Acclimatization of low altitude-bred deer mice (Peromyscus maniculatus) to high altitude

Abstract

A colony of deer mice subspecies (Peromyscus maniculatus sonoriensis) native to high altitude (HA) has been maintained at sea level for 18–20 generations and remains genetically unchanged. To determine if these animals retain responsiveness to hypoxia, one group (9–11 wk old) was acclimated to HA (3,800 m) for 8 wk. Age-matched control animals were acclimated to a lower altitude (LA; 252 m). Maximal O₂ uptake (V̇o_2max) was measured at the respective altitudes. On a separate day, lung volume, diffusing capacity for carbon monoxide (DL_CO), and pulmonary blood flow were measured under anesthesia using a rebreathing technique at two inspired O₂ tensions. The HA-acclimated deer mice maintained a normal V̇o_2max relative to LA baseline. Compared with LA control mice, antemortem lung volume was larger in HA mice in a manner dependent on alveolar O₂ tension. Systemic hematocrit, pulmonary blood flow, and standardized DL_CO did not differ significantly between groups. HA mice showed a higher postmortem alveolar-capillary hematocrit, larger alveolar ducts, and smaller distal conducting structures. In HA mice, absolute volumes of alveolar type I epithelia and endothelia were higher whereas that of interstitia was lower than in LA mice. These structural changes occurred without a net increase in whole-lung septal tissue-capillary volumes or surface areas. Thus, deer mice bred and raised to adulthood at LA retain phenotypic plasticity and adapt to HA without a decrement in V̇o_2max via structural (enlarged airspaces, alveolar septal remodeling) and nonstructural (lung expansion under hypoxia) mechanisms and without an increase in systemic hematocrit or compensatory lung growth.

NEW & NOTEWORTHY Deer mice (Peromyscus maniculatus) are robust and very active mammals that are found across the North American continent. They are also highly adaptable to extreme environments. When introduced to high altitude they retain remarkable adaptive ability to the low-oxygen environment via lung expansion and remodeling of existing lung structure, thereby maintaining normal aerobic capacity without generating more red blood cells or additional lung tissue.

INTRODUCTION

Deer mice (Peromyscus maniculatus) are native to the North American continent at both low and high altitudes (LA and HA, respectively) from below sea level in Death Valley to over 4,300 m above sea level in the high mountains of California and Colorado (20, 63). In previous research, we have shown that when laboratory-bred adult deer mice (descended from wild HA populations) that are similar in age, health status, and developmental history are moved from LA to HA (390 m to >3,000 m) their aerobic performance at HA is initially significantly diminished (18, 55), and then it gradually recovers during an acclimation period. After 6–8 wk of acclimation, their aerobic performance recovers to levels not different from that before transport to HA and is accompanied by an increase in lung mass and volume (18, 62). When tested in normoxia, the HA-acclimated mice had a higher aerobic performance than LA mice without acclimation to HA (55). Furthermore, our previous studies show that the lung mass (18) and volume (55) of deer mice increases with acclimation to HA. Although we find the above results consistently, we have not been able to determine if the larger lung volumes after acclimation are responsible for a greater diffusion of oxygen. We speculated that the recovery of aerobic scope is due to the increase in lung mass and volume, but pulmonary oxygen uptake has not been measured in these mice before and after acclimation to HA. Here, we hypothesize that deer mice, as adults, possess significant capacity for phenotypic plasticity in pulmonary oxygen uptake during adaptation to HA. Our goal was to characterize the adaptive changes in lung structure and cardiopulmonary function in deer mice bred at LA and recently acclimated to HA and to relate these changes to aerobic performance.

A colony of a subspecies (P. maniculatus sonoriensis) native to the mountains of California has been maintained at the University of California Riverside (UCR; 252 m altitude) and the Peromyscus Genetic Stock Center (Columbia, SC; 98 m altitude) for 18–20 generations and has been shown to remain genetically similar (no loss of genetic variation) to the wild strain (Peromyscus Genetic Stock Center, Janet Crossland, unpublished observations). To examine how these mice, maintained at LA, adapt to HA, one group (9–11 wk old) was brought to HA (3,800 m) where they resided for 8 wk before being studied at HA (HA group). Other age-matched deer mice were maintained and studied at UCR at an LA (LA group). Maximal O₂ uptake (V̇o_2max) was measured pre- and postacclimation. Lung function was measured in both groups postacclimation and allometrically compared with that in other adult outbred and inbred laboratory murine strains studied by the same method at the University of Texas Southwestern Medical Center (130 m altitude). Finally, alveolar septal ultrastructure was measured postmortem.

MATERIALS AND METHODS

Animals.

Animal studies were carried out under protocols approved by the Institutional Animal Care and Use Committees of the University of Texas Southwestern Medical Center, the White Mountain Research Station, and UCR. A total of 26 male deer mice (P. maniculatus sonoriensis) were used. Animals were captive-bred at LA (252 m, UCR, CA and 88 m, Columbia, SC) from a colony established from animals caught in the White Mountains of eastern California in 1997 and 18–20 generations removed from wild deer mice. All deer mice underwent measurement of V̇o_2max at LA. Then, one group of deer mice (n = 13, 9–11 wk of age) was brought to HA (3,800 m) at the Barcroft Laboratory of the University of California White Mountain Research Center where they resided for 8 wk before V̇o_2max was measured at HA. Another group of age-matched male deer mice (n = 13) remained simultaneously at LA at UCR for 8 wk before V̇o_2max was measured again at LA. Thereafter, a subset of deer mice (n = 10 per group) underwent postacclimation lung function testing at their respective altitude of residence (HA or LA) 5 to 8 days after the V̇o_2max measurement.

Lung function testing used a rebreathing technique we first established in laboratory murine strains, including outbred male Swiss Webster mice (12 wk old, n = 8, Charles River, Wilmington, MA) and inbred C57BL/6 mice (Charles River) that were bred at sea level and studied at the University of Texas Southwestern in Dallas, TX (130 m altitude). A cohort of female 57BL/6 mice (14 to 54 wk old, n = 9) was used to compare measurements directly by rebreathing and single-breath maneuvers. Another cohort of male 57BL/6 mice (~18 wk old, n = 14) underwent rebreathing lung function testing for allometric comparisons with that in other murine strains and various other species reported in the literature, which used mostly males.

V̇o_2max.

V̇o_2max in deer mice was measured by open-flow respirometry during forced treadmill exercise. Air was supplied either by outlet (UCR) or using a positive pressure pump (Barcroft). Incurrent air was dried by Drierite (Xenia, OH) and scrubbed of carbon dioxide by soda lime. Flow rate was regulated by Porter mass flow controllers (Hatfield, PA) upstream of the treadmill. The working section of the treadmill was enclosed by Plexiglas (dimensions 6 × 7 × 13 cm). Flow rates of 2,300 and 1,550 ml/min [corrected to standard temperature and pressure (760 mmHg)] conditions were used at UCR and Barcroft, respectively. Approximately 150 ml/min of excurrent air was subsampled, dried, and scrubbed of CO₂ before being routed through the oxygen sensor. Oxygen concentration was analyzed with Ametek/Applied Electrochemistry S-3A analyzers (Pittsburgh, PA) and then digitized by Sable Systems UI-2 (Las Vegas, NV) A–D converters and recorded on a Macintosh computer running Warthog Laboratory Helper software (http://www.warthog.ucr.edu).

Body mass was measured before all runs. Mice were then placed on the treadmill and allowed to adjust for 2–4 min. During this time, a reference reading of unbreathed air was obtained. The treadmill was then started at a low speed (~0.1 m/s), and the speed subsequently increased by increments of 0.1 m/s every 30–45 s until the mouse could either no longer maintain position on the tread or V̇o₂ did not increase with increasing speed. At this time, the treadmill was stopped, but V̇o₂ measurements continued for several minutes during the animal’s recovery period before a second reference reading was recorded.

V̇o₂ was calculated from O₂ concentrations using the mode 1 equation in Warthog Laboratory Analyst software (http://www.warthog.ucr.edu)

$${\dot{\text{V}}}_{{\text{O}}_{\text{2}}}=\dot{\text{V}}\frac{\left({\text{F}I}_{{\text{O}}_{\text{2}}}-{\text{F}E}_{{\text{O}}_{\text{2}}}\right)}{1-{\text{F}E}_{{\text{O}}_{\text{2}}}}$$

where V̇ is flow rate (ml/min correcting to standard temperature and pressure), and $F{I}_{O_2}$ and $F{E}_{O_2}$ are incurrent (reference) and excurrent fractional O₂ concentrations, respectively ($F{I}_{O_2}$ was assumed to be 0.2095, and $F{E}_{O_2}$ never fell below 0.2080). An “instantaneous” correction was applied to account for mixing (5) and to resolve short-term metabolic changes better. This is a common adjustment when measuring O₂ consumption in small vertebrates and insects. Correction is necessary because in the small volume of the test chamber on the treadmill, build-up of O₂ and CO₂ can cause measurement errors. V̇o_2max was calculated as the highest 1-min average during the running bout or postexercise recovery period.

Lung function measurements.

On a separate day after measurement of V̇o_2max, lung function was assessed at the animal’s altitude of residence (n = 10 per group). The animal was anesthetized by intraperitoneal injection of ketamine (50 mg/kg) and xylazine (5 mg/kg). The neck, chest, and abdomen were shaved. A longitudinal neck incision exposed the trachea, which was cannulated (20-gauge adaptor) and tied securely (3-0 silk) for mechanical ventilation (Columbus Instruments, Columbus, OH) at a respiratory rate of 150 breaths/min and tidal volume of 8 ml/kg increasing to 10–12 ml/kg as needed to maintain O₂ saturation above 90%. Heart rate and transcutaneous O₂ saturation were monitored using a tail probe (Vet/Ox G2, West Yorkshire, UK). The tracheal cannula was attached to a manifold that switched between the ventilator circuit and a small calibrated glass syringe (1 ml) for delivering precise gas volumes. Total apparatus dead space was 0.065 ml. Airway pressure was recorded using a pressure transducer (model no. 1100, Hans Rudolph, Shawnee, KS) calibrated before each experiment against a manometer. Data were acquired (100 Hz) using a laptop computer, a data acquisition card (PCM-DAS08, Computer Boards, Middleboro, MA), and LabView software (National Instruments, Austin, TX).

Airway pressure-lung volume relationship.

The ventilator was paused and the lungs inflated with a known volume of air using the glass syringe. Each delivered volume was held for 6 s. Airway pressure at 3 s after the initial peak pressure was recorded, followed by expiration to air and switching back to the ventilator circuit. Volumes were delivered in 10-ml/kg increments until airway pressure reached 30 cm H₂O, then repeated in reverse order. Duplicate measurements were averaged.

Rebreathing measurements.

The technique is similar as that described for rats (78). The rebreathing gas mixture consisted of 0.3% CO, 0.5% Ne, 0.8% C₂H₂, and either 40% O₂ or 90% O₂ in balance N₂. The animal breathed 25% O₂ for 2 min before each measurement using 40% O₂ and 100% O₂ for 2 min before each measurement using 90% O₂. Initial gas concentrations were measured by a gas chromatograph (Varian, Micro GC Model 4900, Santa Clara, CA) with two columns (M5AHIBF and PPUHI) and a thermal conductivity detector using 100% high-purity helium as a carrier gas. The ventilator was paused; a test gas volume equivalent to the inflation volume needed to reach 20 cmH₂O airway pressure was delivered to the lungs. Rebreathing was conducted by gently pumping the syringe at 60 strokes/min to mix the test gas with resident lung gas for 5, 7, or 9 s in separate maneuvers. The order of measurement using gas mixture containing either 40% or 90% O₂ was randomized. Precise rebreathing duration was determined from simultaneous airway pressure tracings. Mechanical ventilation resumed at the end of each rebreathing maneuver. The final syringe volume was recorded, then diluted with a known volume (0.7 to 1.0 ml) of 100% N₂ for the measurement of final gas concentrations. Dilution was necessary to ensure a sufficient sample volume for the gas chromatograph. Expired gas concentrations were calculated by considering the dilution ratios. From these measurements, disappearance curves were constructed for Ne, CO, and C₂H₂. Each maneuver was performed in duplicate, and the results were averaged. The interval between successive rebreathing maneuvers was at least 2 min. At the end of rebreathing maneuvers, CO back pressure was determined by delivering 3 ml of 100% O₂ to the lungs, performing the rebreathing maneuver for 60 s, and measuring the final expired CO concentration.

Analysis of physiological data.

The airway pressure-lung volume (PV) curves were analyzed using a standard approach (57). Lung volumes were calculated from Ne dilution and expressed at body temperature and pressure saturated conditions. Compliance was calculated for airway pressures between 10 and 30 cmH₂O. Pulmonary blood flow and lung diffusing capacity for carbon monoxide (DL_CO) were calculated from the slopes of the logarithmic disappearance curves of C₂H₂ and CO, respectively, relative to the corresponding Ne concentration during rebreathing (77). The serial components of DL_CO, alveolar membrane diffusing capacity (DM_CO), and pulmonary capillary blood volume (Vc), were calculated from measurements made at two alveolar O₂ tensions following the Roughton-Forster (RF) technique (54); these values were used to express DL_CO under standardized conditions (DL_CO-std) at a constant alveolar O₂ tension ($P{A}_{O_2}$, 120 mmHg) and hematocrit (0.45) and were plotted with respect to pulmonary blood flow. The gas exchanging alveolar septal tissue volume was calculated by extrapolating the C₂H₂ disappearance curve to time zero (56).

Comparing rebreathing and single-breath measurements.

To validate our rebreathing method in mice, female 57BL/6J mice were studied using both rebreathing and single-breath techniques. Following completion of rebreathing maneuvers as described above, the same volume of test gas mixture as in the rebreathing maneuvers was delivered to the lung; the breath was held for 10 s followed by expiration. The expired gas volume recovered in the glass syringe was measured, then diluted with 1.0 ml of 100% nitrogen, and the gas concentrations measured by gas chromatography. Lung volume was calculated from Ne dilution. DL_CO was calculated from the rate of CO disappearance during breath hold. Breath-hold maneuvers were performed in duplicate in each animal, and the results were averaged.

Terminal procedures.

For tissue harvest, 7 LA and 6 HA deer mice were deeply anesthetized and mechanically ventilated as above. The chest cavity was opened via a midline abdominal incision. Blood was drawn from the left ventricle with a heparinized syringe for measurement of hematocrit. Pentobarbital overdose (150 mg/kg) was administered by intraperitoneal injection. The lungs were removed, fixed by tracheal instillation of 2.5% buffered glutaraldehyde at a constant airway pressure (25 cm H₂O), and immersed in 2.5% glutaraldehyde while maintaining airway pressure.

Morphometric measurements.

Volume of the intact lung was measured by saline immersion (73). The lungs were serially sectioned (3-mm intervals) starting with a random orientation. Four tissue blocks were systematically sampled per animal (1 each from the upper and the lower half of each lung). Tissue blocks were postfixed with 1% osmium tetroxide in 0.1 M cacodylate buffer, treated with 2% uranyl acetate, dehydrated through graded alcohol, and embedded in Spurr (Electron Microscopy Sciences, Hartfield, PA). A 3-level stratified analytical scheme [low-power light microscopy (LM; ×275), high-power LM (×550), and transmission electron microscopy (TEM; ~×21,000)] was employed (14, 24). For low- and high-power LM, each block was sectioned (1 µm) and stained with toluidine blue. One section per block was overlaid with a test grid at the appropriate magnification. From a random start, at least 20 nonoverlapping microscopic fields per block (80 per lung) were systematically examined at ×275 magnification. Using point counting, airways or blood structures >20 µm in diameter (nonfine parenchyma) were excluded from the reference space to estimate the volume density of fine parenchyma, alveolar ducts, and sacs per unit lung volume. At ×550 magnification, 15 nonoverlapping microscopic fields per block (60 per lung) were systematically imaged from a random start. The volume density of alveolar septa per unit lung volume was estimated from the number of points falling on alveolar septa relative to the total number of points falling on fine parenchyma.

For TEM, two blocks (one from each lung) per animal were sectioned (70-nm thickness), mounted on copper grids, and examined at ~×21,000 (JEOL EXII). Thirty nonoverlapping TEM fields per grid were systematically examined (total 60 per animal). Volume densities of epithelia, interstitia, and endothelia were estimated by point counting with the alveolar septum as the reference space. Alveolar epithelial and capillary surface densities per unit septum volume were estimated by intersection counting. At least 300 points or intersections were counted per grid. The volume and surface densities at each level were related through the cascade of levels to the total lung volume to obtain absolute volumes (ml) and surface areas (cm²). The lengths of test lines that transect the tissue-plasma barrier (from air-epithelial interface to the nearest erythrocyte membrane) were measured to estimate the harmonic mean thickness of the tissue-plasma barrier, an anatomical index of resistance of the blood-gas barrier to diffusion. Enough samples, blocks, and fields were examined to reduce the coefficients of variation in the major parameters of volume and surface densities and harmonic mean thickness of the tissue-plasma barrier to ≤10%.

Statistical analysis.

Because of significant differential changes in body mass between groups during acclimation, we show the results (mean ± SD) in both absolute and body mass-specific values. Changes in body mass and V̇o_2max were compared by paired t-test and analysis of covariance. Lung function data were compared between groups using factorial ANOVA or unpaired t-test. Correlation between DL_CO and pulmonary blood flow was evaluated by linear regression analysis; slopes and intercepts were compared between groups by the method of Zar (80). Differences at P ≤ 0.05 were considered significant.

RESULTS

Physiological measurements in deer mice.

Baseline body mass at LA was not different between groups (Table 1). After acclimation, body mass increased in LA deer mice (+12.5% from baseline, P = 0.03) but changed minimally in HA deer mice (+3% from baseline) and became significantly lower than that in LA mice (P = 0.02). As the animals were fed the same diet, the findings likely reflect a lower activity level and/or higher food intake in the LA group. Baseline absolute V̇o_2max at our LA site was borderline (8%) lower in deer mice that subsequently acclimated to HA (P = 0.06). Following acclimation, mean absolute V̇o_2max increased 7% in HA mice and declined 2% in LA mice. Postacclimation absolute V̇o_2max was not significantly different between HA and LA deer mice (Table 1; Fig. 1). The change in V̇o_2max (post/pre ratio) (HA 1.08 ± 0.16, LA 0.99 ± 0.08, mean ± SD) was not significantly different (P = 0.08) but may be biologically meaningful.

Table 1.

V̇o_2max

Exposure	LA	HA	P Value
Number of animals	13	13
Preacclimation
Body mass, g	20.8 ± 2.8	19.0 ± 3.4	0.18
V̇o2max, ml/min	4.52 ± 0.47	4.15 ± 0.48	0.06#
V̇o2max, ml⋅(min·kg)−1	220 ± 24	222 ± 31	0.84
Postacclimation
Body mass, g	23.3 ± 4.5a	19.6 ± 2.75	0.02*
V̇o2max, ml/min	4.43 ± 0.40	4.46 ± 0.73	0.94*
V̇o2max, ml⋅(min·kg)−1	195 ± 31a	228 ± 28	0.008*
Post-/preacclimation V̇o2max ratio	0.99 ± 0.08	1.08 ± 0.16	0.08

Values are mean ± SD. HA, high altitude; LA, low altitude; V̇o_2max, maximal O₂ uptake. Body mass was measured on each day of study.

^*P ≤ 0.05;

^#P = 0.06: HA vs. LA, unpaired t-test;

^aP < 0.002: post- vs. preacclimation, paired t-test.

Fig. 1.

V̇o_2max in deer mice was measured at low altitude (LA) preacclimation and at either low (LA) or high altitude (HA) postacclimation (n = 13 animals each). Preacclimation absolute V̇o_2max was lower in the HA than the LA group at a borderline significance (P = 0.06). Postacclimation V̇o_2max was not significantly different between groups. Mean ± SD. V̇o_2max,maximal O₂ uptake.

Systemic hematocrit and lung function were measured 5–8 days following the postacclimation V̇o_2max measurement (n = 10 deer mice per group, Table 2). Systemic hematocrit was not different between HA and LA deer mice. Lung volume at a given airway pressure was significantly higher in HA deer mice than LA deer mice (Fig. 2). Specific compliance of the respiratory system was not significantly different between HA and LA groups (Table 2).

Table 2.

Lung function

Exposure	LA	HA	P Value
Number of animals	10	10
Terminal body mass, g	22.5 ± 4.1	18.8 ± 3.3	0.02*
Systemic hematocrit, %	42.3 ± 2.4	44.2 ± 3.6	0.27
Respiratory system compliance§, ml⋅(cmH2O)−1	0.021 ± 0.005	0.020 ± 0.006	0.69
Specific respiratory system compliance§, ml⋅(cmH2O⋅ml)−1	0.067 ± 0.027	0.051 ± 0.016	0.08
Rebreathing Mixture Containing 40% O2:
Mean alveolar Po2 (40% O2), mmHg	214 ± 31	125 ± 13	<0.0001*
Expiratory lung volume
ml	0.35 ± 0.09	0.47 ± 0.08	0.003*
ml/kg	16.0 ± 4.7	25.8 ± 7.0	0.003*
Inspiratory lung volume
ml	1.04 ± 0.16	1.21 ± 0.17	0.03*
ml/kg	47.2 ± 10.0	66.2 ± 16.9	0.002*
Pulmonary blood flow
ml/min	3.31 ± 1.15	4.41 ± 2.35	0.19
ml⋅(min·kg)−1	142 ± 47	242 ± 150	0.03*
DLCO-measured
ml⋅(min⋅mmHg)−1	0.0097 ± 0.0034	0.0136 ± 0.0042	0.05*
ml⋅(min⋅mmHg·kg)−1	0.43 ± 0.13	0.74 ± 0.26	0.001*
Septal tissue volume
ml	0.30 ± 0.21	0.20 ± 0.06	0.26
ml/kg	12.9 ± 8.5	10.4 ± 2.5	0.41
Rebreathing Mixture Containing 90% O2:
Mean alveolar Po2 (90% O2), mmHg	469 ± 62	292 ± 30	<0.0001*
Expiratory lung volume
ml	0.36 ± 0.12	0.41 ± 0.06	0.21
ml/kg	15.7 ± 5.8	22.1 ± 4.8	0.006*
Inspiratory lung volume
ml	1.05 ± 0.20	1.14 ± 0.15	0.20
ml/kg	46.6 ± 11.0	62.5 ± 14.6	0.005*
Pulmonary blood flow
ml/min	3.72 ± 1.20	3.72 ± 1.37	0.99
ml⋅(min·kg)−1	164 ± 57	193 ± 53	0.30
DLCO-measured
ml⋅(min⋅mmHg)−1	0.0063 ± 0.0025	0.0067 ± 0.0023	0.73
ml⋅(min⋅mmHg·kg)−1	0.27 ± 0.11	0.36 ± 0.09	0.11
Septal tissue volume
ml	0.18 ± 0.11	0.16 ± 0.07	0.91
ml/kg	7.6 ± 4.7	8.3 ± 3.2	0.84
Derived from the Roughton-Forster Model:
DMCO
ml⋅(min⋅mmHg)−1	0.0255 ± 0.0106	0.0228 ± 0.0111	0.73
ml⋅(min⋅mmHg·kg)−1	1.16 ± 0.43	1.23 ± 0.61	0.82
Pulmonary capillary blood volume
ml	0.0293 ± 0.0099	0.0569 ± 0.0278	0.47
ml/kg	1.34 ± 0.43	2.88 ± 1.30	0.13
DLCO-std
ml⋅(min⋅mmHg)−1	0.0093 ± 0.0024	0.0114 ± 0.0036	0.24
ml⋅(min⋅mmHg·kg)−1	0.43 ± 0.11	0.60 ± 0.14	0.03*

Body mass was measured on the day of study. Mean ± SD. DL_CO, lung diffusing capacity for carbon monoxide; DM_CO, membrane diffusing capacity for carbon monoxide; DL_CO-measured, measured DL_CO; DL_CO-std, standardized DL_CO (hematocrit 0.45, alveolar Po₂ 120 mmHg).

^*P ≤ 0.05: HA versus LA, unpaired t-test or one-way ANOVA.

^§Respiratory system compliance was measured at airway pressures between 10 and 30 cm H₂O.

Fig. 2.

Static airway pressure-lung volume relationships during lung inflation by ambient air showed a significantly larger lung volume in HA-acclimated than LA-acclimated deer mice (10 animals each). Mean ± SD. P = 0.005 HA vs. LA, by repeated measures ANOVA.

A representative set of expiratory gas concentrations obtained during rebreathing maneuvers in one animal is shown in Fig. 3. Complete mixing of the test gas bolus with resident alveolar gas was achieved rapidly during rebreathing; Ne concentration stabilized within 5 s whereas C₂H₂ and CO concentrations declined in a log linear fashion with time, similar as that observed in rats (78, 79), guinea pigs (76), and larger species. As expected at the respective altitudes of study, mean $P{A}_{O_2}$ during rebreathing was lower in the HA than in the LA group (Table 2). Mean CO back pressure was negligible (LA: 0.0016%, HA: 0.0008%).

Fig. 3.

A representative set of logarithmic expiratory gas concentration curves for neon (Ne), acetylene (C₂H₂), and carbon monoxide (CO) obtained during rebreathing maneuvers in one mouse. F_A, fractional alveolar gas concentration; F_A0, initial fractional alveolar gas concentration (at time zero).

While rebreathing the 40% O₂ mixture (mean $P{A}_{O_2}$ of 214 and 125 mmHg, LA and HA, respectively), lung volume and DL_CO were significantly higher in the HA group; pulmonary blood flow also tended to be higher but reached statistical significance only when normalized by body mass (Table 2).

While rebreathing the 90% O₂ mixture (mean $P{a}_O{}_{{}_2}$ 469 and 292 mmHg, LA and HA, respectively), none of the absolute measurements were statistically different between groups; however, body mass-specific lung volumes were significantly higher in the HA group (Table 2). The derived values of DM_CO, Vc, DL_CO-std, and alveolar septal tissue were not significantly different between groups (Table 2) except that body mass-specific DL_CO-std was higher in HA animals (P = 0.03). Average absolute DL_CO at a given pulmonary blood flow was not significantly different between HA and LA groups (Fig. 4).

Fig. 4.

Relationship between average DL_CO and pulmonary blood flow in individual animals measured while rebreathing an inspired gas mixture containing 40% O₂. At a given blood flow, DL_CO was not significantly different between HA and LA deer mice by analysis of the slopes and elevations of regression lines (P = 0.97 and P = 0.61, respectively). DL_CO, lung diffusing capacity for carbon monoxide; HA, high altitude; LA, low altitude.

Compare rebreathing and single-breath measurements in laboratory mice.

Mean end-inspiratory lung volume and DL_CO measured by the rebreathing technique agreed closely with that measured by the single-breath technique (Fig. 5).

Fig. 5.

Correlations between measurements using a rebreathing or a single-breath technique for end-inspiratory lung volume (A) and DL_CO (B) in C57BL/6 mice (9 animals, 2 measurements per animal). Solid line: linear regression. Dashed line: identity. DL_CO, lung diffusing capacity for carbon monoxide.

Allometry.

Mean lung volume, DL_CO, and pulmonary blood flow measured by the rebreathing technique in deer mice and outbred and inbred laboratory mice fit well with the respective allometric relationships across a wide range of species assembled from the literature (Fig. 6); these represent measurements obtained by single-breath, steady state, or rebreathing techniques at rest either under anesthesia (4, 8–11, 13, 17, 19, 23, 26, 29, 31, 33, 34, 36, 38, 43–45, 47, 48, 51, 52, 58, 65, 68, 70, 72, 74) or during conscious spontaneous breathing (1, 3, 12, 25, 39, 42, 53, 77).

Fig. 6.

Allometric relationships for mean lung volume (A), DL_CO (B), and pulmonary blood flow (C) in males of different species reported in the literature. Data on the following murine strains are from the present study: male C57BL/6, and male LA-acclimated deer mice. See text for references to the other source data. DL_CO, lung diffusing capacity for carbon monoxide; LA, low altitude.

Morphometric measurements in deer mice.

Postmortem lung fixation was performed in seven and six deer mice at LA and HA, respectively. Terminal body weight was 11% lower and absolute volume of the fixed lungs was 7% higher in HA mice; neither difference was significant (P = 0.14 and 0.32, respectively). However, body mass-specific lung volume was significantly (34%) higher in HA mice (P = 0.02, Table 3). Morphometric alveolar-capillary hematocrit measured by point counting was 23% higher, in contrast to the similar antemortem systemic hematocrit (Table 2). The arithmetic and harmonic mean septal thicknesses were similar between groups (Table 3). Volume density of fine parenchyma (per unit of lung volume) was significantly higher whereas volume density of nonfine parenchyma was lower in HA than LA deer mice (P < 0.001, Table 4). Volume density of alveolar ducts, but not alveolar sacs, was higher in the HA group (Fig. 7). Volume densities of type I epithelia were 37% higher and type II epithelia 32% lower in HA deer mice; thus, overall volume density of epithelia was unchanged. Volume density of interstitia was significantly (20%) lower in HA-acclimated deer mice whereas that of endothelia was not different between groups. Total volume densities of septum, septal tissue, capillary blood, and alveolar and capillary surface densities (surface area/lung volume ratios) were not different between groups (Table 4).

Table 3.

Morphometric measurements

Exposure	LA	HA	P Value
Number of animals	7	6
Terminal body mass, g	21.4 ± 1.91	19.1 ± 3.26	0.14
Total lung volume, immersion method
ml	0.74 ± 0.08	0.79 ± 0.12	0.32
ml/kg	31.3 ± 5.12	41.8 ± 3.53	0.02*
Morphometric hematocrit, %	58.2 ± 7.2	71.7 ± 6.6	0.01*
Mean arithmetic thickness of septum, µm	1.86 ± 0.31	1.82 ± 0.16	0.79
Mean arithmetic thickness of septal tissue, µm	0.87 ± 0.09	0.84 ± 0.79	0.52
Mean harmonic thickness of blood-gas barrier, µm	0.39 ± 0.05	0.38 ± 0.04	0.53

Values are means ± SD. HA, high altitude, LA, low altitude.

^*P ≤ 0.05: HA versus LA, unpaired t-test.

Table 4.

Volume and surface densities of alveolar septal structures

Volume Density of Alveolar Structure, per unit lung volume
Exposure	LA	HA	P Value
Fine parenchyma	0.87764 ± 0.02534	0.94244 ± 0.02621	0.0009*
Nonfine parenchyma	0.12237 ± 0.02534	0.05757 ± 0.02621	0.0009*
Alveolar ducts	0.27760 ± 0.0203	0.33106 ± 0.02736	0.002*
Alveoli	0.50219 ± 0.02502	0.51920 ± 0.02393	0.24
Septum	0.08706 ± 0.0069	0.08219 ± 0.00953	0.31
Total epithelium	0.01752 ± 0.0021	0.01869 ± 0.00201	0.33
Type I epithelium	0.00971 ± 0.00115	0.01334 ± 0.00116	0.0001*
Type II epithelium	0.00782 ± 0.0016	0.00536 ± 0.0021	0.04*
Interstitium	0.03917 ± 0.0026	0.03153 ± 0.00669	0.02*
Endothelium	0.00841 ± 0.00181	0.00999 ± 0.00161	0.13
Septal tissue	0.0651 ± 0.00387	0.06021 ± 0.00764	0.16
Capillary blood	0.02196 ± 0.00482	0.02199 ± 0.00475	0.99
Surface Density of Alveolar Structure, per unit lung volume/cm
Alveolar surface	955.66 ± 138.81	906.28 ± 83.88	0.46
Capillary surface	548.04 ± 71.79	523.77 ± 62.22	0.53

Values are means ± SD. HA, high altitude; LA, low altitude.

^*P ≤ 0.05: HA versus LA, unpaired t-test.

Fig. 7.

Representative microscopic images in deer mice acclimated to low (LA) or high (HA) altitude. Top: light micrographs of distal lung morphology. Toluidine blue stain (bar = 100, 50, 20 µm, left to right). Bottom: electron micrographs of alveolar septal morphology (bar = 5 µm). Symbol indicates a representative feature: *Mitochondria within a type II epithelial cell. C, capillary erythrocyte; I, interstitial cell; L, lamellar body.

In HA-acclimated deer mice, absolute volume of nonfine parenchyma was 50% lower than that in LA mice (P = 0.004) whereas that of fine parenchyma was borderline higher (15%, P = 0.07). Absolute volumes of the alveolar duct, type I epithelia, and endothelia were significantly higher (by 47%, 48%, and 28%, respectively) in HA deer mice than that in LA deer mice, whereas the volume of interstitia was 15% lower (P = 0.05), and the volume of type II epithelia was nonsignificantly lower (P = 0.08). (Table 5, Fig. 8). Total volume of the septum, extravascular septal tissue, or alveolar-capillary blood was not significantly different between groups. Total alveolar and capillary surface areas were not different between groups (Table 5). These results are consistent with acinar airway enlargement, reduced dimensions of small bronchioles, arterioles and venules, and volume shifts among alveolar septal constituents in HA-acclimated deer mice.

Table 5.

Absolute volumes and surface areas of septal structures

	LA	HA	P Value
Absolute Volume, mm3
Fine parenchyma
mm3	647.176 ± 68.986	746.795 ± 110.114	0.07
ml/kg	30.500 ± 4.231	39.291 ± 2.487	0.001*
Nonfine parenchyma
mm3	90.253 ± 21.353	46.038 ± 23.230	0.004*
ml/kg	4.294 ± 1.312	2.464 ± 1.341	0.030*
Alveolar ducts
mm3	179.414 ± 20.873	263.184 ± 49.582	0.002*
ml/kg	8.435 ± 1.039	13.798 ± 1.311	0.000005*
Alveoli
mm3	396.379 ± 50.495	411.514 ± 62.272	0.64
ml/kg	18.706 ± 3.090	21.661 ± 1.797	0.06
Septum
mm3	64.061 ± 6.989	64.356 ± 5.437	0.93
ml/kg	3.018 ± 0.406	3.423 ± 0.391	0.095
Total epithelium
mm3	12.904 ± 1.868	14.725 ± 2.049	0.12
ml/kg	0.612 ± 0.129	0.779 ± 0.088	0.022*
Type I epithelium
mm3	7.119 ± 0.763	10.558 ± 1.752	0.0006*
ml/kg	0.337 ± 0.056	0.554 ± 0.035	0.00001*
Type II epithelium
mm3	5.785 ± 1.421	4.167 ± 1.630	0.08
ml/kg	0.275 ± 0.082	0.224 ± 0.089	0.31
Interstitium
mm3	28.783 ± 2.359	24.609 ± 4.458	0.05*
ml/kg	1.357 ± 0.171	1.302 ± 0.212	0.62
Endothelium
mm3	6.146 ±1.160	7.845 ± 1.333	0.03*
ml/kg	0.288 ± 0.052	0.417 ± 0.072	0.003*
Septal tissue
mm3	47.833 ± 3.589	47.179 ± 5.266	0.8
ml/kg	2.257 ± 0.294	2.498 ± 0.223	0.13
Capillary blood
mm3	16.228 ± 4.115	17.177 ± 3.442	0.66
ml/kg	0.760 ± 0.173	0.925 ± 0.243	0.18
Absolute Surface Area, cm2
Alveolar surface
cm2	704 ± 123	715 ± 106	0.86
m2/kg	3.332 ± 0.738	3.773 ± 0.335	0.21
Capillary surface
cm2	404 ± 65	411 ± 45	0.83
m2/kg	1.899 ± 0.319	2.19 ± 0.335	0.14

Values are means ± SD. HA, high altitude; LA, low altitude.

^*P ≤ 0.05: HA versus LA, unpaired t-test.

Fig. 8.

Postmortem absolute volumes of the lung, alveolar ducts and alveolar sacs (A), fine and nonfine parenchyma (B), and alveolar septal constituents (C) in deer mice acclimatized to high (HA) or low (LA) altitude (n = 6 and 7 animals, respectively). Mean ± SD. Significant P values by unpaired t-test are shown.

DISCUSSION

Summary of findings.

These are the first detailed examinations of lung function and ultrastructure in a subspecies of deer mice native to HA. Our goal was to determine the adaptive response in deer mice bred at LA and recently acclimated to HA and to relate lung architecture to cardiopulmonary function and aerobic performance. To accomplish this goal, deer mice captive-bred at LA for 18–20 generations were acclimated as adults to HA (3,800 m) for 8 wk whereas matched control deer mice were acclimated to LA (252 m). To assess alveolar-capillary function, we validated a rebreathing technique to measure lung volume, pulmonary blood flow, and diffusion simultaneously in mice at their respective altitude of residence. Major findings are: 1) deer mice maintained their body mass and V̇o_2max at HA relative to LA baseline; 2) HA deer mice did not exhibit an increase in systemic hematocrit that is typical of lowland species acclimatized to HA. However, postmortem anatomical alveolar capillary hematocrit was significantly higher; 3) HA deer mice exhibited a 40% larger static lung volume at a given airway pressure in ambient hypoxia and higher lung volumes and DL_CO measured while rebreathing a 40% O₂ gas mixture but not while rebreathing a 90% O₂ mixture. Specific respiratory system compliance was similar; 4) the relationships of absolute DL_CO-std, DM_CO, and Vc with respect to pulmonary blood flow were not significantly different between groups; 5) postmortem body mass-specific lung volume but not absolute lung volume was larger in HA-acclimated deer mice. Alveolar ducts were enlarged whereas the volumes of small airways and microvessels were smaller without significant differences in whole-lung volumes of extravascular septal tissue, capillary blood, or alveolar-capillary surface areas, compared with LA-acclimated deer mice, i.e., there was no compensatory lung growth in HA mice; 6) HA-acclimated mice exhibited alveolar septal remodeling with higher volumes of alveolar type I epithelia and endothelia, and a lower volume of interstitia, without any change in harmonic mean barrier thickness, an index of blood-gas barrier resistance to diffusion. Thus, deer mice bred and raised to adulthood at LA adapt very well to subsequent HA residence via nonstructural ($P{A}_{O_2}$-dependent increase in lung volume) and structural (distal airspace enlargement and septal remodeling) adjustments such that there is no significant decrement in cardiopulmonary O₂ transport or aerobic capacity at HA compared with control animals kept at an LA.

Critique of methodology.

Body mass and V̇o_2max in our deer mice were in a similar range as that reported previously (18). Using deer mice that have been vivarium-bred at sea level ensures consistent genetic background, age, sex, and living condition among individual animals and avoids the risks of exposing research personnel to Hanta virus that is carried by wild deer mice and is responsible for the lethal human Hantavirus pulmonary syndrome. To ensure that results reflect actual physiological response, baseline measurements were performed at the LA whereas postacclimation measurements were performed at each animal’s place of residence (HA or LA). DL_CO was expressed at standardized $P{A}_{O_2}$ and hematocrit following established methodology. Separate laboratory murine strains were also studied to compare measurements obtained using the rebreathing technique with that using a single-breath technique and to establish that rebreathing measurements fit with the respective allometric relationships among murine strains and across multiple species over a wide range of body sizes. Key morphometric results fall within the body mass-specific range as that previously reported in the literature in laboratory mice (32, 37, 69, 71).

Physiological adaptation to HA in the lungs of deer mice.

Wild deer mice remain active throughout the year in the extreme HA habitat where they must overcome ambient hypoxia, cold temperatures, and food scarcity. They have evolved novel hemoglobin isoforms with increased O₂ binding affinity to surmount partially the low ambient O₂ availability (7, 27, 64); however, these isoforms are not always expressed and are insufficient to support the aerobic demands of deer mice fully, necessitating other strategies to increase O₂ uptake and delivery. Hammond et al. (18) reported that deer mice that were raised and tested at LA exhibit higher aerobic capacity and basal metabolic rate than those that were either raised or acclimated as adults and tested at HA. Aerobic performance at HA is similar between HA-raised deer mice and LA-raised mice acclimated to HA as adults. These results suggest that in addition to genetic and developmental factors, postnatal acclimatization plays an important role in deer mice adaption to HA. Consistent with the present findings, Shirkey and Hammond (62) observed that V̇o_2max of deer mice on arrival at HA is lower than it is after acclimation, indicating adaptation. In our deer mice, body mass was maintained following HA-acclimation but increased significantly in the LA control group. As the animals were fed the same diet, the findings likely reflect lower activity level and/or higher food intake in the LA group. Hammond et al. (50) also demonstrated larger lung volume and heart mass in deer mice acclimated to hypoxia that correlated with aerobic performance, suggesting that increased cardiopulmonary O₂ transport capacities helped HA-acclimated deer mice adapt to ambient hypoxia.

Direct comparisons of postacclimation lung volumes and DL_CO between the HA and LA groups were complicated by a large difference in $P{A}_{O_2}$ at the two altitudes of measurement. In deer mice acclimatized to and studied at HA, static PV relationship while breathing ambient air was clearly elevated; lung volume was significantly higher when measured while rebreathing a 40% O₂ mixture (mean $P{A}_{O_2}$ 125 mmHg) but not while rebreathing a 90% O₂ mixture (mean $P{A}_{O_2}$ 292 mmHg). The differential findings may reflect an acute physiological response to changes in $P{A}_{O_2}$. Both acute and chronic hypoxia induce adaptation in neural ventilatory drive and lung mechanics. In human subjects, acute reductions in $P{A}_{O_2}$ increase total lung capacity, end-expiratory lung volume, residual volume, and static expiratory lung compliance; the effect is thought to be mediated by a loss of smooth muscle tone in distal airways that mildly reduces lung elastic recoil and airway resistance (59) and by a reduction in total pulmonary blood volume because of pulmonary vasoconstriction (41).

Our usual technique for standardizing DL_CO is to calculate DM_CO and Vc using the RF model, then use the derived DM_CO and Vc to express DL_CO at constant conditions (DL_CO-std, at $P{A}_{O_2}$ 120 mmHg, hematocrit 45%). When this was done, DL_CO-std in HA mice was modestly (23%) though nonsignificantly higher. Others have reported modest increases in DL_CO in human lowlanders acclimatized to HA; Agostoni et al. (2) found a 6% higher adjusted DL_CO measured at rest following 2 wk acclimation to 5,400 m altitude. Taylor et al. (67) observed a 10% higher resting DL_CO following ~6 wk acclimation at up to 5,150 m altitude. In our HA-acclimated mice, the lack of significant differences in DL_CO-std, DM_CO, and Vc is internally consistent with postmortem data showing nondifference in whole-lung alveolar tissue-capillary volumes, surface areas, and barrier thickness compared with LA-acclimated mice.

Because O₂ and CO compete for the same binding sites on hemoglobin, at a constant inspired O₂ concentration the lower O₂ tension at HA results in an increase in DL_CO of ~0.35% for each mmHg of reduction in $P{A}_{O_2}$. To verify the nonsignificantly higher DL_CO-std observed in HA mice further, we empirically adjusted the measured DL_CO (DL_CO-measured) to a constant $P{A}_{O_2}$ (120 mmHg) without using DM_CO and Vc: DL_{CO-altitude adjusted} = DL_CO-measured × [1.0 + 0.0035·($P{A}_{O_2}$−120)]. This adjustment indeed confirms a nonsignificantly higher (9%) altitude-adjusted DL_CO in HA mice (LA: 0.0127 ± 0.0037, HA: 0.0138 ± 0.0043 ml·(min·mmHg)⁻¹, P = 0.52). Thus, the higher DL_CO in HA mice measured while rebreathing at a mean $P{A}_{O_2}$ of 125 mmHg (but not 292 mmHg) is likely the result of both the physical effect of hypobaric hypoxia and an acute Po₂-dependent increase in lung volume.

Sources of variability in DL_CO, DM_CO, and Vc should also be noted. There is inherent imprecision in the dilution of a small, expired sample volume. The RF model assumes DM_CO and Vc to be serial independent parameters. However, finite element modeling of alveolar-capillary diffusion (21) has shown intrinsic interdependence between DM_CO and Vc, i.e., increasing Vc recruits capillaries and effective membrane surface for diffusion, thereby increasing DM_CO. The RF model further assumes that DM_CO and Vc are independent of $P{A}_{O_2}$ whereas finite element modeling has shown a decrease in effective DM_CO as $P{A}_{O_2}$ increases (22). Errors arising from model assumptions are normally minor for small variations in $P{A}_{O_2}$ but may be magnified across a large range of $P{A}_{O_2}$ as in this study, a fact that likely accentuated the variability of derived DM_CO and Vc values.

Structural adaptation to HA in the lungs of deer mice.

In contrast to the elevated PV relationship in HA deer mice breathing ambient air (Fig. 2), postmortem absolute volume of the fixed lung was not significantly different between groups. As the animals received supplemental O₂ during the terminal procedure, this finding is consistent with a $P{A}_{O_2}$-dependent change in antemortem lung volume described above. However, body mass-specific lung volume was significant higher in HA animals. There was preferential enlargement of alveolar ducts as has been reported in other species exposed to chronic hypoxia (79) whereas distal conducting structures were smaller such that there were no net gains in whole-lung alveolar septal tissue or capillary volumes or gas exchange surface areas in HA-acclimated deer mice. The lack of whole-lung compensatory tissue-capillary growth or a change in harmonic mean barrier thickness is consistent with the absence of significant differences in DL_CO-std, DM_CO, and Vc between HA and LA deer mice and is consistent with that in HA-acclimated adult canines (30, 31). Findings in adult animals contrast with that in young, growing animals where HA residence or chronic hypoxia exposure before reaching somatic maturity leads to lung volume enlargement and accelerated gains in alveolar-capillary tissue volumes and surface areas (6, 30, 31, 75). In addition to enlarged acinar airways, HA deer mice showed clear evidence of alveolar septal remodeling, including modest shifts in volume densities and absolute volumes of septal constituents toward more type I and less type II epithelia without a change in total epithelia volume and more endothelia with less interstitia to minimize diffusive resistance of the blood-gas barrier and to help prevent a decrement in V̇o_2max at HA.

Systemic hematocrit was similar in our HA- and LA-acclimatized deer mice. In contrast, Hammond et al. (18) reported significantly higher systemic hematocrits and hemoglobin concentrations in LA-raised deer mice acclimatized to the same HA (3,800 m) over 3 wk compared with LA-raised/LA-acclimated deer mice. These divergent findings may be related to differences in the duration of HA exposure or the timing and manner of blood sampling; in the earlier study by Hammond et al. (18) blood was drawn via retro-orbital puncture shortly after the animal exercised to V̇o_2max on the treadmill whereas in the present study blood was drawn via cardiac puncture under anesthesia. Despite a similar systemic hematocrit, postmortem alveolar-capillary hematocrit measured by point-counting was higher in HA-acclimated deer mice, suggesting relatively greater erythrocyte sequestration within alveolar microvasculature. The difference between systemic and morphometric hematocrit partly reflects trapping of erythrocytes as pulmonary microvascular blood flow ceases following the overdose of euthanasia solution. Because systemic hematocrit was not different between HA and LA groups, a higher morphometric hematocrit also suggests higher pulmonary microvascular outflow resistance, perhaps a result of hypoxic pulmonary microvascular reactivity (16), in the HA group.

Other adaptations in O₂ transport.

In our anesthetized HA-acclimated deer mice, body mass-specific pulmonary blood flow was higher than in LA-acclimated animals; the difference may become accentuated when awake and active. The skeletal muscle of native highland deer mice populations exhibit a higher respiratory capacity via more oxidative fibers containing a higher volume density of subsarcolemmal mitochondria situated closer to capillaries (61). Upon acclimation to hypoxia, native highland deer mice exhibit higher diaphragmatic respiratory capacity via increased capillary density (66) and mitochondrial quantity (35) whereas native lowlanders exhibit additional increases in citrate synthase activity and release of reactive O₂ species (35). These peripheral adjustments augment O₂ unloading and tissue diffusion-utilization, counterbalancing the effects of a higher hemoglobin-O₂ affinity that promotes O₂ loading in the lung (28).

Comparison among murine strains.

The arithmetic alveolar septal thickness in LA-acclimated deer mice (1.86 µm) is considerably lower than that in Sv129 mice measured under electron microscopy (3.48 µm, Hsia laboratory, unpublished observations) and CD1 (2.4 µm) (60) and C57/BL6 (~3.4 µm) (69) mice measured under light microscopy whereas the arithmetic thickness of septal tissue barrier in deer mice (0.87 µm) is lower than that in white (1.25 µm) and Japanese Waltzing (1.08 µm) mice (15); differences are at least partly because of a higher alveolar surface area relative to septum volume in deer mice, which facilitates diffusive O₂ uptake. In contrast to the lack of HA-stimulated lung growth in deer mice, Darwin’s leaf-eared mice (Phyllotis darwini) living at HA show enlarged alveolar septal cell volumes and mass-specific surface areas compared with the same species living at sea level (46). Laboratory (C57BL/6) mice also exhibit accelerated lung growth during chronic intermittent hypoxia (49). As the generation of new lung tissue requires higher metabolic energy expenditure than remodeling of existing lung tissue, deer mice are far more efficient at defending O₂ transport via both inborn and hypoxia-induced adaptations.

Conclusion and significance.

We assessed aerobic capacity and lung structure and function in deer mice (P. maniculatus sonoriensis) that have been bred for many generations and raised to adulthood at LA and subsequently acclimated to HA to determine how these animals accommodate their native HA habitat. We found that these deer mice remain well adapted to HA, exhibiting no decrement in V̇o_2max at HA and only modest structural alterations following acclimation. The major adaptive mechanisms during acclimation are alveolar O₂ tension-dependent increases in lung volume, distal airspace enlargement, and alveolar septal remodeling; these changes occur without systemic polycythemia or active alveolar tissue-capillary growth and are sufficient to maintain pulmonary diffusive O₂ uptake and a normal aerobic capacity to meet the demands for survival in a harsh environment. Beyond these structure-function relationships, our study has ecological implications. Over the last century, more than half of the 28 small mammalian species measured across a 3,000-m altitude gradient in Yosemite National Park in California (~100 miles to the west of our study site at White Mountain) have moved up in elevation where daily minimum temperatures are cooler (40). This is consistent with an observed 3°C increase in daily minimum temperatures over the same period. Further increases in minimum temperatures in this region are predicted, which will continue to alter the species migration pattern and the predator/prey relationships at mid to high elevations. Understanding the phenotypic plasticity and species-specific parameters of adaptation will be critical in predicting future distribution and welfare of deer mice populations at HAs.

GRANTS

This work was supported by the National Science Foundation Grant No. 145700 (to K. Hammond) and the National Heart, Lung and Blood Institute Grant Nos. U01-HL-111146 and R01-HL-134373 from the National Institutes of Health (both to C. Hsia).

DISCLAIMERS

The content is solely the authors’ responsibility and does not necessarily represent official views of the funding agencies.

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

K.A.H. and C.C.W.H. conceived and designed research; D.M.D., K.C., H.L., C.Y., J.D., C.D.T., P.R., K.A.H., and C.C.W.H. performed experiments; D.M.D., K.C., H.L., C.Y., J.D., C.D.T., K.A.H., and C.C.W.H. analyzed data; D.M.D., K.C., C.D.T., K.A.H., and C.C.W.H. interpreted results of experiments; D.M.D., K.C., and C.C.W.H. prepared figures; D.M.D., K.C., and C.C.W.H. drafted manuscript; K.A.H. and C.C.W.H. edited and revised manuscript; D.M.D., K.C., C.D.T., K.A.H., and C.C.W.H. approved final version of manuscript.