THE WORLD WITHIN US: LIFE AS A SUPRAORGANISM
All organisms exist in the context of others, evolving together and forming symbiotic relationships. These interactions are not limited to different species sharing an external environment, but also take place within a single organism. Taking humans as an example, we harbour whole microbial communities (consisting of bacteria, viruses and fungi) within us, forming microbiomes on many surfaces of the body (e.g., skin, mucosa, gut). Given that they make up 90% of all cells and the majority of genetic material in the human body, it is no surprise that our microbial symbionts have remarkable effects on us, from aiding in digestion to playing a role in disease.
Recent years have been exciting for scientists seeking to unravel the microbial mysteries locked within us. The advent of next-generation sequencing and advanced bacterial culturing techniques has led to a tidal wave of microbiome-related research. IMMpress consulted Dr. Susan Robertson (see Box 1), a microbial ecologist working at the University of Toronto’s Department of Immunology, to get the inside scoop on what this research entails: How do we find out what functions the microbiome serves in a specific disease? How important is species abundance? How can we culture microbiomes to begin answering such questions?
CONTROLLING A COMPLEX MODEL
There are many approaches to studying the microbiome. As a first step, gaining a broad understanding of the microbial community in your organism through different culturing and sequencing techniques (see Box 2) can identify candidate key players and levels of variation among different treatment groups. This big-picture view can help you narrow down a hypothesis when comparing phenotypes. Next, you’ll want to define your host-related readouts: what are your cells, proteins or metabolites of interest, and how might this be connected to the microbiota? Considering their functions (or the gaps in our knowledge about them) can help you further refine your research aims.
Once you have a particular research question in mind, your experimental set up should be guided by that goal, whether you are developing an animal model or working directly with human samples. For example, you could use a defined community implanted into a germ-free mouse to control for baseline variability and subsequently track changes in a controlled manner. In a paper from 2012, Ze et al. made use of a defined community in vitro to study the role and function of a Firmicutes species, Ruminococcus bromii, found within the human gut microbiota. They discovered that it plays an essential role in breaking down dietary resistant starch, a major food source for gut microbes, which in turn provided necessary nutrients to co-cultured bacteria in vitro. On the other hand, for a more translational disease model aiming to be representative of the human state, a defined community may not be the best choice, as the human microbiota is highly variable. Since the microbiome varies considerably from one individual to another, a large sample size is also crucial. For mouse work, a minimum of 10 animals is ideal, but this number should always be increased when possible. It is also a good practice to use littermates to compare your different genotypes and treatment groups. Regardless of which organism’s microbiome you want to study, you will need to take steps to ensure that rigorous controls are in place to minimize confounding variability in your experimental designs.
|BOX 2: Sequencing the Microbiome|
|16S ribosomal DNA sequencing can give you an idea of the richness and relative abundance of different bacterial taxa, as well as the diversity and metabolic potential in your sample. The resolution is not perfect, though, as only bacteria above a certain threshold in abundance can be detected; obtaining information on specific strains and species is highly unlikely. By comparison, shotgun sequencing can give you more information as it covers all of the genomic DNA in your sample. If you culture isolates from your sample, you can sequence their genomes and get a better idea of which species and strains are present. By doing this, you are also expanding the knowledge of which bacteria may be present in the microbiome.While abundance is an interesting starting point, high abundance does not equate with functional importance. So-called “keystone” species are sparse in the community, but serve unique functions vital for its maintenance. The aforementioned Ruminococcus bromii is one such keystone species, whose essential role in digestion was only shown by performing controlled culture experiments.You can also investigate function at the community level by analyzing expression of certain genes in your sample; culturing different isolates will allow you to perform even more specific phenotyping and genotyping. For example, in Zitomersky et al.’s 2013 paper, the authors were able to culture their target species isolated from human gut biopsies on mucin-coated agar plates, as they knew the target was in close proximity to the mucosal layer and was likely to use mucin as a carbon source. From this, they were able to look at levels of immunomodulatory factors often produced by the species in question, and how these might change in the context of inflammatory bowel disease.|
CULTURING THE MICROBIOME IN VITRO
“Unculturable” was a label given to all bacterial species of the gut microbiome that were deemed too difficult or impossible to grow in vitro. After all, gut bacteria live in very complex, anaerobic environments, which are hard to replicate using traditional culturing methods. Most species that compose our current understanding of gut communities have only recently been cultured. In fact, there is now a shift from “unculturable” to the more optimistic label “as-yet uncultured”, as a variety of new techniques are becoming available to bring more gut microbes into the light. Anaerobic chambers, which are similar to an enclosed biosafety cabinet, but lacking oxygen, are useful in growing pure cultures of a bacterial isolate. Chemostat-based systems, on the other hand, are ideal when growing a community made up of different species. The chemostat is a fluid culture system with a constant input of growth medium and nutrients, and output of waste, very similar to our own gut – food goes in, waste goes out. Emma Allen-Vercoe at the University of Guelph is one of the leading pioneers in the field using chemostat technology with the so-called RoboGut system. Connected to this system is an intricate arrangement of robotics through which you can program the desired pH, oxygen levels, and temperature. The system allows you to grow a defined community for use in vivo, and you can uncover how this community may behave, for example, by modelling inflammation in vitro. The University of Toronto Host Microbiome Research Network now has its very own RoboGut.
Just having a RoboGut is not sufficient however, particularly when you want to culture unknown species. A critical factor in the success of your culture is the media input. Here, a number of different approaches can be taken. The empirical approach makes use of traditional media and trial and error, although this can be laborious when trying to grow an unknown target. The bioinformatics approach is more tailored and works by inferring which culture conditions may be required for a specific target based on its associated sequencing data. Using single-cell sequencing data, researchers estimate the ideal culture conditions based on the target’s genetic profile; for example, a gene for an enzyme involved in the metabolism of a certain nutrient may indicate that adding the nutrient to the culture is important for this bacteria’s growth. Lastly, “culturomics”, developed by Jean-Christophe Lagier et al. (2012), makes use of basic knowledge about growth requirements of the community as well as high throughput technology to test different cultures at the same time. The aim here is to culture as many community members as possible rather than a specific target.
WHAT TO MAKE OF EVERYTHING
In terms of analysis of your data, a solid foundation in statistics and some bioinformatics is valuable. There are a large number of different ways to look at the data and analyze it using multivariate analysis tools. Principle coordinates and principle components analyses are commonly used, but they do have the limitation that they assume linear data. Nonmetric multidimensional scaling on the other hand does not make these assumptions of your data and may be the most ecologically defensible. The beauty of the data you obtain in microbiome research is that you can explore it in a number of different ways to find interesting relations in the data.
Finding out the composition of the gut microbiota of a mouse is no longer limited to DNA fingerprinting and what will grow in your petri dish. With the advent of new technology, we are able to probe deeper into microbial communities and gain a more complete understanding of how they affect host biology and immune responses. Considering the complexity of the microbiome itself – easily on par with the complexity of the immune system – and then taking this into the context of the host, the number of unanswered questions and potential research methods are vast. With all the new techniques being developed, the current field is thrilling for any researcher who has had a taste of the microbiome and realized the potential lurking within it.
Latest posts by Elisabeth Foerster (see all)
- Inflammatory Bowel Disease & Inflammatory Bowel Syndrome - May 6, 2019
- Industrial Endeavours: An Interview with Dr. Daniel Chapman - December 10, 2018
- PharmaCosts: Drug Pricing and Access in Canada - July 20, 2018