitors, such as DNTT, BAALC, CD34, HPGDS, NPR3 and PROM1 were strongly upregulated in CN/AML blasts harbouring both RUNX1 and CSF3R mutations, as compared to the cells prior to leukemia development. Intriguingly, elevated expression of these genes was described previously in RUNX1-mutated de novo AML blasts (Mendler et al., JCO 2012). This genetic signature suggests transformation of hematopoietic progenitors carrying mutated CSF3R into more primitive hematopoietic progenitors after aquisition of RUNX1mutation. We were able to confirm markedly increase of mRNA levels of these genes in hematopoietic cells derived from CN/AML-iPSCs, as compared to CN-iPSCs.

In addition, we found that hematopoietic cells of both CN-iPSCs and CN/AML-iPSCs revealed increased expression of unfolded-protein response (UPR) genes DDIT3 (CHOP), ATF4 and ATF6, as compared to HD-iPSCs. Activation of UPR in hematopoietic cells of CN-ELANEpatients has been previously described by our and other groups. CN/AML-iPSC-derived hematopoietic progenitor cells expressed RUNX1 mRNA at least two-fold higher, as compared to HD- or CN-iPSC-derived cells.

In summary, we established an in vitro cellular model of leukemogenic transformation in CN patients using CN/AML-patient derived hiPSCs that confirmed clinical data of Skokowa et al. (Blood 123:2550, 2014) on a cooperative leukemogenic effect of CSF3R and RUNX1 mutations. Comprehensive analysis of hematopoiesis using this iPSCs model will give us a deeper view into this highly complex signaling network operating during leukemogenic transformation of HSCs in pre-leukemic bone marrow failure syndromes.

Disclosures

No relevant conflicts of interest to declare.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution