Mitochondrial mRNA 3′ cleavage/polyadenylation and RNA editing in Trypanosoma brucei are independent events

Abstract

The organization of the mitochondrial maxicircle genome of Trypanosoma brucei is unique in the close packing of the mRNA genes. For many of them, the 5′ and 3′ ends of adjacent transcripts overlap and formation of the proper 3′ or 5′ end can eliminate a portion of the coding sequence of the adjacent gene. Large, polycistronic transcripts have been detected, suggesting that mechanisms for precise cleavages at both 5′ and 3′ gene boundaries must exist. However, no common sequences near the ends of the mRNAs that could be candidates for control regions have been detected. In addition, nothing is known about how RNA editing interacts with and affects 5′ and 3′ processing and/or polyadenylation. Edited precursor transcripts have been detected, indicating that editing complexes can assemble prior to transcript cleavage. Because editing often initiates near the 3′ end of the mRNA, the assembly of an editing complex in this region may influence the cleavage selection process. In order to determine the extent that RNA editing and 3′ end-processing interact, RNAs were analyzed to determine the extent of editing in precursor RNAs and to determine if unedited transcripts can be cleaved and polyadenylated. Two overlapping RNA junctions were analyzed; the junction between NADH dehydrogenase (ND) subunit 7 and cytochrome oxidase (CO) subunit III, and the junction between CO subunit II and maxicircle unidentified reading frame (MURF) II. For both of these RNAs, editing affects restriction endonuclease recognition sequences, allowing us to analyze editing patterns by differential restriction digests. These analyses suggest that when the gRNA is supplied in trans, RNA editing and cleavage/polyadenylation are independent events and while they may influence one another, one event is not dependent on the other. Conversely, for the COII transcript, where the gRNA is located at the 3′ end of the mRNA and appears to be supplied in cis, edited precursors were not detected. This suggests a requirement for a precise intramolecular interaction for COII editing that cannot form prior to 3′ end-maturation.