Life seems to be ruled by chance: where we are born, what we become, who we meet, how we die. Novozymes Prize winner Gunnar von Heijne has devoted his life to bringing order into the chaos life constantly seems to create. He has become an expert in predicting how life unfolds – and how it folds. He is a pioneer in developing biological prediction methods.
Proteins are the building blocks of all life. Most proteins fold correctly – but when they misfold, diseases can evolve. Understanding how proteins fold and are transported to the right destination inside the cells is crucial to understanding life. Particularly important is how the proteins are embedded in the cell membranes, where they serve as ion channels, pumps and receptors – ensuring that the cells get the right signals from the outside and are maintained sustainably on the inside.
“About half the drugs we use that you can buy in a pharmacy actually target membrane proteins. Of the total number of proteins, maybe 30% are membrane proteins, but more than 50% of the drugs target the membrane proteins, so they are disproportionately important.”
Gunnar von Heijne’s very theoretical approach in often highly experimentally oriented biology has created great revolutions, but why this career path? Chance, he claims. His interest in science arose very early. He grew up in the 1960s – the heydays of science and technology with a space race, computers starting to be developed and much happening in science. Most people were excited about science and technology.
“I remember getting a couple of Christmas books; one was a popular science picture book about mathematics, and I still remember a picture of a French mathematician lying on his back in his office on a couch and looking at the ceiling and thinking deep thoughts about mathematics. That must have made a great impression on me. I must have thought that’s how to spend my life.”
He also got another Christmas book about rockets and space, but the lying-on-the-back image triggered his interest in science. This interest got his chemistry teacher to give von Heijne a stipend in high school to attend the Berzelius Days, an annual event of the Swedish Chemical Society, but spectacular and colourful experiments did not excite him.
“The only thing I remember was a booklet about quantum chemistry, really theoretical chemistry, that I don’t think I had been exposed to in high school, and I thought it was really neat that you could calculate things about molecules. I felt attracted to the theoretical aspects from the very beginning, but I had a better chemistry teacher. So, I was kind of a cryptochemist.”
Although theoretical physics fascinated von Heijne, the inspiring teacher got him to choose the more practical chemistry studies at the Royal Institute of Technology in Stockholm. But when he had to choose a topic for his MSc project and asked his teacher Stig Ljunggren for advice, fate determined that this chemist had a physicist hidden inside.
“I’ve done this microwave spectroscopy all my life, but we measured it, and you can only do this on very simple molecules, and we measured all the molecules you can do it on. So, this has no future. So don’t come to me to do your master project. But instead, the guys who really know the game are the theoretical physicists, so get in touch with the physics department and see if they have anything that you can do.”
However, the theoretical physicists had no great interest in chemistry. Fortunately, a young professor with a certain interest in “chemistry or biology or something” had just been hired. Clas Blomberg later also became von Heijne’s PhD supervisor. However, Blomberg would not tell von Heijne which project to choose.
“He said: ‘You have to find your own project and your own problems to work on.’ Which is totally crazy: how can you just set a PhD student loose like that? But that is how it was in that department, and for me it was great, because we were a couple of PhD students in that group, and we were all shopping around for projects, reading papers and talking: ‘Maybe we can calculate this, maybe we can make a model of that.’”
Picked up by a Nobel laureate
The inspiration for what would become Gunnar von Heijne’s lifelong interest came by serendipity. He had a very good French teacher in high school and because he liked French he took an evening class during his PhD studies and subscribed to a popular French science magazine: La Recherche.
“One day I happened to read a little article, a one-pager, about the signal hypothesis. There was a little cartoon, and it didn’t make sense to me because it had some strange interactions between a protein chain and a membrane that I thought couldn’t be right. So I looked up the original paper and read a bit more and realized that here is another thing that maybe we can make a simple theoretical model of.”
The cartoon was from a paper that is now very famous and that later led to Günter Blobel winning a Nobel Prize. In the paper, Günter Blobel and Bernhard Dobberstein showed the first aspects of how proteins are secreted from a cell. As it turned out, von Heijne was the first to try to model this; Günter Blobel picked up on his paper and von Heijne was invited to present the model at a meeting in the United States.
“Everybody was there. This was when the field started, and the people there were all young scientists who drove this field for the next 20 years. I had never been to anything like it before. Everyone had 10–15 minutes to present and no time for introductions. So, it was super intense and super exciting and, of course, you know about the American way of creating excitement. So this really opened a completely new world for me.”
Big in pharma
The speculation at that time was that the secreted proteins all start with a short sequence, 20–25 amino acids at the beginning, that serves as an address label and thus tells the cell if, for example, the protein should be secreted from the cell. After the meeting in the United States, von Heijne tried to improve his models but could not do any experiments, so he felt stuck.
“But again, I was lucky. People were starting to sequence DNA then. I had the idea that I could collect some of these sequences and compare them and see whether they have anything in common that would define what a molecular address label looks like. So I did this: I collected maybe 20–30 of these sequences and compared them.”
What emerged was a canonical design of this little address label in the beginning of these proteins.
“If you count from the cleavage site, from the side where this peptide is cleaved from the protein, we called the amino acid right before the cleavage site the ‘minus one’ position, and the one two positions over the ‘minus three’ position. These two invariably turned out to be small amino acids.”
Together with Henrik Nielsen and Søren Brunak at the Technical University of Denmark, von Heijne eventually developed more advanced prediction methods that not only predict cleavage sites but also identify entire segments as being signal peptides. Today all the big biotechnology and pharmaceutical companies have built these signal peptide prediction tools into their own sequence analysis pipelines. This knowledge is essential in producing protein-based medical products on a large scale using microorganisms.
The positive-inside rule
The next step was natural for von Heijne, the theoretician. While looking at this very early part of the protein – the signal peptide – von Heijne would also look at the rest of the protein. Many proteins that have signal peptides are membrane proteins, meaning that they must have a signal peptide that targets them to the cell membrane.
“Instead of being completely secreted across the membrane, segments of the protein become inserted into the membrane. By just looking beyond the signal sequence and looking at other parts of the protein, I started to see that membrane proteins have segments of very non-polar, hydrophobic amino acids that eventually end up spanning the membrane.”
The membrane-spanning segments fold up as helices inside the double-layered membrane. But between the helix segments, the proteins have stretches that form loops – either outside or inside the cell. This time, von Heijne’s ability to predict led to more material for biochemistry textbooks.
“I think I originally looked at 10–15 proteins. But even this small number showed that the inside loops have a very different amino acid composition than the outside loops. In particular, what differs is that positively charged amino acids are much more abundant in the inside loops. So this became known as the positive-inside rule.”
Gunnar von Heijne’s positive-inside rule for the first time described which way a protein would face when it is inserted into the membrane: which parts would remain facing inside and which parts would be transported across the membrane to face the outside of the cell. Like the signal peptide predictions, this rule turned out to have lasting effects, since it enabled both science and industry to predict the structure of membrane proteins and thus be able to engineer new ones.
Ad on a wall
Gunnar von Heijne’s career then took a new and unexpected turn. In 1995, while his children were still young enough to go abroad for a semester, von Heijne and his wife spent a winter in Los Angeles visiting Bill Wickner’s laboratory at UCLA.
When I arrived with my laptop, he looked a little wary and suggested that I do something different for a change. He suggested I get my hands dirty in the lab. I decided to test the role of positively charged residues for membrane protein topology.
Importantly, the experiments showed that membrane proteins can be turned on their head in the membrane by relocating positively charged residues. More importantly, this experience moved von Heijne’s research in a direction no one could have foreseen – even 6 months earlier. He became an experimentalist. The trip also brought about another piece of luck.
“At a group meeting in Bill Wickner’s laboratory, I glanced at the wall and saw a little ad from Henrik Garoff, who had just become a professor of molecular biology at the Karolinska Institutet in Stockholm. He was advertising for faculty to join his department. And I thought, maybe this is my chance.”
Inside out – outside in
Gunnar von Heijne sent a fax to Henrik Garoff asking whether there was any chance he could do some experiments in his department. The next day he got a fax: “When do you want to come?” Soon after that, von Heijne moved to the Karolinska Institutet and immediately started setting up a wet lab.
“That was one of the best pieces of luck I have had throughout my career. It’s great to do theory and it’s great to do prediction and you can find things by theoretical studies, but to actually see things work in the real world and to design an experiment and see the system behave as you thought it might behave. Or perhaps even more interesting, if it doesn’t behave as you thought it would behave.”
At Karolinska, von Heijne did many experiments to determine the fine details of the positive-inside rule and how the interplay between hydrophobic and charged regions affects the folding of membrane proteins. Some years later, he became a professor at Stockholm University and soon after that, he stumbled across a very interesting aspect of membrane proteins.
“Normally you would think that one part is always on the outside, while another is always inside, since the outside part has a different function than the inside part. But we stumbled across a class of membrane proteins that randomly inserts either way. How could, and why would, nature design a protein of that kind?”
Initially, von Heijne thought that dual typology proteins were an exception, but in a series of landmark publications, he and his colleagues showed that dual typology proteins are actually very common and especially among proteins that can transport small molecules back and forward across the membrane in and out of the cell. For example, cells use this transport mechanism to neutralize the effect of toxic compounds.
The perfect system
Gunnar von Heijne’s new mix of basic theory with experiments has turned out to be a potent one that has solved longstanding quandaries and invented new standards in the field. At a meeting at the University of Illinois Urbana-Champaign, von Heijne walked back to the hotel with his colleague Steve White.
He had developed a new hydrophobicity scale for amino acids, where they were listed according to their tendency to stay in the membrane versus staying in the aqueous phase outside the membrane. What he needed was a way to measure inside the cell instead of in a test tube with purified proteins.
When White explained this, von Heijne realized that they had already developed the perfect system. This spawned long-term collaboration between them to construct a new biological hydrophobicity scale.
“This has given chemists invaluable insight into what it takes to make a segment of amino acids insert into a membrane – knowledge that is important in designing new enzymes for bioengineering and biomedical purposes.”
However, once more – by chance – von Heijne changed his focus to start working on soluble proteins. He read a paper – this time by a group in Kyoto, Japan led by Koreaki Ito - describing short segments of a protein that have an uncanny ability to glue themselves into the tunnel of the protein production unit – the ribosome.
“So these peptides fit into little pockets in the tunnel, and they basically hold on to the tunnel walls and prevent further elongation of the protein chain. They arrest translation, leading to their name: arrest peptides. It was speculated that these arrest peptides might be pulled out of the ribosome by some external force and allow protein synthesis to continue.”
Gunnar von Heijne’s idea was to use these arrest peptides as force sensors: little devices that can be implanted in a protein that essentially measure forces acting on the protein chain at the precise time when the arrest happens. When the force of the growing proteins pulls sufficiently, the arrest peptides will dislodge.
“And this works. Now if we put a little arrest peptide right after the protein, when the protein gets into a location where it can start to fold, the folding process itself starts to pull on the arrest peptide. So now we have a force-measuring device we can engineer into any protein and look at the forces acting inside the cell on proteins as they are coming out of the ribosome.”
No particular reason
Gunnar von Heijne’s journey from theoretical physics to molecular cell biology has pushed the whole field forward towards more physical thinking and more rationally designed experiments that can get new essential information from a very complex system.
“I think I was lucky because there weren’t that many people coming from my kind of background, so there was very little competition. Nobody else was crazy enough to do these theoretical things. Everybody was just working at the bench. Molecular cell biology is such a rich field that being able to think about the more basic sort of fundamental issues is a good thing.”
According to von Heijne, life is chaotic, and life events very often happen by chance, pushing you one way or another. Or you choose to go one way over another for no particular reason.
“Nevertheless, maybe from a deep psychological viewpoint, doing science is in a way trying to take control, trying to understand enough about the system that you can predict and control what’s going to happen. I don’t know if that’s anywhere near true, but it’s kind of an interesting dichotomy in life: being a scientist and also being a living person.”
Although von Heijne has spent his whole career predicting, he had not predicted that he would receive the Novozymes Prize. However, he knew from his many years as a Chair and Secretary of the Nobel Committee for Chemistry that prize recipients are often surprised when receiving the good news.
Typically, the news triggers one of three reactions. Some think it is a practical joke and take a while to be convinced. Then some think and may say that it was about time they got the prize. Finally, the members of the third and largest group are astonished and speechless that their colleagues have honoured them in this way. I belong to this last group and am deeply moved that researchers from outside my field have found me a worthy recipient of the Prize.
The 2018 Novozymes Prize was awarded to professor Gunnar von Heijne fra Stockholms University, who is director of the SciLifeLab National Cryo-EM Facility.