The team collected homogenized oolong tea waste from a local processing facility, blending discarded material from two Phoenix Dancong and two Ya Shi Xiang varieties. To standardize the substrate, they brewed the waste in boiling ultrapure water to simulate industrial tea extraction, then loaded 1.9 kilograms of the prepared dregs into each of three insulated, aerated laboratory reactors. Crucially, no external microbial culture was introduced at any point; decomposition relied entirely on the native consortium already living in the fresh tea dregs. The researchers maintained moisture at roughly 90 percent, mixed the substrate three times daily, and monitored temperature, pH and moisture at multiple time points, while strict sterile-handling and swab tests confirmed that cross-contamination never occurred.

What emerged from the sequencing data was a tightly choreographed ecological drama in three acts. In the first, a mesophilic colonization phase spanning days 0 to 7, the bacterial phylum Pseudomonadota exploded from 17.4 percent to a striking 93.4 percent relative abundance by day 3. Genera such as Enterobacter, Pantoea and Erwinia dominated, all fast-growing generalists equipped to feast on the monosaccharides and organic acids abundant in freshly processed tea waste. The community showed high activity in glycolysis and the pentose phosphate pathway, and carried an abundance of genes encoding pectinases of the GH28 family, ideal for opening the pectin-rich walls of young plant tissue. A visible signature of this metabolic burst was a viscous mucus coating on the leaves, evidence of extracellular polymeric substances and biofilm formation that help retain enzymes and create microenvironments for cooperative degradation.

As the easy carbon was consumed, the system heated up. Between days 8 and 15, internal temperatures climbed from the ambient 22 degrees Celsius to a peak of 30 to 33 degrees, a modest but ecologically meaningful rise the authors call a thermotolerant transition. Pseudomonadota’s grip loosened to about 75 percent, while heat-tolerant Actinomycetota and Bacillota rose together to 20 percent of the community by day 14. The fungal phylum Ascomycota, barely detectable earlier, climbed to roughly 2.5 percent abundance, coinciding with visible white mold colonies. Although fungi remained numerically minor, the researchers caution that their true contribution may be underestimated, since fungal genomes of 30 to 40 million base pairs are far larger than bacterial ones and are underrepresented in metagenomic read counts.

Functionally, this second phase marked a pivotal shift from simple to complex substrates. The team recorded a 2.3-fold enrichment in glycoside hydrolases, particularly GH5 cellulases and GH13 amylases, indicating that the community was pivoting toward cellulose and starch as soluble sugars dwindled. Analysis against the Pathogen-Host Interactions database revealed a suite of genes typically labeled virulence factors in plant pathogens such as Magnaporthe and Fusarium. In a composting context, the authors argue, these genes likely encode plant cell-wall degrading enzymes like cutinases, cellulases and laccases rather than true virulence traits, a saprophytic toolkit for attacking recalcitrant polymers. In other words, fungi may act as catalysts for the initial breach of lignocellulose, softening structures that bacteria then process downstream.

The final act, a maturation phase from days 16 to 35, brought the deepest reorganization. Pseudomonadota fell to 45.2 percent, while Actinomycetota and Bacteroidota together rose to 27.8 percent by day 35. The fungal signal faded below 1 percent, suggesting that fungi do most of their work during the thermotolerant window. Taking center stage were specialized lignocellulose degraders, notably Cellulomonas hominis and Microbacterium galbinum, whose genomic signatures featured hemicellulases of the GH43 family and carbohydrate-binding modules CBM48 and CBM13. Microbial diversity peaked in this phase, with the Shannon index climbing steadily and a dense co-occurrence network of 33 genera and 275 mostly positive associations forming around the Cellulomonas and Microbacterium hubs, a topology consistent with cooperative cross-feeding in which primary cellulose degraders release oligosaccharides that secondary consumers ferment.

Metabolite profiling tied the ecology to chemistry. Early-dominant Pseudomonadota correlated strongly with labile compounds such as glucose and gallic acid, while later Actinomycetota and Bacteroidota tracked lignin-derived phenolics and complex flavonoid glycosides. Ascomycota correlated with polyphenolic intermediates like chicoric acid derivatives during the transition phase, aligning with a proposed role in modifying aromatic structures. Core-pan genome analysis added a genomic dimension: the pan-genome expanded 2.8-fold from day 0 to day 35 while the shared core genome contracted by 61 percent, charting a shift from a community of generalist opportunists to a consortium of substrate-specific specialists, with genes for motility and signal transduction declining as competition eased.

The headline discovery, however, is the system’s apparent self-purification. By day 35, antibiotic resistance genes had declined by 78 percent, and mobile genetic elements had all but vanished. Virulence-associated genes from both bacterial and fungal sources, prominent in the early phases, dropped to minimal levels. The authors propose two plausible mechanisms: competitive exclusion, as ARG-harboring Enterobacteriaceae fell from 78 percent to 12 percent relative abundance, and abiotic degradation of extracellular DNA under the oxidative stress of prolonged aerobic decomposition. The biosafety implications are considerable, because applying compost to farmland can otherwise seed soil microbiomes with resistance genes, a growing concern in the global spread of antibiotic resistance.

The researchers are careful to frame their findings as genomic potential rather than proven activity, since metagenomics identifies which genes are present, not whether they are being expressed. They also note limitations: the small laboratory reactors cannot fully replicate industrial composting piles of one to five tons, fungal taxonomy requires deeper resolution through ITS sequencing and metatranscriptomics, and the network correlations do not establish direct metabolic interactions. Pilot-scale validation of at least 100 kilograms, the team says, is essential before the results translate to industry. Still, the reproducible succession pattern offers a practical roadmap. Early Pseudomonadota dominance could serve as a bioindicator of efficient process initiation, while the documented cross-kingdom dynamics suggest that bioaugmentation strategies should combine bacterial and fungal lignocellulose degraders. If the self-purification trend holds at scale, oolong tea waste, one of agriculture’s most stubborn residues, could become a low-cost, biosafe feedstock for a circular economy, degraded and detoxified by the very microbes it carries.

Subject of Research: Microbial succession and detoxification during aerobic degradation of oolong tea waste

Article Title: Structured succession of a cross-kingdom microbiome drives degradation and detoxification of oolong tea waste

Article References: Ye, G., Yu, X., Ismaiah, M. J., Ke, R., Zeng, Z., Wang, J., Lan, S., Yu, H., Liu, H., Xie, S., Leung, K. S., Zhang, L., Lee, J. C.-Y., & Habimana, O. (2026). Structured succession of a cross-kingdom microbiome drives degradation and detoxification of oolong tea waste. Frontiers in Microbiomes, 5, Article 1871476. https://doi.org/10.3389/frmbi.2026.1871476

Image Credits: AI Generated

DOI: 10.3389/frmbi.2026.1871476

Keywords: oolong tea waste, microbial succession, metagenomics, lignocellulose degradation, antibiotic resistance genes, composting, Pseudomonadota, Ascomycota, Cellulomonas, carbohydrate-active enzymes, waste valorization, circular economy