Transcriptomic Investigation in CRISPR/Cas9-Mediated GRIK1 -, GRIK2 -, and GRIK4 -Gene-Knockout Human Neuroblastoma Cells

: The glutamate ionotropic kainate receptors, encoded by the GRIK gene family, are composed of four subunits and function as ligand-activated ion channels. They play a critical role in regulating synaptic transmission and various synaptic receptors’ processes, as well as in the pathophysiology of schizophrenia. However, their functions and mechanisms of action need to be better understood and are worthy of exploration. To further understand the exact role of the kainate receptors in vitro, we generated kainate-receptor-knockout (KO) isogenic SH-SY5Y cell lines using the CRISPR/Cas9-mediated gene editing method. We conducted RNA sequencing (RNA-seq) to determine the differentially expressed genes (DEGs) in the isogenic edited cells and used rhodamine-phalloidin staining to quantitate filamentous actin (F-actin) in differentiated edited cells. The RNA-seq and the Gene Ontology enrichment analysis revealed that the genetic deletion of the GRIK1 , GRIK2 , and GRIK4 genes disturbed multiple genes involved in numerous signal pathways, including a converging pathway related to the synaptic membrane. An enrichment analysis of gene–disease associations indicated that DEGs in the edited cell lines were associated with several neuropsychiatric disorders, especially schizophrenia. In the morphology study, fluorescent images show that less F-actin was expressed in differentiated SH-SY5Y cells with GRIK1 , GRIK2 , or GRIK4 deficiency than wild-type cells. Our data indicate that kainate receptor deficiency might disturb synaptic-membrane-associated genes, and elucidating these genes should shed some light on the pathophysiology of schizophrenia. Furthermore, the transcriptomic profiles for kainate receptor deficiency of SH-SY5Y cells contribute to emerging evidence for the novel mechanisms underlying the effect of kainate receptors and the pathophysiology of schizophrenia. In addition, our data suggest that kainate-receptor-mediated F-actin remodeling may be a candidate mechanism underlying schizophrenia.

The kainate receptors encoded by the GRIK gene family are composed of five subunits (GRIK1, GRIK2, GRIK3, GRIK4, and GRIK5) and function as ligand-activated ion channels [6].The kainate receptors dominate brain regions and play critical roles in synaptic plasticity, transmission, learning, and memory [4,7,8].Several studies show that abnormal kainate receptor expression in the brain was observed in subjects with schizophrenia [9][10][11][12][13].Rare variations in a group of genes linked to synaptic development, function, and plasticity were burdened in patients with schizophrenia [14].Notably, several reports identified rare novel SynBio 2024, 2 57 mutations of the GRIK gene family, suggesting a potential role for rare and significant effects of mutations of the GRIK gene family for susceptibility to schizophrenia [15][16][17][18].Recently, we identified four ultra-rare truncating mutations, including two frameshift deletion mutations (GRIK1 p.Phe24fs and GRIK1 p.Thr882fs ) and two nonsense mutations (GRIK2 p.Arg300Ter and GRIK4 p.Gln342Ter ), in four unrelated patients with schizophrenia [19].Taken together, rare pathologic mutations of the genes encoding kainate receptor protein alter biological processes of synaptic function in patients with schizophrenia.
Innovative CRISPR-based approaches have been used for studying the molecular mechanisms of schizophrenia in cellular models [20].Functional genomic studies using cell models carrying the deleterious kainate receptor gene mutations are necessary to understand the role of kainate-receptor-interacting genes and how they contribute to the etiology of schizophrenia.To explore the novel mechanism underlying the effect of kainate receptors, we used the CRISPR/Cas9 genome editing system to create the isogenic kainatereceptor-gene-deficiency SH-SY5Y cell lines.We conducted RNA sequencing (RNA-seq) to determine the differentially expressed genes (DEGs) in these isogenic edited cell lines.Furthermore, we induced these edited SH-SY5Y cells into the differentiated forms and compared the morphology of edited differentiated cells with the wild type.

Results
2.1.Generation of the Isogenic Kainate-Receptor-Gene-Knockout (KO) SH-SY5Y Cell Lines with CRISPR/Cas9 Editing Isogenic GRIK1-KO, GRIK2-KO, and GRIK4-KO cell lines were generated with CRISPR/Cas9 genome editing from SH-SY5Y cell lines.The gRNAs were designed to target the unique sequences in the GRIK1, GRIK2, and GRIK4 genes (Figure 1A-C, Supplementary Table S1) according to the CHOPCHOP online design website (http://chopchop.cbu.uib.no/; last accessed on 4 December 2023).The potential off-target sites for these gRNAs are listed in Supplementary Table S1.Considering the off-target effects of the CRISPR/Cas9, we carried out a confirmation with Sanger sequencing and revealed null off-target results in each edited cell line (Supplementary Figure S1).The pCas-guide vector containing target gRNA was transfected to SH-SY5Y cells.After single-cell isolation, five mutant cell lines (GRIK1 p.L25Pfs*?/WT , GRIK1 p.L25Pfs*?/p.L25Pfs*?, GRIK2 p.L301Ffs*?/WT , GRIK2 p.L301Ffs*?/p.L301Ffs*?, and GRIK4 p.H343Afs*?/p.H343Afs*?) were found.The edited cell lines were confirmed with Sanger sequencing (Figure 1A-C) and RT-qPCR (Figure 1D).An analysis of highly variable STR markers can be used in authenticating human cell lines.Here, the edited cell lines were authenticated by analyzing the 16 STR markers using the AmpFLSTR™ Identifiler™ Plus PCR Amplification Kit, and the STR allelic profile of each edited cell line is shown in Supplementary Figure S2.Based on 16 STR loci comparisons, we found 16 STR loci among edited cell lines, and SH-SY5Y wild-type (WT) cells were concordant.

Confirmation of Schizophrenia-Associated Genes in GRIK1-KO, GRIK2-KO, and GRIK4-KO SH-SY5Y Cell Lines with RT-qPCR
For RNA-seq data verification, six schizophrenia-associated genes (ARC, GRIA2, GRIA4, GABRB3, GRM8, and KCNJ3) in the synaptic membrane pathway were selected for verification with biological replicated cells.We compared the mRNA expression levels of these six genes in biologically replicated cells with an RT-qPCR assay, and the fold changes in these gene expressions between edited cells and WT cells are shown in Figure 3.
24, 2, FOR PEER REVIEW of these six genes in biologically replicated cells with an RT-qPCR assay, and t changes in these gene expressions between edited cells and WT cells are shown in 3.

Figure 3. RT-qPCR assay showing the expression of six schizophrenia-associated gene GRIA2, GRIA4, GABRB3, GRM8, and KCNJ3) in edited SH-SY5Y cells (GRIK1-KO, GRIK2-K GRIK4-KO) and WT cells. The GAPDH gene was used as the endogenous gene for norma
The data are expressed as fold change to WT ± SD (* p < 0.05, n = 6).

Cell Morphology of GRIK1-KO, GRIK2-KO, and GRIK4-KO SH-SY5Y Differentiate Lines
Studies demonstrate that glutamate receptors could regulate actin-based plast dendritic spines [21,22].Here, we aimed to determine, with an assay, whether glutamate receptors regulate the actin-based cytoskeleton.Rhodamine-conjugate loidin was used to detect the F-actin cytoskeleton in differentiated edited cells.duced GRIK1-KO, GRIK2-KO, and GRIK4-KO SH-SY5Y cells into the differe forms by treating RA and BDNF sequentially.Images of phalloidin-labeled cells w lected, and the different morphologies of GRIK1-KO, GRIK2-KO, and GRIK4-K SY5Y differentiated cell lines compared with the wild type are shown in Figure

Cell Morphology of GRIK1-KO, GRIK2-KO, and GRIK4-KO SH-SY5Y Differentiated Cell Lines
Studies demonstrate that glutamate receptors could regulate actin-based plasticity in dendritic spines [21,22].Here, we aimed to determine, with an assay, whether kainate glutamate receptors regulate the actin-based cytoskeleton.Rhodamine-conjugated phalloidin was used to detect the F-actin cytoskeleton in differentiated edited cells.We induced GRIK1-KO, GRIK2-KO, and GRIK4-KO SH-SY5Y cells into the differentiated forms by treating RA and BDNF sequentially.Images of phalloidin-labeled cells were collected, and the different morphologies of GRIK1-KO, GRIK2-KO, and GRIK4-KO SH-SY5Y differentiated cell lines compared with the wild type are shown in Figure 4.

Synaptic Membrane and Schizophrenia-Associated Genes in GRIK1-KO, GRIK2-KO, and GRIK4-KO SH-SY5Y Cells
Given that the GO enrichment analysis demonstrated that the genetic deletion of the GRIK1, GRIK2, or GRIK4 gene jointly disturbed a signal pathway, the synaptic membrane (GO:0097060), we presumed that kainate-receptor-regulated synaptic membrane genes could be involved in synaptic dysfunction in schizophrenia pathogenesis.The synaptic membrane is a specialized area on either the presynaptic or postsynaptic side of a synapse, the space between a nerve fiber of one neuron and another, a muscle fiber, or a glial cell

Synaptic Membrane and Schizophrenia-Associated Genes in GRIK1-KO, GRIK2-KO, and GRIK4-KO SH-SY5Y Cells
Given that the GO enrichment analysis demonstrated that the genetic deletion of the GRIK1, GRIK2, or GRIK4 gene jointly disturbed a signal pathway, the synaptic membrane (GO:0097060), we presumed that kainate-receptor-regulated synaptic membrane genes could be involved in synaptic dysfunction in schizophrenia pathogenesis.The synaptic membrane is a specialized area on either the presynaptic or postsynaptic side of a synapse, the space between a nerve fiber of one neuron and another, a muscle fiber, or a glial cell [31].The present study identified multiple DEGs associated with the synaptic membrane in GRIK1-KO, GRIK2-KO, and GRIK4-KO SH-SY5Y cells.Interestingly, multiple identified DEGs associated with the synaptic membrane, such as ARC, GRIA2, GRIA4, GABRB3, GRM8, and KCNJ3, have been implicated in the pathophysiology of schizophrenia.ARC dysregulation contributes to various neurological and cognitive disorders and schizophrenia [14,[32][33][34].Recently, we generated an ARC-KO HEK293 cell line and conducted a transcriptomic and proteomic analysis to identify the DEGs related to the synaptic membrane [35].Previous studies of schizophrenia have demonstrated the presence of a different AMPA receptor expression in the thalamus [36].We recently found that rare pathogenic mutations of the GRIA1, GRIA2, and GRIA4 genes might contribute to the pathogenesis of schizophrenia in some subjects [37].The down-regulation of GABRB3 may contribute to the pathophysiology and clinical manifestations of schizophrenia through altered oscillation synchronization in the superior temporal gyrus [38].An association analysis revealed the genetic association of GRM8 and KCNJ3 with schizophrenia in the Han Chinese population [39,40].
Studies indicate that kainate receptors are critical mediators of the pre-and postsynaptic actions of neurotransmitters, although the mechanisms underlying such effects remain unclear [41].Several synapse-associated proteins have been identified as interacting components for the kainate receptors [42].Kainate receptors have been linked to a number of neuropsychiatric disorders, such as schizophrenia, bipolar disorder, mental retardation, and autism [43,44].Identifying proteins interacting with kainate receptors is essential to unravel kainate-receptor-mediated signaling in neuropsychic disorders.According to our RNA-seq with an enrichment analysis of gene-disease associations, the DEGs in these three edited cell lines were associated with several neuropsychiatric disorders, especially schizophrenia.Thus, we hypothesize that kainate receptor deficiency may destroy the formation and functional integrity of synapse-associated components for the neuronal processes that are deficient in individuals with schizophrenia.These findings suggest that kainate-receptor-regulated synaptic membrane genes could possibly be implicated in synaptic dysfunction in the pathophysiology of schizophrenia.

F-Actin Abnormalities in GRIK1-KO, GRIK2-KO, and GRIK4-KO SH-SY5Y Cells
Evidence demonstrated that the schizophrenia brain reduced dendritic spine density and altered synaptic plasticity [45,46].Mounting evidence suggests that actin remodeling is critical to synaptogenesis, synaptic plasticity, and the development of neurites in developing neurons [47,48].For example, dynamic actin filaments formed dendritic spines during development and their structural plasticity at mature synapses [48].Extensive studies describe evidence for regulatory mechanisms of actin dynamics in dendritic spines [47,49].Bhambhvani and colleagues identified a reduced protein expression of F-actin in the anterior cingulate cortex of elderly patients with schizophrenia, consisting of reduced dendritic spine density and altered synaptic plasticity in schizophrenia [50].According to the reported RNA-seq data, Kimoto and colleagues found that levels of actin-and mitochondrial-oxidative-phosphorylationrelated transcripts were significantly altered in subjects with schizophrenia [51].The above evidence consisted of the altered dendritic spine morphology in schizophrenia, which may be linked to abnormalities in the regulation of actin cytoskeletal dynamics [52].Notably, a study suggests how glutamate receptors regulate actin-based plasticity in dendritic spines [21].Another study suggests that the glutamate receptor agonist kainate induces the rearrangement of actin filaments in ameboid microglia [22].Our previous genetic study demonstrated that rare pathologic mutations of the GRIK gene family play a potential role in conferring susceptibility to schizophrenia [19].In the present study, the fluorescent images demonstrated less F-actin expressed in differentiated SH-SY5Y cells with GRIK1, GRIK2, or GRIK4 deficiency than in differentiated WT cells.Taken together, kainate glutamate receptors involved in the actin cytoskeleton may be linked to the pathophysiology of schizophrenia.Our findings suggest that kainate glutamate receptors could possibly regulate the actin-based cytoskeleton, which is essential for maintaining dendritic spine morphology and density in the pathophysiology of schizophrenia.

Materials and Methods
4.1.CRISPR/Cas9-Directed Genome Editing of the Isogenic SH-SY5Y Cell Lines and a Single Edited Cell Isolation CRISPR/Cas9-directed genome editing and single-edited cell isolation were performed following previously described methods [35].In brief, the pCas-Guide vector carrying the guide RNA (gRNA) guide sequence was generated using the method described in the manufacturer's protocols (Origene, Rockville, MD, USA).The SH-SY5Y cells (Sigma catalog no.94030304) were transfected with a pCas-Guide vector carrying gRNA using a Neon electroporation transfection system (Invitrogen, Carlsbad, CA, USA).One week after transfection, genomic DNA (gDNA) of harvested cells purified with the Gentra Genomic DNA Purification kit (QIAGEN, Germantown, MD, USA) was subjected to genomic PCR and a T7 endonuclease assay.The single-edited cell on a QIAscout array was isolated with the QIAscout device according to the manufacturer's protocols (QIAGEN), and the isolated cells were processed for further cultivation and clonal expansion.The gDNA of the clonally expanded cells, extracted using the PDQeX Nucleic Acid Extractor (MicroGEM, Southampton, UK), was used for PCR reactions and fluorescence-based Sanger sequencing to find correctly edited cells.

Human Cell Line Identification
The gDNA was purified from the isogenic cell line using a DNeasy Blood & Tissue Kit according to the manufacturer's protocols (QIAGEN).The gDNA was amplified using an AmpFLSTR™ Identifiler™ Plus PCR Amplification Kit (Thermo Fisher Scientific Inc., Waltham, MA, USA), and the short tandem repeat (STR) and PCR products were analyzed with DNA Analyzer 3730XL (ThermoFisher Scientific Inc.).The calling of STR alleles by aligning unknown fragments with a ladder of STR fragments of known allele sizes was analyzed with GeneMapper Software v4.0 (ThermoFisher Scientific Inc.).Total RNA preparation, RNA-seq, and DEG identification were performed following previously described methods [35].A GO enrichment analysis of DEGs was conducted using clusterProfiler (v3.10.1).DEGs associated with KEGG pathways were annotated according to the KEGG database [53].An enrichment analysis of gene-disease associations was performed using the DisGeNET database [54].RT-qPCR assays were performed using the comparative ∆∆Ct method to validate the differential gene expression [55].The expression levels of GRIK1, GRIK2, GRIK3, GRIK4, GRIK5, GRIA2, GRIA4, GABRB3, GRM8, and KCNJ3 were assayed using the QuantStudio 3 real-time PCR system in combination with continuous SYBR Green detection (ThermoFisher Scientific Inc.).The primer sequences for GRIK1, GRIK2, GRIK3, GRIK4, GRIK5, GRIA2, GRIA4, GABRB3, GRM8, and KCNJ3 are listed in Supplementary Table S9.The target gene ARC (Hs01045540_g1, FAM™ dyelabeled TaqMan™ MGB probe) and the endogenous gene GAPDH (Hs02786624_g1, VIC™ dye-labeled TaqMan™ MGB probe) were measured using TaqMan™ gene expression assays according to the manufacturer's protocol (ThermoFisher Scientific Inc.).All tests were performed six times.Statistically significant differences between edited and wild-type (WT) cells were those with a p value < 0.05.

Differentiation of the SH-SY5Y Cells
The SH-SY5Y cells were seeded at an initial density of 10 4 cells/cm 2 in culture dishes previously coated with 0.05 mg/mL of collagen (Collaborative Biomedical Products, Bedford, MA, USA).Retinoic acid (RA, Sigma-Aldrich, St. Louis, MO, USA) was added the day after plating at a final concentration of 10 uM in DMEM with 10% fetal calf serum.After five days in the presence of RA, cells were washed three times with DMEM and incubated with 50 ng/mL of BDNF in DMEM without fetal calf serum for seven days.

Immunocytochemistry
Cultured cells were fixed in 4% paraformaldehyde in PBS (pH 7.4) for 20 min at room temperature, washed three times with PBS containing 0.1% Triton X-100, and blocked for 40 min with PBS containing 1% bovine serum albumin and 0.1% Triton X-100.The samples were incubated with a primary antibody (anti-MAP) diluted at 1:250 in a blocking buffer overnight at 4 • C, washed with PBST three times, and then incubated with secondary antibodies conjugated with fluorescence diluted at 1:500 in a blocking buffer for 1 h at room temperature.After that, the samples were washed with PBST three times, the cell nucleus was labeled with DAPI, and the rhodamine phalloidin (R415, Invitrogen) detected F-actin.Images were acquired with a fluorescence microscope, Axio Vert.A1 (Zeiss, Jena, Germany), and analyzed with ZEN 2 software (Zeiss).

Conclusions
We identified several kainate-receptor-regulated genes involved in multiple signal pathways, especially regarding the synaptic-membrane-associated genes and neuropsychiatric disorders, especially schizophrenia.The association between kainate receptors and DEGs we identified is a fascinating but enigmatic protein that warrants further study.Therefore, the transcriptomic profiles for GRIK1-KO, GRIK2-KO, and GRIK4-KO SH-SY5Y cells contribute to emerging evidence for the novel mechanisms underlying the effect of kainate receptors and molecular pathways of the pathophysiology of schizophrenia.In addition, we suggest that kainate-receptor-mediated F-actin remodeling may be a candidate mechanism underlying schizophrenia.

Figure 4 .
Figure 4. Representative phase-contrast and fluorescent images of differentiated SH-SY5Y cells harboring GRIK1, GRIK2, or GRIK4 deficiency.(A) Phase-contrast.(B) Fluorescent images.Cells are labeled green with the neuronal marker MAP2, red with the cytoskeleton F-actin, and blue with DAPI.

Figure 4 .
Figure 4. Representative phase-contrast and fluorescent images of differentiated SH-SY5Y cells harboring GRIK1, GRIK2, or GRIK4 deficiency.(A) Phase-contrast.(B) Fluorescent images.Cells are labeled green with the neuronal marker MAP2, red with the cytoskeleton F-actin, and blue with DAPI.

Table 2 .
Enrichment for the neuropsychiatric-disorder-associated genes among DEGs based on the DisGeNET database.