The Integrative Human microbiome project: a mile stone in the understanding of the gut microbiome

KEYWORDS:

• Microbiome
• microbiota
• human microbiome project
• dysbiosis
• metabolome
• therapeutics
• metabolomics
• ecological niche

1. Introduction

In the past several years, innovative technologies have allowed for the identification and quantification of the human microbiome composition and function [1]. A PubMed search of the terms: ‘microbiota’[MeSH Terms] OR ‘microbiota’[All Fields] OR ‘microbiome’[All Fields] yields 71,075 articles (search performed on May 19^th, 2020). The first search result was from 1956. From the years 1956 to 2001, less than 80 search results a year were seen. In the year 2019, there were 14,339 search results. These numbers are a testament to dramatically increased research activities, which in turn has enhanced our knowledge of the relationship of the human microbiome with health and disease [2].

The term microbiota is used to define the population of microorganisms (bacteria, fungi and viruses) that live in an ecological niche. The term microbiome is used to define the collective genomes of these microbes living in an ecological niche. The term metabolome is used to define the complete set of small-molecule metabolites within a biological sample. Metabolomics is the study of chemical processes involving the products of metabolism. Metagenomics is the collective study of all the genetic material within a microbiome niche. The term symbiont is used to define an organism participating in a symbiotic relationship with a host. The term pathobiont is used to define a potentially pathogenic organism that causes diseases under a given set of circumstances such as Clostridioides difficile infection (CDI) [2,3].

There isn’t a clear understanding of the constituents of a ‘healthy microbiome’. It is estimated that the commensal bacteria have approximately 1,000-fold more genes than the human genome, and account for an estimated 1–3% of an average adults’ body mass [4]. Dysbiosis is defined as an ‘imbalance’ in the gut microbial community due to the gain or loss of community members or changes in relative abundance of microorganisms from the usual composition in a prior healthy state or as compared to a healthy individual. Dysbiosis has been associated with several conditions, such as CDI, inflammatory bowel disease (IBD), irritable bowel syndrome, asthma, atherosclerosis, obesity and metabolic disorders, asthma and allergic disorders, autism, amongst others [1,5].

2. Body

With the advent and understanding that microorganisms have a tremendous role in human health and disease; the National Institutes of Health announced the human microbiome project in 2007 [3]. There have been two phases of this project which have provided resources and methods for discoveries that link interactions between humans and their microbiomes to health-related outcomes [6].

The overarching goal of the human microbiome project has been to understand the range of genetic and physiological diversity of the microbiome along with the factors that influence the distribution and evolution of the constituent microbiota [3]. The mission of the human microbiome project is to utilize its data to identify new ways to determine microbiome related predisposition to diseases. An additional aim of the project has been to define the parameters needed to design, implement and monitor strategies for manipulating the microbiota with the hope to optimize its performance in the context of an individual’s physiology and disease state(s) [3]. The recently completed second phase of this project known as the integrative human microbiome project, comprised studies of dynamic changes in the microbiome and host under three different medical conditions namely: pregnancy and preterm birth; prediabetes and inflammatory bowel disease. The research in this second phase has begun to expound mechanisms of host–microbiome interactions. These studies provide unique data resources and have created a paradigm for future multi-omic studies of the human microbiome in different disease states [6].

Several important lessons and novel findings have been discovered from studying the microbiome with the two phases of the human microbiome project. Studies within the first phase of the project identified ecological niches in different organs such as the vagina, skin, gut and others have yielded nucleotide sequences of microbes from a large number of isolates to enable further research. Also, protocols to support reproducible body-wide sampling and data generation for microbiome studies were developed [7].

The first phase elucidated and confirmed notions that the host microbiome evolves throughout the lifespan of the host (Figure 1) [6]. These data demonstrated that every human being appeared to carry their own individualized composition of microbial strains. The initial gut bacteria, or ‘founder species’, in the newborn period are known to have low diversity and simple composition. A number of environmental factors, including maternal colonization, mode of delivery, host genetic make-up, diet, illness, cause dynamic shifts in the early microbiome [6,8]. The microbiome differs between populations and can persist in a state for years or undergo transition. The microbiome stabilizes over the course of the lifespan into a mature stable state. In healthy adults, the microbiome is highly diverse and well-differentiated [1]. The microbial diversity and composition vary in different ecological niches of the body with a greater diversity in the gut compared to other parts of the body.

Figure 1. The NIH Human Microbiome Project-1 (HMP1) focused on the characterization of microbial communities from numerous body sites (oral, nasal, vaginal, gut, and skin) in a baseline study of healthy adult subjects. The 2nd phase HMP2 expanded the initial findings to three longitudinal cohort studies of representative microbiome-associated conditions: pregnancy and preterm birth (vaginal microbiomes of pregnant women), inflammatory bowel diseases (gut microbiome) and prediabetes (gut and nasal microbiomes) [6]. Reprinted from Nature 2019 under Creative commons open access license [6].

An important finding of the first phase of the human microbiome project is that just the taxonomic composition of the microbiome in the absence of the knowledge of the microbial molecular function was not a sufficient correlate with a host physiologic or disease phenotype. With this finding in mind, studies in the second phase of the human microbiome project were designed to explore microbiome – host interactions in addition to the molecular activity of the microbes (Figure 1) [6].

A multi-omic microbiome pregnancy initiative research group characterized the microbiomes of pregnant women to study the effect of microbiome on risk of pre-term birth. Over 200,000 samples were collected from over 1,500 women and infants from different maternal and neonate sites. Analyses performed on a subset of these samples included not only 16S rRNA gene taxonomic analysis but also metagenomic and meta-transcriptomic sequencing with cytokine profiling, lipidomics analysis, and bacterial genome analyses. Women who experienced spontaneous pre-term births had lower abundance of Lactobacillus crispatus and a higher abundance of Sneathia amnii, Prevotella-related clades, a Lachnospiraceae taxon known as BVAB1, and a Saccharibacteria bacterium known as TM7-H1 [6,9]. An association of low vitamin D with these microbiome changes and pre-term birth was also seen [10].

The inflammatory bowel disease multi’omics database project enrolled and followed 132 individuals for one year in the second phase of the human microbiome project. Stool metagenomes, meta-transcriptomes, metaproteomes, viromes, metabolomes, host exomes, epigenomes, transcriptomes, and serological profiles were obtained and correlated with clinical activity over time [11]. The prospective nature of the sample collection and recruitment of patients both during both active and quiescent periods of disease allowed collection of longitudinal data. These data demonstrated that no metagenomic species were significantly different between samples from individuals with inflammatory bowel disease and those from controls. Depletion of obligate anaerobes including Faecalibacterium prausnitzii and Roseburia hominis and the enrichment of facultative anaerobes such as E. coli was seen in Crohn’s disease. Microbial compositions were dysbiotic during an inflammatory bowel disease flare and tended to revert to more baseline composition when disease was not active [11]. The next steps would be to form and validate models with better predictive biomarkers of inflammatory bowel disease progression and outcomes.

Diabetes mellitus type 2 is characterized by insulin resistance and a complex host–microbiome interactions. The integrated personal ’omics project followed 106 healthy and pre-diabetic individuals during periods of health, respiratory infections and other perturbations such as weight changes [12].

Subjects who were insulin-resistant had different microbial patterns at baseline from those who were insulin-sensitive. Individuals with weight changes showed thousands of specific molecular and microbial changes during these perturbations, and insulin-resistant and insulin-sensitive individuals responded very differently to perturbations. There were significant increases in microbes of the Verrucomicrobiaceae family (specifically Akkermansia muciniphila) in response to weight gain in the insulin-sensitive participants but not in those who were insulin-resistant [12].

Both the phases of the human microbiome project have led to a wealth of information with an estimated 42 terabytes of multi-omic data. These data are available at the data coordination center for the National Institutes of Health (http://ihmpdcc.org). Other resources including the database of genotypes and phenotypes (https://www.ncbi.nlm.nih.gov/gap/), the sequence read archive (https://www.ncbi.nlm.nih.gov/sra) and the metabolomics workbench (https://www.metabolomicsworkbench.org/) are available. Data models and the associated metadata produced by both phases of the human microbiome project are available freely at https://github.com/ihmpdcc/osdf-schemas.

An enhanced understanding of the microbiome has translated to therapies, several of which have been in clinical trials. Fecal microbiota transplantation has been shown to be efficacious for prevention of recurrent CDI with success rates over 80% [13]. Standardized microbiome restoration therapies for CDI are now in clinical trials and have shown promising results [14]. These therapies have shown to be promising for other infectious indications including multi-drug resistant infections and recurrent multi-drug resistant urinary tract infections [15,16]. Similar therapies have been in clinical trials and studies for other indications including IBD, autism, obesity, diabetes mellitus, nonalcoholic fatty liver disease and primary sclerosing cholangitis with variable results [17].

The results of the studies done as a part of the human microbiome project demonstrate that the gut microbes are an integral component of human physiology and disease pathogenesis. There is tremendous variability amongst individuals and larger studies with more diverse populations, and longitudinal measurements of both the microbiome and the metabolome are needed to develop models to validate causative associations. Understanding these causative associations will gives us the opportunities to translate to therapeutic options with whole microbiome restoration such as fecal microbiota transplantation, therapies with defined bacterial consortia and perhaps single strain bacteria

Declaration of interest

S Khanna receives research grants from Rebiotix, Inc, personal fees from Shire Plc, Premier Inc, Facile Therapeutics, ProbioTech Inc. The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

Reviewer disclosures

Peer reviewers on this manuscript have no relevant financial or other relationships to disclose.

Additional information

Funding

This paper was not funded.