The population raising companion animals is steadily increasing, with 5.52 million households, accounting for 25.7% of all households in South Korea, reported raising companion animals as of the end of 2022 (Hwang and Lee, 2023). The companion animal-related industry is also continuously growing, with the market size in South Korea reaching 2.92 trillion KRW in 2021 and projected to expand to 4.12 trillion KRW by 2028 (KREI, 2024). With increasing interest in the health and welfare of companion animals, their average lifespan is rising, and it is estimated that aging is already underway among companion dogs in South Korea (Jung, 2018).

Elderly dogs are particularly vulnerable to various diseases because of a decline in physiological function, leading to the frequent onset and progression of medical conditions. In particular, the eyes are one of the primary organs most frequently affected by diseases in elderly dogs, with glaucoma being recognized as a representative condition (Hwang and Son, 2021). Glaucoma is a severe condition that damages retinal neurons, ultimately leading to blindness. The primary cause of glaucoma is increased intraocular pressure (IOP), and the disease is classified as congenital, primary, or secondary glaucoma based on its manifestation. Primary angle-closure glaucoma is the most common form of glaucoma in dogs (Park and Jeong, 2023). Glaucoma occurs in all dog breeds; however, the incidence rate varies depending on the breed. Breeds reported to be particularly susceptible to glaucoma include the American Cocker Spaniel, Basset Hound, Shiba Inu, and Shih Tzu (Gelatt and MacKay, 2004; Kato et al., 2006; Park et al., 2012; Yun et al., 2022). The differences in incidence rates suggest a potential genetic predisposition, prompting active research efforts to identify genes associated with glaucoma development. However, most studies have only proposed the locations of candidate genes predicted to be associated with glaucoma. In contrast, the precise pathogenic mechanisms and causative genes have yet to be identified. Accordingly, it is necessary to identify candidate genes associated with glaucoma, a major disease in elderly dogs, and conduct studies to predict the risk of disease development using these genetic markers. Therefore, this study aims to utilize the Axiom^TM Canine HD Array to identify candidate genes and single nucleotide polymorphisms (SNPs) associated with primary angle-closure glaucoma in companion dogs and to establish foundational data for developing SNP markers capable of predicting the risk of disease onset.

Animals

DNA samples used for SNP marker selection were obtained from oral epithelial cells and blood samples of 95 dogs comprising 26 breeds, including mixed breeds raised domestically in South Korea. To validate the sequences of the selected SNP markers, oral epithelial cells, and blood samples were collected from 95 dogs of 10 breeds identified in previous studies, including those by Gelatt and MacKay (2004), Kato et al. (2006), Park et al. (2012), and Yun et al. (2022), as breeds reported to develop glaucoma, either domestically or internationally. The sample information is presented in Table 1. Genomic DNA for analysis was extracted using the AccuPrep^® Genomic DNA Extraction Kit, following the protocol recommended by the manufacturer.

Table 1. Samples of domestic dog breeds used for study.

Method	Breed	No. of sample	Breed	No. of sample
SNP chip analysis	Maltese	20	Cocker Spaniel	1
	Mixed dog	20	Italian Greyhound	1
	Poodle	10	Jindo Dog	1
	Shih Tzu	6	Long haired Dachshund	1
	Bichon Frise	5	Miniature Pinscher	1
	Pomeranian	5	Miniature Schnauzer	1
	Yorkshire Terrier	4	Pointer	1
	Chihuahua	3	Schnauzer	1
	French Bulldog	3	Siberian Husky	1
	Dachshund	2	Spitz	1
	Golden Retriever	2	Standard Poodle	1
	Border Collie	1	Welsh Corgi	1
	Chow Chow	1	Wheaten Terrier	1
Sanger sequencing	Maltese	13	Yorkshire Terrier	11
	Shih Tzu	13	Jindo Dog	9
	Pomeranian	13	Cocker Spaniel	5
	Poodle	12	Shiba Inu	4
	Golden Retriever	12	Miniature Pinscher	3

SNP data analysis and marker selection

The extracted DNA was analyzed for SNP genotypes using the Axiom^TM Canine HD Array (Applied Biosystems^TM, USA), and a total of 730,754 SNPs were obtained through the Axiom Analysis Suite Software (Applied Biosystems^TM, USA). Quality control (QC) was performed using the PLINK 1.9 program (Purcell et al., 2007; Chang et al., 2015) to ensure accurate analysis. The QC criteria were as follows: sample call rate < 90%, SNP call rate < 90%, and Hardy-Weinberg Equilibrium (HWE) p-value < 1 × 10^-7. SNPs that did not meet these criteria were excluded from the analysis. In the study by Oliver et al. (2019), BICF2P928441, an SNP reported to be significantly associated with primary angle-closure glaucoma through genome-wide association studies (GWAS), was utilized in the present study. The information for BICF2P928441 matched the SNP chip genome reference data used in this study, and AX-167199290, which had the same chromosome number and position information, was considered the same SNP and used for analysis. Genes within 500 kb upstream and downstream of the SNP were investigated, and the AX-167199290 SNP was identified in the TNS1 gene. As a result, TNS1 was selected as a candidate gene, and SNP data from TNS1 were collected and analyzed. To analyze the genetic diversity of SNP data, minor allele frequency (MAF), observed heterozygosity (Ho), expected heterozygosity (He), and polymorphism information content (PIC) values for each SNP were calculated using the snpReady package in the R program. Based on the Canis lupus familiaris gene annotation (GTF) (version CanFam3.1.104) and SNP information (chromosome and position) provided by the Ensembl database (https://asia.ensembl.org/index.html), the structural details of each SNP, such as missense and intron variants, were annotated. The final SNP markers were selected based on the genetic diversity analysis and annotation results.

Identification and impact prediction of nsSNPs

To predict functional changes in proteins caused by non-synonymous SNPs (nsSNPs) and evaluate their deleteriousness, SIFT (Sorting Intolerant From Tolerant; http://sift-dna.org) and PolyPhen-2 (Polymorphism Phenotyping v2; http://genetics.bwh.harvard.edu/pph2/) were used. The amino acid sequences used for analysis were collected through the Ensembl database (https://asia.ensembl.org/index.html), and amino acid mutations were indicated according to the analysis format. To interpret the biological functions of candidate genes containing the selected nsSNPs, Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis and Gene Ontology (GO) analysis were conducted using the Database for Annotation, Visualization, and Integrated Discovery (DAVID; http://david.abcc.ncifcrf.gov). Through KEGG pathway analysis, pathways associated with candidate genes were identified, and GO analysis was performed to determine the relationships between the genes and Biological Processes (BP), Molecular Functions (MF), and Cellular Components (CC). To investigate three-dimensional structural variations in proteins caused by nsSNPs, analyses were conducted using ColabFold, an open-source software package based on AlphaFold2 (Mirdita et al., 2022). The amino acid sequence registered in the Ensembl (https://asia.ensembl.org/index.html) database was used as a query sequence for generating a reference protein structure, and the sequence in which an amino acid mutation due to nsSNP appeared was used as a query sequence for generating a mutation protein structure.

Validation of SNP markers through sanger sequencing

Reference sequences registered in the Ensembl database (https://asia.ensembl.org/index.html) were used to validate the sequences of selected SNP markers. Primers were designed using Primer3 (https://primer3.org/) while considering the size of the amplified products (Table 2). For sanger sequencing, PCR was performed using DNA extracted from 95 dogs. The PCR reaction mixture was prepared with a total volume of 15 µL, consisting of the following components: 1 µL of DNA, 0.5 µL of each primer, 1.2 µL of 10 mM dNTPs, 1.5 µL of PCR reaction 10X buffer (10 mM Tris-HCl, 50 mM KCl, 1.5 mM MgCl₂), 10.1 µL of distilled water (D.W), and 0.2 µL of Taq DNA polymerase (Hot start Taq polymerase, 2.5 U/µL). The PCR process began with a pre-denaturation step at 95℃ for 5 minutes, followed by 35 cycles consisting of 95℃ for 30 seconds, 59℃ for 30 seconds, and 72℃ for 1 min. The final extension step was performed at 65℃ for 15 min. Sanger sequencing analysis was conducted using the BigDye^® Terminator v3.1 Cycle Sequencing Kit (Applied Biosystems^TM, USA) and the ABI 3730xl Genetic Analyzer (Applied Biosystems^TM, USA) to verify the nucleotide sequences. The genotypes of the selected SNPs were confirmed using the ABI Sequencing Analysis Software (Applied Biosystems^TM, USA) and the Biological Sequence Alignment Editor (BioEdit ver. 7.2.5, USA).

Table 2. Primer information for sequencing SNP markers associated with primary angle closure glaucoma.

Primer name	Primer sequence (5’-3’)		Size (bp)	A.T (℃)
TNS1_Intron	F	ACCTGTGTCAACTTTGCAACT	669	59
	R	AGACGTGCTAGGGAAGTTTCA
TNS1_Exon	F	GCCTCCTTTAACCCCTGCA	758	59
	R	TCTCCATCCATCAGTCCCTG

Selection of SNP markers through genetic diversity analysis and annotation

From the 730,754 SNPs obtained through analysis using the Axiom^TM Canine HD Array (Applied Biosystems^TM, USA), SNPs that did not meet the QC criteria were removed, resulting in the identification of 65 SNPs located on the TNS1 gene. To confirm the genetic diversity of each SNP, MAF, He, Ho, and PIC were calculated (Table 3). The average values of MAF, He, Ho, and PIC were 0.24, 0.33, 0.27, and 0.27, respectively. The ranges were: MAF (0.04-0.48), He (0.07-0.50), Ho (0.05-0.48), and PIC (0.07-0.37). Candidate SNPs were selected based on high genetic diversity, with criteria of PIC ≥ 0.3, MAF ≥ 0.3, and He and Ho values above average. As a result, 41 SNPs were excluded, and 24 SNPs were selected (Table 3). The positional information of SNPs with high genetic diversity in the candidate gene (TNS1) was annotated to assess their structural impact (Table 4). The analysis identified 23 SNPs as intron variants and 1 SNP as a missense variant. Based on this, the nsSNP AX-167242351, which is expected to affect gene function, and the intron SNP AX-167199290, identified as associated in the study by Oliver et al. (2019), were selected as the final risk prediction markers for primary closed-angle glaucoma.

Table 3. Results of genetic diversity analysis of SNPs associated with primary angle closure glaucoma.

No.	SNP ID	rs number	MAF	He	Ho	PIC
1	AX-167217422	rs23976401	0.07	0.14	0.15	0.13
2	AX-168246902	rs852876458	0.12	0.20	0.17	0.18
3	AX-168210919	rs851491018	0.16	0.27	0.24	0.23
4	AX-167847910*	rs23984491	0.41	0.48	0.39	0.37
5	AX-167801462	rs851538845	0.16	0.27	0.21	0.23
6	AX-167184748*	rs24001495	0.34	0.45	0.41	0.35
7	AX-167846951	rs852237270	0.05	0.09	0.09	0.09
8	AX-168147206	rs23982463	0.04	0.08	0.06	0.08
9	AX-167810094	rs852110940	0.10	0.17	0.15	0.16
10	AX-167991793	rs852370128	0.26	0.38	0.30	0.31
11	AX-168273159	rs853185625	0.06	0.12	0.09	0.11
12	AX-167194690	rs23998390	0.10	0.19	0.16	0.17
13	AX-167823767	rs23998393	0.06	0.11	0.07	0.1
14	AX-167637581*	rs851382779	0.48	0.50	0.42	0.37
15	AX-167985585	rs24039264	0.19	0.31	0.24	0.26
16	AX-167217114*	rs23979462	0.38	0.47	0.32	0.36
17	AX-167404809	rs23986466	0.24	0.36	0.31	0.30
18	AX-167404447	rs23986562	0.09	0.16	0.14	0.15
19	AX-167897987	rs851343494	0.09	0.16	0.14	0.15
20	AX-167841912	rs23986640	0.13	0.23	0.16	0.2
21	AX-167383510*	rs24025679	0.48	0.50	0.39	0.37
22	AX-167923340	rs24025685	0.11	0.19	0.13	0.17
23	AX-168066391*	rs851371034	0.45	0.50	0.40	0.37
24	AX-167604917*	rs851223555	0.48	0.50	0.32	0.37
25	AX-167320296	rs24025719	0.09	0.17	0.15	0.16
26	AX-167172236	rs23978879	0.29	0.41	0.28	0.33
27	AX-168050101	rs23989983	0.07	0.14	0.13	0.13
28	AX-168174642*	rs8731074	0.45	0.50	0.36	0.37
29	AX-167396835*	rs24025767	0.34	0.45	0.33	0.35
30	AX-167797246	rs850592213	0.21	0.33	0.22	0.27
31	AX-167816257	rs24025768	0.22	0.35	0.26	0.29
32	AX-167569583*	rs24025787	0.41	0.48	0.35	0.37
33	AX-167694666	rs23962988	0.12	0.20	0.15	0.18
34	AX-167619949*	rs9220269	0.31	0.42	0.34	0.33
35	AX-167662205	rs8803780	0.11	0.20	0.16	0.18
36	AX-168148502*	rs24025842	0.31	0.42	0.29	0.33
37	AX-167198755*	rs24025844	0.42	0.49	0.43	0.37
38	AX-167497274	rs24025861	0.24	0.37	0.28	0.30
39	AX-167826614	rs24025878	0.20	0.31	0.22	0.27
40	AX-167242351*	rs24025886	0.46	0.50	0.34	0.37
41	AX-168092155	rs9118158	0.19	0.31	0.19	0.26
42	AX-167553235	rs23992915	0.26	0.38	0.28	0.31
43	AX-168011825	rs23975180	0.28	0.41	0.34	0.32
44	AX-167651967*	rs8523148	0.42	0.49	0.33	0.37
45	AX-167227442	rs9080203	0.22	0.34	0.28	0.28
46	AX-167747971	rs852540802	0.15	0.25	0.23	0.22
47	AX-167651717*	rs23967376	0.39	0.48	0.4	0.36
48	AX-167724356	rs23967416	0.23	0.35	0.28	0.29
49	AX-167639625*	rs23991372	0.47	0.5	0.43	0.37
50	AX-167199290*	rs23962261	0.32	0.43	0.46	0.34
51	AX-167933210	rs851921536	0.05	0.09	0.07	0.09
52	AX-168088678	rs850873816	0.04	0.07	0.05	0.07
53	AX-167215217*	rs23967454	0.32	0.43	0.29	0.34
54	AX-167361648*	rs23982770	0.36	0.46	0.39	0.36
55	AX-167702138	rs23982772	0.17	0.29	0.26	0.25
56	AX-167948469	rs23999265	0.19	0.3	0.24	0.26
57	AX-167913500*	rs851530134	0.36	0.46	0.48	0.35
58	AX-167410376	rs23961325	0.22	0.34	0.33	0.28
59	AX-167669027	rs9033620	0.21	0.33	0.32	0.28
60	AX-167667318	rs850779465	0.21	0.33	0.31	0.27
61	AX-167913149*	rs23995410	0.39	0.48	0.4	0.36
62	AX-168247394	rs9232790	0.14	0.24	0.23	0.21
63	AX-167855288*	rs23968050	0.42	0.49	0.44	0.37
64	AX-167527485	rs23999120	0.22	0.34	0.31	0.28
65	AX-167546901*	rs852082637	0.32	0.43	0.36	0.34
Mean		0.24		0.33	0.27	0.27

MAF, minor allele frequency; He, expected heterozygosity; Ho, observed heterozygosity; PIC, polymorphic information content. *SNPs with high genetic diversity..

Table 4. Annotation results of SNPs with high genetic diversity.

No.	SNP ID	rs number	SNP location
1	AX-167847910	rs23984491	Intron variant
2	AX-167184748	rs24001495	Intron variant
3	AX-167637581	rs851382779	Intron variant
4	AX-167217114	rs23979462	Intron variant
5	AX-167383510	rs24025679	Intron variant
6	AX-168066391	rs851371034	Intron variant
7	AX-167604917	rs851223555	Intron variant
8	AX-168174642	rs8731074	Intron variant
9	AX-167396835	rs24025767	Intron variant
10	AX-167569583	rs24025787	Intron variant
11	AX-167619949	rs9220269	Intron variant
12	AX-168148502	rs24025842	Intron variant
13	AX-167198755	rs24025844	Intron variant
14	AX-167242351	rs24025886	Missense variant
15	AX-167651967	rs8523148	Intron variant
16	AX-167651717	rs23967376	Intron variant
17	AX-167639625	rs23991372	Intron variant
18	AX-167199290*	rs23962261	Intron variant
19	AX-167215217	rs23967454	Intron variant
20	AX-167361648	rs23982770	Intron variant
21	AX-167913500	rs851530134	Intron variant
22	AX-167913149	rs23995410	Intron variant
23	AX-167855288	rs23968050	Intron variant
24	AX-167546901	rs852082637	Intron variant

*Oliver et al. (2019), a SNP associated with Primary Angle Closure Glaucoma..

Risk prediction and protein structure analysis

The nsSNP (AX-167242351) selected for this study was evaluated for deleterious effects. Using SIFT, nsSNPs with a SIFT score of 0.05 or lower are considered deleterious. The analysis showed that the SIFT score of AX-167242351 was 0.00, indicating a very low probability and suggesting a potential impact on the structure or function of the protein (Table 5). PolyPhen-2 analysis evaluated nsSNPs with HumDiv and HumVar scores close to one as deleterious. This method was used to assess the deleterious effects of AX-167242351. The analysis showed that AX-167242351 had HumDiv and HumVar scores of 0.005 and 0.002, respectively, with both HumDiv and HumVar predictions being classified as “BENIGN,” suggesting that its impact on protein structure or function is likely minimal (Table 6). To interpret the function of TNS1, the KEGG pathway and Gene Ontology (GO) analyses were performed. The species was initially set as Canis lupus familiaris for KEGG pathway analysis, but no associated pathways were identified. Therefore, the species was reset to Homo sapiens, which has been reported to share various genes with dogs, and the analysis was repeated. The analysis revealed that TNS1 is associated with the integrin signaling pathway (Fig. 1). Functional interpretation using GO analysis identified 6 GO terms (Table 7). The TNS1 gene was found to function in cell-matrix adhesion and focal adhesion primarily. The impact of amino acid variations in the nsSNP (AX-167242351) on protein structure is shown in Fig. 2. The C-to-T variation in AX-167242351 resulted in an amino acid change from serine to leucine at position 9. This change leads to alterations in the residue structure, ultimately affecting overall protein folding.

Table 5. SIFT analysis results of nsSNP.

Gene	SNP ID	rs number	CDS position	Protein position	Amino acid	Codons	SIFT
TNS1	AX-167242351	rs24025886	26	9	S/L	tCg/tTg	0.00

Table 6. Polyphen-2 analysis results of nsSNP.

Gene	Transcript	REF/ALT	HumDiv	HumDiv prediction	HumVar	HumVar prediction
TNS1 (ENSCAFG 00000014575)	ENSCAFT 00000065916	S/L	0.005	BENIGN	0.002	BENIGN

Table 7. GO analysis results of TNS1 gene.

Function	No. of GO terms	GO terms
Biological process (BP)	2	GO:0007044 (cell-substrate junction assembly)
		GO:0010761 (fibroblast migration)
Cell component (CC)	1	GO:0005925 (focal adhesion)
Molecular function (MF)	3	GO:0004721 (phosphoprotein phosphatase activity)
		GO:0005515 (protein binding)
		GO:0046872 (metal ion binding)

Figure 1.KEGG pathway analysis of TNS1 (Tensin1) gene.

Figure 2.Comparison of protein structural variations according to amino acid sequence variations of nsSNPs.

SNP marker sequence verification

Sequence validation and genotyping of the SNP markers were conducted for ten dog breeds in which glaucoma has been reported domestically and internationally. The analysis showed three genotypes (CC, TT, and CT) at AX-167242351 and three genotypes (TT, CC, and TC) at AX-167199290, confirming consistency in the sequence structure for each SNP marker (Fig. 3). The genotypes for each marker were individually analyzed across 95 DNA samples, and the results are summarized in Table 8. The frequencies of the major homozygous (M/M), heterozygous (M/m), and minor homozygous (m/m) genotypes in AX-167242351 were 0.25, 0.45, and 0.29, respectively. For AX-167199290, the frequencies were 0.65, 0.18, and 0.16, respectively. Subsequently, genotype frequency analysis for each SNP marker was conducted for each breed (Table 9 and 10). The analysis revealed that for AX-167242351, the M/M genotype was most frequent in Golden Retrievers, M/m in Poodles, and m/m in Shih Tzus. For AX-167199290, the M/M genotype was most frequent in Pomeranians, M/m in Jindo Dogs, and m/m in Shiba Inus dogs.

Table 8. Genotype frequency analysis of SNP markers associated with Primary Angle Closure Glaucoma.

No.	SNP ID	Allele (major/minor)	MAF	Genotype frequency
				M/M	M/m	m/m
1	AX-167242351	T/C	0.48	0.25	0.45	0.29
2	AX-167199290	C/T	0.26	0.65	0.18	0.16

Table 9. Analysis of genotype frequency by breed for marker AX-167242351.

No.	Breed	No. of sample	Genotype frequency
			M/M	M/m	m/m
1	Maltese	13	0.46	0.08	0.46
2	Shih Tzu	13	0.00	0.31	0.69
3	Pomeranian	13	0.46	0.31	0.23
4	Poodle	12	0.00	0.92	0.08
5	Golden Retriever	12	0.75	0.17	0.08
6	Yorkshire Terrier	11	0.27	0.73	0.00
7	Jindo Dog	9	0.00	0.78	0.22
8	Cocker Spaniel	5	0.00	0.60	0.40
9	Shiba Inu	4	0.50	0.50	0.00
10	Miniature Pinscher	3	0.67	0.33	0.00

Table 10. Analysis of genotype frequency by breed for marker AX-167199290.

No.	Breed	No. of sample	Genotype frequency
			M/M	M/m	m/m
1	Maltese	13	0.46	0.15	0.38
2	Shih Tzu	13	0.69	0.23	0.08
3	Pomeranian	13	0.85	0.15	0.00
4	Poodle	12	0.75	0.25	0.00
5	Golden Retriever	12	0.75	0.00	0.25
6	Yorkshire Terrier	11	0.64	0.00	0.36
7	Jindo Dog	9	0.56	0.44	0.00
8	Cocker Spaniel	5	0.80	0.20	0.00
9	Shiba Inu	4	0.25	0.25	0.50
10	Miniature Pinscher	3	0.33	0.33	0.33

Figure 3.Genotyping and sequence verification of SNP markers by sequencing. Genotype of AX-167242351 (A). Genotype of AX-167199290 (B).

This study conducted SNP chip analysis on 95 dogs of 26 breeds raised in South Korea to select SNP markers for predicting the risk of primary closed-angle glaucoma. A total of 65 SNPs in the candidate gene (TNS1) were identified, and genetic diversity analysis was performed. Markers with a PIC value above 0.5 are reported to have high polymorphism and high utility, while those with a PIC value between 0.25 and 0.5 are considered to have moderate utility (Botstein et al., 1980). Therefore, to select SNPs with high genetic diversity, 24 candidate SNPs were identified based on having a PIC value ≥ 0.3, an MAF ≥ 0.3, and He and Ho values above average. Annotation analysis of the candidate, SNPs, revealed one missense variant located in an exon of the gene that could potentially affect protein structure or function and increase disease susceptibility. Based on this, the nsSNP (AX-167242351) and the associated intron SNP (AX-167199290) identified by Oliver et al. (2019) were selected as the final SNP markers for risk prediction.

To assess the impact of the selected nsSNPs (AX-167242351), both SIFT and PolyPhen-2 were used to evaluate their deleteriousness. This is to validate the deleteriousness of the nsSNP using various algorithms. SIFT analysis indicated that the variant might affect protein structure or function, whereas PolyPhen-2 analysis suggested its impact was likely minimal. This discrepancy is likely due to the different algorithms used in each method, leading to different conclusions regarding the same nsSNPs. Therefore, further analysis using different methods is required.

KEGG pathway analysis showed that the TNS1 gene is associated with the integrin signaling pathway. Integrins are transmembrane proteins that play key roles in cell-matrix adhesion and activation of various intracellular signaling pathways. They are particularly known to play a role in normal eye development and have been implicated in ocular diseases (Mrugacz et al., 2021). TNS1 is a focal adhesion molecule that links the actin cytoskeleton to integrins and forms a signaling complex through multiple binding domains (Shih et al., 2015). This integrin signaling pathway regulates the actin cytoskeleton in trabecular meshwork cells of the eye. It has been reported to be associated with astrocyte migration and microglial activation in the optic nerve head of patients with primary open-angle glaucoma, potentially leading to an imbalance in aqueous humor outflow through the trabecular meshwork. Consequently, this chain of actions increases intraocular pressure, resulting in damage to the optic nerve (Zhong et al., 2013). In addition, ex vivo studies in human eye models have shown that actin cytoskeletal remodeling within trabecular meshwork cells is a key determinant of aqueous humor outflow and intraocular pressure (Soundararajan et al., 2024). Elevated intraocular pressure is a major cause of glaucoma, and based on this, TNS1, which acts as a link between the actin cytoskeleton and the extracellular matrix through focal adhesion, is considered a candidate gene. Therefore, mutations in this gene may contribute to the development of glaucoma and other ocular diseases by disrupting normal interactions with integrins. However, because this analysis was based on Homo sapiens information, additional functional studies specifically targeting dogs are necessary.

Protein structure analysis confirmed that the nsSNP induced structural changes in TNS1. These alterations in protein folding may result in dysfunction, potentially impairing the normal role of the TNS1 protein and affecting its interaction within the integrin signaling pathway. However, because this result is based on protein structural variations, additional studies involving binding and interaction information between proteins are necessary.

For sequence validation and genotyping of the two selected SNP markers, an analysis was performed on ten dog breeds in which glaucoma was reported domestically and internationally. Analysis of genotype frequencies across breeds for each SNP marker showed that there were differences in genotype frequencies across breeds for both markers. This suggests that given the risk alleles for these markers, there may be potential differences in disease risk across breeds, and that these markers may be used as risk prediction markers. However, since there has been no previous study on the risk alleles by genotype of the TNS1 gene, further research using sufficient sample sizes and phenotypic data is required.

Pets do not benefit from the public health insurance system when visiting veterinary clinics, and pet owners must bear the full cost of medical services. This creates a significant financial burden, which potentially hinders the provision of stable healthcare services. Consequently, the need for pet insurance to alleviate medical expenses has increased; however, the lack of a systematic structure has led to very low adoption rates. Elderly dogs are particularly vulnerable to various diseases because of the decline in bodily function, which makes disease onset and progression common. Since this age group incurs higher medical expenses, managing and preventing major diseases are necessary. The present study used SNP chip data from domestically raised dogs to identify candidate genes and SNP markers associated with primary closed-angle glaucoma, a major disease affecting elderly dogs. Further research using sufficient sample size and phenotypic information could enable early management and prevention of diseases by predicting the risk of onset, thus alleviating the financial burden on pet owners. Additionally, using genotype frequencies by breed and individual dogs as a basis for assessing disease risk could contribute to developing a systematic insurance system and advancing the medical industry.

This study is based on the author’s Master’s thesis.

Conceptualization, H.S.K.; methodology, J.H.J., G.H.L., Z.B., H.S.K.; investigation, J.H.J., G.H.L., Z.B.; writing-original draft preparation, J.H.J., G.H.L.; writing-review and editing, J.H.J., G.H.L., H.S.K.; supervision, H.S.K.; project administration, H.S.K.; funding acquisition, H.S.K.

None.

This study was conducted with the approval of the Animal Ethics Committee of Hankyong National University (No. 2024-3).

Not applicable.

Not applicable.

Not applicable.

No potential conflict of interest relevant to this article was reported.