Decoding host–microbiota communication in the gut – now we're flying!

Although we have been aware for centuries that our intestines are colonised with microbes, it is only recently that we are beginning to grasp the significance of their presence. That this is a research area growing at a tremendous pace is evidenced by searching Pubmed for the term ‘intestinal microbiome’. In 2005, such a search returns 100 manuscripts, while in 2015 it yields 2037, representing a 20‐fold increase in research activity in this field over the past decade. We now know that our gut microbiome is comprised of >1000 different species, with microbiota‐associated genes outnumbering our own by ∼100:1, arguably making us more bacterial in nature than human. Furthermore, the balance of these microbial communities is highly dynamic, depending on many factors including, age, race, diet, health status, and even the route by which we entered the world. We are also increasingly appreciating how the gut microbiota are vitally important in controlling many aspects of our physiology. For example, intestinal bacteria are implicated in setting our basal metabolic tone, educating the immune system, and even influencing central functions, such as appetite control and mood (Shreiner et al. 2015). Within the intestine they influence practically all aspects of physiology, including fluid, electrolyte and nutrient transport, motility, mucosal immunity and barrier function. Indeed, the impact of the microbiome on mammalian physiology is illustrated by observations from germ‐free mice, which under normal conditions, appear relatively normal but which, when tested, reveal distinct deficits (Lynch et al. 2015). On the basis of emerging data, one might think that while it may be our parents’ genes that make us human, it is our microbiome, through enhancing cognition, metabolism and immunity, that confers us with superhuman abilities!

The host and its microbiota are in constant communication with one another, engaged in a molecular conversation that is largely mediated by soluble factors derived from both sides. On the microbiota side, signals are sent via bioactive toxins and metabolites or through direct exchange of biochemical and genetic material with the host. From the host side, epithelial‐derived substances such as mucus, cytokines and antimicrobial peptides have an important input to the conversation. The recruitment of mucosal immune cells, enteric nerves and mesenchymal cells in response to both host and microbiota‐derived signals adds further complexity to the repertoire. Given the millions of years over which our relationship has evolved, it is perhaps not surprising that when our microbiota become sick then so too do we, and vice versa. Under such conditions, the nature of our conversation dramatically changes, as is evident in many pathological conditions, including inflammatory, cardiovascular, metabolic and psychological disorders (Walsh et al. 2014). It is clear that a better understanding of the language by which we communicate with our microbiota is a key step to understanding human health and disease.

Large scale metagenomic and metabolomic studies in mice and humans are making considerable progress into defining how different populations of intestinal bacteria are altered in disease states (Franzosa et al. 2015), but how this translates into altered host physiology is still poorly understood. Significant insights have been gained through studying the mouse microbiome in vivo, coupled to more reductionist approaches with isolated epithelial cells in vitro. From this work, it is clear that mechanisms underlying microbe–host communication are extremely complex and will require careful and rigorous interrogation at the most fundamental cellular and molecular levels in order to fully understand them. However, to facilitate such studies, new high throughput models are required that enable rapid screening of the actions of different microbes and their metabolites on epithelial cell function. In a paper published in the current issue of The Journal of Physiology, Shanbhag et al. (2017) go some way to addressing this by demonstrating the feasibility of Drosophila larval midgut as a model to study microbiota–epithelial communication in vivo.

The midgut of fruit flies has several characteristics that make it useful for studies of host–microbe interactions. As in humans, the epithelium is derived from stem cells which differentiate into a number of different cell populations. Similar to humans, Drosophilia midgut epithelium is polarised and differentially expresses transport proteins at the apical and basal poles. Several of these transport proteins have been identified and appear to provide a pathway for HCO₃ ⁻ secretion into the lumen, a process which serves to neutralise acid entering from the foregut (Shanbhag & Tripathi, 2009). Also similar to humans, the activity of this transport pathway is driven by a basolateral ATPase pump. Thus, even though the precise identities of the transport proteins involved may be different, Drosophila midgut provides a good model for basic studies into the molecular regulation of epithelial transport. It should also be noted that, in contrast to human epithelial cells, which are separated by paracellular pores, Drosophila larval enterocytes are fused together by septae. Consequently, they do not have a paracellular pathway and responses to luminal microbes can be considered to be due to direct actions on the epithelium.

Using the Drosophilia midgut model, Shanbhag et al. (2017) employ elegant microperfusion and electrophysiological approaches to demonstrate how the presence of microbiota in the lumen induces apical membrane conductances, accompanied by enhanced KCl absorption. As a further example of how Drosophilia midgut can be employed to study host–microbe interactions, the authors demonstrate that a secreted fungal toxin, destruxin, forms non‐selective channels in the basal membrane, depleting cations, and depolarising the cell. Thus, Shanbhag et al. elegantly demonstrate how luminal bacteria can fundamentally alter intestinal epithelial transport function, effects which would in turn be expected to influence important aspects of enterocyte physiology, such as acid–base balance, cell volume regulation, and transepithelial fluid and electrolyte transport.

As we attempt to gain a better understanding of how the intestinal microbiota communicates with us, models such as that described by Shanbhag and co‐workers provide useful tools for decoding the conversation. Compared to rodents, studies in Drosophilia do not require ethical approval, are inexpensive, and have the capacity for a significantly higher throughput. This is important when considering strategies to study how distinct microbial species and their metabolic products influence epithelial function. Moreover, as the authors demonstrate, Drosophilia midgut provides an in vivo tissue that is very accessible and tractable. Through microperfusion approaches, apical and basolateral bathing solutions can be easily manipulated and consequent changes in transport and barrier function readily measured with microelectrodes. Furthermore, manipulation of specific genes is more easily achieved than in mammalian models, as are studies of enterocyte function under germ‐free conditions. Thus, it seems that much can be learned from Drosophilia larval midgut about how we might communicate with our microbiome in conditions of health and disease. However, it must still be kept in mind that fly physiology is somewhat different to that of humans and that this model only serves as an indicator for the design of more advanced translational studies in mammalian systems.