Climate warming could challenge the ability of endotherms to thermoregulate and maintain normal body temperature (Tb), but the effects of warming summer temperatures on activity and thermoregulatory physiology in many small mammals remain poorly understood. We examined this issue in deer mice (Peromyscus maniculatus), an active nocturnal species. Mice were exposed in the lab to simulated seasonal warming, in which an environmentally realistic diel cycle of ambient temperature (Ta) was gradually warmed from spring conditions to summer conditions (controls were maintained in spring conditions). Activity (voluntary wheel running) and Tb (implanted bio-loggers) were measured throughout, and indices of thermoregulatory physiology (thermoneutral zone, thermogenic capacity) were assessed after exposure. In control mice, activity was almost entirely restricted to the night-time, and Tb fluctuated ∼1.7°C between daytime lows and night-time highs. Activity, body mass and food consumption were reduced and water consumption was increased in later stages of summer warming. This was accompanied by strong Tb dysregulation that culminated in a complete reversal of the diel pattern of Tb variation, with Tb reaching extreme highs (∼40°C) during daytime heat but extreme lows (∼34°C) at cooler night-time temperatures. Summer warming was also associated with reduced ability to generate body heat, as reflected by decreased thermogenic capacity and decreased mass and uncoupling protein (UCP1) content of brown adipose tissue. Our findings suggest that thermoregulatory trade-offs associated with daytime heat exposure can affect Tb and activity at cooler night-time temperatures, impacting the ability of nocturnal mammals to perform behaviours important for fitness in the wild.

As global temperatures continue to rise as a consequence of the effects of climate change, there is a growing need to understand how warming ambient temperatures (Ta) impact the physiology, health and distribution of animal species (Mitchell et al., 2018; Pörtner and Knust, 2007). In endotherms, warming Ta can limit body heat dissipation and threaten their ability to maintain body temperature (Tb). Although species can vary in the magnitude of Tb variation considered normal, excessive increases in Tb (hyperthermia) can cause a range of cellular and physiological problems that lead to pathology and death (Speakman and Król, 2010; Yan et al., 2006). Indeed, the high Ta recently experienced by some tropical and desert endotherms has increased T towards dangerous and even lethal temperatures (Danner et al., 2021; McKechnie and Wolf, 2019). Less severe increases in Ta can constrain locomotor activity, reproduction and various other important performance traits, likely as a result of limits to metabolic heat production (Bao et al., 2020; Speakman and Król, 2010; Tapper et al., 2020). However, while our understanding of the effects of climate warming is improving for some endotherms, such as tropical and desert endotherms (Lovegrove et al., 2014; McKechnie and Wolf, 2010; Tewksbury et al., 2008) and some large temperate mammals (Hetem et al., 2014; McCain and King, 2014), the impacts of climate warming remain largely unknown for many other species (Levesque et al., 2016).

Climate warming may present some distinct challenges for small temperate mammals. Small mammals tend to have relatively narrow thermoneutral zones (TNZs), defined as the range of Ta over which resting metabolic rate is lowest, bounded by lower (LCT) and upper critical temperatures (UCT). For example, small mammals such as rats and mice have TNZs between 28 and 32°C or a narrow portion of that range (Chappell, 1985; Heldmaier, 1989; Hock and Roberts, 1966; Parker, 1988; Porter and Kearney, 2009; Romanovsky et al., 2002). Considering the increasing summer temperatures in many temperate regions, some small mammals could potentially spend much of the daily cycle outside their TNZ – above it during hot daytime periods (necessitating increased heat dissipation, evaporative water loss, etc.) but still below it during cool night-time periods (necessitating thermogenesis, heat retention, etc.) (Gordon, 2012; Mitchell et al., 2018; Rosenmann and Morrison, 1974; Song et al., 1987; Van Sant and Hammond, 2008). The consequences of chronic exposure to such conditions are poorly understood. Chronic exposure to hot temperatures leads to well-known processes of heat acclimation, characterized by an improvement in heat dissipation capacity and heat tolerance (Glaser, 1949; Horowitz, 2001; Racinais et al., 2019; Sareh et al., 2011). By contrast, cold acclimation can increase thermogenic capacity and heat retention (Rezende et al., 2004; Stager et al., 2020; Van Sant and Hammond, 2008). Chronic exposure to daily cycles of hot and cool temperatures may present the possibility for a thermoregulatory trade-off, whereby plastic improvements for coping with daytime heat are associated with a reduced ability to cope with cool temperatures at night. This possibility has received relatively little attention but could have a significant role in the impacts of climate warming on small temperate endotherms. This issue may be significant for nocturnal species that need to remain active at night, particularly if they are unable to completely avoid exposure to high daytime Ta in burrows (Murray and Smith, 2012; Tracy and Walsberg, 2002). Indeed, if high daytime Ta leads to plastic adjustments that reduce tolerance of cool temperatures below the TNZ, night-time activity could be impaired at temperatures that might otherwise be considered favourable (Bonebrake et al., 2020; McCain and King, 2014).

The deer mouse (Peromyscus maniculatus) is a useful species for understanding the potential impacts of climate warming on small temperate mammals. Deer mice are nocturnal and highly active, and they inhabit many different environments and habitats across North America (Bedford and Hoekstra, 2015). They build relatively short, shallow burrows that provide only partial protection from summer daytime temperatures (Hayward, 1965; Hu and Hoekstra, 2017). In recent years, the latitudinal distribution of deer mice has been shifting northward in Canada (southern Ontario and Quebec) and the mid-western USA, where they have often been displaced by the white-footed mouse (Peromyscus leucopus) (Fiset et al., 2015; Myers et al., 2005, 2009; Roy-Dufresne et al., 2013; Walsh et al., 2016). The potential role of the effects of summer warming on thermoregulatory physiology in these range shifts is poorly understood. To gain insight into this issue, our objective here was to examine how exposure to environmentally realistic patterns of summer warming impacts thermoregulation and activity. This was achieved by measuring Tb and activity during exposure to warming diel temperature cycles, representative of the transition from spring to mid-summer in an area where deer mouse populations are declining, and comparing patterns of variation with those of mice held at temperatures representing mid-spring. We also sought to examine whether thermoregulatory trade-offs associated with heat acclimation might contribute to the effects of summer warming on Tb and activity.

Animals and husbandry conditions

Deer mice, Peromyscus maniculatus (J. A. Wagner 1845), were live-trapped in Sherman traps in prairie and farmed habitats near Kearney (Buffalo County, NE, USA, in the area surrounding 40°49′48.0″N, 99°05′02.6″W). Mice were then transported to McMaster University (Hamilton, ON, Canada) and held in standard mouse cages (containing 7090 Teklad Sani-Chips® animal bedding; Envigo, Indianapolis, IN, USA) at ∼25°C, 14 h:10 h light:dark photoperiod, and with unlimited access to water and rodent chow (Teklad 22/5 Rodent Diet formula 8640; Envigo). Seven independent families were bred in captivity to produce first-generation lab progeny, which were then raised to adulthood (∼6–18 months of age) in the same conditions. Animal husbandry conditions and experimental protocols were approved by the McMaster University Animal Research Ethics Board according to guidelines from the Canadian Council on Animal Care.

Experimental treatment groups

Adult first-generation mice were randomly divided into two experimental groups: one subjected to a constant diel temperature cycle representative of late spring, and one subjected to warming diel temperature cycles representative of the transition from spring to mid-summer. To inform the temperatures used in each experimental group, we measured environmental temperatures at a site near Lincoln (Lancaster County, NE, USA, at GPS coordinates 40.74950247590457, −96.81771341527244) in 2020. Deer mice were abundant at this site until recently, but the site is now populated primarily by white-footed mice (Jay F. Storz, University of Nebraska, personal communication), potentially as a consequence of range shifts resulting from climate change (Myers et al., 2005; Roy-Dufresne et al., 2013). An iButton temperature logger (DS1922L, Embedded Data Systems, Lawrenceburg, KY, USA) was deployed at ground level under the shelter of a large wood pile, representing a mouse microhabitat with strong trapping success, and temperature was recorded every 2 h. The temperature data acquired were used to determine average daytime high and average night-time low temperature for each week over the period of summer warming from May to July (Fig. 1A), and these values were used to inform the exposure temperatures used in the two experimental groups. The ‘spring control’ group was exposed to diel temperature variation representative of late May, with temperatures of 27.5°C during the daytime (07:30–21:30 h local time) and 13°C during the night-time (21:30–07:30 h local time) for 8 weeks (10 males, 12 females) (Fig. 1B, top). The ‘summer warming’ group was exposed to temperatures of 27.5°C in the daytime and 13°C during the night-time for 1 week, after which the daytime and night-time temperatures were increased each week by 1.5°C and 1.0°C, respectively, for weeks 2–8 (9 males, 5 females) (Fig. 1B, bottom). Therefore, the summer warming group was subjected to increases in average daytime and night-time temperatures as well as a modest increase in daily temperature variation (18°C versus 14.5°C difference between day and night) compared with the spring control group. Each experimental treatment was replicated twice, with multiple cages per replicate.

Fig. 1.

Experimental treatment groups were modelled from realistic patterns of diel temperature variation and summer warming in the native environment of deer mice. (A) Environmental temperatures in Nebraska at a site where deer mice were once abundant but have been largely displaced by white-footed mice. Ambient temperature (Ta) was recorded every 2 h at ground level under shelter in microhabitat where there is typically strong trapping success for Peromyscus mice. (B) Patterns of temperature variation used in the current study in the ‘spring control group’ (top) and the ‘summer warming group’ (bottom). See Materials and Methods for additional details.

Fig. 1.

Experimental treatment groups were modelled from realistic patterns of diel temperature variation and summer warming in the native environment of deer mice. (A) Environmental temperatures in Nebraska at a site where deer mice were once abundant but have been largely displaced by white-footed mice. Ambient temperature (Ta) was recorded every 2 h at ground level under shelter in microhabitat where there is typically strong trapping success for Peromyscus mice. (B) Patterns of temperature variation used in the current study in the ‘spring control group’ (top) and the ‘summer warming group’ (bottom). See Materials and Methods for additional details.

Shortly before exposures began, mice were implanted with temperature loggers and/or thermosensitive passive integrated transponder (PIT) tags (as described below) to provide subcutaneous temperature measurements that we used as a proxy for Tb. After full recovery from these implantations, mice were moved in pairs into standard mouse cages that contained in-cage running wheels (Starr life Sciences, Oakmont, PA, USA), and cages were moved into a rodent incubator with automated temperature and photoperiod control (model RIS28SSD, Powers Scientific, Pipersville, PA, USA). Mice were first held at standard holding temperature (25°C) for at least 1 week to become accustomed to the incubator and running wheel. Mice were then subjected to diel temperature cycles as described above. The transition from daytime to night-time temperatures began at lights on/off, taking approximately 10 min. In mice implanted with temperature loggers (see below), Tb was measured every 10 min. From these data, we determined the average value at each time of day across the last 5 days of each week for each mouse. We also determined the daytime average Tb, the night-time average Tb, the night-time minimum Tb, and the average magnitude of daily Tb variation for each week for each mouse. Wheel rotations were measured every hour using digital counters, recorded using a GoPro camera set, and a red LED light was used to enable visualization during the dark phase. We thus determined the average number of total daily wheel rotations as an index of daily activity for each week. Food and water consumption over 24 h were measured 4–7 days of each week (which required that we weigh and replace food and water at 14:30 h local time), and the average daily value for each week is reported here. Body mass was measured on days 3 and 7 of each week, concurrent with cage changes (at 14:30 h local time), and the average value for each week is reported here. Otherwise, husbandry conditions were the same as those described above. After completing the 8 week temperature exposures, mice were used for respirometry measurements before being sampled for tissues, as described below.

Implantation of temperature loggers and PIT tags

A subset of mice in the spring control group (7 males, 3 females) and summer warming group (5 males, 5 females) were implanted with temperature loggers to facilitate the continuous recording of Tb during exposures (DST nano-T, Star-Oddi, Garðabær, Iceland). Anaesthesia was induced with 3% isoflurane in O2, and a surgical plane of anaesthesia was maintained with 1.5–2% isoflurane. The ventral surface of the neck was shaved and disinfected with iodine solution, and the mouse was transferred to a heating pad to maintain Tb. A 2–3 cm incision was made along the midline of the neck, the skin was carefully separated by blunt dissection from the underlying muscle to access the subcutaneous interscapular region, and the logger was inserted between the scapulae. The incision was then closed using an interrupted subcuticular suture (6-0 Vicryl). These mice were also implanted with thermosensitive PIT tags shortly after logger implantation (same location as described below), and the animal was removed from anaesthesia, and recovered in a new clean cage (Envigo Teklad 7070C Certified Diamond Dry) that was warmed with a heating pad. The surgery took approximately 20 min to complete. Carprofen (5 mg kg−1 in saline solution) was given by subcutaneous injection during surgery as well as 24 and 48 h after surgery for analgesia. Mice were closely monitored for 7 days post-surgery before experimental treatments began, by which time the incision site had healed well, and mice had resumed their normal behaviour.

All remaining mice were implanted with thermosensitive PIT tags (micro LifeChips with Bio-therm technology; Destron Fearing, Dallas, TX, USA) to facilitate individual identification and the measurement of Tb during respirometry (see below). Mice were lightly anaesthetized using 2% isoflurane and given a subcutaneous injection of 2 mg kg−1 ketoprofen in saline for analgesia. The fur on the left side of the abdomen was cleaned using 70% ethanol, and the PIT tag was injected subcutaneously into the abdominal region near the leg using a sterile trocar (Destron Fearing). Mice were then removed from anaesthesia and returned to their cage to recover. The implantation took approximately 5 min to complete, and mice were monitored for 7 days to ensure complete recovery before experimental treatments.

Respirometry measurements

We conducted two sets of respirometry measurements in each treatment group at the completion of the 8 week exposures. The first set of measurements were made in all mice, in which we examined the metabolic and thermoregulatory responses to acute warming at rest, using respirometry and whole-animal plethysmography techniques we have used in previous studies (Ivy and Scott, 2017a,b). Measurements were made in a cylindrical respirometry chamber (from Data Sciences International, St Paul, MN, USA) that was ∼700 ml in volume and was connected to a sealed reference chamber via a pneumotachograph (for measuring breathing frequency as described below). The respirometry chamber contained a metal platform that elevated mice above any faeces and urine that accumulated during the experiment. The entire respirometry apparatus was placed inside a Peltier-effect drop-in cabinet (Sable Systems International, Las Vegas, NV, USA) in which temperature was controlled using a Peltier-effect temperature controller (PELT-5, Sable Systems International). Dry incurrent air flowing at 600 ml min−1, controlled using precision flow meters (Sierra Instruments, Monterey, CA, USA) and a mass flow controller (MFC-4, Sable Systems International), was first passed through a stainless-steel coil inside the Peltier cabinet and then supplied to the respirometry chamber. Gas composition was measured continuously in both incurrent and excurrent airflows subsampled at 200 ml min−1. The O2 fraction of incurrent air (which was dry and free of CO2) was measured using a galvanic fuel cell O2 analyser (FC-10, Sable Systems International). For excurrent air, water vapour pressure was measured using a thin-film capacitive water vapour analyser (RH-300, Sable Systems International), the gas stream was then dried with pre-baked Drierite, O2 fraction was measured as above, and CO2 fraction was measured using an infrared CO2 analyser (CA-10, Sable Systems International). Measurements of O2 and CO2 fractions and water vapour were used to calculate rates of O2 consumption (O2) and evaporative water loss (EWL) using the appropriate equations with a known incurrent flowrate as described previously (Lighton, 2008). The ambient temperature (Ta) in the chamber was continuously recorded with a thermocouple (TC-2000, Sable Systems International). Breathing frequency (f­R) was measured using whole-body plethysmography from pressure oscillations in the animal chamber relative to the reference chamber using a differential pressure transducer (DP103, Validyne, Northridge, CA, USA). Tb was measured every minute using a PIT tag reader (Global Pocket Reader, Destron Fearing). Data were acquired using a PowerLab 16/32 and LabChart 8 Pro software (ADInstruments, Colorado Springs, CO, USA).

The measurements of metabolic and thermoregulatory responses to acute warming began by placing mice in the chamber, with Ta initially set to 26°C. Mice were allowed to adjust for at least 60 min until stable patterns of O2 consumption, water loss and breathing were observed. Mice were then kept at 26°C for an additional 20 min, followed by exposure to stepwise increases in temperature of 2°C every 20 min, ending with 20 min at 40°C. Measurements of O2, EWL rate, fR and Tb were determined using the last 10 min at each Ta and when responses had reached a steady state. After measurements were completed, mice were returned to their home cage in the appropriate chronic temperature exposure, held for 2 more days, and either used for a second set of measurements or sampled (see below).

The second set of measurements were made in a subset of mice in the spring control group (7 males, 3 females) and summer warming group (5 males, 2 females) to determine thermogenic capacity using similar methods to those we have used previously (Tate et al., 2017). These measurements were conducted ∼48 h after the first set of measurements above. Mice were placed in a respirometry chamber (∼600 ml) inside a temperature-controlled cabinet held at −5°C. Heliox (21% O2, 79% He) passed through a stainless-steel coil within the cabinet and was then delivered to the metabolic chamber at 1000 ml min−1, with incurrent flowrate controlled using precision flow meters (Sierra Instruments) and a mass flow controller (MFC-2, Sable Systems International). Mice were exposed to these conditions for ∼15 min and O2 consumption rate was measured throughout as described above. Thermogenic capacity was defined as the maximal O2 consumption rate (O2,max) achieved over a 30 s period during the trial, which generally occurred after ∼2–6 min in the chamber. Tb was measured before and at the end of the trial, confirming that Tb was reduced by at least 5°C at the end of the trial.

Euthanasia and sampling

Euthanasia was performed with an overdose of gaseous isoflurane followed by decapitation. Blood samples were collected, haemoglobin content was measured using Drabkin's reagent (according to the instructions from the manufacturer, Sigma-Aldrich, Oakville, ON, Canada), and blood was centrifuged in capillary tubes at 12,700 g for 5 min to measure haematocrit. Interscapular brown adipose tissue (iBAT) and one gastrocnemius muscle were dissected, weighed, frozen in liquid N2 and stored at −80°C. The entire lungs, heart ventricles, liver, both kidneys, stomach and small intestine were dissected and weighed.

Western blotting

Protein content of uncoupling protein 1 (UCP1) and citrate synthase (CS) were measured in iBAT by western blotting as previously described (Hayward et al., 2022; Robertson et al., 2019). Tissues were first powdered using a liquid N2-cooled mortar and pestle, and ∼20 mg of powdered tissue was homogenized with a PowerGen 125 homogenizer (Thermo Fisher Scientific, Mississauga, ON, Canada) in cold RIPA buffer containing 150 mmol l−1 NaCl, 50 mmol l−1 Tris, 1% Triton X-100, 0.5% deoxycholic acid, 0.l% SDS and 1 tablet/10 ml of cOmplete™ Mini Protease Inhibitor Cocktail (Sigma-Aldrich) at pH 8.0. Samples were homogenized 3 times for 30 s each, interspersed by periods of at least 3 min on ice between each homogenization. Samples were then centrifuged at 10,000 g at 4°C for 20 min and the supernatant was retained. Proteins in the supernatant were denatured at 95°C for 5 min in Laemmli buffer containing 10% 2-mercaptoethanol (Bio-Rad, Mississauga, ON, Canada). Total protein concentration was determined using a detergent compatible (DC) protein assay kit (Bio-Rad); 2 µg of total protein was loaded into each lane of pre-cast 12% sodium dodecyl sulfate-polyacrylamide gels (Bio-Rad) and separated for ∼30 min at 100 V, followed by ∼45 min at 150 V using a Mini-Protein Tetra system (Bio-Rad). Proteins were then transferred from the gel to polyvinylidene difluoride (PVDF) membranes for 7 min at 25 V and 2.5 A using a Trans Blot Turbo Transfer System (Bio-Rad). Membranes were cut at roughly 40 kDa between the expected location of protein bands for UCP1 (30 kDa) and CS (52 kDa). Membranes were blocked overnight at 4°C using 1% bovine serum albumen (BSA; Sigma-Aldrich) in phosphate-buffered saline containing Tween-20 (PBST; 1.5 mmol l−1 NaH2PO4, 8.1 mmol l−1 Na2HPO4, 145.5 mmol l−1 NaCl, 0.05% Tween-20 at pH 7.4). Membranes were then incubated for 1 h at 4°C with rabbit IgG primary antibodies in PBST containing 1% BSA. Specifically, the bottom half of each membrane was incubated with primary antibody against UCP1 (1:500 dilution; UCP11-A, Alpha Diagnostics International, San Antonio, TX, USA) and the top half with primary antibody against CS (1:500 dilution; PA5-22126, Invitrogen, Thermo Fisher Scientific). Membranes were then rinsed 3 times with PBST for 10 min each, and then incubated at 4°C for 1 h with goat anti-rabbit IgG secondary antibody conjugated with horseradish peroxidase (1:5000 dilution; 31466, Invitrogen, Thermo Fisher Scientific). Membranes were developed in Clarity ECL Substrate (Bio-Rad) and band intensity was detected by chemiluminescence using a ChemiDoc MP Imaging System and Image Lab Software (Bio-Rad). A common protein sample was included in each gel to account for any variation in membrane protein transfer efficiency, and target protein abundance in all other samples was determined relative to this common sample. Target protein abundance in each sample was also normalized to total protein, determined by Coomassie Blue staining. Protein abundance data reported here are expressed relative to the mean value in the spring control group.

Enzyme activity assays

Gastrocnemius muscle was powdered using a liquid N2-cooled mortar and pestle, and ∼20 mg of powdered muscle was homogenized using a glass tissue grinder in 20 volumes of homogenization buffer (100 mmol l−1 KH2PO4, 5 mmol l−1 EDTA and 0.1% Triton X-100 at pH 7.2). Cytochrome c oxidase (COX) and 3-hydroxyacyl-CoA dehydrogenase (HOAD) activities were assayed shortly after homogenization, and CS and lactate dehydrogenase (LDH) activities were measured after storage of homogenate at −80°C. Activity was assayed at 37°C by measuring the change in absorbance over time (CS, 412 nm; COX, 550 nm; HOAD and LDH, 340 nm) under the following conditions (in mmol l−1 unless otherwise stated): CS – 40 Tris, 0.5 oxaloacetate, 0.23 acetyl-CoA, 0.1 5,5′-dithio-bis-(2-nitrobenzoic acid) (DTNB), pH 8.0; COX – 100 KH2PO4, 0.15 reduced cytochrome c, pH 7.0; HOAD – 100 triethylamine hydrochloride, 0.1 acetoacetyl-CoA, 0.28 NADH, pH 7.0; LDH – 40 Tris, 0.28 NADH, 3 pyruvate, pH 7.4. All enzyme assays were run in triplicate. Enzyme activity was determined by subtracting the control rate (measured without key substrate) from the rate measured in the presence of all substrates. Preliminary experiments verified that substrate concentrations were saturating.

Statistics

Linear mixed-effects models were performed using the lme4 package (Bates et al., 2015) in R (version 4.2.0) (http://www.R-project.org/). We tested for fixed effects and interactions of experimental treatment group, sex and week for daytime average Tb, night-time average Tb, night-time minimum Tb, daily Tb variation, daily wheel rotations, food and water consumption, and body mass. For the more frequent measures of Tb (measured every 10 min) and wheel running (measured every hour), we tested for fixed effects and interactions of sex, week and time of day within each treatment group. For measurements of the metabolic and thermoregulatory responses to acute warming, we tested for fixed effects and interactions of experimental treatment group, sex and Ta. In all the above cases, mouse family and individual/cage (nested within family) were also included as random effects (individual/cage thus accounted for repeated measures). For measurements of thermogenic capacity and of organs/tissues, we tested for fixed effects and interactions of experimental treatment group and sex, and included mouse family as a random effect. For the statistical analysis of O2, EWL rate and organ/tissue mass, body mass was also included as a covariate and tests were carried out on the absolute values of these traits; however, these data are reported relative to body mass as is conventional in the literature. The lmerTest package was used to generate ANOVA tables containing F-statistics and P-values for fixed effects and interactions (Kuznetsova et al., 2017), the full results of which are reported in Tables S1–S5. ANOVA P-values for key findings are reported in the Results, in many cases along with fixed-effect estimates (β) and s.e. to illustrate effect size. When main effects or interactions were significant, we often used the package emmeans (version 1.7.4; https://CRAN.R-project.org/package=emmeans) to make post hoc pairwise comparisons using the Tukey method (we did not perform post hoc comparisons for the repeated measurements of Tb and wheel running made throughout the day, because of the large volume of data). P<0.05 was considered to be significant.

Summer warming led to a dysregulation of Tb

There was a strong diel cycle in Tb in the spring control group (Fig. 2A), in which Tb rose during the night-time active phase and declined during the daytime inactive phase (main effect of time of day, P<0.001), with highest Tb shortly after lights off at roughly 22:00 h local time (β±s.e., 0.66±0.53°C) and lowest Tb shortly after lights on from 07:40 h to 09:50 h local time (−1.31±0.53°C for each). There was a transient rise in Tb at ∼14:30 h local time when all health checks, cage changes and measurements were done (see Materials and Methods for details). There were some significant differences in Tb between weeks (main effect of week, P<0.001), but the absolute variation in Tb between weeks was generally modest, and the overall pattern of daily Tb variation was similar across weeks (week×time, P=0.999). Females had greater Tb than males overall (β±s.e., 0.41±0.44°C; main effect of sex, P<0.001) and a more pronounced diel cycle in Tb (sex×time, P=0.003).

Fig. 2.

Summer warming altered the diel cycle of body temperature (Tb).Tb was measured every 10 min using temperature loggers implanted subcutaneously in the interscapular area. The average diel Tb cycle is shown for each week as means±s.e.m. in (A) the spring control group (N=10 mice) and (B) the summer warming group (N=10 mice). The grey shaded area reflects the dark night-time phase when mice are active, and the white areas reflect the light daytime phase of inactivity. Ta during daytime and night-time in each week is shown in Fig. 1.

Fig. 2.

Summer warming altered the diel cycle of body temperature (Tb).Tb was measured every 10 min using temperature loggers implanted subcutaneously in the interscapular area. The average diel Tb cycle is shown for each week as means±s.e.m. in (A) the spring control group (N=10 mice) and (B) the summer warming group (N=10 mice). The grey shaded area reflects the dark night-time phase when mice are active, and the white areas reflect the light daytime phase of inactivity. Ta during daytime and night-time in each week is shown in Fig. 1.

In contrast, there were major changes in the diel cycle of Tb between weeks in the summer warming group (Fig. 2B), which drove a significant main effect of week (P<0.001) and a significant week×time interaction (P<0.001). Daytime Tb started increasing at week 4, and the increase extended across the daytime phase in weeks 6–8 (Fig. 2B). Surprisingly, there were also pronounced drops in Tb during the hours after the transition to cooler night-time temperatures in weeks 6–8 (albeit with some Tb recovery as the night-time progressed), even though night-time Ta was greater in weeks 6–8 than in week 1. As a result, there was a complete reversal in the diel Tb cycle in weeks 6–8, with much greater Tb during the daytime inactive phase than during the night-time active phase. In week 8, for example, Tb reached its peak towards the end of the light phase (β±s.e., 5.98±0.65°C at 20:00 h local time), and it then dropped sharply with cooling Ta after lights off and reached its nadir at 22:10 h (−0.46±0.65°C). Sex did not alter the strong changes in the diel Tb cycle between weeks (sex×week×time, P=0.999).

The appreciable differences in the diel Tb cycle between the spring control group and the summer warming group were reflected in average and minimum measures of Tb and in daily Tb variation (Fig. 3). There were significant treatment×week interactions (all P<0.001) for average daytime Tb, average night-time Tb, minimum night-time Tb and the average magnitude of daily Tb variation. Average daytime Tb did not change between weeks in the spring control group, but it increased progressively in the summer warming group from week 4 onwards, peaking at week 8 (β±s.e., 2.45±0.25°C) (Fig. 3A). Average night-time Tb and minimum night-time Tb were also stable across weeks in the spring control group but decreased significantly in the summer warming group in week 8 (−1.27±0.29°C) and weeks 7 (−2.10±0.55°C) and 8 (−2.28±0.55°C), respectively (Fig. 3B,C). The magnitude of daily Tb variation was stable at <3°C in the spring control group but increased appreciably in the summer warming group from week 6 onwards, peaking at week 8 (4.49±0.51°C) (Fig. 3D). Females tended to have greater average daytime Tb than males (0.44±0.19°C; sex effect, P<0.001), but not average or minimum night-time Tb (sex effects, P=0.855 and 0.992) or average magnitude of daily Tb variation (sex effect, P=0.459), and all the treatment×week interactions were similar between sexes (treatment×week×sex, all P≥0.408) (Fig. S1).

Fig. 3.

Summer warming altered average Tb and Tb variation. (A) Average daytime Tb, (B) average night-time Tb, (C) minimum night-time Tb and (D) the magnitude of daily Tb variation in the spring control group (grey circles; N=10 mice) and the summer warming group (black squares; N=10 mice), shown as means±s.e.m. Data were analysed statistically using linear mixed-effects models followed by pairwise comparisons using the Tukey method. *Significant pairwise difference between treatments within each week (P<0.05). Within the summer warming group, weeks not sharing a letter are significantly different in pairwise comparisons (P<0.05) (there were no significant differences between weeks in the spring control group). Ta during daytime and night-time in each week is shown in Fig. 1.

Fig. 3.

Summer warming altered average Tb and Tb variation. (A) Average daytime Tb, (B) average night-time Tb, (C) minimum night-time Tb and (D) the magnitude of daily Tb variation in the spring control group (grey circles; N=10 mice) and the summer warming group (black squares; N=10 mice), shown as means±s.e.m. Data were analysed statistically using linear mixed-effects models followed by pairwise comparisons using the Tukey method. *Significant pairwise difference between treatments within each week (P<0.05). Within the summer warming group, weeks not sharing a letter are significantly different in pairwise comparisons (P<0.05) (there were no significant differences between weeks in the spring control group). Ta during daytime and night-time in each week is shown in Fig. 1.

Summer warming reduced activity and body mass

Voluntary wheel running, which was used as a metric of routine activity, was almost exclusively restricted to the night-time active phase (time of day effects, P<0.001) (Fig. 4). There was no significant variation in wheel running across weeks in the spring control group (week effect, P=0.481; week×time, P=0.999) (Fig. 4A). In contrast, there was a significant effect of week on wheel running in the summer warming group (P<0.001), with highest levels in weeks 3–5 but lowest levels in week 8 (Fig. 4B). There was also a significant week×time interaction in the summer warming group (P=0.010), likely driven by week 8, in which levels of activity were lowest in the early hours after the transition to cooler night-time Ta but increased thereafter (Fig. 4B). This variation in hourly measures of wheel running was reflected in measures of total daily wheel running (Fig. 5), for which there was a significant effect of week (P=0.029) that was driven by significant declines in wheel running in the summer warming group in week 8 (β±s.e., −14,213±13,381 rotations day−1). There was no significant effect of sex on total daily wheel running (sex effect, P=0.379) and sex did not alter the effects of the treatment (treatment×sex and treatment×week×sex, all P≥0.935).

Fig. 4.

Summer warming altered wheel running across the daily cycle. The number of wheel revolutions was measured every hour for each cage, each of which contained 2 mice. The average diel cycle of wheel running is shown as means±s.e.m. in (A) the spring control group (N=8 cages) and (B) the summer warming group (N=5 cages). The grey shaded area reflects the dark night-time phase when mice are active, and the white areas reflect the light daytime phase of inactivity. Ta during daytime and night-time in each week is shown in Fig. 1.

Fig. 4.

Summer warming altered wheel running across the daily cycle. The number of wheel revolutions was measured every hour for each cage, each of which contained 2 mice. The average diel cycle of wheel running is shown as means±s.e.m. in (A) the spring control group (N=8 cages) and (B) the summer warming group (N=5 cages). The grey shaded area reflects the dark night-time phase when mice are active, and the white areas reflect the light daytime phase of inactivity. Ta during daytime and night-time in each week is shown in Fig. 1.

Fig. 5.

Summer warming reduced total daily wheel running. Average daily wheel rotations for each cage (each of which contained 2 mice) were measured each week for the spring control group (grey circles; N=8 cages) and the summer warming group (black squares; N=5 cages), and data are shown as means±s.e.m. Data were analysed statistically using linear mixed-effects models followed by pairwise comparisons using the Tukey method. Within the summer warming group, weeks not sharing a letter are significantly different in pairwise comparisons (P<0.05) (there were no significant differences between weeks in the spring control group). Ta during daytime and night-time in each week is shown in Fig. 1.

Fig. 5.

Summer warming reduced total daily wheel running. Average daily wheel rotations for each cage (each of which contained 2 mice) were measured each week for the spring control group (grey circles; N=8 cages) and the summer warming group (black squares; N=5 cages), and data are shown as means±s.e.m. Data were analysed statistically using linear mixed-effects models followed by pairwise comparisons using the Tukey method. Within the summer warming group, weeks not sharing a letter are significantly different in pairwise comparisons (P<0.05) (there were no significant differences between weeks in the spring control group). Ta during daytime and night-time in each week is shown in Fig. 1.

Summer warming also led to a decrease in body mass (Fig. 6A). There was a significant treatment×week interaction (P<0.001) for body mass, which increased slightly in week 8 in the spring control group but decreased in weeks 7 and 8 in the summer warming group (β±s.e., −3.30±0.70 and −3.88±0.71 g, respectively) (Fig. 6A). Variation in body mass coincided with variation in food consumption (Fig. 6B), as reflected by a significant treatment×week interaction (P=0.042), driven by declines in the summer warming group in week 8 (−0.15±0.07 g food day−1). The summer warming group also exhibited different patterns of water consumption to the spring control group (treatment effect, P=0.033; treatment×week, P<0.001), increasing water consumption from week 6 onwards, peaking at week 8 (0.66±0.20 ml water day−1) (Fig. 6C). There were no significant overall effects of sex on body mass, food consumption or water consumption (sex effects, all P≥0.254), and sex did not alter the effects of treatment on these traits (treatment×sex, all P≥0.326; treatment×week×sex, all P≥0.254).

Fig. 6.

Summer warming reduced body mass and daily food consumption, and increased daily water consumption. Body mass (A) was the average of measurements on days 3 and 7 of each week, and food (B) and water consumption (C) over 24 h were the average of measurements over days 4–7 of each week for the spring control group (grey circles; N=16 mice, 8 cages, and 8 cages, respectively) and the summer warming group (black squares; N=10 mice, 5 cages, and 5 cages), and data are shown as means±s.e.m. Data were analysed statistically using linear mixed-effects models followed by pairwise comparisons using the Tukey method. *Significant pairwise difference between treatments within each week (P<0.05). Within each treatment group, weeks not sharing a letter are significantly different in pairwise comparisons (P<0.05) (there were no significant differences in food or water consumption between weeks in the spring control group). Ta during daytime and night-time in each week is shown in Fig. 1.

Fig. 6.

Summer warming reduced body mass and daily food consumption, and increased daily water consumption. Body mass (A) was the average of measurements on days 3 and 7 of each week, and food (B) and water consumption (C) over 24 h were the average of measurements over days 4–7 of each week for the spring control group (grey circles; N=16 mice, 8 cages, and 8 cages, respectively) and the summer warming group (black squares; N=10 mice, 5 cages, and 5 cages), and data are shown as means±s.e.m. Data were analysed statistically using linear mixed-effects models followed by pairwise comparisons using the Tukey method. *Significant pairwise difference between treatments within each week (P<0.05). Within each treatment group, weeks not sharing a letter are significantly different in pairwise comparisons (P<0.05) (there were no significant differences in food or water consumption between weeks in the spring control group). Ta during daytime and night-time in each week is shown in Fig. 1.

Summer warming led to plastic adjustments in thermoregulatory physiology

The patterns of variation during the 8 week treatments showed that mice in the summer warming group may be heat stressed during the daytime but less able to maintain Tb and activity at cooler night-time temperatures. We reasoned that this pattern of variation could be explained by a thermoregulatory trade-off in which plastic responses to daytime heat come at the cost of reducing the ability to cope with cold. We examined this possibility by assessing the metabolic and thermoregulatory responses to acute heat exposure at rest, as well as thermogenic capacity during acute cold exposure.

Measurements of metabolism and thermoregulatory physiology during acute warming revealed several differences between treatment groups (Fig. 7). In general, O2, EWL rate, fR and Tb were lowest at intermediate Ta (∼28–34°C depending on the trait) and increased as Ta reached or approached 40°C (Ta effect, P<0.001 for all). However, the summer warming group had lower O2 (β±s.e., −0.54±0.29 ml O2 min−1; treatment effect, P=0.009), EWL rate (−2.4±1.3 ml water min−1; P<0.001), fR (−25±16 breaths min−1; P=0.006) and Tb (−1.9±0.6°C; P<0.001) across a range of Ta up to 36 or 38°C. These traits converged with values in the spring control group at a Ta of 40°C (treatment×Ta, P=0.006 for EWL rate and P=0.008 for Tb). Males were somewhat more responsive to increases in Ta than females for O2 and fR (Ta×sex, P=0.021 and 0.017), but sex did not alter the effects of treatment on any of the four measured traits (treatment×sex, all P≥0.547; treatment×week×sex, all P≥0.133) (Fig. S2).

Fig. 7.

Summer warming reduced resting O2 consumption rate, evaporative water loss rate, breathing frequency and Tb. Rates of resting O2 consumption (O2; A) and evaporative water loss (EWL; B), breathing frequency (fR; C) and Tb (D) were measured across a range of Ta for the spring control group (grey circles; N=22 mice) and the summer warming group (black squares; N=13 mice), and data are shown as means±s.e.m. Note, for one individual in each group, technical issues precluded fR measurements at a subset of Ta (26 and 38°C in the spring control group and 36°C in the summer warming group). Data were analysed statistically using linear mixed-effects models followed by pairwise comparisons using the Tukey method. O2 and EWL rate are shown relative to body mass here, but absolute rates were used for statistical analyses. *Significant pairwise differences between treatment groups within each Ta (P<0.05). Within each treatment group, Ta not sharing a letter are significantly different in pairwise comparisons (P<0.05).

Fig. 7.

Summer warming reduced resting O2 consumption rate, evaporative water loss rate, breathing frequency and Tb. Rates of resting O2 consumption (O2; A) and evaporative water loss (EWL; B), breathing frequency (fR; C) and Tb (D) were measured across a range of Ta for the spring control group (grey circles; N=22 mice) and the summer warming group (black squares; N=13 mice), and data are shown as means±s.e.m. Note, for one individual in each group, technical issues precluded fR measurements at a subset of Ta (26 and 38°C in the spring control group and 36°C in the summer warming group). Data were analysed statistically using linear mixed-effects models followed by pairwise comparisons using the Tukey method. O2 and EWL rate are shown relative to body mass here, but absolute rates were used for statistical analyses. *Significant pairwise differences between treatment groups within each Ta (P<0.05). Within each treatment group, Ta not sharing a letter are significantly different in pairwise comparisons (P<0.05).

Thermogenic capacity, measured as the maximal rate of O2 consumption during acute exposure to −5°C in heliox (i.e. thermogenic O2,max), was reduced by summer warming (Fig. 8A). There was a significant treatment effect on thermogenic capacity (P=0.014), in which mice in the summer warming group had capacities only ∼47% on average of those in the spring control group (β±s.e., −4.6±2.2 ml min−1), but there was no significant effect of sex on this trait (P=0.278). This variation in thermogenic capacity between treatment groups was associated with parallel variation in the mass of iBAT, a key thermogenic tissue in small mammals (−133±35 mg; treatment effect, P<0.001) (Fig. 8B). Not only was iBAT smaller in the summer warming group but it also contained a lower abundance of UCP1 (−0.62±0.22 arbitrary units, a.u.; treatment effect, P=0.003) (Fig. 8C). The abundance of CS (a common marker of mitochondrial volume) was similar between treatment groups (treatment effect, P=0.826), such that the summer warming group had a lower ratio of UCP1 to CS (−0.77±0.21; P<0.001) (Table 1; Fig. S3). However, there were no effects of treatment on the mass or activity of metabolic enzymes (CS, COX, HOAD or LDH) in the gastrocnemius (Table 1). The mass of the stomach (−142±48 mg; P=0.015) and kidneys (−35±15 mg; P=0.049) was reduced in the summer warming group, but treatment had no significant effect on the mass of the lungs, heart ventricles, liver or small intestine (treatment effects, all P≥0.400) (Table 1). Haematocrit was reduced (−5.7±2.5%; P=0.007) and mean cell haemoglobin content was increased (2.6±3.7 g dl−1; P=0.029) in the summer warming group, but the effect of treatment on blood haemoglobin content was not significant (P=0.066) (Table 1).

Fig. 8.

Summer warming reduced thermogenic capacity and induced phenotypic changes in thermogenic tissues. (A) Thermogenic capacity (measured as the maximal rate of O2 consumption during acute cold exposure, O2,max), (B) mass of interscapular brown adipose tissue (iBAT) and (C) abundance of uncoupling protein 1 (UCP1) in iBAT (in arbitrary units, a.u., expressed relative to the average value in the spring control group; see Fig. S3 for representative western blot). Data for the spring control group (grey circles; N=10, 20 and 16 mice, respectively) and the summer warming group (black squares; N=7, 13 and 12 mice) are shown with bars indicating means±s.e.m. and symbols representing individual values for males (♂) and females (♀). Data were analysed statistically using linear mixed-effects models. *Significant effect of treatment (P<0.05).

Fig. 8.

Summer warming reduced thermogenic capacity and induced phenotypic changes in thermogenic tissues. (A) Thermogenic capacity (measured as the maximal rate of O2 consumption during acute cold exposure, O2,max), (B) mass of interscapular brown adipose tissue (iBAT) and (C) abundance of uncoupling protein 1 (UCP1) in iBAT (in arbitrary units, a.u., expressed relative to the average value in the spring control group; see Fig. S3 for representative western blot). Data for the spring control group (grey circles; N=10, 20 and 16 mice, respectively) and the summer warming group (black squares; N=7, 13 and 12 mice) are shown with bars indicating means±s.e.m. and symbols representing individual values for males (♂) and females (♀). Data were analysed statistically using linear mixed-effects models. *Significant effect of treatment (P<0.05).

Table 1.

Tissue phenotypes

Tissue phenotypes
Tissue phenotypes

The physiological impacts of chronic exposure to warming temperatures are poorly understood in many endothermic species (Levesque et al., 2016; Mitchell et al., 2018). Here, we show that seasonal warming simulating the transition from spring to mid-summer leads to disruptions in thermoregulation that may have detrimental health impacts on deer mice. Summer warming led to a dysregulation of Tb that culminated in a complete reversal of the diel pattern of Tb variation, with Tb rising to extreme highs during the day and dropping to extreme lows at night (Figs 2 and 3). This reduced ability to maintain Tb at cool night-time temperatures was associated with reductions in thermogenic capacity and the mass and UCP1 content of iBAT (Fig. 8), which likely arose as part of a heat acclimation response to summer warming. Reductions in Tb likely constrained normal behaviour during the night-time active phase, in association with reductions in voluntary wheel running, food consumption and body mass (Figs 46). These findings suggest that thermoregulatory trade-offs may underlie the effects of warming summer temperatures on small nocturnal mammals, in which acute exposure and/or acclimatization to daytime heat may limit the ability to remain active and perform behaviours important for fitness at cooler night-time temperatures.

The exposure of deer mice in the summer warming group to daytime heat was likely a strong stimulus for heat acclimation. Daytime Ta rose to 35–38°C in weeks 6–8 (Fig. 1), which appears to be hotter than the TNZ of deer mice (Brower and Cade, 1966; McNab and Morrison, 1963) (Fig. 7). This was associated with pronounced increases in daytime Tb (Figs 2 and 3), consistent with previous results in domestic mice and rats exposed to prolonged heat stress (Kuroshima et al., 1982; Sareh et al., 2011). Summer warming also led to physiological adjustments that are typical of heat acclimation in other species, such as reductions in resting metabolic rate, which can be associated with reductions in metabolic heat generation (Guo et al., 2020; Shido et al., 1994) (Fig. 7). It is possible that heat acclimation induced other changes to help deer mice cope with high daytime temperatures, including an increased capacity for evaporative heat dissipation (Horowitz, 2001; Sareh et al., 2011). However, the ability of such mechanisms to attenuate the rise in Tb during the day may be limited, based on our observation that daytime Tb always exceeded Ta (Figs 2 and 3). Indeed, a recent study of desert species of Peromyscus and other cricetids suggested that they have a limited capacity for evaporative cooling, potentially because saliva spreading (the dominant means of EWL in rodents) is a far less effective means of evaporative cooling than panting (Ramirez et al., 2022). Therefore, the later stages of summer warming likely imposed significant heat stress on deer mice.

Daytime heat exposure likely contributed to the pronounced drop in Tb at cooler night-time temperatures (Figs 2 and 3). Deer mice exhibit natural diel variation in Tb like many other nocturnal rodents, characterized by higher Tb during the night-time active phase (Beaudry and McClelland, 2010; Weinert and Waterhouse, 1998). However, this diel pattern of Tb variation was reversed in the later stages of summer warming, when the transition to cool night-time temperatures led to a transient period of hypothermia (Tb ∼34°C). In domestic mice acclimated to 25°C, a single acute bout of heat exposure followed by a return to cooler Ta is known to induce hypothermia, and the magnitude and duration of hypothermia depends on how high Tb rises during heat exposure (Leon et al., 2005; Wilkinson et al., 1988). Therefore, the pronounced drop in Tb at night in the later stages of summer warming can be at least partly attributed to the residual effects of heat exposure the previous day, potentially compounded by the modest increase in daily temperature variation in the summer warming group compared with the spring control group (18°C versus 14.5°C temperature difference between day and night). How such effects of acute heat exposure might change after prolonged heat acclimation is unclear, but acclimation to warm temperatures within the TNZ (30°C) hinders recovery from hypothermia rather than improving it (Leon et al., 2005). It is therefore possible that plastic responses to prolonged heat exposure in the later stages of summer warming accentuated the initial drop in Tb or slowed Tb recovery overnight. For example, heat acclimation could have enhanced heat dissipation and exaggerated the drop in Tb at cool Ta (Ta that would otherwise be considered relatively mild). Heat acclimation could have also reduced the capacity for heat generation and thus slowed the rate of rewarming to recover Tb at cool Ta overnight.

The latter possibility was supported by our observation that prolonged exposure to daytime heat reduced thermogenic capacity (Fig. 8). This observation is consistent with previous findings in various other small mammals, in which heat acclimation has been shown to reduce thermogenic capacity (Guo et al., 2020; Kuroshima et al., 1982). The reduction in thermogenic capacity observed here was associated with a reduction in the mass and UCP1 content of iBAT (Fig. 8), the primary site of non-shivering thermogenesis, consistent with previous findings in other species (Kuroshima et al., 1982; Saha et al., 2000; Zhang et al., 2012). Whether shivering capacity was also affected by summer warming is unclear, but there were no significant effects on metabolic capacity in the gastrocnemius – a large hindlimb muscle involved in shivering and locomotion (Table 1). Nevertheless, these findings suggest that heat acclimation induces plastic adjustments that might constrain the ability of deer mice to rapidly recover Tb at cool night-time temperatures below the TNZ. Future comparisons between acclimation groups on the effects of acute heat exposure followed by cool recovery on Tb and metabolism (and how the latter relates to thermogenic capacity) would provide greater insight into this possibility.

The prolonged reduction in night-time Tb likely constrained normal behaviour and contributed to reducing activity levels in deer mice (Figs 4 and 5). Running activity was restricted to the night-time when Ta was cooler. Findings of previous studies suggest that running activity was probably not impaired by the direct effects of increases in night-time Ta, which rose to 18–20°C in weeks 6–8 of summer warming. Firstly, in previous studies of deer mice, total distance and time of voluntary wheel running did not vary between ambient temperatures of 3, 10 and 25°C (Chappell et al., 2004). Similarly, voluntary wheel running is only modestly affected by chronic exposure to constant temperatures from ∼20 to 30°C in house mice (Fregly et al., 1957; Vaanholt et al., 2007). Secondly, forced endurance capacity is maintained from ∼15 to 30°C in deer mice, and performance is not impaired until Ta is increased above 35°C (Eizenga et al., 2022). It is more likely that decreases in running activity during the later stages of summer warming resulted from the prolonged decrease in night-time Tb, as reductions in Tb are known to reduce locomotor performance (Rojas et al., 2012). Reductions in Tb could have also reduced the motivation to run and be active. It is also possible that reductions in running activity reinforced reductions in Tb at cool night-time temperatures, given that heat produced by muscles during exercise can make important contributions to whole-animal thermogenesis at Ta below the TNZ (Chappell et al., 2004).

The reduction in body mass in the later stages of summer warming represents another potential adverse health impact of daytime heat exposure (Fig. 6). This reduction in body mass could have been caused by the concurrent decrease in food consumption in the summer warming group, which occurred even though mice had unlimited access to food. Indeed, prolonged heat exposure has frequently been observed to decrease food consumption and/or body mass in several species (Dikstein et al., 1970; Guo et al., 2020; Harikai et al., 2003; Saha et al., 2000; Sareh et al., 2011; Shido et al., 1994; Vaanholt et al., 2007; Yamauchi et al., 1983). However, the drop in food consumption may have also been a response to decreased metabolic demands associated with reductions in running activity, and may not have contributed to the fall in body mass. It is instead possible that prolonged increases in EWL led to desiccation, if the observed increases in water consumption (Fig. 6) were insufficient to meet increases in water demand, such that total body water and body mass were reduced. Either way, reductions in body mass might be expected to reduce the ability of deer mice to invest resources into growth and reproduction, which could have corresponding impacts on fitness in the wild.

The effects of warming temperatures observed here could have important implications for how climate change is impacting deer mice in the wild. Deer mouse populations have declined in several regions of North America, where they are being displaced by white-footed mice (Fiset et al., 2015; Myers et al., 2005, 2009; Roy-Dufresne et al., 2013; Walsh et al., 2016). This is likely true for the deer mouse population studied here, which are from a region of Nebraska adjacent to sites where deer mice used to be abundant but are now scarce. Our data here suggest that the current temperatures during peak summer in this region of Nebraska may lead to a dysregulation of Tb, a reduced ability to remain active and a loss of body mass. These effects of summer heat might be expected to reduce fitness if they decrease the ability to forage and avoid predators or impair growth and reproduction. Daytime temperatures exceeding 30°C are not uncommon in this region in peak summer (NOAA, 2022), so it is possible that deer mice were exposed to similar conditions in the past. The detrimental impacts of summer heat could help explain why deer mice have been unable to outcompete white-footed mice in some parts of their range. Furthermore, such detrimental impacts of summer heat may be accentuated by future increases in summer temperature as climate change intensifies. The thermoregulatory trade-offs that underlie the observed effects of summer heat are likely to be relevant to many other small nocturnal mammals, and may help explain the underlying reasons why the latitudinal distributions of many species are shifting northward in response to climate change.

The authors would like to thank Jay Storz for assisting in the collection of environmental temperature data in Lincoln Nebraska as well as the animal care technicians from the Central Animal Facility at McMaster University and all members of the Scott and McClelland labs for help and support while completing this work.

Author contributions

Conceptualization: G.R.S.; Methodology: L.D.F., O.H.W., E.J.G., G.R.S.; Validation: L.D.F., G.R.S.; Formal analysis: L.D.F.; Investigation: L.D.F., O.H.W., E.J.G.; Resources: G.R.S.; Writing - original draft: L.D.F.; Writing - review & editing: L.D.F., O.H.W., E.J.G., G.R.S.; Supervision: G.R.S.; Funding acquisition: G.R.S.

Funding

This work was supported by a Natural Sciences and Engineering Research Council of Canada (NSERC) Discovery Grants to G.R.S. (RGPIN-2018-05707). O.H.W. was supported by an NSERC Vanier Canada Graduate Scholarship. G.R.S. is supported by the Canada Research Chairs Program.

Data availability

The data that support the findings of this study are publicly available in the Figshare repository at https://doi.org/10.6084/m9.figshare.22148369.v1.

Bao
,
M.-H.
,
Chen
,
L.-B.
,
Hambly
,
C.
,
Speakman
,
J. R.
and
Zhao
,
Z.-J.
(
2020
).
Exposure to hot temperatures during lactation in Swiss mice stunts offspring growth and decreases future reproductive performance of female offspring
.
J. Exp. Biol.
223
,
jeb223560
.
Bates
,
D.
,
Machler
,
M.
,
Bolker
,
B.
and
Walker
,
S.
(
2015
).
Fitting linear mixed-effects models using {lme4}
.
J. Stat. Softw.
56
,
1
-
48
.
Beaudry
,
J. L.
and
McClelland
,
G. B.
(
2010
).
Thermogenesis in CD-1 mice after combined chronic hypoxia and cold acclimation
.
Comp. Biochem. Physiol. Part B Biochem. Mol. Biol.
157
,
301
-
309
.
Bedford
,
N. L.
and
Hoekstra
,
H. E.
(
2015
).
The natural history of model organisms: Peromyscus mice as a model for studying natural variation
.
Elife
4
,
e06813
.
Bonebrake
,
T. C.
,
Rezende
,
E. L.
and
Bozinovic
,
F.
(
2020
).
Climate change and thermoregulatory consequences of activity time in mammals
.
Am. Nat
196
,
45
-
56
.
Brower
,
J. E.
and
Cade
,
T. J.
(
1966
).
Ecology and physiology of Napaeozapus insignis (Miller) and other woodland mice
.
Ecology
47
,
46
-
63
.
Chappell
,
M. A.
(
1985
).
Effects of ambient temperature and altitude on ventilation and gas exchange in deer mice (Peromyscus maniculatus)
.
J. Comp. Physiol. B
155
,
751
-
758
.
Chappell
,
M. A.
,
Garland
,
T.
,
Rezende
,
E. L.
and
Gomes
,
F. R.
(
2004
).
Voluntary running in deer mice: speed, distance, energy costs and temperature effects
.
J. Exp. Biol.
207
,
3839
-
3854
.
Danner
,
R.
,
Coomes
,
C.
and
Derryberry
,
E.
(
2021
).
Simulated heat waves reduce cognitive and motor performance of an endotherm
.
Ecol. Evol.
11
,
2261
-
2272
.
Dikstein
,
S.
,
Kaplanski
,
Y.
,
Koch
,
Y.
and
Sulman
,
F. G.
(
1970
).
The effect of heat stress on body development in rats
.
Life Sci.
9
,
1191
-
1200
.
Eizenga
,
M.
,
Flewwelling
,
L.
,
Warrier
,
T.
and
Scott
,
G.
(
2022
).
Thermal performance curve of endurance running at high temperatures in deer mice
.
J. Exp. Biol
.
226
,
jeb244847
.
Fiset
,
J.
,
Tessier
,
N.
,
Millien
,
V.
and
Lapointe
,
F.-J.
(
2015
).
Phylogeographic structure of the white-footed mouse and the deer mouse, two Lyme disease reservoir hosts in Québec
.
PLoS ONE
10
,
e0144112
.
Fregly
,
M. J.
,
Marshall
,
N. B.
and
Mayer
,
J.
(
1957
).
Effect of changes in ambient temperature on spontaneous activity, food intake and body weight of goldthioglucose-obese and nonobese mice
.
Am. J. Physiol. Content
188
,
435
-
438
.
Glaser
,
E. M.
(
1949
).
Acclimatization to heat and cold
.
J. Physiol
110
,
330
.
Gordon
,
C.
(
2012
).
Thermal physiology of laboratory mice: defining thermoneutrality
.
J. Therm. Biol.
37
,
654
-
685
.
Guo
,
Y.-Y.
,
Hao
,
S.
,
Zhang
,
M.
,
Zhang
,
X.
and
Wang
,
D.
(
2020
).
Aquaporins, evaporative water loss and thermoregulation in heat-acclimated Mongolian gerbils (Meriones unguiculatus)
.
J. Therm. Biol.
91
,
102641
.
Harikai
,
N.
,
Tomogane
,
K.
,
Miyamoto
,
M.
,
Shimada
,
K.
,
Onodera
,
S.
and
Tashiro
,
S.
(
2003
).
Dynamic responses to acute heat stress between 34 C and 38.5 C, and characteristics of heat stress response in mice
.
Biol. Pharm. Bull.
26
,
701
-
708
.
Hayward
,
J. S.
(
1965
).
Microclimate temperature and its adaptive significance in six geographic races of Peromyscus
.
Can. J. Zool.
43
,
341
-
350
.
Hayward
,
L.
,
Robertson
,
C. E.
and
McClelland
,
G. B.
(
2022
).
Phenotypic plasticity to chronic cold exposure in two species of Peromyscus from different environments
.
J. Comp. Physiol. B
192
,
335
-
348
.
Heldmaier
,
G.
(
1989
).
Seasonal acclimatization of energy requirements in mammals: functional significance of body weight control, hypothermia, torpor and hibernation
. In
Energy Transformation in Cells and Animals
(ed. W. Wieser), pp.
130
-
139
.
Thieme Stuttgart
.
Hetem
,
R. S.
,
Fuller
,
A.
,
Maloney
,
S. K.
and
Mitchell
,
D.
(
2014
).
Responses of large mammals to climate change
.
Temperature
1
,
115
-
127
.
Hock
,
R. J.
and
Roberts
,
J. C.
(
1966
).
Effect of altitude on oxygen consumption of deer mice: relation of temperature and season
.
Can. J. Zool.
44
,
365
-
376
.
Horowitz
,
M.
(
2001
).
Heat acclimation: phenotypic plasticity and cues to the underlying molecular mechanisms
.
J. Therm. Biol.
26
,
357
-
363
.
Hu
,
C. K.
and
Hoekstra
,
H. E
. (
2017
).
Peromyscus burrowing: a model system for behavioral evolution
. In
Seminars in Cell & Developmental Biology
, pp.
107
-
114
.
Elsevier
.
Ivy
,
C. M.
and
Scott
,
G. R.
(
2017a
).
Ventilatory acclimatization to hypoxia in mice: methodological considerations
.
Respir. Physiol. Neurobiol.
235
,
95
-
103
.
Ivy
,
C. M.
and
Scott
,
G. R.
(
2017b
).
Control of breathing and ventilatory acclimatization to hypoxia in deer mice native to high altitudes
.
Acta Physiol.
221
,
266
-
282
.
Kuroshima
,
A.
,
Yahata
,
T.
,
Doi
,
K.
and
Ohno
,
T.
(
1982
).
Thermal and metabolic responses of temperature-acclimated rats during cold and heat exposures
.
Jpn. J. Physiol.
32
,
561
-
571
.
Kuznetsova
,
A.
,
Brockhoff
,
P. B.
and
Christensen
,
R. H. B.
(
2017
).
lmerTest package: tests in linear mixed effects models
.
J. Stat. Softw
82
,
1
-
26
.
Leon
,
L. R.
,
DuBose
,
D. A.
and
Mason
,
C. W.
(
2005
).
Heat stress induces a biphasic thermoregulatory response in mice
.
Am. J. Physiol. Integr. Comp. Physiol.
288
,
R197
-
R204
.
Levesque
,
D. L.
,
Nowack
,
J.
and
Stawski
,
C.
(
2016
).
Modelling mammalian energetics: the heterothermy problem
.
Clim. Chang. Responses
3
,
1
-
11
.
Lighton
,
J. R. B
. (
2008
).
Measuring Metabolic Rates: A Manual for Scientists
.
Oxford University Press
.
Lovegrove
,
B.
,
Canale
,
C.
,
Levesque
,
D.
,
Fluch
,
G.
,
Řeháková-Petrů
,
M.
and
Ruf
,
T.
(
2014
).
Are tropical small mammals physiologically vulnerable to Arrhenius effects and climate change?
Physiol. Biochem. Zool.
87
,
30
-
45
.
McCain
,
C. M.
and
King
,
S. R. B.
(
2014
).
Body size and activity times mediate mammalian responses to climate change
.
Glob. Chang. Biol.
20
,
1760
-
1769
.
McKechnie
,
A. E.
and
Wolf
,
B. O.
(
2010
).
Climate change increases the likelihood of catastrophic avian mortality events during extreme heat waves
.
Biol. Lett.
6
,
253
-
256
.
McKechnie
,
A. E.
and
Wolf
,
B. O.
(
2019
).
The physiology of heat tolerance in small endotherms
.
Physiology
34
,
302
-
313
.
McNab
,
B. K.
and
Morrison
,
P.
(
1963
).
Body temperature and metabolism in subspecies of Peromyscus from arid and mesic environments
.
Ecol. Monogr.
33
,
63
-
82
.
Mitchell
,
D.
,
Snelling
,
E. P.
,
Hetem
,
R. S.
,
Maloney
,
S. K.
,
Strauss
,
W. M.
and
Fuller
,
A.
(
2018
).
Revisiting concepts of thermal physiology: Predicting responses of mammals to climate change
.
J. Anim. Ecol.
87
,
956
-
973
.
Murray
,
I. W.
and
Smith
,
F. A.
(
2012
).
Estimating the influence of the thermal environment on activity patterns of the desert woodrat (Neotoma lepida) using temperature chronologies
.
Can. J. Zool.
90
,
1171
-
1180
.
Myers
,
P.
,
Lundrigan
,
B. L.
,
Kopple
,
R. V.
and
Lacey
,
E. A.
(
2005
).
Climate change and the distribution of Peromyscus in Michigan: is global warming already having an impact
. In
Mammalian Diversification: From Chromosomes to Phylogeography (A Celebration of the career of James L. Patton)
(ed.
E.A.
Lacey
and
P.
Myers
), pp.
101
-
125
.
Berkeley
:
University of California Press
.
Myers
,
P.
,
Lundrigan
,
B. L.
,
Hoffman
,
S. M. G.
,
Haraminac
,
A. P.
and
Seto
,
S. H.
(
2009
).
Climate-induced changes in the small mammal communities of the Northern Great Lakes Region
.
Glob. Chang. Biol.
15
,
1434
-
1454
.
NOAA
(
2022
).
Local Climatological Data. National Oceanic and Atmospheric Administration, National Centers for Environmental Information
.
Parker
,
K. L.
(
1988
).
Effects of heat, cold, and rain on coastal black-tailed deer
.
Can. J. Zool.
66
,
2475
-
2483
.
Porter
,
W. P.
and
Kearney
,
M.
(
2009
).
Size, shape, and the thermal niche of endotherms
.
Proc. Natl. Acad. Sci. USA
106
,
19666
-
19672
.
Pörtner
,
H. O.
and
Knust
,
R.
(
2007
).
Climate change affects marine fishes through the oxygen limitation of thermal tolerance
.
Science
315
,
95
-
97
.
Racinais
,
S.
,
Sawka
,
M.
,
Daanen
,
H.
and
Périard
,
J. D
. (
2019
).
Heat acclimation
. In
Heat stress in sport and exercise
, pp.
159
-
178
.
Springer
.
Ramirez
,
R. W.
,
Riddell
,
E. A.
,
Beissinger
,
S. R.
and
Wolf
,
B. O.
(
2022
).
Keeping your cool: thermoregulatory performance and plasticity in desert cricetid rodents
.
J. Exp. Biol.
225
,
jeb243131
.
Rezende
,
E. L.
,
Chappell
,
M. A.
and
Hammond
,
K. A.
(
2004
).
Cold-acclimation in Peromyscus: temporal effects and individual variation in maximum metabolism and ventilatory traits
.
J. Exp. Biol.
207
,
295
-
305
.
Robertson
,
C. E.
,
Tattersall
,
G. J.
and
McClelland
,
G. B.
(
2019
).
Development of homeothermic endothermy is delayed in high-altitude native deer mice (Peromyscus maniculatus)
.
Proc. R. Soc. B
286
,
20190841
.
Rojas
,
A. D.
,
Körtner
,
G.
and
Geiser
,
F.
(
2012
).
Cool running: locomotor performance at low body temperature in mammals
.
Biol. Lett.
8
,
868
-
870
.
Romanovsky
,
A. A.
,
Ivanov
,
A. I.
and
Shimansky
,
Y. P.
(
2002
).
Selected contribution: ambient temperature for experiments in rats: a new method for determining the zone of thermal neutrality
.
J. Appl. Physiol.
92
,
2667
-
2679
.
Rosenmann
,
M.
and
Morrison
,
P.
(
1974
).
Maximum oxygen consumption and heat loss facilitation in small homeotherms by He-O2
.
Am. J. Physiol. Content
226
,
490
-
495
.
Roy-Dufresne
,
E.
,
Logan
,
T.
,
Simon
,
J. A.
,
Chmura
,
G. L.
and
Millien
,
V.
(
2013
).
Poleward expansion of the white-footed mouse (Peromyscus leucopus) under climate change: implications for the spread of Lyme disease
.
PLoS One
8
,
e80724
.
Saha
,
S. K.
,
Ohno
,
T.
,
Tsuchiya
,
K.
and
Kuroshima
,
A.
(
2000
).
Adaptive modification of membrane phospholipid fatty acid composition and metabolic thermosuppression of brown adipose tissue in heat-acclimated rats
.
Int. J. Biometeorol.
43
,
163
-
168
.
Sareh
,
H.
,
Tulapurkar
,
M. E.
,
Shah
,
N. G.
,
Singh
,
I. S.
and
Hasday
,
J. D.
(
2011
).
Response of mice to continuous 5-day passive hyperthermia resembles human heat acclimation
.
Cell Stress Chaperones
16
,
297
-
307
.
Shido
,
O.
,
Sakurada
,
S.
,
Kohda
,
W.
and
Nagasaka
,
T.
(
1994
).
Day—Night changes of body temperature and feeding activity in heat-acclimated rats
.
Physiol. Behav.
55
,
935
-
939
.
Song
,
C. W.
,
Rhee
,
J. G.
and
Haumschild
,
D. J.
(
1987
).
Continuous and non-invasive quantification of heat-induced changes in blood flow in the skin and RIF-1 tumour of mice by laser Doppler flowmetry
.
Int. J. Hyperth
3
,
71
-
77
.
Speakman
,
J. R.
and
Król
,
E.
(
2010
).
Maximal heat dissipation capacity and hyperthermia risk: neglected key factors in the ecology of endotherms
.
J. Anim. Ecol.
79
,
726
-
746
.
Stager
,
M.
,
Senner
,
N. R.
,
Tobalske
,
B. W.
and
Cheviron
,
Z. A.
(
2020
).
Body temperature maintenance acclimates in a winter-tenacious songbird
.
J. Exp. Biol.
223
,
jeb221853
.
Tapper
,
S.
,
Nocera
,
J. J.
and
Burness
,
G.
(
2020
).
Experimental evidence that hyperthermia limits offspring provisioning in a temperate-breeding bird
.
R. Soc. open Sci
7
,
201589
.
Tate
,
K. B.
,
Ivy
,
C. M.
,
Velotta
,
J. P.
,
Storz
,
J. F.
,
McClelland
,
G. B.
,
Cheviron
,
Z. A.
and
Scott
,
G. R.
(
2017
).
Circulatory mechanisms underlying adaptive increases in thermogenic capacity in high-altitude deer mice
.
J. Exp. Biol.
220
,
3616
-
3620
.
Tewksbury
,
J. J.
,
Huey
,
R. B.
and
Deutsch
,
C. A.
(
2008
).
Putting the heat on tropical animals
.
Science
320
,
1296
.
Tracy
,
R. L.
and
Walsberg
,
G. E.
(
2002
).
Kangaroo rats revisited: re-evaluating a classic case of desert survival
.
Oecologia
133
,
449
-
457
.
Vaanholt
,
L. M.
,
Garland
,
T.
,
Daan
,
S.
and
Visser
,
G. H.
(
2007
).
Wheel-running activity and energy metabolism in relation to ambient temperature in mice selected for high wheel-running activity
.
J. Comp. Physiol. B
177
,
109
-
118
.
Van Sant
,
M. J.
and
Hammond
,
K. A.
(
2008
).
Contribution of shivering and nonshivering thermogenesis to thermogenic capacity for the deer mouse (Peromyscus maniculatus)
.
Physiol. Biochem. Zool.
81
,
605
-
611
.
Walsh
,
S. E.
,
Woods
,
W. E.
and
Hoffman
,
S. M. G.
(
2016
).
Effects of range contraction and habitat fragmentation on genetic variation in the woodland deer mouse (Peromyscus maniculatus gracilis)
.
Am. Midl. Nat.
176
,
272
-
281
.
Weinert
,
D.
and
Waterhouse
,
J.
(
1998
).
Diurnally changing effects of locomotor activity on body temperature in laboratory mice
.
Physiol. Behav.
63
,
837
-
843
.
Wilkinson
,
D. A.
,
Burholt
,
D. R.
and
Shrivastava
,
P. N.
(
1988
).
Hypothermia following whole-body heating of mice: effect of heating time and temperature
.
Int. J. Hyperth
4
,
171
-
182
.
Yamauchi
,
C.
,
Fujita
,
S.
,
Obara
,
T.
and
Ueda
,
T.
(
1983
).
Effects of room temperature on reproduction, body and organ weights, food and water intakes, and hematology in mice
.
Jikken Dobutsu
32
,
1
-
11
.
Yan
,
Y.-E.
,
Zhao
,
Y.-Q.
,
Wang
,
H.
and
Fan
,
M.
(
2006
).
Pathophysiological factors underlying heatstroke
.
Med. Hypotheses
67
,
609
-
617
.
Zhang
,
L.
,
Zhang
,
H.
,
Zhu
,
W.
,
Li
,
X.
and
Wang
,
Z.
(
2012
).
Energy metabolism, thermogenesis and body mass regulation in tree shrew (Tupaia belangeri) during subsequent cold and warm acclimation
.
Comp. Biochem. Physiol. Part A Mol. Integr. Physiol.
162
,
437
-
442
.

Competing interests

The authors declare no competing or financial interests.

Supplementary information