APC/CCdh1-Rock2 pathway controls dendritic integrity and memory
Edited by Masatoshi Takeichi, RIKEN Center for Developmental Biology, Kobe, Japan, and approved March 16, 2017 (received for review September 26, 2016)
Significance
Disruption of neuronal dendrites causes cognitive impairment in Alzheimer’s disease (AD). Rock2, a kinase of the Rho family of proteins, is a dendrite destabilizer that accumulates in the AD brain. However, why Rock2 aberrantly aggregates, causing neuronal integrity loss, is unknown. Here, we show that Rock2 protein stability is controlled by the ubiquitin ligase APC/CCdh1. Accordingly, APC/CCdh1 loss of function in adult neurons increases Rock2 protein and activity, causing dendrite disruption in the cortex and hippocampus, along with memory loss and neurodegeneration, in mice. These effects are abolished by inhibition of Rock2 activity. Thus, the APC/CCdh1-Rock2 pathway may be a novel therapeutic target against neurodegeneration.
Abstract
Disruption of neuronal morphology contributes to the pathology of neurodegenerative disorders such as Alzheimer’s disease (AD). However, the underlying molecular mechanisms are unknown. Here, we show that postnatal deletion of Cdh1, a cofactor of the anaphase-promoting complex/cyclosome (APC/C) ubiquitin ligase in neurons [Cdh1 conditional knockout (cKO)], disrupts dendrite arborization and causes dendritic spine and synapse loss in the cortex and hippocampus, concomitant with memory impairment and neurodegeneration, in adult mice. We found that the dendrite destabilizer Rho protein kinase 2 (Rock2), which accumulates in the brain of AD patients, is an APC/CCdh1 substrate in vivo and that Rock2 protein and activity increased in the cortex and hippocampus of Cdh1 cKO mice. In these animals, inhibition of Rock activity, using the clinically approved drug fasudil, prevented dendritic network disorganization, memory loss, and neurodegeneration. Thus, APC/CCdh1-mediated degradation of Rock2 maintains the dendritic network, memory formation, and neuronal survival, suggesting that pharmacological inhibition of aberrantly accumulated Rock2 may be a suitable therapeutic strategy against neurodegeneration.
Sign up for PNAS alerts.
Get alerts for new articles, or get an alert when an article is cited.
The correct formation and long-term maintenance of the dendritic network are essential for the normal functioning of the brain. In the adult brain, dendrite stability confers mature neurons with the ability to maintain long-term dendritic arbor integrity and integration within networks (1). Loss of dendrite stability is associated with psychiatric disorders and neurodegenerative diseases. Dendritic disruption and loss of dendritic spines and synapses have been reported in schizophrenia and depression, as well as in neurodegenerative conditions, including Alzheimer’s disease (AD), and after an excitotoxic insult during stroke (2, 3). The serine-threonine Rho protein kinase (Rock), an effector of the RhoA GTPase (4), is a central regulator of the microtubule cytoskeleton in neurons. Rock is known to modify the number, morphology, and stability of dendrites in a variety of neuronal cell types, including cortical neurons (1, 5). The overactivation of Rock signaling antagonizes dendrite stability, whereas Rock inhibition promotes microtubule assembly and restores dendritic arbor complexity (6–8). Consistent with these results, Rock has been considered a promising drug target for central nervous system disorders (4). However, the molecular mechanism that regulates Rock abundance and activity in neurological disorders is unknown.
The anaphase-promoting complex/cyclosome (APC/C) is a multisubunit E3 ubiquitin ligase that plays a critical role in controlling both cell-cycle progression and important functions in postmitotic neurons (9, 10). APC/C is activated by two alternative regulatory subunits, namely Cdh1 and Cdc20. In the developing brain, APC/C is involved in the regulation of neuronal differentiation and survival, glial differentiation and migration, axonal growth and patterning, and synapse formation and plasticity (9, 10). In mature neurons, Cdh1 is the main activator of the APC/C ligase (11). However, whereas in brain development the functions of APC/CCdh1 are well-understood (9, 10), its potential functions in the adult brain are largely unknown.
Here we describe that conditional knockout (cKO) of Cdh1 in the pyramidal neurons of the cortex and hippocampus of the adult brain induces dendrite arbor and structure disruption and dendritic spine and synapse loss, which results in impaired learning and memory as well as neurodegeneration. We also found that the expression level and biochemical activity of Rock2, but not Rock1, was increased in damaged brain areas of Cdh1 cKO mice. Furthermore, we show that APC/CCdh1 targets Rock2, but not Rock1, for degradation in the brain. Administration of the clinically approved Rock inhibitor fasudil to Cdh1 cKO mice prevented dendrite disruption, dendritic spine loss, impaired memory and learning, and neurodegeneration. Together, these data reveal an APC/CCdh1-Rock2 pathway that regulates structural stability and functional integrity of dendrites, thus posing Cdh1 as a key molecular factor in the pathogenesis of neurodegenerative disorders.
Results
Cdh1 Deficiency Causes Dendritic Network Disruption and Impaired Neuronal Connectivity, Leading to Memory and Learning Impairment in the Adult Brain.
Cdh1 is essential for neurogenesis and cortical size during brain development (12). To study the function of Cdh1 in the adult brain, here we generated Cdh1 cKO mice by mating mice harboring a floxed allele of the Cdh1 gene (13) with CaMKIIα-Cre mice, which express Cre recombinase from the third postnatal week in a subset of glutamatergic pyramidal neurons, including nearly all CA1 hippocampal neurons and in scattered cortical and other neurons throughout the forebrain (14). Consistent with the temporal activity of the CamKIIα promoter (14), immunoblotting analyses revealed depletion of Cdh1 levels in the cortex and hippocampus, but not in the cerebellum, from postnatal day 25 and continued into adulthood (Fig. 1A). As from postnatal day 120 the brain weight was reduced (Fig. 1 B and C and Fig. S1A), as was the cortex (Fig. 1D and Fig. S1B) and the CA1 layer of the hippocampus, which were thinner in Cdh1 cKO (Fig. 1E and Fig. S1C). This effect was a consequence of decreased number, not soma size, of neurons, as judged by stereological counting of NeuN+ cells (Fig. S1D). Thus, inactivation of APC/CCdh1 in neurons alters age-related growth of the cortex and hippocampus.
Fig. 1.
Fig. S1.
Whereas neurogenesis is the main determinant of embryonic brain growth (12, 15), dendrite length and dendritic arbor complexity are key determinants of adult brain size (16). We found a reduction in dendrite density in the cortex (Fig. 2A and Fig. S2 A and C) and hippocampal CA1 layer (Fig. 2B and Fig. S2 B and C) of Cdh1 cKO mice. Cdh1 knockdown (siCdh1) reduced the dendrite length of primary cortical neurons (Fig. S2D). Furthermore, the dendrite disruption in Cdh1 cKO mice at 120 d (Fig. 2 A and B and Fig. S2C) was not observed in the cerebellum (Fig. S2E), where Cdh1 levels were unchanged (Fig. 1A and Fig. S2F). In addition, dendritic complexity (Fig. S2G) of cortical pyramidal neurons was greatly reduced in Cdh1 cKO mice. Thus, loss of Cdh1 triggers dendrite disruption and reduces dendrite arborization, suggesting that Cdh1 is essential for dendritic network integrity and stability in the adult brain.
Fig. 2.
Fig. S2.
Golgi impregnation analyses revealed that Cdh1 cKO mice displayed lower spine density than controls (Fig. 2C and Fig. S2H). Moreover, the presynaptic proteins vesicular glutamate transporter 1 (VGlut1) and synaptotagmin 1 (SYT1) and postsynaptic markers glutamate receptor subunit NR2B and postsynaptic density protein 95 (PSD95) were strongly reduced in the cortex and hippocampus of Cdh1 cKO mice, indicating synapse loss (Fig. 2D and Fig. S2I). To evaluate the functionality of the neural pathway integrity, we recorded cortical electrical activity in the left hemisphere of mice after sciatic stimulation (17). A marked decrease in the amplitude of evoked potentials but not in latency was observed (Fig. 2F), reminiscent of the dysfunctional neural network connectivity that is observed during neurodegeneration (17). In agreement with this, Cdh1 deficiency induced neuronal apoptosis (Fig. S3 A and B). Thus, Cdh1 deficiency in pyramidal neurons disrupts the dendritic network, leading to dendritic spine and synapse loss, impaired functional brain connectivity, and neurodegeneration.
Fig. S3.
Next, we assessed whether Cdh1 cKO mice show impaired learning, memory, cognition, and anxiety. We observed no differences in motor coordination (Fig. S4A), but learning and memory, as judged by validated tests (18), were impaired in Cdh1 cKO mice (Fig. S4 B and C). These results indicate that Cdh1 loss in the adult cortex and hippocampus triggers learning and spatial memory deficits. Because psychiatric disorders and dementia include anxiety (19), we next performed tests (20, 21) and found that Cdh1 cKO mice showed impaired locomotion/explorative activity and higher levels of anxiety (Fig. S4 D and E). Altogether, these data indicate that Cdh1 depletion in pyramidal neurons of the adult brain impaired hippocampus-dependent spatial learning and memory, reduced locomotion and exploration activities, and increased levels of anxiety, all of which are consistent clinical signs of psychiatric diseases and AD (2, 3). Thus, Cdh1 loss-mediated dendrite arbor disruption in the adult brain may be involved in the pathogenesis of these neurological disorders.
Fig. S4.
APC/CCdh1 Triggers the Ubiquitin-Dependent Degradation of the Dendrite-Destabilizing Protein Rock2.
The activation of Rho signaling through Rock plays the role of a central mediator of dendrite destabilization (1, 4, 22). We noticed that both Rock isoforms, Rock1 and Rock2, contain a conserved KEN box motif, which targets proteins for ubiquitination by the ubiquitin ligase APC/CCdh1 (9). Hence, we analyzed Rock1 and Rock2 protein levels in the brains of Cdh1 cKO and wild-type mice using specific and validated antibodies (Fig. S5A) (23). We found both Rock1 and Rock2 proteins to be expressed in the cortex and hippocampus of wild-type mice (Fig. 3A). However, Rock2, but not Rock1, increased at 120 and 360 d in Cdh1 cKO mice, an effect that was correlated with dendrite disruption (Fig. 3A and Fig. S5B). Because this result suggests that Cdh1 may regulate levels of Rock2 but not those of Rock1, we next silenced Cdh1 in primary neurons. siCdh1 caused an increase in Rock2 but not in Rock1 (Fig. S5C). Inhibition of APC/C activity in cortical primary neurons (Fig. S5C) and cortical and hippocampal slices (Fig. S5D) also increased Rock2, but not Rock1, protein abundances.
Fig. 3.
Fig. S5.
Next, we ascertained whether Rock2 was regulated by APC/CCdh1 activity via ubiquitination and proteasomal degradation. Immunoprecipitation of the APC/C core subunit APC3, in Cdh1 cKO or control mice, followed by immunoblotting against Rock1 or Rock2, revealed that APC/CCdh1 forms a complex with Rock2 but not with Rock1 (Fig. 3B). Because the APC/CCdh1 recognition motif, the KEN box, is conserved both in Rock1 and Rock2, we next aimed to resolve this apparent paradox. In cancer cells, Rock1 is present in the cytosol and Rock2 both in the nucleus and cytosol (23, 24); however, APC is present only in the nucleus (25, 26). Nucleus–cytosol fractionation of primary cortical neurons revealed that Rock2 is present in the nucleus and, more abundantly, in the cytosol; however, Rock1 is exclusively present in the cytosol (Fig. S5E). Interestingly, APC3 and Cdh1 were found in the nucleus, but not in the cytosol (Fig. S5E). The modestly shifted band of Cdh1 in the cytosol (Fig. S5E) likely reflects a hyperphosphorylated—inactive—form of Cdh1 (27). Furthermore, APC3 immunoprecipitation in the nuclear and cytosolic neuronal fractions, followed by immunoblotting against Rock1 and Rock2, revealed that the interaction between APC3 and Rock2 only occurred in the nucleus; however, no interaction between APC3 and Rock1 occurred either in the nucleus or the cytosol (Fig. S5E). Together, these data indicate that, at least in primary neurons and in the in vivo brain, Rock2, but not Rock1, is an APC/CCdh1 substrate.
To ascertain whether APC/CCdh1 targets Rock2 protein for degradation, cortical and hippocampal slices from wild-type mice were first incubated in the presence of the proteasome inhibitor MG132, which resulted in Rock2 protein accumulation (Fig. S5F), indicating that Rock2 normally undergoes proteolysis. Rock2 immunoprecipitation confirmed substantial Rock2 ubiquitination in the hippocampus and cortex of wild-type mice; however, the levels of ubiquitinated Rock2 were significantly lower in cKO Cdh1 mice (Fig. 3C). To confirm Rock2 ubiquitination, we then performed in-cell (HEK293T) ubiquitination assays. Rock2 showed a smeared pattern of ubiquitinated bands that was substantially attenuated in the presence of proTAME (Fig. 3D). Expression of mutant Rock2, in which the KEN motif was substituted by an AAA sequence (Rock2-mut), disrupted the interaction of Rock2 with Cdh1 (Fig. 3E) and attenuated the smeared pattern of ubiquitinated bands in immunoprecipitated Rock2 (Fig. 3F), indicating the direct participation of the KEN box in Rock2 interaction with Cdh1. Altogether, these data demonstrate that, by recognizing the KEN box, APC/CCdh1 ubiquitinates Rock2, targeting it for proteasomal degradation.
APC/CCdh1 Controls Dendritic Network Integrity via the Regulation of Rock2 Activity.
Given that APC/CCdh1 regulates Rock2 levels and activity, we next were prompted to investigate whether the control of dendritic network integrity by APC/CCdh1 occurred via Rock2 activity. First, we aimed to ascertain whether the increased protein levels of Rock in the brain of Cdh1 cKO mice correlated with Rock2 activity, as measured by its ability to specifically phosphorylate Thr853 of myosin phosphatase myosin-binding subunit (MBS) (23). As shown in Fig. S5G, neuronal Cdh1 loss triggered Rock2 activation in the hippocampus and cortex of the adult brain, indicating that APC/CCdh1 regulates Rock2 levels and activity in the adult brain. Next, we took advantage of fasudil, a clinically approved drug that, by inhibiting Rock activity, improves the clinical outcome of ischemic stroke patients (28). We found that i.p. administration of fasudil for 2 mo starting at postnatal day 30 strongly inhibited Rock2 activity in the cortex and hippocampus, as judged by its ability to fully prevent Thr853 MBS phosphorylation (Fig. S5G). Immunostaining against the neuronal markers NeuN and Map2 of brain sections of Cdh1 cKO mice revealed that fasudil treatment partially rescued the reduced thickness and neuronal number of the cerebral cortex (Fig. 4 A and C and Fig. S6A) and hippocampal CA1 layer (Fig. 4 B and D and Fig. S6A). Furthermore, fasudil prevented Cdh1 depletion-induced dendrite disruption in both the cortex and CA1 layer of the hippocampus (Fig. 4E and Fig. S6B) of the adult brain, as revealed by Map2 immunostaining. To confirm these results using a more sensitive technique, Cdh1 cKO mice were cross-bred with mice expressing YFP as a volume label in pyramidal neurons of the hippocampus and in layer 5 of the cerebral cortex (29). Immunostaining against GFP confirmed the dendrite disruption and neuronal loss in the cortex (layer 5) and hippocampus (CA1 layer) (Fig. 4F and Fig. S6C) of Cdh1 cKO mice, which were partially prevented by fasudil. Finally, we observed that fasudil also prevented dendritic spine loss in pyramidal neurons in the cortex of Cdh1 cKO mice (Fig. 4G and Fig. S6D). Interestingly, we noticed that fasudil caused a slight loss of neurons in control mice, as revealed by GFP staining in the cortex and hippocampus (Fig. 4F and Fig. S6C), likely reflecting that trace amounts of Rock activity are essential for optimal neural structure integrity. In addition, it should be noted that fasudil inhibits both Rock1 and Rock2 activities. However, selective silencing of Rock2 (Fig. S5A) abolished the neurite disruption observed in Cdh1–knocked-down primary neurons (Fig. S6E), as judged by quantification of the average neurite length (Fig. S6F) and number (Fig. S6G) per neuron. This suggests that the impact of Rock1, which is present in Rock2-silenced neurons (Fig. S5A), on Cdh1 loss-mediated dendrite disruption is negligible, at least in primary neurons. In contrast, selective silencing of Rock1 (Fig. S5A) had no effect on Cdh1 loss-mediated neurite disruption (Fig. S6 E–G). Thus, although we cannot unambiguously disregard a possible effect of Rock1 inhibition by fasudil in vivo, our data suggest that its effect on dendritic integrity is specifically mediated by Rock2 inhibition. Altogether, our data indicate that Cdh1 depletion in pyramidal neurons of the adult brain increases Rock2 protein and activity, leading to dendrite disruption, dendritic spine loss, and neurodegeneration in the adult brain. Interestingly, fasudil mitigated the locomotion/explorative activity impairment and attenuated anxiety caused by Cdh1 (Fig. 5 A and B and Fig. S6H), as well as improved the learning and memory impairment caused by Cdh1 loss (Fig. 5C). It should be noted that fasudil slightly worsened locomotion and anxiety (Fig. 5 A and B and Fig. S6H) and learning and memory (Fig. 5C) in control mice, confirming the importance of baseline Rock activity for neural integrity and function. Together, our data indicate that depletion of Cdh1 in pyramidal neurons increases Rock2 protein levels and activity in adulthood, causing dendrite network disruption, anxiety, and impaired learning and memory.
Fig. 4.
Fig. 5.
Fig. S6.
Discussion
We describe a signaling pathway in which the E3 ubiquitin ligase APC/CCdh1 controls the stability and integrity of neuronal dendrites in the adult brain by destabilizing the Rho kinase protein Rock2. Thus, knockout of Cdh1, specifically in the pyramidal neurons of the cerebral cortex and hippocampus, induces dendrite structure disruption and dendritic spine and synapse loss, resulting in impaired neuronal connectivity, deficits in learning and memory, and neurodegeneration. Furthermore, we identify that the mediator of dendrite destabilization, Rock2, is an APC/CCdh1 substrate in vivo by direct interaction between Cdh1 and Rock2 through its KEN box. Thereby, depletion of Cdh1 in pyramidal neurons triggers accumulation and activation of Rock2 in the affected areas of the adult brain. Finally, administration of the clinically approved Rock inhibitor fasudil prevents dendrite disruption, impaired learning and memory, and neuronal loss induced by Cdh1 knockout. Our results therefore demonstrate an APC/CCdh1-Rock2 signaling pathway that regulates structural and functional integrity of the dendritic network in the adult forebrain.
Interestingly, at postnatal day 25, Cdh1 cKO mice exhibited normal, mature dendrite structure and branching in cortical and hippocampal CA1 pyramidal neurons. This contrasts with the reduced cortical size and microcephaly phenotype upon deletion of Cdh1 seen at embryonic stages (12). Thus, in the developing brain, the stabilization of dendrites critically depends on synapse formation, whereas in the mature nervous system, dendritic network stability depends on the microtubule cytoskeleton (1). Nevertheless, the normal dendrite structure observed in Cdh1 cKO mice at postnatal day 25 was not maintained afterward, when dendrite disruption, loss of dendritic branches, and dysfunction of neural network connectivity occurred. These changes caused a thinning of anatomical layers in the cortex and hippocampus in adult mice, as it has been reported that dendrite disruption and reduced dendritic arbor complexity determine brain size in adult animals (16). Furthermore, the reduction in dendritic spine density and synapses altered synaptic connectivity that, in the hippocampus, likely contributed to the anxiety and impaired learning, cognition, and memory phenotypes that recapitulate the alterations seen in a variety of psychiatric and neurodegenerative disorders (2, 3). Thus, APC/CCdh1 maintains structural and functional integrity of dendritic networks, suggesting that Cdh1 is important for the molecular pathogenesis of memory disorders.
The identification of the microtubule-destabilizing protein Rock2 as an APC/CCdh1 substrate may have important implications for our understanding of the mechanism behind dendrite stability in the adult brain. Dendritic microtubules are enriched in microtubule-associated protein Map2, which promotes microtubule polymerization and dendritic arbor stabilization (1, 30, 31). Even though Rock2 disrupts dendrite architecture through several possible mechanisms, it is known that loss of Map2-mediated microtubule instability is an important contributing factor in Rock2-induced dendrite arbor disruption (1, 5). Thus, Rock2 phosphorylates Map2 at Ser1796 (32), a critical residue at the microtubule-binding region, thus reducing its ability to bind microtubules for correct assembly (32, 33). We show that conditional KO of Cdh1 promoted an age-dependent reduction in Rock2 levels, thus correlating with Map2 loss and dendrite disruption. Notably, reduction of dendritic arbor complexity and loss of dendritic spines were prevented by administration of the Rock2 inhibitor fasudil to Cdh1 cKO mice. Therefore, we can conclude that APC/CCdh1 promotes the degradation of Rock2 to ensure proper stability of dendrites and maintenance of dendritic arbors and synapses in the adult brain.
Plasticity of neural circuits relies on dynamic changes in the cytoskeleton of dendritic spines (34), in which Rock2 plays a pivotal role. The RhoA-Rock2 pathway mediates remodeling of dendritic spine morphology and density downstream of glutamate receptor activation (35, 36). Therefore, the APC/CCdh1-Rock2 signaling pathway is consistent with the previously described effect of APC/CCdh1 on synaptic plasticity, including long-term potentiation, mGluR-dependent long-term depression, and homeostatic synaptic plasticity (10, 37–39).
Human neuropathology data reveal that dendritic defects in AD—including dendrite disruption, reduced arbor complexity, and loss of spines—are widespread and occur early in the disease (40). Interestingly, we have observed these features in Cdh1 cKO mice. It has also been suggested that the pathological outcome of neurodegenerative conditions is not necessarily originated by neuronal loss but by subtle changes in dendrites and spines and/or synapses that limit neuronal functionality within the network (3). Moreover, the gradual loss of microtubule mass in neurons is thought to occur by destabilization and depolymerization of microtubules, leading to dendrite disruption (41). Our findings that APC/CCdh1 controls the stability of Rock2 in neurons support this notion, and identify a potential therapeutic target against AD. Interestingly, although no effective therapy is known to combat the progression of AD, studies in animal models provide strong evidence that maintaining dendritic network integrity may ease the symptoms and slow down disease progression (3, 42). Furthermore, Rock2 levels increase in the very early stages of AD and remain elevated throughout the course of the disease (43), and Rock2 inhibition reduces amyloid-β levels (43, 44) and attenuates amyloid-β–induced neurodegeneration (45). Therefore, the reduction in dendrite disruption, and the improvement in learning and memory observed after fasudil administration to Cdh1 cKO mice, further supports the notion that this clinically approved Rock inhibitor drug should be considered as a drug to treat AD.
In conclusion, here we describe an APC/CCdh1-Rock2 signaling pathway that regulates structural and functional integrity of the dendritic network in the adult forebrain, which may have important implications for the pathology of psychiatric and neurodegenerative disorders. Importantly, we have previously described that glutamate receptor overactivation, a hallmark of neurodegenerative diseases, promotes Cdk5-induced Cdh1 phosphorylation (9, 27) and inactivation, which may explain the accumulation of Rock2 that has been found in these disorders (43). Therefore, pharmacological inhibition of aberrantly accumulated Rock2 with fasudil treatment may be a suitable therapeutic strategy against these neurological diseases. Beyond neural tissue, it is known that Rock activity, through its actions on cytoskeletal dynamics, promotes tumor cell invasion and metastasis (46, 47). Whether the APC/CCdh1-Rock2 pathway also represents a new mechanism in cancer progression is a tempting possibility that remains to be elucidated.
Materials and Methods
CamKIIalpha-Cre–Mediated Cdh1 Conditional Knockout Mice.
The Cdh1lox/lox-targeted mouse model (13) was crossed with transgenic mice carrying the gene encoding Cre recombinase under the control of the Camk2α-cre promoter (14) (SI Materials and Methods).
All animals were bred and maintained at the Animal Experimentation Service of the University of Salamanca in accordance with Spanish legislation (RD 53/2013). Procedures and protocols have been approved by the research Bioethics Committee of the University of Salamanca.
Cell cultures and transfections.
Primary cultures of cortical neurons (12) and transfections (27) were prepared as previously described (SI Materials and Methods).
Coimmunoprecipitation assay.
Immunoprecipitation of endogenous (39) and exogenous (27) proteins was performed as previously described (SI Materials and Methods).
Rock2 ubiquitination assay.
Tissue slices were incubated with MG132 (20 µM) for 1 h, and cortex and hippocampus were microdissected, lysed, and immunoprecipitated with anti-Rock2 antibody, followed by immunoblotting with anti-ubiquitin antibody (39) (SI Materials and Methods).
Immunohistochemistry.
Immunohistochemistry was performed according to a previously published protocol (48) (SI Materials and Methods).
Neuronal counting.
Brain areas of interest were identified as in Paxinos and Watson (49) in NeuN-stained 40-μm-thick sections and used for cell counting by an author blinded to genotype and treatment (50, 51) (SI Materials and Methods).
Behavioral studies.
One hundred and twenty-day-old mice were handled for 3 d to acclimate them to the experimenter before subjecting them to the experimental procedures. All behavioral procedures were video-recorded and scored by an individual blind to the genotype of the mouse (SI Materials and Methods).
Statistical Analysis.
The results are expressed as means ± SEM. A one-way ANOVA followed by the Bonferroni post hoc test was used for pairwise comparisons within multiple samples. The Student’s t test was used to compare the means of two independent groups. In all cases, P values < 0.05 were considered significant. Statistical analysis was performed using SPSS Statistics 22.0 for Macintosh.
SI Materials and Methods
CamKIIalpha-Cre–Mediated Cdh1 Conditional Knockout Mice.
The Cdh1lox/lox-targeted mouse model has been described previously (13). For the conditional inactivation of Cdh1 in glutamatergic pyramidal neurons of the cortex and hippocampus of the adult brain, we used transgenic mice carrying the gene encoding Cre recombinase under the control of the Camk2α-cre promoter [B6.Cg-Tg(Camk2α-cre)T29-1Stl/J; The Jackson Laboratory] (14). To generate Cdh1loxP/loxPCamk2α-cre (referred to as Cdh1 cKO) and Cdh1loxP/loxP control mice, mice harboring a Cdh1loxP/loxP allele were crossed with Camk2α-cre mice. When indicated (control-YFP and Cdh1 cKO-YFP), double-transgenic mice Cdh1loxP/loxPCamk2α-cre (Cdh1 cKO) were crossed with Thy1-YFP-H [B6.Cg-TgN(Thy-YFP)HJrs/J; The Jackson Laboratory]. The resulting progeny selectively express yellow fluorescent protein (YFP) in pyramidal neurons in the hippocampus and in layer 5 of the cerebral cortex. YFP fills the entire dendritic tree, providing a Golgi-like stain (29). Mice were kept on a C57BL/6J background. Tail DNA was analyzed by PCR using the following primer sets: Cdh1 mutant mice (286 bp amplified), forward primer 5′-AGCATGGTGACCGCTTCATCC-3′ and reverse primer 5′-TGGCTGGGGGACTTCTCATTTTCC-3′ (13); CamKIIα transgene (150 bp amplified), forward primer 5′-CCGGTTATTCAACTTGCCACC-3′ and reverse primer 5′-CTGCATTACCGGTCGATGCAAC-3′; and Thy1-YFP transgene (415 bp amplified), forward primer 5′-ACAGACACACACCCAGGACA-3′ and reverse primer 5′-CGGTGGTGCAGATGAACTT-3′. PCR products were separated on agarose gels (3%) and Midori Green Advance-stained DNA fragments (Nippon Genetics Europe) were analyzed under a UV source, using the Bio-Rad Universal Hood II Molecular Imager System. All animals were bred and maintained at the Animal Experimentation Service of the University of Salamanca in accordance with Spanish legislation (RD 53/2013). Procedures and protocols have been approved by the research Bioethics Committee of the University of Salamanca. To inhibit Rock activity, a dosage of 20 mg/kg body weight fasudil (Selleck Chemicals) was injected intraperitoneally every other day into 30-d-old mice. Mice were killed at 60 d after the treatment began. Control mice received saline solution (43).
Cell cultures and transfections.
Primary cultures of cortical neurons were prepared from embryonic day 14.5 mouse embryo cortices (12). Cells were seeded at 2.0 × 105 cells per cm2 in DMEM (Sigma) supplemented with 10% (vol/vol) FCS (Roche Diagnostics) and incubated at 37 °C in a humidified 5% CO2-containing atmosphere. At 4 h after plating, the medium was replaced with Neurobasal medium (Invitrogen, Thermo Scientific) supplemented with 2% B27 (Invitrogen) and glutamine (4 mM; Invitrogen). At 8 to 9 d in culture, neurons were transfected with plasmids (27) and/or siRNA using Lipofectamine RNAiMAX (Invitrogen) following the manufacturer’s instructions, and cells were used after 72 h. Specific knockdown of proteins was achieved by using small interfering double-stranded ribonucleotides (siRNAs) to target the coding sequence of the mouse Cdh1 (27), Rock1, and Rock2 (52) mRNA. When indicated, specific knockdown of Cdh1 was performed by using pSuper-neo.gfp (Oligoengine) including the small hairpin sequences for luciferase (shControl) or Cdh1 (shCdh1) (27).
Immunocytochemistry.
Neurons grown on glass coverslips were fixed with 4% (vol/vol; in PBS) paraformaldehyde and immunostained with anti-Map2 (1:100; AP-20; Sigma-Aldrich) and anti-GFP (1:1,000; Sigma-Aldrich). Immunolabeling was detected by using Alexa 568-conjugated goat anti-mouse (1:500; Jackson ImmunoResearch) or Alexa 568-conjugated goat anti-rabbit (1:500; Molecular Probes, Invitrogen). Coverslips were washed, mounted in SlowFade Light Antifade Reagent (Invitrogen) on glass slides, and examined using a spectral laser confocal microscope (TSC-SL; Leica Microsystems). Quantification of the average dendrite length (Map2 staining) and neurite length and number (GFP staining) per neuron was performed using the plugin NeuronJ 1.4.0 (ImageJ, version 1.47; National Institutes of Health). Values are mean ± SEM from 60 neurons per group measured in three different neuronal cultures.
Protein extracts.
After the animals were killed, the brain was quickly removed from the skull, and cerebral cortex, cerebellum, and hippocampus were obtained. These brain tissues were homogenized in RIPA lysis buffer (1% SDS, 2 mM EDTA, 2 mM EGTA, and 50 mM Tris, pH 7.5), supplemented with protease and phosphatase inhibitors (100 µM phenylmethylsulfonyl fluoride, 50 µg/mL anti-papain, 50 µg/mL pepstatin, 50 µg/mL amastatin, 50 µg/mL leupeptin, 50 µg/mL bestatin, 1 mM o-vanadate, 50 mM NaF, and 50 µg/mL soybean trypsin inhibitor) and boiled for 5 min. Extracts were centrifuged at 17,500 × g at 4 °C for 30 min. The supernatants were collected and stored at −80 °C until use. The amount of protein in each sample was measured using the BCA Protein Assay Kit (Pierce, Thermo Scientific).
Western blotting.
Aliquots of tissue lysates (40 to 60 µg protein) were subjected to SDS/PAGE on 6, 8, 10, or 15% acrylamide gels (Mini-PROTEAN; Bio-Rad) including BenchMark (Invitrogen, Thermo Scientific) or Dual Color Standards (Bio-Rad) as prestained protein ladders. The resolved proteins were transferred electrophoretically to nitrocellulose membranes (Hybond ECL; GE Healthcare Life Sciences). Membranes were blocked with 5% (wt/vol) low-fat milk in 20 mM Tris, 500 mM NaCl, and 0.1% (wt/vol) Tween 20 (pH 7.5) for 1 h. After blocking, membranes were immunoblotted with anti-Cdh1 (AR38; a gift from J. Gannon, Cancer Research UK, London), anti-GAPDH (6C5; Ambion), anti-VGlut1 (317G6; Synaptic Systems), anti-SYT (Synaptic Systems), anti-PSD95 (6G6-1C9; Affinity BioReagents), anti-NR2B (a gift from A. Fernández-López, University of León, León, Spain), anti-NeuN (A-60; Merck Millipore), anti-Map2 (AP-20; Sigma-Aldrich), anti-Tuj1 (Abcam), anti-active caspase-3 (Asp175; Cell Signaling Technology), anti-RhoA (26C4; Santa Cruz Biotechnology), anti-Rock2 (H-85; Santa Cruz Biotechnology), anti-Rock1 (H-85; Santa Cruz Biotechnology), anti-APC3 (35/CDC27; BD Pharmingen, BD Biosciences), anti-ubiquitin (Abcam), anti-hemagglutinin (HA; 2-2.2.14; Thermo Scientific), anti-myc (Sigma-Aldrich), anti-MBS (BioLegend), anti-phospho(Thr853)-MBS (MyBioSource), and anti-GFP (Sigma-Aldrich) antibodies, at dilutions ranging from 1:500 to 1:1,000, overnight at 4 °C. GAPDH was used as loading control. After incubation with horseradish peroxidase-conjugated goat anti-rabbit IgG (Pierce, Thermo Scientific) or goat anti-mouse IgG (Bio-Rad), membranes were immediately incubated with enhanced chemiluminescence SuperSignal West Dura (Pierce) for 5 min before exposure to Kodak XAR-5 film for 1 to 5 min, and the autoradiograms were scanned. Band intensities were quantified using ImageJ software (27).
Coimmunoprecipitation assay.
For immunoprecipitation of endogenous proteins, tissue slices were first cross-linked with 1.2% formaldehyde for 7 min and quenched with 1.25 M glycine/PBS. The cortex (Cx) and hippocampus (Hy) were microdissected and lysed in ice-cold lysis buffer (50 mM Tris⋅HCl, pH 7.5, 150 mM NaCl, 2 mM EDTA, and 1% Nonidet P-40, supplemented with the phosphatase and protease inhibitors cited above). Lysates were incubated with anti-APC3 (1 μg) for 4 h at 4 °C, followed by the addition of 30 μL protein G Sepharose (GE Healthcare Life Sciences) for 2 h at 4 °C. Immunoprecipitates were extensively washed with lysis buffer and detected by Western blotting analysis (39).
For immunoprecipitation of exogenous proteins, 293T cells were transfected with plasmids encoding Cdh1 fused to HA (HA-Cdh1) (11) and wild-type Rock2 fused to myc (Myc-Rock2-wt; a gift from M. Olson, Beatson Institute for Cancer Research, Glasgow, United Kingdom) or KEN box-mutated Rock2 fused to myc (Myc-Rock2-mut) using Lipofectamine 2000 (Invitrogen). Cells were cross-linked and lysed as described above. Lysates were incubated with anti-HA agarose beads (Invitrogen) for 2 h at 4 °C and extensively washed with lysis buffer. Immunoprecipitated proteins were detected by Western blot analysis (27).
Rock2 ubiquitination assay.
Tissues slices were incubated with MG132 (20 µM) for 1 h, and cortex and hippocampus were microdissected and lysed in buffer containing 0.1% SDS, 1% Nonidet P-40, 0.5% Na-deoxycholate, 150 mM NaCl, 50 mM Tris⋅HCl (pH 7.4), 2 mM EDTA, 40 mM N-ethylmaleimide, and 1 mM DTT, supplemented with the phosphatase and protease inhibitors cited above (39). Immunoprecipitation of endogenous Rock2 was performed with anti-Rock2 antibody, followed by immunoblotting with anti-ubiquitin antibody.
HEK293T cells were cotransfected with HA-ubiquitin (HA-Ub) and Myc-Rock2-wt or Myc-Rock2-mut (KEN box mutated to AAA) for 24 h. Cells were pretreated with MG132 (20 µM for 2 h) either in the absence or presence of the APC/C inhibitor proTAME (10 µM) and immunoprecipitated with anti-Myc agarose beads followed by immunoblotting with Myc and HA antibodies (39).
Immunohistochemistry.
Animals were deeply anesthetized by i.p. injection of a mixture (1:4) of xylazine hydrochloride (Rompun; Bayer) and ketamine hydrochloride/chlorbutol (Imalgene; Merial) using 1 mL of the mixture per kg of body weight, and then perfused intraaortically with 0.9% NaCl followed by 5 mL/g body weight of Somogy’s fixative [4% (wt/vol) paraformaldehyde, 0.2% (wt/vol) picric acid in 0.1 M phosphate buffer, pH 7.4]. After perfusion, brains were dissected out sagittally in two parts and postfixed, using Somogy’s fixative, overnight at 4 °C. Brain blocks were then rinsed successively for 10 min, 30 min, and 2 h with 0.1 M phosphate buffer solution (PBS; pH 7.4) and sequentially immersed in 10, 20, and 30% (wt/vol) sucrose in PBS until they sank. After cryoprotection, 40-µm-thick sagittal sections were obtained with a freezing-sliding cryostate (Leica; CM1950 AgProtect) and collected in 0.05% sodium azide (wt/vol) in 0.1 M PBS. Sections were rinsed in 0.1 M PBS three times each for 10 min and then incubated in (i) 1:1,000 anti-NeuN (A-60; Merck Millipore) or 1:500 anti-Map2 (AP-20; Sigma-Aldrich) in 0.2% Triton X-100 (Sigma-Aldrich) and 5% goat serum (Jackson ImmunoResearch) in 0.1 M PBS for 72 h at 4 °C; (ii) fluorophore-conjugated secondary antibodies (Jackson ImmunoResearch) in 0.05% Triton X-100 and 2% goat serum in 0.1 M PBS for 2 h at room temperature; or (iii) 0.5 µg/mL DAPI in PBS for 10 min at room temperature (48). After rinsing with PBS, sections were mounted with Fluoromount (Sigma-Aldrich) aqueous mounting medium.
Imaging.
Sections were examined with epifluorescence and appropriate filter sets using a microscope (Nikon; Eclipse Ti-E inverted microscope) equipped with a precentered fiber illuminator (Nikon; Intensilight C-HGFI) and B/W CCD digital camera (Hamamatsu; ORCA-ER). Confocal images were acquired using a spectral laser confocal microscope (TCS-SL; Leica Microsystems) and a commercial inverted confocal microscope (TCS SP5; Leica Microsystems). Large fields of view were acquired with an HCX Plan Apo CS2 40× oil objective (N.A. 1.30) with a pixel size of ∼300 nm at a scan speed of 400 Hz using three line averages. High-resolution images were acquired using an HCX Plan Apo CS2 63× oil objective (N.A. 1.40) with a pixel size of 100 nm and a z-step size of 290 nm at a scan speed of 400 Hz using three line averages.
Immunohistochemistry digital images (RGB images) were used for measuring the thickness of the cortex and hippocampal CA1 layer, using the NIH image-processing package ImageJ (version 1.47). Values are mean ± SEM from 20 measurements from four different animals (n = 4).
The degeneration of dendrites in the cortex and CA1 layer of the hippocampus was assayed by analyzing the density of Map2-positive dendrites in three sections per animal. Fluorescence 8-bit images were acquired as z stacks using an HCX Plan Apo CS2 63× oil objective and an inverted confocal microscope. Images were exported into ImageJ in tiff format for processing. Before image analysis, a maximum-intensity projection over z-series projections spanning 18 to 19 µm was performed. Images were converted to grayscale 8-bit images and brightness/contrast was adjusted using the ImageJ “auto” function. All Map2-positive dendrites were automatically delineated using the “auto setting threshold” (default method) and “dark background” functions of ImageJ. Thresholded images were subsequently quantified as percent area (area fraction) using the “analyze-measure” function, which represents the percentage of pixels in the image that have been highlighted (% area) (53). Values are mean ± SEM from 20 measurements from four different animals (n = 4).
Stereological counting.
Areas of interest in the cortex and hippocampal CA1 layer were identified as in Paxinos and Watson (49) in NeuN-stained 40-μm-thick sections and used for cell counting by an author blinded to genotype and treatment. Neurons (NeuN+ cells) were quantified using the Cavalieri method—which determines the reference volume contained in the cells of interest (Vref)—and the optical dissector method—which determines the density of neurons (i.e., the number of NeuN+ cells per µm3) within Vref (Nv). Vref was determined every 6th section, and Nv every 12th section. The size of the counting frame was 50 μm × 50 μm (cortex) or 25 µm × 25 µm (hippocampal CA1 layer) in a dissector height of 12 µm. With these counting frame areas, we first established to sample ∼200 to 250 sites per brain region per animal to give a coefficient of error (CE) <0.1 using the smoothness factor m =1. The total number of neurons (N) was defined as Vref × Nv (50). These analyses were carried out using a Leica TSC-SL microscope equipped with a 4× objective lens (for Vref) and a 60× objective lens with a 1.4 numerical aperture condenser (for Nv) and NIS-Elements AR software (version 4.20.00; Nikon). The microscope was also equipped with a motor-driven stage to move within the x and y axes and an attached microcator to determine the z axis.
Neuron number was also quantified using the isotropic fractionator method (51). The cortex and CA1 layer of the hippocampus were fixed and disrupted by homogenization. The nuclear suspension obtained after centrifugation (4,000 × g) was collected and immunostained with NeuN antibody (1:300) followed by incubation with the secondary antibody for 3 h at room temperature. Aliquots from the nuclear suspension were introduced into a hemocytometer (Neubauer chamber), and the average neuronal nucleus (NeuN+) density was quantified using a fluorescence microscope (Provis AX70; Olympus).
Golgi staining.
Animals were deeply anesthetized and intraaortically perfused as described above. Brains were incubated in a dichromate-Colonnier solution [3% K2Cr2O7 (wt/vol), 3% glutaraldehyde (vol/vol)] for 7 d followed by cleaning with distilled water, and then impregnated by immersion in 0.75% AgNO3 (wt/vol) for 3 d. Coronal sections (100 µm2) were obtained using a vibratome and mounted onto slides and coverslipped with an Entellan solution (Merck Millipore). Golgi staining was quantified using ImageJ. Grayscale 8-bit images were acquired using a microscope (Nikon; inverted microscope) equipped with a B/W CCD digital camera (Hamamatsu). Images were exported into ImageJ in tiff format and brightness/contrast was adjusted using the ImageJ “auto” function. Golgi staining was thresholded using the “auto setting threshold” (default method) function and subsequently quantified as percent area (area fraction) using the “analyze-measure” function, which represents the percentage of pixels in the image that have been highlighted (% area) (53). Values are mean ± SEM from 15 measurements from three different animals (n = 3).
Terminal deoxynucleotidyl transferase dUTP nick end-labeling assay.
TUNEL assay was performed in brain sections following the manufacturer’s protocol (Roche Diagnostics). Brain sections, fixed as above, were preincubated in TUNEL buffer containing 1 mM CoCl2, 140 mM sodium cacodylate, and 0.3% Triton X-100 in 30 mM Tris buffer (pH 7.2) for 30 min. After incubation at 37 °C with the TUNEL reaction mixture containing terminal deoxynucleotidyl transferase (800 U/mL) and nucleotide mixture (1 μM) for 90 min, sections were rinsed with PBS and counterstained with Cy3-streptavidin (Jackson ImmunoResearch) (48).
Electrophysiological measurements.
One hundred and twenty-day-old mice were anesthetized and placed in a mouse stereotaxic frame (model 1900; Kopf) with a digital coordinate readout system (Wizard 550 Readouts; Anilam). Once the surgical area was disinfected, a small cranial burr hole was drilled (model 1911; Kopf) with a 0.50-mm drill bit through the skull over the left primary motor cortex (+1.98 mm anterior and +2.0 mm lateral to bregma), where the recording concentric tungsten electrode (1 Mohm/kHz; World Precision Instruments) was placed. The sciatic nerve in the contralateral paw was dissected to place a self-building stimulation bipolar electrode. We recorded both spontaneously, to record the cortical activity over 1 min and the evoked potential after stimulation of the sciatic nerve. Square-wave pulses of 0.1 nA and 0.1 ms with a frequency of 0.5 Hz were administered by using a stimulator (Master-8; AMPI Equipment) and an isolator (ISO-Flex; AMPI Equipment) at the sciatic nerve with the appropriate stimulation electrode. At the same time, cortical activity was recorded, from 60 to 90 stimulation cycles. Digital treatment of the signal was identical in both cases: an initial band-pass filter between 0.3 Hz and 10 kHz and analog-to-digital conversion with a sampling frequency of 6,250 Hz with an interface (Power1401-Plus; CED Products). A minimum of 30 sweeps was averaged out to obtain evoked potentials. Amplitudes and latencies of evoked potentials were obtained from the averages of recordings. Latencies were calculated as the time elapsed between the onset of stimulus and the peak of the evoked response, whereas amplitude was measured as the waves’ peak-to-peak voltage difference. Latency mainly reflects nerve demyelinization in the pathway and amplitude correlates with neurite and neural pathway integrity (17).
Behavioral studies.
One hundred and twenty-day-old mice were handled for 3 d to acclimate them to the experimenter before subjecting them to the experimental procedures. Mice were placed in the experiment room at least 60 min before beginning any behavioral protocol. Unless indicated otherwise, all experimental environments were thoroughly cleaned with 70 and 30% ethanol between trials and allowed to dry. All behavioral procedures were video-recorded and scored by an individual blind to the genotype of the mouse.
Rotarod test.
Motor balance and coordination were analyzed using the rotarod test. Mice were trained for 3 d. All determinations were carried out at the same time each day and the test was performed during a four-trial testing session. Mice were allowed to stay for 300 s on a five-lane accelerating rotating rod (model 47600; Ugo Basile) with a continuous accelerating rotation speed from 4 to 40 rpm, increasing 4 rpm every 30 s and reaching a final speed at 270 s. The latency to fall off the rotarod was measured during this period, annotating the time the animal stayed on the rotating rod. Data from tests in which animals completed three turns without walking were disregarded (48).
Lashley III maze test.
Animals were tested in the Lashley III maze for analysis of learning and memory (54). This maze consists of a start box, four interconnected alleys, and a goal box. On 3 successive days, animals received a day of acclimation to adapt to the maze, followed by 2 training days. On the acclimation day, each mouse was confined in each of the first two alleys of the maze for 4 min, and in the final alley (close to the goal box) for 6 min. On the training day, each animal was placed in the start box and allowed to freely navigate the maze. To initiate testing, each animal was picked up by the tail and placed in the start box of the maze. Each animal’s home cage was placed at the end of the goal box, with the acrylic door raised so that the animal could enter it after traversing the maze. The trial was initiated by raising the start box door when an animal approached it. The trial was finished when the animal entered its home cage and then the door was lowered and the home cage was returned to the cart. Latency to enter the goal box and number of errors (a wrong turn or a retracing of the animal’s pathway was considered an error) were recorded. The frequencies of defecation were also recorded. Mice were given one trial per week during 4 consecutive weeks.
Radial arm maze test.
The eight-arm radial maze (Stoelting) was used to evaluate spatial learning and memory, including short-term working memory and long-term reference memory (18). During acclimation, training, and testing phases of the procedure, animals had 18 h of food deprivation to increase the saliency of food pellets located at the end of each bated arm. During acclimation (1 d), animals were allowed to explore the eight-arm radial maze with randomly placed food pellets throughout the maze. During the subsequent training day, animals were placed in the maze facing arm 1 and the four baited arms were located such that two of these arms were adjacent (1, 2, 4, and 7) and the other arms were 90° apart from these arms (3, 5, 6, and 8), which were closed. The same four of eight arms were baited, and the other four arms were never baited. The training session lasted until all food pellets had been retrieved or 5 min had elapsed. During testing, all arms of the radial maze were open. One testing session occurred each week for each animal and lasted until all food reward was retrieved or 5 min had elapsed. Latency to enter eight arms was recorded. Arm entry was counted when all four legs of an animal entered an arm. Entries into unbaited arms were counted as reference memory errors, and reentries into previously baited arms were counted as working memory errors. In contrast to spatial working (short-term) memory, spatial reference memory has more capacity and lasts longer (18). Mice received one trial per week for 4 consecutive weeks (four trials in total).
Open-field test.
Mice were assessed for novelty-induced locomotor activity/exploration in an open field (40 cm × 40 cm × 35 cm high) divided into two regions (64% border and 36% center) and placed in an ANY-box base (Stoelting) including a camera to track the animals and a height-adjustable, infrared photobeam array to detect rearing behavior. The animals were placed in the center of the apparatus and video-recorded for 5 min. The percentage of time spent in each region, distance moved, freezing (immobility), and number of rears (raising the forepaws, denoting vertical exploration) were scored in 5-min sessions (20).
Hole-board test.
Mice were assessed for exploratory behavior in a black box (40 cm × 40 cm × 35 cm high) with 16 holes (3-cm diameter) equally spaced in the floor and placed in the ANY-box base (Stoelting), which includes a camera to track the animals and an infrared photobeam array to detect automatically nose pokes through the holes. Mice were placed facing one corner of the apparatus and observed for 5 min. The exploratory behavior was measured as the number of head dips into holes. Distance moved was also recorded (21).
Statistical Analysis.
The results are expressed as means ± SEM. A one-way ANOVA followed by the Bonferroni post hoc test was used for pairwise comparisons within multiple samples. The Student’s t test was used to compare the means of two independent groups. In all cases, P values < 0.05 were considered significant. Statistical analysis was performed using SPSS Statistics 22.0 for Macintosh.
Acknowledgments
We thank Prof. Sergio Moreno (Institute of Functional Biology and Genomics, CSIC) for the generous supply of the Cdh1lox/lox-targeted mouse model. We also appreciate the technical assistance of Monica Resch and Monica Carabias. The confocal images of fixed samples were taken at the Bordeaux Imaging Center, which is a service unit of CNRS, INSERM, and the University of Bordeaux and a member of the national infrastructure France BioImaging. This work was funded by Instituto de Salud Carlos III Grants PI15/00473, BAE14/0005, RD12/0014/0007, and RD16/0019/0018 (to A.A.) and RD12/0014/0001 and RD16/0019/0001 (to J.C.); FEDER (European Regional Development Fund); Ministerio de Economía y Competitividad Grant SAF2016-78114-R (to J.P.B.); and European Union’s Horizon 2020 Research and Innovation Programme Grant 686009 (to A.A. and J.C.). V.B.-J. (PI15/00473) and M.D.-E. (CP0014/00010) are supported by the Instituto de Salud Carlos III. I.S.-M. is supported by the Junta de Castilla y León (Spain).
Supporting Information
Supporting Information (PDF)
- Download
- 1.70 MB
References
1
AJ Koleske, Molecular mechanisms of dendrite stability. Nat Rev Neurosci 14, 536–550 (2013).
2
VA Kulkarni, BL Firestein, The dendritic tree and brain disorders. Mol Cell Neurosci 50, 10–20 (2012).
3
JN Cochran, AM Hall, ED Roberson, The dendritic hypothesis for Alzheimer’s disease pathophysiology. Brain Res Bull 103, 18–28 (2014).
4
N Hensel, S Rademacher, P Claus, Chatting with the neighbors: Crosstalk between Rho-kinase (ROCK) and other signaling pathways for treatment of neurological disorders. Front Neurosci 9, 198 (2015).
5
M Amano, M Nakayama, K Kaibuchi, Rho-kinase/ROCK: A key regulator of the cytoskeleton and cell polarity. Cytoskeleton 67, 545–554 (2010).
6
Z Li, L Van Aelst, HT Cline, Rho GTPases regulate distinct aspects of dendritic arbor growth in Xenopus central neurons in vivo. Nat Neurosci 3, 217–225 (2000).
7
AY Nakayama, MB Harms, L Luo, Small GTPases Rac and Rho in the maintenance of dendritic spines and branches in hippocampal pyramidal neurons. J Neurosci 20, 5329–5338 (2000).
8
H Chen, BL Firestein, RhoA regulates dendrite branching in hippocampal neurons by decreasing cypin protein levels. J Neurosci 27, 8378–8386 (2007).
9
A Almeida, Regulation of APC/C-Cdh1 and its function in neuronal survival. Mol Neurobiol 46, 547–554 (2012).
10
J Huang, A Bonni, A decade of the anaphase-promoting complex in the nervous system. Genes Dev 30, 622–638 (2016).
11
A Almeida, JP Bolaños, S Moreno, Cdh1/Hct1-APC is essential for the survival of postmitotic neurons. J Neurosci 25, 8115–8121 (2005).
12
M Delgado-Esteban, I García-Higuera, C Maestre, S Moreno, A Almeida, APC/C-Cdh1 coordinates neurogenesis and cortical size during development. Nat Commun 4, 2879 (2013).
13
I García-Higuera, et al., Genomic stability and tumour suppression by the APC/C cofactor Cdh1. Nat Cell Biol 10, 802–811 (2008).
14
L Minichiello, et al., Essential role for TrkB receptors in hippocampus-mediated learning. Neuron 24, 401–414 (1999).
15
V Depaepe, et al., Ephrin signalling controls brain size by regulating apoptosis of neural progenitors. Nature 435, 1244–1250 (2005).
16
YN Jan, LY Jan, Branching out: Mechanisms of dendritic arborization. Nat Rev Neurosci 11, 316–328 (2010).
17
R Zhou, PJ Abbas, JG Assouline, Electrically evoked auditory brainstem response in peripherally myelin-deficient mice. Hear Res 88, 98–106 (1995).
18
S Sharma, S Rakoczy, H Brown-Borg, Assessment of spatial memory in mice. Life Sci 87, 521–536 (2010).
19
VR Porter, et al., Frequency and characteristics of anxiety among patients with Alzheimer’s disease and related dementias. J Neuropsychiatry Clin Neurosci 15, 180–186 (2003).
20
L Prut, C Belzung, The open field as a paradigm to measure the effects of drugs on anxiety-like behaviors: A review. Eur J Pharmacol 463, 3–33 (2003).
21
H Takeda, M Tsuji, T Matsumiya, Changes in head-dipping behavior in the hole-board test reflect the anxiogenic and/or anxiolytic state in mice. Eur J Pharmacol 350, 21–29 (1998).
22
H Murakoshi, H Wang, R Yasuda, Local, persistent activation of Rho GTPases during plasticity of single dendritic spines. Nature 472, 100–104 (2011).
23
T Tanaka, et al., Nuclear Rho kinase, ROCK2, targets p300 acetyltransferase. J Biol Chem 281, 15320–15329 (2006).
24
Z Ma, et al., Interaction between ROCK II and nucleophosmin/B23 in the regulation of centrosome duplication. Mol Cell Biol 26, 9016–9034 (2006).
25
S Tugendreich, J Tomkiel, W Earnshaw, P Hieter, CDC27Hs colocalizes with CDC16Hs to the centrosome and mitotic spindle and is essential for the metaphase to anaphase transition. Cell 81, 261–268 (1995).
26
F Meghini, et al., Targeting of Fzr/Cdh1 for timely activation of the APC/C at the centrosome during mitotic exit. Nat Commun 7, 12607 (2016).
27
C Maestre, M Delgado-Esteban, JC Gomez-Sanchez, JP Bolaños, A Almeida, Cdk5 phosphorylates Cdh1 and modulates cyclin B1 stability in excitotoxicity. EMBO J 27, 2736–2745 (2008).
28
M Shibuya, S Hirai, M Seto, S Satoh, E Ohtomo, Effects of fasudil in acute ischemic stroke: Results of a prospective placebo-controlled double-blind trial. J Neurol Sci; Fasudil Ischemic Stroke Study Group 238, 31–39 (2005).
29
G Feng, et al., Imaging neuronal subsets in transgenic mice expressing multiple spectral variants of GFP. Neuron 28, 41–51 (2000).
30
P De Camilli, PE Miller, F Navone, WE Theurkauf, RB Vallee, Distribution of microtubule-associated protein 2 in the nervous system of the rat studied by immunofluorescence. Neuroscience 11, 817–846 (1984).
31
G Huber, A Matus, Differences in the cellular distributions of two microtubule-associated proteins, MAP1 and MAP2, in rat brain. J Neurosci 4, 151–160 (1984).
32
M Amano, et al., Identification of Tau and MAP2 as novel substrates of Rho-kinase and myosin phosphatase. J Neurochem 87, 780–790 (2003).
33
H Yamamoto, K Fukunaga, E Tanaka, E Miyamoto, Ca2+- and calmodulin-dependent phosphorylation of microtubule-associated protein 2 and tau factor, and inhibition of microtubule assembly. J Neurochem 41, 1119–1125 (1983).
34
P Hotulainen, CC Hoogenraad, Actin in dendritic spines: Connecting dynamics to function. J Cell Biol 189, 619–629 (2010).
35
C Cerri, et al., Activation of Rho GTPases triggers structural remodeling and functional plasticity in the adult rat visual cortex. J Neurosci 31, 15163–15172 (2011).
36
A Martino, et al., Rho GTPase-dependent plasticity of dendritic spines in the adult brain. Front Cell Neurosci 7, 62 (2013).
37
M Li, et al., The adaptor protein of the anaphase promoting complex Cdh1 is essential in maintaining replicative lifespan and in learning and memory. Nat Cell Biol 10, 1083–1089 (2008).
38
AK Fu, et al., APC(Cdh1) mediates EphA4-dependent downregulation of AMPA receptors in homeostatic plasticity. Nat Neurosci 14, 181–189 (2011).
39
J Huang, Y Ikeuchi, M Malumbres, A Bonni, A Cdh1-APC/FMRP ubiquitin signaling link drives mGluR-dependent synaptic plasticity in the mammalian brain. Neuron 86, 726–739 (2015).
40
PA Adlard, JC Vickers, Morphologically distinct plaque types differentially affect dendritic structure and organisation in the early and late stages of Alzheimer’s disease. Acta Neuropathol 103, 377–383 (2002).
41
C Conde, A Cáceres, Microtubule assembly, organization and dynamics in axons and dendrites. Nat Rev Neurosci 10, 319–332 (2009).
42
SS Jiao, et al., Edaravone alleviates Alzheimer’s disease-type pathologies and cognitive deficits. Proc Natl Acad Sci USA 112, 5225–5230 (2015).
43
JH Herskowitz, et al., Pharmacologic inhibition of ROCK2 suppresses amyloid-β production in an Alzheimer’s disease mouse model. J Neurosci 33, 19086–19098 (2013).
44
Y Zhou, et al., Nonsteroidal anti-inflammatory drugs can lower amyloidogenic Abeta42 by inhibiting Rho. Science 302, 1215–1217 (2003).
45
Y Song, X Chen, LY Wang, W Gao, MJ Zhu, Rho kinase inhibitor fasudil protects against β-amyloid-induced hippocampal neurodegeneration in rats. CNS Neurosci Ther 19, 603–610 (2013).
46
N Rath, MF Olson, Rho-associated kinases in tumorigenesis: Re-considering ROCK inhibition for cancer therapy. EMBO Rep 13, 900–908 (2012).
47
AJ Ridley, Rho GTPase signalling in cell migration. Curr Opin Cell Biol 36, 103–112 (2015).
48
C Rodríguez, et al., Neovascularization and functional recovery after intracerebral hemorrhage is conditioned by the Tp53 Arg72Pro single-nucleotide polymorphism. Cell Death Differ 24, 144–154 (2017).
49
G Paxinos, C Watson The Rat Brain in Stereotaxic Coordinates (Academic/Elsevier, 6th Ed, Boston, 2007).
50
C Schmitz, PR Hof, Design-based stereology in neuroscience. Neuroscience 130, 813–831 (2005).
51
S Herculano-Houzel, R Lent, Isotropic fractionator: A simple, rapid method for the quantification of total cell and neuron numbers in the brain. J Neurosci 25, 2518–2521 (2005).
52
YH Park, G Wood, DL Kastner, JJ Chae, Pyrin inflammasome activation and RhoA signaling in the autoinflammatory diseases FMF and HIDS. Nat Immunol 17, 914–921 (2016).
53
D Komnig, JB Schulz, A Reich, BH Falkenburger, Mice lacking Faim2 show increased cell death in the MPTP mouse model of Parkinson disease. J Neurochem 139, 848–857 (2016).
54
LD Matzel, et al., Individual differences in the expression of a “general” learning ability in mice. J Neurosci 23, 6423–6433 (2003).
Information & Authors
Information
Published in
Cover image: As the 21st president of the US National Academy of Sciences, Ralph J. Cicerone served as a steadfast champion of science. During his tenure from July 1, 2005, to June 30, 2016, Cicerone conducted informed and balanced discussions with politicians and policymakers concerning many complex and controversial scientific issues facing the nation. At the same time, he remained a strong advocate for providing independent scientific advice to the government and the public—a goal central to the Academy's mission since its founding in 1863. See the Retrospective on Cicerone's life and legacy by Veerabhadran Ramanathan on pages 4273–4274. Photograph by Mark Finkenstaedt and image courtesy of the National Academy of Sciences.
Classifications
Copyright
Freely available online through the PNAS open access option.
Submission history
Published online: April 10, 2017
Published in issue: April 25, 2017
Keywords
Acknowledgments
We thank Prof. Sergio Moreno (Institute of Functional Biology and Genomics, CSIC) for the generous supply of the Cdh1lox/lox-targeted mouse model. We also appreciate the technical assistance of Monica Resch and Monica Carabias. The confocal images of fixed samples were taken at the Bordeaux Imaging Center, which is a service unit of CNRS, INSERM, and the University of Bordeaux and a member of the national infrastructure France BioImaging. This work was funded by Instituto de Salud Carlos III Grants PI15/00473, BAE14/0005, RD12/0014/0007, and RD16/0019/0018 (to A.A.) and RD12/0014/0001 and RD16/0019/0001 (to J.C.); FEDER (European Regional Development Fund); Ministerio de Economía y Competitividad Grant SAF2016-78114-R (to J.P.B.); and European Union’s Horizon 2020 Research and Innovation Programme Grant 686009 (to A.A. and J.C.). V.B.-J. (PI15/00473) and M.D.-E. (CP0014/00010) are supported by the Instituto de Salud Carlos III. I.S.-M. is supported by the Junta de Castilla y León (Spain).
Notes
This article is a PNAS Direct Submission.
Authors
Competing Interests
The authors declare no conflict of interest.
Metrics & Citations
Metrics
Citation statements
Altmetrics
Citations
Cite this article
114 (17) 4513-4518,
Export the article citation data by selecting a format from the list below and clicking Export.
Cited by
Loading...
View Options
View options
PDF format
Download this article as a PDF file
DOWNLOAD PDFLogin options
Check if you have access through your login credentials or your institution to get full access on this article.
Personal login Institutional LoginRecommend to a librarian
Recommend PNAS to a LibrarianPurchase options
Purchase this article to access the full text.