Scientists have taken another step in their quest to understand the bare genetic essentials of life.
A team led by US research entrepreneur Craig Venter has created a semi-synthetic, functioning bacterium in the lab that has fewer than 500 genes.
This minimal number is lower than in any known free-living bug in nature.
The group says its investigations aim to push the boundaries of fundamental knowledge and could lead to novel means to make new drugs and other chemicals.
"Our long-term vision has been to design and build synthetic organisms on demand where you can add in specific functions and predict what the outcome is going to be," said Daniel Gibson, who is a co-author on a paper describing the latest work in Science Magazine.
"We think these cells would be a very useful chassis for many industrial applications, from medicine to biochemicals, biofuels, nutrition and agriculture," he told reporters.
The team reported its first semi-synthetic organism in 2010.
In that project, the scientists constructed in the lab the entire "genetic software" of Mycoplasma mycoides, a microbe that lives in cattle and other ruminants.
This artificial package of DNA was then transplanted into the cell of another Mycoplasma species that had been emptied of its genome, and "booted up". The engineered bug, dubbed Syn 1.0, duly started to divide.
In the new paper, Dr Venter and his colleagues report how they have now reduced the biochemical instructions in this organism to the bare minimum.
After a long series of trial and error experiments, the Mycoplasma microbe, now dubbed Syn 3.0, can operate on just 473 genes - about half the number found in the wild bug, and about 50 fewer than in the related Mycoplasma genitalium, which has the smallest set of genes in any independent organism known to science.
By way of comparison, more complex organisms such plants and animals can have many tens of thousands of genes driving their biology.
They say they have now come to recognise the role of many "quasi-essential" genes - those needed for robust growth but not absolutely required for life. The filtering has also retained genes that perform vital functions in a kind of back-up to each other; thus, one or other of a pair of genes could be seen as superfluous, but one absolutely has to stay or the organism will die.
Dr Venter used an aviation metaphor: "If you know nothing about aeroplanes and you're looking at a Boeing 777, and you're trying to find out the function of parts just by removing them, and you remove the engine from the right wing - the airplane can still fly and land. So, you might say that's a non-essential component, but you won't discover the essentiality until you remove the second one.
"And that's what's happened over and over again in biology, where we would have what appeared to be a non-essential component until we removed its counterpart."
Of Syn 3.0's 473 necessary genes, 149 are a mystery - the team does not know their function, and experiments are underway to close that knowledge gap.
The scientists stress that this minimal genome applies only to their semi-synthetic organism. Context is everything. Other microbes will live in different types of environment, with different ways of operating.
A bug that powers itself via sunlight and photosynthesis will not have the same essential set of genes, for example, as an organism that processes methane to derive its chemical energy.
Laurence Hurst is the director of the Milner Centre for Evolution at the University of Bath, UK.
His team stated in 2006 that research on minimal genomes was underestimating what was essential in a cell.
"It was gratifying to see that our prediction that the minimal genome would be larger than previously predicted because of hidden essential genes is indeed borne out. These come about because there are often two routes to the same end," Prof Hurst told BBC News.
"Just as you can close a file on a computer by going to the menu bar and clicking 'quit' or going to the keyboard and typing 'command Q', so too genomes have two means to do the same thing. You can only get rid of one to retain functionality. As soon as one is lost the other becomes essential.
"Looking to the future, the claim is that this could lead the way to a new form of synthetic biology, in which genomes are designed rather than simply modified. The possibilities are exciting but whether this is the best and most cost effective route remains to be seen.
"A complete network analysis of how such a simple system works would, however, make for an excellent starting position to predict what modifications could be successfully incorporated."