Calcium and Calcineurin-NFAT Signaling Regulate Granulocyte-Monocyte Progenitor Cell Cycle via Flt3-L

Abstract Maintenance of myeloid progenitor cells is controlled by complex regulatory mechanisms and is orchestrated by multiple different transcription factors. Here, we report that the activation of the transcription factor nuclear factor of activated T cells (NFAT) by calcium-sensing protein calcineurin inhibits the proliferation of myeloid granulocyte–monocyte progenitors (GMPs). Myeloid progenitor subtypes exhibit variable sensitivity to induced Ca2+ entry and consequently display differential engagement of the calcineurin-NFAT pathway. This study shows that inhibition of the calcineurin-NFAT pathway enhances the proliferation of GMPs both in vitro and in vivo and demonstrates that calcineurin-NFAT signaling in GMPs is initiated by Flt3-L. Inhibition of the calcineurin-NFAT pathway modified expression of the cell cycle regulation genes Cdk4, Cdk6, and Cdkn1a (p21), thus enabling rapid cell cycle progression specifically in GMPs. NFAT inhibitor drugs are extensively used in the clinic to restrict the pathological activation of lymphoid cells, and our data reveal for the first time that these therapies also exert potent effects on maintenance of the myeloid cell compartment through specific regulation of GMP proliferation. Stem Cells 2014;32:3232–3244


INTRODUCTION
Regulation of myeloid hematopoiesis plays a key role in the maintenance of innate immune responses. The nuclear factor of activated T cells (NFAT) family of transcription factors has been recently been identified as an important player in the renewal of various myeloid cell subsets [1,2]. The NFAT family has five members, of which the activation cascade of NFAT1-4 is driven by increased levels of intracellular Ca 21 . Ca 21 is sensed by calmodulin, which activates calcineurin-mediated dephosphorylation of NFAT resulting in translocation of NFAT into the nucleus [3][4][5]. Apart from its crucial and well-described role in embryogenesis [6,7] and T cells [3], the calcineurin-NFAT pathway controls several innate immune functions of dendritic cells (DCs), macrophages, mast cells, megakaryocytes, and osteoclasts [2,8]. NFAT signaling also regulates apoptosis of terminally differentiated DCs [9], further promoting maintenance of the steady-state. In contrast, during infection, the calcineurin-NFAT pathway is required for effective neutrophil responses to Candida [10] and effective macrophages responses to Leishmania [11].
As well as being involved in myeloid cell functions, NFAT now appears to be important for myeloid compartment development [1] and megakaryopoiesis [12]. Although myeloid cells are present in mice lacking calcineurin-NFAT signaling [13], NFAT deficiency leads to progressive abnormalities including extramedullary hematopoiesis in the spleen and reduced numbers of hematopoietic stem cells (HSCs) in bone marrow (BM) [14]. In humans, there is parallel evidence for involvement of NFAT in the differentiation of immunomobilized CD34 1 HSCs [15]. However, in each case, the mechanisms underlying these effects of NFAT are unknown. Myeloid hematopoiesis proceeds from HSCs, through multipotent progenitors (MPPs), common myeloid progenitors (CMPs), and finally to GMPs that give rise to fully committed myeloid cells [16]. Within the myeloid lineage, NFAT negatively regulates the differentiation of megakaryocytes [12,17] and induces development of osteoclasts [18]. Genes regulating the cell cycle have been identified as NFAT targets in T cells, [19] stem cells [20], and in embryonic development and lineage specification [21], but not in the myeloid compartment. Indeed, the main networks regulating proliferation and differentiation in hematopoietic progenitors have also been identified [22], but a role for NFAT in the maintenance of myeloid progenitor cells has not previously been reported.
The hematopoietic process is controlled by growth factors and cytokines, including SCF, IL-3, and IL-6 [23], and specifically, in the case of myeloid cells by G-CSF, M-CSF, GM-CSF [24,25], and Flt3-L [16,26,27]. Interestingly, SCF, IL-3, IL-6, and GM-CSF signaling all increase the levels of intracellular Ca 21 in HSCs [28][29][30]. Furthermore, IL-3 and GM-CSF signaling are associated with phospholipase Cc (PLCc 2 ), the main driver of increases in intracellular Ca 21 levels [30], and M-CSF [25] and G-CSF [31] phosphorylate PLCc 2 in BM progenitors, while also being critical determinants of cell lineage commitment. Whether induction of Ca 21 release in any of these instances results in NFAT activation is unknown, as are the potential downstream effects on cell function. It has been shown that Flt3 ligation activates PLCc 2 [26,32], but the association with Ca 21 release, calcineurin and NFAT translocation are currently unknown. The possible link between Flt3/Flt3-L signaling and NFAT induction is particularly intriguing since Flt3-L is a key growth factor for hematopoietic progenitors and also initiates the main signaling pathway responsible for in vivo steadystate differentiation of DCs [33][34][35][36][37]. Flt3 is expressed on MPPs, CMPs, and GMPs [16,36,37], while also sustaining progenitor expansion [38,39], and promoting the growth of colony-forming units (CFU) [40]. Flt3-L is a key cytokine responsible for both development of myeloid cells [33,34,41] and promotion of inflammatory immune responses [42]. Furthermore, expression of Flt3-L has been reported on GMPs [36,37] and Flt3 signaling is known to directly impact on GMP development [43]; however, the mechanism underlying this process remains poorly defined.
Here, we report that NFAT is both present and functional within myeloid progenitors, and directly inhibits the proliferation of GMPs. In addition, we reveal that NFAT mobilization can be triggered by Flt3-L signaling specifically in GMPs, thus providing compelling evidence of a role for NFAT in myeloid hematopoiesis, which has direct implications for the therapeutic inhibition of NFAT in the clinic.

Animals
C57BL/6, C57BL/6-ppp3r1 tm1stl /J-(Cnb1 flox ), Mx1-cre, and CD45.1 mice were from The Jackson Laboratory and were maintained under specific pathogen-free conditions. Cyclosporine A (CsA) (4 mg) or FK506 (0.4 mg) were injected intraperitoneally. C57BL/6(B6)-CD45.1/CD45.2 heterozygote recipients were lethally irradiated with 12Gy in two doses separated by 4 hours and reconstituted with five million BM cells consisting of a mixture of 50% Cnb1 flox/flox Mx1-cre and 50% age-and sex-matched CD45.1. Cells from mice were analyzed 8 weeks after reconstitution. To induce Cnb1-knock out (KO) in Cnb1 flox/flox Mx1-cre mice, animals were treated with five intraperitoneal injections of 250 lg of polyI:C (Invivogen) every other day for 10 days. All mice were bred in the Biological Resource Centre (A*STAR, Singapore) and handled according to institute guidelines under the approval of the Institutional Animal Care and Use Committee.

PLCc Phosphorylation Analysis by Flow Cytometry
Lineage-negative cells were cultured for 4 hours and then stimulated for 2 or 10 minutes with rmFlt3-L (1lg/ml). Progenitor subsets were labeled with antibodies, cells were fixed and incubated with anti-pPLCc 1 -FITC and total PLCc 1 -PE and analyzed using the FlowCellect PLCc 1 Activation Dual Detection Kit (Merck Millipore, Billerica, MA, www.emdmillipore.com) according to manufacturer's instructions. Alternatively, cells were fixed and incubated with anti-pPLCc 2 -FITC (BD Phosphoflow, BD Bioscience, Mississauga ON, www.bdbiosciences.com), according to the kit instructions, using Phosphoflow Perm Buffer III.
Isolation and Culture of Primary Human CD34 1 Cells CD34 1 umbilical cord blood cells (CB) were obtained from fullterm healthy deliveries after informed consent and purified as described previously [44,45]. Cells were cultured in 24-well plates (2 3 10 4 cells per well) in serum-free X-vivo 15 medium (BioWhittaker, Lonza, Walkersville, MA, www.lonza.com) supplemented with 100 ng/ml Flt3-L in presence or absence of CsA (2lg/ml). Approval was obtained from the Medical University of Vienna Institutional Review Board.

Immunofluorescence Labeling
GMPs were fixed after sorting or cultured for 24 hours in HSC medium followed by stimulation with ionomycin (500 ng/ml) for 15 minutes. Cells were fixed with paraformaldehyde (2%) before permeabilization in 0.5% Saponin, and blocking for 1 hours with 3% bovine serum albumin. Cells were incubated for 1 hours (37 C) with anti-NFAT2 antibody (10 mg/ml) (Thermo Scientific, Waltham, MA, www.thermoscientific.com) followed by secondary antibody (AF633 goat anti-mouse IgG, Life Technologies, Molecular Probes) at 1 mg/ml and DAPI at 2 mg/ml. Cellular localization of NFAT2 was visualized using an Olympus FV1000 confocal microscope.

Cell Proliferation Analysis
For in vitro bromodeoxyuridine (BrdU) assays, cells were pulsed with 1 mM BrdU for 1 hours. For in vivo BrdU assays, mice were injected with CsA (4 mg/mouse), FK506 (0.4 mg/mouse), or vehicle. At the 48 hours time-point, mice were killed and BM analyzed. BrdU (1.5 mg/mouse) was injected 1 or 18 hours before analysis. BM was lineage-depleted and progenitors populations labeled and gated as described in the sorting strategy above. Cells were fixed and labeled using a BrdU flow cytome-try kit (BD Biosciences). Proliferation was also assessed by CFSE dilution (Life Technologies, Molecular Probes), cells were labeled with 2 lM CFSE following the manufacturer's protocol.

Quantitative Real-Time PCR
Total cellular RNA was extracted by Trizol (Invitrogen) phase separation followed by purification using RNeasy Mini/Micro kit (Qiagen), or by using the Arcturus PicoPure RNA Isolation Kit. Reverse transcription was carried out using high-capacity cDNA Reverse Transcription Kit with RNase Inhibitor (Applied Biosystems), or with SuperScript III First Strand Synthesis System for RT-polymerase chain reaction (PCR) (Invitrogen). Realtime PCR was carried out with primers listed in the Supporting Information using GoTaq qPCR Master Mix (Promega).

Microarray Hybridization and Analysis
Total RNA was extracted using a double extraction protocol. ssDNA was prepared, fragmented, and labeled according to the Affymetrix protocol. Fragmented ssDNAs were hybridized to the standard arrays for 17 hours at 45 C; the arrays were then washed and stained using the fluidics station and then scanned using GeneChip Scanner 3000. The gene expression data were then filtered for only probes where the associated gene had a valid NCBI Entrez Gene ID to restrict data to well annotated genes. Gene ontology terms were used to identify genes involved in regulation of cell cycle and transcriptional regulation of differentiation and hematopoiesis. These genes were then tested using a series of two-way analysis of variance (ANOVA) to identify genes that differed in their expression levels due to time or treatment. Processing of the data used Accelrys Pipeline Pilot with visualizations in TIBCO Spotfire. All microarray data files are available for free download at the Gene Expression Omnibus (GEO accession number: GSE47208, http://www.ncbi.nlm.nih.gov/geo. Detailed procedure is described in Supporting Information Methods.

Statistical Analysis
Unless specified differently in the legend, all values are shown as means 6SEM. Student's t-test was used to identify significant differences between groups. For all tests, the 0.05 confidence level was considered statistically significant. In figures, *denotes p < .05, **denotes p < .01, and ***denotes p < .001 in an unpaired Student's t-test.

Calcineurin-NFAT Inhibitors Cyclosporin A and FK506 Selectively Increase Proliferation and Numbers of GMPs In Vivo
We have previously observed that calcineurin inhibitors enhance myelopoiesis [1]. To identify which hematopoietic progenitors are regulated by calcineurin-NFAT signaling, mice were treated with the calcineurin-NFAT inhibitor drugs CsA or Tacrolimus (FK506). After 48 hours of CsA treatment, we detected no change in the percentage of HSCs, MPPs and CMPs among lineage-negative BM cells, whereas percentages of GMPs were significantly increased (Fig. 1A, 1E). The proliferation rate of GMPs was then assessed by BrdU incorporation and DNA content analysis (Fig. 1B). The percentage and total number of proliferating BrdU-positive  Figure 1A shows the gating strategy used. Increased cell cycle progression upon in vivo administration of both CsA and FK506 was specifically observed in GMPs, whereas CMPs were not affected (Fig. 1A, 1C, 1E). Treatment with calcineurin-NFAT inhibitors did not significantly change the total numbers of BM cells or splenocytes (Supporting Information Fig. 2A, 2B), while numbers of LSKs, MPPs, CMPs and GMPs in BM show similar trend as percentage obtained after both inhibitors treatment (Supporting Information Fig.  2C). Accordingly, when BM cells were cultured in methylcellulose medium containing SCF, IL-3, and IL-6, BM from mice treated with CsA gave rise to significantly higher numbers of granulocytic and monocytic CFU (CFU-GM) than did the BM of untreated animals (Fig. 1F).
These data indicate that GMPs but not LSKs, MPPs, or CMPs exhibit increased proliferation after inhibition of the calcineurin-NFAT pathway in vivo.
CsA Promotes the Proliferation of Flt3-L Stimulated Human CD34 1 Umbilical Cord Blood Cells In Vitro Flt3-L promotes myeloid and DC differentiation when added to serum-free suspension cultures of human CD34 1 hematopoietic progenitor cells [49]. We studied the effects of calcineurin inhibitor addition on the in vitro proliferation of Flt3-  L-dependent human hematopoietic precursors. CD34 1 progenitors were cultured in serum-free medium supplemented with Flt3-L in presence or absence of CsA for 3 days. CsA addition increased total cell number (Fig. 3A), percentage of dividing cells (Fig. 3B) as well as increased the rate of CFSE dye dilution (Fig. 3C). Therefore, CsA promotes the Flt3dependent proliferation initiation of hematopoietic progenitor cells.   Table 1). The GO processes and pathway enrichment analysis are shown (Supporting Information Fig. 4A, 4B). We detected increased expression of genes controlling the main cell cycle check points, as well as upregulation of several genes responsible for myeloid cell differentiation (Fig. 4A).
To confirm the relevance of these trends in CsA treated patients, we have reanalyzed gene expression data comparing PBMCs collected of healthy donors and from stable kidney recipients under immunosuppressant monotherapy [50]. The analysis showed significant activation of hematopoiesis and proliferation (Supporting Information Fig. 5A-5C). Furthermore, we compared our mouse array with the patient obtained data. Supporting Information Figure 6 shows that significant IPA processes induced with CsA treatment are similar in both mice and human cells.
Lineage-determining transcription factors were downregulated, suggesting a lower rate of differentiation in the presence of the inhibitors. In contrast, kinases including Cdk4 and Cdk6 were expressed at increased levels when the calcineurin-NFAT pathway was inhibited. To determine how calcineurin-NFAT inhibitor treatment affected transcription in different progenitor subpopulations, the expression of the DEGs identified by microarray analysis was measured in sorted HSCs, MPPs, CMPs, and GMPs cultured for 24 hours in HSC medium with Flt3-L in the presence or absence of CsA or FK506. The expression of the main kinases regulating the cell cycle G 0 checkpoint, Cdk4 and Cdk6, was significantly downregulated during differentiation toward GMPs (Fig. 4B, 4C). Conversely, expression of key inhibitors of cell cycle progression, including Cdkn1a (p21), increased with differentiation toward GMPs (Fig. 4D). Comparable changes in Cdk4, Cdk6, and Cdkn1a (p21) expression were observed in the progenitors analyzed immediately after sorting (Supporting Information Fig. 7A-7C). The sorting strategy used and purity achieved is shown in Supporting Information Figure  1A, 1B. These data clearly suggested a decrease in the selfrenewal rate of progenitors during the process of differentiation. Figure 4E, 4F show the relative changes in expression of Cdk4 and Cdk6 mRNAs in different progenitor populations following calcineurin-NFAT inhibition. Cdk4 and Cdk6 expression in GMPs remained significantly higher in the presence of inhibitors, and accordingly, expression of Cdkn1a (p21) was downregulated (Fig. 4G). This finding again indicated that GMPs are the sole progenitor target affected by CsA or FK506 treatment.
We next sorted HSCs, MPPs, CMPs, and GMPs from untreated mice and stimulated these cells with Flt3-L in vitro in the presence or absence of CsA or FK506 before assessing their proliferation by flow cytometry 1-2 days later using CFSE staining (Fig. 4H, 4I) or BrdU incorporation and DNA content analysis (Fig. 4J). In vitro, the progenitor populations showed different proliferation rates, with GMPs replicating the least when inhibitors were absent (Fig. 4H). In contrast, when CsA or FK506 was added, GMPs substantially increased their proliferation rate (Fig. 4I). GMPs exclusively responded to CsA and FK506 treatment by significantly increasing their proliferation rate relative to GMPs in control cultures (Fig. 4J). Total cell numbers from these cultured progenitors show similar trends (Supporting Information Fig. 2E).

Flt3-L Mediates Activation of the Calcineurin-NFAT Pathway in GMPs
Since calcineurin and NFAT members are expressed in multiple different hematopoietic progenitors, we next assessed whether the main myeloid growth factor Flt3-L might be involved in triggering calcineurin-NFAT signaling to regulate the cell cycle and proliferation rate of these populations. LSKs, CMPs, and GMPs were sorted from BM and loaded with Fluo4-NW, before being stimulated with Flt3-L, ionomycin, or thapsigargin and assessed for changes in intracellular Ca 21 levels using a spectrophotometer (Fig. 6A). We observed that Flt3-L-induced Ca 21 release was effectively blocked by addition of Ca 21 chelator BAPTA (Fig. 6B). We also confirmed our findings by using flow-cytometry to identify changes in intracellular Ca 21 levels in different populations of BM progenitors stained, loaded with INDO-1 and exposed to Flt3-L (Fig. 6C). In agreement with the high sensitivity of GMPs to intracellular Ca 21 release induced by ionomycin treatment ( Fig. 5C; Supporting Information Fig. 9A), GMPs also displayed marked increases in intracellular Ca 21 levels upon Flt3-L stimulation (Fig. 6A, 6C), while LSKs and CMPs did not increase the levels of intracellular Ca 21 upon Flt3-L stimulation (Supporting Information Fig. 9B). Flt3-L, which is expressed on GMPs as well as MPPs and CMPs (Supporting Information Fig. 9C), is associated with PLCc as a possible inducer of Ca 21 entry; so we next measured expression and phosporylation of PLCc 1 and PLCc 2 in myeloid cell progenitors (Fig. 6D-6G). Steady-state phosphorylation of PLCc 1 (pPLCc 1 ) was observed to increase in parallel with cell differentiation from Lin 2 , Sca-1 1 , c-Kit 1 progenitors (pooled HSCs and MPPs, referred to as LSK) to GMPs (Fig. 6E-6G) and was further significantly elevated in GMPs by 10 minutes stimulation with Flt3-L ( Fig. 6E-6G; Supporting Information Fig. 9D). On contrary, PLCc 2 seems to be heavily phosphorylated in steady state in freshly isolated GMPs (Fig. 6D), and we have not observed increase in phosphorylation after FLT3-L trigger (Supporting Information Fig.  9E, 9F). To confirm that Flt3-L stimulation results in NFAT translocation in myeloid progenitors, lineage-depleted, cKIT 1 -enriched cells were isolated from BM and immediately transduced with the NFAT luciferase reporter construct. Treatment with Flt3-L resulted in significantly increased luciferase activity, thereby reflecting functional NFAT stimulation (Fig. 6H).
In summary, we demonstrate that Flt3-L mobilized Ca 21 in GMPs and increased phosphorylation of PLCc 1 . Moreover, primary cKIT 1 -enriched BM cells transduced with an NFAT reporter gene confirmed functional translocation of NFAT upon Flt3-L stimulation. Taken together, these findings show that GMP cell cycle regulation is regulated by Flt3-L activation of PLCc 1 and induction of Ca 21 entry, which in-turn activates calcineurin and NFAT translocation. Calcineurin-NFAT signaling subsequently modulates the expression of genes that affect the cell cycle progression of GMPs so that these cells can proliferate more rapidly than other progenitor populations (Fig. 6I).

DISCUSSION
In the current report, we identified and characterized a new pathway that regulates the cell cycle specifically in GMPs (Fig.  6I). In vivo treatment with CsA or FK506 facilitated the cycling of GMPs, leading to a rapid increase in GMP numbers in the BM. Conditional Cnb1-knockout mice were used to confirm these findings by mixing BM cells from control and Cnb1knockout mice and engrafting them into irradiated recipients. Soon after knockout induction with polyI:C, the majority of GMPs was observed to originate from the calcineurinimpaired donor cells, while the ratio of CMPs remained as injected. These data indicate a specific in vivo effect of calcineurin-NFAT signaling on GMPs proliferative potential. Consistent with these data, sorted, GMPs proliferated less than HSCs, MPPs and CMPs when left untreated during in vitro culture, but addition of CsA or FK506 to these cultures was sufficient to significantly increase GMP proliferation.
To dissect the mechanism of enhanced GMP proliferation under NFAT inhibition, we performed a microarray analysis of gene expression on progenitor cells following brief differentiation with Flt3-L in the presence of calcineurin-NFAT inhibitors. We observed similar changes in PBMCs from CsA treated patients. Global gene expression analysis from mouse BM progenitors revealed marked changes in expression of cell cycle regulation genes as a result of CsA treatment, later validated specifically in GMPs. An important observation was the gradual downregulation of Cdk4 and Cdk6 and upregulation of Cdkn1a (p21) gene expression with the differentiation of cells from HSCs through MPPs, CMPs, and GMPs. This suggests a link between cell differentiation and progression through the cell cycle, mediated by NFAT and triggered by Flt3-L. Our results indicate that the process of slowing down the cell cycle during differentiation is perturbed by calcineurin-NFAT inhibition, particularly at the level of GMPs, both in vitro and in vivo. Therefore, we conclude that the calcineurin-NFAT pathway plays a key role in inhibiting GMP proliferation to regulate myeloid cell differentiation. Our results are supported by other studies in which cell cycle gene regulation was suggested to be important for HSC quiescence [52]. Furthermore, excessive progenitor proliferation led to the exhaustion of HSCs [53,54], which has recently been linked to NFAT Fric, Lim, Mertes et al.
Here, we provide evidence that Nfat1, 2, 4, and calcineurin (Cnb1) are ubiquitously expressed in murine HSCs, MPPs, CMPs, and GMPs. Similarly, others have reported NFAT expression in human CD34 1 immunomobilized progenitors, which hinted at a potential role in myeloid differentiation [15]. We show for the first time that GMPs are exquisitely sensitive to induced Ca 21 release which is required for NFAT activation. Furthermore, NFAT was efficiently translocated to the nucleus following activation in cKit 1 -enriched progenitors. Thus, the calcineurin-NFAT pathway is both present and functional in hematopoietic progenitors, especially in GMPs. Several studies have aimed to assess the role of calcineurin-NFAT signaling in myeloid development: Gallo et al. observed small and nonsignificant increases in myeloid cell numbers upon conditional KO induction of Cnb1 in HSCs, concluding that Cnb1 is not necessary for development of the myeloid compartment [13]. In a different experimental setting, we found that progenitors expressing the VIVIT peptide inhibitor of calcineurin-NFAT [62], give rise to increased numbers of myeloid cells. DCs and CD11b 1 Gr1 1 myeloid cells derived from calcineurin-NFAT-impaired progenitors possessed a substantial developmental advantage over their control counterparts when engrafted into irradiated mice [1]. Similarly, increased numbers of GM-CFU were shown when enriched human CD34 1 progenitors were treated with FK506 in vitro [63]. Congruent with these observations, we here demonstrated that CsA promotes Flt3-L-dependent initial cell proliferation of purified human CD34 1 progenitor cells.
Notably, we illustrated that the calcineurin-NFAT pathway in GMPs is triggered by Flt3-L through pPLCc 1 and Ca 21 release, thus revealing another mechanism by which this growth factor regulates myeloid development. PLCc 1 phosphorylation followed by NFAT signaling has been shown to be essential in T cell development, activation, and survival [64], while other growth factors such as G-, M-, and GM-CSF regulate lineage commitment through PLCc 2 [25,30,31]. The source of Ca 21 driving these processes remains to be elucidated, since PLCc can induce both exogenous flux as well and endogenous release [65]. The relative contributions made by PLCc 1 and PLCc 2 to intracellular or extracellular Ca 21 entry will also require further analysis, since cooperation of Ca 21 entry from both compartments is particularly important for NFAT translocation [4,66].
Flt3-L-induced NFAT signaling leads to cell cycle progression in GMPs that is controlled via a coordinated program of NFATregulated changes in expression of cell cycle control genes. Possible roles for other NFAT binding partners have yet to be investigated. In this study, we show that activation of GMPs with Flt3-L induces phosphorylation of PLCc 1 and consequently stimulates intracellular Ca 21 release, which does not occur in HSCs, MPPs or CMPs. This Ca 21 release initiates translocation of Nfat2 to the nucleus and the transcription of target genes. In addition, we showed that the frequency of cells expressing pPLCc 1 , the main regulator of store-operated Ca 21 release, is increased in parallel with progenitor differentiation. The role played by Flt3 signaling in normal hematopoiesis was previously thought to be mediated solely by activation of Stat5, RAS/MAPK and PI3K [67]. However, deregulation of Flt3 signaling by activating mutation is present in one third of acute myeloid leukemia cases where expansion of GMPs occurs [26,32]. The role of pPLCc 1 , Ca 21 , and NFAT signaling with respect to Flt3 signaling in this disease is currently unknown.
Ca 21 signaling in immune cells is known to have two qualitatively different outcomes: a short peak of Ca 21 release results in immunological synapse formation and granule exocytosis, while the type of prolonged Ca 21 signaling induced by growth factors or other cytokines has been shown to enhance NFAT-dependent transcription. This has been reported in T cells [68,69] and also during embryonic development [70]. The regulatory role of Flt3 in steady state development of GMPs has been suggested [43], leading us to hypothesize that this regulation might involve the NFAT pathway. Flt3 directly regulates HSC quiescence and homeostasis, [67,70] as well as DC development [33][34][35][36]. Flt3 is also expressed in highly proliferating MPPs [22,71,72] and the more differentiated CMPs and GMPs [36,37,43]. The role of Flt3 signaling is clearly linked to regulation of hematopoietic progenitor numbers, as shown in both Flt3 and Flt3-L KO models [34,39], but to date, the involvement of Flt3 in the early events of steady state hematopoiesis has not been fully appreciated. Further support for our results comes from the finding that an activating mutation in Flt3 leads to the development of a myeloproliferative disorder [73] that is characterized by increased numbers of GMPs and an accumulation of mature myeloid cells [74][75][76].

CONCLUSIONS
Our data reveal a novel role for Flt3 signaling in NFAT activation and regulation of myelopoiesis. Investigation of the underlying mechanisms by which NFAT inhibition can increase myelopoiesis uncovered direct regulation of cell cycle control genes Cdk4, Cdk6, and Cdkn1a (p21) by NFAT, specifically in GMPs (Fig. 5I). The calcineurin-NFAT pathway is a therapeutic target in multiple conditions including donor organ rejection, graft-versus-host disease, and autoimmune disorders. While T cells are the main cell type known to be subject to CsA and FK506 immunosuppression, here we provide evidence of an important influence of these drugs on myeloid cell hematopoiesis via direct effects on GMP proliferation. Clearly, the full range of effects these drugs exert on the immune system is not fully appreciated, though the prevalence of side effects following their administration is driving new research into their mechanisms of action. Improved knowledge of the roles played by NFAT during myeloid hematopoiesis will provide insight into clinical studies aiming to better understand homeostatic regulation.