|Dr. D. Allan Drummond in his office at the University of Chicago|
This month, we spoke to Dr. D. Allan Drummond, our Researcher of the Month (RotM), regarding his work about proteins. Dr. Allan has a Ph.D in Computation and Neural Systems and heads a team of scientists as Assistant Professor of Biochemistry and Molecular Biology as well as the Department of Human Genetics, University of Chicago. His team works on finding out the effects of errors in protein making mechanism in our cells and their effect in disorders such as Amyotrophic Lateral Sclerosis (ALS).
In this interview, Dr. Allan simplified for us his recent findings about tRNA modification that were recently published in PLoS Biology
CTS: What is the aim of studying tRNA modification? What information can we seek from studying tRNAs?
AD: We study tRNAs and their modifications to understand translation, a fundamental biological process shared by all life. Cells need proteins, the main molecular workers in biology. To make proteins, ribosomes in the cell translate the genetic information in messenger RNA using tRNAs. Modifications to tRNAs potentially change how translation works, with all sorts of consequences (to cellular health, and even to the evolution of genomes) so we’re interested in understanding that.
|English: Codon-anticodon pairing exemplified for a tRNA Ala (Photo credit: Wikipedia)|
tRNAs are interesting because they physically embody the genetic code that maps triplet codons to amino acids. tRNAs have a triplet anticodon on one end (like ‘GUU’) that recognizes the triplet codon in mRNA (like ‘AAC’), and at their other end, they’re charged with an amino acid which, most of the time, corresponds to the codon (in this case, asparagine).
I say “most of the time” because ribosomes make mistakes, sometimes allowing a tRNA to recognize a different codon than the one it’s supposed to. In our study, we’re interested in the errors that ribosomes make as they translate.
CTS: How has this study contributed to existing scientific knowledge?
|Drosophila melanogaster, fruit-fly. |
(Photo credit: Wikipedia)
Making the educated guess, based on our own study and previous work, that a tRNA modification is involved, we measure the tRNA modification levels for multiple species, and find that these levels change systematically. Then we show how just these modification level changes can be sufficient to explain the change in translational accuracy and genome changes. And we make a prediction, which the data support, that the changes in tRNA modification levels as flies develop from eggs into larva into pupae into adults should correspond with changes in the codons used in genes that are turned on during each developmental stage.
CTS: How was translational accuracy tested before Akashi's method was introduced?
AD: Not very cleanly! Akashi’s work was really a remarkable advance. Some papers had made hypotheses about how selection on translational accuracy should work, and tested those hypotheses. For example, you might guess that the more amino acids in a protein (the longer it is), the more likely an error is to occur before translation is complete, so there should be stronger selection to prevent errors in long proteins. The problem, here, is that long proteins and short proteins differ in all kinds of other ways: they can be produced at different levels in the cell, have different functions, and so on. What Akashi did was come up with a way to test for differences within a protein, by comparing certain codons to other codons in the same sequence. His method naturally controls for all these other differences, so it really was a profound advance. I’ve been using his method, or variations on it, for many years, and feel it’s one of the most satisfying methods developed in the study of molecular evolution. That’s one of the reasons we named our modified measure the “Akashi selection score,” to do a bit more than just credit his work.
CTS: You have mentioned that queuosine modification is dependent on the intake of queuine, which is again obtained from bacteria. Isn't there is back up mechanism for this? What happens if queuine is not available for some reasons?
AD: Short answer: we don’t know!
At many points in the paper, we remind the reader that while queuine is not made by eukaryotes, it doesn’t necessarily need to be obtained by eating. Animal guts are filled with bacteria--the gut microbiome--and these bacteria are constantly growing, dying, and shedding their contents into the gut, where they can be absorbed by the host. So it may be extremely difficult to deprive animals of queuine!
In the lab, it’s possible, however. In studies where animals have been entirely deprived of queuine/queuosine, by raising them in the absence of bacteria, the animals do pretty well, at least in the short term. That lack of a clear phenotype is one of the main reasons queuosine’s function remains unclear. Provide biologists with a clear phenotype, and they’ll go berserk until they’ve figured out what causes that phenotype. No clear phenotype = sad biologists = much slower progress.
That lack of a dramatic phenotype could be because queuosine modification make a big difference, but under conditions we haven’t checked yet, like starvation or severe stress or when being chased by large predators or by amorous potential mates.
Lack of a phenotype could also be just lack of human-measurable phenotype. Natural selection can see tiny differences that humans can’t. For example, if flies deprived of queuine (say, due to a mutation that prevents queuine absorption) produce 0.01% fewer offspring on average than wild-type (“normal”) flies, we’d never know; they look the same to us humans and our tools. It’s like being 0.01mm taller than your brother -- might be true, but whose ruler would you use? But in the wild, standard models tell us that queuine-deprived mutant fruit fly and its descendants would quite efficiently go extinct over evolutionary time, outcompeted by the wild-type flies. So seeing “nothing” in the lab does not mean there’s no biologically (= evolutionarily) important effect. That queuosine modification of tRNA is maintained in almost all eukaryotes indicates that there are, indeed, evolutionarily important effects. We’ve just got to find them.
And that’s the long answer to “what happens if queuine is not available”? Maybe something, maybe nothing in the short term, but maybe something that just looks like nothing given our scientific limitations.
CTS: We do not quite know the detailed role of queuosine in tRNA translation. How could we do that? Is your lab working this aspect too?
AD: You’re right. One of us (Tao Pan) is a hard-core RNA biologist who’s made a career out of studying tRNA, RNA modifications, and translational fidelity, so you can bet we’re going deeper. The deep mechanistic aspects of how queuosine alters translation will likely need to wait for the ribosome jockeys to get interested. And that’s one of my hopes, that our paper helps spur interest and new studies from other groups. I think it’s remarkable to see such a set of interlocking questions, spanning diet, the microbiome, speciation, translation, and genome evolution. There’s plenty of work to be done!
CTS: Moving away from the paper, in your personal opinion, do you think that other studies such as those involving cancer, Alzheimers etc. get more media attention , whereas fundamental studies do not?
AD: That’s an empirical question, not a matter of opinion. I think, of studies receiving media attention, more of them do tend to be focused on cancer and Alzheimer’s disease than basic science, but I also think that’s not quite what you’re asking.
The question I sense you’re asking is, should cancer/disease studies garner more media attention than fundamental studies? Or, does it bother me if the media doesn’t seize on what we’ve found and broadcast it to the world?
Short answer: not really. Broad media attention is not the goal, and doesn’t really help us or the science.
Media attention tends to follow studies that appeal to the consumers of those media. The Financial Times has few readers who would, or should, care about our results. The impact of what we’ve done on those readers’ lives for the next several years is probably zero.
However, other media channels exist. Publishing this work was my first experience sending a relatively high-profile paper into the social media world. It got plenty of attention, more than I ever expected! Most of this attention was on Twitter, from scientists sharing it amongst themselves. And that’s great, in my view; attention is just awareness at this stage, and serves primarily to ensure that people read and scrutinize what we’ve done, pull out the gems, locate the flaws, maybe even change the way they think about their research.
As a basic scientist, I crave attention 10-20 years from now. The dream paper, for me, is one that becomes an indispensable part of human knowledge because it turns out to be a pivotal insight, an influential line of thought spawning a whole field of inquiry, a robust result that holds up in the face of scrutiny, a useful advance. At this stage in science, I don’t think we can predict those features (pivotal-ness, influence, robustness, utility) when the paper comes out. Media attention for anything other than entertainment value is premature. We have to let the clock run and see what happens.
That said, I do love the idea that there’s a person out there -- a flight attendant for Lufthansa, say, or a human resources manager for a software firm in Bangalore -- who stumbles on our paper, downloads it (open access!), and sits there, on their break, pushing through the thickets of jargon and unfamiliar abbreviations, and grasps the outline of what we’ve done and what it might mean, and smiles in a way they haven’t ever.
Zaborske JM, DuMont VL, Wallace EW, Pan T, Aquadro CF, & Drummond DA (2014). A nutrient-driven tRNA modification alters translational fidelity and genome-wide protein coding across an animal genus. PLoS biology, 12 (12) PMID: 25489848