Zonulin and intestinal-barrier regulation

Zonulin provides an unusually instructive example of how understanding of intestinal-barrier regulation can evolve.

The pathway emerged from work led by Alessio Fasano and colleagues investigating mechanisms that regulate the paracellular route across the intestinal epithelium. Zonulin was described as an endogenous modulator capable of inducing reversible changes in tight-junction organisation and intestinal permeability.^¹–³

This work contributed to an important conceptual shift.

The intestinal barrier is not simply intact or disrupted. Its permeability can be actively regulated in response to signals arising at the mucosal interface.

But the zonulin story has also become more complex with time. Questions concerning molecular identity, related proteins, assay specificity and the interpretation of circulating measurements have made it increasingly important to distinguish the biological pathway from the biomarker used to study it.

That distinction is central to understanding what zonulin research has established — and what remains uncertain.

A pathway for reversible barrier opening

Tight junctions are dynamic molecular complexes connected to intracellular signalling pathways and the actin cytoskeleton.

Their organisation can therefore change without epithelial destruction.

Early work on zonulin described a signalling pathway capable of producing reversible modulation of intercellular tight junctions and increasing paracellular permeability.^¹–⁴ The response involves intracellular signalling and cytoskeletal reorganisation rather than simply mechanical disruption of the epithelial layer.

This distinction matters.

An increase in paracellular permeability is not necessarily evidence of tissue damage.

A regulated opening of the paracellular route may occur transiently in response to environmental or physiological signals. Its biological significance depends on context, magnitude, duration and the ability of the epithelial system to restore its previous state.

The relevant question is therefore not simply whether permeability increases.

It is whether permeability remains appropriately regulated.

Signals from the intestinal environment

The zonulin pathway also illustrates how the intestinal epithelium responds to its environment.

Bacterial exposure was among the early experimental stimuli shown to induce zonulin secretion. El Asmar and colleagues demonstrated host-dependent zonulin release after exposure of the small intestine to bacteria, accompanied by altered barrier function.^⁴

Gliadin provides another important experimental and clinical context.

Studies associated with coeliac disease showed that gliadin exposure can trigger zonulin release and changes in intestinal permeability, helping establish a mechanistic connection between luminal sensing and tight-junction regulation.^²,³,⁵

These observations should not be interpreted as evidence that every microbial exposure or dietary protein produces pathological permeability.

Their broader importance lies in demonstrating that the epithelial barrier is responsive.

The intestine does not merely separate the luminal environment from the body.

It senses that environment and changes its behaviour accordingly.

From zonula occludens toxin to zonulin

The discovery of zonulin itself arose from an earlier observation.

Research on Vibrio cholerae had identified zonula occludens toxin, a bacterial protein capable of reversibly modifying tight-junction permeability. This prompted the search for an endogenous human counterpart capable of regulating the same biological system.^¹,²

Zonulin was subsequently described as such an endogenous regulator.

Later molecular work identified zonulin as pre-haptoglobin 2, the uncleaved precursor of haptoglobin 2.^⁶

This finding provided an important molecular anchor for the pathway and linked zonulin biology to the haptoglobin system.

Yet it did not close the molecular question.

Subsequent studies suggested that proteins related structurally or immunologically to pre-haptoglobin 2 may also contribute to what has been described experimentally or clinically as zonulin activity.

From one molecule to a zonulin family

The molecular interpretation became more complicated when investigators examined what commonly used antibodies and commercial assays labelled as “zonulin” were actually detecting.

Scheffler and colleagues reported that a widely used commercial zonulin ELISA did not detect recombinant pre-haptoglobin 2 as expected. Their experiments identified properdin as a potential target and proposed that it could represent another member of a broader zonulin family.^⁷

Further work has proposed additional structurally related candidates, including coagulation factor X and CD5 antigen-like protein.^⁸

The resulting concept of zonulin-family proteins is scientifically useful because it separates two propositions that are often conflated:

1. pre-haptoglobin 2 has been identified as zonulin; and 2. biological or immunological signals reported as “zonulin” may not always represent pre-haptoglobin 2 alone.

This distinction does not invalidate the biological pathway.

It changes the precision with which some observations should be interpreted.

A pathway is not its assay

This is where zonulin becomes especially important as an example of scientific inference.

Evidence for a biological mechanism and evidence for the analytical validity of a biomarker assay are different forms of evidence.

The experimental literature supports zonulin-related regulation of tight junctions and paracellular permeability.^¹–⁶

That does not establish that every commercial assay marketed as measuring zonulin quantitatively measures pre-haptoglobin 2 or the activity of that pathway.

Ajamian and colleagues questioned the interpretation of serum zonulin as a direct marker of intestinal mucosal barrier function, while the work of Scheffler and colleagues raised specific concerns about the molecular target of a widely used ELISA.^⁷,⁹

This creates an essential inference boundary:

a measured “zonulin” concentration is not automatically equivalent to zonulin-pathway activity, and neither is automatically equivalent to intestinal permeability.

These may be biologically related observations.

They are not interchangeable measurements.

The analytical details of zonulin assays, their relationship to functional permeability measures and their interpretation in human studies require separate consideration. Those questions belong primarily to the measurement literature and are addressed separately within the Permeability section of BIRNA Research.

Experimental models allow the pathway to be interrogated

Experimental models provide an opportunity to investigate zonulin-related biology in ways that are difficult to achieve in humans.

This has been directly relevant to my own research.

Using wild-type and zonulin-transgenic mice, we investigated the effects of bovine colostrum on intestinal barrier-related outcomes, microbiota and behaviour. In subsequent transcriptomic work, we examined intestinal gene-expression responses in the same broader experimental context.^¹⁰,¹¹

The transcriptomic study identified treatment-associated changes in genes and pathways relevant to tight-junction biology and inflammatory signalling. Importantly, the transcriptional response differed between female and male animals.^¹¹

These findings provide mechanistic information about biological responses occurring in a model with altered zonulin-related signalling.

They do not demonstrate that bovine colostrum produces the same molecular effects in humans.

Nor does altered expression of tight-junction-related genes itself constitute a direct functional measurement of intestinal permeability.

The model allows a mechanism to be interrogated.

Translation requires additional evidence.

Human evidence presents a different problem

Human studies provide direct relevance to human biology, but causal inference is substantially more difficult.

This distinction was evident in our systematic review and meta-analysis examining zonulin-dependent intestinal permeability in children diagnosed with mental disorders.^¹²

The literature suggested differences in zonulin-related measurements in some populations, but the studies varied in design, clinical population and methodology.

The findings therefore could not establish a simple causal sequence in which increased zonulin leads to increased intestinal permeability and subsequently produces a mental disorder.

Several inferential steps would be required to support such a claim.

First, the measured biomarker would need to validly represent the relevant biological process.

Second, altered intestinal permeability would need to be demonstrated.

Third, temporal and mechanistic evidence would be required to establish that the permeability change contributed causally to the clinical phenotype.

An association at one level cannot supply evidence missing at another.

Zonulin within the microbiota–gut–brain context

The same caution becomes especially important when zonulin is discussed in relation to the microbiota–gut–brain axis.

Altered intestinal permeability has been proposed as one possible route through which luminal, microbial and immune signals could influence systemic physiology and, indirectly, neural function.

That hypothesis is biologically plausible.

But plausibility is not demonstration.

A study reporting a difference in serum zonulin alongside a neurological or psychiatric phenotype does not, by itself, establish:

microbiota alteration → zonulin activation → increased intestinal permeability → systemic signalling → altered brain function.

Each arrow represents a separate scientific claim.

Each requires evidence.

This is precisely why zonulin is so useful as an example of the difference between an attractive mechanistic narrative and a demonstrated biological sequence.

Regulation is not synonymous with pathology

The discovery of zonulin helped demonstrate that intestinal permeability can be regulated.

That finding should not be transformed into the opposite oversimplification: that activation of a permeability-regulating pathway necessarily represents disease.

Barrier physiology requires flexibility.

A living interface must respond to microbes, nutrients, immune signals and other environmental information while maintaining sufficient compartmentalisation to protect the internal environment.

Transient permeability changes may therefore occur within normal physiology.

The clinically important question is whether the response is appropriate, reversible and proportionate to context — or whether regulation becomes persistent, exaggerated or otherwise maladaptive.

This distinction between regulated permeability and dysregulated permeability is more informative than treating permeability itself as pathological.

What the zonulin story teaches us

Zonulin has become important for reasons that extend beyond one protein.

Its scientific history illustrates several principles central to barrier research.

Permeability is regulated. The paracellular route can change dynamically in response to biological signals.

Regulation is contextual. Barrier opening is not automatically barrier failure.

Mechanism and measurement are different. Evidence for a regulatory pathway does not validate every assay used to study it.

Molecular identity can become more complex as methods improve. The movement from zonulin as a single molecular entity towards a possible family of related proteins reflects refinement rather than scientific failure.

Experimental models and human biomarkers provide different evidence. Neither can simply substitute for the other.

Association does not complete a causal pathway. Every transition from biomarker to permeability, from permeability to systemic physiology and from systemic physiology to disease requires its own evidence.

The zonulin story therefore provides more than a mechanism for intestinal-barrier regulation.

It provides a lesson in how biological evidence should be interpreted.

The original discoveries opened an important scientific question.

The work that followed made the answer more complex.

That is not a reason to discard the pathway.

It is a reason to describe it more precisely.