By Thiago Carvalho
“Antimicrobial peptides kill microbes. They are positively charged molecules and that attracts them to the negatively charged membranes of bacteria and fungi. They insert themselves into the membrane and create pores or they just rip it apart, a bit like a detergent. Some of them open a hole for others to stab the bacteria in the heart. But most of the time, just opening a hole does the job,” said Mark Hanson. Hanson has been interested in the innate immune response since his Master’s thesis work with Steve Perlman at the University of Victoria, in British Columbia. There he worked on the evolution of the immune system in a “weird, non-conventional fly species that has a nematode parasite and a defensive symbiont. We were curious about how the immune system adapted to this three-way interaction.”
Hanson became particularly interested in the function and evolution of antimicrobial peptides, a topic he then pursued in his Ph.D. at the EPFL in Switzerland with Bruno Lemaitre, where he continues to work as a postdoctoral fellow (he will soon move to the Longdon Viral Ecology and Evolution at the University of Exeter). This year, Hanson was the first-author of two papers reviewed through Review Commons and published in PLOS Pathogens (“The Drosophila Baramicin polypeptide gene protects against fungal infection”) and PLOS Genetics (“Repeated truncation of a modular antimicrobial peptide gene for neural context”). We asked him how the projects got started.
Mark Hanson: He (Lemaitre) brought me into this project that had already been in the lab for a couple of years, and I hit the ground running. Thanks to CRISPR, we were able to study these short peptides that no one had ever worked on before in a systematic way. We could systematically delete them and study their actual function in an organism instead of in a petri dish. We shifted the paradigm of what we understood about these peptides because their behaviour in a live organism is very different from what you see in a petri dish. We saw this remarkable specificity of a single antibiotic gene being the key determinant between life and death in defence against specific bacteria, specific pathogens. That work, characterizing the peptide antibiotic response, of fruit flies is what I’ve continued to this day and it’s the topic of the work we’ve submitted to Review Commons.
Review Commons: It’s amazing what you can find on the internet – I was reading your thesis. You started out studying a more obscure corner of the Drosophila genus and then you switched to Drosophila melanogaster. In your view, what are the relative strengths of the comparative approach and the model organism approach?
I think when you’re on the outside looking, you can be stunned by the value of a model organism and the tools and techniques that you can use. It opens up your way of thinking to ask questions that you never would when you are working on a non-model organism. At the same time, just working with a model organism can give you these false impressions about the universality of principles.
One thing that the coronavirus pandemic has brought to the fore is that we really have a very poor understanding of how different species respond to the same infection. That’s highlighted by the fact that, at the start of this, you would never have predicted that the natural reservoirs of a bat virus would be things like mink or North American deer or cats – but not dogs, and not mice unless you make human ACE2 receptor transgenic mice. We use this idea that phylogenetic relatedness is a good predictor of how a species is going to respond to a pathogen and that they are going to respond more similarly the more related they are – except when that breaks down, which it does very, very often.
Some of my work in Drosophila, including in this project, has taught me that different species respond with very different sensitivities to the same pathogens. That goes back to the original question of what is the value of model organisms and it’s that they’ve given us this framework that we can use to explore questions. But the value of non-model organisms, and this comparative approach, is that it tells you how universal that is. Without that framework, you wouldn’t be able to ask these questions in the non-models either. The non-models tell you more about the universality of your principles, but the models give you the framework, the skeleton that stabilizes the field.
Speaking of reference frameworks, you mentioned in a recent Twitter thread about your work on Drosocin that something fundamental had been missed because of problems with the reference gene annotation, leading to a “24-year-old mystery”.
It’s not that the reference genome was wrong. The reference genome is an N = 1 and so there are times when it’s the weird one, it’s the outlier. It’s the oddball. This is one of those cases, where it turns out that the reference genome has a rare allele of something that was segregating in the wild and that just happened to be fixed in the reference line that became the Drosophila melanogaster reference genome line. But the fly lines that people were working with, particularly in the immunity field – none of them had that allele. It wasn’t simply because we were working with a few lines. There were four or five standard genetic backgrounds, it’s just that none of them had that allele. Everyone had observed this peptide, which had a particular mass in proteomic spectra, but no one could map it back to the reference genome, because the reference genome didn’t encode the same allele of the peptide. We stumbled across it. We’d started doing this systematic mutation of immune response genes, these antibiotic encoding genes, and when we made one of the mutants, we noticed that this unidentified peptide disappeared from the blood after immune challenge.
Finding oddball peptides seems to be something you’re specializing in. How did you start working on Baramicin?
Baramicin (as we named it) is this effector gene that is part of the Drosophila immune response. I got interested in it because of a transcriptome that we had in flies under stress, and this gene came up as a big hit. That project didn’t go anywhere, but it got me to look at the gene and I realized that this was a really weird gene. It encodes a precursor protein that is chopped up into multiple pieces. We describe it as a bit like a protein-based operon.
It’s a tool that the genome uses to produce five different things – maybe more if you count their subpeptides – from a single promoter.
If you look at the original papers on what are the most concentrated peptides in the blood of Drosophila after infection, products from this one gene make up about one-third of all peptides seen after infection. It’s a huge element of the immune response that we knew nothing about. It seemed to us that if we want to understand the way that Drosophila fights infection, we need to know what is making up about a third of the peptides induced after infection, right?
That seems reasonable.
We then realized that one of our colleagues, Steve Wasserman and his group, had made mutations in this gene because they were also interested in it. We ended up collaborating with them to share the reagents and the workload and help solve the curiosity of this gene. Then we talked to another colleague and realized that they were working on it independently. That was Dominique Ferrandon and his group – they were also working on this gene, so it was a hot topic, apparently. We were very collaborative with his group as well.
You and your colleagues establish Baramicin’s role in specific immunity against fungi. But the final figure in the paper is very different from the others.
I did a natural infection with Aspergillus, where we just exposed the flies to fungal spores. In our hands, that strain is not a particularly strong pathogen, it doesn’t kill wild-type flies at all. When I went to check the flies, one hundred percent of Baramicin mutant males in the vial had this erect wing phenotype – and zero of the females. I thought that was weird, and I noted it down. The next experiment I did, with a pathogen challenging the Toll pathway, it happened again. There’s a supplemental table that goes with that final figure, and you can see that once I realized it, I kept track of this phenotype in every single experiment that I did.
That sets up a good transition to the second paper you submitted through Review Commons, which looks at antimicrobial peptides in a neural context.
After we observed this erect wing phenotype we thought there must be something else to this gene, beyond just being an immune antifungal. The first thing you need to know is that this was classically an immune gene. If you looked across species, each one that had the gene, Baramicin was always immune induced. But there are also duplicates, some species have extra copies of Baramicin, and in Drosophila melanogaster there are two extra copies – and neither of them is induced by infection anymore. We don’t know what the function is, but those genes also lose the antifungal regions in their coding sequence, they’re truncated. They have just this one domain with an unknown function, IM24. In the case of Baramicin-C, we saw that it was very specific to the hindgut and glial cells in the brain. With Baramicin-B, we saw an even more striking set of observations. If you suppress its expression with a neural driver then you get this severe mortality and developmental defects. The flies are clumsy and they fail to open their wings properly – which I think is probably a locomotor defect. When flies eclode from the pupa they are soft and squishy and they “manually” inflate their wings by pumping air through their body and that inflates the cuticle in a way that there is room for the body to move around inside that shell. I think they are so uncoordinated because of their developmental and locomotor defects that they can’t inflate their wings and the cuticle hardens before they get the job done.
“In our case, the refereed preprints were essentially the version that was submitted and accepted by the journals.”Mark Hanson
Those were two papers you published this year that were reviewed through Review Commons. How did you first hear about Review Commons?
That’s a good question… I think it might have been through my supervisor, Bruno Lemaitre, and through Twitter.
What did you think of the peer review process at Review Commons? Was it different from the traditional journal review process?
No. Honestly, if there was anything different, I’d say it was better because there isn’t this sense that the reviewers are trying to decide if the paper is a good fit for a specific journal. The reviews can just focus on the science instead of including all these side comments about the potential impact. What’s also kind of nice is that by doing it in this generalist way is that if you submit to a regular journal, a journal with a specific focus, they’re going to have a pool of editors that have that specific focus. If your manuscript is a cross-discipline paper it can make it harder for them to analyze it, and it can make it harder for them to recruit the relevant reviewers. By leaving that editorial decision to a larger pool of EMBO Press editors for Review Commons, I think that there is a broader pool of potential reviewers and potential editors that might be the best fit for each manuscript.
You’re the first first-author that we’ve spoken to in this series of interviews. You are now at the stage where you’re on the job market for a faculty position. In your experience, how are faculty recruitment committees looking at preprints?
I think they’re considered not as a full paper – and I don’t think they should be. But I think it’s nice to show that you have a mature product. You can do more than just say “in preparation” or “in review”. You can put that work out there and if they are curious, they can go and take a look. Even if they’re busy, and they just want to skim it, they can see the rigor of the work, the length, the scope. I think that preprints have become a required element of competition amongst junior scientists. There’s significant pressure to share your work via preprints. It means that you have additional things that you can put on your CV.
In your experience, were there major changes between the revised refereed version of the preprints and the final journal papers?
I’d say that, in our case, the refereed preprints were essentially the version that was submitted and accepted by the journals. There were formal issues of copyediting and formatting that were resolved, but the content was finalized at that point.
“To me, journals are helping to filter.”Mark Hanson
What do you think is the role of scientific journals in 2022?
I think that the journal has an essential role. The journal is this measure of rigor, quality, and assessment. Scientists, just like anyone else, can get caught in the hype that social media feeds on and it’s very tempting to take a paper’s claims at face value. I don’t think that the peer reviews attached to the refereed preprint are there for the general public. The referee reports are there for the very niche case of interested individuals. The journals and the editors have a key role in the quality assurance of the peer-review process – to make sure that when the authors answer the reviewers’ comments, they’ve answered them adequately. Like any human being, they can be wrong sometimes, but it’s this additional element of security to make sure that work that goes out is rigorous and accurate – and is not getting to where it is based on the sexiness of the phrasing of the title or because someone popular tweeted about it. With preprints, you have no way to separate the wheat from the chaff other than to devote your time to reading it. To me, journals are helping to filter where we believe that work situates itself in the publishing landscape and that is very helpful.