This essay is an adjunct to GeneGuessr, and uses the principle "learning more than you need to makes learning easier". The point is that learning in context makes isolated facts become relevant, makes them connected. Iron metabolism is a learning arc. It organises information. It provides a context.

Iron Metabolism

Iron: Essential and Lethal


Lewis Acid: An electron acceptor


Fenton Chemistry: Fe²⁺ + H₂O₂ → Fe³⁺ + OH⁻ + OH•


Why Iron?

A Lewis acid is an electron acceptor - a chemical species hungry for electron density. Iron excels at this. With its partially filled d-orbitals, iron attracts electron-rich molecules, it polarizes bonds, and it stabilizes reaction intermediates. Harnessed within a protein active site or porphyrin ring, this makes iron a superb biological catalyst.

Iron offers redox versatility. Fe²⁺ (ferrous) and Fe³⁺ (ferric) interconvert easily at physiological potentials, allowing iron to shuttle electrons in respiration, activate oxygen for biosynthesis, and make and break bonds in countless enzymes. The two oxidation states give controllability. They prefer different coordination geometries. This gives proteins a conformational "switch" linking shape to iron's redox state. Changing shape in a protein becomes control of redox state.

Why iron over other metals? Partly abundance. Iron is the most common transition metal in Earth's crust. But more importantly, iron's redox potential sits in a sweet spot: reactive enough to do useful chemistry, not so extreme that it's uncontrollable. Copper and manganese can substitute in some roles, but neither matches iron's combination of availability, tunability, and coordination flexibility.

Reaction Specificity

The same reactivity that makes iron useful makes it dangerous. When free Fe²⁺ encounters hydrogen peroxide, a routine byproduct of aerobic metabolism, Fenton chemistry occurs:

Fe²⁺ + H₂O₂ → Fe³⁺ + OH⁻ + OH•

The hydroxyl radical (OH•) is among the most reactive species in biology. It attacks lipids, proteins, and DNA indiscriminately, with a diffusion-limited reaction rate. (OH•) is small. It diffuses fast. It destroys whatever it touches first. There's no enzyme that detoxifies it; the only defense is preventing its formation.

This is why free iron is intolerable. Every system described in this essay - transferrin, ferritin, hepcidin, the IRE/IRP network, frataxin - exists to ensure iron is never unbound . From the moment iron crosses the gut epithelium to the moment it's buried in a ferritin core or inserted into heme, it passes hand-to-hand between chaperones. The entire architecture of iron metabolism is a solution to Fenton chemistry.

The chemical scaffolding around iron is essential: it tunes iron's redox potential, selects which substrates can approach, and directs reactivity toward specific bonds. Free iron is indiscriminately promiscuous. Useful chemistry requires constraints.

Obtaining Iron

Nutritional Iron


Enterocyte : An absorptive cell lining the small intestine. Enterocytes live only 3–5 days before being shed.


Apical membrane : The surface of an epithelial cell facing the lumen (here, the gut contents).


Dietary Iron Absorption

Dietary iron arrives in two forms. Heme iron, from meat, poultry, and fish, enters enterocytes intact via a dedicated transporter, then releases its iron inside the cell after heme oxygenase cleaves the porphyrin ring. Non-heme iron, the dominant form in plants and fortified foods, takes a longer route.

Non-heme iron in food is mostly Fe³⁺. The apical membrane transporter DMT1 (divalent metal transporter 1) only accepts Fe²⁺. A ferrireductase, DCYTB, sits on the gut's brush border and reduces ferric to ferrous iron before DMT1 can pull it into the cell. This reduction step is rate-limiting and explains why vitamin C (a reducing agent) enhances iron absorption while phytates and tannins (which stabilize Fe³⁺) inhibit it.

Enhancers and Inhibitors of Iron Absorption

The requirement for reduction before transport explains why certain dietary factors enhance or inhibit iron absorption.

Vitamin C (ascorbic acid) is a reducing agent. In the acidic environment of the stomach and upper intestine, it donates electrons to Fe³⁺, converting it to Fe²⁺. This is precisely the form DMT1 accepts. Vitamin C also chelates iron in a soluble complex, preventing it from forming insoluble ferric hydroxides as pH rises further down the gut. The effect is substantial: consuming vitamin C with a meal can increase non-heme iron absorption two- to threefold. This is why iron supplements are often taken with orange juice, and why iron-fortified cereals pair well with fruit.

Phytates (inositol hexaphosphate) work in the opposite direction. Found abundantly in whole grains, legumes, nuts, and seeds, phytates carry six phosphate groups that bind metal ions avidly. They form stable, insoluble complexes with Fe³⁺ that resist reduction and cannot be absorbed. The iron passes through the gut and is lost. Populations relying heavily on unrefined grains for calories are at higher risk of iron deficiency despite adequate total iron intake. Traditional food preparation methods, such as soaking, sprouting, and fermentation, activate phytases that degrade phytate, inadvertently improving mineral bioavailability.

Tannins (polyphenolic compounds in tea, coffee, red wine, and some fruits) also chelate iron, though by a different mechanism. Their multiple hydroxyl groups coordinate Fe³⁺, forming large complexes that are poorly absorbed. A cup of tea with a meal can reduce iron absorption by 60% or more. The effect is specific to non-heme iron. Heme iron, entering the enterocyte through its own transporter with the porphyrin ring intact, largely escapes these interactions.

Calcium inhibits absorption of both heme and non-heme iron, though the mechanism remains debated. It may compete at a shared intracellular step rather than at the apical membrane.

The practical implications are straightforward. For those needing to maximize iron absorption: pair iron-rich foods with vitamin C, avoid tea or coffee with meals, and consider how grains are prepared. For those managing iron overload: the same factors can be leveraged in reverse.

Iron's Fate

Once inside the enterocyte, iron faces a decision point. It can be stored locally in ferritin, where it may be lost when the cell sloughs off in 3–5 days, or it can cross the basolateral membrane into the bloodstream. This exit depends entirely on ferroportin, the only known cellular iron exporter in the body.

This architecture reveals a fundamental constraint: the body has no regulated excretion pathway for iron. Losses are passive and fixed, sloughed gut cells, desquamated skin, trace urinary iron, menstrual blood. Roughly 1–2 mg daily. Absorption is therefore calibrated to replace only what's lost. The enterocyte, with its short lifespan and ferroportin-dependent export, functions as a gatekeeper. Iron that enters the enterocyte but doesn't get exported is lost as the gut epithelium renews.

This is the body's one opportunity to regulate iron balance: the gut epithelium. Everything downstream - transport, storage, recycling - operates as a closed loop. Loss of iron is passive, with no known active mechanism to lose excess iron.

Lactoferrin & Nutritional Immunity


Nutritional Immunity: The body deliberately creates a deficiency, to starve bacteria of iron.


Bacteria need iron as strongly as we do. It's essential for their electron transport chains, ribonucleotide reductase (DNA synthesis), and oxidative stress defenses. Evolution has turned this dependency into a weapon to combat bacteria.

Lactoferrin is an iron-binding glycoprotein secreted at every vulnerable interface: milk, tears, saliva, mucus, the neutrophil granules that flood infection sites. Its iron affinity is extraordinary. Lactoferrin binds Fe³⁺ tightly even at low pH, precisely the acidic conditions found in infected and inflamed tissue. Where transferrin would release its iron, lactoferrin holds on.

This is nutritional immunity: deliberate iron starvation as host defense. Rather than attacking pathogens directly, the body denies them a nutrient they cannot do without.

The strategy extends systemically. During infection, inflammatory cytokines (particularly IL-6) drive hepcidin release from the liver. Hepcidin degrades ferroportin, thereby trapping iron inside macrophages and enterocytes. Serum iron drops. Transferrin saturation falls. To a bacterium attempting to proliferate in the bloodstream, the host has suddenly become a desert.

This explains a clinical puzzle: the anemia of chronic disease. Patients with persistent infections or inflammatory conditions become anemic despite adequate iron stores. Iron is present but imprisoned. Macrophages are full of it. The body is choosing infection control over oxygen transport. When the infection has passed, iron can be made available again.

The Iron Economy

A profit and loss account for iron starts with considering the passive losses:

Total: roughly 1–2 mg/day in men, potentially double that in menstruating women. This is why absorption is calibrated to ~1–2 mg/day—it's just replacing obligate losses.

Timeline to deficiency

The body has a buffer: ferritin stores. A typical adult male has ~1,000 mg in reserve; women often have 300–500 mg (or less, depending on menstrual losses and diet).

If you cut off dietary iron completely:

Stage What's happening Timeline
Store depletion Ferritin falls, no symptoms yet Weeks to months
Iron-deficient erythropoiesis Transferrin saturation drops, new RBCs get shortchanged Months
Frank anemia Hemoglobin falls, symptoms appear 6 months to 2+ years

The variance is huge. A male blood donor with good stores might not notice for over a year. A menstruating teenager with marginal stores could become symptomatic in a few months. The closed economy has reserves from months to years precisely because there's no way to rapidly increase uptake if stores crash.

Transferrin & Systemic Transport

Our iron ion has crossed the gut epithelium via ferroportin. Now it faces a problem: Fe²⁺ is soluble but dangerous (Fenton chemistry), while Fe³⁺ is safer but nearly insoluble at physiological pH. Left alone, ferric iron precipitates into useless rust.

Transferrin solves this. The protein binds two Fe³⁺ atoms in deep clefts, each coordinated by tyrosine, histidine, and aspartate residues plus a synergistic carbonate anion. The grip is tight at blood pH. Iron is locked away from water, oxygen, and peroxide. Transferrin is the bloodstream's iron chaperone: it keeps the metal soluble, non-reactive, and flagged for delivery.

Transferrin's Conformational Change

Transferrin does not simply grip iron passively. The protein undergoes a substantial structural rearrangement upon binding, and this change is how cells distinguish loaded from empty transferrin.

Each transferrin lobe consists of two domains connected by a hinge. In the absence of iron, the domains remain relatively open, the binding cleft exposed and accessible. When Fe³⁺ enters along with a synergistic carbonate anion, the two domains rotate toward each other by roughly 50 to 60 degrees, closing around the metal like a jaw. The iron becomes buried, coordinated tightly by tyrosine, histidine, and aspartate residues from both domains. Solvent is excluded. The metal is locked away.

This closed conformation is what the transferrin receptor recognizes. TfR1 binds iron-loaded (holo) transferrin with high affinity and iron-free (apo) transferrin with low affinity at the neutral pH of the bloodstream. The receptor is effectively reading the protein's shape to determine whether it carries cargo worth internalizing.

After endocytosis, protons flood the endosome. The lower pH weakens the iron-protein interaction, but it also shifts the conformational equilibrium. The lobes spring open, releasing Fe³⁺ into the vesicle lumen. Apo-transferrin, now in its open conformation, remains bound to the receptor (their affinity is actually higher at acidic pH). The complex recycles to the cell surface together. At neutral extracellular pH, apo-transferrin's affinity for TfR drops and it dissociates, free to collect more iron.

Requesting Iron

Cells that need iron display transferrin receptors (TfR1) on their surface. The loaded transferrin-TfR complex is internalized by endocytosis. Inside the endosome, proton pumps drop the pH. This acidification weakens transferrin's grip, releasing Fe³⁺ into the vesicle lumen. A ferrireductase converts it to Fe²⁺, and DMT1 - the same transporter family used in the gut - moves iron across the endosomal membrane into the cytoplasm. The now empty transferrin, still bound to its receptor, recycles back to the cell surface where neutral pH releases it to collect more iron.

This cycle is elegant and economical. Transferrin is reused hundreds of times. Iron never exists free in plasma. And delivery is targeted: only cells expressing TfR receive iron.

Iron Storage and Regulation

Ferritin & Storage

Transferrin handles transport, but the body needs a vault. Ferritin is that vault - a hollow protein shell that converts reactive iron into inert mineral.

The ferritin shell assembles from 24 subunits into a spherical cage roughly 12 nm across. Iron enters through channels in the shell as Fe²⁺, but Fe²⁺ is precisely what the cell cannot tolerate in quantity. Ferritin solves this with ferroxidase activity: catalytic sites on the H-subunits oxidize Fe²⁺ to Fe³⁺, which then migrates to the interior and crystallizes as ferrihydrite, an iron oxide mineral. A single ferritin shell can store up to 4,500 iron atoms in this form.

Protein Pdb Id Role Of note:
Ferritin 1FHA Iron storage Single shell stores ~4,500 iron atoms as a mineral

Ferritin 24-subunit assembly →

The mineral core is the point. Iron packed as ferrihydrite is dense, stable, and unreactive. It cannot participate in Fenton chemistry. The protein shell meanwhile controls access. Iron doesn't leak out spontaneously. Mobilization requires reduction back to Fe²⁺ and chaperoning through the shell's channels, a regulated process that matches release to cellular need.

This is the body's iron buffer. When intake exceeds demand, ferritin fills. When demand exceeds intake, ferritin empties. The reserves buy time, months to years of protection against dietary shortfall.

A clinical note: serum ferritin is mostly iron-free, secreted by mechanisms still debated. But its concentration tracks body iron stores reliably enough to serve as the standard clinical marker. Low serum ferritin means depleted reserves. Elevated ferritin signals overload - or inflammation, since the body ramps up ferritin production as part of the inflammatory response, independent of iron status.

Iron Pathologies

One reason to study pathologies is to understand what has gone wrong, so that interventions to correct the problem can be made, but another angle is to look at pathologies, here pathologies of iron metabolism, because they shed light on what is going on in health.

Hemochromatosis is the disease of always on iron import. Patients absorb 3–4 mg of iron daily instead of 1–2 mg. The excess has nowhere to go, since there's no active excretion pathway to compensate, so iron accumulates over decades. Iron deposits in liver (cirrhosis), heart (cardiomyopathy), pancreas (diabetes), and joints. The underlying defect in hereditary hemochromatosis is usually a mutation in HFE, a protein that helps the liver sense iron status. Without this proper sensing the liver fails to produce enough hepcidin. Iron is imported on an ongoing basis despite reserves being high.

The "chroma" in 'hemochromatosis' refers to skin, not blood. The original description by Trousseau in 1865 described a patient with diabetes, pigmented cirrhosis, and bronze-colored skin, leading to the term "bronze diabetes." In 1889, Von Recklinghausen determined that the increased pigment in the liver was iron, and because internal bleeding was wrongly considered to be the source of this liver coloration, he called the disease "hemochromatosis."

The treatment for the excess iron is elegantly crude: remove blood. Phlebotomy - bloodletting - is the primary therapy. Each unit of blood withdrawn carries 200–250 mg of iron locked in hemoglobin. The body regenerates the lost red cells, pulling iron from overloaded stores to synthesize new heme. Weekly phlebotomy over months can normalize iron levels. Hemochromatosis patients often become regular blood donors. Their blood is perfectly normal, and donation serves double duty as treatment and supply.

Anemia of chronic disease is the opposite problem: a depletion of available iron. Patients with chronic infections, autoimmune conditions, or malignancy become anemic despite adequate iron stores. Their macrophages are full of iron; their ferritin is normal or elevated. But serum iron is low because hepcidin, driven up by inflammatory cytokines, has degraded ferroportin throughout the body. Iron is present but imprisoned. The body is prioritizing pathogen starvation over erythropoiesis.

Iron deficiency anemia is simpler: a genuine depletion. Stores exhausted, transferrin saturation low, insufficient iron for heme synthesis. Red cells emerge from the marrow small (microcytic) and pale (hypochromic), their hemoglobin tanks half-empty.

β-thalassemia produces iron overload through a different route to hemochromatosis. Ineffective erythropoiesis—precursor cells dying in the marrow before maturing. This suppresses hepcidin for the body senses "not enough red cells" and tries to increase iron availability. Meanwhile, patients require chronic transfusions. Each unit of blood delivers 200–250 mg of iron. Hepcidin suppressed, transfusional iron flooding in, no excretion: these patients can accumulate iron faster than hemochromatosis patients. Chelation therapy becomes a lifelong essential.

One of the used chelating agents is Deferoxamine, a siderophore - a bacterial iron-scavenging molecule, originally isolated from Streptomyces pilosus . It's a hydroxamate compound: a chain of alternating hydroxamic acid groups that wrap around Fe³⁺. Bacteria evolved these molecules to steal iron from hosts. We've repurposed the bacterial weapon.

Hepcidin: Master Regulator

One variable connects the four pathologies: hepcidin, a 25-amino-acid peptide hormone from the liver. Hepcidin's only known target is ferroportin. When hepcidin binds ferroportin, the complex is internalized and degraded. No ferroportin, no iron export from enterocytes, from macrophages, from any cell.

This gives the body centralized control over the single valve that matters. Hepcidin rises when iron stores are replete: ferroportin disappears, absorption stops, recycled iron stays locked in macrophages. Hepcidin falls when iron is needed: ferroportin returns to cell surfaces, dietary iron flows into plasma, macrophages release their stores.

Inflammation stops iron export by upregulating hepcidin. IL-6 directly induces hepcidin transcription, overriding iron-sensing signals. The result is hypoferremia, low serum iron, even when total body iron is adequate.

The hepcidin-ferroportin axis explains why iron metabolism has so few failure modes. There's essentially one control point. Disrupt hepcidin sensing (hemochromatosis) and the gate stays open. Overproduce hepcidin (inflammation) and the gate stays shut.

Within the Cell

Hepcidin governs the body's iron economy, the flux between gut, plasma, and macrophage. But once iron crosses the cell membrane, it enters a different jurisdiction. The cell must solve the same problem internally that the body solves: how to deliver a reactive metal to the processes that need it without letting it roam free.

The challenge is most acute in mitochondria. These organelles consume the bulk of cellular iron, building heme and assembling the iron-sulfur clusters that drive respiration. They also harbor the highest oxygen concentrations and the lipid-rich membranes most vulnerable to peroxidation. Everything required for Fenton disaster converges in one organelle.

Cells have evolved their own regulatory layer that's faster and more local than hepcidin signaling. When this regulation fails, the consequences play out not as systemic overload or anemia, but as something more targeted: the death of the cell.

Frataxin & Mitochondrial Iron

Mitochondria are the cell's largest iron sink. The final step of heme synthesis, ferrochelatase inserting Fe²⁺ into protoporphyrin IX, occurs in the mitochondrial matrix. So does the assembly of iron-sulfur clusters, the ancient cofactors that make the electron transport chain possible. A cell starved of mitochondrial iron cannot make ATP efficiently.

But mitochondria are also where Fenton chemistry is most likely to ignite. Oxygen concentrations are high (the terminal electron acceptor is right there). Iron flux is intense. The inner membrane is dense with polyunsaturated fatty acids, exquisitely sensitive to lipid peroxidation. This organelle lives at the edge of catastrophe.

Frataxin is the chaperone that keeps iron under control inside the mitochondrial matrix. It binds iron, delivers it to the machinery assembling Fe-S clusters, and prevents the accumulation of uncommitted reactive iron. Frataxin doesn't transport iron into mitochondria, it manages iron once it arrives.

Friedreich's ataxia reveals what happens without this management. Patients carry mutations that reduce frataxin expression, typically through a GAA trinucleotide expansion that silences transcription. The result is paradoxical: mitochondrial iron overload alongside iron-sulfur cluster deficiency . Iron enters the mitochondrion but it cannot be handed off to the biosynthetic machinery. Instead it accumulates in reactive form. The electron transport chain, starved of its Fe-S clusters, falters. Meanwhile, uncommitted iron catalyzes exactly the oxidative damage that working frataxin guards against.

The clinical picture follows: neurodegeneration (neurons are mitochondria-dependent), cardiomyopathy (the heart never rests), diabetes (pancreatic β-cells are metabolically intense). Frataxin deficiency is a slow Fenton catastrophe playing out first in the most oxygen-hungry tissues.

Aconitase

The TCA cycle lives in the mitochondrial matrix, and its third step depends on iron, not in heme, but in a [4Fe-4S] cluster.

Aconitase catalyzes the isomerization of citrate to isocitrate, a seemingly modest rearrangement: a hydroxyl group moves one carbon down the chain. But the mechanism requires first removing water (dehydration to cis-aconitate), then adding it back at a different position. The Fe-S cluster makes this possible.

Three of the cluster's four iron atoms are anchored to the protein by cysteine residues. The fourth iron, exposed, coordinatively unsaturated, is the business end. It binds the citrate substrate directly, acting as a Lewis acid to polarize the hydroxyl group and facilitate its departure. This is iron doing what the opening section promised: accepting electron density, stabilizing intermediates, enabling chemistry that would otherwise be too slow.

The cluster also illustrates something important about Fe-S cofactors. Electrons in a [4Fe-4S] assembly are delocalized across multiple iron and sulfur atoms. The cluster can accept or release an electron without violent reorganization at any single metal center. It's a stable parking place for electrons, buffered against the reactivity that makes free iron dangerous. In aconitase, the cluster isn't passing electrons down a transport chain; it's holding steady, providing a stable Lewis-acidic iron for catalysis.

The cluster is also a vulnerability. Oxidative stress can damage the cluster. Superoxide in particular attacks the exposed iron. When the cluster degrades, aconitase activity vanishes, and the protein doesn't simply sit idle. It becomes something else entirely. We'll return to this later.

Ferroptosis


Ferroptosis: An iron based equivalent of apoptosis - programmed cell death. Especially relevant to mitochondria.


The Frataxin section described why mitochondria are vulnerable: they have high oxygen, intense iron flux, and membranes rich in polyunsaturated fatty acids. Ferroptosis is what happens when that vulnerability is realized.

Ferroptosis is an iron-dependent form of regulated cell death, distinct from apoptosis. The mechanism is direct Fenton chemistry enacted on membranes. Iron catalyzes the peroxidation of polyunsaturated fatty acids (PUFAs) in membrane phospholipids, generating lipid radicals that propagate in chain reactions. The membrane loses integrity. The cell dies.

Protein Pdb Id Role Of note:
GPX4 6ELW Prevents lipid peroxidation Its inhibition triggers ferroptosis

Normally, GPX4 (glutathione peroxidase 4) prevents this. GPX4 reduces lipid hydroperoxides to harmless alcohols, breaking the chain reaction before it propagates. As long as GPX4 is active and supplied with glutathione, ferroptosis is held in check. Inhibit GPX4, deplete glutathione, or overwhelm the system with iron and the brake releases.

This makes ferroptosis exquisitely sensitive to iron status. Cells with higher labile iron pools are more susceptible. Mitochondria, for the reasons already discussed, are often where the damage concentrates.

Ferroptosis is now implicated across pathology. Neurodegeneration: neurons are mitochondria-dense and irreplaceable. Ischemia-reperfusion injury: blood flow returns, oxygen floods tissue, iron that leaked from damaged cells catalyzes a peroxidation wave. And cancer, where the therapeutic logic inverts: tumor cells with high iron and metabolic stress may be selectively vulnerable. So for treatments, induce ferroptosis in the tumor; inhibit it in the stroke. The same process, opposite goals.

IRE/IRP System: Cellular Regulation

As we've seen, the liver hormone Hepcidin controls iron at the systemic level, governing flux across tissues. But individual cells face their own moment-to-moment fluctuations. They need a faster, local system.

The IRE/IRP system is that system, and it operates entirely post-transcriptionally, which makes it fast. No signals travel to the nucleus. The cell adjusts iron handling by controlling mRNAs already present in the cytoplasm.

Iron-responsive elements (IREs) are stem-loop structures in the untranslated regions of specific mRNAs. Iron-regulatory proteins (IRP1 and IRP2) recognize and bind these loops, but only when intracellular iron is low.

The consequences depend on where the IRE sits:

The logic works as follows: When iron is scarce, the cell needs more iron coming in and less going into storage. IRP binding delivers exactly this: transferrin receptor mRNA stabilized (more uptake), ferritin mRNA silenced (less storage). And when iron is abundant, IRPs release. Ferritin translation proceeds. Transferrin receptor mRNA degrades. The cell shifts from acquisition to sequestration.

The reason post-transcriptional control matters is that transcription takes minutes to hours. The IRE/IRP response is nearly instantaneous. mRNAs are already there, ribosomes are already scanning. The cell can rebalance iron handling within a translational cycle.

IRP2 is straightforward: an RNA-binding protein degraded when iron is high. IRP1 is stranger. It has another life, one we've already encountered.

IRP1

We left aconitase with a promise: when the iron-sulphur cluster degrades, something else happens.

Here is what happens. The same polypeptide, identical amino acid sequence, refolds. A cleft that held the [4Fe-4S] cluster closes. A new surface becomes exposed, one that recognizes RNA stem-loops. The TCA cycle enzyme becomes IRP1, the iron-regulatory protein that governs ferritin and transferrin receptor expression.

This is not a minor conformational shift. The two forms have substantively different structure, different binding partners, different cellular locations, different functions. Yet they share a genome entry, a transcript, a ribosome's output.

Some call this "moonlighting", a protein with a second job. The framing implies a primary role and a side hustle. But a different way to look at it is that the same protein has two states, each with its own function. With cluster, Aconitase participates in the TCA cycle. Without cluster, IRP1 regulates iron homeostasis.

When iron is abundant, clusters are stable, and this protein functions as aconitase. When iron is scarce, cluster assembly falters, existing clusters degrade, and the protein accumulates in its now RNA-binding form.

The Fe-S cluster is both a cofactor for enzymatic activity and a entity sensed for iron. Its presence or absence reports cellular iron status directly, converting a metabolic signal into a change in gene expression. This also connects with the earlier points about coupling protein conformation to iron coordination. The extreme version shown here is that lacking iron coordination entirely the aconitase protein reconfigures as IRP1. The reverse directional conformational changes, from IRP1 to aconitase, are connected with loading the Fe-S cluster into the protein.

Cofactor Chemistry

The preceding sections traced iron's journey: absorption, transport, storage, cellular uptake, regulation. But we've been deliberately vague about what iron becomes once it arrives. What molecular forms does the metal actually take?

The answer is not simply "heme." Heme dominates our thinking because it's ubiquitous: hemoglobin, catalases, myoglobins, cytochromes, superoxide dismutase, chlorophyll. But heme represents one end of a spectrum. Iron in biology exists across a range of coordination environments, from loosely bound ions buffered by a few amino acid side chains, through iron-sulfur clusters, to the precision catalysis of a porphyrin macrocycle.

Each configuration reflects a trade-off. Looser coordination means faster exchange, more reactivity, greater vulnerability to Fenton chemistry. Tighter coordination means stability, specificity, but higher biosynthetic cost. The cell deploys different configurations for different tasks.

This section examines that spectrum. We begin with metallothioneins, not because they handle iron, but because they don't. The contrast clarifies what makes iron's chaperones so distinctive.

Metallothioneins & Metal Buffering


Vicinal Cysteines: Side by side or nearby to each other cysteine amino acid resiues in a protein sequence. From the Latin word for neighbouring or adjacent.


Not all metal-binding proteins handle iron. Metallothioneins are the cell's buffer for zinc, copper, and cadmium. They illustrate what iron chaperones are not .

Protein Pdb Id Role Of note:
Metallothionein 1MCT Zn/Cu buffering (not iron) Included to prevent confusion with iron chaperones

Metallothioneins are small, ancient, and ubiquitous across nearly all life. The human versions are roughly 60 amino acids, of which about 20 are cysteine, an extraordinary cysteine density. These cysteines aren't scattered randomly. They appear in Cys-X-Cys and Cys-Cys motifs, clustered in two domains. The thiolate sulfurs of these vicinal cysteines form cages that coordinate metal ions: seven zinc atoms, or more copper atoms in a different geometry.

The repetitive, cysteine-dominated architecture is the structural signature. When you encounter a small protein with this pattern, cysteine-rich, few other coordinating residues, repetitive motifs, you're likely looking at zinc or copper handling, not iron.

Why the distinction? It comes down to hard and soft. In coordination chemistry, "hardness" describes how tightly an atom holds its electron cloud. Hard species, small, highly charged, low polarizability, include Fe³⁺, oxygen donors, and nitrogen donors. They form strong electrostatic bonds with each other. Soft species, larger, lower charge density, more polarizable, include Cu⁺, Zn²⁺, and thiolate sulfur. Soft metals prefer soft ligands; hard metals prefer hard ligands.

Thiolate sulfur from cysteine is soft. It matches well with soft or borderline metals: Zn²⁺, Cu⁺, Cd²⁺. Iron, particularly Fe³⁺, is harder. It prefers oxygen and nitrogen donors. Histidine imidazoles, aspartate and glutamate carboxylates, tyrosine phenolates. Iron-sulfur clusters are an exception, but there the sulfur is inorganic sulfide locked in a rigid scaffold, not flexible cysteine side chains.

Manganese occupies an interesting middle ground. Mn²⁺ is chemically similar to Fe²⁺. It has comparable ionic radius and similar coordination preferences. Some enzymes can use either metal. Some bacteria have exploited this similarity, evolving to run on manganese precisely because hosts sequester iron during infection. When the body wages nutritional immunity by withholding iron, manganese-dependent pathogens slip through.

Calcium lies at the hard extreme. Ca²⁺ is large, highly charged, and exclusively coordinates oxygen—carboxylate side chains, backbone carbonyls, water molecules. The EF-hand motif, found in calmodulin and troponin C, is the classic calcium-binding architecture: a helix-loop-helix where the loop provides a cage of oxygen donors. No sulfur, no nitrogen. Shematrins, proteins that template calcium carbonate deposition in mollusc shells, use yet another strategy: repetitive, glycine- and serine-rich sequences that organize the mineral interface.

Iron's chaperones look different again. Frataxin, already discussed, uses a negatively charged surface—acidic residues, not cysteines. PCBPs (poly-rC-binding proteins) shuttle iron in the cytoplasm through mechanisms still being characterized, but again without metallothionein-like cysteine cages.

Metal-binding is not generic. Evolution built separate systems for separate metals. Much is down to the hardness or softness of the metal ions.

Iron-Sulfur Clusters

Iron-sulfur clusters are ancient and are in all likelihood older than proteins themselves. At hydrothermal vents, iron and sulfide precipitate spontaneously into mineral lattices structurally similar to biological [4Fe-4S] clusters. Life may have begun with these inorganic catalysts, only later wrapping them in peptides, then proteins.

The simplest biological forms are [2Fe-2S] clusters: two iron atoms bridged by two sulfide ions, typically coordinated by four cysteine residues from the protein. [4Fe-4S] clusters are cubes with four irons, four sulfides at alternating vertices, held by four cysteines. [3Fe-4S] clusters are [4Fe-4S] with one iron removed, sometimes a degradation product, sometimes a functional form. The notation tells you the stoichiometry: [Fe-S] means iron and inorganic sulfide, not cysteine sulfur.

What makes clusters useful? First, tunability. The redox potential of a cluster depends on its protein environment, the nearby charges, hydrogen bonds, solvent access. The same [4Fe-4S] core can be tuned across a wide voltage range depending on its scaffold.

Second, the capacitor property mentioned earlier. Electrons in a cluster are delocalized across multiple iron and sulfur atoms. The cluster can accept or donate an electron without dramatic reorganization at any single center. This buffers the system against the violent chemistry of one-electron transfers.

Third, modularity. Clusters can be assembled, inserted, removed, and rebuilt. The IRP1 story, aconitase becoming an RNA-binding protein when its cluster degrades, demonstrates that an Fe-S cluster can be treated as a module.

The respiratory chain uses both Fe-S clusters and irons in heme, sometimes in the same complex. Complex I is loaded with Fe-S clusters, eight of them in all, passing electrons from NADH to ubiquinone. Complex II has three; Complex III contains the Rieske [2Fe-2S] cluster. But as electrons approach oxygen, hemes take over: cytochrome c, cytochrome b, the a/a₃ pair in Complex IV. Clusters handle early electron transfers; hemes dominate closer to oxygen, where their constrained porphyrin environment controls the dangerous final reduction.

Minimal iron-binding motifs

Consider what came before elaborate protein scaffolds or even Fe-S clusters. A short peptid, just a few amino acids, can coordinate an iron ion. The peptide provides ligands; iron provides the Lewis acidity. Together they form a crude catalytic unit.

Only two amino acids contain sulfur. Cysteine has a thiol (-SH) that readily deprotonates to thiolate (-S⁻), a strong metal ligand. Methionine has a thioether (-S-CH₃), which coordinates more weakly and doesn't deprotonate. Cys-Cys or Cys-X-Cys motifs, coordinating iron through thiolate, trend toward Fe-S cluster chemistry.

But sulfur isn't essential for iron binding. Many non-heme iron enzymes use the 2-His-1-carboxylate facial triad : two histidine imidazoles and one aspartate or glutamate carboxylate, arranged on one face of an octahedron. This leaves coordination sites open for substrate and oxygen. The motif is widespread. It's found in dioxygenases, hydroxylases, oxidases, and it contains no sulfur at all.

So minimal iron-binding configurations follow different paths:

The pattern: local sequence provides two or three ligands, iron completes its coordination sphere with water or substrate, and catalysis becomes possible. The specificity is poor. Side reactions abound. But it works, minimally.

From this rough chemistry, a gradient can be climbed. Longer peptides provide more coordinating residues, more second-shell interactions, better exclusion of water. Folded proteins add precise geometry, access channels, conformational control. Fe-S clusters, requiring dedicated biosynthetic machinery (the ISC and CIA pathways), offer good specificity and tunability. Porphyrins, rigid macrocycles at the far end of the spectrum, provide maximal constraint and exquisite specificity, but at high biosynthetic cost.

Fe-S clusters balance cost, flexibility, and performance in a way that neither bare metal ions nor heme can match. This is perhaps why they remain so central to metabolism, and why they appear so early in any reconstruction of life's origins.

Heme Synthesis

Heme sits at the far end of the specificity gradient. Where Fe-S clusters offer modularity and tunability, heme offers constraint. The porphyrin ring is a rigid, planar macrocycle that locks iron into a precise coordination geometry, controls its redox potential, and prevents its escape.

The biosynthetic pathway reflects this commitment. Eight enzymatic steps, split between mitochondria and cytoplasm, build the porphyrin ring from simple precursors. It begins in the mitochondrial matrix: succinyl-CoA from the TCA cycle condenses with glycine to form δ-aminolevulinic acid (ALA). This is the committed step, catalyzed by ALA synthase and regulated by heme. Feedback inhibition prevents overproduction.

ALA exits to the cytoplasm. Two molecules condense to form porphobilinogen. Four porphobilinogens assemble into a linear tetrapyrrole, which cyclizes. Side chains are modified, methyls, vinyls, propionates arranged in precise positions. The pathway returns to the mitochondrion for the final steps: oxidation of protoporphyrinogen to protoporphyrin IX, then iron insertion.

Protein Pdb Id Role Of note:
Ferrochelatase 1HRK Inserts iron into protoporphyrin

Ferrocheletase

Ferrochelatase catalyzes this final step. It slots Fe²⁺ into the center of the waiting tetrapyrrole, displacing two protons. The product is heme. Iron is now locked into a scaffold that defines its chemistry.

Why such elaborate biosynthesis? Because the porphyrin isn't merely holding iron. It's tuning it. The rigid macrocycle prevents geometric rearrangement during redox cycling. The conjugated π system modulates electron density. Substituents on the ring periphery fine-tune the redox potential. And crucially, the tetrapyrrole cage keeps iron from escaping to wreak Fenton havoc.

Cells that synthesize heme must also export it. FLVCR1 and FLVCR2 are heme transporters that move the finished cofactor across membranes, out of the mitochondrion, out of the cell when necessary. Erythroid precursors, synthesizing heme at enormous rates, depend on this trafficking to balance production with globin availability.

Ligand Field & Spin States

Here is a puzzle. Ferrous iron (Fe²⁺) in solution reacts with oxygen rapidly and irreversibly, yielding ferric iron (Fe³⁺) and superoxide. This is oxidation, the reaction that rusts iron and spoils wine. Yet hemoglobin binds oxygen millions of times without being destroyed. The iron remains ferrous. The oxygen remains intact. When conditions change, O₂ leaves and the protein is ready for another cycle. How?

The answer lies in spin states, a quantum mechanical property that ligand field theory explains. The iron ion changes in size in a highly controlled way, related to its spin state.

Iron has six electrons in its 3d orbitals. These electrons can arrange themselves in two ways. In the high-spin state, electrons spread across all five d orbitals, maximizing unpaired spins. The iron is relatively large, paramagnetic, and reactive. In the low-spin state, electrons pair up in the lower-energy orbitals. The iron contracts, becomes diamagnetic, and behaves very differently toward incoming ligands.

Which state iron adopts depends on its ligand field, the electronic environment created by surrounding atoms. Weak-field ligands (water, chloride) leave iron high-spin. Strong-field ligands force electrons to pair, producing low-spin iron.

In deoxyhemoglobin, the iron sits slightly above the porphyrin plane, coordinated by four nitrogens of the ring and one histidine (the "proximal" histidine) from below. Five-coordinate, no sixth ligand, high-spin. The iron is too large to fit snugly in the porphyrin's central hole.

When oxygen binds, everything changes. O₂ is a strong-field ligand. It forces the iron into the low-spin state. The iron ion shrinks, drops into the plane of the porphyrin ring, and pulls the proximal histidine with it. This movement, a fraction of an angstrom, propagates through the protein and triggers the cooperative transition that makes hemoglobin such an effective oxygen carrier.

The spin transition does something else: it stabilizes the Fe²⁺-O₂ complex against electron transfer. In the low-spin state, iron's electrons are tightly paired and less available for the one-electron transfer that would produce destructive superoxide. The bound oxygen is held in a kind of electronic stalemate, neither fully reduced nor fully released.

The protein contributes further protection. The distal histidine, hovering above the binding site, donates a hydrogen bond to the bound O₂. This stabilizes a particular geometry, the bent "end-on" configuration, that favors reversible binding over oxidation. The hydrophobic pocket excludes water, which would catalyze the unwanted electron transfer. And the porphyrin ring, rigid and conjugated, fine-tunes the iron's redox potential to sit precisely where reversible binding is favored.

Free heme in aqueous solution lacks these protections. Water approaches, the geometry is unconstrained, and oxidation proceeds within milliseconds. The globin fold is not decoration. It is the machinery that makes reversible oxygen binding possible.

This illustrates a broader principle: iron's chemistry is not fixed by the metal alone. The ligand field, the protein environment, the exclusion of solvent, all determine whether iron oxidizes its substrate, binds it reversibly, or catalyzes some other reaction entirely. The same element, in different settings, performs different chemistry. Control of the ligand field is control of the reaction.

Heme Degradation & Iron Recycling

The essay began with absorption: 1–2 mg of iron crossing the gut epithelium daily. But this trickle is maintenance, not supply. The body's actual iron economy runs on recycling.

In humand red blood cells circulate for about 120 days, then senesce. Their membranes stiffen, their surface markers change, and splenic macrophages recognize and engulf them. Each erythrocyte carries over a billion heme groups. Each heme contains one iron atom. The macrophage now holds a vast quantity of the metal that must never be free.

Heme oxygenase is the enzyme that cracks open the porphyrin ring. It cleaves the α-methene bridge of heme, releasing three products: biliverdin (the green pigment, soon reduced to yellow bilirubin), carbon monoxide (here acting as a signaling molecule, one of the few contexts where CO is deliberately produced), and Fe²⁺. The iron that was carefully inserted by ferrochelatase is now liberated.

But that iron is not free. The macrophage immediately sequesters the iron into ferritin for storage, or through ferroportin for export. From the macrophage surface, exported iron is captured by transferrin and delivered to cells expressing transferrin receptors. The cycle restarts.

The numbers are striking. Roughly 25 mg of iron passes through this recycling loop daily, mostly from hemoglobin, with smaller contributions from other heme proteins and Fe-S clusters turning over. Compare this to the 1–2 mg absorbed from diet. Recycling provides more than 90% of daily iron supply.

This is why the body has no excretion pathway. Iron is too valuable and too dangerous to let drift away. The architecture described throughout this essay, transferrin, ferritin, hepcidin, ferroportin, the IRE/IRP system, exists to maintain a closed loop. Iron is passed chaperone to chaperone, returning endlessly to varied metabolic work, carrying oxygen, transferring electrons, and catalyzing essential reactions.


Protein Pdb Id Role Of note:
Myoglobin 1MBO O₂ storage in muscle First protein structure ever solved (1958)
Hemoglobin (HbA) 1A3N O₂ transport, CO₂ transport, Bohr effect
Fetal Hemoglobin (HbF) 1FDH Placental O₂ transfer Reactivating HbF is a treatment strategy for sickle cell
Sickle Hemoglobin (HbS) 2HBS Polymerization, sickle cell disease One amino acid change (Glu→Val) creates a disease and malaria resistance
Carbonic Anhydrase 1CA2 CO₂ ↔ bicarbonate interconversion Among the fastest enzymes known (~10⁶ reactions/sec)
Transferrin 1A8E Serum iron transport
Transferrin Receptor (TfR1) 1CX8 Cellular iron uptake
Ferritin 1FHA Iron storage Single shell stores ~4,500 iron atoms as a mineral
Ferroportin 6W4S Sole cellular iron exporter Only known iron export channel; hepcidin's target
Hepcidin 1M4F Master regulator of systemic iron A 25-amino-acid peptide controls body iron homeostasis
DMT1 5WME Intestinal iron absorption
Ferrochelatase 1HRK Inserts iron into protoporphyrin
Heme Oxygenase-1 1N45 Heme degradation, iron recycling Produces CO as a signaling molecule
Frataxin 1EKG Mitochondrial iron chaperone Deficiency causes Friedreich's ataxia
IRP1 (Aconitase) 2B3Y Iron sensing / TCA cycle enzyme Fe-S cluster present → aconitase
IRP1 (RNA Bound) 2B3Y Iron sensing / TCA cycle enzyme Fe-S cluster absent → RNA-binding IRP
IRP2 3SNP Iron-responsive RNA binding
GPX4 6ELW Prevents lipid peroxidation Its inhibition triggers ferroptosis
HFE 1A6Z Hepcidin regulation Resembles MHC class I but binds transferrin receptor, not peptides
Lactoferrin 1LFG Iron sequestration, antimicrobial
Cytochrome b5 Reductase (DCYTB) Ferrireductase at gut surface
FLVCR1 Heme export
Metallothionein 1MCT Zn/Cu buffering (not iron) Included to prevent confusion with iron chaperones
IL-6 1ALU Inflammatory cytokine; induces hepcidin Four-helix bundle; signals through gp130 receptor complex

Structural Highlights

Hepcidin - Remarkably small for a hormone with such systemic reach. Just 25 amino acids folded into a simple hairpin, stabilized by four disulfide bonds that cross-brace the structure. The disulfide density is extraordinary - eight cysteines in 25 residues means nearly a third of the peptide is devoted to structural integrity. This rigidity likely matters for receptor recognition.

Ferritin - A hollow sphere assembled from 24 subunits with octahedral symmetry. Iron enters through channels at the 3-fold axes, narrow enough to exclude most molecules but sized for Fe²⁺. The H-subunits contain ferroxidase centers where Fe²⁺ is oxidized; L-subunits provide nucleation sites for the mineral core. The interior surface is negatively charged, attracting iron cations inward.

Ferrochelatase - Contains a [2Fe-2S] cluster, which seems almost paradoxical: an iron-sulfur cluster in the enzyme that inserts iron into heme. The cluster isn't directly catalytic but is essential for activity, possibly positioning the enzyme correctly in the mitochondrial membrane or sensing iron status.

GPX4 - Uses selenocysteine at its active site, not cysteine. Selenium's lower pKa and higher reactivity make it better suited for reducing lipid hydroperoxides. GPX4 is also monomeric, unusual among glutathione peroxidases, which allows it to access bulky lipid substrates embedded in membranes.

HFE - Structurally mimics MHC class I proteins, complete with association with β2-microglobulin, but the peptide-binding groove is closed and non-functional. Evolution repurposed an immune scaffold for iron sensing. HFE binds transferrin receptor, competing with transferrin and modulating iron uptake.

Frataxin - A compact α-β sandwich with a striking acidic ridge: a surface patch dense with glutamate and aspartate residues. This negatively charged platform binds Fe²⁺ ions and delivers them to the iron-sulfur cluster assembly machinery. The structure explains why frataxin is a chaperone, not a storage protein - it presents iron rather than sequestering it.

Lactoferrin vs Transferrin - Both have two lobes that clamp shut on Fe³⁺. But lactoferrin keeps its iron at pH values where transferrin releases. The difference traces to specific residues near the binding site: lactoferrin has additional hydrogen bonds and a slightly different anion-binding arrangement that maintains grip in acidic, infected tissue.

IRP1/Aconitase - The structural switch is dramatic. With its [4Fe-4S] cluster intact, the protein is globular with a deep active-site cleft. When the cluster degrades, an entire domain rotates outward, exposing an RNA-binding surface that was previously buried. Same polypeptide, different fold, different function.

Metallothionein - Almost no regular secondary structure - just two metal-thiolate clusters wrapped in loops. The α-domain holds four metal ions; the β-domain holds three. The protein is essentially a scaffold for cysteine residues to coordinate metals. Its lack of conventional structure is the point: maximum metal-binding density in minimum space.

IL-6 - A four-helix bundle cytokine, a common fold for signaling molecules of this class. It binds first to IL-6 receptor (IL-6R), then this complex recruits gp130, the shared signal transducer. The ternary complex triggers JAK/STAT signaling that drives hepcidin transcription in hepatocytes - connecting inflammation directly to iron sequestration.

Terminology


Enterocyte : An absorptive epithelial cell lining the small intestine. Enterocytes are columnar, polarized cells with distinct apical (gut-facing) and basolateral (blood-facing) surfaces. Their apical membrane is covered in microvilli, the "brush border" that vastly increases surface area for nutrient absorption. In iron metabolism, enterocytes are the gatekeepers: they take up dietary iron across the apical membrane, decide whether to store it locally in ferritin or export it to the bloodstream, and are shed every 3–5 days, taking any stored iron with them. This short lifespan makes the enterocyte a disposable checkpoint - iron that enters but isn't exported is simply lost when the cell sloughs off.


Apical membrane : The surface of an epithelial cell that faces the lumen (the interior cavity of a tube or organ). In enterocytes, the apical membrane faces the gut contents and is where dietary nutrients first contact the absorptive machinery. It is structurally and functionally distinct from the basolateral membrane, which faces the bloodstream. This polarity matters for iron: DMT1 and DCYTB sit on the apical membrane to import iron from digested food, while ferroportin sits on the basolateral membrane to export iron into circulation. The two membranes have different protein compositions, ensuring that transport is directional - nutrients flow from gut to blood, not the reverse.


Porphyrin : A flat, ring-shaped molecule built from four smaller rings (pyrroles) linked in a square. The center is a cavity perfectly sized to grip a metal ion. With iron inserted, a porphyrin becomes heme; with magnesium, chlorophyll. The rigid, conjugated structure constrains the metal's geometry and tunes its chemistry.


Redox potential : A measure of how readily a molecule gains or loses electrons, expressed in volts. A higher (more positive) potential means the molecule is a stronger oxidizer - it wants electrons. A lower (more negative) potential means it gives electrons away more easily. Iron's redox potential sits in a middling range: reactive enough to participate in electron transfers, not so extreme that reactions become uncontrollable.


Endocytosis : The process by which a cell engulfs material from outside. The plasma membrane dimples inward, surrounds the target, and pinches off to form an internal vesicle called an endosome . In iron metabolism, transferrin bound to its receptor enters cells this way. The endosome then acidifies, triggering iron release.


5' UTR / 3' UTR : Regions of messenger RNA that flank the protein-coding sequence but aren't themselves translated into protein. The 5' UTR ("five-prime untranslated region") sits upstream, where ribosomes first land. The 3' UTR sits downstream, after the stop codon. Both regions can contain regulatory elements - like the iron-responsive elements (IREs) that control whether a transcript is translated or degraded.


Ligand : Any molecule or ion that binds to a central metal atom. In iron chemistry, ligands include the nitrogen atoms of histidine side chains, the oxygen atoms of aspartate and glutamate, and the sulfur atoms in cysteine or inorganic sulfide. The ligands surrounding iron determine its reactivity, redox potential, and which substrates can approach. The term comes from the Latin ligare , to bind.


Lewis acid : An electron acceptor - a chemical species with an empty or partially empty orbital hungry for electron density. Iron excels as a Lewis acid because its partially filled d-orbitals can accept electrons from nearby molecules, polarizing bonds and stabilizing reaction intermediates. This property makes iron useful as a catalyst but dangerous when uncontrolled.


Chelation : The gripping of a metal ion by a molecule with multiple binding sites, like a claw (from the Greek chele , crab's claw). Chelating agents wrap around metal ions, sequestering them. In iron overload, chelation therapy uses drugs like deferoxamine to capture excess iron and allow its excretion. The multiple attachment points make chelation much stronger than single-point binding.


Macrophage : A large phagocytic immune cell that engulfs debris, dead cells, and pathogens. In iron metabolism, splenic macrophages are the recycling centers: they consume senescent red blood cells, digest the hemoglobin, extract the iron, and either store it in ferritin or export it via ferroportin. A single macrophage processes millions of iron atoms daily.


Cytokine : A small signaling protein released by cells to communicate with other cells, particularly in immune responses. IL-6 is the cytokine most relevant here: released during inflammation, it travels to the liver and induces hepcidin production, linking infection status to iron availability. Cytokines are the immune system's chemical vocabulary.


Erythropoiesis : The production of red blood cells (erythrocytes), occurring primarily in bone marrow. Erythropoiesis is the body's largest consumer of iron - roughly 25 mg daily goes into synthesizing new hemoglobin. When erythropoiesis is ineffective (precursor cells dying before maturation, as in thalassemia), the body senses "not enough red cells" and suppresses hepcidin to increase iron availability.