A broad diversity in oxygen affinity to haemoglobin

Abstract

Oxygen affinity to haemoglobin is indicated by the p50 value (pO₂ at 50% O₂Hb) and critically determines cellular oxygen availability. Although high Hb-O₂ affinity can cause tissue hypoxia under conditions of well O₂ saturated blood, individual differences in p50 are commonly not considered in clinical routine. Here, we investigated the diversity in Hb-O₂ affinity in the context of physiological relevance. Oxyhaemoglobin dissociation curves (ODCs) of 60 volunteers (18-40 years, both sexes, either endurance trained or untrained) were measured at rest and after maximum exercise (VO₂max) test. At rest, p50 values of all participants ranged over 7 mmHg. For comparison, right shift of ODC after VO₂max test, representing the maximal physiological range to release oxygen to the tissue, indicated a p50 difference of up to 10 mmHg. P50 at rest differs significantly between women and men, with women showing lower Hb-O₂ affinity that is determined by higher 2,3-BPG and BPGM levels. Regular endurance exercise did not alter baseline Hb-O₂ affinity. Thus, p50 diversity is already high at baseline level and needs to be considered under conditions of impaired tissue oxygenation. For fast prediction of Hb-O₂ affinity by blood gas analysis, only venous but not capillary blood samples can be recommended.

Figure 1

Baseline parameters. 60 volunteers of both sexes and either endurance trained or untrained. (a, b) VO₂ values revealed elevated levels in trained groups of women (a) and men (b). (c) Statistics for VO₂max values comparing either trained versus untrained groups and women versus men. (d) Haemoglobin levels by blood gas analysis of capillary probes. N = 15 per group. Two-way ANOVA with post hoc Bonferroni test served to test for significance. Adjusted p values are indicated.

Figure 2

Capillary blood gas analysis. Estimated p50_e values (a), H⁺-level (b), pCO₂ (c) and Standard bicarbonate (d) values were obtained from trained versus untrained women and men. Estimated p50_e values from capillary probes were correlated with pCO₂ (e) and pH (f), both known determinants that affect O₂-affinity to haemoglobin. N = 15 per group. Two-way ANOVA with post hoc Bonferroni test served to test for significance. Adjusted p values are indicated (a–d). In correlation analysis, the Pearson correlation coefficient R is indicated. Two tailed test with confidence interval 95% served to test for significance (e, f).

Figure 3

Venous blood gas analysis. Calculated p50_c values (a), H⁺-level (b), pCO₂ (c) and Standard bicarbonate (d) values were obtained from trained versus untrained women and men. Calculated p50_c values from venous probes were correlated with pCO₂ (e) and pH (f). N = 15 per group. Two-way ANOVA with post hoc Bonferroni test served to test for significance. Adjusted p values are indicated (a–d). In correlation analysis, the Pearson correlation coefficient R is indicated. Two tailed test with confidence interval 95% served to test for significance (e, f).

Figure 4

Oxyhaemoglobin dissociation curves (ODC) for trained versus untrained women and men. ODCs of women and men that were either trained or untrained were recorded from venous blood samples and plotted using the “specific binding with Hill slope” model. Data fit to the Hill slope algorithm as shown by R squared values. ODCs shown, depict of 136 (a), 149 (b), 173 (c) and 151 (d) data points resulting from 15 single ODCs per group. Oxygen affinity to haemoglobin is indicated by p50 as well as p75 and p25 values. (e) Statistics for p50 values indicate no alteration in oxygen affinity to haemoglobin by fitness levels; however, Hb-O₂ affinity is lower in women than men. (f) The Hill slope declines when ODC shifts to the right. The higher affinity of oxygen to haemoglobin (lower p50) in men results in a more precipitous slope. (e, f) N = 15 per group. Two-way ANOVA with post hoc Bonferroni test served to test for significance. Adjusted p values are indicated.

Figure 5

Lower Hb-O₂ affinity in women compared to men is explained by 2,3-BPG. (a) 2,3-BPG levels were normalised to haemoglobin levels, a significant difference was found between sexes, but not fitness levels. (b) Correlation of 2,3-BPG level and p50 values revealed significance for women and men as well. The Pearson correlation coefficient R is indicated. Two tailed test with confidence interval 95% served to test for significance. (c) Representative Western blot of pooled blood probes to visualise BPGM level. Values were normalised to GAPDH, a glycolytic enzyme that is well expressed in erythrocytes. (d) Statistics of BPGM protein level. A and D: N = 15 per group. Two-way ANOVA with post hoc Bonferroni test served to test for significance. Adjusted p values are indicated.

Figure 6

Oxyhaemoglobin dissociation curves (ODC) in women versus men before versus. after VO₂max test. Individual venous blood samples were used to monitor the ODCs by tonometry under isocapnic conditions at 37 °C. Measurements were performed after arterialisation of blood using carbogen gas (95% O₂, 5% CO₂) followed by deoxygenation (95% N₂, 5% CO₂) up to a blood oxygen saturation of lower than 20%. N = 30 women and 30 men before exercise; N = 15 women and 15 men after exercise.

Figure 7

Experimental p50 and effectors of Hb-O₂ affinity before and after exercise in women and men. (a) Oxygen affinity to haemoglobin is lowered during exercise to ensure that oxygen demand corresponds to cellular availability. Hb-O₂ affinity is higher in men versus women at rest; however, after exercising this difference is nullified. Note: ODCs were recorded at isocapnic conditions and 37 °C, thus, p50 values are only attributed to pH in this setting. (b) Body temperature elevates during exercise, while men reached higher levels. (c, d) In vivo, at rest, capillary blood pH (c, as indicated by H⁺-level) and lactate level (d) were similar in both sexes. After VO₂max test, men showed higher proton level (lower pH) that is explained by a stronger increase in lactate. (e, f) Capillary pCO₂ (e) and standard bicarbonate (f) before and after exercise in women and men. (a) N = 30 for women and men before exercise; N = 15 after exercise. (b–f) N = 30 per group. Two-way ANOVA with post hoc Bonferroni test served to test for significance. Adjusted p values are indicated.

Figure 8

Correlation of experimental p50 (p50exp) values with predicted p50e/c values from blood gas analysis. (a, b) Correlation of p50 values as shown in Fig. 4 with estimated p50e values derived from capillary blood samples (a) and calculated p50c values from venous blood samples (b). The Pearson correlation coefficient R is indicated. Two tailed test with confidence interval 95% served to test for significance.