10.2.1 The concept of totipotency
Many somatic plant cells, including some fully differentiated types (e.g. leaf mesophyll), provided they contain intact nuclear, plastid and mitochondrial genomes, have the capacity to regenerate into whole plants. This phenomenon is totipotency, an amazing developmental plasticity that sets plant cells apart from most of their animal counterparts, and was ﬁrst demonstrated by Steward and Reinert in the 1950s. Often totipotency is revealed when cells or tissues are disturbed or removed from their normal environment and, for example, placed onto artiﬁcial media in tissue culture. A differentiated plant cell that is selectively expressing its genetic information can instead initiate expression of the program required for generation of an entire new plant. Many plants have been regenerated from single cells, but not all plant cells are totipotent; some are terminally differentiated, often because of partial or complete genome loss. We can generalise by saying that most plants at most stages of the life cycle have some populations of cells that are totipotent. Totipotency is of course also a property of normal undifferentiated cells, for example in meristems.
The ﬁrst step in expression of regenerative totipotency is for mature cells to re-enter the cell cycle and resume cell division — a process known as dedifferentiation. This may lead directly to organised development, such as occurs in the epidermal cells of immature hypocotyls of Trifolium (Maheswaran and Williams 1985) where somatic embryos develop (direct embryogenesis), or formation of shoots or roots (direct organogenesis). Alternatively, there may be an intervening callus stage from which organised structures can later be induced to develop — referred to as indirect organogenesis.
Expression of totipotency depends on competence, by which we mean the ability of cells to be induced along a particular developmental pathway, and determination, in which cells become irreversibly committed to a particular pathway. Convolvulus explants display an initial competence to follow two possible developmental pathways — root or shoot formation. Later, once induction of, say, shoots has begun, cells become determined and transfer to conditions that normally induce root formation are now ineffective. However, formation of callus does not necessarily guarantee subsequent organogenesis or that the direction of organogenesis can be controlled. Commitment of root primordia in cereal callus cultures often seems to be irreversible and a high proportion of cells become terminally differentiated root cap cells that secrete the mucus normally associated with caps of intact roots. Calluses which do not lead to regeneration are a common occurrence.
One intriguing question is whether expression of totipotency is a phenomenon of a single cell or normally results from the collective interaction of a cluster of cells (Williams and Maheswaran 1986). We might expect somatic embryogenesis to be an ideal experimental system because normal zygotic embryogenesis always starts from a single cell, the fertilised egg. Perhaps surprisingly, groups of hypocotyl cells from very young embryos may contribute collectively to formation of an embryo ‘bud’; on the other hand, single epi-dermal cells from more mature tissues can divide to produce an embryo. The ability to undergo direct organogenesis may be linked to developmental age of explant tissue, with cells progressively losing this potential as they mature. Fully mature cells, if they retain any capacity for dedifferentiation, tend to exhibit totipotency via indirect organogenesis. Loss of totipotency is probably due to genetic (physical changes to chromosomes, for example loss of DNA, nucleotide sub-stitution, endopolyploidy) or epigenetic (changes in gene expression as a consequence of development, for example DNA methylation) blocks.