Putting things together

Aging is considered as one of the largest risk factors for both diseases and mortality (Flatt and Partridge, 2018). Age-related diseases include sensory changes (e.g. hearing loss), chronic- (e.g. high blood pressure) and neurodegenerative disorders (e.g.

Alzheimer’s disease) among others (Jaul and Barron, 2017). Furthermore, both the number- and proportion of elderly people are expected to increase at an accelerating rate in the future (the United Nations estimated a 230% increase – from 901 million to 2.092 billion – in the population of people of 60 years or above from 2015 to 2050;

United Nations, 2015). Both longevity and healthspan (defined as the “period of time during which humans and non-human animals are generally healthy and free from serious chronic illness”; quote from Wallis et al., 2018) are direct consequences of (a healthy) aging and are equally as important from both social and economic points of view.

Firstly, dogs have a much shorter life expectancy than humans – 10-12 years on average in dogs (Adams et al., 2010; Proschowsky et al. 2003; Leroy et al., 2015) vs.

72 years in humans (2016 estimation; WHO, 2018) – making aging-related experiments much shorter in time. Secondly, dog breeds show a huge phenotypic and genetic variability (Leroy et al., 2009; Ostrander et al., 2017), which can also be observed in longevity (e.g. Inoue et al., 2018; Jimenez, 2016). Dogs also have more shared ancestral sequences with humans than rodents do (Lindblad-Toh et al., 2005). Furthermore, they also share the same, rich environment that humans live in (unlike laboratory animals, which are kept in a highly controlled environment) and therefore near-identical environmental factors influence their everyday life, aging and life expectancy (Hoffman et al., 2018) than those influence the aging of humans. Finally, dogs have multiple, spontaneously developing age-related diseases late in their lives that can be considered as analogues of multiple human diseases (e.g. the canine cognitive dysfunction, which is similar to Alzheimer’s disease; see Chapagain et al., 2018 for

15 Partly based on Jónás, D., Sándor, S., Tátrai, K., Egyed, B. Kubinyi, E., (2019). The genetic background of longevity based on whole-genome sequence data of two methuselah dogs. Submitted.

more details). The undeniable relevance of this final point and the previously mentioned other advantages of the companion dog make this species an ideal candidate model animal for the study of the process of aging. In addition, it is also expected that the obtained results are more applicable to humans than those conducted on other laboratory animals.

The study of aging and longevity is of great importance. It enables us to better understand the process of aging itself, which in turn allows the promotion of a healthy lifestyle among the general population and allows mankind to successfully cope with the long-term socio-economical consequences of an aging population. These studies have recently gained a lot of attention (e.g. Chapagain et al. 2018; Mongillo, Araujo, et al. 2013; Szabó et al. 2016; Wallis et al. 2016), both due to the dog’s potential as a model animal (Adams et al. 2000; Araujo, Studzinski, and Milgram 2005; Cummings et al. 1996) and to expand veterinary applications (Landsberg, Hunthausen, &

Ackerman, 2003). Our results support the increasing amount of evidence (e.g. Cotman and Head 2008; Cummings et al. 1996; Head 2013; Kaeberlein, Creevy, and Promislow 2016; Neus Bosch et al. 2012; Schütt et al. 2016) that describes the domestic dog as a good translational model for the study of aging. Both dogs and humans spontaneously develop similar medical conditions that increase the risk of death, as they age, such as cancer (Hoffman, Creevy, Franks, O’Neill, and Promislow 2018). We have observed similar changes in behaviour as well in a convenient sample of pet dogs (Szabó et al 2018, Chapter 7). This body of evidence may reflect a naturally occurring and age-related cognitive decline, which follows similar pathways in dogs as it does in humans (Head et al., 1995; Head, 2013), and results for example in decreased capacity of emotional processing (Smit et al, 2019, Chapter 10) and reversal learning (Piotti et al 2018, Chapter 9).

The brain mechanisms behind these changes have been also suggested to be similar in dogs and humans (see age-related changes in characteristics of EEG sleep spindles, Iotchev et al. 2019, Chapter 16). Beta-amyloid accumulation has been observed in dogs as early as 9 years of age in the brain region of the prefrontal cortex and from 14 years on in the entorhinal cortex (Head et al. 2000). Atrophy of the cerebral cortex and enlargement of the brain ventricles have also been described in MRI (Magnetic Resonance Imaging) scans of a group 16 years old dogs (Kimotsuki et al., 2005) and in morphometry studies of the cerebral ventricles of both young and old dogs (2-5 years vs 10-12 years, González-Soriano et al. 2001). One study looked at MRI scans of a group of 18 dogs aged between 4 and 15 years (with age treated as continuous variable), observing a non-linear relationship between age and brain ventricles enlargement, cortical atrophy. Interestingly, individual differences were also detected, as one 6 year old dog in the sample was as severely affected as the 14 year old dogs in the group; this suggests that some individuals can develop age-related brain degeneration before others and this can be as early as 6 years of age (Su et al., 1998). The same dogs had been tested for their cognitive performance in a different study (Head, Callahan, Muggenburg, Cotman, & Milgram, 1998), but no relationships between cognitive findings and brain changes were reported (Su et al., 1998). The dogs had been tested for cognitive performance up to 2 years prior to the MRI scanning (Head, Callahan,

Muggenburg, et al. 1998; Su et al. 1998) and the results of the cognitive tests were reported according to 3 distinct age groups (young: < 5 years, middle aged: 5-10 years, old: 10+ years which made any comparison difficult to interpret. Studzinski et al. (2006) also investigated cognitive decline in dogs; they observed spatial deficits starting from the age of 6 and reported that age alone predicted 48.2% of the variability in learning in a task to assess memory. Overall, a large body of evidence indicates that functional decline in cognitive domains, such as learning, memory, executive function, and spatial function, occurs similarly in dogs and humans as they age (for reviews, Cotman &

Head, 2008; Head et al., 2013). It should however be noted that, in this body of research, it is difficult to evaluate the effect of age alone as these findings are rarely adjusted to the lifespan variability that is due to factors such as breed, size, or weight (Szabó et al., 2016), effects we were aware of and controlled for it with equation (Szabó et al 2018, Chapter 7) or specifically chosen samples (Smit et al, 2019, Chapter 0; Piotti et al 2018, Chapter 9, Bognár et al 2018, Chapter 11).

In addition to the physiological reasons described above, the use of family dogs as models for aging research is also supported by ecological reasons. Through their unique domestication history, dogs have adapted to a specific niche, i.e. the human social environment (Topál et al. 2009). However, a large proportion of the research on dog cognitive decline involves purpose-bred and raised laboratory animals (Head, 2013 for a review). It should be noted that laboratory conditions do not mimic the animals’

natural environment (Wood, Desjardins, & Fernald, 2011). There is some evidence that family dogs (i.e. dogs living with humans as pets) and purpose-bred research dogs (i.e.

kennel reared domestic dogs) diverge in their performance during some cognitive tasks (Lazarowski & Dorman, 2014). Indeed, it has been argued that dogs residing long-term in kennel environments may be affected by cognitive deficits, due to the lack of stimulation provided by their living environment (Miklósi & Topál, 2011; Mongillo, Araujo, et al., 2013; Turcsán et al., 2019). Thus, results from canine cognitive tests, when performed in conditions similar to those of the human environment, are likely to have strong ecological validity. Family dogs, which share the same living environment as humans, are promising subjects for research that is both clinically relevant and provide the necessary vertical integration of findings originated from invertebrate and rodent models (Waters, 2011). Consequently, studies on family dogs’ cognitive aging have begun to emerge (e.g. Chapagain et al., 2017; González-Martínez et al., 2013;

Heckler et al. 2014; Mongillo et al., 2013; Piotti et al., 2017; Wallis et al., 2016; Wallis et al., 2017). One remaining issue is that some of these tests still require prolonged and complex procedures (discussed below); this limits the replicability of such studies, especially outside of the laboratory setting (Heckler et al., 2014 for a discussion). The importance of replicability is being discussed in psychology research, as replications and data reproducibility are necessary to generalise research findings to the general population (Westfall, Judd, & Kenny, 2015) and further efforts should be made to increase the replicability of research. We have succeeded in developing a simple, relatively quick reversal learning test for family dogs in two versions, facilitating future longitudinal tests (Piotti et al 2018, Chapter 9).

In our studies we found numerous demographic and environmental differences between dog age groups based on the owners’ reports, emphasizing the importance of taking into account these age-related changes in future studies. We have reported that over 40% of a convenient sample of pet dogs had experienced one or more traumatic events in their lifetime that still has an impact on their behaviour (Wallis et al 2018, Chapter 10). Since stress affects the physical, mental and social health of the animal, managing an animal’s stress after a traumatic event, as well as attempting to prevent the occurrence of the event in the first place, is essential in order to improve healthspan and wellbeing in dogs. Owners should be made aware of the personality trait characteristics of their dog, and the methods they use to cope with stressful experiences, as well as the most common types of trauma and their risk factors. In this way, they can better support their dog when stress is unavoidable, but also can attempt to reduce their dog’s exposure to stress in order to diminish any negative impacts on their healthspan.

Since personality and coping style can also change with age within an individual (Kubinyi et al 2009, Chapter 12, Wallis et al 2018, Chapter 13.1), owners should learn how to read their dogs behavior to better understand their specific needs at all life stages, as well as how to prevent the development of unwanted negative behaviors in their dogs for example by using positive training techniques or by adopting older dogs that show less problematic behaviour compared to young individuals (Turcsán et al 2017).

Not only cognition and personality but also social behaviours were found to be affected by age (Howse et al., 2018; Rosado et al., 2012; Szabó et al., 2016). Social rank is likely to change with age (Kubinyi et al 2019, Chapter 15; Bonanni et al. 2010, 2017; Cafazzo et al. 2010; Pal et al. 1998) which might affect at least some cognitive abilities such as observational learning (Pongrácz, 2014; Pongrácz et al., 2008) and also leadership (Ákos et al. 2014, Chapter 14).