The Intestinal Barrier: Junctions, Mucus and Permeability
A tight junction is a regulated channel, not a seal. Claudin-2 forms a cation-selective pore and is raised by TNF and IL-13 in inflammatory bowel disease, so barrier failure in this tissue is usually an active change in junction composition rather than structural damage \u2014 the epithelium opens a channel, it does not simply tear. The corollary is a warning about how this is measured: occludin-null mice have morphologically normal tight junctions, so measuring a junction protein is not measuring barrier function. Behind the junctions sit the mucus layer and the antimicrobial peptides that do most of the actual separating \u2014 MUC2-null mice develop spontaneous colitis, which establishes that barrier failure alone is sufficient for disease. And in front of them sits the only part of this map that can be manipulated rather than measured: who is allowed into the tissue. Click any protein for the matching Assay Genie In Vivo antibody or ELISA kit.
Composition, not integrity. The claudin family sets what a tight junction lets through, and they do not all do the same thing: claudin-2 forms a cation-selective pore rather than sealing, and it is upregulated by TNF and IL-13 in inflammatory bowel disease. A "leaky gut" in this setting is therefore a junction that has been reprogrammed rather than broken, which matters because the two have different treatments. ZO-1 is the scaffold anchoring the junction to actin and the most commonly imaged marker; EpCAM holds the sheet itself together, and EpCAM mutations cause congenital tufting enteropathy. The caution that belongs beside all of them: occludin-null mice have normal-looking junctions, so junction protein levels are a weak proxy for the thing anyone actually wants to know.
Most of the separation is chemical rather than physical. In health the MUC2 mucus layer keeps bacteria tens of micrometres from the colonic epithelium, and MUC2-null mice develop spontaneous colitis with no other insult \u2014 barrier failure alone is sufficient. REG3\u03b3 enforces that separation chemically under IL-22 control, and \u03b1-defensin 5 and lysozyme are the Paneth-cell equivalents in the small intestine, where Paneth-cell dysfunction is one of the more reproducible findings in ileal Crohn's disease. These are not markers of the barrier; they are the barrier.
Choose the permeability read-out carefully. Zonulin is the most used serum proxy and the least agreed \u2014 what the commercial assays detect is genuinely disputed \u2014 while LBP is the more interpretable one, because it rises when bacterial products actually reach the circulation and therefore measures a consequence rather than a component. Both are downstream of the barrier rather than of it, with their own clearance and production confounds, which is why a functional permeability test belongs alongside them rather than being replaced by them. Calprotectin is the third read-out and the one already in routine clinical use: what it measures is neutrophil presence in the lumen, which is worth remembering whenever it is treated as a proxy for something else.
The In Vivo tie-in, and where this map is thin. 16 of the 38 nodes carry a functional-grade antibody, and almost none of them is in the barrier itself \u2014 EpCAM is the single epithelial exception. What is blockable is who reaches the tissue and what they secrete. That side is unusually well covered: \u03b14\u03b27 (LPAM-1), the \u03b27 chain in two independent clones, \u03b14 (CD49d, clone PS/2) and CD62L are all functional grade, so the vedolizumab manipulation and the natalizumab manipulation can be run side by side and told apart \u2014 and two clones on one chain is what shows an effect to be epitope-independent. Alongside them sit CD3 (145-2C11), CD4 (GK1.5), CD8 (Ly-2), CD25, B220 (RA3-6B2), CD19, CD40 (FGK4.5, ultra-low endotoxin, which is not a formality in a tissue defined by its bacterial load), IL-10, TGF-\u03b2 and the VivoGenie anti-IL-17A. Stated plainly, the barrier is the read-out: occludin, ZO-1, claudin-2, zonulin, MUC2, REG3\u03b3, defensin 5, LBP, pIgR, secretory IgA, calprotectin, MAdCAM-1, CCL25, FOXP3 and ROR\u03b3t are ELISA or CLIA only; CCR9, CD103 and lysozyme and IL-22 are research grade; IL-23 and IL-17RA are biosimilar. Two cautions on the cytokine band: IL-17 maintains the barrier as well as inflaming it, which is why anti-IL-17 works in psoriasis and worsened Crohn's disease, and IL-22's receptor is absent from leukocytes entirely, so it is a repair cytokine that a cytokine-centric reading would misfile. And the \u03b27 reagents cannot separate homing from retention on their own, since the same chain pairs with \u03b1E as CD103 \u2014 pairing them with the \u03b14 and \u03b1E reagents is what makes that separation possible. For research use only; not for use in diagnostic or therapeutic procedures.
Every protein node links to a product — ELISA kit, In Vivo antibody or research antibody.