The metabotropic glutamate type 1 (mGlu1) and type 5 (mGlu5) receptors, the only members of group I mGlu receptors, are implicated in synaptic plasticity and mechanisms of feedback control of glutamate release. They exhibit nearly complementary distributions throughout the central nervous system, well evident in the cerebellum, where mGlu1 receptor is most intensely expressed while mGlu5 receptor is not. Despite their different distribution, they show a similar subcellular localization and use common transducing pathways. We recently described the Grm1crv4 mouse with motor coordination deficits and renal anomalies caused by a spontaneous mutation inactivating the mGlu1 receptor. To define the neuropathological mechanisms in these mice, we evaluated expression and function of the mGlu5 receptor in cerebral and cerebellar cortices. Western blot and immunofluorescence analyses showed mGlu5 receptor overexpression. Quantitative reverse transcriptase-polymerase chain reaction results indicated that the up-regulation is already evident at RNA level. Functional studies confirmed an enhanced glutamate release from cortical cerebral and cerebellar synaptosomes when compared with wild-type that is abolished by the mGlu5 receptor-specific inhibitor, 2-methyl-6-(phenylethynyl) pyridine hydrochloride (MPEP). Finally, acute MPEP treatment of Grm1crv4/crv4 mice induced an evident although incomplete improvement of motor coordination, suggesting that mGlu5 receptors enhanced activity worsens, instead of improving, the motor-coordination defects in the Grm1crv4/crv4 mice.
The metabotropic glutamate (mGlu) receptor family comprises 8 subtypes that are subdivided into 3 groups on the basis of their sequence similarities and G-protein coupling. Group I consists of the mGlu1 and mGlu5 receptors, coded by the Grm1 and the Grm5 genes, respectively. Both receptors couple to polyphosphoinositide hydrolysis, with ensuing mobilization of intraterminal Ca2+ ions (Nicoletti et al. 2011). Although the apparent overlapping of their cell signaling may indicate interplay between the 2 receptors and common functional effects, tissue-specific distributions or physiological versus pathological conditions account for differences in functional roles. Both receptors are widely expressed in the central nervous system, although mGlu1 is highly expressed in the cerebellum, thalamus, and CA3 hippocampal region, while mGlu5 is concentrated in the CA1 and CA3 pyramidal cells, striatum, and cortex (Ferraguti and Shigemoto 2006). mGlu1 and mGlu5 receptors have different patterns of expression and differently contribute to the brain development. During human cortical development, their expression and cellular distribution are differentially regulated. This differential expression is consistent with the recent experimental studies, suggesting a prominent role for mGlu5 receptor in the regulation of proliferation, differentiation, and survival of neural progenitor cells (Castiglione et al. 2008; Boer et al. 2010). Indeed, these receptors differentially affect the maturation of cells in the cerebellum: mGlu1 and mGlu5 receptors appear to be linked to cell survival of Purkinje cells and granule cells, respectively. In turn, mGlu5 receptors on granule cells seem to influence the Purkinje cells indirectly, through a synaptic or a paracrine mechanism (Catania et al. 2001). In pathological conditions, mGlu1 and mGlu5 receptors show different functional contributions.
Genetic mutations affecting the Grm1 gene in the mouse, 11 reported to date (Mouse Genome Database, The Jackson Laboratory, Bar Harbor, United States of America), and the finding of patients with autoantibodies against mGlu1 receptors (Sillevis Smitt et al. 2000; Coesmans et al. 2003; Marignier et al. 2010) indicate this receptor as involved in cerebellar ataxia, although mutations of the GRM1 gene have not been reported to date in human ataxia (Rossi et al. 2010). On the other end, mice lacking the mGlu5 receptor do not exhibit ataxia but a phenotype mainly characterized by impaired spatial learning (Lu et al. 1997), and a GRM5 gene deletion was found in a family with attention deficit/hyperactivity disorder, whose phenotype closely resembled those of the Grm5 null mice (Elia et al. 2009). In addition, ataxia and motor impairment characterize also the Fragile X syndrome, a genetic neuropathy associated with the Fmr1 gene. Currently, there is evidence supporting a role for both receptor subtypes in the pathological mechanisms acting in the Fmr1 knockout mice (Thomas et al. 2011). Finally, in spite of differences in their expression profiles and functions, the mGlu1 and mGlu5 receptors are finely regulated and strongly connected with each other and specific, deleterious events can highlight this relationship.
In this study, we examined in Grm1crv4 mice, mouse mutants lacking the mGlu1 receptor (Conti et al. 2006; Puliti et al. 2011), the expression and function of the other member of group I mGlus, the mGlu5 receptor, to assess the possibility of compensatory mechanisms. We focused on the cerebral and cerebellar cortices, as structures critically involved in cognition, and motor coordination in which group I mGlu receptors function has previously been examined in detail (Nicoletti et al. 2011). We predicted a change in mGlu5 receptor expression and function as a correlate of the altered glutamate signaling that likely accompanies the absence of the mGlu1 receptor in the Grm1crv4 mice. This hypothesis was also sustained by our previous studies in which we demonstrated that the lack of mGlu1 receptors in the Grm1crv4 mice significantly affected the amount of glutamate released by depolarizing stimulus in the cerebral cortex (Musante et al. 2008).
Our results show that, in the Grm1crv4 mice, the absence of mGlu1 receptor produces an overexpression of mGlu5 receptors in the cerebral and cerebellar cortices, and that this altered expression is associated with an enhanced glutamate release from cortical cerebral and cerebellar synaptosomes when compared with wild-type (wt). Indeed, acute 2-methyl-6-(phenylethynyl) pyridine hydrochloride (MPEP) treatment of the Grm1crv4/crv4 mice induced an evident although incomplete improvement of motor coordination. Overall, these results illustrate the importance of both group I mGlu receptors in controlling glutamate release from presynaptic terminals of cerebral and cerebellar cortices, and unveil mGlu1 and mGlu5 receptors interplay that can result in molecular compensatory mechanisms under pathological conditions.
Material and Methods
Animals and In Vivo Treatments
The crv4 mutation is a spontaneous recessive mutation occurred in the BALB/c/Pas inbred strain. It consists of a retrotransposon long terminal repeat (LTR) fragment insertion that disrupts the splicing of the mGlu1 receptor (Grm1) gene and causes the absence of the protein (Conti et al. 2006). Grm1crv4/crv4 homozygous mice present mainly with motor-coordination deficits. Affected (Grm1crv4/crv4) and control (Grm1+/+) mice are maintained on the same genetic background by intercrossing Grm1+/crv4 mice at the animal facility of the National Institute of Cancer Research (Genoa, Italy). DNA was extracted from wt and Grm1crv4 mice tails and was amplified via polymerase chain reaction (PCR) using 2 couples of primers. The first one (5′-GAGTGTTCACTAGTTCACCCAAGA-3′ and 5′-TCAGGCAACAATAAGGCAAG-3′) that flanks the insertion and amplifies a product of 688 and 498 bp for the crv4 mutant and the wt alleles, respectively. The second couple of primers (5′-TGTCAGGGATGAACTGAAAGAA-3′ and 5′-GCAGCTCAATTCCCAACAAT-3′) amplifies a genomic fragment of 250 bp specific for the LTR insertion of the mGlu1 receptor gene (see Musante et al. 2010). All experiments were performed by using adult animals (3 months of age, on average). For acute treatments, 6 Grm1crv4/crv4 mice at 8 weeks of age were given an intraperitoneal injection of 10 mg/kg 2-methyl-6-(phenylethynyl) pyridine hydrochloride (MPEP) dissolved in saline.
The experimental procedures were approved by the ethical committee for animal experimentation of the National Institute of Cancer Research in Genoa, in accordance with the European legislation (European Communities Council directive of 24 November 1986, 86/609/EEC). All efforts were made to minimize animal suffering and to reduce the number of animals necessary to produce reliable results.
Preparation of Synaptosomes
The animals were sacrificed by decapitation, and the cortices and the cerebella were rapidly dissected at 0–4 °C. The experimental procedures were in accordance with the European legislation (European Communities Council Directive of 24 November 1986, 86/609/ EEC) and were approved by Italian Health Ministry according to the Italian legislation on animal experimentation (protocol no. 29823-10). All efforts were made to minimize animal suffering and to reduce the number of animals necessary to produce reliable results. Purified synaptosomes were prepared essentially according to Dunkley et al. (1986), with minor modifications. The tissue was homogenized in 10 volumes of 0.32 M sucrose, buffered to pH 7.4 with Tris (final concentration 0.01 M) using a glass Teflon tissue grinder (clearance 0.25 mm). The homogenate was centrifuged at 1000 × g for 5 min to remove nuclei and debris, and the supernatant was gently stratified on a discontinuous Percoll gradient (6, 10, and 20% v/v in Tris-buffered sucrose) and centrifuged at 33 500 × g for 5 min. The layer between 10% and 20% Percoll (synaptosomal fraction) was collected and washed by centrifugation. The synaptosomal pellets were always resuspended in a physiological medium having the following composition (mM): NaCl, 140; KCl, 3; MgSO4, 1.2; CaCl2, 1.2; NaH2PO4, 1.2; NaHCO3, 5; 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid, glucose, 10; pH 7.2–7.4.
Experiments of Release from Superfused Synaptosomes
Synaptosomes were incubated in the presence of 1-[7,8 3H]d-aspartate ([3H]d-ASP, final concentration 30 nM) at 37 °C, for 15 min, in a rotary water bath. After the labeling period, identical portions of the synaptosomal suspensions were layered on microporous filters at the bottom of parallel superfusion chambers (Ugo Basile, Comerio, Varese, Italy) thermostated at 37 °C (Raiteri and Raiteri 2000) and superfused at 0.5 mL/min−1 with standard physiological solution. Synaptosomes were first equilibrated during 36 min of superfusion and then superfusion fractions were collected according to the following scheme: Two 3-min fractions (basal release), one before (t = 36–39 min) and one after (t = 45–48 min), and a 6-min fraction (t = 39–45 min; evoked release). Synaptosomes were transiently (90 s) exposed, at the end of the first fraction collected (t= 39 min), to high K+ containing medium (12 mM, NaCl substituting for an equimolar concentration of KCl) in the presence or in the absence of (RS)-3,5-dihydroxyphenylglycine (3,5-DHPG); negative allosteric modulators were present from 8 min before agonists, while the positive allosteric modulator 3,3′-difluorobenzaldazine (DFB) was added concomitantly with the depolarizing stimulus. Collected fractions and superfused synaptosomes were counted for radioactivity or analyzed for the endogenous neurotransmitter content.
Real-time Quantitative Reverse Transcriptase-Polymerase Chain Reaction (qRT-PCR) Assay
Total RNA was extracted from the cerebral and cerebellar cortices by using Trizol reagent (Invitrogen, Life Technologies, Carlsbad, CA, United States of America) according to the manufacturer's protocol. All RNA samples were treated with RNase-free DNase (Roche Applied Science, Mannheim, Germany) and quantified afterward by measuring the optical density (NanoDrop ND-2000 Spectrophotometer, NanoDrop Technologies). One microgram of total RNA was reverse transcribed using The Advantage RT-for-PCR Kit (Clontech Laboratories, Mountain View, CA, United States of America). Real-time PCR was performed in an IQ5 Multicolor Real-Time PCR Detection System using iQ SYBR Green Supermix (Bio-Rad, Hercules, CA, United States of America). PCR amplifications were performed in triplicate using 2:l of diluted cDNA in 25 µL final reaction mixture for mouse Grm5 along with parallel measurements of mouse glyceraldehyde-3-phosphate dehydrogenase (“Gapdh”) cDNA (an internal control). Primers for Grm5 (Grm5_F1: 5′-AGCAAGTGATCAGAAAGACTCG-3′; Grm5_R1: 5′-GTCACAGACTGCAGCAGAGC-3′) and Gapdh (Gapdh_F3: 5′-ATTGTCAGCAATGCATCCTG-3′, Gapdh_R3: 5′-ATGGACTGTGGTCATGAGCC-3′) were designed with Primer3 (http://www.genome.wi.mit.edu/cgi-bin/primer/primer3_www.cgi). The PCR conditions were 95 °C for 20 s, 60 °C for 20 s, 72 °C for 20 s, for 45 cycles.
The relative expression levels of each mRNA were calculated using the ΔΔCt method (Livak and Schmittgen 2001) normalizing to Gapdh and relative to the wt samples. The presence of one product of the correct size was verified by 1.5% agarose gel electrophoresis and melting curve analyses containing a single melt curve peak. Data from Grm1crv4/crv4 mice were calculated as relative to wild types, and values from wild types were normalized to 100%.
Cerebral and cerebellar cortices, and isolated cerebral cortex synaptosomes from Grm1crv4/crv4 and age-matched wt mice from the same breeding, were obtained and homogenized in lysis buffer (10 mM Tris, pH 8.8, 20% glycerol, 2% sodium dodecyl sulfate (SDS), 0.1 mM ethylenediaminetetraacetic acid, 5% β-mercaptoethanol). Protein concentration of each sample was determined using the Bradford method (Bio-Rad). Ten microgram of total protein was separated on 7% gel by means of SDS–polyacrylamide gel electrophoresis. The concentration of proteins in each sample fell on the linear portion of the curve. A triplicate analysis for each lysate sample was performed. Electroblotted proteins were monitored using Naphtol blue black staining (Sigma-Aldrich, MO, United States of America). Membranes were then incubated with the following antibodies: Anti-mGlu1 receptor mouse monoclonal antibody (1:2500; BD Biosciences, San Jose, CA, United States of America); anti-mGlu5 receptor rabbit monoclonal (1:10 000, Epitomics, CA, United States of America); and mouse monoclonal anti-Gapdh antibody (1:10 000; Millipore, Billerica, MA, United States of America). Anti-mGlu5 receptor antibodies recognized a band of ∼140 kDa. Specificity of anti-mGlu5 receptor antibodies was verified with immunoblotting using brain protein extracts obtained from mGlu5 receptor deficient mice (data not shown). After incubation with peroxidase-coupled secondary antibodies, protein bands were detected by using a western blotting detection system (ECL Advance™; Amersham Biosciences, Piscataway, NJ, United States of America). Bands were detected and analyzed for density using an enhanced chemiluminescence system (Versa-Doc 4000; Bio-Rad) and QuantityOne software (Bio-Rad). All of the protein bands used were normalized for the Gapdh level in the same membrane.
For immunofluorescence, brain and cerebellum were taken after 4% buffered paraformaldehyde perfusion and washed in 10% sucrose in phosphate buffer, pH 7.4, overnight. Tissues were then embedded in an optimum cutting temperature cryo-embedding matrix (OCT; Miles Scientific, Naperville, IL, United States of America), snapfrozen in a mixture of isopentane and dry ice, and stored at −80°C. Brain and cerebellum frozen sections (5-μm thick) were cut using a cryostat. The sections were incubated with 1:200 anti-mGlu5 receptor rabbit monoclonal antibody (Epitomics). As secondary fluorescent-labeled antibodies, we used the Alexa Fluor 488 goat anti-rabbit IgG (1:70, Invitrogen). Specificity of antibody labeling was demonstrated by the lack of staining after substituting proper control immunoglobulins (rabbit primary Ab isotype control from Invitrogen) for the primary antibodies. Slides were mounted with Fluorsave Reagent (Calbiochem, VWR International, Milano, Italy). Images were acquired by a Zeiss Axioscope 40 FL microscope, equipped with an AxioCam MRc5 digital videocamera and an immunofluorescence apparatus (Carl Zeiss SpA), and recorded by AxioVision software 4.3.
Assessment of Motor Coordination after In Vivo Treatment
Motor performance was evaluated with a rotarod apparatus (accelerating model; Ugo Basile). A group of 6 Grm1crv4/crv4 animals (3 females and 3 males) were analyzed at 8 weeks of age. Before the first test, mice were trained on the rotarod for 60 s at 4 rpm at constant speed, and then, placed on the accelerating rod at a starting speed of 4 rpm, reaching a final speed of 40 rpm in 300 s. The animals were allowed to stay on the rod for a maximum of 300 s and their time on the rod was recorded. Mice were tested for 2 trials at 20 min intervals for 3 consecutive days before treatment. On the day of treatment, mice were tested starting from 1 h from treatment. Other tests were performed 24 h, 4 days, and 1 week after treatment.
Footprint analyses were performed according to a previously described protocol (Conti et al. 2006). In brief, mice were trained to cross an illuminated alley (5 cm wide, 80 cm long) and to reach a dark box located at the end of the alley. Their footpads were then coated with a nontoxic, water-based paint, and the floor of the alley was covered with white paper. This test was repeated 3 times per mouse the day before treatment, and 3 times at 2 h from treatment. Analogous tests were performed with wt animals. The footprints were scanned, and randomly labeled with progressive numbers. The footprints were then blindly evaluated by an independent operator who, on the base of deviation of foot placement, was able to establish if they were obtained by wt, or Grm1crv4/crv4 mice before or after treatment.
Calculation and Data Analysis
For experiments of release from synaptosomes, the amount of radioactivity released into each superfusate fraction was expressed as a percentage of the total synaptosomal tritium content at the start of the fraction collected (fractional efflux). The K+-evoked release of [3H]d-ASP in the absence or in the presence of drugs was estimated by subtracting the neurotransmitter content into the fractions corresponding to the basal release from those corresponding to the evoked release. Analysis of variance (ANOVA) was performed followed by Tukey–Kramer test or Dunnett multiple-comparison test, as appropriate; direct comparisons were performed by applying Student's t-test. Data were considered significant for P < 0.05. Appropriate controls with antagonists were always run in parallel.
For western blot and qRT-PCR analyses, relative expression levels of Grm5 versus Gapdh were calculated. Data were statistically analyzed using 2-tailed unpaired Student's t-test and presented as percentage of control values. Differences were considered statistically significant at P < 0.05.
1-[7,8 3H]d-aspartate ([3H]d-ASP specific activity 16.3 Ci/mmol) was purchased from Amersham Radiochemical Center. (RS)-3,5-d3,5-DHPG, 2-methyl-6-(phenylethynyl) pyridine hydrochloride (MPEP), 7-(hydroxyimino)cyclopropa[b]chromen-1a-carboxylate ethyl ester (CPCCOEt) and DFB were obtained from Tocris Cookson (Bristol, United Kingdom).
mGlu1 Receptor Absence Unveils a mGlu5-Dependent Component of the 12-mM K+-Evoked Release of [3H]d-ASP in Cortical Glutamatergic Synaptosomes
The release of glutamate from synaptosomes isolated from the cortex of Grm1crv4/crv4 mice and from age-matched animals was studied by monitoring the release of preloaded [3H]d-ASP, a radioactive tracer that allows a reliable measure of endogenous glutamate storage and release from glutamatergic nerve terminals (see Grilli et al. 2004). Our previous observations demonstrated that the absence of mGlu1 receptors because of the crv4 genetic mutation failed to cause a significant change to the spontaneous release of preloaded [3H]d-ASP from Grm1crv4/crv4 cortical synaptosomes when compared with control synaptosomes, but significantly potentiated the 12-mM K+-evoked release of tritium in mutated mouse (Musante et al. 2008). Furthermore, the absence of mGlu1 receptor did not prevent the presynaptic mGlu5 autoreceptor-mediated releasing functions as suggested by the finding that the broad spectrum agonist 3,5-DHPG added at 0.3 μM, a concentration shown to selectively activate mGlu5 receptor subtypes (Musante et al. 2008), could further facilitate the 12-mM K+-evoked tritium exocytosis from Grm1crv4/crv4 cortical synaptosomes (Musante et al. 2008).
Figure 1 shows that the release induced by 12 mM K+ in mutated cortical synaptosomes is reduced by the mGlu5 receptor selective antagonist MPEP. Considering that, at the concentration applied, MPEP failed to modify the 12-mM K+-evoked [3H]d-ASP release in control cortical synaptosomes (Musante et al. 2008, 2010), this result could suggest the presence of presynaptic mGlu5 autoreceptors in the constitutively activate form, because of the genetic mutation. Of interest, the concomitant addition of the positive allosteric mGlu5 receptor modulator DFB (1 μM) could not modify the 12-mM K+-evoked release of [3H]d-ASP, suggesting that allosteric ligands are inactive in the absence of orthosteric agonists. At the concentration applied, DFB was previously reported to unveil mGlu5 receptor-mediated releasing effect (Luccini et al. 2007).
Cortical Cerebellar Glutamatergic Synaptosomes are Endowed with Release Regulating Presynaptic mGlu1 Autoreceptors
The existence of presynaptic mGlu1 and mGlu5 autoreceptors controlling glutamate release in cerebellum has been poorly investigated (Ferraguti et al. 2008). Experiments were therefore aimed to evaluate whether and to what extent 3,5-DHPG could modify the release of glutamate (monitored as release of [3H]d-ASP) from purified nerve terminals isolated from the cortex of the mouse cerebellum. Figure 2 shows that 3,5-DHPG enhances in a concentration-dependent manner the release of [3H]d-ASP evoked by 12 mM K+ from cerebellar glutamatergic mouse terminals, the maximum effect being observed when synaptosomes were exposed to 30 μM 3,5-DHPG. Considering that low µM concentration of 3,5-DHPG (0.3 μM) were previously found to preferentially activate mGlu5 autoreceptors while higher concentration (10–50 µM) were required to unveil mGlu1 receptor-mediated effect in the brain cortex (Musante et al. 2008), we speculated that mGlu1, but not mGlu5, receptors could be present in cerebellar cortical synaptosomal membranes. Accordingly, CPCCOEt, the selective antagonist of mGlu1 receptor, almost totally counteracted the positive effect 30 μM 3,5-DHPG exerted on 12-mM K+-evoked [3H]d-ASP exocytosis, while MPEP (1 μM) was inactive (Fig. 3). Notably, CPCCOEt, but not MPEP, also significantly reduced the release of [3H]d-ASP elicited by high K+ (Fig. 3), compatible with the idea that cortical cerebellar glutamatergic nerve endings could be endowed with mGlu1 autoreceptors in the constitutive activate form.
Recent data in literature have proposed that mGlu1 receptor-mediated effects on glutamate release are secondary to cannabinoid receptor activation acting as a retrograde messenger (Landucci et al. 2009). Whether such a cascade of events also could take place in the cortex and the cerebellum remains to be established, since the technical features of the experimental approach here used (the continuous up-down superfusion of synaptosomes) does not allow to detect indirect effects due to compounds endogenously produced and released. Future study will be needed to address this aspect.
mGlu1 Genetic Absence Unveils the Existence of Release-Regulating mGlu5 Autoreceptors in Grm1crv4/crv4 Cortical Cerebellar Synaptosomes
Figure 4 shows that the absence of mGlu1 receptor protein, because of the crv4 genetic mutation, caused a significant reduction of the 12-mM K+-evoked release of [3H]d-ASP from cerebellar synaptosomes when compared with age-matched wt animals. Furthermore, the 12-mM K+-evoked release of tritium from Grm1crv4/crv4 cerebellar synaptosomes was potentiated by concomitant addition of 0.3 µM 3,5-DHPG, while 30 µM 3,5-DHPG was inactive. Finally, in Grm1crv4/crv4 animals, CPCCOEt could not further inhibit K+-evoked glutamate exocytosis, while the release of [3H]d-ASP caused by 12 mM K+/0.3 µM 3,5-DHPG was significantly reduced by 1 µM MPEP (Fig. 5). All together these observations indicate that, in Grm1crv4/crv4 cerebellar glutamatergic terminals, presynaptic low-affinity mGlu1 autoreceptors could be replaced by presynaptic high-affinity mGlu5 autoreceptors.
mGlu5 Receptor mRNA is Increased in Grm1crv4/crv4 Mice Lacking the mGlu1 Subtype
In the absence of mGlu1 receptors, changes in mGlu5 receptors expression may be predicted, and this could play a role in the observed alterations of glutamate release from brain and cerebellar cortices presynaptic terminals of Grm1crv4/crv4 mice.
As first attempt to investigate this point, we tested the Grm5 gene expression in 5 Grm1crv4/crv4 animals and age-matched controls using qRT-PCR with specific Grm5 primers in these tissues. A nearly 10-fold increase of Grm5 mRNA expression was observable in the cerebellar cortex of mutated versus the wt mice (P< 0.05). Analogously, the Grm5 mRNA expression evaluated in the cerebral cortex of mutated and age-matched wt mice, indicating a 4-fold increase of Grm5 (P< 0.05) (Fig. 6). These data are indicative of a signaling alteration and suggest that expression of both Grm1 and Grm5 genes could have a coordinated regulation.
mGlu5 Receptor Protein is Overexpressed in the Homozygous Grm1crv4/crv4 Mice
As it is possible that mRNA levels are altered in the absence of consistent changes in the expression of corresponding coded proteins, we tested if altered expression of mGlu5 receptors could be also detectable in the cerebral and cerebellar cortices at protein level.
The mGlu5 receptor protein expression was evaluated by western blotting in cerebral and cerebellar cortical homogenates of 7 Grm1crv4/crv4 animals compared with age-matched wt animals obtained from the same breeding. A statistically significant increase of mGlu5 receptor protein was observed in the cerebral cortex (P< 0.05; Fig. 7A, D), and more evident in the cerebellar cortex (over 180% increase; Fig. 7C, F). To test if mGlu5 receptor protein increase observed in total tissue may be consistent with overexpression on nerve terminals, the mGlu5 protein expression was also evaluated in cerebral cortical synaptosomal membranes of 3 Grm1crv4/crv4 animals and age-matched controls. The results showed an increase of mGlu5 expression, although not statistically significant (Fig. 7B, E).
Altogether, these results indicate that in the absence of mGlu1 receptors in the motor-coordination deficient Grm1crv4/crv4 mice, the other member of the group I of mGlu receptors, mGlu5, is overexpressed, and that this overexpression is well detectable in the cerebral and cerebellar cortices.
The enhanced mGlu5 expression observed in the synaptosomal membranes suggests their presence at nerve terminals, but does not allow to conclude whether these receptors preferentially locate presynaptically. Indeed, synaptosomes have often sealed postsynaptic membrane components so that the exact location (presynaptic vs. postsynaptic) of receptor protein could not be speculated.
mGlu5 Receptor Protein Overexpression is Evident in Specific Areas of Brain and Cerebellum in the Homozygous Grm1crv4/crv4 Mice
The expression of mGlu5 receptor protein in Grm1crv4/crv4 mice was also investigated by immunofluorescence. Due to the wide expression of the mGlu5 receptors normally present in the brain of wt animals, in Grm1crv4/crv4 animals, we could perceive only a mild overexpression: A diffuse slight increase in the cerebral cortex and a more evident increase in the hippocampus.
In the cerebellar cortex of adult wt animals, only low levels mGlu5 expression are detectable in granule cells and in a small population (10%) of Golgi cells, and no expression has been reported in Purkinje cells (Neki et al. 1996; Negyessy et al. 1997; Ferraguti and Shigemoto 2006). Thus, the overexpression of mGlu5 receptor was well appreciable in Grm1crv4/crv4 cerebella. Specific overexpression was actually detectable in the granular layer of the cerebellar cortex, and, based on cell morphology, granule cells were the most overexpressing cell type, while Purkinje cells did not show enhanced expression (Fig. 8 and Supplementary Fig. S1). To note, although low levels of expression are present in adulthood, during development mGlu5 receptors are well expressed and largely contribute to differentiation and survival of the cerebellar cortex neurons (Casabona et al. 1997; Catania et al. 2001). Thus, high levels of mGlu5 receptors expression are reminiscent of immature cerebellum state and suggest that an arrest of cerebellum development may occur in mGlu1 receptors lacking Grm1crv4 mice. These results raise the possibility that the loss of activity of mGlu1 from one side, and the overexpression of mGlu5 receptors on the other side occurring at critical times of development may contribute to the pathophysiology of motor=coordination deficits in the Grm1crv4/crv4 animals.
Characterization of Grm1crv4/crv4 Mice After Acute MPEP Treatment
To examine the relationship between changes in the expression of mGlu5 receptors and motor symptoms associated with Grm1crv4/crv4 mice, we treated mice with a single injection drug, MPEP, that blocks mGlu5 receptors, behaving as negative allosteric modulator of these receptors (Gasparini et al. 1999). It is noteworthy that MPEP was injected at doses (10 mg/kg) that are behaviourally active but do not affect spontaneous locomotion in mice (Herzig and Schmidt 2004). We performed the treatment once, and observed the results obtained on the phenotype starting from 1 h later, then the day after, and at 4 and 7 days from treatment.
In gross observation, the treated Grm1crv4/crv4 mutants were viable. Their behavioural phenotypes resembled the major features normally seen in Grm1crv4/crv4 mice before treatment, including abnormal head movements and ataxic gait. However, treated mice exhibited less prominent toe dragging during walking, a steadier gait, and a more even pace while moving. Motor coordination in mutant mice was then assessed more in details by the rotarod test and the footprint test. The rotarod test did not show a significant difference between the performance of mutant mice obtained before and after treatments (Supplementary Fig. S2A). However, the rotarod test performance of Grm1crv4/crv4 mice mostly depends more on the muscular strength of forelegs that apparently seemed to be the same before and after treatment, than on the coordination of hindlegs. On the contrary, the footprint test allowed to appreciate gait changes in mutated mice after treatments (Supplementary Fig. S2B).
Wt mice had a normal gait and used only the front part of the paws to walk. Contrarily, Grm1crv4/crv4 mice walked on their whole paw both before and after treatment.
The wt mice had a narrow-based stance with steady close-proximity hindlimb footprints. On the contrary, Grm1crv4/crv4 footprints were featured with markedly wide-based stance, frequent off-line stumbling (arrow in the figure), separated hindlimb prints (better appreciated in the figure inset), and very evident toe dragging (indicated by arrowhead in the figure). The comparison of the footprint pattern in Grm1crv4/crv4 mice before and after treatment showed evident signs of improvement in treated mice, with steady gaits, few or absent off line stumbling, and less prominent toe dragging.
The availability of mouse models carrying distinct inactivating mutations of genes coding for mGlu1 and mGlu5 receptors represents an important tool to determine the individual properties of the 2 group I mGlus. The use of these mice has, for example, shown a prominent role for mGlu1 receptor in cerebellar long-term depression, for mGlu5 receptor in long-term potentiation in the hippocampus (Aiba, Chen, et al. 1994; Aiba, Kano, et al. 1994; Conquet et al. 1994), and for both receptors in neuroprotection (Nicoletti et al. 1999, 2011; Pellegrini-Giampietro 2003; Pshenichkin et al. 2008). However, these studies have not investigated whether expression and functions of the surviving group I mGlu receptors are altered in each mutant mouse. Such information is clearly relevant to interpret the mouse phenotype and understand the function(s) of the receptors. Here, we studied the contributions of group I mGlu receptors in the control of glutamate release from presynaptic nerve endings of cerebral and cerebellar cortices, and the alterations of mGlu5 receptors-controlled glutamate release in the absence of mGlu1 receptors in Grm1crv4/crv4 mice. The expression of mGlu5 receptor transcripts and of the encoded protein in Grm1crv4/crv4 mice was also evaluated.
The presence of mGlu1 and mGlu5 receptors on the terminal endings of the cerebral cortex has been already proved (Musante et al. 2008, 2010). The present functional data led us to conclude that group I mGlu receptors are also located presynaptically on cerebellar glutamatergic terminals, and that contribution of mGlu5 receptor subtype in both cortex and cerebellum could vary in mice bearing the Grm1crv4 mutation, when compared with the wt animals. Replacement of mGlu1 by mGlu5 receptors on the control of glutamate release from terminal ends can occur in both cortex and cerebellum of these mice.
Our expression results demonstrated that the lack of mGlu1 receptor leads to the overexpression of the remaining group I mGlu receptor, mGlu5, detectable at both mRNA and protein levels in both areas. A coordinated regulation of mRNA expression of both genes may account for the overexpression of Grm5 mRNA in the absence of the Grm1 molecules. In fact, similarities present at mRNA and coded proteins, a high degree of similarity in terms of exon/intron arrangement, and a conserved long-range structure of the genomic structures of Grm1 and Grm5, both in human and mouse, suggest that group I mGlu receptors have been generated by gene duplication from a common ancestor (Corti et al. 2003; Crepaldi et al. 2007; Menconi et al. 2011). Overall, these observations strongly indicate that they could still conserve a common mechanism of mRNA expression regulation. According to the enhancement of mRNA expression, and as emerged by western blotting, the encoded Grm5 protein was overexpressed in both cerebral and cerebellar cortices of Grm1crv4/crv4 mice. This protein overexpression may interfere with cellular signaling and functions by affecting different mechanisms. The presence of enhanced quantities of proteins may influence receptor trafficking to the cell membrane, and, consequently, differently modulate the triggered signaling. In this regard, recent studies suggest that calmodulin binding to mGlu5 receptor, conversely to mGlu1, inhibits mGlu5 receptor endocytosis and stabilizes the mGlu5 receptor protein on the cell surface (Lee et al. 2008; Choi et al. 2011).
Thus, replacement of mGlu1 by mGlu5 receptors, with concomitant mGlu5 enhanced expression, in the Grm1crv4 mutated mice may have relevant consequences on the triggered molecular signaling. Both mGlu1 and mGlu5 receptors are coupled to polyphosphoinositide hydrolysis which stimulates intracellular calcium release, but they differ in the kinetics of intracellular Ca2+ response. Indeed, Nash et al. (2002) recently demonstrated that mGlu5 receptors density influences the Ca2+ oscillation frequency in transfected cells, suggesting that the expression level of this receptor should qualitatively affect the physiological responses associated with the receptor.
Activation of mGlu5 receptors has been shown to specifically support the survival of progenitors undergoing differentiation into neurons (Castiglione et al. 2008). This regulation may involve the generation of oscillatory increases in intracellular calcium waves induced by activation of the mGlu5 subtype but not mGlu1 (Kawabata et al. 1998). mGlu5 receptors are highly expressed in the early postnatal life and then progressively decrease with age in most brain regions (Catania et al. 1994; Romano et al. 1996; Casabona et al. 1997). Although both group I mGlus contribute to the cerebral cortex development, that is, they are both expressed in the pyramidal cells in the late prenatal development, mGlu5 receptors exhibit a prominent role in differentiation and survival of neural progenitor cells, and during adulthood their expression seems to be particularly present in cells capable of renewing (Boer et al. 2010). In this view, persistency of mGlu5 receptor expression in the cerebral cortex may be indicative of its immature state.
Based on the results obtained from immunofluorescence analyses, the enhanced expression of mGlu5 receptors in cerebellum seems to involve the granular layer, particularly affecting the granule cells. During development mGlu5 receptors are highly expressed in cerebellum and granule cells, then progressively decrease with age (Catania et al. 1994; Romano et al. 1996; Casabona et al. 1997). In this respect, persistency of mGlu5 receptor expression in adulthood is reminiscent of an immature state of mGlu5-expressing cells and compatible with a block of cerebellar cortex differentiation in the Grm1crv4 mice. Accordingly, the persistent expression of mGlu5 receptors in granule cells during adulthood may have deleterious effects on their functions. Evidence indicate that the presence of intact granule cells and normal formation of parallel fibers-Purkjnie cells synapses are, in turn, prerequisite for climbing fibers synapse elimination (reviewed by Hashimoto et al. 2009). Thus, persistent expression of mGlu5 receptors in granule cells may have an additional indirect effect on the number of climbing fibers and regression of multiple climbing fibers innervation of Purkinje cells, which notably is impaired in Grm1 knockout mice (Kano et al. 1997). Indeed, granule cell are known to influence Purkinje cell differentiation and survival by releasing glutamate and/or neurotrophins (Baptista et al. 1994; Morrison and Mason 1998; Hirai and Launey 2000). It is possible that a reduction of dendritic growth of Purkinje cells, as reported for the Grm1 knockout mice (Aiba A, M Kano et al., 1994), is also present in the Grm1crv4 mice. In this view, the overexpression of mGlu5 receptors in granule cells could be interpreted as a tentative to increase dendritic growth and enhance Purkinje cell differentiation and survival by releasing more glutamate.
Interestingly, a switch in the expression of members of the same mGlu receptor subgroup in mouse genetic models and in experimental autoimmune encephalomyelitis (EAE) mice, models of multiple sclerosis (MS), has been already shown. Fazio et al. (2008) reported that, in EAE mice, the mGlu1 receptor expression decreased accompanied by a similar quantitative increase of the mGlu5 receptor. Analogously, in MS patients, the presence of mGlu1 receptor is reduced but mGlu5 receptor expression enhanced. An acute treatment of EAE mice with mGlu1 receptor enhancers significantly improved their motor coordination, whereas treatment with the mGlu5 receptor antagonists, including MPEP, had no effect. Contrary to what observed in Grm1crv4 mice, in EAE mice changes are restricted to Purkinje cells and their dendritic arborization and do not seem to affect granule cells.
Lyon et al. (2008) reported that the 2 members of mGlus group II, the mGlu2 and mGlu3 receptors, exhibit a mutual expression pattern in the respectively knockout mice: mGlu2 mRNA was increased in mGlu3 receptor knockout mice, and vice versa. In that case, they hypothesized that a mutual, albeit partial, compensatory response to the loss of a group II mGlu receptor may contribute to the relatively mild behavioral phenotype seen in both mice (for review, see Harrison et al. 2008).
Analogously to what reported by Lyon and co-workers for group II of mGlu receptors, it could be possible that, in the absence of mGlu5, the expression and function of mGlu1 receptor can be enhanced. However, our previous data did not evidence a mGlu1 receptor-dependent compensatory response in the release of glutamate from nerve endings of cerebral cortex in Grm5 knockout mice (Musante et al. 2008), suggesting that mechanisms of compensations between the 2 members of group I of mGlu receptors could not be mutual, at least concerning glutamate release from nerve endings in the cerebral cortex.
Overall, the observed overexpression of mGlu5 receptors and replacement of mGlu1 by mGlu5 receptors on the control of glutamate release from terminal ends may be different aspects of a common compensatory mechanism, albeit partial, following the loss of one member of group I mGlu receptors. This functional compensation, observable at the tissue level, is not associated with a rescue of the affected Grm1crv4 phenotype. On the contrary, the presence in the Grm1crv4 mice of active mGlu5 receptors, more sensitive than in control mice to the agonists action, may have deleterious consequences. Although able to mediate glutamate release, the 2 mGlu1 and mGlu5 receptors may play different roles in cerebral and cerebellar cortices. The concentration–effect relationship of 3,5-DHPG on glutamate release in Grm1crv4 unveils that concentration of agonist that were inactive in wt animals became effective in mutated mice. Assuming that this also occurs when glutamate is the autoreceptor agonist, one can speculate that physiological concentrations of glutamate, in the absence of mGlu1 receptors, can activate the mGlu5 autoreceptors and a molecular signaling normally not active in the wt animals leading to deleterious consequences. In keeping with this hypothesis, we then attempted to examine a possible causal relationship between changes in mGlu5 receptors and motor symptoms associated with Grm1crv4 by treating homozygous Grm1crv4/crv4 mice with MPEP, which behaves as negative allosteric modulator of mGlu5 receptors (Gasparini et al. 1999). After acute MPEP treatment, behavioral phenotype analysis revealed that Grm1crv4/crv4 mice still have a clinical ataxic gait, with significantly wide base width for hind feet in comparison with that in the wt mice. The rotarod test did not reveal any improvement of mice performances. However, treated mice had less off-line stumbling, and less prominent toe dragging than before treatments, indicating a steadier gait and a more even pace while moving. In fact, treated mice had better performances on footprint tests, indicating partially improved coordination and movement ability. These results suggest that the levels of mGlu5 receptors in the cerebellum correlate with the severity of ataxic motor behaviour in Grm1crv4/crv4 mice. However, increase of mGlu5 receptor expression and enhanced release of glutamate in cerebral and cerebellar cortices are not the sole biological defect that impairs the neuronal function in the Grm1crv4/crv4 mice. Other mechanisms mediated by mGlu1 receptor, independent of the mGlu5 receptor signaling pathway, are involved in maintaining the integrity of cerebral and cerebellar functions. In summary, on the base of our data, obtained by a single dose treatment with MPEP, it is difficult to conclude on how much the blocking of mGlu5 receptors activity in the Grm1crv4/crv4 mice may induce the phenotype reversal, though our data suggest that mGlu5 receptors activity might worsen, instead of improving, the motor-coordination defects in the Grm1crv4/crv4 mice. In keeping with this hypothesis, and on the base of our data, it would be interesting to see the effects of a chronic treatment of these animals with mGlu5 receptors antagonists. Indeed, considering the availability of mice lacking the mGlu5 receptors (Lu et al. 1997), a further step of our work will be to cross the Grm1crv4 mice with the Grm5ko mice to get a reduction of mGlu5 receptors expression in double mutants and to study the effects of a permanent reduction (i.e. half dose) expression of mGlu5 receptors on the Grm1crv4 mice motor-coordination defects.
This work was supported by RFPS-4-631972 grant ‘Genetic Bases of Birth Defects’ from the Italian Ministry of Health to R.R. and by grant from the Italian ‘Ministero dell'Istruzione, dell'Università e della Ricerca Scientifica’ (MIUR) to A.P.
Conflict of Interest: None declared.