Groundbreaking study suggests that mutations driving evolution are informed by the genome, not random
Empirical evidence supports a new theory of how mutations arise
University of Haifa
A groundbreaking study published in this week’s issue of PNAS by scientists from Israel and Ghana shows that an evolutionarily significant mutation in the human APOL1 gene arises not randomly but more frequently where it is needed to prevent disease, fundamentally challenging the notion that evolution is driven by random mutations and tying the results to a new theory that, for the first time, offers a new concept for how mutations arise.
Implications for biology, medicine, computer science, and perhaps even our understanding of the origin of life itself, are potentially far reaching.
A random mutation is a genetic change whose chance of arising is unrelated to its usefulness. Only once these supposed accidents arise does natural selection vet them, sorting the beneficial from the harmful. For over a century, scientists have believed that a series of such accidents has built up over time, one by one, to create the diversity and splendor of life around us.
However, it has never been possible to examine directly whether mutations in the DNA originate at random or not. Mutations are rare events relative to the genome’s size, and technical limitations have prevented scientists from seeing the genome in enough detail to track individual mutations as they arise naturally. To overcome this, Prof. Adi Livnat of the University of Haifa, director of the Sagol Lab for Evolution Research, lead author Dr. Daniel Melamed and the team developed a new ultra-accurate detection method and recently applied it to the famous HbS mutation, which protects from malaria but causes sickle-cell anemia in homozygotes. Results showed that the HbS mutation did not arise at random, but emerged more frequently exactly in the gene and population where it was needed. Now, they report the same nonrandom pattern in a second mutation of evolutionary significance.
The new study examines the de novo origination of a mutation in the human APOL1 gene that protects against a form of trypanosomiasis, a disease that devastated central Africa in historical times and until recently has caused tens of thousands of deaths there per year, while increasing the risk of chronic kidney disease in people with two copies. If the APOL1 mutation arises by chance, it should arise at a similar rate in all populations, and only then spread under Trypanosoma pressure. However, if it is generated nonrandomly, it may actually arise more frequently where it is useful. Results supported the nonrandom pattern: the mutation arose much more frequently in sub-Saharan Africans, who have faced generations of endemic disease, compared to Europeans, who have not, and in the precise genomic location where it confers protection. “The new findings challenge the notion of random mutation fundamentally,” said Livnat.
From random mutation to natural simplification
Historically, there have been two basic theories for how evolution happens—(1) random mutation and natural selection, and (2) Lamarckism—the idea that an individual directly senses its environment and somehow changes its genes to fit it. Lamarckism has been unable to explain evolution in general, so biologists have concluded that mutations must be random.
Livnat’s new theory moves away from both of these concepts, proposing instead that two inextricable forces underlie evolution. While the well-known external force of natural selection ensures fitness, a previously unrecognized internal force operates inside the organism, putting together genetic information that has accumulated over generations in useful ways.
To illustrate, take fusion mutations, a type of mutation where two previously separate genes fuse to form a new gene. As for all mutations, it has been thought that fusions arise by accident: one day, a gene moves by error to another location and by chance fuses to another gene, once in a great while leading to a useful adaptation. But Livnat’s team has recently shown that genes do not fuse at random. Instead, genes that have evolved to be used together repeatedly over generations are the ones that are more likely to get fused. Because the genome folds in 3D space, bringing genes that work together to the same place at the same time in the nucleus with their chromatin open, molecular mechanisms fuse these genes rather than others. An interaction involving complex regulatory information that has gradually evolved over generations leads to a mutation that simplifies and “hardwires” it into the genome.
In the PNAS paper, they argue that fusions are a specific example of a more general and extensive internal force that applies across mutation types. Rather than local accidents arising at random locations in the genome disconnected from other genetic information, mutational processes put together multiple meaningful pieces of heritable information in many ways. Genes that evolved to interact tightly are more likely to be fused; single-letter RNA changes that evolved to occur repeatedly across generations via regulatory phenomena are more likely to be “hardwired” as point mutations into the DNA; genes that evolved to interact in incipient networks, each under its own regulation, are more likely to be invaded by the same transposable element that later becomes a master-switch of the network, streamlining regulation, and so on. Earlier mutations influence the origination of later ones, forming a vast network of influences over evolutionary time.
“Previous studies examined mutation rates as averages across genomic positions, masking the probabilities of individual mutations. But our studies suggest that, at the scale of individual mutations, each mutation has its own probability, and the causes and consequences of mutation are related,” says Livnat. “At each generation, mutations arise based on the information that has accumulated in the genome up to that time point, and those that survive become a part of that internal information.” This vast array of interconnected mutational activity gradually hones in over the generations on mutations relevant to the long-term pressures experienced, leading to long-term directed mutational responses to specific environmental pressures, such as the malaria and Trypanosoma–protective HbS and APOL1 mutations.
New genetic information arises in the first place, they argue, as a consequence of the fact that mutations simplify genetic regulation, hardwiring evolved biological interactions into ready-made units in the genome. This internal force of natural simplification, together with the external force of natural selection, act over evolutionary time like combined forces of parsimony and fit, generating co-optable elements that themselves have an inherent tendency to come together into new, emergent interactions. “Co-optable elements are generated by simplification under performance pressure, and then engage in emergent interactions—the source of innovation is at the system level,” said Livnat. “Understood in the proper timescale, an individual mutation does not arise at random nor does it invent anything in and of itself.”
Redefining how evolution works
The potential depth of evolution from this new perspective can be seen by examining other networks. For example, the gene fusion mechanism—where genes repeatedly used together across evolutionary time are more likely to be fused together by mutation—echoes chunking, one of the most basic principles of cognition and learning in the brain, where pieces of information that repeatedly co-occur are eventually chunked into a single unit. Yet fusions are only one instance of a broader principle: diverse mutational processes respond to accumulated information in the genome, combining it over generations into streamlined instructions. This view recasts mutations not as isolated accidents, but as meaningful events in a larger, long-term process.
The study was funded by the John Templeton Foundation, the Israel Science Foundation, and the Sagol Network through the Sagol Lab for Evolution Research. The opinions expressed in the publication are those of the authors and do not necessarily reflect the views of the John Templeton Foundation.
Journal
Proceedings of the National Academy of Sciences
Article Title
De novo rates of a Trypanosoma-resistant mutation in two human populations
