INTRODUCTION: High-altitude [>2400 m (7874 ft)] acclimatization has been well studied with physiological adaptations like reductions in body weight and exercise capacity. However, despite the significance of moderate altitude [MA, 1524–2438 m (5000–8000 ft)], acclimatization at this elevation is not well described. We aimed to investigate differences in mice reared at MA compared to sea level (SL). We hypothesized that MA mice would be smaller and leaner and voluntarily run less than SL mice.
METHODS: C57BL/6 mice reared for at least three generations in Laramie, WY [2194 m (7198 ft), MA], were compared to C57BL/6J mice from Bar Harbor, ME [20 m (66 ft), SL]. We quantified body composition and exercise outputs as well as cardiopulmonary morphometrics. Subsets of MA and SL mice were analyzed to determine differences in neuronal activation after exercise.
RESULTS: When body weight was normalized to tibia length, SL animals weighed 1.30 g ⋅ mm-1 while MA mice weighed 1.13 g · mm-1. Total fat % and trunk fat % were higher in MA mice with values of 41% and 39%, respectively, compared to SL mice with values of 28% and 26%, respectively. However, no differences were noted in leg fat %. MA animals had higher heart mass (119 mg) and lower lung mass (160 mg) compared to SL mice heart mass (100 mg) and lung mass (177 mg). MA mice engaged in about 40% less voluntary wheel-running activity than SL animals.
DISCUSSION: Physiological differences are apparent between MA and SL mice, prompting a need to further understand larger scale implications of residence at moderate altitude.
O’Connor AE, Hatzenbiler DM, Flom LT, Bobadilla A-C, Bruns DR, Schmitt EE. Physiological and morphometric differences in resident moderate-altitude vs. sea-level mice. Aerosp Med Hum Perform. 2023; 94(12):887–893.
High-altitude acclimatization, typically defined as chronic exposure to elevations above 2400 m (7874 ft), results in many different physiological adaptations. Several of these adaptations are also apparent in human populations that have resided for generations at high altitudes, including physiological adaptations like reductions in body weight and changes in body composition, 17 as well as differences in neuronal activity that can affect the cardiovascular system. 3 While these and other physiological adaptations are important contributors to physiology, 23 , 24 much less is known about the physiological outcomes of exposure to chronic hypoxic conditions at moderate altitude (MA). MA is defined as elevations between 1524–2438 m (5000–8000 ft). Over 500 million people live at MA across the world, 28 yet few studies have quantified adaptations at these altitudes and how they differ from low and high altitudes. Moreover, the literature lacks normative values for preclinical models (using animals for research), despite five research universities in the western United States (Colorado State University, University of New Mexico, University of Colorado, U.S. Air Force Academy, and University of Wyoming), all located between 1524–2438 m (5000–8000 ft).
Despite a lack of preclinical data, in the 1980s and 1990s, several studies were conducted at the University of Wyoming on human subjects living in Laramie, WY, compared to sea-level (SL) residents. Both populations were sent to simulated high-altitude conditions by hypobaria. MA natives were at an advantage during early adaptation to hypobaric hypoxia compared to low-altitude natives since they did not experience symptoms of acute mountain sickness. 16 It was also discovered that SL individuals had higher serum cortisol and respiration rates compared to MA persons. 15 Next, it is well documented that exercise capacity decreases with an ascent in altitude, 5 with increases in ventilation, decreases in gas exchange, and increases in cardiac output. 4 Even at MA, it has been reported that decreases in red blood cell oxygen transport can lead to lower heart rates and decreased lactate accumulation during exercise. 13 Despite the evidence of physiological differences at MA in humans, animal studies are lacking to identify if these differences are also evident preclinically.
The following experiments took place in Laramie, WY, at the MA of 2194 m (7200 ft) where the air exerts a pressure of 580 mmHg (22.8 inHG, 0.763 atm). We aimed to accomplish the following goals: 1) to quantify differences in body composition and cardiopulmonary morphometrics between MA and SL mice; 2) to quantify differences in voluntary and forced exercise metrics between MA mice and SL mice; and 3) to determine if neuronal activation during exercise differed between MA and SL mice within three regions of the brain reward circuitry (core subregion of the nucleus accumbens, prefrontal cortex, and hypothalamus) to understand the motivation to engage in running at MA. We hypothesized that MA mice would be smaller and leaner and voluntarily run less compared to SL mice. We report that at MA, some physiological differences are apparent in the mouse model, suggesting that a better characterization of preclinical models in MA conditions are needed.
METHODS
Animals
Adult C57BL/6 mice 3–4 mo of age were used for the following experiments. Mice were purchased from Jackson Laboratories (Bar Harbor, ME), then bred and housed in Laramie for at least three generations. These mice were designated as MA. SL animals were purchased from Jackson Laboratories, then quarantined for 48 h, then immediately underwent experimental protocols, and were sacrificed within 1 wk of arrival to MA, except for one cohort of animals that underwent 14 d of wheel-running. Mice were housed at the University of Wyoming [Laramie, WY; 2194 m (7200 ft)] on a normal 12-h light–dark cycle and supplied with ad libitum access to standard rodent chow and water. All protocols followed the standards of humane animal care and were approved by the University of Wyoming Institutional Animal Care and Use Committee. Male and female sexes have been combined to account for underpowered study if separated by sex.
Materials and Procedures
Body weight (BW), body composition, and morphometric measurements were obtained from sedentary animals at MA (N = 15) and SL (N = 16). To quantify body composition, dual energy X-ray absorptiometry was used following a protocol previously established by our group. 18 Following dual energy X-ray absorptiometry, mice were humanely euthanized (FatalPlus; pentobarbital). Tibia length (TL) was collected by caliper. Heart weights were collected before separating the right ventricle (RV) from the left ventricle (LV) and septum. Wet lung weights were collected along with hind limb complex (gastrocnemius, solus, and plantaris) weight. All procedures were performed by the same lab technician to minimize variance in our data.
SL mice (N = 10) arrived from Jackson Laboratory and were given 48 h to acclimate. Age-matched MA mice (N = 11) and SL mice underwent a time-to-exhaustion treadmill protocol as previously described. 1 Briefly, mice underwent a ramped protocol where the speed increased every 3 min by 3 m ⋅ min-1 until animals were no longer able to keep up with the speed of the treadmill. After the treadmill exhaustion test, animals were individually housed with a voluntary running wheel (Columbus Instruments, Columbus, OH) to be used for 14 d as previously described. 2 Wheel-running data were collected daily in revolutions and converted to km ⋅ d-1 and mice were checked to ensure the wheel was functioning properly.
MA mice (N = 4) and SL mice (N = 4) were singly housed with a running wheel (Columbus Instruments) to be used voluntarily as described above. SL mice and MA mice had access to voluntary wheels for 3 d prior to euthanasia and collection of brains for analysis. On Day 4, all mice voluntarily ran for approximately 2 h during the nocturnal hours between 19:00 h and 21:00 h, then were immediately deeply anesthetized with a 2/1 combination of ketamine (BW, Vedco, St. Joseph, MO) and xylaxine (Akorn, Gurnee, IL), perfused with phosphate-buffered saline followed by 3.7% formaldehyde (Fisher Scientific, Hampton, NH), and brains were postfixed for 24 h. Brains were sliced in 1/4 series of 50 μm sections using a cryostat and stored in phosphate buffered saline (PBS)-sucrose-azide solution before performing immunohistochemistry. The three regions of interest were the prefrontal cortex (Bregma 3.08-1.34 mm), nucleus accumbens core (Bregma 1.94-0.62 mm), and hypothalamus (Bregma −0.34 to −2.06 mm). Free-floating sections were rinsed in PBS-Triton (0.25%) and incubated in normal goat serum and primary antibodies (Antiphospho-c-Fos, marker for neuronal activation, Cell Signaling Technology, #5348, 1:1000, Danvers, MA; Anti-NeuN, neuronal marker, #MAB377, 1:1000, Millipore Sigma, Burlington, MA) overnight at 4°C. After multiple PBS-Triton rinses, sections were incubated in Alexa-Fluor conjugated secondary antibodies (1:1000, Life Technologies, Carlsbad, CA). We used super-resolution confocal imaging (confocal 980, Zeiss United States, Dublin, CA) paired with the LAS X Huygens HyVolution deconvolution software (Scientific Volume Imaging, Netherlands) to visualize the neurons within each region of interest and their colocalization with c-Fos; the marker for neuronal activation. To minimize observer bias while quantifying the number c-Fos+ and NeuN+ neurons, we automatized the acquisition tiling of the 3 × 3 (512 × 512 pixels) array of images at 63× magnification in the green channel for c-Fos (488 nm) and far-red channel for NeuN (638 nm). The stitching was also automatized, creating a 0.5 mm2 nine-tile mosaic of a portion of the NAcore, with a composite 1536 × 1536-pixel field. Once 3 × 3 tiled arrays were collected and stitched, they were manually checked for accuracy and fidelity. Next, semiautomated cell counts were performed for each channel (c-Fos and NeuN) using IMARIS software (Oxford Instruments, Concord, MA). During this process, c-Fos+ and NeuN+ cells were initially marked in an automated fashion based on cell body or nucleus size using the spots function (7 µm for c-Fos and NeuN) and the intensity of the signal in that channel (voxel intensity). All automated counts for each channel were then checked manually by an investigator who was blind to treatment groups. We acquired up to four mosaics per animal. In all three regions, % c-Fos activated was calculated by taking the number of c-Fos tagged activated neurons, divided by the number of NeuN tagged, multiplied by 100%.
Statistical Analyses
Voluntary running wheel data were analyzed by a repeated measures analysis of variance with pairwise t-test comparisons. All other outcome variables were analyzed by Student’s t-test. P-value was set at 0.05 α-priori. Power analysis for studies involving inbred strains of mice require a minimum of 6 animals to demonstrate significant differences between groups, yielding a power of at least 0.80.
RESULTS
BW did not differ by elevation (P = 0.85). However, when normalized to TL (mm) to account for body size, SL animals weighed more than MA (P = 0.01). Total fat % and trunk fat % were higher in MA mice compared to SL (P < 0.0 and P < 0.0, respectively) but no differences were noted in leg fat % (P = 0.89). Hind leg complex (HLC; mg) weight was not different in SL vs. MA animals (P = 0.09). When normalized to BW, SL mice had smaller HLC compared to MA (P = 0.03), an effect likely driven by the smaller body mass at MA. However, when normalized to TL, HLC mass showed no difference in MA and SL animals (P = 0.45). In addition, MA mice had higher cardiac mass compared to SL (P = 0.01), which persisted when normalized to BW (P < 0.00). However, when whole heart was normalized to TL, no differences were observed in cardiac mass (P = 0.48). The RV was dissected from the LV and septum to quantify ventricle-specific weights by altitude. LV and RV mass did not differ by altitude (P = 0.17 and P = 0.96, respectively). No differences were observed between MA and SL animals when the LV or RV was normalized to BW (P = 0.97 and P = 0.10, respectively), nor was significance observed between MA and SL mice when LV or RV was normalized to TL (P = 0.06 and P = 0.93, respectively). SL and MA did not differ in lung weights (P = 0.07); however, when normalized to BW (P = 0.04) and TL (P < 0.0), SL animals had bigger lungs compared to MA. All body composition and cardiopulmonary morphometric data can be found in
Table I
.
Next, we quantified voluntary wheel running (km/d) over 14 d. Mice born and bred at MA ran significantly less over the 2-wk period compared to SL mice (P < 0.0) (
Fig. 1A
). These differences in voluntary wheel running distances began to emerge by Day 8 and continued until Day 14. In total, MA mice ran 40% less than SL counterparts. Despite engaging in significantly different volumes of exercise over a 14-d period, time to exhaustion was similar between SL and MA animals (P = 0.82) (
Fig. 1B
).
Citation: Aerospace Medicine and Human Performance 94, 12; 10.3357/AMHP.6234.2023
Given the robust differences in voluntary wheel running engagement, we wanted to test whether MA mice experience less neuronal activity and reward integration compared to SL mice. A separate cohort of animals that engaged in voluntary wheel running was used for neuronal analysis (N = 4 per group). These mice were active and running immediately prior to euthanasia, permitting activation of reward centers in the brain. We quantified neuronal activation in the nucleus accumbens core (
Fig. 2A
), the prefrontal cortex (PFC) (
Fig. 2B
), and the hypothalamus (
Fig. 2C
), three important regions within the brain’s reward circuitry. There were no differences detected in % c-Fos activation between SL and MA mice in the nucleus accumbens core (P = 0.50), in the PFC (P = 0.50), or in the hypothalamus (P = 0.43).
Citation: Aerospace Medicine and Human Performance 94, 12; 10.3357/AMHP.6234.2023
DISCUSSION
Residence at high altitude results in a multitude of morphometric and physiological differences compared to SL. However, despite a significant global population and research institutions that reside at MA, little is known about acclimation at MA, and no normative preclinical data exist. Here, we aimed to quantify normative values for several physiological parameters ranging from body size and composition, cardiopulmonary morphometrics, and voluntary running-wheel activity to shed light on the differences in physiology at MA of 2194 m (7198 ft). We report that when BW was normalized to body size, SL animals weighed more than MA animals. Total fat % and trunk fat % were higher in MA mice compared to SL, but no differences were noted in leg fat % and muscle mass. Differences in cardiopulmonary morphometrics were noted in MA animals having higher heart mass and lower lung mass compared to SL mice. MA mice engaged in less voluntary wheel-running activity than SL animals via unclear mechanisms that are likely not due to differences in reward-seeking behavior since differences in activation of the brain reward circuitry were not detected.
Cardiopulmonary acclimatization to chronic hypoxia is characterized by a variety of functional differences that render the heart and lungs better able to maintain homeostasis under the low-oxygen environment. One of the well-studied adaptations to chronic HA residence are changes to cardiac structure and function. In 1956, Canepa et al. first reported that healthy men and women living at high altitude have some degree of pulmonary hypertension and right ventricular hypertrophy (RVH). 22 This report was later confirmed in several geographical populations. 29 Mean pulmonary arterial pressure (PAP) is elevated even at MA, 21 thus we expected that MA mice would have mildly elevated RV mass. However, we did not find that MA mice had higher RV or RV/BW compared to SL controls. While the degree of hypoxic stress may be responsible for the lack of RVH in our mice, it may also be that species differences are responsible for a lack of RVH. Indeed, high-altitude mammals, including rabbits (mountain viscacha; Lagidium peruanum), do not have an elevated pulmonary vasoconstriction response and as such have low PAP and no RVH. 7 Further, we note that in some cases, such as in a case report from a high-altitude dweller at 3000 m (9843 ft), that RV hypertrophy does not always accompany elevated PAP. 27 The differences in systemic blood pressure between high- and lowlanders are more complex, with differences between populations. While Tibetan highlanders are known to have elevated blood pressure compared to lowland counterparts, Andean highlanders have lower blood pressure. 19 Consistent with lower systemic blood pressure, Sherpas have smaller LV mass, volumes, and wall thickness. 26 While heart weight and heart/BW were higher in MA mice compared to SL, differences in LV mass were not apparent. We suggest that future studies comprehensively quantify cardiac and pulmonary function at MA to understand the mechanisms of cardiac adaptation in acclimatization to this elevation.
Regular aerobic exercise has numerous health benefits. Studies assessing the impact of exercise on health have been used preclinically to uncover novel mechanisms about acute and chronic effects of exercise, leading to notable discoveries in physiology and pharmacology. 6 However, to capitalize on this model at MA, the scientific community requires normative values for how much MA animals engage in voluntary exercise. We tested whether mice born and bred at MA had different voluntary physical activity levels compared to SL mice. Our results indicate that MA mice have lower voluntary running-wheel activity compared to age-matched SL counterparts. When these running values are compared to a hallmark study published in 2010 by Lightfoot and colleagues that quantified activity of inbred strains of mice on voluntary wheels, 12 we find that the SL mice in our study run a comparable daily amount to those SL mice purchased from the same vendor. However, in contrast, the MA mice in our study ran less than SL counterparts. SL mice in our study ran 6.5 ± 2.7 km ⋅ d-1 with MA mice running even less at 4 ± 2.9 km ⋅ d-1.
Another explanation for the differences in voluntary activity at MA compared to SL could be attributed to a reduced drive to exercise. The dopaminergic system controls motivation and neural rewards, as well as motor movement, such as physical activity. 30 It has been postulated that the dopaminergic network contributes to the control and regulation of physical activity. 11 Mice born and bred at MA could have altered dopaminergic profiles. To test if the lowered activity levels in our MA mice are due to hypoxic stress resulting from an increase in altitude, we examined neuronal activation within the brain’s reward circuitry, modulated in part by dopaminergic transmission. We did not find any differences in % c-Fos activation between SL and MA mice in the nucleus accumbens core, in the PFC, or the hypothalamus, perhaps because our study was underpowered and we combined sexes. The literature has suggested sex differences in the physiological responses and neuro-cardiovascular control of mice when placed under intermittent and chronic hypoxic conditions. 10 This includes decreased vasoconstriction in female mice, as well as sex differences in the respiratory modulation of sympathetic activity. 25 Additionally, we caution that we did not control for estrous cycling in the female mice, and it is likely that estrogen could regulate hypoxia sensitivity, at least acutely. 8 Moreover, the number of animals included in the neuronal activation study is low and requires an increase to correlate neuronal activation with running behavior. We suggest that further experimentation is needed to determine the role that estrogen plays in hypoxia.
Several limitations warrant discussion. We only analyzed one inbred strain of mouse and we combined sexes, which should be considered when translating and predicting some of these findings either to other preclinical models or to human populations. Any conclusions drawn from the results of 8 total animals (4 per group) should be taken with an abundance of caution. We also did not measure cardiac function in either group of mice, which would have perhaps given a more in-depth analysis of how MA impacts physiology. Finally, it should be mentioned that there are species differences in response to hypoxia. For example, cattle and pigs are susceptible to hypoxia, while sheep and dogs are less sensitive. 20 By studying various species and their genetic profiles, “omic” data, and our knowledge of physiological principles, we can start to put the puzzle pieces together to better understand and elucidate physiological mechanisms that contribute to adaptations to hypoxia.
A multitude of environmental, social, and lifestyle characteristics differ between MA and SL. Unlike previous reports in human populations where some or even many of these variables differ, our work does not have influence from other confounding variables seen in real-world circumstances. For example, MA is a hypoxic environment, yet other variables differ at MA and SL in addition to hypoxia. Environmental variables like food consumption, rural vs. urban living, ultraviolet light, and ambient temperature all affect physiology. The mice used in this study are considered genotypically identical, ate the same food, were on the same light cycle, and were offered the same environment to voluntarily exercise; therefore, they only differed with respect to the altitude at which they were raised. It would be worth investigating if the differences in our animals persisted if the SL mice had been given a longer opportunity to acclimate to MA or send our MA mice to SL to determine what physiological changes occurred. In addition, continued research needs to further understand the health consequences of MA compared to SL as this could impact the risk for disease progression 14 and implicate drug delivery. 9 Continued work should examine why such differences occur and the extent of these differences between males and females. Eventually, we hope such findings can begin to be applied to human populations and could one day predict health outcomes at different altitudes to improve the well-being and health of people who live at and travel to moderate-to-high elevations.
Voluntary running wheel distances and time to exhaustion in MA and SL mice. A) An overall difference was observed when comparing MA and SL voluntary running wheel data over a 14-d period. B) No difference was observed in time to exhaustion between MA and SL animals. Data were analyzed by t-test. N = 10–11 per group.
Immunohistochemistry staining displaying c-Fos activation in brain regions between SL and MA mice. Neuronal activation was quantified using neuronal marker NeuN (staining of active and inactive neurons) and c-Fos (active neurons) antibodies, followed by imaging using confocal microscopy. A) Representative images for NeuN and c-Fos and merged images in the nucleus accumbens core with no difference in % c-Fos activation between MA and SL mice. B) Representative images for NeuN and c-Fos and merged images in the prefrontal cortex with no difference in % c-Fos activation between MA and SL mice. C) Representative images for NeuN and c-Fos and merged images in the hypothalamus with no difference in % c-Fos activation between MA and SL mice. Data were analyzed by t-test. N = 4 per group.
Contributor Notes
