Many reports have revealed the importance of the sympathetic nervous system (SNS) in the control of the bone marrow environment. However, the specific role of neuropeptide Y (NPY) in this process has not been systematically studied. Here we show that NPY‐deficient mice have significantly reduced hematopoietic stem cell (HSC) numbers and impaired regeneration in bone marrow due to apoptotic destruction of SNS fibers and/or endothelial cells. Furthermore, pharmacological elevation of NPY prevented bone marrow impairments in a mouse model of chemotherapy‐induced SNS injury, while NPY injection into conditional knockout mice lacking the Y1 receptor in macrophages did not relieve bone marrow dysfunction. These results indicate that NPY promotes neuroprotection and restores bone marrow dysfunction from chemotherapy‐induced SNS injury through the Y1 receptor in macrophages. They also reveal a new role of NPY as a regulator of the bone marrow microenvironment and highlight the potential therapeutic value of this neuropeptide.
Neuropeptide Y maintains hematopoietic stem cells in the bone marrow by instructing the stromal microenvironment.
NPY deficiency causes impairment of HSC survival and bone marrow regeneration.
Bone marrow dysfunction by NPY deficiency is due to the destruction of SNS nerve fibers and/or EC reduction.
NPY mediates Y1 receptor activation in macrophages, contributing in turn to HSC survival.
Treatment with NPY or Y1 agonists prevents cisplatin‐induced bone marrow nerve injury.
The bone marrow endosteal and perivascular microenvironments support the function of hematopoietic stem cells (HSCs) (Ding et al, 2012). These bone marrow microenvironments consist of functional compartments that regulate HSC retention, survival, self‐renewal, and differentiation (Arai et al, 2004; Scadden, 2006; Sugiyama et al, 2006; Butler et al, 2010; Lévesque et al, 2010). Endothelial cells (ECs) and nestin+ mesenchymal stem cells (MSCs) found in the perivascular microenvironment regulate HSCs survival and bone marrow regeneration (Butler et al, 2010; Méndez‐Ferrer et al, 2010; Ehninger & Trumpp, 2011; Doan et al, 2013), and the sympathetic nervous system (SNS) and macrophages are positive regulators of ECs and nestin+ MSCs, which retain HSCs (Winkler et al, 2010; Chow et al, 2011).
Recent studies have reported that bone marrow innervations regulate HSC microenvironments. Sympathetic nerve fibers synapsed on perivascular cells are associated with HSC hibernation in bone marrow (BM). Moreover, bone marrow nerve injury induces impairment of HSC regeneration (Méndez‐Ferrer et al, 2008; Yamazaki et al, 2011; Lucas et al, 2013). Neurotransmitters released from the SNS also play a key regulatory role in bone marrow microenvironments. In particular, release of acute physiological stress‐induced neurotransmitters such as substance P, dopamine, and norepinephrine regulate regeneration, motility, and proliferation of bone marrow cells (Katayama et al, 2006; Spiegel et al, 2007; Hong et al, 2009).
Neuropeptide Y (NPY) is a neurotransmitter that is often released by sympathetic nerves in the central or peripheral nervous system, and also functions as a orexigenic peptide inducing food intake (Zukowska‐Grojec, 1995; Kalra & Kalra, 2004; Kuo et al, 2007a). Previous studies have demonstrated that NPY regulates immune cell homeostasis, bone homeostasis, or vascular remodeling through Y receptors expressed in bone marrow cells, especially macrophages, osteoblasts, and ECs (Wheway et al, 2005; Kuo et al, 2007b; Lee & Herzog, 2009). Despite these associations between NPY and bone marrow cells, the specific role of NPY in the HSC microenvironment has not been fully characterized.
Here we show that NPY deficiency causes impairment of HSC survival and bone marrow regeneration, as well as integrity of SNS fibers and/or ECs, and that pharmacological NPY increase leads to HSC regeneration through the Y1 receptor, suggesting a new role of NPY in the bone marrow microenvironment. Furthermore, NPY treatment prevents cisplatin‐induced deficits in bone marrow function through the Y1 receptor in macrophages, indicating the potential for regulating NPY activity as a new therapeutic intervention for chemotherapy‐induced sensory neuropathy or bone marrow dysfunction.
NPY deficiency in the HSC microenvironment
We first confirmed the lack of NPY expression in BM of NPY knockout (NPY−/−) mice (Fig 1A) and then analyzed the bone marrow cell composition in wild‐type (WT) and NPY−/− mice. Although no significant difference in the numbers of bone marrow nucleated cells (BMNCs) was observed between WT and NPY−/− mice, the percentage of phenotypic Lin−Sca1+c‐Kit+ (LSK) cells, LSKCD48−CD150+ long‐term HSCs (LT‐HSCs), and CD45−Lin−CD31−Sca1+CD51+ MSCs was significantly reduced in NPY−/− mice (Fig 1B–D). The numbers of hematopoietic stem/progenitor cells (HSPCs) in peripheral blood (PB) also were significantly decreased (Fig 1E). To confirm whether reduced numbers of HSPCs in the PB were derived from their reduced numbers in the BM, we performed migration assays. The results showed no significant difference between migrated LSK cells of WT and NPY−/− mice with or without SDF‐1α (Fig 1F). These results indicated that NPY‐deficient HSPCs had normal motility and migration capacity and that reduction of HSPCs in the PB of NPY−/− mice was likely derived from their reduced numbers in the BM. We next investigated whether the reduced HSC numbers in the BM of NPY−/− mice was a cell‐autonomous effect in vivo and in vitro. LSK cells of WT and NPY−/− mice were stained with annexin V, resulting in no difference of apoptosis between the two groups (Fig 1G). To further confirm apoptosis in vitro, LSK cells sorted from the BM of WT mice were treated with NPY siRNA for 48 h, and the effects on reduced NPY expression were confirmed (Fig 1H). Apoptosis and expression of pro‐ or anti‐apoptotic genes were not changed between the two groups (mice treated with control or NPY siRNA) (Fig 1I and J). Therefore, the reduction of HSCs in NPY−/− mice was not due to cell‐autonomous apoptosis. Cell cycle analysis using an antibody to Ki‐67 showed that 20% more LSK cells were quiescent (in the G0 phase) in NPY−/− mice. However, the S‐G2‐M phase did not change (Fig 1K). Of note, NPY deficiency did not alter the number of macrophage subsets, B cells, T cells, and complete blood counts (Supplementary Fig S1A–D and Supplementary Table S1). Moreover, NPY deficiency did not affect survival (Fig 1D) or apoptosis (Supplementary Fig S1E–G) of osteoblasts differentiated from MC3T3‐E1 pre‐osteoblasts. These results suggested that NPY deficiency led to the impairment of HSC survival, but did not affect their differentiation in the bone marrow microenvironment.
To investigate further the contribution of NPY in the mediation of HSC engraftment and repopulating efficiency in the marrow microenvironment, we transplanted WT CD45.1+ BMNCs into lethally irradiated WT or NPY−/− mice. Transplantation into NPY−/− mice resulted in significantly reduced mouse survival compared to WT mice, and bone marrow regeneration was severely reduced as well (Fig 2A and B). LSK cells also were significantly decreased in the BM of NPY−/− mice after transplantation (Fig 2C). Homing to the BM and migration to the appropriate microenvironment are required for engraftment of transplanted HSCs (Lapidot et al, 2005). We therefore performed a homing assay in transplanted WT and NPY−/− mice (Fig 2D). The result showed that NPY did not affect the homing of hematopoietic progenitors to the BM (Fig 2E and F), suggesting that the reduced HSC survival in NPY−/− mice is independent of homing. To further confirm HSC repopulating activity and the role of NPY in this phenotype, we injected 5‐fluorouracil (5FU), which causes ablation of most hematopoietic cells, into WT or NPY−/− mice to induce repopulation from remaining HSCs. NPY−/− mice showed marked reductions in mouse survival, the number of BMNCs and complete blood counts, and percentage of HSCs at day 12 after 5FU injection (Fig 2G–I and Supplementary Table S1). Overall, these results indicated that NPY was required for HSC survival and bone marrow regeneration.
Next, we transplanted WT or NPY−/− BMNCs into lethally irradiated WT mice to determine the effect of NPY deficiency in BMNCs on bone marrow regeneration. The results showed no significant difference in mouse survival, repopulating efficiency, and the percentage of LSK cells in BM of each group after transplantation (Fig 2J–L). Moreover, NPY−/− BMNCs had a normal homing ability to the BM (Fig 2M). These results indicated that NPY deficiency in bone marrow cells did not affect bone marrow regeneration.
Destruction of SNS fibers and/or ECs by NPY absence
Sympathetic nerves are required for HSC survival and bone marrow regeneration (Katayama et al, 2006; Méndez‐Ferrer et al, 2008; Lucas et al, 2013). Previous studies reported that SNS injury impairs hematopoietic function by reducing nestin+ cells and ECs, and autonomic nervous system dysfunction affected the endothelial dysfunction (Lucas et al, 2013; Amiya et al, 2014). We therefore hypothesized that NPY−/− mice have defective SNS fibers in BM since NPY is a sympathetic neurotransmitter in the autonomic system. As expected, the results showed reduction of bone marrow SNS fibers in NPY−/− mice by staining with an antibody to the catecholaminergic enzyme tyrosine hydroxylase (Th) (Fig 3A). Moreover, the BM of NPY−/− mice contained fewer CD31+ ECs compared to WT mice (Fig 3B). Taken together, these findings revealed that NPY deficiency caused impairment of HSC survival and bone marrow regeneration by reducing the SNS fibers and/or CD31+ ECs. To further reveal cause and effect relationships between SNS fibers and decreased numbers of ECs, as well as the expression of NPY, WT mice were injected with 6OHDA, which induces SNS fibers defects. 6OHDA‐treated mice showed reduction of NPY expression and Th+ SNS fibers (Fig 3C and D); low expression of NPY co‐localized with Th+ SNS fibers in BM (Fig 3E). The numbers of ECs also were decreased in the BM of 6OHDA‐treated mice (Fig 3F). Therefore, these results indicated that destruction of SNS fibers affected ECs survival as well as NPY expression.
Previous studies reported that the p53 tumor suppressor gene in sympathetic neurons affects neuronal survival and that NPY is involved in the cell death mechanism (Trimmer et al, 1996; Corvino et al, 2012; Gonçalves et al, 2012). We first confirmed apoptosis levels in BM of WT or NPY−/− mice and found increased apoptosis in NPY−/− mice (Fig 4A). To gain more specific insights, we measured expression of p53 and several pro‐ or anti‐apoptotic genes, which are the major checkpoints for the cell death pathway, and found increased p53 signaling and expression of pro‐apoptotic genes in BM of NPY−/− mice (Fig 4B and C). Furthermore, apoptotic Th+ cells or CD31+ ECs in vivo were increased in the BM of NPY−/− mice (Fig 4D). NPY siRNA‐exposed neurons or BM ECs in vitro showed reduction of NPY expression, also leading to increased apoptosis and p53 signaling, or expression of pro‐apoptotic genes (Fig 4E–H). These findings indicated that NPY deficiency caused p53‐dependent apoptosis of SNS fibers and/or CD31+ ECs in the BM, resulting in impairment of HSC survival.
NPY/Y1 regulation in bone marrow microenvironment cell survival
We next investigated whether NPY or the Y1 receptor was involved in survival of SNS fibers and marrow microenvironment cells. To establish the appropriate NPY or Y1 agonist injection time, we first confirmed age‐dependent changes of SNS fiber expression in BM of WT or NPY−/− mice and found significant reduction of Th+ nerves in BM of 4‐week‐old NPY−/− mice (Fig 5A). Therefore, 2‐week‐old WT or NPY−/− mice were treated with NPY or a Y1 agonist ([Leu31, Pro34]NPY) intraperitoneally (i.p.) for 4 weeks. NPY treatment into NPY−/− mice significantly prevented loss of LSK cells and LT‐HSCs in the BM of these mice (Fig 5B). The density of Th+ fibers and the number of CD31+ ECs in BM of NPY−/− mice also was increased by NPY treatment (Fig 5C and D). Similar effects were observed in Y1 agonist‐injected NPY−/− mice (Fig 5E–G). Moreover, increased apoptosis in BM of NPY−/− mice was decreased with NPY or Y1 agonist treatment (Fig 5H). These data indicated that NPY/Y1 receptor regulation affected HSC survival by improving apoptosis of SNS fibers and marrow microenvironment cells.
We next imposed pharmacological Y1 receptor blockade in Nes‐GFP mice (Méndez‐Ferrer et al, 2010) by treating with the Y1 antagonist (BIBP3226) for 3 consecutive days, followed by treatment with 5FU (Supplementary Fig S2A). Y1 antagonist treatment caused reduction in mouse survival and hematopoietic function after 5FU injection (Supplementary Fig S2B and C and Supplementary Table S2). Moreover, we found reduced numbers of nestin‐GFP+ cells and CD31+ ECs and increased apoptosis levels compared to control mice treated with 5FU (Supplementary Fig S2D and E). Of note, SNS fiber expression in BM was not changed by 5FU challenge in each group, similar to previous results (Supplementary Fig S2F and Lucas et al, 2013). Next, since ECs are known to express Y1 receptors (Kuo et al, 2007b), ECs cultured from BM were treated with the Y1 antagonist to confirm a direct role of the Y1 receptor in EC survival. The results showed increased apoptosis in ECs subjected to Y1 receptor blockade (Supplementary Fig S2G), indicating that the NPY/Y1 receptor system played a key role in the survival of HSCs and marrow microenvironment cells.
The NPY/Y1 receptor pathway in bone marrow injury
Chemotherapy causes acute bone marrow injury and impairs HSC function or bone marrow regeneration (Noach et al, 2000; Banfi et al, 2001; Lucas et al, 2013). In particular, chemotherapy drugs such as cisplatin and vincristine induce sympathetic neuropathy by reducing the expression of Th fibers (Lucas et al, 2013). Based on these concepts and our findings above, we predicted that NPY might protect against chemotherapy‐induced neuropathy or bone marrow dysfunction through the Y1 receptor. To investigate this hypothesis, we first evaluated the expression of the Y1 receptor in major bone marrow microenvironment cells such as macrophages, osteoblasts, endothelial cells, and nestin+ cells. Expression of Y1 receptor in macrophages and osteoblasts was increased compared to endothelial cells and nestin+ cells (Fig 6A). Therefore, we focused on the Y1 receptor in these cells to study the mechanism of NPY‐mediated BM recovery in chemotherapy‐induced bone marrow injury. Lyz2‐cre recombined in the myeloid cell lineage (Clausen et al, 1999), Col1a1‐cre (Baek et al, 2009) and Y1fl/fl mice (Howell et al, 2003) was used to conditionally deplete Y1 receptor in macrophages and osteoblasts. Y1 receptor mRNA levels were markedly reduced in macrophages cultured from Lyz2‐cre; Y1fl/fl mice and osteoblasts sorted from Col1a1‐cre; Y1fl/fl mice, respectively (Fig 6B and C). Next, we treated control, Lyz2‐cre; Y1fl/fl, or Col1a1‐cre; Y1fl/fl mice with seven cycles of cisplatin for the induction of neuropathy (Aloe et al, 2000). In some animals, NPY was administered during cisplatin chemotherapy (Fig 6D). No significant difference in the numbers of BMNCs was observed between the groups (Fig 6E). Of note, chemotherapy‐induced HSC impairments were recovered in NPY‐treated control or Col1a1‐cre; Y1fl/fl mice (Fig 6F). In addition, NPY treatment improved the sensory neuropathy, reduced density of Th+ fibers, and reduced the number of CD31+ ECs in cisplatin‐treated control or Col1a1‐cre; Y1fl/fl mice after the bone marrow recovery period. Of interest, Lyz2‐cre; Y1fl/fl mice did not show NPY‐mediated protection of LT‐HSCs, Th+ fibers, and ECs due to the absence of Y1 receptor in macrophages (Fig 6G–I). Similar effects were observed in cisplatin‐induced BM apoptosis (Fig 6J). These results suggested that NPY promoted neuroprotection and bone marrow dysfunction from cisplatin‐induced injury through Y1 receptors in macrophages. Although previous reports have demonstrated that chemotherapy in patients with cancer affects the NPY levels (Cağlar et al, 2005), our data showed no significant difference in NPY expression of BM (Fig 6K).
To further confirm the possibility of neuroprotection by NPY through macrophage Y1 receptors, we cultured macrophages from the BM of control or Lyz2‐cre; Y1fl/fl mice and treated these macrophages with NPY. Next, conditioned medium (CM) harvested from each group was applied to PC12 cells to determine the influence on neural differentiation. CM derived from NPY‐treated control macrophages was able to induce differentiation of PC12 cells toward neurons, compared to CM derived from control macrophages. However, cells exposed to CM derived from NPY‐treated Y1 receptor‐deficient macrophages did not show neural differentiation capacity (Supplementary Fig S3A), indicating that neuroprotection by NPY is mediated by factors secreted from Y1 receptor‐stimulated macrophages. Next, to reveal which factors released from macrophages induced neuroprotection, we screened and compared the CM of control or Y1 receptor‐deficient macrophages with and without NPY for 50 different secreted cytokines using an antibody‐based mouse cytokine array (Supplementary Fig S3B). The CM of control macrophages with NPY revealed stronger signals in 4 array spots in comparison with the CM of control macrophages alone (Supplementary Fig S3C). We also confirmed the mRNA levels of these factors in control or Y1 receptor‐deficient macrophages with or without NPY treatment and found up‐regulated TGF‐β expression in control macrophages with NPY. However, Y1 receptor‐deficient macrophages with NPY did not evaluate TGF‐β expression in both CM and cells (Supplementary Fig S3D and E).
TGF‐β plays a key role in the regulation of neuronal survival, differentiation, and repair processes in the nervous system and is produced through the PI3K/Akt/mTOR/eIL4E signaling pathway (Krieglstein et al, 2002; Xiao et al, 2008; Knöferle et al, 2010; Vogel et al, 2010). Another study reported that NPY promoted TGF‐β production by activating PI3K pathway via Y1 receptor (Zhou et al, 2008). Based on these previous studies and our data, we confirmed PI3K pathway in control or Y1 receptor‐deficient macrophages treated with and without NPY. The results indicated that PI3K levels increased in control macrophages with NPY compared to control macrophages alone. Downstream signaling of PI3K pathway such as Akt/mTOR/eIL4E also was activated, but not in Y1 receptor‐deficient macrophages treated with NPY (Supplementary Fig S3F). These findings suggested that TGF‐β released by NPY‐mediated Y1 receptor stimulation in macrophages promoted neuroprotection from cisplatin‐induced injury. Recently, non‐myelinated Schwann cells were shown to maintain stem cells in the bone marrow microenvironment by activating TGF‐β. To confirm the possibility of neuroprotection by TGF‐β secreted from Schwann cells, Schwann cells were cultured from the sciatic nerve of adult C57BL/6 mice (Supplementary Fig S4A). The cultured adult Schwann cells expressed Y1 receptors, although its expression was fewer than macrophages (Supplementary Fig S4B). TGF‐β expression in NPY‐treated Schwann cells showed no significant difference compare to control Schwann cells, as well as in CM (Supplementary Fig S4C and D). These results indicated that NPY/Y1 receptor in Schwann cells did not induce TGF‐β secretion.
Next, to evaluate further whether NPY treatment prevents cisplatin‐induced mouse death and impairment of BM, BMNCs were transplanted into cisplatin‐treated mice with or without NPY treatment after lethal irradiation (Fig 7A). NPY treatment enhanced mouse survival, rescued bone marrow function after transplantation (Fig 7B and C and Supplementary Table S3), and also restored the numbers of CD31+ ECs in cisplatin‐treated mice (Fig 7D). Collectively, our findings strongly support the concept that NPY is required for the maintenance of bone marrow function by protecting SNS fibers and bone marrow microenvironment cell survival through TGF‐β secreted from Y1 receptors in macrophages, offering a new potential therapeutic intervention for chemotherapy‐induced bone marrow impairment.
Recent studies have suggested that the bone marrow environment is regulated by the SNS (Katayama et al, 2006; Méndez‐Ferrer et al, 2008; Yamazaki et al, 2011; Lucas et al, 2013). However, the mechanisms through which neurotransmitters released from the SNS regulate bone marrow microenvironments remain largely unknown. Here we elucidate an unknown function of NPY in regulating the bone marrow HSC microenvironment. NPY‐deficient mice displayed significantly decreased numbers of HSCs and impaired regeneration in BM after bone marrow transplantation, although these same mice exhibited no effect on homing of hematopoietic progenitors to the BM. Previous studies reported that NPY is produced by sympathetic nerve fibers, as well as bone marrow hematopoietic cells (Wheway et al, 2005; Kuo et al, 2007b). To determine the possibility of bone marrow dysfunction by NPY‐deficient marrow cells, we transplanted WT or NPY−/− BMNCs into lethally irradiated WT mice. The results showed no significant difference in mouse survival, repopulating efficiency in BM of WT or NPY−/− BMNCs transplanted recipient mice, as well as homing to BM. Moreover, reduced HSC numbers in BM of NPY‐deficient mice were not due to a cell‐autonomous effect by NPY absence. These results demonstrated that bone marrow microenvironment cells were related to HSC survival in NPY‐deficient mice, further supporting the concept that NPY produced by sympathetic nerve fibers was required for HSC survival and bone marrow regeneration. NPY‐deficient mice also exhibited reduced MSCs, suggesting that NPY may mediate MSC survival. More studies are required to determine the specific role of NPY in the regulation of BM MSCs.
In this study, NPY deficiency caused destruction of SNS fibers and/or a decrease in the number of ECs preserving HSC survival. Moreover, NPY absence or Y1 receptor inhibition in ECs directly affected ECs apoptosis. Notably, the reductions of SNS fibers and bone marrow microenvironment cells were protected by NPY or Y1 agonist injection in NPY−/− mice. In contrast, inhibition of the Y1 receptor showed impairment of HSCs and bone marrow microenvironment cell survival after 5FU treatment. Overall, these results support the notion that an interaction of NPY and the Y1 receptor could mediate the survival of HSCs residing in the perivascular microenvironment by the regulation of bone marrow SNS fibers, and that this interaction maintains survival of ECs and nestin+ cells, which in turn is associated with HSC retention (Supplementary Fig S5). Recently, a very small population of αSMA‐expressing macrophages also was shown to maintain BM HSPCs under steady state (Ludin et al, 2012). However, this cell population was not changed in BM of NPY−/− mice, as well as other macrophage subsets such as CD11b+F4/80+Ly6G+ and F4/80+CD115+Gr1−CD169+. These results indicated that NPY deficiency did not affect HSCs maintained by macrophages.
Chemotherapy‐induced neuropathy is an important side effect of cancer therapy. In addition, chronic bone marrow damage by chemotherapy accompanies impaired hematopoietic regeneration, and this leads to reduced recovery of bone marrow microenvironment cells (Cavaletti & Marmiroli, 2010; Lucas et al, 2013). In this study, we found that NPY treatment restored bone marrow function in cisplatin‐treated mice. These results suggest that NPY warrants further study regarding its potential therapeutic use for chemotherapy‐induced bone marrow abnormalities. Additionally, our findings indicated that cisplatin‐induced SNS injury was prevented by NPY treatment, which reduced cisplatin‐induced apoptosis of BM cells. This protective effect was not found in macrophages of Y1 receptor‐deficient mice. In vitro experiment evaluating PC12 cell differentiation toward neurons found that PC12 cells exposed to CM derived from NPY‐treated control macrophages showed neural differentiation capacity, but not in Y1 receptor‐deficient macrophages treated with NPY. These results highlight the fact that trophic or bioactive factors secreted from macrophages through NPY/Y1 regulation mediates neural protection. We also found up‐regulated TGF‐β secretion or expression by NPY treatment. Τhe PI3K/Akt/mTOR/eIL4E signaling pathway that produces TGF‐β also was activated in macrophages with NPY. These findings are supported by previous studies showing that TGF‐β promotes neuronal survival and differentiation (Krieglstein et al, 2002; Xiao et al, 2008; Knöferle et al, 2010; Vogel et al, 2010). TGF‐β released from non‐myelinated Schwann cells or extracellular matrix in bone marrow has been shown to maintain stem cell survival or recruiting (Wu et al, 2010; Yamazaki et al, 2011). We confirmed expression of Y1 receptor in Schwann cells; however, NPY treatment did not increase TGF‐β secretion from Schwann cells. Taken together, these results indicated that NPY‐induced TGF‐β activation in macrophages is responsible for neuroprotection and HSC survival in bone marrow. Further studies are needed to reveal additional roles of NPY/Y1 receptor regulated TGF‐β secretion from macrophages on neural protection in the bone marrow microenvironment.
Our findings provide proof‐of‐concept that NPY or Y1 agonists have a potential clinical utility as neuroprotective agents for patients treated with chemotherapy. Importantly, NPY has the clinical benefit of being a stable peptide that is synthesized naturally or can be made easily through recombinant technology. Therefore, the modulation of the endogenous bone marrow HSC microenvironment by NPY could be a more effective therapeutic approach than direct transplantation using BM or stem cells. These findings also suggest that future studies investigating the role of NPY in other tissue stem cell microenvironments (for example, the adult neural stem cell microenvironment) may provide new insights in the fields of stem cell research and cell therapy.
Materials and Methods
Six‐ to eight‐week‐old male or female NPY−/− mice, C57BL/6 mice, or C57BL/6.SJL (BoyJ) mice were purchased from the Jackson Laboratory. Nes‐GFP mice (Méndez‐Ferrer et al, 2010) have been described previously. NPY−/− mice and control mice were maintained on a 129S1/SvImJ background. Lyz2‐cre (Clausen et al, 1999) (The Jackson Laboratory), 2.3‐kB Col1a1‐cre (Baek et al, 2009), and Y1fl/fl mice (Howell et al, 2003) were used to delete the Y1 receptor in osteoblasts and macrophages specifically. The block randomization method was used to allocate the animals to exper imental group. To eliminate bias, investigators were blinded during data collection and data analysis. Mice were housed at a 12‐h day–night cycle with free access to tap water and food pellets. All mouse studies were approved by the Kyungpook National University Institutional Animal Care and Use Committee.
To induce bone marrow ablation and force proliferation of quiescent HSCs, we injected 5FU (Sigma; 250 mg per kg body weight, i.v.) under isoflurane anesthesia. Cisplatin (Enzo; 10 mg per kg body weight once per week) was used to induce SNS injury, and mice received i.p. injections of cisplatin for 7 weeks as described (Aloe et al, 2000). One hour after the 7th injection, some mice were euthanized for analysis; 2 weeks after the last injection of cisplatin (to allow for full bone marrow recovery), mice were tested for sensory neuropathy. To assess neuroprotection from cisplatin, mice were injected i.p. with NPY (Bachem, H‐6375) daily during the 7‐week cisplatin treatment period. To investigate the role of the Y1 receptor in the bone marrow microenvironment, 10 nM NPY (Bachem, H‐6375) was applied to macrophages in vitro. Three days later, CM from macrophages was collected. Cultured PC12 cells (Korean cell line bank, 21721) in DMEM supplemented with 10% FBS and 1% penicillin and streptomycin (all from Gibco) were exposed to CM derived from NPY (10 nM)‐treated macrophages to determine its influence on differentiation. Seven days later, the percentage of PC12 cells with two or more dendrites was scored under an Olympus IX71 microscope. NPY siRNA (GE Dharmacon) was treated in LSK cells, neurons differentiated from PC12 cells, and BM ECs and osteoblasts differentiated from MC3T3‐E1 cells (ATCC, CRL‐2594) for 48 h. For in vivo experiments, NPY−/− or Nes‐GFP mice were injected with NPY, Y1 agonist, or Y1 antagonist (50 μg per kg body weight, i.v.). After 1 h, the blood and BM were harvested and analyzed.
Bone marrow transplantation
For assessment of bone marrow microenvironment function of NPY−/− mice, mice were lethally irradiated (10 Gy, two split doses) and injected with 1 × 106 BM from CD45.1 (BoyJ) mice. In some experiments, lethally irradiated CD45.1 mice were transplanted by 1 × 106 BM cells from NPY−/− mice and control mice intravenously. Eight weeks later, the percentage of donor‐derived cells present in the blood was measured by flow cytometry. For homing assays, lethally irradiated WT and NPY−/− mice were injected with 5 × 106 BM from CD45.1+ (BoyJ) mice. After 24 h, the mice were sacrificed, and bone marrow cells were harvested and assayed for competitive transplantation or the presence of CFU‐C as described (Miller et al, 2008). For competitive transplantation, BM collected from the tibiae and femurs of recipient mice and 106 CD45.2+ BM of C57BL/6 were transplanted into lethally irradiated second recipient mice; sixteen weeks after transplantation, the percentage of CD45.1+ cells in the blood of the recipient mice was quantified (Yuan et al, 2005).
BM was flushed from the tibiae and femurs of each mouse. Red blood cells (RBCs) were lysed once for 5 min at 4°C in 0.15 M NH4Cl (StemCell Technologies), washed once with PBS (Gibco), and counted using a hemocytometer. For HSC, MSC, or osteoblast detection, Lin+ cells were removed by magnetic depletion using biotinylated lineage‐specific antibodies (CD5, CD45R, CD11b, Gr‐1, and Ter‐119), followed by depletion with MACs beads conjugated to a monoclonal anti‐biotin (Miltenyi Biotec). For staining of HSCs, Lin− cells were stained with phycoerythrin PE‐Cy7‐conjugated antibodies to Sca1 (558162), APC‐conjugated antibodies to c‐Kit (553356), FITC‐conjugated antibodies to CD48 (557484), and PE‐conjugated antibodies to CD150 (561540), all from BD Biosciences. For MSCs and osteoblasts staining, the chopped bone fragments and BM were digested with collagenase type I (3 mg/ml; Sigma), and Lin− cells were stained with APC‐Cy7‐CD45 (557659), APC‐CD31 (551262), PE‐Cy7‐Sca1 (558162), or PE‐CD51 (551187), all from BD Biosciences. Cells were further stained with streptavidin‐pacific blue (PB) (Invitrogen, S11222). Osteoblasts were sorted. Macrophages subsets, T cells, and B cells were detected by staining with antibodies to PE‐CD11b (BD Biosciences, 557397), FITC‐Ly6G (BD Biosciences, 551160), APC‐F4/80 (AbD Serotec, MCA497), APC cy7‐Gr1 (BD Biosciences, 557661), FITC‐CD169 (AbD Serotec, MCA884F), PE‐CD115 (eBioscience, 12‐1152), αSMA (Abcam, ab66133), PE‐CD3 (BD Biosciences, 555275), and PB‐B220 (BD Biosciences, 558108). Data were collected on a BD LSRII system and AriaIII (BD Biosciences) and analyzed using FlowJo software (Tree Star).
Immunofluorescence staining of bone marrow sections
Frozen bone marrow sections were prepared and immunostained according to a previously published method (Kawamoto, 2003). Bone marrow sections were fixed using dry ice/hexane. Sections were incubated with one primary antibody followed by incubation with secondary antibody conjugated with either Alexa488 or Alexa594 (Life Technologies). Immunofluorescence data were obtained and analyzed using a laser scanning confocal microscope equipped with Fluoview SV1000 imaging software (Olympus FV1000; Japan). Antibodies used were as follows: NPY (Abcam, ab30914, 1:250 dilution), Th (Millipore, AB152 or AB318, 1:250 dilution), and CD31 (BD Biosciences, 550300, 1:50 dilution). TUNEL assay was performed using the In Situ Cell Detection Kit, TMR red (Roche) following the manufacturer's instructions.
Comparisons between two groups were performed with Student's t‐test. In cases where more than two groups were compared to each other, a one‐way analysis of variance (ANOVA) was used, followed by Tukey's HSD test. Comparisons of overall survival were performed using a log‐rank test. All statistical analyses were performed using SPSS statistical software. P < 0.05 was considered to be significant.
MHP and HKJ designed and performed the experiments and wrote the paper. WKM, WWL, and JEL performed the experiments and analyzed the data. EHS and JSB interpreted the data and reviewed the paper. JSB designed the study and wrote the paper. HA, HH, and GNE generated and provided Col1a1‐cre, Y1fl/fl, and Nes‐GFP mice.
Conflict of interest
The authors declare that they have no conflict of interest.
Source Data for Supplementary Figure S3
This work was supported by the Bio & Medical Technology Development Program (2012M3A9C6049913) and Basic Science Research Program (2014R1A2A1A10051107) of the National Research Foundation of Korea (NRF) funded by the Ministry of Science, ICT & Future Planning, Republic of Korea. This research was also supported by a grant of the Korea Health Technology R&D Project through the Korea Health Industry Development Institute (KHIDI), funded by the Ministry of Health & Welfare, Republic of Korea (HI14C1636).
FundingBio & Medical Technology Development Program2012M3A9C6049913
- © 2015 The Authors