Biology has been pretty successful at creating life, but now chemistry wants a crack at it

It's strange to think that in nearly all the places we know life exists, it is in the same form. Whether lurking in the dark of the deep ocean or business-suited and brandishing a briefcase, the basic unit of life is the same: the cell. Though the contents may vary slightly, cells make up everything from bacteria to bankers.

We're all composed of the same stuff - carbon, hydrogen and a few other elements bundled into neat little packages. All of biology is, in a sense, just the same old cells divided up between everything that lives on this planet.

Well, chemistry has a few things to say about that. There are those who would argue that life doesn't have to be quite so limited - cells, yes, but not as we know them. 'There is this movement, and it keeps coming up in the history of science, that says life is essentially a manifestation of some basic property of matter, and therefore given the right conditions, it will occur,' says chemist Steve Mann, whose research at the University of Bristol, UK, focuses on creating artificial cells and protocells. 'At the moment there's a sense of excitement that one could do it using alternative strategies. It doesn't mean that the thing's going to get out of the test tube and crawl around. We're talking about very minimal aspects of life.'

Starting from scratch

This isn't synthetic biology exactly; it's less about dissecting and reassembling what life has to offer, and more about assembling a new version of life altogether. There are, however, certain ground rules - and, like it or not, life has to be compartmentalised somehow. 

In single-celled organisms, the outer membrane separates the machinery of the cell from the outside environment, maintaining the right conditions for cell processes and reactions. Its function in multicellular organisms is more complex - it separates one cell from another, allowing specialisation and division of labour.

Demonstrating compartmentalisation of a basic form, Mann has developed a new type of cell model that lacks the phospholipid membranes common to natural cells.1 His team's 'microdroplets' appear as oil in an emulsion; they are drops of liquid that are kept from mixing with their watery environment by a kind of phase separation. The droplets, or protocells, contain oppositely charged nucleotides and peptides that attract each other to hold the compartments together. 

Because this approach circumvents the need for a membrane - and therefore the need for that membrane to assemble from complex molecules - it represents a possible scenario for the emergence of life on Earth. Although it's difficult to say how feasible this scenario really is, says Mann. 'I'm not so interested in making strong claims about the origin of life on the Earth,' he says. 'What I'm more interested in is how we realise minimal life in a laboratory and how we would use such a thing for technological development.' The trouble is defining 'minimal life'.

Minimal life

It's a question that others have been struggling with too, including Neal Devaraj at the University of California, San Diego in the US. He says the scientific community needs to think about what would truly constitute an artificial cell. Metabolism? Evolution? Compartmentalisation is somewhere near the top of the list of minimal requirements, and for his own part, Deveraj has been working on exactly the kind of membrane-bound structures that Mann shunned in his microdroplets work. 

In a study published in January this year, Devaraj and collaborator Itay Budin, at Harvard University in Boston, US, showed how artificial cell membranes can self-assemble from synthetic phospholipid components resembling those in real cells.2

The team's approach is only superficially biomimetic, as it uses a reaction that would never occur in nature. 'What we're trying to say is forget about what's prebiotically plausible,' says Devaraj. 'Just taking any kind of chemistry at our disposal, can we make something that would be considered an artificial cell?' 

Their reaction takes place in an emulsion containing an azide oil and an alkyne lipid, which snap together in a click chemistry reaction - the copper-catalysed azide-alkyne cycloaddition popularised by Nobel prize winning chemist Barry Sharpless - to make artificial phospholipids. The oil droplets feed the reaction until, over a matter of hours, they are consumed to leave micrometre-scale vesicles enclosed in artificial phospholipid membranes. 

But Devaraj is already thinking about how he can check off other items on the list. Without wanting to disclose too much about his future plans, he says he is now looking at ways to make these basic compartments grow, reproduce and achieve more complex functions, using unnatural chemistries.

Mann is also pursuing membrane-bound strategies, straying even further from those employed by natural cells. Last year, his team used silica nanoparticles to make porous, self-assembling membranes. These silicon shells were devoid of organic components but capable of housing DNA, and, like real cells, providing space for enzyme-catalysed reactions to take place in their pores. This permeability is not only crucial for enzyme activity, but also allows cells to ship materials, like proteins, in and out.

Booting out carbon

Such inorganic approaches appeal to those keen to draw a line between natural biological cells and so-called 'chemical cells', by taking carbon out of the equation completely. Lee Cronin, a chemist at the University of Glasgow, UK, is in the non-carbon camp and builds his chemical cells from building blocks called polyoxometalates (POMs). 

Through ion exchange reactions, these large molecular anions can form membranes with very little prompting.3 Cronin's POMs are made up of a range of metals, plus oxygen and normally phosphorus or nitrogen. The POMs are first paired with small cations like H+. This salt is then injected into an organic cation droplet - containing comparatively larger cations - and the small cations that were originally paired with the POMs are replaced by the larger ones. These new POMs are insoluble in the bulk solution, and therefore aggregate together to form an insoluble boundary that acts as a membrane surrounding a 'cell', or as Cronin calls it, an iCHELL. Like Mann's silicon shells, iCHELLs can be made porous by using POMs with tiny holes - Mo154 nanowheels, for instance, have nanoscale pores.

What's more, repeating the reaction inside the original droplet - by squirting in more cation and then more of the same or a different POM - forms a 'cell within a cell', paving the way for more complex compartmentalisation and partitioning between and within reactions. 'You could pass molecules from one region to the other and increase the reducing potential, say, from substrate to product and actually engineer different reaction pathways,' Cronin says. 'So you could engineer a kind of metabolism, or a way of splitting water.' 

One limitation of this work is that the iCHELLs have to be made at high ionic strength, but they can be uniformly mass produced using a microfluidic device, as long as they remain in a similarly ionic environment.

Towards artificial evolution

An artificial metabolism might represent a step towards more lifelike properties, but what about everything else that goes into making a cell? When can we really say that scientists have reinvented life itself? 

Perhaps genome king Craig Venter has already done it - with his synthetic cell. But many scientists complain that Venter hasn't created anything new. While he may have made his genome from scratch, the string of code was essentially copied from a bacterium, given a cursory edit, and dumped into a vacated bacterial cell. Life? Only as we already knew it. 

One proposed definition of life is that it must be able to evolve autonomously. But the ability to mutate and evolve brings with it all the kinds of ethical and environmental concerns that haunt bioethicists in their sleep. 

From Mann's point of view, one potential advantage of the chemical cell approach over that of synthetic biology is that it doesn't rely on biology's propagating material, DNA, and therefore circumvents many of these concerns, as well as a whole load of technological ones. 'You could leave out that evolutionary capacity, but have the complexity of the interacting machinery inside the compartments,' he says.

Cronin, on the other hand, argues that life, in the strictest sense, has to be able to evolve. But he's confident that he'll be able to create artificial cells that can do just that. So confident, in fact, that during his July 2011 talk at the TEDGlobal conference in Edinburgh, UK, he set himself a deadline of two years. 'I wanted to challenge everyone's view that making living systems was a timescale beyond normal chemistry,' he says. 'I'm pretty excited because I think the key thing we are going to do in the next year or two is demonstrate in a device that we can do exactly what I said in the TED talk.' And he plans to do this all without carbon. Of course, he's not saying how.

Divide and conquer

So artificial evolution may not be far off, but how do you replicate an artificial cell? Under a simple model, adding material to the membrane should eventually force it to a point where it falls apart, just to reduce its surface energy - straightforward cell division. But when a living cell produces daughter cells, it's not always as easy as divvying up its contents between the two; although DNA has to be copied, molecules like proteins are sometimes asymmetrically inherited. 

Christine Keating, a chemist at Pennsylvania State University in the US, has been working on a model of cell division that accounts for this asymmetry, finding her inspiration in the protein sorting that occurs in yeast.4 'When the mother cell gets old, she has these clumpy denatured proteins but when baby cells form, they don't inherit the bad proteins - only the good proteins,' she says. 'I was fascinated by this idea that the mother cells are able to decide, "Let me keep the crappy ones and you have the nice ones". And I thought, "We could do that".' 

Her lipid-lined vesicles make use of phase separation between two well characterised polymers - polyethylene glycol (PEG) and dextran - and are able to weed out damaged proteins. Denatured proteins have exposed hydrophobic residues that mean they're happier with hydrophobic PEG, while the others side with dextran. Because the vesicles split asymmetrically along the phase boundary, the good proteins are separated from the bad ones, just like in the yeast.

Keating's phase separation model deserves further attention though. In her lipid vesicles, PEG and dextran act as an 'artificial cytoplasm' that phase separates to produce two plainly visible microcompartments, but whether real cells partition in the same way remains an open question in biology. Their interiors are so stuffed full with thousands of different molecules that the boundaries between what might be multiple phases would be almost impossible to spot. 

Questions like these - about the very nature of cells and how they organise themselves - remind us that biology has got it all worked out in ways we still don't understand. The level of complexity is astounding; thus, most scientists working in the area are still attempting to distil the basic elements that can represent life.

But it's the bigger picture that Mann is interested in - and he's taking a systems thinking approach to trying to understand it.5 'I'm interested in viewing the chemical cell as an organisational problem,' says Mann. 'It's about how these components are put together to generate metabolism and information transfer and energy flow and all these things, not just the components themselves.' So as our understanding of life continues to evolve, can we continue to define it by its components, like items on a shopping list? Perhaps only when we have a better understanding of the chemistry of life itself can we really begin to approximate it in the chemistry lab.

Hayley Birch is a science writer based in Bristol, UK

The fusion approach

Another route to making artificial cells involves fusing chemistry and electronics. At the Ruhr University Bochum in Germany, John McCaskill's team is working on a hybrid approach in which computer chips interface directly with chemical components in cell-like systems.6

'The whole entity forms the cell,' explains McCaskill. 'You can regard our computer chip as a sort of sea of electronics, and local pieces of those electronics can choose to interact with local pieces of the chemistry in the same way as local molecules inside a cell can interact with one another.' The electronic components are at the same scale as the chemical ones, with electrodes forming parts of the electrochemical cell's structure - acting as membranes that prevent material from moving in and out, for instance.

Currently, the team can use their electrochemical cell to drive self-replication of DNA-like molecules inside the cell structures. These molecules switch states, zipping together or pulling apart based on the pH of their environment, which can be altered by an electronic signal.

McCaskill is now working towards a hybrid system that can replicate its entire 'cell' structure on its own. 'We'll present this medium of chemicals and naked electronics and then the cell-like structures will proliferate by themselves, making use of both the electronics and the chemistry to do that.' At least, he adds, that's the plan.