Good old-fashioned (anti)sense

Two decades ago, in the early years of recombinant DNA technology, scientists posited the elegant proposal that expression of a eukaryotic target gene could be blocked by engineering production of a complementary “antisense” RNA that would base pair with the target transcript and somehow interrupt its function (1). And it worked, in organisms as divergent as fruit flies (2), cellular slime molds (3), and plants (4, 5), but not always as expected. In plants, for example, occasionally an ectopically expressed “sense” strand was as effective as antisense at suppressing expression of the target gene (6). An underlying naiveté lay in the reasonable expectation from the central dogma of genetics that RNA normally behaves itself as an intermediate between gene and protein and leaves the regulatory business to transcription and translation factors. A few reports of small regulatory RNAs in Escherichia coli (7, 8) were not recognized at the time as harbingers of unforeseen RNA-based regulatory schemes that lurk in the depths of diverse cells. In the ensuing years, it has become obvious that both eukaryotic and prokaryotic organisms possess intricate regulatory systems based on noncoding RNAs (ncRNAs) (6, 9, 10). Those early antisense experiments in transgenic eukaryotes inadvertently tapped into unrecognized endogenous pathways of surprising complexity and remarkable epigenetic consequence (6, 10). The observed effects on gene suppression were not, as expected, mediated by a physical block resulting from extensive base pairing with the target transcript: the active species are now recognized as highly processed 22-nt fragments that act in the context of an elaborate protein complex (10). In this issue of PNAS, Dühring et al. (11) report a natural regulatory event in a cyanobacterium that comes close to the early expectations for straightforward antisense regulation: a 177-nt ncRNA is transcribed in antisense to …

*E-mail: sgolden{at}tamu.edu