Plant Molecular Biology 14: 727-733, 1990. © 1990 Kluwer Academic Publishers. Printed in Belgium. 727 The minimum functional length of pre-mRNA introns in monocots and dicots Gregory J. Goodall and Witold Filipowicz Friedrich Miescher Institute, Postbox 2543, CH-4002 Basel Switzerland Received 31 October 1989; accepted in revised form 9 January 1990 Key words: intron length, protoplast, splicing, transient expression Abstract When exogenous genes are to be expressed in transgenic plants, their RNAs must be correctly processed. To gain information useful for predicting whether foreign introns will be accurately spliced, we have analysed the processing of an artificial gene in maize and Nicotiana plumbaginifolia protoplasts. A synthetic plant intron, devised to contain the elements necessary for pre-mRNA splicing in dicots, was found to be efficiently spliced in a monocot (maize) transient expression system. A series of deletion mutants of the synthetic intron was constructed to assess the minimum functional intron length. In both monocots and dicots this was found to be between 70 and 73 nt. This length requirement is similar to that seen in vertebrates, but significantly greater than that in fungi and insects. Introduction Certain features of nuclear pre-mRNA introns, in particular the splice site consensus sequences, have been conserved throughout the evolution of many organisms, while other features are specific to groups of species (for reviews see [ 1, 2]). For example, the yeast Saccharomyces cerevisiae re- quires the sequence UACUAAC as a branch site [ 1, 2, 3], metazoa require a polypyrimidine tract between the branch site, which is loosely defined, and the 3' splice site [4, 5], and plants require AU-rich sequences and a loosely defined branch site, but no polypyrimidine tract [6]. In addition, the lengths of introns also differ between the various phyla. Plant introns mostly fall between 70 and 1000 nt in length, vertebrate introns from 70 to thousands of nt, fungal introns between 40 and 100 nt and insect introns between 50 and 100 nt [7]. The ability to predict whether a particular se- quence will be recognized as an intron in plants is important to those who wish to express heterol- ogous genes in transgenic plants. In addition, defmition of the minimum size of a functional intron is crucial to the interpretation of some ex- perimental results. The work reported here was in part prompted by the report of Martinez-Zapater etal. [8] who have made a preliminary assess- ment of the usefulness of the Drosophila P-element as a transposon tagging system in plants. They found that introns 1 and 2 of the P-element trans- posase gene are not spliced in transgenic plants (the test gene lacked intron 3), and proposed that the failure of these introns to be spliced must reflect the absence of plant-specific sequence ele- ments or structures necessary for splicing. We suspected that these two introns, 58 and 53 nt in length [9] respectively, were simply too short to be spliced in plants.