Identification of genes associated with longevity in dogs: 9 candidate genes described in Cavalier King Charles Spaniel

Korec, E., Ungrová, L., Kalvas, J., & Hejnar, J. (2024). Identification of genes associated with longevity in dogs: 9 candidate genes described in Cavalier King Charles Spaniel. Veterinary and Animal Science, 100420. https://doi.org/10.1016/j.vas.2024.100420

Summary by: Charlie Clarke

Introduction

The Cavalier King Charles Spaniel (CKCS) has a median lifespan of 9.9 years, and an even shorter healthspan (the years of life when a dog is generally healthy) if we include the years prior to death where they are affected by one or more disorders. Typically, smaller dogs live longer than larger dogs, but the CKCS longevity is similar to that of the Cane Corso, a much larger breed. The King Charles Spaniel, which is approximately the same size as the CKCS and its closest relative, lives a median of 12 years. This discrepancy in median longevity between these two breeds is dramatic and worth investigating. Mitral valve disease and syringomyelia are the leading causes of death for the CKCS. Together, these health disorders account for approximately 37% of all dog deaths. Interestingly, about a quarter of CKCS live to be older than 13, which is significantly older than the median longevity of the breed. This paper aims to compare the oldest 25% of CKCS to CKCS that are between 2 and 7 years old to understand what protective genes are insulating the longer lived spaniels. 

Methods

Oral DNA samples were collected from 69 CKCS in the Czech Republic. Of the original 69 dogs, 28 samples were discarded for not passing DNA quality control metrics. This left 17 dogs in the reference group and 24 dogs in the long lived group. The young dogs in the reference group will be continuously monitored to make sure that they don’t actually belong in the long lived group. The researchers then conducted a genome wide association study (GWAS) to see what mutations are present in the long lived Cavalier. None of the dogs in the study were assessed for their health. They were entirely selected based on their lifespan. 

Results

After the genome wide association analysis, 31 mutations passed the significance threshold. Of those 31 mutations, 15 mutations were found to be located within 9 genes. 12 of the 15 mutations were found within introns, regions of the gene that will be removed before the protein is made. 2 were found to be within exons, the coding portion of the gene that affects the makeup of the protein. Finally, 1 was found to be in the untranslated region, which is not included in the final protein but is typically involved with the regulation of protein synthesis. The introns and the untranslated regions may seem unimportant because they won’t contribute to the final protein, but they are all places where the cell can either speed up or slow down the synthesis of the protein, impacting the equilibrium of each protein within the cell. This can result in too much protein or not enough protein to carry out a specific task. These perturbations are oftentimes more critical than mutations that affect the actual shape of the protein. 

All 9 genes that were implicated in the GWAS were localized to chromosomes 12, 33, and 44. Two single-nucleotide mutations found on chromosome 34 within the B3GALNT1 and NLRP1 genes had the greatest association with longevity. The remaining 7 genes showed a smaller association and are, PARP14, IQCJ-SCHIP1, COL9A1, COL19A1, SDHAF4, B3GAT2, and DIRC2. B3GALNT1, NLRP1, and IQCJ-SCHIP1 are located on chromosome 34, PARP14 and DIRC2 are located on chromosome 33, and COL9A1, COL19A1, SDHAF4, and B3GAT2 are located on chromosome 12.  The last thing that the researchers did was identify the frequency of alleles seen within each population. For example COL9A1 mutation 1 has 3 possible allele combinations: AA, GA, or GG. What they found was that some of the genes showed that 2 copies of the mutation were required for longevity while others only needed one copy of the mutation for protection. If we look at the two most significant genes, the mutation in B3GALNT1 shows that 0% of the long lived dogs had the AA genotype and that only 2.44% of the dogs had a GG genotype and were short lived. Nearly the exact same distribution was seen in the NLRP1 gene. Interestingly, the only gene that actually affected the protein structure was found in PARP14 and had no significant effect on the protein’s structure. 

Figure 1. Genotype distributions for B3GALNT1 and NLRP1. 

Discussion 

The biological relevance of each of these genes can tell us more information about the biological processes that could be disordered in the CKCS. B3GALNT1 is an enzyme that plays a critical role in glycosylation of both proteins and lipids. This enzyme is critical in cell adhesion and signaling responses. It has been implicated in age-related disorders, including neurodegenerative diseases and cancer. NLRP1 plays an important role in the immune system. It is one of the proteins found in the inflammasome, which are associated with aging related complications in humans. NLRP1 recognizes damage to DNA and is a major mediator of cellular aging. PARP14 is involved with a signaling pathway associated with cancer, inflammation, DNA repair, gene transcription, and cell death. PARP inhibitors are commonly used in treatment for cancer in humans. IQCJ-SCHIP1 is a fusion transcript in the brain and has never been described as associated with longevity. The COL9A1 gene encodes a collagen protein involved in the structure and function of cartilage. Dysfunction of this gene is typically linked to the development of hip osteoarthritis. COL19A1 is another collagen protein that forms the extracellular matrix. Collagens maintain tissue integrity and function. Most importantly, dogs affected with mitral valve disease showed a dysfunction in collagen in the affected tissue. SDHAF4 is one protein within Complex II. Dysfunction of Complex II can promote oxidative stress and contribute over time to cellular damage. B3GAT2 regulates the biosynthesis of HNK-1 carbohydrates and has never been previously associated with longevity. DIRC2 is located in proximity to PARP14 and in humans is associated with renal carcinoma and controls cellular homeostasis. 

The mutations on chromosomes 12, 33, and 34 are in close proximity to each other indicating that the longevity in CKCS may be affected by large chromosomal regions. These haplotypes (a group of genes on a chromosome close together) may also have a larger effect because they tend to be inherited together as a set rather than one gene at a time. This genomic analysis points less towards one specific gene as the culprit for shortened longevity, and more towards a collection of genes that all influence the same trait. 
These findings are very interesting, but have some important limitations to consider. GWAS is most reliable with as many data points as possible. This study, having only 17 reference dogs and 24 long lived dogs, is an incredibly small sample size when a gold standard GWAS for a complex trait such as longevity may have 100,000 samples or more. Additionally, if we decide to start selecting for some of these genes, we could accidentally end up creating another bottleneck event that doubles up on other harmful genes. While the genes indicated in this study contribute to shortened longevity in the CKCS, this paper also indicates that they can be inherited together and therefore likely be selected for by breeding to older stud dogs and extensive testing for mitral valve disease. Therefore, while these findings are very interesting as guides for future research, discarding otherwise good breeding candidates from the breeding pool for not having the longevity markers would be unwise at this time.

This work by the Functional Dog Collaborative is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.