The human intestinal microbiota plays essential roles in our health [1]. Extensive gene exchange occurs between the different bacteria in this diverse community 2, 3, 4. Recent technological advances and large-scale genomic and metagenomic datasets are rapidly accelerating our understanding of how this phenomenon shapes the ecology and evolution of this complex community and how it affects ecosystem properties relevant to the human host.

There are many types of mobile genetic elements (MGEs) in bacteria (defined as DNA regions capable of moving within genomes and/or transferring between cells), reviewed in Ref. [5]. MGEs can vary widely in size (from less than 1 kb to larger than 1 Mb) and can be parasitic, selfish, or advantageous for their host cell. Common types include extrachromosomal plasmids, conjugative transposons (CTs, also known as integrative conjugative elements or ICEs), integrative and mobilizable elements (IMEs), bacteriophages, and insertion sequences. While some MGEs mediate their own transfer across cells (such as conjugative plasmids and CTs), others are mobilizable (which rely on several mechanisms such as conjugation by machinery encoded in co-resident MGEs) [5]. Mechanisms of cell-to-cell spread are reviewed in Ref. [6]. In addition to the backbone genes necessary for their replication (for plasmids and phages) and spread, MGEs often harbor genes encoding other functions (‘cargo genes’) that can confer a fitness advantage to cells that acquire them (Figure 1). However, MGEs also impose a fitness cost on host cells due to burdens associated with MGE replication, regulation, and gene expression, especially for genes encoding expensive public goods. MGEs can overcome this burden by several mechanisms 7, 8.

In this review, we discuss recent advances in our understanding of how MGEs shape the ecology and evolution of our intestinal microbiota, focusing on plasmids, CTs, and IMEs that carry potentially fitness-conferring functions. Parasitic MGEs in the gut, such as bacteriophages, have been reviewed in Refs. 5, 9, 10. Historically, knowledge in MGE biology (gene content, regulation, mechanisms, etc.) is heavily skewed towards proteobacteria (Pseudomonadota), which are minor members of the gut microbiota and have MGE repertoires that are vastly different from those of the dominant gut symbionts (Bacteroidales and Firmicutes/Bacillota) 11••, 12••, 13. Although some insights learned from Proteobacteria extend to MGEs in the gut microbiota, recent studies indicate that several other features are unique to the dominant gut-associated microbes, while many other aspects of MGE biology in the gut remain poorly understood. Exchange of MGEs has been shown to frequently occur in the gut microbiota, with an increasing number of examples showing that it enables rapid adaptation to dynamic environmental challenges. Experimental and computational approaches to study horizontal gene transfer (HGT) in microbial communities have been reviewed in Refs. [14] and [15]. Table 1 summarizes common methods with an emphasis on recently developed approaches 16, 17, 18••, 19•, 20, 21, 22, 23.

The future design of microbiome therapeutic interventions with predictable outcomes will require a thorough understanding of how MGEs shape ecological interactions in the gut in health and disease, and how they evolve over short timescales. Understanding these phenomena will enable us to predict how microbiome interventions will progress and whether therapeutic effects will be stable or transient. It will enable tackling questions such as: how can we control antibiotic resistance spread in a population [24]? How do we design a reliable and stable precision microbiome engineering treatment to introduce a desired trait, such as the ability to degrade a harmful metabolite [25]? Which gene functions from non-industrialized microbiomes should we reintroduce to industrialized microbiomes to promote a desired outcome, such as decreased inflammation [1]? How do we design the best consortium to recolonize a patient’s microbiome leading to a predictable and stable outcome, for example, during cancer immunotherapy treatment [26]?