Abstract

   Cyclin Y (CCNY), which is a cyclin protein known to play a role in cell
   division, is unexpectedly and thus interestingly expressed in
   non-proliferating neuronal cells. There have been only a few studies
   reporting the neuronal functions of CCNY in synapse remodeling and
   hippocampal long-term potentiation. Therefore, we here provide global
   and comprehensive information on the putative functions of CCNY in
   biological and functional pathways in neuronal systems. We adopted
   high-throughput RNA-sequencing technology for analyzing transcriptomes
   regulated by CCNY and utilized bioinformatics for identifying putative
   molecules, biological processes, and functional pathways that are
   possibly connected to CCNY functions in hippocampal neuronal cells of
   rats. We revealed that several enriched annotation terms and pathways
   associated with CCNY expression within neurons, including apoptosis,
   learning or memory, synaptic plasticity, actin cytoskeleton, focal
   adhesion, extracellular matrix-receptor interaction and chemokine
   signaling pathway are targeted by CCNY. In addition, the mRNA levels of
   some genes enriched for those annotation terms and pathways or genes
   reported to be altered in Alzheimer’s disease mouse model were further
   validated by quantitative real-time PCR in hippocampal neuronal cells.
   The present study provides an excellent resource for future
   investigations of CCNY functions in neuronal systems.

Introduction

   Cyclin Y (CCNY) is one of the members of the cyclin family that has
   been known to regulate cell division in proliferating cells
   [[34]1–[35]3]. CCNY was originally identified as an interacting protein
   of the cyclin-dependent kinase CDK14/PFTK1 via a yeast two-hybrid
   screen [[36]4]. Its role has been investigated in the field of cancer
   biology by showing that CCNY regulates glioma and lung cancer cell
   proliferation [[37]5, [38]6]. In addition, CCNY played an essential
   role in the maintenance of mammary stem/progenitor cell properties
   [[39]7] and the control of adipogenesis and lipid production [[40]8].
   Furthermore, CCNY was a key factor for the development of Drosophila,
   including larval growth, pupal development and metamorphosis [[41]2].

   Interestingly, CCNY has been shown to play roles in non-dividing
   neuronal cells. Role of CCNY in the nervous system was first described
   in C. elegans as a regulator for synapse formation and elimination
   [[42]9, [43]10], and it was also found in the mammalian nervous system
   as a negative regulator for hippocampal long-term potentiation (LTP)
   [[44]11], the most widely studied cellular basis of learning and memory
   [[45]12–[46]15]. Investigating the function of CCNY in the
   non-proliferating neuronal cells is intriguing since CCNY has been
   generally known for its role in proliferating cells. Although a few
   studies reported on the role of CCNY in the nervous system
   [[47]9–[48]11], the mechanistic and signaling information on how CCNY
   functions in the brain remains mostly unknown. In this study, we
   provide candidate molecules, biological processes and functional
   signaling pathways that might be regulated by CCNY, a relatively novel
   molecule whose function has been rarely investigated.

   RNA sequencing (RNA-seq), which is a recent revolutionary tool
   providing an accurate and precise measurement of transcript levels, has
   been widely applied for systematic, comprehensive, and global analysis
   of transcriptome in various species [[49]16–[50]18]. This
   next-generation high-throughput sequencing technology has provided an
   unbiased approach for investigating pathophysiology of
   neurodegenerative diseases [[51]19–[52]22]. In this study, the RNA-seq
   technique, bioinformatics, and quantitative real-time PCR (qRT-PCR)
   have been adopted to extract molecular profiles that are regulated by
   CCNY in hippocampal neuronal cells and provide invaluable information
   on putative biological processes, molecular functions and functional
   signaling pathways that CCNY may be involved in hippocampal neuronal
   system. The extensive and essential resources provided in the present
   study will serve as a platform for future investigations of CCNY
   function in neuronal systems.

Materials and methods

Cell culture

   HEK 293T cells were grown in DMEM (HyClone) supplemented with 10% fetal
   bovine serum. Hippocampal neuron cultures were prepared from E18
   Sprague-Dawley (SD) rat embryos and maintained for 14–21 days in vitro
   (DIV) [[53]11]. All experiments handling animals and their embryos were
   performed in accordance with the guidelines and regulations of the
   Korea Institute of Science and Technology (KIST). All experimental
   protocols were approved by the KIST Institutional Animal Care and Use
   Committee (IACUC; approval number 2016–065).

DNA constructs

   The same constructs from our previous study [[54]11] were used for
   CCNY-WT-EGFP, FUGW-CCNY-WT, and FUGW-CCNY-shRNA.

Immunocytochemistry

   For staining endogenous PSD-95, hippocampal neurons on coverslips were
   fixed with 4% paraformaldehyde/4% sucrose in phosphate-buffered saline
   (PBS) for 15–20 min at room temperature and permeated with 0.1%
   TritonX-100 in PBS for 10 min at room temperature. Neurons were then
   incubated with mouse anti-PSD-95 (MA1-046, Thermo fisher scientific,
   1:200) in PBS containing 5% normal donkey serum for 1 hr at room
   temperature. Anti-mouse Cy3-conjugated secondary antibody (1:300) was
   applied for 45 min at room temperature. Coverslips were then mounted on
   slide glasses for imaging.

Production of lentivirus

   Lentivirus expressing EGFP, CCNY-WT-EGFP or CCNY-shRNA-EGFP was
   generated as described in our previous study [[55]11]. Briefly,
   lentiviral vector FUGW, FUGW harboring CCNY-WT or CCNY-shRNA, the
   packaging vector Δ8.9, and VSVG envelope glycoprotein vector were
   co-transfected into HEK 293T cells using X-tremeGENE HP DNA
   transfection reagent (Roche). Thirty six to 48 hours after
   transfection, supernatants containing the lentivirus were harvested,
   aliquoted, and stored at −80°C.

Sample preparation for RNA-seq

   Cultured hippocampal neurons were infected with lentivirus expressing
   EGFP, CCNY-WT-EGFP or CCNY-shRNA-EGFP at DIV5-6, and the neuronal cell
   lysates were harvested at DIV14 for total RNA isolation and subsequent
   RNA-seq.

RNA extraction, cDNA library construction, RNA-Seq and data analysis

   RNA-seq was performed as described [[56]23]. Total RNA was isolated
   using the RNeasy kit (Qiagen, Valencia, CA), the RNA-seq library was
   prepared using the TruSeq RNA Sample Prep Kit (Illumina, San Diego, CA,
   USA) and the sequencing was performed based on Illumina NextSeq500
   platform to generate 150-bp paired-end reads. The sequenced reads were
   mapped to the Rat genome (rn4) using TopHat 2, and the gene expression
   levels were calculated using Cufflinks [[57]24, [58]25]. The cuffdiff
   module in the cufflinks package was used to select differentially
   expressed genes (DEGs) from the RNA-seq data which cover the total
   17,066 genes. Meanwhile, the FPKM value of each gene was floored to 1,
   and log2-transformed for further analysis. Heat maps were constructed
   using Mev [[59]26]. Statistical analyses and graph construction were
   performed using R (v. 3.1.0) and PYTHON (v. 2.7.6).

Bioinformatic analysis

   To extract the over-represented (enriched) biological annotation terms
   and pathways from DEGs, the Database for Annotation, Visualization and
   Integrated Discovery (DAVID) Bioinformatics Resources v6.7
   ([60]https://david.ncifcrf.gov) [[61]27, [62]28] was applied for the
   Gene ontology (GO) analysis on the basis of three categories, including
   biological process, cellular component and molecular function and also
   for the Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment
   analysis [[63]29, [64]30]. The enriched GO terms and enriched KEGG
   pathways were first identified based on the uncorrected P-values, and
   only the terms and pathways that are likely related to mainly neuronal
   functions and broadly cell proliferation were represented as figures
   ([65]S2–[66]S4 Figs). Some of the identified terms and pathways from
   the [67]S2–[68]S4 Figs are also presented as Tables, including the
   individual genes belonging to each term and pathway ([69]S1 and [70]S2
   Tables). RNA-seq data were closely investigated for all individual
   genes belonging to the terms and pathways listed in [71]S1 and [72]S2
   Tables. Then, genes showing apparent disparity in the transcript levels
   between control, CCNY-WT overexpression, and CCNY knockdown samples
   were chosen for quantitative real-time PCR (qRT-PCR) validations. Only
   the genes showing statistical significance between the samples of
   control, CCNY-WT overexpression, and CCNY knockdown in mRNA levels
   quantified by qRT-PCR were presented as data in this study. The result
   of Cxcl1 gene showing no significance between the samples was included
   in the data as a control, which is consistent with the reported study
   [[73]31].

RNA-seq data access

   RNA-seq data have been deposited in the NCBI Gene Expression Omnibus
   (GEO) under the accession number [74]GSE84850.

Quantitative Real-Time PCR (qRT-PCR) and analysis

   Cultured hippocampal neurons infected with lentivirus expressing EGFP,
   CCNY-WT-EGFP or CCNY-shRNA-EGFP for 7–8 days were harvested. Total RNA
   was extracted using RNAiso Plus (TaKaRa, Japan) according to the
   manufacturer’s instructions and was reverse transcribed into cDNA using
   the the PrimeScript II 1^st strand cDNA synthesis kit (TaKaRa, Japan).
   The qRT-PCR was performed using Power SYBR Green PCR Master Mix (Thermo
   fisher scientific). The reaction mixture contained 0.5 μl of cDNA
   corresponding to 75 ng of total RNA, 150 nM of each gene-specific
   primers and 2x Power SYBR Green PCR Master Mix in a total volume of 20
   μl. The cycling parameters of StepOnePlus Real-Time PCR System (Applied
   Biosystems) were as follows: 95°C for 10 min, followed by 40 cycles of
   95°C for 15 sec and 60°C for 1 min. Relative expression level was
   calculated according to the 2^−ΔΔCT algorithm based on the expression
   of glyceraldehyde 3-phosphate dehydrogenase (GAPDH) gene, which did not
   show differential expression among samples in the study. Experiments
   were independently performed 3 times with duplicates each time. Primer
   sets used for qRT-PCR are listed in [75]S3 Table.

Immunoblot analysis

   Equal amounts of protein denatured in SDS sample buffer were applied to
   SDS-PAGE, transferred onto a PVDF membrane, and applied to immunoblot
   reactions. Protein bands were visualized by a chemiluminescence method
   (Millipore or Thermo fisher scientific) and an imaging documentation
   system (ImageQuant LAS 4000, GE healthcare). Primary antibodies against
   CCNY (Proteintech group), GFP (Roche) or β-tubulin (Abcam) were used.

Results

Experimental model system for RNA-seq based transcriptome analysis

   CCNY is unexpectedly expressed in non-proliferating neuronal cells
   [[76]11]. However, there are only few reports on the function of CCNY
   in neuronal systems, which include its regulation of synapse remodeling
   and hippocampal LTP [[77]9–[78]11]. Therefore, further investigation on
   the neuronal functions of CCNY is indispensable. For this aim, we first
   searched for systematic information of gene sets that are possibly
   regulated by CCNY in the hippocampal neurons; the region of hippocampus
   was chosen based on the previous report on the function of CCNY in
   hippocampal LTP [[79]11]. We lentivirally overexpressed or knockdowned
   CCNY in the primary cultured hippocampal neurons ([80]Fig 1a) and
   confirmed that CCNY mRNA ([81]Fig 1b; EGFP control, 1.0 ± 0.012;
   CCNY-WT, 27.87 ± 5.365; CCNY-shRNA, 0.37 ± 0.063) and protein levels
   ([82]Fig 1c) were significantly enhanced and reduced in neurons
   overexpressing and knocking down CCNY, respectively. Consistent with
   the previous report showing the existence of CCNY in the postsynaptic
   subcellular fraction [[83]11], CCNY wild-type (CCNY-WT) localizes
   adjacent to the endogenous postsynaptic density protein-95 (PSD-95)
   ([84]Fig 1d), supporting the value of systematic analysis on putative
   neuronal functions of CCNY.

Fig 1. Experimental model systems for RNA-seq based transcriptome analysis.

   [85]Fig 1
   [86]Open in a new tab

   (a) Cultured hippocampal neurons expressing EGFP (Lenti-EGFP Ctrl),
   CCNY-WT-EGFP (Lenti-CCNY-WT-EGFP) or CCNY-shRNA-EGFP
   (Lenti-CCNY-shRNA-EGFP) via lentiviral expression system. (b,c)
   Relative levels of mRNA (b) and protein (c) of CCNY from the neurons
   infected with lentivirus expressing EGFP, CCNY-WT-EGFP or
   CCNY-shRNA-EGFP. n = 5 from 3 independent experiments. **p<0.01
   relative to control, ***p<0.005 relative to control, student’s t test.
   Refer to the [87]S9 Fig for the full-length blots of (c). (d) CCNY-WT
   exists in the spines near the endogenous PDS-95 in cultured hippocampal
   neurons, supporting the value of the study on systematic analysis for
   putative neuronal functions of CCNY. Scale bars, 20 and 2 μm for the
   whole neuronal and enlarged images, respectively.

Up- and down-regulated genes by CCNY in the hippocampal neurons

   We next aimed to obtain the profile of genes that are regulated by CCNY
   in hippocampal neurons and thus performed RNA-seq from each hippocampal
   neuronal samples that are CCNY overexpressed or CCNY knocked down. Gene
   expression level was calculated as the fragments per kilobase of
   transcript per million fragments mapped (FPKM). We found that the
   log[2]FPKM of CCNY is enhanced in the CCNY overexpressed neurons
   whereas it is reduced in the CCNY knocked down neurons ([88]S1a Fig).
   The fold change (fc) was calculated by subtracting the FPKM value of
   EGFP control from the FPKM value of CCNY-EGFP overexpression or CCNY
   shRNA-mediated knockdown samples. Differentially expressed genes (DEGs)
   were selected by setting |log[2]fc| ≥ 0.3 and |log[2]fc| ≥ 0.33 for
   CCNY overexpression and knockdown, respectively. With these criteria,
   the total of 442 up-regulated and 375 down-regulated DEGs in
   CCNY-WT-overexpressing neurons and 529 up-regulated and 671
   down-regulated DEGs in CCNY-shRNA-knockdown neurons were identified
   ([89]S1b Fig), and the differential expression profiles demonstrating
   up- and down-regulations by CCNY overexpression ([90]Fig 2a) or
   knockdown ([91]Fig 2b) were displayed.

Fig 2. Up- and down-regulated genes by CCNY in the hippocampal neurons.

   [92]Fig 2
   [93]Open in a new tab

   (a,b) Heatmaps of differentially expressed genes (DEGs) for up- or
   down-regulated genes by CCNY overexpression (a) or knockdown (b).
   Selected gene names are listed on the right side of the heatmaps. Red
   indicates up-regulated gene expression level, whereas green indicates
   down-regulated gene expression level. The values of log[2]FPKM were
   normalized to the value ranges from minimum -2.0 to maximum +2.0.

CCNY-mediated regulation of biological processes, including apoptosis and
learning or memory

   To better understand the function of CCNY in neuronal cells, we used
   the Database for Annotation, Visualization and Integrated Discovery
   (DAVID) functional annotation tool for conducting the Gene ontology
   (GO) analysis on the basis of three categories, including biological
   process, cellular component, and molecular function [[94]32] from each
   sets of DEGs shown in [95]Fig 2. Positive regulation of apoptosis in
   biological process, extracellular region/space in cellular component,
   hormone activity, cytokine activity, pattern binding, polysaccharide
   binding, glycosaminoglycan binding, growth factor activity, protein
   dimerization activity, voltage-gated ion channel activity, peptide
   receptor activity, metal ion transmembrane transporter activity, and
   carbohydrate binding in molecular function were selected as the GO
   terms that are significantly up-regulated by CCNY overexpression while
   down-regulated by CCNY knockdown ([96]S2a Fig). In addition, regulation
   of apoptosis and regulation of cell proliferation in biological process
   and extracellular space in cellular component were selected as the GO
   terms that are both significantly up-regulated by CCNY knockdown and
   down-regulated by CCNY overexpression ([97]S2b Fig). Using qRT-PCR, we
   further revealed that the mRNA expression levels of genes such as
   Acvr1c, Crh, Crhr1, Fcgr2a, Gch1, Gnrh1, Mmp9, Rxfp2, Sphk1, Btc,
   Nupr1, and Chek2 belonging to the (positive) regulation of apoptosis in
   our analysis ([98]S1 Table) were regulated by CCNY ([99]Fig 3a). These
   data suggest a role of CCNY in apoptosis, which is highly plausible
   based on the previous reports showing the regulation of synaptic
   plasticity such as LTP and long-term depression (LTD) by a signaling
   pathway involving apoptotic molecules [[100]33, [101]34].

Fig 3. CCNY affects the biological processes related to apoptosis, and
learning or memory in hippocampal neurons.

   [102]Fig 3
   [103]Open in a new tab

   (a,b) Validation of several genes in GO terms regulated by CCNY.
   qRT-PCR validations of several genes belonging to the term “(positive)
   regulation of apoptosis” in up-regulated GO terms by CCNY
   overexpression or down-regulated GO terms by CCNY knockdown (a), the
   term “learning or memory” in down-regulated GO terms by CCNY knockdown
   and the term “regulation of synaptic plasticity” in down-regulated GO
   terms by CCNY knockdown (b). Refer the [104]S1 Table.

   In addition to the terms mentioned above, the GO terms related to
   neuronal functions, including learning or memory, regulation of
   synaptic plasticity, neuron development, regulation of neurological
   system process, and/or positive regulation of glutamatergic synaptic
   transmission in biological process were also significantly enriched in
   the down-regulated DEGs by CCNY knockdown or overexpression ([105]S2a
   Fig), which could be predicted from the previous reports showing the
   CCNY functions in synapse formation, elimination and plasticity
   [[106]9–[107]11]. Several genes (Bche, Klk8, Pcdh8, Slc6a1, Drd2, and
   Hrh3) belonging to the terms such as learning or memory, and synaptic
   plasticity ([108]S1 Table) were further validated to be regulated by
   CCNY using qRT-PCR ([109]Fig 3b).

CCNY-mediated regulation of pathways for the regulation of actin
cytoskeleton, focal adhesion, and Extracellular Matrix (ECM)-receptor
interaction

   We next carried out the Kyoto encyclopedia of genes and genomes (KEGG)
   pathway enrichment analysis [[110]29, [111]30] to identify enriched
   metabolic or signaling pathways in each sets of DEGs shown in [112]Fig
   2. Regulation of actin cytoskeleton and chemokine signaling pathway
   were selected as the KEGG pathways that are up-regulated by CCNY
   overexpression while down-regulated by CCNY knockdown ([113]S3a Fig).
   In addition, neuroactive ligand-receptor interaction, calcium signaling
   pathway, cytokine-cytokine receptor interaction, focal adhesion,
   ECM-receptor interaction, axon guidance, melanoma, antigen processing
   and presentation were identified as the KEGG pathways that are
   significantly down-regulated by CCNY knockdown ([114]S3a Fig). DNA
   replication, systemic lupus erythematosus, and SNARE interactions in
   vesicular transport were significantly enriched as up-regulated KEGG
   pathways by CCNY knockdown, whereas neuroactive ligand-receptor
   interaction, allograft rejection, and autoimmune thyroid disease were
   significantly enriched as down-regulated KEGG pathways by CCNY
   overexpression ([115]S3b Fig).

   Since the role of CCNY in AMPA receptor trafficking and LTP [[116]11]
   is known to be mediated by actin cytoskeleton [[117]35–[118]38] and
   CCNY localizes adjacent to the PSD ([119]Fig 1d) [[120]11], we further
   validated the genes ([121]S2 Table) belonging to the pathways for
   regulation of actin cytoskeleton (Iqub, Itgb5, Itgb8, Pik3r5, Vav2;
   Arpc1b, Chrm4, Chrm5, Gsn; Mylpf, Wasl) ([122]Fig 4a; [123]S3, [124]S5
   and [125]S6 Figs) and focal adhesion/ECM-receptor interaction (Flt1,
   Met, Reln, and Sdc4) ([126]Fig 4b; [127]S7 and [128]S8 Figs),
   respectively, by qRT-PCR. Moreover, previous studies have reported CCNY
   as an inhibitory regulator for LTP [[129]11], which has been assumed as
   a cellular model for learning and memory, and the altered levels of
   several cytokines in memory-deficit Alzheimer’s disease (AD) mouse
   model [[130]31] that led us to further validate several genes (Ccl2,
   Ccl7 and Cxcl1; [131]S2 Table) belonging to chemokine signaling pathway
   in [132]S3a Fig and additional cytokine genes (Ccl3, Ccl5 and Ccl11)
   reported in the cytokines-AD study [[133]31] ([134]Fig 5).
   Interestingly, we found that the cytokines that were altered in the AD
   mouse model [[135]31] were also shown to be regulated by CCNY,
   suggesting a possible role of CCNY in the AD mouse model.

Fig 4. CCNY affects the pathways for the regulation of actin cytoskeleton,
focal adhesion, and Extracellular Matrix (ECM)-receptor interaction in
hippocampal neurons.

   [136]Fig 4
   [137]Open in a new tab

   (a,b) Transcript levels of several genes from “regulation of actin
   cytoskeleton” pathway (a) and “focal adhesion” and “ECM-receptor
   interaction” pathways (b) were analyzed by qRT-PCR. n = 3–5 from 3
   independent experiments. *p<0.05 relative to control, **p <0.01
   relative to control, ***p <0.001 relative to control, ns, not
   significant, student’s t test. Refer the [138]S2 Table.

Fig 5. CCNY regulates genes related to Alzheimer’s Disease (AD) in
hippocampal neurons.

   [139]Fig 5
   [140]Open in a new tab

   qRT-PCR results of several cytokine genes in the chemokine signaling
   pathway ([141]S3a Fig) or genes related to AD from CCNY overexpression
   or knockdown samples. n = 3–5 from 3 independent experiments. *p<0.05
   relative to control, **p <0.01 relative to control, ***p <0.001
   relative to control, ns, not significant, student’s t test. Refer the
   [142]S2 Table.

Analysis of overlapping DEGs regulated by CCNY overexpression and knockdown

   Among 442 DEGs that were up-regulated by CCNY overexpression, 153 and
   11 genes were up- and down-regulated, respectively, by CCNY knockdown.
   Among 375 DEGs that were down-regulated by CCNY overexpression, 145 and
   6 genes were down- and up-regulated, respectively, by CCNY knockdown
   ([143]Fig 6a). mRNA expression levels of several genes (Prtn3, Kcnk13,
   Kcnj10, Ifi30, Crh, Ccl2, Ccl7 and Hand2) out of 11 that were both
   up-regulated by CCNY overexpression and down-regulated by CCNY
   knockdown ([144]Fig 6b and 6c), and a gene (Nrl) out of 6 that were
   both up-regulated by CCNY knockdown and down-regulated by CCNY
   overexpression ([145]Fig 6b and 6d) were further validated by qRT-PCR.

Fig 6. Analysis of overlapping DEGs controlled by CCNY overexpression and
knockdown.

   [146]Fig 6
   [147]Open in a new tab

   (a) Venn diagram showing the overlaps among the four categories. (b-d)
   Some genes are oppositely regulated by CCNY overexpression and
   knockdown. (b) Heatmap for the genes overlapping up-regulated by CCNY
   overexpression and down-regulated by CCNY knockdown (upper). Heatmap
   for the genes overlapping down-regulated by CCNY overexpression and
   up-regulated by CCNY knockdown (lower). (c,d) Several genes from (b)
   were validated by qRT-PCR. n = 3–5 from 3 independent experiments.
   *p<0.05 relative to control, **p <0.01 relative to control, ***p <0.001
   relative to control, ns, not significant, student’s t test. (e,f)
   Heatmaps for the genes overlapping up-regulated by CCNY overexpression
   and up-regulated by CCNY knockdown (e) and the genes overlapping
   down-regulated by CCNY overexpression and down-regulated by CCNY
   knockdown (f).

   The 153 DEGs that were both up-regulated by CCNY overexpression and
   knockdown ([148]Fig 6e) and the 145 DEGs that were both down-regulated
   by CCNY overexpression and knockdown ([149]Fig 6f) were further
   analyzed for GO term enrichment ([150]S4 Fig). It implies that a
   certain expression level of CCNY is important for physiological
   maintenance of the GO terms identified in [151]S4 Fig.

Discussion

   In the present study, for the first time, we demonstrated the
   expression patterns of transcripts targeted by the cyclin protein CCNY
   in non-proliferating hippocampal neuronal cells. By conducting
   bioinformatic analysis, including GO terms and KEGG pathways with the
   RNA-seq data, we examined the CCNY-mediated changes in transcriptome
   pattern and made subsequent validations on the transcript changes with
   qRT-PCR in hippocampal neuronal cells. We analyzed total of eight
   categories of the genes targeted by CCNY overexpression and/or
   knockdown ([152]Fig 6a): up- or down-regulated by CCNY overexpression,
   up- or down-regulated by CCNY knockdown, up- or down-regulated by both
   CCNY overexpression and knockdown, up-regulated by CCNY overexpression
   and down-regulated by CCNY knockdown, and up-regulated by CCNY
   knockdown and down-regulated by CCNY overexpression. We thoroughly
   examined the patterns of analysis results and found that several GO
   terms and KEGG pathways should be considered significantly valuable for
   further investigations on their roles particularly in respect to the
   CCNY-mediated functional phenotypes in neuronal systems.

   Through the RNA-seq, transcriptome analysis and qRT-PCR, we presented
   that actin cytoskeleton, learning or memory, focal adhesion,
   ECM-receptor interaction, apoptosis, chemokine signaling, and
   AD-related cytokines are affected by CCNY in neuronal cells. Based on
   our previous report on the CCNY function in LTP [[153]11] which has
   been considered as a cellular mechanism for learning and memory and the
   findings on CCNY regulation of several AD-related cytokines ([154]Fig
   5), it is highly plausible that CCNY plays a role in learning and
   memory-deficit neurological diseases such as AD [[155]39].

   The actin cytoskeleton is abundant in the dendritic spines and is an
   essential factor for spine structure and plasticity [[156]40–[157]44].
   In addition, actin dynamics has been reported to play an important role
   in controlling AMPA receptor trafficking and in bidirectional synaptic
   plasticity such as LTP and LTD [[158]35–[159]38, [160]45, [161]46].
   Importantly, CCNY, which was shown to be localized in spines ([162]Fig
   1d) [[163]11], has been suggested as an inhibitory regulator for AMPA
   receptor delivery to synapses during LTP-inducing stimulation
   [[164]11]. Furthermore, in the present study, the pathway for
   regulation of actin cytoskeleton was identified to be oppositely
   regulated by CCNY overexpression and knockdown, and several genes
   involved in the pathway were qRT-PCR-validated as being regulated by
   CCNY ([165]S3 Fig and [166]Fig 4a). Therefore, it will be important to
   provide direct experimental evidence on the role of CCNY in the actin
   dynamics by investigating the cellular and molecular mechanisms
   underlying the coordination between CCNY and actin cytoskeleton
   signaling for neuronal structural and functional plasticity.

   The terms and pathways such as DNA replication, mitotic cell cycle, and
   regulation of cell proliferation were also identified by the DAVID
   analysis in our experimental systems. Since our culture system used in
   this study is neurons enriched but glia repressed, those terms and
   pathways that are related to cell division and are isolated as being
   targeted by CCNY further suggest and support the idea that cyclin
   proteins, including CCNY, may play roles in neuronal functions by
   controlling synapses, synaptic plasticity and/or memory [[167]47]
   plausibly through the formulation of neuron-specific functional
   molecular networks.

   Transcript expression level of individual genes that we displayed in
   the present study may indicate the protein expression (mRNA
   translation) level of the individual genes. Given that CCNY plays a
   role in synaptic plasticity, the genes isolated as being regulated by
   CCNY in this study could be involved in the maintenance of synaptic
   plasticity and memory (re)consolidation, which require new protein
   synthesis [[168]48–[169]52]. In other words, those genes that did not
   exhibit any changes by CCNY should not be overlooked since they may be
   involved in the early stage of synaptic plasticity, which does not
   require a process of protein synthesis. Therefore, it will be valuable
   to explore and compare the transcriptome profile changes controlled by
   CCNY in an activity-dependent manner with several temporal windows
   after given an activity.

   The global and comprehensive bioinformatic analysis of transcriptome
   controlled by CCNY and consecutive validations with qRT-PCR in neuronal
   cells revealed some valuable molecules as being regulated by CCNY in
   neuronal cells and provided useful information on the neuronal and/or
   synaptic role of CCNY by suggesting several functional pathways. It
   will require in future to experimentally elucidate the CCNY functions
   in nervous systems in vitro and in vivo.

   The GO and KEGG pathway analysis adopted in this study provide a
   meaningful biology associated with a list of genes from a large number
   of genes through the systematical classifications of DEGs and the
   statistical over-representation (enrichment) to the functional
   pathways. However, we cannot overlook the plausibility of other
   biological meanings that could be missed by the enrichment analysis
   since the analysis only relies on the known genes and known functions,
   it cannot reveal unknown functions even in the known genes. Moreover,
   some biological pathways have been more investigated than others, which
   relatively builds up more database for the enrichment analysis and thus
   likely analyzed with more significance than others [[170]27, [171]28,
   [172]53, [173]54].

Supporting information

   S1 Fig. RNA-seq based CCNY expression and Differentially Expressed
   Genes (DEGs).

   (a) Expression level of CCNY. The values of log[2]FPKM were used. (b)
   The number of DEGs that were up- or down-regulated by CCNY
   overexpression or knockdown.

   (PDF)
   [174]Click here for additional data file.^ (9.8KB, pdf)
   S2 Fig. Gene Ontology (GO) analysis of CCNY expression level-responsive
   DEGs.

   (a) GO terms of DEGs up-regulated by CCNY overexpression or
   down-regulated by CCNY knockdown were analyzed. (b) GO terms of DEGs
   up-regulated by CCNY knockdown or down-regulated by CCNY overexpression
   were analyzed. *p<0.05, significantly enriched GO terms in DEGs.

   (PDF)
   [175]Click here for additional data file.^ (153.2KB, pdf)
   S3 Fig. KEGG pathway enrichment analysis of CCNY expression
   level-responsive DEGs.

   (a,b) KEGG pathways were analyzed from the DEGs up-regulated by CCNY
   overexpression or down-regulated by CCNY knockdown (a) and the DEGs
   up-regulated by CCNY knockdown or down-regulated by CCNY overexpression
   (b). *p<0.05, significantly enriched KEGG pathways in DEGs. The Y-axes
   indicate the pathway categories, and the X-axes indicate the enrichment
   of the pathways.

   (PDF)
   [176]Click here for additional data file.^ (34.2KB, pdf)
   S4 Fig. GO term enrichment analysis of DEGs in [177]Fig 6e and 6f.

   GO analysis for the 153 DEGs that were both up-regulated by CCNY
   overexpression and knockdown and for the 145 DEGs that were both
   down-regulated by CCNY overexpression and knockdown. *p<0.05,
   significantly enriched GO terms in DEGs.

   (PDF)
   [178]Click here for additional data file.^ (206KB, pdf)
   S5 Fig. KEGG pathways of regulation of actin cytoskeleton.

   Red stars mark up-regulated mRNAs targeted by CCNY overexpression in
   cultured hippocampal neurons.

   (PDF)
   [179]Click here for additional data file.^ (255KB, pdf)
   S6 Fig. KEGG pathways of regulation of actin cytoskeleton.

   Red stars mark down-regulated mRNAs targeted by CCNY knockdown in
   cultured hippocampal neurons.

   (PDF)
   [180]Click here for additional data file.^ (254.1KB, pdf)
   S7 Fig. KEGG pathways of focal adhesion.

   Red stars mark down-regulated mRNAs targeted by CCNY knockdown in
   cultured hippocampal neurons.

   (PDF)
   [181]Click here for additional data file.^ (237KB, pdf)
   S8 Fig. KEGG pathways of ECM-receptor interaction.

   Red stars mark down-regulated mRNAs targeted by CCNY knockdown in
   cultured hippocampal neurons.

   (PDF)
   [182]Click here for additional data file.^ (183.6KB, pdf)
   S9 Fig. Original blots for immunoblot analysis in [183]Fig 1c.

   (PDF)
   [184]Click here for additional data file.^ (84.6KB, pdf)
   S1 Table. Examples of genes that belong to GO terms in [185]S2 Fig.

   (PDF)
   [186]Click here for additional data file.^ (13KB, pdf)
   S2 Table. Examples of genes that belong to KEGG pathways in [187]S3
   Fig.

   (PDF)
   [188]Click here for additional data file.^ (13.2KB, pdf)
   S3 Table. Oligonucleotides used for quantitative real-time PCR.

   (PDF)
   [189]Click here for additional data file.^ (99.6KB, pdf)

Acknowledgments