ALTITUDE ACCLIMATIZATION, HEMOGLOBIN-OXYGEN AFFINITY, AND CIRCULATORY OXYGEN TRANSPORT IN HYPOXIA

Abstract

In mammals and other air-breathing vertebrates that live at high altitude, adjustments in convective O₂ transport via changes in blood hemoglobin (Hb) content and/or Hb-O₂ affinity can potentially mitigate the effects of arterial hypoxemia. However, there are conflicting views about the optimal values of such traits in hypoxia, partly due to the intriguing observation that hypoxia-induced acclimatization responses in humans and other predominantly lowland mammals are frequently not aligned in the same direction as evolved phenotypic changes in high-altitude natives. Here we review relevant theoretical and empirical results and we highlight experimental studies of rodents and humans that provide insights into the combination of hematological changes that help attenuate the decline in aerobic performance in hypoxia. For a given severity of hypoxia, experimental results suggest that optimal values for hematological traits are conditional on the states of other interrelated phenotypes that govern different steps in the O₂-transport pathway.

1. INTRODUCTION

When humans ascend to high-altitude, the initial acclimatization response to environmental hypoxia involves a consistent suite of hematological changes, including an increase in hemoglobin (Hb) concentration and a reduction in Hb-O₂ affinity (Monge and Leon-Velarde, 1991; Storz et al., 2010b; Ivy and Scott, 2015; Storz and Scott, 2019). The hypoxia-induced increase in [Hb] initially stems from a reduction in plasma volume (hemoconcentration) and – after a period of weeks – increased red blood cell production, which is sustained by renal synthesis and release of erythropoietin. The hypoxia-induced reduction in Hb-O₂ affinity stems from changes in red cell metabolism that increase the intraerythrocytic concentration of 2,3-diphosphoglycerate (DPG [also referred to as biphosphateglycerate]), an allosteric cofactor with inhibitory effects on Hb-O₂ binding. Many previous workers assumed that such responses must be physiologically beneficial in hypoxia (Aste-Salazar and Hurtado, 1944; Lenfant et al., 1968; Eaton et al., 1969; Lenfant et al., 1969; Lenfant et al., 1971; Frisancho, 1975), in keeping with the intuitive expectation that acclimatization responses to environmental stimuli are generally well-matched to the range of conditions experienced by a given species.

There are sound reasons to expect that the above-mentioned hematological changes could be physiologically beneficial under conditions of environmental hypoxia. An increased [Hb] increases arterial O₂ content (C_aO₂) for a given O₂ partial pressure (PO₂), thereby augmenting blood O₂ conductance if cardiac output remains constant. Likewise, a reduced Hb-O₂ affinity promotes O₂-unloading to metabolizing tissues, which could conceivably compensate for suboptimal O₂ delivery (Lenfant et al., 1971). Intriguingly, however, the typical acclimatization responses to hypoxia in species like humans that have predominantly lowland ancestries are frequently not aligned in the same direction as evolved phenotypic changes in species that are long-term highland natives. For example, high-altitude species that are manifestly well-adapted to chronic hypoxia do not generally have highly elevated [Hb] and instead tend to exhibit values similar to those observed in lowland relatives living at or near sea level (Storz et al. 2010b; Storz and Scott 2019). Likewise, in contrast to the hypoxia-induced reduction in Hb-O₂ affinity that is typical of humans and other lowland mammals, many high-altitude vertebrates have evolved genetically based increases in Hb-O₂ affinity in comparison with lowland relatives (Natarajan et al., 2015b; Natarajan et al., 2016; Storz, 2016; Jendroszek et al., 2018; Natarajan et al., 2018; Zhu et al., 2018; Signore et al., 2019; Storz, 2019; Signore and Storz, 2020). How can we reconcile these contrasting patterns of acclimatization and evolutionary change in lowlanders and highlanders?

There is a growing appreciation that acclimatization responses to environmental hypoxia are a fallible guide for inferring adaptive value (Storz et al., 2010b; Dempsey and Morgan, 2015). Indeed, several aspects of hypoxia adaptation in highland natives appear to involve genetically based changes that counteract or compensate nonadaptive acclimatization responses (Storz et al., 2010b; Storz and Scott, 2019, 2021). Another possibility is that the adaptive value of changes in any one trait may be conditional on physiological context and patterns of phenotypic integration, and may therefore differ in species with lowland and highland ancestries. For example, a particular hematological response to hypoxia may be beneficial in the context of a native highlander physiology that is characterized by evolved changes in numerous, interrelated traits, whereas a qualitatively distinct acclimatization response may be more beneficial in the context of a native lowland physiology.

Here we review theoretical and experimental results that suggest hypotheses about the adaptive significance of hypoxia-induced changes in hematological traits. We highlight case studies involving rodents and humans that provide insights into the combination of hematological changes that help attenuate the expected decline in aerobic performance in hypoxia.

2. The O₂-transport pathway and the challenge of environmental hypoxia

In animals that rely on aerobic metabolism to meet their energy requirements, sustained ATP synthesis requires the O₂-transport system to match O₂ supply (delivery of O₂ to tissue mitochondria) and metabolic demand (chemical utilization of O₂ by oxidative phosphorylation in the mitochondrial electron transport chain). This balance is maintained by ventilatory and cardiovascular adjustments that jointly ensure an uninterrupted supply of O₂ to the cells of metabolizing tissues. The transport of O₂ from atmospheric air to cellular mitochondria involves four serially integrated transfer steps: (i) ventilation of the lungs; (ii) diffusion of O₂ across the alveolar-capillary barrier (i.e., from inspired air into the blood); (iii) circulatory transport of O₂ in the blood; and (iv) diffusion of O₂ from capillary blood to cellular mitochondria, where it is then used as the terminal electron acceptor in aerobic respiration. Changes in PO₂ across these alternating convective and diffusive steps provide the driving force for O₂ transport and account for the overall pressure gradient between the PO₂ of inspired air and the PO₂ in the mitochondria.

At high altitude, the reduced barometric pressure diminishes the O₂ pressure head of inspired air. Thus, to ensure an adequate O₂ flux to metabolizing tissues, physiological adjustments are required to minimize the inevitable decrement in PO₂ across successive transfer steps of the O₂-transport pathway (Bouverot, 1985; Scott and Milsom, 2006; Ivy and Scott, 2015; McClelland and Scott, 2019; Storz and Scott, 2019).

2.1. Convective O₂ transport by the cardiovascular system

As a description of convective O₂ transport by the cardiovascular system, the Fick Principle states that the rate of O₂ consumption by perfused tissue is equal to the product of blood flow to the tissue and the difference in O₂ content between the arterial blood entering the tissue capillary bed and the venous blood leaving the capillary bed. Accordingly, the rate of O₂ consumption (V̇O₂) is the product of cardiac output (Q) and the difference in the O₂ contents of arterial and venous blood (C_aO₂ and C_vO₂, respectively):

$$\dot{V}{\text{O}}_2=Q\times \left({\text{C}}_{\text{a}}{\text{O}}_2-{\text{C}}_{\text{v}}{\text{O}}_2\right).$$ (1.1)

This equation can also be written in terms of PO₂ and the blood-O₂ capacitance coefficient (βbO₂), which quantifies the amount of O₂ that is unloaded for a given arterial-venous difference in PO₂:

$$\dot{V}{\text{O}}_2=Q\times \beta {\text{bO}}_2\times \left(P_{\text{a}}{\text{O}}_2-P_{\text{v}}{\text{O}}_2\right),$$ (1.2)

where βbO₂ = (C_aO₂ – C_vO₂)/(P_aO₂ – P_vO₂), the arterial-venous difference in O₂ content divided by the arterial-venous difference in PO₂.

The blood-O₂ capacitance coefficient, βbO₂, is determined by both blood Hb content and the O₂-binding properties of Hb (O₂-affinity and cooperativity). Whereas the O₂ capacitance of air or water is constant for a given temperature, the O₂ capacitance of the blood changes as a function of PO₂ due to the cooperativity of Hb-O₂ binding (ligation of a given heme in deoxyHb increases O₂-affinities of the remaining unliganded hemes and, conversely, ligand release by a given heme in oxyHb reduces O₂ affinities of the remaining liganded hemes). Consequently, Hb has a high O₂ affinity in the pulmonary capillaries where PO₂ is high, and a reduced affinity in the systemic capillaries where PO₂ is much lower.

2.2. Changes in convective O₂ transport are mediated by changes in the shape and position of the O₂-equilibrium curve

The O₂-equilibrium curve provides a way of graphically depicting the Fick principle when the vertical axis is expressed in terms of blood O₂ content. As illustrated in Fig. 1, the blood O₂ capacitance coefficient, βbO₂, is defined as the slope of the line joining the arterial and venous points on the sigmoidal curve. An increase in the slope of the line increases the arterial-venous difference in O₂ content (C_aO₂ – C_vO₂) for a given arterial-venous difference in O₂ tension (P_aO₂ – P_vO₂). Thus, increasing βbO₂ (via changes in Hb concentration and/or changes in Hb-O₂ affinity) increases the quantity of O₂ transported to the tissue for a given difference in PO₂ between the sites of O₂ loading and unloading.

Figure 1.

Schematic illustration of blood O₂ transport. (A) O₂-equilibrium curve under the physicochemical conditions prevailing in arterial blood (a, solid curve, open symbol) and venous blood (v, dashed curve, closed symbol). The curve is a plot of blood O₂ content (y-axis) versus PO₂ (x-axis), with paired values for arterial and venous blood connected by a solid line. C_aO₂ - C_vO₂ denotes the arterial–venous difference in blood O₂ content, P_aO₂ - P_vO₂ denotes the corresponding difference in PO₂, βbO₂ denotes the blood O₂ capacitance coefficient (see text for details), Q denotes cardiac output, and VO₂ denotes the rate of O₂ consumption. On the right-hand side of the graph, the area of the rectangle is proportional to total O₂ consumption, which can be enhanced by increasing Q and/or by increasing βbO_2. Increases in βbO₂ produce a corresponding increase in C_aO₂ - C_vO₂ through shifts in the shape or position of the O₂-equilibrium curve.

2.3. Hypoxia-induced changes in [Hb] and its physiological significance

When the reduced PO₂ of inspired air results in a corresponding reduction in the PO₂ of arterial blood (P_aO₂), the decrement in blood O₂ conductance can be minimized by increasing cardiac output (Q) and/or βbO₂ (Fig. 1). All else being equal, an increase in blood [Hb] augments blood O₂ conductance by increasing C_aO₂, which can contribute to enhanced aerobic performance under conditions of environmental hypoxia as well as exercise-induced tissue hypoxia (Kanstrup and Ekblom, 1984; Ekblom and Berglund, 1991; Mairbaurl, 1994; Samaja et al., 2003; Calbet et al., 2006; Saunders et al., 2013). However, an excessively high [Hb] (polycythemia) can be counterproductive because the associated increase in blood viscosity compromises cardiac output, thereby offsetting the benefit of the elevated C_aO₂ for tissue O₂ delivery. The nonlinear relationship between hematocrit and aerobic exercise performance is well-documented in mice under normoxic conditions (Schuler et al., 2010). Although similar performance curves have not been measured in hypoxia, modeling results predict that the optimal hematocrit for exercise in hypoxia is not too different from that at sea-level, but there are more complex interaction effects with red cell [DPG] and deviations from the optimal hematocrit in either direction are predicted to more strongly curtail $\dot{V}{\text{O}}_{2\max }$ (i.e., the phenotypic optimum is more sharply defined in hypoxia than in normoxia)(Winslow, 1988). Experimental data for rodents and humans indicate that the effect of changing hematocrit or blood [Hb] is highly dependent on changes in other cardiovascular traits during acclimatization (Gonzalez and Kuwahira, 2018).

Experiments on rats that involved isovolumic exchange transfusions revealed that changes in hematocrit had very different effects on exercise performance in acclimated and nonacclimated animals (Gonzalez et al., 1994). These experiments measured the maximum rate of O₂ consumption ($\dot{V}{\text{O}}_{2\max }$), an integrated measure of aerobic performance that reflects the overall flux capacity of the O₂-transport system (McClelland and Scott, 2019; Scott and Dalziel, 2021). Increasing hematocrit in nonacclimated rats from ~45% to 60% attenuated the expected decline in hypoxic $\dot{V}{\text{O}}_{2\max }$ by increasing C_aO₂ and the arterial-venous difference in O₂ content, changes which more than offset the associated decrease in cardiac output. By contrast, reducing hematocrit in acclimated rats from ~60% to 45% increased cardiac output but had no detectable effect on $\dot{V}{\text{O}}_{2\max }$. The inverse relationship between hematocrit and cardiac output is not explained by changes in blood viscosity, and may instead reflect other changes that increase systemic vascular resistance (Gonzalez et al., 1994; Gonzalez and Kuwahira, 2018). Studies in humans have also demonstrated that hypoxic $\dot{V}{\text{O}}_{2\max }$ is unaffected by acute, isovolumic reductions of [Hb] in polycythemic highlanders (Tufts et al., 1985; Winslow et al., 1985) and in altitude-acclimatized lowlanders (Sarnquist et al., 1986; Schaffartzik et al., 1993; Calbet et al., 2002; Calbet et al., 2003). It appears that the increase in C_aO₂ associated with hypoxia-induced polycythemia is generally offset by a corresponding reduction in maximal cardiac output and a reduction in the fraction of cardiac output that is directed to contracting skeletal muscle at maximal exercise (Calbet et al., 2006; Gonzalez and Kuwahira, 2018). Although studies of acclimatized highland Tibetans and lowland Han Chinese subjects at 4,200 m revealed that natural variation in [Hb] is negatively associated with $\dot{V}{\text{O}}_{2\max }$ (Simonson et al., 2015), there is no direct evidence to suggest that the observed association reflects a causative relationship. In subjects that had lower $\dot{V}{\text{O}}_{2\max }$ due to hypoxia-related deficiencies in convective O₂ transport, an elevated [Hb] may simply represent a secondary response to renal tissue hypoxia (the stimulus for erythropoietin production). Also, since Tibetan highlanders maintain elevated blood volumes in conjunction with non-elevated [Hb] in hypoxia (Stembridge et al., 2019), it seems likely that measures of total circulating Hb mass would be positively correlated with $\dot{V}{\text{O}}_{2\max }$, as is generally the case for exercise performance in normoxia (Mairbaurl, 2013). Since measurements of blood [Hb] in g dL⁻¹ are affected by plasma volume (and therefore only measure blood-O₂ carrying capacity per unit volume of cardiac output), measurements of total Hb mass provide a more relevant index of blood-O₂ carrying capacity when investigating determinants of $\dot{V}{\text{O}}_{2\max }$.

In summary, an elevated [Hb] does not generally make a significant contribution to hypoxic $\dot{V}{\text{O}}_{2\max }$ in rats or humans that are already acclimated/acclimatized to hypoxia (Calbet et al., 2006; Gonzalez and Kuwahira, 2018). At face value, this seems consistent with the general absence of polycythemia in longterm highland natives. Given that [Hb] is a hypoxia-responsive trait that reflects overall levels of tissue O₂ delivery (specifically, O₂-delivery to the kidney, which regulates the synthesis and release of erythropoietin in an O₂-dependent manner), the absence of polycythemia in highland natives may simply reflect the robustness of the O₂-transport pathway to a reduction in inspired PO₂. In other words, highland species may typically possess non-elevated [Hb] because they have evolved compensatory mechanisms to sustain adequate levels of tissue O₂ delivery in spite of low inspired PO₂, thereby dampening the hypoxic stimulus that induces the erythropoietin-mediated increase in [Hb] (Storz, 2021; Storz and Cheviron, 2021).

2.4. Optimal P₅₀ in hypoxia

Adjustments in Hb-O₂ affinity provide an alternative means of increasing βbO₂ by shifting the steepest portion of the O₂-equilibrium curve within the operational range of the in vivo PO₂’s at the sites of O₂ loading and unloading (Bouverot, 1985; Samaja et al., 2003). Although reduced Hb-O₂ affinity promotes O₂ unloading in the systemic circulation, it can jeopardize arterial O₂ saturation (S_aO₂) under conditions of environmental hypoxia. High-altitude animals therefore face the challenge of optimizing the trade-off between O₂ loading at the respiratory surfaces (which contributes to convective O₂ transport) and O₂ unloading in the systemic capillaries (which contributes to tissue O₂ extraction). The relative magnitudes of the opposing changes in O₂ loading and O₂ unloading determine the effect of shifting the O₂-equilibrium curve on $\dot{V}{\text{O}}_{2\max }$, and this effect is expected to vary as a function of inspired PO₂.

For a given degree of hypoxia, is it preferable to have a left-shifted curve, a right-shifted curve, or neither? The obvious answer to this question is that it is preferable to have a left-shifted curve in the pulmonary capillaries and a right-shifted curve in the peripheral circulation. In fact, this is what actually happens due to differences in pH, CO₂, and temperature between the lungs and capillaries of working muscle. Although operating conditions for Hb change during circulatory transit of red blood cells, intrinsic oxygenation properties of the protein remain unaltered, and such properties determine the set point for the modulation of O₂-affinity by temperature and allosteric effectors. Thus, insights into how the optimal blood P₅₀ varies as a function of inspired PO₂ is key to understanding the possible adaptive significance of evolved changes in Hb-O₂ affinity in highland taxa.

The graphical framework shown in Fig. 2 reveals how shifts in the O₂-equilibrium curve can be expected to affect tissue O₂ delivery in hypoxia. We can see that under normoxia (P_aO₂=10.7 kPa [80 torr]), a reduced Hb-O₂ affinity (right-shifted O₂-equilibrium curve) results in a larger βbO₂ than an increased Hb-O₂ affinity (left-shifted O₂-equilibrium curve)(Fig. 2A). That is, it produces a larger increase in the slope of the line connecting the arterial and venous points on the curve, thereby maximizing tissue O₂ delivery (C_aO₂ - C_vO₂) for a given difference in PO₂ between the sites of O₂ loading and unloading (P_aO₂ - P_vO₂). By contrast, under hypoxia (P_aO₂= 3.6 kPa [27 torr]), an increased Hb-O₂ affinity (left-shifted curve) produces the largest increase in βbO₂ because the arterial-venous difference in PO₂ spans a steeper portion of the curve (Fig. 2B). Fig. 2B also shows that a left-shifted curve preserves a higher P_vO₂ in hypoxia (Woodson 1988). The higher mean capillary PO₂ maintains a steeper pressure gradient for O₂ diffusion to the cells of working muscle.

Figure 2.

O₂-equilibrium curves showing the effect of changes in Hb-O₂ affinity on tissue O₂ delivery under conditions of normoxia (A) and hypoxia (B). In panel B, the PO₂ for venous blood (P_vO₂) is marked by a vertical grey line that extends to the x-axis. P₅₀, the PO₂ at which Hb is 50% saturated. Each change in Hb-O₂ affinity produces a shift in P_vO₂, but the PO₂ of arterial blood (P_aO₂) is assumed to remain constant. Note that in normoxia (A) the right-shifted curve maximizes βbO₂ and preserves a higher P_vO₂ (an overall index of tissue oxygenation). In hypoxia, by contrast, the left-shifted curve maximizes βbO₂ and preserves a higher P_vO₂ relative to the right-shifted curve.

The above considerations are consistent with theoretical and experimental results which suggest that a reduced Hb-O₂ affinity is generally beneficial for tissue O₂ delivery during exercise in normoxia or moderate hypoxia, whereas an increased Hb-O₂ affinity is beneficial in severe hypoxia (Turek et al., 1973; Turek et al., 1978; West and Wagner, 1980; Bencowitz et al., 1982; Willford et al., 1982; Samaja et al., 1986; Samaja et al., 2003; Scott and Milsom, 2006). However, other theoretical results suggest that $\dot{V}{\text{O}}_{2\max }$ is relatively insensitive to changes in blood P₅₀ over a broad range of prevailing PO₂’s (Wagner 1996a,b, 1997). Since changes in Hb-O₂ affinity might affect interactions with traits that govern different steps in the O₂-transport pathway, it is difficult to predict the effect of such changes on systematic O₂ transport in the intact organism (Wearing and Scott, 2021).

2.5. Hypoxia-induced changes in Hb-O₂ affinity: effects of red cell DPG concentration and acid-base balance

In humans and other mammals, the hypoxia-induced reduction in Hb-O₂ affinity (as measured by the ‘standard P₅₀’, pH 7.40, PCO₂ = 40 torr, 37°) is largely attributable to an increase in the red cell concentration of the allosteric effector DPG, which is a metabolite of glycolysis. The increase in red cell DPG concentration mainly stems from a hypoxia-induced increase in ventilation. The resultant respiratory alkalosis stimulates red cell glycolytic activity which, in turn, increases DPG synthesis (Rapoport et al., 1977). Since DPG preferentially binds and stabilizes deoxyHb, an increase in the relative concentration of Hb liganded with DPG reduces Hb-O₂ affinity by shifting the allosteric equilibrium in favor of the low-affinity (deoxy) T-state (Torrance et al., 1970/71; Lenfant et al., 1971; Mairbaurl et al., 1986; Mairbaurl et al., 1993; Mairbaurl, 1994). The hypoxia-induced stimulation of erythropoiesis can also reduce Hb-O₂ affinity by producing a downward shift in the mean age of circulating red blood cells, as newly produced red cells have higher [DPG] than older cells (Mairbaurl et al., 1990; Mairbaurl, 1994; Samaja et al., 2003). Hypoxia-induced respiratory alkalosis has offsetting effects on Hb-O₂ affinity in vivo: the increase in red cell [DPG] allosterically reduces Hb-O₂ affinity, but – depending on the severity of hypoxia – this may be counterbalanced by the increase in red cell pH which promotes an increased affinity via the Bohr effect (Winslow et al., 1984; Mairbaurl et al., 1990; Samaja et al., 1997; Samaja et al., 2003). In a study conducted at Rifugio Regina Margherita, located on the summit of Punta Gnifetti (4559 m) in the Pennine Alps, in vivo O₂-equilibrium curves were measured in climbers over a 5-day acclimatization period. Results revealed that the affinity-reducing effect of increased red cell [DPG] was offset by the affinity-increasing effect of respiratory alkalosis, such that in vivo blood P₅₀ in acclimatized subjects was unchanged from the pre-acclimatization value (Mairbaurl et al., 1990). During the 1981 American Medical Research Expedition to Everest, blood gases and in vivo O₂-equilibrium curves were measured in climbers over the course of several weeks at different stages of the ascent and subsequent descent. At elevations above ~6000 m, the affinity-increasing effect of respiratory alkalosis more than counteracted the DPG effect. On the summit (8848 m), extreme P_aCO₂ (1.00 kPa [=7.5 torr]) and arterial pH (>7.7) yielded an estimated in vivo blood P₅₀ of 2.59 kPa (=19.4 torr) (Winslow et al., 1984). According to Winslow (2007:124), “Without this striking increase in O₂ affinity, it is likely that Everest could not be climbed without supplemental O₂”.

The general direction of evolved (genetically based) changes in Hb-O₂ affinity in high-altitude vertebrates runs counter to the DPG-mediated acclimatization effect (which right-shifts the O₂-equilibrium curve) and is more consistent with the alkalosis-mediated effect (which left-shifts the curve)(Storz, 2016, 2019). There is a surprisingly strong, positive relationship between Hb-O₂ affinity and native elevation in birds (based on in vitro measurements of purified Hb), but the pattern is far more equivocal in mammals. In any mammalian group for which sufficient data exist, it is not difficult to find exceptions. Among carnivores, for example, Tibetan wolves (Canis lupus laniger) have evolved an increased Hb-O₂ affinity and enhanced Bohr effect relative to lowland canids (Signore et al., 2019), but snow leopards (Panthera uncia) have Hb oxygenation properties indistinguishable from those of house cats (Janecka et al., 2015). Likewise, among rodents, there are some alpine and subalpine taxa that have evolved increased Hb-O₂ affinities in comparison with lowland relatives (Storz et al., 2010a) and other taxa that do not conform to any consistent altitudinal trend (Revsbech et al., 2013). Comparative data can provide a rich source of adaptive hypotheses that can be tested with manipulative experiments, but it is important to recognize that most evolutionary trends are not completely uniform and consistent. Also, measurements of standard P₅₀’s for purified Hbs are informative about evolved changes in protein function, but it is also of interest to know to what extent such measurements reflect in vivo blood P₅₀’s under natural conditions.

3. EMPIRICAL CASE STUDIES

The most direct means of assessing whether the optimal Hb-O₂ affinity varies in relation to ambient PO₂ is to measure performance curves that relate $\dot{V}{\text{O}}_{2\max }$ to blood P₅₀ under normoxia and varying degrees of hypoxia. The case studies described below provide insights into how changes in blood P₅₀ affect physiological performance in hypoxia, and suggest hypotheses about the conditions in which changes in one direction or the other may be adaptive.

3.1. Exercise performance in rats with pharmacologically altered Hb-O₂ affinity

McCanse et al. (1999) and Henderson et al. (2000) used a combination of sodium cyanate treatments and blood transfusions in rats to assess the effect of increasing Hb-O₂ affinity on exercise performance in normoxia and varying degrees of hypoxia. Cyanate increases Hb-O₂ affinity by irreversibly carbamylating the amino termini of the α- and β-chains of the α₂β₂ Hb tetramer. In normoxia, cyanate-treated rats had lower $\dot{V}{\text{O}}_{2\max }$ than controls, which was clearly attributable to a lower tissue O₂ extraction (Fig. 3A). A comparison of the O₂-equilibrium curves in Fig. 3A indicates that the P_vO₂ of the ‘high-affinity’ cyanate-treated rats was considerably lower than that of the controls, reflecting a diminished PO₂ diffusion gradient between capillary blood and muscle cells. In severe hypoxia, by contrast, the difference in $\dot{V}{\text{O}}_{2\max }$ between the treatment and control groups all but disappeared (Fig. 3B). This is explained by the fact that the increased Hb-O₂ affinity in cyanate-treated rats helped safeguard S_aO₂, thereby minimizing the drop in C_aO₂ in spite of the dramatically reduced P_aO₂. Consequently, the increased rate of convective O₂ transport compensated for the corresponding drop in tissue O₂ extraction. At even more severe levels of hypoxia, it is possible that the increased Hb-O₂ affinity would confer a greater performance advantage if the associated increase in maximal convective O₂ delivery (the product of C_aO₂ and cardiac output) more than offsets the decrease in tissue O₂ extraction. This would be consistent with results of experiments that documented a significantly higher survival rate of cyanate-treated rats relative to controls when exposed to a barometric pressure of 29.7 kPa (=223 torr, equivalent to an altitude of >9000 m)(Eaton et al., 1974). Nonetheless, results of these experiments show that a left-shifted O₂-equilibrium curve negatively affects $\dot{V}{\text{O}}_{2\max }$ over a broad range of PO₂ values when [Hb] is experimentally held constant. The limiting factor is O₂ diffusion from capillary blood to muscle cells, as indicated by a strong, positive correlation between $\dot{V}{\text{O}}_{2\max }$ and P_vO₂ in the full set of experimental animals (McCanse et al., 1999; Henderson et al., 2000).

Figure 3.

Johansen plots showing how traits that govern circulatory O₂ delivery and tissue O₂ extraction contribute to VO₂max in Sprague-Dawley rats. Data are shown for rats tested in normoxia (A) and hypoxia (B). Left-hand panels show O₂-equilbrium curves based on data for the control group (solid black line) and the treatment group with pharmacologically increased Hb-O₂ affinity (solid grey line). In both A and B, the area of the box in the right-hand panel denotes VO₂max elicited by aerobic exercise (value inside the box in ml min⁻¹ g⁻¹). Graphs were plotted using data from experimental groups 2 and 3 in Henderson et al. (2000).

3.2. Thermogenic performance in highland and lowland deer mice with evolved differences in Hb-O₂ affinity

North American deer mice (Peromyscus maniculatus) are distributed from sea level to altitudes >4300 m and common-garden experiments have revealed that highland natives have evolved enhanced performance capacities in hypoxia in comparison with lowland relatives (Storz et al., 2019). Highland natives have higher capacities for both aerobic thermogenesis and exercise in hypoxia due to a combination of evolved and plastic changes in numerous respiratory, cardiovascular, and metabolic traits (Cheviron et al., 2012; Cheviron et al., 2014; Lui et al., 2015; Scott et al., 2015; Ivy and Scott, 2017; Lau et al., 2017; Mahalingam et al., 2017; Tate et al., 2017; Dawson et al., 2018; Ivy and Scott 2018; Nikel et al., 2018; Scott et al., 2018; Storz et al., 2019; Ivy et al., 2020; Mahalingam et al., 2020; Tate et al., 2020).

In hypoxia, the higher thermogenic $\dot{V}{\text{O}}_{2\max }$ of highlanders relative to lowlanders is largely attributable to increased cardiac output (Fig. 4), a trait that exhibits an accentuated plastic response to hypoxia exposure in highland mice (Tate et al., 2017; Tate et al., 2020). Lowland mice exhibit a higher increase in [Hb] during hypoxia acclimation in comparison with highland mice, which accounts for the higher upper asymptote of the O₂-equilbrium curve shown in Fig. 4. However, highland mice maintain a higher S_aO₂ in hypoxia, which minimizes the decline in C_aO₂ caused by the reduction in P_aO₂. Common garden experiments revealed that the highland vs. lowland difference in S_aO₂ is genetically based, and appears to stem largely from differences in Hb-O₂ affinity (Tate et al., 2017; Ivy et al., 2020; Tate et al., 2020).

Figure 4.

Johansen plot showing how traits that govern circulatory O₂ delivery and tissue O₂ extraction contribute to hypoxic VO₂max in highland deer mice (Permomyscus maniculatus) and lowland white-footed mice (P. leucopus). The left-hand panel shows O₂-equilbrium curves based on data for highland deer mice (grey) and lowland white-footed mice (black). In the right-hand panel, the area of the box denotes thermogenic VO₂max elicited by cold exposure (value inside the box in ml min⁻¹ g⁻¹). Replicated from Wearing and Scott (2021) and based on experimental data from Tate et al. (2020) and Ivy et al. (2020).

Highland deer mice have evolved an increased Hb-O₂ affinity in comparison to lowland conspecifics and closely related congeners such as P. leucopus (Storz et al., 2009; Storz et al. 2010a; Natarajan et al. 2015a; Jensen et al., 2016; Ivy et al., 2020). Protein-engineering experiments have identified and functionally characterized causative mutations in the α- and β-chains of the α₂β₂ Hb tetramer, and have provided detailed insights into the biophysical mechanisms responsible for the affinity-enhancing effects (Natarajan et al., 2013). The key question is whether the genetically based changes in Hb-O₂ affinity translate into changes in whole-animal physiological performance in hypoxia? If so, are the changes in blood P₅₀ sufficient to improve hypoxic $\dot{V}{\text{O}}_{2\max }$ on their own, or do they only confer a physiological benefit in conjunction with changes in other traits? Previous work involving reciprocal-transplant experiments involving wild-derived strains of deer mice revealed that high-altitude natives with high Hb-O₂ affinities have higher $\dot{V}{\text{O}}_{2\max }$ in hypoxia in comparison with low-altitude conspecifics (Chappell and Snyder, 1984; Chappell et al., 1988). This documented association between Hb-O₂ affinity and hypoxic $\dot{V}{\text{O}}_{2\max }$ provides physiological context for interpreting altitudinal patterns of allele frequency variation in the underlying globin genes (Snyder et al., 1988; Storz et al., 2009; Storz et al. 2010a; Storz et al., 2012; Natarajan et al., 2015a).

To identify intermediate links in the chain of causation between Hb function and whole-animal aerobic performance, Wearing et al. (Wearing et al., 2021) produced interpopulation hybrids of highland and lowland mice using an F₂ intercross breeding design. This design effectively randomizes associations between allelic α- and β-globin variants against an admixed genetic background. They then examined phenotypic effects of the Hb variants on cardiorespiratory and hematological traits (measured in vivo) and thermogenic $\dot{V}{\text{O}}_{2\max }$ in hypoxia. These experiments confirmed that the elevated Hb-O₂ affinity contributes to an enhanced S_aO₂ in hypoxia (Tate et al., 2017; Tate et al., 2020), but – in contrast to previous studies – they revealed no statistically significant association with thermogenic $\dot{V}{\text{O}}_{2\max }$ in hypoxia (Wearing et al., 2021). One possible explanation for the apparent discrepancy is that the experiments of Chappell and colleagues tested the effects of affinity-enhancing Hb variants on a ‘highland’ background, whereas Wearing et al. (2021) tested the effects of similar variants on admixed highland/lowland backgrounds. It is therefore possible that the physiological effects of affinity-enhancing Hb variants are conditional on changes in other phenotypes that are characteristic of highland mice and that were not manifest in the admixed population examined by Wearing et al. (2021). To investigate this possibility, Wearing et al. (2021) used the experimental data to parameterize an in silico model of the O₂-transport pathway and measured main effects and interaction effects of Hb-O₂ affinity and tissue O₂ diffusion capacity (D_TO₂) on $\dot{V}{\text{O}}_{2\max }$. Specifically, they used a sensitivity analysis to determine the effects of increasing D_TO₂ on $\dot{V}{\text{O}}_{2\max }$ at Hb-O₂ affinities characteristic of mice with representative highland and lowland α/β Hb genotypes. Results revealed that increasing D_TO₂ by 50% increases $\dot{V}{\text{O}}_{2\max }$, but the effect was greater in mice with the high-affinity Hb genotype (10.7%) than in mice with the low-affinity genotype (7.7%)(Fig. 5A). The effect of Hb-O₂ affinity was accentuated when D_TO₂ was increased above 37%, when PvO₂ (and thus venous O₂ saturation) fell to zero at the lower Hb-O₂ affinity (=higher P₅₀) (Fig. 5B). Results of this study suggest that increases in Hb-O₂ affinity only contribute to the enhancement of hypoxic $\dot{V}{\text{O}}_{2\max }$ if accompanied by a corresponding increase in tissue O₂ diffusion capacity. Importantly, highland deer mice have evolved changes in muscle microvasculature and mitochondrial oxidative capacity that should increase tissue O₂ diffusion capacity (Lui et al., 2015; Scott et al., 2015; Mahalingam et al., 2017; Tate et al., 2017; Nikel et al., 2018; Scott et al., 2018; Mahalingam et al., 2020; Tate et al., 2020). Thus, available data suggest that genetically based increases in both Hb-O₂ affinity and tissue O₂ diffusion capacity contributed to the enhanced aerobic capacity of high-altitude deer mice in hypoxia. The experimental and modeling results also demonstrate how the adaptive value of changes in one trait may be conditional on changes in other traits that govern different steps in the O₂-transport pathway (Wearing et al., 2021; Wearing and Scott 2021).

Figure 5.

Effects of increasing tissue O₂ diffusing capacity (D_TO₂) on hypoxic VO₂max and blood PO₂’s inferred from mathematical modeling of the O₂-transport pathway in deer mice. (A) Relative changes in hypoxic VO₂max and (B) changes in alveolar (P_AO₂), arterial (P_aO₂), and venous (PvO₂) in response to relative increases in D_TO₂. Effects were modeled using blood P₅₀’s of deer mice with representative high- and low-altitude Hb genotypes (mean values and standard errors of the mean are denoted with bold and fine lines, respectively). Reproduced from Wearing et al. (2021).

3.3. Exercise performance in humans with high-affinity Hb mutants

To test whether an elevated Hb-O₂ affinity improves aerobic performance in hypoxia, Dominelli et al. (Dominelli et al., 2020) conducted exercise trials involving human subjects with high-affinity Hb mutants and a control group with normal Hb. The high-affinity Hb (HAH) subjects possessed Hb Malmö (β97His→Gln) or Hb San Diego (β109Val→Met), with standard P₅₀’s of 2.0–2.3 kPa (15–17 torr) compared to 3.5 kPa (26 torr) for wildtype human Hb. The exercise trials involved acute exposure to an atmosphere of 15% O₂ (equivalent to an altitude of ~3000 m) and test subjects did not undergo any prior acclimation.

In both the HAH and control groups, hypoxia exposure resulted in an expected reduction in exercise performance compared to that achieved in normoxia. However, the HAH group exhibited a significantly attenuated decline in $\dot{V}{\text{O}}_{2\max }$ relative to controls (4 ±5% vs 12 ±5% decline). The attenuated decline in the HAH group was not explained by changes in ventilation, heart rate, or P_aO₂ relative to the control group. Instead, the HAH group had a significantly higher S_aO₂ in hypoxia, which translated into a greater C_aO₂. Subjects with the lowest decrements in S_aO₂ had the lowest decrement in hypoxic $\dot{V}{\text{O}}_{2\max }$, as indicated by the significant negative correlation between the change in S_aO₂ and the change in $\dot{V}{\text{O}}_{2\max }$ (Fig. 6). The HAH group also exhibited a higher [Hb] relative to controls. At sea level, polycythemia is commonly observed in subjects with high-affinity Hb mutants, as it represents a compensatory erythropoietic response to impaired tissue oxygenation (Shepherd et al., 2019). Although the elevated [Hb] of HAH subjects also contributed to the augmentation of C_aO₂ in hypoxia, there was no significant correlation between [Hb] and $\dot{V}{\text{O}}_{2\max }$ in the HAH group or in the control group. It thus appears that the superior performance of the HAH group in hypoxia is primarily attributable to the maintenance of a higher S_aO₂, which in turn prevents C_aO₂ from falling more than it would otherwise. The resultant enhancement of blood O₂ conductance presumably more than offsets the associated impairment in tissue O₂ unloading.

Figure 6.

Relationship between the change in VO_2max between exercise trials in normoxia vs. hypoxia and the corresponding change in oxyHb saturation. Data are for human subjects with normal or high-affinity Hbs. Reproduced with permission from Dominelli et al. (2020).

In contrast with the study of human HAH subjects with Hb Malmö and Hb San Diego (Dominelli et al., 2020), a similar human study that tested the effects of pharmacologically increasing Hb-O₂ affinity detected no net effect on $\dot{V}{\text{O}}_{2\max }$ (Stewart et al., 2021). One possible explanation for the discrepancy between the two studies is that the HAH subjects in the study by Dominelli et al. (2020) were born with high-affinity Hbs and they had a lifetime to develop compensatory physiological adjustments (possibly involving pulmonary gas exchange or tissue O₂ diffusion capacity). In other words, HAH subjects were internally acclimated to the possession of high-affinity Hb. By contrast, acute increases in Hb-O₂ affinity that are achieved via pharmacological manipulations, as in the study by Stewart et al. (2021), do not permit long-term adjustments in interrelated respiratory or cardiovascular traits. Consequently, it may be that the two studies tested biochemically analogous modifications of Hb function in different physiological contexts.

4. CONCLUSIONS

The case studies involving rats, deer mice, and humans provide insights into the combination of hematological changes that can contribute to the enhancement of convective O₂ transport in hypoxia, thereby minimizing the expected decrement in aerobic performance in situations where VO₂max is limited by O₂ supply. Such insights suggest hypotheses about the phenotypic changes that we might expect to observe in highland species that are presumably well-adapted to chronic hypoxia. In the case of [Hb] – a highly labile, hypoxia-responsive trait – observed patterns of altitudinal variation among populations or species may not tell us much about how the phenotypic optimum varies according to ambient PO₂, as observed [Hb] values may be broadly reflective of overall levels of tissue oxygenation and stimulus-response characteristics of the erythropoietin system. By contrast, observed patterns of altitudinal variation in Hb-O₂ affinity among populations or species are potentially informative about how optimal P₅₀ varies as a function of ambient PO₂, provided that the measured values reflect genetically based variation.

Under conditions of severe hypoxia, modeling results suggest that an increased Hb-O₂ affinity may improve aerobic performance if it helps safeguard S_aO₂, but only if it is accompanied by an increased tissue O₂ diffusion capacity so that the augmentation of C_aO₂ translates into a corresponding enhancement of tissue O₂ extraction. This insight informs expectations about patterns of phenotypic correlations in nature and predicts that high-altitude species that exhibit an increased Hb-O₂ affinity will also exhibit changes in muscle microvasculature and/or oxidative capacity that enhance tissue O₂ diffusion capacity.

Abbreviations

[Hb]

hemoglobin concentration

[O₂]_a

arterial O₂ concentration

C_aO₂

arterial O₂ content

C_vO₂

venous O₂ content

C_aO₂-C_vO₂

arterio-venous difference in blood O₂ content

P_aO₂

arterial partial pressure of O₂

P_vO₂

venous partial pressure of O₂

P_aO₂-P_vO₂

arterio-venous difference in blood partial pressure of O₂

P ₅₀

partial pressure of O₂ at which hemoglobin is half saturated

βbO₂

blood capacitance coefficient

Q

cardiac output

VO_2max

maximum rate of O₂ consumption

S_aO₂

arterial O₂ saturation

D_TO₂

tissue diffusion O₂ capacity