The Neurexin I Alpha/Neurexophilin Axis as An Anti-Proliferative Factor In Human Cord Blood and Murine Bone Marrow Hematopoiesis.

Abstract 2627

The membrane-bound receptors Neurexin I alpha (NRXN1α) and Dystroglycan (DAG1) serve as ligands for each other, and the Neurexophilins (NXPHs) are thought to function as natural antagonists for this interaction. This signaling axis centered on NRXN1α, DAG1 and the three highly homologous NXPHs, plays a well-defined role in neuronal development and synapse development. The failure of DAG1 knock-out mice to develop a mesoderm led us to hypothesize that this axis plays a role in hematopoiesis. Consistent with previous research, DAG1 is heavily (>95%) expressed in human (hu) umbilical cord blood (CB) CD34+ cells, including the primitive CD38- subpopulation. NRXN1α is expressed at similar levels in these populations as well. Hematopoietic cell populations are more precisely defined in the murine (mu) system, and adult mu bone marrow (BM) exhibits a similar trend to hu CB with NRXN1α expression being higher in the more primitive cells and decreasing as the cells differentiate. A high concentration of Neurexophilins is present in the hu CB plasma (900ng/ml), though it is absent in adult mu blood plasma and fetal bovine serum. Neither blocking DAG1 with a well-described blocking antibody, nor stimulation with recombinant NRXN1α has any observable effect on stimulation of hematopoietic progenitor cell (HPC) proliferation in vitro. This led us to investigate the other receptor present in the axis, NRXN1α. We first used alpha-latrotoxin, an excitatory toxin for NRXN1α. While no stimulatory or inhibitory proliferation effect was observed under normal growth conditions, alpha-latrotoxin did promote HPC survival under growth factor starved conditions as effectively as SDF1α/CXCL12 (a known HPC survival enhancing factor), suggesting a possible protective role for the DAG1-NRXN1α axis during stress. In contrast, we identified the Neurexophilins as anti-proliferative agents for immature hematopoietic cells. When injected in vivo, recombinant NXPH1 acts in a myelo- and lympho-suppressive manner. Absolute numbers of mu HPCs were suppressed in a dose- and time-dependent manner, with the maximal dosage decreasing HPC numbers by over 50%. The decrease in absolute colony number was mirrored by a decrease in the cycling population of HPCs in mice treated with NXPH1, as measured by the high specificity activated tritiated thymidine kill assay. Peripheral blood neutrophils, lymphocytes, monocytes and platelets, were also decreased in a time-dependant fashion after in vivo exposure to NXPH1. Intriguingly, the most primitive LTR-HSCs behave in the opposite manner, with a three-fold increase in cycling status induced via the Rapamycin sensitive mTOR pathway. During in vitro culture NXPH1 is only able to suppress colony forming ability of unseparated HPCs treated with both GM-CSF and SCF in combination in comparison to stimulation by other cytokines. This response is most likely due to increased NRXN1α expression in response to stimulation by GM-CSF and SCF. Given the high level of DAG1 expression on colony forming hematopoietic cells, we hypothesized that endogenous DAG1 may be competing with NXPH1 for NRXN1α binding. We thus exposed cells to a well studied DAG1 blocking antibody or plated CD34+ cells in single cell culture. The in vitro results were then able to recapitulate the results observed in vivo and a 50% decrease in colony forming ability is observed in combination with the following cytokines: GM-CSF, IL-3, GM-CSF+SCF, and IL-3+SCF as well as under maximally stimulatory conditions (IL-3, GM-CSF, SCF, and Epo for hu cells and SCF, PWMSCM, and Epo for mu cell). The ability of NXPH1 to inhibit proliferation of CD34+ cells in single cell culture suggests that NXPH1 acts directly on HPCs. The anti-proliferative function of the NXPHs are consistent with previous literature correlating NXPH expression in cancers with negative prognoses and these data may provide a mechanism for the immune system evasion of cancer. Additionally, the high concentration of NXPH in CB plasma may help explain the relatively long time to engraftment of transplanted CB cells. More broadly, these results demonstrate the presence and function of a tightly regulated signaling axis centered on NRXN1α in hematopoiesis.