Some examples of "heresies" concerning the dogma are :

. Genetic information can also travels from RNA to DNA (reverse transcription)

. Some DNA genomic sequences that are translated into proteins via an intermediate messenger RNA (mRNA) are found scattered by the insertion of non-coding sequences (introns). Introns are cleaved from the precursor mRNAs to give the mature RNA support for protein synthesis (splicing)

. The elucidation of the replication cycles of infectious protein entities called prions and of small plant RNA viruses, devoided of proteins, called viroids may also lead to the finding of new deviations of the dogma

. Protein splicing in which an internal polypeptide (Intein) is excised from a protein precursor and the terminal polypeptides (Exteins) are ligated together resulting in two or more proteins from a single primary translation product

In the late 80s an interesting observation was obtained by several laboratories, including ours. When comparing the sequences of the corresponding DNA and mRNAs or between the DNA gene (which was intron-free) and the protein encoded in plant mitochondria from higher plants, some important differences were observed. Several uridine residues in mRNAs were detected where C residues were supposed to be present as deduced from the mitochondrial gene sequence. This process was called RNA editing. Further biochemical studies in our group and in a german laboratory showed that the C to U modification involved a deamination mechanism of the base cytidine to become uridine.

RNA editing has been described in several organisms; from trypanosomes, where it was discovered for the first time, to animal cells, plant organelles and some viruses like the hepatitis delta virus and paramyxoviruses. Therefore, due to the mRNA modifications introduced by the RNA editing process, it is not possible to deduce the precise sequence of a given protein from the genomic DNA. The biochemical mechanism of this process may be quite different: in trypanosomes it involves the deletion and insertion of U tracts; in paramyxovirus, RNA editing is related to the "stuttering" of the viral RNA polymerase during replication. In the case of hepatitis delta virus as well as in human intestinal cells and plant mitochondria a deamination process leading to A-to-I and C-to-U modifications are at the origin of this process.

While studying plant mitochondrial gene expression we became interested in a consequence of the dysfunction of these organelles, the cytoplasmic male sterility (CMS), which has been observed in a wide variety of higher plants and is characterized by the very low level or the complete absence of pollen production. This genetic trait is important because most plants of agronomical interest are hermaphrodites and to obtain hybrid plants is necessary to eliminate the pollen production organs. CMS is widely used in agriculture to obtain hybrid seeds since it overcomes the problem of the high cost of hand and mechanical emasculation. CMS results from the coordinated action of nuclear genes able to induce sterility with the mitochondrial genes of a sensitive cytoplasm.

This nucleo-cytoplasmic incompatibility leading to a mitochondrial dysfunction, is associated with mitochondrial DNA rearrangements producing chimeric "new" polypeptides formed by sequences of different origins, as studied in detail in maize and petunia. Mitochondria is the main source of cell energy under the form of ATP to be used for fundamental biochemical pathways in plant tissues, while chloroplasts generates energy mainly for the biosynthesis of reserve products in photosynthetic tissues. Mitochondrial function is assured, in part, by the genes encoded in the organelle genome, and therefore, the presence of modified polypeptides may disturb its crucial bioenergetic function. It has been suggested that the CMS phenotype affects essentially the function of anthers because of the high requirement of energy required by this tissue.

Our aim was to create male sterile plants by addressing modified mitochondrial proteins to the organelles in order to induce CMS. For that purpose the RNA editing process seemed very attractive to produce artificially male sterile plants. While other laboratories used different approaches to attain this goal, we thought that a mitochondrial dysfunction produced by a modified protein interfering with the organelle function, should dramatically affect pollen production. The mitochondria from other plant organs, less energy consuming, should overcome the consequences of integrating this "foreign" protein.

To test this hypothesis we performed an experiment using a gene encoding the subunit 9 of the mitochondrial enzyme ATP synthase (atp9) which plays a crucial role in the production of ATP. A functional ATP9 protein is normally produced from a mRNA which has been modified by RNA editing by changing four residues and shortening the protein of 6 carboxy-end residues compared to the gene encoded form (mentioned here as "unedited protein"). In normal mitochondria the "unedited ATP9 protein" was not detected. Assuming that the unedited atp9 mRNA when translated into a protein should give a nonfunctional ATP9 subunit, we constructed nuclear transgenic tobacco plants with either the edited and unedited forms of atp9 under the control of a strong promoter and terminator.

In summary, the possibility to produce artificial male-sterile plants, highly useful for crop breeding in agriculture, has been obtained using several approaches. We think that the system developed in our laboratory, using an unedited mitochondrial gene plus the antisense approach to restore fertility, represents a model closer to a physiological system as compared with models using foreign, degradative enzymes to attain the same goal. In fact, we use a homologous plant gene and we exploit a physiological process, RNA editing, which plays a crucial role in plant mitochondria gene expression.