Researchers at the University of Cambridge have devised a method for labeling the proteome of specific tissues and at specific stages of an organism's development.
Detailed in a paper published this week in Nature Biotechnology, the approach – named SORT-M for stochastic orthogonal recording of translation with chemoselective modification – uses tRNA-synthetase/tRNA pairs to insert unnnatural amino acids throughout an organism's proteome. By combining this tRNA technique with tools for directing gene expression, the researchers were able to target these amino acids to specific organism tissues – in this case the ovaries of fruit flies.
A number of previous studies have used similar approaches to label target proteomes with unnatural amino acids. These amino acids – most often lysine derivatives – are similar enough to natural amino acids to be taken up by an organism's internal RNA machinery, but are modified to include functional groups allowing for their detection via approaches like fluorescence or click chemistry.
A primary drawback to such labeling systems, however, is that the unnatural amino acids must compete with natural amino acids for incorporation into the organism's proteins, and, therefore, to allow for efficient enough incorporation, organisms must be grown using minimal media under starvation conditions, which can alter their development.
"Most commonly used methods use tRNA and synthetases that are already in the cells, and they just add unnatural amino acids that basically compete with the natural amino acids for the same tRNA synthetase," said Ambra Bianco, an author on the paper and post-doc in the lab of Cambridge researcher Jason Chin, who led the effort.
The Cambridge team, on the other hand, used a tRNA and synthetase pair that operate orthogonally to the naturally existing system. By inserting this system into organisms of interest, they were able to label the proteome without disrupting its normal functioning, Bianco said.
"We can basically grow the cells in the very same environment that they would normally grow in," she told ProteoMonitor, adding that this was "especially relevant" for doing it in whole organisms as opposed to cell cultures.
"Embryonic development and going from the egg to the embryo is a very complicated process, and it cannot proceed very well on a minimal medium," she said. "So if we had to grow an animal on a minimal medium, it would be smaller and have very few progeny. Instead, we basically can do normal growth, which allows us to study all the normal development processes and incorporate unnatural amino acids without interfering in the growth and development of the animal."
By targeting expression of the synthetase to particular tissues and particular developmental stages, the researchers are able to activate this orthogonal system in specific proteomes of interest. In the case of the Nature Methods work, they targeted expression to fruit fly ovaries, demonstrating that they could specifically label the ovary proteome, allowing them to analyze these proteins without the need for dissection, which, Bianco said, in addition to being time consuming and tedious can also alter the physiology of a sample.
Fruit fly was an ideal model for the technique given the variety of tools already in place for targeting expression of the synthetase, she noted. Specifically, the researchers generated flies with their orthogonal synthetase, PylRS, on a promoter dependent on the transcription factor GAL4. A wide variety of fly lines have been developed each expressing GAL4 in specific tissues or at specific developmental stages, Bianco said, and by expressing the GAL4-dependent PylRS in various fly lines exhibiting different GAL4 expression patterns, the researchers can label different portions of the fly proteome.
While this GAL4 system has been best developed in fruit flies, it has also been used in other organisms including mice and works, Bianco noted.
"The advantage of flies is that it is very well established and so there are many tools that are ready, but the theory and technology could be used in other animals," she said.
In addition to targeting various tissues and stages of development, the approach is also amenable to inclusion of a variety of different functional groups. The Cambridge team focused primarily on incorporation of a dual luciferase reporter that allowed them to detect labeled proteins using fluorescence on a 2D gel. However, they also demonstrated the incorporation of tetrazine probes, which could be used for pulling down labeled proteins using click chemistry.
Another potential use, Bianco noted, could be for protein interaction work, where amino acids incorporating photo-crosslinkers could facilitate tissue specific interaction studies.
Although the researchers did not extensively investigate whether there was any bias to the process in terms of levels of amino acid incorporation across different types of proteins, Bianco said that an analysis of a random subset of proteins revealed no such bias.
The researchers also examined using mass spec whether labeling resulted in changes to the organism's proteome, finding, she said, that protein expression was the same in the wild type and labeled organisms.
In addition to fruit flies, the Cambridge team also demonstrated the technique in human HEK293 cells and Escherichia coli.
Currently, Bianco said, they are using the method primarily as a cataloging tool for mapping different tissues at different stages of development.
"We don't at the moment have a defined question that we are trying to answer," she said. "We're just trying to broaden out knowledge of the different proteomes of different tissues."