Journal of Mammalogy, 88(1):24–31, 2007

HEMOGLOBIN FUNCTION AND PHYSIOLOGICAL ADAPTATION

TO HYPOXIA IN HIGH-ALTITUDE MAMMALS
JAY F. STORZ*
School of Biological Sciences, University of Nebraska, Lincoln, NE 68588, USA

Understanding the biochemical mechanisms that enable high-altitude animals to survive and function under

Downloaded from https://academic.oup.com/jmammal/article/88/1/24/927083 by guest on 22 December 2021

conditions of hypoxic stress can provide important insights into the nature of physiological adaptation. Evidence
from a number of high-altitude vertebrates indicates that modifications of hemoglobin function typically play
a key role in mediating an adaptive response to chronic hypoxia. Because much is known about structure–
function relationships of mammalian hemoglobins and their physiological role in oxygen transport, the study of
hemoglobin variation in high-altitude mammals holds much promise for understanding the nature of adaptation
to hypoxia from the level of blood biochemistry to the level of whole-organism physiology. In this review I 1st
discuss basic biochemical principles of hemoglobin function and the nature of physiological adaptation to high-
altitude hypoxia in mammals. I then discuss a case study involving a complex hemoglobin polymorphism in
North American deer mice (Peromyscus maniculatus) that illustrates how integrative studies of protein function
and fitness-related physiological performance can be used to obtain evolutionary insights into genetic
mechanisms of adaptation.

Key words: adaptation, altitude, deer mouse, ecological physiology, evolutionary physiology, hemoglobin, hypoxia, natural
selection, oxygen transport, Peromyscus maniculatus

High-altitude environments present a number of physiolog- different subunit polypeptides are encoded by different sets
ical challenges for endothermic animals, as they are character- of duplicated genes that are located on different chromosomes
ized by a lower partial pressure of oxygen (PO2 ) and lower (Hardison 2001). Because much is known about structure–
ambient temperatures compared to low-altitude environments function relationships of mammalian hemoglobins and their
at similar latitudes. The reduced PO2 at high altitude results in role in oxygen transport (reviewed by Perutz 1983, 2001;
reduced oxygen loading in the lungs such that the blood may Poyart et al. 1992; Weber and Fago 2004), the study of hemo-
not carry a sufficient supply of oxygen to the cells of respiring globin variation in species that are native to high altitude pro-
tissues (Bencowitz et al. 1982; Bouverot 1985; Turek et al. vides a unique opportunity to understand the nature of genetic
1973). This reduced level of tissue oxygenation can impose adaptation to hypoxic stress from the level of blood bio-
severe constraints on aerobic metabolism and may therefore chemistry to the level of whole-organism physiology. In this
influence an animal’s food requirements, water requirements, review I 1st provide some background information about hemo-
the capacity for sustained locomotor activity, and the capacity globin function and the nature of physiological adaptation to
for internal heat production. high-altitude hypoxia. I then discuss a case study involving a
Although the genetic basis of hypoxia tolerance has yet to be complex hemoglobin polymorphism in deer mice (Peromyscus
fully elucidated in any vertebrate species, evidence from a maniculatus) that illustrates how integrative studies of protein
number of mammals, birds, and amphibians indicates that function and fitness-related physiological performance can be
modifications of hemoglobin function often play a key role used to obtain evolutionary insights into genetic mechanisms
in mediating an adaptive response to high-altitude hypoxia of adaptation.
(Perutz 1983). In all vertebrates other than cyclostomes, the
hemoglobin protein is a heterotetramer, composed of 2 a-chain
and 2 b-chain polypeptides. In mammals and birds, the CIRCULATORY ADJUSTMENTS TO HYPOXIC STRESS
When atmospheric air is drawn into the alveoli of the lungs,
oxygen is under a higher partial pressure than in the pulmonary
* Correspondent: jstorz2@unl.edu capillaries, and it therefore diffuses across the respiratory
membrane into the arterial bloodstream. Once oxygen has en-
Ó 2007 American Society of Mammalogists tered the bloodstream, it is immediately bound to hemoglobin
www.mammalogy.org in the red blood cells for transport to the oxygen-consuming
24
February 2007 SPECIAL FEATURE—PHYSIOLOGICAL ADAPTATION TO HIGH ALTITUDE 25

cells of respiring tissues. The gas exchange ends at the tissue

capillaries as oxygen, released by hemoglobin, diffuses across
the capillary walls through the interstitial fluid to the cells. At
the same time, CO2 and other metabolic end-products enter the
bloodstream and are transported to the lungs by the opposite
route.
At high altitude, the arterial PO2 is reduced compared to what
it would be in an oxygen-rich sea-level environment and it
becomes critically important to minimize the corresponding
reduction in tissue oxygenation. In the cascade of PO2 across
different compartments of the gas-exchange system, there are 2

Downloaded from https://academic.oup.com/jmammal/article/88/1/24/927083 by guest on 22 December 2021

main steps where circulatory adjustments can help minimize
the inevitable reduction in tissue PO2 : the gradient between
alveolar gas and arterial blood, and that between capillary
blood and the tissues. The PO2 gradient between alveolar gas
and arterial blood is normally attributable to a small amount of
venous admixture and unequal matching of ventilation to per-
fusion in the lungs (that is, a mismatch between the diameter of
the airways and the diameter of the pulmonary blood vessels).
The PO2 gradient between capillary blood and the tissues results
from unloading of oxygen in the tissue capillary bed. Tissue
gas exchange begins at the arterial inlet to the capillary bed,
and the PO2 falls rapidly from the arterial side to the venous
FIG. 1.—A schematic representation of the oxygen dissociation
side as oxygen diffuses from the high PO2 of the blood to the
curve under physiochemical conditions prevailing in arterial blood
low PO2 of the interstitial fluid. A meaningful estimate of mean (open circle) and mixed venous blood (solid circle). The y-axis
capillary PO2 and the gradient to the cells can be obtained from measures the oxygen concentration in the blood (CO2 ) and the x-axis
measurements of arterial and mixed venous PO2 . The arterial- measures the partial pressure of oxygen in the blood (PO2 ). CaO2 and
mixed-venous PO2 gradient can be minimized by increasing CvO2 are the oxygen concentrations in arterial and mixed venous
the circulatory conductance of oxygen in the blood. In high- blood, respectively. PaO2 and PvO2 are the partial pressures of oxygen
altitude mammals, one of the primary mechanisms for increas- in arterial and mixed venous blood, respectively. The slope of the line
ing the circulatory conductance of oxygen involves increasing joining the arterial and mixed venous points on the curve denotes the
the oxygen-binding affinity of hemoglobin. blood oxygen capacitance coefficient (bbO2 in equations 2 and 3).

ADAPTIVE MODIFICATION OF HEMOGLOBIN This capacitance coefficient is defined as the slope of the line
FUNCTION IN HYPOXIA-TOLERANT MAMMALS connecting the arterial point to the mixed venous point on the
When the arterial PO2 is reduced because of high-altitude oxygen-hemoglobin dissociation curve (ODC; Fig. 1). Because
hypoxia, the transport of oxygen by blood has to serve 2 inter- of the nonlinear relationship between oxygen concentration and
related functions: it must maintain a sufficient flux of oxygen to PO2 in blood (which gives rise to the sigmoid shape of the
meet metabolic demand, and it must also maintain an adequate ODC), the capacitance coefficient bbO2 is not constant.
pressure gradient for oxygen diffusion from the lungs to the The maintenance of an adequate pressure gradient for tissue
cells of respiring tissues (Bouverot 1985; Monge and León- oxygenation can be understood by rearranging equation 2 as
Velarde 1991). The 1st of these 2 functions is described by the follows:
following Fick’s convection equation: PvO2 ¼ PaO2 2f1=½bbO2 ðQb=VO2 Þg; ð4Þ
V_ O2 ¼ QbðCaO2
CvO2 Þ; ð1Þ where PvO2 is viewed as the critical pressure at the vascular
supply source for oxygen diffusion into the cells of respiring
where V_ O2 is the rate of oxygen consumption, Qb is the total
tissues (Bouverot 1985). The product bbO2 (Qb/VO2 ) is the
cardiac blood flow, and CaO2 and CvO2 are the oxygen con-
specific oxygen blood conductance. Under hypoxia, an
centrations in arterial and mixed venous blood, respectively.
This is equivalent to the following: increased oxygen blood conductance helps to maintain
a sufficient driving force for oxygen diffusion to the tissues.
V_ O2 ¼ Qb  bbO2 ðPaO2
PvO2 Þ; ð2Þ One of the most important mechanisms to compensate for
reduced arterial PO2 at high altitude involves shifting the shape
where PaO2
PvO2 is the arterial-mixed-venous PO2 difference,
and position of the ODC (Luft 1972). The ODC describes how
and bbO2 , called the blood oxygen capacitance coefficient
the reversible binding of oxygen by hemoglobin depends on
(Dejours et al. 1970), is defined by the ratio
PO2 in the blood. At low PO2 in the bloodstream, the arterial and
bbO2 ¼ ðCaO2
CvO2 Þ=ðPaO2
PvO2 Þ: ð3Þ mixed venous points on the ODC would be shifted leftward to
26 JOURNAL OF MAMMALOGY Vol. 88, No. 1

Downloaded from https://academic.oup.com/jmammal/article/88/1/24/927083 by guest on 22 December 2021

FIG. 2.—Theoretical influence of a change in hemoglobin–oxygen
affinity on the oxygen dissociation curve under moderate hypoxia
(open symbols) and severe hypoxia (closed symbols). The y-axis
measures the oxygen concentration in whole blood (CO2 ) and the x-
axis measures the partial pressure of oxygen (PO2 ). The symbols
labeled with arrows denote the values for mixed venous blood, and the
unlabeled symbols denote values for arterial blood. For clarity, FIG. 3.—The oxygen dissociation curves of llamas and vicuñas are
differences in arterial-mixed-venous oxygen concentrations are left-shifted relative to the curves of other terrestrial mammals that are
identical in each of the 3 oxygen dissociation curves. At a given not physiologically adapted to chronic hypoxia. The left-shifted curves
level of hypoxia, note how changes in hemoglobin–oxygen affinity are attributable to the fact that these 2 high-altitude species have
influence PO2 in mixed venous blood (arrows) and the blood oxygen hemoglobins with especially high oxygen-binding affinities (Hall et al.
capacitance coefficient (the slope of the line joining the arterial and 1936). Modified from Schmidt-Nielsen (1990).
mixed venous points on each oxygen dissociation curve). Modified
from Bouverot (1985). Physiological experiments by Banchero and Grover (1972)
clearly demonstrate the advantage of a right-shifted ODC (high
the steeper portion of the curve (Fig. 2). As a result, the slope P50) under conditions of moderate hypoxia and the advantage
of the line joining the arterial and mixed-venous points (the of a left-shifted ODC (low P50) under conditions of severe
capacitance coefficient, bbO2 in equations 2 and 3) would hypoxia. In comparisons between llamas (Lama glama; a high-
undergo a dramatic increase. This increase in bbO2 would altitude native with low P50) and sheep (Ovis aries, a species
produce an automatic increase in the blood oxygen conduc- that has a comparatively high P50), blood oxygen capacitance
tance, thereby preventing PO2 from falling too low under was higher in sheep at simulated altitudes of 1,600–2,800 m,
hypoxia. The position and shape of the ODC, and therefore but was much higher in llamas at the maximal simulated
blood oxygen capacitance, can be modified by changes in altitude of 6,400 m. Over the full range of altitudes, the decline
hemoglobin concentration in the blood or by changes in the in PvO2 was only 8 torr in llamas compared to 26 torr in sheep.
oxygen-binding affinity of hemoglobin. In mammals, the For any given species it is difficult to specify the exact
former mechanism appears to be more important in the accli- altitude at which a left-shifted ODC becomes advantageous.
mation response to hypoxia in species that are native to However, because small mammals are generally characterized
lowland environments, whereas the latter mechanism appears by higher mass-specific metabolic rates (and therefore, in-
to be more important in high-altitude natives that are gene- trinsically high oxygen demands), the altitude at which a left-
tically adapted to chronic hypoxia (Bartels and Baumann 1977; shifted curve becomes advantageous should be lower than in
Bullard 1972; Bunn 1980; Hochachka and Somero 2002; the case for large mammals (Snyder 1981; Turek et al. 1973).
Lenfant 1973; Monge and León-Velarde 1991). Based on theoretical considerations and experimental results,
Under conditions of moderate hypoxia, a right-shifted ODC alpine mammals that are genetically adapted to high-altitude
positions the arterial and mixed venous points on a steeper hypoxia might be expected to have chronically left-shifted
slope, thereby increasing bbO2 (Fig. 2). This increase in bbO2 is ODCs (low P50). This prediction is borne out by surveys of
expected to increase PvO2 , the overall index of tissue blood oxygen affinity in a diverse range of terrestrial mammals,
oxygenation. By contrast, under severe hypoxia, a left-shifted including species that inhabit high-altitude environments and
ODC positions the arterial and mixed venous points on those that live in the hypoxic conditions of subterranean
a steeper slope (Shappell and Lenfant 1975; Turek et al. burrows (Bullard 1972; Hall et al. 1936; Monge and León-
1973; Fig. 2). The rightward shift in the ODC involves an Velarde 1991). For example, high-altitude natives such as
increase in the PO2 required for half-saturation of hemoglobin llamas (L. glama) and vicuñas (Vicugna vicugna) are
(P50), whereas the leftward shift involves a decrease in P50. characterized by left-shifted ODCs relative to other terrestrial
February 2007 SPECIAL FEATURE—PHYSIOLOGICAL ADAPTATION TO HIGH ALTITUDE 27

TABLE 1.—Comparison of blood oxygen affinities (as indexed by which suppresses 2 binding sites for 2,3-biphosphoglycerate
P50) and amino acid differences in the a- and b-globin subunits of per tetramer. Among the Andean camelids, the vicuña inhabits
hemoglobin in 1 lowland camelid (Camelus dromedarius) and 4 high- the highest elevational zone (4,500–5,000 m) and it also exhibits
altitude camelid species (Lama guanicoe, L. glama, L. pacos, and the highest blood oxygen affinity (P50 ¼ 17.5). The especially
Vicugna vicugna). Data for humans are provided for comparison. P50 high oxygen affinity of vicuña hemoglobin appears to be
is the partial pressure of oxygen in the bloodstream at which he-
attributable to an Ala!Thr substitution at a130 and a His!Asn
moglobin is 50% saturated. Differences in the sequence of a- and b-
globin polypeptides among the different species are from Piccinini substitution at b2 (Clementi et al. 1994; Piccinini et al. 1990;
et al. (1990). The position of the amino acid residues in the primary Poyart et al. 1992). One important conclusion of these molec-
structure of the globin polypeptides is given in parentheses. Modified ular studies is that a small number of amino acid substitutions
from Poyart et al. (1992). at key positions may be sufficient to adapt the functional
properties of hemoglobin to the hypoxic conditions of high

Downloaded from https://academic.oup.com/jmammal/article/88/1/24/927083 by guest on 22 December 2021

Camelus Lama Vicugna Homo altitude (Poyart et al. 1992).
dromedarius guanicoe L. glama L. pacos vicugna sapiens
With respect to possible adaptive modifications of hemoglo-
P50a 21.5 22.2 20.3 20.3 17.5 26.0 bin function, we generally know what to expect in the case of
Amino acid differences mammalian species such as the vicuña that are exclusively
a-chain restricted to high-altitude environments. In contrast, it is not clear
8(A6) Thr Ala Ala Ala Ala Thr what we should expect in the case of species that inhabit a broad
0(A8) Val Ile Ile Ile Val Val range of different altitudes across their geographic range. The
23(B4) Glu Asp Asp Asp Asp Glu deer mouse (P. maniculatus) is one such species. P. maniculatus
120(H3) Ser Ala Ala Ala Ala Ala
122(H5) His His Asp His His His
has one of the broadest altitudinal distributions of any North
130(H13) Ala Ala Ala Ala Thr Ala American mammal, occurring in lowland prairie and deserts as
b-chain
well as alpine environments at elevations over 4,300 m (Hall
2(NA2) His Asn Asn Asn Asn His
1981; Hock 1964). Because deer mice are characterized by one
76(E20) Asn Ser Ser Ser Ser Ala of the most complex and extensive hemoglobin polymorphisms
a of any mammal, this species is an exemplary study organism for
See Piccinini et al. (1990) for details regarding the experimental measurement of P50.
elucidating how modifications of hemoglobin function contrib-
ute to adaptive variation in physiological performance under
mammals that are not physiologically adapted to chronic hypoxic stress (reviewed by Snyder 1981). Indeed, the adaptive
hypoxia (Fig. 3). Similarly, birds that fly at extremely high significance of hemoglobin polymorphism in deer mice served
altitudes typically have left-shifted ODCs relative to their as the focus for a brief but prolific research program in
lowland sister taxa (Black and Tenny 1980; Monge and León- physiological genetics by the late Lee R. G. Snyder and his
Velarde 1991; Petschow et al. 1977). The high blood oxygen colleagues during the 1970s and 1980s.
affinity of high-altitude mammals is generally attributable to
the possession of hemoglobin with an intrinsically high
oxygen-binding affinity or a reduced responsiveness toward HEMOGLOBIN POLYMORPHISM IN DEER MICE AND
organic phosphates such as 2,3-biphosphoglycerate that ITS ROLE IN PHYSIOLOGICAL ADAPTATION TO
stabilize the low-affinity, deoxygenated conformation of HIGH-ALTITUDE HYPOXIA
hemoglobin (Brewer and Eaton 1971). Given that high-altitude species generally have hemoglobins
The family Camelidae provides an interesting case study of with higher oxygen binding affinities than those of their
hemoglobin variation in relation to altitude. The camelid fam- lowland relatives, an obvious question is whether high- and
ily comprises 6 extant species that are distributed in South low-altitude populations of wide-ranging species such as P.
America, North Africa, and Central Asia. The Asian and Afri- maniculatus are similarly characterized by divergent fine-
can forms, Camelus bactrianus and C. dromedarius, are re- tuning of hemoglobin function. Available evidence for P.
stricted to lowland deserts, whereas the South American forms, maniculatus suggests that modifications of hemoglobin func-
Lama glama, L. guanicoe, L. pacos, and V. vicugna, live at tion do indeed play a role in physiological adaptation to dif-
altitudes of 2,000–5,000 m in the Andes. The P50 values of all 6 ferent elevational zones. Biochemical studies (Snyder 1985;
camelid species are in the range of 17–22 torr (Table 1), which Snyder et al. 1982, 1988) revealed a strong, negative correla-
is low relative to other mammals of comparable size (Jürgens tion between P50 values and the native altitude of different
1989; Piccinini et al. 1990). It thus appears that camelids of the populations of P. maniculatus across western North America
high Andes were preadapted to high-altitude hypoxia in the (Fig. 4). Thus, mice that are native to alpine and subalpine
sense that they possessed a blood biochemistry that allowed environments tend to have higher blood oxygen affinities
them to colonize alpine environments without extensive (lower P50 values) than their lowland counterparts. Given that
modifications of the ancestral condition. However, the Andean the oxygen-binding affinity of hemoglobin is largely deter-
camelids have hemoglobins with even higher oxygen affinities mined by variation in the primary structure of the individual
than those of the Asian or African camels. This biochemical subunit polypeptides, this presumably adaptive altitudinal
difference appears to be related to a His!Asn amino acid variation in blood oxygen affinity may be associated with
substitution at b2 (the 2nd residue of the b-globin polypeptide), allelic variation in the a- or b-globin genes.
28 JOURNAL OF MAMMALOGY Vol. 88, No. 1

Downloaded from https://academic.oup.com/jmammal/article/88/1/24/927083 by guest on 22 December 2021

FIG. 4.—Variation in blood oxygen affinity (as measured by half-
saturation of hemoglobin [P50]) among subspecies of Peromyscus
maniculatus from different altitudes across western North America.
Based on data compiled by Snyder et al. (1988).
FIG. 5.—Variation in blood oxygen affinity, as measured by half-
saturation of hemoglobin (P50), among congenic strains of deer mice
that carry different a-globin haplotypes in identical-by-descent
Making the connection between genotype and phenotype.— condition (i.e., homologous alleles at each gene were derived from
Genetic analysis of 3 subspecies of P. maniculatus with the a single allele carried by the common ancestor of each strain). The 9
broadest altitudinal ranges (nebracensis, sonoriensis, and congenic strains were derived from wild-caught samples of 2 deer
rufinus) revealed that variation in blood oxygen affinity is mouse subspecies (Peromyscus maniculatus nebracensis from Mesa
strongly associated with allelic variation at 2 closely linked County, Colorado, and P. m. sonoriensis from Mono County,
gene duplicates that encode the a-chains of adult hemoglobin California) that have the broadest altitudinal distributions. In each
(Chappell et al. 1988; Chappell and Snyder 1984). In P. case, P50 values were measured after acclimation to a uniform altitude
maniculatus, the 2 loci encoding the a-chains of adult of 340 m. The same genotypic rank-order of P50 values was observed
hemoglobin, Hba and Hbc, are each polymorphic for 2 main for population samples of subspecies nebracensis, rufinus, and
sonoriensis, and for congenic strains of nebracensis and sonoriensis
classes of electrophoretically detectable protein alleles, Hba0,
tested at high altitude (3,800 m). Two-locus a-globin genotypes are
Hba1, Hbc0, and Hbc1 (Snyder 1978a, 1978b, 1980, 1981). abbreviated as follows: 0/0 ¼ a0c0/a0c0, 0/1 ¼ a0c0/a1c1, and 1/1 ¼
Alleles at the 2 genes are characterized by a highly nonrandom a1c1/a1c1. Based on data compiled by Chappell and Snyder (1984).
pattern of association: the a0 and c0 alleles almost always occur
together on the same haplotype, and likewise for the a1 and c1
alleles. In the parlance of population genetics, the 2 a-globin 2 important types of physiological performance: capacity for
genes are characterized by nearly complete linkage disequilib- sustained activity (aerobic capacity) and internal heat pro-
rium. Consequently, alleles at the 2 genes most commonly duction (thermogenic capacity). This measure of aerobic
occur in the following diploid combinations: a0c0/a0c0, a0c0/ metabolism showed a striking pattern of variation among mice
a1c1, and a1c1/a1c1. The 3 genotypes exhibited a highly con- with different a-globin genotypes: V_ O2max was highest for a0c0/
sistent rank-order of P50 values when tested under both high- a0c0 mice when tested at an altitude of 3,800 m, whereas V_ O2max
and low-altitude conditions: mice with the a0c0/a0c0 genotype was highest for a1c1/a1c1 mice when tested at 340 m (Chappell
exhibited the lowest P50 value (the most left-shifted ODC), et al. 1988; Chappell and Snyder 1984; Fig. 6). Effects of the a-
mice with the a1c1/a1c1 genotype exhibited the highest P50 globin genotypes on the distal physiological phenotype (V_ O2max )
value (the most right-shifted ODC), and the a0c0/a1c1 double appear to stem directly from effects on blood O2 affinity,
heterozygotes had an intermediate blood oxygen affinity because no genotypic effects were detected for other aspects of
(Fig. 5). To control for genetic background, these experiments blood biochemistry such as the CO2-Bohr effect (the regression
were based on congenic strains of mice that carried different 2- coefficient of log-P50 on pH), PCO2 (the partial pressure of
locus a-globin haplotypes in identical-by-descent condition CO2 at 50% oxygen saturation in the blood), blood buffering
(i.e., homologous alleles at each gene were derived from capacity (the regression coefficient of log-PCO2 on pH),
a single allele carried by the common ancestor of each strain— erythrocyte 2,3-biphosphoglycerate concentration, hematocrit,
Chappell and Snyder 1984; Chappell et al. 1988). or hemoglobin concentration (Chappell et al. 1988; Chappell
In addition to the effects on blood biochemistry, the and Snyder 1984).
phenotypic effects of these a-globin genes also were manifest This variation in V_ O2max likely has important fitness conse-
at the level of whole-organism physiology. In the context of quences in high-altitude deer mice. As stated by Chappell and
adaptation to high-altitude hypoxia, one especially important Snyder (1984:5487), ‘‘A mouse capable of attaining a higher
measure of physiological performance is V_ O2max , which is V_ O2max can exercise more vigorously without incurring de-
defined as the maximal rate of oxygen consumption elicited by bilitating oxygen debt and/or it can maintain body temperature
aerobic exercise or cold exposure. V_ O2max sets the upper limit on by means of aerobic thermogenesis at lower ambient tempera-
February 2007 SPECIAL FEATURE—PHYSIOLOGICAL ADAPTATION TO HIGH ALTITUDE 29

strong directional selection in high-altitude populations of

small-bodied, endothermic animals such as deer mice. In fact,
such selection has been empirically documented in a survivor-
ship study of high-altitude deer mice in the White Mountains of
eastern California (Hayes and O’Connor 1999).
Altitudinal patterns of hemoglobin polymorphism.— The
‘‘high oxygen affinity’’ a0c0/a0c0 genotype is associated with
superior physiological performance under hypoxic conditions
at 3,800 m above sea level, but is associated with poor per-
formance (relative to the a1c1/a1c1 genotype) in the oxygen-
rich environment at 340 m. In both altitudinal extremes, the

Downloaded from https://academic.oup.com/jmammal/article/88/1/24/927083 by guest on 22 December 2021

a0c0/a1c1 heterozygotes were generally intermediate with re-
spect to both P50 and V_ O2max (Chappell et al. 1988; Chappell and
Snyder 1984). This rank order of genotypic effects is probably
due to the fact that the possession of high-affinity a02b2 or
c02b2 hemoglobin isoforms facilitates pulmonary oxygen
loading in high-altitude environments with low PO2 , but hin-
ders the release of oxygen in low-altitude environments with
relatively high PO2 . On the basis of these physiological trade-
offs between oxygen transport efficiency at different altitudes,
the high-affinity a0c0/a0c0 genotype should be favored in high-
altitude populations, whereas the low-affinity a1c1/a1c1 geno-
type should be favored in low-altitude populations. In fact,
electrophoretic surveys of a-globin variation in deer mice from
western North America have revealed a pattern consistent with
this idea: the high-affinity a0c0 haplotype is present at relatively
high frequency in high-altitude environments (.2,750 m),
whereas the a1c1 haplotype is either fixed or nearly fixed in
low-altitude environments (,1,750 m; Fig. 7).
This system represents a unique case where fitness-related
variation in whole-organism physiology can be related to
a relatively simple biochemical phenotype (blood oxygen
affinity) that has a well-characterized genetic basis. The study
FIG. 6.—Variation in maximal rate of oxygen consumption (V_ O2max )
of deer mouse hemoglobins at the DNA sequence level thus
among deer mice (Peromyscus maniculatus) during treadmill exercise
and cold exposure at high altitude (3,800 m). The experiments were provides a unique opportunity to elucidate the molecular
based on population samples of subspecies nebracensis, rufinus, and underpinnings of physiological adaptation. In fact, a recent
sonoriensis, as well as congenic strains of nebracensis and sonoriensis study of sequence variation in the a-globin gene duplicates of
that carried different a-globin haplotypes in identical-by-descent P. maniculatus has identified the specific amino acid changes
condition. Modified from Chappell and Snyder (1984). that are responsible for the divergent fine-tuning of hemoglobin
function between different elevational zones (J. F. Storz, in
tures. Many behavior patterns of deer mice, including foraging, litt.) By comparing the DNA sequences of functionally distinct
courtship, territorial defense, and predator avoidance probably a-globin alleles that exhibit frequency differences between
necessitate substantial exertion . . ..’’ Variation in thermogenic high- and low-altitude populations, this study revealed that
capacity may have especially important fitness consequences in adaptive modifications of protein function are primarily
subalpine and alpine environments. Because deer mice do not attributable to the independent or joint effects of 5 amino acid
hibernate, they rely heavily on metabolic heat production to replacement mutations in the Hba gene that modulate steric
maintain a constant body temperature (Wickler 1980). Thus, hindrance to oxygen binding: 50(CD15)His/Pro, 57(E6)Gly/
severe cold exposure can elicit the maximal rate of heat pro- Ala, 60(E9)Ala/Gly, 64(E13)Asp/Gly, and 71(EF1)Gly/Ser.
duction (thermogenic capacity), which in an aerobic organism Detailed functional studies are now required to assess whether
is reflected by V_ O2max (Hayes 1989; Rosenmann et al. 1975; the additive or epistatic effects of all 5 Hba mutations are
Wickler 1980). Because V_ O2max is markedly impaired under required to produce the adaptive shift in hemoglobin–oxygen
conditions of high-altitude hypoxia (Rosenmann and Morrison affinity. Alternatively, if just 1 or 2 mutations are directly
1975; Ward et al. 1995), high-altitude endotherms face a double responsible for the shift in hemoglobin–oxygen affinity, then
bind as thermogenic capacity is compromised in an environ- the remaining mutations may have undergone altitudinal
ment where thermoregulatory demands are the most severe. It changes in allele frequency simply as a result of genetic
therefore seems likely that aerobic performance is subject to hitchhiking.
30 JOURNAL OF MAMMALOGY Vol. 88, No. 1

Downloaded from https://academic.oup.com/jmammal/article/88/1/24/927083 by guest on 22 December 2021

FIG. 7.—Geographic variation in the frequency of 2-locus a-globin haplotypes of Peromyscus maniculatus. The a0c0 haplotype is present at
relatively high frequency in samples from high-altitude localities (e.g., the Southern Rockies of Colorado and other montane regions). By contrast,
the a1c1 haplotype is present at relatively high frequency in low-altitude localities along the West Coast and in the plains. Based on data compiled
by Snyder et al. (1988).

ACKNOWLEDGMENTS M. Horvath, and R. W. Bullard, eds.). Academic Press, New

York.
I thank 2 anonymous reviewers for helpful comments and
BUNN, H. F. 1980. Regulation of hemoglobin function in mammals.
suggestions. My work on high-altitude adaptation is funded by the
American Zoologist 20:199–211.
National Science Foundation (DEB-0614342), the National Institutes
CHAPPELL, M. A., J. P. HAYES, AND L. R. G. SNYDER. 1988. Hemo-
of Health (F32 HL68487-01), the Alfred P. Sloan Foundation, as well
globin polymorphisms in deer mice (Peromyscus maniculatus):
as a Layman Award and an Interdisciplinary Research Grant from the
Nebraska Research Council. physiology of beta-globin variants and alpha-globin recombinants.
Evolution 42:681–688.
CHAPPELL, M. A., AND L. R. G. SNYDER. 1984. Biochemical and
physiological correlates of deer mouse a-chain hemoglobin poly-
LITERATURE CITED morphisms. Proceedings of the National Academy of Sciences
BANCHERO, N., AND R. F. GROVER. 1972. Effect of different levels of 81:5484–5488.
simulated altitude on O2 transport in llama and sheep. American CLEMENTI, M. E., S. G. CONDO, M. CASTAGNOLA, AND B. GIARDINA.
Journal of Physiology 222:1239–1245. 1994. Hemoglobin function under extreme life conditions. Europe-
BARTELS, H., AND R. BAUMANN. 1977. Respiratory function of an Journal of Biochemistry 223:309–317.
hemoglobin. International Review of Physiology 14:107–134. DEJOURS, P., W. F. GAREY, AND H. RAHN. 1970. Comparison of
BENCOWITZ, H. Z., P. D. WAGNER, AND J. B. WEST. 1982. Effect of ventilatory and circulatory flow rates between animals in various
change in P50 on exercise tolerance at high altitude: a theoretical physiological conditions. Respiratory Physiology 9:108–117.
study. Journal of Applied Physiology 53:1487–1495. HALL, E. R. 1981. The mammals of North America. 2nd ed. Vol. 1.
BLACK, C. P., AND S. M. TENNY. 1980. Oxygen transport during John Wiley & Sons, Inc., New York.
progressive hypoxia in high-altitude and sea-level waterfowl. HALL, F. G., D. B. DILL, AND E. S. GUZMAN BARRON. 1936.
Respiratory Physiology 39:217–239. Comparative physiology in high altitudes. Journal of Cellular and
BOUVEROT, P. 1985. Adaptation to altitude-hypoxia in vertebrates. Comparative Physiology 8:301–313.
Springer-Verlag, Berlin, Germany. HARDISON, R. 2001. Organization, evolution, and regulation of the
BREWER, G. J., AND J. W. EATON. 1971. Erythrocyte metabolism: globin genes. Pp. 95–116 in Disorders of hemoglobin: genetics,
interaction with oxygen transport. Science 171:1205–1211. pathophysiology, and clinical management (M. H. Steinberg, B. G.
BULLARD, R. W. 1972. Vertebrates at altitude. Pp. 209–225 in Forget, D. R. Higgs, and R. L. Nagel, eds.). Cambridge University
Physiological adaptations: desert and mountain (M. K. Yousef, S. Press, Cambridge, United Kingdom.
February 2007 SPECIAL FEATURE—PHYSIOLOGICAL ADAPTATION TO HIGH ALTITUDE 31

HAYES, J. P. 1989. Altitudinal and seasonal effects on aerobic metab- ROSENMANN, M., P. MORRISON, AND D. FEIST. 1975. Seasonal changes
olism of deer mice. Journal of Comparative Physiology, B. Bio- in the metabolic capacity of red-backed voles. Physiological
chemical, Systemic, and Environmental Physiology 159:453–459. Zoology 48:303–310.
HAYES, J. P., AND C. S. O’CONNOR. 1999. Natural selection on thermo- SCHMIDT-NIELSEN, K. 1990. Animal physiology: adaptation and
genic capacity of high-altitude deer mice. Evolution 53:1280–1287. environment. 4th ed. Cambridge University Press, Cambridge,
HOCHACHKA, P. W., AND G. N. SOMERO. 2002. Biochemical adaptation: United Kingdom.
mechanism and process in physiological evolution. Oxford SHAPPELL, S. D., AND C. J. M. LENFANT. 1975. Physiological role of
University Press, New York. the oxyhemoglobin dissociation curve. Pp. 841–871 in The red blood
HOCK, R. J. 1964. Physiological responses of deer mice to various cell (D. M. Surgenor, ed.). 2nd ed. Academic Press, New York.
native altitudes. Pp. 59–72 in The physiological effects of high SNYDER, L. R. G. 1978a. Genetics of hemoglobin in the deer mouse,
altitude (W. H. Weihe, ed.). Macmillan, New York. Peromyscus maniculatus. I. Multiple a- and b-globin structural loci.
JÜRGENS, K. D. 1989. Strategien der Anpassung des Sauerstofftrans- Genetics 89:511–530.

Downloaded from https://academic.oup.com/jmammal/article/88/1/24/927083 by guest on 22 December 2021

portsystems von Säugetieren an das Leben in großen Höhen. SNYDER, L. R. G. 1978b. Genetics of hemoglobin in the deer mouse,
Peromyscus maniculatus. II. Multiple alleles at regulatory loci.
Naturwissenschaften 76:410–415.
Genetics 89:531–550.
LENFANT, C. 1973. High altitude adaptation in mammals. American
SNYDER, L. R. G. 1980. Closely linked alpha-chain hemoglobin loci in
Zoologist 13:447–456.
Peromyscus maniculatus and other animals: speculations on the
LUFT, U. C. 1972. Principles of adaptations to altitude. Pp. 143–156
evolution of duplicate loci. Evolution 34:1077–1098.
in Physiological adaptations: desert and mountain (M. K. Yousef,
SNYDER, L. R. G. 1981. Deer mouse hemoglobins: is there genetic
S. M. Horvath, and R. W. Bullard, eds.). Academic Press, New York.
adaptation to high altitude? BioScience 31:299–304.
MONGE, C., AND F. LEÓN-VELARDE. 1991. Physiological adaptation to
SNYDER, L. R. G. 1985. Low P50 in deer mice native to high altitude.
high altitude: oxygen transport in mammals and birds. Physiological Journal of Applied Physiology 58:193–199.
Reviews 71:1135–1172. SNYDER, L. R. G., S. BORN, AND A. J. LECHNER. 1982. Blood oxygen
PERUTZ, M. F. 1983. Species adaptation in a protein molecule. affinity in high- and low-altitude populations of the deer mouse.
Molecular Biology and Evolution 1:1–28. Respiration Physiology 48:89–105.
PERUTZ, M. F. 2001. Molecular anatomy and physiology of SNYDER, L. R. G., M. A. CHAPPELL, AND J. P. HAYES. 1988. a-Chain
hemoglobin. Pp. 174–196 in Disorders of hemoglobin: genetics, hemoglobin polymorphisms are correlated with altitude in the deer
pathophysiology, and clinical management (Steinberg, M. H., B. G. mouse, Peromyscus maniculatus. Evolution 42:681–688.
Forget, D. R. Higgs, and R. L. Nagel, eds.). Cambridge University TUREK, Z., F. KREUZER, AND L. J. C. HOOFD. 1973. Advantage or
Press, Cambridge, United Kingdom. disadvantage of a decrease of blood oxygen affinity for tissue
PETSCHOW, D., I. WURDINGER, R. BAUMANN, J. DUHM, G. BRAUNITZER, oxygen supply at hypoxia. A theoretical study comparing man and
AND C. BAUER. 1977. Causes of high blood O2 affinity of animals rat. Pflügers Archiv 342:185–187.
living at high altitude. Journal of Applied Physiology 42:139–143. WARD, M., P. J. S. MILLEDGE, AND J. B. WEST. 1995. High altitude
PICCININI, M., T. KLEINSCHMIDT, K. D. JÜRGENS, AND G. BRAUNITZER. physiology and medicine. Chapman & Hall Medical, London,
1990. Primary structure and oxygen-binding properties of the United Kingdom.
hemoglobin from guanaco (Lama guanacoe, Tylopoda). Biological WEBER, R. E., AND A. FAGO. 2004. Functional adaptation and
Chemistry Hoppe-Seyler 371:641–648. its molecular basis in vertebrate hemoglobins, neuroglobins and
POYART, C., H. WAJCMAN, AND J. KISTER. 1992. Molecular adaptation cytoglobins. Respiratory Physiology and Neurobiology 144:141–159.
of hemoglobin function in mammals. Frontiers in Respiratory WICKLER, S. J. 1980. Maximal thermogenic capacity and body
Physiology 90:3–17. temperatures of white-footed mice (Peromyscus) in summer and
ROSENMANN, M., AND P. MORRISON. 1975. Metabolic response of winter. Physiological Zoology 53:338–346.
highland and lowland rodents to simulated high altitudes and cold.
Comparative Biochemistry and Physiology, A. Comparative
Physiology 51:523–530. Special Feature Editor was Edward J. Heske.