Address correspondence and reprint requests to Lucio Annunziato, MD, Division of Pharmacology, Department of Neuroscience, School of Medicine, University of Naples, “Federico II”, Bldg 19, Via S. Pansini 5, 80131 Naples, Italy. E-mail: firstname.lastname@example.org or Laura Cerchia, PhD, Istituto per l'Endocrinologia e l'Oncologia Sperimentale del CNR “G. Salvatore”, Via S. Pansini 5, 80131 Naples, Italy. E-mail: email@example.com
Ret receptor tyrosine kinase is the signaling component of the receptor complex for the family ligands of the glial cell line-derived neurotrophic factor (GDNF). Ret is involved in the development of enteric nervous system, of sympathetic, parasympathetic, motor and sensory neurons, and it is necessary for the post-natal maintenance of dopaminergic neurons. Ret expression has been as well demonstrated on microglia and several evidence indicate that GDNF regulates not only neuronal survival and maturation but also certain functions of microglia in the brain. Here, we demonstrated that the plant lectin Griffonia (Bandeiraea) simplicifolia lectin I, isolectin B4 (IB4), commonly used as a microglial marker in the brain, binds to the glycosylated extracellular domain of Ret on the surface of living NIH3T3 fibroblasts cells stably transfected with Ret as well as in adult rat brain as revealed by immunoblotting. Furthermore, confocal immunofluorescence analysis demonstrated a clear overlap in staining between pRet and IB4 in primary microglia cultures as well as in adult rat sections obtained from control or post-ischemic brain after permanent middle artery occlusion (pMCAO). Interestingly, IB4 staining identified activated or ameboid Ret-expressing microglia under ischemic conditions. Collectively, our data indicate Ret receptor as one of the IB4-reactive glycoconjugate accounting for the IB4 stain in microglia under physiological and ischemic conditions.
RET (Rearranged during transfection) gene encodes a transmembrane glycoprotein belonging to the receptor tyrosine kinase family that consists of three functional regions: the extracellular region including four N-terminal cadherin-like domains (CLD1-4) followed by a single cysteine rich domain (CRD), the transmembrane region, and the intracellular region formed by a bipartite tyrosine kinase domain.
Ret is activated by a complex consisting of a glial cell line-derived neurotrophic factor (GDNF) family ligand and a glycosyl phosphatidylinositol-anchored co-receptor, GDNF family receptors α (GFRα) (Trupp et al. 1996; Airaksinen and Saarma 2002).
In the nervous system, Ret signaling regulates the development of enteric, sympathetic, parasympathetic, motor, and sensory neurons, and is necessary for the post-natal maintenance of dopaminergic neurons (Runeberg-Roos and Saarma 2007). GDNF signaling exerts potent pro-survival actions on midbrain dopaminergic neurons (Lin et al. 1993), spinal cord motoneurons (Henderson et al. 1994), noradrenergic neurons of the locus coeruleus (Arenas et al. 1995), cerebellar Purkinje cells (Mount et al. 1995), cholinergic neurons of the basal forebrain (Williams et al. 1996), as well as peripheral sensory and autonomic neurons (Buj-Bello et al. 1995). Interestingly, GDNF powerfully and selectively supports the survival of a subpopulation of nociceptive dorsal root ganglion (DRG) neurons that terminate in Lamina II of the spinal dorsal horn (Bennett et al. 1998; Wang et al. 2003). These GDNF-responsive neurons are of small diameter and can be specifically identified by the binding of the plant lectin Griffonia (Bandeiraea) simplicifolia lectin I, isolectin B4 (IB4). Colocalization studies revealed that essentially the entire IB4-positive population of DRG neurons with small body displayed immunoreactivity for Ret (Molliver et al. 1997; Leclere et al. 1998). More recently, it has been demonstrated by our research group that GDNF treatment stimulates Ret-dependent signaling in a population of IB4-positive cells in the pyramidal layer of the CA3 region in hippocampal organotypic cultures (Boscia et al. 2009a).
Lectins are a group of heterogeneous plant proteins of non-immune origin binding to the carbohydrate portion of glycoproteins and glycolipids (Barondes 1988). They have been used to characterize the glycosylation status specifically in a variety of tissues (Balding and Gold 1975; Brabec et al. 1980; Franz et al. 2006). Among the larger family of lectins, the lectin IB4 is a 114 kDa glycoprotein belonging to a family of five tetrameric type I isolectins, and is selective for terminal α-d-galactosyl residues (Goldstein and Winter 1999). Although IB4 binds to endothelial cells (Alroy et al. 1987) and selective populations of neurons (Kawai et al. 2001), it is largely employed in histopathology for specifically identifying microglial cells in the brain of several species including rodents and humans (Streit 1990; Pearse et al. 2004; Lunemann et al. 2006). Interestingly, GDNF receptors, Ret and GFRα1, are expressed in IB4-positive microglial cells (Honda et al. 1999; Rémy et al. 2001), and more importantly, GDNF administration, acting trough Ret receptor have a modulatory role in microglial activities including proliferation and survival (Chang et al. 2006; Boscia et al. 2009a).
Up to now the glycoproteins that are recognized by IB4 have not been well characterized.
Here, we asked whether the ability of IB4 to identify different Ret-expressing cell populations in the brain (neurons as well as microglial cells) could be because of a direct binding of the lectin to the glycosylated extracellular domain of Ret. Using NIH3T3 cells stably transfected with the Ret receptor, we demonstrate that IB4 strongly binds Ret on surface of living cells and binds the glycosylated mature form of the receptor when used as a probe in western blotting analyses. The co-existence of IB4-binding and activated Ret staining was explored in primary microglia cultures as well as in adult rat sections obtained from control or ischemic brain at 3 days following pMCAO insult. Collectively, our data propose that isolectin IB4 recognizes the functional receptor for GDNF, Ret under normal as well as ischemic conditions in the brain.
Materials and methods
Male Sprague-Dawley rats (250 to 270 g; Charles River, Calco, Italy) were housed in a temperature- and humidity-controlled colony room under diurnal/lighting conditions. Animal handling was in accordance with the International Guidelines for Animal Research and the experimental protocol was approved by the Animal Care and Use Committee of ‘Federico II’ University of Naples.
Rats were divided into 3 groups: (1) control (n = 3); (2) sham-operated (n = 3); and (3) ischemic (n = 3). The latter received a permanent occlusion of the middle cerebral artery (pMCAO) for 3 days. pMCAO was performed as previously described (Boscia et al. 2009b). The left pMCAO was performed by electrocoagulation with a bipolar electrocauterizer (Diatermo, GIMA, Milan, Italy). The body temperature was monitored with a rectal probe and maintained at 37 ± 0.5°C until awakening. Sham-operated animals underwent the same procedures except for middle cerebral artery electrocoagulation. Cerebral blood flow was monitored in the cerebral cortex ipsilateral to the occluded middle cerebral artery with a laser Doppler flowmeter.
Parental NIH3T3 cells, NIH/MEN2A, and NIH/MEN2B, expressing the human Ret9C634Y and Ret9M918T mutant proteins responsible for multiple endocrine neoplasia (MEN) type 2A and 2B, respectively, were cultured as previously described, and when required, with appropriate selection pressure (Cerchia et al. 2005).
Primary rat microglia cultures
Microglia cultures were prepared from primary rat mixed glial cell cultures, as previously described (Boscia et al. 2009b). Briefly, cerebral cortices isolated from post-natal day 1 rat brain were first dissociated enzymatically in a solution containing 0.125% trypsin and 1.5 mg/mL DNase (Sigma-Aldrich, St. Louis, MO, USA) and then mechanically in Dulbecco's modified Eagle's medium supplemented with 10% fetal bovine serum, 1% penicillin-streptomycin, 2 mmol/L l-glutamine (Invitrogen, Carlsbad, CA, USA) (normal medium). Cell pellet was re-suspended and plated in tissue culture flasks in normal medium at 37°C in a humidified, 5% CO2 incubator. Once confluent (after 7–9 days), the microglia were separated by mechanical shaking of flasks on a orbital shaker for 60 min at 200 rpm and plated onto poly-d-lysine (Sigma-Aldrich, Milan, Italy)-coated coverslips. This procedure yields 98% IB4-FITC or OX42-positive cells
Immunoprecipitation and western blot analysis
Cell extract preparation, immunoprecipitation, and immunoblotting were performed as previously described (Esposito et al. 2008).
For immunoprecipitation analysis, cell extracts were incubated with anti-Ret (C-19) antibodies (Santa Cruz Biotechnology Inc, Santa Cruz, CA, USA). The cell lysates or immunoprecipitates were subjected to 8% sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS-PAGE). Gels were electroblotted into polyvinylidene difluoride membranes (Millipore Co., Bedford, MA, USA), and filters were probed with anti-Ret (H-300) antibodies (Santa Cruz Biotechnology Inc), anti-(Tyr1062-phosphorylated) Ret rabbit polyclonal antibodies (indicated as pRet, Santa Cruz Biotechnology Inc) or Isolectin GS-IB4 directly conjugated to biotin (IB4-Biotin, Invitrogen). For IB4-Biotin incubation, the filters were first blocked overnight in 100 mM Tris- buffered saline pH 7.5 (TBS) with 8% (w/v) no-fat dry milk, 0.1% tween 20 at 4°C and then incubated with IB4-Biotin (1 : 2000) in TBS, 0.1% tween 20 for 1 h at 25–27°C.
Proteins were visualized with peroxidase-conjugated secondary antibodies or streptavidin-peroxidase-conjugated (Amersham-Pharmacia Biosciences LTD, Uppsala, Sweden) using the enhanced chemiluminescence system. Where indicated, filters have been stripped as described (Cerchia et al. 2003).
Cell treatment with IB4-Biotin
Parental NIH3T3 cells, NIH/MEN2A, or NIH/MEN2B cells were plated at equal density on 10-cm dishes, washed twice in ice-cold phosphate-buffered saline (PBS) and incubated with 1 mg/mL IB4-Biotin in PBS for 25 min at 4°C with agitation. The reaction was stopped with 50 mM Ammonium Chloride for 10 min at 4°C before cell lyses. To recover biotinilated protein, cell lysates were incubated with immobilized streptavidin beads (Pierce Biotechnology, Rockford, IL, USA) for 1 h at 4°C. Samples were washed four times with PBS and denatured in Laemmli buffer for 5 min at 100°C before 8% SDS-PAGE.
Immunofluorescence staining and confocal microscopy
Confocal immunofluorescence procedures in cells or sections were performed as previously described (Boscia et al. 2008, 2012; Esposito et al. 2011). Rats were anesthetized intraperitoneally with chloral hydrate (300 mg/kg) and perfused transcardially with 4% wt/vol paraformaldehyde and in phosphate buffer. The brains were sectioned coronally at 60 μm on a vibratome. Cell cultures were fixed in 4% wt/vol paraformaldehyde in phosphate buffer for 30 min. After blocking, cells or sections were incubated with primary antibodies for 24 h or 48 h, respectively. The primary antibodies used in these experiments were the following: polyclonal anti-(Tyr1062-phosphorylated) Ret (1 : 200), monoclonal anti-OX-42 (1 : 500, Millipore, Billerica, MA, USA), monoclonal anti-NeuN (1 : 1000, Millipore); monoclonal anti-phosphorylated-p44/42 mitogen-activated protein kinase E10 (indicated as pERK, 1 : 500, Cell Signaling, Beverly, MA, USA). Microglia was also identified using Isolectin Bandeiraea Simplicifolia B4 (IB4) directly conjugated to FITC (FITC-IB4; 1 : 200; Sigma, Milan, Italy). Subsequently, sections or cells were incubated with corresponding fluorescent-labeled secondary antibodies (Alexa 488- or Alexa 594-conjugated anti-mouse or anti-rabbit IgGs, Invitrogen, Monza, Italy). Immunohistochemical experiments with IB4-FITC in NIH3T3 cell lines have been performed without permeabilizing cells during blocking procedure. Images were observed using a Zeiss LSM510 META/laser scanning confocal microscope (Carl Zeiss Microscopy, Jena, Germany). Single images were taken with an optical thickness of 0.7 μm and a resolution of 1024X1024. Confocal serial optical sections were acquired as z-stacks to image a single cells in 0.40 μm steps with total depth covering 10 μm.
Data are expressed as mean ± SEM of values obtained in three separate experiments. Statistical comparisons between controls and treated groups were performed using the one-way analysis of variance followed by Newman Keul's test. p <0.05 was considered significant.
Results and Discussion
The expression of GDNF receptors, Ret and GFRα1, has been described not only in neuronal but also in microglial cells (Honda et al. 1999; Rémy et al. 2001). Accordingly, it has been reported that GDNF-Ret-dependent signaling may not only have potent trophic actions on a wide variety of neuronal populations (Sariola and Saarma 2003) but also modulates different microglial activities including survival and proliferation (Chang et al. 2006; Boscia et al. 2009a). In this study, to explore the co-existence of immunosignal for activated Ret and IB4-binding, confocal immunofluorescence staining with anti-pRet antibodies was combined with IB4-FITC labeling in rat primary microglia in the presence or absence of GDNF stimulation. When IB4-positive microglia was exposed to 10 ng/mL GDNF for 30 min, pRet immunoreactivity was intensely detected along the cell plasma membrane when compared to unstimulated cells (Fig. 1ai and v). Interestingly, a clear overlap staining between pRet and IB4 labeling was observed at the plasma membrane level after GDNF stimulation (Fig. 1ai–viii). To control for a possible cross-reactivity in double-labeling experiments with IB4-FITC in microglia, pRet primary antiserum was replaced with the irrelevant neuronal anti-NeuN antibody either in the absence or in presence of GDNF stimulation. In these experimental conditions, no staining was observed with NeuN antibody both in control or stimulated cells, whereas an intense IB4-FITC labeling along the plasma membrane was observed following GDNF exposure (Fig. 1aix–xvi). To further confirm the specificity of pRet and IB4-FITC signals, we performed double-labeling experiments with IB4-FITC and phospho-ERK antibodies. As expected, GDNF stimulation up-regulated both IB4-FITC and pERK labeling if compared to the controls, but only IB4-FITC was detected on microglial plasma membrane (Fig. 1axvii–xxiv). Colocalization of pRet with IB4-FITC was further examined by z-stack confocal microscopy. Indeed, Z-stack images of a representative single microglial cell clearly indicated that pRet and IB4-FITC signals co-localized at the plasma membrane level following GDNF stimulation (Fig. 1b). Quantitative analysis of pRet and IB4 staining in the cytosol and plasma membrane compartments of microglial cells revealed that GDNF exposure significantly up-regulated the number of cells displaying pRet and IB4-FITC labeling on the plasma membrane when compared to unstimulated cells (Fig. 1c). Beside this parallel behavior following the neurotrophin application, pRet and IB4 displayed distinct intracellular staining under control or stimulated conditions. For instance, though GDNF increased pRet signal in the cytosol, the entirely population of microglia was labeled intracellularly by the lectin IB4 both in the absence or presence of GDNF (Fig. 1c).
Thus, we asked whether the correlation among Ret immunoreactivity and the IB4-labeling could be because of the ability of Isolectin IB4 to recognize the glycosylated extracellular domain of Ret receptor. To this we aim, we used NIH3T3 cells stably expressing the RetC634Y (NIH/MEN2A) or RetM918T (NIH/MEN2B) mutants. RetC634Y is mutated in the extracellular domain and forms spontaneously active homodimers on the cell surface, whereas RetM918T is identical to the wild-type receptor in its extracellular domain and, in the absence of the ligand and co-receptor, remains monomeric (Santoro et al. 1995; Cerchia and de Franciscis 2006). Both mutants, in their mature form, are abundantly N-glycosylated and the predicted N-glycosylation sites in the human extracellular domain of Ret are not evenly distributed but also the majority of them (9 of 12) appear downstream of the Ca2+ coordination site in the CLD1 (Kjaer and Ibáñez 2003).
NIH/MEN2A and NIH/MEN2B cells have a similar morphology that mirrors the Ret-dependent human pheochromocytoma phenotype of MEN2 syndromes (Santoro et al. 1995; Cerchia and de Franciscis 2006). Confocal analysis revealed that NIH/MEN2A and NIH/MEN2B-Ret-transfected cells displayed an intense staining upon IB4 labeling, with most of the cells displaying a clear IB4 binding particularly confined along the plasma membrane (Fig. 2a–c). In contrast, IB4 staining was scarcely detected in non-transfected NIH3T3 cells (Fig. 2d).
Further, lysates from NIH3T3, NIH/MEN2A, and NIH/MEN2B cells were immunoprecipitated with anti-Ret antibodies and visualized with IB4 staining. As shown in Fig. 2e, IB4-specific enhanced chemiluminescence was associated with the 170-kDa band corresponding to the mature and completely glycosylated form of the Ret protein present on the cell surface but not to the 150-kDa immature and partially glycosylated form of Ret, present in the endoplasmic reticulum.
IB4 is a carbohydrate-binding protein thus, as expected, it binds several proteins in lysates from parental NIH3T3 as well as from Ret-transfected NIH/MEN2A and NIH/MEN2B cells in addition to Ret (Fig. 3a). Further, to verify the ability of IB4, once incubated on living cells, to bind the mature glycosylated form of Ret exposed on membrane of NIH/MEN2A and NIH/MEN2B cells, lysates from cells prior incubated with IB4-Biotin, were precipitated with Streptavidin-beads and subjected to western blotting. As shown, a specific band at the height of mature Ret protein was detected with both anti-Ret (Fig. 3b) and anti-pRet (Fig. 3c) antibodies only in transfected cells, thus confirming the ability of IB4 to bind the activated mutant receptors in the glycosylation state.
It has been reported the presence of physiologically stimulated Ret in adult rat brain indicating that Ret plays a physiological role in specific structures of the adult central nervous system (Colucci-D'Amato et al. 1996). We confirmed the presence of phosphorylated Ret in lysates from adult rat brain (Fig. 4ai) and demonstrated that in addition to recognize different proteins in total brain lysates (Fig. 4aii), IB4 binds to Ret receptor as revealed by immunoblot with anti-Ret antibodies and IB4-staining of total brain proteins immunoprecipitated with anti-Ret antibodies (Fig. 4aiii). Taken together the results clearly indicate Ret as one of the glycoproteins directly recognized by lectin IB4. In light of these findings, it is possible to speculate that the different population of nociceptive neurons that have been described to express Ret receptor and be responsive to GDNF neurotrophic effects in the DRG (Molliver et al. 1997; Leclere et al. 1998) and enteric system (Thacker et al. 2006), are selectively labeled by IB4 since the lectin IB4 is able to bind Ret receptor. A similar explanation might account for the IB4 labeling of a subpopulation of GDNF-responsive cells we recently described in the hippocampus (Boscia et al. 2009a). However, whether the other neuronal subtypes that express Ret and are responsive to GDNF are labeled by IB4 lectin remains to be ascertained.
Currently, IB4 is largely employed as a technically easy and reliable identifying tool for microglia at various stages of activation both in cultures and rat brain tissues (Colton et al. 1992; Boscia et al. 2009a). Indeed, the IB4 isolectin specifically recognizes microglial cells in the resting state under physiological conditions, or in the activated, ameboid or phagocytic states during pathological conditions, such as cerebral ischemia (Boscia et al. 2009b). On the basis of this observation, our findings suggested that Ret glycoprotein might represent one of the IB4-reactive glycoconjugate accounting for the IB4 stain in microglia under physiological and ischemic conditions. Immunofluorescence experiments performed with anti-pRet antibody revealed that Ret receptor is constitutively activated and intensely expressed in ramified cells resembling microglia-like morphology throughout the brain (Fig. 4b). Double-labeled experiments revealed that pRet was selectively co-expressed in the ramified OX-42-positive microglia (Fig. 4c).
Next, we explored whether IB4 staining also identified activated or ameboid Ret-expressing microglia under ischemic conditions. As already described, 3 days after pMCAO there was a significant increase in the number of ameboid microglia as well as in the intensity of IB4 staining within the ischemic core (Boscia et al. 2009b). Interestingly, in this region, a large number of ameboid microglial cells were intensely labeled with both pRet and IB4 staining (Fig. 4dvii–ix). Quantitative analysis of double-labeled cells in the ischemic core indicated that among Ret-positive cells 94 ± 4% colocalize with IB4 and among IB4-positive cells 77.5 ± 3.8% colocalize with Ret.
GDNF has been reported to provide potent neuroprotective effects against neuronal injury following excitotoxic insults, including cerebral ischemia (Wang et al. 1997). In this regard, an important role is played by GDNF-dependent Ret activation in microglia (Miyazaki et al. 2002; Boscia et al. 2009a).
Although microglial cells have been reported to act as scavenger cells or mediator of inflammatory responses in several neurodegenerative states (Saijo and Glass 2011), our findings support the hypothesis that Ret activation in microglia might provide a trophic signaling after brain damage in the post-ischemic brain.
In conclusion, our data indicate that Ret receptor is one of the IB4-reactive glycoconjugate accounting for the IB4 stain in microglia not only under physiological but also under ischemic conditions.
This study was supported by the following grants: COFIN 2008; Ricerca Sanitaria RF-FSL352059 Ricerca finalizzata 2006; Ricerca Oncologica 2006; Progetto Strategico 2007; Progetto Ordinario 2007 (all to L Annunziato), and from CNR, from AICR No 11-0075 (LC), MIUR grant, MERIT RBNE08YFN3_001 (VdF), AIRC No 11781 (LC), supported in part by the Compagnia di San Paolo and from the Italian Ministry of Economy and Finance to the CNR for the Project FaReBio di Qualità. CLE is recipient of a FIRC fellowship. The author thanks Dr Rosaria Gala for technical help with ischemia surgery in rats.