By Jo Marchant

Oil droplets that creep purposefully through their watery environment, metabolize fuel, sense their surroundings and perhaps even replicate--could these be precursors to life? That's the claim of a chemist with a controversial approach to modeling how Earth's first organisms scraped themselves together.

Theories about how life started range from fortuitous chemistry around hydrothermal vents on the sea floor to the delivery of precursor molecules from outer space. But there is little hope of finding geological evidence for this momentous event: Earth's crust is continuously being recycled, with the oldest known rocks dating to only 3.8 billion years ago. By that point, life was flourishing and relatively complex.

So another way to investigate what happened is to try repeating it--to build basic life forms, called protocells, in the lab. Attempts to do this have generally involved using stripped-down versions of biological cells, and have assumed that certain building blocks, such as RNA, are already present. But Martin Hanczyc at the University of Southern Denmark in Odense is looking for life-like behaviors somewhere much simpler: in drops of oil. He described his work at a Royal Society discussion on the origins of life, held in London on February 21.

Oily character

Hanczyc's first round of experiments used nitrobenzene oil. To give the droplets a "metabolism," he put them into a highly alkaline solution (pH 12) and fuelled them with a chemical called oleic anhydride, which converts to oleic acid on contact with water. This reaction lowered the pH at the boundary of droplets, creating an uneven surface tension that caused them to move autonomously through the liquid (see video). Meanwhile, convection inside the droplets brought fresh supplies of oleic anhydride to the surface.

The droplets can "sense" their environment--moving through a pH gradient to seek out the highest possible pH (In this video, the blue dye indicates a higher pH). And by putting the fuel in the water, with a chemical catalyst in the oil, the droplets can absorb fuel from their surroundings. "You get immortal droplets," says Hanczyc. "As long as you feed them, they keep moving."

The system bears more than a passing resemblance to salad dressing. But Hanczyc insists that it has the potential for great things, with a self-contained body, embedded metabolism and the ability to avoid equilibrium. "If the avoidance of equilibrium by a structure is the most fundamental prerequisite for life, then this model could be considered as a type of primitive life that could have been possible on the early Earth," he says.

Other types of behavior are possible too. The drops tend to circle each other without touching, which Hanczyc sees as evidence for rudimentary chemical communication: "They share a chemical language." In so far unpublished work, Hanczyc and his colleagues have also shown that the droplets' past actions can influence their future ones, which could be interpreted as a primitive form of memory.

Creating watery compartments within oil droplets would allow for more complex structures. And the team is working on creating instabilities within the droplets that cause them to self-divide--with subsequent feeding or fusion, that might lead to a primitive replication cycle.

To demonstrate how all this might have happened on early Earth, Hanczyc is now recreating his droplets using ingredients that would have been around when life started, including mineral oil--a mix of simple hydrocarbons called alkanes--and the simple organic compound hydrogen cyanide (HCN), which reacts with water to form biological precursors such as amino acids and nucleobases. These droplets show many of the same behaviors, he told the meeting.

Software needed

Reaction among delegates was mixed. Geologist Norm Sleep from Stanford University in California says that he isn't aware of any examples of such autonomous structures forming in geological systems, but doesn't rule out that it could happen.

The necessary organic molecules would have been around, he says, perhaps deep beneath the Earth's surface or on the


But biochemists argue that the lack of genetic information in the droplets means that they would never develop into anything more complex. "You need to put software into the hardware," says Philipp Holliger of the MRC Laboratory of Molecular Biology in Cambridge, UK, who recently suggested that before cells evolved, RNA could have replicated within liquid-filled microcompartments inside ice.

Hanczyc disagrees. "I think you will get quite complex structures," he says. "You don't have DNA or RNA, but the necessary information is embedded in the chemistry of the system." Characteristics would be passed to daughter droplets on division, though he concedes that without being formally encoded, these would be dependent on the environment, and could easily be lost.

Jack Szostak, a biologist at Harvard Medical School in Boston, Mass., is generally thought to be closest to building an artificial life form in the lab. He and his colleagues work with fatty-acid molecules that, in water, spontaneously form vesicles with membranes similar to those of biological cells, and support the replication of an added-in DNA template. Szostak told the Royal Society meeting that his protocells can grow and divide, and even compete with each other. For example, vesicles with phospholipids in their membranes--a key component of actual cell membranes--grow faster than those without.

He seems bemused by Hanczyc's oil-based approach. "It's an interesting, simple system," he says. "I don't see any relevance to the origin of life. But you never know."

Hanczyc, however, argues that oil-based life is worth taking seriously, if only as a reminder about the weird variety of forms that life might take. If not involved in our own life history, he argues, oil organisms could be alive alongside us, either buried deep on Earth, or elsewhere in the Solar System--for example on Titan, Saturn's largest moon, where HCN and other hydrocarbons are abundant.

"This behavior is very easy to find, just by throwing things into a pot," he says, pointing out that many biological reactions happen more easily in oil than in water. "We shouldn't be biased by the fact that we are all water-based organisms."