Illuminating error

Yesterday’s first-year biology practical involved forcing the laboratory work-horse bacterium Escherichia coli to take up a circular piece of DNA called pGLO.

pGLO contains a few genes, but the most interesting of these DNA sequences encodes a protein from a jellyfish. This protein fluoresces green under UV light, and goes by the thoroughly unimaginative name of green fluorescent protein (GFP). The production of the GFP is regulated by a second DNA sequence, this time taken from a bacterium. The normal job of this regulatory DNA sequence is to switch on and off the production of enzymes that degrade an uncommon sugar called arabinose. Since bacteria don’t often encounter arabinose, the enzymes are not produced unless there is actually some arabinose around to eat. This save resources.

The system on the pGLO is therefore quite artificial: it patches together regulatory DNA from the arabinose-degrading genes, with protein-coding DNA from the jellyfish. The advantage of this is that arabinose will switch on the production of the GFP, which is much easier to see than the arabinose-degrading enzymes.

So the green, fluorescing bits in the plates below show where the bacteria are detecting arabinose and being tricked into making GFP.

The only thing is, they shouldn’t be! The plates these bacteria are growing on do indeed contain arabinose, as you might expect, but they also contain lots of glucose. Arabinose is very much second-best as far as the bacteria are concerned: they would much prefer to eat glucose than arabinose. If there’s glucose present as well as arabinose, the regulatory DNA sequence should switch off the production of the arabinose-degrading enzymes. In these bacteria with pGLO, the glucose should therefore switch off the production of GFP, and this plate should be dark as pitch.

The reason for the unexpected glowing bits is interesting. If you squint, you’ll see that there are four or five fluffy things on the plate as well as the glowing bacteria. These were not supposed to be there. They are fungal colonies, probably of Penicillium; contaminants that have gotten into the plates from the air, possibly ferried into the lab on the clothes of an unsuspecting undergraduate with a fridgeful of mouldy bread.

As the fungi have grown across the plates, half-smothering the bacteria, they have gobbled up all the glucose. The bacteria, starved of their preferred sugar, respond by switching on their arabinose-degrading enzymes and any other genes that happen to be controlled by that same regulatory DNA. And they glow.

Happy accidents are often more illuminating than the “expected results” from practicals that “work”.