Excessive tubulin polyglutamylation causes neurodegeneration and perturbs neuronal transport
2018; Springer Nature; Volume: 37; Issue: 23 Linguagem: Inglês
10.15252/embj.2018100440
ISSN1460-2075
AutoresMaria M. Magiera, Satish Bodakuntla, Jakub Žiak, Sabrina Lacomme, Patricia Marques Sousa, Sophie Leboucher, Torben J. Hausrat, Christophe Bosc, Annie Andrieux, Matthias Kneussel, Marc Landry, A. Calas, Martin Balaštík, Carsten Janke,
Tópico(s)Microtubule and mitosis dynamics
ResumoArticle12 November 2018free access Transparent process Excessive tubulin polyglutamylation causes neurodegeneration and perturbs neuronal transport Maria M Magiera Corresponding Author [email protected] orcid.org/0000-0003-4847-3053 Institut Curie, CNRS UMR3348, PSL Research University, Orsay, France Université Paris-Saclay, CNRS UMR3348, Université Paris Sud, Orsay, France Search for more papers by this author Satish Bodakuntla orcid.org/0000-0002-0448-7683 Institut Curie, CNRS UMR3348, PSL Research University, Orsay, France Université Paris-Saclay, CNRS UMR3348, Université Paris Sud, Orsay, France Search for more papers by this author Jakub Žiak Department of Molecular Neurobiology, Institute of Physiology, Czech Academy of Sciences, Prague 4, Czech Republic Faculty of Science, Charles University, Prague 2, Czech Republic Search for more papers by this author Sabrina Lacomme Bordeaux Imaging Center, BIC, UMS 3420, Université Bordeaux, Bordeaux, France Search for more papers by this author Patricia Marques Sousa Institut Curie, CNRS UMR3348, PSL Research University, Orsay, France Université Paris-Saclay, CNRS UMR3348, Université Paris Sud, Orsay, France Search for more papers by this author Sophie Leboucher Institut Curie, CNRS UMR3348, PSL Research University, Orsay, France Université Paris-Saclay, CNRS UMR3348, Université Paris Sud, Orsay, France Search for more papers by this author Torben J Hausrat Center for Molecular Neurobiology (ZMNH), University Medical Center Hamburg-Eppendorf, Hamburg, Germany Search for more papers by this author Christophe Bosc Grenoble Institut des Neurosciences, GIN, Université Grenoble Alpes, Grenoble, France Inserm U1216, Grenoble, France Search for more papers by this author Annie Andrieux Grenoble Institut des Neurosciences, GIN, Université Grenoble Alpes, Grenoble, France Inserm U1216, Grenoble, France Search for more papers by this author Matthias Kneussel Center for Molecular Neurobiology (ZMNH), University Medical Center Hamburg-Eppendorf, Hamburg, Germany Search for more papers by this author Marc Landry Interdisciplinary Institute for Neuroscience, CNRS UMR5297, Université Bordeaux, Bordeaux, France Search for more papers by this author André Calas Interdisciplinary Institute for Neuroscience, CNRS UMR5297, Université Bordeaux, Bordeaux, France Search for more papers by this author Martin Balastik Department of Molecular Neurobiology, Institute of Physiology, Czech Academy of Sciences, Prague 4, Czech Republic Search for more papers by this author Carsten Janke Corresponding Author [email protected] orcid.org/0000-0001-7053-2000 Institut Curie, CNRS UMR3348, PSL Research University, Orsay, France Université Paris-Saclay, CNRS UMR3348, Université Paris Sud, Orsay, France Search for more papers by this author Maria M Magiera Corresponding Author [email protected] orcid.org/0000-0003-4847-3053 Institut Curie, CNRS UMR3348, PSL Research University, Orsay, France Université Paris-Saclay, CNRS UMR3348, Université Paris Sud, Orsay, France Search for more papers by this author Satish Bodakuntla orcid.org/0000-0002-0448-7683 Institut Curie, CNRS UMR3348, PSL Research University, Orsay, France Université Paris-Saclay, CNRS UMR3348, Université Paris Sud, Orsay, France Search for more papers by this author Jakub Žiak Department of Molecular Neurobiology, Institute of Physiology, Czech Academy of Sciences, Prague 4, Czech Republic Faculty of Science, Charles University, Prague 2, Czech Republic Search for more papers by this author Sabrina Lacomme Bordeaux Imaging Center, BIC, UMS 3420, Université Bordeaux, Bordeaux, France Search for more papers by this author Patricia Marques Sousa Institut Curie, CNRS UMR3348, PSL Research University, Orsay, France Université Paris-Saclay, CNRS UMR3348, Université Paris Sud, Orsay, France Search for more papers by this author Sophie Leboucher Institut Curie, CNRS UMR3348, PSL Research University, Orsay, France Université Paris-Saclay, CNRS UMR3348, Université Paris Sud, Orsay, France Search for more papers by this author Torben J Hausrat Center for Molecular Neurobiology (ZMNH), University Medical Center Hamburg-Eppendorf, Hamburg, Germany Search for more papers by this author Christophe Bosc Grenoble Institut des Neurosciences, GIN, Université Grenoble Alpes, Grenoble, France Inserm U1216, Grenoble, France Search for more papers by this author Annie Andrieux Grenoble Institut des Neurosciences, GIN, Université Grenoble Alpes, Grenoble, France Inserm U1216, Grenoble, France Search for more papers by this author Matthias Kneussel Center for Molecular Neurobiology (ZMNH), University Medical Center Hamburg-Eppendorf, Hamburg, Germany Search for more papers by this author Marc Landry Interdisciplinary Institute for Neuroscience, CNRS UMR5297, Université Bordeaux, Bordeaux, France Search for more papers by this author André Calas Interdisciplinary Institute for Neuroscience, CNRS UMR5297, Université Bordeaux, Bordeaux, France Search for more papers by this author Martin Balastik Department of Molecular Neurobiology, Institute of Physiology, Czech Academy of Sciences, Prague 4, Czech Republic Search for more papers by this author Carsten Janke Corresponding Author [email protected] orcid.org/0000-0001-7053-2000 Institut Curie, CNRS UMR3348, PSL Research University, Orsay, France Université Paris-Saclay, CNRS UMR3348, Université Paris Sud, Orsay, France Search for more papers by this author Author Information Maria M Magiera *,1,2, Satish Bodakuntla1,2, Jakub Žiak3,4, Sabrina Lacomme5, Patricia Marques Sousa1,2,†, Sophie Leboucher1,2, Torben J Hausrat6, Christophe Bosc7,8, Annie Andrieux7,8, Matthias Kneussel6, Marc Landry9, André Calas9, Martin Balastik3 and Carsten Janke *,1,2 1Institut Curie, CNRS UMR3348, PSL Research University, Orsay, France 2Université Paris-Saclay, CNRS UMR3348, Université Paris Sud, Orsay, France 3Department of Molecular Neurobiology, Institute of Physiology, Czech Academy of Sciences, Prague 4, Czech Republic 4Faculty of Science, Charles University, Prague 2, Czech Republic 5Bordeaux Imaging Center, BIC, UMS 3420, Université Bordeaux, Bordeaux, France 6Center for Molecular Neurobiology (ZMNH), University Medical Center Hamburg-Eppendorf, Hamburg, Germany 7Grenoble Institut des Neurosciences, GIN, Université Grenoble Alpes, Grenoble, France 8Inserm U1216, Grenoble, France 9Interdisciplinary Institute for Neuroscience, CNRS UMR5297, Université Bordeaux, Bordeaux, France †Present address: Harvard Medical School, Children's Hospital Boston, Boston, MA, USA *Corresponding author. Tel: +33 1 69863127; Fax: +33 1 69863017; E-mail: [email protected] *Corresponding author. Tel: +33 1 69863127; Fax: +33 1 69863017; E-mail: [email protected] EMBO J (2018)37:e100440https://doi.org/10.15252/embj.2018100440 See also: V Shashi et al (December 2018) and A Akhmanova & CC Hoogenraad (December 2018) PDFDownload PDF of article text and main figures. Peer ReviewDownload a summary of the editorial decision process including editorial decision letters, reviewer comments and author responses to feedback. ToolsAdd to favoritesDownload CitationsTrack CitationsPermissions ShareFacebookTwitterLinked InMendeleyWechatReddit Figures & Info Abstract Posttranslational modifications of tubulin are emerging regulators of microtubule functions. We have shown earlier that upregulated polyglutamylation is linked to rapid degeneration of Purkinje cells in mice with a mutation in the deglutamylating enzyme CCP1. How polyglutamylation leads to degeneration, whether it affects multiple neuron types, or which physiological processes it regulates in healthy neurons has remained unknown. Here, we demonstrate that excessive polyglutamylation induces neurodegeneration in a cell-autonomous manner and can occur in many parts of the central nervous system. Degeneration of selected neurons in CCP1-deficient mice can be fully rescued by simultaneous knockout of the counteracting polyglutamylase TTLL1. Excessive polyglutamylation reduces the efficiency of neuronal transport in cultured hippocampal neurons, suggesting that impaired cargo transport plays an important role in the observed degenerative phenotypes. We thus establish polyglutamylation as a cell-autonomous mechanism for neurodegeneration that might be therapeutically accessible through manipulation of the enzymes that control this posttranslational modification. Synopsis Excessive accumulation of polyglutamylation, a posttranslational modification of the neuronal microtubule cytoskeleton, leads to degeneration of a variety of neurons in mouse brains, and perturbs axonal transport. Removal of two deglutamylating enzymes in mouse brain leads to hyperglutamylation due to unopposed glutamylase activity in the entire brain. Hyperglutamylation can cause the degeneration of various neurons in the mouse brain. Concomitant deletion of the main brain glutamylase can prevent hyperglutamylation-induced neurodegeneration in a cell-autonomous manner. The microtubule-severing enzyme spastin is not responsible for hyperglutamylation-induced neurodegeneration. Axonal transport is perturbed in neurons with tubulin hyperglutamylation. Introduction The microtubule cytoskeleton is a key structural component of neurons, where it carries out a multitude of specialized functions. Microtubules are essential for establishing and maintaining neuronal polarity (Craig & Banker, 1994), regulating neuronal morphology (Liu & Dwyer, 2014), transporting cargo (Franker & Hoogenraad, 2013), and scaffolding signaling molecules to form signaling hubs (Dent & Baas, 2014). Consequently, dysfunction of microtubules can lead to neurodevelopmental disorders as well as to neurodegeneration. The role of the microtubule cytoskeleton in neuronal dysfunction has been widely recognized; however, it was usually associated with abnormalities in a variety of microtubule-interacting proteins (Millecamps & Julien, 2013; Brady & Morfini, 2017), the most prominent being the aggregation of the microtubule-associated protein tau in Alzheimer's disease (Wang & Mandelkow, 2016). More recently, the discovery of patients with mutations in genes encoding the building blocks of microtubules—the α- and β-tubulins—revealed that alterations of the microtubules themselves could be causative for a range of neurodevelopmental and neurodegenerative disorders (Chakraborti et al, 2016), most likely by subtly but significantly altering microtubule properties (Belvindrah et al, 2017). It is thus conceivable that, in a more general manner, mechanisms involved in fine-tuning intrinsic microtubule properties and functions could play causative roles in neuronal dysfunctions. One such candidate mechanism is dysregulation of tubulin posttranslational modifications (PTMs) that are expected to control many microtubule functions in cells, by either changing the physical properties of microtubules or by regulating their interactions with other cellular components (Janke, 2014). MTs can carry a large number of PTMs, such as acetylation, polyglutamylation, polyglycylation, and detyrosination. Polyglutamylation, a PTM that is particularly enriched on neuronal microtubules (Janke & Bulinski, 2011), adds variable numbers of glutamate residues as secondary branches to the main tubulin chain (Eddé et al, 1990), thus generating a range of graded signals. This predestines polyglutamylation to be a fine-tuning mechanism of microtubule functions, a principle that has recently been proven for the best-studied function of this PTM, the control of spastin-mediated microtubule severing (Lacroix et al, 2010; Valenstein & Roll-Mecak, 2016). Polyglutamylation is further expected to regulate intracellular trafficking; however, so far only initial insights into the role of this PTM in molecular-motor tuning are available (Sirajuddin et al, 2014). The recent discovery of a large number of tubulin-modifying enzymes now allows to directly address the role of tubulin PTMs (Janke, 2014). Cytosolic carboxypeptidases (CCPs) (Kalinina et al, 2007; Rodriguez de la Vega et al, 2007) are a class of tubulin-modifying enzymes that remove acidic amino acid residues from the carboxy-termini of peptide chains (Rogowski et al, 2010; Berezniuk et al, 2012), thus controlling two major types of PTMs: First, they reverse polyglutamylation, a PTM catalyzed by polyglutamylases from the tubulin–tyrosine ligase-like (TTLL) family (Janke et al, 2005; van Dijk et al, 2007), and second, CCPs can remove C-terminal, gene-encoded glutamate residues, which converts α-tubulin into Δ2- and Δ3-tubulins (Rogowski et al, 2010; Aillaud et al, 2016) or affects other proteins with acidic C-termini (Tanco et al, 2015). Impaired deglutamylase activity was initially linked to neurodegeneration in a mouse model with early loss of the Purkinje cells in the cerebellum, the Purkinje cell degeneration (pcd) mouse (Mullen et al, 1976). This mouse carries a mutation in the CCP1 gene (Fernandez-Gonzalez et al, 2002), which causes accumulation of polyglutamylation in the cerebellum (Rogowski et al, 2010). However, whether deregulation of this PTM directly causes the degeneration of the Purkinje cells, whether impaired deglutamylase activity is universally deleterious to neurons, and which are the underlying molecular mechanisms have remained open questions. In the present study, we establish excessive polyglutamylation as a general, cell-autonomous cause of neurodegeneration that induces transport defects in neurons. We further show that by manipulating enzymes catalyzing polyglutamylation, it is possible to protect neurons against hyperglutamylation-induced neurodegeneration. Results Excess of TTLL1-mediated polyglutamylation causes neurodegeneration in a cell-autonomous manner In pcd mice, mutation of the deglutamylase CCP1 leads to strongly upregulated levels of polyglutamylation in brain regions known to degenerate, such as the cerebellum or the olfactory bulb (Mullen et al, 1976; Rogowski et al, 2010). While this suggested that excessive polyglutamylation could be causative for neurodegeneration in these mice, it was unclear whether polyglutamylation damages neurons in a cell-autonomous as opposed to a tissue-wide manner. To answer this question, we generated Ccp1flox/flox L7-cre mice. In the brain, the L7 promoter is only active in Purkinje cells (Rico et al, 2004); therefore, CCP1 is deleted specifically in these neurons. At 4 months of age, Ccp1flox/flox L7-cre mice showed ataxia comparable to Ccp1−/− or pcd mice, and had lost most, though not all of the Purkinje cells (see Fig 1A and F). The partial survival of the Purkinje cells at 4 months is most likely due to the late activation of the L7 promoter in certain regions of the cerebellum (Rico et al, 2004), and indeed, at 18 months, all Purkinje cells have degenerated (see Fig 1A and F, Appendix Fig S1). This demonstrates that the loss of CCP1 is a cell-intrinsic cause of neurodegeneration. Figure 1. Excessive tubulin polyglutamylation induces Purkinje cell degeneration in a cell-autonomous manner Fate of Purkinje cells in mice with Purkinje-cell-specific knockout of Ccp1. Ccp1flox/flox L7-cre mice show a massive but incomplete degeneration of Purkinje cells (calbindin-stained; red) at 4 months, while all Purkinje cells are degenerated at 18 months. Scale bar: 500 μm. Schematic representation of enzymatic tubulin polyglutamylation (red points are glutamate residues that are posttranslationally added to C-terminal tails of α- and β-tubulin). In neurons, TTLL1 is generating a large share of the overall polyglutamylation (Janke et al, 2005), and CCP1 is reversing this PTM. Loss of CCP1 induces TTLL1-mediated hyperglutamylation, which can be avoided by TTLL1 inactivation. Fate of Purkinje cells in mice with combinatorial knockout alleles. Knockout of Ccp1 (pcd mouse; Ccp1−/−) results in a complete degeneration of Purkinje cells (calbindin-stained; red), while the additional knockout of Ttll1 selectively in the Purkinje cells (L7-cre) protects the entire Purkinje cell layer from degeneration up to 18 months. Scale bar: 500 μm. Nissl staining of the Purkinje cell layer of the brains from various knockout mice shown in (C) confirms the absence of Purkinje cells in Ccp1−/− and the presence of a wild-type-like Purkinje cell density in Ccp1−/− Ttll1flox/flox L7-cre mice. Scale bar: 20 μm. Purkinje cell layer (PC) and granule cell layer (GC) are indicated. Immunoblot of tissue extracts from cerebella of Ttll1−/− and wild-type mice, probed with the polyE, anti-∆2-tubulin, and 12G10 antibodies. Increasing amounts of extracts were loaded for detection of polyglutamylated proteins with polyE. Schematic representation of the experimental paradigm applied in this figure. The thick line in the cerebella symbolizes the Purkinje cell layer. Download figure Download PowerPoint We next asked whether accumulation of polyglutamylation, rather than the impairment of the second enzymatic activity of CCP1, the removal of gene-encoded C-terminal acidic amino acids from tubulins or other proteins (Rogowski et al, 2010; Tanco et al, 2015), is causative for the degeneration of neurons lacking CCP1. To specifically reduce polyglutamylation exclusively in the Purkinje cells of pcd mice, we deleted the major brain polyglutamylase, TTLL1 (Janke et al, 2005), in these neurons by combining the pcd (Ccp1−/−), Ttll1flox/flox, and L7-cre (Rico et al, 2004) alleles (Fig 1B). While in pcd mice Purkinje cells degenerate mostly in the first month after birth (Mullen et al, 1976), Ccp1−/−Ttll1flox/floxL7-cre mice had an entirely preserved Purkinje cell layers even at older ages up to 18 months (see Fig 1C, D and F, Appendix Fig S1). The absence of Purkinje cell degeneration throughout the entire lifespan of these mice demonstrates that excessive polyglutamylation catalyzed by TTLL1 is the unique cause of neurodegeneration in pcd (Ccp1−/−) mice. The result also excludes the possibility that the removal of gene-encoded glutamate residues by CCP1 is implicated in the disease mechanism, as TTLL1 cannot interfere with this type of modification. This is illustrated by the unaltered levels of ∆2-tubulin in the cerebellum of Ttll1−/− mice (Fig 1E). Finally, the deletion of TTLL1 exclusively in Purkinje cells (Fig 1F) demonstrates that this mechanism functions in a cell-autonomous manner, thus excluding a tissue-wide effect that might have protected these cells in Ccp1−/− Ttll1−/− mice (Berezniuk et al, 2012). Having established TTLL1-mediated polyglutamylation as the leading cause of neurodegeneration in mice lacking CCP1, we ultimately wanted to know whether this role is related to the polyglutamylation of tubulin, or whether other proteins can also be polyglutamylated by TTLL1. We compared cerebellar extracts from Ttll1−/− and wild-type mice by immunoblot with the polyE antibody, which recognizes extended polyglutamate chains irrespective of the protein they are attached to (Rogowski et al, 2010). At normal protein load, Ttll1−/− brain extracts are void of polyE-positive proteins, while wild-type brains show a unique, strong protein band corresponding to α-tubulin (Fig 1E). Even when we overloaded the electrophoresis gel heavily with the extracts (up to 20x), we could not detect any additional protein band apart from α- and β-tubulin (Fig 1E). This shows that in cerebellum, tubulin is by far the most predominant substrate of TTLL1, therefore implying that the absence of Purkinje cell degeneration in Ccp1−/−Ttll1flox/floxL7-cre mice is directly caused by the reduction of tubulin polyglutamylation in these neurons. Taken together, our results demonstrate that the degeneration of Purkinje cells in the absence of the deglutamylase CCP1 is a cell-autonomous process, which is induced by excess of tubulin polyglutamylation catalyzed by TTLL1 (Fig 1B and F). Excess of polyglutamylation can induce degeneration of different neuronal populations In the pcd mouse, only some cell types and brain regions undergo degeneration, such as Purkinje cells in the cerebellum, mitral cells in the olfactory bulb (Mullen et al, 1976; Greer & Shepherd, 1982), thalamic neurons (O'Gorman & Sidman, 1985), and photoreceptors (LaVail et al, 1982; Bosch Grau et al, 2017). However, no obvious signs of neurodegeneration are seen in cerebral cortex and hippocampus, regions that are primarily affected in a range of late-onset human neurodegenerative disorders (Brettschneider et al, 2015). Based on findings in pcd mice, we had suggested that these regions are protected due to a strong expression of a second deglutamylase with similar specificity, CCP6 (Rogowski et al, 2010). This is supported by the observation that initially increased polyglutamylation levels in these regions return to normal in adult mice (Figs 2A and EV1). To test whether such a compensatory mechanism exists, we generated mice that effectively lacked both deglutamylases in the central nervous system using two different strategies (see Methods; referred to as Ccp1−/−Ccp6−/− hereafter). Elevated levels of tubulin polyglutamylation persisted in cerebral cortex and hippocampus of these mice, and even increased further upon aging (see Figs 2A and G, and EV1). As expected from the known enzymatic activities of CCP1 and CCP6, levels of Δ2-tubulin were altered as well, mainly in young animals, while other tubulin PTMs were not, or only marginally, affected (Fig EV1). Click here to expand this figure. Figure EV1. Immunoblot analyses of tubulin posttranslational modifications in brains of deglutamylase-deficient mice (complement to Fig 2A)Representative analysis of posttranslational modifications of tubulin in different regions of the nervous system. Wild-type, Ccp1−/−, Ccp6−/−, and Ccp1−/−Ccp6−/− mice at 3 weeks, 2 and 5 months are compared. Protein extracts from different regions of the central nervous system were adjusted for equal load with the anti-tubulin antibody 12G10, and then systematically probed for glutamylation (GT335), polyglutamylation (polyE), ∆2-tubulin, detyrosinated tubulin (detyr-tub), and acetylation (6-11B-1). We observed an increase in GT335 and polyE labeling, as well as a decrease in ∆2-tubulin labeling in Ccp1−/− and Ccp1−/−Ccp6−/− mice at 3 weeks in all brain regions. In contrast, at 5 months, ∆2-tubulin levels are equal in all genotypes, and hyperglutamylation (polyE) persists only in cerebellum and olfactory bulb of Ccp1−/− mice, thus confirming our previous observations (Rogowski et al, 2010). In contrast, in double-knockout Ccp1−/−Ccp6−/− mice, hyperglutamylation persists in all brain regions up to 5 months. This indicates that in Ccp1−/− mice, the late-expressed CCP6 (Rogowski et al, 2010) compensates for the absence of Ccp1 in older mice. All other tubulin modifications remain unaltered. Immunoblots boxed in green are shown in the Fig 2A. The α- and β-tubulin bands are indicated. Note that only polyE and GT335 recognize both α- and β-tubulins. Download figure Download PowerPoint Figure 2. Excessive polyglutamylation causes neurodegeneration of the cerebral cortex Immunoblot analyses of tubulin polyglutamylation (antibody: polyE) in extracts of different brain regions of wild-type, Ccp1−/−, Ccp6−/−, and Ccp1−/−Ccp6−/− mice. In 3-week-old mice, hyperglutamylation is observed in cerebral cortex, cerebellum, and hippocampus of Ccp1−/− and Ccp1−/−Ccp6−/− mice. In 2- and 5-month-old mice, hyperglutamylation is restricted to the cerebellum in Ccp1−/− mice; however, it persists in all three brain regions of the Ccp1−/−Ccp6−/− mice (complete analysis in Fig EV1; α- and β-tubulin bands are indicated). Nissl-stained frontal sections of wild-type, Ccp1−/−, Ccp6−/−, and Ccp1−/−Ccp6−/− mice at 1 and 5 months, showing age-dependent thinning (arrowheads) of the cortex in Ccp1−/−Ccp6−/− mice (complete analysis in Appendix Fig S2A). Quantification of the thickness of motor and somatosensory cortex of 1- and 5-month-old mice shown in Appendix Fig S2. The single values shown in Appendix Fig S2B were averaged, and values from each genotype-specific group were tested against the average values of the remaining three genotypes (Appendix Fig S2C). Only the cortical thickness of 5-month-old Ccp1−/−Ccp6−/− mice is significantly decreased (mean ± SEM; Mann–Whitney t-test). Immunohistochemistry of brain sections with anti-MAP2 antibody for visualization of neurons (layer V) and apical dendrites (layer IV) and with anti-GFAP antibody for reactive astrocytes (layer V). The complete analysis is shown in Figs EV2A and EV3A. Scale bars: 50 μm. Quantifications of number of MAP2-positive neurons in layer V and of number of apical dendrites in layer IV (Fig EV2B) reveal reduced neuron number in layer V and reduced dendritic density in layer IV of 5-month-old Ccp1−/−Ccp6−/− mice. Quantification of the number of reactive astrocytes in layer V (Fig EV3B) shows a significant increase over wild-type in Ccp1−/− and Ccp6−/− mice; however, a much stronger increase is seen in Ccp1−/−Ccp6−/− mice. Complete analyses are shown in Figs EV2C and EV3C (mean ± SEM; Student's t-test). Immunohistochemistry of brain sections stained with SMI-32 antibody (a representative section of one brain per genotype group is shown, complete analysis in Appendix Fig S3). Neuronal SMI-32 labeling indicates increased axonal damage in the cortex of Ccp1−/−Ccp6−/− mice at 5 months. Scale bar: 100 μm. Schematic representation of the experimental paradigm applied in this figure. Download figure Download PowerPoint To determine the impact of the Ccp1−/−Ccp6−/− double knockout on brain anatomy, we measured the thickness of motor and somatosensory cortices. While cortical thickness appeared normal at 1 month of age, it was significantly reduced at 5 months (Fig 2B and C, Appendix Fig S2), suggesting a degenerative process as opposed to a developmental defect. To confirm this, we quantified the number of pyramidal neurons (MAP2) and reactive astrocytes (GFAP) in cortical layer V as well as the density of apical dendrites of the pyramidal neurons (MAP2) in cortical layer IV (Figs 2D and EV2A and B, and EV3A and B). These analyses revealed a significantly decreased number of pyramidal neurons and their apical dendrites in the cortex of aging Ccp1−/−Ccp6−/− mice (Figs 2E and EV2C), which was inversely correlated with a strong increase in reactive astrocytes (Figs 2E and EV3C). Decreased numbers of neurons and apical dendrites are hallmarks of cortical degeneration [e.g., in the premature aging mouse model, SAMP10 (Shimada et al, 2006), or in amyotrophic lateral sclerosis (Geevasinga et al, 2016)], and increase in reactive astrocytes is also consistent with neurodegeneration (Hol & Pekny, 2015). Finally, labeling with SMI-32 antibody (Kim et al, 2010) is strongly increased in the cerebral cortex of 2- and 5-month-old Ccp1−/−Ccp6−/− mice, but normal at 1 month (Fig 2F and Appendix Fig S3). Considering that SMI-32 is a marker of altered axonal transport and axonal damage (Watson et al, 1991), the progressive degeneration of the cerebral cortex could be linked to axonal defects. Moreover, as these defects coincide with persistent tubulin hyperglutamylation in Ccp1−/−Ccp6−/− cerebral cortices (Fig 2G), our observations suggest a more general role of hyperglutamylation in degeneration of neurons. Click here to expand this figure. Figure EV2. Quantification of neuron and apical-dendrite densities in cortex (complement to Fig 2D and E) Immunohistochemistry of sagittal sections of frontal cerebral cortex of 5-month-old wild-type, Ccp1−/−, Ccp6−/−, and Ccp1−/−Ccp6−/− mice stained with anti-MAP2 antibody to label pyramidal neurons. The number of positive cells was quantified in randomly placed squares (40,000 μm2) in layer V of motor cortex (dotted boxes). Three square areas were quantified in each section. The number of apical dendrites was determined in layer IV of the motor cortex by counting the average number of intersections of apical dendrites with three randomly drawn 200-μm lines (dotted lines) as shown in (B). Two to four sections per animal were analyzed in three mice of each genotype. Scale bars: 100 μm. CC indicates corpus callosum. Example of the analysis paradigm used for the quantification of apical dendrites in layer IV (upper panels) and neuron number in layer V (lower panels). Scale bars: 50 μm. Analyses of number of MAP2-positive neurons in layer V, and number of apical dendrites in layer IV. Mean values ± SEM are plotted, and significance was tested by Student's t-test. All P-values are indicated. A summary of these diagrams focussing on the significant changes is shown in Fig 2E. Download figure Download PowerPoint Click here to expand this figure. Figure EV3. Quantification of reactive astrocytes in cortex (complement to Fig 2D and E) Immunohistochemistry of sagittal sections of frontal cerebral cortex of 5-month-old wild-type, Ccp1−/−, Ccp6−/−, and Ccp1−/−Ccp6−/− mice stained with anti-GFAP antibody to label reactive astrocytes. The number of positive cells was quantified in randomly placed squares (40,000 μm2) in layer V of motor cortex (dotted boxes) as shown in (B). Three square areas were quantified in each section. Two to four sections per animal were analyzed in three mice of each genotype. Scale bars: 100 μm. CC indicates corpus callosum. Example of the quantification of the number of reactive astrocytes in (A). Scale bar: 50 μm. Analyses of numbers of GFAP-positive astrocytes in layer V. Mean values ± SEM are plotted, and significance was tested by Student's t-test. All P-values are indicated. A summary of this diagram focussing on the significant changes is shown in Fig 2E. Download figure Download PowerPoint To determine the cellular basis of the observed degeneratio
Referência(s)