Investigating the Gut Microbiome Through Diet and Spatial Biology

The human gut contains trillions of microorganisms that impact digestion, immune functions, and overall health¹. Despite major breakthroughs, we’re only beginning to understand the full extent of the microbiome’s influence on health and disease. Advances in next-generation sequencing and spatial biology have opened new windows into this complex environment, yet many questions remain. This article highlights two recent studies exploring how diet influences microbial community composition and introduces new methodologies for gut microbiome research.

Diet Influences Microbiome Compositions
Since the start of microbiome research, scientists have studied the factors that shape gut microbial composition and their effects on human health. One important area of investigation has been understanding how diet shapes the gut microbiome. A recent study from microbiome experts at the Segata Lab examined variations in gut microbiomes among vegan, vegetarian, and omnivore diets². “We wanted to see if we can find a gut microbial signature for the three different diet patterns,” explained Gloria Fackelmann, lead author and postdoctoral fellow in the Segata Lab of Computational Metagenomics at the University of Trento.

Fackelmann and colleagues analyzed data from 21,561 individuals across several multinational cohorts.
This included participants in the UK and US from the ZOE PREDICT nutritional research program^3,4, alongside two publicly accessible datasets with individuals from Italy^5,6. Using a meta-analytical approach and key tools like MetaPhlAn 4, they gained valuable insights from the different metagenomic profiles. “One of the surprising results came when we ran the machine learning to predict the diet patterns based on the gut microbes,” shared Fackelmann. With a mean AUC of 0.85, the results indicated that microbial profiles could effectively distinguish omnivores, vegetarians, and vegans with high accuracy.

One interesting finding was the identification of microbes associated with the vegan gut microbiome, some of which are commonly found in soil and plant microbiomes. Notably, specific microbes used in agriculture to enhance crop yield were also present in vegan gut samples. This observation suggests a possible link between plant-based diets and the introduction of these microbes into the gut, although it remains unclear whether they establish long-term colonization or are transient. This finding parallels a documented presence of Bifidobacteria in dairy consumers, which is potentially acquired from foods like yogurt.

The study also examined the correlation between gut microbes and the healthy Plant Dietary Index (hPDI)⁷. This index evaluates diet patterns by assigning positive scores to nutrient-dense and reverse scores to processed and animal-based foods. The researchers found that certain gut microbes are consistently linked to healthier diets, regardless of whether someone is vegan or omnivore. In particular, red meat consumption was linked to microbes associated with poorer cardiometabolic health, while plant-based diets fostered beneficial microbes supporting gut health and metabolic function.

During the analysis, the researchers also revealed significant differences in gut microbial composition. “Gut microbial diversity was highest in the omnivores, then came the vegetarians, and was lowest in the vegans,” noted Fackelmann. While higher diversity is often thought to be beneficial, it depends on the types of microbes present. Fackelmann hypothesized that excluding major food groups, particularly dairy, could reduce microbial transfer from food to the gut and contribute to lower diversity in vegans. However, this remains a hypothesis needing further study.

This diet-based microbiome work was nicely complemented by a recent publication led by Fackelmann’s colleague, Niccolò Carlino. In Carlino’s study, the team analyzed over 2,500 food metagenomes from 50 countries to examine microbial diversity and its connection to the human microbiome⁸. They identified over 10,000 metagenome-assembled genomes, including previously undescribed species. Their findings suggest that food-associated microbes contribute approximately 3% to the adult gut microbiome, with some microbes potentially transferring from food to humans. Additionally, the team curated a database named cFMD (curatedFoodMetagenomicData), which was created to enable further research on food microbiomes.

Looking ahead, Fackelmann shared that the team is working to scale up microbiome studies by enrolling more participants and examining microbial functions across diverse dietary patterns. Though briefly discussed in the paper, the next phase of research aims to determine whether gut health is more influenced by the species present or the functional genes they carry. This addresses questions that were previously difficult to explore using older methods like 16S rRNA studies. Developments in shotgun sequencing enable researchers like Fackelmann to analyze microbial genes and their potential roles.

Innovations with Spatial Multi-omics
Advancements in gut microbiome research have led to more sophisticated techniques, significantly enhancing the resolution of microbial analysis. One such innovation is Microbiome Cartography (MicroCart), a multi-omics spatial framework designed to investigate host-microbiome interactions in situ⁹. This new approach integrates spatial proteomics, transcriptomics, and glycomics from host and microbial components.

“We created this technology to enable people to quantify microbial compositional change, but at the spatial resolution,” stated Bokai Zhu, first author of the study and postdoctoral researcher at the Ragon Institute at MGH, MIT, and Harvard. While gut microbiome composition has been well studied through sequencing, its spatial organization was not as well characterized. This technology gap prompted Zhu to initiate the development of MicroCart, a project he started several years ago in Garry Nolan’s lab at Stanford. Zhu’s goal was to design a tool that would allow researchers to study bacteria and human cells to better understand their interactions in this complex environment.

Through dedicated efforts, Zhu developed MicroCart to generate high-resolution maps of tissue samples. This system integrates multiplexed imaging (MIBI), spatial sequencing (GeoMx DSP), and mass spectrometry imaging (MALDI-MSI) for precise spatial analysis. Zhu emphasized that one of MicroCart’s strengths is that this approach is straightforward and can be easily adapted for use on other technologies. Additionally, this work introduced an optimized 16S rRNA probe design pipeline that enhances specificity and efficiency in detecting bacterial species. These improved probes allow researchers to target bacteria taxa while preserving biological structures for downstream analysis.

The application of MicroCart to a murine colitis model allowed the research team to identify numerous changes in host and microbial cells. Among these findings, increased immune cell infiltration was observed in the large intestine during colitis, particularly macrophages and monocytes associated with structural changes in smooth muscle and fibroblast activity. In addition, various bacterial populations corresponded with distinct metabolic shifts. For instance, Firmicutes influence phosphatidylethanolamine and fatty acid metabolism, while Bacteroidetes and Proteobacteria contribute to lipid and cholesterol metabolism. Furthermore, immune and epithelial responses exhibited distinct spatial organization, with monocytes responding to bacterial diversity shifts and goblet cells interacting with local mucus production.

One unexpected finding was that during colitis, bacterial communities became more homogenized in localized areas rather than being spatially diverse, as seen in a healthy microbiome. This suggests that disease disrupts the spatial organization of gut bacteria, leading to distinct shifts in different locations within the colon. Zhu explained that this highlights the complexity of the microenvironment, where multiple cell types interact in a coordinated biological process. He emphasized that focusing on microbial-host interactions, rather than individual cells, could provide a more effective approach to understanding and influencing these dynamics.

While MicroCart provided valuable insights, merging these varied datasets was a complex task. According to Zhu, there were two main challenges in integrating this multi-omics data. The first was purely experimental, which involved aligning different spatial data sources from diverse technologies. Due to the technological differences, some data had to be manually aligned to ensure it corresponded to the same tissue region. The second challenge was computational. To address this, the team developed an ensemble model using random forests, enabling dynamic weighting and integration of multiple omics layers. Since protein and RNA data offer distinct insights for disease prediction, their machine-learning approach optimized the contribution of each data type. This work was built on Zhu’s broader research in multi-omics integration and utilized techniques he learned from previous studies.

As for the future of MicroCart, Zhu shared that his current work focuses on analyzing host response and microbiome spatial patterns by integrating image morphology with RNA readouts. This includes examining key tissue features, such as cytokine secretions. “My overall interest is trying to extract more information from tissue samples,” explained Zhu. To achieve this, he combines experimental innovations like MicroCart with computational algorithms he has developed. Ultimately, Zhu emphasized that his goal is to integrate these data sources for a more comprehensive understanding of tissue biology.



References

1. Shahab, M., & Shahab, N. (2022). Coevolution of the human host and gut microbiome: Metagenomics of microbiota. Cureus, 14(6), e26310. https://doi.org/10.7759/cureus.26310
2. Fackelmann, G., Manghi, P., Carlino, N., Heidrich, V., Piccinno, G., Ricci, L., et al. (2025). Gut microbiome signatures of vegan, vegetarian, and omnivore diets and associated health outcomes across 21,561 individuals. Nature Microbiology, 10(1), 41. https://doi.org/10.1038/s41564-024-01870-z
3. Asnicar, F., Berry, S. E., Valdes, A. M., Nguyen, L. H., Piccinno, G., Drew, D. A., et al. (2021). Microbiome connections with host metabolism and habitual diet from 1,098 deeply phenotyped individuals. Nature Medicine, 27(2), 321. https://doi.org/10.1038/s41591-020-01183-8
4. Berry, S. E., Valdes, A. M., Drew, D. A., Asnicar, F., Mazidi, M., Wolf, J., et al. (2020). Human postprandial responses to food and potential for precision nutrition. Nature Medicine, 26(6), 964. https://doi.org/10.1038/s41591-020-0934-0
5. Tarallo, S., Ferrero, G., De Filippis, F., Francavilla, A., Pasolli, E., Panero, V., et al. (2022). Stool microRNA profiles reflect different dietary and gut microbiome patterns in healthy individuals. Gut, 71(7), 1302. https://doi.org/10.1136/gutjnl-2021-325168
6. De Filippis, F., Pasolli, E., Tett, A., Tarallo, S., Naccarati, A., De Angelis, M., et al. (2019). Distinct genetic and functional traits of human intestinal Prevotella copri strains are associated with different habitual diets. Cell Host & Microbe, 25(3), 444. https://doi.org/10.1016/j.chom.2019.01.004
7. Satija, A., Bhupathiraju, S. N., Rimm, E. B., Spiegelman, D., Chiuve, S. E., Borgi, L., Willett, W. C., Manson, J. E., Sun, Q., & Hu, F. B. (2016). Plant-based dietary patterns and incidence of type 2 diabetes in US men and women: Results from three prospective cohort studies. PLOS Medicine, 13(6), e1002039. https://doi.org/10.1371/journal.pmed.1002039
8. Carlino, N., Blanco-Míguez, A., Punčochář, M., et al. (2024). Unexplored microbial diversity from 2,500 food products across 120 countries. Cell, 187(20), 5775–5795.e15. https://doi.org/10.1016/j.cell.2024.07.039
9. Zhu, B., Bai, Y., Yeo, Y. Y., Lu, X., Rovira-Clavé, X., Chen, H., Yeung, J., Nkosi, D., Glickman, J., Delgado-Gonzalez, A., Gerber, G. K., Angelo, M., Shalek, A. K., Nolan, G. P., & Jiang, S. (2025). A multi-omics spatial framework for host-microbiome dissection within the intestinal tissue microenvironment. Nature Communications, 16(1). https://doi.org/10.1038/s41467-025-56237-7