henhaobio
Sign in

bioRxivImmunologyPreprint

Intestinal uptake regulates T cell responses to dietary antigens

Ya-Ting Chang, Henry H. Le, Andrew Morrison, Ryan Kong, E.A. Schulman, Tsui-Wen ChouiD, Jamie BlumiD

PDFDOI
In-depth full-text digest available · Jump to digest ↓

Abstract

Dietary proteins induce antigen-specific immune responses, leading to oral tolerance or food allergies. While mouse models of allergic sensitization rely on antigen context and adjuvant signaling, exposure to purified proteins is usually sufficient for tolerance development in studies with soluble model antigens. In previous work, we discovered that the maize protein zein induces robust antigen-specific intestinal Tregs following normal chow exposure, but unexpectedly does not elicit a T cell response when delivered as a purified protein. Here, we investigated the mechanisms underlying these differential T cell responses, considering both biochemical properties of zein and its food matrix, as well as intestinal antigen processing. We focused on three zein preparations that induced different frequencies of antigen-specific intestinal T cells. While in vitro studies showed comparable presentation by dendritic cells, in vivo studies revealed stark differences in intestinal uptake. We identified differences in protein solubility and active sampling mechanisms as cooperative drivers of intestinal zein uptake. Different zein preparations could be sampled by either goblet cells, a route previously described for the model antigen OVA, or M cells, a route hypothesized for dietary proteins and established for transport of pathogenic bacteria. These findings also extend to peanuts, where the allergen Ara h 1 was selectively transported by goblet or M cells depending on food preparation, establishing examples of naturally occurring dietary M cell ligands. Finally, solubilizing zein increased intestinal uptake, and correspondingly feeding a diet with solubilized zein led to more intestinal zein-specific T cells compared to feeding the insoluble form. Overall, these findings suggest that intestinal uptake of dietary antigens is a regulated and context-dependent determinant of antigen-specific T cell induction, with implications for development of tolerance-restoring immunotherapies.

In-depth summary

Claude Sonnet 4.6 · xhigh effort · 8 Nov 2026 · from the full text of version 2 (9 Sept 2026) · CC BY 4.0

Intestinal uptake, set by protein solubility and goblet versus M cell sampling, limits how strongly different zein preparations induce antigen-specific intestinal T cells in mice.

Context-dependent T cell responses to food are usually attributed to adjuvants that activate antigen-presenting cells, while factors shaping the extent and route of epithelial sampling are far less resolved. This work shows that different preparations of one protein are taken up differently, and that uptake can limit tolerance-associated Treg induction.

Key results

  1. ShownCommercial purified zein (cZein) did not induce zein-specific intestinal T cells, whereas processed zein (pZein) and corn gluten meal (CGM) did. Adding back the non-zein remainder (spentCGM) to pZein did not raise T cell numbers, arguing against an adjuvant explanation.

    Mouse feeding experiments: cZein N=4-6/group, pZein vs cZein N=6-7/group, CGM/pZein/spentCGM N=3-4/group. Authors report T cell development varied 10-fold across preparations of the same protein; induced cells were Tregs. · Dietary or protein context is critical for αZein…; Discussion; Figure legends

  2. ShownTop zein isoforms, epitope presence, in vitro T cell activation, size and zeta potential were similar across preparations, and APC activation did not track T cell induction. Solubility and SEM morphology (pZein versus cZein) differed. Subcutaneous delivery gave comparable T cell induction.

    LCMS proteomics showed shared top isoforms; in vitro activation N=3/group; size and zeta N=5/group; solubility N=3/group, with pZein over twice as soluble as cZein. Subcutaneous injection gave comparable T cell induction across preparations. · Protein composition and particle properties…; Development of zein-specific T cells is limited…; Figure legends

  3. ShownFecal recovery after gavage (FRAG) was high for cZein and depleted for pZein and CGM. Organoid assays showed CGM crossing via M cells and pZein via goblet cells. M cell depletion with αRANKL increased fecal zein after CGM gavage, and FimH, latex beads or αGP2 blocked CGM transport.

    FRAG N=9-14/group; organoid transport CGM vs pZein N=4/group; soluble vs insoluble fractions N=3 to 5/group; inhibitor co-exposure N=3-5/group. αRANKL-treated mice had significantly more fecal zein than control antibody. · Development of zein-specific T cells is limited…; Figure legends

  4. ShownRaising cZein solubility by saccharide conjugation (SOL-cZein) reduced fecal recovery, shifted transport to goblet cells, and increased zein-specific intestinal T cells versus the mock-treated control (CTRL-cZein), without changing particle size, charge or in vitro T cell activation.

    SOL-cZein vs CTRL-cZein: solubility N=3/group; FRAG N=6-7/group; organoid transport N=3/group; feeding gave significantly more zein-specific T cells (Fig 4F; group size not stated in text). · Intestinal uptake can be modified by food…; Figure legends

  5. ShownRoasting peanut changed the route of Ara h 1 uptake. Ara h 1 from raw peanut was mainly transported by goblet cells, whereas from roasted peanut it showed more M cell translocation, and αRANKL increased its fecal recovery after roasted peanut gavage.

    Organoid transport raw vs roasted N=3-4/group; solubility by ELISA N=6/group. Roasted peanut had lower Ara h 1 solubility in organoid media but slightly higher in simulated intestinal fluid; αRANKL raised fecal Ara h 1. · Intestinal uptake can be modified by food…; Figure legends

Methods, data and reproducibility

Mice were fed diets containing zein, OVA or gliadin preparations, and antigen-specific intestinal T cells were measured. Zein and peanut preparations were characterized biochemically and biophysically, and uptake was assessed by fecal recovery after gavage and polarized goblet/M cell organoid transwells.

  • Mouse feeding experiments comparing zein preparations (cZein, pZein, CGM, spentCGM, SOL-cZein, CTRL-cZein) and OVA or gliadin diets, measuring intestinal antigen-specific T cells · mouse (C57BL/6) · n = N=3 to 7/group depending on experiment
  • FRAG gavage assays for zein and Ara h 1, including αRANKL M cell depletion versus control antibody · mouse · n = N=9-14/group (zein preparations); N=6-7/group (cZein variants)
  • Polarized intestinal organoid transwell assays with DAPT-induced goblet cells or RANKL plus TNFα-induced M cells · not stated in text (mice used in vivo) · n = N=3 to 5/group
  • LCMS proteomics, particle size, zeta potential, solubility and SEM characterization of zein preparations and peanut · n = N=3 to 6/group depending on assay
Methods & tools
LCMS-based proteomics, Scanning electron microscopy (SEM), Fecal recovery after gavage (FRAG) assay, Polarized intestinal organoid transwell assay, αRANKL antibody M cell depletion, In vitro APC and zein-specific TCR T cell line co-culture, Simulated intestinal fluid solubility assay, ELISA for Ara h 1, Particle size and zeta potential measurement, Saccharide conjugation for zein solubilization
Reproducibility gaps
The text gives no code repository, deposited proteomics or other data accession, or software versions; the in vitro T cell activation reporter lines and preparation sources are described only by name.
Limitations (5)
  • Initial attempts to measure zein in serum or intestinal tissue lacked sensitivity, so uptake was inferred from fecal recovery, which the authors note is consistent with either intestinal uptake or digestion. · stated by the authors
  • The site of T cell presentation for M cell-sampled antigen is unresolved, as is whether goblet and M cells hand antigen directly to RORγt APCs; the identity of the direct GP2 ligand (zein versus an insoluble CGM component) is also unknown. · stated by the authors
  • The authors note it is unclear why microbiome dependence appears only for some food antigens; they propose, but do not test, that microbial effects on M cell sampling could explain it. · stated by the authors
  • Mechanistic transport data rest on in vitro organoid assays with small groups (N=3 to 5/group) and on co-exposure inhibitors (FimH, beads, αGP2), which support but do not prove a GP2-dependent route. · noted in this summary
  • Saccharide conjugation may alter properties beyond solubility, so the solubility-to-uptake link is not isolated. The peanut analysis has no T cell readout, and why roasting makes Ara h 1 an M cell ligand is unexplained. · noted in this summary

Read this if you study oral tolerance or food allergy and want evidence that antigen sampling route and food form, not adjuvant activity, shape dietary T cell responses.

Citations
0

Reproducibility ledger

Nobody has reported trying to reproduce this paper yet. Did you replicate an experiment or run the code? Your report saves the next lab a week.

Sign in to file a reproducibility report.

Related papers

Similarity Map
SelectedSame fieldCross-field
THISBlum '26Lockhart '23Rudnitsky '25Zhou '25Ansaldo '26Canesso '22
81% semantic matchImmunology
Jamie Blum, Ryan Kong, E. A. Schulman+7 more·Science Immunology·Mar 6, 2026· 14 citations