Proteins and protein coding genes increase in length

The article in PNAS demonstrates that the distributions of protein lengths and their corresponding genes follow log-normal distributions across the whole tree of life. To do this, 9,913 different proteomes and 33,627 genomes were analyzed. Log-normal distributions typically arise as a result of multiplicative processes. Following Ockham’s razor principle, the researchers modeled the evolution of gene length distributions as multiplicative stochastic processes. In fact, they modeled the action of all genetic operators combined in relation to sequence length. Starting from LUCA, i.e., the hypothesized last universal common ancestor from which the three domains of life — the Bacteria, the Archaea, and the Eukarya — originated, the researchers found both theoretically and observationally that the average gene lengths have evolved exponentially over evolutionary time across different species. Furthermore, they discovered a scaling-invariant mechanism of gene growth across the entire tree of life, where the variance directly depends on the mean protein length. By representing all the species captured in the 33,627 genomes, the team was able to observationally verify the predictions and, moreover, show that the average gene length is a very good surrogate for organismal complexity. In a pure exercise of quantitative biology, Dr. Bartolo Luque from the Polytechnic University of Madrid added: “From knowing the average length of protein-coding genes in a species, we can calculate the whole distribution of gene length within that species.”

When representing the evolution of the average protein lengths versus their corresponding gene lengths across different species, it is observed that they evolve simultaneously in prokaryotes, because there are almost no non-coding sequences in their genes. However, once the average gene length reaches 1,500 nucleotides, the proteins decouple from the multiplicative process of gene growth, and the average protein length stabilizes after the onset of the eukaryotic cell at about 500 amino acids in a clear threshold, marking the appearance of the eukaryotic cell. From that point onward, and unlike what happens with proteins, the average gene length continues to increase as it did in prokaryotes, due to the presence of non-coding sequences.

Algorithmic phase transition

A critical phenomena analysis then concluded that a phase transition, well studied in the physics of magnetic materials, occurred at a critical gene length of 1,500 nucleotides. This marked eukaryogenesis and divides the evolution of life into two distinct phases: a coding phase (Prokarya) and a non-coding phase (Eukarya). Additionally, characteristic phenomena of these transitions are observed, such as critical slowing down, where the system’s dynamics become trapped in many metastable states around the critical point. “This is corroborated in early protists and fungi,” said Dr. Fernando Ballesteros from the University of Valencia.

Moreover, “the phase transition was algorithmic,” added Professor Jordi Bascompte from the University of Zurich. In the coding phase, in a scenario close to LUCA, with short proteins, increasing the length of proteins and their corresponding genes was computationally simple. However, as the protein lengths grew, the search for longer proteins became unfeasible. This tension caused by genes that grew at the same rate as before while proteins could not was resolved continuously but abruptly with the incorporation of non-coding sequences into the genes. With this innovation, the algorithm for searching for new proteins rapidly reduced its computational complexity, becoming non-linear through the spliceosome and the nucleus, which separated transcription and splicing from translation. This happened at the critical point of phase transition, which this study dates to 2.6 billion years ago.

The study recently published in PNAS not only answers essential questions, but is interdisciplinary, combining computational biology, evolutionary biology, and physics. “It has the potential to interest a wide audience across many disciplines and serve as a foundation for other groups to explore different research avenues, such as energy or information theory,” emphasized Dr. Enrique Muro of the Institute of Organismic and Molecular Evolution at Mainz University. The eukaryotic cell, the most significant increase in complexity in the history of life’s evolution on Earth, emerged as a phase transition and unlocked the path toward other major transitions — such as multicellularity, sexuality, and sociability — that shaped life on our planet as we know it today.

Published on Friday in the International Journal of Obesity, the research is the first to test in humans the relationship between HFHS diets, particularly those high in refined sugar and saturated fat, and first-person spatial navigation. Spatial navigation is the ability to learn and remember a path from one location to another, a process that can approximate the health of the brain’s hippocampus.

Dr Dominic Tran from the Faculty of Science’s School of Psychology led the research, which found HFHS diets have a detrimental effect on some aspects of cognitive function. It is likely those effects centre on the hippocampus, the brain structure important for spatial navigation and memory formation, rather than acting across the entire brain.

“The good news is we think this is an easily reversible situation,” Dr Tran said. “Dietary changes can improve the health of the hippocampus, and therefore our ability to navigate our environment, such as when we’re exploring a new city or learning a new route home.”

The research team recruited 55 university students aged between 18 and 38. Each participant completed questionnaires capturing their intake of sugary and fatty foods. They also had their working memory tested in a number recall exercise, and their body mass index (BMI) recorded.

The experiment itself required participants to navigate a virtual reality maze and locate a treasure chest six times. The maze was surrounded by landmarks that participants could use to remember their route. Their starting point and the location of the treasure chest remained constant in each trial.

If participants found the treasure in less than four minutes, they continued to the next trial. If they failed to find the treasure in this time, they were teleported to its location and given 10 seconds to familiarise themselves with that location before the next trial.

A seventh trial removed the treasure chest from the virtual maze but asked participants to find and mark its former location based purely on memory. Those with lower levels of fat and sugar in their diets were able to pinpoint the location with a higher degree of accuracy than those who consumed these foods multiple times a week.

“After controlling for working memory and BMI, measured separately to the experiment, participants’ sugar and fat intake was a reliable predictor of performance in that final, seventh, test,” Dr Tran said.

Dr Tran said the results highlight the importance of making good dietary choices to maintain healthy brain function.

“We’ve long known eating too much refined sugar and saturated fat brings the risk of obesity, metabolic and cardiovascular disease, and certain cancers. We also know these unhealthy eating habits hasten the onset of age-related cognitive decline in middle age and older adults.

“This research gives us evidence that diet is important for brain health in early adulthood, a period when cognitive function is usually intact,” Dr Tran said.

Dr Tran said the sample group used in this research was not representative of the wider population, but the findings still apply more broadly.

“It’s likely our participants were a little healthier than the general population and we think, if our sample better represented the public, the impact of diet on spatial navigation would likely be even more pronounced.”

Dr Tran is a recipient of an Australian Research Council Discovery Early Career Research Award (DECRA).