Humans host trillions of microorganisms in their guts, which collectively make up the gut microbiome. These microorganisms have proven to be essential modulators of human health. We commonly study the gut microbiome by identifying which bacteria are present in human fecal samples and determining which genes these bacteria have. However, different bacterial genes in these microbiomes are expressed to different levels, which largely determines the functions of these organisms. While we have been able to study which genes are transcribed into RNA in microbiomes, we have never been able to study which RNAs are translated into proteins. Unfortunately, technologies to study translation have never previously been adapted to work on microbiomes.
Ribosome profiling (Ribo-Seq) is a technology used to study translation by sequencing ribosome-bound RNAs (Ingolia et al, 2009). Ribo-Seq has been instrumental in studying translational regulation and predicting proteins, especially small proteins. Thus far, ribosome profiling has been limited to organisms that can be grown in bulk. As many bacteria can be difficult to grow from the microbiome and many require very specific growth conditions, most bacteria in the gut microbiome have not been ribosome profiled. Even if many bacteria were grown in bulk, performing Ribo-Seq on a large number of these organisms would be technically challenging. Additionally, ribosome profiling on bacteria grown in artificial conditions would not be representative of the proteins that are translated by these bacteria under natural conditions in the microbiome. Therefore, we aimed to perform ribosome profiling on fecal samples without isolating and growing bacteria one-by-one. To achieve this, we made key protocol modifications to ribosome profiling to create MetaRibo-Seq. MetaRibo-Seq fundamentally is ribosome profiling; however, it takes several measures to address challenges of ribosome profiling of fecal samples, such as fecal sample storage, sample purity, and RNA input requirements.
Once we developed and validated MetaRibo-Seq, we were eager to test it out on human fecal samples. One of our major goals was to see if we could use MetaRibo-Seq to determine which open reading frames within a metagenome were being translated. While most “normal length” genes are well annotated using traditional bioinformatic pipelines, we know that small open reading frames (less than 150 nucleotides in length) are often overlooked by annotation programs and are hard to identify. In recent work, we developed a bioinformatic approach to predict small open reading frames from metagenomes (Sberro et al, Cell, 2019), but for the most part, we did not have data to show that small proteins were being synthesized from these small genes. Through the development of MetaRibo-Seq, we were able to overcome this challenge and thus used MetaRibo-Seq to predict thousands of additional families of small proteins in the fecal microbiome. Looking ahead, it is now possible to study translation in microbiomes, a level of regulation we could not study before, as well as predict more genes using evidence of translation. We hope that this method aids researchers in their search for the myriad exciting functions of microbiomes.
The study can be found here.
Ingolia, N. T., Ghaemmaghami, S., Newman, J. R. S. & Weissman, J. S. Genome-wide analysis in vivo of translation with nucleotide resolution using ribosome profiling. Science 324, 218–223 (2009).
Sberro, H. et al. Large-Scale Analyses of Human Microbiomes Reveal Thousands of Small, Novel Genes. Cell 178, 1245-1259.e14 (2019).