Philipp Selenko

Philipp, you started your diploma-work at the IMP a quarter of a century ago.

Yes, but I never looked at it as a diploma thesis and TJ never made me feel as a junior member of the team. The group was small and we all had the same goal: To decipher the role(s) of SET-domain proteins in biology. In 1996, this was completely uncharted territory and we were all committed to changing that. We spent numerous nights and weekends at the IMP and Jo’s Disco effectively became our second home. The sofa was uncomfortable and way too small for catching some sleep in between experiments. Nonetheless, we happily shared this routine (and the sofa) with many other IMP night owls, of which there were many in those days.

As a result, everyone knew everyone, not only on a personal but also on a scientific level. We all discussed each other’s projects and planning as well as troubleshooting became a communal exercise. Those were great times and I learned so much about science and about being a scientist. It was probably right there, in Joe’s disco, in one of those nights, that I realized that this is exactly how I wanted to do science: passionate, committed, competitive and amongst like-minded people. The IMP way. I was hooked.

You are probably also still in touch with Thomas Jenuwein, your former supervisor. Are you still in touch with colleagues from your time at the IMP?

Yes, absolutely. With many of them and the friendships that formed in those years last until today. It’s quite amazing how quickly we reconnect when we meet up or run into each other by chance. The IMP spirit is alive right away and it often feels like no time has passed between evenings in Jo’s disco and now. Because the old IMP was rather small and crowded, these interactions happened very naturally. They supported the social glue that connected the community. Legendary parties, summer outings and winter ski trips added to that glue. I firmly believe that these social aspects played - and continue to play - an important role in the scientific success of the IMP.

How did you end up at the IMP in the first place?

I was a chemistry student, so I never actually studied the biological aspect of things. I worked as a part-time technician in a protein chemistry lab at the university to finance my studies, which really taught me a lot about biology at a time when most of my classes were on physical, theoretical and analytical chemistry. It was as a technician that I first heard about the IMP. Everyone in the lab I was working in praised the quality of the science that was done at the IMP and what a unique place it was within the academic landscape of Austria. So I was really intrigued to get to know it. I started as a summer intern, first with Denis Barlow and then with TJ. The rest is history as they say.

Can you briefly outline the project you did for your master's thesis? 

In 1996, TJ’s major emphasis was on genetics’ approaches rather than on protein chemistry, which was my background and, at that time, a rather rudimentary area of expertise. My task was to develop polyclonal antibodies against two SET domain proteins, SUV39H and EZH, and my limited experience in recombinant protein production and purification seemed like a good fit. When you generate such antibodies, there are long waiting periods in between individual immunisation routines and so I thought to use this time to explore whether SUV39H and EZH specifically interacted with other proteins in cells. At that time, the available literature on similar proteins with suspected chromatin-modifying functions was limited and it was rather straightforward to having read all the relevant papers. This turned out to be key in identifying a binding partner for SUV39H that was simply in the right size range of another protein that I had read about. Moreover, there already existed an antibody against this protein that I could use to confirm my hypothesis. TJ was fairly reluctant to ask for an aliquot of this antibody, to put it mildly, because, after all, what are the chances of finding ‘a needle in the haystack’ based on similar size? He was right, of course, but in this particular case, I was more stubborn … and got lucky. This is how the HP1/M31 – SUV39H interaction was discovered.

I probably learned the three most important lessons in my scientific career from this episode: One, always be on top of the available literature on a given subject because, two, it will enable you to make smarter decisions in your experiments. Three, there is always a fair amount of luck involved. However, chances of getting lucky are greater if you adhere to one and two. I still spend a crazy amount of time reading wide and broad, and I enforce the same in my team. Often in a manner of stubbornness that reminds me of TJ.

Your next decision was to move to Heidelberg, for a PhD at EMBL.

Yes, I always knew that I wanted to do a PhD in Structural Biology and to understand things at the highest possible level of resolution. At the same time, I was always attracted to the physics aspects of methods in Structural Biology, both from a theoretical and practical point of view, and the overall ‘exactness’ of biophysical measurements. I was also determined to pursue Nuclear Magnetic Resonance (NMR) spectroscopy because I loved the spin-physics behind it and enjoyed getting immediate results on protein samples in solution rather than having to wait for protein crystals to grow as is required for X-ray crystallography.  Back in those days, biomolecular NMR spectroscopy was not a particularly strong research area in Austria and so I decided to seek a PhD position abroad. After interviewing at many places - another hugely important aspect of decision-making processes in my opinion: to sample widely - I settled for the European Molecular Biology Laboratory (EMBL) in Heidelberg and joined the biomolecular NMR group of Michael Sattler in 1998.  Michael was 33 years old at that time and the youngest EMBL group leader ever. Considering that I had just consolidated my own stubbornness against TJ’s legendary one, my new boss was in for a ride. 

I learned the trade of NMR spectroscopy at EMBL and while I enjoyed the technical side of things, I often felt that the ‘bio’ aspect in biomolecular NMR fell a bit short. Being rooted in physics and with many protagonists being trained in physics, biology was often secondary to methodological aspects, especially in the inner circle of the community. Similar to the IMP, the EMBL was very much a biology-oriented institute and I enjoyed these complementary aspects a lot. However, I also realised that biophysical methods such as NMR are often detached from the genuine interests of biologists and that both disciplines often fail to find common grounds or to communicate in a way that both find engaging and interesting. In the early years of the millennium, there was a fairly perpetual divide between biology and physics, and I somehow felt that I wanted to bridge that gap in my PostDoc.

Ultimately, of course, the idea of running high-resolution NMR experiments on isotope-labeled proteins inside live cells was instigated by a screaming baby in a one-bedroom apartment in Boston, Massachusetts, were I eventually found myself as a PostDoc, husband and father.

We are now talking about your time in Gerhard Wagner’s lab at the Harvard Medical School. How was the transition to Boston?

We arrived in Boston in 2003 with a newborn baby and no money. Looking back, I don’t know how we managed but what I can say for sure is that old IMP (and EMBL) friends who had moved to New York and Boston earlier, kept us alive and sane. Their help was invaluable and I am forever grateful for their support.

We all know that having a baby changes everything. This is especially true for scientists who are not used to working fixed times or schedules. At Harvard, I was one of 25 other PostDocs and I felt that each one of them was double as smart as I was. Moreover, almost none of them had kids. Everyone was there to discover something great, of high impact, something that enabled us to start our own, independent careers. Despite this pressure, the atmosphere in my PostDoc lab was highly collegial. However, it was very clear that we all needed to be productive to secure a job and that we would likely compete for the same limited set of jobs in the field of biomolecular NMR spectroscopy. I needed to do something radically different. 

I started developing the idea of exploiting the NMR isotope-effect as a selective visualisation filter to detect specifically labelled proteins against the backdrop of all other, unlabelled components of native cells. Similar to visualising different fluorescently labelled proteins by microscopy, I wanted to use NMR to derive atomic-resolution structural information about proteins directly in live cells. Because NMR works at physiological temperatures and in solution, and because it only employs radio-frequency pulses for the excitation and detection of individual atomic nuclei in a non-destructive manner, I reasoned that it should be possible to monitor proteins over extended periods of time while they perform their cellular functions. Once again, I followed my three IMP rules. I read as much as I could find about the physical properties of intracellular environments, planned my experiments accordingly and hoped for a portion of good luck. Once again, it worked out.

Some other IMP benefits proved essential. I had settled on Xenopus laevis oocytes for my first proof-of-principle experiments, which I learned to manipulate with the help of Marc Kirschner, Tim Mitchison and Joan Ruderman. All three had strong ties to the IMP and it was wonderful to be able to rely on these connections for getting the project started.  

You moved back to Europe to set up your first independent group at the FMP in Berlin.

Yes, we moved to Berlin in 2008 and it took another eight years before we were able to routinely perform in-cell NMR measurements in cultured mammalian cells, including human cell lines, which had always been my ultimate goal. We are now advancing these efforts to more specialised cell types, for which we employ induced pluripotent stem cells and dedicated differentiation protocols. I am proud to say that we managed to combine cutting edge NMR technologies with advanced methods in cell biology to arrive at a wholistic understanding of cellular processes in health and disease.

Obviously, in-cell NMR spectroscopy has come a long way since you started.

When I started at Harvard, everything had to be developed from scratch and I had no protocols to follow, nor points of reference. By now, in-cell NMR spectroscopy is firmly established and many groups are using the method. Every big NMR meeting is hosting a dedicated in-cell NMR session, which is fantastic. Still, many physicists find cells even more intimidating than proteins, and I agree. Cellular complexity can be both mesmerizing and overwhelming.

Back to your lab at the Weizmann Institute: your team explores pathophysiological protein states that lead to diseases such as Parkinson's. 

At the moment, we are particularly interested in human neurodegenerative disorders that are, in essence, structural disorders. Whether it's Alzheimer's, Parkinson's, Huntington's disease, or ALS, in each of these disorders a different protein changes its structure into a toxic form of amyloid fibril, which causes problems in different cell types and regions of the brain. We are using in-cell NMR to look at initial stages of these disease processes and to answer why they are so often cell type-specific. In other words, we are aiming to uncover why certain cellular environments promote these pathological structural rearrangements whereas others don’t. Clearly, if we understand these first steps in their respective cellular contexts, we will be much better equipped to interfere with them.

The IMP recently celebrated its 25th anniversary and moved to a new building. Do you have any concluding remarks about its past, present and future?  

I follow the IMP’s developments very closely and consider myself a proud alumnus. Also, because many of my former EMBL PhD-student colleagues are currently group leaders at the institute or at IMBA, including Alex Stark, Manuel Zimmer, Julius Brennecke and Ulrich Elling. I was an IMP student when Max Birnstiel was director and Jan-Michael Peters and Jürgen Knoblich joined the institute as junior group leaders. I vividly remember their interview talks when they first laid out their scientific visions and, clearly, their careers and achievements pay tribute to the excellent research environment that the IMP represents. I witnessed how Kim Nasmyth and his team resolved the mysteries of meiosis and cytokinesis, after having deciphered the roles of cyclin-dependent protein kinases in mitosis and cell-cycle regulation. All of this while following Max Birnstiel as IMP director and consolidating the institute’s leading role in the international research community. Fantastic achievements on their own but truly stunning accomplishments when considered as a whole. I am inspired by Meinrad Busslinger’s success to continuously mesmerize the scientific community with surprising new findings about PAX transcription factors and at awe about recent developments in cellular and molecular imaging spearheaded at the institute.     

While many of these transitions into the highest academic echelons may be considered classical trajectories of brilliant scientists, there is one IMP career that I follow with the utmost respect: the one of Karl Mechtler. From the very beginning of handling the solid-phase peptide synthesis efforts at the IMP, to the early days of ‘electrospray’ mass spectrometry following a visit by Matthias Mann and Kim’s bold and visionary move to bestow Karl with the task of establishing this technology at the institute, to his current accomplishments in the field of single cell proteomics, which are nothing short of heroic. To me, Karl’s career exemplifies the essence of the IMP spirit: Scientific excellence above everything else. 

Interview by Heidi Hurtl, published in May 2021.