Neuronal apoptosis and reversible motor deficit in dominant-negative GSK-3 conditional transgenic mice
2007; Springer Nature; Volume: 26; Issue: 11 Linguagem: Inglês
10.1038/sj.emboj.7601725
ISSN1460-2075
AutoresRaquel Gómez‐Sintes, Félix Hernández, Analı́a Bortolozzi, Francesc Artigas, Jesús Ávila, Paola Zaratin, Jean Pierre Gotteland, José J. Lucas,
Tópico(s)Genetics and Neurodevelopmental Disorders
ResumoArticle17 May 2007free access Neuronal apoptosis and reversible motor deficit in dominant-negative GSK-3 conditional transgenic mice Raquel Gómez-Sintes Raquel Gómez-Sintes Centro de Biología Molecular 'Severo Ochoa', CSIC/UAM, Madrid, Spain Search for more papers by this author Félix Hernández Félix Hernández Centro de Biología Molecular 'Severo Ochoa', CSIC/UAM, Madrid, Spain Search for more papers by this author Analía Bortolozzi Analía Bortolozzi Departamento de Neuroquímica y Neurofarmacología, Instituto de Investigaciones Biomédicas de Barcelona (CSIC), IDIBAPS, Barcelona, Spain Search for more papers by this author Francesc Artigas Francesc Artigas Departamento de Neuroquímica y Neurofarmacología, Instituto de Investigaciones Biomédicas de Barcelona (CSIC), IDIBAPS, Barcelona, Spain Search for more papers by this author Jesús Avila Jesús Avila Centro de Biología Molecular 'Severo Ochoa', CSIC/UAM, Madrid, Spain Search for more papers by this author Paola Zaratin Paola Zaratin Istituto di Ricerche Biomediche 'A. Marxer', LCG-RBM/Serono Discovery, Colleretto Giacosa, Italy Search for more papers by this author Jean Pierre Gotteland Jean Pierre Gotteland Merck Serono International, Geneva, Switzerland Search for more papers by this author José J Lucas Corresponding Author José J Lucas Centro de Biología Molecular 'Severo Ochoa', CSIC/UAM, Madrid, Spain Search for more papers by this author Raquel Gómez-Sintes Raquel Gómez-Sintes Centro de Biología Molecular 'Severo Ochoa', CSIC/UAM, Madrid, Spain Search for more papers by this author Félix Hernández Félix Hernández Centro de Biología Molecular 'Severo Ochoa', CSIC/UAM, Madrid, Spain Search for more papers by this author Analía Bortolozzi Analía Bortolozzi Departamento de Neuroquímica y Neurofarmacología, Instituto de Investigaciones Biomédicas de Barcelona (CSIC), IDIBAPS, Barcelona, Spain Search for more papers by this author Francesc Artigas Francesc Artigas Departamento de Neuroquímica y Neurofarmacología, Instituto de Investigaciones Biomédicas de Barcelona (CSIC), IDIBAPS, Barcelona, Spain Search for more papers by this author Jesús Avila Jesús Avila Centro de Biología Molecular 'Severo Ochoa', CSIC/UAM, Madrid, Spain Search for more papers by this author Paola Zaratin Paola Zaratin Istituto di Ricerche Biomediche 'A. Marxer', LCG-RBM/Serono Discovery, Colleretto Giacosa, Italy Search for more papers by this author Jean Pierre Gotteland Jean Pierre Gotteland Merck Serono International, Geneva, Switzerland Search for more papers by this author José J Lucas Corresponding Author José J Lucas Centro de Biología Molecular 'Severo Ochoa', CSIC/UAM, Madrid, Spain Search for more papers by this author Author Information Raquel Gómez-Sintes1, Félix Hernández1, Analía Bortolozzi2, Francesc Artigas2, Jesús Avila1, Paola Zaratin3, Jean Pierre Gotteland4 and José J Lucas 1 1Centro de Biología Molecular 'Severo Ochoa', CSIC/UAM, Madrid, Spain 2Departamento de Neuroquímica y Neurofarmacología, Instituto de Investigaciones Biomédicas de Barcelona (CSIC), IDIBAPS, Barcelona, Spain 3Istituto di Ricerche Biomediche 'A. Marxer', LCG-RBM/Serono Discovery, Colleretto Giacosa, Italy 4Merck Serono International, Geneva, Switzerland *Corresponding author. Centro de Biología Molecular 'Severo Ochoa', CSIC/UAM, Campus UAM de Cantoblanco, Madrid 28049, Spain. Tel.: +34 91 497 3595/8073; Fax: +34 91 497 8087; E-mail: [email protected] The EMBO Journal (2007)26:2743-2754https://doi.org/10.1038/sj.emboj.7601725 PDFDownload PDF of article text and main figures. ToolsAdd to favoritesDownload CitationsTrack CitationsPermissions ShareFacebookTwitterLinked InMendeleyWechatReddit Figures & Info Increased glycogen synthase kinase-3 (GSK-3) activity is believed to contribute to the etiology of chronic disorders like Alzheimer's disease and diabetes, thus supporting therapeutic potential of GSK-3 inhibitors. However, sustained GSK-3 inhibition might induce tumorigenesis through β-catenin-APC dysregulation. Besides, sustained in vivo inhibition by genetic means (constitutive knock-out mice) revealed unexpected embryonic lethality due to massive hepatocyte apoptosis. Here, we have generated transgenic mice with conditional (tetracycline system) expression of dominant-negative-GSK-3 as an alternative genetic approach to predict the outcome of chronic GSK-3 inhibition, either per se, or in combination with mouse models of disease. By choosing a postnatal neuron-specific promoter, here we specifically address the neurological consequences. Tet/DN-GSK-3 mice showed increased neuronal apoptosis and impaired motor coordination. Interestingly, DN-GSK-3 expression shut-down restored normal GSK-3 activity and re-established normal incidence of apoptosis and motor coordination. These results reveal the importance of intact GSK-3 activity for adult neuron viability and physiology and warn of potential neurological toxicity of GSK-3 pharmacological inhibition beyond physiological levels. Interestingly, the reversibility data also suggest that unwanted side effects are likely to revert if excessive GSK-3 inhibition is halted. Introduction Glycogen synthase kinase-3 (GSK-3) is a serine/threonine kinase that is present in most tissues and that is particularly abundant in the central nervous system (CNS) (Woodgett, 1990). There are two isoforms of the enzyme termed GSK-3α and GSK-3β that are encoded by two different genes (Woodgett, 1990). GSK-3 is known to participate in multiple signaling pathways coupled to receptors for a variety of signaling molecules such as insulin or wnt among many others (Jope and Johnson, 2004). Regulation of GSK-3 is characterized by the fact that the enzyme is active in resting conditions with activation of the different signaling pathways, leading to GSK-3 inhibition by its phosphorylation on the Ser21 or Ser9 residue of GSK-3α and GSK-3β, respectively (Grimes and Jope, 2001). GSK-3 phosphorylation substrates include cytoskeletal proteins, transcription factors, and metabolic regulators, thus leading to a prominent role of GSK-3 in cellular architecture, gene expression, cell division and fate decision, and apoptosis among others (Grimes and Jope, 2001; Jope and Johnson, 2004). Aberrantly increased GSK-3 activity is believed to play a key role in the pathogenesis of chronic metabolic disorders like type-II diabetes (Eldar-Finkelman, 2002), as well as of CNS conditions such as mood disorders and Alzheimer's disease (Avila et al, 2004; Jope and Johnson, 2004). With regard to GSK-3 and neurodegeneration, increased GSK-3 activity has been reported to result in neuronal apoptosis and GSK-3 inhibitors have been shown to exert antiapoptotic and neuroprotective effects in many different cell and mouse models (Pap and Cooper, 1998; Hetman et al, 2000; Beurel and Jope, 2006). Accordingly, potent and specific GSK-3 inhibitors are currently under development in view of their therapeutic potential (Cohen and Goedert, 2004; Frame and Zheleva, 2006; Gould et al, 2006). However, there are two main concerns that counteract the predicted usefulness of GSK-3 inhibitor therapies. On the one hand, the prominent role of GSK-3 in the adenomatous polyposis coli (APC)–β-catenin destruction complex implies that inhibition of GSK-3 could possibly lead to tumor promotion through the activation of β-catenin (Polakis, 2000). On the other hand, despite the above mentioned predicted antiapoptotic effect of GSK-3 inhibition, in vivo sustained inhibition by genetic means (constitutive knock-out mice) revealed an unexpected embryonic lethality due to massive hepatocyte apoptosis (Hoeflich et al, 2000). In this regard, pharmacological inhibition of GSK-3 in cell lines has also been shown to facilitate apoptosis triggered by certain stimuli (Beyaert et al, 1989; Song et al, 2004; Beurel and Jope, 2006). In order to explore the consequences of sustained GSK-3 inhibition in adult tissues, here we have generated transgenic mice carrying a transgene encoding a dominant negative (DN) form of GSK-3β (DN-GSK-3) under control of a tetracycline-controlled conditional promoter. Due to the binary nature of the conditional transgenic system, these mice become a powerful tool because transgene transactivation can be directed to different tissues upon crossing with any of the available driver mice that express tTA (Tet-Off) under control of different promoters (Lewandoski, 2001). To specifically address the neurological consequences of a sustained decrease in GSK-3 activity, we used driver mice with postnatal neuron-specific expression (CamKII-tTA mice; Mayford et al, 1996). Here, we report that sustained expression of DN-GSK-3 in the resulting double transgenic mice (Tet/DN-GSK-3 mice) leads to decreased GSK-3 activity and concomitant decreased phosphorylation of the microtubule-associated protein tau, a well-characterized GSK-3 substrate. Interestingly, Tet/DN-GSK-3 mice grew normally and no evidence of tumor formation was obtained either after gross anatomical examination or after analysis of brain morphology. However, apoptosis was detected in Tet/DN-GSK-3 mice in brain regions involved in motor control. Accordingly, Tet/DN-GSK-3 mice also showed a behavioral deficit in motor coordination. Finally, DN-GSK-3 transgene shutdown by doxycycline administration resulted in normal GSK-3 activity and in full reversal of the motor and of the neuronal apoptosis phenotypes. Results Mouse design Constitutive knock-out of GSK-3β in mice is known to result in embryonic lethality (Hoeflich et al, 2000). As an alternative genetic approach to explore the consequences of a sustained decrease in GSK-3 activity in adult tissues, we decided to generate mice with conditional transgenic expression of a dominant-negative (DN) form of GSK-3. More precisely, the K85R mutant form of GSK-3β (DN-GSK-3) was chosen in view of its previously shown efficacy in decreasing GSK-3 activity (Dominguez et al, 1995). The tetracycline-regulated system can be used for conditional gene expression in mice (Gingrich and Roder, 1998; Lewandoski, 2001). By using this system, we have previously generated inducible mouse models of neurodegenerative diseases (Yamamoto et al, 2000; Lucas et al, 2001). Similarly, here we decided to generate mice with the DN-GSK-3 transgene linked to a tetracycline responsive (tetO) promoter (Figure 1A). These mice can then be bred with mice with tissue-specific expression of the tetracycline transactivator (tTA, also known as Tet-Off). Then, double transgenic mice will express DN-GSK-3 in a tetracycline-repressible manner (Figure 1B). There are many available mouse lines expressing tTA (Lewandoski, 2001). Here, we aimed to explore the neurological consequences of sustained GSK-3 inhibition by using mice that express tTA under control of a postnatal neuron-specific promoter (Cam-KII-tTA mice; Mayford et al, 1996). Figure 1.Generation of Tet/DN-GSK-3 mice and mapping of transgene expression. (A) Diagram showing the structure of the DN-GSK-3 construct used for transgenesis. The bidirectional Bi-tetO promoter is followed in one direction by the K85R-GSK-3β (DN-GSK-3) cDNA sequence with a myc epitope, and by the β-galactosidase (β-gal) sequence with a nuclear localization signal (NLS) in the other direction. (B) Generation of the conditional transgenic Tet/DN-GSK-3 mice (lines 1–4) by breeding the four DN-GSK-3 founder mice (R1, R2, R3 and R4 mice) with mice expressing tTA under control of the CamKIIα promoter (tTA mice). (C) Western blot detection of β-gal in striatum, cortex and hippocampus of 2-month-old Tet/DN-GSK-3 (R1, R2, R3 or R4) mice and control mice. (D–F) β-gal immunohistochemistry on sagittal brain sections from a 2-month-old Tet/DN-GSK-3 mouse (line 2). (D) striatum, (E) cortex and (F) hippocampus. ac, anterior commisure; Cx, cortex, I–VI, cortical layers; cc, corpus callosum; DG, dentate gyrus. Scale bar in panel E corresponds to 300 μm in panels (E, F), and to 200 μm in panel (D). Download figure Download PowerPoint To generate the DN-GSK-3 construct for transgenesis, the sequence of the K85R mutant form of GSK-3β was cloned into a cassette containing the bidirectional tetO (Bi-tetO) promoter linked to a β-galactosidase (β-gal) reporter in divergent orientation (Figure 1A). Our previous experience in generating conditional transgenic mice with the tet-regulated system indicates that the site of insertion and/or the copy number of the tetO construct influences the final pattern and level of transactivation by tTA (Yamamoto et al, 2000; Lucas et al, 2001). The β-gal reporter sequence in the DN-GSK-3 construct permits, on one hand, a quick analysis of the pattern of transgene expression in the double transgenic (Tet/DN-GSK-3) mice from the different founder lines by X-gal staining or by immunohistochemistry against β-gal. On the other hand, it also allows to test the efficiency of transgene silencing after tetracycline administration (Yamamoto et al, 2000; Diaz-Hernandez et al, 2005; Engel et al, 2006). Generation of mice with conditional expression of DN-GSK-3 in forebrain neurons Four independent transgenic mouse founder lines were obtained that carried the DN-GSK-3 construct and they were termed R1, R2, R3 and R4 (Figure 1B). To study the effect of sustained GSK-3 inhibition in adult neurons, these mouse lines were then bred with the Cam-KII-tTA mouse line. In good agreement with the postnatal nature of the driver neuronal promoter, the percentage of double transgenic mice was close to the expected 25% for all founder lines (R1: 29/151, R2: 56/177, R3: 37/167, R4: 54/200 in the first analyzed litters). The level and pattern of expression of the reporter transgene β-gal was then analyzed in 2-month-old Tet/DN-GSK-3 mice by Western blot and by immunohistochemistry. Tet/DN-GSK-3 mice resulting from line R3 showed the lowest level of transactivation by Western blot (Figure 1C) and, as evidenced by immunohistochemistry, this was restricted to the ventral striatum (data not shown). Lines R1, R2 and R4, on the other hand, led to β-gal expression in a spatial pattern very similar to that of endogenous CamKIIα, with expression evident in cortex, hippocampus, striatum, amygdala, and olfactory bulb neurons (Figure 1C–F and data not shown). Since CamKIIα is known to be expressed to a lesser extent in spinal cord (Liang et al, 2004), we also analyzed this structure, but no expression was detected either by Western blot, immunohistochemistry, or X-Gal staining (Supplementary Figure 1). Similarly and as expected, no transgene expression was detected in other brain regions with no expression of CamKIIα, such as cerebellum, brainstem, or thalamus nor in peripheral tissues (Supplementary Figure 1 and data not shown). Expression was highest in the striatum, where the vast majority of neurons showed transgene expression (Figure 1C and D). On the other hand, in the cortex and in the hippocampus, expression was restricted to certain neuronal subpopulations (Figure 1E and F). More precisely, cortical expression was located mainly in layer II–III neurons and hippocampal expression was essentially restricted to neurons in the CA1 field. In good agreement with the β-gal immunohistochemistry data, striatum was the only analyzed brain region, where increased total GSK-3β (endogenous plus transgenic DN-GSK-3β; 0.9 fold increase, P<0.05) was detected by Western blot (Figure 2A), and where some neurons expressed DN-GSK-3β above threshold for detection by immunofluorescence with anti-myc antibody (data not shown). Since R1, R2, and R4 Tet/DN-GSK-3 mice gave a similar pattern and level of transactivation, the three lines were analyzed in parallel. For all the biochemical, histological, and behavioral determinations, the three lines exhibited very similar phenotypes and we will refer to them indistinguishably as Tet/DN-GSK-3 mice. Figure 2.Decreased GSK-3 activity in Tet/DN-GSK-3 mice. (A) Western blot detection of GSK-3β, phosphorylated GSK-3 (pSer21/9 GSK-3α/β), AT-8 and PHF-1 phospho-tau epitopes, and β-catenin levels in homogenates from striatum of Tet/DN-GSK-3 mice and wild-type (wt) littermates. (B) In vitro GSK-3 activity assay performed on striatum, cortex, hippocampus and cerebellum homogenates from Tet/DN-GSK-3 mice and wt littermates (**P<0.001). Download figure Download PowerPoint Tet/DN-GSK-3 mice grew normally and they showed no differences in body weight with respect to their wild-type littermates. Tet/DN-GSK-3 mice also showed a normal lifespan and no evidence of tumor formation was obtained either after gross anatomical examination or after analysis of brain morphology. Reduced GSK-3 activity in the striatum of Tet/DN-GSK-3 mice To verify that the forebrain expression of DN-GSK-3 resulted in decreased GSK-3 activity, we performed GSK-3 enzymatic activity assays on brain lysates as well as Western blot determination of the phosphorylated forms of GSK-3 and of its well-established substrate tau. In good agreement with the fact that striatum was the brain region with the highest level of transgene expression, GSK-3 activity was significantly reduced on striatal homogenates of Tet/DN-GSK-3 mice compared with those of wild-type mice (Figure 2B). A tendency toward decreased activity was observed also in cortical and hippocampal homogenates, but this did not reach statistical significance (Figure 2B). Similarly, no decrease was observed in non-expressing brain regions such as cerebellum. We then analyzed by Western blot the striatal level of the inactive forms of GSK-3 that results from phosphorylation on Ser9 of the β isoform and on Ser21 of the α isoform (p-GSK-3) (Figure 2A). Interestingly, expression of DN-GSK-3β resulted in a dramatic increase (3.2-fold, P<0.001) not only in phospho-Ser9-GSK-3β but also in phospho-Ser21-GSK-3α. Since total GSK-3α levels are not altered in Tet/DN-GSK-3 mice (data not shown), this strongly suggests that the observed net reduction in GSK-3 activity is probably due in part to the previously reported mechanism that amplifies GSK-3 inhibition (Jope, 2003; Zhang et al, 2003). This mechanism is based on increased phosphorylation at the Ser21/9 residues in response to sustained inhibition of GSK-3 either by pharmacological or genetic means, even if the latter is performed in a single isoform (e.g. GSK-3β knock-out fibroblasts; Hoeflich et al, 2000). A moderate increase in the levels of inactive Ser21/9 phospho-GSK-3 was also observed by Western blot in the other two analyzed transgene expressing brain regions, hippocampus, and cortex. However, as for the GSK-3 activity assay data, the cortical and hippocampal increases in Ser21/9 phospho-GSK-3 did not reach statistical significance after normalizing for total GSK-3 levels (data not shown). In good agreement with the marked striatal decrease in GSK-3 activity, the level of phosphorylation of tau was reduced in striatal homogenates of Tet/DN-GSK-3 mice. This was evidenced by Western blot with the AT-8 (72% reduction, P<0.01) and the PHF-1 (46% reduction, P<0.01), antibodies (Figure 2A) that recognize two independent phospho-epitopes that are known to be phosphorylated by GSK-3 (Lovestone et al, 1994). We also analyzed the level of β-catenin that is also a GSK-3 substrate. Since β-catenin phosphorylation by GSK-3 favors its degradation by the proteasome (Aberle et al, 1997), we reasoned that its levels might be increased in Tet/DN-GSK-3 mice. However, as shown in Figure 2A, β-catenin levels were not changed in these mice. We then explored by immunofluorescence the tissue distribution of the detected decrease in tau phosphorylation (Supplementary Figure 2). As expected, given the axonal localization of tau, the decrease in phospho-tau staining occurs diffusely throughout the striatal neurophil. This strongly suggests that the decrease takes place predominantly in the axons of medium size spiny neurons, that represent more than 90% of the neurons in the striatum (see staining with the DARPP-32 marker in Supplementary Figure 2B and E) and that have long axons that project outside the striatum. However, the somas of few large interneurons were also detected by PHF-1 immunofluorescence and the intensity of this staining was diminished in Tet/DN-GSK-3 mice (Supplementary Figure 2A, D, G, and I). By double labeling immunofluorescence we were able to identify these neurons as a subset of the choline acetyltransferase (ChAT)-positive large interneurons (Supplementary Figure 2G–J). In summary, decreased tau phosphorylation seems to take place mainly in the axons of medium size spiny neurons and also in a subset of ChAT-positive large interneurons. Apoptosis detection in the striatum and cortex of Tet/DN-GSK-3 mice Since apoptosis of liver cells was the most prominent phenotype in GSK-3β knock-out mice (Hoeflich et al, 2000), we wondered whether expression of DN-GSK-3 in neurons of Tet/DN-GSK-3 mice might also result in apoptosis. In this regard, although no obvious atrophy takes place in the brain of Tet/DN-GSK-3 mice, in immunofluorescence experiments we observed that some myc-positive neurons showed fragmented nuclei (not shown). To explore in a quantitative manner the incidence of neuronal apoptosis in the forebrain of Tet/DN-GSK-3 mice, we performed cleaved caspase-3 and TUNEL stainings. As anticipated, Tet/DN-GSK-3 mice showed a marked increase in the number of cleaved caspase-3-positive cells in the striatum (Figure 3B–D). Although decreased enzymatic activity was detected only in the striatum, we also analyzed apoptosis in the other brain regions with robust transgene expression (cortex and hippocampus), because apoptosis might take place in specific neuronal subpopulations that show relatively higher expression of the transgene an/or that are particularly vulnerable to apoptosis triggered by decreased GSK-3 activity. In fact, the number of cleaved caspase-3 positive cells was also significantly increased in the cortex of Tet/DN-GSK-3 mice (Figure 3A and D). As expected, this takes place in the external layers (Figure 3A), where transgene expressing neurons are located (see Figure 1E). No significant difference in the number of cleaved caspase-3-positive cells was detected in the hippocampus or in non-expressing regions such as cerebellum (Figure 3D). Besides, we performed double immunofluorescence experiments that confirmed that caspase-3-positive cells were also detected with anti β-gal (Figure 3E and F) and with anti-Neu-N (not shown) antibodies, thus proving that the detected dying cells are transgene expressing neurons. Similar results were obtained with TUNEL staining (Figure 3G and H). Figure 3.Apoptosis detection in striatum and cortex of Tet/DN-GSK-3 mice. (A–D) Immunohistochemical detection and quantification of cleaved caspase-3-positive cells. (A–C) Low-magnification pictures of showing cleaved caspase-3-positive cells (indicated by arrows) in cortex (A), dorsal striatum (B) and ventral striatum (C) of a Tet/DN-GSK-3 mouse. (i–vi) Magnifications of the cleaved caspase-3-positive cells shown in panels (A–C). Scale bar in panel (A) corresponds to 200 μm in panels (A–C). (D) Histogram showing the incidence of cleaved caspase-3 immunopositive cells in striatum, cortex, hippocampus and cerebellum of Tet/DN-GSK-3 mice and their wild-type (wt) littermates. Data are presented as the mean±s.e.m. number of immunopositive cells per 7 mm2 in a 30 μm section (n=9, *P<0.05). (E–F) Double labeling immunofluorescence with anti-cleaved caspase-3 (E) and anti-β-gal (F) antibodies in the striatum of a Tet/DN-GSK-3 mouse. Scale bar in panel (E) corresponds to 10 μm in panels (E and F). (G, H) TUNEL staining in the striatum of a wt (G) and of a Tet/DN-GSK-3 (H) mouse. Scale bar in panel (G) corresponds to 80 μm in panels (G, H); cc, corpus callosum; GP, globus pallidus; CPu, caudate putamen; ac, anterior commisure; NA, nucleus accumbens; Tu, olfactory tubercle. Download figure Download PowerPoint Tet/DN-GSK-3 mice show reduced motor coordination The striatum and the cerebral cortex, the brain regions where increased apoptosis was detected in Tet/DN-GSK-3 mice, are part of the basal ganglia circuit involved in motor control. For this reason we decided to analyze Tet/DN-GSK-3 mice in various tests of motor coordination. We first analyzed the Tet/DN-GSK-3 mice in the rotarod apparatus (Figure 4A–C). After pretraining of mice at constant speed, the rotarod was set to accelerate from 4 to 40 r.p.m. over 5 min and mice were tested four times at 1-h intervals. As shown in Figure 4A, Tet/DN-GSK-3 mice showed a marked deficit in the three first accelerating trials. On average, across the four tests, only 38.6±4.6% of Tet/DN-GSK-3 mice versus 58.8±4.1% of the wild-type mice remained on rod when it reached maximal speed (Figure 4B), and the total time on rod was 262.1±12.2 s for wild type and 187.66±13.4 s for Tet/DN-GSK-3 mice (P<0.007; Figure 4C). The motor coordination deficit in Tet/DN-GSK-3 mice was confirmed in the vertical pole test that measures the time needed to descend along a rough-surfaced pole and detects striatal-dependent motor deficits (Matsuura et al, 1997). As shown in Figure 4D, Tet/DN-GSK-3 mice required almost double amount of time to descend as compared with the wild-type mice (43.15±9.44 versus 25.36±3.62 s; P<0.03). We then performed analysis of stride length in the footprint test. As shown, in Figure 4E, Tet/DN-GSK-3 mice also showed a reduction in stride length (56.31±0.87 versus 53.17±1.07 mm; P<0.03) and higher maximal difference on stride length (23.56±1.95 versus 28.82±1.41 mm; P<0.02), thus further demonstrating a deficit in motor coordination. Finally, to confirm the specificity of the observed motor coordination deficit, we analyze the general motor activity of Tet/DN-GSK-3 mice in the open field test (Figure 4A). Interestingly, no difference between control and Tet/DN-GSK-3 mice was observed either during the first minute (transferal arousal) or in the rest of the testing session. Figure 4.Tet/DN-GSK-3 mice show impaired motor coordination. Motor behavior tests were performed on 3-month-old control and Tet/DN-GSK-3 mice. (A–C) Rotarod. (A) Performance in the four accelerating trials. (B) Average percentage of mice staying on rod in the four trials as the rod accelerates. (C) Histogram showing the mean±s.e.m. latencies to fall from rod in the four trials per genotype (*P<0.05). (D) Time to descend in the vertical pole test (*P<0.03). (E) Average stride length (*P<0.03) and maximal difference in stride length (*P<0.02) in the footprint test. (F) Open field locomotion test. The number of squares entered was counted during the 5 min testing period. Performance in the first minute is also depicted to distinguish the effect of the transferal arousal from the total activity. Download figure Download PowerPoint Reduced dopamine-dependent behavior in Tet/DN-GSK-3 mice The motor deficit in Tet/DN-GSK-3 mice might be explained at least in part by the increased incidence of apoptosis in basal ganglia regions, cortex and striatum. On the other hand, dopamine (DA) neurotransmission from midbrain to the striatum is a key determinant of the activity of striatal neurons and, as a consequence, of motor behavior. Interestingly, striatal GSK-3 activity has recently been shown to be an important mediator of DA action on striatal function and behavior (Beaulieu et al, 2004). It is therefore possible that the observed motor deficit also results from decreased DA-dependent striatal function and behavior in Tet/DN-GSK-3 mice. To explore whether Tet/DN-GSK-3 mice indeed show reduced striatal activation and reduced motor activity in response to DA, wild-type and Tet/DN-GSK-3 mice were challenged with amphetamine to induce a classic DA-dependent behavioral response. As shown in Figure 5A, in vivo microdialysis revealed that systemic administration of amphetamine (2.5 mg/kg, i.p.) induced a similar increase in extracellular DA levels in both groups of mice. Two-way ANOVA revealed a significant effect of amphetamine (F1,15=3.76, P<0.00002), but no significant effect of genotype or genotype × amphetamine interaction. The maximal effect on striatal DA levels was 224±41% of baseline in wild-type mice and 229±39% in transgenic mice. However, despite equivalent striatal DA release, this amphetamine challenge increased locomotor activity in wild-type but not Tet/DN-GSK-3 mice (Figure 5B). In good agreement, the amphetamine-induced expression of c-Fos (a marker of neuronal activity) in striatum was markedly lower in Tet/DN-GSK-3 (Figure 5C–F). Altogether, these results indicate that (a) the dopaminergic (presynaptic) component of the nigrostriatal pathway is unaltered in Tet/DN-GSK-3 mice and (b) the reduction of DA-dependent behaviors in transgenic mice can be attributed to an attenuation of postsynaptic dopaminergic signalling resulting from the stratial expression of DN-GSK-3. Figure 5.Reduced striatal c-Fos induction and reduced motor activity in response to a dopaminergic challenge. Amphetamine (2.5 mg/kg, i.p.) was administered to induce DA release in the striatum. (A) In vivo microdialysis: wild-type (wt) and Tet/DN-GSK-3 mice (n=4 and 8, respectively) were stereotaxically implanted with microdialysis probes in the striatum. Dialysate fractions were collected every 20 min and the concentration of DA in dialysate samples was determined by HPLC. (B) Open field locomotion 31–50 min after saline or amphetamine i.p. injection to wt and Tet/DN-GSK-3 mice (n=4 and 8, respectively). (C–E) Representative images of immunohistochemistry of c-Fos in the striatum after 90 min of a saline injection to a wt mouse (C) or amphetamine (2.5 mg/kg i.p) injections to a wt mouse (D) and a Tet/DN-GSK-3 mouse. (F) H
Referência(s)