Image courtesy of Malia Kuo.
Into the world of proteins
You, me, a worm, and a cow. What makes us different? Shape, size, personality, and innumerable other characteristics create visible differences, of course––but all of that is ultimately founded on the invisible world of proteins.
Proteins are the microscopic tools of life, each one serving a specific function related to communication, catalysis, structure, storage, and every other aspect of cellular business. By interacting with other proteins and biological molecules, proteins generate all of life’s characterizing features: birth, death, cognition, and reproduction, just to name a few.
We can imagine understanding an organism as a composite of its protein inventory. Almost all of its characteristics and behaviors depend on the type, number, and activity of its proteins. In fact, although we often think about evolution solely in terms of DNA mutations creating new characteristics, that relationship between DNA and tangible features depends entirely on proteins, meaning proteins play a crucial role as mediators of evolution.
A serendipitous encounter
Propelled by a protein-focused perspective, Yansheng Liu found himself watching Gunter Wagner’s presentation on the curious case of cow cancer at a seminar on Yale’s West Campus. Wagner, an evolutionary biologist, was interested in understanding cancer by comparing it between species. To understand why cows are less prone to cancer than humans, Wagner had been studying their gene expression—an indirect measure of protein levels—to draw connections between changes in expression and resistance to cancer.
Gene expression can be used to approximate protein levels because of the “central dogma” of molecular biology: a single sequence of DNA, the primary set of instructions, is transcribed into many copies of RNA, the same instructions in a slightly different language. The RNA is then read and used to build proteins—the final, functional product. Measuring gene expression normally means measuring RNA levels, and it had long been conveniently assumed that protein levels were proportional to RNA levels due to the central dogma.
But this is not always the case. Protein levels don’t always follow RNA levels, and Wagner was well aware of the long-standing debate about how well RNA and protein levels correlate. He knew the value of measuring protein levels themselves. However, at the time, there was no practical method to quantify protein levels across the whole collection of proteins in a cell. This lack of technology forced him and almost everyone else in the field to rely on RNA-based gene expression to approximate protein levels.
This is where Liu comes in. As a proteomicist or someone who studies the world of proteins, he, like Wagner, had a keen interest in the biodiversity of proteins. “I was so inspired by Gunter’s talk,” Liu said. “He’s using this very different angle to compare species to get a clue about human beings…and I quickly related [it] to some of my previous work about the diversity between human individuals and suggested that we could…cover different species [with proteomics].”
Liu wanted to go further than gene expression. Rather than approximate protein levels with gene expression, why not study biodiversity at the protein level itself? And for the first time, he brought the technology to answer this question. He had been part of a team building a tool to measure protein levels called DIA-MS, a variant of standard mass spectrometry that offers advantages in reproducibility and accuracy. With this method, he brought the tools and the experience to explore a new frontier with Wagner: investigating biodiversity not with approximations of protein levels but with direct measurements.
Where to go and what to do?
While Wagner’s original presentation focused on cancer, the two saw value in expanding their scope to perform an initial survey of protein-centric biodiversity across mammals. With the powerful ability to quantify complete protein profiles across species, Liu and Wagner were now faced with the difficult task of choosing what questions to ask.
To both, it was clear that they must provide an answer to the fundamental debate: how well does RNA-based gene expression correlate with protein levels? They could test the validity of an assumption that the field had been relying on for decades, now in multiple species.
Beyond this highly practical aspect of the RNA-protein relationship, they also wanted to investigate the evolutionary history of the RNA and protein profiles. Could these two intimately intertwined yet distinct bodies evolve together across the tree of life? Or do intervening mechanisms disrupt the tethers between the two, separating the evolution of RNA levels from protein levels? We know of many possible sources of disruption, such as those that affect RNA stability or modify or degrade protein independently of RNA levels. Even further, we must consider the most significant difference between proteins and their nucleic acid cousins, DNA and RNA: proteins are the final, functional product! They’re made to interact with molecules or other proteins; can these interactions selectively constrain or accelerate only protein evolution and not RNA?
Consistent with their interest in protein biodiversity, the researchers were also curious about how their answers to these previous questions might vary between different species and individuals of the same species. Liu had previously revealed significant variability in the protein profiles of different humans, and they now had the chance to extend this work across species.
Finally, they considered what unique insight they could gain from proteins as opposed to DNA or RNA. Liu expressed interest in phosphorylation sites—spots on proteins that can bind or release phosphate molecules to activate or deactivate the protein. These sites are responsible for complex signaling pathways that regulate everything from cell growth and death to movement and secretion, so they receive much attention for understanding cell regulation or designing protein-inhibiting drugs. Liu and Wagner now had the rare opportunity to catalog the biodiversity of phosphorylation sites based on actual proteins rather than DNA or RNA.
From fundamental questions of molecular biology to structural biochemistry to evolution, the new technologies and the unexpected collaboration between an evolutionary biologist and proteomicist thrust a new probe into previously murky waters of biology. With so much inbuilt potential, they had no reason to constrain themselves to one field of questions. “You have to use your biological intuition to understand what nature is trying to tell us here, and that’s fundamentally a creative process,” Wagner said. The world of proteins had much to tell about every field of biology, and Liu and Wagner were there to listen.
Discoveries from the protein world
From their venture, the team recorded an invaluable dataset of protein diversity amongst mammals. With this in hand, we can finally understand what differentiates you and me from cows on the protein level, with applications to tracking our evolutionary histories or informing medical research. They also used this data to determine that RNA and protein levels are moderately well correlated, though far from perfect. The good news is that the correlation does not invalidate decades of prior gene expression work, but it still sheds light on the importance of going directly to the source: proteins.
Further, they discovered correspondence between variability in RNA and protein levels, suggesting that despite the many possible disruptions, RNA and protein levels do tend to coevolve. The tethers between these two spheres of molecular biology overall remain strong, though the exact relationship is protein-function-dependent.
For example, some classes of proteins—such as those involved in protein degradation—show little variation in both RNA levels and protein levels, while other classes—such as those filling the extracellular region—feature high variation in both, contributing to increased evolvability. However, certain proteins defy this trend. Proteins involved in large protein complexes feature less protein variability than RNA variability because intimate dependence on other proteins pressures individual proteins to be less variable.
Overall, protein profiles were slightly more variable than their RNA counterparts, suggesting that evolution can occur on the scale of proteins rather than solely on DNA/RNA. Furthermore, Liu and Wagner discovered more variability amongst phosphorylation sites relative to protein profiles. This variability may reflect the evolutionary value of tightly regulating protein activity or of generating new cell signaling possibilities. The team also constructed a network of coevolution amongst phosphorylation sites, providing insight into the complex interactions between signaling pathways.
Finally, Liu and Wagner established that variability in protein profiles between species mirrors variability within species. In other words, if levels of a protein are highly variable between humans, they are also likely to be variable between humans and other species.
We have learned much about the protein world, but what do these results mean for biology on a broader scale? For one, they tell us that we can reasonably trust gene expression while still acknowledging that protein levels are more variable because of their functional role. The data also reveals significant diversity in protein profiles not only between species but between individuals as well. And most importantly, Liu and Wagner have opened doors for an incredible array of studies. Evolution can now be interpreted not just from scars in the genome but by studying the proteins themselves—the tools that perform the tasks that evolution evaluates. Biodiversity, disease, and cell biology can all be approached from a more protein-function-centric perspective. In essence, a new method of understanding biology awaits us.
Pondering the future of the field, Wagner concluded, “It’s still very difficult. It’s pioneering, but it’s clear that this is the direction it has to go in the long run.” Trekking through the protein world may remain challenging for years to come, but it promises to light the way toward a pivotal new understanding of diversity and evolution.
Ba, Q., Hei, Y., Dighe, A., Li, W., Maziarz, J., Pak, I., Wang, S., Wagner, G. P., & Liu, Y. (2022). Proteotype coevolution and quantitative diversity across 11 mammalian species. Science Advances, 8(36). https://doi.org/10.1126/sciadv.abn0756
Wagner, G. (2019, November 25). Can Cows Teach us how to beat Cancer Malignancy? Nature Ecology and Evolution. Retrieved October 12, 2022, from https://ecoevocommunity.nature.com/posts/56631-can-cows-can-teach-us-how-to-beat-cancer-malignancy