Reversion-reporter transgenes to analyze all six base-substitution pathways in Arabidopsis.

To expand the repertoire of Arabidopsis (Arabidopsis thaliana) mutation-reporter transgenes, we constructed six mutant alleles in the same codon of the ß-glucuronidase-encoding GUS transgene. Each allele reverts to GUS+ only via a particular one of the six transition/transversion pathways. AcV5 epitope tags, fused carboxyl terminal to the inactive GUS- proteins, enabled semiquantitative immunoassays in plant protein extracts. Spontaneous G:C→T:A transversions, previously not measured using reporter transgenes, were quite frequent. This may reflect mispairing of adenine with 8-oxoguanine in DNA attacked by endogenous oxyradicals. Spontaneous G:C→A:T was modest and other reversions were relatively low, as reported previously. Frequencies of ultraviolet C-induced TT→TC and TC→TT reversions were both high. With increased transgene copy number, spontaneous G:C→T:A reversions increased but ultraviolet C-induced reversions decreased. Frequencies of some reversion events were reduced among T4 versus T3 generation plants. Based on these and other analyses of sources of experimental variation, we propose guidelines for the employment of these lines to study genotoxic stress in planta.

Reciprocal chromosome translocation associated with TDNA-insertion mutation in Arabidopsis: genetic and cytological analyses of consequences for gametophyte development and for construction of doubly mutant lines.

Chromosomal rearrangements may complicate construction of Arabidopsis with multiple TDNA-insertion mutations. Here, crossing two lines homozygous for insertions in AtREV3 and AtPOLH (chromosomes I and V, respectively) and selfing F1 plants yielded non-Mendelian F2 genotype distributions: frequencies of +/++/+ and 1/1 2/2 progeny were only 0.42 and 0.25%. However, the normal development and fertility of double mutants showed AtPOLH-1 and AtREV3-2 gametes and 1/1 2/2 embryos to be fully viable. F2 distributions could be quantitatively predicted by assuming that F1 selfing produced inviable (1,2) and (+,+) gametophytes 86% of the time. Some defect intrinsic to the F1 selfing process itself thus appeared responsible. In selfing AtREV3 (+/2 ) single mutants, imaging of ovules and pollen showed arrest or abortion, respectively, of half of gametophytes; however, gametogenesis was normal in AtREV3 ( 2/2 ) homozygotes. These findings, taken together, suggested that T-DNA insertion at AtREV3 on chromosome I had caused a reciprocal I-V translocation. Spreads of meiosis I chromosomes in selfing AtREV3 (+/2 ) heterozygotes revealed the predicted cruciform four-chromosome structures, which fluorescence in situ hybridization showed to invariably include both translocated and normal chromosomes I and V. Sequencing of the two junctions of T-DNA with AtREV3 DNA and the two with gene At5g59920 suggested translocation via homologous recombination between independent inverted-repeat T-DNA insertions. Thus, when crosses between TDNA-insertion mutants yield anomalous progeny distributions, TDNA-linked translocations should be considered.