We have described earlier (Goebel et al., MNL 1986) the surprisingly large number of phenotypic distortions that were found in maize plants of the line A188 regenerated via somatic embryogenesis. In an attempt to characterise the molecular basis for this variation, we have undertaken a series of experiments to determine whether, for example, gene methylation status is altered as a consequence of tissue culture stress and whether somaclonal variation can be related to methylation changes.
One of the first experiments was to compare methylation status in phenotypically normal and abnormal plants. By digestion with methylation-sensitive enzymes such as MspI, HpaII, DpnI, Sau3A and MboI and subsequent probing with known gene sequences, the results showed that no correlation exists between obvious symptoms of somaclonal variation and methylation status. Plants which appear to be phenotypically normal can be shown, with different genes, to be methylation-abnormal. Conversely, dependent on the gene used, phenotypically abnormal plants can appear to be genetically identical to controls.
We further characterised plants by producing 'lines' of plants; i.e., the progeny of plants all derived from a single embryo, and subsequently compared methylation status not only between 'lines' (i.e., different embryo-derived plants) but within the progeny of the same line. Plants have been examined for up to 5 generations. Although culture conditions were identical for all the original calli, highly significant differences have been found, not only between sibling plants but between lines derived from different embryos on the same callus, as well as plants from embryos produced on different calli. The only consistent pattern appeared to be within the progeny of plants. Therefore, once the methylation status has been changed, then the alteration appears to be stable for the subsequent generations.
As methylation sequences are also assumed to be mutational 'hotspots', RFLP analysis is also being undertaken in an attempt to further characterise genetic alterations. Once again, the amount of variation was significantly much greater than would have been expected. In an attempt to determine which factors during tissue culture are responsible for inducing these mutations, different components of the procedure are being examined, both on callus cultures as well as on direct regenerants. As such, we have found that highly significant differences in genetic status can be produced by the media, some of the more widely used media being considerably more mutagenic than others. There is also evidence emerging for genetic alterations occurring during the process of regeneration from callus to plants, as opposed to induction from immature embryos.
The amount of variation that we have found so far is considerable and has led us to suspect that this may be a reflection not just of tissue culture per se, but a characteristic of the line A188. This work is now being repeated with a number of different maize lines as well as with rice, enabling us to determine what differences, if any, can be found between protoplast and callus derived plants.
Peter Brown, Elke Goebel and Horst Lorz
Return to the MNL 62 On-Line Index
Return to the Maize Newsletter Index
Return to the Maize Genome Database Page