Abstract

Podocalyxin (PodxL) is a CD34 family member previously identified to mark hematopoietic stem cells (HSCs) and other progenitor cells. Previously, we discovered PodxL as a potent erythropoietin (EPO) response gene and demonstrated to promote egression of immature reticulocytes from bone marrow into circulation. PodxL is upregulated in several cancers, including myeloid and lymphoid leukemia. Herein, we aim to define the functional role of PodxL in hematopoiesis - specifically myelopoiesis - by employing conditional PodxL knock out (KO) mouse models. Hematopoietic-specific deletion was achieved using Cre mice with a Vav1 driver and myeloid-specific deletion was achieved with Lyzm2 - Cre driver. We confirmed the deletion of exons 3-7 at the gene, transcript and protein levels using PCR, RT-qPCR and western blotting, respectively. Peripheral blood analysis revealed no difference in blood cell lineages for either KO mouse strain. At steady state, colony forming unit-granulocyte/macrophage (CFU-GM) assay also showed no difference between the KO strains and wild type. In order to examine the functional role of PodxL during stress myelopoiesis, PodxL-/- ; Vav1-Cre mice were treated with 5-Fluorouracil (5FU), a chemotherapeutic agent induces myeloablation. Notably, during rebound of neutrophils, the PodxL-/- ; Vav1-Cre mice showed a sharp increase in neutrophil counts at day 12.5, which at later time points reverted to normal levels comparable to wild type mice. Previously, our in silico analyses combined with outcomes from truncated EpoR knock-in alleles had revealed that PodxL is a potential STAT5 transcriptional target. Here, we tested if G-CSF induces PodxL expression in hematopoietic progenitors. In vivo, G-CSF significantly induced PodxL expression four fold. We then tested the role of PodxL in G-CSF induced neutrophil formation in vivo. Both KO strains (Podxl-/-;Vav1-Cre and Podxl-/-;Lyzm2-Cre) and wild type were treated with G-CSF (125ug/kg/day) for 5 days. Peripheral blood analysis revealed increased neutrophil and monocyte levels in the PodxL-/-;Vav1-Cremice. In order to then determine a possible role of PodxL at the progenitor level, CFU-GM assays were performed. PodxL-/- ; lyzm2-Cre mice had increased colony forming capabilities but there was no difference in PodxL-/-;Vav1-Cre mice compared to wild type. Our results imply that PodxL is playing a negative regulatory role in stress myelopoiesis. Interestingly, the deletion of PodxL in hematopoietic progenitors (Vav1-Cre) resulted in enhanced migration of neutrophils, whereas deletion of PodxL in myeloid compartment (Lyzm2-Cre) resulted in decreased neutrophil migration. This may be in part due to a compensatory effect by CD34 in the hematopoietic compartment. To dissect the molecular mechanism of PodxL during stress myelopoiesis, upon in vivo G-CSF treatment, bone marrow derived hematopoietic progenitors were isolated and PodxL protein was immunoprecipitated. LC-MS/MS proteomic analysis was performed to identify the interacting partners with PodxL. Rap-1A, a small GTPase and member of the RAS family, was among the top interacting proteins. Rap-1A has been shown to promote adhesion and migration of myeloid cells. The association of PodxL with Rap-1A was further confirmed in hematopoietic progenitors by immunoprecipitation and western blotting. To determine if the interaction of PodxL directly regulates Rap-1A activity, a GTP-Rap-1A activity assay was performed in response to G-CSF, GM-CSF and IL-3. Rap-1A activity was significantly elevated in hematopoietic progenitors upon G-CSF treatment in PodxL-/-:Vav1-Cre mice compared to wild type, followed by IL3; however, GM-CSF did not affect Rap-1A activity. In conclusion, our results indicate an important functional role for PodxL in stress myelopoiesis, a function likely mediated via Rap-1A. A complete understanding of the PodxL/Rap-1A axis may reveal important molecular insights into G-CSF-induced mobilization of neutrophils and provide mechanistic understanding into the pathological role of PodxL in aggressive cancers, including leukemia, which in turn may facilitate identification of novel therapeutic targets in PodxL associated cancers.

Disclosures

No relevant conflicts of interest to declare.

Author notes

*

Asterisk with author names denotes non-ASH members.