Jeffrey Perkel has been a scientific writer and editor since 2000. He holds a PhD in Cell and Molecular Biology from the University of Pennsylvania, and did postdoctoral work at the University of Pennsylvania and at Harvard Medical School.
Cat poop is not generally something to be saved. For Jonathan Eisen, professor of evolution and ecology and of medical microbiology and immunology at University of California, Davis, would ask that you do so anyway.
In May 2015, Eisen cofounded, with staff scientist Holly Ganz, “Kittybiome,” a crowdfunded effort described as: “The first kitty-zen science project using the latest DNA sequencing technologies to explore the microbes that live in and on kitties.”
The goal, according to the project’s Kickstarter page, is “to better understand how microbiomes differ among cats, whether those differences reveal insights into cat behavior and biology, and how the kitty microbiome depends on and may shape the health of your cat.” As Eisen puts it, “People send us their cat poop, and we sequence the microbiome.”
Sounds straightforward. But as Eisen explains, it really isn’t.
There are many steps on the road to microbiome analysis, and all of them can skew the resulting profiles.
Sample preparation and DNA extraction methods for the microbial inhabitants of cat feces are different from those used on, say, ocean water or soil; choose the wrong method, and a key group of organisms may simply disappear, or at least change in abundance (see, e.g., [8,9]). Sequencing library preparation methods can selectively enrich or deplete some sequences relative to others, and the analytical algorithms used to interpret the data can likewise introduce biases.
Microbiome sequencing, Eisen says, “is like going to an island with a vacuum cleaner that has a shredder in front of it and [then] slurping up a sample and getting … these tiny little bits out the other side and hoping that you can figure out what the abundance of different insects was. If your filter has something in it that sticks to dragonfly wings, they’re gone.”
Still, more researchers are determined to try. A recent state-of-the-field analysis in Nature Microbiology reported that between 2012 and 2014, federal funding agencies spent some $920 million on microbiome research in the United States alone, focusing heavily on human microbiota and animal models, and relatively little, for instance, on tools development [1]. In late 2015, researchers published in Science and Nature a call to action for large-scale, systematic surveys of the global microbial community [2,3].
Here, we review some of the newest tools and techniques available to drive such research.
Purifying microbiome genomes and transcriptomes
Sequencing is not the only tool microbiome researchers have at their disposal, Eisen says. But it is “incredibly useful.” Yet obtaining a pure sample of the microbiome can be difficult, says G. Brett Robb, scientific director for RNA and genome editing at New England Biolabs: Samples often are contaminated by host or other neighboring cells. “If you want to look at the DNA in the microbiome, in many samples it’s going to be swamped by things you don’t want to sequence.”
NEB’s NEBNext® Microbiome DNA Enrichment Kit circumvents that problem by using a methyl DNA-binding protein to selectively bind and filter out host genomic sequences. Bacterial (and organellar) genomes, Robb explains, generally lack the methylcytosine mark frequently found in eukaryotic genomic DNA. As a result, a simple purification step is sufficient to separate the two.
The kit also enables researchers to purify host DNA away from microbial (or organelle) contents, Robb notes. In one recent example, researchers at Washington University and the University of Notre Dame used this approach to sequence baboon DNA from fecal samples, a method the researchers dubbed FecalSeq [4].
NEB also has recently developed a complementary method for sequencing microbiome transcripts. Prokaryotic mRNAs are marked at their 5’ ends with a nucleotide triphosphate—a mark that is missing from most eukaryotic mRNAs and also from tRNA and rRNA. Using vaccinia capping enzyme (VCE) and a biotinylated form of GTP, the method—called Cappable-seq—tags triphosphorylated transcripts with a biotin tag, enabling researchers to isolate bacterial mRNA (or at least, its 5’ ends) and thus probe the microbial transcriptome [5].
According to Robb, Cappable-seq and the NEBNext kit are complementary. “Whereas the enrichment kit allows you to look at who’s there at a detailed level [in a sample], what Cappable-seq allows is to see what those microbes are actually doing, because it reads out what genes are turned on.” That said, researchers will have to wait to try Cappable-seq, as the reagents are not yet commercially available to run it; NEB sells VCE, but the biotinylated GTP is still in development, Robb says. Commercialization is anticipated “within the year” states Robb.
Informatics tools
After microbial nucleic acids have been isolated and sequenced, the challenge is interpreting the results. In general, researchers have two options. Using 16S ribosomal gene sequences, they can compile a census of the microbial taxathat are present and their abundances—information that can be used to identify large-scale shifts in microbial diversity. Alternatively, they can interrogate whole-metagenome shotgun sequencing data to decipher the metabolic potential of a population and (potentially) reassemble their genomes.
The most popular tools for 16S rRNA analysis are open-source tools QIIME and Mothur, Esien says. Branded packages also exist, including Thermo Fisher Scientific’s Ion 16S Metagenomics Kit, a free web- and cloud-based package.
HUMAnN2, co-developed by Eric Franzosa, a research associate in the Huttenhower laboratory at Harvard School of Public Health, attempts to link phylogeny and function. The original HUMAnN was “a pipeline for efficiently and accurately profiling the presence/absence and abundance of microbial pathways in a community from metagenomic or metatranscriptomic sequencing data,” according to the HUMAnN2 website. In other words, given a set of DNA or RNA sequences from a microbial community, it works out what molecular functions that community is capable of. Powered by a growing set of microbial isolate genomes, HUMAnN2 expands on that by associating particular activities with specific organisms. “Increasingly, there’s a demand to know both,” says Franzosa—what functions a community can perform, and which organisms provide those features.
Another functional tool is MetaPath. Developed by Mihai Pop, a University of Maryland associate professor of computer science, MetaPath uses metagenomics data to work out how the metabolic activities a given community can perform change across conditions. In one test, for instance, Pop’s team used MetaPath to identify differences in fatty acid biosynthetic pathways in obese and lean individuals [7].
Community behavior
Though many microbiome studies use DNA sequencing, not all do. Pieter Dorrestein, a professor at Skaggs School of Pharmacy and Pharmaceutical Sciences at University of California, San Diego, probes microbiome communities using mass spectrometry.
According to Dorrestein, the microbiome field is undergoing a transition from microbial census-taking to understanding community function. “And the output of the community is chemistry.”
Dorrestein has a fleet of mass spectrometers dedicated to this work, from triple-quads and MALDI-TOFs to Orbitraps and FTICRs. In one recent study, his team combined quadrupole-time-of-flight and MALDI-TOF mass spectrometry with 16S ribosomal RNA sequencing to create a 3D map of microbes, metabolites, peptides and proteins on the skin of two individuals [6]. Each individual was sampled at 400 sites on the body, twice. In total, some 17 or 18 million spectral features were collected, all needing to be organized, visualized and annotated.
“What we did was molecular networking, providing an infrastructure to create a map of all detectable chemical space,” Dorrestein says.
Manipulating the microbiome
Still other researchers are working on methods to manipulate the microbiome from the inside out. Pamela Silver, a member of Harvard University Wyss Institute of Biologically Inspired Engineering, is one such researcher.
According to David Riglar, a postdoctoral fellow in Silver’s lab, the team has two basic goals, engineering synthetic microbial communities and developing microbial sensors. Riglar’s project focuses on creating microbial sensors capable of recognizing and reporting on inflammatory processes in the host gut.
Current diagnostic methods for monitoring the gut, Riglar explains, “are quite crude,” relying on either colonoscopy or fecal analysis. The former is highly invasive; in the latter, “there are very few biomarkers that will survive through the gut so you can test them in the feces.” Riglar’s synthetic sensing circuit represents an alternative strategy.
Using “commensal” (as opposed to laboratory) E. coli, he inserted two synthetic elements. The first is a promoter and reporter gene that remembers whether it is on or off and stably maintains that state; the second is a trigger (such as a ligand-activated transcription factor) that can respond to something in the environment, such as a metabolic indicator of inflammation.
As Riglar explains, the idea is simple: As the sensor bacteria pass through the gut, they essentially sniff for the metabolite in their environment. If it is present, the reporter switches on, inducing expression of a reporter, such as beta-galactosidase, that can be detected in the animal’s feces.
Ultimately, the team hopes to use such strains to report on the biology of the microbiome itself. Much of what researchers know about the microbiome, Riglar says, is limited to DNA content. “It’s difficult to know what’s happening functionally.” But with tools such as these, he continues, it should be possible to “eavesdrop” on the microbial community. “It’s a way of putting your sensor as close as possible to the source.”
References
[1] Stulberg, E, et al., “An assessment of US microbiome research,” Nature Microbiology, 1:15015, 2016.
[2] Alivisatos, AP, et al., “A unified initiative to harness Earth's microbiomes,” Science, 350:507-8, 2015. [PMID: 26511287]
[3] Dubilier, N, McFall-Ngai, M, Zhao, L, “Create a global microbiome effort,” Nature, 526:631-4, 2015. [PMID: 26511562]
[4] Chiou, K, Bergey, CM, “FecalSeq: Methylation-based enrichment for noninvasive population genomics from feces,” bioRxiv,2015.
[5] Ettwiller, L, et al., “A novel enrichment strategy reveals unprecedented number of novel transcription start sites at single base resolution in a model prokaryote and the gut microbiome,” BMC Genomics, 17:199, 2016. [PMID: 26951544]
[6] Bouslimani, A, et al., “Molecular cartography of the human skin surface in 3D,” PNAS, 112:E2120-9, 2015. [PMID: 25828778]
[7] Liu, B, Pop, M, “MetaPath: Identifying differentially abundant metabolic pathways in metagenomics datasets,” BMC Proceedings, 5(Suppl 2):S9, 2011. [PMID: 21554767]
[8] Morgan, JL, Darling, AE, Eisen, JA, “Metagenomic sequencing of an in vitro-simulated microbial community,” PLOS ONE, 5:e10209, 2010. [PMID: 20419134]
[9] Choo, JM, Leong, LEX., Rogers, GB, “Sample storage conditions significantly influence faecal microbiome profiles,” Scientific Reports, 5:16350, 2015. [PMID: 26572876]
Image: Shutterstock images