Figure 1. Evolving a fluorescent protein in the laboratory.

(A) Mihajlovic et al. studied a gene called cogfp that codes for a fluorescent protein that can emit both blue and green light. They prepared plasmids containing two active copies of the gene (arrows), and randomly introduced mutations (red lines) into the copies. The plasmids were then inserted into E coli. (oval shapes with tails), which were sorted into those that emitted mostly blue light, those that emitted mostly green light, and those that emitted strongly at both wavelengths (represented here as turquoise). The cells which shone the brightest were selected, their plasmids were removed and the cycle was repeated again. (B) Mihajlovic et al. carried out the experiment on two populations: bacteria which contained two active copies of the cogfp gene (double-copy population; right), and bacteria which contained one active copy and one inactivated copy of cogfp (single-copy population; left). According to Ohno’s hypothesis, the single-copy population will experience adaptive conflict: mutations that improve green fluorescence will lead to a reduction in blue fluorescence, and vice versa. Consequently, these bacteria will become only marginally brighter over the course of evolution. In the double-copy population, one gene can evolve to increase green fluorescence while the other can evolve to increase blue fluorescence, resulting in these bacteria becoming significantly brighter over time. (C) However, Mihajlovic et al. found that – contrary to what Ohno’s hypothesis would suggest – the increase in brightness was essentially the same for the two populations. This happened because one of the two active genes in the double-copy population had been inactivated (marked with X) by deleterious mutations (red line) during evolution – an effect known as Ohno’s dilemma.