Different Regulation of N-Cadherin and Cadherin-11 in Rat Hippocampus

Abstract

Cadherin-mediated specific cell adhesion is an important process in brain development as well as in synaptic plasticity in the adult brain. In this study the authors quantified mRNA levels of N-cadherin and cadherin-11 in different brain regions for the first time. In hippocampus N-cadherin mRNA levels were very high at embryonic stages and decreased during further development, whereas cadherin-11 mRNA levels were highest at postnatal stages. However, N-cadherin protein level was not altered during hippocampal development and cadherin-11 protein was low at embryonic but high at postnatal and adult stages. In cultured hippocampal neurons both cadherins became colocalized and recruited to synaptic sites during ongoing differentiation, with especially high accumulation of cadherin-11 at synapses. These data hint at a critical role of N-cadherin at early embryonic stages and early synaptogenesis, whereas cadherin-11 might be more important for further stabilization of synapses in the postnatal period and adulthood.

Keywords::

• cadherin
• cell adhesion
• development
• hippocampus
• synapse
• synaptogenesis

INTRODUCTION

The central nervous system develops from the relatively simple neural tube to a highly complex structure forming functional neuronal circuits. This includes formation of brain nuclei and layers, dendritic and axonal outgrowth, axonal path finding and formation, and maturation of synaptic connections leading finally to functional systems. Cadherins, Ca²⁺-dependent cell adhesion molecules, are involved in many aspects of these processes (Huntley et al., 2002; Bamji, 2005; Takeichi and Abe, 2005; Halbleib and Nelson, 2006; Redies, 2000).

Classic cadherins are single-span transmenbrane glycoproteins associated with adherens junctions of many cell types (Angst et al., 2001; Shapiro et al., 1995; Steinberg and McNutt, 1999; Yap et al., 1997). The extracellular domain consists of five tandemly repeated domains of ∼110 amino acids (EC1–EC5) with intercalated Ca²⁺ binding sites. The ectodomain forms a rod-like structure responsible for trans-interaction with cadherins of opposing cell surfaces (Shapiro et al., 1995; Nagar et al., 1996). The intracellular domain is highly conserved and is critically involved in linkage of cadherins to the actin cytoskeleton via adaptor molecules, the catenins (Yap et al., 1997; Kemler and Ozawa, 1989).

The functional role of cadherins is not limited to mechanical adhesion between cells, rather it extends to an involvement in many signaling events. In neurons cadherins are localized at or around synapses at puncta adherentia, holding pre- and postsynaptic membrane in the correct position (Uchida et al., 1996). Selective surface expression of cadherins and homophilic cadherin-cadherin interaction is thought to be one major mechanism of axonal fasciculation and path finding as well as specific synapse formation. Neuronal expression of particular cadherins and cadherin-mediated specific adhesion may provide a recognition code important for correct brain development (Fannon and Colman, 1996; Yamagata et al., 1995). It has become clear that the function of cadherins extends beyond their developmental role to regulation of activity-dependent changes in synaptic strength. Thus, cadherins play a critical role in long-term potentiation (LTP), which is considered to be the molecular and cellular basis for memory formation and learning (Tang et al., 1998).

In the hippocampus nine classic cadherins are expressed, namely N-, E-, R-cadherin, cadherin-6, -8, -9, -10, -11, and -14 (Bekirov et al., 2002). Two of these cadherins, i.e., N-cadherin and cadherin-11, have been investigated in more detail regarding their role in LTP formation. Whereas N-cadherin is supposed to promote LTP, cadherin-11 seems to inhibit LTP. Experimental evidence for this statement comes from data of Tang et al. (1998), who could show that application of N-cadherin antibodies or peptides specific for the adhesion motif in the extracellular domain of N-cadherin during the high-frequency stimulation period inhibits the potentiated neuronal response in hippocampal slice cultures. In contrast, deficiency of cadherin-11 leads to enhanced LTP formation (Manabe et al., 2000). In our previous work we proposed a model of how the different Ca²⁺ affinities of both cadherins (cadherin-11: K_d ∼ 0.2 mM; N-cadherin: K_d ∼ 0.7 mM) could explain the opposing effects on LTP by displaying different binding properties during activity-dependent transient reduction of Ca²⁺ in the synaptic cleft (Baumgartner et al., 2003; Heupel et al., 2008).

In this study we quantified N-cadherin and cadherin-11 mRNA levels in hippocampal development and could show that N-cadherin is expressed highly at embryonic stages and decreased during further development. In contrast, cadherin-11 expression was particularly low at embryonic stages and highest at postnatal stages. Additionally, both cadherins showed different localization patterns in cultured hippocampal neurons depending on the maturation state of the neurons. These data hint at a critical role of cadherin-11 in hippocampus where it might be especially important in differentiated neurons for stabilization of glutamatergic synapses.

METHODS

Ethics Statement

All animal experiments were performed in compliance with the guidelines for the welfare of experimental animals issued by the Federal Government of Germany, the National Institutes of Health, and the Max Planck Society. The described experiments with rat tissue and cultured cells/neurons are approved by the local ethics committee (Ulm University; ID Number: O.103).

Cell Culture

Hippocampal cell cultures were prepared from fetal rat hippocampi (embryonic day 19) of Sprague-Dawley rats (Crl:CD; Charles-River Lab., Wilmington, MA, USA). Briefly, hippocampi were dissected and dissociated by treatment with 2.5% trypsin (Invitrogen, USA) for 20 min at 37°C, washed several times with Hanks’ buffered salt solution (HBSS) (PAA, Austria) and treated with 0.01% DNaseI (Invitrogen). Finally, cell suspension was sieved (100 μm pore size) and cells plated onto coverslips coated with poly-l-lysine (Sigma) at a density of 10,000 cells/cm². Cells were initially cultured in Dulbecco's modified Eagle's medium (DMEM) (Invitrogen) supplemented with 10% fetal calf serum (Invitrogen) and 50 U/ml penicllin/50 μg/ml streptomycin (Invitrogen). After 4 h, medium was replaced by Neurobasal medium with 2% B27 supplement (Invitrogen) and antibiotics (concentrations as stated above). Cultures were maintained at 37°C, 5% CO₂, 95% air in a humidified atmosphere (Heraeus Instruments, Germany).

Immunocytochemistry

Cultured hippocampal neurons were fixed with 4% formaldehyde for 20 min at 4°C after 1, 7, 16, and 35 days in vitro. After washing with phosphate-buffered saline (PBS) (3 × 5 min) cells were permeabilized with 0.1% Triton-X-100 in PBS, washed again (3 × 5 min), and blocked for unspecific binding with 0.1% ovalbumin +0.5% fish skin gelatine in PBS for 30 min at room temperature (RT). The primary antibodies were applied at 4°C overnight. Rabbit anti-N-cadherin (Biozol) was used at a final dilution of 1:1200, mouse anti-cadherin-11 (ZyMed) at a dilution of 1:900, and the guinea pig anti-vesicular glutamate transporter 1 (vGlut1) (Chemicon) at a dilution of 1:6000. After further washing, cells were incubated with a mixture of goat anti-rabbit Alexa Fluor 488, goat anti-mouse Alexa Fluor 568, and goat anti-guinea pig Alexa Fluor 647 (dilutions 1:900; Molecular Probes, USA) for 45 min at RT. After final rinsing for 3 × 5 min with PBS, cells were mounted in Mowiol 4.88 (Hoechst, Germany). Images were acquired on a Zeiss Axioscope2 microscope equipped with an AxioCam MRm camera (Zeiss) and analyzed using AxioVision software (Zeiss) and Adobe Photoshop (Adobe Systems, USA). vGlut1 immunoreactivity (detected by Alexa Fluor 647) was visualized in the final images in blue for better comparison with the red and green signals from cadherin-11 and N-cadherin antibodies, respectively. Representative images were shown from at least three independent experiments. Controls were performed by omission of the primary antibodies.

Quantitative Real-Time Reverse Transcriptase–Polymerase Chain Reaction (RT-PCR)

Rats at 18, 19, and 20 days of gestation, postnatal days 4 and 6, and 9 to 10 weeks of age were sacrificed by CO₂. Brains were taken out and, after removal of meninges, further dissected with the help of a stereoscopic microscope. Brains at 9 to 10 weeks of age were either used as whole organ (including brain stem) for RNA purification or were further dissected. The telencephalon was cut off and both hemispheres separated in the midline. From the medial side the striatum and hippocampus could clearly be identified and both regions were completely excised. After removal of remaining neuronal tissue, the whole cortex from the frontal to occipital pole was used for analysis. From brains of embryonic and postnatal stages only the hippocampus was dissected. Excised tissue was either processed immediately or transferred to RNAlater (Applied Biosystems Ambion, Austin, TX, USA). RNA was isolated using the RNeasy Extraction Kit (Qiagen, Germany). Tissue stored in RNAlater was cut into 30-mg pieces, transferred to RLT lysis buffer, and homogenized using a Potter S (B. Braun Biotech International) prior to RNA extraction. Residual DNA was digested either on-column during RNA purification (RNase-Free DNase kit; Qiagen) or afterwards (DNase I; Fermentas, Germany). Success of DNase treatment was assessed by PCR. RNAs were then either purified by standard phenol-chloroform extraction or DNase I was inactivated by heat and EDTA. cDNA synthesis was carried out with SuperScriptIII reverse transcriptase (Invitrogen) and an oligo (dT)₂₅ VN primer.

Real-time PCR was performed with 94°C, 15 s, 55°C, 30 s, and 72°C, 30 s for 50 cycles with 2 μl 1:10 diluted cDNA, the QuantiTect SYBR Green PCR Kit (Qiagen) and the following primers (Eurofins MWG Operon, Germany) in a LightCycler 1.0 (Roche Applied Science, Mannheim, Germany): CycA-fwd (5′-AGC ACT GGG GAG AAA GGA TT-3′) and CycA-rev (5′-AGC CAC TCA GTC TTG GCA GT-3′) (Bonefeld et al., 2008), Rpl13A-fw (5′-ACA AGA AAA AGC GGA TGG TG-3′) and Rpl13A-rev (5′-TTC CGG TAA TGG ATC TTT GC-3′) (Bonefeld et al., 2008), βCat-fwd (5′-CTG AGA AAC TTG TCC GAT GC-3′) and βCat-rev (5′-CGG TAA TGT CCT CCC TGT-3′), N-Cad-fwd (5′-CCA CGC TGA GCC ACA AT-3′) and N-Cad-rev (5′-CCG CTG CTG GAG GAG TT-3′), Cad1-fwd (5′-CGA CCA GCA CCC AAC AGT-3′) and Cad11-rev (5′-TGC CCT CAT AAC CAT AGA TT-3′). Specificity of primers was ascertained with local sequence alignment using BLAST algorithm against GenBank (National Center for Bioinformatics [NCBI]) nucleotides. No homologies to other cadherin mRNAs were found. All quantifications were normalized to the geometric mean of CycA and Rpl13A, which were the most stable of eight reference genes tested. Analysis of real-time RT-PCR data was carried out according to the method described by Rutledge and Côté (2003). Briefly, primer pair efficiency was determined by measurement of a cDNA dilution series and the formula E = 10^−slope. Crossing points were always calculated by the Fit points method of the LightCycler software and a fluorescence threshold (F_t) of 0.1. Original fluorescence was then calculated by the formula F₀ = F_t/E^Ct. This value was divided by the geometric mean of the F₀ values of both reference genes to obtain the expression of the gene of interest relative to the expression of the reference genes. For comparison of gene expression in different brain regions values were further normalized to ”whole brain” and for expression during development to adult hippocampus.

Electrophoresis and Western Blotting

Rat hippocampi of various developmental stages and ages were excised and dissolved in Laemmli sample buffer (Laemmli, 1970). After heating to 95°C for 5 min, samples were subjected to sodium dodecyl sulfate (SDS)–polyacrylamide electrophoresis (7.5% gels). Protein content was quantified according to Heinzel et al. (1965). For immunoblotting, proteins were transferred in Kyhse-Andersen transfer buffer (Kyhse-Andersen, 1984) to Hybond nitrocellulose membranes (Amersham, Braunschweig, Germany), which were blocked with 5% low fat milk for 3 hours at RT in phosphate-buffered saline (pH 7.4) and incubated with the respective primary antibody overnight at 4°C (rabbit anti-N-cadherin 1:3000, Biozol; mouse anti-cadherin-11 1:5000, ZyMed; mouse anti-β-catenin 1:3000, BD Biosciences). As secondary antibodies horseradish peroxidase (HRP)-labeled goat anti-mouse or goat anti-rabbit immunoglobulin G's (IgGs) (Dianova, Hamburg, Germany) were used. Bound immunoglobulins were visualized by the enhanced chemiluminescence technique (Amersham). Representative blots of at least three independent experiments are shown.

Statistics

Data are presented as mean ± standard deviation. Differences in mRNA levels were assessed using the Mann-Whitney test.

RESULTS

Quantification of N-Cadherin and Cadherin-11 mRNAs in Various Regions of the Rat Brain

In order to compare N-cadherin and cadherin-11 expression in rat brain, we quantified the respective mRNA levels in different brain regions using quantitative real-time RT-PCR. For that purpose, RNA was purified from adult rat whole brain, isolated cortex, cerebellum, striatum, and hippocampus (n = 4). The mRNA levels of both cadherins in each region were measured relative to whole brain. As shown in Figure 1A N-cadherin expression was highest in cortex (120% ± 76% in relation to whole brain), lower in hippocampus (69% ± 38%), in cerebellum (62% ± 12%), and statistically significant lower in striatum (30% ± 23%). Cadherin-11 expression, however, was strong in hippocampus (112% ± 26%), cortex (75% ± 15), and striatum (69% ± 32%) and significantly lower in cerebellum (29% ± 3%) (Figure 1B).

Figure 1. Quantification of (A) N-cadherin and (B) cadherin-11 mRNA levels in various regions of the rat brain using real-time RTPCR. Values were normalized to whole brain. Note high expression of N-cadherin in cortex, hippocampus, and cerebellum and low expression in striatum. Cadherin-11 expression was highest in hippocampus and lowest in cerebellum. n = 4; *p <0.05.

Quantification of N-Cadherin and Cadherin-11 mRNAs during Hippocampal Development

Since cadherin-11 was expressed very high in hippocampus, we further focused on cadherin expression of this interesting brain region. N-Cadherin and cadherin-11 mRNA levels were quantified by real-time RT-PCR in hippocampus at different developmental stages. RNA was purified from embryonic (E18–E20; n = 6), postnatal (P4–P6; n = 6), and adult (9 weeks; n = 6) hippocampi. mRNA levels were normalized to the adult stage (Figure 2). N-Cadherin was expressed highest at the embryonic stage, namely 357% ± 125% (of adult hippocampi) and remained high postnatally, namely 288% ± 80%, but decreased significantly at the adult stage (embryonic versus adult, p < 0.01; postnatal versus adult, p < 0.01). In contrast, cadherin-11 was expressed to similar extent in embryonic (111% ± 42%) and adult hippocampi but was up-regulated at postnatal stages to 249% ± 47% (p < 0.01).

In addition, β-catenin expression was determined for comparison. β-Catenin displayed a very high expression at embryonic stages (384% ± 145%) and at postnatal stages (438 ± 76%) compared to adult hippocampi (p < 0.01). This expression pattern was similar to that of N-cadherin.

Figure 2. Quantification of (A) N-cadherin, (B) cadherin-11, and (C) β-catenin mRNA levels in developing rat hippocampus. Values were normalized to adult hippocampus. N-cadherin and β-catenin were expressed higher at embryonic (E18–E20) and postnatal (P4–P6) stages compared to the adult hippocampus, whereas cadherin-11 expression was highest in the postnatal hippocampus. n = 6; *p < 0.01.

N-Cadherin and Cadherin-11 Proteins

N-Cadherin and cadherin-11 proteins were investigated during hippocampal development by Western blotting. The same time points of development were chosen as for determination of the respective mRNA levels by quantitative RT-PCR, i.e., embryonic, postnatal, and adult. Figure 3 shows representative blots (n = 3) for N-cadherin, cadherin-11, and β-catenin. Whereas N-cadherin and β-catenin immunoreactivities were similar at all three developmental stages, cadherin-11 immunoreactivity was very weak at the embryonic stage and strong at postnatal and adult stages. Thus, the data obtained above for N-cadherin and cadherin-11 expression at the mRNA level were only partially reflected at the protein level. Whereas N-cadherin mRNA decreased during development, the protein level was not altered. This is also true for cadherin-11, showing from postnatal stage to adulthood a decrease of the mRNA while protein level remained high. Interestingly, at the embryonic stage cadherin-11 protein level was very low, with the mRNA amount being in the same range as in the adult stage (with high protein levels).

Figure 3. Representative Western blots for N-cadherin, cadherin-11, and β-catenin in developing rat hippocampus (n = 3). Whereas N-cadherin and β-catenin protein content were similar at all stages of hippocampal development investigated, cadherin-11 content was very low at the embryonic stage.

Taking together the data on the mRNA and protein levels, N-cadherin seems to be important in hippocampus already during early stages of development, whereas cadherin-11 becomes more important at postnatal and adult stages. The different time course of expression of these two cadherins during development hints at different roles of both proteins in building up the typical hippocampal cytoarchitecture.

N-Cadherin and Cadherin-11 in Cultured Hippocampal Neurons

Hippocampal development requires specific neuronal interaction and synapse formation. To gain more information about the role of N-cadherin and cadherin-11 in synaptogenesis, cultured hippocampal neurons were investigated after different time periods of differentiation in vitro for the subcellular localization of both cadherins by immunocytochemistry. After 1 day in vitro (DIV), outgrowth of neuronal processes started and lamellipodia were formed (Figure 4A). N-Cadherin and cadherin-11 immunoreactivities were detected in the perikaryon, neuronal processes, and lamellipodia in a punctate staining pattern. In most cases, N-cadherin– and cadherin-11–positive puncta did not colocalize. Also after 7 DIV, N-cadherin and cadherin-11 puncta were observed all over the cytosol without obvious colocalization (Figure 4B). During further maturation of synapses (16 DIV; Figure 4C), especially cadherin-11 became recruited to synapses, as shown by colocalization of cadherin-11 and a marker for glutamatergic synapses, the vesicular glutamate transporter 1 (vGlut1). Some N-cadherin–immunopositive dots also colocalized with vGlut1; however, N-cadherin was distributed additionally within the dendrites in a punctate staining pattern. Localization of cadherin-11 at glutamatergic synapses became most evident after 35 DIV when the patterns of cadherin-11 and vGlut1 immunolabeling were nearly identical (Figure 4D). As already described for neurons of 16 DIV, N-cadherin was found not only to colocalize with cadherin-11 and vGlut1 but was also localized within the whole dendritic compartment.

Figure 4. Double and triple immunolabeling of cultured hippocampal neurons after 1, 7, 16, and 35 days in vitro (A, B, C, D, respectively) for N-cadherin (visualized in green), cadherin-11 (visualized in red), and vesicular glutamate transporter 1 (vGlut1; visualized in blue). Note punctate staining pattern for N-cadherin and cadherin-11 without colocalization of both proteins in neurons of 1 and 7 days in vitro. In longer differentiated neurons (C: 16 days; D: 35 days) both cadherins become partially colocalized and recruited to glutamatergic synapses, which was most evident for cadherin-11. n = 3.

These data show that both cadherins are expressed early in cultured hippocampal neurons and are not colocalized in little differentiated neurons. During further differentiation cadherin-11 and N-cadherin colocalize and become recruited to glutamatergic synapses. Whereas cadherin-11 is exclusively localized at these synapses, N-cadherin was found additionally all over the dendritic compartment in a punctate staining pattern.

DISCUSSION

Cadherin-mediated specific cell adhesion plays a critical role in many aspects of brain development, such as axonal path finding, formation of brain nuclei, layer formation and synaptogenesis, as well as in synaptic remodeling in the mature brain. The expression pattern of many cadherins has been intensively studied during brain development by in situ hybridization especially in the chicken and in the mouse brain. The majority of cadherins investigated is expressed in a spatially restricted manner in fiber tracts, brain nuclei, and neural circuits during different stages of development (Redies, 2000; Redies and Takeichi, 1993; Simonneau and Thiery, 1998; Suzuki et al., 1997; Kimura et al., 1996). Whereas cadherin mRNA localization has been mapped in detail previously, in this study we quantified mRNA levels of cadherins in various regions of the adult rat brain, to our knowledge, for the first time. We could show that in the adult rat brain N-cadherin mRNA was ubiquitously expressed in the areas investigated, with highest expression in cortex and lowest in striatum. In contrast, cadherin-11 mRNA levels were highest in hippocampus and lower in cortex and striatum (not statistically significant) and significantly lower in cerebellum. These quantitative measurements support the expression pattern observed by Suzuki et al. (1997) using in situ hybridization. Cadherin-11 and N-cadherin also differed in their expression pattern during hippocampal development. mRNA levels of N-cadherin were highest in the embryonic and postnatal periods and decreased in adulthood, whereas cadherin-11 mRNA level was especially high in the postnatal period. On the protein level, no decrease of both cadherins in adult hippocampus could be observed but low cadherin-11 expression at embryonic stages was verified. The discrepancy between cadherin mRNA levels and protein levels might be explained by an altered protein half-life or a regulation via miRNA normally repressing cadherin mRNA translation. These data support the idea that N-cadherin is very important during all developmental stages in many brain regions and hint at a particular critical role of cadherin-11 in hippocampus at postnatal and adult stages.

It has become clear that cadherin-11, besides its well-characterized function in the mesenchyme, displays a neuronal function. Cadherin-11 has been found to be present at glutamatergic synapses in the hippocampus and loss of cadherin-11 leads to enhanced LTP formation, suggesting a critical role of cadherin-11 for synaptic function (Manabe et al., 2000; Paradis et al., 2007). In this study we investigated cadherin-11 in cultured hippocampal neurons in comparison to N-cadherin. In young neurons, 1 and 7 days in vitro (DIV), both proteins were localized all over the cytosol in a mutually exclusive manner, whereas they become recruited to glutamatergic synapses in neurons of 16 and 35 DIV where both proteins partially colocalize. Interestingly, in neurons of 35 DIV cadherin-11 was found nearly completely localized at glutamatergic synapses, whereas N-cadherin was additionally distributed in a punctate pattern all over the dendritic compartment.

Whereas the function of N-cadherin at synapses has been extensively studied (Arikkath and Reichardt, 2008; Mendez et al., 2010; Mysore et al., 2008; Tai et al., 2008), little is known about the synaptic role of cadherin-11. N-Cadherin was the first cadherin identified at synapses (Beesley et al., 1995; Cifuentes-Diaz et al., 1994) and shown to be localized around the active zone where it holds pre- and postsynaptic membranes in position (Uchida et al., 1996). Later many other cadherins were found to be associated with synapses as well. It is currently unclear if different members of the cadherin family act together at one individual synapse or if each cadherin is expressed only in a subset of synapses. Benson and Tanaka (1998) could show that in cultured hippocampal neurons N-cadherin is localized to nearly all early immature synapses, whereas during further maturation it became recruited to glutamatergic synapses and was absent from γ-aminobutyric acid-ergic (GABAergic) synapses. Also N-cadherin and E-cadherin were shown to display a mutually exclusive distribution in stratum lucidum of hippocampus (Fannon and Colman, 1996). Taken together, some cadherins may fulfill a common function at an individual synapse, whereas others have quite distinct functions. As already mentioned above, in the case of N-cadherin and cadherin-11, both cadherins were found at glutamatergic synapses of the hippocampus but they might differ with respect to their temporal expression during synaptogenesis. Our real-time PCR data show that cadherin-11 expression is very low at embryonic stages, whereas N-cadherin expression is high at this time period. This fits to the view that N-cadherin is implicated in initial stages of synapse formation (Benson and Tanaka, 1998; Togashi et al., 2002) and we hypothesize that cadherin-11 might be more important at mature synapses. However, N-cadherin is not only important for synaptogenesis but also fulfills important functions at mature synapses, such as in the process of synaptic plasticity (Arikkath and Reichardt, 2008; Tai et al., 2008). In this context, it is interesting to mention that we could show previously that binding properties of individual cadherins differ with respect to their Ca²⁺ dependency. Cadherin-11 displays a higher affinity to Ca²⁺ compared to N-cadherin (K_d ∼ 0.2 mM and K_d ∼ 0.7 mM, respectively), which may have significant consequences for synaptic adhesion during synaptic activity. For example, a drop of extracellular Ca²⁺ from physiological levels to 0.3 mM will lead to a loss of N-cadherin binding activity of 85% but only to a loss of 30% of binding activity of cadherin-11 (Baumgartner et al., 2003; Heupel et al., 2008). Such a drop of extracellular Ca²⁺ has been suggested to occur temporarily in the synaptic cleft during high synaptic activity such as induction of long-term potentiation (Krnjevic et al., 1980; Nicholson et al., 1978; Rusakov, 2001; Rusakov and Fine, 2003). Thus, during synaptic activity synapses will be stabilized mainly via cadherin-11 and not by N-cadherin.

CONCLUSION

In conclusion, our data suggest that both N-cadherin and cadherin-11, which has been little characterized in the nervous system as yet, exert an important function at glutamatergic synapses of the hippocampus. Whereas N-cadherin is expressed very early and is known to be important for initiation of synapse formation, cadherin-11 expression is relatively low in early hippocampus, suggesting cadherin-11 playing not such an important role in early processes of synaptogenesis. Cadherin-11 becomes recruited to mature synapses where it might serve to stabilize synapses more or less independent of neuronal activity.

Declaration of interest: The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.