RUNX1 and NF-E2 upregulation is not specific for MPNs, but is seen in polycythemic disorders with augmented HIF signaling

Transcription factor runt-related transcription factor 1 (RUNX1, also known as AML1) is the principal regulator of mammalian hematopoiesis. Aberrant RUNX1 expression in the hematopoietic lineage, generated by multiple mechanisms (translocations, gain-of-function mutations, and gene amplification), is thought to be causative of leukemic transformation.¹  However, mutations of RUNX1 are rare in chronic myeloproliferative neoplasms (MPNs).² 

Transcription factor nuclear factor, erythroid-derived 2 (NF-E2) is a target of RUNX1 that is essential for the regulation of erythroid and megakaryocytic maturation and differentiation and expression of globin genes.³ 

It has been reported that increased RUNX1 expression in granulocytes is present in all 3 classical MPNs, that is, polycythemia vera (PV), essential thrombocythemia, and primary myelofibrosis. It has been suggested to be specific for MPN,⁴  and that elevated NF-E2 promotes erythropoietin (EPO)-independent erythroid maturation of PV hematopoietic stem cells in vitro.^5,6  A mouse model overexpressing the NF-E2 transgene in hematopoietic cells was reported to be a new model of MPN.⁷ 

Polycythemic states can be divided into primary polycythemias, characterized by intrinsically hyperproliferative erythroid progenitors that are hypersensitive to EPO, and secondary polycythemias, wherein erythroid progenitors respond normally to EPO but circulating EPO is elevated or inappropriately normal for the level of increased red cell mass.^8,9  Examples of primary polycythemias are PV, gain-of-function mutations of the EPO receptor causing a phenotype of primary familial and congenital polycythemia (PFCP), and some congenital disorders of hypoxia sensing that may share features of both primary and secondary polycythemias, as exemplified by Chuvash polycythemia.¹⁰  In this report, we examined the possibility that increased transcripts of RUNX1 and NF-E2 may also be present in other primary polycythemic states.

We prospectively recruited 26 subjects with various primary and secondary polycythemias (Table 1) using approved University of Utah (23 subjects) and Palacky University Hospital (3 subjects) Institutional Review Board informed consent in accordance with the Declaration of Helsinki. Patients’ granulocytes and mononuclear cells were separated from peripheral blood, as previously described.¹¹  Mouse embryos and yolk sacs were analyzed as described.¹² 

Mononuclear cells were isolated from peripheral blood and subjected to in vitro colony-forming assay, as previously described.¹³  Erythroid burst-forming unit colonies (BFU-Es) were scored by standard morphologic criteria.

Total RNA was isolated using TRI-reagent (Molecular Research Center, Cincinnati, OH) and treated with DNA-free DNase Treatment and Removal Reagents (Ambion, Life Technologies, NY). DNA-free RNA was reverse-transcribed using a SuperScript VILO cDNA Synthesis Kit (Invitrogen/ Life Technologies, NY) according to the manufacturer’s instructions and used for quantitative real-time polymerase chain reaction as described.¹⁶ 

The phenotypes and causative mutations of 26 polycythemic patients are depicted in Table 1. All primary polycythemic patients had erythroid progenitor hypersensitivity to or independent of EPO (Figure 1A); all secondary polycythemic subjects had normal BFU-E EPO sensitivity (data not shown). To assess whether the putative mechanism underlying the intrinsic hypersensitivity of erythroid progenitors to EPO in PV^5,6  is unique to PV or shared with other polycythemia states, we analyzed RUNX1 and NF-E2 expression in hypersensitive BFU-Es.

All examined PV patients, polycythemia patients with defects of hypoxia sensing (2 unrelated subjects with Chuvash polycythemia, 1 polycythemic patient homozygous for the VHL^P138L mutation,¹⁶  and 1 patient with the HIF2A^M535V gain-of-function mutation¹⁸ ), and 1 patient with a heterozygous single-nucleotide polymorphism in the LNK gene (rs147341899; LNK^I257T) had elevated RUNX1 and NF-E2 gene transcripts in their BFU-Es (Figure 1A). RUNX1 and NF-E2 gene transcripts were also increased in granulocytes of these patients and in granulocytes of another gain-of-function HIF2A mutant (HIF2A^G537R) patient¹⁸  from whom RNA from BFU-Es was unavailable (Figure 1B).

We tested whether increased RUNX1 and NF-E2 gene transcripts characterize all primary polycythemias. PFCP-derived BFU-Es and/or granulocytes did not have increased levels of these transcripts in cells with the EPOR^Q434X mutation¹⁴  nor in BFU-Es from PFCP patients with EPOR^5967insT mutation¹⁵  (granulocytes were unavailable) (Figure 1A-B).

We next examined granulocytes from 2 Croatian polycythemic patients with a homozygous VHL^H191D exon 3 gene mutation whose erythroid progenitors were not hypersensitive to EPO¹⁹  and found RUNX1 transcripts, but not NF-E2 transcripts, increased (Figure 1B). We observed similar results in 2 compound heterozygotes for VHL^T124A and VHL^L188V mutations. These 2 polycythemic siblings had hypersensitive erythroid colonies¹⁷  and increased RUNX1, but not NF-E2, transcripts in granulocytes (Figure 1B). RNA from their BFU-Es was unavailable for testing.

All 6 unrelated subjects with secondary polycythemia had normal RUNX1 and NF-E2 gene transcripts in granulocytes (Figure 1B).

We then examined granulocyte transcripts of the hypoxia-inducible factor (HIF)-regulated genes TFRC, SLC2A1, HK1, PDK1, VEGF, and BNIP3 and found them to be increased in all PV patients and all studied polycythemic patients with increased RUNX1 gene transcripts, but not in polycythemic EPOR^Q434X patients or 6 patients with secondary polycythemia (Figure 1B). EPOR^5967insT granulocytes were unavailable.

To further support our hypothesis that HIF signaling regulates RUNX1 expression, we analyzed Hif1α^−/− whole mouse embryo (Hif1α deficiency is embryonic lethal by day 11) and hematopoietic tissue, ie, murine Hif1α^−/− yolk sac (E9.5).¹²  The Runx1 transcript was down-regulated (Figure 1Bx), confirming that HIF directly or indirectly regulates RUNX1.

In summary, we found increased expression of RUNX1 in all patients with augmented HIF signaling, including PV patients. Further, in some, but not all, patients with augmented HIF signaling, we also detected elevated NF-E2 transcripts. Hypersensitive erythroid progenitors derived from patients with PV and augmented HIF signaling, but not PFCP, share elevated expression of RUNX1 and NF-E2. This suggests that HIF-mediated mechanisms, by which erythroid progenitors in PV and polycythemias with augmented hypoxia sensing, may exert their EPO hypersensitivity by up-regulation of RUNX1 and NF-E2. However, this is not a mechanism of erythroid EPO hypersensitivity in PFCP. The augmented HIF signaling in PV has not been well described; however, in our preliminary report, we observed increased HIF activity, the so-called “Warburg effect,” in PV patients.²¹  We conclude that increased expression of RUNX1 and NF-E2 is not specific for PV and other MPNs and is not universal for all primary polycythemic disorders such as PFCP, but it is present in those primary polycythemias with augmented HIF signaling.

This work was supported by grant 1P01CA108671-O1A2 (National Cancer Institute, Bethesda, MD) awarded to the Myeloproliferative Disorders Consortium (Principal Investigator, Ron Hoffman) Project 1, Leukemia and Lymphoma Society (Principal Investigator, J.T. Prchal), Education for Competitiveness Operational Programme projects CZ.1.07/2.3.00/20.0164 (K.K.) and CZ.1.07/2.3.00/30.0041 (L.L.), Czech Science Foundation Project P305/11/1745 (Principal Investigator, V. Divoky), and Palacky University (LF_2013_010).

Contribution: K.K. performed the research, analyzed data, and wrote the paper; L.L. performed some research, analyzed data, and wrote the paper; F.L. performed some research and reviewed the paper; J.S. prepared and purified RNA from Hif1α^−/− mice; M.H. contributed to the research and reviewed the paper; V.D. wrote the paper and provided financial support; and J.T.P. conceived the study, wrote the paper, and provided financial support.

Conflict-of-interest disclosure: The authors declare no competing financial interests.

Correspondence: Josef T. Prchal, Division of Hematology, 30 N 1900 E, 5C402 SOM, University of Utah, Salt Lake City, UT 84132; e-mail: josef.prchal@hsc.utah.edu.