SEARCH

SEARCH BY CITATION

Keywords:

  • axonal transport;
  • CD166;
  • motor neuron;
  • neurite outgrowth;
  • neurotrophin receptor;
  • signaling endosome

Abstract

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

J. Neurochem. (2012) 121, 575–586.

Abstract

Cell adhesion molecules of the immunoglobulin superfamily (IgCAMs) have been shown to modulate growth factor signaling and follow complex trafficking pathways in neurons. Similarly, several growth factors, including members of the neurotrophin family, undergo axonal retrograde transport that is required to elicit their full signaling potential in neurons. We sought to determine whether IgCAMs that enter the axonal retrograde transport route co-operate with neurotrophin signaling. We identified activated leukocyte cell adhesion molecule (ALCAM), a protein involved in axon pathfinding and development of the neuromuscular junction, to be associated with an axonal endocytic compartment that contains neurotrophins and their receptors. Although ALCAM enters carriers that are transported bidirectionally in motor neuron axons, it is predominantly co-transported with the neurotrophin receptor p75NTR toward the cell body. ALCAM was found to specifically potentiate nerve growth factor (NGF)-induced differentiation and signaling. The extracellular domain of ALCAM is both necessary and sufficient to potentiate NGF-induced neurite outgrowth, and its homodimerization is required for this novel role. Our findings indicate that ALCAM synergizes with NGF to induce neuronal differentiation, raising the possibility that it functions not only as an adhesion molecule but also in the modulation of growth factor signaling in the nervous system.

Abbreviations used
ALCAM

activated leukocyte cell adhesion molecule

BSA

bovine serum albumin

CAM

cell adhesion molecule

DMEM

Dulbecco’s minimum essential medium

EDL

extensor digitorum longus

GDNF

glial cell-derived neurotrophic factor

HC

carboxy-terminal fragment of tetanus toxin

Ig

immunoglobulin

LAL

levator auris longus

MION

monocrystalline iron oxide nanoparticles

NGF

nerve growth factor

NMJ

neuromuscular junction

p75NTR

p75 neurotrophin receptor

PBS

phosphate-buffered saline

pERK1/2

phosphorylation of ERK1/2

pTrkA

phosphorylation of TrkA

SDS

sodium dodecyl sulfate

Trk

tropomyosin-receptor-kinase

Development of the nervous system requires differentiating neurons to grow axons over great distances following long- and short-range signals that are controlled by distinct guidance systems. In general, long-distance acting molecules, such as growth factors, are soluble and bind receptors on the neuronal surface. In contrast, short-range membrane-associated molecules interact with surrounding cells and extracellular matrix, allowing for growth through diverse and changing environments. However, the boundaries between long-range diffusible ligands and short-range contact-mediated factors have blurred. Cell adhesion molecules (CAMs) have increasingly been shown to participate in signaling cascades with classical receptors, or as ligands and receptors themselves (Maness and Schachner 2007; Schmid and Maness 2008). For example, the best characterized immunoglobulin (Ig) CAM, neural cell adhesion molecule, is a glial cell-derived neurotrophic factor (GDNF) co-receptor, and is involved in neurite outgrowth and neuron survival in cooperation with fibroblast growth factor signaling (Saffell et al. 1997; Paratcha et al. 2003).

Neurotrophins are released by target tissues and activate tropomyosin-receptor-kinase (Trk) and p75 neurotrophin receptor (p75NTR) to mediate neuronal growth and survival. Activated receptor complexes undergo long-range axonal transport, which is required to elicit their full survival response (Zweifel et al. 2005; Cosker et al. 2008). In motor and sensory neurons, this axonal compartment is specifically entered by an atoxic carboxy-terminal binding fragment of the tetanus toxin (HC) (Lalli and Schiavo 2002). This property has been exploited to purify these organelles using magnetic affinity chromatography (Fig. 1) and has allowed the determination of their composition via proteomic analysis (Deinhardt et al. 2006b). In addition to the neurotrophin receptors TrkB and p75NTR, and small GTPases controlling progression along the endosomal pathway (Deinhardt et al. 2006b), this strategy yielded several plasma membrane proteins, of which many were found to be CAMs (see Table 1). Considering the growing evidence that CAMs have diverse roles in growth factor signaling, we investigated whether the CAMs identified in the proteome of HC-containing carriers undergo retrograde transport with neurotrophins and their receptors, and whether or not they can modulate their signaling.

image

Figure 1.  ALCAM undergoes axonal retrograde transport in HC-containing endosomal carriers. (a) MIONs were activated with heterobifunctional cross-linkers, allowing the immobilization of HC or BSA via sulfhydryl groups (Deinhardt et al. 2006b). Modified MIONs underwent binding and internalization in motor neurons. Axonal carriers containing MIONs were purified from post-nuclear supernatants and their proteome analyzed by mass spectrometry (Table 1) (Deinhardt et al. 2006b) or western blot (b). A portion of ALCAM and p50/dynamitin present in the cell lysate (input) were associated with HC-MION beads, but not with BSA-coated beads (FT, flow through). (c) Consecutive frames from a movie of HC555 and αALCAM488 (scale bar, 10 μm). Arrowheads indicate a retrograde transport vesicle containing HC and ALCAM moving towards the cell body (black arrow). Asterisks indicate HC transport carriers that do not contain ALCAM. Squares mark stationary organelles that contain both HC and ALCAM (closed) or ALCAM alone (open). (d) The speed distribution profile of ALCAM carriers overlaps with that of HC (n = 183 ALCAM and n = 231 HC carriers, recorded in 38 axons from eight cultures). (e) The box and whisker plot shows that a higher proportion of αALCAM488-positive carriers pause when compared with HC-containing vesicles.

Download figure to PowerPoint

Table 1.  Plasma membrane proteins associated with the axonal retrograde transport compartment. Components of integrin, selectin and cadherin binding complexes are coloured in orange, green and yellow respectively. IgCAMs are coloured in blue and ALCAM, which has been investigated in this work, is in bold. Other cell membrane proteins identified are also shown. NCAM, CAR and Junction plakoglobin had already been confirmed by other techniques to associate with the HC-endosomeThumbnail image of

In this study, we focused on activated leukocyte cell adhesion molecule (ALCAM), known also as CD166, a classical IgCAM with two distal variable Ig domains, three constant type-2 Ig domains, followed by a transmembrane region and a short cytoplasmic tail (Swart 2002). ALCAM expression is developmentally regulated in a wide-range of tissues and has been associated with cells undergoing dynamic growth and migration (Swart et al. 2005). In the developing nervous system, ALCAM is involved in axonal outgrowth and pathfinding (Ott et al. 1998, 2001; Avci et al. 2004). However, research has mainly focused on its role as a substrate permissive to neurite elongation (Kawauchi et al. 2003; Avci et al. 2004). Our work reveals for the first time that ALCAM is not merely a substrate for neurite outgrowth, but participates in neurotrophin signaling. We found that ALCAM undergoes bidirectional movement in axons, and that it colocalizes with p75NTR during retrograde transport. Over-expression and down-regulation studies revealed a novel role for ALCAM in the potentiation of neurite outgrowth in PC12 cells, which occurs by enhancing TrkA phosphorylation in response to nerve growth factor (NGF).

Methods

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

Endosome purification

Amino-derivatized monocrystalline iron oxide nanoparticles (MION) (Moore et al. 1997) were coupled to cysteine-rich tagged HC (Deinhardt et al. 2006b) or bovine serum albumin (BSA; Sigma-Aldrich, Gillingham, Dorset, UK) (Fig. 1a). Briefly, 4 mg MIONs were incubated with 1 mM EDTA and 4 mM succinimidyl 4-[N-maleimidomethyl]cyclohexane-1-carboxylate (Thermo Fisher Scientific, Cramlington, Northumberland, UK) for 30 min at 22°C followed by 2 h at 4°C on a rotor wheel. Activated MIONs were purified on a PD10 column (GE Healthcare, Little Chalfont, Buckinghamshire, UK) and divided into two batches, which were labeled for 48 h at 4°C with 200 μg HC or BSA previously reduced with 1 mM Tris (2-carboxyethyl) phosphine (Pierce) for 30 min at 22°C. The reaction was blocked by addition of 10 mM β-mercaptoethanol overnight. Conjugated MIONs were purified on Sephacryl S100HR (GE Healthcare). Figure 1a shows a schematic of the procedure used for the purification of HC-MIONs endosomes (Deinhardt et al. 2006b). Ventral horn motor neuron cultures (Arce et al. 1999) were incubated with conjugated MIONs at 37°C in complete medium for 1 h. After cooling on ice, cells were scraped in Hank’s balanced salt solution pH 7.4 supplemented with a protease inhibitor cocktail (Roche Diagnostics, Burgess Hill, West Sussex, UK), centrifuged at 170 g for 5 min at 4°C and resuspended in breaking buffer (0.25 M sucrose, 10 mM HEPES–KOH, pH 7.2, 1 mM EDTA, 1 mM magnesium acetate and protease inhibitors). Neurons were passed 15 times through a cell cracker (18 μm clearance; European Molecular Biology Laboratories, Heidelberg, Germany) and clarified at 690 g for 10 min at 4°C. AS columns (Miltenyi Biotech, Bisley, Surrey, UK) were equilibrated with breaking buffer supplemented with 0.4% BSA, placed inside a SuperMACS II (Miltenyi Biotech), and loaded with the post-nuclear supernatant. After washing with breaking buffer, columns were removed from the magnetic field and elution was carried out with breaking buffer containing 300 mM KCl. Proteins were precipitated and analyzed by sodium dodecyl sulfate (SDS)–polyacrylamide gel electrophoresis using 4–12% gradient gels (Life Technologies, Paisley, UK).

Mass spectrometry and protein identification

After SDS–polyacrylamide gel electrophoresis, bands were excised from the Coomassie blue-stained gel and in-gel digestion was performed as previously described (Brun et al. 2007). Gel pieces were then sequentially extracted with 5% (vol/vol) formic acid solution, 50% acetonitrile, 5% (vol/vol) formic acid, and acetonitrile. After drying, the tryptic peptides were resuspended in 0.5% aqueous trifluoroacetic acid. Samples were injected into a CapLC nanoLC system (Waters, Elstree, Hartfordshire, UK) and first pre-concentrated on a 300 μm × 5 mm pre-column (PepMap C18; Dionex, Camberley, Surrey, UK). The peptides were then eluted onto a C18 column (75 μm × 150 mm; Dionex). Chromatographic separation was performed using a gradient transition from solution A (2% acetonitrile, 98% water and 0.1% formic acid) to solution B (80% acetonitrile, 20% water and 0.08% formic acid) over 60 min at a flow rate of 200 nL/min. The LC system was directly coupled to a mass spectrometer (QTOF Ultima; Waters). MS and MS/MS data were acquired and processed automatically using MassLynx 4.0 software (Waters). Database searching was performed using MASCOT 2.1 software (Matrix Science). Two databases were used: an in-house list of well-known contaminants (keratins and trypsin) and an updated compilation of the SwissProt and Trembl databases.

Axonal transport assays

Motor neurons were exposed to an antibody against the ALCAM ectodomain labeled using the Monoclonal Antibody Labeling Kit (Life Technologies) alone or in combination with an AlexaFluor-conjugated p75NTR antibody (Deinhardt et al. 2007), HC555 (Lalli and Schiavo 2002) or LysoTracker (Life Technologies) at 37°C for 30 min. Neurons were washed with imaging medium (Dulbecco’s minimum essential medium without phenol red, DMEM-), riboflavin, folic acid and penicillin/streptomycin, supplemented with 30 mM HEPES–NaOH pH 7.4), placed in a 37°C humidified chamber and an image taken every 5 s with a Zeiss LSM 510 confocal laser scanning microscope using a 63× 1.4 NA Plan-Apochromat oil-immersion objective (Lalli and Schiavo 2002). Tracking was performed on time lapse sequences using AQM software as previously described (Bohnert and Schiavo 2005).

Immunofluorescence

PC12 cells were fixed in 4% paraformaldehyde in phosphate-buffered saline (PBS), quenched with 50 mM NH4Cl pH 7.5, washed with PBS and mounted directly, or surface ALCAM was labelled with a goat antibody directed to ALCAM ectodomain and a corresponding fluorescent secondary antibody. Alternatively, neurons were incubated with a fluorescent antibody against the extracellular domain of the protein of interest, for example, ALCAM (R&D Systems, Abingdon, Oxfordshire, UK) or p75NTR (Deinhardt et al. 2007) at 37°C for 1–2 h, acid washed (Deinhardt et al. 2006a) and then processed for immunofluorescence.

Neurite outgrowth quantification

PC12 cells were maintained in DMEM (Life Technologies) supplemented with 7.5% horse serum, 7.5% fetal bovine serum, 4 mM glutamine, 1% penicillin/streptomycin and transfected with Lipofectamine 2000 (Life Technologies) according to manufacturer’s instructions. The transfection medium was changed after 4 h to differentiation medium (DMEM supplemented with 0.1% horse serum, 0.1% fetal bovine serum, 4 mM glutamine) with or without the appropriate differentiation stimulus [NGF (Life Technologies), 100 ng/mL, dibutyryl-cAMP (Sigma-Aldrich), 0.1 mM or 0.5 mM, GDNF (Life Technologies), 100 ng/mL in the presence of 10 μg/mL GDNF receptor alpha chimera (GFRα-Fc; R&D Systems)]. Human Fc (Hu-Fc), CD6-Fc or ALCAM-Fc chimeras (R&D Systems) were used at 1–10 μg/mL. For knockdown experiments, PC12 cells were treated with 1 μg of shRNA vector (CSHCTR001-LvmH1) or shRNA ALCAM clone 1 (RSH051336-1-LvmH1; OS377393, Genecopoeia, Source BioSciences, Nottingham, Nottinghamshire, UK) pre-mixed with 1.5 μl of Lipofectamine 2000 in Opti-MEM I Reduced Serum Medium, GlutaMAX (Life Technologies). After 5 h, cells were returned to normal medium for 48 h and then differentiated for 24 h in DMEM supplemented with 4 mM glutamine, 1% penicillin/streptomycin and 100 ng/mL NGF.

Cells were fixed in 4% paraformaldehyde in PBS at 1, 3, 5 or 7 days after transfection as indicated. Cells were processed for immunofluorescence and nuclei were stained with 1 : 2000 DRAQ5 (BioStatus, Shepshed, Leicestershire, UK). At least 10 random fields were selected per coverslip and an average of 48 cells were counted per condition. Cells were regarded as exhibiting neurite outgrowth if the longest neurite was at least twice the widest diameter of the soma. In the knockdown experiments optimum cell viability was obtained after 1 day NGF treatment, therefore we classified cells as exhibiting neurite outgrowth whose longest neurite was at least as wide as the soma.

Quantitative RT-PCR

PC12 cells were seeded in 6-well plates to 70% confluence in DMEM supplemented with 7.5% horse serum, 7.5% fetal bovine serum, 4 mM glutamine and transfected with shRNA vector and ALCAM shRNA as above. Cells were left in normal medium for 72 h or after 48 h recovery transferred to serum-free DMEM supplemented with 100 ng/mL NGF for 24 h. The medium was then removed from the cells and total RNA extracted as per Qiagen instructions (Crawley, West Sussex, UK). First-strand cDNA was generated using the Vilo kit (Life Technologies). The cDNA was then used in qPCR reaction with the following primers: rat ALCAM forward TGAGGAGTTCATGTTTTACTTACCA and reverse CGTCTGTCAGTGTGTAAGTGTTTG; rat tubulin forward CAGAGCCATTCTGGTGGAC and reverse GCCAGCACCACTCTGACC mixed with Fast SYBR® Green (Life Technologies) as instructed by manufacturer and then run on an Applied Biosystems 7900HT PCR.

Western blotting

HEK293 cells were seeded in a 6-well plate to 70% confluence in DMEM supplemented with 10% fetal bovine serum. Cells were transfected the following day as for PC12 cells and left to recover for 72 h. Cells were harvested in cold PBS, washed twice and then lysed in 10 mM Tris–HCl pH 8.0, 150 mM NaCl, 1 mM EDTA 1% NP40 supplemented with HALT™ Protease and Phosphatase Inhibitor cocktail (Thermo Fisher Scientific), vortex mixed at 4°C for 30 min followed by 15 min of centrifugation at 16 200 g. The supernatant was removed, protein concentration determined and 10 μg of lysate subjected to western blotting. The polyvinylidene difluoride membrane was probed with a goat anti-ALCAM antibody (R&D Systems) at 1 : 300 or anti-actin monoclonal antibody (Sigma AC-15) at 1 : 5000 dilution for loading control overnight, washed three times in TBST and incubated for a further 1 h in the appropriate horseradish peroxidase-conjugated secondary antibody (Dako, Ely, Cambridgeshire, UK). Blots were developed with ECL (GE Healthcare).

Signaling experiments

Cells starved of serum and neurotrophins for 5 h were stimulated with 100 ng/mL NGF for the indicated times, washed once with PBS prior to addition of 100 μL of high SDS-containing lysis buffer (2% SDS, 0.05 Tris–HCl pH 6.8) at 100°C, sonicated three times for 10 s and then analyzed by western blot. Quantification of phosphorylated proteins was carried out using ImageJ and normalized according to the total protein in the sample. The change in phosphorylation observed by western blot is likely to be under-estimated because of the limited transfection efficiency (∼40%).

Statistical analysis

Neurite outgrowth was analyzed using one-way anova followed by a post hoc Bonferroni’s test on selected groups or using an unpaired t-test. For the signaling experiments, predicted intensity values were generated using a model that removed experiment and time-trend effects allowing data to be pooled across all experiments and time points (Fig. 5b). anova was then used to determine whether the difference between Dendra2 and ALCAM-D predicted phosphorylation values were significant.

Results

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

ALCAM undergoes retrograde transport in HC-containing axonal carriers

To confirm the association of ALCAM with HC-containing axonal carriers, motor neurons were incubated with HC- or BSA-conjugated MIONs at 37°C and the endosomal compartment containing internalized MIONs was purified by magnetic affinity chromatography (Fig. 1a). The proteins present on organelles eluted from the magnetic column were then analyzed by western blot. As shown in Fig. 1b, ALCAM was found associated with the HC beads, but not with the BSA control pellet. Similarly, p50/dynactin, a component of the dynein motor complex (Fig. 1b), the neurotrophin receptors TrkB and p75NTR (Deinhardt et al. 2006b) and the small GTPases Rab5 and Rab7 (Deinhardt et al. 2006b) were also found specifically in the HC-MION pellet. This result confirmed the association between ALCAM and HC-containing endosomes, which are known to undergo axonal transport (Lalli and Schiavo 2002).

A fluorescent antibody directed towards the ALCAM extracellular domain (αALCAM488) was used to visualize the endocytosis and axonal transport of ALCAM (Fig. 1c and Figure S1). In spinal cord motor neurons, ALCAM undergoes bidirectional transport and a proportion of ALCAM positive puncta moving in the retrograde direction colocalize with HC (% overlap ± SD; 59.7% ± 11.4) (Fig. 1c; n = 3 independent experiments). Importantly, the direct conjugation of this antibody with fluorophores does not alter its ability to recognize ALCAM (Figure S1), and this staining is specific, because no signal is detected in motor neurons isolated from ALCAM knockout mice (Figure S2) (McKinnon et al. 2000; Weiner et al. 2004). As shown in Fig. 1d, the speed distribution of ALCAM is similar to that of HC-containing carriers, although ALCAM transport endosomes have an increased tendency to pause when compared with HC carriers. This is shown as a distinct peak at 0 μm/s in the ALCAM speed distribution profile (Fig. 1d; green line) and was confirmed by quantifying the proportion of carriers that pause during transport (Fig. 1e).

Further analysis of the properties of axonal transport in spinal cord motor neurons revealed that more than half of the ALCAM-positive carriers (59.8% ± 9.3) also contain the neurotrophin receptor p75NTR, as shown by a fluorescently labeled anti-p75NTR antibody (αp75NTR555) (Fig. 2a; n = 3 independent experiments). This high level of colocalization was also found in the cell body at later time points (Fig. 2b).

image

Figure 2.  ALCAM is co-transported with p75NTR and CAMs in non-acidic organelles. (a) Consecutive frames showing co-transport of ALCAM and p75NTR (arrowheads). (b) ALCAM (pseudocolored in white in the lower left panel) also colocalizes with p75NTR in vesicles in the cell body (white arrowheads). There is little or no colocalization of ALCAM with LAMP2, although rare LAMP2-positive endosomes that also contain p75NTR are visible (empty arrowhead). (c) LysoTracker and αALCAM488 are transported along axons toward the cell body (empty and white arrows, respectively) in different compartments. (d) αALCAM488 and LysoTracker are not found in the same endosomal organelles in the cell body. Scale bars, 10 μm.

Download figure to PowerPoint

In contrast, very limited overlap was observed between ALCAM or p75NTR, and LAMP2, a lysosome marker, suggesting that transported ALCAM is not targeted for degradation (Fig. 2b). Co-transport experiments with αALCAM488 and LysoTracker supported the conclusion that ALCAM does not enter a degradative route, because only a minor fraction of transported ALCAM colocalized with acidic compartments in axons (11.7% ± 12.6; Fig. 2c) and no colocalization with Lysotracker was observed in the cell body (Fig. 2d). These results demonstrate that ALCAM undergoes axonal transport in a non-acidic compartment, which contains neurotrophin receptors.

ALCAM and other CAMs share the same axonal transport compartment

Further characterization of ALCAM-containing endosomes revealed that ALCAM was transported together with canine adenovirus 2 (Fig. 3a). The uptake and axonal transport of this neurotrophic virus are mediated by the CAM family member coxsackievirus and adenovirus receptor (Salinas et al. 2010), which has been previously shown to associate with this transport compartment in neurons (Salinas et al. 2009). Partial overlap with neural cell adhesion molecule, another member of the CAM family found in the HC proteome (Table 1), was observed in the cell body (Fig. 3b).

image

Figure 3.  ALCAM and other CAMs share the same axonal transport compartment. Panel (a) shows consecutive frames from a transport experiment after incubation with αALCAM488, HC555 and CAV-2-Cy3. CAV-2 and HC-positive endosomes, moving towards the cell body contain or lack ALCAM (white or empty arrowheads respectively). (b) Internalized αALCAM488 and p75NTR colocalize with NCAM in motor neuron soma as revealed by an antibody directed to its intracellular domain (NCAMIc). ALCAM and p75NTR are found in organelles with (empty arrowheads) or without NCAM (white arrowheads). Scale bars, 10 μm. (c) ALCAM does not directly contribute to retrograde transport. Speed analysis performed in wild type, ALCAM+/− and ALCAM−/− motor neurons revealed no significant difference in the speed distribution profile of HC transport in the three genotypes (wild type: n = 92 carriers, 22 axons, 4 independent cultures; ALCAM+/−: n = 100 carriers, 17 axons, 3 independent cultures; ALCAM−/−: n = 101 carriers, 22 axons and 4 independent cultures).

Download figure to PowerPoint

However, ALCAM does not directly contribute to retrograde transport per se, because the speed distribution of fast axonal retrograde carriers containing HC (Fig. 3c), and by inference TrkB (Deinhardt et al. 2006b), p75NTR (Deinhardt et al. 2007), brain-derived neurotrophic factor (Deinhardt et al. 2006b) and Rab7 (Deinhardt et al. 2006b), was unaltered in motor neurons lacking ALCAM. As shown in Fig. 3c, the speed distribution profiles of retrograde transport carriers quantified in motor neurons isolated from wild type, heterozygous (ALCAM+/−) or homozygous (ALCAM−/−) ALCAM knockout mice (McKinnon et al. 2000; Weiner et al. 2004) were almost completely overlapping.

In summary, these results suggest that ALCAM undergoes axonal transport in an endosomal compartment containing CAMs and neurotrophin receptors, but is not required for the recruitment of molecular motors, nor directly modulates the transport properties of this axonal retrograde transport compartment.

ALCAM potentiates NGF-induced neurite outgrowth

Based on the presence of neurotrophin receptors and ALCAM in the same transport compartment, we went on to investigate whether the latter has a role in neurotrophin signaling. PC12 cells were transfected with ALCAM tagged with the fluorescent protein Dendra2 (ALCAM-D) (Gurskaya et al. 2006) or Dendra2 alone, and treated or not with 100 ng/mL NGF. ALCAM-D is expressed on the cell surface where it was recognized by an antibody directed toward the ALCAM extracellular domain, as well as in internal organelles, which are not labeled by the antibody under non-permeabilizing conditions (Fig. 4a). Interestingly, upon transfection with ALCAM-D, the percentage of PC12 cells that displayed neurites was significantly increased when compared with cells transfected with Dendra2 alone (Fig. 4b; p = 0.0059 at 3 days). Although this trend was already visible after 1 day of NGF treatment, it did not reach statistical significance (Fig. 4b; p = 0.0996). In contrast, no effect of ALCAM on neuritogenesis was found in absence of NGF (Fig. 4c), suggesting that ALCAM over-expression per se does not directly initiate neurite outgrowth.

image

Figure 4.  ALCAM enhances NGF-induced neurite outgrowth. (a) PC12 cells were transfected with Dendra2 or ALCAM-D and differentiated in the presence of 100 ng/mL NGF for 3 days (scale bar, 10 μm). Fluorescence is cytosolic in Dendra2-transfected cells whilst it is predominantly at the plasma membrane in ALCAM-D cells where the fusion protein is recognized by an anti-ALCAM antibody (αALCAM). ALCAM-D organelles, which are not labeled by this antibody in non-permeabilized conditions, are also visible (arrowheads). (b) ALCAM-D significantly potentiated the percentage of cells that displayed neurite outgrowth at 3 days post-transfection (DPT) (p = 0.0059; n = 3; error bars = SEM). (c) Quantification of neurite outgrowth in the presence of specific differentiation stimuli. ALCAM-D did not induce neuronal differentiation in the absence of differentiation stimuli (−) or in response to dibutyryl-cAMP (dbcAMP; 0.1 or 0.5 mM) 5 days after transfection (5DPT), but significantly potentiated it in response NGF at 3DPT and increased outgrowth following treatment with GDNF at 7DPT. Data were analyzed using one-way anova with Bonferroni’s multiple comparison test (*p < 0.05, ***p < 0.001; n = 4; error bars = SEM). (d–f) ALCAM knockdown was followed both at the level of mRNA by quantitative real-time PCR (d) and protein level by western blot (e). (f) Neurite outgrowth upon NGF treatment was significantly impaired in PC12 cells down-regulated for ALCAM when compared with vector-treated control cells. Data were analyzed using an unpaired t-test (*p < 0.05, error bars = SEM).

Download figure to PowerPoint

To further investigate this process, we transfected cells with ALCAM-D or Dendra2 and compared their neurite growth response to a variety of differentiation stimuli (Fig. 4c). Strikingly, ALCAM over-expression had no effect on neurite outgrowth observed in response to dibutyryl-cAMP, a cell-permeable cAMP analog. This suggests that ALCAM does not act on a signaling cascade controlled by cAMP (Fig. 4c). However, in addition to robustly increasing NGF-induced neurite elongation, ALCAM over-expression also potentiated GDNF-induced differentiation at 7 days post-transfection (Fig. 4c; p = 0.0433), suggesting that ALCAM acts as a modifier of specific growth factor-dependent differentiation pathways.

The specificity of these gain-of-function experiments is confirmed by a loss-of-function approach in which ALCAM has been down-regulated in PC12 cells using an shRNA strategy (Fig. 4d–f). Cells were transfected for two days with an shRNA vector, targeting rat ALCAM, and then differentiated for one day in the presence of 100 ng/mL NGF. These conditions, which have been selected to maximize cell viability, yielded a robust, yet partial, knockdown of ALCAM both at mRNA (Fig. 4d) and protein (Fig. 4e) levels, as shown by quantitative real-time PCR and western blot, respectively. Crucially, under these conditions, NGF-dependent neurite outgrowth was significantly impaired in PC12 cells in which ALCAM was down-regulated when compared with vector-treated controls (Fig. 4f).

Both cis- and trans- interactions of ALCAM contribute to NGF-dependent neurite outgrowth

We generated truncation mutants in order to determine which domains of ALCAM were required for the potentiation of NGF-dependent neurite outgrowth (Fig. 5a). Deletion of the cytoplasmic domain of ALCAM, a short 34 amino acid region with no known interaction motifs, had no effect on this process (Fig. 5a). Surprisingly, truncations of the most distal extracellular domains in ΔD1-ALCAM-D and ΔD1D2-ALCAM-D, which have been previously shown to abolish ALCAM interactions in trans (van Kempen et al. 2001), had only a limited impact on differentiation (Fig. 5a). Indeed, over-expression of these mutants still potentiated neurite outgrowth in response to NGF, although to a lesser extent than the full-length protein (Fig. 5a). However, the extracellular domain of ALCAM is essential for this effect, because a CD8-ALCAM chimera containing its transmembrane and cytoplasmic domains was not capable of potentiating NGF-dependent neurite outgrowth (Fig. 5a). The differential effects of the ALCAM-D constructs on neurite outgrowth were not attributable to their level of expression (Figure S3).

image

Figure 5.  The extracellular domain of ALCAM is required for potentiation of NGF-dependent neurite outgrowth. (a) Neurite outgrowth in PC12 cells transfected with ALCAM mutants (lower panel) and differentiated with 100 ng/mL NGF for 3 days (upper panel). Truncations in the extracellular domain (ΔD1-ALCAM-D and ΔD1D2-ALCAM-D) reduced the potentiating effect of ALCAM. In contrast, removal of the entire intracellular domain (ΔC-ALCAM-D) did not. Consistently, CD8-ALCAM-D, which lacks the ALCAM extracellular domain did not potentiate neurite outgrowth. Data analyzed using one-way anova with Bonferroni’s multiple comparison test (*p < 0.05, ***p < 0.001; n = 3–8; error bars = SEM). (b) Quantification of Dendra2-transfected cells with extended neurites 3 days after transfection in response to 100 ng/mL NGF in the presence of 1 μg soluble ALCAM-Fc or human-Fc (Hu-Fc). A significant difference was seen between samples treated with ALCAM-Fc and Hu-Fc (p = 0.0285). Data were analyzed using an unpaired t-test (*p < 0.05; n = 4; error bars = SEM). (c) No significant potentiation of neurite outgrowth was observed in the presence of soluble CD6-Fc (1 μg/mL) in either Dendra2 or ALCAM-D transfected cells treated with NGF for 3 days. Data were analyzed using unpaired t-test analysis (n = 3; error bars = SEM).

Download figure to PowerPoint

To our knowledge, this is the first molecular characterization of ALCAM function in neurite outgrowth and suggests that the protein itself does not engage in downstream signaling via its cytoplasmic tail; rather, ALCAM appears to act as a modifier of differentiation pathways involving different classes of trophic factors, such as neurotrophins and GDNF. In support of this possibility, incubation of PC12 with the soluble ALCAM extracellular domain is sufficient to potentiate NGF-dependent neurite outgrowth (Fig. 5b). In contrast, the glycoprotein CD6, a well-known heterophilic ALCAM binding partner (Bowen et al. 1995), was incapable of potentiating this process (Fig. 5c). These results illustrate that distinct ALCAM ligands differentially affect the ALCAM-mediated potentiation of neurite outgrowth induced by NGF.

ALCAM over-expression affects TrkA signaling

We then assessed the effects of ALCAM-D or Dendra2 in NGF-dependent signaling by treating starved PC12 cells with 100 ng/mL of NGF and analyzing the phosphorylation of TrkA and ERK1/2 (pTrkA and pERK1/2) (Klesse and Parada 1999). As shown in Fig. 6a, ALCAM-D over-expressing PC12 cells displayed increased pTrkA levels at 3, 15 and 30 min after NGF activation when compared with Dendra2-expressing cells (Fig. 6a). These differences are highly significant for pTrkA (Fig. 6b; upper panel; p = 0.00038), but not for pERK1/2 (Fig. 6b; lower panel; p = 0.19) (see Discussion section).

image

Figure 6.  ALCAM enhances TrkA phosphorylation in response to NGF. (a) pTrkA and pERK1/2 are shown at 0, 3, 5 and 30 min after addition of 100 ng/mL NGF to PC12 cells starved for 5 h. ALCAM-D expressing cells show higher TrkA phosphorylation than control cells in response to NGF. Four independent experiments are assessed in the lower panel using one-way anova (p = <0.0004). However, Bonferroni’s multiple comparison test failed to detect a significant effect of transfection. (b) An alternative statistical approach that removes variability attributed to different experiments and timepoints after NGF stimulation is shown in the box and whisker plot for TrkA (upper panel) and ERK1/2 phosphorylation (lower panel) (au, arbitrary units). A significant increase in TrkA phosphorylation in ALCAM-D transfected cells was found in comparison to Dendra2 controls (***p < 0.001). In contrast, no significant difference in ERK1/2 phosphorylation was detected.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

In this study, we have shown that ALCAM undergoes axonal transport in an endosomal compartment associated with neurotrophin signaling and provided evidence that ALCAM has a cell-autonomous role in promoting differentiation by potentiating TrkA phosphorylation and NGF signaling. Our findings contribute to the growing body of evidence suggesting that the interplay between classical IgCAMs and neurotrophic signaling is a widespread function of this protein family, rather than the exception. Many CAMs have similar properties to ALCAM that may be important for growth factor signaling. These properties include site-specific cleavage to release soluble ectodomains that alter adhesive functions, but could also modulate growth factor responses, as shown for the ALCAM ectodomain in NGF signaling and neurite outgrowth.

We were unable to determine the exact mechanism by which ALCAM potentiates neurotrophin signaling, although we did find an increase in TrkA phosphorylation. No direct binding between TrkA and ALCAM was detected by immunoprecipitation (data not shown). However, ALCAM could increase TrkA activation in several ways that would not require direct interaction, including promoting the local concentration of growth factors at the cell surface, or altering the distribution of neurotrophin receptors on the membrane.

We have shown that ALCAM undergoes long-range axonal transport in a non-degradative compartment associated with neurotrophin receptors, as well as other CAMs. Moreover, we did not find evidence to suggest that the transported protein is degraded in the cell body. As the ALCAM extracellular domain is required for potentiating NGF-signaling, and ALCAM does not synergize with the secondary messenger cAMP, it is possible that ALCAM’s role in potentiating TrkA signaling occurs at the plasma membrane or within signaling endosomes, as suggested previously for neurotrophin receptors (Moises et al. 2007; Cosker et al. 2008).

Our results also have implications for the study of the role of ALCAM in other tissues and in pathological conditions, such as cancer. ALCAM is over-expressed in a diverse range of cancers where it has been associated with tumor growth and metastasis (Swart et al. 2005; Ofori-Acquah and King 2008). Soluble ALCAM and ALCAM truncation mutants disrupt ALCAM adhesion and alter metalloprotease activity and cancer cell migration (van Kilsdonk et al. 2008). Moreover, we showed that ALCAM potentiates growth factor signaling, a condition found in many types of cancers and linked with tumor progression (Spencer-Cisek 2002). Importantly, certain forms of cancer in which ALCAM is over-expressed, such as epithelial ovarian cancer (Rosso et al. 2007), also show high levels of expression of NGF and TrkA (Campos et al. 2007). Our results predict that ALCAM over-expression in these cells would potentiate TrkA signaling, possibly contributing to cell survival and enhanced tumor progression.

ALCAM could function as a site of integration for signaling, adhesion and migration. Its clustering at the plasma membrane at points of cell-cell contact could provide additional spatio-temporal control of growth factor receptor signaling at these specific sites. These signaling hotspots may provide a positive feedback mechanism so that growth and survival signals are potentiated at specific points of cell–cell interaction, which in turn could help in the stabilization of neuronal networks during development.

Further work is still needed to determine the role of endogenous ALCAM in established neurotrophin-dependent processes using cells from ALCAM−/− mice as well as knockdown experiments. Based on the lack of overt developmental effects in ALCAM−/− mice (McKinnon et al. 2000; Weiner et al. 2004), a common occurrence in single CAM knockouts caused by the compensatory effect of other CAMs (Cremer et al. 1994; Cohen et al. 1998), and the finding that ALCAM is up-regulated in response to injury (Fournier-Thibault et al. 1999); we tested whether its absence influences sprouting at the neuromuscular junction in response to botulinum neurotoxin A (Angaut-Petit et al. 1990). No significant difference in sprouting was seen in mice lacking ALCAM (Figure S4), suggesting that ALCAM does not play a role in this process. However, botulinum neurotoxin A-induced nerve terminal sprouting may not rely on the neurotrophin pathway potentiated by ALCAM, or alternatively, other proteins may compensate for its loss in this context.

A better experimental model to validate our studies and extend our conclusions to other neurotrophins could be to test the effects of ALCAM on the plasticity of the dendritic network in response to TrkB receptor signaling (Horch 2004). It would be also very interesting to check the effects of ALCAM deletion in conjunction with other neurotrophin modifiers, such the scaffolding protein Kidins220/ARMS (Iglesias et al. 2000; Kong et al. 2001).

In conclusion, the results of this study provide strong evidence that CAMs possess functions beyond their adhesive properties and undergo complex and long-range endocytic pathways in neurons.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

We thank Anna Moore, Department of Radiology, Massachusetts General Hospital, for providing MIONs; Anne Gonzalez de Peredo, CEA, Grenoble, for the MS analysis; Katrin Deinhardt, New York University School of Medicine, for the purification of axonal endosomes; Joshua Weiner, University of Iowa, for providing the ALCAM−/− mice and Gavin Kelly, Cancer Research UK London Research Institute, for help with statistical analysis. AW was in receipt of a PhD Studentship from the Wellcome Trust. LG is the Graham Watts Senior Research Fellow supported by the Brain Research Trust. MT was supported by the Marie Curie RTN ‘ENDOCYTE’’ from the European Union FP6 Program and GS by Cancer Research UK. The authors have no actual or potential conflict of interest to declare.

References

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information
  • Angaut-Petit D., Molgo J., Comella J. X., Faille L. and Tabti N. (1990) Terminal sprouting in mouse neuromuscular junctions poisoned with botulinum type A toxin: morphological and electrophysiological features. Neuroscience 37, 799808.
  • Arce V., Garces A., de Bovis B., Filippi P., Henderson C., Pettmann B. and deLapeyriere O. (1999) Cardiotrophin-1 requires LIFRbeta to promote survival of mouse motoneurons purified by a novel technique. J. Neurosci. Res. 55, 119126.
  • Avci H. X., Zelina P., Thelen K. and Pollerberg G. E. (2004) Role of cell adhesion molecule DM-GRASP in growth and orientation of retinal ganglion cell axons. Dev. Biol. 271, 291305.
  • Bohnert S. and Schiavo G. (2005) Tetanus toxin is transported in a novel neuronal compartment characterized by a specialized pH regulation. J. Biol. Chem. 280, 4233642344.
  • Bowen M. A., Patel D. D., Li X. et al. (1995) Cloning, mapping, and characterization of activated leukocyte-cell adhesion molecule (ALCAM), a CD6 ligand. J. Exp. Med. 181, 22132220.
  • Brun V., Dupuis A., Adrait A., Marcellin M., Thomas D., Court M., Vandenesch F. and Garin J. (2007) Isotope-labeled protein standards: toward absolute quantitative proteomics. Mol. Cell Proteomics 6, 21392149.
  • Campos X., Munoz Y., Selman A., Yazigi R., Moyano L., Weinstein-Oppenheimer C., Lara H. E. and Romero C. (2007) Nerve growth factor and its high-affinity receptor trkA participate in the control of vascular endothelial growth factor expression in epithelial ovarian cancer. Gynecol. Oncol. 104, 168175.
  • Cohen N. R., Taylor J. S., Scott L. B., Guillery R. W., Soriano P. and Furley A. J. (1998) Errors in corticospinal axon guidance in mice lacking the neural cell adhesion molecule L1. Curr. Biol. 8, 2633.
  • Cosker K. E., Courchesne S. L. and Segal R. A. (2008) Action in the axon: generation and transport of signaling endosomes. Curr. Opin. Neurobiol. 18, 270275.
  • Cremer H., Lange R., Christoph A. et al. (1994) Inactivation of the N-CAM gene in mice results in size reduction of the olfactory bulb and deficits in spatial learning. Nature 367, 455459.
  • Deinhardt K., Berninghausen O., Willison H. J., Hopkins C. R. and Schiavo G. (2006a) Tetanus toxin is internalized by a sequential clathrin-dependent mechanism initiated within lipid microdomains and independent of epsin1. J. Cell Biol. 174, 459471.
  • Deinhardt K., Salinas S., Verastegui C., Watson R., Worth D., Hanrahan S., Bucci C. and Schiavo G. (2006b) Rab5 and Rab7 control endocytic sorting along the axonal retrograde transport pathway. Neuron 52, 293305.
  • Deinhardt K., Reversi A., Berninghausen O., Hopkins C. R. and Schiavo G. (2007) Neurotrophins redirect p75NTR from a clathrin-independent to a clathrin-dependent endocytic pathway coupled to axonal transport. Traffic 8, 17361749.
  • Fournier-Thibault C., Pourquie O., Rouaud T. and Le Douarin N. M. (1999) BEN/SC1/DM-GRASP expression during neuromuscular development: a cell adhesion molecule regulated by innervation. J. Neurosci. 19, 13821392.
  • Gurskaya N. G., Verkhusha V. V., Shcheglov A. S., Staroverov D. B., Chepurnykh T. V., Fradkov A. F., Lukyanov S. and Lukyanov K. A. (2006) Engineering of a monomeric green-to-red photoactivatable fluorescent protein induced by blue light. Nat. Biotechnol. 24, 461465.
  • Horch H. W. (2004) Local effects of BDNF on dendritic growth. Rev. Neurosci. 15, 117129.
  • Iglesias T., Cabrera-Poch N., Mitchell M. P., Naven T. J., Rozengurt E. and Schiavo G. (2000) Identification and cloning of Kidins220, a novel neuronal substrate of protein kinase D. J. Biol. Chem. 275, 4004840056.
  • Kawauchi D., Kobayashi H., Sekine-Aizawa Y., Fujita S. C. and Murakami F. (2003) MuSC is involved in regulating axonal fasciculation of mouse primary vestibular afferents. Eur. J. Neurosci. 18, 22442252.
  • van Kempen L. C., Nelissen J. M., Degen W. G., Torensma R., Weidle U. H., Bloemers H. P., Figdor C. G. and Swart G. W. (2001) Molecular basis for the homophilic activated leukocyte cell adhesion molecule (ALCAM)-ALCAM interaction. J. Biol. Chem. 276, 2578325790.
  • van Kilsdonk J. W., Wilting R. H., Bergers M., van Muijen G. N., Schalkwijk J., van Kempen L. C. and Swart G. W. (2008) Attenuation of melanoma invasion by a secreted variant of activated leukocyte cell adhesion molecule. Cancer Res. 68, 36713679.
  • Klesse L. J. and Parada L. F. (1999) Trks: signal transduction and intracellular pathways. Microsc. Res. Tech. 45, 210216.
  • Kong H., Boulter J., Weber J. L., Lai C. and Chao M. V. (2001) An evolutionarily conserved transmembrane protein that is a novel downstream target of neurotrophin and ephrin receptors. J. Neurosci. 21, 176185.
  • Lalli G. and Schiavo G. (2002) Analysis of retrograde transport in motor neurons reveals common endocytic carriers for tetanus toxin and neurotrophin receptor p75NTR. J. Cell Biol. 156, 233239.
  • Maness P. F. and Schachner M. (2007) Neural recognition molecules of the immunoglobulin superfamily: signaling transducers of axon guidance and neuronal migration. Nat. Neurosci. 10, 1926.
  • McKinnon P. J., McLaughlin S. K., Kapsetaki M. and Margolskee R. F. (2000) Extracellular matrix-associated protein Sc1 is not essential for mouse development. Mol. Cell. Biol. 20, 656660.
  • Moises T., Dreier A., Flohr S., Esser M., Brauers E., Reiss K., Merken D., Weis J. and Kruttgen A. (2007) Tracking TrkA’s trafficking: NGF receptor trafficking controls NGF receptor signaling. Mol. Neurobiol. 35, 151159.
  • Moore A., Weissleder R. and Bogdanov A., Jr (1997) Uptake of dextran-coated monocrystalline iron oxides in tumor cells and macrophages. J. Magn. Reson. Imaging 7, 11401145.
  • Ofori-Acquah S. F. and King J. A. (2008) Activated leukocyte cell adhesion molecule: a new paradox in cancer. Transl. Res. 151, 122128.
  • Ott H., Bastmeyer M. and Stuermer C. A. (1998) Neurolin, the goldfish homolog of DM-GRASP, is involved in retinal axon pathfinding to the optic disk. J. Neurosci. 18, 33633372.
  • Ott H., Diekmann H., Stuermer C. A. and Bastmeyer M. (2001) Function of Neurolin (DM-GRASP/SC-1) in guidance of motor axons during zebrafish development. Dev. Biol. 235, 8697.
  • Paratcha G., Ledda F. and Ibanez C. F. (2003) The neural cell adhesion molecule NCAM is an alternative signaling receptor for GDNF family ligands. Cell 113, 867879.
  • Rosso O., Piazza T., Bongarzone I. et al. (2007) The ALCAM shedding by the metalloprotease ADAM17/TACE is involved in motility of ovarian carcinoma cells. Mol. Cancer Res. 5, 12461253.
  • Saffell J. L., Williams E. J., Mason I. J., Walsh F. S. and Doherty P. (1997) Expression of a dominant negative FGF receptor inhibits axonal growth and FGF receptor phosphorylation stimulated by CAMs. Neuron 18, 231242.
  • Salinas S., Bilsland L. G., Henaff D., Weston A. E., Keriel A., Schiavo G. and Kremer E. J. (2009) CAR-associated vesicular transport of an adenovirus in motor neuron axons. PLoS Pathog. 5, e1000442.
  • Salinas S., Schiavo G. and Kremer E. J. (2010) A hitchhiker’s guide to the nervous system: the complex journey of viruses and toxins. Nat. Rev. Microbiol. 8, 645655.
  • Schmid R. S. and Maness P. F. (2008) L1 and NCAM adhesion molecules as signaling coreceptors in neuronal migration and process outgrowth. Curr. Opin. Neurobiol. 18, 245250.
  • Spencer-Cisek P. A. (2002) The role of growth factors in malignancy: a focus on the epidermal growth factor receptor. Semin. Oncol. Nurs. 18, 1319.
  • Swart G. W. (2002) Activated leukocyte cell adhesion molecule (CD166/ALCAM): developmental and mechanistic aspects of cell clustering and cell migration. Eur. J. Cell Biol. 81, 313321.
  • Swart G. W., Lunter P. C., Kilsdonk J. W. and Kempen L. C. (2005) Activated leukocyte cell adhesion molecule (ALCAM/CD166): signaling at the divide of melanoma cell clustering and cell migration? Cancer Metastasis Rev. 24, 223236.
  • Weiner J. A., Koo S. J., Nicolas S., Fraboulet S., Pfaff S. L., Pourquie O. and Sanes J. R. (2004) Axon fasciculation defects and retinal dysplasias in mice lacking the immunoglobulin superfamily adhesion molecule BEN/ALCAM/SC1. Mol. Cell. Neurosci. 27, 5969.
  • Zweifel L. S., Kuruvilla R. and Ginty D. D. (2005) Functions and mechanisms of retrograde neurotrophin signalling. Nat. Rev. Neurosci. 6, 615625.

Supporting Information

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

Appendix S1. Supplemental methods.

Figure S1. Direct conjugation of the anti-ALCAM antibody (αALCAM) to fluorophores does not alter the ability of this antibody to recognize native ALCAM.

Figure S2. αALCAM does not bind to motor neurons isolated from ALCAM knockout mice.

Figure S3. Expression of ALCAM-D constructs does not reflect the level of neurite outgrowth in response to NGF A.

Figure S4. ALCAM does not modulate BoNT/A-induced nerve terminal sprouting in vivo.

As a service to our authors and readers, this journal provides supporting information supplied by the authors. Such materials are peer-reviewed and may be re-organized for online delivery, but are not copy-edited or typeset. Technical support issues arising from supporting information (other than missing files) should be addressed to the authors.

FilenameFormatSizeDescription
JNC_7658_sm_FigS1-S4.pdf12202KSupporting info item

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.