Cells depend on a host of carefully assembled protein complexes for survival. The origins of these complexes, and cells’ dependence on them, present something of an evolutionary puzzle: since every protein must be present for the machine to function at all, how could these complexes have been built over time by the gradual processes of evolution?
“These complexes were cobbled together from simpler ancestors during evolution by subtle tinkering, accidental gene duplications, and breaking rather than gaining new functions.”
Joseph W. Thornton
This issue – the so-called problem of “irreducible complexity” – has become a favorite argument for critics of evolution. But by carefully dissecting the history of one complex cellular machine, a team of scientists has revealed just how easily complexity in molecular machines can evolve. Their work is published online January 8, 2012, in the journal Nature.
“This work provides a clear and decisive account, supported by strong experimental evidence, for how a molecular machine evolved increased complexity through simple genetic processes,” says Howard Hughes Medical Institute early career scientist Joseph W. Thornton of the University of Oregon, one of the study’s lead scientists.
Thornton, along with University of Oregon collaborator Tom Stevens and two graduate students who came up with the idea for the project, focused on a small complex of proteins that pumps hydrogen ions across membranes within cells to maintain the proper level of acidity. The so-called vacuolar proton-ATPase (V-ATPase) is essential in all but the simplest of cells. For their experiments, the research took advantage of variations between species in a crucial part of the complex. Although each pump contains a ring of six molecules that physically transports the ions across membranes, the ring in animals and plants is assembled from two different kinds of proteins, whereas in fungi, three distinct proteins form the ring.
The researchers wanted to know how the fungal version of V-ATPase had evolved to be more elaborate. To address this question, the team recreated the ring proteins as they existed 800 million to 1 billion years ago, just before and just after the requirement for the three-protein ring evolved. They inferred the DNA sequences that encoded the ancestral proteins by gathering the sequences of 139 present-day ring proteins and used advanced statistical methods and a powerful computer cluster to trace evolution backwards along the tree of life. The researchers then synthesized genes for the ancestral protein sequences and introduced them into modern yeast cells. When the yeast cells expressed the ancient proteins, the team was able to study their functions.
They found that the third protein in the ring of fungi originated when one of the ancient proteins was duplicated, a common occurrence in cells due to errors during DNA copying. The pre-duplication ancestral proteins turned out to be more versatile than their descendants. When the scientists expressed the two ancient ring proteins in yeast whose own genes for all three descendant proteins had been deleted, function was restored. In contrast, all three of the resurrected genes from after the duplication were required, and each could compensate for the loss of only a single ring protein gene. The researchers concluded that the functions of the ancestral protein were partitioned among the duplicate copies: the complexity of the ring increased not because new functions were gained but because some of the ancestral functions were lost among each of its descendants.
Thornton’s team found an explanation for the evolution of the proteins’ specialization in their structures. The ancestral protein was able to occupy almost any position in the ring, but the two daughter proteins could fill only certain slots, because each one had lost different interfaces that the ancestor used to interact with some of the other ring components. Each duplicate could only occupy a specific position within the V-ATPase ring, and both became absolutely required for the complex to assemble properly.
Finally, Thornton and his colleagues sought the specific mutations that caused the duplicated ring proteins to lose some of their ancestors’ capacities. They re-introduced into the reconstructed ancestral protein each historical mutation from the period when the three-protein ring evolved and then repeated their yeast experiments. They found that a single mutation from the lineage leading to each duplicate was sufficient to compromise part of the ancestral protein’s functions and recapitulate the need for a three-protein ring.
“This kind of process – generalist ancestors whose physical interfaces degenerate to yield complexes of specialist proteins—is likely to be widespread in the evolution of molecular assemblies,” Thornton said. “Many molecular machines are composed of proteins in the same family, which means they were generated by gene duplication. And small mutations that lead to the loss of specific interactions with other proteins are very common.”
This is the first study of its kind, and additional studies of other molecular machines will be needed to test this hypothesis, Thornton said. But for now, the story of the V-ATPase illustrates the creative power of simple genetic processes in evolution.
“It turns out to be quite misleading to call these things molecular ‘machines,’ because they’re not precision-engineered devices in which everything does its part perfectly.” Thornton said. “These complexes were cobbled together from simpler ancestors during evolution by subtle tinkering, accidental gene duplications, and breaking rather than gaining new functions.”