Respiratory physiology — gas exchange and transport

Intuition [Beginner]

Every cell in your body needs oxygen and must dispose of carbon dioxide. The respiratory system solves this problem by bringing air and blood into close contact across the enormous surface area of the lungs.

Air travels from the nose and mouth down the trachea, through branching bronchi and bronchioles, and into tiny sacs called alveoli. An adult lung contains roughly 300 million alveoli, providing a gas-exchange surface area about the size of a tennis court (70 square metres).

Blood arriving at the lungs via the pulmonary artery carries oxygen-poor, carbon dioxide-rich blood from the tissues. Across the thin alveolar walls (just one cell thick), oxygen diffuses from the air into the blood while carbon dioxide diffuses in the opposite direction. The now oxygen-rich blood returns to the heart via the pulmonary veins and is pumped to the systemic circulation.

The volume of air moved in and out per minute is called minute ventilation. At rest, a typical adult breathes 12 times per minute with a tidal volume of 500 mL, giving a minute ventilation of 6 L/min. Not all of this air reaches the alveoli — about 150 mL per breath fills the conducting airways (anatomical dead space) and does not participate in gas exchange.

Visual [Beginner]

The oxygen-haemoglobin dissociation curve shows how readily haemoglobin binds oxygen at different partial pressures. The horizontal axis is the partial pressure of oxygen ($P_{O_2}$) in the blood. The vertical axis is the percentage of haemoglobin binding sites occupied by oxygen (saturation).

The curve has a characteristic sigmoid (S) shape. At high $P_{O_2}$ (as in the lungs, ~100 mmHg), haemoglobin is nearly 100% saturated. At lower $P_{O_2}$ (as in active tissues, ~40 mmHg), saturation drops to about 75%, meaning oxygen is released where it is needed most.

The Bohr effect shifts the curve to the right when CO2 levels rise or pH falls. This means that active tissues (which produce CO2 and acid) cause haemoglobin to release more oxygen — a beneficial match between supply and local demand.

Worked example [Beginner]

Calculate alveolar $P_{O_2}$ at high altitude using the alveolar gas equation.

Given: barometric pressure at altitude $P_B=500$ mmHg, $P_{AC{O}_2}=36$ mmHg, respiratory quotient $R=0.85$.

Step 1. Subtract water vapour pressure (47 mmHg at body temperature):

$$P_{I{O}_2}=(P_B-47)\times F_{I{O}_2}=(500-47)\times 0.21=453\times 0.21\approx 95.1\text{ mmHg}.$$

Step 2. Apply the simplified alveolar gas equation:

$$P_{A{O}_2}=P_{I{O}_2}-\frac{P_{AC{O}_2}}{R}=95.1-\frac{36}{0.85}=95.1-42.4\approx 52.7\text{ mmHg}.$$

At sea level ($P_B=760$ mmHg), the same calculation gives $P_{A{O}_2}\approx 100$ mmHg. At altitude, the reduced barometric pressure cuts alveolar $P_{O_2}$ nearly in half, which is why acclimatisation (increased ventilation, increased red blood cell production) is essential.

Check your understanding [Beginner]

Formal definition [Intermediate+]

Dalton's law states that in a gas mixture, the total pressure equals the sum of the partial pressures of each component gas. For atmospheric air:

$$P_B=P_{N_2}+P_{O_2}+P_{H_{2}O}+P_{C{O}_2}+\dots$$

At sea level, $P_B=760$ mmHg and the fractional concentration of oxygen in dry air is $F_{I{O}_2}=0.2093$.

Fick's law of diffusion governs gas transfer across the alveolar membrane:

$${\dot{V}}_{gas}=\frac{D\cdot A\cdot (P_1-P_2)}{T},$$

where ${\dot{V}}_{gas}$ is the rate of gas diffusion, $D$ is the diffusion coefficient (which depends on gas solubility and molecular weight), $A$ is the surface area, $T$ is the membrane thickness, and $(P_1-P_2)$ is the partial pressure difference across the membrane. The lung maximises $A/T$: enormous surface area (70 m${}^2$) with minimal thickness (0.3 $\mu$m).

The alveolar gas equation predicts alveolar $P_{O_2}$:

$$P_{A{O}_2}=F_{I{O}_2}(P_B-P_{H_{2}O})-\frac{P_{AC{O}_2}}{R},$$

where $P_{H_{2}O}=47$ mmHg at 37 degrees C, $P_{AC{O}_2}$ equals arterial $P_{C{O}_2}$ (because CO2 diffuses rapidly across the alveolar membrane), and $R$ is the respiratory quotient (${\dot{V}}_{C{O}_2}/{\dot{V}}_{O_2}$), typically 0.8-0.85 on a mixed diet.

CO2 transport

Carbon dioxide is transported in three forms: dissolved CO2 (7-10%), carbaminohaemoglobin bound to haemoglobin (10-20%), and bicarbonate ($HC{O}_3^{-}$, 70-80%). The bicarbonate pathway involves the enzyme carbonic anhydrase inside red blood cells:

$$C{O}_2+H_{2}O\rightleftharpoons H_{2}C{O}_3\rightleftharpoons H^{+}+HC{O}_3^{-}.$$

Bicarbonate exits the red blood cell in exchange for chloride (the chloride shift or Hamburger phenomenon), maintaining electroneutrality. In the lungs, the reaction reverses and CO2 is exhaled.

Ventilation-perfusion matching

Efficient gas exchange requires that each alveolus receive both fresh air (ventilation, $\dot{V}$) and blood (perfusion, $\dot{Q}$). The ventilation-perfusion ratio $\dot{V}/\dot{Q}$ quantifies this matching. In an ideal lung, $\dot{V}/\dot{Q}\approx 0.8$ at the alveolar level (slightly less ventilation than perfusion because of the physiological shunt).

Gravity causes a gradient: in upright posture, both ventilation and perfusion increase from apex to base, but perfusion increases more. The apex has high $\dot{V}/\dot{Q}$ (wasted ventilation); the base has low $\dot{V}/\dot{Q}$ (wasted perfusion). Regional hypoxia causes local vasoconstriction (hypoxic pulmonary vasoconstriction), diverting blood away from poorly ventilated alveoli — a mechanism unique to the pulmonary circulation.

Key theorem with proof [Intermediate+]

Theorem (Oxygen transport capacity). The total oxygen content of arterial blood equals the sum of haemoglobin-bound oxygen and dissolved oxygen. For a normal adult with haemoglobin concentration [Hb] = 15 g/dL, arterial saturation $S_a{O}_2=97.5\%$, and arterial $P_{O_2}=100$ mmHg, the total arterial oxygen content is approximately 19.8 mL O2 per dL blood.

Proof. Each gram of haemoglobin binds 1.34 mL O2 at full saturation (the Hufner constant). The haemoglobin-bound oxygen is:

$$[O_2{]}_{Hb}=[\text{Hb}]\times 1.34\times S_a{O}_2=15\times 1.34\times 0.975=19.6\text{ mL/dL}.$$

Dissolved oxygen follows Henry's law: $[O_2{]}_{dissolved}=0.003\times P_{O_2}=0.003\times 100=0.3\text{ mL/dL}$.

Total: $19.6+0.3=19.9{\text{ mL O}}_2\text{/dL}$. $\square$

This calculation shows why haemoglobin is essential. Without it, dissolved oxygen alone (0.3 mL/dL) would be far too little to meet metabolic demands. Haemoglobin increases oxygen-carrying capacity by roughly 65-fold.

The Fick equation relates oxygen consumption to cardiac output:

$${\dot{V}}_{O_2}=\text{CO}\times (C_a{O}_2-C_v{O}_2),$$

where $C_a{O}_2$ is arterial oxygen content and $C_v{O}_2$ is mixed venous oxygen content. At rest, the arteriovenous oxygen difference is about 5 mL/dL, giving ${\dot{V}}_{O_2}=5\text{ L/min}\times 5\text{ mL/dL}=250\text{ mL/min}$, matching measured resting oxygen consumption.

Exercises [Intermediate+]

Pulmonary mechanics — compliance, surface tension, airway resistance, and the work of breathing [Master]

Pulmonary gas exchange begins not at the alveolar membrane but one step upstream, in the mechanical apparatus that moves air into and out of the lung. The thoracic cage acts as a bellows; the lung tissue acts as a passive elastic structure that follows the bellows; the airways act as a branching resistive network that throttles airflow; and the surfactant film coating the alveolar lining acts as a variable surface-tension agent that prevents the small alveoli from collapsing into the larger ones. Each of these four elements is governed by a clean physical law, and the integrated mechanical behaviour of the lung is the sum of their actions.

Compliance. The static pressure-volume relationship of the lung is captured by the lung compliance $C_L=\Delta V/\Delta P$, defined as the change in lung volume per unit change in transpulmonary pressure. A healthy adult lung at functional residual capacity has $C_L\approx 0.2L/cm{\mathrm{H}}_2\mathrm{O}$, meaning the lung accepts about $200\mathrm{mL}$ of additional air for every $\mathrm{cm}{\mathrm{H}}_2\mathrm{O}$ of transpulmonary pressure applied. Compliance is not a constant; the pressure-volume curve is sigmoidal, with a steep middle region (the physiological operating range) and flat upper and lower regions (the extremes where the elastic limits of tissue and the resistance to alveolar opening dominate). The curve exhibits hysteresis: the inflation path lies below the deflation path, because opening collapsed alveoli requires more pressure than keeping them open. The integrated area between the inflation and deflation curves is the work dissipated as heat during one breath cycle.

Compliance has two anatomical contributions. Tissue compliance arises from the elastin and collagen fibres of the lung parenchyma, which behave like a non-linear spring with characteristic recoil pressure. Surface compliance arises from the air-liquid interface inside each alveolus, where surface tension at the curved interface generates a pressure tending to collapse the alveolus. The two contributions are roughly comparable in magnitude in a normal lung. In emphysema, destruction of alveolar walls eliminates elastic recoil and increases compliance (a more compliant, less elastic lung that empties poorly); in pulmonary fibrosis, deposition of dense scar tissue stiffens the parenchyma and decreases compliance (a stiff lung that requires high pressures to inflate).

Surface tension and the Laplace law. Treating each alveolus as a thin-walled sphere with internal radius $r$ and surface tension $\gamma$, the Laplace law gives the pressure across the alveolar wall as $P=2\gamma /r$ — the smaller the sphere, the greater the pressure for the same surface tension. This poses a critical stability problem. If two alveoli of different radii are connected by an airway and both have the same surface tension, the smaller one (higher internal pressure) will empty into the larger one (lower internal pressure), collapsing entirely. A pure water-air interface has $\gamma \approx 70dyne/cm$, far too high to permit stable alveolar populations of mixed size.

The lung solves this problem with pulmonary surfactant, a phospholipid-protein mixture (~ 90% phospholipid, predominantly dipalmitoylphosphatidylcholine [DPPC], with surfactant proteins SP-A, SP-B, SP-C, SP-D) secreted by type II pneumocytes onto the alveolar lining. Surfactant has the critical property that its surface tension is area-dependent: when the alveolus is small, the surfactant molecules are densely packed and surface tension drops dramatically (to as low as $5dyne/cm$ or less); when the alveolus is large, the molecules spread out and surface tension rises back toward water values. Substituting this area-dependent $\gamma (A)$ into the Laplace law makes the small-alveolus pressure equal to or less than the large-alveolus pressure, eliminating the collapse instability. The bridge to alveolar stability is exactly this — surfactant converts the unstable bare-water-interface lung into a stable population of mixed-size alveoli.

Surfactant deficiency is the molecular basis of neonatal respiratory distress syndrome (NRDS or hyaline membrane disease), in which premature infants (born before type II pneumocyte maturation around week 32 of gestation) cannot maintain surfactant production. Their alveoli progressively collapse over hours after birth, work of breathing rises, and gas exchange fails. Exogenous surfactant administration (Survanta, Curosurf, Infasurf — beractant, poractant alfa, calfactant — animal-derived preparations introduced in the late 1980s) has transformed the prognosis of these infants from one of the leading causes of neonatal mortality to a manageable condition with high survival. Adult acute respiratory distress syndrome (ARDS) involves a different mechanism (inflammatory injury to the alveolar-capillary barrier with secondary surfactant inactivation) but shares the loss-of-surfactant pathophysiology in part.

Airway resistance and the Poiseuille law. Air flowing through the conducting airways encounters resistance governed by the Poiseuille law for laminar flow through a cylindrical tube of radius $r$ and length $\ell$,

$$R=\frac{8\eta \ell }{\pi r^4},$$

where $\eta$ is the dynamic viscosity of the gas. The fourth-power dependence on radius is decisive: a $50\%$ reduction in radius produces a $16$-fold increase in resistance. This is the mechanical foundation of obstructive pulmonary disease. Asthma involves bronchoconstriction (smooth-muscle-mediated narrowing of the medium-sized airways) and airway-wall oedema, both reducing $r$ and dramatically increasing $R$. Chronic obstructive pulmonary disease (COPD) involves a combination of fixed structural narrowing (chronic bronchitis) and loss of radial traction on the small airways (emphysema), again raising $R$ but in a less reversible fashion.

Total airway resistance in a healthy adult is approximately $1-2\mathrm{cm}{\mathrm{H}}_{2}O/(L/s)$ at quiet breathing. Resistance is not distributed uniformly along the airway tree. The largest contributor is the medium-sized airways (segmental bronchi, roughly generations 4-8 of the branching tree); the small airways (terminal bronchioles, generations 16-19) contribute only a small fraction of total resistance in health because their enormous parallel number compensates for the small individual radius. This non-uniformity has a clinical consequence: small-airway disease (the early lesion of COPD) can be present without measurable change in spirometric forced expiratory volume, because the small airways' contribution to total resistance is small even when many are diseased. Specialised tests (oscillometry, single-breath nitrogen washout) probe the small-airway compartment.

Where flow becomes turbulent (high Reynolds number, typically in the upper airways during exercise or in pathological narrowings producing local jets), the Poiseuille relation fails and resistance becomes flow-dependent. Turbulent flow obeys a quadratic pressure-flow relation $\Delta P\propto {\dot{V}}^2$ rather than the laminar linear relation, and the pressure cost of moving air rises faster than expected.

Work of breathing. The mechanical work performed by the respiratory muscles during one breath cycle is

$$W=\oint PdV,$$

the integral of pressure with respect to volume around the inflation-deflation loop. In quiet breathing, $W$ is small (about $0.3-0.5\mathrm{J}$ per breath) and the respiratory muscles consume only $\sim 3\%$ of total body oxygen uptake. The work has three components: elastic work (overcoming tissue and surface recoil), resistive work (overcoming airway and tissue resistance), and inertial work (negligible at quiet breathing frequencies). At rest, elastic work dominates; in obstructive lung disease, resistive work dominates and can rise to $30\%$ of total body oxygen uptake at severe exacerbation — a self-limiting trap in which the act of breathing consumes a substantial fraction of the oxygen it delivers.

The optimal breathing frequency minimises the sum of elastic and resistive work per unit alveolar ventilation. Slow deep breaths minimise the number of dead-space-filling cycles (favouring efficiency) but maximise elastic work per breath (because the tidal volume excursion moves up the steep part of the pressure-volume curve); fast shallow breaths minimise elastic work per breath but waste a higher fraction of each breath on dead space. The body chooses an intermediate frequency (around $12$-$15$ breaths per minute at rest) that approximately minimises total work for the required alveolar ventilation — a clean optimisation problem that the medullary respiratory centres solve automatically.

Static and dynamic lung volumes. Spirometric measurement of the static lung volumes provides a clinical window into the mechanical state of the lung. Tidal volume ($V_T\approx 500\mathrm{mL}$) is the volume of one breath at rest. Inspiratory reserve volume ($\mathrm{IRV}\approx 3000\mathrm{mL}$) is the additional volume that can be inspired above tidal end-inspiration; expiratory reserve volume ($\mathrm{ERV}\approx 1000\mathrm{mL}$) is the additional volume that can be exhaled below tidal end-expiration. Residual volume ($\mathrm{RV}\approx 1200\mathrm{mL}$) is the air remaining in the lung after maximal exhalation, kept open by the chest-wall recoil that resists complete collapse. Vital capacity ($\mathrm{VC}=V_T+\mathrm{IRV}+\mathrm{ERV}\approx 4500\mathrm{mL}$) is the total displaceable volume. Total lung capacity ($\mathrm{TLC}=\mathrm{VC}+\mathrm{RV}\approx 5700\mathrm{mL}$) is the maximal lung volume. Functional residual capacity ($\mathrm{FRC}=\mathrm{ERV}+\mathrm{RV}\approx 2200\mathrm{mL}$) is the equilibrium volume at end-tidal expiration, where the inward elastic recoil of the lung balances the outward recoil of the chest wall.

Dynamic flow measurements add the temporal dimension. Forced vital capacity ($\mathrm{FVC}$) is the total volume exhaled during a maximal forced expiration. Forced expiratory volume in one second (${\mathrm{FEV}}_1$) is the portion of FVC expelled in the first second. The ratio ${\mathrm{FEV}}_1/\mathrm{FVC}$ is a sensitive index of airway obstruction: a normal value is above $0.75$, while in obstructive lung disease (asthma, COPD) the ratio falls because expiratory airflow is throttled by the increased airway resistance. In restrictive lung disease (pulmonary fibrosis, neuromuscular weakness), both ${\mathrm{FEV}}_1$ and $\mathrm{FVC}$ fall together and the ratio is preserved or even elevated. This spirometric dichotomy between obstructive and restrictive patterns is the foundational diagnostic step in clinical pulmonology, and reflects directly the compliance and resistance mechanics described above.

Gas exchange and alveolar physiology — diffusion, the A-a gradient, and V/Q matching [Master]

The integrated mechanical apparatus of the previous section delivers fresh air to the alveolar surface; the role of this section's machinery is to move oxygen across the alveolar-capillary membrane into the blood and carbon dioxide out, then to distribute that gas-exchange function across the lung in a way that matches the regional perfusion delivered by the pulmonary circulation.

Fick's law on the alveolar-capillary membrane. The diffusive flux of a gas through a homogeneous membrane is given by Fick's first law,

$$J=-DA\frac{dC}{dx},$$

where $J$ is the molar flux (mol per unit time), $D$ is the diffusion coefficient (cm${}^2$/s), $A$ is the membrane surface area, $C$ is the gas concentration, and $x$ is the spatial coordinate normal to the membrane. For gas exchange in the lung, it is more convenient to express the driving force as a partial-pressure difference; this gives the lung-physiology form

$${\dot{V}}_{\mathrm{gas}}=\frac{D_{m}A\beta }{T}(P_A-P_c)=D_L(P_A-P_c),$$

where $D_m$ is the membrane diffusion coefficient, $\beta$ is the gas solubility (Bunsen coefficient), $T$ is the membrane thickness, $D_L$ is the lung diffusing capacity (the lumped membrane-and-blood diffusion conductance), and $P_A-P_c$ is the difference between alveolar partial pressure and mean pulmonary-capillary partial pressure. The product $A/T$ is the geometric driver of high pulmonary diffusion — a $70{\mathrm{m}}^2$ surface across a $\sim 0.3\mu \mathrm{m}$ thick membrane gives $A/T\sim 2\times 10^8\mathrm{cm}$, an enormous diffusion conductance.

The diffusion coefficient $D$ for a gas in the membrane scales as $\sqrt{1/M}\beta$, where $M$ is the molecular weight and $\beta$ the solubility. For O${}_2$ ($M=32$, $\beta \approx 0.024$) and CO${}_2$ ($M=44$, $\beta \approx 0.57$), the solubility ratio dominates and CO${}_2$ diffuses through the alveolar membrane about $20$ times faster than O${}_2$ per unit partial-pressure difference. This is the foundational reason CO${}_2$ does not develop an alveolar-arterial gradient in the way O${}_2$ does: even a low diffusion driving pressure suffices to maintain CO${}_2$ equilibration across the membrane.

Capillary transit and equilibration kinetics. A red blood cell traverses the pulmonary capillary in $\sim 0.75\mathrm{s}$ at rest (the transit time, $t_c$). During this transit, the capillary partial pressure $P_c(t)$ rises from the mixed-venous value $P_v$ toward the alveolar value $P_A$ according to the Bohr integral,

$$P_c(t)=P_A-(P_A-P_v)\exp \left(-\frac{D_{L}t}{\beta {\dot{Q}}_c}\right),$$

where ${\dot{Q}}_c$ is the volumetric flow of the capillary segment and $\beta$ the effective gas-capacitance coefficient of blood (which for O${}_2$ is dominated by the haemoglobin binding curve and is therefore highly non-linear in P_O_2, complicating the integral in practice). For O${}_2$ in a normal lung at rest, equilibration is essentially complete within $\sim 0.25\mathrm{s}$, leaving a margin of safety: the transit time can fall to a third of resting value (during heavy exercise, when cardiac output triples and capillary transit shortens correspondingly) before incomplete equilibration begins to limit gas exchange. The diffusion-limited regime, in which $P_c(t_c)<P_A$ at the end of transit, occurs only at extreme exercise, at altitude, or in disease that reduces $D_L$ (pulmonary fibrosis, emphysema, vascular bed loss). CO is the prototypical diffusion-limited gas: it binds haemoglobin so tightly that blood acts as an infinite sink, the back-pressure $P_c$ never rises measurably, and uptake is purely diffusion-limited. Clinically, this allows $D_{LCO}$ (the single-breath diffusing capacity for carbon monoxide) to be measured as a direct probe of the membrane and capillary blood compartments.

The alveolar-arterial oxygen gradient. Pure-physiology and clinical practice both report the alveolar-arterial gradient $A-a=P_{A{O}_2}-P_{a{O}_2}$, the difference between the alveolar partial pressure of oxygen (predicted from the alveolar gas equation) and the arterial partial pressure of oxygen (measured from blood). In a perfectly homogeneous, fully equilibrated lung, the A-a gradient would be zero. In a real lung breathing room air at sea level, the gradient is approximately $5-10\mathrm{mmHg}$ in healthy young adults and rises with age to $15-20\mathrm{mmHg}$ by the seventh decade. Three mechanisms widen the gradient: anatomical shunt (bronchial and Thebesian venous return mixing with arterial blood, a small physiological contribution); V/Q mismatch (regions with low V/Q ratio contributing under-oxygenated blood to the arterial circulation); and diffusion limitation (incomplete capillary equilibration, normally negligible).

Clinical interpretation of the A-a gradient distinguishes hypoxaemia caused by alveolar hypoventilation (gradient normal: the alveolar and arterial oxygen both fall together) from hypoxaemia caused by V/Q mismatch, shunt, or diffusion limitation (gradient widened: the arterial drops faster than the alveolar). This is the central diagnostic dichotomy in arterial blood gas interpretation.

V/Q matching, dead space, and shunt. The local efficiency of gas exchange at any region of lung depends on the local ventilation-perfusion ratio ${\dot{V}}_A/\dot{Q}$. In the limit ${\dot{V}}_A/\dot{Q}\to \infty$ (ventilation without perfusion: dead space), inspired air enters but no blood is available to receive the oxygen, and the alveolar partial pressures equal the inspired-gas partial pressures ($P_{A{O}_2}\approx 150\mathrm{mmHg}$, $P_{AC{O}_2}\approx 0$). In the limit ${\dot{V}}_A/\dot{Q}\to 0$ (perfusion without ventilation: shunt), blood arrives but no fresh gas is available, and the blood partial pressures equal the mixed-venous values ($P_{a{O}_2}\approx 40\mathrm{mmHg}$, $P_{aC{O}_2}\approx 46\mathrm{mmHg}$). The healthy lung sits between these extremes, with ${\dot{V}}_A/\dot{Q}\approx 0.8$ as the population-mean ratio. Regional values vary; the distribution width ${\sigma }_{\log \dot{V}/\dot{Q}}$ is a clinically useful summary measure of V/Q heterogeneity.

The multiple inert gas elimination technique (MIGET, Wagner-Saltzman-West 1974) measures the full V/Q distribution by infusing a mixture of inert gases of varying solubility, measuring their retention in arterial blood and excretion in expired gas, and inverting a fifty-compartment linear model to recover the distribution. The technique is the gold standard for V/Q distribution characterisation; it shows that healthy lungs have a tight unimodal distribution centred near ${\dot{V}}_A/\dot{Q}=1$, while diseased lungs show broadening, bimodality (with shunt and dead-space modes), or systematic shifts.

The West zones. John West showed in 1964 that the upright human lung has a gravitational gradient of perfusion: pulmonary arterial pressure rises with depth below the apex, and at the apex it may fall to a level comparable to or below the alveolar pressure. He divided the lung into three zones based on the relative magnitudes of pulmonary arterial pressure $P_a$, alveolar pressure $P_A$, and pulmonary venous pressure $P_v$. In Zone 1 (apex), $P_A>P_a>P_v$: alveolar pressure compresses the capillaries, no perfusion occurs, and the region behaves as dead space. In Zone 2 (mid-lung), $P_a>P_A>P_v$: flow is driven by the arterial-to-alveolar pressure difference (a Starling-resistor regime, analogous to a vascular waterfall). In Zone 3 (base), $P_a>P_v>P_A$: flow is conventional and driven by the arterial-to-venous pressure difference. A theoretical Zone 4 at the very base accounts for compression of vessels by lung tissue weight, partially closing the gradient. The zonal model is a clean physical mechanism for the basal-to-apical perfusion gradient observed in radionuclide imaging, and explains why apical V/Q is high (over-ventilated relative to perfusion) and basal V/Q is low (under-ventilated relative to perfusion). The Bohr-Enghoff dead-space equation, $V_D/V_T=(P_{aC{O}_2}-P_{EC{O}_2})/P_{aC{O}_2}$, exploits CO${}_2$ partial-pressure differences between arterial and mixed-expired gas to estimate the total physiological dead space (anatomical plus alveolar), giving a single-number summary that integrates over the zonal heterogeneity.

Hypoxic pulmonary vasoconstriction. Unlike the systemic circulation, where hypoxia produces vasodilatation, the pulmonary circulation responds to local alveolar hypoxia with vasoconstriction. The mechanism involves direct sensing of low $P_A{O}_2$ by pulmonary arterial smooth-muscle cells (mitochondrial reactive-oxygen-species and Kv channel inhibition pathways have been implicated), causing local vasoconstriction that diverts blood away from poorly ventilated alveoli toward better-ventilated regions. This is a feedback-matching mechanism that reduces V/Q mismatch and limits the A-a gradient in regional disease. In chronic generalised hypoxia (high altitude, chronic obstructive lung disease) it becomes maladaptive: global pulmonary vasoconstriction raises pulmonary arterial pressure, eventually producing pulmonary hypertension and right-heart strain (cor pulmonale).

CO${}_2$ transport and the chloride shift. Carbon dioxide moves from tissues to lung in three forms that together solve the high-volume, low-driving-pressure transport problem. About $7-10\%$ travels as dissolved CO${}_2$ (Henry's-law contribution); $10-20\%$ travels as carbamino-haemoglobin ($\mathrm{Hb}-\mathrm{NH}-\mathrm{CO}{\mathrm{O}}^{-}$), the carbamino adduct between CO${}_2$ and the $\alpha$-amino terminus of haemoglobin (the same chemistry that contributes to the Bohr effect); the remaining $70-80\%$ travels as bicarbonate. The bicarbonate pathway operates inside the red blood cell, where carbonic anhydrase catalyses the hydration $\mathrm{C}{\mathrm{O}}_2+{\mathrm{H}}_2\mathrm{O}\to {\mathrm{H}}_2\mathrm{C}{\mathrm{O}}_3$ and the rapid ionisation ${\mathrm{H}}_2\mathrm{C}{\mathrm{O}}_3\to {\mathrm{H}}^{+}+\mathrm{HC}{\mathrm{O}}_3^{-}$. The H${}^{+}$ is buffered by haemoglobin (which is a substantially better buffer in its deoxygenated than its oxygenated form — the molecular basis of the Haldane effect, the reciprocal partner of the Bohr effect: deoxygenation increases CO${}_2$ carrying capacity, and oxygenation decreases it, providing the lung-side enhancement of CO${}_2$ release as the haemoglobin loads oxygen). The HCO${}_3^{-}$ exits the red cell on the anion exchanger AE1 (band-3, $\mathrm{C}{\mathrm{l}}^{-}/HC{\mathrm{O}}_3^{-}$ antiporter) in exchange for Cl${}^{-}$ — the Hamburger chloride shift. In the lungs the reaction reverses, CO${}_2$ is regenerated and exhaled, and Cl${}^{-}$ shifts back out. The Henderson-Hasselbalch equation $\mathrm{pH}=\mathrm{p}{K}_a+\log ([\mathrm{HC}{\mathrm{O}}_3^{-}]/[\mathrm{C}{\mathrm{O}}_2])$ with $\mathrm{p}{K}_a=6.1$ governs the resulting acid-base balance of blood; $[\mathrm{HC}{\mathrm{O}}_3^{-}]$ is the slow renal-controlled variable, and dissolved $[\mathrm{C}{\mathrm{O}}_2]\approx 0.03\times P_{C{O}_2}mmol/L$ is the fast respiratory-controlled variable.

Oxygen transport and the haemoglobin saturation curve — Hill kinetics, the Bohr effect, and 2,3-BPG [Master]

The third pillar of pulmonary physiology is the binding chemistry of oxygen to haemoglobin in the red blood cell. Without haemoglobin, dissolved oxygen alone would deliver only about $0.3mL/dL$ to tissues at sea level — roughly $2\%$ of the metabolic demand even at rest. Haemoglobin raises the carrying capacity sixty-fold and, through its cooperative binding, ensures that the same molecule that loads quickly in the lungs also unloads where needed in tissues. This unit cross-links to 16.06.01 for the bioinorganic chemistry of the haem iron itself; the present treatment focuses on the binding equilibria and their physiological consequences.

Cooperative binding and the Hill equation. Each haemoglobin tetramer (${\alpha }_2{\beta }_2$ in adult Hb A) carries four haem groups, each capable of binding one O${}_2$. If the four sites bound independently, the saturation curve would follow the simple Henderson-Hasselbalch hyperbola $S=P/(K+P)$ (a Langmuir isotherm). Empirically, the curve is sigmoidal, indicating positive cooperativity: binding at one site enhances the affinity at the remaining sites. The standard phenomenological description is the Hill equation,

$$S=\frac{P^n}{P_{50}^n+P^n},$$

where $S$ is the fractional saturation, $P$ is the partial pressure of O${}_2$, $P_{50}$ is the half-saturation pressure ($\approx 26.6\mathrm{mmHg}$ for adult human blood at standard conditions), and $n$ is the Hill coefficient. For human Hb A, $n\approx 2.7$ — substantially less than the upper bound of $4$ (which would correspond to all four sites binding in a single concerted step), reflecting partial cooperativity in which the transition between low- and high-affinity states is gradual.

The Hill equation is a fit, not a derivation; the underlying molecular mechanism is captured by the Monod-Wyman-Changeux (MWC) two-state allosteric model. Each tetramer is in dynamic equilibrium between two conformational states: the T (tense) state, with low O${}_2$ affinity ($P_{50,T}\approx 100\mathrm{mmHg}$), and the R (relaxed) state, with high affinity ($P_{50,R}\approx 1\mathrm{mmHg}$). Each state binds O${}_2$ with site-independent affinity (no cooperativity within a state). Cooperativity arises because oxygen binding stabilises the R state — each successive ligand bound shifts more tetramers from T to R. The fraction of T molecules drops from $\sim 1$ at zero saturation to $\sim 0$ at full saturation, and the apparent affinity rises smoothly with saturation. The MWC model derives the Hill coefficient $n$ and the curve shape from three parameters: the T and R affinities and the equilibrium constant $L=[T_0]/[R_0]$ at zero saturation. Fitting these to data gives the Hill coefficient as an emergent quantity rather than an independent parameter. The alternative Koshland-Némethy-Filmer (KNF) sequential model treats each binding event as inducing a local conformational change at one subunit that propagates to its neighbours; KNF and MWC predict slightly different curve shapes for high-resolution kinetic data, and current evidence favours a hybrid in which both concerted (T-R) and sequential (subunit-subunit) transitions occur.

The Adair four-step constants. A more granular description treats each of the four binding events as an independent equilibrium,

$$\mathrm{Hb}+n{\mathrm{O}}_2\rightleftharpoons {\mathrm{Hb}({\mathrm{O}}_2)}_n,n=1,2,3,4,$$

with stoichiometric constants $K_1,K_2,K_3,K_4$. The Adair equation expresses fractional saturation as

$$S=\frac{K_{1}P+2K_1{K}_2{P}^2+3K_1{K}_2{K}_3{P}^3+4K_1{K}_2{K}_3{K}_4{P}^4}{4(1+K_{1}P+K_1{K}_2{P}^2+K_1{K}_2{K}_3{P}^3+K_1{K}_2{K}_3{K}_4{P}^4)}.$$

Experimentally, $K_4/K_1\approx 100$ — the fourth oxygen binds about $100$ times more readily than the first, the quantitative signature of cooperativity. The Adair constants are the most general phenomenological description and reduce to the Hill equation in a particular limit and to non-cooperative independent binding when $K_1=K_2=K_3=K_4$.

The Bohr effect. Christian Bohr observed in 1904 that increased CO${}_2$ partial pressure shifts the haemoglobin saturation curve rightward, decreasing O${}_2$ affinity at any given P_O_2. The molecular mechanism, worked out by Perutz in the 1970s on the basis of crystal structures, involves two protonation events at the T state. (1) Increased H${}^{+}$ concentration (lower pH, the fixed-acid Bohr effect) protonates histidine-146 of the $\beta$ chain, forming an intra-subunit salt bridge to aspartate-94 of the same chain that stabilises the T state. (2) Increased CO${}_2$ (the carbamino Bohr effect) forms carbamino adducts on the $\alpha$-amino terminus of each $\alpha$ chain ($\mathrm{R}-\mathrm{N}{\mathrm{H}}_2+\mathrm{C}{\mathrm{O}}_2\to \mathrm{R}-\mathrm{NH}-\mathrm{CO}{\mathrm{O}}^{-}+{\mathrm{H}}^{+}$), simultaneously stabilising T and contributing to the acid load. The two contributions are quantitatively comparable.

The physiological consequence is offloading-optimised binding. In peripheral tissues with high CO${}_2$ and low pH (the metabolic signature of active cells), the curve shifts rightward and haemoglobin releases additional oxygen at the local P_O_2. In the pulmonary capillaries, the opposite shift occurs as CO${}_2$ leaves the blood and pH rises, increasing the affinity and enhancing pulmonary loading. The Bohr effect thus produces an automatic match between local O${}_2$ delivery and local metabolic activity — a feedback structure analogous to the hypoxic pulmonary vasoconstriction of the previous section, but operating at the binding-chemistry level inside the red blood cell.

A quantitative diagnostic measure is the Bohr coefficient $\partial \log P_{50}/\partial \mathrm{pH}\approx -0.48$ at $37^{\circ }\mathrm{C}$: a $0.1$ unit drop in pH raises $P_{50}$ by about $11\%$, releasing an additional $\sim 5\%$ of bound oxygen at typical tissue P_O_2.

2,3-bisphosphoglycerate. The red blood cell synthesises a unique allosteric effector, 2,3-bisphosphoglycerate (2,3-BPG, formerly called 2,3-DPG), via a side-branch off the glycolytic pathway (the Rapoport-Luebering shunt). 2,3-BPG binds in the central cavity between the two $\beta$ chains of the T state but not the R state, stabilising T and shifting the saturation curve rightward. At its normal concentration in red cells ($\approx 5mmol/L$), 2,3-BPG raises $P_{50}$ from the stripped-haemoglobin value of $\sim 13\mathrm{mmHg}$ to the in-vivo value of $\sim 27\mathrm{mmHg}$ — a major component of physiological oxygen-affinity tuning.

2,3-BPG concentration adapts to chronic conditions over hours to days. Chronic hypoxia (high altitude, chronic lung disease, severe anaemia) upregulates 2,3-BPG, further rightward-shifting the curve to favour tissue offloading; stored banked blood progressively loses 2,3-BPG (depleted to $\sim 10\%$ of normal after a week of storage), increasing affinity and reducing tissue oxygen delivery from transfused units. Fetal haemoglobin (Hb F, ${\alpha }_2{\gamma }_2$ in fetus and newborn) binds 2,3-BPG poorly because the $\gamma$ subunit lacks the histidine residues that anchor 2,3-BPG in adult Hb A; this gives Hb F a leftward-shifted curve ($P_{50,\mathrm{HbF}}\approx 19\mathrm{mmHg}$ vs. $\sim 27$ for Hb A), favouring placental oxygen transfer from maternal to fetal blood. The switch from Hb F to Hb A occurs in the months around birth via the BCL11A and other transcriptional regulators.

Cooperativity as molecular biophysics, and the cross-link to 16.06.01. The cooperative binding chemistry of haemoglobin is the canonical case study in allosteric regulation. The haem iron itself — a five-coordinate Fe(II) in deoxyhaemoglobin, becoming six-coordinate with bound O${}_2$ — is treated in detail at 16.06.01 as a representative metalloenzyme active site. That unit covers the spin-state change of the iron upon oxygen binding (from high-spin S=2 in deoxy to low-spin S=0 in oxy), the proximal histidine F8 that links the iron to the protein scaffold, the distal histidine E7 that hydrogen-bonds to bound O${}_2$ and discriminates against CO binding, and the role of porphyrin-ring conformational changes (the "iron-out-of-plane" displacement of the T state pulling on the proximal histidine to communicate ligand state to the protein) in transmitting binding information from the haem to the protein interface. The present unit treats the same chemistry at the tetrameric and integrated-physiology level: how the haem-level events sum across four subunits to produce the cooperative saturation curve, how the curve is tuned by Bohr, CO${}_2$, 2,3-BPG, and temperature, and how it is exploited for matched delivery.

The structural mechanism (Perutz 1970) is the T-R quaternary transition: the two $\alpha \beta$ dimers of the tetramer rotate $\sim 15^{\circ }$ relative to one another between T and R states, breaking and reforming a set of intersubunit salt bridges at the ${\alpha }_1{\beta }_2$ interface. The rotation is concerted across the tetramer in the MWC limit, partial and sequential in the KNF limit. Crystallography of intermediate-saturation states (Brzozowski-Walford-Whitlow-Brunori) supports a more complex picture with multiple substates of partial cooperativity. The current consensus is that the T and R extremes are real and well-defined, but transit between them passes through populated intermediate quaternary states whose statistical weights depend on saturation, pH, and allosteric effector concentrations.

Control of breathing — chemoreceptor feedback and the medullary respiratory pattern generator [Master]

The integration of all the preceding physiology — the mechanical apparatus delivering ventilation, the alveolar-capillary surface transferring gas, and the haemoglobin binding chemistry distributing oxygen — depends on a control system that adjusts ventilation moment by moment to match metabolic demand. The control system is implemented in three layers: the medullary pattern generator that produces the basic respiratory rhythm; the peripheral and central chemoreceptors that report blood-gas status to the medulla; and the higher cortical inputs that allow voluntary modulation of the otherwise autonomic respiratory drive.

The medullary respiratory pattern generator. The basic rhythm of breathing originates in two coupled neuronal networks in the medulla oblongata. The pre-Bötzinger complex (preBötC), located in the ventrolateral medulla, contains an oscillator of pacemaker and follower neurons that generates the inspiratory drive; the retrotrapezoid nucleus / parafacial respiratory group (RTN/pFRG) generates a complementary expiratory drive that becomes active mainly during active expiration. Smith and Feldman, with collaborators, identified preBötC in 1991 as the site of essential respiratory rhythmogenesis by lesion studies and brainstem slice preparations that retain rhythmic activity in vitro. Specific lesions to preBötC abolish breathing in vivo; smaller lesions produce ataxic breathing.

The rhythm propagates to motor outputs through several pathways. Inspiratory motor neurons in the cervical spinal cord (C3-C5) drive the phrenic nerve to the diaphragm — the principal inspiratory muscle. Intercostal motor neurons in the thoracic spinal cord drive the external intercostals (inspiratory) and internal intercostals (expiratory at active expiration). Bulbar motor neurons drive the upper-airway dilator muscles (genioglossus, palatal, laryngeal abductors) that must open during inspiration to prevent collapse of the compliant pharyngeal airway under the negative-pressure drive — the failure of this opening underlies obstructive sleep apnoea.

The preBötC oscillation is modulated by inputs from many sources: chemoreceptor signals (described below), pulmonary stretch receptors (Hering-Breuer reflex, terminating inspiration when the lung is fully expanded), upper-airway irritant receptors (cough, sneeze), arterial baroreceptors (cross-coupling to circulation), the cortex (voluntary control), and the limbic system (emotion-driven breathing changes). The integration of these inputs into a single coherent breathing pattern is the operational definition of "respiratory control" as a clinical entity.

Peripheral chemoreceptors: the carotid and aortic bodies. Specialised type I (glomus) cells in the carotid bodies (at the bifurcation of each common carotid artery) and the aortic bodies (along the aortic arch) sense arterial blood gases and pH. They respond primarily to a drop in arterial P_O_2 (with substantial response only below $P_a{O}_2\approx 60\mathrm{mmHg}$ — a steep curve at the low end), to a rise in arterial $P_{C{O}_2}$, and to a fall in arterial pH. The transduction mechanism is partly resolved: low P_O_2 inhibits a K${}^{+}$ background current in glomus cells through a putative oxygen-sensitive enzyme (heme oxygenase-2 and the AMPK-LKB1 pathway have both been implicated), depolarising the cell, opening voltage-gated Ca${}^{2+}$ channels, and releasing neurotransmitters (acetylcholine, ATP, dopamine) onto sensory afferents that travel via the carotid sinus nerve (a branch of the glossopharyngeal CN IX) and the aortic sinus nerve (a branch of the vagus CN X) to the nucleus of the solitary tract (NTS) in the medulla.

The carotid body's response curve to $P_a{O}_2$ is highly non-linear: below $60\mathrm{mmHg}$ the firing rate rises steeply, providing a strong ventilatory drive that compensates for hypoxia. The cooperative O${}_2$ binding chemistry of haemoglobin combined with the steep carotid-body response gives the body two co-acting low-$P_a{O}_2$ defences. Critically, the carotid body is the only receptor in the body whose primary input is low $P_a{O}_2$ rather than $P_{C{O}_2}$; this makes it indispensable when the central chemoreceptors are blunted by chronic CO${}_2$ retention (severe COPD) — patients in that state become dependent on hypoxic drive, and supplemental oxygen administered too aggressively can suppress ventilation. The clinical lesson is to titrate supplemental oxygen carefully in chronic CO${}_2$-retaining patients to avoid carbon dioxide narcosis.

Central chemoreceptors and the CO${}_2$-driven loop. The dominant minute-to-minute ventilatory drive in healthy individuals comes not from peripheral $O_2$ sensing but from central $\mathrm{C}{\mathrm{O}}_2$ sensing. Central chemoreceptors lie on the ventral surface of the medulla (at the retrotrapezoid nucleus and adjacent regions) and sense the partial pressure of CO${}_2$ in the brain interstitial fluid indirectly via the resulting H${}^{+}$ concentration. Because CO${}_2$ crosses the blood-brain barrier freely whereas charged species (H${}^{+}$, HCO${}_3^{-}$) cross slowly, a rise in arterial $P_{C{O}_2}$ produces a corresponding rise in brain CO${}_2$ and a fall in brain interstitial pH, exciting the central chemoreceptors and increasing ventilatory drive. The relationship between $P_{a}C{O}_2$ and ventilation is steep: at constant P_O_2, a rise of $1\mathrm{mmHg}$ in $P_{a}C{O}_2$ produces a $2-3L/min$ rise in ventilation in healthy subjects. The slope of this ventilatory response to CO${}_2$ curve is the clinical measure of central chemosensitivity, depressed by opioids, sleep, and brainstem lesions.

The loop closes: a small rise in $P_{a}C{O}_2$ → increased ventilation → CO${}_2$ wash-out → return of $P_{a}C{O}_2$ toward setpoint. Under healthy conditions the loop maintains $P_{a}C{O}_2$ within $\pm 2\mathrm{mmHg}$ around its setpoint of $\sim 40\mathrm{mmHg}$ across a fourfold range of metabolic CO${}_2$ production. In chronic hypercapnia (severe COPD, obesity hypoventilation), the central chemoreceptors adapt: brain interstitial bicarbonate rises to buffer the chronic CO${}_2$, restoring brain pH toward normal at the elevated $P_{C{O}_2}$ setpoint, and the central drive falls. This adaptive desensitisation is what shifts the patient's ventilatory drive onto hypoxic peripheral sensing and accounts for the carbon-dioxide-narcosis risk just described.

Voluntary versus autonomic control. Higher cortical centres (motor cortex, supplementary motor area) project to the medullary respiratory neurons and to the spinal motor pools, providing voluntary modulation that overrides the autonomic rhythm. Voluntary control is essential for speech, swallowing, breath-holding, and the act of inhaling test substances, and is dramatic in scope: a healthy subject can voluntarily hyperventilate to $P_{a}C{O}_2\approx 15\mathrm{mmHg}$ or breath-hold to $P_{a}C{O}_2\approx 60\mathrm{mmHg}$. But the autonomic drive eventually wins — at the limits of breath-hold (the "break point"), the urge to breathe becomes overwhelming and respiration resumes regardless of voluntary effort. The break point is not at any extreme $P_a$ value but at a more complex integrated state in which carotid-body afferents and central chemoreceptors together fire above a threshold that the cortex cannot suppress.

The two-system architecture (autonomic + voluntary) gives the respiratory system its distinctive position among autonomic functions: it is the only "vegetative" function subject to direct conscious override at all times. This duality has both clinical and philosophical aspects — Ondine's curse, the central-hypoventilation syndrome arising from focal medullary lesions or from PHOX2B mutations, is the loss of automatic breathing with preservation of voluntary breathing, producing patients who breathe normally while awake but apnoeic when asleep.

Sleep-disordered breathing. Sleep alters the entire respiratory control system. The cortical voluntary input drops out; the central chemoreceptors' setpoint rises (sleeping $P_{a}C{O}_2$ runs $\sim 3-5\mathrm{mmHg}$ higher than waking); muscle tone in upper-airway dilators falls; ventilatory responsiveness to hypoxia and hypercapnia is reduced. Three patterns of sleep-disordered breathing result. Obstructive sleep apnoea is the failure of upper-airway dilator activation, allowing the negative-pressure inspiratory drive to collapse the pharynx; the patient continues to make respiratory effort but moves no air, until arousal restores muscle tone. Central sleep apnoea is the failure of medullary drive itself, with no respiratory effort during the apnoeic episode; common in heart failure (the prolonged circulation time delays chemoreceptor feedback and produces the periodic-breathing pattern of Cheyne-Stokes respiration) and in opioid use. Mixed apnoea combines features of both. The clinical syndromes of obstructive sleep apnoea (loud snoring, fragmented sleep, daytime hypersomnolence, cardiovascular sequelae including hypertension and atrial fibrillation) and Cheyne-Stokes (the crescendo-decrescendo waxing and waning of tidal volume during sleep in heart failure) directly reflect the control-system architecture described above.

Synthesis. The four-level architecture of pulmonary physiology — mechanical bellows delivering ventilation, alveolar-capillary diffusion transferring gas, haemoglobin chemistry distributing oxygen, and chemoreceptor feedback adjusting drive — is the foundational reason the respiratory system can match minute-to-minute metabolic demand across the full physiological range from resting metabolic rate to maximal aerobic exercise. The central insight is that each level operates on a clean physical or chemical law (Laplace for surface tension, Poiseuille for laminar flow, Fick for diffusion, the Hill / MWC equations for cooperative binding, mass-action for the carbonic-anhydrase buffer reaction) and the level-to-level connections are matched by feedback mechanisms that the system has tuned across vertebrate evolution. Putting these together, the respiratory system identifies a metabolic rate (set by tissue oxygen consumption and CO${}_2$ production) with a ventilation rate (set by the chemoreceptor-medullary loop), and identifies the alveolar gas composition with the arterial blood gas composition via the V/Q distribution and the haemoglobin binding curve. The bridge is from chemistry to physics to physiology, mediated at each step by a quantitative law that holds across the operating range. This pattern generalises to other gas-transport systems — the cardiovascular oxygen delivery loop, the cellular respiration ATP-coupling loop, the muscle-contraction calcium loop — and the pattern recurs whenever physiology demands fast, robust, automatic matching of supply to demand.

Full proof set [Master]

Proposition 1 (Bohr integral for capillary equilibration). Let $P_c(t)$ be the partial pressure of a gas in pulmonary capillary blood as a function of transit time $t$, with $P_c(0)=P_v$ (mixed-venous value) at the start of the capillary and the alveolar partial pressure $P_A$ taken as constant along the capillary. Assume Fick's law $dN/dt=D_L(P_A-P_c)$ for the molar uptake rate (where $D_L$ is the lung diffusing capacity for the gas and $N$ is moles of gas in the capillary segment), and assume that gas content in blood is related to partial pressure by $N=\beta {\dot{Q}}_{c}t{P}_c$ with effective capacitance $\beta$ (linear approximation, valid for inert gases and the dissolved fraction of O${}_2$; for the haemoglobin-bound fraction, $\beta$ is locally linear over a small partial-pressure range). Then

$$P_c(t)=P_A-(P_A-P_v)\exp \left(-\frac{D_{L}t}{\beta {\dot{Q}}_c}\right).$$

Proof. In a capillary segment of length $\ell$, residence time $t$, and volumetric flow ${\dot{Q}}_c$, the partial pressure $P_c(t)$ evolves under Fick uptake. Differentiating the gas-content relation $N(t)=\beta {\dot{Q}}_{c}t{P}_c(t)$ at fixed flow (and noting the simplification that for a steady-state co-current segment the relevant kinetics describes time along the streamline rather than time at a fixed Eulerian point), the relevant kinetic equation is

$$\beta {\dot{Q}}_c\frac{d{P}_c}{dt}=D_L(P_A-P_c).$$

This is a first-order linear ODE in $P_c$ with constant coefficients (under the constant-$P_A$ assumption). Define $\tau =\beta {\dot{Q}}_c/D_L$ as the equilibration time constant. The ODE becomes $\tau d{P}_c/dt=P_A-P_c$, with solution $P_c(t)=P_A+(P_c(0)-P_A)\exp (-t/\tau )=P_A-(P_A-P_v)\exp (-t/\tau )$. $\square$

Proposition 2 (Hill equation from MWC two-state allostery in the symmetric limit). Consider a haemoglobin tetramer in equilibrium between a T (low-affinity, dissociation constant $K_T$ at each site) and an R (high-affinity, $K_R$) state, with intrinsic equilibrium $L=[T_0]/[R_0]$ in the absence of ligand. Assume site-independence within each state (MWC concerted-transition model). Then the fractional saturation is

$$S(P)=\frac{(P/K_R)(1+P/K_R{)}^3+Lc(P/K_R)(1+cP/K_R{)}^3}{(1+P/K_R{)}^4+L(1+cP/K_R{)}^4},$$

where $c=K_R/K_T$. In the limit of large allosteric ratio ($L\gg 1$, $c\ll 1$) and at intermediate saturations, $S$ approximates the Hill equation $S\approx P^n/(P_{50}^n+P^n)$ with apparent $n\approx 1+3(1-c)/(1+c)\in (1,4]$, the Hill coefficient.

Proof. In MWC with site-independence per state, the binding polynomial of one state is $(1+P/K{)}^4$ for a tetramer. Summing over T and R states weighted by their equilibrium populations gives the binding-polynomial sum in the denominator above; the numerator follows by differentiating with respect to $\log P$. At low saturation, only the T state contributes appreciably (high $L$), giving an effective Langmuir curve with the T affinity. At high saturation, only the R state contributes (binding shifts the equilibrium toward R), giving the R affinity. The transition between the two regimes is what produces the apparent cooperativity. In the limit $L\to \infty$, $c\to 0$, the apparent Hill coefficient saturates at $n=4$ (Hill's original concerted-binding limit); for finite $L$ and $c$ the apparent $n$ is between $1$ and $4$. Differentiating $\log [S/(1-S)]$ with respect to $\log P$ at the inflection point and evaluating in the MWC limit yields the explicit formula stated. Numerical fits to adult human Hb A data give $L\approx 10^5$ to $10^6$, $c\approx 0.01$ to $0.05$, and apparent $n\approx 2.7$. $\square$

The hybrid MWC-KNF treatment (Pauling 1935; Koshland-Némethy-Filmer 1966; Imai 1979) supports a mixed-mechanism description in which the dominant route is concerted with a measurable sequential contribution; for present purposes the MWC limit captures the essential structure.

Connections [Master]

• Cardiovascular physiology 18.02.01. Determines pulmonary blood flow (perfusion) and therefore the denominator of the ventilation-perfusion ratio. Cardiac output appears directly in the Fick equation linking oxygen consumption to oxygen extraction. The Frank-Starling mechanism of cardiac output regulation is the upstream partner to the alveolar V/Q control discussed at master tier here. Pending shipped status; on shipping, will be promoted to formal prerequisite.

• Cellular respiration and oxidative phosphorylation 17.04.03 pending. Sets the metabolic demand for oxygen at the mitochondrial electron-transport chain and produces the carbon dioxide that drives respiratory feedback. The entire respiratory system exists to serve mitochondrial ATP synthesis; the oxygen consumption and CO${}_2$ production rates determined by cellular respiration appear as the inputs to the alveolar gas equation and to the Fick oxygen-extraction calculation. Pending shipped status.

• Bioinorganic metalloenzymes 16.06.01. The bioinorganic chemistry of the haem iron — the spin-state transition between high-spin Fe(II) and low-spin Fe(II)-O${}_2$, the proximal-distal histidine architecture that discriminates O${}_2$ from CO, and the porphyrin-ring iron-out-of-plane displacement that communicates ligand binding to the protein scaffold — sits at the chemistry side of the protein covered here at the physiology side. The cooperativity treated in the master subsection on Hill kinetics arises from the haem-by-haem mechanical coupling described in 16.06.01.

• Nervous system 18.05.01 pending. Provides the motor pathways from the medullary respiratory pattern generator to the phrenic and intercostal motor neurons, and the afferent pathways from the carotid and aortic bodies (CN IX, CN X) to the nucleus of the solitary tract. The control-of-breathing subsection at master tier is fundamentally a neuroscience topic dressed in physiological language.

• Renal physiology 18.08.01 pending. Compensates for chronic respiratory acid-base disturbances via the bicarbonate-reabsorption mechanism in the proximal tubule and the H${}^{+}$ secretion mechanism in the distal tubule. Chronic respiratory acidosis (CO${}_2$ retention) triggers renal bicarbonate retention that partially corrects pH over days; chronic respiratory alkalosis triggers the opposite. The renal compensation is the slow partner to the fast chemoreceptor compensation discussed here.

• Endocrine regulation 18.07.01. Thyroid hormone modulates basal metabolic rate and therefore CO${}_2$ production; progesterone (rising in pregnancy) stimulates the central chemoreceptors and explains the physiological hyperventilation of pregnancy (resting $P_{a}C{O}_2\approx 30\mathrm{mmHg}$ in third trimester); erythropoietin (EPO) drives red-cell mass and therefore haemoglobin concentration. Each of these hormonal axes feeds into the integrated oxygen-delivery system.

• Diffusion equations and reaction kinetics [02.13.NN, pending]. The mathematical-biology layer underneath the gas-exchange physiology. The Bohr integral derived in Proposition 1 is a clean application of first-order linear ODE theory; the MWC and Hill equations are exercises in equilibrium chemistry on a small state space. The V/Q distribution and the MIGET inversion are exercises in measure theory and ill-posed inverse problems. These mathematical foundations live in math §02 and physics §11.

Historical & philosophical context [Master]

The quantitative understanding of pulmonary gas exchange developed in roughly four phases over a century and a half. The first phase, in the mid-nineteenth century, established that gases obey simple physical laws (Dalton's law of partial pressures, Henry's law of solubility, Fick's law of diffusion in 1855 ^{[Fick 1855]}) and that blood is oxygenated in the lungs by the diffusion of dissolved O${}_2$ across the alveolar membrane. Adolf Fick's 1855 paper formulated diffusion as the steady-state limit of mass conservation and a constitutive law relating flux to concentration gradient, providing the equation that governs gas exchange to this day.

The second phase, around the turn of the twentieth century, established the chemistry of haemoglobin-oxygen binding. Bohr, Hasselbalch, and Krogh in 1904 ^{[Bohr 1904]} published the seminal paper showing that increased CO${}_2$ decreased haemoglobin's O${}_2$ affinity — the Bohr effect. Christian Bohr (father of the physicist Niels Bohr) was Professor of Physiology at Copenhagen; his collaboration with the future Nobel-laureate August Krogh produced the modern quantitative picture of oxygen transport. Hill's 1910 paper formulating the eponymous cooperative-binding equation ^{[Hill 1910]} — Archibald Vivian Hill's first major contribution, predating his Nobel-winning muscle work — introduced the framework that still bears his name. Adair's 1925 work ^{[Adair 1925]} generalised the binding analysis to the four-step stoichiometric description that remains the most general phenomenological treatment.

The third phase, in the mid-twentieth century, established the mechanical, V/Q-distribution, and acid-base aspects of the integrated respiratory physiology. J. S. Haldane (father of J. B. S. Haldane and Naomi Mitchison, an extraordinary scientific family) developed gas-analysis methods and characterised CO${}_2$-driven ventilatory drive in the 1900s-20s. Fenn, Rahn, and Otis at Rochester ^{[Rahn-Fenn 1955]} formulated the alveolar gas equation and the O${}_2$-CO${}_2$ diagram in the 1940s-50s, codifying the relationship between inspired gas, alveolar ventilation, and arterial gas tensions. West and Wagner, beginning in the 1960s and especially with the multiple-inert-gas elimination paper of 1974 ^{[Wagner-Saltzman-West 1974]}, introduced the distribution-based view of V/Q matching that transformed respiratory physiology from compartment-based to distribution-based science.

The fourth phase, since the 1970s, has elucidated molecular and cellular mechanisms behind the integrative physiology. Perutz's structural work on haemoglobin (1959, 1970) ^{[Perutz 1970]} resolved the T-R quaternary transition and the structural basis of cooperativity, allostery, and the Bohr effect. The MWC (Monod-Wyman-Changeux 1965) ^{[MWC 1965]} and KNF (Koshland-Némethy-Filmer 1966) ^{[KNF 1966]} models of allostery were articulated in parallel, with subsequent work establishing their relative roles in different systems. Smith, Feldman, and Ellenberger identified preBötC as the medullary rhythmogenic site in 1991 ^{[Smith-Feldman 1991]}. Genetic dissection of central-chemoreceptor function and respiratory-control mutations (PHOX2B in central congenital hypoventilation syndrome; ATOH1 and other transcription factors in respiratory development) continues into the current decade.

The respiratory system illustrates the biological principle of structure-function correspondence: the enormous surface area and extreme thinness of the alveolar membrane are structural adaptations that maximise Fick diffusion, and the cooperative binding curve of haemoglobin is a chemical adaptation that combines fast loading at high alveolar P_O_2 with substantial unloading at modest tissue P_O_2. The Bohr effect, hypoxic pulmonary vasoconstriction, and chemoreceptor feedback are nested feedback loops that produce automatic matching of supply to demand at three different timescales — molecular (Bohr), local (vasoconstriction), and whole-body (chemoreceptor-medullary). The hierarchical integration across these timescales is the defining systems-physiology pattern of the respiratory system.

Bibliography [Master]

• Adair, G. S., "The hemoglobin system. VI. The oxygen dissociation curve of hemoglobin", J. Biol. Chem. 63 (1925), 529-545.

• Bohr, C., "Uber die spezifische Tatigkeit der Lungen bei der respiratorischen Gasaufnahme", Skand. Arch. Physiol. 16 (1904), 402-412.

• Bohr, C., Hasselbalch, K. & Krogh, A., "Ueber einen in biologischer Beziehung wichtigen Einfluss, den die Kohlensaurespannung des Blutes auf dessen Sauerstoffbindung ubt", Skand. Arch. Physiol. 16 (1904), 402-412.

• Fenn, W. O. & Rahn, H. (eds.), Handbook of Physiology, Section 3: Respiration (American Physiological Society, 1964).

• Fick, A., "Ueber Diffusion", Ann. Phys. Chem. 94 (1855), 59-86.

• Haldane, J. S., Respiration (Yale University Press, 1922).

• Hill, A. V., "The possible effects of the aggregation of the molecules of haemoglobin on its dissociation curves", J. Physiol. 40 (1910), iv-vii.

• Koshland, D. E., Némethy, G. & Filmer, D., "Comparison of experimental binding data and theoretical models in proteins containing subunits", Biochemistry 5 (1966), 365-385.

• Monod, J., Wyman, J. & Changeux, J.-P., "On the nature of allosteric transitions: a plausible model", J. Mol. Biol. 12 (1965), 88-118.

• Perutz, M. F., "Stereochemistry of cooperative effects in haemoglobin", Nature 228 (1970), 726-739.

• Rahn, H. & Fenn, W. O., A Graphical Analysis of the Respiratory Gas Exchange (American Physiological Society, 1955).

• Smith, J. C., Ellenberger, H. H., Ballanyi, K., Richter, D. W. & Feldman, J. L., "Pre-Botzinger complex: a brainstem region that may generate respiratory rhythm in mammals", Science 254 (1991), 726-729.

• Wagner, P. D., Saltzman, H. A. & West, J. B., "Measurement of continuous distributions of ventilation-perfusion ratios", J. Appl. Physiol. 36 (1974), 588-599.

• West, J. B., Respiratory Physiology: The Essentials, 10th ed. (Lippincott Williams & Wilkins, 2015).

• West, J. B. & Wagner, P. D., "Ventilation-perfusion relationships", in Comprehensive Physiology (2011).

• Boron, W. F. & Boulpaep, E. L., Medical Physiology, 3rd ed. (Elsevier, 2017), Ch. 26-32.