IGFBP-3: Binding Protein Biology, Regulation, Measurement

What is IGFBP-3 and what does it do?

IGFBP-3 (insulin-like growth factor binding protein 3) is the most abundant IGF binding protein in human blood and the principal carrier of circulating IGF-1 and IGF-2. Almost no IGF-1 circulates free; the large majority sits inside a ternary complex of roughly 150 kDa built from IGF-1, IGFBP-3 and the acid-labile subunit (ALS). That complex is too large to cross the capillary barrier, which converts a growth factor with a half-life of minutes into one that persists for hours. The same protein also acts independently of IGF ligands at the cell surface and inside the nucleus.

Here is the simplification worth correcting straight away. IGFBP-3 is not a passive sponge. It is a protease-sensitive, growth-hormone-responsive glycoprotein with its own signalling repertoire, and any assay figure reports total immunoreactive protein rather than intact binding capacity. What follows covers structure, ternary complex assembly, transcriptional and post-translational regulation, IGF-independent signalling, how the six binding proteins differ from one another, and how research laboratories actually measure the thing.

What happens to IGF-1 once IGFBP-3 captures it in the bloodstream?

Three things change at once: distribution, clearance and receptor access. Free IGF-1 is a vanishingly small share of the total pool, commonly reported as well under one percent of circulating immunoreactive IGF-1. The rest is partitioned between binary complexes (ligand plus a single binding protein, around 50 kDa, able to cross the endothelium) and the ternary complex, which is confined to the vascular space.

Clearance differences are the headline. The literature commonly reports a circulating half-life of roughly ten minutes for free IGF-1, around twenty to thirty minutes for binary complexes, and in the region of twelve to fifteen hours once the ligand is locked into the 150 kDa assembly. That is a two-order-of-magnitude extension achieved without altering the growth factor itself.

There is a safety dimension to the carriage as well. IGF-1 cross-reacts with the insulin receptor at high free concentrations, and sequestration inside the ternary complex is the main reason a large circulating reservoir of an insulin-like peptide does not produce continuous hypoglycaemia. Sequestration is also why total IGF-1 and bioactive IGF-1 can move in opposite directions. A sample can carry a high total ligand concentration and still deliver very little signal to the type 1 IGF receptor if the carrier pool is intact and ALS is plentiful.

For anyone working with IGFBP-3 in a culture or serum context, that distinction between total pool and available pool is the single most consequential fact on this page.

How is the binding protein built, and why does its structure matter?

The IGFBP3 gene sits on the short arm of chromosome 7, and the secreted mature protein is a 264-amino-acid chain organised into three structural regions: a cysteine-rich N-terminal domain, a poorly conserved central linker, and a C-terminal domain with a thyroglobulin type-1 fold. Both the N-terminal and C-terminal domains contribute contact residues to the IGF binding site, which is why isolated fragments bind ligand far more weakly than the intact molecule.

Eighteen cysteine residues form the disulphide scaffolding that holds those domains in register. That matters practically: reducing conditions during sample preparation destroy ligand binding, which is why native or non-reducing formats are required for ligand blotting while denaturing formats are fine for immunodetection.

The protein carries three N-linked glycosylation sites, and glycosylation is why IGFBP-3 runs as a doublet in the region of 40 to 45 kDa on SDS-PAGE despite a non-glycosylated core closer to 29 kDa. Seeing that doublet is a useful sanity check on sample integrity.

The C-terminal domain also contains a basic, heparin-binding stretch that overlaps the nuclear localisation signal and provides the ALS contact surface. Those functions are physically layered on top of one another, which means one short motif determines whether a given molecule of IGFBP-3 joins a ternary complex, sticks to cell-surface glycosaminoglycans, or gets imported into the nucleus. Most summaries treat carriage and nuclear signalling as separate stories. Structurally they compete for the same patch of protein.

What is the ternary complex with IGF-1 and ALS?

The ternary complex is a 1:1:1 assembly of an IGF ligand, a binding protein and ALS, and it forms in a strict order. ALS will not bind a free carrier. It only engages the binary complex, so ligand must load first. ALS itself is a leucine-rich repeat glycoprotein encoded by IGFALS, produced in the liver under growth hormone control, and heavily glycosylated.

Only two of the six binding proteins can complete this assembly: IGFBP-3 and IGFBP-5. In adult serum the overwhelming majority of ternary complexes are built on IGFBP-3 simply because it is present in far greater quantity.

Human genetics shows how load-bearing the arrangement is. Inactivating mutations in IGFALS produce acid-labile subunit deficiency, in which total circulating IGF-1 and IGFBP-3 fall dramatically because the reservoir can no longer be stabilised, yet the growth phenotype is comparatively mild. That dissociation between a collapsed circulating pool and near-normal stature is one of the clearest natural demonstrations that tissue-level IGF signalling is driven more by local production than by the size of the blood reservoir. It is also a caution against reading a low carrier concentration as a direct index of growth factor action.

What regulates IGFBP-3 levels, and how tightly is it linked to growth hormone?

Growth hormone is the dominant regulator. GH binding to its receptor activates JAK2 and phosphorylates STAT5b, which drives hepatic transcription of IGF1, IGFBP3 and IGFALS more or less in parallel. That shared transcriptional control is why the three move together, and why STAT5b loss-of-function produces GH insensitivity with low IGF-1, low binding protein and low ALS at the same time. Receptor-level defects such as those underlying Laron syndrome produce the same pattern from one step upstream.

Nutrition modulates the axis independently. Energy restriction and protein deficiency reduce hepatic GH receptor expression and lower the circulating pool, which is part of why IGFBP-3 falls in catabolic states, critical illness and malnutrition even when GH secretion is high.

Transcriptional inputs extend well beyond the somatotropic axis. The IGFBP3 promoter was identified in the mid-1990s as a direct transcriptional target of p53, linking the gene to the DNA damage response. Retinoic acid acting through retinoid receptors, transforming growth factor beta, vitamin D metabolites and several antiproliferative signals also induce expression in various cell types, while insulin tends to suppress it.

Age is the other large variable in any dataset. Circulating concentrations rise through childhood, peak around puberty alongside the pubertal IGF-1 surge, and decline gradually through adult life. Comparing IGFBP-3 across age groups without age-stratified reference intervals is one of the most common analytical errors in the published literature, and it is entirely avoidable.

Which proteases cleave the carrier, and what changes when they do?

Proteolysis is the fast regulatory layer, and it operates on a timescale transcription cannot match. Limited cleavage within the central linker region produces fragments with sharply reduced ligand affinity, which releases IGF-1 locally without changing the amount of immunoreactive protein present.

Several protease families do this. Pappalysin-2 (PAPP-A2) cleaves IGFBP-3 and IGFBP-5 specifically. Prostate-specific antigen, a kallikrein-family serine protease, cleaves it in seminal plasma and prostatic tissue. Matrix metalloproteinases, plasmin and cathepsin D contribute in other compartments, and pregnancy serum shows marked proteolytic activity that fragments a large share of the circulating pool.

The interpretive consequence is sharp. Immunoassays detect fragments if the antibody epitopes survive cleavage, so total IGFBP-3 can read normal or high while functional binding capacity has collapsed. Human PAPP-A2 deficiency is the textbook illustration: affected individuals show elevated total IGF-1 and elevated total carrier protein alongside reduced free IGF-1 and growth failure, because nothing is releasing the ligand from the complex.

A practical implication that rarely makes it into method sections: if an experimental condition involves serum from a proteolytically active source, an immunoassay result and a ligand blot can disagree by a wide margin, and the ligand blot is the one reporting binding competence. Running both is the only way to tell apart a change in protein quantity from a change in protein function.

What are the IGF-independent actions of this binding protein?

Some actions require no ligand at all. The C-terminal nuclear localisation signal is recognised by importin-beta, and nuclear import has been demonstrated in multiple cell types. Once inside, the protein interacts with retinoid X receptor alpha and modulates retinoid-dependent transcription, a route that connects it to antiproliferative and apoptotic programmes rather than to growth factor carriage.

At the cell surface, the literature describes binding to a type V transforming growth factor beta receptor identified as LRP-1, with downstream Smad-dependent signalling, as well as an interaction with the mitochondrial-derived peptide humanin. IGF-independent apoptosis induction has been reported in cell systems where type 1 IGF receptor signalling is blocked or absent, which is the experimental design required to separate the two mechanisms.

Treat these findings as cell-type dependent and context dependent rather than universal. The honest summary is that carriage is well characterised and quantitatively dominant, while the intracellular repertoire is real, reproducible in defined systems, and still being mapped.

How do the six IGF binding proteins compare?

All six high-affinity binding proteins bind IGF-1 and IGF-2 with affinities exceeding those of the type 1 IGF receptor, but they differ in abundance, regulation and secondary function. The table below is the short version for anyone selecting which one to measure or neutralise in an experiment.

How is IGFBP-3 measured in a research laboratory?

Immunoassay is the workhorse. Immunoradiometric assays, enzyme-linked immunosorbent assays and chemiluminescent platforms all quantify total immunoreactive protein, and liquid chromatography tandem mass spectrometry methods have been developed where fragment-level discrimination or freedom from antibody interference is required. Western ligand blotting, in which labelled IGF is used to probe non-reduced samples, remains the reference approach for assessing binding competence rather than quantity.

Three pre-analytical and analytical issues dominate the reproducibility problem. First, epitope masking: when the protein is locked inside a ternary complex, antibody access to some epitopes is restricted, so results vary with the dissociation step a given method uses. Second, assay harmonisation: platforms calibrated against different preparations and using different antibody pairs generate results that are not interchangeable, so longitudinal datasets must stay on a single platform. Third, fragment cross-reactivity, which inflates apparent concentrations of IGFBP-3 in proteolytically active samples such as pregnancy serum.

Sample handling is comparatively forgiving. The circulating protein shows less diurnal variation than growth hormone itself and tolerates ordinary serum separation, but repeated freeze-thaw cycles and prolonged ambient storage degrade recovery, and haemolysed samples should be rejected. Results should always be expressed against age-matched and sex-matched reference intervals generated on the same platform, with molar units used where the ratio to IGF-1 is being calculated.

IGFBP-3 was among the candidate markers examined during growth hormone biomarker development work; the method ultimately adopted for anti-doping purposes relies on IGF-1 together with procollagen type III N-terminal propeptide rather than the binding protein.

Why do engineered IGF-1 analogues bind the carrier so poorly?

Because they were designed to. Long R3 IGF-1 (IGF-1 LR3) carries a thirteen-residue N-terminal extension and an arginine substituted for glutamate at position 3, and des(1-3)IGF-1 simply lacks the first three residues. Both modifications disrupt the N-terminal contacts that binding proteins rely on, so affinity for IGFBP-3 and its relatives drops by orders of magnitude while affinity for the type 1 IGF receptor is largely preserved.

That design choice has a specific laboratory purpose. In cell culture, serum and conditioned media contain endogenous binding proteins that sequester native IGF-1 unpredictably, making concentration-response data difficult to reproduce between labs and between serum lots. An analogue that evades IGFBP-3 produces a far more stable effective concentration at the receptor, which is why these variants became standard reagents in serum-free and low-serum media development long before they appeared in any other context. The flip side is that any extrapolation from analogue data to native ligand biology must account for the missing carrier layer.

What if the numbers do not line up?

What if two assay platforms report different concentrations on the same sample?

Treat the discrepancy as expected rather than as an error until calibration is ruled out. Binding protein immunoassays differ in antibody pairs, calibrator material and sample pre-treatment, and cross-platform agreement for IGFBP-3 has historically been imperfect. Re-run both arms of a comparison on one platform, report the platform in the methods, and never merge historical data generated on a different system into a single trend line.

What if total protein reads normal but biological effect looks exaggerated?

Suspect proteolysis before suspecting the assay. Fragmented carrier retains immunoreactivity while losing ligand affinity, so total concentration can look untouched while free IGF-1 rises. A non-reducing ligand blot showing fragment bands below the usual 40 to 45 kDa doublet, or a parallel free IGF-1 measurement, will separate the two possibilities quickly.

What if serum lot changes shift a cell culture result?

Check the binding protein content of the new lot before redesigning the experiment. Serum carries variable quantities of IGFBP-3 and related species, and a lot change can alter free ligand availability enough to move an entire response curve. This is the specific confound that drove adoption of binding-protein-resistant IGF analogues in culture work.

What if a sample sat at room temperature overnight?

Document it and treat the result as provisional. Prolonged ambient exposure and repeated freeze-thaw cycles degrade recovery and can generate fragments that behave differently across assay formats. Protein degradation is not visible to the eye and will not announce itself in the readout.

The uncomfortable truth about reading a single number

Let us be direct about this: a standalone IGFBP-3 concentration tells you much less than most datasets imply. It is a composite of growth hormone output, hepatic synthetic capacity, nutritional state, age, proteolytic activity and assay idiosyncrasy, and the widely used molar ratio to IGF-1 is a crude surrogate for free ligand that has been repeatedly criticised because it ignores ALS, the other five binding proteins and fragmentation entirely. Interpreted alone, it is a weak signal. Interpreted alongside IGF-1, ALS and an age-matched reference interval, it becomes genuinely informative.

Nothing described here is guidance for any living subject. All compounds discussed on this page are sold for laboratory research use only and are not for human consumption. Anyone with a question about an animal should talk to their veterinarian.

How was this page researched?

This explainer was assembled from the published endocrinology and protein biochemistry literature on the insulin-like growth factor axis: primary research and review articles covering binding protein structure and glycosylation, ternary complex assembly and ALS biology, growth hormone signalling through JAK2 and STAT5b, characterised protease families and their substrate specificity, documented IGF-independent nuclear and cell-surface actions, and published method comparisons for immunoassay and mass spectrometry quantification of IGFBP-3. Human genetic syndromes referenced here, including acid-labile subunit deficiency, STAT5b loss of function and pappalysin-2 deficiency, are drawn from the clinical genetics literature describing those conditions. No first-party laboratory testing was performed for this page, and no numeric value has been asserted where the literature reports a range rather than a point estimate. Real Peptides publishes lot certificates of analysis for catalogue compounds.

Related reading across the IGF and laboratory verification hubs

For the applied side of this biology, the IGF-1 LR3 research hub collects the structural and experimental background, with supporting detail in what IGF-1 LR3 is, what IGF-1 actually does, the Long R3 modification explained, the development history, the study summaries, the frequently asked questions page, the 2026 research news roundup, the reported side effect literature, the safety profile overview, the published safety studies, the strength and performance literature, the oral stability and taste discussion, and the comparison pages covering CJC-1295 versus IGF-1 LR3, IGF-1 LR3 versus ipamorelin and sermorelin versus IGF-1 LR3. Laboratory handling and verification are covered in the peptide storage guide, the degradation pathways explainer, the lyophilised versus liquid stability comparison, the cold storage notes for IGF-1 LR3, the transport and shipping considerations and the milligram-per-millilitre math walkthrough. Analytical documentation is explained in how to read a certificate of analysis, HPLC peptide analysis, HPLC versus mass spectrometry, mass spectrometry verification and research quality standards, with regulatory framing in what research use only means, published lot certificates, sourcing guidance and shipping coverage.

The reason IGFBP-3 keeps appearing in endocrinology papers decades after it was first purified is that it sits exactly where the axis becomes interesting: it is simultaneously the reservoir, the brake and, in the nucleus, a signal in its own right. Most molecules in biology do one job well. This one does three, and the assay on your bench can only see the first. Any experiment that treats the carrier as background noise rather than as a variable is measuring a system it has not fully defined.