Research

RNAs can be distinguished by their nucleotide sequences, which confer diverse properties. There are tens to hundreds of thousands of RNA species, and for a long time we had to study them one by one. With the advent of next-generation sequencers, sequences can now be determined in massive parallel (hundreds of millions to billions of reads). This enables determination of RNA identity from sequence information and quantification of RNA abundance from read counts in genome-wide manner (transcriptome analysis).

Ribosome profiling (Ribo-Seq) leverages this capability to globally assess translation on mRNAs. In this approach, cell lysates are treated with RNases, which digest most RNA; however, ribosomes protect the region of mRNA they occupy. The protected fragments are sequenced, and the reads are mapped to the genome to identify where and how many ribosomes were bound, i.e., which mRNAs were being translated and to what extent.

We use Ribo-Seq as a core technology and integrate biochemistry and molecular biology with fluorescence microscopy to reveal how RNA-mediated regulation shapes cellular physiology. We aim for all lab members to become versatile researchers who can perform both wet-lab experiments and dry-computational analyses.

Where does translation occur within cells? In eukaryotes, mRNAs transcribed in the nucleus are exported to the cytoplasm and translated when encountering ribosomes. Is translation reaction evenly distributed throughout the cytoplasm?

Recent work has shown that mRNAs are transported to specific subcellular locations and undergo local translation. Local translation can efficiently localize proteins by transporting a small number of mRNA molecules and producing many protein molecules on site, rather than transporting proteins themselves. For example, translation occurs on the surface of mitochondria, and newly synthesized proteins are imported into mitochondria co-translationally. In elongated cells such as neurons, local translation at distal regions (e.g., axons) is important for cellular function, and dysregulation is linked to neurological diseases.

However, comprehensive measurement of local translation has been challenging, and information on which mRNAs are translated where and how much has been limited. We recently developed a general method, APEX-Ribo-Seq, and applied it to 14 organelles to build a large-scale dataset—a local translation atlas (preprint). Analyses of this dataset are beginning to reveal regulatory principles and biological significance that were not apparent from previous studies.

How is local translation controlled? And what happens when it is defective? We will ansser these questions, combining comprehensive approaches with microscopic analysis.

Producing proteins in the right amount at the right time is essential. If proteins are synthesized when they are not needed, they are not only wasteful but can also be detrimental to cells. Thus, after transcription, mRNAs must often be stored in a translationally repressed state until the appropriate timing.

A key mechanism thought to contribute to this storage is the formation of RNP granules—assemblies by the condensation of RNAs and proteins. Because these structures lack lipid membranes, they are also referred to as membraneless organelles. Many types of RNP granules exist in cells, with highly diverse compositions; some are observed only in disease contexts. Recent studies suggest that mRNAs may be sequestered within such assemblies to regulate translation.

How are mRNAs selectively retained or released? How is mRNA localization connected to translation? And what are the cellular consequences when this system fails? To address these questions, we will harness granule transcriptome analysis, combining biochemical purification/labeling of granules with RNA-Seq (review).

We actively collaborate using Ribo-Seq. Through collaborations, we have studied translational regulation in diverse biological contexts, including stress responses, RNA modifications, immunity, neuroscience/learning, development/differentiation, aging, and intraspecies variation. We also work with a wide range of organisms—not only human cultured cell lines but also iPS cells, budding yeast, mice and organoids, zebrafish, Drosophila, plants, and amoebae. (Once a lysate is prepared, the experimental workflow is largely shared.) We will continue to expand our collaborative network going forward.