The research, led by Xuanjing Yu, Xinyu Jiang, and colleagues at Yunnan University’s Bio-X Center for Interdisciplinary Innovation and Yunnan Vaccine Laboratory, took advantage of a randomized clinical trial of mRNA booster vaccines. The team selected an outcome-balanced subset of 124 participants—62 who later developed a documented COVID-19 case during follow-up and 62 who did not—matched across sex, vaccine product, and age category. Peripheral blood mononuclear cells collected immediately before vaccination (Day 0) and again 24 hours later (Day 1) were stained with a metal-conjugated antibody cocktail and analyzed by mass cytometry, a technique that detects individual metal isotopes rather than overlapping fluorescent signals, allowing researchers to measure dozens of markers on millions of single cells simultaneously.

Seventeen major immune-cell populations were manually gated from the resulting data, spanning the full spectrum of circulating immunity: classical, intermediate, and non-classical monocytes; myeloid and plasmacytoid dendritic cells; natural killer cells; gamma-delta T cells; NKT-like cells; naive, central memory, effector memory, and terminally differentiated effector memory subsets of both CD4 and CD8 T cells; and B cells. Cell abundances were transformed using a centered log-ratio (CLR) method to account for the compositional nature of blood-cell counts, and marker expression was summarized as sample-level median arcsinh-transformed intensities. Composite module scores were built from markers associated with type I interferon signaling, Th1 chemokines, pro-inflammatory cytokines, CCL2–CCR2 axis, antigen presentation, activation, and cytotoxicity, alongside individual readouts for global H3K27ac—a histone modification marking active promoters and enhancers—and m6A, the most prevalent internal modification of eukaryotic mRNA.

The results revealed a strikingly rapid, lineage-specific redistribution of circulating immune cells. Within 24 hours, classical and intermediate monocytes showed the largest positive shifts in relative abundance, consistent with mobilization of the myeloid compartment in response to vaccine-derived signals. In contrast, plasmacytoid dendritic cells, natural killer cells, and selected CD8 and unconventional T-cell populations showed prominent negative shifts. The four major CD4 T-cell populations—naive, central memory, effector memory, and terminally differentiated effector memory—remained comparatively stable on this relative-composition scale. These changes were accompanied by distinct, lineage-restricted alterations in intracellular and phenotypic markers: classical monocytes displayed prominent changes in MCP-1, arginase-1, and H3K27ac; intermediate monocytes showed changes in CCR2, CD38, and IL-4; pDCs exhibited changes in granzyme B and CXCR3; and NK cells showed marked alterations in CD69 and CD56.

Crucially, the team found that early changes in global m6A and H3K27ac signals correlated with relative-abundance changes in selected innate and T-cell populations. m6A changes correlated positively with classical and intermediate monocyte abundance but inversely with pDC and NK-cell abundance. H3K27ac changes were also inversely correlated with pDC and NK-cell abundance. Within the T-cell compartment, both m6A and H3K27ac changes were inversely related to CD4 naive T-cell abundance, while m6A changes were positively associated with CD4 effector memory abundance but inversely associated with CD8 naive and CD8 central memory abundance. These correlations suggest that global modification signals track distinct differentiation-state trajectories across the immune landscape, though the authors caution that correlation does not establish causation.

When the researchers examined whether these early cellular and molecular changes predicted subsequent neutralizing-antibody responses, the picture became more nuanced. Both groups showed broad increases in serum neutralization against a panel of SARS-CoV-2 variants—including the prototype strain, Alpha, Beta, BA.4/5, BF.7, BQ.1.1, XBB.1.5, XBB.1.16, XBB.1.9.1, and JN.1—measured longitudinally through Day 180 using a vesicular stomatitis virus-based pseudovirus neutralization assay. However, after excluding antibody measurements obtained on or after documented breakthrough infection and applying Benjamini–Hochberg correction across the complete 340-test family of cell population–marker–strain combinations, no early H3K27ac- or m6A-associated signal change remained significantly associated with Day 14 or Day 180 neutralizing-antibody responses. The strongest nominal association—an inverse relationship between early H3K27ac change in CD4 central memory cells and the BF.7 response—did not survive correction.

The most compelling finding emerged when the team compared early immune trajectories between participants who subsequently developed documented breakthrough infection and those who did not. Major-cell abundance trajectories did not differ significantly between the two groups, indicating that the broad redistribution of circulating immune populations was shared regardless of later infection status. However, a focused screen of 13 prespecified feature readouts across all 17 major immune-cell populations—comprising 221 tests—revealed five feature–cell-population combinations that remained significant after global multiple-testing correction. Four of these were concentrated within the CD4 naive T-cell compartment: the IFN–Th1 chemokine-associated, cytotoxic–Th1-associated, and CCL2–CCR2-associated marker scores, and the m6A-associated readout. The fifth involved the cytotoxic–Th1-associated score in CD4 central memory cells. Relative to non-BTI participants, individuals who subsequently developed breakthrough infection showed smaller—more negative—Day 0-to-Day 1 changes in each of these features.

The authors emphasize that these composite scores represent combinations of measured markers rather than direct functional assays. Signal intensity varied substantially among constituent markers within CD4 naive cells: granzyme B was robustly detected, while MCP-1, CXCL9, and CD57 showed measurable intermediate-level signals, but IFN-γ, CD86, and CXCR3 exhibited relatively low absolute intensities. Unsupervised clustering further resolved 11 phenotypic subclusters within the CD4 naive compartment, demonstrating substantial internal heterogeneity, though the significant associations were identified at the total population level and cannot yet be assigned to specific subclusters. A sensitivity analysis adjusting for age, sex, and vaccine product using HC3 robust standard errors confirmed the direction of the BTI-associated differences for all four CD4 naive features.

The study’s implications extend beyond the immediate findings. By demonstrating that dynamic early immune states—captured within 24 hours of vaccination—can capture outcome-associated variation not evident from major-cell abundance alone, the work supports a broader paradigm shift in systems vaccinology: rather than relying on static baseline measurements or single downstream readouts, researchers may need to characterize the full temporal trajectory of immune responses to understand why some individuals remain protected while others experience breakthrough infection. The authors note that breakthrough infection is influenced by factors beyond measured immune features, including exposure intensity, circulating viral variants, mucosal immunity, and behavioral factors, and that illness-triggered surveillance may have missed asymptomatic infections. Future studies integrating antigen-specific assays, absolute cell counts, single-cell RNA sequencing, chromatin-accessibility profiling, and transcript-resolved m6A analysis will be necessary to define the biological pathways underlying these early CD4 naive T-cell associations and to establish their clinical relevance as correlates of vaccine-response heterogeneity.

Subject of Research: Mass cytometry study of early immune remodeling after mRNA booster vaccination and its association with breakthrough infection

Article Title: Mass cytometry of early immune remodeling after mRNA booster vaccination reveals exploratory CD4 naive T-cell associations with documented breakthrough infection

Article References: Yu, X., Jiang, X., Xue, Y., Ji, Y., Li, C., Zhang, Z. S., & Cheng, R. (2026). Mass cytometry of early immune remodeling after mRNA booster vaccination reveals exploratory CD4 naive T-cell associations with documented breakthrough infection. Frontiers in Cellular and Infection Microbiology, 16, Article 1966930. https://doi.org/10.3389/fcimb.2026.1966930

Image Credits: AI Generated

DOI: 10.3389/fcimb.2026.1966930

Keywords: mRNA vaccine, mass cytometry, breakthrough infection, CD4 naive T cells, H3K27ac, m6A, Mass, cytometry, early, immune, remodeling, mRNA

The finding matters because gestational diabetes affects a substantial share of pregnancies worldwide and raises risks for both mother and child, including future type 2 diabetes. Previous observational work had already hinted that pregnant women who develop the condition tend to have lower levels of butyrate in their circulation. What remained unclear was whether this deficiency was merely a byproduct of the disease or a genuine driver of it, and if the latter, through which molecular route butyrate might act. The new study set out to answer both questions with a combination of human data, animal experiments and mechanistic work at the level of chromatin.

The clinical arm of the study was a nested case-control analysis that measured serum butyrate and active GLP-1, the glucagon-like peptide-1 hormone released by intestinal cells after meals to stimulate insulin secretion, in pregnant women. The results were striking in their consistency: women who went on to develop gestational diabetes had significantly lower levels of both butyrate and active GLP-1, and this difference was detectable not only in mid-pregnancy, when the condition typically manifests, but already in early pregnancy. That early timing is important, because it suggests that a depleted butyrate-GLP-1 axis precedes the onset of disease rather than simply following from it, raising the possibility of prediction and even prevention.

To test causality, the team turned to a mouse model of gestational diabetes induced by a high-fat diet combined with antibiotics, a regimen that disrupts the gut microbial community and mimics the metabolic disturbances seen in patients. When the animals received a probiotic supplement of Clostridium butyricum, a bacterium renowned for its butyrate production, their metabolic picture improved markedly. Hyperglycaemia eased, insulin resistance declined, and perhaps most intriguingly, the disruption of circadian glucose rhythms that accompanies gestational diabetes was partially corrected. Microbial profiling of the treated animals confirmed that the supplement enriched the community of butyrate-producing bacteria and raised systemic butyrate levels, tying the metabolic benefits directly to the microbial metabolite.

The mechanistic heart of the study lies in the colon. Butyrate has long been known to act as an inhibitor of histone deacetylases, enzymes that remove acetyl groups from histone proteins and, in doing so, keep stretches of DNA tightly packed and transcriptionally silent. When the researchers examined colon tissue from the treated mice, they found that Clostridium butyricum supplementation reduced the abundance of one particular deacetylase, HDAC3, and increased the global acetylation of histone H3 at lysine 27, a chemical mark known as H3K27ac that flags active genes and their regulatory elements. In other words, the microbial metabolite appeared to be physically opening up the genome in intestinal cells.

Which genes were being opened? To find out, the team performed integrated RNA sequencing and H3K27ac chromatin immunoprecipitation sequencing in enteroendocrine cells, the hormone-producing specialists of the gut lining, treated with sodium butyrate and with agents that modulate HDAC3. The multi-omics analysis converged on a single compelling candidate: secretagogin, or SCGN, a calcium-binding protein expressed in enteroendocrine cells. Butyrate enhanced both the transcription of SCGN and the deposition of H3K27ac at its promoter, indicating that the metabolite directly activates the gene through the same epigenetic mechanism. Because SCGN has been implicated in the regulated secretion of peptide hormones, the connection to GLP-1 was an obvious next step.

The functional experiments confirmed it. Treating enteroendocrine cells with butyrate, or knocking down HDAC3 with molecular tools, increased SCGN expression and boosted the secretion of GLP-1. Conversely, when the researchers forced cells to overexpress HDAC3, the SCGN-GLP-1 axis was suppressed, but this suppression could be reversed by co-treatment with butyrate. The epigenetic brake and the microbial metabolite were locked in a push-pull relationship: HDAC3 silences the hormone-secreting programme, and butyrate releases it. This is a textbook example of how a dietary and microbial signal can be translated into a durable change in gene expression, one that ultimately shapes whole-body metabolism.

Genetic validation in living animals provided the final, decisive test. The team used mice that carry only one functional copy of the HDAC3 gene, a condition known as haploinsufficiency, which reduces the protein’s levels without eliminating it entirely. If the benefits of Clostridium butyricum truly flow through HDAC3 inhibition, then animals with less HDAC3 to begin with should gain little from the probiotic. That is precisely what the researchers observed: in HDAC3-haploinsufficient mice, the metabolic improvements and the upregulation of SCGN driven by the bacterium were abolished. The genetic experiment closes the causal loop and rules out the possibility that butyrate works through unrelated pathways.

Taken together, the study delineates what the authors describe as a novel gut microbiota-epigenetic-hormonal axis. Gut bacteria produce butyrate; butyrate inhibits HDAC3 in the intestinal lining; HDAC3 inhibition raises H3K27ac at the SCGN promoter; SCGN activation potentiates GLP-1 secretion; and GLP-1 improves glucose homeostasis for mother and fetus. Each link in this chain was tested independently, from the human correlation through the mouse intervention to the chromatin-level biochemistry and the genetic knockout. The architecture of the pathway is elegant precisely because it is specific: rather than a diffuse anti-inflammatory effect, the metabolite acts on a defined enzyme at a defined genomic target to produce a defined hormonal outcome.

The translational implications are considerable. HDAC3 and the downstream SCGN-GLP-1 pathway now stand as candidate targets for preventing or treating gestational diabetes, whether through probiotic supplementation with butyrate-producing strains such as Clostridium butyricum, dietary strategies that favour butyrate-producing communities, or future drugs that modulate HDAC3 activity in the gut. The authors note that their work was supported by the Beijing Municipal Natural Science Foundation, the National Natural Science Foundation of China and related programmes, and the study was conducted under ethics approvals for both the human cohort and the animal experiments. As with any mouse study, the road to clinical application will require trials in pregnant women, but the early-pregnancy signature of low butyrate and low GLP-1 offers a tantalising window for intervention long before the disease declares itself. For a condition that shapes the lifelong metabolic health of two generations at once, a microbial molecule that rewrites gene expression in the gut may prove to be one of the most consequential discoveries in pregnancy metabolism in years.

Subject of Research: How the gut microbial metabolite butyrate improves gestational glucose homeostasis through HDAC3 inhibition and epigenetic activation of the SCGN/GLP-1 axis

Article Title: The gut microbiota-derived butyrate improves gestational glucose homeostasis through HDAC3 inhibition and epigenetic activation of the SCGN/GLP-1 axis

Article References: Han, W., Zhang, Y., Huang, J., Lu, Y., Yin, X., Zhao, Y., Yuan, X., Yang, R., Zhang, L., Zheng, W., & Li, G. (2026). The gut microbiota-derived butyrate improves gestational glucose homeostasis through HDAC3 inhibition and epigenetic activation of the SCGN/GLP-1 axis. Journal of Translational Medicine. https://doi.org/10.1186/s12967-026-08720-y

Image Credits: AI Generated

DOI: 10.1186/s12967-026-08720-y

Keywords: gestational diabetes, butyrate, gut microbiota, HDAC3, epigenetics, GLP-1, secretagogin, Clostridium butyricum, histone acetylation, H3K27ac, enteroendocrine cells, insulin resistance