Single bacterial strain recruits microbial allies to break down tobacco leaf polysaccharides over 36 months

When tobacco leaves are harvested, they are far from finished. Before they can be manufactured into a usable product, they must undergo a prolonged period of aging, during which a hidden workforce of microorganisms gradually dismantles the large, stubborn molecules locked inside the leaf. A new study published in Advanced Biotechnology has now traced this microbial demolition project in unprecedented detail, following treated and untreated leaves for a full 36 months and revealing how a single added bacterium, Microbacterium testaceum, can orchestrate the entire process by recruiting other key species to the task.

The research, led by Yichao Hu and Yuwen Wang with colleagues at Hunan Agricultural University and partner institutions in China, focused on four polysaccharides that dominate the chemistry of cured tobacco: starch, pectin, cellulose, and hemicellulose. These macromolecules matter enormously for leaf quality. Starch that survives aging produces acrolein when the leaf burns, generating an unpleasant withering odor. Excess pectin releases irritating methanol at high temperatures and forms hydrophilic colloids that alter the leaf’s physical behavior. Undegraded cellulose and hemicellulose produce paste-like flakes during combustion, disrupting how evenly the leaf burns and reducing its flexibility and filling power.

Yet polysaccharides are not simply enemies of quality. Their breakdown products are precursors of aromatic compounds that give aged tobacco its desirable character. Fiber-derived disaccharides can be oxidized into vanillin, while galacturonic acid from pectin can be converted into furanone. The challenge, therefore, is not to eliminate polysaccharides but to tune their content into an optimal range, allowing the beneficial transformations while avoiding the sensory and chemical defects caused by residues. Effective regulation of this balance is considered one of the key routes to improving raw tobacco quality.

To test whether a functional microbe could accelerate this tuning, the team turned to Microbacterium testaceum No. 2, a strain previously isolated from B2F-grade tobacco in Liangshan, Sichuan Province. Laboratory screening had shown that this strain produces four classes of polysaccharide-degrading enzymes: amylase, pectinase, cellulase, and hemicellulase. The researchers grew the bacterium in liquid culture to produce a whole-cell fermentation broth, then sprayed it onto single-varietal K326 tobacco leaves at a ratio of 4 milliliters per 100 grams of leaf. A control group received sterile water under identical conditions. Both groups were then stored in the cave storage facility of the Changsha Cigarette Factory in Hunan, where natural aging proceeded under stable conditions averaging 18.6 degrees Celsius and 60 percent relative humidity.

The sampling design was unusually ambitious for this kind of study. Rather than tracking quality for a few weeks, the team collected material at 3 hours, 2 months, 6 months, and 36 months after treatment, drawing cores from the centers of 200-kilogram compressed bales at standardized heights, with six replicates per treatment at every time point, for a total of 54 samples. Chemical analysis by continuous flow analysis and gas chromatography tracked polysaccharides, sugars, nitrogen, nicotine, potassium, and chloride, while a second portion of each sample was frozen at minus 80 degrees Celsius for DNA extraction and high-throughput sequencing of the bacterial 16S rRNA gene on an Illumina NovaSeq platform.

The chemical results were striking. Compared with the sterile-water controls, the Microbacterium-treated leaves showed substantially enhanced degradation of all four polysaccharides: pectin fell by 22.77 percent more, starch by 30.61 percent more, cellulose by 21.39 percent more, and hemicellulose by 17.35 percent more over the aging period. The timing of degradation differed by substrate. Starch declined fastest during the 2-to-6-month window, reaching a degradation rate of 17.4 percent, whereas pectin degraded most rapidly later in aging, with a peak rate of 45.46 percent between 6 and 36 months. Meanwhile, total sugars and potassium declined over time, while total nitrogen, reducing sugars, nicotine, and the derived nitrogen-to-alkali and potassium-to-chloride ratios remained statistically unchanged.

Sequencing revealed that the added strain did far more than contribute its own enzymes. Bacterial richness and diversity, measured by Chao1, ACE, Shannon, and Simpson indices, peaked at 2 months and then declined with further aging, but remained consistently higher in the treated leaves than in controls at 2, 6, and 36 months. Principal coordinate analysis showed that the community structure of treated and control leaves diverged significantly at every sampling point, from 3 hours onward. At the phylum level, Proteobacteria, Firmicutes, and Bacteroidetes together accounted for more than 85 percent of the community, and Proteobacteria became markedly more abundant in treated leaves, particularly during the later stages of aging. Genera such as Pseudomonas and Pantoea dominated early, while Ralstonia and Proteus rose to prominence between 6 and 36 months.

To understand the architecture behind these shifts, the researchers constructed molecular ecological networks using random matrix theory, classifying nodes by their within-module and among-module connectivity to identify keystone taxa. The networks of both groups simplified as aging progressed, but the treated leaves consistently maintained more nodes and higher average clustering coefficients than controls at 2, 6, and 36 months, indicating a denser, more complex web of bacterial interactions. Negative correlations, which signal competitive relationships, rose from 45.06 percent in the untreated baseline to between 49 and 69 percent in treated samples, suggesting that the added strain intensified competition for resources. Most tellingly, the treated networks contained 32 keystone operational taxonomic units against only 21 in the controls, and several of these keystone taxa, including members of Delftia and Proteus, showed significant negative correlations with starch and cellulose content, implying that their proliferation directly promoted polysaccharide breakdown.

Random forest modeling reinforced this picture, identifying Streptomyces, Delftia, Romboutsia, and Sphingomonas as significant predictors of starch content, and Staphylococcus, Steroidobacter, Proteus, and Agathobacter as predictors of pectin. The authors propose that Microbacterium testaceum acts less as a solo degrader and more as an ecological architect: by altering the leaf’s microenvironment, it recruits and empowers a consortium of keystone taxa whose combined metabolic activity dismantles macromolecules far more efficiently than any single strain could. The researchers caution that the precise signaling mechanisms behind this recruitment remain unknown, and they plan metabolomic comparisons of treated and untreated leaves to identify the key mediator molecules. If those signals can be identified and harnessed, the strategy could extend beyond tobacco, offering a template for steering microbial communities in any aging or fermentation process where breaking down recalcitrant polysaccharides is the goal.

Subject of Research: Microbial community regulation of polysaccharide degradation during post-harvest tobacco leaf aging

Article Title: Microbacterium testaceum facilitates polysaccharide decomposition during post-harvest aging of tobacco leaves by recruiting keystone bacterial taxa

Article References: Hu, Y., Wang, Y., Qin, T., Chen, W., Ma, T., Lei, J., Fu, Q., Feng, X., Han, Z., & Li, J. (2025). Microbacterium testaceum facilitates polysaccharide decomposition during post-harvest aging of tobacco leaves by recruiting keystone bacterial taxa. Advanced Biotechnology, 3(4), Article 32. https://doi.org/10.1007/s44307-025-00086-4

Image Credits: AI Generated

DOI: 10.1007/s44307-025-00086-4

Keywords: Microbacterium testaceum, tobacco aging, polysaccharide degradation, keystone taxa, microbial ecology, Delftia, Proteus, molecular ecological networks, 16S rRNA sequencing, functional strains, pectin, starch