How We Found the Missing 70% of Glucose That Disappears at High Altitude

The experiment should have had a clean answer. Mice breathing low-oxygen air clear glucose from their bloodstream dramatically faster than mice in normal conditions. Standard imaging - PET/CT scanning - should be able to show which organs are absorbing the extra glucose. It is how researchers map metabolism in living animals.

But when Yolanda Marti-Mateos and Isha Jain ran those scans, the math did not work. After accounting for every major organ - muscle, liver, brain, adipose tissue - 70% of the increased glucose clearance remained unexplained. Something was consuming large amounts of sugar and was invisible to their imaging system.

This is the story of how that mystery was solved, as described by Marti-Mateos and Jain in the context of research published in Cell Metabolism.

The suspect hiding in plain sight

Red blood cells became a suspect for a simple reason: PET imaging cannot track them. They are constantly moving. Whatever glucose they absorb does not show up as a localized signal the way muscle or liver uptake would.

Red blood cells seemed unlikely candidates for a meaningful metabolic role. They have no nucleus, no mitochondria. They cannot do oxidative metabolism. They are primarily hemoglobin-filled sacks whose job is carrying oxygen. But they are also the most abundant cell type in the body, they depend entirely on glucose for energy since they cannot use other fuel sources, and in chronic hypoxia their numbers increase substantially.

To test whether RBCs were the missing piece, the team used approaches that have fallen somewhat out of fashion in modern molecular biology: phlebotomy and transfusion. They repeatedly bled hypoxic mice to keep their red blood cell counts at baseline levels. Blood glucose normalized. Then they transfused extra red blood cells into normal mice breathing room air. Blood glucose dropped. The evidence was unambiguous: more red blood cells meant lower blood sugar.

How hypoxia reprograms the bone marrow

Individual red blood cells from hypoxic mice were taking up more glucose per cell than those from normal animals. The team measured the abundance of GLUT1 - a glucose transporter protein - at the cell surface and found it significantly elevated in hypoxic mice's RBCs.

That created a puzzle. Mature red blood cells cannot make new proteins - they have no nucleus to transcribe genes from. So how were they producing more GLUT1?

The answer was in the bone marrow. Red blood cells are produced continuously from stem cells, maturing over about two weeks before entering circulation, where they live for approximately three months. The team labeled all existing red blood cells with biotin for three days, then moved mice into hypoxia. After four weeks, they separated old (biotin-labeled) and new RBCs and measured their GLUT1 levels. Only the newly produced cells showed elevated GLUT1. The old cells were unchanged. Hypoxia was reprogramming developing cells in the bone marrow to arrive in circulation carrying more glucose transporters - and they maintained that programming for their entire three-month lifespan.

What the cells do with the glucose

Hypoxic red blood cells metabolize glucose faster, converting it rapidly to a molecule called 2,3-diphosphoglycerate (2,3-DPG). This molecule binds hemoglobin and causes it to release oxygen more readily to tissues - which is exactly what's needed when oxygen is scarce at altitude. The glucose isn't just being burned for energy. It is being converted into a signaling molecule that improves oxygen delivery.

The molecular switch controlling this is elegant. Under normal oxygen conditions, key glycolytic enzymes are physically sequestered at the red blood cell membrane, held in place by binding to a membrane protein called Band 3. This keeps glucose metabolism slow. When oxygen levels drop and hemoglobin releases its oxygen, the deoxygenated hemoglobin changes shape and competes with the glycolytic enzymes for the Band 3 binding site, freeing those enzymes to run at full speed.

The team confirmed this mechanism using crosslinking proteomics, STED microscopy, and proximity ligation assays in both mouse and human red blood cells - finding the same basic mechanism in both species.

The therapeutic implication

If red blood cells act as a glucose sink in low-oxygen conditions, could that mechanism be harnessed therapeutically? The team tested three approaches in mouse diabetes models: exposing diabetic mice to actual hypoxia, transfusing extra red blood cells, and administering HypoxyStat - a small molecule developed in Jain's lab that increases hemoglobin's oxygen affinity, creating functional tissue hypoxia without changing the air the animal breathes.

All three reversed hyperglycemia. HypoxyStat performed as well as or better than existing diabetes drugs in these mouse models. It works through a completely different mechanism from current treatments, which primarily target insulin production or sensitivity in conventional metabolic tissues.

The research moved from hypothesis to submitted manuscript in under a year - unusually fast for work of this scope. Marti-Mateos and Jain attribute that pace partly to embracing older experimental techniques like phlebotomy and transfusion that gave direct answers, and partly to recognizing early when specialized expertise was needed and finding the right collaborators in D'Alessandro and Doctor.

As with the companion research paper from the same group, important caveats apply: this is mouse work, HypoxyStat has not been tested for diabetes in humans, and the long-term safety of pharmacologically induced tissue hypoxia requires careful clinical evaluation before any therapeutic application could be considered.