Computational Analysis of CRISPR-Cas9 Mediated Therapeutic Targeting in Alzheimer’s Disease

INTRODUCTION

The progressive neurodegenerative disorder Alzheimer’s disease (AD) leads to permanent destruction of memory functions, cognitive abilities, and behavioral patterns.^[1,2] The disease serves as the primary reason for dementia, which affects 60–80% of dementia patients and generates major healthcare costs worldwide.^[2] The core pathological elements of AD consist of amyloid-beta (Aβ) plaques located outside cells and tau protein tangles found inside cells, together with synaptic damage and ongoing brain inflammation.^[3] The available medical treatments for Alzheimer's disease help manage symptoms, but they do not address the core biological processes that lead to disease progression.^[4] The last twenty years of genomic research have identified critical genetic elements that determine Alzheimer's disease susceptibility.^[5]The combination of APP and PSEN1 and PSEN2 gene mutations results in early-onset familial Alzheimer's disease, but APOE ε4 allele variants increase the risk of developing late-ons et alzheimer's disease.^[2-5] Multiple risk loci have been identified through Genome-Wide Association Studies (GWAS) and meta-analyses, which include CLU, BIN1, PICALM, SORL1, and CR1 genes that influence lipid metabolism and synaptic plasticity, and immune system regulation.^[2-6] The discovery of these genetic elements shows potential for developing gene-specific treatments which could alter disease progression through specific molecular pathways.

The clustered regularly interspaced short palindromic repeats–CRISPR-associated protein 9 (CRISPR-Cas9) genome editing system has revolutionized DNA sequence modification in living cells because of its precise and efficient operation.^[7-9] The technology demonstrates strong potential to treat genetic disorders, which include neurodegenerative diseases. The CRISPR system allows scientists to eliminate disease-triggering genetic mutations, repair genetic mutations, and manage gene expression that affects AD.^[10]The success of CRISPR-based AD therapy depends on developing suitable guide ribonucleic acids (gRNAs) and performing. thorough off-target assessments and understanding gene interactions and pathway responses.^[11]The development of CRISPR-based therapeutic methods for Alzheimer's disease treatment needs computational biology tools to evaluate and enhance before laboratory testing.^[12]Researchers use gene expression data together with pathway analysis and network modelling to identify vital targets, predict treatment outcomes, and simulate therapeutic responses.^[13] The development of gRNA designs becomes possible through tools like CHOPCHOP and CRISPOR, CasOFFinder, and Cytoscape, which enable researchers to predict gene-editing effects on cellular pathways using Boolean and dynamic models.^[14] The treatment of Alzheimer's disease requires genetic analysis because the condition involves multiple genetic factors, which need multiple gene targets for successful therapy.

The research conducts a complete computational assessment of CRISPR-Cas9 genome editing for AD treatment. The researchers performed database searches and literature reviews to identify candidate genes before using AD brain tissue expression data to create gRNAs that demonstrated high specificity and minimal off-target activity. The researchers performed pathway enrichment analysis and gene regulatory-Cas9 editing of APP, PSEN1, APOE, and TNF genes to create enhanced network simulations, which demonstrated how gene editing would impact biological systems. The research shows that combining CRISPR effects on amyloid processing, inflammation, and lipid metabolism could result in superior gene therapies for Alzheimer’s disease. This study does not represent clinical gene editing but rather provides an in-silico prioritization framework.

MATERIAL AND METHODS

RESULTS

Differential gene expression analysis in Alzheimer’s brain tissue

The GSE33000 dataset (n = 624 prefrontal cortex samples) showed that AD patients had widespread changes in their gene expression patterns when compared to healthy controls. The researchers found 376 genes that showed significant differential expression (adjusted p-value < 0.05; |log₂FC| > 1) between AD brains and control brains. The researchers found that APP and PSEN1, TNF, and CLU showed increased expression in AD brains, but SORL1 and BIN1 showed decreased expression. The volcano plot displayed the most important genes that showed significant changes. The top 20 differentially expressed genes showed different expression patterns between AD patients and control subjects through heatmap clustering [Figure 1 and Table 2].

Close

Figure 1:

Volcano plot of differentially expressed genes in Alzheimer’s disease (AD). Highlights upregulated and downregulated genes in AD brain tissue (GSE33000). Significant genes like APP, PSEN1, TNF, and CLU are clearly visualized.

Export to PPT

Table 2: Differential expression summary (Top 10)

Rank	Gene	log2 FC	Adjusted p-value
1	APP	1.73	2.4 × 10-6
2	PSEN1	1.51	3.1 × 10-5
3	TNF	1.44	7.8 × 10-5
4	CLU	1.32	1.2 × 10-4
5	BIN1	–1.28	2.3 × 10-4
6	CR1	1.22	4.1 × 10-4
7	SORL1	–1.18	6.7 × 10-4
8	PICALM	1.16	9.9 × 10-4
9	APOE	1.05	1.5 × 10-3
10	PSEN2	1.02	2.1 × 10-3

Adjusted p-value < 0.05. FC: Fold change

Guide RNA design and off-target analysis

The researchers used CHOPCHOP and CRISPOR tools to create gRNAs for APP, PSEN1, TNF, and APOE. The chosen gRNAs demonstrated high on-target performance (≥86) and low off-target potential. The CRISPOR score for the APP gRNA (5’ GTGCTGAGCGTCTTCAGACG 3’) reached 89.5 while showing no predicted coding off-target sites. The researchers successfully designed allele-specific correction gRNAs, which could transform APOE4 into APOE3 while minimizing potential off-target effects, thus proving the practicality of precision editing [Figure 2 and Table 3].

Close

Figure 2:

CRISPR gRNA efficiency and off-target score, presented as a heatmap. It summarizes on-target scores (gRNA efficacy), off-target counts (specificity), and gRNA type (knockout or correction, noted at the bottom) gRNA: Guide ribonucleic acid

Export to PPT

Table 3: gRNA designs and Off-target scores

Gene	gRNA (5’→3’)	On target score	Predicted high risk off targets
APP	GTGCTGAGCGTCTTCAGACG	89.5	0
PSEN1	CGTACGAGACGTTTGTCTGA	91.2	1 (intronic)
APOE4→3	GACCTGCGGAGGTTGGACAA	86.7	0
TNF	CAGGTTCTCTAGATGGCACA	88.0	2 (non coding)

gRNA: Guide ribonucleic acid

Functional enrichment and pathway analysis

The GO and KEGG pathway enrichment analysis of differentially expressed and CRISPR-targeted genes demonstrated significant enrichment of multiple AD related pathways (FDR < 0.01). The analysis identified Aβ formation and tau-protein kinase activity, cholesterol transport, endocytosis, and pro-inflammatory signalling as the most significant pathways. The Enrich analysis demonstrated that the Nuclear Factor kappa B (NF-κB) -TNF/NF-κB signalling pathway and synaptic vesicle cycling showed the most significant enrichment in neuroinflammatory processes. The results validate APP and PSEN1, APOE, and TNF as therapeutic targets for AD treatment [Figure 3, 4 and Table 4].

Close

Figure 3:

Heatmap of top 20 differentially expressed genes. Displays normalized expression (Z-score) across 10 AD and control samples. Clusters gene expression into disease-specific signatures. Genes such as BIN1, SORL1, and APOE show distinct expression trends. AD: Alzheimer’s disease

Export to PPT

Close

Figure 4:

Bar plot of top enriched pathways (–log₁₀ FDR). Shows top KEGG/GO pathways enriched among DEGs and CRISPR targets. Includes: Amyloidogenesis, tau signaling, neuroinflammation, cholesterol metabolism, endocytosis, and synaptic signaling. FDR: False discovery rate.

Export to PPT

Table 4: Pathway enrichment (Top 6)

Pathway	–log10 FDR	Key genes
Tau pathway	14.8	MAPT, PSEN1, PSEN2
Cholesterol metabolism	14.5	APOE, CLU, SORL1
Amyloidogenesis	12.7	APP, BACE1, PSEN1
Neuroinflammation	10.4	TNF, CR1
Synaptic signalling	9.1	BIN1, PICALM
Endocytosis	6.1	PICALM, SORCS1

FDR: False discovery rate

Network modelling and visualization

The Cytoscape platform enabled researchers to construct a gene regulatory network comprising 15 nodes and 18 directed edges, which represented known relationships between selected genes and their associated pathways. The network revealed that APP, PSEN1, APOE, and TNF function as central hub genes, controlling Aβ production, tau phosphorylation, cholesterol regulation, neuroinflammation, and synaptic dysfunction. The network analysis through node centrality demonstrated that APP and APOE function as primary regulators which makes them suitable targets for therapeutic intervention [Figure 5].

Close

Figure 5:

Gene regulatory network (Cytoscape-style). Nodes represent genes and pathways; edges represent regulatory interactions. Central regulators: APP, APOE, PSEN1, TNF. Outputs: Aβ tau, cholesterol, synapse, inflammation.

Export to PPT

Boolean simulation of CRISPR editing scenarios

Boolean models simulated how CRISPR interventions would affect five Alzheimer’s disease-related phenotypic outcomes. The five pathways operated in an active state when the system operated under normal conditions. The removal of APP genes blocked both amyloid and tau protein production, while TNF knockout specifically reduced neuroinflammatory responses. The APOE4 to APOE3 correction improved cholesterol handling and reduced inflammation to some extent. The results indicate that using CRISPR to edit multiple genes simultaneously produces better therapeutic outcomes than editing a single gene at a time [Table 5 and Figure 6].

Table 5: Summary of key findings

Intervention	Aβ	Tau	Inflamm- ation	Chole- sterol	Synapse
Wild-Type	ON	ON	ON	ON	ON
APP knockout	OFF	OFF	ON	ON	ON
TNF knockout	ON	ON	OFF	ON	ON
APOE4→3 Correction	ON	ON	OFF	OFF	ON

Close

Figure 6:

Boolean simulation heatmap of CRISPR-Cas9 interventions. Rows: Editing scenarios (WT, APP KO, TNF KO, APOE4→3 correction). Columns: Functional outputs (Aβ tau, neuroinflammation, cholesterol, synapse). The binary matrix (ON/OFF) clearly shows the effects of each intervention. AD: Alzheimer’s disease

Export to PPT

The results support the hypothesis that CRISPR-Cas9 gene editing can be a feasible and targeted therapeutic approach for Alzheimer’s disease. The network-based simulation confirms the potential for personalized, multi-gene editing strategies tailored to patient-specific risk profiles.

DISCUSSION

The neurodegenerative disorder AD continues to spread as an untreatable condition that requires new therapeutic methods that focus on genetic and molecular disease mechanisms.^[2-15] The research used CRISPR-Cas9 gene editing to study AD by performing a complete computational analysis, which combined gene expression data with guide RNA development and network modelling and Boolean simulation methods.^[16] The research shows that CRISPRCas9 gene editing of APP and PSEN1, APOE, and TNF genes can effectively modify essential AD-related biological pathways, which include Aβ formation, tau phosphorylation, inflammation, and lipid metabolism disturbances. Although CRISPR-Cas9 has shown remarkable potential for therapeutic genome editing, its clinical utility in AD remains largely experimental.^[15] Current applications are mainly limited to in-vitro neuronal systems and animal models due to major translational challenges, including safe delivery across the blood–brain barrier, off-target genome alterations, long-term safety considerations, and ethical concerns regarding permanent genetic modification. Therefore, the present study should be considered a computational framework for target prioritization and mechanistic exploration rather than direct evidence of clinical treatment feasibility.

The GSE33000 dataset showed that AD patients exhibited elevated expression of APP and PSEN1, and TNF in their prefrontal cortex, which matches previous research about these genes' involvement in amyloid processing and immune system activation.^[17,18] The decreased expression of SORL1 and BIN1 genes confirms their essential roles in endosomal trafficking and synaptic function.^[19,20] The observed expression patterns confirm our gene selection approach and demonstrate why researchers should focus on both established (APP) and emerging (TNF) disease-causing genes in AD research.^[18] In this study, we created high-performance gRNAs with minimal off-target effects for each candidate gene through modern CRISPR technology. The in-silico results showed that APOE4 to APOE3 allele-specific correction proved effective because it reduced cholesterol problems and neuroinflammatory responses.^[2-21] The treatment method uses precision medicine to benefit people who have the APOE4 allele because it targets their specific genetic risk. Mechanistically, CRISPR-Cas9 editing relies on guide RNA sequences that direct the Cas9 endonuclease to specific genomic loci, where targeted DNA cleavage occurs. Gene disruption (e.g., APP or TNF knockout) is typically achieved through non-homologous end joining (NHEJ), which introduces insertions or deletions leading to functional gene silencing.^[22] In contrast, allele-specific correction strategies, such as APOE4 to APOE3 conversion, aim to exploit homology-directed repair or next-generation precision editing approaches, including base or prime editing.

The targeted genes show extensive connections through biological pathways that control AD development according to pathway enrichment and gene network analysis.^[22] The biological processes of lipid transport (APOE, CLU) and amyloidogenesis (APP, PSEN1) and synaptic vesicle trafficking (BIN1, PICALM) and inflammation (TNF, CR1) are all connected through these genes.^[23] The regulatory network design showed how APP and APOE function as central regulatory elements in the system. Boolean simulations showed that individual gene knockout operations produce specific effects on disease progression, where APP knockout decreased both Aβ and Tau levels, and TNF knockout blocked neuroinflammatory reactions.^[24] The research indicates that complete therapeutic success requires multiple gene editing targets.

Despite the promising computational predictions presented in this study, several real-world barriers currently limit the clinical translation of CRISPR-based therapies for Alzheimer’s disease. One of the major challenges is the efficient delivery of genome-editing systems across the blood-brain barrier (BBB), which restricts the penetration of large biomolecules into the central nervous system.^[11-25]

Viral vectors such as adeno-associated viruses (AAVs), lipid nanoparticles, and exosome-based delivery platforms are under active investigation, yet each presents limitations related to immunogenicity, payload size, tissue specificity, and long-term safety.^[26] Additional concerns include potential off-target editing, immune responses to Cas proteins, and ethical considerations surrounding permanent genomic modification in neuronal tissues. Therefore, while the present study provides a computational framework for identifying promising targets, substantial technological and safety advancements are required before clinical application becomes feasible. Prospective human studies involving CRISPR-based genome editing for AD are not yet feasible at a clinical scale. A realistic translational pathway would involve sequential validation beginning with computational modelling, followed by experiments in induced pluripotent stem cell (iPSC)-derived neurons and animal models, before consideration of early-phase clinical trials. This staged approach is essential to establish safety, delivery efficiency, and therapeutic efficacy before human application.

The promising in silico results need to be evaluated against multiple factors that limit their application. The binary models used in our simulations fail to represent complete gene expression levels and post-translational modifications. The central nervous system delivery of CRISPR requires additional research to address both immune response risks and ethical concerns about germline and somatic editing before medical applications become possible. The study provides computational methods for designing CRISPR treatments in neurodegenerative diseases while creating opportunities for laboratory testing in suitable cell and animal models. The research demonstrates that CRISPR-Cas9 gene editing technology can effectively targ et alzheimer's disease-causing pathways. The combination of transcriptomic data analysis with guide RNA optimization and systems-level modelling demonstrates how precision gene therapies can target multiple disease-causing genes effectively. Researchers need to develop combination delivery methods and conduct extended safety and effectiveness tests while confirming their results through human iPSC-derived neurons and AD mouse models. The Boolean modelling framework employed in this study provides a simplified systems-level representation of pathway activity using binary ON/OFF states. While this approach facilitates conceptual understanding and hypothesis generation, it does not capture quantitative gene expression dynamics, graded signalling responses, or post-translational regulatory mechanisms. Future studies integrating dynamic or probabilistic modelling may provide higher biological fidelity.

CONCLUSION

The research through computational methods shows that CRISPR Cas9 gene editing technology holds promise for treating AD through disease modification. Our research combines multiple methods to demonstrate that the four genes APP, PSEN1, APOE, and TNF represent critical targets because they control amyloidogenesis and tau pathology, lipid regulation, and neuroinflammatory responses. The designed gRNAs show high on-target performance (≥ 86) while minimizing predicted coding off-target effects, which makes precise in vivo editing possible. Network and pathway analysis show that single-gene modifications create specific therapeutic effects, but complete disease treatment requires multiple gene editing approaches. The Boolean model predicts that APP knockout will stop Aβ and tau production, while TNF knockout will reduce inflammation, and APOE4 to APOE3 conversion will restore cholesterol balance and reduce inflammation. In summary, this computational study establishes a systematic in-silico pipeline for evaluating CRISPR-Cas9 strategies targeting Alzheimer’s disease-associated genes. The findings suggest that combined editing of APP, PSEN1, APOE, and TNF may theoretically influence amyloid pathology, neuroinflammation, and lipid dysregulation. Nevertheless, these observations represent predictive computational outcomes and must be validated through experimental and preclinical studies before clinical application. The framework presented here aims to support future translational research toward safe and effective precision gene therapies for neurodegenerative diseases.