University of Groningen
Understanding the gut ecosystem: bugs, drugs & diseases
Vich Vila, Arnau
DOI:
10.33612/diss.102587978
IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite from it. Please check the document version below.
Document Version
Publisher's PDF, also known as Version of record
Publication date: 2019
Link to publication in University of Groningen/UMCG research database
Citation for published version (APA):
Vich Vila, A. (2019). Understanding the gut ecosystem: bugs, drugs & diseases. University of Groningen. https://doi.org/10.33612/diss.102587978
Copyright
Other than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).
Take-down policy
If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim.
Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons the number of authors shown on this cover page is limited to 10 maximum.
General discussion
The microbiome field is following a similar trajectory to that seen in the field of hu-man genetics during the “genome-wide association studies (GWAS) era”. This trend is marked by a large number of early discoveries and a subsequent bloom in reported condition–trait associations, followed by a rapid increase in cohort sizes, methodological standardization, functional validations and, in some cases, clinical implementation of the discoveries1. In the microbiome field, however, it looks like that lessons learnt from
GWAS are accelerating the transitions between each of these stages and preventing some of the pitfalls experienced in the past.
At the time of writing this thesis, the field of human gut microbiome research is be-ing boosted by the availability of larger datasets, which provide great resources for the expansion of our current knowledge and for the discovery of new microbial species and features. Due the microbiome’s great clinical potential, as discussed later in this chapter, there is a growing industry for the development of treatments targeting the gut microbiota. However, we should not overestimate our knowledge of the microbiome: although great advances have been made in the characterization of the gut microbiota, and its impact on the human health, more efforts are still needed to understand the gut ecosystem and how to modulate it.
Following the analogy of the GWAS era, the work presented in this thesis was accom-plished in the transition between the “blooming associations” phase and the “increase in cohort sizes” phase. Our work provides a detailed characterization of the inflammatory bowel disease (IBD) gut microbiota, identifying phenotypic characteristics associated to specific microbial signatures and linking the usage of common medication to changes in the gut microbiome composition.
Discussion
Over the past 20 years, major advances have been made in understanding the physiopa-thology of IBD. The use of different layers of “omics” data have helped to identify impor-tant mechanisms involved in the disease, revealing new potential targets for the treatment and/or prevention of IBD2–4. While in the first decade of the 2000’s efforts were
concen-trated on characterizing the genetic architecture of the disease, in the recent years the avail-ability of bigger cohorts and improvements in sequencing techniques have helped with fine-mapping and led to the discovery of new genetic regions implicated in the disease5,6.
The identification of multiple variants in genes involved in recognition of and interaction with the microbiota7–9, together with evidence from mouse models that administration of
antibiotics and germ-free conditions reduce the incidence of colitis10,11, have shifted the
IBD research community’s focus onto the study of the role of the gut microbiota. As a part of the initial efforts to understand the human gut microbiota and its dis-ease-associated changes, in chapter 2 we describe the gut dysbiosis in patients with IBD. Although the taxonomic resolution that can be achieved by 16S sequencing is limited, changes in the gut microbial community are evident at family and genus level. Consistent with previous studies8,12, we show that the IBD microbial signature is characterized by
de-creased microbial richness together with a depletion of anaerobic species and the inde-creased abundance of facultative anaerobic bacteria. Increased sample size in combination with the opportunity to perform metagenomic sequencing allowed us to pinpoint the relevant bacterial species in gut dysbiosis (chapter 3). We concluded that the gut ecosystem in patients with IBD shows a loss of microbial species, for example decreased abundance of Faecalibacterium prausnitzii, Roseburia hominis and Bifidobacterium spp13. On the other
hand, bacteria such Escherichia coli and Fusobacterium species are enriched in IBD14,15.
We have also demonstrated that the gut ecosystem in IBD shows changes in its met-abolic potential. Whereas the metabolism of carbohydrates and the biosynthesis of ami-no-acids is depleted in IBD, microbial pathways involved in oxidative stress are enriched8
(chapter 3). As a consequence of the reduction of certain species involved in fibre fermen-tation, the production of short-chain fatty acids in the gut is lower in patients with IBD16.
Recently, untargeted analysis of faecal metabolomics profiles in patients with IBD revealed that dysbiosis can also be seen at metabolic level17. Similar to the patterns observed at the
taxonomic level, a decrease in metabolite richness and the differential abundances of cer-tain molecules are the main distinctive features in IBD patients compared to controls. For example, the increase of primary bile acid metabolites in the faecal samples is indicative of changes in bile acid transformation, a process known to be mediated by the gut microbio-ta18 and found to be altered in patients with IBD (figure 1).
The microbial signature in IBD
In the next sections I put our work into perspective, point out its limitations and discuss the directions that, in my opinion, the field should follow in order to translate our knowl-edge of the gut ecosystem into the design of microbial-based therapeutic tools for treating complex diseases such as IBD.
Fig 1.
Faecal microbiota in patients with IBD.
Figure showing a representation of the factors infl uencing the gut microbiota composition and a summary of the main characteristics observed in the faecal samples from patients with IBD in comparison to non-IBD samples.
Despite the fact that microbial features allow us to distinguish patients with IBD from the general population, the gut microbial composition of IBD patients shows large in-ter-individual variation. This variation cannot solely be explained by the two common IBD subtypes, ulcerative colitis (UC) and Crohn’s disease (CD). In fact, our research team and others, have reported that UC and CD show a substantial overlap in the associated host genetics signatures5,25,26 as well as in terms of gut dysbiosis (chapters 2 and 3).
We observed that a significant proportion of this variation (~3%) is driven by the disease location. The microbial profiles of patients with colonic CD show similarities with the microbiota of patients with UC. In contrast, patients with inflammation in the ileum or with an intestinal resection affecting the ileocecal valve show the most extreme dysbiotic profile (chapters 2, 3 and 5). This pattern has also been described in a longitudinal study of 109 patients with IBD and 9 controls21. Although the number of participants was limited,
Halfvarson and colleagues studied the microbial variation over time and compared it to the standard/normal variation of non-IBD participants. Concordant with our observations, patients with resection in the ileum showed the most dysbiotic profile characterized by the largest variation in microbial composition over time. This may be explained by the role of the intestinal ileum in bile acid reabsorption: inflammation or resection of the ileum can result in an increased concentration of bile acids in the colon, which then induces changes in the gut ecosystem and in the intestinal transit time27. In chapter 5 we report that this
group of patients show a significant decrease in the absolute number of bacteria in their faecal samples. Differences in microbial densities of samples are correlated with the relative abundance of several bacterial species and can therefore have a confounding effect on the discovery of microbe–trait associations.
Disease severity and inflammatory activity in the gut, usually reported as disease scores or biomarker levels, are also associated with specific microbial signatures. Moreover, as
Patients with IBD show a heterogeneous microbial profile
Taxonomic, functional and metabolic features associated with IBD provide both a better understanding of the role of microbes in the disease and a resource in identifying new tential therapeutic targets. As described in chapter 3, distinct microbiome features can po-tentially be used to assist in the diagnosis of the diseases. Faecal taxonomic profiles, either deriving from metagenomic or 16S sequencing data, can discriminate patients with IBD from the general population and, more importantly, from people with other disorders pre-senting similar gastrointestinal symptoms such as irritable bowel syndrome12,19–22.
More-over, recent studies have shown that faecal metabolites have great predictive power17,23,24.
It is important to note, however, that most of these predictive tools have been validated in established IBD cohorts, and as a consequence their accuracy is probably overestimat-ed. Therefore, tools aiming to assist in the diagnosis or prediction of IBD should also be tested in new-onset and treatment naive patients and validated in longitudinal population cohorts.
Understanding the link between host genetics and the gut microbiota is a promising key for unlocking several disease mechanisms. In chapter 2 we observed that non-IBD indi-viduals with a higher genetic risk for IBD show a lower abundance of Roseburia species. This finding suggests that host genetics can shape microbial composition. Roseburia is generally considered a beneficial bacterium due to its capacity for butyrate production, and
Roseburia and butyrate depletion is observed in many host disorders. Therefore, genetic risk
transmitted through a decrease of beneficial bacteria can serve as an early marker for the increased risk of developing the disease. To characterize potential host–microbiota inter-actions in more detail, in chapter 4 we made use of sequencing technologies that provide better resolution data for the host genetics (exome-sequencing) and of the gut microbiota (shotgun metagenomic sequencing). The combination of these two data types resulted in the identification of several novel genetic variants associated with the abundance of bacte-rial species and microbial pathways. Most of the correlations found were in IBD risk loci and genes involved in the immune response.
The study of host–microbiota interaction has so far led to the discovery of a limited number of mechanisms and is characterized by low replication rates between studies30–36.
This can be explained by several factors: 1) the lack of gold-standards in the microbiome field, which complicates comparisons between studies; 2) the multidimensionality of the data, where the association of thousands of genetic polymorphisms with hundreds of mi-crobial features inflates the number of potential false positives; 3) the mimi-crobial inter-in-dividual variation driven by multiple environmental confounding factors; 4) the relatively small effect size of the observations and 5) the relatively small size of the cohorts.
Despite this, there are associations that are consistent over cohorts and methods. The most consistent association is that between polymorphisms in the LCT gene and the abun-dance of Bifidobacterium species. The large effect size of this association is probably driven by the potential of Bifidobacterium to metabolize lactose in the context of aberrant lactase enzymatic activity in the host. In the context of IBD, genetic variants in the NOD2 gene have been linked to an increased abundance of bacteria belonging to the Enterobacteriace-ae family37 and a suggestive increase in the enterobactin biosynthetic pathway31. In mice,
NOD2-deficient animals show intestinal dysbiosis and induction of colitis. Moreover, a
recent study combining different IBD cohorts (using the cohort presented in chapter 2 as
Exploring host-microbiome interactions
also shown in the study of population cohorts28,29, medication use and common lifestyle
factors can also contribute to the microbial inter-individual variation in patients with IBD (chapters 2 and 3).
Overall, the heterogeneity observed in the gut microbiota of patients with IBD is not limited to the classical phenotypes (UC or CD), therefore in-depth phenotypic charac-terization is needed to capture the whole clinical spectrum of IBD and identify potential confounders in the study of the human microbiome.
Studies of cross-sectional cohorts, such as those presented in this thesis, have been insuf-ficient to determine the exact role of the microbiome in IBD. Therefore, the “chicken-or-egg” question remains: Are changes in the gut microbiota cause for or consequence of the disease?
One of the most accepted hypotheses on the aetiology of IBD is that an environmental disruption triggers an impaired immune response to the microbiota in individuals with genetic susceptibility, and this disrupts the gut homeostasis and results in an inflammatory response2,40,41. The expansion of facultative anaerobe species, together with the decreased
abundance of obligate anaerobic bacteria, could indicate that the oxidative environment generated in inflammatory conditions is one of the main drivers of the dysbiosis. The presence of reactive oxygen species in addition to the immune response of the host may act as an ecological pressure that reshapes gut microbial species composition. Consistent with this hypothesis, in chapter 3 we describe that the changes in the gut microbiota are not limited to its composition but are also present in the intra-species richness (or strain richness) and the increased abundance of certain virulence factors. The in-depth charac-terization of bacterial species enriched in the context of IBD, e.g. E. coli and R. gnavus, have shown an enrichment of strains carrying genes that confer adaptive advantage in inflammatory conditions15,42.
This may suggest that dysbiosis is merely a consequence of the diseases. However, is there also evidence that microbes are the triggering factor of the diseases? Several studies have reported that gut microbes can exhibit both anti-inflammatory and pro-inflammatory effects in the host3. Enrichment of pathobionts has been described as a characteristic of
IBD. Nonetheless, the causal relation of a single species provoking the disease in humans has not been proven. However, the inoculation of certain intestinal strains derived from patients with IBD has been shown to induce colitis in mice models43. Other mechanisms,
Microbial changes and IBD: cause or consequence?
validation) showed that variants in NOD2 were associated with a decrease in Roseburia ge-nus and Faecalibacterium prausnitzii abundances38. Interestingly, besides the discrepancies
between studies, NOD2-microbiota associations seem to capture different characteristics of a dysbiotic microbial ecosystem.
I anticipate that in the near future the results from microbiome quantitative trait analy-ses performed by big consortia39 will bring new insights into gene–microbiota interactions.
The biological meaning of these associations, however, will still need functional validation. Moreover, the integration of longitudinal data with tissue-specific interactions, for exam-ple combining colonic expression profiles with microbial composition in-situ, will help reveal local effects. Finally, some host–microbiome interactions can be evident in different molecular levels, for example metabolic dependencies. Therefore, integration of metabolite measurements will improve the characterization of human–microbial interactions.
In addition to direct implication of the gut microbiota in the development of disease, mi-crobes can also play a significant role in mediating the effect of treatment of host pathol-ogies. There is a growing interest in drug–microbiota interactions as it is becoming evi-dent that microbes can alter the efficiency of certain medication as well as modulate their side-effects. Modulation of the gut microbiota could thus become a complement of several therapies in the future.
In our efforts to investigate confounding factors in the study of the gut microbiota, we discovered that the use of proton-pump inhibitors (PPIs) is associated to several microbial changes (chapters 7 and 8). PPIs are a commonly used group of medications prescribed
Pharmacomicrobiomics: the role of the microbiota in the advances of personalized medicine
like the colonization of oral bacteria such as Klebsiella pneumoniae in the gut, have also been suggested as a potential mechanism to induce immune response and IBD44,45.
More-over, longitudinal data on new-onset treatment-naive UC patients has linked the baseline microbiome composition and its dynamics to differences in disease progression46. Overall,
these findings prevent us from ruling out the hypothesis that the microbiota is implicated in the development and progression of the diseases. Nonetheless, these results have to be interpreted carefully since associative studies do not indicate causality and mechanisms proven in mice cannot be directly translated to humans.
Longitudinal studies have, so far, helped to confirm the patterns observed in cross-sec-tional cohorts but, more importantly, have provided a dynamic perspective on the gut crobiota in health and disease. Despite recent efforts to characterize the host and mi-crobiota dynamics in patients with IBD, this has not been sufficient to determine which factor(s) trigger the diseases47, highlighting the complexity of IBD yet again. As described
in the previous section, the disease heterogeneity and its unpredictable dynamics pose a challenge for the design of optimal studies.
An alternative hypothesis is that the involvement of the gut microbiota in IBD is not directly related to its overall composition, but rather to its resilience and stability. An in-creasing incidence of IBD is associated with a westernized lifestyle. The western lifestyle may influence, already early in life, the maturation of the host immune system and the constitution of the gut microbial community, possibly resulting in a less stable gut eco-system. In this scenario, intrinsic or extrinsic perturbations may induce disruption of the gut homeostasis that generates a cascade of host immune-responses against the intestinal microbiota. Thus it is possible that exposures in early life are the key determinants of the risk of developing the disease2,48.
To conclude, our current knowledge on the gut microbiota in the context of IBD suggest that the dysbiosis seen in patients is partially driven by the disease and is influenced by inflammation and other derived consequences such as medication use, surgery or changes in intestinal transit. However, the microbiome’s composition and stability, and its interac-tion with the host, could play an essential role in the development and progression of the disease, and is yet to be revealed.
Current microbiome research still has technical and biological limitations that pose a great challenge to the translation of the research findings into clinical applications. Therefore, our current knowledge of the human microbiota is just the “tip of the iceberg” of a more complex (eco)system. In the coming sections, I suggest different strategies that in my opin-ion will help to tackle the current limitatopin-ions, keeping in mind that our end goal is that mi-crobiome research translates into improvements in medical practice and, as a consequence, improvements in human society.
to treat heartburn and are often complementary to other medications. The main action of PPIs is to inhibit acid secretion in the stomach. While epidemiological studies have related the use of this medication with an increased risk of enteric infections, we hypothesized that the reduced acidity in the stomach may promote the colonization of oral bacteria in the gut. This hypothesis is based on the observation that the faecal samples of PPI-users show an increased abundance of typically oral microbial species. Other researchers have also confirmed this association in independent cross-sectional and longitudinal studies49–53.
Interestingly, in a large cohort, Schmidt and colleagues found that although the transmis-sion of oral bacterial strains to the gut is observed in healthy individuals, increased levels of colonization are observed in disease phenotypes such as colorectal cancer and rheumatoid arthritis54.
Other types of medications, such as metformin and selective serotonin reuptake inhibi-tors also reveal associations with specific faecal microbial signatures (chapter 8). The close relation between use of medication and disease interferes with the assessment of causality in drug–microbiota relations. Recently, in-vitro models have provided evidence of different mechanisms in the microbe–medication bidirectional relation. For example, several chem-ical compounds found in human medications have exhibited antibiotic effects against gut bacterial strains55. In addition, human gut microbiota have been shown to have the ability
to metabolize certain types of medication56,57. Strikingly, the combined metabolic actions
of two human gut bacterial species, Eggerthella lenta and Eubacterium faecalis, have been implicated in the efficiency of L-dopa treatment in patients with Parkinson’s disease58.
Likewise, it has been suggested that the microbiota composition can be determinant of the success of immune checkpoint inhibitor therapy in colorectal cancer59.
To summarise, the gut microbiota is emerging as an important factor in the field of personalized medicine. The capacity of microbes to metabolize the chemical compounds present in drugs has an impact on therapy efficacy and potential side effects. While the study of compositional signatures of the gut microbiome has identified interesting links between microbes and medication, in my opinion the in-depth study of microbial genes, enzymes and metabolites will be the inflection point in the field of pharmacomicrobiomics.
In the past 5 years our knowledge of the gut microbiota community has increased dra-matically. However, this is probably true for just a part of this complex ecosystem. For historical and technical reasons, efforts to characterize the gut microbiota have focused on the identification of bacterial taxonomic groups. So far, multiple studies have been able to characterize which bacteria are more prevalent in the human gut and in which relative proportion they are present60–62. Despite this, recent studies combining a large number of
human metagenomic samples from all across the world and combining sequencing data with in-vitro techniques have shown that a relevant proportion of bacterial species in the human gut remain uncharacterized63–67. Although it is estimated that these new findings
have increased the mappability of faecal metagenomes up to, on average, 80% of the reads, our knowledge of the human bacteriome is still not complete68. Moreover, if we extrapolate
this to other less studied human microbial communities, the proportion of unknown bac-teria will certainly be larger.
Importantly, most of the studies investigating the gut microbiota rely on faecal material as a representation of the gut ecosystem. However, it has already been shown that faeces contains a mixture of microbial populations that are mainly enriched in bacteria from the distal gut69,70. Thus, it is possible that a relevant proportion of the unknown bacterial
rich-ness in the gut is represented by bacteria deriving from the proximal intestine.
To determine to which degree we are underestimating gut microbial diversity, it will be necessary to profile mucosal biopsies from different locations in the gastrointestinal tract. Moreover, the use of large cohorts will also help to reveal low prevalence bacterial groups and their implications. Finally, improvements in metagenomic sequencing (e.g. deeper and longer reads) are necessary to characterize intra-species genetic variation via the recovery of microbial genomes by de novo assembly.
In addition to the bacterial component of the human microbiota, there is increasing interest in other members of this ecosystem, e.g. fungi, small eukaryotes and viruses. Our current knowledge of the diversity and roles of these groups within the human microbiota is very limited. For example, it is believed that the number of viruses in the gut is greater than the number of bacteria, and, interestingly, bacteriophages have been proposed to play an important role in regulating the bacterial population in the gut71. In addition, studies
of patients with IBD have suggested that gut dysbiosis also implicates fungal72 and viral
species73, but the implications of these species (as either cause or consequence) has not
been elucidated. Strikingly, a longitudinal study of patients with IBD has revealed that some patients exhibit viral spikes preceding periods of active disease47. Although capturing
all this information involves overcoming many technical challenges, it is becoming evident that the study of the gut ecosystem cannot be limited to the bacterial composition: it needs an integrative approach to capture the full microbial diversity and relations.
Another understudied component of the microbiota is its genes and functional charac-teristics. Shotgun metagenomic sequencing has facilitated not only the study of the mi-crobial taxa but also the possibility of investigating the genomic content of the mimi-crobial community. To date, most large population studies have focused on the characterization of gene families and predicted pathways. However, metagenomic datasets also have great potential for investigating the presence of previously uncharacterized genes. The discovery
Thus far, most of the research exploring the human microbiota has been performed consid-ering the microbial community as a collection of multiple independent traits or taxa. This has allowed us to explore its composition and structure and to identify the key species and characteristics associated to different human conditions or diseases, all under the partial assumption that the human microbiota is a relatively constant feature that rarely changes in normal conditions. However, in contrast to the “trait” model, the microbiota constitutes multiple ecosystems in the human body78,79. Therefore, adopting more complex models
based on the principles of ecology and evolution will help us understand the complexity of the human–microbiota interactions.
The members of a microbial ecosystem are not independent. The gut microbiota, for example, is heavily controlled and modulated by the host immune system and constantly exposed to multiple environmental factors (diet, medication, etc.). In addition, microbes establish variable types of relations between each other: metabolic dependencies, competi-tion mechanisms and cooperacompeti-tion strategies80. Correlations between bacterial abundances
or between microbes and human traits in cross-sectional cohorts are probably not sufficient to disentangle these complex relations.
Change of perspective: the microbiome is an ecosystem not just a trait
of new genes can reveal new bacterial functions that can be relevant for human health, such as antibiotic resistance mechanisms. To illustrate this potential, a recent study has shown that the gut microbiota harbours the potential to convert human blood types74. Moreover,
other elements of microbial genetics, such as gene mutations, regulatory elements or the transfer of genomic content between species (i.e. horizontal gene transfer), are of great relevance in the dynamics, adaptation and evolution of the gut microbiota, and they should be considered when performing time-series experiments.
The study of faecal metabolites in combination with gut microbial composition and host genetics will improve our understanding of the microbial communities, their functions and their interactions with the host75–77. Moreover, changes in the abundance of metabolites
can be indicative of several disorders17,23,24. For example, it has been shown that patients
with IBD show a significant alteration of their faecal metabolomic profiles. However, a large proportion of the differentially abundant compounds in patients with IBD are still uncharacterized17,47. Therefore, more research is needed in order to reveal the relevance
of faecal metabolites in the context of IBD and to investigate whether these changes are merely reflecting functional dysbiosis in the gut or if some of these features can provide a better understanding of disease causality.
In summary, combining in-depth characterization of bacterial genomes with the study of other members of the gut microbiota such as viruses and fungi will lead to a better under-standing of the key players in gut microbial ecosystems and the discovery of new species and metabolic functions, which together can reveal relevant new features for host health.
Fig 2.
Dynamics of the gut microbiota.
Representation of the temporal changes on the gut microbiota composition and richness (y-axis). The dashed line indicates the predicted dynamics based on the sampling frequency. The top panel illustrates the dynamic of the gut microbiota in a healthy individual. Antibiotics use and infection periods represent perturbations in the gut ecosystem (blue gradient). The monitoring of microbial changes during these periods and the time that ecosystem takes to reach its stationary state (stability) can be used to estimate resilience properties. Bottom panel refl ects the dynamics of the gut ecosystem in patients with IBD: char-acterized by more frequent and profound changes. Disease activity (red gradient) represents perturba-tion periods which are associated with changes in microbial composiperturba-tion and richness.
Additionally, although the majority of ecosystems tend to a stationary/stable state, the gut microbiota is exposed to multiple environmental disruptions during the life of the host81. The capacity of the gut ecosystem to recover from these perturbations – resilience
– is probably one of the key characteristics to identify “healthy” microbial populations82.
Therefore, in the future, it is important that we incorporate these kinds of measurements in population and disease cohorts. The integration of multi-omic data in a longitudinal set-up can help us gain better insight into the dynamics and normal fluctuations of host–micro-biota interactions, for example by combining daily stool collection and metabolic profiling with host exposures and genetics. The induction of controlled perturbations and the meas-urement of the timespan that a gut microbial ecosystem needs to recover to its initial state could be used as estimation of its resilience. Resilience properties, therefore, can be one of the main characteristics that help to discriminate healthy ecosystems from those that are unstable and consequently prone to develop diseases (figure 2).
Another important aspect of an ecosystem is its evolution. Multi-omic data from birth cohorts is becoming available83–86, and periodic measurements of these cohorts will give
us an insight into the evolution of the gut microbiota. Epidemiologic studies have already established a link between early-life events and the development of autoimmune diseases and IBD. It would therefore be interesting to see how disruptive events during the child development, such as infections or use of antibiotics, affect the consolidation of the gut microbiota community and whether this has a direct or indirect relation with the develop-ment of health disorders.
Another promising strategy to investigate the microbial ecosystem is “organs-on-chip” technology87. Although still in its infancy, the simulation and monitoring of human tissues
in chips is already a reality. The integration of the microbiota into this system can make a big difference in how we study microbial communities, allowing the consideration of multiple taxonomic groups at the same time in the context of different host genetic back-grounds. Ideally, if such a system can maintain different human ecosystems in a controlled and monitored way, we could study local human–microbiota interactions and facilitate the study of the changes induced by environmental factors such as the presence of certain compounds (like drugs) or pathogens.
To sum up, redefining the microbiota as a dynamic network in which each of the mem-bers is related to the rest of the community and to the host will bring a better definition of what is a healthy microbiota, and consequently improve the way in which we currently approach therapeutic options such as faecal microbiota transplantation (FMT), probiotics and prebiotics supplementation.
Microbiome research still lacks gold standards. Although several efforts are currently being established88–90, a community-wide efforts are needed to establish a set of guidelines or best
practices for microbiome research. This will not only improve the quality of the research, it will also increase the reusability of data and the reproducibility of the scientific findings.
As described in the introduction of this thesis, the lack of previous knowledge on the structure and composition of the microbial ecosystem, in combination with the potential artefacts introduced in each step of the microbiome workflow, pose an important challenge to the biological interpretation of the data obtained. So far, the research community does not have a standard protocol that guarantees a high degree of conservation of the microbial ecosystem characteristics from sampling to sequencing. Therefore, to minimize bias and improve comparability between cohorts, it is important to annotate relevant information in each of the protocol steps. Storage conditions (e.g. use of buffers, conservation temper-atures) and sample characteristics (Bristol Stool Chart, quantity of DNA isolated, etc.) are known to be important confounding factors. In the context of biomedical research, the collection of metadata should not be limited to the sample and technical characteris-tics, but also include host information such as life-style information and disease-specific sub-phenotypes. Although more information is better, technical and ethical considerations can limit the extent to which metadata can be made publicly available. In those cases, researchers should prioritize collecting and releasing those phenotypes that, based on pre-vious research, have been shown to have bigger contribution to the inter-sample variation. While, in my opinion, most of the standardization efforts should be put into the sam-ple and metadata collection, computational protocols should also incorporate guidelines and consensus for best practice. Currently, there are multiple tools and pipelines availa-ble for the taxonomic and functional identification for both amplicon-based and shotgun metagenomic data. Studies comparing methods usually conclude that the selection of one tool versus another depends on the researcher’s interest and the compromises between sensitivity versus specificity, in addition to computational speed91,92. While it is difficult
to envision (in the near future) a tool that will solve all the limitations of the current bioinformatic approaches, I think that a way to overcome these constraints would be to develop frameworks to integrate different computational strategies. These would provide a confidence score for each prediction: for example, combining reference-based methods (both marker-gene-based and k-mers-based methods) with de-novo assembly approaches for taxonomic characterization. Moreover, the availability of open-source software should motivate developers and data scientists into collaborative – rather than competitive – ef-forts in order to improve and optimize existing methods.
Most microbiome research and tools rely on databases containing genomes and micro-bial genes. The availability of large metagenomic sequenced cohorts was provided a new resource of previously unknown microbial genomes, providing better taxonomic and func-tional profiles. Considering this, it is important to actively maintain databases by curating microbial reference genomes and by incorporating the high-quality genomes derived from isolation experiments and culturomics as well as metagenomic predictions93. Expanding
these references will help reveal the diversity of many complex microbial ecosystems as well as facilitating the characterization of samples by reducing the uncertainties in taxonomic classifications. This task is particularly urgent in the study of other microorganisms, e.g. viruses, since current databases are extremely incomplete.
Setting gold standards for the statistical methods for analysing microbiome data will improve the discovery of trait–microbiota associations. The use of compositional data
facil-The understanding of the human microbiota is necessary to define the concept of “health” and identify risk factors or potential causes of disease. Replacing or modulating the mi-crobiota seems more feasible and – so far – more ethical than editing the human genome. In consequence, there is a big interest in translating the current knowledge on the human microbial ecosystem into clinical applications103. While the previous sections focussed on
the limitations in the current microbiome research and how to overcome (some of) them in the near future, in this section I would like to highlight the potential and clinical impli-cations of the microbial work.
In my opinion, there are three ways in which microbiota could be used in medicine: 1) microbiota transplantation to replace a damaged ecosystem with a healthier one in order to cure a disease; 2) modulation of the microbiota, for example by enhancing host-beneficial properties of microbes or inhibiting the expression of virulence mechanisms of certain bacteria, as a complement to therapy; and 3) the use of faecal microbiota as a biomarker or diagnostic/screening tool to improve early detection of host disorders.
FMT is probably one of the most promising therapeutics so far. The capacity to treat
Clostridium difficile infections with this technique has opened the door for multiple trials
in other conditions, including UC104,105. In the case of UC, the first reports have
suggest-ed that FMT could rsuggest-educe disease activity, but not its overall remission106. Although the
treatment seems to be effective for C. difficile infection, the exact mechanisms by which the microbial ecosystem is re-established after transplantation is still unknown. A study using sterile faecal filtrates suggested that the observed efficacy of FMT in patients with C.
Dif-ficile infection could be mediated by bacteriophages or metabolites rather than bacteria107.
Investigation of the host–microbiota interactions and microbial ecosystem properties of the recipient may help to increase the efficacy of the treatment in other disorders. More-over, protocols with strict selection of donors and extensive screening of the microbiota
Steps towards microbial-target therapies
itates the comparison of multiple samples with different characteristics, including different sequencing depths. However, a lack of quantitative methods still poses a challenge for interpreting results from microbiome studies94–97, and several statistical approaches have
been suggested to overcome these limitations98–102. In the future, quantitative
measure-ments, either accompanying the metagenomic samples (bacterial counts as part of mini-mum-required meta-data) or as a part of the study replication (direct count of the amount of certain trait-associated taxa), will be important to assess the biological relevance of microbe–trait correlations. Alternatively, the development of high-throughput single-cell microbial sequencing approaches could enable us to study microbial communities in a high resolution, quantitative manner.
Overall, I think that it is time to bring microbiome research closer to the actual (mi-cro)biology. The continuous development of new (and better) experimental methods, the standardization of procedures and, finally, the combination of sequencing techniques with laboratory experiments are necessary to reveal biological relations.
The study of the gut microbiota in patients with IBD has provided a better understanding of the gut ecosystem. As discussed in the previous sections of this chapter, the dysbiosis observed in these patients, which occurs at different molecular levels, gives us the oppor-tunity to investigate new treatment and diagnostic options. However, the exact role of the microbiota in the development and the progression of the disease is still unknown. While the results obtained from cross-sectional studies may help us to treat the diseases, in my opinion, the study of longitudinal data will provide information that could be used to pre-vent flares. Consequently, two essential questions need to be addressed in the future: What is the role of the microbiota in the transition from remission to active disease? And can we predict and/or prevent these episodes by targeting the gut microbiota?
The first time-series studies have shown that patients with IBD show, on average, a less stable gut microbiota than non-IBD individuals. Patient microbiome stability is dependent
Future of gut microbiome research in IBD: a dynamic perspective.
pre-transplantation should help to dissipate ethical and clinical concerns. At the time of writing this thesis chapter, the U.S. Food and Drug Administration agency had issued a warning regarding the use of FMT and the transmission of multi-drug resistant organisms. Selective targeted approaches require a personalized understanding of the microbial eco-systems. As I described earlier, pharmacomicrobiomics represents a good example in which the modulation of the gut microbiota can have clinical impact. For example, the efficiency of certain therapies could be improved by not only adjusting the treatment to the host and microbiota characteristics, but also by stimulating or inhibiting microbial features. The use of probiotics could also be promising, but it has been shown that bacterial engraftment rates are personalized, indicating that the recipient gut ecosystem is relevant in achieving colonization69,70. Alternatively, phages and microbial metabolites can be a new resource
for non-invasive treatment options108. In the case of IBD, the observed dysbiosis indicates
the relevant targets to restore and maintain. For example, restoring some of the depleted mechanisms such as short-chain fatty acid and vitamin biosynthesis, as well as bile-acid metabolism, could help ameliorate patients’ symptoms and/or reduce inflammation.
Finally, I believe that the most immediate application of metagenomic data as a clinical application is its use as a biomarker or diagnostic tool. For example, use of sequencing technologies can improve the detection of pathogens, resistance mechanisms or virulence factors in the context of infection outbreaks109. In addition, the study of the gut microbiota
in recent years has led to large catalogue of faecal characteristics associated with multiple host disorders. As I discussed in chapter 3, we have shown that changes in the microbial composition in faecal samples can be used as a biomarker to assist the diagnosis of IBD. The implementation of such a test could potentially reduce the number of colonoscopies, resulting in lower burden for patients, in terms of invasive testing and time to diagnosis, and society in terms of reduced costs. Moreover, the design of a microbiome-based test could be expanded to other diseases such as colorectal cancer or applied to evaluate an individual’s response to medical treatments.
The work presented in this thesis has provided a better understanding of the gut microbial ecosystem, identifying a wide range of factors associated to its composition. In the future, new insights into microbe–microbe and host–microbe interactions, both at different mo-lecular levels and over time, will lead to the transition from a “microbiome-as-trait” model to an ecosystem perspective on human microbial communities. Considering that microbi-ome research is already becoming a keystone in the development of personalized medicine approaches, we can expect that its evolution will translate to the prevention and better treatment of many human disorders.
Final remarks
on the disease activity and disease phenotypes, e.g. inflammation location and whether the microbial composition experiences transitions between “healthy” and “dysbiotic” status21,47.
Current research, however, is not able to determine which factors precede disease events (flares). A mismatch between the timespan over which these changes occur and the fre-quency at which sampling has traditionally been performed (bi-weekly or monthly) may explain the current limitations to the detection of relevant predictors (figure 2).
I believe that the combination of periodic sampling (weekly or monthly) in large IBD cohorts with intervals of daily monitoring will help us determine which changes occur before flares. Periodic sampling by established biobanks will help determine the long-term effects of disease events or disruptions, e.g. surgery and antibiotic use, as well as capture a broader spectrum of disease heterogeneity. Daily sampling will provide insights into the stability and resilience of the gut microbiota, allowing us to link these characteristics to different disease dynamics. However, it is important to acknowledge that study designs like the one proposed above represent a big logistical challenge that will require close collabo-ration between patients, doctors and researchers.
Finally, the interpretation of the dynamics and evolution of the gut microbiota in the con-text of IBD dynamics is one of the major goals for understanding the disease. However, the study of the gut microbiota should not be performed without considering its context. Therefore, as I highlighted in the previous sections, attempts to describe microbial changes in a particular human ecosystem, including IBD, will benefit from the integration of multiple layers of information (such as lifestyle, genetics, transcriptomics and metabolites of the host) with compositional, transcriptional and quantitative measurements of the microbiota.
1. Weissbrod, O., Rothschild, D., Barkan, E. & Segal, E. Host genetics and microbiome associations through the lens of genome wide association studies. Curr. Opin. Microbiol. 44, 9–19 (2018).
2. Kaplan, G. G. & Ng, S. C. Understanding and Preventing the Global Increase of Inflammatory Bowel Disease. Gastroenterology 152, 313-321.e2 (2017). 3. Cohen, L. J., Cho, J. H., Gevers, D. & Chu, H. Genetic Factors and the Intestinal Microbiome Guide Development of Microbe-Based Therapies for Inflammatory Bowel Diseases. Gastroenterology 156, 2174–2189 (2019).
4. Ashwin N. Ananthakrishnan. Epidemiology and risk factors for IBD. Nat. Rev. Gastroenterol. Hepatol. 12, 205–217 (2015).
5. Liu, J. Z. et al. Association analyses identify 38 susceptibility loci for inflammatory bowel disease and highlight shared genetic risk across populations. Nat. Genet. 47, 979–986 (2015).
6. Huang, H. et al. Fine-mapping inflammatory bowel disease loci to single-variant resolution. Nature 547, 173–178 (2017).
7. Jostins, L. et al. Host-microbe interactions have shaped the genetic architecture of inflammatory bowel disease. Nature 491, 119–124 (2012). 8. Morgan, X. C. et al. Dysfunction of the intestinal microbiome in inflammatory bowel disease and treatment. Genome Biol. 13, R79 (2012). 9. Goethel, A. et al. Nod2 influences microbial resilience and susceptibility to colitis following antibiotic exposure. Mucosal Immunol. 12, 720 (2019). 10. Madsen, K. L. et al. Antibiotic therapy attenuates colitis in interleukin 10 gene-deficient mice. Gastroenterology 118, 1094–1105 (2000). 11. Sellon, R. K. et al. Resident Enteric Bacteria Are Necessary for Development of Spontaneous Colitis and Immune System Activation in Interleukin-10-Deficient Mice. Infect. Immun. 66, 5224–5231 (1998).
12. Gevers, D. et al. The treatment-naive microbiome in new-onset Crohn’s disease. Cell Host Microbe 15, 382–392 (2014).
13. Machiels, K. et al. A decrease of the butyrate-producing species Roseburia hominis and Faecalibacterium prausnitzii defines dysbiosis in patients with ulcerative colitis. Gut 63, 1275–1283 (2014).
14. Strauss, J. et al. Invasive potential of gut mucosa-derived Fusobacterium nucleatum positively correlates with IBD status of the host. Inflamm.
Bowel Dis. 17, 1971–1978 (2011).
15. Palmela, C. et al. Adherent-invasive Escherichia coli in inflammatory bowel disease. Gut 67, 574–587 (2018).
16. Parada Venegas, D. et al. Short Chain Fatty Acids (SCFAs)-Mediated Gut Epithelial and Immune Regulation and Its Relevance for Inflammatory Bowel Diseases. Front. Immunol. 10, (2019).
17. Franzosa, E. A. et al. Gut microbiome structure and metabolic activity in inflammatory bowel disease. Nat. Microbiol. 4, 293 (2019).
18. Duboc, H. et al. Connecting dysbiosis, bile-acid dysmetabolism and gut inflammation in inflammatory bowel diseases. Gut 62, 531–539 (2013).
19. Pascal, V. et al. A microbial signature for Crohn’s disease. Gut 66, 813–822 (2017).
20. Pasolli, E., Truong, D. T., Malik, F., Waldron, L. & Segata, N. Machine Learning Meta-analysis of Large Metagenomic Datasets: Tools and Biological Insights. PLOS Comput. Biol. 12, e1004977 (2016).
21. Halfvarson, J. et al. Dynamics of the human gut microbiome in inflammatory bowel disease. Nat. Microbiol. 2, 17004 (2017). 22. Zhou, Y. et al. Gut Microbiota Offers Universal Biomarkers across Ethnicity in Inflammatory Bowel Disease Diagnosis and Infliximab Response Prediction. mSystems 3, (2018).
23. Yachida, S. et al. Metagenomic and metabolomic analyses reveal distinct stage-specific phenotypes of the gut microbiota in colorectal cancer.
Nat. Med. 25, 968 (2019).
24. Zhou, W. et al. Longitudinal multi-omics of host–microbe dynamics in prediabetes. Nature 569, 663 (2019).
25. Cleynen, I. et al. Inherited determinants of Crohn’s disease and ulcerative colitis phenotypes: a genetic association study. The Lancet 387, 156–167 (2016). 26. Uniken Venema, W. T., Voskuil, M. D., Dijkstra, G., Weersma, R. K. & Festen, E. A. The genetic background of inflammatory bowel disease: from correlation to causality. J. Pathol. 241, 146–158 (2017).
27. Tiratterra, E. et al. Role of bile acids in inflammatory bowel disease. Ann. Gastroenterol. 31, 266–272 (2018).
28. Zhernakova, A. et al. Population-based metagenomics analysis reveals markers for gut microbiome composition and diversity. Science 352, 565–569 (2016). 29. Falony, G. et al. Population-level analysis of gut microbiome variation. Science 352, 560–564 (2016).
30. Turpin, W. et al. Association of host genome with intestinal microbial composition in a large healthy cohort. Nat. Genet. 48, 1413–1417 (2016). 31. Bonder, M. J. et al. The effect of host genetics on the gut microbiome. Nat. Genet. 48, 1407–1412 (2016).
32. Goodrich, J. K. et al. Genetic Determinants of the Gut Microbiome in UK Twins. Cell Host Microbe 19, 731–743 (2016).
33. Wang, J. et al. Genome-wide association analysis identifies variation in vitamin D receptor and other host factors influencing the gut microbiota. Nat. Genet. 48, 1396–1406 (2016).
34. Davenport, E. R. et al. Genome-Wide Association Studies of the Human Gut Microbiota. PLOS ONE 10, e0140301 (2015). 35. Blekhman, R. et al. Host genetic variation impacts microbiome composition across human body sites. Genome Biol. 16, 191 (2015). 36. Kolde, R. et al. Host genetic variation and its microbiome interactions within the Human Microbiome Project. Genome Med. 10, 6 (2018). 37. Knights, D. et al. Complex host genetics influence the microbiome in inflammatory bowel disease. Genome Med. 6, 107 (2014).
38. Aschard, H. et al. Genetic effects on the commensal microbiota in inflammatory bowel disease patients. PLOS Genet. 15, e1008018 (2019). 39. Wang, J. et al. Meta-analysis of human genome-microbiome association studies: the MiBioGen consortium initiative. Microbiome 6, 101 (2018). 40. Kostic, A. D., Xavier, R. J. & Gevers, D. The Microbiome in Inflammatory Bowel Diseases: Current Status and the Future Ahead.
Gastroenterology 146, 1489–1499 (2014).
41. Ni, J., Wu, G. D., Albenberg, L. & Tomov, V. T. Gut microbiota and IBD: causation or correlation? Nat. Rev. Gastroenterol. Hepatol. 14, 573–584 (2017).
42. Hall, A. B. et al. A novel Ruminococcus gnavus clade enriched in inflammatory bowel disease patients. Genome Med. 9, (2017).
43. Palm, N. W. et al. Immunoglobulin A Coating Identifies Colitogenic Bacteria in Inflammatory Bowel Disease. Cell 158, 1000–1010 (2014). 44. Schirmer, M. et al. Linking the Human Gut Microbiome to Inflammatory Cytokine Production Capacity. Cell 167, 1125-1136.e8 (2016). 45. Atarashi, K. et al. Ectopic colonization of oral bacteria in the intestine drives TH1 cell induction and inflammation. Science 358, 359–365 (2017). 46. Schirmer, M. et al. Compositional and Temporal Changes in the Gut Microbiome of Pediatric Ulcerative Colitis Patients Are Linked to Disease Course. Cell Host Microbe 24, 600-610.e4 (2018).
47. Lloyd-Price, J. et al. Multi-omics of the gut microbial ecosystem in inflammatory bowel diseases. Nature 569, 655 (2019).
48. Bernstein, C. N., Burchill, C., Targownik, L. E., Singh, H. & Roos, L. L. Events Within the First Year of Life, but Not the Neonatal Period, Affect Risk for Later Development of Inflammatory Bowel Diseases. Gastroenterology 156, 2190-2197.e10 (2019).
49. Castellani, C. et al. The Influence of Proton Pump Inhibitors on the Fecal Microbiome of Infants with Gastroesophageal Reflux—A Prospective Longitudinal Interventional Study. Front. Cell. Infect. Microbiol. 7, (2017).
50. Hojo, M. et al. Gut Microbiota Composition Before and After Use of Proton Pump Inhibitors. Dig. Dis. Sci. 63, 2940–2949 (2018). 51. Jackson, M. A. et al. Proton pump inhibitors alter the composition of the gut microbiota. Gut 65, 749–756 (2016).
52. Reveles, K. R., Ryan, C. N., Chan, L., Cosimi, R. A. & Haynes, W. L. Proton pump inhibitor use associated with changes in gut microbiota composition. Gut 67, 1369–1370 (2018).
53. Takagi, T. et al. The influence of long-term use of proton pump inhibitors on the gut microbiota: an age-sex-matched case-control study. J. Clin.
Biochem. Nutr. 62, 100–105 (2018).
54. Schmidt, T. S. et al. Extensive transmission of microbes along the gastrointestinal tract. eLife 8, e42693 (2019). 55. Maier, L. et al. Extensive impact of non-antibiotic drugs on human gut bacteria. Nature 555, 623–628 (2018).
56. Zimmermann, M., Zimmermann-Kogadeeva, M., Wegmann, R. & Goodman, A. L. Separating host and microbiome contributions to drug pharmacokinetics and toxicity. Science 363, eaat9931 (2019).
57. Zimmermann, M., Zimmermann-Kogadeeva, M., Wegmann, R. & Goodman, A. L. Mapping human microbiome drug metabolism by gut bacteria and their genes. Nature 1 (2019). doi:10.1038/s41586-019-1291-3
58. Rekdal, V. M., Bess, E. N., Bisanz, J. E., Turnbaugh, P. J. & Balskus, E. P. Discovery and inhibition of an interspecies gut bacterial pathway for Levodopa metabolism. Science 364, eaau6323 (2019).
59. Routy, B. et al. Gut microbiome influences efficacy of PD-1–based immunotherapy against epithelial tumors. Science 359, 91–97 (2018). 60. The Human Microbiome Project Consortium (last) et al. Structure, function and diversity of the healthy human microbiome. Nature 486, 207–214 (2012). 61. Arumugam, M. et al. Enterotypes of the human gut microbiome. Nature 473, 174–180 (2011).
62. Donaldson, G. P., Lee, S. M. & Mazmanian, S. K. Gut biogeography of the bacterial microbiota. Nat. Rev. Microbiol. 14, 20–32 (2016). 63. Almeida, A. et al. A new genomic blueprint of the human gut microbiota. Nature 1 (2019). doi:10.1038/s41586-019-0965-1
64. Forster, S. C. et al. A human gut bacterial genome and culture collection for improved metagenomic analyses. Nat. Biotechnol. 37, 186 (2019). 65. Nayfach, S., Shi, Z. J., Seshadri, R., Pollard, K. S. & Kyrpides, N. Novel insights from uncultivated genomes of the global human gut microbiome. Nature 1 (2019). doi:10.1038/s41586-019-1058-x
66. Pasolli, E. et al. Extensive Unexplored Human Microbiome Diversity Revealed by Over 150,000 Genomes from Metagenomes Spanning Age, Geography, and Lifestyle. Cell 176, 649-662.e20 (2019).
67. Zou, Y. et al. 1,520 reference genomes from cultivated human gut bacteria enable functional microbiome analyses. Nat. Biotechnol. 37, 179 (2019). 68. Thomas, A. M. & Segata, N. Multiple levels of the unknown in microbiome research. BMC Biol. 17, 48 (2019).
69. Zmora, N. et al. Personalized Gut Mucosal Colonization Resistance to Empiric Probiotics Is Associated with Unique Host and Microbiome Features. Cell 174, 1388-1405.e21 (2018).
70. Suez, J. et al. Post-Antibiotic Gut Mucosal Microbiome Reconstitution Is Impaired by Probiotics and Improved by Autologous FMT. Cell 174, 1406-1423.e16 (2018).
71. Shkoporov, A. N. & Hill, C. Bacteriophages of the Human Gut: The ‘Known Unknown’ of the Microbiome. Cell Host Microbe 25, 195–209 (2019). 72. Sokol, H. et al. Fungal microbiota dysbiosis in IBD. Gut 66, 1039–1048 (2017).
74. Rahfeld, P. et al. An enzymatic pathway in the human gut microbiome that converts A to universal O type blood. Nat. Microbiol. 1 (2019). doi:10.1038/s41564-019-0469-7
75. Michelini, S. et al. A reverse metabolic approach to weaning: in silico identification of immune-beneficial infant gut bacteria, mining their metabolism for prebiotic feeds and sourcing these feeds in the natural product space. Microbiome 6, 171 (2018).
76. Zierer, J. et al. The fecal metabolome as a functional readout of the gut microbiome. Nat. Genet. 50, 790 (2018).
77. Chen, H. et al. A Forward Chemical Genetic Screen Reveals Gut Microbiota Metabolites That Modulate Host Physiology. Cell 177, 1217-1231.e18 (2019). 78. Dethlefsen, L., McFall-Ngai, M. & Relman, D. A. An ecological and evolutionary perspective on human–microbe mutualism and disease. Nat.
2007 4497164 (2007).
79. Gilbert, J. A. & Lynch, S. V. Community ecology as a framework for human microbiome research. Nat. Med. 1 (2019). doi:10.1038/s41591-019-0464-9 80. Ley, R. E., Peterson, D. A. & Gordon, J. I. Ecological and Evolutionary Forces Shaping Microbial Diversity in the Human Intestine. Cell 124, 837–848 (2006).
81. Faust, K., Lahti, L., Gonze, D., de Vos, W. M. & Raes, J. Metagenomics meets time series analysis: unraveling microbial community dynamics.
Curr. Opin. Microbiol. 25, 56–66 (2015).
82. Relman, D. A. The human microbiome: ecosystem resilience and health. Nutr. Rev. 70 Suppl 1, S2–9 (2012).
83. Stewart, C. J. et al. Temporal development of the gut microbiome in early childhood from the TEDDY study. Nature 562, 583–588 (2018). 84. Vatanen, T. et al. The human gut microbiome in early-onset type 1 diabetes from the TEDDY study. Nature 562, 589 (2018).
85. Vatanen, T. et al. Variation in Microbiome LPS Immunogenicity Contributes to Autoimmunity in Humans. Cell 165, 842–853 (2016). 86. Zhong, H. et al. Impact of early events and lifestyle on the gut microbiota and metabolic phenotypes in young school-age children. Microbiome
7, 2 (2019).
87. Jalili-Firoozinezhad, S. et al. A complex human gut microbiome cultured in an anaerobic intestine-on-a-chip. Nat. Biomed. Eng. 1 (2019). doi:10.1038/s41551-019-0397-0
88. Knight, R. et al. Best practices for analysing microbiomes. Nat. Rev. Microbiol. 16, 410–422 (2018).
89. Sczyrba, A. et al. Critical Assessment of Metagenome Interpretation—a benchmark of metagenomics software. Nat. Methods 14, 1063–1071 (2017). 90. Costea, P. I. et al. Towards standards for human fecal sample processing in metagenomic studies. Nat. Biotechnol. 35, 1069–1076 (2017). 91. Vollmers, J., Wiegand, S. & Kaster, A.-K. Comparing and Evaluating Metagenome Assembly Tools from a Microbiologist’s Perspective - Not Only Size Matters! PLOS ONE 12, e0169662 (2017).
92. McIntyre, A. B. R. et al. Comprehensive benchmarking and ensemble approaches for metagenomic classifiers. Genome Biol. 18, 182 (2017). 93. Lagier, J.-C. et al. Culturing the human microbiota and culturomics. Nat. Rev. Microbiol. 16, 540 (2018).
94. Gloor, G. B., Macklaim, J. M., Pawlowsky-Glahn, V. & Egozcue, J. J. Microbiome Datasets Are Compositional: And This Is Not Optional.
Front. Microbiol. 8, (2017).
95. Gloor, G. B., Wu, J. R., Pawlowsky-Glahn, V. & Egozcue, J. J. It’s all relative: analyzing microbiome data as compositions. Ann. Epidemiol. 26, 322–329 (2016). 96. Nayfach, S. & Pollard, K. S. Toward Accurate and Quantitative Comparative Metagenomics. Cell 166, 1103–1116 (2016).
97. Tsilimigras, M. C. B. & Fodor, A. A. Compositional data analysis of the microbiome: fundamentals, tools, and challenges. Ann. Epidemiol. 26, 330–335 (2016).
98. Props, R. et al. Absolute quantification of microbial taxon abundances. ISME J. 11, 584–587 (2017).
99. Mandal, S. et al. Analysis of composition of microbiomes: a novel method for studying microbial composition. Microb. Ecol. Health Dis. 26,
27663 (2015).
100. Friedman, J. & Alm, E. J. Inferring Correlation Networks from Genomic Survey Data. PLOS Comput. Biol. 8, e1002687 (2012). 101. Rivera-Pinto, J. et al. Balances: a New Perspective for Microbiome Analysis. mSystems 3, e00053-18 (2018).
102. Pereira, M. B., Wallroth, M., Jonsson, V. & Kristiansson, E. Comparison of normalization methods for the analysis of metagenomic gene abundance data. BMC Genomics 19, 274 (2018).
103. Knox, N. C., Forbes, J. D., Van Domselaar, G. & Bernstein, C. N. The Gut Microbiome as a Target for IBD Treatment: Are We There Yet?
Curr. Treat. Options Gastroenterol. (2019). doi:10.1007/s11938-019-00221-w
104. Wilson, B. C., Vatanen, T., Cutfield, W. S. & O’Sullivan, J. M. The Super-Donor Phenomenon in Fecal Microbiota Transplantation. Front.
Cell. Infect. Microbiol. 9, (2019).
105. Mcilroy, J. R. et al. Current and future targets for faecal microbiota transplantation. Hum. Microbiome J. 11, (2019).
106. Costello, S. P. et al. Effect of Fecal Microbiota Transplantation on 8-Week Remission in Patients With Ulcerative Colitis: A Randomized Clinical Trial. JAMA 321, 156–164 (2019).
107. Ott, S. J. et al. Efficacy of Sterile Fecal Filtrate Transfer for Treating Patients With Clostridium difficile Infection. Gastroenterology 152, 799-811.e7 (2017).
108. Cisek, A. A., Dąbrowska, I., Gregorczyk, K. P. & Wyżewski, Z. Phage Therapy in Bacterial Infections Treatment: One Hundred Years After the Discovery of Bacteriophages. Curr. Microbiol. 74, 277–283 (2017).
