A research team from Kiel University, DESY, and the Center for Structural Systems Biology (CSSB) has reconstructed the common ancestor of two bacterial enzymes to better understand how related proteins can evolve into distinct forms with different functions. Led by Holger Sondermann, a professor at Kiel University and head of the Structural Microbiology group at DESY, the team recreated a protein that no longer exists in nature. Their findings, published in Science Advances, offer new insights into the evolutionary process by which proteins can gradually develop new functions over time. The two enzymes studied belong to a group called nucleases, which are proteins that break down specific molecules made of the building blocks of genetic material. One enzyme, called NrnC, breaks down small molecules made of two linked DNA or RNA building blocks. Another enzyme, diDNase, performs a similar function but is specialized in processing molecules made of two DNA building blocks. The researchers aimed to understand how these two enzymes, which have similar origins, came to perform such different functions. This study builds on years of research by the Sondermann group. Scientists compared proteins from various bacteria to identify which features remained unchanged over time and which evolved. Earlier work by Sofia Mortensen, a researcher in the group, revealed a set of proteins closely related to known enzymes but with different molecular targets. The team initially tried to explain these differences by making targeted changes to the proteins. However, these modifications often disrupted the proteins' stability or function, forcing the researchers to explore a new approach. Instead of focusing only on the proteins that exist today, the team looked back at their evolutionary history using a technique called ancestral sequence reconstruction. This method uses the genetic sequences of modern proteins and their evolutionary relationships to predict what their common ancestors might have looked like. Using this approach, the researchers reconstructed a protein sequence that may have existed long ago. They then synthesized this ancient protein in the lab and studied its properties. To analyze the structure of the reconstructed protein, the team used X-ray crystallography at DESY's PETRA III X-ray source, specifically the P11 beamline. This technique allows scientists to determine the three-dimensional structure of proteins. They also used facilities at CSSB, operated by the European Molecular Biology Laboratory (EMBL), to study the protein's biophysical properties. The results showed that the reconstructed ancestor had characteristics that were intermediate between the two modern enzymes. Over time, two distinct evolutionary paths emerged: one enzyme became more specialized in DNA, while the other evolved to handle both DNA and RNA. The study highlights that the development of new protein functions is not always the result of a single, dramatic change. Instead, gradual modifications over time can lead to significant functional differences. Mortensen led the research and characterized the enzyme families, analyzing their structures and functions. The team also collaborated with researchers at the National Center for Biotechnology Information (NCBI) at the US National Institutes of Health (NIH) to reconstruct the ancestral protein sequence. For the researchers, this study is a starting point for further exploration. The biological role of DNA dinucleotides—small molecules made of two linked DNA building blocks—remains largely unknown. The team plans to investigate their function in different bacteria and how their levels are regulated within cells. They are particularly interested in whether these enzymes help bacteria defend against viruses. While the findings could eventually open new paths in biomedical research, Sondermann notes that much work remains. "First, we need to understand what biological function these molecules actually have," he emphasizes.