How Does Carbon Monoxide Bind to Hemoglobin?

Carbon monoxide binds to hemoglobin at exactly the same spot where oxygen normally attaches: the iron atom sitting at the center of each heme group. The problem is that hemoglobin’s affinity for CO is roughly 250 times greater than its affinity for oxygen, so even small amounts of the gas can occupy binding sites and refuse to let go. This lopsided competition is what makes CO poisoning so dangerous, but the molecular details of how it happens reveal a surprisingly rich story about protein shape, cellular energy, and even the body’s own deliberate use of CO as a signaling molecule.

Where CO Meets Iron

Each hemoglobin molecule carries four heme groups, and each heme contains a single iron atom held in place by a ring-shaped structure called a porphyrin. In normal breathing, an oxygen molecule slides into an open coordination site on that iron and forms a reversible bond. CO targets the very same site. When CO arrives, it coordinates with the iron in an almost identical fashion, forming a bond that is far more stable. The result is carboxyhemoglobin (COHb), a hemoglobin molecule that is effectively locked out of oxygen transport for as long as the CO remains attached.

The geometry of the heme pocket matters. A key amino acid called the distal histidine sits near the binding site and influences how easily different gas molecules can attach. Researchers tested this by engineering hemoglobin mutants in which the distal histidine was replaced with a smaller amino acid, glycine. Removing that bulky side chain caused significant changes in oxygen and CO affinity in myoglobin and in hemoglobin’s alpha subunits, but had surprisingly little effect on the beta subunits. This means the two types of chains in hemoglobin do not regulate gas binding in the same way, and the architecture of the pocket around the iron shapes CO’s ability to latch on.

A systematic review of CO poisoning diagnostics puts the affinity difference at up to 250-fold, which aligns with the classic textbook figure. That number is not a fixed constant across all conditions, since temperature, pH, and the structural state of hemoglobin all nudge it. But as a practical matter, it means that even a CO concentration of a few hundred parts per million in inhaled air can steadily convert a meaningful fraction of your hemoglobin into COHb over the course of an hour or two.

The Shape-Shift That Makes Things Worse

If CO simply sat on one of hemoglobin’s four heme sites and did nothing else, the damage would be limited to losing a quarter of your oxygen-carrying capacity. But hemoglobin is a cooperative molecule. When one subunit binds a ligand, it nudges the other three subunits into a higher-affinity shape (the R state), making them grip their own ligands more tightly. This is great when the ligand is oxygen, because it helps hemoglobin load up efficiently in the lungs and release efficiently in tissues. With CO in the mix, though, the same cooperativity turns toxic.

When CO occupies one or two heme sites, the remaining sites still bind oxygen, but they hold onto it much more tightly than they normally would. The oxygen-hemoglobin dissociation curve shifts to the left. In plain terms, hemoglobin picks up oxygen fine in the lungs but becomes reluctant to release it in tissues where cells are starving for it. So CO does double damage: it directly blocks binding sites and it makes the remaining oxygen-carrying sites less useful.

The allosteric behavior of hemoglobin also differs depending on whether CO or oxygen is the bound ligand. Research comparing the two has shown that the structural transition between the “tense” (T) and “relaxed” (R) conformations happens at different rates. At 25°C, hemoglobin with three CO molecules bound switches from R to T more slowly than hemoglobin with three oxygen molecules bound. And the reverse transition, T to R, is faster for CO-bound hemoglobin. The net effect is that CO-loaded hemoglobin spends more time in the high-affinity R state, reinforcing that reluctance to release whatever oxygen remains.

CO Does Not Stop at Hemoglobin

Hemoglobin gets most of the attention in CO poisoning, and for good reason: it is the most abundant heme protein in the body and the easiest to measure. But CO also binds to other iron-containing proteins, and some of this binding may explain why the severity of CO poisoning does not always correlate neatly with COHb levels measured in the blood.

One critical target is cytochrome c oxidase, the final enzyme in the mitochondrial electron transport chain. This enzyme is what your cells use to consume oxygen and produce energy. CO inhibits it by binding to its heme iron, effectively shutting down cellular respiration even if some oxygen is still being delivered. A study in a rat model found that CO exposure significantly decreased heart cytochrome oxidase activity and reduced the amount of a key enzyme subunit, even when there was no detectable tissue hypoxia from hemoglobin impairment alone. The authors suggested that this direct mitochondrial damage could underlie the cardiac dysfunction seen in CO poisoning.

⁴

CO also binds to myoglobin, the oxygen-storage protein in muscle tissue. Because myoglobin has an even higher affinity for CO than hemoglobin does in some conditions, muscle cells can accumulate CO and lose their oxygen reserves. The combination of hemoglobin blockade, mitochondrial poisoning, and myoglobin saturation explains why CO is so much more dangerous than a simple reduction in oxygen-carrying capacity would suggest.

⁵

Neurological Harm and Delayed Symptoms

The brain is especially vulnerable to CO for several reasons: it has high metabolic demand, limited energy reserves, and it is packed with iron-containing enzymes. Acute CO poisoning can cause confusion, loss of consciousness, seizures, and death. But a particularly insidious feature is delayed neurological injury, where patients who initially seem to recover develop cognitive problems, personality changes, or movement disorders days to weeks later.

The mechanism behind these delayed symptoms involves more than simple oxygen deprivation. When CO clears and oxygen returns, the reperfusion itself triggers oxidative stress. Reactive oxygen species damage cell membranes, especially in the lipid-rich white matter of the brain. Inflammatory cascades follow, with damage to the lining of blood vessels and further injury from the immune response. Hyperbaric oxygen therapy, which involves breathing pure oxygen at pressures above normal atmospheric levels, has been proposed as a way to counteract these processes by reducing oxidative stress and accelerating CO clearance.

⁶

Clearing CO From the Blood

CO leaves hemoglobin by the same route it arrived: competitive displacement. Flooding the lungs with oxygen tips the equilibrium back in oxygen’s favor, gradually prying CO off the iron sites. On room air (about 21% oxygen), the half-life of COHb is roughly four to five hours. Breathing 100% oxygen at normal atmospheric pressure cuts that dramatically. A study of CO-poisoned patients treated with 100% oxygen found a mean COHb half-life of about 74 minutes, though there was wide individual variation ranging from 26 to 148 minutes. Interestingly, the only factor that significantly predicted clearance speed was the partial pressure of oxygen in arterial blood. Age, sex, smoking status, severity of initial acidosis, and even the initial COHb level did not influence how fast it dropped.

⁷

Hyperbaric oxygen pushes the half-life down further, to roughly 20–30 minutes in some protocols. However, access to hyperbaric chambers is limited, treatment must begin quickly to be most effective, and the evidence for its superiority over normobaric oxygen in preventing long-term neurological damage is still debated among toxicologists.

Why Pulse Oximeters Get Fooled

Standard pulse oximeters work by shining two wavelengths of light through your finger and measuring how much is absorbed. They can distinguish between oxygenated and deoxygenated hemoglobin because those two forms absorb light differently. The problem is that carboxyhemoglobin absorbs light at a wavelength very similar to oxyhemoglobin. A standard two-wavelength oximeter cannot tell the difference, so it reads a CO-poisoned patient’s oxygen saturation as normal or near-normal, even when a large fraction of hemoglobin is bound to CO and completely unable to carry oxygen.

This is one reason CO poisoning is called “the great mimicker.” Patients may arrive in an emergency department with headache, nausea, and dizziness, and a reassuring SpO₂ reading on the monitor. Specialized CO-oximeters that use additional wavelengths of light can distinguish COHb, but they are not standard equipment everywhere. The gold-standard measurement remains a venous or arterial blood gas analyzed by a laboratory co-oximeter, which directly quantifies what percentage of hemoglobin is bound to CO.

Fetal Hemoglobin Binds CO Even More Tightly

Fetal hemoglobin (hemoglobin F) evolved to pull oxygen across the placenta from the mother’s blood, which means it naturally has a higher oxygen affinity than adult hemoglobin. That same property extends to CO. Under identical laboratory conditions, fetal hemoglobin binds both oxygen and CO with higher affinity than adult hemoglobin, while maintaining the same degree of cooperativity between its subunits.

⁸

This has practical implications for pregnant women and newborns. A fetus exposed to CO through the mother’s bloodstream will accumulate COHb to higher levels and clear it more slowly. Even a CO exposure that produces only mild symptoms in the mother can put the fetus at serious risk. Newborns, who still carry significant amounts of hemoglobin F, remain more vulnerable than older children or adults for the same biochemical reason.

Your Body Produces CO on Purpose

CO is not only an external poison. Your body makes it constantly as a normal byproduct of breaking down heme. When red blood cells reach the end of their lifespan (about 120 days), macrophages in the spleen and liver recycle their hemoglobin. The enzyme heme oxygenase splits the heme ring open, producing biliverdin (which becomes bilirubin, the yellow pigment in bruises and bile), free iron, and one molecule of CO.

⁹

This endogenous CO is not wasted. It functions as a signaling molecule in the brain and other tissues, influencing neurotransmitter release, blood vessel tone, and inflammatory responses. Two forms of heme oxygenase handle the work: a constitutive version that runs at a steady baseline, and an inducible version that ramps up during stress. The CO produced by these enzymes has been shown to affect inflammation and cell survival pathways, which is why researchers have explored inhaling low doses of CO as a therapeutic intervention for conditions ranging from organ transplant rejection to pulmonary hypertension.

¹⁰

In a healthy nonsmoker, endogenous CO production maintains COHb levels of roughly 0.5–1.5%. This background level is harmless and appears to be biologically useful.

Chronic Exposure and Smokers

Cigarette smoke contains significant amounts of CO, and regular smokers walk around with chronically elevated COHb levels. A study of workers in London found that smokers had mean COHb levels of 5–8%, compared with 1–3% in nonsmokers. Heavier smoking generally correlated with higher levels, though there was considerable individual variation. Sedentary workers also tended to have higher COHb than manual laborers, likely because physical activity increases breathing rate and helps clear CO faster.

¹¹

Chronic heavy smokers can push their COHb levels as high as 10%, a concentration that would prompt a trip to the emergency department if detected in a nonsmoker. Case reports describe polycythemia (an overproduction of red blood cells) in heavy smokers as the body tries to compensate for the oxygen-carrying capacity lost to CO.

Elephant Seals and Natural CO Tolerance

Humans are not the only animals that live with elevated COHb. Northern elephant seals routinely carry COHb levels that would alarm a human physician. Adult elephant seals averaged about 8.7% COHb, with a maximum recorded value of 10.4%. Even juveniles and pups showed levels around 7–8%.

¹²

These levels held steady over hours of sampling, suggesting they are not caused by brief exposure events but by a consistently high rate of heme-protein turnover. Elephant seals have exceptionally large blood volumes and high hemoglobin concentrations, so the sheer mass of heme being recycled produces a lot of CO. Researchers calculated that this baseline COHb reduces the seals’ total body oxygen stores by about 7%, which is a real cost for an animal that makes extreme breath-hold dives. The prevailing theory is that the trade-off may be worthwhile because CO, at these levels, could protect tissues from ischemia-reperfusion injury: the oxidative damage that occurs when blood flow returns to oxygen-starved tissues after a dive. These COHb values are comparable to the highest endogenous levels recorded in critically ill human patients with hemolytic anemia, and to doses used in early human clinical trials of inhaled CO therapy, which were tolerated without adverse effects.

¹²

Experimental Antidotes That Outcompete Hemoglobin

Current treatment for CO poisoning is essentially just oxygen, delivered as fast as possible. Researchers have been working on something more targeted: engineered proteins that can snatch CO directly off hemoglobin by binding it even more tightly than hemoglobin does.

One approach uses a modified version of neuroglobin, a heme protein naturally found in the brain. By mutating a key amino acid and stabilizing the protein’s surface, researchers created a molecule (Ngb-H64Q-CCC) that binds CO about 500 times more strongly than hemoglobin. When incubated with CO-saturated hemoglobin in a test tube, it reduced the half-life of COHb from over 200 minutes (exposed only to air) to as little as 0.11 minutes for cell-free hemoglobin and 0.41 minutes for hemoglobin inside intact red blood cells. In CO-poisoned mice, intravenous infusion of this engineered neuroglobin restored heart rate and blood pressure, improved survival, and the CO-loaded neuroglobin was cleared through the kidneys.

¹³

Another research group tested modified hemoglobin molecules themselves as scavengers. Two chemically altered hemoglobins, called StHb and NEMHb, improved survival from 0% to 100% in a lethal CO poisoning mouse model, outperforming a myoglobin-based approach that achieved 60% survival. These modified hemoglobins were well tolerated in 48-hour toxicity assessments.

¹⁴

A third strategy takes a different angle altogether: instead of adding a competing protein, small-molecule drugs could change hemoglobin’s own shape to make it release CO faster. Researchers have designed synthetic compounds that act as allosteric effectors, shifting hemoglobin toward a conformation that weakens its grip on CO. Early versions had the side effect of destroying red blood cells, but newer analogs have been developed that enhance CO release from carboxyhemoglobin without causing hemolysis.

¹⁵

Sickle Cell Hemoglobin and CO

Sickle cell disease is caused by a single amino acid substitution in the beta chain of hemoglobin, which causes deoxygenated hemoglobin molecules to stack into rigid fibers (polymers) that distort red blood cells into their characteristic sickle shape. It was long assumed that CO, like oxygen, could not bind to hemoglobin while it was locked in these polymers, since the binding sites are buried. But experimental evidence tells a different story. When researchers studied the melting of sickle hemoglobin polymers in the presence of CO, they found that standard models in which CO binds only after the polymer dissolves could not explain the data. CO appears to bind directly to the polymer form during the melting process itself.

¹⁶

This finding is more than a biochemical curiosity. If CO can access the heme iron even inside a sickle polymer, it could in principle shift the equilibrium away from polymerization, since CO-bound hemoglobin adopts the R-state conformation that resists stacking. Some researchers have investigated whether very low doses of inhaled CO could reduce sickling episodes, though this remains speculative and the therapeutic window between benefit and toxicity is razor-thin. The biology of CO binding, it turns out, keeps revealing wrinkles that the simple story of “gas meets iron” does not capture.