Scientists have mimicked evolution in the laboratory for the first time using droplets of oil.
The researchers behind the study claim their work proves that a non-biological system composed of chemicals can be made to evolve.
The findings mark an important step towards creating synthetic life and may also help scientists to explain how the first biological cells appeared on Earth more than 3.6 billion years ago.
Scientists used a robot to find the ‘fittest’ droplets in experiments by selecting desirable properties like division (b), vibration (e) and movement (f) that over time led to the droplets becoming more stable
Evolution was long believed to be a process that only biological creatures were capable of, but recent research aimed towards creating synthetic life has begun to question that idea.
Professor Lee Cronin, regius chair of chemistry at the University of Glasgow who led the new research, used a robot to create tiny oil droplets from a mixture of four chemicals.
Each droplet was dropped into a petri dish of water and analysed for three different types of ‘fitness’ over the course of a minute using video cameras.
The robot then selected the droplet that performed best and the chemical composition of this was used to replicate the experiment, tweaking the mixture slightly each time.
Over the course of 21 generations, the oil droplets became more stable in the watery world in which they were being dropped.
This, according to Professor Cronin, mimicked the process of natural selection, which Charles Darwin proposed for driving evolution.
But rather than pressure from the environment and other species driving the evolution of the droplets, it was the robot.
Professor Cronin said: ‘This is the first time that an evolvable chemical system has existed outside of biology.
‘In recent years, we’ve learned a great deal about the process of biological evolution through computer simulations.
‘However, this research provides the possibility of new ways of looking at the origins of life as well as creating new simple chemical life forms.’
The robotic system initially created oil droplets at random using four chemicals, choosing from octanol, diethyl phthalate, pentanol and either octanoic acid or dodecane.
The researchers selected these chemicals with the aim of producing droplets that had a wide range of motility – the ability to move spontaneously and actively – stability, the ability to divide, viscosity and density.
In total they examined 225 unique combinations of these chemical by placing four drops of each mixture into a dish.
The robot was programmed to look for those that moved around the dish the most, the amount they vibrated in the water and their ability to divide over the course of one minute.
The best performing droplet was then selected and its chemical formula used as the basis of a second generation of droplets and again the ‘fittest’ droplet was selected to form the basis of the next generation.
The scientists adapted a 3D printer to create a robot that was able to mix and tweak the oil compounds before using a camera to select those that performed best in a series of tests in a Petri dish (Evo arena) full of water
They found that droplets without octanoic acid were best at dividing and moving around the dish but some octanoic acid is required for the best vibration response.
In total 21 generations of droplets were created and they became more stable, dividing into many, fast moving but distinct droplets.
The researchers believe their set up can be used to create chemical systems that will evolve freely without the need for the robot.
In their paper in Nature Communications, they said: ‘While most of the characteristics of ‘life’ are wholly dependent on the robot in the current system, we propose that these dependencies could be removed stepwise, in serial evolutionary experiments, to move the chemical system towards full autonomy, avoiding the all-or-nothing barrier that currently plagues the study of minimal chemical replicators.
‘It is therefore our intention to focus follow-up work towards the creation of artificial life-like assemblies, driven by a thesis that is chemically agnostic, thereby opening up investigations of new evolvable chemical systems.’
The findings could also help to explain how complex mixtures of chemicals on the early Earth came together to form the oily membranes found in every living biological cell.
Scientists believe these first formed capsules inside which the first biochemical processes began to evolve.
Four droplets of each mixture were deposited by the robot (top left) into a Petri dish (bottom left) and then monitored for stability in the water and to see whether they divided (bottom right), moved or vibrated
‘We are interested in exploring how life started on planet earth, which was almost certainly due to a large number of high probable events,’ said Professor Cronin.
‘But it probably took a long time – between 100 million and a billion years. So to speed up the ‘search’ we have used a cheap-3D printer platform and turned it into a chemical robot that can automatically do precise experiments.
‘By hacking together this kit we have in effect built a highly sophisticated machine that can fully automate a life cycle in the life of a chemical proto-cell model and used the robot to explore lots of different types of ingredients to try and come up with ‘interesting’ recipes that show ‘life-like’ behaviours, which include cell-division, movement, vibration, wobbling, clustering and so on.
‘We can show that statistically, the chances of droplet evolution happening at the origin of life is higher than a complete biological cell just ‘springing into existence’.
‘This means that the probability of life just spontaneously forming just went up dramatically. It is not an impossible miracle anymore.’
Professor Cronin added that the technique could also be adapted to help scientists develop new types of chemical compounds for use in industry and medicine.
He said: ‘Biological evolution has given rise to enormously complex and sophisticated forms of life, and our robot-driven form of evolution could have the potential to do something similar for chemical systems.
‘This initial phase of research has shown that the system we’ve designed is capable of facilitating an evolutionary process, so we could in the future create models to perform specific tasks, such as splitting, then seeking out other droplets and fusing with them.
‘We’re also keen to explore in future experiments how the emergence of unexpected features, functions and behaviours might be selected for.’