In 1978, Tom Cech started his lab at the University of Colorado and the project he set out on was to study the processing of ribosomal RNA (rRNA) in the ciliated protozoa, Tetrahymena. Tetrahymena was chosen because its rRNA genes are amplified about 10,000 times, so it is likely to provide a ready source of rRNA precursors. Of note is that the 26S rRNA gene in Tetrahymena is interrupted by an intron of 400 nucleotides.

A nuclear extract was developed to produce radioactive precursors (Zaug and Cech, 1980). A distinct 9S RNA molecule was evident in the gels and this turned out to be the 400-nucleotide intron. The appearance of this internal sequence indicated that RNA splicing was occurring in the nuclear extracts.

The obvious next thing was to characterize the splicing of the intron. To produce a 32P-labelled RNA substrate, labelled RNA was phenol extracted from nuclei and characterized by gel electrophoresis. In a wonderfully understated paragraph the discovery of RNA catalysis is reported (Cech et al., 1981):

“We hoped to use the unspliced pre-rRNA ... as a substrate to assay for splicing activity in nuclear extracts. This approach proved to be impossible, however, because excision of the intervening sequence occurred when the isolated RNA was incubated in transcription cocktail at normal salt concentration in the absence of a nuclear extract.”

The in vitro transcribed RNA still underwent the self-splicing reaction

Papers rarely describe the actual circumstance of a discovery. In this case Art Zaug had carried out an in vitro splicing assay with a control lane containing no extract. The splicing reaction still took place. Exhaustive efforts to remove any protein still bound to the RNA after phenol extraction did not affect the reaction but this was not considered definite proof of the absence of a protein.

In a later paper, however, Kruger et al. (1982) described the cloning of a portion of the Tetrahymena rRNA gene containing the intron into an Escherichia coli plasmid. The in vitro transcribed RNA still underwent the self-splicing reaction, thereby proving that no protein is necessary.

Quite remarkably, Cech et al. (1981) also reported the correct mechanism of the reaction. GTP was found to be a required co-factor in the reaction, in which it forms a standard 3′–5′ linkage with the 5′ end of the intron, suggesting that the 3′-OH of GTP initiates the reaction by attacking the phosphodiester bond at the 5′ splice site.

In addition to the revolutionary implications this result has for biochemistry, there are two additional points to be made. First, the paper by Cech et al. (1981) is science at its best. A bound protein catalyst would have been interesting but not revolutionary. Everything possible to prove or disprove the bound protein hypothesis was done. Second, that RNA can be a catalyst has profound implications for the origin of life. Protein enzymes might have appeared but they could not replicate, so there must have been a ribozyme that could catalyse its own replication. Such ribozymes were soon discovered (Doudna and Sjostak, 1989).