Introduction Graft-versus-host disease (GVHD) is a serious complication that markedly influences the prognosis of patients undergoing hematopoietic stem cell transplantation (HSCT). The gastrointestinal (GI) tract is not only a principal target organ but also a central driver of GVHD pathogenesis. Disruption of the intestinal barrier can permit translocation of damage-associated molecular patterns (DAMPs) and pathogen-associated molecular patterns (PAMPs), potentially triggering GVHD.

Diamine oxidase (DAO), an enzyme responsible for histamine degradation, is highly expressed in the small intestine, placenta, and kidneys, with the greatest activity in the intestinal mucosa. DAO is localized at the tips of intestinal villi and functions intracellularly. Injury to the villous architecture reduces DAO synthesis and secretion, making DAO a potential biomarker of intestinal function. Circulating DAO activity correlates with mucosal DAO expression and is regarded as a useful indicator of intestinal integrity in various clinical settings.

The gut microbiota is likewise essential for maintaining intestinal homeostasis, and alterations in specific bacterial taxa have been linked to GVHD development. In this retrospective study, we investigated the relationships among GVHD onset, longitudinal changes in DAO activity and concentration, and fecal microbiota composition in HSCT recipients.

Methods Nineteen patients who underwent HSCT at Sapporo Medical University Hospital between 2018 and 2022 were retrospectively analyzed. Serum, plasma, and fecal samples were collected before and after transplantation and stored for subsequent assays. Plasma DAO activity was measured using a colorimetric assay, whereas serum DAO concentration was determined by enzyme-linked immunosorbent assay (ELISA). Fecal DNA was extracted and analyzed by next-generation sequencing. Microbiota profiling—including α- and β-diversity analyses—was performed with the QIIME 2 platform.

Results The cohort comprised 9 males and 10 females (median age, 54 years). Nine patients received myeloablative conditioning and 10 received reduced-intensity conditioning. Stem cell sources were related peripheral blood (n = 4), unrelated peripheral blood (n = 8), unrelated bone marrow (n = 4), and cord blood (n = 5). Nine patients developed grade 2–4 GVHD requiring systemic therapy: seven with grade 2 skin GVHD, two with grade 3 GI or hepatic GVHD, and one with grade 4 skin GVHD (with some overlap). GVHD occurred between day 18 and day 67 post-transplantation (median, 35 days).

Both serum DAO concentration and plasma DAO activity displayed marked inter-individual variability. To evaluate intra-individual trends, values were expressed relative to the pre-transplant baseline (set at 100%). Plasma DAO activity decreased significantly on day 14 post-transplantation (P < 0.01) and then gradually recovered. In the GVHD group, recovery by day 28 tended to be delayed compared with the non-GVHD group (P = n.s.). Conversely, serum DAO concentration was significantly higher in the GVHD group on day 28 (P < 0.05). This divergence between activity and concentration likely reflects leakage of catalytically inactive DAO protein from damaged mucosal epithelial cells, indicating subclinical injury preceding clinical GVHD.

Microbiota analysis showed that butyrate-producing bacteria were predominant before transplantation in the GVHD group but were markedly depleted thereafter, accompanied by an increase in inflammation-associated Coriobacteriaceae. Similar but less pronounced shifts were observed in the non-GVHD group. Overall microbial diversity declined in all patients after transplantation; however, a particularly severe reduction occurred in the GVHD group, characterized by loss of butyrate producers and enrichment of pro-inflammatory taxa, suggesting diminished microbiota functionality.

Conclusion Serial monitoring of plasma DAO activity and serum DAO concentration provides insight into intestinal function after HSCT. Patients who develop GVHD demonstrate pronounced loss of microbiota diversity and function, implying that intestinal dysbiosis may contribute to GVHD pathogenesis.

This content is only available as a PDF.
2025
Sign in via your Institution