Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Cell-autonomous reduction of CYFIP2 is insufficient to induce Alzheimer's disease-like pathologies in the hippocampal CA1 pyramidal neurons of aged mice

Cell-autonomous reduction of CYFIP2 is insufficient to induce Alzheimer's disease-like... ANIMAL CELLS AND SYSTEMS 2023, VOL. 27, NO. 1, 93–101 https://doi.org/10.1080/19768354.2023.2192263 Cell-autonomous reduction of CYFIP2 is insufficient to induce Alzheimer’s disease-like pathologies in the hippocampal CA1 pyramidal neurons of aged mice a,b a a,b a,b a a,b Ruiying Ma , Yinhua Zhang , Huiling Li , Hyae Rim Kang , Yoonhee Kim and Kihoon Han a b Department of Neuroscience, Korea University College of Medicine, Seoul, Republic of Korea; BK21 Graduate Program, Department of Biomedical Sciences, Korea University College of Medicine, Seoul, Republic of Korea ABSTRACT ARTICLE HISTORY Received 9 February 2023 Cytoplasmic FMR1-interacting protein 2 (CYFIP2) is an evolutionarily conserved multifunctional Revised 7 March 2023 protein that regulates the neuronal actin cytoskeleton, mRNA translation and transport, and Accepted 8 March 2023 mitochondrial morphology and function. Supporting its critical roles in proper neuronal development and function, human genetic studies have repeatedly identified variants of the KEYWORDS CYFIP2 gene in individuals diagnosed with neurodevelopmental disorders. Notably, a few CYFIP2; Alzheimer’s disease; recent studies have also suggested a mechanistic link between reduced CYFIP2 level and hippocampal CA1; excitatory Alzheimer’s disease (AD). Specifically, in the hippocampus of 12-month-old Cyfip2 pyramidal neuron; heterozygous mice, several AD-like pathologies were identified, including increased levels of conditional knock-out Tau phosphorylation and gliosis, and loss of dendritic spines in CA1 pyramidal neurons. However, detailed pathogenic mechanisms, such as cell types and their circuits where the pathologies originate, remain unknown for AD-like pathologies caused by CYFIP2 reduction. In this study, we aimed to address this issue by examining whether the cell-autonomous reduction of CYFIP2 in CA1 excitatory pyramidal neurons is sufficient to induce AD-like phenotypes in the hippocampus. We performed immunohistochemical, morphological, and biochemical analyses in 12-month-old Cyfip2 conditional knock-out mice, which have postnatally reduced CYFIP2 expression level in CA1, but not in CA3, excitatory pyramidal neurons of the hippocampus. Unexpectedly, we could not find any significant AD-like phenotype, suggesting that the CA1 excitatory neuron-specific reduction of CYFIP2 level is insufficient to lead to AD-like pathologies in the hippocampus. Therefore, we propose that CYFIP2 reduction in other neurons and/or their synaptic connections with CA1 pyramidal neurons may be critically involved in the hippocampal AD-like phenotypes of Cyfip2 heterozygous mice. Introduction 2019; Zweier et al. 2019; Begemann et al. 2021;Kang et al. 2023). The two members of the cytoplasmic FMR1-interacting At the molecular level, CYFIP1 and CYFIP2 have a high protein family, CYFIP1 and CYFIP2, are evolutionarily amino acid sequence homology and both are involved conserved proteins whose genetic variants are causally in the regulation of cellular actin cytoskeleton dynamics associated with numerous brain disorders, including as a critical component of the heteropentameric autism spectrum disorders, intellectual disability, Wiskott–Aldrich syndrome protein family verprolin- schizophrenia, and epilepsy (Schenck et al. 2001; homologous protein (WAVE) regulatory complex (WRC) Abekhoukh and Bardoni 2014; Zhang, Lee, et al. (Lee Y et al. 2017; Rottner et al. 2021). Additional func- 2019). Specifically, in the case of CYFIP2, de novo var- tions in neurons, such as regulation of mRNA translation iants have recently been identified in individuals and transport (Napoli et al. 2008; De Rubeis et al. 2013; diagnosed with neurodevelopmental disorders and Cioni et al. 2018), and regulation of mitochondrial func- early-onset epileptic encephalopathy characterized by tion and morphology (Kanellopoulos et al. 2020; Kim GH developmental regression, intellectual disability, seizures, et al. 2020) have also been identified, but unlike actin muscular hypotonia, and microcephaly (Nakashima et al. regulation via the WRC, these functions may be 2018;Peng etal. 2018; Lee et al. 2019; Zhong et al. CONTACT Kihoon Han neurohan@korea.ac.kr Supplemental data for this article can be accessed online at https://doi.org/10.1080/19768354.2023.2192263. © 2023 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. The terms on which this article has been published allow the posting of the Accepted Manuscript in a repository by the author(s) or with their consent. 94 R. MA ET AL. mediated by both shared and distinct interactors of pathologies in aged mice remains largely unknown. CYFIP1 and CYFIP2, respectively (Lee Y et al. 2020;Ma In particular, considering the complex interactions et al. 2022). Collectively, these molecular functions of between different neuronal and non-neuronal cell CYFIP1 and CYFIP2 can contribute to the morphological types and the trans-synaptic spread of pathology in and functional changes in neuronal synapses observed AD (Tzioras et al. 2023), investigating the cell types in both Cyfip1 and Cyfip2 mutant mice (Bozdagi et al. and their circuits where CYFIP2-dependent pathologies 2012; Pathania et al. 2014; Han et al. 2015; Davenport originate is a critical step toward understanding the et al. 2019; Lee SH et al. 2020; Zhang et al. 2020; Kim pathogenic mechanisms. In this study, we aimed to NS et al. 2022). address this issue by examining the AD-like pheno- Moreover, several lines of evidence indicate the differ- types of aged (12-month-old) Cyfip2 conditional ential roles of CYFIP1 and CYFIP2 in vivo, including the knock-out (cKO) mice, which have postnatally reduced embryonic and perinatal lethality of Cyfip1-null mice CYFIP2 expression level selectively in CA1, but not in and Cyfip2-null mice, respectively (Chung et al. 2015; CA3, excitatory pyramidal neurons of the hippocampus. Han et al. 2015; Zhang et al. 2019a), and different brain Surprisingly, unlike Cyfip2 het mice, there was no overt regional and cell-type distributions of CYFIP1 and AD-like phenotype in the hippocampal CA1 region of CYFIP2 mRNAs and proteins (Zhang, Kang, et al. 2019b; aged Cyfip2 cKO mice. Therefore, our results suggest Ma et al. 2022). In particular, CYFIP1, but not CYFIP2, is that cell-autonomous reduction of CYFIP2 is insufficient also expressed in non-neuronal cells, such as astrocytes, for AD-like pathologies in CA1 pyramidal neurons and microglia, and oligodendrocytes, in the brain (Domin- that other neurons and/or their synaptic connections guez-Iturza et al. 2019; Silva et al. 2019; Habela et al. with CA1 pyramidal neurons are also critically involved 2020; Haan et al. 2021; Ma et al. 2022). in the hippocampal AD-like phenotypes of Cyfip2 het Beyond the genetic association between CYFIP2 and mice. neurodevelopmental disorders, recent studies have suggested a mechanistic association between the Materials and methods reduction of CYFIP2 and Alzheimer’s disease (AD) (Tiwari et al. 2016; Ghosh et al. 2020). Specifically, Mice CYFIP2 protein levels were reduced in the postmortem The Cyfip2 cKO mice and Thy1-YFP mice used in this forebrain of patients with AD and in the hippocampus study were previously described (Lee SH et al. 2020; and cortex of AD model mice (Tiwari et al. 2016). More- Zhang et al. 2020). The mice were fed ad libitum and over, the protein levels of amyloid precursor protein housed under a 12 h light–dark cycle. All experiments (APP), β-site APP cleaving enzyme 1 (BACE1), and were performed using aged (12-month-old) male calcium/calmodulin-dependent protein kinase IIα Cyfip2 cKO mice and their littermate controls. (CaMKIIα) were up-regulated in the hippocampal synap- tosomal fraction of conventional Cyfip2 heterozygous +/– (Cyfip2 , het) mice. Consistently, in the hippocampus Fluorescence immunohistochemistry of aged (12-month-old) Cyfip2 het mice, several AD-like pathologies have been observed, including increased Fluorescence immunohistochemistry was performed as levels of Tau phosphorylation and gliosis, and significant previously described (Lee B et al. 2017; Yu et al. 2021). loss of dendritic spines in CA1 pyramidal neurons (Ghosh Mice were anesthetized with isoflurane and transcardially et al. 2020). Mechanistically, it has been proposed that perfused with heparinized (20 units/mL) phosphate- reduced expression level of CYFIP2 induces aberrant buffered saline (PBS), followed by 4% paraformaldehyde local mRNA translation of several AD-related proteins (PFA) in PBS. The brains were extracted and post-fixed (i.e. APP, BACE1, and CaMKIIα) at the synaptic compart- overnight in 4% PFA. Following post-fixation, brain ment, thereby leading to the overproduction of Aβ and tissue was washed with PBS and cryoprotected with hyperphosphorylation of Tau in Cyfip2 het mice (Ghosh 30% sucrose in PBS for 48 h. The brain tissues were et al. 2020). Under normal conditions, CYFIP2, as shown frozen in an O.C.T compound (SAKURA Tissue-Tek, for CYFIP1 (Napoli et al. 2008; De Rubeis et al. 2013), 4583) and sectioned (60 µm) using a cryostat microtome may repress the translation of these mRNAs by forming (Leica, CM3050S). The primary antibodies used for immu- an inhibitory complex with the RNA-binding protein nohistochemistry were AT-8 (Phospho-Tau [Ser202, fragile X messenger ribonucleoprotein (FMRP) and the Thr205], Invitrogen, #MN1020), CYFIP1 (Sigma-Aldrich, eukaryotic initiation factor 4E (eIF4E). #AB6046), CYFIP2 (Abcam, #ab95969), GFAP (Abcam, However, compared to CYFIP2 function and dysfunc- #ab4674), Iba1 (Synaptic System, #234-006), NeuN tion in the developing brain, its role in AD-like (Abcam, #ab177487; Millipore, #MAB377). F-actin was ANIMAL CELLS AND SYSTEMS 95 visualized by Alexa Fluor 488-conjugated Phalloidin (Invi- Results trogen, #A-12379). The samples were washed with 0.1% CA1 excitatory pyramidal neuron-specific Triton X-100 in PBS and blocked with PBS containing reduction of CYFIP2 in the hippocampus of aged 3% bovine serum albumin (BSA) and 0.5% Triton X-100. Cyfip2 cKO mice The high-resolution image acquisition was performed using a Zeiss LSM800 confocal microscope equipped To investigate whether cell-autonomous CYFIP2 with a 20×/0.8 objective lens, 8-bit image depth, and reduction in excitatory pyramidal neurons is sufficient snapshot mode focused on maximum intensity. Whole to induce AD-like phenotypes in the hippocampal CA1 brain regions were obtained using a slide scanner region, we crossed floxed-Cyfip2 mice with CaMKIIα-Cre floxed/floxed (Zeiss Axio Scan.Z1). The regions of the stratum oriens mice to generate Cyfip2 cKO (Cyfip2 ; CaMKIIα- (SO) and stratum radiatum (SR) were defined as 100 Cre) mice as previously described (Zhang et al. 2020). µm and 100–200 µm away from the cell body area The CaMKIIα-Cre line (T29-1) used in this study starts (stratum pyramidale, SP) of CA1, respectively. The expressing Cre recombinases in forebrain excitatory values of at least two brain sections were averaged for neurons during the third to fourth postnatal weeks, each mouse. and especially in the hippocampus, Cre expression is restricted mainly to the CA1 region (Tsien et al. 1996). floxed/floxed Notably, to securely obtain control (Cyfip2 ) Dendritic spine analysis and Cyfip2 cKO progeny mice, we crossed male floxed/floxed floxed/floxed The dendritic spine analysis was performed as pre- Cyfip2 mice with female Cyfip2 ; viously described (Han et al. 2013;Choietal. 2015; CaMKIIα-Cre mice (Figure 1(A)) to avoid unwanted germ- Hong et al. 2022). Mice were deeply anesthetized line recombination, which was recently reported in male, with isoflurane and transcardially perfused with but not female, T29-1 mice (Luo et al. 2020). Further- heparinized (20 units/mL) PBS followed by 4% PFA more, we designed an additional primer set for in PBS. The brains were extracted and post-fixed genotyping PCR to detect germline deletion of floxed overnight in 4% PFA. After post-fixation, coronal sec- exon 6 of the Cyfip2 gene (Lee SH et al. 2020), which tions (100 µm thickness) of the hippocampal region indeed produced an expected PCR band from some floxed/Δexon6 were obtained using a vibratome (VT1000S, Leica). portions of progeny mice (i.e. Cyfip2 mice) floxed/floxed The sections were collected and stored in 50% gly- when we crossed male Cyfip2 ;CaMKIIα-Cre floxed/floxed cerol in 2 × PBS at −20 C until further processed. mice with female Cyfip2 mice as a test Blocking, permeabilization, and anti-GFP (Abcam, (Figure 1(B) and (C)). Using this primer set, we #AB13970) primary and Alexa Fluorconjugated (anti- confirmed that the control and Cyfip2 cKO mice used chicken Alexa Fluor-488, Jackson ImmunoResearch in this study did not have a germline deletion of the Labs, #703-545-155) secondary antibody incubation floxed Cyfip2 exon 6. were performed as described above. Finally, the sec- Fluorescence immunohistochemical analysis showed tions were mounted on slide glasses with mounting that CYFIP2 protein levels were selectively reduced in media (Biomeda, M02). Images of dendritic spines the hippocampal CA1, but not CA3, of 12-month-old in the secondary or tertiary branches (apical or Cyfip2 cKO mice (Figure 1(D)), which was expected basal dendrites of YFP-positive CA1 pyramidal from CA1-restricted Cre expression in T29-1 mice. Mean- neurons in the hippocampus) were acquired by con- while, CYFIP1 protein levels in the CA1 region were com- focal microscopy (Zeiss LSM800) using 63×/1.2 water parable between control and Cyfip2 cKO mice, immersion objective lens, 8-bit image depth, and Z- suggesting that there is no compensatory increase in stack function with 0.93 µm intervals, followed by CYFIP1 level in the hippocampus of aged Cyfip2 cKO Z-stack projection of maximum intensity. Images mice (Figure 1(E)). were analyzed using ImageJ software. For quantifi- cation of dendritic spines, mushroom spines were defined as protrusions with heads and with a width No AD-like immunohistological phenotype in the greater than length. Stubby spines were defined as hippocampal CA1 region of aged Cyfip2 cKO mice protrusions without a neck. The rest of the protru- sions with heads were categorized as thin spines. A previous study showed that several AD-like immu- The values of six to eight neurons were averaged nohistological phenotypes were significantly exacer- for each mouse. bated in the hippocampal CA1 region of 12-month- Additional information of Materials and Methods is old Cyfip2 het mice compared with age-matched included in supporting online material. wild-type (WT) mice (Ghosh et al. 2020). These 96 R. MA ET AL. Figure 1. No AD-like immunohistological phenotype in the hippocampal CA1 region of aged Cyfip2 cKO mice. (A) The breeding f/f f/f scheme for the control (Cyfip2 ) and Cyfip2 conditional knock-out (Cyfip2 ;CaMKIIα-Cre, cKO) mice. (B) The breeding scheme to test the partial germline recombination of male CaMKIIα-Cre mice. (C) Design of primers to detect floxed and Δexon6 Cyfip2 f/f floxed/Δexon6 alleles (left panel). Results of PCR for the tail genomic DNA isolated from Cyfip2 and Cyfip2 mice (right panel). Note f/f that the primer set (a + c) does not produce the expected ∼1.1 kbp band from the DNA sample of Cyfip2 mice due to the short elongation time of the PCR. (D) Fluorescence immunohistochemistry images and quantification showing CA1-specific reduction of CYFIP2 in the hippocampus of aged Cyfip2 cKO mice. CA, cornu ammonis; DG, dentate gyrus; NS, not significant; SO, stratum oriens; SP, stratum pyramidale; SR, stratum radiatum. (E, F) Normal CYFIP1 and phospho-Tau (AT-8) levels in the hippocampal CA1 region of aged Cyfip2 cKO mice. (G, H) Normal density and total intensity of astrocytes (GFAP-positive) and microglia (Iba1-positive) in the hippocampal CA1 region of aged Cyfip2 cKO mice. N =7–8 mice. phenotypes include increased levels of phospho-Tau [GFAP] and ionized calcium-binding adapter molecule immunoreactivity (measured by monoclonal AT-8 1 [Iba1] antibodies, respectively). Therefore, we antibody) and gliosis (both for astrocytes and micro- performed fluorescence immunohistochemical ana- glia, as measured by glial fibrillary acidic protein lyses of these AD markers in the hippocampal CA1 ANIMAL CELLS AND SYSTEMS 97 region of 12-month-old Cyfip2 cKO mice and their littermate controls. However, there were no significant differences in phospho-Tau levels between control and Cyfip2 cKO mice (Figure 1(F)). Moreover, neither GFAP nor Iba1 positive cell number or total intensity was significantly altered in Cyfip2 cKO mice compared to control mice (Figure 1 (G) and (H)). Reduced number of immature dendritic spines and increased F-actin levels in the basal dendrites of CA1 pyramidal neurons of aged Cyfip2 cKO mice Dendritic spines are small dendritic protrusions that rep- resent the most excitatory postsynapses in the brain (Penzes et al. 2011). Loss of dendritic spines is another key feature of AD (Dorostkar et al. 2015) and was signifi- cantly aggravated in CA1 pyramidal neurons of 12- month-old Cyfip2 het mice compared to age-matched WT mice (Ghosh et al. 2020). In particular, the number of mature-type mushroom spines, but not that of imma- ture-type thin spines, was reduced in the apical den- drites of CA1 pyramidal neurons in aged Cyfip2 het mice. Therefore, we analyzed the dendritic spines of CA1 pyramidal neurons in 12-month-old control and Cyfip2 cKO mice. Dendritic spines were visualized by crossing Cyfip2 cKO mice with Thy1-YFP mice (Feng et al. 2000) that sparsely express yellow fluorescent protein (YFP) in CA1 pyramidal neurons (Figure 2(A)). We separately analyzed the basal and apical dendrites of the neurons (Figure 2(B)). In the basal dendrites, we found that the total number of dendritic spines was sig- nificantly reduced in Cyfip2 cKO neurons compared to control neurons (Figure 2(C)). However, unlike the mush- room spine-specific reduction in Cyfip2 het mice (Ghosh et al. 2020), the spine reduction in Cyfip2 cKO mice was mainly attributed to a decrease in immature-type thin spines. Moreover, in the apical dendrites, neither total Figure 2. Dendritic spine and F-actin changes in the hippocam- density nor morphologically-based categorization of pal CA1 region of aged Cyfip2 cKO mice. (A) Visualization of CA1 dendritic spines was altered in Cyfip2 cKO neurons com- pyramidal neurons in Thy1-YFP mice by sparse expression of yellow fluorescent protein (YFP). CA, cornu ammonis; DG, pared to control neurons (Figure 2(C)), suggesting that dentate gyrus; SO, stratum oriens; SP, stratum pyramidale; SR, there is no AD-like dendritic spine phenotype in CA1 pyr- stratum radiatum. (B) Representative confocal images of dendri- amidal neurons of aged Cyfip2 cKO mice. We also com- tic spines in the basal and apical dendrites of CA1 pyramidal pared head size for both thin and mushroom spines neurons of aged control and Cyfip2 cKO mice. Examples of den- between the control and Cyfip2 cKO neurons and dritic spines in each morphologically-based categorization (thin, found no significant differences in the basal or apical stubby, and mushroom) are indicated by arrows with different colors. (C) Quantification of dendritic spine number in the dendrites (Figure 2(D)). basal (upper panel) and apical (lower panel) dendrites. NS, not F-actin is a key cytoskeletal component of dendritic significant. (D) Quantification of dendritic spine head size in spines and is directly associated with their formation, basal (upper panel) and apical (lower panel) dendrites. (E) Repre- maintenance, and dynamics (Spence and Soderling sentative confocal images and quantification of F-actin levels in 2015). As a critical component of the WRC, CYFIP2 is the hippocampal CA1 region of aged control and Cyfip2 cKO mice. N =4–8 mice. involved in actin regulation in various cellular 98 R. MA ET AL. compartments, including neuronal dendritic spines were comparable between control and Cyfip2 cKO (Rottner et al. 2021). Specifically, it has been previously mice. Notably, we found that neither APP nor CaMKIIα shown increased F-actin levels in the medial prefrontal protein levels were significantly altered in Cyfip2 cKO cortex (mPFC) of young adult Cyfip2 het and Cyfip2 mice compared to control mice (Figure 3(B)). Further- cKO mice (Lee SH et al. 2020; Zhang et al. 2020). There- more, the levels of PSD-95, another synaptic protein fore, we also measured F-actin levels in the hippocampal whose mRNA stability and translation are regulated by CA1 region and found an increase in the stratum oriens FMRP (Zalfa et al. 2007), and FMRP itself were normal (SO), but not in the stratum radiatum (SR), of aged Cyfip2 in the CA1 synaptosome of aged Cyfip2 cKO mice. cKO mice compared to control mice (Figure 2(E)). Taken together, these results suggest that there is no overt AD-like pathology in the hippocampal CA1 region of 12-month-old Cyfip2 cKO mice. Normal expression of synaptosomal APP and CaMKIIα in the hippocampal CA1 region of aged Cyfip2 cKO mice Discussion Overexpression of AD-related proteins, such as APP and In this study, we combined immunohistochemical, mor- CaMKIIα, in the synaptic compartment due to aberrant phological, and biochemical approaches to understand local mRNA translation, has been proposed as a molecu- whether the cell-autonomous reduction of CYFIP2 in lar mechanism underlying AD-like pathologies in Cyfip2 excitatory pyramidal neurons is sufficient to induce het mice (Tiwari et al. 2016; Ghosh et al. 2020). Therefore, AD-like pathologies in the hippocampal CA1 region. we analyzed the protein levels of APP and CaMKIIα in the However, none of the results showed a significant AD- hippocampus of 12-month-old control and Cyfip2 cKO like phenotype in aged Cyfip2 cKO mice, in contrast to mice. We prepared a crude synaptosomal fraction from the severe phenotypes observed in Cyfip2 het mice the dissected hippocampal CA1 region and performed (Tiwari et al. 2016; Ghosh et al. 2020). Therefore, our immunoblotting (Figure 3(A)). Consistent with the results suggest that other neurons and/or their synaptic immunohistological analysis, CYFIP2 levels were connections with CA1 pyramidal neurons are also criti- reduced in hippocampal CA1 synaptosomal lysates cally involved in the hippocampal AD-like phenotypes from Cyfip2 cKO mice compared to control mice of Cyfip2 het mice. Additional genetic or viral tools to (Figure 3(B)). As expected, the WAVE1 protein, another reduce CYFIP2 protein levels in specific or combinatorial component of the WRC, was also reduced in Cyfip2 neurons of the hippocampal circuit will help us further cKO lysates because the stability of WAVE1 is inter- address this issue. dependent with that of CYFIP2 (Han et al. 2015; Zhang Based on previous findings, we speculate on some et al. 2020; Kang et al. 2023). In contrast, CYFIP1 levels mechanisms that explain the lack of an AD-like Figure 3. Normal synaptosomal expression levels of AD-related proteins, APP and CaMKIIα, in the hippocampal CA1 region of aged Cyfip2 cKO mice. (A) Schematic diagram showing the dissection of the CA1 region of the mouse hippocampus. (B) Representative immunoblot images and quantification of the expression levels of CYFIP2, WAVE1, CYFIP1, APP, CaMKIIα, PSD-95, and FMRP proteins in CA1 synaptosomal fraction of aged Cyfip2 cKO mice compare to control mice. Protein levels were normalized by either a neuron- specific protein, neuron-specific enolase (NSE), or GAPDH. NS, not significant. N = 5 mice. ANIMAL CELLS AND SYSTEMS 99 Figure 4. Schematic diagrams summarizing the different hippocampal phenotypes among aged wild-type (WT), Cyfip2 het, and Cyfip2 cKO mice. CYFIP2 protein levels are reduced in both CA1 and CA3 excitatory and inhibitory neurons in Cyfip2 het mice, but they are only reduced in CA1 excitatory neurons in Cyfip2 cKO mice. In the hippocampal CA1 region, AD-like pathologies, such as gliosis and dendritic spine loss, are observed in Cyfip2 het mice. In contrast, no AD-like phenotype is observed in Cyfip2 cKO mice. Ex., excitatory; Inh., inhibitory. phenotype in Cyfip2 cKO mice (Figure 4). CYFIP2 mRNAs synaptic functions in neurons of Cyfip2 cKO mice could and proteins are predominantly expressed in neurons preserve local mRNA translation within the normal compared to non-neuronal cells in the brain and are range. However, in Cyfip2 het mice, more profound detected in both excitatory and local inhibitory changes in synaptic activity, due to both presynaptic neurons (Zhang, Kang, et al. 2019b; Lee SH et al. 2020; and postsynaptic reduction of CYFIP2, and changes in Ma et al. 2022). Moreover, CYFIP2 is expressed in the CYFIP2-FMRP-eIF4E complex may congruently lead neurons of the hippocampal CA3 region as well as of to aberrant mRNA translation and overproduction of other brain regions that can directly form synaptic con- AD-related proteins. Further investigations of the mol- nections with CA1 excitatory pyramidal neurons (Han ecular composition and function of the CYFIP2-FMRP- et al. 2015; Lee SH et al. 2020). As CYFIP2 regulates eIF4E complex in Cyfip2 het and Cyfip2 cKO neurons axonal and presynaptic development and function are needed to test this hypothesis. (Cioni et al. 2018; Kim GH et al. 2020), it is conceivable Considering the aforementioned speculations regard- that both presynaptic and postsynaptic compartments ing the mechanisms involved, another plausible expla- are functionally affected in CA1 pyramidal neurons of nation for the observed phenotypic difference Cyfip2 het mice, thereby ultimately leading to synaptic between Cyfip2 het and Cyfip2 cKO CA1 pyramidal loss. Meanwhile, the postsynapse-specific CYFIP2 neurons could be the delayed onset of disease in reduction in CA1 pyramidal neurons of Cyfip2 cKO Cyfip2 cKO mice compared to Cyfip2 het mice. Due to mice may be insufficient to induce such changes. Furthe- the preservation of normal presynaptic inputs in CA1 more, considering the concept of trans-synaptic propa- pyramidal neurons of Cyfip2 cKO mice, it is possible gation of pathology in AD (Tzioras et al. 2023), bi- that the neurons will take a longer time to exhibit AD- directional spread of AD pathology through synaptic like phenotypes compared to the neurons of Cyfip2 het connections with other abnormal neurons may synergis- mice. This hypothesis can be examined by investigating tically worsen the phenotypes of CA1 pyramidal neurons the hippocampus of older (e.g. 18-month-old) Cyfip2 of Cyfip2 het mice compared to those of Cyfip2 cKO mice. cKO mice and age-matched control mice. At the molecular level, local mRNA translation of AD- We observed a reduced number of thin, but normal related proteins by the CYFIP2-FMRP-eIF4E complex can stubby and mushroom spines in the basal dendrites of be differentially affected in the CA1 pyramidal neurons CA1 pyramidal neurons of aged Cyfip2 cKO mice. of Cyfip2 het and Cyfip2 cKO mice (Ghosh et al. 2020). However, there was no significant change in the Indeed, we observed normal synaptosomal levels of number of thin, stubby, and mushroom spines in the APP and CaMKIIα proteins in the hippocampal CA1 of apical dendrites. The basal dendrite-specific decrease Cyfip2 cKO mice, unlike their increased levels in Cyfip2 in dendritic spine number was also observed in the het mice (Tiwari et al. 2016). The interaction of CYFIP1 layer 5 neurons of the mPFC of young adult Cyfip2 cKO with FMRP and eIF4E is regulated by synaptic activity mice (Zhang et al. 2020), although the underlying mech- (De Rubeis et al. 2013). Therefore, it can be speculated anism remains unknown. Additionally, a previous study that even with CYFIP2 reduction, relatively normal found that overexpression of CYFIP2 increases excitatory 100 R. MA ET AL. Davenport EC, Szulc BR, Drew J, Taylor J, Morgan T, Higgs synapse number in cultured hippocampal neurons at 14 NF, Lopez-Domenech G, Kittler JT. 2019. Autism and days in vitro (Davenport et al. 2019), suggesting that schizophrenia-associated CYFIP1 regulates the balance of CYFIP2 dosage may be an important factor regulating synaptic excitation and inhibition. Cell Rep. 26(8):2037– the development and maintenance of excitatory 2051.e6. synapses in hippocampal neurons. Our results also indi- De Rubeis S, Pasciuto E, Li KW, Fernandez E, Di Marino D, Buzzi cate that regulation of the F-actin dynamics via the WRC A, Ostroff LE, Klann E, Zwartkruis FJ, Komiyama NH, et al. 2013. CYFIP1 coordinates mRNA translation and cytoskele- may be a key mechanism, as we observed changes in F- ton remodeling to ensure proper dendritic spine formation. actin levels and downregulation of WAVE1 in the hippo- Neuron. 79(6):1169–1182. campal CA1 region of Cyfip2 cKO mice. Dominguez-Iturza N, Lo AC, Shah D, Armendariz M, Vannelli A, In conclusion, our results suggest that excitatory Mercaldo V, Trusel M, Li KW, Gastaldo D, Santos AR, et al. neuron-specific reduction of CYFIP2 is insufficient to 2019. The autism- and schizophrenia-associated protein CYFIP1 regulates bilateral brain connectivity and behaviour. induce AD-like pathologies in the hippocampal CA1 Nat Commun. 10(1):3454. region, warranting further investigation into the neur- Dorostkar MM, Zou C, Blazquez-Llorca L, Herms J. 2015. onal circuit-dependent mechanisms of AD pathology Analyzing dendritic spine pathology in Alzheimer’s induced by CYFIP2 reduction. Beyond identifying the disease: problems and opportunities. Acta Neuropathol. pathogenic mechanisms of AD, these investigations 130(1):1–19. will potentially provide critical neuronal targets in the Feng G, Mellor RH, Bernstein M, Keller-Peck C, Nguyen QT, Wallace M, Nerbonne JM, Lichtman JW, Sanes JR. 2000. brain for novel therapeutic approaches to AD. Imaging neuronal subsets in transgenic mice expressing multiple spectral variants of GFP. Neuron. 28(1):41–51. Ghosh A, Mizuno K, Tiwari SS, Proitsi P, Gomez Perez-Nievas B, Disclosure statement Glennon E, Martinez-Nunez RT, Giese KP. 2020. Alzheimer’s No potential conflict of interest was reported by the author(s). disease-related dysregulation of mRNA translation causes key pathological features with ageing. Transl Psychiatry. 10 (1):192. Funding Haan N, Westacott LJ, Carter J, Owen MJ, Gray WP, Hall J, Wilkinson LS. 2021. Haploinsufficiency of the schizophrenia This work was supported by National Research Foundation of and autism risk gene Cyfip1 causes abnormal postnatal hip- Korea (NRF) grants funded by the Korea Government Ministry pocampal neurogenesis through microglial and Arp2/3 of Science and ICT (NRF-2018M3C7A1024603 and NRF- mediated actin dependent mechanisms. Transl Psychiatry. 2021R1A2C4001429 to KH, NRF-2022R1I1A1A01053508 to YZ) 11(1):313. and by a Korea University grant (K2206011 to KH). Habela CW, Yoon KJ, Kim NS, Taga A, Bell K, Bergles DE, Maragakis NJ, Ming GL, Song H. 2020. Persistent Cyfip1 expression is required to maintain the adult subventricular References zone neurogenic niche. J Neurosci. 40(10):2015–2024. Han K, Chen H, Gennarino VA, Richman R, Lu HC, Zoghbi HY. Abekhoukh S, Bardoni B. 2014. CYFIP family proteins between 2015. Fragile X-like behaviors and abnormal cortical dendri- autism and intellectual disability: links with fragile X syn- tic spines in cytoplasmic FMR1-interacting protein 2-mutant drome. Front Cell Neurosci. 8:81. mice. Hum Mol Genet. 24(7):1813–1823. Begemann A, Sticht H, Begtrup A, Vitobello A, Faivre L, Banka S, Han K, Holder JL, Schaaf CP, Lu H, Chen H, Kang H, Tang J, Wu Z, Alhaddad B, Asadollahi R, Becker J, Bierhals T, et al. 2021. Hao S, Cheung SW, et al. 2013. SHANK3 overexpression New insights into the clinical and molecular spectrum of causes manic-like behaviour with unique pharmacogenetic the novel CYFIP2-related neurodevelopmental disorder properties. Nature. 503(7474):72–77. and impairment of the WRC-mediated actin dynamics. Hong N, Park JS, Kim HJ. 2022. Synapto-protective effect of Genet Med. 23(3):543–554. lithium on HIV-1 Tat-induced synapse loss in rat hippocam- Bozdagi O, Sakurai T, Dorr N, Pilorge M, Takahashi N, Buxbaum pal cultures. Anim Cells Syst (Seoul). 26(1):1–9. JD. 2012. Haploinsufficiency of Cyfip1 produces fragile X-like Kanellopoulos AK, Mariano V, Spinazzi M, Woo YJ, McLean C, phenotypes in mice. PLoS One. 7(8):e42422. Pech U, Li KW, Armstrong JD, Giangrande A, Callaerts P, Choi SY, Pang K, Kim JY, Ryu JR, Kang H, Liu Z, Kim WK, Sun W, et al. 2020. Aralar sequesters GABA into hyperactive mito- Kim H, Han K. 2015. Post-transcriptional regulation of chondria, causing social behavior deficits. Cell. 180 SHANK3 expression by microRNAs related to multiple neu- (6):1178–1197.e20. ropsychiatric disorders. Mol Brain. 8(1):74. Kang M, Zhang Y, Kang HR, Kim S, Ma R, Yi Y, Lee S, Kim Y, Li H, Chung L, Wang X, Zhu L, Towers AJ, Cao X, Kim IH, Jiang YH. Jin C, et al. 2023. CYFIP2 p.Arg87Cys causes neurological 2015. Parental origin impairment of synaptic functions and defects and degradation of CYFIP2. Ann Neurol. 93(1):155– behaviors in cytoplasmic FMRP interacting protein 1 (Cyfip1) deficient mice. Brain Res. 1629:340–350. Kim GH, Zhang Y, Kang HR, Lee SH, Shin J, Lee CH, Kang H, Ma Cioni JM, Wong HH, Bressan D, Kodama L, Harris WA, Holt CE. R, Jin C, Kim Y, et al. 2020. Altered presynaptic function and 2018. Axon-Axon interactions regulate topographic optic number of mitochondria in the medial prefrontal cortex of tract sorting via CYFIP2-dependent WAVE complex function. adult Cyfip2 heterozygous mice. Mol Brain. 13(1):123. Neuron. 97(5):1078–1093.e6. ANIMAL CELLS AND SYSTEMS 101 Kim NS, Ringeling FR, Zhou Y, Nguyen HN, Temme SJ, Lin YT, Rottner K, Stradal TEB, Chen B. 2021. WAVE regulatory complex. Eacker S, Dawson VL, Dawson TM, Xiao B, et al. 2022. Curr Biol. 31(10):R512–R517. CYFIP1 dosages exhibit divergent behavioral impact via dia- Schenck A, Bardoni B, Moro A, Bagni C, Mandel JL. 2001.A metric regulation of NMDA receptor complex translation in highly conserved protein family interacting with the fragile mouse models of psychiatric disorders. Biol Psychiatry. 92 X mental retardation protein (FMRP) and displaying selec- (10):815–826. tive interactions with FMRP-related proteins FXR1P and Lee B, Zhang Y, Kim Y, Kim S, Lee Y, Han K. 2017. Age-depen- FXR2P. Proc Natl Acad Sci USA . 98(15):8844–8849. dent decrease of GAD65/67 mRNAs but normal densities Silva AI, Haddon JE, Ahmed Syed Y, Trent S, Lin TE, Patel Y, of GABAergic interneurons in the brain regions of Shank3- Carter J, Haan N, Honey RC, Humby T, et al. 2019.Cyfip1 hap- overexpressing manic mouse model. Neurosci Lett. loinsufficient rats show white matter changes, myelin thin- 649:48–54. ning, abnormal oligodendrocytes and behavioural Lee SH, Zhang Y, Park J, Kim B, Kim Y, Lee SH, Kim GH, Huh YH, inflexibility. Nat Commun. 10(1):3455. Lee B, Kim Y, et al. 2020. Haploinsufficiency of Cyfip2 causes Spence EF, Soderling SH. 2015. Actin out: regulation of the lithium-responsive prefrontal dysfunction. Ann Neurol. 88 synaptic cytoskeleton. J Biol Chem. 290(48):28613–28622. (3):526–543. Tiwari SS, Mizuno K, Ghosh A, Aziz W, Troakes C, Daoud J, Lee Y, Kim D, Ryu JR, Zhang Y, Kim S, Kim Y, Lee B, Sun W, Han Golash V, Noble W, Hortobagyi T, Giese KP. 2016. K. 2017. Phosphorylation of CYFIP2, a component of the Alzheimer-related decrease in CYFIP2 links amyloid pro- WAVE-regulatory complex, regulates dendritic spine duction to tau hyperphosphorylation and memory loss. density and neurite outgrowth in cultured hippocampal Brain. 139(Pt 10):2751–2765. neurons potentially by affecting the complex assembly. Tsien JZ, Chen DF, Gerber D, Tom C, Mercer EH, Anderson DJ, Neuroreport. 28(12):749–754. Mayford M, Kandel ER, Tonegawa S. 1996. Subregion- and Lee Y, Zhang Y, Kang H, Bang G, Kim Y, Kang HR, Ma R, Jin C, cell type-restricted gene knockout in mouse brain. Cell. 87 Kim JY, Han K. 2020. Epilepsy-and intellectual disability- (7):1317–1326. associated CYFIP2 interacts with both actin regulators and Tzioras M, McGeachan RI, Durrant CS, Spires-Jones TL. 2023. RNA-binding proteins in the neonatal mouse forebrain. Synaptic degeneration in Alzheimer disease. Nat Rev Biochem Biophys Res Commun. 529(1):1–6. Neurol. 19(1):19–38. Lee Y, Zhang Y, Ryu JR, Kang HR, Kim D, Jin C, Kim Y, Sun W, Yu X, Zhang R, Wei C, Gao Y, Yu Y, Wang L, Jiang J, Zhang X, Li J, Han K. 2019. Reduced CYFIP2 stability by Arg87 variants Chen X. 2021. MCT2 overexpression promotes recovery of causing human neurological disorders. Ann Neurol. 86 cognitive function by increasing mitochondrial biogenesis (5):803–805. in a rat model of stroke. Anim Cells Syst (Seoul). 25(2):93– Luo L, Ambrozkiewicz MC, Benseler F, Chen C, Dumontier E, 101. Falkner S, Furlanis E, Gomez AM, Hoshina N, Huang WH, Zalfa F, Eleuteri B, Dickson KS, Mercaldo V, De Rubeis S, di Penta et al. 2020. Optimizing nervous system-specific gene target- A, Tabolacci E, Chiurazzi P, Neri G, Grant SG, et al. 2007.A ing with Cre driver lines: prevalence of germline recombina- new function for the fragile X mental retardation protein tion and influencing factors. Neuron. 106(1):37–65.e5. in regulation of PSD-95 mRNA stability. Nat Neurosci. 10 Ma R, Pang K, Kang H, Zhang Y, Bang G, Park S, Hwang E, Ryu (5):578–587. JR, Kwon Y, Kang HR, et al. 2022. Protein interactome and Zhang Y, Kang H, Lee Y, Kim Y, Lee B, Kim JY, Jin C, Kim S, Kim H, cell-type expression analyses reveal that cytoplasmic Han K. 2019a. Smaller body size, early postnatal lethality, FMR1-interacting protein 1 (CYFIP1), but not CYFIP2, associ- and cortical extracellular matrix-related gene expression ates with astrocytic focal adhesion. J Neurochem. 162 changes of Cyfip2-null embryonic mice. Front Mol (2):190–206. Neurosci. 11:482. Nakashima M, Kato M, Aoto K, Shiina M, Belal H, Mukaida S, Zhang Y, Kang HR, Han K. 2019b.Differential cell-type- Kumada S, Sato A, Zerem A, Lerman-Sagie T, et al. 2018. expression of CYFIP1 and CYFIP2 in the adult mouse hippo- De novo hotspot variants in CYFIP2 cause early-onset epi- campus. Animal Cells Syst (Seoul). 23(6):380–383. leptic encephalopathy. Ann Neurol. 83(4):794–806. Zhang Y, Kang HR, Lee SH, Kim Y, Ma R, Jin C, Lim JE, Kim S, Napoli I, Mercaldo V, Boyl PP, Eleuteri B, Zalfa F, De Rubeis S, Di Kang Y, Kang H, et al. 2020. Enhanced prefrontal neuronal Marino D, Mohr E, Massimi M, Falconi M, et al. 2008. The activity and social dominance behavior in postnatal fore- fragile X syndrome protein represses activity-dependent brain excitatory neuron-specificCyfip2 knock-out mice. translation through CYFIP1, a new 4E-BP. Cell. 134 Front Mol Neurosci. 13:574947. (6):1042–1054. Zhang Y, Lee Y, Han K. 2019c. Neuronal function and dysfunc- Pathania M, Davenport EC, Muir J, Sheehan DF, Lopez- tion of CYFIP2: from actin dynamics to early infantile epi- Domenech G, Kittler JT. 2014. The autism and schizophrenia leptic encephalopathy. BMB Rep. 52(5):304–311. associated gene CYFIP1 is critical for the maintenance of Zhong M, Liao S, Li T, Wu P, Wang Y, Wu F, Li X, Hong S, Yan L, dendritic complexity and the stabilization of mature Jiang L. 2019. Early diagnosis improving the outcome of an spines. Transl Psychiatry. 4:e374. infant with epileptic encephalopathy with cytoplasmic Peng J, Wang Y, He F, Chen C, Wu LW, Yang LF, Ma YP, Zhang FMRP interacting protein 2 mutation: case report and litera- W, Shi ZQ, Xia K, et al. 2018. Novel West syndrome candidate ture review. Medicine (Baltimore). 98(44):e17749. genes in a Chinese cohort. CNS Neurosci Ther. 24(12):1196– Zweier M, Begemann A, McWalter K, Cho MT, Abela L, Banka S, 1206. Behring B, Berger A, Brown CW, Carneiro M, et al. 2019. Penzes P, Cahill ME, Jones KA, VanLeeuwen JE, Woolfrey KM. Spatially clustering de novo variants in CYFIP2, encoding 2011. Dendritic spine pathology in neuropsychiatric dis- the cytoplasmic FMRP interacting protein 2, cause intellec- orders. Nat Neurosci. 14(3):285–293. tual disability and seizures. Eur J Hum Genet. 27(5):747–759. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Animal Cells and Systems Taylor & Francis

Cell-autonomous reduction of CYFIP2 is insufficient to induce Alzheimer's disease-like pathologies in the hippocampal CA1 pyramidal neurons of aged mice

Download PDF
9 pages

Loading next page...
 
/lp/taylor-francis/cell-autonomous-reduction-of-cyfip2-is-insufficient-to-induce-9hejYasr83

References (88)

Publisher
Taylor & Francis
Copyright
© 2023 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group
ISSN
1976-8354
eISSN
2151-2485
DOI
10.1080/19768354.2023.2192263
Publisher site
See Article on Publisher Site

Abstract

ANIMAL CELLS AND SYSTEMS 2023, VOL. 27, NO. 1, 93–101 https://doi.org/10.1080/19768354.2023.2192263 Cell-autonomous reduction of CYFIP2 is insufficient to induce Alzheimer’s disease-like pathologies in the hippocampal CA1 pyramidal neurons of aged mice a,b a a,b a,b a a,b Ruiying Ma , Yinhua Zhang , Huiling Li , Hyae Rim Kang , Yoonhee Kim and Kihoon Han a b Department of Neuroscience, Korea University College of Medicine, Seoul, Republic of Korea; BK21 Graduate Program, Department of Biomedical Sciences, Korea University College of Medicine, Seoul, Republic of Korea ABSTRACT ARTICLE HISTORY Received 9 February 2023 Cytoplasmic FMR1-interacting protein 2 (CYFIP2) is an evolutionarily conserved multifunctional Revised 7 March 2023 protein that regulates the neuronal actin cytoskeleton, mRNA translation and transport, and Accepted 8 March 2023 mitochondrial morphology and function. Supporting its critical roles in proper neuronal development and function, human genetic studies have repeatedly identified variants of the KEYWORDS CYFIP2 gene in individuals diagnosed with neurodevelopmental disorders. Notably, a few CYFIP2; Alzheimer’s disease; recent studies have also suggested a mechanistic link between reduced CYFIP2 level and hippocampal CA1; excitatory Alzheimer’s disease (AD). Specifically, in the hippocampus of 12-month-old Cyfip2 pyramidal neuron; heterozygous mice, several AD-like pathologies were identified, including increased levels of conditional knock-out Tau phosphorylation and gliosis, and loss of dendritic spines in CA1 pyramidal neurons. However, detailed pathogenic mechanisms, such as cell types and their circuits where the pathologies originate, remain unknown for AD-like pathologies caused by CYFIP2 reduction. In this study, we aimed to address this issue by examining whether the cell-autonomous reduction of CYFIP2 in CA1 excitatory pyramidal neurons is sufficient to induce AD-like phenotypes in the hippocampus. We performed immunohistochemical, morphological, and biochemical analyses in 12-month-old Cyfip2 conditional knock-out mice, which have postnatally reduced CYFIP2 expression level in CA1, but not in CA3, excitatory pyramidal neurons of the hippocampus. Unexpectedly, we could not find any significant AD-like phenotype, suggesting that the CA1 excitatory neuron-specific reduction of CYFIP2 level is insufficient to lead to AD-like pathologies in the hippocampus. Therefore, we propose that CYFIP2 reduction in other neurons and/or their synaptic connections with CA1 pyramidal neurons may be critically involved in the hippocampal AD-like phenotypes of Cyfip2 heterozygous mice. Introduction 2019; Zweier et al. 2019; Begemann et al. 2021;Kang et al. 2023). The two members of the cytoplasmic FMR1-interacting At the molecular level, CYFIP1 and CYFIP2 have a high protein family, CYFIP1 and CYFIP2, are evolutionarily amino acid sequence homology and both are involved conserved proteins whose genetic variants are causally in the regulation of cellular actin cytoskeleton dynamics associated with numerous brain disorders, including as a critical component of the heteropentameric autism spectrum disorders, intellectual disability, Wiskott–Aldrich syndrome protein family verprolin- schizophrenia, and epilepsy (Schenck et al. 2001; homologous protein (WAVE) regulatory complex (WRC) Abekhoukh and Bardoni 2014; Zhang, Lee, et al. (Lee Y et al. 2017; Rottner et al. 2021). Additional func- 2019). Specifically, in the case of CYFIP2, de novo var- tions in neurons, such as regulation of mRNA translation iants have recently been identified in individuals and transport (Napoli et al. 2008; De Rubeis et al. 2013; diagnosed with neurodevelopmental disorders and Cioni et al. 2018), and regulation of mitochondrial func- early-onset epileptic encephalopathy characterized by tion and morphology (Kanellopoulos et al. 2020; Kim GH developmental regression, intellectual disability, seizures, et al. 2020) have also been identified, but unlike actin muscular hypotonia, and microcephaly (Nakashima et al. regulation via the WRC, these functions may be 2018;Peng etal. 2018; Lee et al. 2019; Zhong et al. CONTACT Kihoon Han neurohan@korea.ac.kr Supplemental data for this article can be accessed online at https://doi.org/10.1080/19768354.2023.2192263. © 2023 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. The terms on which this article has been published allow the posting of the Accepted Manuscript in a repository by the author(s) or with their consent. 94 R. MA ET AL. mediated by both shared and distinct interactors of pathologies in aged mice remains largely unknown. CYFIP1 and CYFIP2, respectively (Lee Y et al. 2020;Ma In particular, considering the complex interactions et al. 2022). Collectively, these molecular functions of between different neuronal and non-neuronal cell CYFIP1 and CYFIP2 can contribute to the morphological types and the trans-synaptic spread of pathology in and functional changes in neuronal synapses observed AD (Tzioras et al. 2023), investigating the cell types in both Cyfip1 and Cyfip2 mutant mice (Bozdagi et al. and their circuits where CYFIP2-dependent pathologies 2012; Pathania et al. 2014; Han et al. 2015; Davenport originate is a critical step toward understanding the et al. 2019; Lee SH et al. 2020; Zhang et al. 2020; Kim pathogenic mechanisms. In this study, we aimed to NS et al. 2022). address this issue by examining the AD-like pheno- Moreover, several lines of evidence indicate the differ- types of aged (12-month-old) Cyfip2 conditional ential roles of CYFIP1 and CYFIP2 in vivo, including the knock-out (cKO) mice, which have postnatally reduced embryonic and perinatal lethality of Cyfip1-null mice CYFIP2 expression level selectively in CA1, but not in and Cyfip2-null mice, respectively (Chung et al. 2015; CA3, excitatory pyramidal neurons of the hippocampus. Han et al. 2015; Zhang et al. 2019a), and different brain Surprisingly, unlike Cyfip2 het mice, there was no overt regional and cell-type distributions of CYFIP1 and AD-like phenotype in the hippocampal CA1 region of CYFIP2 mRNAs and proteins (Zhang, Kang, et al. 2019b; aged Cyfip2 cKO mice. Therefore, our results suggest Ma et al. 2022). In particular, CYFIP1, but not CYFIP2, is that cell-autonomous reduction of CYFIP2 is insufficient also expressed in non-neuronal cells, such as astrocytes, for AD-like pathologies in CA1 pyramidal neurons and microglia, and oligodendrocytes, in the brain (Domin- that other neurons and/or their synaptic connections guez-Iturza et al. 2019; Silva et al. 2019; Habela et al. with CA1 pyramidal neurons are also critically involved 2020; Haan et al. 2021; Ma et al. 2022). in the hippocampal AD-like phenotypes of Cyfip2 het Beyond the genetic association between CYFIP2 and mice. neurodevelopmental disorders, recent studies have suggested a mechanistic association between the Materials and methods reduction of CYFIP2 and Alzheimer’s disease (AD) (Tiwari et al. 2016; Ghosh et al. 2020). Specifically, Mice CYFIP2 protein levels were reduced in the postmortem The Cyfip2 cKO mice and Thy1-YFP mice used in this forebrain of patients with AD and in the hippocampus study were previously described (Lee SH et al. 2020; and cortex of AD model mice (Tiwari et al. 2016). More- Zhang et al. 2020). The mice were fed ad libitum and over, the protein levels of amyloid precursor protein housed under a 12 h light–dark cycle. All experiments (APP), β-site APP cleaving enzyme 1 (BACE1), and were performed using aged (12-month-old) male calcium/calmodulin-dependent protein kinase IIα Cyfip2 cKO mice and their littermate controls. (CaMKIIα) were up-regulated in the hippocampal synap- tosomal fraction of conventional Cyfip2 heterozygous +/– (Cyfip2 , het) mice. Consistently, in the hippocampus Fluorescence immunohistochemistry of aged (12-month-old) Cyfip2 het mice, several AD-like pathologies have been observed, including increased Fluorescence immunohistochemistry was performed as levels of Tau phosphorylation and gliosis, and significant previously described (Lee B et al. 2017; Yu et al. 2021). loss of dendritic spines in CA1 pyramidal neurons (Ghosh Mice were anesthetized with isoflurane and transcardially et al. 2020). Mechanistically, it has been proposed that perfused with heparinized (20 units/mL) phosphate- reduced expression level of CYFIP2 induces aberrant buffered saline (PBS), followed by 4% paraformaldehyde local mRNA translation of several AD-related proteins (PFA) in PBS. The brains were extracted and post-fixed (i.e. APP, BACE1, and CaMKIIα) at the synaptic compart- overnight in 4% PFA. Following post-fixation, brain ment, thereby leading to the overproduction of Aβ and tissue was washed with PBS and cryoprotected with hyperphosphorylation of Tau in Cyfip2 het mice (Ghosh 30% sucrose in PBS for 48 h. The brain tissues were et al. 2020). Under normal conditions, CYFIP2, as shown frozen in an O.C.T compound (SAKURA Tissue-Tek, for CYFIP1 (Napoli et al. 2008; De Rubeis et al. 2013), 4583) and sectioned (60 µm) using a cryostat microtome may repress the translation of these mRNAs by forming (Leica, CM3050S). The primary antibodies used for immu- an inhibitory complex with the RNA-binding protein nohistochemistry were AT-8 (Phospho-Tau [Ser202, fragile X messenger ribonucleoprotein (FMRP) and the Thr205], Invitrogen, #MN1020), CYFIP1 (Sigma-Aldrich, eukaryotic initiation factor 4E (eIF4E). #AB6046), CYFIP2 (Abcam, #ab95969), GFAP (Abcam, However, compared to CYFIP2 function and dysfunc- #ab4674), Iba1 (Synaptic System, #234-006), NeuN tion in the developing brain, its role in AD-like (Abcam, #ab177487; Millipore, #MAB377). F-actin was ANIMAL CELLS AND SYSTEMS 95 visualized by Alexa Fluor 488-conjugated Phalloidin (Invi- Results trogen, #A-12379). The samples were washed with 0.1% CA1 excitatory pyramidal neuron-specific Triton X-100 in PBS and blocked with PBS containing reduction of CYFIP2 in the hippocampus of aged 3% bovine serum albumin (BSA) and 0.5% Triton X-100. Cyfip2 cKO mice The high-resolution image acquisition was performed using a Zeiss LSM800 confocal microscope equipped To investigate whether cell-autonomous CYFIP2 with a 20×/0.8 objective lens, 8-bit image depth, and reduction in excitatory pyramidal neurons is sufficient snapshot mode focused on maximum intensity. Whole to induce AD-like phenotypes in the hippocampal CA1 brain regions were obtained using a slide scanner region, we crossed floxed-Cyfip2 mice with CaMKIIα-Cre floxed/floxed (Zeiss Axio Scan.Z1). The regions of the stratum oriens mice to generate Cyfip2 cKO (Cyfip2 ; CaMKIIα- (SO) and stratum radiatum (SR) were defined as 100 Cre) mice as previously described (Zhang et al. 2020). µm and 100–200 µm away from the cell body area The CaMKIIα-Cre line (T29-1) used in this study starts (stratum pyramidale, SP) of CA1, respectively. The expressing Cre recombinases in forebrain excitatory values of at least two brain sections were averaged for neurons during the third to fourth postnatal weeks, each mouse. and especially in the hippocampus, Cre expression is restricted mainly to the CA1 region (Tsien et al. 1996). floxed/floxed Notably, to securely obtain control (Cyfip2 ) Dendritic spine analysis and Cyfip2 cKO progeny mice, we crossed male floxed/floxed floxed/floxed The dendritic spine analysis was performed as pre- Cyfip2 mice with female Cyfip2 ; viously described (Han et al. 2013;Choietal. 2015; CaMKIIα-Cre mice (Figure 1(A)) to avoid unwanted germ- Hong et al. 2022). Mice were deeply anesthetized line recombination, which was recently reported in male, with isoflurane and transcardially perfused with but not female, T29-1 mice (Luo et al. 2020). Further- heparinized (20 units/mL) PBS followed by 4% PFA more, we designed an additional primer set for in PBS. The brains were extracted and post-fixed genotyping PCR to detect germline deletion of floxed overnight in 4% PFA. After post-fixation, coronal sec- exon 6 of the Cyfip2 gene (Lee SH et al. 2020), which tions (100 µm thickness) of the hippocampal region indeed produced an expected PCR band from some floxed/Δexon6 were obtained using a vibratome (VT1000S, Leica). portions of progeny mice (i.e. Cyfip2 mice) floxed/floxed The sections were collected and stored in 50% gly- when we crossed male Cyfip2 ;CaMKIIα-Cre floxed/floxed cerol in 2 × PBS at −20 C until further processed. mice with female Cyfip2 mice as a test Blocking, permeabilization, and anti-GFP (Abcam, (Figure 1(B) and (C)). Using this primer set, we #AB13970) primary and Alexa Fluorconjugated (anti- confirmed that the control and Cyfip2 cKO mice used chicken Alexa Fluor-488, Jackson ImmunoResearch in this study did not have a germline deletion of the Labs, #703-545-155) secondary antibody incubation floxed Cyfip2 exon 6. were performed as described above. Finally, the sec- Fluorescence immunohistochemical analysis showed tions were mounted on slide glasses with mounting that CYFIP2 protein levels were selectively reduced in media (Biomeda, M02). Images of dendritic spines the hippocampal CA1, but not CA3, of 12-month-old in the secondary or tertiary branches (apical or Cyfip2 cKO mice (Figure 1(D)), which was expected basal dendrites of YFP-positive CA1 pyramidal from CA1-restricted Cre expression in T29-1 mice. Mean- neurons in the hippocampus) were acquired by con- while, CYFIP1 protein levels in the CA1 region were com- focal microscopy (Zeiss LSM800) using 63×/1.2 water parable between control and Cyfip2 cKO mice, immersion objective lens, 8-bit image depth, and Z- suggesting that there is no compensatory increase in stack function with 0.93 µm intervals, followed by CYFIP1 level in the hippocampus of aged Cyfip2 cKO Z-stack projection of maximum intensity. Images mice (Figure 1(E)). were analyzed using ImageJ software. For quantifi- cation of dendritic spines, mushroom spines were defined as protrusions with heads and with a width No AD-like immunohistological phenotype in the greater than length. Stubby spines were defined as hippocampal CA1 region of aged Cyfip2 cKO mice protrusions without a neck. The rest of the protru- sions with heads were categorized as thin spines. A previous study showed that several AD-like immu- The values of six to eight neurons were averaged nohistological phenotypes were significantly exacer- for each mouse. bated in the hippocampal CA1 region of 12-month- Additional information of Materials and Methods is old Cyfip2 het mice compared with age-matched included in supporting online material. wild-type (WT) mice (Ghosh et al. 2020). These 96 R. MA ET AL. Figure 1. No AD-like immunohistological phenotype in the hippocampal CA1 region of aged Cyfip2 cKO mice. (A) The breeding f/f f/f scheme for the control (Cyfip2 ) and Cyfip2 conditional knock-out (Cyfip2 ;CaMKIIα-Cre, cKO) mice. (B) The breeding scheme to test the partial germline recombination of male CaMKIIα-Cre mice. (C) Design of primers to detect floxed and Δexon6 Cyfip2 f/f floxed/Δexon6 alleles (left panel). Results of PCR for the tail genomic DNA isolated from Cyfip2 and Cyfip2 mice (right panel). Note f/f that the primer set (a + c) does not produce the expected ∼1.1 kbp band from the DNA sample of Cyfip2 mice due to the short elongation time of the PCR. (D) Fluorescence immunohistochemistry images and quantification showing CA1-specific reduction of CYFIP2 in the hippocampus of aged Cyfip2 cKO mice. CA, cornu ammonis; DG, dentate gyrus; NS, not significant; SO, stratum oriens; SP, stratum pyramidale; SR, stratum radiatum. (E, F) Normal CYFIP1 and phospho-Tau (AT-8) levels in the hippocampal CA1 region of aged Cyfip2 cKO mice. (G, H) Normal density and total intensity of astrocytes (GFAP-positive) and microglia (Iba1-positive) in the hippocampal CA1 region of aged Cyfip2 cKO mice. N =7–8 mice. phenotypes include increased levels of phospho-Tau [GFAP] and ionized calcium-binding adapter molecule immunoreactivity (measured by monoclonal AT-8 1 [Iba1] antibodies, respectively). Therefore, we antibody) and gliosis (both for astrocytes and micro- performed fluorescence immunohistochemical ana- glia, as measured by glial fibrillary acidic protein lyses of these AD markers in the hippocampal CA1 ANIMAL CELLS AND SYSTEMS 97 region of 12-month-old Cyfip2 cKO mice and their littermate controls. However, there were no significant differences in phospho-Tau levels between control and Cyfip2 cKO mice (Figure 1(F)). Moreover, neither GFAP nor Iba1 positive cell number or total intensity was significantly altered in Cyfip2 cKO mice compared to control mice (Figure 1 (G) and (H)). Reduced number of immature dendritic spines and increased F-actin levels in the basal dendrites of CA1 pyramidal neurons of aged Cyfip2 cKO mice Dendritic spines are small dendritic protrusions that rep- resent the most excitatory postsynapses in the brain (Penzes et al. 2011). Loss of dendritic spines is another key feature of AD (Dorostkar et al. 2015) and was signifi- cantly aggravated in CA1 pyramidal neurons of 12- month-old Cyfip2 het mice compared to age-matched WT mice (Ghosh et al. 2020). In particular, the number of mature-type mushroom spines, but not that of imma- ture-type thin spines, was reduced in the apical den- drites of CA1 pyramidal neurons in aged Cyfip2 het mice. Therefore, we analyzed the dendritic spines of CA1 pyramidal neurons in 12-month-old control and Cyfip2 cKO mice. Dendritic spines were visualized by crossing Cyfip2 cKO mice with Thy1-YFP mice (Feng et al. 2000) that sparsely express yellow fluorescent protein (YFP) in CA1 pyramidal neurons (Figure 2(A)). We separately analyzed the basal and apical dendrites of the neurons (Figure 2(B)). In the basal dendrites, we found that the total number of dendritic spines was sig- nificantly reduced in Cyfip2 cKO neurons compared to control neurons (Figure 2(C)). However, unlike the mush- room spine-specific reduction in Cyfip2 het mice (Ghosh et al. 2020), the spine reduction in Cyfip2 cKO mice was mainly attributed to a decrease in immature-type thin spines. Moreover, in the apical dendrites, neither total Figure 2. Dendritic spine and F-actin changes in the hippocam- density nor morphologically-based categorization of pal CA1 region of aged Cyfip2 cKO mice. (A) Visualization of CA1 dendritic spines was altered in Cyfip2 cKO neurons com- pyramidal neurons in Thy1-YFP mice by sparse expression of yellow fluorescent protein (YFP). CA, cornu ammonis; DG, pared to control neurons (Figure 2(C)), suggesting that dentate gyrus; SO, stratum oriens; SP, stratum pyramidale; SR, there is no AD-like dendritic spine phenotype in CA1 pyr- stratum radiatum. (B) Representative confocal images of dendri- amidal neurons of aged Cyfip2 cKO mice. We also com- tic spines in the basal and apical dendrites of CA1 pyramidal pared head size for both thin and mushroom spines neurons of aged control and Cyfip2 cKO mice. Examples of den- between the control and Cyfip2 cKO neurons and dritic spines in each morphologically-based categorization (thin, found no significant differences in the basal or apical stubby, and mushroom) are indicated by arrows with different colors. (C) Quantification of dendritic spine number in the dendrites (Figure 2(D)). basal (upper panel) and apical (lower panel) dendrites. NS, not F-actin is a key cytoskeletal component of dendritic significant. (D) Quantification of dendritic spine head size in spines and is directly associated with their formation, basal (upper panel) and apical (lower panel) dendrites. (E) Repre- maintenance, and dynamics (Spence and Soderling sentative confocal images and quantification of F-actin levels in 2015). As a critical component of the WRC, CYFIP2 is the hippocampal CA1 region of aged control and Cyfip2 cKO mice. N =4–8 mice. involved in actin regulation in various cellular 98 R. MA ET AL. compartments, including neuronal dendritic spines were comparable between control and Cyfip2 cKO (Rottner et al. 2021). Specifically, it has been previously mice. Notably, we found that neither APP nor CaMKIIα shown increased F-actin levels in the medial prefrontal protein levels were significantly altered in Cyfip2 cKO cortex (mPFC) of young adult Cyfip2 het and Cyfip2 mice compared to control mice (Figure 3(B)). Further- cKO mice (Lee SH et al. 2020; Zhang et al. 2020). There- more, the levels of PSD-95, another synaptic protein fore, we also measured F-actin levels in the hippocampal whose mRNA stability and translation are regulated by CA1 region and found an increase in the stratum oriens FMRP (Zalfa et al. 2007), and FMRP itself were normal (SO), but not in the stratum radiatum (SR), of aged Cyfip2 in the CA1 synaptosome of aged Cyfip2 cKO mice. cKO mice compared to control mice (Figure 2(E)). Taken together, these results suggest that there is no overt AD-like pathology in the hippocampal CA1 region of 12-month-old Cyfip2 cKO mice. Normal expression of synaptosomal APP and CaMKIIα in the hippocampal CA1 region of aged Cyfip2 cKO mice Discussion Overexpression of AD-related proteins, such as APP and In this study, we combined immunohistochemical, mor- CaMKIIα, in the synaptic compartment due to aberrant phological, and biochemical approaches to understand local mRNA translation, has been proposed as a molecu- whether the cell-autonomous reduction of CYFIP2 in lar mechanism underlying AD-like pathologies in Cyfip2 excitatory pyramidal neurons is sufficient to induce het mice (Tiwari et al. 2016; Ghosh et al. 2020). Therefore, AD-like pathologies in the hippocampal CA1 region. we analyzed the protein levels of APP and CaMKIIα in the However, none of the results showed a significant AD- hippocampus of 12-month-old control and Cyfip2 cKO like phenotype in aged Cyfip2 cKO mice, in contrast to mice. We prepared a crude synaptosomal fraction from the severe phenotypes observed in Cyfip2 het mice the dissected hippocampal CA1 region and performed (Tiwari et al. 2016; Ghosh et al. 2020). Therefore, our immunoblotting (Figure 3(A)). Consistent with the results suggest that other neurons and/or their synaptic immunohistological analysis, CYFIP2 levels were connections with CA1 pyramidal neurons are also criti- reduced in hippocampal CA1 synaptosomal lysates cally involved in the hippocampal AD-like phenotypes from Cyfip2 cKO mice compared to control mice of Cyfip2 het mice. Additional genetic or viral tools to (Figure 3(B)). As expected, the WAVE1 protein, another reduce CYFIP2 protein levels in specific or combinatorial component of the WRC, was also reduced in Cyfip2 neurons of the hippocampal circuit will help us further cKO lysates because the stability of WAVE1 is inter- address this issue. dependent with that of CYFIP2 (Han et al. 2015; Zhang Based on previous findings, we speculate on some et al. 2020; Kang et al. 2023). In contrast, CYFIP1 levels mechanisms that explain the lack of an AD-like Figure 3. Normal synaptosomal expression levels of AD-related proteins, APP and CaMKIIα, in the hippocampal CA1 region of aged Cyfip2 cKO mice. (A) Schematic diagram showing the dissection of the CA1 region of the mouse hippocampus. (B) Representative immunoblot images and quantification of the expression levels of CYFIP2, WAVE1, CYFIP1, APP, CaMKIIα, PSD-95, and FMRP proteins in CA1 synaptosomal fraction of aged Cyfip2 cKO mice compare to control mice. Protein levels were normalized by either a neuron- specific protein, neuron-specific enolase (NSE), or GAPDH. NS, not significant. N = 5 mice. ANIMAL CELLS AND SYSTEMS 99 Figure 4. Schematic diagrams summarizing the different hippocampal phenotypes among aged wild-type (WT), Cyfip2 het, and Cyfip2 cKO mice. CYFIP2 protein levels are reduced in both CA1 and CA3 excitatory and inhibitory neurons in Cyfip2 het mice, but they are only reduced in CA1 excitatory neurons in Cyfip2 cKO mice. In the hippocampal CA1 region, AD-like pathologies, such as gliosis and dendritic spine loss, are observed in Cyfip2 het mice. In contrast, no AD-like phenotype is observed in Cyfip2 cKO mice. Ex., excitatory; Inh., inhibitory. phenotype in Cyfip2 cKO mice (Figure 4). CYFIP2 mRNAs synaptic functions in neurons of Cyfip2 cKO mice could and proteins are predominantly expressed in neurons preserve local mRNA translation within the normal compared to non-neuronal cells in the brain and are range. However, in Cyfip2 het mice, more profound detected in both excitatory and local inhibitory changes in synaptic activity, due to both presynaptic neurons (Zhang, Kang, et al. 2019b; Lee SH et al. 2020; and postsynaptic reduction of CYFIP2, and changes in Ma et al. 2022). Moreover, CYFIP2 is expressed in the CYFIP2-FMRP-eIF4E complex may congruently lead neurons of the hippocampal CA3 region as well as of to aberrant mRNA translation and overproduction of other brain regions that can directly form synaptic con- AD-related proteins. Further investigations of the mol- nections with CA1 excitatory pyramidal neurons (Han ecular composition and function of the CYFIP2-FMRP- et al. 2015; Lee SH et al. 2020). As CYFIP2 regulates eIF4E complex in Cyfip2 het and Cyfip2 cKO neurons axonal and presynaptic development and function are needed to test this hypothesis. (Cioni et al. 2018; Kim GH et al. 2020), it is conceivable Considering the aforementioned speculations regard- that both presynaptic and postsynaptic compartments ing the mechanisms involved, another plausible expla- are functionally affected in CA1 pyramidal neurons of nation for the observed phenotypic difference Cyfip2 het mice, thereby ultimately leading to synaptic between Cyfip2 het and Cyfip2 cKO CA1 pyramidal loss. Meanwhile, the postsynapse-specific CYFIP2 neurons could be the delayed onset of disease in reduction in CA1 pyramidal neurons of Cyfip2 cKO Cyfip2 cKO mice compared to Cyfip2 het mice. Due to mice may be insufficient to induce such changes. Furthe- the preservation of normal presynaptic inputs in CA1 more, considering the concept of trans-synaptic propa- pyramidal neurons of Cyfip2 cKO mice, it is possible gation of pathology in AD (Tzioras et al. 2023), bi- that the neurons will take a longer time to exhibit AD- directional spread of AD pathology through synaptic like phenotypes compared to the neurons of Cyfip2 het connections with other abnormal neurons may synergis- mice. This hypothesis can be examined by investigating tically worsen the phenotypes of CA1 pyramidal neurons the hippocampus of older (e.g. 18-month-old) Cyfip2 of Cyfip2 het mice compared to those of Cyfip2 cKO mice. cKO mice and age-matched control mice. At the molecular level, local mRNA translation of AD- We observed a reduced number of thin, but normal related proteins by the CYFIP2-FMRP-eIF4E complex can stubby and mushroom spines in the basal dendrites of be differentially affected in the CA1 pyramidal neurons CA1 pyramidal neurons of aged Cyfip2 cKO mice. of Cyfip2 het and Cyfip2 cKO mice (Ghosh et al. 2020). However, there was no significant change in the Indeed, we observed normal synaptosomal levels of number of thin, stubby, and mushroom spines in the APP and CaMKIIα proteins in the hippocampal CA1 of apical dendrites. The basal dendrite-specific decrease Cyfip2 cKO mice, unlike their increased levels in Cyfip2 in dendritic spine number was also observed in the het mice (Tiwari et al. 2016). The interaction of CYFIP1 layer 5 neurons of the mPFC of young adult Cyfip2 cKO with FMRP and eIF4E is regulated by synaptic activity mice (Zhang et al. 2020), although the underlying mech- (De Rubeis et al. 2013). Therefore, it can be speculated anism remains unknown. Additionally, a previous study that even with CYFIP2 reduction, relatively normal found that overexpression of CYFIP2 increases excitatory 100 R. MA ET AL. Davenport EC, Szulc BR, Drew J, Taylor J, Morgan T, Higgs synapse number in cultured hippocampal neurons at 14 NF, Lopez-Domenech G, Kittler JT. 2019. Autism and days in vitro (Davenport et al. 2019), suggesting that schizophrenia-associated CYFIP1 regulates the balance of CYFIP2 dosage may be an important factor regulating synaptic excitation and inhibition. Cell Rep. 26(8):2037– the development and maintenance of excitatory 2051.e6. synapses in hippocampal neurons. Our results also indi- De Rubeis S, Pasciuto E, Li KW, Fernandez E, Di Marino D, Buzzi cate that regulation of the F-actin dynamics via the WRC A, Ostroff LE, Klann E, Zwartkruis FJ, Komiyama NH, et al. 2013. CYFIP1 coordinates mRNA translation and cytoskele- may be a key mechanism, as we observed changes in F- ton remodeling to ensure proper dendritic spine formation. actin levels and downregulation of WAVE1 in the hippo- Neuron. 79(6):1169–1182. campal CA1 region of Cyfip2 cKO mice. Dominguez-Iturza N, Lo AC, Shah D, Armendariz M, Vannelli A, In conclusion, our results suggest that excitatory Mercaldo V, Trusel M, Li KW, Gastaldo D, Santos AR, et al. neuron-specific reduction of CYFIP2 is insufficient to 2019. The autism- and schizophrenia-associated protein CYFIP1 regulates bilateral brain connectivity and behaviour. induce AD-like pathologies in the hippocampal CA1 Nat Commun. 10(1):3454. region, warranting further investigation into the neur- Dorostkar MM, Zou C, Blazquez-Llorca L, Herms J. 2015. onal circuit-dependent mechanisms of AD pathology Analyzing dendritic spine pathology in Alzheimer’s induced by CYFIP2 reduction. Beyond identifying the disease: problems and opportunities. Acta Neuropathol. pathogenic mechanisms of AD, these investigations 130(1):1–19. will potentially provide critical neuronal targets in the Feng G, Mellor RH, Bernstein M, Keller-Peck C, Nguyen QT, Wallace M, Nerbonne JM, Lichtman JW, Sanes JR. 2000. brain for novel therapeutic approaches to AD. Imaging neuronal subsets in transgenic mice expressing multiple spectral variants of GFP. Neuron. 28(1):41–51. Ghosh A, Mizuno K, Tiwari SS, Proitsi P, Gomez Perez-Nievas B, Disclosure statement Glennon E, Martinez-Nunez RT, Giese KP. 2020. Alzheimer’s No potential conflict of interest was reported by the author(s). disease-related dysregulation of mRNA translation causes key pathological features with ageing. Transl Psychiatry. 10 (1):192. Funding Haan N, Westacott LJ, Carter J, Owen MJ, Gray WP, Hall J, Wilkinson LS. 2021. Haploinsufficiency of the schizophrenia This work was supported by National Research Foundation of and autism risk gene Cyfip1 causes abnormal postnatal hip- Korea (NRF) grants funded by the Korea Government Ministry pocampal neurogenesis through microglial and Arp2/3 of Science and ICT (NRF-2018M3C7A1024603 and NRF- mediated actin dependent mechanisms. Transl Psychiatry. 2021R1A2C4001429 to KH, NRF-2022R1I1A1A01053508 to YZ) 11(1):313. and by a Korea University grant (K2206011 to KH). Habela CW, Yoon KJ, Kim NS, Taga A, Bell K, Bergles DE, Maragakis NJ, Ming GL, Song H. 2020. Persistent Cyfip1 expression is required to maintain the adult subventricular References zone neurogenic niche. J Neurosci. 40(10):2015–2024. Han K, Chen H, Gennarino VA, Richman R, Lu HC, Zoghbi HY. Abekhoukh S, Bardoni B. 2014. CYFIP family proteins between 2015. Fragile X-like behaviors and abnormal cortical dendri- autism and intellectual disability: links with fragile X syn- tic spines in cytoplasmic FMR1-interacting protein 2-mutant drome. Front Cell Neurosci. 8:81. mice. Hum Mol Genet. 24(7):1813–1823. Begemann A, Sticht H, Begtrup A, Vitobello A, Faivre L, Banka S, Han K, Holder JL, Schaaf CP, Lu H, Chen H, Kang H, Tang J, Wu Z, Alhaddad B, Asadollahi R, Becker J, Bierhals T, et al. 2021. Hao S, Cheung SW, et al. 2013. SHANK3 overexpression New insights into the clinical and molecular spectrum of causes manic-like behaviour with unique pharmacogenetic the novel CYFIP2-related neurodevelopmental disorder properties. Nature. 503(7474):72–77. and impairment of the WRC-mediated actin dynamics. Hong N, Park JS, Kim HJ. 2022. Synapto-protective effect of Genet Med. 23(3):543–554. lithium on HIV-1 Tat-induced synapse loss in rat hippocam- Bozdagi O, Sakurai T, Dorr N, Pilorge M, Takahashi N, Buxbaum pal cultures. Anim Cells Syst (Seoul). 26(1):1–9. JD. 2012. Haploinsufficiency of Cyfip1 produces fragile X-like Kanellopoulos AK, Mariano V, Spinazzi M, Woo YJ, McLean C, phenotypes in mice. PLoS One. 7(8):e42422. Pech U, Li KW, Armstrong JD, Giangrande A, Callaerts P, Choi SY, Pang K, Kim JY, Ryu JR, Kang H, Liu Z, Kim WK, Sun W, et al. 2020. Aralar sequesters GABA into hyperactive mito- Kim H, Han K. 2015. Post-transcriptional regulation of chondria, causing social behavior deficits. Cell. 180 SHANK3 expression by microRNAs related to multiple neu- (6):1178–1197.e20. ropsychiatric disorders. Mol Brain. 8(1):74. Kang M, Zhang Y, Kang HR, Kim S, Ma R, Yi Y, Lee S, Kim Y, Li H, Chung L, Wang X, Zhu L, Towers AJ, Cao X, Kim IH, Jiang YH. Jin C, et al. 2023. CYFIP2 p.Arg87Cys causes neurological 2015. Parental origin impairment of synaptic functions and defects and degradation of CYFIP2. Ann Neurol. 93(1):155– behaviors in cytoplasmic FMRP interacting protein 1 (Cyfip1) deficient mice. Brain Res. 1629:340–350. Kim GH, Zhang Y, Kang HR, Lee SH, Shin J, Lee CH, Kang H, Ma Cioni JM, Wong HH, Bressan D, Kodama L, Harris WA, Holt CE. R, Jin C, Kim Y, et al. 2020. Altered presynaptic function and 2018. Axon-Axon interactions regulate topographic optic number of mitochondria in the medial prefrontal cortex of tract sorting via CYFIP2-dependent WAVE complex function. adult Cyfip2 heterozygous mice. Mol Brain. 13(1):123. Neuron. 97(5):1078–1093.e6. ANIMAL CELLS AND SYSTEMS 101 Kim NS, Ringeling FR, Zhou Y, Nguyen HN, Temme SJ, Lin YT, Rottner K, Stradal TEB, Chen B. 2021. WAVE regulatory complex. Eacker S, Dawson VL, Dawson TM, Xiao B, et al. 2022. Curr Biol. 31(10):R512–R517. CYFIP1 dosages exhibit divergent behavioral impact via dia- Schenck A, Bardoni B, Moro A, Bagni C, Mandel JL. 2001.A metric regulation of NMDA receptor complex translation in highly conserved protein family interacting with the fragile mouse models of psychiatric disorders. Biol Psychiatry. 92 X mental retardation protein (FMRP) and displaying selec- (10):815–826. tive interactions with FMRP-related proteins FXR1P and Lee B, Zhang Y, Kim Y, Kim S, Lee Y, Han K. 2017. Age-depen- FXR2P. Proc Natl Acad Sci USA . 98(15):8844–8849. dent decrease of GAD65/67 mRNAs but normal densities Silva AI, Haddon JE, Ahmed Syed Y, Trent S, Lin TE, Patel Y, of GABAergic interneurons in the brain regions of Shank3- Carter J, Haan N, Honey RC, Humby T, et al. 2019.Cyfip1 hap- overexpressing manic mouse model. Neurosci Lett. loinsufficient rats show white matter changes, myelin thin- 649:48–54. ning, abnormal oligodendrocytes and behavioural Lee SH, Zhang Y, Park J, Kim B, Kim Y, Lee SH, Kim GH, Huh YH, inflexibility. Nat Commun. 10(1):3455. Lee B, Kim Y, et al. 2020. Haploinsufficiency of Cyfip2 causes Spence EF, Soderling SH. 2015. Actin out: regulation of the lithium-responsive prefrontal dysfunction. Ann Neurol. 88 synaptic cytoskeleton. J Biol Chem. 290(48):28613–28622. (3):526–543. Tiwari SS, Mizuno K, Ghosh A, Aziz W, Troakes C, Daoud J, Lee Y, Kim D, Ryu JR, Zhang Y, Kim S, Kim Y, Lee B, Sun W, Han Golash V, Noble W, Hortobagyi T, Giese KP. 2016. K. 2017. Phosphorylation of CYFIP2, a component of the Alzheimer-related decrease in CYFIP2 links amyloid pro- WAVE-regulatory complex, regulates dendritic spine duction to tau hyperphosphorylation and memory loss. density and neurite outgrowth in cultured hippocampal Brain. 139(Pt 10):2751–2765. neurons potentially by affecting the complex assembly. Tsien JZ, Chen DF, Gerber D, Tom C, Mercer EH, Anderson DJ, Neuroreport. 28(12):749–754. Mayford M, Kandel ER, Tonegawa S. 1996. Subregion- and Lee Y, Zhang Y, Kang H, Bang G, Kim Y, Kang HR, Ma R, Jin C, cell type-restricted gene knockout in mouse brain. Cell. 87 Kim JY, Han K. 2020. Epilepsy-and intellectual disability- (7):1317–1326. associated CYFIP2 interacts with both actin regulators and Tzioras M, McGeachan RI, Durrant CS, Spires-Jones TL. 2023. RNA-binding proteins in the neonatal mouse forebrain. Synaptic degeneration in Alzheimer disease. Nat Rev Biochem Biophys Res Commun. 529(1):1–6. Neurol. 19(1):19–38. Lee Y, Zhang Y, Ryu JR, Kang HR, Kim D, Jin C, Kim Y, Sun W, Yu X, Zhang R, Wei C, Gao Y, Yu Y, Wang L, Jiang J, Zhang X, Li J, Han K. 2019. Reduced CYFIP2 stability by Arg87 variants Chen X. 2021. MCT2 overexpression promotes recovery of causing human neurological disorders. Ann Neurol. 86 cognitive function by increasing mitochondrial biogenesis (5):803–805. in a rat model of stroke. Anim Cells Syst (Seoul). 25(2):93– Luo L, Ambrozkiewicz MC, Benseler F, Chen C, Dumontier E, 101. Falkner S, Furlanis E, Gomez AM, Hoshina N, Huang WH, Zalfa F, Eleuteri B, Dickson KS, Mercaldo V, De Rubeis S, di Penta et al. 2020. Optimizing nervous system-specific gene target- A, Tabolacci E, Chiurazzi P, Neri G, Grant SG, et al. 2007.A ing with Cre driver lines: prevalence of germline recombina- new function for the fragile X mental retardation protein tion and influencing factors. Neuron. 106(1):37–65.e5. in regulation of PSD-95 mRNA stability. Nat Neurosci. 10 Ma R, Pang K, Kang H, Zhang Y, Bang G, Park S, Hwang E, Ryu (5):578–587. JR, Kwon Y, Kang HR, et al. 2022. Protein interactome and Zhang Y, Kang H, Lee Y, Kim Y, Lee B, Kim JY, Jin C, Kim S, Kim H, cell-type expression analyses reveal that cytoplasmic Han K. 2019a. Smaller body size, early postnatal lethality, FMR1-interacting protein 1 (CYFIP1), but not CYFIP2, associ- and cortical extracellular matrix-related gene expression ates with astrocytic focal adhesion. J Neurochem. 162 changes of Cyfip2-null embryonic mice. Front Mol (2):190–206. Neurosci. 11:482. Nakashima M, Kato M, Aoto K, Shiina M, Belal H, Mukaida S, Zhang Y, Kang HR, Han K. 2019b.Differential cell-type- Kumada S, Sato A, Zerem A, Lerman-Sagie T, et al. 2018. expression of CYFIP1 and CYFIP2 in the adult mouse hippo- De novo hotspot variants in CYFIP2 cause early-onset epi- campus. Animal Cells Syst (Seoul). 23(6):380–383. leptic encephalopathy. Ann Neurol. 83(4):794–806. Zhang Y, Kang HR, Lee SH, Kim Y, Ma R, Jin C, Lim JE, Kim S, Napoli I, Mercaldo V, Boyl PP, Eleuteri B, Zalfa F, De Rubeis S, Di Kang Y, Kang H, et al. 2020. Enhanced prefrontal neuronal Marino D, Mohr E, Massimi M, Falconi M, et al. 2008. The activity and social dominance behavior in postnatal fore- fragile X syndrome protein represses activity-dependent brain excitatory neuron-specificCyfip2 knock-out mice. translation through CYFIP1, a new 4E-BP. Cell. 134 Front Mol Neurosci. 13:574947. (6):1042–1054. Zhang Y, Lee Y, Han K. 2019c. Neuronal function and dysfunc- Pathania M, Davenport EC, Muir J, Sheehan DF, Lopez- tion of CYFIP2: from actin dynamics to early infantile epi- Domenech G, Kittler JT. 2014. The autism and schizophrenia leptic encephalopathy. BMB Rep. 52(5):304–311. associated gene CYFIP1 is critical for the maintenance of Zhong M, Liao S, Li T, Wu P, Wang Y, Wu F, Li X, Hong S, Yan L, dendritic complexity and the stabilization of mature Jiang L. 2019. Early diagnosis improving the outcome of an spines. Transl Psychiatry. 4:e374. infant with epileptic encephalopathy with cytoplasmic Peng J, Wang Y, He F, Chen C, Wu LW, Yang LF, Ma YP, Zhang FMRP interacting protein 2 mutation: case report and litera- W, Shi ZQ, Xia K, et al. 2018. Novel West syndrome candidate ture review. Medicine (Baltimore). 98(44):e17749. genes in a Chinese cohort. CNS Neurosci Ther. 24(12):1196– Zweier M, Begemann A, McWalter K, Cho MT, Abela L, Banka S, 1206. Behring B, Berger A, Brown CW, Carneiro M, et al. 2019. Penzes P, Cahill ME, Jones KA, VanLeeuwen JE, Woolfrey KM. Spatially clustering de novo variants in CYFIP2, encoding 2011. Dendritic spine pathology in neuropsychiatric dis- the cytoplasmic FMRP interacting protein 2, cause intellec- orders. Nat Neurosci. 14(3):285–293. tual disability and seizures. Eur J Hum Genet. 27(5):747–759.

Journal

Animal Cells and Systems – Taylor & Francis

Published: Dec 11, 2023

Keywords: CYFIP2; Alzheimer’s disease; hippocampal CA1; excitatory pyramidal neuron; conditional knock-out

There are no references for this article.