Muscarinic Receptor Type-1 Regulates Centrally Hematopoietic Stem Cell Mobilization By Granulocyte-Colony Stimulating Factor Via the Hypothalamic-Pituitary Axis

Hematopoietic stem cells (HSCs) give rise to the cellular constituents of the blood throughout the lifetime. Transplantation of healthy HSCs is the only curative therapy for many hematologic disorders. Collection of these cells is most commonly achieved by administration of granulocyte colony stimulating factor (G-CSF), which mobilizes HSCs from specialized niches within the bone marrow (BM) to peripheral blood for collection. The mechanisms behind G-CSF-induced mobilization remains incompletely understood and may have broad implications about how healthy or cancer cells migrate. We have previously identified a role of signals from the sympathetic nervous system in regulating HSC mobilization (Katayama et al. Cell, 2006) and more recent studies have revealed cholinergic nerves from the parasympathetic nervous system regulated prostate cancer metastasis via the muscarinic receptor type-1 (Chrm1; Magnon et al. Science, 2013). This led us to explore whether Chrm1 could also regulate the migration of HSCs during enforced mobilization. Although steady-state HSC numbers are not altered in the BM of Chrm1^−/− mice, G-CSF-induced mobilization of HSPCs to the blood was reduced by ~54% (p=0.008) compared to wild-type (WT) mice. Competitive transplantation of WT and Chrm1^−/− blood revealed reduced long-term (16 weeks) repopulating capacity in mutant mice, confirming that the mobilization of functional HSCs was impaired. Although Chrm1 was robustly expressed in the prostate, we did not detect Chrm1 expression by qPCR in HSCs or sorted putative cellular niche constituents (Nestin⁺ cells, osteoblasts, endothelial cells, and macrophages), suggesting that cholinergic activity was located outside the BM. We thus tested the hypothesis that Chrm1 expression in the brain alters HSC behavior in the BM. To this end, we inhibited Chrm1 with pirenzepine, a specific antagonist that does not cross the blood-brain barrier. Whereas intra-peritoneal pirenzepine administration did not affect HSC mobilization, intra-cerebral pirenzepine administration significantly reduced mobilization (by ~45%, p=0.003), to a similar extent as the knockout mouse. These results suggest that cholinergic signals from the central nervous system (CNS), rather than the peripheral nervous system were required for robust mobilization of HSCs. To investigate the mechanisms by which CNS signals were relayed to HSCs within the BM, we generated chimeric parabions in which WT and Chrm1^−/− mice were surgically joined and subsequently mobilized with G-CSF. We found that a paired WT rescued the mobilization defect of Chrm1^−/− mice, suggesting that the CNS to BM relay was mediated by a circulating factor. Chrm1 was highly expressed in the hypothalamus of WT brains by qPCR and immunofluorescence. Chemical denervation of the hypothalamus by neonatal monosodium glutamate treatment reduced HSC mobilization (by ~32%, p=0.003), identifying a role for hypothalamic signals in G-CSF mobilization of HSCs. Since the hypothalamus regulates pituitary hormone release, we evaluated their levels. We found that plasma ACTH levels were modestly reduced, combined with a significant reduction in circulating corticosterone (by ~50%, p=0.002) and a dramatic reduction of corticosterone in BM extracellular fluid (by ~98%, p=0.007), suggesting a deficiency in glucocorticoids (GC). Treatment of WT mice with metyrapone, an inhibitor of GC synthesis, significantly reduced HSC mobilization (by ~25%, p<0.05). Furthermore, GC administration rescued the HSC mobilization deficit of Chrm1^−/− mice. However, administration of GC above the physiological levels inhibited G-CSF mobilization efficiency. Immunofluorescence analyses revealed reductions in nuclear translocation of the GC receptor (Nr3c1) in HSCs derived from Chrm1^−/− mice, a hallmark of nuclear receptor activity. In addition, GC enhanced the migration (by ~2 fold, p<0.05) of WT colony forming progenitors in vitro, suggesting cell-autonomous signaling of GC with their receptor may be necessary during G-CSF mobilization. Ongoing studies are evaluating whether the niche or hematopoietic cells requires GC receptors using Nr3c1-floxed mice. These data are to our knowledge the first direct demonstration of CNS regulation of HSC mobilization and uncover a critical function for physiological levels of GC in enabling HSC migration.