Annals of Botany 94(4):481-495 (2004)
The Effect of Stress on Genome Regulation and Structure
Andreas Madlung and Luca Comai

Genome methylation and control of heterochromatin
The silencing of DNA sequences that are potentially hazardous to the organism is considered an important housekeeping function. Heterochromatin is commonly regarded as such ‘silent’ DNA. It consists of large regions of repetitive nucleotide sequences and transposons, many of which are more or less degenerate. At least some heterochromatic sequences serve an important role. While not coding for proteins, ribosomal RNA genes are needed for ribosome synthesis, while centromeres and telomeres are essential for the stability of chromosomes. Transcriptionally silent and densely packaged DNA is necessary for proper function during chromosome segregation and cell division. Transposons, however, must be suppressed because they constitute two dangers for the genome: (1) their repeated units can cause spurious homologous recombination; and (2) their ability to transpose can lead to disruption or misregulation of important genes. Both of these dangers are suppressed by heterochromatinization. Since the heterochromatic state is mitotically stable it serves as an epigenetic mark that designates these regions as heterochromatin through multiple cell cycles.

Transposons. Together, DNA and RNA transposons comprise a large component of most genomes, and can be considered as intracellular pathogens. Indeed, the difference between transposons and viruses is blurred in the gypsy class of retroelements (Bucheton, 1995) and in other plant viruses and retrotransposons (Richert-Poggeler and Shepherd, 1997; Harper et al., 2002). RNA- or retrotransposons, also known as class I transposons, replicate via an RNA intermediate. This is in contrast to class II transposons (DNA transposons), which replicate via a cut-and-paste mechanism. Retrotransposons are divided into two subgroups. While all retrotransposons encode a reverse transcriptase some are flanked by long terminal repeats (LTRs) while others are not (non-LTR elements). Depending on the position of the gene for the enzyme integrase involved in transposition, the LTR-containing group can further be subdivided into Ty3/gypsy-like, or Ty1/copia-like retrotransposons (Galun, 2003). In contrast to terminal inverted repeats (TIRs) of class II transposons, LTRs are direct repeats, usually several hundred base pairs in length (Galun, 2003). LTRs contain the elements’ promoter and enhancer (Pauls et al., 1994). Recombination can result in the separation of one LTR from the protein-coding sequences of the transposon it is flanking and produce so called Solo LTRs. Ty1/copia-like transposons initiate transcription within the LTR (Voytas and Boeke, 2002). This has important implications for the role that Solo LTRs may play in the genome with respect to the potential control of neighbouring genes. Transposons may also play an important role in the evolution of gene function and may be involved in the restructuring of genomes due to their ability to restructure or rearrange chromosomes (Agrawal et al., 2001; Witte et al., 2001).

Allopolyploidization and hybridization as a cause for genomic stress
The fourth example of a stress inflicted upon a plant is that of invasion of a foreign genome by way of fertilization with pollen of a different species. In most cases, plants have evolved barriers preventing the fusion of its gametes with those of individuals of a different species. However, interspecific hybridization between close relatives can occur, although it normally results in sterile offspring. But in some cases such as crosses in which chromosome doubling occurs before or immediately after hybridization, fertile progeny can arise, which, owing to the presence of duplicated parental genomes, are called allopolyploids. In allopolyploids intergenomic recombination of the homologous chromosomes (those contributed from the two different parental species) is infrequent (Comai  et al., 2003). The chromosomes of the two original species are instead retained independently throughout subsequent generations. Allopolyploidization is an important process through which new species may theoretically arise quickly. However, estimates on how many species have arisen through allopolyploidy vary (for a recent review on the evolution of plant polyploids, see Liu and Wendel, 2003). Although direct comparisons between allopolyploids and their progenitors have only been conducted for few species (Liu and Wendel, 2003), established allopolyploids are often vigorous in growth and high in seed yield and fertility. Indeed, many of today’s crop plants are of allopolyploid origin.