A systematic, complexity-reduction approach to dissect the kombucha tea microbiome

  1. Xiaoning Huang
  2. Yongping Xin
  3. Ting Lu

  • Curated by eLife

    eLife logo

    Evaluation Summary:

    This work will be of interest for researchers studying the functions of microbial communities, microbial ecology and interactions. Using the Kombucha tea (KT) microbiome as a case study, Huang et al. provide a framework for simplifying complex communities into core communities that capture aspects of complex communities. Authors demonstrated that core communities can facilitate a mechanistic understanding of how microbes interact, especially when member species are individually culturable. The work presents a fresh, novel approach for the coarse-grained analysis of complex microbiomes.

    (This preprint has been reviewed by eLife. We include the public reviews from the reviewers here; the authors also receive private feedback with suggested changes to the manuscript. Reviewer #1 agreed to share their name with the authors.)

Reviewed by

Read the full article See related articles

Listed in

Log in to save this article

Abstract

One defining goal of microbiome research is to uncover mechanistic causation that dictates the emergence of structural and functional traits of microbiomes. However, the extraordinary degree of ecosystem complexity has hampered the realization of the goal. Here, we developed a systematic, complexity-reducing strategy to mechanistically elucidate the compositional and metabolic characteristics of microbiome by using the kombucha tea microbiome as an example. The strategy centered around a two-species core that was abstracted from but recapitulated the native counterpart. The core was convergent in its composition, coordinated on temporal metabolic patterns, and capable for pellicle formation. Controlled fermentations uncovered the drivers of these characteristics, which were also demonstrated translatable to provide insights into the properties of communities with increased complexity and altered conditions. This work unravels the pattern and process underlying the kombucha tea microbiome, providing a potential conceptual framework for mechanistic investigation of microbiome behaviors.

Article activity feed

  1. Version published to 10.7554/elife.76401 on eLife
  2. eLife

    Evaluation Summary:

    This work will be of interest for researchers studying the functions of microbial communities, microbial ecology and interactions. Using the Kombucha tea (KT) microbiome as a case study, Huang et al. provide a framework for simplifying complex communities into core communities that capture aspects of complex communities. Authors demonstrated that core communities can facilitate a mechanistic understanding of how microbes interact, especially when member species are individually culturable. The work presents a fresh, novel approach for the coarse-grained analysis of complex microbiomes.

    (This preprint has been reviewed by eLife. We include the public reviews from the reviewers here; the authors also receive private feedback with suggested changes to the manuscript. Reviewer #1 agreed to share their name with the authors.)

  3. eLife

    Reviewer #1 (Public Review):

    The authors identified a core community consisting of one bacterium species and one yeast species. The core community qualitatively captured properties of complex communities, such as population composition, metabolite profiles, and capability of pellicle formation, even in altered conditions. The work is interesting and provides useful guidance for future work.

    The authors selected 5 bacterial and 5 fungal species, and made 25 2-species communities. Strikingly, in most binary cocultures, bacteria and yeast coexisted. B2 (Komagataeibacter intermedius) combined with any of the five yeasts could form pellicles. The authors then focused on B2Y1 (Y1 = Brettanomyces bruxellensis, the most predominant yeast species in the native samples). The two species stably coexisted in both broth and pellicle, although at a steady state, bacteria dominated pellicles while yeast dominated the broth. Despite divergent initial ratios, chemical dynamics were remarkably consistent. Using monoculture fermentation, the authors deduced inter-species metabolic interactions: Y1 consumes sucrose, generating glucose (which is consumed by both yeast and bacteria) and fructose (which is consumed by yeast but not bacteria). Y1 also generates ethanol. Bacteria in the presence of glucose and ethanol or acetate forms pellicles. Although B2 alone cannot use ethanol, with glucose, ethanol is used and acetate is produced. When culturing the five yeast and the five bacterial species together, overall patterns are similar to the pattern of the core community, although differences do exist (in part due to high variability in yeast digesting sugars and in bacterial conversion of ethanol to acetate).

  4. eLife

    Reviewer #2 (Public Review):

    This study demonstrated a unique approach to reduce the complexity when analyzing the dynamics and function of microbial communities, by using the kombucha tea microbiome as a model system. The approach is highly streamlined and consists of several major steps.

    1. Analysis of the composition of the original (complex) microbiome to identify dominant species (bacteria or yeast).
    2. Definition of target metabolic profiles as the surrogate of microbiome function.
    3. According to the analysis in #1, construct pair-wise (bacteria + yeast) communities to identify the pairs that most closely recapitulate the target metabolic profiles defined in #2. The top candidate is considered as the potential core microbiome of the original microbiome.
    4. In-depth analysis of the top candidate in terms of both mechanistic underpinnings and its functions.
    5. Evaluating the robustness of the core microbiome in response to perturbations (growth conditions and introduction of other members in the original microbiome).

    At the conceptual level, finding the appropriate level of abstraction is critical for dissecting the dynamics of complex microbiomes, given their high dimensionality. To this end, the work presents a fresh, novel approach for the coarse-grained analysis of complex microbiomes. The approach allows systematic dissection of the target microbiome to establish its core microbiome, which well captures the function of the original microbiome. It is quite remarkable that this approach works so well (at least for the KT microbiome).

  5. eLife

    Reviewer #3 (Public Review):

    This study provides a great example of dissecting the function of complex microbiota. Huang et al. developed an approach to reduce the complexity of a natural microbial community, the resulting minimal core community is amenable to mechanistic studies while still retaining key structural and biochemical features of the original community. The authors use kombucha tea (KT) microbial community to i) characterize several KTs with respect to their microbial composition and physicochemical profile, ii) define the most prominent community functions (pellicle formation, fermentation of sucrose with formation of ethanol, acetate and other metabolites), and iii) select and study bacteria-yeast subcommunities that best capture functions of original KT.

    The study is thorough, systematic and well-executed. Conclusions are generally supported by the data. The presented approach can serve as a blueprint and step-by-step manual for extracting mechanistic insights from a complex system. This work is of high quality, is clearly presented and will benefit the scientific community with practical ideas. The value of this work is not only in its approach but also in the insights it provides into the connection between metabolic function and growth dynamics of individual species of KT culture. That said, there are also a few minor limitations.

    First, this approach is particularly well suited for 'in vitro' communities, such as fermented food, where the metabolic input and output relatively clearly define the function of the community. However, the functions of most communities are defined by their interaction within their natural hosts and ecosystems and are not easily defined or quantified, even with regard to major nutrient sources and metabolic products. Therefore, applying this methodology inherently imposes a narrow choice of representative community function, which will bias the core selection and conclusions. While not at all preventing important scientific discoveries, it may be prudent to take this bias into account.

    Second (and I understand that this may be a feasibility issue), the authors infer contributions of 1 and 2 species communities to 10 species community behavior. However, the power of these interpretations is limited by the lack of quantified species-level compositional dynamics of the 10 species community (amounts and growth of individual yeast and bacteria species are not clear). In addition, emergent community properties (which cannot be inferred from analyzing subcommunities in isolation), could have been missed because of qualitative rather than quantitative approach of comparisons between communities of different complexities.

    I want to stress that these limitations are not taking away from the utility of the study, but are mentioned here to be considered when extrapolating the results obtained with simple communities to the original consortia.

  6. Version published to 10.1101/2022.01.12.475982v1 on bioRxiv