Lifetime Observation of Cognition and Physiological Parameters in Male Mice

Kahnau, Pia; Guenther, Anja; Boon, Marcus Nicolaas; Terzenbach, Janine Denise; Hanitzsch, Eric; Lewejohann, Lars; Brust, Vera

doi:10.3389/fnbeh.2021.709775

ORIGINAL RESEARCH article

Front. Behav. Neurosci., 01 September 2021

Sec. Learning and Memory

Volume 15 - 2021 | https://doi.org/10.3389/fnbeh.2021.709775

This article is part of the Research Topic Home Cage-based Phenotyping in Rodents: Innovation, Standardization, Reproducibility and Translational Improvement View all 24 articles

Lifetime Observation of Cognition and Physiological Parameters in Male Mice

$\r\nPia Kahnau*$ Pia Kahnau^1*

Anja Guenther²

Marcus Nicolaas Boon^3,4

Janine Denise Terzenbach⁵

Eric Hanitzsch⁵

Lars Lewejohann^1,6

Vera Brust⁵

¹Laboratory Animal Science, German Centre for the Protection of Laboratory Animals (Bf3R), German Federal Institute for Risk Assessment (BfR), Berlin, Germany
²Department of Animal Behaviour, Bielefeld University, Bielefeld, Germany
³Department for Electrical Engineering and Computer Science, Modeling of Cognitive Processes, Technische Universität Berlin, Berlin, Germany
⁴Exzellenzcluster Science of Intelligence, Technische Universität Berlin, Berlin, Germany
⁵Behavioral Phenotyping Unit, University of Osnabrück, Osnabrück, Germany
⁶Animal Behavior and Laboratory Animal Science, Institute of Animal Welfare, Freie Universität Berlin, Berlin, Germany

Laboratory mice are predominantly used for one experiment only, i.e., new mice are ordered or bred for every new experiment. Moreover, most experiments use relatively young mice in the range of late adolescence to early adulthood. As a consequence, little is known about the day-to-day life of adult and aged laboratory mice. Here we present a long-term data set with three consecutive phases conducted with the same male mice over their lifetime in order to shed light on possible long-term effects of repeated cognitive stimulation. One third of the animals was trained by a variety of learning tasks conducted up to an age of 606 days. The mice were housed in four cages with 12 animals per cage; only four mice per cage had to repeatedly solve cognitive tasks for getting access to water using the IntelliCage system. In addition, these learner mice were tested in standard cognitive tests outside their home-cage. The other eight mice served as two control groups living in the same environment but without having to solve tasks for getting access to water. One control group was additionally placed on the test set-ups without having to learn the tasks. Next to the cognitive tasks, we took physiological measures (body mass, resting metabolic rate) and tested for dominance behavior, and attractivity in a female choice experiment. Overall, the mice were under surveillance until they died a natural death, providing a unique data set over the course of virtually their entire lives. Our data showed treatment differences during the first phase of our lifetime data set. Young learner mice showed a higher activity, less growth and resting metabolic rate, and were less attractive for female mice. These effects, however, were not preserved over the long-term. We also did not find differences in dominance or effects on longevity. However, we generated a unique and valuable set of long-term behavioral and physiological data from a single group of male mice and note that our long-term data contribute to a better understanding of the behavioral and physiological processes in male C57Bl/6J mice.

Introduction

Cognition comprises information processing mechanisms which enable decision making, i.e., perception, memory, and learning (McEwen, 2007; Gruszka et al., 2010; Roelofs et al., 2016). Cognitive abilities thereby enable adaptation to the social and physical environment of an individual. Especially complex or constantly changing environments are better coped with and exploited more successfully by species and individuals with an increased cognitive performance (e.g., Sol and Lefebvre, 2000; Lee, 2003; Dunbar and Shultz, 2007; Boucherie et al., 2019). Accordingly, a higher cognitive ability holds the potential to favor reproductive success and survival (Papini, 2002; Dukas, 2004). However, cognition is also associated with costs as the processing of information requires nervous tissue, which is energetically expensive to develop and maintain (Laughlin et al., 1998; Niven and Laughlin, 2008; Hollis and Kawecki, 2014). Adding up on this constitutive investment are the induced costs of building and maintaining particular memories (Snell-Rood, 2013). The trade-off between costs and benefits of cognition can elegantly be demonstrated by the fact that cognitive abilities are usually not maxed out under natural selection. Indeed, evidence from many different species shows that cognitive abilities can be substantially improved by artificial selection (e.g., Tryon, 1940; Brandes, 1988; Mery and Kawecki, 2002). In most habitats, however, it is generally assumed that natural selection prevents a permanent improvement of cognitive abilities due to trade-offs with other fitness related traits (Buchanan et al., 2013). Thus, an evolved species has a cognitive range within which an individual must incur the corresponding physiological costs depending on the energetic expenditure of brain activity. Studies linking cognition directly to physiological, reproductive, or survival traits are limited (e.g., Cole et al., 2012; Huebner et al., 2018). Therefore, we still know little about how these trade-offs actually shape cognitive abilities or, on an individual level, affect how an individual uses its cognitive abilities. Here, we follow a cohort of male mice throughout life with the aim to study the effects of a cognitively demanding life on different aspects of physiology, reproduction, and survival.

By provisioning an environment which constantly held new cognitive challenges to some but not all of the mice, we wanted to test for the consequences of such different lifestyles. Based on the above mentioned trade-offs, cognitively stimulated mice might have a reduced or slower growth and in addition or alternatively a higher energy turn-over as compared to non-stimulated mice. The energy consumption of an individual can be determined by measuring its metabolic rate, i.e., oxygen uptake and carbon dioxide release, indicating how much energy is produced by aerobic respiration (Brown, 2004). While a direct influence of learning performance or use of cognitive abilities on metabolic rate to our knowledge has not yet been investigated, the metabolic rate in this study was measured four times throughout the mice’s life. We wanted to investigate a possible relation between different learning environments suggesting a higher metabolic rate in males which were cognitively stimulated repeatedly throughout their lives.

Besides these possible physiological contrasts between the differently stimulated mice, we assessed an aspect of reproductive success. Only male mice were included in the study and we tested them twice in a female choice task where potential mates were allowed to freely choose to spend time in close proximity with individuals of the different testing groups. Some studies in insects and vertebrates have shown that females choose mates with better cognitive skills reflected in males’ courtship behavior, performance in foraging or in diet-dependent morphological traits [reviewed in Boogert et al. (2011)]. Additionally, males of species with a complex and competitive sexual environment may have to process complex sensory information and display learned abilities in courting females (Byrne and Rice, 2006; Dukas, 2006; Griffith and Ejima, 2009). In an experimental evolution study in fruit flies cognitive performance of males declined under the absence of sexual selection (Hollis and Kawecki, 2014). Consequently, we might suggest an impact of the different learning environments applied to our tested mice in relation to their attractiveness toward females. Still, as their potential differences in cognitive abilities are not directly on display in our testing context, it is difficult to predict whether the females are able to include these traits in their decisions. In addition to female choice, male-male competition is a core principle of sexual selection. In order to measure whether or not the cognitive stimulation affected intrasexual competition, we performed direct observations of aggressive behavior within the social groups of male mice.

Measuring longevity is a straightforward way to test if the learning environment has fitness consequences. Under natural conditions, longevity may be directly influenced by differences in foraging success or predator avoidance (e.g., Madden et al., 2018). Accordingly, higher cognitive abilities have repeatedly been found to be positively linked to survival in the wild (see Morand-Ferron, 2017 for an overview). However, under captive conditions, differences in predator avoidance or foraging success are unlikely to occur because animals are usually protected from predation and provided with food ad libitum. Nevertheless, the link between physiological condition and cognitive function could affect longevity even under captive conditions, especially if differences in early environments influence physiological and cognitive development through phenotypic plasticity (Pravosudov et al., 2005; Loi et al., 2017). Accordingly, replicate populations of fruit flies selected for an improved learning ability have shown a pronounced reduction in longevity and conversely, lines selected for extended longevity showed a reduction in learning ability (Burger et al., 2008). Our tested mice were kept over their whole lifespan enabling us to directly test for an effect of the applied learning treatment on longevity. In accordance with former laboratory studies, we expect a reduced lifespan in mice kept under cognitively stimulating conditions as compared to the non-stimulated mice. Besides this direct measurement of longevity, we investigated the telomere length of the mice at a later stage in life. Telomeres comprise of repeated and non-coding DNA strands forming the ends of each chromosome and have been linked to rates of aging and age-related diseases in aging human and non-human individuals (Von Zglinicki, 2002; Brouilette et al., 2003; Benetos et al., 2004; Martin-Ruiz et al., 2006). In addition, several studies such as for example by Yaffe et al. (2011) showed that telomere length can also serve as a marker for cognitive aging in humans with shorter telomeres going along with reduced cognitive abilities. While the effects of stressful environments on telomere length and cognitive decline are well described, it is not yet known if an environment that imposes elevated cognitive processes like applied in our study can affect telomere lengths.

Considering all the above, we hypothesize that cognition in male mice throughout life affects physiology (body mass development, metabolic rate), sexually selected traits (competitive ability, male attractiveness), and longevity (measured by actual survival and telomere length) across life. A unique feature of our study is that mice were kept in groups in a home-cage based test apparatus, the IntelliCage (IC, New Behavior) system. This way, individual testing paradigms could be applied to each animal enabling us to form social groups that contained both, learner and non-learner mice. Previous studies show that mice are able to solve learning tasks within the IC and also other parameters such as activity patterns can be measured within the system (Galsworthy et al., 2005; Mechan et al., 2009; Krackow et al., 2010; Endo et al., 2011; Voikar et al., 2018). Home-cage based test systems offer the advantage of testing animals without daily interference of researchers and in accordance with the animals natural activity phases over a long period of time [reviewed in Voikar and Gaburro (2020)].

Before going into further detail, we wish to emphasize that this study comprises data from three initially independent Master theses which all focus on the costs of cognition. We reverted to the same group of mice in all theses which allowed us to compile a unique lifetime observation data set presented in the current study. In this set-up, we consequently had to face slight variations in experimental procedures due to the changing experimenters whereas the same overarching research question allows the data to be presented as a single long-term study with three consecutive phases. In summary, we present a unique and comprehensive lifetime observation of 48 male mice, which to our knowledge has no comparison in the present literature record.

Materials and Methods

Animals and Housing Conditions

48 male C57BL/6J mice (Charles River, Sulzfeld, Germany) arrived at Osnabrück University at the age of 21 days. The mice were randomly separated in four housing groups, 12 mice per cage, which were kept stable over the lifetime of the mice. One day after arrival, the mice received an RFID transponder (radio frequency identification ISO FDX-B 2.12 × 12 mm, Planet ID GmbH, Essen, Germany), which was implanted subcutaneously into the neck area under isoflurane anesthesia. Each social group of 12 mice was further randomly assigned to one of three treatment groups: The learner mice (L), which had to solve various cognitive tasks, the non-learner (NL), which had no tasks to solve, and the equipment control group mice (EC), which also had no tasks to solve, but were exposed to handling similar to that of the L mice during testing. For visual identification, each mouse was assigned a unique two-color code which was applied to the tail skin with lacquer painting pens (edding 750).

During their growth phase up to an age of 73 days, body mass was measured every 3 days. Adult mice were weighted weekly during the process of cage cleaning up to an age of 757 days, except for a non-experimental phase between 192 and 302 days of age. During handling (tail-handling), the tail color-codes were additionally renewed if necessary. The mice were kept at 22 ± 2°C at 56 ± 15% humidity. The dark/light cycle was 12 h each, with light hours between 8:00 am and 8:00 pm. Pellet food (Altromin International, 1314) was available ad libitum at all times. At the age of 325 days, the mice moved to another animal facility within the university. The housing conditions were similar but with an additional half hour of sunrise simulation and the last half hour of the light cycle simulating sunset. Within these facilities mice were kept in different cage types throughout their life as described in detail below. Mice were kept in standard home-cages after the last experiment until they died a natural death. However, to keep animal welfare standards mice were euthanized as soon as signs of pain or suffering were detected.

IntelliCage

The IntelliCage (IC, NewBehavior) is a home-cage in which various cognition tasks can be applied to individual animals with minimal invasion by experimenters. Each IC contained bedding (Allspan, Olympia, 2 cm high), paper as nesting material and four red mouse houses (“TheMouseHouse;” Tecniplast) for shelter. The houses were placed directly under a central feeding rack. Water was available in eight dispensers arranged in the four cage corners (Figure 1). Each corner is only accessible to one mouse at a time. Each corner contains a presence sensor, one RFID antenna and an airpuff valve for mild punishment (0.5 bar). Access or denial to water can be granted to individual animals identified via an implemented RFID sensor, enabling individual operant conditioning in group housed animals. Each corner contains a nosepoke-sensor and a door per water dispenser. The doors can be opened by a nosepoke detected by the nosepoke-sensor. By assigning mice access to individual corners or water dispensers at certain times or in certain orders, tasks of different levels of difficulty could be applied. Nosepokes at non-rewarded dispensers were punished with a 1 s and 0.5 bar airpuff in certain conditions.

FIGURE 1

Figure 1. Schematic figure of an IntelliCage from the top with four conditional corners marked by numbers. Each corner contains a radio frequency identification antenna (gray), a presence-sensor (black), an airpuff valve (light blue) and two water dispensers (dark blue). In the cage center four shelters (red) are arranged under the food rack.

During their stays in the IC, different programs of various difficulty were applied to the L mice allowing and denying access to different corners or water dispensers (Table 1). As water was only offered in the cage corners, we assume that the mice were primarily entering the corners for fluid intake. Accordingly, attempts to visit corners and drink at certain water dispensers reflects the animals’ foraging effort. Therefore, we use the term “foraging behavior” to describe the behavior of corner visiting for getting access to water. The IC tasks got more and more difficult over time including patterns in which the rewarded corner or side within a corner changed after each successful drinking event, like for example to the opposite corner or in a clock- or anticlockwise manner. The difficulty of the IC tasks was increased by allowing the L mice access to water in all four corners, but the L mice received an airpuff in addition to water in three (incorrect) corners. Only in one (correct) corner, drinking was possible without receiving an airpuff. But drinking within the correct corner on the incorrect side leads to a corner change (for example clock- or anticlockwise). If the L mice drank on the correct side within the correct corner, the correct corner did not change (see Supplementary Material for full details on all IC cognition tasks). Some of the IC tasks were repeated, adding up to 51 learning tasks applied to the L mice in the IC. Both the NL and EC mice had access to water at all times but were punished by an airpuff if they stayed for longer than 15 s in a single corner to avoid a mouse occupying a corner and blocking the access for other individuals for too long. In addition, while the NL and EC mice had 15 s to drink, the L mice in the different IC tasks had 8–10 s to drink water within the correct corner or on the correct side. In order to drink again, all mice had first to leave the corner and re-enter it or visit another corner. In total, the mice were kept in the ICs in three phases of different lengths (Table 1, IC phase 1 = 32 tasks, IC phase 2 = 16 tasks, and IC phase 3 = 3 tasks). Even though the L mice had less time within the correct corner to drink and were not able to drink in all corners without punishment relative to the NL and EC mice, we compared the foraging behavior of the three treatment groups with each other. This was carried out by analyzing the number of corner visits that were performed (per week) by the mice. A visit was evaluated as soon as a mouse entered a corner, the RFID antenna registered the RFID transponder of the mouse and at the same time the presence sensor registered the presence of the mouse.

TABLE 1

Table 1. Learning IntelliCage tasks assigned to learner mice within the IntelliCage.

Standard Home-Cage

Mice were kept in type IV Macrolon cages (59 cm × 59 cm × 20 cm) per social group. Food and water were available ad libitum. Each home-cage contained two red houses, a plastic tube, and paper for nesting and bedding. No tests were applied directly in the home-cage, but mice were transferred to different tests over short time periods during their stays in the home-cage (for an overview of the timeline see Figure 2 and Supplementary Table 1).

FIGURE 2

Figure 2. Body mass of mice over their lifetime. Shown are the mean body masses of the three treatment groups. At 305 days of age, the mice moved to another animal facility. Repeated experiments were conducted: IC, IntelliCage; FC, female choice test; BM, Barnes maze test; RMR, resting metabolic rate; TLM, telomere length measurement. When not in the ICs, the mice were kept in standard home-cages within their social groups.

Barnes Maze Test

The Barnes maze test (BM) was carried out to test for spatial learning performance of L mice outside the IC. It uses the natural aversion of mice to open and exposed areas (Crawley, 1985) and measures how many errors they make to escape from such an area over the course of multiple trials. We used a round gray plastic platform of 100 cm diameter mounted about 120 cm above the ground as the exposed area. The platform contained 12 holes of 4 cm diameter evenly distributed along its edges, which could be closed via opaque black plexiglass lids. One hole could be connected via a 4 cm diameter PVC tube to a cage (Makrolon type III, filled with bedding transferred from the cage the mouse was housed in) placed under the platform. Visual cues (a bottle, a cloth, a metal stand, and a box) arranged around the platform served as spatial orientation cues. The test was performed during the light phase at 22 ± 1°C and with 54 ± 4% humidity in the experimental rooms. The illumination on the platform was 10–20 lux. Prior to each trial, the platform was cleaned with ethanol (70%).

For each trial, a single L mouse was placed in a start-cylinder (diameter: 10.5 cm, high: 21 cm) in the middle of the platform for 1 min. After this acclimatization phase, the start-cylinder was removed and the trial started. The number of attempts to enter covered holes to escape from the platform were evaluated from video recordings (Logitech HD webcam, Software: VirtualDub Version 1.9.11) to investigate learning performance. The BM test was performed twice at the age of 92 and 573 days. In the first run each trial lasted 5 min. Each mouse performed ten trials, two trials per day with an inter-trial interval of 30 min on five consecutive days. In the second run the test was performed in a different room and with different positions of the escape hole and the spatial cues surrounding the platform, but the same apparatus with the same illumination conditions was used. For the second run, each mouse performed six 3-min trials, two trials per day with an inter-trial interval of 30 min over three consecutive days. In both runs, if a mouse did not find the open hole within the given time, it was gently guided to the escape hole by the researcher’s hand. To control for a possible handling-effect, each L mouse was assigned to an EC mouse. After the L mouse finished the BM test, the assigned EC mouse was placed on the platform of the maze for exactly the same time it took the L mouse to find the open hole. Contrary to the L mouse, the matched EC mouse could not escape from the platform, as all 12 holes were closed with the black opaque plexiglass.

For further cognitive stimulation outside the IC, a T-maze test, a tone conditioning test (carried out during the first IC phase) and a labyrinth experiment (after the third IC phase) were carried out. However, the results of these tests were not sufficiently conclusive to be included in this work due to methodological problems (e.g., we only know now why the T-Maze did not work: Habedank et al., 2021). Nevertheless, it is important to mention that the control group had other experiences outside the home-cage besides the BM test (for more information about the test procedure see the supplements).

Female Choice Tests

To test for possible differences of attractiveness to females between the three treatment groups of mice, the males were subjected to two female choice tests (FC), one at an age of 152 days (shortly after IC phase 1) and a second one at an age of 590 days (several months after IC phase 3). Females used in the first test were C57BL/6J mice naive to cognitive experiments (N = 16) and 152 days old. Females in the second test were the first generation offspring of C57BL/6J X BALB/C and subjected to cognitive tests including an IC phase themselves prior to the choice tests (N = 12, each female conducted a maximum of two choice tests). At the time of testing the females were 304 days old. The estrus status was not determined. Nevertheless, we must note that the estrus status may affect the female’s behavior. In each test, one female was introduced into a type III Makrolon cage filled with fresh bedding for a 1 h habituation time. Accordingly, three males, one of each treatment group were randomly placed in three similar cages divided in half by a perforated clear plexiglass wall. The female’s cage was connected to the three empty half compartments of the males’ cages via PVC tubes. The location of the female and the time spent with each male over the course of 24 h was automatically recorded using light barriers installed in each tube.

Observations of Agonistic Behavior

To test whether the outcome of agonistic encounters between mice within each social group was influenced by the treatments, live observations on fighting behavior were conducted. Mice within the social groups in general lived rather peacefully together throughout their whole lives. At no time any mouse had to be removed from a group due to social incompatibility. The only times when agonistic behaviors occurred more frequently were during the weekly cage cleaning events, when mice were transferred to clean cages. Cage cleaning and accordingly live observations of agonistic behavior took place once a week between 08:30 am and 11:30 am. The mice of one social group were placed in random order into an observation cage containing only fresh bedding, food, and water. One minute after all mice were placed in this cage, they were observed by an experimenter for 30 min. Within this time, the mice usually calmed down and agonistic behavior almost ceased to occur. The experimenter noted down the IDs of mice involved in fights as well as who won and lost each fight. The behavior fight began as soon as two mice began to circle one another with body contact. During a fight, the mice pushed one another with their paws and bodies or bit, sometimes even pushing the opponent with its back to the ground. Fights could be accompanied by vocalizations, too. A fight ended when one of the two opponents turned its head away from the other. The loser mouse was the one which turned away first. Fights between three or more mice were not included in the data set, as identification of individuals as well as the determination of winners and losers were less straightforward. Observations of agonistic behaviors were done when the mice were between 373 and 759 days old.

Resting Metabolic Rate

To test whether the different treatments of mice led to differences in their metabolism, respirometry measurements were repeatedly taken throughout their life. Respirometry is a method for determining the total energy turnover of an organism. The release of carbon dioxide is determined in relation to oxygen absorption. This allows determining the metabolic rate of an organism. The resting metabolic rate (RMR) measurement was carried out at the Department of Animal Behavior at Bielefeld University. The measurement was performed at 20 ± 1°C under low light conditions in an open flow system by measuring the oxygen consumption and carbon dioxide production with a continuous inflow (45 l/h, Mass Flow Meter FM-360, Tylan Corp., Torrance, CA, United States) of external fresh air. External air was transferred under ambient pressure to two transparent Plexiglas measuring chambers (14 cm × 20.5 cm × 14 cm), which were placed in a climate chamber (Rubarth Appaerate, Laatzen, Germany) and contained paper (for excretion absorption) but no water or food. Both chambers were located in a way that animals could not see, hear, or smell each other. For drying, the air was first pumped to two cooling devices (M&C Cooler, Ratingen, Germany) and then transferred to a molecular sieve. The oxygen consumption and carbon dioxide production were analyzed by an O₂ analyzer (Oxzillar FC, Sable Systems, Henderson, NV, United States) and CO₂ analyzer (Maihak AG, Hamburg, Germany). As a control, we compared dried outside air against the carbon dioxide and oxygen concentrations measured in the outflow of the metabolic chambers, in which the test animals rested. An initial control period of 10 min was used to assure the stability of the system. Over a period of 2.5 h, each animal’s oxygen consumption and carbon dioxide production were measured across six periods of 10 min each. Between measurements, 1-min control intervals were interspersed to allow correction for system drifting if necessary. The specific RMR (KJ/(d^∗kg)^–1) was calculated from the 3-min interval with the lowest, stable oxygen consumption throughout the measurement periods. Within the apparatus, two animals were measured simultaneously between 08:30 am and 05:00 pm. The evening before the measurement the mice were separated in type III Macrolon cages (with nesting and bedding, water and food from the home-cage) to habituate them to being separated during measurement. Directly before and after the measurement, body mass was measured. After RMR measurement, the mice were placed back into their home-cages. The RMR was measured four times, at an age of 138, 308, 482, and 665 days.

Telomere Lengths Measurement

To investigate whether repeated cognitive stimulation affects the length of telomeres, from all 44 mice still alive (15 L mice, 14 NL mice and 15 EC mice) at an age of 699 days, a blood sample was taken from the tail vein. Previous studies showed a link between telomere length and aging, age-related diseases or cognitive aging (Von Zglinicki, 2002; Brouilette et al., 2003; Benetos et al., 2004; Martin-Ruiz et al., 2006; Yaffe et al., 2011). By repeated cognitively stimulating the L mice, we assumed a difference in telomere length between the three treatments. Therefore, in our study telomere length should serve as a longevity marker (in addition to the classical survival analysis). DNA was extracted out of the leukocytes according to the instructions of the UltraClean^® Blood DNA Sample Kit (Non-Spin, MO BIO Laboratorie, Inc.) and used in a telomere real time quantitative polymerase chain reaction (RT qPCR) performed with a BioRad PCR system consisting of the BioRad CFX96^TM RealTime System as optical reaction module and BioRad C1000 Touch^TM Thermal Cycler following Cawthon et al. (2003) and Callicott and Womack (2006). From each sample DNA and from a control DNA (single copy gene 36B4), three replicates were prepared. The dilutions of the sample DNA were: 100, 20, 4, 0.8, and 0.16 ng/5 μl. The dilutions of the control DNA were: 20, 4, 0.8, 0.16, and 0.032 ng/5 μl. In addition, a zero-control (Null Template Control-NTC) was prepared with three replicates. 10 μl GoTag^®qPCR Master Mix, 1.8 μl telg-Primer, 1.8 μl telc-Primer and 1.4 μl PCR-water were added to the DNA samples. 10 μl GoTag^®qPCR Master Mix, 1.8 μl forward-Primer, 1.8 μl reverse-Primer and 1.4 μl PCR-water were added to the control samples. The relative telomere length was then determined from the C_T value of the sample DNA and the C_T value of the control DNA.

Data Analysis

The following data analysis and visualization was performed with R (R Core Team, 2020; version 3.4.2) and Python (version 3.8), using the panda package (version 1.2.3).

Body Mass Measurement

For body mass analysis four linear mixed effects models (package nlme, Pinheiro et al., 2020) with weight (log transformed), housing condition (housed in the IC vs. housed in the home-cage) and age set as fixed effects were calculated for each of the three IC phases and in the home-cage. Animal ID served as a random effect. Model assumptions were checked visually by Q–Q plots and by plotting fitted versus residual values. To test for effects between treatment groups within housing conditions, pairwise post hoc comparisons were conducted (package emmeans, Lenth, 2020).