Laboratory of Neuroimmunology and Regenerative TherapyUniversity of Nebraska Medical Center, Omaha, Nebraska, USA
Department of Pharmacology and Experimental NeuroscienceUniversity of Nebraska Medical Center, Omaha, Nebraska, USA
Center for Translational Neurodegeneration and Regenerative Therapy, Shanghai Tenth People's Hospital, Tongji University School of Medicine, Shanghai, China
Department of Pathology and MicrobiologyUniversity of Nebraska Medical Center, Omaha, Nebraska, USA
Laboratory of Neuroimmunology and Regenerative Therapy, Departments of Pharmacology and Experimental Neuroscience and Pathology and Microbiology, 985930 Nebraska Medical Center, Omaha, Nebraska 68198-5930, USA
Author contributions: B.Z.: conception and design, collection and/or assembly of data, data analysis and interpretation, and manuscript writing; D.X.: conception and design, collection and/or assembly of data, and data analysis and interpretation; X.D., Q.C., and H.P.: collection and/or assembly of data and data analysis and interpretation; Y.H.: data analysis and interpretation and manuscript writing; Y.L., B.J., L.Z., Y.W., and K.L.W.: collection and/or assembly of data; W.B.T. and J.D.: provision of study material and data analysis and interpretation; W.D.: data analysis and interpretation; S.D. and J.Z.: Conception and design, financial support, data analysis and interpretation, manuscript writing, and final approval of manuscript.
Disclosure of potential conflicts of interest is found at the end of this article.
First published online in STEM CELLS EXPRESSSeptember 17, 2012
Chemokine CXCL12 is widely expressed in the central nervous system and essential for the proper functioning of human neural progenitor cells (hNPCs). Although CXCL12 is known to function through its receptor CXCR4, recent data have suggested that CXCL12 binds to chemokine receptor CXCR7 with higher affinity than to CXCR4. However, little is known about the function of CXCR7 in hNPCs. Using a primary hNPC culture system, we demonstrated that CXCL12 promotes hNPC survival in the events of camptothecin-induced apoptosis or growth factor deprivation, and that this effect requires both CXCR7 and CXCR4. Through fluorescence-activated cell sorting analysis and immunocytochemistry, we determined that CXCR7 is mainly localized in the early endosome, while CXCR4 is more broadly expressed at the cell surface and on both early and recycling endosomes. Furthermore, we found that endocytosis is required for the prosurvival function of CXCL12. Using dual-color total internal reflection fluorescence microscopy and immunoprecipitation, we demonstrated that CXCR7 quickly trafficks to plasma membrane in mediating CXCL12 endocytosis and colocalizes with CXCR4 after CXCL12 treatment. Investigating the molecular mechanisms, we found that ERK1/2 endocytotic signaling pathway is essential for hNPC survival upon apoptotic challenges. Consistent with these findings, a significantly higher number of apoptotic NPCs were found in the developing brain of CXCR7 knockout mice. In conclusion, CXCL12 protects hNPCs from apoptotic challenges through CXCR7- and CXCR4-mediated endocytotic signaling. Since survival of hNPCs is important for neurogenesis, CXCR7 may become a new therapeutic target to properly regulate critical processes of brain development. STEM CELLS2012;30:2571–2583
Chemokines play crucial roles in the central nervous system (CNS) during development. There are approximately 50 chemokines interacting with over 20 chemokine receptors . Among them, CXCL12 has been shown to play an important role in cell migration, proliferation, and survival . Although CXCR4 has long been believed the unique receptor of CXCL12, recent data have suggested that CXCR7 is also a receptor for CXCL12, with 10-fold higher affinity than CXCR4 . Exactly what role CXCR7 plays toward the function of CXCL12 remains unclear. CXCR7 by itself does not trigger G-protein-dependent signaling but can heterodimerize with CXCR4 and regulate CXCL12-mediated G-protein signaling . In addition, CXCL12 induces the activation of ERK1/2 and AKT through CXCR7 in astrocytes and Schwann cells .
Like CXCL12, CXCR7 is also widely expressed in the CNS. During mouse embryonic development, CXCR7 mRNA is first observed at embryonic day 11.5 (E11.5) and increases strongly between E15 and E18 in the marginal zone/layer I. At postnatal day 1 (P1), CXCR7 decreases rapidly and at P7 only scarce signals can be detected . In the cortex, CXCR7 is expressed in GABAergic precursors and in reelin-expressing Cajal-Retzius cells. Also CXCR7 is abundant in neural precursors forming the cortical plate . This expression pattern suggests that CXCR7 may have an important role in the development of the CNS. CXCR7 regulation of cell migration has been well-documented and has been shown to regulate the migration of primordial germ cells in developing zebrafish  and interneuron in mouse [9, 10]. In addition to the migration, CXCR7 has been shown to promote cancer cell survival [11–13], although little is known about how CXCR7 plays an antiapoptotic role in the CNS.
To determine the function of CXCR7 in neural progenitor cells (NPCs), we used a primary human NPC (hNPC) culture that we initially characterized in 2004 . Since then we have extensively documented the survival , differentiation , and proliferation of NPCs . Here, we further investigated the role of CXCR7 in hNPC survival and the molecular mechanism involved. We found that CXCL12 promotes hNPC survival through the coordination between CXCR7 and CXCR4. Furthermore, CXCR7 and CXCR4 are associated with endosome in hNPCs and the survival of hNPCs is dependent on endocytosis of CXCL12 that activates ERK1/2 signaling in endosomes. The revealing of these molecular events may have strong implications to help us better understand hNPC survival during apoptotic challenges.
CXCL12 Enhances hNPC Survival During Camptothecin-Induced Apoptosis or Growth Factor Deprivation
CXCL12 plays crucial roles in the CNS through inducing NPC migration, proliferation, and neuronal axon projection . However, little is known about the effect of CXCL12 on NPC survival. To determine the function of CXCL12 on hNPC survival, we pretreated hNPCs with CXCL12 for 2 hours and then treated with 10 μM camptothecin for an additional 4 hours. Camptothecin is a cytotoxic chemical that causes DNA damage, thus serving as an apoptosis inducer. Camptothecin dramatically increased the number of TUNEL-positive hNPCs compared with untreated control (Fig. 1A, 1B) and CXCL12 alleviated camptothecin-induced apoptosis in a dose-dependent manner (10–100 ng/ml) (Fig. 1C–1F). To confirm the apoptosis of hNPCs, we also determined cleaved poly-ADP ribose polymerase (PARP, an apoptosis indicator) in the cell lysates at similar experimental settings. Camptothecin increased the levels of cleaved PARP. When treated with increasing concentrations (10–100 ng/ml) of CXCL12, the camptothecin-induced PARP cleavage showed a dose-dependent decrease (Fig. 1G, 1H), which confirmed that CXCL12 alleviates camptothecin-induced apoptosis in hNPCs.
To test the extent of the functional effect of CXCL12 on hNPC survival, we treated hNPC with CXCL12 during ongoing apoptosis. CXCL12 was added to the culture 1 hour post camptothecin treatment while hNPCs were treated with camptothecin for 4 hours or 8 hours. Adding CXCL12 after camptothecin treatment decreased the number of TUNEL-positive hNPCs in a dose-dependent manner compared with camptothecin treatment alone (supporting information Fig. S1A–S1F). Similarly, adding CXCL12 after camptothecin treatment decreased the levels of cleaved PARP in a dose-dependent manner compared with camptothecin treatment alone (supporting information Fig. S1G, S1H). Together, these results demonstrated the extent of how CXCL12 enhances hNPC survival during apoptotic challenge or ongoing apoptosis.
Next, we tested whether CXCL12 could promote hNPC survival during growth factor deprivation. hNPCs were deprived of growth factors for 48 hours with or without the treatment of CXCL12. CXCL12 decreased the number of apoptotic hNPCs in a dose-dependent manner (supporting information Fig. S2A–S2F). Similarly, growth factor deprivation for 48 hours induced PARP cleavage, whereas treatment with increasing concentrations of CXCL12 reduced PARP cleavage in a dose-dependent manner (supporting information Fig. S2G, S2H). Taken together, these data suggest that CXCL12 blocks apoptosis and helps hNPC survival during DNA damage or growth factor deprivation.
Both CXCR7 and CXCR4 Are Required for the Antiapoptosis Function of CXCL12 in hNPCs
To determine the mechanism involved in CXCL12-induced hNPC survival, we investigated the involvement of CXCR7 and CXCR4, receptors for CXCL12 in hNPCs. We used siRNA targeting CXCR7 and CXCR4. Both siRNAs effectively silenced their corresponding receptors, as evaluated by real-time reverse-transcription polymerase chain reaction (real time RT-PCR) and Western blotting of CXCR7 (supporting information Fig. S3A, S3C) and CXCR4 (supporting information Fig. S3B, S3D). Three days after siRNA transfection, hNPCs were pretreated with CXCL12 for 2 hours and then treated with camptothecin for 4 hours. Apoptotic levels of hNPCs were determined by TUNEL assay and PARP cleavage. In TUNEL assay, CXCR7 or CXCR4 silencing blocked the antiapoptotic effect of CXCL12 (Fig. 2A–2J). Similarly, CXCL12 reduced the levels of cleaved PARP in camptothecin-challenged hNPCs in a dose-dependent manner, whereas silencing of CXCR7 or CXCR4 abolished the effect (Fig. 2K–2M). These data suggest that both CXCR7 and CXCR4 are required for the antiapoptosis function of CXCL12 in hNPCs.
CXCR7 and CXCR4 Are Associated with Endosome in hNPCs
To determine the mechanism of how CXCR7 and CXCR4 mediate hNPC survival, we evaluated the expression and subcellular localization of CXCR7 and CXCR4 in hNPCs. Flow cytometry analysis revealed that CXCR7 is mostly expressed in the cytosol with little expression on the cell surface, whereas CXCR4 is strongly expressed both on the cell surface and in the cytosol (Fig. 3A–3D). We further labeled hNPCs with antibodies specific to Nestin (green, NPC marker) and either CXCR7 (red) or CXCR4 (red). CXCR7 immunoreactivity was mostly localized to the cytosol, whereas CXCR4 immunoreactivity could be found both on the cell surface and in the cytosol (Fig. 3E, 3F). To further determine the association between CXCR7, CXCR4, and subcellular organelles, we transfected hNPCs with vector expressing CXCR7-mcherry or CXCR4-EGFP and stained the cells with EEA1 (early endosome marker), Rab4 (recycling endosome marker), or Rab11 (recycling endosome marker). CXCR7 mostly colocalized with EEA1, with little colocalization with Rab4 or Rab11 (supporting information Fig. S4A–S4C). In contrast, CXCR4 was evenly colocalized with EEA1, Rab4, and Rab11 (supporting information Fig. S4D–S4F). Together, these data suggest that both CXCR4 and CXCR7 are associated with endosome in hNPCs, and CXCR4 has a stronger association with recycling endosome compared with CXCR7.
Endocytosis Is Necessary for the Antiapoptotic Function of CXCL12 in hNPCs
Because little CXCR7 immunoreactivity was found on the cell surface, CXCR7 likely functions through G-protein-coupled receptor (GPCR)-independent pathways. Therefore, we tested whether the endocytosis-related signaling pathway is involved in the antiapoptosis function of CXCR7. We pretreated hNPCs with monodansylcadaverine (MDC) (10 μM), an endocytosis inhibitor for 1 hour. MDC completely blocked the endocytosis of fluorescent-labeled CXCL12 (supporting information Fig. S5). Treatment of hNPCs with MDC (10 μM) neither significantly changed the apoptotic levels of hNPCs (Fig. 4A, 4D) nor did it change the levels of camptothecin-induced apoptosis (Fig. 4B, 4E). In contrast, inhibition of endocytosis by MDC abolished the antiapoptotic effect of CXCL12 on hNPCs (Fig. 4C, 4F, 4G). To further test the effect of endocytosis inhibition on the antiapoptotic effect of CXCL12, we used Western blotting to determine the levels of cleaved PARP. CXCL12 reduced the levels of cleaved PARP in camptothecin-challenged hNPCs in a dose-dependent manner, whereas pretreatment with MDC abolished the effect (Fig. 4H, 4I). These data suggest that endocytosis is required for the antiapoptotic function of CXCL12 in hNPCs.
Different Endocytotic Properties of CXCR7 and CXCR4 in Mediating CXCL12 Endocytosis
Next, we investigated the molecular events associated with CXCR7- and CXCR4-mediated CXCL12 endocytosis. To selectively visualize events within 100 nm of the plasma membrane, where CXCR7 and CXCR4 initiate CXCL12 endocytosis, we used the total internal reflection fluorescence microscope (TIRFm) . We transfected hNPCs with CXCR7-mcherry and CXCR4-EGFP, and then examined endocytosis of CXCL12 using a dual-color TIRFm system. Time-lapse images were captured every 3 seconds for 20 minutes. We identified CXCR7-mcherry fluorescent puncta under TIRFm illumination, which suggests that CXCR7-positive vesicles are in proximity to the plasma membrane. In addition, we observed that CXCR7 puncta were much more mobile than CXCR4 puncta (supporting information Fig. 5 Video 1). The addition of CXCL12 (100 ng/ml) at 2.5 minutes appeared to enhance the movement of CXCR7 to the plasma membrane (Fig. 5A, 5B; supporting information Fig. 5 Video 1 and Video 2). CXCR7 fluorescence intensity of the whole cell measured within ∼ 100 nm of the membrane by TIRF microscopy increased significantly after CXCL12 treatment (Fig. 5C). In contrast, the fluorescence intensity of CXCR4 did not have a significant change after CXCL12 treatment in the duration of 17.5 minutes under recording (Fig. 5D). These data indicate that CXCR7 may play a more active role in mediating CXCL12 endocytosis, at least at the early moments of endocytosis.
Increased Interaction Between CXCR7 and CXCR4 After CXCL12 Endocytosis
To further study CXCL12 endocytosis, we used Alex647-labeled CXCL12 and used confocal microscopy to monitor the localization of CXCL12 during endocytosis. hNPCs were also cotransfected with CXCR7-mcherry and CXCR4-EGFP to study their association with CXCL12 at different time points of endocytosis. CXCL12 colocalized with CXCR7 as early as 5 minutes, consistent with the TIRFm data. At 2 hours after CXCL12 treatment, colocalization among CXCL12, CXCR7, and CXCR4 was evident, and at 6 hours more CXCL12, CXCR7, and CXCR4 colocalization was found (supporting information Fig. S6A). The close colocalization indicates a close association, and possibly protein-protein interaction between CXCR7 and CXCR4. To test whether CXCR7 and CXCR4 have physical interaction, we used immunoprecipitation (IP) with anti-CXCR4 antibody followed by Western blotting for CXCR7. Little CXCR7 was detected in the precipitated protein mix of the untreated hNPCs (supporting information Fig. S6B). However, after CXCL12 treatment, CXCR7 increased in a time-dependent manner (supporting information Fig. S6B), suggesting increased protein-protein interaction between CXCR7 and CXCR4 upon CXCL12 treatment. Furthermore, to test whether transfected CXCR7-mcherry or CXCR4-EGFP colocalized with endogenous proteins, we transfected hNPCs with either CXCR7-mcherry or CXCR4-EGFP and used immunocytochemistry to determine the endogenous CXCR7 or CXCR4 levels. Cells transfected with CXCR7-mcherry were immunostained with endogenous CXCR4, and cells transfected with CXCR4-GFP were immunostained with endogenous CXCR7. The confocal imaging showed after CXCL12 treatment, CXCR7-mcherry and CXCR4-EGFP colocalized with endogenous CXCR4 and CXCR7, respectively (supporting information Fig. S6C).
CXCL12 Mediates hNPC Survival Through the ERK1/2 Signaling Pathway
Activation of ERK1/2 by CXCL12 promotes cell survival on retinal ganglion cells . Furthermore, recent data have suggested that CXCR7 recruits β-arrestin2 and increase ERK1/2 phosphorylation in cytoplasmic vesicles in HEK293 cells . To further investigate the molecular mechanisms by which the CXCL12-induced endocytosis promotes cell survival, we tested whether CXCL12 mediates hNPC survival through ERK1/2 signaling pathway. We pretreated hNPCs with ERK1/2 inhibitor PD98059 (20 μM). Inhibition of ERK1/2 significantly reduced the antiapoptotic effect of CXCL12 on hNPCs (Fig. 6A, 6B), suggesting a critical role of ERK1/2 in CXCL12-mediated hNPC survival.
To determine whether endocytotic signaling is involved in ERK1/2 activation, we pretreated hNPCs with MDC (10 μM) for 1 hour and determined ERK1/2 phosphorylation upon CXCL12 treatment through Western blotting. Quantification of Western blotting suggested that MDC blocks ERK1/2 activation at the time points between 5 and 60 minutes but not at earlier time points (Fig. 6C, 6D). This result is consistent with a previous report that showed the CXCL12-mediated ERK1/2 activation at early time points is mainly through GPCR signaling and the sustained activation of ERK1/2 at later time points is through endocytotic signaling [22, 23]. To further test whether ERK1/2 activation is dependent on CXCR7- or CXCR4-mediated endocytotic signaling, we transfected hNPCs with specific siRNA targeting either CXCR7 or CXCR4, and determined CXCL12-induced ERK1/2 activation through Western blotting. CXCL12 induced ERK1/2 phosphorylation in hNPCs. CXCR7 silencing significantly reduced CXCL12-mediated ERK1/2 phosphorylation at later time points (5–60 minutes) (Fig. 6E, 6F). In contrast, CXCR4 silencing blocked both early (2 minutes) and late (5–15 minutes) ERK1/2 activation (Fig. 6G, 6H). These data suggest that sustained ERK1/2 activation by CXCL12 is dependent on both CXCR4 and CXCR7. Furthermore, hNPCs were transfected with CXCR7-mcherry and treated with CXCL12 (100 ng/ml) for 30 minutes. Then we detected the ERK1/2 activation in hNPCs by immunocytochemistry. The confocal imaging showed after treatment with CXCL12 for 30 minutes, the ERK1/2 was activated at CXCR7-positive endosomes (supporting information Fig. S7). Taken together, this data demonstrated that CXCL12 enhances hNPC survival through CXCR7- and CXCR4-mediated endocytotic signaling and ERK1/2 activation.
Increased Apoptosis of mNPCs in the Developing Brain of CXCR7 Deficient Mice
To further determine the relevance of CXCR7 in NPC survival in vivo, we obtained an established CXCR7 knockout mouse model . The migration defects of interneuron in CNS development of CXCR7 knockout mice have been well-documented . However, the survival of mNPCs during the developmental stage of CXCR7 knockout mice is unclear. To determine whether there is any defect of mNPC survival in the CXCR7 knockout mice, we obtained floating brains sections from E16.5 embryos. Immunohistochemical analysis of Nestin and TUNEL-positive cells in the lateral and medial ganglionic eminences revealed a significantly higher percentage of TUNEL-positive mNPCs in the CXCR7 knockout mice compared with the wild-type control mice (Fig. 7). These in vivo data suggest that CXCR7 plays a crucial role in mNPC survival during the development of mouse brain.
CXCL12 and its receptor CXCR4 are constitutively expressed during the development of the CNS and play important roles in establishing and maintaining CNS homeostasis . Recent data have suggested that CXCL12 binds to chemokine receptor CXCR7 with higher affinity than to CXCR4. However, little is known about the function of CXCR7 in hNPCs. Here, we demonstrated that CXCL12 promotes hNPC survival during chemical-induced DNA damage or growth factor deprivation (Fig. 1, supporting information Figs. S1, S2). More importantly, CXCR7 and CXCR4, two cognate receptors for CXCL12, are both required for the antiapoptotic function of CXCL12, because siRNA silencing of either receptor abrogated the protective effect of CXCL12 on hNPCs against chemical-induced DNA damage (Fig. 2, supporting information Fig. S3). Interestingly, CXCR7 mediates NPC survival through interaction with CXCR4 (supporting information Fig. S6) and an endocytotic signaling that involves ERK1/2 activation (Figs. 4, 6, supporting information Fig. S4, S5, S7). The relevance of CXCR7 to NPC survival is further demonstrated by in vivo data showing defects of NPC survival in the developing brain of CXCR7 knockout mice (Fig. 7). To our knowledge, this is the first report demonstrating that CXCR7 is required for hNPC survival.
Using a hypoxia model, Bakondi et al. demonstrated that CXCL12 promotes NPC survival through CXCR7 . In light of the study, we have expanded the apoptotic stimuli to chemical-induced DNA damage and growth factor deprivation and found that CXCL12 also promotes the survival of hNPCs (Fig. 1, supporting information Figs. S1, S2). Furthermore, we have found that CXCR7 and CXCR4 are both required for the antiapoptotic function of CXCL12. These results are consistent with the original report  in that both studies have found that CXCR7 is required for NPC survival. However, the results on CXCR4 indicate that the CXCR4 functionality is more variable in different NPC models. The difference in the requirement of CXCR4 in NPC survival may be due to the specificity of the siRNA versus inhibitors used in the studies, different apoptotic challenges, or there is a species-specific effect for CXCR4.
CXCR7 is highly conserved between mammalian species and widely expressed in a variety of cell types . Interestingly, the expression pattern of CXCR7 varies among different cell types. In MCF-7 cells, CXCR7 is expressed both on the cell surface and in the cytosol , but in human T lymphocytes, there is little CXCR7 expression on the cell surface and most expression localized to early endosome . In the CNS, high levels of CXCR7 have been found on the cell surface of astrocytes . On the contrary, CXCR7 is mostly expressed in recycling vesicles (Rab4-positive cells), but not on the cell surface in neurons . Our study finds that CXCR7 expression is unique in hNPCs in that CXCR7 is mostly expressed in cytosol, whereas CXCR4 is expressed both on the cell surface and in cytosol (Fig. 3). Furthermore, the subcellular location of CXCR7 is different from CXCR4. CXCR7 is localized in early endosome (EEA1 positive) with little expression in recycling vesicles (Rab11 and Rab4 positive), whereas CXCR4 is expressed in early endosome as well as recycling vesicles (supporting information Fig. S4). These unique expression patterns of CXCR7 and CXCR4 in hNPCs suggest that they may have similar downstream signaling in early endosome but CXCR4 may be more prone to traffick to the plasma membrane through the recycling vesicles.
The expression patterns of CXCR7 and CXCR4 in early endosome also indicate their role in CXCL12 endocytosis. Indeed, we have determined that endocytosis is necessary for the antiapoptotic function of CXCL12 in hNPCs (Fig. 4). We visualized the molecular events associated with CXCR7- and CXCR4-mediated CXCL12 endocytosis using TIRFm and confocal microscopy and found that CXCR7 fluorescent puncta moved more quickly and trafficked faster to the cell membrane in response to CXCL12 treatment than CXCR4 puncta. In contrast, CXCR4 stimulates CXCL12 endocytosis in a slower manner, at least in the first 17.5 minutes after CXCL12 treatment there is no significant decrease of surface CXCR4 receptors (supporting information Fig.5, Video 1, Video 2). In contrast, the confocal microscopy (supporting information Fig. S6) indicates that the endocytosis of CXCL12 by CXCR4 does occur, likely in a delayed manner after CXCL12 treatment. These data suggest that the CXCR7 receptor has a more rapid dynamic of internalization, which is consistent with previous studies in CXCR7 deficient mice suggesting that interneurons uptake CXCL12 through CXCR7 and prevent the rapid desensitization of CXCR4 receptors [9, 10]. Together, these data illustrate a CXCR7- and CXCR4-mediated endocytotic signaling pathway that guards hNPCs against apoptosis.
Our data suggested that CXCL12 protects hNPCs from apoptosis and both CXCR7 and CXCR4 are required for the antiapoptotic functions of CXCL12. CXCR4 is known to regulate survival of CNS cell types including glioma cells, neural, and oligodendrocyte progenitors [29, 30]. Recent evidence indicates that the receptor CXCR7 is also involved in survival signaling [31, 32]. In line with these reports, we found that CXCR7 deletion has a detrimental effect on NPCs, as there was significantly higher numbers of apoptotic NPCs in the developing brains of CXCR7 deficient mice compared to those of wild-type mice (Fig. 7). CXCR4 enhances survival through GPCR. However, whether CXCR7 functions through GPCR is more debatable. Recent evidence has suggested that CXCR7 and CXCR4 are able to form CXCR7/CXCR4 heterodimers , which is likely to boost the signal transduction of each other. Through confocal live imaging, we show the colocalization of CXCR7 and CXCR4 intracellularly after CXCL12 endocytosis (supporting information Fig. S6A). However, the exact functions of CXCL12 endocytosis as well as CXCR7/CXCR4 heterodimerization remain to be determined.
Previous studies suggested that ERK is an effector in the endocytotic signaling pathways and treatment with endocytosis inhibitor blocks ERK activation . Our study has extended the endocytotic signaling pathways and ERK1/2 activation to the antiapoptotic function of CXCL12 through ERK1/2. It is unclear whether CXCL12-activated ERK phosphorylation is through a classic GPCR pathway or an endocytotic signaling pathway. Previous study suggests that GPCR activation that leads to ERK phosphorylation occurs in the early stages of endocytosis signaling, mostly in the first 2 minutes, while ERK activation mediated by β-arresin2 takes place in a delayed manner, from 5 minutes onward . In our study, CXCL12 induces sustained activation of ERK1/2, which is dependent on CXCR7 or CXCR4 based on the time point indicated. Silencing of CXCR7 may inhibit endocytotic ERK activation while silencing CXCR4 may inhibit both GPCR-dependent ERK activation and endocytotic ERK activation. This is in agreement with the confocal microscopy that showed ERK activation colocalized with CXCR7-positive vesicles after CXCL12 treatment (supporting information Fig. S7). Interestingly, after CXCL12 endocytosis, colocalization between CXCR7 and CXCR4 increased in the endosome, suggesting that CXCR7 and CXCR4 may act in concert in mediating downstream endocytotic signaling. The physical interaction between CXCR7 and CXCR4 were also confirmed by IP. However, the exact mechanism by which CXCR4 interacts with CXCR7 and triggers endocytotic signaling remains unclear. Biologically, the endocytosis signaling pathways are slower than the G-protein pathways , but more constant and stable. These characteristics of endocytosis may be more beneficial for cells in response to apoptotic challenges.
In summary, our studies demonstrate that CXCL12 protects hNPCs from apoptotic challenges, such as camptothecin or growth factor deprivation. The antiapoptotic function of CXCL12 is dependent on CXCR7- and CXCR4-mediated endocytosis, ERK1/2 activation, and may involve interaction between CXCR7 and CXCR4. These molecular events may be used to develop potential therapeutic targets for promoting hNPC survival during apoptotic challenges.
MATERIALS AND METHODS
Antibodies for Western blotting: CXCR7, CXCR4, PARP, phospho-ERK1/2, total ERK1/2, and β-actin protein levels were detected using Anti-CXCR7 (Abcam; 1:1,000), anti-CXCR4 (1:1000; Abcam, Cambridge, MA, www.abcam.com), anti-PARP, anti-phospho-ERK1/2, anti-total-ERK1/2 (1:1,000; Cell Signaling Technologies, Danvers, MA, www.cellsignal.com), and anti-β-actin (1:100,000; Sigma Aldrich, St. Louis, MO, www.sigmaaldrich.com), respectively. Recombinant human CXCL12 was obtained from R&D Systems Minneapolis, MN, www.rndsystems.com. CXCL12-Alex647 was purchased from ALMAC Sciences, Craigavon, United Kingdom, www.almacgroup.com.
Human NPCs were isolated and cultured from human brain tissue. Briefly, hNPCs were cultured in 75-cm flasks and grown in suspension in neurosphere initiation medium, which consisted of X-Vivo 15 (LONZA, Basel, Switzerland, www.lonza.com) with N2 supplement (Invitrogen, Carlsbad, CA, www.invitrogen.com), neural cell survival factor-1 (LONZA), basic fibroblast growth factor (20 ng/ml, Sigma-Aldrich), epidermal growth factor (20 ng/ml, Sigma-Aldrich), and leukemia inhibitory factor (10 ng/ml, EMD Millipore Corporation, Billerica, MA, www.emdmillipore.com). All studies using human subjects were performed in full compliance with the University of Nebraska Medical Center and National Institutes of Health's ethical guidelines.
CXCR7 knockout mice were kindly provided by Dr. Liang Zhou (Northwestern, Chicago, IL) . All mice were housed and bred at University of Nebraska Medical Center, in accordance with ethical guidelines for care and use of laboratory animals set forth by the National Institutes of Health. All procedures were approved by the Institutional Animal Care and Use Committee of the University of Nebraska Medical Center.
siRNA Silencing of CXCR7 and CXCR4
siRNA silencing in hNPCs was performed as previously described . Briefly, predesigned siRNA duplex targeted against CXCR7 and CXCR4 was purchased from Applied Biosystems Inc. hNPCs were transfected with 100 nM siRNA duplex for 72 hours in the presence of siIMPORTER (EMD Millipore) according to the manufacturer's instructions. Nonspecific control siRNAs from Applied Biosystems Inc. were also transfected at the same concentration as controls to CXCR7 and CXCR4 siRNA.
RNA Extraction and TaqMan Real-Time RT-PCR
Total RNA was isolated with TRIzol Reagent (Invitrogen) and RNeasy Kit according to the manufacture's protocol (Qiagen Inc., Valencia, CA, www.qiagen.com). Primers used for real-time RT-PCR include CXCR7 and CXCR4 and glyceraldehyde 3-phosphate dehydrogenase (GAPDH) from Applied Biosystems Inc. Real-time RT-PCR was carried out using the one-step quantitative TaqMan assay in a StepOne Real-Time PCR system (Applied Biosystems Inc., Foster City, CA, www.appliedbiosystems.com). Relative CXCR7 and CXCR4 mRNA levels were determined and standardized with a GAPDH internal control using comparative ΔΔCT method. All primers used in the study were tested for amplification efficiencies and the results were similar.
IP and Western Blotting
Cell lysates were collected using IP buffer and precleared using protein G (Thermo Scientific Inc., Waltham, MA, www.thermoscientific.com). Then, precleared cell lysates were incubated with monoclonal antibody of CXCR4 (R&D Systems) overnight on a rotator. Rabbit antibody of CXCR7 (Abcam) was used to detect the CXCR7-CXCR4 interaction using Western blotting.
Immunocytochemistry and Confocal Microscopy
For immunofluorescence staining, hNPCs were fixed using 4% paraformaldehyde (PFA) and permeabilized with 0.5% Triton-X and 3% bovine serum albumin (BSA) in phosphate buffered saline (PBS). After washing, hNPCs were incubated with primary antibodies overnight. Antibodies used for immunostaining include monoclonal anti-CXCR7 (R&D Systems; 1:100), anti-rab11 (1:100; BD Biosciences, San Diego, CA, www.bdbiosciences.com), anti-rab4 (BD Biosciences; 1:100), and anti-EEA1 (BD Biosciences; 1:150). Cultures were washed and then secondary antibodies, anti-mouse IgG (coupled with Alexa Flour 488, Invitrogen), or anti-rabbit IgG (coupled with Alexa Fluor 568, Invitrogen), were added for 1 hour at room temperature. Nuclear DNA was labeled with 4',6-diamidino-2-phenylindole (DAPI) (Invitrogen) for 10 minutes after the secondary antibody at room temperature. Coverslips were mounted on glass slides with mounting medium (Sigma-Aldrich). Triple immunostaining was examined by a Zeiss META 510 confocal microscope (Carl Zeiss MicroImaging, LLC, www.zeiss.com). For CXCR7 and CXCR4 trafficking experiments, full-length human CXCR7 and CXCR4 were subcloned into pmcherry-N1 (Clontech Laboratories, Inc., Mountain View, CA, www.clontech.com) and pEGFP-N1 (Clontech), respectively. hNPCs were seeded into 35-mm glass-bottom dishes and transfected with CXCR7-mcherry and CXCR4-EGFP using Lipofectamine LTX reagent (Invitrogen). Twenty-four hours after transfection, hNPCs were deprived of growth factors overnight and then treated with CXCL12-Alex647 (100 ng/ml). The colocalization of CXCR7, CXCR4, and CXCL12 was detected using a Zeiss Meta 510 confocal microscope (Carl Zeiss MicroImaging, LLC).
Flow Cytometry Assay
hNPCs were cultured in six-well plates at the density of 0.5 million/well. Cells were resuspended in 3% fetal bovine serum (FBS)-PBS and the cell surface expression of CXCR7 and CXCR4 was detected using phycoerythrin (PE)-conjugated CXCR7 monoclonal antibody and allophycocyanin (APC)-conjugated CXCR4 monoclonal antibody from R&D, respectively. A matching mouse IgG isotype control was used. For intracellular expression of CXCR7 and CXCR4, cells were permeabilized with a permeabilization buffer before antibody incubation. Data were acquired on a FACSCalibur flowcytometer (BD Biosciences) using CellQuest Software (BD Biosciences). At least 10,000 cells were analyzed per sample.
In Vitro TUNEL Assay
hNPCs were seeded on poly(D-lysine)-coated coverslips at the density of 0.1 million/well. After treatment, cells were fixed and permeabilized with 0.5% Triton-X, and the apoptotic cells were determined by in situ cell death detection kit with Fluorescein (Roche, Basel, Switzerland, www.roche-applied-science.com) according to the manufacture's protocol. Images were taken using a Nikon Eclipse E800 fluorescent microscope. All obtained images were imported into Image-ProPlus, version 7.0 (Media Cybernetics, Inc., Rockville, MD, www.mediacy.com) for quantifying number of apoptotic cells. The assessors were blinded during image acquisition or quantification.
Free-Floating Immunohistochemistry and Image Analysis
The brains from E16.5 wild-type or CXCR7 knockout embryos were rapidly removed and immersed in freshly depolymerized 4% PFA for 48 hours and then cryoprotected by 30% sucrose for 48 hours before sectioning. Fixed, cryoprotected brains were frozen and sectioned in the horizontal plane at 30 μm using a Cryostat (Leica Microsystems Inc., Bannockburn, IL), with sections collected serially in PBS. Antibody to Nestin was used for the detection of NPCs. Apoptotic cells were determined by in situ cell death detection kit with TMR red (Roche) according to the manufacture's protocol. Images were taken using a Zeiss Meta 710 confocal microscope (Carl Zeiss MicroImaging, LLC) (×20 object, tile scan 2X2 mode). Eleven brain section images from three litters of CXCR7 wild-type or knockout mice were imported into Image-ProPlus, version 7.0 for quantifying levels of TUNEL/Nestin double-positive staining. Apoptotic NPCs in the lateral and medial ganglionic eminences in the developing cerebrum were analyzed.
An objective-type TIRF microscope was used to observe CXCR7- and CXCR4-mediated CXCL12 endocytosis. The setup consisted of two solid state lasers (488 nm and 561 nm), a high-numerical-aperture (NA) oil immersion objective (×60, 1.45 NA), an inverted microscope (IX71; Olympus, Tokyo, Japan, www.olympusamerica.com), an Electron Multiplying Charge Coupled Device (EMCCD) (Hamamatsu, Japan, hamamatsucameras.com), and temperature controller (Live Cell Instrument, Korea, www.chamlide.com). Images were acquired every 3 seconds, CXCL12 (100 ng/ml) was added to hNPCs after recording for 2.5 minutes, and then the recording was continued for another 17.5 minutes. All images and movies were acquired and analyzed using Metamorph software (Universal Imaging Corporation, Downingtown, PA, www.moleculardevices.com).
Data were analyzed by one-way analysis of variance (ANOVA) followed by the Tukey's test for pairwise comparisons or by two-way ANOVA followed by post-tests using GraphPad Prism software. Significance was considered with a p value less than .05. All experiments were performed with at least three donors to account for any donor-specific differences. Assays were performed at least three times in triplicate or quadruplicate.
We thank Dr. Changhai Tian, Dr. Xiqian Lan, Lijun Sun, and Hamilton Vernon, who provided technical support for this work. We thank Dr. Liang Zhou at Northwestern University who provided the CXCR7 knockout mice. We thank Janice A. Taylor and James R. Talaska of the Confocal Laser Scanning Microscope Core Facility at the University of Nebraska Medical Center for providing assistance with confocal microscopy and the Nebraska Research Initiative and the Eppley Cancer Center for their support of the Core Facility. This work was supported in part by research grants by the National Institutes of Health: R01 NS 41858-01, R01 NS 061642-01, 3R01NS61642-2S1, R21 MH 083525-01, P01 NS043985, P20 RR15635-01, and National Natural Science Foundation of China # 81028007 (J. Z.), Research to Prevent Blindness and EY10542 (W.T.), and National Institutes of Health R21 NS066841 (H.P.).
DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
The authors indicate no potential conflicts of interest.