Info

Repeats were in direct orientation apart from Rogalski et al. 2006. Most recombinant genomes contained two engineered repeats apart from 1three 418 bp and two 174 bp repeats (Iamtham and Day 2000). 2Restriction fragment from pACYC184 plasmid. IR = inverted repeat. RBS = ribosome binding site. HPPD = 4-hydroxyphenylpyruvate dioxygenase.

Repeats were in direct orientation apart from Rogalski et al. 2006. Most recombinant genomes contained two engineered repeats apart from 1three 418 bp and two 174 bp repeats (Iamtham and Day 2000). 2Restriction fragment from pACYC184 plasmid. IR = inverted repeat. RBS = ribosome binding site. HPPD = 4-hydroxyphenylpyruvate dioxygenase.

DNA (Fig. 11a) while recombination between inverted repeats reverses the orientation of the intervening DNA. Both length and number of repeated sequences influence recombination frequency. Whereas two 418 bp direct repeats were ineffective in deleting intervening DNA, three 418 bp repeats promoted high frequencies of excision to leave a single 418 bp direct repeat (Iamtham and Day 2000). Intermediate forms containing two copies of the 418 bp repeat were not detected indicating that once activated the homologous recombination pathway goes to completion (Fig. 11b). A variety of direct repeats promote excision (Table 1) but a systematic study on the relationship between repeat length, DNA sequence and recombination frequency has not been carried out. Whereas recombination was barely detected between two 3'UTR psb A direct repeats of 418 bp (Iamtham and Day 2000), excision was promoted by repeated ~200 bp 16S rrn promoter elements in L. sativa (Lelivelt et al. 2005), and N. tabacum (Zou et al. 2003), and 210 bp rbc L 3' UTR repeats in N. tabacum (Dufourmantel et al. 2007). Recombination between 232 bp invertedpsbA 3' UTR repeats has been shown in N. tabacum plastids (Rogalski et al. 2006). Excision can take place at any time during the transformation process and non-deleterious transgenes flanked by direct repeats, might be stabilised by the high copy number of plastid DNA once homoplasmy has been reached. Promoter regions of ~120 bp are duplicated in a number of plastid transformation vectors with no apparent reports of instability (Zoubenko et al. 1994) hinting at a lower size limit for efficient recombination.

Fig. 11. Homologous recombination between engineered direct repeats (arrows) in plastid DNA. a) Recombination between two 649 base atpB promoter repeats (Kode et al. 2006) excises the rbcL gene and a marker gene (not shown). b) Recombination between three 418 bp 3' psbA repeats leaves a single 418 bp 3' psbA sequence. Once recombination is activated the pathway appears to go to completion because intermediates containing two repeats are rare and were not found (Iamtham and Day 2000). c) The recombination event shown in a) gives rise to pale green sectors (light areas) which appear randomly throughout leaf development.

Fig. 11. Homologous recombination between engineered direct repeats (arrows) in plastid DNA. a) Recombination between two 649 base atpB promoter repeats (Kode et al. 2006) excises the rbcL gene and a marker gene (not shown). b) Recombination between three 418 bp 3' psbA repeats leaves a single 418 bp 3' psbA sequence. Once recombination is activated the pathway appears to go to completion because intermediates containing two repeats are rare and were not found (Iamtham and Day 2000). c) The recombination event shown in a) gives rise to pale green sectors (light areas) which appear randomly throughout leaf development.

The types of intervening genes excised by flanking direct repeats might also influence the accumulation of recombination products. Selection on plastid fitness might be expected to promote the division of plastids from which foreign genes with a negative impact on plastid functions and division have been deleted. The finding that excision mediated by recombination between duplicated 649 bp atpB 5' regulatory regions allows the isolation of mutant defective plastid genomes lacking the rbcL gene (Fig. 11a; Kode et al. 2006) indicates that products of deleterious recombination events can be isolated under suitable conditions; in this case sucrose was provided in the media to allow non-photosynthetic growth. Recombination-mediated excision of rbcL and segregation of plastid genomes gives rise to pale-green sectors allowing the process to be monitored. Pale-green sectors of varying sizes are visualised in leaves (Fig. 11c, lighter areas represent pale-green sectors) indicating the recombination and segregation pathways are active throughout leaf development. The variable sizes and random appearance of pale-green sectors reflect random spontaneous excision events combined with stochastic replication and segregation of plastid genomes.

In E. coli, a minimum sequence length of 23-27 bp is considered to be required for efficient homologous recombination via the recBC-dependent pathway (Shen and Huang 1986). A minimum identical stretch of 150 to 200 bases appears to be required for homologous recombination in C. reinhardtii (Newman et al. 1992). Recombination between engineered direct repeats of 216 bp (Cerutti et al. 1995) has been observed in C. reinhardtii (Cerutti et al. 1995). In another study recombination was not detected between 100 bp or 230 bp direct repeats but frequent recombination was detected between 483 bp repeats (Fischer et al. 1996). This is probably because the recombination assay for the 216 bp repeats relied on restoration of gene function giving rise to green sectors and was more sensitive than the loss of antibiotic resistance assay involving the 230 bp direct repeats flanking aadA (Fischer et al. 1996). However, sequence-dependent differences in recombination rates between direct repeats cannot be ruled out (Fischer et al. 1996).

Site-specific recombination is an alternative to homologous recombination for manipulating plastid genomes. The Cre site-specific recombinase from the P1 bacteriophage of E. coli mediates strand-exchange between 34 bp LoxP sites (Sternberg and Hamilton 1981). When Cre is introduced into plastids it recombines LoxP sites as intended (Corneille et al. 2001; Hajdukiewicz et al. 2001). Unexpectedly, Cre recombinase also appears to stimulate recombination between ~120 bp direct repeats comprised of the 16S rrn promoter region (Corneille et al. 2001; Hajdukiewicz et al. 2001). The WT N. tabacum plastid genome lacking LoxP sites is stable in the presence of Cre recombinase (Corneille et al. 2003). This suggests that creation of double strand breaks at LoxP sites by Cre recombi-nase stimulates native recombination events in plastids. Cre-stimulated illegitimate recombination events between a LoxP site and a recombination hotspot in the promoter region of the rps7/3'rps12 operon were also reported (Hajdukiewicz et al. 2001; Corneille et al. 2003). The recombination hotspot contained multiple copies of a TATTA sequence (Hajdukiewicz et al. 2001). Short repeats are often associated with recombination hotspots in plastid DNA. The role of short multiple 18 to 37 bp repeats near a recombination hot spot in C. reinhardtii was addressed by deleting them. Their deletion did not reduce recombination frequency (Newman et al. 1992) indicating the repeats were not responsible for increased recombination. The observation that Cre-cleaved DNA ends are recombinogenic might suggest that the natural ends of the linear DNA molecules found in plastids (Oldenburg and Bendich 2004b; Scharff and Koop 2006) are protected by proteins or secondary structures, for example, DNA loops or hairpins.

Spontaneous excision of an 868 bp sequence following apparent recombination between 16 bp imperfect direct repeats (5' GTACTGc/tGCTCTCCAA) was reported to accompany plastid transformation in N. tabacum (see Section 5 above; Staub and Maliga 1994). This might indicate that some plastid sequences of less that 20 bp are effective substrates for recombination. Evolutionary comparisons of plastid genomes have found DNA sequence inversions in the large single copy region that distinguish related species of flowering plants (Doyle et al. 1992). Analyses of the end points of an inversion in rice plastid DNA relative to N. ta-bacum suggest recombination events between sequences as short as 16 bp in length (Hiratsuka et al. 1989). In another study, short inverted repeats of 7-11 bp were found to be associated with inversions of the intervening 4 bp region (Kelchner and Wendel 1996). These studies indicate recombination events between short repeated stretches of nucleotides. Whether the same recombination pathway acts on the very short (10-20 bp) and longer (~200 bp and above) substrates remains to be determined. These questions can only be addressed once mutants in specific plastid recombination pathways have been isolated.

Was this article helpful?

0 0

Post a comment