Raising new questions about the workings of natural selection, scientists have found protein structures whose complexity appears to serve no useful purpose, which runs counter to long-held beliefs that evolution always moves toward greater effectiveness. 

In a study published Wednesday in Nature, U.S. and Swedish researchers examined hundreds of homodimers — complexes made of two identical proteins — and found they have structures that can grow larger and more complex over evolutionary time despite achieving no increased functionality. 

The specific structures at issue are regions of the proteins that are hydrophobic and likely exist to shield vulnerable areas of the proteins from water, which could damage the molecules. The hydrophobic regions appear to increase in size and complexity even though the change makes no difference in protecting the proteins, according to the study. 

These bloated homodimers defy traditional explanations of natural selection by demonstrating that not all evolutionary complexity is functional, according to Joseph Thornton, a professor of human genetics and ecology and evolution at the University of Chicago and the senior author of the paper.

“The classic explanation is that elaborate structures must exist because they confer some functional benefit on the organism, so natural selection drives ever-increasing states of complexity,” Thornton said, noting the greater benefits of a complex eye over a simple eye. “But at the molecular level, we found that there are other simple mechanisms that drive the build-up of complexity.” 

One mechanism identified by the researchers is universally applicable to all living beings because it relies on how the basic genetic code works. Proteins are limited in the percentage of hydrophobic amino acids they can contain while remaining stable, but that limit is easily exceeded by how often mutations create hydrophobic amino acids and corresponding weak points — encouraging the useless complexes to stick around through evolution.

Proteins across nature can be overwhelmed in this way because more than 40% of codons — three-base genetic instructions for amino acids — encode hydrophobic amino acids, whereas the limit for proteins is far smaller. 

A remarkable aspect of the findings is that they weren't uncovered sooner, because the lack of corresponding functionality in the development of the structures is not uncommon, said Georg Hochberg, a microbiologist at the Max Planck Institute for Terrestrial Microbiology in Germany and the study’s lead author.

“I think people could have done this 15 years ago,” said Hochberg, who was affiliated with the University of Chicago at the time of the study. “It doesn’t seem like such a big jump looking back, but somehow no one had noticed this before.”

The lead author said he and his research team were studying interactions between different types of proteins when they stumbled across the “strange-looking structures” of a steroid hormone receptor protein that took the form of a homodimer. A mutated version of the receptor did not bond with a copy of itself but had an amino-acid “scar” not present in the two-protein complex.

After tracing the evolutionary history of the homodimer back about 450 million years and conducting a series of experiments, the researchers found that forming a homodimer did not create any functional benefit and instead covered the protein’s hydrophobic surfaces. The single-protein counterpart escaped the need of forming a homodimer by evolving the scar, which covered the weak point.

The team then compared 466 homodimers to the steroid hormone receptor and found that 83% had an interface that was more hydrophobic, implying that most same-protein pairs in their sample have developed useless complexes.

The authors said their research explains only why the useless complexes persist through evolution and not how they are first created. Hochberg declined to speculate on their origins but said he is pursuing the question in new research, alongside their consequences in some of life’s most crucial proteins.

The article, “A hydrophobic ratchet entrenches molecular complexes,” was published Dec. 9 in Nature.

The authors of the study were Georg Hochberg, Brian Metzger and Joseph Thornton, University of Chicago; Yang Liu and Arthur Laganowsky, Texas A&M University; and Erik Marklund, Uppsala University. The lead author was Georg Hochberg.