Irisin is real, but most human studies used kits that missed it

What is irisin?

Irisin is a peptide cleaved from FNDC5 (fibronectin type III domain containing 5), a transmembrane protein expressed in skeletal muscle. Irisin belongs to the myokines, muscle-secreted signaling molecules such as IL-6, myostatin, and BDNF, through which contracting muscle communicates with other organs ^[5]⁠^[8].

Review after review describes the same proposed upstream pathway: exercise activates PGC-1α in muscle, PGC-1α drives FNDC5 expression, and FNDC5 is cleaved and released into circulation as irisin ^[3]⁠^[7]. That part of the story, the gene, the protein, and the cleavage event, is not what fell apart.

The dispute is downstream: how much irisin circulates in a human, how reliably it tracks with exercise, and what it does once it reaches fat, bone, heart, or immune tissue.

Why can't most irisin blood measurements be trusted?

Commercial irisin ELISA kits, the near-universal human assay for years, produced no band at irisin's expected size in a 2015 validation study and cross-reacted with unrelated serum proteins ^[2].

The kits were built on polyclonal antibodies. The validation study tested four of them by western blot against human and animal serum, using recombinant glycosylated and non-glycosylated irisin as positive controls ^[2]. None of the kits produced an immune-reactive band at the size expected for real irisin, and all four showed prominent cross-reactivity with unrelated serum proteins ^[2].

Mass spectrometry run on the same samples did find a real FNDC5-derived signature, at roughly 20 kDa, which the commercial antibodies missed entirely ^[2].

The authors concluded that their findings "call into question all previous data obtained with commercial ELISA kits for irisin" ^[2]. The caveat is not a footnote. Any study from that period, whether an exercise trial, an obesity comparison, or a disease-association cohort, that reported an irisin concentration or correlation from one of these kits cannot be assumed to have measured irisin at all.

Does exercise raise irisin?

Exercise is proposed to raise irisin through a PGC-1α to FNDC5 signaling axis in contracting muscle, a pathway the review literature describes consistently ^[3]⁠^[5]⁠^[7]. Most of the human dose-response numbers behind that claim come from the now-questioned ELISA kits.

How reliably irisin tracks with exercise in humans is therefore unsettled. None of this changes the case for exercise, which has plenty of other well-documented mechanisms behind it.

What health effects are proposed for irisin?

Reviews propose roles for irisin in muscle, bone, heart, lung, kidney, immune, and vascular tissue, none of them established as a causal human outcome. The proposed roles include:

• Muscle and bone. Mitochondrial quality control and osteoblast activity in musculoskeletal tissue ^[6].
• Cardiac fibrosis. A Sirt1 and TGF-β/Smad3-linked signaling pathway in cardiac fibroblasts ^[3].
• Organ protection. Protective signaling described across heart, lung, and kidney models ^[4].
• Immune cells. Modulation of immune cell proliferation and differentiation ^[8].
• Blood vessels. Involvement in oxidative stress and endothelial signaling pathways studied in atherosclerosis research ^[9].

Reviews published since the antibody problem surfaced have not abandoned irisin, but their language has visibly softened. These are legitimate hypotheses worth tracking. Almost all of the data feeding them are animal, cell-culture, or ELISA-based human correlations, precisely the layer the 2015 validation work put in doubt ^[2]. For a comparable case, see why osteocalcin's hormonal claims lack independent replication.

Is irisin a useful marker for sarcopenia?

Not yet: a 2025 systematic review concluded only that irisin "may serve as a potential diagnostic marker" for sarcopenia in older adults, not that it does ^[1].

The review searched four databases, identified 364 candidate studies, and could include only 21 that met its criteria: 12 in humans and 9 in animals or cells ^[1].

That hedge comes from one of the more recent and methodologically deliberate reviews available. It fairly summarizes where the field stands: an interesting signal sitting on a small verified base, with no established clinical utility yet.

Can a supplement or cold exposure raise irisin?

Claims that a supplement, cold-exposure protocol, or training regimen raises irisin are unverified unless a validated, mass-spectrometry-grade measurement stands behind them.

A product or article claiming to raise irisin by a specific number is, at minimum, citing a figure that cannot be checked against the assay problem. At worst, it is repeating a number generated by a kit shown to detect the wrong protein ^[2].

A specific irisin blood number is not worth shopping for, tracking, or treating as a metabolic score until the field agrees on how to measure it.

How can you judge whether an irisin number is reliable?

Ask which method measured the irisin. Discount a number from a pre-2015 commercial ELISA kit, and give more weight to a mass spectrometry result, though that body of work remains small.

The same habit of asking which method produced a number helps when reading a peptide certificate of analysis.

What is still unknown about irisin?

Irisin's true blood concentration, its effects on human tissue, and the validity of the older correlational studies all remain open, a gap most consumer coverage does not acknowledge:

• Older correlations. Whether the older human associations, irisin tied to obesity, mood, reproductive conditions, or bone density, would hold up if rerun with a validated assay is unknown. Until they are re-measured with mass spectrometry, they belong in "unresolved," not "confirmed" or "disproven."
• Blood concentration. A 2021 review of the broader "irisin system" lists irisin's physiological blood concentration among the field's unresolved questions ^[7]. Its resting and post-exercise levels in human blood are unsettled, a striking admission for a molecule with hundreds of published papers behind it.
• Human tissue effects. The receptor-level effects proposed for irisin on adipocytes, osteoblasts, and cardiac fibroblasts come from animal and cell models. Whether they translate into confirmed human tissue effects, or remain a plausible extension of rodent data, has not been shown directly in people ^[1]⁠^[3]⁠^[6].