Extra-intestinal regulation of the gut microbiome: The case of C. elegans TGFß/SMA signaling

Rebecca Choi, Dan Kim, Stacy Li, Meril Massot, Vivek Narayan, Samuel Slowinski, Hinrich Schulenburg, and Michael Shapira

All authors contributed equally to the writing of this chapter.

Introduction: Caenorhabditis elegans as a model for studying the holobiont

C. elegans is a bacterivore. Reproductive populations are typically found in environments rich in rotting organic matter, such as compost or fallen rotting fruit, which are also rich in fast growing bacteria (Felix and Braendle 2010). This co-existence suggests evolutionary interactions between C. elegans and various bacteria, some as food, others as persistent gut inhabitants. However, decades of culturing worms in the lab with E. coli as a sole food source—an easily cultivable and nourishing bacterium, yet poor colonizer—left a gap in our understanding of the interactions between worms and their microbes. This is now being corrected with studies characterizing the C. elegans gut microbiome, in wild isolates or in wildtype worms raised in the lab in natural-like microcosms (Berg et al. 2016a; Dirksen et al. 2016). The same advantages making C. elegans a generally useful model organism, such as its simple body plan, transparency, fast development, and short lifespan, also make it useful for microbiome research; several studies have thus demonstrated bacterial acceleration of worm development, modulation of host metabolism, and enhancement of pathogen resistance (Montalvo-Katz et al. 2013; Watson et al. 2014; Berg et al. 2016b; Dirksen et al. 2016; Samuel et al. 2016). Moreover, the ability to work with genetically homogenous populations of self-fertilizing worms facilitates the identification of gene effects on microbiome composition and function, by minimizing interindividual variation. This is starting to yield insights into the role of host genetics in shaping gut microbiome structure and function, as discussed in this chapter (Berg et al. 2016b, 2019). Furthermore, the genetic tractability of C. elegans enables facile generation of transgenic worms expressing transgenes from tissue-specific promoters, enabling the study of the role of intertissue communication in shaping and responding to the gut microbiome. In the following sections, we will examine what can be achieved in this model and how it may serve to shed light on the factors and principles that shape the gut microbiome.

The C. elegans gut microbiome and the factors that shape it

Work in the past few years has shown that C. elegans harbors a diverse yet characteristic gut microbiome, similar in worms raised in different environments (both in the wild and in natural-like microcosms), and significantly different from the respective environmental communities (Berg et al. 2016a; Dirksen et al. 2016; Zhang et al. 2017). A core gut microbiome was described, consisting of bacterial families known for their flexible metabolism and fast growth, and dominated by bacteria of the Enterobacteriaceae and Pseudomonadaceae families (Berg et al. 2016a; Zhang et al. 2017). As a rule, members of the C. elegans gut microbiome are beneficial, and those examined demonstrated contributions to infection resistance and acceleration of development (Montalvo-Katz et al. 2013; Berg et al. 2016b; Dirksen et al. 2016; Samuel et al. 2016). In some cases, underlying mechanisms for beneficial contributions are known, such as the priming of p38 MAPK-dependent immune gene expression, enabled by the commensal Pseudomonas mendocina (Montalvo- Katz et al. 2013), or the production of antifungal and antibacterial compounds by several other Pseudomonas isolates (Dirksen et al. 2016; Kissoyan et al. 2019). Other beneficial contributions may be attributed to metabolic interactions between the worm and its commensals. Whole genome sequence analysis of a representative selection of bacteria from the C. elegans gut microbiome suggested that the community as a whole can provide all the nutrients required for C. elegans growth, reproduction, and survival (Zimmermann et al. 2019). Among those, members of the Pseudomonas and Ochrobactrum genera were found to be capable of providing the full range of metabolites necessary to support worm growth, including vitamin B12, which is essential for nematode development and fertility, and could not be produced by any of the other sequenced microbiome members (Watson et al. 2014,2016; Zimmermann et al. 2019). Metabolic network analysis based on bacterial genomic data revealed specific metabolic modules that were significantly associated with bacterial ability to colonize worms and affected population growth. For example, fermentation of pyruvate to (S)-acetoin was positively correlated with bacterial ability to colonize. This fermentation pathway produces diacetyl as an intermediate, which is known to attract worms and promote feeding activity (Ryan et al. 2014). Thus, prolonged time spent on lawns producing diacetyl could increase the likelihood of colonization. Additional associations were revealed, forming hypotheses to be tested. Similar metabolic analyses performed on transcriptomic and proteomic data from worms colonized with Ochrobactrum species indicated a role for host amino acids, carbohydrates, vitamin, and folate metabolism, as well as N-glycan and serotonin/ octopamine biosynthesis, in shaping the worm interactions with microbes (Yang et al. 2019). Together, these genomic analyses suggest that metabolic interactions between the host and the microbiome are fundamental for determining microbiome composition, extending an emerging theme describing metabolic control over microbial community assembly (Goldford et al. 2018).

Other studies, utilizing compost microcosms, have shown that additional factors are important for shaping the worm gut microbiome. First, temperature-dependent host-associated processes were found to uncouple temperature-dependent changes in the worm gut microbiome from temperature-dependent changes in the environmental microbiome. This was taken to represent effects of host physiology on microbiome composition (Berg et al. 2016a). In addition, network analysis comparing gut microbiomes to the respective environmental communities indicated that interspecies interactions between microbiome members also played a role in shaping microbiome composition, with bacteria of core gut families identified as hubs for competitive interactions (Berg et al. 2016a). What underlies such physiological or competitive interactions could still be metabolic interactions, but alternatively, such interactions could be attributed to host immunity, bacterial toxins, or all of the above together.

Potentially underlying all interactions between the host and its microbiome is genetics, both of the host and of its microbes. Supporting this, a study based on 16S sequencing, which compared the composition of gut microbiomes in different C. elegans strains and related species raised in identical microcosm environments, identified significant contributions of host genetics to microbiome composition (Berg et al. 2016b). Constrained multivariate analysis estimated host genetics as responsible for 12% of the overall variation in microbiome composition. This is likely an underestimate, due to the limited phylogenetic resolution offered by 16S sequence data. Indeed, subsequent functional evaluation of gut isolates, most of which of the Enterbacter cloacae clade, demonstrated that host genetics determined the function (or lack thereof) of commensals of the same clade, otherwise indistinguishable by sequence.

The intestinal niche

Host genetics contributes to defining the intestinal niche—the structure and surface proteins of the intestinal epithelium, chemical composition, nutrient availability, and the presence of antimicrobial proteins. It can further define host behavior, affecting its interactions with the environment. The C. elegans gut is slightly acidic; luminal pH cycles between 4.4 and about 6 in a period of 45 seconds, linked to the defecation cycle, with acidification enabled by the VHA-6 proton transporter (Allman et al. 2009; Bender et al. 2013). Oxygen levels in the worm gut have not been measured, but considering the small size of the intestine, lack of apparent compartmentalization, and its frequent opening to the outside world, it is likely to be aerobic. Composition of the worm gut microbiome supports this, with more than a few obligate aerobes (Berg et al. 2016a). Mucin secretion and glycosylation are also important characteristics of the intestinal niche. In humans, heavily glycosylated mucin proteins serve as food for a subset of gut bacteria, chief among them is Akkermansia muciniphila, which makes up to 3% of the human colon microbiome. By liberating mucin-associated oligosaccharides, Akkermansia cross-feeds additional bacteria (Belzer et al. 2017; Umu et al. 2017; Van Herreweghen et al. 2018). Glycosylation patterns of mucins further diversify environments, with sugars that can be digested only by bacteria with the suitable enzymes (Etienne-Mesmin et al. 2019). Highlighting the importance of such glycosylation patterns, disruption of the fucosyltransferase gene fut2 causes extensive changes in gut microbiome composition in mice (Kashyap et al. 2013). Mucus also serves as a selective substrate for adhesion of bacteria with suitable surface proteins, potentially enhancing gut colonization by slow-growing taxa (Juge 2012; McLoughlin et al. 2016). In C. elegans electron micrographs of the intestine show what appears to be a mucus layer overlaying intestinal microvilli (Figure 8.IB). However, the proteins that make this layer are not known (although several mucin homologs are identified in the C. elegans genome). Nevertheless, glycosylation is prevalent, and glycosylation patterns, determined by different enzymes, affect localization of secreted proteins, including lectins (Maduzia et al. 2011). Lectins, further discussed below, are sugar and lipid binding proteins, which play conserved roles in host innate immunity (Vaishnava et al. 2011; Hoving et al. 2014; Casals et al. 2018), and were shown to contribute to immune protection also in C. elegans (O’Rourke et al. 2006; Irazoqui et al. 2010; Simonsen et al. 2011; Dierking et al. 2016).

Host immunity and its role in shaping the intestinal niche

Host immunity is a dominant factor among those which could shape gut microbiome structure and function. In fact, some researchers have gone as far as suggesting that immunity evolved first and foremost under selection to recognize and manage complex communities of beneficial microbes (Hedrick 2004; McFall-Ngai 2007). Innate immunity is the more conserved branch of the immune system, shared between vertebrates and invertebrates alike. Lectins represent a significant part of C. elegans innate immunity, making a large gene family (283 members), of which many are known to be expressed in the gut, and most encode secreted proteins (Pees et al. 2016). Lectins, in particular C-type lectins, are also among the genes most strongly and reproducibly induced during C. elegans responses to pathogens, but different subsets are induced by different pathogens, as well as by nonpathogenic commensals (Shapira et al. 2006; Wong et al. 2007; Pees et al. 2016; Berg et al. 2019; Yang et al. 2019). Interestingly, studies using fluorescent reporters for lectin gene expression demonstrated differential expression, with some C-type lectins preferentially localized to the posterior gut (e.g. clec-52 and clec-60) and others (clec-66) localized to both the anterior, as well as the posterior portions of the gut (Irazoqui et al. 2010;

C. elegans colonization by its Enterobacter commensal

FIGURE 8.1 C. elegans colonization by its Enterobacter commensal. (A) A scanning electron micrograph of a mature worm. (B) A transmission electron micrograph showing initial colonization of the worm intestine by E. cloacae (C) A fluorescent image showing tdTomato-expressing E. cloacae colonizing the anterior gut. One worm is outlined. Asterisks mark the head region. Images are not to scale.

Pees et al. 2016). Considering the simple cylindrical structure of the intestine and its small size, such local specializations are surprising, and their functional significance is still unknown. Lysozymes, enzymes capable of hydrolyzing the bacterial cell wall, are another conserved family of immune effectors shown to play important roles in C. elegans immune responses. At least one member of this family (ilys-3) also demonstrated differentially localized intestinal expression (Irazoqui et al. 2010). In addition, worms express antimicrobial peptides (AMPs) divided between several families. Some are conserved, such as members of the defensin gene family (named abf genes), while others are specific to worms. Caenopores are among the latter, and encode saposin-like peptides with demonstrated pore-forming antibacterial activities (Roeder et al. 2010). Most members of this family are expressed in the intestine, and several (namely spp-3, spp-5, spp-8 and spp-18) were further shown to be upregulated in worms raised in the presence of their normal commensals, as compared to standard E. coli food bacteria (Berg et al. 2019; Yang et al. 2019). In mammals, AMPs are among the fastest evolving gene families, and are thought to be at the forefront of an arms race with bacterial resistance (Peschel and Sahl 2006). As an outcome of AMP diversification in this arms race, different members of AMP families tend to have selective efficiency against different microbes (typically shown for pathogens), and different host strains demonstrate selective susceptibility to bacterial pathogens hinging on their particular AMP cocktail (Salger et al. 2016; Schmitt et al. 2016; Romoli et al. 2017). Such selectivity may affect also beneficial microbes or commensals.

The innate immune system enables rapid responses to nonself, typically based on the identification of molecular patterns associated with pathogens (PAMPs), or just microbes (MAMPs), as well as molecular patterns associated with downstream damage (DAMPs). Recognition of such molecular patterns is achieved by pattern recognition receptors (PRR), such as Toll-like. RIG-I-like, or NOD-like receptors, or the inflammasome complex. The innate immune system is suited to quickly respond to changes in microbial composition, enabling it to dynamically shape the intestinal niche, affecting both nutrient availability and antimicrobial agents (Thaiss et al. 2016). For example, TLR signaling was shown to regulate intestinal fucosylation (Pickard et al. 2014); also, the inflammasome complex, in different configurations, was shown to regulate epithelial AMP production, and its disruption caused intestinal dysbiosis (Hu et al. 2015; Levy et al. 2015). Whereas the function of the innate immune system is phylogenetically conserved, its activation mechanisms in different organisms vary, suggesting that its interactions with gut microbes may also follow distinct trajectories. For example, the first PRR to be identified, drosophila’s Toll, is activated by a peptide generated downstream to soluble PRRs (Michel et al. 2001; Gottar et al. 2002); in contrast, vertebrate Toll-like receptors (or similar PRRs) often bind MAMPs directly (Botos et al. 2011). Furthermore, work in C. elegans so far provided only circumstantial evidence for PAMP/MAMP recognition (Twumasi- Boateng and Shapira 2012), and the one Toll homolog. TOL-1, is largely dispensable for immune responses (Pujol et al. 2001; Tenor and Aballay 2008). Instead, several lines of evidence suggest that responses to DAMPs, detection of pathogens by chemosensory neurons, and a surveillance mechanism guarding conserved targets of bacterial toxins (e.g. translation, cellular respiration) may play more dominant roles in C. elegans immunity, both against pathogens and potentially also in its interactions with gut commensals (Pujol et al. 2008; Dunbar et al. 2012; Melo and Ruvkun 2012; Meisel et al. 2014; Zugasti et al. 2014; Zhang et al. 2015).

While microbial recognition in C. elegans is not fully understood, much more is known about subsequent activation of signaling cascades that coordinate intestinal immune responses to different microbes. These include the p38 MAP kinase pathway, insulin/insulin-like growth factor signaling (IIS), and TGF-(3 signaling, and their downstream transcription factor mediators, which include SKN-1, ATF-7, DAF-16, as well as ZIP-2, HLH-30, and ELT-2 (Mallo et al. 2002; Shivers et al. 2010; Hoeven et al. 2011; Visvikis et al. 2014; Block et al. 2015; Reddy et al. 2016; Tjahjono and Kirienko 2017; Lee et al. 2018). Whereas more is known about the involvement of these pathways in antipathogen responses, new research has begun to describe their roles in host-microbiome interactions. The role of TGF|3 signaling is the focus of the subsequent sections. In addition, recent work has unraveled new roles for the central intestinal regulator ELT-2 in regulating immune gene expression (as well as affecting reproduction) during interactions with nonpathogenic Ochrobactrum commensals, and further suggested involvement of the IIS pathway (Yang et al. 2016; Yang et al. 2019).

< Prev   CONTENTS   Source   Next >