Neurochem Res (2013) 38:1285–1294 DOI 10.1007/s11064-013-1048-6
ORIGINAL PAPER
Differential Expression of Micro-Heterogeneous LewisX-Type Glycans in the Stem Cell Compartment of the Developing Mouse Spinal Cord Michael Karus • Eva Hennen • Dina Safina • Alice Klausmeyer • Stefan Wiese • Andreas Faissner
Received: 10 September 2012 / Revised: 10 April 2013 / Accepted: 12 April 2013 / Published online: 30 April 2013 Ó Springer Science+Business Media New York 2013
Abstract Complex glycan structures and their respective carrier molecules are often expressed in a cell type specific manner. Thus, glycans can be used for the enrichment of specific cell types such as neural precursor cells (NPCs). We have recently shown that the monoclonal antibodies 487LeX and 5750LeX differentially detect the LewisX (LeX) glycan on NPCs in the developing mouse forebrain. Here, we analysed the staining pattern of both antibodies during late embryonic mouse spinal cord development. At E13.5 both antibodies strongly label the central canal region. Along these lines they detect the LeX glycan primarily on Nestinpositive NPCs at that age. Moreover, we show that spinal cord NPCs cultured as free floating neurospheres display a high immunoreactivity to both antibodies. In that context, we also demonstrate that the 487LeX antibody can be used to deplete a subpopulation of neurosphere forming NPCs from a mixed E13.5 spinal cord cell suspension. Towards the end of embryogenesis the overall immunoreactivity to both antibodies increases and the staining appears very diffuse. However, the 5750LeX antibody still labels the central canal region. The increase in immunoreactivity correlates with an
Special Issue: Elisabeth Bock. M. Karus E. Hennen D. Safina A. Klausmeyer S. Wiese A. Faissner (&) Department for Cell Morphology and Molecular Neurobiology, Ruhr-University Bochum, Bochum, Germany e-mail:
[email protected] M. Karus E. Hennen D. Safina International Graduate School of Neuroscience, Ruhr-University Bochum, Bochum, Germany A. Klausmeyer S. Wiese Group for Molecular Cell Biology, Ruhr-University Bochum, Bochum, Germany
expression increase of the extracellular matrix molecules Tenascin C and Receptor Protein Tyrosine Phosphatase b/f, two potential LeX carrier proteins. In line with this, immunoprecipitation analyses confirmed Tenascin C as a LeX carrier protein in the embryonic mouse spinal cord. However, the immunoreactivity to both antibodies appears only to be marginally affected in the absence of Tenascin C, arguing against Tenascin C being the major LeX carrier. In conclusion our study gives some novel insights into the complex expression of LeX glycans and potential carrier proteins during the development of the mouse spinal cord. Keywords Complex glycans Extracellular matrix Neural stem cells Stem cell niche Tenascin-C Receptor protein tyrosinphosphatase-b/f
Introduction Complex glycan structures are expressed throughout the development of the central nervous system (CNS) and are known to influence the behaviour of neural precursor cells (NPCs) in terms of proliferation and differentiation [26]. For example, the complex DSD-1-chondroitinsulfate epitope recognized by the monoclonal antibody 473HD has been identified as a novel marker [24] that is involved in NPC-proliferation and differentiation [20, 21] and up-regulated in response to CNS-lesion [7, 19]. The LewisX glycan (LeX; also known as stage specific embryonic antigen-1 or CD15) represents a well-known glycan that consists of a galactose b1-4-linked and fucose a1-3-linked to N-acetylglucosamine. LeX is primarily expressed by NPCs during early stages of CNS development and in adulthood [3, 4, 11, 15]. Meanwhile, several LeX carrier proteins such as Tenascin C (Tnc), Phosphacan, L1-CAM,
123
1286
and b1-Integrin have been identified [1, 11, 12, 22, 25]. Since the LeX glycan is expressed on the cell surface of NPCs, it can be used as a biomarker for the selective enrichment of e.g. NPCs from mixed cell suspensions. Indeed, we recently demonstrated that the LeX-specific monoclonal antibody 5750LeX can be used to enrich for multipotent NPCs from E14.5 telencephalic cell suspensions [11]. Moreover, the 5750LeX and the 487LeX, yet another LeX-specific monoclonal antibody, can be applied for the immuno-depletion of neurosphere forming cells [11]. However, when analysed by glycan-array analysis both antibodies exhibit different affinities to the LeX glycan depending on the special context of the glycan structures [11]. In line with this, the immunoreactivity to these antibodies within the developing forebrain overlaps only to some extent [11]. Here, we investigated the expression of the LeX glycan throughout mouse spinal cord development and in spinal cord derived NPC cultures. We show that the 5750LeX and the 487LeX antibodies used in this study differentially label both spinal cord tissue and freshly dissociated spinal cord cells, suggesting that both antibodies can not only be used to selectively enrich for distinct cell populations of the embryonic forebrain, but also of the spinal cord. Moreover, we identified the glycoprotein Tnc as a novel LeX carrier protein in the embryonic spinal cord.
Materials and Methods Animals Experiments were performed according to international rules using either timed-pregnant wildtype NRMI or Tncmutant animals [10]. The genotype of Tnc-mutant animals was determined as described [23]. All animals were kept under standard housing conditions. The age of the embryos was determined following the Theiler Stages, and the day of the vaginal plug was considered as E0.5. Immunological Reagents In the following the primary antibodies and the respective dilutions used in this study are listed. The monoclonal antibodies were: anti-Nestin (1:500; mouse IgG; clone rat-401; Chemicon, Hofheim, Germany), anti-a-Tubulin (1:10,000; mouse IgG; clone DM1a; Sigma-Aldrich), anti-bIII Tubulin (1:500; mouse IgG; clone SDL3D10; Sigma-Aldrich), antiLewisX (1:300; rat IgM; clone 487, [22]), anti-LewisX (1:300; rat IgM; clone 5750, [11]), monoclonal antibody 4860 (1:100; rat IgM [5]). The polyclonal antibodies were: antiRPTPb/f (1:1,000; rabbit, batch Kaf13/5) [8], anti-Tnc (1:2,000; rabbit, batch Kaf14/1 [9]). The anti-Tnc antibody
123
Neurochem Res (2013) 38:1285–1294
used in this study is specific for tenascin-C and does not react with other tenascin proteins [6]. Immunohistochemistry Pregnant mice were sacrificed by cervical dislocation, and the embryos were immediately removed and fixed in 4 % (w/v) paraformaldehyde (PFA) at 4 °C. After fixation the embryos were transferred to 20 % (w/v) sucrose in DEPCtreated water for cryoprotection. Finally, the tissue was embedded in Tissue Tec Freezing Medium (Jung, Nussloch, Germany) and cut into 16 lm thin sections on a cryostat CM3050S (Leica, Solms, Germany). The cryosections were rehydrated and blocked for 1 h at room temperature (RT) with PBT1 (PBS ? 1 % (w/v) BSA ? 0.1 % (v/v) Triton X-100)/1.7 % (w/v) NaCl-PBS (PBS ? 0.9 % (w/v) NaCl) (1:1) ? 10 % (v/v) normal goat serum. Next, the incubation with the primary antibodies diluted in PBT1 ? 5 % (v/v) normal goat serum was carried out overnight at 4 °C. The next day, the sections were washed three times with PBS and subsequently incubated with species-specific antibodies coupled with either Cy2 (1:250) or Cy3 (1:500) (Dianova, Hamburg, Germany) diluted in PBS/A (PBS ? 0.1 % (w/v) BSA) for 3 h at RT. In order to visualize the cell nuclei, Hoechst 33528 (Sigma) was included (diluted 1:105 in PBS) into the secondary antibody solution. Finally, the sections were washed three times with PBS and mounted with ImmuMount (Invitrogen). Immunocytochemistry The analysis of cell markers followed established protocols [18]. After removal of the culture medium adherent cells were briefly washed twice with PBS/A. For the detection of the LeX-epitope the incubation with the primary antibody diluted in PBS/A was directly carried out for 30 min at RT. Then, the cells were washed again three times with PBS/A and fixed with 4 % (w/v) PFA for 10 min at RT. To detect intracellular epitopes, the fixation was performed prior to the incubation with the primary antibodies diluted in PBT1. After incubation with the primary antibody, the cells were washed three times with PBT1 and the incubation with either Cy3- or Cy2- (1:500)-coupled species specific secondary antibodies (Dianova) diluted in PBS/A was carried out at RT for 30 min. Hoechst 33528 (1:105) was additionally added to visualize the nuclei. Finally, the cells were washed twice with PBS and mounted in PBS/glycerine (2:1). Neurosphere Culture Cultivation of primary mouse spinal cord NPCs was performed as previously described [13, 14]. In brief, E13.5 lumbar spinal cord tissue was enzymatically dissociated
Neurochem Res (2013) 38:1285–1294
using papain for 30 min. The enzymatic reaction was stopped by adding an equal amount of ovomucoid. Then, the tissue was gently triturated, and the cells were pelleted by centrifugation for 5 min at 200g. The pellet was resuspended in neurosphere medium (DMEM/F12 (1:1), 2 % B27, 1 % L-glutamine, 1 % penicillin/streptomycin). Finally, 1250 cells were plated per ml medium in the presence of 10 ng/ml FGF2, 10 ng/ml EGF, and 0.25 U/ml heparin. The cells were kept at 37 °C and 6 % CO2 for 1 week. After that, the number of neurospheres was determined by counting all neurospheres in the culture flask. Western-Blot Analysis Spinal cord tissue or neurosphere cells were homogenized and solubilized by mechanical agitation in 4 °C cold cell lysis buffer (50 mM Tris–HCl pH 7.4, 150 mM NaCl, 5 mM EDTA, 5 mM EGTA, 1 % (v/v) Triton X-100, 0.1 % (v/v) Na-desoxycholate, 0.1 % (v/v) SDS, 1 mM Na3VO4, 1 mM NaF) and incubated for further 30 min on ice. Tenascin-C was isolated from mouse P7 to P14 brains and affinity purified using a J1/tn2 antibody column [16]. The protein concentration was determined using a protein quantification kit according to the manufacturer’s instructions. 10 lg protein samples were loaded on an 8 % (v/v) polyacrylamide gel or for purified Tnc 10 ng was used for a 6-8 % (v/v) polyacrylamide gradient gel and transferred to a PVDF membrane under semi-dry conditions for 2 h with approximately 1.5 mA/cm2. Immunoprecipitation experiments were performed as previously described [11]. After the transfer the membrane was blocked with 5 % (w/v) skim milk powder in TBS for 1 h at RT. The primary antibodies were diluted in 5 % (w/v) skim milk powder in TBS/T (TBS ? 0.05 % (v/v) Tween-20) and incubated at 4 °C overnight. Subsequently, the membrane was washed three times with TBS/T for 10 min, and the incubation with HRP-coupled secondary antibodies diluted in 5 % (w/v) skim milk powder in TBS/T was carried out at RT for 1 h. Finally, the membrane was washed again three times with TBS, and the signal was detected using enhanced chemiluminescence reagent and X-ray films. Immunodepletion Assay Complement-mediated immunocytolysis was performed as previously described [11]. In brief, acutely dissociated cells were diluted to a concentration of 1,000 cells/ll in MEM containing 0.2 % BSA and incubated with guinea-pig complement (Linaris, Wertheim, Germany) and mAb 487LeX for 1 h. As a control, an aliquot of cells was only incubated with the guinea-pig complement. The quality of immunodepletion for mAb 487LeX positive cells was assessed by immunhistochemistry.
1287
Statistics All data are given as mean plus/minus SD. Statistical analyses for the results of the immunodepletion experiments were performed using the paired two-tailed Student’s T Test and the p-values are given as * (p \ 0.05), ** (p \ 0.005), *** (p \ 0.001).
Results We recently demonstrated that the use of different anti-LeX antibodies detecting the LeX-glycan in the developing brain results in different staining patterns. Moreover, these antibodies exhibit different affinities to diverse LeX containing glycans in a glycan array [11]. Based on these observations we compared the expression of the 487LeX- and 5750LeX-epitope during mouse embryonic spinal cord development, reasoning that both epitopes might display differential expression patterns also during spinal cord development. Immunohistochemical analyses revealed a general up-regulation of both epitopes between E13.5 and E18.5 (Fig. 1a–f). However, while the 487LeX antibody primarily labels the central canal region and the ventral mantle zone at E13.5 (Fig. 1a), the 5750LeX antibody strongly binds to the central canal region at that stage (Fig. 1b). At E15.5 both antibodies strongly label great portions of the spinal cord (Fig. 1c, d). Note that the 5750LeX antibody still binds to the central canal region, while the 487LeX only weakly labels this area (Fig. 1c, d). At E18.5 both epitopes share a similar expression pattern within the spinal cord, although the 5750LeX antibody still labels the central canal region to some extent (Fig. 1e, f). Next, we investigated the expression levels of both epitopes via Western-blot analysis in detergent extracts derived E12.5–P10 spinal cord tissue. Alpha-Tubulin served as loading control. Consistent with the immunohistochemical data, the Westernblot analysis revealed a general increase in the expression levels with ongoing development. Interestingly, while the 487LeX antibody recognizes an additional epitope at around 100 kDa at later embryonic and postnatal stages (E18.5; P6; P10), the 5750LeX antibody does not (Fig. 1g, h). Additionally, we performed immunoprecipitation experiments on E13.5 spinal cord and brain protein extracts using the 5750LeX antibody. The subsequent Western-blot analysis demonstrated that Tnc is a LeX carrier protein in both the embryonic spinal cord and brain (Fig. 1i). Moreover, the 5750-IP-fraction displayed a high immunoreactivity to the 487LeX antibody. Immunoprecipitation using the isotype-matched 4860 antibody that recognizes a glycolipid proved the specificity of our immunoprecipitation experiments (Fig. 1j). Next, we analysed the expression of both epitopes on a single cell level. For that purpose we dissociated embryonic spinal cord tissue and plated single cells on a poly-DL-Ornithine
123
1288
Neurochem Res (2013) 38:1285–1294
Fig. 1 LewisX expression during late embryonic mouse spinal cord development and Tnc (a–f) Frontal spinal cord sections labelled against the LeX-glycan using the monoclonal antibodies 487LeX (a, c, e) and 5750LeX (b, d, f). Between E13.5 and E18.5 the overall LeX expression increases and changes from a locally defined signal at the central canal towards a more diffuse staining throughout the spinal cord. At E13.5 both antibodies label the central canal region. However, while the 487LeX-immunoreactivity around the central canal drops towards the end of embryogenesis, the 5750LeXimmunoreactivity in this area is still visible at E18.5. The sections were counterstained with Hoechst, in order to visualize the nuclei. Scale bar: 200 lm. g, h) Western-blot analysis of spinal cord detergent extracts (10 lg/lane), showing the overall increase of LeX expression throughout spinal cord development. The usage of either antibodies results in a double band above 250 kDa. Yet, the 487LeX-antibody additionally recognizes a LeX-glycan at around 100 kDa. i, j Western blot of immunoprecipitates obtained from E13.5 mouse spinal cord, of E13.5 mouse brain extracts and of affinity column-purified Tnc (10 ng) showing decoration of Tenascin-C (revealed with polyclonal Tnc-antibody) with 5750LeX (i) and copresence of the 5750LeX- and the 487LeX-epitopes within the same immunoprecipitation sample (j). The Mab 4860 detects a glycolipid expressed in oligodendrocyte membranes [5] and was used as an independent antibody control for the immunoprecipitations. Input lanes in (i) show the amounts of Tnc in the protein extracts
substrate for 2 h. Thereafter, the cells were immunocytochemically analysed. First we determined the relative number of 487LeX–/5750LeX-positive cells between E13.5 and
123
E18.5. While the percentage of 487LeX-positive cells was 12.4 ± 3.4 % (n = 4) at E13.5 (Fig. 2a), the percentage of 5750LeX-positive cells was only 5.6 ± 1.1 % (n = 4)
Neurochem Res (2013) 38:1285–1294
Fig. 2 Cell-type specific expression of the LeX-glycan in the embryonic mouse spinal cord (a, b) Freshly dissociated E13.5 spinal cord cells stained against the LeX-glycan and the marker proteins Nestin (NPCs) and bIII-Tubulin (young neurons). Quantification of 487LeX- and 5750LeX-positive cells throughout late embryonic mouse spinal cord development revealed an increase in LeX-expressing cells between E13.5 and E18.5. c, d, e, f At E13.5 both antibodies detect the LeX-glycan on NPCs. Some bIII-Tubulin-positive neurons also stain positive for the 487LeX-epitope but not for 5750LeX (see f). Cells were counterstained with Hoechst, in order to visualize the nuclei. Scale bar: 25 lm. Quantifications in a and b were performed from three independent experiments with at least 100 cells per experiment
(Fig. 2b). Within the next 2 days the 487LeX-positive population slightly increased to 16.1 ± 6.8 % (n = 4) (Fig. 2a). In contrast the 5750LeX-positive population nearly doubled to 10.6 ± 2.1 % (n = 4) (Fig. 2b). At E18.5 the relative numbers of 487LeX- and 5750LeX-positive cells were similar [487: 25.7 ± 8.4 (n = 4); 5750: 22.5 ± 4.3 (n = 4)] (Fig. 2a, b). Using cell type specific markers, we observed that the 487LeX antibodies primarily labelled Nestin-positive cells at E13.5 (Fig. 2c). However, we also documented some bIII-Tubulin-positive cells that co-expressed the 487LeXepitope (Fig. 2e). In contrast to the 487LeX antibody the
1289
5750LeX antibody exclusively labelled Nestin-positive NPCs at E13.5. We never observed any bIII-Tubulin-positive neurons that stained also positive for the 5750LeX-epitope (Fig. 2d, f). These data demonstrate that both epitopes are mainly expressed in the precursor cell compartment. Since we observed a strong expression of both epitopes in the NPC compartment, we next analysed, whether these epitopes are also enriched in spinal cord NPCs cultures. For that purpose we cultivated E13.5 spinal cord NPCs as free floating neurospheres in the presence of the commonly used mitogens FGF2 and EGF. After 1 week neurospheres of about 200 lm diameter were easily detectable and were further processed for either immunohistochemical or Western-blot analysis. The immunohistochemical analysis revealed a strong expression of both epitopes especially at the outer rim of the neurospheres (Fig. 3a, b). Western-blot analysis of neurosphere detergent extracts confirmed the association of both epitopes with proteins expressed in spinal cord NPC cultures (Fig. 3c). Note, that the 487LeXantibody again recognized an epitope at 100 kDa. In contrast, the 5750LeX-antibody did not (Fig. 3c). The strong immunoreactivity for the 487LeX antibody within the NPC compartment prompted us to examine, whether the 487LeX could be used to deplete the neurosphere forming cell population in dissociated E13.5 spinal cord cell suspensions via immuncytolysis. Therefore, we incubated freshly dissociated E13.5 spinal cord cells with the 487LeX antibody in the presence and absence of a commercially available immune-complement. To verify the efficient immuncytolysis, we plated some cells on a poly-DL-ornithine substrate for 2 h and determined the relative number of 487LeX-positive cells. In the control situation several 487LeX-positive cells were easily detectable. In contrast, the addition of the complement effectively reduced the number of 487LeX-positive cells (Fig. 3d, e). In parallel experiments we plated control and immuno-depleted cells under clonal conditions and monitored the neurosphere formation capacity in the presence of EGF and FGF2. In the control condition several neurospheres formed after 1 week. However, the immunodepleted cell population gave rise to significantly less neurospheres after 1 week [control: 117.7 ± 14.7; complement 58.7 ± 11.6 (n = 3; p = 0.0055)] (Fig. 3f–h). In order to get a first idea of potential LeX-carrier proteins during spinal cord development, we analysed the overall expression of the known LeX-carrier proteins RPTPb/f and Tnc. We found that both proteins are only weakly expressed between E12.5 and E13.5 and became up-regulated towards the end of embryogenesis (Figs. 2e, 4a, b), reflecting the temporal expression of the LeX-glycan. Thus, we next asked, whether the immunoreactivity for the 487LeX antibody or the 5750LeX antibody might be altered in the Tnc-deficient E15.5 spinal cord. However, we did not observe any changes in the
123
1290
Neurochem Res (2013) 38:1285–1294
Fig. 3 Expression of the LeX-glycan in spinal cord NPC cultures (a, b) Photomicrographs of neurosphere sections stained against the LeXglycan using the 487LeX- and the 5750LeX-antibody. Both antibodies strongly labelled the outer rim of the neurospheres. In contrast the neurosphere core displayed only a weak LeX expression level. The sections were counterstained with Hoechst, in order to visualize the nuclei. c Western-blot analysis of neurosphere detergent extracts, showing a strong expression of both epitopes in spinal cord NPC cultures. Again, the 487LeX-antibody detects an additional band around 100 kDa. a-Tubulin served as loading control. d, e Freshly dissociated spinal cord cells under control conditions and after immuno-depletion
of 487LeX-expressing cells, showing successful ablation of 487LeXpositive cells. The cells were counterstained with Hoechst, in order to visualize the nuclei. f, g Neurospheres grown in the presence of EGF and FGF2 from E13.5 spinal cord dissociated cells. In the control condition the cells gave rise to several neurospheres of about 100–150 lm size (f). After ablation of 487LeX-expressing cells, the number of neurospheres in culture was strongly reduced (g). h Quantification of neurospheres (from 3 independent experiments) grown under clonal conditions for 1 week in culture revealed a significant decrease after immuno-depletion. Scale bar: 25 lm (a, b); 50 lm (d, e); 100 lm (f, g). Significance is indicated with **p \ 0.01
immunoreactivity and in Western blot when comparing Tnc-deficient and wildtype E15.5 littermates (Fig. 4c–g). As there were only little if any changes in immunoreactivity for the LeX epitopes in Tnc-deficient animals we next asked whether RPTPb/f and its isoforms could be one carrier of LeX epitopes during embryonic development. We further analysed E15.5 as a representative stage and found that the immunreactivities for RPTPb/f, phosphacan appear to partially overlap with the immunreactivities for the LeX epitopes 487LeX and 5750LeX (Fig. 5a). In contrast to this observation we found only some overlap between the immunoreactivities for Tnc and the 5750LeX-epitope (Fig. 5b), although our immunoprecipitation experiments
clearly identified Tnc as a LeX-glycan carrier (Fig. 1i). We next used spinal cord sections from E15.5 and co-labeled for RPTPb/f and the respective LeX epitopes 487LeX and 5750LeX and found that there are possibly overlapping areas around the central canal and the future white matter (Fig. 5c, d).
123
Discussion In this study we convey an expression analysis of the LeXglycan and some of its known carrier proteins throughout
Neurochem Res (2013) 38:1285–1294
1291
Fig. 4 Expression of the LeX carrier proteins RPTPb/f and Tenascin C throughout mouse spinal cord development. a, b Western-blot analyses of mouse spinal cord protein extracts, demonstrating an overall increase in the expression levels of the known LeX carrier proteins RPTPb/f and Tnc between E12.5 and P10. c–f Wildtype and Tncdeficient frontal E15.5 spinal cord sections stained against the LeX-glycan using the 487LeXand 5750LeX-antibody. The loss of the LeX carrier protein Tnc does not affect the immunoreactivity for both antibodies. The sections were counterstained with Hoechst, in order to visualize the nuclei. g 5750LeX epitope detection in a Western-blot of E13.5 spinal cord detergent extracts from Tnc ?/?, Tnc ?/- and Tnc -/- embryos. Scale bar in C: 100 lm
the development of the mouse spinal cord. Moreover, we demonstrate that the usage of the two LeX-specific monoclonal antibodies 487LeX and 5750LeX results in
similar yet not completely congruent LeX-glycan expression patterns. We also show that the LeX-glycan is primarily expressed within the NPC compartment in vivo and
123
1292
Neurochem Res (2013) 38:1285–1294
Fig. 5 Comparative expression analysis of RPTPb/f and the LeX-glycan in the E15.5 mouse spinal cord. a Western blot analysis of E15.5 mouse spinal cord protein extracts showing that the LeX-epitopes 487LeX and 5750LeX co-migrate with RPTPb/f and Phosphacan. b 5750LeX-immunoreactivity only marginally overlaps with Tnc-immunoreactivity at E15.5. Immunoreactive bands primarily migrate in different Mr-ranges. c Coimmunohistochemical staining for RPTPb/f (green) and 5750LeX (red). The overlap appears in yellow colour. d Co-immunohistochemical staining for RPTPb/f (green) and 487LeX (red). The overlap appears in yellow colour. Scale bar: 100 lm
in vitro. Using the 487LeX antibody we successfully depleted a subpopulation of neurosphere forming cells from a mixed E13.5 mouse spinal cord cell suspension. Finally, we identified Tnc as a novel LeX carrier protein in the embryonic mouse spinal cord. Recent studies have already reported on the expression of the LeX glycan in the embryonic and adult mouse spinal cord [15, 17]. However, our study extends these findings, since it is the first to give a comprehensive overview of the temporal expression of this particular glycan during the development of the mouse spinal cord. With regard to its cell-type specificity the LeX-glycan is primarily expressed by multipotent NPCs in the developing and adult CNS [3, 4, 11, 15]. However, it can also be found on more restricted progenitor cells such as oligodendrocyte precursor cells [11] or even on postmitotic neurons (this study, and a
123
previous report [2]). Interestingly, different LeX-specific antibodies differentially label these cell populations [11]. Therefore, it appears feasible to either specifically enrich or to deplete for distinct cell types with the help of LeXspecific antibodies. As a proof of principle we used the 487LeX antibody to eliminate the neurosphere forming cells from a mixed E13.5 cell population. Surprisingly, in contrast to our previous study using cortical cells [11], we did not observe a full reduction in neurosphere formation capacity after immuno-depletion of spinal cord cells. This can either indicate an incomplete immuno-depletion or hint towards neurosphere forming NPC subpopulations. A previous study, that provided evidence for both a LeX-positive and a LeX-negative NPC population during embryonic mouse spinal cord development, speaks in strong favor of the latter explanation [15]. For future
Neurochem Res (2013) 38:1285–1294
studies it would be of great interest to further characterise these subpopulations. The usage of antibodies for the detection of cell surface antigens such as the LeX-glycan is not only interesting for the isolation and characterisation of distinct cell populations from a developing tissue. Antibodies recognizing cell type specific markers on the cell surface might have important biomedical applications as well. For example, current stem cell differentiation protocols do not yield 100 % pure cell populations, leaving a risk for tumour formation. Thus, the knowledge of the cell-type specific expression of cell surface glycans and the application of glycan specific antibodies may help to circumvent these problems. With respect to the epitope identity we previously showed that the two LeX-specific monoclonal antibodies 487LeX and 5750LeX detect distinct subtypes of LeX containing glycans. Moreover, both antibodies differentially label sections of the developing forebrain as well as different cell populations in mixed murine glial cultures [11]. In the developing spinal cord we also observed a partially overlapping immunoreactivity for both antibodies. While the 487LeX antibody only labelled the central canal region until the end of neurogenesis (E13.5), the 5750LeX antibody labelled this region at the beginning of gliogenesis as well (E15.5–E18.5). This indicates that these antibodies might detect the LeX-glycan on different carrier proteins and therefore target distinct subpopulations of cells. The fact that the usage of both antibodies in the Western-blot analysis yields similar but not identical results in terms of protein bands supports this interpretation. Looking on the overall LeX expression patterns we noticed two aspects. First, the expression of the LeX-glycan appears rather defined at early stages of spinal cord development (Fig. 1c, d). Second, the expression of this glycan does not only increase, its localisation also becomes increasingly diffuse with ongoing maturation. While a prominent expression of membrane bound molecules (e.g. b1 Integrin) likely accounts for the rather defined pattern at early developmental stages, the shift towards a more diffuse pattern presumably reflects spatio-temporal expression changes of known LeX carrier molecules secreted into the extracellular space. Along these lines, we observed an increase of the known secreted LeX carrier proteins Tnc and Phosphacan, a secreted splice variant of the receptor protein tyrosine phosphatase RPTPb/f. With respect to RPTPb/f we documented a strong overlap with the LeX-glycan in the E15.5 spinal cord. Moreover, RPTPb/f appeared to co-segregate with the LeX-glycan in our Western-blot analyses. However, a clear evidence based on e.g. immunoprecipitation analyses for RPTPb/f being a major LeX carrier protein at that age is still missing. Regarding Tnc our co-immunoprecipitation experiments revealed that Tnc is a LeX-carrier protein in the embryonic mouse spinal cord, at least to some extent. Yet, since the immunoreactivity for the LeX-glycan
1293
does not appear to be notably changed in the absence of Tnc, it is questionable whether Tnc is really the main LeX carrier, at least at the time point investigated. In line with this the main immunoreactivities for Tnc and the 5750LeX-epitope overlap only to some degree in the Western-blot analysis. In addition, we previously showed that Tnc is, except for the central canal region, widely expressed at E18.5 in the embryonic spinal cord [13]. However, since the 5750LeX antibody labels this region at this time point, whereas the 487LeX does not, the 5750LeX likely detects the LeX-glycan on a different carrier molecule in this region, again arguing against Tnc as the main 5750LeX-epitope carrier. Thus, to further address the identity of the major LeX carrier, analyses of e.g. RPTPb/f mutant animals would be necessary. Moreover, it is also conceivable that an as yet unidentified molecule, possibly either a protein or even a lipid might be the most prominent LeX carrier in the developing mouse spinal cord. At this point, the answer to this question remains a subject for further studies. Taken together our study adds new aspects concerning the expression of the LeX-glycan and its carrier proteins during embryonic and early postnatal spinal cord development. These data can be used as a solid basis for the identification of LeX carrier molecules within the spinal cord and for further analyses on the functional role of LeX-type glycans for NPC behaviour during spinal cord development. Acknowledgments MK, EH and DS were supported through the PhD program of the International Graduate School of Neuroscience (IGSN), the Research School at the Ruhr-University supported by the DFG (GSC 98/1), and MK received additional support from the Wilhelm and Gu¨nther Esser Foundation. We thank Ann-Kathrin Urbanowitz for kind help. We gratefully acknowledge grant support to AF from the Ruhr-University (President’s special programme call 2008) and from the Federal Ministry of Research and Education (BMBF, Grant 01GN0504).
References 1. Allendoerfer KL, Magnani JL, Patterson PH (1995) FORSE-1, an antibody that labels regionally restricted subpopulations of progenitor cells in the embryonic central nervous system, recognizes the Le(x) carbohydrate on a proteoglycan and two glycolipid antigens. Mol Cell Neurosci 6:381–395 2. Brito C, Escrevente C, Reis CA, Lee VM, Trojanowski JQ, Costa J (2007) Increased levels of fucosyltransferase IX and carbohydrate Lewis(x) adhesion determinant in human NT2 N neurons. J Neurosci Res 85:1260–1270 3. Capela A, Temple S (2002) LeX/ssea-1 is expressed by adult mouse CNS stem cells, identifying them as nonependymal. Neuron 35:865–875 4. Capela A, Temple S (2006) LeX is expressed by principle progenitor cells in the embryonic nervous system, is secreted into their environment and binds Wnt-1. Dev Biol 291:300–313 5. Czopka T, Hennen E, von Holst A, Faissner A (2009) Novel conserved oligodendrocyte surface epitope identified by monoclonal antibody 4860. Cell Tissue Res 338:161–170
123
1294 6. Czopka T, Von Holst A, Schmidt G, Ffrench-Constant C, Faissner A (2009) Tenascin C and tenascin R similarly prevent the formation of myelin membranes in a RhoA-dependent manner, but antagonistically regulate the expression of myelin basic protein via a separate pathway. Glia 57:1790–1801 7. Dobbertin A, Rhodes KE, Garwood J, Properzi F, Heck N, Rogers JH, Fawcett JW, Faissner A (2003) Regulation of RPTPbeta/phosphacan expression and glycosaminoglycan epitopes in injured brain and cytokine-treated glia. Mol Cell Neurosci 24:951–971 8. Faissner A, Clement A, Lochter A, Streit A, Mandl C, Schachner M (1994) Isolation of a neural chondroitin sulfate proteoglycan with neurite outgrowth promoting properties. J Cell Biol 126: 783–799 9. Faissner A, Kruse J (1990) J1/tenascin is a repulsive substrate for central nervous system neurons. Neuron 5:627–637 10. Forsberg E, Hirsch E, Frohlich L, Meyer M, Ekblom P, Aszodi A, Werner S, Fassler R (1996) Skin wounds and severed nerves heal normally in mice lacking tenascin-C. Proc Natl Acad Sci USA 93:6594–6599 11. Hennen E, Czopka T, Faissner A (2011) Structurally distinct LewisX glycans distinguish subpopulations of neural stem/progenitor cells. J Biol Chem 286:16321–16331 12. Hennen E, Faissner A (2012) LewisX: a neural stem cell specific glycan? Int J Biochem Cell Biol 44:830–833 13. Karus M, Denecke B, Ffrench-Constant C, Wiese S, Faissner A (2011) The extracellular matrix molecule tenascin C modulates expression levels and territories of key patterning genes during spinal cord astrocyte specification. Development 138:5321–5331 14. Karus M, Samtleben S, Busse C, Tsai T, Dietzel ID, Faissner A, Wiese S (2012) Normal sulfation levels regulate spinal cord neural precursor cell proliferation and differentiation. Neural Dev 7:20 15. Kelly TK, Karsten SL, Geschwind DH, Kornblum HI (2009) Cell lineage and regional identity of cultured spinal cord neural stem cells and comparison to brain-derived neural stem cells. PLoS ONE 4:e4213 16. Rigato F, Garwood J, Calco V, Heck N, Faivre-Sarrailh C, Faissner A (2002) Tenascin-C promotes neurite outgrowth of embryonic hippocampal neurons through the alternatively spliced
123
Neurochem Res (2013) 38:1285–1294
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
fibronectin type III BD domains via activation of the cell adhesion molecule F3/contactin. J Neurosci 22:6596–6609 Sabourin JC, Ackema KB, Ohayon D, Guichet PO, Perrin FE, Garces A, Ripoll C, Charite J, Simonneau L, Kettenmann H, Zine A, Privat A, Valmier J, Pattyn A, Hugnot JP (2009) A mesenchymal-like ZEB1(?) niche harbors dorsal radial glial fibrillary acidic protein-positive stem cells in the spinal cord. Stem Cells 27:2722–2733 Sirko S, Akita K, Von Holst A, Faissner A (2010) Structural and functional analysis of chondroitin sulfate proteoglycans in the neural stem cell niche. Methods Enzymol 479:37–71 Sirko S, Neitz A, Mittmann T, Horvat-Brocker A, von Holst A, Eysel UT, Faissner A (2009) Focal laser-lesions activate an endogenous population of neural stem/progenitor cells in the adult visual cortex. Brain 132:2252–2264 Sirko S, von Holst A, Weber A, Wizenmann A, Theocharidis U, Gotz M, Faissner A (2010) Chondroitin sulfates are required for fibroblast growth factor-2-dependent proliferation and maintenance in neural stem cells and for epidermal growth factordependent migration of their progeny. Stem Cells 28:775–787 Sirko S, von Holst A, Wizenmann A, Gotz M, Faissner A (2007) Chondroitin sulfate glycosaminoglycans control proliferation, radial glia cell differentiation and neurogenesis in neural stem/ progenitor cells. Development 134:2727–2738 Streit A, Faissner A, Gehrig B, Schachner M (1990) Isolation and biochemical characterization of a neural proteoglycan expressing the L5 carbohydrate epitope. J Neurochem 55:1494–1506 Talts JF, Wirl G, Dictor M, Muller WJ, Fassler R (1999) TenascinC modulates tumor stroma and monocyte/macrophage recruitment but not tumor growth or metastasis in a mouse strain with spontaneous mammary cancer. J Cell Sci 112(Pt 12):1855–1864 von Holst A, Sirko S, Faissner A (2006) The unique 473HDChondroitinsulfate epitope is expressed by radial glia and involved in neural precursor cell proliferation. J Neurosci 26: 4082–4094 Yanagisawa M, Taga T, Nakamura K, Ariga T, Yu RK (2005) Characterization of glycoconjugate antigens in mouse embryonic neural precursor cells. J Neurochem 95:1311–1320 Yanagisawa M, Yu RK (2007) The expression and functions of glycoconjugates in neural stem cells. Glycobiology 17:57R–74R
